This article is mentioned in:

Introduction

Cervical cancer remains one of the leading causes of cancer-related mortality in women despite medical advances and the availability of vaccination programmes (1). Promoter hypermethylation of tumor suppressor genes (TSGs) has long been contested as a probable cause of cancer and its progression (2). Several TSGs are inactivated by this mechanism (3). The tissue inhibitor of metalloproteinases-2 (TIMP2) is known to antagonize matrix metalloproteinase activity and to suppress tumor growth, angiogenesis, invasion and metastasis. The TIMP2 gene is known to be expressed in normal human tissues, whereas its expression is downregulated in glioblastomas and metastatic lung tumors (4). TIMP2 overexpression had an inhibitory effect on tumor growth and angiogenesis in a breast cancer mouse model (5). In addition, overexpression of TIMP2 has been shown to restrict the invasiveness of various tumor cell types in vitro (6,7). Suzuki et al (8) observed methylated TIMP2 in the colorectal cancer cell line RKO. However, methylation of TIMP2 was not commonly found in primary colorectal tumors. The frequency of the hypermethylated TIMP2 gene in cervical cancer was found to be 47% by Ivanova et al (9), whereas methylation-specific PCR (MSP) and sodium bisulfite analysis of genomic DNA of the HeLa cell line revealed an unmethylated promoter and expression of TIMP2. SiHa and Caski cervical cancer cell lines had a methylated promoter with downregulated expression of corresponding gene activity and methylation of the 5′ region of the TIMP2 gene. Therefore, in the present study, we aimed to study the role of the promoter hypermethylation and associated TIMP2 expression in cervical cancer cell lines.

Materials and methods

Cell culture

The cervical cancer cell lines HeLa, SiHa and Caski were procured from NCCS (Pune, India) and maintained in RPMI-1640 (Sigma, St. Louis, MO, USA) supplemented with 10% FBS (Life Technologies, Israel) and 10,000 units of penicillin and streptomycin (Sigma) at 37°C and 5% CO2 (10).

Treatment of cell lines

HeLa, SiHa and Caski cell lines were treated with 5-aza-2′-deoxycytidine (Sigma) (positive control) at 20-μM concentration for 4 days with change of media along with 5-aza-2′-deoxycytidine every 48 h. Untreated cells were used as a control to analyse the promoter methylation status of the TIMP2 gene.

Methylation specific PCR (MSP)

DNA was isolated using standard phenol:chloroform extraction and quantified using an ND1000 spectrophotometer (Thermo Scientific). DNA (1 μg) was subjected to bisulfite modification using EZ gold methylation kit (Zymo Research). Bisulfite-modified DNA was used for MSP of the TIMP2 gene with a set of primers (9) spanning regions 1919–1987 (−325 to −257), relative to the transcription start site. MSP was performed as detailed by Ivanova et al (9). The PCR products were analysed on 3% agarose gel. MSP was carried out in duplicate.

Reverse transcription PCR (RT-PCR)

Total RNA was isolated from cultured cells using TRI reagent (Sigma) and was treated with RNAse-free DNAseI (Fermentas) to eliminate any DNA contamination. cDNA was synthesized using a First Strand Revertaid cDNA synthesis kit (Fermentas). RT-PCR was carried out for TIMP2, β-actin and GAPDH genes using the following primers: TIMP2 (forward: 5′-TGGCCT TTATATTTGATCCACAC-3′, reverse: 5′-AAAAATCCAAAC GGAAACAAAAT-3′); β-actin (forward: 5′-CATGTACGT TGCTATCCAGGC-3′, reverse: 5′-CTCCTTAATGTCACG CACGAT-3′); GAPDH (forward: 5′-CAAGGTCATCCA TGACAACTTTG-3′, reverse: 5′-GTCCACCACCCTGTTGCT GTAG-3′), under the following conditions: initial denaturation at 94°C for 3 min followed by 28 cycles (94°C for 30 sec, 60°C for 45 sec, 72°C for 45 sec) and final extension at 72°C for 5 min. β-actin and GAPDH were regarded as the internal control. Densitometric analysis of the bands was performed using 1D analysis software with average density as a parameter, calculated using intensity (INT U)/mm2 with a sensitivity of 10 (Bio-Rad, Hercules, CA, USA). Fold change in expression was calculated for each band using the formula: Fold change = average density of test gene/average density of internal control.

Results

MSP

MSP for the TIMP2 gene with primers specific to methylated DNA was carried out with untreated and 5′ aza-2′-deoxycytidine treated cells, which resulted in the amplification of a 68-bp amplicon in untreated cells of the HeLa, SiHa and Caski cell lines, whereas no such band was observed in 5′ aza-2′-deoxycytidine treated cells. Primers specific for unmethylated DNA responded only with samples treated with 5′ aza-2′-deoxycytidine, resulting in amplification of the 68-bp amplicon (Fig. 1A).

Figure 1 (A) M, methylation-specific band; U, unmethylation-specific band; C, control samples; A, 5-aza-2′-deoxycytidine treated samples. From left: lane 1, 100 bp DNA ladder; lanes 2–3, TIMP2 MSP in the HeLa cell line; lanes 4–5, TIMP2 MSP in the SiHa cell line; lanes 6–7, TIMP2 MSP in the Caski cell line. (B) RT-PCR was carried out to analyse the expression of TIMP2 in the HeLa, SiHa and Caski cell lines, respectively. From left: lane 1, 100 bp DNA ladder; lanes 2–3, comparative expression analysis of TIMP2, β-actin and GAPDH in the HeLa cell line; lanes 4–5, comparative expression analysis of TIMP2, β-actin and GAPDH in the SiHa cell line; lanes 6–7, comparative expression analysis of TIMP2, β-actin and GAPDH in the Caski cell line. (C) Comparative densitometric analysis of TIMP2 expression with β-actin and GAPDH in the HeLa, SiHa and Caski cell lines.

