Sphingosine 1-phosphate (S1P), a bioactive sphingolipid involved in diverse biological processes, is generated by sphingosine kinase (SphK) and acts via intracellular and/or extracellular mechanisms. We used biochemical, pharmacological, and physiological approaches to investigate in rat myometrium the contractile effect of exogenous S1P and the possible contribution of SphK in endothelin-1 (ET-1)-mediated contraction. S1P stimulated uterine contractility (EC50 = 1 μM and maximal response = 5 μM) by a pertussis toxin-insensitive and a phospholipse C (PLC)-independent pathway. Phosphorylated FTY720, which interacts with all S1P receptors, except S1P2 receptors, failed to mimic S1P contractile response, indicating that the effects of S1P involved S1P2 receptors that are expressed in myometrium. Contraction mediated by S1P and ET-1 required extracellular calcium and Rho kinase activation. Inhibition of SphK reduced ET-1-mediated contraction. ET-1, via ETA receptors coupled to pertussis toxin-insensitive G proteins, stimulated SphK1 activity and induced its translocation to the membranes. Myometrial contraction triggered by ET-1 is consecutive to the sequential activation of PLC, protein kinase C, SphK1 and Rho kinase. Prolonged exposure of the myometrium to S1P downregulated S1P2 receptors and abolished the contraction induced by exogenous S1P. However, in these conditions, the tension triggered by ET-1 was not reduced, indicating that SphK activated by ET-1 contributed to its contractile effect via a S1P2 receptor-independent process. Our findings demonstrated that exogenous S1P and SphK activity regulated myometrial contraction and may be of physiological relevance in the regulation of uterine motility during gestation and parturition.
contractility is an essential function of the uterus in the control of gestation. Myometrium, the smooth muscle of the uterus, needs to be quiescent during gestation to avoid premature expulsion of the fetus and to strongly contract at parturition for delivery. Thus fine regulation of this physiological process is absolutely required for normal enrolment of gestation. It is important to investigate the mechanisms that regulate the physiology of the myometrium in an attempt to treat or to prevent preterm birth. The mechanisms involved in the regulation of uterine contractility are complex and, although intensively investigated, are still poorly understood.
In different smooth muscles (54), including myometrium (19, 44), the increases in intracellular calcium concentrations play a major role in the contraction. The following two major mechanisms are associated with regulation of cellular calcium concentrations: release of calcium from intracellular stores or increase in calcium influx triggered by specific calcium channels such as voltage-operated calcium channels (VOCCs; see Ref. 54). In rat myometrium, these channels are particularly involved in calcium influx stimulated by G protein-coupled receptor (GPCR; see Ref. 19).
Different contractile agonists acting via GPCR stimulate the phospholipase C (PLC) pathway, which catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate into inositol 1,4,5-trisphosphate (InsP3). InsP3 increases the release of calcium from intracellular stores, which is soon followed by an entry of calcium across the plasma membrane (2), leading to the activation of myosin light chain kinase (MLCK) involved in myosin light chain (MLC) phosphorylation required for actin-myosin interaction and contraction (55). The stimulation of PLC also produces diacylglycerol (DAG), which activates protein kinase C (PKC). It has been reported that, in human myometrium, contraction is regulated by PKC (10, 38). PKC contributes to the process referred to as calcium sensitization (46, 55). This process can be the result of the inhibition of myosin light chain phosphatase (MLCP) leading to an increased MLC phosphorylation required for the contraction. The increase of calcium sensitivity occurs through PKC and Rho kinase activation, which act either singly or cooperatively to inhibit MLCP (55). These kinases can inhibit MLCP directly or via CPI-17 activation (39, 54).
It is well established that PKC regulates not only protein kinases but also sphingosine kinase (SphK), a lipid kinase involved in GPCR and receptor tyrosine kinase responses (28, 50, 55). This lipid kinase phosphorylates sphingosine to produce sphingosine 1-phosphate (S1P). Two forms of SphK have been cloned: SphK1 is a 42- to 49-kDa protein, whereas SphK2 is a 66-kDa protein. These proteins appear to contain multiple regulatory domains and putative phosphorylation sites for kinases such as PKC (28, 50). Ample evidence indicates that S1P modulates intracellular pathways that are important for diverse biological processes, including cell growth, survival, motility, and contraction. S1P has been proposed to act not only as an extracellular mediator via GPCR but also as an intracellular messenger. In this context, S1P has been reported to be a calcium-releasing factor (28, 48, 50, 55, 59). Recent progress in the identification of specific GPCR that can account for the extracellular effect triggered by S1P has improved the understanding of the mechanisms of action of this lipid. Five cognate GPCR have been identified (S1P1–5). The S1P1 receptor couples selectively to the Gi signaling pathway, whereas the sphingosine 1-bisphosphate (S1P2) and sphingosine 1-trisphosphate (S1P3) receptors couple to the Gi, Gq, and G12/13 signaling pathways (28, 50). Although the role of S1P in the regulation of diverse cellular functions emerged, its role in uterine smooth muscle is unknown.
