Sphingosine-1-Phosphate Prevents Tumor Necrosis Factor-–Mediated Monocyte Adhesion to Aortic Endothelium in Mice
http://www.100md.com
动脉硬化血栓血管生物学 2005年第5期
From the Division of Endocrinology & Metabolism and Cardiovascular Research Center (D.T.B., S.S., M.E.H., A.W., N.F., C.C.H.), Department of Chemistry (J.J.C., T.L.M.), Department of Biochemistry & Molecular Genetics (M.D.D.), and Department of Pharmacology (K.W.K., K.R.L., C.C.H.), University of Virginia, Charlottesville, Va; Division of Cardiology (P.S.T.), Stanford University, Palo Alto, Calif.
Correspondence to Catherine C Hedrick, PhD, Cardiovascular Research Center, University of Virginia, 415 Lane Rd, MR5 G123, Charlottesville, VA 22908. E-mail cch6n@virginia.edu
Abstract
Objective— Endothelial activation and monocyte adhesion to endothelium are key events in inflammation. Sphingosine-1-phosphate (S1P) is a sphingolipid that binds to G protein-coupled receptors on endothelial cells (ECs). We examined the role of S1P in modulating endothelial activation and monocyte–EC interactions in vivo.
Methods and Results— We injected C57BL/6J mice intravenously with tumor necrosis factor (TNF)- in the presence and absence of the S1P1 receptor agonist SEW2871 and examined monocyte adhesion. Aortas from TNF-–injected mice had a 4-fold increase in the number of monocytes bound, whereas aortas from TNF- plus SEW2871-treated mice had few monocytes bound (P<0.0001). Using siRNA, we found that inhibiting the S1P1 receptor in vascular ECs blocked the ability of S1P to prevent monocyte–EC interactions in response to TNF-. We examined signaling pathways downstream of S1P1 and found that 100 nM S1P increased phosphorylation of Akt and decreased activation of c-jun.
Conclusions— Thus, we provide the first evidence that S1P signaling through the endothelial S1P1 receptor protects the vasculature against TNF-–mediated monocyte–EC interactions in vivo.
We examined the role of sphingosine-1-phosphate (S1P) in modulating monocyte–endothelial cell (EC) interactions. S1P and a specific S1P1 receptor agonist prevented monocyte adhesion to aorta in tumor necrosis factor (TNF)-–treated mice in vivo. We provide the first evidence to our knowledge that S1P signaling through the endothelial S1P1 receptor protects the vasculature against TNF-–mediated monocyte–EC interactions in vivo.
Key Words: endothelium ? sphingosine-1-phosphate ? inflammation ? Endothelium differentiation gene (Edg) receptors
Introduction
Inflammation is a hallmark of atherosclerosis and diabetes.1 Monocyte–endothelial interactions are key initiating events of inflammation.2 Activated monocytes release tumor necrosis factor (TNF)- that mediates a variety of pathological vascular responses.3 Monoclonal antibody therapies to reduce TNF- have reduced inflammation in several chronic diseases,4 but these therapies do not prevent the initiation of inflammation.
Sphingosine-1-phosphate (S1P) is a biologically active sphingolipid in circulation that evokes a variety of cellular responses, including cell migration, vascular maturation, and lymphocyte homing.5 However, the role of S1P in vascular inflammation is unknown. S1P binds to 5 G protein-coupled receptors (Edg receptors) on endothelium and activates these receptors in the low nanomolar range.6,7 In the vasculature, S1P receptors are found on both endothelial cells (ECs) and monocytes.8,9 S1P1 is involved in endothelial migration and blood vessel formation.10 S1P3 regulates heart rate in mice11 and vascular tone through activation of endothelial nitric oxide synthase (eNOS).12,13
In the current study, we report that S1P prevents TNF-–mediated monocyte adhesion to intact aorta and to cultured primary isolates of aortic endothelial cells. We found that the protective, antiinflammatory action of S1P is primarily mediated through binding to the endothelial S1P1 receptor.
Methods
Detailed Methods for all experiments can be found online. Please see http://atvb.ahajournals.org.
Results
S1P and an S1P1 Receptor Agonist Prevent TNF-Mediated Monocyte–Endothelial Interactions
We injected C57BL/6J (B6) mice with 0.5 μg of recombinant murine TNF- intravenously, and either 5 mg/kg of the selective S1P1 receptor agonist, SEW2871 (5-(4-Phenyl-5-trifluoromethylthiophen-2-yl)-3-(3-trifluoromethylphenyl)-1,2,4-oxadiazole), or saline plus 0.2% fatty acid-free bovine serum albumin as a vehicle control. SEW2871 is a selective agonist for S1P111 and is 30-fold less potent than S1P at S1P1, with no agonist activity at S1P2 or S1P3 at concentrations up to 10 micromolar.11 At 2 hours after injection, aortas were harvested. Subgroups of aortas received 100 ng/mL pertussis toxin (PTX) to uncouple S1P1/S1P3 Gi receptor signaling.14 Some aortas were treated for 4 hours with 10 μmol/L VPC23019 a S1P receptor antagonist with a Ki for S1P1 of 25 nM; the molecule is 50-fold less potent in blocking S1P3, but is inactive at S1P2.15 Aortas were incubated with fluorescently labeled mouse monocytes as previously described.16 Injection of SEW2871 entirely prevented TNF-–mediated monocyte–EC adhesion in mice (+TNF+SEW2871; Figure 1a), suggesting that S1P activation of the endothelial S1P1 receptor regulates monocyte–endothelial adhesion. The notion that S1P1 is the receptor regulating monocyte–endothelial interactions in response to S1P is further strengthened by the fact that both the S1P1 receptor antagonist VPC23019and PTX completely reversed the inhibitory action of SEW2871 on monocyte adhesion to aorta. We did observe that PTX alone caused a slight increase in monocyte adhesion to aorta (Figure 1a). However, by phase microscopy we did not observe gross changes in EC morphology in response to PTX.
Figure 1. S1P prevents TNF-mediated monocyte:EC interactions in vivo. A, C57BL/6J mice were injected intravenously with saline+0.2%FAFBSA (saline), 500 ng murine TNF for 2 hours (+TNF), or with 500 ng murine TNF plus 5 mg/kg SEW2871, a S1P1 agonist for 2 hours (+TNF+SEW2871). All aortas were harvested and incubated overnight in DMEM containing 1% heat-inactivated FBS in the absence or presence of 100ng/ml pertussis toxin (+PTX) to uncouple Gi signaling. Subgroups of aortas were also incubated ex vivo for 4 hours with 10 μM VPC23019(+VPC23019), an S1P1 receptor antagonist, immediately before performing a monocyte adhesion assay directly on the aortic tissue. *Significantly higher than saline control, P<0.0001; ** significantly lower than TNF, P<0.0001; #significantly higher than saline control, P<0.05 by ANOVA. Mean ±SE of 5 mice per group. B, C57BL/6J mice were injected intravenously with saline+0.2%FAFBSA (saline), or with 500 ng murine TNF for 2h (+TNF). Aortas were harvested and incubated overnight in DMEM containing 1% heat-inactivated FBS in the absence or presence of 100 ng/ml pertussis toxin (+PTX) to uncouple Gi signaling. Aortas were incubated 4 hours ex vivo with 100 nM S1P (+S1P). Subgroups of aortas were also incubated for 4 hours ex vivo with 10 μM VPC23019(+VPC23019), an S1P1 receptor antagonist, or with 100 nM SEW2871 (+SEW2871), immediately before performing a monocyte adhesion assay directly on the aortic tissue. A subgroup of aortas was also incubated with 100 nM 12SHETE (+12SHETE) as a lipid control. *Significantly higher than saline control, P<0.0001; ** significantly lower than TNF, P<0.0001; #significantly higher than saline control, P<0.05 by ANOVA. Mean ±SE of 5 mice per group. C, Representative images from aorta of C57BL/6J mice injected IV with 500 ng recombinant murine TNF (left) or with 500 ng recombinant murine TNF plus 5 mg/kg SEW2871 for 2h (right). After injections, aortas were harvested and immediately incubated with fluorescently-labeled monocytes at 37°C for 30 minutes. Images show one representative aorta per experimental group. Fluorescent monocytes can be seen adhering to aortic endothelium. Note the reduction in the number of monocytes bound to the aorta in which the mouse has received an IV injection of the S1P1 agonist SEW2871 (right).