RT-PCR

RT-PCR for the TIMP2 gene was carried out for the HeLa, SiHa and Caski cell lines to assess the impact of promoter hypermethylation on the expression of the TIMP2 gene in untreated cells. Expression of the TIMP2 gene was found to be normal in the HeLa cell line. In the case of the SiHa cell line, the TIMP2 gene exhibited downregulation, which was not found to be significant at RT-PCR level, whereas TIMP2 was found to be markedly upregulated in the Caski cell line, which was confirmed following a comparison of two internal controls (Fig. 1B). As per the analysis, the expression of the TIMP2 gene was found to be 3.7- and 1.6-fold in the Caski cell line, 0.95- and 0.98-fold in the HeLa cell line, and 0.76- and 0.83-fold (Fig. 1C) compared to β-actin and GAPDH, respectively.

Discussion

The TIMP2 gene, an endogenous inhibitor of matrix metalloproteinase (MMP), plays a significant role in cell invasion and tumorigenesis and is downregulated or silenced in various human cancer cell lines. The percentage ratio of TIMP2 expression compared to that of MMP has always been debated. High expression of both TIMP2 and MMP9 has been observed in invasive cancer cells of the cervix and the surrounding stromal cells, whereas the expression of both TIMP2 and MMP9 is hardly detectable in cervical intraepithelial neoplasias (CIN) by hybridization and in situ RT-PCR (11,12). The fact that TIMP2 is an inhibitor of MMPs means that it belongs to a significant family of tumor suppressor genes whose function requires comprehensive evaluation. During our ongoing study on DNA methylation, we selected the TIMP2 gene and focused on a possible reversal of hypermethylation and subsequent reactivation.

Since there was only one published study by Ivanova et al available (9) relating TIMP2 promoter hypermethylation to cervical cancer and, specifically, cervical cancer cell lines, a TIMP2 promoter hypermethylation study spanning the 1919-1987 region, i.e., −325 to −257 relative to the transcription start site was undertaken. However, the results were very dissimilar to those already reported. The promoter of the TIMP2 gene was found to be methylated in the HeLa cell line using MSP, which was confirmed after treatment of the HeLa cell line with 5′ aza-2′-deoxycytidine leading to the disappearance of the 68 bp methylation-specific band. Primers specific for unmethylated DNA responded only to 5′ aza-2′-deoxycytidine treated samples, which again confirmed the promoter hypermethylation, contrary to the existing study. Similarly, methylation of the TIMP2 gene was confirmed in the SiHa and Caski cell lines. Upon meta-analysis of the TIMP2 sequence (GenBank accession number: U 44381), eight CpG sites were found in the 1919–1987 (−325 to −257) region and 13 CpG sites were found in total when the range was extended to −350 to −240, as in the study carried out by Ivanova et al (9), relative to the transcription start site (Fig. 2). The CpG sites within the −350 to −240 region in the HeLa cell line were found to be unmethylated, correlating with the normal expression of the TIMP2 gene in the HeLa cell line. In the case of the SiHa cell line, seven CpG sites were hemimethylated, three were methylated and three were found to be unmethylated, whereas in the Caski cell line all CpG sites were found to be methylated with the exception of one or two sites that correlated with the downregulated gene expression of TIMP2 in the SiHa and Caski cell lines. In the present study, the transcription of the TIMP2 gene was found to be highly upregulated in the Caski cell line despite the promoter being hypermethylated. High expression in the Caski cell line was also confirmed with another housekeeping gene, GAPDH, in addition to β-actin; however, there was a consistent expression of the TIMP2 gene in the HeLa cell line, whereas the downregulated expression was observed in the SiHa cell line despite promoter hypermethylation, which was moderately observable at the RT-PCR level.

Figure 2 (|) CpG site. Eight CpG sites exist between the 1919 and 1987 regions of a 2691-bp-long sequence with the exon sequence extending from the 2244 to the 2669 region of the TIMP2 gene, exon 1 (GenBank accession number: U 44381).

High expression of TIMP2 in HPV-16 positive SiHa and Caski cell lines has been reported (13) but discussion is lacking on the effect and role of promoter methylation status of the TIMP2 gene in these cervical cancer cell lines. Our study clearly showed that the CpG sites within the 1919–1987 (−325 to −257) region, relative to the transcription start site of the TIMP2 promoter, do not regulate the expression of the TIMP2 gene in the Caski and HeLa cell lines. However, the expression in the SiHa cell line indicated downregulation but was not found to be significant. Five CpG sites still remain in the −350 to −240 region, relative to the transcription start site, of which two CpG sites are situated upstream of −325 and the remaining three CpG sites are situated downstream of −247. The methylation profile of specific CpG sites within a promoter, located towards the transcription start site are involved in the regulation of gene expression (14). Since the TIMP2 gene is expressed in cervical cancer cell lines, the involvement of other CpG sites situated downstream of −247, including the remaining three CpG sites is significant. It is speculated that other CpG sites located towards the transcription start site play a more significant role in regulating the expression of the TIMP2 gene in cervical cancer cell lines and specifically the Caski cell line. In conclusion, we did not find a clear-cut correlation between promoter hypermethylation and expression of TIMP2 gene in cervical cancer cell lines, indicating that the interplay of methylation and location of CpG sites in the promoter near the transcription site may be important for gene expression.

Acknowledgements

This study was supported by the Council of Scientific and Industrial Research (CSIR), New Delhi, India.

References