Identification of tumor suppressor genes (TSG) silenced by methylation uncovers mechanisms of tumorigenesis and identifies new epigenetic tumor markers for early cancer detection. Both nasopharyngeal carcinoma (NPC) and esophageal carcinoma are major tumors in Southern China and Southeast Asia. Through expression subtraction of NPC, we identified Deleted in Liver Cancer 1 (DLC1)/ARHGAP7 (NM_006094) – an 8p22 TSG as a major downregulated gene. Although expressed in all normal tissues, DLC1 was silenced or downregulated in 11/12 (91%) NPC, 6/15 (40%) esophageal, 5/8 (63%) cervical and 3/9 (33%) breast carcinoma cell lines. No genetic deletion of DLC1 was detected in NPC although a hemizygous deletion at 8p22–11 was found by 1-Mb array-CGH in some cell lines. We then located the functional DLC1 promoter by 5′-RACE and promoter activity assays. This promoter was frequently methylated in all downregulated cell lines and in a large collection of primary tumors including 89% (64/72) NPC (endemic and sporadic types), 51% (48/94) esophageal, 87% (7/8) cervical and 36% (5/14) breast carcinomas, but seldom in paired surgical marginal tissues and not in any normal epithelial tissue. The transcriptional silencing of DLC1 could be reversed by 5-aza-2′-deoxycytidine or genetic double knock-out of DNMT1 and DNMT3B. Furthermore, ectopic expression of DLC1 in NPC and esophageal carcinoma cells strongly inhibited their colony formation. We thus found frequent epigenetic silencing of DLC1 in NPC, esophageal and cervical carcinomas, and a high correlation of methylation with its downregulation, suggesting a predominant role of epigenetic inactivation. DLC1 appears to be a major TSG implicated in the pathogenesis of these tumors, and should be further tested as a molecular biomarker in patients with these cancers.
Nasopharyngeal carcinoma (NPC) is an upper respiratory tract tumor endemic in Southern China and Southeast Asia, including Hong Kong and Singapore. It is rare in other parts of the world including the US, except for an intermediate incidence in some North African nations (sporadic NPC) (reviewed by Tao et al. (2006)). This striking geographical and ethnic distribution suggests a strong association with specific genetic and environmental factors (Hildesheim and Levine, 1993). Although the molecular basis of NPC development is still poorly elucidated, NPC pathogenesis has been shown to be strongly associated with Epstein–Barr virus (EBV), whether endemic or sporadic (Pathmanathan et al., 1995; Raab-Traub 2002). Consumption of salted fish in early childhood, tobacco smoking (Yu and Yuan 2002) and genetic predisposition (Lu et al., 1990; Feng et al., 2002; Xiong et al., 2004) have also been suggested as risk factors.
Previous molecular studies of NPC revealed multiple genetic alterations on 3p, chromosome 9, 11q, 12q, 13q, 14q and 16q (Chen et al., 1999; Lo et al., 2000; Tao et al., 2006). Searches for putative tumor suppressor genes (TSG) in these regions have identified only several candidates, like p16INK4a on 9p (Lo et al., 1995; Gulley et al., 1998). Genetic mutations of known TSGs, including TP53 and RB1, are infrequent in NPC (Effert et al., 1992; Spruck, III et al., 1992; Sun et al., 1992). Meanwhile, epigenetic abnormalities, alone or together with genetic alterations, have been shown to be frequently involved in NPC, such as the promoter methylation of BLU and RASSF1A at 3p21 (Qiu et al., 2004; Zhou et al., 2005), TSLC1 on 11q (Hui et al., 2003), GADD45G on 9q22 (Ying et al., 2005) and PCDH10 (Ying et al., 2006). These limited findings suggest that additional cancer-related genes are yet to be identified in NPC in the reported regions, or in other unidentified loci.
To identify epigenetically silenced candidate TSGs genome-wide in NPC, we used suppression subtractive hybridization (SSH) to screen for downregulated genes and identified Deleted in Liver Cancer 1 (DLC1) (NM_006094, also known as ARHGAP7, FLJ21120, HP, STARD12 or p122-RhoGAP). DLC1 was firstly identified as a candidate TSG for hepatocellular carcinoma and mapped to 8p22–p21.3 (Yuan et al., 1998). It has high sequence similarity (86%) with the rat p122RhoGAP, a GTPase-activating protein for the Rho family of proteins involved in cell cytoskeleton organization and other functions (Homma and Emori, 1995; Etienne-Manneville and Hall, 2002; Moon and Zheng, 2003). More recent studies have demonstrated that DLC1 is an important TSG, being inactivated by aberrant promoter methylation in multiple carcinomas, including liver, colon, gastric, lung, breast and prostate carcinomas (Kim et al., 2003; Plaumann et al., 2003; Wong et al., 2003; Yuan et al., 2003a, 2003b, 2004), while no study has been performed for NPC yet. Here, we report on the epigenetic alterations of DLC1 in NPC and other tumors (esophageal and cervical carcinomas) which have not been studied for this gene. We also assessed the gene's ability to inhibit NPC and esophageal carcinoma cell colony formation.
Identification of DLC1 as a major downregulated gene in NPC but broadly expressed in normal tissues
Poly A+ RNA samples of a well-characterized NPC cell line, CNE-2 (Qiu et al., 2004), and normal adult upper respiratory tissue (trachea) were subtracted for differentially expressed genes using SSH. Normal trachea was chosen as a control since it is readily available and anatomically close to the site of NPC (nasopharynx). In previous experiments (Qiu et al., 2004; Ying et al., 2005; Ying et al., 2006), we confirmed that the gene expression pattern of trachea is similar to that of nasopharynx. Subtracted SSH-PCR products with different sizes were observed in gels between the subtracted and unsubtracted cDNA populations in all three experiments, indicating that the SSH was effective. After cloning the subtracted cDNA, randomly chosen clones, representing candidate genes possibly downregulated in CNE-2, were identified by sequencing and BLAST analysis.
