Life Sciences, Vol. 49, pp. 29-37 Printed in the U.S.A.
Pergamon Press
PHOSPHOTYROSINE PHOSPHATASEACTIVITY IN HUMANPLATELETS Henry M. Smilowitz, L i l i Aramli, Dong Xu and Paul M. Epstein Department of Pharmacology University of Connecticut Health Center Farmington, CT 06030 (Received in final form May i, 1991)
Summary Using O-phosphotyrosine as a substrate, human platelets were shown to contain a highly active phosphotyrosine phosphatase (PTPase) activity. This a c t i v i t y was potently inhibited by vanadate, molybdate, and HgCl2. A b o u t 80% of the PTPase activity was particulate. When Triton-solubilized PTPase a c t i v i t y from whole platelets was applied to a DEAE Sephacel column about 40% came through unbound. The a c t i v i t y that bound was eluted by a NaCl gradient as a broad, heterogeneous peak. The possibility is raised for the existence of multiple forms of phosphotyrosine phosphatases in human platelets. That one or more of these forms may be regulated by activators of platelet aggregation and secretion, such as thrombin and collagen, is discussed. Protein tyrosine phosphorylation is recognized as important to the regulation of cell growth and proliferation (1). Protein tyrosine kinases are also found in differentiated non-proliferating cells (2) however, t h e i r physiological role is not yet clearly defined. Phosphotyrosine phosphatases are also becoming appreciated as important regulators of cell function (3,4) and disease states (5). It has been demonstrated, for example, in lymphocytes, that the CD45 PTPase can regulate the a c t i v i t y of the tyrosine kinase pp561ck (6) which is thought to play a role in several lymphocyte specific processes (7); in this case the notion is raised that the interaction of PTPase with cell surface receptors may modulate its function thereby regulating intracellular tyrosine phosphorylation of proteins (8). Platelets have h i g h levels of protein tyrosine kinase a c t i v i t y (9-II) suggesting that protein tyrosine phosphorylation and dephosphorylation may be important for platelet function. In intact platelets, activation with the physiological agonists thrombin and collagen stimulates tyrosine phosphorylation of several proteins (12-14). Similarly, sodium orthovanadate also stimulates an increase in tyrosine phosphorylation, secretion, and aggregation of platelets (15,16). These results raise the possibility that thrombin and collagen may activate protein-tyrosine kinase and/or inhibit PTPase via receptor-mediated mechanisms. Since PTPases could be the target of activators responsible for bringing about these physiological changes in platelets, we examined some of the properties of platelet PTPase a c t i v i t y and its sensitivity to modulation by agents known to affect PTPases in other
systems.
0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press Dlc
30
Platelet Tyrosine Phosphatase
Vol. 49, No. i, 1991
Methods Platelet Preparation. Bags of human platelets were purchased f r o m the American Red Cross, Farmington, CT. A bag of platelets was emptied into two plastic centrifuge tubes (= 30 ml/tube) and lO0 ~l of 40 mM s a l i c y l i c acid was added to each tube. After a 20 min incubation at room temperature, the tubes were centrifuged at 2500 rpm for 45 sec in a Sorvall RCSC centrifuge (SS34 rotor) to remove red blood c e l l s . This was repeated i f the f i r s t centrifugation did not remove a l l the red blood c e l l s . The platelets were then centrifuged