.5 a). 1.424 I. . :T’l": o‘éf ’ “Ir“x‘v fun-.33 5. " :38.” '11.: 62.5: .«r‘ ' ’ .J’. - I Yaw" ‘33., 0.;40 7713.33: . . ’ L3,“, 3,, ’39." 1‘93, 3'13,;. Jih‘“ I, 1.15:: ’ ‘ "3' "' “(£533. 4 if.) 5,, 3 " 5,3,3 3439:}; Air“, uh . WW. 1'3: 3 3' 0‘4 , ‘1 ~35; 3' 'L,..n'/L._ _":1 . ‘ ’_‘ 1:3‘?‘ .'3U' :“1‘:: A -133. . 3 - . .- _':“.:>~ . . 3 3-,: 133,; 31'”: n4- ""' ifiytr.nl‘f. ,Ifiiil' "(if H.- .3 3. “W xO-m ‘ " {Hag-7‘3 3' s' :3,» not. 3.3,. a /»II "193: S “'7' (WI! 15:15}; 4:. y; . 1“? {$5 3r ‘3' ., r; 3‘3}; I”, ,5 3: 3., S ‘ 3 3 -| . f f , f -l .2 '5 3 5 -€ , A 3' '39:“ ‘ P‘ - rn- " .5351: ,r:3 " r k'fin‘ou ‘iYIH 3 . - v mfg-'5; ' “N _ .3 u ‘ » 2!?"4 “9.7:, 414%; ~-u-' .4..- uwz‘r‘; 4 ~L::‘ $27.7. ....,-'. —' JJ'AT'“_UV~-‘}1r‘-‘ .3 ' .cifi' -: ‘ ”mi :54}! GAN STATE mllllmmu lulllllllilllll 1293 00901 9385 This is to certify that the dissertation entitled Basset Hound Hereditary Thrombopathy: Measurement of Cytosolic Calcium, pH and 1,2-Diacylglycerol Production presented by Jennifer Sue Thomas has been accepted towards fulfillment of the requirements for Ph . D . degree in Pa tho lo gy $333M ajor professor Date 2/19/91 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before due due. DATE DUE DATE DUE DATE DUE MSU Is An Affirmative Action/Equal Opportunity Institution ammo-9.1 .‘i‘f 4 BASSET ROUND HEREBITARY THROMBOPATHY: MEASUREMENT OF PLATELET CYTOSOLIC CALCIUM, pH AND 1,2-DIACYLCLYCEROL PRODUCTION By Jennifer Sue Thomas A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pathology 1991 ABSTRACT nassar nouns uznznrrAnx THROHBOPATHY: MEASUREMENT or PLATELET CYTOSOLIC CALCIUM, pH AND 1,2-DIACYLGLICEROL rnonucrrou By Jennifer Sue Thomas Basset hound hereditary thrombopathy is an autosomal recessive defect characterized by hemorrhagic diathesis. Affected platelets have ‘normal quantities of membrane GPIIb-IIIa and bind fibrinogen following .ADP stimulation. Affected platelets undergo shape change but do not aggregate in response to agonists which depend on phospholipase C and thromboxane production for full platelet activation (ie. ADP, PAP and‘ calcium ionophore A23187). They aggregate in response to agonists which bypass phospholipase C (ie. high concentrations of thrombin and phorbol ester). Secretion of dense granule ATP occurs irrespective of aggregation. Following stimulation, affected platelets produce thromboxane; however, the production is not under normal regulatory control. Affected platelets yield normal phosphorylation of the 20kDa and 47kDa proteins and decreased phosphorylation of a 64-67kDa protein of unknown function and identity. The aim of this dissertation was the further investigation of events in platelet activation and their relationship to Basset hound hereditary thrombopathy. First, it was found that affected platelets had normal resting [Ca++]1. Using aequorin-loaded platelets, there were no consistent differences between control and affected post-stimulation [Ca'H’]i. Using fura 2-loaded platelets, there were no consistent differences between control and affected post-stimulation [Ca++]i in the presence of external Ca++. In the absence of external Ca++, affected platelets had decreased post- stimulation [Ca++]1 in response to PAF, thrombin and ionomycin. Secondly, affected platelets had normal resting pHi. In response to agonists which do not induce aggregation in affected platelets (ie. ADP, PAF, A23187 and thromboxane mimetic U466l9), there was decreased post- stimulation pHi in affected platelets relative to control platelets. Following stimulation with thrombin, there was no difference in pHi between affected and control platelets. Finally, 1,2-diacylglycerol levels were normal in affected platelets at rest or following stimulation with ADP, PAF, A23187, thrombin or phorbol ester. In response to U466l9/epinephrine, affected platelets had increased 1,2- diacylglycerol production relative to control platelets. Although these findings help localize the platelet defect in Basset hound hereditary thrombopathy, they cannot explain the disorder based on current understanding of normal platelet function. Copyright by JENNIFER SUE THOMAS 1991 ACKNOWLEDGEMENTS I would like to express my sincere thanks to the members of my graduate guidance committee, Drs. Thomas C. Bell, Douglas W. Estry, Janver D. Krehbiel, Mary F. McConnell and William S. Spielman. Special thanks are due to Dr. Bell for his commitment, advise and support. I am deeply indebted to Dr. McConnell for her help and patience in the laboratory as well as her careful editing of this manuscript. I would also like to thank my parents, Beatrice H. and Benjamin D. Thomas, for their loving support. They have always encouraged me to strive to be the best I can be, no matter where that may take me. Finally, I wish to thank my sisters, Lorrie A. Schartow, Gale L. Ceccato and Patricia A. Curtis, for their faith, understanding and encouragement . TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . xii LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . xv LIST OF ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . xviii INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . 1 CHAPTER I: LITERATURE REVIEW . . . . . . . . . . . . . .'. . . . 7 PLATELET PRODUCTION . . . . . . . . . . . . . . . . . . . . . 9 Megakaryocytopoiesis . . . . . . . . . . . . . . . . . . . 9 Regulation of Megakaryocytopoiesis and Platelet Production 10 PLATELET STRUCTURE . . . . . . . . . . . . . . . . . . . . . . 12 Mitochondria and Glycogen . . . . . . . . . . . . . . . . . 12 Granules . . . . . . . . . . . . . . . . . . . . . . . . . 13 Cytoskeleton . . . . . . . . . . . . . . . . . . . . . . . 15 Dense Tubular System . . . . . . . . . . . . . . . . . . . 18 Open Canalicular System . . . . . . . . . . . . . . . . . . 19 Plasma Membrane . . . . . . . . . . . . . . . . . . . . . . 19 GPIaIIa . . . . . . . . . . . . . . . . . . . . . . . . 21 GPIb . . . . . . . . . . . . . . . . . . . . . . . . . . 21 GPIc-IIa . . . . . . . . . . . . . . . . . . . . . . . . 23 GPIIb-IIIa . . . . . . . . . . . . . . . . . . . . . . . 23 GPIV . . . . . . . . . . . . . . . . . . . . . . . . . . 27 vi vii GPV GPIX . GMP-IAO PTAl . THE PLATELET RESPONSE Adhesion Shape Change Aggregation . The Release Reaction Secretion of Granules The Release of Arachidonate Metabolites SIGNAL TRANSDUCTION OR PROCESSING Adenylate Cyclase and cAMP Production . Phospholipase C and Membrane Phosphoinositide Hydrolysis Phosphoinositide Metabolism Phospholipase C Mediated Hydrolysis of PI G protein Regulation of PLC Inositol Phosphates: Metabolism and Role as Second Messengers . Production of Diacylglycerol and Its Role as a Second Messenger . . . . . Role of Phosphatidic Acid and Lysophosphatidic Acid as Second Messengers . . Ca++ AND ITS ROLE IN SIGNAL TRANSDUCTION . Platelet Storage Sites for Ca++ Non-Exchangeable Ca++ Pools Exchangeable Ca++ Pools Maintenance of Resting [Ca++]1 27 27 28 28 29 29 31 33 36 37 39 AZ 43 46 47 49 50 52 SS 57 58 58 58 59 60 viii Ca++ Influx Ca++ Efflux Ca++ Sequestration . Agonist Stimulated Ca++ Fluxes CaTT-Dependent Processes During Platelet Activation . CaTT/Calmodulin Dependent Protein Kinase . Calcium Dependent Protease . Role of Ca++ in the Cytoskeleton . THE ROLE OF PROTEIN KINASE C IN PLATELET ACTIVATION Protein Kinase C Diacylglycerol and Phorbol Activation of PKC Phosphorylation of Proteins by PKG The Stimulatory Effects of PKC in Platelet Activation . The Inhibitory Effects of PKC on Platelet Activation THE ROLE or NaT/H" EXCHANGE IN CELL ACTIVATION . Characteristics of NaT/H+ Exchange Na+/H+ Exchange and Platelet Activation . NaT/H+ Exchange and Ca++ Mobilization NaT/H+ Exchange and Arachidonic Acid Metabolism Summary of NaT/H+ Exchange . SUMMARY OF THE EFFECTS OF INDIVIDUAL PLATELET AGONISTS . Adenosine Diphosphate (ADP) . Epinephrine . Platelet-Activating Factor (PAF) Thrombin Calcium Ionophore A23187 and Ionomycin Endoperoxide/Thromboxane Mimetic U46619 . 61 61 62 63 67 67 68 70 71 71 71 73 75 76 79 79 81 82 8h 86 87 88 91 94 '96 98 100 ix Phorbol . BASSET HOUND HEREDITARY THROMBOPATHY . CHAPTER 2: RESTING AND STIMULATED PLATELET CALCIUM INTRODUCTION . MATERIALS AND METHODS Experimental subjects . Reagents Platelet Collection . Measurement of [Ca++]1 Using Fura 2 . Measurement of [Ca++]i Using Aequorin . Statistical Analysis RESULTS Resting [Ca++]1 . Adenosine Diphosphate (ADP) Platelet-Activating Factor (PAF) Thrombin Ionomycin . Phorbol Myristate Acetate (PMA) DISCUSSION . CHAPTER 3: RESTING AND POST-STIMULATION CYTOSOLIC pH INTRODUCTION . . . . MATERIALS AND METHODS Experimental Subjects . Reagents Measurement of pHi Using BCECF 102 103 112 112 117 117 118 119 .120 122 123 124 126 128 131 135 138 142 146 168 168 175 175 175 176 Statistical Analysis . . . . . . . . . . . . . . . . . . 178 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . 178 Resting pHi . . . . . . . . . . . . . . . . . . . . . . . 179 Buffer . . . . . . . . . . . . . . . . . . . . . . . . . 180 ADP . . . . . . . . . . . . . . . . . . . . . . . . . 182 Platelet-Activating Factor (PAF) . . . . . . . . . . . . 185 Thrombin .'. . . . . . . . . . . . . . . . . . . . . . . 188 Ionomycin . . . . . . . . . . . . . . . . . . . . . . . . 190 U46619/Epinephrine . . . . . . . . . . . . . . . . . . . 192 Dye Leakage Rates . . . . . . . . . . . . . . . . . . . . 196 DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . 196 CHAPTER 4: 1,2-DIACYLGLYCEROL PRODUCTION . . . . . . . . . . . 219 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . 219 MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . 222 Experimental Subjects . . . . . . . . . . . . . . . . . . 222 Reagents . . . . . . . . . . . . . . . . . . . . . . . . 219 Preparation of [3H]Arachidonic Acid Labelled Platelets . 223 Separation of Lipids by Thin Layer Chromatography . . . . 225 Statistical Analysis . . . . . . . . . . . . . . . . . . 226 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . 227 Lipid TLC Results . . . . . . . . . . . . . . . . . . . . 228 Resting Values . . . . . . . . . . . . . . . . . . . . . . 228 Buffer . . . . . . . . . . . . . . . . . . . . . . . . . 230 Adenosine Diphosphate (ADP) . . . . . . . . . . . . . . . 232 Platelet Activating Factor (PAF) . . . . . . . . . . . . 235 Thrombin . . . . . . . . . . . . . . . . . . . . . . . . 238 xi Calcium Ionophore A23187 046619/Epinephrine Phorbol Myristate Acetate (PMA) DISCUSSION . CHAPTER 5: SUMMARY AND CONCLUSIONS LIST OF REFERENCES 239 241 242 245 260 267 Table 10 11 12 13 LIST OF TABLES Page Resting [Ca++]1 (uM) in control and affected aequorin- loaded platelets . . . 126 Resting [Ca++]i (nM) in control and affected fura 2- loaded platelets . . 127 Peak [Ca++]1 (uM) following stimulation with 10uM ADP in control and affected aequorin-loaded platelets . . . 128 Peak [Ca++]1 following stimulation with ADP in control and affected fura 2-loaded platelets . 130 Peak [Ca++]i (uM) following stimulation with PAP in control and affected aequorin-loaded platelets . . . . . . 132 Peak [Ca++]i in response to stimulation with PAP in control and affected fura-2 loaded platelets . 134 Peak [Ca++]1 (uM) following addition of thrombin in control and affected aequorin-loaded platelets . . . . . . 13S Peak [Ca++]1 (nM) following thrombin stimulation in control and affected fura 2-1oaded platelets . . 137 Peak [Ca++]1 (uM) following addition of ionomycin in control and affected aequorin- -1oaded platelets . . . . . . . 139 Peak [Ca++]1 (nM) in response to ionomycin in control and affected fura 2-loaded platelets . 141 Peak [Ca++]1 (uM) in response to 3uM PMA in control and affected aequorin-loaded platelets . . . . . 144 Peak [Ca++]1 in response to 3uM PMA in control and affected fura 2-loaded platelets 145 Summary of resting and post-stimulation [Ca++]1 results in control and affected platelets loaded with aequorin . 161 xii 14 15 16 17 18 19 20 21 22 23 24 25 26 27 xiii Summary of resting and post-stimulation [Ca++]i in control and affected platelets loaded with fura 2 Resting pH1 of control and affected platelets in non- stirred and stirred BCECF-loaded platelets . The change in pHi from resting levels in control and affected BCECF-loaded platelets following the addition of buffer . . . . . . . . . . . . . . . . . . . The change in pHi relative to resting values in control and affected BCECF-loaded platelets following the addition of lOuM ADP in the presence of external Ca and fibrinogen . The change in pfli relative to resting levels in control and affected BCECF-loaded platelets following stimulation with PAP . . . . . . . . . . . . . . . . . . . . The change in pHi relative to resting levels in control and affected BCECF-loaded platelets following stimulation with thrombin . . . . . . . . . . . . . . . . The change in pHi relative to resting levels in control and affected BCECF-loaded platelets stimulated with luM ionomycin . . . . . . . . . . The phi change relative to resting levels in control and affected BCECF-loaded platelets following stimulation with U46619 and epinephrine . . . . . . . . . Summary of the change in pHi from resting levels in control and affected BCECF-loaded platelets Comparison of values from the resting platelet samples of control and affected 38- AA- labelled platelets Measurement of baseline 1,2-DG following the addition of buffer alone to control and affected 3H-AA-la‘belled platelets Comparison of 1,2-DG production in control and affected 3H-AA-labelled platelets following the addition of lOuM ADP. Ratio of post-stimulation CPM : resting CPM Comparison of 1,2-DG production in control and affected 3H-AA-labelled platelets following stimulation with luM PAF. Ratio of post-stimulation CPM: resting CPM . Comparison of 1,2-DG production in control and affected 3H-AA-la‘belled platelets following stimulation with 0.01uM PAF. Ratio of post-stimulation CPM: resting CPM . . . 162 180 181 183 186 189 191 194 215 230 231 232 235 237 28 29 30 31 xiv Comparison of 1,2-DG production in control and affected 3H-AA-labelled platelets following stimulation with 0.3U/m1 thrombin. Ratio of stimulated CPM: resting CPM . . Comparison of 1,2-DG production in control and affected 3H-AA-la'belled platelets following stimulation with luM calcium ionophore A23187. Ratio of 1,2—DG CPM following agonist stimulation : 1,2-DG CPM following addition of buffer only . . . . . . . . . . . . . . . Comparison of 1,2-DG production in control and affected H-AA-labelled platelets following stimulation with 2.5uM U46619/luM epinephrine. Ratio 1,2-DG CPM following agonist-stimulation: baseline 1,2-DC CPM following addition of buffer . . . . . . . . . . . . Comparison of 1,2-DG production in control and affected 3H-AA-labelled platelets following the addition of 3uM PMA. Ratio of post-stimulation 1,2-DG CPM : resting 1,2-DG CPM . . . . . . . . . . . . 238 240 242 243 LIST OF FIGURES Figure Page 1 Representative luminescence tracings for the response of aequorin-loaded affected and control platelets to lOuM ADP in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . 129 2 Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to lOuM ADP in the presence of external Ca++. Arrow - agonist addition . 131 3 Representative luminescence tracings of aequorin-loaded affected and control platelets in response to luM PAF in the presence of external Ca++. Arrow - agonist addition . 133 4 Representative fluorescence tracings of fura 2-1oaded control and affected platelets in response to luM PAF in the presence of external Ca++. Arrow - agonist addition . 135 5 Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 0.25U/m1 thrombin in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . 136 6 Representative fluorescence tracings of fura 2-loaded affected and control platelets in response to 0.25U/ml thrombin in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . 138 7 Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 2uM ionomycin in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . . . . . . . 140 8 Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to SuM ionomycin in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . . . . . . . 142 9 Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 3uM PMA in the presence of external Ca++. Arrow - addition of agonist . . . . . . . . . . . . . . . . . . . . . . . . . 143 10 11 12 13 14 15 16 17 18 19 20 xvi Representative fluorescence tracings of fura 2-1oaded control and affected platelets in response to 3uM PMA in the presence of external Ca++. Arrow - addition of agonist Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to the addition of buffer in the presence of external Ca++ and fibrinogen. Arrow — agonist addition . Representative fluorescence tracings of BCECF-loaded control and affected platelets following+the addition of lOuM ADP in the presence of external Ca Arrow - agonist addition . . . . . . . . . Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to luM PAF in the presence of external Ca++. Arrow - agonist addition . Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to 0.1U/m1 thrombin in the presence of external Ca++. Arrow - agonist addition . Representative fluorescence tracings of BCECF-loaded control and affected BCECF-loaded platelets in response to luM ionomycin in the presence of external Ca++. Arrow - agonist addition . Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to luM U46619/luM epinephrine in the presence of external Ca++ and fibrinogen. Arrow - agonist addition . . . Representative TLC plate. Lane 1 - oleate; lane 2- monoglyceride; lane 3 - cholesterol; lane 4 - 1,3-diolein; lane 5 - 85% 1,3-diolein, 15% 1,2-diolein; lane 6 - platelet sample + diolein mix; lane 7 - platelet sample without added standard; lane 8 - triolein; lane 9 - cholesterol oleate; lane 10 - phosphatidylcholine Comparison of the baseline 1,2-DG (CPM) following the addition of buffer in control and affected 3H-AA-labelled platelet samples. Values represent mean 1 SD Comparison of 1,2-DG production in response to lOuM ADP in control and affected 3H-AA-labelled platelet samples. Ratio of post-stimulation CPM : resting CPM. Values represent mean 1 SD . . . . . . . . . . 1,2-DG production in response to lOuM ADP and buffer in 3H-AA-labe11ed affected platelet. Values represent mean i SD . . . . . . . . . . . . . . . . and fibrinogen. 145 182 184 187 189 192 195 229 231 233 234 21 22 23 24 25 26 27 28 29 30. xvii 1,2-DG production in response to lOuM ADP and buffer in H-AA-labelled control platelets. Values represent mean i SD Comparison of 1, 2- DG production in response to luM PAF 8- AA- labelled control and affected platelets. Values represent mean + SD. * - significantly different from control, p < 0.05 fimparison of 1, 2- DC production in response to 0.01uM PAF 8- AA- labelled control and affected platelets. Values represent mean + SD. * - significantly different from control, p < 0.05 . Comparison of 1,2-DG production in response to 0.3U/m1 thrombin in 3H-AA-labelled control and affected platelets. Values represent mean 1 SD. * - significantly different from control, p < 0.05 . . . . . fimparison of 1,2-DG production in response to luM A23187 H- arachidonic acid labelled control and affected platelets. Values represent mean i SD . Comparison of 1,2-DG production in response to U46619/epinephrine in 3H-AA-labelled control and affected platelets. Values represent mean i SD. * - significantly different from control, p < 0.05 . Comparison of 1,2-DG production in response to 3uM PMA in H-AA-labelled control and affected platelets. Values represent mean 1 SD Comparison of 1,2-DG production in response to 3uM PMA and buffer in affected 3H-AA-labelled platelets. Values represent mean i SD . . . . . . . . Comparison of 1,2-DG production in response to 3uM PMA and buffer in 3H-AA-labelled control platelets. Values represent mean i SD . . . . . . . . Schematic model for platelet activation 234 236 237 239 240 243 244 244 245 266 ADP . BFU-M BSS . [Cd++]1 cAMP cAMP- PDE CDP- DG CPU-M CPM DMSO DTS EDTA EGTA FSBA HETE LIST OF ABBREVIATIONS . . adenosine diphosphate burst forming unit— -megakaryocyte Bernard Soulier syndrome . concentration of cytosolic free ionized calcium cyclic adenosine monophosphate . cAMP phosphodiesterase . cytidine diphosphate-diacy1g1ycerol . colony forming unit- -megakaryocyte . counts per minute . diacylglycerol dimethyl sulfoxide . dense tubular system . ethylene diamine tetraacetate ethylene glycol tetraacetic acid 5'-p fluorosulfonylbenzoyl adenosine . glycoprotein hydroxyeicosatetraenoate inositol diphosphate inositol trisphosphate inositol tetrakisphosphate . . . . lysophosphatidic acid . megakaryocyte- colony stimulating activity . megakaryocyte- colony stimulating factor . . . myosin light chain kinase . open canalicular system . . phosphatidic acid platelet activating factor prostaglandin E1 prostaglandin 62 . prostaglandin H2 ..phospyaosdyltnp81261 . phosphatidylinositol 4- -phosphate . phosphatidylinositol 4, 5- -bisphosphate . . protein kinase C phospholipase A2 . . . phospholipase C . phorbol myristate acetate platelet poor plasma . . platelet rich plasma . thrombocytosis- stimulating factor . . . . . . . . . thromboxane A2 . . thromboxane 82 . vonWillebrand's disease vonWillebrand's factor xviii INTRODUCTION Platelets play an important part in the maintenance of normal hemostasis. In the past few decades, a great deal of research has focused on platelet function and biochemistry due to the potential role of platelets in the pathogenesis of atherosclerosis, thrombosis and other cardiovascular diseases; major causes of illness and death in Western cultures. Platelets are also frequently studied as a model for cellular physiology and biochemistry because they are readily available and have signal transduction pathways linked to identifiable surface receptors. Some of the current knowledge about platelet function and biochemistry has resulted from studies on humans or animals with hereditary or congenital platelet disorders. By identifying an enzyme, glycoprotein or other structure which is missing or non-functional in affected platelets, it is possible to come to a better understanding of the role played by that component in normal platelet function. In the 1970's, a veterinarian identified such a disorder in a group of related purebred Basset hounds which has been termed Basset hound hereditary . thrombopathy (BHT). Examination of the pedigrees of affected dogs suggest that BHT has an autosomal recessive pattern of inheritance. BHT is characterized by hemorrhagic diathesis which is associated with markedly defective in vitro aggregation. l 2 Platelets from dogs affected with BHT share many abnormal clinical and functional characteristics with platelets from humans affected with Glanzmann's thrombasthenia, leading to the initial hypothesis that BHT was an animal model for this human disorder.300 Glanzmann's thrombasthenia is a hereditary platelet dysfunction characterized by abnormal glycoprotein IIb-IIIa (GPIIb-IIIa) content or function. GPIIb- IIIa is a surface membrane glycoprotein complex that acts as the platelet fibrinogen receptor. It is the binding of fibrinogen to two adjacent platelets which mediates platelet aggregation. Platelets from dogs affected with BHT, however, were demonstrated to have normal GPIIb- IIIa content using two dimensional gel electrophoresis.299 Following stimulation, affected platelets were also found to bind 125I-la‘belled298 and gold-labelled fibrinogen119 to the same extent as normal canine platelets, indicating that the GPIIb-IIIa complex is functional and that the defect must lie in some, as yet unidentified, post-fibrinogen binding event(s). Because BHT platelets do not aggregate in spite of normal fibrinogen binding, it was hypothesized that affected platelets have a defect in signal transduction. Signal transduction is the process by which an external signal is transmitted into the cell to elicit an appropriate physiologic response. As a general rule, signal transduction involves the following sequence of events: (1) binding of an external stimulus (agonist) to a specific surface receptor; (2) regulation of the activity of an effector pathway(s) by the activated receptor; (3) alteration of the production or function of intracellular second messengers by the effector system(s); and (4) mediation by second 3 messengers of the activity, structure or phosphorylation of cellular proteins. The final result is the modulation of the cellular response. In an attempt to further characterize the disorder and to localize the biochemical defect in BHT, an extensive examination of affected platelet aggregation and ATP dense granule secretion was performed.254 For this study, as well as all additional studies, a wide range of agonists were used in order to activate different effector pathways. It was found that affected platelets only aggregate in response to those agonists which cause activation independent of phospholipase C activation and thromboxane production. In affected platelets, secretion of dense granules does not occur in response to all agonists; however, it occurs independently of the ability to aggregate and is dependent on or concurrent with thromboxane production. In studies concurrent to this dissertation research, the resting and post-stimulation phosphorylation of affected and control platelet proteins was measured to assess the generation of intracellular second messengers.255 It was found that both affected and control platelets have similar phosphorylation patterns of the major, identified cytosolic proteins. Unexpectedly, it was discovered that affected platelets do not phosphorylate a 64-67kDa protein to the same extent as control platelets. This protein has not been previously described.and its functional significance is unknown.255 Finally, in additional studies, thromboxane production was measured in affected and control platelets.256 Thromboxane is produced 4 from membrane arachidonic acid following platelet activation and acts as a positive feedback stimulus to further enhance platelet activation. It was found that affected platelets have increased production of thromboxane in response to those agonists which stimulate secretion and decreased production of thromboxane in response to those agonists which do not stimulate secretion. This indicates that the pathway for thromboxane production is intact in affected platelets but not under proper regulatory control. It also suggests that secretion in affected platelets is strongly dependent on thromboxane production and that aggregation is not directly correlated to thromboxane production. The aim of the current dissertation is to further explore the biochemistry of signal transduction in platelets from affected and control canine platelets in order to identify and/or characterize a defective pathway(s) or enzyme(s). At the time of the initiation of this dissertation research, the studies on GPIIb-IIIa,299 fibrinogen binding298'119 and aggregation/secretion254 in BHT had been completed and the study on protein phosphorylation was underway in the laboratory.255 Thus, it was determined that measurement of resting and post-stimulation cytosolic free ionized calcium ([Ca++]1) was the next important event to be investigated due to the central role of Ca++ as a second messenger in platelet activation. Intracellular Ca++ is integral in determining the activity of a number of effector enzymes and the organization of cytoskeletal proteins. It was anticipated that there would be decreased Ca++ mobilization in affected platelets in response to those agonists which do not cause platelet aggregation. 5 Following the conclusion of the calcium studies for this dissertation, the concurrent data on protein phosphorylation and the preliminary results on the thromboxane production in affected platelets became available. Therefore, it was determined that the measurement of platelet cytosolic pHi fluxes was an appropriate area of investigation to best integrate available data. Increased cytosolic pHi frequently accompanies platelet activation and is associated with activation of phospholipase A2 (the enzyme responsible for arachidonic acid release and subsequent thromboxane production), release of internal Ca++ and organization of the platelet cytoskeleton. In addition, the adrenergic receptor is a 64kDa which is closely associated with, if not actually part of, the Na+/H+ antiport which is involved in the regulation of pHi. If the adrenergic receptor were the same as the poorly phosphorylated 64-67kDa protein in affected platelets, then abnormal alkalinization responses would be anticipated in activated BHT platelets. Finally, because of a suspected role in the mediation of abnormal responses in BHT, the measurement of phospholipase C activity was judged to be essential for the final portion of the dissertation research. Previous aggregation studieszsa had shown that aggregation and complete platelet activation occur in BHT when phospholipase C is bypassed, indicating that phospholipase C was not being normally activated in response to those agonists unable to cause aggregation. In contrast, the data on protein phosphorylation, Ca++ mobilization and thromboxane production suggest that the second messengers generated by phospholipase C were being produced. In order to assess phospholipase C activity, the production of 1,2-diacy1glycerol was measured. 1,2-Diacylglycerol is 6 one of the second messengers generated by the phospholipase C mediated hydrolysis of membrane phosphoinositides. In summary, this study supports the hypothesis that BHT affected platelets have a defect in several biochemical pathways and it details signal transduction defects that relate to cell dysfunctions. CHAPTER 1: LITERATURE REVIEW Platelets occupy a vital role in the maintenance of normal hemostasis in concert with the coagulation system and the vascular wall. In response to a break in the integrity of the vessel wall, platelets form a primary hemostatic plug by adhering to the subendothelial tissue and secondarily recruiting additional platelets. In addition to forming the primary plug, platelets play an important part in the coagulation cascade by providing a phospholipid surface to accelerate the activation of several coagulation factors which convert circulating fibrinogen to insoluble fibrin. The fibrin then enmeshes and stabilizes the primary p1u8.184,403 Platelets respond to a wide variety of both inhibitory and stimulatory substances. Platelet plasma membrane receptors bind to specific external stimuli and subsequently transmit the signal into the cell. This culminates in a distinct pattern of physiologic reactions commonly termed the platelet response. These reactions can be elicited both in_21;;g and in vivg and include: (1) adhesion of platelets to a foreign surface; (2) shape change from a smooth disc to a sphere with numerous pseudopodial projections; (3) aggregation by the formation of platelet-to-platelet bridges; and (4) the release reaction characterized by the secretion of the contents of cytoplasmic granules and the production of arachidonate metabolites.212'5’77'403J‘78 7 8 Although platelets respond to a complex variety of agonists and antagonists, only a limited number of effector pathways and second messengers are utilized to produce a cellular reaction.212 None of the physiologic agonists can penetrate the plasma membrane of intact platelets.478 Instead, exogenous stimuli bind to a specific receptor(s) on the platelet surface. The activated receptor transmits this signal into the interior of the cell by regulating the activity of specific effector pathways such as enzyme systems or membrane ion channels. These effector systems alter intracellular second messengers which then mediate the activity, phosphorylation or structure of intracellular proteins. This alteration in the cellular protein triggers the physiologic response. The process by which an external stimulus is translated into the cellular response is termed signal transduction or processing.212'377'403 Amplification or inhibition of the signal can occur at any step due to the multiple points of interaction or communication between the different arms of the signal transduction pathway. Activation of platelets causes the production and/or release of substances which act as positive or negative feedback to respectively amplify or inhibit the platelet response.9-103'212 Although a large amount of information about platelet activation and function has been published in the past few decades, there are still many unanswered questions. Most of the published information has been limited to studies in platelets from humans, rats or rabbits. The material presented in this literature review will attempt to summarize the pertinent findings. It is assumed that canine platelets act in a similar manner to that published; however, relatively little has been 9 published on canine platelets. Known deviations in the activity or behavior of canine platelets from human platelets will be described and discussed. PLATELET PRODUCTION The average circulating platelet number in normal human blood is 248,000 platelets/ul, with a range of 148,000—348,000/ul.461 The normal circulating platelet number in dog blood is similar with an average of 300,000 platelets/ul and a range of 200,000 to 500,000/ul.183 Human platelets in peripheral blood have a mean survival time of six to ten days403 while dog platelets have a shorter survival time of only four to seven days.182 Platelets are removed from the circulation by consumption in the hemostatic process, removal by the reticuloendothelial system as part of the normal aging process or destruction as part of a pathologic immune-mediated disorder. In a healthy animal, the number of circulating platelets is relatively constant and is maintained by the production and release of platelets from the precursor megakaryocytes.182 Megakaryocytopoiesis Megakaryocytopoiesis is a developmental cycle involving at least five stages: (1) the commitment of a pluripotential stem cells to megakaryocyte differentiation; (2) megakaryocyte mitotic activity to expand cell numbers; (3) variable numbers of endomitotic divisions to increase nuclear number; (4) cytoplasmic growth with the acquisition of 10 organelles and platelet specific proteins; and (5) formation of demarcation membranes within the megakaryocyte cytoplasm to release platelets into the circulation.253'374 Committed stem cells form two functionally identifiable classes of megakaryocyte precursor cells: (1) the burst forming unit-megakaryocyte (BFU-M); and (2) the colony forming unit-megakaryocyte (CFU-M). BFU-M are differentiated by producing multiple clusters of megakaryocytes which contain larger numbers of megakaryocytes than the CFU-M. The next developmental stage is a small, mononuclear cell which cannot be identified morphologically but which can be identified by the expression of phenotypic platelet specific markers. These cells are a transition cell between the progenitor cell and the megakaryocyte.144 Megakaryocytes are morphologically identifiable in the bone marrow by their large size and their lobulated nuclei. The DNA content of the 'megakaryocytes varies, with most cells having ploidy values from 8 to 64N. The size and number of platelets produced from each cell is dependent on the nuclear polyploidization of the megakaryocyte.253 Regulation of Megakaryocytopoiesis and Platelet Production The control of megakaryocytopoiesis and platelet production can be divided into two distinct phases: (1) a proliferative, mitotic phase which involves the BFU-M and the CFU-M and is responsible for maintaining megakaryocyte numbers; and (2) a non-proliferative phase stage which involves the non-mitotic development of megakaryocytes and culminates in the release of platelets.253.374 11 Mitotic division of the megakaryocyte precursors is not directly influenced by circulating platelet numbers; instead, this phase is regulated by, and serves to maintain, megakaryocyte and progenitor cell numbers.253 CFU-M requires the presence of a factor(s) with megakaryocyte colony stimulating activity (M-CSA) to promote the formation of colonies of megakaryocytes in vitro. Whether this activity is due to the presence of a single megakaryocyte colony stimulating factor (M-CSF) or to a collection of factors with M-CSA has not been determined.144 Factors affect megakaryocytopoiesis by having M-CSA themselves or by promoting CFU-M proliferation in concert with other substances. These factors include granulocyte-macrophage colony stimulating factor, interleukin 3, interleukin 1, interleukin 6, erythropoietin and thrombopoietin.144 The non-proliferative phase of megakaryocyte development, on the other hand, is responsive to circulating platelet numbers. The development is characterized by the appearance of cytoplasmic organelles, the mitotic division of the nucleus without division of the cytoplasm (endomitosis), the appearance of membrane antigens and platelet release. Thrombocytopenic animals contain a substance in their urine and plasma, termed thrombocytosis-stimu1ating factor (TSF) or thrombopoietin. TSF promotes megakaryocyte development in at least three ways: (1) stimulation of the commitment of progenitor cells to the megakaryocyte line; (2) stimulation of megakaryocyte endomitotic divisions to increase cell size and ploidy and (3) shortening of the megakaryocyte maturation time. Whether TSF is a single cytokine or a 12 collection of cytokines has not been determined. The site of production of thrombopoietin is not known.u‘l‘»182 PLATELET STRUCTURE Canine and human platelets are of a similar volume (7.6-8.3 f1)184 and, in the resting state, they circulate as discoid cells which are 2-3 um in diameter.403 Platelets are anucleate and contain only a small amount of DNA in the mitochondria.“03 They have limited ability to synthesize proteins.169 On a dry weight basis, platelets contain approximately 50% protein, 8.5% carbohydrate and 41.5% lipid and other constituents.184 Microscopic examination of Wright's stained blood smears reveals granular organelles which are dispersed throughout the cytoplasm. Electron microscopic examination of platelets defines the unit (plasma) membrane, the open canalicular system, the dense tubular system, mitochondria, glycogen stores, three types of secretory granules and cytoskeletal elements.452 Mitochondria and Glycogen Resting platelets do not require much energy to maintain their anabolic processes, but stimulated or activated platelets expend large amounts of catabolic energy. ATP is essential to maintain platelet activation with the different platelet responses varying in their energy requirement. The catabolic ATP demands, in order of increasing need, are shape change, aggregation, alpha— and dense-granule secretion, lysosomal granule secretion and arachidonic acid liberation.171 13 Platelets have large stores of glycogen which can be utilized for ATP production.169'171 Platelets contain relatively few mitochondria and their number and size are smaller than the mitochondria present in muscle cells.169'266 Although oxidative metabolism produces approximately 80% of the total ATP needed in the resting state, platelets have a well-developed glycolytic system which can fully compensate for any increase in energy demand associated with activation. Simultaneous inhibition of oxidative phosphorylation and glycolysis or glucose starvation of cells severely inhibits platelet responses. Granules The platelet cytosol contains three identifiable types of secretory granules: (1) alpha granules; (2) dense granules; and (3) lysosomal granules. Alpha granules are the most numerous of all the granules. On ultrastructure they appear as membrane bound round or oval structures with a moderately electron-dense matrix.181"266 They contain a variety of proteins which function in coagulation, inflammation and wound healing. These proteins can be divided into three groups: (1) proteins specific to platelets (platelet factor 4, platelet derived growth factor, 9 -thromboglobulin);. (2) proteins which function in the coagulation cascade (fibrinogen, factor V, factor VIII/von Willebrand's factor); and (3) proteins not specific to platelets which have a variety of functions (thrombospondin, fibronectin, albumin, a z-plasmin l4 inhibitor, histidine-rich glycoprotein, high-molecular weight kininogen, ¢ 'anti-trypsin, a 2-macroglobulin) .403 Dense granules are less numerous than alpha granules and appear as round to oval structures with an electron dense core surrounded by an electron-lucent area. They contain primarily adenine nucleotides (ATP and ADP), amines (serotonin, histamine and catecholamines) and bivalent cations (Ca++ and Mg""*’).18“'266 All species have relatively high levels of ADP, ATP and serotonin in their dense granules while the quantity of divalent cations varies. Human platelet dense granules contain mainly Ca++ while canine platelet dense granules contain relatively equal quantities of Ca++ and Mg++.267 Within the granules, Ca++, inorganic pyrophosphate, ATP and ADP form insoluble high molecular complexes. Except for serotonin, which is freely exchangeable, the contents of the dense granules are almost totally unexchangeable with their cytoplasmic counterparts.403 Lysosomal granules morphologically resemble alpha granules. They contain four types of acid hydrolases: (1) acid proteases (endo- and exopeptidases); (2) acid glycosidases (p -hexosaminidase, p - glucoronidase, a -mannosidase and galactosidase); (3) acid phosphatase; and (4) aryl sulfatases.403 The release of lysosomal granules requires a strong stimulus and only partial release occurs. The function of these granules is unknown. Platelets phagocytize extracellular material; however, the formation of a phagolysosome and subsequent destruction of the phagocytized material does not occur.480 Instead, the lysosomes release their contents into the external media where they 15 may play a role in the clearing of thrombi, the inactivation and degradation of heparin and the inflammatory response.169»403 Cytoskeleton The platelet cytoskeleton consists of four components: (1) the membrane skeleton; (2) the microtubular coil; (3) actin and myosin filaments; and (4) a variety of associated proteins.130 The cytoskeleton maintains the normal platelet resting discoid shape and participates in the formation of a contractile system necessary for shape change, pseudopod formation, contraction and granule secretion.184 The membrane skeleton is located directly under the plasma membrane and is composed of a network of actin filaments. It is a submembranous lining that stabilizes the plasma membrane and helps maintain its shape.131 The membrane skeleton pool of actin is separate from the cytoplasmic pool that is involved in contraction. Unlike the cytoplasmic pool, the membrane skeleton actin is resistant to Ca++- induced depolymerization. Actin binding protein links the membrane skeleton with the membrane glycoproteins (CP) Ib-IX, Ia, IIa, and a glycoprotein of Mr 250,000 of unknown function.128'129 The attachment of the membrane skeleton to CPIb-IX determines its location within the membrane, thereby regulating the ability of GPIb-IX to bind to von Willebrand's factor following injury to the vessel wall. Because of its attachment to both CPIb (one of the thrombin receptors) and GPIa (a proposed collagen receptor), it has been proposed that the membrane skeleton also plays a role in signal transduction.131 During platelet 16 aggregation, activation of the intracellular Ca++-dependent protease (calpain) causes the hydrolysis of actin binding protein. This releases the membrane skeleton from the membrane glycoproteins and permits reorganization of the cytoskeleton.132 The second component of the cytoskeleton, the microtubular coil, consists of a single strand of polymerized tubulin coiled upon itself.30 The microtubule coil is located under the membrane skeleton and helps maintain the resting discoid shape.452 During platelet activation, the platelet organelles are compressed into the cell center and surrounded by a constricting rim of microtubules and microfilaments. Some studies indicate that the microtubule coil dissolves upon platelet activation and reassembles in its new location,406 while others suggest that microtubule disassembly-reassembly is not necessary for platelet activation.“55 In support of the latter statement, taxol treatment of platelets, which stabilizes microtubules and prevent their disassembly, does not inhibit platelet shape change or the centralization of organelles.“52 The third component of the cytoskeleton is the actin and myosin filaments. Actin comprises approximately 15-20% of the platelet protein.130 It is divided into two distinct pools, one associated with the membrane cytoskeleton as already discussed and the other associated with the platelet cytosol. In the resting state, 40-50% of the total actin is present in the polymerized state. The remainder is present as monomeric actin complexed to a protein, profilin, which stabilizes the actin. The polymerized actin forms a network which helps maintain the 17 cell shape and the distribution of the organelles dispersed in the cytoplasm. Following activation, 70-80% of the total actin becomes polymerized.130 This network of polymerized actin is present as bundles of filaments associated with actin binding protein. Polymerized actin is found in the forming pseudopodia as well as in the cytoplasmic network. The cytoplasmic network is rich in actin and myosin and is associated with the centralizing organelles.478 Once platelet activation occurs and the membrane skeleton is released from the plasma membrane, the cytoplasmic actin filaments become associated with the membrane CPIIb-IIIa. This association only occurs if aggregation is present. It does not occur if platelets are activated but aggregation is prevented.308'448 Myosin is present in the resting platelet primarily as monomers which are diffusely distributed throughout the cytoplasm. During platelet activation, increasing concentrations of cytosolic free calcium activate the Ca++-dependent enzyme, myosin light chain kinase (MLCK). Activated MLCK phosphorylates the 20kDa light chain of myosin, allowing myosin to polymerize and interact with actin.478 The interaction of myosin with actin activates the myosin-associated.Mg++-dependent ATPase1 and the subsequent generation of a contractile force results from the sliding of the myosin and actin filaments past each other.228 Following stimulation with most agonists, myosin phosphorylation is closely correlated with shape change.95 The subsequent actin-myosin association also plays a role in the centralization of cytoplasmic granules and secretion.142 18 Finally, there are a variety of other cytoplasmic structural proteins associated with the cytoskeleton. These proteins can be divided into four basic classes: (1) those that complex monomeric actin and help stabilize it (ie. profilin); (2) those that cross-link actin filaments and regulate the formation of an organized network (ie. actin binding protein ands -actinin); (3) those that stabilize actin filaments (ie. tropomyosin); and (4) those that cap actin filaments and restrict their length (ie. gelsolin and 235K protein).478 Dense Tubular System The dense tubular system (DTS) is an internal membrane system analogous to the sarcoplasmic reticulum in muscle cells. There is no direct communication between the DTS and the external environment, although the DTS does form junctional complexes with the open canalicular system.450 The role of these complexes is not known. The DTS is derived from the megakaryocyte endoplasmic reticulum and has been demonstrated to act as an internal calcium (Ca++) storage site.51 Numerous studies have identified an energy—driven Ca++-ATPase on the membranes of the DTS which mediates Ca++ uptake and internal sequestration.]-O7'111'166v265 The DTS also contains a receptor-operated Ca++ channel that mediates Ca++ release into the cytoplasm following appropriate stimulation. This channel is separate from the Ca++-ATPase transport system and does not require ATP to function. Ca++ release via this channel is not induced by increasing concentrations of cytosolic free ionized calcium ([Ca++]1) but it is stimulated by the production of the water soluble second messenger inositol trisphosphate (1P3) from the 19 hydrolysis of membrane phosphatidylinositol 4,5-bisphosphate (PIP2).2 Finally, the DTS is the major site in the platelet for both cyclooxygenase and thromboxane synthetase activity.69'227 Open Canalicular System The open canalicular system (008) consists of tortuous, interconnected invaginations of the plasma membrane within the platelet cytosol which extend from one surface of the platelet to the other.453 These channels serve as one mechanism by which cytoplasmic granules are released to the external environment during secretion.454 The OCS also is a route for the internalization of substances from the plasma,453 a site for the clearance of membrane receptor complexes,230 and a membrane reserve that can be evaginated onto the platelet surface during platelet activation.118 Plasma Membrane The platelet plasma membrane is a typical lipid bilayer characteristic of many cell types. It's two main functions are to provide: (1) a permeability barrier to limit the transport of substances into and out of the platelet cytosol; and (2) a matrix to support the surface receptors involved in cell interactions and contact.26 The membrane is composed of 35% lipids, 57% proteins and 8% carbohydrates. The lipids consist of primarily the phospholipids and cholesterol. The phospholipids are arranged in a fluid bilayer with their hydrophilic heads oriented toward the aqueous internal cytoplasm or external 20 environment and their hydrophobic tails oriented towards the center of the bilayer. The portion of the bilayer in contact with the external environment contains primarily the neutral phospholipids (phosphatidylcholine and sphingomyelin) while the inner portion of the membrane in contact with the cytosol contains primarily the negatively charged phospholipids (phosphatidylethanolamine, phosphatidylserine and phosphatidylinositol).370 At physiologic temperatures, the platelet plasma membrane is fluid. Cholesterol is dissolved in the phospholipids and helps modulate the fluidity of the bilayer.370 Distributed throughout the plasma membrane bilayer are proteins and glycoproteins which function as specific surface receptors, transport systems or enzymes. The hydrophobic portions of the proteins or glycoproteins interact with the center of the bilayer while the hydrophilic portions interact with either the polar surfaces of the bilayer or the aqueous external or internal regions.370 In the normal physiologic state, these proteins or glycoproteins are able to move laterally over the surface of the membrane; however, they are not easily removed from the membrane except with the use of detergents. Specific platelet membrane glycoproteins reported to be involved in the processes of adhesion, aggregation or activation include CPIa-IIa, CPIb, CPIc-IIa, CPIIb-IIIa, GPIV, CPV, GPIX, cup-140 and warn-215 21 CPIaoIIa CPIa-Ila exists as a heterodimer and is part of a superfamily of adhesion receptors termed the integrins.215 The complex is related to the very late activation antigen 2 (VLA-2) present on the surface of activated T lymphocytes, with CPIa being analogous to the alpha subunit and CPIIa being analogous to the beta subunit. Studies on purified GPIa-IIa indicate that the complex plays a role in mediating the Mg++- dependent adhesion of platelets to collagen.400'401 Evidence supporting the role of CPIa in collagen-induced platelet activation comes from studies on the platelets from a person with a deficiency of membrane CPIa. This patient had a mild clinical bleeding disorder characterized in the laboratory as a failure of platelets to respond to collagen stimulation.282 Recent studies have reported that CPIa is actually two distinct proteins, Ia and Ia*, which can be separated by two-dimensional nonreduced/reduced gel electrophoresis. Ia* may be a granule constituent that is only present on the surface of activated platelets. It is this glycoprotein which is absent in the patient with the bleeding disorder.35 GPIb GPIb is a major sialoglycoprotein expressed on the surface of non- activated platelets. It consists of alpha and beta chains and it forms a non-covalent complex with CPIX. CPIb also forms a complex with the cytoskeletal protein, actin binding protein, linking it to the cytoplasmic membrane skeleton as discussed above.78'128'129 In 22 experimental isolation procedures, CPIb consistently coisolates with several poorly defined cellular components. This has lead to the proposal that the CPIb complex plays a role in signal transduction.78 At physiologic shear rates, the alpha chain of CPIb functions in the von Willebrand factor (vWF)-mediated adhesion to the exposed subendothelium.197 The exact mechanism by which vWF mediates this adhesion is not known; however, vWF binds readily to heparin-like glycosaminoglycans and collagen, both of which are present in the subendothelium. Binding to the subendothelium induces a change, possibly conformational, in vWF, which allows it to bind to CPIb and subsequently cause contact-mediated platelet activation.78'215 An autosomally recessive deficiency of the membrane complex CPIb- IX has been identified in humans and is termed Bernard-Soulier syndrome (BSS).260 It is characterized by mucocutaneous bleeding of variable severity. It clinically resembles von Willebrand disease (vWD), a disorder associated with the decreased function or production of vWF. Unlike vWD, however, the defect in BSS is not corrected by the addition of normal vWF since affected platelets are not able to bind vWF.260 BSS is characterized by the inability of affected platelets to adhere to the exposed subendothelium. In vigrg aggregation in response to stimulation with ADP, collagen or epinephrine is normal while aggregation in response to thrombin is delayed.142 CPIb contains a binding site for thrombin on its alpha chain.197 The importance of this site in platelet activation in response to thrombin is not clear. As mentioned, the platelets from humans with BSS aggregate in response to 23 thrombin, although the rate of aggregation is reduced.260 In addition, studies have shown that the binding of thrombin to CPIb is not directly coupled to activation of the major effector enzymes systems including phospholipase C, protein kinase C or phospholipase A2.185 Platelets from BSS patients and neuraminidase-treated platelets have a reduced surface charge due to decreased sialic acid density. This is responsible for decreased electrophoretic mobility and, presumably, the shortened survival time of these affected or altered platelets.ll‘7'286 Studies on platelets from humans with BSS have shown that there is also altered organization of the phospholipid bilayer, suggesting a role for CPIb in governing membrane fluidity.“81 GPIc-IIa CPIc is a relatively minor membrane glycoprotein which forms a heterodimer complex with CPIIa. The complex is a member of the integrin superfamily and is analogous to the very-late-activation antigen 3 (VLA- 3) on T-lymphocytes. It mediates adhesion to fibronectin and may interact with cytoskeletal actin filaments.78v215 GPIIb-IIIa The CPIIb-IIIa complex is a major constituent of the platelet plasma membrane and is another member of the integrin superfamily. CPIIb is composed of two disulfide-linked alpha and beta subunits while CPIIIa consists of a single polypeptide chain.307 The two glycoproteins 24 exist as a Ca++-dependent heterodimer complex.63 Studies have identified a number of binding sites for Ca++ on CPIIb.307 In resting platelets at physiologic temperatures, the complex can be reversibly dissociated by short (less than 10 minutes) incubation with the divalent cation chelator ethylenediamine-tetraacetate (EDTA). Longer incubation periods cause irreversible dissociation of the complex.126 The CPIIb-IIIa complex contains the activation dependent fibrinogen receptor and, as such, plays an essential role in platelet aggregation.303 This role of the complex was established based on the following findings: (1) platelets with little or no membrane CPIIb-IIIa cannot bind fibrinogen; (2) platelets can bind approximately 40,000 molecules of fibrinogen, the same as the number of molecules of CPIIb- IIIa; (3) isolated CPIIb-IIIa adsorbed to plastic can bind fibrinogen in the presence of Ca++; and (4) several monoclonal antibodies directed against portions of the CPIIb-IIIa complex can inhibit fibrinogen binding.473 Monoclonal antibodies that recognize only the CPIIb-IIIa complex bind to both resting and stimulated platelets, indicating that an intact GPIIb-IIIa complex is present on resting platelets.261 Following activation, additional CPIIb-IIIa complexes are exposed on the platelet surface. Whether these complexes originate from increased exposure of the open canalicular system.