Endothelin-1 (ET-1) is a vasoactive peptide of 21 amino acids. ET-1 produces its effects via interaction with ETA and ETB receptors, which belong to the GPCR family (8). In human myometrium, ET-1-induced myometrial contraction involves PKC activation (10, 38). In rat myometrium, ET-1-mediated contraction is regulated via a dual coupling of ETA receptors with Gi for the inhibition of the adenylyl cyclase pathway, resulting in a decrease in the cAMP, which is a relaxant messenger, and with Gq/G11 for the activation of the PLC/PKC/calcium pathway (16, 20). Because PKC is able to stimulate SphK activity and because S1P stimulates smooth muscle contraction (43, 55), we have analyzed the possible contribution of SphK in ET-1-mediated uterine contraction compared with the effect of exogenous S1P. We demonstrated that exogenous S1P, via interaction with S1P2 receptors, was able to stimulate myometrial contraction and that activation of SphK1 by ET-1 contributed to its contractile effect, most probably via an intracellular mechanism.
MATERIALS AND METHODS
[γ-32P]ATP (2,000 Ci/mmol) was purchased from Perkin-Elmer (Les Ulis, France). The myo-[3H]inositol (20 Ci/mmol) was from Pharmacia Biotechnology (Les Ulis, France). S1P and DL-threodihydrosphingosine were from Biomol (Tebu, Le Perray-en-Yvelines, France). Silica gel plates were purchased from Whatman (Kent, UK). BQ-123 and ET-1 were from Neosystem (Strasbourg, France). Phosphorylated FTY720 was synthesized by Novartis (Basel, Switzerland). Phorbol 12,13-dibutyrate (PDBu), leupeptin, aprotinin, phenylmethylsulfonyl fluoride, pertussis toxin (PTX), N,N-dimethylsphingosine, sphingosine, U-73122, Y-27632, 4-deoxypyridoxine, β-estradiol 3-benzoate, and rabbit polyclonal anti-S1P2 receptors were obtained from Sigma Chemicals (St. Louis, MO). U-0126 and rabbit polyclonal antibody to activated extracellular signal-regulated kinase (ERK) were from Promega (Madison, WI). Ro-31–8220 and the second antibodies, anti-goat IgG conjugated to horseradish peroxidase, were from Calbiochem (Meudon, France). The second antibodies, anti-rabbit IgG conjugated to horseradish peroxidase, were from Cell Signaling (Ozyme, Saint-Quentin en Yvelines, France). Rabbit polyclonal antibodies to total ERK were from Zymed Laboratories (San Francisco, CA). Rabbit anti-Rho and goat anti-actin antibodies were from Santa Cruz Biotechnology (Tebu, Le Perray-en-Yvelines, France). All other chemicals were of the highest grade available.
Prepubertal Wistar female rats (Janvier, France), 21 days old, were housed for 9 days in an environmentally controlled room before use. Chow and water were available ad libitum. Rats were treated by intraperitoneal injection of 30 μg of estradiol for the last 2 days and were killed the next morning at 30 days old by carbon dioxide inhalation. All the treatments were performed in accordance with the principles and procedures outlined in the European guidelines for the care and use of experimental animals. The uteri were removed immediately, and the myometrium was prepared free of endometrium as previously described (39).
Methods for recording uterine contractile response.
The contractile activity of isolated myometrial strips was measured with an isometric transducing device. The segments were attached to a vertical holder under resting tension of 0.2–0.3 g. Tissues were equilibrated in organ bath for 30–60 min at 37°C in 10 ml of Krebs-Ringer bicarbonate buffer containing 0.8 mM CaCl2 and 5 mM glucose (gas phase 95% O2 + 5% CO2). When the baseline did not change, pharmacological agents to be tested were added. Contractile activity was measured as area under the contraction curve obtained after 2 min exposure to each concentration of the indicated agent. We used the maximal tetanic contraction induced by 2 nM ET-1 as a reference response (∼6 cm2). Cumulative dose-response curves were calculated as the percentage of ET-1 response and plotted against the log of the agonist concentration. One rat was used for each experiment, which was reproduced with three different animals.
