Hypermethylation of Promoter Region of LATS 1-a CDK Interacting Protein in Oral Squamous Cell Carcinomas-a Pilot Study in India

Genes that act to suppress the cell division are called “tumor suppressors” because the loss of function of such genes results in uncontrolled proliferation of cells leading to tumorigenesis. One of the best known examples for tumor suppressor gene is the p53 gene (Sharma et al., 2014). In the presence of DNA damaging agents (like toxic chemicals present in tobacco and related products for example that can induce mutation) cellular tumor suppressor proteins block cyclins A, E and D (Lin et al., 2014), and cyclin dependent kinases (CDK) 2, 4 and 6 to arrest the cell cycle (Cooks et al., 2014; Gao et al., 2014). Loss of function of tumor suppressor genes due to mutation(s) or loss of expression of tumor suppressor


Introduction
Genes that act to suppress the cell division are called "tumor suppressors" because the loss of function of such genes results in uncontrolled proliferation of cells leading to tumorigenesis. One of the best known examples for tumor suppressor gene is the p53 gene (Sharma et al., 2014). In the presence of DNA damaging agents (like toxic chemicals present in tobacco and related products for example that can induce mutation) cellular tumor suppressor proteins block cyclins A, E and D (Lin et al., 2014), and cyclin dependent kinases (CDK) 2, 4 and 6 to arrest the cell cycle (Cooks et al., 2014;Gao et al., 2014). Loss of function of tumor suppressor genes due to mutation(s) or loss of expression of tumor suppressor genes due to epigenetic modification causes the cell cycle machinery to undergo uncontrolled proliferation even in the presence of DNA damaging agents resulting in carcinogenesis Liu et al., 2011;Adduri et al., 2014). Several studies, both in vitro and in vivo, have clearly associated mutations and epigenetic modification in tumor suppressor genes with transformation potential, progression and metastasis (Aylon et al., 2011;Rivlin et al., 2011;Mak et al., 2014).
Promoter hypermethylation is one of the most common forms of epigenetic silencing of tumor suppressor genes and is believed to be more frequently involved in cancerous lesions than mutations (Chan et al., 2008). Hypermethylation of several tumor suppressor genes involved in cell cycle regulation such as p16 INK4a ,p15 INK4b ,p14 ARF ,p73,APC and BRCA1 have been reported in many types of tumors (Asokan et al., 2014;Kanherkar et al., 2014), albeit with variations in percent prevalence amongst them (Christoph et al., 2007;Moran et al., 2012). The percent occurrence of hypermethylation, hence, provides a tumor specific gene signature that may be used for sensitive detection of grade and stage of tumor types (Moran et al., 2012). For example, hypermethylation profile of MGMT, a gene involved in DNA repair has proven its usefulness in predicting the clinical prognosis and response to therapy in malignant glioma, gastric cancer, and lung cancer (Vallot et al., 2011).
The gene LATS1 (Large Tumor Suppressor, isoform 1) is a novel tumor suppressor molecule, the protein product of which localizes to the mitotic apparatus of a dividing cell (Hao et al., 2009). LATS1 protein forms complexes with the cyclin dependent kinase-CDK1, essential for the cells to progress from G2 phase to mitotic phase. Binding of LATS1 with CDK1 reduces the activity of CDK1 and thus blocks the cells from entering mitosis (Bothos et al., 2005). Since LATS1 negatively regulates cell division, it is called as a tumor suppressor. Mice that lacks LATS1 gene develop soft-tissue sarcomas and ovarian stromal cell tumors besides being highly sensitivity to carcinogenic agents (Bothos et al., 2005). Hypermethylation and downregulation of LATS1 gene promoter has been reported in breast cancers (56.7%) (Takahashi et al., 2005), astrocytomas (63.6%) (Jiang et al., 2006), myxoid liposarcomas, leiomyosarcomas, and malignant fibrous histiocytomas (Hisaoka et al., 2002;Visser S et al., 2010). Although oral squamous cell carcinoma (OSCC) is one of the most prevalent forms of cancer in India (Mishra et al., 2014) among all reported cancers, the methylation status of LATS1 gene promoter has not been explored. Hence, we decided to investigate the methylation status of LATS1 to understand: a) its prevalence, and b) its association with histopathological differentiation.

