Biochimica
et Biophysics
0 1992 Elsevier
BBALIP
Science
Acta,
279
1124 (1992) 279-287
Publishers
B.V. All rights reserved
0005-2760/92/$05.00
53858
Activation of phospholipase A 2 and P-thromboglobulin release in human platelets: comparative effects of thrombin and fluoroaluminate stimulation Mario Stasi ‘, Paolo Gresele ‘, Serena Porcellati ‘, Elisabetta Quero Goracci ” Giuseppe G. Nenci ’ and Gianfrancesco I’ lstituto
di Biochimica
e Chimica
Medica,
Utzil’ersiti
Ut~kersitrl
(Revised
Key words:
Arachidonate
release;
di Perugia,
di Perqia,
Perugia
Perugia
(Italy)
A,; Thrombin;
” Istituto
di Semeiotica
Medica,
(Italy)
(Received 29 July 1991) manuscript received 12 November
Phospholipase
and
‘,
Fluoroaluminate;
1991)
G-protein;
cGMP; (Human
platelet)
Several reports have suggested that the activity of platelet phospholipase A, is modulated by GTP-binding protein(s) whose nature and properties need to be defined. Fluoroaluminate is known to activate G-proteins and this leads to a number of cellular responses including the activation of phospholipases. This paper demonstrates that human platelets, prelabelled with r3Hlarachidonic acid, produce free arachidonic acid when stimulated with fluoroaluminate and this effect is time- and dose-dependent. The production of arachidonic acid is not inhibited by neomycin, a PI-cycle inhibitor, but is completely abolished by mepacrine, an inhibitor of both phospholipase A, and C. At low concentration of fluoroaluminate (10 mM NaF) phospholipase A, but not phospholipase C is activated. In addition, fluoroaluminate treatment releases /34hromboglobulin t/3-TG) and this effect is not inhibited by acetylsalicylic acid. Under identical conditions both neomycin and mepacrine suppress the release of arachidonic acid and &TG induced by thrombin. Sodium nitroprusside, which increases cGMP levels in platelets, inhibits arachidonic acid liberation and @-TG release in thrombin-stimulated platelets but has no effect in fluoroaluminate-treated platelets; cGMP was reported to suppress phospholipase C activation. These results are consistent with the hypothesis that, in thrombin-stimulated platelets, the liberation of arachidonic acid and P-TG are strictly dependent on the activation of phospholipase C. We have also provided evidence for the existence of a phospholipase A, activated by a G-protein which is independent from the degradation of phosphoinositides and, contrary to phospholipase C, it is not down regulated by cGMP.
Introduction It is well with several arachidonic This event synthesis of on platelet
established that the stimulation of platelets agonists leads to the production of free acid from membrane phospholipids [1,2]. is considered a rate-limiting step for the eicosanoids [31, which play important roles function as well as on the interaction of
Abbreviations: AA, arachidonic acid; ASA, acetylsalicylic acid; P-TG, P-thromboglobulin; PA, phosphatidic acid; PC, acid-stable choline phosphoglycerides; PGI,, prostacyclin; PI, phosphatidylinositol; SNP, sodium nitroprusside; WLP, washed labelled platelets; WP, washed platelets. Correspondence: ica, Universita
G. Goracci, Istituto di Biochimica e Chimica Meddi Perugia, Via del Giochetto, 06100 Perugia, Italy.
platelets with other blood cells and with the vascular wall [4]. In stimulated platelets, it is almost generally accepted that the mobilization of arachidonic acid from membrane phospholipids is mainly the consequence of an increased activity of phospholipase A, [5]. However, the identity of platelet phospholipase(sI A,, their subcellular localization, their specificity for substrates and their regulation are not completely defined [6]. In human platelets, the highest enzymic activity has been measured in the cytosolic fraction and it has been shown that this enzyme is Ca2+-dependent [7] but a consistent activity is also associated with membrane preparations [X]. The specific activity of phospholipase A, in human platelets is much lower than that measured in platelets from other animal species as rat and rabbit [9]. In rat platelets the major part of phospholipase A, activity is
280 localized in the cu-granules [Y] and is secreted after stimulation with various agonists [lO]. In addition, a membrane-bound enzyme has been also purified from rat platelets and it has been shown that this enzyme shares common properties and structure with that secretcd after stimulation with thrombin [ 111. Recently, Kramer et al. [12] have reported the structure of a C’a’ ‘-dependent human platelet phospholipase A2 which is also secreted in the active form. A novel type of high molecular weight, cytosolic phospholipase A1 has been purified from platelets [ 1% 151, and seems to be very similar to the recently cloned phospholipase A, from U9.37 cells [Ih]. In fact, these enzymes exhibit prcfercncc for arachidonic acid and are activated by less than micromolar concentration of Ca”. These properties arc consistent with the hypothesis that this phospholipase A, type may be involved in the receptor-mediated release of arachidonic acid in platelets [ 13-161. However, the molecular mechanisms responsible for receptor-mediated activation of phospholipase A, in platelets and in many other cell types are still not completely known. The thrombin-induced platelet response has been extensively studied and it has been shown that this agonist induces shape change, secretion of granular contents and aggregation. The thrombin-receptor interaction activates phospholipasc C which produces inositol phosphates and diacylglycerols from inositol phospholipids [ 171. Diacylglycerols activate protein kinase C and inositol 1,4,5_trisphosphate (IP,) increases by the mobilization of cytosolic Ca’ + concentration this ion from intracellular calcium stores [18]. The activation of phospholipase C due to thrombinreceptor interaction appears to be mediated by a GTP-binding protein (G-protein) from the Gq class [IO]. It has been also shown that the stimulation of human platelets with thrombin increases the levels of cGMP [20] which in turn inhibits platelet activation and it is thus considered a feed-back regulator [21]. The inhibitory effect of cGMP is most likely exerted at the level of the activation of phospholipase C because the preincubation of platelets with sodium nitroprusside (SNP), which activates the soluble guanylate cyclase. or with 8-Br-cGMP, a membrane-permeable stable cGMP analogue, reduces the production of diacylglycerol [22] and that of inositol phosphates from phosphoinosirides in thrombin-stimulated platelets [23]. The stimulation of platelets with thrombin increases also phospholipase A 1 activity and it has been proposed that this event might bc secondary to the activation of phospholipasc C being the IP,-mediated inthe trigger for crease of cytosolic Ca’+ concentration the activation of phospholipase A, [24]. On the other hand, other studies have shown that the sole elevation of Ca’ + concentration is not sufficient for activating phospholipase A, in thrombin-stimulated platelets [25].