at 2500 rpm for lO min at room temperature. The supernates were discarded and the platelets were gently resuspended (using plastic pipets) in 25 ml of buffer containing lO mM PIPES pH 6.5, 0.2 mM EGTA, 145 mM NaCl, 5 mM KCI, and 5.5 mM dextrose. The platelets were repelleted again at 2500 rpm, lO min, resuspended in 2 ml of the same buffer, aliquoted and frozen at -70°C. Protein concentrations of the packed, frozen p l a t e l e t pellets ranged from 28-45 mg/ml. Phosphotyrosine Phosphatase (PTPase) A c t i v i t y . PTPase a c t i v i t y was assayed by s l i g h t modification of the methods of Mustelin et a l . (6) and Leis and Kaplan (17). The basic assay was run in a 20 ~l volume in 1.5 ml microfuge tubes containing lO0 mM sodium acetate buffer pH 6.0, I0-20 ~g freshly thawed p l a t e l e t protein (protein based on the Bio-Rad Coomassie dye binding assay using bovine serum albumin as standard), lO mM O-phosphotyrosine as substrate, and 0.2% I r i t o n X-lO0. The reaction was i n i t i a t e d by the addition of the p l a t e l e t preparation, incubated at 37°C and terminated by the addition of 60 ~l 25% t r i c h l o r a c e t i c acid followed by 20 ~l of bovine serum albumin (lO mg/ml). Precipitated protein was removed by centrifugation a f t e r incubation on ice (microfuge, 3 min) and the supernatants were used for the measurement of liberated inorganic phosphate using the malachite green method (IB). Of the lO0 ~l of terminated reaction mixture, lO ~l was removed and phosphate content determined by mixing with 300 ~l of 1.2 N HCl and lO0 ~l of malachite green solution (3 volumes of 0.2% malachite green and l volume of I0% ammonium molybdate solution in 4 N HCI). After 5-I0 min, 200 ~l of each sample mixture was loaded into a 96 well dish and read in an ELISA plate reader (Titertek Multiskan P l u s MKll) at 595 nm. At 595 nm there is approximately 80% of the maximum absorption seen at 650 nm. Standard phosphate curves (0.05 nmoles to lO nmoles) were run with each experiment using KH2PO4. Materials. O-phosphotyrosine and d i t h i o t h r e i t o l ( D l l ) were from Boehringer Mannheim. Heparin sodium (lO0 U/mg) was from Nutritional Biochemical. NaF was f r o m J . l ~. Baker. MgCl2, MnCl2, ZnSO4, HgCl2, ammonium molybdate, sodium acetate, and sodium orthovanadate (Na3V04) w e r e f r o m Fisher. Protease i n h i b i t o r s , E D I A , spermine tetrahydrochloride, Triton X-lO0, s a l i c y l i c acid, trichloroacetic acid, iodoacetic acid, bovine serum albumin ( f a t t y acid free No. A-7030), and malachite green were from Sigma. Thrombin was obtained from Sigma (No. 1-9135) as a lyophilized powder from human plasma and was freshly reconstituted with d i s t i l l e d water just prior to use. Results and Discussion Using the PIPase assay method described, we could routinely detect the release of as l i t t l e as 50 pmoles of Pi from O-phosphotyrosine [ t y r ( P ) ] . Fig. l shows that using 20 ug of whole p l a t e l e t protein, the reaction is linear for at least 20 min. The specific a c t i v i t y of the PIPase for tyr(P) in whole platelets was approximately 150 nmol/min/mg protein. This is 2-3 fold lower than PIPase a c t i v i t y in human lymphocytes (6) and human astrocytoma cells (l?) which both have specific a c t i v i t i e s of approximately 400 nmol/min/mg protein when assayed under the same conditions.
Vol. 49, No. i, 1991
Platelet Tyrosine Phosphatase
31
4000 O~
E 3000 0
E c-
2000 O= ," 0.. 0
1000
r"
13,.
