Integrative Physiology

Sphingosine 1-Phosphate Produced by Sphingosine Kinase 2 Intrinsically Controls Platelet Aggregation In Vitro and In VivoNovelty and Significance

Nicole Urtz, Florian Gaertner, Marie-Luise von Bruehl, Sue Chandraratne, Faridun Rahimi, Lin Zhang, Mathias Orban, Verena Barocke, Johannes Beil, Irene Schubert, Michael Lorenz, Kyle R. Legate, Andrea Huwiler, Josef M. Pfeilschifter, Christian Beerli, David Ledieu, Elke Persohn, Andreas Billich, Thomas Baumruker, Michael Mederos y Schnitzler, Steffen Massberg
https://doi.org/10.1161/CIRCRESAHA.115.306901
Circulation Research. 2015;117:376-387
Originally published June 30, 2015

Jump to

Abstract

Rationale: Platelets are known to play a crucial role in hemostasis. Sphingosine kinases (Sphk) 1 and 2 catalyze the conversion of sphingosine to the bioactive metabolite sphingosine 1-phosphate (S1P). Although platelets are able to secrete S1P on activation, little is known about a potential intrinsic effect of S1P on platelet function.

Objective: To investigate the role of Sphk1- and Sphk2-derived S1P in the regulation of platelet function.

Methods and Results: We found a 100-fold reduction in intracellular S1P levels in platelets derived from Sphk2−/− mutants compared with Sphk1−/− or wild-type mice, as analyzed by mass spectrometry. Sphk2−/− platelets also failed to secrete S1P on stimulation. Blood from Sphk2-deficient mice showed decreased aggregation after protease-activated receptor 4-peptide and adenosine diphosphate stimulation in vitro, as assessed by whole blood impedance aggregometry. We revealed that S1P controls platelet aggregation via the sphingosine 1-phosphate receptor 1 through modulation of protease-activated receptor 4-peptide and adenosine diphosphate–induced platelet activation. Finally, we show by intravital microscopy that defective platelet aggregation in Sphk2-deficient mice translates into reduced arterial thrombus stability in vivo.

Conclusions: We demonstrate that Sphk2 is the major Sphk isoform responsible for the generation of S1P in platelets and plays a pivotal intrinsic role in the control of platelet activation. Correspondingly, Sphk2-deficient mice are protected from arterial thrombosis after vascular injury, but have normal bleeding times. Targeting this pathway could therefore present a new therapeutic strategy to prevent thrombosis.

Introduction

The primary physiological function of platelets is to stop bleeding from sites of vascular injury by adhering to exposed extracellular matrix proteins and by formation of aggregates that plug damaged blood vessels. During this process, platelets release proaggregatory mediators from their characteristic storage granules, including platelet-derived adenosine diphosphate (ADP),1 as well as the arachidonic acid (AA) derivatives thromboxane A2 (TBXA2) and prostaglandins.2 Particularly, TBXA2 and ADP act together in a positive feedback loop to amplify and sustain the recruitment of circulating platelets into a growing hemostatic plug.3 Although platelet activation and recruitment are essential for hemostasis, they may also promote vascular disease complications, including myocardial infarction and stroke, which remain the leading causes of morbidity and mortality in most industrialized countries.4 Therefore, detailed knowledge of the versatile activation mechanisms of platelets offers new perspectives in the prophylaxis and treatment of cardiovascular diseases.

Apart from TBXA2, another lipid mediator released by platelets is sphingosine 1-phosphate (S1P). Sphingosine kinase 1 (Sphk 1) and 2 catalyze the formation of S1P from Sph and adenosine triphosphate (ATP). The bioactive lipid metabolite S1P is involved in the control of diverse cellular processes in different cell types, including cell proliferation, cell survival and migration, cytoskeletal rearrangements, cell adhesion and inflammation, as well as angiogenesis.5–10 S1P not only exerts this wide range of effects predominantly by binding to 1 of the 5 known G-protein–coupled membrane receptors (S1PR1–S1PR5)11 but also acts as a second messenger inside cells.12 The concentrations of S1P are high in blood and lymph fluid, but low in tissues. Maintenance of a steep blood-to-tissue S1P gradient is of critical importance for trafficking and immune surveillance by adaptive immune cells, particularly T cells.13

Intracellular S1P levels are low in many cell types and are tightly regulated via the irreversible cleavage of S1P by S1P-lyase or via dephosphorylation to Sph by S1P-phosphatases.14 However, unlike most other blood cells, platelets contain high Sphk activities15 but at the same time lack S1P-lyase activity and are therefore able to store and release high amounts of S1P.16,17 Erythrocytes also lack the S1P degrading machinery, but they have only low constitutive Sphk activity.18 As a consequence, platelets were originally considered to be the major source of blood S1P. However, studies showing that thrombocytopenic mice having normal plasma S1P concentrations have challenged this concept.13,17 Therefore, although platelets store and release S1P, red blood cells and also vascular endothelial cells are currently considered to be the major cellular contributors of plasma S1P,19 establishing and maintaining the steep blood-to-tissue S1P gradient.20 Hence, given its redundancy in maintenance of blood S1P levels, the role of platelet-derived S1P still remains unclear.

Here, we addressed the importance of platelet-derived S1P for the intrinsic control of distinct platelet functions. We report that activated but not resting platelets release S1P and that Sphk2, but not Sphk1, is responsible for S1P production in platelets. Platelet S1P generation is dispensable for S1P homeostasis in blood; however, it is essential for cell autonomous control of platelet activation. Correspondingly, we demonstrate that Sphk2 null platelets produce reduced amounts of TBXA2 in response to activation and fail to aggregate normally both in vitro and in response to vascular injury in vivo, while platelet adhesion remains unaffected. We show that S1P directly initiates whole blood aggregation via S1P receptor 1 expressed by platelets and facilitates platelet aggregation in response to protease-activated receptor 4-peptide (PAR4-P) and ADP. High concentrations of exogenous S1P rescue defective aggregation of Sphk2 null platelets. Together, our findings identify S1P released by platelet Sphk2 acting via S1P receptor 1 (S1PR1) as a novel cell autonomous pathway intrinsically regulating platelet functions.

Methods

Full Methods are described in the Online Data Supplement.

Animals

Specific pathogen-free wild-type (WT) C57BL/6J mice were purchased from Charles River Laboratories. Sphk1- and Sphk2-deficient mice on the C57BL/6 background were a kind gift from Andrea Huwiler, Institute of Pharmacology, University of Bern, Switzerland, whereas Sphk1- and Sphk2-deficient mice on the Balb/c background21 were a kind gift from Novartis Institutes for Biomedical Research, Basel, Switzerland.

All mice used in the experiments were 8 to 14 weeks old and weight and sex matched. All experimental procedures performed on animals were approved by the local legislation on protection of animals (Regierung von Oberbayern, Munich).

