Basic fibroblast growth factor-stimulated arachidonic acid release in rat pancreatic acini: Sequential action of tyrosine kinase, phospholipase C, protein kinase C and diacylglycerol

Cell. Signal. Vol. 8, No. 7, pp. 487-496, 1996 ISSN 0898-6568/96 $15.00 Copyright © 1996 Elsevier Science Inc. PII S0898-6568(96)00104-0

ELSEVIER

Basic Fibroblast Growth Factor-Stimulated Arachidonic Acid

Release in Rat Pancreatic Acini: Sequential Action of Tyrosine Kinase, Phospholipase C, Protein Kinase C and Diacylglycerol Lipase

Wei Hou, Yoshiyuki Arita and Jean Morisset* DI~PARTEMENT DE MI~DEC1NE,

FACULTI~ DE MI~DECINE UNIVERSITI~ DE SHERBROOKE, SHERBROOKE, QUI~BEC, CANADA J1H 5N4

ABSTRACT. This study was performed to evaluate the effect of human recombinant basic fibroblast growth factor on arachidonic acid release from rat pancreatic acini and to determine the cellular mechanism involved. From enzymatic assays, basic fibroblast growth factor did not significantly stimulate phospholipase A2 activity, whereas it significantly increased diacylglycerol lipase activity. Validity of phospholipase A2 or diacylglycerol lipase inhibitors was confirmed by their ability to inhibit phospholipase A2 or diacylglycerol lipase activities. Basic fibroblast growth factor increased intracellular accumulation and extracellular release of arachidonic acid from metabolically labelled acinar cells in a concentration- and time-dependent manner. This effect was maximal with 50 pM basic fibroblast growth factor and became significant after a 5-min incubation period. The protein tyrosine kinase inhibitor, 0.5 mM genistein, inhibited arachidonic acid release in basic fibroblast growth factor-stimulated acini, whereas 100 IxM vanadate, a protein tyrosine phosphatase inhibitor, enhanced arachidonic acid release. Two phospholipase A2 inhibitors, mepacrine and aristolochic acid, failed to attenuate basic fibroblast growth factor-stimulated arachidonic acid release. A diacylglycerol lipase inhibitor RHC 80267 at 150 p~M and 50 ~M com- pletely inhibited 50 pM basic fibroblast growth factor-induced intracellular accumulation and extracellular release of arachidonic acid, respectively. Furthermore, basic fibroblast growth factor stimulated arachidonic acid release was also inhibited by 10 IxM U73122 and by 100 nM staurosporine, phospholipase C and protein kinase C respective inhibitors. Wortmannin, an inhibitor of basic fibroblast growth factor-stimulated phospholipase D, did not affect arachidonic acid release. 100 nM 413-phorbol 12-myristate 13-acetate also increased arachidonic acid release, an effect also inhibited by staurosporine. Taken together, these data demonstrate activation of diacylglycerol lipase and arachidonic acid release in pancreatic acini upon stimulation by basic fibroblast growth factor, and strongly indicate that arachidonic acid release in response to basic fibroblast growth factor depends upon the sequential action of tyrosine kinase, phospholipase C, protein kinase C and diacylglycerol lipase but not from phospholipase A2 nor phospholipase D activation. Copyright © 1996 Elsevier Science Inc. CELL SIGNAL 8;7:487--496, 1996.

KEY WORDS. bFGF, Arachidonic acid, Tyrosine kinase, Phospholipase A2, Phospholipase C, DAG lipase, Phospholipase D, Pancreatic acini

I N T R O D U C T I O N

Fibroblast growth factor (FGF) was originally identified from extracts of pituitary and brain that stimulated growth of 3T3 cells [1, 2]. The FGFs are a family of at least seven different polypeptides that share 30-40% homology. Basic

* Author to whom all correspondence should be addressed. Abbreviations: bFGF-basic fibroblas~ growth factor; PLC-phospholipase

C; PLA2-phospholipase a2; PLD- phospholipase D; DAG lipase~zliacyl- glycerol lipase; PKC-protein kinase C; PPH-phosphatidate phosphohydrolase; PEt-phosphatidylethanol; PlP2~hosphatidylinositol 4,5-bisphosphate; IPs-inositol 1,4,5-trisphosphate; PMA-4[3-phorbol 12 myristate 13 acetate; ETYA-eicosatetraynoic acid; SBTI-soybean trypsin inhibitor; TLC-thin layer chromatography; TCA-trichloroacetic acid.

Received 16 March 1996; and accepted 10 June 1996.

FGF (bFGF) remains the most studied member of this family and is highly mitogenic for a variety of mammalian cell types. In addition to its mitogenic activity, bFGF is involved in neuron survival, stimulation of cell migration, wound repair, embryo development, bone formation, anglo- genesis and malignant transformation [3, 4, 5]. bFGF oper- ates through specific cell surface tyrosine kinase receptors and activates second messengers in a cascade originating from their intrinsic receptor tyrosine kinase activity [6, 7]. However, comparable roles have never been established in the pancreas. Recently, the presence of low-and high-affin- ity receptors was reported on pancreatic acini and suggested bFGF as an unrecognized pancreatic secretagogue that may

488 W. Hou et al.

participate in the regulation of pancreatic exocrine func- tions [8]. bFGF was also shown to stimulate DNA synthesis in short-term cultured pancreatic cells, but the underlying mechanism for such an effect still remains unknown [9].

It has been reported that the protein substrates for bFGF- induced tyrosine kinase activity could be the bFGF receptor per se and PLCy [10]. PLC can hydrolyze the phosphatidylinositol 4,5-bisphosphate (PIP2) to produce inositol 1,4, 5-trisphosphate (IP3) and 1,2-DAG. DAG activates PKC and IP3 mobilizes Ca 2+ from intracellular stores [11, 12]. Among second messengers that can be formed through the activation of the various known effectors, arachidonic acid has been recognized as an important factor [13]. In the exocrine pancreas, arachidonic acid liberation in response to caerulein [14] or CCK-octapeptide [15] has been clearly demonstrated and both studies suggested its involvement in enzyme secretion. Arachidonic acid can be generated via two major pathways, one of which is PLA2, and the other involves a PiP2 metabolite, DAG, which is catalyzed by DAG lipase to liberate arachidonic acid [16, 17, 18]. In the exocrine pancreas, arachidonic acid may derive from a PLC-catalyzed breakdown of phosphatidylinositol followed by DAG lipase, but not from the involvement of PLAz in phosphatidylinositol degradation [14]. Another source would be a phospholipase A action on phosphatldylcholine [15]. Among other reported pathways, arachidonic acid may be generated from the activation of phospholipase As followed by lysophospholipase B [19] or through the sequential activation of PLD, phosphatidate phosphohydrolase (PPH) and di- and monoglyceride lipase [20].

