THY1 is a candidate tumour suppressor gene with decreased ... - Nature

Oncogene (2005) 24, 6525–6532

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

THY1 is a candidate tumour suppressor gene with decreased expression in metastatic nasopharyngeal carcinoma Hong Lok Lung1, Dhinoth Kumar Bangarusamy2, Dan Xie3,9, Arthur Kwok Leung Cheung1, Yue Cheng1,10, Mande Kuppusamy Kumaran2,11, Lance Miller2, Edison Tak-Bun Liu2, Xin-Yuan Guan3, Jonathan Shuntong Sham3, Yan Fang4, Liqiong Li5, Nancy Wang5, Alexey I Protopopov6, Eugene R Zabarovsky6, Sai Wah Tsao7, Eric J Stanbridge8 and Maria Li Lung*,1 1 Department of Biology, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong (SAR), People’s Republic of China; 2Genome Institute of Singapore, Biomedical Sciences Institutes, Singapore; 3The Department of Clinical Oncology, University of Hong Kong, Pokfulam, Hong Kong (SAR), People’s Republic of China; 4Cancer Institute, Sun Yat-Sen University, Guangzhou, People’s Republic of China; 5Department of Pathology and Laboratory Medicine, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642, USA; 6Microbiology and Tumor Biology Center, Center for Genomics and Bioinformatics, Karolinska Institute, Stockholm, Sweden; 7Department of Anatomy, University of Hong Kong, Pokfulam, Hong Kong (SAR), People’s Republic of China; 8Department of Microbiology and Molecular Genetics, University of California, Irvine, CA 92697, USA

Using oligonucleotide microarray analysis, THY1, mapping close to a previously defined 11q22–23 nasopharyngeal carcinoma (NPC) critical region was identified as showing consistent downregulated expression in the tumour segregants, as compared to their parental tumour-suppressing microcell hybrids (MCHs). Gene expression and protein analyses show that THY1 was not expressed in the NPC HONE1 recipient cells, tumour segregants, and other NPC cell lines; THY1 was exclusively expressed in the non-tumourigenic MCHs. The mechanism of THY1 gene inactivation in these cell lines was attributed to hypermethylation. Clinical study showed that in 65% of NPC specimens there was either downregulation or loss of THY1 gene expression. Using a tissue microarray and immunohistochemical staining, 44% of the NPC cases showed downregulated expression of THY1 and 9% lost THY1 expression. The frequency of THY1 downregulated expression in lymph node metastatic NPC was 63%, which was significantly higher than in the primary tumour (33%). After transfection of THY1 gene into HONE1 cells, a dramatic reduction of colony formation ability was observed. These findings suggest that THY1 is a good candidate tumour suppressor gene in NPC, which is significantly associated with lymph node metastases. Oncogene (2005) 24, 6525–6532. doi:10.1038/sj.onc.1208812; published online 20 June 2005 Keywords: chromosome 11; THY1; nasopharyngeal carcinoma; microcell hybrid; oligo-microarray *Correspondence: M Lung; E-mail: [email protected] 9 Current address: Cancer Institute, Sun Yat-Sen University, Guangzhou, People’s Republic of China 10 Current address: Genetics Branch, National Cancer Institute, National Naval Medical Center, Bethesda, MD 20889-5105, USA 11 Current address: Physiological and Genomics Medicine Group, Imperial College, London W12 ONN, UK Received 29 March 2005; revised 28 April 2005; accepted 2 May 2005; published online 20 June 2005

