Mobilization of arachidonic acid in thrombin-stimulated human platelets

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

V. G . MAHADEVAPPAAND FRANK SICILIA Department of Nutritional Sciences, University of Guelph, Guelph, Ont., Canada NIG 2 WI Received May 16, 1989 MAHADEVAPPA, V. G., and SICILIA,F. 1990. Mobilization of arachidonic acid in thrombin-stimulated human platelets. Biochem. Cell Biol. 68: 520-527. In the present work we investigated the effect of serine esterase inhibitors such as 2-nitro-4-carboxyphenyl N,Ndiphenylcarbarnate (NCDC) and phenylmethylsulfonyl fluoride (PMSF), as well as the effect of mepacrine on thrombininduced mobilization of arachidonic acid (AA) in human platelets. The inhibitor NCDC (0.6 mM) completely abolished the thrombin-induced activation of phospholipase C, phospholipase A,, and transacylase enzymes, whereas the pretreatment of platelets with PMSF (2 mM) resulted in a highly selective inhibition of phospholipase A, and transacylase activities, with no marked effect on thrombin-induced activation of phospholipase C. The thrombin-induced release of [ 3 ~from ] phosphatidylcholine ~ ~ and phosphatidylinositol was reduced by 90 and 56%, respectively, in the presence of PMSF. This inhibitor also caused a parallel inhibition in the accumulation of [,HIM (85%) with little effect on thrombin-induced formation of [3~]phosphatidicacid (5%), whereas mepacrine (0.4 mM) caused a selective inhibition of phospholipase A, and transacylase activities with concomitant stimulation of [3~]phosphatidic acid formation in intact human platelets. These results demonstrate that NCDC and PMSF (serine esterase inhibitors) do not affect agonist-induced activation of phospholipases that mobilize arachidonic acid through a common site. Our results further demonstrate that the inhibition of [,HIM release observed in the presence of NCDC, PMSF, and mepacrine is primarily due to their direct effects on enzyme activities, rather than due to their indirect effects through formation of complexes between inhibitors and membrane phospholipids. Based upon these results, we also conclude that the combined hydrolysis of phosphatidylcholine and phosphatidylinositol by phospholipase A, serves as a major source for eicosanoid biosynthesis in thrombin-stimulated human platelets. Key words: deacylation, phospholipids, thrombin, platelets, phospholipase A,. MAHADEVAPPA, V. G., et SICILIA,F. 1990. Mobilization of arachidonic acid in thrombin-stimulated human platelets. Biochem. Cell Biol. 68 : 520-527. Dans ce travail, nous examinons l'effet d'inhibiteurs de la serine esttrase, tels le 2-nitro-4-carboxyphenyl N,Ndiphenylcarbamate (NCDC) et le phCnylm6thylsulfonyl fluorure (PMSF), et I'effet de la mtpacrine sur la mobilisation, induite par la thrombine, de I'acide arachidonique (AA) dans les plaquettes humaines. Le NCDC (0,6 mM) abolit complhtement l'activation, induite par la thrombine, de la phospholipase C, de la phospholipase A, et de la transacylase, tandis que le prktraitement des plaquettes avec le PMSF (2 mM) produit une inhibition fortement stlective de la phospholipase A, et de la transacylase, sans exercer d'effet marque sur I'activation, induite par la thrombine, de la phospholipase C. En presence du PMSF, la libbation, induite par la thrombine, de I'AA-~Hde la phosphatidylcholine et du phosphatidylinositolest res ectivement rtduite de 90 et de 56%. Cet inhibiteur entraine Cgalement une inhibition parallele de l'accumulation de I'AA- r3H (85%) sans exercer d'effet important sur la formation, induite par la thrombine, de l'acide [3~]phosphatidique(5%) alors que la mbpacrine (0,4 mM) inhibe de fagon selective la phospholipase A, et la transacylase avec stimulation concomitante de la formation de l'acide [3~]phosphatidiquedans les plaquettes humaines intactes. Ces resultats demontrent que le NCDC et le PMSF (inhibiteurs de la skrine esterase) n'affectent pas l'activation, induite par un agoniste, des phospholipases qui mobilisent I'AA a travers un site cornmun. Nos rksultats dtmontrent egalement que l'inhibition de la liberation de I ' M - 3 ~ observte en prksence du NCDC, du PMSF et de la mkpacrine est d'abord due a l'effet direct de ces inhibiteurs sur les activitks enzymatiques plut6t qu'8 leurs effets indirects par formation de complexes entre les inhibiteurs et les phospholipides membranaires. A la lumikre de ces rkultats, nous concluons Cgalement que l'hydrolyse combink de la phosphatidylcholine et du phosphatidylinositol par la phospholipase A, sert de source majeure A la synthtse des eicosanoldes dans les plaquettes humaines stirnulees par la thrombine. Mots clks : ddeacylation, phospholipides, thrombine, plaquettes, phospholipase A,. [Traduit par la revue]

Introduction The initial activation of phospholipases serves as the rate-limiting step in the biosynthesis of various eicosanoids (Irvine 1982). The two major metabolites formed from AA via the action of cyclooxygenase and lipoxygenase(s) in stimulated platelets include TxA2 and 12-HETE. These metabolites appear to play predominant roles in a variety of platelet responses including shape change, secretion, and aggregation (Lagarde 1988; Lagarde et al. 1989; Siess 1989). Agonists such as thrombin, collagen, platelet-activating factor, and ADP have been frequently employed to study a wide range of platelet responses (Siess 1989). Of these, thrombin appears to be the most potent physiological ABBREVIATIONS: NCDC, 2-nitro-4-carboxyphenyl N,N-diphenylcarbamate; PMSF, phenylmethylsulfonyl fluoride; AA, arachidonic acid; TxA,, thromboxane A,; 12-HETE, 12-hydroxyeicosatetraenoicacid; PIP,, phosphatidyl4.5-bisphosphate; PLC, phospholipase C; DG, diacylglycerol; IP,, inositol 1,4,5-trisphosphate; PKC, protein kinase C; PLA,, phospholipase A,; MG, monoacylglycerol; FFA, free fatty acid; ACD, acid-citrate-dextrose; PRP, platelet-rich plasma; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PI, phosphatidylinositol; PS, phosphatidylserine; TLC, thin-layer chromatography; PA, phosphatidic acid; dpm, disintegrations per minute (60 dpm = 1 Bq). his paper is dedicated to Dr. Morris Kates in honour of his valuable contributions to biochemistry in Canada.