Currently, there is convincing evidence to point to activation of tyrosine kinase as an early event involved in cellular responses to bFGF [6, 21, 22]. Most previous studies on bFGF-induced intracellular signalling events have largely focused on its role in phosphoinositide hydrolysis, PKC activation or calcium mobilization; however, there is no con- sensus of opinion and no definitive conclusions that can be drawn from these studies [10, 23, 24, 25]. It seems that activation of PLC and/or PKC by bFGF remains a cell- or tissue-specific event, with little attention being paid to the mechanism of bFGF-induced arachidonic acid release. Data from two studies suggested that stimulation of arachidonic acid release from bovine aortic arch endothelial cells and Swiss 3T3 cells by bFGF is independent of PLC activation but mediated via PLA2 [26, 27]. Recently, bFGF has been shown to stimulate pancreatic acinar cell phospholipase D (PLD) activity [28], but its implication in PLA2 activation is still unknown in this organ.

Therefore, by determining PLA2 and DAG lipase enzyme activities and by blocking selectively PLC-DAG lipase, PKC, PLD and PLA2 pathways, we investigated the potential mechanisms by which bFGF acts to release arachidonic acid in pancreatic acinar cells. We used selective inhibitors of potential enzymes involved in arachidonic acid release as tools to distinguish arachidonic acid release from different pathways. Our data provide strong evidence for the first time, that upon occupancy of the cell surface bFGF recep-

tor, PLC-DAG lipase pathway is sequentially activated and free arachidonic acid is increased in acinar cells and release from them. This process does not seem to involve PLA2 nor PLD pathways but the sequential activation of tyrosine kinase, PLC, PKC and DAG lipase.

MATERIALS A N D METHODS Materials

Bovine serum albumin (BSA, Fraction V and BSA fatty acid free), soybean trypsin inhibitor type 2-S (SBTI), N-2- hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), standards for thin layer chromatography (TLC), propranolol, 5,8,11,14-eicosatetraynoic acid (ETYA), arachidonic acid (5,8,11,14-eicosatetraenoic acid), orthovanadate, sucrose, ethyleneglycol-bis-(13-aminoethylether) N,N,N', N'-tetraacetic acid (EGTA), ethylenediamine tetraacetic acid (EDTA), leupeptin, pepstatin, phenylmethylsulfonyl fluoride (PMSF). aristolochic acid and mepacrine were purchased from Sigma (St. Louis, MO). Phosphatidylethanol (PEt) was from Avanti Polar Lipids (Birmingham, AL). Pu- rified collagenase (1,424 units/mg) was from Worthington Biochemicals (Freehold, NJ). RHC-80267 was from Calbio- chem (La Jolla, CA). Staurosporine was from Kyowa Hakko USA Inc. (New York, NY). U73122 was from Biomol Res Lab Inc.. Wortmannin was a gift from Sandoz, Canada. Genistein was purchased from INC. Biochemical, Aurora, Ohio, USA. Human recombinant basic flbroblast growth factor was from Collaborative Research Incorporated (Mass, USA). Silica gel G TLC plates (LK6D) were from What- man (Clifton, NJ). Solvents for TLC and silica gel (28-200 mesh) were from Fisher (Pittsburg, PA). pH] myristic acid (56 Ci/mmol), 1-stearyl-2-p4C] arachidonyl-glycerol (56 mCi/mmol) and 1-palmitoyl-2-piC] arachidonoyl-phosphatidylcholine (53 mCi/mmol) were from Amersham (Arling- ton, IL). [3H]-arachidonic acid (221 Ci/mmol) was from NEN. Male Sprague-Dawley rats (200-240 g) were from our own colony.

METHODS Preparation of Pancreatic Acini

Pancreases from rats fasted overnight were removed and trimmed of fat and mesentery. A suspension of pancreatic acini was prepared as reported by Peikin [29]. Acini from 5 pancreases were resuspended in 32 ml of an enriched HEPES-buffered solution [(in raM) 24.1 HEPES, 98 NaCI, 6 KC1, 2.1 KH2PO4, 0.5 CaC12, 1.2 MgC12, 5 sodium py- ruvate, 5 sodium fumarate, 5 sodium glutamate, and 11.4 glucose, as well as 0.01% (wt/vol) SBTI, 2.5% (vol/vol) glu- tamine, 1% (vol/vol) essential vitamin mixture, and 1% (vol/vol) BSA, adjusted to pH 7.4]. For experiments with [3H]-myristic acid incorporation, fatty-acid-free BSA was used at concentration of 0.5% (wt/vol) in the same HEPES- buffered solution as described above. For experiments with [3HI- arachidonic acid incorporation, HEPES.buffered solution without BSA was used.

bFGF and Arachidonic Acid Release 489

Uptake of [3H]. Arachidonic Acid into Pancreatic Acinar Cells

Acini from 5 pancreases were resuspended in 32 ml of HEPES-buffered solution and divided into flasks of 5 ml each. Acinar cells in each flask were incubated with [3H]- AA (2 IxCi/ml) for 120 min at 37°C. At the start of the incubation period (1 rain) and every 30 min thereafter, 1 ml of acini suspension was removed from each flask followed by a 30 s centrifugation to discard the supernatant. The arachidonic acid incorporation was ended by the addition of 2 ml of 5% trichloroacetic acid (TCA) to the pellets. The mixture of the acinar cells and TCA was vortexed vigorously and centrifuged again. The radioactivity present in the supematant and in the TeA-precipitated materials was then determined after resuspension of the pellet in 1 N NaOH and addition of scintillation fluid and expressed as percent of the total radioactivity present in acini.

