HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: M800143-JLR200v1 49/11/2323    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Google Scholar Articles by Zhao, C. Articles by Bourgoin, S. G. PubMed PubMed Citation Articles by Zhao, C. Articles by Bourgoin, S. G. Social Bookmarking                      What's this?

Journal of Lipid Research, Vol. 49, 2323-2337, November 2008
Copyright © 2008 by American Society for Biochemistry and Molecular Biology

Specific and overlapping sphingosine-1-phosphate receptor functions in human synoviocytes: impact of TNF-

Chenqi Zhao^*, Maria J. Fernandes^*, Mélanie Turgeon^*, Sabrina Tancrède^*, John Di Battista, Patrice E. Poubelle^* and Sylvain G. Bourgoin^1,*

^* Centre de Recherche en Rhumatologie et Immunologie, Centre de Recherche du CHUQ-CHUL, Départements d'Anatomie-Physiologie et Médecine, Faculté de Médecine, Université Laval, Québec, Canada, G1V 4G2
Département de Rhumatologie et Immunologie, Centre Universitaire McGill, Montréal, Québec, Canada, H3A 1A1

Published, JLR Papers in Press, July 24, 2008.

This project is supported by research grants from Canadian Institutes of Health Research and the Arthritis Society of Canada (S.G.B.) and the Canadian Arthritis Network Post-Doctoral Fellowship Award (C.Z.).

¹ To whom correspondence should be addressed. e-mail: sylvain.bourgoin{at}crchul.ulaval.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Sphingosine-1-phosphate (S1P), via interaction with its G protein-coupled receptors, regulates various physiological and pathological responses. The present study investigated the role of S1P/S1P receptor signaling in several functional responses of human fibroblast-like synoviocytes (FLSs) that may contribute to the pathogenesis of rheumatoid arthritis (RA). We report that FLSs express the S1P₁, S1P₂, and S1P₃ receptors. Moreover, exogenously applied S1P induces FLS 1) migration, 2) secretion of inflammatory cytokines/chemokines, and 3) protection from apoptosis. Using specific S1P receptor agonists/antagonists, we determined that S1P stimulates FLS migration through S1P₁ and S1P₃, induces cytokine/chemokine secretion through S1P₂ and S1P₃, and protects from cell apoptosis via S1P₁. The S1P-mediated cell motility and cytokine/chemokine secretion seem to be regulated by the p38 mitogen-activated protein kinase (MAPK), p42/44 MAPK, and Rho kinase signal transduction pathways. Interestingly, treatment of FLSs with tumor necrosis factor- increases S1P₃ expression and correlates with the enhancement of S1P-induced cytokine/chemokine production. Our data suggest that S1P₁, S1P₂, and S1P₃ play essential roles in the pathogenesis of RA by modulating FLS migration, cytokine/chemokine production, and cell survival. Moreover, the cytokine-rich environment of the inflamed synovium may synergize with S1P signaling to exacerbate the clinical manifestations of this autoimmune disease.

Supplementary key words inflammation • G protein-coupled receptor • interleukin-8 • cell migration • apoptosis • mitogen-activated protein kinase • Rho kinase

Abbreviations: COX-2, cyclooxygenase-2; EDG, endothelial differentiation gene; FLS, fibroblast-like synoviocyte; GM-CSF, granulocyte-monocyte colony-stimulating factor; IL, interleukin; IP-10, interferon--inducible protein 10; LPA, lysophosphatidic acid; MAPK, mitogen-activated protein kinase; MCP-1, monocyte chemotactic protein-1; MIP-1, macrophage inflammatory protein-1; PGE₂, prostaglandin E₂; RA, rheumatoid arthritis; RANTES, regulated on activation normal T cells expressed and secreted; S1P, sphingosine-1-phosphate; SNP, sodium nitroprusside; TNF-, tumor necrosis factor-

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Rheumatoid arthritis (RA) is a chronic inflammatory disorder that primarily affects diarthroidal joints, leading to their progressive destruction. The disease is characterized by hyperplasia of fibroblast-like synoviocytes (FLSs) and a massive infiltration of inflammatory cells in the affected joint. FLSs comprise the synovial lining, a thin membrane in direct contact with cartilage and bone. In RA, FLSs play a key role in the pathogenesis (as reviewed in Ref. 1). FLSs increase in number, invade adjacent tissues, and produce proinflammatory cytokines, chemokines, and matrix metalloproteinases that promote inflammation and joint destruction (2). Proinflammatory cytokines, such as tumor necrosis factor- (TNF-) and interleukin-1β (IL-1β), appear to play a key role in the stimulation of FLSs toward this aggressive phenotype. There is mounting evidence, however, that the activation of FLSs can be maintained in the absence of inflammatory cytokines (3), suggesting the contribution of other proinflammatory molecules to the pathogenesis of RA.

Sphingosine-1-phosphate (S1P) is a biologically active sphingolipid that transmits potent signals through five G protein-coupled EDG (endothelial differentiation gene) receptors, namely, S1P₁/EDG-1, S1P₂/EDG-5, S1P₃/EDG-3, S1P₄/EDG-6, and S1P₅/EDG-8 (4). Among the five S1P receptors, S1P₁, S1P₂, and S1P₃ are widely expressed in various tissues, whereas the expression of S1P₄ and S1P₅ is mainly confined to cells of the immune system and nervous system, respectively (5). S1P receptors activate a variety of heterologous signaling pathways through coupling with multiple G proteins (G_q, G_i/o, and G_12/13) that regulate both physiological (cell growth, differentiation, migration, and survival) and pathological processes (angiogenesis, cancer, and autoimmunity) (as reviewed in Ref. 6).

Activated platelets are the main source of S1P (7). However, other cell types, such as neutrophils and mononuclear cells, also produce S1P (8, 9). S1P synthesis and secretion are stimulated by inflammatory mediators such as TNF- (10). Virtually all cells that participate in immune responses express multiple S1P receptors (as reviewed in Ref. 11). Whereas S1P enhances the expression of inflammation-related genes (12), S1P signaling through the S1P₁ receptor also controls T cell migration and tissue distribution, as well as the initiation of early events in the differentiation of T cells into effector states (13).

