The role of salt-inducible kinase 2 in hepatocellular carcinoma

Abstract

Salt-inducible kinase 2 (SIK2), an AMP-activated serine/threonine protein kinase, regulates cellular metabolism by phosphorylating transducer of regulated CREB activity (TORC) and inhibits CREB-mediated gene expression. Recent evidence revealed that SIK2 was frequently overexpressed in cancers and contributes to cancer cell survival, proliferation, chemo-resistance, and metastasis. However, little is known about its function in hepatocellular carcinoma (HCC), which has an extremely high mortality rate due to its detrimental effect on liver function and the lack of effective treatment. In this study, I showed that SIK2 was overexpressed in HCCs. Stable overexpression of SIK2 in HCC cells promoted cisplatin resistance. Moreover, overexpression of SIK2 drastically enhanced cancer invasion, whereas silencing of SIK2 suppressed. Western blotting analysis proved that SIK2 upregulated mesenchymal markers, suggesting that it promotes epithelial-mesenchymal transition (EMT). RT-qPCR analysis and gelatin zymography revealed that SIK2 promoted the activity of MMP-2 and MMP-9. SIK2 also upregulated the MAPK/ERK pathway upon starvation and the replenishment of growth factors. Taken together, these data suggested that SIK2 promotes HCC invasion through upregulation of EMT, MMPs, and ERK. In the next chapter, I investigated whether SIK2 is a centrosome kinase in HCC as described previously in ovarian cancer. Immunofluorescence confocal microscopy showed that SIK2 was mainly localized in the cytoplasm of HCC cells while it is strongly localized at the centrosome when activated. The constitutively active mutant, SIK2 T175D, and the phosphorylated SIK2 at T175 were exclusively localized at the centrosomes. Oppositely, the kinase-dead mutation, SIK2 K49M, did not specifically localize at the centrosome and was found mainly in the cytoplasm, suggesting that the kinase activity of SIK2 is crucial for its localization to the centrosomes. Using a panel of truncation mutants, the kinase domain of SIK2 was shown to be crucial for its centrosome targeting. Besides, co-immunoprecipitation coupled with mass spectrometry analysis of SIK2 showed that SIK2 interacted with PP2A and cofilin. Overexpression of SIK2 reduced the phosphorylation of cofilin, increased G-actin/F-actin ratio, and the formation of lamellipodia, suggesting that SIK2 regulated cofilin and mediated de novo formation of the actin filament, thereby promoting invasion. Given that PP2A has been reported to dephosphorylate cofilin and SIK2 has been shown to form a holoenzyme complex with PP2A, SIK2 may upregulate the cofilin pathway by modulating the phosphatase activity of PP2A. Taken together, this study revealed the mechanisms of how SIK2 promotes HCC invasion and its role as a centrosome kinase, proving SIK2 to be a molecular target for metastatic HCC treatment.