Secreted Frizzled-Related Protein 4 Is Silenced by Hypermethylation and Induces Apoptosis in B-Catenin – Deficient Human Mesothelioma Cells

Research Article

Biao He,1 Amie Y. Lee,1 Sina Dadfarmay,1 Liang You,1 Zhidong Xu,1 Noemi Reguart,1,2 Julien Mazieres,1,3 Iwao Mikami,1 Frank McCormick,1 and David M. Jablons1

1Thoracic Oncology Laboratory, Department of Surgery, Comprehensive Cancer Center, University of California, San Francisco, California; 2Medical Oncology Service, Institut Catala` d’Oncologia, Hospital Germans Trias i Pujol, Barcelona, Spain; and 3Department Innovation Therapeutique et Oncologie Moleculaire, Institut National de la Sante et de la Recherche Medicale U563, Institut Claudius Regaud, Toulouse, France

Abstract acute lymphoblastic leukemia (7), and renal clear cell carcinoma The secreted frizzled-related proteins (SFRPs) function as (8). Forced expression of Dkk-1 in brain tumor (9) and cervical negative regulators of Wnt signaling and have important cancer (10), or Dkk-3 in non–small cell lung cancer (4) and implications in tumorigenesis. Frequent promoter hyper- osteosarcoma (11), inhibits cell growth. Down-regulation of SFRPs methylation of SFRPs has been identified in human cancer. was found in several cancers, including mesothelioma (12), breast Restoration of SFRP function attenuates Wnt signaling and cancer (13), cervical cancer (14), and gastric cancer (15, 16). Loss induces apoptosis in a variety of cancer types. Wnt signaling of SFRP family expression was also found to be associated with is known to inhibit apoptosis through activation of BB-catenin/ promoter hypermethylation in mesothelioma and colorectal Tcf–mediated transcription. Recently, we identified aberrant cancer tissue samples (12, 16–18). Furthermore, restoration of Wnt activation as a result of Dishevelled overexpression in SFRPs in colon cancer cell lines carrying downstream mutations h malignant mesothelioma. Here, we report that silencing suppressed Wnt/ -catenin-dependent transcription and induced of SFRP4 is correlated with promoter hypermethylation in apoptosis, suggesting that Wnt signaling may be regulated in a BB-catenin–deficient mesothelioma cell lines. Reexpression of quantitative manner at different levels (18). However, there are few h SFRP4 in these BB-catenin–deficient mesothelioma cell lines reports that have linked -catenin-independent pathway(s) to the blocks Wnt signaling, induces apoptosis, and suppresses Wnt-dependent apoptotic inhibition (19). Here, we report that h growth. Conversely, knocking down SFRP4 by small interfer- SFRP4 is methylation silenced in -catenin-deficient mesothelio- ing RNA in cell lines expressing both SFRP4 and BB-catenin ma cell lines. Furthermore, we show that reexpressing SFRP4 in stimulates Wnt signaling, promotes cell growth, and inhibits these cell lines down-regulates Dishevelled (Dvl), induces apopto- h chemodrug-induced apoptosis. Our results suggest that sis, and suppresses cell growth, suggesting that -catenin- methylation silencing of SFRP4 may play an important role independent pathway(s) may be important for the apoptotic in aberrant Wnt activation in mesothelioma even in the inhibition caused by Wnt activation. absence of BB-catenin. Our data also suggest that BB-catenin– independent noncanonical pathway(s) may be involved in the Materials and Methods apoptotic inhibition caused by activation of Wnt signaling. Cell Lines. The human mesothelioma cell lines H28 and H2052 were (Cancer Res 2005; 65(3): 743-8) obtained from American Type Culture Collection (Manassas, VA). The human mesothelioma cell line MS-1 was obtained from NIH (Bethesda, Introduction MD). These cell lines were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, penicillin (100 IU/mL), and streptomycin (100 Ag/mL). The activation of the canonical Wnt pathway is mediated by Normal human small airway epithelial cells (primary culture) were h -catenin (1, 2). However, noncanonical pathways act indepen- obtained from Clonetics (Walkersville, MD) and cultured in Clonetics h dently of -catenin (1, 2). At least two classes of Wnt antagonists SAGM Bullet kit. All cells were cultured at 37jC in a humid incubator with have been reported (1). The first class, including secreted Frizzled- 5% CO2. related protein (SFRP) family and Wnt inhibitory factor 1, binds Semiquantitative Reverse Transcription-PCR. Total RNA from cell directly to Wnt ligands. The second class, including the Dickkopf lines was isolated using Qiagen RNeasy Mini Kit (Valencia, CA). Reverse (Dkk) family, binds to low-density lipoprotein receptor–related transcription-PCR (RT-PCR) was done in GeneAmp PCR system 2700 protein 5/6. Recently, both SFRP and Dkk families have been (Applied Biosystems, Foster City, CA) using One-Step RT-PCR kit from implicated in oncogenesis. For example, the production of Dkk-1 Invitrogen Life Technologies (Carlsbad, CA). Primers for RT-PCR were by myeloma cells is associated with the presence of lytic bone obtained from Operon Technologies, Inc. (Alameda, CA). Primer sequences lesions in patients with multiple myeloma (3). Dkk-3 has been for amplifying the human SFRP4 were described previously (17). Glyceraldehyde-3-phosphate dehydrogenase was used as internal found silenced by methylation in non–small cell lung cancer (4–6), control. Methylation and Sequencing Analysis. Genomic DNA from cell lines was extracted using DNA STAT-60 reagent (TEL-TEST, Inc., Friendswood, TX). Bisulfite modification of genomic DNA was done by using EZ DNA Requests for reprints: David M. Jablons, Thoracic Oncology Laboratory, methylation kit (Zymo Research, Orange, CA). Bisulfite-treated genomic Department of Surgery, Comprehensive Cancer Center, University of California, DNA was amplified using primers described previously for human SFRP4 1600 Divisadero Street, C322C, Campus Box 1674, San Francisco, CA 94115. Phone: 415-353-7502; Fax: 415-502-3179; E-mail: [email protected]. (17). The amplified 230-bp product corresponds to À384 to À154 in the #2005 American Association for Cancer Research. SFRP4 promoter region (the start codon ATG of SFRP4 is defined as 0). www.aacrjournals.org 743 Cancer Res 2005; 65: (3). February 1, 2005