Several lines of evidence strongly suggest that S1P contributes to the pathogenesis of RA. S1P is produced by neutrophils, the most abundant cell type in synovial fluids of RA patients. Also, TNF-, a cytokine that plays a key role in the pathogenesis of RA, and other inflammatory mediators induce the synthesis and secretion of S1P. High levels of S1P have, indeed, been reported in the synovial fluid of patients with RA (14). Not only is the synthesis and secretion of S1P induced by proinflammatory mediators but, in turn, S1P also triggers the expression of inflammatory genes such as cyclooxygenase-2 (COX-2), and a super-production of prostaglandin E₂ (PGE₂) by FLSs (14). Moreover, the expression of sphingosine kinase, the enzyme that generates S1P by phosphorylation of sphingosine (15), is significantly increased in the cells derived from RA B lymphoblastoid cells (16). These findings from in vitro experiments ought to be at least partly representative of the in vivo physiological function of S1P in the RA synovium.

Because FLS activation is partly cytokine-independent and S1P is known to induce gene expression in FLSs, we sought to fully characterize the role of the S1P/S1P receptor signaling pathway in FLS function. The objective of the present study was thus to determine whether the S1P/S1P receptor signaling pathway is involved in RA. Herein, we report that FLSs express the S1P₁, S1P₂, and S1P₃ receptors. We also provide evidence that these S1P receptors trigger specific and overlapping functional responses. S1P₁ was essential for the survival of FLSs, and S1P₁ and S1P₃ stimulated FLS migration. In contrast, the activation of S1P₂ and S1P₃ enhanced cytokine/chemokine secretion. In addition, S1P₃ expression by FLSs was enhanced by TNF-, and S1P-mediated chemokine secretion was super-induced in TNF--primed FLSs. These results suggest that in addition to S1P₁, S1P₂ and S1P₃ may also play important roles in the pathogenesis of RA.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reagents
S1P was purchased from Biomol (Plymouth Meeting, PA). The specific S1P_1/3 receptor antagonist VPC23019 was obtained from Avanti Polar Lipids, Inc. (Alabaster, AL). The S1P₁ agonist SEW2871, the S1P₂ antagonist JTE-013, and the S1P₃ antagonist CAY10444, were from Cayman Chemical (Ann Arbor, MI). TNF- was from PeproTech, Inc. (Rocky Hill, NJ). Human IL-8 and IL-6 ELISA kits were purchased from BioSource International, Inc. (Camarillo, CA). The apoptosis assay kit (Annexin V-FITC) was from BD Pharmingen (Oakville, ON, Canada). The cell proliferation assay kit (CellTiter 96® Aqueous One Solution Cell Proliferation) was purchased from Promega (Madison, WI). Sodium nitroferricyanide (III) dihydrate [sodium nitroprusside (SNP)], and SYBR® Green JumpStart^TM Ready Mix kit were obtained from Sigma (St. Louis, MO). TRIzol Reagent was from Invitrogen (Burlington, ON, Canada). Inhibitors of p42/44 mitogen-activated protein kinase (MAPK) PD98059, of p38 MAPK SB203580, and of Rho kinase Y27632 were purchased from Calbiochem (San Diego, CA). S1P₁ and S1P₃ antibodies were from Cayman Chemical. Antibodies to total and phosphorylated forms of p42/44 MAPK, of p38MAPK, and of JNK were purchased from Cell Signaling Technology (Beverly, MA). The human cytokine/chemokine LINCOplex kit was purchased from Linco Research (St. Charles, MO). Cell culture reagents were purchased from Wisent, Inc. (St-Bruno, QC, Canada).

Cell culture
Human primary FLSs were obtained from RA patients, who were diagnosed according to the criteria developed by the American College of Rheumatology (17) and were undergoing joint surgery on the knee or hip. Cells were maintained under standard conditions (37°C and 5% CO₂) and grown in DMEM supplemented with 10% FBS, penicillin (100 IU), and streptomycin (100 µM). Cells were used at passages 5 to 15.

Cell treatment
Semi-confluent cells were starved with serum-free medium for 24 h before treatment. At the moment of cell treatment, the culture medium was replaced with fresh serum-free medium containing various concentrations of the tested compounds as indicated in the details presented below.

Semi-quantitative RT-PCR and real-time PCR analyses
Total cellular RNA was extracted using TRIzol reagent according to the manufacturer's instructions. RNA (0.5–1 µg) was reverse-transcribed using random priming and the Superscript II Reverse Transcriptase system (Invitrogen). Oligonucleotides used as primers were designed to amplify specific cDNA sequences. Primer sequences and PCR conditions are as follows: S1P₁ (778 bp product), sense, 5'-GGA-AGG-GAG-TAT-GTT-TGT-GGC-3', antisense, 5'-TGA-CGT-TTC-CAG-AAG-ACA-TA-3'; S1P₂ (425 bp product), sense, 5'-AGT-GGC-CAT-TGC-CAA-GGT-CAA-G-3', antisense, 5'-TAG-TGG-GCT-TTG-TAG-AGG-A-3'; S1P₃ (466 bp product), sense, 5'- AGG-GAG-GGC-AGT-ATG-TTC-G-3', antisense, 5'-GCC-ACA-TCA-ATG-AGG-AAG-AGG-AT-3'; S1P₄ (672 bp product), sense, 5'-ATC-ACG-CTG-AGT-GAC-CTG-CTC-A-3', antisense, 5'-TGC-GGA-AGG-AGT-AGA-TGA-3'; and S1P₅ (658 bp product), sense, 5'-CTA-CTG-TCG-GGG-CCG-CTC-AC-3', antisense, 5'-CGG-TTG-GTG-AAC-GTG-TAG-ATG-A-3'. To ensure linear cDNA amplification, different amplifying cycles were tried. The experiments revealed linear amplification within 35 cycles. A total of 35 PCR cycles were run at 94°C (denaturation, 1 min), 63°C for S1P_1–3, 61°C for S1P_4–5 (annealing, 1 min), and 72°C (extension, 1 min). The ribosomal protein RPLP0 mRNA was used as an internal PCR control. RPLP0 (248 bp product) primer sequences are as follows: sense, 5'-GTT-GTA-GAT-GCT-GCC-ATT-G-3', antisense, 5'-CCA-TGT-GAA-GTC-ACT-GTG-C-3'. The PCR products were subjected to electrophoresis on an 0.8% agarose gel and visualized by ethidium bromide staining. Densitometry analysis was used for band quantification using the software Alphamage 2000. The results were expressed as a ratio of the band intensity relative to the corresponding RPLP0 band obtained by amplification of the same template cDNA. Semi-quantitative real-time PCR was also conducted to examine the mRNA expression of S1P_1–3 receptors. In real-time PCR experiments, we used the same primers as for RT-PCR to amplify S1P_1–3. The thermal cycling conditions were as follows: 95°C (initial denaturation, 3 min) followed by 40 cycles of 95°C (denaturation, 15 s), 63°C for S1P_1–3 (annealing, 20 s), and 72°C (extension, 20 s).

