Deletion of the WWOX Gene and Frequent Loss of its Protein Expression in Human Osteosarcoma

By Jilong Yang, MD, PhD and Wei Zhang, PhD

Introduction

The WWOX gene has been proposed as a tumor suppressor gene, and loss of its expression has been found in many cancers. Loss of the WWOX expression has been associated with more aggressive phenotypes and poorer prognosis in breast cancer, non-small cell lung cancer, bladder cancer, gastric cancer, and sporadic meningioma.^1-7 Loss of WWOX expression has also been found in pancreatobiliary cancers, but there is no correlation with prognosis, suggesting that loss of WWOX expression may be an early event in these cancers.⁸ Restoration of WWOX expression in breast and lung cancers cells that lack endogenous WWOX expression resulted in significant caspase-mediated apoptosis, growth inhibition, and blocked tumor development in athymic nude mice.^9-10 In ovarian cancer WWOX expression abolishes the tumorigenicity in vivo and decreases attachment to fibronectin via integrin alpha^4,11 The WWOX protein has been shown to interact with c-Jun, p53, p73, AP-2, and E2F-1 and may play a central role in tumor suppression through transcriptional regulation of apoptosis pathway.¹²

Supporting evidence for the WWOX gene as a tumor suppressor gene was obtained in a genetically engineered mouse model where targeted deletion of the WWOX gene resulted in development of osteosarcomas in juvenile WWOX _/_ mice and lung papillary carcinomas in adult WWOX _/_ mice.¹¹ This finding also implicated the WWOX gene in the development of osteosarcoma, but there is no information regarding the status of the WWOX gene in human osteosarcoma.

Osteosarcoma is the most common primary non-haemopoietic malignant bone tumor with complex karyotypes and a highly unstable genome exhibiting both numerical- and structural-chromosomal instability. In this study, which was supported by the Liddy Shriver Sarcoma Initiative, we investigated the role of the WWOX gene in the pathogenesis of osteosarcoma by examining the gene copy number status of the WWOX gene and WWOX protein expression in primary osteosarcoma tissues. We collected fresh osteosarcoma tissues, isolated genomic DNA from these tissues, and performed whole-genome array Comparative Genomic Hybridization (aCGH) to evaluate the copy number changes of the WWOX gene.  We also collected FFPE tissues with clinical information to detect the expression of the WWOX protein by immunohistochemistry. The results from this study are in press in an upcoming issue of the journal Cancer Letters.¹³

Results

To gain insight into the global genetic alterations and specifically the gene copy number status of the WWOX gene in osteosarcoma, we carried out a high-density genome-wide aCGH profiling with genomic DNA isolated from 10 fresh osteosarcoma tissues from 9 patients. The aCGH results were analyzed for recurrent alterations.

When all ten samples were analyzed together for recurrent gene copy number alterations, we observed several major regions of deletions and amplifications (Figure 1A). These regions included 1,162 genes with high level amplifications and 146 genes with deletion (FDR <10%). To determine whether the recurrent gene copy alterations we observed are representative, we downloaded an independent aCGH data (GSE9654) (Figure 1B) from a cohort of ten osteosarcoma patients.¹⁴ Strikingly, the overall recurrent gene copy alteration patterns of these two independent populations in two different countries (China and Canada) were very similar, suggesting distinct genetic alterations underlying the pathogenesis of osteosarcoma.

We next zeroed in on the WWOX gene, which is located on chromosome 16q23.3-q24.1. The WWOX gene was not on the recurrent copy number alteration list from our analysis.  We then examined the WWOX gene locus on aCGH data. Deletion was only observed in 3 of 10 samples (Figure 2A). Our analysis of the Squire et al dataset¹⁴ showed that the WWOX gene was deleted in 3 of ten samples and amplified in 1 sample (Figure 2B). The data suggested that deletion of the WWOX gene occurs in 30% of human osteosarcomas.

