JOURNAL OF CELLULAR PHYSIOLOGY 149277-283 (1991)

Signal Transduction by bFGF, But Not TCFP1, Involves Arachidonic Acid Metabolism in Endothelial Cells VERONIQUE FAFEUR, ZHI PING JIANG, AND PETER BOHLEN*

Medical Research Division, American Cyanamid, Pearl River, New York 10965 (V.F., P.B.); Department of Biochemistry, University of Zurich, CH-8057 Zurich, Switzerland (Z.P.I.) We investigated the stimulation of early cellular events resulting from the interaction of the growth factor basic FGF (bFGF) and of the growth inhibitor transforming growth factor beta-type 1 (TGFPl), with their specific receptors on bovine endothelial cells. At mitogenic concentrations, bFGF stimulated the rapid release of arachidonic acid and its metabolites from ('H)-arachidonicacid labeled cells. When arachidonic acid metabolism was stimulated by addition of the calcium ionophore A231 87, the effect of bFGF was amplified. Nordihydrogua'iaretic acid, an inhibitor of the Iipoxygenase pathway of arachidonic acid metabolism, decreased the mitogenic effect of bFGF, whereas indomethacin, an inhibitor of the cyclooxygenase pathway, was ineffective. These findings suggest that metabolism of arachidonic acid to lipoxygenase products may be necessary for the mitogenic effect of bFGF. Basic FGF did not stimulate the production of inositol phosphates from cells labelled with myo-(2-'H)-inositolnor did it induce calcium mobilization, as measured by fura-2 fluorescence, indicating that bFGF does not activate phosphoinositide-specific phospholipase C in endothelial cells, but rather, that bFGF-induced arachidonic acid metabolism is mediated by another phospholipase. TGFPl, which inhibits basal and bFGF-induced endothelial cell growth, had no effect on arachidonic acid metabolism and inositol phosphate formation and did not prevent bFGF-induced arachidonic acid metabolism. These results suggest that the inhibitory action of TGFPl on endothelial cell growth occurs through different mechmisms. Basic fibroblast growth factor (bFGF) and transforming growth factor beta-type 1 (TGFP1) are potent stimulator and inhibitor, respectively, of endothelial cell proliferation (Baird and Bohlen, 1990; FraterSchroder e t al., 1986; Baird and Durkin, 1986). These factors modulate cell proliferation after binding to receptors in the plasma membrane of endothelial cells (Moscatelli, 1987; Fafeur et al., 1990). Little is known of the intra-cellular events that transduce the signals generated by the binding of bFGF or TGFPl to their respective receptor. Several groups have reported that activation of tyrosine kinase activity is a n early event in cellular mitogenic response to FGFs (basic or acidic FGF) (Huang and Huang, 1986; Coughlin et al., 1988; Friesel et al., 1989) and protein substrates of FGF-induced tyrosine kinase activity may include the FGF receptor itself and a phospholipase C (Burgess et al., 1990). Stimulation of tyrosine kinase activity by growth factors may induce phospholipase C-mediated hydrolysis of phosphoinositides and the generation of the second messengers inositol triphosphate and diacylglycerol. Conflicting results have been obtained regarding a role for phosphoinositide hydrolysis in the mechanism of action of bFGF. While bFGF was found to stimulate the formation of inositol phosphates (Brown et al., 1989) and the production of diacylglycerol (Tsuda e t al., 1985; Kaibuchi et al., 1986) in 3T3 fibroblasts, bFGF did not (0

1991 WILEY-LISS, INC.

stimulate inositol phosphate production in hamster fibroblasts (Magnaldo et al., 1986),in bovine epithelial lens cells (Moenner et al., 1987), or in PC12 cells (Sharma and Dahiya, 1989). The finding that bFGF can induce a mitogenic response without activation of phosphoinositide metabolism implies the existence of other growth-signaling pathway(s). There is also little known as to how the TGFp signal is transduced into the interior of the cell. Studies suggest that TGFPl does not stimulate tyrosine kinase phosphorylation (Libby et al., 1986) and S6 kinase activation (Like and Massague, 1986). Furthermore, it does not interfere with early molecular events induced by other growth factors, phosphoinositide hydrolysis, activation of protein kinase C, Naf/Ht antiport activity (Chambard and Pouyssegur, 1988). A recent study suggests G-protein activation (Howe et al., 1990) as a potential mechanism of TGFp signal transduction. Many hormones, neurotransmitters, and growth factors stimulate arachidonic acid metabolism in cells. Arachidonic acid metabolism may provide a further means for receptor-mediated signal transduction. After being released from phospholipids through activation of specific phospholipases (Lapetina, 19891, arachidonic Received April 10, 1991; accepted June 7, 1991 *To whom reprint requestsicorrespondence should be addressed.