Western blot analyses
To monitor the activation of p42/44 MAPK, p38 MAPK, and JNK, cells were exposed to S1P (5 µM) for various times (5–30 min). Where indicated, cells were pretreated with PD98059 (25 µM), SB203580 (10 µM), and Y27632 (10 µM) for 30 min prior to S1P stimulation (5 µM, 15 min). Cells were lysed in boiling sample buffer [50 mM Tris/HCL (pH 6.8), 10% (v/v) glycerol, 50 mM DTT, 4% (v/v) SDS] and boiled for 7–10 min. Equal amounts of protein (50 µg) were separated by 10% SDS-PAGE.

For analysis of S1P₁/S1P₃ receptor expression, cell membrane fractions were prepared as described previously (18). Briefly, cells were treated with accutase, harvested in ice-cold KCl-Hepes relaxation buffer (50 mM Hepes, 100 mM KCl, 5 mM NaCl, 1 mM MgCl₂, 0.5 mM EGTA, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 2 mM PMSF, and 1 mM sodium orthovanadate, adjusted to pH 7.2), and sonicated for 20 s, and the samples were centrifuged for 7 min at 1,000 g. Unbroken cells and nuclei were discarded, and the supernatants were ultracentrifuged at 180,000 g for 45 min in a Beckman TL-100 ultracentrifuge. Membrane pellets were washed once and resuspended in a small volume of solubilization buffer containing 0.25 M Na₂HPO₄, 0.3 M NaCl, 2.5% SDS, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 2 mM PMSF, and samples were assayed for protein content with Pierce Coomassie Brilliant Blue Protein Assay reagent. To determine whether TNF- can regulate the expression of S1P₃, cells were cultured for 24 h in the presence of TNF- (100 ng/ml). Protein samples (150 µg) were separated by gradient (7.5–20%) SDS-PAGE.

Proteins were later transferred from polyacrylamide gel to methanol-soaked Immobilon polyvinylidene difluoride membranes (Millipore, Bedford, MA). Primary antibody incubation was performed overnight at 4°C. The membranes were then washed three times and incubated with appropriate HRP-conjugated secondary antibodies at room temperature for 1 h. Membranes were washed three times, and antibody-antigen complexes were revealed using ECL® according to the manufacturer's instructions (Perkin Elmer Life Sciences, Boston, MA). The ECL signal was visualized by developing the signal on a film.

Wound-closing assay
Cells were plated at a concentration of 5 x 10⁴ cells/ml in 12-well plates. After routine starvation, a plastic pipette tip (200 µl) was drawn across the center of the well to produce a clean wound area. Free cells were removed and the medium was replaced with serum-free medium containing S1P, selective agonists/antagonists of S1P receptors, and the specific inhibitors of p44/42 MAPK, p38 MAPK, and Rho kinase at indicated concentrations. Immediately following scratch wounding (0 h) and after incubation for 24 h, the wound-closing process was photographed with an inverted microscope (Nikon TE300). The cells that had migrated into the wound area were examined and monitored with MetaMorph software.

ELISA and Luminex 100 multiplex array assay
Cells plated at a concentration of 5 x 10⁴ cells/ml in 24-well plates were stimulated with S1P or the S1P₁ agonist SEW2871 at indicated concentrations or times. Where indicated, cells were pretreated with the S1P receptor antagonists or the inhibitors of p42/44 MAPK, p38 MAPK, and Rho kinase for 30 min prior to stimulation with S1P. To evaluate the effect of TNF- on S1P-mediated cytokine secretion, cells were pretreated with TNF- for 2 h, 8 h, and 24 h and washed extensively prior to stimulation with S1P (5 µM, 24 h). Cell culture supernatants were collected and stored at –80°C until the cytokine ELISAs were performed. IL-8 and IL-6 in all samples were monitored in duplicate, according to the manufacturer's protocol. Optical densities were determined using a SoftMaxPro40 plate reader at 450 nm. The results were compared with a standard curve that was generated using known concentrations (pg/ml) of IL-8 and IL-6. The results were expressed in pg/ml.

For the Luminex 100 multiplex array assay, cells were stimulated with S1P (5 µM, 24 h) with or without pretreatment of TNF- (100 ng/ml, 8 h), and cell supernatants were collected for multiple cytokine and chemokine analyses. The multiplex bead-based sandwich immunoassay kit was used to measure the levels of IL-1, IL-1β, IL-8, IL-15, Eotaxin, granulocyte-monocyte colony-stimulating factor (GM-CSF), interferon--inducible protein 10 (IP-10), monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein-1 (MIP-1) and MIP-1β, and regulated on activation normal T cells expressed and secreted (RANTES).

Apoptosis assay and proliferation assay
For the apoptosis assay, cells were stimulated with 5 µM S1P or 0.1–3 µM SEW2871 for 8 h prior to stimulation with 1 mM SNP for an additional 16 h. Cells were then harvested, and apoptosis was evaluated using Annexin V-FITC labeling according to the manufacturer's instructions. Cells were then analyzed by flow cytometry. For the proliferation assay, cells were cultured with or without S1P (1 nM–5 µM) in triplicate in 96-well microplates at a concentration of 2 x 10⁵/ml in DMEM without serum. After 96 h, cell viability was assessed by measuring the conversion of the CellTiter 96® Aqueous One Solution Reagent [3-(4,5-dimethythizol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] to a water-soluble formazan product that absorbs light at 490 nm and is proportional to the number of live cells.

