26
Biochimica et Biophysica Acta, 1054(1990) 26-32
Elsevier BBAMCR 12742
Stimulation of human platelets by collagen occurs by a N a ÷ / H ÷ exchanger independent mechanism Sunil Joseph 1, Winfried Siffert 2, Gertie Gorter 1 and Jan-Willem N. Akkerman 1 J Department of Haematology, University Hospital Utrecht, Utrecht (The Netherlands) and 2 Max Planck Institut fiir Biophysik, Frankfurt (F.R.G.)
(Received 18 January 1990)
Key words: Platelet activation; Sodium/proton exchanger; Protein kinase C; Collagen; Secretion
In stimulated human platelets dense-granule secretion in response to the 'weak agonists' ADP, adrenaline, platelet activating factor and low concentrations of thrombin as well as Ca 2 ÷ mobilisation in response to thrombin are enhanced by a N a ÷ / H + exchanger. In the present study the role of this antiport in collagen stimulated human platelets was examined. While stimulation of platelets loaded with the fluorescent intracellular pH-sensitive dye, bis-carboxyethyl-5(6)-carboxyfluorescein (BCECF) with thrombin resulted in the activation of the N a + / H + exchanger, activation of this antiport did not occur in collagen-stimulated platelets. The lack of antiport activity in response to collagen using BCECF-loaded platelets correlated with the lack of any functional role of the antiport in collagen stimulated platelets. In the presence of a N a t / H ÷ exchange inhibitor, ethylisopropylamiloride, neither collagen-induced platelet aggregation or dense-granule secretion was affected. Furthermore, while the removal of extracellular Na + (Naext), ÷ a condition that also prevents activation of the antiport, inhibited dense-granule secretion in response to a low concentration of thrombin, collagen-induced secretion was potentiated. This potentiatory effect could not be attributed to changes in either the membrane potential or in collagen-induced phospholipase C or protein kinase C activity. The present results indicate that in contrast to the 'weak agonists' (1) collagen-induced platelet activation does not require activation of the N a + / H + exchanger and (2) Nae+t per se is an inhibitor of collagen-induced secretion.
Introduction In common with most other cell types, human platelets maintain large transmembrane gradients of Na + and K t [1]. Na t in particular is involved in many steps in platelet activation since it enhances binding of adrenaline to a2-adrenergic receptors [2], modulates adenylate cyclase activity [3], alters intracellular Ca 2t ([Ca2+]i) transport processes [4] and affects the membrane potential [5,6]. In addition, Na t is involved in the regulation of platelet cytosolic p H (pHi) via the plasma membrane N a t / H + exchanger [7,8]. This antiport
Abbreviations: BCECF, bis-carboxyethyl-5-(6)-carboxyfluorescein; TxA2, thromboxane A2; BCECF-AM, 2'-7'-bis-carboxyethyl-5-(6)carboxyfluorescein tetraacetoxymethylester; fura-2/AM, fura-2 acetoxymethylester;EIPA, ethylisopropylamiloride;PRP, platelet-rich plasma; 5-HT, 5-hydroxytryptamine;PGI2, prostacyclin; NMDG +, N-methyl-D-glucamine;PKC, protein kinase C. Correspondence (present address): S. Joseph, Department of Clinical Pharmacology, Royal Postgraduate Medical School, Du Cane Road, London W12 OHS, U.K.
