Sphingosine Kinase-Dependent Activation of Endothelial Nitric Oxide Synthase by Angiotensin II

【摘要】  Objective- In addition to their role in programmed cell death, cell survival, and cell growth, sphingolipid metabolites such as ceramide, sphingosine, and sphingosine-1-phosphate have vasoactive properties. Besides their occurrence in blood, they can also be formed locally in the vascular wall itself in response to external stimuli. This study was performed to investigate whether vasoactive compounds modulate sphingolipid metabolism in the vascular wall and how this might contribute to the vascular responses.

Methods and Results- In isolated rat carotid arteries, the contractile responses to angiotensin II are enhanced by the sphingosine kinase inhibitor dimethylsphingosine. Endothelium removal or NO synthase inhibition by N -nitro- L -arginine results in a similar enhancement. Angiotensin II concentration-dependently induces NO production in an endothelial cell line, which can be diminished by dimethylsphingosine. Using immunoblotting and intracellular calcium measurements, we demonstrate that this sphingosine kinase-dependent endothelial NO synthase activation is mediated via both phosphatidylinositol 3-kinase/Akt and calcium-dependent pathways.

Conclusions- Angiotensin II induces a sphingosine kinase-dependent activation of endothelial NO synthase, which partially counteracts the contractile responses in isolated artery preparations. This pathway may be of importance under pathological circumstances with reduced NO bioavailability. Moreover, a disturbed sphingolipid metabolism in the vascular wall may lead to reduced NO bioavailability and endothelial dysfunction.

Sphingolipid metabolites can be formed locally in the vascular wall. AT 1 receptor stimulation by angiotensin II induces a sphingosine kinase-dependent endothelial NO synthase activation via phosphatidylinositol 3-kinase/Akt and calcium-dependent pathways. This may be of importance during, or alternatively under pathological circumstances leading to, reduced NO bioavailability and endothelial dysfunction.

【关键词】  sphingosine kinase sphingosinephosphate angiotensins nitric oxide synthase vasoconstriction

Introduction

Sphingolipids such as sphingomyelin are a major constituent of cellular plasma membranes. Various stimuli activate enzymes involved in the sphingolipid metabolism. Sphingomyelinase catalyzes the hydrolysis of sphingomyelin to form ceramide. 1,2 The sequential action of ceramidase and sphingosine kinase converts ceramide to sphingosine and sphingosine-1-phosphate (S1P), and ceramide synthase and S1P phosphatase can reverse this process to form ceramide from S1P. 3,4 The sphingomyelin metabolites ceramide, sphingosine, and S1P are biologically active mediators that play important roles in cellular homeostasis. In this regard, ceramide and sphingosine on the one hand and S1P on the other hand frequently have opposite biological effects. For example, ceramide and sphingosine are generally involved in apoptotic responses to various stress stimuli and in growth arrest, 5,6 whereas S1P is implicated in mitogenesis, differentiation, and migration. 7,8 This homeostatic system is frequently referred to as the ceramide/S1P rheostat. 9 It can be hypothesized that this rheostat also plays a role in vascular contraction and relaxation because S1P, sphingosine, and ceramide are potentially counteracting, vasoactive compounds. 10,11

The molecular basis of ceramide effects has not been explored fully but is believed to involve stress-activated protein kinases, protein phosphatases such as protein phosphatases 1 and 2, guanylyl cyclase, and charybdotoxin-sensitive K + channels. 11,12 The molecular basis of S1P effects has been characterized in more detail. S1P can act on specific G protein-coupled receptors, of which 5 subtypes have been identified thus far, termed S1P 1-5. These receptors couple to intracellular second messenger systems including intracellular Ca 2+, adenylyl cyclase, phospholipase C, phosphatidylinositol 3 (PI3)-kinase, protein kinase Akt, mitogen-activated protein kinases, and Rho- and Ras-dependent pathways. 13 The cardiovascular system primarily expresses the receptor subtypes S1P 1-3, and within the vasculature they are expressed in both vascular smooth muscle and endothelial cells. 14 S1P can cause elevation of intracellular Ca 2+ in both cell types, 10,15,16 which is likely to be the basis of contractile effects in smooth muscle but can also cause smooth muscle relaxation via activation of endothelial NO synthase (eNOS) and subsequent production of NO. 17

