1001-1009 5/4/08 13:59 Page 1001 INTERNATIONAL JOURNAL OF ONCOLOGY 32: 1001-1009, 2008 Hyaluronan-mediated motility: A target in oral squamous cell carcinoma YUKIO YAMANO1, KATSUHIRO UZAWA1,3, KEIJI SHINOZUKA1, KAZUAKI FUSHIMI1, TAKASHI ISHIGAMI1, HITOMI NOMURA1, KATSUNORI OGAWARA1, MASASHI SHIIBA1, HIDETAKA YOKOE3 and HIDEKI TANZAWA1-3 1Department of Clinical Molecular Biology, Graduate School of Medicine, 2Center of Excellence (COE) Program in The 21st Century, Graduate School of Medicine, Chiba University; 3Division of Dentistry and Oral-Maxillofacial Surgery, Chiba University Hospital, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan Received December 27, 2007; Accepted February 21, 2008 Abstract. To characterize cancer-related gene expression Cancers of the oral cavity account for 40% of head and changes in oral squamous cell carcinomas (OSCCs), we neck cancers including squamous cell carcinoma (2). Oral compared the gene expression profiles in OSCC-derived squamous cell carcinomas (OSCCs) have been identified cell lines with human normal oral keratinocytes (HNOKs). as major global threats to public health, the prognosis of Microarray analysis identified 166 genes that were up-regu- which have not improved in decades (3). With the currently lated in OSCC-derived cell lines. Gene ontology analysis available clinical assessment and treatment methods, a third showed that cancer-related function had the highest signi- of all cases of oral cancer are diagnosed only after becoming ficance. Among the genes mapped to the cancer-related locally advanced (4,5). In addition, aggressive treatment of network with the highest significance, the receptor for OSCCs is controversial because it can lead to severe morbidity hyaluronan-mediated motility (RHAMM) was evaluated and disfigurement. As a result, many patients with OSCCs further for mRNA and protein expression in the OSCC are either overtreated or untreated, with substantial personal cell lines, primary OSCCs. Overexpression of RHAMM and socioeconomic implications. A number of etiologic protein was observed in all cell lines compared to HNOKs. factors have been implicated in the development of OSCCs, Immunohistochemical analysis showed highly expressed such as the use of tobacco, alcohol, or the presence of incom- RHAMM in primary OSCCs, whereas most corresponding patible prosthetic materials (6,7). However, some patients normal tissues had no or significant down-regulation of develop OSCC without risk factors, suggesting that host protein immunoreactivity. Real-time quantitative reverse susceptibility may play a role. Molecular alterations in a transcriptase-polymerase chain reaction data agreed with number of oncogenes and tumor suppressor genes associated the protein expression. Moreover, the RHAMM expression with the development of OSCC could be important clues status was correlated with the TNM stage (P<0.001). The for addressing these problems (7,8). Comprehensive and results suggested that RHAMM expression may be correlated exhaustive expression studies of numerous genes, including with tumor aggressiveness and offer clues to the development functionally unknown genes, are essential to understand the of new treatments for human OSCCs. complexity and polymorphisms of OSCC. Microarray technologies have been used widely for Introduction comprehensive gene expression analysis that enables researchers to simultaneously analyze changes in thousands Head and neck cancers are the sixth most common cancers of genes and identify significant patterns. However, previous worldwide and major causes of morbidity and mortality (1). microarray studies performed in OSCCs have produced lists of differentially expressed genes (9-11) but failed to establish how these genes form regulatory networks. Moreover, these _________________________________________ studies have ignored the genes that do not pass the randomly or empirically determined criteria for gene selection. We Correspondence to: Dr Katsuhiro Uzawa, Chiba University, 1-8-1 performed microarray analysis using high-density Affymetrix Inohana, Chuo-ku, Chiba 260-8670, Japan Human Genome-U133A plus 2.0 GeneChip arrays containing E-mail: [email protected] 54,675 probe sets (Affymetrix, Santa Clara, CA) to compare gene expression patterns among the OSCC-derived cell lines Key words: oral squamous cell carcinoma, receptor for and human normal oral keratinocytes (HNOKs). We also hyaluronan-mediated motility, pathway analysis, microarray adopted a computational tool, Ingenuity Pathway Analysis analysis, qRT-PCR, Western blot, immunohistochemistry (IPA) software (Ingenuity Systems, Mountain View, CA) to identify regulatory networks of differentially expressed genes related to OSCCs. 