One of the most frequently identified genes, represented in 5% of all clones, was DLC1 (NM_006094), encoding a RhoGAP-domain containing protein. DLC1 was reported to be a functional TSG frequently inactivated by promoter methylation in multiple carcinomas (Kim et al., 2003; Plaumann et al., 2003; Wong et al., 2003; Yuan et al., 2003a, 2003b, 2004). DLC1 is located at 8p22 – a region where we also detected heterozygous deletion in some esophageal cell lines by conventional CGH (Tang et al., 2001), but no study of DLC1 has been performed in esophageal carcinoma yet. Thus, in this study, we included an investigation of DLC1 alteration in other common cancers including esophageal, cervical and breast carcinomas.
We then examined the genetic deletion of DLC1 in our completed high-resolution (1-Mb) array-CGH data obtained with the Wellcome Trust Sanger Institute whole-genome arrays (www.sanger.ac.uk/Projects/Microarrays) (Hurst et al., 2004; Ying et al., 2006). Although a hemizygous deletion at 8p22–11 was detected in 6/11 (55%) NPC and 4/10 (40%) esophageal cancer cell lines, DLC1 itself was actually not located within the deleted region in most cell lines including CNE-2, except for two esophageal cell lines (Seng, Ying and Tao, unpublished). Further deletion analyses of DLC1 using differential-PCR (Qiu et al., 2004) in genomic DNA from several silenced cell lines and NPC tumors did not reveal any homozygous deletion of DLC1. These results indicate that epigenetic but not genetic defect is the dominant mechanism of DLC1 downregulation in NPC.
We then examined the expression profile of DLC1 in human normal adult and fetal tissues by semi-quantitative reverse transcriptase–polymerase chain reaction (RT-PCR). Expression of DLC1 was readily detected in all normal tissues including nasopharynx and esophagus (Figures 1a, 2c), which are the corresponding normal tissues for NPC and esophageal carcinoma.
Functional location of the DLC1 promoter
Although DLC1 methylation has been well studied in several tumors, its promoter has not been located functionally. The DLC1 sequence upstream of the exon 1 (variant 2, NM_006094) was retrieved from the NCBI database and analysed. We found that the region spanning the putative promoter and exon 1 is a typical CpG-island (Gardiner-Garden and Frommer 1987): GC content, 65.5%; observed/expected CpG ratio, 0.76; with a total of 84 CpG sites in a 756-bp region (Figure 1b). Several potential transcription factor (TF)-binding sites, including those for E2F, STAT, p53, Sp1 and heat-shock factor (HSF), were predicted in this putative promoter using the TF search programs TFSEARCH (www.cbrc.jp/research/db/TFSEARCH) and MotifSearch (motif.genome.jp) (Figure 1c), indicating that this promoter could be p53-, STAT-, Sp1- and E2F-regulated and might even respond to stress stimuli.
We further determined the transcription start sites of DLC1 transcripts (NM_006094) using 5′-RACE. The results showed that the start sites in different normal tissues (trachea, lung and heart) were different from the published data (NM_006094), but still located downstream of the putative DLC1 promoter (Figure 1b). However, these different transcription start sites would not affect the open reading frame of the DLC1 transcripts.
To further assess whether the putative DLC1 promoter could function to drive transcription, three regions of the promoter were cloned and linked to a luciferase reporter construct. Promoter activities of these fragments were assessed by transient transfection in DLC1-expressing or -silent cell lines, including two NPC (C666-1 and HK1), two cervical carcinoma (C33A) and one Hodgkin lymphoma (HD-MY-Z) cell line. All three constructs could drive the transcription of target gene, although at different efficiencies, with the largest fragment (−1926/+152) having the highest activity, followed by the second largest (-1198/+152) and the smallest (−480/+152) (Figure 1c), indicating that the putative DLC1 promoter is indeed functional in cells.
Frequent silencing of DLC1 by methylation in cell lines of NPC, esophageal and cervical carcinomas
We next examined DLC1 (NM_006094) expression and the methylation status of the identified promoter in a panel of carcinoma cell lines not studied before for DLC1. Absent or low DLC1 expression was detected in 11/12 (91%) of NPC, 6/15 (40%) of esophageal, 5/8 (63%) of cervical and 3/9 (33%) of breast carcinoma cell lines (Figure 2a and 2b). Meanwhile, DLC1 was readily detectable in five immortalized normal epithelial cell lines (NP69, NE1, NE3, FK16A and FK16B) and three normal nasopharyngeal tissue samples (Figure 2c).
The silencing of DLC1 was not due to the lack of transcriptional factors since the transfected exogenous DLC1 promoter was still active in a DLC1-silenced cell line (HK1), as in other DLC1-expressing cell lines (Figure 1c). Therefore, we examined the role of promoter methylation in the silencing of DLC1. Using methylation-specific PCR (MSP), DLC1 methylation was detected in all the cell lines with downregulated or silenced expression, while unmethylated promoter was detected in expressing cell lines (Figure 2a and 2b). Notably, all of the NPC cell lines except for one had methylated DLC1. None of the normal control cell lines or normal nasopharyngeal and esophageal tissues showed methylation (Figure 2c).
