Wnt2 as a new therapeutic target in malignant pleural mesothelioma
Version of Record online: 17 MAY 2005
Copyright © 2005 Wiley-Liss, Inc.
International Journal of Cancer
Volume 117, Issue 2, pages 326–332, 1 November 2005
How to Cite
Mazieres, J., You, L., He, B., Xu, Z., Twogood, S., Lee, A. Y., Reguart, N., Batra, S., Mikami, I. and Jablons, D. M. (2005), Wnt2 as a new therapeutic target in malignant pleural mesothelioma. Int. J. Cancer, 117: 326–332. doi: 10.1002/ijc.21160
- Issue online: 16 AUG 2005
- Version of Record online: 17 MAY 2005
- Manuscript Accepted: 28 FEB 2005
- Manuscript Received: 31 DEC 2004
- Larry Hall Memorial Trust
- The Kazan, McClain, Edises, Abrams, Fernandez, Lyons and Farrise Foundation
- Fondation pour la Recherche Medicale
- Association Nationale pour le Traitement a Domicile des Insuffisants Respiratores
- monoclonal antibody
Malignant mesothelioma of the pleura (MPM) is a highly aggressive neoplasm with a poor prognosis and limited treatment options. A better understanding of its pathogenesis is essential to developing alternative therapeutic strategies. We previously demonstrated that the Wnt signaling pathway is activated in MPM through the overexpression of disheveled proteins. To extend our knowledge of Wnt signaling activation in MPM, we performed Wnt-specific microarrays in normal pleura and MPM. We found that the most common event in MPM was the upregulation of Wnt2. We inhibited Wnt2 by siRNA and a monoclonal anti-Wnt2 antibody and analyzed their effects on apoptosis and downstream signaling effectors. We then assessed the antiproliferative effects of the Wnt2 antibody and Alimta, one of the current standard treatments of MPM. We confirmed Wnt2 overexpression at the mRNA and protein level in MPM cell lines and tissues. We then demonstrated that inhibition of Wnt2 by siRNA or a monoclonal antibody induces programmed cell death in MPM cells. We next analyzed the effects of the anti-Wnt2 antibody and of Alimta on MPM cell proliferation. We found that although Wnt2 antibody by itself had less antiproliferative potency than Alimta, the two in combination had substantially more activity than Alimta alone. We thus propose that inhibition of Wnt2 is of therapeutic interest in the development of more effective treatments for MPM. © 2005 Wiley-Liss, Inc.
Malignant pleural mesothelioma (MPM) is a highly aggressive and challenging cancer arising primarily from the pleural lining of the lung. Approximately 3,000 patients are diagnosed with MPM in the United States annually and the incidence of this tumor is predicted to increase dramatically over near term, peaking around 2020.1 Since MPM usually presents at an advanced stage, a curative resection is rarely possible. Radiotherapy has failed to show clinical benefit as a single treatment modality, and the administration of chemotherapy is mostly restricted to the advanced stage with limited efficiency.2 Alternative strategies based on pleural injections of recombinant cytokines have similarly proven unsatisfactory.3 Since current interventions offer only limited benefit and overall survival is low, there is an urgent need to develop new therapeutic agents based on a greater understanding of MPM's underlying molecular mechanisms.
The Wnt family of secreted glycoproteins is a group of signaling molecules broadly involved in developmental processes and oncogenesis.4, 5 Nineteen human Wnt proteins have thus far been identified. Transduction of Wnt signals is triggered by the binding of Wnt ligands to 2 distinct families of cell-surface receptors: the frizzled (Fz) receptor family and the LDL-receptor-related protein (LRP) family.6 Intracellularly, Wnt signaling activates disheveled (Dvl) proteins, which inhibit glycogen synthase kinase-3β (GSK-3β) phosphorylation of β-catenin, leading to its cytosolic stabilization. Stabilized β-catenin then enters the cell nucleus and associates with LEF/TCF transcription factors. β-catenin-Tcf/Lef induces transcription of important downstream target genes, many of which have been implicated in cancer.
