Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid that mediates cellular functions by ligation via G protein-coupled S1P receptors. In addition to its extracellular action, S1P also has intracellular effects; however, the signaling pathways modulated by intracellular S1P remain poorly defined. We have previously demonstrated a novel pathway of intracellular S1P generation in human lung endothelial cells (ECs). In the present study, we examined the role of intracellular S1P generated by photolysis of caged S1P on EC barrier regulation and signal transduction. Intracellular S1P released from caged S1P caused mobilization of intracellular calcium, induced activation of MAPKs, redistributed cortactin, vascular endothelial cadherin, and β-catenin to cell periphery, and tightened endothelial barrier in human pulmonary artery ECs. Treatment of cells with pertussis toxin (PTx) had no effect on caged S1P-mediated effects on Ca2+ mobilization, reorganization of cytoskeleton, cell adherens junction proteins, and barrier enhancement; however, extracellular S1P effects were significantly attenuated by PTx. Additionally, intracellular S1P also activated small GTPase Rac1 and its effector Ras GTPase-activating-like protein IQGAP1, suggesting involvement of these proteins in the S1P-mediated changes in cell-to-cell adhesion contacts. Downregulation of sphingosine kinase 1 (SphK1), but not SphK2, with siRNA or inhibition of SphK activity with an inhibitor 2-(p-hydroxyanilino)-4-(p-chlorophenyl) thiazole (CII) attenuated exogenously administrated S1P-induced EC permeability. Furthermore, S1P1 receptor inhibitor SB649164 abolished exogenous S1P-induced transendothelial resistance changes but had no effect on intracellular S1P generated by photolysis of caged S1P. These results provide evidence that intracellular S1P modulates signal transduction in lung ECs via signaling pathway(s) independent of S1P receptors.
- Ca2+, cell-to-cell adhesion
sphingosine-1-phosphate (S1P), a bioactive sphingolipid, secreted by erythrocytes, activated platelets, and other cells, regulates a wide range of biological processes such as cell growth, differentiation, proliferation, motility, and barrier function (10, 26, 31, 37, 38). S1P exerts dual action in cells. It is a natural ligand for five G protein-coupled receptors, S1P receptors1–5 (10, 24, 38, 40, 49), and also acts as an intracellular second messenger (4, 14, 25, 37, 50). The receptors also recognize dihydro-S1P (3, 29, 38). In cells and tissues, formation of S1P from sphingosine (Sph) is catalyzed by SphK1 and SphK2 isoforms (18, 26). S1P, however, can be degraded by S1P phosphatases to Sph and S1P lyase to hexadecenal and ethanolamine phosphate (18, 30, 38, 53). In the vascular endothelium, S1P plays an important role in protecting lungs from agonist- or sepsis-induced pulmonary leak and lung injury, as demonstrated using both in vitro and in vivo models (8, 9, 22, 28, 33). The mechanisms by which S1P regulates the integrity of the endothelial barrier, however, still remain unclear. S1P modulates endothelial cytoskeleton, particularly cortactin, an actin-binding protein, in the formation of a strong cortical actin ring in the cell periphery (8, 10, 17). It stimulates MAPKs and focal adhesion kinase, causing rearrangement of vascular endothelial (VE)-cadherin, paxillin, catenins, and zonula occludens (ZO)-1 in endothelial cells (ECs) (8, 23, 36). Additionally, S1P is a potent agonist for the regulation of intracellular calcium (4, 14, 20, 23, 24, 35, 50), which in turn plays an important role in the activation of several kinases including AKT and formation of stress fibers (41, 42, 45). Finally, S1P activates Rac1, a critical regulator of nonmuscle cytoskeleton, cell-to-cell and cell matrix-tethering forces (12, 36). In addition to its extracellular action, S1P can function as an intracellular signaling molecule. Photolysis of intracellular caged S1P mobilized cytosolic Ca2+ from thapsigargin-sensitive stores that was independent of S1P receptors in HEK-293, SKNMC, and HepG2 cells (25) and also inhibited cell motility in breast cancer cells (47).