Printed in Canada / Imprim6 au Canada

MAHADEVAPPA AND SICILIA

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

agonist. Initial interaction of thrombin with its specific receptor present on the platelet surface membrane triggers the activation of PIP2-specific PLC (EC 3.1.4.10) (Mauco et al. 1979; Rittenhouse-Simmons 1979). This initial activation of PLC leads to the formation of several bioactive products including DG, IP3, and AA metabolites, which appear to possess a wide range of intracellular messenger functions (Nishizuka 1984a, 1984b; Berridge 1987; Majerus et al. 1988; Lagarde 1988; Siess 1989). For example, the IP3 formed from PIP2 appears to be responsible for the mobilization of c a 2 + from intracellular stores into the cytosol (Berridge 1987). The increase in cytoplasmic c a 2 + and the DG formed during PLC-mediated hydrolysis of PIP2 appear to be coupled to the activation of a specific phospholipid-, DG-, and ca2+-dependent enzyme, PKC, which phosphorylates a 47-kDa protein in platelets (Nishizuka 1984a, 1984b; Siess 1989). More recent evidence suggests that the agonist-induced activation of PLC is regulated by specific GTP-binding proteins (Casey and Gillman 1988). Furthermore, it has been suggested that the mobilization of c a 2 + from intracellular stores and the activation of PKC that results in response to agonists play a critical role in the activation of PLA2, the key enzyme in the mobilization of AA in stimulated platelets (Lapetina 1986; Berridge 1987; Fuse and Tai 1988; Feinstein and Halenda 1988; Siess 1989). However, recent studies have also led to the speculation that PLA2 may itself be regulated by one or more GTP-binding proteins that affect the activity of PLA2 directly and that affect various ion channels (Feinstein and Halenda 1988). The source and pathways that are responsible for the mobilization of AA in stimulated platelets have been extensively studied in our laboratory and by others (Mahadevappa and Holub 1986, 1987; Broekman 1986; Mahadevappa 1987). Although the majority of the AA released in thrombin-stimulated platelets comes from the hydrolysis of phospholipids through a route involving PLA2, a small but metabolically significant amount of AA is released through the DG lipase pathway. The latter route involves a sequential removal of stearic and arachidonic acids from DG by the action of DG and MG lipases, respectively (Bell et al. 1979; Chau and Tai 1981; Prescott and Majerus 1983; Mahadevappa and Holub 1986). In the past, several compounds including serine esterase inhibitors have been used as specific PLC, PLA2, and DG lipase inhibitors to evaluate the importance of the pathways responsible for the mobilization of AA from various phospholipids in stimulated platelets (Vanderhoek and Feinstein 1979; Walenga et al. 1980, 1981; Holmsen et al. 1984; Smith et al. 1985; Mahadevappa and Holub 1986,1987; Mahadevappa 1987). These compounds have served as useful tools in studies involving platelet AA metabolism, despite their potential nonspecific effects. In the present work we studied the effect of two serine esterase inhibitors (NCDC and PMSF) and quinacrine hydrochloride (mepacrine) on thrombin-induced mobilization of AA from platelet phospholipids. The inhibition of AA release resulting in the presence of these compounds appears t o be due to their direct effects on phospholipases that release AA from platelet phospholipids rather than due to a depletion of substrates, which had been thought to result from complexes formed between phospholipid substrates

