Fig. 5

DUB1 associates with TAZ and modulates TAZ stability in gastric cancer cells A: Analysis of the intracellular localization of DUB1 and TAZ by immunofluorescence staining. MGC803 cells were cultured in normal medium before fixation. The intracellular localization of DUB1 (green) and TAZ (red) is shown. Nuclei (blue) were stained with 4’,6-diamidino-2-phenylindole (DAPI). B: DUB1 and TAZ were mainly localized in the nucleus in MGC803 cells. A subcellular protein fractionation kit (Thermo Scientific, 78,840) was used for nuclear/cytoplasmic fractionation. Tubulin and histone-3 were used as the cytoplasmic and nuclear controls, respectively. C: The Co-IP assay revealed an association between endogenous DUB1 and TAZ in MGC803 cells. MGC803 cells were harvested with RIPA lysis buffer. Co-IP was performed using an antibody as indicated. D-E: DUB1 domain structure and domain deletion mutants used in the study (full length, USP domain, central domain and CTD). Full-length TAZ and TAZ deletion mutants used in the study (full length, ΔTA domain, ΔWW domain, ΔTBD). F: DUB1 was found to interact with TAZ through its USP domain. HEK293 cells were transfected with 2 µg of Myc-TAZ together with full-length or mutant Flag-DUB1 (full length, USP domain, central domain and CTD). After 24 h, the cells were harvested with NP-40 lysis buffer. Co-IP was performed using an anti-Myc antibody. The possible DUB1 interacting domains were detected with an anti-Flag antibody. G: TAZ was found to interact with DUB1 through its WW domain. HEK293 cells were transfected with 2 µg of Flag-DUB1 together with full-length or mutant Myc-TAZ (full length, ΔTA domain, ΔWW domain, ΔTBD). After 24 h, the cells were harvested with NP-40 lysis buffer. Co-IP was performed using an anti-Flag antibody. The possible DUB1 interacting domains were detected with an anti-Flag antibody. H: In the presence of the proteasome inhibitor MG132, the effect of DUB1 on TAZ did not result in a further increase in the TAZ protein level. MGC803 cells were transfected with 50 nM siDUB1 or siControl. After 24 h, the cells were treated with 10 mM MG132/vehicle for 6 h. Cell lysates were prepared for western blot analysis. The results are representative of three independent experiments. I: In the presence of the proteasome inhibitor MG132, the stabilizing effect of DUB1 on TAZ did not result in a further increase in the TAZ protein level. HEK293 cells were transfected with 0.5 µg of the Flag or Flag-DUB1 plasmid. After 24 h, the cells were treated with 10 mM MG132/vehicle for 6 h. Cell lysates were prepared for western blot analysis. The results are representative of three independent experiments. J: DUB1 overexpression increased the TAZ half-life in HEK293 cells. HEK293 cells were transfected with 0.5 µg of the Flag-tag or Flag-DUB1 plasmid. After 24 h, the cells were treated with 100 mM cycloheximide/vehicle for the indicated times. Cell lysates were prepared for western blot analysis. The results are representative of three independent experiments. The relative density of the TAZ protein band was measured by ImageJ software. K: DUB1 depletion decreased the TAZ half-life in MGC803 cells. MGC803 cells were transfected with 50 nM siControl or siDUB1. After 24 h, the cells were treated with 100 mM cycloheximide/vehicle for the indicated times. Cell lysates were prepared for western blot analysis. The results are representative of three independent experiments. The relative density of the TAZ protein band was measured by ImageJ software