Dual-Specificity Tyrosine Phosphorylation-Regulated Kinase 3 Loss Activates Purine Metabolism and Promotes Hepatocellular Carcinoma Progression

Cancer cells metabolize different energy sources to generate biomass rapidly. The purine biosynthetic pathway was recently identified as an important source of metabolic intermediates for these processes. However, very little was known about the regulatory mechanisms of purine metabolism in hepatocellular carcinoma (HCC). We explored the role of dual-specificity tyrosine (Y) phosphorylation-regulated kinase 3 (Dyrk3) in HCC metabolism. Dyrk3 was significantly down-regulated in HCC compared with normal controls. Its introduction in HCC cells markedly suppressed tumor growth and metastasis in xenograft tumor models. Mass spectrometric analysis of metabolites suggests that the effect of Dyrk3 on HCC occurred at least partially through down-regulating purine metabolism, as evidenced by the fact that inhibiting purine synthesis reverted the HCC progression mediated by the loss of Dyrk3. We further provide evidence that this action of Dyrk3 knockdown requires nuclear receptor coactivator 3 (NCOA3), which has been shown to be a coactivator of activating transcription factor 4 (ATF4) to target purine pathway genes for transcriptional activation. Mechanistically, Dyrk3 directly phosphorylated NCOA3 at Ser-1330, disrupting its binding to ATF4 and thereby causing the inhibition of ATF4 transcriptional activity. However, the phosphorylation-resistant NCOA3-S1330A mutant has the opposite effect. Interestingly, the promoter activity of Dyrk3 was negatively regulated by ATF4, indicating a double-negative feedback loop. Importantly, levels of Dyrk3 and phospho-NCOA3-S1330 inversely correlate with the expression of ATF4 in human HCC specimens. Conclusion: Our findings not only illustrate a function of Dyrk3 in reprograming HCC metabolism by negatively regulating NCOA3/ATF4 transcription factor complex but also identify NCOA3 as a phosphorylation substrate of Dyrk3, suggesting the Dyrk3/NCOA3/ATF4 axis as a potential candidate for HCC therapy.