mediates the exchange of extracellular Na + (Naext) + for intracellular H t ([H t ] i ) and is important for the cell in two ways: (i) it stabilizes the p H i in resting cells through removal of intracellularly produced H t [reviewed in 9], and (ii) it raises p H i during cell stimulation which enhances different platelet activation processes [reviewed in 10]. For example, [Ca2+]i mobilisation in response to arachidonic acid and low doses of thrombin [10-12] and thromboxane A 2 (TxA2) formation, aggregation and secretion in response to the 'weak agonists', ADP, adrenaline and Platelet Activating Factor (PAF) [13-18] are both enhanced by an increase in p H i. At present little is known about the role of the N a t / H t exchanger in collagen-induced human platelet activation, an aspect addressed in the present study. Compared to the other platelet agonists, collagen is unique in that activation requires the adhesion of platelets to the fibres of collagen [19,20] and is accompanied by little [Ca2t]i mobilisation [21,22]. In addition, compared to the 'weak agonists' which are totally dependent upon endogenously formed TxA 2 to amplify platelet aggregation and secretion, collagen is only partially dependent upon TxA 2 [23,24]. On the other hand,
0167-4889/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
27 collagen is almost as potent as thrombin in triggering aggregation and secretion, both processes requiting considerable hydrolysis of metabolic ATP with concomitant production of protons [25,26]. Thus, it is also possible that in collagen-activated platelets the Na + / H + exchanger may play a role in both protecting the cytosol against acidification and in enhancing intracellular signal transduction mechanisms via its capacity to raise pH i . The present paper, however shows that in contrast to the 'weak agonists' and thrombin, collagen fails to activate the N a + / H + exchanger despite apparent normal activation of one of the transduction mechanisms that control this antiport. Materials and Methods
Materials 2'-7'-bis-carboxyethyl-5-(6)-carboxyfluorescein tetraacetoxymethylester (BCECF-AM), fura-2 acetoxymethylester (fura-2/AM) and the calcium ionophore, ionomycin were purchased from Calbiochem. Horm-collagen was purchased from Hormon-Chemie, F.R.G.. Bovine thrombin, indomethacin and phorbol-12-myristate 13-acetate (PMA) were purchased from Sigma. Ethylisopropylamiloride (EIPA) was a kind gift from Dr.T.Burckhardt, Max-Planck Institute fur Biophysik, Frankfurt, F.R.G.. Human fibrinogen was obtained from Kabi-Vitrum, Sweden. [14C]5-hydroxytryptamine (5-HT) and [32p]orthophosphate were purchased from Amersham International, U.K.. Prostacyclin (PGI2) was purchased from Upjohn, Kalamazoo, U.S.A.. Materials for gel electrophoresis were from Bio-Rad. All other reagents were purchased from Sigma. Methods Preparation of platelets. Venous blood (9 vol) was obtained by venepuncture from healthy volunteers who had denied taking any medication in the previous two weeks, into 3.2% trisodium citrate (1 vol). Platelet-rich plasma (PRP) was prepared by centrifugation of the blood at 200 x g for ten rain at room temperature. Washed platelets were prepared by acidification of the PRP with 0.11 M citric acid to pH 6.5, followed by the addition of PGI 2 (30 nM final concentration). The PRP was then centrifuged at 1500 × g for 15 min at room temperature. The resulting platelet pellet was resuspended in a pH 6.5 buffer composed of 36 mM citric acid, 103 mM NaC1, 5 mM KC1, 0.5 mM CaCI 2, 5 mM glucose plus 0.35% bovine serum albumin, 0.05 U / m l hirudin and 20 /~g/ml apyrase. For the [Ca2+]i and [14C]5HT secretion experiments, platelets were resuspended in the pH 6.5 buffer at a platelet count of 3 • 108/ml and incubated with respectively, 3/~M fura2 / A M (in the absence of albumin) and 0.1 /~Ci/ml [14C]5HT (57 Ci/mol) for 45 rain at 37 ° C. For the pHi experiments, platelets were resuspended in 2 ml of the