We performed a series of parallel studies using 100 nM S1P ex vivo. A group of mice were injected with 0.5 μg TNF-, and the aortas were harvested and incubated ex vivo with 100 nM S1P for 4 hours before incubation with mouse monocytes. We did not inject a bolus amount of S1P in vivo because S1P injection intravenously is toxic to mice.17 However, incubation of the aorta ex vivo with 100 nM S1P also completely prevented TNF-–stimulated monocyte adhesion to endothelium (+TNF+S1P; Figure 1b). Again, treatment of aorta with PTX or with VPC23019blocked the antiinflammatory action of S1P (Figure 1b). We incubated one group of aortas ex vivo with SEW2871 for 4 hours as a direct comparison to S1P. SEW2871 worked similarly to S1P (Figure 1b). There appeared to be no difference in SEW2871 action whether it was added ex vivo (as in panel b) or injected in vivo (as in panel a). As a lipid control, we incubated aortas in 100 nM 12SHETE, an eicosanoid product of the 12/15-lipoxygenase pathway that modulates monocyte adhesion to endothelium,18 and saw no change in TNF-–mediated monocyte adhesion. Representative images of monocytes adhering to aorta from TNF-–treated mice and TNF- plus SEW2871-treated mice are shown in Figure 1c. Taken together, these data illustrate that S1P is potently antiinflammatory in the nanomolar range, and that S1P1 is the receptor that is targeted by S1P to inhibit monocyte adhesion to endothelium.
S1P Blocks Monocyte Adhesion to Endothelial Cells
Murine aortic endothelial cells (MAECs) were incubated with several concentrations of S1P in the presence of TNF-. TNF-–treated MAECs bound 20±3 monocytes/field. Incubation of TNF-–treated MAECs with 1 nM S1P for 4 hours reduced TNF-–mediated adhesion by only 10%, whereas incubation of MAECs with 10 nM S1P for 4 hours reduced TNF-–mediated monocyte adhesion by 90%. S1P at 100 nM concentration completely blocked TNF-–mediated monocyte adhesion to that observed for untreated cells (5±1 monocytes bound/field). Addition of 1 to 100 nM S1P in the absence of TNF- as a control did not change adhesion, with an average of 5±1 monocytes bound/field. Thus, we used 100 nM S1P for 4 hours for our studies in vitro.
We used a parallel plate flow chamber system in which monocytes were allowed to flow over a confluent monolayer of MAECs at a shear stress of 0.75 dynes/cm2. This flow rate is the maximum rate that allowed for quantitative measurement of monocyte adhesion in our system. S1P pretreatment of MAECs (100 nM for 4 hours) reduced monocyte adhesion to TNF-–activated endothelium under flow conditions by 95% (Figure 2). Taken together, our data illustrate that S1P at physiological concentrations is potently antiinflammatory and prevents monocyte adhesion to endothelium in response to TNF-.
Figure 2. S1P prevents monocyte adhesion to ECs under conditions of flow. MAECs at passage 2 from C57BL/6J mice were treated for 4 hours with 10 U/ml murine TNF (+TNF), or TNF plus 100 nM S1P (+TNF+S1P) and then used in a flow chamber assay. Cells were incubated in media containing 1% heat-inactivated FBS (No treatment) or plus 100 nM S1P (+ S1P alone) as controls. In the flow chamber, WEHI 78/24 mouse monocytes (1.0x106/ml) in M199 containing 1% heat-inactivated FBS were allowed to flow over confluent monolayers of MAECs at a shear stress of 0.75dynes/cm2 for 5 minutes in a parallel plate flow chamber. The number of monocytes that firmly adhered during the 5-minute flow assay were counted. Data represent the mean±SE of 3 experiments performed in duplicate. *Significantly higher than the no treatment control, P<0.0001; **significantly lower than TNF, P<0.0001 by ANOVA.
S1P Modulation of Endothelial Chemokines and Adhesion Molecules
Endothelial vascular cell adhesion molecule (VCAM)-1, intercellular adhesion molecule-1 (ICAM-1), and E-selectin mediate monocyte adhesion to endothelium. TNF- significantly upregulated expression of these 3 adhesion molecules on MAECs, yet treatment of MAECs for 4 hours with 100 nM S1P did not significantly modify expression of any of these adhesion molecules (data not shown).
JE/MCP-1 and KC are pro-inflammatory chemokines that promote monocyte–endothelial interactions.19,20 Levels of KC (the mouse ortholog of interleukin [IL]-8) and JE/MCP-1 mRNA were measured using quantitative polymerase chain reaction. TNF- (10 U/mL for 4 hours) significantly increased mRNA expression of both KC and JE/MCP-1 20-fold and 40-fold, respectively (Figure Ia, available online at http://atvb.ahajournals.org). S1P (100 nM for 4 hours) dramatically reduced KC and JE/MCP-1 mRNA to levels observed for untreated MAECs (Figure Ia). TNF- significantly increased KC and JE/MCP-1 secretion from MAECs by several fold (Figure Ib). Interestingly, at 4 hours, protein levels of KC were reduced only 10% by 100 nM S1P, and JE/MCP-1 levels were unchanged (Figure Ib). We did not examine levels of chemokines secreted by MAECs after 4 hours. Thus, within the 4-hour timeframe of our study, S1P does not alter the TNF-–mediated increase in adhesion molecule expression on endothelium, and only modestly impacts secretion of KC.
Identification of S1P1 as an Antiinflammatory Receptor Pathway in Endothelium
Mouse aortic ECs express mRNA for only S1P1, S1P2, and S1P3 (Figure II, available online at http://atvb.ahajournals.org). We have found expression of only S1P1, S1P2, and S1P3 in human aortic ECs (data not shown). We transfected human umbilical vein ECs because of their ease of transfection with siRNA to target S1P1 and S1P3; transfection rates approached 95%. S1P1-targeted siRNA, but not S1P3 siRNA, significantly reduced S1P action by 70% (Figure 3). A scrambled siRNA control had no effect. We confirmed these data using primary human monocytes (data not shown). Expression of target S1P1 mRNA was decreased by 70% and target S1P3 mRNA by 55% (Figure III, available online at http://atvb.ahajournals.org). Furthermore, the siRNA constructs were specific for their target mRNAs (Figure III) because there was no decrease in expression of other S1P receptors. Taken together, these data indicate that S1P1 is the primary endothelial S1P receptor that mediates the antiinflammatory effect of S1P.
Figure 3. Identification of S1P1 receptor as antiinflammatory in ECs. HUVEC were transfected with control scrambled siRNA (siSCR), S1P1 siRNA (siS1P1), or S1P3 siRNA (siS1P3) as described in Materials and Methods. At 48 hours after transfection, HUVECs were treated with 10 U/ml TNF (+TNF) or TNF plus 100 nM S1P (+TNF+S1P) for 4 hours at 37°C. HUVECs were then used in a static monocyte adhesion assay. No treatment: untransfected control HUVEC maintained in media. + S1P: 100 nM S1P alone as a control. *Significantly higher than the no treatment control, P<0.0001; **significantly lower than the TNF value for each corresponding siRNA, P<0.001; $significantly lower than siS1P3+TNF, P<0.01 by ANOVA. Data represent the mean±SE of 8 experiments.
Downstream Signaling of S1P
TNF- activates both c-jun and NFB in endothelium.21 Interestingly, eNOS inhibits endothelial IL-8 production by blocking JNK-mediated activation of AP-1.22 Treatment of human umbilical vein ECs with 100 nM S1P did not reduce TNF-–mediated activation of NFB in endothelium (data not shown). However, S1P prevented TNF-–mediated activation of a c-jun-luciferase reporter plasmid (Figure 4). S1P significantly increased phosphorylation of Akt (Figure 5). Akt phosphorylation by S1P upregulates eNOS activity.23 Our results suggest that the protective action of S1P on endothelium is caused, at least in part, by upregulation of Akt phosphorylation and decreased activation of c-jun.
Figure 4. S1P prevents c-Jun activation in primary endothelium. HUVECs were transfected with pFA2-c-jun plasmid for 48 hours. HUVECs were then treated for 2 hours with TNF (+TNF) in the absence or presence of 100 nM S1P (+S1P) for 30 minutes. No treatment: untransfected control HUVECs maintained in media. Cell lysates were obtained after incubation, and luciferase c-Jun transcriptional activity was assayed using a luminometer. *Significantly higher than no treatment, P<0.001, **significantly lower than TNF, P<0.0001 by ANOVA.
Figure 5. Phosphorylation of Akt by S1P. HUVECs were treated for 4 hours with 10 U/ml murine TNF (+TNF), or TNF plus 100 nM S1P (+TNF+S1P). No treatment: untransfected control HUVEC maintained in media. Total cell protein lysates were harvested as described in Materials and Methods. Samples were analyzed by gel electrophoresis and immunoblotting for total Akt and phosphorylated Akt. Bands were detected using chemiluminescence and were quantified using densitometry.