“66 or fusion of the alpha granules with the 89 plasma membrane remains to be determined. Recent studies have shown that the membrane CPIIb-IIIa complexes on resting platelets are actively 25 recycled into the open canalicular system, to secretory granules and back to the surface membrane.446 The GPIIb-IIIa complex on resting platelets cannot bind fibrinogen. Some stimulation of the platelet is required to expose the fibrinogen binding site on the complex. This finding is supported by the discovery of monoclonal antibodies that only recognize the activated CPIIb-IIIa complex.373 The exposure of the fibrinogen binding site may be dependent on a conformational change in CPIIb-IIIa or a rearrangement in the surrounding membrane.215'307 Using studies on permeabilized platelets, it has been proposed that there are two distinct routes leading to fibrinogen binding site exposure.371 The first route involves hydrolysis of membrane phosphoinositides, the production of second messengers and the subsequent elevation in [Ca++]i. The second route is Ca++-independent and may involve regulation by G proteins. Alternatively, it has been proposed that aggregin (a lOOkDa membrane protein) sterically interferes with the CPIIb-IIIa complex. Proteolytic digestion of aggregin (i.e. chymotrypsin) or the binding of ADP to aggregin induces a conformational change in the protein which leads to the exposure or unmasking of the fibrinogen binding site on CPIIb/IIIa.82v83 Once activated, the CPIIb-IIIa complex can bind to a variety of adhesive proteins including fibrinogen, fibronectin, vitronectin, collagen and vwr.215 CPIIb-IIIa recognizes the peptide sequence Arg- Cly-Asp (RCD) on these adhesive proteins; however, optimal binding of fibrinogen may require recognition of addition sequences.307'322 26 Because of its relative abundance and/or its higher affinity, fibrinogen is the primary mediator of platelet aggregation.215 In the absence of fibrinogen, alternative adhesive proteins (such as vWF) can bind to GPIIb-IIIa and sustain aggregation.105 The CPIIb-IIIa complex has numerous other functions besides the mediation of fibrinogen binding. First, CPIIb-IIIa binds to the cytoskeletal proteins talin and actin filaments following the onset of platelet aggregation.181v308'429 This interaction connects the external adhesive proteins to the internal contractile proteins generated after activation and facilitates the process of clot retraction. Secondly, the binding of fibrinogen to CPIIb-IIIa initiates Na+/H+ exchange and subsequent TxA2 production in epinephrine-stimulated platelets by some, as yet, undefined mechanism.20 Thirdly, the CPIIb-IIIa complex itself appears to play'role in Ca++ influx in both resting and stimulated p1atelets.52'316'468'469 Lastly, recent studies have demonstrated activation-induced phosphorylation of CPIIIa which may be involved in the induction of granule secretion.207r296 Clanzmann's thrombasthenia is an autosomal recessive disorder identified in human platelets which is characterized by a deficiency or altered function of the CPIIb-IIIa complex.260 The disorder is clinically characterized by variable severity of mucocutaneous bleeding. Laboratory findings include abnormal clot retraction and failure of platelets from affected individuals to aggregate in response to any of the physiologic agonists. Affected platelets adhere to subendothelial 27 tissue and bind vWF, but they fail to spread or to recruit additional platelets to form a primary hemostatic plug.26o GPIV GPIV is a major component of the platelet membrane which is also identified as CPIIIb. It has been recently reported to be the receptor for thrombospondin8 and may play an auxiliary role in aggregation. Platelets release thrombospondin from their alpha granules during secretion which then binds to GPIV in a Ca++-dependent manner. The thrombospondin also binds to fibrinogen and may form bridges which stabilize the platelet aggregates.297 Other studies identify GPIV as a receptor involved in collagen adhesion.415 GPV The role of CPV in platelet function is not known.78 In vivg, CPV is cleaved by thrombin to release a water-soluble fraction, leading to the suggestion that it serves as a thrombin receptor. Recent studies have shown, however, that a monoclonal antibody to CPV does not inhibit thrombin activation, thereby negating its role in thrombin-induced platelet activation.36 GPIX CPIX is found tightly bound to CPIb in the platelet membrane but its role in this complex is unknown.78 28 GNP-140 GNP-140 is a membrane protein within the secretory granules which is redistributed to the platelet plasma membrane following platelet activation. Monoclonal antibodies to CMP-l40 serve as markers for activation. The role of CMP-14O is not known; however, it appears to be related to a gene family of receptors called selectins. This family includes ELAM-l, a cytokine-inducible receptor on endothelial cells that mediates neutrophil binding.259 A monoclonal antibody directed against CMP-14O inhibits collagen- and thrombin-induced platelet aggregation, suggesting a role for CMP-14O in mediating platelet aggregation.297 PTAl PTAl is a heavily sialated glycoprotein of Mr 67kDa which is present on the plasma membrane of platelets and lymphocytes. It is believed that the associated sialic acid residues that protrude from the plasma membrane strongly contribute to the surface charge along with similar residues associated with CPIb. Monoclonal antibodies directed against PTAl initiate platelet aggregation and secretion. It has been proposed that this platelet activation involves protein kinase C (PKC) and subsequent phosphorylation of the 47kDa protein. During activation by the monoclonal antibody, the PTAl antigen is phosphorylated. This antigen is phosphorylated during activation by the physiologic agonists collagen and thrombin and the non-physiologic agonist phorbol. The role of PTAl in platelet activation in V119 is not known.367 29 THE PLATELET RESPONSE The primary role of platelets in hemostasis is to form a platelet plug in response to injury to thevessel wall. Although platelets respond to a wide variety of stimuli, they do so by eliciting a limited series of physiologic events termed the basic platelet response: adhesion of platelets to the site of injury, shape change, aggregation, and the secretion of granule contents and release of arachidonate metabolites. The number of these responses elicited depends on: (1) the number and type of stimulating agonists; and (2) the concentration of agonist(s) in contact with the platelets.377.403 Adhesion Platelets will adhere to almost any foreign surface. Whether the platelets become further activated depends on the surface involved. In xigrg, unactivated platelets which adhere to a surface can.either detach or remain adhered.29 1g xivg, platelets rarely, if ever, come in contact with foreign surfaces since circulating plasma proteins readily adsorb to such surfaces. The degree of adhesion and subsequent activation of the circulating platelets depends on the protein adsorbed. Albumin inhibits adhesion while fibrinogen, vWF, fibronectin, vitronectin, thrombospondin, a z-macroglobulin and laminin all promote adhesion.295 Platelets adhere to surfaces in a process mediated by plasma membrane glycoproteins. At high shear rates, adhesion of platelets to 30 the subendothelium is probably mediated by membrane CPIb and vWF.295'354 Studies in_21§19 indicate that at high flow rates the exposure of vWF to a foreign surface causes an alteration in vWF, allowing it to bind to CPIb which is already exposed on the membrane of resting platelets. This leads to contact-induced platelet activation and the exposure of the fibrinogen binding site on CPIIb-IIIa. Fibrinogen, vWF and the other adhesive proteins are able to bind to the complex and mediate spreading or platelet-to-platelet adhesions.264 The role of CPIb and vWF in platelet adhesion is substantiated by the absence of platelet adhesion in patients with Bernard-Soulier syndrome.260 The role of GPIa in platelet adhesion is less clear. ln_xigg, fibrillar collagen is the most likely subendothelial element that initiates platelet adhesion since it is a strong inducer of platelet aggregation.309 As discussed, CPIa has been identified as a collagen receptor. Studies on the platelets of a patient with a congenital deficiency of CPIa have shown decreased adhesion to the subendothelium as well as absence of shape change and aggregation in response to collagen. The abnormal adhesion appears to be due to defect in spreading and not in the initial adherence. These studies suggest that vWF mediates the initial attachment of the platelet to the subendothelium while collagen mediates the subsequent spreading.282'283 Recent studies have shown that platelets adhere to collagen in both a Mg++-dependent and a bivalent-cation-independent manner. The Mg++- dependent adhesion is mediated by CPIa. The mechanism mediating the bivalent-cation-independent adhesion is not known. It has been proposed that this adhesion is mediated by vWF or some other adhesive protein 31 which can bind to collagen and act as a bridge connecting collagen to the platelet.476 Adhesion requires metabolic energy, a functioning contractile mechanism and an external source of bivalent cations. It causes activation similar to that caused by a strong agonist such as thrombin. Adhesion is not dependent on the formation of thromboxane A2 (TxAz) or the release of ADP.295 The initial phase of adhesion is the contact phase which is followed by the spreading phase where there is increased contact between the platelet membrane and the surface. One group has proposed that there are three steps involved in the process of adhesion: (1) the attachment of vWF to the subendothelium; (2) the binding of vWF to the platelet; and (3) platelet spreading and aggregate formation.309 This theory is supported by studies using monoclonal antibodies. Antibodies against CPIb block adherence to the subendothelium at low and high shear rates while antibodies against CPIIb/IIIa decrease adherence at high shear rates only. This suggests that CPIb is the primary receptor for adhesion while CPIIb/IIIa mediates spreading since it is only at high shear rates that spreading becomes important.104 Shape Change Shape change is the initial, reversible response of platelets in suspension to a stimulus.377 Mbrphologically, it is characterized by the loss of the platelet's discoid shape, transition to a spherical shape, the extension of pseudopods or filopodia and the movement of the internal organelles to the cell center where they are surrounded by a 32 ring consisting of the constricting microtubular coil and actin filaments.452 Shape change is accompanied by actin polymerization, both as the formation of bundles within the filopodia and asva network around the centralizing granules.478 The formation of pseudopods is inhibited by a compound, cytochalasin B, which caps actin filaments. The subsequent addition of thrombin, causes granule centralization because the agonist-induced increase in [Ca++]1 still supports the formation of a contractile cytoskeleton containing actin and myosin. These findings suggest the importance of actin polymerization in the extension of filopodia and the importance of the actin-myosin interaction in granule centralization.142 Shape change occurs within a few seconds of stimulation, requires metabolic energy and is independent of the presence of external divalent cations (ie. Ca++ and/or Mg++) or fibrinogen.172'377 Granule centralization is not dependent on the release of arachidonic acid metabolites. Cyclooxygenase inhibition by aspirin blocks granule centralization only in those cases where there is also inhibition of myosin light chain phosphorylation, such as with the weak agonist epinephrine.142'143 Physiologically, shape change precedes aggregation and secretion; however, shape change may occur without subsequent aggregation and secretion. Shape change and granule centralization are correlated with the phosphorylation of the 20kDa myosin light chain.95'142 Myosin light chain phosphorylation is stimulated by increasing cytosolic [Ca++]1 in a calmodulin-dependent manner.96v163 Phosphorylation of myosin light 33 chain may also occur independent of an increase in cytosolic Ca”.153 It has been proposed that this phosphorylation occurs at a site separate from the Ca++-dependent phosphorylation site and is mediated by PKC.276 Other studies have demonstrated that increasing cytosolic pH (pHi) and an influx of external Na+ plays a role in cytoskeletal reorganization and, therefore, may play a role in the initiation of shape change.231 Shape change and granule centralization are not a prerequisite for aggregation and secretion. This is best demonstrated by those agonists which are selective activators of PKC (ie. phorbol myristate acetate or synthetic diacylglycerides).142 These agonists cause some degree of pseudopod formation, possibly due to PKG-mediated stimulation of actin polymerization; however, granule centralization does not occur in spite of the initiation of aggregation.142 In addition, shape and granule centralization are not limited to activated platelets. Exposure of platelets to cold or to antimitotic drugs (such as vincristine or colchicine) cause shape change due to the disassembly of the microtubule coil. Since there is no associated myosin light chain phosphorylation, there is no granule centralization.142v452 Aggregation In the physiologic setting, aggregation generally follows shape change, though it requires a stronger stimulus than shape change. In jxitgg, two types of aggregation responses can be demonstrated: (1) reversible aggregation which is not accompanied by secretion; and (2) irreversible aggregation which is accompanied by secretion and the 34 release of arachidonate metabolites.377 Aggregation requires fibrinogen (or some other adhesive protein), extracellular divalent cations (Ca++ and/or Mg++), exposure of the fibrinogen binding site and platelet-to- platelet collisions.377'478 The process of aggregation itself may be a passive, non-energy requiring process once shape change and exposure of the fibrinogen binding sites occur.403 Aggregation is proposed to occur when a dimeric fibrinogen molecule binds to two adjacent platelets, creating a platelet-to- platelet bridge.303'478 Binding of fibrinogen to the platelet membrane requires some activation-induced alteration in the CPIIb-IIIa complex which subsequently permits fibrinogen to attach. The binding of fibrinogen to its receptor is specific, saturable and inhibited by the chelation of external Ca++ and Mg++. Fibrinogen binding is also inhibited by metabolic inhibitors, prostaglandin E1 (PCEl) and pH < 6.5.303 Fibrinogen is a glycoprotein hexamer consisting of three non- identical pairs of chains (alpha, beta and gamma).377 It contains two low affinity binding sites on the alpha chains and two high affinity binding sites on the gamma chains. The relative contribution of the different binding sites to in vivg platelet aggregation is not clear; however, studies indicate that the alpha chain bindings sites are not required for irreversible fibrinogen binding.302 Irreversible fibrinogen binding is defined as the permanent attachment of fibrinogen to the platelet which is unaltered following incubation with EDTA and is dependent on cytoskeletal activation and association with CPIIb/IIIa.304 finale initial, reversible fibrinogen binding is dependent on an intact (SPIIb/IIIa complex, irreversible fibrinogen binding is independent of an 35 intact CPIIb-IIIa complex. Irreversible binding may be mediated by secondary interactions between fibrinogen and CPIIb, GPIIIa or some alternate membrane constituent.302 Exposure of the fibrinogen binding site may involve a change in the conformation of the CPIIb-IIIa complex or in the surrounding membrane arrangement.215 It has been demonstrated that the exposure of the fibrinogen binding site is a Ca++-dependent process.373 Inositol 1,4,5-trisphosphate (1P3) and increased [Ca++]i induce fibrinogen binding site exposure in saponin-permeabilized platelets. This effect is blocked by cyclooxygenase inhibition, suggesting a dependence on the Ca++-dependent production of arachidonic acid metabolites. Stable CTP analogues and diacylglycerol also induce exposure of the fibrinogen binding site in a similar system.281‘v371 These studies suggest exposure of the fibrinogen receptor is partially dependent on G proteins and activation of PKC.377 Epinephrine induces fibrinogen receptor exposure in the absence of any synergistic agonists. Whether this is due to a localized Ca++ flux, the action of a C protein or some other second messenger system has yet to be determined.372 Alternately, it has been proposed that the ADP receptor, aggregin, plays a role in the exposure of the fibrinogen binding site. This lOOkDa membrane protein is located in close proximity to the CPIIb-IIIa complex. In the resting state, aggregin is proposed to sterically interfere with the CPIIb-IIIa complex. Binding of ADP causes a conformational change in aggregin, thereby allowing exposure of the fibrinogen binding site. This conformational change in aggregin can 36 also be induced by proteolysis by chymotrypsin or high concentrations of thrombin.81'82'83 The proteolysis induced by thrombin is not a direct cleavage; instead, it is due to the thrombin-stimulated activation of the intracellular protease calpain.321 Numerous studies have shown that fibrinogen binding alone is not sufficient to sustain aggregation, demonstrating the need for one or more post-fibrinogen binding events. Two monoclonal antibodies to CPIIb-IIIa (AP3 and Tab) allow fibrinogen binding to platelets but inhibit aggregation and secretion.281 Platelets exposed to the cold or formalin fixed after stimulation with ADP are able to bind fibrinogen but are unable to aggregate.305 This suggests that normal membrane fluidity or flow may be necessary for aggregation, possibly as a prerequisite for receptor mobility. The Release Reaction The release reaction is characterized by the secretion of the contents of cytoplasmic granules and the formation of arachidonate metabolites. The release reaction comprises two major positive feedback loops that serve to amplify the platelet response: (1) the release of ADP from dense granules; and (2) the release of arachidonate metabolites.103'377 This amplification is most important in the stimulation of full platelet activation by agonists of intermediate or weak strength. The release reaction does not play a role in the primary responses of adhesion, shape change and reversible aggregation in response to any of the agonists. The contribution of each of the 37 release pathways to the overall response can be studied ig_yi§gg by the addition of: (l) cyclooxygenase blockers to inhibit arachidonate metabolism; and (2) ADP-scavenging enzymes (apyrase, creatinine phosphokinase) or ADP-receptor blockers (ATP) to prevent feedback by released ADP.377 Secretion of Granules Secretion of the contents of both alpha and dense granules occurs following stimulation with sufficient doses of any of the physiologic agonists. Secretion of lysosome enzymes, however, requires stimulation with strong agonists such as high concentrations of thrombin or collagen. Relative to alpha and dense granules, secretion of lysosome enzymes occurs at a slower rate and never involves complete release of the lysosomal contents. Secretion of all types of granules requires energy and may be inhibited by metabolic inhibitors.359'403 Two potential mechanisms for granule release to the external environment have been proposed. First, following platelet activation the membrane of the storage granules may fuse with the membrane of the open 454 ~An alternate mechanism, demonstrated in bovine canalicular system. platelets (which do not have a formed OCS) and proposed to also exist in human platelets, involves the fusion of granules directly with the plasma membrane.31o'451' Electron microscopic examination of platelets stimulated using synthetic diacylglycerol has shown swelling of granules, fusion of the swollen granules to each other and the eventual fusion of granules with both the OCS and the plasma membrane.142 38 Secretion has frequently been associated with activation of phospholipase C (PLC) and the subsequent production and/or release of intracellular second messengers.17o’222'382 Following stimulation with agonists of weak or intermediate strength, secretion is fully or partially dependent on the secondary activation of PLC by the released arachidonate metabolites. Secretion in response to these agonists is fully or partially blocked by the use of cyclooxygenase inhibitors.16'68 Granule secretion is not, however, strictly dependent on PLC activation. This is demonstrated by the addition of thrombin to saponin- permeabilized platelets which causes secretion without any measurable activation of no.2“ In addition, inhibition of PLC-mediated 'diacylglycerol formation by a compound which inhibits G protein function, CDI‘D S, in saponin-permeabilized platelets does not inhibit thrombin-stimulated dense granule secretion.57 Cytosolic Ca++ also plays a role in mediating platelet secretion. Permeabilized platelets can be induced to secrete the contents of their granules by increasing the [Ca++] in the external medium which directly increases [Ca"""]1.20l"205 The addition of the calcium ionophore, ionomycin, to intact, aspirin-treated platelets causes elevation of [Ca++]1 and subsequent secretion in the absence of measurable TXA2 formation and in the absence of PLC activation.339 The addition of calcium ionophore A23187 to platelets from patients with Clanzmann's thrombasthenia causes dense granule secretion independent of both TxAz production and aggregation.232 One role for Ca++ is the Ca++-dependent activation of MLCK and subsequent phosphorylation of myosin light chain. As discussed earlier, this leads to interaction of myosin and actin 39 filaments and the generation of a contractile force which plays an important role in the process of secretion.1l‘2'203 A recent report has shown impaired release of intracellular Ca++ and decreased phosphorylation of myosin light chain in the platelets from two humans with secretion defects, once again demonstrating the close association between cytosolic Ca++ and secretion.325 Secretion can also occur at basal cytosolic [Ca"""]1.33“'335'360 This may result from activation of PKC since numerous studies have shown that the phosphorylation of the 47kDa protein is closely associated with secretion.3'285'291’377*400 One study showed that PKC activation facilitates the membrane fusion necessary for granule secretion.139 In xivg, Ca++ mobilization and PKC activation most likely work in synergy to promote granule secretion.159'191"380 One researcher has proposed a two step model for platelet secretion. The initial release of Ca++ and myosin light chain phosphorylation cause shape change and granule centralization. The increased [Ca++]1 primes PRC for the next step where activation of PKC then modulates platelet secretion.377 The Release of Arachidonate Metabolites Stimulated platelets release arachidonate from the various membrane phospholipids. The major pathway for arachidonic acid release involves the hydrolysis of phospholipids (primarily phosphatidylcholine but also phosphatidylethanolamine, phosphatidylinositol and phosphatidylserine) at the fatty acyl 2 position by phospholipase A2 4O (PLAZ).122'245-319'320'434 PLA2 is a Ca++-dependent enzyme”,397 that is activated by directly increasing [Ca++]1 in permeabilized platelets151 or by increasing [Ca++]1 in intact platelets using calcium ionophores.339 One report identified at least two separate forms of PLA2, one specific for phosphatidylcholine which is dependent on increased [Ca++]1 and one specific for phosphatidylethanolamine which is Ca'H’-independent.19 Factors other than cytosolic Ca++ play a role in regulating PLA2 activity. PLA2 activity is maximal in an alkaline environment.27 Activation of Na+/H+ exchange by agonist-stimulation causes cytosolic alkalinization which potentiates activation of PLA2.6'411 Activators of PKC enhance arachidonic acid release by modulating the activity of PLA2.7'122v150v152 In addition, other studies suggest that G proteins serve a regulatory function in PLA2 activity.6'7'137'280'396 G proteins are a family of related guanine-nucleotide binding proteins that mediate the interaction between the surface receptor and an intracellular effector system. There are two proposed mechanisms by which G proteins regulate PLA2: (l) G protein-dependent activation of PLC with the production of (l,4,5)IP3 and the eventual release of Ca++; or (2) direct G protein activation of PLA2.by some, as yet, unidentified C protein.53 Two other routes for activation-induced arachidonic acid release exist in platelets. The first is PLC-mediated hydrolysis of phosphoinositol with the production of the second messenger 1,2- diacylglycerol (1,2-DG). Arachidonic acid is then released by the sequential action of diglyceride and monoglyceride lipases.76 The 41 second pathway involves the phosphorylation of 1,2-DC to phosphatidic acid (PA) and subsequent activation of a Ca++odependent PLA2.38 In the resting cell, the concentration of arachidonic acid is maintained at a low level by the activity of arachidonate scavenging enzymes which incorporate free arachidonate into membrane phospholipids by the sequential action of the enzymes arachidonyl-CoA synthase and arachidonyl-CoA lysophospholipid acyltransferase. Decreased activity of either of these enzymes could lead to increased levels of free arachidonic acid. One study showed that PKC activation inhibits both of these enzymes 136 while a separate study showed no effect.173 Once freed from the membrane phospholipids, arachidonic acid may be oxygenated by either the cyclooxygenase or the lipoxygenase pathway. In the cyclooxygenase pathway, arachidonic acid is converted to prostaglandins 62 and “2 (PGGz/PGHZ). These endoperoxides are metabolized by thromboxane synthetase to produce the unstable metabolite TxAz.“03 All of these metabolites are potent platelet agonists and are an important part of the amplification mechanism. They cause activation of PLC379 and Ca++ mobilization/‘7'50 The lipoxygenase pathway causes the production of 12- hydroxyeicosatetraenoate (lZ-HETE) and 15-hydroxyeicosatetraenoate (15- HETE) as well as several other dihydroxylated products.447 The role of these metabolites in platelet activation is unclear. Some evidence indicates an inhibitory role for these products. l2-HETE inhibits PLA2 activity, decreasing the release of arachidonic acid from membrane 42 phospholipids and causing inhibition of aggregation.369 Other evidence suggests a stimulatory role for the lipoxygenase metabolites. Lipoxygenase inhibitors inhibit both Ca++ flux and aggregation in response to ADP, indicating that lipoxygenase metabolites play a stimulatory role in ADP-induced aggregation and Ca++ mobilization.“6 SIGNAL TRANSDUCTION OR PROCESSING None of the physiologic agonists are able to penetrate the platelet plasma membrane; instead, they bind to surface receptors.478 Signal transduction is the process by which an external stimulus (ie. agonist binding) is transmitted through the plasma membrane and translated into a cellular response. The binding of an agonist to a surface receptor stimulates specific effector systems, such as enzymes or ion channels, which then regulate the production of cellular second messengers. It is these second messengers which actually modulate the physiologic response by altering the conformation, enzymatic activity or phosphorylation of cellular proteins.377 Although platelets are stimulated or inhibited by a wide range of substances, there appears to be a finite number of effector systems available for signal transduction. It is often difficult to correlate a specific effector system or second messenger with a specific cellular response since many events are occurring within a very short period of time following platelet activation. In addition, there are multiple points of cross- over or checks and balances between the different effector systems. Enzyme systems which play a major role in signal transduction include the following: (1) adenylate cyclase which controls the production of 43 the second messenger cyclic AMP (cAMP); (2) PLC which mediates the hydrolysis of membrane phosphoinositides and subsequent production of the second messengers (1,4,5)IP3 and 1,2-DC; and (3) PKC which phosphorylates a variety of cellular substances. Important second messengers include (1,4,5)IP3, 1,2-DC and Ca++. Ion channels which play a role in platelet activation include Ca++ channels and the Na+/H+ antiport. Adenylate Cyclase and cAMP Production Cyclic AMP production is regulated by adenylate cyclase while cAMP removal is regulated by cAMP phosphodiesterases. Increased cAMP is caused by agents that stimulate adenylate cyclase (ie. prostacyclin, PGEI, prostaglandin D2 and adenosine) and/or inhibit phosphodiesterase activity.9'403 Decreased cAMP is caused by agents that inhibit adenylate cyclase (ie. epinephrine, ADP and thrombin) and/or stimulate phosphodiesterase activity.“03 As a general rule, increased cAMP is inhibitory to platelet function while decreased cAMP facilitates platelet activation.9'377 Increased cytosolic cAMP inhibits platelet adhesion, shape change, secretion and aggregation.376 Adenylate cyclase produces cAMP from ATP in a reaction coupled to surface receptors by G proteins which mediate the interaction between receptors and effector systems. Adenylate cyclase is regulated by two distinct G proteins: (1) G8 which serves a stimulatory role; and (2) Ci which serves an inhibitory role. Cs and C1 are heterotrimeric proteins with similar beta and gamma subunits but different alpha subunits. 44 Cholera toxin ADP ribosylates the alpha subunit of CS, leading to permanent activation. Pertussis toxin ADP ribosylates the alpha subunit of C1 which uncouples Gi from the receptor and prevents agonist-induced inhibition of adenylate cyclase.53’377 Na+ and other monovalent cations inhibit both the basal and stimulated activity of adenylate cyclase. The data suggests that the cations interfere with coupling between membrane receptors, G protein and the catalytic portion of adenylate cyclase. Na+ and CTP are also necessary for maximal inhibition of adenylate cyclase by epinephrine.405 Binding of an agonist to its receptor leads to the dissociation of GDP from the heterotrimeric G protein and the binding of CTP to the alpha subunit. Binding of CTP alters the alpha unit in some manner, possibly conformational, and allows it to dissociate from the beta-gamma units. The dissociated alpha subunit is mobile in the membrane where it interacts with the target effector system while the beta-gamma subunits remain relatively stationary. The effect is terminated by the slow hydrolysis of CTP to GDP by the GTPase inherent in the alpha subunit. The alpha subunit then reassociates with the beta-gamma units. Nonhydrolyzable analogues of CTP (ie. Cpp(NH)p or CTPgammaS) bind at the GTP site and cause persistent G protein activation while nonhydrolyzable analogues of GDP (ie. GDIB S) bind at the GDP site and cause persistent inhibition. 53 : 377 Although it is clear that cAMP inhibits platelet function, the exact mechanisms involved are unclear. Cyclic AMP activates cAMP- dependent kinase(s) which phosphorylate, and therefore modulate, the 45 activity of a number of enzymes and proteins.9 Increased concentrations of cAMP increase the phosphorylation of at least four different proteins of Mr SlkDa (P51), 36kDa (P36), 24kDa (P24) and 22kDa (P22). P51 and P24 are phosphorylated very rapidly, but their roles in the inhibitory action of cAMP is not known. P24 has been identified as the beta subunit of CPIb which plays an important role in vWF mediated adhesion and is the site for the attachment of the membrane skeleton. Whether this phosphorylation plays a physiologic role is not known.133 P22 has been identified as the protein thrombolamban which is proposed to play a role in regulating the Ca++-ATPase pump which sequesters Ca++ into the DTS.3 Recent studies, however, have shown that P22 does not interact with the Ca++-ATPase present in the platelet membranes125 and that the 22kDa protein is actually related to a family of CTP-binding proteins. This study has also demonstrated that the cAMP-dependent kinase phosphorylates at least 9 different proteins. The identity and function of these proteins is not known.458 Numerous studies have shown that increased cAMP interferes with Ca++omediated stimulatory responses. cAMP inhibits the agonist-induced rise in [Ca++]1 by limiting influx from the external environment350'433 and/or suppressing release from internal stores.43'240'350'433 This inhibition of release may be due to a regulatory effect on the (1,4,5)IP3-induced Ca++ release from the'DTS.112'268 The release of internal Ca++ by calcium ionophores, which bypass the physiologic Ca++ release process, is not blocked by cAMP, indicating a specific role for cAMP in mediating this process.350 Cyclic AMP also stimulates Ca++ sequestration into the DTS."3'120'167 Others studies, however, have 46 shown no stimulation of Ca++ transport in platelet microsomes or membrane vesicles in response to cAMP.330v449 Increased cAMP has numerous other inhibitory effects. First, PKC activity is decreased due to cAMP-mediated inhibition of the potentiating effects of secreted ADP213 or some other non-ADP-mediated events.7l"213 Secondly, cAMP inhibits PLC-induced hydrolysis of membrane phosphoinositides by some, as yet, unidentified mechanism.205v344,376'430'472 Thirdly, cAMP stimulates the formation of phosphoinositol 4-phosphate (PIP), possibly by inhibiting phosphatidylinositol kinase. This causes decreased availability of polyphosphoinositides for hydrolysis by PLC and leads to decreased production of the second messengers (1,4,5)IP3 and 1,2-DG.73'220 In addition, increased cAMP inhibits fibrinogen receptor exposure.206'376 Finally, cAMP activates a cAMP-dependent kinase which phosphorylates MLCK.165 If the Ca++-calmodulin complex is present, this phosphorylation does not affect enzyme activity. If the Ca++-ca1modulin complex is absent, phosphorylation causes a decrease in MLCK activity and inhibits the subsequent activation of MLCK by Ca'H'.88 Phospholipase C and Membrane Phosphoinositide Hydrolysis One of the main effector systems coupled to the action of stimulatory agonists is the PLC-mediated hydrolysis of membrane phosphoinositides to produce the second messengers (1,4,5)IP3 and 1,2- DG. Additional second messengers produced as a result of this reaction 47 include other inositol phosphates, PA and lysophosphatidic acid (lysoPA). Phosphoinositide Metabolism Quantitatively, phosphatidylinositol (PI) is a minor phospholipid, comprising only 597% of total platelet phospholipids.31v252 Of that, approximately 15 and 5% are PIP and PIP2 respectively.377 Three features of the phosphoinositides point to an important role in platelet function: (1) a strongly acidic polar head group; (2) a preponderance of arachidonic acid; and (3) a very active metabolic cycle. The phosphoinositides contain a glycerol backbone with a polar head group containing a myo-inositol ring in position 3 and fatty acids bound to positions 1 and 2 on the glycerol backbone. The myo-inositol ring can be esterified with one or two phosphate groups to yield the polyphosphoinositides, PIP and PIP2.249 The l-stearoyl 2-arachidonyl species comprises 71% of the total phosphoinositides, compared to 47% of phosphatidylethanolamine and 10% of phosphatidylcholine.244 This positions arachidonic acid in a site where it is readily cleaved by PLA2 to participate in the platelet response. It has been proposed that the platelet enzyme l-acyl-glycerophosphoryl-inositol acyltransferase is partially responsible for the relative abundance of arachidonic acid at position 2.195 The phosphoinositides have a very active metabolism in both the resting and stimulated state. They are in a continual “futile cycle" involving an ATP-dependent interconversion of the different 48 phosphoinositides.249 The phosphoinositides exist in a metabolically homogeneous pool within the platelet, with relatively free exchange between the polyphosphoinositides and PI.“31 PI and PIP are phosphorylated by PI-kinase and PIP-kinase respectively. These kinases are Mg++-dependent and are bound to the plasma membrane.198 Phosphate groups are removed from the PIP and PIP2 by P1 phosphomonoesterases. These enzymes are located in both the cytosol and the membranes. They require Mg++ at the physiologic concentrations for maximal activity.37'338 This “futile cycle" is ongoing in the resting platelet and does not play a direct role in the agonist-stimulated breakdown of membrane phosphoinositides; however, it is important in maintaining the proper equilibrium of the different phosphoinositides for activation to occur.249 The ratio of PIP2:PA and PIP:PA remains relatively constant following agonist stimulation, suggesting that the activity of both the kinases and the phosphomonoesterases are tightly regulated in stimulated cells as well as in resting cells.404 PI can be formed from 1,2-DC by the following sequence of reactions: (1) phosphorylation of DC in position 3 to yield PA; (2) conversion of PA to cytidine diphosphate-diacylglycerol (CDP-DG) by CTP- DG cytidyl transferase; and (3) addition of an inositol ring to CDP-DG by P1 synthase to yield PI.21‘9'368 This pathway becomes an important mechanism following platelet activation to recycle the 1,2-DG Illolecule.2“9'368 49 Phospholipase C Mediated Hydrolysis of PI The phosphoinositides are primarily positioned on the cytoplasmic side of the plasma membrane, placing them at the correct physical location to play a role in signal transduction.368 Binding of an agonist to a specific receptor causes the PLC-mediated phosphodiesteric cleavage of the phosphoinositides to yield the corresponding inositol phosphate and 1,2—DC. Physiologic agonists having surface receptors which directly modulate PLC activity include thrombin, PAF, TxA2 and vasopressin.58'106'174'269'272'312'337'378 Agonist-induced stimulation of PLC activity is independent of extracellular Ca++ and Mg'H'.383 The initial response of agonist binding is a rapid decrease in PIP2 (within 10 seconds of stimulation) followed by increased formation of PIP2.37'277 The decrease in PIP2 could result from: (1) increased hydrolysis by PLC; (2) decreased phosphorylation of PIP by PIP kinase; or (3) increased dephosphorylation of PIP2 by phosphomonoesterase.249'377 There is simultaneously increased production of the inositol phosphates within 5 to 10 seconds of agonist addition,94'340v341'343v375'3782418 suggesting that the decrease in PIP2 is primarily due to hydrolysis by PLC. Phosphoinositide-specific PLC is not a single enzyme but is composed of multiple forms located in the cytoplasm and bound to the platelet membrane. The role of the different subtypes of the PI- specific-PLC is not currently known. They vary in terms of molecular weight, substrate specificity, pH optima and degree of Ca++- 50 dependence.18'25'53'238'343 An early study has suggested that the majority of PLC activity is present in the cytosol and requires millimolar concentrations of Ca++ for maximal activity.23 PLC is not activated by Ca++ alone, since the calcium ionophore A23187-mediated increase in cytosolic Ca++ is unable to initiate the hydrolysis of PIP2.337 These findings preceded the identification of: (l) a membrane- bound PLC which shows greater activity for the hydrolysis of the polyphosphoinositides and is active at Ca++ levels present within the resting platelet cytosol;25'248'343 and (2) a soluble PLC which shows preferential hydrolysis of PIP2 and is activated by G proteins.17 As a general rule, the identified PLC enzymes optimally hydrolyze PIP2 at low concentrations of Ca++ but hydrolyze PIPZ, PIP and PI equally well at millimolar concentrations of Ca++. This means that initial receptor- mediated activation of PLC induces the preferential hydrolysis of the polyphosphoinositides, causing the release of membrane bound Ca++37 and the production of the second messengers (1,4,5)IP3 and l,2-DC.377'418 (1,4,5)IP3 mediates the release of intracellular stores of Ca++ (see section on IP3-mediated events). The increased [Ca++]1 is associated with PLC-mediated hydrolysis of PI in addition to hydrolysis of polyphosphoinositides. G Protein Regulation of PLC Brass53 lists the following five criteria as evidence that a C protein, termed G , modulates PLC activity: (1) guanine nucleotides stimulate PIP2 hydrolysis and/or 1P3 and DC production in permeabilized cells or isolated platelet membranes; (2) guanine nucleotides mimic the 51 effects of agonists known to stimulate PLC; (3) agonist-induced PI hydrolysis is enhanced by the addition of CTP to permeabilized cells or membranes; (4) the addition of pertussis toxin or GDP analogues inhibit agonist-induced phosphoinositide hydrolysis; and (5) the extent of ADP- ribosylation of G proteins by pertussis toxin correlates with the inhibitory effects of the toxin on agonist-induced phosphoinositide hydrolysis. Each of these criteria have not been shown to operate in all tissues.53 Many studies in platelets implicate the important role of GP- mediated activation of PLC.53’57'280'399 Partially purified PLC from human platelet membranes is stimulated by the addition of purified G proteins from brain tissue, suggesting a role for C protein regulation.223 The addition of the GTP analogue guanosine 5'-({- thio)triphosphate (GTPyS]) to platelet membrane suspensions or to permeabilized platelets increases PLC activity and 1,2-DG and/or IP3 production.]'7'2“’57'160'175 In permeabilized platelets, CTP analogues potentiate the stimulatory effects of thrombin.175 Addition of a GDP analogue (GDIB S), on the other hand, inhibits the ability of thrombin to stimulate PLC in permeabilized platelets.57 Caution should be used in the interpretation of these results since some studies have suggested that both CTP- and GDP-analogues have effects on platelet activation independent of G protein-mediated events?!"210 As mentioned in the section on adenylate cyclase, pertussis toxin ADP-ribosylates the alpha unit of G1 and subsequently uncouples it from the receptor. Pertussis toxin also inhibits thrombin-induced release of 52 intracellular Ca++ 293 and 1,2-DG production57 by ADP-ribosylating membrane proteins, suggesting that a pertussis toxin-sensitive G protein similar to G1 functions in thrombin-stimulated PLC activation. This effect is overcome by using higher concentrations of thrombin.21'57 These findings indicate that low concentrations of thrombin activate PLC by a pertussis toxin-sensitive C protein while high concentrations of thrombin activate platelets by a pathway independent of PI hydrolysis57 or by a PLC pathway modulated by a pertussis toxin-resistant C protein.21 Pertussis toxin has minimal to no effect on PAF- or TxA2 mimetic U44069-stimulated GTPase activity in platelet membranes, indicating that the G protein regulating PLC activity in response to these agonists is pertussis resistent and therefore different from the adenylate cyclase C1.174 This finding has been supported by studies utilizing permeabilized platelets in which U46619-stimulated activation of PLC was unaffected by the addition of pertussis toxin. This study has also suggested that the TxA2 receptors are coupled to PLC by a pertussis-resistent G P Gp coupled to thrombin activation.58 which is different than the pertussis-sensitive Inositol Phosphates: Metabolism and Role as Second Messengers The PLC-mediated hydrolysis of the phosphoinositides causes the rapid production of a variety of inositol phosphates.377 Hydrolysis of PIP2 yields primarily (1,4,5)IP3 and lesser quantities of (1,2-cyclic 4,5)IP3.94'418 (1,4,5)IP3 is rapidly converted to either: (1) inositol 1,4-bisphosphate [(1,4)IP2] by a specific 5-phosphomonoesterase; or (2) inositol l,3,4,5-tetrakisphosphate [(l,3,4,5)IP4] by a specific 3- 53 kinase.84'418 This 3-kinase is stimulated by Ca++ in a calmodulin- dependent fashion.93'348 (l,3,4,5)IP4 is subsequently converted to inositol 1,3,4-trisphosphate [(l,3,4)IP3] by a specific 5- phosphomonoesterase.84 (1,3,4)IP3 is metabolized by phosphatases to yield either inositol 1,3 phosphate or inositol 3,4 phosphate.377 (1,2- cyclic 4,5)IP3 is also hydrolysed by S-phosphomonoesterase to yield inositol 1,2-cyclic bisphosphate; however, this reaction occurs at a 10- fold slower rate than that of the non-cyclic (1,4,5)IP3, leading to the relatively greater accumulation of (1,2-cyclic 4,5)IP3 over time.84'87-418 (1,2-cyclic 4,5)IP3 is not phosphorylated by 3-kinase to a cyclic 1P4 metabolite.8a The various isomers of inositol bisphosphate/inositol cyclic bisphosphate and inositol monophosphate/inositol cyclic monophosphate are produced by either the PLC-mediated hydrolysis of PIP or PI respectively or the sequential degradation of the polyphosphoinositides/ cyclic polyphosphoinositides. The metabolism of the inositol phosphates is very complex and involves a number of alternate pathways with enzymes differing in requirements for Mg++ and sensitivity to lithium inhibition.377 The inositol cyclic phosphates are metabolized via pathways separate from the non-cyclic inositol phosphates. The two pathways converge when inositol cyclic 1,2 monophosphate is metabolized to inositol non-cyclic monophosphate by a hydrolase. The final step in both metabolic pathways is the dephosphorylation of inositol monophosphate to yield free inositol which recycled for use in the cell.377 54 (1,4,5)IP3 is a water-soluble molecule that is released into the cytoplasm and implicated in the release of intracellular Ca++ from the platelet DTS. The addition of (1,4,5)IP3 to either permeabilized platelets or platelet membrane vesicles causes the rapid release of non- mitochondrial stores of Ca""".11'12’55'113'242 A monoclonal antibody against 1P3 blocks the thrombin-induced release of Ca++ from membrane vesicles, supporting a role for (1,4,5)IP3 in Ca++ release. Although the mechanism by which (1,4,5)IP3 causes Ca++ mobilization is not known, it has been proposed that it binds to specific sites on the DTS and activates a Ca++ channel.377 This IP3-gated Ca++ channel is distinct from the Ca++-ATPase pump2 and is not dependent on ATP or phosphorylation of membrane proteins. A recent study, which characterized the (1,4,5)IP3 binding site on platelet membranes using radiolabelled 1P3, has shown that Ca++ release could be blocked without inhibiting 1P3 binding. This suggests that the IP3-gated Ca++ channel has a region regulating Ca++ release that is distinct from the binding site.292 (1,4,5)IP3 also stimulates Ca++ release from plasma membrane vesicles, suggesting an additional role for (1,4,5)IP3 in facilitating Ca++ influx from the external environment.328 Numerous other inositol phosphate metabolites are formed during platelet activation. (1,2-cyclic 4,5)IP3 causes Ca++ mobilization in permeabilized platelets, although it is less effective than the non- cyclic (1,4,5)IP3.462 Since the relative proportion of (1,2-cyclic 4,5)IP3 increases with time, it has been proposed that (1,4,5)IP3 is more important in early Ca++-mobilization in platelet activation while (1,2-cyclic 4,5)IP3 becomes more important in late Ca++ mobilization.418 55 One study which compared intracellular [Ca++]i with inositol phosphate production has suggested that: (l) (1,4,5)IP3 alone cannot account for the sustained elevation in Ca++ seen following agonist stimulation; and (2) other metabolites ,such as (1,3,4)IP3 or (l,3,4,5)IP4, must also play a role in Ca++ mobilization.94 Numerous reports have proposed a role for (l,3,4,5)IP4 in the influx of external Ca++ in other cell types; however, its role in platelets is not known.306 The majority of 1P3 produced during platelet activation is the (l,3,4)-IP3 isomer. In other cells types, (1,3,4)IP3 has been shown to cause Ca++ release from internal stores, though it appears to be 30-fold less effective than (1,4,5)IP3. It's role in platelet Ca++ mobilization is not known.418 The addition of (1,4,5)IP3 to permeabilized platelets causes shape change, aggregation and secretion. The addition of cyclooxygenase inhibitors prevents aggregation and secretion, but not shape change. This suggests that Ca++ mobilization induced by (1,4,5)IP3 is unable to induce secretion and aggregation by itself. These effects appear to be partially mediated by the production of endoperoxide metabolites by the Ca++-sensitive PLA2.12 Production of Diacylglycerol and Its Role as a Second Messenger 1,2-Diacylglycerol is a neutral lipid that is rich in stearic and arachidonic acids. Only DC with the 1,2-sn-configuration plays a role in platelet activation. 1,2-DC remains in the membrane where it activates the enzyme PRC.377 56 As discussed above, the initial response of agonist-induced activation of PLC is the preferential hydrolysis of PIP2 and the production of small quantities of 1,4,5)IP3 and 1,2—DC. (1,4,5)IP3 is released into the cytosol where it mediates the release of internal stores of Ca++. Activation of additional Ca++-sensitive PLC hydrolyses both polyphosphoinositides and phosphoinositide. While the quantity of (1,4,5)IP3 released from the hydrolysis of PIP2 is sufficient to mobilize Ca++, the majority of 1,2-DC production comes directly from the hydrolysis of PI rather than from conversion of P1 to PIP2 with subsequent hydrolysis of the PIP2.435'463 Other potential sources for 1,2-DC production in stimulated platelets include the metabolism of triglycerides, PA, phospholipids other than PI and monoglycerides; however, these sources do not contribute a significant quantity of DC to the overall platelet response.342 The concentration of 1,2-DC is low in resting platelets.342 Following stimulation, 1,2-DC levels increase 2-3 fold by 30 seconds and gradually decrease to basal levels by 5 minutes. 1,2-DC production can be detected as early as 5 seconds.318 Platelets contain two different pathways to assure that DC does not accumulate. The majority of the DC is rapidly phosphorylated by diacylglycerol kinase to PA. The remainder is metabolized by the sequential action of diglyceride and monoglyceride lipases to yield arachidonic acid and a glycerol backbone.“0'l‘1v76 57 Role of PA and LysoPA as Second Messengers Phosphatidic acid is generated from the phosphorylation of DC by diacylglycerol kinase. One study which compared the time relationship between Ca++ mobilization and the production of PA has suggested that PA acts as a Ca++ ionophore.179 The addition of a synthetic diacylglycerol, l-oleoyl-2-acetoyl glycerol (OAC), to permeabilized platelets causes the release of Ca++ from the DTS.56'65 This release is not affected by cyclooxygenase inhibition and not duplicated by specific activators of PKC. It has been suggested that the increased [Ca++]1 is due to the formation of PA or lysoPA from OAC and that these metabolites enhance Ca++ mobilization in response to (1,4,5)IP3.56 The addition of high concentrations of PA to intact platelets stimulates platelet aggregation and enhances thrombin-induced aggregation and secretion. PA-induced platelet activation is inhibited by cyclooxygenase blockers, indicating that these effects are mediated by cyclooxygenase products and that PA acts to enhance the activity of PLA2 by increasing [Ca++]1.214 LysoPA is formed from the deacylation of PA by a specific PLA2. Since PA frequently contains arachidonic acid at the sn-2 position, its deacylation to lysoPA serves as a source of arachidonic acid.38 LysoPA acts as a potent Ca++ ionophore 1n_21;;n.141 When added to intact platelets, lysoPA induces aggregation, secretion, and protein phosphorylation.140 It causes platelet activation in a manner similar to calcium ionophore A23187. There is initial Ca++ mobilization which causes subsequent aggregation, inositol phosphate production and protein 58 phosphorylation. Cyclooxygenase inhibition decreases the aggregation response and blocks inositol phosphate production, suggesting that PLC activation by lysoPA is dependent on the Ca++-dependent activation of PLA2 and production of arachidonic acid metabolites.444 08++ AND ITS ROLE IN SIGNAL TRANSDUCTION Ca++ plays a pivotal role in many aspects of platelet activation. Extracellular Ca++ is required to maintain the heterodimer GPIIb-IIIa complex on the platelet surface. This complex is essential for fibrinogen binding and aggregation.63 Intracellular Ca++ fluxes are important in the platelet responses of shape change and secretion as already discussed. Platelets contain approximately 20nmol of Ca'H'/108 cells. It is primarily located in intracellular organelles or bound to membranes or cytoplasmic proteins.62 Platelets utilize various mechanisms to actively maintain the resting [Ca++]1 at approximately lOOnM, some 10,000 fold less than the surrounding plasma.51 The agonist-stimulated increase in [Ca++]1 results from release of internal stores or from influx of external Ca++. Platelet Storage Sites for Ca++ Non-exchangeable Ca++ Pools The vast majority of platelet Ca++ is stored within the dense granules. This Ca++ is released into the external environment following appropriate stimulation. The quantity of Ca++ stored in, and 59 subsequently secreted from, the dense granules is species-dependent.266 This pool of Ca++ is poorly exchangeable with the other pools of intracellular Ca++ or extracellular Ca++ in the resting platelet.62'273 Exchangeable Ca++ Pools Unstimulated platelets contain two exchangeable pools of Ca++. One is a slowly exchangeable pool that contains 54% of the exchangeable Ca++, is not affected by incubation with the Ca++ chelator ethylene glycol tetraacetic acid (EGTA) and is most likely intracellular. The second pool contains 46% of the exchangeable Ca++, is removed by incubation with EGTA and is most likely bound directly to or near the platelet surface.59 Further examination of the surface-bound pool of Ca++ has shown that the surface of unstimulated platelets contain two classes of binding sites: a high affinity class with approximately 57,000 sites and a low affinity class with approximately 460,000 sites.59 Platelets from patients with Clanzmann's thrombasthenia have reduced numbers of both type of bindings sites, supporting the role of GPIIb-IIIa as a major surface Ca++-binding site on unstimulated platelets.60 Following stimulation with ADP or epinephrine, there is an increase in the amount of Ca++ bound to the platelet surface and examination of the membrane demonstrates the appearance of additional binding sites. These sites develop in both normal and Clanzmann's thrombasthenic platelets, suggesting that these new sites are not located on GPIIb-IIIa but represent some other protein or glycoprotein.59'6O In spite of the 60 presence of lmM external Mg++, removal of the Ca++ from the platelet surface causes decreased aggregation and secretion in response to weak agonists. This indicates that external Ca++ bound to GPIIb/IIIa is required for maximal responses to stimulation with weak agonists.6o In resting platelets, the intracellular pool of exchangeable Ca++ behaves as if it is present in two distinct sites: (1) a rapidly exchangeable pool (t1/2-17 minutes) located in the cytosol; and (2) a slowly exchangeable pool (t1/2- 300 minutes) located in the mitochondria and the DTS. The size of the rapidly exchangeable pool is independent of the external [Ca++]. The size of the slowly exchangeable pool varies with the external [Ca++]. Studies using an inhibitor of mitochondrial Ca++ uptake have shown no alteration in the size of the slowly exchanging pool, implicating the DTS as the primary site of the slowly exchangeable pool.51 Maintenance of Resting [Ca++] Resting platelets maintain an average [Ca++]1 of approximately lOOnM. Examination of single resting platelets show that the internal Ca++ is located in non-homogeneous zones within the cytoplasm, with a continuous Ca++-gradient of increasing concentrations approaching the plasma membrane.428 Platelets maintain these resting levels by limiting influx, stimulating sequestration into the DTS and promoting efflux.51 61 Ca++ Influx The rate of Ca++ influx into non-stimulated platelets is maximal at an external Ca++ concentration 20-fold less than that present in normal plasma.51 Ca++ influx occurs through Ca++-selective channels by passive movement down a concentration gradient.51 Examination of unstimulated platelets from patients with Clanzmann's thrombasthenia has shown a decreased rate of Ca++ influx, without any significant effect on the total size of the cytosolic exchangeable pool of Ca++ or on [Ca++]1. This implicates the GPIIb-IIIa complex in passive Ca++ influx. Whether the CPIIb-IIIa complex itself serves as the channel or whether the complex is needed for the functioning of a distinct Ca++ channel in intact platelets has not been determined;52 however, studies using GPIIb-IIIa incorporated into phospholipid vesicles have shown that GPIIb-IIIa acts as a passive Ca++ channel.347 Ca++ Efflux The control of Ca++ efflux in platelets is not clear. In the resting platelet, Ca++ efflux is not directly linked to Ca++ influx and does not occur by simple Ca++/Ca++ exchange. Ca++ efflux is: (1) dependent on a source of energy; (2) stimulated by calmodulin; and (3) competitively inhibited by the cations Gd+++ and La+++.51 In most cells, Ca++ efflux across the plasma membrane is controlled by a Ca++- ATPase pump and/or a NaI/Ca++ exchange mechanism which exchanges extracellular Na+ for intracellular Ca++. The driving force for this exchange is the inwardly directed electrochemical Na+ gradient which is 62 maintained by an energy dependent Na+-K+-ATPase on the plasma membrane.365 There is question over the existence of a Na+/Ca++ in intact platelets, though one study did show the presence of a Na+/Ca++ exchanger in platelet plasma membrane vesicles.329 In intact platelets, an increase in cytoplasmic Na+ and/or a decrease in external Na+ causes an increase in [Ca++]i36S or in the quantity of Ca++ in the intracellular exchangeable pools,54 indicating that the membrane Na+ gradient does affect resting Ca++ homeostasis. This Na+/ Ca++ exchanger appears to be relatively dormant in the resting platelet, but may be activated by increased [Ca++]1.36S There is also controversy over the location of the Ca++-ATPase pump in platelets. Early studies detected Ca++-ATPase activity only in internal membrane fractions associated with the DTS and/or the OCS;107'111'166-265 however, recent reports have demonstrated the presence of two distinct platelet Ca++-ATPases. One of the Ca++ pumps appears to be located in the internal membranes while the other appears to be located in the plasma membranes.11“'115v116 Ca++ Sequestration Numerous studies have demonstrated the presence of Ca++oATPase activity in the membranes of the platelet DTSllli265 and/or within specialized structures associated with the OCS which take up and release Ca++ in a manner similar to calciosomes in non-muscle tissues.107'166 The platelet Ca++-ATPase identified on the internal membranes is similar to the ATPase in the sarcoplasmic reticulum of muscle cells, although there are some functional and structural differences.100'102-107v149 It requires Mg++, membrane phospholipids and ATP for optimum function and 63 it is stimulated by increasing [Ca++] up to 50uM. Ca++—ATPase-driven Ca++ sequestration within the DTS is stimulated by a number of processes including calmodulin,102'356 increased cytosolic cAMPl‘B'lzo'167 and PKC activation.471'472'473 Inhibition of Ca++-ATPase activity occurs following agonist-induced stimulation of PLC and the subsequent hydrolysis of membrane PIPZ. The mechanism of this inhibition has not been determined.330 Agonist-Stimulated Ca++ Fluxes A variety of agonists, including thrombin, PAF, vasopressin, endoperoxides and ADP, stimulate an increase in [Ca++]1.331 Historically, a number of techniques have been used to measure Ca++ flux in platelets. Currently, the most common techniques used to measure platelet [Ca++]1 include: (1) the fluorescent Ca++ chelator dyes fura 2, indo l and quin2; and (2) the luminescent photoprotein aequorin. The fluorescent dyes are believed to be more indicative of the average [Ca++]1 while aequorin is believed to respond to local zones of Ca++ flux. 79 ,97 , 188 Agonist-stimulated elevations in [Ca++]i arise from increased influx across the plasma membrane, release of Ca++ bound to the plasma membrane or release of Ca++ from the DTS. The internal Ca++ flux measured in the presence of external Ca++ is composed of both internal mobilization and influx from the external environment. The internal Ca++ flux measured in the absence of external Ca++ represents Ca++ release from internal stores. It should be kept in mind that the 64 absence of external Ca++ may affect the coupling of the agonist/receptor to the internal Ca++ discharge mechanism, secondarily inhibiting the release of internal stores of Ca++.331 When Ca++ is absent from the external media, the agonist-induced increase in [Ca++]i is attenuated and transient when compared to the change in [Ca following stimulation in the presence of external H] i Ca""".154'155'192'2“3'313'331 When platelets are stimulated in the presence of external Ca++, the peak [Ca++]1 is generally larger and prolonged in nature. This indicates a role for the influx of external Ca++ in sustaining the rise in [Ca++]i.313 The use of stop-flow cytometry in combination with the fluorescent dyes to measure [Ca++]1 have enabled researchers to study the sub-second kinetics of Ca++ flux in stimulated platelets. In the presence of external Ca++, stimulation of platelets by thrombin, PAF, vasopressin, or the TxA2 mimetic U46619 causes an increase in [Ca++]1 which is_ delayed in onset by 200-400ms. In the presence of EGTA, which binds external Ca++, this response is delayed by an additional 60-100ms, implying that the influx of external Ca++ precedes the release of Ca++ from internal stores and is not dependent on intracellular Ca++ release. In addition, the delay between agonist addition and the initiation of a Ca++ flux indicates that there are one or more biochemical steps between agonist binding and the release and/or influx of Ca++. These biochemical steps most likely involve the production of intracellular second messengers. Contrary to the other agonists tested, ADP stimulation causes an increase in Ca++ which is delayed in onset by only 65 20ms in the presence of external Ca++. When EGTA is present in the buffer, ADP causes Ca++ fluxes which are similar to those initiated by the other agonists in similar conditions,351'352 suggesting a similar mechanism(s) for the release of internal Ca++. The lack of significant delay between ADP-binding and Ca++ influx in the presence of external Ca++ suggests that the ADP receptor is closely associated with or coupled to the plasma membrane Ca++ channel, possibly via a G protein- mediated mechanism.331'352 A role for an, as yet unidentified, G protein in regulating influx through a plasma membrane Ca++ channel is also suggested by studies using Al+++ and Fl'l. These substances, which directly stimulate G proteins, cause an elevation in [Ca++]1 when added to platelets.331 The mechanism for the agonist-induced Ca++ influx across the plasma membrane is unclear. There are two major types of calcium channels in the membranes of cells: (1) voltage-dependent channels; and (2) receptor-operated channels. Studies using organic Ca++ channel antagonists and/or membrane depolarization agents have shown that voltage-dependent Ca++ channels do not play a role in the agonist- induced Ca++ influx,1°9'243'349v477 leading to the conclusion that the Ca++ influx is mediated by receptor-operated channels.1£"“2'151"192 A role for second messengers is indicated by the lag period in the kinetic studies discussed above. Second messengers implicated in cells other than platelets include (1,4,5)IP3 and (1,3,4,5)1P4; however, kinetic studies have shown that influx precedes internal release. This argues against a role for 1P4 which is formed after the release of (1,4,5)IP3.352 (1,4,5)IP3 causes the release of Ca++ from platelet 66 plasma membrane vesicles, suggesting a role in Ca++ influx.328 PIPZ may also mediate the influx of Ca++ through the receptor-operated channels, since it is more potent than (1,4,5)IP3 in inducing Ca++ mobilization from platelet microsomes.2“3'289 The plasma membrane receptor-operated Ca++ channel may be closely associated with the GPIIb-IlIa complex. Monoclonal antibodies against the CPIIb-IIIa complex inhibit Ca++ influx, but not internal mobilization, following stimulation with weak agonists; however, Ca++ influx in response to strong agonists is not affected. It has been proposed that the antibody causes stearic hindrance of an adjacent Ca++ channel since platelets from patients with Clanzmann's thrombasthenia, which are lacking the complex, have normal Ca++ fluxes. This Ca++ channel is important in influx in response to weak agonists while strong agonists appear to stimulate influx via a different route.316 It has also been proposed that it is the binding of fibrinogen to the CPIIb- IIIa complex or the incorporation of the GPIIb-IIIa complex into the cytoskeleton which opens a pathway for Ca++ influx and is blocked by the monoclonal antibody.469 As discussed in the section on the inositol phosphates, (1,4,5)IP3 and (1,2 cyclic 4,5)IP3 are major stimuli for the receptor-mediated release of Ca++ from the DTS.331 Intracellular Ca++ may also be released from the membranes themselves. Studies have shown that the hydrolysis of membrane PIP2 correlates with a decrease in chlortetracycline fluorescence in stimulated platelets, indicating a release of membrane-bound Ca++;64 however, it is possible that in the 67 physiological state the Ca++ binding sites on PIP2 are actually occupied by Mg++ which is present in relatively higher concentrations in the platelet cytoplasm.377 Ca++-Dependent Processes During Platelet Activation Ca++ plays a role in a number of effector pathways during platelet activation. These effector pathways include: (1) Ca++-calmodulin dependent protein kinase; (2) Ca++-dependent proteases; and (3) the enzymes PLC, PLA2 and PKC.212 PLC and PLA2 have been discussed above. PKC will discussed in a later section. Intracellular Ca++ also plays an important part in the organization of cytoskeletal proteins.377 Ca++-Calmodulin Dependent Protein Kinase Calmodulin, a Ca++-binding protein present in high concentrations in the platelet cytoplasm, is a primary intracellular Ca++ receptor. Once formed, the Ca++-calmodulin complex interacts with a variety of calmodulin-binding proteins which then modulate the cellular response. Ten different calmodulin binding proteins have been identified in platelet preparations and include MLCK, phosphorylase kinase, caldesmon, calmodulin-dependent kinase and phosphoprotein phosphatase.436 The role of many of these proteins in platelet activation is not known. MLCK is an enzyme whose activity is regulated by the Ca++- calmodulin complex.96'142'163'164 Activation of MLCK causes the phosphorylation of the 20kDa myosin light chain.177v353 The initiation 68 of shape change correlates with phosphorylation of this 20kDa protein.95 Myosin phosphorylation is associated with an increase in the actin- activated ATPase activity, the association of myosin with the cytoskeleton and the assembly of myosin into filaments.377 The association of myosin with actin and the activation of the ATPase leads to the generation of a contractile force which an important role in the centralization of cytoplasmic granules, secretion and clot retraction.142 MLCK activity is modulated by substances other than the Ca++- calmodulin complex. Increased cAMP causes the phosphorylation of MLCK which, in the absence of Ca++-calmodulin, inhibits MLCK activity and decreases phosphorylation of the myosin light chain.88 In the absence of the Ca++-calmodulin complex, PRC phosphorylates MLCK at two different sites and inhibits subsequent calmodulin binding.178 The physiologic significance of this finding is not clear since it has not been demonstrated to occur in intact platelets.377 CalciumaDependent Protease Platelets contain two types of Ca++-dependent neutral proteases, more commonly known as calpains, which have similar substrate specificities but different Ca++ requirements. Calpain I hydrolyses its substrates at Ca++ concentrations of only 1 to lOuM while calpain 11 requires Ca++ concentrations of 0.1 to 1mM.366 Calpain I is an 80kDa protein which remains intracellular during platelet activation. Upon platelet aggregation, it is cleaved to a 76-78kDa protein which appears 69 to be the active form.357 Calpain II is associated with the intracellular membranes in the resting state. Following activation, it becomes exposed on the external portion of the membrane.366 The role of Ca++-dependent protease in platelet activation is not clear. Calpains are activated when platelet aggregation is stimulated by a variety of agonists. If aggregation is prevented by not stirring the platelet suspension, the protease is not activated even though the platelets are still activated.134 Ca++-dependent protease hydrolyses actin binding protein and P235, permitting reorganization of the platelet cytoskeleton during platelet aggregation.132 Other studies have shown that calpain activation may occur without aggregation, although calpain activation does not induce platelet secretion and aggregation.110 Calpain may also have a role in the exposure of fibrinogen binding sites. One recent report has shown that calpain cleaves aggregin in thrombin-stimulated platelets, allowing exposure of the fibrinogen binding site.321 Another study utilizing a calpain inhibitor has suggested that calpain is involved in the activation of PLC and TxAz synthetase.426 Opposite results were demonstrated in a different study using calpain inhibitors in which calpain had no effect on PLC or adenylate cyclase activity.61 70 Role of Ca++ in the Cytoskeleton Ca++ has important, though sometimes opposing, roles in the organization and composition of the platelet cytoskeleton.377 In resting platelets, the majority of the actin is maintained as monomers by association with the protein profilin. The interaction between actin and profilin is stabilized by Ca++.235 Gelosin is another protein which interacts with actin to regulate actin filament length. In the resting platelet, gelosin is poorly complexed to actin; however, when [Ca++]i increases during activation, gelosin tightly complexes to actin. When gelosin is added to preformed actin filaments, the association of gelosin with actin causes the shortening of filaments. When added to actin monomers, the association of gelosin with actin increases the rate of formation of actin filaments by stabilizing the small oligomeric nuclei.234 Since nuclei formation is the rate limiting step in the formation of actin filaments,377 the increasing interaction of gelosin and actin during periods of elevated [Ca++]1 may enhance the actin polymerization that occurs during platelet activation.216 The organization and structure of the cytoskeleton in resting and stimulated platelets is partially controlled by the Ca++-dependent interaction of various proteins (ie. actin binding protein, a -actinin and P 235) with actin.377 The role of calpain and MLCK in cytoskeletal organization were discussed above. 