Effects of Lysophosphatidic Acid on Proliferation and Cytosolic Ca ++ of Human Adult Vascular Smooth Muscle Cells in Culture

Thrombosis Research 94 (1999) 317–326

REGULAR ARTICLE

Effects of Lysophosphatidic Acid onProliferation and Cytosolic Ca11 of HumanAdult Vascular Smooth Muscle Cells in CultureIsabelle Gennero1, Jean-Marie Xuereb2, Marie-Francoise Simon3,Jean-Pierre Girolami4, Jean-Louis Bascands4, Hugues Chap3, Bernard Boneu2 and Pierre Sie1

1Service d’Hematologie, Fac. Pharmacie; 2Lab. de Recherchesur la Thrombose, Hopital Purpan; 3INSERM U326; 4INSERM U388, Toulouse, France.

(Received 1 September 1998 by E. Angles-Cano; revised/accepted 18 December 1998)

findings support the proposal that LPA releasedAbstractfrom activated platelets is a mediator for smoothmuscle cell response at the site of vessel injury inLysophosphatidic acid (LPA) is a lipid mediatorhumans. 1999 Elsevier Science Ltd. All rightsgenerated by activated platelets and having variousreserved.effects on numerous cell types. We investigated

some effects of 1-oleyl LPA on vascular smoothKey Words: Lysophosphatidic acid; Smooth muscle cells;muscle cells cultured from adult human normalIntracellular calcium; Mitogenic effect; Atherosclerosisarteries. At micromolar concentrations, LPA in-

duced a mitogenic effect ([3H]-thymidine incorpo-ration and cell proliferation) on quiescent cells, Lysophosphatidic acid (LPA; 1-acyl-2-hydroxy-without an additional growth factor being required. sn-glycerol-3-phosphate) has attracted aThis effect was equipotent to that of 10% fetal calf great deal of attention in studies on biolog-serum, and it was accompanied by early (5 minutes) ical effects and signal transduction in various celland late (1–3 hours) phosphorylation of mitogen- types [1–3]. This simple phospholipid is a criticalactivated protein kinase. LPA inhibited cell migra- intermediate in de novo lipid biosynthesis [1,3,4]tion through collagen coated membranes, with or and is rapidly produced in significant quantitieswithout platelet-derived growth factor BB as che- upon platelet activation [4–7]. LPA displays multi-moattractant. LPA induced a typical biphasic Ca21 ple biological activities depending on the targetsignal response made up of a rapid first phase due cell type. These include proliferation of fibroblaststo Ca21 release from intracellular stores followed by and cultured rat vascular smooth muscle cellsa second wave due to external Ca21 influx. These (VSMC), intestinal smooth muscle cell contraction,

platelet aggregation, Ca21 mobilization, neurite re-traction, or inhibition of myeloma cell growth (seePresented in part at the XVIth Congress of the Internationalreference [3] for primary references). After intra-Society for Thrombosis and Haemostasis, Florence, Italy, 6–12

June, 1997. venous injection into animals, LPA elicits strongAbbreviations: LPA, lysophosphatidic acid; VSMC, vascular vasoactive effects depending on the species [8].smooth muscle cells; MAPK, mitogen-activated protein kinases;

Putative LPA cell surface receptors recently havePDGF, platelet-derived growth factor; PBS, phosphate-buffered-saline; FCS, fetal calf serum. been recognized among novel or previously de-Corresponding author: P. SIE, Lab. d’Hematologie, Pavillon scribed orphan G-protein linked receptors [9–11].Lefebvre, Hopital Purpan, 31059 Toulouse Cedex, France. Tel:

At least three distinct G-proteins (Gi, signaling the133 (5) 61 77 90 78; Fax: 133 (5) 61 49 76 10; E-mail: ,[email protected].. Ras-MAPK cascade; Gq, which links the receptor

0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved.PII S0049-3848(99)00004-3

