Comparison of Sphingosine 1-Phosphate–Induced Intracellular Signaling Pathways in Vascular Smooth Muscles
Differential Role in Vasoconstriction
Sphingosine 1-phosphate (S1P), a lipid released from activated platelets, influences physiological processes in the cardiovascular system via activation of the endothelial differentiation gene (EDG/S1P) family of 7 transmembrane G protein–coupled receptors. In cultured vascular smooth muscle (VSM) cells, S1P signaling has been shown to stimulate proliferative responses; however, its role in vasoconstriction has not been examined. In the present study, the effects of S1P and EDG/S1P receptor expression were determined in rat VSM from cerebral artery and aorta. S1P induced constriction of cerebral artery, which was partly dependent on activation of p160ROCK (Rho-kinase). S1P also induced activation of RhoA in cerebral artery with a similar time course to contraction. In aorta, S1P did not produce a constriction or RhoA activation. In VSM myocytes from cerebral arteries, stimulation with S1P gives rise to a global increase in [Ca2+]i, initially generated via Ca2+ release from the sarcoplasmic reticulum by an inositol 1,4,5-trisphosphate–dependent pathway. In aorta VSM, a small increase in [Ca2+]i was observed after stimulation at higher concentrations of S1P. S1P induced activation of p42/p44mapk in aorta and cerebral artery VSM. Subtype-specific S1P receptor antibodies revealed that the expression of S1P3/EDG-3 and S1P2/EDG-5 receptors is 4-fold higher in cerebral artery compared with aorta. S1P1/EDG-1 receptor expression was similar in both types of VSM. Therefore, the ability of S1P to act as a vasoactive mediator is dependent on the activation of associated signaling pathways and may vary in different VSM. This differential signaling may be related to the expression of S1P receptor subtypes.
Several studies have now shown that the bioactive sphingolipid, sphingosine 1-phosphate (S1P), is likely to play an important role in regulating cellular processes via activation of specific signal transduction pathways.1–3⇓⇓ In the short-term, S1P-induced activation of several intracellular signaling pathways occurs, such as an increase in the intracellular calcium concentration ([Ca2+]i),4 activation of the monomeric GTP-binding protein, p21RhoA (RhoA),5 and activation of p42/p44 mitogen-activated protein kinase (p42/p44mapk).6,7⇓ Through activation of these signaling pathways, S1P can induce longer term effects, such as proliferation, differentiation, and cell migration.1
It is now clear that S1P-induced intracellular effects occur predominantly through activation of selective S1P receptors on the plasma membrane.2 These receptors, belonging to the G protein–coupled receptor superfamily, originally known as EDG (endothelial differentiation gene) receptors.8 Several isoforms have now been cloned and S1P1/EDG-1,5 S1P2/EDG-5,9 S1P3/EDG-3,10 S1P4/EDG-6,11 and S1P5/EDG-812 have high affinities for S1P with EC50s in the nmol/L range. Recent studies have investigated the delineation of activation for S1P-induced transduction pathways with specific S1P receptor subtypes.13,14⇓ In transfected cell lines, it has been demonstrated that increases in [Ca2+]i are partly inhibited by pertussis toxin, suggesting involvement of Gi,13 although this effect can also occur via activation of Gq14 and production of inositol 1,4,5-trisphosphate (InsP3). Cell lines overexpressing S1P3 receptors, compared with S1P1, have a more pronounced increase in [Ca2+]i.13,15⇓ As with other receptors that couple to heterotrimeric G proteins, S1P receptors also activate the monomeric G protein, p21RhoA (RhoA)5 possibly via G12.16 A recent study with S1P3 receptor knockout mice has demonstrated that, whereas S1P-induced InsP3 production was significantly decreased in embryonic fibroblast cells, RhoA activation was unaffected.17 Through these in vitro studies, an understanding of selective S1P receptor signaling is emerging.
