Vol. 10, 2459–2465, April 1, 2004 Clinical Cancer Research 2459

The Tumor Suppressor WWOX at FRA16D Is Involved in Pancreatic Carcinogenesis

Tamotsu Kuroki,1 Sai Yendamuri,1 Furthermore, transfection with WWOX inhibited colony for- Francesco Trapasso,1 Ayumi Matsuyama,1 mation of pancreatic cancer cell lines by triggering apoptosis. Rami I. Aqeilan,1 Hansjuerg Alder,1 Conclusion: These results indicate that the WWOX gene 1 1 2 may play an important role in pancreatic tumor devel- Shashi Rattan, Rossano Cesari, Maria L. Nolli, opment. Noel N. Williams,3 Masaki Mori,4 5 1 Takashi Kanematsu, and Carlo M. Croce INTRODUCTION 1 Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, Pancreatic cancer is one of the most common human can- Pennsylvania; 2Areta International srl, Gerenzano, Italy; 3Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania; cers and carries a poor prognosis (1). Although complete sur- 4Department of Surgery, Medical Institute of Bioregulation, Kyushu gical resection is the only therapeutic option for a long-term University, Beppu, Japan; and 5Department of Surgery, Nagasaki survival, the 5-year survival rate with curative resection is University, Graduate School of Biomedical Sciences, Nagasaki, Japan Ͻ20% (2). The precise molecular mechanisms of development and/or progression of pancreatic cancer remain unknown, al- ABSTRACT though multiple genetic and epigenetic changes have been de- tected in pancreatic cancer. Therefore, additional elucidation of Purpose: WWOX (WW domain containing oxidoreduc- the molecular mechanisms involved in pancreatic cancer is tase) is a tumor suppressor gene that maps to the common urgently needed for more effective treatment. fragile site FRA16D. We showed previously that WWOX is The tumor suppressor gene FHIT spans the FRA3B fragile frequently altered in human lung and esophageal cancers. site, the most active common fragile site of the The purpose of this study was to delineate more precisely the , and abnormal FHIT transcripts have been de- role of WWOX in pancreatic carcinogenesis. tected in many human cancers, including pancreatic cancer Experimental Design: We analyzed 15 paired pancreatic (3–6). Additionally, our previous study demonstrated that FHIT adenocarcinoma samples and 9 pancreatic cancer cell lines overexpression enhanced the susceptibility of pancreatic cancer for WWOX alterations. Colony assay and cell cycle analysis cells to exogenous inducers of apoptosis (7). Recently, in the were also performed to evaluate the role of the WWOX as a WWOX (WW domain containing oxidoreductase), a candidate tumor suppressor gene. tumor suppressor gene, chromosome 16q23.3–24.1 was iso- Results: Loss of heterozygosity at the WWOX locus was lated. This gene maps at the location of FRA16D, one of the observed in 4 primary tumors (27%). Methylation analysis most active common chromosomal fragile sites (8–11). The showed that site-specific promoter hypermethylation was WWOX gene is similar to the FHIT gene in that both are detected in 2 cell lines (22%) and treatment with the de- Ͼ1 Mb and encompass very active common fragile sites, both methylating agent 5-aza-2؅-deoxycytidine demonstrated an genes show frequent allelic loss and/or loss of homozygous increase in the expression of WWOX. In addition, 2 primary deletion region in several human cancers, and both frequently tumor samples (13%) showed promoter hypermethylation show aberrant transcripts (5, 10, 11). Because increased risk of including the position of site-specific methylation. Tran- pancreatic cancer is associated with cigarette smoking and be- scripts missing WWOX exons were detected in 4 cell lines cause the carcinogens present in tobacco smoke cause deletions (44%) and in 2 tumor samples (13%). Real-time reverse at the FHIT/FRA3B loci (5, 6), we hypothesized that the WWOX transcription PCR revealed a significant reduction of gene may play a role in pancreatic cancer. In this report, we WWOX expression in all of the cell lines and in 6 primary describe that WWOX is altered frequently in pancreatic cancer tumors (40%). Western blot analysis showed a significant by genetic and/or epigenetic changes and is involved in pancre- reduction of the WWOX protein in all of the cell lines. atic carcinogenesis.