Results

Sphk2 Is the Major Isoform in Platelets and Responsible for S1P Generation and Secretion

Several previous studies have demonstrated that platelets store and secrete S1P into the plasma,16,22 however, the molecular pathways involved have remained unclear. Therefore, we first focused on identifying the Sphk isoform(s) responsible for the formation and release of S1P by platelets. To achieve this, we analyzed Sphk1 and Sphk2 mutant mice. As reported previously, red blood cell and leukocyte counts were normal, whereas platelet counts were reduced in Sphk2-deficient mice, but not in Sphk1-deficient mice (on C57B6/J and Balb/c background) because of a defect in proplatelet shedding by Sphk2-deficient megakaryocytes.23 We purified platelets from WT, Sphk1−/−, or Sphk2−/− mice and analyzed S1P levels in platelet lysates by liquid-chromatography/mass spectrometry. Whereas platelets lacking Sphk1 show an increase in intracellular S1P levels, S1P was hardly detectable in Sphk2-deficient platelets (Figure 1A). We did not find any differences in the gross platelet morphology, the number of platelet mitochondria, or in the number of specific vesicles/granules between WT and mutant mice, indicating that the inability of Sphk2-deficient platelets to synthetize and secrete S1P does not result from or lead to a disturbed organelle composition (Figure 1B). We also analyzed the mRNA expression of both Sphks in WT, Sphk1- and 2-deficient platelets and found that the loss of 1 Sphk isoform does not affect the expression of the other Sphk isoform (N. Urtz, unpublished data, 2007). To further substantiate the role of both Sphk isoforms for platelet S1P synthesis and secretion, we incubated WT, Sphk1-, or Sphk2-deficient platelets with [3H]-D-erythro-sphingosine and analyzed platelet supernatants for the presence of S1P by thin layer chromatography. Resting WT and Sphk1−/− platelets each secreted low amounts of S1P into the medium. This S1P release was markedly increased in response to α-thrombin stimulation (Figure 1C). In contrast, platelets lacking Sphk2 failed to secrete S1P even after stimulation by α-thrombin. Taken together, these findings have 2 major implications: (1) They show that Sphk2 is the major isoform responsible for S1P generation in platelets, and (2) they demonstrate that Sphk1 cannot compensate for a loss of Sphk2 in platelets.

Figure 1.
Figure 1.

Sphingosine kinase 2 (Sphk2) is the major isoform in murine platelets. A, Intracellular sphingosine 1-phosphate (S1P) levels were assayed by liquid-chromatography/mass spectrometry from wild-type (WT), Sphk1-, and Sphk2-deficient platelets. Values represent mean±SD, n=3/group. Statistical significance was calculated using Mann–Whitney rank-sum test. B, Electron microscopic evaluation of WT, Sphk1-, and Sphk2-deficient platelets showing normal morphology; note the mitochondria (black arrow M) and the specific granules (black arrow SG) in the cytoplasm. Bar, 2 μm; n=5/group. C, Sphk2-deficient platelets fail to secrete S1P into the medium. 3H-S1P secretion from unstimulated or α-thrombin–stimulated platelets (time points indicated on the top) isolated from WT, Sphk1-, and Sphk2-deficient mice were analyzed by thin layer chromatography (TLC). Shown is an autoradiogram of a TLC plate representative of 2 independent experiments. Left, The position of 3H-S1P and 3H-S on the TLC plate is indicated.

Sphk1 Is the Major Isoform in Red Blood Cells and Contributes to Plasma S1P

We next examined the role of both Sphk isoforms for S1P production in red blood cells, which are thought to be the main source of plasma S1P. To address this, we isolated red blood cells from WT, Sphk1-, or Sphk2-deficient mice and determined S1P levels by liquid-chromatography/mass spectrometry. Interestingly, red blood cells lacking Sphk2 show intracellular S1P levels similar to WT cells. In contrast, Sphk1-deficient red blood cells fail to generate S1P (Figure 2A). We also performed Sphk activity assays in red blood cells following the addition of [33P]-ATP. [33P]-S1P was extracted from cell lysates and analyzed by thin layer chromatography. We could not detect any S1P in Sphk1−/− red blood cells, whereas WT or Sphk2−/− red blood cells revealed normal S1P generation (Figure 2B).

Figure 2.
Figure 2.

Sphingosine kinase 1 (Sphk1) is the major isoform in red blood cells and regulates plasma sphingosine 1-phosphate (S1P) in mice. A, Intracellular S1P levels from wild-type (WT), Sphk1-, and Sphk2-deficient red blood cells were assayed by mass spectrometry (MS). Values represent mean±SD; n=3/group. B, Sphk activities were measured in WT (black bar), Sphk1-deficient (grey bar), and Sphk2-deficient (white bar) red blood cells in the presence of 33P-γ-ATP. Shown is the quantification of the generation of 33P-S1P; n=4/group. C, S1P plasma levels in WT, Sphk1-deficient, and Sphk2-deficient mice analyzed by MS; n=3/group. Statistical significances were calculated using Mann–Whitney rank-sum test.

Because S1P production in platelets and red blood cells is regulated by different Sphks, we next explored how S1P production by the 2 isoforms would affect plasma S1P levels. We found that in Sphk1-deficient mice S1P plasma concentrations are reduced by about 50% when compared with WT (Figure 2C). This finding is consistent with data reported previously by others.21,24 In contrast, loss of Sphk2 did not reduce but rather yielded a moderate increase of plasma S1P levels (Figure 2C), a finding that is in line with previous reports.21,25

In red blood cells, Sphk1, but not Sphk2, is therefore essential for the catalytic conversion of Sph to S1P. The reduced plasma S1P levels in Sphk1-deficient mice support the concept that red blood cells are a main source of plasma S1P. In contrast, platelet S1P production, which depends on Sphk2, is dispensable for maintenance of S1P homeostasis in the blood. This is consistent with the low steady state release of S1P by platelets, which is increased dramatically only when platelets are activated (Figure 1C).

Sphk2-Deficient Platelets Release Decreased Amounts of AA and Its Metabolite Thromboxane A2

Rather than contributing to general blood S1P homeostasis, S1P synthesis by activated platelets might act locally by providing an intrinsic and cell autonomous control of platelet function in scenarios of vessel damage. To test this hypothesis, we first analyzed whether the inability to synthetize and secrete S1P might affect the capability of Sphk2−/− platelets to release potent platelet agonists. To this end, we examined the AA release from WT, Sphk1−/−, or Sphk2−/− platelets after preincubation with [3H]-AA. On activation with α-thrombin for the indicated time points, release of AA into the supernatants was measured (Figure 3A). Loss of Sphk1 had no significant effect on the release of AA (except 30 minutes after stimulation) and its metabolite TBXA2 by resting or thrombin-activated platelets when compared with WT platelets (Figure 3B). In contrast, release of AA, and in particular of TBXA2, was significantly reduced in activated platelets obtained from Sphk2-deficient mice compared with WT or Sphk1-deficient mice (Figure 3A and 3B). No difference in TBXA2 release was observed in response to collagen, whereas ADP did not trigger any TBXA2 generation in either WT or Sphk null platelets (N. Urtz, unpublished data, 2014).

Figure 3.
Figure 3.

Sphingosine kinase 2 (Sphk2)-deficient murine platelets release decreased amounts of arachidonic acid (AA) metabolites. A, Mildly decreased AA release after α-thrombin activation. Wild-type (WT; black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) platelets were left unstimulated (0 minutes) or induced with α-thrombin, and release of AA was analyzed at the indicated time points. Shown are the mean values±SD from 1 experiment representative of 3 independent experiments; n=4/group. B, Decreased thromboxane A2 (TBXA2) release after α-thrombin activation in Sphk2-deficient platelets. WT (black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) platelets were left unstimulated (0 minutes) or induced with α-thrombin, and release of TBXA2 in the supernatant was analyzed at the indicated time points. Shown are the mean values±SD; n=6/group. C, Normal adenosine triphosphate (ATP) release after α-thrombin activation. WT (black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) platelets were left unstimulated (0 minutes) or induced with α-thrombin, and release of ATP in the supernatant was analyzed at the indicated time points. Shown are the mean values±SD from 1 experiment representative of 2 independent experiments, n=3/group. Statistical significances were calculated using Mann–Whitney rank-sum test.

To also define the role of platelet S1P for the release of dense granules, we measured the ATP concentration in the supernatants of platelets isolated from WT and Sphk1- or Sphk2-deficient mice. As demonstrated in Figure 3C, loss of Sphk2 does not impair dense granule storage or depletion but rather leads to a mild increase in ATP release in response to thrombin activation. Together our findings suggest that loss of Sphk2 in platelets is associated with a severe defect in the release of the AA metabolite TBXA2, whereas release of dense granules is not affected.