We further validated the MSP results by high-resolution methylation analysis on every CpG site in the identified DLC1 promoter (Figure 1b), using bisulfite genomic sequencing (BGS). The result confirmed the MSP analysis (Figure 2d). Thus, a high correlation between downregulation and DLC1 methylation was observed in tumor cell lines of multiple tissue types. Interestingly, three small regions containing CpG sites (CpG site #10, #16–18, #34) were much less frequently methylated. These regions all contain Sp1-binding elements, indicating that Sp1 might physically protect these binding sites from methylation, as we observed previously for the Q promoter of EBV (Tao et al., 1998, 2006).
Pharmacologic and genetic demethylation restores DLC1 expression
To determine whether methylation directly mediates the silencing of DLC1, two NPC cell lines, CNE-1 and CNE-2, were treated with 5-aza-2′-deoxycytidine (Aza), a DNA methyltransferase inhibitor. After the treatment, DLC1 expression levels were dramatically increased, along with obvious decrease in methylated alleles of DLC1 (Figure 3a). High-resolution methylation analysis by BGS on Aza-treated cell lines confirmed the significant demethylation of the promoter (Figure 3b). These results demonstrate that CpG methylation of the DLC1 promoter directly silences its expression in NPC. As a control, NPC cell line C666-1 with unmethylated DLC1 promoter was also treated with Aza but no significant increase of DLC1 expression was detected (Figure 3a), indicating that Aza does not lead to increased DLC1 expression through demethylation-independent mechanisms.
DLC1 could also be activated in the completely methylated colorectal cancer cell line HCT116, by genetic demethylation through double knock-out of both DNMT1 and DNMT3B (DKO cell line) (Rhee et al., 2002) (Figure 3c). Concomitantly, unmethylated DLC1 alleles were strongly induced in these DKO cells. Trace DLC1 expression and some unmethylated alleles were also detected in HCT116 cells with the single knockout of DNMT1 (1KO), but no expression and no unmethylated allele were detected in cells with the single knockout of DNMT3B (3BKO).
Frequent DLC1 methylation in multiple primary carcinomas
We next studied DLC1 methylation in a large collection (n=188) of primary tumors, including NPC, esophageal and cervical carcinomas. Fourteen breast carcinomas were also included since previous studies of DLC1 methylation in this cancer was mainly performed on cell lines (Yuan et al., 2003a). We detected DLC1 methylation in 61/68 (90%) endemic NPC tumors, all of the three nude mice-passaged undifferentiated NPC tumors from North Africa (C15, C17 and C18), and 3/4 sporadic NPC tumors in Caucasian patients from Baltimore, US (Figure 4a), with no methylation detected in any normal nasopharyngeal tissue (Figure 2c). Forty-eight out of 94 (51%) esophageal carcinomas were methylated (one additional tumor had only weak methylation), while only two of the 94 paired surgical marginal tissues had weak methylation (Figure 4b) and none of the eight normal esophageal tissues showed methylation (Figure 2c). Furthermore, DLC1 was methylated in 5/14 (36%) of primary breast carcinomas (Figure 4c), and 7/8 (87%) of primary cervical carcinomas (Figure 4d). These results clearly demonstrate that DLC1 methylation is tumor-specific, and that the methylation frequencies of these primary tumors were similar to their corresponding cell lines.
DLC1 inhibits the colony formation of NPC and esophageal carcinoma cells
To evaluate the role of DLC1 as a TSG in carcinomas, we examined the inhibitory effect of ectopic DLC1 expression on tumor cells. An expression vector with a full-length DLC1 gene was transfected into NPC (CNE-1 and HK1) and esophageal carcinoma cell line EC109 which previously showed virtually no DLC1 expression, with TP53 as a positive control. The efficiency of colony formation of transfected cells was evaluated by monolayer culture. A significant reduction of colony formation efficiencies was observed in cells transfected with pcDNA3.1(+)/DLC1 (down to 14% of vector control) or pcDNA3.1(+)/TP53 (down to 33% of vector control), compared to pcDNA3.1(+) vector only (Figure 5). Thus, DLC1 indeed has growth inhibitory activity and can function as a tumor suppressor for NPC and esophageal carcinoma, as previously reported for other carcinomas.
To identify candidate TSGs genome-wide in NPC, we used SSH to screen for downregulated genes in this tumor and identified DLC1 (NM_006094) to be a major downregulated gene. DLC1 promoter methylation strongly correlated with its downregulation in NPC cell lines. We further showed that DLC1 is frequently inactivated by promoter methylation not only in both endemic and sporadic NPC tumors, but also in additional common cancers such as esophageal and cervical carcinomas. We also showed that DLC1 functions as a TSG in NPC and esophageal carcinoma cells. As DLC1 methylation has not been reported in these three tumors before, our results add more candidates to the reported list of cancers (liver, breast, colon, gastric, lung and prostate) with epigenetic abnormalities of this TSG (Kim et al., 2003; Plaumann et al., 2003; Wong et al., 2003; Yuan et al., 2003a , 2003b, 2004). Interestingly, in one of our independent studies screening for downregulated genes in NPC cell lines using whole-genomic microarray expression profiling combined with pharmacologic demethylation, DLC1 was also identified as one of the downregulated target genes (Tao et al., manuscript in preparation).