MATERIALS AND METHODS Received 8/26/03; revised 12/16/03; accepted 1/2/04. Cell Lines and Tissues. Human pancreatic cancer cell Grant support: USPHS Grant CA56036 from the National Cancer lines Panc1, AsPc1, CFPAC1, MiaPaca2, Capan2, BxPc3, Institute. Hs766T, and Su.86.86 were obtained from the American Type The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked Culture Collection. PSN1 was obtained from the European advertisement in accordance with 18 U.S.C. Section 1734 solely to Collection of Cell Cultures. All of the cell lines were maintained indicate this fact. in RPMI 1640 with 10% fetal bovine serum at 37°C and 5% Note: T. Kuroki and S. Yendamuri contributed equally to this study. CO . Tumors and corresponding noncancerous tissues were Requests for reprints: Carlo M. Croce, Kimmel Cancer Institute, 2 Jefferson Medical College, Thomas Jefferson University, BLSB Room obtained from 15 patients who underwent surgery for pancreatic 1050, 233S 10th Street, Philadelphia, PA 19107-5799. Phone: (215) 503- adenocarcinoma. The tissue samples were excised and were 4645, Fax: (215) 923-3528; E-mail: [email protected]. immediately stored at Ϫ80°C. DNA was extracted from each of

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. 2460 WWOX Alterations in Pancreatic Carcinoma