Sphk2-Deficient Platelets Accomplish Normal Adhesion In Vitro and In Vivo

We subsequently asked whether these findings translate into a defect in platelet functions. Adhesion of resting platelets to exposed extracellular matrix proteins at sites of injured vascular walls is the first step in primary hemostasis. To test the role of Sphk2-dependent S1P release in this process, we performed in vitro adhesion assays, in which isolated and fluorescently labeled platelets were perfused over chamber slides coated with different matrix proteins, including fibrinogen, von Willebrand factor, and laminin. We found that both Sphk1- and Sphk2-deficient platelets adhere normally to fibrinogen, von Willebrand factor, or laminin (Figure 4A and 4B). Similar results were obtained when we used coating with fibrillar collagen (Online Figure I).

Figure 4.
Figure 4.

Sphingosine kinase 2 (Sphk2)-deficient platelets show normal adhesion in vitro and in vivo. A (Top), Representative photomicrographs of adherent single platelets (green) after 5 minutes of perfusion over fibrinogen (fbg) in vitro. Lower, Representative photomicrographs of small adherent platelet aggregates (red) 30 minutes after vessel denudation in vivo. Bars, 100 μm. B, In a flow chamber fluorescently labeled wild-type (WT; black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) platelets were perfused over different immobilized matrix proteins (laminin [lam], fbg, and von Willebrand factor [vWf]). Values represent mean±SEM; n=6/group. C, Sphk2-deficient platelets show normal adhesion in vivo. Adhesion of fluorescently labeled WT (black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) platelets after wire-induced injury of the arteria femoralis were analyzed in the corresponding recipient mice at the indicated time points. Values represent mean±SEM, n=5/group. Statistical significances were calculated using Mann–Whitney rank-sum test.

To further analyze the impact of Sphk-dependent S1P generation in vivo, we next evaluated platelet adhesion to the vessel wall in response to endothelial disruption in mice. To achieve this, we induced transluminal mechanical injury of the femoral arteries using a flexible wire as described previously.26,27 This model is known to promote platelet adhesion in a multistep process involving collagen, von Willebrand factor, and fibrinogen.27 Fluorescently labeled platelets from WT, Sphk1-, or Sphk2-deficient donors were infused intravenously into genotype-matched recipients. Platelet adhesion was monitored by intravital epifluorescence microscopy at different time points after denudation. Consistent with our in vitro findings, platelets lacking Sphk1 or Sphk2 adhered normally to the injured vessel wall throughout the entire observation period (Figure 4A and 4C). Together, these data suggest that Sphk2-dependent intrinsic S1P release is dispensable for platelet adhesion in injured arteries in vivo.

S1P Intrinsically Controls Platelet Aggregation via S1P Receptor S1PR1

S1P has been reported to induce platelet shape change and aggregation, the steps ensuing platelet adhesion during thrombosis and hemostasis.22,28 Indeed, using impedance aggregometry of whole blood from healthy human donors, we observed that S1P provokes a weak, but significant and dose-dependent aggregation of platelets. The proaggregatory effects of S1P are first seen at concentrations just above those reported in plasma and reach a maximum with the highest S1P concentration tested (50 μmol/L; Figure 5A). Next, we examined the S1P receptor subtypes involved in S1P-induced platelet aggregation. We have previously shown that the platelet/megakaryocyte lineage predominantly expresses the S1PR subtypes S1PR1, as well as S1PR2, and S1PR4, but not S1PR3 or S1PR529. SEW2781, a potent and selective S1PR1 agonist, triggered a dose-dependent platelet aggregation similar to that observed with S1P, suggesting that S1PR1 predominantly mediates S1P-dependent platelet aggregation, whereas other S1PRs are dispensable (Figure 5B). Consequently, W146, which inhibits S1PR1, but not the S1PR2 antagonist JTE013, dose dependently blocked S1P-stimulated whole blood aggregation (Figure 5C; Online Figure II).

Figure 5.
Figure 5.

Sphingosine 1-phosphate (S1P) controls platelet aggregation via S1P receptor 1 (S1PR1) in human blood. A, S1P triggers whole blood aggregation. Whole blood was incubated with different S1P concentrations or treated with vehicle alone as control (ctrl) and platelet aggregation was analyzed; n=10 donors. B, S1PR1 agonist SEW2781 triggers aggregation. Whole blood was incubated with different SEW2781 concentrations or treated with vehicle alone as ctrl and platelet aggregation was analyzed; n=6 donors. C, S1PR1 antagonism prevents S1P-induced aggregation. Blood was pretreated with different W146 concentrations or treated with vehicle alone (ctrl). Subsequent platelet aggregation was induced with S1P; n=6 donors. D, S1PR1 affects PAR4-P and adenosine diphosphate (ADP)-induced aggregation. Whole blood was treated either with vehicle alone or with the S1PR1 antagonist W146 (1 μmol/L), and platelet aggregation was induced with collagen (col), ADP, and protease-activated receptor 4-peptide (PAR4-P); n=10 donors. Statistical significances were calculated using Mann–Whitney rank-sum test. AU indicates arbitrary units; and AU*min, AUC (area under the curve).

We then tested whether endogenous S1P released by platelets in response to activation by prototypic platelet agonists also modulates aggregation via S1PR1. To address this, human whole blood was pretreated with W146 to inhibit the S1PR1 or with vehicle alone. Inhibition of S1PR1 led to a significant impairment of PAR4-P and ADP-triggered platelet aggregation, whereas aggregation in response to collagen remained unaffected (Figure 5D). Together, this indicates that Sphk2-dependent S1P acts as an intrinsic and cell autonomous regulator, facilitating platelet aggregation to PAR4 agonists and ADP but not to collagen in a process involving S1PR1.

Sphk2-Deficient Platelets Reveal Impaired Aggregation

Next, we further evaluated the above hypothesis that platelets intrinsically control their aggregation via Sphk2-dependent S1P generation and release. To test this, whole blood was taken from WT or Sphk1-deficient or Sphk2-deficient mice and exposed to collagen, PAR4-P, or ADP to trigger aggregation. As shown in Figure 6A, loss of Sphk2 was associated with a severe (40%) reduction in PAR4-P, as well as ADP-induced aggregation, whereas platelet aggregation in response to collagen was only moderately affected. In contrast, Sphk1-deficient platelets aggregated normally in response to all agonists tested (Figure 6A). These findings were consistent across different genetic backgrounds of mice (N. Urtz, unpublished data, 2011). Consistent with previous reports,30 dilution of blood from WT mice with platelet poor plasma was not sufficient to phenocopy the defect in platelet aggregation observed in Sphk2-deficient mice (N. Urtz, unpublished data, 2011), indicating that the reduced platelet numbers in these mutants are unlikely to be responsible for their defect in PAR4-P and ADP-induced aggregation.

Figure 6.
Figure 6.

