Silencing of the retinoid response gene TIG1 by promoter hypermethylation in nasopharyngeal carcinoma†
Version of Record online: 28 SEP 2004
Copyright © 2004 Wiley-Liss, Inc.
International Journal of Cancer
Volume 113, Issue 3, pages 386–392, 20 January 2005
How to Cite
Kwong, J., Lo, K.-W., Chow, L. S.-N., Chan, F. L., To, K.-F. and Huang, D. P. (2005), Silencing of the retinoid response gene TIG1 by promoter hypermethylation in nasopharyngeal carcinoma. Int. J. Cancer, 113: 386–392. doi: 10.1002/ijc.20593
This work was carried out within the Hong Kong Cancer Genetics Research Group.
- Issue online: 18 NOV 2004
- Version of Record online: 28 SEP 2004
- Manuscript Accepted: 29 JUL 2004
- Manuscript Received: 14 APR 2004
- Kadoorie Charitable Foundations
- Hong Kong Research Grant Council. Grant Numbers: CUHK4067/02M, CUHK4410/03M
- RGC Central Allocation. Grant Number: HKUST3/03C
- tazarotene-induced gene 1 (TIG1);
- nasopharyngeal carcinoma;
- tumor suppressor
Tazarotene-induced gene 1 (TIG1) and Tazarotene-induced gene 3 (TIG3) are retinoid acid (RA) target genes as well as candidate tumor suppressor genes in human cancers. In our study, we have investigated the expression of TIG1 and TIG3 in nasopharyngeal carcinoma (NPC). Loss of TIG1 expression was found in 80% of NPC cell lines and 33% of xenografts, whereas TIG3 was expressed in all NPC samples and immortalized nasopharyngeal epithelial cells. In order to elucidate the epigenetic silencing of TIG1 in NPC, the methylation status of TIG1 promoter was examined by genomic bisulfite sequencing and methylation-specific PCR (MSP). We have detected dense methylation of TIG1 5′CpG island in the 5 TIG1-negative NPC cell lines and xenograft (C666-1, CNE1, CNE2, HONE1 and X666). Partial methylation was observed in 1 NPC cell line HK1 showing dramatic decreased in TIG1 expression. Promoter methylation was absent in 2 TIG1-expressed NPC xenografts and the normal epithelial cells. Restoration of TIG1 expression and unmethylated alleles were observed in NPC cell lines after 5-aza-2′-deoxycytidine treatment. Moreover, the methylated TIG1 sequence was detected in 39 of 43 (90.7%) primary NPC tumors by MSP. In conclusion, our results showed that TIG1 expression is lost in the majority of NPC cell lines and xenografts, while promoter hypermethylation is the major mechanism for TIG1 silencing. Furthermore, the frequent epigenetic inactivation of TIG1 in primary NPC tumors implied that it may play an important role in NPC tumorigenesis.
Nasopharyngeal carcinoma (NPC) is a serious health problem in southern China because it has an unusually high incidence among our population. The annual male crude incidence rate per 100,000 in Hong Kong is 24.4 (Hong Kong Cancer Registry, 1999), contrasted to a frequency of <1 case per 100,000 persons in Caucasians.1 Previous etiology studies demonstrated that the development of NPC might be attributable to a complex interaction of genetic factors, dietary exposure to carcinogenic constituents in preserved foods and Epstein-Barr virus (EBV) infection. In the last 10 years, epigenetic inactivation of tumor-suppressor genes in human cancers has been extensively studied. Promoter hypermethylation is one of the major mechanisms to inactivate tumor suppressor genes and cancer-related genes in human cancer.2 Our group has previously shown that several tumor suppressor genes and cancer-related genes (p16, RASSF1A, RARβ2, DAP-kinase,EDNRB and TSLC1) were frequently silenced by promoter hypermethylation in NPC.3, 4, 5, 6, 7
Retinol, or vitamin A, is indispensable for embryonic development, growth, vision and survival of vertebrate.8 The vitamin A metabolite retinoic acid (RA) mediates multiple biological processes, including cell proliferation and differentiation, by modulating the rate of transcription of numerous target genes. Over the last quarter century, more than 532 genes have been put forward as regulatory targets of RA. The regulatory control of some targets is direct, driven by a liganded heterodimer of retinoid receptors bound to a DNA response element of the target genes. In others, the control is indirect, reflecting the actions of intermediate transcription factors, nonclassical associations of receptors with other proteins or even more distant mechanisms.9
Among a number of RA target genes in various systems, tazarotene-induced gene 1 (TIG1) and tazarotene-induced gene 3 (TIG3) were identified in skin and found as tumor suppressor genes in human cancers.10, 11, 12