Statistical analysis
Unless otherwise stated, experiments were performed three times and results presented are expressed as mean values ± SE. In Table 1 and Fig. 10C, results are expressed as mean values ± SD. Statistical significance of the difference between treated and untreated samples was determined by t-test (two-tailed P value). For the dose response and time course studies, statistical significance between untreated samples and samples treated with S1P, S1P₁ receptor agonist, TNF-, and kinase inhibitors (dose response experiments) and between samples treated at 0 h with those treated at indicated time points (time course study) was determined by ANOVA, Dunnett's Multiple Comparison Test. Calculations were made with the Prism 4.0 software. P values less than 0.05 were considered statistically significant.

View larger version (46K): [in this window] [in a new window]   Fig. 10. Regulation of S1P3 receptor expression by tumor necrosis factor- (TNF-). A: Concentration-dependent response of S1P3 mRNA expression to TNF- (RT-PCR). Cells were treated with TNF- (2 h, at indicated concentrations) prior to RNA extraction and RT-PCR analyses. Results are presented as a representative agarose gel (upper panels) and/or as ratios (means ± SE, n = 3) of S1P3 and RPLP0 (lower panels). B: Real-time PCR analysis of S1P3 mRNA expression upon TNF- treatment. Cells were incubated with TNF- (100 ng/ml) for 2 h prior to RNA extraction and real-time PCR analyses. C: Western blot analysis of S1P3 protein expression upon TNF- treatment. Cells were incubated with or without TNF- (100 ng/ml) for 24 h. Cell lysates were subjected to 7.5–20% gradient SDS/PAGE, and samples were probed with antibodies to S1P3, HA-tag, and flotillin. Blot shown is a representative example from two independent experiments with similar results (upper panel). Bands corresponding to S1P3 were quantified densitometrically, and their levels were normalized with respect to flotillin. Data are shown as the ratio of arbitrary units for S1P3 to flotillin and as means ± SD (n = 2, lower panel). For statistical comparative analysis, we compared nontreated cells to those treated with TNF-. *P < 0.05; **P < 0.01.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

View larger version (33K): [in this window] [in a new window]   Fig. 1. Expression of sphingosine-1-phosphate1 (S1P1), S1P2, and S1P3 mRNA in human fibroblast-like synoviocytes (FLSs). Semi-quantitative RT-PCR (A) and real-time PCR (B) analyses of S1P1, S1P2, and S1P3 mRNA expression in FLSs. Total RNA from resting-state FLSs was extracted for RT-PCR or real-time PCR. The reaction performed without oligonucleotide primers was used as a negative control, and RPLP0 was used as an internal control. Data shown are representative of three separate experiments and are expressed as means ± SE.

Induction of human FLS migration by S1P via S1P₁ and/or S1P₃ receptors
S1P is known to induce cell migration. We therefore investigated whether S1P could enhance the migration of FLSs using a wound-closing assay. As shown in Fig. 2 , cell migration was stimulated by S1P at a concentration of 1 µM or 5 µM, by 1.7 ± 0.3-fold (P < 0.05) and 3.2 ± 0.2-fold (P < 0.01) above the nontreated control, respectively.

View larger version (21K): [in this window] [in a new window]   Fig. 2. Stimulation of human FLS motility by S1P. A clean wound area was made on a monolayer of FLSs. After free cells were removed, the wound was allowed to close for 24 h in serum-free medium containing S1P (0.1–5 µM). The wound-closing process was photographed at 0 h and at 24 h (A). The data shown are representative of three separate experiments. The migration index corresponds to the percentage of cells that migrated in the presence of S1P over that of nontreated cells (B). For statistical comparative analyses, we compared cells treated with and without S1P. Data shown are means ± SE of three independent experiments. *P < 0.05; **P < 0.01.

View larger version (26K): [in this window] [in a new window]   Fig. 3. Effect of selective S1P receptor agonists/antagonists on FLS motility. A: Effect of S1P1 agonist SEW2871 on cell motility. After scratching the monolayer and removing free cells, the wound was allowed to close for 24 h in serum-free medium containing SEW2871 (1–10 µM). The wound-closing process was photographed at 0 h and at 24 h (upper panels). The data shown are representative of three separate experiments. Migrated cell numbers were expressed as percentage of nontreated cells (lower panel). For statistical comparative analyses, we compared cells treated with and without SEW2871. B: Effect of the S1P1/3 antagonist VPC23019, S1P2 antagonist JTE-013, and S1P3 antagonist CAY10444 on cell motility. The wound was allowed to close for 24 h in serum-free medium containing S1P (5 µM) in the presence or absence of VPC23019 (5 µM), JTE-013 (5 µM), and CAY10444 (5 µM). Migrated cell numbers were expressed as percentage of nontreated cells. For statistical comparative analyses, samples treated with S1P were compared with those treated with S1P+VPC23019/JTE-013/CAY10444. Data shown are means ± SE of three independent experiments. **P < 0.01.

View larger version (15K): [in this window] [in a new window]   Fig. 4. S1P-mediated interleukin-8 (IL-8) and IL-6 secretion. A, B: Dose response effect of S1P on IL-8/IL-6 secretion. FLSs were treated with the indicated concentrations of S1P for 24 h before collecting supernatants for cytokine quantification. C, D: Kinetics of IL-8/IL-6 secretion. Cells were incubated with S1P (5 µM) for the indicated times, and the cell culture supernatants were collected for the ELISA assay. Experiments were repeated three times, and the results are displayed as mean value ± SE. For statistical analyses, samples incubated with diluents were compared with those treated with S1P at the indicated concentrations (A, B), or times (C, D). *P < 0.05; **P < 0.01.