the 9 cell lines and from the 15 paired pancreatic tissues ac- Biosystems Prism 377 DNA sequencing system. Two primer cording to methods described previously (12). RNA was ex- sets were designed to amplify the entire open reading frame. tracted with the Trizol reagent (Invitrogen Life Technologies, Methylation Analysis. Genomic DNA was treated with Inc., Carlsbad, CA) per the manufacturer’s recommendations. sodium bisulfite according to methods described previously Loss of Heterozygosity (LOH) Analysis. Allelic losses (15). Briefly, 1 ␮g of the genomic DNA was denatured by 2 M were analyzed using a PCR approach, with primers amplifying NaOH, and then incubated in 3 M sodium bisulfite and 10 mM polymorphic microsatellites internal to WWOX at loci D16S3029, hydroquinone for 17 h at 55°C. Bisulfite-treated DNA was D16S3096, D16S504, and D16S518, as described elsewhere (13). extracted using a genomic DNA clean-up kit (Promega, Madi- Briefly, the primer sequences were obtained from the Genome son, WI). Modified DNA was amplified using nested PCR that Database and primers were labeled using 5Ј-fluorescein phosphora- amplify the putative promoter region, including a CpG-rich midite or 5Ј-tetrachlorofluorescein phosphoramidite for microsat- area surrounding the translation start codon (11). Primers for Ј ellite loci, as described by Ishii et al. (14). PCR was performed on the nested PCR amplification were 5 -TAGTTTTTATTAT- Ј Ј 50 ng of DNA for each sample using the conditions described for TATTAGTTTTTATTATT-3 (forward primer) and 5 -ACT- Ј the mutation search. PCR products were denatured in formamide CATCCTCACTATCCATATCAT-3 (reverse primer) for the Ј for 5 min at 95°C and then loaded on a 6% denaturing gel on the first PCR, and 5 -AGTTTTTATTATTATGAGTTTTTATTA- Ј Ј Applied Biosystems 373 DNA sequencer. LOH was analyzed AAT-3 (forward primer) and 5 -CCATATCATCCAAC- Ј using the Applied Biosystems Prism Genescan and the Applied CCCACATA-3 (reverse primer) for the second PCR. PCR was ␮ ϫ Biosystems PRISM GENETYPER ANALYSIS software (Perkin- performed in 50- l reaction volumes containing 1 PCR buffer II (Perkin-Elmer, Branchburg, NJ), 2 mM MgCl , 0.2 mM of Elmer/Applied Biosystems). Cases were defined as having LOH 2 each deoxynucleoside triphosphate, 10 pmol of each primer, 1 when an allele peak signal from tumor DNA was reduced by 50% unit of AmpliTaq Gold polymerase (Perkin-Elmer), and 50 ng compared with the normal counterpart. bisulfite-modified DNA. PCR cycling conditions were 10 min at Mutation Screening of the WWOX Gene. Nine pairs of 95°C, 35 cycles of a 30-s denaturation at 94°C, a 30-s annealing primers for individual exon amplification and screening were at 55°C, a 60-s extension at 72°C, and a final extension step of used for PCR screening with genomic DNA. The primers for 5 min at 72°C. PCR products were subcloned into the pGEM-T each exon are specified in GenBank (accession numbers Easy vector (Promega), and six clones were sequenced. AF325423–325432). The PCRs were performed with the same Real-Time Reverse Transcription PCR. Real-time re- conditions as those described for the LOH analysis. The PCR verse transcription PCR was performed using primers designed products were purified using the QIAquick PCR purification kit to span intron 6. The forward primer on exon 6 was 5Ј-CGTG- (Qiagen, Inc., Valencia, CA), and sequencing reactions and CAGCATTTTGCTGAAG-3Ј and