DAG Lipase Assay

Cell extracts were prepared as described by Cybulsky [30]. Acini were pretreated with DAG lipase inhibitor RHC 80267 at 150 IxM in the appropriate groups for 15 min. A 15-min incubation followed in the presence or absence of 50 pM bFGF. At the end of the incubation, acini were centrifuged and the pellet obtained was washed twice with an homogenization buffer containing 50 mM HEPES, 0.25 M sucrose, 1 mM EDTA, 1 mM EGTA, 20 btM leupeptin, 20 tzM pepstatin, 0.1 mM PMSF, and 0.01% SBTI (w/v), pH 7.4. Acinar cells were disrupted in a glass-glass homogenizer and the homogenate was centrifuged at 1,000 g for 10 min to separate membrane components. The supematant was collected and proteins were determined by the method of Bradford [31]. Assay of DAG lipase was adapted from Pres- cott and Majerus [32]. The substrate mixture of 1-stearyl- 2-[14C] arachidonyl-glycerol and unlabelled DAG were dried under nitrogen and suspended in 0.05% Triton X-100. After sonication for 20 s, 5 btl aliquot of the substrate mixture was distributed into tubes to give final 200 IzM concentration of DAG. The assay was initiated by adding the enzyme source containing 16 p~g of protein and 7 mM CaCI2 and terminated by adding ethanol containing 2% acetic acid. Incubation was carried out for 30 min at 37°C. Sam- ples containing lipid standard were applied on thin layer chromatography plates. A solvent system consisting of hex- ane/diethyl ether/acetic acid (80/20/2) was used to separate arachidonic acid which was identified as migrating with authentic standards detected using I2 vapour. Areas containing arachidonic acid were scraped and radioactivity was determined in a liquid scintillation counter. Results are expressed as DPM/mg protein.

Phospholipase .42 Assay

PLA2 assay was performed according to the method described by Jelsema [31]. After they were incubated with 5 I~M Cch or 50 pM bFGF for 15 rain in the presence or absence

of inhibitors, pancreatic acini were washed twice with an homogenization buffer containing 10 mM HEPES (pH 7.4), 1 mM EDTA, 1 mM EGTA, 0.34 M sucrose, 10 jxg/ml leupeptin, 0.01% SBTI (wt/vol) and 1 mM PMSF. After homogenization in a glass-glass homogenizer, nuclei and debris were removed by centrifugation at 500 g for 5 min. The low-speed supernates were spun at 20,000 rpm for 1 h. The supernates thus obtained were assayed for protein content [31] and for cPLAz activity using L-a-l-palmitoyl-2-[~4C - arachidonyl] phosphatidylcholine (14C-PAPC) as substrate. Unlabelled dipalmitoyl phosphatidylcholine (0.08 mg/ml) was added to the radiolabelled IiC-PAPC (107 dpm/ml) and the solvent (toluene/ethanol, 1/1) evaporated under N2. The substrate mixtures were resuspended in 0.12 M Tris- HCI, pH 8.8, and solubilized by sonication. They were incubated for at least 2 h at 37°C before use to allow rean- nealing. PLA2 assay was then initiated by adding 20 btl of substrate, 75 ~g of cytosolic protein in a reaction buffer, pH 8.8, containing 30 mM Tris-HCl, 5 mM CaCI2, 40 mM MgCI2, 0.6 mM NaC1, 4 mM glutathione in a total volume of 350 bt 1. Incubations were carried out at 37°C under slow agitation for 15 min. Reactions were stopped by addition of 1.25 ml of Dole's reagent (isopropanol/n-heptane/1 N H2SO4, 40/10/1). The enzymatically released 14C-arachidonate was extracted by addition of 1.5 ml of n-heptane and 1.0 ml of H20. Samples were vortexed vigorously and centrifuged at 1,000 g for 10 min. Silica gel of 150 mg (28-200 mesh, Fisher) were added to 1.5 ml of upper phase to remove the remaining radiolabelled substrate. Samples were vortexed, centrifuged at 1,000 g for 10 min, and the radioactivity of 1-ml aliquot of supematant was determined in a scintillation counter after the addition of 10 ml of scintillation cocktail. PLA2 activity was expressed as nmol of 14C-AA released/mg of proteins/min. Background control values were subtracted.

Determination of Arachidonic Acid Release

The method described by Sato was modified and used to measure arachidonic acid [34]. Pancreatic acinar cells were labelled with pH]-arachidonic acid (2 IzCi/ml) for 90 mm at 37°C in the HEPES-buffered solution without BSA. The labelled acinar cells were washed three times with incubation buffer and resuspended in the incubation medium for an additional 30 min without the labelled arachidonic acid. The acinar cells were then incubated in the presence or absence of bFGF at the indicated concentrations in the HEPES-buffered solution containing 0.5% (wt/vol) fatty- acid-free BSA. ETYA was added in the incubation buffer prior to and during the stimulation to prevent arachidonic acid from being further hydrolyzed into its metabolites and therefore to facilitate observation of arachidonic acid release. Arachidonic acid and other lipids were extracted sep- arately from acini and from the incubation medium by adding methanol/chloroform/HC1 (200/200/1, vol/vol/vol) to the acinar cells and to a 200 bd aliquot of incubation medium. The mixture was well vortexed and the phases were separated by centrifugation after adding H:O. The upper

490 W. Hou et al.

aqueous phase was discarded. The samples of the lower or- ganic phase with standards added were dried under a stream of nitrogen, redissolved in 50 Ixl of chloroform, applied on TLC plates and developed in a solvent system consisting of petroleum ether/diethyl ether/acetic acid (60/45/1, vol/vol/ vol). Arachidonic acid was identified as migrating with an authentic arachidonic acid standard detected by I2 vapour. Areas containing arachidonic acid were scraped and radioactivity was determined in a liquid scintillation counter. In- tracellular arachidonic acid was expressed as a percent of total radioactivity into the acini and extracellular arachidonic acid as a percent of radioactivity in the phospholipids. Ra- dioactivity in phospholipids was calculated by subtracting the radioactivity in intracellular arachidonic acid from that of total radioactivity.

Determination of Phosphatidylethanol (PEt)

PEt was determined according to a method described previously [35]. After pancreatic acinar cells were labelled with pH]-myristic acid (5 txCi/ml) for 1 h at 37°C in HEPES-buffered solution containing 0.5% fatty-acid-free BSA and washed, they were incubated with or without bFGF for 20 rain in the medium containing 1.0% ethanol (vol/vol). At the end of incubation period, 1 ml of acini was removed and processed as described [35].