pH 6.5 buffer (minus albumin) and incubated at 37 °C with 6/xM BCECF-AM for 30 rain. For the [32p]phosphatidic acid/ protein phosphorylation experiments, platelets were resuspended in the pH 6.5 buffer at a platelet count of approx. 2- 109/ml and incubated with 0.25 mCi/ml carrier-free [32p]orthophosphate for 90 rain at 37 ° C. At the end of the respective incubation periods, 30 nM PGI 2 was added and the platelet suspensions centrifuged at 1500 × g for 15 min. The platelets were subsequently resuspended in a Hepes pH 7.4 buffer composed of 10 mM Hepes, 140 mM NaC1, 5 mM KC1, 1 mM MgS04, 1 mM CaC12, 5 mM Glucose plus 0.35% bovine serum albumin and 0.05 U / m l hirudin to a platelet count of (4-6). 109/ml for the [Ca2+]i and pH i experiments (minus albumin) and (5-6)- 109/ml for the aggregation and dense granule secretion experiments. [32p]-labelled platelets were resuspended to a platelet count of (8-9)- 109/ml in the absence of albumin. Measurement of platelet aggregation and dense-granule secretion. Platelets were diluted to a final platelet count of (2.5-3.0)-108/ml using the Hepes pH 7.4 buffer containing either Na +, K + or N-methyl-Dglucamine (NMDG+), in which K+or NMDG ÷ isotonically replaced Na ÷. The concentration of Nae+xt in either the K ÷ or NMDG+-buffers was lower than 7 mM: for the purpose of this study, these buffers are described as being essentially ' Naext-free. + ' In some experiments the platelets were pretreated with the cyclooxygenase inhibitor, indomethacin (10/~M) to prevent the actions of endogenously produced TxA 2. At the start of each incubation, fibrinogen (0.25 mg/ml) was added. Platelet aggregation was carried out at 37°C and monitored as an increase in light transmission in an aggregometer (Payton, Canada). Dense granule secretion was monitored by the secretion of the specific granule marker, [14C]5HT as described by Krishnamurthi et al. [24]. The concentration of agonist employed was either threshold to sub-maximal or maximal, defined with respect to the extent of [14C]5HT secretion. Measurement of [Ca 2 +] i and pHi. Fluorescence measurements were carried out at 37°C on a Hitachi F-3000 fluorescence spectrophotometer. The excitation and emission wavelengths were 495 nm and 530 nm, respectively for BCECF, and 340 nm and 500 nm, respectively for fura-2. Before the start of each measurement, the platelets were diluted to a final platelet count of (1.52.0)-108/ml using either the Na ÷' K + or NMDG ÷Hepes pH 7.4 buffers. After establishment of a stable baseline, the platelets were pretreated with indomethacin (10/~M) 1 min prior to vehicle or agonist addition. Calculation of the peak [Ca2+] i levels in nM was carried out as described by Pollock et al. [21]. Calibration of the BCECF fluorescence signal as a function of pHi was as described by Zavoico et al. [7]: the pH i results were expressed as the maximum change in pH i that occurred 2 rain after agonist or vehicle addition.
28
Measurement of phosphatidic acid formation and 47 kDa protein phosphorylation. 32p-labelled platelets were diluted to a final platelet count of 4. 108/ml using either the Na +, K ÷ or N M D G + - H e p e s p H 7.4 buffer. The platelets were then incubated at 37°C for 1 min prior to the addition of indomethacin (10 #M), followed 1 rain later by vehicle or agonist addition. Aliquots were removed at various time points and either processed for SDS polyacrylamide gel electrophoresis (10% gels) to determine the extent of phosphorylation of a 47 kDa protein or subjected to organic extraction and thin-layer chromatography for the determination of phosphatidic acid, as described previously [25,26]. Statistical analysis. Analysis of the data for statistical significance was carried out using the Student's t-test for un-paired data. Unless otherwise indicated, the results represent the means ± S.E. of 9-12 determinations from 3 separate experiments using platelets from different donors. Results
~~0.02)
(-0.02~+0.01)
1" T Veh
+0.05 I Changein pll~ (-0.02:~0.01) 0.00
? ? ~
Thr
Veh
T 1"
Coil
Veh
CDB
l min
I,
J
0.00 02) EIPA
(-0.02"~t.01)
EIPA Coil
Thr
.
(-0.03-+0.01)
EIPA CDB
Fig. 1. Effect of collagen and thrombin on platelet p H i. Indomethacin (10 pM)-pretreated BCECF-loaded platelets were incubated with either DMSO vehicle (Veh, 0.1% final volume) or 25 pM EIPA for 1 rain prior to thrombin (Thr, 0.5 U/ml), collagen (Coil, 20 /~g/ml) collagen dilution buffer (CDB, 2% final volume) addition. Resting pH i was pH 7.24+0.01. The figures in brackets beside each trace denote the maximum change in p H i 2 rain after CBD or agonist addition. The fluorescencetraces shown are representativeones from 3 separate experimentsusing platelets from different donors.