TIG1 is a candidate tumor suppressor gene of human prostate cancer.11 It was demonstrated that the decreased expression of TIG1 in prostate carcinoma cells was associated with an increase in the malignant potential. The restoration of TIG1 expression in the highly malignant prostate cancer cell line, which does not express TIG1, greatly reduced the invasiveness in vitro and the tumorigenicity in nude mice.11
TIG3 acts as a tumor suppressor gene based on its homology to H-rev 107, a known class II tumor suppressor that inhibits growth of cancer cell lines in vitro.12, 13 The gene is expressed in various tissues, while expression is reduced in human cancer cell lines.12 With regard to epidermal hyperproliferation and development of skin cancers, TIG3 expression was found to be diminished in psoriasis lesions and basal cell carcinoma and significantly lower in advanced squamous cell carcinoma.14 In breast cancer cell lines, retinoid-dependent TIG3 induction is observed in lines that are growth suppressed by retinoid but not in retinoid nonresponsive lines. Transient over-expression of TIG3 in cell line T47D inhibited colony expansion.12 Recently, it is suggested that TIG3 is regulated by all-trans retinoic acid (ATRA) via retinoid receptors in certain aerodigestive tract cancer cells, and its induction by ATRA is associated with the suppression of anchorage-independent growth.15
In our study, we investigated the 2 RA target tumor suppressor genes (TIG1 and TIG3) in NPC. We analyzed mRNA expression of TIG1 and TIG3 in normal nasopharyngeal epithelia, NPC cell lines and xenografts. Moreover, we investigated the methylation status of TIG1 in normal nasopharyngeal epithelia and primary NPC samples. Based on our findings, we found that promoter hypermethylation is the key mechanism to inactivate TIG1 expression in NPC.
Material and methods
Cell lines, xenografts and primary tumors
Five NPC cell lines (C666-1, CNE1, CNE2, HK1 and HONE1), 3 xenografts (X2117, X666 and XNPC8), an immortalized nasopharyngeal epithelial cell line (NP69)16 and 2 normal nasopharyngeal epithelial outgrowth cultures (NP1 and NP2) were included in our study. The NPC cell lines were maintained in RPMI-1640 medium supplemented with 10% FBS, while NP69 was grown in keratinocyte serum-free medium (Invitrogen, Carlsbad, CA). The cells were incubated in a humidified atmosphere of 5% CO2 at 37°C. Forty-three cases of archived paraffin-embedded primary tumors and 4 cases of normal nasopharyngeal mucosa were retrieved from the pathology bank of the Department of Anatomical and Cellular Pathology at The Chinese University of Hong Kong. The histological diagnoses of the specimens were confirmed by a pathologist. Genomic DNA and total RNA of a human prostate immortalized cell line (PNT2), 4 human prostate cancer cell lines (LNCaP, DU145, CWR22 and C4-2B), were also included in our study.
To examine the correlation of promoter hypermethylation and expression of TIG1 gene, the NPC cell line (C666-1, CNE1, CNE2 and HONE1), shown to have a lack of TIG1 expression. was subject to 5-aza-2′-deoxycytidine and Tricostatin A treatment. Cells were plated with 5-aza-2′-deoxycytidine or Tricostatin A (Sigma Chemical Co., St. Louis, MO) and incubated for 4 days. The medium and the drug were replaced every 24 hr and cells were harvested for RNA and DNA extraction 4 days after treatment.
For each primary tumor, 50 serial sections (5 μm thick) of paraffin embedded specimen were subject to microdissection manually or by laser-captured microdissection using a PixCell LCM system (Arcturus Engineering, Mountain View, CA) under the guidance of a pathologist. Neoplastic cells of tumor samples were isolated and collected for DNA extraction. DNA was extracted as described previously.17
Reverse transcription polymerase chain reaction (RT-PCR)
The mRNA expression of TIG1 and TIG3 in NPC samples and immortalized nasopharyngeal epithelial cells were examined by RT-PCR analysis. Total RNA was isolated using TRIZOL reagent (Invitrogen) according to the manufacturer's protocol. Two micrograms of total RNA from each sample was subject to cDNA synthesis using Superscript II reverse transcriptase (Invitrogen). The primers and conditions for RT-PCR analysis were listed in Table I. Transcript of GAPDH was amplified as control.