Maintenance of EC barrier integrity is critical for vessel wall homeostasis and normal organ function. We recently demonstrated that exogenous S1P was rapidly converted to intracellular S1P in ECs, thereby raising the possibility that a novel pathway(s) may exist (32). In an attempt to understand these pathway(s), we uploaded ECs with photolyzable caged S1P that can be released upon UV illumination directly into the cytoplasm. Our data presented here with caged S1P support a role for intracellular S1P in elevating calcium levels, MAPK and Rac1/IQGAP1 activation, cytoskeleton remodeling, and modulation of barrier function in human pulmonary artery ECs (HPAECs). The responses of released S1P from its cage were pertussis toxin (PTx)-insensitive support of the concept of intracellular actions of S1P that are independent of S1P receptors (25).
MATERIALS AND METHODS
HPAECs and endothelial basal media (EBM-2) were obtained from Lonza (San Diego, CA). PBS was from Biofluids (Rockville, MD). Glass-bottom Microwell dishes (MatTek, Ashland, MA), Fura-2 AM ester, Fluo-4 AM ester, pluronic acid (F-127), mounting media, BAPTA-AM, and precast Tris-Glycine PAAG (Invitrogen, Eugene, OR), S1P (Avanti Polar Lipids, Alabaster, AL), caged S1P (Toronto Research Chemicals, Toronto, ON, Canada), gadolinium chloride (Sigma, St. Louis, MO), thapsigargin (TG), inhibitors SP600125, SB203580, PD98050, NSC23766, and PTx (Calbiochem, San Diego, CA), SphK inhibitor, compound II, (Cayman Chemical, Ann Arbor, MI), scrambled siRNA and siRNA for SphK1 (Santa Cruz Biotechnology, Santa Cruz, CA), and GeneSilencer (Genlantis, San Diego, CA) were commercially obtained. Adenoviral constructs, vector control, and SphK1-flag (Dn) were generated at the services of the University of Iowa Gene Transfer Vector Core (Iowa City, IA). S1P1 receptor inhibitor SB649146 was kindly provided by GlaxoSmithKline. Cell lysis buffer (Cell Signaling, Danvers, MA) and immunobilon-P, 0.45 μm (Millipore, Bedford, MA), were also purchased.
Anti-SphK1 antibody was purchased from Abcam (rabbit, cat. no. ab37980, Cambridge, MA). Anti-ERK1 (rabbit, cat. no. sc-93), anti-ERK2 (rabbit, cat. no. sc-154) anti-phosphospecific-ERK (mouse, cat. no. sc-7383), anti-JNK (rabbit, cat. no. sc-571), anti-phosphospecific-JNK (mouse, cat. no. sc-6254), anti-p38 (rabbit, cat. no. sc-535) anti-phosphospecific-p38 (mouse, cat. no. sc-7973), and anti-cortactin (rabbit, cat. no. sc-11408) antibodies were purchased from Santa Cruz Biotechnology. Anti-VE-cadherin (rabbit, cat. no. 160840; Cayman Chemical), anti-β-catenin (mouse, cat. no. 610153), anti-Flag (rabbit, cat. no. F7425; Sigma), secondary antibodies Alexa Fluor 488 (mouse, cat. no. A21202 and rabbit, cat. no. A11034), Alexa Fluor 568 (rabbit, cat. no. A11036) (Invitrogen), goat anti-rabbit- (cat. no. 170-6515) or anti-mouse- (cat. no. 170-6516) IgG (H+L) horseradish peroxidase conjugates (Bio-Rad, Hercules, CA) were commercially obtained.
HPAECs cultured in EBM, were maintained at 37°C and 5% CO2, and grown to contact-inhibited monolayers that revealed typical cobblestone morphology. Cells were then detached with 0.05% trypsin, resuspended in fresh medium, and cultured on gold electrodes for electrical resistance determinations, on glass coverslips for calcium measurement or fluorescent microscopy studies, or on 100-mm culture dishes for immunoblotting.
Loading of ECs with caged S1P.
