Integrin β3 directly inhibits the Gα13-p115RhoGEF interaction to regulate G protein signaling and platelet exocytosis

Abstract

The integrins and G protein-coupled receptors are both fundamental in cell biology. The cross talk between these two, however, is unclear. Here we show that β₃ integrins negatively regulate G protein-coupled signaling by directly inhibiting the Gα₁₃-p115RhoGEF interaction. Furthermore, whereas β₃ deficiency or integrin antagonists inhibit integrin-dependent platelet aggregation and exocytosis (granule secretion), they enhance G protein-coupled RhoA activation and integrin-independent secretion. In contrast, a β₃-derived Gα₁₃-binding peptide or Gα₁₃ knockout inhibits G protein-coupled RhoA activation and both integrin-independent and dependent platelet secretion without affecting primary platelet aggregation. In a mouse model of myocardial ischemia/reperfusion injury in vivo, the β₃-derived Gα₁₃-binding peptide inhibits platelet secretion of granule constituents, which exacerbates inflammation and ischemia/reperfusion injury. These data establish crucial integrin-G protein crosstalk, providing a rationale for therapeutic approaches that inhibit exocytosis in platelets and possibly other cells without adverse effects associated with loss of cell adhesion.

Fig. 1. Integrin β₃ cytoplasmic domain inhibits Gα₁₃-p115RhoGEF interaction in purified protein binding assays and in platelets.

a Binding of purified GST-tagged RGS domain of p115RhoGEF (GST-p115RGS) to purified Histidine-tagged Gα₁₃ (His-Gα₁₃) coated on microtiter plates is inhibited by increasing concentrations of purified GST-β₃ cytoplasmic domain fusion protein (β₃CD) compared with control GST, 3 independent experiments. b, c Inhibition of the binding of purified GST-p115RGS to microtiter well-coated His-Gα₁₃ (b) and the binding of His-Gα₁₃ to microtiter well-coated GST-p115RGS (c) by increasing concentration of mP6 peptide or a scrambled peptide with the same amino acid composition (mP6 Scr), 3 independent experiments. d, e Co-immunoprecipitation of p115RhoGEF and Gα₁₃ in WT (C57BL/6J) and β₃^-/- mouse platelets stimulated with thrombin (0.035 U/ mL) for increasing lengths of time. d A representative Western blot; e, quantification of data from 4 independent experiments. f, g Co-immunoprecipitation of p115RhoGEF and Gα₁₃ in resting (0) and 0.03 U/mL thrombin-stimulated WT mouse platelets pre-treated with RGDS or control RGES peptide (2 mM). f A representative western blot; g quantification of data from four independent experiments. h, i Co-immunoprecipitation of p115RhoGEF and Gα₁₃ in wild-type mouse platelets pre-treated with control peptide (myr-FAAAKL) or mP6 peptide (200 μM, DMSO), stimulated with thrombin (0.03 U/ mL) for 10 seconds. h A representative Western blot; i Quantification of co-immunoprecipitation of Gα₁₃. Data are from three independent experiments. Platelets used in (d–g) were pre-treated with 500 μM aspirin to exclude the differential influence of secondary TXA₂ production on Gα₁₃ pathway of WT and β₃^-/- platelets. All data are shown as mean ± SEM. Statistical analysis was done using Student’s t-test, two tailed. Source data are provided as a Source Data file.

Fig. 2. The role of β₃ integrins in inhibiting thrombin-induced RhoA activation.

a, b Rhotekin-RBD bead pull down assay was used to analyze RhoA activation in WT or β₃^-/- mouse platelets stimulated with thrombin (0.03 U/mL) at various time points, 4 independent experiments. c, d RhoA pull down assay with RGDS or RGES-pre-treated (2 mM, 3 min, 22 °C) mouse platelets stimulated with thrombin (0.03 U/mL) for the indicated time points, four independent experiments. e, f RhoA pull down assay with 0.03 U/ mL thrombin-stimulated mouse platelets pre-treated with control peptide or mP6 peptide (100 μM, DMSO) for 3 min at 22 °C, three independent experiments. a, c, e Representative western blots; b, d, f quantification of western blot images as analyzed using ImageJ. All data are shown as mean ± SEM. Statistical significance was determined using Student’s t test, (f) paired Student’s t test, two tailed. Source data are provided as a Source Data file.