Effect of collagen on pH i Fig. 1 demonstrates that a maximal concentration of collagen (20 # g / m l ) failed to activate the N a + / H + exchanger as assessed using indomethacin (10 #M) pretreated BCECF-loaded platelets. While collagen caused an initial decrease in p H i , this decrease was due to the acidic nature of the collagen vehicle (Horm-collagen dilution buffer, CDB) which produced an essentially similar decrease in p H i ( P > 0.05 compared to collagen control). In contrast to collagen a maximal concentration of thrombin (0.5 U / m l ) increased p H i by 0.11 ± 0.02 p H units. The possibility that the CDB might be preventing or masking the operation of the N a + / H ÷ exchanger was discounted as the CDB had no significant ( P > 0.05) effect on the thrombin-induced p H i response (result not shown). In the presence of EIPA (25/~M), a specific inhibitor of the N a + / H ÷ exchanger [7,9], the thrombin-induced increase in pH i was prevented and pronounced acidification occurred. N o such EIPA mediated acidification occurred following collagen stimulation when compared to the collagen vehicle, which would have been expected if collagen had activated the antiport.
Effect of EIPA and the removal of Na+x, on collagen-induced platelet aggregation and dense-granule secretion. The failure of collagen to activate the N a + / H + exchanger was in-directly confirmed in the following experiments. Previous studies have shown that activation of the N a + / H + exchanger enhances 'TxA2-dependent'-platelet aggregation and secretion in response to low concentrations of thrombin and the 'weak agonists' [14-18]. Fig. 2 demonstrates that like the 'weak
agonists', a low concentration of collagen (5 g g / m l ) similarly induces 'TXAE-dependent'- platelet aggregation and secretion as indomethacin inhibited both responses. However both collagen-induced platelet responses were not inhibited by 5-25 # M EIPA (results with 25 /xM EIPA Fig. 2), which would have been expected if the antiport had been activated. Similarly, inhibiting the antiport by suspending platelets in media in which Nae+xt was isotonically replaced by Ke+xt or
Veh
Coil
Indo $$
$$
Coil
EIPA Coil $$ (7_+1)*
l~min
~
(65+3)
(67±3)
Fig. 2. Effect of EIPA on collagen-inducedplatelet aggregation and [14C]5HTsecretion.Plateletswereincubated with either DMSO vehicle (Veh), 10 pM indomethacin (Indo) or 25 pM EIPA for 1 rain prior to 5/~g/ml collagen (Coil) addition. The traces shown are typical of 3 similar experiments using platelets from 3 different donors. The figures beside each trace represent the % [14C]5HTsecreted into the platelet supernatant 3 rain after collagen addition. EIPA alone released less than 5% [Z4C]5HT. * P
Biochimica et Biophysica Acta, 1054(1990) 26-32
Elsevier BBAMCR 12742
Stimulation of human platelets by collagen occurs by a N a ÷ / H ÷ exchanger independent mechanism Sunil Joseph 1, Winfried Siffert 2, Gertie Gorter 1 and Jan-Willem N. Akkerman 1 J Department of Haematology, University Hospital Utrecht, Utrecht (The Netherlands) and 2 Max Planck Institut fiir Biophysik, Frankfurt (F.R.G.)
(Received 18 January 1990)
Key words: Platelet activation; Sodium/proton exchanger; Protein kinase C; Collagen; Secretion
In stimulated human platelets dense-granule secretion in response to the 'weak agonists' ADP, adrenaline, platelet activating factor and low concentrations of thrombin as well as Ca 2 ÷ mobilisation in response to thrombin are enhanced by a N a ÷ / H + exchanger. In the present study the role of this antiport in collagen stimulated human platelets was examined. While stimulation of platelets loaded with the fluorescent intracellular pH-sensitive dye, bis-carboxyethyl-5(6)-carboxyfluorescein (BCECF) with thrombin resulted in the activation of the N a + / H + exchanger, activation of this antiport did not occur in collagen-stimulated platelets. The lack of antiport activity in response to collagen using BCECF-loaded platelets correlated with the lack of any functional role of the antiport in collagen stimulated platelets. In the presence of a N a t / H ÷ exchange inhibitor, ethylisopropylamiloride, neither collagen-induced platelet aggregation or dense-granule secretion was affected. Furthermore, while the removal of extracellular Na + (Naext), ÷ a condition that also prevents activation of the antiport, inhibited dense-granule secretion in response to a low concentration of thrombin, collagen-induced secretion was potentiated. This potentiatory effect could not be attributed to changes in either the membrane potential or in collagen-induced phospholipase C or protein kinase C activity. The present results indicate that in contrast to the 'weak agonists' (1) collagen-induced platelet activation does not require activation of the N a + / H + exchanger and (2) Nae+t per se is an inhibitor of collagen-induced secretion.