71 THE ROLE OF PROTEIN KINASE C IN PLATELET ACTIVATION Protein Kinase C Protein kinase C is a Ca++-dependent enzyme which requires the phospholipid phosphatidylserine for its activity.194 In the resting cell, the majority of PKC is located in an inactive form in the cytoplasm. Following cell activation, most of the PRO is transferred from the cytoplasm to the membrane.208'465 The binding of PKC to the membrane is inhibited by the removal of Ca++ 355 and appears to be due to the synergistic action of both Ca++ and l,2-DG.380’465 Although enzyme translocation from the cytoplasm to the membrane accompanies platelet activation, studies have shown that PKC activation can occur without translocation. The role of the cytoplasmic, activated PKC has not been determined.11‘7v301 Several species of PKC have been identified in mammalian tissue. To date, two subtypes PKG have been identified in human platelets.427'l“‘1 The role of the different subtypes in platelet function is not known. Diacylglycerol and Phorbol Activation of PKC In platelets, 1,2-DC activates PKC by increasing the affinity of PKC for Ca++, allowing enzyme activation at Ca++ concentrations present in resting cells.193’414 It has been proposed that phosphatidylserine binds Ca++ and subsequently creates a surface in the membrane for PKC 72 binding. The addition of sn-l,2-DG then causes a change, possibly conformational, in PRC, leading to reversible enzyme activation.138 lnnxigzg, PKC can be activated by synthetic diacylglycerols and phorbol esters independently of PLC-induced formation of 1,2-DC. Synthetic diacylglycerols are chemically modified sn-l,2-diacy1glycerols which intercalate into the membranes of intact platelets to activate PKC.194'224 Phorbol esters appear to have a similar structure to the diacylglycerols and to be able to bind PKC, causing activation.72 Although phorbol ester- and synthetic DC-induced activation of PKC are frequently used in vitgo to simulate the in vivn activation of PKC, there are some significant differences. First, phorbol esters cause permanent activation of PKC377 by inducing the proteolytic cleavage of the 80kDa PKG into an active SlkDa protein enzyme which is Ca++- and phospholipid-independent.l‘16'417 Secondly, the synthetic diacylglycerols are more dependent on secreted ADP for platelet activation. The addition of apyrase to synthetic DG-stimulated platelets inhibits aggregation and protein phosphorylation. This inhibition is significantly less in phorbol stimulated platelets.10 Thirdly, it has been proposed that phorbol esters primarily cause platelet aggregation by a direct effect on altering platelet membranes rather than by activating PKC.187 Finally, the addition of synthetic diacylglycerols, but not phorbol, to platelets is associated with an increased [Ca++]1 due to its metabolism to the second messengers PA and lysoPA.56 73 Phosphorylation of Proteins by PKC In platelets, the major substrate of PKC that has been identified is a 47kDa cytoplasmic protein.362 In resting platelets, this protein consists of two or three poorly phosphorylated components as identified by two-dimensional electrophoresis and isoelectric focusing. Following stimulation, seven to nine phosphorylated components have been identified.180:239 Though traditionally associated with secretion, a variety of specific functions for the phosphorylated 47kDa protein have been proposed. Purified 47kDa from unstimulated platelets inhibits actin polymerization while phosphorylated 47kDa protein from stimulated platelets has no inhibitory effects. This suggests a role for the 47kDa protein in the cytoskeletal reorganization that accompanies platelet activation.158 PKC activation is associated with actin incorporation into the cytoskeleton and the subsequent formation of pseudopods. Another proposed role for the 47kDa protein is that of a lipocortin. Lipocortins are a group of proteins which inhibit PLAZ. It has been suggested that phosphorylation of the 47kDa protein suppresses its anti-PLA2 activity, allowing PLA2 activation to occur.424 Recent studies have shown that gamma thrombin causes phosphorylation of the 47kDa protein without the measurable production of arachidonic acid that would be expected if PLAQ activation had occurred.91 This suggests that substances other than the 47kDa protein control PLA2 activity. 74 A third function proposed for the 47kDa protein is that of 5- phosphomonoesterase, an enzyme that cleaves the 5 position phosphorous from the Ca++-mobilizing (1,4,5)IP3, (1,2cyc1ic 4,5)IP3 and (1,3,4,5)IP4 to yield relatively inactive metabolites. Gel electrophoresis and peptide mapping comparisons of purified 5-phosphomonoesterase and the 47kDa protein of platelets indicate that they are identical proteins.85 It has been suggested that PKG—mediated phosphorylation of the 47kDa protein increases its activity as a 5-phosphomonoesterase, leading to the termination of the stimulatory Ca++ signal generated by the inositol phosphates;85 however, there has been no demonstration of enzymatic activity by purified platelet 47kDa protein.377 In addition, inhibition of PKC activation by staurosporine blocks thrombin-stimulated 47kDa phosphorylation without causing measurable alteration in the production of the inositol phosphates.443 substrates for PKC other than the 47kDa protein have been identified in platelets. PKC phosphorylates the light (20kDa) chain of myosin in phorbol ester-stimulated platelets199 at a site distinct from that phosphorylated by the CaI+-calmodulin dependent MLCK.275 Thrombin- stimulated platelets phosphorylate myosin at two separate sites. One site is phosphorylated rapidly and is mediated by MLCK. The second site is more slowly phosphorylated by PKG.276 The physiologic significance of this PKG-mediated phosphorylation of myosin has yet to be determined. Phorbol treatment of platelets is also associated with phosphorylation of three high molecular weight cytoskeletal proteins: (1) actin binding protein; (2) talin (P235); and (3) myosin heavy chain. This supports a 75 role for PKC in the regulation of contractile or cytoskeletal events in platelet activation.236 The Stimulatory Effects of PKC in Platelet Activation The role of PKC in platelet activation is very complex and unclear since PKC has been reported to have both positive and negative feedback effects.377 Numerous reports have demonstrated that Ca++ and PKC activation act synergistically to induce full platelet activation.139'll‘2'194 In aspirin-treated platelet rich plasma, the addition of phorbol esters 10-15 seconds prior to the addition of a variety of agonists (including epinephrine, ADP, U46609, collagen, PAF or vasopressin) potentiates both secretion and aggregation.381 The combination of subthreshold concentrations of synthetic diacylglycerol and calcium ionophore induces secretion.194 Finally, thrombin-induced aggregation and secretion are potentiated by subthreshold concentrations of phorbol ester."70 Low concentrations of a variety of agonists which mobilize Ca"'+ also potentiate phorbol-induced aggregation and secretion. It has been suggested that the initial mobilization of Ca++ causes the translocation of PKC to the membrane where it is primed for activation by phorbol ester or 1.2-DG,330 The mechanism by which PKC potentiates platelet activation is not clear. PKC causes alterations in the membrane GPIIb-IIIa complex, leading to exposure of the fibrinogen binding site and subsequent fibrinogen binding.371 PKC also activates Na+/H+ exchange, causing 76 cytoplasmic alkalinization which then potentiates platelet activation.386 PKC activation enhances the Ca++-dependent arachidonic acid release following stimulation with calcium ionophore A23187.6'122'150'152 Addition of phorbol ester by itself does not cause any measurable release of arachidonic acid. Phorbol esters and synthetic diacylglycerol, however, cause a dose- and time-dependent potentiation of arachidonic acid release in response to A23187, even in platelets treated with aspirin and creatine kinase/creatinine phosphate to prevent secondary PLC activation.122 This suggests that PKC promotes the Ca++-dependent PLA2 release of arachidonic acid by either increasing the enzyme sensitivity to Ca++ or making more phospholipid substrate and/or PLA2 available. One study has proposed that this potentiation is partly mediated by Na+/H+ exchange and cytosolic alkalinization6 while a different study has shown that Na+/H+ exchange is not involved.150 PKC may enhance arachidonic acid release through a G protein-mediated stimulation of PLA2.5'7 Phorbol-induced PKC activation inhibits both arachidonyl-CoA synthase and arachidonyl-CoA lysophosphatide acyltransferase, causing decreased incorporation of arachidonic acid into membrane phospholipids and increased availability of arachidonic acid for the production of endoperoxides and TxAz.136 The Inhibitory Effects of PKC on Platelet Activation PKC has numerous inhibitory functions in platelet activation.377 It has been proposed that the determination of whether the overall 77 effect of PKC activation is inhibitory or stimulatory depends on the type of agonist used and the incubation time.209 Secretion induced by all agonists tested is inhibited by long (5 minutes) periods of phorbol incubation. In non-aspirin treated platelet rich plasma, short (10 seconds) periods of pretreatment with phorbol ester inhibit TxA2 production and secretion in response to stimulation with weak agonists (ADP, epinephrine and PAF) which are partially dependent on TxAz production for full activation. It has been proposed that this effect is due to inhibition of the metabolism of arachidonic acid to TxAz. If TxA2 production is blocked by pretreatment of platelets with aspirin, phorbol pretreatment potentiates secretion in response to these same agonists in a TxAz-independent manner.209 PKC activation has multiple inhibitory effects on Ca'*'+ mobilization. Treatment of platelets with phorbol or 1,2-DC does not induce increased [Ca++]1 as measured using the fluorescent dye indicators (ie. quin2, fura 2 and indo 1); however, a Ca++ peak is detected using aequorin.437 This suggests that phorbol or 1,2-DC stimulates a localized zone of Ca++1 elevation, possibly associated with the plasma membrane. Pretreatment of platelets with phorbol esters inhibits elevations in [Ca++]1 induced by thrombin, ADP, epinephrine, TxA2 mimetic U46619, PAP and vasopressin.117izl‘o'2-79'371 Phorbol ester or diacylglycerol treatment of platelets inhibits influx through the plasma membrane receptor-operated Ca++ channel.13'311'433 Activation of PKC also inhibits the agonist-induced release of internal stores of Ca++, possibly by modulating the formation and/or degradation of (l,4,5)IP3.l‘33 Following stimulation with thrombin or ADP, activation 78 of PKC stimulates resequestration of the released Ca++. This sequestration occurs independent of any effect on cAMP or (1,4,S)IP3.471'472'473 Finally, PKC activation potentiates Ca++ efflux from the cell following stimulation with thrombin, possibly by stimulating the plasma membrane Ca++-ATPase pump.315’331 In addition to a negative effect on Ca++ mobilization, PKC activation inhibits phosphoinositide hydrolysis as measured by decreased production of PA, 1,2-DC or inositol phosphates in response to ADP, PAF, U46069, thrombin or collagen.39'380'421'442 One study has proposed that PKC inhibits the G protein coupled to PLC.5 By itself, stimulation with phorbol esters or synthetic diacylglycerol has no effect on platelet levels of inositol phosphates, indicating no direct effect of PKC on the hydrolysis of the phosphoinositides.442 Treatment with either phorbol or 1,2-DC actually increase the levels of the polyphosphoinositides, suggesting that PKC modulates the activity of the enzymes involved in the 'phosphoinositide cycle' (ie. the specific kinases or phosphatases).442 Finally, PMA inhibits the production of cAMP in response to PGEl, prostaglandin D2 and forskolin. PMA also blocks the inhibition of cAMP production by thrombin and epinephrine while it has no effect on ADP- induced inhibition of cAMP production. These findings suggest that the overall effect of PKC activation is inhibition of adenylate cyclase; however, the effects depend on the agonist used.46o Previous reports have shown that a primary role of PKC is to inhibit the agonist-induced 79 inhibition of adenylate cyclase by phosphorylating the inhibitory G protein (G1) associated with adenylate cyclase.28'196 THE ROLE or Na+/H"’ EXCHANGE IN CELL ACTIVATION Regulation of intracellular pH is vital to the proper functioning of all cells. The typical cell has a large interior-negative membrane potential, causing the passive influx of H+ ions into and/or efflux of negatively charged ions (ie OH' or HCO3') out of the cell. To maintain resting cytosolic pH (pHi), cells must have a mechanism(s) to actively extrude H+ or to actively accumulate HCO3’.246 A variety of cells respond to stimulation by increasing pHi. This also requires a process for actively extruding H+ and/or accumulating HCO3'. One mechanism involves anion channels which exchange Cl' for HCO3' in either a Na+— dependent or -independent manner. Relatively few investigations have examined the role of anion channels in platelets. Cl'/HCO3' exchange has been shown to play a role in the regulation of pHi by some investigators294 while others have found no effect of anion exchange on resting or stimulated pH1.77 In recent years, Na+/H+ exchange is the primary mechanism implicated in the modulation of platelet p111.386 Characteristics of Na+/H+ Exchange Although the principle function of Na+/H+ exchange is to regulate pHi, it has other important roles in cellular metabolism. The exchanger is involved in signal transduction in response to those agonists which affect cell function by altering pHi. Since the exchanger transports 80 Na+, it controls cell volume regulation. Na+/H+ exchange is also involved in the net transport of solutes across epithelia such as the renal proximal tubule or the small intestine.246 The stoichiometry of the Na+/ H+ antiport is 1:1. Since the drive for the transport of H+ out of the cell comes from an inwardly directed electrochemical Na+ gradient, the antiport is not directly dependent on the production of energy. The Na+ gradient must be maintained by an energy-requiring, plasma membrane Na+-K+-ATPase system.246 Na+/H+ exchange is electroneutral, independent of the membrane potential386 and fairly specific for Na+ and Li+. It is inhibited by amiloride or its derivatives.237 In vitro, this exchange is commonly inhibited using four different experimental manipulations: (l) the addition of amiloride derivatives to directly block the exchange; (2) replacement of external Na+ to eliminate the inwardly directed Na+ gradient; (3) increasing extracellular H+ concentration to block H+ efflux; and (4) treatment of cells with a Na+-KI-ATPase inhibitor to decrease the inwardly directed Na+ gradient. The Na+/H+ antiport is relatively quiescent in the resting platelet; however, addition of amiloride derivatives to non-stimulated platelets causes a slow decrease in pHi of 0.05.“75 This indicates that the antiport is responsible for the continuous removal of metabolically generated protons. The activity of the antiport is greatly increased when the cytoplasm is acidified.237 This suggests that the antiport has two binding sites: (1) a site that exchanges Na+ for H+; and (2) a site where H+ binding modifies the activity of the antiport.386 Stimulation 81 with either phorbol esters, 1,2-DC or thrombin causes an increase in pHi which requires external Na+ and is inhibited by amiloride derivatives or PKC inhibitors.387'391'392 The latter findings indicate that PKC also activates the antiport, possibly by altering the setpoint of the exchanger so that it responds to a more alkaline pHi by increasing its H+ extrusion.386 The antiport is activated by other agonists, such as ADP or epinephrine, in a process independent of PKC activation. Epinephrine causes a cytoplasmic alkalinization which is blocked by amiloride derivatives, suggesting that the increase in pHi is mediated by Na+/H+ exchange. This alkalinization is also inhibited by an antibody which blocks fibrinogen binding, indicating a dependence on fibrinogen receptor expression.20 Finally, preliminary findings in a recent study suggest that the Na+/H+ exchanger is regulated by two different G proteins: (1) a G protein stimulated by epinephrine which increases Na+/H+ exchange; and (2) a G protein stimulated by fluoride which inhibits Na+/H+ exchange.389 Na+/H+ Exchange and Platelet Activation Cell activation is often accompanied by increased pHi. Cytosolic alkalinization represents a balance between the metabolic generation of protons, the buffering capacity of the cytoplasm and the extrusion of H+ from the cell.“75 Platelet alkalinization and activation of Na+/H+ exchange occurs following stimulation with thrombin, PAF, ADP, epinephrine, vasopressin, TxA2 mimetic U46619 and calcium ionophore A23187.45'390'394'409'413'423'475 Artificial alkalinization of the platelet cytosol by use of the monovalent cation ionophores, monensin or 82 nigericin, or by addition of NHQCl does not, by itself, induce platelet activation as assessed by the measurement of Ca++ mobilization, PA production, aggregation or protein phosphorylation.385'398'47S Artificial alkalinization does sensitize platelets to subsequent activation by thrombin.384 This suggests that cytoplasmic alkalinization enhances the platelet responses to the subsequent addition of agonists. The addition of amiloride derivatives or the removal of external Na+ prevent Na+/H+ exchange. These manipulations also inhibit aggregation in response to ADP, PAF, collagen, phorbol esters, vasopressin and calcium ionophore A23187,“5'388'390'l‘23 suggesting that a functional Na+/H+ exchange mechanism is essential for full aggregation responses. Other groups dispute this requirement for Na+/H+ exchange. One study has shown that ADP-induced aggregation is inhibited only when using amiloride concentrations much higher than required for inhibition of the antiport. They have proposed that the inhibition demonstrated in the previous studies is due to secondary effects of amiloride.135 A second study has shown that amiloride derivatives inhibit agonist- induced aggregation in doses 100 times greater than those required for inhibition of the antiport, once again pointing to secondary inhibitory effects of amiloride.176 NaI/H+ Exchange and Ca++ Mobilization A considerable amount of information suggests that Ca++ mobilization and Na+/H+ exchange are interrelated. Increased [Ca++]1 83 precedes cytosolic alkalinization and may activate the antiport. First, the addition of the non-physiologic calcium ionophores, which increase membrane permeability and cause increased [Ca++]i, stimulates Na+/H+ exchange.390 Secondly, removal of Ca++ from the external media inhibits alkalinization in response to PAF stimulation. This decreased alkalinization corresponds with a decrease in [Ca++]i secondary to impaired Ca++ influx.“5 Finally, simultaneous measurements of pHi and Ca++ flux demonstrate that there is an initial decrease in pHi following stimulation and that the subsequent increase in pHi lags behind the increase in [Ca"""]i.99'398'47S In contrast, a recent study has shown that Ca++ mobilization occurs simultaneously with activation of the antiport. It was proposed that the initial decrease in measured pH is actually due to platelet shape change which interferes with the measurement of the fluorescence of the dye used to measure pHi.393 Other evidence suggests that activation of the Na+/H+ antiport modulates the Ca++ fluxes. Blockage of the antiport by the addition of amiloride derivatives or the removal of external Na+ markedly reduces the release of Ca++ from internal stores and the influx of external Ca++ in response to thrombin.1“5'38“'395 This inhibition is not due to decreased hydrolysis of the membrane phosphoinositides by PLC and can be overcome by artificial alkalinization of the cytosol.395 ca++ mobilization in response to ADP, U46619 and PAP is also partially dependent on activation of NaI/H+ exchange.45'394 It is not known how pHi affects Ca++ influx across the plasma membrane. (l,4,5)IP3-induced Ca++ release from internal stores is potentiated by an alkaline pH1.55 It has been prOposed that an alkaline pHi either enhances the binding of 84 (1,4,5)IP3 to its receptor or interferes with the breakdown of (1.4.5)IP3 to its inactive metabolites.386 In contrast, other studies have demonstrated that Ca++ mobilization occurs independent of Na+/H+ exchange and cytosolic alkalinization. Ca++ mobilization in response to thrombin occurs in spite of cytosolic acidification and/or inhibition of Na+/H+ exchange.176'358'398'474 Simultaneous flow cytometric measurements of thrombin-induced changes in cytosolic pHi and Ca++ flux indicate that the initial acidification phase correlates with the increasing [Ca++]i while the subsequent alkalinization appears to be independent of the Ca++ changes.99 Na+/H+ Exchange and Arachidonic Acid Metabolism Removal of external Na+ inhibits platelet aggregation and secretion following stimulation with the weak agonists ADP, epinephrine and low concentrations of thrombin. These agonists are dependent on the formation of arachidonic acid metabolites to induce secondary aggregation and secretion. These effects can be duplicated by treating platelets with aspirin to inhibit cyclooxygenase, suggesting a role for extraplatelet Na+ in the arachidonic acid release pathway.86 Blockage of Na+/H+ exchange by the addition of amiloride derivatives, the removal of extraplatelet Na+ or the acidification of external pH also inhibits arachidonic acid mobilization using these same agonists.412 Inhibition of NaI/H+ exchange blocks the activation of PLC by weak agonists but not by high concentrations of thrombin, indicating that PLC activity is not 85 directly affected by Na+/H+ exchange. Instead, it has been proposed that weak agonists cause Na+/H+ exchange-mediated activation of PLA2. PLA2 then mobilizes a very small pool of arachidonic acid which can only be detected using gas chromatography/mass spectrometry. The subsequent production of cyclooxygenase metabolites acts as an amplification loop to activate PLC.409 It has also been proposed that weak agonists stimulate PLA2 activity by a combination of local alkalinization, due to the activation of the Na+/H+ exchange, and the release of a small pool of membrane Ca'H'.411 In vigzg evaluation of PLA2 shows that enzyme activity is potentiated by increasing pH up to a maximum of 8.0 in the presence of ca”.27 Mobilization of arachidonic acid in response to PAF is also partially dependent on Na+/H+ exchange.413 Stimulation of phosphoinositide hydrolysis by epinephrine is blocked by either: (1) inhibition of PLA2 or cyclooxygenase activity; (2) inhibition of NaI/H+ exchange; or (3) prevention of fibrinogen binding by removal of fibrinogen, addition of a monoclonal antibody against GPIIb-IIIa or incubation with EDTA. Similarly, epinephrine- induced cytosolic alkalinization is inhibited not only by prevention of Na+/H+ exchange by amiloride derivatives, but also by a monoclonal antibody to the GPIIb-IIIa complex.20 These findings suggest that epinephrine stimulates PLC activity in a manner dependent on PLA2 activation, fibrinogen binding and cytosolic alkalinization. More importantly, these results indicate that weak agonists initiate exposure of fibrinogen binding sites and subsequent fibrinogen binding concomitant with activation of Na+/H+. The mechanism by which fibrinogen binding mediates cytosolic alkalinization is not clear. Some 86 data suggests that the az-adrenergic receptor is closely associated with, if not actually a part of, the Na+/H+ exchanger.410 It has been proposed that binding of agonist to the adrenergic receptors causes a local alkalinization but that the production of measurable alkalinization also requires fibrinogen binding.410 Summary of Na+/H+ Exchange There are at least two pathways for agonist-induced activation of Na+/H+ exchange in platelets. The first occurs following the addition of strong agonists such as thrombin. The binding of agonist to a receptor activates PLC and stimulates the production of the second messengers (1,4,5)IP3 and 1,2-DC. (1,4,5)IP3 mobilizes Ca"'+ while 1,2- DC activates PKC which, in turn, stimulates Na+/H+ exchange. The second pathway occurs following activation by weak agonists such as epinephrine. The binding of agonist activates Na+/H+ exchange by some undefined mechanism that also requires fibrinogen binding. This local alkalinization, in combination with a local Ca+I elevation, activates PLA2. The release of prostaglandin endoperoxides and TxA2 acts as a positive feedback loop to activate PLC and stimulate subsequent platelet activation.386'410 While strong agonists cause the greatest activation of Na+/H+ exchange, cytosolic alkalinization is not essential for platelet activation to occur in response to these agonists. The increase in pHi just enhances the process of signal transduction. This enhancement of signal processing is much more important in the response of platelets to weak agonists like ADP or epinephrine.386'410 87 SUMMARY OF THE EFFECTS OF INDIVIDUAL PLATELET AGONISTS Platelets respond to a large number of physiologic and non- physiologic agonists. The following summary will discuss individually the agonists used in the research for this dissertation. It should be kept in mind that in vivn, platelets are stimulated by multiple agonists simultaneously. Numerous studies have demonstrated synergistic responses of different platelet agonists. The addition of subthreshold doses of agonists which do not cause platelet activation alone can induce the full platelet response when added in combination.377 Agonists are divided into three groups depending on their ability to cause release and irreversible aggregation. Weak agonists (ie. epinephrine and ADP) require the release of cyclooxygenase metabolites and primary aggregation to induce irreversible aggregation and secretion. Secretion is almost completely inhibited by cyclooxygenase blockers. Agonists of intermediate strength (ie. PAF, vasopressin, prostaglandin endoperoxides/TXA2 mimetics) induce secretion independent of prior aggregation or release of arachidonate metabolites. Treatment of platelets with cyclooxygenase inhibitors or agents which remove extracellular ADP reduce secretion and aggregation. Strong agonists (ie. high concentrations of thrombin) stimulate secretion and aggregation in a manner that is not affected by cyclooxygenase inhibitors.377 Low concentrations of strong agonists act like intermediate or weak agonists, but high concentrations of weak or intermediate agonists do not act like strong agonists. 88 Adenosine Diphosphate (ADP) ADP is considered a weak physiologic agonist. The binding of ADP to specific platelet receptors has two separate effects: (1) inhibition of the activation of adenylate cyclase by stimulatory agonists; and (2) stimulation of shape change, fibrinogen binding site exposure and aggregation.377 Studies using 5'-p-fluorosulfonylbenzoyl adenosine (FSBA), an affinity analogue of ADP, have shown that ADP covalently binds to a lOOkDa protein, called aggregin, on the plasma surface.81v82'83 This binding is associated with inhibition of shape change and aggregation but not inhibition of adenylate cyclase activity.123'32a These findings suggest that there are two distinct membrane receptors for ADP, one coupled to adenylate cyclase and the other coupled to the stimulation of aggregation and shape change. Two separate receptors have yet to be identified. It has been proposed that aggregin sterically inhibits the fibrinogen binding site on the membrane GPIIb/IIIa complex. Binding of ADP to this receptor causes a conformational change in aggregin which permits exposure of the fibrinogen binding site and subsequent fibrinogen binding to the GPIIb-IIIa complex.81'82'83 The use of FSBA to block ADP binding at the receptor level also blocks fibrinogen binding and aggregation in response to the prostaglandin endoperoxides or to epinephrine while shape change following stimulation with the endoperoxides is not affected. This suggests that small quantities of ADP are essential in epinephrine- and endoperoxide-stimulated platelets to allow exposure of the fibrinogen binding site.82 Fibrinogen binding 89 induced by high concentrations of thrombin is not affected by FSBA, indicating that thrombin uses a different mechanism for exposure of the fibrinogen binding sites.82 As discussed above, it has been proposed that ADP initiates platelet activation by stimulating the release of a small amount of arachidonic acid by a Na+/H+ exchange-dependent activation of PLA2. The released arachidonic acid is metabolized by cyclooxygenase to yield the prostaglandin endoperoxides and TxA2 which amplify the platelet response by subsequently activating PLC.409'411 Inhibition of NaTI/H+ exchange blocks the release of arachidonic acid but not the occurrence of primary aggregation.412 These findings support the theory that binding of ADP to aggregin exposes the fibrinogen binding site, an event essential for the process of primary aggregation. The process of secondary or irreversible aggregation is dependent on the generation of arachidonate metabolites in a Na+/H+ exchange-dependent manner. When platelets are treated with cyclooxygenase inhibitors to block the endoperoxide/TxA2-mediated activation of PLC, the addition of ADP induces a rapid increase in [Ca.++]1.92'12l"290’394 This indicates that Ca"'+ mobilization in response to ADP is partially independent of the release of arachidonate metabolites. Kinetic studies have shown that, in the presence of external Ca++, this mobilization occurs without the lag phase found following stimulation by other agonists.351'352 The rapid response negates the role of a receptorcmediated production of a diffusible second mediator to induce the Ca++ influx. Instead, the response suggests more direct coupling between the receptor and the 90 plasma membrane Ca++ channel, possible via a G protein-mediated mechanism.331 The increase in [Ca++]1 is partially inhibited when platelets are incubated in EGTA-buffer, indicating a role for the influx of external Ca++ as well as the release of internal stores of Ca++ in the overall Ca++ response. Kinetic studies have shown that, in the absence of external Ca++, ADP-stimulated platelets have a significant lag period between the addition of agonist and the initiation of a measurable change in [Ca++]1. The lag period suggests a role for the production of a diffusible second messenger in the internal Ca++ response.351'352 Ca++ mobilization in response to ADP is not affected by incubation with FSBA. Therefore, the receptor mediating Ca++ mobilization and adenylate cyclase activation by ADP is not aggregin, the receptor that regulates ADP-induced fibrinogen binding and aggregation.324 Whether or not ADP activates PLC independent of the generation of arachidonic acid metabolites is under debate. One study has demonstrated that ADP stimulation of cyclooxygenase-inhibited platelets mobilizes Ca++ without inducing phosphoinositide metabolism.12h A separate study has shown that cyclooxygenase-inhibition by aspirin blocks the production of the inositol phosphates in platelets stimulated by ADP. This finding suggests that PI hydrolysis is initiated secondary to the production of endoperoxides/TxAz and not directly by ADP.409 On the other hand, it has been demonstrated that ADP stimulation of cyclooxygenase-inhibited platelets does not induce measurable PI hydrolysis while it does cause increased production of (1,4,5)IP3 and 1,2-DC. The lack of measurable PI metabolism is attributed to a low 91 level of PLC activation, allowing PIP2 resynthesis from PIP to compensate for the PIP2 hydrolyzed by PLC.92 Epinephrine Epinephrine is a physiologic stimulant which is classified as a weak agonist. It has two separate effects on platelet responses: (1) induction of aggregation and/or potentiation of aggregation induced by other agonists; and (2) inhibition of adenylate cyclase.377 Studies using the selective antagonist radioligand 3H methyl-yohimbine have shown that these effects are mediated through a single receptor, the «2- adrenergic receptor.158'271 Though mediated by a single receptor, aggregation and inhibition of adenylate cyclase are two separately controlled events. Observations in a family with a defect in the number of az-adrenergic receptors have shown that fewer receptors are required for inhibition of adenylate cyclase than for stimulation of aggregation.326 In addition, pretreatment of platelets with epinephrine for 5-30 minutes decreases the ability of epinephrine to subsequently initiate or potentiate aggregation; however, pretreatment with epinephrine does not affect the ability of epinephrine to inhibit adenylate cyclase.270 Platelets also have pZ-adrene'rgic receptors which, unlike the a 2- adrenergic receptors, inhibit platelet function by increasing the adenylate cyclase-mediated production of cAMP.464 The relative effects of epinephrine on platelet function are species dependent and determined by the densities of a2- and pz-adrenergic receptors.202 92 Epinephrine inhibits adenylate cyclase in a G1 protein-dependent manner. Binding of epinephrine to the a2-adrenergic receptor induces a CTP-dependent dissociation of the alpha subunit of Gi from the beta- gamma subunits of G1 with the subsequent inhibition of adenylate cyclase activity. Pertussis toxin ADP ribosylates G1, preventing the receptor- mediated dissociation and the subsequent inhibition of adenylate cyclase.53 Inhibition of adenylate cyclase by epinephrine is dependent on the presence of monovalent cations, preferably Na+.53'405'410 In contrast to most other physiologic agonists, epinephrine does not induce platelet shape change.383 The ability of epinephrine, by itself, to cause aggregation is questionable. In buffers containing physiologic [Ca++], epinephrine does not cause any aggregation, leading one group to propose that epinephrine only potentiates activation by other agonists and is not a true aggregating agent.219 Other observations have shown that epinephrine causes full aggregation in a medium containing low [Ca++] and added fibrinogen.201 The mechanism by which epinephrine potentiates aggregation is not clear. It has been proposed that epinephrine activates platelets by the Na+/H+ exchange-dependent activation of PLA2 in a manner similar to ADP.409'410’411 There are, however, some differences. Fibrinogen binding appears essential for this activation to occur since it is blocked by monoclonal antibodies against the CPIIb-IIIa complex.20 Some evidence suggests that epinephrine alone induces exposure of the fibrinogen binding sites and subsequent fibrinogen binding.372'445 Other studies have demonstrated that binding of FSBA to the ADP receptor 93 inhibits epinephrine—induced fibrinogen binding, suggesting an essential role for ADP in epinephrine induced platelet activation.83'123 Previous studies have used ADP-mobilizing enzymes (ie apyrase or creatine phosphokinase) which may have left very small trace amounts of ADP in the preparation. Investigations have shown that epinephrine increases the avidity of the ADP receptor for ADP, making trace amounts of ADP effective in initiating exposure of the fibrinogen receptor. By itself, epinephrine does not cause an elevation in [Ca++]1 as measured using the fluorescent chelator dyes quin2 or indo 1. It does cause a measurable increase in [Ca++]1 using aequorin in buffers containing Ca++, but not in buffers containing EGTA.7O'75'219'438 This suggests that epinephrine causes a local membrane-associated, influx of external Ca++ without any direct effects on the release of internal Ca++. When added in combination with other agonists, epinephrine potentiates Ca++ influx, Ca++ release from internal stores, protein phosphorylation of both the 20kDa and 47kDa proteins, Na+/H+ exchange and PA formation.75'219’290'402 The exact mechanism(s) for the epinephrine-induced potentiation of platelet activation is not known. This potentiation is independent of PLC activation290 since epinephrine alone cannot activate PLC.383 The stimulatory effects on PLC, as indicated by PA production, are not inhibited by the prevention of NaI/H+ exchange, suggesting that some pathway besides Na+/H+ exchange must exist to mediate the epinephrine effects on platelet activation.“02 94 The potentiation of Ca++ mobilization is independent of both fibrinogen binding and TxA2 production.317 Platelet-Activating Factor (PAF) PAF is a phospholipid and a physiologic agonist of intermediate potency.377 Although a PAF receptor has not been isolated, binding studies using 3H-labelled PAF have shown that canine platelets have a high affinity PAF receptor on their platelet surface. In addition, the specific binding of PAF to canine platelets indicates that the receptor is of similar number, capacity and affinity to that found in human platelets.186 Pretreatment with PAF inhibits the ability of platelets to respond to subsequent stimulation with PAP or other agonists. Washing the platelets to remove external agonist does not restore sensitivity, suggesting that the agonist induces an uncoupling of the receptor from the effector system.272 At high concentrations, PAF acts as a detergent and non-specifically alters membranes.364 Binding of PAF to its receptor mediates the hydrolysis of the membrane phosphoinositides with the production of the second messengers (1,4,5)IP3 and 1,2-DG.106'22]*'375'383 PAF also causes mobilization of Ca++. Measurements of [Ca++]1 in Ca++- and EGTA-containing buffers have shown that the majority of the increase in [Ca++]1 comes from influx of external Ca++.155'432 Kinetic studies have demonstrated a lag period between PAF-binding and Ca++ mobilization, suggesting a requirement for the production of second messengers in the generation of the Ca++ signal.351'352 Cyclooxygenase inhibition only mildly decreases the 95 increase in [Ca++]i, therefore the Ca++ flux is relatively independent of TxAz production.155 The role of secreted ADP and of cyclooxygenase-mediated production of prostaglandin endoperoxides/TxAz in PAF-induced aggregation and secretion is unclear. Shape change and primary reversible aggregation occur independent of ADP and arachidonic acid metabolism. Irreversible aggregation is strictly dependent on ADP release while its dependence on TxAZ is variable.250 In response to low concentrations of PAF, activation of the cyclooxygenase pathway is essential for irreversible aggregation and secretion.68'407 In response to high concentrations of PAF, the findings differ. Some studies have shown that PAF-induced aggregation is minimally affected by treatment with cyclooxygenase' inhibitors while secretion of dense granules is partially inhibited.250’257'258 Other evidence indicates that both aggregation and secretion are partially blocked by inhibition of cyclooxygenase.155 Finally, incubation of platelets with a specific TxA2 antagonist blocks both secretion and irreversible aggregation, suggesting that arachidonic acid metabolism is required for full platelet activation to occur in response to PAF. This same study has shown that inhibition of Na+/H+ exchange causes partial inhibition of TxA2 production and subsequent secretion, suggesting that PAF stimulates the release of arachidonic metabolites by two mechanisms, one dependent on Na+/H+ and one independent of Na+/H+ exchange.413 As noted with ADP and epinephrine, PAF inhibits the activation of adenylate cyclase by stimulatory agonists. This inhibition is dependent 96 on external Na+ and the presence of guanine nucleotides, suggesting a Gi-mediated mechanism similar to that discussed under the epinephrine- induced inhibition of adenylate cyclase.162'459 Thrombin Thrombin is a strong physiologic agonist which utilizes two distinct mechanisms to activate platelets: a receptor-mediated and a proteolytic process.263 Thrombin interacts with the platelet surface at two, and probably more, separate sites. It binds to CPIb in a hormone- like manner and causes the proteolytic cleavage of GPV.36'157'197 The platelets from patients with Bernard Soulier syndrome have delayed, but maximal, responses to thrombin.185'26o This finding suggests that CPIb and CPV are not essential for the initiation of thrombin-mediated activation. Instead, these glycoproteins increase the rate of activation. Functional studies have also shown that platelets have two types of thrombin receptors. One is resistant to inactivation by chymotrypsin and mediates PLC activation, PKC activation, Ca++ mobilization, aggregation and secretion. The second receptor is inactivated by the protease chymotrypsin and is resistant to activation by gamma thrombin, a derivative of alpha thrombin which lacks clotting activity. This site mediates adenylate cyclase inhibition and arachidonic acid mobilization.262 Thrombin initiates platelet activation by stimulating PLC activity21'3l‘v90'262'346 as a result of both receptor-mediated and proteolytic processes.263'346 The hydrolysis of the 97 polyphosphoinositides occurs very rapidly and is the triggering event for platelet activation.37'378 Investigations in saponin-permeabilized platelets have shown that thrombin can cause platelet aggregation and secretion by some unidentified mechanism which is independent of phosphoinositide hydrolysis and PLC activation.57'226 Thrombin also stimulates Ca++ mobilization in platelets.438 Studies comparing Ca++ mobilization in the presence and absence of external Ca++ have shown that the majority of the increase in [Ca++]1 comes from influx of external Ca++. The influx is partially blocked by cyclooxygenase inhibition while the release of Ca++ from internal stores is unaffected.192 Removal of external Ca++ partially inhibits aggregation with no effect on secretion.243 Kinetic studies have shown a delay between the binding of thrombin to its receptor and the initiation of Ca++ mobilization. This suggests that the receptor coupling to both the plasma membrane Ca++ channel and the internal stores of Ca++ is not direct and requires the generation of second messengers.352'477 Observations on the effect of the ADP receptor antagonist FSBA have demonstrated that thrombin causes fibrinogen receptor exposure, irrespective of the presence of external ADP. It has been proposed that thrombin causes the cleavage of aggregin, leading to exposure of the fibrinogen binding site and subsequent aggregation.83 This cleavage is not due to direct proteolytic action by thrombin itself; rather, it is secondary to the activation of calpain by the thrombin-stimulated increase in [Ca++]1.321 98 Stimulation of platelets with thrombin induces activation of Na+/H+ exchange and marked cytosolic alkalinization.98'392 This activation of Na+/H+ exchange is not a direct action of thrombin; instead, it is due to the PLC-mediated production of second messengers and secondary activation of PKC.386 Cytosolic alkalinization is not essential for platelet activation following stimulation with high concentrations of thrombin.4'99'176'358'398'474 Stimulation of platelets with thrombin also causes the production of arachidonic acid metabolites secondary to the agonist-induced increase in [Ca++]1. The release of these metabolites is not essential for platelet activation since cyclooxygenase inhibition does not prevent aggregation, secretion or PLC activation following stimulation with thrombin.345 Finally, thrombin inhibits adenylate cyclase activation by a mechanism which is receptor-mediated and dependent on G protein activation.21'262 Calcium Ionophores A23187 and Ionomycin Calcium ionophores A23187 and ionomycin are non-physiologic agonists that transport Ca++ across cell membranes and release Ca++ from internal stores.121v457 Observations in platelets in a Ca++-free environment indicate that ionophores can release sufficient internal stores of Ca++ to initiate secretion and aggregation, though the presence of external Ca++ potentiates the response.121'313'438'457 This elevation in [Ca++]1 is not affected by inhibition of cyclooxygenase or the removal of ADP, therefore it is a primary effect of calcium ionophores.339 99 Ca++ ionophores activate both PLA2 and PLC as assessed by the measurement of the production of TxAZ, arachidonic acid, PA and the hydrolysis of membrane phosphoinositides.34'311"339 The activation of PLAz-mediated arachidonic acid release is Ca++-dependent and requires micromolar [Ca++]i.314 The activation of PLC is not strictly due to increased [Ca""*']:l..225'339 Studies have shown that PLC activation in response to the calcium ionophores is due to the Ca++-dependent activation of PLAZ, the production of arachidonic acid metabolites and the release of dense granule ADP which act as positive feedback loops to activate PLC,337,339,444 Although cyclooxygenase inhibition prevents TxAz production and PLC activation in platelets stimulated by calcium ionophores, it does not completely block platelet activation. In platelets treated with cyclooxygenase inhibitors, secretion and aggregation still occur following stimulation with the calcium ionophores; however, the responses are inhibited.339v444 These findings suggest that the Ca++ ionophores partially activate platelets via some Ca++-dependent pathways which are independent of the positive feedback loops involving TxA2 and PLC. Calcium ionophores cause phosphorylation of both the 20kDa myosin light chain and the 47kDa protein.225'361 Phosphorylation of both proteins can occur in the presence of inhibitors of cyclooxygenase, suggesting Ca++-mediated activation of their respective kinases.225 100 Endoperoxide/Thromboxane Mimetic U46619 Thromboxane A2 and the prostaglandin endoperoxides are potent physiologic agonists which cause the platelet responses of shape change, secretion and aggregation. They also amplify the platelet response following stimulation with other agonists. Unfortunately, the short half life of both TxA2 (90 seconds) and PGHZ (5 minutes) preclude their use for in vitgg studies.156 This has lead to the development of a number of stable mimetics. The stable TxAZ/PGHZ mimetic U46619 induces shape change, aggregation and secretion.278 Canine platelets differ from other species in their response to U46619. When U46619 alone is added to canine platelets, it induces only shape change. Full aggregation requires the prior addition of a subthreshold dose of epinephrine to U46619.66 The platelet TxAz/PGHZ receptor has been isolated and characterized.108 Binding studies have shown that it is located on the plasma membrane and/or the DTS membranes.363 Recent observations suggest that there are actually two subtypes of the TxAz/PGHZ receptor which mediate separate functions. One subtype mediates shape change while the other mediates aggregation.107 Pretreatment of platelets with U46619 causes decreased responsiveness to subsequent stimulation.with additional doses of U46619 but not to stimulation with other agonists.233 The decreased responsiveness results from: (1) initial receptor desensitization due to 101 the uncoupling of receptor from G protein; and (2) delayed downregulation and loss of receptor sites.274'419 Binding of U46619 to the TxA2/PGH2 receptor activates PLC independent of arachidonic acid metabolism and external Ca""".337'379 Activation of platelet PLC by U46619 is accompanied by relatively poor release of arachidonic acid337 since the desensitization and downregulation of TxAz/PGHZ receptors may inhibit the binding, and therefore the positive feedback effects, of endogenously produced endoperoxides and TxA2.379 Activation of PLC by U46619 is regulated by a pertussis toxin-resistant G protein58 and is not affected by inhibition of Na+/H+ exchange.409 In U46619-stimulated platelets, removal of external ADP is accompanied by partial inhibition of aggregation and PLC activation.337'379 Evidence using the ADP analogue FSBA indicates that external ADP is required to stimulate exposure of the fibrinogen binding site. The addition of FSBA to U46619-stimulated platelets blocks fibrinogen binding and subsequent aggregation without affecting shape change.83 U46619 stimulates Ca"'+ mobilization that is not affected by cyclooxygenase inhibition.l“7'50 Ca++ mobilization occurs in both the presence and absence of external Ca++, though the response is greater when external Ca++ is present. Kinetic studies have shown a time delay between the binding of agonist and the increase in [Ca++]1, indicating the presence of one or more biochemical steps between binding and the 102 release and/or influx of Ca++.352 A recent study has shown that inhibition of Na+/H+ exchange decreases Ca++ mobilization in response to U46619. It was proposed that this inhibition occurs at the level of (1,4,5)IP3-mediated release from internal stores. Prevention of Na+/H+ exchange inhibits cytosolic alkalinization, negating the potentiating effect of increased pHi on (1,4,5)IP3-mediated Ca++ release.394 Finally, studies on platelet membrane preparations have shown that U46619 inhibits both basal and PGEl-stimulated adenylate cyclase activity in a C protein-dependent manner. It has not been determined whether this inhibition occurs in intact platelets.15 Phorbol Ester Phorbol esters are non-physiologic agonists which cause both aggregation and secretion.377 The endogenous second messenger 1,2-DC increases the affinity of PKC for Ca"'+ and phospholipid, causing PKC activation.71'72 Phorbol esters have a structure similar to 1,2-DC and are believed to intercalate into the membrane and cause activation of PKC. In contrast to 1,2-DC, the phorbol esters are not readily metabolized by the cell and cause permanent PKC activation.377 The aggregation and secretion patterns following stimulation with phorbol esters are unlike patterns obtained following stimulation with most other agonists. Phorbol esters do not cause shape change, but they cause aggregation that is relatively slow and occurs after a variable delay period.381'479 Aggregation requires external Ca++, although 103 concentrations lower than those needed by other agonists are required.187 Phorbol-induced aggregation is not dependent on the production of arachidonic acid metabolites and is not significantly affected by cyclooxygenase inhibition.187'456 Partial inhibition of aggregation occurs following the addition of apyrase to remove external ADP or following stimulation of cAMP production.10'187 This inhibition is much more pronounced in platelets stimulated by the synthetic 1,2-DC than by phorbol esters, highlighting a difference between phorbol- and endogenous 1,2-DG-induced platelet activation.10 Secretion occurs well after the onset of aggregation and is not accompanied by granule centralization on electron microscopic examination.