The role of Wnt signaling in cancer was first suggested 20 years ago with the discovery of Wnt-1 as an integration site for mouse mammary tumor virus (MMTV) in mouse mammary carcinoma.7 Increasing evidence over time suggests the Wnt signaling pathway is involved in tumor development and/or progression.4, 8 Numerous reports have demonstrated aberrant activation of the Wnt signaling pathway in disparate human cancers, including colorectal cancer,9, 10 head and neck carcinoma,11 melanoma12 and leukemia.13
Recent data have further elucidated the role of the Wnt pathway in MPM. First we demonstrated that Wnt signaling is activated through disheveled and β-catenin overexpression in MPM.14 Second, Abutaily et al.15 confirmed by immunohistochemistry that β-catenin accumulates in the nucleus in MPM. Third, Wnt1 has been found to be overexpressed in mesothelioma cell lines and blocking its activity leads to apoptosis even in β-catenin-deficient cell lines.16, 17 Lastly, we demonstrated that the secreted frizzled-related protein (sFRP) gene family that antagonizes Wnt signaling is silenced by promoter methylation in MPM cell lines and tissues.18
To extend our knowledge of Wnt signaling activation in MPM, we performed Wnt-specific microarrays in normal pleura and MPM and found that the most common event was the overexpression of Wnt2 in malignant tissues. We further demonstrate that inhibiting Wnt2 using either siRNA or a monoclonal antibody leads to programmed cell death in MPM. We propose that inhibition of Wnt2 is of therapeutic interest in MPM especially when used in combination with conventional chemotherapy.
Material and methods
Cell lines and tissue samples
Mesothelioma cancer cell lines were obtained from the following sources: LRK1A and REN through a generous gift from Dr. Steven Albelda (University of Pennsylvania, Philadelphia, PA), NCI-H2052, H28 and H513 from American Type Culture Collections (ATCC, Manassas, VA), MS-1 and NCI-H290 from NIH (Frederick, MD) and LP9 were from the Cell Culture Core Facility at Harvard University (Boston, MA). We should notice that LP9 are mesothelial cells that are not activated by SV40. All cell lines except LP9 were cultured in RPMI-1640 supplemented with 10% fetal bovine serum, penicillin (100 IU/ml) and streptomycin (100 μg/ml). LP9 was cultured in M199 containing 15% medium plus 10 ng/ml EGF and 0.4 μg/ml HC. All cells were cultured at 37°C in a humid incubator with 5% CO2. Fresh mesothelioma tissues and adjacent normal pleural tissues from patients undergoing primary resection of their tumors were collected at the time of surgery and immediately snap-frozen in liquid nitrogen (IRB H8714-22942-01). These tissue samples were kept at −170°C in a liquid nitrogen freezer prior to use.
Gene expression array
Gene expression profiling was analyzed using a custom array designed to profile the expression of genes involved in and downstream of Wnt signaling with the AmpoLabelling-LPR Kit protocol (GEArray Q Human Wnt Signaling Pathway Gene Array; SuperArray, Frederick, MD). Briefly, total RNA isolated from the selected tissues was subjected to an RT reaction and cDNA probes were subsequently labeled with Biotin-16-dUTP (Roche, Mannheim, Germany), denatured and hybridized overnight in hybridization tubes containing the Wnt-specific arrays. Detection was done with a chemiluminescent reaction by using a CDD camera. Images of spots were converted in numerical data using software provided by the manufacturer. Expression data were matched against the gene list provided by the manufacturer.
Wnt2 monoclonal antibody
The anti-Wnt2 mouse monoclonal antibody (IgG1) was custom-made at Rockland (Gilbertsville, PA). Antigen used to raise this monoclonal antibody was a synthetic peptide corresponding to amino acid 49–63 of the human Wnt2 (Ac-SSQRQLCHRHPDVMR-amide). The antigen was chosen bioinformatically based on its hydrophilicity, antigenicity, accessibility, sequence homology and N-terminal vicinity.19 The test bleed was screened twice by ELISA using the peptide, and the parental clones and the subclones were screened by Western blot analysis. The monoclonal antibody was affinity-purified by using protein G and kept at −80°C. Seize X mammalian Immunoprecipitation Kit (Pierce Biotechnology, Rockford, IL) was used to precipitate Wnt2 protein from cell line extracts according to the manufacturer's protocol and followed by Western blotting.
