SUMO化调控GATA2稳定性以控制子宫浆液性癌的侵袭

Background: Uterine serous carcinoma (USC) is an aggressive uterine cancer subtype that is responsible for 40% of uterine cancer deaths. We recently demonstrated that USCs lose expression of the zinc-finger transcription factor GATA2 which drives USC invasion, predicts USC recurrence, and is closely correlated with poor cancer-related and overall survival. However, the mechanistic basis of GATA2 downregulation in USC remains unknown. Methods: GATA2 RNA transcripts were correlated to patient outcomes derived from The Cancer Genome Atlas (TCGA). GATA2 gene methylation was assessed in isolated genomic DNA from patient USC tumors by bisulfite sequencing. Ark1 and Ark2 USC cell lines were utilized for in vitro studies. GATA2 half-life was measured using cycloheximide chase experiments paired with inhibitors of phosphorylation (staurosporine), acetylation (C646), SUMOylation (ML792), cysteine peptidase activity (N-Ethylmaleimide/NEM), and proteasome-dependent degradation (MG132). Anti-GATA2 immunoprecipitations were performed with custom anti-GATA2 antibodies (Im et al, 2005). Depletion of PIAS2, SENP1, and SUMO2/3 was performed using commercially available siRNAs. Levels of GAPDH, GATA2, PIAS2, SENP1, and SUMO2/3 were measured by western blotting. USC invasion was assessed using Matrigel-coated membrane transwell inserts. Results: GATA2 IHC protein levels showed no correlation with GATA2 gene body or proximal promoter DNA methylation, and GATA2 transcript levels did not predict patient outcome across TCGA USC cases. Post-translationally, USC GATA2 protein half-life was approximately 60 minutes after cycloheximide treatment. Co-treatment with the SUMOylation inhibitor ML792 or the proteasome inhibitor MG132 prolonged GATA2 half-life compared with vehicle, whereas NEM treatment, which inhibits de-SUMOylation and de-ubiquitination, shortened GATA2 half-life. Direct GATA2 SUMOylation was confirmed by anti-GATA2 immunoprecipitation and SUMO2/3 western blot. A candidate siRNA-based approach found that depletion of the E3 SUMO ligase PIAS2 elevated GATA2 levels, while depletion of the SUMO peptidase SENP1 reduced GATA2 levels. siRNA-mediated SUMO2/3 depletion significantly increased GATA2 levels in USC cells and suppressed USC invasion in vitro compared to siScramble controls. Conclusion: USC GATA2 levels are determined by post-translational mechanisms. GATA2 protein has a rapid 60-minute half-life determined by SUMOylation and proteasome-mediated degradation. Depletion experiments support PIAS2 as the GATA2-targeting E3 SUMO ligase and SENP1 as the GATA2-targeting SUMO peptidase. SUMO inhibition increased USC GATA2 levels and suppressed USC invasion, suggesting that SUMO-targeting agents may suppress USC spread through upregulation of GATA2 levels.