Sphingosine kinase 2 (Sphk2)-deficient platelets show impaired aggregation. A, Whole blood from wild-type (WT; black bars), Sphk1-deficient (grey bars), and Sphk2-deficient (white bars) mice was triggered either with collagen (col), protease-activated receptor 4-peptide (PAR4-P), or with adenosine diphosphate (ADP) and aggregation was analyzed; n=10/group. B, Blocking of sphingosine 1-phosphate receptor 1 (S1PR1) decreases thromboxane A2 (TBXA2) release. WT platelets were pretreated with vehicle (ctrl) or with 1 μmol/L W146 and induced with α-thrombin. Release of TBXA2 in the supernatant was analyzed. Values present mean±SEM; n=5/group. C, Exogenous TBXA abolishes PAR4-P–induced impaired aggregation in Sphk2-deficient platelets. Whole blood from WT (black bars) or Sphk2-deficient (white bars) mice were induced either with PAR4-P alone or simultaneously with TBXA2, and aggregation was analyzed. Values present mean±SEM; n=5/group. D, Impaired integrin αIIbβ3 activation of Sphk2-deficient platelets. WT (black bars) and Sphk2-deficient (white bars) platelets were left unstimulated (nonstimulated) or induced with adenosine diphosphate (ADP) and α-thrombin, respectively, and binding of an specific antibody to active αIIbß3 was measured by flow cytometry. Shown are the mean values±SD; n=5/group. E, ADP and S1P work synergistically. Whole blood from WT (black bars) or Sphk2-deficient (white bars) mice were induced either with ADP alone or simultaneously with S1P, and aggregation was analyzed. Values present mean±SEM; n=5/group. Statistical significance was calculated using Mann–Whitney rank-sum test.

But how does S1P/S1PR1 control platelet aggregation in response to PAR4 agonists and ADP? PAR4-P dependent irreversible platelet aggregation has been demonstrated to partially rely on subsequent TBXA2 release.31 We observed that Sphk2-deficient platelets as well as platelets treated with the S1PR1 antagonist W146 showed an impaired TBXA2 release following stimulation by α-thrombin (Figures 3B and 6B). Accordingly, we tested whether addition of the stable TBXA2 analogue U46619 could compensate for the diminished aggregation observed in Sphk2-deficient platelets. Indeed, as shown in Figure 6C, addition of 1 μmol/L exogenous U46619 during PAR4 activation completely rescued the aggregation defect of Sphk2-deficient platelets to a level comparable with the normal aggregation response of WT littermates. Thus, our data suggest that impaired TBXA2 release in Sphk2 null platelets leads to defective PAR4-induced platelet aggregation in a process involving S1PR1.

In contrast, TBXA2 generation is unlikely to contribute to enhanced ADP-induced aggregation as blocking of TBXA2 production with cyclooxygenase inhibitors did not affect ADP-induced aggregation in WT platelets (N. Urtz, unpublished data, 2014). Platelet aggregation in response to ADP is mediated by 2 G-protein–coupled receptors, P2Y1 and P2Y12. Since activation of P2Y1 triggers phospholipase C–dependent calcium release from internal stores, we next measured intracellular calcium response in ADP-stimulated Sphk2-deficient platelets. Unexpectedly, we observed no alteration of calcium responses in Sphk2-deficient and control platelets (Online Figure III), indicating that Sphk2/S1P signaling does not act via modulation of calcium mobilization following ADP activation. Both the ADP receptor P2Y12 and the S1PR1 couple to G-protein subunit Gαi. ADP signaling via P2Y12 plays a crucial role in the activation of platelet αIIbβ3 integrin and subsequent platelet aggregation. Hence, we measured αIIbβ3 integrin activation in WT and Sphk2-deficient platelets by flow cytometry in the presence and absence of platelet agonists (Figure 6D). Interestingly, αIIbβ3 integrin activation by ADP stimulation was significantly reduced in Sphk2-deficient compared with WT platelets and likely explains their reduced aggregation after ADP induction (Figure 6A). Finally, we tested whether addition of exogenous S1P could compensate for the diminished aggregation observed in Sphk2-deficient platelets after ADP-induced aggregation. Indeed, as shown in Figure 6E, addition of 10 μmol/L exogenous S1P during ADP activation completely rescued the aggregation defect of Sphk2-deficient platelets to a level comparable with the normal aggregation response of WT littermates.

Together, the above in vitro data have at least 3 important implications: (1) the endogenous Sphk2-dependent S1P release by activated platelets fosters PAR4-induced and ADP-induced aggregation, (2) a defective TBXA2 release as well as a diminished αIIbβ3 integrin activation contributes to impaired aggregation of Sphk2 null platelets in response to PAR4-P and ADP, and (3) S1P facilitates ADP-induced aggregation by supporting αIIbβ3 integrin activation.

Sphk2-Deficiency Reduces Arterial Thrombosis in a FeCl3 Model

Finally, to substantiate the above findings in vivo, we evaluated the role of intrinsic S1P release in platelet aggregation using a mouse arterial thrombosis model. Briefly, we induced arterial thrombosis using ferric chloride (FeCl3) as described27 and monitored platelet aggregation by intravital microscopy for 1 hour. Consistent with our in vitro findings, loss of Sphk2 provided some protection against arterial thrombosis. Although the frequency of occlusive thrombus formation was only moderately (nonsignificantly) affected in Sphk2-deficient mice compared with WT and Sphk1-deficient mice, thrombus stability was severely impaired in Sphk2 null mutants (Figure 7A and 7B). After initial occlusion, arteries from Sphk2-deficient mice rapidly re-established blood flow (within 6 minutes), in contrast to the stable thrombus formation observed in WT (19 minutes) and Sphk1-deficient (13 minutes) mice (Figure 7B). To further evaluate whether these findings are solely platelet-dependent, we reconstituted platelet-depleted WT mice with platelets isolated from either WT or Sphk2-deficient mice (Methods in Online Data Supplement). Although mice transfused with WT platelets showed normal thrombus formation after FeCl3 injury, mice transfused with Sphk2-deficient platelets had a severe reduction in both thrombus incidence, as well as stability (Figure 7C and 7D).

Figure 7.
Figure 7.

Effect of sphingosine kinase 2 (Sphk2) on carotid arterial thrombosis in mice. A, Sphk2 deficiency reduces arterial thrombosis triggered with FeCl3. The Kaplan–Meier curve shows the percent of arteries without occlusion as a function of time after injury; mice per group: n=15 (wild-type, WT), n=10 (Sphk1−/−/Sphk2−/−). Statistical significance was calculated using linear rank test. B, Sphk2 deficiency reduces thrombus stability. Representative intravital microscopy images of thrombus after 12 and 25 minutes are shown in WT, Sphk1-deficient, and Sphk2-deficient mice (left). Bars, 100 μm; dotted white lines mark the vessel walls; white arrows mark a channel with reestablished blood flow in the thrombus. The duration of occlusion indicates the thrombus stability (right). Values present mean±SEM; n=15 (WT), n=10 (Sphk1−/−), or n=10 (Sphk2−/−). C, Arterial thrombosis formation triggered with FeCl3 performed in platelet-depleted WT mice transfused either with WT or with Sphk2-deficient platelets. The Kaplan–Meier curve shows the percent of arteries without occlusion as a function of time after injury; n=5/group. Statistical significance was calculated using linear rank test. D, Sphk2-deficient platelets reduce thrombus stability. The duration of occlusion is indicating the thrombus stability. WT mice transfused with WT or with Sphk2-deficient platelets (plts). Values present mean±SEM; n=5/group. Statistical significance was calculated using Mann–Whitney rank-sum test. E, Tail bleeding time measured after transection of the tail tip of WT and Sphk2-deficient mice. Values present mean±SEM; n=8/group. Statistical significance was calculated using Mann–Whitney rank-sum test.

These findings clearly establish that intrinsic S1P release through platelet Sphk2 controls platelet aggregation and thrombus growth in vivo. Interestingly, Sphk2 deficiency does not affect physiological hemostasis as indicated by an unaltered tail bleeding time in Sphk2-deficient mice (Figure 7E).