Preparation of the different myometrial fractions.
For each preparation, myometrial strips from 8–10 rat uteri were pooled and divided into equal parts. Myometrium were incubated in the presence of the different molecules to be tested for the times indicated in the figure legends. To stop the reactions, the tissues were frozen in liquid N2 before being homogenized with an Ultra Turrax. To determine SphK activity, tissues were homogenized in the SphK buffer containing 20 mM Tris (pH 7.4), 20% glycerol, 1 mM mercaptoethanol, 1 mM EDTA, phosphatase inhibitors (in mM: 15 NaF, 1 sodium orthovanadate, and 40 glycerol phosphate), protease inhibitors [10 μg/ml leupeptin, 10 μg/ml aprotinin, 10 μg/ml trypsin inhibitor, and 1 mM phenylmethylsulfonyl fluoride (PMSF)], and 0.5 mM 4-deoxypyridoxine, a pyridoxal phosphate analog that inhibits the pyridoxal-dependent S1P lyase. The homogenates were centrifuged at 700 g for 5 min at 4°C. The pellets were eliminated, and the SphK1 activity was measured in the supernatants. In some experiments, SphK1 activity was measured in membrane and cytosolic fractions obtained from supernatants by a centrifugation at 100,000 g for 20 min at 4°C. The membrane fractions were resuspended in SphK buffer. Protein concentrations in the different fractions were determined by the Bradford assay.
Measurement of SphK1 activity.
SphK1 activity contained in rat myometrial supernatant, cytosolic, or membrane fractions was analyzed, as described (23), in 200 μl of SphK buffer containing 50 μM sphingosine, [γ-32P]ATP (5 μCi, 1 mM), 100 mM MgCl2, 0.25% Triton X-100, which abolishes SphK2 activity (30), and 25 μg proteins. The solutions were incubated for 30 min at 37°C as described (23, 29). The reactions were stopped by the addition of 800 μl of chloroform-methanol-concentrated HCl (100:200:1). After vigorous vortexing, 250 μl of chloroform and 250 μl of KCl were added for phase separation. The samples were vortexed and centrifuged 5 min at 700 g. Labeled lipids contained in 100 μl of the organic phase were separated by TLC on silica gel using chloroform-acetone-methanol-acetic acid-water (50:20:15:10:5). The radioactive sphingosine phosphate was visualized by exposing the plate overnight on a storm phosphorimager. Specific SphK activity was expressed as picomoles S1P per minute per milligram of protein, and total activity in each fraction was expressed as picomoles S1P per minute per total protein.
Western blot analysis.
Myometrial strips were homogenized in solubilization buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 100 mM NaF, 10% glycerol, 10 mM Na4P2O7, 200 μM Na3VO4, and 10 mM EDTA), protease inhibitors (10 μg/ml leupeptin, 10 μg/ml aprotinin, and 1 mM PMSF), and 1% Triton X-100. The homogenate was centrifuged at 700 g for 5 min. For RhoA, SphK1, or S1P2 receptor analysis, the tissues were solubilized in a similar buffer without Triton X-100. The 700-g supernatants were centrifuged at 100,000 g for 20 min at 4°C. The membrane fractions were resuspended in the solubilization buffer containing 1% Triton X-100. Protein concentrations in the different fractions were determined by the Bradford assay. Detergent-extracted proteins were heated for 5 min at 95°C with Laemmli sample buffer and analyzed by 10% (wt/vol) SDS-PAGE. The separated proteins were transferred to nitrocellulose sheets and probed with polyclonal anti-SphK1 (dilution 1:1,000 vol/vol), anti-RhoA (dilution 1:500 vol/vol), anti-activated ERK (1:5,000 vol/vol), anti-total ERK (1:5,000 vol/vol), anti-S1P2 receptor (dilution 1:500 vol/vol), or anti-actin (dilution 1:1,000 vol/vol) antibodies. The immunoreactive bands were visualized by an enhanced chemiluminescence system after incubation with horseradish peroxidase-conjugated anti-rabbit IgG and quantified with a densitometer (Molecular Dynamics, Sunnyvale, CA).
Measurement of [3H]inositol phosphates.
