Talin-1 is the principal platelet Rap1 effector of integrin activation

Abstract

Ras-related protein 1 (Rap1) is a major convergence point of the platelet-signaling pathways that result in talin-1 binding to the integrin β cytoplasmic domain and consequent integrin activation, platelet aggregation, and effective hemostasis. The nature of the connection between Rap1 and talin-1 in integrin activation is an important remaining gap in our understanding of this process. Previous work identified a low-affinity Rap1-binding site in the talin-1 F0 domain that makes a small contribution to integrin activation in platelets. We recently identified an additional Rap1-binding site in the talin-1 F1 domain that makes a greater contribution than F0 in model systems. Here we generated mice bearing point mutations, which block Rap1 binding without affecting talin-1 expression, in either the talin-1 F1 domain (R118E) alone, which were viable, or in both the F0 and F1 domains (R35E,R118E), which were embryonic lethal. Loss of the Rap1-talin-1 F1 interaction in platelets markedly decreases talin-1-mediated activation of platelet β1- and β3-integrins. Integrin activation and platelet aggregation in mice whose platelets express only talin-1(R35E, R118E) are even more impaired, resembling the defect seen in platelets lacking both Rap1a and Rap1b. Although Rap1 is important in thrombopoiesis, platelet secretion, and surface exposure of phosphatidylserine, loss of the Rap1-talin-1 interaction in talin-1(R35E, R118E) platelets had little effect on these processes. These findings show that talin-1 is the principal direct effector of Rap1 GTPases that regulates platelet integrin activation in hemostasis.

Graphical abstract

Figure 1.

Rap1 binding to talin-1 F1 domain contributes to integrin activation in platelets. (A) Generation of mice harboring Tln1 R118E mutation. Two guide RNAs were used to target Tln1 exon 3. A silent mutation corresponding to the PAM sequence of gRNA2 was inserted into the donor single-stranded DNA oligo to prevent re-editing. (B) Sequencing chromatogram of mutated region of Tln1(R118E) gene. (C) Expression of talin-1 mutant in Tln1-R118E platelets was assayed by using western blotting. Results are representative of 3 independent experiments, n = 3 mice each time. (D) Surface expression of αIIb, β3, α2, α5, and β1 integrins in Tln1-R118E platelets was measured by using flow cytometry. Bar graph represents mean fluorescence intensity (MFI) ± standard error of the mean (n = 6 mice). Two-tailed Student t test. (E-G) Impaired integrin activation in Tln1-R118E platelets. Flow cytometry assay to measure binding of GPIX-labeled platelets in whole blood to the JonA/PE antibody (E-F) or the Alexa Fluor 488–coupled 9EG7 antibody (G) in response to agonist stimulation. Bar graphs represent MFI ± standard error of the mean (n = 6 mice, representative of ≥3 independent experiments). Two-way analysis of variance with Tukey posttest. *P < .05; ***P < .001. ns, not significant.

Figure 2.

Blockade of Rap1 binding to both F0 and F1 domains in talin-1 prevents integrin activation in platelets in Tln1-mR35E,R118E mice. (A) Expression of talin-1 mutant in Tln1-mR35E,R118E platelets was assayed by using western blotting. Results are representative of 3 independent experiments, n = 3 mice each time. (B) Surface expression of αIIb, β3, α2, α5, and β1 integrins in Tln1-mR35E,R118E platelets was measured by using flow cytometry. Bar graph represents mean fluorescence intensity (MFI) ± standard error of the mean (n = 6 mice). Two-tailed Student t test. (C-D) Impaired integrin activation in Tln1-mR35E,R118E. Flow cytometry assay to measure binding of GPIX-labeled platelets in whole blood to JonA/PE antibody (C) or Alexa Fluor 488–coupled 9EG7 antibody (D) in response to agonist stimulation. Bar graphs represent MFI ± standard error of the mean (n = 6 mice, representative of ≥3 independent experiments). Two-way analysis of variance with Tukey posttest. (E) Representative aggregation responses of Tln1-mR35E,R118E platelets stimulated with various concentrations of agonists (indicated by arrows). (F) Intravital microscopy studies to monitor hemostatic plug formation after laser injury to the saphenous vein in Tln1-mR118E and Tln1-mR35E,R118E mice. The experiment was terminated at the end of 10 minutes to avoid excessive loss of blood. Individual bleeding times with median were plotted for 3 experimental groups. Data depicted are determinations in 3 control mice (17 injury sites), 4 Tln1-mR118E mice (42 injury sites), and 5 Tln1-mR35E,R118E mice (39 injury sites). Statistical significance was assayed by a one-way analysis of variance with Tukey posttest. ***P < .001. ns, not significant.

