RGT, a synthetic peptide corresponding to the integrin beta 3 cytoplasmic C-terminal sequence, selectively inhibits outside-in signaling in human platelets by disrupting the interaction of integrin alpha IIb beta 3 with Src kinase

Abstract

Mutational analysis has established that the cytoplasmic tail of the integrin beta 3 subunit binds c-Src (termed as Src in this study) and is critical for bidirectional integrin signaling. Here we show in washed human platelets that a cell-permeable, myristoylated RGT peptide (myr-RGT) corresponding to the integrin beta 3 C-terminal sequence dose-dependently inhibited stable platelet adhesion and spreading on immobilized fibrinogen, and fibrin clot retraction as well. Myr-RGT also inhibited the aggregation-dependent platelet secretion and secretion-dependent second wave of platelet aggregation induced by adenosine diphosphate, ristocetin, or thrombin. Thus, myr-RGT inhibited integrin outside-in signaling. In contrast, myr-RGT had no inhibitory effect on adenosine diphosphate-induced soluble fibrinogen binding to platelets that is dependent on integrin inside-out signaling. Furthermore, the RGT peptide induced dissociation of Src from integrin beta 3 and dose-dependently inhibited the purified recombinant beta 3 cytoplasmic domain binding to Src-SH3. In addition, phosphorylation of the beta 3 cytoplasmic tyrosines, Y(747) and Y(759), was inhibited by myr-RGT. These data indicate an important role for beta 3-Src interaction in outside-in signaling. Thus, in intact human platelets, disruption of the association of Src with beta 3 and selective blockade of integrin alpha IIb beta 3 outside-in signaling by myr-RGT suggest a potential new antithrombotic strategy.

Figure 1

Intraplatelet localization of the membrane-permeable peptides. Platelets incubated with FITC-conjugated peptides (250 μM) for 30 minutes and analyzed by flow cytometry and fluorescence microscopy. (A) Fluorescence histograms of platelets treated with FITC-conjugated myristoylated RGT peptide (FITC-myr-RGT, closed histogram) or with FITC-conjugated RGT peptide (FITC-RGT, open histogram) were analyzed by flow cytometry. (B) Platelets were treated with FITC-myr-RGT peptide or FITC-RGT peptide and allowed to spread on immobilized fibrinogen for 60 minutes. The same microscopic fields were analyzed by differential interference contrast (DIC) microscopy as well as confocal fluorescence microscopy (fluorescence). Figure shows representative images made by a Zeiss LSM510 confocal microscope with a 63× plan-apochromat DIC oil-immersion objective with Pascal software. (C) Z-Stack scanning was performed on FITC-myr-RGT-treated platelets with intervals of 1.2 μm (from 1 to 4). The fluorescence density profiles are shown below each picture.

Figure 2

Quantitative analysis of the effect of myristoylated RGT peptide on platelet stable adhesion and spreading on immobilized fibrinogen. (A) Platelets were added to microtiter wells precoated with fibrinogen and allowed to adhere for 60 minutes at 37°C. The phosphatase activity in supernatants (open columns) or in adherent platelets (closed columns) was quantified by a PNPP assay. Data of 3 experiments (mean ± SD) were presented as the ratio of the phosphatase activity of platelet samples over blank. Peptide concentrations: *62.5 μM; **125 μM; ***250 μM. (Inset) The phosphatase activity of myr-RGT-treated platelets adherent to immobilized fibrinogen after removal of the peptide from the buffer. (B) Microphotographs of platelets adherent on fibrinogen and treated with DMSO (B1), myristic acid and RGT peptide at a concentration of 250 μM (B2), scrambled myr-GRT at 250 μM (B3), myr-RGT at 62.5 μM (B4), 125 μM (B5), and 250 μM (B6). DIC indicates differential interference contrast microscopy; IF: immunofluorescence assay with anti-integrin β3 antibody. (C) Platelets were incubated in suspension with different treatments as indicated for 30 minutes at 37°C. The phosphatase activity in supernatants (□) or in remaining platelet suspensions () was quantified by a PNPP assay. Data are arranged as in panel A.

Figure 3

Effect of myr-RGT on fibrin clot retraction and platelet aggregation. (A) Washed platelets were resuspended in HEPES buffer and incubated with different peptides or their vehicles as indicated. Then 2 mg/mL of human fibrinogen was added and fibrin clot formation was initiated by adding 1 U/mL of thrombin. Clot retraction was monitored over time, and photographs of the clots were taken at different time points (bottom panel). Peptide concentrations: *62.5 μM; **125 μM; ***250 μM. The histograms of the clot size were generated from the photographs by calculating the ratio of the surface area of the retracted clot versus that of the initial clot. (B) Aggregation of nontreated or peptide-treated platelets was induced in an aggregometer at 37°C under constant stirring (1000 rpm) by ADP (2 μM), ristocetin (1.25 mg/mL) in PRP, or by thrombin (0.1 U/mL) in washed platelets preincubated with (A) DMSO, (B) myristic acid and RGT peptide at a concentration of 250 μM, (C) scrambled myr-GRT at 250 μM, myr-RGT at concentrations of (D) 62.5 μM, (E) 125 μM, and (F) 250 μM, respectively. (Inset) Platelets in plasma treated with 250 μM of (F) myr-RGT, (C) scrambled myr-GRT, or (A) vehicle DMSO were stimulated by TRAP (10 μM) to aggregate.