Introduction In our previous study, suppression of tumourigenicity in the nasopharyngeal carcinoma (NPC) HONE1 cells was observed after the transfer of an additional intact human chromosome 11 by microcell-mediated chromosome transfer (MMCT). NPC tumour segregants derived from HONE1/chromosome 11 microcell hybrids (MCHs) exhibited a delayed latency period in tumour formation compared with that of the parental HONE1 cells (Cheng et al., 2000). The delayed tumour appearance may be associated with loss or inactivation of wildtype tumour suppressor genes (TSGs) from the normal donor chromosome 11. Based on this hypothesis, we have successfully refined a tumour suppressive region to 1.8 Mb at chromosome band 11q13 (Cheng et al., 2002) and three critical regions of 0.36, 0.44, and 0.3 Mb for tumour suppression at 11q22–23 (Lung et al., 2004). Differential expression analysis was performed for those non-tumourigenic MCHs and their matched tumourigenic segregant cell lines. One interesting gene was identified from this approach, THY1, which is located close to the previously defined 11q22-23 NPC critical region (Lung et al., 2004). It has been suggested that THY1 is associated with tumour suppression in human ovarian cancer (Abeysinghe et al., 2003). However, direct evidence that THY1 is a candidate TSG in ovarian cancer is still lacking (Abeysinghe et al., 2004). THY1 is a surface glycoprotein of 25–28 kDa (Bonewald et al., 1984), which is expressed on the cytoplasmic membrane of different cell types (Crawford and Barton, 1986). THY1 triggers a variety of cellular functions including proliferation, lymphokine release, differentiation, and apoptosis. Despite extensive investigation, the exact function and physiologic role of THY1 in the cell remains unknown (Abeysinghe et al., 2004). In this study, differential expression analysis identified on an oligonucleotide array containing nearly 19 000 human probes, was performed for four pairs of

THY1 in nasopharyngeal carcinoma HL Lung et al

6526

non-tumourigenic MCHs and their matched tumourigenic segregant cell lines. Reverse transcription–polymerase chain reaction (RT–PCR) and Western blot analyses were used to validate the microarray results. THY1 was identified as one of the differentially expressed genes and the tumour suppressive function of THY1 in NPC was examined. RT–PCR and Western blot analyses were used to study the gene and protein expression in HONE1 and three other NPC cell lines including HK1, CNE1, and HNE1 cells. Additional investigations were performed to determine whether the THY1 gene inactivation was attributed to promoter hypermethylation. A putative THY1 promoter region was identified by computer analysis and the methylation status of the THY1 promoter in the four NPC cell lines was examined by methylation-specific PCR (MSP). The clinical relevance of THY1 in NPC was examined in NPC biopsies and in tissue microarray (TMA) studies by RT–PCR and immunohistochemistry approaches. To examine the possible growth suppressive activity of THY1 in NPC, HONE1 cells were transfected with the wild-type THY1 gene and the colony formation assay was performed. A pCR3.1 expression vector and a tetracycline inducible expression vector, pETE-Bsd, were utilized in the THY1 functional studies.

Results Microarray analysis in HONE1 cells, chromosome 11 MCHs and their tumour segregants Gene expression profiles of non-tumourigenic chromosome 11 MCHs and their tumourigenic revertants were analysed by competitive hybridization on a 19 K oligomicroarray. In total, sixteen 19 K microarrays were screened: the recipient HONE1 cells versus four HONE1/ intact chromosome 11 MCHs (HK11.8, HK11.12, HK11.13, and HK11.19) and the four hybrids versus their tumour segregants (HK11.8-3TS, HK11.12-2TS, HK11.13-1TS, and HK11.19-4TS). Genes upregulated in MCHs after chromosome transfer and downregulated in the tumour revertants are presumably putative TSGs (Robertson et al., 1999). THY1 at 11q22.3–q23 was found to be differentially expressed under our selection criteria in three out of the four pairs of cell lines. The expression ratios of each duplicate and the mean ratios of the duplicates from the hybrid pairs are summarized in Table 1. A comparison of the expression profile of HONE cells with the four hybrids showed a nearly twofold increased expression of THY1 in all the MCHs except in HK11.13 (Table 1). On the other hand, the mRNA levels of THY1 were significantly decreased in all tumour segregants except HK11.13-1TS (Table 1). RT–PCR analysis of THY1 in HONE1 cells, chromosome 11 MCHs and their tumour segregants, NPC cell lines, and NPC biopsy specimens The differential expression of THY1 identified by the microarray study was verified by RT–PCR. The same Oncogene

