7‘9"an Clo-0.0.1 "an"; ‘ In at" "my v. c 1.: 1:: 221:0 fllu§ b- mn-N In pm {.0 «p -u, .3 r, .‘_.....‘.,.7 “fur ‘1“ ,., lllllllllllll}lllllllHHHIUIIIIIIIIllllllHl(llllllllllllHll 1293 01020 1592 This is to certify that the thesis entitled IDENTIFICATION OF PTA-l ON CANINE PLATELETS USING A HUMAN MONOCLONAL ANTIBODY, LEO-Al presented by MARY NINA DIPINTO, VMD has been accepted towards fulfillment of the requirements for MASTERS degree in SCIENCE JMWQMZ Major professor Date NOVEMBER 17. 1994 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Mlchlgan State Unlverslty PLACE DI RETURN BOXtomnwombehockwtfiom yournoord. To AVOlD FINES Man on or More data duo. DATE DUE DATE DUE DATE DUE MSU In An Affirmative Wan-I Opportunity 1m Wanna-m IDENTIFICATION OF PTA-1 ON CANINE PLATELETS USING A HUMAN MONOCLONAL ANTIBODY, LEO A-1 By Mary Nina DiPinto, VMD A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Pathology 1994 ABSTRACT IDENTIFICATION OF PTA-1 ON CANINE PLATELETS USING A HUMAN MONOCLONAL ANTIBODY, LEO A-1 By Mary Nina DiPinto, VMD Basset Hound Hereditary Thrombopathy (BHT), an autosomal recessive canine platelet function defect, causes a complete failure of primary platelet aggregation in affected animals. Previous studies have demonstrated a markedly decreased phosphorylation of a 65 kDa protein on SDS-PAGE in BHT platelets. This protein may be represented by PTA-1, a 65 kDa surface glycoprotein of human platelets and T-Iymphocytes, which is important in their activation. A human monoclonal antibody, Leo A-1, was utilized to identify PTA-1 on canine platelets, and to evaluate canine platelet responses. Western blotting revealed that Leo A-1 bound to BHT and control canine platelets at a higher molecular weight than in man under non-reducing conditions only. In radioligand binding studies, Leo A—1 demonstrated lower affinity for canine platelets than human. Leo A-1 did not induce platelet aggregation in normal or affected dogs. ACKNOWLEDGMENTS I wish to express my appreciation and thanks to the members of my graduate committee, Drs. Thomas G. Bell, Douglas W. Estry, Kenneth Schwartz, and Vilma Yuzbasiyan-Gurkan, for their encouragement, guidance, and support. Many thanks also to Dr. Edward C. Mather, Associate Dean for Research and Graduate Studies, without whose help this project would never have been completed. I would like to especially acknowledge the endless trust, loyalty, love, and cooperation of Agatha, Bertha, Buster, Button, Guinivere, Kewpie, Lacey, Lily, Max, Pippin, Sparrow, and Squire. Without them, of course, there would have been no project. Special thanks must also go to Jim for everything. TABLE OF CONTENTS LIST OF TABLES ...................................... v LIST OF FIGURES ...................................... vi LIST OF ABBREVIATIONS ................................ viii CHAPTER 1 INTRODUCTION ............................. 1 Basset hound hereditary thrombopathy ........... 10 Summary ............................... 21 CHAPTER 2 MATERIALS AND METHODS .................... 24 Experimental subjects ....................... 24 Part 1. Identification of platelet-thymocyte antigen (PTA—1) on canine platelets .................. 24 Experimental design ..................... 24 Monoclonal antibodies .................... 25 Preparation of platelet-rich plasma ............ 26 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis ........................ 26 Western blotting ........................ 28 Binding studies ......................... 30 Platelet surface labelling and immunoprecipitation . 33 Part II. Platelet function studies with Leo A-1 ....... 35 Experimental design ..................... 35 Platelet isolation ........................ 36 Gel filtration of platelets ................... 36 Platelet aggregation studies ................ 37 CHAPTER 3 RESULTS .................................. 38 Western blot analysis ....................... 38 Direct binding studies ....................... 44 Immunoprecipitation ................. I ....... 58 Effects of Leo A-I on platelet aggregation in vitro . . . . 61 CHAPTER 4 DISCUSSION .......... ’ ..................... 66 REFERENCES ......................................... 71 iv Table 1 Table 2a Table 2b Table 3 Table 4a Table 4b LIST OF TABLES Direct binding of 250 ng/ml 125I Leo A-1 to the platelet surface on man and dog. Binding of 125l Leo A-1 to the surface of human platelets. 5 x 108 platelets were incubated with con- centrations of 125l Leo A-1 ranging from 25 to 1000 ng/ml. Binding of 125I Leo A-1 to the surface of normal canine platelets. 5 x 108 platelets were incubated with 12“’I Leo A-1 in concentrations ranging from 25 to 1000 ng/ml. Direct binding of 125I Leo A-l to the surface of normal canine platelets. 2 x 107 platelets were incubated with increasing concentrations of 125I Leo A-1 ranging from 250 to 3000 ng/ml. Maximum aggregation (%), slope, time to maximal aggregation (min), of human, normal canine, and BHT- affected gel-filtered platelets to increasing concentra- tions of Leo A-1. Maximum aggregation (%), slope, and time to maxi- mum aggregation (min) of human, normal canine, and BHT-affected platelet-rich plasma to increasing con- centrations of Leo A-1. 45 49 50 54 64 65 Figure 1 Figure 2 Figure 3 Figure 4a Figure 4b Figure 5 Figure 6a LIST OF FIGURES Typical aggregation curve for normal canine platelet- rich plasma stimulated by 10 pM ADP. The arrow indicates the point of agonist addition, (a) shape change, (b) maximal aggregation. Proposed mechanisms of antibody-mediated platelet activation. (A) platelet surface antigen, (U) platelet Fc receptor, (Y) antibody molecule“. Typical aggregation curves for canine control and BHT-affected platelet-rich plasma stimulated by 10 [1M ADP. Western blot of non-reduced platelet proteins incu- bated with Leo A-1 monoclonal antibody. Lane A: BHT-affected dog, Lane B: normal dog, Lane C: human, Lane D: molecular weight standard, Lane 1: molecular weight standard, Lane 2: human platelet proteins. Lanes 1 and 2 are stained with an India ink total protein stain to evaluate efficacy of protein transfer to the nitrocellulose membrane. Western blot of reduced platelet proteins incubated with Leo A-1 monoclonal antibody. Lane A: BHT- affected dog, Lane B: normal dog, Lane C: human, Lane D: molecular weight standard. Direct binding of 250 ng/ml 12"I Leo A-1, HB 43, or IV 3 to the platelet surface in man and dog. A and B: human platelets, C and D: control canine platelets, E and F: BHT-affected platelets. Klotz plot depicting 1”I Leo A-1 binding to the surface of human platelets. vi 10 14 41 43 47 52 Figure 6b Figure 7 Figure 8 Klotz plot depicting 125I Leo A-1 binding to the surface of normal canine platelets. Autoradiograph following SDS-PAGE of immunopre- cipitated PTA-1 from canine and human surface ”5|- labelled platelets with Leo A-1. Lane A: human whole platelet Iysate, Lane B: normal canine whole platelet Iysate, Lane C: BHT whole platelet Iysate, Lane D: normal canine + Leo A-l, Lane E: BHT + Leo A-1, Lane F: human + Leo A-1, Lane G: normal canine + H8 43, Lane H: BHT + H3 43, Lane I: human + H8 43. The arrow represents PTA-1 immunoprecipitated from human platelet proteins. Platelet aggregation induced by Leo A-1 antibody. The aggregation curves illustrated demonstrate the response of gel-filtered human platelets to decreasing concentrations Leo A-1. a: 4 pg/ml, b: 2 ,ug/ml, c: 1 pg/ml, d: 0.4 pg/ml, e: 0.2 pg/ml. For comparison, aggregation trace f represents the response of normal and BHT-affected canine platelets to 4 pg/ml Leo A-I. vii 56 60 63 1,4,5 IP3 ADP BCIP BHT BSA Caz+ cAMP CPM DAG DEA GFP LIST OF ABBREVIATIONS inositol 1,4,5-triphosphate adenosine diphosphate 5-bromo-4-chloro-3-indoloyl phosphate Basset hound hereditary thrombopathy bovine serum albumin ionized calcium cyclic AMP counts per minute diacylglycerol diethanolamine gel-filtered platelets hydrogen peroxide monoclonal