Study On Epigenetic Mechanisms Of Glioma

[Background of LRRC4]LRRC4 (GenBank accession number is AF196976), a novel member of LRR superfamily, was identified by us at 7q31-32 employing a strategy of combination EST-mediated screening and 5'RACE. We found that LRRC4 specifically expresses in human and mouse normal brain tissues by Northern-blot and RT-PCR assay. But its expression was not only deleted in several malignant glioma cell lines, but also deleted or down-regulated in 87.5% primary gliomas biopsies. It had the potential to suppress tumorigenesis of U251 malignant glioma cells in vivo and cell proliferation in vitro by a tetracycline-inducible expression system. It was also found that LRRC4 requires a functional LRR cassette domain to suppress U251 cell proliferation and invasion by regulating the ERK/AKT/NF-κB and JNK2/c-Jun/p53 pathways. In conclusion, LRRC4 is not only a brain-specific gene but also a tumor suppressor gene for glioma. In order to get further proof for a suppressor, molecular mechanism of LRRC4 down-regulation in glioma had been analyzed in primary gliomas and glioma cell lines. However, we did not find any genetic alteration such as mutation, deletion and rearragment in LRRC4 coding region.[Cloning and location of LRRC4 promoter]In order to explore the mechanism of LRRC4 down-regulation in glioma, we carried out research on LRRC4 transcription regulation. The LRRC4 promoter region in the 5'end of humans was predicted using the PromoterInspector and PromoterScan programs. The CpG island was found using CpGplot from the European Molecular Biology Open Software Suite. The three program prediction results have some overlapping region. Then a genomic DNA fragment that spanned positions-2475 to-101 relative to the initiation codon ATG of LRRC4 was amplified by PCR. The PCR product was cloned into pGL3-enhancer luciferase report vector and was named as pGL3--2475/-101. Luciferase activity analysis showed that pGL3 -2475/-101 had similar luciferase activity as pGL3-control.The result suggested that the region-2475/-101 include the LRRC4 promoter. In order to further locate the LRRC4 promoter, four deletion constructs were created (pGL3-1483/-101, pGL3-835/-101, pGL3-293/-101 and pGL3-835/-293), originating from the construct pGL3-2475/-101. Then they were transfected into Cos7 and Hela cells and we found the region-835/-293 was necessary for LRRC4 promoter activity. In order to verify the result, we constructed pGL3-835/-293/eGFP plasmid and it was transfected into Cos7 and Hela cells. The result showed that eGFP expression could be drived by the sequence-835/-293 in Cos7 and Hela cells. In a word, we successfully cloned and located the LRRC4 promoter through the study above.[Methylation of LRRC4 promoter in glioma】Bioinformatics analysis shows that the LRRC4 promoter region has no TATA box or CAAT box. But it has high G/C content (approximately 70%) and characteristics of a CpG island. The result suggested the possibility that LRRC4 might be regulated through changes in the methylation status. Two glioma cell lines,30 primary glioma biopsies and three tissue specimens of normal brain were examined for LRRC4 methylation by MS-PCR. Both SF767 and SF126 cell lines showed complete methylation of the LRRC4 promoter. And the LRRC4 promoter was free from methylation in the three normal brain tissue samples, but was methylated to different extents in the 30 primary gliomas. To determine a more detailed map of the methylation in the LRRC4 promoter, we performed bisulfite sequencing around the promoter region of the LRRC4 gene in two glioma biopsies, two glioma cell lines and a normal brain tissue studied above. The results revealed densely methylated CpG sites within the promoter regions compared to normal brain tissue. All the results suggested that LRRC4 methylation is a tumor-specific event. It may distinguish tumors from normal tissue and serve as a promising biomarker for diagnosis. 