Discussion
Monocyte–endothelial interactions are initiating events in vascular inflammation. In the current study, we show that S1P prevents TNF-–mediated monocyte adhesion to endothelium. This is the first report to our knowledge that indicates that S1P binding to the S1P1 receptor on endothelium activates a potent antiinflammatory signaling cascade and prevents initiation of inflammation in aorta in response to TNF-. In that TNF- is a primary cytokine produced during inflammation, the action of the S1P1 receptor in the vasculature has broad therapeutic implications.
S1P receptor agonists modulate cell migration, lymphocyte homing, and vascular permeability. Recent reports have identified that FTY720, a nonselective S1P receptor agonist, modulates lymphocyte homing.24–26 FTY720 is a sphingosine analog that is rapidly phosphorylated in vivo to yield an active metabolite that is an agonist at 4 of 5 S1P receptors. FTY720 regulates vascular permeability, thereby protecting against acute lung injury in mice.27 However, SEW2871 is the first known specific agonist of the S1P1 receptor. SEW2871, like S1P, has been shown to modulate lymphocyte homing and influence cell migration.11 Our data support a new role for S1P1 agonists in mediating monocyte–EC interactions in inflammation. Possibly, S1P1 agonists exert antiinflammatory effects in monocytes as well as in endothelium. The role of S1P1 agonists, including SEW2871 and FTY720, in modulating monocyte trafficking to tissues is also an important concept that will be examined.
In the current study, we did not inject mice in vivo with a bolus amount of S1P. Bolus injection of S1P is toxic to mice, probably because of cardiovascular effects mediated through S1P3.17 Although slow infusion of S1P may be possible,28 we did not find this method relevant for studying acute responses to TNF- in vivo. We were unable to confirm downregulation of S1P1 receptor protein in response to siRNA because of lack of specific antibodies to S1P1 and S1P3. However, we did observe a 70% knockdown of S1P1 receptor mRNA, so we predict that the S1P1 receptor protein levels are also significantly reduced.
Our data point to S1P1 as being the primary endothelial S1P receptor that prevents monocyte adhesion to aorta. We made this conclusion based on results from a number of different assays using pharmacological reagents as well as molecular siRNA. In Figure 3, we found a small, yet significant, effect of the S1P3 siRNA on TNF-–mediated monocyte adhesion. We did not have specific S1P3 pharmacological reagents for our studies. Although the bulk of our data point to S1P1 as the primary S1P receptor preventing TNF-–mediated monocyte–EC interactions, we cannot absolutely rule out some contribution of S1P3. This is especially important in light of the fact that S1P3, like S1P1,23 has recently been shown to regulate eNOS activity.13 We will examine the role of S1P3 in regulating these early inflammatory processes.
The mechanisms by which S1P exerts its antiinflammatory effect are still unclear. We examined expression of adhesion molecules on endothelium in response to S1P. TNF- upregulates expression of VCAM-1 and ICAM-1 on the endothelial surface.29,30 S1P at 100 nM did not reduce the TNF-–mediated increase in endothelial VCAM-1 or ICAM-1 expression. Previous studies31,32 reported that S1P induced endothelial VCAM-1 and E-selectin expression. These investigators used very high concentrations of S1P (1 to 20 μmol/L). We recapitulated this finding, but only when S1P was added to ECs at concentrations >5 μmol/L (data not shown). In the bloodstream, physiological levels of free S1P are in the nanomolar range6,7 and the reported Kd values for S1P binding to the S1P1 receptor are 1 to 10 nanomolar. Thus, it is conceivable that very high concentrations of S1P act on ECs through some other signaling pathway. Xia et al reported that S1P at concentrations >1 μmol/L activated NFB.31 These investigators also reported that TNF- activated sphingosine kinase in endothelium (the enzyme that generates S1P from sphingosine), causing increased production of S1P that leads to erk, c-jun, and NFB activation. We also observed NFB activation in endothelium by S1P concentrations of 1 μmol/L or greater (data not shown). However, at concentrations of S1P <500 nM, we found no activation of NFB, and a decrease in c-jun activation in response to TNF- stimulation of endothelium. Again, because we are able to recapitulate these investigators’ findings when using micromolar concentrations of S1P, we believe that the concentration of S1P in circulation is quite important for preventing inflammatory processes. Thus, the antiinflammatory action of S1P is through nanomolar concentrations of S1P primarily activating S1P1 on endothelium. Xia et al reported that pertussis toxin (to uncouple Gi signaling of G protein-coupled receptors (GPCRs) as shown in Figure 1) did not inhibit the effect of S1P on adhesion molecule expression, suggesting that S1P at higher concentrations acts intracellularly on some non-S1P receptor pathway.31 However, we did not measure sphingosine kinase activation in the current study, so we do not know if injection of TNF- in vivo increased sphingosine kinase activity. However, in light of our data supporting a decrease in monocyte adhesion in response to S1P, this would seem unlikely.
A second mechanism of action of S1P on endothelium could be to reduce endothelial production of pro-inflammatory chemokines, including IL-8 and MCP-1. We found that TNF- increased endothelial production of both JE/MCP-1 and KC. S1P modestly reduced protein expression of KC (by 10%), but not JE/MCP-1. We did not examine chemokine expression at later time points. IL-8/KC has been shown to be regulated by NFB, p38 MAP kinase, and by Jun kinase pathways through activation of the transcription factor AP-1. We have previously reported endothelial IL-8 activation by p38 MAP kinase and AP-1 activation in diabetes.20 Activation of these signaling pathways in endothelium may be the result of increased oxidative stress in response to TNF-. We did not examine p38 activation by TNF- and S1P in the current study in that inhibition of p38 MAP kinase by 5 μmol/L SB203580 had no impact on S1P action in a monocyte adhesion assay (data not shown). However, we did find that S1P reduced activation of Jun kinase (Figure 4). Thus, our data suggest that endothelial KC/IL-8 mRNA levels may be reduced by S1P through inhibition of JNK and AP-1. Studies are ongoing to further investigate this finding.
Garcia et al have recently reported that 500 nM S1P enhanced endothelial barrier function in the pulmonary vasculature.33 This enhanced barrier protection in EC was caused by a tighter rearrangement of the cytoskeleton that was mediated by S1P. It is unclear whether enhanced barrier function in EC prevents monocyte–EC interactions in the vessel wall, although a correlation is quite possible. Sessa et al have shown that S1P regulates Akt and eNOS expression.23 Thus, signaling mechanisms downstream of S1P1 that include activation of PI3 kinase and Akt may be protective against monocyte–endothelial interactions. Indeed, reports have suggested that eNOS prevents leukocyte–endothelial interactions and may enhance barrier function.34,35 Additional evidence that points to a role for eNOS in S1P action in endothelium is a report by Natarajan et al that suggested that nitric oxide production by the endothelium prevented Jun kinase-mediated activation of AP-1.22 Michel et al have also reported that S1P caused vasodilation in rat arteries through rapid phosphorylation and activation of eNOS.12 Studies are underway in the laboratory to examine the possible role of eNOS in the antiinflammatory action of S1P on monocyte–EC interactions.
In conclusion, we have identified that S1P signaling to the S1P1 receptor on EC protects the vasculature against TNF-–mediated monocyte–EC interactions. Identification of this potent antiinflammatory action of S1P and understanding its signaling mechanism may be of great therapeutic benefit for a number of inflammatory diseases.
Acknowledgments
The authors thank Melissa A Marshall and Jeremy Mauldin (University of Virginia) for technical assistance with mouse studies. The authors also thank Dr Brett Blackman (University of Virginia) for the gift of human umbilical vein endothelial cells, and Dr Erik Hewlett (University of Virginia) for the gift of PTX.
This work was supported by grants from the Juvenile Diabetes Research Foundation (C.C.H.), National Institutes of Health R01 HL079621 (C.C.H.), National Institutes of Health R01 GM067958 (K.R.L.), and National Institutes of Health F31 GM064101 (M.D.D.).
References
Ross R. Atherosclerosis–an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Berliner JA, Navab M, Fogelman AM, Frank JS, Demer LL, Edwards PA, Watson AD, Lusis AJ. Atherosclerosis: basic mechanisms. Oxidation, inflammation, and genetics. Circulation. 1995; 91: 2488–2496.
Aggarwal BB, Shishodia S, Ashikawa K, Bharti AC. The role of TNF and its family members in inflammation and cancer: lessons from gene deletion. Curr Drug Targets Inflamm Allergy. 2002; 1: 327–341.
Reimold AM. New indications for treatment of chronic inflammation by TNF-alpha blockade. Am J Med Sci. 2003; 325: 75–92.
Spiegel S, Milstien S. Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol. 2003; 4: 397–407.