521

and inhibitors. Our results further indicate that the majority of the labelled AA released into the FFA fraction following stimulation with thrombin was derived from PC. Materials and methods [5,6,8,9,11,12-15-3~(~)]~rachidonic acid (specific activity, 94.5 Ci/mmol; 1 Ci = 37 GBq) was purchased from New England Nuclear (Boston, MA). Thrombin, neutral lipid and phospholipid reference standards, 2' ,7'-dichlorofluorescein,mepacrine, NCDC, and PMSF were all obtained from Sigma Chemical Co. (St. Louis, MO). Precoated thin-layer silica gel H plates (H-60) were from E. Merck (Darmstadt, Federal Republic of Germany). 3-Amino-l(3-tyrifluoromethylphenyl)-2-pyrazoline hydrochloride (BW755C), a dual selective inhibitor of cyclooxygenase and lipoxygenase(s), was donated by the Wellcome Research Laboratories (Kent, U.K.). All solvents and chemicals employed were of analytical grade. Isolation, prelabelling, and preparation of platelet suspensions Blood was drawn from the antecubital veins of healthy human volunteers into vacutainer tubes containing ACD. Citrated blood was centrifuged at 200 x g for 7 rnin at room temperature and the resulting PRP was transferred to polypropylene tubes. PRP was then incubated with tritiated AA (2.5 pCi/20 mL PRP) in a shaking water bath at 37°C for 2 h. Incubation of platelets with radioactive AA for 90-120 rnin resulted in a significant incorporation of this fatty acid into the sn-2 position of all major phospholipids including PC, PE, PI, and PS. However, the relative distribution of labelled AA was not in proportion to the mass of this fatty acid and in different phospholipids. Labelled platelets were isolated from the PRP by centrifugation at 1600 x g for 13 min at room temperature. The resulting pellet was washed with Tris-EDTA buffer (0.02 M Tris, 0.15 M NaCl, and 0.002 M Na2EDTA, pH 7.4) and resuspended in ca2+-freeTris-glucose buffer (0.02 M Tris, 0.15 M NaCl, and 0.005 M glucose). Platelets were counted in a Coulter counter and the density of platelets was adjusted to 4 x 10' platelets/mL in the final platelet suspension. Incubation of platelet suspensions and extraction of lipids One-millilitre aliquots of platelet suspensions (4 x 10' platelets/mL) were routinely stimulated with thrombin (2 U/mL) for 3 min in a shaking water bath maintained at 37OC in the presence and absence of various inhibitors. In the presence of BW755C (80 pM), an extensive accumulation of free AA was observed in response to the action of thrombin. Incubation of platelets with this compound before stimulation with thrombin results in about 95-98% inhibition of the cyclooxygenase and lipoxygenase(s), with no effect on agonist-induced activation of phospholipases that release AA from phospholipids. In these experiments platelet suspensions were preincubated with BW755C (vehicle was saline) for 1 rnin at 37OC and then stimulated with thrombin for 3 min, and reactions were terminated by the addition of 3.75 mL chloroform-methanol (1:2, v/v). This experimental design allowed the comparison of the effect of NCDC, PMSF, and mepacrine on the release of AA from different phospholipids, as well as on the accumulation of free AA in the FFA fraction. In these experiments platelet suspensions were incubated with NCDC, PMSF, or mepacrine in the presence of BW755C for 2 rnin and then stimulated with thrombin for 3 min, and reactions were terminated by the addition of chloroform-methanol (1:2, v/v). Lipids were extracted and purified essentially by the method of Bligh and Dyer (1959). The lower chloroform phases were dried under nitrogen and resuspended in exactly 1 mL of chloroform-methanol (2: 1, v/v). One half of the lipid extract was used for the analysis of phospholipids, free AA, and DG by three-dimensional TLC and the other half was used for the analysis of PI and PS by onedimensional TLC.

BIOCHEM. CELL BIOL. VOL. 68, 1990

TABLE1. Effect of NCDC (0.6 mM) on thrombin-induced AA mobilization from human platelet phospholipids Net loss of ['HIM (dprn) Incubation conditions

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

Platelets + BW755C + thrombin Platelets + BW755C + NCDC + thrombin

PC

PE

PIn

PS

7424 k 324(0%)

0

3297 + 308(0%)

1658 k 602(0%)

NOTE:Labelled platelets were stimulated with thrombin (2 U) for 3 min in the presence and absence of NCDC. Lipids were extracted, purified, and analyzed as described in Materials and methods. Results represent the net loss of radioactivity following stimulation with thrombin. Numbers in parentheses indicate percent inhibition of ['HIAA loss from different phospholipids in the presence of NCDC. Results represent means + SE (n = 3). "The net loss of [%]AA was determined by subtracting the net gain in ['HIDG and ['HIPA from the total loss from ['HIPI.

TABLE2. Effect of NCDC (0.6 mM) on thrombin-induced gain in tritiated radioactivity in the PE, PA, and FFA fractions Net gain in radioactivity (dpm) Incubation conditions Platelets + BW755C + thrombin Platelets + BW755C + NCDC + thrombin

PE

PA

AA

1100k 304(0%)

1724k 20(0%)

9429 k 35 1(0%)

132k 20(92%)

O(100Vo)

0(100%)

NOTE:Numbers in parentheses indicate percent inhibition in the presence of NCDC (0.6 mM). For experimental details, see legend to Table I. Results represent means + SE ( n = 3).

Chromatography of phospholipids and neutral lipids The separation of both phospholipid and neutral lipid classes was achieved on precoated silica gel H plates by three-dimensional TLC with the exception of PI and PS. Platelet lipid extracts, dissolved in small volumes of chloroform-methanol (2: 1, v/v), were applied as a single spot on the plate and the plates were then developed in chloroform - methanol - concd ammonium hydroxide (65:35:5.5, by volume) up to 1 cm from the top. After this initial development, the plates were removed, dried under nitrogen for 30 min, and developed perpendicular to the first direction in chloroform - methanol - acetone - acetic acid - water (30: 10:30:5:2.5, by volume). The individual phospholipid and FFA spots were detected by spraying only two thirds of the plate with 2' ,7 ' -dichlorofluorescein in the direction employed for the second development. Bands corresponding to PC. PE. PI plus PS, PA, and free AA were scraped into scintillation vials containing 1.5 mL water and 13.5 mL Aquasol, and the radioactivity was determined in a liquid scintillation counter (Beckman LS 7800). The unsprayed portion of the plate, in which most of the neutral lipids including DG appeared, was developed in the opposite direction to the one used in the second development (from the top) with heptane isopropyl ether - acetic acid (60:40:3, by volume), and the band corresponding to DG was scraped and counted for radioactivity. As PI and PS comigrated in the above solvent systems, these phospholipids were separated from the other half of the lipid extracts by one-dimensional TLC using chloroform - methanol acetic acid - water (50:37.5:3.5:2, by volume) as described previously from our laboratory (Mahadevappa 1987). Following this separation, the PI and PS bands were scraped and counted for radioactivity.