Kienast et al. 1261 have shown that fluoride, which activates G-proteins [27], induces a dose-dependent aggregation of intact human platelets together with an increased degradation of phosphoinositides due to the activation of phospholipase C. It has also been reported that fluoroaluminate (AIF,-) stimulates the production of thromboxane B2 (TxB, 1 and IP, when added to a human platelet suspension [2X]. Marc recently, Silk et al. [29] have shown that the incubation of human platelet membranes with AIF,- or GTP-y-S stimulates the release of arachidonic acid from cndogcnous phospholipids. These observations suggest that the direct activation of G-proteins by nonhydrolyzable GTP analogs or by fluoroaluminate results in the stimulation of both phospholipase C and A2 activities, an effect similar to that obtained with the stimulation of platelets with thrombin. However, it is not known whether the activation of phospholipase A,, due to the direct activation of Gproteins, requires necessarily the previous stimulation of a phosphoinositide-specific phospholipase C similarly to what happens after thrombin stimulation [17]. Platelet activation induced by fluoroaluminatc was reported to be inhibited by cGMP due to the interference of this cyclic nucleotide with G-protein phospholipase C [Xl]. However, it is not known whether cGMP interferes with the release of arachidonic acid in fluoroaluminate-stimulated platelets. In the present study. we have compared the effects of fluoroaluminate and thrombin on the mobilization of arachidonic acid and on the release of /3-TG, an a-granule protein. from intact human platelets. Our results provide evidence that the activity of phospholipasc AZ can be stimulated by a mechanism mediated by a G-protein, not affected by cGMP, and does not necessitate the previous activation of phospholipase C. In addition, the release of /3-TG induced by fluoroaluminate stimulation in human platelets shares some similarities with the liberation of arachidonic acid. Materials
and Methods
Materials [5,6,8,9,1 1,12,14,15-3H]arachidonic acid (spec. rad. 191-220 Ci/mmol) was purchased from NEN Rcsearch Products (Boston, MA, USA). PGI,, bovine serum albumin (BSA, fraction V), apyrase, neomycin and sodium nitroprusside were from Sigma Chemical (USA). Bovine thrombin (Topostasin) was from Roche (Basel, Switzerland). Silica Gel 60A plates (Whatman International Ltd. Maidstone, U.K.) were used for thin-layer chromatography. Preparation qf platelets lahelled with I~‘Hlarachidonic acid Blood (50-60 ml) was obtained from healthy voluntcers using citrate-dextrose as anticoagulant (1 : 6 vol-
281 umes of blood). Platelets were pelleted in the presence of 0.2 PM prostacyclin (PGI,) and 50 U/ml heparin and then resuspended in 10 ml of Tyrode’s buffer containing 0.35% bovine serum albumin, apyrase and PGI,. This procedure was repeated twice to obtain washed platelets (WP). The incubation of WP with labelled arachidonate (lo-15 PCi) was carried out for 1 h at 37°C as previously described [31,32]. Labelled platelets were washed according to Kinlough-Rathbone et al. [33] using a Tyrode-BSA buffer without Ca’+ and Mg *+ ions. At the end, washed labelled platelets (WLP) were resuspended in the same medium and the final concentration was adjusted at 2-3 . 10” platelets/ml. Stimulation of lahelled platelets WLP suspension (0.5 ml> was placed in the cell of a dual channel aggregometer (Elvi Logos, Milan, Italy) and then stimulated with 20 ~1 of a solution containing NaF and AICl,, in a molar ratio 5000: 1, or with thrombin 0.5 U/ml. When indicated, samples were preincubated at 37°C with 10 mM neomycin for 10 min or with 0.5 mM mepacrine for 2 min or with 20 @M nitroprusside for 1 min. Platelet aggregation was monitored by Born’s method [34]. At the end of the stimulation period, 0.15 ml of 0.1 M EDTA (pH 7.4) were added to the cell and its content was then immediately poured into a tube containing 2.4 ml of chloroform/ methanol (1 : 2, v/v> ]321. Extraction and analysis of lipids Lipids were extracted according to Bills et al. [3]. Free arachidonate was isolated by TLC using petroleum ether (30”-50”)/ diethyl ether/ acetic acid (70 : 30 : 1, v/v). Thromboxane B, was isolated by TLC on silica Gel 60A plates using benzene/ dioxane/ acetic acid (40:20:2, v / v ) as developing solvent [3]. Unlabelled arachidonic acid and thromboxane B, was added before the separation by TLC. Phospholipid classes were separated by bidimensional TLC with exposure to HCI fumes between the first and the second dimension [35]. The areas containing acid-stable cholinephosphoglycerides (I-alkyl-2 acyl-+ 1,2-diacyl-sn-glycero-3phosphocholine), phosphatidic acid and phosphatidylinositol were scraped off. The radioactivity was determined by liquid scintillation counting (Emulsifier-safe, Packard Instrument Company, IL, USA) and quenching corrected by external standardization. Determination of p-TG Aliquots of WP were stimulated with either fluoroaluminate or thrombin in an aggregometer cuvette at 37°C upon stirring for different time intervals. The samples were then immediately centrifuged at 12 000 x g for 2 min and the supernatant stored at - 80°C for subsequent assay. Total platelet content of P-TG was
