Fig 2: Pharmacologic inhibition of USP15 exacerbates RIII in mice model. a C57 BL/6J mice received an intraperitoneal injection of 0.1 mg/kg USP15-IN-1 for 3 consecutive days before 14 Gy WBI and 30 min after WBI. Similarly, the control mice were exposed to 14 Gy WBI and received injections of vehicle (an equivalent volume of normal saline) at the same frequency. b Kaplan-Meier survival analysis of mice after 14 Gy WBI (n = 15 per group). ***P < 0.001 for vehicle vs. 0.1 mg/kg USP15-IN-1. c Body weights of irradiated mice in each group after 14 Gy WBI. d Small intestinal tissue samples were collected from the control and irradiated mice treated with vehicle/USP15-IN-1 at 6 h and 3.5 days after 14 Gy WBI. e, f Intestine length of mice in each group at 3 days post-WAI. g Representative H&E-stained, PAS-stained, Ki-67-stained, TUNEL-stained, 8-OHdG-stained and γ-H2AX-stained intestinal sections of various groups. Scale bar: 100 μm. h Villus height of mice at 3 days post-WAI. i Number of PAS+ cells per villus at 3 days post-WAI. j Number of Ki67+ cells per crypt and surviving crypts per circumference at 3 days post-WAI. k Number of TUNEL+ cells per crypt at 6 h post-WAI. l intensityof 8-OHdG at 6 h post-WAI. m Number of γ-H2AX+ cells per villus at 6 h post-WAI. *P < 0.05; **P < 0.01; ***P < 0.001; n = 6/group

Fig 3: MYEOV attenuates miRNA-mediated suppression of TGFBR2 and USP15. a Binding sites of miR-30c-2-3p in TGFBR2-3′-UTR and USP15-3′-UTR. b Relative luciferase activities in indicated cells transfected with miR-30c-2-3p mimic and luciferase reporters containing TGFBR2-3’UTR or USP15-3′UTR (each bar represents the mean ± SD derived from three independent experiments, two-tailed Student’s t test. *P < 0.05; ns, not significant). c The reporters containing TGFBR2-3′-UTR or USP15-3′-UTR were co-transfected with miR-30c-2-3p inhibitor, and luciferase activities were assessed after 48 h (each bar represents the mean ± SD derived from three independent experiments, two-tailed Student’s t test. *P < 0.05; ns, not significant). d Effects of silencing TGFBR2 or USP15 in indicated cells on luciferase activities of the TGF-β responsive reporter (each bar represents the mean ± SD derived from three independent experiments, one-way ANOVA followed by Dunnett’s multiple comparison test. *P < 0.05). e Effects of silencing TGFBR2 or USP15 on cell invasion as measured by Transwell invasion assay in the indicated NSCLC cells (each bar represents the mean ± SD derived from three independent experiments, one-way ANOVA followed by Dunnett’s multiple comparison test. *P < 0.05). f Luciferase assay of reporters for pGL3-TGFBR2-3′-UTR or pGL3-USP15-3′-UTR in indicated cells, co-transfected with MYEOV-cDNA, MYEOV-ATGmut or MYEOV-miRmut (each bar represents the mean ± SD derived from three independent experiments, one-way ANOVA followed by Dunnett’s multiple comparison test. *P < 0.05; ns, not significant). g WB analysis performed for TGFBR2 and USP15 with β-actin as loading control. h Model diagram of proposed MYEOV-mediated regulation of TGF-β signaling in NSCLC

Fig 4: USP15 knockdown enhances the sensitivity of ERα+ BC cells to tamoxifen.A, B Cells were exposed to USP15 siRNA/shRNA and tamoxifen (TAMO, 20 μM/L). Cell viability was detected by adding MTS. C Colony formation assay was done. D, E EdU staining assay was done and images were captured by fluorescence microscope. *p < 0.05, **p < 0.01, ***p < 0.001 versus each vehicle control. F T47D cells were treated with tamoxifen, USP15 siRNA or the combination of the two drugs. Cell apoptosis was tested by flow cytometry for apoptotic cells. G The pooled data has been showed. **p < 0.01 versus each vehicle control. H And protein was collected for western blot to evaluate Bcl-2 expression. I The treated cells were subjected to FACS for cell distributions. J The percent of cell number on each phase was showed. K Proteins were extracted from T47D treated with USP15 siRNA and tamoxifen, subjected to western blot for Cyclin D1, Rb, p-Rb, p27 expression. L A proposed mechanism for USP15 to regulate the ERα level in breast cancer.

Fig 5: Clinical relevance of MYEOV with expression levels of TGFBR2 and USP15 and activation of TGF-β signaling. a Expression of MYEOV is associated with TGFBR2 and USP15 expression levels and localization of SMAD3 in clinical NSCLC specimens. Two representative cases (high and low MYEOV) are shown. b Percentage of specimens showing high- or low- MYEOV expression in relation to the expression levels of TGFBR2, USP15 and nuclear or cytoplasmic SMAD3. The χ2-test was used to analyze statistical significance. *P < 0.05. c GSEA plot showing MYEOV expression in relation to metastasis-associated gene signature and TGF-β-activated gene signature in the TCGA lung cancer dataset