Table 1 Microarray detection of THY1 in HONE1 versus MCHs and MCHs versus tumour segregants Cell lines 11.8/HONE1 11.8-3TS/11.8 11.12/HONE1 11.12-2TS/11.12 11.13/HONE1 11.13-1TS/11.13 11.19/HONE1 11.19-4TS/11.19

Expression ratioa 1.8702, 0.5686, 2.3329, 0.4578, 0.7835, 0.8892, 2.0286, 0.4820,

2.7533 0.5537 2.6434 0.4509 1.4379 1.058 1.9872 0.5696

Average 2.3 0.56 2.5 0.45 1.1 0.97 2.0 0.53

a Expression ratio is Cy5/Cy3, for dye-swap Cy3/Cy5 the expression ratio is the reciprocal

batch of RNAs used in microarray analysis was employed and quantification revealed a good correlation between the fold changes obtained in the microarray and RT–PCR (Figure 1a and Table 1). No expression of THY1 was detected in four NPC cell lines (HONE1, HK1, CNE1, and HNE1). THY1 was induced in HK11.8, HK11.12, and HK11.19, but not in HK11.13 (Figure 1a). When comparing the gene expression between the hybrids HK11.8, HK11.12, and HK11.19 and their tumour segregants, complete loss of mRNA expression of THY1 was observed in their tumour segregants (Figure 1a). THY1 was expressed in the mouse chromosome 11 donor cells, MCH556.15, and this cell line was used as a positive control for THY1. A9 cells were also included in the RT–PCR analysis; results showed that THY1 primers were specific for human chromosome 11 and do not amplify the murine sequences (Figure 1a). In order to investigate the clinical significance of THY1 in NPC, 17 NPC specimens were analysed for THY1 gene expression using RT–PCR. A total of 11/17 (65%) (samples 2, 3, 5, 6, 10, 11, 13, 14, 15, 16, 17) showed no or decreased THY1 expression, when compared with the immortalized normal nasopharyngeal epithelial cell line NP460 and MCH556.15 (Figure 1b). Protein analysis of THY1 in HONE1 cells, chromosome 11 MCHs and their tumour segregants, and other NPC cell lines The protein expression of THY1 in the same set of cell lines was analysed on Western blots. Similar to the findings from RT–PCR, HONE1, HK1, CNE1, and HNE1 did not express THY1 protein, but the protein was expressed in MCHs HK11.8, HK11.12, HK11.19, and the chromosome 11 donor cells (Figure 1c), whereas HK11.13 expressed a very low level of THY1 protein. The upper band in MCH556.15 belongs to the mouse A9 background, since the A9 cell lysate only shows the upper band in the Western blot (data not shown). The THY1 protein was downregulated to undetectable levels in all four tumour revertants. The results of the Western blot and RT–PCR data are concordant.


6527

Figure 1 (a) RT–PCR analysis of gene expression levels of THY1 in HONE1, MCH556.15, the four chromosome 11 MCHs and their tumour segregants, and three other NPC cell lines and an A9 mouse cell line. (b) THY1 gene expression analysis of NPC specimens. The immortalized normal nasopharyngeal epithelial cell line NP460 and MCH556.15 were used as positive controls. The DNA ladder (L) indicates the sizes of the PCR bands, which are shown on the right (bp). GAPDH served as an internal control. (c) Western blot analysis of THY1 protein in same cell lines in Figure 1a. The sizes of protein bands are indicated on the right. a-tubulin was used as the internal control

THY1 promoter methylation analysis in NPC cell lines To investigate the inactivation mechanism of THY1 in NPC, we studied the methylation status of THY1 promoter in four NPC cell lines by MSP analysis. Figure 2a shows amplification of methylated sequences in the HONE1, HK1, CNE1, and HNE1 cells, while an unmethylated signal was observed in HONE1, CNE1, and the donor cells. Subsequent re-expression of the gene in HONE1 cells was observed after treatment with the demethylation agent 5-aza-20 -deoxycytidine (Figure 2b). THY1 expression in NPC TMA The clinical relevance of THY1 protein expression in NPC was further analysed using NPC samples by immunohistochemistry on the TMA. Informative TMA cases were observed in 79 of the 100 cases. Since the staining index of THY1 expression in nine informative normal nasopharyngeal mucosas was no less than six, therefore, we designated the staining index of 6–9 as the normal expression of THY1 (Figure 3a and b). Staining index 1–4 was depicted as downregulated expression (Figures 3c and d), while staining index 0 was designated as loss of expression. The TMA expression data are summarized in Table 2. In 20 cases