5-Aza-2V-deoxycytidine (Sigma, St. Louis, MO) treatment was done as Cruz, CA). Anti-caspase-3 antibody was purchased from Oncogene (Cam- described previously (17). For amplification of SFRP4 genomic region, bridge, MA). Anti-c-Myc antibody was purchased from Cell Signaling primers were designed according to the SFRP4 genomic sequence in Technology (Beverly, MA). Anti-h-actin antibody was purchased from chromosome 7. Untreated genomic DNA isolated from the cell lines were Sigma-Aldrich Corp. (St. Louis, MO). Anti-h-catenin antibody was used as template. The primer sequences and the length of fragments purchased from Transduction Laboratories (Lexington, KY). For detecting h amplified are listed in Table 1. The PCR products were extracted from alteration of -catenin, cytosolic extracts were prepared and examined as the agarose gel using Qiagen QIAquick Gel Extraction Kit and were described previously (20). RNA Interference. RNA interference experiments were done as subsequently sequenced at the DNA Sequencing Core Facility, Compre- described previously (19). The ion-exchange high-performance liquid hensive Cancer Center, University of California (San Francisco, CA). chromatography–purified small interfering RNA (siRNA; SFRP4 siRNA Western Blotting. Standard protocol was used. Anti-Dvl-3 and anti- and nonsilencing siRNA control, >97% pure) were purchased from Qiagen- Survivin antibodies were obtained from Santa Cruz Biotechnology (Santa Xeragon (Germantown, MD). The targeted sequence of SFRP4 siRNA is 5V- AAGTCCCGCTCATTACAAATT-3V, corresponding to +701 to +721 of the human SFRP4 cDNA sequence (the start codon ATG is defined as +1). To Table 1. Primers used for amplifying SFRP4 analyze proliferation, 3 Â 104 cells were transfected with siRNA in a 24-well genomic region plate. After transfection, viable cells (trypan blue exclusion) were collected by trypsinization and counted at various time points. Experiments were Primer Sequence Size (bp) done in triplicate. Apoptosis Analysis. Cells were harvested by trypsinization and stained 1F:5V-ATGTTCCTCTCCATCCTAGT-3V 350 using an Annexin V FITC Apoptosis Detection Kit (Oncogene) according to R: 5V-TGGTTGTACATCTTCATGAG-3V the manufacturer’s protocol. Then, stained cells were immediately analyzed 2F:5V-CTCATGAAGATGTACAACCA-3V 420 by flow cytometry (FACScan, Becton Dickinson, Franklin Lakes, NJ). R: 5V-TAGCAACACCTGTTTACTGC-3V Alimta (LY231514, multitargeted antifolate, pemetrexed) was supplied by 3F:5V-GGGAGCCTTCGATGATGTTT-3V 470 Eli Lilly (Indianapolis, IN). Alimta was diluted in sterile physiologic solution R: 5V-GGTAACCTGGGCCAGATAAA-3V at a concentration of 10 mg/mL. The solution was divided into aliquots, 4F:5V-TTATCCAGGCTTTATCTGGC-3V 430 stored at À80jC, and diluted in culture medium before each experiment. R: 5V-CCAAGCACAAATCGATGTGA-3V Transient Transfection and Colony Formation Assay. For transient 5F:5V-TCACATCGATTTGTGCTTGG-3V 400 transfection experiments, cells (2 Â 105) were plated in six-well plates 24 V V R: 5-TAATCCCAGCAGTTTGGGAA-3 hours before transfection. LipofectAMINE 2000 (Invitrogen Life Technolo- V V 6F:5-TTCCCAAACTGCTGGGATTA-3 390 gies) was used to