antibody to human IgG high performance liquid chromatography Hepes-Tyrodes-Albumin immunoglobulin G monoclonal antibody to human Fc receptors dissociation constant kflodahon viii LEO-A1 mA MoAb NBT ”9 PAF PBS PGE1 PKC PLC PMA PMSF PRP PTA-1 SDS SDS-PAGE TBS TxA2 U #9 vWF TxB2 monoclonal antibody to PTA-1 molar milliamperes monoclonal antibody nitro blue tetrazolium nanograms platelet activating factor phosphate buffered saline prostaglandin E1 protein kinase C phospholipase C phorbol myristate acetate phenyl methyl sulfonyl flouride platelet-rich plasma platelet-thymocyte antigen-1 sodium dodecyl sulfate sodium dodecyl sulfate polyacrylamide gel electrophoresis Tris buffered saline thromboxane A2 units micrograms von Willebrand’s factor thromboxane B2 CHAPTER 1 INTRODUCTION This thesis project arose as a sequel to a study of an inherited, severe hemorrhagic diathesis in purebred Basset hounds that has been ongoing at three universities since the late 19705. The support for extended study of this and other platelet disorders in humans and other mammalian species was in response to the major contributions that such studies have made to the current understanding of platelet biochemistry and physiology. Support for the initial investigation of Basset Hound Hereditary Thrombopathy (BHT) arose because of the superficial similarity of this disease to a specific inherited human platelet disorder. It has since been determined that BHT is not a model for any known human platelet disorder but represents a unique and as yet undefined platelet defect. Because a functional disturbance of a 65 kDa protein had been identified in BHT, the initial purpose of this project was to determine if a protein analogous to the human 65 kDa platelet-thymocyte antigen (PTA-1) is present and functional in the canine. Further, it is anticipated that the project may serve to refine what is known of canine platelet function and may be of value in increasing our understanding of the BHT defect. 2 Hemostasis is a complex interaction between platelets, soluble coagulation factors, and the vascular endothelium. Exposure of the subendo- thelium following vessel injury initiates the platelet response, which culminates in the formation of the primary hemostatic plug. The platelet response is a sequence of precisely coordinated events in which platelets adhere to the exposed subendothelium of the injured vessel, change shape, spread, begin to aggregate on the adherent platelet layer, and secrete the contents of their cytoplasmic granules. The concurrent production and release of arachidonic acid metabolites is also important in the recruitment of additional platelets into the forming coagulum. During activation, alterations in the platelet membrane provide phospholipid surfaces that enhance the activation of a number of soluble coagulation factors, amplifying the coagulation cascade, and initiating the conversion of soluble circulating fibrinogen to an insoluble fibrin monomer. Polymerization of fibrin then stabilizes the platelet aggregate, resulting in secondary hemostasiSI'sa. Adhesion of platelets to subendothelial components such as collagen and glycosaminoglycans is directly dependent upon the presence of platelet surface receptors for adhesive proteins such as von Willebrand’s factor (vWF), fibrinogen, collagen, and fibronectin‘. The major function of the surface glycoprotein lb-IX is to bind vWF in the subendothelium and it is essential for the initial adhesion of platelets and the formation of the platelet monolayer covering the site of injury. Even though non-activated circulating platelets do not bind soluble vWF in plasma, they will adhere to vWF once the latter protein 3 becomes reconformed on the endothelial surface”. However, all other platelet adhesion molecules only become functional receptors in response to platelet agonist-mediated activations. This may occur in response to soluble agonists that are present in the surrounding milieu, such as ADP and thrombin, or agonists that are exposed during vessel injury, such as collagen. The interaction of these extracellular "agonists" with their specific platelet surface receptors through the signal transduction process initiates platelet activation, exposure and induction of receptors for adhesion molecules, and recruitment of additional platelets to the site of injury. An important example of receptor induction is demonstrated by the fibrinogen receptor, glycoprotein lIb-llla (GP lIb-llla). This receptor is composed of two surface glycoproteins and is a member of the integrin supergene family. These proteins exist as a complex on the platelet surface, and at 50,000 copies per platelet, are the most abundant platelet receptors’. Platelet activation leads to conformational changes in this protein complex that permit the binding of fibrinogen”'“. Fibrinogen-receptor binding is calcium dependent, and is essential for platelet aggregation. Once bound to its receptor, fibrinogen is thought to cross- Iink adjacent platelets by virtue of its dimeric structure. GP IIb-Illa possesses many functional domains, and has the ability to bind other adhesion molecules besides fibrinogen, such as vitronectin, fibronectin, and von Willebrand's factorleFl‘z. Primary platelet aggregation, which consists of adhesion, shape change, and fibrinogen binding, is a reversible phenomenon. If the initiating stimulus is 4 of sufficient strength, primary aggregation will proceed to an irreversible secondary phase, which is accompanied by, and is dependent on, the platelet release response“? Platelet release involves both the secretion of alpha and dense granule contents and the production/release of arachidonic acid from membrane phospholipids. The contents of platelet alpha granules include coagulation factors (fibrinogen, factor V, factor VII), growth factors, mediators (thrombospondin, fibronectin, platelet factor 4), and vWF. Dense granules contain primarily ions (calcium and magnesium), adenine nucleotides (ADP and ATP), pyrophosphates, and serotonin4'18'19. Thus, the release response ensures that high concentrations of agonists, calcium, and adhesive proteins become localized on the surface of the forming platelet plug. Platelet agonists are classified as "weak", "intermediate", or "strong" based upon their ability to provoke irreversible secondary aggregation and the release response‘3'15. The optical aggregometer is the method most frequently used to study platelet aggregation in vitro. This instrument measures changes in light transmission when platelet agonists are added to stirred platelet suspensions, ”'2‘. Platelets respond to and the results are displayed on a strip chart recorder most agonists by an initial shape change converting the resting discoid morphology to a spiny sphere. This conformational alteration produces a small but detectable change in light transmission. Phorbol esters and epinephrine are agonists that do not induce shape change prior to inducing aggregation. As aggregation proceeds with the formation of the platelet clump, light transmis- sion through the suspension increases to a plateau (see Figure 1). Primary 5 aggregation is reversible, and not associated with platelet secretion, so a plateau phase may not be attained. However, if the agonist is strong, or present in high enough concentration, the release reaction will occur, resulting in irreversible secondary aggregation. Figure 1 outlines a typical curve generated during platelet aggregation stimulated by ADP, and measured by an aggregometer. The platelet response is ultimately determined by the balance between excitatory and inhibitory signals mediated by cell surface receptors. The signal transduction process, initiated by the binding of an agonist to its receptor, is mediated by transducer molecules known as G proteins. G proteins act to transmit the signal across the plasma membrane and regulate the function of specific effector mechanisms such as enzymes or ion channelse'zs'”. Although platelets respond to a wide variety of stimuli, they possess only a finite number of effector mechanisms by which these cellular reactions can be mediated. These effector pathways determine cellular responses by controlling the production of intracellular second messengers, which then modulate the structure, activity, or phosphorylation of intracellular proteins"6"3"6'”. The signal transduction process is tightly regulated and contains many backup systems that permit communication between different parts of the platelet activation sequence. This arrangement provides opportunities for modulation (i.e., amplification or inhibition) of the signal at any step. Figure 1. Typical aggregation curve for normal canine platelet-rich plasma stimulated by 10 pM ADP. The arrow indicates the point of agonist addition, (a) shape change, (b) maximal aggregation. 