【LRRC4 expression induced by 5-Aza-CdR]To demonstrate a functional association between LRRC4 promoter methylation and its gene inactivation, a DNA demethylating agent, 5-Aza-2'-deoxycytidine (5-Aza-CdR), was used to treat SF126 and SF767 cell lines. Methylation-specific PCR was used to examine the methylation status changes of LRRC4 promoter in SF126 and SF767 cell lines. LRRC4 mRNA expression in SF126 and SF767 cell lines treated by 5-Aza-CdR was detected by RT-PCR. The results indicated that LRRC4 promoter aberrant hypermethylation can be reversed and LRRC4 expression can be induced by 5-Aza-CdR. The results demonstrated that LRRC4 promoter methylation suppressed its expression and promoter methylation is the important molecular mechanism of LRRC4 inactivation in glioma. At the time, we also observed the effect on glioma cell lines cell growth and cell cycle of 5-Aza-CdR.5-Aza-CdR displayed a growth inhibitory effect on SF126 and SF767 cells and cell cycles were blocked at G0/G1 phase after 5-Aza-CdR treatment. Taken together, glioma cell lines SF126 and SF767 cell growth could be inhibited and cell cycles could be blocked by 5-Aza-CdR; methylation can be reversed and LRRC4 expression can be induced by 5-Aza-CdR. All these data suggests that LRRC4 may serve as a demethylation therapeutic target in glioma.[Methylation interruptting SP1 and E2F1 binding with LRRC4 promoter]To explore the mechanism of promoter methylation-mediated LRRC4 inactivation in glioma, transcription factors binding sites of LRRC4 promoter was analyzed by MatInspector program. There were several transcription factors in LRRC4 promoter. Five transcription factors binding sites were chosen for EMSA. However, specific binding bands could be detected only at E2F1(-655/-631) and SP1(-568/-547) binding sites. The result implied that E2F1 and SP1 involved in LRRC4 transcription regulation. In order to observe the effect on transcription factors binding of LRRC4 promoter methylation, oligonucleotide probes for E2F1(-655/-631) and SP1(-568/-547) binding sites were treated by methylase M.SssI and EMSA were carried out. Previous specific binding band disappeared. The results suggested that DNA methylation interrupted transcription fators E2F1 and SP1 binding with DNA. We also performed CFHP to verify the result in vivo. In glioma cells untreated by 5-Aza-CdR, E2F1 and SP1 binding with LRRC4 promoter could not be observed. However, when glioma cells were treated by 5-Aza-CdR for four days, we could observe the E2F1 and SP1 binding with LRRC4 promoter. Therefore we draw the conclusion that LRRC4 transcription are regulated by transcription factors E2F1 and SP1 and methylation interferes E2F1 and SP1 binding with LRRC4 promoter, which suppresses LRRC4 trancription. [Histone modification correlated with LRRC4 promoter methylation]Both DNA methylation and histone modification are not isolated epigenetic events. There are cross-talks between them. In order to analyze histone modifation correlated with LRRC4 promoter methylation, we carried out CHIP in glioma cells treated or untreated by 5-Aza-CdR using H3 acetylation, trimethyl-H3(lys4) and trimethyl-H3(Lys9) antibodies. In glioma cells untreated by 5-Aza-CdR, LRRC4 promoter could be amplified only from DNA immunoprecipitated by trimethyl-H3 (Lys9) antibodies. In glioma cells treated by 5-Aza-CdR for four days, the level of trimethyl-H3(Lys9) around LRRC4 promoer was decreased while the level of H3 acetylation and trimethyl-H3(lys4) were increased. On the basis of these data, we draw the following conclusions that trimethyl-H3(Lys9) is correlated with LRRC4 promoter methylation, which suppresses LRRC4 expression and H3 acetylation and trimethyl-H3(lys4) are correlated with LRRC4 promoter demethylation, which activates LRRC4 expression.DNA methylation patterns appear to have tumor-type specificity. Genome-wide different CpG island methylation status in different type tumor constructs tumor-specific DNA methylation profile. Geome-wide profile of DNA methylation in gliomas not only contributes to uncover all around the molecular mechanism of glioma initiation and development, but also provide very important and valuable clue such as biomarker for early diagnosis