Figure 4

Effect of myr-RGT on soluble fibrinogen binding to platelets. Platelets were preincubated with different peptides or their vehicle, and binding of Alexa Fluor 488-conjugated fibrinogen (100 μg/mL) to platelets was measured by flow cytometry after the addition of 20 μM of ADP. (A) Representative histograms. Fibrinogen bound to platelets treated with DMSO, myristic acid, and RGT at 250 μM (M + RGT), scrambled myr-GRT at 250 μM (myr-GRT), myr-RGT at 250 μM (myr-RGT), or with RGDS peptide (1 mM). The background fibrinogen binding was assessed using platelets treated without ADP (Control). (B) Statistical data were derived from quantitative results (means and SD) calculated from the ratios of mean fluorescence intensity (samples/control) of 3 separate experiments.

Figure 5

Effect of myr-RGT on platelet CD62P expression in the presence of agonists. The expression of CD62P on nontreated or peptide-treated platelets stimulated with agonists was analyzed by flow cytometry using an FITC-labeled monoclonal anti-CD62P antibody. (A) Data shown in the left pattern (mean ± SD) were derived from the ratio of the geometric mean fluorescence intensity measured for anti-CD62P antibody binding to thrombin-treated platelets, preincubated for 30 minutes with DMSO vehicle, nonmyristoylated RGT peptide (250 μM) plus myristic acid (M + RGT), scrambled myr-GRT (250 μM), or myr-RGT (250 μM), versus resting platelets (without thrombin treatment) and obtained from 3 separate experiments. Data shown in the right pattern are a representative figure of CD62P expression in the presence of thrombin on platelets preincubated with DMSO (fine line), myr-RGT (normal line), or on resting platelets (in the absence of thrombin; thick line). (B) Expression of CD62P on peptide-treated platelets stimulated with ADP under stirring conditions.

Figure 6

Effect of myr-RGT on thrombin-induced phosphorylation of integrin β3 cytoplasmic Y⁷⁴⁷ and Y⁷⁵⁹. Washed platelets were preincubated for 30 minutes with 250 μM of the different peptides as indicated. Thrombin (0.05 U/mL) was added to induce platelet aggregation with stirring at 1000 rpm for 1 minute. The platelets were then lysed in SDS-PAGE sample buffer and analyzed by Western blotting using monoclonal antibodies directed against the integrin β3 subunit extracellular domain (SZ21), β-actin, and polyclonal antibodies specific for the β3 integrin cytoplasmic sequences containing phosphorylated Y⁷⁴⁷ or Y⁷⁵⁹ residue, respectively.

Figure 7

Effect of myr-RGT on the interaction of integrin β3 cytoplasmic domain with Src or talin. (A) Platelets preincubated with 250 μM of myr-RGT or scrambled myr-GRT were lysed with lysis buffer, and the lysates of untreated or peptide-treated platelets were analyzed with an immunoprecipitation procedure as follows. The lysates were incubated with SZ21 antibody or nonspecific mouse IgG. After washing, the immune complexes were subjected to SDS-PAGE and probed by Western blotting using monoclonal antibodies SZ22 or 327 against integrin αIIb or c-Src, respectively (lanes 1-5). In another set of experiments, immunoprecipitation was performed with monoclonal antibody 327 and blotted with SZ21 or 327 (lanes 6-10). Representative results of 3 experiments are shown. (B) Glutathione-Sepharose 4B beads coated with GST-wild-type integrin β3 cytoplasmic tail fusion protein were incubated overnight with purified His-Src-SH3 in the presence of peptides as indicated. After wash, protein complexes were subjected to Western blot analysis with anti-His or anti-GST antibodies. Peptide concentrations: *62.5 μM; **125 μM; ***250 μM; ****500 μM. (C) Increasing concentrations of purified GST-Src-SH3 or GST protein were added to the microtiter wells coated with RGT or GRT peptide (20 μg/mL). Binding of the purified proteins to the peptides was detected by incubation with mouse anti-GST antibody, followed by horseradish peroxidase-conjugated antimouse Ig antibody. Specific binding was normalized by subtracting the OD (optical density) values of the blank wells from that of the sample wells. Results were presented as percentage of the maximal binding. Data were organized as binding of GST-Src-SH3 to RGT peptide (●), GST-Src-SH3 to GRT peptide (■), GST protein to RGT peptide (△), and GST protein to GRT peptide (□). (D) Glutathione-Sepharose 4B beads coated with GST-integrin β3 cytoplasmic tail fusion proteins were incubated overnight with platelet lysates in the presence of peptides as indicated at 4°C before being lysed by SDS sample buffer. Talin was detected with the monoclonal antibody 8d4. Anti-GST antibody binding was used to verify the loading of the β3 cytoplasmic tail fusion proteins. The increased electrophoretic mobility of GST-β3-741 documents the 21 residue truncation of this fusion protein.