Discussion

Platelets are known to produce and release S1P. However, the exact molecular source of platelet S1P and the role of intrinsic S1P generation in platelet function have not been studied in detail. We now report that Sphk2 is the main isoform that generates S1P in platelets and that platelets lacking Sphk2 are unable to secrete S1P into the extracellular space. We show that S1P alone initiates platelet aggregation via the S1PR1 and that this S1P/S1PR1 pathway facilitates PAR4-P and ADP-induced aggregation. In addition, Sphk2 null platelets release lower amounts of TBXA2 and show an impaired activation of αIIbβ3 integrin translating into diminished aggregation in response to stimulation with PAR4-P and ADP. Consistent with the aggregation defect of Sphk2-deficient platelets in vitro, loss of platelet Sphk2 protects partially against arterial thrombosis in vivo without affecting hemostasis.

Sphk2-deficient mice do not show an obvious phenotype and pharmacological Sphk2 inhibitors, which are currently tested for treatment of pancreatic cancer in patients, could thus develop into promising new drug tools relevant for various diseases.32 Therefore, modulating the platelet Sphk2/S1P pathway could represent an interesting novel antithrombotic strategy.

Platelets possess constitutive Sphk activity33 and lack S1P-lyase, an enzyme that irreversibly degrades S1P.14,18 Together, this pattern of enzyme activities leads to the generation and storage of high S1P levels in platelets. Since increased S1P levels are observed in the serum compared with the plasma,16,34 platelets were originally thought to constitute the main rheostat of plasma S1P. However, we show here that loss of Sphk2 does not lead to a reduction in plasma S1P concentration, suggesting that platelet-derived S1P is dispensable for plasma S1P homeostasis. This is in line with several studies demonstrating that red blood cells13,17,18 and endothelial cells19 act as the primary sources of plasma S1P and are responsible for the establishment of the steep blood-to-tissue S1P concentration gradient in vivo.35 In line with this notion we observed that Sphk1 is the predominant source of red blood cell-derived S1P, and that loss of Sphk1 is associated with a reduction in plasma S1P concentrations. Hence, our data provide direct evidence that red blood cells not only store and transport S1P generated elsewhere (eg, by endothelial cells)17,18,20 but also act as potent manufacturers of S1P controlling S1P homeostasis in the blood.

Since platelets are dispensable for homeostasis of S1P in blood plasma, what is the role of platelet-derived S1P? The irrelevance of platelet-derived S1P for plasma S1P homeostasis does not exclude a role of this mediator in the intrinsic control of platelet function. Indeed, we show here that activated platelets generate and release S1P into their local microenvironment. In line with this finding, others have reported previously that platelets generate and release S1P after activation with thrombin, collagen, phorbol ester, and to a lesser extent with ADP,15,22,36–38 suggesting that they likely contribute to the production of local S1P gradients that may have effects on local processes rather than impacting on bulk plasma S1P levels. Platelets are not only essential for primary hemostasis and repair of the endothelium after vascular injury but also act as major cellular players of arterial thrombosis triggering myocardial infarction and stroke. The first step of arterial thrombosis is the adhesion of platelets to exposed extracellular matrix. Our data show that the reduced S1P content of Sphk2-deficient platelets does not influence their ability to adhere to various substrates in a flow chamber model in vitro or in a denudation model of the mouse femoral artery in vivo. Following platelet adhesion, the subsequent steps leading to arterial thrombosis are platelet aggregation and clot formation. Platelet adhesion to immobilized fibrinogen as well as platelet aggregation in the presence of soluble fibrinogen both depend on integrin αIIbβ3. However, while adhesion to fibrinogen can occur without prior αIIbβ3 activation, aggregation largely depends on inside-out integrin activation. This renders aggregation more susceptible to Sphk2-dependent integrin activation and likely explains why aggregation is affected in null mutants, despite normal platelet adhesion to fibrinogen.

The role of S1P for platelet aggregation and coagulation remains controversial. There is evidence that S1P enhances platelet aggregation by inducing platelet shape change through an increase in phospholipase C activity.22,39,40 Other studies have reported that S1P inhibits rather than supports platelet aggregation.41 However, the latter data are based on in vitro experiments using washed platelets or platelet-rich plasma. In our present study, we performed whole blood aggregation assays to mimic physiological conditions as closely as possible. Münzer et al42 recently reported that Sphk1-deficient platelets show elevated degranulation (release of alpha and dense granules) and aggregation in response to low-dose agonist stimulation. Based on their findings, the authors speculated that this was because of reduced levels of intracellular platelet S1P and proposed that platelet S1P acts as a negative regulator of platelet aggregation. However, the authors did not measure platelet S1P levels in Sphk1-deficient platelets in their study. In fact, we show here by liquid-chromatography/mass spectrometry that intracellular S1P levels are not reduced but rather increased in Sphk1-deficient platelets compared with WT controls or Sphk2-deficient mice. Hence, platelet S1P facilitates rather than inhibits platelet activation, explaining the increased aggregation reported in Sphk1-deficient platelets.42

We show that S1P mainly induces a dose-dependent aggregation response via the S1PR1 in whole blood. A recent study showed that aggregation of washed human platelets was reduced not only by inhibition of S1PR1, but also in the presence of the S1PR2 antagonist JTE-01340. However, washing of platelet affects extracellular S1P concentrations, which in turn may have a substantial effect on S1PR surface expression. Hence, the differences between our findings and that obtained by Randriamboavonjy et al are likely explained by the different experimental conditions applied. Notably, we observed no platelet aggregation response when washed platelets were incubated with S1P (N. Urtz, unpublished data, 2009). This is in line with previous data from Yatomi et al,22 who showed that S1P stimulates aggregation only after the addition of exogenous adhesion proteins. In their study, subthreshold concentrations of S1P and weak platelet agonists, such as ADP, synergistically elicit aggregation. Hence, taken together our present findings support the concept that S1P acts as a platelet agonist via S1PR1.

During activation, platelets release or translocate a panoply of >300 proteins and lipids.1,2,43 Among the lipid fraction, AA and its metabolite TBXA2, are known to act as intrinsic agonists, supporting platelet aggregation by amplifying the secretion of platelet constituents from lysosomes, α-granules, and dense granules.3 We show here that both the loss of Sphk2-dependent S1P production by platelets and inhibition of S1PR1 results in defective agonist-induced TBXA2 release by platelets suggesting that S1P intrinsically controls the generation and release of this AA metabolite. Interestingly however, in the absence of additional platelet agonists, S1P alone fails to activate cyclooxygenase 1 and consequently fail to synthesize and release TBXA2, highlighting its role as a supportive rather than potent agonist facilitating platelet activation (N. Urtz, unpublished data, 2014). Importantly, defective release of TBXA2 results in attenuated aggregability of Sphk2-deficient platelets in response to PAR4-P. Notably, it has been shown that blocking TBXA2 formation results in a failure to release S1P from platelets36 suggesting a reciprocal crosstalk of 2 agonistic lipid pathways operating in platelets.

Although our data clearly support the role of secreted S1P in amplifying platelet aggregation, S1P might also act as an intracellular second messenger as previously reported for other cell types.44 We have recently shown that intracellular Sphk2/S1P signaling is required for expression and overall activities of most Src family kinases present in the megakaryocyte/platelet lineage, including Lyn, Fyn, Src, Yes, and Fgr.23 Indeed, Lyn has been demonstrated to regulate TBXA2-synthesis in platelets.45 Thus, reduced Lyn expression/activity might further contribute to the reduced TBXA2 release by Sphk2 null platelets. In summary, our present findings identify a novel agonistic signaling pathway downstream of PAR4 receptors, involving Sphk2-dependent S1P release leading to the release of the potent platelet agonist TBXA2 supporting platelet aggregation.