For each experiment, six rats were used. Myometrial strips (25 mg) were labeled with 5 μCi of myo-[2-3H]inositol (0.4 μM) in 800 μl of fresh buffer for 4 h, essentially as described previously (39). Tissues were incubated for 10 min in 1 ml of fresh buffer containing 10 mM LiCl before exposure to the agents indicated. Reactions were stopped by immersion of the myometrial strips in liquid nitrogen. Tissues were homogenized in 1.5 ml of cold 7% (wt/vol) trichloroacetic acid and centrifuged at 3,000 g for 20 min at 4°C. The trichloroacetic acid-soluble supernatants were extracted with diethyl ether and applied to a column of anion exchange resin (AG1-X8, formate form, 200–400 mesh). Total inositol phosphates were eluted with 12 ml of 1 M ammonium formate/0.1 M formic acid (39). The 3H content of the fractions was determined by scintillation counting in Quicksafe (Zinsser Analytic). Results were expressed as counts per minute per 100 milligram of tissue.
All results presented derived from experiments repeated at least three times. Results are expressed as means ± SD.
S1P stimulated myometrial contraction.
Data from the laboratory have clearly demonstrated that different signaling pathways are involved in the modulation of myometrial contraction (39). In the present study, we analyzed the effect of exogenous S1P on myometrial contraction. When rat myometrial strips were incubated with increased concentrations of S1P, the smooth muscle contracted in a dose-dependent manner (Fig. 1A), with an EC50 = 2 μM and a maximal response at 7 μM (Fig. 1C). Contractions obtained with S1P were oscillatory (Fig. 1A) in contrast to the tetanic response triggered by ET-1 (Fig. 1B). The intensity of maximal contraction triggered by S1P was about one-half of the maximal contraction displayed by ET-1 (Fig. 1D).
Involvement of membrane receptors coupled to PTX-insensitive G protein in exogenous S1P-induced myometrial contraction.
Presently, the study of S1P receptor pharmacology remains limited by the lack of selective agonists and antagonists for these receptors. We used phosphorylated FTY720, which interacts with all S1P receptors excepted S1P2 (31, 50). Incubation of the myometrial strips in the presence of this agent failed to stimulate uterine contractility by itself at a concentration as high as 50 μM (Fig. 2A) and did not modify contraction due to S1P (Fig. 2B compared with Fig. 1C). The data suggested the involvement of S1P2 receptors in S1P-mediated myometrial contraction. These receptors are coupled to different G proteins sensitive or insensitive to PTX (55). In rat myometrium, contraction mediated by the activation of transmembrane receptors involved Gi and/or Gq proteins (16, 20, 24). In the presence of PTX, which strongly ADP ribosylated the Gi proteins in rat myometrium and blocked its interaction with cognate receptors (53), the contractile effect of ET-1 was reduced, indicating that Gi contributed to ET-1-mediated contraction (Fig. 2C). By contrast, the contraction resulting from S1P remained unaffected by PTX treatment (Fig. 2C), demonstrating that myometrial contraction induced by S1P was mediated via S1P receptors coupled to a G protein different from the Gi family.
Role of extracellular calcium, PLC, and Rho kinase in the contractile effect of ET-1 and S1P.
Our previous data demonstrated that uterine contraction triggered by different agonists, including ET-1, required PLC activation (1, 20, 24, 32, 39). However, contraction may also involve PLC-independent but calcium-dependent processes (35). Data shown in Fig. 3A demonstrated that, in contrast to ET-1, S1P used at concentration up to 50 μM failed to stimulate the production of inositol phosphates in rat myometrium. Thus S1P induced myometrial contraction independently from the PLC pathway and subsequent intracellular calcium release.
Because extracellular calcium concentration is a major regulator of myometrial contraction (44), we tested the influence of extracellular calcium on S1P and ET-1 effects. Data demonstrated that the incubation of myometrial strips in a calcium-free medium blocked contraction triggered by the two agonists (Fig. 3B). Myometrial contraction was associated not only with MLCK activation in the presence of calcium but also with the inhibition of MLCP, resulting in the increase of MLC phosphorylation that can be regulated by the activation of RhoA/Rho kinase (47, 55). We examined the role of Rho kinase in S1P- and ET-1-mediated contraction. Tension evoked by S1P was fully abolished when the myometrium were incubated in the presence of Y-27632, an inhibitor of Rho kinase, which inhibited by 44% the contraction triggered by ET-1 (Fig. 3B). Rho kinase is activated by RhoA, a monomeric G protein, that is localized in the cytosol in an inactive state in association with GDP dissociation inhibitor (GDI). GDI interacts with RhoA (12) and RhoA-GDI, the inactive complex, is maintained in the cytosol (4). The activation of RhoA by different agonists, including ET-1, leads to its dissociation from GDI and its translocation to the membranes to activate Rho kinase in fibroblasts (13). As expected, the data described in Fig. 3C showed that both ET-1 and S1P induced an increase (∼2-fold over basal) in the level of RhoA present in the membrane fraction. Altogether, these results indicated that myometrial contraction induced by S1P and ET-1 was totally dependent on extracellular calcium. The contraction induced by S1P was fully dependent on RhoA/Rho kinase, whereas ET-1-mediated contraction is only partially regulated by this kinase.
