BACKGROUND: 5-fluorouracil (5-FU) is widely used in the treatment of gastric and colorectal cancer. Recent microrarray studies associated different gene lists with 5-FU resistance. A major challenge in the genomic era is to find the most validated genes, and to decipher the regulatory networks responsible for the expression changes in a set of co-regulated transcripts. Our aim was to find genes repeatedly associated with 5-FU resistance, and to identify transcription factors (TFs) having overrepresented binding sites (TFBSs) in the promoter regions of genes associated with 5-FU resistance. MATERIALS AND METHODS: The analyzed data originated from 5 different publications describing genome-wide gene expression patterns associated with 5- FU resistance in gastric and colorectal cancer. First, a data warehouse containing all genes associated with resistance was set up. 39 genes were identified which were repeatedly associated with resistance. Of these, using the EZ-Retrieve web service, proximal promoter sequences were available for 33 genes. The MotifScanner software was used to detect TFBSs in this set of sequences. RESULTS: A total of 200 different TFBSs were identified. Using the statistics tool of the Java program TOUCAN, 4 binding sites were found to be significantly overrepresented: NFKappaB50 (p = 0.01), EGR2 (p = 0.027), EGR3 (p = 0.007), and NGFIC (or EGR4) (p = 0.001). These genes intercept apoptotic pathways at multiple locations in the tumor cells. CONCLUSION: We identified a consensus gene list associated with 5-FU resistance, performed an in silico comparative promoter analysis, and highlighted the potential implication of some TFs in the development of chemoresistance.
BACKGROUND:5-fluorouracil (5-FU) is widely used in the treatment of gastric and colorectal cancer. Recent microrarray studies associated different gene lists with 5-FU resistance. A major challenge in the genomic era is to find the most validated genes, and to decipher the regulatory networks responsible for the expression changes in a set of co-regulated transcripts. Our aim was to find genes repeatedly associated with 5-FU resistance, and to identify transcription factors (TFs) having overrepresented binding sites (TFBSs) in the promoter regions of genes associated with 5-FU resistance. MATERIALS AND METHODS: The analyzed data originated from 5 different publications describing genome-wide gene expression patterns associated with 5- FU resistance in gastric and colorectal cancer. First, a data warehouse containing all genes associated with resistance was set up. 39 genes were identified which were repeatedly associated with resistance. Of these, using the EZ-Retrieve web service, proximal promoter sequences were available for 33 genes. The MotifScanner software was used to detect TFBSs in this set of sequences. RESULTS: A total of 200 different TFBSs were identified. Using the statistics tool of the Java program TOUCAN, 4 binding sites were found to be significantly overrepresented: NFKappaB50 (p = 0.01), EGR2 (p = 0.027), EGR3 (p = 0.007), and NGFIC (or EGR4) (p = 0.001). These genes intercept apoptotic pathways at multiple locations in the tumor cells. CONCLUSION: We identified a consensus gene list associated with 5-FU resistance, performed an in silico comparative promoter analysis, and highlighted the potential implication of some TFs in the development of chemoresistance.