mediate transfection using 5.0 Ag SFRP4 cDNA construct R: 5V-CGTTGCCAAAGTTGGCTTCA-3V in pCDNA3 vector (kindly provided by Dr. Amir Rattner) or 5.0 Ag empty 7F:5V-TCGGTGCAAGTGTAAAAAGG-3V 490 pCDNA3 vector as control according to the manufacturer’s protocol. R: 5V-TAATTGTCTGGGCCTATGGG-3V Transfected cells were striped and plated on 10 cm cell culture dishes at 48 8F:5V-AGAATTAAGGCCCATAGGCC-3V 350 hours after transfection. The cells were then selected by G418 (400 Ag/mL). R: 5V-AAGTAGGCAGCCCAAACAGA-3V Colonies were stained with 0.5% methylene blue and were counted 3 weeks 9F:5V-TCTGTTTGGGCTGCCTACTT-3V 450 after the transfection. R: 5V-ATCAAGGTCGTTCAAAGGCC-3V Statistical Analysis. Data are mean F SD. The Student’s t test was used 10 F: 5V-GTCATTCTCTGGGGTGTTAT-3V 370 for comparing activities of different constructs and treatments. R: 5V-CTGATGCTTGAGCTTCTTCT-3V 11 F: 5V-AGAAGAAGCTCAAGCATCAG-3V 480 R: 5V-TTTACATCCACCACCGTTGT-3V 12 F: 5V-ACAACGGTGGTGGATGTAAA-3V 480 Results and Discussion R: 5V-CAGTATGTGCCACTCTAAAA-3V SFRP4 Is Silenced by Promoter Hypermethylation in 13 F: 5V-TTTTAGAGTGGCACATACTG-3V 470 BB-Catenin–Deficient Human Mesothelioma Cell Lines. Previ- R: 5V-GAGCATTAACAGCAGAGAAT-3V ously, we showed that human mesothelioma cell lines H28 and MS-1 14 F: 5V-GCCCCATAAGTTGATGCAGT-3V 390 lack cytosolic h-catenin protein and have no Tcf/Lef transcriptional R: 5V-CCATTCATCTCTCCAGATAG-3V 15 F: 5V-CTATCTGGAGAGATGAATGG-3V 550 activity (19). To further assess the role of upstream Wnt signaling in R: 5V-TTATGCAGTCAGATATGTGT-3V these cells, we examined SFRP4 expression (Fig 1A). We found that 16 F: 5V-ACACATATCTGACTGCATAA-3V 460 the SFRP4 transcript was missing in both H28 and MS-1 cell lines. As R: 5V-AACCCTGTATACAGTAGTGT-3V a control, we found that SFRP4 was expressed in a normal primary 17 F: 5V-AAGCCAGTGCACACTACTGT-3V 330 cell culture small airway epithelial cell and a mesothelioma cell line R: 5V-TCCTTTACCCCATAATCCAC-3V H2052 (Fig. 1A). To investigate cause of the missing SFRP4 transcript 18 F: 5V-TGGTGGATGTGTGGATTATG-3V 380 in these two h-catenin-deficient cell lines, we first examined the V V R: 5-GTCGAATGATAGATTCCTGG-3 possibility of deletion or rearrangement of the gene by genomic 19 F: 5V-AGGCCGTTAAGAGTGTAGCT-3V 560 sequencing (Fig. 1B). We were able to amplify fragments covering the R: 5V-GAGAGCATAGGCAATAGCAA-3V 20 F: 5V-TTGCTATTGCCTATGCTCTC-3V 470 whole genomic region of SFRP4 gene (including both exons and R: 5V-CTGTGCTATGACACACTTCT-3V introns) from both cell lines H28 and MS-1 (data not shown). 21 F: 5V-AGAAGTGTGTCATAGCACAG-3V 450 Sequencing of these fragments showed no molecular abnormalities, R: 5V-CCCCATAATGCAACATTCCT-3V such as deletions, mutations, or rearrangements, when compared 22 F: 5V-AGGAATGTTGCATTATGGGG-3V 360 with the wild-type genomic sequence (chromosome 7; Fig. 1B). R: 5V-TATCTCACCCCTTTATGGTG-3V Aberrant promoter hypermethylation has been found to be an 23 F: 5V-CCAAGTCACCTGGTACATGA-3V 370 important mechanism in the inactivation of tumor suppressor R: 5V-CCGCTTTGGAAACTAGTTAG-3V genes in cancer (21). Therefore, we next analyzed the methylation status of the CpG islands in the SFRP4 promoter