Murata N, Sato K, Kon J, Tomura H, Yanagita M, Kuwabara A, Ui M, Okajima F. Interaction of sphingosine 1-phosphate with plasma components, including lipoproteins, regulates the lipid receptor-mediated actions. Biochem J. 2000; 352 (Pt 3): 809–815.
Kimura T, Sato K, Malchinkhuu E, Tomura H, Tamama K, Kuwabara A, Murakami M, Okajima F. High-density lipoprotein stimulates endothelial cell migration and survival through sphingosine 1-phosphate and its receptors. Arterioscler Thromb Vasc Biol. 2003; 23: 1283–1288.
Fueller M, Wang dA, Tigyi G, Siess W. Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate. Cell Signal. 2003; 15: 367–375.
Kluk MJ, Hla T. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochim Biophys Acta. 2002; 1582: 72–80.
Liu Y, Wada R, Yamashita T, Mi Y, Deng CX, Hobson JP, Rosenfeldt HM, Nava VE, Chae SS, Lee MJ, Liu CH, Hla T, Spiegel S, Proia RL. Edg-1, the G protein-coupled receptor for sphingosine-1-phosphate, is essential for vascular maturation. J Clin Invest. 2000; 106: 951–961.
Sanna MG, Liao J, Jo E, Alfonso C, Ahn MY, Peterson MS, Webb B, Lefebvre S, Chun J, Gray N, Rosen H. Sphingosine 1-phosphate (S1P) receptor subtypes S1P1 and S1P3, respectively, regulate lymphocyte recirculation and heart rate. J Biol Chem. 2004; 279: 13839–13848.
Dantas AP, Igarashi J, Michel T. Sphingosine 1-phosphate and control of vascular tone. Am J Physiol Heart Circ Physiol. 2003; 284: H2045–H2052.
Nofer JR, van der GM, Tolle M, Wolinska I, von Wnuck LK, Baba HA, Tietge UJ, Godecke A, Ishii I, Kleuser B, Schafers M, Fobker M, Zidek W, Assmann G, Chun J, Levkau B. HDL induces NO-dependent vasorelaxation via the lysophospholipid receptor S1P3. J Clin Invest. 2004; 113: 569–581.
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 G(i), G(q), and G(12) families of heterotrimeric G proteins. J Biol Chem. 1999; 274: 27351–27358.
Davis MD, Clemens JJ, Macdonald TL, Lynch KR. Sphingosine 1-phosphate analogs as receptor antagonists. J Biol Chem. 2005; 280: 9833–9841.
McEvoy LM, Jutila MA, Tsao PS, Cooke JP, Butcher EC. Anti-CD43 inhibits monocyte-endothelial adhesion in inflammation and atherogenesis. Blood. 1997; 90: 3587–3594.
Forrest M, Sun SY, Hajdu R, Bergstrom J, Card D, Hale J, Keohane CA, Meyers C, Milligan J, Mills S, Nomura N, Rosenbach M, Shei GJ, Singer II, Tian M, West S, White V, Xie J, Rosen H, Proia R, Doherty G, Mandala S. Immune Cell Regulation and Cardiovascular Effects of Sphingosine 1-Phosphate Receptor Agonists in Rodents are Mediated via Distinct Receptor Sub-Types. J Pharmacol Exp Ther. 2004; 758–768.
Patricia MK, Kim JA, Harper CM, Shih PT, Berliner JA, Natarajan R, Nadler JL, Hedrick CC. Lipoxygenase products increase monocyte adhesion to human aortic endothelial cells. Arterioscler Thromb Vasc Biol. 1999; 19: 2615–2622.
Lee H, Shi W, Tontonoz P, Wang S, Subbanagounder G, Hedrick CC, Hama S, Borromeo C, Evans RM, Berliner JA, Nagy L. Role for peroxisome proliferator-activated receptor alpha in oxidized phospholipid-induced synthesis of monocyte chemotactic protein-1 and IL-8 by endothelial cells. Circ Res. 2000; 87: 516–521.
Srinivasan S, Bolick DT, Hatley ME, Natarajan R, Reilly KB, Yeh M, Chrestensen C, Sturgill TW, Hedrick CC. Glucose regulates IL-8 production in aortic endothelial cells through activation of the p38 mitogen-activated protein kinase pathway in diabetes. J Biol Chem. 2004; 279: 31930–31936.
Lerner-Marmarosh N, Yoshizumi M, Che W, Surapisitchat J, Kawakatsu H, Akaike M, Ding B, Huang Q, Yan C, Berk BC, Abe J. Inhibition of tumor necrosis factor--induced SHP-2 phosphatase activity by shear stress: a mechanism to reduce endothelial inflammation. Arterioscler Thromb Vasc Biol. 2003; 23: 1775–1781.
Natarajan R, Gupta S, Fisher BJ, Ghosh S, Fowler AA III. Nitric oxide suppresses IL-8 transcription by inhibiting c-Jun N- terminal kinase-induced AP-1 activation. Exp Cell Res. 2001; 266: 203–212.
Morales-Ruiz M, Lee MJ, Zollner S, Gratton JP, Scotland R, Shiojima I, Walsh K, Hla T, Sessa WC. Sphingosine 1-phosphate activates Akt, nitric oxide production, and chemotaxis through a Gi protein/phosphoinositide 3-kinase pathway in endothelial cells. J Biol Chem. 2001; 276: 19672–19677.
Xie JH, Nomura N, Koprak SL, Quackenbush EJ, Forrest MJ, Rosen H. Sphingosine-1-phosphate receptor agonism impairs the efficiency of the local immune response by altering trafficking of naive and antigen-activated CD4+ T cells. J Immunol. 2003; 170: 3662–3670.
Brinkmann V, Davis MD, Heise CE, Albert R, Cottens S, Hof R, Bruns C, Prieschl E, Baumruker T, Hiestand P, Foster CA, Zollinger M, Lynch KR. The immune modulator FTY720 targets sphingosine 1-phosphate receptors. J Biol Chem. 2002; 277: 21453–21457.
Matloubian M, Lo CG, Cinamon G, Lesneski MJ, Xu Y, Brinkmann V, Allende ML, Proia RL, Cyster JG. Lymphocyte egress from thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature. 2004; 427: 355–360.
Peng X, Hassoun PM, Sammani S, McVerry BJ, Burne MJ, Rabb H, Pearse D, Tuder RM, Garcia JG. Protective Effects of Sphingosine 1-Phosphate in Murine Endotoxin-Induced Inflammatory Lung Injury. Am J Respir Crit Care Med. 2004; 1245–1251.
Mandala S, Hajdu R, Bergstrom J, Quackenbush E, Xie J, Milligan J, Thornton R, Shei GJ, Card D, Keohane C, Rosenbach M, Hale J, Lynch CL, Rupprecht K, Parsons W, Rosen H. Alteration of lymphocyte trafficking by sphingosine-1-phosphate receptor agonists. Science. 2002; 296: 346–349.
Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J. 1995; 9: 899–909.
VandenBerg E, Reid MD, Edwards JD, Davis HW. The role of the cytoskeleton in cellular adhesion molecule expression in tumor necrosis factor-stimulated endothelial cells. J Cell Biochem. 2004; 91: 926–937.
Xia P, Gamble JR, Rye KA, Wang L, Hii CS, Cockerill P, Khew-Goodall Y, Bert AG, Barter PJ, Vadas MA. Tumor necrosis factor-alpha induces adhesion molecule expression through the sphingosine kinase pathway. Proc Natl Acad Sci U S A. 1998; 95: 14196–14201.
Rizza C, Leitinger N, Yue J, Fischer DJ, Wang DA, Shih PT, Lee H, Tigyi G, Berliner JA. Lysophosphatidic acid as a regulator of endothelial/leukocyte interaction. Lab Invest. 1999; 79: 1227–1235.
Dudek SM, Jacobson JR, Chiang ET, Birukov KG, Wang P, Zhan X, Garcia JG. Pulmonary endothelial cell barrier enhancement by sphingosine 1-phosphate: roles for cortactin and myosin light chain kinase. J Biol Chem. 2004; 279: 24692–24700.
Hayashi T, Sumi D, Juliet PA, Matsui-Hirai H, Asai-Tanaka Y, Kano H, Fukatsu A, Tsunekawa T, Miyazaki A, Iguchi A, Ignarro LJ. Gene transfer of endothelial NO synthase, but not eNOS plus inducible NOS, regressed atherosclerosis in rabbits. Cardiovasc Res. 2004; 61: 339–351.