0 5 r - 90 Light Transmission 1 Min 8 Several enzyme systems play a pivotal role in platelet signal transduction: 1) phospholipase C (PLC), which mediates the hydrolysis of membrane inositol phospholipids, resulting in generation of the second messengers inositol 1,4,5 triphosphate {(1,4,5) 1P3}, and sn-1,2-diacylglycerol (1,2 DAG); 2) protein kinase C (PKC), which phosphorylates various intracellular proteins; 3) phospholipase A2, which cleaves arachidonic acid from membrane phospho- lipids; and 4) adenylate cyclase, which controls the generation of cyclic AMP (CAMP), an important inhibitory second messenger‘"3"7. Physiologic platelet agonists such as thrombin, platelet activating factor (PAF), and thromboxane A2 (TxAZ) transmit their signal across the platelet membrane via the receptor-mediated hydrolysis of membrane phosphoinositides by phospholipase C (PLC). This generates the second messengers 1,2 DAG and 1,4,5 IP3, which activate protein kinase C (PKC) and increase cytosolic calcium concentrations”. These events are associated with the phosphoryla- tion of the 47 kDa substrate of PKC and the 20 kDa myosin light chain. The 47 kDa protein has been purified”, and its cDNA sequencedzz'. This protein is called "pleckstrin", and it may represent the lP3 5'-monoesterase that is phosphorylated and activated by PKC“, although its function is still controver- sial22"23'2‘. Platelets may also be activated by non-physiologic agonists such as phorbol myristate acetate (PMA), a tumor-promoting phorbol ester. PMA intercalates into the lipid bilayer of the platelet plasma membrane, and directly activates PKC by virtue of its structural similarity to 1,2 DAG”. Free arachidonic acid cleaved from membrane phospholipids is metabolized to TxA2 9 and other eicosanoids, which function as intra- and inter-cellular messengers. Arachidonic acid and its cyclooxygenase metabolites are of major importance in normal platelet function. Weak physiologic agonists such as ADP, epineph- rine, and low concentrations of PAF and thrombin require arachidonic acid to be liberated and metabolized to TxA2 for irreversible aggregation and dense granule secretion to occur”. An interesting class of platelet agonists is represented by antiplatelet lgG monoclonal antibodies (MoAb), autoantibodies, and alloantibodies, all of which are capable of inducing platelet aggregation and secretion. These immuno- globulins can bind via their Fab components to specific platelet glycoproteins or to proteins that are adsorbed to the platelet surface”. Of this group, the most is known about monoclonal antibodies, which have been extensively used to identify, quantify, and isolate platelet membrane glycoproteins“‘". The majority of platelet-activating MoAb are of the lgGl subclass“. It has been shown that these MoAb also require binding of the antibody to the platelet Fc receptor to initiate platelet aggregation and release”. Platelets express approximately 1500 Fc receptors of a single class (Fclel) on the plasma membrane“. MoAb can initiate platelet activation following Fc receptor occupation on adjacent platelets (termed interplatelet activation), or, through antibody binding to the F“, and Fc receptors on the same platelet (intraplatelet activation)“. These mechanisms of antibody-mediated platelet activation are diagrammed below. There is also evidence to suggest that intraplatelet Fc receptor cross-linking may also be an important element in 10 MoAb activation of platelets“. Platelet activation by MoAb binding appears to be mediated by G protein activation of PLC and the generation of 1,2 DAG”. A B I_ntr_ap1atelet Interplatelet Figure 2. Proposed mechanisms of antibody-mediated platelet activation. ( A) platelet surface antigen, (U) platelet Fc receptor, (Y) antibody molecule“. Basset hound hereditary thrombopathy The study of inherited disorders of platelet function has contributed a great deal to the current understanding of platelet biochemistry and physiology. Identification of absent or dysfunctional glycoproteins, enzymes, and other structures has helped to clarify their role in normal platelet function. An inherited, severe hemorrhagic diathesis in purebred Basset hounds was first described in 1979”. Animals affected with the disorder exhibit clinical signs of platelet dysfunction such as mucosal bleeding, petechiation, easy bruising, and excessive hemorrhage associated with estrus cycles and the loss of deciduous teeth. Pedigree analysis of 92 related animals“, as well as breeding studies conducted in our own colony strongly support an autosomal recessive mode of inheritance. To date, no similar disorder has been described in man or other animal species. 1 1 The initial investigation of BHT centered around its potential as an animal model for Glanzmann's thrombasthenia, a hereditary disease of human platelets in which the membrane GP llb-llla complex is either absent or dysfunctional”. As the fibrinogen receptor, GP llb-llla is essential for platelet-to-platelet adhesion during aggregation". Affected dogs share many clinical features with Glanzmann’s patients, such as prolonged bleeding times, normal coagulation profiles and von Willebrand’s factor levels, normal platelet numbers and morphology, as well as the failure of primary platelet aggregation in response to ADP or collagen. It was found, however, that affected BHT platelets have normal GP Ilb-lllA content”, and bind I-125 or gold-labelled fibrinogen to the P3“31 . Also, unlike same extent as control platelets when stimulated by AD Glanzmann’s patients, affected Basset hounds demonstrate normal clot retraction. Therefore, it was apparent that BHT is not a model for Glanzmann's thrombasthenia, but represents a unique platelet defect involving an as yet undefined post-fibrinogen binding event(s). The function of the different platelet activation pathways was investigat- ed in BHT by measuring aggregation and ATP dense granule secretion in response to a variety of physiological and non-physiological agonists. These included ADP, PAF, thrombin, collagen, the thromboxane-mimetic U46619 plus epinephrine, PMA and the calcium ionophore A-2318732. Although affected platelets changed shape in response to all physiologic agonists tested, they did not aggregate unless activated by PMA or high concentrations of thrombin, both of which are able to bypass phospholipase C and the production of the 12 arachidonic acid metabolite TxA2‘3. The method by which PMA-mediated activation of protein kinase C (PKC) induces aggregation is currently unknown. Figure 3 illustrates typical aggregation curves for control and BHT-affected platelet-rich plasma following stimulation by ADP. Dense granule ATP secretion from affected platelets occurred in response to some, but not all, agonists, and the release response was not linked to the ability to aggregate. ATP secretion in the affected platelets was also inhibited by aspirin to a greater degree than in controls, suggesting an increased dependence on arachidonate metabolites”. In BHT, the evaluation of TxA2 production, when measured as the inactive metabolite Tsz, revealed that affected platelets produce significantly more Tsz than controls in response to those agonists that induce secretion, but decreased Tsz in response to those agonists that do not cause secretion. These studies confirm the hypothesis that dense granule release of ATP is strongly linked to TXA2 production in affected platelets. Although the TxA2 production pathway appears to be present and functional in BHT platelets, it does not appear to be properly regulated”. In addition, dense granule secretion occurs earlier and more rapidly in affected platelets“. 