Doublecortin-like kinase1 (DCLK1), a microtubule-associated protein (MAP), recently gained interest in the cancer research field. A recent meta-analysis, including gastric cancer (GC), showed that DCLK1 overexpression correlates with epithelial-to-mesenchymal transition (EMT), advanced and poorly differentiated GC, lymph node metastasis and reduced overall survival.
Our analysis of the stomach adenomas (STAD) data-set from the Cancer Genome Atlas (TCGA), showed that DCLK1-high expressing tumours significantly clustered within the genomic stable molecular subtype and the histologically diffuse type. Both subtypes have the worst overall survival and poor therapeutic response. Gene-set enrichment analysis showed DCLK1-high tumours correlate with EMT-up regulated gene signatures, stromal influx and matrix remodelling and are downregulated for immune-signatures. We are currently evaluating DCLK1 expression off 300 stomach cancer patients by immunohistochemistry on tissue microarrays. Preliminary results show that DCLK1 protein levels are higher in diffuse-type GCs compared to intestinal-type GCs.
We established a DCLK1-overexpressing MKN1 gastric cancer cell-line. The overexpression resulted in increased migration and invasion in vitro and in vivo. These findings support our TCGA-STAD data analysis where high DCLK1 levels correlated with EMT, chemokines, and stromal- and immune cell markers. Strikingly, we observed an overall increase in chemokine secretion when DCLK1 is overexpressed, ex vivo. CXCL12 is the one of the main upregulated chemokines; this is further supported by findings in the TCGA-STAD data set, which shows that DCLK1 and CXCL12 expression levels significantly correlate with each other. Furthermore, a DCLK1-inhibitor reversed migration, invasion and chemokine secretion in the DCLK1-overexpressing MKN1 cells to parental MKN1 cell levels, in vitro and in vivo. This suggests that DCLK1 could be a good target for poor prognosis GCs with high DCLK1 levels..
Thus far, the signalling cascade in which DCLK1 promotes cancer progression, by initiating EMT and increasing chemokine secretion is poorly understood. To answer these questions we will use mass spectrometry based studies on proteomics, phospho-proteomics and secretomic analysis, comparing parental vs DCLK1-overexpressing cells, with and without DCLK1-inhibitor.