In all experiments where inhibitors of PLC, PLD, PLA2, PKC, DAG lipase, tyrosine kinase and protein tyrosine phosphatase were used, acini were always preincubated with each inhibitor for a 15-rain period and then incubated for the indicated time periods in the presence or absence of bFGF with the different inhibitors. These inhibitors were chosen to neutralize potential enzymes involved in arachidonic acid release. ETYA, a cyclooxygenase and lipoxygenase inhibitor [36], was used to limit arachidonic acid degradation for a better evaluation of its intracellular accumulation and extracellular release. The PLC inhibitor U73122 [37], the PLD inhibitor wortmannin [38], the PPH inhibitor propranolol [39,401 were used, in combination with the DAG lipase inhibitor RHC 80267 [41], to distinguish arachidonic acid generated from PLC and/or PLD activated pathways. Two PLA2 inhibitors, mepacrine [42] and aristolochic acid [43], were also used to assess the involvement of PLA> Fi- nally, the tyrosine kinase inhibitor genistein [44] was used to evaluate the importance of tyrosme phosphorylation of the cell surface receptor.

Data were analysed by analysis of Variance (ANOVA) and Fisher and Scheff~ tests. Significance was taken at the 5% level (P < 0.05).

RESULTS Uptake of [3HJ.Arachidonic Acid into Pancreatic Acini

As shown in Fig. 1A, uptake of [3H]-arachidonic acid into pancreatic acini was linear for 120 min and the incorporation of radioactivity into pancreatic phospholipids was also linear for 90 min (Fig. 2B). In light of these initial data, all subsequent studies involved a 90-rain labelling period with

1'21 A e~ . . 1 ,0

, , , =~ =~ 0,8

..= ~ 1

. ~ 0,6

• m,q ~

I~ ~. 0,2 . . . . ,

.•.• ,.-, 60

== ~ 50 ,,,= =. ~ 4o

"~ "-.---" 0 t _ .

I I I I I

1 30 6 0 90 1 2 0

T ime (min)

FIGURE 1. Time course of [3H]-arachidonic acid accumulation into pancreatic acini (A), and phospholipids (B). Pancreatic acini were labelled with [3H]-arachidonic acid as described in Materials and Methods. At the indicated time (A, B), 1 ml of acini suspension was removed and the acini were treated with 2 ml of 5% TCA. Radioactivity in the supernatant (A) represents free [3H]-arachidonic acid in the cells and that in the TCA-precipitated material (B) represents [3H]-arachidonic acid into phospholipids. Results represent means ± SE of 5 different experiments.

PH]-arachidonic acid. Intracellular accumulation of free [3H]-arachidonic acid into pancreatic acini and its extracellular release by bFGF were significantly improved by 10 IxM ETYA, the cyclooxygenase and lipoxygenase inhibitor (data not shown). From this observation, this inhibitor was systematically used in all upcoming studies.

D A G Lipase and PLA2 Assays

Enzymatic assays of both enzymes were performed to determine whether they could be significantly activated by bFGF and if so, whether described related inhibitors can inhibit their activities, respectively. Table 1 shows that bFGF significantly increased DAG lipase activity by 141%, an effect which was totally prevented by the DAG lipase inhibitor RHC 80267 as well as by the PLC inhibitor U 73122. How-


.'R-

= ~ ~ ,o0

N #

, - , , , • ~ 0

N '~ "= 200 D ~ / ,

.~--= ,0

~ e

= . -

bFGF Time (rnin)

FIGURE 2. Time course (A, B) and bFGF-dose response curve (C, D) on intracellular [3H]-arachidonic acid accumulation and extracellular release. After a 15-min pretreatment with 10 ttM ETYA, labelled pancreatic acini were incubated for different time periods with 50 pM bFGF (A, B) or for 20 min with increasing concentrations of bFGF (C, D). Intracellular and extracellular [3H]-arachidonic acid were measured as described in the Methods section. In these experiments, results represent means ___ SE of 5 different experiments. *." significantly different from their respective control at p < 0.05. O - - O : basal; 0 - - 0 : bFGF 50 pM.

ever, our data also indicate that bFGF did not activate PLA2 activity. Furthermore, the observation that carbamylcholine significantly increased PLA2 activity by 153%, an effect totally inhibi ted by mepacrine and aristolochic acid, two PLA2 inhibitors, clearly demonstrates that our PLA2 assay is accurate and valid and that bFGF can not act ivate this enzyme in pancreat ic acini. 5 IxM carbamylchol ine also significantly increased D A G lipase activity and this increase was

totally inhibi ted by 150 IzM R H C 80267 (data not shown). Mepacrine or aristolochic acid did not affect bFGF- or carbamylcholine-induced D A G lipase activity (data not shown). Since these data on PLAz and D A G lipase activities have clarified the efficacy and specificity of the inhibitors used, the other studies were performed using arachidonic acid release as measures of PLAz and D A G lipase activities in asso- ciat ion with the appropriate inhibitors.

Time Course and bFGF-Concentration Dependency on Intracellular Accumulation and Extracellular Release of [3HJ-Arachidonic Acid

As shown in Fig. 2A, bFGF significantly increased [3H]-arachidonic acid intracellular accumulat ion above basal values as early as 5 min for up to 60 rain. Similarly, bFGF also increased significantly above basal values within 5 min and for at least 60 min, the [3H]-arachidonic acid release into the incubat ion medium (Fig. 2B). The intracellular accumulat ion of [3H]-arachidonic acid was bFGF-concent ra t ion dependent with a threshold effect at 5 pM and a maximal effect at 50 pM (Fig. 2C). The effect of increasing concentrations of bFGF on [3H]-arachidonic acid extracellular release is depicted in Fig. 2D; as for intracellular arachidonic acid accumulation, bFGF exhibi ted an initial significant increase at 5 pM with a maximal effect at 50 pM.