We evaluated whether the expression of WWOX protein is altered in osteosarcoma by immunohistochemistry.  For this purpose, we acquired 55 cases of FFPE tissues from patients whom we also obtained clinical information. The normal cutaneous, muscular, and skeletal tissues exhibited strong positive staining in the cytoplasm (Figures 3A-3B).  The WWOX protein was negative in 61.8% (34/55) cases (Figure 3I) and positive in 38.2% (21/55) (Figures 3C-3H).  The counts of weak positive (Figures 3G-3H) and moderate positive (Figures 3E-3F) were 10 (18.2%, 10/55) and 6 (10.9%, 6/55), respectively. Consistently with what had been observed, for the positive cases, the WWOX protein was located predominantly in the cytoplasm. The staining was very strong in 5 cases and in these case both cytoplasm and nucleus were positive (Figures 3C-3D). In contrast to what was reported in some cancer types such as breast cancer, the WWOX expression in osteosarcoma had no correlation with the prognosis and other clinical information including age, tumor site, pathological type, PTNM staging, tumor recurrence, metastasis, disease free survival, or the overall survival. This pattern is similar to results in pancreatobiliary cancers where the loss or reduced WWOX expression did not predict tumor progression or patient survival.⁸

Our results showed that the WWOX protein is lost in more than half of the osteosarcoma tissues and in accordance with the gene deletion. Thus, the WWOX gene deletion and loss of expression appear to be the mechanism for loss of function of this tumor suppressor gene. All these results suggest that loss of WWOX through deletion and loss of expression is likely involved in the early stage of pathogenesis of human osteosarcoma.

Conclusions and Future Directions

As far as we can tell, this is the first investigation regarding the role of the WWOX gene in the pathogenesis of human osteosarcoma. We are reporting this new finding in an upcoming issue of Cancer Letters.¹³ Because the loss of WWOX expression is common but the gene deletion is not so frequent, epigenetic changes of the WWOX gene such as promoter methylation might also be involved in the pathogenesis of osteosarcoma. Future investigation into the epigenetic regulation of the WWOX gene will shed more light into the early event leading to the loss of the WWOX tumor suppressor gene and provide new therapeutic opportunities for osteosarcomas with the emerging drugs that reverse the cancer associated epigenetic alteration.

V6N5 ESUN Copyright © 2009 Liddy Shriver Sarcoma Initiative.

Characterization of WWOX Tumor Suppressor Gene in Osteosarcoma

By Jilong Yang, MD, PhD and Wei Zhang, PhD

Introduction

Osteosarcoma is the most common primary malignancy of bone in children and adolescents. It is an aggressive neoplasm composed of spindle cells producing osteoid, and can occur in any bone, but most frequently occurs in the long bones of the arms and legs near the growth plates, such as the thigh bone above the knee, the shin bone below the knee, and the upper arm bone. Currently, the standard therapy for osteosarcoma patients includes surgery with pre- and post-surgery chemotherapy consisting of a combination of adriamycin, cisplatin, ifosfamide, and methotrexate. The combination of these drugs and surgery has significantly improved the 5-year disease-free survival rate up to 60-70% during the last 30 years. However, the total survival and disease-free survival is still very low.

Background

The causes of osteosarcoma are not known. Aqeilan et al demonstrated that the loss of both alleles of WWOX gene results in osteosarcomas in some early postnatal mice, while loss of one allele significantly increases the incidence of spontaneous and chemically induced tumors.¹ Histological examination of the knockout mice showed focal lesions along the bone diaphysis with increased numbers of osteoblast and enlarged cells in cartilage matrix with multiple nucleoli, suggesting proliferation of the progenitors arising from the periosteum. These tumors developed in young postnatal mice in the absence of any carcinogenic treatment, suggesting a role of WWOX as a tumor suppressor in the bone. Although WWOX gene alteration has been implicated in chromosomal alterations in many types of human malignancies including breast, lung, prostate, esophageal, pancreatic, and stomach carcinoma, there is no report on the role of the WWOX gene in osteosarcoma.