|Genes||Primer sequences||Products sizes (bp)||Annealing temperature (°C)||Number of cycle||Reference|
|TIG1 (Region 1)||M1L: 5′-GTAGTACGGGCGGGTCGC-3′||118||67||35||21|
|TIG1 (Region 2)||M2L: 5′-CGTTTATGTAGTTTCGTCGGTAAC-3′||227||60||35|
Real-time quantitative RT-PCR
To elucidate the restoration of TIG1 expression in C666-1 cells after 5-aza-2′-deoxycytidine treatment, real-time quantitative RT-PCR was performed. In our study, pool sample of normal nasopharyngeal epithelial outgrowths was recruited as normal control. The first strand cDNA was prepared using TaqMan® reverse transcription reagents (Applied Biosystems, Foster City, CA). The PCR primers and conditions were listed in Table I. Using the SYBR® Green RT-PCR Kit (Applied Biosystems), PCR products were stained with SYBR Green and analyzed using an I-cycler (Bio-Rad, Hercules, CA). All reactions were done in triplicate. Relative mRNA of target gene of each sample was normalized with β-actin and calculated using the 2[-Delta Delta C(T)] method18 that compared the mRNA amount of each sample to that of a pool sample of normal nasopharyngeal epithelia (labeled normal NP).
Genomic bisulfite sequencing
To investigate the methylation pattern of TIG1 promoter, genomic DNAs from the nasopharyngeal samples were subjected to bisulfite sequencing. Genomic DNAs were modified by bisulfite treatment using CpGenome DNA Modification Kit (Intergen, New York, NY). The sequences of putative promoter and exon 1 of TIG1 gene (NM_002888) were obtained from UCSC Human Genome Browser Gateway (http://genome.ucsc.edu/cgi-bin/hgGateway) and the CpG island was then identified by CpG Island Searcher (http://ccnt.hsc.usc.edu/cpgislands/). The criteria of the 5′ CpG island is: GC content >60%, ratio of CpG to GpC > 0.6 and minimum length 200 bp. The PCR primers, conditions and expected product sizes were listed in Table I. For bisulfite sequencing, PCR amplification was performed on 100 ng of bisulfite-modified genomic DNA. The amplified fragments were cloned and 8–10 clones of each sample were sequenced with Big Dye Terminator Reaction Mix and analyzed by ABI 3100 Genetic Analyzer (Applied Biosystems).
Methylation specific polymerase chain reaction (MSP)
The promoter methylation status of TIG1 in the nasopharyngeal and prostatic samples was also investigated by methylation specific PCR (MSP) as described previously.19 Two regions of the TIG1 5′ CpG island were analyzed by MSP. Primer sequences for both methylated and unmethylated alleles, annealing temperatures and the expected PCR product sizes of the genes are summarized in Table I. One microliter of bisulfite-modified DNA from the samples was subject to PCR amplification. In vitro methylated DNA (IVD) (Intergen) was served as a control for methylated sequences, while the DNA from peripheral blood lymphocytes (PBL) was served as a control for unmethylated sequences. All MSP reactions were duplicated. Twenty-five microliters of PCR products were loaded onto a 10% nondenaturing polyacrylamide gel, stained with ethidium bromide and visualized under UV illumination.
Loss of TIG1 expression in NPC cell lines and xenografts
We have examined the mRNA expression of TIG1 and TIG3 genes in 5 NPC cell lines, 3 xenografts and an immortalized nasopharyngeal epithelial cell line (NP69) by RT-PCR. Expression of TIG1 was completely lost in 4/5 of NPC cell lines (80%; C666-1, CNE1, CNE2 and HONE1) and 1/3 of NPC xenografts (33%; X666). Highly reduced TIG1 mRNA expression was observed in a NPC cell line HK1. On the other hand, TIG3 was expressed in all NPC cell lines, xenografts and NP69 epithelial cells (Fig. 1).
Aberrant promoter hypermethylation of TIG1 in NPC cell lines and xenografts
To assess whether the loss of TIG1 expression in NPC was resulted from promoter hypermethylation, the methylation status of the 5′-CpG island of TIG1 gene was determined by genomic bisulfite sequencing. The CpG island of TIG1 starts from 5′ promoter (−350) to transcription start site (+1), and further downstream to the entire exon 1 (+303). Within the CpG island, 20 CpG sites were examined by genomic bisulfite sequencing. From our data, dense methylation of TIG1 5′-CpG island was found in 4 NPC cell lines (C666-1, CNE1, CNE2 and HONE1) and a xenograft (X666) with undetectable TIG1 expression. No aberrant methylation of TIG1 promoter was detected in TIG1-expressing NPC xenografts (X2117 and XNPC8), an immortalized nasopharyngeal epithelial cell line (NP69) and normal nasopharyngeal epithelial outgrowths (NP1 and NP2) (Fig. 2a). Scattered methylated CpG sites were found in the TIG1 promoter of the NPC cell line HK1 that showed weak expression of TIG1.