the reverse primer on exon 7 analysis were performed using the Applied Biosystems Prism was 5Ј-AGTTGCTGCGTTGCACACA-3Ј. The probe sequence BigDye terminator reaction chemistry on a Perkin-Elmer Gene 5Ј-AGGCCAAGAATGTGCCTCTTCATGTGC-3Ј was labeled Amp PCR system 9600 and the Applied Biosystems Prism 377 with the reporter dye VIC. 6-Carboxytetramethylrhodamine was DNA sequencing system (Applied Biosystems, Inc., Foster used as the quencher. ␤-Actin was used as an internal control. City, CA). Thirty-five cycles were run with an annealing temperature of Nested Real-Time Reverse Transcription PCR Analysis 62°C and an elongation time of 30 s. All of the reactions were of the WWOX Transcripts. cDNA was synthesized from 2 performed in an ABI PRISM 7000 Sequence Detection System ␮g of total RNA and real-time reverse transcription PCR was (Applied Biosystems). Each sample was performed in triplicate. performed as described previously (13). Glyceraldehyde-3- After normalization of the samples with the respective ␤-actin phosphate dehydrogenase amplification served as a control for control, a t test was used to compare the quantitation of WWOX cDNA quality. One ␮l of cDNA was used for the first PCR expression. All of the analyses were performed with the Statis- amplification with WWOX-specific primers (forward primer, tical Package for the Social Sciences for Windows (SPSS, Inc., 5Ј-AGTTCCTGAGCGAGTGGACC-3Ј; reverse primer, 5Ј- Release 10.0.7, 2000). The t tests were considered statistically TTACTTTCAAACAGGCCACCAC-3Ј) in a volume of 50 ␮l significant if P Ͻ 0.05. ؅ containing 20 pmol of each primer, 2.5 mM MgCl2, 1.5 mM 5-Aza-2 -Deoxycytidine (5-AZAC) Treatment. To de- deoxynucleoside triphosphate mix, 1ϫ PCR buffer, and 2 units termine whether treatment of a methylated cell line with a of AmpliTaq Gold (Perkin-Elmer). PCR cycles included one demethylating agent induced WWOX expression, we treated the cycle at 95°C for 8 min followed by 30 cycles at 94°C for 30 s, pancreatic cancer cell line Hs766T with 5-AZAC. Cells were 57°C for 30 s, and 72°C for 1 min, with a final extension step plated and incubated for 3 days with 1 ␮M of 5-AZAC. The at 72°C for 5 min in a Perkin-Elmer Gene Amp PCR system medium and the drug were replaced every 24 h. RNA obtained 9600. One ␮l of the first PCR amplification product was used from wild-type and 5-AZAC-treated cells was used to make for a second PCR amplification with WWOX-nested primers cDNA for real-time reverse transcription PCR analysis. Primers (forward primer, 5Ј-AGGTGCCTCCACAGTC-3Ј; reverse and probe spanning intron 8 were used as described elsewhere primer, 5Ј-GTGTGTGCCCATCCGCTCT-3Ј) under the same (11). Analysis of data was performed as described previously. conditions as those of the first PCR. The amplified products Western Blot. Proteins were extracted from all 9 of the were analyzed by electrophoresis on a 1.5% agarose gel. DNA pancreatic cancer cell lines by lysing cells in a Triton extraction bands corresponding to the normal- and abnormal-sized WWOX buffer containing 10 mM Tris-Hcl (pH7.4), 5 mM NaCl, 5 mM transcripts were excised from the gel, purified using the QIA- EDTA, 10% glycerol, 1% Triton X-100, 1 mM phenylmethyl- quick gel extraction kit (Qiagen), and sequenced on the Applied sulfonyl fluoride, 1 ␮g/ml, and 1 mM sodium orthovanadate.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. Clinical Cancer Research 2461