318 I. Gennero et al./Thrombosis Research 94 (1999) 317–326

to phospholipase C; and G12/13, which mediates Rho Nitrocellulose membranes were from Amersham(Little Chalfont, Buckinghamshire, UK), and CDPactivation) are involved in the cellular response to

LPA [3]. Star chemiluminescent reagent was from New En-gland Biolabs Ltd. (Beverly, MA, USA). Type ILPA accumulates rapidly in thrombin-stimulated

platelets [12], mainly through phospholipase A2 at- collagen from calfskin was from Boehringer Mann-heim (Mannheim, Germany).tack of newly formed phosphatidic acid [5,6], and

is released extracellularly. Since it is produced andreleased by activated platelets during blood clot- 1.2. Cell Culturesting, LPA is a normal constituent of serum. Itscirculating level is in the range of 1–5 mM in an VSMCs were isolated by the explant techniquealbumin-bound form. Palmitoyl- and oleoyl-LPA from small pieces of normal mammary arteries ob-are predominant. The local LPA concentration in tained during vascular surgery and cultured as pre-the immediate vinicity of a platelet plug is expected viously described [13]. After first confluence, cellsto be much higher, where it may exert paracrine were dissociated by trypsin digestion, subculturedeffects on VSMC at the site of arterial vessel injury, in plastic flasks, and routinely grown in DMEMthereby participating in physiological or pathologi- containing 10% FCS. VSMCs were used betweencal processes as a platelet-derived lipid mediator. passages 2 and 7.

In the present study, using primary cultures of Primary VSMCs and subcultures were identifiedhuman adult VSMC, we examine the effects of by their characteristic “hill and valley” pattern ofLPA on cell migration, proliferation, and early re- growth and by positive immunofluorescence stain-sponses, such as changes in free cytosolic Ca21 and ing with an anti-human a actin antibody.mitogen-activated protein kinase (MAPK) phos-phorylation. 1.3. [3H]-Thymidine Incorporation

VSMCs were seeded in 96-well dishes in DMEM1. Materials and Methods containing 10% FCS at 104 cells per well. After 48

hours, the medium was changed for serum-free1.1. Materials DMEM containing 0.2% BSA to make them quies-

cent. After 72 hours, the wells were refilled with1-oleoyl-sn-glycero-3-phosphate (18:1 LPA), fatty fresh serum-free DMEM containing 0.2% BSAacid–free bovine serum albumin (BSA), thapsigar- and LPA at various concentrations or 10% FCSgin and neomycin were from Sigma Chemical Co. as a positive control and cultured for 40 hours(St. Louis, MO, USA). Fetal calf serum (FCS), Dul- without medium change. As in a preliminary study,becco’s Modified Eagle’s Medium (DMEM), and we had determined that label incorporation begantrypsin were from Seromed (Berlin, Germany). at about 18 hours after FCS refeeding, [3H]-thymi-

dine (1 mCi/well) was added at 16 hours for the[3H]-thymidine was from New Life Sciences Prod-last 24 hours. Then, cells were washed twice withucts (Boston, MA, USA), fura-2 AM from Molecu-PBS, trypsinized, aspirated on Titertek filters, andlar Probes Inc. (Eugene, OR, USA). Plastic materialwashed and dissolved in scintillation liquid forfor culture and inserts for migration experimentscounting (Betamatic SL 30; Kontron Instrumentswere from Nunclon Inter Med (Roskilde, Denmark).(Rotkreuz, Switzerland).The anti-human a-actin antibody and the anti-mouse

IgG-FITC conjugate were from DAKO (Copenha-gen, Denmark). Other chemicals were from Sigma. 1.4. Cell ProliferationPhosphate buffered saline (PBS: NaCl 137 mM,Na2HPO4. 8.1 mM, KCl 2.68 mM, KH2PO4, 1.47 mM, VCSMs were subcultured in 24-well plates at 2.104

pH 7.4) was purchased from Seromed. Acrylamide, per well, and, after 48 hours, they were made quies-bisacrylamide and products for electrophoresis were cent by serum deprivation for 72 hours as describedfrom BioRad (Hercules, CA, USA). The antibody above. Wells were then refilled with fresh serum-against phosphorylated p44 and p42 MAPKs was free DMEM containing 0.2% BSA with various

concentrations of LPA or 10% FCS as control. Asfrom Promega Corporation (Madison, WI, USA).

