Fig 2: tnFGFR1 activates KIT and FLT3 in primary leukemia cells. Western blot analysis of KIT and FLT3 expression in leukemic cells from primary, tnFGFR1 transduced stem cells shows upregulation of both genes, compared to those cells transduced with BCR-FGFR1m (A). A similar correlation was seen in primary human AML cells that either express or do not express tnFGFR1 (B). (C) qRT-PCR detection of KIT, FLT3 and FGFR1 mRNA expression for human AML samples used for western blotting in (B) shows increased levels for all three genes in tnFGFR1 expressing cells. The preferential activation of KIT and FLT3 is further validated by flow cytometry analysis of primary leukemia cell from mouse models (D). Using two different shRNAs targeting tnFGFR1 (E, above) in primary tnFGFR1 transformed bone marrow cells, qPCR analysis (N = 3) shows extensive knockdown of tnFGFR1, Flt3, and Kit mRNA compared with cells transduced with a scrambled shRNA (SCR). Western blot analysis of the same cells (E, below) shows effective reduction in FLT3 and KIT protein levels in cells treated with either shRNA. ChIP analysis in primary tnFGFR1 transformed cells isolated from mouse bone marrow, using primers P1 from an upstream intergenic region, P2 within the Kit (F, left) or Flt3 (F, right) promoters and P3 within intronic regions of the genes, shows enrichment only for the P2 defined promoter regions (N = 3). ChIP was performed using antibodies that are specific to tnFGFR1 and also to the MYC-tag that was included in the transfection construct, both of which show the same result. ChIP using IgG alone was performed as a background control. Kaplan-Meier analysis of an AML cohort (N = 422, GSE37642) demonstrates a highly significant decrease in overall survival in patients with high expression levels of both FLT3 and GZMB compared with those showing low-level expression. No significant change in survival was found for different levels of expression of FGFR1 or KIT (G). ns = not significant. *p = 0.05, **p = 0.01, ***p = 0.001

Fig 3: BC interacts with STS1 and STS2 HP domains and promotes phosphorylation of FLT3 and cKIT by inhibiting STS1 and STS2. (A,B) Molecular docking analysis of the binding of BC with STS1 (A) or STS2 (B). Left: The representative docking pose of BC with the HP domain STS1 (5VR6) or STS2 (5WDI). BC and the key residues of protein receptor are shown as sticks. The red dashed lines indicate the hydrogen bonds formed. Right: The 2D docked conformation showing the hydrophobic contacts (red eyelash shape) between BC and the HP domain of STS1 or STS2. (C,D) Capillary electrophoresis analysis of the binding of BC to STS1 (C) or STS2 (D). BC, STS1 or STS2 samples were prepared in 20 mM PBS at pH 7.0, and electrophoresis was run under 20 kV voltage. Red (C) or blue (D) arrows represent STS1 or STS2 protein peaks, respectively. (E,H) HEK293T cells were transiently transfected with FLT3 (E) or cKIT (H) constructs, and 48 h later, cells were stimulated with 10 µg/mL BC for different periods of time at 37 °C. Phosphorylated FLT3 (p-FLT3) or cKIT (p-cKIT) and endogenous STS1/STS2 proteins were detected by Western blot (n = 3). GAPDH was used as internal control. (F,G) FLT3 construct was transfected into HEK293T cells together with or without pcDNA-STS1 (F) or -STS2 (G). Then, 48 h later, cells were stimulated with BC for 2 min at 37 °C. The overexpressed proteins and phosphorylated FLT3 were detected by Western blot (n = 3). Vec indicates empty vector control. (I,J) cKIT construct was transfected into HEK293T cells together with or without Flag-tagged STS1 (I) or STS2 (J). Then, 48 h later, cells were stimulated with BC for 2 min at 37 °C. The overexpressed proteins and phosphorylated cKIT were detected by Western blot (n = 3). Vec indicates empty vector control. All data are shown as mean ± SEM, ** p < 0.01 (Student’s t test).

Fig 4: BC increases the proportion of primitive HSC population and colony-forming ability of human CD34+ cord blood cells in vitro. (A,B) Flow cytometry analysis of FLT3 phosphorylation in CD34+ UCB cells cultured in vitro with or without BC (10 µg/mL) treatment for 2–15 min. (A) Flow cytometry histogram showing the intensity of p-FLT3 Tyr591 versus cell counts. (B) The mean fluorescence intensity (MFI) analysis of p-FLT3 Tyr591 from three independent experiments. (C) Representative flow cytometry profiles to show the percentage of CD34+CD38- and CD34+CD38-CD45RA-CD90+ cells in CD34+ UCB cells cultured in vitro with or without BC treatment for 4 days or 7 days. Gates shown identify the CD34+CD38- and CD34+CD38-CD45RA-CD90+ populations from CD34+ cells. (D,E) The proportion and absolute number of CD34+CD38- and CD34+CD38-CD45RA-CD90+ cells in (C). Data represent three independent experiments. (F) Colony counting of HSPCs derived CFUs from 2 × 103 CD34+ UCB cells cultured in vitro with or without BC treatment for 7 days. Data represent triplicate experiments. All data error bars are shown as mean ± SEM, * p < 0.05, ** p < 0.01 (Student’s t test).

Fig 5: BC promotes the proliferation of mouse BMMNCs and the expansion of their HSPC population in vitro. (A,B) Flow cytometry analysis of FLT3 phosphorylation in mouse BMMNCs cultured in vitro with or without BC (10 µg/mL) treatment for 5–15 min. (A) Flow cytometry histogram showing the intensity of p-FLT3 Tyr591 versus cell counts. (B) The mean fluorescence intensity (MFI) of p-FLT3 Tyr591 from three independent experiments. (C) Cell counting of mouse BMMNCs cultured in vitro with or without BC treatment at different timepoints, which was presented as relative cell number to the cell number at day 0. Data represent three independent experiments. (D) Representative flow cytometry profiles to show the percentages of LSK (Lineage-Sca-1+cKIT+) in mouse BMMNCs cultured in vitro with or without BC treatment for 4 days or 7 days. Gates shown identify the LSK populations from Lineage- (Lineage negative cells) population. (E) The proportion and absolute number of LSK population in (D). Data represent three independent experiments. (F) Colony counting of HSPCs derived CFUs from 1 × 104 BMMNCs cultured in vitro with or without BC treatment for 4 days. CFU-GM: colony-forming unit of granulocyte and macrophage; BFU-E: burst-forming unit-erythroid; CFU-GEMM: colony-forming unit of granulocyte, erythrocyte, monocyte and megakaryocyte. Data represent triplicate experiments. All data are shown as mean ± SEM, * p < 0.05, ** p < 0.01 (Student’s t test).