Table 1 Summary of alteration to WWOX Cell line Reverse or case no. Allelic loss Methylationa transcription-PCR Aberrant productb Panc1 Exon 7-8 homozygous Ϫ37 Absence — deletion MiaPaca2 — None Aberrant Exon 1*-6* Deletion Capan2 — None Aberrant ϩ N Exon 1*-8* Deletion Hs766T — Ϫ37 Aberrant Exon 1*-8* Deletion

1 Retention Ϫ37, Ϫ43, Ϫ53 Normal — 2 Retention None Absence — 3 LOHc None Normal — 5 LOH Ϫ37 to Ϫ148 Normal — 6 Retention None Aberrant ϩ normal Exon 2-8 deletion 7 LOH None Normal — 14 LOH None Normal — a Cytosine position of methylated CpG site relative to the ATG start codon. b Represents a partial deletion of the exon. c LOH, loss of heterozygosity.

Concentration of proteins in the lysates was measured using the equimolar quantity of pMV-7/WWOX according to the manufac- Bio-Rad protein assay reagent (Bio-Rad Inc., Richmond, CA) turer’s protocol. G-418 was added to the medium at a concentration according to the manufacturer’s protocol. Normal human pan- of 400 ␮g/ml. Colonies were stained with 500 ␮g of crystal violet creas tissue lysate was obtained from ProSci, Inc. (Poway, CA). in 20% methanol and a colony count was performed. Equal amounts of protein (20 ␮g) were boiled in Laemmli WWOX Transfection and Cell Cycle Analysis. Lipo- sample buffer, separated in a 4–20% SDS-PAGE gel (Ready fectamine was used to transfect equimolar quantities of the Gels; Bio-Rad, Inc.), and transferred to nitrocellulose mem- empty pMV-7 vector and pMV-7/WWOX into 105 cells of branes (Bio-Rad, Inc.). Filters were blocked in Tris-buffered AsPc1 cells per well of a six-well plate. Cells were collected saline with 5% skim milk for 2 h. Primary antibody exposure after 48 h and washed in PBS solution. DNA was stained with was performed for 2 h at room temperature with mouse mono- propidium iodide (50 ␮g/ml) and analyzed with a FACScan clonal anti-WWOX and anti-␤-actin (Amersham Biosciences) flow cytometer (Becton-Dickinson, San Jose, CA) interfaced antibodies. Peroxidase-linked secondary goat antimouse IgG with a Hewlett-Packard computer (Palo Alto, CA). Cell cycle antibodies (Amersham Biosciences) and the enhanced chemilu- data were analyzed with the CELL-FIT program (Becton Dick- minescence Western Blotting Detection kit (Amersham Bio- inson). sciences) were used for detection. Monoclonal antibodies against the human WWOX protein were elicited (16). Briefly, we generated a glutathione S-transferase/WWOX fusion protein RESULTS containing the first 171 amino acids of the WWOX NH2 termi- LOH and Mutation Analysis. We searched for LOH nus fused in frame to the glutathione S-transferase protein. The using four microsatellite markers within WWOX. The loci WWOX insert was obtained by reverse transcription-PCR from D16S3029, D16S3096, and D16S504 are located in intron 8, normal esophagus as described. The amplification was per- and D16S518 is located in exon 1 (13). All 15 of the cases were formed with 5Ј-GATGGCAGCGCTGCGCTACGCGGGG-3Ј informative for at least one of the loci examined, and in 4 cases as a forward primer and 5Ј-TCATCA TTCTTCTAAAAT- (27%) we detected LOH (Table 1). Representative results of GCGTGACACTC-3Ј as a reverse primer. Subsequently, the LOH analysis are shown in Fig. 1A. To determine whether the amplified insert was subcloned into the SmaI restriction site of WWOX gene was the target of functional inactivation in pan- pGEX-2TK; the pGEX-2TK-WWOX plasmid then was se- creatic adenocarcinoma, we screened for somatic mutations in quenced to exclude random mutations in the insert. Clones the WWOX gene by direct sequencing of PCR products of all of producing the WWOX-specific antibodies were isolated by test- the WWOX exons in tumor cells, in normal counterparts for each ing their supernatants on Western blot strips prepared with total cancer, and in 9 human pancreatic cancer cell lines. In the Panc1 proteins purified from either wild-type or pMV-WWOX tran- cancer cell line, exons 7 and 8 were not amplified, indicating a siently transfected HEK293 cells. WWOX monoclonal antibod- homozygous deletion. We confirmed this homozygous deletion ies were used at 1:1000. of exons 7 and 8 in Panc1 by duplex PCR (data not shown). All WWOX Transfection and Colony Assay. WWOX of the exons were amplified in the other 8 cell lines and in 15 cDNA was cloned in sense orientation into pMV-7, a eukaryotic paired samples. No somatic point mutations were identified in expression vector (17). This vector has a neomycin gene that the cell lines and clinical samples. provides G-418 resistance for the selection of transfectants and Methylation Analysis of the WWOX Gene Promoter a long terminal repeat promoter that controls the expression of Region. To evaluate the methylation status of the promoter the transgene. Lipofectamine (Invitrogen) was used to transfect region of WWOX in pancreatic adenocarcinoma, we used a AsPc1 and Panc1 cells with 5 ␮g of empty vector and an sodium bisulfite sequence approach. Two of 9 pancreatic cell

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. 2462 WWOX Alterations in Pancreatic Carcinoma

Fig. 1 Analysis of WWOX gene in pancreatic cancer. A, representative loss of heterozygosity analysis of case 3 at the D16S518 locus. The microsatellite locus is identified below the chromatogram. T, tumor; N, corresponding normal tissue. Arrow indicates loss of heterozygosity at D16S518. B, direct sequencing chromatogram of PCR products from Hs766T. The CpG sites are indicated by a bar and the complementary strand at methylated cytosine are indicated by an asterisk. C, reverse transcription PCR analysis of WWOX in pancreatic cancer cell lines. Arrow indicates location of normal transcripts and arrowheads indicate location of aberrant transcripts. Glyceraldehyde-3-phosphate dehydrogenase amplification served as a control for cDNA quality. NC, negative control. D, representative reverse transcription PCR analysis from two cases, 2 and 6. Case 2 shows the absence of WWOX transcript in tumor sample. Case 6 shows the aberrant transcript with normal transcript of WWOX in tumor sample.

