CANCER-RELATED GENE REGULATION MECHANISMS

Cancer is a disease characterized by deregulation of cellular functions. The means by which these changes occur may differ among cancer types. However, common to all cancers are that they have changed the genetic programs that determine their characteristics. Genetic programs are mainly regulated by the action of transcription factors. Changes in transcription factor activity and modification of promoter functions are therefore important changes in the development of cancer. In the first paper a bioinformatical method is developed that identify transcription factors important for a given tumor type. The basic assumption is that genes that show co-expression are likely to be regulated by the same set of transcription factors. The method consists on several steps. First co-expressed genes are identified using an arbitrary clustering algorithm. Genes with similar promoters are then identified and enrichment for evolutionarily conserved binding sites is determined. We show by this method that transcription factors known to be important for the development of a subtype of leukemia (AML) may be identified and, by the analysis of a time series data, that biologically relevant time dependent interactions between transcriptions factors and genes may be identified. The second paper focus on the promoter sequences. It is well established that DNA modulation by methylation affects the accessibility of transcription factors and that this may inhibit gene activity. To identify genes potentially regulated by promoter methylation we treated bladder cancer cell lines with the drug 5-aza-2’-cytidine that specifically inhibits methylation and thereby reactivates methylated promoters. Our major finding is that many genes shows patchy promoter methylation patterns that frequently affect high scoring and evolutionarily conserved transcription factor binding sites. In addition we identified several genes that may be of importance for bladder cancer development. The third paper is entirely focused on microRNAs. MicroRNAs are small RNA molecules with an approximate size of 20bp that function in analogy to transcription factors. MicroRNAs do, however, not bind to the promoter but to the mRNA and thereby inhibit translation or induce mRNA degradation, in both cases resulting in down regulation of gene activity. We use micro array technology to study alterations in microRNA expression in bladder cancer. We show that microRNA expression is associated with tumor stage and that mir-10a is associated with Ta tumors, that mir-7 is associated with FGFR3 mutations, and that mirs 452 and 452* is associated with lymph node metastases and bad prognosis