Cancer Res 2005; 65: (3). February 1, 2005 744 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2005 American Association for Cancer Research. SFRP4 in bb-Catenin–Deficient Mesothelioma region in these cell lines. Both cell lines, H28 and MS-1, lacking Table 2. SFRP4 methylation silencing and cytosolic SFRP4 expression were found to be hypermethylated (Fig. 1C). In h-catenin status in human cell lines contrast, no hypermethylation was seen in the control cells, small airway epithelial cell and H2052, that express SFRP4.In Cell lines SFRP4 Cytosolic addition, we confirmed that the SFRP4 expression was restored h-catenin after the demethylating agent, 5-aza-2V-deoxycytidine, treatment Expression Methylation in MS-1 and H28 cells (Fig. 1D). Therefore, our results indicate that the SFRP4 expression is silenced in h-catenin-deficient Normal mesothelioma cell lines and the silencing correlates with dense SAEC + À + CpG methylation of the SFRP4 promoter. Mesothelioma To further investigate the relationship between loss of SFRP4 MS-1 À + À expression and h-catenin level, we compared the expression and H28 À + À methylation status of SFRP4 with h-catenin levels in a variety H2052 + À + H513 + À + of cell lines (Table 2). In all the cancer cell lines expressing h H290 À ++ -catenin that we examined, 9 of 18 (50%) showed methylation Non–small cell lung cancer silencing of SFRP4, 3 of 18 (16.7%) showed SFRP4 down- A549 + À + regulation and partial methylation, and 6 of 18 (33.3%) expressed H460 À ++ SFRP4 and showed no methylation. This result indicates that H838 À ++ silencing of SFRP4 also correlates with the promoter methylation H1703 À ++ in cancer cells expressing h-catenin. Taken together, our data Glioma suggest that loss of SFRP4 expression in h-catenin-deficient U87 À ++ Breast cancer mesothelioma cells is not likely a result of loss of h-catenin. BB MCF7 À ++ Restoration of SFRP4 in -Catenin–Deficient Cell Lines Colon cancer Induces Apoptosis and Suppresses Cell Growth. Our finding of RKO* À ++ methylation silencing of SFRP4 in H28 and MS-1 cells raises the HCT116* À ++ possibility that upstream Wnt signaling is still functional in HT29* À ++ apoptotic inhibition in cells lacking h-catenin. To answer this SW480* + + (Partial) + question, we restored the SFRP4 expression in these h-catenin- LoVo* + + (Partial) + deficient cells. Interestingly, restoration of SFRP4 was able to kill CaCo2* + + (Partial) + these cells; this cell killing was largely due to apoptosis induction DLD1* + À + Liver cancer (Fig. 2A). In contrast, when we transfected the SFRP4 construct PLC/PRF/5* + À + into the mesothelioma cell line H2052, where SFRP4 is not SNU182* + À + silenced and the promoter is not methylated, no apoptosis NOTE: Semiquantitative RT-PCR was used to examine SFRP4 expression, methylation-specific PCR was used to examine methylation status, and Western analysis was used to examine cytosolic h-catenin expression. *Refs. (17, 18).