Recent studies have indicated the importance of S1P in the cardiovascular system, particularly its essential role in vascular development18 and potential role in cardiovascular diseases.19,20⇓ S1P is synthesized and stored by resting platelets and, after platelet activation, is released in high concentrations (nmol/L to μmol/L range).21 Both endothelial cells and VSM cells can be exposed to significant levels of S1P in vivo. In primary cultured VSM cells, S1P has a proliferative effect.19,20⇓ In addition to these mitogenic effects, release of S1P from activated platelets may also modulate more rapid events, such as vasoconstriction. Intravenous injections of S1P results in decreased blood flow22 and contraction in renal, mesenteric, and basilar arteries.23,24⇓ The expression of S1P receptors and the signal transduction pathways activated by S1P in freshly isolated blood vessels have not yet been determined.
In the present study, we have compared 2 different types of smooth muscle: rat cerebral artery and rat thoracic aorta. Freshly isolated rat cerebral arteries express S1P1, S1P2, and S1P3 receptors. S1P induces a release of Ca2+ from intracellular stores and subsequent contraction. S1P can also activate RhoA in rat cerebral artery with a similar time course. In contrast, freshly isolated rat aorta expresses lower levels of S1P2 and S1P3 receptors compared with cerebral artery. In these cells, S1P at a physiological concentration range produces a negligible increase in [Ca2+]i and does not activate RhoA. These results suggest an important role for S1P in regulating contractility of selective blood vessels after platelet activation. Evidence suggests that this select activation of signaling pathways may be partly dependent on the relative expression of specific S1P receptor subtypes.
Materials and Methods
Tissue Preparation and Tension Measurement
Sprague-Dawley rats (≈250 g; Harlan Olac, Oxford, UK) were maintained in accordance with Home Office (UK) regulations and were euthanized by inhalation of CO2 followed by cervical dislocation. Thoracic aorta and cerebral arteries (middle and basilar) were placed into ice-cold HEPES-buffered Krebs solution. Arteries were cleaned of connective tissue, and the endothelium was removed by gentle rubbing of the lumen. Endothelium-denuded arterial rings were attached to a force transducer (AE801, SensoNor AS). After determining contractile responses to 143 mmol/L K+, rings were stimulated with either S1P or ET-1. In some experiments, the rings were incubated with 5 μmol/L S1P to obtain a stable plateau contraction before Y-27632 was applied.
Enzymatic Dissociation of VSM Myocytes
Smooth muscle cells from cerebral arteries and aorta were isolated by enzymatic dissociation. Briefly, tissues were transferred to a Ca2+-free Hanks solution consisting of (in mmol/L) 55 NaCl, 80 Na glutamate, 5.6 KCl, 2 MgCl2, 10 glucose, and 10 HEPES (pH 7.4). Arteries were incubated for 20 minutes in solution containing 0.3 mg/mL papain (Sigma), 1 mg/mL dithioerythritol (Sigma), and 1 mg/mL BSA. Tissues were further incubated for 5 minutes in Hanks solution (37°C) supplemented with 1 mg/mL BSA, 0.8 mg/mL collagenase (Sigma), and 100 μmol/L CaCl2. After several washes, tissues were triturated. Isolated smooth muscle cells were stored on glass-bottom dishes maintained at 4°C and used within 6 hours. Only cells that had an elongated morphology were used for imaging.
Imaging of [Ca2+]i
Individual VSM myocytes were loaded with 2 μmol/L Fura-2 AM for 30 minutes in a solution containing (in mmol/L) 130 NaCl, 5.6 KCl, 1 MgCl2, 1.7 CaCl2, 11 Glucose, and 10 HEPES (pH 7.4) followed by a 20 minute de-esterification period. A Zeiss Axiovert 200 inverted microscope, equipped with a cooled CCD camera (Photometrics) and a polychromatic illumination system (T.I.L.L. Photonics), was used to capture fluorescence images with excitations at 340 and 380 nm. The ratio of the fluorescence intensity between the pair of frames (FR340/380) was calculated after background subtraction. The MetaFluor 4.6 software (Universal Imaging Corporation) controlled the illuminator and camera, and performed image ratioing and analysis. Results are expressed as F340/380 ratio. Experiments were performed at room temperature (22 to 24°C).