HPAECs grown on either cell culture dishes, glass-bottom 35-mm dishes, or gold electrodes were preloaded with caged S1P (intracellularly, “in”), or caged S1P was added directly into the media (outside loading, “out”). Time for saturating the cells with caged S1P and UV flashing was established in preliminary experiments. Serum-free and phenol red-free EBM medium was used in all experiments. Intracellular loading was performed in the dark (diffused red light, Littlight lamp) with 1 μM caged S1P for 15 min, and cells were washed thrice. For the extracellular application, 1 μM caged S1P was added into the incubation media before UV stimulation as indicated. The dishes or electrodes were then flashed with UV light (254 nm) for 20 s at 5-cm distance, the shortest and most efficient time period and distance as determined for the release of S1P by a Spectroline ENF-240-C (Spectronics, New York) and a UV-light source (0.2 A).
Measurement of intracellular Ca2+ concentration.
HPAECs were grown in complete EBM as monolayers on glass coverslips or glass-bottom 35-mm dishes using a basic phenol red-free medium (in mM): 116 NaCl, 5.37 KCl, 26.2 NaHCO3, 1.8 CaCl2, 0.81 MgSO4, 1.02 NaHPO4, 5.5 glucose, 10 HEPES/HCl pH 7.40. Cells were loaded for 15 min with 5 μM Fura-2 AM (spectrofluorimetry) or with Fluo-4 AM (confocal microscopy) in the above media, in the presence of 0.1% BSA and 0.03% pluronic acid F-127 at 37°C in a cell-culture incubator as recommended by the manufacturer. Intracellular calcium was monitored with an Aminco-Bowman Series 2 luminescence spectrometer (SLM/Aminco, Urbana, IL) at excitation wavelengths of 340 and 380 nm and emission wavelength of 510 nm. Confocal microscopy imaging experiments were performed in real time using a Leica DMI6000 and ×63 (NA 1.40) objective on a SP5 resonant scanner laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany). Photo excitation of Fluo-4 was achieved by illumination at 488 nm (Ar laser), and emitted light was acquired at 510–540 nm. Data acquisition (5 Hz) was performed with Leica Confocal software. Data analysis was performed with NIH ImageJ software (27).
Preparation of cell lysates and immunoblotting.
ECs were grown on 100-mm culture dishes, and 18 h before the experiment the cells were transferred to serum-free medium. After treatment with caged S1P, dishes were rinsed with ice-cold PBS in the presence of 1 mM sodium orthovanadate. ECs were lysed in 1 ml of lysis buffer containing 1% phosphatase inhibitor cocktail, scraped off the dishes, sonicated on ice with a probe sonicator (15 s), and centrifuged at 5,000 g in a microfuge (4°C for 5 min), and protein concentrations of the supernatants were determined using Pierce protein assay kit. The supernatants, adjusted to 0.5–1.0 mg protein/ml (cell lysates) were denatured by boiling in 2× SDS sample buffer for 5 min and analyzed on 10% SDS-PAGE gels. Protein bands were transferred overnight (25 V, 4°C) on the PVDF (Millipore) membrane, probed with primary and secondary antibodies, and immunodetected by enhanced chemiluminescence (ECL Kit, Amersham). The blots were scanned (UMAX Power Lock II) and quantified by ImageJ software (27).
HPAECs grown on slide chambers were fixed with 3.7% paraformaldehyde in PBS for 10 min and permeabilized for 4 min in 3.7% paraformaldehyde containing 0.25% Triton X-100. In some experiments aimed at Rac1, permeabilization was performed by methanol treatment for 4 min at −20°C. Cells were then rinsed and incubated for 30 min in TBS with Tween (TBST) blocking buffer containing 1% BSA followed by incubation with primary antibodies (1:200 dilution in blocking buffer, 1 h). After being thoroughly rinsed with TBST, cells were then stained with Alexa Fluor secondary antibodies (1:200 dilutions in blocking buffer, 1 h). The washed slides were prepared using mounting media and examined with a Nikon TE 2000-S fluorescence microscope and Hamamatsu digital camera (Japan) using a ×60 oil-immersion objective lens and MetaVue software (Universal Imaging, West Chester, PA).
Infection and transfection of HPAECs.