Potential Role of Tyrosine Kinase in bFGF-induced [3HJ.Arachidonic Acid Accumulation and Release

As shown in Figs. 3A and B, 0.5 mM genistein, a protein tyrosine kinase inhibitor, totally inhibi ted both intracellular accumulat ion and extracellular release of PH]- arachidonic acid induced by bFGF. In contrast, 100 IzM

TABLE 1. Cytosolic PLA2 and DAG lipase activities

PLA2 activity DAG lipase activity nmol/mg protein/min dpm/mg protein

Control 12.5 _+ 0.5 50 pM bFGF 13.9 + 0.6 50 pM bFGF + 10 IxM U 73122 50 pM bFGF + 150 IzM RHC 80267 5 IxM Cch 31.7 -+ 1.9" 5 IzM Cch + 150 ~M 13.3 + 0.6 Mepacrine 5 IxM Cch + 50 txM 12.8 _+ 0.7 Aristolochic acid

57.8 -+ 3. 139.8 -+ 9.7* 61.6 -+ 3.4

58.3 + 3.8

Results for PLA2 and DAG lipase assays are from 6 experiments and expressed as means _+ SE. Enzymatic assays of PLA2 and DAG lipase were conducted as described in the Methods section. DAG lipase activity was measured by using 1-stearyl-2-[14C] arachidonyl-sn- glycerol as substrate and sytosolic protein extracted from acini as enzyme source. The reaction was carried out in 50 mM HEPES buffer, pH 7.4, at 37°C for 30 min. PLA2 activity was measured by using L-ot-l-palmitoyl-2-[14C-arachidonyl] phosphatidylcholine (~+C-PAPC) as substrate and cytosolic protien extracted from acini as enzyme source. The reaction was carried out in a 30 mM Tris-HC1 buffer, pH 8.8, at 37°C for 15 min. For all experiments where inhibitors were used, acini were pretreated with the inhibitors for 15 min. For each experiment, each value was measured in triplicate, bFGF, basic fibroblast growth factor; Cch, carbamylcholine. *: p < 0.05 vs their respective control.

492 W. Hou et al.

,'R. 300

e.a . ~

O

l o o _=

o

N 200

0

, 4 : 1 ~

~ ~oo

_=

~ 0

bFGF (50 pM) + + +

genistein (0.5 mM) + vanadate (100 pM) +

FIGURE 3. Effects of inhibitors of protein tyrosine kinase and protein tyrosine phosphatase on bFGF-stimulated intracellular [3H]-arachidonic acid accumulation and extracellular release. Pancreatic acini prelabeled with [3H]-arachidonic acid for 90 min were treated for 15 min with 0.5 mM of genistein and 100 ~M of orthovanadate prior to and during a 20 min stimulation with 50 pM bFGF. Intracellular (A) and extracellular (B) [3H]- arachidonic acid were measured as described in the Methods section. In these experiments, results represent means + SE of 5 different experiments. *: significantly different from control at p < 0.05; **: significantly different from the other groups at p < 0.05.

orthovanadate, a protein tyrosine phosphatase inhibitor, significantly enhanced bFGF-stimulated [3H]-arachidonic acid accumulation and release (Figs. 3A and B).

Possible Implication of PLA2 in bFGF.induced [3H]-Arachidonic Acid Accumulation and Release

The origin of arachidonic acid release in the pancreatic acinar cells may be through the sequential action of PLC and DAG lipase on phosphatidylinositol [14] or from the breakdown of phosphatidylinositol, phosphatidylethanolamine or phosphatidylcholine via PLA2 [15]. In this study, two known inhibitors of PLA2, mepacrine and aristolochic acid [42, 43], were used to evaluate the potential implication of PLA2. Both inhibitors selectively inhibited 5 I-tM carbamylcholine-induced PLA2 activities (Table 1) and were therefore used to investigate the involvement of PLA2 in arachi-

• • • _ _. _ . .~. 2 0 0 c •

"~ 2oo g 200

. . . .

o . . . . . . .

bFGF 5~(pM} + + + + + bFGF 50 (pM) ÷ + + + + m e p a c d n e ( p . M ) - 10 50 100 150 Aristolochic lo 50 loo 200

acid (pM)

FIGURE 4. Effects of the PLAz inhibitors, mepacrine (A, B) and aristolochic acid (C, D) on bFGF-stimulated intraceUular PH]- arachidonic acid accumulation and extracellular release. Pancre- atic acini prelabelled with [3H]-arachidonic acid for 90 min were treated with increasing concentrations of mepacrine (A, B) or aristolochic acid (C, D) for 15 min prior to and during a 20-min stimulation with 50 pM bFGF. Intracellular and extraceUular pH]-arachidonic acid were measured as described in the Meth- ods section. In these experiments, results represent means ± SE of 5 different experiments. *: significantly different from basal at p < 0.05.

donic acid release. Mepacrine, 10, 50, 100 and 150 >M, had no significant effect on basal intracellular [3H]-arachidonic acid accumulation; however, at 100 and 150 IzM, the inhibitor significantly reduced basal extracellular [3H]- arachidonic acid release into the incubation medium (data not shown). As shown in Fig. 4A, increasing concentrations of mepacrine from 10 to 15 btM had no effect on bFGF-induced intracellular [3H]-arachidonic acid accumulation in pancreatic acini. The inhibitor also failed to de- crease bFGF-induced extracellular [3H]-arachidonic acid release at concentrations from 10 to 100 btM but reduced that release to basal values at the highest concentrations of 150 txM (Fig. 4B), an effect comparable to that observed under basal conditions. As indicated in Fig. 4C and 4D, aristolochic acid, another PLA2 inhibitor, also failed to reduce bFGF-stimulated intracellular [3H]-arachidonic acid accumulation in pancreatic acini and its release into the incubation medium. These data also suggest that bFGF does not cause PLA2 activation in rat pancreatic acini and further support our findings with enzymatic assay of PLA>

Potential Implication of PLC, PLD, PKC and DAG Lipase in bFGF-induced [3H].Arachidonic Acid Accumulation and Release

Since we demonstrated that bFGF did not seera to activate PLA2, the bFGF-induced arachidonic acid accumulation and release could then be under the sequential action of PLC-PKC-DAG lipase or PLD-PPH-DAG lipase or both. The implication of DAG lipase was further investigated with the selective DAG lipase inhibitor RHC 80267 [43]. As indicated in Fig. 5A, RHC 80267 significantly inhibited bFGF-induced pH]-arachidonic acid accumulation into