Results The use of a dual inhibitor of cyclooxygenase and lipoxygenase, BW755C, serves as a useful tool in the evaluation of AA mobilization from membrane phospholipids in intact cells, as it does not affect the agonist-induced activation of phospholipases (Smith et al. 1985; Mahadevappa and Holub 1986). This approach also eliminates the potential feedback effects that the derived eicosanoids may have on the activation of various phospholipases and related phospholipid changes. In the present study the above inhibitor was used in combination with serine esterase inhibitors (NCDC and PMSF) and mepacrine, to determine their effects on the mobilization of AA induced by thrombin. The effect of NCDC on thrombin-induced release of AA from labelled platelets was investigated and the results obtained are summarized in Table 1. We have consistently observed that treatment of labelled platelets with BW755C (1100 pM) alone does not result in any significant shift in the radioactivity between different platelet phospholipids (data not shown). However, exposure of platelets to thrombin (2 U) in the presence of BW755C (80 pM) resulted in a significant loss of radioactivity from labelled PC (7424 dpm), PI (3297 dpm), and PS (1658 dpm) (Table 1). Under our experimental conditions, the loss of labelled AA from PC observed in response to the action of thrombin (2 U) accounted for about 20% of the total radioactivity present in unstimulated labelled PC, while the net loss of

MAHADEVAPPA AND SICILIA

TABLE 3. Effect of PMSF (2 mM) on AA mobilization from thrombin-stimulated human platelet phospholipids Net loss of [ 3 ~ ] A A (dpm) Incubation conditions

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

Platelets + BW755C + thrombin Platelets + BW755C + PMSF + thrombin

PC

*

23 046 2460(0%)

2 360 + 1369(90%)

PE

PIa

PS

0

9328 657(0%)

1700 300(0%)

0

4276 + 235(56%)

*

*

0(100%)

NOTE:Numbers in parentheses indicate percent inhibition of I~H]AAloss in the presence of PMSF (2 mM). For experimental conditions, see legend to Table 1. Results represent means + SE ( n = 3). T h e net loss of [ 3 ~ was ] determined ~ ~ by subtracting the net gain in I~H]DGand [ 3 ~ from ] ~the~ total loss from [ 3 ~ ] ~ ~ .

TABLE 4. Effect of PMSF (2 mM) on thrombin-inducedgain in tritiated radioactivity in the PE, PA, and FFA fractions Net gain in radioactivity (dpm) Incubation conditions Platelets + BW755C + thrombin Platelets + BW755C + PMSF + thrombin

PE

PA

AA

4436 k 178(0%)

5656 + 112(0%)

29 309 + 702(0%)

*

1398 358(68%)

5371 k 173(5%)

4 286 + 104(85%)

NOTE:Numbers in parentheses indicate percent inhibition of thrombin-induced gain in radioactivity in the presence of PMSF (2 mM). For experimental conditions, see legend to Table 2. Results represent means + SE ( n = 3).

labelled AA from PI (total loss of radioactivity from PI minus total gain of radioactivity in DG and PA) represented 60% of the total loss from this phospholipid and 31% of the combined loss from PC and PI. We found no loss of radioactivity from PE following stimulation of platelets with thrombin, although the total PE fraction had contained a substantial amount of radioactivity. In the presence of NCDC (0.6 mM), the thrombin-induced release of labelled AA from PC and PI, but not from PS, was completely abolished (Table 1). Several lipid fractions including PE, DG, PA, and FFA fractions gained radioactivity following stimulation of platelets with thrombin (Table 2). The observed thrombin-induced increase in these lipid fractions was also abolished in the presence of NCDC (Table 2). It was found that 88% of the radioactivity lost from PC and PI in response to the action of thrombin was recovered in the FFA fraction in the presence of BW755C (Table 2). In addition, we consistently found that 30-40% of the radioactivity lost from PI appeared in the newly formed [ 3 ~ ] ~ ~ . On the other hand, PMSF, which is also a serine esterase inhibitor, selectively inhibited the mobilization of AA induced by thrombin without affecting the thrombininduced activation of PLC. The results obtained on the effect of PMSF are summarized in Tables 3 and 4. In the presence of PMSF (2 mM), the thrombin-induced loss of radioactivity from PC, PI, and PS was reduced by 90, 56, and 100%, respectively. Furthermore, the gain in radio-

activity observed following stimulation of platelets with thrombin in the PE, PA and FFA fractions was reduced by only 68,5, and 85%, respectively, in the presence of PMSF (2 mM). A slight increase in [ 3 ~was] observed ~ ~ in the presence of PMSF (data not shown). Unlike NCDC, this serine esterase inhibitor had no marked inhibitory effect on PLC-induced formation of [ 3 ~ that ] resulted ~ ~ in response to the action of thrombin, although the net accumulation of [ 3 ~ ] A A in the FFA fraction was reduced by 85% in the presence of PMSF (2 mM). However, the net loss of [ 3 ~from ] PI~ was ~ reduced only by 56% in the presence of PMSF and this decrease in radioactivity was reflected in the FFA fraction (Table 4). It appears that the loss of [ 3 ~ ] A A from PC and the gain in PE observed in response to the action of thrombin are more sensitive to inhibition by PMSF. Since much of the radioactivity gained in platelet PE following thrombin stimulation is derived from PC via the CoA-independent transacylase reaction, the inhibition of [ 3 ~ ] A A loss from PC and the inhibition of [3~]AA gain in PE by PMSF account for the total inhibition of [ 3 ~ ] A A loss from PC (Tables 3 and 4). These differences in the presence and absence of PMSF indicate that 30-3 1% of the AA released from platelet phospholipids in response to the action of thrombin is derived from PI, which is similar to that observed in experiments with NCDC. We also examined the effect of mepacrine (0.4 mM) on thrombin-induced mobilization of AA from individual phospholipids and the results obtained are summarized in