HPAECs grown to ∼80% confluence were infected with 5 pfu/ml purified adenoviral empty vector and adenoviral vector containing cDNA for SphK1-flag dominant negative. After infection (24 h) the virus-containing medium was replaced with EBM, and the experiments were carried out. In separate experiments HPAECs grown to ∼50% confluence were transfected with 50 nM scrambled siRNA and SphK1 siRNA in serum-free EBM-2 medium according to the manufacturer's recommendation. After 3 h posttransfection, complete EGM-2 medium containing 10% FBS was added, and the cells were cultured for an additional 72 h.
RNA isolation and real-time RT-PCR.
Total RNA was isolated from HPAECs grown on 35-mm dishes using Trizol reagent according to the manufacturer's instruction. iQ SYBR Green Supermix was used to do the real-time measurements using iCycler by Bio-Rad. 18S (sense, 5′-GTAACCCGTTGAACCCCATT-3′, and antisense, 5′- CCATCCAATCGGTAGTAGCG-3′) was used as a housekeeping gene to normalize expression. The reaction mixture consisted of 0.3 μg of total RNA (target gene) or 0.03 μg of total RNA (18S rRNA), 12.5 μl of iQ SYBR Green, 2 μl of cDNA, 1.5 μM target primers, or 1 μM 18S rRNA primers, in a total volume of 25 μl. For all samples, reverse transcription was carried out at 25°C for 5 min, followed by cycling to 42°C for 30 min and 85°C for 5 min with iScript cDNA synthesis kit. Amplicon expression in each sample was normalized to its 18S rRNA content.
Measurement of transendothelial cell electrical resistance.
HPAECs were seeded on gold electrodes (8 wells, 10 electrodes/well) to ∼95% confluence, electrodes were treated with caged S1P as described above, and transendothelial electrical resistance (TER) was measured across the EC monolayer. To estimate differences between cell-to-cell and cell-to-matrix components, total TER was resolved into values reflecting resistance to current flow beneath the cell layer (α) and resistance to current flow between adjacent cells (Rb), utilizing the method of Giaever and Keese (11), which models the endothelial monolayer mathematically.
ANOVA with Student-Newman-Keuls test was used to compare means of clearance rates of two or more different treatment groups. The level of significance is P < 0.05 unless otherwise stated. Data are expressed as means ± SE.
Intracellular S1P modulates Ca2+ signaling in HPAECs.
It is well established that agonist-mediated calcium signaling is initiated through PLC/IP3 mechanisms, which induces calcium release from the endoplasmic reticulum and it triggers activation of store-operated calcium entry, resulting in Ca2+ influx from extracellular media (51). As S1P is a potent modulator of calcium signaling (23, 35, 41, 42), we examined the effect of S1P on mobilization of intracellular calcium release. There was a dose-dependent release of intracellular calcium by S1P. S1P at 10, 100, 500 and 1,000 nM corresponded to [Ca2+]i at 99 ± 11, 237 ± 14, 334 ± 19, and 365 ± 21 nM, respectively. Treatment of cells with BAPTA (25 μM), an intracellular Ca2+ chelator, for 1 h completely abolished changes in [Ca2+]i induced by S1P [control with no BAPTA - 100%; with BAPTA (25 μM) plus S1P: (1 μM), 66 ± 10%; (10 μM), 23 ± 11%; (25 μM), 9 ± 6%]. Pretreatment of HPAECs with gadolinium (Gd3+), a specific blocker of store-operated calcium channels (14, 50, 51), partially attenuated S1P-induced [Ca2+]i response (compare control - 100%, vs. Gd3+ plus S1P - 65 ± 4%), whereas treatment of cells with 5 μM TG released Ca2+ from endoplasmic reticulum, and subsequent addition of S1P caused no further change in [Ca2+]i. Furthermore, exogenous addition of S1P or its release from caged S1P (out) into incubation media or release intracellularly from loaded caged S1P (in) caused rapid and significant increase in [Ca2+]i (Fig. 1, A–C). UV light alone did not cause any changes in [Ca2+]i (Fig. 1B). Pretreatment of HPAECs with TG (Fig. 1D), BAPTA (Fig. 1G), or in the presence of Gd3+- or Ca-free media (Fig. 1, E and F) attenuated intracellularly released S1P-induced changes in [Ca2+]i. It is well known that most S1P effects on cell signaling act through G protein-coupled receptors (10, 24, 29, 38, 49). Pretreatment of HPAECs with PTx, a Gi-protein receptor blocker, in a time-dependent manner, attenuated extracellular S1P-induced changes in [Ca2+]i (Fig. 2A) but had no effect on [Ca2+]i induced by intracellular S1P (Fig. 2B). These data suggest that intracellular S1P directly induces Ca release from the endoplasmic reticulum and modulates [Ca2+]i in a Gi-independent manner in HPAECs.
