GATOR2-dependent mTORC1 activity is a therapeutic vulnerability in FOXO1 fusion-positive rhabdomyosarcoma

Abstract

Oncogenic FOXO1 gene fusions drive a subset of rhabdomyosarcoma (RMS) with poor survival; to date, these cancer drivers are therapeutically intractable. To identify new therapies for this disease, we undertook an isogenic CRISPR-interference screen to define PAX3-FOXO1-specific genetic dependencies and identified genes in the GATOR2 complex. GATOR2 loss in RMS abrogated aa-induced lysosomal localization of mTORC1 and consequent downstream signaling, slowing G1-S cell cycle transition. In vivo suppression of GATOR2 impaired the growth of tumor xenografts and favored the outgrowth of cells lacking PAX3-FOXO1. Loss of a subset of GATOR2 members can be compensated by direct genetic activation of mTORC1. RAS mutations are also sufficient to decouple mTORC1 activation from GATOR2, and indeed, fusion-negative RMS harboring such mutations exhibit aa-independent mTORC1 activity. A bisteric, mTORC1-selective small molecule induced tumor regressions in fusion-positive patient-derived tumor xenografts. These findings highlight a vulnerability in FOXO1 fusion-positive RMS and provide rationale for the clinical evaluation of bisteric mTORC1 inhibitors, currently in phase I testing, to treat this disease. Isogenic genetic screens can, thus, identify potentially exploitable vulnerabilities in fusion-driven pediatric cancers that otherwise remain mostly undruggable.

Keywords: Cancer; Drug therapy; Genetics; Oncology; Signal transduction.

Figure 1. An isogenic screen identifies PAX3-FOXO1 dependencies.

(A) Immunoblot of lysates from P3F⁺ and P3F^KD cells demonstrating decreased abundance of both PAX3-FOXO1 and its target FGFR4. (B) Differential growth of P3F⁺ and P3F^KD cells in 2D (left: CellTiterGlo after 72 hours of tissue culture; n = 10 wells per condition) or 3D (right: crystal violet quantitation of soft agar colony formation in 0.4% agarose over 4 weeks; n = 3 wells per condition) growth. (C) Schematic of CRISPR-interference screen to identify PAX3-FOXO1 genetic dependencies. Each condition was carried out in experimental duplicate. (D) Screen results. Gene-level scores are plotted on the basis of log2 fold-change in P3F⁺ (x-axis) or P3F^KD (y-axis) conditions. Color indicates FDR-adjusted P value by Mann-Whitney test for significance of fold-change in the P3F⁺ condition. Black asterisks represent negative control sgRNA. The black box indicates genes selected for further study. (E) Results from the box in D arranged by physical interactions identified in the STRING database (20). Outlines correspond to selectivity calls from the DepMap database (19). Coloring indicates the P value, by Mann-Whitney test, for stronger CERES gene effect score in FP (n = 6) rather than FN (n = 5) cell lines in the DepMap. (F) Schematic of the role of the GATOR2 complex, including MIOS and WDR24, in regulation of mTORC1. Positive regulators of mTORC1 are shown in green.

Figure 2. Reducing PAX3-FOXO1 dosage decreases GATOR2 dependence in RMS.

(A) Schematic of in vitro competitive fitness assays. Equal numbers of differentially labeled cells were plated after puromycin selection, then maintained in puromycin for 12 days of growth. Separate competitions were carried out to assess the effects of GATOR2 loss in P3F⁺ cells (filled circles) or P3F^KD cells (open circles). (B) FP cell lines RMS13 and Rh41 show suppressed growth after GATOR2 knockdown, compared with the control (filled bars), assessed by 1-way ANOVA and post hoc Dunnett’s test. The experiment was repeated with cells that harbored combined knockdown of GATOR2 or control by sgRNA and knockdown of PAX3-FOXO1 by shRNA (open bars). Loss of PAX3-FOXO1 partially rescued cells from GATOR2 knockdown, assessed by 2-way ANOVA and post hoc Sidak’s test comparing the effects of GATOR2 knockdown between P3F⁺ and P3F^KD cells. (C) NSG mice were implanted with 2 × 10⁶ Rh30 cells transduced with sgRNA targeting WDR59 (green) or the nontargeting control (gray) after puromycin selection for 7 days and monitored for tumor development. Rates of tumor growth are shown (n = 6 mice per condition). (D) Tumor volumes after 32 days of growth at time of euthanasia compared by a 2-tailed t test. (E) Schematic of in vivo competitions. A 4:1 excess of P3F^KD cells to P3F⁺ cells, transduced with either nontargeting sgRNA (gray) or sgWDR59 (green), was implanted in NSG mice. (F) Relative compositions of sgCTL (gray) or sgWDR59 (green) tumors after 21 days of in vivo growth. Δ, change.*P < 0.05; ***P < 0.001; ****P < 0.0001.

Figure 3. Loss of GATOR2 abrogates mTORC1 activation and slows cell cycle progression in RMS.

