.1. s . \c : .5. . . I g»... . . . . .. . .. numfifi. EK§¥N£EHNAJ . .. «a. . . . 1.5%. f5... :l 6.. 5’53”}; - ' 3‘: ; l» 1 . ll . 1...... m. z ”49 .1 i ., . . . .. .9? .U. 443...;3%¢ : . . . .44.“? . i I a .\ 1 , .. u-N. V .. . . W ....... , 3...? e n . -- {$14. ...4,................ ... .. . 5‘ 7.“. 1 NJ? . . - .....h..z. .. .. :. ..;.. . s... 4., . . .... @mev..mw.fi$$mh4.mflfim.a fifiawwwgmfiw! . . . . . . . .. r a umfiflfi .... x. .. . Xx . .. . . . . . t“: . .. . .c. rmAMribr ..1W.\.P.§ 4...... u. «a... Q... .. t... , : . . _ ... , . .,..!L.u.2r..rm.. l. 2 4...... Ian Jihummniuw raw aemflumw .. . . . .. .. .. . . . 1&4"? ..n “NJ-vi... "4.4mm. .. 3.4” am... . . m. . .. . . .. z . . . afinwfimuuaw. . . .... . . ... . . .. . Luv .. . t Lawn”. .7 h . . .. . .H . .. .. 1...». .. M....m..n..u.u..w.o...m~ .. .. an. . . . . ., . , .. . ....~..~..m.i:.i it...“ 7...- .32 c . .. . 2... AH . z . ‘u' )l 1... .. .I aw......my...sn... .. u in: .ru‘.‘v-r . as a. .344 a . . E 4 . . fink... . 3...... .23“... .25....) ii..- . . a . .. L! in II . 0.2. 1.1“ . . . 4...... W #338. . - l .. .w! ........I. . 11...”. . 1.... . h. .5! w... 1 . .. 05.. y I . .Afllv‘hlr. .091 .. . , . . . 1.54%....» 19min...“ .. Eur! .. . . . .u. 34 . .9 . .wt ..J....... . . ...........\r:m.. 1.2!... , ...l....-.. . $31.14-! . 5-4.4.3542... . .:. . .. .. .t. (A. \ httl f . 1 «I! . 53:315.. :45 \‘I a 1 o .v i lt’uof' 26.14...va .... tannin; K .3 f.l.. 1-; LEAN-v. .8 gift} .I I 3|... u. .glilp. b}! .s. . .a: .u 1" .4 nA|o"l .34.... .. "input V l ‘4 .1 4‘73“}: l l... . .. .Juh‘wa $9.034: .‘ .1. 121:2. E rm 5. '. . i '1'. I . n '9 I..A ,...I‘1‘|‘ ‘ 'v . ..ubl 51- ".14.. 0.1:... m7 Illi'llli'lllllllllllllllllll'llllllfllll 3 1293 01592 84 LIBRARY Michigan State University This is to certify that the thesis entitled IDENTIFICATION OF THE MOLECULAR DEFECT OF GLYCOPROTEIN Ib ALPHA IN A PATIENT WITH BERNARD-SOULIER SYNDROME presented by Aida A. Abujoub has been accepted towards fulfillment of the requirements for law degree in ClinicaLLaboratory Sciences Maid," Major profegdé- j Date May 2. 4297 0.7639 MS U is an Affirmative Action/Equal Opportunity Institution ___ ____. _L. — ..____ _ _. PLACE It RETURN BOXtomnovothb chockomtrom yourneord. TO AVOID FINES Mun on or bdoro date due. DATE DUE DATE DUE DATE DUE MSU I. An Affirmative Action/Equal Opportunlty lnotltwon Wanna-9.1 IDE? IDENTIFICATION OF THE MOLECULAR DEFECT OF GLYCOPROTEIN Ibo: IN A PATIENT WITH BERNARD-SOULIER SYNDROME BY Aida A. Abujoub A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Medical Technology Program 1997 lDl ml by CO CH dis he ABSTRACT IDENTIFICATION OF THE MOLECULAR DEFECT OF GLYCOPROTEIN Ibo: IN A PATIENT WITH BERNARD-SOULIER SYNDROME BY Aida A. Abujoub Bernard Soulier Syndrome (BSS) is a rare congenital bleeding disorder. The glycoprotein (GP) Ib.IX-V complex, the receptor for von Willebrand factor, was found to be absent or dysfunctional in patients with 888. At the molecular level a defect in any of the genes that encode GPIba, GPIbB, GPIX, and possibly GPV could prevent assembly and cell surface expression of the entire complex. In this research a family with B88 is being investigated to determine the molecular basis of GPIba absence. The propositus’ platelets lack GPlba as documented by two dimensional gel electrophoresis. Polymerase chain reaction (PCR) did not reveal any gross rearrangement (insertion or deletion) within GPlba gene. Single stranded conformation polymorphism analysis indicated that a mutation could be present at the 3’ end of GPIba gene. Sequencing analysis identified a point mutation that changed amino acid proline 481 to a serine a change which could play a role in the development of the disease. PCR, restriction digest and sequencing analyses showed that the mother is heterozygote for a 39 bp repeat, whereas the propositus is homozygote for this repeat. To my mother and father Adla and Abdel-Hamid Samara, who instilled in me that knowledge is the greatest achievement. To my wonderful husband Amin for all his support, love and help. To my precious children, Rawan and Shadi. This is for you all iii ACKNOWLEDGMENTS I would like to thank my advisor Dr. Douglas Estry for his time, guidance and financial support. Also I would to thank Dr. Joan Mattson for providing us with the blood samples and clinical data. A special thanks to Dr. John Gerlach and Dr. Paul Coussens for their guidance, suggestions, comments and for giving me the opportunity to use their labs to conduct my experiments. My deepest appreciation goes to my husband Dr. Amin Abujoub for his help, support and encouragement. iv L810 L810 Inuod Mate TABLE OF CONTENTS List of Tables ......................................................... vii List of Figures ........................................................ viii Introduction ............................................................ 1 Overview of hemostasis ............................................. 1 Bernard-Soulier Syndrome .................... ~. ...................... 5 Normal Biochemical and Molecular Characteristics of GPIb-IX-V ........... 6 GPIb-IX complex and va. ......................................... 11 Structural and Functional Defects of platelets in BSS ..................... l4 Platelet structure abnormality in 888 ........................... 14 Platelet function abnormality .................................. 14 Platelet membrane glycoprotein abnormalities .................... 15 Molecular Defects Associated with BSS ............................... 16 Polymorphism in GPIb-IX .......................................... 23 Family History ................................................... 30 Objectives ...................................................... 30 Materials and Methods ................................................... 33 Sample collection and DNA isolation ................................. 33 DNA amplification ................................................ 35 ReSL Di Fl Agarose gel electrophoresis ......................................... 37 PCR products purification .......................................... 37 Heteroduplexing analysis ........................................... 38 Single Stranded Conformation Polymorphism (SSCP) .................... 39 Cloning ......................................................... 40 Preparation of large scale DNA ..................................... 46 Sequencing ...................................................... 49 Sequencing analysis ............................................... 50 Results ............................................................... 51 Polymerase Chain Reaction ......................................... 51 Hetreoduplexing .................................................. 53 Single Stranded Conformation Polymorphism (SSCP) .................... 54 Cloning ......................................................... 59 Sequencing ...................................................... 62 Discussion ............................................................ 74 Localization of mutation(s) by heteroduplexing and Single Stranded Conformation Polymorphism ........................................ 76 Sequencing and Sequence Analysis ................................... 79 Future Research ........................................................ 82 List of references ....................................................... 84 vi Tabl: IBM: Tab! Tab Tar LIST OF TABLES Table 1. Mutations found in BSS. .......................................... 24 Table 2. Gene frequency of molecular weight polymorphism in Japanese and American population .............................................. 27 Table 3. Tandem repeat polymorphism of individuals with ATG 145. .............. 28 Table 4. Frequency distribution of GPIb-a alleles .............................. 29 Table 5. Sequences of primers used for amplification and sequencing. ............. 35 vii Fig Fit Fig Fig Fig Fi; Fi Fi Fi Fi LIST OF FIGURES Figure 1. Assembly of GPIb-IX complex. ..................................... 8 Figure 2. Polymorphism of GPIb-IX complex. ............................... 25 Figure 3. Pedigree of the studied family. ..................................... 32 Figure 4. Cloning vector pUCl9 ............................................ 41 Figure 5. Purification of the closed circular plasmid DNA by ultracentrifugation using CsCl-ethedium bromide gradient centrifugation method. ............ 48 Figure 6. PCR amplification of the first and the second fragments. ................ 52 Figure 7. PCR amplification of the third fragment. ............................. 55 Figure 8. Heteroduplexing analysis gel ....................................... 56 Figure 9. Restriction digest of the first and the second PCR fragments. ............. 57 Figure 10. Restriction digest of the third PCR fragment. ........................ 58 Figure 11. SSCP of the first and second fragments. ............................ 60 Figure 12. SSCP analysis of the third fragment. ............................... 61 Figure 13. Comparison of the third fragment nucleic acid sequences. .............. 64 Figure 14. Comparison of the third fragment peptide sequences. .................. 