We also examined TIG1 promoter methylation in NPC samples by methylation-specific PCR (MSP). Two sets of MSP primers flanking the 5′-promoter region (MSP1L and MSP1R) and exon 1 (MSP2L and MSP2R) were used in our study. The location of MSP primers is illustrated in Figure 2a. Similar to the findings from genomic bisulfite sequencing, complete methylation was detected in C666-1, CNE1, CNE2, HONE1 and X666 by MSP analysis of these regions. Partial methylation was detected in HK1 cells, and unmethylation was detected in 3 TIG1-expressing nasopharyngeal samples (X2117, XNPC8 and NP69) (Fig. 2b).
Previously, TIG1 was shown to be downregulated in human prostate cancers and have been first identified as a tumor suppressor gene in prostate cancer.11 In order to examine the epigenetic inactivation of TIG1 in prostate cancer, we investigated the mRNA expression and promoter methylation of TIG1 gene on an immortalized human prostate cell line (PNT2) and 4 prostate cancer cell lines (LNCaP, DU145, CWR22 and C4-2B). TIG1 mRNA expression was only detected in the line PNT-2 and a prostate cancer cell line DU145. No TIG1 expression was observed in other 3 prostate cancer cell lines (LNCaP, CWR22 and C4-2B) (Fig. 3a). By genomic bisulfite sequencing, dense TIG1 promoter methylation was detected in TIG1-negative prostate cancer cell line (LNCaP), whereas no methylation was observed in TIG1-expressing prostate cancer cell line (DU145) (Fig. 3a). By MSP, TIG1 promoter methylation was detected in TIG1-negative prostate cancer cell lines (LNCaP, CWR22 and C4-2B), while no methylation was observed in TIG1-expressed samples (PNT-2 and Du145) (Fig. 3b).
Reactivating TIG1 expression after demethylating agent treatment
To confirm that promoter hypermethylation of NPC cell lines are correlated with the loss of TIG1 expression, we studied the effect of 5-aza-2′-deoxycytidine, a demethylating agent, on TIG1 expression. After treating the NPC cell line C666-1 (which showed transcriptionally silencing and dense promoter methylation of TIG1) with 5-aza-2′-deoxycytidine, TIG1 gene was reexpressed and unmethylated alleles of TIG1 promoter were detected (Fig. 4a,b). In Figure 4a, we found that TIG1 expression in C666-1 was only partial restored after treatment with 1–5 μM 5-aza-2′-deoxycytidine. The level of TIG1 reexpression did not reach the same level as the normal nasopharyngeal sample. By bisulfite sequencing, we detected only about half of alleles are unmethylated after demethylating agent treatment. Aside from C666-1, the expression of TIG1 gene was also observed in the other 3 NPC cell lines (CNE1, CNE2 and HONE1) with promoter hypermethylation after 5-aza-2′-deoxycytidine treatment (Fig. 4c). No TIG1 gene expression were found in the NPC cell lines (C666-1, CNE1, CNE2 and HONE1) treated with Tricostatin A (Fig. 4d).
Aberrant promoter hypermethylation of TIG1 gene in primary NPC tumors
To address the incidence of the promoter hypermethylation of TIG1 in primary tumors, we examined 43 primary NPC tumors and 4 normal nasopharyngeal mucosa by MSP in the 5′-promoter region (MSP1L and MSP1R). Aberrant promoter hypermethylation of TIG1 was found in 39/43 (90.7%) of primary NPCs but not in all 4 normal nasopharyngeal epithelia. The representative results of MSP analysis are shown in Figure 5.
