Fig 1.
Specific binding of IGF2 to ανβ3.
a, Binding of soluble αvβ3 to IGF2 in ELISA-type binding assays. Wells of 96 well microtiter plates were coated with increasing concentrations of IGF2. Soluble recombinant αvβ3 (5 μg/ml) was incubated with immobilized IGF2 in Tyrode-HEPES buffer containing 1 mM Mg2+. The αvβ3 bound to IGF2 was measured using anti-integrin β3 mAb. The data are shown as means +/- SEM of triplicate experiments. b, CHO cells expressing recombinant αvβ3 (β3-CHO cells) bound to IGF2 while CHO cells and CHO cells expressing αvβ1 (β1-CHO) did not. Wells 96 well microtiter plates were coated with IGF2 at increasing concentrations. The wells were incubated with β3-CHO, β1-CHO and CHO cells in serum free DMEM buffer (105 cells/well). The bound cells were measured. The data are shown as means +/- SEM of triplicate experiments. c, Antibody against αvβ3 (7E3) and cyclic RGDfV blocked the adhesion of β3-CHO cells to IGF2. Wells 96 well microtiter plates were coated with IGF2 at 50 μg/ml. β3-CHO cells (105 cells/well) were incubated with the immobilized IGF2 plus 7E3 or cyclic RGDfV in Tyrode-HEPES buffer containing 1 mM Mg2+, the bound cells were measured. The data are shown as means +/- SEM of triplicate experiments.
Fig 2.
Integrin binding is required for IGF2-induced cell proliferation.
a, Cell proliferation was enhanced by IGF2 in β3-CHO cells better than CHO cells. Cells (2 x 104 cells/well) were incubated for 48 hrs in polyHEMA-coated plates. Cell proliferation was measured by MTS assays. The data are shown as means +/- SEM of triplicate experiments. b, IGF2-induced cell proliferation was reduced by cyclic RGDfV. Cells (2 x 104 cells/well) were incubated with IGF2 (100 ng/ml) in combination with cyclic RGDfV for 48 hrs in polyHEMA-coated plates. Cell proliferation was measured by MTS assays. The data are shown as means +/- SEM of triplicate experiments. c. IGF2 induces phosphorylation of IGF1R in a dose and time-dependent manner. Cells were serum-starved in serum free DMEM for 4 hrs, and treated with WT IGF2 (10 or 100 ng/ml) for 10 min or 1 hr in a polyHEMA coated plates. Cell lysates were analyzed by western blotting. Density of the bands were quantified using ImageJ software and p-IGF1R/t-IGF1R was calculated. d, IGF2 induces signals in β3-CHO cells. Cells were serum-starved in serum free DMEM for 4 hrs, and treated with WT IGF2 for 10 minutes in a polyHEMA coated plates. Cell lysates were analyzed by western blotting. Density of the bands were quantified using ImageJ software and p-IGF1R/t-IGF1R or p-AKT/t-AKT was calculated.
Fig 3.
The C-domain is critically involved in integrin binding and the integrin-binding defective IGF2 mutants are defective in signaling functions and antagonistic (dominant-negative).
a. Alignment of IGF1, IGF2 and insulin. Previous studies showed that the critical amino acid residues for integrin binding in IGF1 are present in the C-domain (Arg36 and Arg37, in blue) (5). Thus several Arg residues at positions 24, 30, 34, 37, 38 and 40 (in red) of IGF2 in and around the C-domain were selected for mutagenesis. b and c. Mutations in the predicted integrin-binding interface of IGF2 suppress integrin binding of IGF2. Wells of 96-well microtiter plate were coated with WT and mutant IGF2. Then β3-CHO cells and CHO cells were incubated in Tyrode-HEPES buffer containing 1 mM Mg2+. The bound cells were measured. The data are shown as means +/- SEM of triplicate experiments. d, R24E/R37E/R38E, R34E/R37E/R38E, and R24E/R34E/R37E/R38E are defective in inducing phosphorylation of the IGF1R, AKT and ERK1/2 in β3-CHO cells. Cells were treated with WT IGF2, R24E/R37E/R38E, R34E/R37E/R38E, or R24E/R34E/R37E/R38E at 100 ng/ml for 10 min. Cell lysates were analyzed by western blotting. e, R24E/R37E/R38E, R34E/R37E/R38E, and R24E/R34E/R37E/R38E are defective in inducing proliferation of β3-CHO in MTS assays. β3-CHO cells were incubated with WT IGF2, R24E/R37E/R38E, R34E/R37E/R38E, or R24E/R34E/R37E/R38E at increasing concentrations for 48 hrs in 96-well plate coated with polyHEMA. The data are shown as means +/- SEM (n = 6). f, R24E/R37E/R38E, R34E/R37E/R38E, and R24E/R34E/R37E/R38E inhibited cell proliferation induced by WT IGF2 (25 ng/ml) in MTS assays. The data are shown as means +/- SEM (n = 6).
Fig 4.
IGF2 binds to α6β4 and induces proliferation of α6β4-CHO cells while integrin-binding defective IGF2 mutants are defective in these functions.
a, α6β4-CHO cells bind to WT IGF2 in adhesion assay. Wells of 96-well microtiter plates were coated with WT IGF2. α6β4-CHO cells were incubated in Tyrode-HEPES buffer containing 1 mM Mg2+, and bound cells were measured. The data are shown as means +/- SEM of triplicate experiments. b, WT IGF2 enhanced proliferation of α6β4-cells but not β1-CHO cells. α6β4-CHO and β1-CHO cells were incubated for 48 hrs with increasing concentrations of WT IGF2 in polyHEMA-coated plates. The data are shown as means +/- SEM (n = 6). c, Integrin-binding defective IGF2 mutants were defective in inducing proliferation of α6β4-CHO cells in MTS assays while WT IGF2 induced it. α6β4-CHO cells (2 x 104 cells/well) were incubated for 48 hrs with WT or mutant IGF2 (250 ng/ml) in polyHEMA-coated plates. The data are shown as means +/- SEM (n = 6). d, Binding of IGF2 mutants to the immobilized IGF1R ectodomain. Wells of 96-well microtiter plate were coated with recombinant human soluble IGF1R at 1 μg/ml in PBS for 1 h at 37°C, and the remaining protein-binding sites were block by incubating with 1 mg/ml BSA for 1 h at room temperature. WT and mutant IGF2 (2.5 μg/50 μl in PBS) were added to the wells and incubated in PBS/0.05% Tween 20 at room temperature for 1 hr. After washing with PBS/0.05% Tween 20, wells were incubated with anti-5His antibody conjugated with HRP, then peroxidase substrate. The data are shown as means +/- SEM (n = 3)