“56 The specific effects of phorbol on signal transduction were covered in the section on PKC and will not be repeated. BASSET HOUND HEREDITARY THROMBOPATHY Basset hound hereditary thrombopathy (BHT) is an inherited platelet disorder in purebred Basset hounds. Although a similar disorder has not been identified in humans or other animals, BHT presents a unique opportunity to increase our understanding of normal platelet function. The disease was first identified in the 1970's when a practicing veterinarian requested some advice about a group of Basset hounds with hemorrhagic diathesis. Clinically, affected dogs present with signs typical of a platelet disorder. Findings include petechiation, easy bruising, mucosal hemorrhages and excessive bleeding associated with estrous cycles and the shedding of deciduous teeth. The 104 initial investigation showed that affected dogs have prolonged bleeding times and complete absence of aggregation, either primary or secondary, in response to stimulation with ADP.32’191 The examination of the pedigrees and in viggg aggregation studies on a group of 92 related Basset hounds suggested an autosomal recessive pattern of inheritance.32 Further investigations on 45 of these dogs identified numerous similarities with Clanzmann's thrombasthenia in humans.32 Normal laboratory findings identified in both disorders include coagulation factor parameters, vWF levels, platelet number and morphology on light microscopy and platelet morphology on transmission electron microscopy. Abnormal findings include prolonged bleeding time, decreased in viggo retention in glass bead columns and absence of aggregation to any concentration of ADP tested. One difference is clot retraction time which is normal in platelets from dogs with BHT but markedly prolonged or absent in Clanzmann's thrombasthenic platelets. Clot retraction depends on the attachment of both cytosolic actin filaments and extracellular fibrin to the GPIIb-IIIa complex on the surface of the platelet pseudopods. The subsequent generation of a contractile force by the interaction of myosin and actin pulls the pseudopods in toward the cell, also pulling in the attached fibrin.142 In addition to failure of primary aggregation in response to ADP, platelets from dogs affected with BHT have abnormal release of dense granule ATP. When platelets from affected dogs are stimulated with lOuM ADP, they change shape and release ATP from the dense granules. The release occurs six times faster than the release in control dog 105 platelets. Sequential transmission electron microscopy have confirmed the aggregation and secretion defects. It was proposed that the lack of aggregation is due to a defect in either CPIIb-IIIa complex itself or its functional ability to bind fibrinogen.300 Due to a clinical resemblance to Clanzmann's thrombasthenia, contact activation was measured in platelets dogs with BHT and control dogs using whole mount electron microscopy. Affected platelets attach poorly to a foreign surface and have decreased contact-induced shape change when compared to control canine platelets. The addition of ADP causes a five-fold increase in adherent platelets and a four-fold increase in spread forms. Aggregate formation is not noted. These findings are similar to those observed in platelets from a human with Clanzmann's thrombasthenia.251 Based on the hypothesis that the defect in BHT may represent an inability of affected platelets to bind fibrinogen, the initial investigations focused on the examination of membrane GPIIb-IIIa content and ability to bind fibrinogen. Membrane CPIIb-IIIa content was evaluated using three different techniques: (1) crossed immunoelectrophoresis; (2) two-dimensional nonreduced-reduced electrophoresis; and (3) O'Farrell two-dimensional electrophoresis. There was no difference between affected and control platelets in membrane glycoprotein content using either crossed immunoelectrophoresis or two-dimensional electrophoresis. There was also no difference in membrane glycoproteins using O'Farrell's two-dimensional electrophoresis; however, some low molecular weight lipids and/or 106 sialoglycoproteins were missing in samples from BHT platelets. The significance of these missing components is not known.299 Normal platelet GPIIb-IIIa content does not prove that the complex is functional and able to bind fibrinogen. Fibrinogen binding was therefore measured in affected and control canine platelets. The platelets were stimulated with ADP in the presence of 125I-fibrinogen and the quantity of surface bound radioactivity measured. There was no significant difference between the two groups of dog platelets using this technique.298 Fibrinogen binding was also evaluated using scanning electron microscopy. Platelets from normal and affected dogs were stimulated with ADP, allowed to adhere and spread and labelled with either albumin-gold or fibrinogen-gold. The binding of fibrinogen-gold was quantitatively and morphologically identical in affected and control platelets.119 These investigations demonstrated that BHT platelets are able to bind fibrinogen; however, they also suggest that fibrinogen binding alone is not sufficient to maintain aggregation. Instead, there appears to be one or more post-fibrinogen binding events necessary for aggregation. The necessity for some, as yet, undefined post-fibrinogen binding events has been reported by other groups. One study has demonstrated that formalin fixed, ADP stimulated platelets are able to bind fibrinogen but are unable to aggregate.305 Another group has described two monoclonal antibodies that inhibit ADP stimulated aggregation without inhibiting fibrinogen binding.281 107 Another study by Catalfamo et al67 on affected platelets from BHT dogs has reported no fibrinogen binding in response to ADP and impaired rate of fibrinogen binding in response to thrombin. The reason for the discrepancy between the results from this study and the results by Patterson et al298 and Estry et al119 discussed above is likely due to the difference in experimental techniques used. Fibrinogen binding in the study by Catalfamo et al67 was measured using a fibrinogen-coated bead agglutination technique.80 This technique depends on the formation of bead agglutination after the platelets bind to the beads. It is possible that this technique actually measures the ability of platelets to aggregate rather than their ability to bind fibrinogen. If platelets bind to the fibrinogen coated beads but do not bind to each other, it is difficult to imagine how the beads can agglutinate. At this point it was determined that BHT is not an animal model for Clanzmann's thrombasthenia, but represents a unique disorder involving a defect in the undefined post-fibrinogen events essential for platelet aggregation. This suggests the importance of transmembrane signal processing and intracellular biochemical changes in the platelet response of aggregation. The importance of cAMP as an inhibitory second messenger in platelets lead one group to investigate its role in BHT.48'49 They found that basal levels of cAMP are increased 30-90% in affected platelets relative to control dog platelets. The addition of forskolin, a substance which directly activates adenylate cyclase, causes marked increase in concentrations of cAMP in affected platelets but not in 108 normal dog platelets.49 These findings were further investigated by evaluation of cAMP phosphodiesterase (cAMP-PDE), the enzyme responsible for the removal of cAMP. Resting levels of cAMP-PDE activity are similar in affected and control dog platelets. Affected platelets are more sensitive to the inhibitory effects of forskolin and MIX, a competitive inhibitor of cAMP-PDE. The post-stimulation concentrations of cAMP require greater time to return to non-stimulated levels in affected platelets than in control platelets. In addition, the addition of epinephrine or PG12 respectively decrease or increase the forskolin stimulated levels of cAMP. In combination, these findings suggested that the G protein regulation of adenylate cyclase is intact in affected and the defect resides in the regulatory control of cAMP-PDE.48 Aggregation, ATP dense granule secretion and aspirin sensitivity were examined in platelets from dogs affected with BHT. The aggregation study showed that affected platelets: (1) change shape but do not undergo primary aggregation in response to ADP, PAF and calcium ionophore A23187; (2) have full aggregation responses to high concentrations of thrombin although aggregation is delayed in onset and significantly slower than in control canine platelets; and (3) have phorbol ester-stimulated aggregation responses similar to those in control platelets. The ATP secretion study showed that affected platelets: (l) secrete ATP in response to stimulation with ADP, thrombin, phorbol ester and high concentrations of PAF; and (2) do not secrete ATP in response to calcium ionophore A23187 and low concentrations of PAF. Aspirin treatment of affected and control platelets demonstrated: (1) almost total inhibition of secretion in 109 response to ADP in both affected and control platelets; and (2) partial inhibition of secretion in response to PAF, thrombin and PMA that is more pronounced in affected platelets. These results suggest that at least one effector pathway is intact in BHT and can elicit full platelet activation. This pathway is stimulated by PMA and high concentrations of thrombin and may involve PKC. In addition, the results of aspirin inhibition indicate that affected platelets are more sensitive to cyclooxygenase inhibition than control platelets and have a greater reliance on the release of arachidonic acid metabolites.254 Another research group also measured aggregation responses in platelets from dogs with BHT.67 They reported similar results following ADP and thrombin stimulation. They also showed that stimulation of affected platelets with either collagen or arachidonic acid causes shape change but not primary aggregation. In contrast to the results discussed above, this group reported that aggregation occurs in response to high concentrations of A23187 in gel-filtered,affected platelets. The concentration of A23187 used in this study has been shown in our laboratory to actually cause cell lysis in gel-filtered platelets. Phosphorylation of platelet proteins was measured in platelets from dogs with BHT in response to a variety of agonists.255 There is no significant differences between affected and control platelets in the phosphorylation patterns of the 20kDa myosin light chain and the 47kDa protein. Shape change correlates closely with phosphorylation of myosin light chain; however, there is no consistent correlation between aggregation and/or secretion and the phosphorylation of either the 20 or 110 47kDa proteins. Interestingly, there is significantly decreased phosphorylation of a 64-67kDa protein in affected platelets. The identity and significance of this protein is not known.255 Finally, TxA2 production, as indicated by production of the stable metabolite Tsz, was measured in affected and control platelets in response to stimulation with a wide variety of agonists.256 The basal concentrations of TxB2 in platelet rich plasma are markedly greater in affected samples than in control samples. Stimulation with ADP, thrombin or high concentrations of PAF causes rapid production of significantly greater concentrations of TxB2 in affected samples. Stimulation with A23187 or low concentrations of PAF causes the production of significantly decreased concentrations of Tx32 in affected samples when compared to control samples. As expected, PMA stimulation does not produce significant increases in Tx32 concentrations in either group. It is interesting to note that those agonists which cause dense granule secretion also cause increased production of Tx32 in affected samples when compared to control samples. On the other hand, those agonists which do not cause dense granule secretion produce significantly decreased quantities of TxB2 in affected samples relative to control samples. These findings suggest that the mechanism for the production of arachidonic acid metabolites is intact in affected platelets; however, the release of arachidonic acid metabolites does not appear to be under the normal regulatory control. The addition of exogenous arachidonic acid to affected and control platelets produce similar quantities of Tsz, indicating that the pathway for the 111 production of TxA2 from arachidonic acid is intact and under normal regulation in affected platelets. In conclusion, the platelet disorder in BHT does not resemble any thrombopathy reported to date. The findings discussed above suggest that there is a defect in some post-fibrinogen binding event(s) leading to failure of platelet aggregation following stimulation with ADP, PAF, calcium ionophore A23187 or low concentrations of thrombin. In contrast, high concentrations of thrombin or PMA cause full platelet activation. Identified abnormalities in BHT platelets include decreased phosphorylation of a 64-67kDa protein, abnormal secretion of ATP from dense granules, greater dependance on the production of arachidonic acid metabolites, altered regulation of TxA2 production and increased cytosolic cAMP concentrations. None of these findings can adequately explain the defect in BHT based on current understanding of the process of platelet activation, necessitating further investigations of the biochemical pathways involved in platelet signal transduction. MEASUREMENT OF RESTING AND STIMULATED PLATELET CALCIUM INTRODUCTION Calcium ions are intracellular messengers which play a central role in the transmission of signals within living cells.44 In platelets, Ca++ is important in the physiologic responses of shape change, secretion and aggregation.377 The concentration of cytosolic free ionized calcium ([Ca++]i) within resting platelets is maintained at approximately lOOnM.51'331'335 Spatial distribution studies on individual platelets have demonstrated that the intracellular free ionized Ca++ is not distributed homogeneously within the cytoplasm. Instead, it exists in two different gradients: (1) a continuous gradient with increasing [Ca++]i towards the plasma membrane; and (2) discontinuous gradients with plateaus in restricted regions of the endoplasm. It has been suggested that the relatively high concentration of Ca++ under the plasma membrane results from compartmentation of ion movement by the cytoskeleton while the discontinuous gradients are associated with the internal storage sites.“28 Following stimulation, the gradients disappear and there is a uniform increase in [Ca++]i. 112 113 Platelets actively maintain resting [Ca++]i by: (1) promotion of Ca++ efflux across the plasma membrane; (2) energy-requiring sequestration of Ca++ into intracellular storage sites; and (3) limitation of the influx of external Ca++. The process of Ca++ efflux in platelets is not well understood; however, it is affected by changes in the membrane Na+ gradient, implying a role for a Na+/Ca++ exchange mechanism.5“'365 Recent studies have also shown that Ca++ efflux is regulated by a plasma membrane Ca++-ATPase pumplll"115'116 which is distinct from the Ca++-ATPase present on the internal membranes. The plasma membrane pump is related to the plasma membrane Ca++-ATPase present on erythrocytes and hepatocytes.116 Ca++ sequestration is regulated by a Ca++-ATPase located on the DTS which is related to the Ca++ pump located on the sarcoplasmic reticulum of muscle cells.107'111'265 It has a higher affinity for Ca++ and a lower requirement for ATP than the plasma membrane pump.114 Ca++ influx occurs by passive diffusion down a concentration gradient through Ca++- selective channels in the plasma membrane. The rate of influx is maximal at Ca++ concentrations 20-fold less than the levels normally found in plasma.51 A wide variety of agonists induce an increase in [Ca++]1.331'438 Agonist-stimulated Ca++ fluxes result from the prolonged influx of external Ca++ and/or the rapid release of Ca++ from exchangeable 428 C8++ internal storage sites. influx occurs along receptor-operated, not voltage-dependent, channels in the plasma membrane.14,154,243'349'476 These channels may be closely associated with, if not actually part of, the membrane CPIIb-IIIa complex.315v469 114 Kinetic studies have shown that there is a lag period between the addition of agonist and the onset of a measurable increase in [Ca++]i, suggesting a role for one or more biochemical steps between agonist binding and the receptor-operated influx of Ca++. These same studies have also demonstrated that Ca++ influx precedes internal Ca++ mobilization, suggesting that Ca++ influx and internal release are two distinct events.351’352 The internal release of Ca++ is mediated by the production of second messengers generated by the agonist-stimulated hydrolysis of membrane phosphoinositides.331 (1,4,5)IP3 is the primary second messenger implicated in the receptorooperated release of Ca++ from the DTS.11’12'242 Historically, a variety of experimental techniques have been used to measure Ca++ fluxes, intracellular storage pools and cytosolic [Ca++]1 in platelets. These include: (1) chlortetracycline; (2) 45Ca++; (3) fluorescent chelator dyes; and (4) luminescent photoproteins. Chlortetracycline is a fluorescent compound which, after binding Ca++, increases its fluorescence and becomes associated with the platelet membranes. Unfortunately, its intracellular location is not known. Although useful for measuring Ca++ flux in some circumstances, chlortetracycline does not reflect changes in [Ca++]1 and may not always accurately reflect Ca++ metabolism in the platelet."‘"97 45Ca++ is used to estimate the size and exchange rate in the surface bound and intracellular Ca++ pools; however, it also does not reflect [Ca++]1.377 Currently, the most common techniques utilized to measure platelet [Ca++]1 include a family of fluorescent dyes and the luminescent 115 photoprotein aequorin. The fluorescent dyes are polycarboxylic Ca++ chelators related to EGTA and include the compounds quin2, fura 2 and indo l. The acetoxymethyl (AM) ester form of these dyes is lipophilic and readily penetrates the plasma membrane of the platelets. Once in the cytosol, the ester is cleaved by cytosolic esterases to form the free acid which is then trapped within the cell. The free acid binds to divalent cations in 1:1 stoichiometry. The Ca++-bound dye increases its fluorescence relative to the unbound form. The change in fluorescence can be measured and is proportional to the corresponding change in [Ca‘*"*']i_.79'97*148'331 Quin2 was the first such dye synthesized. Unfortunately, due to its low extinction coefficient and fluorescent quantum yield, large cytosolic concentrations of quin2 are required to produce fluorescence signals significantly greater than background autofluorescence and noise. This high concentration of quin2 buffers intracellular free Ca++. In addition, the cleavage of quin2-AM generates potentially harmful concentrations of formaldehyde, protons, and acetate ions.79'97'333 Finally, quin2 has a relatively high affinity for Ca++ and rapidly becomes saturated, losing resolution at high concentrations of Ca‘H'.148 Quin2 does not accurately measure [Ca++]1 above 2-3uM, though it is effective at measuring levels in the normal resting range.79'148 These problems with quin2 lead to the generation of the newer dyes fura 2 and indo 1. These dyes have three advantages over quin2: (1) they produce 30-fold greater fluorescence; (2) they have improved selectivity for Ca++ over other divalent cations; and (3) they yield a change in either the emission or the excitation fluorescence wavelength 116 following Ca++ binding. The former two features significantly lower the quantity of cytosolic dye needed to produce a measurable signal, decreasing Ca++ buffering and the generation of deleterious substances. The latter feature permits calculation of [Ca++]1 based on the ratio of fluorescence of the wavelengths of Ca++-bound and -unbound fura 2 and indo 1. The use of the ratio method to calculate [Ca++]i partially negates the effects of cell size, intracellular dye concentration or instrument inefficiency which affect the calculation of [Ca++]1 using the single wavelength technique.79'97’1l‘8'425 Finally, the newer dyes have greater resolution, allowing the accurate calculation of [Ca++]i up to S-lOuM.79 The luminescent photoprotein aequorin has also been frequently used to measure [Ca++]1 in platelets. Aequorin is a polypeptide of 20,000 daltons which contains a bound chromophore. When this complex binds to free Ca++, the chromophobe becomes oxidized and releases C02 and a photon of light of wavelength 469 nm. The amount of light produced is quantitated by a photomultiplier tube and used to calculate the [Ca++]1. The response of aequorin to [Ca++]1 is exponential over the range of 10'7 to lO'aM, therefore a small increase in [Ca++]i in a local region gives a greater increase in the luminescence signal than if that same amount of Ca++ were evenly distributed throughout the cytosol.“4'97'188 It has been suggested that aequorin detects local zones of Ca++ flux, possibly associated with the plasma membrane, while the fluorescent chelator dyes detect average [Ca"""]1_.188'190 At the initiation of the research for this dissertation, it had 117 been determined that the platelet defect in Basset hound hereditary thrombopathy is not related to abnormalities in membrane GPIIb/IIIa structure or function.119'298’299 Instead, the defect in BHT appears to be associated with post-fibrinogen binding events involved in the process of signal transduction. It was decided to measure the resting and post-stimulation [Ca++]i due to the pivotal role of Ca++ as an intracellular second messenger in signal transduction. [Ca++]1 was measured using both aequorin and fura 2 since the two experimental techniques measure different aspects of Ca++ metabolism. Fura 2 was chosen over indo 1 because fura 2 has been more commonly used to measure [Ca++]1 in platelets in suspensions. Platelets were stimulated using a similar range of agonists as those used for the aggregation/secretion study254 to determine if there was any association between Ca++ flux and the abnormalities identified in the aggregation and/or secretion of affected platelets. MATERIALS AND METHODS Experimental Subjects For the measurement of [Ca++]1 using fura 2-loaded platelets, the affected group consisted of five Basset hounds with BHT. There were one male and four female dogs. .The control group consisted of one male and two female Basset hounds without evidence of BHT. Due to the unavoidable loss of dogs and the replacement with different dogs, the experimental groups for the measurement of [Ca++]1 using aequorin-loaded platelets were different. The affected group consisted of four Basset 118 hounds with BHT which were a subset of the affected group for the fura 2 studies. There were one male and three female dogs. The control group consisted of two female and one male Basset hounds with no evidence of BHT. Only one of the female dogs had been used in the fura 2 study. All dogs were free from signs of clinical disease other than those associated with BHT. They did not receive any medication during the study and had been vaccinated against rabies, distemper, leptospirosis, adenovirus, parainfluenza and parvovirus. They were housed in kennels in the Veterinary Clinical Center at Michigan State University, were fed a commercial dry dog chow (Hill's Pet Products, Topeka, KS) and received regular exercise. Reagents Adenosine diphosphate (ADP) was purchased from Sigma Chemical Company (St. Louis, MO). Stock solution (2.6mM) was mixed in 20mM Tris- saline, aliquoted and stored at -20°C. Platelet-activating factor (PAF) was purchased from Calbiochem Corporation (San Diego, CA). Stock solution (1.0 mg/ml) was mixed in ethanol and stored at -20°C. Thrombin was purchased from Sigma Chemical Company, mixed in distilled water and stored in aliquots at -20°C until use. Phorbol-12-myristate-l3-acetate (PMA) was purchased from Calbiochem. Stock solution (1.56 mM) was mixed in ethanol and stored at -20°C. Ionomycin was purchased from Calbiochem. Stock solution (2mM) was mixed in dimethyl sulfoxide (DMSO), aliquoted and stored at 4°C. Fura 2-AM was purchas‘d from Calbiochem. Stock solution (1mg fura 2-AM in 249ul anhydrous DMSO) was 119 aliquoted and stored in a dessicator jar at -20°C. Aequorin was purchased from the Mayo Clinic Foundation (Rochester, MN). Stock solution (1mg aequorin in 333ul Ca++-free water containing 7mM EGTA) was aliquoted into plastic cuvettes and stored at -20°C. Sepharose 4-B was purchased from Sigma Chemical Company. Digitonin was purchased from Sigma Chemical Corporation, mixed in de-ionized distilled water to a final concentration of lOmg/ml and stored at 4°C. Prostaglandin E1 (PGEl) was purchased from Sigma Chemical Company. Stock solution (lmM) was mixed in ethanol and stored at -20°C. All other chemicals were of reagent grade. Stock solutions were diluted at the time of use and maintained on ice. ADP was diluted with 0.9% NaCl. Thrombin was diluted with distilled water. PAF, ionomycin and PMA were diluted with balanced salt solution (l36mM NaCl, 5.4 mM KCl, 0.44 mM KHZPOQ, 0.34 mM NazHPOA and 5.6 mM glucose, pH 7.4). Final concentrations of ethanol or DMSO in the platelet solutions did not exceed 0.5% v/v. Platelet Collection To limit variability in experimental results, blood must be collected from healthy dogs which are not excited or receiving any medication.32 For all procedures, blood was obtained by jugular venipuncture using an 18 gauge needle into a plastic syringe containing 3.8% trisodium citrate at a ratio of 9ml of blood to lml anticoagulant. The dogs are conditioned to calmly accept this procedure and minimal physical restraint is required. If the collection was disrupted, the 120 original sample was discarded and a second sample collected. To obtain platelet rich plasma (PRP), the blood was separated by centrifugation 3-5 times at 1,324 x g for 60 seconds, removing a small quantity of PRP between each spin. After removal of sufficient quantities of PRP, platelet poor plasma (PPP) was obtained by centrifugation of the remainder of the blood at 1,324 x g for 13 minutes. For all platelet isolation procedures, plastic (polystyrene, polyethylene or polypropylene) labware was used. Measurement of [Ca++]1 Using Pure 2 [Ca++]1 was measured by modification of the technique described by Daniel et al.97 PRP was acidified by the addition of 9ul 1M citric acid/ml PRP. The suspension was incubated at room temperature for 5 minutes prior to centrifugation for 15 minutes at 800 x g. The supernatant was discarded and the platelet pellet resuspended in 4ml of autologous PPP acidified by the addition of 36ul 1M citric acid. Fura 2-AM was added from stock solution to yield a final concentration of 4uM and the platelet sample incubated at 37°C for 30 minutes. The suspension was centrifuged at 800 x g for 15 minutes, the supernatant discarded and the pellet resuspended in 1 ml Ca++-Mg++-free Tyrode‘s buffer (137mM NaCl, 2.7mM xc1, 5.5mM glucose, 0.35mM NazHPoa, lOmM HEPES and 0.2% bovine serum albumin, pH 7.4). The platelets were gel-filtered by passing the suspension over a 10ml Sepharose 4B column equilibrated with the same buffer. The eluted platelet suspension was manually counted using the Unopette system and a Neubauer hemocytometer and 121 diluted to l x 108 platelets/ml using the same buffer. MgClz (final concentration 0.8mM) was added and the suspension divided into 1.5ml aliquots into plastic spectrophotometer cuvettes, then covered by parafilm. The cuvettes were maintained at room temperature until 10 minutes prior to the measurement of [Ca++]1 at which time they were warmed to 37°C. Five minutes prior to measuring [Ca++]i, either lmM CaC12 or lmM EGTA was added along with a teflon coated magnetic stir bar. Fluorescence was measured in a Perkin-Elmer MPF 66 spectrofluorometer using an excitation wavelength of 339nm and an emission wavelength of 502nm. The platelet suspension was stirred and resting fluorescence recorded. Agonist (60ul) was added and maximum post-stimulation fluorescence recorded. Agonists used in this study included ADP, PAF, thrombin, PMA and ionomycin. The [Ca++]1 was calculated based on the formula: [Ca++] - Kd [(F'Fmin)/(Fmax'F)] where Rd is the dissociation constant of fura 2 (assumed to be 224nM), F is the measured fluorescence of the sample, Fmax is the sample fluorescence following addition of 4mM Ca++ and lysis by 50uM digitonin, and Fmin is the sample fluorescence following addition of 6mM EGTA/75mM Tris base to the lysed platelet suspension. Autofluorescence of the platelet suspension was determined by measuring fluorescence in similarly prepared platelet samples not incubated with fura 2-AM. None of the agonists used had significant autofluorescence at the wavelength settings used for these experiments; Leakage of dye from the samples was determined by centrifuging representative platelet samples for 1 minute at 11,600 x g. The fluorescence of the supernatant was measured in the presence and absence of external Ca++ to determine the Ca++ sensitivity of the leaked dye. The external dye was found to be 122 composed of both Ca++-sensitive and Ca++-insensitive forms. The leakage rate was relatively small and the resting [Ca++]i relatively constant over the time span of the experiments. For these reasons, it was decided to not correct for external dye in the calculations. Measurement of [Ca++]1 Using Aequorin The platelets were prepared using a modification of the technique described by Yamaguchi et a1.467 PRP was acidified by the addition of 9ul 1M citric acid/ml PRP. The sample was incubated at room temperature for 5 minutes prior to centrifugation for 15 minutes at 800 x g. The supernatant was discarded and the platelet pellet gently resuspended in 10 m1 HEPES buffered saline (140mM NaCl, 2.7mM KCl, 0.1% bovine serum albumin, 0.1% glucose and 3.8mM HEPES, pH 7.6) containing luM PGEl and SmM EGTA. The sample was incubated for 5 minutes at room temperature prior to centrifugation at 800 x g for 15 minutes. The supernatant was discarded, the pellet gently resuspended in 90ul of the same buffer at which time lOul stock aequorin solution was added. DMSO was added over a period of 7.5 minutes at a rate of lul DMSO/1.5 minutes to yield a final concentration of 6% by volume. The suspension was incubated for 2 minutes at room temperature, diluted to 1 ml using HEPES buffered saline without PGEI or EGTA and gel-filtered by passing the sample over a 10ml Sepharose 43 column equilibrated with the same buffer. The gel-filtered platelets were manually counted using the Unopette system and a hemocytometer and diluted to a final concentration of 1.5 x 108 platelets/ml using the same buffer. MgC12 (final concentration 0.8mM) was added and the sample divided into lml aliquots in plastic 123 aggregometer cuvettes, then covered with parafilm. Five minutes prior to the measurement of [Ca++]i, the cuvettes were warmed to 37°C, a teflon coated magnetic stir bar inserted and either lmM Ca012 or lmM EGTA added. The cuvettes were placed into a Platelet Ionized Calcium Aggregometer (PICA, Chronolog, Haverton, PA) and stirred at 900 rpm. After maintaining a steady resting baseline, 40u1 of agonist was added and both aggregation and peak dye luminescence were recorded on a dual channel Omniscribe recorder (Houston Instruments, Austin, TX). The total platelet aequorin luminescence (Lmax) was determined by lysis of a four-fold diluted platelet sample containing lmM Ca++ by lSOul of Triton X-100. The [Ca++]i was calculated from the maximum peak of the aequorin luminescence signal as described in the PICA instruction manual, using the calibration curve of log L/Lmax vs log [Ca++]. The calibration curve was supplied with each lot of aequorin and.was constructed using a medium containing lSOmM KCl, SmM PIPES and lmM MgClz. Statistical Analysis For the fura 2 studies, the resting and post-stimulation [Ca++]1 in response to each agonist was measured on three separate occasions in all dogs. For the aequorin studies, the resting and post-stimulation [Ca++]1 in response to each agonist was measured on two separate occasions in all dogs. For each dog, the [Ca++]i were averaged for each concentration of agonist to yield a representative value which was used in the statistical comparisons. All data was analyzed using a computer software statistical/graphics program (Statgraphics, STSC, Rockville, MD). Due to the small sample sizes, it was found that the distribution- 124 free non-parametric were not effective. For that reason, all data was compared using the standard parametric tests. Control and affected values were compared using the student's t test for independent samples while the stimulated and resting values within a group were compared using the t test for paired samples. RESULTS The agonists used for the calcium studies were the same ones used in the aggregation and dense granule secretion study on affected platelets which had been previously reported254 and included PAF, ADP, thrombin, PMA and a calcium ionophore. To evaluate Ca++ fluxes, the calcium ionophore ionomycin was used instead of A23187 because A23187 has fluorescent properties at the wavelengths used for the fura 2 studies which interfered with sample fluorescence. Results will be presented according to agonist. The agonist concentrations listed in the tables are the final concentrations in the platelet samples. The concentrations of agonists were chosen based on aggregation studies on similarly prepared control platelets. When only one concentration of agonist was used, it was a concentration found to give maximal aggregation response. When two concentrations of agonist were used, the higher concentration was found to yield maximal aggregation response while the lower concentration was found to yield sub-maximal to no aggregation response. It was found that ionomycin has a relatively narrow range of activity separating non-activation, activation and cell lysis. The difference in the concentrations used 125 for the aequorin and fura 2 studies reflect a different sensitivity of the two different platelet preparations to ionomycin activation. When loaded with fura 2, the canine platelets had increased aggregation responses to agonist stimulation, as indicated by lower threshold for agonist concentration and increased aggregation rate and maximum percent aggregation, when compared to similarly manipulated platelets not loaded with fura 2. Compared to non-loaded platelets, the fura 2-loaded platelets also had an increased incidence of spontaneous shape change following the initiation of suspension stirring. Increased aggregation and secretion responses have also been reported in human platelets loaded with fura 2.218 Similar aggregation findings were not noted in aequorin-loaded platelets, although the platelets did have an increased incidence of spontaneous shape change when compared to non- loaded platelets. The [ca++]1 listed in all tables is the maximal or peak post- stimulation [Ca++]i measured in the presence or absence of external Ca++. As mentioned above, the peak [Ca++]1 in the presence of external Ca++ is indicative of Ca++ flux due to both release from internal stores and influx. The [Ca++]1 in the presence of external EGTA is solely indicative of Ca++ release from internal stores. The figures of fluorescence or luminescence are representative tracing for both control and affected platelets. The quantity of external fura 2 that had leaked from the platelets during the course of the experiment was assessed by centrifugation of 126 platelet suspensions at set time periods and measurement of the fluorescence of the supernatant. The leakage rate for control platelets was 3.7%/hr while the leakage rate for affected platelets was 3.5%/hr. The fluorescence of the supernatant showed minor increases in fluorescence following the addition of excess Ca++, suggesting that it was composed of both Ca++-sensitive and -insensitive forms. Resting [Ca++]1 There was no significant difference between resting [Ca++]1 in control or affected platelets using aequorin in the presence or absence of external Ca++. The measured [Ca++]i in the presence of external Ca++ was significantly larger than the measured [Ca++]1 in the absence of external Ca++, but there was no significant difference between affected and control platelets. The results are summarized in Table 1. Table l. Resting [Ca++]1 (uM) in control and affected aequorin-loaded platelets. External Buffer Control Affected (n - 3) (n — 4) lmM ca” 3.6 i- 0.4 3.2 1 0.3 lmM EGTA 2.2 i 0.3 2.0 i 0.2 Values - mean (uM) 1 SD; n - number of dogs tested. 127 Using fura 2-1oaded platelets, there was also no significant difference between resting [Ca++]i in affected and control platelets in either Ca++-or EGTA-containing buffer. As with aequorin, the [Ca++]i in the presence of external Ca++ was greater than the [Ca++]1 in the absence of external Ca++ in both groups of platelets. The resting [Ca++]1 was compared at the beginning and the end of each day's series of experiments to assess the effect of the leakage of dye from the cells. There was no significant difference in either affected or control platelets, suggesting a minimal effect of dye leakage on the calculated [Ca++]i. The results are presented in Table 2. Table 2. Resting [Ca++]i (nM) in control and affected fura 2-loaded platelets. External Buffer Control Affected (n - 4) (n ' 5) lmM Ca++ Initial [0a++]1 . 133.9 1 13.5 146.0 i 19.9 Final [0a++]i 130.6 1 16.0 140.0 i 9.8 lmM EGTA Initial [0a++]i ' 86.6 i 14.6 78.7 i 9.9 Final [Ca++]1 75.7 i 7.1 73.7 i 9.1 Values - mean (nM) i SD; n - number of dogs tested. 128 Adenosine Diphosphate (ADP) While there was no significant difference between affected and control aequorin-loaded platelets in the peak [Ca++]i following stimulation with high concentrations (lOuM) of ADP in either Ca++- or EGTA-containing buffer, there was a marked trend for the values to be greater in the control dogs. In Ca++-containing buffer, the post- stimulation [Ca++]i was significantly increased over resting values in both groups of platelets. Although a distinct post-stimulation, luminescent peak was measurable in EGTA-containing buffer, the difference between resting and post-stimulation [Ca++]1 was not statistically significant. The results are summarized in Table 3. There was no consistent difference in the appearance of the luminescence tracings in control or affected platelets; however, 50% of the control samples had a biphasic curve. This response was never noted in affected samples. See Figure 1 for representative tracings. Table 3. Peak [Ca++]1 (uM) following stimulation with lOuM ADP in control and affected aequorin-loaded platelets. External Buffer Control Affected (n - 3) (n - 4) lmM Ca++ 6.0 i 1.1 # 4.6 i 0.4 #' lmM EGTA 2.5 i 0.4 2.2 i 0.4 Values - mean (uM) i SD; n - number of dogs tested; # - significantly different from resting, at p < 0.05. 129 affected control relative luminescence units a L l |-60 sacs-l Figure 1. Representative luminescence tracings for the response of aequorin-loaded control and affected platelets to lOuM ADP in the presence of external CaT+. Arrow - addition of agonist. Using fura 2 as the Ca++-indicator, there was significantly greater post-stimulation [ca++]1 in affected platelets in response to both low (2uM) and high (lOuM) concentrations of ADP when Ca++ was present in the external buffer. This difference was no longer found when EGTA was present in the external buffer. In this situation, there was a trend for lower post-stimulation [ca++]1 in affected platelets, but the difference was not statistically significant. The stimulated [Ca++]1 was significantly increased over resting [Ca++]1 in all experimental groups. See Table 4 for a summary of the results. 130 Table 4. Peak [Ca++]1 following stimulation with ADP in control and affected fura 2-loaded platelets. Buffer Agonist Control Affected (n-4) (n- 5) lmM Ca'H' lOuM ADP 311.0 1 19.1 # 399.5 f 26.9 * # 2uM ADP 279.3 i 66.7 # 341.6 i 23.3 * # lmM EGTA lOuM ADP 125.2 i 15.4 # 114.0 i 14.4 # 2uM ADP 116.7 i 25.8 # 100.6 i 12.5 # Values - mean (nM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # — significantly different from resting, p < 0.05. In EGTA-containing buffer, the appearance of the fluorescence curves in response to both concentrations of ADP was similar in affected and control platelets. The curves were characterized by a sharp increase in fluorescence with a slower decline to resting levels. In Ca++-containing buffer, the ADP-induced fluorescence curves differed in shape between affected and control platelets. The curves in affected platelets had a broader peak with a slower decline toward resting when compared to the curves in control platelets. See Figure 2 for representative tracings. 131 ' I ' affected control I H \ relatlve fluorescence units ' 1 time 1«so secs-I Figure 2. Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to lOuM ADP in the presence of external Ca++. Arrow - agonist addition. Platelet-Activating Factor (PAF) Using aequorin as the Ga++ indicator, there was no significant difference between control and affected platelets in peak [Ca++]1 following stimulation with either high (1.0uM) or low (0.01uM) concentrations of PAF in.the presence or absence of external oath There was a trend for greater post-stimulation [Ca++]1 in control platelets in response to either concentration of PAF in EGTA-containing buffer or low concentrations of PAF in Ca++-containing buffer. The post-stimulation [ca++]1 was significantly increased relative to resting values in affected and control platelets following the addition of all concentrations of PAF tested. See Table 5 for a summary of the results. 132 Table 5. Peak [Ca++]i (uM) following stimulation with PAF in control and affected aequorin-loaded platelets. Buffer Agonist Control Affected (n- 3) (n-h) lmM 0a++ 1.0 uM PAF 16.7 i 2.9 # 16.6 i 2.8 # 0.01uMPAF 9.13:1.8 # 7,411.03: lmM EGTA 1.0 uM PAF 5.2 i 0.8 # 4.2 i 0.5 # 0.01 uM PAF 5.0 i 0.9 # 4.0 :t 0.6 # Values - mean (uM) i SD; n - number of dogs tested; # - significantly different from resting, p < 0.05. There was no consistent difference in the appearance of the luminescent tracings. See Figure 3 for representative tracings. In contrast, using fura 2 as the Ca++ indicator, the post- stimulation [Ca++]1 was significantly increased in affected platelets relative to control platelets following the addition of high concentrations of PAF in the presence of external Ca++. There was no difference between the two groups of platelets in response to low concentrations of PAF. In EGTA-containing buffer, the results were reversed. The peak post-stimulation [Ca++]1 was significantly decreased in affected platelets relative to control platelets in response to high and low concentrations of PAF. In both the presence and absence of external Ca++, the post-stimulation [Ca++]1 was significantly increased over resting levels in response to all concentrations of PAF in control 133 . affected control relative luminescence units :1. met .11. time 0 sacs-I Figure 3. Representative luminescence tracings of aequorin-loaded affected and control platelets in response to luM PAF in the presence of external Ca++. Arrow - agonist addition. and affected platelets. See Table 6 for a summary of the results. There was no difference in the appearance of the fluorescence curves between control and affected platelets. In the presence of external Ca++, the fluorescence curves were characterized by a sharp peak which quickly decreased to a level greater than resting. When external Ca‘++ was removed, the curves were characterized by a sharp increase which then decreased gradually toward resting levels. See Figure 4 for representative tracings. 134 Table 6. Peak [Ca++]1 in response to stimulation with PAF in control and affected furs-2 loaded platelets. Buffer Agonist Control Affected (n - 4) (n - 5) lmM ca” 1.0 uM PAF 676.8 5: 98.7 # 842.9 1- 47.8 * # 0.01 uM PAF 485.9 i 74.5 # 465.2 i 71.1 # lmM EGTA 1.0 uM PAF 248.6 3: 25.5 # 176.6 3: 9.0 * # 0.01 uM PAF 218.1 1 18.2 # 164.8 i 17.9 s # Values - mean (nM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from resting, p < 0.05. relative fluorescence unlts A. .A _A—‘ 'w‘V—‘T affected v' '7 tr. 4 A“ control [-50 secs-l tune Figure 4. Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to luM PAF in the presence of external Ca++. Arrow - addition of agonist. 135 Thrombin Using aequorin-loaded platelets, there was no significant difference between control and affected platelets in the post- stimulation [Ca++]i following the addition of high (0.25U/m1) or low (0.01U/m1) concentrations of thrombin in either the presence or absence of external Ca++. There was a trend for the post-stimulation [Ca++]1 to be greater in the control platelets than in the affected platelets. In all cases, post-stimulation [Ca++]1 was significantly elevated over resting concentrations. See Table 7 for a summary of the results. The luminescence tracings were similar in appearance in both control and affected platelets. See Figure 5 for representative curves. Table 7. Peak [Ca++]i (uM) following addition of thrombin in control and affected aequorin-loaded platelets. Buffer Agonist Control Affected (rt-3) (rt-4) lmM Ca++ 0.25U/ m1 Thrombin 13.4 i 3.3 # 12.1 i 0.9 # 0.01U/ m1 Thrombin 6.1 i 0.7 # 5.5 i 1.0 # lmM EGTA 0.25U/ ml Thrombin 5.5 i 1.4 # 3.9 t 0.4 # 0.01U/ ml Thrombin 5.1 i 1.5 # 3.5 i 0.3 # Values - mean (uM) i SD; n! number of dogs tested; # - significantly different from resting, p < 0.05. 136 . affected control “—1)- . A . time r60 SECS’I relative luminescence units Figure 5. Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 0.25U/m1 thrombin in the presence of external Ca++. Arrow - addition of agonist. Using fura 2 as the Ca++ indicator, there was no significant difference in the post-stimulation [Ca++]1 between control and affected platelets following the addition of either concentration of thrombin. In EGTA-containing buffer, there was significantly decreased post- stimulation [CaT+]1 in affected platelets relative to control platelets following stimulation with high concentrations of thrombin. Following the addition of low concentrations of thrombin, there was a trend for higher post-stimulation [Ca++]1 in control platelets; however, the difference was not statistically significant. In both groups, all post- stimulation values were significantly increased relative to resting [Ca++]1. See Table 8 for a summary of the results. 137 Table 8. Peak [Ca++]1 (nM) following thrombin stimulation in control and affected fura 2-1oaded platelets. Buffer Agonist Control Affected (n - 4) (n - 5) lmM Ca'H' 0.250/ ml Thrombin 744.4 1 88.7 4 696.2 '1 53.2 # 0.01U/ ml Thrombin 292.1 i 42.6 # 299.4 i 31.2 # lmM EGTA 0.25U/ m1 Thrombin 235.3 i 5.5 # 161.8 .+. 15.3 * # 0.01U/ ml Thrombin 155.4 i 27.9 # 124.3 i 20.9 # Values - mean (nM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from resting, p < 0.05. The appearance of the fluorescence curves in control and affected platelets was similar in the EGTA-containing buffers. In response to high concentrations of thrombin, there was a sharp fluorescence peak which gradually decreased to baseline levels. Following low concentrations of thrombin, the fluorescence peak was broader but otherwise similar. In Ca++-containing buffer, the fluorescence curves appeared slightly different in response to high concentrations of thrombin. The affected platelets reached a similar peak to that of control platelets; however, the level of fluorescence remained relatively stable in control platelets while it tended to gradually decrease in affected platelets. Following stimulation with low concentrations of thrombin, the fluorescence curves were similar in appearance. See Figure 6 for representative fluorescence curves. 138 control *<:v¥=:=r;.aanuu~‘-v affected relaflve fluorescence unfls “"19 l—so sacs-l Figure 6. Representative fluorescence tracings of fura 2-1oaded affected and control platelets in response to 0.25U/ml thrombin in the presence of external Ca++. Arrow - addition of agonist. Ionomycin Using aequorin-loaded platelets, there was no significant difference between control and affected platelets in the post- stimulation.[Ca++]1 in response to high (2uM) or low (0.5uM) concentrations of ionomycin in the presence of external Ca++. In both groups, the post-stimulation [Ca++]1 was significantly increased over resting values. In EGTA-containing buffer, the peak [ca++]1 was significantly decreased in affected platelets relative to control platelets following the addition of high concentrations of ionomycin. The post-stimulation [Ca++]1 of both affected and control platelets was 139 significantly increased over resting values. There was no significant difference between affected and control dogs in post-stimulation [Ca++]i following the addition of low concentrations of ionomycin. Though there was a trend for the stimulated [Ca++]1 to be increased relative to resting values, the difference was not statistically significant. See Table 9 for a summary of the results. There was no consistent difference between affected and control platelets in the appearance of the luminescence tracings in any of the experimental groups in response to ionomycin. See Figure 7 for representative tracings. Table 9. Peak [Ca++]1 (uM) following addition of ionomycin in control and affected aequorin-loaded platelets. Buffer Agonist Control Affected (n - 3) (n - 5) lmM Ca++ 2.0 uM ionomycin 10.1 i 2.0 # 10.6 i 2.1 # 0.5 uM ionomycin 5.6 i 1.3 # 5.3 i 0.8 # lmM EGTA 2.0 uM ionomycin 3.2 i 0.3 # 2.6 i 0.3 * # 0.5 uM ionomycin 2.7 i 0.3 2.3 i 0.3 Values - mean (uM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from resting, p < 0.05. 140 ,. affected control relative luminescence units F r time [-60 sacs-I Figure 7. Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 2uM ionomycin in the presence of external Ca++. Arrow - addition of agonist. Using fura 2-loaded platelets in the presence of external ca++, there was no significant difference between control and affected platelets in the maximum post-stimulation [ca++]1 in response to high (5uM) or low (2uM) concentrations of ionomycin. When compared to resting values, the post-stimulation [Ca*+]1 was significantly increased in both affected and control platelets. In EGTA-containing buffer, the post-stimulation [Ca++]i was significantly decreased in affected platelets in response to high concentrations of ionomycin. There was no significant difference between the two groups following the addition of low concentrations of ionomycin. All post-stimulation values for [ca++]1 were significantly increased over resting levels. See Table 10 for a summary of the results. 141 Table 10. Peak [Ca++]i (nM) in response to ionomycin in control and affected fura 2-loaded platelets. Buffer Agonist Control Affected (n - 4) (n - 5) lmM 0a++ 5.0 uM Ionomycin 817.6 1' 252.2 # 609.4 i 47.5 # 2.0 uM Ionomycin 315.8 i 37.8 # 318.0 i 19.7 # lmM EGTA 5.0 uM Ionomycin 141.6 i 1.3 # 119.3 i 14.7 * # 2.0 uM Ionomycin 110.4 i 9.0 # 96.9 i 8.8 # Values - mean (nM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from resting, p < 0.05. There was no significant difference in the appearance of the fluorescence tracings between affected and control platelets. In the presence of external Ca++, there was a relatively rapid increase to peak fluorescence levels which were maintained throughout the test period. In the absence of external Ca++, there was a similar increase to peak fluorescence levels; however, the fluorescence levels gradually decreased over the test period. See Figure 8 for representative curves. 142 control £3 affected relaflve fluorescence unlts time '- 50 SECS‘ Figure 8. Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to SuM ionomycin in the presence of external Ca‘H'. Arrow - addition of agonist. Phorbol Myristate Acetate (PMA) Using aequorin as the ca” indicator, the post-stimulation [Ca'H'11 in response to 3uM PMA was significantly less in the affected platelets than in control platelets in the presence of external Ca‘H'. The post- stimulation [Ca‘H’h in both groups was significantly increased over resting concentrations . In EGTA-containing buffer, there was no significant difference between the two groups of platelets, although there was a general trend for higher post-stimulation [Ca'H']1 in control platelets. Though a luminescence peak was always detectable, the 143 calculation of [Ca++]1 yielded values that were not statistically different from resting concentrations. See Table 11 for a summary of the results. There was no difference in appearance of the luminescence tracings between affected and control platelets. See Figure 9 for representative tracings. T— affected control relative luminescence units [-60 secs-I Figure 9. Representative luminescence tracings of aequorin-loaded control and affected platelets in response to 3uM PMA in the presence of external Ca++. Arrow - addition of agonist. 144 Table 11. Peak [Ca++]1 (uM) in response to 3uM PMA in control and affected aequorin-loaded platelets. Buffer Control Affected (n - 3) (n - 4) Ca++ 6.6 i 1.1 # 4.7 t 0.5 * # lmM EGTA 2.7 i 0.4 2.3 i 0.4 Values - mean (uM) i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from resting, p < 0.05. In contrast to the increased [Ca++]i using aequorin, the PMA- induced post-stimulation [Ca++]1 was decreased compared to resting values in both control and affected fura 2-loaded platelets in the presence of external Ca++. The decrease was of similar magnitude in both affected and control platelets. In an EGTA-containing buffer, the post-stimulation [Ca++]i was also decreased relative to resting levels in response to PMA in both groups; however, the decrease was only statistically significant in the control platelets. See Table 12 for a summary of the results. There was no difference between the affected and control platelets in the appearance of the fluorescent tracings. In the presence of external Ca++, the curves were characterized by a moderate decrease in fluorescence which was maintained for the time period recorded. In the absence of external Ca++, the decrease in fluorescence was less distinct and, in some cases, absent. See Figure 10 for representative tracings. 145 Table 12. Peak [Ca++]i in response to 3uM PMA in control and affected fura 2-loaded platelets. External Buffer Control Affected (n - 4) (n - 5) lmM Ca'H’ 101.4 i 6.8 # 97.5 i 15.6 # lmM EGTA 56.2 i 4.2 # 59.1 i 7.6 Values - mean (nM) i SD; n - number of dogs tested; # - significantly different from resting, p < 0.05. WW relative fluorescence units time [—60 sacs—i ' Figure 10. Representative fluorescence tracings of fura 2-loaded control and affected platelets in response to 3uM PMA in the presence of external CaI+} Arrow - addition of agonist. 146 DISCUSSION Changes in [Ca++]i play a pivotal role in a number of important events or processes in platelet activation. First, increased [Ca++]i stimulates MLCK activity which causes the subsequent phosphorylation of myosin light chain.353 Phosphorylation of myosin light chain is correlated with granule centralization, the generation of a contractile ++]i force and shape change.95’142 Secondly, elevated [Ca acts synergistically with PKC to promote granule secretion.159'194'380 Thirdly, cytosolic Ca++ is essential for the proper functioning of a number of enzymes which have important roles in platelet activation. Increased [Ca++]i is associated with activation of PLA2397 and the Ca++- dependent proteases (calpains)366 while resting to increased [Ca++]1 is required for PLC23'25 and PKC194 activity. Finally, cytosolic Ca++ is involved in the organization and composition of the platelet cytoskeleton.377 Resting and post-stimulation [Ca++]1 was measured in the platelets from dogs with Basset hound hereditary thrombopathy (BHT) and from control Basset hounds because previous findings suggest that the platelet disorder in BHT is due to a defect(s) in signal transduction or processing. In the current study, two different techniques were used to measure [Ca++]1: (l) the fluorescent dye fura 2; and (2) the luminescent photoprotein aequorin. It has been proposed that fura 2 and aequorin provide unique, but complimentary, information about changes in [Ca++]1.190 Aequorin responds to regional elevations in [Ca++]1 with a corresponding exponential increase in the luminescent signal. While 147 aequorin effectively detects local zones of Ca++ change, it may not accurately reflect the average [Ca++]i.97 Although aequorin is primarily located in the cytosol and not in the internal organelles, it has been proposed that bulk of the intracellular aequorin is closely associated with the plasma membrane and not distributed diffusely throughout the cytosol.190 In contrast, fura 2 is distributed throughout the cytosol.190 In some cells, fura 2 compartmentalizes in organelles; however, this has not been found to be a problem in platelets in suspension.79*190 While fura 2 is not able to detect local Ca++ changes, it provides a better indication of average cytosolic Ca++ fluxes.188'190 The difference between the pool of Ca++ being measured using the two techniques is best illustrated by a comparison of the signals obtained using aequorin- and fura 2-loaded platelets. Aequorin gives a rapid luminescence peak which generally drops back to baseline levels within 60-90 seconds, irrespective of the presence or absence of external Ca++. The rapid decrease in the aequorin luminescent signal cannot be explained by irreversible oxidation of the dye following reaction with Ca++ since the addition of a second agonist following the initial luminescence peak elicits another luminescent response.190 It has been proposed that the shortness of the signal is due to rapid diffusion and/or sequestration of Ca++ from the region of dye location.190 Fura 2 gives a rapid peak of fluorescence which tends to plateau before slowly decreasing toward baseline. This plateau is sustained much longer in the presence of external Ca++. 