Figure 1.
Expression of Siglec-7 in human platelets.
(A), (B) Siglec-7 expression in platelets was analyzed by flow cytometry (n = 20). Platelets (stimulated with TRAP or vehicle control) were labeled with anti-CD41a and anti-Siglec-7 mAbs. Membrane expression of Siglec-7+ was increased after TRAP-induced platelet activation (*significant difference, of %CD41a+Siglec-7+ between TRAP-induced platelets activation vs resting platelets, t-test, p<0.05). (C) Membrane and intracellular localization of Siglec-7 in platelets. Platelets before and after permeabilization (stimulated with TRAP or vehicle control) were labeled with anti-CD41a, anti-CD62P (positive control for permeabilized platelets) and anti-Siglec-7 mAbs. Gating on the CD41a+ population, intra-platelet expression of Siglec-7 is significantly higher than its membrane expression (in both stimulated and unstimulated) (*, # significant difference (t-test, p<0.05) between %CD41a+CD62P+ or %CD41a+Siglec-7+, respectively, vs non-permeabilized resting platelets. (D) Distribution of Siglec-7 in platelets analyzed by confocal microscopy. Immunofluorescence labeling with anti–CD41a, and anti-Siglec-7 mAbs, and overlay (top to bottom). Siglec-7 staining in permeabilized (right panel) and non-permeabilized platelets (left panel) shows intracellular expression is more important than membrane expression. Scale bars = 6 µm. (E) Labeling with anti-tubulin and anti Siglec-7 antibodies and overlay (top to bottom). Siglec-7 is expressed on platelet membranes and tubulin is stained in permeabilized platelets. Tubulin was labeled to demarcate platelet borders and Siglec-7 is mostly observed in intracellular compartments. Scale bars = 10 µm.
Figure 2.
Siglec-7 localization in human platelets.
(A) Siglec-7 does not colocalize with markers for dense granules, lysosomes, T-granules or endosomes. From top to bottom: intra-platelet Siglec-7 was co-stained with serotonin (dense granules), M6P (endosomes), TLR-9 (T-granules), and LAMP-1 (lysosomes) using two distinct colored secondary antibodies. Insets represent magnified regions shown in yellow boxes. Scale bars = 10 µm, except insets where scale bars = 2 µm. (B) Scattergram, scatter plot, and colocalization analysis of Serotonin/Siglec-7; M6P/Siglec-7; TLR-9/Siglec-7; and LAMP-1/Siglec-7 (left to right). Siglec-7 did not colocalize with any proteins (Pearson correlation analysis, p>0.05). (C) Siglec-7 colocalized with CD62P, an α-granule marker. Scale bars = 10 µm except insets bars, which = 2 µm. Pearson correlation analysis, p<0.05.
Figure 3.
Characterization of Siglec-7 as a novel marker of platelet activation.
(A),(C). Agonist-induced platelet activation (n = 10). Platelets were stimulated with different agonists: TRAP (PAR-1), PAR-4 activator, collagen (β2α1 integrin and GPVI-FcR gamma). Membrane MFI expression of CD62P (A) and Siglec-7 (B) was correlated under all experimental conditions (PCC = 0.995, p = 0.002) (C). (D),(E). Kinetics of CD62P (D) and Siglec-7 (E) expression on platelet membranes (n = 5). Membrane expression of Siglec-7 and CD62P in platelets under un-stimulated and TRAP stimulated conditions was analyzed by flow cytometry. MFI changes between CD62P and Siglec-7 were correlated over time (unstimulated: PCC = 0.842; p = 0.016; stimulated by TRAP: PCC = 0.958; p = 0.01). *, #, ¥, ¤: significant differences (analysis of variance, p<0.05) between MFI of Siglec-7 or CD62P marker over time vs 0, 30, 60, and 180 min respectively; ‡: t-test, p<0.05) MFI of Siglec-7 or CD62P marker in TRAP-induced platelets activation vs resting platelets (expression of CD62P and Siglec-7 in MFI was reported for unstimulated conditions at 0 min, which was considered as 100%).
Figure 4.
Effects of Siglec-7 ligand, ganglioside, on platelet function.
(A) Platelet activation upon stimulation with GD2 (n = 10). *: Significant difference, t-test, p<0.05, changes in membrane expression of CD62P in platelets stimulated by TRAP vs resting platelets. There was no significant change in the expression of CD62P between platelets stimulated with GD2 or vehicle control (t-test, p>0.05). Similar results were observed for GD3 and GT1b stimuli. Thus, gangliosides show no effect on platelet activation. (B) Platelet aggregation (n = 3) analyses. Incubation of platelets with either GD2 or vehicle did not induce platelet aggregation or alter their response to ADP stimulation (10 µM). (C) Platelet secretion analysis: Platelet secretion of serotonin, sCD40L, and RANTES induced by GD2 stimulation (n = 10). The secretion of resting platelets was slightly lower than both the vehicle and GD2-stimulated platelets; however, there was no significant difference between the latter two groups (t-test, p>0.05).
Figure 5.
Effect of GD2 on platelet death.