Introduction In common with most other cell types, human platelets maintain large transmembrane gradients of Na + and K t [1]. Na t in particular is involved in many steps in platelet activation since it enhances binding of adrenaline to a2-adrenergic receptors [2], modulates adenylate cyclase activity [3], alters intracellular Ca 2t ([Ca2+]i) transport processes [4] and affects the membrane potential [5,6]. In addition, Na t is involved in the regulation of platelet cytosolic p H (pHi) via the plasma membrane N a t / H + exchanger [7,8]. This antiport
Abbreviations: BCECF, bis-carboxyethyl-5-(6)-carboxyfluorescein; TxA2, thromboxane A2; BCECF-AM, 2'-7'-bis-carboxyethyl-5-(6)carboxyfluorescein tetraacetoxymethylester; fura-2/AM, fura-2 acetoxymethylester;EIPA, ethylisopropylamiloride;PRP, platelet-rich plasma; 5-HT, 5-hydroxytryptamine;PGI2, prostacyclin; NMDG +, N-methyl-D-glucamine;PKC, protein kinase C. Correspondence (present address): S. Joseph, Department of Clinical Pharmacology, Royal Postgraduate Medical School, Du Cane Road, London W12 OHS, U.K.
mediates the exchange of extracellular Na + (Naext) + for intracellular H t ([H t ] i ) and is important for the cell in two ways: (i) it stabilizes the p H i in resting cells through removal of intracellularly produced H t [reviewed in 9], and (ii) it raises p H i during cell stimulation which enhances different platelet activation processes [reviewed in 10]. For example, [Ca2+]i mobilisation in response to arachidonic acid and low doses of thrombin [10-12] and thromboxane A 2 (TxA2) formation, aggregation and secretion in response to the 'weak agonists', ADP, adrenaline and Platelet Activating Factor (PAF) [13-18] are both enhanced by an increase in p H i. At present little is known about the role of the N a t / H t exchanger in collagen-induced human platelet activation, an aspect addressed in the present study. Compared to the other platelet agonists, collagen is unique in that activation requires the adhesion of platelets to the fibres of collagen [19,20] and is accompanied by little [Ca2t]i mobilisation [21,22]. In addition, compared to the 'weak agonists' which are totally dependent upon endogenously formed TxA 2 to amplify platelet aggregation and secretion, collagen is only partially dependent upon TxA 2 [23,24]. On the other hand,
0167-4889/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
27 collagen is almost as potent as thrombin in triggering aggregation and secretion, both processes requiting considerable hydrolysis of metabolic ATP with concomitant production of protons [25,26]. Thus, it is also possible that in collagen-activated platelets the Na + / H + exchanger may play a role in both protecting the cytosol against acidification and in enhancing intracellular signal transduction mechanisms via its capacity to raise pH i . The present paper, however shows that in contrast to the 'weak agonists' and thrombin, collagen fails to activate the N a + / H + exchanger despite apparent normal activation of one of the transduction mechanisms that control this antiport. Materials and Methods
Materials 2'-7'-bis-carboxyethyl-5-(6)-carboxyfluorescein tetraacetoxymethylester (BCECF-AM), fura-2 acetoxymethylester (fura-2/AM) and the calcium ionophore, ionomycin were purchased from Calbiochem. Horm-collagen was purchased from Hormon-Chemie, F.R.G.. Bovine thrombin, indomethacin and phorbol-12-myristate 13-acetate (PMA) were purchased from Sigma. Ethylisopropylamiloride (EIPA) was a kind gift from Dr.T.Burckhardt, Max-Planck Institute fur Biophysik, Frankfurt, F.R.G.. Human fibrinogen was obtained from Kabi-Vitrum, Sweden. [14C]5-hydroxytryptamine (5-HT) and [32p]orthophosphate were purchased from Amersham International, U.K.. Prostacyclin (PGI2) was purchased from Upjohn, Kalamazoo, U.S.A.. Materials for gel electrophoresis were from Bio-Rad. All other reagents were purchased from Sigma. Methods Preparation of platelets. Venous blood (9 vol) was obtained by venepuncture