DELINEATING REGULATORY PATHWAYS OF NOVEL CANCER STEM CELL MARKER, DCLK1-S, FOR TARGETING COLON CANCER STEM CELLS

Colon cancer is third most prevalent cancer in the United States, with significant health consequences for the patient and society. Cancer stem cells are believed to be resistant to conventional treatments resulting in relapse of the disease. Targeting cancer stem cells in addition to conventional therapy may result in better outcomes for patients. Colon cancer stem cells express several stem cell markers, including DCLK1. Recent reports suggest that the 5’ promoter of DCLK1 is increasingly methylated and silenced in colon cancers; however significant levels of DCLK1-protein are expressed by colon cancer cells and human adenocarcinomas, suggesting that DCLK1 measured in colon cancers is likely transcribed from an alternative promoter. As my first goal, this possibility was investigated and led us to discover that colon cancers express a short isoform of DCLK1 from a promoter within intron V, while normal colons mainly express long isoforms of DCLK1 from 5’promoter. The loss of DCLK1 expression in cancer cells has been reported to result in the loss of proliferative/tumorigenic/metastatic potential of colon cancer cells. RNAi methods used so far, target both isoforms of DCLK1, therefore my second goal was to use shRNA knockdown methods to specifically target DCLK1-S isoform, in order to delineate the biological role of cancer specific DCLK1-S isoform. The results of our studies suggest that DCLK1-S mediates the activity of transcription factor, NFATC2 (via the NFATC2 53SPPS56 motif), which in turn results in NFATC2 binding and activation of the COL3A1 promoter which enhances the invasive potential of colon cancer cells. The 5’(α)-promoter is differentially methylated in normal human colons vs. human colorectal cancers, however the methylation status of the IntronV(β)-promoter does not change. Therefore for my third goal, I investigated if differential expression of DCLK1-S in normal colons vs. human colorectal cancers was perhaps due to differences in transcriptional activity of the promoter in normal vs. cancer cells. Our studies demonstrate for the first time that FOXD3 is a potent transcriptional inhibitor of the IntronV(β)-promoter, resulting in the absence of DCLK1-S expression in normal human colons.