278

FAFEUR ET AL.

acid is metabolized through the cyclooxygenase pathway, generating prostaglandins and thromboxanes, and through the lipoxygenase pathway, leading to the formation of hydroxyeicosatetraenoic acids (HETEs) and leukotrienes (Needleman et al., 1986). These metabolites have a large spectrum of biological activities. Endothelial cells synthesize predominantly prostacyclin (PGI,), as well as other eicosanoids, including prostaglandin E2 and various HETEs (Greenwald et al., 1979). Recently, Setty et al., (1987a) have suggested that lipoxygenase rather than cyclooxygenase products stimulate the basal proliferation of endothelial cells and in support of this conclusion i t was found that 15-HETE (Setty et al., 198713) and leukotriene C4 (Modat et al., 1987) are mitogenic for these cells. In this study, we investigated the effects of bFGF on arachidonic acid metabolism, inositol phosphate formation, and intra-cellular calcium mobilization in endothelial cells. In addition, we tested the hypothesis that TGFPl can inhibit the mitogenic effect of bFGF by interfering with early events induced by bFGF.

MATERIALS AND METHODS Materials Basic FGF was prepared from bovine brain as previously described (Gospodarowicz et al., 1984). TGFPl was prepared from human platelets as described by Assoian et al., (1983). Indomethacin was purchased from Sigma and nordihydroguaiaretic acid (NDGA) from Fluka. ('HI-arachidonic acid (80 Ci/mmol), myo(2-'H)-inositol (10 Ci/mmol), and ('HI-thymidine (20 Ci/mmol) were purchased from Amersham. Bradykinin and the ionophore A23187 were obtained from Boehringer Mannheim. Fura-2/acetoxymethylester was obtained from Molecular Probes (Junction City, OR). Cell proliferation assays Adult bovine aortic arch endothelial cells were isolated and subcultured as previously described (Gospodarowicz et al., 1984). For assays, cells were seeded in 24-well plates (8,000 cells per well) in 0.5 ml of Dulbecco's Modified Eagle Medium (DMEM, Gibco) containing 10% calf serum (Hyclone, Logan, UT). Following 2 h of incubation, agents to be tested (in DMEM containing 0.5% bovine serum albumin (BSA)) were added and the cells were grown for 5 days. Freshly prepared stock solutions of nordihydroguaiaretic acid or indomethacin in ethanol were diluted in DMEM containing 0.5% BSA just before adding to the cells. The final concentration of ethanol in the cell culture medium (C 0.1%) had no effect on the cell growth. At the end of the experiments, cells were detached from the culture dishes with trypsin and counted in a Coulter particle counter. DNA synthesis DNA synthesis was assayed by measuring incorporation of ('HI-thymidine into DNA a s previously described (Fafeur et al., 1990). Briefly, cells (10,000 cells/well) were grown in medium containing 10% calf serum for 24 h; the medium was then replaced with serum-free medium containing (3H)-thymidine (0.4 pCiiwel1). Twenty microliter aliquots of agents to be tested diluted in DMEM containing 0.5% BSA were

added and the cells were further incubated for 30 h. Thymidine incorporated into trichloroacetic acid (TCA)-precipitable material was assayed by liquid scintillation spectrometry.