319I. Gennero et al./Thrombosis Research 94 (1999) 317–326

in preliminary studies, the time for doubling cell solved on Laemmli SDS polyacrylamide gel (9%acrylamide), electro-transferred to a nitrocellulosepopulation was about 72 hours after FCS refeeding;membrane, and the phosphorylated p42/44 MAPKscells were cultured for 72 hours without mediumwere detected by using CDP Start chemilumines-change. The cells were then washed twice with PBS,cent reagent.trypsinized, and counted using an electronic particle

counter (Coulter counter 2M, Miami, FL, USA).Cell viability was assessed by the trypan blue exclu- 1.7. Ca11 Mobilization Induced by LPAsion test.

The intracellular concentration of calcium, [Ca21]i,was determined by use of the fluorescent Ca21 indi-1.5. Migration Assayscator fura 2 acetoxymethyl ester (fura-2 AM) aspreviously described [15].Migration was measured using a 24-well insert sys-

VSMCs (105 cells/ml) suspended in DMEM con-tem containing a polycarbonate filter with pore sizetaining 10% FCS were subcultured on 2037-mmof 8 mm. The filters were coated with 100 mg/mLglass coverslips for 24 hours. Confluent cell-coatedtype I collagen in 0.2% (v/v) acetic acid overnight.coverslips were rinsed with HEPES buffer (10 mMConfluent monolayers of VSMCs were washedHEPES, 145 mM NaCl, 2.5 mM KH2PO4, 1 mMtwice in PBS and briefly exposed to a trypsin solu-CaCl2, 1 mM MgSO4, 10 mM glucose, pH 7.4, con-tion. Cells were washed in DMEM supplementedtaining BSA 0,1%) before loading with fura 2-AMwith 0.5% FCS, resuspended in serum-free DMEM(1 mM in HEPES buffer) for 45 minutes at 378C.containing 0.2% BSA, and put in the top chamberAfter two washes, the coverslips were mounted into(5.104 cells/well) of the insert. The lower chamberthe experimental chamber of the spectrofluorimetercontained the same buffer with or without PDGFand perfused at a rate of 6 ml/min with HEPES(40 ng/mL) and with or without LPA (0–50 mM). Thebuffer equilibrated at 378C containing various testmigration assay was allowed to proceed overnightsubstances: 100 mM neomycine, a nonspecific phos-at 378C in humidified atmosphere of 5% CO2 in air.pholipase C inhibitor, 2 mM EGTA, a Ca21 chela-At the end of the period, the cells on both sidestor, 1.5 mM thapsigargin, an inhibitor of the endo-of the filter were fixed with methanol and stainedplasmin reticulum Ca21 ATPases. Fluorescencewith Giemsa. The cells on the upper surface weremeasurements were made using a Spex Fluorilogmechanically removed and the cells remaining onspectrofluorometer (Spex Industry Inc., Edison,the underside of the filter were counted under NY, USA), set for alternative dual wavelength exci-

3200 magnification. Nine fields were counted per tation at 340 and 380 nm. The light emitted at 520filter, and all experiments were run in duplicate. nm was collected by a photomultiplier and passed

to a Spex system microcomputer, which averaged1.6. Immunoblots of the emission collected over a 0.50-second periodMitogen-Activated Protein Kinases at each excitation wavelength. Autofluorescence of

