Genome-Wide DNA Methylation Analysis In Primary Hepatocellular Carcinoma Without Infection Of HBV, HCV And The Expression Verification Of Partial Functional Genes

Objective The morbidity and mortality of Hepatocellular carcinoma (HCC) are ranked in the forefront of the various types of neoplastic diseases. DNA methylation, as one of the most important epigenetic regulation modification in tumor, has become a research hot spot, recently. In this study, we examined the differences of the whole genome DNA methylation, mRNA transcription and protein expressing in cancerous tissue and there matched adjacent non-tumor tissue to explore the relationships between methylation profile and HCC.Method 1.To obtain the microarray data:(1) Eighteen HCC samples and there paired adjacent tissues confirmed by the pathology detection were collected from the Specimen Library of Affiliated Tumor Hospital and Department of Hepatobiliary Surgery in First Affiliated Hospital of Guangxi Medical. All the patients signed the informed consent for the study. Bisulfite conversion was performed after isolating the genomic DNA. Then we used Illumina Human Methylation 27k platform together with the methylation array data which met our study criteria from the TCGA database, performed background correction, quality control, and normalization by RnBeads package written in R language. (2) The genes of interest were chosen by the analysis of Gene Ontology and KEGG_PATHWAY.2. Validation experiments:(1) We used the Methylated Primer Design module of PyroMark Assay Design 2.0 to design the forward, reverse, and sequencing primers for the genes of interest, and the polymerase chain reaction (PCR) assay was applied to verify the specificity of the amplification products. The methylation level of cancerous and there matched adjacent tissue were detected by the PyroMark Q96 Pyrosequencing platform manufactured by QIAGEN. (2) New samples were collected to perform PCR and Immunohistochemistry (IHC) assay to explore the expression of the genes. All of the statistical works were completed in R environment.Result l.The clinical characteristics:There were 18 participants involved in our methylation array study. Seventeen of them were men, and 12 men had smoking history,6 men had drinking history. The average age 54.9±11.5 years.2. The results of methylation array:P-values on the site level were computed using the limma method. I.e. hierarchical linear models from the limma package were employed and fitted using an empirical Bayes approach. Furthermore, we used a cutoff of completely changed methylation status combined with |Delta Beta|>0.2 to identify 444 genes which contains 271 hypomethylated sites and 223 hypermethylated sites.3. Functional annotation-. The hypermethylated genes were involved in the activities of developing, cell process, and cell motility, etc. The hypomethylated genes, however, were involved in the process of keratinization, cell communication, and response to stimulus. Unsupervised hierarchical clustering revealed natural clustering by diagnosis demonstrating that DNA methylation profiles distinguished each group. We choose 2 genes (LRSAM1 and VAV1) to validate the methylation level based on the functional analysis and literature research, and found LRSAM1 and VAV1 were hypomethylated in cancerous tissue. The results were consistent with methylation array data and had a good linear correlation with Beta value.4. The results of RT-qPCR showed that mRNA level of LRSAM1 (0.55±1.42, p=0.011) and VAV1 (0.84±1.82, p=0.011) were significantly increased in cancerous tissue. 5. We used immunohistochemistry to further verify the protein expression of the above mentioned genes. There were two proteins which were LRSAM1 82% (37/45)-42%(13/31) and VAV1 (57% 26/46)-41%(15/37) differentially expressed in cancerous tissues compared with adjacent non-tumor tissues.Conclusion 1. This work primarily established the profile of the uninfected (HBV、HCV) hepatocellular carcinoma, and pave the road for further studying the functions and mechanisms.2. To the best of our knowledge, this is the first time that LRSAM1 and VAV1 are found to be hypomethylated in HCC tissue, and we validated the mRNA and protein expression level by RT-qPCR and IHC, respectively.