Physiologically, the vascular wall is exposed to S1P as a constituent of HDLs 18,19 or on its release by activated platelets. 20 The experimental addition of exogenous ceramide or S1P imitates this. However, studies in several cell types and tissues demonstrate that various stimuli can elicit local ceramide and S1P formation, which then act in an autocrine or paracrine manner. 1,21-23 Therefore, it was the aim of the present study to determine whether known vasoconstrictive compounds may exert their vascular effects at least in part by modulating the ceramide/S1P rheostat. For this purpose, we have used the specific sphingosine kinase inhibitor dimethylsphingosine (DMS) 24 to block S1P formation. Using this approach, we show that the important vasoactive modulator angiotensin II (Ang II) exerts its effects on isolated rat carotid arteries at least partly via the ceramide/S1P rheostat in the endothelium. This may be of importance in further understanding the underlying pathophysiology of various vascular diseases associated with endothelial dysfunction, such as atherosclerosis and hypertension.

Methods

For contraction experiments, we used rat carotid arteries. For all other experiments (DAF-2DA NO assay, [Ca 2+ ] i measurements, immunoblotting, and real-time quantitative polymerase chain reaction), the bEnd.3 endothelial cell line was used. For detailed methods, please see the online-only supplement, available at http://atvb.ahajournals.org.

Results

Effect of Sphingosine Kinase Inhibition on Vascular Contraction

In the contraction experiments, the mean normalized diameter of a total number of 82 carotid artery preparations was 1028±8 µm. The maximum contraction evoked by KCl (100 mmol/L) amounted to 4.0±0.5 mN/mm segment length, and there was no significant difference in KCl-induced maximal contractile force between the compared groups. In endothelium-denuded preparations, KCl responses amounted to 2.6±0.2 mN/mm segment length. DMS (10 µmol/L) and VPC 23019 (10 µmol/L) had no influence on the pretension of the preparations. Preincubation of the vessels with DMS (10 µmol/L) had no significant effect on the potency or efficacy of KCl or phenylephrine ( Figure 1A and 1 B). However, DMS induced a leftward shift of the concentration-response curve for Ang II (pEC 50 9.11±0.05 versus 8.57±0.04 for control; n=7 to 8) without significantly affecting the efficacy ( Figure 1 C). To directly compare the results with and without endothelial denudation, data in supplemental Figure IA (available online at http://atvb. ahajournals.org) are normalized to the contractile response obtained by the third 100 mmol/L KCl. Preincubating the vessel with the NOS inhibitor N -nitro- L -arginine (L-NNA) (100 µmol/L) mimicked the effect of DMS on Ang II-induced contraction (pEC 50 9.17±0.20), although there was a more substantial increase in E max (102.2±3.7% versus 78.4±1.7% for control; n=7). More importantly, there was no additional effect of DMS when applied simultaneously with L-NNA. Removal of the endothelium resulted in an effect similar to that observed for the Ang II-induced contraction in the presence of L-NNA (supplemental Figure IA). Preincubation of the vessel with the S1P 1 /S1P 3 receptor antagonist VPC 23019 (10 µmol/L) resulted in a significant increase in E max (3.20±0.26 versus 2.53±0.13 mN/mm for control; n=6) and a small, although not significant, leftward shift of the curve for Ang II (supplemental Figure IB). The AT 2 receptor antagonist PD123319 (10 µmol/L) did not show any effect (data not shown).

Figure 1. Contractile responses in the isolated rat carotid artery for KCl (A), phenylephrine (PhE) (B), and Ang II (C) in the presence of vehicle (DMSO) or DMS (10 µmol/L). Contractile force is presented as mN/mm segment length. DMS or vehicle was added to the organ bath 30 minutes before the construction of the concentration-response curve for indicated agonists. Values are given as mean±SEM (n=4 to 8).