estimated in the supernatant of aliquots of WP after the addition of 5 ~1 Triton X-100 (Sigma, St. Louis, MO, USA). Further processing and assay of the samples were carried out as described [36] by ELISA (Boehringer Biochemica, Italy). Determination of cGMP in washed platelets cGMP was assayed in 1 ml of WP suspension (2.5. 10’ platelets) after incubation with 20 PM nitroprusside for 1 min or its vehicle followed by the addition of 1 ml of 1 M ice-cold perchloric acid. After centrifugation, the clear supernatant was neutralized with KOH and the precipitate was removed. The solution was freeze-dried and stored at -20°C until used for radioimmunological assay (Amersham, U.K.). Statistical analysis Statistical analysis of the data was carried out by ANOVA followed by multiple comparison (Sheffe’s test). Results The incubation of platelets with labelled arachidonic acid led to its incorporation into membrane phospholipids and the distribution of the radioactivity among the various phospholipid classes was similar to that previously reported [32]. When labelled platelets were stimulated with fluoroaluminate, a slow aggregation was observed which reached the maximal amplitude after 15-20 min (Fig. 1). Under identical conditions, thrombin (0.5 U/ml> induced a more rapid aggregation which was maximal after 3 min (data not shown). The maximal amplitude of platelet aggregation, observed after stimulation with thrombin for 20 min, was more than two times greater than that obtained with fluoroaluminate in the same conditions.
3
50, .
-60
% .s
minutes
Fig. 1. Effect of fluoroaluminate on the release of arachidonic acid and aggregation of prelabelled human platelets. Human platelets were prelabelled with [‘Hlarachidonic acid (S.R. 190 Ci/mmol) washed as described under Materials and Methods, and incubated with 21.1 mM NaF+4.2 FM AICI, for the indicated time at 37°C with continuous stirring. These data are from one experiment in duplicate representative of three others which gave similar results.
282
NoF (mM) Fig. 3. Effect Time of stimulation Fig. 2. Effect
of tluoroaluminate
on the release of a-TG
platelets. Washed platelets were stimulated presence of 4.2 FM by ELISA.
AICI,
of labelled
(min)
and /3-TG
with 21, I mM NaF in the
was measured
Data are from 01x experiment
in human
in duplicate
in the medium representative
of another which gave similar results.
platelets. (S.R.
of fluoroaluminate
arachidonic Platelets
prelahelled
with
on the mobilization
radioactive
and washed as described
Platelets were then incubated
concentrations P-TG
were
I90 Ci/mmol)
Methods.
concentration
acid and on the release of P-TG under
refer to one experiment
washed
in duplicate
and
Materials
for IO min with the indicated
1 in
of NaF in the presence of AICI
was assayed using unlabelled
in human
arachidonate
the ratio 5000:
human
representative
platcleta.
I,
Data
of another which
gave similar results.
Previously we have reported that the stimulation of WLP with 0.5 U/ml thrombin caused a rapid release of labelled arachidonic acid from endogenous phospholipids being this effect already significant after 10 s from the stimulation and the maximal liberation of labelled arachidonic acid was observed after 3 min [32]. The stimulation of WLP with fluoroaluminate caused an increase of the radioactivity of arachidonic acid which was time-dependent and almost paralleled the extent of platelet aggregation. As shown in Fig. I, 20 min after the addition of fluoroaluminate the radioactivity of arachidonic acid was more than IO-times greater than its initial level but was lower than that found when platelets were stimulated with thrombin in the same conditions. The incubation of WLP for 20 min without addition of fluoroaluminate did not cause any significant change of the arachidonic acid radioactivity. A time-dependent release of P-TG was also observed from platelets stimulated with fluoroaluminate (Fig. 2) which was also maximal after 15-20 min. The mobilization of radioactive arachidonic acid from prelabelled platelets and the release of P-TG were dependent on fluoroaluminate concentration (Fig. 3). The maximal effect was observed with 30-40 mM NaF in the presence of AICI,. At higher concentrations of fluoroaluminate a slight decrease of both free arachidonic acid production and /?-TG release was observed (Fig. 3). Another series of experiments was performed for studying the effect of 10 mM neomycin or 0.5 mM mepacrine on the liberation of arachidonic acid from platelets stimulated with fluoroaluminate or thrombin. At this concentration neomycin is known to inhibit the degradation of phosphoinositides by phospholipase C [37], whereas mepacrine (0.5 mM) inhibits both phospholipase C and AI [38]. In these experiments, both
the concentration of fluoroaluminate and the duration of the stimulation were chosen to have comparable production of labelled arachidonic acid with respect to samples stimulated with thrombin. The results shown in Table I demonstrate that neomycin significantly reduced the release of arachidonic acid in thrombinstimulated but not in fluoroaluminate-stimulated platelets. Mepacrine completely blocked the release of arachidonic acid both in thrombin- or fluoroaluminatestimulated WLP (Table 1).