induction was observed (Fig. 2B). The blockade of Wnt signaling by SFRP4 restoration in H28 cells was confirmed by analyzing the expression of Wnt signaling effectors (Fig. 2C). Dvl-3, a Wnt intracellular mediator, as well as Survivin, an apoptosis inhibitor and also a Wnt downstream target gene (22), were both down- regulated after restoration of SFRP4 in H28 cells. The cell line MS- 1 also showed similar apoptotic and downstream effects after restoration of SFRP4 (data not shown). As a control, in H2052 cells, no change in cytosolic h-catenin, Dvl-3, and Survivin was observed after transfection of the SFRP4 construct, consistent with the presence of endogenous SFRP4 in these H2052 cells (Fig. 2C). Furthermore, after selection of drug-resistant colonies for 3 weeks, we found that the colony numbers of SFRP4-transfected cells Figure 1. Correlation of promoter methylation with silencing of SFRP4 in h-catenin-deficient mesothelioma cell lines H28 and MS-1. A, semiquantitative significantly decreased compared with that of empty vector- RT-PCR results for SFRP4 expression. B, sequencing of genomic region of transfected cells (P < 0.001 for both MS-1 and H28; Fig. 3A and B). the SFRP4 gene (both exons and introns). Part of the intron between exons 2 and 3 of the SFRP4 gene from cell line H28. C, bisulfite-sequencing analysis However, in H2052 cells, no significant difference was seen in of the SFRP4 promoter region. Unmethylated (5) and methylated (n) CpG colony numbers between SFRP4 and empty vector-transfected islands in the promoter region. We sequenced three individual clones of PCR cells after drug selection (P = 0.45; Fig. 3C). Taken together, these products amplified from bisulfite-treated genomic DNA for each sample. D, reactivation of SFRP4 expression by 5-aza-2V-deoxycytidine treatment. data suggest that the silencing of SFRP4 in mesothelioma H28 and Semiquantitative RT-PCR was done after 5-aza-2V-deoxycytidine treatment MS-1 cells may be critical for their survival and that h-catenin- (1.0 Amol/L for 96 hours). www.aacrjournals.org 745 Cancer Res 2005; 65: (3). February 1, 2005