5
10
15
Time
20
(min)
FIG. l Time course of PTPase a c t i v i t y . PTPase a c t i v i t y was assayed as described in Methods using 20 ~g of whole platelet protein. The PTPase a c t i v i t y calculated from the slope of the curve is 157 nmole/min/mg protein. We tested a variety of agents shown to affect PTPase a c t i v i t y in other systems (19) for t h e i r effect on this a c t i v i t y in platelets. The results are shown in Fig. 2. Some PTPases have been shown to be dependent on or activated by Mg2+, Mn2+, or Ca2+ divalent cations for a c t i v i t y (2p~21) and others not (22,23). In the case of platelets, neither Mg2÷, Mn2 , nor Ca2÷ had any significant effect on the a c t i v i t y and EDTA was s l i g h t l y stimulatory. In an average of 5 different experiments, EDTA stimulated the a c t i v i t y by 19 ± 6% SE, Alkaline phosphatases have been shown, in some cases, to be capable of hydrolyzing phosphate From tyr(P) (19,24,25). However, s i n c e alkaline phosphatases are divalent metal ion-requiring enzymes that are potently inhibited by EDTA (19,24), these results indicate that the a c t i v i t y we are measuring in platelets is not alkaline phosphatase. Some PTPases have been shown to be completely inhibited by micromolar concentrations of Zn2+ (20,22,26,27) and others stimulated by i t (21). Micromolar concentrations of Zn2+ had almost no effect on platelet PTPase a c t i v i t y , but at higher concentrations, Zn2+ was a weak inhibitor with an ICso ~ 0.5 mM (Figs. 2 and 3). Sodium orthovanadate and ammonium molybdate are inhibitors of PTPases in most systems (21,22,25,28,29), and platelet PTPase was completely inhibited by both of these agents (Fig. 2). Compared with other PTPases, the platelet PTPase was extremely sensitive to inhibition by sodium orthovanadate, with an IC50 = 0.25 ~M (Fig. 3). By comparison, for example, sodium orthovanadate was 1200 times more potent as an i n h i b i t o r of human platelet PTPase than of human lymphocyte PTPase when assayed under the same conditions (6). NaF is a well known inhibitor of serine and threonine phosphatases (30). Most forms of PTPase are r e l a t i v e l y insensitive to NaF inhibition (17,22,23,25,27,31), although some forms of the enzyme, especially those associated with an acid phosphatase a c t i v i t y , are uncharacteristically sensitive to inhibition by NaF (19,25,32,33). Platelets contain PTPase a c t i v i t y that is r e l a t i v e l y insensitive to inhibition by NaF. At l mM, NaF inhibited p l a t e l e t PTPase by only 15% (Figs. 2 and 3) and at a concentration as high as lO0 mM NaF PTPase was inhibited by only about 40% (Fig. 3).
32
Platelet Tyrosine Phosphatase
Vol, 49, No. i, 1991
140 120
>,
".~ ~ 100 '~.
0 0
mr,.) I~
60
°~..~.,. ° 40
I.O.
20
}}
-
oz
z
o
FIG. 2 Effect of agents on PTPase a c t i v i t y . PTPasea c t i v i t y was measured as described in Methods using 12 ~g of whole platelet protein for 30 min at 37°C. Mean control a c t i v i t y was 141 nmole/min/mg protein. All agents were added just prior to the start of the enzyme assay, to a f i n a l concentration of l mM, except sodium tartrate, which was added to a f i n a l concentration of 5 mM. Results represent the mean ± the range of duplicate determinations. As with alkaline phosphatases, there are examples of acid phosphatases (phosphomonoesterases) which possess intrinsic PTPase a c t i v i t y (17,19,25,29,32,33). Many of these acid phosphatases are r e l a t i v e l y specific for the dephosphorylation of tyrosine residues. They have l i t t l e or no a c t i v i t y towards proteins phosphorylated on serine or threonine residues (17,1g,25,29,32), nor w i l l they dephosphorylate any phosphorylated amino acids other than tyr(P) (17,25). Several different tissues have been found to contain three d i s t i n c t forms of acid phosphatase: two high molecular weight forms (Mr's > 200,000 and = lO0,O00) which are primarily of lysosomal origin and characterized by s e n s i t i v i t y to inhibition by t a r t r a t e , and one low molecular w e i g h t f o r m (Mr : 15,000) which is primarily a soluble, cytoplasmic enzyme, resistant to t a r t r a t e inhibition (29,34). All of the tartrate-sensitive (32-34), and some of the tartrate-resistant, acid phosphatases, such as t h a t in bone (25), are also inhibited by NaF. Platelet PTPase was only p a r t i a l l y and very weakly inhibited by t a r t r a t e , with 5 mM t a r t r a t e producing only about 20% inhibition (Fig. 2). Since platelet PTPase is essentially insensitive to both NaF and tartrate, i t suggests that this platelet PTPase a c t i v i t y is d i s t i n c t from both the tartrate-sensitive high molecular weight acid phosphatases, as well as from the tartrate-resistant phosphatase isolated from bone (25). However, a low molecular weight acid phosphatase-associated PTPase has been isolated from bovine heart which shows some properties similar to the platelet enzyme. The bovine heart enzyme is inhibited by micromolar concentrations of molybdate and vanadate and insensitive to inhibition by t a r t r a t e , NaF, ZnCl2, MnCl2 and EDTA (29).