of normal nasopharyngeal mucosa, all the nine informative cases showed normal expression of THY1 protein. However, in 70 informative NPC cases, 44% (31/70) cases showed downregulated expression of THY1 and 9% (6/70) revealed loss of expression of this protein. The frequency of downregulated expression of THY1 in metastatic neck nodes of NPC was 63% (17/27), which was significantly higher than that in primary NPC (33%) (14/43) (Po0.05, Table 2). THY1 transfection and colony formation assay Functional evidence of the growth suppressive activity of THY1 was assessed by colony formation assay. Four replicates of independent transfections were performed for HONE1 cells. A significant decrease in number of neomycin-resistant colonies occurred in the cell lines, which were transfected with the wild-type THY1 cDNA, as compared to vector alone (Figure 4a). There was a 90% decrease in colony number in the THY1 transfectants. To further test the inhibitory effect of THY1 on colony formation, a recently developed tetracyclineinducible transgene system (Protopopov et al., 2002) was utilized to study the functional role of THY1. HONE1-2 cells producing tetracycline trans-activator tTA were utilized for transfections. In the absence of doxycycline, the THY1 protein was expressed Oncogene


6528

Figure 2 (a) MSP analysis of THY1 promoter methylation in four NPC cell lines (HK1, CNE1, HNE1, and HONE1), MCH556.15, and the universal methylated DNA. Sizes of PCR products are shown on the right. The methylated allele was observed in all NPC cell lines. (b) Re-expression of THY1 in HONE1 cells after treatment with 5 mM 5-aza-20 -deoxycytidine was monitored by RT–PCR analysis

Figure 3 Immunohistochemical staining of THY1 in NPC TMA tissues (a–d). (a and b) Normal expression of THY1 with a staining index more than 6 was observed in 2 primary NPC specimens. (c) A lymph node metastatic NPC showed downregulation of THY1 expression with a staining index less than 6. (d) Downregulated expression of THY1 was observed in another lymph node metastatic NPC, as compared to its adjacent normal lymphocytes

Table 2 Expression of THY1 in primary and lymph node metastatic NPC tissues

Normal Primary NPC Metastatic NPC

Oncogene

Normal expression

Downregulated expression

Loss of expression

Total

9 (100%) 26 (60%) 7 (26%)

0 (0%) 14 (33%) 17 (63%)

0 (0%) 3 (7%) 3 (11%)

9 43 27

Po0.05

Po0.05

Not significant


6529

significant change of colony number was observed between the THY1 and vector-alone transfected HONE1-2 cells (Figure 4a). Negative controls without addition of vector DNA to HONE1-2 cells showed no expression of THY1 (Figure 4b) and no formation of colonies with or without doxycycline treatment (data not shown).

Discussion

Figure 4 Colony formation assay of THY1 transfectants. (a) HONE1 cells were transfected with pCR3.1-THY1 or pCR3.1, and HONE1-2 cells were transfected with pETE-Bsd-THY1 or pETEBsd in the presence or absence of doxycycline (dox). The colony forming ability was calculated by comparing the number of colonies in THY1 transfectants to that of vector alone. Each treatment was performed in quadruplicate. *Designates a significant difference from vector alone (Po0.005). (b) Western blot analysis of the THY1-transfected HONE1-2 cells and vector-alone transfectants. HONE1-2 cells were transfected in the absence of and the presence of dox. Western blotting was performed 3 days after transfection

(Figure 4b) and a significant reduction in number of blasticidin-resistant colonies was observed in HONE1-2 cells transfected with THY1 cDNA, as compared to vector alone (Figure 4a). The addition of doxycycline inhibited THY1 protein expression (Figure 4b). No