Patient samples
Thirteen patients with oral squamous cell carcinoma with confirmed histopathological diagnosis were included in the study after obtaining informed consent from patients. Biopsy tissues were collected in 2ml microfuge tubes containing 1.5ml of RNA Save reagent and transferred to the refrigerator. 24h later the samples were transferred to freezer and stored until transported to the laboratory.

DNA extraction and methylation sensitive restriction digestion
Tissues were washed once with cold 1X PBS (Phosphate Buffered Saline) to remove traces of RNA Save (Biological Industries Israel Beit-Haemek Ltd. Kibbutz Beit-Haemek, 25115, Israel) and were lysed in 500µl of lysis buffer to extract genomic DNA as described elsewhere (Mehta, 2014). 250ng of genomic DNA was digested with methylation sensitive type II restriction endonuclease, HpaII (New England Biolabs, Ipswich, MA, USA). The digestion was performed with 10U of enzyme in a 20µl reaction volume at 37°C for 4h, after which the enzyme was inactivated by incubating the reaction mix at 65°C for 20m. Methylated DNA (EpiTect Control DNA, cat # 59655, Qiagen) was included as positive control for enzyme digestion.

Methylation sensitive polymerase chain reaction (MS-PCR)
The Hpa II digested tumor DNA samples were amplified with a set of LATS1 promoter specific primers. Primers were designed flanking 500bp of promoter sequence upstream of translational start site, since this region is rich in CCGG content (Figure 1). The following set of primer sequences was used: Left primer: AAGTAAAACCTCTTGTCACACC, Right primer: CAATCCTTTTCCCCATTG. PCR was performed in triplicates with the above LATS1 promoter set of primers on HpaII digested or undigested tumor samples and control DNA. Control DNA was obtained from peripheral blood of a healthy volunteer. The following conditions were used: initial denaturation at 94°C for 5 minutes, followed by 35 cycles of denaturation at 94°C for 45 seconds, primer annealing at 55°C for 45 seconds, primer extension at 72°C for 45 seconds, with a final extension at 72°C for 5 minutes.

Promoter analysis and primer designing
The sequence of promoter region of LATS1 was acquired from the public database (www.ncbi.nlm.nih. gov/gene), and 1000 base pairs upstream of transcription start site was analyzed for the relative presence of CG (Cytosine and Guanine) (Figure 1). A total of fifty six CG motifs were identified, of which six were within the CCGG tetranucleotide recognition motif of HpaII, type 2 restriction enzyme. Based on the location of the CCGG motifs, forward and reverse primers were designed to encompass the first three CCGG motifs upstream of the transcriptional start site.  DOI:http://dx.doi.org/10.7314/APJCP.2015.16.4.1599 Restriction digestion of genomic DNA with Hpa II and Polymerase chain reaction to detect hypermethylation of LATS1 promoter region 250 nanograms of genomic DNA from each of the sample were subjected to Hpa II digestion. Hpa II recognizes unmethylated tetranucleotide motif CCGG and cuts after the first C. Incubating the genomic DNA with Hpa II results in complete digestion of all unmethylated CCGG motifs in the entire genome including the promoter region of LATS1, which has six CCGG motifs. The fragmented promoter of LATS1 cannot be amplified by the flanking primers, as the primers were designed to amplify the intact promoter region only. In case, if the C residue in CCGG motif were to be methylated the Hpa II enzyme would fail to recognize the CCGG motif, which as a result maintains the promoter region intact (uncut). This unfragmented and intact promoter region serves as an effective template and hence the promoter region gets amplified in PCR.
Following digestion of OSCC DNA samples with HpaII, 25 nanogram aliquots of the samples were used as template to amplify the promoter region. Polymerase chain reaction (PCR) was performed in triplicates under conditions mentioned in the methods section. Gel analysis of the PCR product showed amplification in 7 out of 13 (54%) tumor DNA samples, which indicated hypermethylation of the promoter region ( Figure 2, top panel, lanes 2, 4, 6, 8, 10, 11) in these samples. The specificity of LATS1 promoter amplification was confirmed by migration of the DNA bands in agarose gel with reference to positive control. Undigested OSCC samples were amplified along with positive control DNA under similar condition as control for the presence of DNA in all the samples, which showed amplification in all tumor DNA samples (Figure 2, bottom panel). Similar data was obtained in the rest of triplicate PCR amplifications (data not shown).

