Scavenger receptor BI modulates platelet reactivity and thrombosis in dyslipidemia

Abstract

Hypercholesterolemia is associated with increased platelet sensitivity to agonists and a prothrombotic phenotype. Mechanisms of platelet hypersensitivity are poorly understood; however, increased platelet cholesterol levels associated with hypercholesterolemia were proposed as leading to hypersensitivity. Scavenger receptor class B type I (SR-BI) in the liver controls plasma high-density lipoprotein (HDL) levels, and SR-BI-deficient mice display a profound dyslipoproteinemia. SR-BI is also expressed on platelets, and recent studies have suggested a role for SR-BI in platelet function; however, its role in hemostasis is unknown. Our present studies demonstrated that non-bone marrow-derived SR-BI deficiency and the dyslipidemia associated with it lead to platelet hyperreactivity that was mechanistically linked to increased platelet cholesterol content. Platelet-specific deficiency of SR-BI, on the other hand, was associated with resistance to hyperreactivity induced by increased platelet cholesterol content. Intravital thrombosis studies demonstrated that platelet SR-BI deficiency protected mice from prothrombotic phenotype in 2 types of dyslipidemia associated with increased platelet cholesterol content. These novel findings demonstrate that SR-BI plays dual roles in thrombosis and may contribute to acute cardiovascular events in vivo in hypercholesterolemia.

Figure 1

Agonist-dependent modulation of platelet aggregation of SR-BI^−/−–deficient mice. (A-F) Platelet aggregation in PRP from WT and SR-BI^−/− mice was induced by selective PAR4-AP, ADP, or convulxin and was optically monitored. (B,D,F) Representative aggregation curves in response to PAR4-AP (60μM to 150μM; B), or ADP (1μM; D), or convulxin (500 ng/mL; F) are shown, respectively. (A,C,E) Quantifications of the aggregation data are expressed as maximal amplitude of aggregation within 5 minutes after adding the agonist (mean ± SEM), and data are presented as a typical result of at least 3 independent experiments. *P < .05, **P < .01.

Figure 2

SR-BI deficiency modulates platelet integrin α_IIbβ₃ activation and P-selectin expression in the absence of changes in platelet expression of major receptors. (A-B) Platelets in PRP from WT and SR-BI^−/− mice were stimulated with selective protease-activated receptor 4–activating peptides (PAR4-AP), ADP, or convulxin (CVX). Platelet integrin α_IIbβ₃ activation and P-selectin (CD62) expression were determined using the PE-conjugated antibody for murine α_IIbβ₃ in the activated conformation (JON/A) or murine P-selectin. (C) Platelets were isolated by gel filtration from mice of indicated genotypes, and activation of integrin α_IIbβ₃ was assessed by FACS analysis. (A) Flow cytometry histograms from representative experiments are shown. (B-C) Quantification of FACS analysis data presented as mean ± SEM of at least 3 independent experiments. *P < .05, **P < .01. (D) Expression of thrombin receptors PAR4 and PAR3 and ADP receptors P2Y1 and P2Y12 were assessed by Western blotting. (E) Expression of collagen receptor GPVI and α_IIbβ₃ integrin. Platelets were incubated with FITC-labeled anti-α_IIbβ₃ or anti-GPVI antibody and analyzed by flow cytometry. Data are presented as mean ± SEM of at least 4 independent experiments.

Figure 3

Non–bone marrow–derived SR-BI deficiency leads to platelet hyperreactivity and increased platelet cholesterol content. (A) WT platelets in PRP either from WT or SR-BI^−/− recipients were stimulated with ADP, PAR4-AP, or convulxin (CVX), and platelet integrin α_IIbβ₃ activation and P-selectin expression were determined. (B-C) Platelet aggregation in PRP isolated from chimeric WT or SR-BI^−/− mice with WT bone marrow was induced by 120μM PAR4-AP and was optically monitored. (B) Quantification of the aggregation data. Data are expressed as maximal amplitude of aggregation within 5 minutes after adding the agonist. (C) Representative aggregation curves are shown. (D-E) Platelets in PRP from chimeric WT or SR-BI^−/− mice with WT bone marrow (BM) in panel D as well as from WT and SR-BI^−/− mice in panel E were stained with 50 μg/mL filipin to label unesterified cholesterol and analyzed by flow cytometry. All above data are presented as mean ± SEM of at least 3 independent experiments. (F) WT platelets were isolated by gel-filtration from pooled blood of WT chimeras and loaded with cholesterol in vitro by incubation with 75μM cholesterol-chelated MβCD. Incubation with alpha-cyclodextrin (αCD) was used as control. After cholesterol loading, platelets were stimulated with ADP, thrombin, or CVX, and platelet integrin α_IIbβ₃ activation was determined by FACS analysis. Data are presented as mean ± SEM of measurements after 3 separate cholesterol loading, which were repeated twice. *P < .05, **P < .01, ***P < .001.