71 viii INTRODUCTION Overview of hemostasis: Hemostasis is the process by which the body prevents the loss of blood from it’s vascular system. This process involves platelets, blood vessels, coagulation factors, fibrinolytic proteins and inhibitors that maintain a state of fluidity in the normal circulation as well as help form a physical barrier that limits the loss of blood from sites of vascular injury (Ratnof et al., 1984). The balance of hemostasis must be maintained to prevent bleeding or pathologic thrombosis. Physical and biochemical properties of blood vessels help to maintain normal circulation. The physical properties include vasoconstriction, vasodilation and changes in vessel permeability. They separate plasma factors from the highly reactive elements present in the deeper layers of the vessel wall. These elements include adhesive proteins such as: a) collagen and von Willebrand factor (va) which promote platelet adhesion to the injured vessel wall (Hession et al.,1990). b) tissue factor, a membrane protein located in fibroblasts and macrophages that triggers blood coagulation (Hessian et al.,1990). Biochemical properties of the blood vessel include inhibition of blood coagulation by synthesizing and releasing heparan sulfate, which accelerates the activity of antithrombin ill. 2 disc cytn of1 SCt' gr: SU 2 III, a plasma anticoagulant and the vascular endothelium synthesis of prostacyclin (PGIZ) and nitric oxide which cause vasodilation (Weiss et al., 1979) Platelets play a fundamental role in this hemostatic process. Platelets are 2 4pm, disc-shaped cells formed in the bone marrow by fragmentation of the megakaryocyte cytoplasm. Normal whole blood contains 150,000-400,000 platelets /uL and the life span of the average platelet is 95-105 days. The platelet has a variety of functions. These include adhesion, aggregation, secretion and concentration of plasma clotting factors on both the membrane and within granules. Platelets adhere to an injured vessel and then spread on the exposed subendothelial surface. Adhesion at high shear rates depends on the presence of va and the platelets surface receptor complex GPIb-IX-V. Platelets do not adhere to normal endothelial cells but, in the case of injury, the adhesive proteins in the subendothelial matrix of blood vessels facilitate platelet adhesion to subendothelial collagen. Von Willebrand factor is made up of multiple functional domains, including domains involved in binding to GPIb and to the subendothelium of blood vessels (Clemetson et al., 1982). Thus va present in the subendothelium may directly mediate platelet adhesion and the plasma and/or platelet form of va may bridge platelets to other subendothelial components such as collagen and microfibrils. It is postulated that the association of va with matrix constituents may alter its properties and permit its direct binding to GPIb (Handin et al., 1989). Secretion is the release of a wide variety of substances that can stimulate or inhibit platelets, contribute to cell adhesive events and modulate growth of cells of the vessel wall. Most of these substances are either secreted from preformed storage granules (alpha and c bodi: plate brou and diff l 98 V\\' suc pla pie in- fo 3 and dense granules) or synthesized de novo from membrane phospholoipids. The dense bodies release ADP (Greenburg et al., 1988), a weak agonist, which recruits other platelets to the forming aggregate. ADP induces secretion only when platelets are brought into close contact during aggregation (Charo et al., 1977). Thromboxane A2 (TXAZ) is generated as a consequence of hydrolysis of membrane phospholipids (Kroll et al., 1989). It is a highly potent aggregation stimulating substance. TXA2 is produced by the cyclooxygenase pathway from arachidonate. When added to platelets in vitro, thromboxane causes shape change, aggregation, secretion, phosphoinositide hydrolysis and an increase in the cytosolic calcium (Gerrad et al., 1981). Once formed, TXA2 diffuses across the platelet plasma membrane and activates other platelets (Brass et al., 1987). The alpha (or) granules of platelets contain adhesive proteins such as fibrinogen, va, fibronectin, vitronectin, thrombospondin. They also contain growth modulators such as platelet factor 4 and platelet-derived growth factor (PDGF) (Snyder 1989), which play an important role in smooth muscle cell proliferation that may occur consequent to platelet interaction with vessel walls. Many coagulation proteins are present in a granules and are released by platelets. The alpha granules contain factor V which is involved in the generation of thrombin (Wencel-Drake et al., 1986). Platelet aggregation is the process by which platelets interact with one another to form a hemostatic plug. Both fibrinogen and GPIIb-IIIa are required for aggregation to occur (Peerschke 1985, Phillips et a1. 1988). GPIIb-IIIa complex belongs to the integrin family of adhesion receptors and functions as the platelet receptor for fibrinogen, fibronectin and va (Phillips et al., 1988). GPIIb-IIIa is the most abundant glycoprotein on platelet surface (50,000 - 80,000 copies / platelet). Patients with Glanzman’s The: fiulu aaha urak. flbnnc 1987) recept mem'r fibnnr sugge comp 81... 19 4 Thrombasthenia have defective or abnormal amounts of GPIIb-IIIa, thus their platelets fail to bind fibrinogen and fail to aggregate (Nurden et al., 1974). Soluble fibrinogen does not interact with resting platelets, but binds to platelets activated by various agonists (strong agonists such as thrombin and thromboxane A2, weak agonists such as ADP). Thrombin and thromboxane A2 induce exposure of fibrinogen receptors by activation of phospholipase C and protein kinase C (Shattil et al., 1987). The intracellular molecular mechanism of the exposure of platelet fibrinogen receptor is not well understood but platelet activation might induce a change in the membrane micro environment around GPIIb-IIIa, allowing access of ligands, such as fibrinogen, to their binding site (Coller 1986). Another more recent theory theory suggests that platelet activation causes a conformational change within the GPIIb-IIIa complex thereby inducing formation or exposure of a ligand-binding pocket (Golden et al., 1990). The importance of platelet membrane receptors is illustrated by the occurrence of hemorrhage in different diseases such as Bernard-Soulier Syndrome (BSS) where the adhesion process is impaired due to the fact that the patient’s platelets lack the GPIb-IX- V complex (Nurden et al., 1983), in von Willebrand disease in which va is decreased or defective resulting in decreased platelet adhesion (Weiss et al., 1978), or Glanzman’s thrombasthenia where GPIIb-IIIa is absent or defective therefore preventing aggregation (Nurden et al., 1974). In addition the platelet plasma membrane plays an important hemostatic role in cell-cell interaction and in concentrating coagulation factors thus assisting in the formation of a fibrin clot. Although, the various components of the enzyme-substrate serve hiad panu Iierr infer plate next esui S§r1 recs bot] eta ncn CO] lit 5 complexes that constitute the coagulation pathway can interact in the fluid phase, the rate of reaction increases by at least 1000 fold in association with a phospholipid surface (Nurden et al., 1975). The platelet plasma membrane acts as a phospholipid surface and serves to enrich the local concentrations of reactants such as factor X, V and prothrombin. In addition the platelet (Jr-granule contains some of the factor V, which when secreted participates in this process (Hoffman et al., 1991). Bernard-Soulier Syndrome Bernard-Soulier Syndrome (BSS) was first described in 1948 with the report of an infant who had severe mucocutaneous bleeding, a prolonged bleeding time with a normal platelet count and abnormally large platelets (Bernard J. and Soulier J. ,1948). Over the next ten years additional patients were reported. In 1964 a review of 14 patients established the diagnostic criteria for the disease (Alagille et al., 1964). Bernard-Soulier Syndrome is a congenital bleeding disorder that is usually inherited in an autosomal recessive manner. Consanguinty is often noticed, and the disease is equally common in both sexes (Bernard 1983). Bernard-Soulier syndrome is characterized by giant platelets with an absent or dysfunctional platelet membrane receptor GPIb-IX-V complex ( Nurden et al.,1981; Ware et al.,1993), prolonged bleeding time, thrombocytopenia in most patients, large platelets, normal coagulation time and normal clot retraction. Platelet counts in these patients commonly range from 50,000/uL to near normal (normal value 150,000 - 400,000/uL). However platelet counts as low as 20,000/ul have been reported (Alagille et al.,1964). The illness is both rare and ubiquitous, it has been observed in Europe (Italy, 6 Britain, France, Poland, Austria, Norway and Greece), in Asia (India, Japan, Kuwait and Turkey), in America (United States and Mexico), and in Africa (South Africa and Reunion) (Bernard 1983). The signs of the disease are usually noticed during the first months of life, but a few exceptions in which onset of symptoms occurs later have been reported (Bernard 1983). Hemorrhage, the only clinical sign of the disease, is varied. Epistaxes, ecchymoses and purpura are frequent. Menorrhagia and digestive hemorrhages are often observed. Interestingly hematuria and hemorrhages due to circumcission and retinal hemorrhages are rare (Bernard J. 1983). Normal Biochemical and Molecular Characteristics of GPIh-IX-V: The GPIb-IX-V complex is a major sialoglycoprotein on the platelet surface and contributes to the net negative charge of the platelet surface. GPlb is a high molecular weight GP composed of two subunits. Cleavage of intermolecular disulfide bonds results in the separation of a small 22 kDa, B-subunit from a large, 148 kDa, a-subunit (Modderman et al.,1992). The outer N-terminal 45 kDa region of GPIba forms a compact globular domain containing binding sites for both va and thrombin (Wicki et al., 1985). Approximately 25,000 copies of the GPIb-IX complex and half as many of GPV are present on the platelet surface. Each of the three polypeptides «,0 and IX span the platelet plasma membrane once, with the carboxyl terminus in the cytoplasm and the amino terminus toward the exterior (Lopez et a1. 