Figure 2. Restoration of SFRP4 induces apoptosis in h-catenin-deficient mesothelioma cells. A and B, Annexin V analysis of apoptosis after restoration of SFRP4 in H28 and H2052 cells, respectively. FL1-H (X axis), Annexin V-FITC staining; FL3-H (Y axis), propidium iodide staining. C, analysis of downstream effectors after SFRP4 restoration in H28 and H2052 cells. Semiquantitative RT-PCR was used to examine the SFRP4 expression. Cytosolic proteins were prepared and used in the Western blots. h-actin served as loading control.

independent mechanism(s) might be responsible for the apoptosis the h-catenin level correlated inversely with SFRP4 level in and growth suppression induced by blocking Wnt signaling in endometrial sarcomas (25). Both Survivin and c-Myc have been these cells. h-catenin/Tcf transcription was thought to be required known to promote cell proliferation and inhibit apoptosis. for Wnt-mediated cell survival and it inhibits apoptosis by Consistently, we observed increased growth rate in SFRP4 siRNA- preventing cytochrome c release and subsequent caspase activa- treated H2052 cells (Fig. 4B). To examine whether increased Survivin tion (23, 24). To our knowledge, our report is the first level by SFRP4 siRNA treatment inhibits apoptosis, we treated demonstration that h-catenin is not necessary for SFRP-induced H2052 cells with a chemodrug, Alimta, 1 day after transfection with apoptosis in cancer. control and SFRP4 siRNA. Alimta is a novel multifunctional Knockdown of SFRP4 by siRNA in SFRP4-Expressing Cells antifolate antimetabolite (26) and is now considered a standard in Promotes Cell Growth and Inhibits Chemodrug-Induced the treatment of mesothelioma. After 3 days of Alimta treatment Apoptosis. To further explore the function of SFPR4 in canonical (at a concentration of 1 Ag/mL), we observed significant apoptosis and noncanonical Wnt signaling pathways during tumorigenesis, induction (>30%; P < 0.03) in H2052 cells transfected with control we used siRNA to silence SFRP4 expression in H2052 cells that siRNA but not with SFRP4 siRNA (Fig. 4C). We also observed up- express both SFRP4 and h-catenin (Fig. 4). Interestingly, in H2052 regulation of the cleaved (active) form of caspase-3 (apoptosis cells treated with SFRP4 siRNA, we found increased expression of activator) and down-regulation of Survivin (apoptosis inhibitor) cytosolic h-catenin, Wnt target genes: c-Myc and Survivin (Fig. 4A). after Alimta treatment in control siRNA-transfected H2052 cells Our explanation is that siRNA knockdown of SFPR4 reduces the (Fig. 4D). In contrast, Survivin level was significantly up-regulated Wnt-inhibiting effect of SFRP4 and allows more Wnt molecules to in SFRP4 siRNA-transfected cells. Although Alimta down-regulated bind to its receptor complex, which in turn enhances the canonical the Survivin level, it was still higher than that in control siRNA- pathway. Enhanced activation of the canonical pathway results in transfected cells before Alimta treatment (Fig. 4D), consistent with stabilization of more free h-catenin, which accumulates in the no apoptosis induction and caspase-3 activation by Alimta in these cytoplasm. This result is also supported by a recent observation that SFRP4 siRNA-transfected cells. Taken together, our data show that

Figure 3. Restoration of SFRP4 suppresses the growth of h-catenin-deficient mesothelioma cells. Colony formation assay usingcelllinesMS-1(A),H28(B),andH2052 (C). Columns, average colony numbers in triplicate experiments; bars, SD.