I
TABLE
WLP
were
/3-TG
release was measured
prepared
with AIF, U/ml
(21.1 mM
Thrombin
preincuhated
with
as described
under WLP AICI,)
NaF+3.2
PM
for 3 min.
When
OS mM
mepacrine
for 20 min or with 0.5 WLP
Data
P-TG
(ng/IOH
platelets) I.6 (II = 14)
were
10 mM
are the means I S.D.
AA production
release platelets)
312+_ 100 (II = 6)
(,I = 0) I’ 1’0 + 452 (II = 6) A”
I 434 * 24x
AIF,-
50.4f
Neomycin + AIF,
55.0 * 26.5 (It = 6) ~’ 11 7.‘) + 2.x (n = 4)
2
55. I * IS.5 07 = 4) ‘I
2316*443(?l=
4) ,’
‘
404 +_ IO0
( ,I = 4) ‘
Mepacrine
+ AIF,
Thrombin
IS.1 01=
or WP
for 2 min or with
(nCi/lO” 4.9*
Resting
and Methods.
or WP were stimulated
indicated.
neomycin for IO min before stimulation. Treatment
Materials
in WP.
8) ‘I
&I
I32 or=
4) ”
Neomycin + Thrombin Mepacrine Thrombin
0.7 01 = 4)
6.‘) *
0. I (12 = 4)
384 +
16.72
F value Total content (Triton 466.89
3.7+ +
(n = 8) ng/lOx
X-100
20 (II = 4)
L
43.70 extraction)
platelets.
of P-TG
P < tt.0001
,’ P < 0.05 vs. resting: ” P < 0.05 vs. AIF,
in WP was 2YX2.S i for
the
two groups.
; ’ P < 0.05vs. thrombin.
283 350
Similarly, the release of /3-TG induced by fluoroaluminate was not reduced by the preincubation of WP with neomycin but it was almost completely blocked by the preincubation with mepacrine. On the contrary, both neomycin and mepacrine significantly inhibited the release of /3-TG when thrombin was the agonist (Table I). The preincubation of platelets with 200 PM acetylsalicylic acid (ASA) had no effect on the release of /3-TG both in fluoroaluminateand thrombinstimulated WP (1468 rt 132 vs. 1444 rt: 248 for fluoroaluminate and 1944 + 128 vs. 2316 + 444 for thrombin). The stimulation of WL,P with AIF,- at a relatively low concentration (10 mM NaF) caused a significant increase of the radioactivi~ of unesterified arachidonic acid but did not affect the radioactivity of phosphatidic acid (Table II). In the same conditions, the radioactivity of phosphatidylinositol was reduced, whereas the radioactivity of acid-stable choline phosphoglycerides (PC) underwenti to a slight decrease. Both effects were particularly significant when WLP were stimulated with higher concentrations of AlF,- (21 mM NaF) which was also able to increase the radioactivity associated with phosphatidic acid. The latter effect is generally considered an indirect estimation of phospholipase C activity because diglycerides produced from phosphoinositides are efficiently converted to PA by diglyceride kinase [391. The stimulation of WLP with fluoroaluminate (21 mM NaF) caused a significant increase of the labelling of TxB, with respect to unstimulated samples or stimulated by a lower concentration of fluoroaluminate (10 mM NaF). The preincubation of WLP with 200 PM ASA abolished the production of labelled TxB, but had no effect on the radioactivity of PA, PI and PC in fluoroaIuminate-stimulated platelets. When platelets were exposed to ASA, the radioactivity associated with arachidonic acid had a significant increase (Table II). As shown in Fig. 4A, the treatment of platelets with neomycin did not affect the reduction of PC radioactivTABLE
I
OPC
A
II f
+ 100
+
;
4
B
X
Fig. 4. Effect of neomycin and mepacrine on the radioactivity of acid-stable choline phosphoglycerides (PC), ph~sphatidylinositol (PI1 and phosphatidic acid (PA). Platelets were incubated with radioactive arachidonate (S.R. 240 Ci/mmoll for 1 h. Washed labelled platelets were stimulated with 0.5 U/ml thrombin for 3 min or with fluoroaluminate (21.1 mM NaF+4.2 yM AlCl,) for 20 min. When indicated WLP were preincubated with 10 mM neomycin for 10 min or with 0.5 M mepacrine for 2 min. Data are the meaniS.D. of the results from two experiments in duplicate.