1987, Lopez et a1. 1988, Hickey et a1. 1989). The three polypeptides and GPV belong to the leucine rich family of glycoproteins and have a structural domain defined by one or more tandem repeats of a conserved leucine-rich sequence and conserved sequences flanking the repeats. GPlba .7 has 7 leucine rich repeats (LRR). GPIbB and GPIX each have one LRR, whereas GPV has 15 LR (Lopez 1994; Sae-Tung et.al,1996). The GPIb-IX complex consists of two globular domains with an intrvening rod- like region (Figure 1, adapted from Lopez 1994). The smaller globular domain represents the amino-terminal ligand-binding part of GPIba, and the larger globular domain consists of the GPIba transmembrane and cytoplasmic domains associated with GPIbB and GPIX. The rod-like region that connects the two globular domains consists primarily of the macroglycopeptide part of GPIboc. The structure of GPIba can be divided into four distinct domains: 1) A globular domain at the amino terminus that contains the seven leucine rich repeats (LRR), consists of approximately 300 amino acids and has two N-glycosylation sites and few, if any, O-linked carbohydrate chains. This domain accounts for most, if not all, of the va binding to the complex and also contains a binding site for thrombin. This region also contains all of the intrachain disulphide bonds in GPIbct, having seven of its nine cysteines. The cysteines form three disulphide bonds, leaving one free thiol. A disulphide bond links Cys-4 at the amino-terminal and Cys-17 just prior to the first LRR (Fox et al., 1988). Two more disulphide bonds follow the seven LR and are formed by a pairing between Cys-209 with Cys-248 and Cys-211 with Cys-264. Because of the proximity of Cys 209 and Cys-211, the structure that results from this disulfide pairing consists of two loops, these loops are critical for the ability of the complex to bind va. 2) The macroglycopeptide is an 84KD peptide that separates the amino-terminal globular domain from the plasma membrane. The sequence of this region is dominated by threonine, serine and proline. Most of the threonine and serine residues are — M wwi Figure 1. Assembly of GPIb-IX complex. This schematic diagram shows the different structure motifs of GPIb-alpha and its covalent association with GPIb-beta. (Adapted from Lopez 1994). 9 glycosylated. Glycosylation accounts for approximately half of the molecular weight of the GPIba polypeptide determined by SDS-polyacrylamide gel electrophoresis. Most of the carbohydrate chains in this region are modified by sialic acid which makes GPIba the major contributor to the sialic acid concentration on the platelet surface (Judson et a1. 1982, Tsuji et al. 1983, Korrel et a1. 1984). This concentration was estimated to be eleven times that of the erythrocyte (Maddoff et al., 1964), giving the platelet surface a very high negative charge which generates electrostatic repulsion that keeps platelets apart as they circulate in the blood stream. The extensive glycosylation and high proline content of this region prevent the formation of extended secondary structure (Shogren et al., 1989). The structure of the GPIba resembles a palm tree, with the ligand binding domain representing the leafy top and the macroglycopeptide representing its rigid trunk. The elongated nature of this receptor is possibly one of the structural features that underlie its unique response to shear stress. Following the macroglycopeptide and preceding the hydrophobic transmembrane domain is a sequence of approximately 30 amino acids that has very few sites for O- glycosylation. This region contains the protease sensitive site. Cleavage at this site releases a water-soluble extra cytoplasmic domain of GPlba known as glycocalicin. 3) The hydrophobic transmembrane segment consists of 29 amino acids flanked with charged amino acids which may serve to anchor the protein in the membrane (Lopez et al., 1987). 4) The cytoplasmic domain is approximately 100 amino acids. This domain connects the GPIb to the cortical cytoskeleton of the platelet. The interaction of GPIb-IX complex with the cytoskeleton plays a role in maintaining the distribution of the complex on the platelet surface and may be necessary for the proper spatial relations of the 10 complex with other proteins on the plain of the plasma membrane (Fox et al., 1989). The structure of GPIbB and GPIX shows approximately 60% sequence similarity (identical or conserved residues) in their extra cytoplasmic domains but diverge completely in their cytoplasmic domains. The major differences between the overall structure of these smaller subunits and that of GPIba are the presence of the macroglycopeptide domain in GPIba, the number of LRR, and a much longer cytoplasmic domain in GPIba. The cytoplasmic domain of GPIbB is approximately 34 amino acids and that of GPIX ranges between 6-8 amino acids. The disulphide pairings in GPIbB and GPIX, shown in figure 1 above, are drawn from homology with GPIba and with other members of the leucine-rich family (N came et al., 1989; Lopez 1994). Each of the leucine rich glycoproteins in the complex GPIb-IX-V is encoded by its own gene. The a chain has been cloned, completely sequenced (Lopez et al., 1987;Wenger et al.,1988), and localized to chromosome 17 (Wenger et al., 1989). Also, the cDNA for GPIX has been cloned, sequenced (Hickey et al., 1989; Miller et al., 1992), and localized to chromosome 3 (Hickey et al., 1993). The gene encoding GPV has been cloned (Lanza et al., 1993), but its chromosomal location has not been determined (Lopez, 1994). GPIbB has been localized to chromosome 22 and it was cloned and sequenced (Lopez et al., 1988 b). GPIboc and GPV each have one intron , where as GPIX has two introns, all of which are in the 5' untranslated regions of the gene immediately upstream of the ATG initiation codon, which leaves the coding sequence uninterrupted (Lopez, 1994). l1 GPIb-IX complex and va Platelet adhesion to va in the subendothelium is the major function of the GPIb— IX complex, evidence for this function is as follows: 1) 2) 3) 4) In the congenital absence of GPIb, as in 888, platelets fail to bind to va when stimulated with Ristocetin. Ristocetin is a peptide antibiotic that is used to diagnose BSS. Some data indicates that it binds to the receptor whereas other data shows that it binds to the ligand. (Coller et a1. 1977, Coller et a1. 1978, Jenkins et al. 1979) Selective proteolytic digestion of GPIb decrease ristocetin-induced va binding. Monolonal antibodies against GPIb selectively inhibit ristocetin-induced va dependent platelet agglutination (Oscar et al.,1984). Inhibition of platelet binding to immobilized va in the presence of monoclonal antibodies against the GPIb-IX complex (Weiss et al., 1986). Although va circulates freely in plasma, under normal conditions it does not associate with GPIb-IX. The binding appears to require a change in either the ligand (th) or the receptor (GPIb-IX), induced either by shear or by immobilization of va on the subendothelial matrix. For platelet aggregation to occur, shear stresses must reach levels high enough to allow the GPIb-IX/va interaction to occur in the fluid phase, with a subsequent events resulting in the formation of a platelet aggregate. Shear stresses high enough to induce the fluid-phase interaction between GPIb-IX complex and va are not found in normal physiologic situations (Maoke et al., 1988). It appears that either one or both va and GPIb-IX molecules are acting as mechanoreceptors capable of sensing shear (th) and adapting a conformation (GPIb—IX) favorable for the interaction to occur 1 2 (Peterson et al., 1987). Many studies have been done to identify the site on GPIb-IX complex to which va binds. Synthetic peptides that block certain areas of GPIba were used for this purpose (V icenti et al., 1990). These studies along with others that involved recombinant GPIba polypeptide (Cruz et al., 1992) suggest that the va binding domain comprises more than simple linear sequences. The studies together with naturally occurring mutations provided evidence that the two loops formed by two disulphide bonds in the carboxyl terminus of this region are essential for va binding to GPIb-IX complex. The first loop appears to regulate the binding of va to the second loop, possibly through steric effects. Another possibility is that discontinuous stretches of sequences from both loops combine to form the binding site for va (Lopez, 1994). The GPIb-IX complex plays a role in stabilizing the plasma membrane and thus helps maintain the discoid shape of platelets (Fox et al., 1988). This occurs by interaction between the cytoplasmic portion of GPIb-IX and the submembraneous cytoskeleton. Platelets from individuals with BSS lack this membrane-cytoskeleton interaction which may be the cause of large and irregularly-shaped cells (Okita et al., 1985). Membrane fluidity studies have shown that BSS platelets are much more deformable than normal platelets (McGill et al., 1984) and form surface filopods three times longer than normal ones when aspirated into micropipettes (White et al., 1984). This could be due to the absence of interaction between GPIb-IX and actin binding protein (ABP). This loss of membrane stability may predispose platelets to early destruction when exposed to the rigours of traversing the circulation which could be the reason for low platelet counts in these patients (Hartwig et al., 1991). Anchorage of the GPIb-IX