Cancer Res 2005; 65: (3). February 1, 2005 746 www.aacrjournals.org

Wnt-induced apoptotic inhibition has rarely been reported. Recently, we found that blockade of Wnt-1 signaling induces apoptosis partially through the Dvl/c-Jun-NH2-kinase–dependent pathway (19, 28), the role of which in Wnt signaling is not well understood (29). In this study, we show that the restoration of SFRP4 in h-catenin-deficient mesothelioma cell lines induces apoptosis, suppresses cell growth, and down-regulates Wnt signaling. Moreover, down-regulation of SFRP4 in h-catenin-intact mesothelioma cells inhibits drug-induced apoptosis and stimulates cell growth through further enhancing canonical Wnt signaling pathway. Therefore, our data suggest that SFRP4 may induce apoptosis not only through the canonical pathway but also through h-catenin-independent noncanonical pathways, such as c-Jun-NH2- kinase-dependent pathways. Recently, Suzuki et al. (17) reported that SFRPs were preferen- tially hypermethylated in human colorectal cancers carrying downstream mutations and that those colorectal cancer cells retained sensitivity to upstream Wnt signaling (18), suggesting that upstream regulation of Wnt signaling may be actively and importantly involved in tumorigenesis even in colorectal cancer Figure 4. Silencing SFRP4 by siRNA in SFRP4-expressing H2052 cells. carrying downstream activating mutations. Our current finding of A, analysis of canonical Wnt pathway downstream effectors after SFRP4 h siRNA treatment. Semiquantitative RT-PCR was used to confirm siRNA-directed SFRP4 silencing by promoter hypermethylation in -catenin- down-regulation of the SFRP4 expression. B, analysis of cell proliferation deficient mesothelioma cells, together with our recent report of after SFRP4 siRNA treatment. Numbers of viable cells transfected with SFRP4 Dvl (an upstream mediator of Wnt signaling) overexpression in siRNA (100 nmol/L; 5) and nonsilencing control siRNA (100 nmol/L; o). Points, means; bars, SD. C, graph of apoptosis induction by Alimta in mesothelioma (20), provides more evidence that upstream Wnt siRNA-treated H2052 cells analyzed by Annexin V staining. Columns, signaling may be important during the development of cancer, mean percentage of apoptotic cells; bars, SD. Experiments were done in triplicate, and a total of 2 Â 104 cells were analyzed in each individual especially mesothelioma. experiment. D, Western blot analysis of apoptosis pathway effectors after Alimta and siRNA treatment. In all Western blots, cytosolic proteins were prepared and h-actin served as loading control. Acknowledgments Received 5/25/2004; revised 11/19/2004; accepted 11/29/2004. Grant support: Larry Hall Memorial Trust and Kazan, McClain, Edises, Abrams, knockdown of SFRP4 in SFRP4-expressing mesothelioma cells Fernandez, Lyons & Farrise Foundation. further enhances canonical Wnt signaling, inhibits apoptosis, and The costs of publication of this article were defrayed in part by the payment of page promotes cell growth, suggesting a critical role of SFRP4 during the charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. development of mesothelioma. We thank Drs. Amir Rattner (Department of Molecular Biology and Genetics, Previous studies have shown that SFRPs can sensitize cells to School of Medicine, Johns Hopkins University, Baltimore, MD) for kindly providing the h SFRP4 cDNA construct and Hiromu Suzuki (Sidney Kimmel Comprehensive Cancer proapoptotic stimuli through -catenin down-regulation (27). Center at Johns Hopkins, Baltimore, MD) for kindly providing the primer sequences of However, involvement of h-catenin-independent mechanism(s) in methylation analysis and RT-PCR for SFRP4.

References 8. Kurose K, Sakaguchi M, Nasu Y, et al. Decreased in most invasive carcinomas except of the medullary expression of REIC/Dkk-3 in human renal clear cell type. Oncogene 2001;20:5810–7. 1. Giles RH, van Es JH, Clevers H. Caught up in a Wnt carcinoma. J Urol 2004;171:1314–8. 14. Ko J, Ryu KS, Lee YH, et al. Human secreted frizzled- storm: Wnt signaling in cancer. Biochim Biophys Acta 9. Shou J, Ali-Osman F, Multani AS, Pathak S, Fedi P, related protein is down-regulated and induces apoptosis 2003;1653:1–24. Srivenugopal KS. Human Dkk-1, a gene encoding a Wnt in human cervical cancer. Exp Cell Res 2002;280:280–7. 2. Taketo MM. Shutting down Wnt signal-activated antagonist, responds to DNA damage and its over- 15. To KF, Chan MW, Leung WK, et al. Alterations of cancer. Nat Genet 2004;36:320–2. 3. Tian E, Zhan F, Walker R, et al. The role of the Wnt- expression sensitizes brain tumor cells to apoptosis frizzled (FzE3) and secreted frizzled related protein signaling antagonist DKK1 in the development of following alkylation damage of DNA. Oncogene (hsFRP) expression in gastric cancer. Life Sci osteolytic lesions in multiple myeloma. N Engl J Med 2002;21:878–89. 2001;70:483–9. 2003;349:2483–94. 10. Mikheev AM, Mikheeva SA, Liu B, Cohen P, Zarbl H. A 16. Caldwell GM, Jones C, Gensberg K, et al. The Wnt 4. Nozaki I, Tsuji T, Iijima O, et al. Reduced expression of functional genomics approach for the identification of antagonist sFRP1 in colorectal tumorigenesis. Cancer REIC/Dkk-3 gene in non-small cell lung cancer. Int putative tumor suppressor genes: Dickkopf-1 as sup- Res 2004;64:883–8. J Oncol 2001;19:117–21. pressor of HeLa cell transformation. Carcinogenesis 17. Suzuki H, Gabrielson E, Chen W, et al. A genomic 5. Tsuji T, Nozaki I, Miyazaki M, et al. Antiproliferative 2004;25:47–59. screen for genes upregulated by demethylation and activity of REIC/Dkk-3 and its significant down- 11. Hoang BH, Kubo T, Healey JH, et al. Dickkopf 3 histone deacetylase inhibition in human colorectal regulation in non-small-cell lung carcinomas. Biochem inhibits invasion and motility of Saos-2 osteosarcoma cancer. Nat Genet 2002;31:141–9. h Biophys Res Commun 2001;289:257–63. cells by modulating the Wnt- -catenin pathway. Cancer 18. Suzuki H, Watkins DN, Jair KW, et al. Epigenetic 6. Kobayashi K, Ouchida M, Tsuji T, et al. Reduced Res 2004;64:2734–9. inactivation of SFRP genes allows constitutive WNT expression of the REIC/Dkk-3 gene by promoter-hyper- 12. Lee A, He B, You L, et al. Expression of the secreted signaling in colorectal cancer. Nat Genet 2004;36: 417–22. methylation in human tumor cells. Gene 2002; 282:151–8. frizzled-related protein gene family is downregulated 19. You L, He B, Uematsu K, et al. Inhibition of Wnt-1 7. Roman-Gomez J, Jimenez-Velasco A, Agirre X, et al. in human mesothelioma. Oncogene 2004;23:6672–6. signaling induces apoptosis in h-catenin-deficient Transcriptional silencing of the Dickkopfs-3 (Dkk-3) 13. Ugolini F, Charafe-Jauffret E, Bardou VJ, et al. WNT mesothelioma cells. Cancer Res 2004;64:3474–8. gene by CpG hypermethylation in acute lymphoblastic pathway and mammary carcinogenesis: loss of 20. Uematsu K, Kanazawa S, You L, et al. Wnt pathway leukaemia. Br J Cancer 2004;91:707–13. expression of candidate tumor suppressor gene SFRP1 activation in mesothelioma: evidence of Dishevelled