Consolidated data of hypermethylation of LATS1 promoter and histopathology
All tissue samples that were used in the present study were subjected to histopathological analysis and their diagnosis confirmed independently. Of the thirteen OSCC samples that were analyzed, six were well differentiated, three were moderately differentiated and the remaining four were early invasive lesions. When the hypermethylation status was corroborated with histopathological finding, LATS1 promoter hypermethylation was observed in three of six well differentiated OSCCs, one of three moderately differentiated OSCCs and three of four early invasive lesions (Table 1). No association between the hypermethylation and location of tumor, or age and gender of the patients could be determined due to smaller size of the samples.

Discussion
In the present study, we have used methylation sensitive restriction enzyme digestion to determine the methylation profile of the promoter region of LATS1, a tumor suppressor gene, in oral squamous cell carcinomas. Our data demonstrates promoter hypermethylation in a significant proportion of cases (54%, n=7 out of 13 OSCCs). No association between the location of OSCC lesion, age and gender of patients and LATS1 hypermethylation could be made due to smaller size of the samples that were analyzed. The sample size  was intentionally chosen to include only a few cases in each of the three categories-early invasive, moderately differentiated and well differentiated OSCCs, so as to first explore the occurrence of LATS1 hypermethylation in OSCC lesions as no prior data on this gene was available in OSCCs. Although bisulfite modification based methylation-specific PCR (MSP) is widely used to determine the hypermethylation status of genes, the assay is difficult to optimize for all loci and requires column based purification procedures and primers for both modified and unmodified DNA. The methylation sensitive restriction enzyme technique, on the other hand, is fast and equally sensitive in detecting the methylation status of genes. This method is also proven to be sensitive in several studies including OSCCs and hence was used in the present study . Besides, the methylation sensitive PCR on HpaII digested DNA was performed in triplicates with similar end results, based on which we believe that the technique is unlikely to have influenced the higher detection of LATS1 hypermethylation in the present study. The observation of hypermethylation in one lesion with epithelial dysplasia, two early invasive, and one moderately differentiated OSCC samples suggests that the loss of expression of LATS1 may be one of the early events during carcinogenesis. This may be considered as significant as only thirteen OSCC samples were tested altogether. This percent occurrence of LATS1 hypermethylation is similar to the prevalence pattern observed in breast cancers (56.7%) (Takahashi et al., 2005) and astrocytomas (63.6%) (Jiang et al., 2006). The lack of promoter hypermethylation in the rest of the samples indicates that the hypermethylation may not be a targeted event in OSCCs. Such heterogeneity in promoter hypermethylation has also been observed in several genes in several cancers including OSCCs. For example, the cell cycle regulators p15 and p16 are known to be hypermethylated in 23% of OSCCs, mismatch repair genes-hMLH1 and MGMT are hypermethylated in 8% and 41% of OSCCs respectively, while the homotypic cell-cell adhesion gene-E-Cadherin is hypermethylated in 35% of OSCCs (Marcinkiewicz et al., 2014). The above percent occurrence may be considered as a relevant indicator for heterogeneity of hypermethylation observed in the present study as the authors in the earlier study had analyzed OSCC samples obtained from patients who belonged to the southern region of India from where the samples used in the present study were also collected. P o s i t i v e a s s o c i a t i o n b e t w e e n p r o m o t e r hypermethylation and lymph node involvement and metastasis has been observed in other cancers (Peters et al., 2014). In addition, decrease in the levels of LATS1 has also been associated with poor prognosis in breast cancers (Takahashi et al., 2005). This association of LATS1 expression with clinical outcome adds significance to our finding of more frequent hypermethylation in early invasive and moderately differentiated OSCCs than well differentiated OSCCs, as it opens up the possibility of LATS1 methylation status being used as a prognostic indicator and diagnostic biomarker. However, this requires systematic evaluation in a statistically significant number of OSCCs that have been staged and graded. Taken together, the present explorative study has identified for the first time hypermethylation of LATS1 promoter region in histopathologically defined OSCCs.