Fig 2: Depletion of USP19 delays tumor growth in vivo. (a) Scheme of xenograft experiments using SKNMC inducible cell lines. (b) Doxycycline-induced shRNA knockdown against USP19 in vivo. Two mice with engrafted tumors of ~200 mm3 (shCtr and shUSP19#1) were treated for five days with doxycycline. Tumor lysates were analyzed by western blotting with anti-USP19. (c) Tumor growth rate of indicated cell lines and treatment of subcutaneously injected SKNMC cells, error bars as SD. (d) Tumors of two representative mice transplanted with SKNMC shUSP19#1 cells and treated with doxycycline or PBS. (e) Immunohistochemical analysis of representative sections doxycycline or PBS treated SKNMC shUSP19#1 tumors using an USP19 antibody.

Fig 3: USP19 mechanism of action.An in silico study was performed in order to analyze the relationship between USP19 expression levels and different pathway activation status. A USP19 mRNA expression among primary breast carcinomas according to their intrinsic subtype. Expression analysis showed a consistent upregulation in luminal A and B subtypes compared with basal-like and Her2 subtypes. B Luminal A/B primary breast cancers divided into low (n = 77) or high (n = 209) USP19 mRNA expression levels. C Significantly activated pathways among Luminal A/B tumors with high USP19 mRNA expression (n ≥ 77, SAM test, p < 0.01). Western blotting was performed in order to analyze LRP6 protein expression in breast cancer cells upon USP19 genetic manipulation. D Top: Western blot quantification in control or USP19 silenced MDAMB231 cells (n = 6, one-way ANOVA, Dunnett’s multiple comparison test. shRNA#1 p = 0.0416 and shRNA#2 p = 0.0102), bottom: representative image of a blot. E Top: Western blot quantification in MCF7 cells overexpressing control or GFP-tagged USP19 constructs (n = 5, one-way ANOVA, Dunnett’s multiple comparison test. WT p = 0.0484, C506S p = 0.8469 and WTΔTM p = 0.9968), bottom: representative image of a blot. F Wound-healing assays were performed in order to analyze endogenous LRP6 silencing effects in MCF7 cells overexpressing WT or C506S mutant versions of USP19. Cells were stably transduced with a control vector (‘ctrol’, PLKO.1 empty vector), or shRNAs targeting LRP6 (sh#1 and sh#2). Scratching with a pipette tip made a gap on a monolayer of the different cell cultures, and time-lapse imaging monitored the number of migrating cells across the border. The graph shows the gap covered area (mm2) after 8 h (n = 3, Kruskal–Wallis and Dunn’s multiple comparison test for WT or C506S overexpressing MCF7 cell lines, analyzed separately. WT overexpressing MCF7 cell line: sh#1 p = 0.0341 and sh#2 p = 0.2021. C506S overexpressing MCF7 cell line: sh#1 p = 0.7422 and sh#2 p > 0.9999).

Fig 4: EWS-FLI1 ubiquitination is modulated by USP19. (a) EWS-FLI1 interacts with USP19. 3xflag-EWS-FLI1 and 3xmyc-USP19 were co-expressed in HEK293T cells for 48 h. After co-immunoprecipitation, lysates were analyzed by western blotting as indicated. (b) USP19 depletion increases EWS-FLI1 ubiquitination. A673 cells were transiently incubated with three different siRNAs against USP19 or a control siRNA for 24 h followed by co-expression of 3xflag-EWS-FLI1 and HA-ubiquitin for 48 h. Ubiquitination of EWS-FLI1 was analyzed by western blot analysis using anti-HA antibody. (c) EWSR1 also immunoprecipitates with USP19. 3xflag-EWS-FLI1, 3xflag-EWSR1 or 3xflag-FLI1 were co-expressed with 3xmyc-USP19 in HEK293 cells for 48 h. After co-immunoprecipitation, lysates were analyzed by western blotting as indicated. (d) USP19 depletion increases EWSR1 monoubiquitination. A673 cells were transiently incubated with two different siRNAs against USP19 or a control siRNA for 24 h followed by co-expression of 3xflag-EWSR1 and HA-ubiquitin for 48 h. Ubiquitination of EWSR1 was analyzed by western blot analysis using anti-HA antibody. Full-length blots of all immunoprecipitates are presented in Supplementary Fig. S6.

Fig 5: Validation and characterization of USP19 as a regulator of cell migration.A Workflow used to select a candidate regulatory gene. MDAMB231 cells were stably transduced with control empty vector shRNA (control) or two different shRNAs (#1 & #2) targeting USP19. B Efficiencies of shRNA-mediated knockdown were confirmed by RT-PCR (top, n = 5, one-way ANOVA, Dunnett’s multiple comparison test. shRNA#1 p = 0.0329 and shRNA#2 p = 0.0013) and Western Blotting (middle and bottom, n = 4, one-way ANOVA, Dunnett’s multiple comparison test. shRNA#1 p = 0.0227 and shRNA#2 p = 0.0006). (C) Crystal violet (CV) staining was used to determine cell growth over time. Cells were seeded onto wells and allowed to attach. At the indicated time points, cells were fixed and then stained at the end of the experiment. The graph on the top shows the mean relative CV absorbance every 24 h. Doubling time was calculated for control and USP19 silenced cell lines on the bottom (n ≥ 3, Kruskal–Wallis, Dunn’s multiple comparison test. shRNA#1 p > 0.9999 and shRNA#2 p > 0.9999). The migratory potential was evaluated by two different experiments. D Transwell assay: After 24 h of incubation, USP19-depleted cells were stained for microscopic examination and the number of migratory cells was compared to control cells. The graph shows the number of migratory cells per transwell membrane (n = 4, one-way ANOVA, Dunnett’s multiple comparison test. shRNA#1 p = 0.0187 and shRNA#2 p = 0.0001). E Wound-healing assays: scratching with a pipette tip made a gap on a monolayer of the different cell cultures, and time-lapse imaging monitored the number of migrating cells across the border. After 8 h, cells exhibited different levels of migration. The graph on the left shows the gap covered area (mm2) after 8 h (n = 16, one-way ANOVA, Dunnett’s multiple comparison test. shRNA#1 p < 0.0001 and shRNA#2 p < 0.0001) and the images on the right show representative areas in a wound-healing experiment at the indicated time points. Scale bar = 100 μm.