Kuhlencordt PJ, Rosel E, Gerszten RE, Morales-Ruiz M, Dombkowski D, Atkinson WJ, Han F, Preffer F, Rosezweig A, Sessa WC, Gimbrone MA, Ertl G, Huang PL. The role of endothelial nitric oxide synthase (eNOS) in endothelial activation: insights from eNOS-knockout endothelial cells. Am J Physiol Cell Physiol. 2004; C1195–C1202.(David T. Bolick; Suseela )
Correspondence to Catherine C Hedrick, PhD, Cardiovascular Research Center, University of Virginia, 415 Lane Rd, MR5 G123, Charlottesville, VA 22908. E-mail cch6n@virginia.edu
Abstract
Objective— Endothelial activation and monocyte adhesion to endothelium are key events in inflammation. Sphingosine-1-phosphate (S1P) is a sphingolipid that binds to G protein-coupled receptors on endothelial cells (ECs). We examined the role of S1P in modulating endothelial activation and monocyte–EC interactions in vivo.
Methods and Results— We injected C57BL/6J mice intravenously with tumor necrosis factor (TNF)- in the presence and absence of the S1P1 receptor agonist SEW2871 and examined monocyte adhesion. Aortas from TNF-–injected mice had a 4-fold increase in the number of monocytes bound, whereas aortas from TNF- plus SEW2871-treated mice had few monocytes bound (P<0.0001). Using siRNA, we found that inhibiting the S1P1 receptor in vascular ECs blocked the ability of S1P to prevent monocyte–EC interactions in response to TNF-. We examined signaling pathways downstream of S1P1 and found that 100 nM S1P increased phosphorylation of Akt and decreased activation of c-jun.
Conclusions— Thus, we provide the first evidence that S1P signaling through the endothelial S1P1 receptor protects the vasculature against TNF-–mediated monocyte–EC interactions in vivo.
We examined the role of sphingosine-1-phosphate (S1P) in modulating monocyte–endothelial cell (EC) interactions. S1P and a specific S1P1 receptor agonist prevented monocyte adhesion to aorta in tumor necrosis factor (TNF)-–treated mice in vivo. We provide the first evidence to our knowledge that S1P signaling through the endothelial S1P1 receptor protects the vasculature against TNF-–mediated monocyte–EC interactions in vivo.
Key Words: endothelium ? sphingosine-1-phosphate ? inflammation ? Endothelium differentiation gene (Edg) receptors
Introduction
Inflammation is a hallmark of atherosclerosis and diabetes.1 Monocyte–endothelial interactions are key initiating events of inflammation.2 Activated monocytes release tumor necrosis factor (TNF)- that mediates a variety of pathological vascular responses.3 Monoclonal antibody therapies to reduce TNF- have reduced inflammation in several chronic diseases,4 but these therapies do not prevent the initiation of inflammation.
Sphingosine-1-phosphate (S1P) is a biologically active sphingolipid in circulation that evokes a variety of cellular responses, including cell migration, vascular maturation, and lymphocyte homing.5 However, the role of S1P in vascular inflammation is unknown. S1P binds to 5 G protein-coupled receptors (Edg receptors) on endothelium and activates these receptors in the low nanomolar range.6,7 In the vasculature, S1P receptors are found on both endothelial cells (ECs) and monocytes.8,9 S1P1 is involved in endothelial migration and blood vessel formation.10 S1P3 regulates heart rate in mice11 and vascular tone through activation of endothelial nitric oxide synthase (eNOS).12,13
In the current study, we report that S1P prevents TNF-–mediated monocyte adhesion to intact aorta and to cultured primary isolates of aortic endothelial cells. We found that the protective, antiinflammatory action of S1P is primarily mediated through binding to the endothelial S1P1 receptor.
Methods
Detailed Methods for all experiments can be found online. Please see http://atvb.ahajournals.org.
Results
S1P and an S1P1 Receptor Agonist Prevent TNF-Mediated Monocyte–Endothelial Interactions
We injected C57BL/6J (B6) mice with 0.5 μg of recombinant murine TNF- intravenously, and either 5 mg/kg of the selective S1P1 receptor agonist, SEW2871 (5-(4-Phenyl-5-trifluoromethylthiophen-2-yl)-3-(3-trifluoromethylphenyl)-1,2,4-oxadiazole), or saline plus 0.2% fatty acid-free bovine serum albumin as a vehicle control. SEW2871 is a selective agonist for S1P111 and is 30-fold less potent than S1P at S1P1, with no agonist activity at S1P2 or S1P3 at concentrations up to 10 micromolar.11 At 2 hours after injection, aortas were harvested. Subgroups of aortas received 100 ng/mL pertussis toxin (PTX) to uncouple S1P1/S1P3 Gi receptor signaling.14 Some aortas were treated for 4 hours with 10 μmol/L VPC23019 a S1P receptor antagonist with a Ki for S1P1 of 25 nM; the molecule is 50-fold less potent in blocking S1P3, but is inactive at S1P2.15 Aortas were incubated with fluorescently labeled mouse monocytes as previously described.16 Injection of SEW2871 entirely prevented TNF-–mediated monocyte–EC adhesion in mice (+TNF+SEW2871; Figure 1a), suggesting that S1P activation of the endothelial S1P1 receptor regulates monocyte–endothelial adhesion. The notion that S1P1 is the receptor regulating monocyte–endothelial interactions in response to S1P is further strengthened by the fact that both the S1P1 receptor antagonist VPC23019and PTX completely reversed the inhibitory action of SEW2871 on monocyte adhesion to aorta. We did observe that PTX alone caused a slight increase in monocyte adhesion to aorta (Figure 1a). However, by phase microscopy we did not observe gross changes in EC morphology in response to PTX.
Figure 1. S1P prevents TNF-mediated monocyte:EC interactions in vivo. A, C57BL/6J mice were injected intravenously with saline+0.2%FAFBSA (saline), 500 ng murine TNF for 2 hours (+TNF), or with 500 ng murine TNF plus 5 mg/kg SEW2871, a S1P1 agonist for 2 hours (+TNF+SEW2871). All aortas were harvested and incubated overnight in DMEM containing 1% heat-inactivated FBS in the absence or presence of 100ng/ml pertussis toxin (+PTX) to uncouple Gi signaling. Subgroups of aortas were also incubated ex vivo for 4 hours with 10 μM VPC23019(+VPC23019), an S1P1 receptor antagonist, immediately before performing a monocyte adhesion assay directly on the aortic tissue. *Significantly higher than saline control, P<0.0001; ** significantly lower than TNF, P<0.0001; #significantly higher than saline control, P<0.05 by ANOVA. Mean ±SE of 5 mice per group. B, C57BL/6J mice were injected intravenously with saline+0.2%FAFBSA (saline), or with 500 ng murine TNF for 2h (+TNF). Aortas were harvested and incubated overnight in DMEM containing 1% heat-inactivated FBS in the absence or presence of 100 ng/ml pertussis toxin (+PTX) to uncouple Gi signaling. Aortas were incubated 4 hours ex vivo with 100 nM S1P (+S1P). Subgroups of aortas were also incubated for 4 hours ex vivo with 10 μM VPC23019(+VPC23019), an S1P1 receptor antagonist, or with 100 nM SEW2871 (+SEW2871), immediately before performing a monocyte adhesion assay directly on the aortic tissue. A subgroup of aortas was also incubated with 100 nM 12SHETE (+12SHETE) as a lipid control. *Significantly higher than saline control, P<0.0001; ** significantly lower than TNF, P<0.0001; #significantly higher than saline control, P<0.05 by ANOVA. Mean ±SE of 5 mice per group. C, Representative images from aorta of C57BL/6J mice injected IV with 500 ng recombinant murine TNF (left) or with 500 ng recombinant murine TNF plus 5 mg/kg SEW2871 for 2h (right). After injections, aortas were harvested and immediately incubated with fluorescently-labeled monocytes at 37°C for 30 minutes. Images show one representative aorta per experimental group. Fluorescent monocytes can be seen adhering to aortic endothelium. Note the reduction in the number of monocytes bound to the aorta in which the mouse has received an IV injection of the S1P1 agonist SEW2871 (right).
We performed a series of parallel studies using 100 nM S1P ex vivo. A group of mice were injected with 0.5 μg TNF-, and the aortas were harvested and incubated ex vivo with 100 nM S1P for 4 hours before incubation with mouse monocytes. We did not inject a bolus amount of S1P in vivo because S1P injection intravenously is toxic to mice.17 However, incubation of the aorta ex vivo with 100 nM S1P also completely prevented TNF-–stimulated monocyte adhesion to endothelium (+TNF+S1P; Figure 1b). Again, treatment of aorta with PTX or with VPC23019blocked the antiinflammatory action of S1P (Figure 1b). We incubated one group of aortas ex vivo with SEW2871 for 4 hours as a direct comparison to S1P. SEW2871 worked similarly to S1P (Figure 1b). There appeared to be no difference in SEW2871 action whether it was added ex vivo (as in panel b) or injected in vivo (as in panel a). As a lipid control, we incubated aortas in 100 nM 12SHETE, an eicosanoid product of the 12/15-lipoxygenase pathway that modulates monocyte adhesion to endothelium,18 and saw no change in TNF-–mediated monocyte adhesion. Representative images of monocytes adhering to aorta from TNF-–treated mice and TNF- plus SEW2871-treated mice are shown in Figure 1c. Taken together, these data illustrate that S1P is potently antiinflammatory in the nanomolar range, and that S1P1 is the receptor that is targeted by S1P to inhibit monocyte adhesion to endothelium.