148 A comparison of resting [Ca++]i yielded no significant differences between the platelets of control and affected dogs using either aequorin-loaded or fura 2-loaded platelets, suggesting that affected platelets maintain normal resting Ca++ homeostasis. It is important to note the difference in the magnitude of the [Ca++]1 using the two techniques. The [Ca++]i calculated using aequorin was 20 to 30 fold greater than the value calculated using fura 2. The values for resting [Ca++]i in the current study were within the range of resting [Ca++]i measured using aequorin188 and fura 2218'313'315 in human platelets. The discrepancy between the [Ca++]i calculated using fura 2 and aequorin can be partially explained by the effect of intracellular Mg++ on the aequorin signal. Mg++ competitively binds to aequorin without producing measurable luminescence and causes decreased sensitivity to Ca++.79'44 Though the [Mg++] in canine platelets is not known, the calibration curve used to calculate [Ca++]i for the current study was constructed in a medium containing lmM Mg++.. Recent studies in human platelets have shown that the cytosolic [Mg++] is closer to 0.3mM. In the reported studies, calculation of [Ca++]i using an aequorin calibration curve constructed in a medium containing 0.25mM Mg++ yields resting platelet [Ca++]1 of approximately 300nM, much closer to the most frequently published [Ca++]1 of approximately lOOnM in fura 2-loaded platelets.440 The calculation of [Ca++]1 in fura 2-loaded platelets is also affected by the [Mg++], but to a much lesser extent than aequorin since fura 2 has a relatively high selectivity for Ca++ compared to Mg++. The dissociation constant (Kd) for fura 2 used to calculate 149 [Ca++]1 in the present study also assumed an intracellular Mg++ concentration of lmM.79 The absolute value for [Ca++]i calculated using both dyes should be interpreted with caution. The fluorescence of fura 2 is dependent on the microviscosity of the medium surrounding the dye, therefore the fluorescence of fura 2 in the platelet cytosol differs from the fluorescence of the same amount of dye in the lysed platelet solution. Since the equation used to calculate [Ca++]1 is based on the comparison of dye fluorescence in the platelet cytosol to the fluorescence of dye in the lysed platelet suspension, the calculation of absolute [Ca++]i has inherent error.148'332 In aequorin-loaded platelets, the dye localizes in regions of high Ca"'+ concentrations, such as under the plasma membrane. Therefore, the [Ca++]1 measured does not reflect average [Ca'*"*']i.188'190 The absolute value for [Ca++]1 was not of primary interest in the current study since the study was designed to compare resting and post-stimulation [Ca++]1 in control and affected platelets using the same experimental techniques for each group. A second interesting feature of the current study was the significant difference between [Ca++]1 in Ca++- and EGTA-containing buffers in both aequorin- and fura 2-1oaded platelets. It was most likely a feature of experimental design rather than a true physiologic difference since platelet resting [Ca++]1 is independent of large changes in external [Ca*""].51'331 One study that compared the extent and rate of decay of the aequorin signal in the presence and absence of external Ca++ has shown that the aequorin loaded into platelets is 150 partly accessible to external Ca++. It was not determined whether this accessibility reflects the natural distribution of Ca++ in the cytoplasm and/or plasma membrane or is due to alterations in the platelet handling of Ca++ secondary to the permeabilization procedures used to load aequorin.217 Other groups have proposed that the difference in resting [Ca++]i in Ca++- and EGTA-containing buffers reflects the effect of a few "leaky platelets" on the luminescence signal. Since the [Ca++]i is measured in a platelet suspension, it is possible that a small number of injured cells have increased membrane permeability, thereby exposing their intracellular aequorin to the [Ca++] present in the external environment. In Ca++-containing buffer, these cells would be exposed to a relatively high [Ca++]. The kinetics of the aequorin reaction make the luminescence signal of platelets in suspension susceptible to dominance by a few damaged cells with large luminescence signals.19o'l‘l‘O The difference in resting [Ca++]1 between Ca++- and EGTA- containing buffers was also noted in the fura 2 preparations. As with aequorin, the difference may be partly due to damage to cells during experimental manipulation and increased membrane permeability to Ca++. Fura 2 is not as susceptible as aequorin to dominance of the fluorescence signal by a few ”leaky platelets” in the platelet suspensions. Studies on quin2-loaded platelets have shown that the difference between resting [Ca++]1 in the presence and absence of external Ca++ is mainly due to buffering of internal Ca++ by quin2 itself. When external Ca++ is present, the buffering of quin2 is partly compensated for by the subsequent influx of external Ca”.190 This buffering effect is not as significant in fura 2-loaded platelets since 151 lower concentrations of fura 2 are loaded.u‘8v332 Another aspect for consideration is the effect of the fura 2 itself. Low concentrations of cytosolic fura 2 potentiate the aggregation and secretion responses of human platelets.218 This finding was duplicated in the current study during pilot aggregation trials on fura 2-loaded canine platelets. The fura 2-loaded platelets had increased aggregation responses to all agonists tested when compared to non-loaded platelets. The loaded platelets also had a greater tendency, in the presence of external Ca++ and Mg++, to undergo spontaneous shape change upon stirring, suggesting that [Ca++]1 is elevated. Finally, in the current study, the difference in the resting [Ca++]1 could be partially explained by leakage of fura 2 into the external medium. Due to the relatively low leakage rate, the calculation of [Ca++]1 did not correct for the presence of external dye. The fluorescence of the external dye was only mildly affected by the presence or absence of Ca++, indicating that the form of dye in the supernatant consisted of both Ca++-sensitive and Ca++-insensitive forms. If leakage of dye was the main factor involved in the difference between the resting [Ca++]1 in the presence and absence of external Ca++, it would be anticipated that resting [Ca++]1 would increase over the time period that the experiments were run. This was not found; instead, there was no significant difference in the resting [Ca++]i calculated at the beginning and the end of each day's set of experiments. ADP is a weak physiologic agonist which requires primary aggregation and the release of arachidonic acid metabolites for secretion and irreversible aggregation to occur.377 It has been proposed that ADP activates platelets by stimulating Na+-H+ exchange and 152 causing a local, membrane-associated increase in pHi. The localized alkalinization activates PLA2 which mediates the release of small quantities of arachidonic acid. The arachidonic acid is metabolized via the cyclooxygenase pathway to yield endoperoxides and TxA2 which amplify the response by binding to membrane receptors and activating PLC.409'4101412 Inhibition of cyclooxygenase blocks the production of arachidonic acid metabolites in response to ADP but does not prevent primary aggregation412 or Ca++ mobilization92'229’290’394, suggesting these responses are independent of the TxA2 pathway. Kinetic studies on Ca++ mobilization in fura 2-loaded platelets have demonstrated that, in the presence of external Ca++, ADP stimulation induces an almost instantaneous increase [Ca‘*""]j_.351'352 This suggests that there is a direct association, independent of the production of diffusible second messengers, between the ADP receptor and the plasma membrane Ca++ channel. The nature of this association is not known; however, some evidence implicates the involvement of a G protein.331'352 In the absence of external Ca++, there is a lag period between ADP binding and Ca++ release, suggesting a requirement for the production of second messengers to mediate the release of Ca++ from internal stores.351'352 The ADP-induced release of internal Ca++ is partially inhibited by agents which block Na+/H+ exchange.394 In the present study, a comparison of ADP-induced post- stimulation [Ca++]1 in EGTA-containing buffer yielded no significant differences between control and affected platelets using either fura 2 or aequorin, though there was a trend for values to be decreased in 153 affected platelets. These findings suggest that affected platelets have normal to slightly decreased ability to mobilize internal Ca++. In the presence of external Ca++, there was significantly increased post- stimulation [Ca++]i in fura 2-loaded affected platelets relative to control platelets. These findings indicate that affected platelets have increased ability to stimulate Ca++ influx following stimulation with ADP when compared to control platelets. In contrast, in the presence of external Ca++, affected aequorin-loaded platelets had decreased post- stimulation [Ca++]1 relative to control platelets. These findings suggest that affected platelets have normal to decreased influx of external Ca++ across the plasma membrane in response to ADP. It must be remembered that the Ca++ signal measured using aequorin is short lived and likely represents the flux in submembranous regions. The Ca++ signal measured by fura 2 is longer-lived and represents the overall cellular response.188o190 Caution must be used in the interpretation of changes in [Ca++]i in the presence of external Ca++ due to the differing effects of aggregation on the signals measured using aequorin- or fura 2-1oaded platelets. Control platelets are able to aggregate in response to ADP. Affected platelets do not undergo primary or secondary aggregation in response to ADP, though shape change and the rapid secretion of dense granule ATP occur.254 The fura 2 fluorescence signal requires a- homogeneous suspension for accurate measurement and is artificially suppressed by the formation of platelet aggregates in cell suspensions.190 Although aggregation was not noted by gross examination of the control samples, the presence of microaggregates of the control 154 platelets cannot be ruled out. It has been shown in human platelets that ADP induces the formation of microaggregates in washed platelets in the absence of exogenous external fibrinogen.190 The presence of microaggregates would artificially suppress the fluorescence signal in control platelets relative to affected platelets. The formation of microaggregates could also explain the difference in the appearance of the fluorescence curves between affected and control platelets. The more rapid decrease of the fluorescence signal in the control group could coincide with the formation of aggregates. In aequorin-loaded platelets, the luminescence signal itself is not affected by aggregation;189 however, a recent study in aequorin-loaded platelets has shown that the stimulation of Ca++-mobilization by weak agonists, such as ADP, requires aggregation-dependent amplification for full Ca++ responses to occur.229 Lack of aggregation, therefore, would suppress the luminescence signal in the affected platelets relative to control platelets. PAF is a physiologic agonist of intermediate strength. Secretion occurs independent of prior aggregation or the production of arachidonic acid metabolites; however, treatment of platelets with cyclooxygenase inhibitors or ADP scavengers reduces the extent of the aggregation and secretion responses.377 PAF causes activation of PLC, hydrolysis of membrane phosphoinositides and the production of the second messengers (1,4,5)IP3 and l,2-DG.106'283'37S Kinetic studies have demonstrated a lag period between agonist binding and the initiation of increased [Ca++]1, suggesting a requirement for the production of second messengers to mediate both Ca++ release from internal stores and the 155 influx of external Ca'H'.352 PAF-induced Ca++ mobilization is almost completely independent of TxA2 production and cyclooxygenase inhibition.15S In the presence of external Ca++, the post-stimulation [Ca++]i in response to high concentrations of PAF was increased in fura 2-1oaded affected platelets when compared to control platelets. There was no difference between the two groups using aequorin-loaded platelets. As discussed under ADP, part, if not all, of the difference between the two groups of fura 2-loaded platelets may be due to the formation of platelet aggregates in the control samples which would artificially lower the fluorescence signal. In affected platelets, PAF stimulation causes shape change but not primary or secondary aggregation.254 Therefore, the fluorescence signal would not be affected by the formation of aggregates. The lack of difference using aequorin-loaded platelets is consistent with this hypothesis since the aequorin luminescent signal is not directly affected by aggregation;190 however, the presence of aggregation in the control samples may reinforce or amplify the Ca++ response to PAF.229‘ In EGTA-containing buffer, post-stimulation [Ca++]1 was decreased in affected platelets following stimulation with either concentration of PAF, suggesting that affected platelets do not mobilize internal ca++ stores as well as control platelets. In spite of lower post-stimulation [Ca++]1, affected platelets mobilize sufficient quantities of intracellular Ca++ to cause shape change254 and the Ca++-dependent phosphorylation of the 20kDa myosin light chain similar to control 156 platelets.255 Phosphorylation of myosin light chain is associated with activation of MLCK,177'3S3 a Ca++-calmodulin dependent enzyme.164 Thrombin is a strong physiologic agonist that induces secretion and aggregation responses which are not affected by cyclooxygenase inhibition.377 Thrombin stimulates platelet by causing activation of PLC.90'262'263'346 Thrombin also initiates platelet aggregation and secretion independent of PLC activation and phosphoinositide activation.57'226 Thrombin mobilizes intracellular and extracellular Ca++, though the bulk of the increase in post-stimulation [Ca++]1 comes from the influx of external Ca++.192'243'438 Kinetic studies have shown a lag period between the addition of thrombin and the start of Ca++ mobilization, suggesting that it is the production of second messengers which actually mediate the mobilization of both internal and external Ca++.351'352'476 In the presence of external Ca++, the thrombin-induced post- stimulation [Ca++]1 was not significantly different between affected and control groups using either aequorin- or fura 2-1oaded platelets. It is important to note that thrombin causes aggregation and dense granule secretion in both affected and control platelets, negating the selective effect of aggregation on the Ca++ response in control platelets.254 In EGTA-containing buffer, the post-stimulation [Ca++]1 was significantly decreased in affected platelets loaded with fura 2. The same trend was seen using aequorin, though the difference was not statistically significant. This indicates that affected platelets mobilize Ca++ as well as control platelets only when Ca++ influx from the external 157 environment occurs while Ca++ mobilization from internal stores is impaired. The amount of internal Ca++ release is large enough to cause shape change and the Ca++-dependent phosphorylation of the 20kDa protein.255 Ionomycin and calcium ionophore A23187 are non-physiologic agonists that transport Ca++ across cell membranes.457 The ionophores cause sufficient Ca++ release from the DTS to initiate aggregation and secretion independent of the influx of external Ca++; however, the presence of external Ca++ potentiates the responses.121'457 The ++ii activates a number of Ca++- ionophore-induced elevation in [Ca dependent enzymes, including PLA2. Activation of PLA2 mediates the release of the arachidonic acid metabolites314 from membrane phospholipids and the subsequent activation of PLC, hydrolysis of membrane phosphoinositides and production of intracellular second messengers.337-339'444 The ionophore-induced activation of PLC is blocked by cyclooxygenase inhibition while the increase in [Ca++]1 is not prevented.337’339 In the presence of external Ca++, there was no significant difference between affected and control platelets in the post- stimulation [Ca++]1 in response to ionomycin using aequorin- or fura 2- loaded platelets. There was a tendency for the post-stimulation [Ca++]£ to be decreased following stimulation with high concentrations of ionomycin in fura 2-loaded affected platelets; however, the difference was not statistically significant due to the wide range of responses within the groups. Stimulation of affected platelets with calcium 158 ionophore A23187 causes shape change without significant aggregation or secretion of dense granule ATP.254 Since, in the current study, the high concentration of ionomycin caused visible aggregation of control platelets but not affected platelets, it is possible that the formation of aggregates artificially lowered the fluorescence signal selectively in the control platelets. This suggests that post-stimulation [Ca++]i in the control platelets could be higher than the calculated value and that the presence of aggregates negated any difference in the post- stimulation Ca++ response to ionomycin. In EGTA-containing buffer, there was significantly decreased [Ca++]i both aequorin- and fura 2-1oaded affected platelets following stimulation with high concentrations of ionomycin, indicating decreased Ca++ release from internal stores. Calcium ionophores cause a non- receptor-mediated release of Ca++ from the DTS which is separate from the receptor-mediated release by the inositol phosphates.ll'313'377 In spite of a defect in the internal mobilization of Ca++ in response to ionomycin, the amount of Ca++ released by calcium ionophores is sufficient to cause shape change and phosphorylation of the 20kDa and 47kDa proteins.255 Phorbol myristate acetate is a non-physiologic agonist which causes platelet activation by direct stimulation of PKC by increasing the affinity of PKC for Ca++ and phospholipid.71'72 It was originally reported that phorbol esters initiate platelet activation in a Ca++- independent manner since the fluorescent dyes quin2 or fura 2 do not detect any increase in sample fluorescence in PMA-stimulated 159 platelets.117'439 Studies using aequorin-loaded platelets have identified a post-stimulation Ca'H' peak in response to phorbol esters.1171439 The mechanism for this PMA-induced Ca'H' flux is not known. The marked discrepancy between the [Ca'H'h measured using aequorin and that measured using the fluorescent chelator dyes is not easily resolved but illustrates the fact that the techniques are measuring different aspects of Ca'H' metabolism. It has been suggested that aequorin is measuring local elevations in [Ca‘H’]1, most likely submembranous in origin, which are not detectable by the use of fura 2 01' quin2 . 190 In the presence of external Ca++, there was increased post- Stimulation [CaHh in response to PMA using aequorin-loaded platelets sii-lllilar to that reported in the literature."37 This response was significantly decreased in the affected platelets when compared to c‘Tntzrol platelets. In the presence of external EGTA, there was no 81gtlificant difference between the two groups of platelets or between the stimulated and resting [Ca'Hli within each group. This suggests two coInclusions: (1) the primary component of the PMA-induced Ca'H' is the influx of external Ca'H’ in both control and affected platelets; and (2) the influx of external Ca'H' is decreased in affected platelets in re Sponse to PMA. In the presence of external Ca++, the post-stimulation [Ca++]1 in f‘Jra 2-1oaded platelets was significantly decreased when compared to testing [Ca++]1. This decrease was of similar magnitude in both groups of platelets. This finding was expected because phorbol enhances Ca'H' 160 resequestration into internal storage sites315'471'472’473 and Ca” efflux across the plasma membrane.315'331 In addition, phorbol esters cause aggregation and secretion in both affected and control platelets,254 therefore the fluorescence signal in both groups would be equally altered. In summary, the measurement of post-stimulation [CaHh in the absence of external Ca‘H' has shown that affected platelets do not mobilize internal stores of Ca++ as well as control platelets following stimulation with PAF, thrombin and high concentrations of the calcium 1<>l§l<>phore ionomycin. The same trend was seen following stimulation with ADP; however, the difference was not statistically significant. The rasults are summarized in Tables 13 and 14. Thrombin, PAF and ADP do not directly cause release of internal Ca“; instead, these agonists require the production of intracellular 8e<2ond messengers351'352 which initiate a receptor-mediated release of intracellular Ca'H' from the DTS.331 There is also a small quantity of Ca“ which may be released from the membrane phospholipids in response to agonist stimulation.6t‘ Possible explanations for the defective release of internal Ca‘H’ in affected platelets include: (1) decreased pl'l‘oduction of second messengers or production of defective second uleSsengers which normally mediate Ca‘H' release from the DTS (primarily 1 - 4 ,5-IP3); (2) decreased ability of the DTS to respond to the second theSsengers; or (3) decreased storage of releasable Ca‘H' within the DTS. 1“- contrast, calcium ionophores bypass the physiologic, receptor- mediated release of internal Ca'H'.350 It has been shown that the 161 Table 13. Summary of resting and post-stimulation [Ca'H’]i results in control and affected platelets loaded with aequorin. p < 0.05; (+) - aggregation occurs; (+/-) - aggregation may oCCur; (-) - aggregation does not occur. Agonist Concentration Control Affected (n - 3) (1’1 " 4) lmM External Ca'H' Resting - 3.6 i 0.4 (-) 3.2 i 0.3 (-) ADP 10 uM 6.0 i 1.1 (+) 4.6 i 0.4 (-) PAF 1 uM 16.7 i 2.9 (+) 16.6 i 2.8 (-) 0.01 uM 9.1 :t 0.7 (+/-) 7.4 i 1.0 (-). Thrombin 0.25U/ml 13.4 :t 3.3 (+) 12.2 i' 0.9 (+) 0.01U/ml 6.1 i 0.7 (+/-) 5.5 i 1.0 (-) Ionomycin 2 uM 10.1 t 2.0 (+) 10.6 i 2.1 (7) 0 5 uM 5.6 i 1.3 (+/-) 5.3 1': 0.8 (-) PMA 3 uM 6.6 i 1.1 (+) * 4.7 1' 0.5 (+) 1‘93 External EGTA Resting - 2.2 i 0.3 (-) 2.0 i 0.2 (-) ADP 10 uM 2.5 i 0.4 (-) 2.2 i 0.4 (-) PAF 1 uM 5.2 i 0.8 (-) 4.2 i 0 5 (-) 0.01 uM 5.0 i 0.9 (-) 4.0 :L‘ 0 6 (-) Thrombin 0.25U/ml 5.5 i 1.4 m 3.9 i 0.4 H 0.01U/ml 5.1 i 1.5 (—) 3.5 i 0.3 (-) Ionomycin 2 uM 3.2 i: 0.3 H * 2.6 i 0.3 H 0.05 uM 2.7 i 0.3 (-) 2.3 i 0.3 (-) PMA 3 uM 2.7 i 0.4 (-) 2.3 i 0.4 (-) \ values - mean (uM) i’ S.D.; * - significantly different from control, or may not 162 Table 14. Summary of resting and post-stimulation [Ca++]1 in control and affected platelets loaded with fura 2. Agonist Concentration Control Affected (n - 4) (n - 5) lmM External 0a++ H- \D m, A I v Resting - 130.6 3 16.0 (-) 140.0 ADP 10 uM 311.0 i 19.1 (+/-) '* 399.5 i 26.9 (-) PAF 1 uM 676.8 i 98.7 (+) * 842.9 i 47.8 (+/-) 0.01 uM 485.9 i 74.5 (+/-) 465.2 1 71.1 (-) Thrombin 0.25U/ml 744.4 i 88.7 (+) 696.2 i 53.2 (+) 0.01U/ml 292.1 i 42.6 (+/-) 299.4 i 31.2 (-) Ionomyéin 5 uM 817.6 1 252.2 (+) 609.4 i 47.5 (-) 2‘uM 315.8 i 37.8 (+/-) 318.0 i 19.7 (-) PMA 3 uM 101.4 i 6.8 (+) 97.5 i 15.6 (+) lmM External EGTA Resting - ' 75.7 i 7.1 (-) 73.7 i 9.1 (-) ADP 10 uM 125.2 i 15.4 (-) 114.0 i 14.4 (-) PAF 1 uM 248.6 1 25.5 (-) * 176.6 i 9.0 (-) 0.01 uM 218.1 1 18.2 (-) * 164.8 1 17.9 (-) Thrombin 0.25U/ml 235.3 i 5.5 (~) * 161.8 i 15.3 (-) 0.01U/m1 155.4 i 27.9 (-) 124.3 i 20.9 (-) Ionomycin 5 uM 141.6 1 1.3 (-) * 119.3 i 14 7 (~) 2 uM 110.4 1 9 0 (-) 96.9 i 8 8 (-) PMA 3 uM 56.2 i 4.2 (-) 59.1 i 7.6 (-) Value - mean (nM) i S.D.; * - significantly different from control, p < 0.05; (+) - aggregation occurs; (+/-) - aggregation may or may not occur; (-) - aggregation does not occur. 163 increase [Ca++]1 following stimulation with ionomycin is not inhibited by blockage of cyclooxygenase or by removal of external ADP, suggesting that the mobilization of Ca++ is independent of the secondary stimulation of PLC and the production of second messengers.339 Therefore, the decreased post-stimulation [Ca++]i in affected platelets following the addition of ionomycin suggests that there is a defect in the storage of releasable Ca++ in the DTS. Affected platelets have increased basal intracellular concentrations of cAMP, possibly due to defective regulation of the cAMP phosphodiesterase activity.48’49 Increased cAMP could partially explain the Ca++ defect in affected platelets since cAMP inhibits agonist- induced increases in [Ca++]i by blocking both the influx of external Ca++ and the release of Ca++ from internal stores.331'350'433 Cyclic AMP also stimulates resequestration of Ca++ into the DTS following agonist-induced release.l‘3'1207167 Increased cAMP does not affect the post-stimulation [Ca++]1 following stimulation with ionomycin in platelets treated with cyclooxygenase inhibitors.350 This suggests that the decreased post-stimulation [Ca++]1 in response to ionomycin in the platelets from dogs with BHT is not due to increased cAMP; however, affected platelets were not treated with cyclooxygenase inhibitors to eliminate any secondary effects due to the generation of arachidonate metabolites. The situation in Ca++-containing buffer, where the post- stimulation [Ca++]1 represents a combination of internal release and influx of external Ca++, is less clear. Using aequorin-loaded 164 platelets, there was a trend for post-stimulation [Ca++]1 to be similar to slightly decreased in affected platelets when compared to control platelets in response to all of the agonists tested; however, the difference was only statistically significant following the addition of PMA. These findings suggest that the ability of affected platelets to generate an agonist-induced influx of external Ca++ is similar to slightly decreased when compared to control platelets. Unfortunately, the absence of aggregation in affected platelets negates the aggregation-dependent reinforcement of platelet Ca++ flux which plays a role in the Ca++ response to weak agonists.229 In the presence of external Ca++, the difference between affected and control platelets using fura 2 varied according to the agonist; however, the data overall suggests that Ca++ influx is normal to slightly increased in affected platelets. The results must be viewed cautiously because of the effect of aggregation, which artificially decreases the fluorescence signal. In general, stimulation with agonists which cause aggregation of control but not affected platelets was associated with significantly increased post-stimulation [Ca++]1 in affected platelets. Stimulation with agonists which aggregated both groups of platelets did not cause any significant difference in post- stimulation [Ca++]1 between affected and control platelets. One exception was the response to ionomycin when there was a trend, though not statistically significant, for post-stimulation [Ca++]1 to be lower in affected dogs. This was the only agonist tested that caused visible aggregation only in control platelets where the post-stimulation [Ca++]i was higher in control than affected platelets. It would be expected 165 that aggregation would, if anything, cause decreased [Ca++]1. It is interesting to speculate on the different Ca++ responses of fura-2 loaded control and affected platelets to agonist stimulation in the presence of external Ca++. One explanation is the role ADP which is released from the dense granules and could positively feedback on the membrane Ca++ channel to enhance Ca++ influx. Thrombin, ADP and PAF, which cause normal to increased post-stimulation [Ca++]i in affected platelets in spite of decreased release of internal Ca++, also cause the release of dense granules.254 The calcium ionophores, which cause normal to decreased post-stimulation [Ca++]i in BHT affected platelets, do not cause significant dense granule secretion in affected platelets.254 A second explanation is the role of released TxA2, which acts as another amplification signal to enhance platelet activation. In response to all agonists which cause increased production of TxB2 in affected platelets,256 the post-stimulation [Ca++]i was normal to slightly increased. Calcium ionophore A23187 was the only agonist tested which causes decreased production of Tx82 in affected platelets when compared to control platelets. To definitely determine the effects of aggregation, ADP secretion and TxA2 production on the Ca++ response in the presence of external Ca++ would require additional studies utilizing: (1) non-aggregating preparations which still contain external Ca++ (ie. monoclonal antibodies against GPIIb—IIIazal); (2) ADP- . mobilizing enzymes; (3) and cyclooxygenase inhibitors respectively. In conclusion, although affected platelets have some identifiable abnormalities in Ca++ mobilization, the defects cannot explain the other 166 pathologic findings in BHT. In spite of decreased mobilization of internal Ca"'+ in affected platelets, the quantity released is large enough to cause shape change254 and similar phosphorylation of the 20kDa and 47kDa proteins in affected and control platelets in non-aggregating, EDTA-containing preparations.256 Shape change, secretion and aggregation have different Ca"’+ thresholds, with shape change requiring the lowest [Ca++]1 and aggregation requiring the highest [Ca'*""]i.331'335 Therefore, normal shape change and 20kDa protein phosphorylation is not contradictory to the decreased internal Ca++ mobilization response in affected platelets. There is no apparent correlation between the post-stimulation Ca++ response and the platelet responses of aggregation and secretion, though direct comparison of results is not possible due to variation in the concentration of external Ca++ in the different investigations. In the presence of external Ca++, affected platelets do not have any consistent abnormality in agonist-stimulated Ca++ mobilization; yet, affected platelets have marked defects in aggregation and secretion.254 For example, in the presence of external Ca++, PAF produced the largest Ca++ fluxes using both fura 2- and aequorin-loaded platelets. Yet, PAF does not induce aggregation in affected platelets. Ionomycin, which does not induce aggregation or secretion in affected platelets, produced Ca++ levels comparable to thrombin, which stimulates all of these responses. There is also no apparent correlation between Ca++ mobilization in the presence of external Ca++ and TxB2 production in affected platelets.256 Since PLAZ, the enzyme which regulates arachidonate release, is activated by increasing cytosOlic [Ca++]1,27'151'339'397 any alteration 167 in activity could be associated with changes in Ca++ metabolism. In spite of the lack of consistent differences in post-stimulation [Ca++]i between control and affected platelets in the presence of external Ca++, affected platelets have significantly increased Tx82 production following stimulation with high concentrations of PAF, ADP and thrombin and significantly decreased Tx32 following stimulation with low concentrations of PAP and calcium ionophore.256 CHAPTER 3: RESTING AND POST-STIMULATION PLATELET CYTOSOLIC pH INTRODUCTION In order to function and survive, cells must maintain cytosolic pH (pHi) within the physiologic range. Although partially dependent on maintenance of a physiologic external pH, cells must also have an internal mechanism(s) for regulating pHi irrespective of external pH. First, cells generate H+ during metabolic processes (ie. ATP hydrolysis and glycolysis) which must be removed. Secondly, the typical cell has a large interior-negative membrane potential which promotes the passive cytosolic influx of H+ and efflux of OH' and/or HCO3'. If these two processes are not reversed, they lead to toxic cytosolic acidification. To maintain a physiologic pHi, therefore, cells must have a mechanism(s) for actively removing H+ or accumulating HCO3'.146'245 A variety of cells respond to stimulation by increasing pHi above resting levels. In platelets, cytosolic alkalinization occurs following stimulation with thrombin,“'98'99'145'”7391340214741“75 PAF,“S phorbol esters,287’47S synthetic diacylglycerols,387 the calcium ionophores A23187 or ionomycin,287'475 arachidonic acid,1“5'287 TxA2 mimetic U466l9394 and ADP.“5 As with maintenance of resting pHi, the production of post-stimulation alkalinization requires a mechanism(s) for actively extruding H+ and/or accumulating HCO3'. The degree of alkalinization 168 169 depends on the amount of H+ generated by energy-producing metabolic activity, the buffering capacity of the cytosol and the capacity of the platelet to extrude H+,475 Cellular mechanisms for regulation of resting and post-stimulation pHi include Na+/H+ exchange, Na+-independent HCO3’/Cl‘ exchange and Na+- dependent HCO3'/Cl’ exchange. Relatively little has been published in platelets on the role of the latter two exchange mechanisms. One study has demonstrated that a Na+-independent HCO3'/Cl' exchange is active in thrombin-stimulated platelets maintained in a Na+-deficient buffer; however, when external Na+ is present, the alkalinization due to anion exchange is minimal relative to the alkalinization resulting from Na+/H+ exchange.294 A different study has shown that anion exchange is not involved in the regulation of pHi in resting or stimulated platelets. It was suggested that anion exchange plays a role in the restoration of resting pHi following recovery from cytosolic alkalinization.77 The extrusion of H+ in platelets is primarily mediated by Na+/H+ exchange.77 The exchanger does not directly require energy; instead, it is coupled to the influx of Na+ into the cell along its concentration gradient. The inwardly directed electrochemical Na+ gradient is maintained by a separate plasma membrane Na+-K+-ATPase which exchanges intracellular Na+ for extracellular K+ in an energy-requiring manner.246 Inhibition of the ATPase by ouabain causes an increase in the cytosolic concentration of Na+ and a secondary inhibition of Na+/H+ exchange.237 The NaI/H+ exchanger has a 1:1 stoichiometry.246 It is specific for the cations Na+ and Li+ and does not transport significant amounts of K+, 170 Rb+, Cs+, choline ions or N-methyl-D-glucamine.237’386 Amiloride and its derivatives inhibit the exchanger in a competitive manner with Na+, suggesting that the binding of amiloride and Na+ are mutually exclusive.237 The Na+/H+ exchanger or antiport is relatively inactive at physiologic pHi, although the addition of amiloride to resting platelets causes a slow decrease in pHi. This suggests a role for Na+/H+ exchange in the extrusion of H+ ions which are continuously generated during basal platelet metabolic processes.475 When platelets are stimulated, they expend a large amount of energy.169 The energy-generating reactions release H+ and cause rapid cytosolic acidification. In thrombin-stimulated platelets, this cytosolic acidification is concentration-dependent and independent of Na+/H+ exchange or Ca++ mobilization, suggesting that the accumulation of cytosolic H+ is one of the earliest, measurable responses to thrombin activation.99 In thrombin-stimulated platelets, cytosolic acidification is followed immediately by sustained alkalinization which is primarily mediated by Na+/H+ exchange.99'475 The activity of the exchanger is greatly enhanced by cytosolic acidification,237 suggesting that H+ has a role in the regulation of the exchanger. It has been proposed that the exchanger has two internal binding sites for H+, one that regulates antiport activity and one that exchanges internal H+ for external Na+. The HI-dependent activation of Na+/H+ exchange provides a mechanism for the rapid extrusion of cytosolic H+, protecting against deleterious accumulations of H+. The exchange has an apparent setpoint or 171 threshold. When the cytosolic concentration of H+ decreases below a certain level, the rate of Na+/H+ exchange becomes very slow. This protects the platelet against the generation of potentially harmful intracellular alkalosis.2“6'386 The acidification of the cytosol generated during agonist- stimulation of platelets is not large enough to explain the subsequent alkalinization phase where the pHi increases above resting pHi. There must be additional mechanisms by which agonist-stimulation leads to activation of Na+/H+ exchange. Some studies indicate that activation of PKC is involved in the agonist-mediated stimulation of Na+/H+ exchange.387'391'475 PKC, or one of its phosphorylated products, alters the regulatory site on the exchanger, shifting the setpoint or threshold so that it is activated by lower cytosolic concentrations of H+. Agonist-stimulated increases in [Ca++]i may also play a role in activation of NaI/H+ exchange. Simultaneous measurement of pHi and [Ca++]1 has shown that agonist-stimulated Ca++ mobilization precedes measurable cytosolic alkalinization.99'35813981474 In addition, stimulation of platelets with calcium ionophore A23187, which causes the release of internal stores of Ca++, induces a Na+ influx consistent with activation of Na+/H+ exchange.390 It has been proposed that increased [Ca++]i does not directly regulate the exchanger. Instead, Ca++ most likely plays a role in the activation of the Ca++-dependent PKC.2l‘6'386 Finally, the Na+/H+ antiport is regulated by mechanisms other than H+, PKC and Ca++. Stimulation of platelets with weak agonists, such as epinephrine or ADP, is associated with activation of Na+/H+ exchange 172 independent of PKC activation. It has been proposed that binding of weak agonists to surface receptors activates the antiport in a fibrinogen-binding-dependent manner, producing a local region of alkalinization. The increase in pHi, in combination with a local increase in [Ca++]1, stimulates PLA2 which mediates the subsequent release of arachidonic acid products. These arachidonate metabolites amplify the platelet response by activating PLC, resulting in the production of the second messengers (l,4,5)IP3 and 1,2-DG.20:409.411.412 Although most studies agree that Na+/H+ exchange and cytosolic alkalinization accompany platelet activation, there is considerable disagreement as to whether they are required for platelet activation. Artificial induction of cytosolic alkalinization by monovalent cation ionophores or by the addition of NHgCl does not, by itself, induce platelet activation as indicated by measurement of Ca++ mobilization, PA production, aggregation or protein phosphorylation.3851387'398 Artificial alkalinization does enhance platelet responses to agonist: induced activation as indicated by aggregation and/or measurement of Ca++ mobilization.384 Treatment of platelets with the cation ionophores monensin or nigericin induces shape change in a manner dependent on the external pH, suggesting that increasing pHi has a direct effect on cytoskeletal assembly.231 Numerous studies have demonstrated that activation of the Na+/H+ exchange and cytosolic alkalinization play a vital role in the platelet activation response. Most of these studies have investigated the effect of innxitng inhibition of NaI/H+ exchange on some physiologic response 173 which occurs during platelet activation. Techniques used to inhibit Na+/H+ exchange include: (1) removal of external Na+; (2) addition of amiloride or one of its derivatives; (3) elevation of extracellular H+ concentration; or (4) inhibition of Na+-KI-ATPase to remove the inwardly directed Na+ gradient. The addition of amiloride to platelet samples inhibits the rate of ADP shape change and spreading.231 Inhibition of Na+/H+ exchange blocks ADP- and epinephrine-induced activation of PLA2 and PLC.409'411'412 Inhibition of the antiport also inhibits secretion,86'413 Ca++ mobilizationl‘si145138473871395 and aggregation388'390'423 in response to a variety of agonists including PAF, ADP, thrombin, calcium ionophores, the TxAZ mimetic U46619 and vasopressin. The effect of Na+/H+ exchange and subsequent cytosolic alkalinization on Ca++ mobilization may be two-fold. First, the (l,4,5)IP3-mediated release of Ca++ from internal stores is enhanced in alkaline pH.55 Secondly, it has been proposed that an increase in pHi enhances Ca++ influx, possibly by modulating the production of inositol Phosphate metabolites.386 In contrast, other studies have shown that Na+/H+ exchange and cytosolic alkalinization are not essential events for Ca++ mobilization, shape change or aggregation in response to thrombin4'99v176v358'474 or ADP.135'398 It has been proposed that the inhibition of the platelet responses noted in the studies discussed in the previous paragraph are not due to inhibition of cytosolic alkalinization but are due to secondary effects of the experimental techniques used to block Na+/H+ exchange/"135'176 These latter studies do not dispute the proposal that pHi plays a modulatory role in platelet activation. 174 Although the role of cytosolic alkalinization in platelet activation is still uncertain, resting and post-stimulation pHi were measured in affected and control platelets using the fluorescent dye BCECF (2,7-biscarboxyethyl-5(6)-carboxyf1uoroscein) for a variety of reasons. First, at this point in the project it was determined that affected platelets have aggregation and secretion responses which show increased sensitivity to cyclooxygenase inhibition, suggesting increased dependance on the production of arachidonic acid metabolites.254 As mentioned above, stimulation of Na+/H+ exchange plays an important role in the activation of PLAZ and subsequent production of arachidonate metabolites, particularly in response to weak agonists. Any alteration in pHi could have secondary effects on PLA2 activity. Secondly, as discussed in the previous chapter, affected platelets have decreased mobilization of internal Ca"'+ in response to stimulation with a variety of agonists. Since an alkaline pHi potentiates the (1,4,5)IP3-mediated release of Ca++ from internal stores,55 a decrease in pHi could potentially inhibit the release of internal Ca++. Finally, the phosphorylation studies on affected platelets identified a 64-67kDa protein which is not phosphorylated normally.255 The role and function of this protein has not been identified. The e2-adrenergic receptor on human platelets is a 64kDa protein327 which may be closely linked to, if not actually part of, the Na+/H+ exchange mechanism.410 If the 64-67kDa protein which is abnormally phosphorylated in affected platelets is the 132-adrenergic receptor, then any alteration in its structure or function would have a major effect on pHi and subsequent platelet activation. 175 MATERIALS AND METHODS Experimental Subjects The affected group consisted of five dogs with BHT. There were four female and one male dogs. The control group consisted of four normal Basset hounds. There were one male and three female dogs. All dogs were housed in kennels in the Michigan State University Veterinary Clinical Center, fed a commercial dry dog chow (Hill's Pet Products, Topeka, KS) and exercised regularly. The dogs were free from signs of clinical disease other than those associated with BHT and were not receiving any medication during the study. Reagents BCECF-AM was purchased from Calbiochem Biochemicals (San Diego, CA). Stock solution (1mg in 227ul anhydrous DMSO) was aliquoted and stored at —20°C in a dessicator jar. Nigericin was purchased from Sigma Chemical Company (St. Louis, MO). Stock solution (5mM) was mixed in ethanol and stored at 4°C in a dessicator jar. U46619 was purchased from Sigma. Stock solution (5.2mM) was mixed in ethanol and stored at -20°C. Fibrinogen was purchased from Helena Laboratories (Beaumont, TX) and stored in the dessicated form at 4°C. Each day before use, 10mg fibrinogen was dissolved in 0.5ml HBSS. Epinephrine was obtained from Parke-Davis (Morris Plains, NJ). Stock solution (1mg/ml in sealed lml ampules) was stored at room temperature. Thrombin, PAF, ADP, digitonin 176 and PCEl were purchased, mixed and stored as described in the previous chapter. Measurement of pH1 Using BCECF pHi was measured by modification of the technique described by Zavoico et a1.“75 PRP was isolated as described in the previous chapter (pages 119-120). PGEl (luM final concentration) was added to the PRP and the sample incubated at room temperature for 5 minutes prior to centrifugation at 800 x g for 15 minutes. The supernatant was discarded and the platelet pellet resuspended in 10ml Ca++-Mg++-free Tyrode's buffer (137mM NaCl, 2.7mM KCl, 5.5mM glucose, 0.35mM Na2HP04, lOmM HEPES, pH 7.4) to which 1M citric acid was added at a ratio of 9ul acid to lml buffer. BCECF-AM (1.5uM final concentration) was added from stock solution and the suspension incubated at 37°C for 30 minutes. The suspension was centrifuged for 15 minutes at 800 x g. The supernatant was discarded and the platelet pellet resuspended in lml of Ca++-Mg++- free Tyrode's buffer containing 0.02% bovine serum albumin. The platelet suspension was passed over a 10ml Sepharose 48 column equilibrated with the same buffer. The gel-filtered platelets were manually counted using a Neubauer hemocytometer and diluted to a final concentration of 7.5 x 107 platelets/ml using the same buffer. MgC12 (0.8mM final concentration) was added and the suspension divided into 1.75ml aliquots and placed into polystyrene spectrofluorometer cuvettes covered with parafilm. The cuvettes were maintained at room temperature for the remainder of the experimental procedures. Five minutes prior to measuring sample fluorescence, a teflon-coated stir bar was added along 177 with either: (1) lmM CaClz; (2) lmM CaC12 and 0.2mg/ml fibrinogen; or (3) lmM EGTA. The cuvettes were placed into a Perkin Elmer MPF 66 spectrofluorometer. The samples were stirred and a resting (baseline) fluorescence recorded using an excitation wavelength of 495nm and an emission wavelength of 530nm. Agonist was added (70ul) and the change in fluorescence recorded over a nine minute period. The pH of the samples was determined by comparing measured fluorescence to calibration curves generated each day by modification of the technique described by Thomas et al.420 A small quantity of gel-filtered platelets were resuspended in K+—HEPES buffer (140mM KCl, 5.5mM glucose, lOmM HEPES, pH 7.4) to a final concentration of 7.5 x 107 platelets/ml. The monovalent cation ionophore nigericin (2uM final concentration) was added to allow equalization of internal and external pH. The external pH was measured using a pH probe. The sample fluorescence was measured as the external pH was varied from 6.7 to 7.7 by the stepwise addition of KOH. The suspension was lysed by the addition of 50uM digitonin and the fluorescence measured as the pH was varied from 7.7 to 6.7 by the stepwise addition of HCl. The amount of dye leakage from the cells was calculated by centrifugation of the platelet samples in Ca++-Mg++-free Tyrode's buffer at 11,600 x g for 60 seconds at set time periods during the experimental runs. The fluorescence of the supernatant (indicative of the portion of sample fluorescence due to external dye) was compared to the total fluorescence in lysed platelet suspension (indicative of fluorescence due to both internal and external dye). The leakage of dye was calculated as % leakage/hr. The presence of external dye was not corrected for in the calculation of pHi. 178 Statistical Analysis The resting and post-stimulation pHi were measured in response to each agonist on two separate occasions for each dog. The post- stimulation pHi was recorded as the maximum change in pHi from the corresponding resting value. The values for each dog were averaged to calculate a representative value which was used in the statistical comparisons. All data was analyzed using a computer software statistical/graphics program (Statgraphics, STSC, Rockville, MD). Control and affected values were compared using the student's t test for independent samples. Changes in pHi following the addition of agonist and the addition of buffer only were compared using paired t tests. RESULTS As in the previous chapter, the results will be presented according to agonist. The agonists included those used in the Ca++ studies (PAF, ADP, thrombin and ionomycin) as well as additional agonists (TxAz mimetic U46619 and epinephrine). The concentration of agonists were chosen based on aggregation studies on platelets loaded with BCECF and treated in a similar manner to that used for the pH studies. The aggregation studies showed that both control and affected BCECF-loaded platelets had a marked tendency to undergo spontaneous shape change following the initiation of stirring in the aggregometer. When two concentrations of agonist were used, the high concentration was believed to stimulate platelet activation independent of TxA2 production while the lower concentration was believed to be partially dependent on 179 TxA2 production for full activation to occur. When only one concentration of agonist was tested, it was a concentration found to give maximal aggregation responses. The concentration of agonist listed in the tables is the final concentration in solution. The buffer used to dilute the agonists was also added to stirred platelet suspensions to assess its effects on resting pHi. The values for post-stimulation pHi are listed as the change in pHi from the corresponding, stirred resting value in the same sample. The figures are representative fluorescence tracings for both control and affected platelets. In the fluorescence tracings, it is important to note that BCECF increases its fluorescence as the pHi of the solution increases. With the exception of thrombin, post-stimulation pHi was measured with and without the addition of exogenous fibrinogen to assess the role of external fibrinogen in the alkalinization process. The aggregation studies indicated that the addition of fibrinogen enhanced the rate and extent of the platelet aggregation response in control platelets. The pHi response was also measured in the presence and absence of external Ca++ to assess the role of external Ca++ in the regulation of Na+/H+ exchange. The addition of external EGTA (lmM final concentration) prevented aggregation in both groups of platelets. Resting pHi Resting pHi was measured in both stirred and non-stirred platelet suspensions. There was no difference between affected and control platelets in the resting pHi of stirred or non-stirred suspensions. In 180 control platelets, the pHi of the non-stirred suspensions was significantly increased relative to the pHi of the same suspensions following the initiation of stirring. This same trend was also found in affected platelets; however, the difference between non-stirred and stirred pHi was not statistically significant. The results are summarized in Table 15. Table 15. Resting pHi of control and affected platelets in non-stirred and stirred BCECF-loaded platelets. Suspension Control Affected m-4> m-S) Non-Stirred 7.23 i 0.03# 7.20 i 0.03 Stirred 7.18 i 0.02 7.18 i 0.02 Values - mean pHi i SD; n - number of dogs tested ; # - significantly different from stirred pHi, p < 0.05. Buffer , The addition of the buffer used to dilute the agonists was added to the platelet suspensions to serve as a negative control for comparison of post-stimulation pHi in both groups of platelets. The addition of buffer caused an immediate decrease in sample fluorescence in affected and control platelets which was interpreted from the calibration curve 181 as a decrease in sample pHi. The addition of 70ul buffer to 1.75m1 of platelet suspension caused a 3.8% dilution by volume. A comparison of the average resting fluorescence of samples pre- and post-addition of buffer showed a 4.0 i 1.6% decrease in fluorescence in the affected platelets and a 2.8 i 1.2% decrease in fluorescence in control platelets. The appearance of the fluorescent tracings were similar in both groups of platelets and were characterized by a sharp decrease in sample fluorescence which remained constant or gradually decreased further over the 9 minute time span during which sample fluorescence was recorded. See figure 11 for representative tracings. There was no significant difference between affected and control platelets in the pHi response following the addition of buffer. See Table 16 for a summary of results. Table 16. The change in pHi from resting levels in control and affected BCECF-loaded'platelets following the addition of buffer. External buffer Control Affected (n - 4) (n - 5) Ca++/ fibrinogen -0.04 i 0.02 -0.06 i 0.02 Values - mean i SD; n - number of dogs tested ; external buffer contained lmM Ca++ and 0.2 mg/ml fibrinogen. 182 -‘V' ' .1“ control ' .. «MM w .. IMW - f affected relative fluorescence units * i . i—100 secs! tune Figure 11. Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to the addition of buffer in the presence of external Ca'H’ and fibrinogen. Arrow - agonist addition. In an external buffer containing both fibrinogen and Ca'H', ADP stimulation caused significantly different pHi changes in control and affected. The control platelets had an increase in p111 following ADP stimulation. The change in pHi was significantly different than that following the addition of buffer. The degree of alkalinization appeared to parallel the occurrence of aggregation; The average change in pHi in those samples with grossly visible aggregation was 0.17 i 0.06 while the average change in pHi in those samples without visible aggregation was only 0.01 i 0.05. The affected platelets had an imediate decrease in 183 sample fluorescence, interpreted as a decrease in pHi, following agonist addition which was similar to the decrease following the addition of buffer. Though the decrease in pHi following ADP was not as marked as that following buffer, the difference between the two was not statistically significant. Grossly visible aggregation was never noted in the affected samples. See Table 17 for a summary of the results. Table 17. The change in pHi relative to resting values in control and affected BCECF-loaded platelets following the addition of lOuM ADP in the presence of external Ca++ and fibrinogen. Agonist Control Affected (n - 4) (n - 5) lOuM ADP 0.09 i 0.06 # -0.03 i 0.03 * Buffer -0.04 i 0.02 -0.06 i 0.02 Values - mean i SD; * - significantly different from control, p < 0.05; # - significantly different from buffer, p < 0.05. The appearance of the fluorescence tracings were markedly different in affected and control dogs. In control dogs, there was an immediate decrease in fluorescence following ADP addition which was followed by a gradual increase to fluorescence levels above resting. In affected dogs, there was an immediate decrease in fluorescence following agonist addition similar to the decrease noted following the addition of buffer. 184 The fluorescence remained stable or increased slightly over the time period recorded. See Figure 12 for representative tracings. ’ :control 2 I 'E 3 . O o I C 0 0 m 2 o 2 g i 35 - affected 0 h i vh~fifiTHIu+Hhfip A 1 time _ |-100 secs-I Figure 12. Representative fluorescence tracings of BCECF-loaded control and affected platelets following the addition of lOuM ADP in the presence of external Ca'"+ and fibrinogen. Arrow - agonist addition. It has been proposed that the weak agonist epinephrine requires fibrinogen binding for measurable increases in pH1.20 Since ADP is also a weak agonist, it was assumed that fibrinogen binding was required to obtain measurable alkalinization following ADP stimulation. In.platelet samples from two of the control dogs, the pHi response to ADP was measured in the presence and absence of exogenously added external fibrinogen. The average change in pH1 from resting in the presence of external fibrinogen was 0.08 while the average change in pHi in the 185 absence of external fibrinogen was -0.04. The latter value was not significantly different from the response following the addition of buffer alone. Therefore, the pHi response to ADP was not measured in any of the other dogs in platelet suspensions either lacking fibrinogen or containing EGTA which inhibits fibrinogen binding and aggregation. Platelet-Activating Factor (PAF) The pHi responses to a high concentration of PAF (luM) were measured in platelet suspensions in the presence of external Ca++, with and without added fibrinogen, and in the presence of external EGTA. When external Ca++ was present, the control platelets had much greater elevations in post-stimulation pHi than affected platelets. The increase in pHi was significantly larger in the presence of external fibrinogen, where all of the samples had grossly visible aggregation, than in the absence of added external fibrinogen, where only 1 out of 8 . samples had grossly visible aggregation. The affected platelets had significantly less alkalinization in response to PAF stimulation than control platelets. The affected platelets had similar changes in pHi irrespective of the presence or absence of fibrinogen. The pHi changes in all groups were significantly different than those obtained following the addition of buffer only. Grossly visible aggregation was not noted in any of the affected samples. When external Ca++ was removed by the addition of EGTA, there was no significant difference in pHi changes between control and affected platelets. In both groups, the post- stimulation alkalinization in the absence of external Ca++ was 186 significantly larger than the pHi measured following the addition of buffer only. The change in pHi in response to low concentrations of PAF (0.01uM) was measured in the presence of external Ca++ and fibrinogen. The affected platelets had significantly decreased post-stimulation alkalinization when compared to control platelets. The pHi changes in both groups of platelets were similar to those obtained following stimulation with high concentrations of PAF. Grossly visible aggregation was not noted in any of the platelet samples. See Table 18 of a summary of results. Table 18. The change in pHi relative to resting levels in control and affected BCECF-loaded platelets following stimulation with PAF. Agonist Conc External Control Affected Buffer (n - 4) (n - 5) Buffer Ca++/fibrinogen -0.04 i 0.2 -0.06 i 0.02 PAF luM Ca++/ fibrinogen 0.25 i 0.04 #@ 0.12 i 0.03 *# Ca++ 0.20 i 0.02 # 0.12 i 0.03 *# EGTA 0.05 t 0.02 # 0.04 i 0.02 # PAF 0.01uM Ca‘H/ fibrinogen 0.23 :i: 0.04 # 0.10 i 0.04 *# Values - mean 1 SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from buffer, p < 0.05; @ - significantly different from non-fibrinogen- containing buffer, p < 0.05. 187 The appearance of the fluorescence tracings were similar in control and affected. In Ca++-containing buffer, there was an immediate decrease in fluorescence after the addition of PAF which was followed by a sharp increase to fluorescence levels greater than resting. In control platelets, there was a tendency for the fluorescence to gradually increase over the measured time period. In affected platelets, this fluorescence remained relatively constant through the measured time period. See figure 13 for representative tracings. In EGTA-containing buffer, both control and affected platelets had an immediate decrease in sample fluorescence after the addition of PAF which was followed by an increase in fluorescence to levels greater than resting. This level of sample fluorescence stayed constant or gradually decreased over the measured time period. control . ' f . ' 9mm . affected relative fluorescence units um, i400 sscq Figure 13. Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to luM PAF in the presence of external Ca++. Arrow - agonist addition. 188 Thrombin The change in pHi in response to high concentrations of thrombin (0.6U/m1) was measured in the presence and absence of external Ca++ while the change in pHi in response to lower concentrations of thrombin (0.1U/m1) was measured only in the presence of external Ca++. Although there was a trend for the agonist-induced alkalinization to be greater in control platelets, there was no significant difference between affected and control platelets. Interestingly, the alkalinization response following stimulation with the lower concentration of thrombin was greater than that found following stimulation with the high concentration of thrombin. This difference was similar in both groups of dogs. In all cases, the change in pHi was significantly different than the change in pHi following the addition of buffer alone. See Table 19 for a summary of the results. The appearance of the fluorescence tracings was similar in both groups of platelets. In Ca++-containing buffer, the addition of agonist was accompanied by an immediate decrease in fluorescence which was followed by an increase in sample fluorescence above resting levels. See Figure 14 for representative fluorescence tracings. In EGTA- containing buffer, the addition of agonist was accompanied by an immediate decrease in fluorescence which was followed by an increase in fluorescence to levels similar to resting values. 189 Table 19. The change in pHi relative to resting levels in control and affected BCECF-loaded platelets following stimulation with thrombin. Agonist Conc. External Control Affected Buffer (n - 4) (n - 5) Buffer Ca++/fibrinogen -0.04 i 0.02 -0.06 i 0.02 Thrombin 0.6U/m1 ca” 0.07 i 0.03 # 0.04 i 0.02 # EGTA 0.02 i 0.02 # 0.01 i 0.03 # Thrombin 0.1U/ml Ga“ 0.20 i 0.03 # 0.16 '3: 0.02 # Value - mean i SD; n - number of dogs tested ; # - significantly different from addition of buffer only, p < 0.05. ‘ saw" “"“"“Wurdwwioiflh' ' ,' ' fl relative fluorescence units time [-100 secs] Figure 14. Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to 0.1U/ml thrombin in the presence of external Ca++. Arrow - agonist addition. 190 Ionomycin The pHi changes in response to ionomycin were measured in Ca++- containing buffers in the presence and absence of fibrinogen and in EGTA-containing buffer. In the presence of external Ca++, the agonist- induced alkalinization was significantly lower in affected platelets when compared to control platelets. Within each group, there was no significant difference between the pHi changes obtained in the presence or absence of fibrinogen. All of the control dog platelet samples had visible aggregation in the presence of exogenous fibrinogen and no visible aggregation in the absence of fibrinogen. Visible aggregation was not noted in any of the affected samples. In the absence of external Ca++, there was no significant difference between control and affected platelets in the pHi response to ionomycin. There was no statistical difference between the pHi change in response to ionomycin in the presence of EGTA and the pHi change in response to the addition of buffer, indicating that there was no significant pHi response to ionomycin stimulation in the absence of external Ca++. See Table 20 for a summary of the results. The appearance of the fluorescence curves in the Ca++-containing buffers differed slightly between control and affected platelets. In both groups, the addition of ionomycin was accompanied by an immediate ' decrease in fluorescence followed by a relative slow increase in fluorescence to levels greater than resting. In affected platelets, the fluorescence tended to level off and maintain a constant intensity while in control platelets, the fluorescence tended to increase over the 191 entire time period measured. See Figure 15 for representative tracings. The appearance of the fluorescence curves in EGTA-containing buffers was similar in both groups of platelets and was characterized by a sharp decrease in fluorescence which remained constant or slightly increased. Table 20. The change in pHi relative to resting levels in control and affected BCECF-loaded platelets stimulated with luM ionomycin. Agonist External Control Affected Buffer (n - 4) (n - 5) Buffer Ca++/ fibrinogen -0.04 i 0.02 -0.06 i 0.02 Ionomycin Ca++/ fibrinogen 0.40 i 0.10 # 0.17 i 0.11 *# 0a++ 0.38 i 0.06 # 0.16 i 0.14 *# EGTA -0.03 i 0.02 -0.03 i 0.03 Values - mean i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from addition of buffer, p < 0.05. 192 control I . ‘ . i '8 . _ . l KM ' ' " * ' .. 1 a affected i relative fluorescence unlts time [-100 secs- Figure 15. Representative fluorescence tracings of BCECF-loaded control and affected BCECF-loaded platelets in response to luM ionomycin in the presence of external Ca'H'. Arrow - agonist addition. U46619/Epinephrine The pH1 changes in response to luM U46619 were measured in Ca'H/fibrinogen-containing buffers in the presence and absence of luM epinephrine. Without the addition of subthreshold concentrations of epinephrine, there was no significant difference between the two groups of platelets in the pHi response following the addition of U46619. The response was similar to that recorded when buffer alone was added, suggesting no significant effect of U46619, by itself, on pHi. In the presence of external epinephrine and fibrinogen, there was a significantly decreased alkalinization response to U46619 in the 193 affected platelets relative to the alkalinization response in control platelets. When compared to pHi changes following the addition of buffer only, both control and affected platelets had a significant alkalinization response. In the control platelets, it is important to note that when gross aggregation was visible, the mean pHi change was 0.21 i 0.07. When gross aggregation was not visible, the mean pHi change was only 0.08 i 0.04. Gross aggregation was never noted in affected platelets. In the presence of external epinephrine without added fibrinogen, there was no significant difference in the pHi response in the affected platelets when compared to the control platelets. The response in affected platelets was similar irrespective of the presence or absence of external fibrinogen. In the control platelets, there was a tendency, though not statistically significant, for increased alkalinization in the presence of external fibrinogen relative to the absence of fibrinogen. In the presence of EGTA, there was no significant difference between control and affected platelets in the pHi response. The response in both groups was not significantly different than the response following the addition of buffer only. The pHi response to low concentrations of U46619 (0.05uM) was measured in the presence of external fibrinogen and epinephrine. There was a significantly decreased alkalinization response in the affected platelets when compared to the control platelets. In the control platelets, the post-stimulation pHi was significantly more alkaline than the pH1 following the addition of buffer. In the affected platelets, the post-stimulation pHi was actually lower than the resting pHi; 194 however, it was more alkaline than the pHi following the addition of buffer alone. See Table 21 for a summary of the results. Table 21. The pHi change relative to resting levels in control and affected BCECF-loaded platelets following stimulation with U46619 and epinephrine. ‘ Agonist (Conc.) External Buffer Control Affected (n - 4) (n - 5) Buffer Ca++/fibrinogen -0.04 i 0.02 -0.06 i 0.02 U46619 (luM) Ca++/fibrinogen/epi 0.13 i 0.08 # -o.01 i 0.03 *# Ca++/epi 0.01 i 0.02 -0.01 i 0.02 Ca++/fibrinogen -0.06 i 0.03 -0.07 i 0.03 EGTA -0.05 i 0.02 -0.07 i 0.03 U46619 0.5uM) Ca++/fibrinogen/epi 0.07 i 0.04 # -0.02 i 0.02 10.5uM Epi Ca++/fibrinogen -0.05 i 0.01 -0.05 i 0.02 Values - mean i SD; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from addition of buffer, p < 0.05; concentration of epinephrine in external buffer - luM. The appearance of the fluorescence tracing were different in control and affected platelets. In buffer containing both Ca++ and epinephrine, the control platelets showed a decrease in fluorescence followed by a gradual increase to levels similar to or greater than resting values. The increase in control platelets was greatly potentiated by the addition of exogenous fibrinogen. In a similar 195 buffer, the affected platelets had a rapid decrease in sample fluorescence followed by a gradual increase toward the resting levels. See Figure 16 for representative tracings. In EGTA-containing buffer or in Ca++-containing buffer without external epinephrine, there was no difference in the appearance of the fluorescence curves between affected and control platelets. In all cases, the curves were characterized by an immediate decrease in fluorescence which remained relatively constant or slowly increased toward resting values. g is 5 - control 0 o D c . O 8 I . ‘6 - *" 2 o D .2 g e i 2 p i , i affected time [-100 secs- Figure 16. Representative fluorescence tracings of BCECF-loaded control and affected platelets in response to luM U46619/luM epinephrine in the presence of external Ca'"+ and fibrinogen, Arrow - agonist addition. In the presence of external Ca++ and fibrinogen, there was no significant difference between control and affected platelets in the pHi 196 response to 10.5uM epinephrine. In both cases, there was an immediate decrease in fluorescence which was similar to that found following the addition of buffer alone, indicating no specific effect of epinephrine on pHi. See Table 21 for a summary of the results. Dye Leakage Rates There was no significant difference between affected and control platelets in the rate of leakage of BCECF from the cytosol. In the affected platelets, the leakage rate was 9.7 i 0.9% /hr. while in the control platelets the rate was 8.7 i 0.9% /hr.. DISCUSSION Stimulation of platelets is usually accompanied by cytosolic alkalinization which is primarily mediated by a Na+/H+ exchange mechanism.3861475 The importance of the role of cytosolic alkalinization in platelet activation is not clear. Early studies were hampered by a lack of readily available techniques to measure pHi, since platelets, as well as other blood cells, are too small for microinjection of indicators or impalement by pH-sensitive microelectrodes.336 The pHi is conveniently measured using fluorescent probes and spectroscopic techniques. One of the first fluorescent probes developed to measure pHi was 6-carboxyf1uorescein. 