The whole-cell mode of the patch clamp technique was used to record currents evoked by S1P in smooth muscle cells from cerebral arteries. Patch pipettes (resistances, 4 to 8 MΩ) were filled with solution containing, in mmol/L (140 KCl, 0.1 CaCl2, 2 MgCl2, 10 HEPES, 1.1 EGTA, 2 ATP; pH 7.2). Whole-cell currents were measured with an Axoclamp 2A (Axon Instruments Inc) and recorded at a membrane potential of −30 mV. S1P was externally applied by low-pressure ejection about 50 to 100 μm away from the cell being recorded. Currents were captured and stored on digital audio tape using a digital tape recorder (DTR 1200, Biologic). Data were analyzed using a Tandon computer (Cambridge Electronic Design software, version 6.0).
Frozen tissue was added to lysis buffer and homogenized at 4°C in a Braun homogenizing vessel as previously described.25 Protein was measured using a Lowry assay (Biorad) to ensure equal protein loading. In addition, samples were checked using Coomassie Blue–stained gels, and membranes were stained with Ponceau Red to confirm protein loading. Whole-cell homogenates were used for immunoblotting with Gα-specific antibodies (anti-Gαq, anti-Gα12, and anti-Gα13, Santa Cruz Biotechnology) and anti-p42/p44mapk antibodies (phospho- and pan-) (New England Biolabs). For S1P receptor immunoblots, homogenates were centrifuged at 100 000g for 1 hour and the supernatant discarded. Proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to a nitrocellulose membrane (Biorad) as previously described.25 The membranes were immunoblotted with primary antibodies followed by detection with horseradish peroxidase–conjugated secondary antibodies. The immunoreactive bands were visualized using enhanced chemiluminescence and quantitated with an imaging densitometer (Biorad GS-690). Only blots that were entirely nonsaturated pixels (as determined using Multianalyst software, Biorad) were quantitated to ensure linearity of densitometric analysis.
Monoclonal anti-S1P3/EDG-3 and anti-S1P2/EDG-5 receptor antibodies were obtained commercially from Exalpha Biologicals and characterized previously. They do not cross-react with S1P2/EDG-5 or S1P3/EDG-3, respectively, or with LPA1/EDG-2, LPA2/EDG-4, or S1P4/EDG-6. HEK cell overexpressing S1P1/EDG-1 receptor also did not show any reactivity with the S1P3/EDG-3 or S1P2/EDG-5 receptor antibodies (see the online data supplement available at http://www.circresaha.org). In addition, S1P3/EDG-3 receptor antibodies did not show any reactivity in HEK cells overexpressing S1P1/EDG-1 or S1P2/EDG-5. S1P2/EDG-5 receptor antibodies also did not show any reactivity in HEK cells overexpressing S1P1/EDG-1 or S1P3/EDG-3 (see online data supplement). Purified polyclonal anti-S1P1/EDG-1 receptor antiserum was raised in rabbit against the COOH-terminal (peptide sequence KDEGDNPETIMSSGNVNSSS) of the human S1P1/EDG-1 receptor protein. This sequence is specific to the S1P1/EDG-1 receptor. The S1P1/EDG-1 antibody did not show any reactivity against HEK cells overexpressing S1P3/EDG-3 or S1P2/EDG-5 receptors.
Rho Activation Assay
The Sepharose-bound GST-Rhotekin-Rho Binding Domain (RBD) was used (Upstate Biotechnology). Tissues were dissected and denuded as described. In order to obtain sufficient tissue, the cerebral arteries from 5 rats were pooled for each sample. Three aortae were used per sample. Cerebral arteries and thoracic aorta were stimulated for 5 or 10 minutes with either S1P (10 μmol/L) or ET-1 (100 nmol/L) and quickly frozen in liquid nitrogen. Frozen tissues were homogenized in buffer containing 50 mmol/L Tris.HCl (pH 7.2), 1% Triton X-100, 0.5% deoxycholic acid, 0.1% SDS, 500 mmol/L NaCl, 10 mmol/L MgCl2, 20 μg/mL each of leupeptin and aprotinin, and 1 mmol/L phenylmethylsulfonyl fluoride (PMSF). Homogenates were clarified by centrifugation and incubated with the Sepharose-bound GST-Rhotekin RBD for 45 minutes at 4°C. Beads were washed 3 times and GTP-RhoA bound to beads was eluted by boiling in Laemmli’s SDS buffer. Proteins from each extract were separated by SDS-PAGE and transferred to nitrocellulose membrane, followed by immunoblotting with a specific mouse monoclonal anti-RhoA antibody (Santa Cruz Biotechnology) and a horseradish peroxidase-conjugated secondary antibody. Densities of bands corresponding to precipitated RhoA (GTP-RhoA) and total RhoA were quantitated with an imaging densitometer (Biorad GS-690). The amount of GTP-RhoA was normalized for the total amount of RhoA present in each sample.