Metabolism of (3H)-arachidonicacid Cells were seeded in 35 mm dishes a t a density of 50,000 cells per 2 ml of DMEM containing 10% calf serum. The following day, 0.5 pCi of (3H)-arachidonic acid was added to the culture media and the cells were further incubated for 24 h. The cells were then washed twice with 2 ml of Hank's buffered saline solution (HBSS) and allowed to equilibrate in 1ml of HBSS for 1h in a water-bath at 37"C, with gentle shaking. Media were then replaced by 1ml of HBSS containing bFGF or TGFPl and the cells were further incubated for 1h. In some experiments, the cells were incubated with the growth factors for 15 min and then the calcium ionophore A23187 (1 pM) was added for a n additional 1h. Stock solutions of the ionophore A23187 (10 mM) were prepared in dimethyl sulfoxide (DMSO) and diluted in HBSS. Control cells received the same final concentration of DMSO. At the end of experiments, cells were placed on ice, supernatants were collected, and the released radioactive material was counted in a liquid scintillation counter. Metabolism of (3H)-inositol-labeledlipids Cells were seeded in 60 mm dishes at a density of 600,000 cells per 6 ml of DMEM containing 10% calf serum. The following day, media were replaced by Medium-199 (Gibco) containing 1%calf serum and the cells were labeled with 15 pCi of my0-(2-~H)-inositol for 48 h. The cells were then washed twice with 2 ml of HBSS and allowed to equilibrate in 1 ml of HBSS at 37°C for 30 min. Incubations of the cells in the presence or the absence of growth factors or bradykinin was then performed for 30 sec at 37°C. In experiments were LiCl was used, cells were incubated with LiCl (10 mM) during the last 10 min of the equilibration period. Agents (bFGF, TGFP1, or bradykinin) were then added to the media containing LiCl and the cells were incubated for a n additional 10 min. Reactions were terminated by aspirating the media from the dishes and adding 2 ml of ice-cold TCA (15%, w/v) and the dishes were incubated at 0°C for 30 min to allow for extraction of the water soluble inositol phosphates. TCA-soluble extracts were removed and the dishes washed twice with 1ml of water which was then added to the extract. The extracts were washed four times with ethyl ether and neutralized with 0.1 M NaOH. An 8 ml aliquot of the upper aqueous phase was removed for analysis and inositol phosphates were separated by anion exchange chromatography using columns containing 1ml Dowex 1-X8 as previously described (Derian and Moskowitz, 1986). Measurement of intra-cellular calcium The measurements of intra-cellular calcium on cell suspension were performed using the fluorescent dye fura-2. Suspensions of cells were obtained by a brief trypsin-EDTA treatment of monolayers from 60 mm dishes. The concentration of cells in the suspension was adjusted to lo6 cells per ml using DMEM containing 1%

279

bFGF, TGFpl AND ARACHIDONIC ACID METABOLISM

A

B

11

A 2 3 1 87

10

5

T

C

F

0

C

F+T

C

F

F+T

T

Fig. 1. Effects of bFGF (A) and TGFpl (B) on (3H)-arachidonicacid metabolism. Cells (50,000 cellsidish) labeled with (3H)-arachidonicacid were incubated for 1 h with bFGF (4 ngiml) and/or TGFpl (1 ngiml), in the absence (white columns) or presence (grey columns) of the ionophore A23187 (0.5 I*.giml).Data are expressed as cpm released into incubation media. Values are means S.D. of triplicate incubations and experiment was repeated twice with similar results. Absence of error bars indicates that the S.D. was below the size of the symbol. C, control; F, bFGF; T, TGFp1.

*

calf serum. The cells were then kept in a CO, incubator and mixed by a gentle stirring. Loading of the cells with fura-2 was achieved by incubating the cell suspension a t 37°C for 45 min with 1.25 pM fura-2. For fluorimetric measurement, 1 ml of the cell suspension was centrifuged very briefly at 13,OOOg and the pellet resuspended i n 50 p1 of phosphate buffer saline (PBS) containing 25 mM Hepes. The cell suspension was then transferred to a thermostatically controlled (37°C) holder containing 2 ml of PBS and the fluorescent signal was measured with continuous stirring. Fluorescence was monitored in a Shimadzu fluorescence spectrophotometer with a n excitation wavelength of 339 nm and a n emission wavelength of 492 nm. Intracellular Ca++ was calculated using the formula of Tsien et al. (1982).