The WWOX gene encodes a WW domain-containing oxidoreductase. The gene is located at chromosome 16q23.1-16q23.2, a region that spans the second most common human fragile site, FRA16D. The WWOX protein contains two N- terminal WW domains and a central short chain oxidoreductase-like domain. The two N-terminal WW domains are important for protein–protein interactions that form protein complexes involved in a variety of cellular processes such as transcriptional regulation and protein stability. Analysis of several WWOX interacting candidates revealed that WWOX, via its first WW domain, associates with the proline-rich motif PPxY found in a number of transcription factors including p73, AP2gamma, and Jun. WWOX associates with its targets in the cytoplasm and prevents their translocation into the nucleus, thus regulating their transcriptional activity. The recent observations indicate that loss of WWOX expression in human tumors may contribute to their pathogenesis. Bednarek et al. showed that ectopic WWOX expression strongly inhibits anchorage independent growth of breast cancer cell lines in soft agar.² WWOX overexpression induces a dramatic inhibition of tumorigenicity of MDA-MB-435 breast cancer cells in mice. Subsequently, Fabbri et al. utilized a high-titer adenoviral vector that encodes WWOX (Ad-WWOX) to restore WWOX expression in lung cancer cell lines that lack expression of endogenous WWOX.³ Using this technique, overexpression of WWOX inhibited cell growth in three different WWOX-negative human lung cancer cell lines, but not in a WWOX-positive lung cell line. Similar experiments in pancreatic, prostate and breast cancer cell lines supported these results. WWOX overexpression in these studies resulted in caspase-mediated apoptosis. Thus, multiple lines of evidence, both in cell culture and in athymic mice, suggested that WWOX functions as a tumor suppressor. Aqeilan et al recently generated mice carrying a targeted deletion of the WWOX gene. Using conventional gene targeting methods, a mouse lacking exons 2, 3, and 4 of the mouse WWOX gene was generated. Analysis of the WWOX mutant mice demonstrated that WWOX functions as a tumor suppressor. Four of 13 juvenile WWOX_/_ knockout mice developed osteosarcoma. These tumors developed in young postnatal mice in the absence of any carcinogenic treatment, suggesting a role of WWOX as a tumor suppressor in the bone. Altogether these data indicate that WWOX is a bona fide tumor suppressor and that inactivation of one WWOX allele is sufficient for tumorigenesis.

Purpose of the Specific Investigation

In our study, we propose to test the hypothesis that the WWOX gene is an osteosarcoma tumor suppressor gene that is inactivated in human osteosarcoma. We will perform the following experiments in this study. 1. To examine homozygous deletion, loss of heterogeneity (LOH), methylation of WWOX promoter, and mutation of WWOX in osteosarcoma cell lines and patient tumor tissues. 2. To investigate the role of WWOX in cell apoptosis and proliferation by correlating WWOX gene deletion, mutation, and expression status with expression for P73, BAX, Bcl-2, Bcl-xL, casepase-3, casepase-9, Ki-67 and PCNA. We will use standard molecular and pathological methods including PCR analysis, sequencing, Immunoblotting and immunohistochemistry, methylation-specific PCR (MSP), DHPLC and flow cytometric analysis. The tissue bank of our hospital has accumulated 35 frozen osteosarcoma tissues and there are about 150 paraffin-embedded archival blocks available for this investigation. Our preliminary investigations in human osteosarcoma tissues detected WWOX gene deletion and promoter methylation. These studies support our hypothesis. We anticipate that this study will improve our understanding of the molecular mechanisms of osteosarcoma development by determination whether the WWOX gene is key tumor suppressor gene and a new target for gene therapy development for osteosarcoma.

Research and experimental design

We will test our hypothesis through the following research steps:

1. To determine the status of homozygous deletions, loss of heterozygosity, and mutation of WWOX gene in bone and osteosarcoma cell lines, S180S37, U-2OS, SW 1353, and MG-63.
2. To determine the methylation status of WWOX promoter region in bone and osteosarcoma cell lines.
3. To determine gene deletion and mutation of WWOX gene in osteosarcoma tissues.
4. To determine expression levels of WWOX protein and related proteins such as P73, BAX, Bcl-2, Bcl-xL, casepase-3, casepase-9, Ki-67 and PCNA on a tissue microarray. We will correlate the WWOX status with clinical information of the patients as well as the apoptosis and proliferation markers.

The potential significance of this investigation

Osteosarcoma is among the most common non-hematologic primary malignant tumors of adults and children. Although an increased number of molecular studies have been initiated in recent years, no clear results or new therapeutic treatments have been obtained. This first study of the biological effects of WWOX expression in osteosarcoma cells will determine the role of the reduced WWOX signal in osteosarcoma pathobiology and whether the WWOX gene is an important tumor suppressor gene in osteosarcoma. Our proposed study will explore whether drugs capable of reactivating the epigenetically silenced WWOX gene could be effective in treatment of osteosarcoma. Restoration of WWOX protein expression may lead to induction of apoptosis and suppression of tumorigenicity. The delineation of this signal pathway will not only help to clarify the role of a significant signal mediator in osteosarcoma but also may lead to identification of therapeutic targets for osteosarcoma.

V5N3 ESUN Copyright © 2008 Liddy Shriver Sarcoma Initiative.

Grant Funding

The Liddy Shriver Sarcoma Initiative funded this study in June 2008. The study was made possible, in part, by several generous donations from Herbert Blodgett in loving memory of his wife, Merril, by generous donations from Una O'Hagen in memory of her son Sean Keane, and by generous donations from the Kleftis family in memory of Gregory.