View larger version (12K): [in this window] [in a new window]   Fig. 5. Effect of selective S1P receptor antagonists on S1P-induced IL-8 and IL-6 secretion. Cells were incubated with or without 5 µM VPC23019, JTE-013, and CAY10444 for 30 min prior to stimulation with S1P (5 µM) for 24 h. Cell culture supernatants were collected and IL-8 (A) and IL-6 (B) were quantified by ELISA. The results are displayed as mean value ± SE (n = 3). For statistical comparative analyses, we compared the samples stimulated with S1P to those treated with S1P+VPC23019/JTE-013/CAY10444. *P < 0.05; **P < 0.01.

View larger version (10K): [in this window] [in a new window]   Fig. 6. Protection of FLSs from apoptosis by S1P and the S1P1 agonist SEW2871. Human FLSs were pretreated for 8 h with 5 µM S1P (A) or the indicated concentrations of the S1P1 receptor agonist SEW2871 (B) prior to the addition of 1 mM sodium nitroprusside (SNP) for 16 h. Apoptotic cells were analyzed by Annexin V-propidium iodide (PI) labeling. Total apoptotic cells represent the cells that were Annexin V+/PI+ and Annexin V+/PI–. Early apoptotic cells correspond to the percentage of cells that were Annexin V+/PI–. The results are displayed as mean values ± SE (n = 3). For statistical comparative analyses, we compared samples incubated with SNP to those treated with SNP+S1P/SEW2871. *P < 0.05; **P < 0.01; ***P < 0.001.

View larger version (72K): [in this window] [in a new window]   Fig. 7. S1P-induced activation of p42/44 mitogen-activated protein kinase (MAPK) and p38 MAPK. A: Effect of S1P on the activation of p38 MAPK. Cells were stimulated with S1P (5 µM) for up to 30 min, and cell lysates were analyzed by Western blot analysis. Blot shown is a representative example from three independent experiments with similar results. Bands corresponding to phosphorylated p38 (15 min time point) were quantified densitometrically, and were normalized with respect to the total amounts of p38 (n = 4). Data are shown as the ratio of arbitrary units for the phosphorylated p38 to total p38. B: Effect of S1P on the activation of p42/44 MAPK. Cells were stimulated with S1P (5 µM) for up to 30 min, and cell lysates were analyzed by Western blot analysis. Blot shown is a representative example from three independent experiments with similar results. Bands corresponding to phosphorylated p42/44 (15 min time point) were quantified densitometrically, and were normalized with respect to the total amounts of p42/44 (n = 4). Data are shown as the ratio of arbitrary units for the phosphorylated p42/44 to total p42/44. C–F: Effect of p42/44 MAPK, p38 MAPK, and Rho kinase inhibitors on S1P-induced activation of p42/44 MAPK and p38 MAPK. Cells were pretreated with 25 µM PD98059 (C, D), 10 µM SB203580 (C, D), and 10 µM Y27632 (E, F) for 30 min prior to stimulation with 5 µM S1P for 15 min. Cell lysates were analyzed by Western blot analysis. Data shown are representative of three separate experiments and are expressed as means ± SE. In panel D, bands corresponding to phosphorylated p42/44 were quantified densitometrically, and were normalized with respect to the total amounts of p42/44. For statistical comparative analyses, we compared nontreated samples to those treated with S1P (*P < 0.05), and samples treated with S1P to those treated with S1P+PD98059 (P < 0.05).

View larger version (8K): [in this window] [in a new window]   Fig. 8. MAPKs and Rho kinase regulation on S1P-induced FLS migration. After scratching the monolayer and removing free cells, the wound was allowed to close for 24 h in serum-free medium containing 5 µM S1P, with or without PD98059 (A), SB203580 (B), or Y27632 (C), at the indicated concentrations. Migrated cell numbers were expressed as percentage of nontreated cells. Data shown are means ± SE of three independent experiments. For statistical comparative analyses, the samples treated with S1P and S1P+inhibitors were compared. **P < 0.01.

View larger version (17K): [in this window] [in a new window]   Fig. 9. Role of p42/44 MAPK, p38 MAPK, and Rho kinase in S1P-induced cytokine secretion. Cells were treated with S1P (5 µM) for 24 h, in the presence or absence of PD98059 (A, D), SB203580 (B, E), or Y27632 (C, F), at the indicated concentrations. Secreted IL-8 (A–C) and IL-6 (D–F) were quantified by ELISA. Data shown are means ± SE of three independent experiments. For statistical comparative analyses, the samples incubated with S1P and S1P+inhibitors were compared. *P < 0.05; **P < 0.01.

Regulation of S1P₃ receptor expression by TNF-

Effect of TNF- on S1P-induced cytokine/chemokine secretion in human FLS
The next series of experiments was designed to examine the effect of a proinflammatory environment, established by a pretreatment of FLSs with TNF-, on S1P-induced secretion of cytokines/chemokines that is relevant to RA pathogenesis (27). Starved FLSs were treated with S1P (5 µM) for 24 h with or without a pretreatment of TNF- (100 ng/ml) for 8 h. Cell culture supernatants were analyzed using a Luminex 100 array assay (Table 1). Among the 11 cytokines/chemokines tested, IL-1, IL-1β, IL-15, Eotaxin, GM-CSF, MIP-1, and MIP-1β were not secreted in response to S1P, TNF-, or their combination. In contrast, significant amounts of IL-8, MCP-1, and RANTES were secreted following stimulation with S1P and, most importantly, S1P-mediated cytokine synthesis was strongly enhanced in TNF--primed FLSs. The release of IL-8, MCP-1, and RANTES was increased 5.0 ± 0.3 (P < 0.01)-, 1.7 ± 0.2 (P < 0.05)-, and 66 ± 6.4 (P < 0.01)-fold, respectively, in TNF--primed and S1P-stimulated samples as compared with unprimed cells stimulated with S1P. Moreover, S1P or TNF- priming did not stimulate the secretion of IP-10, but their combination resulted in a significant secretion of this chemokine.

