Oncogenic role of a novel GC fusion and its targetability. (A) Schematic diagram of G3BP1 and CSF1R rearrangement and Sanger sequencing validation of the GC fusion. Exons 1 to 8 of G3BP1 were fused to exons 12 to 22 of CSF1R. Red line, breakpoint. (B) Fluorescence in situ hybridization study and schematic diagram showing rearrangement of G3BP1 (left) and CSF1R (right) in the diagnostic BM of GC patient. Normal BM sample was used as the control. Scale bar, 10 μm. For both the G3BP1 and CSF1R loci, the normal cell has 2 red (5') and 2 green (3') signals, whereas the GC cell has 1 red (5') and 2 green (3') signals, indicating the loss of the red (5') signal in GC cell. The transcription of G3BP1 and CSF1R occurs in different directions, necessitating an inversion for the formation of a functionally effective gene fusion. (C) Schematic representation of the lentiviral vectors. (D) M07E and TF1 cells transduced with EV, CSF1R, or GC lentivirus (72 hours) were cultured without cytokine (GM-CSF) for 4 hours, and cell lysates were prepared for western blotting using indicated antibodies. Representative results from 3 independent experiments. (E) M07E and TF1 cells transduced with EV, CSF1R, or GC lentivirus were grown in medium without cytokine (GM-CSF), and cell viability was determined by CellTiter-Glo Luminescent Cell Viability Assay. Results are presented as mean ± standard deviation (SD) from 3 independent experiments. (F) Left, representative flow cytometric analysis of human CD45⁺ (hCD45⁺) cells in BM, liver, and sarcoma of non-obese diabetic-severe combined immunodeficiency IL2rγ null (NSG) mice intravenously injected with TF1-EV or TF1-GC cells (3e6 cells) for 28 days. Right, quantification of hCD45⁺ cells. ∗∗ and ∗∗∗ indicate P < .01 and P < .001 vs EV, respectively, by Student t test. EV, n = 3; GC, n = 3. (G) GC patient cells (from diagnostic BM) and TF1-GC cells were treated with dasatinib, pexidartinib, and ruxolitinib. DMSO was used as the control. After 72 hours, cell viability was determined using the CellTiter-Glo Luminescent Cell Viability Assay. IC₅₀ was calculated by GraphPad Prism. (H) TF1-GC cells were treated with 100 nM dasatinib, 1000 nM pexidartinib, and 1000 nM ruxolitinib for 24 hours, and the induction of cell apoptosis was measured by Caspase-Glo 3/7 assay. ∗∗∗ indicate P < .001 vs DMSO by 1-way analysis of variance (ANOVA) followed by Dunnett test. Results are expressed as mean ± SD from 3 independent experiments. (I) TF1-GC cells were treated with 100 nM dasatinib, 1000 nM pexidartinib, and 1000 nM ruxolitinib for 4 hours, and phosphorylation of CSF1R, STAT5, and ERK1/2 was detected by western blotting. Representative results of 2 independent experiments. (J) Experimental design of in vivo drug treatment. NSG mice were intravenously injected with 3e6 TF1-Luc-GC cells and treated with DMSO, 20 mg/kg dasatinib, and 40 mg/kg pexidartinib daily for 21 days by oral gavage starting from day 7 after cell injection. Leukemia burden was monitored by bioluminescence imaging (BLI) every 7 days from day 7 to day 42 after transplant. Survival of mice was measured after discontinuing treatment. (K) Follow-up of BLI of mice detected by in vivo imaging system spectrum. (L) Quantification of bioluminescence signal (total flux) of mice over time. ∗, ∗∗, and ∗∗∗∗ indicate P < .05, P < .01, and P < .0001 vs DMSO, respectively, by 1-way ANOVA followed by Dunnett test (days 7, 14, 21, and 28 after transplant) or Student t test (day 35 after transplant). Results are presented as mean ± SD after log transformation. n = 4 for each group. (M) Kaplan-Meier survival curves of mice. ∗∗ indicates P < .01 vs DMSO by log-rank test. n = 4 for each group. DMSO, dimethyl sulfoxide; EGFP, enhanced green fluorescent protein; GM-CSF, granulocyte-macrophage colony-stimulating factor; IRES, internal ribosome entry site; SFFV, spleen focus-forming virus.