complex to the skeletal 13 framework of the cell may also affect the ability of the complex to mediate platelet adhesion. This could happen in two ways, first attachment of the complex to the submembraneous region may result in an orderly spatial distribution of the receptor on the platelet surface. This arrangement could be important for the Optimal interaction of the complex with its ligand va, which binds to platelet best when it is in the very large multimer form (Federici et al., 1989). Second, the attachment of complex to cytoskeleton may regulate the affinity with which va binds to the complex, independent of any distribution effect (Coller et al, 1982). Glycoprotein V is another platelet membrane glycoprotein that is rich in leucine, has large amounts of carbohydrate and provides antigenic determinants on the platelet surface for antiplatelet antibodies (Gerald et al., 1993). GPV is an 82 KDa glycoprotein which was first identified because it is a substrate for thrombin on intact platelets. Thrombin cleavage causes liberation of a large (MW 69,000) water soluble fragment of GPV (Modderman et al.,1992). Therefore, GPV is a logical candidate to be a thrombin receptor. However, experiments have shown that even platelets that lack GPV (on a congenital, proteolytic or immunologic basis) respond to thrombin in a nearly normal fashion. This implies that GPV may influence the effect of thrombin on platelets, but is not likely to be an actual thrombin receptor. Glycoprotein V is not involved directly in platelet adhesion and has non-covalent association with the GPIb-IX complex but present in close proximity to it (Gerald et al.,1990). Modderman et a1. 1 992, found that GPV and GPIb-IX form a noncovalent complex in the platelet membrane. It is unknown why in BSS GPV is deficient along with GPIb-IX, but perhaps all three proteins must participate in some or all steps (synthesis, 14 processing, and assembly) that lead to the expression of the mature complex on the platelet surface ( Gerald et al.,1990). Structural and Functional Defects of Platelets in BSS: Platelet structure abnormality in BSS: Large platelets are the most prominent morphological feature of BSS. Mattson et a1 1984, studied the size of platelets in two homozygote BSS siblings and their heterozygote parents. Platelets from the two affected siblings had a mean diameter of 3.13 um and 4.14 um as measured on Wright Giemsa stained smears, whereas normal controls had platelet diameters ranged from 0.83 - 2.33 um. In the same study , diarnetrs of spread BSS platelets were larger than fully spread controls as evidenced by both scanning electron microscopy and transmission electron microscopy. Platelets appear to be more spherical and more deformable. As previously indicated the increased deformability may be related to the lack of transmembrane linkages to the cytoskeleton. Platelet survival was markedly shortened in several studies (Najean et a]. 1963, Grottum et a1. 1969, Cullurn et al. 1967). Platelet function abnormality: In 1972, it was noted that although BSS platelets aggregated normally in response to adenosine diphosphate (ADP), epinephrine, and collagen, platelets failed to aggregate when stimulated with va and ristocetin (Clemetson et al.,1982). Ristocetin induces va binding to normal platelets and this is absent or markedly decreased in BSS. In vitro normal platelets adhere to immobilized va in the absence of ristocetin whereas 15 BSS platelets do not. This may be analogous to platelet adhesion to subendothelium bound va in vivo. Von Willebrand factor binding stimulated by ADP or thrombin, which occurs on the GPIIb-Illa complex rather than GPIb, is normal in BSS platelets, this is consistent with other evidence that the GPIIb-Illa receptor functions normally (Peerschke et al., 1980). The response of BSS platelets to (Jr-thrombin is decreased, demonstrated by decreased thrombin binding and decreased aggregation. It has been shown that GPIb contains one or more high affinity thrombin binding sites and the diminished response of BSS platelets may be related to the absence of these sites. Platelet membrane glycoprotein abnormalities: A plasma membrane abnormality of BSS platelets was first suggested in 1969 by observing a decrease in electrophoretic mobility that was associated with a decrease in sialic acid ( Grottum et al., 1969). In 1975, Nurden and Caen demonstrated a decreased amount of the platelet membrane GPIb. These results were confirmed by single and two dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) of 1251 labeled platelets. These early observations and other studies were important in establishing that GPIb was a platelet receptor for va (George et al., 1984). Further studies using H3 borohidride to radiolabel platelet surface carbohydrate residues demonstrated that, in addition to GPIb, BSS platelets were deficient in GPIX and GPV (Nurden et al., 1983). GPIb and GPIX form a 1:1 noncovalent complex (Du et al., 1979). This complex is linked to the membrane skeleton by an actin binding protein (Okita et al., 1985). Although GPV is a substrate for thrombin it has no known physiological role. 16 The deficiency of GPV in the BSS platelets may indicate an association with the GPIb- IX complex (Colman et al.,1994; Moddennan et al.,1992). Molecular Defects Associated with BSS: The classic form of BSS is characterized by the absence of the GPIb-IX-V complex. However, other variants have been described that show a wide variety of molecular differences. In some cases, relatively normal levels of all four chains (GPIba, GPIbD, GPIX and GPV) have been detected with the defect eventually being localized to a point mutation in the outer domain of GPIba (Miller et al., 1992). In other cases , the amounts of all the chains are reduced in parallel suggesting that there is a coordinated expression, or the expression level of one of the chains regulates the amount of the mature complex (Clemetson et al., 1993). A few rare cases have demonstrated an imbalance between the chains expressed ' where GPIb was totally absent but low amounts of GPIX have been found (Hourdille et al., 1990). In some cases, at low levels of all the chains, there was clearly less GPIX than the other three (Clemetson et al., 1993). There are rare cases that seem to be due to a double heterozygote defect in GPIba chain (Ware et al., 1990). Other cases have been produced by a double heterozygote point mutation in the gene encoding GPIX (Wright et.al, 1993). A defect in one of the four genes that encodes GPIb, IX or GPV could alter or prevent the assembly and cell surface expression of the entire va receptor complex (Finch et al., 1990). A defect in the promoter for one of these genes could lead to a complete lack of expression without there being any problem in the translated, structural region (Clemetson et al., 1993). It was suggested (Lopez et al., 1992) that GPV is not l7 necessary for the surface expression of the other chains but that all three chains of GPIb- IX are necessary for an efficient surface expression of the complex. The underlying molecular basis of BSS is currently unclear and may in fact be heterogeneous. As mentioned before, in BSS, deficiencies of membrane GPIb have been associated with comparable deficiencies in GPIX and GPV (N urden et al., 1993). At the molecular level, different studies support the heterogeneous nature of underlying defects that may lead to an abnormal va receptor. Ware et a1. 1990, studied a case in which the GPIb-IX-V complex was absent from the platelet surface, but a truncated fragment could be detected on western blots. Sequencing of GPIba showed a mutation of TGG to TGA that changed tryptophan 343 to a stop codon. This resulted in a truncated protein missing the membrane-spanning section of GPIba. Hence it could not be anchored to the membrane and associate with other chains. It may be that the truncated fragment is present in the cytoplasm and is then degraded or secreted. This case is further complicated by the fact that it appears to be a double heterozygote. The Polymerase Chain Reaction (PCR) amplification of the region of the gene containing the mutation gave two sequences, one allele containing the mutation and the other allele being normal. It is proposed that the second allele must contain a defect causing a complete absence of expression of GPIba, and when combined with the first mutated allele, only the truncated fragment is expressed. The defect in the second allele has still not been demonstrated. Marco et a1. 1990, had identified another BSS variant, termed BSS Bolzano, which leads to the typical symptoms of BSS. In this case, GPIba appears to be present but in decreased levels. The difference between this variant and the classic form of BSS 18 lies in the presence of a smaller (105 KD) piece of GPIba detected on western blots. Binding of thrombin was normal implying that only va binding was affected. This could be due to a mutation in the N-terrninal region of GPIba chain that affects the va binding site and most likely increases susceptibility to proteolytic degradation. This might occur as a result of the mutation providing a new cleavage site for a protease or by changing the conformation of a domain of the protein, making an already existing cleavage site more accessible. In 1992 Miller et al. described an autosomal dominant variant of the disease. A point mutation in GPIba, where a C at position 259 had been substituted by a T, converted a highly conserved leucine at position 57 to phenylalanine within the first leucine rich repeat in one of the alleles. In this case an abnormal rather than absent GPIba was detected which resulted in abnormal va binding and the typical clinical features of BSS. This mutation could also increase the degradation rate of GPIba. The abnormal GPIba didn’t affect the assembly of GPIb-IX complex on the platelet surface. This case indicates the importance of the integrated LRR.. Salle et a1. 