Abstract
Abstract 225
Cancer cells are characterized by abnormal DNA methylation, including overall genomic hypomethylation with concurrent region-specific hypo- and hyper-methylation, causing aberrant activation of some genes and the silencing of others. Three DNA methyltransferase (DNMT) enzymes catalyze DNA methylation in eukaryotic cells, DNMT1, DNMT3A, and DNMT3B. We discovered previously that cancer cells exhibit aberrant splicing of the DNMT3B gene, which produces transcripts containing premature stop codons that encode truncated proteins lacking the catalytic domain. When we bred transgenic mice expressing DNMT3B7, one of the aberrantly spliced DNMT3B isoforms found most commonly in cancer cells, with the Eμ-Myc mice, a mouse model for B cell lymphomas, we observed an acceleration of mediastinal lymphomagenesis along with changes in the expression of several genes involved in oncogenesis. The acceleration in tumorigenesis was associated with global DNA hypermethylation, and further analyses showed that these changes in DNA methylation were heterogeneous in tumors derived from Eμ-Myc/DNMT3B7 mice, a phenomenon reminiscent of human tumors. We hypothesized that DNMT3B7 altered DNA methylation by functioning as a dominant negative isoform of full-length endogenous mouse Dnmt3b, and therefore tested a second mouse model that has defects in DNA methylation.
The introduction of Dnmt3b heterozygosity (Dnmt3b+/−) into the Eμ-Myc background accelerated mediastinal lymphomagenesis to an even greater extent, with more than 90% of the Eμ-Myc/Dnmt3b+/− mice developing mediastinal lymphomas within the first 120 days. This was also associated with an increase in global DNA methylation as measured by liquid chromatography-mass spectrometry, to a larger extent than in the Eμ-Myc/DNMT3B7 mice. Interestingly, the tumors from Eμ-Myc mice themselves showed global hypermethylation when compared to non-transformed cells from Eμ-Myc mice, suggesting that the transformation of cells that express Myc is a key aspect in the induction of global DNA hypermethylation.
These observations led us to the hypothesis that Myc-mediated tumorigenesis is particularly sensitive to changes in DNA methylation. Brenner et al. demonstrated that Myc binds to Dnmt3a/b and recruits the methyltransferases to promoter regions of Myc targets, leading to DNA hypermethylation in these regions. We have also found previously that DNNMT3B7 binds with full-length DNMT3B, by co-immunoprecipitation studies. We hypothesize that either in the presence of DNMT3B7 or with Dnmt3b heterozygosity, Myc-Dnmt3a/b binding at promoters is enhanced, which leads to hypermethylation and repression of gene expression. Using Mycbp, a gene that was repressed in Eμ-Myc/DNMT3B7 tumors, we demonstrated that its promoter region was hypermethylated in both Eμ-Myc/DNMT3B7 and Eμ-Myc/Dnmt3b+/− tumors. The E-box, a conserved sequence located ∼100bp upstream of the transcription start site that Myc binds specifically, was hypomethylated in the Eμ-Myc/DNMT3B7 tumors, suggesting that there was an enrichment of Myc binding at this region. Chromatin immunoprecipitation analyses confirmed increased binding of Myc at the E-box of Mycbp in the Eμ-Myc/DNMT3B7 tumors. Furthermore, we also demonstrated that Myc expression induced all the three DNA methyltransferases, suggesting that Myc-mediated lymphomagenesis may occur using a feedback loop which enhances expression of the DNA methyltransferases to regulate particular genes involved in tumorigenesis. This study offers an insight into the mechanism behind Myc-mediated tumorigenesis and provides evidence for the central role played by changes in DNA methylation patterns in this process.
Disclosures:
No relevant conflicts of interest to declare.
A novel role for BRCA1 in regulating breast cancer cell spreading and motility