As shown in Fig. 11A , after priming FLSs with TNF- for 2 h, 8 h, and 24 h, S1P-induced IL-8 secretion was super-induced. To determine the relevance of the S1P receptor(s) to this process, FLSs were treated with the S1P_1/3 receptor antagonist VPC23019. VPC23019 had no effect on the release of IL-8 by TNF--primed FLSs, but almost completely inhibited the enhanced secretion of cytokine induced by S1P in these cells (Fig. 11B). S1P-mediated super-production of IL-8 in TNF--primed cells was also completely abolished by a selective S1P₃ antagonist, CAY10444 (Fig. 11C). The data suggest that TNF- enhances S1P receptor (S1P₃) expression and S1P₃-dependent responses in human FLSs.

View larger version (25K): [in this window] [in a new window]   Fig. 11. Super-induction of S1P-induced IL-8 secretion by cell priming with TNF-. A: Kinetics of TNF- pretreatment on S1P-induced IL-8 secretion. Cells were pretreated with TNF- (100 ng/ml) for different time lengths, as indicated, prior to stimulation with 5 µM S1P for 24 h. Cell culture supernatants were harvested for IL-8 measurement. Results are presented as means ± SE (n = 3). For statistical comparative analyses, we compared the samples incubated with S1P to those treated with S1P+TNF-. **P < 0.01; ***P < 0.001. B, C: Effect of S1P1/3 antagonist VPC23019 and S1P3 antagonist CAY10444 on S1P-induced super-production of IL-8 after TNF- priming. Cells were stimulated with TNF- (100 ng/ml) for 8 h before stimulation with S1P (5 µM) for another 24 h in the presence/absence of 5 µM VPC23019 and CAY10444. The results are presented as means ± SE (n = 3). For statistical comparative analyses, we compared the samples treated with TNF-+S1P to those incubated with TNF-+S1P+VPC23019/CAY10444. *P < 0.05; ***P < 0.001.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The observation that S1P induces FLS migration is significant in the context of RA because one of the hallmarks of RA is synovial hyperplasia. This pro-migratory effect of S1P was mimicked by the S1P₁-specific agonist SEW2871 but abolished by the S1P_1/3 antagonist VPC23019 or the S1P₃ selective antagonist CAY10444. The results suggest a prevailing role for S1P₁ and S1P₃ receptors in this process. These observations are in agreement with previous reports that identified a role for S1P₁ and S1P₃ in the migratory response to S1P in other cell types (28–30). Because FLSs express S1P₂ and because this receptor is known to decrease growth factor-mediated cell migration (31), we hypothesized that blocking this receptor would increase cell migration. Our results reveal, however, that blocking the S1P₂ receptor with the selective antagonist JTE-013 had no significant effect on S1P-induced FLS migration. These data suggest that the activation of S1P₂ by S1P does not act as a negative regulator of S1P-induced migration in human FLSs. Whether S1P₂ can counteract growth factor-mediated or CC or CXC chemokine-mediated migration of FLSs remains to be elucidated.

A key feature of RA is the infiltration of immune cells such as neutrophils, monocytes, and T lymphocytes into the joints due to the large amount of CC and CXC chemokines (32–34) produced by activated cells of the synovial lining. In particular, IL-8 exhibits selective chemotactic activity for neutrophils, whereas MCP-1, MIP-1, MIP-1β, and RANTES primarily attract monocytes (19). To gain insight into additional molecular mechanisms by which S1P may contribute to RA pathogenesis, we profiled the secretion of inflammatory CC and CXC chemokines by FLSs stimulated with S1P. FLSs do not secret detectable levels of cytokines or chemokines, except for low amounts of MCP-1 in the resting state. Upon treatment with S1P, the secretion of IL-8, IL-6, MCP-1, and RANTES was strongly induced, thereby suggesting that S1P can contribute to and/or amplify the secretion of chemokines by cells of the inflamed synovium. Because immune cells express a wide repertoire of chemokine receptors, including IL-8, MCP-1, SDF-1, IP-10, and RANTES (35), our results suggest that S1P-mediated cytokine secretion may contribute to the recruitment and retention of inflammatory cells in RA. Chemokines such as MCP-1, SDF-1, IP-10, and RANTES enhance the migration and proliferation of FLSs and upregulate matrix metalloproteinase production by FLSs (27), indicating a direct role for FLSs in the destructive phase of RA beyond the regulation of immune cell trafficking.

Studies using S1P receptor agonists/antagonists revealed that S1P₂ and S1P₃ are probably involved in S1P-mediated cytokine/chemokine secretion in FLSs. The ability of the S1P₁ agonist SEW2871 to induce cell migration but not cytokine/chemokine secretion implies that S1P-driven FLS migration is independent of and not secondary to synthesis of CC or CXC chemokines (27).

In contrast to the induction of FLS migration and cytokine/chemokine secretion by S1P, this bioactive lipid had no effect on FLS proliferation (data not shown). Our observations differ from those of Kitano et al. (14), who reported that S1P induces FLS proliferation. The discrepancy between these two observations may be partly explained by the different experimental conditions. Kitano et al. performed their experiments in the presence of serum as opposed to our serum-starved culture of FLSs during the proliferation assay.

Because proliferation is scarce in RA synovium (36), dysregulation of apoptosis has been proposed to explain synovial hyperplasia (37). We also provide direct evidence for the inhibition of FLS apoptosis by S1P. Indeed, S1P appears capable of increasing cell survival and inhibiting apoptosis of various other cell types (38–40), including B lymphoblastoid cells derived from patients with RA (16).

TNF- plays a key role in the pathogenesis of RA. Khoa et al. (41) reported that TNF- can cross-talk with the G protein-coupled receptor adenosine A_2A receptor. Because S1P receptors are coupled to G proteins, we sought to determine whether the cross-talk between TNF- and G protein-coupled receptors is a general phenomenon. We provide direct evidence for the significant enhancement of S1P-induced cytokine/chemokine secretion, including IL-6, IL-8, MCP-1, and RANTES in FLSs primed with TNF-. Moreover, S1P and TNF- alone were not able to stimulate IP-10 secretion. Their combination, however, resulted in an impressive induction of IP-10 production, suggestive of cross-talk between TNF- and S1P receptors. These observations are in agreement with the findings of Kitano et al. (14), who reported that S1P and TNF- signaling pathways synergize. The increase in S1P-induced cytokine secretion by TNF- can be partly explained by the induction of S1P₃ expression by TNF-. Although a large number of genes upregulated by TNF- have been identified in FLSs (26), this is the first report of the regulation of S1P₃ receptor expression by TNF-, raising the possibility of a causal relationship between enhanced expression of the S1P₃ receptor and production of chemokines by S1P following priming of FLSs with TNF-. Indeed, the S1P_1/3 receptor antagonist VPC23019 and the selective S1P₃ antagonist CAY10444 did not reduce TNF--induced cytokine secretion but totally blocked that induced by S1P in TNF--primed FLSs. Together, these observations suggest that the proinflammatory environment potentiates some of the functional outcomes of the S1P/S1P receptor signaling pathway.