BIOCHEM. CELL BIOL. VOL. 68, 1990

TABLE 5. Effect of mepacrine (0.4 mM) on thrombin-induced AA mobilization from human platelet phospholipids Net loss of I~H]AA(dpm) Incubation conditions

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

Platelets + BW755C + thrombin Platelets + BW755C mepacrine + thrombin

PC

PE

PI'

PS

30 132 + 1154(0%)

0

15 060 + 107(0%)

2233 + 123(0%)

2 573 k 1234(92%)

0

1 483 + 858(90%) 2316 + 308(0%)

NOTE:Numbers in parentheses indicate percent inhibition of thrombin-induced loss of ['H]AA in the presence of mepacrine (0.4 mM). For experimental conditions, see legend to Table 1. Results represent means ? SE ( n = 3). 'The net loss of ['HIAA was determined by subtracting the net gain in ['HIDG and [ 3 ~ from ] ~the ~ total loss from ['HIPI.

TABLE6. Effect of mepacrine (0.4 mM) on thrombin-induced gain in tritiated radioactivity in the PE, PA, and FFA fractions Net gain in radioactivity (dpm) Incubation conditions Platelets + BW755C + thrombin Platelets + BW755C + mepacrine + thrombin

PE

4288 + 282(0%)

0

PA

AA

*

7 480 & 107(0%)

31 23 1 400(0%)

12 527 + 927(0%)

1 1 15 k 22(96%)

NOTE:Numbers in parentheses indicate percent inhibition in thrombin-induced gain in radioactivity in the presence of mepacrine (0.4 mM). For experimental details, see legend to Table 2. Results represent means f SE ( n = 3).

Tables 5 and 6). Mepacrine (0.4 mM), a frequently used antimalarial drug, suppressed the thrombin-induced release of [ 3 ~ ] A A from PC and PI by 92 and 90%, respectively. This compound had no effect on [ 3 ~ ] A loss A from PS. In none of our experiments did stimulation of platelets with thrombin result in any marked loss of radioactivity from labelled PE. In contrast, a substantial gain in radioactivity in this phospholipid fraction was apparent in platelets stimulated with thrombin. The increases observed in the radioactivity of PE and FFA fractions following stimulation with thrombin were reduced by 100 and 96070, respectively, in the presence of mepacrine. In contrast, a significant enhancement in the formation of thrombin-induced [ 3 ~occurred ] ~ in~ the presence of mepacrine (Table 6). The lack of inhibition of PA formation by mepacrine was evident even at concentrations lower than 0.4 mM employed in this study (data not shown). Our results also indicate that the inhibition of AA release from phospholipids observed in the presence of mepacrine did not result from a simple depletion of phospholipid substrates, owing to formation of a complex between phospholipids and mepacrine as reported previously (Dise et al. 1982). We found no phospholipid complexes on our thin-layer chromatographic plates developed in the solvent systems described in this report. Moreover, the inhibition of AA accumulation in the FFA fraction observed in the presence of mepacrine demonstrate that the actual release of AA from phospholipids was affected in the presence of this compound or NCDC or PMSF.

Discussion It has been reported previously that the combined release of AA from phospholipids such as PC, PI, and PE by a route involving PLA2 serves as the major pathway for eicosanoid biosynthesis in stimulated platelets (Bills et al. 1977; Walengaet al. 1981; Billah et al. 1981; McKean et al. 1981; Billah and Lapetina 1982; Smith et al. 1985; Mahadevappa and Holub 1986; Broekman 1986; Mahadevappa and Holub 1987; Mahadevappa 1987). Furthermore, we have demonstrated that the DG lipase pathway, which involves a sequential removal of stearic and arachidonic acids by DG and MG lipases, respectively, from DG derived from phosphoinositides, provides only 15-20% of the total AA released in thrombin-stimulated human platelets (Mahadevappa and Holub 1986). In addition, PI may also release AA directly through the action of PLA, in thrombin-stimulated platelets (Billah and Lapetina 1982). The use of a dual inhibitor of cyclooxygenase and lipoxygenase in the present work permitted the examination of the effect of NCDC and PMSF (serine esterase inhibitors), and mepacrine (a specific PLAz inhibitor), on thrombininduced loss of [ 3 ~ from ] ~individual ~ platelet phospholipids and the comparison of these results with the accumulation of nonesterified [ 3 ~ ] A Awhich , was readily detected in the FFA fraction. This approach eliminated the possible interpretation that the inhibition of AA release caused by these compounds may be due to substrate depletion rather than due to a direct inhibition of phospholipases that release AA, as these compounds have a tendency to