Role of Sphingosine Kinase in Ang II-Induced NO Release In Vitro

Ang II concentration-dependently increased NO production in the bEnd.3 cell line ( Figure 2 ). DMS and VPC 23019 had no effect on basal NO production (1.00±0.10 [n=10] and 0.98±0.07 [n=6], respectively). Preincubation of the cells with 10 µmol/L DMS or 10 µmol/L VPC 23019 inhibited Ang II-induced NO production to approximately basal level. L-NNA 100 µmol/L further diminished NO production. As a positive control, Ca 2+ ionophore A23187 (2.5 µmol/L) induced an NO response of &2.5-fold of basal, which was not significantly influenced by DMS ( Figure 2 ). The 1 -adrenoreceptor agonist phenylephrine did not induce NO production in bEnd.3 cells (data not shown).

Figure 2. NO formation measured directly in bEnd.3 endothelial cells with the use of the specific fluorescent NO probe DAF-2DA. After loading, cells were preincubated with DMS (10 µmol/L), L-NNA (100 µmol/L), VPC 23019 (10 µmol/L), vehicle (DMSO, distilled water, and DMSO, respectively), or none. Afterward, cells were stimulated with the positive control Ca 2+ ionophore A23187 (2.5 µmol/L), Ang II (1, 10, and 100 nmol/L), or vehicle (DMSO and distilled water, respectively). Values are calculated with the use of the mean increase in fluorescence, measured every 2 minutes over a period of 70 minutes. NO levels are expressed as fold of basal and mean±SEM (n=6 to 19). * P <0.05. Note the differential right y axis for Ca 2+ ionophore A23187 data.

Effects of Sphingosine Kinase Inhibition on [Ca 2+ ] i Changes

Ang II concentration-dependently increased [Ca 2+ ] i in the bEnd.3 cell line. Preincubation of the cells with 10 µmol/L DMS prevented the Ang II-induced Ca 2+ increase completely. The Ang II-induced Ca 2+ release was also inhibited by the AT 1 receptor blocker telmisartan (10 nmol/L) but not by 100 nmol/L PD123319, an AT 2 receptor-specific antagonist. Preincubation with 10 µmol/L DMS did not influence the Ca 2+ ionophore A23187 (2.5 µmol/L)-induced increase in [Ca 2+ ] i ( Figure 3 ).

Figure 3. Intracellular Ca 2+ measurements in bEnd.3 cells. After loading with fluo-4 AM, cells were preincubated with DMS (10 µmol/L), the AT 1 receptor antagonist telmisartan (10 nmol/L), the AT 2 receptor antagonist PD123319 (100 nmol/L), vehicle (DMSO, distilled water, and distilled water, respectively), or none. Cells were then stimulated with Ang II (1, 10, and 100 nmol/L), Ca 2+ ionophore A23187 (2.5 µmol/L), or vehicle under constant measuring of fluorescence. With the use of Triton and EGTA, the maximal and minimal fluorescent responses were determined, and changes in intracellular Ca 2+ concentrations ( [Ca 2+ ] i ) were calculated. Ca 2+ levels are expressed in nmol/L and are mean±SEM (n=4 to 9). * P <0.05. Note the differential right y axis for Ca 2+ ionophore A23187 data.

Role of Akt in Ang II-Induced eNOS Activation

To investigate the role of the PI3-kinase/Akt pathway in Ang II-induced sphingosine kinase activity and subsequent eNOS activation, we stimulated bEnd.3 cells with 100 nmol/L Ang II or 20 ng/mL vascular endothelial growth factor (VEGF) in either the presence or absence of 10 µmol/L DMS or the PI3-kinase inhibitor wortmannin (200 nmol/L). In a pilot study we investigated the time dependency of Ang II-induced and VEGF-induced (as a positive control) 25 phosphorylation of Akt and eNOS. This revealed that the maximal phosphorylation occurred at a time point of 2.5 minutes. Ang II (100 nmol/L) induced Akt phosphorylation to an extent similar to that of VEGF, which was inhibited by DMS. DMS had no influence on basal level of Akt or eNOS phosphorylation (data not shown). Ang II (100 nmol/L) induced eNOS phosphorylation, which was also inhibited by DMS. The PI3-kinase inhibitor wortmannin abolished both Akt and eNOS phosphorylation. As a loading control, the bands for the antibody directed against the general protein -tubulin are shown ( Figure 4 ).