Differential gene expression levels were identified by microarray analysis of THY1 at 11q22.3–q23. The differential gene expression levels identified by microarray analysis of THY1 were validated by RT–PCR. RT–PCR showed undetectable THY1 levels in HONE1, HK11.13, and HK11.13-1TS after 45 PCR cycles. This may explain the inconsistent expression of THY1 in the HK11.13 hybrid pairs obtained from the microarray analysis. The genotyping of this particular hybrid in our previous studies (Lung et al., 2004) may provide an explanation for the absence of THY1 gene expression. The donor chromosome loci represented by microsatellite markers D11S908 and D11S2077, which map close to the THY1 physical location, are absent in HK11.13, consistent with the THY1 gene not being transferred to this particular hybrid or being lost after MMCT. As a result, the time to appearance of tumours of this particular clone was the shortest (25–35 days) and the in vivo tumour growth kinetics was the most rapid among the four MCHs (average 27.5–50 days) (Cheng et al., 2000). None of the tumourigenic NPC cell lines, including HONE1, HK1, CNE1, HNE1 cells and the tumour segregants, expresses THY1. The THY1 gene and protein expression is only observed in the donor cells and the HONE/chromosome 11 hybrids. Loss of THY1 expression in NPC cell lines may be attributed to the THY1 promoter hypermethylation, as at least one of the alleles was methylated in all four NPC cell lines and demethylation restored the THY1 gene expression in HONE cells. This suggests that THY1 may belong to the class of genes that predisposes to cancer through haploinsufficiency in the hemizygous state and, therefore, will not show a second mutation in the remaining wild-type allele in tumours (Fero et al., 1998; Tang et al., 1998). The in vitro colony formation assay provides functional evidence that introduction of wild-type THY1 is sufficient to inhibit the colony formation ability of THY1-null HONE1 cells using both the pCR3.1 and the inducible pETE-Bsd vector systems. The specificity of the growth suppressive effect of THY1 was supported by the absence of suppression in the presence of doxycycline, when the transgene was not expressed. The majority of NPC specimens show downregulated or lost THY1 gene and protein expression, while all normal nasopharyngeal mucosa samples showed normal expression of this protein in TMA analysis. The frequency of downregulated or loss of expression of Oncogene


6530

THY1 in metastatic lymph node NPC was 74% (20/27), which is significantly higher than that in primary tumours in the nasopharynx (40%). This suggests that the inactivation of THY1 protein expression might be associated with lymph node metastasis. THY1 has been reported to possess cell adhesion function in T cells adhering to the bone marrow stroma (Barda-Saad et al., 1999), and surface expression of THY1 increases focal adhesion in fibroblasts (Barker et al., 2004; DiazRomero et al., 2005). An increasing body of evidence now indicates that aberrant tumour cell adhesion is causally involved in tumour progression and metastasis (Cavallaro and Christofori, 2001). Besides regulation of focal adhesion, THY-1 surface expression also regulates cytoskeletal organization and migration by modulating the activity of p190 RhoGAP and Rho GTPase (Barker et al., 2004). Except for the initial THY1 association with ovarian cancer, its importance in other human cancers has yet to be reported. It is extremely significant that ours is the first report providing validation for the importance of THY1 in clinical specimens. The current study also provides the first reported evidence for its inactivation by hypermethylation. In addition, the growth suppressive function of THY1 in NPC was demonstrated with two different expression vector systems. Thus, these results provide direct evidence that THY1 is a good candidate TSG. Because of the strong association of THY1 with metastatic NPC, it may also be a good candidate biomarker associated with tumour metastasis and progression.