S1P Blocks Monocyte Adhesion to Endothelial Cells
Murine aortic endothelial cells (MAECs) were incubated with several concentrations of S1P in the presence of TNF-. TNF-–treated MAECs bound 20±3 monocytes/field. Incubation of TNF-–treated MAECs with 1 nM S1P for 4 hours reduced TNF-–mediated adhesion by only 10%, whereas incubation of MAECs with 10 nM S1P for 4 hours reduced TNF-–mediated monocyte adhesion by 90%. S1P at 100 nM concentration completely blocked TNF-–mediated monocyte adhesion to that observed for untreated cells (5±1 monocytes bound/field). Addition of 1 to 100 nM S1P in the absence of TNF- as a control did not change adhesion, with an average of 5±1 monocytes bound/field. Thus, we used 100 nM S1P for 4 hours for our studies in vitro.
We used a parallel plate flow chamber system in which monocytes were allowed to flow over a confluent monolayer of MAECs at a shear stress of 0.75 dynes/cm2. This flow rate is the maximum rate that allowed for quantitative measurement of monocyte adhesion in our system. S1P pretreatment of MAECs (100 nM for 4 hours) reduced monocyte adhesion to TNF-–activated endothelium under flow conditions by 95% (Figure 2). Taken together, our data illustrate that S1P at physiological concentrations is potently antiinflammatory and prevents monocyte adhesion to endothelium in response to TNF-.
Figure 2. S1P prevents monocyte adhesion to ECs under conditions of flow. MAECs at passage 2 from C57BL/6J mice were treated for 4 hours with 10 U/ml murine TNF (+TNF), or TNF plus 100 nM S1P (+TNF+S1P) and then used in a flow chamber assay. Cells were incubated in media containing 1% heat-inactivated FBS (No treatment) or plus 100 nM S1P (+ S1P alone) as controls. In the flow chamber, WEHI 78/24 mouse monocytes (1.0x106/ml) in M199 containing 1% heat-inactivated FBS were allowed to flow over confluent monolayers of MAECs at a shear stress of 0.75dynes/cm2 for 5 minutes in a parallel plate flow chamber. The number of monocytes that firmly adhered during the 5-minute flow assay were counted. Data represent the mean±SE of 3 experiments performed in duplicate. *Significantly higher than the no treatment control, P<0.0001; **significantly lower than TNF, P<0.0001 by ANOVA.
S1P Modulation of Endothelial Chemokines and Adhesion Molecules
Endothelial vascular cell adhesion molecule (VCAM)-1, intercellular adhesion molecule-1 (ICAM-1), and E-selectin mediate monocyte adhesion to endothelium. TNF- significantly upregulated expression of these 3 adhesion molecules on MAECs, yet treatment of MAECs for 4 hours with 100 nM S1P did not significantly modify expression of any of these adhesion molecules (data not shown).
JE/MCP-1 and KC are pro-inflammatory chemokines that promote monocyte–endothelial interactions.19,20 Levels of KC (the mouse ortholog of interleukin [IL]-8) and JE/MCP-1 mRNA were measured using quantitative polymerase chain reaction. TNF- (10 U/mL for 4 hours) significantly increased mRNA expression of both KC and JE/MCP-1 20-fold and 40-fold, respectively (Figure Ia, available online at http://atvb.ahajournals.org). S1P (100 nM for 4 hours) dramatically reduced KC and JE/MCP-1 mRNA to levels observed for untreated MAECs (Figure Ia). TNF- significantly increased KC and JE/MCP-1 secretion from MAECs by several fold (Figure Ib). Interestingly, at 4 hours, protein levels of KC were reduced only 10% by 100 nM S1P, and JE/MCP-1 levels were unchanged (Figure Ib). We did not examine levels of chemokines secreted by MAECs after 4 hours. Thus, within the 4-hour timeframe of our study, S1P does not alter the TNF-–mediated increase in adhesion molecule expression on endothelium, and only modestly impacts secretion of KC.
Identification of S1P1 as an Antiinflammatory Receptor Pathway in Endothelium
Mouse aortic ECs express mRNA for only S1P1, S1P2, and S1P3 (Figure II, available online at http://atvb.ahajournals.org). We have found expression of only S1P1, S1P2, and S1P3 in human aortic ECs (data not shown). We transfected human umbilical vein ECs because of their ease of transfection with siRNA to target S1P1 and S1P3; transfection rates approached 95%. S1P1-targeted siRNA, but not S1P3 siRNA, significantly reduced S1P action by 70% (Figure 3). A scrambled siRNA control had no effect. We confirmed these data using primary human monocytes (data not shown). Expression of target S1P1 mRNA was decreased by 70% and target S1P3 mRNA by 55% (Figure III, available online at http://atvb.ahajournals.org). Furthermore, the siRNA constructs were specific for their target mRNAs (Figure III) because there was no decrease in expression of other S1P receptors. Taken together, these data indicate that S1P1 is the primary endothelial S1P receptor that mediates the antiinflammatory effect of S1P.
Figure 3. Identification of S1P1 receptor as antiinflammatory in ECs. HUVEC were transfected with control scrambled siRNA (siSCR), S1P1 siRNA (siS1P1), or S1P3 siRNA (siS1P3) as described in Materials and Methods. At 48 hours after transfection, HUVECs were treated with 10 U/ml TNF (+TNF) or TNF plus 100 nM S1P (+TNF+S1P) for 4 hours at 37°C. HUVECs were then used in a static monocyte adhesion assay. No treatment: untransfected control HUVEC maintained in media. + S1P: 100 nM S1P alone as a control. *Significantly higher than the no treatment control, P<0.0001; **significantly lower than the TNF value for each corresponding siRNA, P<0.001; $significantly lower than siS1P3+TNF, P<0.01 by ANOVA. Data represent the mean±SE of 8 experiments.
Downstream Signaling of S1P
TNF- activates both c-jun and NFB in endothelium.21 Interestingly, eNOS inhibits endothelial IL-8 production by blocking JNK-mediated activation of AP-1.22 Treatment of human umbilical vein ECs with 100 nM S1P did not reduce TNF-–mediated activation of NFB in endothelium (data not shown). However, S1P prevented TNF-–mediated activation of a c-jun-luciferase reporter plasmid (Figure 4). S1P significantly increased phosphorylation of Akt (Figure 5). Akt phosphorylation by S1P upregulates eNOS activity.23 Our results suggest that the protective action of S1P on endothelium is caused, at least in part, by upregulation of Akt phosphorylation and decreased activation of c-jun.
Figure 4. S1P prevents c-Jun activation in primary endothelium. HUVECs were transfected with pFA2-c-jun plasmid for 48 hours. HUVECs were then treated for 2 hours with TNF (+TNF) in the absence or presence of 100 nM S1P (+S1P) for 30 minutes. No treatment: untransfected control HUVECs maintained in media. Cell lysates were obtained after incubation, and luciferase c-Jun transcriptional activity was assayed using a luminometer. *Significantly higher than no treatment, P<0.001, **significantly lower than TNF, P<0.0001 by ANOVA.
Figure 5. Phosphorylation of Akt by S1P. HUVECs were treated for 4 hours with 10 U/ml murine TNF (+TNF), or TNF plus 100 nM S1P (+TNF+S1P). No treatment: untransfected control HUVEC maintained in media. Total cell protein lysates were harvested as described in Materials and Methods. Samples were analyzed by gel electrophoresis and immunoblotting for total Akt and phosphorylated Akt. Bands were detected using chemiluminescence and were quantified using densitometry.
Discussion
Monocyte–endothelial interactions are initiating events in vascular inflammation. In the current study, we show that S1P prevents TNF-–mediated monocyte adhesion to endothelium. This is the first report to our knowledge that indicates that S1P binding to the S1P1 receptor on endothelium activates a potent antiinflammatory signaling cascade and prevents initiation of inflammation in aorta in response to TNF-. In that TNF- is a primary cytokine produced during inflammation, the action of the S1P1 receptor in the vasculature has broad therapeutic implications.