ity induced by the stimulation of WLP with fluoroaluminate. On the contrary, the exposure of platelets to mepacrine almost completely abolished the loss of the
I1
Effect offluoroaluminate concentration and acetylsalicylic acid on the mobilization of lahelled arachidonic acida its conversion to fhromboxune B, and on the ~ud~oacti~i~ of phosphatidic acid, phosphutidylinositol and ~c~d-~tab~eChofinephos~~loRlyceridPs WLP were prepared as described under Materials and Methods and then stimulated with fluoroaluminate for 20 min at the indicated concentrations. Ratio between NaF and AlQ, concentrations was 5000: 1. When indicated WLP were preincubated with 200 PM acetylsalicylic acid (ASA) for 10 min at 37°C before stimulation. Data are the means+S.D. for one experiment representative of another one. Both experiments were performed in quadruplicate. Treatment
nCi/lO’
platelets
AA Resting AIF; (10 mM NaF) ASA+ AIF,- (10 mM NaF) AIF,- (21 mM NaF) ASA+ AIF,- (21 mM NaFl F value P < 0.0001 for all groups.
11.6+1.3 25.2+ 1.5 29.7 + _ 2.0 68.7+ 3.6 81.8 + _- 7.5 878.09 ’ P < 0.05 vs. resting;
TxB,
a ” a,h .JXc
PA
2.4 & 0.4 2.7 i 0.3 0.6 i 0.2 9.9 * 0.8 il,h 1.8kO.5 c 281.31
” P < 0.05 vs. AIF;
9.2+ 8.6i 7.7+ 17.2* 14.6 + 43.82
PI 1.6 1.6 1.0 1.3 ‘,’
I .6 a.h
PC
74.7k5.1 66.0 + 3.4 62.9; 4.2 48.9 + - 4.9 46.0 + 5.3 53.05
(10 mM NaFl; ’ P < 0.05 vs. AIF,-
a a Kh 0
(21 mM NaFf.
259.02 244.7 * 21x.5* 191.7k 203.0 + 6262
8.7 11.5 7.8 6.2 a.h 6.8 a.h
2K4 radioactivity from this phospholipid class (V = ns vs. resting and P < O.OSvs. AIT;;--stimulated). At the indicated conditions, the stimulation of WLP with thrombin reduced the radioactivity associated with PC to a similar extent as that obtained in fluoroaluminatestimulated platelets. The treatment of WLP with neomycin or mepacrinc before the stimulation with thrombin abolished the loss of labelled arachidonate from PC (k’ = 11s vs. resting: P < 0.05 vs. thrombinstimulat~d~~ 3oth fluoroaluminate and thrombin induced a decrease of the radioactivity of phosphatidylinosito1 (P < 0.05 vs. resting), whereas increased that associated with phosphatidic acid (PA) (P < 11.05 vs. resting) (Fig. 4B). The pr~in~uhati~~n of WLP with neomycin reduced the loss of radiolabelled arachidonic acid from PI (P < 0.05 vs. AIF,--stimulated and versus resting). Neomycin effect was especially evident in thrombin-stimulated platelets (P = ns vs. resting) whereas the radioactivity of acid-stable choiine PhosphogIyceridcs (PC) underwent to a slight decrease, The pre~ncubati~~n of WLP with either neomycin or mepacrine reduced the increase of PA radioactivity induced by fluloroaluminate or thrombin (P < 0.05). We have also evaluated the effect of cGMP on the release of @-TG and on the liberation of labellcd axachidonic acid from endogenous pbosphol~pids of platelets stimulated with thrombin or with fluoroaluminate. The level of cGMP in platelets is increased by the treatment with sodium nitroprusside (SNP) which produces nitric oxide, an activator of soluble guanylate cyclase [40]. in our experimental conditiuns the incubation of platelets with 20 PM SNP for I min induced a 2-J-fold increase of cGMP and greatly inhibited the aggregation of WLP stimulated with fluoroaluminate (21 mM
Treatment Rest kg AIF,SNP -f AlF., Thrombin SNP + Thrombin
F value
release
(ng/lU” platelets)
6.1 f vJ.4* 72.4 + 79.5 * 45.4i 43.68
272 & 120 rn = 41 1zlixJr 7fi(v=41:1 1278 &. 72 (It = 43 ;’ 1Y32 * 256 (n = 4) ” I 296 $ 60 (n = 3) il.h 7(1.83
2.2(8 = 71 I I .9 1x.4 23.Y 14.7
P < O.fKIOl for the two groups. thmmhin.