Antibody incubation with cells
Cells were plated in 6-well plates or 10 cm dishes 1 day before experiments. Then normal media was replaced by media containing antibodies at various concentrations and the cells were incubated at 37°C in a humid incubator with 5% CO2. At various time points, the cells were collected using standard protocols for further analysis. The control antibody used was mouse IgG1 MOPC21 from Sigma-Aldrich (St. Louis, MO).
Cells were plated into a 6-well plate with media without antibiotics 24 hr before testing. The ion-exchange HPLC-purified siRNAs (Wnt2 siRNA and nonsilencing siRNA control) were purchased from Qiagen-Xeragon (Germantown, MD). The lyophilized siRNAs were dissolved in annealing buffer and reheated to 95°C for 1 min followed by 1 hr at 37°C incubation. We followed the protocol described by Elbashir et al.20 After siRNA transfection, plates were incubated for 3 days at 37°C before further analysis. The siRNA specific for human Wnt2 was derived from the mRNA sequence (5′-GAAGATGGGAAGCGCCAAG-3′) of human Wnt2. The control (nonsilencing) siRNA does not target any known mammalian gene (5′-AATTCTCCGAACGTGTCACGT-3′).
Standard protocol was used. Anti-Wnt2, Anti-Dvl3 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA), anti-β-actin antibody from Cell Signaling Technology (Beverly, MA) and anti-β-catenin antibody from Transduction Laboratories (Lexington, KY).
Cells were collected 72 hr after transfection, then processed for determination of cell surface annexin-V and propidium iodide (PI) contents (Apotarget; BioSource International, CA) according to the manufacturer's protocol. With the use of an annexin-V-PI double-staining regime, 3 populations of cells are distinguishable in 2 color flow cytometry: nonapoptotic cells: annexin-V- and PI-negative; early apoptotic cells: annexin-V-positive and PI-negative; necrotic or late apoptotic cells: annexin-V- and PI-positive. Then cells were immediately analyzed by flow cytometry (FACScan; Becton Dickinson, San Jose, CA).
Alimta (MTA) was supplied by Eli Lilly (Indianapolis, IN). Alimta was diluted in sterile physiologic solution at a concentration of 10 mg/ml. The stock was divided into aliquots, stored at −80°C and diluted in culture medium before each experiment. Wnt2 antibody was stored at 4°C and used as previously described. Cell proliferation was determined by measuring metabolic activity of tetrazolium conversion (Cell Titer 96 assay; Promega, Madison, WI). Briefly, 5,000 cells were plated per well in a 96-well plate and culture medium containing increasing concentrations of both drugs was added. Plates were incubated at 37°C for 72 hr in a CO2 incubator. Then a solubilization-stop solution was added and the absorbance was recorded with a fluorescence plate reader at a wavelength of 570 nm. Each experiment was repeated at least 3 times in triplicate.
Data shown represent mean ± SD. Unpaired t-test in the Excel was used for comparing different treatments and cell lines.
Wnt2 is overexpressed in mesothelioma cell lines and tissues
In order to analyze the activation of the Wnt pathway in mesothelioma, we performed gene expression profiling. We used a custom array designed to profile systematically the expression of genes involved in the Wnt pathway and its downstream signaling. We analyzed 16 matched samples (malignant tissue and normal adjacent pleura) obtained from 8 patients who had undergone surgery for MPM, but otherwise received no previous treatment (i.e., radiation, chemotherapy, targeted therapy) that could have compromised our results (Fig. 1). We found that numerous Wnt genes (Wnt1, Wnt2, Wnt5) and Wnt-related genes (c-myc, cyclin D1, c-jun) were upregulated. In contrast, Wnt8a and some Wnt antagonists (DKK1, sFRP2 and sFRP4) were downregulated. Among the 96 genes studied, Wnt2 upregulation in MPM was the most common event, presenting in 6 out of the 8 patients.