To identify the signaling molecules involved in the functional responses of FLSs to S1P, the activation of the classical signaling proteins involved in G protein-coupled receptor signaling was investigated. We suggest that the main pathway that regulates S1P-induced IL-8 and IL-6 secretion is Rho/Rho kinase related, and that p38 MAPK is also involved in this process. On the other hand, p42/44 MAPK, p38 MAPK and Rho kinase, but not JNK, are all involved in S1P-induced FLS motility. The results suggest that the coupling of S1P receptors to various heterotrimeric G proteins and, consequently, distinct downstream signaling pathways lead to downstream pathological phenomena, such as FLS invasion into cartilage and bone and recruitment of immune cells into the inflamed synovium. We previously reported that a lysophospholipid that is structurally related to S1P, lysophosphatidic acid (LPA), can also stimulate FLS cytokine secretion and migration by interacting with its cognate receptors (42). The functional responses induced by LPA, however, differ from those induced by S1P in that LPA-induced cytokine synthesis is strongly dependent on p42/44 MAPK activity, whereas signal transduction through p42/44 MAPK is dispensable for LPA-mediated FLS migration. As reported previously (24, 25), p42/44 MAPK is slightly phosphorylated in synovial cells from RA patients, even in the absence of stimulation. Basal levels of p42/44 MAPK may result from autocrine activation by unknown stimulatory factor(s) released by synoviocytes in culture.

In summary, the present study provides evidence that S1P/S1P receptor signaling may contribute to RA pathogenesis by stimulating FLS migration and cytokine/chemokine secretion and by inhibiting apoptosis. Although S1P receptors exert receptor subtype-specific responses, we also demonstrated a redundancy of function of the three S1P receptors expressed by FLSs. Furthermore, our data suggest that upregulation of S1P₃ receptor expression and enhanced S1P-induced cytokine secretion by TNF--primed FLSs may amplify the inflammatory process in RA. Moreover, our results suggest a role for S1P/S1P receptor signaling in the development and progression of RA, and also suggest that S1P₃ may represent a critical player in the events that take place in the RA synovium. Thus, subtype-specific antagonists for S1P receptors could be novel therapeutic modalities for limiting inflammation in the destructive phase of RA.

	ACKNOWLEDGMENTS

 
The authors would like to express thanks for the use of the Bioimaging Platform of the Centre de Recherche en Infectiologie, Centre de Recherche du CHUQ-CHUL, for Luminex analysis.

Manuscript received March 17, 2008 and in revised form June 19, 2008 and in re-revised form July 18, 2008.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

1. Pap, T., U. Muller-Ladner, R. E. Gay, and S. Gay. 2000. Fibroblast biology. Role of synovial fibroblasts in the pathogenesis of rheumatoid arthritis. Arthritis Res. 2: 361–367.[CrossRef][Medline]

2. Feldmann, M., F. M. Brennan, and R. N. Maini. 1996. Rheumatoid arthritis. Cell. 85: 307–310.[CrossRef][Medline]

3. Muller-Ladner, U., J. Kriegsmann, B. N. Franklin, S. Matsumoto, T. Geiler, R. E. Gay, and S. Gay. 1996. Synovial fibroblasts of patients with rheumatoid arthritis attach to and invade normal human cartilage when engrafted into SCID mice. Am. J. Pathol. 149: 1607–1615.[Abstract]

4. Kluk, M. J., and T. Hla. 2002. Signaling of sphingosine-1-phosphate via the S1P/EDG-family of G-protein-coupled receptors. Biochim. Biophys. Acta. 1582: 72–80.[Medline]

5. Rosenfeldt, H. M., J. P. Hobson, S. Milstien, and S. Spiegel. 2001. The sphingosine-1-phosphate receptor EDG-1 is essential for platelet-derived growth factor-induced cell motility. Biochem. Soc. Trans. 29: 836–839.[CrossRef][Medline]

6. Hla, T. 2004. Physiological and pathological actions of sphingosine 1-phosphate. Semin. Cell Dev. Biol. 15: 513–520.[CrossRef][Medline]

7. Yatomi, Y., F. Ruan, S. Hakomori, and Y. Igarashi. 1995. Sphingosine-1-phosphate: a platelet-activating sphingolipid released from agonist-stimulated human platelets. Blood. 86: 193–202.[Abstract/Free Full Text]

8. Spiegel, S., and S. Milstien. 2002. Sphingosine 1-phosphate, a key cell signaling molecule. J. Biol. Chem. 277: 25851–25854.[Free Full Text]

9. Yang, L., Y. Yatomi, Y. Miura, K. Satoh, and Y. Ozaki. 1999. Metabolism and functional effects of sphingolipids in blood cells. Br. J. Haematol. 107: 282–293.[CrossRef][Medline]

10. Xia, P., L. Wang, J. R. Gamble, and M. A. Vadas. 1999. Activation of sphingosine kinase by tumor necrosis factor-alpha inhibits apoptosis in human endothelial cells. J. Biol. Chem. 274: 34499–34505.[Abstract/Free Full Text]

11. Lin, D. A., and J. A. Boyce. 2006. Lysophospholipids as mediators of immunity. Adv. Immunol. 89: 141–167.[CrossRef][Medline]

12. Lin, C. I., C. N. Chen, P. W. Lin, and H. Lee. 2007. Sphingosine 1-phosphate regulates inflammation-related genes in human endothelial cells through S1P1 and S1P3. Biochem. Biophys. Res. Commun. 355: 895–901.[CrossRef][Medline]