13 Figure 3. Typical aggregation curves for canine control and BHT-affected platelet-rich plasma stimulated by 10 pM ADP. III 100 "I" Control % Light Transmission Affected 1 mln 15 It was hypothesized that the BHT defect might involve a dysfunction in the signal transduction system and/or the generation of second messengers within the platelet. Due to the importance of cAMP as an inhibitory second messenger, Boudreaux et al. examined both the concentrations of cAMP and the activity of the cAMP phosphodiesterase in control and affected platelets. It was found that affected platelets had increased resting levels of cAMP33. It was concluded that affected platelets have functionally intact regulatory G proteins that regulate cAMP production. cAMP phosphodiesterase, which catalyzes cAMP removal and breakdown, is also functional. In BHT, however, the control of this enzyme appears to be impaired“. Although concentrations of cAMP are slightly but significantly increased in affected BHT platelets, it is highly unlikely that this alteration is responsible for the failure of primary aggregation, particularly since all other post-activational events, including phosphorylation of the 20 and 47 kDa proteins, occur normally. One hypothesis for this elevation is that it is merely a reflection of an underlying primary control defect within the affected platelet. It would be useful to determine if the derangements in arachidonate metabolism seen in affected platelets also produce increased concentrations of those prostanoids (i.e., PGE1/PGE2) that act to increase cAMP. Studies in our laboratory examined signal transduction more closely in control and affected platelets through the measurement of a number of end products: 1) resting and post-stimulation cytosolic free ionized calcium concentrations”, 2) cytosolic pH changes in response to agonist stimulation”, 16 3) PLC activity through the generation of the intracellular second messenger 1,2 DAG35, and 4) resting and post-stimulation phosphorylation of the 20 and 47 kDa proteins in 32-P—labelled platelets”. The activity of many intracellular effector systems is regulated by cytosolic free ionized calcium concentrations. Calcium also affects the organization of platelet cytoskeletal proteins‘3'36. The measurement of resting and post-stimulation cytosolic calcium concentrations in fura-2 or aequorin- loaded platelets indicated that calcium fluxes across the plasma membrane are normal in affected platelets. However, affected platelets released less calcium from internal stores than controls”. Despite this, the amount of calcium released from internal stores in affected platelets was adequate to support calcium-dependent processes such as shape change and the activation of the myosin light chain kinase, in the absence of external calcium”. Cytosolic alkalinization frequently accompanies platelet activation, and 37-33, release of internal appears to potentiate phospholipase A2 activation calcium, and cytoskeletal organization”. Regulation of cytosolic pH is dependent upon the Na*/H+ antiport, which is closely associated with the 64 kDa alphaz-adrenergic receptor in the platelet plasma membrane“. Measure- ment of cytosolic pH indicated that the Na*/H+ antiport is able to generate adequate alkalinity during activation to support phospholipase A2 activity in affected platelets”. PLC-mediated hydrolysis of membrane inositol phospholipids generates two important second messengers in the platelet signal transduction pathway: 17 1,2 DAG and (1 ,4,5)IP3. PLC is activated through agonist-specific receptors which are coupled to the enzyme via G proteins”. The activity and integrity of this enzyme system were indirectly examined by monitoring the production of 1,2 DAG in affected and control platelets in response to agonist stimulation. Affected and control dogs produced similar quantities of 1,2 DAG, indicating that PLC activity is not impaired in affected platelets”. Stimulation of platelets by agonists results in the phosphorylation of a number of proteins. Following electrophoresis and autoradiography of human platelets, agonist-induced phosphorylation is prominent in two protein bands. One is the 20 kDa myosin light chain, which is phosphorylated following activation of its calcium-dependent kinase“. The other band is the 47 kDa pleckstrin, the main substrate of PKC”. Phosphorylation of the 20 and 47 kDa proteins in response to a range of agonists was similar in control and affected 32-P labelled canine platelets”. Phosphorylation of the 20 kDa band correlated with shape change, but phosphorylation of neither the 20 nor the 47 kDa protein correlated with the ability to aggregate. However, a consistent finding on autoradiographs was markedly decreased phosphorylation of a 65 kDa band in affected platelets compared with controls. This difference was evident both before and after agonist stimulation. Minimal time-dependent phosphorylation of this 65 kDa band was observed only in response to those agonists capable of inducing aggregation in affected platelets: PMA or thrombin”. In the previous study, it was not possible to determine if this 65 kDa area was a single band, or several bands that migrated closely together. 18 Subsequent work that I performed in our laboratory utilizing gradient and lower %T resolving gels with more sensitive silver/copper stain techniques has generated data that supports the conclusion that this 65 kDa band migrates as, and is representative of, a single band. The decreased intensity of this band on autoradiography may represent either a missing protein or the inability to phosphorylate an existing band. This phenomenon of reduced phosphorylation raises several possibilities that could explain the results observed: 1) that this 65 kDa protein is defective or dysfunctional and cannot be phosphorylated normally, 2) the protein is present, but in reduced amounts, 3) the protein kinase responsible for phosphorylation of the 65 kDa protein is defective or dysfunctional, or 4) that the protein kinase is present in reduced concentra- tions. When reduced SDS-PAGE gels stained with Coomassie blue from affected and control dogs were examined both visually and densitometrically, no difference in the 65 kDa band could be appreciated between the two groups. These data tend to favor the possibility that either the 65 kDa protein or its specific protein kinase are defective or dysfunctional, rather than present in reduced amounts. If the latter were true, one would have expected to detect a densitometric difference in the band or bands in the Coomassie-stained gels. Currently, the identity, cellular location, and function of the protein in this 65 kDa band remains unknown, and its relationship to the BHT defect is a mystery. Several proteins of similar size have been identified on human platelets. One is the alphaz—adrenergic receptor, a 64 kDa glycoprotein which is associated with the Na“/H+ antiport“. It is unknown if phosphorylation of 19 this protein occurs, or if phosphorylation is important in the regulation of alphaz-adrenergic receptor activity. With the very low copy numbers present on human platelets (only 200 copies/platelet“), it is unlikely to be detected and observed by SDS-PAGE. A likely