Materials and methods Cell lines The recipient NPC HONE1 cell line, donor chromosome 11 mouse hybrid cell line (MCH556.15), four HONE1/chromosome 11 MCH cell lines (HK11.8, HK11.12, HK11.13, and HK11.19), and their tumour segregants (HK11.8-3TS, HK11.12-2TS, HK11.13-1TS, and HK11.19-4TS) were used. Three other NPC cell lines (HK1, CNE1, and HNE1) were maintained as previously described (Lung et al., 2004). The NP460 is a telomerase-immortalized nasopharyngeal epithelial cell line established from primary epithelial outgrowth from nonmalignant nasopharyngeal biopsies. It was cultured in a 1 : 1 mixture of defined KSFM media (Invitrogen, Carlsbad, CA, USA) and EpiLife (Sigma, St Louis, MO, USA). NP460 is near diploid in karyotype and is non-tumourigenic, when injected into nude mice (manuscript submitted). Tissue specimens For THY1 gene expression analysis, 17 primary NPC biopsies were collected from Queen Mary Hospital in 2001. For preparation of the NPC TMA, 80 NPC specimens and 20 specimens of nasopharyngeal mucosa from nonNPC diseases were collected in the Cancer Institute of Sun Yat-Sen University between 1997 and 2000. Ages of the 80 carcinoma patients ranged from 14 to 72 years (mean, 47 years old). The male and female ratio is 4.3 : 1. The tumour specimens encompassed 50 primary NPC cases without metastasis and 30 lymph node metastatic NPC. All NPC cases Oncogene

selected in this study were poorly differentiated squamous cell carcinomas. Total RNA isolation Total RNA was extracted from the cell lines using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA). Total cellular RNA from NPC biopsies was prepared by the TRIzol reagent (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s manual. Gene expression analysis using oligonucleotide microarray hybridization A human oligonucleotide library (Sigma-Genosys, Woodlands, TX, USA) containing synthesized 60-mers of 18 912 genes was supplemented with additional 284 custom-designed oligonucleotide probes representing approximately 250 known and predicted genes (http://www.ensembl.org/) that mapped to the 11q13 and 11q22 regions. These oligonucleotides were spotted on glass slides using a custom-built microarray spotter at the Genome Institute of Singapore. The incorporation of Cy3 and Cy5 to cDNAs, competitive hybridizations, and processing of array images were performed as described in Lin et al. (2004). The gene expression analysis was carried out using analytical tools on the GIS Microarray Database (GIS MAdb, http://gismadb.gis.a-star.edu.sg). Stringency was set so that only data sets with expression ratios higher than 1.4 in hybrids/HONE1 cells and ratios less than 0.6 in hybrids/tumour segregants for both duplicates were selected for further studies. Reverse transcription–polymerase chain reaction In all, 1 mg of total RNA was reverse transcribed with 200 U of SuperScriptII reverse transcriptase (Invitrogen, Carlsbad, CA, USA). For THY1 cDNA PCR amplifications, sense and antisense oligonucleotides were designed with the Primer3 primer design program (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) and the sequences were 50 -GAC CCGTGAGACAAAGAAGC-30 and 50 -GCCCTCACACTT GACCAGTT-30 . Details of the procedures performed, are described previously (Lung et al., 2004). Identification of the putative THY1 promoter The putative THY1 promoter was identified based on the study of Qiu et al. (2004). To define the THY1 promoter, the upstream sequence of the THY1 exon 1 was retrieved from UCSC Genome Browser (http://genome.ucsc.edu, extracted fragment NT_033899). A moderately GC-rich putative THY1 promoter was identified by computer analysis (http://www. genomatix.de), which is 718 bp, located at 22 837 541– 22 838 258 of fragment NT_033899. The 30 end region spanning this putative promoter fulfilled the criteria of a CpG island (16): CpG island starts at 402 bp and ends at 714 bp; GC content, 70%; observed/expected CpG ratio, 0.602; a total of 22 CpG sites in a 313 bp region. Methylation-specific PCR (MSP) analysis The design of primer pairs for methylated and unmethylated DNA was based on MethPrimer (www.urogene.org/methprimer) primer design program using guidelines for MSP primer selection (Li and Dahiya, 2002). The sequences were 50 TATTTTTATATTAATGCGGGATCGT-30 and 50 -CGATT ACTACACCCAACTCGAA-30 for methylation primers and 50 -TTATTTTTATATTAATGTGGGATTGT-30 and 50 -TCC