and prognosis, or useful target for therapy. Though some progress has been made in research on methylation genes in gliomas, candidate genes approach was adopted in most study to aim at several or a flock of genes. So far, no study is aimed directly at genome wide to analyze completely and systematically aberrant methylation genes in gliomas. In some degree, lacking high-throughput methods is the bottleneck in this investigation.In this study, a novel and high-throughput technique called MeDIP combination with CpG island microarrays were applied to screen aberrant methylation genes in 6 glioma biopsies(T1,T2,T3,T4,T5,T6) and 4 normal brain tissue(N1,N2,N3,N5). Under our standardization, we identified 562 hypermethylation loci and 108 hypomethylation loci, including 325 hypermethylation genes and 74 hypomethylation genes. They distribute evenly on every chromosome and susceptible chromosome is not found. The differential methylation genes involve in extensive biological process such as cell communication, signal transduction, cell adhesion, cell migration, apoptotic program, metabolic process, transport, post-translational protein modification, nervous system development, and etc. At the same time they also refer to several cell pathways such as MAPK signaling pathway, Wnt signaling pathway, Jak-STAT signaling pathway and Cell cycle. Therefore, the differential methylation genes may contribute to initiation and development of glioma.In order to validate the results of microarrays,8 hypermethylation genes (ANKDD1A, SST, SIX3, GAD1, PHOX2B, HIST1H3E, PCDHA13 and PCDHA8), whose methylation region is a CpG island and locate in promoter, were chosen to verify and investigate in 40 glioma biopsies,11 normal brain tissue and 4 glioma cell lines using MassArray system. ANKDD1A, SST, SIX3, GAD1, PHOX2B, HIST1H3E, PCDHA13 and PCDHA8 methylation level in normal tissue is 13.29%±2.12%,9.72%±2.42%,16.34%±4.13%, 11.81%±3.76%,23.06%±4.49%,21.39%±3.13%,31.46%±4.28% and 31.66%±12.77% respectively. However, ANKDD1 A, SST, SIX3, GAD1, PHOX2B, HIST1H3E, PCDHA13 and PCDHA8 methylation level in glioma biopsies is 36.24%±23.71%,35.08%±19.44%,51.36%±21.25%, 29.22%±15.13%,50.38%±19.15%,41.67%±25.56%,55.77%±21.77% and 69.42%±19.49% respectively; and in glioma cell lines is 76.14%±27.99%,47.64%±31.61%,72.55%±25.49%,62.21%±16.80%, 22.44%±18.00%,63.05%±20.35%,58.09%±30.00% and 55.00%± 21.29% respectively. Except PCDHA13, methylation level of genes in glioma biopsies and glioma cell lines is significantly increased compared to normal brain tissue (p<0.01). Except PCDHA8 and PCDHA13, methylation level of genes in glioma cell lines is higher than those in glioma biopsies(p<0.01).There is a group of probes against LRRC4 on CpG island microarray of NimbleGen. The probes covering region is a CpG island and located LRRC4 translation initiation site upstream from 1472bp to 2218bp. The CpG island microarray screening results showed that the region aginst LRRC4 is unmethylated in glioma. To furtherly verify methylation of LRRC4 active promoter region spanned positions-835 to-293 relative to the initiation codon ATG of LRRC4 in glioma, MassArray is also carried out to detect this region in 40 glioma biopsies,11 normal brain tissue and 4 glioma cell lines studied above. The LRRC4 promoter average methylation level is 15.47%±3.96%,38.29%±21.69% and 68.57%±25.46% in normal brain tissue, glioma biopsies and glioma cell lines, respectively. Methylation level of LRRC4 active promoter in glioma biopsies and glioma cell lines is significantly increased compared to normal brain tissue (p<0.01). The result is in accordance with our previous study. Based on the research, we can draw the conclusion that LRRC4 is one of the methylated genes in glioma and we can classify LRRC4 into DNA methylation profile of glioma. Taken together, we have identified aberrant methylation genes in gliomas. This not only lays a foundation for mapping geome-wide aberrant DNA methylation in gliomas, but also provides important clues to find biomarker for diagnosis and prognosis, or useful target for therapy.