candidate is Platelet Thymocyte Antigen (PTA-1), which shares many features with the poorly phosphorylated 65 kDa protein seen in platelets from affected Basset hounds. A 65 kDa membrane glycoprotein, PTA-1 was first identified on T-Iymphocytes in the mid-19805 using monoclonal antibody, Leo A-1. Initially thought to be a T cell lineage-specific antigen, studies indicated that this protein was important in T cell activation“. However, subsequent investigation revealed that PTA-1 was also expressed on the surface of human platelets and their megakaryocytic precursors. In addition, an intracytoplasmic pool of PTA-1 was demonstrated to be present in membrane-bound vacuolar structures and the canalicular system in a fashion similar to that described for GP lIb-llla”. Biochemical characterization of PTA-1 shows that the identical protein is present on T lymphocytes and platelets (J. Scott, unpublished data). PTA-1 is a 65 kDa molecule that is composed of a 35-40 kDa protein backbone, with the remainder consisting of heavily sialated N-linked carbohydrate moieties. Partial sequencing of the amino terminus reveals that PTA-1 exhibits no identity with any other known platelet glycoprotein. Scatchard analysis of competitive binding assay data from human platelets indicated that there are approximately 1200 Leo A-1 binding sites per platelet”. This is a low copy number, and is 20 comparable to that seen with the platelet Fc receptor, which is also involved in certain types of antibody-mediated platelet activation“. Leo A-1 monoclonal antibody is a powerful platelet agonist, capable of stimulating both platelet aggregation and secretion. The binding of Leo A-1 induces irreversible platelet aggregation that is not preceded by shape change”. The aggregation response is distinct, and divided into two phases: an initial "slow" phase, which occurs independent of granule secretion and does not require fibrinogen; and a subsequent "fast" phase, during which the release reaction occurs”. This latter phase most resembles the platelet response seen with the secretion-dependent weak physiologic agonists. In dose-response evaluations of Leo A-1 induced aggregation, the time to onset of aggregation was inversely proportional to the concentration of antibody used. At very low antibody concentrations (i.e., < 0.3 pg/ml), this lag phase was as long as 5 minutes”. Leo A-1, like other platelet activating MoAb, appears to require antibody interaction with the platelet lg Fc receptor for the aggregation and release response to occur”. Previous studies suggest that platelet activation induced by the binding of Leo A-1 may be mediated by protein kinase C, rather than indirectly via a cyclooxygenase-dependent pathway or by the release of dense granule ADP, as agents that interfere with the activation of protein kinase C inhibit Leo A-1 induced platelet aggregation and secretion”. Leo A-1 aggregation curves are typical of those observed with other platelet activating MoAbs, and closely resemble those generated when platelets are stimulated 21 with phorbol ester. As previously stated, phorbol esters activate protein kinase C directly and irreversibly, unlike MoAb which act via G protein stimulation of PLC to activate PKC. Aggregation responses induced by phorbol esters are characterized by a lack of shape change, a slow aggregation response, and delayed granule secretion“. Both phorbol and Leo A—1 stimulate the phosphor- ylation of a 47 kDa protein of unknown function that is a substrate of protein kinase C5556. In addition, PTA-1 antigen is itself phosphorylated in response to the binding of Leo A-1 as well as phorbol. This response is inhibited by blockers of protein kinase C, further strengthening the case for the importance of this enzyme as a mediator of Leo A—1 platelet activation. Summary In summary, three metabolic derangements following activation have been identified in affected platelets: 1) altered arachidonate metabolism with overproduction of TxAz; 2) early, rapid dense body secretion, which is linked to 1); and 3) increased concentrations of cAMP. These derangements are most likely secondary manifestations of an underlying primary control defect because none of these alterations would result in complete failure of primary aggrega- tion. The 65-67 kDa protein, which is poorly phosphorylated in platelets from dogs affected with BHT, may play an important role in the pathogenesis of the defect. This protein exhibits physical and functional similarity to human platelet-thymocyte antigen (PTA-1). The purpose of the studies outlined in this 22 thesis was to determine if a protein analogous to the PTA-1 antigen in man was present in the dog. If this were the case further investigations may determine if PTA-1 represents the dysfunctional mechanism in BHT. This potential relationship was investigated by determining if the monoclonal antibody developed against human PTA-1, Leo A-1, exhibits functional and immunologi— cal cross-reactivity with canine platelet proteins. This was accomplished by: 1) functional studies investigating platelet aggregation responses to Leo A-1 monoclonal antibody in control and BHT affected Basset hounds; 2) the use of Western blotting techniques and radioligand binding analysis to determine if cross-reactivity existed between Leo A-1 and platelet proteins in control and BHT affected dogs, and if the specificity of any cross-reactivity was associated with the 65-67 kDa area; 3) if successful cross-reactivity was observed in the 65-67 kDa area, Leo A-1 would be utilized to immunoprecipitate the protein for further characterization in control and BHT affected dogs. The success in accomplishing these experimental procedures was dependent upon demonstration of the cross—reactivity of the human monoclonal antibody Leo A-1 with a canine platelet protein. Although cross-reactivity 23 between canine platelet proteins and monoclonal antibodies to analogous human proteins has been demonstrated”, it could not be assumed that: 1) the dog possesses an analogue to PTA-1, or 2) that even if such an analogue exists, it will bind Leo A-1 antibody. The experimental design was developed to first address these two issues. CHAPTER 2 MATERIALS AND METHODS Experimental subjects Experimental subjects consisted of Basset hounds from a colony established at Michigan State University for the study of Basset Hound Hereditary Thrombopathy (BHT). This colony is composed of 8 dogs, ranging in age from 3 to 7 years, of known genotype for the disorder. In the colony there is only a single known normal Basset hound; other healthy dogs of various breeds were recruited as a source of additional normal canine platelets. All canine subjects were receiving no medication (including heartworm preventative) at the time of blood collection. Human platelet proteins were utilized for comparison, and to link the study to established identification procedures for PTA-1. Part I. Identification of platelet-thymocyte antigen (PTA-1) on canine platelets Experimental design Preliminary experiments in this section were designed to identify PTA-1 in canine platelet preparations by immunodetection utilizing Western blotting. The Western blotting technique enables the detection of specific proteins in 24 25 complex mixtures following separation by gel electrophoresis. Platelet preparations from a total of 3 affected Basset hounds, 3 normal dogs (1 Basset and 2 non-Bassets), and a human being were used in the immunoblotting studies with human Leo A-1 monoclonal antibody. Each immunoblot compared human, normal dog, and BHT platelets under either reducing or non-reducing conditions. Non-specific binding of the primary and secondary antibody was evaluated in separate immunoblots utilizing HB 43, a murine monoclonal antibody to human lgG, as a negative control. Studies were then performed using 125l-labelled Leo A-1 to evaluate antibody binding to human, BHT, and