Figure 5.
Figure 5. Warfarin impairs hematopoiesis via the periostin/ integrin β3/ AKT axis. (A) Number of LKS cells expressing integrin β3 in control mice (red circles), control mice treated with periostin (blue squares), or mice treated with warfarin (green triangles) or warfarin plus periostin (purple inverted triangles) for 14 days (P = .01 for control vs warfarin and P = .003 for warfarin vs warfarin plus periostin; ANOVA, Tukey Test, n = 9-10). The data were generated in 2 independent experiments. (B) Protein lysates of 293T cells, transfected with an integrin β3-overexpressing construct and grown in CM from vehicle- or warfarin-treated macrophages for 7 days, were subjected to coimmunoprecipitation with an anti-integrin β3 antibody, and coimmunoprecipitation of periostin was assessed by immunoblotting, as indicated. Anti-IgG-isotype antibody was used as control in the co-immunoprecipitation. The input (left) shows the presence of integrin β3 (92 kDa) and periostin (90 kDa), and GAPDH (38 kDa) was used to control for equal loading. The CM had been concentrated before use. The data are representative of 3 independent experiments. (C) Adhesion of 100 000 Lin− Actin- DsRed+ cells to vehicle- or warfarin-treated macrophages in the presence or absence of periostin (P = .03; ANOVA, Tukey Test). Cells were allowed to adhere for 6 hours before being counted. The data were generated in 2 independent experiments. (D) Number of Lin− pAKT+ cells per femur of control mice or mice treated with warfarin 14 days after initiation of treatment (P = .037, t test, n = 3). (E-F) Representative confocal images (scale bar depicts 10 μm) (E) and quantification of nuclear pAKT+ (P = .028, t-test, n = 10) (F) in LKS cells stained with pAKT (green) and DAPI (blue) after 15 hours of culture in CM harvested from vehicle or warfarin-treated macrophages. The experiment was performed twice. (G) Total number of hematopoietic cells 4 days after plating of 1 × 106 Lin− cells from vehicle- or warfarin-treated mice on vehicle- or warfarin-treated macrophages in the presence of the AKT inhibitor MK-2206 (5 μM; P = .002 for the impairment of the increase of the cell number by MK-2206; ANOVA, Tukey Test, n = 4-5). The data were generated in 3 independent experiments. (H-I) Representative confocal images (scale bar depicts 10 μm) (H) and quantification of nuclear pAKT+ (P = .02, t-test, n = 10) (I) in LKS cells stained with pAKT (green) and DAPI (blue) after 15 hours of culture in CM harvested from vehicle-treated macrophages in the presence or absence of 0.05 μM cilengitide. The experiment was performed twice. (J) Relative abundance of annexin+ DAPI+ LKS cells in control mice treated with vehicle (red circles) or cilengitide (blue squares) or mice treated with warfarin (green triangles) or warfarin and cilengitide (purple inverted triangles) relative to vehicle-treated control mice (P = .0003 for control plus vehicle vs control plus cilengitide; ANOVA, Tukey Test, n = 5).

Warfarin impairs hematopoiesis via the periostin/ integrin β3/ AKT axis. (A) Number of LKS cells expressing integrin β3 in control mice (red circles), control mice treated with periostin (blue squares), or mice treated with warfarin (green triangles) or warfarin plus periostin (purple inverted triangles) for 14 days (P = .01 for control vs warfarin and P = .003 for warfarin vs warfarin plus periostin; ANOVA, Tukey Test, n = 9-10). The data were generated in 2 independent experiments. (B) Protein lysates of 293T cells, transfected with an integrin β3-overexpressing construct and grown in CM from vehicle- or warfarin-treated macrophages for 7 days, were subjected to coimmunoprecipitation with an anti-integrin β3 antibody, and coimmunoprecipitation of periostin was assessed by immunoblotting, as indicated. Anti-IgG-isotype antibody was used as control in the co-immunoprecipitation. The input (left) shows the presence of integrin β3 (92 kDa) and periostin (90 kDa), and GAPDH (38 kDa) was used to control for equal loading. The CM had been concentrated before use. The data are representative of 3 independent experiments. (C) Adhesion of 100 000 Lin− Actin- DsRed+ cells to vehicle- or warfarin-treated macrophages in the presence or absence of periostin (P = .03; ANOVA, Tukey Test). Cells were allowed to adhere for 6 hours before being counted. The data were generated in 2 independent experiments. (D) Number of Lin− pAKT+ cells per femur of control mice or mice treated with warfarin 14 days after initiation of treatment (P = .037, t test, n = 3). (E-F) Representative confocal images (scale bar depicts 10 μm) (E) and quantification of nuclear pAKT+ (P = .028, t-test, n = 10) (F) in LKS cells stained with pAKT (green) and DAPI (blue) after 15 hours of culture in CM harvested from vehicle or warfarin-treated macrophages. The experiment was performed twice. (G) Total number of hematopoietic cells 4 days after plating of 1 × 106 Lin− cells from vehicle- or warfarin-treated mice on vehicle- or warfarin-treated macrophages in the presence of the AKT inhibitor MK-2206 (5 μM; P = .002 for the impairment of the increase of the cell number by MK-2206; ANOVA, Tukey Test, n = 4-5). The data were generated in 3 independent experiments. (H-I) Representative confocal images (scale bar depicts 10 μm) (H) and quantification of nuclear pAKT+ (P = .02, t-test, n = 10) (I) in LKS cells stained with pAKT (green) and DAPI (blue) after 15 hours of culture in CM harvested from vehicle-treated macrophages in the presence or absence of 0.05 μM cilengitide. The experiment was performed twice. (J) Relative abundance of annexin+ DAPI+ LKS cells in control mice treated with vehicle (red circles) or cilengitide (blue squares) or mice treated with warfarin (green triangles) or warfarin and cilengitide (purple inverted triangles) relative to vehicle-treated control mice (P = .0003 for control plus vehicle vs control plus cilengitide; ANOVA, Tukey Test, n = 5).

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal