UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Surface chemical studies of human platelets Chiu, Basil 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1983_A1 C49_7.pdf [ 9.17MB ]
Metadata
JSON: 831-1.0095845.json
JSON-LD: 831-1.0095845-ld.json
RDF/XML (Pretty): 831-1.0095845-rdf.xml
RDF/JSON: 831-1.0095845-rdf.json
Turtle: 831-1.0095845-turtle.txt
N-Triples: 831-1.0095845-rdf-ntriples.txt
Original Record: 831-1.0095845-source.json
Full Text
831-1.0095845-fulltext.txt
Citation
831-1.0095845.ris

Full Text

SURFACE CHEMICAL STUDIES OF HUMAN PLATELETS by BASIL CHIU ,Sc, University of Wisconsin-Stevens Point, 1974 M.Sc, Medical College of Wisconsin, 1977 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES THE DEPARTMENT OF PATHOLOGY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA SEPTEMBER, 1983 0 Basil Chiu, 1983 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. B a s i l Chiu Department of PATHOLOGY  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 14 Oct. , 1983 ABSTRACT The purpose of this project is to investigate the surface properties of platelet discocytes, echinocytes and spherocytes. Normal "non-sticky" discoid shaped platelets (discocytes) can be transformed by ADP into irregularly shaped echinocytes which are "sticky" and aggregate easily in media containing Ca and fibrinogen. A model is examined here in which an echinocyte attains i t s "sticky" properties by evagination of a surface-connected canalicular system. Platelets also evaginate this canalicular system upon hypotonic shock, in which case the platelets swell up to form spherocytes. By comparing the properties of the different geometric forms of platelets insight into the nature of "stickiness" was sought. The sur-face areas of the discocyte and spherocyte measured microscopically were —8 2 found to be 16.4 and 36.7x10 cm respectively while that of the echinocyte —8 2 was estimated to be 23.7x10 cm using surface chemical analysis. Electron microscopic examination showed that the canalicular system may not be totally evaginated in the echinocyte. Although i t was found that the spherocyte could s t i l l be agglutinated passively by ristocetin i t had completely lost i t s a b i l i t y to aggregate. Microelectrophoretic studies I | j | revealed 8 and 6 fold increases in the density of Ca and Mg binding sites respectively on the echinocyte surface relative to the discocyte. The spherocyte on the other hand had lost most of i t s Ca binding sites. Electrokinetic analysis of l i v e , fixed and neuraminidase or alkaline phos-phatase treated platelets showed major differences in charge as well as amino, s i a l i c acid and phosphate group densities among the discocyte, echinocyte and spherocyte. The evaginated canalicular membrane surfaces of the latter two were also different. SDS-PAGE of platelets radiolabelled - i i i -via lactoperoxidase iodination, periodate-borohydride t r i t i a t i o n or neuraminidase/galactose oxidase-borohydride t r i t i a t i o n failed to show any difference in the gel patterns between the three platelet forms. No new glycoprotein species appeared during the transformations. The presence of fibrinogen interferes in a concentration related manner with lactoperoxidase iodination of GP-III on the echinocyte surface. An overall picture is presented here showing differences between the surface properties of plate-let discocytes, echinocytes and spherocytes. The accumulated evidence sug-gests that changes in the whole platelet surface occur during activation and the model of a cloistered "sticky" membrane may be an oversimplification. - iv -TABLE OF CONTENTS Page Chapter 1 GENERAL INTRODUCTION 1 Chapter 2 PLATELET AGGREGATION AND MORPHOLOGY 8 2.1 INTRODUCTION 9 2.2 MATERIALS AND METHODS 11 2.2.1 Collection of Blood 11 2.2.2 Preparation of Platelets 12 2.2.3 Preparation of Echinocytes and Spherocytes . . 12 2.2.4 Cocaine Induced Spheres 12 2.2.5 Enumeration of Platelets 12 2.2.6 Fixation of Platelets 12 2.2.7 Measurement of Platelet Sizes 12 2.2.8 Electron Microscopy 13 2.2.9 Tests for the Integrity of Platelet Forms 14 2.2.10 Platelet Aggregation 15 2.2.11 Lectin Studies 16 2.2.12 St a t i s t i c a l Methods 16 2.2.13 Materials 16 2.3 RESULTS 20 2.3.1 Platelet Dimensions and Morphology 20 2.3.2 Platelet Integrity and Release of Contents 24 2.3.3 Platelet Aggregation 41 2.4 DISCUSSION 67 2.4.1 Platelet Dimensions and Morphology . 67 2.4.2 Platelet Integrity and Release of Contents 68 2.4.3 Platelet Aggregations 70 2.4.4 Summary of Chapter 2 77 Chapter 3 MICROELECTROPHORESIS 78 3.1 INTRODUCTION 79 3.2 MATERIALS AND METHODS 86 3.2.1 Microelectrophoresis 86 3.2.2 Fixation of Platelets 87 3.2.3 Platelet Surface S i a l i c Acid 87 3.2.4 Platelet Surface Phosphate Groups 88 3.2.5 St a t i s t i c a l Methods 89 3.2.6 Materials 89 - V -Page 3.3 RESULTS 90 3.3.1 Platelet Electrophoretic Mobilities 90 3.3.2 Platelet Surface S i a l i c Acid 98 3.3.3 Platelet Surface Phosphate Groups 102 3.3.4 Cocaine Spheres 110 3.3.5 Calcium and Magnesium Ion Binding 110 3.4 DISCUSSION 121 3.4.1 Platelet Electrophoretic Mobilities 121 3.4.2 Platelet Surface S i a l i c Acid 125 3.4.3 Platelet Surface Phosphate Groups 128 3.4.4 Cocaine Spheres 129 3.4.5 Calcium and Magnesium Ion Binding 131 3.4.6 A Hypothesis of the Exposed Membranes 135 Chapter 4 RADIOCHEMICAL LABELLING 141 4.1 INTRODUCTION 142 4.2 MATERIALS AND METHODS 148 4.2.1 Surface Labelling of Platelets 148 4.2.2 Isolation of Membrane 151 4.2.3 Gel Electrophoresis 151 4.2.4 Surface Labelling in the Presence of Fibrinogen 152 4.2.5 Materials 153 4.3 RESULTS 154 4.3.1 Iodination Experiments 154 4.3.2 Trit i a t i o n Experiments 172 4.3.3 Surface Labelling in the Presence of Fibrinogen 177 4.4 DISCUSSION 195 4.4.1 Iodination Experiments 195 4.4.2 Tritiation Experiments 199 4.4.3 Surface Labelling in the Presence of Fibrinogen 201 4.4.4 Summary of Chapter 4 202 Chapter 5 SUMMARY AND CONCLUSION 203 REFERENCES 209 APPENDIX, 225 - v i -LIST OF TABLES Table Page 1 Physical Dimensions of Washed Platelets 23 2 Electrophoretic Mobilities of Platelets 91 3 Apparent Platelet Surface Charge 92 4 Platelet Surface Amino Groups 93 5 Platelet Surface S i a l i c Acid 101 6 Platelet Surface S i a l i c Acid . . . . 103 7 Platelet Surface S i a l i c Acid 104 8 Platelet Surface Phosphate Groups 107 9 Platelet Surface Phosphate Groups 108 10 Platelet Surface Phosphate Groups 109 11 S i a l i c Acid and Phosphate Groups of Cocaine Spheres I l l 12 Calcium Ion Binding Sites 118 13 Magnesium Ion Binding Sites 119 I | | | 14 S t a t i s t i c a l Comparison of Ca /Mg Binding Sites 120 15 Comparison of Calcium Ion Bindings 133 16 Platelet Surface Properties 136 (Table 16 continued) 137 (Table 16 continued) 138 17 Highlights of Differences Between Discocytes, Echinocytes and Spherocytes 208 - v i i -LIST OF FIGURES Figure Page 1 Example of a typical aggregation tracing 17 2 Phase photomicrographs of different forms of platelets . . . 22 3 Electron micrographs of platelet discocytes 26 (Fig. 3 continued) 28 4 Electron micrographs of platelet echinocytes 30 (Fig. 4 continued) 32 5 Electron micrographs of platelet spherocytes 34 (Fig. 5 continued) 36 6 Electron micrographs of cocaine spheres 38 (Fig. 6 continued) 40 7 LDH leakage from spherocytes 43 8 Effect of ADP concentration on aggregation velocity . . . . 46 9 Example of aggregation tracings showing effect of ADP concentration 48 10 Effect of fibrinogen concentration on aggregation velocity . 50 11 Effect of divalent cation concentration on aggregation velocity 52 12 Examples of aggregation tracings showing the effects of divalent cations 54 I | 13 Inhibitory effect of Mn on platelet aggregation 57 14 Agglutination of platelet spherocytes and discocytes by ristocetin 59 15 Aggregation tracings from normal and neuraminidase treated platelets 61 16 Aggregation responses to lectins 64 17 Examples of aggregation responses to lectins 66 18 The pH-electrophoretic mobility profiles of fixed platelets 95 19 The pH-electrophoretic mobility profiles of fresh platelets 97 20 The time-release curve of s i a l i c acid 100 - v i i i -Figure Page 21 The time-release curve of phosphate groups 106 I [ j j 22 Ca and Mg concentration on electrophoretic mobility . . . 113 I [ | | 23 Double reciprocal plots to find Ca and Mg binding densities 115 (Fig. 23 continued) 117 24 SDS-PAGE of 1 2 5 I - l a b e l l e d platelets 156 1 ? 5 25 SDS-PAGE of I-labelled membranes 158 125 131 26 SDS-PAGE of a mixture of I-labelled discocyte and I-labelled discocyte 161 27 SDS-PAGE of a mixture of 1 2 5 I - l a b e l l e d discocyte and 1 3 1 I -labelled echinocyte 163 125 131 28 SDS-PAGE of a mixture of I-labelled discocyte and I-labelled spherocyte 165 17 5 131 29 SDS-PAGE of If I double labelled discocyte 167 30 SDS-PAGE of 1 2 5 I / 1 3 1 I double labelled echinocyte 169 1 ? 5 1 31 31 SDS-PAGE of 1/ I double labelled spherocyte 171 32 SDS-PAGE of tr i t i a t e d platelets (periodate) 174* 33 SDS-PAGE of tr i t i a t e d platelets (neuraminidase/galactose oxidase 176" 34 SDS-PAGE of pseudo-double labelled platelets (periodate) . . . 17:9 35 SDS-PAGE of pseudo-double labelled platelets (neuraminidase/ galactose oxidase) 181 36 SDS-PAGE of discocytes iodinated in the presence of fibrinogen 183 37 SDS-PAGE of echinocytes iodinated in the presence of fibrinogen 185 38 SDS-PAGE of spherocytes iodinated in the presence of fibrinogen 187 39 SDS-PAGE of echinocyte iodinated in the presence of albumin . 189 40 SDS-PAGE of echinocyte iodinated in the presence of fibronectin 191 41 Effect of fibrinogen concentration on the iodination of echinocytes 194 - ix -LIST OF ABBREVIATIONS ADP Adenosine diphosphate Beta-Tg Beta-thromboglobulin CPM Counts per minute DTT Dithiothreitol EPM Electrophoretic mobility GP Glycoprotein JBA Jequirity bean agglutinin LDH Lactate dehydrogenase MW Molecular weight NAD Nicotinamide adenine dinucleotide PAGE Polyacrylamide gel electrophoresis PRP Platelet rich plasma RBC Red blood c e l l R C A 6 0 Castor bean agglutinin, MW=60,000 RCA 1 2 ( ) Castor bean agglutinin, MW=120,000 SDS Sodium dodecyl sulfate TEMED Tetramethylethylenediamine WGA Wheat germ agglutinin - X -ACKNOWLEDGEMENTS I wish to express my deep and sincere gratitude to Dr. D.E. Brooks for his constant guidance and support throughout t h i s project. I also wish to thank Drs. J.G. Foulks, R.H. Pearce and P.E. Reid, members of my Super-visory Committee, for t h e i r advice and encouragement. I would l i k e to thank the l a t e Dr. E. Anderson as well as Mr. J . Cavanagh, Mr. C. Ramey and Ms. R. Rupps for t h e i r expert t e c h n i c a l a s s i s t -ance. I am also g r a t e f u l to fellow graduate students, e s p e c i a l l y J . Janzen and R. Snoek, for t h e i r h e l p f u l advice. The author i s f i n a n c i a l l y supported by Medical Research Council grant MT-5759. SURFACE CHEMICAL STUDIES OF HUMAN PLATELETS - 1 -CHAPTER 1 GENERAL INTRODUCTION - 2 -Platelets are circulating formed elements of mammalian blood that are essential for normal hemostasis. Their functions can generally be divided into two groups. The f i r s t is hemostatic, i.e. the physical occlusion of damaged blood vessels by masses of aggregated platelets. The second function is thromboplastic, i.e. the participation of the chemical constituents of the platelets in the blood coagulation mechanism. When the blood vessel wall is transected, platelets come into contact with the basement membrane as well as collagen in the subendothelial tissues around the cut margin. They then rapidly undergo shape change and degranulation. Platelets are normally discoid in shape (a form referred to as a discocyte) but during the shape change reaction they become irregular in shape with pseudopods extending out from a more or less spherical central body (echinocyte). Adhesion of the platelet to the injured vessel wall is f i r s t initiated with the pseudopods and then the platelets spread out over the exposed subendothelial surfaces (Mohammad and Mason, 1981; Frojmovic and Milton, 1982). During this process platelet degranulation occurs and substances released such as adenosine diphosphate (ADP) and serotonin help stimulate other platelets to undergo shape change and adhere to the f i r s t layer of platelets. This "second wave" of adherent platelets also release their granules and causes more platelets to adhere and accumulate around the injured site u n t i l a hemostatic platelet plug is formed, stopping further loss of blood through the injured vessel (Zucker, 1972). By 30 sec after injury the hemostatic plug w i l l be well formed and can grow to several times the diameter of the blood vessel. In the central part of the plug, activated platelets that have lost most of their granules form a cytoplasmic mass in which individual platelets are no longer distinguishable (Wester ^ t al., 1977). - 3 -During platelet activation factor 3 becomes available for i t s thrombo-plastic function. Platelet factor 3 is a phospholipid associated with the platelet membrane which is not detected until activation occurs. Platelet factor 3 forms a complex with factor IXa, factor VIII and calcium to activate factor X, following which i t forms a complex with factor Xa, factor V and calcium to convert prothrombin into thrombin (Zwaal, 1978). Other substances released from platelet granules during activation such as platelet factor 4 (anti-heparin activity), platelet fibrinogen and factor VIII may also contribute to the coagulation cascade. The time course of platelet-induced clotting is such that f i b r i n is not observed in the hemostatic plug until about 3 min after the i n i t i a l activating event. However by 30 min most of the platelets in the plug have disintegrated and have been replaced by a massive accumulation of f i b r i n (Wester et a l . , 1977). The introduction of the aggregometer allowed platelet aggregation to be studied in vitro in detail (Born 1962; Michal and Born, 1971). An aggrego-meter is a simple photometric device that measures the turbidity of a stirred suspension of platelets. A beam of light shines through the platelet sus-pension and the amount of light passing through is measured. Normal discoid platelets have the maximum turbidity and the least amount of light passes through. During aggregation the platelets clump together into aggregates and the suspension clears optically, allowing more light to pass through. A measure of the rate of aggregation is therefore provided by the rate of increase in transmission of light through the platelet suspension. Among the agents which can cause aggregation, the most well studied one is ADP. Upon exposure to ADP, the platelets undergo a shape change from discocyte to echinocyte. Providing the right conditions exist, the platelets - 4 -w i l l aggregate. Right conditions include the presence of the divalent cations calcium or magnesium, and fibrinogen (Zucker, 1972). A certain amount of agitation i s also required to bring sufficient numbers of platelets into contact with each other. The aggregometer has a b u i l t - i n magnetic stirrer for this purpose. Shape change alone does not require any of these conditions. The f i r s t event in the stimulation of platelets by ADP is the binding of this molecule onto the surface of the platelets (Born and Feinberg, 1975; Legrand and Caen, 1978). A yet-to-be understood sequence of events then t r i g -gers the discocyte-echinocyte transformation. At the same time, the surface of the platelet becomes "sticky." The development of platelet "stickiness" is a process which parallels shape change but not directly caused by the discocyte to echinocyte transformation i t s e l f (Zucker, 1972; Barnhart, 1978). The biochemical nature of "stickiness" is s t i l l uncertain and remains to be defined. It may involve the increased, saturable binding of fibrinogen onto the ADP-stimulated platelet surface in the presence of Ca + + or Mg (Mustard et a l . , 1978; Marguerie et a l . , 1979; Bennett and Vilaire, 1979; Peerschke et a l . , 1980). Surface properties associated with changes in electrophoretic mobility (Hampton and Mitchell, 1974; Seaman and Vassar, 1977; Stoltz, 1979) or in the amount of neuraminidase-susceptible s i a l i c acid on the platelet surface (Motemed et a_l., 1976; Ku and Wu, 1977) have also been suggested as being involved. Throughout the following sections, these and other properties that may be involved in platelet "stickiness" w i l l be explored. After echinocyte formation and the appearance of "stickiness", the platelets begin a two stage aggregation process. First is primary or reversible aggre-gation in which the platelets adhere to each other to form aggregates. Some - 5 -i n t e r n a l r e o r g a n i z a t i o n i n the p l a t e l e t s occurs: o rgane l les move towards the c e n t r a l part of the ech inocyte and the c i r c u m f e r e n t i a l bundle of microtubules a l s o s h i f t s i n t e r n a l l y . The degree of change va r i e s from ech inocyte t o ech inocyte but no granules are r e l ea sed at t h i s point (White, 1972). When the concen t ra t i on of the aggregat ion agent i s not high enough or the temperature i s below 30°C , the p l a t e l e t s w i l l e ven tua l l y d i saggregate and r e v e r t t o the d i s c o i d shape (Zucker , 1972). On the other hand, i f the concen t ra t i on of the aggregat ing agent i s high enough and the temperature i s i d e a l , then secondary or i r r e v e r s i b l e aggregat ion proceeds a f t e r pr imary aggregat ion (White, 1972). At l o c a t i o n s i n the aggregates where the p l a t e l e t s are t i g h t l y squeezed t oge the r , t h e i r organel les and m i c r o tubu l e bundles beome more c e n t r a l i z e d and f i n a l l y contents of the p l a t e l e t granules are r e l ea sed . The p l a t e l e t s i n the aggregates subsequent ly beg in t o d i s i n t e g r a t e (Rodman, 1971; White, 1972). the r e l e a s e r e a c t i o n induced by agents such as thrombin, a rach idon i c a c i d and A23187 can be s t ud i ed wi thout aggregat ion by not s t i r r i n g the p l a t e l e t suspens ion t o promote aggregat ion (Holmsen, 1977). Aggregat ion induced by ADP i s u s u a l l y not f o l l owed by r e l e a s e a c t i o n (Holmsen, 1977). P l a t e l e t s have three kinds of g ranu les , alpha g ranu les , dense bodies and lysoscmes (Skaer, 1981; Nurden et a l . , 1982). Dense bodies are few i n number and e l e c t r on dense. They con ta in l a r ge quan t i t i e s of s e r o t o n i n , i n o r g a n i c ca lc ium and phosphorus as w e l l as ADP and ATP. The alpha granules are l e s s e l e c t r o n opaque and conta in numerous p ro te i n s i n c l u d i n g f i b r i n o g e n , f i b r o n e c t i n , thrombospondin, f a c t o r V H I - r e l a t e d a n t i g e n , f a c t o r V, low a f f i n i t y and h igh a f f i n i t y ( a f f i n i t y f o r hepar in ) p l a t e l e t f a c t o r 4, be t a -th rombog lobu l in , p l a t e l e t growth f a c t o r , chemotact ic f a c t o r and b a c t e r i c i d a l - 6 -factor. Platelet lysosomes which are also not electron dense contain various proteinases, glycosidases , esterases and phosphatases. Serotonin is usually used as a monitor for dense granule release while bet a-thrombo globulin and platelet factor 4 levels reflect alpha granule release. Transformation from a discocyte to an echinocyte requires the platelets to increase their apparent surface areas to provide for the formation of the pseudopods . The shape change occurs almost instantaneously after exposure to an aggregating agent and is completed within seconds. The discocyte must therefore have a mechanism to provide for the increase during this short time period. The most l i k e l y mechanism seems to be the provision of excess surface area within the discocyte. The surf ace-connected canalicular system (open canalicular system) is a l i k e l y source. This open canalicular system is a labyrinth of membrane channels that are connected to the platelet surface with openings to the outside (White and Clawson, 1980). The evagination of this membrane system could therefore provide the additional surface membrane required for transformation (Morgenstern andKho, 1977 ; Frojmovic and Milton, 1982). A comparison of the surface properties of platelet discocytes and echinocytes may therefore provide some information on the properties of "sticky" membranes and/or the open canalicular membrane system. The degree to which the two are related remains to be seen. Another way to study the invaginated open canalicular system i s to evaginate i t by means of hypotopnic stress. By lowering the osmotic pressure of the suspending medium, large swollen spherical platelets (spherocytes) are creatred with their surface-connected canalicular system apparently evaginated (Milton and Frojmovic, 197 9) . Treatment of the discocyte with cocaine w i l l also produce a spherical form of platelet (Behnke, 1979) but one which i s much - 7 -sma l le r i n s i z e than the spherocyte . Hereaf ter these w i l l be r e f e r r e d t o as coca ine spheres to d i s t i n g u i s h them from the spherocytes produced by hypoton ic s t r e s s . The general term spheres w i l l i nc lude both spherocytes and cocaine s pher es . The p ro jec t desc r ibed h e r e i n was aimed at comparing the su r face featu res of the d i s c o c y t e , e ch i nocy te , spherocyte and cocaine sphere. P r o p e r t i e s s t ud i ed i n c l uded e l c t r o k i n e t i c behav iour, su r face s i a l i c a c i d , phosphate and amino groups, ca lc ium and magnesium b i nd i ng and s u r f a ce l a b e l l i n g . Aggregat ion w i t h var ious agents was a l so s t u d i e d . CHAPTER 2 PLATELET AGGREGATION AND MORPHOLOGY - 9 -2.1 INTRODUCTION Mild hypotonic shock to platelets has been used in the blood banking com-munity to test for platelet v i a b i l i t y during storage. Platelets are very resistant to hypotonic stress. When introduced into a hypotonic environment with osmolarity as low as 150 mOsm, the platelets w i l l change into an echinoid shape but can slowly reverse back into the discoid form (Fantl, 1966; Handin eit a l . , 1970; Kim and Baldini, 1974; Odink, 1976; Milton and Frojmovic, 1977a). Under more severe hypotonic stress platelets w i l l turn into swollen spheres or spherocytes. Zucker-Franklin (1969) suspended platelets in d i s t i l l e d water to obtain spherocytes for microfibril studies. Milton and Frojmovic (1977b) found that by lowering the osmolarity to as much as 60 mOsm they could produce spherocytes stable for at least an hour. They also suggested that the osmotic spherocyte formation provides an opportunity for the study of the surface-connected canalicular system. Treatment of platelets with cocaine results in a total loss of aggreg-abil i t y (Aledort and Niemetz, 1968; O'Brien, 1976) and produces a spherical form of platelet (Mannucci and Sharp, 1967; Behnke, 1970). Other properties of this cocaine-induced sphere have yet to be explored. In this chapter, the morphology and dimensions of the normal platelet discocyte, the ADP transformed echinocyte, the hypotonically induced spherocyte and the cocaine-induced sphere w i l l be described using the results from phase and electron microscopy. ++ As discussed earlier, aggregation of platelets by ADP requires Ca or Mg + + ions as well as fibrinogen. Aggregation studies to find the optimum ADP, Ca + +, Mg + + and fibrinogen concentrations required are reported in this chapter as preparation for later experiments. - 10 -Aggregation induced by a variety of other agents was also examined, mainly to test the aggregability of the spherocyte. Of these, ristocetin is of particular interest because i t passively agglutinates rather than pharmaco-logically inducing platelet aggregation, and a plasma co-factor (von Willebrand's factor) is also required (Phillips, 1980). Another group of agents, the lectins, were also of special interest. It w i l l be shown here that removal of surface s i a l i c acid with neuraminidase can modify the plate-let's response to different lectins. - 11 -2.2 MATERIALS AND METHODS 2.2.1 Collection of Blood Blood was drawn from healthy human volunteers using plastic syringes by venipuncture and anticoagulated with 0.38% sodium citrate. Volunteers were mostly from within the Pathology Department but i n a few cases arrangements were made through the local Red Cross. Red Cross blood was collected i n citrate-phosphate-dextrose (Masouredis, 1972). Plastic ware was used i n a l l the handling of blood and platelets. Samples remained at room temperature at a l l times . 2.2.2 Preparation of Platelets Within one hour after the collection of blood, platelet-rich plasma (PRP) was prepared by centrifugation at 120xg for 15 min (Day et a l . , 1976). Platelets were then isolated from the PRP by centrifugation at 1200xg for 15 min. They were washed twice in a calcium and magnesium free Tyrode's solution made up of 136.75 mM sodium chloride, 2.68 mM potassium chloride, 11.90 mM sodium bicarbonate, 0.36 mM sodium dihydrogen phosphate and 5.55 mM glucose. The pH was 7.4 unless otherwise indicated. To prevent platelet loss during the isolation and washing procedures, the PRP and Tyrode's solution were f i r s t acidified with c i t r i c acid to pH 6.5 according to the method of Zucker and Grant (1978). After wshing, the platelets were resuspended in pH 7.4 Tyrode's solution. It was found that washing with the acidified Tyrode's solution was essential. If platelets are centrifuged in a pH 7.4 solution they will aggregate. Fibrinogen as well as calcium and magnesium ions were added separately as required on an individual experiment basis. - 12 -2.2.3 Preparation of Echinoctyes and Spherocytes Echinocytes were prepared by the addition of 2x10 fi n a l concentration of ADP to suspensions of platelets at room temperature. The suspensions were then gently mixed by inverting the test tube twice. Vigorous mixing or shaking was avoided. Spherocytes were prepared by the addition of water to isotonic suspensions of platelets (in PRP or Tyrode's solution) in a ratio of three parts water to one part suspension (Milton and Frojmovic, 1979). Water was added slowly through the period of about a minute and vigorous mixing was also avoided. The osmolality of the final suspension was therefore lowered to about 75 mOsmol. 2.2.4 Cocaine Induced Spheres Platelet suspensions were incubated for half an hour with cocaine at 10 mM final concentration (Behnke, 1970) to produce the cocaine spheres. Sphering occurred at both room temperature and 37°C. 2.2.5 Enumeration of Platelets Platelets were counted using a hemocytometer under phase contrast microscopy. 2.2.6 Fixation of Platelets Platelets in various forms were fixed at room temperature in 0.35% glutar-aldehyde in Tyrode's solution. After one hour the platelets were spun down and resuspended with fresh glutaraldehyde solution for overnight fixation at 4°C. Fixation with glutaraldehyde was found to be very rapid (Vassar e_t ajL., 1972). 2.2.7 Measurement of Platelet Sizes The morphology of fresh and fixed platelets was examined under phase contrast microscopy. Photomicrographs were taken with a Carl Zeiss Photo-- 13 -microscope II using a 40x objective and overall magnification of 800x. A grid of O.l^jLwas also photographed in the same manner. To measure the sizes of the platelets, the negatives from the photomicrographs were projected onto a piece of paper using a photographic enlarger. Diameters measured from the outside edges and the inside edges of the diffraction ring of each platelet were averaged as described by -Frojmovic and Panjwani (1976). They were compared to that of the grid traced out in the same way. Dimensions for the spherical forms of platelets were calculated from their measured radii (r) using simple geometric formulae for spheres: Area = 4t<r2 3 and Volume = 4/3-"rr Dimensions for the discocytes were calculated according to Frojmovic and Panjwani (1976) using a model of an oblate spheroid: Area = (^/2)d2 + (n/4)t 2{(l+r)/(l-r)} log R _ 1 2 and Volume = (fl/6)d t where d = diameter t = thickness and R = t/d 2.2.8 Electron microscopy Glutaraldehyde fixed platelets were postfixed with 2% osmium tetroxide in 0.1M cacodylate buffer (pH 7.35) for 1 hour. After dehydration in a graded series of ethanol solutions the platelets were embedded in Epon 812. Thin sections were cut with an ultramicrotome equipped with a diamond knife and contrast of the sections enhanced by staining with uranyl acetate and lead citrate. Platelet samples from three healthy volunteers were viewed with a Phillips EM 300 electron microscope and photographs taken. - 14 -2.2.9 Tests for the Integrity of Platelet Forms 2.2.9.1 Lactate Dehydrogenase (LDH) Leakage LDH assays were performed according to the kinetic method of Kachmar (1970). This method is based on the reverse reaction of pyruvate (substrate) to lactate and monitored as a decrease in absorbance at 340 nm when NADH is oxidized to NAD. LDH Pyruvate + NADH >NAD + lactate The unit of enzyme activity is the Wroblewski Unit which is the drop of absorbance per minute per volume of sample. Supernatants from discocyte, echinocyte and spherocyte suspensions were assayed for the leakage of the enzyme. Total LDH in platelets was assayed using supernates from frozen and thawed samples. More detailed time studies for spherocytes were also performed. 2.2.9.2 Beta-Thromboglobulin (beta-Tg) Release Beta-thromboglobulin was assayed using an Amersham radioimmunoassay k i t . Release of beta-Tg from echinocytes and spherocytes was monitored by comparing supernatant and platelet contents released by freeze-thawing. Some discocytes were also stimulated to release their granules by the addition of thrombin at 0.4 NIH units/ml. 14 2.2.9.3 C-Serotonin Release 14 Platelet (discocyte) dense bodies were f i r s t loaded with C-serotonin by 14 the method of Clark and Harms (1978). About 25jj.Ci of C-serotonin were incubated with each m i l l i l i t e r of PRP for one hour at room temperature. The platelets were then washed as before and resuspended in Tyrode's solution. 14 . . After transformation to echinocytes or spherocytes C-serotonin in the - 15 -supernatants and p l a t e l e t s were moni tored us ing l i q u i d s c i n t i l l a t i o n count ing. Samples were mixed w i th A toml i gh t s c i n t i l l a n t f l u i d (New England Nuc l e a r , Boston) i n a r a t i o of 1 ml sample t o 5 ml Atoml ight and then counted w i th a Beckman LS-233 s c i n t i l l a t i o n counter. 2.2.10 P l a t e l e t Aggregat ion P l a t e l e t aggregat ion was s t ud i ed w i t h a Born-Michal Aggregometer (London, England). The v e l o c i t y of aggregat ion was measured as the r a t e of decrease i n the o p t i c a l den s i t y of the p l a t e l e t suspension (Born and C ro s s , 1963). Th is was done by drawing a l i n e tangent t o the steepest part of the s lope of the aggregat ion t r a c i n g recorded by the chart recorder ( F i g . 1.). The s lope of t h i s l i n e was measured as the v e l o c i t y of aggregat ion and had an a r b i t r a r y u n i t of inches (of chart paper) per minute. A l l aggregat ion experiments were o 8 done at 37 C and at a concen t ra t i on of about 2.5x10 p l a t e l e t s / m l of Ty rode ' s s o l u t i o n . A se r ie s of experiments was undertaken t o determine the e f f e c t s of d i f f e r e n t concentrat ions of ADP and f i b r i n o g e n as w e l l as ca lc ium and magnesium ions on p l a t e l e t aggregat ion. In order to make i n t e r - s amp le comparisons po s s i b l e a no rma l i z a t i on procedure s i m i l a r t o that used by F ro jmov ic (1973) was adopted. A concen t ra t i on at which maximum aggregat ion v e l o c i t y always occur red was r e f e r r e d to as producing a 100% v e l o c i t y . V e l o c i t i e s at other concent rat ions s t u d i e d were then expressed as percentages of t h i s maximum v e l o c i t y . I n t h i s way, percentage vs concent ra t i on curves from d i f f e r e n t samples cou ld be combined. ADP concen t ra t i on s tud ies were performed with p l a t e l e t s i n Ty rode ' s s o l u t i o n con ta i n i n g 0.5 mg/ml of f i b r i n o g e n and 4 mM CaCl„ _., . 0 0 <* 2. F ib r i nogen - 16 -concentration studies were done using platelets i n Tyrode's solution containing 4 mM Ca + + and then 2x10 ^M (final concentration) of ADP added as aggregating agent. Ca + and Mg + + concentration studies were performed using platelets i n Tyrode's solution containing 0.5 mg/ml of fibrinogen. ADP added was also 2x10 "*M. Mn + + and S r + + were also investigated for their a b i l i t i e s to support aggregation induced by ADP. A panel of aggregating agents including ADP, thrombin, epinephrine, A23187, arachidonic acid and ristocetin was used to survey the aggregability of the spherocyte and cocaine sphere. 2.2.11 Lectin Studies Lectins and ADP were used to study the effect of neuraminidase treatment g on the aggregation of platelets. Platelets (2.5x10 /ml) were incubated with 0.04 IU/ml of neuraminidase (Vibrio cholerae) for 90 min at 37° in Tyrode's solution (see Chapter 3 for more information). The l i s t of lectins used i n these experiments includes wheat germ agglutinin (WGA) , jeguirity bean agglutinin (JBA) and the two castor bean agglutinins (RCA^Q a n d R C A ^ Q ) -Sugars N-acetyl-D-glucosamine, N-acetyl-D-galactosamine and D-galactose were used for blocking experiments. 2.2.12 S t a t i s t i c a l Methods A l l s t a t i s t i c a l methods were according to Kalbfleisch (1974). 2.2.13 Materials Fibrinogen was prepared and kindly supplied by Mr. Johan Janzen of our laboratory by precipitation with polyethylene glycol 6000 (Janzen, 1983). The fibrinogen was further purified by a f f i n i t y chromatography using lysine-sepharose 4B and collagen-sepaharose 4B to remove plasminogen and fibronectin - 17 -Fig. 1. Example of a typical aggregation tracing. Aggregation velocity was found by drawing a line tangent to the steepest part of the tracing. The slope of this line was taken as the aggregation velocity. ADP is introduced into the platelet suspension as indicated by arrow. A rapid shape change phase occurred before aggregation. - 18 -- 19 -respectively. SDS-PAGE demonstrated no detectable contaminant and the preparation was considered to be 99% pure. ADP, arachidonic acid, epinephrine, ristocetin sulfate, WGA, JBA, D-galactose, N-acetyl-D-glucosamine, N-acetylgalactosamine, NADH and sodium phruvate were a l l from Sigma (St. Louis, Missouri). 14 The beta-throm bo globulin RIA k i t and C-serotonin were from Amersham (Arlington Heights, I l l i n o i s ) . The neuraminidase, both castor bean agglutinins and A23187 were from Calbiochem (La J o l l a , California). Thrombin was from Park, Davis and Co. of Detroit, Michigan. Cocaine hydrochloride was purchased through the UBC Department of Family Practice pharmacy from Allen and Hanburys of Toronto, Ontario. Arachidonic acid and A23187 were dissolved i n ethanol. A l l others were dissolved i n isotonic saline (pH 7.4) solution. Appropriate ethanol controls were included i n a l l relevant experiments. Calcium, magnesium, strontium and manganese were added as the chloride. - 20 -2.3 RESULTS 2.3.1 Platelet Dimensions and Morphology Under phase microscopy, platelet discocytes could be seen to tumble through the f i e l d of view demonstrating their discoid shape. No differences were observed i n the morphology of discocytes in pi at el et-ri ch plasma or i n Tyrode's solution. Therefore the washing procedure proved to be quite sati sfactory. Addition of 2x10 "*M of ADP to the platelet suspension produced the echinocytes (Fig. 2). They appeared more "stationary" than the discocytes. Careful observations by focusing the microscope up and down showed long slender pseudopods which were extremely d i f f i c u l t to photograph. Echinocytes appeared as single platelets i n suspension. Small aggregates would form i f the suspension was shaken or stirred even i n the absence of Ca + + and fibrinogen. Hypotonic shock at 75 mOsm produced the spherocytes which were much larger than the spheres induced by treatment with 10 mM of cocaine (Fig. 2). Table 1 shows the measured diameters for the discocyte, spherocyte and -4 cocain sphere. The discocyte has a mean diameter of 3.18x10 cm and a -4 thickness of 0.89x10 cm. The thickness to diameter ratio i s therefore —8 2 0.28. Its surface area and volume were calculated to be 16.48x10 cm and 4.71x10 cm respectively. Since the discocyte tumbles through the f i e l d of view, only platelet images considered as presenting the edge-on view were measured for thickness. Discocytes i n PRP had similar dimensions. Table 1 also shows the spherocyte has a 123% and a 345% increase i n surface area and volume respectively over the discocyte. The cocaine sphere i s much smaller than the spherocyte and i t s increase in volume and area over the discocyte is not as large. No differences were found between fixed and non-fixed - 21 -Fig. 2. Phase photomicrographs of different forms of platelets. (a) Discocytes (xl200 magnification*) (b) Echinocytes (xl400 magnification) (c) Spherocytes (xlOOO magnification) (d) Cocaine spheres (xl200 magnification) *Magnification here and in a l l following figures depicts magnification of the platelets as they appear on the photographic prints. - 23 -Table 1. Physical Dimensions of Washed Platelets* Diameter"1" (cm) Surface Area (cm2) Volume (cm3) Discocyte** 3.18X10-4 (+0.03xl0 _ 4) + + 16.48xl0-8 (+0. 35xl0"8) 4.71xl0 - 1 2 (+0.14xl0-2) Spherocyte 3. 42x10 - 4 (+0. 017X10-4) 36. 7xl0 - 8 (+0.36xl0-8) 20. 9 4 x l 0 - 1 2 (+0.31xl0 - 1 2) Cocaine spheres 2. 90X10-4 (+0.025X10-4) 26. 38xl0 - 8 (+0.45xl0-8) 12. 77xl0 1 2 (+0. 33x10"12) *Measured diameter of discoycte i n PRP=3. 17xl0"4cm (+0 . 035X10-4) +1000 platelets measured in each category **Thickness for di scocyte=0. 89X10-4 cm (+0.06X10-4 cm; n=130) Standard deviation - 24 -platelets. Addition of ADP (2x10 M^) to the spherocytes and cocaine spheres produced no further change. Transmission electronmicrographs of platelet discocytes, echinocytes, spherocytes and the cocaine spheres are shown in Figs. 3 to 6. At least 20 fields of view for each type of platelet were examined. A l l the platelets appeared to have retained their complement of granules. The discocytes (Fig. 3) ranged in appearance from discs (top view) to cigar shapes (edge-on view). The echinocytes (Fig. 4) were irregular in shape with long pseudopods. The spherocytes (Fig. 5) were large and swollen but the plasma membranes were intact. The spherocyte cytoplasm was less dense and had an "empty" appear-ance. There were large vacuoles apparent besides the storage granules. Because of the "empty" appearance of the cytoplasm the plasma membrane as well as the granules and vacuoles were highlighted better than in the discocytes and echinocytes. The cocaine spheres (Fig. 6) were round but smaller than the spherocyte. The surface-connected canalicular system was obvious in the discocyte. Numerous openings to the outside could be seen. The canalicular system in the echinocytes was dilated. In the cocaine spheres dilation of the canalicular channels was even more obvious. No opening connecting the dilated channels of the echinocyte and cocaine spheres to the outside could be seen. 2.3.2 Platelet Integrity and Release of Contents The supernatants of washed discocytes showed no sign (zero) of lactate dehydrogenase (LDH) activity. After the platelet suspension had been frozen and thawed once to damage the platelets, LDH activity was detected in the supernatant at 1.047 Wroblewski unit per ml per 10 platelets (+0.350 s.d.; n=3). This is considered as the total platelet LDH activity. After - 25 -F i g . 3. E l e c t r o n micrographs of p l a t e l e t discocytes. (a) x33600 magnification (b) x22000 magnification - 27 -Fig. 3 (continued) (c) schematic representation of (a) (d) schematic representation of (b) Two prominent features are the channels of the surface-connected canalicular system (CS) with openings on the platelet surface (arrows) and the alpha granules (AG). Microtubules (MT) from the circumplatelet bundles can be seen. Also visible are glycogen deposits (GY) as well as an occasional mitrochondrion (MC). - 29 -F i g . 4. E l e c t r o n micrographs of p l a t e l e t echinocytes. (a) xl6900 magnification (b) x27600 magnification - 31 -F i g . 4 (continued) (c) schematic representation of (a) (d) schematic representation of (d) The echinocytes are i r r e g u l a r i n shape with long pseudopods. The surface-connected c a n a l i c u l a r system (CS) i s d i l a t e d . The granules (AG) as well as the glycogen deposits (GY) can s t i l l be seen. - 33 -Fig. 5. Electron micrographs of platelet spherocytes. (a) x29500 magnification (b) x20800 magnification - 35 -Fig. 5 (continued) (c) schematic representation of (a) (d) schematic representation of (b) The platelets are swollen but the plasma membranes appear intact. There are swollen vacuoles (VO) as well as storage granules (AG). - 37 -Fig. 6. Electron micrographs of the cocaine spheres. (a) xl8200 magnification (b) x20800 magnification - 38 -- 39 -Fig. 6 (continued) (c) schematic representation of (a) (d) schematic representation of (d) The platelets here are spherical but much smaller than the spherocytes. Storage granules (AG) are present and channels of the canalicular system (CS) can be seen. - 41 -transformation to echinocytes with ADP and to spheres with cocaine the super-natant remained negative. A more detailed timed study of LDH leakage from the spherocytes was done and shown in Fig. 7. After 90 min of hypotonic shock about 3.87% (+0.80%) of the total platelet LDH activity was found in the super-natant. The percentage increased to 10.19% (+^ 1.70%) after 2-^hours and 25.51% (+1.23%) after 4 hours. 14 The platelet release experiments consisted of two parts, the C-serotonin released from dense bodies and beta-thromboglobulin (beta-Tg) released from alpha-granules. It was found that practically no release of 14 C-serotonin had occurred after echinocyte and spherocyte transformations. Nearly a l l of the radioactivity (97.8% in echinocyte and 97.5% in spherocyte) could be recovered from the pellets after centrifugation. The mean beta-Tg content of washed human discocytes from two determinations was 22.21 jug/109 platelets (22.23 and 22.19^ug/109 platelets). About 13% of this was found in the supernatant of the washed platelets as free beta-Tg. It was possible that some release of beta-Tg had occurred during the washing procedure. Additional releases of 4% during echinocyte transformation and 27% during spherocyte transformation were also found. If the discocytes were stimulated with 0.4 NIH units/ml of thrombin, 94% of their beta-Tg contents was released. 2.3.3 Platelet Aggregations Fig. 8 demonstrates how aggregation velocity increased with ADP concen-tration. The curve was constructed from aggregation measurements of five different samples of washed platelets suspended in Tyrode's solution containing 0.5 mg/ml of fibrinogen and 4 mM Ca + . Aggregation velocity was about maximum at 2x10 "*M ADP. Aggregations at this concentration were therefore - 42 -Fig. 7. LDH leakage from the spherocytes (as percentage of total platelet LDH activity) as a function of time of suspension in hypotonic Tyrode's solution. Values are means of three series of experiments; error bars indicate +1 standard deviation. Vo total LDH release 1 - 44 -considered as 100% in a l l samples for the normalization procedure mentioned before. This concentration was also used throughout this project for echino-cyte transformation. An example of the actual aggregation tracings showing the effect of different ADP concentration is shown in Fig. 9. The effect of fibrinogen concentration on aggregation velocity is shown in Fig. 10. In this case fibrinogen and 4 mM of Ca + + ions were added to the platelet suspensions before ADP (2x10 "JM). Maximum velocity achieved at 0.5 mg/ml of fibrinogen was designated as 100% according to the normalization procedure above. Fig. 10 is a composite of five different series of concen-tration experiments. The curves showing the effect of divalent cation concentration on the aggregation velocity are shown in Fig. 11. Fibrinogen at 0.5 mg/ml together with either C a + + or Mg + + ions were added to the platelet suspensions before ADP (2x10 M). A biphasic effect was observed in both the Ca and Mg curves with maximum aggregation at 4 mM of Ca + + ion. Velocity of this optimum concentration therefore serves as the 100% aggregation velocity for both the Ca + + and Mg + + series. In this way, i t can be demonstrated that the a b i l i t y of Mg + + ion to support aggregation is only about half that of Ca + + ion throughout the concentrations tested. Examples of individual aggregation tracings are shown in Fig. 12. Two more divalent cations, strontium and manganese were tested for their a b i l i t i e s to support ADP induced platelet aggregation. In the presence of 0.5 mg/ml of fibrinogen, Sr ions at concentrations between 0-20 mM failed to support ADP (2x10 "*M) induced aggregation (Fig. 12). If added to a system ++ . ++ already containing 4 mM of Ca ions, Sr would neither augment nor ++ . reduce the supportive role of Ca ions in ADP induced aggregation. - 45 -Fig. 8. The effect of ADP concentration on the aggregation velocity of washed platelets suspended in Tyrode's solution containing 0.5 mg/ml of fibrinogen and 4 mM Ca + +. Aggregation velocity at 2x10 "*M ADP is taken as 100% and velocities at lower concentrations are expressed as percentages of i t . Values are means of five series of experiments; error bars indicate +1 standard deviation. - 47 -F i g . 9. Examples of aggregation tracings showing the e f f e c t of ADP concentration on p l a t e l e t aggregation v e l o c i t y . ADP at (a) 2x10 M; (b) 2x10 M or (c) 1x10 M was added at arrow. 1 MIN. time - 49 -Fig. 10. The effect of fibrinogen concentration on the aggregation velocity of platelets. Washed platelets were suspended in Tyrode's solution containing fibrinogen and 4 mM Ca + ions. Aggregation was induced by 2x10 M^ of ADP. Aggregation velocity at 0.5 mg/ml of fibrinogen was taken as 100% and velocities at lower concentrations were expressed as percentages of i t . Values are from means of five series of experiments; error bars indicate +1 standard deviation. Fibrinogen cone, (mg/ml) - 51 -Fig. 11. The effect of divalent cation concentration on the ++ ++ aggregation velocity of platelets: Ca (-•-), Mg ( - O - ) and S r + + ( -A-) . Washed platelets were suspended in Tyrode's solution containing 0.5 mg/ml fibrinogen and one of the cations. Aggregation was induced by 2x10 "*M of ADP. Aggregation velocity at 4 mM Ca + + ions is taken as 100%. Values are means from five series of experiments; error bars indicate +1 standard deviation. °/ol CATION CONC. (mM) - 53 -Fig. 12. Examples of aggregation tracings showing the effects of different divalent cations. Platelets were suspended in Tyrode's solution containing 0.5 mg/ml of fibrinogen and 4 mM (a) Ca + + ions; (b) Mg + + ions or (c) S r + + ions. ADP at 2x10 ~*M was added as indicated by arrows. v— < >~ 1 MIN. time - 55 -It was found that Mn + + ion was an inhibitor of platelet aggregation. Aggregation induced by 2x10 "*M of ADP in the presence of 0.5 mg/ml of fibrinogen and 4 mM Ca ions was greatly reduced by 2.5 mM and completely inhibited by 10 mM Mn + + ions (Fig. 13). Aggregation induced by 0.1 NIH -4 unit/ml of thrombin or by 1x10 M of arachidonic acid was also fully . . ++ . inhibited by 10 mM of Mn ions. A panel of aggregating agents including ADP (2x10 ~*M with 0.5 mg/ml of fibrinogen and 4 mM of Ca ions), epinephrine (5x10 M^ with 0.5 mg/ml of fibrinogen and 4 mM of Ca + + ions), thrombin (0.1 NIH unit/ml), A23187 -5 -3 (1x10 M) and arachidonic acid (1x10 M) were tested for their a b i l i t i e s to aggregate spherocytes. They a l l failed to do so. These aggregating agents at the concentrations indicated, would strongly aggregate normal discoid platelets. The cocaine spheres also failed to aggregate upon stimulations with ADP, thrombin and arachidonic acid (same concentrations as above). Although spherocytes have lost their a b i l i t y to aggregate their reaction to ristocetin, an agglutinating agent, is quite different. When 1.5 mg/ml of ristocetin alone was added to either discocyte or spherocyte there was no response. If plasma was added to provide for the von Willebrand's factor then agglutination occurred in both spherocyte and discocyte samples. In fact, the agglutination of the spherocytes was twice the rate (velocity) of the disco-cytes (Fig. 14). Plasma was added at a ratio of 0.1 ml per ml of platelet suspension. Under phase microscopy discocytes treated with neuraminidase s t i l l retained their discoid shape. Tested at 2x10 "*M of ADP (with 0.5 mg/ml of fibrinogen and 4 mM of Ca + + added) the aggregation rates were the same for the treated and non-treated discocytes (Fig. 15). - 56 -Fig. 13. The inhibitory effect of Mn on platelet aggre-gation. Washed platelets were suspended in Tyrode's solution ++ ++ containing 4 mM Ca and 0.5 mg/ml of fibrinogen. Mn at (a) 0 mM; (b) 2.5 mM or (c) 10 mM was then added. Aggregation was induced by adding 2x10 "*M of ADP as indicated by arrows. 1 1 y > 1 MIN. time - 58 -Fig. 14. Agglutination of platelet (a) spherocytes and (b) discocytes by ristocetin. Discocytes were suspended in iso-tonic and spherocytes in hypotonic Tyrode's solution. Plasma was added to each sample at 0.1 ml of plasma per ml of platelet suspension, to provide von Willebrand's factor. Values are means from two series of experiments. Ristocetin Cone, (mg/ml) - 60 -Fig. 15. Aggregation tracings from normal (left) and neuraminidase treated (right) platelets. Both were suspended in Tyrode's solution containing 0.5 mg/ml of fibrinogen and 4 mM Ca + + ions. Aggregation was induced by 2x10 "*M of ADP as indicated by arrows. ; . i 1 MIN. time - 6 2 -Wheat germ agglutinin (WGA) aggregated normal discocytes strongly. Afte neuraminidase treatment, the aggregation velocities dropped substantially at a l l the concentrations tested (Fig. 1 6 ) . Typical aggregation tracings are shown in Fig. 1 7 . Aggregation was inhibited by EDTA or EGTA (both at 2 . 2 5 mg/ml) as well as by N-acety1-glucosamine ( 5 0 mM) but galactose and N-acetyl galactosamine at this concentration were without effect. Both the castor bean agglutinins ( R C A ^ Q and R C A ^ Q ) aggregated normal platelets only to a small extent but after neuraminidase treatment aggregati was greatly enhanced (Figs. 1 6 and 1 7 ) . Aggregation was also inhibited by 2 . 2 5 mg/ml of EDTA or EGTA. Aggregation by RCA & 0 was inhibited by both N-acetyl-galactosamine ( 5 0 mM) and D-galactose ( 5 0 mM). Aggregation by R C A ^ Q was inhibited only by D-galactose while N-acetyl-galactosamine had no effect Neither were affected by N-acety1-glucosamine. Jeguirity bean agglutinin (JBA) did not aggregate non-treated platelets and after neuraminidase treatment the platelets were aggregated moderately (Fig. 1 6 ) . The aggregation was inhibited by 2 . 2 5 mg/ml of EDTA or EGTA as well as 5 0 mM D-galactose. N-acetyl-galactosamine and N-acety1-glucosamine were without effect. Aggregation of glutaraldehyde fixed discocytes by WGA could not be demon strated with the aggregometer but under the microscope numerous small aggregates of two or three platelets were seen. - 63 -Fig. 16. Aggregation responses of normal (-•-) and neuraminidase treated (-O-) platelets to the lectins. - 64 -Lectin Cone, (jjg/ml) - 65 -Fig. 17. Examples of platelet aggregation responses to lectins. (i) Normal (a) and neuraminidase-treated (b) platelets aggregated by WGA ( i i ) Normal (c) and neuraminidase-treated (d) platelets aggregated by R C A ^ Q ( i i i ) Normal (e) and neuraminidase-treated (f) platelets aggregated by RCA 1 2 Q A l l lectins were added at 100 u,g/ml as indicated by arrows. - 66 -i t 1 MIN. time - 67 -2.4 DISCUSSION 2.4.1 Platelet Dimension and Morphology -A The mean diameter of the discocyte measured here 3.18x10 cm i s similar -L to 3.2x10 cm reported by Milton and Frojmovic (1979). The R value of 0.28 for human discocyte measured at room temperature was close to that of 0.26 measured at 37°C by Frojmovic and Panjwani (1976) . Although there was a small difference between the R values, the resulting difference i n surface area —8 —8 2 (16.48x10 vs 16.42x10 cm ) was minimal. Born et jil (1978) reported R=0.25 for rabbit platelets. Using electronic particle counting Gear (1981) -12 3 found the human platelet volume to be 4.94x10 cm and Yamazaki and -12 3 Motcmiya (1980) found a similar value 4.66x10 cm . These two values are -12 3 quite compatible with 4.71x10 cm reported i n this chapter. Hypotonic stress at 75 mOsm produced spherocytes with a mean diameter of -4 3.42x10 cm. An identical value was found by Milton and Frojmovic (197 9). Recently Boneu et a l . (1982) found hypotonic treatment at 120 mOsm increased platelet diameter to 3.07x10"^ cm. The surface area and volume of the echinocyte cannot be found from the geometrical means used here because of i t s irregular shape. However i t will be shown i n Chapter 3, from indirect calculations, that the surface area of the echinocyte is probably close to that of the cocaine sphere. Born et al (1978), using a high speed centrifugation radiolabel dilution technique found that -12 3 rabbit platelets had a mean volume of about 5x10 cm . They found no change in platelet volume after echinocytic transformation. The morphologies of the discocyte, echinocyte and the cocaine sphere under electron microscopy were not unlike those reported elsewhere (White, 1972; - 68 -Behnke, 1970). The surface-connected canalicular system was present in the discocyte, echinocyte and the cocaine spheres. The channels of the system were dilated in the echinocyte and severely dilated in the cocaine spheres. Presence of the canalicular system in the echinocyte suggested that evagination of the system was not total. This could be supported by the fact that the surface area of the spherocyte appears to be considerably larger than that of the echinocyte (Chapter 3). Our original hypothesis, that the increase in membrane surface area originates from the canalicular system does not dictate the degree of evagination of the canals. In fact White and Clawson (1980) suggested that the canalicular system within the cytoplasm is so tortuous and interwoven that complete evagination would result in the total destruction of the platelet. Platelets spreading on a surface can attain a surface area of —8 2 70x10 cm (Frojmovic and Milton, 1982), double that of the spherocyte. This implies that even in the spherocyte evagination may not have been total. The large empty vacuoles in the spherocytes might be swollen canalicular channels. Platelet morphology and surface area related to surface chemistry w i l l be discussed further in Chapter 3. 2.4.2 Platelet Integrity and Release of Contents The enzyme LDH is a cytoplasmic enzyme (Gogstad ejt al_., 1981) and has been used extensively as a marker for cytoplasmic leakage from platelets (Milton and Frojmovic, 1979; Fratantoni and Poindexter, 1981; Ostermann et a l . , 1982; Sturk et a l . , 1982). The possibility that the transformation of platelets, especially during hypotonic shock could result in leakage or breaks in the plasma membrane is of considerable concern here. If even a small number of platelets burst, released material could coat the surfaces of the remaining intact ones and the resulting properties could be misinterpreted as platelet - 69 -surface changes during transformation. At 60 mOsm Milton and Frojmovic (1979) found that considerable enzyme leakage had occurred after only 10 min of hypo-tonic shock. In the present work i t was found that LDH leakage was not measur-able after 1 hour of stress at 75 mOsm (Fig. 7) and that the amount remained 3 below 4% for 90 min, the time required to .perform the H -borohydride labelling procedure in Chapter 4. Much shorter time periods (20-30 min) were needed to do c e l l electrophoresis (Chapter 3). Therefore within the time periods of the experiments the platelet spherocytes appeared to be intact by this criterion. Beta-Tg is present in the alpha-granules and the sole source of plasma beta-Tg is from the platelets. Increased amounts of beta-Tg in the plasma indicate vascular events involving in vivo platelet activation (Kaplan and Owen, 1981). The radioimmunoassay kit developed by Amersham is intended for use with plasma but can easily be adopted to assay platelet beta-Tg with l i t t l e change in the assay protocol. This kit has already been used by a number of groups for in vitro platelet experiments as an indicator for alpha-granule release (Ludlam and Cash, 1976; Bolton et^ a l . , 1980; Gogstad et^  a l . , 1981). Kaplan and Owen (1981) have reported that the mean platelet 9 9 beta-Tg content is 17.7^ig/10 platelets, a l i t t l e lower than 22.2jxg/10 platelets found here. It seems that i t is extremely easy to induce beta-Tg release. It was recommended by the manufacturer and others (Ludlam and Cash, 1976) that the blood should be collected in EDTA and handled at 4°C. This was not possible in our circumstances and a considerable amount of beta-Tg was released during washing and hypotonic shock as a result. Examination of the platelets under electron microscopy, on the other hand, did not show the platelets to be grossly depleted of their granules. No release from the dense bodies was evident as monitored using 14C -serotonin. - 70 -2.4.3 Platelet Aggregations The effect of ADP concentration on platelet aggregation has been well studied. The results shown in Fig. 8 were not significantly different from those found by others (Mills et a l . , 1968; Tangen et a_l., 1971; Frojmovic, 1973; Stibbe and Holmsen, 1977; Marguerie et a l . , 1979). It is seen that 2x10 M^ of ADP gives the maximum aggregation velocity. This concentration was therefore used throughout this project. In most of the experiments in Chapters 3 and 4 shape change (to echinocyte), and not aggregation, was desired. Fibrinogen and divalent cations were withheld from the platelets to avoid aggregation. Born (1970) has shown that the speed and extent of shape change also depend on ADP concentration. Michal and Born (1971) found that the velocity of aggregation was directly related to the velocity of shape change. In the present work i t was also found, based on the magnitude of the i n i t i a l decrease in light transmission preceding the increase due to aggregation, that higher ADP concentrations brought about larger shape change responses. The effect of fibrinogen concentration on platelet aggregation induced by ADP has also been well studied and Fig. 10 is similar to concentration curves reported by others (McLean e_t al., 1964; Deykin et al. , 1965; Niewiarowski e_t a l . , 1977; Marguerie e_t al., 1979; Harfenist et al., 1980). The fibrinogen concentration study here forms the basis for the fibrinogen adsorption experiments to be described in Chapter 4. Fig. 11 shows the biphasic platelet aggregation response to increasing concentrations of Ca + + and Mg + + ions. Such biphasic responses were also observed by Born and Cross (1964) and Heptinstall (1976). It was found here . . . . ++ ++ . that aggregation velocities increased with Ca or Mg ion concentration - 71 -until the optimum occurred at 4 mM. Further increases i n concentration caused a decrease in aggregation velocity, the reason for which is not clear although i t i s l i k e l y associated with the high ionic strength of the suspending medium at the higher C a + + or Mg + + concentrations. Born and Cross (1964) found the optimum C a + + concentrations. Born and Cross (1964) found the optimum Ca + + concentrations to be 1.7 mM while Heptinstall (1976) found i t at 2.5 mM. They also found, as shown here, that Mg + + ions were less effective i n supporting ADP induced platelet aggregation. The requirement of Ca + + or Mg + + ions for the binding of fibrinogen to ADP activated platelets has been discussed earlier. Marguerie ^ t a l . (1980) found that this requirement i s also biphasic. The calcium or magnesium ion concentration optimal for fibrinogen binding is about 1 mM and binding decreases below or above this concentration. Marguerie ^ t _al. (1980) also found that Mg + + was less effective i n f u l f i l l i n g the requirement for fibrinogen binding than was Ca + +. At the optimum calcium ion concentration, about 3x10^ fibrinogen molecules bind to each platelet i f magnesium i s substituted. The calcium and magnesium ion concentration curves for platelet aggregation (Born and Cross, 1964; Heptinstall, 1976 and reported here) and fibrinogen binding (Marguerie, 1980) are strikingly similar, suggesting the presence of a close relationship between platelet aggregation and fibrinogen binding. The contribution of strontium ions (Sr + +) to platelet aggregation i s not as clear. It was found here that between 0 and 20 mM, S r + + did not support ADP-induced aggregation. Lages et a l . (1975) found that the presence of 0.2 mM of S r + + had l i t t l e or no effect on ADP and adrenalin induced platelet aggregation. On the other hand, Best et a l . (1981) reported that addition of Sr ions alone could result in platelet activation with thromboxane production and serotonin release. Unlike magnesium, calcium and strontium, manganese is not a type IIA element. Its contribution to platelet aggregation is also uncertain. It has been proposed by Bosmann (1972) and Wu and Ku (1978) that sialyltransferase on the surface of platelets may be involved in the platelet aggregation process. The hypothesis is that the enzyme on one platelet interacts with the substrate s i a l i c acid on the surface of another platelet forming a "bridge." This has never been substantiated. This enzyme has a requirement for Mn+ as co-factor. However, present results suggest that Mn + + is an inhibitor of platelet aggregation. It is therefore unlikely that sialyltransferase plays any essential part in platelet aggregation. The optimum concentration of Mn + + ions required by the enzyme is between 0.8 and 2 mM as reported by Wu and Ku (1978) and 20 to 60 mM according to Bosmann (1972). Manganese ions at a few millimolar concentration have an inhibitory effect on the movement of calcium ions across the sacroplasmic reticulum membranes in muscle fibres (Saida and Suzuki, 1981). Platelet activation by various aggregating agents involves the massive movement of compartmentalized calcium into the cytoplasm (Detwiler et: a l . , 1978; Massini £t aj.., 1978). Whether manganese inhibits platelet aggregation by interferring with calcium ion movements or not remains ++ ++ ++ ++ to be determined. The fact that Ca , Mg , Sr and Mn a l l have very different effects on ADP-induced aggregation at the same concentration strongly implies that specific binding of the divalent ions to, presumably, membrane sites is involved in their activity, since generalized electrostatic effects would depend only on the ionic strength, not on the chemical nature of the ions (Diamond and Wright, 1969). - 73 -After spherocyte transformation, the platelets were found to have lost their a b i l i t y to aggregate. The most likely explanation is that hypotonic shock has caused platelet damage to the point that aggregation no longer is possible. The loss of aggregability in cocaine spheres is also expected (O'Brien, 1962; Aledort and Niemetz, 1968). In both types of spheres the loss of aggregability may be attributed to the total disruption of contractile and cytoskeletal elements (Zucker-Franklin, 1969; Nachmias et al_., 1977; Nachmias et a l . , 1979). Membrane changes to be described in Chapter 3 may also be involved. Ristocetin i s not a pharmacological aggregating agent but is a direct agglutinating agent. That i s , no physiological response from the platelet is required to induce agglutination. In fact, ristocetin has been shown to agglutinate formaldehyde fixed platelets (Allain e_t a l . , 1975). Agglutination requires the co-binding of ristocetin and a plasma factor (von Willebrand's factor) to the surface of platelets. The exact nature of the interaction is s t i l l uncertain and a variety of possible models have been reviewed elsewhere (Kirby, 1977; Solum and Peterka, 1977; Coller, 1978; Phillips, 1980). Risto-cetin binds to the surface of platelets (Hashimoto and Suzuki, 1979) and produces a condition that favours the binding of von Willebrand's factor to the platelet surface (Hahsimoto and Suzuki, 1979; Kao e_t a l . , 1979; Schneider et al., 1979). The von Willebrand's factor and/or ristocetin then cross-bridge receptors on the surfaces of adjacent platelets, causing agglutination. The receptor for von Willebrand's factor was found to be associated with the surface glycoprotein designated lb (Phillips, 1980). Patients with Bernard-Soulier syndrome lack this glycoprotein and their platelets f a i l to agglutinate upon exposure to ristocetin and the plasma factor. On the other hand persons - 74 -with von Willebrand 1s disease lack the factor in their plasma and their plate-lets w i l l agglutinate only after external von Willebrand's factor is added. Although spherocytes have lost their a b i l i t y to aggregate they are s t i l l able to agglutinate in response to ristocetin and added plasma. Hypotonic shock has therefore not destroyed the receptors for either ristocetin or von Willebrand's factor. The reason why the rate of agglutination is greater for the spherocyte than the discocyte is uncertain. It may be related to the increase in surface area or platelet volume in the spherocyte, perhaps increasing the number of receptor sites per platelet for either type of molecule. The increase in volume might also lead to an increase in the number of platelet- platelet collisions in the spherocyte suspension, since a larger fraction of the suspension volume would be occupied by the spherocytes. The decrease in negative charge density on the surface of the spherocyte (Chapter 3) might also contribute to the increased aggregation rate. Ristocetin is a positively charged molecule and i t has been proposed that i t s function in the agglutination is to neutralize some of the platelet surface negative charge (Phillips, 1980). If the surface charge density has already been lowered during spherocyte transformation then the action of ristocetin might be much enhanced. Finally, since the spherocytes were suspended in hypotonic Tyrode's solution while the discocytes were in isotonic Tyrode's solution, the low ionic strength might also play a role in f a c i l i t a t i n g agglutination of the spherocytes. As expected, neuraminidase treatment of platelets did not alter their shape or their aggregability. Hovig (1965) as well as Bowles and Brunton (1982) found no alteration in platelet morphology under transmission and scanning electron microscopy after neuraminidase treatment. Greenberg e_t a l . - 75 -(1975) found only a slight enhancement of ADP induced aggregation in treated platelets. The enhancement was most obvious at a low (4.5x10 M^) ADP concentration with pratically no difference being observed at a higher (9x10 M^) concentration. At 2x10 "'M no differences were found here. Greenberg and Jamieson (1974) reported that human platelets can be aggre-gated by WGA and to a lesser extent by RCA. Ganguly and Fossett (1979) found that neuraminidase treatment of platelets reduced the extent of aggregation by WGA as well as the number of WGA binding sites on the platelet surface. Patscheke and Worner (1977) and Nairn et aj.. (1982) observed that removal of terminal s i a l i c acid residues from the platelet surface by neuraminidase actually enhanced the aggregation response to RCA. Similar phenomena were observed by us. The effects of lectin concentration on the aggregation of treated and control platelets was also studied here (Fig. 16). WGA binding can be specifically inhibited by N-acetyl-glucosamine but in the absence of this sugar i t can interact strongly with the terminal s i a l i c acid residues of glycoproteins (Goldstein and Hayes, 1978; Bhavanandan and Katlic, 1979; Peters et a l . , 1979; Wright, 1980). In platelets i t has been demonstrated by a number of workers that WGA binds to glycoprotein-Ib (GP-Ib) on the surface of the c e l l (Ganguly and Fossett, 1979; Marchesi and Chasis, 1979; McGregor _et a l . , 1979; Nairn et a l . , 1982). These experiments involved either the binding of radiolabelled WGA by separated glycoproteins in poly-acrylamide gels or the isolation of solubilized GP-I using WGA-Sepharose affinity chromatography. Caution in interpreting these results was urged by Nairn et a l . (1982). Moreover Rock et a l . (1980) found that Bernard-Soulier syndrome platelets which lack GP-Ib could s t i l l be aggregated by WGA. It therefore seems unlikely that GP-Ib is the only receptor on platelets. The - 76 -hypothesis is that WGA binds to terminal s i a l i c acid residues of the glyco-p r o t e i n s ) on the platelet surface and causes platelet aggregation via the usual cross-linking of adjacent surfaces by the multivalent lectin. Removal of s i a l i c acid by neuraminidase therefore reduces the platelet's response to WGA (Ganguly and Fossett, 1979). The removal of terminal s i a l i c acid residues apparently exposes galactose or N-acety1-D-galactosamine residues. This suggestion is based on the effect of R C A ^ Q , RCAj^o a n <* ^ B A ' a ^ °^ w n i c n have a high affinity for galactose (Goldstein and Hayes, 1978) and a l l of which show enhanced activity with neuraminidase treated platelets (Patscheke et. al., 1977). These galactose-binding lectins probably induce aggregation by the same mechanism(s) as does WGA. That galactose or N-acety1-galactosamine residues can be exposed after neuraminidase removal of terminal s i a l i c acid has been found in many systems (Steck and Dawson, 1974; Wright, 1980). The combined use of neuraminidase and galactose oxidase followed by reduction with tritiated NaBH^ to label c e l l surface glycoproteins has become a standard procedure in recent years (Phillips, 1979). This labelling procedure as applied to platelet membrane studies w i l l be discussed in Chapter 4. Finally, a small degree of agglutination was observed when fixed platelets were exposed to WGA. Similar to our study Ganguly and Fossett (1980) observed agglutination of fixed platelets by WGA microscopically but were unable to demonstrate this with aggregometry. They therefore suggested that two mechanisms, passive agglutination and active aggregation, might be at work in this situation. - 77 -2.4.4 Summary of Chapter 2 The dimensions and morphology of the different forms of platelets were established. The most important point found was that most spherocytes formed via hypotonic shock remained intact for a reasonable period of time, although some release of beta-Tg occurred. The second half of Chapter 2 dealt with platelet aggregation in general, and i t was found that the spherocytes had lost a l l abi l i t y to aggregate. They could only be agglutinated passively by ristocetin and von Willebrand 1s factor. - 78 -CHAPTER 3 MICROE IE CTROPHORE SIS - 79 -3.1 INTRODUCTION Microelectrophoresis or c e l l electrophoresis is a technique used to probe the electrokinetic properties of c e l l surfaces. An in depth theoretical discussion of microelectrophoresis and the electrokinetic behaviour of cells can be found in a review by Seaman, 1975. Briefly, the electrophoretic mobility (EPM) of a c e l l is measured by observing the velocity of i t s motion under the influence of the particular electric f i e l d applied. It is defined as the velocity per unit f i e l d strength and is usually expressed in terms of 2 -1 -1 cm .sec .V . The EPM (ja.) of a c e l l , assuming i t behaves as a smooth particle bearing a uniform charge density, is directly related to it s zeta potential (IS), i.e. the electrostatic potential at the shear plane. The zeta potential i s in turn related to the charge density (c?) apparently located at the plane of shear: where 1/K = double layer thickness ~ 8A at 1=0.15 = viscosity of suspending medium N = Avogadro's number I = ionic strength of suspending medium =1^ C^Z^ .. .th . . c^ = molar concentration of l ionic species n . .th . . z. = valence of I ionic species I £ = dielectric constant k = Boltzman's constant T = absolute temperature e = electronic charge unit (4.8x10""^ esu) - 80 -Equation [1] holds providing the smallest radius of curvature of the particle is much larger than the double layer thickness 1/K. Equation [2] is valid providing z^e^^kT. Both conditions are f u l f i l l e d by platelets suspended in physiological buffers. Combining [1] and [2] gives: CT [3]. 2 The charge density,cr, is expressed as esu/cm but can be converted into the 2 . . number of charges per cm by dividing by e. The surface of most cells contains both positive and negative charge groups but in general the net surface charge is a negative one at physiological pH. The charge groups on the surface of a c e l l attract ions of opposite charge (counter-ions) and repel charges of like sign (co-ions), the net effect being the formation of the diffuse double layer adjacent to the c e l l surface. In some instances i t is found that a fraction of the counterions adsorb to the particle surface reducing the net surface charge density and forming what is known as a Stern layer. There is no evidence for the formation of a Stern layer on biological cells i f only monovalent ions are present in the suspending medium, as no dependence of mobility on the chemical nature of the ions has been found (Heard and Seaman, 1960). When multivalent cations such as Ca + or Mg + + are present, however, binding does occur and the mobility decreases associated with these events can be used to estimate the binding parameters, as w i l l be discussed subsequently. Equations [1] to [3] assume the surface of the particle is smooth and well defined, although i t may be irregular in shape. However, the surfaces of biological cells are not smooth on the scale of double layer dimensions. The li p i d bilayer anchors a diffuse layer of charged and neutral glycoproteins and glycolipids known as the glycocalyx which extends some distance out from the - 81 -plane of the l i p i d head groups. Recently, theories have been developed which explic i t l y model the effects of this layer on the electrophoretic mobility (Donath and Pastushenko, 1979; Wunderlich, 1982; Levine et. _al., 1983). The results provide a modification of [3] in the form of: o-F=^nK [4]. The function F includes terms involving the thickness of the glycocalyx and the average size and volume concentration of the polymer segments in this region. None of these parameters are known for platelets, however. Equation [3] is therefore used to interpret electrophoretic mobilities in this work, the assumption being made throughout that the function F remains constant or changes in a consistent manner among the platelet forms examined as conditions are varied. Like most other types of cells platelets have a negative net surface charge. The electrokinetic behaviour of platelets has been reviewed by Seaman and Brooks (1970), Mason and Shermer (1971) and Seaman (1976). The electro-phoretic mobility of platelets was f i r s t investigated by Abramson in 1928. He suspected that the electrical charge on platelets could have something to do with platelet aggregation and thrombosis. However the lack of information about platelets at that time prevented him from further examining this point. More recently Seaman and Vassar (1966) found that addition of lyag/ml of ADP to platelets in PRP caused an 18% decrease in mobility. They suggested that the decrease in EPM might somehow be associated with aggregation induced by ADP. Hampton and Mitchell (1966) reported a biphasic change in the EPM of platelets (in PRP) induced by ADP. At extremely low concentrations the mob-i l i t y increased, possibly due to the binding of ADP onto the platelet surface. - 82 -TTU »tm v c i n~2 / i i , _ , , „ _ 7M) the e l e c t r o p h o r e t i c m o b i l i t y When ADP was above 5x10 g/ml (about 1x10 v J ddecreased w i t h i n c r e a s i n g concent ra t i on . Th is b iphas i c phenomenon was a l so observed by S t o l t z (1971) as w e l l as K o s z t o l a n y i et a l . (1980) but s e ve r a l groups i n c l u d i n g Grottum (1968), Be t t s et a l . (1968) and Seaman and Vassar (1966) c o u l d not conf i rm t h i s e f f e c t . A drop i n the sur face negat ive charge would mean a drop i n the mutual r e p u l s i v e fo rce between p l a t e l e t s . Th i s might b r i n g about a more f a v o u r i t e c o n i d t i o n f o r p i at e l e t - p i at e l et i n t e r a c t i o n and aggregat ion (Grotum, 1968 and Seaman, 1976). No morpho log ica l s tud ie s were undertaken by these authors to c o r r e l a t e between p l a t e l e t shape change and the decrease i n m o b i l i t y . A French group, on the other hand, u s i n g 2x10 of ADP have r epo r ted that d i s cocy te to ech inocy te t r an s f o rma t i on i n PRP r e s u l t s i n a m o b i l i t y i n c rea se (Boisseau et ^ 1 . , 1977). S i a l i c a c i d i s a major c o n t r i b u t o r to p l a t e l e t s u r f a ce negat ive charge. Jung et _al. (1982) found a l i n e a r c o r r e l a t i o n between p l a t e l e t su r face s i a l i c a c i d content and EPM. I t has been est imated that about 41% of the net negat ive charge detected e l e c t r o k i n e t i c a l l y comes from termina l s i a l i c a c i d res idues (Seaman, 1976). Another 28% der i ves from phosphate groups. Neuraminidase removal of the t e rm ina l s i a l i c a c i d groups has been found t o r e s u l t i n a 40 t o 60% drop i n human p l a t e l e t e l e c t r o p h o r e t i c m o b i l i t y (Madoff _et a l . , 1964; Seaman and Vas sa r , 1966; Bray and Alexander , 1969; S t o l t z and N i c o l a s , 1979). F ran the drop i n m o b i l i t y one can c a l c u l a t e the decrease i n charge den s i t y and p rov i d i ng the su r face area i s known, convert that i n t o the a c tua l number of s i a l i c a c i d molecules r e l ea sed (Seaman and Vas sa r , 1966). One must be aware that t h i s number r e f l e c t s on ly s i a l i c a c i d r e l e a s e d from near the e f f e c t i v e plane of shear. Other s i a l i c a c i d r e s i due s l o c a t e d at d i f f e r e n t d i s tances from t h i s plane c o n t r i b u t e somewhat l e s s to the e l e c t r o k i n e t i c p roper t i e s of - 83 -the platelets and their release can be expected to have l i t t l e effect on the mobility. The total number of s i a l i c acid molecules liberated by the enzyme from the platelet membrane into the supernatant can be assayed chemically. The effective number of s i a l i c acid molecules at the plane of shear, determined using equation [3], can be expressed as a fraction of the total platelet sur-face s i a l i c acid removable by neuraminidase. This fraction is approximately 46% for human red blood cells for example (Cook et al., 1961). It w i l l be shown here that this ratio can be very helpful in the determination of the echinocyte surface area. The enzyme alkaline phosphatase has also been used to remove negatively charged phosphate groups from the surface of platelets (Mehrishi, 1979 and Stoltz e_t al_., 1975). Decreases in mobility of 15 to 30% were reported. Similarly, the number of phosphate groups eliminated at the plane of shear can be calculated based on equation [3] and expressed as a fraction of the total amount of phosphate liberated from the platelet surface by the enzyme. Alkaline phosphatase has a specificity for monester orthophosphates, C-O-P (Fernley, 1971). Fixation of platelets with acetaldehyde has been reported to bring about a 20% increase in negative mobility (Seaman and Vassar, 1966). Fixation of red blood cells with glutaraldehyde also results in an increase in RBC mobility (Vassar e_t a_l., 1972). The aldehydes block the positively charged amino groups eliminating their positive charge (Jentoft and Dearborn, 1979). As a result net negative charge increases. The number of negative charges increased, calculated from the rise in EPM, gives the apparent number of positive amino groups neutralized. - 84 -The electrophoretic mobilities of platelet discocytes compared to those of the platelet spheres as well as the electrokinetic properties of the different forms of platelets modified with neuraminidase, alkaline phosphatase and aldehydes are discussed in this chapter. Electrophoretic mobility measurements have also been used to estimate the number of calcium ion binding sites on cells (Seaman et a l . , 1969). If red cells are suspended in solutions of constant ionic strength containing different concentrations of calcium ions, their mobilities w i l l drop with increasing concentrations of calcium. By plotting the relationship between the decrease in apparent charge density (Ac) and calcium concentration according to the equation: J L_ + _J 1  Ao- ~ 2en + 2enK ' tCo*]-exp(2eX/kT) ' where K = exp(AG/kT)/55.6 2 n = number of binding sites per cm AG = chemical free energy of adsorption [Ca + +] = calcium concentration t = zeta potential of platelet at that calcium concentration the number of calcium ion binding sites as well as the chemical free energy of binding can be evaluated. If the binding obeys the above model, a plot of I/ACT versus the reciprocal of the concentration-zeta potential function w i l l be a straight line. The intercept with the ordinate w i l l give l/(2en) and the slope w i l l represent l/(2enK). This operation can also be used to find the number of magnesium ion binding sites. Attempts had been made previously to determine the number of calcium ion binding sites on platelet surfaces by 45 ++ measuring the amount of Ca adsorbed after an equilibrium incubation - 85 -(Peerschke et a l . , 1980; Taylor and Heptinstall, 1980; Brass and Shattil, 1982). However this technique suffers from the serious drawback that platelets also have an active process of calcium ion internalization or uptake and i t becomes d i f f i c u l t to distinguish between surface associated and internalized ^~*Ca++ (Peerschke et a l . , 1980). Moreover, there is no suitable magnesium isotope for this kind of study. The microelectrophoresis technique described here therefore offers a unique opportunity to investigate calcium and magnesium ion binding to platelet surfaces. - 86 -3.2 MATERIALS AND METHODS 3.2.1 Mi cro ele ctro phor esi s The electrophorecti c mobil ities of platelets were measured i n a cylindrical chamber essentially as described by Seaman and Heard (1961). The chamber was immersed i n a water bath at 25°C and measurements were made at 40V. Electrodes were of silver/silver chloride. Mobilities of between 10 to 20 platelets from each sample were usually measured with typical standard deviations of -4 2 - 1 - 1 +0.07x10 cm .sec .V . Unless indicated a l l the samples were measured in pH 7.4 isotonic Tyrode's solution except for the non-fixed spherocytes which were measured in the hypotonic Tyrode's described earlier (pH 7.4). Platelet samples were usually examined under phase microscopy before electrophoresis. The mobilities of fresh human red blood cells were measured as a control each day to ensure proper working conditions for the apparatus. The EPM of the human RBC at pH 7.4 i n 0.15MNaCl is -1.08x1 O^cm 2.sec" 1 .V_1 with a standard -4 2 - 1 - 1 deviation of +0.05x10 cm .sec .V . Measurements from the daily controls f e l l within one standard deviation of the quoted mean. The pH-mobility profiles were constructed from measurements of platelets suspended i n Tyrode's solutions having different pH values. Adjustments of the pH's were made with HC1 or NaOH shortly before the platelets were introduced into the electrophoresis chamber. ++ ++ The Ca and Mg ion binding experiments were done with platelets suspended i n Tyrode's solutions containing different concentrations of calci um or magnesium chloride. However constant ionic strength and osmolality had to be maintained. With increasing Ca + + or Mg + + concentration, the ionic strength of the solution increases. Therefore the amount of sodium chloride In the solution has to decrease. At the same time, osmolality decreases as a result - 87 -because the ionic strength, I, varies as the square of ionic valence so to maintain I constant as Ca concentration is increased the tonicity of necessity drops. The addition of an appropriate amount of glucose is used to compensate. The resulting changes in viscosity have to be accounted for when calculating surface charge (Seaman et a l . , 1969). Viscosities were measured by a Cannon viscometer. Spherocytes were suspended in a solution with 1/4 isotonicity for a l l measurements. 3.2.2 Fixation of Platelets Fixation of platelets with glutaraldehyde was the same as in Chapter 2. Formaldehyde fixation of platelets was done with 3.7% formaldehyde (final concentration) overnight at 4°C. Some samples of the formaldehyde fixed platelets were further treated with 5 mM sodium borohydride for 30 min at room temperature (Jentoft and Dearborn, 1979). They were then washed and resus-pended in Tyrode's solution. 3.2.3 Platelet Surface Sial i c Acid Surface s i a l i c acid from glutaraldehyde fixed platelets was removed with neuraminidase (Vibrio cholerae). Digestion time was 90 min at 37°C. This incubation period was determined by a time-release study. Platelet concen-tration was 2.5x10 /ml and neuraminidase 0.04 IU/ml. For optimum conditions (Ada et a_l., 1961) the pH of the Tyrode's solution was lowered to 6.5 and 0.1 mM of Ca was added. After digestion the platelets were spun down. The supernatants were used for the chemical assays of released s i a l i c acid and the platelet pellets were resuspended in pH 7.4 Tyrode's solution for microelectro-phoresis. - 88 -The charge density of control and enzyme treated platelets can be calcu-lated from equation [3], The number of s i a l i c acid molecules removed per unit area at the plane of shear can be calculated by dividing the drop in charge density with the electron charge unit (e). The procedure of Culling e_t a_l. (1977) was used to monitor the amount of s i a l i c acid released into the supernatant. Sialic acid was f i r s t oxidized with sodium metaperiodate and then estimated colorimetrically with thio-barbituric acid reagent. N-acetyl-neuraminic acid was used as standard. To find the total s i a l i c acid content, fixed whole platelets were hydro-lysed with 0.1N sulphuric acid at 80°C for 1 hour. Sialic acid released this way was assayed by the Warren (1959) procedure. To test for 0-acetyl substitution at the C-4 position of s i a l i c acid, fixed platelets were saponified with 0.1N potassium hydroxide at room temperature (Reid et a l . , 1975) before neuraminidase digest. Neuraminidase is not able to attack s i a l i c acid residues bearing such a substitution, therefore more s i a l i c acid w i l l be released after saponification i f C-4 is acetylated. 3.2.4 Platelet Surface Phosphate Groups Alkaline phosphatase (calf intestine) was used to liberate terminal phos-phate groups from the surface of glutaraldehyde fixed platelets. Incubation was at 37°C for 45 min using 0.1 mg/ml of the enzyme. Due to the presence of phosphate in Tyrode's solution, platelets were resuspended in pH 10 (optimum pH) Tris-buffered saline before enzyme treatment. After digestion the platelets were spun down, the supernatant was used for chemical assay of phosphate and the platelet pellet was resuspended in pH 7.4 Tyrode's solution for microelectrophoresis - 89 -The method of Chen jet _al. (1956) was used to assay phosphate ions liberated by the enzyme into the supernatant. Phosphate forms a complex with ammonium molybdate and color was developed by reduction of the complex with as corbie acid. 3.2.5 Statistical Methods Linear regressioin and other s t a t i s t i c a l comparisons were according to Kalbfleisch (1974). Further discussion can be found i n the appendix. 3. 2. 6 Materials Neuraminidase was from Calbiochem (La J o l l a , California) and calf intestine alkaline phosphatase from Boehringer-Mannheim (Dorval, Quebec). Both enzymes were tested for the presence of proteolytic activity with azocoll (Rinderknecht et a l . , 1968). The enzymes were incubated with 3 mg/ml of azocoll (Calbiochem) overnight at 37°C under the same conditions, such as pH, medium and concentration, as used i n the platelet experiments. Trypsin was used as standard. No proteolytic activity was detected i n either enzyme. N-acetyl neuraminic acid and trypsin were from Sigma (St. Louis, Missouri). A l l other reagents were from Fisher (FairLawn, N. J.) . - 90 -3.3 RESULTS 3.3.1 Platelet Electrophoretic Mobilities Electrophoretic mobilities of the different forms of platelets are shown in Table 2. Table 3 shows the mobility values converted into number of charges per platelet and per unit area. Shape change from discocyte to echinocyte induced by 2x10 "*M of ADP resulted in a 13% drop in mean mobility and mean electrokinetic charge density. Since the surface area of the echinocyte is not known at this point, one cannot calculate the actual number of charges per echinocyte from the data. We were not able to detect any change in the mobility of platelets stimulated with 1x10 M^ ADP. The spherocyte had the lowest charge density although the mobility for the non-fixed spherocyte was relatively high. This is because non-fixed spherocytes were measured in low ionic strength hypotonic Tyrode's solution. Adding ADP to spherocytes did not change the mobility. Formaldehyde fixation raised platelet mobilities by about 17%. Table 4 shows how these changes are translated into increases in net negative charge density. The apparent number of amino groups on the platelet surface can then be calculated. The surface densities of amino groups on the echinocyte and spherocyte were both lower than on the discocyte. Because of the increase in surface area the number of amino groups per spherocyte actually increased. Not knowing the surface area, the number of amino groups per echinocyte cannot be calculated. The mobilities of fixed platelets further treated with boro-hydride were the same as the untreated ones. Therefore there was no evidence for reversal of the fixation process. The pH-mobility curves for fixed and non-fixed platelets are shown in Fig. 18 and Fig. 19 respectively. The plateau values for glutaraldehyde fixed Table 2. The El e c t r o p h o r e t i c M o b i l i t i e s of P l a t e l e t s i n x l cAcm 2 .sec - 1.V - 1 Non-fixed Glutaraldehyde Formaldehyde P l a t e l e t s Fixed P l a t e l e t s Fixed P l a t e l e t s Discocyte -1.08 -1.41 -1.26 (+0.016) + (+0.020) (+0.012) Echinocyte -0.93 -1.21 -1.10 (+0.035) (+0.008) (+0.007) Spherocyte -1.55* -1.00 -0.90 (+0.016) (+0.010) (+0.007) Cocaine sphere -1.40 (+0.017) *Non-fixed spherocytes measured i n hypotonic Tyrode's s o l u t i o n ; a l l others measured i n regular is o t o n i c Tyrode's. "•"Values i n brackets represent standard error of the mean derived from mobility determinations of at least 10 d i f f e r e n t p l a t e l e t samples i n each category. - 92 -Table 3. Apparent Platelet Surface Charge Discocyte Echinocyte Spherocyte Cocaine sphere Electrophoretic Mobility (xl0 4cm 5.sec- 1.V- 1) -1.082 (+0.016) -0.933 (+0.035) -1.552 (+0.016) -1.405 (+0.017) Negative Charge Density (esu/cm2) 3.74xl03 (+0.056xl03) 3.23xl03 (+0.121xl03) 2.60xl03 (+0.026xl03) 4.87xl03 (+0.058xl03) Number of Charge Groups per cm2 7.80xl0 1 2 (+0.12xl0 1 2) 6.73xl0 1 2 (+0.25xl012) 5.43xl0 1 2 (+0.05xl0 1 2) 10.14xl0 1 2 (+0.12xl012) Surface Area (cm2) 16.4xl0"8 - 36.7xl0 - 8 26.4xl0 - 8 Number of Charge Groups per Platelet 1.28xl06 (+0.019xl06) -1.99xl06 (+0.020xl06) 2.67xl06 (+0.032xl06) Table 4. Platelet Surface Amino Groups Calculated from Charge Density Increases After Formaldehyde Fixation Discocyte Echinocyte Spherocyte Charge density of non-fixed platelets (esu/cm2) 3.74xl03 (+0.056xl03) 3.23xl03 (+0.121xl03) 2.60xl03 (+0.026xl03) Charge density of formaldehyde-fixed platelets (esu/cm2) 4.38xl03 (+0.041xl03) 3.80xl03 (+0.024xl03) 3.12xl03 (+0.023xl03) Increase in charge density (esu/cm2) 6.40xl02 (+0.526xl02)* 5.70xl02 (+0.271xl02)* 5.20xl02 (+0.226xl02)* 9 Ammo groups per cur- 1.33xl01 2 (+0.109xl012) 1.19xl0 1 2 (+0.047xl012) 1.08xl0 1 2 (+0.056xl012) Surface area (cm2) 16.4xl0 - 8 - 36.7xl0- 8 Amino groups per platelet 2.18xl05 (+0.179xl05) - 3.97xl05 (+0.172xl05) *Combined standard deviations from fixed and non-fixed platelet charge densities using the formula (Kalbfleisch, 1974): S=( -s2 + V 1 .c2VS where S is the combined standard deviation, s a and s^ are standard deviations of the two individual components, and n a and nj, are number of samples from the two individual components. - 94 -F i g . 18. The pH-electrophoretic mobility p r o f i l e s of fixed p l a t e l e t discocytes (-•-), echinocytes and spherocytes (-X-). S o l i d l i n e s represent glutaraldehyde fixed p l a t e l e t s and dash l i n e s represent formaldehyde fixed p l a t e l e t s . Values are from means of four series of experiments. E r r o r bars represent one standard deviation. discocyte 4 -I echinocyte 1 spherocyte - 96 -Fig. 19. The pH-electrophoretic mobility profiles of fresh platelet discocytes (-•-), echinocytes (-A-) and spherocytes ( -T- ) . Values are means from three series of experiments. Error bars represent one standard deviation. 1 1 J ^spherocyte -discocyte -ir echinocyte i i 7 8 - 98 -platelets are about 10% higher than those of the formaldehyde fixed ones (Fig. 18 and Table 2). At low pH's two problems were encountered. The f i r s t problem was platelet agglutination which occurred at pH 3 or below. The second problem was irreversible changes in the mobility at low pH. For example, when the pH of the discocyte suspension was brought down to 4.0 from 7.4 the mobility went _4 2 -1 -1 from -1.08 to -0.75x10 cm .sec .V . Upon returning to pH 7.4, the disco-_4 2 - 1 - 1 cyte mobility went back only to -0.98x10 cm .sec .V . Therefore mobilities below pH 4.5 were not explored further. 3.3.2 Platelet Surface Sialic Acid The neuraminidase digest-time curve is shown in Fig. 20. After 30 min. most of the neuraminidase susceptible s i a l i c acid was released from the glutaraldehyde fixed discocytes. An arbitrary time of 90 min (3x30 min) was therefore used for further enzyme digestions. The amount of s i a l i c acid liberated from the platelet surfaces was deter-mined in two ways: f i r s t , by chemically assaying the platelet suspending media and second, by calculation from the drop in platelet electrophoretic mobility. The chemical analysis data is shown in Table 5. No major differences were found in total s i a l i c acid content between the discocyte, echinocyte and spherocyte, total s i a l i c acid being s i a l i c acid liberated by acid hydrolysis. It represents s i a l i c acid from within as well as on the surface of the platelets. Neuraminidase, on the other hand, removed more s i a l i c acid (17-18%) from the echinocyte and spherocyte than from the discocyte surface (Table 5). Neuraminidase digestion of saponified platelets failed to release any more s i a l i c acid than from the non-saponified ones. No 0-acetyl substitution was therefore evident. - 99 -Fig. 20. The time-release curve for the digestion of s i a l i c acid from the surface of fixed platelet discocytes by g neuraminidase. 2.5x10 /ml of platelets were incubated with 0.04 IU/ml of enzyme. Values are means from three series of experiments. Error bars represent one standard deviation. - OOL -- 101 -Table 5. Platelet Surface Sial i c Acid Determined Chemically Discocyte Echinocyte Spherocyte Total s i a l i c acid released by acid hydrolysis: (n=10) j L A g / 1 0 1 0 platelet 205 (+6.0) 203 (+3.8) 204 (+4.9) molecules/platelet 4.00xl07 (+0.117xl07) 3.96xl07 (+0.074xl07) 3.98xl07 (+0.095xl07) Neuraminidase removable s i a l i c acid: (n=18) jxg/1010 platelet 116 (+1.9) 137 (+3.3) 136 (+3.2) molecules/platelet 2.26xl07 (+0.037xl07) 2.67xl07 (+0.064xl07) 2.65xl07 (+0.062xl07) Percent of total s i a l i c acid removed by neuraminidase 56.6% 67.5% 66.7% - 102 -Table 6 shows how the decrease in platelet electrophoretic mobility after neuraminidase treatment translates into molecules of s i a l i c acid removed. By comparing the electrophoresis data to the chemical analysis data (Table 7), i t is obvious that only 2.9% of the neuraminidase susceptible s i a l i c acid on the surfaces of the discocyte and spherocyte contributed to the platelet electro-kinetics. An opportunity is therefore presented to estimate the surface area of the echinocyte. Assuming the 2.9% ratio to be true for the echinocyte as well, then the number of s i a l i c acid molecules removed from the plane of shear per echinocyte can be calculated (line (a) in Table 7). This is found to be 7.74x10 (2.67x10 x 0.029). Since the number of molecules removed per 2 . 12 cm at the plane of shear is known (3.52x10 ), the echinocyte surface area is calculated to be 22.0xl0~8cm2 (7.74xl0 5 ± 3.5xl0 1 2). 3.3.3 Platelet Surface Phosphate Groups The time-digest curve for alkaline phosphatase is shown in Fig. 21. It seems only 15 min is required to liberate the phosphate groups from the fixed discocytes. Therefore the phosphate experiments were carried out with 45 min (3x15 min) of incubation with the enzyme. The amounts of phosphate terminal groups released by alkaline phosphatase as determined chemically and electro-phoretically are shown in Table 8 and Table 9 respectively. The percentage of phosphate groups at the plane of shear as determined electrophoretically is about 3 .9% of the total amount determined chemically (Table 10). Again i f one takes the percentage also to be true for the echinocyte, then i t s area can be —8 2 back calculated as before. This calculation gave 25.3x10 cm , only slightly higher than the area found from the s i a l i c acid calculations. An —8 2 average of the two (23.7x10 cm ) is therefore taken. - 103 -Table 6. Platelet Surface Sialic Acid Determined from the Decrease in Electrophoretic Mobility after Neuraminidase Treatment Discocyte Echinocyte Spherocyte EPM:glutaraldehyde fixed platelet (xi0 4cm 2.sec" 1.V~ 1) -1.41 (+0.020) -1.21 (+0.008) -1.00 (+0.100) EPM:neuraminidase treated fixed platelet (xl0 4cm 2.sec _ 1.V - 1) -0.86 (+0.010) -0.72 (+0.010) -0.70 (+0.007) EPM decrease (xl0 4cm 2.sec _ 1.V _ 1) 0.55 (+0.014)* 0.49 (+0.005)* 0.29 (+0.004)* Decrease in charge density (esu/cm2) 1.91xl03 (+0.047xl03) 1.69xl03 (+0.016xl03) 1.02xl03 (+0.015xl03) Siali c acid molecules removed per cmi 3.98xl0 1 2 (+0.098xl012) 3.52xl0 1 2 (+0.033xl012) 2.12xl0 1 2 (+0.031xl012) Siali c acid molecules removed per platelet 6.53xl05 (+0.161xl05) - 7.80xl05 (+0.114xl05) Electrophoretic mobility (EPM) values are means from 10 series of paired experiments. *Combined standard deviation (same as in Table 4) - 104 -Table 7. Ratio of Platelet Surface Sialic Acid Removed by Neuraminidase as Determined by Microelectrophoresis to That Determined Chemically Discocyte Echinocyte Spherocyte Surface area (cm2) 16.4xl0 - 8 36.7xl0 - 8 Electrophoresis data: Siali c acid molecules removed per cm2 3.98xl0 1 2 3.52xl0 1 2 2.12xl0 1 2 S i a l i c acid molecules removed per platelet (a) 6.53xl05 — 7.80xl05 Chemistry data: Sialic acid molecules removed per platelet (b) 2.26xl07 2.67xl07 2.65xl07 Ratio (a)/(b) 0.0288 - 0.0294 - 105 -Fig. 21. The time-release curve for the digestion of phos-phate groups from the surface of fixed platelet discocytes g by alkaline phosphatase. 2.5x10 /ml of platelets were incubated with 0.01 mg/ml of the enzyme. Values are means from three series of experiments. Error bars represent one standard deviation. - 106 -£]3Pld J e d p9SD6]ey s d n c u g epqdsoiy - 107 -Table 8. Platelet Surface Phosphate Groups Released After Alkaline Phosphatase Treatment Determined Chemically Discocyte Echinocyte Spherocyte Phosphate groups/ platelet 9.44xl06 (+0.199xl06) 12.50xl06 (+0.184xl06) 18.51xl06 (+0.130xl06) n=5 - 108 -Table 9. Platelet Surface Phosphate Groups Determined from the Decrease in Electrophoretic Mobility after Alkaline Phosphatase Treatment Discocyte Echinocyte Spherocyte EPM:glutaraldehyde fixed platelet (xl0 4cm 2.sec- 1.V - 1) -1.41 (+0.020) -1.21 (+0.008) -1.00 (+0.010) EPM:alk. phosphatase treated fixed platelet (xl0 4cm 2.sec _ 1.V - 1) -1.10 (+0.005) -0.94 (+0.008) -0.73 (+0.009) EPM decrease (xl0 4cm 2.sec - 1.V" 1) 0.31 (+0.015)* 0.27 (+0.008)* 0.27 (+0.009)* Decrease in charge density (esu/cm2) 1.09xl03 (+0.052xl03) 0.94xl03 (+0.027xl03) 0.94xl03 (+0.032xl03) Phosphate groups removed per cm2 2.28xl0 1 2 (+0.108xl012) 1.96xl0 1 2 (+0.056xl012) 1.96xl0 1 2 (+0.067xl012) Phosphate groups removed per platelet 3.74xl05 (+0.177xl05) - 7.19xl05 (+0.245xl05) Electrophoretic mobility (EPM) values are means from 10 series of paired experiments. *Combined standard deviation (same as in Table 4) - 109 -Table 10. Ratio of Platelet Surface Phosphate Groups Removed by Alkaline Phosphatase as Determined by Microelectrophoresis to Those Determined Chemically Discocyte Echinocyte Spherocyte Surface area (cm2) 16.4xl0 - 8 36.7xl0 - 8 Electrophoresis data: Phosphate groups removed per cm2 2.28xl0 1 2 1.96xl0 1 2 1.06xl0 1 2 Phosphate groups removed per platelet (a) 3.74xl05 - 7.19xl05 Chemistry data: Phosphate groups 9.44xl06 12.50xl06 18.51xl06 removed per platelet (b) Ratio (a)/(b) 0.0396 0.0389 - 110 -3.3.4 Cocaine Spheres Non-fixed cocaine spheres were treated with neuraminidase and alkaline phosphatase. The amounts of s i a l i c acid and phosphate groups liberated as determined chemically from the supernatant and determined from the reductions in electrophoretic mobility after enzyme treatment are shown in Table 11. Since the electrokinetic properties of the cocaine sphere are much different from the other forms of platelets (Table 3), no attempt was made to compare the fixed cocaine spheres to the other forms of fixed platelet. 3.3.5 Calcium and Magnesium Ion Binding Fig. 22 shows how the electrophoretic mobility of platelets decreases with increasing divalent ion concentration. Fig. 23 is a plot of the double reciprocal relationship between the drop in negative charge density and divalent cation concentration according to the formula given in the Introduction. The number of binding sites calculated from the intercepts in Fig. 23 are shown in Table 12 for calcium ions and Table 13 for magnesium ions. There are twelve times more Ca and eight times more Mg binding sites on the echinocyte than on the discocyte. On the other hand, the amount ++ of Ca binding to the spherocyte is 61% lower than to the discocyte. ++ Statistical comparisons of the binding density and free energy of Ca and Mg + + adsorption to the discocyte, echinocyte and spherocyte are summarized in Table 14. It is evident that the binding densities and free energies of ++ ++ Ca and Mg to the echinocyte are significantly different from those to ++ ++ . . the discocyte. The binding of Ca and of Mg to the discocyte is not significantly different and the same applies for the echinocyte. - I l l -Table 11. S i a l i c A c i d and Phosphate Groups Re leased by Enzymes from Coca ine Sphere* Determined E l e c t r o p h o r e t i c a l l y and Chemica l l y Con t r o l Neuraminidase treatment Ak l a l i n e Phos phat as e t reatment E l e c t r o p h o r e s i s da ta : EPM ( 1 0 4 c m 2 . s e c _ 1 . V - 1 ) n=8 -1 .40 (+0.017) -0.67 (+0.012) -1.00 (+0.018) Char ge dens i t y (esu/ cm 2) 4. 87x10 3 (+0. 058x1c 3 ) 2. 34x10 3 (+0 . 041x10 3) 3.48x10 3 (+0.062x10 3) Number of charges per cm 2 10 .14x10 1 2 (+0 .121x l0 1 2 ) 4 . 88x10 1 2 (+0 . 085x10 1 2 ) 7. 2 5 x10 1 2 (+0 .129x l0 1 2 ) Number of charges per p i at e l et 2.68x10 6 (+0.032x10 6) 1.29x10 6 (+0.022x10 6) 1.91x10 6 (+0.03xl0 6 ) Drop i n number of charges per p l a t e l e t (a) 1.39x106 0. 76x10 6 Chemistry da ta : S i a l i c a c i d molecules r e l e a s e d per p l a t e l e t (b) (n=4) 2. 69x107 (+0. 031x107) Phosphate groups re l ea sed per p l a t e l e t (b) (n=4) 1.27x107 (+0 . 02 3x10 7 ) Ra t i o (a)/(b) 0. 0515 0.0601 *Area=2 6. 4 x 1 0 - 8 c m 2 - 112 -Fig. 22. Influence of calcium and magnesium ion concentration on the electrophoretic mobility of the platelet discocyte, echinocyte, and spherocyte. Each data point represent a mean of 100 mobility determinations. Error bar represent one standard deviation. i - 113 -1-6--1-5--1-1-- 1 0 --0-9-CO -0-8-. Z> -0 7-e -0-6-o> v • X J -1-1-z : - 10 -Q _ -0-9-l±J -0-8-- 0 7 --0-6-spherocyte discocyte echinocyte i i I I I discocyte ^eclinccyte i l l I I I i i i i i 0 2 4 6 8 10 12 14 16 18 20 Cation Cone. (mM) - 114 -Fig. 23. The double reciprocal plots to find the densities of calcium and magnesium ion binding sites on the discocyte and echinocyte. Error bars represent one standard deviation, [cat. ] concentration of either Ca + + or Mg + + - 116 -Fig. 23 (continued). The double reciprocal plot to find the density of calcium ion binding sites on the spherocyte. Error bars represent one standard deviation, [cat. ]concentration of Ca - 117 -- 118 -Table 12. Calcium Ion Binding Sites Discocyte Echinocyte Spherocyte Intercept on ordinate (esu/cm 2) -! 8.15xl0 - 4 (+6.8xl0 _ 4) 9.50xl0~5 (+2.07xl0 - 4) 4.69xl0 - 3 (+1.5xl0~3) Slope 4.54xl0 - 5 (+8.45xl0 - 6) 4.03xl0~5 (+2.24xl0 - 6) 2.43xl0 - 4 (+4.06xl0- 5) Correlation coefficient 0.88 0.98 0.95 Binding sites: Density (cm - 2) 1.28xl0 1 2 (+1.07xl0 1 2) 1.09xl0 1 3 (+2.39xl0 1 3) 2.22X10 1 1 (+7.17xl01 0) Per platelet 2.10xl05 2.58xl0 6* 8.16xl04 -AG (Kcal/mole) 4.09 (+0.11) 2.89 (+0.03) 4.13 (+0.10) *Calculated using the surface area of 23.7x10 "cm . - 119 -Table 13. Magnesium Ion Binding Sites Discocyte Echinocyte Intercept on ordinate (esu/cm 2) -! 9.66xl0 - 4 (+3.49xl0-4) 1.58xl0~4 (+2.55xl0-4) Slope 4.03xl0"5 (+5.32xl0 - 6) 3.57xl0 - 5 (+3.38xl0-6) Correlation coefficient 0.97 0.98 Binding sites: Density (cm - 2) 1.08xl0 1 2 (+3.89xl0 n) 6.60xl0 1 2 (+1.07xl013) Per platelet 1.77xl05 1.56xl06* -AG (Kcal/mole) 4.26 (+0.08) 3.26 (+0.06) *Calculated using the surface area of 23.7x10 °cm% - 120 -Table 14. Statistical Ccmparison of Ca + + and Mg + + Binding to the Discocyte, Echinocyte and Spherocyte (Sig. = significant differences; not=no significant difference) Binding site density Free energy ( G) Ca + +, discocyte vs echinocyte sig. , 0.025 p 0.05 sig-, 0.0005 p 0.001 Ca + +, discocyte vs spherocyte sig. , p 0.005 sig., 0.01 p 0.02 5 Ca + +, echinocyte vs spherocyte sig. , p 0.005 sig-, 0.005 p 0.01 Mg++, discocyte vs echinocyte sig. , 0. 005 p 0. 01 sig-, p 0.0005 discocyte, Ca + + vs Mg + + not, 0.50 p 0.75 not., 0. 50 p 0. 75 echinocyte, Ca + + vs Mg + + not, 0.50 p 0.75 not., 0.50 p 0.75 Statistical analysis using Analysis of Variance Tables and F-dis t r i but ion according to Kalbfleisch (1974). See also the appendix. - 121 -3.4 DISCUSSION 3.4.1 Platelet Electrophoretic Mobilities The electrophoretic mobility of the human discocyte in Tyrode's solution -4 2 - 1 - 1 was found to be -1.08x10 cm .sec .V (non-fixed). This is similar to the mobility of the fresh human RBC. A review of the literature shows a variety of values for human platelet mobility. They are -0.635 (Coller and Zarrabi, 1981), -0.71 (Boisseau et a l . , 1971), -0.806 (Kosztolanyi et a l . , 1980), -1.07 (Hampton and Mitchell, 1974) and -1.17xl0~ 4cm 2.sec _ 1.V - 1 (Seaman and Vassar, 1966) for platelets measured in platelet-rich plasma and -0.84 (Mehrishi, 1970), -0.85 (Seaman and Vassar, 1966), -0.86 (Kosztolanyi et a l . , 1980), -1.08 (Bray and Alexander, 1969), -1.09 (Bosmann, 1972), -1.1 (Ross and Ebert, 1959), -1.14 (Madoff et a l . , 1964), -1.25 (Kirschmann et a l . , 1959), -1.4 (Yamazaki et a l . , 1980) and -1.58xl0~4cm2. sec _ 1.V _ 1 (Shimizu e_t ajL., 1979) for washed platelets measured in simple buffered saline-_4 2 - 1 - 1 type media. Our result of -1.08x10 cm .sec .V for washed platelets is similar to the ones obtained by Bray and Alexander and Bosmann. The great range of mobility values probably reflects differences in the electrophoretic systems involved as well as the conditions of the platelet samples (this point is to be discussed further below). -4 2 -1 -1 An electrophoretic mobility of -1.08x10 cm .sec .V translates 3 2 6 into a charge density of 3.74x10 esu/cm or 1.28x10 net negative charge groups per discocyte (Table 3). The number of net negative charge groups is the actual (total) number of negative charge groups minus the positive charge groups. After stimulation by 2x10 ~*M ADP, there was a 13.7% drop in mobility -4 2 - 1 - 1 from -1.08 to -0.93x10 cm .sec .V . The net negative charge density - 122 -3 3 2 decreased correspondingly from 3.74x10 to 3.23x10 esu/cm (Table 3). As discussed earlier, Seaman and Vassar (1966) reported an 18% drop in human platelet electrophoretic mobility after addition of lyo-g/ml (2.3x10 M^) of ADP. The decrease was concentration dependent and the reduction was up to 40% -4 at 100^jug/ml (2.3x10 M) of ADP. However, they failed to confirm the biphasic mobility behaviour reported by Hampton and Mitchell (1966), Stoltz (1970) and Kosztolanyi et a l . (1980). These workers reported that at low ADP concentrations, the platelet mobility increased up to a concentration of -2 -7 5x10 jxg/ml (1x10 M) after which the mobility decreased with concen-tration. However the maximum mobility increase at 1x10 M^ was only about 8% in a l l three reports. Confirming Seaman (1976) we were able to observe only the decrease in electrophoretic mobility. At 1x10 M^ ADP concentration we were not able to detect any change in platelet mobility. Takano and Suzuki (1981) working on rabbit platelets recorded only decreases in mobility with increasing ADP concentration. Finally, Boisseau a l . (1977) reported an —6 increase in platelet mobility of 10% at 2x10 M ADP concentration. Interestingly, platelet electrophoretic mobilities were a l l measured in platelet-rich plasma in the above papers. This work is the f i r s t to report a change in the mobility of washed platelets induced by ADP. Seaman and Vassar (1966) failed to find any change in the mobility of washed platelets in saline after exposure to ADP. The mobility -0.85 m.sec \v ''.cm for their washed platelets was very low, in fact close to the mobility of our echinocyte. They washed their platelets simply by centrifuging them down in saline. The d i f f i -culties encountered in washing platelets have already been discussed in Chapter 2. By repeating of the Seaman and Vassar washing procedure by us produced platelet aggregates of various sizes. Therefore i t is possible that the above - 123 -authors were actually measuring the mobility of small platelet aggregates or echinocytes. This could well be the reason why their washed platelets would not respond to ADP. The great variety of published platelet mobilities mentioned above may also in part be caused by different platelet handling procedures. Parallel microscopic monitoring of platelet morphology similar to that done here is therefore considered essential. Handling and transformation (to echinocyte and spherocyte) protocols were set up and standardized with the help of electron microscopy and a l l platelet samples were examined under phase microscopy before electrophoresis. _4 2 -1 -1 The spherocyte had an electrophoretic mobility of -1.55x10 cm .sec .V 3 measured in hypotonic Tyrode's solution. The charge density was 2.60x10 2 esu/cm (Table 3). Presumably due to the increase in surface area, the number of charge groups per spherocyte increased by 55%. The hypotonic Tyrode's solution has an osmolality of 75 mOsmol and an ionic strength of 0.0379. The relationship between mobility and ionic strength is an inverse one (see equation in Introduction). Kirschmann et a l . (1959) have examined the effect of low ionic strength on the electrokinetic properties of human platelets. As ionic strength was decreased the mobility increased. They found that the change in mobility was totally reversible. If platelets were suspended in 0.015 ionic strength medium and then resuspended back in an iso-ionic strength medium, the mobility returned to normal. We made similar observations. Discocytes suspended in a 0.04 ionic strength -4 2 isotonic Tyrode's solution had a mobility of about -2.08x10 cm . sec "^.V ^ and after resuspension in regular Tyrode's solution the mobility -4 2 -1 -1 was returned back to -1.08x10 cm .sec .V . No shape change was observed in the discocyte due to change in ionic strength. - 124 -The pH-mobility curves are shown in Fig. 18 and 19. It was hoped that information concerning platelet surface ionogenic groups could be obtained by constructing these curves. However, as a result of the problems encountered at low pH already discussed, the curves cannot be extrapolated below pH 4.5 and values such as pKa for the platelet surfaces cannot be deduced. The mobilities of glutaraldehyde fixed platelets were about 10% higher than the mobilities of formaldehyde fixed platelets. This phenomenon has also been found in red blood cells (Vassar et a l . , 1972) and lymphocytes (Vassar et a l . , 1973). The reason for the apparent additional charge remains unclear. Because of i t s fast action (Vassar et a l . , 1972) glutaraldehyde remained the fixative of choice for microscopic and enzyme digestion studies. Since the 10% increase occurred similarly to the discocyte, echinocyte and spherocyte no correction is required for the enzyme digestion studies. Acetaldehyde fixation tends to agglutinate platelets during fixation. To study the amino groups on platelet surfaces formaldehyde fixation was chosen. The reaction of formaldehyde with cellular components to induce fixation is complex but presumably formaldehyde interacts with amino groups in the following way: R-NH3 + H2C=0 > R'^0H * R - N = C H 2 + H 2 ° The positive charge is thus eliminated (Jentoft and Dearborn, 1979). To test the possibility that reversal of the fixation process could have occurred the platelets were further treated with borohydride (Jentoft and Dearborn, 1979): R-N=CH2^^R-N-CH; - 125 -A stable neutral species is then obtained. If there were significant reversal of the fixation process, the borohydride treated fixed platelets ought to have had a greater increase in mobility than the non-treated ones. We found no differences, indicating no reversal of fixation had occurred. The value obtained here for the number of amino groups per platelet was 2.18x10"* (Table 4) which compares well with those of Stoltz and Nicolas (1979), 3xl0 5, and Mehrishi (1970), 2.2 to 2.5xl0 5. The amino group surface density on the echinocyte is 10.5% lower and on the spherocyte 18.8% lower than on the discocyte. 3.4.2 Platelet Surface Sialic Acid By acid hydrolysis one can find the total s i a l i c acid content in the plate-lets. Mild hydrolysis with l^ESO^ released s i a l i c acid from the surface as well as from within the fixed platelets. The released s i a l i c acid was then assayed. The total s i a l i c acid content in the discocyte was found to be 205 yug/10^^ platelets or 3.99x10^ molecules/platelet. The discocyte, echino-cyte and spherocyte a l l had similar total s i a l i c acid contents (Table 5) showing that during the transformations no s i a l i c acid is lost. Both Motamed et a l . (1976) and Martin et a l . (1982) found 207 ^ig of s i a l i c acid per 10 1 0 platelets by acid hydrolysis. Ku and Wu (1977) found 185 and Stoltz and Nicolas (1979) found about 197yug/10 1 0 platelets. Neuraminidase (Vibrio cholerae) removed 56.6% of the total s i a l i c acid from the d i s c o c y t e . Higher percentages, 67.5% and 66.7% of t o t a l were removed from the ech inocyte and spherocyte r e s p e c t i v e l y (Table 5 ) . In other words, neuraminidase s u s cep t i b l e s i a l i c a c i d goes from 116 yxg/10^ d i s cocy te s to 137yxg/10^ ech inocytes and 136 jig/10^ spherocytes. Ku and Wu (1977) - 126 -found that 48% of the total platelet s i a l i c acid is released by neuraminidase and after ADP (5x10 M^) stimulation the percentage increased to 65%. Motamed et a l . (1976) found 40% for the unstimulated platelets and about 49% for ADP stimulated ones. Peerschke et a l . (1978), on the other hand, found no difference between the amount of s i a l i c acid removed by the enzyme from the surface of unstimulated and ADP stimulated platelets. No explanation was offered by them for the discrepancy. Neuraminidase (Vibrio cholerae) has a molecular weight of 90,000 (Gottschalk and Bhargava, 1971; Behring, 1979). From sedimentation and diffusion coefficients i t has an estimated radius of 32A (Pye and Curtain, 1961). It is the belief of most workers that i t w i l l not gain access to the interior of the platelet and i t s enzymatic action is s t r i c t l y on the platelet surface (Stoltz and Nicholas, 1979; Packham et^  _al., 1980; Coller and Zarrabi, 1981). It is our hypothesis that an enzyme of this size w i l l not be able to penetrate into the surface-connected canalicular system. Electron microscopic studies by Behnke (1968) showed that horseradish peroxidase, which has a molecular weight of 40,000 (Worthington, 1972) did stain the inside surface of some of the canalicular system. White (1970) made a more detailed study of the system by using two electron dense tracers of different sizes. Lanthanum o nitrate (forming 25A diameter particles) penetrates easily into the surface-° connected canalicular system. On the other hand, ruthenium red (forming 50A diameter particles) stains just the outside of platelets and only penetrates "superficially" into the canalicular system. After ADP stimulation, ruthenium red readily penetrates into the canalicular system demonstrating an apparent opening up process. White (1980) further showed that the openings of the - 127 -the channels at the discocyte surface are rather constrictive (see Fig. 3) and may not be as simple as once thought. Our results show quantitative differences between the discocyte and the echinocyte as well as between the discocyte and the spherocyte. The additional s i a l i c acid susceptible to neuraminidase most probably comes from evaginated or newly exposed membrane surfaces. Given white's results and the increase in surface area believed to occur upon echinocyte transformation (Frojmovic and Milton, 1982) i t seems unlikely that neuraminidase penetrates into the discocyte canalicular system. The same argument applies to the enzyme alkaline phosphatase whose molecular weight is considerably larger at 140,000 (Frenette, 1980). Electrophoretic measurements of fixed platelet discocytes showed a 39% drop in mobility after neuraminidase treatment (Table 6). Madoff et a l . (1964) found a 42% decrease while Seaman and Vassar (1966) and Bosmann (1972) found 53% reductions. The decrease in mobility was 40% for the echinocyte and only 29% for the spherocyte (Table 6). This reflects the much lower s i a l i c acid charge density on the spherocyte. Only 2.9% of the neuraminidase susceptible s i a l i c acid on the platelet surface is apparently responsible for the electrokinetic charge properties of the platelet (both discocyte and spherocyte). This figure derives from the 30 fold difference between the amount of s i a l i c acid assayed chemically in the enzyme digest supernatant and that calculated from electrophoretic mobility measurements. Stoltz and Nicolas (1979) found a figure of 3.2%. The percentage is higher in polymorphonuclear leukocytes (10%) and considerably higher s t i l l for red blood cells (46%) (Cook et a l . , 1961 and Vassar et a l . , 1969). If we assume the percentage for the echinocyte to be also 2.9%, then - 128 -the surface area of the echinocyte can be back calculated from the chemistry -8 2 data as described above (Table 7), giving the value of 22.0x10 cm . 3.4.3 Platelet Surface Phosphate Groups Alkaline phosphatase removed 9.44x10 phosphate groups from each fixed discocyte (Table 8). This value is double that of 4.2x10 found by Bik et al . (1982). The enzyme liberates 32% and 96% more phosphate from the surfaces of the echinocyte and spherocyte respectively. The electrophoretic mobility of the discocyte decreased by 22% after enzyme digestion corresponding to a removal of 3.74x10"' phosphate groups from the discocyte plane of shear (Table 9). Mehrishi (1970) found a drop in mobility of between 30 and 37%, corresponding to a removal of 5x10^ groups per platelet. Stoltz (1975) estimated platelet surface phosphate groups to be between 3 and 5x10^ per platelet from microelectrophoresis experiments. Again, by comparing electrophoresis data with chemistry data (Table 10) i t was found that only 3.9% of the phosphate groups removed are apparently responsible for the electrokinetic charge contribution of the discocyte and spherocyte. If i t is assumed that the same percentage applies to the echino-cyte i t s surface area can again be calculated. The value based on phosphatase —8 2 —8 2 digestion data was 25.3x10 cm , quite close to 22.0x10 cm obtained —8 2 from the s i a l i c acid calculations. An average of 23.7x10 cm was therefore taken to be the estimate of the echinocyte surface area accessible to these two enzymes. —8 2 The established area of 23.7x10 cm is larger than that of the disco-cyte but much smaller than that of the spherocyte (Table 1). Presuming that the newly-exposed membrane originates from the surface-connected canalicular - 129 -system, i t is apparent that i t could not be totally evaginated in the echino-cyte. This conclusion is consistent with the morphological evidence for a partially intact canalicular system presented in Chapter 2. In a recent review, Frojmovic and Milton (1982) made an estimate of the surface area of an echinocyte based on microscopic observations. By assuming 2 an echinocyte to have a central spherical body of 13jum with thin cylindrical pseudopods extending out of the central body, they arrived at an area of between 17.7 and 22.4x10 cm . Their pseudopods had a mean radius of 0.075 jdm and total length of between 10 and 20 y.m. The closeness of this estimation to the one we obtained via surface chemical means is remarkable. 3.4.4 Cocaine Spheres Incubating the platelet discocyte with 10 mM of cocaine for half an hour _g 2 produces a "cocaine sphere" with a surface area of 26.4x10 cm . The -4 2 - 1 - 1 electrophoretic mobility increases dramatically from -1.08x10 cm .sec .V -4 2 - 1 - 1 for the discocyte to -1.40x10 cm .sec .V for the cocaine sphere. This 2 corresponds to an apparent 30% increase in charge density (charge per cm ) and a 108% increase in the number of charges per platelet (Table 3) at the plane of shear. Having a surface area slightly larger than that of the echino-cyte, i t has 68% more charge than the echinocyte. This change cannot be attributed to increase in surface area alone (see below). Treatment of the cocaine spheres (unfixed) with neuraminidase resulted in a 52.6% reduction in mobility representing 1.39x10^ s i a l i c acid groups per platelet removed from the plane of shear. This is 66.4% more than in the fixed echinocytes. The chemistry data, on the other hand, shows that the total amount of s i a l i c acid removed from the platelet surface is about the same for the fixed echinocyte and unfixed cocaine sphere. In other words, the - 130 -percentage of s i a l i c acid groups on the platelet surfce situated at the plane of shear whose loss is seen electrophoretically, has increased from 2.9% to 5.1%. Similarly the percentage for the phosphate groups went up from 3.9% to 6.0% (Table 11). Since the total enzyme susceptible s i a l i c acid and phosphate groups on the surfaces of the cocaine sphere and the echinocyte are similar i t is reasonable to believe the total number of surface molecules bearing these groups is similar in the two forms. The change in surface architecture associated with cocaine binding apparently allows a larger fraction of the s i a l i c acid and phosphate groups to contribute to the mobility, hence the increase in surface charge. Cocaine has the molecular structure: 0 Its mechanism as a local anesthetic is s t i l l unclear, however i t is commonly believed to act on the membrane of nerve cells interfering with their function (Papahadjopoulos, 1972; Mather and Cousins, 1979). The lipophilic benzene ring buries i t s e l f into the membrane l i p i d layer and the hydrophilic end remains in solution. It causes expansion of the membrane and conformation changes (Mather and Cousins, 1979). Although cocaine has a positively charged tertiary amine group, i t is doubtful that this w i l l contribute to the electrokinetic properties of the platelet at the plane of shear. It is more - 131 -likely to interact with the negative charges of the phospholipid head groups (Papahadjopoulos , 1972). I t is thought that this molecule's unique position i n the c e l l membrane may somehow interfere with Na+/K+ fluxes and the transmission of nerve signals. Cocaine and related local anesthetics also transform biconcave erythrocytes into cup-shaped forms by disrupting the inside/outside ionic environment of the erythrocytes (Deuticke, 1968). By displacing Ca + + from anionic sites on the membrane and i n the cytoplasm, cocaine and related anesthetics also cause the breakdown of microfilaments, microtubules and cytoskeleton i n cells (Pos te jat jal., 1975 ; Nicolson et a l . , 1976) and platelets (Nachmias et al ., 1977 ; Nachmias et al ., 1979; Davies and Palek, 1982). No matter which i s the mechanism for the sphering of the platelet, the change i n surface electrokinetic properties is overwhelming. 3. 4. 5 Calcium and Magnesium Ion Binding It was estimated that 2.10x10"' Ca + + binding sites are present on the elctrokinetic surface of each discocyte, assuming that equation [3] and [5] model the situation adequately. ADP stimulation of the discocyte increases this figure twelve-fold to 2.58x10^ sites. The density of binding sites 12 2 13 increases eight-fold from 1.28x10 per cm on the discocyte to 1.09x10 2 per cm on the echinocyte. The average binding free energy ( G) at the same time drops from -4.09 to -2.89 Kcal/mole (Table 12) . Using similar procedures ++ 1 2 Seaman et al. (1969) found the Ca binding site density to be 4.97x10 2 12 2 per cm for the human erythrocyte and 3.84x10 per cm for the poly-morphonuclear leukocyte (Table 15). Their binding free energies were -3.8 and -4.0 Kcal/mole respectively. The density of Ca + + binding sites on the discocyte is therefore comparatively low, although the average binding free - 132 -energy i s similar to those of the erythrocyte and leukocyte. After transformation into the echinocyte the density becomes much higher than on the RBC and leukocyte but the binding i s less strong. The estimated distance between adjacent sites decreases from about 9nm to 3nm after echinocyte transformation. Data obtained for Mg + + binding were not significantly different from those of Ca + + although the number of sites i s somewhat lower (Table 13 and 14). After transformation of discocytes into spherocytes by hypotonic shock the density of the Ca + + binding sites decreased by over 80% representing apparently a drastic change in platelet membrane properties. Using the ^ C a + + binding techniques, Taylor and Heptins t a l l (1980) found 2.20xl05, Brass and Shattil (1982) found 4.85xl05, and Peerschke et al. (1980) found 1.66xl0 6 sites per discocyte. Peerschke ej: _al. (1980) 45 ++ found no increase i n Ca binding after ADP stimulation while Brass and Shattil (1982) found a 28% increase. The range of values shows the . . 45 ++ susceptibility of the Ca experiments to calcium internalization. However the microelectrophoresis method used here also has some drawbacks. It only estimates Ca + + and Mg + + binding to the electrokinetic surface of the platelets. It i s therefore l i k e l y to underestimate the total C a + + and Mg + + binding to the whole plasma membrane surface. The estimates obtained from microelectrophoresis depend on an idealized model. They ignore, by depending on equation E3], the distributed nature of the surface charge throughout the glycocalyx. Despite these uncertainties, differences i n the binding free energy and the number of binding sites on various platelet forms are much less l i k e l y to be in error than the individual absolute values. 45 ++ Given their limitations, comparisons of the Ca and the electrophoresis results are d i f f i c u l t to interpret. - 133 -Table 15. Comparison of Calcium Ion Binding on Platelets, Erythrocytes and Polymorphonuclear Leukocyte Density of Binding Sites (cm - 2) - G (Kcal/mole) Discocyte Echinocyte Spherocyte 1.28xl0 1 2 1.09xl0 1 3 2.22X10 1 1 4.09 2.89 4.13 RBC* PMN* 4 . 9 7 x l 0 1 2 3 . 8 4 x l 0 1 2 3.82 4.01 *From Seaman et a l . (1969) - 134 -The necessity for fibrinogen binding to precede platelet aggregation and the requirement of Ca + + and Mg + + for fibrinogen binding to the surface of the activated platelet has already been discussed in previous chapters and will be discussed further i n the next chapter. It has been estimated that between 40,000 (Bennett and Valarie, 1979; Plow and Marguerie, 1980) and 50,000 (Hawiger jat ^ 1., 1980) fibrinogen molecules bind specifically to surface of each echinocyte. This in effect gives a binding ratio of about 50 to 60 Ca + + per fibrinogen molecule. The fibrinogen molecule has been estimated to be about 45nm long and 6. 5nm in diameter (Weisel et _al., 1981). The maximum area per Ca + + binding site on the echinocyte surface i s about 2 . 13 2 9nm ( i . e . the reciprocal of the sites density of 1.09x10 /cm ). Therefore on the average only a maximum of 30 Ca + + could be associated with each fibrinogen molecule. From this estimatin about half of the Ca + + on the echinocyte could be associated with fibrinogen providing fibrinogen molecules maximize thier area of contact with the surface and the Ca + + binding sites are uniformly distributed. The above discussion i s based on the assumption that a l l the Ca + + binding sites are saturated. Obviously, under physiological conditions this i s not the case. The optimum Ca + + concentration for platelet aggregation i s 4mM (Chapter 2). One can back calculate using Fig. 23 and equation [5] that only 5. 7x10^ Ca + + ions bind to each echinocyte at this concentration. Only about 2% of the binding sites are therefore occupied. This also gives a relationship of 1 to 1 .4 Ca + + ion per fibrinogen molecule. - 135 -3.4.6 A Hypothesis Regarding Membrane Exposed by ADP Stimulation or Hypotonic  Swelling Using the above data, a table outlining the surface properties of the platelet discocyte, echinocyte and spherocyte as well as their "new membranes" can be constructed (Table 16). "New membrane" is defined as the increased or evaginated surface areas of the echinocyte and spherocyte. Total negative charge is the apparent net negative charge (Table 3) corrected by adding the number of positively charged amino groups (Table 4). The specific properties of the "new membrane" are calculated by assuming that the surface properties of the original "old membrane" of the platelet remain unchanged during shape transformation and that the additional charged groups which appear are distributed uniformly over the exposed area added during formation of the echinocyte or spherocyte. There seem to be only small differences between the discocyte and echinocyte membrane considering their respective s i a l i c acid/total charge, phosphate/total charge and s i a l i c acid/ phosphate ratios, although absolute quantitative differences are obvious. The spherocyte, and especially the spherocytic "new membrane," on the other hand, are quite different from the others both quantitatively and qualitatively. It was proposed at the beginning of this work that the newly exposed, evaginated membrane of the echinocyte might be responsible for the sticky nature of the platelet surface. In other words, the non-sticky platelet discocyte has a cloistered sticky surface that can be unfolded to produce a sticky platelet. The data presented in this chapter argue against this interpretation. Fir s t l y , the unfolding of the evaginated canalicular membrane system using hypotonic shock produces a spherocyte with completely Table 16a. Platelet Surface Properties by Microelectrophoresis Spherocyte Echinocyte Discocyte Spherocyte New Membrane Echinocyte New Membrane Area (cm2) 16.4xl0~8 36.7xl0"8 20.3xl0"8 23.7xl0"8 7 .3xl0~ 8 Total negative charge -per platelet 2 per cm^  1.50xl06 9.13x10 2.39xl06 6.50x10 0.89xl06 4.38x10 1.88xl06 7.92x10 0 5 .38xl0 6 .21x10 Amino groups -per platelet per cm2 2.18xl05 1.33x10 3.97xl05 1.08x10 1.79xl05 0.88x10 2.82x10!? 1.19x10 0 0 .64xl0 5 .88x10 Sialic acid groups -per platelet per cm2 6.53x10^ 3.98x10 7.80xl05 2.12x10 1.27x10^ 0.62x10 8.34xl05 3.52x10 1 2 .18xl0 5 .48x10 Phosphate groups -per platelet per cm2 3.74xl05 2.28x10 7.19xl05 1.96x10 3.45xl05 1.70x10 4.62x10^ 1.95x10 0 1 ,88xl0 5 .20x10 Sia l i c acid/Total charge Phosphate/Total charge Sia l i c acid/Phosphate 43.5% 25.0% 1.75 32.6% 30.1% 1.08 14.3% 38.8% 0.37 44.4% 24.6% 1.80 47.6% 23.1% 2.07 Table 16b. Platelet Surface Enzyme Released Si a l i c Acid and Phosphate Group (Chemistry Data) Spherocyte Echinocyte Discocyte Spherocyte New Membrane Echinocyte New Membrane Sia l i c acid groups -per platelet 2 . 2 6 x 1 0 2 . 6 5 x l 0 ? , 0.39xl0? 2 . 6 7 x 1 0 0 . 4 1 x l 0 ? , per cm2 1.38x10 0.72x10 0.19x10 1.13x10 0.56x10 Phosphate groups -per platelet 9.44xl06 18.52xl06 9.08x10^ 12.48xl06~ 3.04xl06 per cm2 5.75x10 5.04x10 '4.47x10 5.28x10 4.17x10 Sialic acid/Phosphate 2.39 1.43 0.43 2.14 1.34 - 138 -Table 16c. Platelet Surface Ca + + and Mg + + Binding Sites Echinocyte Discocyte Spherocyte Echinocyte New Membrane C a + + per platelet 2.10xl05 8.16xl04 2.58xl06 2.37xl06 per cm2 1.28xl0 1 2 2.22xlO n 1.09xl0 1 3 3.25xl0 1 3 Mg + + per platelet 1. 77xl0 5 per cm'- 1.08xl0 1 2 1.56xl06 1.38xl06 6.60xl0 1 2 1.89xl0 1 3 - 139 -different surface properties. Secondly, i t does not seem possible that the original discocyte surface ("old membrane") remains unchanged during echino-cyte transformation. This becomes most evident i f the platelet surface Ca + + binding data is considered (Table 16c). Under the sticky invaginated membrane model almost a l l of the Ca + + binding sites on the echinocyte surface must originate from the "new membrane." Yet during spherocyte transformation the total number of sites actually decreases. Not only do no contributions appear from the "new membrane", but in this case Ca + + binding sites are disappearing. This contradiction implies that the spherocyte membrane cannot consist of unchanged echinocyte membrane plus "new membrane." Rather, the average properties of the whole surface are likely to have changed during sphering. Finally, i t was mentioned earlier that about 40,000 fibrinogen molecules bind specifically onto the surface of an activated platelet. Considering the fibrinogen molecule as a cylindrical rod of 45nm long and 6.5nm in diameter (Weisel et al., 1981), 40,000 molecules w i l l cover about 11.7x10 cm i f the molecules l i e f l a t , about half of the total echinocyte surface area. If they are standing up on ends, the 40,000 fibrinogen molecules could occupy an —8 2 area as small as 1.3x10 cm . Since the surface area of the echinocyte _g 2 "new membrane" is about 7.3x10 cm , i t is possible, given a favourable orientation for the fibrinogen molecules, that only "new" membrane could be involved in the binding. However, the Ca binding site density in this region would be extremely high. It remains to be seen whether independent evidence, perhaps ultrastructural, in favour of this model w i l l appear. In conclusion, although i t appears that there are significant differences between the discocyte membrane and the "new membrane" of the echinocytes and spherocytes, i t is not clear that the "new membrane" is responsible for the stickiness of the echinocyte surface. The changes in Ca + +, Mg + + and - 140 -fibrinogen binding sites could just as well involve the whole platelet and not just the "new membrane." e - 141 -CHAPTER 4 RADIOCHEMICAL LABELLING OF PLATELET SURFACE STRUCTURES - 142 -4.1 INTRODUCTION Polyacrylamide gel electrophoresis (PAGE) of radiolabeled platelet membranes dissolved in sodium dodecyl sulphate (SDS) has been an important tool in the understanding of the surface chemistry of platelets. The gel electrophoresis patterns and their interpretations have been discussed and reviewed extensively elsewhere (Jenkins et ^al., 1979; Phillips, 1979; McGregor et a l . , 1979; Nurden et a l . , 1981; Connellan et a l . , 1982; Bowles and Brunton, 1982; Toor ej: a l . , 1982). Three major platelet membrane glycoproteins were originally found by radiolabelling techniques and SDS-PAGE (Phillips, 1979; Phillips, 1980; Okumura and Jamieson, 1976; George et aJL., 1980). They are glycoprotein I (GP-I) with an apparent molecular weight of between 130,000 and 160,000; glycoprotein II (GP-II) with a molecular weight of between 110,000 and 130,000 and glycoprotein III (GP-III) with a molecular weight of between 90,000 and 110,000. Subsequent work revealed another glycoprotein with a molecular weight of between 150,000 and 170,000. This glycoprotein ran very near the GP-I peak and was resolved only through improved SDS-PAGE techniques. This last glycoprotein was named GP-Ia and the original GP-I was then called GP-Ib. Careful SDS-PAGE experiments demonstrated more minor glycoprotein peaks below the GP-III peak: GP-IV (sometimes referred to as GP-IIIb) was found with a molecular weight of between 85,000 and 100,000; GP-V has a molecular weight of between 68,000 and 89,000. If one uses the nomenclature of GP-IIIb for GP-IV, then GP-V becomes GP-IV. A variety of nomenclature systems have been developed in different laboratories. Wide ranges of molecular weight have also been reported in the literatures for the glyco-proteins species, resulting in some d i f f i c u l t y and confusion in the inter-pretation of data. The nomenclature system using GP-Ia, GP-Ib, GP-II, GP-III, - 143 -GP-IV, GP-V and GP-VI generally referred to as the Phillips system (Phillips, 1979) w i l l be used throughout this work. -SDS-PAGE has proved useful for the separation of proteins according to their molecular weights (Reynold and Tanford, 1970; Lasky, 1978). The useful-ness of SDS-PAGE for molecular weight determination depends on the abi l i t y of the anionic detergent SDS to interact with and denature a wide variety of proteins in a similar manner. Native proteins having a wide difference in charge, size and shape characteristics are converted, upon disulfide bond reduction and SDS-binding, to SDS-protein complexes of their constituent polypeptide chains. The complex is a rod-like structure, the length of which varies with the molecular weight of the protein. One interesting aspect of this complex is that different proteins bind identical amounts of SDS on a gram per gram basis. The charge per unit mass of protein is therefore approximately constant. In other words, the charge densities of different SDS-protein complexes are about equal and the electrokinetic properties of the complex w i l l be mainly a function of the protein's molecular weight. The PAGE gel is formed by the cross-linking of acrylamide and bis-acrylamide to give a three-dimensional polymer meshwork. The degree of cross-linking depends on the gel (acrylamide) concentration. This three-dimensional meshwork forms a molecular sieve. During electrophoresis, the smaller proteins w i l l encounter less resistance from the gel matrix than the larger ones and hence migrate farther along the gel. When the distances migrated by different proteins are plotted vs. the logarithm of their molecular weights, i t is generally found that the proteins f a l l on a straight line. Ideally, the molecular weight of an unknown protein can be read off from such a calibration plot, providing identical experimental conditions are used. However, errors in determining - 144 -molecular weights can occur (Lasky, 1978). First, the semi-logarithmic calibration plot may not be linear, especially towards the higher molecular weight region of the plot (top of the gel). Second, the action of SDS may not be perfect (uniform) towards a l l proteins and errors of as much as 15% in the determination of molecular weight can occur (Lasky, 1978). Finally, l i t t l e is known about how the carbohydrate moieties of glycoproteins may affect the molecular weight determination. Three methods are commonly used for the radiolabelling of membrane glyco-proteins (Phillips, 1979 and McGregor et a l . , 1979). The f i r s t method is lactoperoxidase-catalyzed iodination which covalently attaches the radioactive iodine to tyrosine residues of the glycoproteins. This method gives a strong GP-III and a moderate GP-II peak on SDS-PAGE. The second method requires the removal of terminal s i a l i c acid residues from the surface glycoproteins with neuraminidase. Galactose oxidase is then used to oxidize the hydroxyl groups at the C-6 position of the exposed galactose residues into aldehyde groups. 3 Next, H-borohydride is used to reduce the aldehyde groups, resulting in the 3 incorporation of H into the galactose molecules. The third method uses periodate to cleave oxidatively the carbon-carbon bonds between C7-C8 and C8-C9 positions of the terminal s i a l i c acid residues. The aldehyde groups so 3 . . 3 formed are then reduced with H-borohydride resulting in H being incorporated into the s i a l i c acid molecules. The second and the third labelling methods give similar patterns on SDS-PAGE. GP-Ib has the strongest peak while peaks of GP-Ia, GP-II and GP-III are moderately strong. A l l three methods w i l l be used here to detect possible changes in platelet surface glycoproteins after transformation to echinocytes and spherocytes. - 145 -Bunting et_ _al. (1978) using the periodate-tritiation method found no differences between the gel patterns of normal and ADP activated platelets. 125 George «jt _al. (1980) using the diazotized I-diiodosulfanilic acid labelling technique found no difference. This last compound reacts with and labels amino groups on the platelet surface. Preliminary experiments by us 125 . . . using I - l act oper oxidase iodination also fa i l e d to demonstrate any difference between discocyte and echinocyte SDS-PAGE patterns. Therefore double label experiments were undertaken to try to improve the resolution. Sorg and Greczy (1976) have used double radiolabelling in conjunction with SDS-PAGE in lymphocyte studies. They have shown that this can be a valuable technique in detecting small differences between test and control experiments. 125 131 Double labelling with I and I was used with lactoperoxidase-catalyzed iodination. This was done in two ways. F i r s t l y , discocytes were 125 . 1 3 1 labelled with I and echinocytes or spherocytes were labelled with I. They were then mixed, dissolved i n SDS and run on the same gel. This method helps to eliminate gel to gel differences that may have masked small 125 differences i n the patterns. Secondly, I-labelled discocytes were 125 transformed into echinocytes or spherocytes. Then these I-labelled 131 131 echinocytes and spherocytes were relabelled with I. Any I peak i n 125 the gel represents new glycoprotein species exposed after I labelling. Controls were done by double labelling the discocytes without the transformation step. A similar double radio-iodination technique was used by Phillips and Agin (1974) to find the platelet surface thrombin proteolytic site. For the t r i t i a t i o n experiments, double labelling i s achieved by f i r s t "labelling" the discocytes with non-radioactive borohydride using the periodate or neuraminidase-galactose oxidase method. Then after transformation into - 146 -3 echinocytes or spherocytes, the platelets are relabelled with H-borohydride 3 using similar methods. Any H peak in the gel greater than that in the con-trol represents glycoproteins exposed during the transformations. These are termed pseudo-double label experiments. Lactoperoxidase iodination was also used to study the surface component involved in the binding of fibrinogen onto the echinocyte surface. The requirement for fibrinogen in ADP induced aggregation has been discussed earlier. Activation of platelets results in the exposure or appearance of receptors for fibrinogen. Fibrinogen molecules then bind onto the surface of the activated platelets. Calcium or magnesium ions are required for the binding (Chapter 2). The number of specific receptor sites per activated platelet ranges from about 40,000 (Bennett and Vilaire, 1979; Plow and Marguerie, 1980) to 50,000 (Hawiger et a_l., 1980). Peerschke et a l . (1980) found 4000 high affinity and 9000 low affinity sites. Recent studies suggest that the fibrinogen receptor is associated with the GPII-GPIII complex. Four lines of evidence point to this fact. F i r s t l y , patients with Glanzmann's thrombasthenia have platelets that f a i l to bind fibrinogen when activated. Glanzmann's thrombasthenia platelets lack the GPII-GPIII complex. Platelets from patients with Bernard-Soulier syndrome bind fibrinogen normally when activated. They have the GPII-GPIII complex but lack GPIb (Mustard et a l . , 1979; Bennett and Vilaire, 1979; Peerschke et a l . , 1980; Lee et a l . , 1981). Secondly, antibodies to the GPII-GPIII complex block fibrinogen binding (Lee et a l . , 1981; Coller, 1981; DiMinno et a l . , 1981). Thirdly, isolated GPII, GPIII and GPII-GPIII complex when adsorbed to plastic microtitre plates bind to fibrinogen (Nachman and Leung, 1982). The glyco-proteins were isolated using l e n t i l lectin affinity chromatography and were - 147 -found to react with mono-specific anti-GPIIa and I l l b antibodies. Fourthly, by using fibrinogen coupled to a photoreactive agent Peerschke et a l . , (1981) and Bennett e_t a_l. (1981) were able to demonstrate the association of fibrinogen with GPII and GPIII respectively. This agent (methyl-4-azidobenzoimidate) reacts with organic molecules covalently on exposure to light. It can therefore link fibrinogen with i t s receptor, or to structures near the receptor, when the experiment is done in darkness and then exposed to light. When fibrinogen is bound, a fibrinogen-protein supercomplex is formed and identified as a high molecular weight band on SDS-PAGE. The molecular weight of the putative receptor is the difference between the molecular weight of the supercomplex and that of fibrinogen. This method does not demonstrate a ligand-receptor relationship but only a close proximity between the fibrinogen molecule and GPII/III molecule. Morrison and Bayse (1970) have shown that the lactoperoxidase enzyme molecule must form a complex with tyrosine in a stereospecific manner before iodination can occur. It w i l l be shown here that the presence of fibrinogen interferes with the iodination of GPIII in the activated platelet, demon-strating a close relationship between the adsorbed fibrinogen and GPIII. - 148 -4.2 MATERIALS AND METHODS 4.2.1 Surface L a b e l l i n g of P l a t e l e t s 4.2.1.1 Lactoperoxidase Iodination Lactoperoxidase i o d i n a t i o n of p l a t e l e t s was c a r r i e d out according to the methods of P h i l l i p s (1972). P l a t e l e t s were i s o l a t e d and adjusted to 2.5x10° per ml with Tyrode's s o l u t i o n 125 as i n Chapter 2. They were mixed with 0.1 mCi/ml of c a r r i e r - f r e e Na I or 131 Na I and 0.1 mg/ml of lactoperoxidase. Hydrogen peroxide was then added at 0.5 mM f i n a l concentration. Reaction was almost instantaneous at room temperature and af t e r 2 min the p l a t e l e t s were washed free of excess radio-l a b e l . Spherocytes were l a b e l l e d i n hypotonic Tyrode's s o l u t i o n under the same conditions and reagent concentrations. Whole p l a t e l e t s were dissolved i n 1% SDS and reduced with 40 mM d i t h i o t h r e i t o l (DTT) according to Fairbanks (1971). Double l a b e l l i n g was as described above i n the Introduction. In 125 si t u a t i o n s where I l a b e l l e d discocytes had to be transformed and re-131 125 la b e l l e d with I, the p l a t e l e t s were f i r s t washed free of excess I, transformed and then reincubated with fresh lactoperoxidase, ^ 2®2 anC* 131 Na I. Afte r l a b e l l i n g , p l a t e l e t s were counted i n a LKB 1282 Dual-channel Compugamma counter before membrane i s o l a t i o n . - 149 -4.2.1.2 Borohydride T r i t i a t i o n g Platelets (2.5x10 /ml) in Tyrode's solution were f i r s t incubated with 1 mM sodium metaperiodate for 10 min at room temperature to selectively oxidize 3 s i a l i c acid residues. Then, 1 mCi/ml (4.4 mM) sodium H-borohydride was used to reduce and label the s i a l i c acid at room temperature for 30 min (McGregor et a l . , 1979). Other platelets (2.5xl0°/ml) were treated with neuraminidase as outlined in Chapter 3 to release s i a l i c acid and then incubated with galactose oxidase (20 U/ml) at 37°C for 30 min at pH 7.5 to oxidize the exposed galactose residues. 3 Reduction using H-borohydride was as above (McGregor ej: ajL. , 1979). - 150 -The pseudo-double label experiments were done as outlined in the Introduction using cold (5 mM) and then hot borohydride. In these cases "cold labelled" platelets were washed and then transformed with ADP or hypotonic shock. They were treated with metaperiodate or the enzymes under the same conditions as 3 used for the discocyte before relabelling with H-borohydride. Spherocytes were handled and labelled in hypotonic Tyrode's solution but otherwise under the same conditions and reagent concentrations as above. After labelling, a portion of each of the tritiated platelet samples was dissolved in Amersham NCS-Tissue Solubilizer and diluted in an OSC-Organic Counting Scintillant (Amersham) fl u i d . The samples were then counted with a Beckman LS-233 s c i n t i l -lation counter. Changes in counting efficiency were monitored using the - 151 -external standard ratio output of the counter. Other portions of the platelet samples were used for membrane isolation. 4.2.2 Isolation of Membrane Isolation of membranes from labelled platelets was according to the method g of Jamieson j 3 t aj.. (1979). Platelets (2.5x10 /ml) in Tyrode's solution were disrupted using a Braunsonic 1510 sonicator (A. Braun, South San Francisco, California) at 100W for about 30 seconds. Membranes were then isolated using differential centrifugation. The homogenates were i n i t i a l l y centrifuged at 20,000xg for 20 min at 4°C. The pellets containing platelet debris and undisrupted organelles were discarded. The remaining supernatant was again centrifuged at 150,000xg for 1 hr at 4°C. The pellets containing membranes were washed in Tyrode's solution and dissolved in 20 mM Tris-HCl pH 8 buffer containing 2% SDS and reduced with 40 mM dithiothreitol (DTT) according to Fairbanks (1971). Both centrifugation steps were done with a Beckman L5-65 Ultracentrifuge and a SW41 Swinging Bucket Rotor. 4.2.3 Gel Electrophoresis SDS-PAGE was run on 5% cylindrical gels cast according to the method of Fairbanks (1971). The gel solution consists of 5% acrylamide, 0.18% bis-acrylamide, 0.2% SDS, 0.025% tetramethylethylenediamine (TEMED) and 0.15% ammonium persulfate in a buffer system containing 40 mM Tris, 20 mM sodium acetate and 2 mM EDTA at pH 7.4. The gel solution was cast into 7x12.5 mm glass tubes and allowed to harden at room temperature for an hour. About 20 ug of dissolved protein material together with a small amount of bromophenol blue as tracking dye were loaded into each tube. The electrophoresis buffer (pH 7.4) consists of 40 mM Tris, 20 mM sodium acetate, 2 mM EDTA and 2 gm/L of - 152 -SDS. A Hoefer Scientific Instrument (San Francisco, California) gel electro-phoresis chamber model DE102 was used. Electrophoresis was run at 5 mA/tube in a 4°C refrigerator until the tracking dye reached bottom. The gels were removed from the tubes and fixed with 3.5% perchloric acid. BDH protein molecular weight standards were run alongside the membrane samples. They have molecular weights of 53,000; 106,000; 159,000; 212,000 and 265,000. The molecular weight standard gels were stained with coomassie blue according to Fairbanks (1971). Radioactive gels were cut into 1 mm slices using a BioRad gel slicer model 195 (BioRad Laboratories, Richmond, California). Iodinated gel slices were counted with the LKB Dual-channel Compugamma counter. Tritiated gel slices were incubated with 0.6 ml NCS-Tissue Solubilizer at 50°C for 2 hr. Radio-activity was leached out of the gel into solution which was then diluted with 6 ml of OSC-Organic Counting Scintillant fluid for counting in a Beckman LS-233 sc i n t i l l a t i o n counter. These procedures were recommended by the manufacturer Amersham Corp. (Arlington Heights, I l l i n o i s ) . Leaching was complete after 2 hrs, further incubation of the gel with fresh NCS overnight failed to remove any more radioactivity from the gel. 4.2.4 Surface Labelling in the Presence of Fibrinogen In this series of experiments fibrinogen at 0.5 mg/ml was included during iodination of the platelets. Fibronectin and albumin at the same concentration were added to other samples as controls. Calcium ions at 1 mM were added 125 together with the proteins. Labelling with I-lactoperoxidase followed. Membrane isolation and SDS-PAGE were then carried out in the same way as before. - 153 -4.2.5 Materials Fibrinogen used was the same as mentioned in Chapter 2. Fibronectin was prepared and kindly supplied by Mr. Johan Janzen using the method of Vuento and Vaheri (1978). Fibronectin from fresh human citrated plasma was f i r s t bound to a gelatin-Sepharose 4B aff i n i t y column and then eluted with 1.0M L-arginine. Bovine serum albumin was from Miles Laboratories (Elkhart, Indiana). Enzymes: neuraminidase (Vibrio cholerae) was from Calbiochem (La Jolla, California), lactoperoxidase from Sigma (St. Louis, Missouri) and galactose oxidase (Dactylium dendroides) was from Worthington (Freehold, N.J.). 125 131 Carrier-free sodium iodide - I and I as well as tritiated sodium borohydride were obtained from Amersham (Arlington Heights, I l l i n o i s ) . NCS-Tissue Solubilizer and OCS-Organic Counting Scintillant were also from Amersham. SDS-PAGE reagents including SDS, acrylamide, bis-acrylamide, IEMED and DTT were a l l from BioRad (Richmond, California). The molecular weight standard was from BDH (Poole, England). - 1 5 4 -4.3 Results 4.3.1 Iodination Experiments Gel electrophoresis of lactoperoxidase iodinated whole platelet discocytes and echinocytes revealed one major glycoprotein peak (GP-III) with an apparent molecular weight of 100,000 (Fig. 24). A smaller peak with molecular weight of 120,000 corresponding to GP-II was also present. The gel pattern of the spherocyte showed two major peaks, the GP-III peak and an additional peak with a molecular weight similar to that of GP-II (Fig. 24). When the plasma membrane from the iodinated platelets was isolated and electrophoresed, the patterns from the discocyte, echinocyte and spherocyte were a l l similar (Fig. 25). The large 120,000 molecular weight peak in the spherocyte pattern had apparently disappeared although the smaller underlying peak similar to that in the discocyte and echinocyte patterns remained. Gel patterns in Fig. 24 and 25 were typical of seven experiments. Washing of the spherocytes after iodination had no effect on the 120,000 molecular weight protein which was apparently removed during plasma membrane isolation. It appears that two proteins with similar molecular weights are involved: GP-II with a smaller peak size, and a second protein with a larger peak size appearing only on the intact spherocyte. The most likely explanation for the observation is that this second protein is not native to the platelet plasma membrane but an artifact originating from inside the spherocyte. It apparently leaks out and attaches i t s e l f to the surface of the spherocyte during hypotonic shock. The attachment is strong enough to withstand the washing procedure but the rigors of membrane isolation can dislodge this protein from the spherocyte membrane. This observation w i l l be discussed further in section 4.4.1. In light of this phenomenon, a l l SDS-PAGE experiments throughout the rest of this work were done using plasma membranes isolated from radiolabelled platelets. - 155 -Fig. 24. SDS-PAGE of ""^I-labelled platelets. Whole platelet (a) discocytes; (b) echinocytes and (c) spherocytes were dissolved in 1% SDS and reduced in 40 mM DTT. They were then run on 5% gels. o OOO'OVn 000'09-OOO'OS-ooo'oot-O O C t o ooo'ovi-000'09l-5 I o - 157 -Fig. 25. SDS-PAGE of plasma membrane isolated from I-labelled platelet (a) discocyte; (b) echinocyte and (c) spherocyte. Isolated membranes were dissolved and reduced as in Fig. 24. They were then run on 5% gels. r160,000 •140,000 -120,000 •100,000 ;80,000 "60,000 -40,000 - 159 -Gel patterns of isolated platelet membranes from the double iodination experiments are shown in Figs. 26-31. Fig. 27 and 28 demonstrate the f i r s t kind of double labelling experiment in which discocytes were labelled with 125 I. Echinocytes or spherocytes derived from unlabelled discocytes were 131 125 separately labelled with I and then run on the same gels as the I— 125 discocytes. Fig. 26 is a control in which I labelled discocytes were 131 mixed with I labelled discocytes and run on the same gel. Figs. 29-31 125 show the second kind of double labelling experiment in which I labelled discocytes were transformed into echinocytes or spherocytes and then relabelled 131 125 with I. Fig. 29 is the control in which I labelled discocytes were 131 . . . 125 relabelled with I without being transformed. The ratio, R, of the I 131 to 1 counts is shown under each pair of the gel patterns in Fig. 26-31. They w i l l be referred to as R-plots. No major differences can be observed between the patterns in any of these figures. They a l l show one major (GP-III) peak at molecular weight of about 100,000 and a smaller peak (GP-II) with molecular weight of about 120,000. Although some small variations are shown in the R-plots, they are not significant enough to demonstrate unequivocal differences between the 125 131 I- I patterns. The brackets (S-S) in a l l the R-plots in Figs. 26-31 represent +2 standard deviations of their respective means. In theory, i f the 125 131 I and I patterns are identical the R-plot w i l l be a straight line. If there is a significant difference between the patterns i t w i l l show up as a large peak or valley in the R-plot (see under the fibrinogen section below). The above patterns are typical of three series of similar experiments. Quantitation of the counts revealed that during the activation of platelets from discocyte to echinocyte there was a 21.7% (+1.5%) increase in the iodine - 160 -Fig. 26. SDS-PAGE of membranes (reduced) from a mixture of 125 131 I-labelled discocytes and I-labelled discocytes 125 131 run on the same 5% gel. (a) I-pattern; (b) I-pattern 125 131 and (c) ratio, R, of I to I activity in each slice (R-plot). Dotted line marks the mean of the R values and the bracket (s-s) represents +2 standard deviation. P160JOOO •140,000 •120,000 •100,000 •80,000 •60,000 %0,000 CPM > - 162 -Fig. 27. SDS-PAGE of membranes (reduced) from a mixture of 125 131 I-labelled discocytes and I-labelled echinocytes run 125 131 on the same 5% gel. (2) I-pattern; (b) I-pattern and (c) R-plot, as defined for Fig. 26. - 163 -- 164 -Fig. 28. SDS-PAGE of membranes (reduced) from a mixture of 125 131 I-labelled discocytes and I-labelled spherocytes run on the same 5% gel. (a) 1 2 5I-pattern; (b) 1 3 1I-pattern and (c) R-plot, as defined for Fig. 26. CPM > - 166 -Fig. 29. SDS-PAGE of membranes (reduced) from ^"'i-labelled 131 discocytes which were washed and then relabelled with I. 125 131 The membranes were run on 5% gel. (a) I-pattern; (b) I-pattern and (c) R-plot, as defined for Fig. 26. X [160,000 •140,000 •120,000 •100,000 •80,000 •60,000 40,000 o CPM. - 168 -Fig. 30. SDS-PAGE of membranes (reduced) from *~*~J I-label led discocytes which then transformed into echinocytes and re-131 labelled with I. The membranes were run on 5% gel. 125 131 (a) I-pattern; (b) I-pattern and (c) R-plot, as defined for Fig. 26. CPM O O O o o o o o o - 170 -Fig. 31. SDS-PAGE of membranes (reduced) from ^^"'i-labelled discocytes which then transformed into spherocytes and re-131 labelled with I. The membranes were run on 5% gel. (a) 1 2 5I-pattern; (b) 1 3 1I-pattern and (c) R-plot, as defined for Fig. 26. -160,000 [•140,000 120,000 -100P00 -80000 •60,000 %0,000 CPM > cn o o o o cn o o o o - 172 -incorporation. Transformation from discocytes to spherocytes resulted in a o o 125 39.8% (+1.3%) increase. These were whole platelet I counts from parallel single label experiments. The percentages were means of four series of such experiments. Morphological examinations of the platelets under phase microscopy were done before and after labelling to ensure they were in their proper discoid, echinoid or spheroid shapes as desired. There were no problems in this respect. No aggregation was encountered during or after labelling. 4.3.2 Tritiation Experiments Fig. 32 shows the gel patterns of platelet discocytes, echinocytes and spherocytes labelled with the periodate method. Fig. 33 shows them labelled using the neuraminidase-galactose oxidase method. More peaks are revealed in these patterns. The four major ones are GP-Ia; GP-Ib; GP-II and GP-III with apparent molecular weights 150,000, 140,000, 120,000 and 100,000 respectively. A few minor peaks can also be seen. They are GP-IV (which appears as a shoulder on the right side of GP-III); GP-V and GP-VI with molecular weights of 90,000, 80,000 and 65,000 respectively. However the overall patterns for the discocytes, echinocytes and spherocytes do not differ in either t r i t i a t i o n labelling procedure. These patterns are typical of two series of experiments. Quantitation of whole platelet counts showed that the echinocyte and the spherocyte had 0.4% and 0.9% more counts respectively than the discocyte using the periodate t r i t i a t i o n method. While using the neuraminidase-galactose oxidase labelling method the echinocyte had 16.2% and the spherocyte had 21.1% more counts than the discocyte. These percentages were the means of two series of experiments. - 173 -Fig. 32. SDS-PAGE gel patterns of membranes (reduced) from tritiated (a) discocyte; (b) echinocyte and (c) spherocyte. Tritiation was by the periodate method. 10,000 CPM i i 1 i i r160,000 -140,000 -120,000 :100,000 -80,000 -60,000 L40,000 - 1 7 5 -Fig. 33. SDS-PAGE gel patterns of membranes (reduced) from tritia t e d (a) discocyte; (b) echinocyte and (c) spherocyte. Tritiaton was by the neuraminidase/galactose oxidase method. 10,000 CPM i 1 1 1 1 1 1-160,000 -140,000 -120,000 -100,000 -80,000 £0,000 •40,000 CT) - 177 -The gel patterns of the pseudo-double label experiments are shown in Figs. 3 34 and 35. Using the periodate method, l i t t l e or no labelling by H (the 2nd label) has taken place (Fig. 34). With the neuraminidase-galactose oxidase method, small peaks are present in the gel patterns of the echinocytes and spherocytes (Fig. 35) . They correspond to GP-Ia, GP-Ib, GP-II and GP-III. In both Figs. 34 and 35, the patterns of the discocytes served as controls. 3 Here "cold labelled" discocytes were relabelled with H-borohydride without being transformed. Theoretically, i f the cold reaction went to completion, no 3 H labelling should be observed. Morphological examinations of the platelets under phase microscopy were done as in the iodination experiments above. Again, no problems were encountered. 4.3.3 Surface Labelling in the Presence of Fibrinogen The influence of fibrinogen (0.5 mg/ml) on lactoperoxidase catalyzed iodination of the platelet forms is shown in Figs. 36-38. The fibrinogen was added together with the lactoperoxidase and therefore was present during the whole process of iodination. Profiles of the gels of the discocyte (Fig. 36) and spherocyte (Fig. 38) show that fibrinogen did not influence iodination of the discocyte and spherocyte surface. On the other hand, a drop in the peak size of GP-III in the patterns of the echinocytes labelled in the presence of fibrinogen can be seen (Fig. 37). This is also reflected by a peak in the R-plot demonstrating a difference between the control (without fibrinogen) and the test gel at that location. Fibrinogen therefore impaired the labelling of GP-III on the echinocyte but not on the spherocyte or discocyte. Fibronectin (Fig. 40) and albumin (Fig. 39), tested also at 0.5 mg/ml, had no influence on the gel patterns of the echinocytes. This impairment in labelling was also - 178 -Fig. 34. SDS-PAGE gel patterns of membranes (reduced) from the pseudo-double labelled (a) discocyte; (b) echinocyte and (c) spherocyte. Platelet discocytes were f i r s t labelled with 3 cold borohydride, then transformed and relabelled with H-borohydride using the periodate method both times. - 179 -Q_ C_J> 2 0 0 0 0 2 0 0 0 0 (a) ( b ) - 180 -Fig. 35. SDS-PAGE gel patterns of membranes (reduced) from the pseudo-double labelled (a) discocyte; (b) echinocyte and (c) spherocyte. Platelet discocytes were f i r s t labelled with 3 cold borohydride, then transformed and relabelled with H-borohydride using the neuraminidase/galactose oxidase method both times. - 182 -Fig. 36. SDS-PAGE gel patterns of membranes (reduced) from 125 discocyte I-lactoperoxidase labelled (a) without and (b) with fibrinogen present (0.5 mg/ml). (c) Ratio, R, of the radioactivity in (a) to that in (b) for each gel slice (R-plot). Dotted line marks the mean of the R values and the bracket (s-s) represents +2 standard deviation. r160,000 -140,000 •120P00 •100,000 •80,000 •60,000 •40,000 CPM > ~ o o o o o co - 184 -Fig. 37. SDS-PAGE gel patterns of membranes (reduced) from 125 echinocytes I-lactoperoxidase labelled (a) without and (b) with fibrinogen present (0.5 mg/ml); (c) R-plot, the bracket (S-S) is the same as that of the control in Fig. 36. X O —•> ro cn O &> o T 1 r160,000 1-140,000 -120P00 H00,000 -80,000 -60,000 -40,000 CPM Q - 186 -Fig. 38. SDS-PAGE gel patterns of membranes (reduced) from 125 spherocytes I-lactoperoxidase labelled (a) without and (b) with fibrinogen present (0.5 mg/ml); (c) R-plot, as defined for Fig. 36. - 187 -- 188 -Fig. 39. SDS-PAGE gel patterns of membranes (reduced) from 125 echinocytes I-lactoperoxidase labelled (a) without and (b) with albumin present (0.5 mg/ml); (c) R-plot, as defined for Fig. 36. - 189 -- 190 -Fig. 40. SDS-PAGE gel patterns of membranes (reduced) from 125 echinocytes I-lactoperoxidase labelled (a) without and (b) with fibronectin present (0.5 mg/ml); (c) R-plot as defined for Fig. 36. </>h ooo'oy 000'09H 000'08: OO0OOI-oocfozt ooo'on-000091-5 cr - 192 -dependent on the concentration of fibrinogen. The decrease in the number of counts (cpm) from the control (echinocyte labelled with no fibrinogen present) increases in magnitude with fibrinogen concentration (Fig. 41). A plateau was reached at about 0.5 mg/ml. Gel patterns are typical of three series of similar experiments. Values in Fig. 41 are also from three sets of experiments. - 193 -F i g . 41. The e f f e c t of fibrinogen concentraion on the l a b e l l i n g of echinocytes. The decrease i n the number of counts (cpm) per echinocyte from the control (echinocyte l a b e l l e d with no fibrinogen present) i s plotted against fibrinogen concentration. Decrease in CPM per Echinocyte - 195 -4.4 DISCUSSION 4.4.1 Iodination Experiments The gel patterns in Figs. 24-31 are not unlike those found by others for lactoperoxidase iodination (Phillips, 1972; Okumura and Jamieson, 1976; Connellan et^  a l . , 1982). GP-III is most dominant while the GP-II peak is usually small. Those of the GP-Ia and GP-Ib are less obvious and better demonstrated in the t r i t i a t i o n gel patterns. L i t t l e difference can be seen between the gel patterns of single iodinated discocytes and echinocytes (Figs. 24 and 25). However the difference between the gel patterns of the iodinated intact spherocytes (Fig. 24) and that of plasma membrane isolated from iodinated spherocytes (Fig. 25) is obvious. The large peak with an apparent molecular weight of 120,000 had disappeared during the membrane isolation process leaving behind only a small peak similar to that of GP-II observed in the discocyte and echinocyte patterns. Apparently two proteins with similar molecular weights are involved here. The smaller peak belongs to GP-II. The identity of the other remains uncertain although i t must be a product of hypotonic shock. The fact that i t was removed during membrane isolation makes i t less likely to be a native component of the membrane. It i t believed to be a protein originating from the inside of the platelet and released to the outside during hypotonic treatment. Presumably i t was then adsorbed onto the surface of the spherocyte and iodinated. Removal from the platelet surface occurred during membrane isolation. The adsorption would be facilitated by low ionic strength condition of the hypotonic Tyrode's solution (Heard and Seaman, 1960). After membrane isolation they were washed in isotonic Tyrode's solution and desorption might have occurred assisted by the higher salt concentration. - 196 -In 1974, Phillips and Agin reported the presence of a thrombin sensitive protein with a molecular weight similar to that of GP-II in SDS-PAGE. They were not able to confirm the observation in a latter study (Phillips and Agin, 1977) and attribute i t to an artifact. In their i n i t i a l 1974 observation, i t was found that only 30% of the 120,000 molecular weight peak was susceptible to thrombin indicating the presence of two overlapping components with similar molecular weights in the gel. Several ideas were entertained at that point concerning the nature of the thrombin sensitive component. One of the suggestions was platelet factor XIII. Platelet factor XIII accounts for approximately 50% of total factor XIII activity in human whole blood and almost a l l of i t is carried in the platelet cytoplasm (Lopaciuk et a l . , 1976; Walsh, 1981). Leakage of platelet contents such as beta-thromboglobulin and LDH to the outside during hypotonic shock has already been discussed in Chapter 2. Leakage of some factor XIII out of the spherocyte is therefore not totally unexpected. Factor XIII was known to adsorb strongly onto platelet surfaces (Born, 1968). Platelet factor XIII is a dimer of two identical chains each having a molecular weight of about 110,000 (Loewy et a l . , 1961; Ganguly, 1971; Schwartz et a l . , 1971). In an in vitro environment the dimer spontaneously dissociates into i t s two monomer subunits (Loewy ejt a_l., 1961). Platelet factor XIII monomers appear to be a likely candidate for the protein in question here. Unlike plasma factor XIII, platelet factor XIII contains no carbohydrate (Schwartz et a l . , 1971) and therefore w i l l not influence s i a l i c acid results in Chapter 3. On the other hand, the presence of this adsorbed protein may physically interfere with the neuraminidase molecule resulting in underestimation of s i a l i c acid density on the spherocyte surface. However, this seems unlikely because the density of - 197 -the phosphate groups on the spherocyte and the echinocyte surfaces are similar (Table 9) and the molecular weight of alkaline phosphatase is much higher than neuraminidase. In light of the artifact involving intact spherocytes, the rest of the SDS-PAGE experiments were done using plasma membranes isolated from radio-labelled platelets. During the preliminary single iodination experiments (Figs. 24 and 25) no significant difference could be observed between the gel patterns of the discocyte and the echinocyte. The decision was therefore made to u t i l i z e double label procedures in hope that i t might bring out more subtle differences. The gel patterns of the double iodination experiments are shown in Figs. 26-31. It seems clear that the patterns are again similar to each other, since in no case did the ratio R vary by more than two standard deviations from i t s mean value. Apparently no new platelet surface species are labelled on the echinocyte or spherocyte. Fig. 29 shows that iodination of the 125 . tyrosine residues is less than complete. After the i n i t i a l I-iodination, 131 there was considerable tyrosine lef t over to be labelled with I for the second iodination. Fig. 29 represents the control experiment in which disco-cytes were f i r s t labelled with ^ " ' i and then relabelled with ' ^ I without transformation. However, failure of the f i r s t iodination reaction to go to completion w i l l not affect the outcome of the echinocyte and spherocyte experi-125 131 ments since the experimental conditions for both I and I iodinations were kept the same. New species appearing on the echinocyte or spherocyte 131 surface after transformation would be labelled by I together with "old 1 species not yet labelled by ^ " ' i . - 198 -125 Q u a n t i t a t i v e experiments showed that the ech inocy te bound 21.7% more I l a b e l than the d i s cocy te . The est imated i n c r ea s e i n su r face a rea , on the other hand, i s 44% (Chapter 3 ) . There are a number of p o s s i b l e exp lanat ions f o r t h i s d i screpancy. F i r s t l y , the ech inocyte area est imated i n Chapter 3 may be too h i g h . Th is seems u n l i k e l y because the areas obta ined by neuraminidase and a l k a l i n e phosphatase experiments mu tua l l y agree w i th each other and w i t h the F ro jmov i c e s t i m a t i o n . Secondly, 1 actoperox idase may have entered i n t o part of the i n vag i na ted sur face-connected c a n a l i c u l a r system of the d i s cocy te t o ca ta l y ze i o d i n a t i o n t h e r e . In t h i s case the d i f f e r ence betwwen values (counts) ob ta ined f o r the d i s cocy te and ech inocy te w i l l be sma l le r than expected with respect t o the e s t imated d i f f e r e n c e i n su r face areas. Lact oper ox i das e has a mo lecu la r weight of 93,000, l a r g e r than neuraminidase but smal le r than a l k a l i n e phosphatase. As d i scussed i n Chapter 3 f o r neuraminidase and a l k a l i n e phosphatase, i t i s u n l i k e l y t h a t 1 actoperox idase w i l l enter the c a n a l i c u l a r system. T h i r d l y , i f the on l y change i n the b iochemica l nature of the ech inocy te s u r f a ce i s the appearance of "new membrane" from the c a n a l i c u l a r system, t h i s membrane may be d e f i c i e n t , r e l a t i v e t o the d i s cocy te s u r f a c e , i n i o d i n a t a b l e t y r o s i n e r e s i due s . I t has a l ready been shown i n Tab le 16 that the ech inocyte "new membrane" has lower quan t i t i e s of te rmina l s i a l i c a c i d , phosphate and amino groups. F o u r t h l y , i t i s a l so po s s i b l e that the chemical charac te r of the p l a t e l e t su r face may have changed due t o ADP-induced t r an s f o rmat i on such that the o v e r a l l l a b e l l i n g i s reduced. Membrane changes i n v o l v i n g whole p l a t e l e t surfaces such as c a t i o n and f i b r i n o g e n b i nd i ng s i t e s have been d i scussed i n Chapter 3. However i t i s u n ce r t a i n i f the t h i r d , the f o u r t h , or both exp lanat ions best f i t s the d i sc repancy. - 199 -The spherocyte showed a 39.8% i n c rea se i n l a b e l l i n g w h i l e the geometric su r face area a c t u a l l y inc reased by 123% (Chapter 2). The i n c r ea se i n l a b e l l i n g due t o the appearance of the 120,000 molecu la r weight p r o t e i n ( f a c t o r X I I I ) , on the h y p o t h e t i c a l "new membrane" i s t h e r e f o r e minimal. Aga i n one can use reasons given above f o r echinocytes to e x p l a i n t h i s s i t u a t i o n . 