Vol. 49, No. 1, 1991
Platelet Tyrosine Phosphatase
33
100
NaF >, ,~ .>
80
, m
...,o
¢3
Pergamon Press
PHOSPHOTYROSINE PHOSPHATASEACTIVITY IN HUMANPLATELETS Henry M. Smilowitz, L i l i Aramli, Dong Xu and Paul M. Epstein Department of Pharmacology University of Connecticut Health Center Farmington, CT 06030 (Received in final form May i, 1991)
Summary Using O-phosphotyrosine as a substrate, human platelets were shown to contain a highly active phosphotyrosine phosphatase (PTPase) activity. This a c t i v i t y was potently inhibited by vanadate, molybdate, and HgCl2. A b o u t 80% of the PTPase activity was particulate. When Triton-solubilized PTPase a c t i v i t y from whole platelets was applied to a DEAE Sephacel column about 40% came through unbound. The a c t i v i t y that bound was eluted by a NaCl gradient as a broad, heterogeneous peak. The possibility is raised for the existence of multiple forms of phosphotyrosine phosphatases in human platelets. That one or more of these forms may be regulated by activators of platelet aggregation and secretion, such as thrombin and collagen, is discussed. Protein tyrosine phosphorylation is recognized as important to the regulation of cell growth and proliferation (1). Protein tyrosine kinases are also found in differentiated non-proliferating cells (2) however, t h e i r physiological role is not yet clearly defined. Phosphotyrosine phosphatases are also becoming appreciated as important regulators of cell function (3,4) and disease states (5). It has been demonstrated, for example, in lymphocytes, that the CD45 PTPase can regulate the a c t i v i t y of the tyrosine kinase pp561ck (6) which is thought to play a role in several lymphocyte specific processes (7); in this case the notion is raised that the interaction of PTPase with cell surface receptors may modulate its function thereby regulating intracellular tyrosine phosphorylation of proteins (8). Platelets have h i g h levels of protein tyrosine kinase a c t i v i t y (9-II) suggesting that protein tyrosine phosphorylation and dephosphorylation may be important for platelet function. In intact platelets, activation with the physiological agonists thrombin and collagen stimulates tyrosine phosphorylation of several proteins (12-14). Similarly, sodium orthovanadate also stimulates an increase in tyrosine phosphorylation, secretion, and aggregation of platelets (15,16). These results raise the possibility that thrombin and collagen may activate protein-tyrosine kinase and/or inhibit PTPase via receptor-mediated mechanisms. Since PTPases could be the target of activators responsible for bringing about these physiological changes in platelets, we examined some of the properties of platelet PTPase a c t i v i t y and its sensitivity to modulation by agents known to affect PTPases in other
systems.
0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press Dlc
30
Platelet Tyrosine Phosphatase
Vol. 49, No. i, 1991
Methods Platelet Preparation. Bags of human platelets were purchased f r o m the American Red Cross, Farmington, CT. A bag of platelets was emptied into two plastic centrifuge tubes (= 30 ml/tube) and lO0 ~l of 40 mM s a l i c y l i c acid was added to each tube. After a 20 min incubation at room temperature, the tubes were centrifuged at 2500 rpm for 45 sec in a Sorvall RCSC centrifuge (SS34 rotor) to remove red blood c e l l s . This was repeated i f the f i r s t centrifugation did not remove a l l the red blood c e l l s . The platelets were then centrifuged at 2500 rpm for lO min at room temperature. The supernates were discarded and the platelets were gently resuspended (using plastic pipets) in 25 ml of buffer containing lO mM PIPES pH 6.5, 0.2 mM EGTA, 145 mM NaCl, 5 mM KCI, and 5.5 mM dextrose. The platelets were repelleted again at 2500 rpm, lO min, resuspended in 2 ml of the same buffer, aliquoted and frozen at -70°C. Protein concentrations of the packed, frozen p l a t e l e t pellets ranged from 28-45 mg/ml. Phosphotyrosine Phosphatase (PTPase) A c t i v i t y . PTPase a c t i v i t y was assayed