Analysis of Data
Data are expressed as mean±SEM. Significance was tested by means of Student’s t test. A value of P<0.05 was considered significant.
S1P-Induced Intracellular Ca2+ Increase in VSM Myocytes From Cerebral Artery and Aorta
Transient intracellular Ca2+ increases were induced by extracellular application of S1P in myocytes from cerebral artery. Typically, the transients observed in single myocytes from cerebral artery exhibited a rapid increase in [Ca2+]i followed by a slower decline (Figure 1A). The mean amplitude of the peak evoked by 1 μmol/L S1P was 0.45±0.01 ratio units (n=109). In contrast to cerebral artery, stimulation of aortic myocytes with S1P produced only a small Ca2+ increase (Figure 1A) at higher S1P concentrations (maximum 30 μmol/L 0.13±0.03 ratio units; n=32). In both VSM types, 100 nmol/L ET-1 evoked transient Ca2+ increases (Figure 1B). Amplitude of the ET-1 induced Ca2+ responses was similar in the two types of VSM (Figure 1C). Identical intracellular Ca2+ increases were induced in cerebral arteries and aorta after stimulation with 10 mmol/L caffeine (data not shown). S1P-induced increases in [Ca2+]i with an EC50 of 5 nmol/L (Figure 1D). The EC50 for aorta was not calculated as it was in the nonphysiological range and a maximum could not be achieved.
As shown in Figure 2A, Ca2+ mobilization after stimulation with 1 μmol/L S1P propagates throughout the entire cerebral artery myocyte producing contraction. S1P produced Ca2+ oscillations in approximately 50% of the cerebral artery myocytes (not shown). The S1P-stimulated Ca2+ transient observed in cerebral arteries was concentration-dependent and saturable, indicative of a receptor-dependent response. The effects of thapsigargin in the generation of calcium responses to S1P were investigated. Such treatment almost completely abolished the Ca2+ increases evoked by 1 μmol/L S1P (Figure 2B). Removal of external Ca2+ (in the presence of 0.5 mmol/L EGTA) or application of 2 μmol/L nifedipine only slightly reduced the transient peak (Figure 2B). These results indicate that the peak increases in [Ca2+]i evoked by S1P are due to calcium mobilization from intracellular stores. As S1P may stimulate the InsP3 pathway, we used the phospholipase C (PLC) inhibitor U73122 and its inactive analogue U73343. Twenty minutes preincubation of cerebral artery myocytes with U73122 (1 μmol/L) greatly decreased the responsiveness to 1 μmol/L S1P, whereas the inactive analogue, U73343, had no effect on S1P-induced increases in [Ca2+]i (Figure 2B). U73122 alone did not affect the resting Ca2+ level nor the caffeine-induced Ca2+ responses (data not shown), suggesting that it blocked the phosphatidylinositol-specific PLC without depleting the intracellular Ca2+ store. To further determine the Ca2+ signaling cascade evoked by S1P, we determined the effects of ryanodine, an inhibitor of the ryanodine receptor (RyR)/Ca2+-induced Ca2+-release (CICR). Perfusion with 10 μmol/L ryanodine for 30 minutes reduced approximately 60% the amplitude of S1P-induced Ca2+ release (Figure 2B).
