Molecular determinants of centrosome separation at the onset of mitosis are attractive targets for cancer therapy. For example, the centrosome kinase NEK2 plays an essential role in the initiation of centrosome separation via phosphorylating C-NAP1 that mediates centrosome pairing. However, NEK2 activity reaches peak level in late S phase and G2 and not at the onset of mitosis. This suggests that other molecular events are required for initiation of centrosome separation in prophase. %9Here we show that SIK2 plays an essential role in centrosome separation in mitosis. Using a monoclonal antibody against endogenous SIK2 and ectopic expression of myc/flag tagged SIK2, we established that SIK2 is a centrosome kinase. Depletion of SIK2 using siRNA in the ovarian cancer cell line SKOv3 resulted in a significant increase in monopolar spindles compared to control transfected cells (39.4% v 9.7%, respectively, p<0.001, t-test). Overexpression of SIK2-myc/flag resulted in a significant increase of interphase cells with premature separation of the centrosomes (PSC) in flag-positive cells compared to flag-negative cells (15.3%, 3.3%; p=0.01, t-test). These results directly implicated SIK2 in the regulation of centrosome pairing. We confirmed that PKA, a known potent inhibitor of SIK2, is recruited to the centrosome during interphase via its regulator PKAR2\#945;. At the onset of prophase, the centrosome localization of SIK2 was significantly increased while that of PKAR2\#945; became undetectable and this was coupled with the initiation of centrosome separation. Depletion of PKAR2\#945; resulted in a significant increase in PSC in interphase cells compared to control transfected cells (9.3% and 0.8% respectively; p<0.001, t-test). More specifically, inhibiting the centrosome localization of PKAR2\#945; by depleting its anchoring protein AKAP450 also resulted in PSC (10.9% in AKAP450 depleted cells v 1.1% in controls; p<0.001, t-test). Importantly, knockdown of SIK2 rescued the PSC following loss of PKAR2\#945; or AKAP450 suggesting that SIK2 mediated the centrosome separation following loss of centrosomal PKAR2\#945;. Using ectopic expression of C-NAP1-GFP we showed that it co-localized with SIK2 in the centrosome. We also tested the therapeutic potential of SIK2 and found that either the downregulation of SIK2 or its conditional expression in stably transfected SKOv3 cells under doxycyclin control resulted in a significant reduction of cell proliferation (p=0.017 and p<0.001, respectively; t-test). This suggested that an optimal level of SIK2 is required for proliferation. Finally, in a panel of 45 ovarian cancers we found the SIK2 was expressed at low levels and was overexpressed in 29% of cases. These results identify a previously unrecognized pathway for the control of centrosome pairing that may determine the timely onset of centrosome separation in mammalian cells and that may provide a target for therapy in a fraction of ovarian cancers.
Citation Information: In: Proc Am Assoc Cancer Res; 2009 Apr 18-22; Denver, CO. Philadelphia (PA): AACR; 2009. Abstract nr LB-43.
100th AACR Annual Meeting-- Apr 18-22, 2009; Denver, CO