Fig. 3. β₃ integrins negatively regulate thrombin-induced platelet granule secretion.

a Aggregation and ATP secretion traces of washed WT and β₃^-/- platelets stimulated with thrombin. b Quantification of ATP secretion for a, three independent experiments. c Aggregation and ATP secretion traces of washed human platelets pre-incubated with vehicle or 10 μg/mL Eptifibatide for 3 min at 22 °C and stimulated with thrombin. d Quantification of ATP secretion for c, three independent experiments. e Flow cytometry analysis of P-selectin expression on mouse platelets pre-treated with control vehicle or Eptifibatide (20 μg/mL) for 3 min at 22 °C and stimulated with 0.03 U/mL thrombin at 37 °C with stirring at 1000 rpm in an aggregometer. Three independent experiments. All data are shown as mean ± SEM. Statistical significance was determined using Student’s t test, two tailed. Source data are provided as a Source Data file.

Fig. 4. β₃ integrins negatively regulate integrin-independent platelet secretion but stimulate integrin-dependent secretion.

a Aggregation and ATP secretion traces of washed WT and β₃^-/- platelets induced by U46619 at 37 °C. Note the two waves of ATP secretion in WT platelets and a higher single (first) secretion wave in β₃^-/- platelets. b, c Quantification of the first (b) and second (c) waves of secretion as shown in a, three independent experiments. d Aggregation and ATP secretion traces of washed human platelets pre-incubated with vehicle or 10 μg/mL Eptifibatide for 3 min at 22 °C followed by stimulation with U46619 at 37 °C. e Quantification of secretion data from (d), 3 independent experiments. f U46619-induced ATP secretion traces of β₃^-/- platelets pre-incubated with control vehicle or 10 µM Y27632 for 3 min at 22 °C. All data are shown as mean ± SEM. Statistical significance was determined using Student’s t-test, two tailed. Source data are provided as a Source Data file.

Fig. 5. Gα₁₃ stimulates both integrin-independent and integrin-dependent granule secretion.

a Washed WT(Gα₁₃^fl/fl) and Gα₁₃^-/- mouse platelets were solubilized and immunoblotted with antibodies specifically recognizing mouse Gα₁₃, and integrin β₃. b Aggregation and ATP secretion traces of washed WT and Gα₁₃^-/- platelets stimulated with thrombin. c Quantification of secretion as shown in b, four independent experiments. d Aggregation and ATP secretion traces of WT and Gα₁₃^-/- platelets stimulated with 2 μmol/L U46619. e Quantification of ATP secretion as shown in d, six independent experiments. Aggregation assays were performed without exogenously added fibrinogen. All data are shown as mean ± SEM. Data were analyzed using Student’s t-test, (e) paired Student’s t-test, two tailed. Source data are provided as a Source Data file.

Fig. 6. Gα₁₃ antagonistic peptide mP6, inhibits both integrin-independent and integrin-dependent granule secretion.

a, c ATP secretion traces of washed mouse platelets pre-incubated with 20 μM control FAAAKL or mP6 high loading peptide nanoparticles (HLPNs) and then stimulated with 0.03 U/mL thrombin (a) or 3 μmol/L U46619 (c). b, d Quantification of ATP secretion as shown in a, n = 3 or (c), n = 4, independent experiments. e, f Effects of control FAAAKL or mP6 HLPNs (20 μM) on serotonin secretion (e), or on P-selectin expression (f), in mouse platelets stimulated with 0.03U/ mL thrombin in an aggregometer stirring at 37 °C and 1000 rpm. n = 3, independent experiments. The assays were performed using washed platelets without adding fibrinogen. g Plasma serotonin levels in sham, saline, and M3mP6 HLPN treatment groups 24 h after myocardial infarction with reperfusion. Sham group, n = 12; Saline group, n = 14; M3mP6 group, n = 13, independent animals. Data in b, d–f are shown as mean ± SEM and in g are shown as median ± SEM. Statistical significance was determined using Student’s t-test, two tailed. h A schematic of crosstalk between the (G protein-coupled receptor) GPCR-coupled Gα₁₃ pathway and integrin α_IIbβ₃ in regulating RhoA activation and platelet granule secretion. Left panel: GPCR agonists such as thrombin induce Gα₁₃-p115RhoGEF-dependent activation of RhoA, which mediates integrin-independent granule secretion. Center panel: following integrin ligation, Gα₁₃ interacts with the β₃ cytoplasmic domain ExE motif, which inhibits Gα₁₃ binding to p115RhoGEF, resulting in inhibition of p115RhoGEF-mediated RhoA activation and integrin-independent granule secretion. Gα₁₃ binding to β₃, however, also mediates integrin outside-in signaling leading to Src activation and integrin-dependent granule secretion. Right Panel: The β₃-derived mP6 peptide binds to Gα₁₃ to inhibit Gα₁₃ interacting with both p115-RhoGEF and β₃. Thus, mP6 potently inhibits both integrin-independent and integrin-dependent secretion without affecting integrin ligation and primary platelet adhesion/aggregation necessary for normal hemostasis. Source data are provided as a Source Data file.