Involvement of PLC, PKC, SphK, and Rho/Rho kinase in ET-1-mediated contraction.
To investigate the mechanisms by which ET-1 induced myometrial contraction, we used a pharmacological approach. Incubation of the myometrium in the presence of U-73122, an inhibitor of PLC, partially reduced (55%) contraction induced by ET-1 (Fig. 4). ET-1-mediated contraction was unaffected by U-73343, the inactive analog of U73122 (data not shown). Increased intracellular calcium concentrations and/or production of DAG subsequent to PLC activation can be involved in PKC activation (11). Stimulation of PKC is able to contribute to the regulation of smooth muscle contractility (9, 10, 38). The treatment of myometrium in the presence of Ro-31–8220, an inhibitor of PKC, reduced by 43% uterine contractility induced by ET-1 (Fig. 4). The level of the inhibition of ET-1-mediated contraction observed in the presence of U-73122 or Ro-31–8220 was similar to that observed with Y-27632. Moreover, coincubation of myometrium in the presence of U-73122 plus Y-27632 or in the presence of Ro-31–8220 plus Y-27632 did not produce any additive inhibitory effect on the contraction due to ET-1 (Fig. 4), suggesting that PLC, PKC, and Rho kinase operated on a common signaling pathway to regulate myometrial contraction. Our data demonstrated that S1P contracted myometrial tissues (Fig. 1). SphK can be stimulated by GPCR via PKC activation (28, 50). To analyze the potential role of SphK activity in contraction triggered by ET-1, the effect of two SphK inhibitors [dimethylsphingosine (DMS) and DL-threo-dihydrosphingosine (threo-DHS)] was tested. Data illustrated in Fig. 4 showed that the presence of 20 μM threo-DHS or 10 μM DMS caused a marked and similar reduction in ET-1-mediated contraction. These observations suggested that ET-1 was able to stimulate SphK activity, which participated to the contraction triggered by ET-1. The reduction of ET-1-mediated contraction obtained with SphK inhibitors was comparable to that observed when PKC, PLC, or Rho kinase was inhibited.
In an attempt to check the specificity of the different pharmacological inhibitors, we determined the concentrations of these agents that induced half-maximal and maximal inhibition. Data reported in Table 1 showed that the inhibitory effect of Ro-31–8220, U-73122, Y-27632, DMS, and threo-DHS on ET-1-mediated rat myometrial contraction was dose dependent, with IC50 and maximal response values that are well in the range described for specific inhibition of PKC (42), PLC (61), Rho kinase (57) and SphK (22, 58).
Threo-DHS and DMS have been reported to inhibit not only SphK activity but also PKC (21, 56). In rat myometrium, we have analyzed the possible inhibition of PKC by these agents by testing their action on PKC-dependent activation of ERK induced by PDBu, a direct activator of PKC. Our data demonstrated that PDBu stimulated ERK2. Treatment of the tissue in the presence of Ro-31–8220 abolished ERK2 activation triggered by PDBu. In contrast, a similar treatment in the presence of 20 μM threo-DHS or 10 μM DMS did not affect ERK2 activation triggered by PDBu (Fig. 5). The present observation demonstrated that the inhibition of ET-1-mediated tension by DMS or threo-DHS involved SphK inhibition rather than inhibition of PKC.
Altogether, the data suggested that SphK activity is involved in contraction induced by ET-1, which also required a PLC-, PKC-, and Rho kinase-dependent pathway.
ET-1 stimulated SphK1 activity in myometrium.
Because contraction mediated by ET-1 seems to involve SphK activation, we evaluated the capacity of ET-1 to stimulate SphK activity by assaying this lipid kinase activity in myometrial extracts. Reactions were performed in the presence of 0.25% Triton X-100, which blocked SphK2 activity but not SphK1 activity (29). In homogenates prepared from control tissues, basal SphK1 activity was equal to 40 ± 5 pmol S1P·min−1·mg protein−1 (Fig. 6A), whereas, in homogenates from ET-1-stimulated myometrium, SphK1 activity was increased by 2.4- and 2.7-fold after 5 and 10 min incubation, respectively (Fig. 6A). The maximal effect observed at 10 min slightly declined at 15 min. The effect observed at 10 min was dose dependent (EC50 = 11 nM and maximal effect = 100 nM; Fig. 6B). Contraction induced by ET-1 was observed at low concentrations compared with that required to stimulate SphK1. This observation is in line with our previous data obtained with different contractile agonists (1, 24), including ET-1 (20), where a small generation of InsP3 produced by these agents was sufficient to induce a maximal contraction. Alternatively, the difference in the potency of ET-1 to cause contraction and SphK1 activation may be related to the experimental conditions. Indeed, the measurement of SphK1 activity was made in vitro, whereas the contraction was evaluated in intact tissues.