lines (22%; Panc1 and Hs766T) showed methylated CpG site showed methylation at all of the cytosines from Ϫ148 to Ϫ37 in only at the cytosine Ϫ37 bp relative to the ATG start codon of six of six clones (Table 1). 55 CpG sites (Table 1). The result of bisulfite sequence analysis 5-AZAC Treatment. After treatment with 5-AZAC, of Hs766T is shown in Fig. 1B. In the PCR product from the Hs766T cell lines demonstrated an increase in expression of specific primers for bisulfite-treated DNA, methylated cytosine WWOX by real-time reverse transcription-PCR. The WWOX: appears as a G signal in the complementary strand at the CpG actin ratio increased from a mean of 0.34 to 0.44. This differ- dinucleotide only at the Ϫ37 position. Other unmethylated cy- ence was statistically different and represents a 30% increase in tosine nucleotides, including those in the CpG dinucleotides, expression. changed to thymines because of bisulfite modification. This Analysis of the WWOX Transcripts. The mRNA ex- site-specific methylation of a single CpG site of Panc1 and pression of WWOX was analyzed in 9 pancreatic cancer cell Hs766T was detected in six of six and five of six clones, lines and in 15 clinical samples by reverse transcription-PCR. respectively. In primary tumors, case 1 showed methylation at One cell line (Panc1) showed no WWOX transcript. Three cell cytosines Ϫ37, Ϫ43, and Ϫ53 in six of six clones. Case 5 lines (MiaPaca2, Capan2, and Hs766T) exhibited aberrant

Fig. 2 Real-time reverse transcription-PCR analysis of WWOX expression relative to ␤-actin in human pancreatic cancer cell lines. Analysis was performed with the Statistical Package for the Social Sciences for Windows (SPSS, Inc, Release 10.0.7, 2000). The t tests were considered to be statistically significant if P Ͻ 0.05.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. Clinical Cancer Research 2463

Fig. 3 Western blot demon- strating the decreased expres- sion of WWOX in human pan- creatic cancer cell lines. Pro- tein loading normalized with ␤-actin; bars, ϮSD.

transcripts, and Capan2 showed an aberrant band with a sion in the 9 pancreatic cancer cell lines and in the normal normal-sized transcript (Fig. 1C). Sequence analysis of ab- pancreatic tissue control. Although Panc1 showed no expression errant transcripts showed frameshifts in the open reading of the WWOX gene, the remaining 8 cell lines showed a reduc- frame (Table 1). Of 15 primary tumors, 1 case (case 6) tion in expression. Subsequently, we assessed the expression of showed an aberrant transcript plus a normal-sized WWOX WWOX in 15 primary pancreatic adenocarcinomas. Six cases reverse transcription-PCR product (Fig. 1D), and sequence (cases 2, 6, 9, 10, 13, and 14; 40%) showed a reduction in analysis of the aberrant transcript showed deletion of exons expression of WWOX. 2– 8. In 1 case (case 2), no WWOX transcript was detected Western Blot. Fig. 3 shows the results of a Western blot (Table 1). The remaining primary tumors showed only a analysis of the pancreatic cancer cell lines. Although Panc1 normal-sized transcript. showed no expression of WWOX, the other 8 cell lines showed Real-Time Reverse Transcription-PCR Analysis. Fig. a clear reduction in WWOX expression compared with normal 2 summarizes the results of the comparison of WWOX expres- pancreas tissue lysate.

Fig. 4 A, colony assay demonstrating decreased colony formation after WWOX transfection. B, quantification of decreased colony formation by colony count; bars, ϮSD.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. 2464 WWOX Alterations in Pancreatic Carcinoma

Fig. 5 Fluorescence-activated cell sorter analysis after pro- pidium iodide staining demon- strating cell cycle changes after transfection with WWOX. In-

creased apoptotic (Sub G0/G1) fraction is seen in Aspc1 cell lines after WWOX transfection.