We performed Western blot using a commercially available anti-Wnt2 antibody to confirm Wnt2 overexpression in mesothelioma cell lines and tissues. We found that Wnt2 was overexpressed in all 8 tested mesothelioma cell lines, but weakly expressed in LP9, a normal mesothelial cell line (Fig. 2a). Moreover, all 5 freshly resected MPM tissue samples showed increased expression of Wnt2 protein compared with autologous-matched normal pleural tissue (Fig. 2b). It should be noted that these surgical specimens were different from those used in our microarray analysis.
Wnt2 monoclonal antibody leads to cell death and inhibits downstream effectors of β-catenin canonical pathway
We used a custom-made monoclonal antibody targeting Wnt2 as previously described19 to treat 4 mesothelioma cell lines: MS1, H28, H513 and LRK1A. After 3–5 days of incubation with the Wnt2 antibody, we found significant cell death at a concentration of 10 μg/ml in these cell lines, whereas a control antibody (mouse IgG1 MOPC21) had no effect. Flow cytometry analysis subsequently demonstrated that cell death was due largely to induction of programmed cell death (p < 0.005; Fig. 3b). As previously reported, we observed that the anti-Wnt2 antibody-induced apoptosis could be inhibited by a blocking peptide (data not shown).19
We then investigated the ability of the anti-Wnt2 antibody to inhibit the downstream effectors of the Wnt pathway. We previously reported that disheveled-3 was overexpressed in mesothelioma.14 Here, we found that treatment with the anti-Wnt2 antibody decreased the expression level of disheveled-3 (Fig. 4). Cytosolic β-Catenin was also downregulated in the LRK1A cell line but absent in MS1 and H28 as these 2 cell lines are known to be deficient in β-catenin.17
Wnt2 siRNA inhibition leads to cell death and inhibits signaling pathways
Wnt2-targeted small interfering RNA (siRNA) was used to study the effect of Wnt2 mRNA silencing. Similar to the results of the monoclonal anti-Wnt2 antibody experiments, treatment with Wnt2 siRNA for 3 days induced programmed cell death in these cell lines expressing Wnt2. Significant programmed cell death was induced by 100 nM Wnt2 siRNA, but was absent in the nonsilencing siRNA control (100 nM; p < 0.01; Fig. 5). We confirmed the silencing of Wnt2 expression after Wnt2 siRNA treatment (100 nM for 72 hr) by Western blot analysis (Fig. 5). Nonsilencing siRNA served as a control (100 nM for 72 hr). To determine whether the apoptotic effects correlated with the inhibition of Wnt2 signaling, we analyzed the expression of downstream proteins. We found that cytosolic β-catenin and disheveled-3 were downregulated after Wnt2 siRNA treatment.
Effects of Wnt2 Ab, Alimta alone and in combination on cell proliferation
We next assessed the effects of the anti-Wnt2 monoclonal antibody on cell proliferation and compared them with the effects of Alimta. Alimta (LY231514, MTA, pemetrexed) is a novel multifunctional antifolate antimetabolite.21 Recently, a phase 3 trial showed that the combination of Alimta with cisplatine resulted in a significantly increased efficacy compared with cisplatine alone.22 Alimta is now considered a standard in the treatment of surgically nonamenable mesothelioma. MS1 mesothelioma cells were plated in 96-well plates and treated with increasing concentrations of the Wnt2 antibody, Alimta, or both. Three days later, we assessed the cell proliferation by measuring the metabolic activity of cellular enzyme (here the tetrazolium conversion) (Fig. 6). We observed a decrease of around 30% ± 7% in cell proliferation when the Wnt2 antibody was used at a concentration of 10 μg/ml. Alimta treatment induced a decrease in cell proliferation of 41.9% ± 4% at a concentration of 1 μg/ml and of 42.9% ± 2% at a concentration of 10 μg/ml. When used together at the concentration of 10 μg/ml, cell proliferation was inhibited by 51.2% ± 5% (p < 0.005). It is noteworthy that our results may reflect either reduced cell proliferation or increased cell death (or both).