4.4.2 T r i t i a t i o n Experiments Ge l patterns of t r i t i a t e d p l a t e l e t s show more peaks than the i o d i n a t e d samples. A g a i n , t he re i s no apparent d i f f e r e n c e between the gel p r o f i l e s of the d i s c o c y t e , e ch i noc y t e , or spherocyte . The patterns of p l a t e l e t s l a b e l l e d u s i ng the pe r i oda te method and those u s ing the neuramin idase-ga lactose ox idase method a re a l s o s i m i l a r ( F i g s . 32 and 33). They a re not un l i k e those shown elsewhere (McGregor jet al ., 1979; Mosher et _ a l . , 1979; Marches i and Chas is , 1979; P h i l l i p s and A g i n , 1977) The pseudo-double l a b e l experiments u s ing the pe r i oda te t r i t i a t i o n method showed no H - l a b e l l i n g at a l l ( F i g . 34) . Here the d i scocytes were f i r s t l a b e l l e d with cold borohydr ide f o l l o w i n g per iodate o x i d a t i o n . They were then transformed i n t o echinocytes or spherocytes and r e l a b e l l e d w i t h the per iodate 3 method us ing H impl ies that a l l the t e rm ina l s i a l i c a c i d res idues on t h e i r sur faces had a l ready r eac ted . I n other words, the per iodate crossed the membrane or entered the c a n a l i c u l a r system and o x i d i z e d a l l the s i a l i c a c i d i n the membrane. Both borohydride and m eta per iodate are very small molecules and shou ld e a s i l y enter the su r f ace-connected c a n a l i c u l a r system. They a l s o cross membranes and l a b e l the i n t e r i o r m a t e r i a l s of i n t a c t c e l l s (Gahmberg and Andersson, 1977). - 200 -The pseudo-double label experiments using the neuraminidase-galactose oxidase method present another picture (Fig. 35). Small peaks were found corresponding to the surface glycoproteins already demonstrated in the single label gels. Since the molecular weight of neuraminidase is 90,000 and galac-tose oxidase is 42,000 (Worthington, 1972), they would not be expected to enter the canalicular system significantly. Therefore the glycoproteins on the canalicular surface were probably not cold labelled in the discocyte. Even though the cold borohydride had access to the canalicular system, the enzymes required to catalyze the reaction should not have been able to enter. After transformation from discocyte to echinocyte or spherocyte the glyco-3 proteins from these surfaces became available for enzymatic oxidation and H labelling. The fact that the peaks here are relatively small compared to those of the direct labelling experiments (Fig. 33), suggests that the amount of glycoproteins cloistered in the discocyte canalicular system is rather small. Quantitative single label experiments using the periodate method showed that the echinocyte and the spherocyte had only 0.4% and 0.9% more counts respectively than the discocyte. Using the neuraminidase-galactose oxidase 3 method, the echinocyte exhibited 16.2% and the spherocyte 21.1% more H label than the discocyte. The discrepancies can be explained using the same argument as for the pseudo-double labelling experiments discussed above. The periodate had ready access to the invaginated canalicular system. Therefore the amount of membrane surface available for labelling is the same in the discocyte, echinocyte and spherocyte. On the other hand, the enzymes w i l l not be able to enter the canalicular system. More surface w i l l therefore be made - 201 -a v a i l a b l e f o r l a b e l l i n g a f t e r evag ina t i on of the c a n a l i c u l a r system i n the ech inocyte and spherocyte . . . 3 The q u a n t i t a t i v e r e s u l t s from neuramin idase-ga lactose ox idase H l a b e l l i n g . . 125 were qu i t e s i m i l a r t o those of the l a c t ope rox i da se I - l a b e l l i n g exper iments . The percent i n c rea se s i n the t r i t i a t i o n of the ech inocyte and spherocyte sur faces were much lower than the percent increases i n t h e i r sur face areas est imated i n Chapter 2 and 3. Th i s aga in suggests e i t h e r that the den s i t y of the g lycoprote in s on the i n vag i na ted membrane system i s much lower than on the e x t e r n a l plasma membrane or that the average su r face p rope r t i e s of the transformed c e l l s have changed. I t has been r epo r ted by Cazenave j jt _al. (1976) that pe r i oda te (1-10 mM) treatment of r a b b i t p l a t e l e t s i n Ty rode ' s s o l u t i o n caused them to aggregate. However, we found no aggregat ion dur ing the experiments w i t h human p l a t e l e t s . The ch ie f concern here i s that the d i s cocy te s may t u r n i n t o ech inocytes dur ing the l a b e l l i n g procedures. M i c r o s c o p i c examinat ion demonstrated that t h i s d i d not happen. 4. 4. 3 Surface L a b e e l i n g i n the Presence of F i b r i nogen I t was found here t ha t f i b r i n o g e n impairs the i o d i n a t i o n of ech inocy te surface GP- I I I . I t seems that the b i nd ing of f i b r i n o g e n onto the ech inocy te su r face br ings i t i n t o c lose a s s o c i a t i o n w i t h GP- I I I , c l o se enough f o r f i b r i n o g e n (molecular weight 340,000) t o i n t e r f e r e w i t h the enzyme l a c t ope rox i da se which cata lyzes the i o d i n a t i o n . Wh i l e t h i s does not mean that the f i b r i n o g e n recepto r i s a c t u a l l y GP- I I I , i t c e r t a i n l y adds f u r t h e r evidence t o support the i d e a that GP-I I I i s the r ecep to r . The phenomenon i s s p e c i f i c o n l y t o the ech inocy te ( a c t i v a t e d p l a t e l e t ) and not the d i s cocy te or spherocyte . I t i s - 202 -specific to fibrinogen and does not occur with albumin or fibronectin. This last result is consistent with the work of Harfenist (1980) who found that fibronectin is not involved in platelet aggregation and w i l l not bind to activated platelets. It has already been shown in Chapter 2 that ADP induced platelet aggregation requires fibrinogen. Fig. 10 and Fig. 41 are extremely similar to each other. They imply a close relationship between platelet aggregation and fibrinogen binding to or near GP-III on the platelet surface. This experiment failed to demonstrate any association of fibrinogen with the spherocyte surface. It is possible that there is no fibrinogen receptor on the spherocyte or that the receptors have been destroyed during the hypotonic treatment. Presence of the 120,000 molecular weight protein on the spherocyte might also have physically blocked the fibrinogen receptors and/or the C a + + binding sites. This might have caused the drastic drop in the Ca + + binding sites on the spherocyte (section 3.3.5). 4.4.4 Summary of Chapter 4 Different methods have been used to label the platelet discocyte, echino-cyte and spherocyte but none showed any qualitative differences between the gel patterns. Quantitative differences were demonstrated, however, but i t was found that the labelling density on the evaginated surfaces of the canalicular system was lower than on the discocyte surface, presuming the latter was unchanged during shape transformation. A close association between fibrinogen and GP-III was also demonstrated in the echinocyte, the concentration dependence of which closely parallels the fibrinogen dependence of aggregation. - 203 -CHAPTER 5 SUMMARY AND CONCLUSIONS - 204 -The purpose of this project was to investigate the surface properties of the platelet discocyte, echinocyte and spherocyte. A model is examined here in which an echinocyte attains i t s stickiness properties by evagination of the surface-connected canalicular system. Platelets also evaginate their canalicular system upon hypotonic shock treatment to form spherocytes. By comparing the properties of different forms of platelets some insight into the nature of "stickiness" was sought. The following conclusions can be drawn from the results obtained. 5.1 The surface areas of the discocyte and spherocyte were found micro-—8 2 —8 2 scopically to be 16.4x 10 cm , and 36.7x10 cm respectively (Chapter —8 2 2). The surface area of the echinocyte was estimated to be 23.7x10 cm using surface chemical and c e l l electrophoresis techniques (Chapter 3). The surface area of the echinocyte is therefore much smaller than that of the spherocyte. 5.2 Electron microscopic examination showed that the surface-connected canalicular system may not have totally evaginated in the echinocyte, therefore supporting the surface area calculations. 5.3 Aggregometry studies demonstrated a biphasic requirement for the divalent ++ ++ cations Ca or Mg to support ADP induced aggregation. The optimum ++ . ..... . ++ concentration is 4 mM for both cations but Ca is more efficient than Mg ++ ++ in supporting aggregation. Sr does not support aggregation while Mn inhibits aggregation. 5.4 Microelectrophoresis studies revealed an eight fold increase in the density of C a + + binding sites on the platelet surface during discocyte to - 205 -echinocyte transformation. The increase in the density of Mg binding sites was over six fold. The spherocyte which has lost i t s a b i l i t y to aggregate also lost over 60% of i t s Ca binding sites. 5.5 ADP induced aggregation requires the presence of fibrinogen (Chapter 2). The presence of fibrinogen also interferes with the lactoperoxidase labelling of GP-III on the ADP activated platelet surface (Chapter 4). Similar concen-tration dependences were found between the two phenomena. 5.6 Although spherocytes have lost their a b i l i t y to aggregate they can s t i l l be agglutinated with ristocetin and a plasma component, presumably von Willebrand 1s factor. 5.7 Neuraminidase treatment has no effect on ADP induced aggregation but the platelet's responses to different lectins are altered. The removal of terminal s i a l i c acid residues and the exposure of galactose by the enzyme decreases the aggregation response to WGA and increase the responses to RCA and JBA. 5.8 Electrokinetic analysis of live, fixed and neuraminidase or alkaline phosphatase treated platelets showed major differences in charge densities as well as amino, s i a l i c acid and phosphate group densities on the discocyte, echinocyte and spherocyte. The evaginated "new" membrane surfaces of the echinocyte and spherocyte seem also considerably different. Data are summarized in Table 16. 5.9 SDS-PAGE of platelets radiolabelled using a variety of methods including lactoperoxidase (single and double) iodination, periodate-borohydride t r i t i a t i o n and neuraminidase/galactose oxidase-borohydride t r i t i a t i o n failed - 206 -to show any significant differences in the gel patterns of the discocyte, echinocyte and spherocyte. No new glycoprotein species appeared after the transformations but quantitative differences among the three forms of platelets were found in the radiolabelling experiments. 5.10 Considering the amounts of s i a l i c acid removed from the platelet forms by neuraminidase and the quantitative differences in radiolabelling experiments, i t appears that the "new membrane" surfaces of the echinocyte and spherocyte have lower densities of glycoproteins than the discocyte surface. In the original hypothesis, i t was proposed that the discocyte surface-connected canalicular membrane system is inherently "sticky". The apparent increase in surface area associated with the discocyte to echinocyte transformation was thought to occur via the evagination of this membrane system. Since there is no evidence that platelets synthesize membrane, the "pre-formed" membrane of the canalicular system seems to be the most likely, i f not the only, choice for the source of the added area. However, whether this cloistered membrane system is also inherently "sticky" is another matter. Evagination of the canalicular membrane system by hypotonic shock failed to create a spherocyte that was capable of fibrinogen-dependent aggregation, even though i t was found that the spherocyte had a much larger surface area than the echinocyte, indicating a greater degree of evagination. It is shown in Table 16 that the echinocyte "new membrane" has lower densities of the various charge groups than the discocyte surface. Its evagination w i l l therefore cause a dilution effect resulting in the lowering of the average surface charge density on the platelet echinocyte. - 207 -Lowering of surface charge in principle could f a c i l i t a t e platelet-platelet contact and aggregation. However, the spherocyte has an even lower negative charge density than that of the echinocyte but i t s surface is non-sticky and the spherocytes do not aggregate. Therefore i t seems unlikely that "stickiness" results simply from the lowering of platelet surface charge density. Two other properties that can be considered as a characteristic of "stickiness", i.e. the appearance of Ca + +/Mg + + binding sites and of fibrinogen binding sites were observed here. They are unique to the echinocyte and were not observed on the discocyte or spherocyte. GP-III was demonstrated in a l l three forms of platelets but only those on the echinocyte surface associate with fibrinogen molecules. This suggests that a conformational change in GP-III i t s e l f , or in molecules associated with i t , i s a consequence of platelet activation and not necessarily a result only of shape change and area increase. In conclusion, a number of differences were demonstrated here between the surface properties of the platelet discocyte, echinocyte and spherocyte (Table 17). However, the data f a i l to support the model that platelet "stickiness" is a result only of canalicular system evagination or that the cloistered membrane of the canalicular system is inherently "sticky". It appears more likely that changes occur over the whole platelet surface during activation and make them capable of participating in aggregation reactions. - 208 -Table 17. Highlights of Differences* Between Platelet Discocyte, Echinocyte, and Spherocyte Discocyte Echinocyte Spherocyte Area (cm2) 16.4xl0~8 23.7xl0~ 8 36.7xl0~ 8 Negative charge density per cm2 9.13xl0 1 2 7.92xl0 1 2 6.50xl0 1 2 Ammo groups per cm' 1.33xl0 1 2 1.19xl0 1 2 1.08xl0 1 2 S i a l i c acid molecules per cm2 3.98xl0 1 2 3.52xl0 1 2 2.12xl0 1 2 Phosphate groups per cnr- 2.28xl0 1 2 1.96xl0 1 2 1.96xl0 1 2 S i a l i c acid/Total charge 43.5% 44.4% 32.6% Phosphate/Total charge 25.0% 24.6% 30.1% Sialic acid/Phosphate 1.75 1.80 1.08 Ca + + binding sites per cm2 1.28xl0 1 2 1.09xl0 1 3 2.22xlO n Fibrinogen association with membrane No Yes No Aggregation No Yes No Membrane "stickiness" No Yes No *No differences found in their SDS-PAGE patterns. - 209 -REFERENCES - 210 -Abramson, H.A. (1928) The electrophoresis of blood platelets of the horse with reference to their origin and to thrombus formation. J. Expt. Med., 47_: 677-683. Ada, G.L., French, E.L. and Lind, P.E. (1961) Purification and properties of neuraminidase from Vibro cholera. J. Gen. Microbiol., 24:409-421. Aledort, L.M. and Niemetz, J. (1968) Dissociation of platelet aggregation from clot retraction, potassium loss and adenosine triphosphatase activity. Proc. Soc. Expt. Biol. Med., 128:658-661. Allain, J.P., Cooper, H.A., Wagner, R.H. and Brinkhous, K.M. (1975) Platelets fixed with paraformaldehyde: a new reagent for assay of von Willebrand's factor. J. Lab. Clin. Med., 85:318-328. Barnhart, M.I. (1978) Platelet responses in health and disease. Mol. Cell.  Biochem., 22:113-137. Behnke, 0. (1968) Electron microscopical observations on the surface coating of human blood platelets. J. Ultrastructure Res., 24:51-69. Behnke, 0. (1978) Effects of some chemicals on blood platelet microtubules, platelet shape and some platelet functions in vitro. Scand. J. Haemat., 2:123-140. Behring Institute, West Germany (1979). Personal communication. Bennett, J.S. and V i l a r i e , G. (1979) Exposure of platelet fibrinogen receptors by ADP and epinephrine. J. Clin. Invest., 64:1393-1401. Bennett, J.S., Vilaire, G. and Cines, D.B. (1981) Identification of the human platelet fibrinogen receptor by photoaffinity labeling. Blood, 58:188a. Best, L.C., Bowe, E.A. and Russell, G.G. (1981) Strontium ions induced pro-duction of thromboxane B2 and secretion of 5-hydroxytryptamine in washed human platelets. Biochem. Pharm., 30:635-637. Betts, J.J., Betts, J.P. and Nicholson, J.T. (1968) Significance of ADP, plasma and platelet concentration in platelet electrophoretic studies. Nature, 219:1280-1282. Bhavanandan, V.P. and Katlic, A.W. (1979) The interaction of wheat germ agglutinin with sialoglycoprotein. J. Biol. Chem., 254:4000-4008. Bik, T., Sarov, I. and Livne, A. (1982) Interaction between vaccinia virus and human blood platelets. Blood, 59:482-487. Boisseau, M.R., Lorient, M.F., Born, G.V.R. and Michal, F. (1977) Change in electrophoretic mobility associated with the shape change of human blood platelets. Proc. R. Soc. Lond., 196:471-474. - 211 -Bolton, A.E., Amess, J.A.L., Lekhwani, CP. and E l l i o t t , P. (1980) Beta-thromboglobulin content of human blood platelet. Scand. J. Hematol., 25: 25-29. ' Boneu, B., Robert, A., Sie, P., Elkoury, M., Liverato, C, Caranobe, C. and Nouvel, C. (1982) Coulter counter studies of hypotonic-induced macrothrombo-cytosis in normal subjects and in idiopathic thrombocytopenic purpura patients. Br. J. Haematol., 51:305-311. Born, G.V.R. (1962) Aggregation of blood platelets by adenosine diphosphatase and i t s reversal. Nature, 194:927-928. Born, G.V.R. and Cross, M.J. (1963) The aggregation of blood platelets. J. Physiol., 168:178-195. Born, G.V.R. and Cross, M.J. (1964) Effects of inorganic ions and of plasma proteins on the aggregation of blood platelets by ADP. J. Physiol., 170: 397-414. Born, G.V.R. (1968) The platelet membrane and its functions. In First  International Symposium on Metabolism and Membrane Permeability of Erythro- cytes and Thrombocytes. Edited by E. Deutsch, E. Gerlach and K. Moser. Georg Thieme Verlag, Stuttgart, Germany. Born, G.V.R. (1970) Observations on the change in shape of blood platelets brought about by ADP. J. Physiol., 209:487-551. Born, G.V.R. and Feinberg, H. (1975) Binding of adenosine diphosphate to intact human platelets. J. Physiol., 251:803-816. Born, G.V.R., Dearnley, R., Foulks, J.G. and Sharp, D.E. (1978) Quantifi-cation of the morphological reaction of platelets to aggregating agents and of its reversal by aggregating inhibitors. J. Physiol., 280:193-212. Bosmann, B.H. (1972) Platelet adhesiveness and aggregation. Biochim.  Biophys. Acta, 279:456-474. Bowles, D.J. and Brunton, C. (1982) Interactions between platelet-surface glycoproteins. Biochem. J., 202:707-716. Brass, L.F. and Shattil (1982) Increased calcium surface binding and efflux during platelet aggregation. Clin. Res., 30:312A. Bray, B.A. and Alexander, B. (1969) Progressive cleavage of s i a l i c acid from platelets and their electrophoretic mobility. Blood, 34 :523. Bunting, R.W., P eerschke, E.I. and Zucker, M.B. (1978) Human platelet s i a l i c acid content and tritium incorporation after ADP-induced shape change and aggregation. Blood, 52:643-653. Cazenave, J.P., Reimer, H.J., Kinlough-Rathbone, R.L., Packham, M.A. and Mustard, J.F. (1976) Effect of sodium periodate on platelet function. Lab.  Invest., ^4:471-481. - 212 -Chen, P.S., Toribara, T.Y. and Warner, H. (1956) Microdetermination of phosphorus. Anal. Chem., 28:1756-1758. Clark, C. and Harms, C. (1978) Measurement of serotonin uptake and release using r a d i o a c t i v e carbon-14 l a b e l l i n g . In P l a t e l e t Function. Laboratory  Evaluation and C l i n i c a l A p p l i c a t i o n . Edited by D.A. T r i p l e t t . American Society of C l i n i c a l Pathologists, Chicago, pp. 240-242. C o l l e r , B.S. (1978) The e f f e c t s of r i s t o c e t i n and von Willebrand factor on p l a t e l e t electrophoretic mobility. J . C l i n . Invest., 62:1168-1174. C o l l e r , B.S. (1981) Preliminary c h a r a c t e r i z a t i o n of hybridoma antibodies that i n h i b i t f i b r i n o g e n - p l a t e l e t i n t e r a c t i o n s . Blood, 58:191a. C o l l e r , B.S. and Zarrabi, M.H. (1981) P l a t e l e t membrane studies i n the May-Hegglin anomaly. Blood, 58:279-284. Connellan, J.M., Barlow, B., Smith, I. and C a s t a l d i , P.A. (1982) The r o l e of p l a t e l e t surface proteins reacting with heterologous antibodies. Haemostas., 11:109-118. Cook, G.M., Heard, D.H. and Seaman, G.V.F. (1961) S i a l i c acid and the e l e c t r o -k i n e t i c charge of the human erythrocyte. Nature, 191:44-47. C u l l i n g , C.F.A., Reid, P.E., Ramey, C.W., Dunn, W.L. and Clay, M.G. (1977) A method for the simultaneous estimation of free and k e t o s i d i c a l l y bound s i a l i c a cid. Can. J. Biochem., 55:778-782. Davies, G.E. and Palek, J. (1982) The state of a c t i n polymerization i n t e t r a caine treated p l a t e l e t s . Thrombos. Haemostas., 48:153-155. Day, H.J., Holmsen, H. and Zucker, M.B. (1976) Report of the working party on p l a t e l e t s . Thrombos. Haemostas., 36:263-276. Denticke, B. (1968) Transformation and r e s t o r a t i o n of biconcave shape of human erythrocytes induced by amphibilic agents and change of i o n i c environment. Biochim. Biophys. Acta, 163:494-500. Detwiler, T.C., Charo, I.F. and Feinman, R.D. (1978) Evidence that calcium regulates p l a t e l e t function. Thrombos. Haemostas., 40:207-211. Deykin, D. , P r i t z k e r , CR. and Scolnick, E.M. (1965) Plasma co-factor i n adenosine diphosphate induced aggregation of human p l a t e l e t s . Nature, 208: 296-298. Diamond, J.M. and Wright, E.M. (1969) B i o l o g i c a l membranes: the physical basis of ion and nonelectrolyte s e l e c t i v i t y . Ann. Rev. Ph y s i o l . , 31:581-646. DiMinno, G., Thiagarajan, P., Perussia, B., Martinez, J . , Shapiro, S.S. and Murphy, S. (1981) I n h i b i t i o n of ADP, collagen and arachidonic acid dependent fibrinogen binding to human p l a t e l e t s by a monoclonal antibody to Glyco-protein I l b - I I I a . Blood, 58:193a. - 213 -Donath, E. and Pastushenko, V. (1979) Electrophoretical study of c e l l sur-face properties. The influence of the surface coat on the electric potential distribution and on general electrokinetic properties of animal cells. J.  Electroanal. Chem., 104:543-554. Fairbanks, G., Theodore, L.S. and Wallach, D.F.H. (1971) Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochem., 10:2606-2616. Fantl, P. (1966) Osmotic stability of blood platelets. J. Physiol., 198:1-16. Fernley, H.N. (1971) Mammalian alkaline phosphatase. In The Enzymes, Vol. IV. Edited by P.D. Boyer. Academic Press, New York, pp. 417-447. Frantantoni, J.C. and Poindexter, B.J. (1981) Characterization of the plate-let response to exogenous arachidonic acid. Thromb. Res., 22:157-166. Frenette, G. (1980) Boehringer Mannheim Canada, Dorval, Quebec. Personal communication. Frojmovic, M.M. (1973) Quantitative parameterization of the light trans-mission properties of citrated, platelet-rich plasma as a function of plate-let and adenosine diphosphate concentrations and temperature. J. Lab. Clin.  Med., 82:137-153. Frojmovic, M.M. and Panjwani, R. (1976) Geometry of normal mammalian plate-lets by quantitative microscopic studies. Biophys. J., 16:1071-1089. Frojmovic, M.M. and Milton, J.G. (1982) Human platelet size, shape and related functions in health and disease. Physiol. Rev., 62:185-261. Gahnberg, C.G. and Andersson, L.C. (1977) Selective radioactive labeling of c e l l surface sialoglycoprotein by periodate-tritiated borohydride. J. Biol.  Chem., 257:5888-5894. Ganguly, P. (1971) Isolation and properties of a thrombin sensitive protein from human blood platelets. J. Biol. Chem., 246:4286-4290. Ganguly, P. and Fossett, N.G. (1979) Role of surface s i a l i c acid in the inter-action of wheat germ agglutinin with human platelets. Biochem. Biophys. Res.  Commun., 89:1154-1160. Ganguly, P., Gould, N.J. and Sidhu, P. (1979) Interactions of lectins with human platelets. Biochim. Biophys. Acta, 586:574-583. Ganguly, P.J. and Fossett, N.G. (1980) Evidence of multiple mechanisms of interaction between wheat germ agglutinin and human platelets. Biochim.  Biophys. Acta, 627:256-261. Gear, A.R.L. (1981) Preaggregation reactions of platelets. Blood, 58:477-490. - 214 -George, J.N., Lyons, R.M. and Morgan, R.K. (1980) Membrane changes associated with platelet activation. J. Clin. Invest., 66:1-9. Gogstad, G.O., Hagen, I., Korsmo, R. and Solum, N.O. (1981) Characterization of the proteins of isolated human platelet alpha-granules. Biochim. Biophys.  