Screening Of Methylation Profile Of Breast Cancer By Microarray And Investigating Their Expression Levels

BackgroundBreast cancer is the most frequently diagnosed cancer among females. Incidence rateshave been rising in many countries. With the development of molecular biology techniques, Itis used to be hot topic of breast cancer clinical research to find effective biomakers which canreliably forecast disease diagnosis, treatment and prognosis. It has been increasinglyrecognized over the past several years that in cancer epigenetic changes are more prominentthan genetic changes, such as gene mutation and gene deletion. DNA methylation is the mostcommon events of epigenetic changes in tumorigenesis. Abnormal DNA methylation isfrequently associated with transcription and expression of genes. Methylation is frequentlyassociated with transcriptional silencing of tumor suppressor genes and activing of oncogenes,and then have a influence on cell cycle regulation, apoptosis, DNA repair or cell adhesion..What’s more, aberrant methylation seems to be an early event in tumorigenesis, and DNAmethylation is potentially reversible by demethylating agents. Restore expression of specificgenes through change of DNA methylation state may become a target of cancer gene therapy.Thus, the study of tumor DNA methylation status has provided a new way to explore themolecular mechanisms of tumor development. It is also a hot spot of tumor basic research inrecent years.ObjectiveThis study aims to investigate the methylation profile of breast cancer for screen methylation biomarker of breast cancer early detection by MRIA combined with CpG islandarray, explore the relationship between methylation and gene expression by using microarrayand RT-PCR. Hope to provide candidate data for further study about DNA methylation of thebreast cancer, principle theory for novel hypermethylated biomakers for early diagnosis andtherapeutic target for breast cancer.Methods1.Samples of breast tumor tissues and normal breast tissues were collected from15infiltrative ductal carcinoma patients (7stages T1N0M0and8stages T2N0M0) and15non-cancer patients(2cases of mastitis,2cases of breast adenosis,5cases of intraductalpapilloma,6cases of fibroadenoma), respectively. All diagnoses were pathologicallyconfirmed during operation immediately or subsequently. All samples were stored atRNAlater solution and put in an ultra-low temperature refrigerator at-80°C until nucleic acidsextraction. Among these sample,10matched breast tumor tissues and normal breast tissueswere used for micrarray array, the other5pairs samples for validation of methylation level ofcandidates genes.2. We performed methylated-CpG island recovery assay combined with CpG island arrayon10pairs of breast cancer tissues and normal breast tissues to identify the methylationprofile of breast cancer. Direct bisulfite sequencing and combined bisulfite restrictionanalysis (COBRA) were carried out in independent5pairs of cancer and normalsamples to validation of methylation level of candidates genes.3. In order to investigate the methylated genes expression profile of breast cancer, we usehuman whole genome microarray in the10previously pairs of breast cancer tissues andnormal breast tissues. RT-PCR was carried out in independent5pairs of cancer andnormal samples to validation of gene expression of candidates genes.4. To explore the relationship between methylation and gene expression,We use humanwhole genome expression microarray and RT-PCR on4concern genes in the10matchedbreast cancer tissues and normal breast tissues. Results1. Difference DNA methylation profile between breast cancer and normal breast tissueProbes with an intensity>400were selected for further analysis. Probes were selected aspositive if their fold changes in MIRA/Input signaling ratios between cancer versus normalwere above1.2(p<0.01) using SAM software analysis. Genes with at least two positiveprobes were defined to be hypermethylated genes. Then we detected70significantlyhypermethylated genes in breast cancer tissues,5genes of the most significantlyhypermethylated are CCND2、PAX6、MNX1、OTX2and PITX2, including many novelhypermethylated genes in breast cancer model such as ITGA4, NFIX, OTX2and FGF12.These significantly hypermethylated genes involved with cell cycle, differentiation,poliferation, damage and healing, tumorangiogenesis, etc.2. Validation of hot hypermethylated genesTo confirm the reliability of our results of MIRA microarray analysis, four genes wereselected for validation in one breast tumor sample and one non-tumor breast cancer sampleusing direct bisulfite sequencing. Through direct bisulfite sequencing, widespreadmethylations were observed in intragenic regions of WT1, PAX6and ITGA4genes, and inpromoter region of OTX2in breast cancer tissue. Methylations were also observed innontumor breast tissue for OTX2and WT1genes but not for PAX6and ITGA4genes.WT1,OTX2,PAX6and ITGA4were selected for validation in one breast tumor sample andone non-tumor breast cancer sample using COBRA. COBRA assay confirmed that WT1,OTX2and PAX6genes were hypermethylated in breast cancer tissues. Direct bisulfitesequencing and COBRA confirm the reliability of our results of MIRA microarray analysis.3. Interrelationship between DNA methylation and gene expressionHuman whole genome microarray analysis showed that a total of1308transcripts wereidentified as being significantly differentially expressed in breast cancer tissues relative tonormal breast tissues. Compared with the methylation profile of breast cancer, an in-depthanalysis of microarray data revealed that mRNA of some highly methylated genes were toolow to be detected by using gene expression microarray in both cancer and normal breasttissues. RT-PCR assay showed that WT1and PITX2were only weakly expressed in the breast cancer tissues and weren’t expressed in most normal breast tissues. OTX2and PAX6weren’t expressed in both breast cancer tissues and normal breast tissues.Conclusions1. There are multiplicity significantly hypermethylated genes in breast cancer tissues relativeto normal breast tissues, suggest that DNA methylation may play a crucial role in breastcarcinogenesis.2. ITGA4, NFIX, OTX2and FGF12were first identified to be hypermethylated in breastcancer model. It provides principle theory for novel hypermethylated biomakers for earlydiagnosis and therapeutic target for breast cancer.3. Widespread methylations occurred in not only in promoter region of genes, but also inintragenic regions of some genes. The relationship between methylation and geneexpression is complicated and needs further investigations.