1995, reported a case where a three base deletion resulted in the removal of leucine 179, which is located in a highly conserved position of the seventh leucine-rich repeat of GPIba. Three affected siblings were characterized by absence of ristocetin-induced platelet agglutination, although ADP aggregation was normal. Flow cytometry studies showed detectable amounts of all members of GPIb-IX-V complex on the surface of the patient’s platelets, whereas western blot analysis revealed normal levels of GPIX, decreased levels of GPIbB and GPV, and less than 1% of GPIba. Sequencing of GPIboc revealed the absence of leucine 179 which is believed to cause a conformational 19 change in the protein that would account for the lack of binding of most of the monoclonal antibodies tested against GPIba. This mutation could be responsible for the absence of va binding . This case indicates the requirement of the leucine-rich region for the correct exposure of the va binding site as well as for the correct assembly and stability of the GPIb-IX-V complex on the platelet surface. Simsek et al. 1994c, used monoclonal antibodies directed against GPIba and GPIb-IX to show the absence of detectable GPIba and the presence of small amounts of GP-IX. This indicated a defect in the a subunit of GPIb. Sequence analysis revealed a homozygous single base pair mutation, T to A, this led to the substitution of serine for cysteine at position 209 of the mature protein. Cysteine 209 in GPIba was found to be involved in intrachain disulphide bond formation with Cys 248. This interaction is important in forming one of the loops that are critical for the ability of the complex to bind va. So this mutation had affected the proper folding of GPIba , and presumably its surface expression as well as its non covalent association with GPIX and GPV which resulted in the disease. In another case of BSS, Simsek et al. 1994a, analyzed platelet membrane glycoproteins by flow cytometry. This revealed a significant decrease or absence of GPIba and low levels of GPIX and GPV. Sequence analysis demonstrated a homozygous deletion of T 317 resulting in a frame shift mutation. This deletion causes a shift in the reading frame, predicting a premature stop codon after 19 amino acids leading to a severely truncated nonfunctional protein. Ware et al. 1993, suggested that the structural integrity of the leucine-rich repeat is necessary for normal function of the GPIb-IX-V complex and possibly for normal 20 platelet morphology. They studied a patient with BSS Bolzano and found a homozygous mutation which converted Ala 156 to Val within the leucine-rich repeat thus causing a structural abnormality resulting in the loss of conformation-sensitive epitopes. This mutation was responsible for decreased va binding. They suggested that the appropriate conformation maintained by the LRR in the extra cytoplasmic domain may also influence the linkage of the intracytoplasmic tail of GPIba to the cytoskeleton. They concluded that the Ala 156 to Val mutation may be responsible for both the ligand binding defect and the apparent morphology of giant platelets in the Bolzano variant. Kunishima et al. 1994, studied a novel variant phenotype of BSS in a female patient. Her platelets completely lacked the expression of GPIba. However, her platelets’ lysates contained GPIba (23 % of normal value), and her plasma contained the glycocalicin part of GPIba (60% of the normal values). GPIbB and GPIX were expressed in reduced amounts. DNA sequence analysis showed a homozygous single nucleotide substitution from a serine codon (TCA) to a nonsense codon (TAA) at residue 444 in GPIba gene. The mutant gene generated a truncated GPIba molecule lacking the transmembrane region and cytoplasmic tail. This mutation resides in the boundary between the macroglycopeptide and the transmembrane coding region which prevented the formation of the disulphide bond between GPIba and GPIbB. Consequently GPIba may be secreted from megakaryocytes and platelets as a soluble glycocalicin like molecule. Further the truncated GPIba may be loosely associated with other subunits in the cytoplasm. When they cross the plasma membrane, the truncated GPIba with no transmembrane domain is secreted into plasma, while GPIbB and GPIX are inserted into the membrane. 21 Li et al. 1995, studied two cases of BSS in two related individuals (first cousins) who had the clinical features of the disease. A point mutation was identified in codon 129 in GPIba , this mutation lies in the first position of the fifth LRR, changing a leucine (CTC) to a proline (CCC). Transcription of GPIb-IX seemed to be normal but its surface expression was decreased (40% of normal), whereas the GPV level was found to be normal. These findings could indicate the critical importance of the LRR of GPIba, and a defect in this motif could influence surface expression of GPIb-IX-V, hence changing the nature of receptor-ligand interaction. As mentioned earlier BSS could be due to a defect in GPIbcL , GPIbB, GPIX or GPV genes. Double heterozygosity for mutations in the GPIX gene in three siblings with classical BSS have been reported by Wright et al. 1993. Although the expression of GPIba was reduced on the platelet surface, lesions at its locus had been excluded by restriction fragment length polymorphism analysis, whereas defects in GPIbB gene were excluded by single stranded conformation polymorphism analysis. Analysis of the GPIX gene showed two different missense mutations in the coding region of the GPIX gene. First, an A to a G transition in codon 21 resulted in conversion of a conserved aspartic acid to glycine within the amino terminal flanking sequence, and an A to a G change in codon 45 converted a conserved asparagine to serine within the core of LRR motif. Three affected individuals were doubly heterozygous for these mutations, which alter conserved residues in the GPIX LRR motif. Experiments performed in a Chinese hamster ovary (CHO) cell expression system showed that GPIX is a necessary requirement for membrane expression of the GPIb-IX complex, and the same seems to be true for both GPIba, and GPIbB but not apparently for GPV. Wright et al. suggested that these 22 missense mutations cause severe disruption of GPIX structure and it supports evidence that the GPIX molecule, although small, is of critical importance in the formation of the intact, functional complex. The bulk of the GPIX molecule lies outside the plasma membrane, and its extracellular domain is largely composed of a single LRR motif. This study indicates the potential importance of this region. Sae-Tung et al. 1996, took advantage of the study done by Wright et al. 1993, in which double heterozygote mutations in GPIX resulted in BSS. They studied the effect of these mutations on the biosynthesis of the GPIb-IX complex. Alpha and [3 CHO cells were transfected with the mutants and wild type GPIX. Wild type GPIX increased the expression of GPIb, whereas both mutants failed to do so. Cotransfection of wild type and mutant GPIX with GPIbB resulted in detection of wild type of GPIX on CHO surface but not the mutants although these mutant polypeptides were detected in low amounts in the intracellular compartments. Imunoprecipitation studies showed that mutant GPIX associates poorly with GPIbB which prevents the stabilization of GPIb on the cell surface thus resulting in BSS. These studies indicate the importance of the conserved leucine- rich repeat of GPIX in the association between GPIX and GPIb-B, and also the importance of GPIX in stabilizing GPIb-IX complex on the platelet surface. Clemtson et al. 1994, studied another variant of BSS where a homozygous mutation lies in the leucine-rich domain of GPIX. Surface labeling of the platelets and two dimensional gel electrophoresis showed reduced but detectable amounts of GPIb-IX- V. However, there was markedly less GPIX than GPIba, [3, or GPV. This disproportion was confirmed by western blot analysis. Sequence analysis showed no mutations in the GPIba gene, whereas a point mutation (A to G) was found in GPIX converting Asn 45 to 23 Ser within the leucine-rich domain. Both alleles of GPIX contained the same defect, which was confirmed by the appearance of a new site for the restriction enzyme F nu4H1. It seemed that this mutation changed the conformation of the leucine rich domain, thus reducing surface expression of the complex by disturbing the association of its parts. Clemetson et al. 1994, suggested that among other possible functions, the leucine-rich domain present on all the components of GPIb-IX-V may play a role in the assembly and surface expression of the complex. Because this mutation is the same as one of the two single base pair substitutions reported by Wright et al. 1993, Clemetson et al. 1994 proposed that codon 45 in GPIX could be vulnerable to mutation because the two families are geographically widely separated. Table 1 summarizes the cases discussed above and shows the different mutations that are related to BSS. Polymorphism in GPIb-IX Two major protein polymorphisms of GPIb-IX complex have been described, both of which affect GPIba. One is a molecular weight polymorphism due to the presence of a variable number of tandem repeats (VNTR) in the macroglycopeptide region. The other is a dimorphism at codon 145 where a Met (ATG), or a Thr (ACG) can be present. This polymorphism is illustrated in Figure 2, which was adapted with modifications from Lopez 1994. The first polymorphism is molecular weight polymorphism. Size differences in GPIba were first described by Moroi et al.1984, who noted four variants (in Japanese donors) that migrated on non-reduced SDS-polyacrylamide gels with molecular masses ranging from 153 KD to 168 KD. They were designated A, B, C and D in order of Table 1. Mutations found in BSS. 24 Protein Mutation Result Expression on Reference ‘ fl membrane GPIba Deletion of Leu 179 BSS Very low levels of Salle et al., GPIba. 