unloaded cells was found to be ,18% of the emittedSemi-quantitative evaluation of MAPK phosphor- signal and was subtracted from the fura-2–loadedylation (activation) was performed as previously fluorescence ratio (340:380). As previously men-described [14] with minor modifications. VSMCs tioned, [Ca21]i was calculated as [Ca21]i5Kd Xwere subcultured in 6-well plates (105 cells/well) [(R2Rmin)/(Rmax2R)] X l, where Kd (224 nM) is theuntil confluence. Then, they were made quiescent dissociation constant of the complex fura-2-Ca21 andby serum deprivation for 72 hours as above. Two Rmin, Rmax, and l are constant parameters that de-hours before LPA addition, the medium was pend on the optical system used. In our experimentalchanged. At various times after LPA addition, the conditions they were Rmin50.9, Rmax518, and l54.cells were washed twice with PBS, lysed with sam-ple buffer containing 10% sodium dodecyl sulfate 1.8. Statistical Methods(SDS), 10% glycerol, 100 mM dithiotreitol, 5%b mercapto-ethanol, 1% bromophenol blue, 0.5% Results, given as means6SEM, were analyzed usingTris HCl, pH 6.8, and processed for immunoblot Student’s t-test. Results were considered significant

for p,0.05.experiments. Briefly, proteins obtained were re-

320 I. Gennero et al./Thrombosis Research 94 (1999) 317–326

Fig. 1. [3H]-thymidine incorporation by quiescent VSMCtreated with various concentrations of LPA or 10% FCS forcomparison. Values represent the mean6SEM of triplicate

Fig. 2. Effects of LPA on VSMC density. VSMCs seededexperiments. Asterisks denote a significant difference fromat 10. 103/well in 24-well plates were made quiescent forthe control value without LPA, p,0.05.72 hours before stimulation by various concentrations ofLPA or 10% FCS for comparison. At the end of this period,cells were treated for 5 minutes with trypsin and countedin a Coulter counter. Values (mean6SEM) are expressed2. Resultsas the value of the cell number in treated wells comparedwith that in the well without LPA. Asterisks denote aLPA induces a dose-dependent mitogenic effect significant difference from the control well.

on VSMCs. Figure 1 shows the effects of LPAon [3H]-thymidine incorporation in serum-starvedVSMCs. The increase was statistically significant Figure 4 shows the effects of LPA on VSMCat the lowest LPA concentration used (1 mM) and migration. In the absence of PDGF as chemoattrac-maximal at 5 mM (2.5-fold the value obtained in tant, LPA strongly inhibited the passage of the cellsthe absence of LPA and 80% of that obtained with through the collagen-coated filter (Figure 4A). The10% FCS). Figure 2 shows the effect of LPA on same effect also was obtained when using uncoatedcell proliferation. Seventy-two hours after the ad- filter (not shown). The maximal effect (85% inhibi-dition of LPA, the cell number increased dose- tion) was obtained at 1 mM. At this concentration,dependently. The maximum cell number obtained LPA reduced the migration of the cells in a gradientin the presence of 2 mM LPA was 1.6-fold that of of PDGF by 75% (Figure 4B).control cells and close to that obtained with 10%FCS. Thymidine incorporation and the cell numberdeclined at higher LPA concentrations. Cell viabil-ity at 72 hours remained .95% at all LPA concen-trations.

Phosphorylation of MAPK was assayed by West-ern blot analysis using anti-phosphorylated p42-p44 MAPK antibody. Figure 3 illustrates a repre-sentative result out of three different experiments.Stimulation of serum-starved VSMCs with 5 mMLPA resulted in a biphasic increase in MAPK phos-phorylation. After a first peak at 5–10 minutes,phosphorylation rapidly declined until 30 minutes,

Fig. 3. MAPK phosphorylation in VSMCs exposed toafter which a smaller second phase persisted aboveLPA. Quiescent VSMCs were stimulated with 5 mM LPA

basal level over 3 hours. At 5 minutes, LPA in- for various periods of time (0–180 minutes) (top) or withduced MAPK phosphorylation in a dose-depen- various concentrations of LPA (0.5–50 mM) for 5 min-

utes (bottom).dent manner between 0.5 and 20–50 mM.