6531 AATTACTACACCCAACTCAAA-30 for unmethylated primers. In brief, 1 mg genomic DNA was bisulfite-modified using the CpGenomet DNA Modification Kit (Chemicon International, Inc., Temecula, CA, USA) and modified DNA was subjected to PCR. CpGenomet Universal Methylated DNA (Chemicon International, Inc., Temecula, CA, USA) was used as a positive control for methylated DNA. For each PCR, negative controls without DNA templates were included. 5-Aza-20 -deoxycytidine treatment HONE1 cells were treated with 5 mM 5-aza-20 -deoxycytidine (Sigma, St Louis, MO, USA), as previously described (Lung et al., 2004). Western blot analysis Viable NPC cells at 70–80% confluency (5 106) were harvested and cellular protein was extracted and Western blot performed as previously described (Chan and Lung, 2004). In brief, samples of 20 mg cellular protein were fractionated and transferred to polyvinylidene difluoride membranes (pore size: 0.45 mm; Millipore, Billerica, MA, USA). Primary antibody incubation was performed with 1 : 2500 H-110 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for the THY1 protein, and 1 : 5000 Ab-1 (Calbiochem, Darmstadt, Germany) for a-tubulin. The signals were visualized by enhanced chemiluminescence method according to the manufacturer’s instructions (Amersham, Uppsala, Sweden). NPC tissue microarray For construction of the NPC TMA, one sample was selected from each NPC case (50 primary carcinomas and 30 metastatic NPC) and 20 cases of normal nasopharyngeal mucosa, as previously described (Xie et al., 2003). Multiple sections (5 mm thick) were cut from the TMA block and mounted on microscope slides.

A semiquantitative scoring criterion for IHC was used, in which both staining intensity and positive areas were recorded. A staining index (values 0–9) obtained as the intensity of THY1-positive staining (negative ¼ 0, weak ¼ 1, moderate ¼ 2, or strong ¼ 3 scores) and the proportion of immunopositive cells of interest (o10% ¼ 1, 10–50% ¼ 2, >50% ¼ 3 scores) were calculated. Gene transfection and colony formation assay The full-length wild-type THY1 cDNA flanked with BamHI restriction sites (Abeysinghe et al., 2004) was ligated to the pCR3.1 neomycin-resistance-tagged expression plasmid. pCR3.1-THY1 recombinant and control pCR3.1 (Invitrogen, Carlsbad, CA, USA) vector-alone plasmids were transfected into HONE1 cells with Lipofectamine 2000 Reagent (Invitrogen, Carlsbad, CA, USA). A total of 4 105 cells were seeded in six-well plates and were transfected with 1 mg of plasmid DNA for 5 h. The cells were subsequently split into 100 mm dishes. After 14 days of selection in DMEM/10%FCS containing 400 mg/ml neomycin, colonies were fixed and stained with Giemsa to assess the transfection efficiency. Construction of a pETE-Bsd responsive vector and a HONE1 cell line, HONE1-2, producing the tetracycline trans-activator tTA, was described in Protopopov et al. (2002). The full-length THY1 cDNA was inserted into BamHI-digested pETE-Bsd plasmid containing a selectable blasticidin-resistance gene. pETE-Bsd-THY1 and control pETE-Bsd vector were transfected into the HONE1-2 cells, as described above. After 14 days of selection in DMEM/ 10%FCS containing 5 mg/ml blasticidin and 400 mg/ml neomycin with or without 0.5 mg/ml doxycycline, colonies were fixed and stained. Statistical analysis

Immunohistochemical (IHC) staining

The w2 and Fisher’s exact tests were used for analysis of significant differences in THY1 expression level detected by TMA between primary and metastatic NPC, and in colony numbers between control and THY1-transfected cells. Differences were considered statistically significant for P-values o0.05.

Immunohistochemistry studies were performed using the standard streptavidin-biotin-peroxidase complex method as described (Xie et al., 2003). The TMA slides were incubated with rabbit anti-THY1 polyclonal antibody (H-110, Santa Cruz Biotechnology, Santa Cruz, CA, USA, 1 : 500 dilution) overnight. The slides were then incubated with a biotinylated goat anti-rabbit serum for 30 min and subsequently reacted with a streptavidin–peroxidase conjugate and 30 -30 diaminobenzidine. For the negative control, the primary antibody was replaced with normal rabbit IgG.