Intracellular S1P activates MAPKs in HPAECs.
On the basis of earlier results that S1P activates MAPKs (5, 43, 48, 49), we next investigated the effect of intracellular S1P on MAPK activation. As shown in Fig. 3, exogenously added or released S1P as well as S1P released from caged S1P inside the cell stimulated phosphorylation of JNK, p38 MAPK, and ERK. Pretreatment of cells with PTx significantly attenuated exogenously added S1P-induced MAPK activation but had no effect on intracellularly released S1P from caged S1P on phosphorylation of JNK, p38 MAPK, and ERK (Fig. 3, A and B). We and others previously demonstrated that intracellular calcium is critical for phosphorylation of cellular proteins mediated by protein kinases (41, 42, 45). Furthermore, pretreatment of HPAECs with 25 μM BAPTA significantly attenuated MAPK phosphorylation mediated by intracellular S1P released from caged S1P (Fig. 4, A and B). These data suggest that both extracellular and intracellular S1P-induced MAPK activation is dependent on intracellular calcium.
Intracellular S1P modulates cytoskeletal rearrangement in HPAECs.
As exogenous S1P is known to induce cytoskeletal rearrangement of proteins such as cortactin (8, 22, 34), we examined the effect of intracellular S1P released from caged S1P on cortactin reorganization. As shown in Fig. 5, control cells showed a diffused cortactin distribution, with little localization near the cell periphery. Treatment of HPAECs with extracellular S1P or intracellular S1P released from caged S1P induced translocation of cortactin to regions near the cell periphery and areas of membrane ruffle (Fig. 5, top). Pretreatment of cells with PTx attenuated cortactin redistribution by exogenously added S1P or S1P released from caged S1P outside but had no effect on intracellular S1P-mediated cortactin redistribution (Fig. 5, bottom). These data demonstrate for the first time the ability of intracellular S1P to modulate cytoskeletal reorganization in HPAECs.
Intracellular S1P modulates focal adhesion and adherens junction proteins in HPAECs.
Earlier studies have demonstrated the ability of S1P to stimulate redistribution of focal adhesion and adherens junction proteins to cell periphery and membrane ruffles, which in turn regulates endothelial barrier function (8, 17, 23, 36). Having established that intracellular S1P modulates cortactin, we next investigated the effect of intracellular S1P on redistribution of focal adhesion and adherens junction proteins. As shown in Fig. 6A, HPAECs that were exposed to UV flash showed a somewhat diffused localization pattern of VE-cadherin, β-catenin, and ZO-1; however, intracellular S1P promoted reorganization of VE-cadherin and β-catenin toward cell periphery and changes in ZO-1 localization, which were similar to the redistribution of adherens and tight junction proteins by S1P applied outside the cells. These results provide strong evidence for redistribution of adherens junction proteins by intracellular S1P in HPAECs.
Intracellular S1P activates Rac1 and IQGAP1 in HPAECs.
Earlier, it has been shown that exogenous S1P activated Rac1, cytoskeletal rearrangement, and barrier enhancement via S1PR1 in ECs (12, 23, 36). As intracellular S1P modulates [Ca2+]i and activates JNK, p38 MAPK, and ERK that is PTx independent, we next evaluated the role of intracellular S1P on Rac1 and IQGAP1 activation. As shown in Fig. 6B, in control-untreated cells, Rac1 and IQGAP1 were dispersed mainly in the cytosol with minor quantities localized to plasma membrane. However, S1P released from caged S1P induced redistribution of Rac1 and IQGAP1 to cell periphery, which was similar to the S1P applied outside (Fig. 6B). These results demonstrate that S1P generated inside HPAECs from caged S1P activates Rac1 and IQGAP1.
Intracellular S1P modulates endothelial permeability.