321I. Gennero et al./Thrombosis Research 94 (1999) 317–326

Fig. 4. Inhibition of VSMC migration by LPA. Results are expressed in number of cells per high-power field (HPF)counted on the underside of the filter after overnight migration assay. (A) Cells were put in the upper compartment, andbuffer with or without increasing LPA concentrations was placed in the lower compartment. The results refer to the effectof LPA on cell migration without chemoattractant. (B) Same as (A) but PDGF was added in the lower compartment,with or without LPA. The results refer to the effect of LPA on cell migration in a gradient of PDGF. The experimentswere performed in triplicate and expressed as mean6SEM. Asterisks denote a significant difference from spontaneousmigration (no LPA in the lower compartment).

322 I. Gennero et al./Thrombosis Research 94 (1999) 317–326

Figure 5 illustrates the Ca21 mobilization elicited from those found by others [17,18]. This may bedue in part to the use of different animal species.by LPA. Basal [Ca21]i of fura-2–AM loaded VSMCs

was 67610.5 nM. LPA application in Ca21-con- Tokumura et al. [17] have shown that 1-oleoyl-LPA induces the proliferation of cultured VSMCstaining buffer induced a typical biphasic signal,

with an immediate rise peaking within 30 seconds from rat aorta with a maximum effect at high con-centrations (30–100 mM). At 10 or 100 mM, LPAto two- to threefold the baseline, followed by a

secondary increase of similar amplitude and a long also displayed a synergistic effect with epidermalor basic fibroblast growth factors. Natarajan et al.tonic phase sustained as long as LPA stimulation

was applied (Figure 5a). Application of neomycine [18] have shown that exogenous LPA and its imme-diate precursor phosphatidic acid, increase 3H-thy-(100 mM) prevented the response to LPA (not

shown). The magnitude of the response was dose midine incorporation in cultured VSMCs from rab-bit femoral arteries over a concentration range ofdependent (Figure 5b). To determine whether the

increase in [Ca21]i induced by LPA was due to Ca21 10–100 mM. In our study, we were able to detectan effect at 1 mM. More recently, Seewald et al.release from intracellular stores or to an influx

from the extracellular medium, experiments were have shown effects on rat VSMCs similar to thosewe observed with a maximum DNA synthesis at 5repeated in the presence of 2 mM EGTA. In these

conditions, the baseline [Ca21]i of unstimulated mM LPA and a decrease in the effect at higherconcentrations [16]. A similar biphasic effect hascells was slightly reduced. The first phase of the

signal elicited by LPA remained unchanged, but been described on rabbit kidney proximal tubulecells [20] and on glomerular mesangial cells [14].the secondary increase was abolished. When extra-

cellular Ca21 was reintroduced in the presence of It was speculated that this complex response wasdue to the presence of more than one type of recep-LPA, an immediate and sustained increase of [Ca21]i

was observed (Figure 5c). In the absence of Ca21 tor with different affinities for LPA and oppositemitogenic and antimitogenic effects. The same hy-in the extracellular medium, application of thapsi-

gargin induced a transient increase of [Ca21]i indi- pothesis can be made for VSMCs in the presentstudy.cating the release of Ca21 from intracellular pools.

Under these conditions, LPA-induced Ca21 tran- At mitogenic concentrations of LPA, p42-44MAP kinases are phosphorylated according to bi-sients were abolished. Again, when extracellular

Ca21 was readmitted, an immediate Ca21 influx was phasic kinetics in rat-1 fibroblasts [21] and mesan-gial cells [14]. In the present study, we also foundobserved (Figure 5d), indicating that LPA can in-

duce Ca21 influx independently of intracellular biphasic kinetics of MAPK phosphorylation sug-gesting that a discrete pool of enzymes remainsCa21 release.