S1P-Induced Ca2+-Activated Currents
In voltage-clamped cerebral artery myocytes (−30 mV), extracellular application of 1 μmol/L S1P activated outward currents (Figure 2C). These had a mean amplitude of 550±140 pA (n=8) with a reversal potential of −74 mV. In some cells (n=3), a fast transient inward current preceded the larger outward current and had a mean peak amplitude of −590±160 pA (not shown). These currents were inhibited by thapsigargin suggesting that S1P mobilizes Ca2+ from stores and subsequently activates Ca2+-dependent conductances.
S1P Induced Contraction of Cerebral Artery but not of Aorta
The effects of S1P on contraction were examined in denuded cerebral arteries and denuded aortic rings. Exposure of cerebral arteries to 5 μmol/L S1P produced a contractile response followed by prolonged plateau (Figure 3A). The S1P-evoked increase of tension developed slowly compared with KCl (143 mmol/L) or ET-1 (100 nmol/L)–induced contractions (data not shown) and corresponded to 35±5% (n=5) of the ET-1-induced contraction. In contrast, S1P did not change tension in aortic rings (Figure 3A). In cerebral artery myocytes, the time course of the S1P-induced rise in [Ca2+]i was substantially quicker than the prolonged time course of contraction in cerebral artery rings (Figure 3B).
Y-27632 (10 μmol/L) was used to assess the involvement of Rho-kinase in S1P-induced contraction of cerebral artery. Cerebral artery was stimulated with 5 μmol/L S1P and maximum tension achieved. After the addition of Y-27632, the tension decreased to baseline (Figures 3C and 3D).
Stimulation with 10 μmol/L S1P increased the level of GTP-RhoA expressed as a proportion of total RhoA (Figure 3E). The association of the GST-Rhotekin RBD with GTP-RhoA is linearly proportional in smooth muscle homogenates as previously published.26 The increase in GTP-RhoA was 4-fold after 5 minutes of stimulation with S1P. Stimulation with ET-1 in cerebral arteries also resulted in a 5-fold increase, compared with controls, in the proportion of GTP-bound RhoA to total RhoA. In aorta, a 10-minute incubation with S1P did not produce a significant increase in GTP-bound RhoA (1.5-fold). However, ET-1 (100 nmol/L) induced a 5-fold increase in GTP-bound RhoA.
Expression of S1P Receptor Subtypes, Gαq and Gα12/13, in Cerebral Arteries and Aorta VSM
The expression pattern of S1P receptors was examined using specific receptor subtype antibodies. All 3 subtypes examined were expressed in aorta and cerebral artery tissues (Figure 4). Densitometric measurements showed that the level of S1P1 protein expression was similar in both tissues. However, the expression of both S1P2 and S1P3 receptors was increased 4-fold in the cerebral arteries compared with aorta (Figure 4). The relative expression of Gαq, Gα12, and Gα13 was also assessed. Gαq was slightly greater in cerebral arteries compared with aorta (1.4-fold; Figure 4). Expression of Gα12 and Gα13 was also slightly greater in cerebral artery compared with aorta (1.6-fold).
Cerebral artery and aorta were stimulated with 5 μmol/L S1P for 15 and 30 minutes and the activation of p42/p44mapk was determined (Figure 5). p42/p44mapk was stimulated 4-fold after 15 minutes and 5-fold after 30 minutes following incubation with S1P in aorta. In cerebral artery, p42/p44mapk was increased 3-fold after 15 minutes and 4-fold after 30 minutes compared with the total p42/p44mapk present.
Recent studies suggest that S1P exerts diverse physiological actions by activating S1P receptors in cultured VSM cells.18–20⇓⇓ However, the relative contribution of S1P in regulating vascular tone, and the intracellular mechanisms involved, has not yet been addressed. The aim of this study was to determine the effects of S1P on 2 types of freshly isolated vascular smooth muscle. The results of our studies demonstrate that S1P can act as a vasoactive substance in cerebral artery but not in aorta. S1P can produce a global increase of [Ca2+]i by an InsP3-mediated mechanism with a subsequent increase in tension in cerebral artery. S1P stimulation also resulted in an increase in GTP-bound RhoA, indicating RhoA activation, in a time course compatible with a role in cerebral artery vasoconstriction. Inhibition of the downstream effector of RhoA, Rho-kinase, results in a significant reversal in the tension produced by S1P. In aorta, similar concentrations of S1P do not appear to activate RhoA. The differential effects of S1P in different blood vessels may be due to alterations in upstream intracellular signaling proteins or membrane receptors. Although expression of Gα12/13 and Gαq were similar in the 2 types of VSM, the relative expression of S1P receptor subtypes was significantly different. Cerebral arteries expressed S1P1, S1P2, and S1P3 receptor subtypes. Aorta also expressed these 3 subtypes; however, the expression of S1P2 and S1P3 was significantly lower compared with cerebral artery.