1995 GPIba Leu 129 to Pro in the BSS Low levels of GPIb - Li et al., 1995 fifth LRR IX Normal GP V GPIba Deletion of T 317 BSS Very low to absent Simsek et al., resulted in a premature GPIba. 1994a stop codon Low levels of GPIX and V GPIba Cys 209 to Ser BSS Absence of GPIba Simsek et al., Low levels of GPIX 1994c GPIba Ser 444 to stop codon BSS Absence of GPIba Kunishima et Low levels of GPIbB al,. 1994 Low levels of GPIX Truncated GPIba in plasma GPIba Ala 156 to Val in LR BSS Dysfunctional GPIb- Ware et al., Type IX-V. 1993 Bolzano GPIba Leu 57 to Phe in the BSS Abnormal GPIba Miller et al., first LR 1992 GPIba Trp 343 to stop codon BSS Absent GPIba Ware et al., 1990 GPIX Asp 45 to Ser in the BSS Low levels of GPIb- Clemetson et LRR IX-V al., 1994 GPIX Asp 21 to Gly and BSS Low GPIba. Wright et al., Asp 45 to Ser in the No GPIX 1993 LR 25 GPlba "m e e .%mr4smu " O worms-i - O—N-Iflml ‘ - _. ".me /// l/l/l/M Figure 2. Polymorphism of GPIb-IX complex. Length polymorphism involves 13 amino acids, starts at Ser 399 and ends at Thr 411, where n represents number of copies of the repeat. n=1, 2, 3, or 4 for variants D, C, B, and A respectively. The other polymorphism is at residue 145 (EPA-2), where a Met or a Thr can be found. Figure is adapted from Lopez 1994. 26 decreasing molecular mass. The frequency of the variants were quite different between American Caucasian and the Japanese population. In both p0pulations the C variant was the most common, whereas the two largest alleles were uncommon (Table 2). The region accounting for the polymorphism was later identified as the GPIba macroglycopeptide. PCR and sequence analysis of genomic DNA performed on 207 individuals from four ethnic groups (Lopez 1994), showed three different PCR fragment lengths and the polymorphism was shown to result from either one or two duplications of the 39 base pair sequence corresponding to Ser 399 through Thr 411. However the fourth length variant was not found in this study. Each added repeat of 13 amino acids contains five potential sites for glycosylation and thus may add up to 6 KD per repeat to the molecular mass of the polypeptide. Extensive glycosylation coupled with heavy proline content suggested that the repeated sequence has an extended conformation (Lopez 1994). Lopez 1994, also predicted that each repeat would increase the length of the macroglycopeptide and thus the distance of the ligand binding domain from the platelet membrane by about 32 A. Length polymorphism could be a determinant of platelet sensitivity to shear due to moving va—binding domain farther out from the platelet plasma membrane which is likely to affect the fluid forces to which this region is exposed. Ishida et al. 1996, studied the molecular evolution of GPIba polymorphism in Eastern Asian population and they indicated that the frequencies of these polymorphisms differ considerably depending on race and the largest variant with four tandem repeats is almost exclusively present in the Japanese population. This study included different races from Eastern Asia, Japan (n= 103), Korea (n= 101), and China (n= 177). The A isoforrn 27 (four tandem repeats) was detected in all groups, and the T-4 haplotype, which has Met 145 (ATG) and 4 tandem repeats (TRs), held a majority in these three groups. Homozygotes for haplotype T-4 were found among both the Japanese and the Korean populations but not in the Chinese Table 3. However, haplotype T-2 which has Thr 145 (ACG) and 2 TRs and was first identified in European Caucasians, was not detected in this study. Table 2. Gene frequency of molecular weight polymorphism in Japanese and American population. A = highest molecular weight, D = lowest molecular weight Gene frequency“ American Japanese A 1 % 7% B 1 1% l % C 80% 56% D 8% 35% * Data adapted from Lopez 1994. So far these variants had no known functional differences, but whether modest variations (30 A per tandem repeat) in the distance by which the active site of GPIb extend from the surface, influence adhesion and the susceptibility for pathologic thrombi 28 to form is not known (N urden 1995). The other type of polymorphism is human platelet alloantigen (HPA-Z). This polymorphism is within the LRR, it resulted from the presence of a Thr145 (HPA-2a), or Met 145 (I-IPA-2b) in the N-terminal globular domain of the human platelet GPIba (N urden, 1995). Alloantibodies against either dimorphic variant can cause neonatal thrombocytopenia (Bizzaro et al., 1988) or can account for refractoriness to platelet transfusion (Saji et al., 1989). Table 3. Tandem repeat polymorphism of individuals with ATG 145. A= 4 tandem repeats(TR),B=3TR,C=2TR,D=1 TR Genotype of TR“ Population AA AC AD BC BD Total Japanese 3 l3 1 3 l 1 3 1 Korean 1 8 5 0 1 l 5 Chinese 0 9 0 1 1 1 1 ‘Data adapted from Ishida et al. (1996) The genotype frequency of HPA-2a in the white population is 92.6%, that of HPA-Zb is 7.4% (Kuijpers et al., 1992). The size polymorphism and HPA-2 are in linkage disequilibrium. The D variant (1 TR) or C variant (2 TRs) are associated with HPA-2a. The B variant (3 TRs) or C variant (2TR) are associated with HPA-2b . Ishida 29 et al. 1991, found that A or B variants are associated with HPA-2b. Simsek et al. 1994b, studied the genetic association between Met 145 / Thr and the presence of VNTR polymorphism in 106 Caucasian individuals (Table 4). Simsek et al. 1994b, concluded that the two polymorphisms in GPIba, HPA-2 at the N-terrninus and Table 4. Frequency distribution of GPIba alleles Allele HPA-2/ VNTR Frequency“ I HPA-2a / D 0.1 1 II HPA-Za / C 0.82 111 HPA-2b / C <0.01 IV HPA-2b / B 0.07 *Data adapted from Simsek et al., 1994b. the size polymorphism of 13 amino acid sequence 399SEPAPSP'I"1'PEP'I“’“ are genetically linked in the Caucasian population. A possible evolutionary model is that allele type I (with D variant and a Thr 145) was the ancestral stage of the protein, then allele type II was the result of tandem duplication of the 13 amino acids sequence. A mutation from Thr 145 to Met 145 had occurred giving rise to type III. A duplication of the 13 amino acids sequence occurred immediately after this mutation leading to the formation of type IV which has Met 145 and 3 copies of the same 13 amino acids. A variant with 4 repeats 30 had not been found in this group but Simsek hypothesized that A variant would have Met at residue 145 Family History: The family that is studied in this research is of Middle Eastern origin. Consanguinity has been noticed as the parents are cousins. The patient, a female of 22 years age, has all the clinical features of the disease including bleeding episodes, low platelet count (29,000 cells /uL), prolonged bleeding time (greater than 20 minutes) and giant platelets. Aggregation of the patient’s platelets in response to ristocetin showed a value of 16% compared to more than 39% in a normal person. Treatment of patient’s platelets with antibodies against GPIba, showed low reactivity compared to normal. Two dimensional gel electrophoresis indicated the absence of GPIba from patient’s platelets. Other family members show some criteria such as thrombocytopenia and giant platelets, but no bleeding. The family pedigree is shown in Figure 3. Objectives: In this research we studied the GPIba gene in the patient mentioned above and her mother, in an attempt to localize and identify the mutation that could be the cause for an absent GPIba on the patient’s platelets. Understanding the molecular defects in BSS will shed more light on the mechanism by which va interacts with its receptor (GPIb- IX) on the platelet surface. Identifying the defect in the patient’s gene will also help in screening other members of the family for heterozygosity. This could be done by different approaches, such as synthesizing a probe corresponding to the mutation and 3 1 hybridizing it with the questioned gene, or using the restriction fragment length polymorphism (RF LP) technique if the mutation altered a site for a restriction enzyme, or if it generated a new restriction site. ’ ’ ’ v‘.