Fig 2: USP15 expression correlates with disrupted redox signaling in AML.A Normalized enrichment scores for the top 15 upregulated (red, top) and downregulated (blue, bottom), significantly altered gene sets in USP15hi versus USP15low AML patient samples from the BEAT-AML dataset. P-value < 0.05. B Selected gene set enrichment plots for USP15hi and USP15low AML groups. Shown is the enrichment of NPM1-mutated AML genes in USP15hi AML and enrichment of hypoxia-related genes in USP15low AML, (P < 0.00001). NES normalized enrichment score. C Schematic overview of the USP15-KEAP1-NRF2 pathway as it pertains to observations in USP15high and USP15low AML. Elevated USP15 is required to maintain redox homeostasis by directly sustaining KEAP1 protein expression and as a result inhibition of NRF2. In the absence of USP15, KEAP1 expression is diminished resulting in NRF2-mediated redox stress. D Immunoblot of USP15, KEAP1, NRF2, and GAPDH in normal CD34+ cells and a panel of AML cell lines. Representative data from 2 replicate experiments. E An X, Y coordinate dot plot using the quantities of USP15 and KEAP1 determined by mean gray area value in ImageJ, normalized to GAPDH for each sample from panel (D). Linear regression was performed to calculate the R-squared and P-values. The best-fit line is shown.

Fig 3: Inhibition of USP15 with a small molecule induces NRF2-medidated redox stress and impairs leukemic progenitors in vitro.A Immunoblot analysis of USP15, KEAP1, NRF2, and ACTIN/GAPDH in MV4;11 and OCI-AML3 cells treated with vehicle (DMSO, 0 µM) or the indicated concentrations of USP15 inhibitor for 24 h. Data reflects three independent assays. B Left; histogram flow plots of DCFDA analysis of cellular ROS in the indicated AML cells treated for 24 h with DMSO (gray) or 10 µM of USP15 inhibitor (green) with the raw MFI values set shown. Right; the normalized mean-fluorescent intensity of DCFDA in the AML cell lines versus their DMSO controls. A parametric, ratio-paired t test was performed. *P = 0.01. The centerline depicts the median. Data is from four independent cell lines. C Relative NQO1 expression in the indicated AML cell lines treated with vehicle (DMSO) or USP15 inhibitor (10 µM) for 24 h (n = 3 replicates). D Cell-Titer Glo relative luminescence was measured at 72 h with USP15 inhibitor in the indicated AML cell lines. Luminescence detected in inhibitor-treated cells was normalized to the vehicle control cells treated with DMSO (n = 3 replicates per dose). The IC50 values for each cell line are shown. E Total number of colonies formed by MV4;11, OCI-AML3, and MOLM14 cell lines in methylcellulose after 7 days of treatment with the USP15 inhibitor (10 µM). (n = 3 per group). Data reflects two independent assays. F The total number of colonies formed by three independent patient-derived-xenografted (PDX) AML samples and normal CD34+ cells after 7–10 days of treatment with the USP15 inhibitor (10 µM). *P < 0.05. G Representative colony images from healthy CD34+, OCI-AML3 leukemic cell line, and three PDX AML samples treated with DMSO or USP15 inhibitor.

Fig 4: Deletion of USP15 impairs murine AML in vivo but is tolerated during normal hematopoiesis.A Schematic overview of the generation of a murine model of MLL-AF9 myeloid leukemia using Usp15-/- and Usp15+/+ BM cells. BM cells were transduced with retrovirus encoding MLL-AF9 (MA9) and GFP. After the selection of GFP-expressing cells, lethally-irradiated mice were engrafted with the AML cells and then monitored for engraftment and overall survival. B Representative analysis of the leukemic cell burden in the BM of mice four weeks after engraftment with MLL-AF9;Usp15+/+ and MLL-AF9;Usp15-/- AML cells. C Summary of the leukemic cell burden in the BM of mice four weeks after engraftment with MLL-AF9;Usp15+/+ (n = 7 mice per group) and MLL-AF9;Usp15-/- (n = 6 mice per group) AML cells. *P < 0.05. D Kaplan-Meier analysis of overall survival of mice engrafted with MLL-AF9;Usp15+/+ (n = 7) and MLL-AF9;Usp15-/- (n = 6) AML cells. *P = 0.04. E Summary of the leukemic cell burden in the BM of mice at the time of death (“sick”) at ~10 weeks after engraftment with MLL-AF9;Usp15+/+ and MLL-AF9;Usp15-/- AML cells and remaining mice (“non-sick”) at 10 weeks. F Schematic overview of the generation of BM engraftment of Usp15-/- and Usp15+/+ BM cells into lethally-irradiated recipient mice. BM cells (CD45.2) from Usp15-/- and Usp15+/+ mice were engrafted into lethally-irradiated mice (CD45.1) and then BM and PB were analyzed by immunophenotyping. G Representative flow plots of the gating strategy for hematopoietic progenitor populations in the BM of recipient mice, including Lineage-negative (Lin-), Lineage-c-Kit+ (LK), Lineage-c-Kit+Sca1+ (LSK), granulocyte-monocyte progenitors (GMP), megakaryocyte-erythroid progenitors (MEP), and common myeloid progenitors (CMP), and lymphoid progenitors (CLP). H–J Frequency of LK, LSK, CMP, GMP, and MEPs populations from the BM of recipient mice transplanted with Usp15+/+ and Usp15-/- BM cells. (n = 5 mice per group). K Schematic overview of the generation of competitive BM chimeras using Usp15-/- and Usp15+/+ BM cells. BM cells (CD45.2) from Usp15-/- and Usp15+/+ mice and wild-type (WT) BM cells (CD45.1) were mixed at equal proportions and then engrafted into lethally-irradiated mice (CD45.1). Percent of CD45.2+ cells in the PB and BM was measured at the indicated time points. L Peripheral blood chimerism of cBMT recipients at 1, 4, and 7 months post-transplant (n = 10 mice per group). M Frequency of Usp1+/+ and Usp15-/- lineage-committed B-cell (B220), T-cell (CD3e), and myeloid (CD11b) progenitors in the bone marrow at 7-months post-cBMT (n = 10 mice per group).

Fig 5: USP15 knockdown protects against oxidative stress, apoptosis, and inflammation in liver I/R.A The protein expression of USP15 in the livers of normal human and patients undergoing liver transplantation, n = 6. (B) USP15 protein expression in the livers of mice exposed to 1 h of ischemia followed by 0, 1, 3, 6, and 12 h of reperfusion, n = 3. C USP15 protein expression in AML12 cells subjected to 12 h of hypoxia followed by 0, 1, 2, 4, and 8 h of reoxygenation. D–J AAV-TBG-USP15 shRNA was injected into mice followed by liver I/R. D USP15 mRNA expression, n = 6. E H&E staining. Scale bar, 100 µm. F Serum ALT, AST and LDH levels, n = 8. G Liver IL-1ß, IL-6 and TNF-a mRNA levels, n = 6. H Liver CAT, SOD, H2O2, and MDA contents, n = 8. I TUNEL staining. Scale bar, 100 µm. J USP15, p66Shc, Bcl2, Bax, and C-cas-3 protein expression. K AML12 cells were transfected with USP15 siRNA under H/R conditions. USP15, p66Shc, Bcl2, Bax and C-cas-3 protein expression, n = 3. #P < 0.05, ##P < 0.01.