P-TG
AA production (nCi/lOy platrlets) (n In (n (n
= = = =
8) rk 81 i’ 8) :’ 8) &”
;$P < 0.05 vs. resting:
h P < 0.05 VS&
NaF) but had no effect on the aggregation induced by 0.25 U/ml of thrombin (data not shown). These results arc in agreement with those reported by Deana et al. [30]+ On the contrary, the preincubation of WLP with SNP did not affect the production of labclled arachidonic acid nor the release of P-TG in fluoroaluminatcstimulated platelets (Table III) hut significantly reduced the liberation of labelted arachidonic acid and the release of ,&TG when platelets were stimulated with thromb~~. Discussion We have presented evidence that the exposure to ffuor~~alumi~ate of intact human platelets induces the release of radioactive free arachidonic acid from prelahelled cndogcnous phospholipids. This result is in agreement with previous findings indicating that fluoroaluminate is able to induce the production of TxB, from intact human pfatelets [28,41]. The production of arachidonic acid in intact platelets due to ~u~~rua~unlinate stimulntion is a rather slow process if compared to the receptor-mediated stimulation of platelets with thrombin. It is possible that this delayed release of labelled arachidonic acid is partially related to the slow diffusion of fluoroaluminate through the membrane. In fact, it has been shown that the stimulation of permeabiiized rabbit platelets with fluoroaluminate causes a 3-foid increase of arachidonic acid radioactivity already after I min [42]. In addition, Kajiyama et al. [43] have also reported that the stimulation of permcahilized rabbit phtelcts with ffuoroaluminate increases the production of radioactive ara~hidonate but they did not find the same. effect with intact platelets. It is possible that the stimulation period used in their study with intact platelets (7 min) was too short. It has been previously reported that fluoride induces the release of two substances contained in dense granules, ATP [2h] and serotonin [44], from intact human platelets. Confirming the results reported by Rendu et al. [41], here we have shown that the stimulation of intact human platelets with fluoroaluminate is also able to induce the release of /3-TG, an a-granule protein. It seems unlikely that such an effect is secondary to the formation of cyclooxygenase products because it is not inhibited by the preincubation of platelets with ASA. Both the production of arachidonic acid and the release of P-TG are time- and dose-dependent. At relatively high ~~~nc~ntr~~~~ns of ~uoroalum~nat~ (4065 mMn) we have observed a decrease of both effects, This finding is similar to that reported for aggregation and ATP release from human platelet-rich plasma and it might be a consequence of the rise of cAMP produced by fluoroalum~nat~ in this concentration range 1261.
285 It is known that neomycin binds to phosphatidylinositol 4,Sbisphosphate [45] and thus it inhibits its hydrolysis by phospholipase C in thrombin-stimulated platelets [46]. We have shown that the preincubation of platelets with neomycin has no effect either on the liberation of arachidonic acid or on the release of /3-TG induced by fluoroaluminate while mepacrine, an inhibitor of both phospholipase A, and phospholipase C [38], suppresses both effects. On the other hand, the exposure of WLP to neomycin or mepacrine, at the indicated concentrations, abolished the release of arachidonic acid and /?-TG in platelets stimulated with thrombin. These results are consistent with the hypothesis that, in thrombin-stimulated platelets, the activation of the phosphoinositide-specific phospholipase C is absolutely required for the release of arachidonic acid. In fact, we have shown that the preincubation of platelets with neomycin inhibits the changes on the radioactivity of PI and PA and, in the same time has also an inhibitory effect on the release of arachidonic acid from acid-stable cholinephosphoglycerides in thrombin-stimulated platelets (Fig. 4). The stimulation of platelets with fluoroaluminate by-passes the agonist-receptor interaction and induces the cellular response by a direct activation of G-proteins. We have provided evidence that fluoroaluminate is able to increase the radioactivity associated with PA and this indicates that it is able to stimulate the G-protein involved in the activation of phospholipase C. However, the activation of phospholipase D [47] cannot be ruled out. Under the same conditions, fluoroaluminate activates also phospholipase A, because we have observed an increase of arachidonic acid radioactivity and a decrease of the radioactivity associated with PC. Since this effect is inhibited by mepacrine but it is insensitive to neomycin, we can assume that the activation of phospholipase A, by fluoroaluminate is regulated by a G-protein distinct from that involved in the receptor-mediated activation of phospholipase C in thrombin-stimulated platelets. While this paper was in preparation, Tysnes et al. [48] have reported that the treatment of platelets with neomycin in the concentration range 2-8 mM inhibits the binding of thrombin to its receptor. If this would be the only activity of neomycin, the inhibition of arachidonic acid release in thrombin-stimulated platelets preincubated with neomycin would reflect the lack of activation of PI cycle more than its inhibition due to the binding of neomycin to phosphoinositides. However, we have also shown that neomycin 10 mM partially prevents the changes of PI and PA radioactivity in platelets stimulated with fluoroaluminate. This indicates that neomycin, besides the effect at the level of the thrombin receptor, affects also the degradation of phosphoinositides by phospholipase C. At relatively low concentration (10 mM NaF), fluo-