° °

~ . . . . . . . . i o 200 13=

• 0

bFGF 50 pM: bFGF 50pM: " + + + + +

RHCBOL~67 (p .M) : - 10 50 100 t50 inhibitors: SI00~M U10gM WS~IM P200~M

FIGURE 5. Effects of different inhibitors of DAG lipase (A, B) and PLC, PLD and PKC (C, D) on bFGF-stimulated intracellular [3H]-arachidonic acid accumulation and extracellular release. Pancreatic acini prelabelled with [3H]-arachidonic acid for 90 min were treated with increasing concentrations of RHC 80267, a DAG lipase inhibitor (A, B) for 15 min prior to and during a 20 min stimulation with 50 pM bFGF. They were also treated for 15 min (C, D) with 100 nM staurosporine (S), 10 p.M U73122 (U), 50 nM wortmannin (W) or 200 ~M propranolol (P) prior to and during a 20-min stimulation with 50 pM bFGF. Intracellular and extracellular pH]-arachidonic acid were measured as described in the Methods section. In these experiments, results represent means + SE of 5 different experiments. *: significantly different from basal at p < 0.05; **: significantly different from bFGF alone at p < 0.05.

pancreatic acini in a concentration dependent fashion with total inhibition at 150 IzM. Reduction of bFGF-induced [SH]-arachidonic acid release was more sensitive to the inhibitor as 50 txM caused a return to basal value and further inhibition below basal was observed at concentrations of I00 IzM and 150 ~,M (Fig. 5B). These new data clearly serve as further evidence that DAG lipase can be stimulated by bFGF. The results presented in Fig. 5C and 5D give us some indications on the mechanism(s) involved in bFGF- induced [3H]-arachidonic acid accumulation and release from the pancreatic acini. Indeed, the observation that both processes can be equally inhibited by U73122, a selective PLC inhibitor, indicates that bFGF-activation of PLC is a prerequisite for DAG lipase increased activity. The demonstration that staurosporine, a known PKC inhibitor, also reduced significantly bFGF-induced [3H]-arachidonic acid accumulation and release strongly suggests a sequential action of PLC-PKC with DAG as an intermediary mole- cule. The implication of bFGF-induced PLD as a prerequisite for DAG lipase stimulation seems improbable since wortmannin, a known inhibitor of pancreatic PLD, had no effect on bFGF-induced [3H]-arachidonic acid accumulation and release. This was further confirmed by the failure of propranolol, a PPH inhibitor, to inhibit the bFGF- induced [3H]-arachidonic acid accumulation and release.

The selectivity of staurosporine and wortmannin as inhibitors of pancreatic PKC and PLD, respectively, was further evaluated. Indeed, as shown in Fig. 6A and 6B, the implication of PKC in the subsequent activation of DAG

lipase was confirmed from the observation that 4[3-phorbol 12 myristate 13 acetate (PMA), a PKC activator, significantly increased intracellular [3H]-arachidonic acid accumulation and release, an effect totally blocked by the selective inhibitor staurosporine. Furthermore, the demonstration that 50 pM bFGF-activated PLD can be totally inhibited by wortmannin (Fig. 6C) indicates that the failure of wortmannin to block bFGF-induced [3H]-arachidonic acid accumulation and release (Fig. 5C, 5D) was not the re- sult of wortmannin failing to inhibit PLD activity.

D I S C U S S I O N

There are two major pathways which can generate arachidonic acid. One pathway involves PLA2 hydrolysis of membrane phospholipids, whereas the other source of agonist- induced accumulation of arachidonic acid is from DAG, a product of PLC activation [14]. A third pathway may involve PLD which catalyzes the hydrolysis of the terminal diester bond of phosphatidylcholine and possibly of other glycerophosphatides with the formation of phosphatidic acid and choline [46]. Phosphatidic acid can then serve as a substrate for DAG biosynthesis through the action of PPH [40]. In the exocrine pancreas, bFGF has recently been described as an unrecognized pancreatic secretagogue [8] and a mitogenic factor for cultured pancreatic cells [9] as arachidonic acid is emphasized as a newly recognized second mes- senger [13], mediatory role of arachidonic acid in amylase secretion has been suggested [14, 15, 47]. Therefore, it be- comes necessary to demonstrate a bFGF effect on arachidonic acid release and to evaluate clearly the intracellular mechanisms involved for a better understanding of the physiological significance of bFGF. The aim of this study was to investigate the effect of bFGF on arachidonic acid release and furthermore to characterize the intracellular signal transduction pathways that regulate the arachidonic acid release in rat pancreatic acini.

The present study demonstrates for the first time that bFGF activates DAG lipase and stimulates intracellular accumulation and extracellular release of arachidonic acid in rat pancreatic acini. Our data confirm the previous demonstration of arachidonic acid release upon stimulation by bFGF in bovine aortic endothelial cells and Swiss 3T3 cells [26, 27]. The increments in free intracellular arachidonic acid in response to bFGF stimulation were followed almost simultaneously by the extracellular release of the fatty acid. The physiological importance of such arachidonic acid release outside the acinar cells still remains to be investigated. Previous studies have indicated that exogenously applied arachidonic acid can cause inhibition of inositol incorporation into phosphoinositides in rat pancreatic acinar cells [48], elicit Ca 2+ release from intracellular store in pancreatic islets [49], increase the number of available GTP binding proteins and can consequently lead to cell activation [50], or activate PKC in synergy with DAG [5 I]. Finally, arachidonic acid added to pancreatic acini also caused a concentration-dependent increase in amylase release, an effect reproduced by PLA2 [15].

494 W. Hou et al.

.~ 200 I= O

-~ 0 "0

~ 200

- ~ 100

~ 0

PMA smurosporine

B

200

,,,,,

8 loo

r ~

0

bFGF 50 pM

wortmannin

- + +

- +

of the PKC inhibitor, staurosporine, FIGURE 6. Effect on PMA.stimulated pH]-arachidonic acid accumulation and release and of wortmannin, a PLD inhibitor, on bFGF-stimulated phosphatidylethanol production. Pancreatic acini prelabelled with pH]-arachidonic acid for 90 rain were treated with 100 nM staurosporine (A, B) for 15 min prior to and during a 20 min stimulation with 100 nM PMA. Intracellular and extracellular [3H]-arachidonic acid were measured as described in the Meth- ods section. For PEt production (C), acini prelabelled with [3H]- myristic acid for 60 min were treated with 50 nM wortmannin for 15 min prior to and during a 20 min-stimulation with 50 pM bFGF. PEt production was measured as described in the Methods section. In these experiments, results represent means _+ SE of 5 different experiments. *: significantly different from basal at p < 0.05; **: significantly different from PMA or bFGF at p < 0.05.