In line with previous reports, we did not detect TBXA2 generation after ADP stimulation in the presence of physiological calcium concentrations.46 In addition, blocking of TBXA2 synthesis with a cyclooxygenase inhibitor had no impact on ADP-induced aggregation (N. Urtz, unpublished data, 2014). Nevertheless, ADP-induced aggregation of Sphk2-deficient platelets was largely impaired, indicating the existence of additional S1P-dependent regulatory pathways facilitating platelet aggregation independent of TBXA2 release. ADP induces platelet aggregation by activating αIIbβ3 integrin through its receptors P2Y1 (Gαq-coupled) and P2Y12 (Gαi-coupled). P2Y12 receptor signaling highly depends on phosphoinositide-3 kinase activation and is important for sustained activation of αIIbβ3 as well as irreversible platelet aggregation.47 In line with this, we found significantly reduced levels of activated integrin αIIbβ3 in Sphk2-deficient platelets following incubation with ADP. Likewise, S1PR1 is a Gαi-coupled receptor and activation of phosphoinositide-3 kinase has been demonstrated to be an important signaling event downstream of S1PR1, involved in various cellular functions.48 Thus, we speculate that platelet intrinsic S1P might amplify ADP-induced platelet activation via S1PR1 by contributing to phosphoinositide-3 kinase activation and thereby facilitating sustained αIIbβ3 activation. However, we show that Sphk2/S1P-dependent facilitation of ADP-induced platelet aggregation does not involve calcium mobilization. Hence, it is likely that Sphk2/S1P acts via activation of phosphoinositide-3 kinase signaling, which is known to support αIIbβ3 integrin activation and has been reported to act downstream of both P2Y12 and S1PR1.48

The formation and stabilization of a growing hemostatic plug requires the continued full-fledged activation of recruited platelets to prevent those platelets not in direct contact with the wound from loosely associating and eventually detaching from the thrombus. We report here that Sphk2-deficient mice show a reduction in both the extent and duration of occlusion in a FeCl3-induced mouse model of arterial thrombosis in vivo. Several mechanisms are likely to act together accounting for the partial protection of Sphk2-deficient mice from arterial thrombosis, including (1) the loss of a potential direct intrinsic agonistic effect of S1P on platelet aggregation, (2) impaired aggregation capacity after PAR4 activation accompanied by decreased secretion of TBXA2 by Sphk2-deficient platelets and hence the loss of this amplifying signal on newly recruited platelets, and (3) the diminished sensitivity to ADP because of the loss of synergy with S1P-mediated signaling.

In summary, we demonstrate here that the release of S1P by activated platelets into their local micromilieu plays a significant role in regulating important platelet functions including secretion and aggregation in vitro and in vivo. We show that Sphk2 is the Sphk responsible for S1P release by platelets, identifying this enzyme as a new key mediator of platelet homeostasis and potential antithrombotic target. However, since our present findings were obtained in mice, future studies will have to evaluate the relevance of Sphk2/S1P signaling in humans.

Acknowledgments

We thank Raphaela Kutil for her excellent technical assistance.

Sources of Funding

This work was supported by the DFG (German Research Foundation; SFB914, SFB1039, FOR923), the DZHK (German Centre for Cardiovascular Research), the German Cardiac Society and the EU (PRESTIGE).

Disclosures

None.

Footnotes

  • In May 2015, the average time from submission to first decision for all original research papers submitted to Circulation Research was 15.49 days.

  • The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.115.306901/-/DC1.

  • Nonstandard Abbreviations and Acronyms
    AA
    arachidonic acid
    ADP
    adenosine diphosphate
    ATP
    adenosine triphosphate
    PAR4-P
    protease-activated receptor 4-peptide
    S1P
    sphingosine 1-phosphate
    S1PR
    S1P receptor
    Sphk
    sphingosine kinase
    TBXA2
    thromboxane A2
    WT
    wild-type

  • Received May 22, 2015.
  • Revision received June 26, 2015.
  • Accepted June 30, 2015.