Although this study is the first to determine S1P receptor expression in fully differentiated VSM, the relative expression of the S1P receptor subtypes in cerebral artery smooth muscle compared with aorta smooth muscle suggest that different vascular beds may express varying levels of S1P receptor subtypes. As pharmacological tools are not yet available to dissect out roles for each subtype, a direct comparison of expression with function in isolated tissue, such as VSM, is difficult to address. In cell lines it has been established that experimental manipulation of the relative expression of S1P receptor isoforms, for example by overexpression, has functional consequences for specific S1P-induced intracellular effects.5,13,15⇓⇓ This is also the case in cultured aortic VSM cells.20 This study now suggests that in fully differentiated smooth muscle, a differential expression of S1P receptor subtypes may contribute to variations of the cellular responses to S1P in vivo. In rat aorta VSM, the lower relative expression (compared with cerebral artery) of S1P2 and S1P3 correlates with a lack of vasoconstriction and a greatly reduced S1P-induced global increase in [Ca2+]i. Additionally, aorta VSM stimulated with S1P did not produce an activation of RhoA. (Because S1P1, S1P2, and S1P3 receptors are present in aorta, some activation of RhoA may occur although the GST-rhotekin RBD “pulldown” assay utilized in this study would not detect such small increases). In cerebral artery, compared with aorta, the coupling of a greater S1P-induced increase in [Ca2+]i and higher expression level (compared with aorta) of S1P2 and S1P3 agrees with several previous studies in various cell lines demonstrating a direct correlation of S1P3 expression in particular and S1P-induced Ca2+ release.13–15,17⇓⇓⇓ In support of these functional data, S1P receptor subtypes, S1P2 and S1P3, can couple to Gαq13,15–17⇓⇓⇓ linked to phospholipase C activation, whereas S1P1 couples relatively selectively to Gαi.16,27⇓ Results from our study would also suggest that RhoA activation occurs predominantly via S1P2 and S1P3 receptors. Both S1P2 and S1P3 (but not S1P1) can couple to Gα12,16 an upstream effector of RhoA activation.28 In aorta, the expression of S1P receptors but the lack of S1P-induced vasoconstriction suggest that S1P may have other physiological effects in this VSM. S1P-induced stimulation of p42/p44mapk was observed in aorta as well as in cerebral arteries. As activation of MAPK is associated with a proliferative response, S1P may be involved in the initiation of VSM proliferation as suggested in cultured aortic VSM cells,19,20⇓ possibly via S1P1/Gαi-coupled mechanism.27
In cerebral VSM, S1P produced concentration-dependent increases in [Ca2+]i with an EC50 value similar to that of recombinant S1P receptors overexpressed in oocytes.14 The S1P-induced increase in [Ca2+]i is probably the result of Ca2+ release from intracellular thapsigargin-sensitive stores because it fails to occur after inhibition of the sarcoplasmic reticulum Ca2+-ATPase. Moreover, removal of external Ca2+ or inhibition of voltage-gated Ca2+ channels had no significant effect on the S1P-induced Ca2+ response. U73122 completely abolished the [Ca2+]i response to S1P, indicating S1P-induced Ca2+ transients were mediated by PLC activation and subsequent activation of InsP3-gated Ca2+ release channels. Ryanodine decreased the S1P-induced increase in [Ca2+]i, suggesting VSM cells use combined activation of InsP3 receptors and of RyR/Ca2+-induced Ca2+ release channels to release and propagate global increases in intracellular Ca2+. The ability of InsP3 pathway to recruit RyR to generate Ca2+ waves have been previously suggested in cerebral VSM.29 The global increase in [Ca2+]i is further shown by electrophysiological techniques. After S1P application, a global Ca2+ rise presumably activates a large Ca2+-sensitive K+ current,30 promoting repolarization of the cell membrane. This would also negatively regulate contractility, ie, lead toward relaxation. However, in this case the predominating effect is maintained contractility, probably via activation of the RhoA pathway. For comparison, the effects of another vasoconstrictor, ET-1, on the [Ca2+]i in isolated VSM myocytes from rat cerebral artery and aorta was determined. These were similar in both types of VSM. The characteristics of the increase in [Ca2+]i in response to a maximal concentration of ET-1 in cerebral artery was similar to that observed for S1P.