normal dog platelets. Dose-response experiments with radiolabelled antibody were designed to evaluate binding affinity and to potentially estimate the number of Leo A-1 binding sites on canine platelets in comparison to human platelets. If Leo A-1 cross-reacted with canine platelets, the next step was to immunoprecipitate PTA-1 and the canine analogue from 12"l-labelled platelet preparations using Leo A-1 for accurate molecular weight determinations in man and dog. Monoclonal antibodies Leo A-1 monoclonal antibody was the generous gift of Dr. Judith Scott, University of Newcastle, New South Wales, Australia. This antibody, characterized as an lgGIh was developed by immunizing mice to intact, stimulated human T-lymphocytes“. Leo A-1 was precipitated from ascites fluid 26 and purified by HPLC to a concentration of 1.025 mg/ml in PBS as described by Scott”. HB 43 (murine anti-human lgG), and |V3 (murine anti-human Fc) monoclonal antibodies were the generous gift of Dr. Kenneth Schwartz. Preparation of pla telet-rich plasma For all studies, canine whole blood was collected by atraumatic jugular venipuncture into a plastic syringe containing 3.2% trisodium citrate at a ratio of 9 ml of blood to 1 ml of trisodium citrate. Samples from human subjects were obtained by phlebotomy from the median cubital vein into a plastic syringe containing 3.8% trisodium citrate in the same ratio as for canine whole blood. The citrated blood was then transferred to polypropylene plastic tubes, and the platelet-rich plasma obtained by repeated centrifugations at 1324 x g for 60 seconds. Platelet-rich plasma (PRP) was removed after each centrifuga- tion. A manual platelet count was then performed on the PRP using a Unopette and a Neubauer hemocytometer. All platelet preparations were used within four hours of collection. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis Preparation of pla te/et suspensions To minimize platelet activation during the subsequent centrifugation steps, prostaglandin E1 (PGE1) was added to PRP at a final concentration of 1 ”M. The platelets were allowed to rest at room temperature for 5 minutes, and 27 then sedimented by centrifugation at 730 x g for 15 minutes, washed twice, and the platelet pellet resuspended in the original plasma volume of Buffer #1 containing 120 mM NaCI, 13 mM trisodium citrate, and 30 mM glucose, pH 7.0. Between washes, the platelets were resuspended in 5 ml of Buffer #1 and a manual platelet count performed. Following the second wash, the platelet pellet was resuspended to a final count of 9 x 108/ml in Buffer #3 containing 138 mM NaCl, 2.9 mM KCI, 12 mM NaHCO3, 0.36 mM NaH2P04, 5.5 mM glucose, and 1.0 mM EDTA, pH 7.4. An aliquot of this suspension was solubilized by an equal volume of 4% w/v SDS, and the protein content determined by Markwell protein assay. The remaining platelet suspension was solubilized by the addition of an equal volume of a 2x concentrated solubiliza- tion buffer containing 4% w/v SDS, 10% w/v 2-mercaptoethanol, 20% w/v glycerol, 0.004% w/v pyronin—Y as the dye front marker, and 125 mM TRIS HCI, pH 6.8, and incubated at 100°C for 3 minutes. For non-reduced samples, distilled water was substituted for 2-mercaptoethanol in the solubilization buffer. Aliquoted samples were frozen at -70°C until used. SDS-PAGE Samples were electrophoresed according to the method of Laemmli57 through a 1.5 mm thick 3% T acrylamide stacking gel, and a 5-10% T acrylamide linear gradient resolving gel, using a Biorad Protean ll vertical slab gel electrophoresis apparatus (Biorad, Richmond, CA). The linear gradient gel was used to achieve optimal separation of the proteins in the 65 kDa area. 28 Lanes were loaded with 100 pg of protein or 10 pl of either high molecular weight standard or biotinylated high molecular weight standard (Biorad, Richmond, CA). Each gel run contained sample lanes with platelet preparations from a human, normal dog, and BHT-affected dog. The protein standards were prepared according to the manufacturer's instructions by dissolving them in a reducing buffer containing pyronin-Y as a dye front marker. Pyronin-Y is used because it will readily transfer to nitrocellulose during electroblotting, while the more commonly used bromphenol blue will not. The gels were run overnight at 9.5 mA/gel (950 V). Western blotting Electroblotting of the SDS-PAGE gels was performed according to the method described by Kyshe-Andersensfi'59 , using a Jansen semi-dry blotting apparatus. The gel was allowed to equilibrate in a solution of 25 mM TRIS prior to blotting. A section of nitrocellulose membrane cut to the same dimensions as the gel was soaked briefly in distilled water and then laid gently on top of the moist gel. Eighteen pieces of Whatman #1 filter paper were also cut to the exact size as the gel. Three layers were soaked in anode solution #2 (25 mM TRIS, 20% v/v methanol, pH 10.4) and placed gently on top of the nitrocellulose. This was followed by six layers of filter paper soaked in anode solution #1 (0.3 M Tris, 20% v/v methanol, pH 10.4). The gel was then carefully removed from the glass plate, and laid filter paper side down on the anodic plate of the apparatus. Nine layers of filter paper soaked in cathode 29 solution (40 mM 6-amino-n-hexanoic acid in 25 mM TRIS, pH 9.4) were added to the top of the gel. The lid, containing the cathodic plate, was placed on the stacked assembly and connected to the power supply (LKB). The transfer was carried out at a constant current of 8 mA/cm2 of gel at room temperature for 1 hour. Following completion of the blot-transfer, the portion of the membrane corresponding to the non-biotinylated molecular weight standard was cut away and stained with a total protein stain containing India ink to evaluate the efficacy of transfer. Protein detection on the remainder of the nitrocellulose membrane was performed using a commercial kit for mouse antibodies (Amersham, Chicago, IL). All steps were carried out at room temperature on a shaker plate. The blot was first incubated with reconstituted dried milk for one hour to prevent non— specific binding of the antibodies or detection reagents to the nitrocellulose. The blot was then incubated for one hour with the primary mouse antibody (Leo A-1), washed, and incubated with a biotinylated anti-mouse antibody (2° antibody), also for one hour. This preparation was subsequently washed, and incubated with a detection solution containing a streptavidin alkaline phospha- tase conjugate for 20 minutes. This was followed by a final wash, and the addition of a mixture of nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3- indoloyl phosphate (BCIP). NBT and BCIP were used to develop the color and visualize the protein band or bands of interest that may have bound the primary anfibody. 