References

  1. 1.↵
    1. Holmsen H,
    2. Weiss HJ
    . Secretable storage pools in platelets. Annu Rev Med. 1979;30:119–134. doi: 10.1146/annurev.me.30.020179.001003.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Marcus AJ
    . The role of lipids in platelet function: with particular reference to the arachidonic acid pathway. J Lipid Res. 1978;19:793–826.
    OpenUrlPubMed
  3. 3.↵
    1. FitzGerald GA
    . Mechanisms of platelet activation: thromboxane A2 as an amplifying signal for other agonists. Am J Cardiol. 1991;68:11B–15B.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Engelmann B,
    2. Massberg S
    . Thrombosis as an intravascular effector of innate immunity. Nat Rev Immunol. 2013;13:34–45. doi: 10.1038/nri3345.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Hla T
    . Physiological and pathological actions of sphingosine 1-phosphate. Semin Cell Dev Biol. 2004;15:513–520. doi: 10.1016/j.semcdb.2004.05.002.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. McVerry BJ,
    2. Garcia JG
    . In vitro and in vivo modulation of vascular barrier integrity by sphingosine 1-phosphate: mechanistic insights. Cell Signal. 2005;17:131–139. doi: 10.1016/j.cellsig.2004.08.006.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Cyster JG
    . Chemokines, sphingosine-1-phosphate, and cell migration in secondary lymphoid organs. Annu Rev Immunol. 2005;23:127–159. doi: 10.1146/annurev.immunol.23.021704.115628.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Lee JF,
    2. Zeng Q,
    3. Ozaki H,
    4. Wang L,
    5. Hand AR,
    6. Hla T,
    7. Wang E,
    8. Lee MJ
    . Dual roles of tight junction-associated protein, zonula occludens-1, in sphingosine 1-phosphate-mediated endothelial chemotaxis and barrier integrity. J Biol Chem. 2006;281:29190–29200. doi: 10.1074/jbc.M604310200.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Taha TA,
    2. Mullen TD,
    3. Obeid LM
    . A house divided: ceramide, sphingosine, and sphingosine-1-phosphate in programmed cell death. Biochim Biophys Acta. 2006;1758:2027–2036. doi: 10.1016/j.bbamem.2006.10.018.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Massberg S,
    2. Schaerli P,
    3. Knezevic-Maramica I,
    4. Köllnberger M,
    5. Tubo N,
    6. Moseman EA,
    7. Huff IV,
    8. Junt T,
    9. Wagers AJ,
    10. Mazo IB,
    11. von Andrian UH
    . Immunosurveillance by hematopoietic progenitor cells trafficking through blood, lymph, and peripheral tissues. Cell. 2007;131:994–1008. doi: 10.1016/j.cell.2007.09.047.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Sanchez T,
    2. Hla T
    . Structural and functional characteristics of S1P receptors. J Cell Biochem. 2004;92:913–922. doi: 10.1002/jcb.20127.
    OpenUrlCrossRefPubMed
  12. 12.↵
    1. Olivera A,
    2. Spiegel S
    . Sphingosine kinase: a mediator of vital cellular functions. Prostaglandins. 2001;64:123–134.
    OpenUrlPubMed
  13. 13.↵
    1. Pappu R,
    2. Schwab SR,
    3. Cornelissen I,
    4. Pereira JP,
    5. Regard JB,
    6. Xu Y,
    7. Camerer E,
    8. Zheng YW,
    9. Huang Y,
    10. Cyster JG,
    11. Coughlin SR
    . Promotion of lymphocyte egress into blood and lymph by distinct sources of sphingosine-1-phosphate. Science. 2007;316:295–298. doi: 10.1126/science.1139221.
    OpenUrlAbstract/FREE Full Text
  14. 14.↵
    1. Ikeda M,
    2. Kihara A,
    3. Igarashi Y
    . Sphingosine-1-phosphate lyase SPL is an endoplasmic reticulum-resident, integral membrane protein with the pyridoxal 5’-phosphate binding domain exposed to the cytosol. Biochem Biophys Res Commun. 2004;325:338–343. doi: 10.1016/j.bbrc.2004.10.036.
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Tani M,
    2. Sano T,
    3. Ito M,
    4. Igarashi Y
    . Mechanisms of sphingosine and sphingosine 1-phosphate generation in human platelets. J Lipid Res. 2005;46:2458–2467. doi: 10.1194/jlr.M500268-JLR200.
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    1. Yatomi Y,
    2. Igarashi Y,
    3. Yang L,
    4. Hisano N,
    5. Qi R,
    6. Asazuma N,
    7. Satoh K,
    8. Ozaki Y,
    9. Kume S
    . Sphingosine 1-phosphate, a bioactive sphingolipid abundantly stored in platelets, is a normal constituent of human plasma and serum. J Biochem. 1997;121:969–973.
    OpenUrlAbstract/FREE Full Text
  17. 17.↵
    1. Hänel P,
    2. Andréani P,
    3. Gräler MH
    . Erythrocytes store and release sphingosine 1-phosphate in blood. FASEB J. 2007;21:1202–1209. doi: 10.1096/fj.06-7433com.
    OpenUrlAbstract/FREE Full Text
  18. 18.↵
    1. Ito K,
    2. Anada Y,
    3. Tani M,
    4. Ikeda M,
    5. Sano T,
    6. Kihara A,
    7. Igarashi Y
    . Lack of sphingosine 1-phosphate-degrading enzymes in erythrocytes. Biochem Biophys Res Commun. 2007;357:212–217. doi: 10.1016/j.bbrc.2007.03.123.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Venkataraman K,
    2. Lee YM,
    3. Michaud J,
    4. Thangada S,
    5. Ai Y,
    6. Bonkovsky HL,
    7. Parikh NS,
    8. Habrukowich C,
    9. Hla T
    . Vascular endothelium as a contributor of plasma sphingosine 1-phosphate. Circ Res. 2008;102:669–676. doi: 10.1161/CIRCRESAHA.107.165845.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    1. Bode C,
    2. Sensken SC,
    3. Peest U,
    4. Beutel G,
    5. Thol F,
    6. Levkau B,
    7. Li Z,
    8. Bittman R,
    9. Huang T,
    10. Tölle M,
    11. van der Giet M,
    12. Gräler MH
    . Erythrocytes serve as a reservoir for cellular and extracellular sphingosine 1-phosphate. J Cell Biochem. 2010;109:1232–1243. doi: 10.1002/jcb.22507.
    OpenUrlPubMed
  21. 21.↵
    1. Zemann B,
    2. Kinzel B,
    3. Müller M,
    4. Reuschel R,
    5. Mechtcheriakova D,
    6. Urtz N,
    7. Bornancin F,
    8. Baumruker T,
    9. Billich A
    . Sphingosine kinase type 2 is essential for lymphopenia induced by the immunomodulatory drug FTY720. Blood. 2006;107:1454–1458. doi: 10.1182/blood-2005-07-2628.
    OpenUrlAbstract/FREE Full Text
  22. 22.↵
    1. Yatomi Y,
    2. Ruan F,
    3. Hakomori S,
    4. Igarashi Y
    . Sphingosine-1-phosphate: a platelet-activating sphingolipid released from agonist-stimulated human platelets. Blood. 1995;86:193–202.
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Zhang L,
    2. Urtz N,
    3. Gaertner F,
    4. Legate KR,
    5. Petzold T,
    6. Lorenz M,
    7. Mazharian A,
    8. Watson SP,
    9. Massberg S
    . Sphingosine kinase 2 (Sphk2) regulates platelet biogenesis by providing intracellular sphingosine 1-phosphate (S1P). Blood. 2013;122:791–802. doi: 10.1182/blood-2012-12-473884.
    OpenUrlAbstract/FREE Full Text
  24. 24.↵
    1. Allende ML,
    2. Sasaki T,
    3. Kawai H,
    4. Olivera A,
    5. Mi Y,
    6. van Echten-Deckert G,
    7. Hajdu R,
    8. Rosenbach M,
    9. Keohane CA,
    10. Mandala S,
    11. Spiegel S,
    12. Proia RL
    . Mice deficient in sphingosine kinase 1 are rendered lymphopenic by FTY720. J Biol Chem. 2004;279:52487–52492. doi: 10.1074/jbc.M406512200.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    1. Linke B,
    2. Schreiber Y,
    3. Zhang DD,
    4. Pierre S,
    5. Coste O,
    6. Henke M,
    7. Suo J,
    8. Fuchs J,
    9. Angioni C,
    10. Ferreiros-Bouzas N,
    11. Geisslinger G,
    12. Scholich K
    . Analysis of sphingolipid and prostaglandin synthesis during zymosan-induced inflammation. Prostaglandins Other Lipid Mediat. 2012;99:15–23. doi: 10.1016/j.prostaglandins.2012.06.002.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Massberg S,
    2. Brand K,
    3. Grüner S,
    4. Page S,
    5. Müller E,
    6. Müller I,
    7. Bergmeier W,
    8. Richter T,
    9. Lorenz M,
    10. Konrad I,
    11. Nieswandt B,
    12. Gawaz M
    . A critical role of platelet adhesion in the initiation of atherosclerotic lesion formation. J Exp Med. 2002;196:887–896.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Massberg S,
    2. Gawaz M,
    3. Grüner S,
    4. Schulte V,
    5. Konrad I,
    6. Zohlnhöfer D,
    7. Heinzmann U,
    8. Nieswandt B
    . A crucial role of glycoprotein VI for platelet recruitment to the injured arterial wall in vivo. J Exp Med. 2003;197:41–49.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Yatomi Y,
    2. Yamamura S,
    3. Ruan F,
    4. Igarashi Y
    . Sphingosine 1-phosphate induces platelet activation through an extracellular action and shares a platelet surface receptor with lysophosphatidic acid. J Biol Chem. 1997;272:5291–5297.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Zhang L,
    2. Orban M,
    3. Lorenz M,
    4. et al
    . A novel role of sphingosine 1-phosphate receptor S1pr1 in mouse thrombopoiesis. J Exp Med. 2012;209:2165–2181. doi: 10.1084/jem.20121090.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Morowski M,
    2. Vögtle T,
    3. Kraft P,
    4. Kleinschnitz C,
    5. Stoll G,
    6. Nieswandt B
    . Only severe thrombocytopenia results in bleeding and defective thrombus formation in mice. Blood. 2013;121:4938–4947. doi: 10.1182/blood-2012-10-461459.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Wu CC,
    2. Hwang TL,
    3. Liao CH,
    4. Kuo SC,
    5. Lee FY,
    6. Teng CM
    . The role of PAR4 in thrombin-induced thromboxane production in human platelets. Thromb Haemost. 2003;90:299–308. doi: 10.1267/THRO03020299.
    OpenUrlPubMed
  32. 32.↵
    1. Kunkel GT,
    2. Maceyka M,
    3. Milstien S,
    4. Spiegel S
    . Targeting the sphingosine-1-phosphate axis in cancer, inflammation and beyond. Nat Rev Drug Discov. 2013;12:688–702. doi: 10.1038/nrd4099.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Banno Y,
    2. Kato M,
    3. Hara A,
    4. Nozawa Y
    . Evidence for the presence of multiple forms of Sph kinase in human platelets. Biochem J. 1998;335 (Pt 2):301–304.
    OpenUrl
  34. 34.↵
    1. Yatomi Y,
    2. Ozaki Y,
    3. Ohmori T,
    4. Igarashi Y
    . Sphingosine 1-phosphate: synthesis and release. Prostaglandins. 2001;64:107–122.
    OpenUrlPubMed
  35. 35.↵
    1. Schwab SR,
    2. Cyster JG
    . Finding a way out: lymphocyte egress from lymphoid organs. Nat Immunol. 2007;8:1295–1301. doi: 10.1038/ni1545.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. Ulrych T,
    2. Böhm A,
    3. Polzin A,
    4. Daum G,
    5. Nüsing RM,
    6. Geisslinger G,
    7. Hohlfeld T,
    8. Schrör K,
    9. Rauch BH
    . Release of sphingosine-1-phosphate from human platelets is dependent on thromboxane formation. J Thromb Haemost. 2011;9:790–798. doi: 10.1111/j.1538-7836.2011.04194.x.
    OpenUrlCrossRefPubMed
  37. 37.↵
    1. Rink TJ,
    2. Sanchez A,
    3. Hallam TJ
    . Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. Nature. 1983;305:317–319.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Jonnalagadda D,
    2. Sunkara M,
    3. Morris AJ,
    4. Whiteheart SW
    . Granule-mediated release of sphingosine-1-phosphate by activated platelets. Biochim Biophys Acta. 2014;1841:1581–1589. doi: 10.1016/j.bbalip.2014.08.013.
    OpenUrl
  39. 39.↵
    1. Hashizume T,
    2. Sato T,
    3. Fujii T
    . Phosphatidic acid with medium-length fatty acyl chains synergistically stimulates phospholipase C with Ca2+ in rabbit platelets. Biochem Int. 1992;26:491–497.
    OpenUrlPubMed
  40. 40.↵
    1. Randriamboavonjy V,
    2. Badenhoop K,
    3. Schmidt H,
    4. Geisslinger G,
    5. Fisslthaler B,
    6. Fleming I
    . The S1P(2) receptor expressed in human platelets is linked to the RhoA-Rho kinase pathway and is down regulated in type 2 diabetes. Basic Res Cardiol. 2009;104:333–340. doi: 10.1007/s00395-008-0769-1.
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Nugent D,
    2. Xu Y
    . Sphingosine-1-phosphate: characterization of its inhibition of platelet aggregation. Platelets. 2000;11:226–232.
    OpenUrlCrossRefPubMed
  42. 42.↵
    1. Münzer P,
    2. Schmid E,
    3. Walker B,
    4. Fotinos A,
    5. Chatterjee M,
    6. Rath D,
    7. Vogel S,
    8. Hoffmann SM,
    9. Metzger K,
    10. Seizer P,
    11. Geisler T,
    12. Gawaz M,
    13. Borst O,
    14. Lang F
    . Sphingosine kinase 1 (Sphk1) negatively regulates platelet activation and thrombus formation. Am J Physiol Cell Physiol. 2014;307:C920–C927. doi: 10.1152/ajpcell.00029.2014.
    OpenUrlAbstract/FREE Full Text
  43. 43.↵
    1. Siess W
    . Molecular mechanisms of platelet activation. Physiol Rev. 1989;69:58–178.
    OpenUrlFREE Full Text
  44. 44.↵
    1. Urtz N,
    2. Olivera A,
    3. Bofill-Cardona E,
    4. Csonga R,
    5. Billich A,
    6. Mechtcheriakova D,
    7. Bornancin F,
    8. Woisetschläger M,
    9. Rivera J,
    10. Baumruker T
    . Early activation of sphingosine kinase in mast cells and recruitment to FcepsilonRI are mediated by its interaction with Lyn kinase. Mol Cell Biol. 2004;24:8765–8777. doi: 10.1128/MCB.24.19.8765-8777.2004.
    OpenUrlAbstract/FREE Full Text
  45. 45.↵
    1. Cho MJ,
    2. Pestina TI,
    3. Steward SA,
    4. Lowell CA,
    5. Jackson CW,
    6. Gartner TK
    . Role of the Src family kinase Lyn in TxA2 production, adenosine diphosphate secretion, Akt phosphorylation, and irreversible aggregation in platelets stimulated with gamma-thrombin. Blood. 2002;99:2442–2447.
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    1. Cattaneo M,
    2. Gachet C,
    3. Cazenave JP,
    4. Packham MA
    . Adenosine diphosphate(ADP) does not induce thromboxane A2 generation in human platelets. Blood. 2002;99:3868–9; author reply 3869.
    OpenUrlFREE Full Text
  47. 47.↵
    1. Kauffenstein G,
    2. Bergmeier W,
    3. Eckly A,
    4. Ohlmann P,
    5. Léon C,
    6. Cazenave JP,
    7. Nieswandt B,
    8. Gachet C
    . The P2Y(12) receptor induces platelet aggregation through weak activation of the alpha(IIb)beta(3) integrin–a phosphoinositide 3-kinase-dependent mechanism. FEBS Lett. 2001;505:281–290.
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Rakhit S,
    2. Conway AM,
    3. Tate R,
    4. Bower T,
    5. Pyne NJ,
    6. Pyne S
    . Sphingosine 1-phosphate stimulation of the p42/p44 mitogen-activated protein kinase pathway in airway smooth muscle. Role of endothelial differentiation gene 1, c-Src tyrosine kinase and phosphoinositide 3-kinase. Biochem J. 1999;338 (Pt 3):643–649.
    OpenUrl