The human Wnt2 gene, located on chromosome 7q31.3, is highly expressed in fetal lung and weakly expressed in placenta.23 The link between Wnt2 and tumorigenesis was first proposed after data indicated there was Wnt2 (known as int2 at that time) amplification in human cancers.24 Similarly, Wnt2 has been implicated in mouse mammary tumorigenesis through gene amplication.25 Wnt2 was later shown to be upregulated in gastric cancer,26, 27 colorectal cancer28, 29, 30 and melanoma.31 We also demonstrated recently the overexpression of Wnt2 in non small cell lung cancer.19 Based on our gene expression profile findings, we propose here that Wnt2 upregulation is a common event in MPM. This can be due either to elevated transcriptional activity, but also to an increase of mRNA stability or by gene amplification leading to a higher mRNA copy number. We also confirm its overexpression at the protein level.
The mechanisms of MPM pathogenesis have thus far not been fully elucidated. The disease has a long latency period, suggesting that multiple genetic events are required for mesothelial cell transformation.32 Among them, deletions of chromosomal sites, especially in chromosomes 1, 3, 6 and 9, and alterations of both tumor suppressor genes (p16, p53, NF2) and oncogenes (TGF-β, IGF) have all been extensively reported.32 Another hypothesis, strongly supported by epidemiologic and in vivo data, associates simian virus 40 (SV40) with MPM.33 We and others recently highlighted the role of the Wnt pathway in the pathogenesis of MPM.14, 15, 16, 17, 18 Wnt activation in MPM may be linked to SV40 infection since it has been shown that the small t antigen (tag), a viral oncoprotein of SV40, binds to the protein phosphatase PP2A, leading to constitutive activation of the Wnt pathway.34 Dysregulation of β-catenin signaling trough primary β-catenin mutations has been shown to be an important event in several human malignancies, but has not been reported in MPM.14 Taken together, these data suggest that the Wnt pathway is of critical importance in mesothelioma.
To test our hypothesis that the Wnt2-mediated signal is a survival factor involved in the pathogenesis of MPM, we evaluated the effects of a specific anti-Wnt2 monoclonal antibody in various MPM cell lines. We demonstrate that this antibody can induce apoptosis in these cells. Most notably, the apoptotic effect was accompanied by inhibition of canonical Wnt signaling, as evidenced by a decrease in disheveled and β-catenin. Interestingly, the anti-Wnt2 antibody has the potency to induce apoptosis in β-catenin-deficient cell lines (H28 and MS1) similar to its effect on cell lines with intact β-catenin. This observation supports our recent findings that Wnt signaling may play a role in apoptosis inhibition through a noncanonical pathway involving the c-Jun NH2-terminal kinase (JNK).17 As an alternative approach, we transfected the MPM cell lines with specific Wnt2 siRNA and witnessed similar results. We thus demonstrate by 2 different methods that inhibition of Wnt2 in disparate mesothelioma cell lines induces programmed cell death
The new challenge for mesothelioma is to study multimodality approaches. In recent years, better understanding of the biology of mesothelioma has led to the development of new drugs. Recently, EGFR inhibitors, VEGF inhibitors and PDGF inhibitors have entered clinical trials and combinations of novel targeted agents with chemotherapy after radical surgery seem promising.2 Therefore, we analyzed the effects of anti-Wnt2 antibody and combined with Alimta (one of the current standard treatment of MPM) on MPM cell proliferation. We found that although Wnt2 antibody by itself had less antiproliferative potency than Alimta, the 2 in combination had substantially more activity than Alimta alone. In the perspective of future clinical studies, the individual Wnt2 levels should be screened in these patients to ensure that the target is present.
In summary, we propose that Wnt2 is a new and promising target for treatment of MPM. Our findings, including the potency of our antibody in inducing apoptosis, raise its therapeutic interest in the development of more effective treatments for this disease.
Supported by the Fondation pour la Recherche Medicale and ANTADIR (both to J.M.).