(A) RMS13 cells transduced with the indicated sgRNA or shRNA plus sgRNA combination were seeded and then grown in serum-free RPMI-1640 lacking arginine, leucine, and lysine for 3 hours. Cells were either lysed directly (–) or after 10 minutes of stimulation with aa-replete medium (+). Representative immunoblot demonstrates the effects of GATOR2 knockdown and/or PAX3-FOXO1 knockdown on mTORC1 signaling. (B) Quantitation of p70S6K phosphorylation in response to aa stimulation from replicates of immunoblot, as in A. Significance was assessed by 1-way ANOVA and Dunnett’s multiple comparisons test (comparing GATOR2 knockdown to control) or Sidak’s multiple comparisons test (comparing P3F⁺ to P3F^KD cells within each genotype). (C) Quantification of correlation coefficients for mTOR and LAMP2 colocalization by structured illumination microscopy of aa-starved cells fixed immediately or after 10 minutes of stimulation with full RPMI-1640 (n = 10–15 cells imaged per condition). (D) GATOR2 knockdown cells were subjected to a single thymidine block for 18 hours, then washed and plated in full medium for 8 hours prior to fixation and permeabilization. Percentage of cells in G1 was calculated by propidium iodide staining and analysis on a flow cytometer. CTL, control.**P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 4. Activation of mTORC1 overcomes GATOR2 dependence.

(A) Schematic of nutrient-sensing and mitogen pathways that converge on mTORC1 activation. (B) RMS13 cells were transduced with the indicated sgRNA, and mTORC1 activity was assessed by immunoblot after aa starvation and stimulation. (C) Quantification of levels phosphorylated p70S6K normalized to total p70S6K (see Supplemental Figure 4 for 2 additional replicates and quantification of m7-GTP binding by 4EBP1). Differences in aa-stimulated levels of p70S6K phosphorylation across sgDEPDC5 conditions were NS by 1-way ANOVA. (D) Competition assays confirmed that reactivation of mTORC1 can rescue cell growth after knockdown of MIOS or WDR59, but not WDR24, SEH1L, or SEC13. One-way ANOVA with post hoc Dunnett’s test was significant for differences between sgDEDPC5 and sgDEPDC5-WDR24 (P = 0.0271) and sgDEPDC5-SEH1L (P < 0.0001), and was NS for others. (E) Immunoblot of aa-starved and -stimulated RAS-mutant FN (BIRCH, JR1, RD) or PAX3-FOXO1 FP (RMS13, Rh30, Rh41) cells shows increased basal mTORC1 activity in the former; representative immunoblot from 3 independent replicates. Asterisks indicate long exposures of the same blots. (F) Quantitation of phosphorylation of p70S6K and RPS6 under aa starvation conditions in FP and FN RMS cells. Significance of differences assessed by 2-tailed t test. CTL, control; max, maximum. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5. MAP kinase activation downstream of oncogenic RAS is necessary and sufficient for aa-independent activation of mTORC1 in RMS.

(A) PAX3-FOXO1–positive RMS13 cells transduced with WT or mutant NRAS were subjected to aa starvation and stimulation. Representative immunoblot from 3 replicates of lysates and m7-GTP pulldowns demonstrates that NRAS^Q61H is sufficient to drive aa-dependent phosphorylation of p70S6K and RPS6, but not of 4EBP1. (B) Competition assays in NRAS^Q61H expressing RMS13 cells compared with parental cells from Figure 2B; differences between parental and NRAS^Q61H expressing cells assessed by 2-tailed t test. (C) RMS13 or RMS13 cells expressing NRAS^Q61H were incubated with DMSO, 1 μM ulixertinib (ERKib), or 100 nM sapanisertib (TORib) for the duration of a 3-hour aa starvation with or without 10-minute stimulation. Immunoblots demonstrate that sapanisertib suppresses p70S6K phosphorylation in either condition but incompletely suppresses RPS6 phosphorylation. By contrast, ulixertinib restores aa control of RPS6 phosphorylation. (D) Quantification of p70S6K and RPS6 phosphorylation from C (n = 3 independent replicates). Significance of basal p70S6K or RPS6 phosphorylation assessed by 1-way ANOVA and Dunnett’s multiple comparisons test. CTL, control. *P < 0.05; **P < 0.01.

Figure 6. The bisteric mTORC1 inhibitor RMC-6272 induces complete remission in FP RMS PDXs.

(A) FP RMS cell lines were seeded in 96-well plates and treated across a dose range of the allosteric mTOR inhibitor rapamycin and RMC-6272. After 6 days, viability was measured by alamarBlue assay. (B) Representative immunoblot from 3 independent replicates of RMS13 cells treated with 1 nM rapamycin or RMC-6272 shows similar dephosphorylation of p70S6K but poor suppression of cap-dependent translation by rapamycin, as measured by competitive binding of EIF4G and 4EBP1 to EIF4E in an m7-GTP pulldown assay. (C) Waterfall plot demonstrating tumor response after 28 days based on PDX and drug treatment. (D) Tumor growth curves of 2 FP RMS PDXs treated with RMC-6272 at the indicated doses. Differences between vehicle (Veh) and drug treatment were measured by a linear mixed-effects regression model with Dunnett’s multiple comparisons test. (E) Mice harboring the 13759 PDX were euthanized 24 hours after administration of vehicle or the indicated doses of RMC-6272. Immunoblot of whole-cell lysates (WCLs) and m7-GTP pulldowns from excised, flash-frozen PDX demonstrates mTORC1 inhibition with RMC-6272. *P < 0.05; **P < 0.01.