Novelty and Significance

What Is Known?

  • Two sphingosine kinase (Sphk) isoforms produce sphingosine 1-phosphate (S1P) which is a physiological component of human and mouse plasma.

  • Blood platelets generate, store, and release S1P.

What New Information Does This Article Contribute?

  • The Sphk isoform Sphk2 is the major source of S1P in murine platelets.

  • Platelet-intrinsic S1P critically regulates platelet function.

  • Loss of Sphk2 in platelets protects from arterial thrombosis without affecting hemostasis in vivo.

Platelet aggregation is an essential step during thrombus formation and is regulated by the concerted action of platelet agonists. Although it is known that platelets produce and secrete the bioactive lipid metabolite S1P on stimulation, the exact molecular source of platelet S1P and its role in platelet function have not been studied in detail. We identified Sphk2 as the major isoform responsible for S1P production in platelets and demonstrated its pivotal intrinsic role during platelet activation. S1P released by activated platelets triggered platelet aggregation via the S1P-receptor 1, thereby critically amplifying the sustained aggregatory response of classical platelet agonists in vitro. Consequently, Sphk2-deficient mice developed unstable thrombi after vascular injury in vivo. While protecting mice from arterial thrombosis Sphk2 deficiency did not affect physiological hemostasis translating in normal bleeding times in Sphk2-deficient animals. Modulating the platelet Sphk2/S1P pathway could therefore represent an interesting novel antithrombotic strategy.

View Abstract

This Issue

Jump to

Article Tools

  • Print
  • Citation Tools
  • Article Alerts
  • Save to my folders

Share this Article

  • Email
  • Sphingosine 1-Phosphate Produced by Sphingosine Kinase 2 Intrinsically Controls Platelet Aggregation In Vitro and In VivoNovelty and Significance
    Nicole Urtz, Florian Gaertner, Marie-Luise von Bruehl, Sue Chandraratne, Faridun Rahimi, Lin Zhang, Mathias Orban, Verena Barocke, Johannes Beil, Irene Schubert, Michael Lorenz, Kyle R. Legate, Andrea Huwiler, Josef M. Pfeilschifter, Christian Beerli, David Ledieu, Elke Persohn, Andreas Billich, Thomas Baumruker, Michael Mederos y Schnitzler and Steffen Massberg
    Circulation Research. 2015;117:376-387, originally published June 30, 2015
    https://doi.org/10.1161/CIRCRESAHA.115.306901
    del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Related Articles

Cited By...

Subjects