roaluminate induces the liberation of labelled arachidonic acid from platelet phospholipids but does not increase the radioactivity associated with phosphatidic acid (Table II). This observation indicates that, at this concentration, fluoroaluminate activates phospholipase A, but has no effect on the activity of phospholipase C thus supporting the hypotesis that two distinct G-proteins may be responsible for the activation of these phospholipases. These results are in agreement with previous findings reported by Rendu et al. [41] where the stimulation of human platelets with fluoroaluminate (10 mM NaF) did not induce the activation of phosphoinositide-specific phospholipase C but was able to produce thromboxane B?. In our experiments, with fluoroaluminate (IO mM NaF), the conversion of labelled arachidonic acid to TxB, was not observed. This apparent discrepancy with previously reported results [41] might be due to the presence of bovine serum albumin in our medium which is known to reduce the availability of unesterified arachidonic acid for its conversion to TxB, and to other biologically active metabolites [3,49,50]. The hypothesis that two distinct mechanisms are involved in the activation of phosphoinositide phospholipase C and of phospholipase A, is further substantiated by the effect of cGMP, which is known to interfere with phospholipase C activity [22,23], on the release of arachidonic acid which is inhibited in thrombin-stimulated platelets but not in those treated with fluoroaluminate. Finally, it should be mentioned that the liberation of arachidonic acid and the release of P-TG proceed in fluoroaluminate-stimulated platelets in a parallel way and are similarly influenced by mepacrine, neomycin and nitroprusside. This might indicate the existence of common steps in the mechanisms involved in the activation of these two processes. In this study we have not evaluated whether the G-protein which modulates the activity of phospholipase A, is directly coupled to the enzyme or the activation of the enzyme is the consequence of other cellular events. For instance, it has been reported that the stimulation of platelets with NaF leads to the increase of cytosolic Ca2+ concentration [26]. This phenomenon appears to be only partially mediated by IP, because it is dependent upon extracellular Ca2+ [44]. An IP,-independent rise of cytosolic Ca” might be a possible event linking the activation of the G-protein to that of phospholipase A, and to /3-TG release. Indeed, evidence has been reported on the existence of a G-protein-mediated entry of extracellular calcium in other cells [.51,52]. However, Deana et al. [30] have shown that the stimulation of human platelets with fluoroaluminate leads to an increase of cytosolic-free Ca*+ even in the presence of EGTA in the medium and this clearly indicates its origin from intracellular
286
stores. In the same conditions, cGMP reduces the elevation of cytosolic-free CaZt 1301. On the other hand, it has been also suggested that cGMP may inhibit the influx of CaZ+ induced by thrombin in human platelets 1531.Both studies [30,531 however, lead to the conclusion that cGMP suppresses the increase of intracellular Ca”+. Considering that, in our conditions, cGMP has no effect on arachidonic acid release induced by fluoroaluminate, the possibility that the activation of phospholipas~ A, is the consequence of a G-protein mediated increase of intracellular Ca” concentration is unlikely. Our data and those provided by other laboratories with different experimental approaches support the concept that the activation of platelet phospholipase A, does not necessarily require the previous activation of phosphoinositide phospholipase C. Both enzymic activities appear to be modulated by different G-proteins; their nature and their properties need to be better defined by further studies. Finally, Akiba et al. [42] have recently shown that the mobilization of arachidonic acid from permeabilized rabbit platelets, induced by fluoroaluminate or GTP-y-S, is potentiated by phorbol 12-myristate 13acetate (PMA) thus indicating that protein kinase C is also invoived in the G-protein mediated regulation of phosph~~lipase AZ activity. Acknowledgements