RESULTS Basic FGF, but not TGFPl, stimulates arachidonic acid metabolism Exposure of endothelial cells to a mitogenic concentration of bFGF (4 ngiml) for one h stimulated the release of (3H)-labeled material (arachidonic acid and metabolites) from endothelial cells that had been labeled for 24 h with (3H)-arachidonic acid (Fig. 1). In additional experiments we determined that the increases in (3H)-labeled material were detectable after 5 min of incubation, reached a maximum at 30 min and then remained elevated for at least 180 min, the longest time interval studied (data not shown). In contrast to bFGF, TGFPl (1ngiml) had no effect on the release of (3H)-labeled material from endothelial cells (Fig. 1). When bFGF and TGFPl were added together, TGFPl was not able to prevent bFGF-induced release of (3H)labeled material (Fig. 1).The concentration of TGFpl (1 ng/ml) t h a t we used in this experiment is sufficient to inhibit the mitogenic effect of bFGF (FraterSchroder et al, 1986; Baird and Durkin, 1986). The calcium ionophore A23187 has been used in many cell systems to stimulate increase in intracellular calcium and activation of phospholipases (Hong, 1988; Lapetina, 1989). The stimulatory effect of

31

P

al u)

m Q)

-

z

1

0

o !

I

I

I

I

1

.001

.01

.1

1

10

growth

factor

i

(nglml)

Fig. 2. Dose-response of bFGF and TGFpl on (3H)-arachidonicacid metabolism. Cells (50,000 cellsidish) labeled with (3H)-arachidonic acid were incubated for 1 h with increasing concentrations of bFGF (black circles) or TGFpl (white circles) in the presence of the ionophore A23187 (0.5 Fgiml). Values are means 5 S.D. of triplicate incubations.

bFGF on ('H)-arachidonic acid metabolism was observed whether or not the cells were treated with the calcium ionophore A23187, although the effect of bFGF was more pronounced in the presence of the calcium ionophore A23187 (Fig. 1, compare panels A and B). Even in the presence of the ionophore, TGFPl was inactive (Fig. 1B). In a subsequent experiment, performed in the presence of the ionophore, the effect of bFGF on (3H)-labeled material was shown to be dosedependent (Fig. 2), within the same range of concentration inducing cell proliferation. In contrast, at all concentrations tested, TGFpl was without effect on (3H)-arachidonic acid metabolism (Fig. 2).

Effect of inhibitors of arachidonic acid metabolism on bFGF-induced cell proliferation To determine if arachidonic acid metabolism is involved in bFGF or TGFPl action, we investigated the

FAFEUR ET AL

A

75

-

25

I

I

1

.l

1

10

NDGA

(pglml)

Fig. 3. Effects of NDGA and indomethacin on bFGF-induced endothelial cell proliferation. A Doseresponse of NDGA; B dose-response of indomethacin. Cells (8,000 cellsiwell) were grown in the absence (black circles) or in the presence of 4 ngiml bFGF (black squares),or in the presence of bFGF and 1ngiml TGFpl (white circles). Duplicate samples were counted on day 5 after plating. Individual values did not differ by more than 10% of the mean and experiments were repeated a t least twice with similar results.

effects of inhibitors of arachidonic acid metabolism (Rao et al., 1987; Sakai et al., 1988) on bFGF- or TGFpl-modulated cell proliferation. As shown in Figure 3A, a lipoxygenase inhibitor, nordihydrogua'iaretic acid (NDGA) (0.6-10 pg/ml) decreased the mitogenic effect of bFGF (4ngiml) in a dose-dependent manner, with maximal inhibition occurring a t a n NDGA concentration of 2.5 pg/ml. In contrast, the cyclooxygenase inhibitor indomethacin was ineffective a t concentrations up to 10 pg/ml (Fig. 3B). TGFPl(O.05 ng/ml) inhibited by 50% the mitogenic effect of bFGF and this effect was neither modified by NDGA nor by indomethacin (Fig. 3). As shown in Figure 4,NDGA also inhibited in a dose-dependent manner basal and bFGF-stimulated DNA synthesis.

