Genome-wide analysis of DNA methylation identifies novel cancer-related genes in hepatocellular carcinoma.
ABSTRACT Aberrant DNA methylation has been implicated in the development of hepatocellular carcinoma (HCC). Our aim was to clarify its molecular mechanism and to identify useful biomarkers by screening for DNA methylation in HCC. Methylated CpG island amplification coupled with CpG island microarray (MCAM) analysis was carried out to screen for methylated genes in primary HCC specimens [hepatitis B virus (HBV)-positive, n = 4; hepatitis C virus (HCV)-positive, n = 5; HBV/HCV-negative, n = 7]. Bisulfite pyrosequencing was used to analyze the methylation of selected genes and long interspersed nuclear element (LINE)-1 in HCC tissue (n = 57) and noncancerous liver tissue (n = 50) from HCC patients and in HCC cell lines (n = 10). MCAM analysis identified 332, 342, and 259 genes that were methylated in HBV-positive, HCV-positive, and HBV/HCV-negative HCC tissues, respectively. Among these genes, methylation of KLHL35, PAX5, PENK, and SPDYA was significantly higher in HCC tissue than in noncancerous liver tissue, irrespective of the hepatitis virus status. LINE-1 hypomethylation was also prevalent in HCC and correlated positively with KLHL35 and SPDYA methylation. Receiver operating characteristic curve analysis revealed that methylation of the four genes and LINE-1 strongly discriminated between HCC tissue and noncancerous liver tissue. Our data suggest that aberrant hyper- and hypomethylation may contribute to a common pathogenesis mechanism in HCC. Hypermethylation of KLHL35, PAX, PENK, and SDPYA and hypomethylation of LINE-1 could be useful biomarkers for the detection of HCC.
- [show abstract] [hide abstract]
ABSTRACT: We report 3 cases of lymphoid neoplasms with mixed lineage features of T-, NK-, or B-cell marker expression and clonal gene rearrangement for both T-cell receptor and immunoglobulin light chain IgK. A characteristic of our cases was the lack of expression of the specific B-cell transcription factor, Pax5, which is essential for maintaining the identity and function of mature B cells during late B lymphopoiesis. In the absence of Pax5, B cells in vitro can differentiate into macrophages, dendritic cells, granulocytes, and T/NK cells. Methylation analysis of the Pax5 gene in our cases suggests that its inactivation by this epigenetic event in a committed or mature B cell, before plasma cell differentiation, may well be a common pathogenetic mechanism in mature lymphoid neoplasms with expression of multilineage antigens. In particular, case 1 may represent a mixed NK- and B-cell lineage; and cases 2 and 3 may represent mixed T and B-cell lineage, respectively. Aberrations in the DNA methylation patterns are currently recognized as a hallmark of human cancer. Cases with aberrant phenotypes require molecular analysis for lineage assignment. Studies of such cases may be helpful to better elucidate whether they represent a distinct entity with clinical, immunophenotypic, and molecular characteristics or an incidental phenomenon during malignant transformation. Interestingly, these cases were all characterized by poor clinical outcome.Human pathology 05/2009; 40(9):1252-61. · 3.03 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Using real-time quantitative methylation-specific PCR (RTQ-MSP), we quantified methylated p16INK4a sequences and determined the fractional concentrations of circulating tumor DNA in plasma, serum, and peripheral blood cells collected preoperatively, intraoperatively, and postoperatively from 49 patients with hepatocellular carcinoma (HCC). RTQ-MSP was sufficiently sensitive to detect down to 10 genome-equivalents of methylated p16INK4a sequences. Quantitative MSP data were expressed in terms of the methylation index, which was the percentage of bisulfite converted unmethylated and methylated p16INK4a sequences that consisted of methylated p16INK4a sequences. Quantities of methylated p16INK4a sequences were detected in peripheral circulation of 80% (23 of 29) of HCC patients. No significant difference was seen in the detectability and concentrations of methylated p16INK4a sequences (range: 10-4046 genome-equivalents/ml) between preoperative plasma and serum samples from HCC patients. Preoperatively, the p16INK4a methylation indices ranged from 0.2 to 100% and from 0.012 to 0.075% in the patients' plasma and buffy coat samples, respectively. After surgical resection, the median p16INK4a methylation indices in plasma and buffy coat concordantly decreased 12- and 15-fold, respectively. These results demonstrated the clinical usefulness and effectiveness of peripheral blood RTQ-MSP for detecting and monitoring HCC after treatment. Furthermore, none of the intraoperative plasma samples and only two of the intraoperative buffy coat samples were p16INK4a methylation positive. Quantification of epigenetic changes in peripheral blood by RTQ-MSP is useful for the detection and monitoring of HCC.Clinical Cancer Research 04/2003; 9(3):1047-52. · 7.84 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Most hepatocellular carcinomas (HCC) are diagnosed at an advanced stage. Hypermethylation of CpG islands in promoter regions is now recognized as an important early event in carcinogenesis and detection of methylated DNA has been suggested as a potential biomarker for early detection of cancer. There are no studies on epigenetic changes in samples from HCC patients before diagnosis. We explored the possible diagnostic value of aberrant promoter hypermethylation of three tumor suppressor genes in serum DNA for early detection of HCC. Aberrant promoter hypermethylation was investigated in DNA isolated from the serum of 50 HCC patients who provided repeated blood samples before diagnosis and 50 controls enrolled in a cancer screen program in Taiwan. Methylation-specific PCR was used to determine the methylation status of p16, p15, and ras association domain family 1A (RASSF1A). Among cases, aberrant methylation was found in serum DNA 1 to 9 years before clinical HCC diagnosis. RASSF1A had the highest frequency of hypermethylation with 35 (70%) cases having at least one positive sample compared with 22 (44%) for p16 and 12 (22%) for p15. Six subjects were hypermethylation negative for all three genes. For the 50 controls, promoter hypermethylation was found in three and two subjects for RASSF1A and p16, respectively; none had methylation of p15. A receiver operating characteristic curve that included clinical risk factors (age, HBsAg status, anti-hepatitis C virus status, smoking, and alcohol status) and hypermethylation biomarkers gave an overall predictive accuracy of 89% with sensitivity and specificity 84% and 94%, respectively. The analysis of epigenetic changes on RASSF1A, p16, and p15 tumor suppressor genes in serum DNA may be a valuable biomarkers for early detection in populations at high risk of HCC.Clinical Cancer Research 05/2007; 13(8):2378-84. · 7.84 Impact Factor