S1P receptor agonists modulate cell migration, lymphocyte homing, and vascular permeability. Recent reports have identified that FTY720, a nonselective S1P receptor agonist, modulates lymphocyte homing.24–26 FTY720 is a sphingosine analog that is rapidly phosphorylated in vivo to yield an active metabolite that is an agonist at 4 of 5 S1P receptors. FTY720 regulates vascular permeability, thereby protecting against acute lung injury in mice.27 However, SEW2871 is the first known specific agonist of the S1P1 receptor. SEW2871, like S1P, has been shown to modulate lymphocyte homing and influence cell migration.11 Our data support a new role for S1P1 agonists in mediating monocyte–EC interactions in inflammation. Possibly, S1P1 agonists exert antiinflammatory effects in monocytes as well as in endothelium. The role of S1P1 agonists, including SEW2871 and FTY720, in modulating monocyte trafficking to tissues is also an important concept that will be examined.
In the current study, we did not inject mice in vivo with a bolus amount of S1P. Bolus injection of S1P is toxic to mice, probably because of cardiovascular effects mediated through S1P3.17 Although slow infusion of S1P may be possible,28 we did not find this method relevant for studying acute responses to TNF- in vivo. We were unable to confirm downregulation of S1P1 receptor protein in response to siRNA because of lack of specific antibodies to S1P1 and S1P3. However, we did observe a 70% knockdown of S1P1 receptor mRNA, so we predict that the S1P1 receptor protein levels are also significantly reduced.
Our data point to S1P1 as being the primary endothelial S1P receptor that prevents monocyte adhesion to aorta. We made this conclusion based on results from a number of different assays using pharmacological reagents as well as molecular siRNA. In Figure 3, we found a small, yet significant, effect of the S1P3 siRNA on TNF-–mediated monocyte adhesion. We did not have specific S1P3 pharmacological reagents for our studies. Although the bulk of our data point to S1P1 as the primary S1P receptor preventing TNF-–mediated monocyte–EC interactions, we cannot absolutely rule out some contribution of S1P3. This is especially important in light of the fact that S1P3, like S1P1,23 has recently been shown to regulate eNOS activity.13 We will examine the role of S1P3 in regulating these early inflammatory processes.
The mechanisms by which S1P exerts its antiinflammatory effect are still unclear. We examined expression of adhesion molecules on endothelium in response to S1P. TNF- upregulates expression of VCAM-1 and ICAM-1 on the endothelial surface.29,30 S1P at 100 nM did not reduce the TNF-–mediated increase in endothelial VCAM-1 or ICAM-1 expression. Previous studies31,32 reported that S1P induced endothelial VCAM-1 and E-selectin expression. These investigators used very high concentrations of S1P (1 to 20 μmol/L). We recapitulated this finding, but only when S1P was added to ECs at concentrations >5 μmol/L (data not shown). In the bloodstream, physiological levels of free S1P are in the nanomolar range6,7 and the reported Kd values for S1P binding to the S1P1 receptor are 1 to 10 nanomolar. Thus, it is conceivable that very high concentrations of S1P act on ECs through some other signaling pathway. Xia et al reported that S1P at concentrations >1 μmol/L activated NFB.31 These investigators also reported that TNF- activated sphingosine kinase in endothelium (the enzyme that generates S1P from sphingosine), causing increased production of S1P that leads to erk, c-jun, and NFB activation. We also observed NFB activation in endothelium by S1P concentrations of 1 μmol/L or greater (data not shown). However, at concentrations of S1P <500 nM, we found no activation of NFB, and a decrease in c-jun activation in response to TNF- stimulation of endothelium. Again, because we are able to recapitulate these investigators’ findings when using micromolar concentrations of S1P, we believe that the concentration of S1P in circulation is quite important for preventing inflammatory processes. Thus, the antiinflammatory action of S1P is through nanomolar concentrations of S1P primarily activating S1P1 on endothelium. Xia et al reported that pertussis toxin (to uncouple Gi signaling of G protein-coupled receptors (GPCRs) as shown in Figure 1) did not inhibit the effect of S1P on adhesion molecule expression, suggesting that S1P at higher concentrations acts intracellularly on some non-S1P receptor pathway.31 However, we did not measure sphingosine kinase activation in the current study, so we do not know if injection of TNF- in vivo increased sphingosine kinase activity. However, in light of our data supporting a decrease in monocyte adhesion in response to S1P, this would seem unlikely.
A second mechanism of action of S1P on endothelium could be to reduce endothelial production of pro-inflammatory chemokines, including IL-8 and MCP-1. We found that TNF- increased endothelial production of both JE/MCP-1 and KC. S1P modestly reduced protein expression of KC (by 10%), but not JE/MCP-1. We did not examine chemokine expression at later time points. IL-8/KC has been shown to be regulated by NFB, p38 MAP kinase, and by Jun kinase pathways through activation of the transcription factor AP-1. We have previously reported endothelial IL-8 activation by p38 MAP kinase and AP-1 activation in diabetes.20 Activation of these signaling pathways in endothelium may be the result of increased oxidative stress in response to TNF-. We did not examine p38 activation by TNF- and S1P in the current study in that inhibition of p38 MAP kinase by 5 μmol/L SB203580 had no impact on S1P action in a monocyte adhesion assay (data not shown). However, we did find that S1P reduced activation of Jun kinase (Figure 4). Thus, our data suggest that endothelial KC/IL-8 mRNA levels may be reduced by S1P through inhibition of JNK and AP-1. Studies are ongoing to further investigate this finding.
Garcia et al have recently reported that 500 nM S1P enhanced endothelial barrier function in the pulmonary vasculature.33 This enhanced barrier protection in EC was caused by a tighter rearrangement of the cytoskeleton that was mediated by S1P. It is unclear whether enhanced barrier function in EC prevents monocyte–EC interactions in the vessel wall, although a correlation is quite possible. Sessa et al have shown that S1P regulates Akt and eNOS expression.23 Thus, signaling mechanisms downstream of S1P1 that include activation of PI3 kinase and Akt may be protective against monocyte–endothelial interactions. Indeed, reports have suggested that eNOS prevents leukocyte–endothelial interactions and may enhance barrier function.34,35 Additional evidence that points to a role for eNOS in S1P action in endothelium is a report by Natarajan et al that suggested that nitric oxide production by the endothelium prevented Jun kinase-mediated activation of AP-1.22 Michel et al have also reported that S1P caused vasodilation in rat arteries through rapid phosphorylation and activation of eNOS.12 Studies are underway in the laboratory to examine the possible role of eNOS in the antiinflammatory action of S1P on monocyte–EC interactions.
In conclusion, we have identified that S1P signaling to the S1P1 receptor on EC protects the vasculature against TNF-–mediated monocyte–EC interactions. Identification of this potent antiinflammatory action of S1P and understanding its signaling mechanism may be of great therapeutic benefit for a number of inflammatory diseases.
Acknowledgments
The authors thank Melissa A Marshall and Jeremy Mauldin (University of Virginia) for technical assistance with mouse studies. The authors also thank Dr Brett Blackman (University of Virginia) for the gift of human umbilical vein endothelial cells, and Dr Erik Hewlett (University of Virginia) for the gift of PTX.
This work was supported by grants from the Juvenile Diabetes Research Foundation (C.C.H.), National Institutes of Health R01 HL079621 (C.C.H.), National Institutes of Health R01 GM067958 (K.R.L.), and National Institutes of Health F31 GM064101 (M.D.D.).
References
Ross R. Atherosclerosis–an inflammatory disease. N Engl J Med. 1999; 340: 115–126.
Berliner JA, Navab M, Fogelman AM, Frank JS, Demer LL, Edwards PA, Watson AD, Lusis AJ. Atherosclerosis: basic mechanisms. Oxidation, inflammation, and genetics. Circulation. 1995; 91: 2488–2496.
Aggarwal BB, Shishodia S, Ashikawa K, Bharti AC. The role of TNF and its family members in inflammation and cancer: lessons from gene deletion. Curr Drug Targets Inflamm Allergy. 2002; 1: 327–341.
Reimold AM. New indications for treatment of chronic inflammation by TNF-alpha blockade. Am J Med Sci. 2003; 325: 75–92.
Spiegel S, Milstien S. Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat Rev Mol Cell Biol. 2003; 4: 397–407.
Murata N, Sato K, Kon J, Tomura H, Yanagita M, Kuwabara A, Ui M, Okajima F. Interaction of sphingosine 1-phosphate with plasma components, including lipoproteins, regulates the lipid receptor-mediated actions. Biochem J. 2000; 352 (Pt 3): 809–815.