MAHADEVAPPA AND SICILIA

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

form complexes with membrane phospholipids (Dise et al. 1982) because of their chemical nature. The present work also provided a means in the evaluation of the relative importance of different phospholipid sources and pathways for the release of AA in thrombin-stimulated human platelets. In general agreement with previous results from our laboratory (Mahadevappa and Holub 1986,1987; Mahadevappa 1987; Mahadevappa and Sicilia 1988), thrombin alone induced an extensive accumulation of free [ 3 ~ ] A A following the inhibition of cyclooxygenase and lipoxygenase(s) by BW755C. Comparison of the net loss of labelled AA from individual phospholipids with that gained in the nonesterified AA revealed that 80-90% of the AA lost from phospholipids in response to the action of thrombin appeared in the FFA fraction (Tables 1-6). Furthermore, our results demonstrate that 70070 of the [ 3 ~ ] A A accumulated in the nonesterified fatty acid fraction following stimulation with thrombin was derived from PC. This finding is in general agreement with previous reports from our laboratory (Mahadevappa and Sicilia 1988), as well as from others (Walenga et al. 1981; Smith et al. 1985). Interestingly, the stimulation of platelets with thrombin alone resulted in a marked increase in the radioactivity of PE. This increase in PE appears to be due to an increased transfer of [ 3 ~ ] A A from PC to the alkenyl species of PE via the stimulation of a c a 2 + - and CoA-independent transacylase enzyme present in platelets. Similar findings have been previously reported in platelets (Rittenhouse-Simmons and Deykin 1981; Purdon and Smith 1985; Mahadevappa 1987). Recent studies indicate that deacylation of arachidonoyl species of phospholipids in response to agonists in platelets can occur independent of cytoplasmic c a 2 + increases, aggregation, and oxygenated products of AA (Pollock et al. 1986; Vedelago and Mahadevappa 1988; Pich6 and Mahadevappa 1989). The biochemical explanation for the highly selective metabolism of arachidonoyl species of phospholipids in stimulated phospholipids remains unknown. The results obtained on the effect of NCDC (0.6 mM) are in general agreement with those reported previously (Walenga et al. 1980). Walenga et al. (1980) showed that both NCDC and PMSF inhibit agonist-induced activation of PLC and suggested that the inhibition of PLC activity by these compounds was not due to an inactivation of thrombin, but was due to their direct effects on PLC. Our results demonstrate that NCDC, even at much lower concentrations, relative to previous studies, abolished completely the hydrolysis of PC and PI, and the accumulation of PA and free AA, thereby suggesting the inhibition of both PLC and PLA2. However, while the accumulation of [ 3 ~ ] A Ain the 'FFA fraction was inhibited completely (100%), the loss of [ 3 ~ ] A A from PS remained unaffected in the presence of NCDC. These results suggest that PS does not contribute towards the release of AA in thrombinstimulated platelets. In addition, the inhibition of radioactivity increase in PE observed in the presence of NCDC suggests that the thrombin-induced transacylase activity was also abolished by this compound. Although the exact nature of interaction between NCDC and platelet phospholipases remains unknown, the inhibition of these enzymes observed in the presence of NCDC is likely to have occurred from its combined effects on PLC and PLA2. However, the inhibition is not due to the formation of an inhibitor and

525

phospholipid complex, since the actual release of AA was also abolished (Table 2). The inhibition of PLC and PLA2 observed in the presence of NCDC appears, therefore, to occur prior to the coupling of PLC with its GTP-binding protein(s) in the plasma membrane, because its effect on later steps would not have completely abolished both the production of PA and the release of AA. Whether or not the inhibition of phospholipid deacylation observed in the presence of NCDC is due to its general antiproteolytic property or its effects through serine esterase remains unknown. However, the role of proteolysis in the regulation of phospholipid deacylation is being actively considered in our recent studies. It is also very likely that proteolytic products of PKC that are known to be involved in the regulation of PLA2 may be affected in the presence of NCDC. Inhibition of proteolysis by NCDC may result in the dissociation of GTP-binding proteins from plasma membrane phospholipases. However, the reasons for the differences in the behaviour of these two serine esterase inhibitors (NCDC and PMSF) on the deacylation of platelet phospholipids remain unclear. These two inhibitors do not appear to act through a common site. The use of NCDC, however, provided no clue on the relative importance of the various pathways that are responsible for the mobilization of AA from phospholipids in thrombin-stimulated human platelets. Our results regarding the effect of PMSF on thrombininduced activation of PLC and release of AA from phospholipids are not in agreement with those reported previously (Walenga et al. 1980). In this work we found that 85% of the AA released following stimulation with thrombin was inhibited in the presence of PMSF, while the formation of [ 3 ~was] only ~ reduced ~ by 5%. We further noted that the loss of [ 3 ~ ] A A from PC and the gain in PE observed in response to the action of thrombin were both inhibited completely by PMSF, while the net loss of [ 3 ~ ] A A from PI and net accumulation of [ 3 ~ ] A A were partially affected (56 and 85070, respectively). Thus, PMSF appears to selectively affect the release of AA, possibly through inhibition of PLA2 without affecting the activity of PLC. These results indicate that PMSF effects are highly selective to the inhibition of PLA2, which appears to contribute about 85% of the total AA released in thrombin-stimulated platelets (Table 4). Comparison of the accumulation of [ 3 ~and] [ 3~ ~ ]~A A in the presence of PMSF (2 mM) further suggests that hydrolysis of PA by PA-specific PLA2 is either not involved in the release of AA or is a very minor route in the release of AA in thrombin-stimulated human platelets (Neufeld and Majerus 1983; Mahadevappa and Sicilia 1988). However, the presence of PA-specific PLA2 has been previously reported in platelets (Billah et al. 1981). The serine esterase inhibitor PMSF also affected the transfer of labelled AA from PC to PE, thereby suggesting the inhibition of c a 2 + -and CoA-independent transacylase activity. The reduced transfer of [ 3 ~ ] A A from PC to PE observed in this study may have resulted from the inhibition of PLA2. Thus, unlike NCDC, PMSF appears to selectively inhibit the agonist-induced deacylation of platelet phospholipids and this inhibitory effect of PMSF is not due to substrate depletion, since the actual release (monitored through free labelled AA) was affected. The results obtained in regards to the effect of mepacrine on thrombin-induced mobilization of AA and formation of

526

BIOCHEM. CELL

BIOL. VOL.