DLC1 was firstly identified by Yuan et al.(1998) and mapped to 8p22–p21.3 . It has a high sequence similarity (86%) with the rat p122RhoGAP, a GTPase-activating protein for the Rho family of proteins (Homma and Emori, 1995). A RhoGAP domain is present at the carboxyl terminal of the DLC1 protein. The RhoGAP domain catalyses the conversion of active GTP-bound Rho proteins into inactive GDP-bound proteins. The Rho proteins, members of the Ras superfamily, are involved in diverse cellular functions, including cell actin cytoskeleton organization, regulation of transcription and cell cycle, and Ras-mediated oncogenic transformation (Wang et al., 1997; Sahai et al., 2001; Etienne-Manneville and Hall, 2002; Moon and Zheng, 2003). Downregulation of DLC1 may thus activate Ras-mediated signaling pathways, contributing to oncogenesis (Wang et al., 1997; Sahai et al., 2001; Etienne-Manneville and Hall, 2002; Moon and Zheng, 2003).
Compared to other common carcinomas like lung, colon, breast and prostate cancers, NPC is less well studied and its molecular pathogenesis is less well understood, with only few TSGs identified so far in this tumor (Tao et al., 2006). Previous searches for TSGs in NPC mainly used the classic positional cloning approach, focusing on chromosomal regions with gross genetic abnormalities (deletions) detected by low-resolution microsatellite or CGH analyses. Limited success has been achieved this way, implying that additional TSGs exist on regions without gross abnormalities. Meanwhile, other studies using monochromosome transfer and functional analysis have also identified additional NPC TSG loci/candidates, including 3p21.3 (Cheng et al., 1998) and 11q13 (Cheng et al., 2002). Our laboratory has been trying to identity NPC-related TSGs using an integrated genomic and epigenetic approach. In this study, we focused on a downregulated TSG, DLC1 on 8p22, identified by expression subtraction. 8p22 has not been reported previously to have genetic aberrations in NPC. Our recent high-resolution 1-Mb array-CGH study revealed a hemizygous deletion at 8p22–11 in around half of the NPC cell lines examined, however, this deletion does not include the DLC1 gene. The present study shows evidence that DLC1 is frequently inactivated epigenetically in NPC. Furthermore, despite a difference in incidence of endemic and sporadic NPC which suggests possibly different pathogenic mechanisms (Hildesheim and Levine 1993; Raab-Traub 2002; Yu and Yuan 2002), DLC1 methylation is detected in both types of NPC, indicating that the epigenetic inactivation of DLC1 is a common and perhaps necessary step during NPC tumorigenesis.
DLC1 has previously been reported as a functional TSG by colony formation assays in breast, liver and lung cancer cell lines. We also showed that DLC1 functions as a TSG in NPC and esophageal carcinoma cells by colony formation assays. In agreement with previous studies (Plaumann et al., 2003; Yuan et al., 2003b; Teramoto et al., 2004), we also detected DLC1 methylation in a fraction of breast carcinoma cell lines and primary tumors from Southeast Asians. We also noticed that DLC1 is not methylated in any of the immortalized cell lines (immortalized by SV40 T-antigen or HPV E6/E7, or HPV), indicating that DLC1 methylation is not an early event during carcinogenesis.
Our results also showed that the transcription start sites of DLC1 (NM_006094) are different in different normal tissues. However, all the transcripts generated still encode the same protein sequence. Thus, the usage of different transcription start sites would not affect the normal physiologic function of DLC1, or even the transcription efficiencies as shown by our semi-quantitative RT–PCR (Figure 1a). DLC1 also has multiple isoforms and alternative promoters (Wilson et al., 2000), and the present study focused on the transcript variant 2 (NM_006094). Further studies on DLC1 in tumorigenesis need to be carried out on its alternative splicing variants and promoter usages, their possibly different functions, and their genetic mutations. Our preliminary data show that the expression patterns of DLC1 isoforms do vary in both normal and tumor samples, and we have also identified a novel isoform of DLC1 from an alternative promoter (manuscript in preparation).
In summary, we found that DLC1 (NM_006094) is frequently inactivated by promoter methylation in multiple tumors including NPC, esophageal and cervical carcinomas. We also showed that DLC1 could function as a tumor suppressor in NPC and esophageal carcinoma cells. Our results suggest that epigenetic inactivation of DLC1 is one of the important steps underlying the tumorigenesis of NPC, esophageal and cervical carcinomas, and add to the evidence that DLC1 is a general tumor suppressor for multiple tumor types. The high incidence of epigenetic inactivation of DLC1 in NPC, esophageal and cervical carcinomas indicates that DLC1 methylation may be a good molecular marker and epigenetic therapeutic target for these tumors.
Materials and methods
Cell lines, tumor and normal tissue samples
Cell lines used were: 12 NPC (C666-1, CNE-1, CNE-2, HK1, HNE1, HNE2, HNE3, HONE1, TW-01, BM1, 5-8F and 6-10B) (Glaser et al., 1989; Qiu et al., 2004; Ying et al., 2005), 15 esophageal (EC1, EC18, EC109, HKESC1, HKESC2, SLMT-1, KYSE30, KYSE70, KYSE140, KYSE150, KYSE180, KYSE270, KYSE410, KYSE510 and KYSE520) (Tang et al., 2001; Ying et al., 2006), nine breast (MCF7, ZR-75-1, MB-231; T47D, MB-468, MB-435, SKBR3, BT549, and YCC-B2) (Kim et al., 2005; Ying et al., 2005) and eight cervical carcinoma (HeLa, Caski, C33A, SiHa, 808, 866, 879 and 915) (Steenbergen et al., 2004; Ying et al., 2006). NP69, an SV40 T-antigen-immortalized nasopharyngeal epithelial cell line with many features of normal nasopharyngeal epithelial cells was used as a ‘normal’ control for NPC (Tsao et al., 2002). Two immortalized normal esophageal epithelial cell lines (NE1, NE3) (Deng et al., 2004; Ying et al., 2006) were used as ‘normal’ controls for esophageal carcinoma. Two HPV-immortalized human normal foreskin keratinocyte cell lines (FK16A at passage 88 and FK16B at passage 70) (Steenbergen et al., 2004) were also used as controls. These cell lines were routinely maintained in RPMI or DMEM medium, or the Keratinocyte Growth Medium (for FK16A and FK16B only) (Steenbergen et al., 1996). Total RNA and DNA were extracted from cell pellets using TRI Reagent (Molecular Research Centre Inc., Cincinnati, OH, USA) as reported previously (Tao et al., 2002).