Acknowledgements We acknowledge the financial support from the Research Grants Council of the Hong Kong Special Administrative Region, People’s Republic of China: Grant numbers HKUST 6113/01M, CA99/00.SC02, and CA03/04.SC01 to MLL, the Agency for Science Technology and Research, Singapore to EL, and the Swedish Cancer Society, the Swedish Research Council, STINT, the Swedish Institute, the Royal Swedish Academy of Sciences and INTAS to ERZ.

References Abeysinghe HR, Cao Q, Xu J, Pollock S, Veyberman Y, Guckert NL, Keng P and Wang N. (2003). Cancer Genet. Cytogenet., 143, 125–132. Abeysinghe HR, Pollock SJ, Guckert NL, Veyberman Y, Keng P, Halterman M, Federoff HJ, Rosenblatt JP and Wang N. (2004). Cancer Genet. Cytogenet., 149, 1–10. Barda-Saad M, Rozenszajn LA, Ashush H, Shav-Tal Y, Ben Nun A and Zipori D. (1999). Exp. Hematol., 27, 834–844. Barker TH, Grenett HE, MacEwen MW, Tilden SG, Fuller GM, Settleman J, Woods A, Murphy-Ullrich J and Hagood JS. (2004). Exp. Cell Res., 295, 488–496.

Bonewald L, Ades EW, Tung E, Marchalonis JJ and Wang AC. (1984). J. Immunogenet., 11, 283–296. Cavallaro U and Christofori G. (2001). Biochim. Biophys. Acta., 1552, 39–45. Chan KT and Lung ML. (2004). Cancer Chemother. Pharmacol., 53, 519–526. Cheng Y, Chakrabarti R, Garcia-Barcelo M, Ha TJ, Srivatsan ES, Stanbridge EJ and Lung ML. (2002). Genes Chromosomes Cancer, 34, 97–103. Cheng Y, Stanbridge EJ, Kong H, Bengtsson U, Lerman MI and Lung ML. (2000). Genes Chromosomes Cancer, 28, 82–91. Oncogene


6532 Crawford JM and Barton RW. (1986). Lab. Invest., 54, 122–135. Diaz-Romero J, Gaillard JP, Grogan SP, Nesic D, Trub T and Mainil-Varlet P. (2005). J. Cell Physiol., 202, 731–742. Fero ML, Randel E, Gurley KE, Roberts JM and Kemp CJ. (1998). Nature, 396, 177–180. Li LC and Dahiya R. (2002). Bioinformatics, 18, 1427–1431. Lin CY, Strom A, Vega VB, Kong SL, Yeo AL, Thomsen JS, Chan WC, Doray B, Bangarusamy DK, Ramasamy A, Vergara LA, Tang S, Chong A, Bajic VB, Miller LD, Gustafsson JA and Liu ET. (2004). Genome Biol., 5, R66. Lung HL, Cheng Y, Kumaran MK, Liu ET, Murakami Y, Chan CY, Yau WL, Ko JM, Stanbridge EJ and Lung ML. (2004). Int. J. Cancer, 112, 628–635.

Oncogene

Protopopov AI, Li J, Winberg G, Gizatullin RZ, Kashuba VI, Klein G and Zabarovsky ER. (2002). J. Gene Med., 4, 397–406. Qiu GH, Tan LK, Loh KS, Lim CY, Srivastava G, Tsai ST, Tsao SW and Tao Q. (2004). Oncogene, 23, 4793–4806. Robertson GP, Huang HJ and Cavenee WK. (1999). Mol. Cell Biol. Res. Commun., 2, 1–10. Tang B, Bottinger EP, Jakowlew SB, Bagnall KM, Mariano J, Anver MR, Letterio JJ and Wakefield LM. (1998). Nat. Med., 4, 802–807. Xie D, Sham JS, Zeng WF, Lin HL, Che LH, Wu HX, Wen JM, Fang Y, Hu L and Guan XY. (2003). Int. J. Cancer, 107, 896–902.