30 Molecular weights of proteins of interest on the blot were determined by constructing a standard curve from the standards of known molecular weight. Using semi-log paper, weights of the known proteins were plotted on the log axis versus the migration index for each protein. The migration index can be calculated as follows: Migration index = distance of protein migration (cm) distance of dye front migration (cm) This way, the migration index of the protein or proteins of interest can be calculated, and the corresponding molecular weight read from the standard curve. Binding studies Radiaiadina tion of Leo-A 7 Radioiodination of Leo A-1 was accomplished according to the method described by Fraker”. A 12 x 75 mm glass tube was first coated with 0.6 pg of iodogen (1,3,4,6-tetrachloro-3a, 6a-diphenylglycouril) (Sigma Chemical Company) dissolved in methylene chloride to a final concentration of 0.001 mg/ml. The methylene chloride was then allowed to evaporate. A PBS solution containing approximately 25 pg of Leo A-1 was added to the tube, followed by 0.25 millicuries of carrier-free Na++ l125 (DuPont, New England Nuclear). The tube was agitated for 10 minutes at room temperature, and the solution was 31 run over a Sephadex G-25 column to remove the unbound l‘25. The fraction containing the labeled immunoglobulin was collected in the void volume of the column, aliquoted, and stored frozen at -20°C. Direct binding analysis Leo A-1 binding studies were conducted according to the method described by Newman47 with some modifications. The assays were performed in duplicate at room temperature, utilizing PRP prepared as previously described. For each assay, PRP was obtained from a human subject, a normal Basset hound, and a BHT-affected Basset hound; platelet preparations were used within four hours of collection. Prostaglandin E, (PGE, final concentration 1.0 pM) was added to the PRP to minimize platelet activation during the subsequent centrifugation and handling. The platelets were washed twice by centrifugation at 3000 rpm for 10 minutes, followed by aspiration of the supernatant and gentle resuspension of the platelet pellet in the original volume of a buffer containing 3% w/v bovine serum albumin (BSA) in PBS, pH 6.5. A platelet count was performed after the second wash. In an Initial pilot study to evaluate the binding of Leo A-1 to human and canine platelets, 5.0 x 107 platelets/ml were incubated with 50 ng (final concentration 250 ng/ml) of 126l-labelled Leo A-1, HB 43, or IV3 monoclonal antibodies. HB 43 served as a negative control for canine platelets, and IV 3 as a control for low copy number (the number of Fc receptors on the human 32 platelet surface is approximately the same as PTA-1). The final volume of each tube was adjusted to 200 pl with 3% BSA/PBS. The tubes were capped, mixed gently, and incubated on a shaker plate for 1 hour. After incubation, the solution was layered over a 20% sucrose gradient (20% w/v sucrose in PBS), and spun at 8200 x g for 4 minutes. The tip of each tube was then cut off, and radioactivity quantified by counting for 2 minutes in a gamma counter. The value obtained represented the amount of activity bound to the surface of the platelets. 10 pl (50 ng) of 125l—labelled Leo A-I, HB 43, and IV 3 was also counted in the gamma counter representing total activity (i.e., bound plus free). In the later studies, 5.0 x 108 platelets/ml were incubated with 125l- labelled Leo A-1 in concentrations ranging from 5-200 ng (final concentration 25-1000 ng/ml), and the binding assay carried out as described above. A final study was performed on normal dog platelets with radiolabeled antibody concentrations from 50-600 ng (final concentration 250-3000 ng/ml) at an adjusted platelet count of 2 x 107/ml. A Klotz plot65 was used to determine if the platelet surface binding sites had been saturated with Leo A-1, to thereby validate a Scatchard analysis of the data to estimate the number of binding sites/platelet. Linear and non-linear regression of the binding data was performed using the EBDA and LIGAND programs modified for microcomputers“. 33 Platelet surface Iabelling and immunoprecipita tion PRP was prepared from platelet samples from a human donor, a normal dog, and a BHT-affected dog, 1 pM PGE, added, and the samples washed three times by centrifugation at 3000 rpm for 10 minutes. The platelet pellet was gently resuspended in the original plasma volume of a wash buffer containing 12 mM sodium citrate, 30 mM glucose, and 120 mM NaCl, pH 6.5. A manual platelet count was performed after the second wash as described previously. 1 x 109 platelets were surface-labelled using a Iactoperoxidase-catalyzed reaction. Pelleted platelets, after the final wash, were resuspended in 800 pl of wash buffer, and 200 pg Iactoperoxidase (1 mg/ml) was added to the suspension. 0.33 millicuries of 125l was added to each tube, followed by five 5 pl aliquots of 3 mM hydrogen peroxide at 10-second intervals accompanied by gentle agitation. The suspension was then diluted to 4 ml final volume with wash buffer plus 1 pM PGE,, and washed three times as described above. Following the last wash, the platelet pellet was resuspended in 1 ml 3% BSA/PBS with 1 pM added PGE,, and 500 pl aliquots placed into 1.5 ml Eppendorf tubes. A small quantity (<10 pl) was preserved, and solubilized with an equal volume of 2x SDS-PAGE solubilization buffer to be utilized as a total platelet lysate sample. 5 pg of Leo A-1 or negative control H843 (antibody concentration of both MoAb = 1 mg/ml) was then added to appropriately labeled tubes, which were incubated on a shaker plate at room temperature for 1 hour. The platelet suspension was then centrifuged at 8200 x g for 4 minutes, and the superna- 34 tant aspirated and discarded. The pellet was resuspended in 200 pl of lysis buffer (0.1 M PBS, 1% v/v Triton X-100, 1 mM PMSF, 1 mM EDTA, 1% w/v BSA). 20 pg of anti-mouse lgG agarose (200 pg/ml, Sigma Chemical Co.), which had been previously blocked for 1 hour by incubation with a human non- radiolabelled platelet lysate corresponding to 2 x 108 platelets/50 pl, and washed three times in lysis buffer, was then added to the platelet lysate in each Eppendorf tube. The lysate/agarose suspension was incubated on a rocker plate at room temperature for 1 hour. Following this second incubation, the labelled platelet lysate/agarose suspension was washed by centrifugation at 8200 x g for 5 seconds, followed by resuspension in lysis buffer. The lysate was washed and resuspended until the radioactivity of the supernatant decreased to a plateau (about 5 washes). The resultant pellets were resus- pended in 50 pl of SDS-PAGE sample buffer (non-reduced) containing 4% w/v SDS, 20% v/v glycerol, .0004% w/v bromphenol blue, and 125 mM Tris HCL, pH 6.8 and heated to boiling for 5 minutes to elute PTA-1 from the agarose. This suspension was then centrifuged at 8200 x g for 5 seconds. The supernatant was decanted, and total counts estimated by use of a Geiger counter. Aliquots of this supernatant were then loaded into individual wells of an SDS-PAGE slab gel (3% T acrylamide stacking gel, 7%T acrylamide resolving gel). The amount loaded was dependent on the cpm of the supernatant. From the small quantity of previously preserved undiluted labeled platelet lysate containing no agarose, 1 x 10" cpm was loaded/lane for man and dog. The 35 SDS-PAGE gel was run at 7.5 mA overnight, and stained with Coomassie blue to illuminate and determine the molecular weight of the protein bands. For detection of immunoprecipitated proteins by autoradiography, the labelled gels were dried onto porous cellophane in a slab gel drier (Biorad, Richmond, CA). The dried gel was then exposed on Kodak X-OMAT AR film at -80°C for 24-36 hours in a cassette with intensifying screens, and the film developed in Kodak GBX developer. Part II. Platelet function studies with Leo A-1 Experimental design Once binding studies had determined that PTA-1 or a similar protein was indeed present on canine platelets, the goal of the next series of experiments was to see if the function of this protein was the same in man and dog. As Leo A—1 antibody has been shown to be a potent activator of human platelets, functional studies in this section examined the ability of this monoclonal antibody to aggregate canine platelets. Aggregation studies were performed on PRP and gel-filtered platelets from three normal dogs (1 Basset, 2 non-Bassets), three affected dogs, and 3 human subjects. As a control to validate the normalcy of platelet responses, 10 pM ADP and 0.2U/ml thrombin were used in PRP and gel-filtered preparations, respectively. Platelet suspensions were then stimulated with serial dilutions of Leo A-I in PBS with final concentrations in PRP ranging from 0.04-4 pg/ml. In 36 man, Leo A-I is reported to induce half-maximal aggregation at a concentration of 1.5 pg/ml52. Platelet isolation Whole blood was collected, PRP prepared, and a manual platelet count performed as described previously in Part I. Following