It is recognized that initiation of bFGF action depends upon its binding to a specific cell surface tyrosine kinase receptor [6, 7]. Accordingly, we investigated whether tyrosine phosphorylation was involved in bFGF-induced pH]-ara-

chidonic acid accumulation and release. The observations that the tyrosine kinase inhibitor genistein totally inhibited and that the protein tyrosine phosphatase inhibitor orthovanadate significantly enhanced bFGF-induced pH]-arachidonic acid accumulation and release strongly suggest that an initial activation of the bFGF intrinsic receptor tyrosine kinase participates in the bFGF action and precedes the activation of other phospholipase(s) involved in arachidonic acid release.

Fafeur et al. reported that bFGF stimulated the rapid release of arachidonic acid in endothelial cells and believed this process was mediated by PLA2 rather than by the sequential action of PLC and DAG lipase, since bFGF did not stimulate the production of inositol phosphates nor the in- duction of calcium mobilization [26]. Recently, Virdee et al. also demonstrated that recombinant bFGF caused the release of pH]-arachidonic acid from metabolically labelled Swiss 3T3 fibroblasts in a concentration- and time-dependent manner, and suggested that this bFGF-induced [3H]-arachidonic acid release was mediated via a PLA2 pathway as mepacrine, a putative PLA2 inhibitor, suppressed this bFGF effect [27]. In contrast, other studies have also shown that bFGF stimulated accumulation of inositol phosphates [52] and DAG production [23, 53], suggesting a bFGF-induced activation of PLC rather than PLA2. Our present findings that bFGF-induced intracellular accumulation of arachidonic acid and its extracellular release were not inhibited by two PLA2 inhibitors, mepacrine and aristolochic acid, and that enzymatic PLA2 assay did not demonstrate any activation of the enzyme by bFGF, indicate that activation of a PLA2 does not occur in response to bFGF. The reason for this discrepancy between our data and others is not yet clear and may indicate that the mechanisms involved in the control of arachidonic acid release are cell- or tissue-specific.

If arachidonic acid release does not derive from PLA2 hydrolysis of phospholipids, the other source could be from DAG, a product of PLC or PLD-PPH activation. Coupled PLD-PPH pathway is able to generate DAG which, hydrolyzed by DAG lipase, liberates arachidonic acid [19]. In our study, propranolol, a known inhibitor of PPH [40], did not suppress bFGF-induced generation of arachidonic acid, ruling out a role for PLD in arachidonic acid release. Fur- thermore, the failure of wortmannin, a PLD inhibitor [38], to inhibit bFGF-induced arachidonic acid release while in- hibiting PEt production (Fig. 6C), does not support a role for PLD in arachidonic acid metabolism either.

With the absence of PLA2 and the coupled PLD-PPH involvement in arachidonic acid release, the remaining pathway seems to be a PLC-catalyzed breakdown of phospholipids to form DAG, followed by deacylation of DAG by DAG lipase. Such a pathway has been previously demonstrated by Bell [54] and Konrad [45] in human platelets and isolated pancreatic islets, respectively. In our study, bFGF at 50 pM significantly activated DAG lipase activity with enzymatic assay and this activation was totally prevented by RHC 80267. RHC 80267 also successfully inhibited bFGF- induced [3H]-arachidonic acid intracellular accumulation and extracellular release. Thus, a DAG lipase activation by


bFGF is believed to be definitively involved in arachidonic acid release. The next question arising from the above observation regards the mechanism by which bFGF causes the DAG lipase activation and the source of DAG production. Therefore, to ascertain that PLC-DAG lipase activation is the existing pathway, we used the selective PLC inhibitor U73122 capable of total inhibition of CCK-OP-stimulated phosphatidylinositol bisphosphate hydrolysis in pancreatic acini [37]. The total inhibition of bFGF-induced [3H]-arachidonic acid accumulation and release by the PLC inhibitor supports what seems to be a unique pathway in the pancreas for the release of arachidonic acid bFGF. To our knowledge, this is novel for exocrine pancreatic tissue.

A previous study has indicated that intracellular arachidonic acid release can be significantly increased in response to PKC activation by the phorbol ester PMA [15]. Our data certainly confirmed that initial observation which was further emphasized by the finding that the PKC inhibitor staurosporine totally inhibited PMA-stimulated intracellular arachidonic acid accumulation and its extracellular release. Furthermore, the important observation that this PKC inhibitor also inhibited bFGF-induced arachidonic acid accumulation and release suggests that activation of DAG lipase is in line with a previous activation of PLC and PKC; PKC activation is thus an important and obligatory partner in this whole process of DAG lipase activation and arachidonic acid release.

In conclusion, we demonstrated that bFGF activated DAG lipase activity and stimulated intracellular accumulation and extracellular release of arachidonic acid, and eluci- dated the underlying intracellular signalling pathways for the first time in rat pancreatic acini. Our data do not support any implication of PLA2 nor PLD in stimulation of intracellular events leading to arachidonic acid release in rat pancreatic acini. However, our findings indicate that bFGF probably stimulates a unique pathway which originates from activation of the bFGF receptor tyrosine kinase, leading to PLC activation, generation of DAG, which in turn activates PKC which is responsible for DAG lipase increased activity, leading to arachidonic acid release.

The authors thank Madeleine Martel and Claire L. Palardy for secre- tarial assistance. This research was supported by the Medical Research Council of Canada and le Fonds pour la Formation de chercheurs et d'Aide a la Recherche de la Province de Quebec, Grants MT11032 and ER- 1092.

References 1. Gospodarowicz D. (1974) Nature (London) 249, 123-127. 2. Armelin H. A. (1973) Proc. Natl. Acazt. Sci. U.S.A. 70,

2702-2706. ~. Burgess W. H. and Macaig T. (1989) Annu. Rev. Biochem. 58,

575-606. 4. Gospodarowicz D. (1990)J. Clin. Orthoped. Related Res. 58,

231-242. 5. Jaye M., Lyall R. M., Mudd R., Schlessinger J. and Sarver, N.

(1988) EMBO J. 7, 963-969. 6. Coughlin S. R., Barr P. J., Cousens L. S., Fretto L. J. and Wil-

liams L. T. (1988) J. Biol. Chem. 263,988-993.