To further investigate intracellular S1P effects on endothelial barrier function, HPAECs were grown on gold microelectrodes, preloaded with 1 μM caged S1P that was released inside the cell by UV flash. The resulting changes in TER were monitored as an index of endothelial permeability. As shown in Fig. 7A, the release of intracellular S1P resulted in a rapid and significant increase in TER, a reflection of a proportional decrease in EC permeability. This effect is similar to S1P action applied exogenously as described previously by us and others (8, 10, 14, 34). Caged S1P alone added to the incubation media without exposure to UV flash had no effect on TER, suggesting that release of S1P from the cage is critical for its EC barrier enhancement. S1P released into incubation media by UV light demonstrated similar effect as that of intracellular S1P (Fig. 7A). Exposure of cells to UV flash alone did not result in any significant changes in the TER. Additionally, our study demonstrates that the major alteration in TER after S1P treatment, both extracellular and intracellular, was attributable to changes in Rb (cell-to-cell contacts) through increased cell-to-cell adhesion (Fig. 7B). In contrast to Rb, very little change to α (cell-matrix contacts) occurred, which is a measure of the distance between the endothelial plasma membrane and the surface of the electrode. The above results indicated for the first time that intracellular S1P modulates barrier function through the changes in cell-to-cell adhesion contacts.
Downregulation of Sph kinase 1 expression or activity attenuates S1P-induced intracellular calcium and barrier enhancement.
To further understand the role of intracellular S1P, we investigated whether downregulation of Sph kinase 1 (SphK1) expression or activity modulates extracellular action of S1P. Infection of HPAECs with SphK1-flag, (Dn) significantly increased expression of the protein that attenuated S1P-induced intracellular Ca2+ (Fig. 8A). This suggests a role for intracellular S1P generated by SphK1 in [Ca2+]i. Next, we investigated the role of SphK1 on S1P-induced increases in endothelial permeability. Transfection of HPAECs with SphK1 siRNA knocked down SphK1 protein expression as evidenced by real-time PCR and Western blotting (Fig. 8B) and reduced ability of cells to convert Sph to S1P (2). Furthermore, downregulation of SphK1 by siRNA attenuated S1P-induced increase of TER compared with scrambled siRNA (Fig. 8B). However, knockdown of SphK2 had no effect on S1P-induced TER (data not shown). Similarly, inhibiting SphK activity with SphK inhibitor, CII, also prevented S1P-mediated TER changes (Fig. 8B). These results suggest a role for intracellular S1P generated by SphK1 in exogenous S1P mediated [Ca2+]i and endothelial barrier enhancement.
The role of S1P1 receptor and Rac1 on intracellular S1P-mediated endothelial barrier enhancement.
Several signaling pathways including activation of G protein-coupled S1P receptors have been implicated in S1P-induced endothelial barrier function. Having established that intracellularly generated S1P from caged S1P modulates endothelial signaling and that barrier function was PTx independent, we next examined the role of S1P1 receptor and Rac1 activation in intracellular S1P-mediated endothelial barrier regulation. HPAECs were pretreated with SB649146, a S1P1 receptor inhibitor, or NSC23766, a Rac1 inhibitor, and S1P-induced TER changes were measured. Treatment of cells with SB649146 attenuated TER mediated by S1P released outside the cell from caged S1P (control: caged S1P, out - 312 ± 46; SB649146 + caged-S1P, out - 57 ± 17) and had no effect, however, on TER mediated by S1P released inside the cell from caged S1P (Fig. 9A). In contrast to SB649146, treatment of cells with NSC23766, an inhibitor of Rac1, blocked TER changes mediated by extracellular S1P released from caged S1P (data not shown) or intracellular S1P generated from caged S1P (Fig. 9B). These results suggest that intracellular S1P generated from caged S1P modulates endothelial barrier function independent of S1P1 receptor but requires Rac1.