DNA and RNA samples were extracted from 68 primary endemic NPC (Southern China and Southeast Asia Chinese), 14 Asian Chinese breast carcinomas (Tao et al., 1998, 1999; Murray et al., 2004; Qiu et al., 2004; Ying et al., 2005; Zhou et al., 2005), eight cervical carcinomas (Steenbergen et al., 2004), and three normal nasopharyngeal tissues (Srivastava et al., 2000). We also collected four sporadic Caucasian NPC from the Department of Otolaryngology-Head and Neck Surgery, John Hopkins Hospital, Baltimore, which together with three nude mice-passaged undifferentiated NPC tumors derived from North Africans (C15, C17, C18) (Busson et al., 1988), acted as comparison for the endemic type NPC from Southern China and Southeast Asia. A total of 94 esophageal carcinomas of squamous cell carcinoma type from Hong Kong Chinese and their corresponding surgical marginal tissues and eight normal esophageal tissues from healthy individuals (Wong et al., 2006), were collected in the Department of Pathology, University of Hong Kong. Human normal adult and fetal tissue RNA samples were purchased commercially (Tang et al., 2001; Ying et al., 2005, 2006).
Suppression subtractive hybridization
Poly A+ RNA from the CNE-2 cell line was isolated using MessageMaker Reagent Assembly (Invitrogen, Carlsbad, CA, USA). Human normal trachea poly A+ RNA was purchased from Clontech (Clontech, Palo Alto, CA). Suppression subtractive hybridization (SSH) was performed using the PCR-Select cDNA Subtraction Kit (Clontech), according to the manufacturer's protocol, by comparing the trachea poly A+ RNA (as tester) to the CNE-2 poly A+ RNA (as driver). Differentially expressed cDNA sequences between these two RNA populations, representing genes downregulated in CNE-2, were further amplified by nested PCR, and TA-cloned into pCR®2.1-TOPO vector (Invitrogen). The experiment was performed three times independently.
Reverse transcriptase–polymerase chain reaction
RT–PCR primers (V2F: 5′-IndexTermGGACACCATGATCCTAACAC and TJR: 5′-IndexTermAGTCCATTTGCCACTGATGG) for DLC1 transcript variant 2 (NM_006094) were designed across introns to ensure non-genomic amplification. RT–PCR was performed as previously for 35 cycles using the Taq-Gold polymerase and the GeneAmp RNA PCR system (Applied Biosystems, Foster City, CA, USA), with GAPDH as a control (Tao et al., 2002).
We determined the transcription start site of the DLC1 transcript variant 2 (NM_006094) using 5′-RACE version 2.0 (Invitrogen) (Ying et al., 2005). Briefly, the first-strand cDNA was synthesized using primer E11R: 5′-IndexTermATCATGCTCTCTCTTGACCA. After adding the homopolymeric tails to the 3′ ends with terminal deoxynucleotidyl transferase, PCR was performed using Abridged Anchor Primer and a second gene-specific primer E10R2: 5′-IndexTermTAAAGCTGTGCATACTGGGG. The RACE product was enriched by re-amplifying with the Abridged Universal Amplification Primer and E10R: 5′-IndexTermCCGTAGCCAATCACAAGCTT, cloned and sequenced.
Methylation-specific PCR and bisulfite genomic sequencing
DNA bisulfite treatment, bisulfite genomic sequencing (BGS), and Methylation-specific PCR (MSP) were performed as previously described (Tao et al., 1999, 2002). MSP primers detecting the methylated or unmethylated alleles of the DLC1 promoter were: for methylated promoter: DLC1bm11, 5′-IndexTermAACCGAAAAACAACCCGTCG (or DLC1bm1, 5′-IndexTermAAAAAAAACATTCCAACCTTCG) and DLC1bm22, 5′-IndexTermGAGCGAATTGTTTTTCGCGC; for unmethylated promoter: DLC1bu11, 5′-IndexTermAAAACCAAAAAACAACCCATCA (or DLC1bu1, 5′-IndexTermAAAAAACATTCCAACCTTCAACA) and DLC1bu22, 5′-IndexTermTTGGAGTGAATTGTTTTTTGTGT (Figure 1b). These primer pairs have been tested previously for not amplifying any unbisulfited DNA. MSP was performed for 40 cycles using the Taq-Gold polymerase (Applied Biosystems) (Tao et al., 1999; 2002). BGS primers (for the bottom strand) were DLC1BGSb1: 5′-IndexTermccaaataaataccttataaccttta and DLC1BGSb2: 5′-IndexTermgaggtgYggttatgttttggt (Y=C or T). Amplified BGS products were TA-cloned and 5–12 colonies were randomly chosen and sequenced.
5-aza-2′-deoxycytidine (Aza) treatment of NPC cell lines
NPC cells (CNE-1, CNE-2 and C666-1) freshly seeded at 1 × 105 cells/ml were allowed to grow overnight. The culture medium was then replaced with fresh medium containing Aza at a final concentration of 50 μ M (Sigma-Aldrich Corporation, St Louis, MO, USA) (Qiu et al., 2004; Ying et al., 2005). Cells were allowed to grow for 72 h, with changing of Aza-containing medium every 24 h, and then harvested for DNA and RNA extraction. For this treatment, a cell viability of around 70% was retained after 72 h of treatment.
