NG2 interacts with FLT3 to render ligand-independent activation of FLT3 signaling and downregulation of NR3C1 via the AP-1 complex. (A) Experimental design of the NG2 immunoprecipitation and MS analysis. (B) Heat map of NG2 partners detected by MS after NG2 immunoprecipitation. (C) Predicted region for NG2-FLT3 interaction using the iFrag prediction method. (D) FLT3 immunoprecipitation followed by NG2 blotting. (E) Colocalization of NG2 and FLT3 by ImageStream in MLLr B-ALL patient-derived xenograft (PDX). Human leukocyte antigen (HLA) marker depicts the human population. (F) Schematic representation of FLT3 pathway. (G) Immunoblots showing the phosphorylation status of FLT3 and its downstream target STAT5 in NG2^WT and NG2^KO SEM cells. (H) Percentage of live NG2^WT SEM cells treated with DX in the presence or absence of the FLT3 inhibitors sorafenib and midostaurin. (I) Percentage of live NG2^WT, NG2^KO, and NG2^WTAP1^KO SEM cells on treatment with DX. (J) NR3C1 expression detected by qPCR in NG2^WT SEM cells in the presence or absence of sorafenib and midostaurin. (K) NR3C1 expression detected by qPCR in NG2^WT, NG2^KO, and NG2^WTAP1^KO SEM cells. (L) Experimental design for in vivo experiments using NG2-expressing MLL-AF4⁺ PDXs treated with DX alone or in combination with sorafenib (top panel). Tumor burden (percentage of blasts) in PDX-transplanted mice at the beginning (day 0) and at the end (day 13) of treatment in the indicated groups (bottom panel). (M) Experimental design for in vivo experiments using NG2-expressing MLL-AF4⁺ primary cells treated with standard induction therapy (VXL) alone or in combination with sorafenib (top panel). Tumor burden (percentage of blasts) in transplanted mice at the end (day 20) of treatment in the indicated groups (bottom panel). Each point depicts 1 mouse. A mouse is considered in CR when the percentage of blasts in BM ≤0.5%. ∗P < .05, ∗∗P < .01, ∗∗∗P < .001, unpaired Student t test. Ns, not significant.