‘ w 11711 9000i: Family Studied QQQ'. Obi'lflfi u gimtphbhts (mum prm ONotested Pnbdib hehnzjgote not used Figure 3. Pedigree of the studied family. A pedigree of the family studied shows the predicted genotypes based on the phenotypic characteristics of the disease. Materials and Methods Sample collection and DNA isolation Whole blood was collected from the propositus, her mother and a normal control, using ethylenediaminotetraacetic acid (EDTA) as the anticoagulant. The samples were centrifuged at 750 x gravity (g) for 20 minutes (centrifugation steps during DNA isolation were performed at room temperature) . Buffy coats were transferred to 15 mL conical polypropylene screw cap tubes and residual red blood cells were lysed by bringing the volume up to 15 mL with Tris ammonium chloride (140 mM NH4C1, 6 mM Tris-HCl [pH 7.2]) followed by a 5 minutes incubation at 37 °C. Each tube was centrifuged at 200 xg for 20 minutes and the supernatant was aspirated and discarded without disturbing the pellet. The pellet was resuspended with 10 mL calcium/magnesium free Dulbeco’s phosphate buffered saline (DPBS) (Sigma, St. Louis, Missouri) and centrifuged at 200 xg in a microcentrifuge for 10 minutes . The pellet was washed 3 second time by adding 0.75 mL DPBS per 1 mL original whole blood and centrifuged at 225 xg for 20 minutes. The supernatant was decanted and the pellet was resuspended in the residual liquid. White blood cells were lysed by adding 0.375 mL of lysing solution (10 mM Tris [pH 7.6], 0.25 mM EDTA [pH 8.0], 0.75 mM NaCl) per 1 mL original whole blood and 18.75 uL of 20% sodium dodecyl sulfate per lmL lysing solution. Cellular proteins were digested by adding 1.5 ul proteinase K (50 mg/mL) (Boehringer Mannheim, Indianapolis, 33 34 IN) per 1 mL lysis solution. The RNA was removed by adding 100 ug of DNase free RNase (Behringer Manheim, Indianapolis, IN) per 1.0 mL lysis buffer and incubating at 37 °C for 1 hour. Cellular debris which contains protein and RNA was precipitated by the addition of 0.125 mL of 6 M saturated NaCl per 1 mL original whole blood followed by vigorous shaking for 10 seconds and centrifugation for 10 minutes at 850 xg. The supernatant, containing the DNA was decanted into a clean tube without disturbing the pellet containing the cell debris. The supernatant was spun for an additional 10 minutes at 850 xg to further remove any remaining proteins. The supernatant was poured into a clean glass tube, and DNA was precipitated by adding cold (-20 °C) absolute ethanol (2X the volume of supernatant). Samples were mixed by gentle inversion. DNA was collected using a wide bore plastic transfer pipet, transfening the DNA to a 1.5 mL microtube followed by washing twice with 90% cold ethanol by filling the tube with 90 % ethanol and centrifuging at 10,000 xg in a microcentrifuge (Microspin 24S, Sorvall Instrument Dupont). The pellet was dried in a speed vac centrifuge under vacuum for 5-7 minutes and resuspended in 100 uL sterile double deionized water (dde). The concentration of DNA was estimated by preparing a dilution of 1:100 in dde and measuring the optical density (OD) at 260 nm. Calculation of DNA amount was done by using the formula: OD at 260 X 50 X dilution factor = mg/mL DNA. The OD at 280 nm was also measured in order to verify the purity of the DNA where the ratio of 260/ 280 was calculated and a value of 1.8 - 2.0 indicates a clean DNA. 3 5 DNA amplification The gene encoding GPIba was amplified using 3 sets (1 , 2 and 3) of oligonucleotide primers, Table 5, which generate 3 overlapping fragments. Oligonucleotide primers were adapted from Simsek et al, 1994a with some modifications and synthesized at the Macromolecular Structure Facility at Michigan State University using the 394 DNA synthesizer (Applied Biosystem, Foster City, California). Table 5. Sequences of primers used for amplification and sequencing. Set Primer Binding site Sequence" 5' - 3' Expected size Set 1 DEl 554 - 573 GCTGCTCCTGCTGCCAAGCC 704 bp DE2 1256 - 1237 AGCATTGTCCTGCAGCCAGC Set 2 DE3 1091 - 1111 ATGGGCTGGAGAATCTCGACAC 664 bp DE4 1753 - 1730 GGAGTGGGCTCCAGGGTGGTCATG Set 3 DE7 1658 - 1681 AACTCCAAAATCCACTACTGAACC 785 bP DE6 2442 - 2417 CTCAAGGTCCCCAAACCTCCCACC C 36 The first fragment was amplified using oligonucleotide primers DE] and DE2 (primers set 1), the second fragment was amplified using oligonucleotide primers DE3 and DE4 (primers set 2), and the third fragment was amplified using oligonucleotide primers DE7 and DE6 (primers set 3). Amplification of the first and second fragments was accomplished by mixing 200 nanograms of DNA (or an equivalent amount of sterile dde for the negative control) with 200 uM of each deoxynucleotide triphosphate (dNTP), 100 uM of each oligonucleotide primer, 10 uL of 10 X PCR reaction buffer (GIBCO BRL), 1.5 mM MgC12, and 2.0 U Taq Polymerase (GIBCO, BRL), in a 100 uL reaction volume. The PCR reactions were performed using a GeneAMP 9600 thermal cycler (Perkin Elmer Cetus, Norwalk, CT.) as follows: following an initial denaturation step for 5 minutes at 95 °C, DNA was amplified during 35 cycles of 95 °C for 45 sconds, 68 °C for 45 seconds, and 72 °C for 45 seconds. The PCR reactions were completed by a final elongation step at 72 °C for 10 minutes. Amplification of the third fragment was accomplished in a manner similar to the previous PCR protocol with the following modifications. Briefly, 200 nanograms of DNA (or equivalent amount of sterile dde for negative control) was combined with 200 uM of each deoxynucleotide triphosphate (dNTP), 100 uM of each oligonucleotide primer, 10 uL of 10 X PCR reaction buffer (GIBCO BRL), 0.75 mM MgC12, and 4.0 U Taq Polymerase (GIBCO BRL), in a 100 uL reaction volume. The PCR reactions were performed using a GeneAMP 9600 thermal cycler (Perkin Elmer Cetus, Norwalk, CT.). Following an initial denaturation step for 5 minutes at 95 °C, DNA was amplified during 30 cycles of 95 °C for 45 seconds, 65 °C for 45 seconds, and 72 °C for 45 seconds. The PCR reactions were completed by a final elongation step at 72 °C for 10 minutes. 37 Agarose gel electrophoresis Ten uL of the PCR products were mixed with 1 uL of 10X loading dye (0.25% bromophenol blue, 0.25 xylene cyanol , and 15% (w/v) sucrose in ddHZO) and run on 1% (WN) agarose gel (5 x 10 x 1 cm) in 1X Tris acetate EDTA buffer (TAE) (40 mM Tris- 2W434Acetate, 1 mM EDTA [pH 8.0]). One ug of the one Kb molecular weight marker (GIBCO, BRL) was loaded on the gel to estimate the sizes of the amplified products. The gel was run at 75 volts for 2-3 hours. Gels were stained with ethidium bromide (5 ug/mL ethidium bromide in dde) for 20 - 30 minutes with gentle agitation, and destained by soaking in dde for another 20 - 30 minutes with gentle agitation. Gels were transferred to a trans-illuminator with an ultra violet (UV) light source to visualize the amplified products. The gels were documented by photographing using type 667 Polaroid film (Polaroid, Cambridge, Massachusetts). PCR products purification PCR products were purified using WizardTM PCR Preps DNA Purification System for Rapid Purification of DNA Fragments kit (Promega, Madison, Wisconsin) following manufacturer’s recommendations. Briefly, PCR products were run on 0.8% GTG low melting point (LMP) agarose (FMC BioProduct, Rockland, Maine) , gels were electrophoresed, stained and destained under conditions similar to those described above. The appropriate size bands were cut from the gel, transferred to 1.5 mL microtubes, and incubated at 65 °C for 10 - 15 minutes to dissolve the LMP agarose. One mL of resin solution was added to the melted agarose slice and mixed by inversion for 20 second. The mixture was transferred to a 3 mL syringe barrel which was attached to the . 38 minicolurnn assembled at the vacuum manifold. After applying the vacuum to draw the DNA/resin mix into the minicolumn, the column was washed with 2 mL of 80% isopropanol. Columns were dried by centrifugation for 20 seconds at full speed in a microcentrifuge (Microspin 24S. Sorvall Instrument, Dupont). DNA was eluted from the columns by applying 50 uL of pre-warmed (65 °C ) sterile ddeO, incubating for 1 minute at room temperature, and centrifugation for 25 seconds at full speed in a microcentrifuge. Three uL of the purified PCR product were run on a 1% agarose gel as previously described to check the purity and the yield of the DNA. DNA was stored at -20 °C for further usage. Heteroduplexing analysis Using a GeneAMP 9600 thermal cycler ( Perkin Elmer Cetus, Norwalk, CT.) purified PCR products were prepared for heteroduplexing analysis as follows: 15 uL of the PCR products were heated over a period of 4 minute to 94 °C, tubes were held at 94 °C for 10 minutes, cooled down to 25 °C over a 4 minutes period and held at 25 0C for 30 seconds. Each sample was mixed with 3 uL 6X loading dye (0.25% bromophenol blue, 0.25 xylene cyanol , and 15% (w/v) sucrose and 3 uL 8 M urea). Samples were loaded onto 10% urea / 12% polyacrylamide (29 :1 acrylamidezN, N-methylenebisacrylamide)ge1 (20 x 16 x 0.05 cm) in 1X Tris borate EDTA (TBE) ( 0.089 M Tris base, 0.089 M Boric acid 0.002 M EDTA), and electrophoresed for 20 hours at 100 volts. The gels were maintained cool during electrophoresis by running cold tap water in the special jackets which are part of the electrophoresis apparatus (Protean 11, Bio Rad, Hercules, CA). After electrophoresis the gels were fixed in fixing solution (10% absolute 39 ethanol, 0.005% glacial acetic acid) for 20 minutes with gentle agitation. The gels were stained with fresh 0.2% silver nitrate solution (0.2 gm silver nitrate in 100 mL dde) for 30 minutes with agitation, rinsed with dde for 5 seconds, immersed in fresh developing solution (3% NaOH, 0.008% Formaldehyde) for 8-12 minutes with agitation and washed in ddeO for 5 minutes. The gels were transferred to a Whatrnan filter papers (3 MM) and dried on a vacuum dryer for 1 hour at 80 °C. Single Stranded Conformation Polymorphism (SSCP) For SSCP the optimal fragment size is approximately 300 bp. Because the PCR fragments that were generated were bigger than this, the PCR products were digested with different restriction enzymes to generate smaller size fragments. Primer sets 1 and 2 were digested with Ban I rstriction enzyme (Boehringer Mannheim, Indianapolis, Indiana) as followsz26 uL purified PCR products were mixed with 3 uL 10X Buffer A (Tris-acetate 33 mM [pH 8.0], Mg-acetate 10 mM, K-acetate 66 mM, Dithiotheitol (DTT) 0.5 mM), 10 units Ban I and the mixture was incubated at 37°C over night. Primer set 3 was digested with SstI and P51 I restriction enzymes (GIBCO, BRL) as follows: 25 uL of purified PCR products were mixed with 3 uL 10X React 2 buffer (Tris-HCl 50 mM, MgCl2 10 mM, NaCl 50 mM), 10 U 53! I, 10 U Pst I, and the mixture was incubated overnight at 37°C. To check for complete digestion, ten uL of the restriction enzyme digest were mixed with 1 uL of 10X loading dye and loaded on 8% polyacrylamide gel (16.4 mM acrylamide, 0.2 mM bis(N,N’-methy1ene-bis-acrylamide)) (8.6 x 8.5 x 0.1 cm), along with 1 Kb molecular weight marker.