1001-1009 5/4/08 13:59 Page 1002 1002 YAMANO et al: UP-REGULATED GENES IN OSCCs We analyzed the gene expression profiles of OSCCs using Data analysis. GeneChip analysis was performed based on microarray technology to identify genes associated with the Affymetrix GeneChip Manual with Microarray Analysis oral carcinogenesis, which then were analyzed for network Suite 5.0, Data Mining Tool 2.0, and Microarray Database and gene ontology by IPA to identify networks of interacting software. All genes on the GeneChip were globally normalized genes and other functional groups. We also verified experi- and scaled to a signal intensity of 500. The Microarray mentally that the extracellular matrix-associated protein, Analysis Suite software used Wilcoxon's test to generate receptor for hyaluronan-mediated motility (RHAMM), is detected (present or absent) calls and used the calls to OSCC up-regulated in humans and also plays an important statistically determine if a transcript was expressed or not. role in tumor progression. After being filtered through a ‘present’ call (P<0.05), the expression data were analyzed using GeneChip Operating Materials and methods Software 1.1 (Affymetrix) and GeneSpring 6.1 (Silicon Genetics, Redwood City, CA). Fold changes were calculated Tissue specimens and cell lines. Tumors and patient-matched by comparing transcripts between OSCC-derived cell lines normal epithelium were obtained during surgical resection at and HNOKs. Chiba University Hospital after patient informed consent was obtained under a protocol reviewed and approved by the Network and gene ontology analysis. The list of up-regulated institutional review board of Chiba University. The resected genes in OSCC identified by microarray analysis was used tissues were divided into two parts, one of which was frozen for network and gene ontology analysis. Gene accession immediately after removal of the surrounding normal tissues numbers were imported into the IPA software. The genes and stored at -80˚C until protein isolation, and the second were categorized based on location, cellular components, and of which was fixed in 10% buffered formaldehyde solution reported or suggested biochemical, biologic, and molecular for pathologic diagnosis and immunohistochemical staining. functions using the software. The identified genes also were Histopathologic diagnosis of each neoplastic tissue was mapped to genetic networks available in the IPA database performed according to the World Health Organization criteria and ranked by score. The score is the probability that a by the Department of Pathology, Chiba University Hospital. collection of genes equal to or greater than the number in Clinicopathologic staging was determined by the TNM a network can be achieved by chance alone. A score of 3 classification of the International Union against Cancer. All indicates that there is a 1/1000 chance that the focus genes patients had SCC that was histologically confirmed, and are in a network due to random chance. Therefore, scores tumor samples were checked to ensure that tumor tissue was of 3 or higher have a 99.9% confidence level of not being present in >80% of the specimens. The OSCC-derived cell generated by random chance alone. This score was used lines used in this study were HSC-2, HSC-3, HSC-4, Ca9-22, as the cut-off for identifying gene networks. (Human Science Research Resources Bank, Osaka, Japan), H1, and Sa3 (provided by Dr Fujita at Wakayama Medical Protein and mRNA extraction. Protein was extracted when University, Wakayama, Japan). All OSCC-derived cell lines the cells reached 80-90% confluence; they were washed were grown in Dulbecco's modified Eagle's medium/F-12 twice with phosphate-buffered saline (PBS), scraped into a HAM (Sigma-Aldrich Co., St. Louis, MO) supplemented with tube and centrifuged briefly. The cell pellets were incubated 10% fetal bovine serum (Sigma) and 50 units/ml penicillin for 30 min in a lysis buffer containing 7 M urea, 2 M thiourea, and streptomycin (Sigma). Healthy oral gingival specimens 4% w/v CHAPS, and 10 mM Tris pH 8.0, and lysed by were collected from 22- to 35-year-old patients at Chiba sonication (3x10 sec pulses on ice). The sample was centri- University Hospital. The institutional review board of Chiba fuged at 13,000 rpm for 20 min. The supernatant containing University approved all relevant protocols. Five independent the cell proteins then was recovered and the protein concen- HNOKs were primary cultured and maintained in defined tration was measured with a Protein Assay Kit (Bio-Rad keratinocyte-SFM (Gibco-BRL, Gaithusberg, Germany) (12). Laboratories, Hercules, CA) and adjusted to 1 mg/ml with lysis buffer. The pH of the protein sample was adjusted to Affymetrix GeneChip hybridization. Double-stranded cDNA 8.5 with 30 mM Tris-HCl. Total