Repeated applications of LPA (5-minute periods active for a prolonged period. This second phaseof activation seem to be necessary for mitogenicfour times at 3-minute intervals, or 5-minute peri-

ods two times at 30-minute intervals) abolished the effect as reported in hamster fibroblast cells withother mitogenic agents [22].response to LPA but not to 1 mM bradykinin (not

shown), indicating homologous desensitization Our study demonstrates that LPA is mitogenicfor VSMCs without the absolute requirement of(Figure 5e).cooperative peptidic growth factors, as is also thecase for rat fibroblasts, epithelial, or endothelialcells [1]. In fibroblasts, LPA acts through a protein3. Discussionkinase C–independent, pertussin toxin–sensitivesignaling pathway involving the activation of p21rasRelatively few studies have been devoted to the

effects of LPA on smooth muscle cells of vascular and its downstream effectors p74raf-1/MAPK cas-cade [23–25]. Variations around this scheme haveorigin [16–19]. The present study is the first report

of LPA action on VSCMs from normal human been reported in other cell lines [21], and the pre-cise mechanism of LPA signal leading to MAPKarteries. At micromolar concentrations, LPA ap-

pears to be equipotent to 10% FCS in inducing phosphorylation in VSMCs remains to be investi-gated.DNA synthesis and cell division. Optimal LPA

concentrations reported in this study are different At mitogenic concentrations, LPA blocks VSMC

Fig. 5. [Ca21]i signals induced on VSMC by LPA. (a) Typical biphasic [Ca21]i signal response to 1 mM LPA. (b) Dose-response relationship of peak [Ca21]i signal response (% of baseline) to LPA. (c) Effect of EGTA (2 mM) on responseto 1 mM LPA. (d) Effect of the combined application of thapsigargin (TG 1.5 mM) and EGTA (2 mM) on the responseto 1 mM LPA. (e) Autologous desensitization by repeated expositions (5-minute periods) of VSMCs to 1 mM LPA.

324 I. Gennero et al./Thrombosis Research 94 (1999) 317–326

migration in the presence or absence of PDGF as mic reticulum channels (Figure 5d). It is known thatcapacitative Ca21 entry is operative in VSMCs [30].chemoattractant. A similar effect has been de-

In conclusion, we describe some features of thescribed with a parent phospholipid, sphingosine-1-response of VSMCs from human normal arteriesP, also released from activated platelets [26]. LPAto LPA. At micromolar concentrations, LPA blocksinduces cytoskeleton remodeling through a Rho-cell migration, behaves as a complete growth factordependent pathway, leading to contractile re-equipotent to 10% FCS, and induces an extensivesponse in smooth muscle cells, fibroblasts, endothe-mobilization of [Ca21]i. The importance of thislial cells, and neurites [3]. In contrast, LPA inducesplatelet-derived mediator in the natural woundchemotaxis in 3T3 cells through a Ras-mediatedhealing process or in the dynamics of atheroscle-pathway [27]. Our results suggest that, in the pro-rotic plaque remains to be determined.cess leading to the accumulation of VSMCs in the

arterial intima, LPA promotes the proliferation ofthe cells once they have migrated from the me-

The authors thank Dr. G. Fournial and Y. Glock (Department ofdia layer.Vascular Surgery) for providing human mammary arteries.The ability of LPA to induce Ca21 mobilization

has been evaluated in a large variety of cell types,and only few have been found unresponsive [28].Ca21 transients usually are observed within a wide Referencesrange of LPA concentrations, starting as low as 10nM in rat glomerular mesangial cells [19] or human 1. Moolenaar WH. Lysophosphatidic acid, a multi-fibroblasts [28]. In the present study, we found a functional phospholipid messenger. J Biolsignificant [Ca21]i LPA response beginning at 1027