-

0

0

1 .01

.1

NDGA

1

(pg/ml)

Lack of effect of bFGF and TGFPl on inositol Fig. 4. NDGA inhibits bFGF-stimulated DNA synthesis. Cells phosphates formation (10,000 cellsiwell) were incubated in the presence of increasing In endothelial cells labeled with my0-(2-~H)-inositol concentrations of NDGA with (black circles) or without (white circles) for 24 h, a brief exposure (30 sec) to bFGF or TGFPl did 4 ngiml of bFGF. Results are the means f S.D. of triplicate determinot induce production of inositol phosphates (Fig. 5A). nations. These agents also failed to increase cytosolic levels of inositol phosphates after a 10-min exposure in the presence of 10 mM LiC1, a n inhibitor of inositol monoDISCUSSION phosphatase (Hallcher and Sherman, 1980) (Fig. 5B). The present studies suggest that a n increase in In contrast, bradykinin, a vasoactive peptide, which is known to stimulate phosphatidylinositol turnover and arachidonic acid metabolism may be required for calcium mobilization in endothelial cells (Lambert bFGF-dependent mitogenesis, as shown by the ability et al., 1986) caused the induction of inositol phosphate of bFGF to stimulate the rapid release of arachidonic acid and its metabolites. This effect may be mediated by formation as expected (Fig. 5). lipoxygenase metabolites, as indicated by the specific bFGF does not stimulate intra-cellular inhibition of the bFGF mitogenic effect by the lipoxycalcium mobilization genase inhibitor NDGA, but not by the cyclooxygenase By using the fluorescent calcium indicator fura-2, we inhibitor indomethacin. The active arachidonic acid found that bFGF (10 ng/ml) had no effect on the basal metabolites have not been identified in this study, but level of intra-cellular calcium, whereas addition of interestingly, the lipoxygenase products 15-HETE and bradykinin (10 pg/ml) greatly stimulated a transient leukotriene C4 were previously found to be mitogenic increase in intra-cellular calcium (Fig. 6). These results for endothelial cells (Setty et al., 1987b; Modat et al., confirmed the findings of others that bFGF does not 1987). A mechanism by which arachidonic acid may be induce a rise in intra-cellular calcium (NBnberg et al., released from phospholipids involves the sequential 1990).

281

bFGF, TGFPl AND ARACHIDONIC AClD METABOLISM

' -A 6 r

0

x x

4 -

El

5 2

2.5

0

0

C

F

T

BK

C

F

T

BK

Fig, 5. Effect of bFGF, TGFp1, and bradykinin on inositol phosphate formation. Cells (600,000 cellsidish) labeled with my0-(2-~H)-inositolwere treated with diluent (C), 4 ng/ml bFGF (F), 1 ngiml TGFpl (T), or 1 pgiml bradykinin (BK) for 30 sec (A) or for 10 min in the presence of 10 mM LiCl (B). The data show cpm corresponding to inositol monophosphate (white columns), biphosphates (grey columns), and triphosphates (black columns). This experiment was repeated twice with similar results.

\

I

200 nM

BK

1

F

"

Fig. 6 . Effect of bFGF and bradykinin on intra-cellular C a + +mobilization. Traces represent typical responses to addition of 4 ngiml bFGF (F) or 10 pgiml of bradykinin (BK). Scale on right indicates cytoplasmic free Ca++ concentrations.

action of inositol phospholipid-specific phospholipase C and diacylglycerol lipase (Bell et al., 1979). This phospholipase C hydrolyzes phosphatidylinositol biphosphate to produce inositol triphosphate and diacylglycerol, which mediate, respectively, the release of calcium from intra-cellular stores and the activation of protein kinase C. Our data t h a t mitogenic concentrations of bFGF do not increase inositol phosphate formation and intracellular calcium, while the control agent, bradykinin, readily induces those responses indicate that bFGF does not cause activation of a phosphoinositide-specific phospholipase C in endothelial cells. These findings are in agreement with data of other investigators obtained with various cell types (Nbnberg e t al., 1990; Magnaldo et al., 1986; Moenner et al., 1987; Sharma and Dahiya, 19891, but conflict with results obtained in 3T3 cells showing that bFGF stimulated accumulation of inositol phosphates (Brown et al., 1989) and production of diacylglycerol (Tsuda et al., 1985; Kaibuchi et al., 1986). Although it is possible that bFGF may stimulate phosphoinositide metabolism in some, but not all, cell types, there is increasing evidence that biologically active agents, including bFGF (NBnberg et al., 19901, can activate

