Figure 1.
Pim-1 directly interacts with and serine phosphorylates FLT3.
A. Pim-1 interacts with FLT3 in cells. Lysates of MV4-11 and MOLM-14 FLT3-ITD cells, and of EOL-1 FLT3-WT cells, all of which express Pim-1, were immunoprecipitated with anti-Pim-1 and immunoblotted with anti-FLT3, and, reciprocally, immunoprecipitated with anti-FLT3 and immunoblotted with anti-Pim-1. 10% of the total cell lysate (TCL) used for immunoprecipitation is shown as input control. Interaction between Pim-1 and FLT3 was detected in MV4-11 and MOLM-14, and also in EOL-1, indicating that Pim-1 interacts with both FLT3-ITD and FLT3-WT. B. Pim-1 interacts with FLT3 in vitro. Lysates from Ba/F3-ITD and MV4-11 FLT3-ITD cells were incubated with immobilized GST-tagged Pim-1 fusion protein or control GST, followed by immunoblotting with anti-FLT3 antibody. Total cell lysate is shown as input control. GST-Pim-1 recombinant protein, but not GST control, bound to FLT3-ITD, confirming direct binding of Pim-1 kinase to FLT3-ITD in vitro. C. Pim-1 interacts with and phosphorylates FLT3 on serine residues between amino acids 571 and 993. An in vitro kinase assay was performed with purified GST-FLT3 peptide (571-993) containing the Pim-1 consensus phosphorylation site at amino acid 935 (RKRPS) as the substrate and immunoprecipitates from MV4-11 cells with anti-Pim-1 antibody or IgG control on Sepharose A beads as the kinase, in the absence and presence of 10 µM quercetagetin, followed by immunoblotting with the antibodies indicated. Immunoprecipitates with IgG and with beads alone are shown as negative controls. Direct interaction between Pim-1 and FLT3 resulted in serine phosphorylation of FLT3, which was completely inhibited by 10 µM quercetagetin. D. Quercetagetin inhibits FLT3 serine phosphorylation by Pim-1. After treatment with 10 µM quercetagetin for three hours, FLT3-ITD was immunoprecipitated from BaF3-ITD and MOLM-14 cell lines and immunoblotted for phoshoserine and FLT3. Quercetagetin decreased FLT3 serine phosphorylation in both cell lines. E. shRNA knockdown of Pim-1 inhibits serine phosphorylation of FLT3. MOLM-14 cells were transiently transfected with control shRNA or Pim-1 shRNA for 72 hours. FLT3-ITD was immunoprecipitated and then immunoblotted for phosphoserine. The total cell lysates were also immunoblotted with anti-Pim-1 to confirm Pim-1 knockdown. GAPDH is shown as a loading control. Pim-1 knockdown decreased FLT3 serine phosphorylation, consistent with serine phosphorylation of FLT3 by Pim-1.
Figure 2.
Pim-1 stabilizes 130 kDa FLT3-ITD A.
Pim-1 inhibition with quercetagetin destabilizes 130 kDa and stabilizes 150 kDa FLT3. MV4-11 cells were treated with cycloheximide (CHX) in the presence or absence of 10 µM quercetagetin for the indicated time periods and immunoblotted for FLT3 and the loading control GAPDH. The expression ratio of each FLT3 isoform to GAPDH was quantitated by densitometric analysis, then normalized to “0” time point control and the normalized values are shown graphically as the percent of the FLT3 isoform remaining at various time points. B. Pim-1 inhibition with AR339 destabilizes 130 kDa and stabilizes 150 kDa FLT3. Cells were studied as in A, with 500 nM AR339. C. Pim-1 inhibition increases FLT3-ITD ubiquitination. MV4-11 and MOLM-14 cells were incubated with and without 10 µM quercetagetin for three hours and the protein lysates were immunoprecipitated with FLT3 antibody or control IgG and immunoblotted for ubiquitin. Total FLT3 in the protein lysates is also shown as control. A similar effect was seen in MV4-11 cells incubated with AR339 at a range of concentrations. D. Pim-1 inhibition enables glycosylation of 130 kDa FLT3, forming 150 kDa FLT3. MV4-11 cells were treated with 10 uM quercetagetin or DMSO control with and without the glycosylation inhibitor 2-deoxy-D-glucose (2-DG) for 24 hours, and immunoblotted with the indicated antibodies. Expression of 150 kDa FLT3 increased in the presence of quercetagetin, but this increase was abrogated by co-incubation with 2-DG, consistent with Pim-1 inhibition by quercetagetin enabling glycosylation of 130 kDa FLT3 to form 150 kDa FLT3.