ET-1 translocated SphK1 from the cytosol to the membrane fraction.
It has been reported that the increase in SphK1 activity was particularly observed in membrane fractions (40). The subcellular location of SphK1 activity was determined upon fractionation of control and ET-1-stimulated myometrial strips. Data in Fig. 7A demonstrated that, in the absence of ET-1, SphK1 activity in the membrane fractions represented only 25% of total SphK1 activity expressed in the 700-g supernatants. When myometrial strips were stimulated with ET-1, the total SphK1 activity in the 700-g supernatants increased 2.4-fold. This increase concerned particularly the membrane fraction, where SphK1 activity was fourfold more important. To check if the increase of SphK1 activity in the membrane fraction was accompanied by a translocation of SphK1 in this fraction upon ET-1 stimulation, we determined the amount of SphK1 protein in the membrane fraction by Western blot. ET-1 mediated a time-dependent increase of the level of SphK1 in the membrane fraction (Fig. 7B). The maximal effect was observed at 10 min (Fig. 7C). The results indicated that the increase of SphK1 activity in the membrane fractions triggered by ET-1 was associated with its translocation in the membranes. Such relocation was also observed with PDBu or AlF4− (G protein activation) (Fig. 7, B and C), which activated SphK1 (Fig. 8B).
Mechanisms involved in ET-1-stimulated SphK1.
The effects of ET-1 can be mediated through the interaction with ETA and/or ETB receptors (8). We characterized the ET receptors involved in the activation of SphK1 in response to ET-1 in rat myometrium. Data reported in Fig. 8A demonstrated that SphK1 activation triggered by ET-1 was abolished when the tissues were treated in the presence of 1 μM BQ-123, a specific antagonist of ETA receptor. Activation of SphK1 triggered by ET-1 was refractory to a treatment of the myometrium with PTX (Fig. 8B) and then did not involve a Gi protein. Thus ET-1 stimulated SphK1 via ETA receptors coupled to PTX-insensitive G proteins. We also investigated whether the activation of PLC/PKC by ET-1 was involved in the stimulation of SphK1 in rat myometrium (Fig. 8B). Treatment of tissues with Ro-31–8220 abolished the activation of SphK1 resulting from PDBu but also from ET-1 and AlF4−. The activation of SphK1 induced by ET-1 and AlF4− was also strongly inhibited by U-73122, whereas, as expected, the response triggered by PDBu was unaffected. We have also analyzed the role of extracellular calcium in ET-1-mediated SphK1 activation. Treatment of myometrial strips in a calcium-free medium abolished stimulation of SphK1 induced by ET-1. This inhibitory effect was similarly observed with AlF4− or PDBu (Fig. 8B). The data showed that ET-1 and AlF4− stimulated SphK1 activity through a PLC-, PKC- and calcium-dependent process. Data in Fig. 8C showed that treatment of myometrium in the presence of Y-27632 failed to reduce SphK1 activation induced by ET-1, indicating that stimulation of SphK did not require Rho kinase activation by ET-1. This result suggested that ET-1 may independently stimulate both Rho kinase and SphK pathways or that RhoA/Rho kinase may be downstream from SphK in the same cascade. To discriminate between these two possibilities, we analyzed the effect of DMS on the translocation of RhoA in the membranes after stimulation of the tissues by ET-1. The results presented in Fig. 8D showed that the translocation of RhoA observed after 3 and 5 min incubation in the presence of ET-1 was markedly reduced in the presence of DMS. Similar results were obtained when SphK was inhibited in the presence of threo-DHS (data not shown). The data indicated that Rho kinase is downstream from SphK1 activation, whereas PKC and PLC are upstream in the pathway mediating the ET-1 contractile effect.
SphK activation contributed to ET-1-stimulated uterine contraction via a S1P receptor independent process.