13. Dorsam, G., M. H. Graeler, C. Seroogy, Y. Kong, J. K. Voice, and E. J. Goetzl. 2003. Transduction of multiple effects of sphingosine 1-phosphate (S1P) on T cell functions by the S1P1 G protein-coupled receptor. J. Immunol. 171: 3500–3507.[Abstract/Free Full Text]

14. Kitano, M., T. Hla, M. Sekiguchi, Y. Kawahito, R. Yoshimura, K. Miyazawa, T. Iwasaki, H. Sano, J. D. Saba, and Y. Y. Tam. 2006. Sphingosine 1-phosphate/sphingosine 1-phosphate receptor 1 signaling in rheumatoid synovium: regulation of synovial proliferation and inflammatory gene expression. Arthritis Rheum. 54: 742–753.[CrossRef][Medline]

15. Spiegel, S., and S. Milstien. 2003. Sphingosine-1-phosphate: an enigmatic signalling lipid. Nat. Rev. Mol. Cell Biol. 4: 397–407.[CrossRef][Medline]

16. Pi, X., S. Y. Tan, M. Hayes, L. Xiao, J. A. Shayman, S. Ling, and J. Holoshitz. 2006. Sphingosine kinase 1-mediated inhibition of Fas death signaling in rheumatoid arthritis B lymphoblastoid cells. Arthritis Rheum. 54: 754–764.[CrossRef][Medline]

17. Faour, W. H., A. Mancini, Q. W. He, and J. A. Di Battista. 2003. T-cell-derived interleukin-17 regulates the level and stability of cyclooxygenase-2 (COX-2) mRNA through restricted activation of the p38 mitogen-activated protein kinase cascade: role of distal sequences in the 3'-untranslated region of COX-2 mRNA. J. Biol. Chem. 278: 26897–26907.[Abstract/Free Full Text]

18. Marcil, J., D. Harbour, M. G. Houle, P. H. Naccache, and S. Bourgoin. 1999. Monosodium urate-crystal-stimulated phospholipase D in human neutrophils. Biochem. J. 337: 185–192.[CrossRef][Medline]

19. Koch, A. E. 2005. Chemokines and their receptors in rheumatoid arthritis: future targets? Arthritis Rheum. 52: 710–721.[CrossRef][Medline]

20. Ospelt, C., M. Neidhart, R. E. Gay, and S. Gay. 2004. Synovial activation in rheumatoid arthritis. Front. Biosci. 9: 2323–2334.[Medline]

21. Aupperle, K. R., D. L. Boyle, M. Hendrix, E. A. Seftor, N. J. Zvaifler, M. Barbosa, and G. S. Firestein. 1998. Regulation of synoviocyte proliferation, apoptosis, and invasion by the p53 tumor suppressor gene. Am. J. Pathol. 152: 1091–1098.[Abstract]

22. Borderie, D., P. Hilliquin, A. Hernvann, H. Lemarechal, C. J. Menkes, and O. G. Ekindjian. 1999. Apoptosis induced by nitric oxide is associated with nuclear p53 protein expression in cultured osteoarthritic synoviocytes. Osteoarthritis Cartilage. 7: 203–213.[CrossRef][Medline]

23. Anliker, B., and J. Chun. 2004. Cell surface receptors in lysophospholipid signaling. Semin. Cell Dev. Biol. 15: 457–465.[CrossRef][Medline]

24. Ha, J. E., Y. E. Choi, J. Jang, C. H. Yoon, H. Y. Kim, and Y. S. Bae. 2008. FLIP and MAPK play crucial roles in the MLN51-mediated hyperproliferation of fibroblast-like synoviocytes in the pathogenesis of rheumatoid arthritis. Febs J. 275: 3546–3555.[CrossRef][Medline]

25. Harigai, M., M. Hara, M. Kawamoto, Y. Kawaguchi, T. Sugiura, M. Tanaka, M. Nakagawa, H. Ichida, K. Takagi, S. Higami-Ohsako, et al. 2004. Amplification of the synovial inflammatory response through activation of mitogen-activated protein kinases and nuclear factor kappaB using ligation of CD40 on CD14+ synovial cells from patients with rheumatoid arthritis. Arthritis Rheum. 50: 2167–2177.[CrossRef][Medline]

26. Taberner, M., K. F. Scott, L. Weininger, C. R. Mackay, and M. S. Rolph. 2005. Overlapping gene expression profiles in rheumatoid fibroblast-like synoviocytes induced by the proinflammatory cytokines interleukin-1 beta and tumor necrosis factor. Inflamm. Res. 54: 10–16.[CrossRef][Medline]

27. Garcia-Vicuna, R., M. V. Gomez-Gaviro, M. J. Dominguez-Luis, M. K. Pec, I. Gonzalez-Alvaro, J. M. Alvaro-Gracia, and F. Diaz-Gonzalez. 2004. CC and CXC chemokine receptors mediate migration, proliferation, and matrix metalloproteinase production by fibroblast-like synoviocytes from rheumatoid arthritis patients. Arthritis Rheum. 50: 3866–3877.[CrossRef][Medline]

28. Okamoto, H., N. Takuwa, T. Yokomizo, N. Sugimoto, S. Sakurada, H. Shigematsu, and Y. Takuwa. 2000. Inhibitory regulation of Rac activation, membrane ruffling, and cell migration by the G protein-coupled sphingosine-1-phosphate receptor EDG5 but not EDG1 or EDG3. Mol. Cell. Biol. 20: 9247–9261.[Abstract/Free Full Text]

29. Balthasar, S., J. Samulin, H. Ahlgren, N. Bergelin, M. Lundqvist, E. C. Toescu, M. C. Eggo, and K. Tornquist. 2006. Sphingosine 1-phosphate receptor expression profile and regulation of migration in human thyroid cancer cells. Biochem. J. 398: 547–556.[CrossRef][Medline]

30. Ishii, I., N. Fukushima, X. Ye, and J. Chun. 2004. Lysophospholipid receptors: signaling and biology. Annu. Rev. Biochem. 73: 321–354.[CrossRef]