Colony Assay. Fig. 4 illustrates the results of the colony 5-AZAC elevated WWOX expression significantly. In clinical assay. Transfection with the pMV-7/WWOX vector decreased samples, 2 cases showed promoter hypermethylation within the the number of colonies formed compared with the empty promoter region including site-specific methylation at Ϫ37 po- pMV-7 vector. This effect is clear in AsPc1 cells and is dra- sition. Pogribny et al. (21) demonstrated that site-specific meth- matically illustrated in Panc1 cells. ylation within the promoter region was associated with Fluorescence-Activated Cell Sorter Analysis. Signifi- transcriptional inactivation of the p53. These findings suggest cantly greater fractions of transfected cells were found in the that site-specific methylation of the WWOX may be involved in sub-G1 phase by fluorescence-activated cell sorter analysis after pancreatic carcinogenesis. transfection with pMV-7/WWOX compared with the empty Transfection of WWOX into pancreatic cancer cell lines pMV-7 vector (Fig. 5), strongly suggesting a role for WWOX in resulted in apoptosis. Chang et al. (22) reported that murine triggering apoptosis of malignant pancreatic cells. WWOX (WOX1) is a mitochondrial apoptotic protein and an essential partner of p53 in apoptosis. Furthermore, to evaluate DISCUSSION the potential of WWOX as a tumor suppressor gene, we per- In this study, we demonstrated that: (a) WWOX is altered formed a colony-forming assay using tumor-derived cell lines. by deletion and/or aberrant expression in 4 of 9 pancreatic WWOX inhibited the colony formation of pancreatic cancer cell cancer cell lines (44%) and 6 of 15 primary pancreatic adeno- lines. Thus, we conclude that WWOX plays a role in pancreatic carcinomas (40%); (b) promoter hypermethylation of WWOX, carcinogenesis by affecting apoptosis. WWOX may represent a including Ϫ37 position site-specific methylation, is detected in new target for the development of a gene therapy approach to 2 cell lines (22%) and in 2 samples (13%), and treatment with the treatment of pancreatic cancer. the demethylating agent 5-AZAC elevated significantly WWOX expression in Hs766T; (c) all of the cell lines showed low levels of WWOX expression using real-time reverse transcription-PCR REFERENCES and Western blot, and 6 primary cases (40%) showed a statis- 1. Warshaw AL, Fernandez-del Castillo C. Pancreatic carcinoma. tically significant reduction in WWOX expression; and (d) trans- N Engl J Med 1992;326:455–65. fection with WWOX induced apoptosis and suppressed colony 2. Conlon KC, Klimstra DS, Brennan MF. Long-term survival after formation in cell lines. curative resection for pancreatic ductal adenocarcinoma Clinicopatho- logic analysis of 5-year survivors Ann Surg 1996;223:273–9. The WWOX gene was identified recently as a tumor sup- 3. Ohta M, Inoue H, Cotticelli MG, et al. The FHIT gene, spanning pressor gene at 16q23.3–24.1, a chromosome region that spans the chromosome 3p14.2 fragile site and renal carcinoma-associated the common fragile site FRA16D (8–11). Several studies have t(3;8) breakpoint, is abnormal in digestive tract cancers. Cell 1996; revealed alterations of WWOX in several types of human cancers 84:587–97. (8–11, 14). The finding of aberrant WWOX transcripts is con- 4. Mori M, Mimori K, Shiraishi T, et al. Altered expression of Fhit in sistent with the pattern of WWOX alterations described in recent carcinoma and precarcinomatous lesions of the esophagus. Cancer Res studies (10, 13). We found that most aberrant transcripts in 2000;60:1177–82. esophageal and lung cancers lacked exons 5, 6, 7, and 8 (13, 18). 5. Croce CM, Sozzi G, Huebner K. Role of FHIT in human cancer. J Clin Oncol 1999;17:1618–24. These exons encode the oxidoreductase domain of the WWOX 6. Huebner K, Croce CM. FRA3B and other common fragile sites: the protein (19), and their absence may cause loss of function of weakest links. Nat Rev Cancer 2001;1:214–21. WWOX. In this study, all of the aberrant transcripts showed 7. Dumon KR, Ishii H, Vecchione A, et al. Fragile histidine triad absence of these exons. expression delays tumor development and induces apoptosis in human Recent studies have reported that promoter hypermethyla- pancreatic cancer. Cancer Res 2001;61:4827–36. tion of several tumor suppressor genes may be an important 8. Bednarek AK, Laflin KJ, Daniel RL, Liao Q, Hawkins KA, Aldaz mechanism for gene inactivation in several types of cancer (15, CM. WWOX, a novel WW domain-containing protein mapping to hu- Ϫ man chromosome 16q23.3–24.1, a region frequently affected in breast 20). Interestingly, we detected 37 position site-specific meth- cancer. Cancer Res 2000;60:2140–5. ylation in pancreatic cancer cell lines. In 2 cell lines, the position 9. Paige AJW, Taylor KJ, Stewart A, et al. A 700-kb physical map of a of the methylated site was the same as the CpG island. Further- region of 16q23.2 homozygously deleted in multiple cancers and spanning more, treatment of Hs766T cells with the demethylating agent the common fragile site. FRA16D. Cancer Res 2000;60:1690–7.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. Clinical Cancer Research 2465