The results herein described demonstrated that exogenous S1P induced myometrial contraction and that SphK activation triggered by ET-1 contributed to the contractile effect of this peptide. We analyzed if SphK activated by ET-1 participated to the contractile effect of ET-1 via its interaction with membrane S1P receptors. This question was addressed by experimental approaches based on S1P receptor downregulation by prolonged exposure to S1P (17). Incubation of the myometrium in the presence of S1P or ET-1 demonstrated that contraction resulting from exogenous S1P was totally inhibited in S1P-pretreated tissues, whereas contraction resulting from ET-1 was not affected (Fig. 9A). Additionally, in S1P-desensitized tissues, DMS was still able to reduce tension resulting from ET-1. Our data suggested the involvement of S1P receptors, most probably S1P2 receptors, as previously shown in Fig. 2A, in exogenous S1P-mediated contraction but not in ET-1-mediated contraction. We confirm the presence of S1P2 receptors in rat myometrium by an immunological approach (Fig. 9B). In the tissues pretreated with S1P and refractory to S1P-mediated contraction, the level of membrane S1P2 receptors was markedly reduced (Fig. 9B). The data demonstrated that, in rat myometrium, exogenous S1P was able to develop a homologous desensitization that may occur at the level of membrane S1P2 receptors.
The present study provided the first evidence that exogenous S1P and SphK contribute to myometrial contraction in rat myometrium. The sphingolipid mediator S1P stimulates contraction via a mechanism insensitive to PTX treatment. This observation led us to rule out the involvement of S1P 1, 4, and 5 receptors coupled only to the Gi family (55). It has been reported that S1P3 receptors are preferentially coupled to PLC activation compared with S1P2 receptors (55). In rat myometrium, S1P failed to activate PLC, suggesting that the contractile effect of S1P may be mediated via S1P2 receptors. Consistent with this interpretation, phosphorylated FTY720, which interacts with all S1P receptors except S1P2 (31, 50), failed to affect the uterine contraction. Additionally, our data demonstrated that S1P2 receptors are expressed in rat myometrial membrane and their downregulation was associated with a marked reduction in the contractile effect of S1P. The regulation of smooth muscle contraction by S1P2 receptors has also been reported in coronary artery (37). In rat myometrium, the contractile effect of S1P was markedly reduced in the presence of Rho kinase inhibitor, indicating the involvement of this kinase classically stimulated by the monomeric G protein RhoA (47). We demonstrated that S1P was able to translocate RhoA from the cytosol to the membrane fraction where, in turn, it activates Rho kinase (13, 47, 52). Concerning the mechanisms by which GPCR activate RhoA/Rho kinase, different results have been reported in the literature. RhoA can be stimulated by direct interaction with GPCR (3), or by G12/G13 activated by S1P2 receptors (41). More recently, it has been reported that RhoA is stimulated through a direct interaction of activated Gq/11α subunits with p63 RhoGEF independent from PLC-β (30). So far, all myometrial contractile agonists, acting via GPCR, required the activation of the PLC pathway. This work presented S1P as the first agonist able to induce rat myometrial contraction via the activation of RhoA/Rho kinase and independently from the PLC pathway.
Rho kinase is also involved in the contractile effect of ET-1. Indeed, Y-27632 partially reduced myometrial contraction induced by ET-1. The analysis of subcellular distribution of RhoA showed that ET-1 increased the amount of this monomeric G protein in the membrane fraction. The involvement of Rho kinase has also been described in ET-1-mediated contraction in rat pancreatic stellate cells (33), cerebrovascular smooth muscle (26), and oxytocin-induced uterine contraction (51). Our finding provided evidence that the contribution of RhoA/Rho kinase in ET-1-meditated contraction involved the stimulation of PLC/PKC. Indeed, the reduction of myometrial contractility observed when the PKC or the PLC activities were abolished, was not additive with that obtained with the Rho kinase inhibitor. We then investigated how PKC activity regulates the RhoA/Rho kinase-mediated ET-1 contractile effect. Because the activation of PKC stimulates SphK activity (28) and because S1P induced myometrial contraction via RhoA/Rho kinase, we analyzed the potential role of SphK activity in ET-1-mediated contraction. The inhibition of SphK activity by DMS or threo-DHS partially reduced contraction mediated by ET-1. Furthermore, ET-1 was able to stimulate SphK1 activity and to induce translocation of the enzyme from the cytosol to the membranes. The activation of SphK by ET-1 has also been observed in human hepatic stellate cells (15). In rat myometrium, ET-1 activated SphK1 via ETA receptors coupled to PTX-insensitive G protein. This ET-1 signaling pathway was previously described to be involved in the activation of PLC in rat myometrium (20). The inhibition of PLC or PKC resulted in a marked reduction of SphK1 activation by ET-1. SphK1 