Figure 4. Ang II-mediated Akt and eNOS phosphorylation. bEnd.3 cells were stimulated with Ang II or VEGF for 2.5 minutes with or without preincubation with DMS, wortmannin (WM), or vehicle (DMSO for both) for 30 minutes. Protein extracts were analyzed for phospho-Ser 473 -Akt (pAkt) (A) and phospho-Ser 1177 -eNOS (peNOS) (B) by Western blotting. Loading controls for -tubulin content are shown. All results are representative of 4 experiments. Densitometric analysis of blots is shown, with the phosphorylation of vehicle-treated cells arbitrarily set to 100%.

Expression of S1P Receptor and Sphingosine Kinase Subtypes in bEnd.

3 Cells

The rank order of expression of S1P receptor subtypes in the bEnd.3 cell line, based on the raw Ct values from real-time polymerase chain reaction from 3 independent experiments, was as follows: S1P 1 (29.2±0.6) S1P 2 S1P 4 (34.8±0.5), with S1P 3 and S1P 5 not detectable. SphK2 (29.9±1.0) was expressed higher than SphK1 (34.9±0.5). In comparison, the Ct values for the housekeeping genes HPRT1 and GAPDH were 29.4±0.7 and 21.7±0.5, respectively.

Discussion

S1P, sphingosine, and ceramide are interconvertible sphingolipids that have important effects on cellular homeostasis. S1P has been shown to induce cell growth and survival, 7,8 whereas ceramide and sphingosine, the metabolic precursors of S1P, have been shown to induce apoptosis and growth arrest. 5,6 Accordingly, the dynamic balance between ceramide and sphingosine versus S1P, referred to as the ceramide/S1P rheostat, is thought to be an important determinant of cell fate. 9 We hypothesized that this rheostat may play a role in vascular contraction and relaxation because S1P, sphingosine, and ceramide are potentially counteracting, vasoactive compounds. 10,11 S1P and ceramide, when applied exogenously or administered in vivo, can have differential effects that may be dependent on the type of vascular bed, species, and/or method used to study vascular contraction and relaxation (eg, in vivo, ex vivo, wire myograph, cannulated vessels). It is still unknown whether physiologically relevant vasoactive factors make use of the rheostat by activating 1 or more of the aforementioned key enzymes to exert their vasoactive effects. Therefore, we investigated the role of the rheostat in agonist-induced vascular responses by inhibition of sphingosine kinase rather than by applying sphingolipids exogenously.

Here we show that the presence of the specific competitive sphingosine kinase inhibitor DMS substantially potentiated the Ang II-induced contractile effect. In contrast, the contractile effects of the 1 -adrenoceptor agonist phenylephrine or receptor-independent constriction by KCl were unaffected. Early reports state that DMS may act as a protein kinase C (PKC) inhibitor in vitro 26,27; however, Edsall et al 24 have shown that DMS is a specific sphingosine kinase inhibitor in cellular systems at concentrations up to 50 µmol/L. A PKC-independent action of DMS in monocytes, 10 µmol/L, was reported recently by Lee et al. 28 This is in concurrence with our finding that the PKC inhibitor calphostin C (100 nmol/L) did not affect the Ang II-induced contraction (data not shown). Moreover, when DMS would be a PKC inhibitor in our system, one would, if anything, expect an opposite response (ie, a rightward shift of the concentration-response curve for Ang II and phenylephrine) because PKC activation can be involved in smooth muscle cell contraction. Finally, the fact that the concentration-response curves for phenylephrine and KCl are not influenced by DMS supports a specific effect on sphingosine kinase rather than a nonspecific effect on PKC.