from healthy volunteers who had denied taking any medication in the previous two weeks, into 3.2% trisodium citrate (1 vol). Platelet-rich plasma (PRP) was prepared by centrifugation of the blood at 200 x g for ten rain at room temperature. Washed platelets were prepared by acidification of the PRP with 0.11 M citric acid to pH 6.5, followed by the addition of PGI 2 (30 nM final concentration). The PRP was then centrifuged at 1500 × g for 15 min at room temperature. The resulting platelet pellet was resuspended in a pH 6.5 buffer composed of 36 mM citric acid, 103 mM NaC1, 5 mM KC1, 0.5 mM CaCI 2, 5 mM glucose plus 0.35% bovine serum albumin, 0.05 U / m l hirudin and 20 /~g/ml apyrase. For the [Ca2+]i and [14C]5HT secretion experiments, platelets were resuspended in the pH 6.5 buffer at a platelet count of 3 • 108/ml and incubated with respectively, 3/~M fura2 / A M (in the absence of albumin) and 0.1 /~Ci/ml [14C]5HT (57 Ci/mol) for 45 rain at 37 ° C. For the pHi experiments, platelets were resuspended in 2 ml of the
pH 6.5 buffer (minus albumin) and incubated at 37 °C with 6/xM BCECF-AM for 30 rain. For the [32p]phosphatidic acid/ protein phosphorylation experiments, platelets were resuspended in the pH 6.5 buffer at a platelet count of approx. 2- 109/ml and incubated with 0.25 mCi/ml carrier-free [32p]orthophosphate for 90 rain at 37 ° C. At the end of the respective incubation periods, 30 nM PGI 2 was added and the platelet suspensions centrifuged at 1500 × g for 15 min. The platelets were subsequently resuspended in a Hepes pH 7.4 buffer composed of 10 mM Hepes, 140 mM NaC1, 5 mM KC1, 1 mM MgS04, 1 mM CaC12, 5 mM Glucose plus 0.35% bovine serum albumin and 0.05 U / m l hirudin to a platelet count of (4-6). 109/ml for the [Ca2+]i and pH i experiments (minus albumin) and (5-6)- 109/ml for the aggregation and dense granule secretion experiments. [32p]-labelled platelets were resuspended to a platelet count of (8-9)- 109/ml in the absence of albumin. Measurement of platelet aggregation and dense-granule secretion. Platelets were diluted to a final platelet count of (2.5-3.0)-108/ml using the Hepes pH 7.4 buffer containing either Na +, K + or N-methyl-Dglucamine (NMDG+), in which K+or NMDG ÷ isotonically replaced Na ÷. The concentration of Nae+xt in either the K ÷ or NMDG+-buffers was lower than 7 mM: for the purpose of this study, these buffers are described as being essentially ' Naext-free. + ' In some experiments the platelets were pretreated with the cyclooxygenase inhibitor, indomethacin (10/~M) to prevent the actions of endogenously produced TxA 2. At the start of each incubation, fibrinogen (0.25 mg/ml) was added. Platelet aggregation was carried out at 37°C and monitored as an increase in light transmission in an aggregometer (Payton, Canada). Dense granule secretion was monitored by the secretion of the specific granule marker, [14C]5HT as described by Krishnamurthi et al. [24]. The concentration of agonist employed was either threshold to sub-maximal or maximal, defined with respect to the extent of [14C]5HT secretion. Measurement of [Ca 2 +] i and pHi. Fluorescence measurements were carried out at 37°C on a Hitachi F-3000 fluorescence spectrophotometer. The excitation and emission wavelengths were 495 nm and 530 nm, respectively for BCECF, and 340 nm and 500 nm, respectively for fura-2. Before the start of each measurement, the platelets were diluted to a final platelet count of (1.52.0)-108/ml using either the Na ÷' K + or NMDG ÷Hepes pH 7.4 buffers. After establishment of a stable baseline, the platelets were pretreated with indomethacin (10/~M) 1 min prior to vehicle or agonist addition. Calculation of the peak [Ca2+] i levels in nM was carried out as described by Pollock et al. [21]. Calibration of the BCECF fluorescence signal as a function of pHi was as described by Zavoico et al. [7]: the pH i results were expressed as the maximum change in pH i that occurred 2 rain after agonist or vehicle addition.