Acta, 670:150-162. Goldstein, I.J. and Hayes, CE. (1978) The lectins: carbohydrate-binding proteins of plants and animals. Adv. Carb. Chem. Biochem., 35:127-340. Gottschalk, A. and Bhargava, A.S. (1971) Neuraminidase. In The Enzymes, Vol. V. Edited by P.D. Boyer. Academic Press, New York, pp. 321-342. Greenberg, J. and Jamieson, G.A. (1974) The effect of various lectins on platelet aggregation and release. Biochim. Biophys. Acta, 345:231-242. Greenberg, J., Packham, M.A., Cazenave, J.P. , Reimers, H.J. and Mustard, J.F. (1975) Effect on platelet function of removal of platelet s i a l i c acid by " neuraminidase. Lab. Invest., 32:476-484. Grottum, K.A. (1968) Influence of aggregating agents on electrophoretic mobility of blood platelets from healthy individuals and from patients with cardiovascular disease. Lancet, 1:1406-1408. Hampton, J.R. and Mitchell, J.R.A. (1966) The effect of aggregating agents on the electrophoretic mobility of human platelets. Br. Med. J., 1:1074-1077. Handin, R.I., Fortier, N.L. and Valeri, G.R. (1970) Platelet response to hypo-tonic stress after storage at 4°C and 22°C. Transfusion, 10:305-309. Harfenist, E.J. (1980) Plasma fibronectin is not involved in ADP-induced aggregation of rabbit platelets. Thrombos. Haemostas., 44:108. Harfenist, E.J., Packham, M.A. and Mustard, J.F. (1980) Identical behavior of fibrinogen and asialo-fibrinogen in reactions with platelets during ADP-induced aggregation. Thromb. Res., 20:353-358. Hashimoto, S. and Suzuki, K. (1979) Characterization of interaction among von Willebrand factor, ristocetin and platelet. Acta Haematol. Jap., 42:851-863. Hawiger, J., Parkinson, S. and Timmons, S. (1980) Prostacyclin inhibits mobilization of fibrinogen-binding sites in human ADP and thrombin treated platelets. Nature, 283:195-197. Heard, D.H. and Seaman, G.V.F. (1960) The influence of pH and ionic strength on the electrokinetic stability of the human erythrocyte membrane. J. Gen.  Physiol., 43:635-654. Heptinstall, S. (1976) The use of a chelating ion-exchange resin to evaluate the effects of the extracellular calcium concentration on adenosine diphos-phate induced aggregation of human blood platelets. Thrombos. Haemostas., 36:208-220. - 215 -Holmsen, H. (1977) Prostaglandin endoperoxide - thromboxane synthesis and dense granule secretion as positive feedback loops in the propagation of platelet responses during "the basic platelet reaction." Thrombos.  Haemostas., 38:1030-1041. Hovig, T. (1965) The effect of various enzymes on the ultrastructure, aggreg-ation and clot retraction ab i l i t y of rabbit blood platelets. Thrombos.  Haemostas., 13:84-113. Jamieson, G.A., Okumura, T., Fishback, B., Johnson, M.M., Egan, J.J. and Weiss, H.J. (1979) Platelet membrane glycoproteins in thromboasthenia, Bernard-Soulier syndrome, and storage pool disease. J. Lab. Clin. Med., 93:652-660. Janzen, J. (1983) Ph.D. thesis, Department of Pathology, University of British Columbia. Jenkins, C.S.P., Ali-Briggs, E.F., Zonneveld, G.T.E., Sturk, A. and Clemetson, K.J. (1979) Human blood platelet membrane glycoproteins. Thrombos.  Haemostas., 42:1490-1502. Jentoft, N. and Dearborn, D.G. (1979) Labelling of proteins by reductive methylation using sodium cyanoborohydride. J. Biol. Chem., 254:4359-4365. Jung, S.M., Kinoshita, K., Tanoue, K., Isohisa, I. and Yamazaki, H. (1982) Role of surface negative charge in platelet function related to the hyper-active state in estrogen treated prostatic carcinoma. Thrombos. Haemostas., 47:203-209. Kachmar, J.F. (1970) Enzymes. In Fundamentals of C l i n i c a l Chemistry. Edited by N.W. Tietz. W.B. Saunders Co., Philadelphia, pp. 362-473. Kalbfleisch, J.H. (1974) Introduction to St a t i s t i c a l Methods. Medical College of Wisconsin, Milwaukee, Wisconsin. Kao, K.J., Pizzo, S.V. and McKee, P.A. (1979) Platelet receptors for human factor VHI/von Willebrand's protein. Pro. Nat. Acad. Sci., 76:5317-5320. Kaplan, K.L. and Owen, J. (1981) Plasma levels of beta-thromboglobulin and platelet factor 4 as indices of platelet activation in vivo. Blood, 57: 199-202. Kim, B.K. and Baldini, M.G. (1976) The platelet response to hypotonic shock, its value as an indicator of platelet v i a b i l i t y after storage. Transfusion, 14:130-138. Kirby, E.P. (1977) Factor VHI-associated platelet aggregation. Thrombos.  Haemostas., 38:1054-1072. Kirschmann, C , Katchalsky, A. and de Vries, A. (1959) Electrochemical study of the platelet surface. Arch. Biochem. Biophys., 80:140-149. - 216 -Kosztolanyi, G., Jobst, K., Kellermayer, M. and Ludany, A. (1980) ADP induced surface charge changes of adult and newborn p l a t e l e t s . Br. J. Haematol., 46: 257-262. Ku, C.S.L. and Wu, K.K. (1977) Increase of surface s i a l i c a c i d content during p l a t e l e t a c t i v a t i o n . Blood, 50(Suppl):244. Lages, B., Scrutton, M.C. and Holmsen, H. (1975) Studies on g e l - f i l t e r e d human p l a t e l e t s : i s o l a t i o n and ch a r a c t e r i z a t i o n i n a medium containing no added Ca++, Mg++ or K+. J . Lab. C l i n . Med., 85:811-825. Lasky, M. (1978) Protein molecular weight determination using polyacrylamide gradient gels i n the presence and absence of sodium dodecyl s u l f a t e . In Electrophoresis '78. Edited by N. Catsimpoolas. Elsevier North Holland Press, Amsterdam, pp. 195-210. Lee, H., Nurden, A.T., Thomaidis, A. and Caen, J.P. (1981) Relationship between fibrinogen binding and the p l a t e l e t glycoprotein d e f i c i e n c i e s i n Glanzmann's thromboasthenia Type I and Type I I . Br. J . Haematol., 48:47-57. Legrand, C. and Caen, J.P. (1978) Binding of l^C-ADP to normal and thromb-asthenic p l a t e l e t membranes. Haemostas., 7_:339-351. Levine, S., Levine, M., Sharp, K.A. and Brooks, D.E. (1983) Theory of the el e c t r o k i n e t i c behaviour of human erythrocytes. Biophy. J . , i n press. Loewy, A.G., Dahlberg, A., Dunathan, K. , K r i e l , R. and Wolfinger, H.L. (1961) Fib r i n a s e . Some physical properties. J . B i o l . Chem., 236:2634-2643. Lopaciuk, S., Lovette, K.M., McDonagh, J . , Chuang, H.Y.K. and McDonagh, R.P. (1976) Subcellular d i s t r i b u t i o n of fibrinogen and factor XIII i n human blood p l a t e l e t s . Thrombos . Res. , 8^:453-465. Ludlam, CA. and Cash, J.D. (1976) Studies on the l i b e r a t i o n of beta-thromboglobulin from human p l a t e l e t s i n v i t r o . Br. J . Haematol., 33:239-247. Madoff, M.A., Ebbe, S. and B a l d i n i , M. (1964) S i a l i c acid of human blood p l a t e l e t s . J . C l i n . Invest., 43:870-877. Mannucci, P.M. and Sharp, A.A. (1967) P l a t e l e t volume and shape i n r e l a t i o n to aggregation and adhesion. Br. J . Haematol., 13:604-617. Marchesi, S.L. and Chasis, J.A. (1979) I s o l a t i o n of human p l a t e l e t glyco-proteins. Biochim. Biophys. Acta, 555:442-459. Marguerie, G.A., Plow, E.F. and Edgington, T.S. (1979) Human p l a t e l e t s pos-sess an inducible and saturable receptor s p e c i f i c for fibrinogen. J . B i o l .  Chem., 254:5357-5363. Marguerie, G.A., Edgington, T.S. and Plow, E.F. (1980) Interaction of fibrinogen with i t s p l a t e l e t receptor as part of a multistep reaction i n ADP-induced p l a t e l e t aggregation. J . B i o l . Chem., 10:154-161. - 217 -Marguerie, G.A. and Plow, E.F. (1981) Interaction of fibrinogen with i t s platelet receptor: kinetics and effect of pH and temperature. Biochem., 20:1074-1080. Martin, J.F., Kilbey, R.S. and Paterson, A.D. (1982) Platelet s i a l i c acid in malignant disease. Haemostas., 11:128-131. Mason, R.G. and Shermer (1971) Cell electrophoresis: Reflections of platelet reactive status. In The Platelet. Edited by K.M. Brinkhous, R.W. Shermar and F.K. Mostofi. Williams and Wilkins, Baltimore, Maryland, pp. 123-132. Masouredis, S.P. (1972) Cli n i c a l use of whole blood. In Hematology. Edited by W.J. Williams et a l . McGraw H i l l Book Co., New York, pp. 1308-1319. Massini, P., Glanzmann, R.K. and Luscher, E.F. (1978) Movement of calcium ions and their role in the activation of platelets. Thrombos. Haemostas., 40:212-218. Mather, L.E. and Cousins, M.J. (1979) Local anesthetics and their c l i n i c a l use. Drugs, 18:185-205. McGregor, J.L., Clemetson, K.J., James, E. and Dechavanne, M. (1979) A comparison of techniques used to study externally oriented proteins and glycoproteins of human blood platelets. Thromb. Res., 16:437-452. McGregor, J.L., Clemetson, K.J., James, E., Greenland, T. and Dechavanne, M. (1979) Identification of human platelet glycoproteins in SDS-polyacrylamide gels using 125i_i abelled lectins. Thromb. Res., 16:825-831. McLean, J.R., Maxwell, R.E. and Heartier, D. (1964) Fibrinogen and adenosine diphosphate induced aggregation of platelets. Nature, 202:605-606. Mehrishi, J.N. (1970) Phosphate groups on the surface of human blood plate-lets. Nature, 226:452-453. Mehrishi, J.N. (1970) Positively charged amino groups on the surface of normal and cancer c e l l s . Eur op. J. Cancer, j5:127-137. Michal, F. and Born, G.V.R. (1971) Effect of the rapid shape change of plate-lets on the transmission and scattering of light through plasma. Nature, 231: 220-222. Mills, D.C.B., Robb, I.A. and Roberts, G.C.K. (1968) The release of nucleo-tides, 5-hydroxytryptamine and enzymes from human blood platelets during aggregation. J. Physiol., 195:715-729. Milton, J.G. and Frojmovic, M.M. (1977a) Discocyte-echinocyte transformation in human platelets: geometries of echinocyte prepared with osmotic stress and aggregating agent. Thrombos. Haemostas., 38:278 . Milton, J.G. and Frojmovic, M.M. (1977b) Measurement of amount of invaginated membrane in mammalian platelets by osmotically-induced spherocyte formation. Thrombos. Haemostas., 38:277 . - 218 -Milton, J.G. and Frojmovic, M.M. (1979) Invaginated plasma membrane of human platelets: evagination and measurement in normal and "giant" platelets. J. Lab. Clin. Med., 93^:162-170. Mohammad, S.F. and Mason, R.G. (1981) A human model for study of blood-vascular wall interactions. Arch. Pathol. Lab. Med., 105:62-66. Morgenstern, E. and Kho, A. (1977) Morphometrische Untersuchungen an Blutplattchen. Cytobiol., 15:233-249. Morrison, M. and Bayse, G.S. (1970) Catalysis of iodination by lacto-peroxidase. Biochem. , _9: 2995-3000. Mosher, D.F., Vaheri, A., Choate, J.J. and Gahmberg, C.G. (1979) Action of thrombin on surface glycoproteins of human platelets. Blood, 53:437-445. Motamed, M., Michal, F. and Born, G.V.R. (1976) Increase in s i a l i c acid removable by neuraminidase during the shape change of platelets. Biochem. J_^, 158:655-657. Muraki, H., Watanabe, K., Ando, Y., Toyama, K. and Hasegawa, M. (1977) Reduction of surface negative charge in human platelets by ristocetin. Thrombos. Haemostas., 38:69. Mustard, J.F., Packham, M.A., Kinlough-Rathbone, R.L., Perry, D.W. and Regoeczi, E. (1978) Fibrinogen and ADP-induced platelet aggregation. Blood, 52:453-466. Mustard, J.F., Kinlough-Rathbone, R.L., Packham, M.A., Perry, D.W., Harfenist, E.J. and Pai, K.R.M. (1979) Comparison of fibrinogen association with normal and thrombasthenic platelets on exposure to ADP or chymotrypsin. Blood, 54:987-993. Nachman, R.L. and Leung, L.L.K. (1982) Complex formation of platelet membrane glycoproteins l i b and I l i a with fibrinogen. J. Clin. Invest., 69:263-269. Nachmias, V.T., Sullender, J.S. and Fallon, J.R. (1979) Effect of local anesthetics on human platelets: filopodial suppression and endogenous proteolysis. Blood, 53:63-72. Nachmias, V.T., Sullender, J.S. and Asch, A. (1977) Shape and cytoplasmic filaments in control and lidocaine-treated human platelets. Blood, 50:39-53. Nairn, H.Y., Clemetson, K.J. and Luscher, E.F. (1982) Effects of galactose-binding lectins on human blood platelets. Thromb. Res., 26:431-441. Nicolson, G.L., Smith, J.R. and Poste, G. (1976) Effects of local anesthetics on c e l l morphology and membrane-associated cytoskeleton organization in BALB/3T3 c e l l s . J. Cell Biol., 68:395-402. Niewiarowski, S., Budzynski, A.Z. and Lipinski, B. (1977) Significance of intact polypeptide chains of human fibrinogen in ADP-induced platelet aggregation. Blood, 49:635-644. - 219 -Nurden, A.T., Dupuis, D., Kunicki, T.J. and Caen, J.P. (1981) Analysis of the glycoprotein and protein composition of Bernard-Soulier platelets by single and two-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis. J. Clin. Invest., 67:1431-144Q. Nurden, A.T., Kunicki, T.J., Dupuis, D., Soria, C. and Caen, J.P. (1982) Specific protein and glycoprotein deficiencies in platelets isolated from 2 patients with gray platelet syndrome. Blood, 59:709-717. O'Brien, J.R. (1962) Platelet aggregation. J. Clin. Path., 15:446-455. Odink, J. (1976) Platelet preservation II. The response of human platelet suspensions to hypotonic stress. Thrombos. Haemostas., 36:182. Okumura, T. and Jamieson, G.A. (1976) Platelet glycocalicin. J. Biol. Chem., 251:5944-5949. Ostermann, G., Spangenberg, P. and T i l l , U. (1982) Reconsideration of quantitative phospholipid analysis in human gel-filtered blood platelets. Acta Haemat., 67:13-19. Packham,/ 1A., Guccione, M.A., Kinlough-Rathbone, R.L. and Mustard, J.F. (1980) Platelet s i a l i c acid and platelet survival after aggregation by ADP. Blood, 56:876-880. Papahadjopoulos, D. (1972) Studies on the mechanism of action of local anesthetics with phospholipid model membranes. Biochim. Biophys. Acta, 265: 169-186. Patscheke, H., Brossmer, R. and Worner, P. (1977) D-galactose binding lectins induce a differential response of blood platelets. Biochem. Biophys. Res.  Commun., 75:200-206. Patscheke, H. and Worner, P. (1977) Common activation of aggregation and release reaction of platelets. Thromb. Res., 11:391-402. Peerschke, E.I. and Zucker, M.B. (1978) Shape change and the percentage of si a l i c acid removed by neuraminidase from human platelets. Proc. Soc. Expt.  Biol. Med., 159:54-56. Peerschke, E.I., Grant, R.A. and Zucker, M.B. (1980) Decreased association of calcium with platelets unable to aggregate due to thrombasthenia or prolonged calcium deprivation. Br. J. Haematol., 46:247-256. Peerschke, E.I. and Zucker, M.B. (1980) Relationship of ADP-induced f i b r i n -ogen binding to platelet shape change and aggregation elucidated by use of colchicine and cytochalasin B. Thromb. Haemostas., 48:58-60. Peerschke, E.I., Zucker, M.B., Grant, R.A., Egan, J.J. and Johnson, M.M. (1980) Correlation between fibrinogen binding to human platelets and platelet aggregation. Blood, 55:841-847. - 220 -Peerschke, E.I., Zucker, M.B. and Rotman, A. (1981) Characterization of the ADP induced fibrinogen binding sites on the human platelets using photo-aff i n i t y techniques. Thrombos. Haemostas., 46:193. Peters, B.P., Ebisu, S., Goldstein, I.J. and Flashner, M. (1979) Interaction of wheat germ agglutinin with s i a l i c acid. Biochem., 18:5505-5511. Phill i p s , D.R. (1972) Effect of trypsin on the exposed polypeptides and glycoproteins in the human platelet membrane. Biochem., 11:4582-4588. Ph i l l i p s , D.R. and Agin, P.O. (1974) Thrombin substrates and the proteolytic site of thrombin actin of human-platelet plasma membranes. Biochim. Biophys.  Acta, 352:218-227. Phi l l i p s , D.R. and Agin, P.O. (1977) Platelet plasma membrane glycoproteins. Biochem. Biophys. Res. Commun., 75:940-947. Ph i l l i p s , D.R. (1979) Surface labelling as a tool to determine structure-function relationships of platelet plasma membrane glycoproteins. Thrombos.  Haemostas., 42:1638-1651. Phi l l i p s , D.R. (1980) An evaluation of membrane glycoproteins in platelet adhesion and aggregation. Prog. Hemostas. Thrombos., J5:81-109. Plow, E.F. and Marguerie, G.A. (1980) Participation of ADP in the binding of fibrinogen to thrombin-stimulated platelets. Blood, 56:553-555. Poison, A., Potgieter, M., Largier, J.F., Mears, G.E.F. and Joubert, F.J. (1964) The fractionation of protein mixtures by linear polymers of high molecular weight. Biochim. Biophys. Acta, 82:463-475. Poste, G., Papahadjopoulos, D. and Nicholson, G.L. (1975) Local anesthetics affect transmembrane cytoskeletal control of mobility and distribution of c e l l surface receptors. Proc. Nat. Acad. Sci., 72:4430-4434. Pye, J. and Curtain, CC. (1961) Electrophoretic, sedimentation and diffusion behaviour of crystalline neuraminidase from Vibrio cholerae. J. Gen.  Microbiol., 24:423-425. Reid, P.E., Culling, C.F.A., Tsang, W.C., Clay, M.G., Ramey, C.W. and Dunn, W.L. (1975) The determination of O-acetylated s i a l i c acid in colonic epithelial glycoproteins. Can. J. Biochem., 53:388-391. Reynolds, J. and Tanford, C. (1970) The gross confirmation of protein-sodium dodecyl sulfate complexes. J. Biol. Chem., 245:5161-5165. Rinderknecht, H., Geokas, M.C., Silverman, P. and Haverback, B.J. (1968) A new ultrasensitive method for the determination of proteolytic activity. Clin. Chim. Acta, 21:197-203. - 221 -Rock, G., Ordinas, A., Drouin, J., Wong, S.C. and Jamieson, G.A. (1980) Glycoprotein I is not involved in aggregation of platelets by wheat germ agglutinin. Thromb. Res., 19:725-728. Rodman, N.F. (1971) Morphological basis of platelet function. In The  Platelet. Edited by K.M. Brinkhous, R.W. Shermer and F.K. Mostofi. Williams and Wilkins, Baltimore, Maryland, pp. 55-70. Ross, S.W. and Ebert, R.V. (1959) Microelectrophoresis of blood platelets and the effects of dextran. J. Clin. Invest., 38:155-160. Saida, K. and Suzuki, A. (1981) Effect of on the sarcoplasmic reticulum in skined skeletal muscle fibres. Japan J. Pharmacol., 31:1084-1086. Scialla, S.J., Speckart, S.F., Haut, M.J. and Kimball, D.B. (1979) Alter-nations in platelet surface sialyItransferase activity and platelet aggregation in a group of cancer patients with high incidence of thrombosis. Cancer Res., 39:2031-2035. Schneider, M.D., Jenkins, C.S.P., Kahle, L.H., Sturk, A. and Cate, J.W.T. (1979) Studies on the mechanisms of ristocetin-induced platelet aggregation: binding of factor VIII to platelet. Br. J. Haematol., 43:99-112. Schwartz, M.L., Pizzo, S.V., H i l l , R.L. and McKee, P.A. (1971) The subunit structures of human plasma and platelet factor XIII. J. Biol. Chem., 246: 5851-5844. Seaman, G.V.F. and Heard, D.H. (1961) A microelectrophoresis chamber of small volume for use with biological systems. Blood, 18:599-604. Seaman, G.V.F. and Vassar, P.S. (1966) Changes in the electrokinetic prop-erties of platelets during their aggregation. Arch. Biochem. Biophys., 117: 10-17. Seaman, G.V.F., Vassar, P.S. and Kendall, M.J. (1979) Electrophoretic studies of human polymorphonuclear leukocytes and erythrocytes: the binding of calcium ions within the peripheral regions. Arch. Biochem. Biophys., 135:356-362. Seaman, G.V.F. and Brooks, D.E. (1970) Electrochemical aspects of platelet adhesion and aggregation. In Platelet Adhesion and Aggregation in  Thrombosis: Countermeasures. Edited by E.F. Mammen, G.F. Anderson and M.I. Barnhart. F.K. Schattauer Verlag, Stuttgart, West Germany, pp. 93-107. Seaman, G.V.F. (1976) Electrochemical features of platelet interactions. Thrombos. Res., 8_: (Suppl II): 235-246. Seaman, G.V.F. (1975) Electrokinetic behavior of red cell s . In The Red Blood  Cell. Edited by D.M. Surgenor. Academic Press, New York, pp. 1135-1229. - 222 -Shiimizu, M., Iwaguchi, T., Kikutani, N., Motomiya, T. and Yamazaki, H. (1979) A new automatic method for measuring c e l l electrophoretic mobility. Proc.  Japan Acad., 55:418-422. Skaer, R.J. (1981) Platelet degranulation. In Platelets In Biology and  Pathology. Edited by J.L. Gordon, Elsevier North Holland Biomedical Press, New York, pp. 321-348. Solum, N.O. and Peterka, M. (1977) Interactions of platelets with factor VHI-related protein. In Platelets and Thrombosis. Edited by D.C.B. Mills and F.I. Pareti. Academic Press, New York, pp. 21-31. Sorg, C. and Geczy, C.L. (1976) Antibodies to guinea pig lymphokines III. Reactions with radiolabeled lymphocyte activation products. Eur. J. Immunol., 6:688-693. Steck, T.L. and Dawson, G. (1974) Topographical distribution of complex carbohydrates in the erythrocyte membrane. J. Biol. Chem., 249:2135-2142. Stibbe, J. and Holmsen, H. (1977) Effects of sodium azide on platelet function. Thrombos. Haemostas., 38:1042-1053. Stoltz, J.F. (1971) Electrokinetic charge and platelet aggregation. In Platelet Aggregation. Edited by J. Caen. Masson and Editeurs, Paris, France, pp. 213-218. Stoltz, J.F., Larcan, A. and Nicolas, A. (1975) Physico-chemical study of phosphate groups on the human platelet membrane. Biochem. Expt. Biol., 11: 359-363. Stoltz, J.F. and Nicolas, A. (1979) Analytical study of ionized or ionizable groups of platelet membrane. Blut, 38:103-117. Sturk, A., Burt, L.M., Hakvoort, T., Cate, J.W.T. and Crawford, N. (1982) The effect of storage on platelet morphology. Transfusion, 22:115-120. Takano, S. and Suzuki, T. (1981) Effects of changes in temperature and pH of plasma on the aggregation and electrophoretic mobility of rabbit platelets. Acta Haematol. Jap., 44:124-130. Tangen, 0., Berman, H.J. and Marfey, P. (1971) Gel f i l t r a t i o n , a new tech-nique for separation of blood platelets from plasma. Thrombos. Haemostas., 25:268-278. Taylor, P.M. and Heptinstall, S. (1980) The a b i l i t i e s of human blood plate-lets to bind extracellular calcium and to be aggregated by adenosine d i -phosphate are related. Br. J. Haematol., 46:115-122. Toor, B., McGregor, J.L., McGregor, L. and Clemetson, K.J. (1982) Comparison of the major membrane glycoproteins and proteins of human, rabbit and rat blood platelets. Thromb. Res., 26:317-328. - 223 -Vassar, P.S., Kendall, M.J. and Seaman, G.V.F. (1969) Electrophoresis of human leukocyte. Arch. Biochem. Biophys., 135:350-355. Vassar, P.S., Hards, J.M., Brooks, D.E., Hagenberger, B. and Seaman, G.V.F. (1972) Physicochemical effects of aldehydes on the human erythrocyte. J. Cell Biol., 53:809-818. Vassar, P.S., Hards, J.M. and Seaman, G.V.F. (1973) Surface properties of human lymphocytes. Biochim. Biophys. Acta, 291:107-115. Vuento, M. and Vaheri, A. (1978) Dissociation of fibronectin from gelatin-agarose by amino compounds. Biochem. J., 175:333-336. Walsh, P.W. (1981) Platelets and coagulation proteins. Fed. Proc., 40: 2086-2091. Weisel, J.W., Ph i l l i p s , G.N. and Cohen, C. (1981) A model from electron microscopy for the molecular structure of fibrinogen and f i b r i n . Nature, 289:263-266. Wester, J., van der Veen, J. and Sixma, J.J. (1977) Morphology of the hemo-statis plug. In Platelets and Thrombosis. Edited by D.C.B. Mills and F.I. Pareti. Academic Press, New York, pp. 97-109. White, J.G. (1970) A search for the platelet secretory pathway using electron dense tracers. Am. J. Path., 58:31-49 . White, J.G. (1972) Platelet morphology and function. In Hematology. Edited by W.J. Williams. McGraw-Hill Book Co., New York, pp. 1023-1039. White, J.G. and Clawson, CC. (1980) The surface-connected canalicular system of blood platelets. Am. J. Path., 101:353-364. Worthington Enzyme Manual (1972) Worthington Biochemical Corp., Freehold, New Jersey. Wright, CS. (1980) Crystallographic elucidation of the saccharide binding mode in wheat germ agglutinin and i t s biological significance. J. Mol.  Biol., 141:267-291. Wu, K.K. and Ku, C.S.L. (1978) Stimulation of platelet surface s i a l y l -transferase activity by platelet aggregating agents. Thromb. Res., 13: 183-192. Wunderlich, R.W. (1982) The effects of surface structure on the electro-phoretic mobilities of large particles. J. Colloid Interface Sci., 88: 385-397. Yamazaki, H. and Motomiya, T. (1980) Consumption of large platelets with decrease in adenine nucleotide content in thrombosis, disseminated intra-vascular coagulation and postoperative state. Thrombos. Res., 18:77-88. - 224 -Yamazaki, H., Ryoko, T., Motomiya, T., Jung, S.M. and Sonoda, M. (1980) The increase of the electrophoretic mobility of platelets after laporatomy. Thrombos. Haemostas., 44:43-45. Zucker-Franklin, D. (1969) Microfibrils of blood platelets: their relation-ship to microtubules and the contractile protein. J. Clin. Invest., 48: 165-175. Zucker, M.B. (1972) Platelet function. In Hematology. Edited by W.J. Williams. McGraw-Hill Book Co., New York, pp. 1014-1022. Zucker, M.B. and Grant, .R.A. (1978) Nonreversible loss of platelet aggreg-abil'ity induced by calcium deprivation. Blood, 52:505-514. Zucker, M.B. and Peerschke, E.I. (1980) Specific binding of fibrinogen to platelets: relationship to shape change and aggregation. In Platelets:  Cellular Response Mechanisms and Their Biological Significance. Edited by A. Rotman, F.A. Meyer, C. Gitler and A. Silberberg. John Wiley and Sons, New York, pp. 157-168. Zwaal, R.F.A. (1978) Membrane and l i p i d involvement in blood coagulation. Biochim. Biophys. Acta, 515:163-205. - 225 -APPENDIX STATISTICAL CONSIDERATIONS FOR THE Ca++ AND Mg**  BINDING EXPERIMENTS* The den s i t i e s of the d i v a l en t c a t i o n i c ( C + + ) b i n d i n g s i t e s on p l a t e l e t s were determined acco rd ing t o the f o l l o w i n g equat ion presented i n Chapter 3 + : _!_ _ J _ , _L_ 1 A C " 2en * 2enK tC + +]exp(2eVkT) _ exp(AG/kT)  K ' 55 6 was p l o t t e d aga inst ^c**]exp(2e^/(<T) ' l " i n e a r r e g r e s s i o n was used t o o b t a i n a l i n e whose s lope represents /^enK t n e i n t e r c e p t w i t h the o rd inate gives /^en • ^he c o e f f i c i e n t of v a r i a t i o n ( c . v . ) f o r a p a r t i c u l a r j^cy was equal to the c .v . of the mean m o b i l i t y f o r that p o i n t , i . e : where s .d . = s tandard d e v i a t i o n * A l l s t a t i s t i c a l methods are acco rd ing t o K a l b f l e i s c h (1974). + D e t a i l s of the equat ion and a f u l l l i s t of the symbols can be found i n Chapter 3. - 226 -This i s derived by calculating the absolute value of the f i r s t derivative of l ^ a : S(ka) = S ( A C f ) (ACrf Therefore %{\<j) _ S ( A a ) A O since (VACT) cr = constant * JJ SOu) Hence, the uncertainty i n (lAc) i s given by the uncertainty i n (u), provided both uncertainties are expressed as fractions of the appropriate mean. In situations like the present one, where reciprocal values are plotted against each other, linear regression is valid only if the variances of the data points are not significantly different from each other. To test this, the variance of the f i r s t J&cr data point on each regression line was compared to the variances of a l l the other l^cr data points on the same line point by point. Pairs of variances were tested using the F-distribution and a confidence interval of 95%. It was determined that the variances for the data points along each regression line were not significantly different from each other. The regression lines i n Fig. 23 are therefore valid. S t a t i s t i c a l comparisons for the binding site densities (N) and binding free energy (AG) between different platelet forms and cations were achieved by comparing the regression lines concerned using analysis of variance tables. They are shown below. The number of degrees of freedom i s based on the number of points on each line illustrated i n the figures, each point representing the average of at least 100 individual mobility determinations. Significance between the two intercepts reflects the significance between the binding site densities while that between the slopes reflects the significance between theAG. - 227 -Ca b i n d i n g ; comparing d i s c o c y t e vs ech inocyte : — Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o To ta l 19 3.33 x 10-5 -I nt er ce pt 1 1.64 x IO" 6 6. 14 Slope 1 4. 77 x 10-6 17. 86 Res idua l 16 4. 27 x IO" 6 -I n tercept 0.025 < p < 0.05; s i g n i f i c a n t d i f f e r e n c e . Slope 0.001 < p < 0.0005; s i g n i f i c a n t d i f f e r e n c e . M g + + b i n d i n g ; comparing d i s c o c y t e vs ech inocyte Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o To ta l 19 1.63 x I O - 5 I nt er ce pt 1 1. 32 x I O - 6 1 4 - 23 Slope 1 5. 18 x 10-6 55.69 Resi dual 16 7.44 x 10-7 Int er ce pt Slope 0. 005 < p < 0. 01; p < 0.0005; s i g n i f i c a n t d i f f e r e n c e , s i g n i f i c a n t d i f f e r e n c e . Discocytes ; comparing C a + + vs M g + + b i n d i n g : -Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o To ta l 15 2.46 x 10"5 I nt er ce pt 1 5.33 x 1 0 ~ 8 0. 1525 Slope 1 6. 78 x 10-8 o.1941 Res i dual 12 4. 19 x 10-6 I n tercept 0.50 < p < 0. 75 ; no s i g n i f i c a n t d i f f e r e n c e . Slope 0.50 < p <0.75; no s i g n i f i c a n t d i f f e r e n c e . - 228 -Ech inocyte; comparing C a + + vs M g + + b i n d i n g : — Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o To ta l 15 2.20 x 1 0 - 5 -I nt er ce pt 1 2.91 x 1 0 - 5 0.04 Slope 1 7. 36 x 10~ 8 1.07 Res i dual 12 6.88 x I O - 8 -I n tercept 0.50 Slope 0.25 < p < 0. 75 ; < p < 0. 50; no s i g n i f i c a n t d i f f e r e n c e no s i g n i f i c a n t d i f f e r e n c e . C a + + b i nd i n g ; comparing d i s c o c y t e vs spherocyte : — Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o To ta l 15 4.80 x IO" 4 -I nt er ce pt 1 3.71 x 10-^ 345.57 Slope 1 4.48 x 10 " 5 41.68 Res i dual 12 1.29 x 10-5 -Int er ce pt Slope 0.01 < p < 0. 005; p < 0.025; s i g n i f i c a n t d i f f e r ence s i g n i f i c a n t d i f f e r e n c e C a + + b i n d i n g ; comparing ech inocy te vs spherocyte :-Source of Var iance Degree of Freedom Sum of Squares Mean Square R a t i o Tota l 15 5.21 x K T * — I nt er ce pt 1 4.15 x IO - * 510. 96 Slope 1 4.92 x 10-5 60. 58 Res idua l 12 9.75 X IQ-6 I n tercept p < 0.005; s i g n i f i c a n t d i f f e r ence Slope 0.005 < p < 0.01; s i g n i f i c a n t d i f f e r e n c e . 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0095845/manifest

Comment

Related Items