Smooth muscle contractility is regulated by the Ca2+/calmodulin-dependent phosphorylation of the myosin regulatory light chains (MLCs). Increases in MLC phosphorylation can also occur at constant Ca2+ concentration through inhibition of the MLC phosphatase (ie, Ca2+ sensitization).31 It is now recognized that the p160 Rho-associated protein kinase (Rho-kinase), which is activated by RhoA, can phosphorylate the MLC phosphatase, inhibiting its activity.32 The ability of S1P to stimulate activation of RhoA in cerebral artery smooth muscle has important implications for its potential role as a vasoconstrictor. The prolonged contractile response obtained with S1P is not due to a sustained [Ca2+]i and is likely due to RhoA activation. Rho-kinase inhibition reverses this prolonged contractile response33 to S1P in agreement with a study in canine basilar arteries,24 further demonstrating the involvement of the RhoA/Rho-kinase pathway in S1P-induced vasoconstriction. The ET-1–induced contraction in cerebral arteries was also decreased by Rho-kinase inhibition, as previously reported in other smooth muscles.34 Despite similar Ca2+ release characteristics evoked by S1P and ET-1 in cerebral artery VSM, the contractile response of these vasoconstrictor agonists in whole arteries is different. The prolonged profile of the S1P-induced contraction is similar to that observed previously24 and corresponds to its potential role in cerebral vasospasm, typified by a prolonged vasoconstriction. The ET-1–induced contraction is greater (60% to 70%) than the response to a maximal concentration of S1P. The reason for this difference is not yet clear but may be related to relative receptor efficacy.
In conclusion, this study has demonstrated that physiologically relevant concentrations of S1P can act as a vasoconstrictor in cerebral arteries, but not in aorta. Selectivity of this action may be the result of differential expression of S1P receptor isoforms, previously undetermined in native vascular smooth muscle. Therefore, in vivo, partial occlusion of the cerebral circulation by thrombus formation could result in the local release of S1P from platelets, producing a local vasoconstrictor response. This could further compromise cerebral blood flow and contribute to cerebral ischemia.
This study was funded by The Wellcome Trust.
Original received November 19, 2001; revision received June 20, 2002; accepted June 20, 2002.
- ↵Mattie M, Brooker G, Spiegel S. Sphingosine 1-phosphate, a putative second messenger, mobilizes calcium from internal stores via an inositol trisphosphate-independent pathway. J Biol Chem. 1994; 269: 3181–3188.
- ↵Lee MJ, Van Brocklyn JR, Thangada S, Liu CH, Hand AR, Menzeleev R, Spiegel S, Hla T. Sphingosine-1-phosphate as a ligand for the G protein–coupled receptor EDG-1. Science. 1998; 279: 1552–1555.
- ↵Wu J, Spiegel S, Sturgill TW. Sphingosine 1-phosphate rapidly activates the mitogen-activated protein kinase pathway by a G protein–dependent mechanism. J Biol Chem. 1995; 270: 11484–11488.
- ↵Rakhit S, Conway AM, Tate R, Bower T, Pyne NJ, 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: 643–649.
- ↵Hla T, Maciag T. An abundant transcript induced in differentiating human endothelial cells encodes a polypeptide with structural similarities to G-protein–coupled receptors. J Biol Chem. 1990; 265: 9308–9313.