Cloning the DLC1 promoter, plasmid construction and transfection
Three different regions of the putative DLC1 promoter (-1926/+152) were PCR amplified from normal PBMC DNA using a high-fidelity DNA polymerase, Platinum Pfx (Invitrogen) and TA-cloned. The sequences and orientation of the cloned fragments were verified by sequencing. These fragments were then ligated to the promoter-less plasmid, pGL2-Enhancer vector (Promega, Madison, WI, USA) to create p(-480)DLC1EN, p(-1198)DLC1EN and p(-1926)DLC1EN. Plasmids used for transfection were isolated and purified using EndoFree Plasmid Maxi Kit (Qiagen GmbH, Germany).
The promoter activities of these fragments were tested by transient co-transfection with 2 μg plasmid DNA and 100 ng pRL-SV40 as an internal control using LipofectAMINE 2000 (Invitrogen) in four cell-lines (C33A, C666-1, HK1 and HD-MY-Z) in a 12-well plate. Transfected cells were incubated for 48 h and promoter activities were assessed using the Dual Luciferase Reporter Assay System (Promega). Three independent assays were conducted and the mean±standard deviation (s.d.) values were calculated.
Colony formation assay
CNE-1, HK1 and EC109 cells were seeded at 1–2 × 105 cells/ml in a 12-well plate and allowed to grow for 24 h. Cells were then transiently transfected with 0.5–2 μg of pcDNA3.1(+)/DLC1 (plasmid containing the full-length DLC1 cDNA (NM_006094), a gift from Dr NC Popescu (Yuan et al., 1998)), pcDNA3.1(+)/TP53 (gift from Dr Bert Vogelstein) or pcDNA3.1 vector alone, using LipofectAMINE 2000 or FuGENE 6 (Roche). At 48-h post-transfection, the transfected cells were seeded onto six-well plates at a density of 1–3 × 104 cells/well with G-418 selection (500 μg/ml). After 2–3 weeks, cells were stained with Gentian Violet and examined. Numbers of colonies (with >50 cells/colony) were counted and analysed. The experiment was repeated twice independently. Statistical analysis was performed with Student's t-test, P<0.05 was considered as statistically significant difference.
Busson P, Ganem G, Flores P, Mugneret F, Clausse B, Caillou B et al. (1988). Establishment and characterization of three transplantable EBV-containing nasopharyngeal carcinomas. Int J Cancer 42: 599–606.
Chen YJ, Ko JY, Chen PJ, Shu CH, Hsu MT, Tsai SF et al. (1999). Chromosomal aberrations in nasopharyngeal carcinoma analyzed by comparative genomic hybridization. Genes Chromosomes Cancer 25: 169–175.
Cheng Y, Chakrabarti R, Garcia-Barcelo M, Ha TJ, Srivatsan ES, Stanbridge EJ et al. (2002). Mapping of nasopharyngeal carcinoma tumor-suppressive activity to a 1.8-megabase region of chromosome band 11q13. Genes Chromosomes Cancer 34: 97–103.
Cheng Y, Poulos NE, Lung ML, Hampton G, Ou B, Lerman MI et al. (1998). Functional evidence for a nasopharyngeal carcinoma tumor suppressor gene that maps at chromosome 3p21.3. Proc Natl Acad Sci USA 95: 3042–3047.
Deng W, Tsao SW, Guan XY, Lucas JN, Si HX, Leung CS et al. (2004). Distinct profiles of critically short telomeres are a key determinant of different chromosome aberrations in immortalized human cells: whole-genome evidence from multiple cell lines. Oncogene 23: 9090–9101.
Effert P, McCoy R, Abdel-Hamid M, Flynn K, Zhang Q, Busson P et al. (1992). Alterations of the p53 gene in nasopharyngeal carcinoma. J Virol 66: 3768–3775.
Etienne-Manneville S, Hall A . (2002). Rho GTPases in cell biology. Nature 420: 629–635.
Feng BJ, Huang W, Shugart YY, Lee MK, Zhang F, Xia JC et al. (2002). Genome-wide scan for familial nasopharyngeal carcinoma reveals evidence of linkage to chromosome 4. Nat Genet 31: 395–399.
Gardiner-Garden M, Frommer M . (1987). CpG islands in vertebrate genomes. J Mol Biol 196: 261–282.
Glaser R, Zhang HY, Yao KT, Zhu HC, Wang FX, Li GY et al. (1989). Two epithelial tumor cell lines (HNE-1 and HONE-1) latently infected with Epstein-Barr virus that were derived from nasopharyngeal carcinomas. Proc Natl Acad Sci USA 86: 9524–9528.
Gulley ML, Nicholls JM, Schneider BG, Amin MB, Ro JY, Geradts J . (1998). Nasopharyngeal carcinomas frequently lack the p16/MTS1 tumor suppressor protein but consistently express the retinoblastoma gene product. Am J Pathol 152: 865–869.
Hildesheim A, Levine PH . (1993). Etiology of nasopharyngeal carcinoma: a review. Epidemiol Rev 15: 466–485.
Homma Y, Emori Y . (1995). A dual functional signal mediator showing RhoGAP and phospholipase C-delta stimulating activities. EMBO J 14: 286–291.
Hui AB, Lo KW, Kwong J, Lam EC, Chan SY, Chow LS et al. (2003). Epigenetic inactivation of TSLC1 gene in nasopharyngeal carcinoma. Mol Carcinog 38: 170–178.