The leftward shift of the concentration-response curve for Ang II implies that endogenous S1P, the formation of which is inhibited by DMS, has vasodilatory properties or that ceramide or sphingosine (which may accumulate) have contractile properties in our system. Because NO is the major relaxing factor throughout the vasculature, we investigated whether the leftward shift of the Ang II curve by sphingosine kinase inhibition is attributable to a decrease in NOS activation. Preincubation with the NOS inhibitor L-NNA or removal of the endothelium indeed leads to a similar leftward shift of the concentration-response curve for Ang II. More importantly, DMS in the presence of L-NNA did not further influence the concentration-response curve for Ang II, suggesting that a decreased activation of NOS might indeed mediate the leftward shift of the Ang II concentration-response curve in the presence of DMS. This implies that Ang II under normal circumstances induces NO production, a phenomenon that also has been shown by others. 29,30 The fact that L-NNA, in contrast to DMS, also increases the E max of Ang II might be attributable to inhibition of basal NO production by L-NNA ( Figure 2 ). NO production by Ang II has been attributed to both AT 1 and AT 2 receptor stimulation. The lack of effect of the specific AT 2 antagonist PD123319 in the present study indicates that the Ang II-induced NO production is due to AT 1 receptor stimulation, which is in accordance with the findings of Boulanger et al. 31 To show that indeed the Ang II-induced NO production is inhibited by DMS, we measured NO formation directly in cultured vascular endothelial cells. The bEnd.3 endothelial cell line is known to express relatively high levels of eNOS and therefore is highly suitable to investigate relatively small alterations in eNOS activity. 32,33 Ang II induced a concentration-dependent increase in NO production in the bEnd.3 cell line that could be completely inhibited by DMS and L-NNA. In these experiments, DMS had no influence on the NO production induced by Ca 2+ ionophore A23187, indicating that DMS had no nonspecific influences in this assay. These findings suggest that either Ang II-induced S1P production leads to activation of eNOS or that ceramide and/or sphingosine inhibits eNOS activity. The former explanation is not unlikely because it has been demonstrated before that S1P can lead to NO formation through increased eNOS activity in the endothelium, which can be mediated via both intracellular Ca 2+ mobilization and phosphorylation of Akt and eNOS. 34,35

To test the involvement of Ca 2+ elevation in Ang II-induced eNOS activation via endogenous S1P formation, we measured Ang II-induced changes in [Ca 2+ ] i in the bEnd.3 cells. [Ca 2+ ] i was modestly elevated in bEnd.3 cells after stimulation with Ang II, in a concentration-dependent manner. This rise in [Ca 2+ ] i could be inhibited by DMS, whereas the changes in [Ca 2+ ] i caused by the receptor-independent influx of Ca 2+ by the Ca 2+ ionophore A23187 were not affected by DMS, indicating that DMS has no a-specific effect in this assay. The fact that the Ca 2+ response for Ang II was inhibited by telmisartan but not PD123319 demonstrates again an AT 1 receptor-mediated effect.

The second major pathway leading to increased eNOS activity is via phosphorylation of Akt and eNOS. Ser 1177 phosphorylation of eNOS by Akt (which can be activated by PI3-kinase) increases the sensitivity of eNOS for the Ca 2+ /calmodulin complex by &10 to 15 times and is therefore an important mechanism underlying increased NO production. Both exogenously applied S1P 17,35,36 and Ang II receptor activation 37,38 have been shown to induce Akt and eNOS phosphorylation in cultured endothelial cells. In the present study, Ang II rapidly (within 2.5 minutes) induced phosphorylation of Akt and eNOS that could be inhibited by DMS. Wortmannin, a specific inhibitor of PI3-kinase, also inhibited phosphorylation of Akt and eNOS induced by Ang II. Therefore, it seems that sphingosine kinase activity is important not only for the mobilization of intracellular Ca 2+ but also for the PI3-kinase/Akt pathway in the Ang II-induced activation of eNOS. The latter finding points toward a receptor-mediated phenomenon, and stimulation of both S1P 1 and S1P 3 receptors has been reported to result in increased NO formation via the PI3-kinase/Akt pathway in cultured endothelial cells. 36,39 This indicates that it is most likely S1P that increases eNOS activity via 1 or more types of S1P receptors expressed in the endothelium. Interestingly, a similar signaling mechanism has been shown recently for tumor necrosis factor- -induced eNOS activation in endothelial cells. In this report, the authors showed that silencing S1P 1 and/or S1P 3 receptors by means of siRNA prevents eNOS activation by tumor necrosis factor-. 40 To investigate whether S1P 1 and S1P 3 receptors are involved in the Ang II-induced NO production, we tested whether the novel S1P 1 /S1P 3 receptor antagonist VPC 23019 also augments the contractile effects of Ang II in the rat carotid artery, as seen for DMS and L-NNA. Indeed, VPC 23019, one of the few available S1P receptor antagonists, induced a significant increase in E max and a small, although not significant, leftward shift of the concentration-response curve for Ang II. Moreover, VPC 23019 also inhibited the Ang II-induced production of NO in the bEnd.3 cell line. These data indeed may point toward involvement of S1P receptors, but S1P receptor-independent mechanisms cannot be excluded. A similar sphingosine kinase-dependent formation of NO has recently been shown for the vasodilatory action of acetylcholine, although these effects appeared not to be mediated by S1P receptors. 41 To further investigate the role of S1P receptors, receptor subtypes, or putative intracellular targets, genetic models can be used. With the use of S1P 3 knockout mice, for example, it was recently shown that HDLs, known to carry S1P, and the immunomodulator and S1P receptor agonist FTY720 induce an endothelium- and NO-dependent vasorelaxation via the S1P 3 receptor in vitro and ex vivo. 34,42