Figure 3.
Pim-1 knockdown also destabilizes FLT3-ITD.
A. Silencing Pim-1 with shRNA decreases FLT3-ITD expression. Pim-1 knockdown was performed as described in Materials and Methods, and expression of FLT3, Pim-1 and GAPDH control in Ba/F3-ITD, MOLM-14 and MV4-11 cells was measured by immunoblotting. Deceased expression of FLT3 was seen. (B, C). Pim-1 knockdown decreases half-life of 130 kDa FLT3. FLT3-ITD-expressing cells were treated with cycloheximide (CHX) combined with either 10 µM quercetagetin or DMSO for the times indicated. Expression of FLT3 and GAPDH control in MOLM-14 (B) and MV4-11 (C) cells was measured by immunoblotting. Decreased half-life of 130 kDa FLT3 was seen in both cell lines.
Figure 4.
Mutation of serine 935 to alanine inhibits glycosylation of 130 kDa FLT3-ITD.
Translation of FLT3-ITD in PT67 cells stably expressing the S935 (A), S935A (B) or phosphomimetic S935D (C) or S935E (D) FLT3-ITD constructs was blocked with cycloheximide and cells were collected at the indicated time points. Protein lysates were prepared and FLT3 and GAPDH expression was measured by immunoblotting following SDS-PAGE electrophoresis. A representative immunoblot is shown for each construct. The intensity of FLT3-ITD was quantified densitometrically and normalized in relation to that of GAPDH. Changes in intensity of GAPDH-normalized S935 (A), S935A (B), S935D (C) and S935E (D) 150 kDa and 130 kDa FLT3-ITD over time were then plotted graphically, in relation to 0 time point. The S935A mutant showed absence of 150 kDa FLT3, consistent with findings when glycosylation is inhibited by incubation with 2-deoxy-D-glucose, as shown in Figure 2D. The findings are consistent with the S935A mutation inhibiting glycosylation.
Figure 5.
Pim-1 inhibition decreases FLT3-ITD binding to its chaperone proteins calnexin and HSP90.
Ba/F3-ITD, MV4-11 and MOLM-14 FLT3-ITD cells were incubated with the Pim-1 inhibitor 10 µM quercetagetin or DMSO control for 3 hours, and protein-protein interaction was studied by immunoprecipitation (IP) followed by immunoblotting (IB) with the antibodies indicated. Treatment with quercetagetin disrupted FLT3-ITD binding to calnexin (A) and decreased its binding to HSP90 (B).
Figure 6.
Pim-1 inhibition decreases FLT3 Y591 and STAT5 phosphorylation and Pim-1 expression in FLT3-ITD cells.
Ba/F3-ITD and MV4-11 cells were treated with 10 µM quercetagetin or DMSO control for the times indicated, followed by immunoblotting (IB) with the antibodies indicated.
Figure 7.
Pim-1 inhibition sensitizes FLT3-ITD cells to FLT3 inhibition.
Ba/F3-ITD cells were treated with 100 nM PKC412 with and without 10 uM quercetagetin (top), and with 10 nM sorafenib with and without 10 uM quercetagetin (bottom) and also with quercetagetin alone for 24 and 48 hours, and apoptosis was measured by annexin V/propidium iodide staining, detected by flow cytometry and analyzed with CellQuest Pro software. Percentages of apoptotic cells were significantly greater following both 24-hour (p = 0.009) and 48-hour (p = 0.005) treatment with PKC412 and quercetagetin, compared with PKC412 alone, and following 24-hour (p = 0.03) and 48-hour (p = 0.05) treatment with sorafenib and quercetagetin, compared with sorafenib alone. The cells were also treated with the FLT3 inhibitor quizartinib at 1 nm with and without the Pim-1 inhibitor AR339 at 500 nM, demonstrating increased percentages of apoptotic cells at 24 and at 48 hours (p<0.001 for both).