28
Measurement of phosphatidic acid formation and 47 kDa protein phosphorylation. 32p-labelled platelets were diluted to a final platelet count of 4. 108/ml using either the Na +, K ÷ or N M D G + - H e p e s p H 7.4 buffer. The platelets were then incubated at 37°C for 1 min prior to the addition of indomethacin (10 #M), followed 1 rain later by vehicle or agonist addition. Aliquots were removed at various time points and either processed for SDS polyacrylamide gel electrophoresis (10% gels) to determine the extent of phosphorylation of a 47 kDa protein or subjected to organic extraction and thin-layer chromatography for the determination of phosphatidic acid, as described previously [25,26]. Statistical analysis. Analysis of the data for statistical significance was carried out using the Student's t-test for un-paired data. Unless otherwise indicated, the results represent the means ± S.E. of 9-12 determinations from 3 separate experiments using platelets from different donors. Results
~~0.02)
(-0.02~+0.01)
1" T Veh
+0.05 I Changein pll~ (-0.02:~0.01) 0.00
? ? ~
Thr
Veh
T 1"
Coil
Veh
CDB
l min
I,
J
0.00 02) EIPA
(-0.02"~t.01)
EIPA Coil
Thr
.
(-0.03-+0.01)
EIPA CDB
Fig. 1. Effect of collagen and thrombin on platelet p H i. Indomethacin (10 pM)-pretreated BCECF-loaded platelets were incubated with either DMSO vehicle (Veh, 0.1% final volume) or 25 pM EIPA for 1 rain prior to thrombin (Thr, 0.5 U/ml), collagen (Coil, 20 /~g/ml) collagen dilution buffer (CDB, 2% final volume) addition. Resting pH i was pH 7.24+0.01. The figures in brackets beside each trace denote the maximum change in p H i 2 rain after CBD or agonist addition. The fluorescencetraces shown are representativeones from 3 separate experimentsusing platelets from different donors.
Effect of collagen on pH i Fig. 1 demonstrates that a maximal concentration of collagen (20 # g / m l ) failed to activate the N a + / H + exchanger as assessed using indomethacin (10 #M) pretreated BCECF-loaded platelets. While collagen caused an initial decrease in p H i , this decrease was due to the acidic nature of the collagen vehicle (Horm-collagen dilution buffer, CDB) which produced an essentially similar decrease in p H i ( P > 0.05 compared to collagen control). In contrast to collagen a maximal concentration of thrombin (0.5 U / m l ) increased p H i by 0.11 ± 0.02 p H units. The possibility that the CDB might be preventing or masking the operation of the N a + / H ÷ exchanger was discounted as the CDB had no significant ( P > 0.05) effect on the thrombin-induced p H i response (result not shown). In the presence of EIPA (25/~M), a specific inhibitor of the N a + / H ÷ exchanger [7,9], the thrombin-induced increase in pH i was prevented and pronounced acidification occurred. N o such EIPA mediated acidification occurred following collagen stimulation when compared to the collagen vehicle, which would have been expected if collagen had activated the antiport.
Effect of EIPA and the removal of Na+x, on collagen-induced platelet aggregation and dense-granule secretion. The failure of collagen to activate the N a + / H + exchanger was in-directly confirmed in the following experiments. Previous studies have shown that activation of the N a + / H + exchanger enhances 'TxA2-dependent'-platelet aggregation and secretion in response to low concentrations of thrombin and the 'weak agonists' [14-18]. Fig. 2 demonstrates that like the 'weak
agonists', a low concentration of collagen (5 g g / m l ) similarly induces 'TXAE-dependent'- platelet aggregation and secretion as indomethacin inhibited both responses. However both collagen-induced platelet responses were not inhibited by 5-25 # M EIPA (results with 25 /xM EIPA Fig. 2), which would have been expected if the antiport had been activated. Similarly, inhibiting the antiport by suspending platelets in media in which Nae+xt was isotonically replaced by Ke+xt or
Veh
Coil
Indo $$
$$
Coil
EIPA Coil $$ (7_+1)*
l~min
~
(65+3)
(67±3)
Fig. 2. Effect of EIPA on collagen-inducedplatelet aggregation and [14C]5HTsecretion.Plateletswereincubated with either DMSO vehicle (Veh), 10 pM indomethacin (Indo) or 25 pM EIPA for 1 rain prior to 5/~g/ml collagen (Coil) addition. The traces shown are typical of 3 similar experiments using platelets from 3 different donors. The figures beside each trace represent the % [14C]5HTsecreted into the platelet supernatant 3 rain after collagen addition. EIPA alone released less than 5% [Z4C]5HT. * P