by s l i g h t modification of the methods of Mustelin et a l . (6) and Leis and Kaplan (17). The basic assay was run in a 20 ~l volume in 1.5 ml microfuge tubes containing lO0 mM sodium acetate buffer pH 6.0, I0-20 ~g freshly thawed p l a t e l e t protein (protein based on the Bio-Rad Coomassie dye binding assay using bovine serum albumin as standard), lO mM O-phosphotyrosine as substrate, and 0.2% I r i t o n X-lO0. The reaction was i n i t i a t e d by the addition of the p l a t e l e t preparation, incubated at 37°C and terminated by the addition of 60 ~l 25% t r i c h l o r a c e t i c acid followed by 20 ~l of bovine serum albumin (lO mg/ml). Precipitated protein was removed by centrifugation a f t e r incubation on ice (microfuge, 3 min) and the supernatants were used for the measurement of liberated inorganic phosphate using the malachite green method (IB). Of the lO0 ~l of terminated reaction mixture, lO ~l was removed and phosphate content determined by mixing with 300 ~l of 1.2 N HCl and lO0 ~l of malachite green solution (3 volumes of 0.2% malachite green and l volume of I0% ammonium molybdate solution in 4 N HCI). After 5-I0 min, 200 ~l of each sample mixture was loaded into a 96 well dish and read in an ELISA plate reader (Titertek Multiskan P l u s MKll) at 595 nm. At 595 nm there is approximately 80% of the maximum absorption seen at 650 nm. Standard phosphate curves (0.05 nmoles to lO nmoles) were run with each experiment using KH2PO4. Materials. O-phosphotyrosine and d i t h i o t h r e i t o l ( D l l ) were from Boehringer Mannheim. Heparin sodium (lO0 U/mg) was from Nutritional Biochemical. NaF was f r o m J . l ~. Baker. MgCl2, MnCl2, ZnSO4, HgCl2, ammonium molybdate, sodium acetate, and sodium orthovanadate (Na3V04) w e r e f r o m Fisher. Protease i n h i b i t o r s , E D I A , spermine tetrahydrochloride, Triton X-lO0, s a l i c y l i c acid, trichloroacetic acid, iodoacetic acid, bovine serum albumin ( f a t t y acid free No. A-7030), and malachite green were from Sigma. Thrombin was obtained from Sigma (No. 1-9135) as a lyophilized powder from human plasma and was freshly reconstituted with d i s t i l l e d water just prior to use. Results and Discussion Using the PIPase assay method described, we could routinely detect the release of as l i t t l e as 50 pmoles of Pi from O-phosphotyrosine [ t y r ( P ) ] . Fig. l shows that using 20 ug of whole p l a t e l e t protein, the reaction is linear for at least 20 min. The specific a c t i v i t y of the PIPase for tyr(P) in whole platelets was approximately 150 nmol/min/mg protein. This is 2-3 fold lower than PIPase a c t i v i t y in human lymphocytes (6) and human astrocytoma cells (l?) which both have specific a c t i v i t i e s of approximately 400 nmol/min/mg protein when assayed under the same conditions.
Vol. 49, No. i, 1991
Platelet Tyrosine Phosphatase
31
4000 O~
E 3000 0
E c-
2000 O= ," 0.. 0
1000
r"
13,.
5
10
15
Time
20
(min)
FIG. l Time course of PTPase a c t i v i t y . PTPase a c t i v i t y was assayed as described in Methods using 20 ~g of whole platelet protein. The PTPase a c t i v i t y calculated from the slope of the curve is 157 nmole/min/mg protein. We tested a variety of agents shown to affect PTPase a c t i v i t y in other systems (19) for t h e i r effect on this a c t i v i t y in platelets. The results are shown in Fig. 2. Some PTPases have been shown to be dependent on or activated by Mg2+, Mn2+, or Ca2+ divalent cations for a c t i v i t y (2p~21) and others not (22,23). In the case of platelets, neither Mg2÷, Mn2 , nor Ca2÷ had any significant effect on the a c t i v i t y and EDTA was s l i g h t l y stimulatory. In an average of 5 different experiments, EDTA stimulated the a c t i v i t y by 19 ± 6% SE, Alkaline phosphatases have been shown, in some cases, to be capable of hydrolyzing phosphate From tyr(P) (19,24,25). However, s i n c e alkaline phosphatases are divalent metal ion-requiring enzymes that are potently inhibited by EDTA (19,24), these results indicate that the a c t i v i t y we are measuring