Hurst CD, Fiegler H, Carr P, Williams S, Carter NP, Knowles MA . (2004). High-resolution analysis of genomic copy number alterations in bladder cancer by microarray-based comparative genomic hybridization. Oncogene 23: 2250–2263.
Kim TM, Jeong HJ, Seo MY, Kim SC, Cho G, Park CH et al. (2005). Determination of genes related to gastrointestinal tract origin cancer cells using a cDNA microarray. Clin Cancer Res 11: 79–86.
Kim TY, Jong HS, Song SH, Dimtchev A, Jeong SJ, Lee JW et al. (2003). Transcriptional silencing of the DLC-1 tumor suppressor gene by epigenetic mechanism in gastric cancer cells. Oncogene 22: 3943–3951.
Lo KW, Huang DP, Lau KM . (1995). p16 gene alterations in nasopharyngeal carcinoma. Cancer Res 55: 2039–2043.
Lo KW, Teo PM, Hui AB, To KF, Tsang YS, Chan SY et al. (2000). High resolution allelotype of microdissected primary nasopharyngeal carcinoma. Cancer Res 60: 3348–3353.
Lu SJ, Day NE, Degos L, Lepage V, Wang PC, Chan SH et al. (1990). Linkage of a nasopharyngeal carcinoma susceptibility locus to the HLA region. Nature 346: 470–471.
Moon SY, Zheng Y . (2003). Rho GTPase-activating proteins in cell regulation. Trends Cell Biol 13: 13–22.
Murray PG, Qiu GH, Fu L, Waites ER, Srivastava G, Heys D et al. (2004). Frequent epigenetic inactivation of the RASSF1A tumor suppressor gene in Hodgkin's lymphoma. Oncogene 23: 1326–1331.
Pathmanathan R, Prasad U, Sadler R, Flynn K, Raab-Traub N . (1995). Clonal proliferations of cells infected with Epstein-Barr virus in preinvasive lesions related to nasopharyngeal carcinoma. N Engl J Med 333: 693–698.
Plaumann M, Seitz S, Frege R, Estevez-Schwarz L, Scherneck S . (2003). Analysis of DLC-1 expression in human breast cancer. J Cancer Res Clin Oncol 129: 349–354.
Qiu GH, Tan LK, Loh KS, Lim CY, Srivastava G, Tsai ST et al. (2004). The candidate tumor suppressor gene BLU, located at the commonly deleted region 3p21.3, is an E2F-regulated, stress-responsive gene and inactivated by both epigenetic and genetic mechanisms in nasopharyngeal carcinoma. Oncogene 23: 4793–4806.
Raab-Traub N . (2002). Epstein-Barr virus in the pathogenesis of NPC. Semin Cancer Biol 12: 431–441.
Rhee I, Bachman KE, Park BH, Jair KW, Yen RW, Schuebel KE et al. (2002). DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. Nature 416: 552–556.
Sahai E, Olson MF, Marshall CJ . (2001). Cross-talk between Ras and Rho signalling pathways in transformation favours proliferation and increased motility. EMBO J 20: 755–766.
Spruck III CH, Tsai YC, Huang DP, Yang AS, Rideout III WM, Gonzalez-Zulueta M et al. (1992). Absence of p53 gene mutations in primary nasopharyngeal carcinomas. Cancer Res 52: 4787–4790.
Srivastava G, Wong KY, Chiang AK, Lam KY, Tao Q . (2000). Coinfection of multiple strains of Epstein-Barr virus in immunocompetent normal individuals: reassessment of the viral carrier state. Blood 95: 2443–2445.
Steenbergen RD, Kramer D, Braakhuis BJ, Stern PL, Verheijen RH, Meijer CJ et al. (2004). TSLC1 gene silencing in cervical cancer cell lines and cervical neoplasia. J Natl Cancer Inst 96: 294–305.
Steenbergen RD, Walboomers JM, Meijer CJ, van der Raaij-Helmer EM, Parker JN, Chow LT et al. (1996). Transition of human papillomavirus type 16 and 18 transfected human foreskin keratinocytes towards immortality: activation of telomerase and allele losses at 3p, 10p, 11q and/or 18q. Oncogene 13: 1249–1257.
Sun Y, Hegamyer G, Cheng YJ, Hildesheim A, Chen JY, Chen IH et al. (1992). An infrequent point mutation of the p53 gene in human nasopharyngeal carcinoma. Proc Natl Acad Sci USA 89: 6516–6520.
Tang JC, Wan TS, Wong N, Pang E, Lam KY, Law SY et al. (2001). Establishment and characterization of a new xenograft-derived human esophageal squamous cell carcinoma cell line SLMT-1 of Chinese origin. Cancer Genet Cytogenet 124: 36–41.
Tao Q, Huang H, Geiman TM, Lim CY, Fu L, Qiu GH et al. (2002). Defective de novo methylation of viral and cellular DNA sequences in ICF syndrome cells. Hum Mol Genet 11: 2091–2102.
Tao Q, Robertson KD, Manns A, Hildesheim A, Ambinder RF . (1998). The Epstein-Barr virus major latent promoter Qp is constitutively active, hypomethylated, and methylation sensitive. J Virol 72: 7075–7083.
Tao Q, Swinnen LJ, Yang J, Srivastava G, Robertson KD, Ambinder RF . (1999). Methylation status of the Epstein-Barr virus major latent promoter C in iatrogenic B cell lymphoproliferative disease. Application of PCR-based analysis. Am J Pathol 155: 619–625.
Tao Q, Young LS, Woodman CB, Murray PG . (2006). Epstein-Barr virus (EBV) and its associated human cancers – Genetics, epigenetics, pathobiology and novel therapeutics. Front Biosci 11: 2672–2713.