Chem 1995;270:12949–52.M, very close in its magnitude and dose depen- 2. Moolenaar WH. LPA: A novel lipid mediatordency to that found with rat mesangial cells [19]. with diverse biological actions. Trends CellLPA induced a typical biphasic [Ca21]i waveform Biol 1994;4:213–9.susceptible to homologous desensitization, similar 3. Moolenaar WH, Kranenburg O, Postma FR,to that obtained in VSMCs with other well-estab- Zondag GCM. Lysophosphatidic acid: G-pro-lished agonists. The extent of the initial [Ca21]i rise tein signalling and cellular responses. Currwas higher than that found after LPA stimulation Opin Cell Biol 1997;9:168–73.of rat aorta VSMCs [17]. LPA-induced mitogenesis 4. Gaits F, Fourcade O, Le Balle F, Gueguen G,and Ca21 mobilization can be dissociated and can Gaige B, Gassama-Diagne A, Fauvel J, Sallesuse different pathways in fibroblast cells [23]. The JP, Mauco G, Simon MF, Chap H. Lysopho-complete suppression of Ca21signal in the presence phatidic acid as a phospholipid mediator: Path-of neomycine is in agreement with the crucial role ways of synthesis. FEBS Lett 1997;410:54–8.of phosphoinositide breakdown and rules out LPA- 5. Mauco G, Simon MF, Chap H, Douste-Blazyinduced translocation of Ca21 across the plasma L. Phosphatidic and lysophosphatidic acid pro-membrane by a simple ionophoretic effect, which duction in phospholipase C and thrombin-may occur with high LPA concentrations [29]. Ex- treated platelets. Possible involvment of aperiments in the absence of external Ca21 (Figure platelet lipase. Biochimie 1978;60:653–61.5c) or after thapsigargin treatment (Figure 5d) indi- 6. Eichholtz T, Jalink K, Fahrenfort I, Moolenaarcate that the immediate rise primarily results from WH. The bioactive phospholipid acid is re-Ca21 release from intracellular stores and the sec- leased from activated platelets. Biochem Jondary phase from the influx of extracellular Ca21. 1993;291:677–80.In these experiments, readmission of Ca21 in the 7. Fourcade O, Simon MF, Viode C, Rugani N,buffer perfusing the cells was followed by immedi- Leballe F, Ragab A, Fournie B, Sarda L, Chapate Ca21 influx. This indicates the switch-on of ca- H. Secretory phospholipase A2 generates thepacitative Ca21 channels as a result of emptying novel lipid mediator lysophosphatidic acid inCa21 stores by LPA-induced mobilization (Figure membrane microvesicles shed from activated

cells. Cell 1995;80:919–27.5c), or by thapsigargin inhibition of the endoplas-

325I. Gennero et al./Thrombosis Research 94 (1999) 317–326

8. Tokumura A, Fukazawa K, Tsukatani H. Ef- 18. Natarajan V, Scribner WM, Hart CM, Partha-sarathy S. Oxidized low density lipoprotein-fects of synthetic and natural lysophosphatidicmediated activation of phospholipase D inacids on the arterial blood pressure of differentsmooth muscle cells: A possible role in cellanimal species. Lipids 1978;13:572–4.proliferation and atherogenesis. J Lipid Res9. Hecht JH, Weiner JA, Post SR, Chun J. Ven-1995;36:2005–16.tricular zone gene-1 (vzg-1) Encodes a lyso-

19. Inoue CN, Forster HG, Epstein M. Effects ofphosphatidic acid receptor expressed in neuro-lysophosphatidic acid, a novel lipid mediator,genic regions of the developing cerebral cortex.on cytosolic Ca21 and contractility in culturedJ Cell Biol 1996;135:1071–83.rat mesangial cells. Circ Res 1995;77:888–96.10. Guo Z, Liliom K, Fischer DJ, Bathurst IC,

20. Bahir N, Kuhen K, Taub M. Phosphilipids reg-Tomei LD, Kieffer MC, Tigyi G. Molecularulate growth and function of MDCK cells incloning of a high-affinity receptor for thehormonally defined serum free medium. In

growth factor-like lipid mediator lysophospha- vitro Cell Dev Biol 1992;28A:663–8.tidic acid from Xenopus oocytes. Proc Natl 21. Cook SJ, Rubinfeld B, Albert I, Mccormick F.Acad Sci USA 1996;93:14367–72. RapV12 antagonized Ras-dependent activa-

