Ovarian cancer cells, not normal cells, are damaged by Mirk/Dyrk1B kinase inhibition

Prior studies had shown that the Mirk/dyrk1B gene is amplified/upregulated in about 75% of ovarian cancers, that protein levels of this kinase are elevated in quiescent G0 cells and that Mirk maintains tumor cells in quiescence by initiating rapid degradation of cyclin D isoforms and by phosphorylation of a member of the DREAM complex. Depletion of Mirk/dyrk1B led to increased cyclin D levels, an elevated reactive oxygen species (ROS) content and loss of viability. However, many normal cells in vivo are quiescent, and therefore, targeting a kinase found in quiescent cells might be problematic. In our study, Mirk kinase activity was found to be higher in ovarian cancer cells than in normal cells. Pharmacological inhibition of Mirk/dyrk1B kinase increased cyclin D levels both in quiescent normal diploid cells and in quiescent CDKN2A-negative ovarian cancer cells, but led to more active CDK4/cyclin D complexes in quiescent ovarian cancer cells, allowing them to escape G0/G1 quiescence, enter cycle with high ROS levels and undergo apoptosis. The ROS scavenger N-acetyl cysteine reduced both the amount of cleaved poly(ADP-ribose) polymerase (PARP) and the extent of cancer cell loss. In contrast, normal cells were spared because of their expression of cyclin directed kinase (CDK) inhibitors that blocked unregulated cycling. Quiescent early passage normal ovarian epithelial cells and two strains of quiescent normal diploid fibroblasts remained viable after the inhibition of Mirk/dyrk1B kinase, and the few cells that left G0/G1 quiescence were accumulated in G2+M. Thus, inhibition of Mirk kinase targeted quiescent ovarian cancer cells.