(A) Cell death induced by GD2 stimulation. Platelets untreated or pre-treated with blocking anti-Siglec-7 pAb were incubated with vehicle or GD2. A23187 was used as a positive control for apoptosis. Cell death rate (%) was measured as the percentage of diminution in platelet counts relative to unstimulated platelets. *, #: significant differences (t-test, p<0.05) vs vehicle or GD2 stimulated conditions. Results are representative of five independent experiments. (B)(C) Morphology of platelets analyzed by scanning electron microscopy. (B) The morphology of platelets stimulated with GD2 resembled apoptotic cells as characterized by blebbing, filopod extrusion (1,2) and cell shrinkage (3). (C) Resting platelets have a discoid shape with a smooth cell surface. (D): A23187-induced platelet apoptosis, red (Scale bars = 20 µm) and green (Scale bars = 10 µm) arrows that represent platelet apoptosis.
Figure 6.
Mechanism of GD2-induced platelet apoptosis.
Both platelets pretreated with blocking anti-Siglec-7 PAb and untreated platelets were incubated with vehicle or GD2. A23187 was used as an apoptosis positive control. Different parameters of apoptosis pathways in platelets were analyzed. (A) Expression of TRAIL R1 in platelets stimulated with GD2 (n = 12). *: significant difference (t-test, p<0.05) vs resting platelets. The extrinsic pathway may not be involved in platelet apoptosis induced by GD2. (B) Analyses of phosphatidylserine (PS) exposure (n = 13). The extent of PS exposure by GD2 was reduced by blocking anti-Siglec-7 pAb in a concentration-dependent manner. *, #: Significant difference (t-test, p<0.05) vs vehicle or GD2 stimulated conditions respectively. (C) Platelet microparticle assay (n = 10). PMP formation was calculated with respect to the concentration in vehicle conditions (arbitrarily designated 100%). *, #: significant difference (Wilcoxon paired test, p<0.05) vs vehicle or GD2 stimulated conditions, respectively. (D) ΔΨm depolarization. ΔΨm depolarization resulted in decreased DIOC6(3) accumulation. *, #: Significant difference (t-test, p<0.05) vs vehicle or GD2 stimulated conditions, respectively. Results are representative of five independent experiments. GD2-induced mitochondrial depolarization in platelets treated with anti-FcγRII mAb or PBS control. *, ** significant difference (t-test, p<0.05) of ΔΨm between GD2 stimulated platelets vs unstimulated platelets in the presence or absence of anti-FcγRII mAb. #, ¥: Anti-Siglec-7 PAb + GD2 vs only-GD2 stimulated platelets in the presence or absence of anti-FcγRII mAb (n = 5). NS: Not significant. (E) Western blot demonstrates strong expression of Bax and Bak in GD2-treated human platelets and this expression was prevented by blocking anti Siglec-7 pAb. Results are representative of three independent experiments.
Figure 7.
PI3K inhibitors reduce GD2-induced platelet apoptosis (n = 15).
(A) Platelets were pre-treated with varying concentrations of intracellular pathway inhibitors (DPI for NAPDH oxidase, LY294,002 for PI3K, BIM I for PKC and BAY-11 for NFκB) or DMSO (negative control) followed by stimulation with GD2. Cell death rate (%) was calculated by the percentage of diminution in platelet number in comparison to unstimulated platelets. DPI, BIM I, and LY294,002 prevented cell death in a concentration-dependent manner. *: Significant difference (t-test, p<0.05) between inhibitor vs DMSO treated. The lowest concentration of inhibitors with a significant effect (DPI 1 µM, LY294,002 50 µM, BIM I 10 µM) (no difference at a higher concentration) was selected to treat platelets. (B) Loss of ΔΨm resulted in reduced accumulation of DIOC6(3), and was calculated as the percentage of diminution in DIOC6(3) MFI compared with unstimulated platelets. (C) Phosphatidylserine exposure (left panel: percentage of CD41a+, Annexin V+; right panel: MFI). *: Significant difference (t-test, p<0.05) between inhibitor vs DMSO pretreated platelets. Thus, PI3K inhibitor prevented both ΔΨm depolarization and PS exposure in platelets, while NAPDH oxidase inhibitor prevented only ΔΨm depolarization.
Figure 8.
Ganglioside-induced platelet apoptosis is regulated by platelet P2Y1 and GPIIbIIIa antagonists (n = 10).
Platelets were pre-treated with antagonists for the corresponding platelet receptors: (A) MRS2179 100 µM: P2Y1 antagonist; (B) Tirofiban 10 µM: GPIIbIIIa antagonist; (C). Clopidogrel 476 µM: P2Y12 antagonist; (D) SCH79797 10 µM: PAR-1 antagonist; (E) tcY-NH2 400 µM: PAR-4 antagonist) or DMSO at the same dilution (negative control) followed by stimulation with GD2. The cell death rate (percentage of diminution in platelet counts) and ΔΨm depolarization (percentage of diminution in DIOC6(3) MFI) were prevented by P2Y1 and GPIIbIIIa antagonists. * Significant difference (t-test, p<0.05) between antagonist vs DMSO pretreated platelets.