7. Kuo M. D., Huang S. S. and Huang J. S. (1990) J. Biol. Chem. 265, 16455-16463.

8. Chandraseker B. and Korc M. (1991) Biochem. Biophys. Res. Commun. 177, 166-170.

9. Hoshi H. and Logsdon C. D. (1993) Biochem. Biohpys. Res. Commun. 196, 1202-1207.

10. Burgess W. H., Dionne C. A., Kaplow J., Mudd R., Friesel R., Zilberstein A., Schlessinger J. and Jaye M. (1990) Mol. Cell. Biol. 10, 4770--4777.

11. Berridge M. J. (1984) Biochern. J. 220, 345-360. 12. Streb H., Irvine R. F., Berridge M. J. and Schulz I. (1983) Na.

ture 303, 67-68. 13. Naor Z. (1991) Mol. Cell. Endocr. 80, C181-C186. 14. DixonJ. F. and Hokin L. E. (1984) J. Biol. Chem. 259, 14418--

14425. 15. Pandol S. J., Hsu Y., Kondratenko N. F., Schoeffield-Payne

M. S. and Steinbach J. H. (1991) Am . J. Physiol. 260 (Gas- trointest. Liver Physiol. 23), G423-G433.

16. Irvin R. F. (1982) Biochem. J. 240, 3-16. 17. Van den Bosch H. (1980) Biochim. Biophys. Acta 604, 191-

246. 18. Wang X. D., King J. G. and Smallridge R. C. (1994) Biochim.

Biophys. Acta 1223, 101-106. 19. Exton J. H. (1990)J. Biol. Chem. 265, 1-4. 20. Martin T. W. and Wysolmerski R. B. (1987) J. Biol. Chem.

262, 13086-13092. 21. Huang S. S. and Huang J. S. (1986)J. Biol. Chem. 261, 9568-

9571. 22. Friesel R., Burgess W. H. and Macaig T. (1989) Mol. Cell.

Biol. 9, 1857-1865. 23. Tsuda T., Kaibuchi K., Kawahara Y., Fukuzaki H. and Takai

Y. (1985) FEBS Lett. 191,205-210. 24. Maynaldo I., L'Allemain G., Chambard J. C., Moenner M.,

Barritault D. and Pouyssegur J. (1986) J. Biol. Chem. 261, 16916-16922.

25. Nanberg E., Morris C., Higgins T., Var F. and Rozengurt, E. (1990) J. Cell. Physiol. 143,232-242.

26. Fafeur V., Jiang Z. P. and Bolen P. (1991) J. Cell. Physiol. 149, 277-283.

27. Virdee K., Brown B. L. and Dobson P. R. M. (1994) Biochim. Biophys, Acta 1220, 171-180.

28. Rydzewska G. and Morisset J. (1995) Pancreas 10, 59-65. 29. Peikin S. R., Rothman A. J., Batzri S. and Gardner J. D.

(1978) Am. J. Physiol. 235 (Endocrinol. Metab. Gastrointest. Physiol. 4), E743-E749.

30. Cybulski A.V. (1991) Am. J. Physiol. 261, F427-F436. 31. Bradford M. A. (1976) Anal. Biochem. 72, 248-254. 32. Prescott S. M. and Majerus P. W. (1982)J. Biol. Chem. 258,

764-769. 33. Jelsema C. L. (1987) J. Biol. Chem. 262, 163-168. 34. Sato T., Hashizume T. and Fujii T. (1992) J. Biochem. 112,

756-76l. 35. Rydzewska G., Rossignol B. and Morisset J. (1993) Am. J.

Physiol. 265 (Gastrointest. Liver Physiol. 28), G725-G734. 36. Turk J., Colca J. R. and McDaniel M. L. (1985) Biochim. Bio-

phys. Acta 834, 23-36. 37. Yule D. I. and Williams J. A. (1992) J. Biol. Chem. 267,

13830-13835. 38. Bonser R. W., Thompson N. T., Randall R. W., Tateson J. E.,

Spacey G. D., Hodson H. F. and Garland L. G. (1991) Br. J. Pharmacol. 103, 1237-1241.

39. Sozzani S., Agwu D. E., McCall C. E., O'Flaherty J. T., Schmitt J. D., Kent J. D. and McPhail L. C. (1992) J. Biol. Chem. 267(28), 20482-20488.

40. Witter B. and Kanfer J. N. (1985) J. Neurochem. 44, 155-162. 41. Sutherland C. A. and Amin D. (1982) J. Biol. Chem. 257,

14006-14010. 42. Wallach D. P. and Brown V. J. R. (1981)Biochem. Pharmacol.

30, 1315-1324.

496 W. Hou et al.

43. Spangel B. L., Jarvis W. D., Judd A. M. and MacLeod R. M. (1991) Endocrinology 129, 2886-2894.

44. Aklyama T., Ishida J., Nakagawa S., Ogawa H., Watanabe S., Itoh N., Shibuya M. and Fukami Y. (1987) J. Biol. Chem. 262, 5592-5595.

45. Konrad R. J., Major C. D. and Wolf B. A. (1994) Biochemistry 33, 13284-13294.

46. Cook S. J. and Wakelam M. J. O. (1991) Cell Signal. 3,273- 282.

47. Tsunoda Y. and Owyang C. (1994) Biochim. Biophys. Res. Commun. 203, 1716-1724.

48. Chaudhry A., Laychock S. G. and Rubin R. P. (1987) J. Biol. Chem. 262, 17426-17431.

49. Wolf B. A., Turk J., Sherman W. R. and McDaniel M. L. (1986) J. Biol. Chem. 261, 3501-3511.

50. Abramson S. B., Leszczynska-Piziak J. and Weissmann G. (1991) J. lmmunol. 147, 231-236.

51. Shinomura T., Asaoka Y., Oka M., Yoshida K. and Nishizuka, Y. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 5149-5153.

52. Brown K. D., Blakeley D. M. and Brigstock D. R. (1989) FEBS Lett. 247,227-231.

53. Kaibuchi K., Tsuda, T., Kaibuchi A., Tanimoto T., Yamashita T. and Takai Y. (1986)J. Biol. Chem. 261, 1187-1192.

54. Bell R. L., Kennerly D. A., Stanford W. and Majerus P. W. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 3238-3241.

Documents

Basic fibroblast growth factor-stimulated arachidonic acid release in rat pancreatic acini: Sequential action of tyrosine kinase, phospholipase C, protein kinase C and diacylglycerol