6- Carboxyfluorescein is taken up by the cell and subsequently hydrolyzed in the cytoplasm to a less permeable fluorescent form. Unfortunately, this dye has a low sensitivity at pH > 7.2 and suffers from severe 197 leakage problems.98'336 This lead to the development of a newer fluorescent derivative, 2',7'-bis(carboxyethy1)-5,6~carboxyfluorescein (BCECF). BCECF is available in a non-fluorescent ester form which is lipophilic and readily penetrates across the platelet membrane. Once in the cytosol, the ester group is removed by non-specific esterases to yield a relatively impermeable fluorescent probe. Relative to 6- carboxyfluorescein, BCECF has the advantages of pH sensitivity to 7.8 or 7.9 and decreased permeability.336 The intensity of the fluorescence of BCECF-loaded platelets is directly related to the pHi, with the intensity of fluorescence increasing as the pHi becomes more alkaline.146 BCECF is diffusely distributed throughout the cytosol and, therefore, measures average pHi. Using BCECF as the pHi indicator, resting pHi was measured in control and affected platelets. When the platelet suspensions were not stirred, the pHi of the affected platelets tended to be more acidic than the pHi of the control platelets. Following initiation of stirring, the pHi of the control platelets became significantly more acidic while the pHi of the affected platelets stayed relatively the same. There was no significant difference in the stirred, resting pHi between the two groups. Initial activation of platelets is accompanied by a transient acidification stage.99 It has been proposed that acidification results from the generation of protons due to increased metabolic activity and the production of energy.398'475 Others have proposed that the acidification results from platelet shape change which changes the light 198 transmitting properties of the platelet suspension, thereby decreasing sample fluorescence.393 The data on resting pHi suggest that the affected platelets were already in a state of activation before stirring and were not affected by stirring while the control platelets were activated in some manner by stirring. Previous studies in the laboratory have indicated that affected platelets readily undergo partial activation with handling. The aggregation studies performed on BCECF-loaded control platelets showed the platelets were susceptible to spontaneous shape change upon the initiation of stirring in the aggregometer. It is likely that the same effect occurred upon initiation of stirring in the spectrofluorometer. The buffer used to dilute agonists was added to stirred platelet suspensions to determine what effect, if any, it had on resting pHi. Unexpectedly, the addition of buffer caused an immediate and sustained decrease in measured pHi which was of similar magnitude in both affected and control platelets. This decrease may be partially due to a dilutional effect. Comparison of the percent dilution of the platelet suspension to the percent decrease in the sample fluorescence following the addition of buffer showed similar values in both groups of platelets. Since the spectrofluorometer measures the fluorescence produced by the excitation of the sample by a discrete beam of light, decreasing the concentration of dye-loaded platelets within the beam concomitantly decreases the emission fluorescence measured. Another possible explanation for the decrease in pHi is that the buffer itself was causing some degree of platelet activation; however, this is unlikely since the buffer used was a physiologic solution. Regardless 199 of the cause of the decrease in measured pHi, the magnitude of the change was similar in both control and affected platelets. Therefore, the most accurate assessment of agonist-induced alterations in pHi was to compare the change in pHi from the corresponding stirred, resting value following the addition of agonist to the change in pHi obtained following the addition of buffer without added agonist. ADP initiates platelet activation by mobilizing a small pool of arachidonic acid in a manner dependent on Na+/H+ exchange, the production of a localized zone of alkalinization and PLA2 activation. The mechanism by which ADP activates the exchanger is not known/£91411"412 Inhibition of Na+/H+ exchange prevents PLA2 activation following ADP stimulation while artificial alkalinization of the platelet cytosol reverses the inhibitory effects of the prevention of Na+/H+ exchange.411 Shape change and primary aggregation are not prevented by blockage of Na+/H+ exchange, suggesting that these responses are mediated directly by ADP binding and not by the production of cytosolic alkalinization.412 Studies have shown that ADP-induced irreversible aggregation388 and Ca++ mobilization394 are fully or partially inhibited by experimental manipulations that block Na+/H+ exchange and cytosolic alkalinization. In converse, other studies have demonstrated that cytosolic alkalinization is not essential for ADP- induced aggregation135 and Ca++ mobilization.398 These studies do not dispute that Na+/H+ exchange is important in the release of arachidonic acid and the secondary enhancement of platelet activation in response to ADP; however, it has been suggested that the concentration of amiloride 200 used in the previous studies was higher than necessary to block Na+/H+ exchange and was having secondary effects on activation.135 There were marked differences between control and affected platelets in the pHi response to addition of ADP. In the control platelets hand, addition of ADP caused immediate acidification which was followed by alkalinization to post-stimulation pHi levels higher than resting. This indicates that ADP stimulation of control platelets causes a significant pHi response. In contrast, the affected platelets had an immediate and sustained decrease in sample fluorescence which was not significantly differentfrom the response following the addition of buffer alone. This indicates that ADP stimulation of affected platelets is not associated with any significant acidification or alkalinization of the cytosol. Inhibition of fibrinogen binding in BCECF—loaded platelets blocks alkalinization following stimulation with the weak agonist epinephrine, indicating a requirement for fibrinogen binding in the production of measurable pHi changes.20'410 The lack of measurable alkalinization in control platelets stimulated with ADP in the absence of exogenously added fibrinogen suggests that fibrinogen binding is also essential for a pHi response to ADP. The data presented in the results section further indicate that the production of measurable post-stimulation changes in pH1 depends not only on fibrinogen binding, but also on the close platelet-platelet interactions that occur during aggregation or on some other post-fibrinogen binding event. This conclusion is based on the fact that, in control platelets, the degree of post-stimulation 201 alkalinization was significantly greater in those cases where aggregation was grossly visible when compared to those cases where aggregation was not visible. Since the gross identification of aggregation is a very crude technique, the lack of visible aggregates does not preclude the presence of microaggregates which could account for the small degree of alkalinization measured. In spite of normal ADP-induced fibrinogen binding,199'298 affected platelets do not support primary or secondary aggregation,25h therefore the phi measured in the current study was not affected by platelet aggregation. Comparison of the alkalinization response to ADP stimulation and some of the other measurable responses in BHT affected platelets suggests that there must be a primary defect in some post-fibrinogen binding event(s) other than Na+/H+ exchange. Affected platelets produce significantly greater quantities of Tsz in response to ADP stimulation than control platelets.256 It is impossible to make direct correlations between the pHi data and the Txfiz data since the phi measurements used different experimental conditions (ie. different concentrations of external Ca++). Based on the proposed mechanism of ADP activation presented above, comparison of the pHi and the Tx82 results suggest that the addition of ADP to affected platelets does cause activation of Na+/H+ exchange and the production of a small, localized zone of alkalinization that BCECF is not able to detect. Aggregation or some other post-fibrinogen binding event may be necessary to produce measurable alkalinization. Alternatively, it is possible that the PLA2 present in affected platelets is independent of regulation by pHi. In addition, primary aggregation induced by ADP is not affected by 202 inhibition of Na+/H+ exchange, suggesting that the binding of ADP to surface receptors, the exposure of the fibrinogen binding site and subsequent fibrinogen binding of adjacent platelets is not inhibited by pH1.412 Primary aggregation, which is absent in affected platelets,254 should not be affected by pHi changes. Finally, in spite of no measurable pHi response to ADP, affected platelets mobilize Ca++ as well as control platelets under similar experimental conditions as discussed in the previous chapter. PAF causes receptor-mediated activation of PLCIO5'221'375 and subsequent stimulation of Na+/H+ exchange by at least two separate routes. One mechanism is mediated by PKCl‘S'386 while the second route is independent of PKC.45 Blockage of Na+/H+ exchange inhibits the release of arachidonic acid metabolites following PAF stimulation, suggesting that Na+/H+ exchange is involved in the regulation of arachidonic acid release and/or metabolism.200’410'413 Inhibition of Na+/H+ exchange or artificial acidification of the platelet cytosol also partially inhibits PAF-induced dense granule secretion,410'413 aggregationl‘s'388 and Ca++ mobilization,45 suggesting a significant role for the exchanger in platelet activation. Removal of external Na+ inhibits Tx32 production and serotonin dense granule secretion following stimulation with PAP in buffers containing low (ie. non-physiologic) concentrations of Ca++ but not in buffers containing physiologic (2mM) concentrations of ca”.200 The removal of extracellular Na+ does not affect Ca++ mobilization in platelets treated with cyclooxygenase inhibitors,349 indicating that the inhibition of Ca++ mobilization by 203 blockade of Na+/H+ exchange reported above is due to an effect on arachidonic acid release and not a direct effect on Ca++ fluxes. Unlike ADP, the addition of PAF to control and affected platelets was associated with significantly increased post-stimulation pHi in the presence and absence of external Ca++. In the presence of external Ca++, the degree of cytosolic alkalinization was significantly greater in the control platelets than in the affected platelets. Within the control group, but not the affected group, the degree of alkalinization was significantly greater in those samples containing fibrinogen, where aggregation was grossly visible, than in the samples without fibrinogen, where aggregation was not always visible. This suggests that the production of measurable pHi is partially dependent on aggregation. Unlike ADP, PAF also causes activation of the exchanger independent of aggregation as indicated by the post-stimulation cytosolic alkalinization in the affected platelets in spite of an inability to undergo either primary or secondary aggregation in response to PAF.254 When external Ca++ was removed by EGTA, the alkalinization response was significantly decreased when compared to the response in the presence of external Ca++, but there was no significant difference between control and affected platelets in the post-stimulation pHi. These findings demonstrate that PAF initiates Na+/H+ exchange irrespective of external Ca++, fibrinogen binding and aggregation. Affected and control. platelets show similar phosphorylation of the 47kDa protein in the presence of external EDTA, indicating similar activation of PKC.255 Therefore, it is likely that PKC mediates the stimulation of Na+/H+ exchange when external Ca++ is removed. 204 As discussed for the ADP data, the difference between control and affected platelets in the pHi responses to PAF cannot explain the platelet abnormalities previously identified in BHT. First, it has been shown that inhibition of Na+/H+ exchange does not inhibit primary platelet aggregation in response to PAF;413 yet, affected platelets do not aggregate in response to PAF stimulation.254 In spite of an inability to aggregate, PAF-induced cytosolic alkalinization of affected platelets occurs, though the post-stimulation pHi was decreased when . compared to control platelets. Secondly, one of the primary roles proposed for Na+/H+ exchange is regulation of arachidonic acid mobilization.413 Yet, the alkalinization response was much greater in control platelets in response to high concentrations of PAF while Tx32 production was significantly less than that of control platelets. In addition, PAF stimulation of affected platelets causes an increased production of TxBZ in response to high concentrations of PAF but decreased production of Tx82 in response to low concentrations of PAF when compared to control platelets.256 In contrast, the degree of post- stimulation cytosolic alkalinization was similar following stimulation with both high and low concentrations of PAF in affected platelets. Unfortunately, it is impossible to directly correlate these results since the measurement of TxA2 production was performed in plasma containing 0.32% trisodium citrate (a low Ca++ concentration medium) while the phi changes were assessed in a buffer containing higher concentrations of Ca++ (lmM). ,In platelet suspensions maintained in the presence of physiologic concentrations of external Ca++ (2mM), PAF- induced production of arachidonate metabolites (TxAz) is significantly inhibited when compared to the same measurement in platelets maintained 205 in suspensions containing low concentrations of external Ca++. The mechanism for this Ca++-dependent inhibition of TxA2 production is not known.200 Finally, it was proposed that decreased post-stimulation pHi in affected platelets could partially explain the decrease release of intracellular Ca++ as discussed in the previous chapter; however, this was not supported by the current study. In the presence of external Ca++, there was no significant decrease in Ca++ mobilization in affected platelets relative to control platelets. It was only in the presence of external Ca++ that there was a difference in post-stimulation cytosolic alkalinization between affected and control platelets. The affected platelets had significantly less internal Ca++ released in the presence of external EGTA; yet, there was no significant difference between control and affected platelets in the pHi response to PAF in buffers containing EGTA. Similar to PAF, thrombin causes the activation of PLC, hydrolysis of membrane phosphoinositides and the production of the second messengers (1,4,5)IP3 and 1,2-DG.21'3“'90'252'346 1,2-DC activates PKC which stimulates Na'VH+ exchange, causing cytosolic alkalinization.386'391 Thrombin also stimulates Na+/H+ exchange by some undefined mechanism that is independent of PKC activation.“75 The role of Na+/H+ exchange and subsequent cytosolic alkalinization in thrombin- stimulated platelet activation is under intense debate. Na+/H+ exchange is most important in the platelet responses to low concentrations of thrombin and one study concluded that thrombin concentrations of >0.2U/ml were relatively independent of phi effects.381 Inhibition of Na+/H+ exchange by amiloride attenuates platelet aggregation in response 206 to thrombin388 while inhibition of Na+/H+ exchange by removal of external Na+ decreases aggregation and secretion in response to thrombin.86 Inhibition of Na+/H+ exchange blocks the release of arachidonic acid following stimulation with low concentrations of thrombin.“12 Blockade of Na+/H+ exchange by the removal of external Na+ or by use of amiloride derivatives also partially blocks the influx of external Ca++ and the release of Ca"'+ from internal stores.145’211'38“'387 This inhibitory action on Ca++ mobilization is bypassed by artificial alkalinization of the platelet cytosol.395 These effects on thrombin stimulation are not mediated by inhibition of PLC since manipulations which alter Na+/H+ exchange have no effect on thrombin-induced hydrolysis of membrane phosphoinositides.395v409 Other studies have demonstrated that Ca++ mobilization and platelet activation in response to thrombin occur independent of cytosolic alkalinization or activation of Na+/H+ exchange. These studies have shown that: (1) the increase in [Ca++]i precedes cytosolic ’ alkalinization;99'398 (2) artificial acidification of the platelet cytosol and inactivation of Na+/H+ exchange have minimal effects on the Ca++ mobilization in response to high or low concentrations of thrombin;474 and (3) removal of external Na+ does not affect Ca++ mobilization in aspirin-treated platelets in response to thrombin stimulation.349 A recent study has demonstrated that removal of external Na+ enhances thrombin-induced serotonin secretion due to inhibition of cytosolic alkalinization and other undefined effects of Na+ itself.211 Inhibition of PKC activity blocks thrombin-induced cytoplasmic a1kalinization.without affecting Ca++ mobilization,358 207 suggesting that Ca++ mobilization occurs independently of pHi effects. It has been proposed that amiloride derivatives inhibit thrombin-induced platelet aggregation and Ca++ mobilization at levels lOO-fold greater than those required to inhibit Na+/H+ exchange,176 indicating that the inhibitory effects demonstrated in the previous studies are secondary to other effects of amiloride and not to inhibition of Na+/H+ exchange. Thrombin was the only agonist tested which showed a similar degree of cytosolic alkalinization in control and affected platelets in both the presence and absence of external Ca++. Thrombin was also the only agonist tested which causes aggregation of platelets from both groups of dogs.254 In spite of similar pHi responses, previous studies have demonstrated that affected platelets produce significantly increased quantities of Tsz.256 This suggests that, though Na+/H+ exchange and pHi may modulate PLA2 activity,27'412 they are not the only factors regulating its activity. The smaller increases in pHi measured in the presence of external EGTA in both control and affected platelets indicate a requirement for external Ca++, fibrinogen binding and aggregation for full activation of Na+/H+ exchange following stimulation with thrombin. In the previous chapter, it was shown that thrombin-stimulation of affected platelets in EGTA-containing buffers is associated with decreased release of internal stores of Ca++_relative to control platelets. Since there was no difference in cytosolic alkalinization in response to thrombin, alterations in put responses cannot explain the defective Ca++ mobilization. It should be noted that 0.25U/ml thrombin was used in the 208 Ca++ studies while 0.6U/ml was used in the pHi studies. Both of these are considered concentrations at which the effects of thrombin have been shown to be independent of pH1.387 Ionomycin is a calcium ionophore that activates platelets by increasing membrane permeability to Ca++ 121’313'377'457 and releasing internal stores of Ca++ in a non-receptor-mediated manner.11'313'377 Relatively little has been published regarding the effects of calcium ionophores on Na+/H+ exchange and cytosolic alkalinization. Blockade of Na+/H+ exchange by amiloride derivatives inhibits aggregation in response to low concentrations of A23187390 while platelets activated with calcium ionophore A23187 extrude H+ in a manner compatible with Na+/H+ exchange.391 Whereas perturbation of Na+/H+ exchange inhibits PLA2 activation in response to ADP and epinephrine, it does not have any significant effect on PLA2 activation by A23187.411 Studies using fluorescent dyes to measure phi have demonstrated that the addition of ionomycin causes an initial acidification of the platelet suspension followed by minimal to no cytosolic alkalinization.358v398v47S Contrary to previously published results, the current phi studies on control and affected platelets have demonstrated significant alkalinization in response to ionomycin. This difference may be due to a variety of reasons. First, the concentration of ionomycin used in the current study was approximately lO-fold higher than that used in the published studies. The concentration of agonist chosen was based on aggregation studies in BCECF-loaded control canine platelets showing that luM ionomycin was required to cause strong aggregation responses 209 while 0.luM ionomycin (the concentration used in the published studies) did not reliably induce aggregation. It is important to note that there was a narrow range between ionomycin concentrations that cause no aggregation, aggregation and visible cell lysis. Since concurrent enzyme leakage studies were not performed, it cannot be ruled out that some of the alkalinization measured in the platelet samples represents the leakage of dye from injured or damaged platelets. Since the external pH was more alkaline than the pHi, the fluorescence of external dye would artificially elevate the measured signal. If this were the case, it would be anticipated that levels would be similar in both control and affected platelets. Another explanation for the difference between the current study and the published results was that the published studies used cyclooxygenase inhibitors to eliminate the amplification effects of TxA2 production while the current study did not. In a Ca++-containing buffer, the addition of ionomycin caused significantly less cytosolic alkalinization in affected platelets when compared to control platelets. The absence or presence of exogenously added external fibrinogen had minimal effect on the post-stimulation pHi in either group. As with the other agonists tested, visible aggregation occurred in response to ionomycin in control platelets while affected platelets have been previously demonstrated to not aggregate in response to the calcium ionophore A23187.254 Therefore, part, if not all, of the decrease in cytosolic alkalinization in affected platelets may be due to the absence of aggregation or some other post-fibrinogen binding event. 210 Although there were significant differences between affected and control platelets in the degree of cytosolic alkalinization in response to ionomycin stimulation, the results do not correlate with the other abnormalities detected in BHT. Previous studies have demonstrated that affected platelets release significantly less TxB2 in response to calcium ionophore A23187 than control platelets.256 Decreased post- stimulation pHi could, theoretically, be associated with decreased PLA2 activity and decreased release of arachidonic acid since PLA2 activity is enhanced in alkaline pH;27 however, studies have shown that inhibition of Na+/H+ exchange has no effect on calcium ionophore A23187- induced activation of PLAZ.411 In addition, in the presence of external Ca++, the level of ionomycin-induced cytosolic alkalinization in affected platelets was as large as the alkalinization following stimulation with PAF or thrombin which cause increased production of TxB2 in affected platelets.255 This indicates that alkalinization alone does not mediate the release of arachidonate metabolites. The lack of post-stimulation pHi response in the presence of external EGTA in both affected and control platelets was not expected and indicates that measurable alterations in pHi following stimulation with ionomycin are dependent on the presence of external Ca++. External Ca++ could be required for: (l) fibrinogen binding and aggregation; or (2) potentiation of the Ca++ flux into the platelet cytosol. Since neither control nor affected platelets showed significant alkalinization in response to ionomycin in the absence of external Ca++, the decreased release of internal stores of Ca++ in response to ionomycin in affected platelets does not appear to be secondary to a defect in Na+/H+ 211 exchange. This was expected since ionomycin causes non-receptor- mediated Ca++ release11'313'377 and increased pH has been shown to potentiate (l,h,5)IP3-mediated release of internal Ca'H'.55 U46619 is a synthetic PGHz/TxAz mimetic which causes platelet activation by receptor-mediated stimulation of PLC.337'374 In dogs, full platelet activation in response to U46619 requires the addition of subthreshold doses of epinephrine.66 This finding was supported by the present study in which the addition of U46619 alone caused no significant change in pHi in either control or affected platelets when compared to the addition of buffer only.. This was also supported by the results of the aggregation studies used to establish concentrations of agonists for the pHi measurements. In all control dogs tested, the addition of U46619 by itself caused only shape change while the prior addition of a subthreshold concentration of epinephrine (luM) was required to initiate aggregation or secretion. U46619 stimulates Na+/H+ exchange primarily by the PLC-mediated hydrolysis of membrane PI and the production of 1,24DG.394 U46619- induced hydrolysis of membrane phosphoinositides is not inhibited by in xigxg manipulations that block Na+/H+ exchange, suggesting that U46619- induced activation of PLC is independent of Na+/H+ exchange and subsequent alkalinization.409 U46619-induced aggregation and secretion is also not inhibited by removal of external Na+, indicating that these responses occur independent of Na+/H+ exchange.86 Recent studies have demonstrated that blockage of Na+/H+ exchange partially inhibits 046619- induced Ca++ mobilization. Ca++ mobilization is restored by artificial 212 alkalinization of the platelet cytosol, suggesting a dependence on increased pHi at the level of (1,4,S)IP3-mediated release of Ca++ from internal stores.394 The significance of the U46619-stimulated pHi changes to the defect in BHT is not clear since U46619 has not been used in any of the previous studies. It was used in the current study due to the suggestions that: (1) affected platelets are more susceptible to cyclooxygenase inhibition254; and (2) the pathway leading to the generation of TxA2 is not under proper regulatory control in affected platelets.256 In the presence of external Ca"'+ and epinephrine, affected platelets had weak to no alkalinization responses to U46619. These responses were markedly depressed compared to those in control platelets. It is likely that part, if not all, of the difference was due to the absence of aggregation in affected platelets. This conclusion is supported by the pHi response in control platelets. In the presence of external fibrinogen, the control platelets had greater post-stimulation pHi relative to the situation in the absence of fibrinogen where the post-stimulation pHi was similar to that in affected platelets. The aggregation trials performed in control platelets showed that external fibrinogen was not needed to maintain measurable aggregation; however, the rate of aggregation was slower and the extent of aggregation less in the absence of external fibrinogen. Incubation with EGTA inhibited any significant pHi response in either control or affected platelets stimulated by U46619/epinephrine. These findings were not expected since a recent published study has 213 shown that BCECF-loaded platelets in the presence of external EGTA have an initial decrease in sample fluorescence after addition of the agonist, due to initial shape change, followed by a return to initial resting levels.394 Possible explanations for the lack of measurable pHi responses following the removal of external Ca++ include requirements for: (1) the Ca++-dependent membrane GPIIb-IIIa complex, fibrinogen binding and aggregation; (2) the influx of external Ca++ to potentiate and sustain the intracellular Ca++ release involved in the regulation of Ca++-dependent processes; or (3) Ca++ to mediate U46619 receptor binding. Platelets have been demonstrated to be less responsive to TxA2 or TxA2 mimetics in low Ca++ media.200 Epinephrine is proposed to have a mechanism of action similar to ADP whereby the binding of epinephrine to its receptor initiates local alkalinization by activation of Na+/H+ exchange. This causes the subsequent release of a localized, very small quantity of arachidonic acid.409'411'412 This response is dependent on the exposure of the fibrinogen binding site and subsequent fibrinogenbinding.20 The released arachidonic acid metabolites then amplify the original stimulatory response by activating PLC. In both control and affected platelets, epinephrine did not cause any significant pHi response when compared to the addition of buffer only. Even though measurable cytosolic alkalinization was not identified in response to epinephrine using BCECF, it cannot be ruled out that both control and affected platelets do produce local, membrane- associated activation of Na+/H+ exchange. The lack of measurable 214 alkalinization does not indicate that epinephrine does not play a potentiating role. Examination of U46619 data shows that the prior addition of subthreshold doses of epinephrine was needed to cause full platelet activation. Other studies have shown that inhibition of cytosolic alkalinization by prevention of Na+/H+ exchange does not affect the epinephrine potentiation of thrombin-induced PLC 402 activation, suggesting that the mechanism of epinephrine potentiation of platelet activation is independent of pH effects. In conclusion, the pHi studies yielded significant differences between control and affected platelets as summarized in Table 22. The findings suggest that these alterations are secondary to failure of aggregation and do not explain the primary platelet defect in BHT based on current knowledge of platelet function. Since many events in platelet activation are occurring within a very short time period, it is often difficult to determine which events are primary and which are secondary. In the current study, it is difficult to separate those events which result from activation of Na+/H+ exchange and those events that initiate Na+/H+ exchange. There is no apparent correlation between pHi and the platelet responses of dense granule ATP secretion, Ca++ mobilization and aggregation in affected platelets. ATP secretion occurs in affected platelets in response to ADP and high concentrations of PAF;254 yet, post-stimulation pHi was significantly decreased when compared to control platelets. ATP secretion is absent in affected platelets following stimulation with low concentrations of PAF; yet, in affected platelets there was no significant difference between the p81 response to high or low concentrations of PAF. Calcium ionophores do Table 22. 215 and affected BCECF-loaded platelets. Summary of the change in pHi from resting levels in control Agonist (Conc.) External Buffer Control Affected (n - 4) (n - 5) Buffer Ca++/fibrinogen -0.04 $0 02 (-) -0.06 $0.02 (-) ADP (100M) Ca++/fibrinogen 0.09 $0.06 #(+/-) -0.03 $0.03 * (-) 0.17 $0.06 #(+) 0.01 $0.01 #(-) PAF (luM) Ca++/fibrinogen 0.25 $0.04 #(+) 0.12 $0.03 *#(-) 0a++ 0.20 $0.02 #(+/-) 0.12 $0.03 *#(-) ECTA‘ 0.05 $0.02 #(-) 0.05 $0.02 # (-) PAF (0.01uM) Ca++/fibrinogen 0.23 $0.04 #(+) 0.10 $0.04 *#(-) Thrombin 0a++ 0.07 $0.03 #(-) 0.04 $0.02 # (-) (0.6U/ml) EGTA 0.02 $0.03 #(-) 0.01 $0.03 # (-) Thrombin Ca++ 0.20 $0.03 #(-) 0.15 $0.02 # (-) (0.1U/ml) Ionomycin (luM) Ca++/fibrinogen 0.40 $0.10 #(+/-) 0.17 $0.11 *#(-) 0a++ 0.38 $0.06 #(-) 0.16 $0.14 *#( ) EGTA -0.03 $0.02 (-) -0.03 $0.03 (-) U46619 (luM) Ca++/fibrinogen/ 0.13 $0.08 #(+/-) -0.01 $0.03 *#(-) - luM epinephrine 0.21 $0.07 #(+) 0.08 $0.04 #(-) Ca++/epinephrine 0.01 $0.02 (-) -0.01 $0.02 (-) EGTA -0.05 $0 02 (-) -0.07 $0.03 (-) Epinephrine Ca++/fibrinogen -0.05 $0.01 (-) -0.06 $0.02 (-) (10.5uM) Values - mean change in pHi $ S.D.; n - number of dogs tested; * - significantly different from control, p < 0.05; # - significantly different from addition of buffer, p < 0.05; (+) - visible aggregation occurs; (+/-) - visible aggregation may or may not occur; (-) - visible aggregation does not occur. 216 not cause secretion in affected platelets.254 While the calcium ionophore ionomycin-induced post-stimulation pHi was decreased in affected platelets relative to control platelets, the absolute level was similar to that obtained following stimulation with other agonists which do induce secretion in affected platelets. In terms of Ca++ mobilization, affected platelets had a decreased phi response in the presence of external Ca++; yet, measurement of post-stimulation [Ca++]i in similar experimental conditions shows normal to increased values. In the presence of external EGTA, affected platelets have normal to decreased post-stimulation [Ca++]i relative to control platelets; yet, there was no significant difference in post—stimulation pHi in similar experimental conditions. Enhancement of platelet activation by Na+/H+ exchange and subsequent cytosolic alkalinization is most important in response to weak agonists.386 Activation of Na+/H+ exchange appears to be vital in the activation of PLA2, release of arachidonate metabolites and subsequent activation of PLC in response to weak agonists such as ADP, epinephrine and low concentrations of thrombin.410 In the current study, activation of affected platelets with ADP was associated with a significantly decreased alkalinization response when compared to control platelets. In spite of this finding, it has been shown that affected platelets produce significantly greater quantities of TxB2 than control platelets.256 This suggests that weak agonists do initiate local activation of Na+/H+ exchange which is not detected using BCECF. Alternately, it is possible that affected platelets contain a PLA2 system that is activated independent of changes in phi. 217 The cause of the decreased alkalinization response in affected platelets was not determined in the current study. Since it was seen in response to all agonists but thrombin, the defect is not likely secondary to abnormal agonist binding or receptor coupling to effector systems. While ADP and epinephrine cause direct receptor-mediated activation of Na+/H+ exchange,409v411'412 other agonists activate Na+/H+ exchange in a manner primarily mediated by PLC hydrolysis of phosphoinositides, DG production and PKC activation. Therefore, it cannot be ruled out that there is decreased activation of PKC or decreased PKC activation of Na+/H+ exchange in affected platelets. This theory is not substantiated by similar patterns of phosphorylation of the 47kDa protein in response to agonist stimulationin control and affected platelets.255 Alternately, the decreased post-stimulation alkalinization response in the affected platelets may be due to the lack of the potentiating effects of aggregation. This theory is supported by a number of findings in the current study. First, there were significant differences in the post-stimulation pHi of control samples in the presence and absence of added fibrinogen which appeared to correlate to the presence or absence of visible aggregation. Secondly, thrombin, the only agonist which induced similar pHi responses in control and affected platelets, was also the only agonist tested that causes aggregation of both groups of platelets.254 Finally, there was no significant difference between affected and control platelets in the pHi response to any of the agonists tested in platelet suspensions incubated with 218 external EGTA, a situation which inhibits aggregation and fibrinogen binding in both groups of platelets. CHAPTER 4: 1,2—DIACYLGLYCEROL PRODUCTION INTRODUCTION Platelet activation is associated with the hydrolysis of membrane phosphoinositides due to either: (1) direct stimulation of PLC by an activated receptor-agonist complex; or (2) stimulation of Ca++-dependent PLA2, release of arachidonic acid metabolites and secondary activation of PLC by PGHz/TxA2.337 Initially, PLC preferentially hydrolyses PIP2 to release (1,4,5)IP3 and 1,2-DC.17'25'21‘8’323’343 1P3 production is detectable within seconds of agonist stimulation and the quantity produced is sufficient to initiate internal Ca++ release.340 The elevation in [Ca++]1 stimulates further activation of PLC and the hydrolysis of PI in addition to PIP and PIPZ, leading to the release of the corresponding inositol phosphate and additional quantities of 1,2- DG.435'463 The majority of the membrane phosphoinositides are present as PI rather than PIP or PIP2,337 therefore the bulk of 1,2-DC production comes from the hydrolysis of PI.342 Diacylglycerol is generated within 5 seconds of the addition of agonist and production generally peaks by 30 seconds,318'342 at which time the Concentration of 1,2-DC is elevated 2-3 fold over resting levels.318 Studies in other cells have shown that approximately one- half of the post-stimulation 1,2-DC comes from the breakdown of non- 219 220 phosphoinositide lipids.323 In platelets, potential sources for the formation of 1,2-DC include triglyceride, PA, monoglyceride, PI and phospholipids other than PI including phosphatidylcholine, phosphatidylethanolamine and phosphatidylserine. It has been shown that the level of [3H]arachidonic acid in platelet triglyceride, monoglyceride and phospholipids other than PI does not change following agonist stimulation during the period in which 1,2-DC production is peaking. In platelets, the levels of [3H]arachidonic acid in PA increase during the same time period,342 indicating that the primary source for the activation-associated production of 1,2—DC is PI. It has recently been demonstrated that radiolabelled phosphatidylcholine is not metabolized during agonist-induced platelet activation, supporting the primary role of PI in the production of l,2-DG.39 The concentration of 1,2-DC in resting platelets is actively maintained at low levels318'342 by two different biochemical pathways: (1) phosphorylation to PA by diacylglycerol kinase; and (2) hydrolysis by diglyceride lipase.33 Diacylglycerol kinase catalyses the reaction: 1 ,2-DG + ATP - PA 4» ADP.“22 Studies using diacylglycerol kinase inhibitors or measurement of the breakdown of exogenous diacylglycerols Suggest that the diacylglycerol kinase pathway is the primary route by which platelets metabolize 1,2-00.41.422 The PA formed is then recycled back into the phosphoinositide pathway. PA also plays a role in Platelet activation as a second messenger which amplifies the 8timulatory signal.21l‘ Diglyceride lipase catalyzes the deacylation of the sn-1 fatty acid from 1,2-DC and the subsequent action of monoglyceride lipase catalyzes the hydrolysis of the sn—2 fatty acid, 221 releasing the glycerol backbone which is recycled back into the membrane lipids.76 The rate-limiting enzyme in this pathway is diglyceride lipase. Since 71% of the total platelet phosphoinositol is present as the l-stearoyl, 2-arachidonyl form,244 the primary form of 1,2-DC is also the l-stearoyl, 2-arachidonyl form. Both lipases show some specificity for arachidonyl-containing substrates, therefore this 'pathway plays a role in the release of arachidonic acid.76 A relatively minor pathway for the removal of 1,2-DC is acylation to triacylglycerol . “1 1,2-DC is a neutral lipid that remains within the membrane. It activates PKC377 by increasing the affinity of PKC for ca” and permitting enzyme activation at [Ca'H'h present in the resting p1atelet.193'414 This reaction is dependent on the presence of membrane phosphatidylserine, which binds Ca'H' and provides a surface for PKC binding,138 and the presence of diacylglycerols of the sn-1,2 <:<)nfiguration.224'377 The effects of endogenously produced 1,2-DC can be mimicked in vitro by the use of synthetic diacylglycerols or the tumor promoting phorbol esters which intercalate into the plasma membrane and bind to PKC to induce activation.72’194v224 Based on current knowledge of platelet function, no single defect can explain the platelet abnormalities identified in BHT; however, it was determined that measurement of PLC activity in intact platelets was essential since the enzyme plays a central role in platelet activation. Agonists which stimulate platelet activation independent of PLC alcltzivation and the associated hydrolysis of membrane phosphoinositides 222 (such as phorbol esters337 or high concentrations of thrombin57’226) cause aggregation and secretion of both control and affected platelets. In contrast, those agonists that rely on either direct or secondary activation of PLC for full activation (such as PAF, ADP and calcium ionophores) do not induce aggregation of affected platelets.254 PLC activity can be assessed in vigzg by the measurement of the production of the metabolites of phosphoinositide hydrolysis such as the inositol ‘phosphates, 1,2-DC or PA. Due to the compatibility of experimental techniques with equipment available in the laboratory, 1,2-DC production *was measured following stimulation with the same range of agonists used in previous studies. It was anticipated that there would be decreased 'production of 1,2-DC in response to those agonists dependent on PLC activation for the full range of platelet activation responses. IhKTERIALS AND METHODS lfixperimental Subjects The affected group consisted of three dogs with BHT. There were cine male and two female dogs. The control group consisted of three riormal Basset hounds. There were one male and two female dogs. All (logs were housed in kennels at the Michigan State University Veterinary (Ilinical Center, fed a commercial dry dog chow (Hill's Pet Products, Tepeka, KS) and exercised regularly. The dogs were free from signs of clinical disease other than those associated with BHT and were not receiving any medication during the study. 223 Reagents [3H]Arachidonic acid was purchased from Amersham Corporation (Arlington thts, IL, 250uCi/250u1) and from Dupont/ NEN Products (Boston, MA, 250uCi/2.5ml). The labeled arachidonic acid was stored in its original container at -20°C. Silica 60A (250uM) glass-backed thin layer chromatography (TLC) plates with 19 channels and preabsorbent strips were purchased from Whatman LabSales (Hillsboro, OR). Lipid standards were purchased from Sigma Chemical Company (St. Louis, MO) and included the following: (1) diolein containing 85% 1,3-diolein and 15% 1,2-diolein, (2) 1,3-diolein, (3) phosphatidylcholine, (4) free fatty acid (oleate), (5) monoglyceride, (6) cholesterol oleate, (7) triolein and (8) cholesterol. Each standard was dissolved in lml chloroform and stored in a dessicator jar at -20°C. Scintillation fluid was purchased from Beckman (Arlington thts, IL) and from Research Products International Corp.(Mount Prospect, IL). Organic solvents (chloroform, Inethanol, benzene, diethyl ether and NHAOH) were of HPLC grade and were [aurchased from Aldrich Chemical Company (Milwaukee, WI). Thrombin, PAF, AXDP, calcium ionophore A23187, U46619 and epinephrine were purchased and sstored as discussed in the previous chapters (pages 119-120, 175). JPreparation of [3H]Arachidonic Acid Labelled Platelets 1,2-DC was measured by modification of a technique described by Brass et al.57 Platelets were collected and PRP isolated as described 3111 the previous chapter (page 119-120). PGEl (luM final concentration) ‘Viis added to the PRP and the suspension incubated for 5 minutes at room 224 temperature prior to centrifugation at 800 x g for 15 minutes. The supernatant was discarded and the pellet resuspended in 1.5m1 autologous PPP acidified by the addition of 1M citric acid to yield a final pH 6.5 as measured using a pH meter. A total of 6uCi [3H]arachidonic acid was added and the suspension incubated at 37°C for 1 hour. The suspension ”was passed over a 15ml column of Sepharose 4B equilibrated with Ca++- MgH-free Tyrode's buffer (137mM NaCl, 2.7mM x01, 5.5mM glucose, 0.35m NaZHPOA, lOmM HEPES, pH 7.4) to remove the excess radiolabel. The platelets were manually counted using the Unopette system and a Neubauer ‘hemocytometer. The eluate was diluted to a final concentration of 3 x 108 platelets/ml using the same buffer containing lmM EDTA. .Aliquots (0.5ml) were placed in polypropylene centrifuge tubes, covered land.maintained at room temperature for 15 minutes. Agonist (20u1) was .added and the samples briefly swirled to gently mix the suspension. The :reaction was stopped at 5, 15, 30, 60 and 120 seconds post-stimulation lay the addition of 1.8m1 ice cold chloroform/ methanol solution (1:2 wz/v) which was freshly mixed each day. The buffer used to dilute the algonists was added alone to additional samples and the procedure J:epeated to establish baseline samples for comparison of results iiollowing the addition of agonists. Resting (0 seconds) 1,2-DC levels ‘V'ere established by adding stopper solution to an aliquot of platelet suspension which was not agitated and which did not have anything added tao it. Following the addition of the stopper buffer, the suspensions were vigorously vortexed and maintained at room temperature for 30 allidmutes. The lipids were extracted by the addition of 0.9ml 2M KCl and 0 ~ 9ml chloroform to each aliquot. The suspensions were thoroughly "f post—stimulation CPM : time-matched CPM following the addition of 1>uffer only for the agonists U46619/epinephrine and calcium ionophore 1&23187 since it was observed that the 1,2-DC produced in response to t:hese agonists was relatively small and that the increase above resting ].evels alone was not representative. Each sample contained 0.5ml of the I>latelet suspension with a total of 1.5 x 108 platelets. The mg I>rotein/sample was calculated each day to assure that similar quantities Of platelets were present in the samples. 228 Lipid TLC Results The platelet lipid samples were compared to known lipid standards. The region containing the 1,2-DC radioactivity was determined by co- migration with 1,2-diolein standards. See Figure 17 for a representative TLC plate. The Rf values were calculated over the course of the experiments and the mean values $ SD are as follows: (1) 1,2- diolein - 0.72 $ 0.06; (2) 1,3-diolein - 0.83 $ 0.04; (3) triolein - 0.92 $ 0.03; (4) monoglyceride - 0.08 $ 0.02; (5) phosphatidylcholine - 0.51 $ 0.04; (6) cholesterol oleate - 0.90 $ 0.05; and (7) cholesterol - 0.45 $ 0.06. It was found during the course of the experiments that the Rf values and resolution of the lipid bands were adversely affected by increasing ambient humidity and the subsequent accumulation of moisture *within the plates. Care had to be taken to assure the TLC plates were .adequately dried and maintained in a low humidity environment. liesting Values There was no significant difference between affected and control 1>latelets in the resting 1,2-DC CPM in extracted lipid samples which had 1>een separated on TLC plates. The total radioactivity of all of the I>latelet lipids recovered following the extraction procedure was also nueasured each day on a representative sample prior to lipid separation <311'TLC plates. There was no significant difference between the two groups. The mg protein/sample was measured each day and affected and Control platelets contained similar amounts of protein. The ratio of the resting 1,2-DC CPM/sample : mg protein/sample and the ratio of the 229 sis. iii omflamaWN" O ‘ (I I n. r‘ .,.- O .fi 1 Figure 17. Representative TLC plate. Lane 1 - oleate; lane 2- monoglyceride; lane 3 - cholesterol; lane 4 - 1,3-diolein; lane 5 - 85% 1,3-diolein, 15% 1,2-diolein; lane 6 - platelet sample + diolein mix; lane 7 - platelet sample without added standard; lane 8 - triolein; lane 9 - cholesterol oleate; lane 10 - phosphatidylcholine. resting 1,2-DC CPM/sample : CPM for total extracted lipids/sample were also calculated. Though there was a tendency for the ratios to be lower in affected platelets, the differences between the two groups of platelets were not statistically significant. See Table 23 for a summary of the results. 230 Table 23. Comparison of values from the resting platelet samples of control and affected 3H-AA--labelled platelets. Value Affected Control (rt-3) (rt-3) 1,2-DC (CPM) 144.5 $ 17.4 129.0 $ 37.2 total extracted lipids (CPM) 72738.8 $ 12882.5 77651.2 $ 8867.5 mg protein/sample 0.41 $ 0.04 0.46 $ 0.05 1,2-DC CPM/mg protein 322.8 $ 35.1 299.2 $ 97.8 1,2-DC CPM/CPM of total 0.00206 $ 0.00029 0.00177 $ 0.00030 extracted lipids Values - mean $ SD; n - number of dogs tested. Buffer The addition of buffer without added agonist to the platelet samples of both affected and control dogs caused a decrease in sample CPM which was measurable 5 seconds after the addition of buffer and remained at a relatively constant level throughout all of the time periods tested. The CPM of the post-stimulation samples was not statistically different from resting values and was of similar magnitude in both groups. See Table 24 and Figure 18 for summaries of the results. Table 24. 231 buffer alone to control and affected 3H-AA-labelled platelets. Measurement of baseline 1,2-DC following the addition of Post-stimulation time Control Affected (II-3) (n-3) 0 Seconds (Resting) 155.8 $ 27.9 152.2 $ 39.9 5 Seconds 123.0 $ 16.0 109.2 $ 27.1 15 Seconds 119.5 $ 11.9 109.7 $ 30.0 30 Seconds 128.1 $ 18.9 112.5 $ 27.7 60 Seconds 121.1 $ 17.0 106.4 $ 24.1 120 Seconds 124.6 $ 5.2 110.6 $ 26.3 Values - mean CPM $ SD; n - number of dogs tested. ms - ' - b I g ”H— as AFFECTED _ ~' . -*- ennui I P C) E; lfllr - P : a J- b C s! 1 . b d 2 115:- - ~ .......... O I... . a - . ‘ h d E «r _ P. . ' : 73 - - P1 _ . 1 k! . a ‘ L L ‘ A ‘ l _ ‘ l a an 40 n . m 120 TIME (seams) Figure 18. addition of buffer in control and affected Values represent mean $ SD. samples. Comparison of the baseline 1,2-DC (CPM) following the 3H-AA-la‘belled platelet 232 Adenosine Diphosphate (ADP) The ADP results are presented as the ratio of post-stimulation 1,2- DG CPM : resting 1,2-DC CPM. In both affected and control platelets, the addition of lOuM ADP caused a decrease in sample 1,2-DC CPM which was reflected as a decreased ratio. The decrease in CPM tended to be greater in the affected platelets; however, the difference between control and affected platelets was not significant. Resting and post- stimulation values were compared in affected and control platelets. The difference was statistically significant only in the affected platelets. See Table 25 and Figure 19 for summaries of the results. Table 25. Comparison of 1,2—DC production in control and affected 3H- AA-labelled platelets following the addition of lOuM ADP. Ratio of post-stimulation CPM : resting CPM. Post-stimulation Time Control Affected - (n - 3) (n - 3) 5 Seconds 0.80 $ 0.10 0.73 $ 0.08 @ 15 Seconds 0.83 $ 0.14 0.71 $ 0.11 @ 30 Seconds 0.82 $ 0.09 0.75 $ 0.09 @ 60 Seconds 0.81 $ 0.10 0.74 $ 0.10 @ 120 Seconds 0.79 $ 0.04 0.76 $ 0.07 @ Value - mean $ SD; n - number of dogs tested; @ - significantly different from resting value, p < 0.05. 233 I'loET v r u I If r g t 7 T T f'TW I I rfff I r: if .mmmnn ] IMOO _ —-— cum - a a I: 090 J «c ’ F' . 5 r I 8 0.30E [J g £ ..4‘ ........ i eeeeeeeeeeeeeeeeeee : 0.70 ' .. OoSOEL L 1 LA 1 l I .L 1 L L a L L L L- a an 40 an II m 120 Figure 19. Comparison of 1,2-DC production in response to lOuM ADP in control and affected 3H-AA-labelled platelet samples. Ratio of post- stimulation CPM : resting CPM. Values represent mean $ SD. Comparison of ADP-induced 1,2-DC production (as indicated by the ratio of post-stimulation CPM : resting CPM following the addition of lOuM ADP) to the buffer induced 1,2-DC production (as indicated by the ratio of post-stimulation CPM : resting CPM) yielded no significant differences in either control or affected platelets. See Figures 20 and ' 21 for a summary of the results. 234 I f' V I ' V r ' fr! '* ' I V ' r' ' T I 1.05: - b a I a! m I 0.95- - { _._ m " 8 r . :2 * 3 ‘< 0.85;- 1 E , a 0.75- . .............. eeeee eeeeee - D d L . p 0.63- .1 0.55:. .‘ l J 1 e 1 LL . l l . e l L; . 1+. LLJ A L 1 9 an m m m 120 an TIME (SECONDS) Figure 20. 1,2-DC production in response to lOuM ADP and buffer in 3H- AA-labelled affected platelet. Values represent mean $ SD. 1.05 "f" V I w w v I v ffir 1' v r r I f. w I v v v [30m (RATIO) r'I'UIUIIUIVUUfr'i'Ul'r'Ul 14.11..IllLllllllllll|llllI AAL‘ALAJ‘AlleJLLLAlAAA TIME (SEEMS) Figure 21. 1,2-DC production in response to lOuM ADP and buffer in 3H- AA-labelled control platelets. Values represent mean :9: SD. 235 Platelet-Activating Factor (PAF) The PAF results are presented as the ratio of post-stimulation 1,2- DC CPM : resting 1,2-DC CPM. In response to stimulation with high concentrations of PAF (luM), there was a rapid peak of 1,2-DC production at 5 seconds which was of similar magnitude in both control and affected platelets. By 15 seconds 1,2-DC production was decreasing and by 30 seconds values were similar to resting levels in both groups. At 60 seconds, the ratio was significantly decreased in affected platelets when compared to control samples. Though the difference was not significant, the same trend was seen at 120 seconds. See Table 26 and Figure 22 for a summary of the results. Table 26. Comparison of 1,2-DC production in control and affected 3H- AA-labelled platelets following stimulation with luM PAF. Ratio of post-stimulation CPM: resting CPM. Post-Stimulation Time Control Affected (n-3) (fl-3) 5 Seconds 2.30 $ 0.13 @ 2.49 $ 0.74 @ 15 Seconds 1.67 $ 0.20 @ 1.64 $ 0.37 30 Seconds 1.19 $ 0.24 1.12 $ 0.12 60 Seconds 1.08 $ 0.09 0.92 $ 0.06 * 120 Seconds 1.10 $ 0.15 0.94 $ 0.08 Value - mean $ SD; n - number of dogs tested; * - significantly different from control, p < 0.05; @ - significantly different from resting, p < 0.05. 236 _ ; 1 : d 1 _ f . 3.25 2J5 2.43 2.03 [3HJDG (RATIO) ‘I'lIII'I'Il'W'II'IVUI' llllllll|lll|lllllllllJ Figure 22. Comparison of 1,2-DC production in response to luM PAF in 3H-AA-labelled control and affected platelets. Values represent mean $ SD. * - significantly different from control, p < 0.05. Following stimulation with low concentrations of PAF (0.01uM), the response at 5 and 15 seconds was similar in control and affected platelets, although the peak post-stimulation ratio was not as large as that found following stimulation with luM PAF. By 30 and 60 seconds, the amount of 1,2-DC was similar to resting values and the ratio was significantly decreased in the affected platelets relative to the control samples. See Table 27 and Figure 23 for summaries of the results . 237 Table 27. Comparison of 1,2-DC production in control and affected 3H- AA-labelled platelets following stimulation with 0.01uM PAF. Ratio of post-stimulation CPM: resting CPM. Post-Stimulation Time Control Affected (n - 3) (n ' 3) 5 Seconds 1.54 $ 0.06 @ 1.61 $ 0.28 @ 15 Seconds 1.27 $ 0.11 @ 1.19 $ 0.13 30 Seconds 1.18 $ 0.11 0.92 $ 0.07 * 60 Seconds 1.02 $ 0.02 0.85 $ 0.09 * 120 Seconds 0.92 $ 0.10 0.84 $ 0.11 Values - mean $ SD; n - number of dogs tested; * - significantly different from control, p < 0.05; @ - significantly different from resting value, p < 0.05. llwvltwl‘l‘wrr‘rfr'ww‘rl *mm. [SHJOG (RATIO) 9 TWI'I'I'le'V'I'V'I'I'l lLll'jllllllllllllllllll| Figure 23. Comparison of 1,2-DC production in response to 0.01uM PAF in H-AA-labelled control and affected platelets. Values represent mean $ SD. * - significantly different from control, p < 0.05. 238 THROHBIN The thrombin results are presented as the ratio of post-stimulation 1,2-DC CPM : resting 1,2-DC CPM. Following the addition of 0.3 U/ml thrombin, there was a relatively slow increase in 1,2-DC production which reached its peak by 30 seconds in both control and affected platelets. At 60 and 120 seconds the 1,2-DC production was gradually decreasing toward resting levels in both groups, though there was a tendency for the decrease to be greater in affected platelets. The ratio was significantly decreased in affected platelets relative to control platelets at 120 seconds. See Table 28 and Figure 24 for summaries of the results. Table 28. Comparison of 1,2-DC production in control and affected 3H- AA-labelled platelets following stimulation with 0.3U/ml thrombin. Ratio of stimulated CPM: resting CPM. Post-Stimulation Time Control Affected ‘ (n - 3) (n - 3) 5 Seconds 1.01 $ 0.09 0.92 $ 0.08 15 Seconds 1.31 $ 0.06 @ 1.21 $ 0.15 30 Seconds 1.48 $ 0.17 @ 1.52 $ 0.44 60 Seconds 1.46 $ 0.10 @ ' 1.50 $ 0.03 @ 120 Seconds 1.30 $ 0.06 @ 1.15 $ 0.06 * Values - mean $ SD; n - number of dogs tested; * - significantly different from control, p < 0.05; @ - significantly different from resting, p < 0.05. 239 1 1 q i a 1 1 J J 1 2.00 1.80 {as 33 r'T'VVrIVII'IU'IUjV'VVII O O lllLlllllllllllllJlJllJl 1.40 1.20 [SHIOO (RATIO) 1.00 0.80 TINE (SECONDS) Figure 24. Comparison of 1,2-DC production in response to 0.3U/ml thrombin in 3H-AA-labelled control and affected platelets. Values represent mean i SD. * - significantly different from control, p < 0.05. Calcium Ionophore A23187 The A23187 results are presented as the ratio of 1,2-DC CPM following stimulation with luM A23187 : 1,2-DC CPM following the baseline addition of buffer at matched time periods. Increased production of 1,2-DC was detectable by 5 seconds in both control and affected samples. The increase peaked by 30 to 60 seconds and was gradually decreasing by 120 seconds. There was no significant difference between control and affected platelets at any of the measured time periods. See Table 29 and Figure 24 for summaries of the results. 240 Table 29. Comparison of 1,2-DC production in control and affected 38- AA-labelled platelets following stimulation with luM calcium ionophore A23187. Ratio of 1,2-DC CPM following agonist stimulation : 1,2-DC CPM following addition of buffer only. Post-Stimulation Time Control Affected (n - 3) (n - 3) 5 Seconds 1.08 $ 0.05 1.11 $ 0.04 15 Seconds 1.14 $ 0.06 1.12 $ 0.03 30 Seconds 1.22 $ 0.06 @ 1.15 $ 0.04 60 Seconds 1.22 $ 0.02 @ 1.22 $ 0.07 @ 120 Seconds 1.15 $ 0.09 1.12 $ 0.10 Value - mean $ SD; n - number of dogs tested; @ - significantly different from baseline, p < 0.05. V V V I V f V I r ‘I' V l V V V ' V 1 U ' V I I l 1.30 ' 1.25 1.20 1.15 1.10 13mm (RATIO) 1.05 'jj'Ij'T'T'r'jir'rt'lnij'jit' —. as - llllllllllllllLllllllllLlllllJl 1.00 b A A L l A A L L L A L ' L l I I L A. A . A I .1 TIME (SECOI‘IJS) Figure 25. Comparison of 1,2-DC production in response to luH.A23187 in 3H-arachidonic acid labelled control and affected platelets. Values represent mean.$ SD. 241 U46619/Epinephrine The U46619/epinephrine results are presented as the ratio of 1,2-DC CPM following the addition of 2.5uM U46619/luM epinephrine : baseline 1,2-DC CPM following the addition of buffer at matched time periods. The amount of 1,2-DC in the control platelets peaked at approximately 15 to 30 seconds post-stimulation after which it remained relatively constant over the time period measured. There was an initial sharp increase in 1,2-DC production in affected platelets that was significantly greater than the increase in control platelets. At 15 seconds, the amount of 1,2-DC had decreased, but was still greater than that of the control samples, though the difference was not significant. The amount of 1,2-DC stayed relatively constant at times 30 and 60 seconds. 1,2-DC production peaked again at 120 seconds in the affected platelets to a level that was significantly elevated over values present in control platelets. See Table 30 and Figure 26 for summaries of the results. 242 Table 30. Comparison of 1,2-DC production in control and affected 3H- AA-labelled platelets following stimulation with 2.5uM U46619/luM epinephrine. Ratio 1,2-DC CPM following agonist-stimulation: baseline 1,2-DC CPM following addition of buffer. Post-Stimulation Time Control Affected (n - 3) (n ' 3) 5 Seconds 1.18 $ 0.16 1.49 $ 0.05 *@ 15 Seconds 1.20 $ 0.07 1.34 $ 0.06 @ 30 Seconds 1.23 $ 0.21 1.28 $ 0.14 @ 60 Seconds 1.22 $ 0.06 1.28 $ 0.10 @ 120 Seconds 1.21 $ 0.06 1.37 $ 0.06 *@ Value - mean $ SD; n - number of dogs tested; * - significantly different from control, p < 0.05; @ - significantly different from baseline, p < 0.05. Phorbol Myristate Acetate (PMA) The PMA results are presented as the ratio of 1,2-DC CPM following the addition of 3uM PMA : resting 1,2-DC CPM. In both control and _ affected platelets, there was a rapid decrease in 1,2-DC levels which was of similar magnitude. The 1,2-DC values stayed relatively constant throughout the test period. See Table 31 and Figure 27 for summaries of the results. The values in both control and affected platelets were not statistically different than the 1,2-DC response following the addition of buffer without agonist. See Figures 28 and 29 for summaries of the results. 243 .1 r '— r—f r—r 1 T I—v T'T r V 1' fir rii 7 ., . . . 'l' H H u m 0 O | . WT!" rrrrrrrrrr‘ln 111' ' I'T'T'I'T— *CONTRIL * H J.