Taken together, these data suggest that activation of the endothelial AT 1 receptor by Ang II leads to a modulation of the sphingolipid metabolism, resulting in increased NO production. This is most likely the result of increased sphingosine kinase activity leading to increased production of S1P that subsequently stimulates (an) endothelial S1P receptor(s). Via activation of the PI3-kinase/Akt pathway and Ca 2+ mobilization, eNOS activity is increased, and the resulting NO formation counteracts the Ang II- induced smooth muscle cell contraction ( Figure 5 ). This counteracting effect may be of importance under pathological circumstances with reduced bioavailability of NO such as atherosclerosis and hypertension. Moreover, disturbed regulation of the ceramide/S1P rheostat (eg, reduced sphingosine kinase activity) may be another mechanism leading to reduced NO bioavailability and endothelial dysfunction

Figure 5. Overview of the suggested role of the ceramide/S1P rheostat during Ang II-induced vascular contraction. AT 1 receptor activation in the endothelial cell leads to endogenous formation of S1P via activation of sphingosine kinase (SphK). This subsequently leads to activation of eNOS involving both release of intracellular Ca 2+ and phosphorylation of Akt and eNOS via the PI3-kinase pathway. The resulting formation of NO has a counterbalancing effect on the Ang II-induced contraction in vascular smooth muscle.

Acknowledgments

Disclosures

None.

【参考文献】
  Hannun YA, Luberto C, Argraves KM. Enzymes of sphingolipid metabolism: from modular to integrative signaling. Biochemistry. 2001; 40: 4893-4903.

Kolesnick R. The therapeutic potential of modulating the ceramide/sphingomyelin pathway. J Clin Invest. 2002; 110: 3-8.

Maceyka M, Payne SG, Milstien S, Spiegel S. Sphingosine kinase, sphingosine-1-phosphate, and apoptosis. Biochim Biophys Acta. 2002; 1585: 193-201.

Spiegel S, Milstien S. Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol. 2003; 4: 397-407.

Chatterjee S. Sphingolipids in atherosclerosis and vascular biology. Arterioscler Thromb Vasc Biol. 1998; 18: 1523-1533.

Gulbins E. Regulation of death receptor signaling and apoptosis by ceramide. Pharmacol Res. 2003; 47: 393-399.

Pyne S, Pyne NJ. Sphingosine 1-phosphate signalling in mammalian cells. Biochem J. 2000; 349: 385-402.

Saba JD, Hla T. Point-counterpoint of sphingosine 1-phosphate metabolism. Circ Res. 2004; 94: 724-734.

Cuvillier O, Pirianov G, Kleuser B, Vanek PG, Coso OA, Gutkind S, Spiegel S. Suppression of ceramide-mediated programmed cell death by sphingosine-1-phosphate. Nature. 1996; 381: 800-803.

Bischoff A, Czyborra P, Fetscher C, Meyer zu Heringdorf D, Jakobs KH, Michel MC. Sphingosine-1-phosphate and sphingosylphosphorylcholine constrict renal and mesenteric microvessels in vitro. Br J Pharmacol. 2000; 130: 1871-1877.