Figure 4

Effects of platelet SR-BI deficiency on the platelet reactivity, platelet cholesterol content, platelet counts, and thrombosis in hyperlipidemic conditions of SR-BI^−/− mice. (A) Platelets in PRP from chimeric SR-BI^−/− mice with SR-BI^−/− bone marrow (BM) or WT BM were stimulated with ADP, PAR4-AP, or convulxin (CVX). Platelet integrin α_IIbβ₃ activation was determined by FACS analysis (n ≥ 3). (B-C) Platelet aggregation in PRP from SR-BI^−/− mice with SR-BI^−/− BM or WT BM was induced by 120μM PAR4-AP and was optically monitored. (B) Quantification of the aggregation data. Data are expressed as maximal amplitude of aggregation within 5 minutes after adding the agonist (n ≥ 3). (C) Representative aggregation curves are shown. (D) Platelet cholesterol contents for chimeric SR-BI^−/− mice with WT or SR-BI^−/− BM. Platelets were stained with 50 μg/mL filipin and analyzed by flow cytometry (n ≥ 3). (E) Times to thrombotic occlusion of carotid arteries of chimeric SR-BI^−/− mice with SR-BI^−/− BM or WT BM were measured 2 minutes after topical application of 10% FeCl₃. Carotid arteries were visualized, and in vivo thrombosis formation was assessed by intravital microscopy as described in “Methods” n = 7. (F) Progression of thrombus in carotid arteries is shown. Times after FeCl₃-induced injury are indicated (in minutes). At 10 minutes, the artery from the SR-BI^−/− (WT BM) mice was completely occluded, whereas the SR-BI^−/− (SR-BI^−/− BM) mice showed large thrombi with persistent blood flow. All images were observed under a Leica DM LFS microscope (Leica) with 10×/0.30 objective lens and acquired by a cooled high-speed, color, cooled digital camera (QImaging Retiga EXi Fast 1394) with Streampix high-speed acquisition software. (G-H) Platelet counts in whole blood collected from chimeric SR-BI^−/− mice with SR-BI^−/− BM or WT BM and from chimeric WT mice with WT BM in panel G, as well as from WT or SR-BI^−/− mice in panel H. n ≥ 5. All data are presented as mean ± SEM. *P < .05, **P < .01, ***P < .001.

Figure 5

Effects of platelet SR-BI deficiency on the platelet reactivity, platelet cholesterol content, and platelet counts in normolipidemic conditions of WT mice. (A-B) Platelets in PRP in panel A or isolated by gel filtration in panel B from chimeric WT mice with SR-BI^−/− BM or WT BM were stimulated with ADP, PAR4-AP, thrombin, or convulxin (CVX), and platelet integrin α_IIbβ₃ activation was assessed by FACS analysis. (C-D) Platelet aggregation in PRP isolated from chimeric WT mice with SR-BI^−/− BM or WT BM was induced by ADP (5μM), convulxin (500 ng/mL), or PAR4-AP (150μM) and was optically monitored. (C) Quantification of the aggregation data. Data are expressed as maximal amplitude of aggregation within 5 minutes after adding the agonist. (D) Representative aggregation curves are shown. (E) Platelet cholesterol contents for chimeric WT mice with WT or SR-BI^−/− BM. Platelets were stained with 50 μg/mL filipin and analyzed by flow cytometry. (F) Platelet counts in whole blood collected from chimeric WT mice with SR-BI^−/− BM or WT BM. Data are presented as mean ± SEM of at least 3 independent experiments *P < .05, **P < .01.

Figure 6

Effects of cholesterol loading in vitro on the activation of WT and SR-BI^−/− platelets. WT or SR-BI^−/− platelets were isolated by gel-filtration from pooled blood of WT chimeras and loaded with increasing amounts of cholesterol in vitro by incubation with cholesterol-chelated MβCD. Incubation with αCD was used as control. (A-C) After cholesterol loading platelets were stimulated with ADP, thrombin, or convulxin (CVX), and platelet integrin α_Πbβ₃ activation was determined by FACS analysis. (D) After cholesterol loading, platelets were stained with 50 μg/mL filipin to label unesterified cholesterol and analyzed by flow cytometry. Data are presented as mean ± SEM of measurements after 3 separate cholesterol loading, which were repeated twice. *P < .05, **P < .01, ***P < .001, compared with the same platelets without cholesterol loading. ^#P < .05, ^##P < .01, ^###P < .001, compared with WT platelets with the same concentration of cholesterol loading.

Figure 7

Platelet SR-BI contributes to increased platelet responses and thrombosis in hyperlipidemic apoE^−/− mice. (A) Platelets in PRP from apoE^−/− mice fed Western-type diet and from matched WT mice were stained with 50 μg/mL filipin to label unesterified cholesterol and analyzed by flow cytometry (n = 4). (B-E) ApoE^−/− mice were reconstructed with either apoE^−/−/SR-BI^−/− bone marrow or apoE^−/− bone marrow. One month later, Western diet feeding was started and continued for 6 weeks, and then mice were used for experiments. (B) Agonist-induced platelet integrin α_IIbβ₃ activation was assessed by flow cytometry in platelet isolated by gel filtration. (n ≥ 3). (C-D) Platelet aggregation in PRP from apoE^−/− mice with apoE^−/− BM or apoE^−/−/SR-BI^−/− BM was induced by PAR4-AP and optically monitored. (C) Representative aggregation curves are shown in response to 120μM PAR4-AP. (D) Quantification of the aggregation data expressed as maximal amplitude of aggregation within 5 minutes after induction (n ≥ 3). (E) Times to thrombotic occlusion of carotid arteries of chimeric apoE^−/− mice with apoE^−/− BM or apoE^−/−/SR-BI^−/− BM were measured 2 minutes after topical application of 7% FeCl₃. n = 11. Data are presented as mean ± SEM. * P < .05, **P < .01.