activity is sensitive to the concentrations of intracellular calcium (59). In rat myometrium, ET-1-mediated SphK1 activation required extracellular calcium. Our data demonstrated that Rho kinase inhibition failed to alter SphK1 activation induced by ET-1. By contrast, the inhibition of SphK markedly reduced ET-1-mediated RhoA translocation. These data led us to conclude that RhoA/Rho kinase was downstream from SphK and that RhoA activation is SphK dependent. The activation of RhoA/Rho kinase by SphK has also been described in arterial smooth muscle cells to modulate microvascular tone and myogenic responses (6); however, the mechanism remained unknown. It has been reported that PKC can directly activate RhoA and Rho kinase (13). Our cumulative data indicated that one mechanism by which ET-1 mediated rat myometrial contraction implied the interaction of ET-1 with ETA receptors coupled to PTX-insensitive G protein, leading to the sequential activation of PLC, PKC, SphK1, and RhoA/Rho kinase. The role of PKC via CPI-17 phosphorylation in the increased calcium sensitivity has also been reported (55). Our data demonstrated that the effect of PKC in ET-1-mediated rat myometrial contraction operated at the level of the SphK1/Rho kinase pathway that can be activated by PDBu. However, our group (unpublished data) and others found that this agent failed to contract the rat myometrium (45). These observations provide evidence that direct activation of this PKC signaling pathway is not sufficient to mediate rat myometrial contraction.
Because both exogenous S1P- and ET-1-induced SphK activation converged at the level of Rho kinase activation to regulate myometrial contraction, we investigated the potential role of membrane S1P receptors in the SphK contractile response triggered by ET-1. Ligand binding on GPCR, including S1P receptors (17), results in receptor phosphorylation and subsequent receptor downregulation (27). Based on this GPCR property, our data demonstrated that the downregulation of S1P2 receptors induced by exogenous S1P was associated with a marked reduction of S1P-mediated contraction without any effect on ET-1 contractile effect. These data suggested that the contribution of SphK1 activation in ET-1-mediated contraction occurred independent of an extracellular S1P/S1P receptor mechanism. Data from different groups proposed that SphK and production of intracellular S1P form a second messenger pathway used by membrane receptors for calcium mobilization. Indeed, the inhibition or the downregulation of SphK led to the reduction of agonist-mediated calcium mobilization, whereas the accumulation of intracellular S1P or overexpression of SphK (49) were associated with the increase of agonist-mediated intracellular calcium (49, 59).
In the myometrium, calcium is a major determinant in contractile activity (44, 46). The activation of GPCR stimulates entry of calcium, particularly through VOCCs (19, 44) that can be regulated by SphK in GH4C1 rat pituitary cells (5). SphK also reduces the level of the substrate sphingosine involved in the inhibition of L-type calcium channel conductance in cardiac myocytes (34). Calcium influx is regulated by membrane hyperpolarization resulting from K+ conductance inhibition. A recent study indicated that intracellular S1P suppress K+ conductance and then increases calcium influx (60, 61). Alternatively, the contribution of SphK activation is not the result of consecutive S1P production but rather because of S1P metabolites obtained via S1P lyase activity as reported for the regulation of mitogenesis in mouse F9 embryonic carcinoma cell lines (18). It is also possible that, in cells, S1P can be metabolized to produce ceramide 1-phosphate that may act as an intracellular second messenger much like S1P (25). The mechanism involved in the contribution of SphK in ET-1 contractile effect remained to be determined.
In conclusion, our findings open new insight on the potential role of exogenous S1P and the contribution of the SphK pathway in the regulation of myometrial contractility via a Rho- and/or Rho kinase-dependent signaling pathway. This mechanism was regulated by exogenous S1P independent from PLC and also by a sequential activation of PLC, PKC, and SphK in response to ET-1. It has been reported that the late pregnancy in different species is associated with the upregulation of calcium sensitivity by PKC/CPI-17 (38) and the Rho- and/or Rho kinase-mediated processes (7, 14, 36). Therefore, based on these collective findings, it is likely to consider that sphingolipid signaling pathway may contribute to these multiple mechanisms to facilitate the coordinated regulation of myometrial contractions. A better understanding of this signaling pathway might contribute to develop a new therapy to prevent preterm labor.
This work was supported by grants from the Centre National de la Recherche Scientifique and Université Paris-Sud.
We thank Novartis (Basel, Switzerland) for providing phosphorylated FTY720. We are grateful to Ginette Vilain for expert technical assistance and to Hervé Le Stunff for helpful discussion about the measurement of SphK activity.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society