Czyborra P, Saxe M, Fetscher C, Meyer Zu HD, Herzig S, Jakobs KH, Michel MC, Bischoff A. Transient relaxation of rat mesenteric microvessels by ceramides. Br J Pharmacol. 2002; 135: 417-426.

Ruvolo PP. Intracellular signal transduction pathways activated by ceramide and its metabolites. Pharmacol Res. 2003; 47: 383-392.

Sanchez T, Hla T. Structural and functional characteristics of S1P receptors. J Cell Biochem. 2004; 92: 913-922.

Alewijnse AE, Peters SL, Michel MC. Cardiovascular effects of sphingosine-1-phosphate and other sphingomyelin metabolites. Br J Pharmacol. 2004; 143: 666-684.

Coussin F, Scott RH, Wise A, Nixon GF. Comparison of sphingosine 1-phosphate-induced intracellular signaling pathways in vascular smooth muscles: differential role in vasoconstriction. Circ Res. 2002; 91: 151-157.

Mehta D, Konstantoulaki M, Ahmmed GU, Malik AB. Sphingosine 1-phosphate-induced mobilization of intracellular Ca 2+ mediates rac activation and adherens junction assembly in endothelial cells. J Biol Chem. 2005; 280: 17320-17328.

Dantas AP, Igarashi J, Michel T. Sphingosine 1-phosphate and control of vascular tone. Am J Physiol. 2003; 284: H2045-H2052.

Kimura T, Sato K, Kuwabara A, Tomura H, Ishiwara M, Kobayashi I, Ui M, Okajima F. Sphingosine 1-phosphate may be a major component of plasma lipoproteins responsible for the cytoprotective actions in human umbilical vein endothelial cells. J Biol Chem. 2001; 276: 31780-31785.

Okajima F. Plasma lipoproteins behave as carriers of extracellular sphingosine 1-phosphate: is this an atherogenic mediator or an anti-atherogenic mediator? Biochim Biophys Acta. 2002; 1582: 132-137.

Yatomi Y, Igarashi Y, Yang L, Hisano N, Qi R, Asazuma N, Satoh K, Ozaki Y, Kume S. Sphingosine 1-phosphate, a bioactive sphingolipid abundantly stored in platelets, is a normal constituent of human plasma and serum. J Biochem (Tokyo). 1997; 121: 969-973.

Florio T, Arena S, Pattarozzi A, Thellung S, Corsaro A, Villa V, Massa A, Diana F, Spoto G, Forcella S, Damonte G, Filocamo M, Benatti U, Schettini G. Basic fibroblast growth factor activates endothelial nitric-oxide synthase in CHO-K1 cells via the activation of ceramide synthesis. Mol Pharmacol. 2003; 63: 297-310.

Payne SG, Brindley DN, Guilbert LJ. Epidermal growth factor inhibits ceramide-induced apoptosis and lowers ceramide levels in primary placental trophoblasts. J Cell Physiol. 1999; 180: 263-270. <a href="/cgi/external_ref?access_num=10.1002/(SICI)1097-4652(199908)180:2

Xu CB, Zhang Y, Stenman E, Edvinsson L. D-Erythro-N,N-dimethylsphingosine inhibits bFGF-induced proliferation of cerebral, aortic and coronary smooth muscle cells. Atherosclerosis. 2002; 164: 237-243.

Edsall LC, Van Brocklyn JR, Cuvillier O, Kleuser B, Spiegel S. N,N-Dimethylsphingosine is a potent competitive inhibitor of sphingosine kinase but not of protein kinase C: modulation of cellular levels of sphingosine 1-phosphate and ceramide. Biochemistry. 1998; 37: 12892-12898.

Zachary I, Gliki G. Signaling transduction mechanisms mediating biological actions of the vascular endothelial growth factor family. Cardiovasc Res. 2001; 49: 568-581.

Khan WA, Dobrowsky R, el Touny S, Hannun YA. Protein kinase C and platelet inhibition by D-erythro-sphingosine: comparison with N,N-dimethylsphingosine and commercial preparation. Biochem Biophys Res Commun. 1990; 172: 683-691.

Igarashi Y, Hakomori S. Enzymatic synthesis of N,N-dimethyl-sphingosine: demonstration of the sphingosine: N-methyltransferase in mouse brain. Biochem Biophys Res Commun. 1989; 164: 1411-1416.