protein kinase C without inducing phosphoinositide breakdown through generation of diacylglycerol from other phospholipids, e.g. phosphatidylcholine (Exton, 1990). The existence of alternate pathways for arachidonic acid release from phospholipids is now being increasingly documented. These include activation of a phospholipase C-mediated hydrolysis of phosphatidylcholine rather than phosphatidylinositol biphosphate, yielding diacylglycerol, which can serve as a source of arachidonic acid (Exton, 1990), or activation of phospholipase A,, which can hydrolyze phosphatidylinositol, phosphatidylcholine, or phosphatidylethanolamine (reviewed in Hong, 1988). In endothelial cells, both phospholipase C and phospholipase A,-mediated pathways for arachidonic acid release have been identified. Some agents, such a s bradykinin, can stimulate the release of arachidonic acid through phospholipase A, activation, at least in part independently of phospholipase C activation (Hong and Deykin, 1982). Although it is difficult at present to determine which pathway can be effective in releasing arachidonic acid in endothelial cells, it is possible that a phospholipase A, activity is modulated by bFGF. In support of this possibility we found a potentiation of bFGF-induced arachidonic acid metabolism by the calcium ionophore A23187. This agent stimulates arachidonic acid release without receptor activation by directly increasing intra-cellular calcium ion concentrations. The known requirement of phospholipase A, for calcium (Hong, 1988) and the potentiation of the effect of bFGF obtained with the ionophore suggest a role of bFGF in modulating phospholipase A, activity. Finally, there is evidence that bFGF may stimulate phospholipase A, through a pathway that involves lipocortins. Basic FGF was recently shown to stimulate the phosphorylation of a lipocortinlike protein in bovine epithelial iens cells (Blanquet et al., 1990). Lipocortins have been shown to be substrates for growth factor-mediated tyrosine phosphorylation and it has been proposed that phosphorylation of lipocortins blocks their ability to inhibit phospholipase A, activity (Hirata et al., 1984). These observations suggest a potential mechanism for tyrosine kinaseactivating growth factors such as bFGF (Huang and

282

FAFEUR ET AL

Huang, 1986; Coughlin et al., 1988; Friesel et al., 1989) to modulate phospholipase A, activity via lipocortin phosphorylation. Finally, it is noteworthy that the growth inhibitor TGFPl had no effect on several signal transduction events and that it did not interfere with any of the growth factor-induced potential transducing mechanisms that we studied in endothelial cells. Those results are in agreement with other observations made in different cell systems that TGFPl does not affect known early signalling events, such a s phosphoinositide breakdown, activation of protein kinase C , Na+/H' antiport activity (Chambard and Pouyssegur, 19881, or S6 kinase activation (Like and Massague, 1986). This may indicate that TGFPl uses pathways which either are not shared by growth factors or interfere with growth factor-activated pathways at a point or points distal to the events investigated in this study.