Teramoto A, Tsukuda K, Yano M, Toyooka S, Dote H, Doihara H et al. (2004). Less frequent promoter hypermethylation of DLC-1 gene in primary breast cancers. Oncol Rep 12: 141–144.
Tsao SW, Wang X, Liu Y, Cheung YC, Feng H, Zheng Z et al. (2002). Establishment of two immortalized nasopharyngeal epithelial cell lines using SV40 large T and HPV16E6/E7 viral oncogenes. Biochim Biophys Acta 1590: 150–158.
Wang DZ, Nur-E-Kamal MS, Tikoo A, Montague W, Maruta H . (1997). The GTPase and Rho GAP domains of p190, a tumor suppressor protein that binds the M(r) 120,000 Ras GAP, independently function as anti-Ras tumor suppressors. Cancer Res 57: 2478–2484.
Wilson PJ, McGlinn E, Marsh A, Evans T, Arnold J, Wright K et al. (2000). Sequence variants of DLC1 in colorectal and ovarian tumours. Hum Mutat 15: 156–165.
Wong CM, Lee JM, Ching YP, Jin DY, Ng IO . (2003). Genetic and epigenetic alterations of DLC-1 gene in hepatocellular carcinoma. Cancer Res 63: 7646–7651.
Wong ML, Tao Q, Fu L, Wong KY, Qiu GH, Law FB et al. (2006). Aberrant promoter hypermethylation and silencing of the critical 3p21 tumour suppressor gene, RASSF1A, in Chinese oesophageal squamous cell carcinoma. Int J Oncol 28: 767–773.
Xiong W, Zeng ZY, Xia JH, Xia K, Shen SR, Li XL et al. (2004). A susceptibility locus at chromosome 3p21 linked to familial nasopharyngeal carcinoma. Cancer Res 64: 1972–1974.
Ying J, Li H, Seng TJ, Langford C, Srivastava G, Tsao SW et al. (2006). Functional epigenetics identifies a protocadherin PCDH10 as a candidate tumor suppressor for nasopharyngeal, esophageal and multiple other carcinomas with frequent methylation. Oncogene 25: 1070–1080.
Ying J, Srivastava G, Hsieh WS, Gao Z, Murray P, Liao SK et al. (2005). The stress-responsive gene GADD45G is a functional tumor suppressor, with its response to environmental stresses frequently disrupted epigenetically in multiple tumors. Clin Cancer Res 11: 6442–6449.
Yu MC, Yuan JM . (2002). Epidemiology of nasopharyngeal carcinoma. Semin Cancer Biol 12: 421–429.
Yuan BZ, Durkin ME, Popescu NC . (2003a). Promoter hypermethylation of DLC-1, a candidate tumor suppressor gene, in several common human cancers. Cancer Genet Cytogenet 140: 113–117.
Yuan BZ, Jefferson AM, Baldwin KT, Thorgeirsson SS, Popescu NC, Reynolds SH . (2004). DLC-1 operates as a tumor suppressor gene in human non-small cell lung carcinomas. Oncogene 23: 1405–1411.
Yuan BZ, Miller MJ, Keck CL, Zimonjic DB, Thorgeirsson SS, Popescu NC . (1998). Cloning, characterization, and chromosomal localization of a gene frequently deleted in human liver cancer (DLC-1) homologous to rat RhoGAP. Cancer Res 58: 2196–2199.
Yuan BZ, Zhou X, Durkin ME, Zimonjic DB, Gumundsdottir K, Eyfjord JE et al. (2003b). DLC-1 gene inhibits human breast cancer cell growth and in vivo tumorigenicity. Oncogene 22: 445–450.
Zhou L, Jiang W, Ren C, Yin Z, Feng X, Liu W et al. (2005). Frequent Hypermethylation of RASSF1A and TSLC1, and High Viral Load of Epstein-Barr Virus DNA in Nasopharyngeal Carcinoma and Matched Tumor-Adjacent Tissues. Neoplasia 7: 809–815.
We thank Drs Bert Vogelstein, George Tsao, Dolly Huang, Kaitai Yao, Soh-ha Chan, Yixin Zeng, Ya Cao, Guiyuan Li, Shuen-Kuei Liao and Peter Stern for some cell lines, Cordelia Langford for the high-resolution (1-Mb) array-CGH slides, and Nicholas C Popescu for the DLC1 plasmid. This project was supported by an A*STAR grant (Johns Hopkins Singapore) and a Michael Kadoorie Cancer Genetics Research Fund grant (QT). Richard Ambinder is supported by an NIH grant (P01CA15396).
About this article
Cite this article
Seng, T., Low, J., Li, H. et al. The major 8p22 tumor suppressor DLC1 is frequently silenced by methylation in both endemic and sporadic nasopharyngeal, esophageal, and cervical carcinomas, and inhibits tumor cell colony formation. Oncogene 26, 934–944 (2007). https://doi.org/10.1038/sj.onc.1209839
- tumor suppressor gene
- nasopharyngeal carcinoma
- esophageal carcinoma
- cervical carcinoma
Molecular and Cellular Biochemistry (2017)
Genome Medicine (2016)
Modern Pathology (2014)
Epigenetic identification of receptor tyrosine kinase-like orphan receptor 2 as a functional tumor suppressor inhibiting β-catenin and AKT signaling but frequently methylated in common carcinomas
Cellular and Molecular Life Sciences (2014)
Double restriction-enzyme digestion improves the coverage and accuracy of genome-wide CpG methylation profiling by reduced representation bisulfite sequencing
BMC Genomics (2013)