S1P is a naturally occurring bioactive lipid that has both extracellular and intracellular effects in mammalian cells. Extracellular action of S1P is through its G protein-coupled S1PRs that trigger subsequent activation of downstream targets such as Rho-GTPases, cytoskeletal reorganization, adherens and tight-junction assembly, and focal adhesion formation (8, 21, 29, 37). Also, there is some evidence for intracellular S1P-mediated effects; however, the targets and underlying mechanisms are yet to be fully defined. In the cell, S1P is generated by phosphorylation of Sph catalyzed by SphK1 and SphK2 (18). S1P is catabolized by S1P phosphatases to Sph or by S1P lyase to ethanolamine phosphate and hexadecenal (3, 18, 26, 29, 38, 53). The accumulation of S1P in cells is a natural balance between its synthesis and catabolism. Whereas some of the earlier studies have used inhibitors of SphKs to determine intracellular action of S1P, we have taken the approach of generating intracellular S1P by uploading the cells with caged S1P followed by UV photolysis (32). In the present study, using such an approach, we show that intracellular S1P 1) increases intracellular calcium [Ca2+]i, 2) activates MAPKs, JNK, p38 MAPK and ERK, which are PTx insensitive, 3) induces cortical actin redistribution to cell periphery that is also PTx insensitive, 4) stimulates rearrangement of focal adhesion and tight junction proteins VE-cadherin, β-catenin, and ZO-1, 5) activates small GTPase Rac1 and its effector IQGAP1, and 6) decreases EC permeability or tightens the barrier through the changes in cell-to-cell adhesion contacts.
S1P is a potent modulator of intracellular calcium release in ECs (23, 35, 41, 42). Exogenous S1P mobilizes Ca2+ from intracellular stores either through an IP3-dependent or -independent mechanism (4, 20, 24, 37). Furthermore, S1P-mediated Ca2+ mobilization was independent of mitochondrial Ca2+ stores as evidenced by studies with uptake inhibitor, CCCP, or release inhibitor, cyclosporine (35). Use of caged S1P to evaluate potential effects of intracellular S1P on DNA synthesis has been previously reported (32). Our results demonstrate that photolysis of caged S1P raised intracellular free Ca2+ levels in human lung ECs, which are in agreement with an earlier report that also used caged S1P in SKNMC and HepG2 cells (25). In the present study, caged S1P altered Ca2+ mobilization in HPAECs, which were sensitive to TG and BAPTA. Treatment of cells with TG released Ca2+ from endoplasmic reticulum, and subsequent photolysis of caged S1P inside the cell did not alter [Ca2+]i . However, gadolinium (Gd3+), a specific blocker of store-operated calcium channels, or Ca-free media abolished calcium influx from extracellular space but did not alter Ca2+ release from the endoplasmic reticulum. These data support the idea that S1P released intracellularly directly induces Ca2+ release from the endoplasmic reticulum. Additionally, we show that the caged S1P-induced Ca2+ mobilization was PTx insensitive compared with exogenous S1P-mediated Ca2+ release that was PTx sensitive. In mammalian cells, intracellular S1P is generated by phosphorylation of Sph catalyzed by SphK1 and 2. In the present study, a role for intracellular S1P in [Ca2+]i and endothelial permeability was demonstrated by modulating SphK1 with dominant-negative SphK1, SphK1 siRNA, or SphK1 inhibitor (Fig. 8). Interestingly, inhibition of S1P1 receptor with SB649146 had no effect on endothelial permeability mediated by intracellularly generated S1P from caged S1P inside the cell (Fig. 9A), suggesting a S1P1 receptor-independent mechanism in barrier regulation by intracellular S1P.