Transforming growth factor-beta inhibits endothelial cell proliferation. Biochem. Biophys. Res. Commun., 137t295-302. Friesel, R., Burgess, W.H., and Maciag, T. (1989) Heparin-binding growth factor 1 stimulates tyrosine phosphorylation in NIH 3T3 cells. Mol. Cell. Biol., 9:1857-1865. Gospodarowicz, D., Cheng, J., Lui, G.M., Baird, A,, and Bohlen, P. (1984) Isolation of brain fibroblast growth factor by heparin Sepharose affinity chromatography: Identity with pituitary fibroblast growth factor. Proc. Natl. Acad. Sci. U.S.A., 81:6963-6967. Greenwald, J.E., Bianchine, J.R., and Wong, L.K. (1979) The production of the arachidonate metabolite HETE in vascular tissue. Nature, 281 588-589. Hallcher, L.M., and Sherman, W.R. (1980) The effects of lithium ion and other agents on the activity of myo-inositol-1-phosphatasefrom bovine brain. J . Biol. Chem., 255:10896-10901. Hirata, F., Matsuda, K., Notsu, Y., Hattori, T., and del Carmine, R. (1984) Phosphorylation at a tyrosine residue of lipomodulin in mitogen-stimulated murine thymocytes. Proc. Natl. Acad. Sci. U.S.A., 81t4717-4721. Hong, S.L. (1988) The release of arachidonic acid from cellular lipids. Prog. Allergy 44t99-139. Hong, S.L., and Deykin, D. (1982)Activation ofphospholipases A, and C in pig aortic endothelial cells synthesizing prostacyclin. J. Biol. Chem., 25737151-7154. ACKNOWLEDGMENTS Howe, P.H., Cunningham, M.R., and Leof, E.B. (1990)Inhibition of We wish to thank Dr. Joseph Pfeilschifter, CIBAmink lung epithelial cell proliferation by transforming growth factor-beta is coupled through a pertussis-toxin-sensitive substrate. GEIGY, Basel, Switzerland, for helpful suggestions and Biochem. J., 266t537-543. discussions and Therese Muller-Michel for her techni- Huang, S.S., and Huang, J.S. (1986) Association of bovine braincal assistance. We are grateful to Dr. Magdalena derived growth factor receptor with protein tyrosine kinase activity. Eisinger and Mildred Decker for reviewing the manuJ. Biol. Chem., 261:9568-9571. script. V.F. was supported by a fellowship from Kaibuchi, K., Tsuda, T., Kikuchi, A., Tanimoto, T., Yamashita, T., and Takai, Y. (1986) Possible involvement of protein kinase C and SANOFI, France. calcium ion in growth factor-induced expression of c-myc oncogene in Swiss 3T3 fibroblasts. J. Biol. Chem., 261t1187-1192. LITERATURE CITED Lambert, T.L., Kent, R.S., and Whorton, A.R. (1986) Bradykinin stimulation of inositol polyphosphate production in porcine aortic Assoian, R.K., Komoriya, A., Meyers, C.A., Miller, D.M., and Sporn, endothelial cells. J. Biol. Chem., 261 t15288-15293. M.B. (1983) Transforming growth factor beta in human platelets: Identification of a major storage site, purification, and character- Lapetina, E.G. (1989) The inositide and arachidonic acid signal system. Adv. Exp. Med. Biol. 2613285-293. ization. J . Biol. Chem., 258t7155-7160. Baird, A,, and Bohlen, P. (1990)Fibroblast growth factors. In: Peptide Libby, J., Martinez, R., and Weber, M.J. (1986) Tyrosine phosphorylation in cells treated with transforming growth factor-beta. J . Cell. Growth Factors and Their Receptors: Handbook of Experimental Physiol., 129t159-166. Pharmacology. M.B. Sporn and A.B. Roberts, eds. Springer Verlag, Like, B., and Massague, J . (1986) The antiproliferative effect of type New York, Vol. 95, pp. 369-418. beta transforming growth factor occurs at a level distal from Baird, A,, and Durkin, T. (1986) Inhibition of endothelial cell prolifreceptors for growth-activating factors. J. Cell. Physiol., I29:159eration by type beta-transforming growth factor: Interactions with 166. acidic and basic fibroblast growth factor. Biochem. Biophys. Res. Magnaldo, I., L'Allemain, G., Chambard, J.C., Moenner, M., BarriCommun., 138:476482. tault, D., and Pouyssegur, J. (1986) The mitogenic pathway of Bell, R.L., Kennerly, D.A., Stanford, N., and Majerus, P.W. (1979) fibroblast growth factor is not mediated through polyphosphoinosiDiglyceride lipase: A pathway for arachidonic acid release from tide hydrolysis and protein kinase C activation in hamster fibrohuman platelets. Proc. Natl Acad. Sci. U.S.A., 76:323%3241. blasts. J. Biol Chem, 261t16916-16922. Blanquet, P.R., Paillard, S., and Courtois, Y. (1990) Phosphorylation and lipocortin-like activity of a 34 kD surface protein in lens Modat, G., Muller, A., Mary, A., Gregoire, C., and Bonne, C. (1987) Differential effects of leukotrienes B4 and C4 on bovine aortic epithelial cells: Relation to mitogenesis induced by fibroblast endothelial cell proliferation in vitro. Prostaglandins, 3 3 5 3 1 4 3 8 . growth factor. Growth Factors, 3t15-23. Brown, K.D., Blakeley, D.M., and Brigstock, D.R. (1989) Stimulation Moenner, M., Magnaldo, I., L'Allemain, G., Barritault, D., and Pouyssegur, J. (19871 Early and late mitogenic events induced by FGF on of polyphosphoinositide hydrolysis in Swiss 3T3 cells by recombibovine epithelial lens cells are not triggered by hydrolysis of nant fibroblast growth factors. FEBS Lett., 247t227-231. polyphosphoinositides. Biochem. Biophys. Res. Commun., 146t32Burgess, W.H., Dionne, C.A., Kaplow, J., Mudd, R., Friesel, R., 40. Zilberstein, A,, Schlessinger, J., and Jaye, M. (1990) Characterization and cDNA cloning of phospholipase C-y, a major substrate for Moscatelli, D. (1987) High- and low-affinity binding sites for basic fibroblast growth factor on cultured cells: Absence of a role for low heparin-binding growth factor 1 (acidic fibroblast growth factor)affinity binding in the stimulation of plasminogen activator proactivated tyrosine kinase. Mol. Cell. Biol., 10t4770-4777. Chambard, J.C., and Pouyssegur, J . (1988) TGF-beta inhibits growthduction by bovine capillary endothelial cells. J. Cell. Physiol., 1313123-130. factor-induced DNA synthesis in hamster fibroblasts without affecting the early mitogenic events. J . Cell. Physiol., 1353101-107. NBnberg, E., Morris, C., Higgins, T., Vara, F., and Rozengurt, E. Coughlin, S.R., Barr, J., Coussens, L.S., Fretto, L.J., and Williams, (1990) Fibroblast growth factor stimulates protein kinase C in L.T. (1988) Acidic and basic fibroblast growth factors stimulate quiescent 3T3 cells without calcium mobilization or inositol phostyrosine kinase activity in vivo. J . Biol. Chem., 263:988-993. phate accumulation. J. Cell. Physiol., 143.232-242. Derian, C.K., and Moskowitz, M.A. (1986) Polyphosphoinositide hyNeedleman, P., Turk, J., Jakschik, B. A,, Morrison, A. R., and drolysis in endothelial cells and carotid artery segments. J . Biol. Leflcowitz, J. B. (1986) Arachidonic acid metabolism. Annu. Rev. Chem., 261;3831-3837. Biochem., 55369-102. Exton, J.H. (1990) Signaling through phosphatidylcholine break- Rao, G.H., Kishore, N.P., and White, J.G. (1987) Differential effects of down. J. Biol. Chem., 265t1-4. putative inhibitors on cytosolic and membrane associated platelet Fafeur, V., Terman, B.T., Blum, J., and Bohlen, P. (1990) Basic FGF lipoxygenase. Prostaglandins Leukot. Med., 26t281-290. treatment of endothelial cells down-regulates the 85-kDa TGFp Sakai, K., Fafeur, V., Vulliez-Le-Normand, B., and Dray, F. (1988) receptor subtype and decrease the growth inhibitory response to 12-hydro-peroxyeicosatetraenoicacid (12-HPETE) and 15-HPETE TGFp1. Growth Factors, 3.237-246. stimulate melatonin synthesis in rat pineals. Prostaglandins, Frater-Schroder, M., Muller, G . ,Birchmeier, W., and Bohlen, P. (1986) 35:969-976.