in platelets is not alkaline phosphatase. Some PTPases have been shown to be completely inhibited by micromolar concentrations of Zn2+ (20,22,26,27) and others stimulated by i t (21). Micromolar concentrations of Zn2+ had almost no effect on platelet PTPase a c t i v i t y , but at higher concentrations, Zn2+ was a weak inhibitor with an ICso ~ 0.5 mM (Figs. 2 and 3). Sodium orthovanadate and ammonium molybdate are inhibitors of PTPases in most systems (21,22,25,28,29), and platelet PTPase was completely inhibited by both of these agents (Fig. 2). Compared with other PTPases, the platelet PTPase was extremely sensitive to inhibition by sodium orthovanadate, with an IC50 = 0.25 ~M (Fig. 3). By comparison, for example, sodium orthovanadate was 1200 times more potent as an i n h i b i t o r of human platelet PTPase than of human lymphocyte PTPase when assayed under the same conditions (6). NaF is a well known inhibitor of serine and threonine phosphatases (30). Most forms of PTPase are r e l a t i v e l y insensitive to NaF inhibition (17,22,23,25,27,31), although some forms of the enzyme, especially those associated with an acid phosphatase a c t i v i t y , are uncharacteristically sensitive to inhibition by NaF (19,25,32,33). Platelets contain PTPase a c t i v i t y that is r e l a t i v e l y insensitive to inhibition by NaF. At l mM, NaF inhibited p l a t e l e t PTPase by only 15% (Figs. 2 and 3) and at a concentration as high as lO0 mM NaF PTPase was inhibited by only about 40% (Fig. 3).
32
Platelet Tyrosine Phosphatase
Vol, 49, No. i, 1991
140 120
>,
".~ ~ 100 '~.
0 0
mr,.) I~
60
°~..~.,. ° 40
I.O.
20
}}
-
oz
z
o
FIG. 2 Effect of agents on PTPase a c t i v i t y . PTPasea c t i v i t y was measured as described in Methods using 12 ~g of whole platelet protein for 30 min at 37°C. Mean control a c t i v i t y was 141 nmole/min/mg protein. All agents were added just prior to the start of the enzyme assay, to a f i n a l concentration of l mM, except sodium tartrate, which was added to a f i n a l concentration of 5 mM. Results represent the mean ± the range of duplicate determinations. As with alkaline phosphatases, there are examples of acid phosphatases (phosphomonoesterases) which possess intrinsic PTPase a c t i v i t y (17,19,25,29,32,33). Many of these acid phosphatases are r e l a t i v e l y specific for the dephosphorylation of tyrosine residues. They have l i t t l e or no a c t i v i t y towards proteins phosphorylated on serine or threonine residues (17,1g,25,29,32), nor w i l l they dephosphorylate any phosphorylated amino acids other than tyr(P) (17,25). Several different tissues have been found to contain three d i s t i n c t forms of acid phosphatase: two high molecular weight forms (Mr's > 200,000 and = lO0,O00) which are primarily of lysosomal origin and characterized by s e n s i t i v i t y to inhibition by t a r t r a t e , and one low molecular w e i g h t f o r m (Mr : 15,000) which is primarily a soluble, cytoplasmic enzyme, resistant to t a r t r a t e inhibition (29,34). All of the tartrate-sensitive (32-34), and some of the tartrate-resistant, acid phosphatases, such as t h a t in bone (25), are also inhibited by NaF. Platelet PTPase was only p a r t i a l l y and very weakly inhibited by t a r t r a t e , with 5 mM t a r t r a t e producing only about 20% inhibition (Fig. 2). Since platelet PTPase is essentially insensitive to both NaF and tartrate, i t suggests that this platelet PTPase a c t i v i t y is d i s t i n c t from both the tartrate-sensitive high molecular weight acid phosphatases, as well as from the tartrate-resistant phosphatase isolated from bone (25). However, a low molecular weight acid phosphatase-associated PTPase has been isolated from bovine heart which shows some properties similar to the platelet enzyme. The bovine heart enzyme is inhibited by micromolar concentrations of molybdate and vanadate and insensitive to inhibition by t a r t r a t e , NaF, ZnCl2, MnCl2 and EDTA (29).
Vol. 49, No. 1, 1991
Platelet Tyrosine Phosphatase
33
100
NaF >, ,~ .>
80
, m
...,o
¢3