Figure 3.

Graduated αIIbβ3 activation response in Tln1-mR35E, Tln1-mR118E, Tln1-mR35E,R118E, and Tln1-mKO platelets. (A) Flow cytometry assay to measure binding of GPIX-labeled platelets in whole blood to Jon/A-PE antibody in response to PAR4-AP stimulation. Bar graph represents mean fluorescence intensity (MFI) ± standard error of the mean (n = 6 mice, representative of ≥3 independent experiments). Two-way analysis of variance with Tukey posttest. (B-C) Real-time αIIbβ3 activation assay. JonA/PE binding to washed platelets was recorded continuously for 9 minutes by using flow cytometry in response to PAR4-AP (B) or convulxin (C) stimulation. Arrows indicate addition of agonists. (D) Representative aggregation responses of Tln1-mR118E and Tln1-mR35E,R118E platelets stimulated with various concentrations of agonists. Arrows indicate addition of agonists. *P < .05; ***P < .001. ns, not significant.

Figure 4.

Tln1-mR35E,R118E platelets exhibit impaired αIIbβ3  activation to a similar extent as Rap1a/b-mKO platelets. (A-B) Flow cytometry assay to measure binding of GPIX-labeled platelets in whole blood to JonA/PE antibody in response to PAR4-AP (A) or convulxin (B) stimulation. Bar graphs represent mean fluorescence intensity (MFI) ± standard error of the mean (n = 6 mice, representative of ≥3 independent experiments). (C) Representative aggregation responses of Tln1-mR35E,R118E and Rap1a/b-mKO platelets stimulated with various concentrations of agonists. Curves corresponding to control and Tln1-mR35E,R118E platelets stimulated with PAR4-AP were from the same experiment as those depicted in Figure 3D. Arrows indicate addition of agonists. *P < .05;**P < .01; ***P < .001. ns, not significant.

Figure 5.

Tln1-mR35E,R118E mice exhibit milder defects in hemostasis compared with Rap1a/b-mKO mice, along with preserved capacity of platelets to secrete α-granules and expose surface PS. (A-B) Intravital microscopy studies to monitor hemostatic plug formation after laser injury to the saphenous vein in Tln1-mR35E,R118E and Rap1a/b-mKO mice. Before laser injury, animals were injected with Alexa Fluor 488–conjugated antibodies to GPIX to label circulating platelets and Alexa Fluor 647–conjugated antibodies to fibrin. The experiment was terminated at the end of 10 minutes to avoid excessive loss of blood. (A) Representative images taken 30, 60, and 300 seconds after laser injury. Scale bar, 50 μm. (B) Percentage of bleeding injury sites was plotted over time for each experimental group. Seventeen injury sites in 3 control mice, 39 injury sites in 5 Tln1-mR35E,R118E mice, and 8 injury sites in 2 Rap1a/b-mKO mice were tested. Statistical significance was assayed by using the Gehan-Breslow-Wilcoxon test with Bonferroni correction. (C-D) Comparisons of the 3 pairs of groups were all statistically significant using a family-wise significance level of 5%. Flow cytometry analysis of P-selectin (CD62P) surface expression onto GPIX-labeled platelets in whole blood in response to PAR4-AP (C) or convulxin (D) stimulation. Bar graphs represent Δ mean fluorescence intensity (MFI) ± standard error of the mean (n = 6 mice). Two-way analysis of variance with Tukey post-test. (E) Determination according to flow cytometry of PS exposure onto GPIX-labeled platelets in whole blood in response to dual stimulation with 100 ng/mL of convulxin and 500 μM of PAR4-AP. Bar graph represents the percent mean ± standard error of the mean (n = 6 mice). Two-way analysis of variance with Tukey posttest. **P < .01; ***P < .001. ns, not significant.

Figure 6.

Model for the main final steps in talin-1–dependent platelet integrin activation. (A) Talin-1 is autoinhibited where the F3 integrin–binding site (red disc) is obscured by its interaction with the R9 helical bundle, and the PI(4,5)P₂ binding site in F2 and F3 are masked by the R12 helical bundle. (B) The Rap1-binding sites in F0 and F1 were not part of the autoinhibited structure and therefore available to bind Rap1-GTP. (C) At the membrane, PI(4,5)P₂ can disrupt the F3–R9 and F2–R12 interactions to unveil the talin-1 F3 integrin–binding site (red disc). (D) The binding of talin-1 F3 domain to the integrin β tail induces activation of αIIbβ3 and β1 integrins.