Neomorphic PDGFRA extracellular domain driver mutations are resistant to PDGFRA targeted therapies

Fig. 1. Activating PDGFRA mutations differentially promote malignant phenotypes. a Functional activity of a series of PDGFRA mutations in the transient transfection assay using Ba/F3. The transformation activity of each mutation was tested at least three times. The cell viability in the absence of IL-3 of duplicates was measured 7 days after infection with lentiviral virus carrying PDGFRA WT or mutants and presented relative to that of cells expressing PDGFRA WT (Dunnett’s multiple comparisons test). b The distribution of PDGFRA missense mutations in the GENIE dataset across different domains. Total number of PDGFRA missense mutations per tumor type is shown in the middle of each chart. c Expression of PDGFRA WT and mutants in MCF10A stable cell lines was examined by western blot. d Relative cell proliferation of MCF10A parental and cells stably expressing PDGFRA WT or mutants in the absence of EGF/insulin was measured every 24 h. The cell proliferation of three independent replicates relative to day 0 are presented as mean ± SD (exponential growth equation). e MCF10A parental, PDGFRA WT, and mutant expressing cells were grown 3D in growth factor-reduced Matrigel for 10 days. Bar, 100…

Fig. 2. Y288C protein has altered glycosylation and localizes at endoplasmic reticulum (ER). a Protein lysate of MCF10A expressing PDGFRA WT or Y288C were digested with or without Endo H or PNGase F and the band shift of PDGFRA was analyzed by SDS-PAGE. b Cell surface expression of PDGFRA in MCF10A parental (gray), PDGFRA WT (blue) or Y288C (green) expressing cells was assessed by flow cytometry. c MCF10A cells expressing PDGFRA WT, Y288C, V561D, or D842V were fixed and stained with anti-PDGFRA (green), anti-GM130 (Golgi marker; red), or anti-calnexin (ER marker; red), and DAPI (nucleus; blue) to visualize the subcellular colocalization of PDGFRA with Golgi and ER. Bar, 25 μm. d An OptiPrep gradient ultracentrifugation was used to resolve the membrane fractions (100k) of MCF10A cells stably expressing PDGFRA WT (upper panel) or Y288C (lower panel). 11 fractions were collected. Equal volume of lysate of each fraction were separated by SDS-PAGE and immunoblotted with antibodies against PDGFRA, calnexin (ER marker), GM130 (cis-Golgi marker), and syntaxin 6 (trans-Golgi marker). Heat maps showing the distribution and the relative levels of indicated proteins across different factions were…

Fig. 3. Y288C protein is constitutively active at ER and Golgi. a MCF10A cells stably expressing PDGFRA WT (right) or Y288C (left) were serum-starved for 16 h and stimulated with PDGF-BB (50 ng ml-1) for 15–60 min. Whole cell lysates were resolved by SDS-PAGE and blotted with indicated antibodies to analyze for the activation of PDGFRA and different downstream signaling pathways. Corresponding total proteins were used as loading controls. b Relative cell proliferation of MCF10A parental, PDGFRA WT or Y288C expressing cells in the absence of serum of three independent replicates was assessed for 4 days (exponential growth equation). c Protein lysate of MCF10A expressing PDGFRA WT or Y288C were resolved by SDS-PAGE in the presence or absence of β-mercaptoethanol. Monomeric and dimerized PDGFRA were detected by anti-PDGFRA antibody, and ERK2 was used as loading control. The band intensities of monomers and dimers from three independent experiments were measured by densitometry and normalized to ERK2. The levels of PDGFRA monomers and dimers were presented as mean ± SD relative to WT. The statistical significance of dimer vs the corresponding monomer was assessed by Student’s t-test. d Y288C…

Fig. 4. Y288C protein activates signaling pathways in distinct subcellular compartments. a Phosphorylation of the Y288C receptor at the ER and Golgi was analyzed by western blot of the OptiPrep gradient ultracentrifugation resolved membrane fractions. MCF10A cells stably expressing PDGFRA WT or Y288C were serum-starved for 16 h and treated with or without monensin (500 nM) or BFA (1 μM) for 2 or 4 h respectively. b Cells were fixed and stained with anti-PDGFRA (green), anti-GM130 (Golgi marker; red) or anti-calnexin (ER marker; red), and DAPI (nucleus; blue) to examine the subcellular localization of PDGFRA after treatments. Bar, 25 μm. c Protein was extracted and analyzed for the activities of Akt, STAT3, STAT5, ERK1/2 signaling pathways by western blot with indicated antibodies. Corresponding total proteins were used as loading controls

Fig. 5. Y288C promotes cell proliferation through Akt, ERK1/2, and STAT3. MCF10A cells stably expressing Y288C was transfected with siRNA specifically target, Akt1, MAPK1/3, STAT3, or STAT5, respectively. a The knockdown efficiency was verified by western blot 48 h after transfection. 24 h after transfection, cells were trypsinized, counted and plated for b cell viability assay (Student’s t-test) and c 3D growth in growth factor-reduced Matrigel for 4 days. The relative cell viability of three independent replicates was assessed with Prestoblue 4 days after transfection. Bar, 100 μm. The area of 50 spheroids are presented as mean ± SD (Dunnett’s multiple comparisons test). ***p < 0.001

Fig. 6. PDGFR inhibitors have limited efficacy against Y288C mutation. MCF10A cells expressing PDGFRA mutants were treated with DMSO or 2.44nM-10 μM of a PDGFR inhibitors (Imatinib, Nilotinib, Sunitinib, and Crenolanib), b PI3K/mTOR inhibitor (GSK2126458), MEK inhibitor (GSK1120212B), or STAT3 ASO (AZD9150) for 3 days. Each treatment was performed in triplicate and cell viability is presented as mean ± SD. The IC50 was calculated with Graphpad Prism 6