Kimura T, Sato K, Malchinkhuu E, Tomura H, Tamama K, Kuwabara A, Murakami M, Okajima F. High-density lipoprotein stimulates endothelial cell migration and survival through sphingosine 1-phosphate and its receptors. Arterioscler Thromb Vasc Biol. 2003; 23: 1283–1288.
Fueller M, Wang dA, Tigyi G, Siess W. Activation of human monocytic cells by lysophosphatidic acid and sphingosine-1-phosphate. Cell Signal. 2003; 15: 367–375.
Kluk MJ, Hla T. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochim Biophys Acta. 2002; 1582: 72–80.
Liu Y, Wada R, Yamashita T, Mi Y, Deng CX, Hobson JP, Rosenfeldt HM, Nava VE, Chae SS, Lee MJ, Liu CH, Hla T, Spiegel S, Proia RL. Edg-1, the G protein-coupled receptor for sphingosine-1-phosphate, is essential for vascular maturation. J Clin Invest. 2000; 106: 951–961.
Sanna MG, Liao J, Jo E, Alfonso C, Ahn MY, Peterson MS, Webb B, Lefebvre S, Chun J, Gray N, Rosen H. Sphingosine 1-phosphate (S1P) receptor subtypes S1P1 and S1P3, respectively, regulate lymphocyte recirculation and heart rate. J Biol Chem. 2004; 279: 13839–13848.
Dantas AP, Igarashi J, Michel T. Sphingosine 1-phosphate and control of vascular tone. Am J Physiol Heart Circ Physiol. 2003; 284: H2045–H2052.
Nofer JR, van der GM, Tolle M, Wolinska I, von Wnuck LK, Baba HA, Tietge UJ, Godecke A, Ishii I, Kleuser B, Schafers M, Fobker M, Zidek W, Assmann G, Chun J, Levkau B. HDL induces NO-dependent vasorelaxation via the lysophospholipid receptor S1P3. J Clin Invest. 2004; 113: 569–581.
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 G(i), G(q), and G(12) families of heterotrimeric G proteins. J Biol Chem. 1999; 274: 27351–27358.
Davis MD, Clemens JJ, Macdonald TL, Lynch KR. Sphingosine 1-phosphate analogs as receptor antagonists. J Biol Chem. 2005; 280: 9833–9841.
McEvoy LM, Jutila MA, Tsao PS, Cooke JP, Butcher EC. Anti-CD43 inhibits monocyte-endothelial adhesion in inflammation and atherogenesis. Blood. 1997; 90: 3587–3594.
Forrest M, Sun SY, Hajdu R, Bergstrom J, Card D, Hale J, Keohane CA, Meyers C, Milligan J, Mills S, Nomura N, Rosenbach M, Shei GJ, Singer II, Tian M, West S, White V, Xie J, Rosen H, Proia R, Doherty G, Mandala S. Immune Cell Regulation and Cardiovascular Effects of Sphingosine 1-Phosphate Receptor Agonists in Rodents are Mediated via Distinct Receptor Sub-Types. J Pharmacol Exp Ther. 2004; 758–768.
Patricia MK, Kim JA, Harper CM, Shih PT, Berliner JA, Natarajan R, Nadler JL, Hedrick CC. Lipoxygenase products increase monocyte adhesion to human aortic endothelial cells. Arterioscler Thromb Vasc Biol. 1999; 19: 2615–2622.
Lee H, Shi W, Tontonoz P, Wang S, Subbanagounder G, Hedrick CC, Hama S, Borromeo C, Evans RM, Berliner JA, Nagy L. Role for peroxisome proliferator-activated receptor alpha in oxidized phospholipid-induced synthesis of monocyte chemotactic protein-1 and IL-8 by endothelial cells. Circ Res. 2000; 87: 516–521.
Srinivasan S, Bolick DT, Hatley ME, Natarajan R, Reilly KB, Yeh M, Chrestensen C, Sturgill TW, Hedrick CC. Glucose regulates IL-8 production in aortic endothelial cells through activation of the p38 mitogen-activated protein kinase pathway in diabetes. J Biol Chem. 2004; 279: 31930–31936.
Lerner-Marmarosh N, Yoshizumi M, Che W, Surapisitchat J, Kawakatsu H, Akaike M, Ding B, Huang Q, Yan C, Berk BC, Abe J. Inhibition of tumor necrosis factor--induced SHP-2 phosphatase activity by shear stress: a mechanism to reduce endothelial inflammation. Arterioscler Thromb Vasc Biol. 2003; 23: 1775–1781.
Natarajan R, Gupta S, Fisher BJ, Ghosh S, Fowler AA III. Nitric oxide suppresses IL-8 transcription by inhibiting c-Jun N- terminal kinase-induced AP-1 activation. Exp Cell Res. 2001; 266: 203–212.
Morales-Ruiz M, Lee MJ, Zollner S, Gratton JP, Scotland R, Shiojima I, Walsh K, Hla T, Sessa WC. Sphingosine 1-phosphate activates Akt, nitric oxide production, and chemotaxis through a Gi protein/phosphoinositide 3-kinase pathway in endothelial cells. J Biol Chem. 2001; 276: 19672–19677.
Xie JH, Nomura N, Koprak SL, Quackenbush EJ, Forrest MJ, Rosen H. Sphingosine-1-phosphate receptor agonism impairs the efficiency of the local immune response by altering trafficking of naive and antigen-activated CD4+ T cells. J Immunol. 2003; 170: 3662–3670.
Brinkmann V, Davis MD, Heise CE, Albert R, Cottens S, Hof R, Bruns C, Prieschl E, Baumruker T, Hiestand P, Foster CA, Zollinger M, Lynch KR. The immune modulator FTY720 targets sphingosine 1-phosphate receptors. J Biol Chem. 2002; 277: 21453–21457.
Matloubian M, Lo CG, Cinamon G, Lesneski MJ, Xu Y, Brinkmann V, Allende ML, Proia RL, Cyster JG. Lymphocyte egress from thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature. 2004; 427: 355–360.
Peng X, Hassoun PM, Sammani S, McVerry BJ, Burne MJ, Rabb H, Pearse D, Tuder RM, Garcia JG. Protective Effects of Sphingosine 1-Phosphate in Murine Endotoxin-Induced Inflammatory Lung Injury. Am J Respir Crit Care Med. 2004; 1245–1251.
Mandala S, Hajdu R, Bergstrom J, Quackenbush E, Xie J, Milligan J, Thornton R, Shei GJ, Card D, Keohane C, Rosenbach M, Hale J, Lynch CL, Rupprecht K, Parsons W, Rosen H. Alteration of lymphocyte trafficking by sphingosine-1-phosphate receptor agonists. Science. 2002; 296: 346–349.
Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J. 1995; 9: 899–909.
VandenBerg E, Reid MD, Edwards JD, Davis HW. The role of the cytoskeleton in cellular adhesion molecule expression in tumor necrosis factor-stimulated endothelial cells. J Cell Biochem. 2004; 91: 926–937.
Xia P, Gamble JR, Rye KA, Wang L, Hii CS, Cockerill P, Khew-Goodall Y, Bert AG, Barter PJ, Vadas MA. Tumor necrosis factor-alpha induces adhesion molecule expression through the sphingosine kinase pathway. Proc Natl Acad Sci U S A. 1998; 95: 14196–14201.
Rizza C, Leitinger N, Yue J, Fischer DJ, Wang DA, Shih PT, Lee H, Tigyi G, Berliner JA. Lysophosphatidic acid as a regulator of endothelial/leukocyte interaction. Lab Invest. 1999; 79: 1227–1235.
Dudek SM, Jacobson JR, Chiang ET, Birukov KG, Wang P, Zhan X, Garcia JG. Pulmonary endothelial cell barrier enhancement by sphingosine 1-phosphate: roles for cortactin and myosin light chain kinase. J Biol Chem. 2004; 279: 24692–24700.
Hayashi T, Sumi D, Juliet PA, Matsui-Hirai H, Asai-Tanaka Y, Kano H, Fukatsu A, Tsunekawa T, Miyazaki A, Iguchi A, Ignarro LJ. Gene transfer of endothelial NO synthase, but not eNOS plus inducible NOS, regressed atherosclerosis in rabbits. Cardiovasc Res. 2004; 61: 339–351.
Kuhlencordt PJ, Rosel E, Gerszten RE, Morales-Ruiz M, Dombkowski D, Atkinson WJ, Han F, Preffer F, Rosezweig A, Sessa WC, Gimbrone MA, Ertl G, Huang PL. The role of endothelial nitric oxide synthase (eNOS) in endothelial activation: insights from eNOS-knockout endothelial cells. Am J Physiol Cell Physiol. 2004; C1195–C1202.(David T. Bolick; Suseela )