bFGF, TGFpl AND ARACIIIDONIC ACID METABOLISM

Setty, B.N.Y., Dubowy, R.L., and Stuart, M.J. (1987a) Endothelial cell proliferation may be mediated via the production of endogenous lipoxygenase metabolites. Biochem. Biophys. Res. Commun., 144t345-35 1. Setty, B.N.Y., Graeber, J.E., and Stuart, M.J. (198713) The mitogenic effect of 15- and 12-hydroxyeicosatetraenoicacid on endothelial cells may be mediated via diacylglycerol kinase inhibition. J . Biol. Chem., 262t17613-17622. Sharma, A,, and Dahiya, R. (1989) Basic fibroblast growth factor does

283

not initiate mitogenic signaling via phosphoinositide hydrolysis in PC12 cells. Indian J. Exp. Biol., 27:593-597. Tsien, R.Y., Pozzan, T., and Rink, T.J. (1982) T-cell mitogen causes early changes in cytoplasmic free Cat+ and membrane potential in lymphocytes. Nature, 295:68-71. Tsuda, T., Kaibuchi, K., Kawahara, Y., Fukuzaki, H., and Takai, Y. (1985) Induction of protein kinase C activation and C a ' ' mobilization by fibroblast growth factor in Swiss 3T3 cells. FEBS Letts., 191t205-210.