P A are summarized in Tables 5 and 6. I n the presence of mepacrine (0.4 mM), a n increased accumulation of P A was apparent with concomitant reduction in the accumulation of [ 3 ~ i n ] the~ nonesterified ~ fatty acid fraction (Table 6). A parallel reduction in the loss of ['HIM from P C and PI was also observed (Table 5). These results are in general agreement with those re orted by Lapetina et al. (1981). Increased formation of [ H]PA can result from ( i ) an inhibition of PA-specific PLA2 in the presence of mepacrine (Billah et al. 1981; McCrea et al. 1985; (ii) an increased phosphorylation of DG by a D G kinase in the presence of mepacrine; and (iii) a n inhibition of PA-specific phosphohydrolase (Allan and Michell 1975; Brindley and Bowley 1975). However, the results reported in this study with mepacrine suggest that the route involving PA-specific PLA, does not represent a major source for AA release, although this pathway may provide small amounts of AA in thrombin-stimulated platelets (Lapetina et a/. 1981; McCrea et al. 1985). Nevertheless, much of the A A released is derived from P C and P I via the action of the enzymes of the PLAz type, even though small amounts of AA can be released through the D G lipase pathway in thrombinstimulated platelets (Mahadevappa and Holub 1986; Mahadevappa and Sicilia 1988). Similar t o that observed with NCDC and PMSF, mepacrine also inhibited the transfer of [ 3 ~ ] A A from P C to PE, thereby suggesting the inhibition of c a 2 + -and CoA-independent transacylase. In summary, our results show that NCDC and PMSF (serine esterase inhibitors) inhibit the mobilization of AA, but not through a common site, in thrombin-stimulated platelets. While NCDC inhibited the activities of PLC, PLA,, and transacylase, PMSF caused a selective inhibition of ca2+-dependent PLA2 and ca2+- and CoAindependent transacylase activities without affecting the activity of PLC. O n the other hand, mepacrine inhibited the release of A A into the nonesterified fatty acid fraction with concomitant stimulation of PA. Our results further demonstrate that the inhibition of AA release in the presence of NCDC, PMSF, and mepacrine is not due t o the formation of complexes between inhibitors and phospholipids. Based upon these results, we conclude that the hydrolysis of P C by PLA2 represents the major source for eicosanoid biosynthesis in stimulated platelets.

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

T

Acknowledgements The present work was carried out during the tenure of a Senior Research Scholarship (Career Development Award) from the Canadian Heart Foundation and was supported by a grant from the Heart and Stroke Foundation of Ontario (V.G.M.). The assistance provided by Mrs. Andra Williams during the preparation of this manuscript is greatly appreciated. ALLAN,D., and MICHELL,R.H. 1975. Enhanced synthesis of de novo of phosphatidylinositol in lymphocytes treated with cationic amphiphilic drugs. Biochem. J. 148: 471-478. BELL, R.L., KENNERLY, D.A., STANFORD, N., and MAJERUS, P.W. 1979. Diacylglycerol lipase: A pathway for arachidonate release from human platelets. Proc. Natl. Acad. Sci. U.S.A. 76: 3238-3241. BERRIDGE, M.J. 1987. Inositol triphosphate and diacylglycerol: two interacting second messengers. Annu. Rev. Biochem. 56: 159-193.

68, 1990

BILLAH,M.M., and LAPETINA,E.G. 1982. Formation of lysophosphatidylinositolin platelets stimulated with thrombin or ionophore A23187. J. Biol. Chem. 257: 5196-5200. BILLAH,M.M., LAPETINA, E.G., and CUATRECASAS, P. 1981. Phospholipase A, activity specific for phosphatidic acid: a possible mechanism for the production of arachidonic acid in platelets. J. Biol. Chem. 256: 5399-5403. BILLS,T.K., SMITH,J.B., and SILVER,M.J. 1977. Selective release of arachidonic acid from the phospholipids of human platelets in response to thrombin. J. Clin. Invest. 60: 1-6. BLIGH,E.G., and DYER,W.J. 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37: 911-919. BRINDLEY, D.N., and BOWLEY, M. 1975. Drugs affecting the synthesis of glycerides and phospholipids in rat liver. Biochem. J. 148: 461-469. BROEKMAN, M.J. 1986. Stimulated platelets release equivalent amounts of arachidonic acid from phosphatidylcholine, phosphatidylethanolamine and inositides. J. Lipid Res. 27: 884-891. CASEY,P. J., and GILMAN, A.G. 1988. G protein involvement in receptor-effector coupling. J. Biol. Chem. 263: 2577-2580. CHAU,L.-Y., and TAI,H.-H. 1981. Release of arachidonate from diglyceride in human platelets requires the sequential action of a diglyceride lipase and a monoglyceride lipase. Biochem. Biophys. Res. Commun. 100: 1688-1695. DISE,C.A., BURCH,J.W., and GOODMAN, D.B.P. 1982. Direct interaction of mepacrine with erythrocyte and platelet membrane phospholipid. J. Biol. Chem. 257: 4701-4704. FEINSTEIN, M.B., and HALENDA, S.P. 1988. Arachidonic acid mobilization in platelets: The possible role of protein kinase C and G-proteins. Experientia, 44: 101-104. FUSE,I., and TAI,H.-H. 1988. Does protein kinase C activation mediate thrombin-induced arachidonate release in human platelets? Biochim. Biophys. Acta, 972: 54-59. HOLMSEN. H., DANIEL,J.L., DANGELMAIER, C.A., MOLISH.I., RIGMAIDEN, M., and SMITH,J.B. 1984. Differential effects of trifluoperazine on arachidonic acid liberation, secretion and myosin phosphorylation in intact platelets. Thromb. Res. 36: 419-428. IRVINE, R.F. 1982. How is the level of free arachidonic acid controlled in mammalian cells? Biochem. J. 204: 3-16. LAGARDE, M. 1988. Metabolism of fatty acids by platelets and the functions of various metabolites in mediating platelet function. Prog. Lipid Res. 27: 135-152. LAGARDE, M., GUALDE, N., and RIGAUD,M. 1989. Metabolic interactions between eicosanoids in blood and vascular cells. Biochem. J. 257: 313-320. LAPETINA,E.G. 1986. Inositide-dependent and independent mechanisms in platelet activation. I n Receptor biochemistry and methodology. Vol. 7. Edited by J.W. Putney, Jr. Alan R. Liss, Inc., New York. pp. 271-286. LAPETINA, E.G., BILLAH,M.M., and CUATRECASAS, P. 1981. The phosphatidylinositol cycle and the regulation of arachidonic acid production. Nature (London), 292: 367-369. MAHADEVAPPA, V.G. 1987. Thrombin-induced transfer of arachidonic acid in human platelets is not inhibited by trifluoperazine. Biochem. Biophys. Res. Commun. 144: 821-828. MAHADEVAPPA, V.G., and HOLUB,B. J. 1986. Diacylglycerol lipase pathway is a minor source of released arachidonic acid in thrombin-stimulated human platelets. Biochem. Biophys. Res. Commun. 134: 1327-1333. 1987. Quantitative loss of individual eicosapentaenoylrelative to arachidonoyl-containingphospholipids in thrombinstimulated human platelets. J. Lipid Res. 28: 1275-1280. MAHADEVAPPA, V.G., and SICILIA,F. 1988. The inhibition of arachidonic acid mobilization in human platelets by R59 022, a diacylglycerol kinase inhibitor. Biochim. Biophys. Acta, 961: 309-315.