‘ ' O [3H] 06 (RATIO) .t-l H '3 3 ___1___ lllJl.lllLlllllllllllllllfllll s 3 § ii TIRE (SECONOS) Figure 26. Comparison of 1,2-DC production in response to U46619/epinephrine in 3H-AA-labelled control and affected platelets. Values - mean $ SD. * - significantly different from control, p < 0.05. Table 31. Comparison of 1,2-DC production in control and affected 3H- AA-labelled platelets following the addition of 3uM PMA. Ratio of post- stimulation 1,2-DC CPM : resting 1,2-DC CPM. ' Post-Stimulation Time Control Affected (n - 3) (n - 3) 5 Seconds 0.73 $ 0.17 0.72 $ 0.10 15 Seconds 0.72 $ 0.09 0.79 $ 0.14 30.Seconds 0.76 $ 0.14 0.75 $ 0.13 60 Seconds 0.74 $ 0.11 0.74 $ 0.12 120 Seconds 0.72 $ 0.09 ‘0.80 $ 0.12 Value - mean $ SD; n - number of dogs tested. 244 I’ I r j r r v T 1 1 U f T V ' I —r ' j T ' '— I- 1.06 :- . : x m 1 0.96 - "' L -- cram. 4 A - i O " ‘ '- r .- O.35 r- “ s . : 8 0 76 P ....... . ..... l H . .- e ' . . + E. : 1 ; l P .2 0.56 ‘ 1 ‘ L 1 . . L A e e l U m 12:: U n a ”nut (seems Figure 27. Comparison of 1,2-DC production in response to 3uM PMA in I-I-AA-labelled control and affected platelets. Values - mean $ SD. f t 1 1 I ff 1 ‘ rfi I 1 ' V V I ' ' T I ' j r I - 1 0.96. I. an m l I ..... m I 8 - : F! I E 0.86 - '1 I . . . e 9 ‘ 3 0.75 - °' ‘ - ....... . . , ...... '1 3 . « H + S 0.66 r- ‘: r j ‘ p- 0.55 - A _ t L L A . L ‘ A L L _ - + 1 . _ _ J . . A l— O a 49 m I” II II TIRE ($812305) Figure 28. Comparison of 1,2-DC production in response to 3uM PMA and buffer in affected 3II--AA-labelled platelets. Values - mean $ SD. 245 1.06 0.96 0.86 0.76 [34106 (RATIO) 0.66 [TVUV'TIIIIIUUW"T'V‘l'fi'j'r \ ‘. llleIJllllLLlJllJll'llL 0.56 L a. J. l A A + 1 a A a L A A ALL A kl LJ_ A 0 20 40 an In In! 120 TIME (SEW Figure 29. Comparison of 1,2-DC production in response to 3uM PMA and buffer in 3H-AA-la'belled control platelets. Values - mean i SD. DISCUSSION Activation of phosphoinositide-specific PLC is a central effector pathway in platelet signal transduction.377 Agonists activate PLC either directly by a receptor-mediated process or indirectly via the production of arachidonic acid metabolites which secondarily initiate receptor-mediated activation of PLC.377 PLC stimulates the hydrolysis of membrane phosphoinositides to produce the intracellular second messengers (1,4,5)IP3 and 1,2-DC. Therefore, the production of either of these products can be used to indirectly monitor PLC activity in intact platelets. 246 1,2-DC production was measured in control and affected platelets. Non-aggregating platelet preparations (ie. non-stirred, no exogenous fibrinogen, lmM EDTA) were used to allow direct comparisons with the protein phosphorylation data previously reported255 and to eliminate alterations in 1,2-DC production which occur secondary to the amplification processes resulting from aggregation itself. There was no significant difference between control and affected samples in the mg protein per sample, indicating that the samples being compared contained similar concentrations of platelets. Resting 1,2-DC levels were measured and reported as CPM per sample. There was no significant difference between affected and control platelets in resting 1,2-DC concentrations using this parameter. There was marked day to day variation in the CPM values in both control and affected platelets. The range of values for each dog also changed on separate days. This was reflected in marked day to day variations in the radioactivity for the total quantity of extracted lipids/platelet sample. These findings suggest that there were unavoidable differences in loading of [3H]arachidonic acid into platelet samples which occurred irrespective of uniform experimental loading techniques. Therefore, a better assessment of resting 1,2-DC was the comparison of the ratio of the radioactivity of 1,2-DC recovered from the TLC plates to the radioactivity of the total amount of lipids recovered from the extracted sample. The latter value was calculated in a representative platelet sample prior to separation of the lipids on TLC plates. Since there is only approximately 33% recovery of lipids from the TLC plates (personal communication M. Laposata), the ratio calculated cannot be used to 247 estimate the absolute percentage of total lipids in the resting platelet present as 1,2-DC. The ratio can be used to estimate differences between control and affected platelets in the relative amount of the total resting platelet lipids which is present as 1,2-DC. The data showed a trend for the ratio to be decreased in affected platelets, although the difference between affected and control was not statistically significant. This suggests that affected platelets have a tendency for a decreased percentage of the total lipids to be present as 1,2-DC in resting platelets. It may reflect decreased basal formation of 1,2-DC or increased rate of removal of 1,2-DC. The quantity of 1,2-DC in resting platelets is maintained at relatively low levelsng'342 by two separate enzyme systems that actively remove 1,2-DC. The addition of exogenous diacylglycerols is rapidly followed by conversion to PA by diacylglycerol kinase or to glycerol by the sequential action of diglyceride and monoglyceride lipases. Studies have shown that 80% of exogenous [3H1-labelled diacylglycerols are metabolized within 2.5 minutes, with the majority being removed by diglyceride kinase.41 Inhibition of diglyceride kinase is associated with prolongation of the 1,2-DC signal while inhibition of diglyceride lipase, the rate limiting enzyme in the conversion of 1,2-DC to glycerol,76 is not associated with prolongation of the 1,2-DC signal. Instead, when diglyceride lipase is inhibited, a greater percentage of the exogenously added 1,2-DC is metabolized by diglyceride kinase.“1 248 The amount of 1,2-DC in platelet samples was measured following the addition of 20ul buffer without agonist in order to act as a time- matched baseline for comparison to the amount of 1,2-DC produced following the addition of agonist in the same group of platelet samples. Unexpectedly, there was a decrease in 1,2-DC following the addition of buffer which was evident by five seconds and persisted throughout all time points tested. It seemed unlikely that the response was due to the addition of buffer itself since the buffer was a physiologic solution. It was felt that a more likely source for the decrease was the gentle swirling applied briefly to the platelet samples to mix the suspensions after the addition of buffer or agonist. This hypothesis was examined in a few experiments where platelets samples were swirled but nothing was added. It was found that the shaking alone did cause a decrease in measurable 1,2-DC; however, the experiments were not repeated sufficient times to do statistical comparisons with the change in 1,2-DC following the addition of buffer. The mechanism for the decrease in 1,2-DC is not known and has not, to my knowledge, been previously reported. Possible explanations include decreased basal formation of 1,2-DC or increased activity of diglyceride kinase and/or diglyceride lipase. As discussed above, the post-stimulation production of 1,2-DC is primarily regulated by PLC- mediated hydrolysis of membrane PI.39'342' PLC is a Ca++-dependent enzyme whose activity is modulated by a G protein.53'377 In platelets, a number of forms of PLC have been identified which differ in molecular weight, substrate specificity, pH optima, Ca++ requirements and immunoreactivity.53 Agents that inhibit PLC activity include of a group 249 of positively charged substances which interact with the negatively charged phosphate group on the phosphoinositides. These agents include neomycin, polyamines, phenothiazines and local anesthetics.377 cAMP inhibits an agonist-induced increase in 1,2-DC but does not affect basal 1,2—DC levels.318 Therefore the increased basal cAMP present in BHT affected plateletsag'ag cannot adequately explain the decrease in 1,2-DC following the addition of buffer. Relatively little has been published on diglyceride kinase activity in platelets; however, it is strongly dependent on membrane phospholipids for normal activity318 and can be inhibited by a number of synthetic inhibitors.41 Diglyceride lipase activity is optima at acidic pH, is not affected by Ca++ and/or Mg++ concentrations and is sensitive to sulfhydryl inhibitors.76 In the current study, the 1,2-DC response to buffer was not felt to be significant since the response was similar in control and affected platelets. In hindsight, it would have been desirable to compare all of the post-stimulation 1,2-DC CPM values to the baseline, time-matched values following the addition of buffer. Unfortunately, time-matched baseline values were only available for A23187 and U46619/epinephrine. The contribution of the different pathways of 1,2-DC removal to the decrease in 1,2-DC could be determined by measuring the concurrent production of the metabolites produced. If the decrease is due to increased activity of diglyceride kinase, a concurrent elevation in [3H]PA would be expected. If the decrease is due to increased activity of diglyceride lipase, a concurrent elevation in [3H]glycerol or 2- [3H]oleoylmonoacylglycerol would be expected. Finally, if the decrease 250 is due to decreased activity of PLC, none of these metabolites would be increased. ADP is a weak agonist which requires both cyclooxygenase activity and primary aggregation to induce secretion and irreversible aggregation.377 Previous studies have shown that activation of PLC following stimulation with weak agonists, such as ADP, is dependent on the production of arachidonic acid metabolites.409 ADP stimulation does not cause any measurable phosphorylation of the 20kDa and 47kDa proteins in platelet suspensions when granule secretion and the production of arachidonic acid metabolites is prevented by non-aggregating experimental conditions,127v161 suggesting that ADP does not cause activation of PLC under non-aggregating conditions. In the presence of EDTA to inhibit aggregation, control and affected platelets do not have significantly increased phosphorylation of the 20kDa and 47kDa proteins following ADP stimulation.255 These findings suggest that there is no significant Ca++ mobilization and PKC activation respectively, indirectly indicating that PLC activation does not occur. The current study used a non-aggregating/non-fibrinogen binding preparation (ie non-stirred, no exogenous fibrinogen, lmM EDTA). It was anticipated that ADP stimulation would not cause any measurable increase in 1,2—DC production relative to resting levels. Unexpectedly, it was found that the addition of ADP caused a decrease in sample 1,2-DC which was of similar magnitude in both affected and control platelets. Comparison of the 1,2-DC responses following the addition of ADP and following the addition of buffer alone showed no significant difference 251 in either group. These findings indicate that the decrease in 1,2-DC in response to ADP is not due to a specific effect of ADP but results from the same factors that cause a decrease in measurable 1,2-DC following the addition of buffer alone. PAF is a physiologic agonist that induces receptor-mediated activation of PLC, hydrolysis of membrane phosphoinositides and the production of the second messengers (1,4,5)IP3 and 1,2- D<3.]'O6'221'375'383 The PAF-induced activation of PLC is similar both in the presence of external Ca++/Mg++ and in the presence of external EDTA.383 PAF-induced activation of PLC occurs independent of the release of arachidonic acid metabolites.383 In the presence of external EDTA, the phosphorylation pattern of the 20kDa and 47kDa proteins following addition of PAF is similar in control and affected platelets. Phosphorylation of the 20kDa protein peaks within 30 seconds of agonist addition while phosphorylation of the 47kDa protein peaks within 60 seconds.255 These findings suggest that Ca++ mobilization and PKC activation is similar in control and affected platelets. In the current study, production of 1,2-DC in response to PAF stimulation was very rapid in both control and affected platelets. Production peaked by 5 seconds and gradually decreased over the 2 minute test period. The maximum level of 1,2-DC production was similar in both groups of platelets; however, the amount of 1,2-DC was significantly decreased in affected platelets relative to control platelets by 30 (luM PAP) and 60 seconds (luM and 0.01uM PAF) after agonist addition. 252 Decreased concentrations of 1,2-DC may have resulted from either decreased production by PLC or increased rate of removal of 1,2-DC. Since PAF stimulation of platelets is associated with an agonist-induced uncoupling of the membrane receptor from the PLC effector system,272 it is possible that affected platelets have an enhanced rate of receptor uncoupling. Alternatively, affected platelets may have an increased rate of 1,2-DC metabolism due to enhanced activity of either diglyceride kinase or diglyceride lipase. Thrombin is a physiologic agonist that causes receptor-mediated activation of PLC.21'3“'90'262'3a6 The thrombin-induced activation of PLC and subsequent production of 1,2-DC occurs in either the presence or absence of external Ca++/Mg++ and is independent of the release of arachidonic acid metabolites.383 Previous studies on the thrombin- induced phosphorylation of the 20kDa and 47kDa proteins have shown similar phosphorylation patterns in control and affected platelets, although peak phosphorylation of the 47kDa protein in affected platelets is not as extensive as in control platelets.255 Measurement of 1,2-DC production in control and affected platelets showed no significant differences over the first 60 seconds following the addition of thrombin, though there was much greater variation in the values from the affected platelets. At 120 seconds post-stimulation, the 1,2-DC level was significantly decreased in the affected platelets relative to control platelets. As discussed with PAP, this may result from enhanced uncoupling of the PLC effector system from the membrane receptor or from increased removal of 1,2-DC by diglyceride kinase or diglyceride lipase. 253 The calcium ionophores A23187 and ionomycin are non-physiologic agonists that cause activation of PLC as indicated by the loss of membrane phosphoinositides and the concurrent formation of 1,2-DC and PA.31*'31‘“339 Inhibition of cyclooxygenase and removal of ADP completely block these responses, indicating that the activation of PLC is secondary to Ca++-induced activation of PLA2 and the release of ADP.337'339 Examination of control and affected platelets have yielded similar phosphorylation patterns for both the 20kDa and 47kDa proteins in the presence of external EDTA,255 indirectly indicating similar PLC activity. Phosphorylation of both of these proteins can occur independent of PLC activation due to Ca++-mediated activation of PKC and MLCK.225'361 Since affected platelets have decreased production of Tx82256 and no significant ATP dense granule secretion254 following stimulation with A23187, it was anticipated that there would be decreased activation of PLC in affected platelets. However, measurement of 1,2-DC production showed a similar response in both control and affected platelets. The degree of activation of PLC, as indicated by relative 1,2-DC production, was much less following stimulation with A23187 than following stimulation with PAP or thrombin. U46619 is a PGHZ/TxAz mimetic that causes direct receptor-mediated activation of PLC which is independent of arachidonic acid metabolism and the presence of external Ca++.337'379 In the present study, subthreshold concentrations of epinephrine were added prior to U46619 because canine platelets do not aggregate in response to U46619 by itself.66 Epinephrine does not directly activate PLC; instead, 254 activation of PLC is dependent on the release of arachidonic acid metabolites and ADP secretion.383'409 When compared to baseline 1,2-DC following the addition of buffer, control platelets had relatively constant 1,2-DC concentrations over the entire time period tested following the addition of U46619/epinephrine. In contrast, affected platelets had significantly increased production of 1,2-DC following stimulation with U46619/epinephrine. By five seconds post-stimulation, affected platelets had a peak of 1,2-DC production not found in control platelets. 1,2-DC levels then decreased until 120 seconds post-stimulation at which time there was another, although smaller, peak of production. These findings suggest that affected platelets have increased sensitivity to U46619 and/or epinephrine which is specific for one, or both, of these agonists. Increased sensitivity to either agonist may result from (1) increased receptor number or affinity; or (2) decreased receptor downregulation and desensitization in affected platelets. Pretreatment of platelets with U46619 causes decreased responsiveness to addition of subsequent doses of U46619,233 indicating that receptor downregulation and desensitization normally occur. Phorbol esters (ie PMA) are non-physiologic agonists that intercalate into the membrane and directly activate PKC.71'72'377 The addition of phorbol esters to platelets has no measurable effect on the levels of inositol phosphateshhz or the production of 1,2-DC,39 indicating no direct effect on PLC activity and phosphoinositide metabolism. PKC activation causes decreased production of PA, 1,2-DC 255 and/or inositol phosphates in response to stimulation with ADP, PAF, U46069, thrombin or collagen.39'380'421'442 This indicates that PKC inhibits agonist-induced PLC actiVation and the hydrolysis of the membrane phosphoinositides. One mechanism proposed for this inhibition is a negative effect of PKC on the G protein coupled to PLC.5 In the presence of external EDTA, affected and control platelets have similar phosphorylation of the 20kDa and 47kDa proteins following stimulation with PMA. There is rapid phosphorylation of the 47kDa protein and a slower phosphorylation of the 20kDa protein.255 Unlike the other agonists discussed above, phosphorylation of the 20kDa and 47kDa proteins in response to PMA does not necessarily result from PLC activation. The 47kDa protein is a major substrate for phosphorylation by PKC.362 PKC also phosphorylates the 20kDa protein at a distinct site from that phosphorylated by MLCK.275 There was no significant difference between control and affected platelets in 1,2-DC production following stimulation with PMA. In both groups, the addition of PMA was followed by a decrease in sample CPM which stayed relatively constant over the time periods measured. This response was similar to that found following the addition of buffer without agonist, indicating that it was not due to a specific effect of PMA. These findings suggest that the phosphorylation of the 20kDa and 47kDa proteins demonstrated in the previous study is due to PKC, not PLC, activation. 256 In conclusion, affected and control platelets produced similar quantities of 1,2-DC in response to all agonists tested with the exception of the PGHz/TxAZ mimetic U46619 and/or epinephrine. The combination of U46619/epinephrine caused affected platelets to produce significantly greater quantities of 1,2-DC. It was not possible to make direct comparisons between the current investigation and previous investigations on BHT because of the different experimental conditions; however, it appears that the platelet defects previously identified do not result from lack of PLC activation. A direct comparison between the production of 1,2-DC and the phosphorylation of the 47kDa and 20kDa proteins255 was possible since the studies had been performed under similar experimental conditions. Both investigations, which indirectly assess PLC activity in intact platelets, have demonstrated similar findings. The affected platelets have similar patterns of agonist-induced phosphorylation of the 20kDa and 47kDa proteins; however, there was a general tendency for affected platelets to phosphorylate the proteins less extensively or the dephosphorylate the proteins more rapidly.255 This finding correlates well with the tendency for the production of 1,2-DC to decrease more rapidly in affected platelets when compared to control platelets. This decrease in 1,2-DC could result from either: (1) increased uncoupling of the agonist receptors or the modulatory G protein from PLC; or (2) enhanced metabolism of 1,2-DC by diglyceride kinase and/or diglyceride lipase. Unfortunately, the phosphorylation of platelet proteins in response to U46619/epinephrine has not been done in affected or control platelets to offer a comparison. Based on the measurement of 1,2-DC 257 production, it would be anticipated that affected platelets would show enhanced phosphorylation of the 47kDa and 20kDa proteins. Increased production of 1,2-DC in response to U46619/epinephrine in affected platelets suggests increased sensitivity of PLC to stimulation with these specific agonists. As mentioned earlier, increased platelet sensitivity to U46619 could result from: (1) increased number or affinity of PGHZ/TxAz receptors; or (2) decreased receptor downregulation or desensitization. This finding is significant in view of the increased production of Tx82 in affected platelets following stimulation with thrombin, ADP and high concentrations of PAP and decreased production of Tx82 in response to stimulation with A23187 and low concentrations of PAF.256 Although activation of PLC and the associated production of 1,2-DC is independent of the formation of arachidonate metabolites, the release of arachidonate metabolites does enhance the formation of PA.383 It would be anticipated that increased production of Tsz in combination with enhanced sensitivity to TxA2 mimetics would lead to increased PLC activity in affected platelets. However, there appeared to be no correlation between Tx32 production and PLC activation, as indicated by 1,2-DC production. It is not possible to make direct comparisons between Tx82 production and 1,2-DC formation since the measurement of Tsz was performed in aggregating preparations in the presence of low concentrations of external Ca++ 256 while the measurement of 1,2-DC was performed in non-aggregating conditions in the absence of external Ca++ (lmM EDTA). It would be interesting to measure Tsz production in the 258 presence of EDTA to remove external cations and inhibit aggregation. If under these conditions, affected platelets do not produce increased concentrations of Tsz, then it is possible that decreased activation of PLA2 would be associated with decreased mobilization of internal Ca++ as discussed in the previous chapter. PLAZ is a Ca++-dependent enzyme27’397 which is activated by increasing [Ca"""]i_;122'151 however, PLA2 activity is also influenced by other factors including cytosolic pH,27 PKC activitylzz'lsz'150 and G protein modulation.6v137'280-396 Alternatively, if affected platelets still produce significantly increased quantities of Tsz in the presence of external divalent cation chelators, then the lack of significantly increased effect of the released arachidonic acid metabolites on PLC activation would be due to other factors. It has been demonstrated that affected platelets have increased basal cAMP concentrations.48'49 Increased cAMP inhibits PLC- mediated hydrolysis of membrane phosphoinositides.205'344'430'472 Therefore, the increased cAMP could offset the increased activation of PLC by the endogenously produced arachidonate metabolites. Alternatively, increased 1,2-DC production in response to U46619/epinephrine may have resulted from increased sensitivity of affected platelets to epinephrine. The exact mechanism for epinephrine- mediated potentiation of platelet activation is not well defined. By itself, epinephrine does not stimulate PLC activity383. Stimulation of PLC by epinephrine is dependent on Na+/H+ exchange and the local production of arachidonic acid metabolites.20'409 Epinephrine does potentiate Ca++ influx, Ca++ release from internal stores, protein phosphorylation of the 20kDa and 47kDa proteins, Na+/H+ exchange and PA 259 formation.75'219'290'402 The potentiation of platelet responses can occur independent of PLC,290 Na+/H+ exchange,402 fibrinogen binding317 and TxA2 production.317 To definitively determine whether the response indicates increased sensitivity to epinephrine would require the addition of subthreshold concentrations of epinephrine to platelets suspensions followed by the measurement of 1,2-DC production in response to the other agonists which have been shown to cause similar 1,2-DC production in affected and control platelets. CHAPTER 5: SUMMARY AND CONCLUSIONS Previous studies on BHT have shown that affected platelets have normal quantities of membrane GPIIb-IIIa299 and bind fibrinogen following ADP stimulation.119’298 Affected platelets do not aggregate in response to those agonists which are fully or partially dependent on arachidonic acid metabolism and PLC activation (ie. ADP, PAF and calcium ionophore A23187).254 They do aggregate when phosphoinositide hydrolysis is bypassed (ie. phorbol ester and high concentrations of thrombin). In spite of an inability to aggregate, ATP dense granule secretion occurs in affected platelets in response to all agonists tested except for U46619/epinephrine and calcium ionophore A23187.254 The activation responses in affected platelets are more sensitive to cyclooxygenase inhibition than those in control platelets. Affected platelets also produce significantly greater quantities of Tx32 in response to all those agonists capable of stimulating secretion.256 Phosphorylation of the two major proteins (20 and 47kDa) is similar in control and affected platelets following agonist stimulation. Unexpectedly, affected platelets have decreased phosphorylation of a 64- 67kDa protein of unknown identity or functional significance.255 Finally, basal concentrations of cAMP is increased in affected platelets.“8'49 260 261 In the research for the current dissertation, resting and post- stimulation [Ca++]i was measured. In the presence of external Ca++, affected aequorin-loaded platelets have normal (ie.thrombin, ionomycin, ADP and PAP) to decreased (ie. PMA) Ca++ mobilization relative to control platelets while affected fura 2-loaded platelets have normal (ie. thrombin, ionomycin and PMA) to increased (ie. ADP and PAP) Ca++ mobilization. When external Ca++ is removed by EGTA, affected aequorin- loaded platelets have normal (ie. ADP, PAF, thrombin and PMA) to decreased (ie. ionomycin) post-stimulation [Ca++]i relative to control platelets while affected fura 2-loaded platelets have normal (ie. ADP and phorbol) to decreased (ie. PAF, thrombin and ionomycin) post- stimulation [Ca++]1. There was no difference between control and affected platelets in resting [Ca++]1. These findings suggest that affected platelets have normal Ca++ homeostasis in the resting state. They appear to be able to mobilize Ca++ as well as control platelets when the influx of external Ca++ can occur. When external Ca++ is removed, the internal release of Ca++ is impaired. The previously- identified defects in affected platelets do not appear to be associated with defective Ca++ metabolism since the aggregation, secretion and TxBZ production responses are abnormal in the presence of external Ca++, a situation where the Ca++ fluxes appear to be normal. The cause of the defective mobilization of internal Ca++ was not determined in the current study. Possible explanations include: (1) decreased production of the second messengers that mediate Ca++ release; (2) production of defective or non-functional second messengers; (3) decreased responsiveness of the DTS to the second messengers; or (4) 262 decreased stores of releasable Ca++ within the DTS. The first three could explain the decrease in response to the physiologic agonists. Increased cAMP, which has previously been identified in affected platelets,48’49 decreases Ca++ mobilization in response to the physiologic agonists.331-3SO'433 However, cAMP does not affect Ca++ release in response to the non-physiologic calcium ionophores.350 The decrease post-stimulation [Ca++]1 in response to ionomycin suggests that there is also a defect in the releasable stores of Ca++ within the DTS. The second portion of the dissertation research measured resting and post-stimulation pHi in affected and control platelets. There was no significant difference between control and platelets in resting pHi in stirred samples. In the presence of EGTA, a situation which inhibits fibrinogen binding and prevents aggregation, there was no significant difference between affected and control platelets in post-stimulation pHi. In the presence of external Ca++, there was significantly decreased post-stimulation pHi in affected platelets in response to those agonists (ie. ADP, PAF, ionomycin, U46619/epinephrine) which cause aggregation in control, but not affected, platelets. In response to thrombin, the only agonist tested which causes aggregation in both affected and control platelets, there was no difference between the two groups. These results demonstrate that affected platelets do have a functional NaI/H+ antiport as indicated by the similar pHi responses in control and affected platelets in the presence of external EGTA. The results in the presence of external Ca++ suggest that the decreased 263 post-stimulation pHi in the affected platelets is partially, if not fully, due to the lack of aggregation. In response to the weak agonist ADP, affected platelets do not show any significant post-stimulation increase in pHi; yet, they have increased production of Tsz. It has been proposed that the weak agonists initiate platelet activation by producing a local, membrane associated increase in pHi. This local alkalinization then causes activation of PLAZ, the production of arachidonic acid metabolites and subsequent activation of PLC.409'410'411'412 Since affected platelets produce increased quantities of TxA2 in response to ADP,256 it is likely that a local cytosolic alkalinization is occurring but is not detectable using the fluorescent pH indicator BCECF in platelet suspensions. The final phase of the dissertation research assessed PLC activation in intact platelets by measurement of the production of the second messenger 1,2-DC in a non-aggregating preparation. When compared to control platelets, affected platelets produce similar quantities of 1,2-DC relative to control platelets in response to all agonists tested with the exception of TxA2 mimetic U46619/epinephrine. U46619/epinephrine causes increased production of 1,2-DC in affected platelets. These findings suggest that the defect in BHT does not result from lack of activation of PLC. The increased responsiveness to U46619/epinephrine suggests increased responsiveness to U46619 and/or epinephrine. The results described in the three portions of the dissertation research do not explain the platelet defect in BHT; however, they do 264 help to further localize the defect. Unfortunately, may events occur simultaneously during platelet activation, making it difficult to determine which events are primary and which events are secondary. In addition, the different biochemical pathways involved in the activation sequence are not isolated; instead, there are numerous checks and balances. Figure 30 presents a schematic model of platelet activation. Based on this model, the following conclusions on affected platelets can be drawn from previous studies as well as those reported in this dissertation: l. Membrane CPIIb-IIIa complex is present in normal amounts299 and is able to bind fibrinogen following ADP-stimulation.119'298 2. Na+/H+ exchange is functional and produces sufficient alkalinization to support PLA2 activity. Whether it is under normal con;rol mechanisms needs to be determined using a more sensitive technique than BCECF in platelet suspensions. 3. The pathway leading to the TxA2 production is intact but not under proper regulatory control. Affected platelets are much more dependent on this pathway for activation than control.254 4. Based on measurement of 1,2-DC production, PLC activation is similar in control and affected platelets in response to most agonists. This suggests that the primary defect in BHT must lie in some pathway other than PLC. There is increased activation of PLC by U46619 and/or epinephrine. 5. When PKC is activated directly, full platelet activation occurs.254 The mechanism by which PKC mediates aggregation is not known. 265 6. The influx of external Ca++ across the plasma membrane is normal in affected platelets while the release from internal stores is impaired. The amount released is sufficient to support the Ca++-dependent processes of shape change, phosphorylation of the 20kDa protein and PLA2 activity. 7. Phosphorylation of the major cytoskeletal proteins (20kDa and 47kDa) is normal in non-aggregating preparations.255 Phosphorylation of the 20kDa protein is associated with shape change. The role of the 47kDa protein in platelet activation is under debate. It is not clear how these proteins mediate secretion and aggregation. 8. There is decreased phosphorylation of a 64-67kDa protein of unknown function or identity.255 In conclusion, these findings do not fit any described platelet thrombopathy and cannot be explained by current understanding of platelet function. The results do suggest areas for further . investigation including: 1. Closer examination of the thromboxane pathway including measurement of PLA2 activity, arachidonic acid release and PGHz/TxAz receptor number and affinity. 2. Further investigation of the 64-6IkDa protein which is not phosphorylated as well in affected platelets. This would include protein isolation, production of a monoclonal antibody and protein sequencing. 266 3. Further examination of the process of aggregation. None of the findings thus far can explain the lack of primary aggregation in affected platelets. Inherently, it seems logical that if platelets can bind fibrinogen, they can also aggregate. Therefore, further examination of surface charge and membrane fluidity is warranted. auracomsf MOMS? Figure 30. Schematic model for platelet activation. Modified from Kroll et a1.212 LIST OF REFERENCES Adelstein RS and Conti MA. Phosphorylation of platelet myosin increases actin-activated myosin ATPase activity. Nature 256: 597-598, 1975. Adunyah SE and Dean WL. Effects of sulfhydryl reagents and other inhibitors on Ca2+ transport and inositol triphosphate- induced Ca2+ release from human platelet membranes. J Biol Chem 261: 13071-13075, 1986. Adunyah SE and Dean WL. Regulation of human platelet membrane Ca2+ transport by cAMP- and calmodulin-dependent phosphorylation. Biochim Biophys Acta 930: 401-409, 1987. Agam G, Argaman A and Livne A. Thrombin-induced platelet aggregation is affected by external Na+ independently of the Na /H exchange. FEBS Letts 244: 231-236, 1989. Akiba S, Sato T, Fujii T. Differential effects of phorbol 12- myristate l3-acetate on GTPyS-induced diacylglycerol formation and arachidonic acid liberation in saponin-permeabilized rabbit platelets. Thromb Res 53: 503-512, 1989. Akiba S, Sato T and Fujii T. Enhancement of arachidonic acid liberation by protein kinase C activator is partially dependent on extracellular Na+ in rabbit platelets. FEBS Letts 254: 29-32, 1989. Akiba S, Sato T and Fujii T. Involvement of a guanine- nucleotide-binding protein-mediated mechanism in the enhancement of arachidonic acid liberation by phorbol l2- myristate lB-acetate and Ca2+ in saponin-permeabilized platelets. Biochim Biophys Acta 1044: 291-296, 1990. Asch AS, Barnwell J, Silverstein RL and Nachman RL. Isolation of the thrombospondin membrane receptor. J Clin Invest 79: 1054-1061, 1987. Ashby B, Daniel JL and Smith JB. Mechanisms of platelet activation and inhibition. Hematol/ Oncol Clinics N Amer 4: l- 26, 1990. 267 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 268 Ashby B, Kowalska MA, Wernick E, Rigmaiden M, Daniel JL and Smith JB. Differences in the mode of action of 1-oleoyl-2- acetlyl-glycerol and phorbol ester in platelet activation. J Cyclic Nucleotide Prot Phosphorylation Res 10: 473-483, 1985. Authi KS and Crawford N. Inositol 1,4,S-trisphosphate~induced release of sequestered Ca2+ from highly purified human platelet intracellular membranes. Biochem J 230: 247-253, 1985. Authi KS, Evenden BJ and Crawford N. Metabolic and functional consequences of introducing inositol 1,4,5-trisphosphate into saponin-permeabilized human platelets. Biochem J 233: 709- 718, 1986. Avdonin PV and Altukhova IP. Blocking of receptor-dependent calcium channels of platelets by protein kinase C activator phorbol ester. Biokhimiya 50: 1235-1240, 1985. Avdonin PV, Cheglakov IB, Boogry EM, Suitina-Ulitina IV, Mazaev AV and Tkachuk VA. Evidence for the receptor-operated calcium channels in human platelet plasma membrane. Thromb Res 46: 29-37, 1987. Avdonin PV, Suitina-Ultina IV, Leytin VL and Tkachuk VA. Interaction of stable prostaglandin endoperoxide analogs U46619 and U44069 with human platelet membranes: Coupling of receptors with high-affinity GTPase and adenylate cyclase. Thromb Res 40: 101-112, 1985. Balduini CL, Bertolino G, Norris P, Sinigaglia F, Bisio A and Torti M. Interrelation of platelet aggregation, release reaction and thromboxane A2 production. Biochem Biophys Res Commun 156: 823—829, 1988. ' Baldassare JJ and Fisher GJ. Regulation of membrane- associated and cytosolic phospholipase C activities in human platelets by guanosine triphosphate. J Biol Chem 261: 11942- 11944, 1986. Baldassare JJ, Henderson PA and Fisher GJ. Isolation and characterization of one soluble and two membrane-associated forms of phosphoinositide-specific phospholipase C from human platelets. Biochemistry 28: 6010-6016, 1989. Ballou LR, DeWitt LM and Cheung WY. substrate-specific forms of human platelet phospholipase A2. J Biol Chem 261: 3107: 3111, 1986. Banga HS, Simons ER, Brass LF and Rittenhouse SE. Activation of phospholipase A and C in human platelets exposed to epinephrine. Role of glycoproteins IIb/IIIa and dual role of epinephrine. Proc Natl Acad Sci USA 83: 9197-9201, 1986. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 269 Banga HS, Walker RK, Winberry LR and Rittenhouse SE. Platelet adenylate cyclase and phospholipase C are affected differentially by ADP-ribosylation. Effects on thrombin- mediated responses. Biochem J 252 297-300, 1988. Banno Y, Nagao S, Katada T, Nagata K, Ui M and Nozawa Y. Stimulation by CTP-binding proteins (Gi' Go) of partially purified phospholipase C activity from human platelet membranes. Biochem Biophys Res Commun 146: 861-869, 1987. Banno Y, Nakashima S and Nozawa Y. Partial purification of phosphoinositide phospholipase C from human platelet cytosol: Characterization of its three forms. Biochem Biophys Res Commun 136: 713-721, 1986. Banno Y, Nakashima S, Tohmatsu T, Nozawa Y and Lapetina EG. CTP and GDP will stimulate platelet cytosolic phospholipase C independently of Ca2+. Biochem Biophys Res Commun 140: 728- 734, 1986. Banno Y, Yada Y and Nozawa Y. Purification and characterization of membrane-bound phospholipase C specific for phosphoinositides from human platelets. J Biol Chem 263: 11459-11465, 1988. Barber AJ and Jameson GA. Isolation and characterization of plasma membranes from human blood platelets. J Biol Chem 245: 6357-6365, 1970. Baron BM and Limbird LE. Human platelet phospholipase A2 activity is responsive in vitro to pH and Ca2 variations which parallel those occurring after platelet activation in vivg. Biochim Biophys Acta 971: 103-111, 1988. Bauer S and Jakobs RH. Phorbol ester treatment inpairs hormone-but not stable GTP analogue-induced inhibition of adenylate cyclase. FEBS Letts 198: 43-46, 1986. Baumgartner HR, Muggli R, Tschoop TB and Turitto VT. Platelet adhesion, release and aggregation in flowing blood: Effect of surface properties and platelet function. Thromb Haemostas 35: 124-138, 1976. Behnke O and Zelander T. substructure in negatively stained microtubules of mammalian blood platelets. Exp Cell Res 43: 236, 1967. Bell RL and Majerus PW. Thrombin-induced hydrolysis of phosphatidylinositol in human platelets. J Biol Chem 255: 1790-1792, 1980. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 270 Bell TG, Leader RW, Olson PM, Padgett GA, Penner JA and Patterson WR. Basset hound hereditary thrombopathy: An autosomally, recessively inherited platelet dysfunction with 11 cases in a kindred of 56 dogs. In: Que Medicine, Ryder 0A and Byrd ML (editors). Springer-verlag, New York, 1984. Pages 335-344. Berridge MJ. Inositol trisphosphate and diacylglycerol as second messengers. Biochem J 220: 345-360, 1984. Best L, Bone EA and Russell RGG. Phosphoinositide metabolism in human blood platelets: Effects of two types of divalent cation chelators. Thromb Res 26: 241-247, 1982. Bienz D and Clemetson KJ. Human platelet glycoprotein Ia. One component is only expressed on the surface of activated platelets and may be a granule constituent. J Biol Chem 264: 507-514, 1989. Bienz D, Schnippering W and Clementson KJ. Glycoprotein V is not the thrombin activation receptor on human blood platelets. Blood 68: 720-725, 1986. Billah MM and Lapetina EC. Rapid decrease of phosphatidylinositol 4,5-biphosphate in thrombin-stimulated platelets. J Biol Chem 257: 12705-12708, 1982. Billah MM, Lapetina EG and Cuatrecasas P. Phospholipase A2 activity specific for phosphatidic acid. A possible mechanism for the production of arachidonic acid in platelets. J Biol Chem 256: 5399-5403, 1981. Bishop RW, August J, Petrin JM and Pai JR. Regulation of sn- 1,2-diacy1glycerol second-messenger formation in thrombin- stimulated human platelets. Potentiation by protein kinase C inhibitors. Biochem J 269: 465-473, 1990. Bishop WR and Bell RM. Attenuation of sn-1,2-diacy1glycerol second messengers. Metabolism of exogenous diacylglycerols by human platelets. J Biol Chem 261: 12513-12519, 1986. Bishop WR, Ganong BR and Bell RM. Attenuation of sn-1,2- diacylglycerol second messengers by diacylglycerol kinase. Inhibition by diacylglycerol analogues luggigzg and in human platelets. J Biol Chem 261: 6993-7000, 1986. Blache D, Ciavatti M and Ojeda C. The effect of calcium channel blockers on blood platelet function, especially calcium uptake. Biochim Biophys Acta 923: 401-412, 1987. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 271 Blache D, Ciavatti M, Ponsin G and Nargoet J. Direct evidence for the modulation of human platelet cytosolic free Ca2+ by intracellular cyclic AMP produced with a photoactivatable derivative. Biochem Biophys Res Commun 146: 321-331, 1987. Blinks JR, Wier WG, Hess P and Prendergast FG. Measurement of Ca2+ concentrations in living cells. Prog Biophys Molec Biol 40: 1-114, 1982. Borin ML, Pinelis VG, Azizowa 0A, Kudinov YV, Markov CM, Cragoe EJ and Khodorov BI. Na+/H+ exchange in PAF-stimulated platelets. J Lipid Mediators 1: 257-272, 1989. Borin ML, Pinelis VG, Ivanova MA, Kudinov YV, Azizova 0A, Markov CM and Kjodorov BI. Blockade of ADP-induced Ca2+- signal and platelet aggregation by lipoxygenase inhibitors. FEBS Letts 257: 345-347, 1989. Bosia A, Losche W, Pannocchia A, Treves S, Ghigo D, Till U and Pescarmona G. Regulation of arachidonic acid-dependent Ca++ influx in human platelets. Thromb Haemostas 59: 86-92, 1988. Boudreaux MK, Dodds WJ, Slauson D0 and Catalfamo JL. Evidence for regulatory control of canine platelet phosphodiesterase. Biochem Biophys Res Commun 140: 589-594, 1986. Boudreaux MK, Dodds WJ, Slauson D0 and Catalfamo JL. Impaired cAMP metabolism associated with abnormal function of thrombopathic canine platelets. Biochem Biophys Res Commun 140: 595-601, 1986. Brace LD, Benton DL and LeBreton GC. Thromboxane A2/ prostaglandin H2 mobilizes calcium in human blood platelets. Am J Physiol 249: H1-H7, 1985. Brass, LF. Ca2+ homeostasis in unstimulated platelets. J Biol Chem 259: 12563-12570, 1984. Brass LF. Ca2+ transport across the platelet plasma membrane. A role for membrane glycoproteins IIb and Illa. J Biol Chem 260: 2231-2236, 1985. Brass LF. Regulation of biological responses in platelets. In: W Atherosclerosis; Meyer and Marche (editors), Raven Press, New York, 1989. Pages 59-92. Brass LP. The effect of Na+ on Ca2+ homeostasis in unstimulated platelets. J Biol Chem 259: 12571-12575, 1984. Brass LF and Jose h K. Role for inositol triphosphate in intracellular Ca2 mobilization and granule secretion in platelets. J Biol Chem 260: 15172-15179, 1985. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 272 Brass LF and Laposata M. Diacylglycerol causes Ca release from the platelet dense tubular system: Comparisons with Ca release caused by inositol 1,4,5-triphosphate. Biochem Biophys Res Commun 142: 7-14, 1987. Brass LF, Laposata M, Banga HS and Rittenhouse SE. Regulation of the phosphoinositide hydrolysis pathway in thrombin- stimulated platelets by a pertussis toxin-sensitive guanine nucleotide-binding protein. J Biol Chem 261: 16838-16847, 1986. Brass LF, Shaller CC and Belmonte EJ. Inositol 1,4,5- triphosphate-induced granule secretion in platelets. Evidence that the activation of phospholipase C mediated by thromboxane receptors involves a guanine nucleotide binding protein- dependent mechanism distinct from that of thrombin. J Clin Invest 79: 1269-1275, 1987. Brass LF and Shattil SJ. Changes in surface-bound and exchangeable calcium during platelet activation. J Biol Chem 257: 14000-14005, 1982. Brass LF and Shattil SJ. Identification and function of the high affinity binding sites for Ca2+ on the surface of platelets. J Clin Invest 73: 626-632, 1984. Brass LF and Shattil SJ. Inhibition of thrombin-induced platelet activation by leupeptin. Implications for the participation of calpain in the initiation of platelet activation. J Biol Chem 263: 5210-5216, 1988. Brass LF and Shattil SJ. Interaction of extracellular calcium with the surface membrane of human platelets. Methods Enzymol 169: 355-371, 1989. ’ Brass LF, Shattil SJ, Kunicki TJ and Bennett JJ. Effect of calcium on the stability of the platelet membrane glycoprotein IIb-IIIa complex. J Biol Chem 260: 7875-7881, 1985. Broekman, J. Phosphatidylinositol 4,5-bisphosphate may represent the site of release of plasma membrane-bound calcium upon stimulation of human platelets. Biochem Biophys Res Commun 120: 226-231, 1984. Bruns C, Schachtele C and Marme D. Synthetic diacylglycerols induce a rise of quin2-detectable free intracellular calcium in human platelets. FEBS Letts 221: 23-27, 1987. Bush LR and Smith SG. Antagonism of U46619-induced aggregation of human and canine platelets by four TxA2 receptor antagonists. Thromb Res 44: 377-389, 1986. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 273 Catalfamo JL, Raymond SL, White JG and Dodds WJ. Defective platelet-fibrinogen interaction in hereditary canine thrombopathia. Blood 67: 1568-1577, 1986. Cattaneo M, Canciani MT and Mannucci PM. Human platelet aggregation and release reaction induced by platelet activating factor (PAF-Acether)- Effects of acetylsalicylic acid and external ionized calcium. Throm Haemostas 52: 221- 224, 1985. Carey F, Menashi S and Crawford N. Localization of cyclooxygenase and thromboxane synthetase in human platelet intracellular membranes. Biochem J 204: 847-851, 1982. Carty DJ, Jones GD, Freas DL and Gear ARL. Effect of epinephrine on rapid ADP-induced aggregation, protein phos- phorylation, and cytoplasmic calcium dynamics of platelets: A quenched-flow study. J Lab Clin Med 112: 603-611, 1988. Castagna M. Phorbol esters as signal transducers and tumor promoters. Biol Cell 59: 3-13, 1987. Castagna M, Takai Y, Kaibuchi K, Sano K, Kikkawa U and Nishizuka Y. Direct activation of calcium-activated, phospholipid-dependent protein kinase by tumor-promoting phorbol esters. J Biol Chem 257: 7847-7851, 1982. de Chaffoy de Courcelles D, Roevens P and VanBelle H. Agents that elevate platelet cAMP stimulate the formation of phosphatidylinositol 4-phosphate in intact human platelets. FEBS Letts 195: 115-118, 1986. de Chaffoy de Courcelles D, Roevens P and VanBelle H. Prostaglandin E1 and forskolin antagonize C-kinase activation in the human platelet. Biochem J 244: 93-99, 1987. de Chaffoy de Courcelles D, Roevens P, Van Belle H and DeClerck F. The synergistic effect of serotonin and epinephrine on the human platelet at the level of signal transduction. FEBS Letts 219: 283-288, 1987. Chau LY and Tai HH. Monoglyceride and diglyceride lipases from human platelet microsomes. Biochim Biophys Acta 963: 436-444, 1988. Clemens N, Siffert W and Schneid P. Thrombin-induced cytosolic alkalinization in human platelets occurs without an apparent involvement of HCO3'/Cl' exchange. Pflugers Arch 416: 68-73, 1990. Clemetson RJ and Luscher EF. Membrane glycoprotein abnormalities in pathological platelets. Biochim Biophys Acta 947: 53-73, 1988. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 274 Cobbold PH and Rink TJ. Fluorescence and bioluminescence measurement of cytosolic free calcium. Biochem J 248: 313- 328, 1987. Coller BS. Interaction of normal, thrombasthenic, and Bernard-Soulier platelets with immobilized fibrinogen: Defective platelet fibrinogen interaction in thrombasthenia. Blood 55: 169-178, 1980. Colman RW. Aggregin: a platelet ADP receptor that mediates activation. FASEB J 4: 1425-1435, 1990. Colman RW. Platelet receptors. Hematol/ Oncol Clinics N Amer 4: 27-42, 1990. Colman RW. Platelet activation: Role of an ADP receptor. Sem Hematol 23: 119-128, 1986. Connolly TM, Bansal VS, Bross TE, Irvine RF and Majerus PW. The metabolism of tris- and tetraphosphates of inositol by 5- phosphomonoesterase and 3-kinase enzymes. J Biol Chem 262: 2146-2149, 1987. Connolly TM, Lawing WJ and Majerus PW. Protein kinase C phosphorylates human platelet inositol triphosphate 5'- phosphomonesterase, increasing the phosphatase activity. Cell 46: 951-958, 1986. Connolly TM and Limbird LE. Removal of extraplatelet Na+ eliminates indomethacin-sensitive secretion from human platelets stimulated by epinephrine, ADP and thrombin. Proc Natl Acad Sci USA 80: 5320-5324, 1983. Connolly TM, Wilson DB, Bross TE and Majerus PW. Isolation and characterization of the inositol cyclic phosphate products of phosphoinositide cleavage by phospholipase C. Metabolism in cell-free extracts. J Biol Chem 261: 122-126, 1986. Conti MA and Adelstein S. The relationship between calmodulin binding and phosphorylation of smooth muscle myosin kinase by the catalytic subunit of 3':5'cAMP-dependent protein kinase. J Biol Chem 256: 3178-3181, 1981. Cramer EM, Sabidge GF, Vainchenker W, Berndt MC, Pidard D, Caen JP, Masse JM and Breton-Gorius J. Alpha-granule pool of glycoprotein IIb-IIIa in normal and pathologic platelets and megakaryocytes. Blood 75: 1220-1227, 1990. Crouch MF and Lapetina EG. A role for 61 in control of the thrombin receptor-phospholipase C coupling in human platelets. J Biol Chem 263: 3363-3371, 1988. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. ' a human platelet Ca2+-ATPase. Cell Calcium 10: 289-297, 1989. 102. 275 Crouch MF and Lapetina EG. Phosphorylation of the 47kDa protein in gamma-thrombin stimulated human platelets does not activate phospholipase A2: Evidence against lipocortin. Biochem Biophys Res Commun 141: 459-465, 1986. Daniel JL, Dangelmaier CA, Selak M and Smith JB. ADP stimulates 1P3 formation in human platelets. FEBS Letts 206: 299-303, 1986. . Daniel JL, Dangelmaier CA and Smith JB. Calcium modulates the generation of inositol 1,3,4-triphosphate in human platelets by the activation of inositol 1,4,5-triphosphate 3-kinase. Biochem J 253: 789-794, 1988. Daniel JL, Dangelmaier CA and Smith JB. Formation and metabolism of inositol 1,4,5-triphosphate in human platelets. Biochem J 246: 109-114, 1987. Daniel JL, Molish IR, Rigmaiden M and Stewart G. Evidence for a role of myosin phosphorylation in the initiation of the platelet shape change response. J Biol Chem 259: 9826-9831, 1984. Daniel JL, Purdon AD and Molish IR. Platelet myosin phosphorylation as an indicator of the cellular calcium concentration. Fed Proc 41: 1118, 1982. Daniel JL, Selak MA, Purdon AD and Salganicoff L. Methods for the study of the role of calcium in platelet function. Modern Methods in Pharmacology 4: 185-215, 1987. Davies TA, Dunn JM and Simons ER. Evaluation of changes in cytoplasmic pH in thrombin-stimulated human platelets. Analytical Biochem 167: 118-123, 1987. Davies TA, Weil GJ and Simons ER. Simultaneous flow cytometric measurements of thrombin-induced cytosolic pH and Ca2+ fluxes in human platelets. J Biol Chem 265: 11522-11526, 1990. Dean WL. Purification and reconstitution of a Ca2+ pump from human platelets. J Biol Chem 259: 7343-7348, 1984. Dean WL. Structure function and subcellular localization of Dean WL and Sullivan DM. Structural and functional properties of a Ca2+-ATPase from human platelets. J Biol Chem 257: 14390-14393, 1982. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 276 De Clerck F and de Chaffoy de Courcelles D. Amplification mechanisms in platelet activation. In: filggd_§g11§_§gg Arterigg in fiypgzgeggiog and Atherosglezosis, Raven Press, New York, 1989. De Groot PG and Sixma JJ. Platelet adhesion. Br J Haematol 75: 308-312, 1990. De Marco L, Girolami A, Zimmerman TS and Ruggeri ZM. von Willebrand factor interaction with the glycoprotein IIb/IIIa complex. Its role in platelet function as demonstrated in patients with congenital afibrinogenemia. J Clin Invest 77: 1272-1277, 1986. Dhar A, Paul AK and Shukla SD. Platelet-activating factor stimulation of tyrosine kinase and its relationship to phospholipase C in rabbit platelets: Studies with genistein and monoclonal antibody to phosphotyrosine. Molecul Pharmacol 37: 519-525, 1990. Dorn GW. Distinct platelet thromboxane A2/ prostaglandin H2 receptor subtypes. A radioligand binding study of human platelets. J Clin Invest 84: 1883-1891, 1989. Dorn GW. Isoelectric and mass characterization of human platelet thromboxane A2/ prostaglandin H2 receptors. Biochem Biophys Res Commun 163: 183-188, 1989. Doyle VM and Ruegg UT. Lack of evidence for voltage dependent calcium channels on platelets. Biochem Biophys Res Commun 127: 161-167, 1985. Elce JS, Sigmund L and Fox MJ. Calpain I activation is not correlated with aggregation in human platelets. ‘Biochem J 261: 1039-1042, 1989. Enouf J, Bredoux R, Boizard B, Wautier JL, Chap H, Thomas J, de Metz M and Levy-Toledano S. Simultaneous isolation of two platelet membrane fractions: Biochemical, immunological and functional characterization. Biochem Biophys Res Commun 123: 50-58, 1984. Enouf J, Giraud F, Bredoux R, Bourdeau N and Levy-Toledano 8. Possible role of a cAMP-dependent phosphorylation in the calcium release mediated by inositol 1,4,5-triphosphate in human platelet membrane vesicles. Biochim Biophys Acta 928: 76-82, 1987. Enouf J, Giraud F, Bredoux R, Bourdeau N and Levy-Toledano S. Stimulation of the 23-Kd protein cAMP dependent phosphorylation by inositol 1,4,5-triphosphate in human platelet membrane vesicles. Biochem Biophys Res Commun 145: 139-145, 1987. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 277 Enouf J, Bredoux R, Bourdeau N and Levy-Toledano S. Two different Ca2+ transport systems are associated with plasma and intracellular human platelet membranes. J Biol Chem 262: 9293-9297, 1987. Enouf J, Lompre AM, Bredoux R, Bourdeau N, LaBastie D and Levy-Toledano S. Different sensitivity to trypsin of the human platelet plasma and intracellular membrane Ca2+ pumps. J Biol Chem 263: 13922-13929, 1988. Enyedi A, Sarkadi B, Foldes-Papp Z, Monostory S and Gardos G. Demonstration of two distinct calcium pumps in human platelet membrane vesicles. J Biol Chem 261: 9558-9563, 1986. Erne P, Schachter M, Fabbro D, Miles CMM and Sever PS. Calcium transients in human platelets monitored by aequorin, furs-2 and quin 2: Effects of protein kinase C activation and inhibition. Biochem Biophys Res Commun 145: 66-72, 1987. Escolar G, Leistikow E and White JG. The fate of the open canalicular system in surface and suspension-activated platelets. Blood 74: 1983-1988, 1988. Estry DW, Mattson JC, Oesterle JR, Patterson WR and Bell TG. Analysis of ADP-induced fibrinogen receptors in dogs with Basset hound hereditary thrombopathy: a comparison with normal dog and human platelets. Submitted for publication. Feinstein MB, Egan JJ, Sha'afi RI and White J. The cytoplasmic concentration of free calcium in platelets is controlled by stimulators of cyclic AMP production (PGD2, PGEl, forskolin). Biochem Biophys Res Commun 113: 598-604, 1983. Feinstein MB and Fraser C. Human platelet secretion and aggregation induced by calcium ionophores. Inhibition by PGEl and dibutyryl cyclic AMP. J Gen Physiol 66: 561-581, 1975. Feinstein MB and Halenda SP. Arachidonic acid mobilization in platelets: The possible role of protein kinase C and G- proteins. Experientia 44: 101-104, 1988. Figures WR, Scearce LM, Wachtfogel Y, Chen J, Colman RF and Colman RW. Platelet ADP receptor and a 2-adrenoreceptor . interaction. Evidence for an ADP requirement for epinephrine- induced platelet activation and an influence of epinephrine on ADP binding. J Biol Chem 261: 5981-5986, 1986. Fischer GJ, Bakshian S and Baldassare JJ. Activation of human platelets by ADP causes a rapid rise in cytosolic free calcium without hydrolysis of phosphatidylinositol-4,5-bisphophate. Biochem Biophys Res Commun 129: 958-964, 1985. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 278 Fischer TH and White GC. cAMP-dependent protein kinase substrates in platelets. Biochem Biophys Res Commun 159: 644- 650, 1989. Fitzgerald LA and Phillips DR. Calcium regulation of the platelet membrane glycoprotein IIb-IIIa complex. J Biol Chem 260: 11366-11374, 1985. Folco G, Granstom E and Kindahl H. Albumin stabilizes thromboxane A2. FEBS Letts 82: 321-324, 1977. Fox JEB. Identification of actin-binding protein as the protein linking the membrane skeleton to glycoproteins on platelet plasma membranes. J Biol Chem 260: 11970-11977, 1985. Fox JEB. Linkage of a membrane skeleton to integral membrane glycoproteins in human platelets. Identification of one of the glycoproteins as glycoprotein Ib. J Clin Invest 76: 1673- 1683, 1985. Fox JEB. The platelet cytoskeleton. In: Thrombosis and Haemostasis 1982. Verstraete M, Vermylen J, Lijnen HR and Arnout J (Editors), International Society on Thrombosis and Haemostasis and Leuven University Press, 1987. Pages 175-225. Fox JEB, Boyles JK, Berndt MC, Steffen PK and Anderson LK. Identification of a membrane skeleton in platelets. J Cell Biol 106: 1525-1538, 1988. Fox JEB, Goll DE, Reynolds CC and Phillips DR. Identification of two proteins (actin-binding protein and P235) that are hydrolyzed by endogenous Ca2+-dependent protease during platelet aggregation. J Biol Chem 260: 1060-1066, 1985. Fox JEB, Reynolds CC and Johnson MM. Identification of glycoprotein IB as one of the major proteins phosphorylated during exposure of intact platelets to agonists that activate cyclic AMP-dependent protein kinases. J Biol Chem 262: 12627- 12631, 1987. Fox JEB, Reynolds CC and Phillips DR. Calcium-dependent proteolysis occurs during platelet aggregation. J Biol Chem 258: 9973-9981, 1983. Funder J, Hershco L, Rothstein A and Livne A. Na+/H+ exchange and aggregation of human platelets activated by ADP: The exchange is not required for aggregation. Biochim Biophys Acta 938: 425-433, 1988. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 279 Fuse 1, Iwanaga T and Tai HH. Phorbol ester, 1,2- diacylglycerol, and collagen induce inhibition of arachidonic acid incorporation into phospholipids in human platelets. J Biol Chem 264: 3890-3895, 1989. Fuse I and Tai HH. Stimulation of arachidonate release and inositol-1,3,5-triphophate formation are mediated by distinct G-proteins in human platelets. Biochem Biophys Res Commun 146: 659-665, 1987. Ganong BR, Loomis CR, Hannun YA and Bell RM. Specificity and mechanism of protein kinase C activation by sn-1,2- diacylglycerols. Proc Natl Acad Sci USA 83: 1184-1188, 1986. Gerrard JM, Beattie LL, Park J, Israels SJ, McNichol A, Lint D and Cragoe EJ. A role for protein kinase C in the membrane fusion necessary for platelet granule secretion. Blood 74: 2405-2413, 1989. Gerrard JM, Beattie LL, McCrae JM and Singhroy S. The influence of lysophosphatidic acid on platelet protein phosphorylation. Biochem Cell Biol 65: 642-650, 1987. Gerrard JM, Kindom SE, Peterson DA, Peller J, Krantz RE and White JG. Lysophosphatidic acids. Influence on platelet aggregation and intracellular calcium flux. Am J Pathol 96: 423-438, 1979. Gerrard JM, McNichol A, Klassen D and Israels SJ. Protein phosphorylation and its relation to platelet function. In: 'H-C- .Tin'i '. ‘_11r- ‘1'°1-‘19 ‘ ie_°‘. r-:_ Meyer P and Marche P. (Editors) Raven Press, New York, 1989. Pages 93-114. Gerrard JM and White JG. The influence of aspirin and indomethacin on the platelet contractile wave. Am J Pathol 82: 513-526, 1976. Gerwitz AM and Hoffman R. Human megakaryocyte production: Cell biology and clinical considerations. Hematol/ Oncol Clinics N America 4: 43- 64, 1990. Ghigo D, Treves S, Turrini F, Pannocchia A, Pescarmona G and Bosia A. Role of Na+/H+ exchange in thrombin- and arachidonic acid- induced Ca2+ influx in platelets. Biochim Biophys Acta 940: 141-148.1988. Grinstein S. The intracellular pH of white blood cells: Measurement and regulation. Biochem Cell Biol 66: 245-249, 1988. