| Epigenetic differences become an important issue for explaining cancer formation. Epigenetic modifications control gene expression in a hereditary way. Epigenomics describes the changes in epigenetic modifications on the genome. Recently, epigenomics becomes one of the cutting age fields in cancer research.DNA methylation is the most common and well-studied genomic modifications. DNA methylations regulate a variety of biological events. Different types of cancers have specific DNA methylation patterns. Moreover, DNA methylation patterns change in different stages during cancer formation. Consequently, DNA methylation on CpG islands becomes a genomic identity for a particular cancer. Tongue squamous cell carcinoma (TSCC) is most common oral squamous cancer. But, the mechanism is still unclear. In the past decades, abnormal DNA methylations in several genes have been found to be related to oral squamous cell carcinoma (OSCC) formation. However, most researches focused on genome-wide DNA methylation pattern instead of exploring methylation changes of several genes promoter. There is no report for genome-wide analysis of DNA methylation in TSCC. It is promising to establish genomic DNA methylation map for understanding the mechanism of TSCC, prediagnosis of TSCC and treatment of TSCC. Lack of high-throughput DNA methylation detection techniques is the major reason impeding the development of genomic DNA methylation map. The development of DNA methylation chip provides a powerful method for the research on epigenomics. NimbleGen has a DNA methylation chip best covering the number of CpG islands in human genome, best annotating the chip, and best sensitivity in the DNA methylation chip international market.We will use methylated DNA immunoprecipitation (MeDIP) and NimbleGen HG18 CpG Promoter Chip to analyze DNA methylations in TSCC with this high-throughput method. Comparing the gene expression data between TSCC sample and adjacent normal tissue in DNA methylation Chip, we are able to identify TSCC related genes.DNA methylation Chip detect TSCC genomic DNA methylationIn this study, we collected nine samples from TSCC and their adjacent counterparts as control. After obtaining the genomic DNA from those samples, we mixed nine genomic DNA from TSCC as tumor group and mixed nine genomic DNA from adjacent tissues as control group. By performing MeDIP, we respectively enrich the methylated DNA fragments from two groups followed by hybridizing to DNA mehtylation microarray individually. After data processing, there are 1269 DNA methylation sites only existing in tumor samples, not in normal adjacent samples. Those methylation sites accounts for 330 different genes, spreading in different chromosomes. There are 28 genes (8.48%) showing DNA methylation difference in chromosome 1. Secondly,27 genes (8.18%) in chromosome 19 and only 4 genes (1.21%) show difference in DNA methylation.218 out of the 330 genes (66.1%) have the DNA mehtylation in the promoter region and 70% of them are on the CpG islands.Similarly, in the normal adjacent sample, there are 1385 DNA methylation sites only existing in control tissue residing in 321 different genes. The number and distribution in different chromosomes have larger difference compared to those in tumor sample. Chromosome 19 has the largest number of the genes showing difference in DNA methylation. There are 40 genes (12.46%) in chromosome 19. X chromosome and chromosome 1 have 32 (9.97%) and 23 (7.17%) genes respectively. Chromosome 14 has the least number of the genes showing difference in DNA methylaion, only 2 genes (0.62%).210 out of the 321 genes (65.4%) have the DNA methylation in the promoter resion and 43.6% of them are on the CpG islands.The genes showing different DNA methylation in tumor and normal samples are reside in the same gene family. For example, four keratin protein family genes and six transmembrane protein family members only have DNA methylation in tumor samples. In the other hand, seven melanoma antigen family members only have DNA methylation in adjacent normal samples.We use Gene Ontology analysis website (Gostat) to classify the genes screening out from the DNA methylation chip. These genes are involved in transcription regulation, reproduction, development, metabolism, ion transportation, cell growth, cell differentiation and apoptosis, signal transduction and cell adhesion processes. At the same time, they are also involved in several immune and cancer development related pathways, such as cell communication, purine metabolism, cell adhesion molecules, natural killer cell mediated cytotoxicity, neuroactive ligand-receptor interaction, glycerophospholipid metabolism, leukocyte transendothelial migration and Fc epsilon RI signaling pathway. Those differential genes in DNA methylation might involve in TSCC development.Preliminary results of verifying the TSCC related genes selected from DNA methylation microarray dataWe selected ITIH5 and FBLN1 which are the differential genes in results from DNA methylation Chip of tumor samples. Moreover, we picked RUNX3 which shows differential changes in DNA methylation in adjacent normal samples. MSPCR were performed to detect the changes in DNA methylation both in tumor and normal samples. Although promoter hypermethylation of ITIH5 and FBLN1 were detected in 25% and 20% of normal samples respectively, higher percentage of promoter hypermethylation were found in tumor samples,70%(14/20) and 55% (11/20) respectively. The percentage differences between tumor and normal sample in those two genes are statistically significant. (P=0.004<0.01, P=0.026<0.05). In addition, there are 55% and 15% of DNA methylation in RUNX3 gene in normal and tumor samples respectively. This difference is also statistically significant (P=0.008<0.01). Consequently, preliminary MSPCR results are consistent to DNA methylation chip data.In order to further identify TSCC related genes, we analyzed DNA methylation Chip data and selected six genes (BCL2L14, CDCPl, DIRAS, FBLNl, ITIH5 and RUNX3) which are correlated with carcinogenesis but are not studied previously for further studies in TSCC. We used RT-PCR to analyze the expression level of those six genes in TSCC samples. RT-PCR data showed that the expression level of DIRAS (45% 9/20), FBLN1 (40% 8/20) and ITIH5 (55% 11/20) were down-regulated in TSCC samples (P<0.05). However, BCL2L14 (60% 12/20), CDCP1 (50% 10/20) and RUNX3 (75% 15/20) were up-regulated in TSCC samples. Therefore, those genes might be related to TSCC carcinogenesis.We further correlate the gene expression level with the DNA methylation in FBLN1, ITIH5 and RUNX3 genes. The DNA of FBLN1 (87.5% 7/8) and ITIH5 (90.9% 10/11) genes were methylated in TSCC samples whose expression level of FBLN1 and ITIH5 were down-regulated. Similarly, the DNA of RUNX3 (66.7% 10/15) gene was methylated in the adjacent normal tissue instead of TSCC tumor samples which shows up-regulating RUNX3 expression. Together, abnormal DNA methylation is related to the differential expression of FBLN1, ITIH5 and RUNX3 in TSCC tissues.In summary, this study established the whole genome-wide DNA methylation profile in TSCC tissues and their adjacent counterparts. Upon DNA methylation chip, it helps to develop precise genomic DNA methylation map and provides possible molecular mechanisms for TSCC carcinogenesis. In addition, six genes which are related to TSCC development are also found result from this study and might be able to become the biomarkers for diagnosis, treatment and prognosis of TSCC.
|
PDF Full Text Request