MAHADEVAPPA AND SICILIA

527

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by UNIV CALGARY on 08/01/12 For personal use only.

acid in the major diacyl and ether phospholipids of human MAJERUS,P.W., CONNOLLY,T.M., BANSAL,V.S., INHORN, platelets. J. Biol. Chem. 260: 12 700 - 12 704. R.C., Ross. T.S., and LIPS, D.L. 1988. Inositol phosphates: Synthesis and degradation. J. Biol. Chem. 263: 3051-3054. RITTENHOUSE-SIMMONS, S. 1979. Production of diglyceride from MAUCO,G., CHAP, H., and DOUSTE-BLAZY, L. 1979. Characphosphatidylinositol in activated human platelets. J. Clin. Invest. terization and properties of a phosphatidylinositol phospho63: 580-587. diesterase (phospholipase C) from platelet cytosol. FEBS Lett. RITTENHOUSE-SIMMONS, S., and DEYKIN,D. 1981. Mobilization 100: 367-370. of arachidonic acid in platelets. In Platelets in biology and MCCREA, J.M., ROBINSON,P., and GERRARD,J.M. 1985. pathology. Vol. 2. Edited by J.L. Gordon. Elsevier NorthMepacrine (quinacrine) inhibition of thrombin-induced platelet Holland Biomedical Press, Amsterdam. pp. 349-372. responses can be overcome by lysophosphatidic acid. Biochim. SIESS,W. 1989. Molecular mechanisms of platelet activation. Biophys. Acta, 842: 189-194. Physiol. Rev. 69: 58-178. MCKEAN,M.L., SMITH,J.B., and SILVER,M. J. 1981. Formation SMITH,J.B., DANGELMAIER, C., and M ~ u c oG. , 1985. Measureof lysophosphatidylcholine by human platelets in response to ment of arachidonic acid liberation in thrombin-stimulated thrombin. J. Biol. Chem. 256: 1522-1524. human platelets. Use of agents that inhibit both the cyclooxygenase and lipoxygenase enzymes. Biochim. Biophys. Acta, NEUFELD,E.J., and MAJERUS,P.W. 1983. Arachidonate release and phosphatidic acid turnover in stimulated human platelets. 835: 344-35 1. J. Biol. Chem. 258: 2461-2467. VANDERHOEK, J.Y., and FEINSTEIN,M.B. 1979. Local anesthetics, chlorpromazine and propanolol inhibit stimulusNISHIZUKA, Y. 1984a.The role of protein kinase C in cell surface signal transduction and tumor promotion. Nature (London), 308: activation of phospholipase A, in human platelets. Mol. Pharmacol. 16: 171-180. 693-697. 1984b. Turnover of inositol phospholipids and signal VEDELAGO, H.R., and MAHADEVAPPA, V.G. 1988. Mobilization of arachidonic acid in collagen-stimulated human platelets. transduction. Science (Washington, D.C.), 225: 1365-1370. PICHC,L., and MAHADEVAPPA, V.G. 1989. Aggregation and/or Biochem. J. 256: 981-987. oxygenated products of arachidonic acid are not required for WALENGA,R.W., VANDERHOEK, J.Y., and FEINSTEIN,M.B. collagen-induced deacylation of phosphatidylcholine in human 1980. Serine esterase inhibitors block stimulus-induced mobilizaplatelets. Biochem. J. 263: 143-148. tion of arachidonic acid and phosphatidylinositide-specific POLLOCK,W.K., RINK,T.G., and IRVINE,R.F. 1986. Mobilizaphospholipase C activity in platelets. J. Biol. Chem. 255: tion of arachidonic acid in collagen-stimulated platelets. 6024-6027. Biochem. J. 235: 869-877. WALENGA,R.W., OPAS,E.E., and FEINSTEIN,M.B. 1981. DifPRESCOTT, S.M., and MAJERUS,P.W. 1983.Characterization of ferential effects of calmodulin antagonists on phospholipases A, 1,2-diacylglycerolhydrolysis in human platelets. J. Biol. Chem. and C in thrombin-stimulated platelets. J. Biol. Chem. 256: 258: 764-769. 12 523 - 12 528. PURDON,A.D., and SMITH,J.B. 1985. Turnover of arachidonic