10. Paige AJW, Taylor KJ, Taylor C, et al. WWOX: a candidate tumor 17. Trapasso F, Iuliano R, Boccia A, et al. Rat protein tyrosine phos- suppressor gene involved in multiple tumor types. Proc Natl Acad Sci phatase ␩ suppresses the neoplastic phenotype of retrovirally trans- USA 2001;98:11417–22. formed thyroid cells through the stabilization of p27(Kip1). Mol Cell 11. Bednarek AK, Keck-Waggoner CL, Daniel RL, et al. WWOX, the Biol 2000;20:9236–46. FRA16D gene, behaves as a suppressor of tumor growth. Cancer Res 18. Yendamuri S, Kuroki T, Trapasso F, et al. WW domain containing 2001;61:8068–73. oxidoreductase gene expression is altered in non-small cell lung cancer. Cancer Res 2003;63:878–81. 12. Mori M, Shiraishi T, Tanaka S, et al. Lack of DMBT1 expression in esophageal, gastric and colon cancer. Br J Cancer 1999;79:211–13. 19. Ried K, Finnis M, Hobson L, et al. Common chromosomal fragile site FRA16D sequence: identification of the FOR gene spanning 13. Kuroki T, Trapasso F, Shiraishi T, et al. Genetic alterations of the FRA16D and homozygous deletions and translocation breakpoints in tumor suppressor gene WWOX in esophageal squamous cell carcinoma. cancer cells. Hum Mol Genet 2000;9:1651–63. Cancer Res 2002;62:2258–2260. 20. Esteller M. CpG island hypermethylation and tumor suppressor 14. Ishii H, Baffa R, Numata S, Murakumo et al. The FEZ1 gene at genes: a booming present, a brighter future. Oncogene 2002;21: chromosome 8p22 encodes a leucine-zipper protein, and its expression 5427–40. is altered in multiple human tumors. Proc Natl Acad Sci USA 1999;96: 21. Pogribny IP, James SJ. Reduction of p53 gene expression in human 3928–33. primary hepatocellular carcinoma is associated with promoter region 15. Kuroki T, Trapasso F, Yendamuri S, et al. Promoter hypermethy- methylation without coding region mutation. Cancer Lett 2002;176: lation of RASSF1A in esophageal squamous cell carcinoma. Clin Can- 169–74. cer Res 2003;9:1441–5. 22. Chang N-S, Pratt N, Heath J, et al. Hyaluronidase induction of a 16. Galfre G, Milstein C. Preparation of monoclonal antibodies: strat- WW domain-containing oxidoreductase that enhances tumor necrosis egies and procedures. Methods Enzymol 1981;73:3–46. factor cytotoxicity. J Biol Chem 2001;276:3361–70.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research. The Tumor Suppressor Gene WWOX at FRA16D Is Involved in Pancreatic Carcinogenesis

Tamotsu Kuroki, Sai Yendamuri, Francesco Trapasso, et al.

Clin Cancer Res 2004;10:2459-2465.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/10/7/2459

Cited articles This article cites 22 articles, 13 of which you can access for free at: http://clincancerres.aacrjournals.org/content/10/7/2459.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/10/7/2459.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/10/7/2459. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2004 American Association for Cancer Research.