11. An S, Dickens MA, Bleu T, Hallmark OG, tion of ERL1 and ERK2 by LPA and EGF inGoetzl EJ. Molecular cloning of the human rat-1 fibroblasts. EMBO J 1993;12:3475–85.Edg2 protein and its identification as a func- 22. Kahan C, Seuwen K, Meloche S, Pouyssegur J.tional cellular receptor for lysophosphatidic Coordinate, biphasic activation of 44 mitogen-acid. Biochim Biophys Res Comm 1997;231: activated protein kinase and 56 kinase by619–22. growth factors in hamster fibroblasts. J Biol

12. Gerrard JM, Robinson P. Identification of the Chem 1992;267:13369–75.molecular species of lysophosphatidic acid pro- 23. Van Corven EJ, Groenink A, Jalink K, Eich-duced when platelets are stimulated by throm- holtz T, Moolenaar WH. Lysophosphatidate-bin. Biochim Biophys Acta 1989;1001:282–5. induced cell proliferation: Identification and

dissection of signalling pathways mediated by13. Xuereb JM, Sie P, Boneu B, Constans J. Up-G proteins. Cell 1989;59:45–54.regulation of tissue factor expression by plate-

24. Howe LR, Marshall CJ. Lysophosphatidic acidlet-derived growth factor in human vascularstimulates mitogen-activated protein kinasesmooth muscle cells in culture. Thromb Hae-activation via a G-protein-coupled pathway re-most 1997;78:1520–6.quiring p21ras and p74raf-1. J Biol Chem 1993;14. Gaits F, Salles JP, Chap H. Dual effect of lyso-268:20717–20.phosphatidic acid on proliferation of glomeru-

25. Tigyi G, Dyer DL, Miledi R. Lysophosphatidiclar mesangial cells. Kidney Int 1997;51: 1022–7.acid possesses dual action in cell proliferation.15. Bascands JL, Pecher C, Rouaud S, Emond C,Proc Natl Acad Sci USA 1994;91:1908–12.Leuna Bascands JL, Pecher C, Rouaud S,

26. Bornfeldt KE, Graves LM, Raines EW, Iga-Emond C, Leuna Tack J, Bastie MJ, Burch R,rashi Y, Wayman G, Yamamura S, Yatommi

Regoli D, Girolami JP. Evidence for existenceY, Sidhu JS, Krebs EG, Hakomori SI, Ross

of two distinct bradykinin receptors on rat mes- R. Sphingosine-1-phosphate inhibits PDGF-angial cells. Am J Physiol 1993;264:F548–56. induced chemotaxis of human arterial smooth

16. Seewald S, Sachinidis A, Dusing R, Ko Y, Seul muscle cells: Spatial and temporal modulationC, Epping P, Vetter H. Lysophosphatidic acid of PDGF chemotactic signal transduction. Jand intracellular signalling in vascular smooth Cell Biol 1995;130:193–206.muscle cells. Atherosclerosis 1997;130:121–31. 27. Kundra V, Anand-Apte B, Feig LA, Zetter

17. Tokumura A, Iimori M, Nishioka Y, Kitahara BR. The chemotactic response to PDGF-BB:M, Sakashita M, Tanaka S. Lysophosphatidic Evidence of a role of Ras. J Cell Biol 1995;acids induce proliferation of cultured vascular 130:725–31.smooth muscle cells from rat aorta. Am J Phys- 28. Jalink K, Van Corven EJ, Moolenaar WH. Ly-

sophosphatidic acid, but not phosphatidic acid,iol 1994;267:C204–10.

326 I. Gennero et al./Thrombosis Research 94 (1999) 317–326

is a potent Ca21 mobilizing stimulus for fibro- 30. Pacaud P, Loirand G, Gregoire G, Mironneaublasts. J Biol Chem 1990;265:12232–9. C, Mironneau J. Noradrenaline-activated hep-

29. Hildebrandt JP. Lysophosphatidic acid induces arin-sensitive Ca21 entry after depletion of in-inositol phosphate and calcium signals in exo- tracellular Ca21 store in portal vein smoothcrine cells from the avian nasal salt gland. J muscle cells. J Biol Chem 1993;268:3866–72.Membr Biol 1995;144:49–58.