Lyn and PECAM-1 function as interdependent inhibitors of platelet aggregation

Abstract

Inhibition of platelet responsiveness is important to control pathologic thrombus formation. Platelet-endothelial cell adhesion molecule-1 (PECAM-1) and the Src family kinase Lyn inhibit platelet activation by the glycoprotein VI (GPVI) collagen receptor; however, it is not known whether PECAM-1 and Lyn function in the same or different inhibitory pathways. In these studies, we found that, relative to wild-type platelets, platelets derived from PECAM-1-deficient, Lyn-deficient, or PECAM-1/Lyn double-deficient mice were equally hyperresponsive to stimulation with a GPVI-specific agonist, indicating that PECAM-1 and Lyn participate in the same inhibitory pathway. Lyn was required for PECAM-1 tyrosine phosphorylation and subsequent binding of the Src homology 2 domain-containing phosphatase-2, SHP-2. These results support a model in which PECAM-1/SHP-2 complexes, formed in a Lyn-dependent manner, suppress GPVI signaling.

Figure 1

PECAM-1 and Lyn depend on one another to inhibit platelet responses to GPVI-specific agonists. (A) To control for genetic differences other than PECAM-1 or Lyn, platelets were obtained from age- and sex-matched PECAM-1–positive/Lyn-positive (P^+/+/L^+/+) and PECAM-1–negative/Lyn-positive (P^−/−/L^+/+) offspring of pecam-1^+/−/lyn^+/+ breeding pairs (i), Lyn-positive/PECAM-1–positive (L^+/+/P^+/+) and Lyn-negative/PECAM-1–positive (L^−/−/P^+/+) offspring of lyn^+/−/pecam-1^+/+ breeding pairs (ii), PECAM-1–positive/Lyn-negative (P^+/+/L^−/−) and PECAM-1-negative/Lyn-negative (P^−/−/L^−/−/) offspring of pecam-1^+/−/lyn^−/− breeding pairs (iii), and Lyn-positive/PECAM-1–negative (L^+/+/P^−/−) and Lyn-negative/PECAM-1–negative (L^−/−/P^−/−) offspring of lyn^+/−/pecam-1^−/− breeding pairs (iv). Pairwise combinations of platelets were stimulated in an aggregometer with various doses of CRP as described in “Isolation and stimulation of platelets” under constant stirring conditions. Low, intermediate, and high doses of CRP were empirically determined for each pairwise comparison. A low dose of CRP (range, 0.025-0.1 μg/mL) was defined as the highest dose of CRP that failed to induce aggregation of double-positive platelets for the data shown in subpanels i-ii, or of single-positive platelets for the data shown in subpanels iii-iv. A high dose of CRP (range, 0.2-0.6 μg/mL) was defined as the lowest dose of CRP that induced 100% aggregation of double-positive platelets for the data shown in subpanels i-ii, or of single-positive platelets for the data shown in subpanels iii-iv. An intermediate dose of CRP (range, 0.05-0.3 μg/mL) was between the high and low doses for each experiment. The percentage of aggregation was quantified in multiple pairwise comparisons (n = 3), and results are presented as mean percentage (± SEM) of aggregation. An asterisk identifies a statistically significant difference in platelet aggregation (P < .05). “NS” identifies differences that are not statistically significant. Note that PECAM-1 inhibits platelet responses to intermediate doses of CRP in the presence (i) but not in the absence (iii) of Lyn, and that Lyn is inhibitory in the presence (ii) but not in the absence (iv) of PECAM-1. (B) CRP-induced aggregation responses of platelets from age- and sex-matched mice of all 4 genotypes were compared in a single experiment using a dose of CRP (0.3 μg/mL) that induced an intermediate level of responsiveness from P^+/+/L^+/+ platelets. Note that PECAM-1–deficient, Lyn-deficient, and PECAM-1/Lyn double-deficient platelets are equally hyperresponsive, relative to wild-type platelets, to an intermediate dose of the GPVI-specific agonist CRP.

Figure 2

Lyn is required for PECAM-1 phosphorylation and SHP-2 binding in GPVI-stimulated platelets. Platelets isolated from wild-type (Lyn^+/+) or Lyn-deficient (Lyn^−/−) mice were left unstimulated (Resting) or were stirred in the presence of CRP to achieve 20% or 100% aggregation. (A) Lysates were prepared and immunoblotted with antibodies specific for the phosphorylated form of the C-terminal PECAM-1 ITIM tyrosine residue at position 686 (PECAM-1 pY686), PECAM-1 to show equal antigen loading, or Lyn to show the presence and absence of Lyn in wild-type and Lyn-deficient platelets, respectively. Note that PECAM-1 is tyrosine phosphorylated in CRP-aggregated wild-type but not in Lyn-deficient platelets. (B) Lysates were subjected to IP with normal mouse IgG (NMIgG) as a negative control or with anti–SHP-2. Lysates and IP lysates were immunoblotted with antibodies specific for PECAM-1 or SHP-2. Note that PECAM-1 coimmunoprecipitates with SHP-2 from CRP-aggregated wild-type but not Lyn-deficient platelets. Results shown are representative of 3 independent experiments.