www.aacrjournals.org 747 Cancer Res 2005; 65: (3). February 1, 2005

overexpression and transcriptional activity of h- 24. You Z, Saims D, Chen S, et al. Wnt signaling 27. Kawano Y, Kypta R. Secreted antagonists of catenin. Cancer Res 2003;63:4547–51. promotes oncogenic transformation by inhibiting c- the Wnt signalling pathway. J Cell Sci 2003;116: 21. Baylin SB. Mechanisms underlying epigenetically Myc-induced apoptosis. J Cell Biol 2002;157: 429–40. 2627–34. mediated gene silencing in cancer. Semin Cancer Biol 25. Hrzenjak A, Tippl M, Kremser ML, et al. Inverse 28. Lee AY, He B, You L, et al. Dickkopf-1 antagonizes 2002;12:331–7. correlation of secreted frizzled-related protein 4 and Wnt signaling independent of h-catenin in human 22. Kim PJ, Plescia J, Clevers H, Fearon ER, Altieri DC. h-catenin expression in endometrial stromal sarco- mesothelioma. Biochem Biophys Res Commun 2004; Survivin and molecular pathogenesis of colorectal mas. J Pathol 2004;204:19–27. 323:1246–50. cancer. Lancet 2003;362:205–9. 26. Hanauske AR. The role of Alimta in the treatment of 29. Veeman MT, Axelrod JD, Moon RT. A second 23. Chen S, Guttridge DC, You Z, et al. Wnt-1 signaling malignant pleural mesothelioma: an overview of canon. Functions and mechanisms of h-catenin- inhibits apoptosis by activating h-catenin/Tcell factor- preclinical and clinical trials. Lung Cancer 2004;45 independent Wnt signaling. Dev Cell 2003;5: mediated transcription. J Cell Biol 2001;152:87–96. Suppl 1:S121–4. 367–77.

Cancer Res 2005; 65: (3). February 1, 2005 748 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 2005 American Association for Cancer Research. Secreted Frizzled-Related Protein 4 Is Silenced by Hypermethylation and Induces Apoptosis in β-Catenin− Deficient Human Mesothelioma Cells

Biao He, Amie Y. Lee, Sina Dadfarmay, et al.

Cancer Res 2005;65:743-748.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/3/743

Cited articles This article cites 28 articles, 7 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/3/743.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/3/743.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://cancerres.aacrjournals.org/content/65/3/743. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.