In our study, we have examined 2 retinoid response tumor suppressor genes (TIG1 and TIG3) in NPC. Loss of TIG1 expression was frequently detected in NPC cell lines and xenografts, whereas TIG3 was expressed in all NPC samples. Our results suggested that the loss of TIG1 may play a role in NPC tumorigenesis. Tazarotene-induced gene 1 (TIG1), also known as retinoic acid receptor responder 1 (RARRES1), is a retinoic acid receptor-responsive gene that was originally isolated from the skin and its expression is increased by the synthesis retinoid tazarotene, which is effective in the treatment of hyperproliferative dermatologic disease such as psoriasis.10TIG1 cDNA contains an open-reading frame of coding for a putative protein of 228 amino acids. It appears to be a transmembrane protein with a small N-terminal intracellular region, a single membrane-spanning hydrophobic region, and a large C-terminal extracellular region containing a glycosylation signal.10, 11 The putative TIG1 protein also contains a hyaluronic acid binding motif. It has been suggested that TIG1 functions as a cell adhesion molecule whose expression on the cell surface may lead to better cell-cell contact and reduced proliferation.11 Moreover, TIG1 has been identified as a tumor suppressor gene for human prostate cancer. It was demonstrated that the decreased expression of TIG1 was associated with an increase in the malignant characteristics of both prostate cell lines and tissues. The expression of TIG1 was absent in all malignant prostate cell lines examined in the study, in all carcinoma tissues with Gleason scores of 4–10 and in the majority (73%) of the prostate tissues with Gleason scores of 2–3. Restoration of TIG1 expression in the highly malignant prostatic PC-3M cells, which does not express TIG1 mRNA, greatly reduced their invasiveness in vitro and their tumorigenicity in nude mice.11 Based on this tumor suppressing activities, the function of TIG1 in NPC cells is worth future study.
According to UCSC Human Genome Browser Gateway, the chromosomal location of TIG1 (RARRES1) is located at 3q25.32. Based on our previous studies on allelotyping analysis and comparative genomic hybridization in NPC,23, 24 loss of heterozygosity at this region is frequently found in NPC. In the present study, TIG1 promoter hypermethylation is detected in TIG1-negative NPC cell lines and xenografts, while no TIG1 methylation was detected in TIG1-expressing nasopharyngeal samples. When administrating the demethylating agent, 5-aza-2′-deoxycytidine, to the TIG1-negative NPC cell line (C666-1), TIG1 gene was reexpressed and unmethylated alleles of TIG1 promoter were detected. Reexpression of the TIG1 gene by treatment with demethylating agent was also confirmed in another 3 NPC cell lines (CNE1, CNE2 and HONE1). However, treatment with the histone deacetylase inhibitor, Tricostatin A, was not able to restore the TIG1 gene expression in these cell lines. The results indicated that the silencing of TIG1 gene is closely related to its promoter hypermethylation in NPC. Therefore, besides allelic loss of chromosome 3q25.32, promoter hypermethylation is another mechanism to inactivate TIG1 in NPC.
Promoter hypermethylation of TIG1 gene has been reported in several cancers. Youssef et al. have examined TIG1 methylation and expression status in 53 human cancer cell lines and 74 primary tumors, including leukemia, head and neck, breast, colon, skin, brain, lung and prostate cancer. Loss of TIG1 expression was strongly associated with TIG1 promoter hypermethylation (p < 0.001). Treatment with 5-aza-2′-deoxycytidine restored TIG1 expression in all 8 silenced cell lines tested.25 Moreover, Tokumaru et al. have observed a cancer-specific pattern of TIG1 methylation in approximately 50% of primary tumor prostate, head and neck, and lung cancer by MSP. The authors examined the methylation status of TIG1 in primary prostate cancers, head and neck cancers, and lung cancers. Their results revealed TIG1 methylation in 54.8% prostate cancers, 50% of head and neck cancers and 43.3% of lung cancers.22 Recently, another study demonstrated that the TIG1 downregulation in prostate cancer is due to the methylation of the promoter and CpG island of TIG1. TIG1 was methylated in 52% primary prostate cancers but was not methylated in normal tissues or benign hyperplasia. Positive correlation between the methylation of TIG1 and RAR-beta was also observed in prostate cancer.26 In our present study, we have also confirmed that the promoter hypermethylation of TIG1 is related to its gene silencing in prostate cancer cell lines. Therefore, these findings suggested that promoter hypermethylation is the major mechanism for TIG1 inactivation and the silencing of this potential tumor suppressor gene may be a critical step for cancer development in different tissues.
In conclusion, loss of TIG1 was found in the majority of NPC cell lines and xenografts. Promoter hypermethylation is the key mechanism to inactivate TIG1 expression in NPC. High frequency of TIG1 methylation was also detected in primary NPCs but not in normal nasopharyngeal samples, implying that the inactivation of this potential tumor suppressor gene is essential for NPC tumorigenesis. In order to explore the candidacy of TIG1 as a tumor suppressor gene in NPC, future studies should focus on the function of TIG1 in NPC cell lines in the development of NPC.
- 17Detailed deletion of mapping on the short arm of chromosome 3 in nasopharyngeal carcinoma. Int J Oncol 1994; 4: 1359–64., , , , , .
- 21Hypermethylation silencing of the retinoid-induced gene (TIG1) promoter in a variety of human cancers. Proc Am Assoc Cancer Res 2002; 43: 5530., , , , .