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 280 Grottum RA and Solum NO. Congenital thrombocytopenia with giant platelets: a defect in the platelet membrane. Br J Haematol 16: 277-290, 1969. G kiewicz G, Poenie M and Tsien RY. A new generation of Ca + indicators with greatly improved fluorescence properties. J Biol Chem 260: 3440-3450, 1985. Hack N, Croset M and Crawford N. Studies on the bivalent- cation activated ATPase activities of highly purified human platelet surface and intracellular membranes. Biochem J 233: 661-668, 1986. Halenda SP, Banga HS, Zavoico GB, Lau LF and Feinstein MB. Synergistic release of arachidonic acid from platelets by activators of protein kinase C and Ca2+ ionophores. Evidence for the role of protein phosphorylation in the activation of phospholipase A2 and independence from the Na+/H+ exchanger. Biochemistry 28: 7356-7362, 1989. Halenda SO and Rehm AG. Direct stimulation of arachidonic acid release by calcium in human platelets. Fed Proc 46: 398, 1987. Halenda SP, Zavoico GB and Feinstein MB. Phorbol esters and oleoyl acetoyl glycerol enhance release of arachidonic acid in platelets stimulated by Ca2+ ionophore A23187. J Biol Chem 260: 12484-12491, 1985. Hallam TJ, Daniel JL, Kendrick-Jones J and Rink TJ. Relationship between cytoplasmic free calcium and myosin light .chain phosphorylation in intact platelets. Biochem J 232: 373-377, 1985. Hallam TJ and Rink TJ. Agonists stimulate divalent cation channels in the plasma membrane of human platelets. FEBS Letts 186: 175-179, 1985. Hallam TJ, Sanchez A and Rink TJ. Stimulus-response coupling in human platelets. Changes evoked by platelet-activating factor in cytoplasmic free calcium monitored with fluorescent calcium indicator quin 2. Biochem J 218: 819-827, 1984. Hamburg M, Svensson J, Wakabayashi T and Samuelsson B. Isolation and structure of two prostaglandin endoperoxides that cause platelet aggregation. Proc Natl Acad Sci USA 71: 345-349, 1974. Harmon JT and Jamieson GA. Platelet activation by thrombin in the absence of the high-affinity thrombin receptor. Biochemistry 27: 2151-2157, 1988. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 281 Hashimoto K, Im T, Tatsumi N, Okuda K and Yukioka M. Modulation of actin polymerization by 47,000-da1ton protein of human platelets. Biochem Internat 14: 759-767, 1987. Haslam RJ and Davidson MML. Potentiation by thrombin of the secretion of serotonin from permeabilized platelets equilibrated with Ca2+ buffers. Biochem J 222: 351-351, 1934. Haslam RJ and Davidson MML. Receptor-induced diacylglycerol formation in permeabilized platelets: Possible role for a GTP- binding protein. J Receptor Res 4: 605-629, 1984. Haslam RJ and Lynham JA. Relationship between phosphorylation of blood platelet proteins and secretion of platelet granule constituents. I. Effects of different aggregating agents. Biochem Biophys Res Commun 77: 714-722, 1977. Haslam RJ and Vanderwel M. Inhibition of platelet adenylate cyclase by l-O-a1ky1-2-acety1-sn-glyceryl-3-phosphory1choline (platelet-activating factor). J Biol Chem 257: 6879-6885, 1982. Hatayama K, Kambayashi J, Kawasaki T, Morimoto K, Ohshiro T and Mori T. Interrelationships between secretion, protein phosphorylation and intracellular Ca2+ concentration in platelets stimulated by thrombin or thromboxane A2 analogue. Thromb Res 41: 761-770, 1986. Hathaway DR and Adelstein RS. Human platelet mysoin light chain kinase requires the calcium-binding protein calmodulin for activity. Proc Natl Acad Sci USA 76: 1653-1657, 1979. Hathaway DR, Eaton CR and Adelstein RS. Regulation of human platelet myosin light chain kinase by the catalytic subunit of cyclic AMP-dependent protein kinase. Nature 291: 252-254, 1981. Herbener GH and Dean WL. Immunocytochemical localization of the Ca2+-ATPase polypeptide in human platelets. Biochem Biophys Res Commun 153: 848-854, 1988. Hettasch JM and LeBreton GC. Modulation of Ca2+ fluxes in isolated platelet vesicles: Effects of cAMP-dependent protein kinase and protein kinase inhibitor on Ca2+ sequestration and release. Biochim Biophys Acta 931: 49-58, 1987. Hoffman BB, Michel T, Brenneman TB and Lefkowitz RJ. Interactions of agonists with platelet 112-adrenergic receptors. Endocrinology 110: 926-932, 1982. Holmsem H. Platelet metabolism and activation. Sem Hematol 22: 219-240, 1985. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 282 Holmsen H, Dangelmaier CA and Rongved S. Tight coupling of thrombin-induced acid hydrolase secretion and phosphatidate synthesis to receptor occupancy in human platelets. Biochem J 222: 157-167, 1984. Holmsen H, Kaplan KL and Dangelmaier CA. Differential energy requirements for platelet responses. A simultaneous study of aggregation, three secretory processes, arachidonate liberation, phosphatidylinositol turnover and phosphatidate formation. Biochem J 208: 9-18, 1982. Holmsen H, Setkowsky CA and Day HJ. Effects of antimycin and 2-deoxyglucose on adenine nucleotides in human platelets. Role of metabolic adenosine trisphosphate in primary aggregation, secondary aggregation and shape change in platelets. Biochem J 144: 385-396, 1974. Hornberger W and Patscheke H. Primary stimuli of eicosanoid release inhibit arachidonoyl-CoA sythetase and lysophospholipid acyl transferase. Mechanism of action of hydrogen peroxide and methyl mercury in platelets. Eur J Biochem 187: 175-181, 1990. Houslay MD, Bojanic D and Wilson A. Platelet activating factor and U46609 stimulate a GTPase activity in human platelets which is distinct from the guanine nucleotide regulatory proteins Ns and Ni- Biochem J 234: 737-740, 1986. Hrbolich JK, Culty M and Haslam RJ. Activation of phospholipase C associated with isolated rabbit platelet membranes by guanosine 5'-Py-thiol]triphosphate and by thrombin in the presence of GTP, Biochem J 243: 457-465, 1987. Hunyady L, Balazs S, Cragoe EJ, Spat A and Gardos G. Activation of sodium-proton exchange is not a prerequisite for Ca + mobilization and aggregation in human platelets. FEBS Letts 225: 72-76, 1987. Ikebe M. Phosphorylation of a second site for myosin light chain kinase on platelet myosin. Biochemistry 28: 8750-8755, 1989. Ikebe M, Inagaki M, Kanamaru K and Hidaka H. Phosphorylation of smooth muscle myosin light chain kinase by Ca2+-activated, phospholipid-dependent protein kinase. J Biol Chem 260: 4547- 4550, 1985. Imai A, Ishizuka Y, Kawai K and Nozawa Y. Evidence for coupling of phosphatidic acid formation and calcium influx in thrombin-activated human platelets. Biochem Biophys Res Commun 108: 752-759, 1982. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 283 Imaoka T, Lynham JA and Haslam RJ. Purification and characterization of the 47,000-da1ton protein phosphorylated during degranulation of human platelets. J Biol Chem 258: 11404-11414, 1983. Isenberg WM, Fox JEB and Phillips DR. Platelet aggregation induces a redistribution of talin and the incorporation of talin and GPIIb-IIIa into the actin cytoskeleton. J Cell Biol 107: 256a, 1988. Jain NC. Megakaryocytopoiesis and platelet production, survival, and distribution. In: Schglm's Veterinary Hgmggglggy‘ Fourth edition, Lea and Febiger, Philadelphia, 1986. Pages 431-445. Jain NC. Qualitative and quantitative disorders of platelets. In: W Fourth edition. Lea and Febiger, Philadelphia, 1986. Pages 466-486 Jain NC. The Platelets: Structure, biochemical, and functional aspects. In: Sghalm's Vgtgrinagy Hematology, Fourth edition, Lea and Febiger, Philadelphia, 1986. Pages 446-465. Jandrot-Perrus M, Rendu F, Caen JP, Levy-Toledano and Guillin MC. The common pathway for alpha- and gamma-thrombin-induced platelet activation is independent of GPIb: a study of Bernard-Soulier platelets. Br J Haematol 75: 385-392, 1990. Janero DR, Burghardt B and Burghardt C. Specific binding of 1-O-a1ky1-2-acety1-sn-glycero-3-phosphocholine (platelet- activating factor) to the intact canine platelet. Thromb Res 50: 789-802, 1988. Jerushalmy Z, Englender T and Shaklai M. Phorbol-myristate- acetate-induced platelet aggregation in the presence of inhibitors. Acta Haematol 80: 210-215, 1988. Johnson PC, Ware JA, Cliveden PB, Smith M, Dvorak AM and Salzman EW. Measurement of ionized calcium in blood platelets with the photoprotein aequorin. Comparison with quin 2. J Biol Chem 260: 2069-2076, 1985. Johnson PC, Ware JA and Salzman EW. Concurrent measurement of platelet ionized calcium concentration and aggregation: Studies with the lumiaggregometer. Thromb Res 40: 435-443, 1985. Johnson PC, Ware JA and Salzman EW. Measurement of platelet cytoplasmic ionized calcium concentration with aequorin and fluorescent indicators. Methods in Enzymology 169: 386-415, 1989. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 284 Johnstone IB and Lotz F. An inherited platelet function defect in Basset hounds. Can Vet J 20: 211-215, 1979. Jy W and Haynes DH. Thrombin-induced calcium movements in platelet activation. Biochim Biophys Acta 929: 88-102, 1987. Kaibuchi Y, Takai Y and Nishizuka Y. Cooperative roles of various membrane phospholipids in the activation of calcium- activated phospholipid-dependent protein kinase. J Biol Chem 256: 7146-7149, 1981. Kaibuchi K, Takai Y, Sawamura M, Hoshijima M, Fujikura T and Nishizuka Y. Synergistic functions of protein phosphorylation and calcium mobilization in platelet activation. J Biol Chem 258: 6701-6704, 1983. Kameyama Y, Yoshioka S, Imai A and Nozawa Y. Possible involvement of l-acyl-glycerophosphorylinositol acyltransferase in arachidonic enrichment of phosphatidylinositol in human platelets. Biochim Biophys Acta 752: 244-250, 1983. Katada T, Gilman AG, Watanabe Y, Bauer S and Jakobs KH. Protein kinase C phosphorylates the inhibitory guanine- nuceotide-binding regulatory component and apparently suppresses its function in hormonal inhibition of adenylate cyclase. Eur J Biochem 151: 431-437, 1985. Katagiri Y, Hayashi Y, Yamamoto K, Tanoue K, Kosaki G and Yamazaki H. Localization of von Willebrand factor and the thrombin-interactive domains on human platelet glycoprotein Ib. Thromb Haemostas 63: 122-126, 1990. Kaulen HD and Gross R. Metabolic properties of human platelet membranes II. Thrombin-induced phosphorylation of membrane lipids and demonstration of phosphorylating enzymes in the platelet membrane. Thromb Haemostas 35: 364-376, 1976. Kawamoto S, Bengur AR, Sellers JR and Adelstein RS. In situ phosphorylation of human platelet myosin heavy and light chains by protein kinase C. J Biol Chem 264: 2258-2265, 1989. Keraly CL, Delautier D and Benveniste J. The role of extracellular Ca2+ and Na+ in PAF-acether-induced human platelet activation. Biochem Pharmacol 38: 3415-3421, 1989. Keraly CL, Kinlough-Rathbone RL, Packham MA, Suzuki H and Mustard JF. Conditions affecting the responses of human platelets to epinephrine. Thromb Haemostas 60: 209-216, 1988. Kerry R, Scrutton MC and Wallis RB. Mammalian platelet adrenoreceptors. Br J Pharmacol 81: 91-102, 1984. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 285 Knight DE, Niggli V and Scrutton MC. Thrombin and activators of protein kinase C modulate secretory responses of permeabilized human platelets induced by Ca2+. Eur J Biochem 143: 437-446, 1984. Knight DE and Scrutton MC. Direct evidence for a role for Ca2 in amine storage granule secretion by human platelets. Thromb Res 20: 437-446, 1980. Knight DE and Scrutton MC. Cyclic nucleotides control a system which regulates Ca2+ sensitivity of platelet secretion. Nature 309: 66-68, 1984. Kornecki E, Ehrlich YH, Hardwick DH and Lenox RH. Proteolysis of the platelet surface: Dissociation of shape change from aggregation. Am J Physiol 250: H550-H557, 1986. Kornecki E, Walkowiak B, Naik UP and Ehrlich YH. Activation of human platelets by a stimulatory monoclonal antibody. J Biol Chem 265: 10042-10048, 1990. Kraft AS and Anderson WB. Phorbol esters increase the amount of Ca2+, phospholipid-dependent protein kinase associated with the plasma membrane. Nature 301: 621-623, 1983.725 Krishnamurthi S, Joseph S and Kakkar VV. Effect of phorbol 12-myristate l3-acetate (PMA) on agonist-induced arachidonate release and S-hydroxytryptamine secretion in human platelets. Dependence of effects on agonist type and time of incubation with PMA. Biochim Biophys Acta 927: 429-436, 1987. Krishnamurthi S, Patel Y and Kakkar V. Inhibition of agonist- induced platelet aggregation, Ca2+ mobilization and granule secretion by guanosine 5' -[p -thio]diphosphate and GDP in intact platelets. Evidence for an inhibitory mechanism unrelated to the inhibition of G-protein-GTP interaction. Biochem J 250: 209-214, 1988. Krishnamurthi S, Morgan WA and Kakkar VV. Extracellular Na+ removal enhances granule secretion in platelets-evidence that Na+/H+ exchange is inhibitory to secretion induced by some agonists. FEBS Letts 250: 195-200, 1989. Kroll MH and Schafer AI. Biochemical mechanisms of platelet activation. Blood 74: 1181-1195, 1989. Kroll MH and Schafer AI. Control of platelet protein kinase C activation by cAMP. Biochim Biophys Acta 970: 61-67, 1988. Kroll MH, Zavoico GB and Schafer AI. Second messenger function of phosphatidic acid in platelet activation. J Cell Physiol 139: 558-564, 1989. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 286 Kunicki TJ. Platelet membrane glycoproteins and their function: An overview. Blut 59: 30-34, 1989. Kurth MC and Bryan J. Platelet activation induces the formation of a stable gelosin-actin complex from monomeric gelsolin. J Biol Chem 259: 7473-7479, 1984. Lages B and Weiss HJ. Time dependence of aequorin-indicated calcium levels in stimulated and unstimulated platelets: Evidence for multiple aequorin environments in platelets. Thromb Haemost 62: 1094-1099, 1989. Lanza F, Beretz A, Kubina M and Cazenave JP. Increased aggregation and secretion responses of human platelets when loaded with the calcium fluorescent probes quin2 and furs-2. Thromb Haemostas 58: 737-743, 1987. Lanza F, Beratz A, Stierle A, Hanau D, Kubina M and Cazenave JP. Epinephrine potentiates human platelet activation but is not an aggregating agent. Am J Physiol 255: H1276-H1288, 1988. Lapetina EG. Incorporation of synthetic 1,2-diacylglycerol into platelet phophatidylinositol is increased by cyclic AMP. FEBS Letts 195: 111-114, 1986. Lapetina EG. Platelet-activating factor stimulates the phosphatidylinositol cycle. J Biol Chem 257: 7314-7317, 1982. Lapetina EC and Cuatrecasas P. Stimulation of phosphatidic acid production in platelets precedes the formation of arachidonate and parallels the release of serotonin. Biochim Biophys Acta 573: 394-402, 1979. Lapetina EG and Reep BR. Specific binding of [a -32P]GTP to cytosolic and membrane-bound proteins of human platelets correlates with the activation of phospholipase C. Proc Natl Acad Sci USA 84: 2261-2265, 1987. Lapetina EG, Reep B, Ganong BR and Bell RM. Exogenous sn-l,2- diacylglycerols containing fatty acids function as bioregulators of protein kinase C in human platelets. J Biol Chem 260: 1358-1361, 1985. Lapetina EG, Reep B and Watson SP. Ionophore A23187 stimulates phosphorylation of the 40,000 dalton protein in human platelets without phospholipase C activation. Life Sci 39: 751-759, 1986. 226. 227. 228. 229. 230. 231. 232. 233. 234. 235. 236. 287 Lapetina £6, 81110 J and Ruggiero M. Thrombin induces serotonin secretion and aggregation independently of inositol phospholipids hydrolysis and protein phosphorylation in human platelets permeabilized with saponin. J Biol Chem 260: 7078- 7083, 1985. Largarde M, Guichardant M, Menashi S and Crawford N. The phospholipid and fatty acid composition of human platelet surface and intracellular membranes isolated by high voltage free flow electrophoresis. J Biol Chem 257: 3100-3103, 1982. Lebowitz EA and Cooke R. Contractile properties of actomyosin from human blood platelets. J Biol Chem 253: 5443-5447, 1978. Lecompte T, Potevin F, Champeix, Morel MC, Favier R, Hurtaud MF, Schlegel N, Samama M and Kaplan C. Aequorin-detected calcium changes in stimulated thrombasthenic platelets. Aggregation-dependent calcium movement in response to ADP. Thromb Res 58: 561-570, 1990. Leistikow EA, Barnhart MI, Escolar G and White JG. Receptor- 1igand complexes are cleared to the open canalicular system of surface-activated platelets. Br J Haematol 74: 93-100, 1990. Leven RM, Gonnella PA, Reeber MJ and Nachmias VT. Platelet shape change and cytoskeletal assembly: Effects of pH and monovalent cation ionophores. Thromb Haemostas 49: 230-234, 1983. Levy-Toledano S, Maclouf J, Bryon P, Savariau E, Hardisty RM and Caen JP. Human platelet activation in the absence of aggregation: A calcium-dependent phenomenon independent of thromboxane formation. Blood 59: 1078-1085, 1982. Liel N, Mais DE and Halushka PV. Desensitization of platelet thromoxane Az/ prostaglandin H2 receptors by the mimetic U46619. J Pharmacol Exper Therapeutics 247: 1133-1138, 1988. Lind SE, Yin HL and Stossel TP. Human platelets contain gelsolin. A regulator of actin filament length. J Clin Invest 69: 1384-1387, 1982. Lindberg U and Markey F. Platelet microfilaments and motility. In: 0 MacIntyre DE and Gordon JL (editors). Elsevier, Amsterdam, 1987. Pages 205-238. Litchfield DW and Ball EH. Phosphorylation of high molecular weight proteins in platelets treated with 12-0- tetradecanoylphorbol-13-acetate. Biochem Internat 20: 615- 621, 1990. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 288 Livne A, Grinstein S and Rothstein A. Characterization of Na+/H+ exchange in platelets. Thromb Haemostas 58: 971-977, 1987. Low MG, Carroll RC and Weglick WB. Multiple forms of phosphoinositide-specific phospholipase C of different relative molecular masses in animal tissues. Biochem J 221: 813-820, 1984. Lyons RM and Atherton RM. Characterization of a platelet protein phosphorylated during the thrombin-induced release reaction. Biochemistry 18: 544-552, 1979. MacIntyre DE, Bushfield M and Shaw AM. Regulation of platelet cytosolic free calcium by cyclic nucleotides and protein kinase C. FEBS Letts 188: 383-388, 1985. MacIntyre DE, McNichol A and Drummond AH. Tumor-promoting phorbol esters inhibit agonist-induced phosphatidate formation and Ca2+ flux in human platelets. FEBS Letts 180: 160-164, 1985. Magocsi M, Enyedi A, Sarkadi B and Gardos G. Effects of phosphoinositides on calcium movements in human platelet membrane vesicles. Biochim Biophys Acta 944: 202-212, 1988. Magocsi M, Sarkadi B, Kovacs T and Gardos G. Thrombin-induced activation of calcium transport pathways and their role in platelet functions. Biochim Biophys Acta 984: 88-96, 1989. Mahadevappa VG and Holub BJ. The molecular species composition of individual diacyl phospholipids in human platelets. Biochim Biophys Acta 713: 73-79, 1982. Mahadevappa VG and Sicilia F. Mobilization of arachidonic acid in thrombin-stimulated human platelets. Biochem Cell Biol 68: 520-527, 1990. Mahnensmith RL and Aronson PS. The plasma membrane sodium- hydrogen exchanger and its role in physiological and pathophysiological processes. Circulation Res 57: 773-788, 1985. Majerus PW, Connolly TM, Bansal VS, Inhorn RC, Ross TS and Lips DL. Inositol phosphates: Synthesis and degradation. J Biol Chem 263: 3051-3054, 1988. Manne V. A novel candidate for receptor-coupled phospholipase C purified from human platelets. Oncogene 3: 579-585, 1988. Marche P. Polyphosphoinositides and cell activation. Nouv Rev Fr Hematol 27: 223-228, 1985. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 289 Marcus AJ, Safier LB, Ullman HL, Wong KTH, Broekman J, Weksler BB and Kaplan KL. Effects of acetyl glyceryl ether phosphorylcholine on human platelet function in vitro. Blood 58: 1027-1031, 1981. Mattson JC, Estry DW, Bell TG and Patterson WR. Defective contact activation of platelets from dogs with Basset hound hereditary thrombopathy. Thromb Res 44: 23-28, 1986. Mauco G, Dajeans P, Chap H and Douste-Blazy L. subcellular localization of inositol lipids in blood platelets as deduced from the use of labelled precursors. Biochem J 244: 757-761, 1987. Mazur EM. Megakaryocytopoiesis and platelet production: A review. Exp Hematol 15: 340-350, 1987. McConnell MF. Platelet aggregation, dense granule secretion and transmission electronmicroscopy. In:_£ungtigngl_§nd o em a Abncrm u d e e Thrombopathy, Doctoral Dissertation, Michigan State University, 1989, Pages 124-218. McConnell MF. Phosphorylation of platelet proteins. In: ti nd B c e ca bn rmal es 1 d Hereditary41hrgmhgpathy‘ Doctoral Dissertation, Michigan State University, 1989, Pages 219-298. McConnell MF. Measurement of thromboxane B2 production. In: u t na c e es s on Hereditggy Thrombgpgthy, Doctoral Dissertation, Michigan State University, 1989, Pages 299-365. McCulloch RK, Summers J, Vandongen R and Rouse IL. Role of thromboxane A2 as a mediator of platelet-activating-factor- induced aggregation of-human platelets. Clin Sci 77: 99-103, 1989. McCulloch RK and Vandongen R. Mechanisms of platelet activating factor-induced aggregation and secretion in human platelets. Prostaglandins 39: 13-21, 1990. McEver RP. Properties of GMP-l40, an inducible granule membrane protein of.plate1ets and endothelium. Blood Cells 16: 73-83, 1990. McEver RP. The clinical significance of platelet membrane glycoproteins. Hematol/ Oncol Clinics N Amer 4: 87-105, 1990. McEver RP, Bennett EM and Martin MN. Identification of two structurally and functionally distinct sites on human platelet membrane glycoprotein IIb-IIIa using monoclonal antibodies. J Biol Chem 258: 5269-5275, 1983. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272. 273. 290 McGowan EB and Detweiler TC. Modified platelet responses to thrombin. Evidence for two types of receptors or coupling mechanisms. J Biol Chem 261: 739-746, 1986. McNichol A, Gerrard JM and MacIntyre DE. Evidence for two mechanisms of thrombin-induced platelet activation: One proteolytic, one receptor mediated. Biochem Cell Biol 67: 332-336, 1989. McPherson J, Brownlea and Zucker MB. Effect of monoclonal antibodies against von Willebrand factor and platelet glycoproteins IIb/IIIa on the platelet retention test. Blood 70: 546-550, 1987. Menashi S, Davis C and Crawford N. Calcium uptake associated with an intracellular membrane fraction prepared from human blood platelets by high-voltage, free flow electrophoresis. FEBS Lett 140: 298-301, 1982. Meyers KM. Pathobiology of animal platelets. Adv Vet Sci Comparat Med 30: 131-165, 1985. Meyers KM, Holmsen H and Seachord CL. Comparative study of platelet dense granule constituents. Am J Physiol 243: R454- R462, 1982. Moos M and Goldberg ND. Cyclic AMP opposes IP3-induced calcium release from permeabilized human platelets. Second Messengers and Phosphoproteins 12: 163-170, 1988. Morrison WJ and Shukla SD. Antagonism of platelet activating factor receptor binding and stimulated phosphoinositide- specific phospholipase C in rabbit platelets. J Pharmacol Exper Therapeutics 250: 831-835, 1989. Motulsky HJ, Shattil SJ, Ferry N, Rozansky D and Insel PA. Desensitization of epinephrine-initiated platelet aggregation does not alter binding to the az-adrenergic receptor or receptor coupling to adenylate cyclase. Mole Pharmacol 29: 1- 6, 1986. Motulsky HJ, Shattil SJ and Insel PA. Characterization of¢2- adrenergic receptors on human platelets using [3H]-yohimbine. Biochem Biophys Res Commun 97: 1562-1570, 1980. Morrison.WJ and Shukla SD. Desensitization of receptor- coupled activation of phosphoinositide-specific phospholipase C in platelets: Evidence for distinct mechanisms for platelet- activating factor and thrombin. Molecul Pharmacol 33: 58-63, 1988. Murer EH. Role of platelet calcium. Sem Hematol 22: 313-323, 1985. 274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284. 291 Murray R and Fitzgerald GA. Regulation of thromboxane receptor activation in human platelets. Proc Natl Acad Sci USA 86: 124-128, 1989. Naka M, Nishikawa M, Adelstein RS and Hidaka H. Phorbol ester-induced activation of human platelets is associated with protein kinase C phosphorylation of myosin light chains. Nature 306: 490-492, 1983. Naka M, Saitoh M and Hidaka H. Two phosphorylated forms of myosin in thrombin-stimulated platelets. Arhcives Biochem Biophys 261: 235-240, 1988. Nakamura K, Kambayashi J, Suga K, Hakata H and Mori T. Hydrolysis of polyphosphoinositides in human platelets. Thromb Res 38: 513-525, 1985. Nakano T, Hanasaki K and Arita H. Different effects of two thromboxane A2/prostaglandin H2 receptor ligands, U46619 and S-145, on rabbit platelets. FEBS Letts 234: 309-312, 1988. Nakano T, Hanasaki K and Arita H. Role of protein kinase C in U46619-induced platelet shape change, aggregation and secretion. Thromb Res 56: 299-306, 1989. Nakashima S, Hattori H, Shirato L, Takenaka A and Nozawa Y. Differential sensitivity of arachidonic acid release and 1,2 diacylglycerol formation to pertussis toxin, GDIBS and NaF in saponin-permeabilized human platelets: Possible evidence for distinct CTP-binding proteins involving phospholipase C and A2 activation. Biochem Biophys Res Commun 148: 971-978, 1987. Newman PJ, McEver RP, Doers MP and Kunicki TJ. Synergistic action of two murine monoclonal antibodies that inhibit ADP- induced platelet aggregation without blocking fibrinogen binding. Blood 69: 668-676, 1987. Nieuwenhuis HK, Akkerman JWN, Houdjik WPM and Sixma JJ. Human blood platelets showing no reponse to collagen fail to express surface glycoprotein Ia. Nature 318: 470-472, 1985. Nieuwenhuis HK, Sakariassen KS, Houdijk WPM, Nievelstein PFEM and Sixma JJ. Deficiency of platelet membrane glycoprotein la is associated with a decreased platelet adhesion to ‘ subendothelium: A defect in platelet spreading. Blood 68: 692-695, 1986. Niiya K, Hodson E, Bader R, Byers-Ward V, Koziol JA, Plow EF and Ruggeri ZM. Increased surface expression of the membrane glycoprotein IIb/IIIa complex induced by platelet activation. Relationship to the binding of fibrinogen and platelet aggregation. Blood 70: 475-483, 1987. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 292 Nunez D and Levy-Toledano S. Further characterization of human platelet activation in the absence of aggregation: Phosphorylation or specific proteins and relationship with platelet secretion. Thromb Haemostas 51: 198-203, 1984. Nurden AT. Abnormalities of platelet glycoproteins in inherited disorders of platelet function. In: £1§§212§§_in Biology gnd Pathglggy 111, MacIntyre and Gordon (eds). Elsevier Science Publishers, 1987. Pages 37-94. Ogawa A, Ishikawa Y and Sasakawa S. Cytoplasmic pH dependency of Na+/H+ exchange in human platelets activated with thrombin, arachidonic acid, A23187 and TPA. Thromb Res 55: 683-693, 1989. Ogawa A, Ishikawa Y and Sasakawa S. Effects of changes in intracellular pH induced by platelet storage and treatment with monovalent cation ionophores on platelet functions. Thromb Res 52: 369-379, 1988. Oho C and Takisawa H. Distinct occurrence of phosphatidyl- inositol 4,5-biphosphate induced Ca2+ release and inositol 1,4,S-triphosphate-induced release in ATP-dependent Ca2+- transporting platelet microsomes. J Biochem 100: 911-921, 1986. Olbrich C, Aepfelbacher M and Siess W. Epinephrine potentiates calcium mobilization and activation of protein kinases in platelets stimulated by ADP through a mechanism unrelated to phospholipase C. Cellular Signaling 1: 483-492, 1989. Opstvedt A, Rongved S, Aarsether N, Lellehaug JR and Holmsen H. Differential effects of chlorpromazine on secretion, protein phosphorylation and phosphoinositide metabolism in stimulated platelets. Biochem J 238: 159-166, 1986. O'Rourke F and Feinstein MB. The inositol 1,4,5-trisphosphate receptor binding sites of platelet membranes. Biochem J 267: 297-302, 1990. ' O'Rourke F, Zavoico GB, Smith LH and Feinstein MB. Stimulus- response coupling in a cell-free platelet membrane system. CTP-dependent release of Ca2+ by thrombin, and inhibitiOn by pertussis toxin and a monoclonal antibody that blocks calcium release by 1P3. FEBS Letts 214: 176-180, 1987. Ozaki Y, Yatomi Y, Kariya T and Kume S. Anion channels contribute to the regulation of intracellular pH in human platelets. Thromb Res 53: 221-230, 1989. Packham MA. The behavior of platelets at foreign surfaces. Proc Soc Exper Biol Med 189: 261-274, 1988. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 293 Parise LV, Criss AB, Nannizzi L and Wardell MR. Glycoprotein IIIa is phosphorylated in intact human platelets. Blood 75: 2362-2368, 1990. Parmentier S, Kaplan C, Catimel B and McGregor JL. New families of adhesion molecules play a vital role in platelet functions. Immunol Today 11: 225-227, 1990. Patterson WR, Estry DW, Schwartz KA, Borchert RD and Bell TG. Absent platelet aggregation with normal fibrinogen binding in Basset hound hereditary thrombopathy. Thromb Haemostas 62: 1011-1015, 1989. Patterson WR, Kunicki TJ and Bell TG. Two-dimensional electrophoretic studies of platelets from dogs affected with Basset hound hereditary thrombopathy: A thrombasthenia-like aggregation defect. Thromb Res 42: 195-203, 1986. Patterson WR, Padgett GA and Bell TG. Abnormal release of storage pool adenine nucleotides from platelets of dogs affected with Basset hound hereditary thrombopathy. Thromb Res 37: 61-71, 1985. Pelech SL, Charest DL, Howard SL, Paddon HB and Salari H. Protein kinase C activation by platelet-activating factor is independent of enzyme translocation. Biochim Biophys Acta 1051: 100-107, 1990. Peerschke EIB. Irreversible platelet fibrinogen interactions occur independently of fibrinogen alpha chain degradation and are not mediated by intact platelet membrane glycoprotein IIb-IIIa complexes. J Lab Clin Med 1111: 84-92, 1988. Peerschke EIB. The platelet fibrinogen receptor. Sem Hematol 22: 241-259, 1985. Peerschke EIB and Wainer JA. Examination of irreversible platelet-fibrinogen interactions. Am J Physiol 238: C466, 1985. Peerschke EIB and Zucker MB. Fibrinogen receptor exposure and aggregation of human blood platelets produced by ADP and chilling. Blood 57: 663-670, 1981. Peterson OH. Does inositol tetrakisphosphate play a role in the receptor-mediated control of calcium mobilization? Cell Calcium 10: 375-383, 1989. Phillips DR, Charo IF, Parise LV and Fitzgerald LA. The platelet membrane glycoprotein IIb-IIIa complex. Blood 71: 831-843, 1988. 308. 309. 310. 311. 312. 313. 314. 315. 316. 317. 318. 294 Phillips DR, Jennings LK and Edwards HH. Identification of membrane proteins mediating the interaction of human platelets. J Cell Biol 86: 77-86, 1980. Pietu G, Fressinaud E, Girma JP, Nieuwenhuis HK, Rothschild C and Meyer D. Binding of human von Willebrand factor to collagen and to collagen-stimulated platelets. J Lab Clin Med 109: 637-646, 1987. Polasek J, Richardson M, Moore MA and Blajchman MA. Evidence for an alternative mechanism of human platelet secretion involving peripheralization of secretory granules and formation of membrane-associated multivesicular structures. Thromb Res 45: 771-782, 1987. Poll C, Kyrle P and Westwick J. Activation of protein kinase C inhibits sodium fluoride-induced elevation of human platelet cytosolic free calcium and thromboxane B2 generation. Biochem Biophys Res Commun 136: 381-389, 1986. Pollock WK and MacIntyre DE. Desensitization and antagonism of vasopressin-induced phosphoinositide metabolism and elevation of cytosolic free calcium concentration in human platelets. Biochem J 234: 67-73, 1986. Pollock WK and Rink TJ. Thrombin and ionomycin can raise platelet cytosolic Ca2+ to micromolar levels by discharge of internal Ca2+ stores: Studies using fura-2. Biochem Biophys Res Commun 139: 308-314, 1986. Pollock WK, Rink TJ and Irvine RF. Liberation of [3H]arachidonic acid and changes in cytosolic free calcium in fura-2-loaded human platelets stimulated by ionomycin and collagen. Biochem J 235: 869-877, 1986. Pollock WK, Sage SO and Rink TJ. Stimulation of Ca2+ efflux from fura-2-1oaded platelets activated by thrombin or phorbol myristate acetate. FEBS Letts 210: 132-136, 1987. Powling MJ and Hardisty RM. Glycoprotein IIb-IIIa complex and Ca2+ influx into stimulated platelets. Blood 66: 731-734, 1985. Powling MJ and Hardisty RM. Potentiation by adrenaline of Ca2+ influx and mobilization in stimulated human platelets: Dissociation from thromboxane generation and aggregation. Thromb Haemostas 59: 212-215, 1988. Preiss J, Loomis CR, Bishop WR, Stein R, Niedel JE and Bell RM. Quantitative measurement of sn-1,2-diacy1glycerol present in platelets, hepatocytes, and ras- and sis-transformed normal rat kidney cells. J Biol Chem 261: 8597-8600, 1986. 319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 295 Purdon AD, Patelunas D and Smith JB. Evidence for the release of arachidonic acid through the selective action of phospholipase A2 in thrombin-stimulated human platelets. Biochim Biophys Acta 920: 205-214, 1987. Purdon AD and Smith JB. Turnover of arachidonic acid in the major diacyl and ether phospholipids of human platelets. J Biol Chem 260: 12700-12704, 1985. Puri RN, Zhou F, Bradford H, Hu CJ, Colman RF and Colman RW. Thrombin-induced platelet aggregation involves an indirect cleavage of aggregin by calpain. Arch Biochem Biophys 271: 346-358, 1989. Pytela R, Pierschbacher MD, Ginsberg MH, Plow EF and Rouslahti E. Platelet membrane glycoprotein IIb-IIIa: Member of a family of Arg-Gly-Asp-specific adhesion receptors. Science 231: 1559-1562, 1986. Rana RS and Hokin LE. Role of phosphoinositides in transmembrane signaling. Physiol Rev 70: 115-164, 1990. Rao AK and Kowalska MA. ADP-induced platelet shape change and mobilization of cytoplasmic ionized calcium are mediated by distinct binding sites on platelets: 5'-p- fluorosulfonylbenzoyl-adenosine is a weak platelet agonist. Blood 70: 751-756, 1987. Rao AK, Kowalska MA and Disa J. Impaired cytoplasmic ionized calcium mobilization in inherited platelet secretion defects. Blood 74: 664-672, 1989. Rao AK, Willis J, Kowalska MA, Wachtfogel YT and Colman RW. Differential requirements for platelet aggregation and inhibition of adenylate cyclase by epinephrine. Studies of a familial platelet alphaz-adrenergic receptor defect. Blood 71: 494-501, 1988. Regan JW, Nakata H, DeMarinis RM, Caron MG and Lefkowitz RJ. Purification and characterization of the human platelet e2- adrenergic receptor. J Biol Chem 261: 3894-3900, 1986. Rengasamy A and Feinberg H. Inositol 1,4,5-trisphosphate- induced calcium release from platelet plasma membrane vesicles. Biochem Biophys Res Commun 150: 1021-1026, 1988. Rengasamy A, Sours S and Feinberg H. Platelet Ca2+ homeostasis: Na+-Ca2+ exchange in plasma membrane vesicles. Thromb Haemostas 57: 337-340, 1987. Resink TJ, Dimitrov D, Stucki S and Buhler FR. The influence of activating hormones on human platelet membrane Ca2+-ATPase activity. Biochem Biophys Res Comm 138: 215-223, 1986. 331. 332. 333. 334. 335. 336. 337. 338. 339. 340. 341. 342. 343. 344. 296 Rink TJ. Cytosolic calcium in platelet activation. Experientia 44: 97-100, 1988. Rink TJ. Measurement of cytosolic calcium: Fluorescent calcium indicators. Mineral Electrolyte Metab 14: 7-14, 1988. Rink TJ and Pozzan T. Using quin2 in cell suspensions. Cell Calcium 6: 133-144, 1985. Rink TJ, Sanchez A and Hallam TJ. Diacylglycerol and phorbol ester stimulate secretion without raising cytoplasmic free calcium in human platelets. Nature 305: 317-319, 1983. Rink TJ, Smith SW and Tsien RY. Cytoplasmic free Ca2+ in human platelets: Ca2+ thresholds and Ca-independent activation for shape-change and secretion. FEBS Letts 148: 21-26, 1982. Rink TJ, Tsien RY and Pozzan T. Cytoplasmic pH and free Mg++ in lymphocytes. J Cell Biol 95: 189-196, 1982. Rittenhouse SE. Activation of human platelet phospholipase C by ionophore A23187 is totally dependent upon cyclo-oxygenase products and ADP. Biochem J 222: 103-110, 1984. Rittenhouse SE. Human platelets contain phospholipase C that hydrolyzes polyphosphoinositides. Proc Natl Acad Sci USA 80: 5417-5420, 1983. Rittenhouse SE and Horne WC. Ionomycin can elevate intraplatelet Ca2+ and activate phospholipase A without activating phospholipase C. Biochem Biophys Res Commun 123: 393-397, 1984. Rittenhouse SE and Sasson JP. Mass changes in myoinositol triphosphate in human platelets stimulated by thrombin. J Biol Chem 260: 8657-8660, 1985. Rittenhouse SE and Sasson JP. Measurement of 1P3 mass as a monitor of phospholipase C activation in stimulated human platelets. Nouv Rev Fr Hematol 27: 239-242, 1985. Rittenhouse-Simmons, S. Production of diglyceride from phosphatidylinositol in activated human platelets. J Clin Invest 63: 580-587, 1979. Rock CO and Jackowski S. Thrombin- and nucleotide-activated phosphatidylinositol 4,5-bisphosphate phospholipase C in human platelet membranes. J Biol Chem 262: 5492-5498, 1987. Rubin R and Hoek JB. Inhibition of ethanol-induced platelet activation by agents that elevate cAMP. Thromb Res 58: 625- 632, 1990. ' 345. 346. 347. 348. 349. 350. 351. 352. 353. 354. 355. 356. 357. 297 Ruggiero M and Lapetina EG. Sustained proteolysis is required for human platelet activation by thrombin. Thromb Res 42: 247-255, 1986. Ruggiero M and Lapetina EG. Thrombin and trypsin activate phospholipase C in human platelets via a proteolytic-mediated pathway. Sem Thromb Haemostas 12: 241-243, 1986. Rybak ME, Renzulli LA, Bruns MJ and Cahaly DP. Platelet glycoproteins IIb and IIIa as a calcium channel in liposomes. Blood 72: 714-720, 1988. Ryu SH, Lee SY, Lee KY and Rhee SG. Catalytic properties of inositol triphosphate kinase activation by Ca + and calmodulin. FASEB J 1: 388-393, 1987. Sage SO and Rink TJ. Effects of ionic substitution on [Ca2+]1 rises evoked by thrombin and PAF in human platelets. Eur J Pharmacol 128: 99-107, 1986. Sage SO and Rink TJ. Inhibition by forskolin of cytosolic calcium rise, shape change and aggregation in quin 2-loaded human platelets. FEBS Letts 188: 135-139, 1985. Sage SO and Rink TJ. Kinetic differences between thrombin- induced and ADP-induced calcium influx and release from internal stores in fura-2-loaded human platelets. Biochem Biophys Res Commun 136: 1124-1129, 1986. Sage SO and Rink TJ. The kinetics of changes in intracellular calcium concentration in fura-2-1oaded human platelets. J Biol Chem 262: 16364-16369, 1987. Saitoh M, Naka M and Hidaka H. Modulatory role of myosin light chain phosphorylation in human platelet activation. Biochem Biophys Res Commun 140: 280-287, 1987. Sakariassen KS, Bolhuis PA and Sixma JJ. Human blood platelet adhesion to artery subendothelium is mediated by factor VIII- von Willebrand factor bound to the subendothelium. Nature 279: 636-638, 1979. Salama SE and Haslam RJ. Characterization of the protein kinase activities of human platelet supernatant and particulate fraction. Biochem J 218: 285-294, 1984. Salimath BP, Mittelholzer E, Muggli R, Dieter P and Marme D. Microsomal Ca2+ uptake in human platelets is stimulated by calmodulin. Thromb Res 41: 519-526, 1986. Samis JA, Zboril G and Elce JS. Calpain I remains intact and intracellular during platelet activation. Biochem J 246: 481- 488, 1987. 358. 359. 360. 361. 362. 363. 364. 365. 366. 367. 368. 298 Sanchez A, Alonso MT and Collazos JM. Thrombin-induced changes of intracellular [Ca2+] and pH in human platelets. Cytoplasmic alkalinization is not prerequisite for calcium mobilization. Biochim Biophys Acta 938: 497-500, 1988. Sanchez A. Ca2+-independent secretion is dependent on cytoplasmic ATP in human platelets. FEBS Lett 191: 283-286, 1985. Sanchez A, Hallam TJ and Rink TJ. Trifluoperazine and chlorpromazine block secretion from human platelets evoked at basal cytoplasmic free calcium by activators of C-kinase. FEBS Lett 164: 43-46, 1983. Sano K, Nakamura H, Matsuo T, Kawahara Y, Fukuzoke H, Kaibuchi K and Takai Y. Comparison of the modes of action of Ca2+ ionophore A23187 and thrombin in protein kinase C activation in human platelets. FEBS Letts 192: 4-8, 1985. Sano K, Takai Y, Yamanashi J and Nishisuka Y. A role of calcium-activated, phospholipid dependent protein kinase in human platelet activation. J Biol Chem 258: 2010-2013, 1983. Saussy DL, Mais DE, Baron DA, Pepkowitz SH and Halushka PV. Subcellular localization of a thromboxane A2/ prostaglandin H2 receptor antagonist binding site in human platelets. Biochem Pharmacol 37: 647-654, 1988. Sawyer DB and Anderson 08. Platelet-activating factor is a general membrane perturbant. Biochim Biophys Acta 987: 129- 132, 1989. Schaeffer J and Blaustein MP. Platelet free calcium concentrations measured with fura-2 are influenced by the transmembrane sodium gradient. Cell Calcium 10: 101-113, 1989. Schmaier AH, Bradford HN, Lundberg D, Farber A and Colman RW. Membrane expression of platelet calpain. Blood 75: 1273-1281, 1990. Scott JL, Dunn SM, Jin B, Hallam AJ, Walton S, Berndt MC, Murray AW, Krissansen CW and Burns GF. Characterization of a novel membrane glycoprotein involved in platelet activation. J Biol Chem 264: 13475-13482, 1989. Sekar MC and Hokin LE. The role of phosphoinositides in signal transduction. J Membrane Biol 89: 193-210, 1986. 369. 370. 371. 372. 373. 374. 375. 376. 377. 378. 379. 380. 299 Sekiya F, Takagi J, Sasaki K, Kawajiri K, Kobayashi Y, Sato F and Saito Y. Feedback regulation of platelet function by 12S- hydroxyeicosatetraenoic acid: Inhibition of arachidonic acid liberation from phospholipids. Biochim Biophys Acta 1044: 165-168, 1990. Shattil SJ and Bennett JS. Platelets and their membranes in hemostasis: Physiology and pathophysiology. Ann Internal Med 94: 108-118, 1980. Shattil SJ and Brass LF. Induction of the fibrinogen receptor on human platelets by intracellular mediators. J Biol Chem 262: 992-1000, 1987. Shattil SJ, Budzynski A and Scrutton MC. Epinephrine induces platelet fibrinogen receptor expression, fibrinogen binding, and aggregation in whole blood in the absence of other excitatory agonists. Blood 73: 150-158, 1989. Shattil SJ, Hoxie JA, Cunningham M and Brass LF. Changes in the platelet membrane glycoprotein IIb-IIIa complex during platelet activation. J Biol Chem 260: 11107-11114, 1985. Shaw T. The role of blood platelets in nucleoside metabolism: Regulation of megakaryocyte developement and platelet production. Mutation Research 200: 67-97, 1988. Shukla SD, Franklin CC and Carter MG. Activation of phospholipase C in platelets by platelet activating factor and thrombin causes hydrolysis of a common pool of phosphatidylinositol 4,5-biphosphate. Biochim Biophys Acta 929: 134-141, 1987. Siess W. Mechanisms of platelet inhibition by prostacyclin: New aspects. In: Aghggg§g1§12§1§_gggigg§‘ Vol. 21, Raven Press Ltd., New York, 1990. Pages 159-164. Siess W. Molecular mechanisms of platelet activation. Physiol Rev 69: 58-178, 1989. Siess W and Binder H. Thrombin induces the rapid formation of inositol biphosphate and inositol triphosphate in human platelets. FEBS Letts 180: 107-111, 1985. Siess W, Boehlig B, Weber PC and Lapetina EG. Prostaglandin endoperoxide analogues stimulate phospholipase C and protein phosphorylation during platelet shape change. Blood 65: 1141- 1148, 1985. Siess W and Lapetina EG. Ca2+ mobilization primes protein kinase C in human platelets. Ca2+ and phorbol esters stimulate platelet aggregation and secretion synergistically through protein kinase C. Biochem J 255: 309-318, 1988. 381. 382. 383. 384. 385. 386. 387. 388. 389. 390. 391. 300 Siess W and Lapetina EG. Phorbol esters sensitize platelets to activation by physiological agonists. Blood 70: 1373-1381, 1987. Siess W, Siegel FL and Lapetina EG. Arachidonic acid stimulates the formation of 1,2-diacylglycerol and phosphatidic acid in human platelets. Degree of phospholipase C activation correlates with protein phosphorylation, platelet shape change, serotonin release and aggregation. J Biol Chem 258: 11236-11242, 1983. Siess W, Weber PC and Lapetina EG. Activation of phospholipase C is dissociated from arachidonate metabolism during platelet shape change induced by thrombin or platelet- activating factor. Epinephrine does not induce phospholipase C activation or platelet shape change. J Biol Chem 259: 8286- 8292, 1984. Siffert W and Akkerman JWN. Activation of sodium-proton exchange is a prerequisite for Ca2+ mobilization in human platelets. Nature 325: 456-458, 1987. Siffert W and Akkerman JWN. Cytosolic alkalinization alone is not sufficient for Ca2+ mobilization, phosphatidic acid formation, and protein phosphorylation in human platelets. Biochem Biophys Res Commun 161: 1007-1012, 1989. Siffert W and Akkerman JWN. Na+/H+ exchange as a modulator of platelet activation. Trends Biochem Sci 13: 148-151, 1988. Siffert W and Akkerman JWN. Protein kinase C enhances Ca2+ mobilization in human platelets by activating Na+/H+ exchange. J Biol Chem 263: 4223-4227,.1988. Siffert W, Gengenbach S and Scheid P. Inhibition of platelet aggregation by amiloride. Thromb Res 44: 235-240, 1986. Siffert W, Jakobs KH and Akkerman JWN. Sodium fluoride prevents receptor- and protein kinase C-mediated activation of the human platelet Na+/H+ exchanger without inhibiting its basic pHi-regulating activity. J Biol Chem 265: 15441-15448, 1990. Siffert W, Muckenhoff K and Scheid P. Evidence for a role of Na+/H+ exchange in platelets activated with calcium ionophore A23187. Biochem Biophys Res Commun 125: 1123-1128, 1984. Siffert W and Scheid P. A horbol ester and l-oleoyl-2- acetylglycerol induce Na+/HR exchange in human platelets. Biochem Biophys Res Commun 141: 13-19, 1986. 392. 393. 394. 395. 396. 397. 398. 399. 400. 401. 402. 301 Siffert W, Siffert G and Scheid P. Activation of Na+/H+ exchange in human platelets stimulated by thrombin and a phorbol ester. Biochem J 241: 301-303, 1987. Siffert W, Siffert G, Scheid P and Akkerman JWN. Activation of Na+/H+ exchange and Ca2+ mobilization start simultaneously in thrombin-stimulated platelets. Biochem J 258: 521-527, 1989. Siffert W, Siffert G, Schneid P and Akkerman JWN. Na+/H+ exchange modulates Ca2+ mobilization in human platelets stimulated by ADP and the thromboxane mimetic U46619. J Biol Chem 264: 719-725, 1990. Siffert W, Siffert G, Schneid P, Riemens T, Gorter G and Akkerman JWN. Inhibition of Na+/H+ exchange reduces Ca2+ mobilization without affecting the initial cleavage of phosphatidylinositol 4,5-biphosphate in thrombin-stimulated platelets. FEBS Letts 212: 123-126, 1987. Silk ST, Clejan S and Witkom K. Evidence of CTP-binding protein regulation of phospholipase A2 activity in isolated human platelet membranes. J Biol Chem 264: 21466-21469, 1989. Simon MF, Chap H and Douste-Blazy L. Selective inhibition of human platelet phospholipase A2 by buffering cytoplasmic calcium with the fluorescent indicator quin 2. Evidence for different calcium sensitivities of phospholipases A2 and C. Biochim Biophys Acta 875: 157-164, 1986. Simpson AWM and Rink TJ. Elevation of pHi is not an essential step in calcium mobilization in fura-2-loaded human platelets. FEBS Letts 222: 144-148, 1987. Smith CD, Cox CC and Snyderman R. Receptor-coupled activation of phosphoinositide-specific phospholipase C by an N protein. Science 232: 97-100, 1986. Staatz WD, Peters KJ and Santaro SA. Divalent cation- dependent structure in the platelet membrane glycoprotein Ia- IIa (VLA-2) complex. Biochem Biophys Res Commun 168: 107-113, 1990. Staatz WD, Rajpara SM, Wagner EA, Carter WC and Santaro SA. The membrane glycoprotein Ia-IIa (VLA-2) complex mediates the Mg++-dependent adhesion of platelets to collagen. J Cell Biol 108: 1917-1924, 1989. Steen VM, Cook CA, Tysnes OB and Holmsen H. Potentiation by adrenaline of thrombin-induced elevation of pHi is not essential for synergistic activation of human platelets. FEBS Letts 250: 211-214, 1989. 403. 404. 405. 406. 407. 408. 409. 410. 411. 412. 413. 302 Steen VM and Holmsen H. Current aspects on human platelet activation and responses. Eur J Haematol 38: 383-399, 1987. Steen VM, Tysnes OB and Holmsen H. Evidence for tight metabolic control of the receptor-activated polyphosphoinositide cycle in human platelets. Biochem J 263: 621-624, 1989. Steer ML and Wood A. Inhibitory effects of sodium and other monovalent cations on human platelet adenylate cyclase. J Biol Chem 256: 9990-9993, 1981. Steiner M and Ikeda Y. Quantitative assessment of polymerized and depolymerized platelet microtubules. Changes caused by aggregating agents. J Clin Invest 63: 443-448, 1979. Sturk A, Asyee GM, Schaap MCL, VanMaanen M and ten Cate JW. Synergistic effects of platelet-activating factor and other platelet agonists in human platelet aggregation and release: The role of ADP and products of cyclooxygenase pathway. Thromb Res 40: 359-373, 1985. Suzuki H, Kinlough-Rathbone RL, Packham MA, Mustard JF, Tanoue K and Yamazaki H. 'Localization of fibrinogen during ADP- induced aggregation of washed rabbit platelets. J Histochem Cytochem 38: 869-874, 1990. Sweatt JD, Blair IA, Cragoe EJ and Limbird LE. Inhibitors of Na+/H+ exchange block epinephrine- and ADP-induced stimulation of human platelet phospholipase C by blockade of arachidonic acid release at a prior step. J Biol Chem 261: 8660-8666, 1986. Sweatt JD, Connolly TM, Baron BM and Limbird LE. Involvement of Na+/H+ exchange in human platelet activation. In: Platelgt W221. W Jamieson GA (editor). Alan R Liss, Inc., New York, 1988. Pages 523-557. Sweatt JD, Connolly TM, Cragoe EJ and Limbird LE. Evidence that Na+/ H+ exchange regulates receptor-mediated phospholipase A2 activation in human platelets. J Biol Chem 261: 8667-8673, 1986. Sweatt JD, Johnson SL, Cragoe EJ and Limbird LE. Inhibitors of Na+/H+ exchange block stimulus-provoked arachidonic acid release in human platelets. J Biol Chem 260: 12910-12919, 1985. Sweatt JD, Schwartzberg MS, Frazer M, Cragoe EJ, Blair IA, Reed PW and Limbird LE. Evidence for a role for Na+-H+ exchange in activation of human platelets by PAF. Circulation Res 61: 116-1111, 1987. 414. 415. 416. 417. 418. 419. 420. 421. 422. 423. 303 Takai Y, Hishimoto A, Kikkawa U, Mori T and Nishizuka Y. Unsaturated diacylglycerol as a possible messenger for the activation of calcium-activated, phospholipid-dependent protein kinase system. Biochem Biophys Res Commun 91: 1218- 1224, 1979. Tandon NN, Kralisz U and Jamieson GA. Identification of glycoprotein IV (CD 36) as a primary receptor for platelet- collagen adhesion. J Biol Chem 264: 7576-7583, 1989. Tapley PM and Murray AW. Evidence that treatment of platelets with phorbol ester causes proteolytic activation of Ca+ activated, phospholipid- dependent protein kinase. Eur J Biochem 151: 419-423, 1985. Tapley PM and Murray AW. Modulation of Ca2+-activated, phospholipid-dependent protein kinase in platelets treated with a tumor-promoting phorbol ester. Biochem Biophy Res Commun 122: 158-164, 1984. Tarver AP, King WC and Rittenhouse SE. Inositol 1,4,5- triphosphate and inositol 1,2-cyclic 4,5-triphosphate are minor components of total mass of inositol triphosphate in thrombin-stimulated platelets. J Biol Chem 262: 17268-17271, 1987. Theirauch KH, Gabriel D, Prior G and Schillinger E. Desensitization of human platelets against stimulation by thromboxane mimics-reduction in receptor capacity and loss of functional response. Biomed Biochim Acta 47: 575-578, 1988. Thomas JA, Buchsbaum RN, Zimniak A and Racker E. Intracellular pH measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry 18: 2210-2218, 1979. Tohmatsu T, Hattori H, Nagao S, Ohki K and Nozawa Y; Reversal by protein kinase C inhibitor of suppressive actions of phorbol-12- -myristate-13- acetate on polyphosphoinositide metabolism and cytosolic Ca2+ mobilization in thrombin- stimulated human platelets. Biochem Biophys Res Commun 1134: 868-875, 1986. ' Tohmatsu T, Nakashima S, Hattori H, Suganuma A and Nozawa Y. A role of diacylglycerol kinase in stimulus- secretion coupling of human platelets dissociation of serotonin secretion from Ca2+ mobilization. Thromb Res 47:25-35,1987. Toledo T, Palamo FR and Hockaday TD. The NaI/H+ exchange is required for platelet aggregation induced by AVP (arginine- vasopressin). Thromb Res 55: 389-393,1989. 424. 425. 426. 427. 428. 429. 430. 431. 432. 433. 434. 435. 304 Touqui L, Rothhut B, Shaw AM, Fradin A, Vargaftig BB, Russo- Marie F. Platelet activation- a role for a 40k anti- phospholipase A2 protein indistinguishable from lipocortin. Nature 321: 177-180, 1986. Tsien RY, Rink TJ and Poenie M. Measurement of cytosolic free Ca2+ in individual small cells using fluorescence with dual excitation wavelengths. Cell Calcium 6: 145-157, 1985. Tsujinaka T, Ariyoshi H, Uemura Y, Sakon M, Kambayashi J and Mori T. Potential participation of calpain in platelet activation studied by use of a cell penetrating calapin inhibitor (calpeptin). Life Sci 46: 1059-1066, 1990. Tsukuda M, Asaoka Y, Sekiguchi K, Kikkawa U and Nishizuka Y. Properties of protein kinase C subspecies in human platelets. Biochem Biophys Res Commun 155: 1387-1395, 1988. Tsunoda Y, Matsuno K and Tashiro Y. Spatial distribution and temporal change of cytoplasmic free calcium in human platelets. Biochem Biophys Res Commun 156: 1152-1159, 1988. Tuszymski GP, Daniel JL and Stewart G. Association of proteins with the platelet cytoskeleton. Semin Hematol 22: 303, 1985. Tysnes OB, Steen VM, Frolich KW and Holmsen H. Evidence that chlorpromazine and prostaglandin E1 but not neomycin interfere with the inositol phospholipid metabolism in intact human platelets. FEBS Letts 264: 33-36, 1990. Tysnes OB, Verhoeven AJM, Aarbakke GM and Holmsen H. Phosphoinositide metabolism in resting and thrombin-stimulated human platelets. FEBS Letts 218: 68-72, 1987. Valone FH and Johnson B. Decay of the activating signal after platelet stimulation with 1-O-alky1-2-acet1y-sn-glycero-3- phosphorylcholine: Changes in calcium permeability. Throm Res 40: 385-393, 1985. Valone FH and Johnson B. Modulation of platelet-activating- factor-induced calcium influx and intracellular calcium release in platelets by phorbol esters. Biochem J 247: 669- 674, 1987. Vedelago HR and Mahadevappa VG. Mobilization of arachidonic acid in collagen-stimulated human platelets. Biochem J 256: 981-987, 1988. Verhoeven A, Horvli O and Holmsen H. Is poly-PI turnover increased upon platelet activation? Trends Biochem Sci 11: 67-68, 1986. ' 436. 437. 438. 439. 440. 441. 442. 443. 444. 445. 446. 305 Wallace RW, Tallant EA and McManus MC. Human platelet calmodulin-binding proteins: Identification and Ca2+-dependent proteolysis upon platelet activation. Biochemistry 26: 2766- 2733, 1987. Ware JA, Johnson PC, Smith M and Salzman EW. Aequorin detects increased cytoplasmic calcium in platelets stimulated with phorbol ester or diacylglycerol. Biochem Biophys Res Commun 133: 98-104, 1985. Ware JA, Johnson PC, Smith M and Salzman EW. Effect of common agonists on cytoplasmic ionized calcium concentration in platelets. Measurement with 2-methyl-6-methoxy 8- nitroquindine (Quin 2) and aequorin. J Clin Invest 77: 878- 886, 1986. Ware JA, Saitoh M, Smith M, Johnson PC and Salzman EW. Response of aequorin-loaded platelets to activators of protein kinase G. Am J Physiol 256: C35-C43, 1989. Ware JA, Smith M, Fossel ET and Salzman EW. Cytoplasmic Mg2+ concentration in platelets: Implications for determination of Ca2+ with aequorin. Am J Physiol 255: H855-H859, 1988. Watanabe M, Hagiwara M, Onoda K and Hidaka H. Monoclonal antibody recognition of two subtype forms of protein kinase C in human platelets. Biochem Biophys Res Commun 152: 642-648, 1988. Watson SP and Lapetina EG. 1,2-Diacylglycerol and phorbol ester inhibit agonist-induced formation of inositol phosphates in human platelets: Possible implications for negative feedback regulation of inositol phospholipid hydrolysis. Proc Natl Acad Sci USA 82: 2623-2626, 1985. Watson SP, McNally J, Shipman LJ and Godfrey PP. The action of the protein kinase C inhibitor, staurosporine, on human platelets. Evidence against a regulatory role for protein kinase C in the formation of inositol triphosphate by thrombin. Biochem J 249: 345-350, 1988. Watson SP, Wolf M and Lapetina EG. The formation of [3H] inositol phosphates in human platelets by palmitoyl lysophosphatidic acid is blocked by indomethacin. Biochem Biophys Res Commun 132: 555-562, 1985. Weiss HJ and Lages B. The responses of platelets to epinephrine in storage pool deficiency- Evidence pertaining to the role of adenosine diphosphate in mediating primary and secondary aggregation. Blood 72: 1717-1725, 1988. Wencel-Drake JD. Plasma membrane GPIIb/IIIa. Evidence for a cycling receptor pool. Am J Pathol 136: 61-70, 1990. 447. 448. 449. 450. 451. 452. 453. 454. 455. 456. 457. 458. 459. 306 Westlund P, Edenius C and Lindgren JA. Evidence for a novel pathway of leukotriene formation in human platelets. Biochim Biophys Acta 962: 105-115, 1989. Wheeler ME, Cox AC and Carroll RC. _Retention of the glycoprotein IIb-IIIa complex in the isolated platelet cytoskeleton. J Clin Invest 74: 1080-1089, 1984. White GC, Barton DW, White TE and Fischer TH. Cyclic AMP- dependent protein kinase does not increase calcium transport in platelet microsomes. Thromb Res 56: 575-581, 1989. White JG. Interaction of membrane systems in blood platelets. Am J Pathol 66: 295-312, 1972. White JG. Secretory pathway of bovine platelets. Blood 69: 878-885, 1987. White JG. Ultrastructural modifications in platelet membranes and cytoskeleton following activation. Blood Cells 9: 237- 261, 1983. White JG and Clawson CC. The surface connected canalicular system of blood platelets: A fenestrated membrane system. Am J Pathol 101: 353-364, 1980. White JG and Krumwiede M. Further studies of the secretory pathway in thrombin-stimulated human platelets. Blood 69: 1196-1203, 1987. White JG, Krumwiede M and Sauk JJ. Microtubule reassembley in surface activated platelets. Blood 65: 1494-1503, 1985. White JG, Rao GHR and Estensen RD. Investigation of the release reaction in platelets exposed to phorbol myristate acetate. Am J Pathol 75: 301-314, 1974. White JG, Rao GHR and Gerrard JM. Effects of the ionophore A23187 on blood platelets. Influence on activation and secretion. Am J Pathol 77: 135-150, 1974. White TE, Lacal JC, Reep B, Fischer TH, Lapetina EG and White GC. Thrombolamban, the 22-kDa platelet substrate of cyclic AMP-dependent protein kinase, is immunologically homologous with the Ras family of CTP-binding proteins. Proc Natl Acad Sci USA 87: 758-762, 1990. Williams KA and Haslam RJ. Effects of NaCl and CTP on the inhibition of platelet adenylate cyclase by 1-O-octadecy1-2- acetyl-sn-glyceryl-3-phosphorylcholine (synthetic platelet- activating factor). Biochim Biophys Acta 770: 216-223, 1984. 460. 461. 462. 463. 464. 465. 466. 467. 468. 469. 470. 307 Williams KA, Murphy W and Haslam RJ. Effects of activation of protein kinase C on the agonist-induced stimulation and inhibition of cyclic AMP formation in intact human platelets. Biochem J 243: 667-678, 1987. Williams WJ and Schneider AS. Examination of peripheral blood. In: Hgmatology, 23d Edition. Williams, Beutler, Erslev and Rundles (editors). McGraw-Hill Book Company, New York, 1977. Pages 10-25. Wilson DB, Connolly TM, Bross TE, Majerus PW, Sherman WR, Tyler AN, Rubin LJ and Brown JE. Isolation and characterization of the inositol cyclic phosphate products of phosphoinositide cleavage by phospholipase C. J Biol Chem 260: 13496-13501, 1985. Wilson DB, Neufeld EJ and Majerus PW. Phosphoinositide interconversion in thrombin-stimulated human platelets. J Biol Chem 260: 1046-1051, 1985. Winther K, Klyysner R, Geisler A and Anderson PH. Characterization of human platelet beta-adrenoreceptors. Thromb Res 40: 757-767, 1985. Wolf M, LeVine H, May WS, Cuatrecasas P and Sahyoun N. A model for intracellular translocation of protein kinase C involving synergism between Ca2+ and phorbol esters. Nature 317: 546-549, 1985. ' Woods VL, Wolff LE and Keller DM. Resting platelets contain a substantial centrally located pool of glycoprotein IIb-IIIa complex which may be acessible to some but not other extracellular proteins. J Biol Chem 261: 15242-15251, 1986. Yamaguchi A, Suzuki H, Tanoue K and Yamazaki H. Simple method of aequorin loading into platelets using dimethyl sulfoxide. Thromb Res 44: 165-174, 1986. Yamaguchi A, Tanoue K and Yamazaki H. Secondary signals mediated by GPIIb/IIIa in thrombin-activated platelets. Biochem Biophys Acta 1054: 8-13, 1990. Yamaguchi A, Yamamoto N, Kitagawa H, Tanoue K and Yamakaki H. Ca2+ influx mediated through the GPIIb/IIIa complex during platelet activation. FEBS Letts 225: 228-232, 1987. Yatomi Y, Higashihara M, Tanabe A, Ohashi T, Oka H and Kume S. Synergism between subthreshold concentrations of thrombin and 12-0-tetradecanoylphorbol 13-acetate in platelet activation. Biochem Biophys Res Commun 148: 1025-1029, 1987. 471. 472. 473. 474. 475. 476. 477. 478. 479. 480. 481. 308 Yoshida Y and Nachmias VT. Calcium sequestration in human platelets: Is it stimulated by protein kinase C. Cell Calcium 10: 299-307, 1989. Yoshida K and Nachmias VT. Phorbol ester stimulates calcium sequestration in saponized human platelets. J Biol Chem 262: 16048-16054, 1987. Yoshida K, Stark F and Nachmias VT. Comparison of the effects of phorbol 12-myristate l3-acetate and prostaglandin E1 on calcium regulation in human platelets. Biochem J 249: 487- 493, 1988. Zavoico GB and Cragoe EJ. Ca2+ mobilization can occur independent of acceleration of NaI/H+ exchange in thrombin- stimulated human platelets. J Biol Chem 263: 9635-9639, 1988. Zavoico GB, Cragoe EJ and Feinstein MB. Regulation of intracellular pH in human platelets. Effects of thrombin, A23187, and ionomycin and evidence for activation of Na+/H+ exhange and its inhibition by amiloride derivatives. J Biol Chem 261: 13160-13167, 1986. Zijenah LS, Morton LF and Barnes MJ. Platelet adhesion to collagen. Factors affecting Mg2+-dependent and bivalent- cation-independent adhesion. Biochem J 268: 481-486, 1990. Zschauer A, van Breeman C, Buhler FR and Nelson MT. Calcium channels in thrombin-activated human platelet membrane. Nature 334: 703-705, 1988. Zucker MB and Nachmias. Platelet activation. Arteriosclerosis 5: 2-18, 1985. Zucker MB, Troll W and Belman S. The tumor-promoter phorbol ester (12-0-tetradecanoyl-phorbol-13-acetate), a potent aggregating agent for blood platelets. J Cell Biol 60: 325- 336, 1974. Zucker-Franklin D. Endocytosis by human platelets: Metabolic and freeze fracture studies. J Cell Biol 91: 706-715, 1981. Zwaal R. Membrane and lipid involvement in blood coagulation. Biochim Biophys Acta 515: 163-205, 1978.