This work was supported by a Research grant (to G.G.) from Minister0 d~iI’Universit~ e della Ricerca Scientifica e Tecnologica. References 1 Lapetina, E.G. (lY82) Trends Pharmacol. Sci. 3, L-18. 2 Mauco, G, (1987) in Platelets Response and Metabolism. Vol. 111 (Hoimsen. H. ed.1. DD. ~. 101-l 19. CRC Press. Boca Raton. 3 Bills, T.K., Smith. J.B. and Silver, M.J. (1976) Biochim. Biophys. Acta 424, 303-3 14. 4 Marcus, A.J. (1984) J. Lipid Res. 25, 1511-1.542. 5 Broekman, M.J. (1986) J. Lipid Res. 27, X84-891. h Waite. M. (39873 in Handbook of Lipid Research. Vol. 5, The Phospholipases (Hanahan, D.J., rd.). pp. 1 I 1-i 13,Plenum Press, New York. 7 Kramer, R.M., Checani, G.C., Deykin, A., Pritzker, C.R. and Deykin. L>. (1986) Biochim. Biophys. Acta 87X. 394-403. 8 Trugnan, Cr., Bereziat, G.. Manier, M.C. and Polonovski, J. (1979) Biochim. Biophys. Acta 573, 61-72. Y Aarsman. A.J.. ~unissen-Bijvelt~ J., Van den Koedijk, C.D.M.A., Neys, F.W., Velkleij, A.J. and Van den Bosch, H. (1989) J. Lipid Med, 1. 4%ht. IO Horigome, K., Hayakawa, M., Inoue. K. and Nojima, S. (1987) J. Biochem. 101. 53-61. 11 Hayakawa, M., Kudo. I., Tomita, M. and Inoue, K. 11988) J. Biocbem. 103. 263-266. 12 Kramer, R.M.. Hession, C.. Johansen. B., Hayes, G.. McGray, P.,
13 I4 IS lb
17 IX IY 20 71
22 23 24 25 26 27 28 29 30 31
32
33
34 35 36 37 38 39 40 41 42 43
Pingchan. Chow, E.. Tizard. R. and Pepinsky. R.B. (lY8Y) J. Biol. Chem. 264. 5768-5775. Kim, D.K., Kudo, 1. and Inoue, K. (19Y1) B&him. Biophys. Acta 1083. x0-88. Kim, D.K., Suh. P.G. and Ryu. S.H. (1991 f Biochcm. Biophys. Res. Commun. 174. IXY-IYh. Takayama. K., Kudo, I., Kim, D.K.. Nagata. K., Nozawa, Y. and Inoue. K. (1091) FEBS Lett. 282. 326-330. Clark. J.D.. Cin. L.L., Kriz, R.W.. Ramesha. C.S., Sultzman. CA., Lin, A.Y., Milonc. N. and Kuopf. J.L. (lYY1) tell 65. ffW?-1051. Lapetina. E.G. (19YO) FEBS Lztt. 268, 400-404. Nishizuka. Y. 11984) Nature 308, 6Y3-698. Taylor. S.J.. Chae, H.Z.. Rhee. S.G. and Exton, J.H. (19Yl) Nature 350, 516-518. Haslam. R.J. and McClenaghan, M.D. (1074) B&hem. J. 13X. 317-320. Haslam, R.J.. Solam, S.E., Fox, J.E.B.. Lynham, J.A. and Davidson, M.M.L. (IYXO) in Cellular Response Mechanism and theit Biological Significance (Rotman, A., Meyer, F.A.. Gilter. C. and Silberberg, A., eds.). pp. 213-231. J. Wiley and Sons Ltd. Takai. Y., Kaibuchi, K.. Matusbara. T. and Nishizuka, Y. (IYXI) B&hem. Biophys. Res. Commun. 101. 61-67. Nakashima. S.. Tohmatsu, T.. Hattori. H., Okano. Y. and Nozawa. f. 11986) Biochem. Biophys. Res. C(~rnmun. 135. 1090-I 104. Watson, S.P., Ruggiero. M.. Abraham% S.L. and hpetina. E.G. 11986) J. Biol. Chem. 261. 536X-5372. Crouch, M.F. and Lapetina. E.G. (19X8) Biochem. Biophys. Res. Commun. 15.3. ?I -30. Kienast, J., Arnout. J., Pfiegler, G.. Deckmyn, H. and Vermylen. J. (1987) Blood 69. 859-866. Gilman. A.G. 11984) Ceil 36, 577-579. Fuse, I. and Tai, H. (1987) Biochem. Biophys. Rea. Commun. 146, 659-66s. Silk, S.T., Clejan, S. and Witkom, K. (19XY) J. Biol. Chem. 264. 21466-21469. Deana. R., Ruzzenr, M., Doni. M.G.. Zoccarato. F. and Alexandre. A. (1089) Biochim. Biophys. Acta 1014. X3-206. Porcellati, S.. Costantini. V., Prosdocimi, M., Pistoiesi. R.. Porrovecchio, P.. Nenci, G.G. and Goracci. G. (1987) Thromb. Res. 47, 15-24. Porcellati, S.. Costantini. V.. Prosdocimi, M.. Stasi, M., Pistolesi. R.. Nenci, G.G. and Goracci, G. (1990) Agents and Action 2Y, 364-373. Kinlough-Rathbone, R.L.. Packham, M.A. and Mustard, J.F. (lY83) in Methods in Hemat~)l(lgy (Zimmerm~inn. T.S., cd.), pp, 64-91. Church-Livingstone, Edinburgh. Born, G.V.R. (1062) Nature (London) 194. 927-920. Saunders. R.D. and Horrocks. L.A. (lY84) Anal. Biochem. 143, 697-703. Nenci, G.G., Gresele, P., Agnelii, A. and Parise. P. (IYXI) Thromb. Haemost. 46. 572. Streh, H.. Heslop. J.P.. Irvine, R.F., Schulz, I. and Berridge, M.J. (1985) J. Biol. Chem. 260, 73OY-7315. Hofmann. S.L., Prescott, S.M. and Majerus. P.W. (lY82) Arch. Biochem. Biophys. 215. 237-244. Billah, M.M., Lapetina, E.G. and Cuatrecasas, P. (1979) Biochem. Biophys. Res. Commun. 90, 92-Y% Radomski, M.W., Palmer. R.M.J. and Moncada, S. (1%)) Proc. Nat! .4cad. Sci. USA 87. 5193-5197. Rendu. F.. L.ebret, M.. Tenza, D. and Levy Toledano, S. (19901 Biochem. J. 265, 343-349. Akiba, S., Sate, T. and Fujii, T. (1990) Biochim. Biophys. Acta 1044, 291-2Yh. Kajiyama, Y., Murayama, T. and Nomura. Y. (1989) Arch. Biochem. Biophys. 274, 200-208.
287 44 Brass, L.F., Laposata, M.. Banga, H.S. and Rittenhouse, S.E. (1986) J. Biol. Chem. 261, 16838816847. 45 Schacht, J. (1976) J. Neurochem. 27, 1119-1124. 46 Tysnes, O., Verhoeven, A.J.M. and Holmsen, H. (1987) Biochem. Biophys. Res. Commun. 144, 454-462. 47 Van der Meulen and Haslam, R.J. (1990) Biochem. J. 271, 693-700. 48 Tysnes, O.B.. Johanessen, E. and Steen, V.M. (1991) Biochem. J. 273, 241-243. 49 Colli, S., Maderna, P., Caruso, D., Stragliotto, E., Galli, G. and Tremoli, E. (19891 Eicosanoids 2, 51-57.
50 Porcellati, S., Gresele, P., Stasi, M., Costantini, V., Montanini, I., Nenci, G.G. and Goracci, G. (1989) 19th Meeting of the Federation of European Biochemical Society, July 2-7, Rome, Abst. TU178. 51 Whatley, R.E.. Fennel, D.F., Kurrus, J.A., Zimmerman, G.A., McIntyre. T.M. and Prescott, S.M. (1990) J. Biol. Chem. 265, 15550-15559. 52 Extort, J.H. FASEB J. (1988) 2. 2670-2676. 53 Simon, M.F., Chap, H. and Douste-Blazy. L. (19881 Biochem. Pharm. 37, 1263-1269.