- ↵Im DS, Heise CE, Ancellin N, O’Dowd BF, Shei GJ, Heavens RP, Rigby MR, Hla T, Mandala S, McAllister G, George SR, Lynch KR. Characterization of a novel sphingosine 1-phosphate receptor, EDG-8. J Biol Chem. 2000; 275: 14281–14286.
- ↵An S, Bleu T, Zheng Y. Transduction of intracellular calcium signals through G protein-mediated activation of phospholipase C by recombinant sphingosine 1-phosphate receptors. Mol Pharmacol. 1999; 55: 787–794.
- ↵Ancellin N, Hla T. Differential pharmacological properties and signal transduction of the sphingosine 1-phosphate receptors EDG-1, EDG-3, and EDG-5. J Biol Chem. 1999; 274: 18997–19002.
- ↵Windh RT, Lee MJ, Hla T, An S, Barr AJ, Manning DR. Differential coupling of the sphingosine 1-phosphate receptors Edg-1, Edg-3, and H218/Edg-5 to the Gi, Gq, and G12 families of heterotrimeric G proteins. J Biol Chem. 1999; 274: 27351–27358.
- ↵Ishii I, Friedman B, Ye X, Kawamura S, McGiffert C, Contos JJ, Kingsbury MA, Zhang G, Heller Brown J, Chun J. Selective loss of sphingosine 1-phosphate signaling with no obvious phenotypic abnormality in mice lacking its G protein-coupled receptor, LP(B3)/EDG-3. J Biol Chem. 2001; 276: 33697–33704.
- ↵Tamama K, Kon J, Sato K, Tomura H, Kuwabara A, Kimura T, Kanda T, Ohta H, Ui M, Kobayashi I, Okajima F. Extracellular mechanism through the Edg family of receptors might be responsible for sphingosine-1-phosphate-induced regulation of DNA synthesis and migration of rat aortic smooth-muscle cells. Biochem J. 2001; 353: 139–146.
- ↵Kluk MJ, Hla T. Role of the sphingosine 1-phosphate receptor EDG-1 in vascular smooth muscle cell proliferation and migration. Circ Res. 2001; 89: 496–502.
- ↵Yatomi Y, Ruan F, Hakomori S, Igarashi Y. Sphingosine-1-phosphate: a platelet-activating sphingolipid released from agonist-stimulated human platelets. Blood. 1995; 86: 193–202.
- ↵Tosaka M, Okajima F, Hashiba Y, Saito N, Nagano T, Watanabe T, Kimura T, Sasaki T. Sphingosine 1-phosphate contracts canine basilar arteries in vitro and in vivo: Possible role in pathogenesis of cerebral vasospasm. Stroke. 2001; 32: 2913–2919.
- ↵Tasker PN, Michelangeli F, Nixon GF. Expression and distribution of the type 1 and type 3 inositol 1,4,5-trisphosphate receptor in developing vascular smooth muscle. Circ Res. 1999; 84: 536–542.
- ↵Sakurada S, Okamoto H, Takuwa N, Sugimoto N, Takuwa Y. Rho activation in excitatory agonist-stimulated vascular smooth muscle. Am J Physiol. 2001; 2001: 281:C571–C578.
- ↵Lee M-J, Evans M, Hla T. The inducible G protein–coupled receptor edg-1 signals via the Gi/mitogen-activated protein kinase pathway. J Biol Chem. 1996; 271: 11272–11279.
- ↵Kozasa T, Jiang X, Hart MJ, Sternweis PM, Singer WD, Gilman AG, Bollag G, Sternweis PC. p115 RhoGEF, a GTPase activating protein for Gα12 and Gα13. Science. 1998; 280: 2109–2111.
- ↵Jaggar JH, Nelson MT. Differential regulation of Ca2+ sparks and Ca2+ waves by UTP in rat cerebral artery smooth muscle cells. Am J Physiol. 2000; 279: C1528–C1539.
- ↵Brayden JE, Nelson MT. Regulation of arterial tone by activation of calcium-dependent potassium channels. Science. 1992; 256: 532–535.
- ↵Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M, Yamamori B, Feng J, Nakano T, Okawa K, Iwamatsu A, Kaibuchi K. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science. 1996; 273: 245–248.