It is well recognized that increased vascular EC permeability and edema are hallmarks of many lung inflammatory diseases (21). During acute lung injury, acute respiratory distress syndrome, or thrombocytopenia, increased capillary permeability accelerates fluid and protein extravagation. The effect is reversible with infusion of platelets or platelet-released product, S1P. S1P and its analogs have proven effective in reducing pulmonary leak and barrier dysfunction as shown in in vivo and in vitro models (6, 21, 28, 33). Infusion of S1P in murine and canine models significantly reduced LPS-induced microvascular permeability, inflammation, and phagocyte infiltration (28). Ligation of S1P to S1P1/S1P3 receptors in human lung ECs leads to reorganization of cytoskeletal proteins, enhanced junctional integrity, and tightened endothelial barrier (22). Our data with caged S1P show that intracellularly generated S1P, similar to exogenous S1P, mediates redistribution of VE-cadherin, β-catenin, ZO-1, and Rac1/IQGAP1 to cell periphery (Fig. 6). Unlike previous studies (35), we show here that intracellular S1P-mediated stimulation of [Ca2+]i is PTx insensitive, suggesting a novel underlying mechanism utilized by intracellular S1P as opposed to that mediated by extracellular S1P bound to its cell surface receptors. There are only a few studies that describe calcium release from the endoplasmic reticulum by intracellular S1P (4, 25). However, there are a number of reports on sphingolipid-mediated [Ca2+]i changes, which cannot be attributed to G protein-coupled receptors (4, 25, 50, 51). The presence of a “sphingolipid-calcium release-mediated protein of the ER” was postulated as the Ca channels that appear to be a radically different to either InsP3R or Ryanodine receptors (15, 16, 19, 50).
Several studies have shown that extracellular S1P activates MAPKs, especially p42/p44 MAPK, through Gi-coupled S1P receptors, which are sensitive to PTx (31, 48). Here, we have demonstrated for the first time that S1P released inside the cell from caged S1P stimulated phosphorylation of JNK, p38 MAPK, and ERK. Furthermore, caged S1P-mediated phosphorylation of MAPKs was insensitive to PTx, whereas activation of MAPKs by exogenous S1P was PTx sensitive. Activation of MAPKs by extracellular S1P regulates cellular responses such as DNA synthesis, cell migration, and endothelial capillary tube formation (5). Our results show that inhibition of MAPK by pharmacological inhibitors such as SP600125 (JNK), SB203580 (p38), and PD98050 (ERK) has no effect on S1P-induced redistribution of focal adhesion, tight junction proteins, and Rac1/IQGAP1 redistribution (Supplemental Figs. S1 and S2; supplemental material for this article is available online at the American Journal of Physiology Lung Cellular and Molecular Physiology website). Thus, although the data suggest that both extracellular S1P and intracellular S1P stimulate phosphorylation of JNK, p38 MAPK and ERK, the MAPK signaling pathway is not involved in S1P-induced barrier enhancement in HPAECs.
Although much is known on signaling pathways and extracellular action of S1P via its G protein-coupled receptors, very little is known on the role of intracellular S1P and its targets in mammalian cells. Recent studies suggest potential interaction between S1P and histone acetylase 2 in breast cancer cells (13) and S1P as a cofactor for E3 ubiquitin ligase TRAF2 in HEK-293 cells (1). We have recently demonstrated that intracellularly generated S1P offers protection against LPS-induced lung injury and inflammation in a murine model of acute lung injury (52). Furthermore, EC motility mediated by extracellular S1P was dependent on intracellular S1P production, which was regulated by SphK1 and S1P lyase (2). Our results show that downregulation of SphK1 expression or activity and Rac1, but not S1P1 receptor, attenuated S1P-induced endothelial barrier enhancement (Figs. 8B and 9, A and B), further supporting a role for intracellular S1P in modulating signaling pathways independent of S1P receptors in the endothelium. Mechanisms of intracellular action of S1P are yet to be completely characterized (39); however, the action of phosphatidic acid (PA) generated from phospholipase D signal transduction in OVCAR-3 cells seems to act as a membrane anchor of Rac1 with the COOH-terminal polybasic motif of Rac1 (7) being responsible for the direct interaction with PA. Therefore, it is possible that S1P, like PA, may directly bind to target proteins such as Rac1 (7) and induce dissociation of the guanine nucleotide inhibitor from Rac1 for activation and redistribution to cell periphery (44, 46).
In summary, our results presented here demonstrate a role for intracellular S1P in the elevation of intracellular calcium levels, activation of MAPKs and Rac1/IQGAP1, redistribution of cytoskeletal, focal adhesion and tight junction proteins, and modulation of endothelial barrier function. These results support the notion that intracellular S1P can modulate signaling pathways independent of S1P receptors in the endothelium.
This work was supported by grants from National Institutes of Health R37 HL 079396 and P01 HL 098050 to V. Natarajan.
No conflicts of interest, financial or otherwise are declared by the authors.
- Copyright © 2011 the American Physiological Society