removal of the PRP, the blood was centrifuged at 1324 x g for 13 minutes to separate platelet-poor plasma. The platelet count was then adjusted using the autologous platelet poor plasma to a count of 3 x 10° platelets/ml. 0.5 ml of platelet-rich plasma was then aliquoted into aggregometer cuvettes, covered with parafilm, and the platelets allowed to rest for 30 minutes at room temperature. All platelet preparations were used within four hours of collection. Gel filtration of platelets In the preparation of gel filtered platelets, whole blood was collected in 3.8% trisodium citrate in all subjects, and the PRP collected as described above. Prostaglandin E, was added to a concentration of 1 pM to minimize platelet activation during the subsequent handling steps and gel filtration. The platelets were allowed to rest for 10 minutes at room temperature, and were then pelleted by centrifugation at 800 x g for 15 minutes. The supernatant plasma was discarded, and the pellet gently resuspended in 1-1.5 ml of HEPES- Tyrodes-albumin buffer (HTA) containing 130 mM NaCl, 2.6 mM KCI, 0.42 mM NaH2P04, 5.5 mM glucose, 0.01 mM HEPES, and 3.0 mg/ml BSA, pH 7.2. The 37 platelet suspension was layered onto a polyethylene column containing Sepharose 4-8 (10 ml bed volume) that had been equilibrated at room temperature with HTA. As platelets traversed the void volume of the column, they were detected visually. Two to three ml of gel-filtered platelets were collected, a manual platelet count performed, and the platelet number adjusted to 3 x 10°/ml with HTA. CaCl2 and MgCl2 were added to the platelet suspension to a final concentration of 1 mM and 0.8 mM, respectively, for canine platelets; and for human platelets, both were added at a concentration of 1 mM. The gel-filtered platelets were aliquoted into aggregometer cuvettes, and allowed to rest at room temperature for 30 minutes. Platelet aggregation studies All aggregation studies were performed in a Lumi-aggregometer (Chrono- log Corporation, Havertown, PA) at 37.5°C. Cuvettes containing 0.5 ml aliquots of PRP or gel-filtered platelets were allowed to equilibrate for at least 4 minutes at 37.5°C prior to being placed in the aggregometer. Following placement of the cuvette in the aggregometer, the platelet suspension was allowed to stir at 900 rpm for approximately 30 seconds and 20 pl of agonist (Leo A-1, ADP, or thrombin) was then added. Results were recorded on a dual pen Houston Omniscribe Chart Recorder (Houston Instrument Division of Bausch and Lamb, Inc., Houston, TX). Failure to record an increase in light transmission above baseline when the sample was permitted to stir for at least 15 minutes after the addition of agonist was designated as zero aggregation. CHAPTER 3 RESULTS Western blot analysis To determine if LEO A-1 recognized and bound a platelet protein(s) in man and dog, initial studies were carried out using Western blotting. Reduced and non-reduced platelet protein preparations from a human, normal dog, and a BHT-affected dog were applied‘to a 7-10%T polyacrylamide gel. The gel was electroblotted onto a nitrocellulose membrane, and the immunocomplex(es) identified using a biotinylated secondary antibody directed against murine IgG. The signal was then amplified in a streptavidin-biotinylated alkaline phospha- tase-catalyzed reaction by BCIP and NBT. As seen in Figure 4a, LEO A-1 reacted strongly with a human protein in a broad band from approximately 56 to 62 kDa under non-reducing conditions. Leo A-1 also reacted with non-reduced platelet proteins in normal and affected dogs. In this case, 2 fainter but crisp bands could be appreciated at approxi- mately 67 and 70 kDa, respectively. Great difficulty was encountered in approximating molecular weights for the bands detected on the Western blots. This resulted from the detection of additional bands in the molecular weight standard lane not corresponding to known compounds. Unfortunately, this 38 39 situation was encountered with several batches of biotinylated standards, and was present in all blots performed. As seen in Figure 4b, reduced blots showed no interaction of LEO A-I with platelet proteins in man or dog. However, faint doublet bands could be discerned in the 70-80 kDa area in man and dog that raised concern that the doublet bands seen in the non-reduced blots could represent non—specific binding. A non-reduced blot containing a negative and positive control for the primary and secondary antibodies showed a very faint singlet band in the 70 kDa area. Because of these findings, 3 direct binding study using 12°l-labelled LEO A-1 was performed to determine if the bands seen on the Western blots did indeed represent specific binding of the antibody. Figure 4a. 40 Western blot of non-reduced platelet proteins incubated with Leo A-1 monoclonal antibody. Lane A: BHT-affected dog, Lane 8: normal dog, Lane C: human, Lane D: molecular weight standard, Lane 1: molecular weight standard, Lane 2: human platelet proteins. Lanes 1 and 2 are stained with an India ink total protein stain to evaluate efficacy of protein transfer to the nitrocellulose membrane. 1H WV, 00' “0| o=| OON l we. 17 42 Figure 4b. Western blot of reduced platelet proteins incubated with Leo A-1 monoclonal antibody. Lane A: BHT-affected dog, Lane 8: normal dog, Lane C: human, Lane D: molecular weight standard. 43 44 Direct binding studies In an initial experiment to evaluate binding of LEO A-1 to the platelet surface, 12°l-Iabelled LEO A-1 was added to a suspension of washed platelets from a human, normal dog, and a BHT-affected dog in a final concentration of 250 ng/ml. The adjusted platelet count of the suspension was 2 x 10° platelets/ml. 125l HB 43 (a monoclonal antibody to human lgG) was used as a negative control for canine platelets, and 125I IV 3 (a monoclonal antibody to human Fc receptors) was used as a positive control for low copy number in the human sample. Both of the latter monoclonal antibodies were also used at a final concentration of 250 ng/ml. The platelet suspensions were incubated in an Eppendorf tube with 125I-Iabelled antibody for 1 hour at room temperature. Following the incubation, the bound antibody was separated from the free by centrifugation of the platelets through a 20% sucrose cushion. The tip of the Eppendorf tube containing the platelet pellet was cut off and its radioactivity measured in a gamma counter. This was taken to represent the amount of radioactivity bound to the surface of the platelet. An equivalent solution of 250 ng/ml of 125l-labelled LEO A-1, HB 43, and IV 3 was also counted to represent the total radioactivity originally added to the tube. The percent binding of each antibody was then calculated by dividing the amount of radioactivity (cpm) of the platelet pellet by the total radioactivity (cpm) for each antibody for canine and human. The data for this initial study are presented in Table 1 and Figure 5. As can be seen, 125I Leo A-1 binds to canine platelets, but to a much lesser degree than to human platelets. This could be the result 45 of lower affinity of the human monoclonal antibody for the canine protein and/or a lower receptor copy number on the surface of canine platelets compared to human platelets. Leo A-I binding to human platelets approxi- mates the binding of the positive control, IV 3, whose receptor is present in a comparable copy number to PTA-1. Table 1. Direct binding of 250 ng/ml 12°l Leo A-1 to the platelet surface on man and dog. Subject Human Human Normal Dog Normal Dog BHT BHT LEO A- IV3 LEO A-I HB 43 LEO A-I HB 43 cpm bound 321205 192857 80084 279 79295 274 total cgm 602402 438434 602402 137973 602402 137973 % bound 53 44 13 0.1 13 0.2 46 Figure 5. Direct binding of 250 ng/ml 125l Leo A-1, HB 43, or IV 3 to the platelet surface in man and dog. A and 8: human plate- lets, C and D: control canine platelets, E and F: BHT-affected platelets. mva 47 Scam... ITI mva F