Lee EH, Lee YK, Im YJ, Kim JH, Okajima F, Im DS. Dimethylsphingosine regulates intracellular pH and Ca 2+ in human monocytes. J Pharmacol Sci. 2006; 100: 289-296.

Patzak A, Lai EY, Mrowka R, Steege A, Persson PB, Persson AE. AT 1 receptors mediate angiotensin II-induced release of nitric oxide in afferent arterioles. Kidney Int. 2004; 66: 1949-1958.

Siragy HM, Carey RM. The subtype 2 (AT 2 ) angiotensin receptor mediates renal production of nitric oxide in conscious rats. J Clin Invest. 1997; 100: 264-269.

Boulanger CM, Caputo L, Lévy BI. Endothelial AT 1 -mediated release of nitric oxide decreases angiotensin II contractions in rat carotid artery. Hypertension. 1995; 26: 752-757.

Ghigo D, Arese M, Todde R, Vecchi A, Silvagno F, Costamagna C, Dong QG, Alessio M, Heller R, Soldi R. Middle T antigen-transformed endothelial cells exhibit an increased activity of nitric oxide synthase. J Exp Med. 1995; 181: 9-19.

Govers R, Bevers L, de Bree P, Rabelink TJ. Endothelial nitric oxide synthase activity is linked to its presence at cell-cell contacts. Biochem J. 2002; 361: 193-201.

Nofer JR, van der GM, Tölle M, Wolinska I, von Wnuck LK, Baba HA, Tietge UJ, Gödecke A, Ishii I, Kleuser B, Schäfers M, Fobker M, Zidek W, Assmann G, Chun J, Levkau B. HDL induces NO-dependent vasorelaxation via the lysophospholipid receptor S1P 3. J Clin Invest. 2004; 113: 569-581.

Igarashi J, Bernier SG, Michel T. Sphingosine 1-phosphate and activation of endothelial nitric-oxide synthase: differential regulation of Akt and MAP kinase pathways by EDG and bradykinin receptors in vascular endothelial cells. J Biol Chem. 2001; 276: 12420-12426.

Gonzalez E, Kou R, Michel T. Rac1 modulates sphingosine-1-phosphate-mediated activation of phosphoinositide 3-kinase/Akt signaling pathways in vascular endothelial cells. J Biol Chem. 2006; 281: 3210-3216.

Bayraktutan U. Effects of angiotensin II on nitric oxide generation in growing and resting rat aortic endothelial cells. J Hypertens. 2003; 21: 2093-2101.

Dugourd C, Gervais M, Corvol P, Monnot C. Akt is a major downstream target of PI3-kinase involved in angiotensin II-induced proliferation. Hypertension. 2003; 41: 882-890.

Waeber C, Blondeau N, Salomone S. Vascular sphingosine-1-phosphate S1P 1 and S1P 3 receptors. Drug News Perspect. 2004; 17: 365-382.

De Palma C, Meacci E, Perrotta C, Bruni P, Clementi E. Endothelial nitric oxide synthase activation by tumor necrosis factor alpha through neutral sphingomyelinase 2, sphingosine kinase 1, and sphingosine 1 phosphate receptors: a novel pathway relevant to the pathophysiology of endothelium. Arterioscler Thromb Vasc Biol. 2006; 26: 99-105.

Roviezzo F, Bucci M, Delisle C, Brancaleone V, Di Lorenzo A, Mayo IP, Fiorucci S, Fontana A, Gratton JP, Cirino G. Essential requirement for sphingosine kinase activity in eNOS-dependent NO release and vasorelaxation. FASEB J. 2006; 20: 340-342.

Tölle M, Levkau B, Keul P, Brinkmann V, Giebing G, Schönfelder G, Schäfers M, von Wnuck LK, Jankowski J, Jankowski V, Chun J, Zidek W, van der Giet M. Immunomodulator FTY720 induces eNOS-dependent arterial vasodilatation via the lysophospholipid receptor S1P 3. Circ Res. 2005; 96: 913-920.

作者单位：Department of Pharmacology and Pharmacotherapy, University of Amsterdam, Academic Medical Center, Amsterdam, Netherlands.