(GIBCO, BRL). The gel was electrophoresed in 1X TBE buffer at 110 volts for 2 hours, stained with ethidium bromide (0.5 ug/uL dde) for 10 40 minutes, destained with dde for another 10 minutes and visualized using a trans- illumination with a UV light source. The gel was photographed using polaroid type 667 film (Polaroid, Cambridge, Massachusetts). Single Stranded Conformation Polymorphism (SSCP) analysis was performed (with modifications) as recomended by Ainsworth J. et al 1994,. Briefly, the remaining 20 uL of the restriction digest was mixed with 4 uL of denaturant solution (97.5% deionized formarnide, 4.6 M urea, 0.3% bromphenol blue, 0.3% xylene cyanol, 10 mM EDTA) in order to maintain the denatured DNA as single stranded. To denature the DNA, the mixture was heated to 95 °C for 10 minutes in GeneAMP 9600 thermal cycler (Perkin Elmer Cetus, Norwalk, CT.). The mixture was chilled on ice and loaded immediately on 6% polyacrtlamide (29:1 acrylamide : N, N’-methylene bisacrylarnide) gel with 5% glycerol (v/v) (8.5 x 8.6 x 0.01 cm), the gel was electrophoresed in 1X TBE at 180 volts for 1 hour at room temperature. After electrophoresis the gel was stained in a way similar to that described in the heteroduplexing section above except that the silver nitrate concentration was reduced to 0.1% in ddHZO. The gel was transferred to 3 MM Whatrnanc filterpaper and dried on vacuum dryer as explained in heteroduplexing Cloning The cloning vector pUC19 was used which contains an arnpicillin resistance gene that allows for colony selection when grown on a medirun with ampicillin. pUC19 also has the Lac Z gene which allows for blue versus white color screening (Figure 5). One ug of pUC19 cloning vector (Boehringer Mannheim, Indianapolis, IN) was linearized by digestion with 20 U Sma I restriction enzyme in Buffer A at 37 °C for 2 hours. The pUC19 2.69 kb ED’ Figure 4. Cloning vector pUC19. The diagram represents pUC19 cloning vector, where the genes encoding Beta-lactamase (Amp'), Beta-galactosidase (Mel), and lac repressor (LacI), the origin of replication (ori) and the multiple cloning sites (MCS) where some restriction enzymes recognition sites are illustrated. Also the binding sites for the M13 forward and reverse primers are pointed by an arrow. 42 linearized vector was dephosphorylated to minimize the vector religation to itself during the ligation reaction. Dephosphorylation was performed by combining 27 uL of the digested product, 35 uL of 10X dephosphorylation buffer (500 mM Tris- HCl [pH 9.0], 10 mM MgC12, 1 mM ZnClz, 10 mM spermidine), 4U Calf Intestinal Alkaline Phosphatase (CIAP) (Boehringer Mannheim, Indianapolis, IN) and 284 uL sterile ddeO. The mixture was incubated at 37 °C for 30 minutes followed by addition of another unit of CIAP and continued incubation for another 30 minutes at 37 °C. The enzyme was inactivated by heating at 65°C for 10 nrinutes and the vector was extracted with phenol/ chloroform as follows: 350 uL of phenol/ chloroform (24:1) were added to the mixture, vortexed briefly and centrifuged for 2 minutes in microcentrifuge at full speed. The aqueous phase was transferred to a clean tube and extracted again with phenol/ chloroform as above. The remaining phenol was extracted by adding 350 uL of chloroform/ isoamyl alcohol (24:1) followed by vortexing and centrifugation as above. The aqueous phase was transferred to a clean tube and the vector was precipitated by the addition of 875 uL absolute ethanol, 35 uL 3 M sodium acetate and incubation at -70°C for 30 minutes, followed by centrifugation for 15 minutes at 10,000 xg in rrricrocentrifuge at 4°C. The pellet was washed by adding 250 uL of 70% ethanol followed by centrifugation for 5 minutes at 10,000 xg in microcentrifuge. The pellet was dried in a vacuum centrifuge centrifuge at room temperature for 5 minutes and resuspended in 30 uL sterile ddeO. Since oligonucleotide primers lack the phosphate group at their 5' end, PCR products (insert) were phosphorylated using T4 polynucleotide kinase as follows: 15 uL of purified PCR products were mixed with 28 uL sterile dde, 5 uL of 10X T4 43 polynucleotide kinase buffer (70 mM Tris- HCl [pH 7.6], 100 mM MgC12, 5 mM DTT), 1 uL of 10 mM ATP and 1 uL T4 polynucleotide kinase (New England Bio Lab, Beverly, MA), the mixture was incubated at 37°C for 15 minutes. The reaction was stopped by adding 2 uL of 0.5 M EDTA. The volume of the mixture was expanded to 100 uL by addition of 50 uL Tris-EDTA (TE) buffer (pH 7.6). The DNA was extracted once with 100 uL of phenol/ chloroform, and once with 100 uL of chloroformzisoamyl alcohol as described above. The DNA was precipitated with 250 uL absolute ethanol and 10 uL 3 M sodium acetate as previously described, the pellet was resuspended in 13 uL sterile ddHZO. Ligation of the vector and insert was accomplished in 2 steps, the first step was performed in a small volume to enhance intermolecular ligation between insert and vector. This was accomplished by mixing 7 uL of phosphorylated PCR products, 1 uL dephosphorylated pUC19, l uL 10X ligation buffer (660 mM Tris-HCl, 50 mM MgC12, 10 mM DDT, 10 mM ATP) and l U T4 DNA ligase (Boehringer Manhheim, Indianapolis, IN). The mixture was incubated at 15 °C for 8 hours. In the second step, the volume was expanded to 50 uL to enhance intramolecular ligation within the same molecule by adding 34 uL sterile dde, 5 uL 10X ligation buffer and 1 U T4 DNA ligase enzyme followed by incubation at 15 °C overnight. The volume of the reaction was expanded to 100 uL by adding 50 uL of TE buffer (pH 7.6) and the ligated product was precipitated by adding 250 uL absolute ethanol, 10 uL 3 M sodium acetate. This mixture was incubated at -70 °C for 30 minutes and centrifuged for 15 minutes at 10,000 xg in microcentrifuge at 4 °C. The pellet was washed once with 70% ethanol and dried in a speed vac centrifuge and resuspended in 10 uL sterile dde and used for transformation. 44 Luria-Bertani (LB) media (1% bacto-tryptone, 0.5% bacto-yeast extract, 1% NaCl, 1.5% agar) was used for bacterial growth. Plates were prepared as follows: 5 mg ampicillin/ 100 mL LB media was added to select for cells that have been transformed by the plasmid (since the plasmid has arnpicillin resistance gene), media were poured and allowed to solidify. Forty uL of 0.1 mM isopropylthio-B-D-galactoside (IPTG) and 40 uL of 40 ug/mL 5-bromo-4-chloro-3-indolyl-B-D galactoside (X-gal) were spread on the plate surface to allow for blue /white color screening For transformation, frozen (- 70 °C) DHSa competent cells (Stratagene, La Jolla, CA) were thawed slowly on ice over a period of 20 minutes, 5 uL of the ligation product was mixed with cells and incubated on ice for 30 minutes. The DNA was transformed into the cells by heat shock at 42 °C for 90 seconds followed by a cold shock (on ice)for 2 minutes. Nine hundred fifty uL of LB broth was added to the cells and the cells were allowed to multiply by incubation at 37 °C for 1 hour in a shaking incubator. The cells were collected by centrifugation at 2000 xg in a microcentrifuge for 4 minutes. 800 uL of the supernatant was discarded while the cells were resuspended in the remaining 200 uL of broth. To select for transformants, 50 uL of this volume was plated on an LB agar plates, prepared previously, and left at room temperature for 15 minutes to allow the cells to adsorb to the agar surface and incubated overnight at 37 °C. To analyze the transformants, small-scale plasmid DNA preparations (minipreps) were performed as follows: 5 mL of LB broth with arnpicillin was inoculated with selected single white colonies and incubated at 37°C overnight in a shaking incubator. At the same time a replica plate was inoculated with the same colony to double check for selected single white colonies and to be saved for further DNA purification and 45 preparation of glycerol stocks. Purification of insert DNA was performed using WizardTM Plus Minipreps DNA Purification System (Promega, Madison, WI) following the manufacturer’s recommendations. Briefly, the 5 mL culture was centrifuged at 12,000 xg for 10 minutes. The supernatant was discarded and the pellet was resuspended in 300 uL cell resuspension solution (50 mM Tris-HCl (pH 7.5), 10 mM EDTA, 100 ug/mL Rnase A) and transferred to a 1.5 mL microcentrifirge tube. The cells were lysed with 300 uL of cell lysis solution (0.2 M NaOH, 1% SDS) mixed with 300 uL neutralization solution (1.32 M potassium acetate, pH 4.8). Lysates were centrifuged at 10,000 xg for 5 minutes at room temperature. Meanwhile syringe barrels and minicolumns were assembled at the vacuum manifold. The clear supernatant was transferred to the syringe barrel which contained 1 mL of resuspended resins. After applying the vacuum to draw the DNA/resin mixture into the minicolumn, the column was washed with 2 mL of washing solution (80 mM K-acetate, 8.3 mM Tris-HCl (pH 7.5), 40 uM EDTA, 55% ethanol). The columns were dried by centrifugation for 2 minutes at 10,000 xg in a microcentrifuge. DNA was eluted from the columns by applying 50 uL of pre-warmed (65 °C ) sterile ddeO, incubated for 1 minute at room temperature, followed by centrifugation for 25 seconds at 10,000 xg in a microcentrifuge. DNA from the Minipreps was used to analyze for the presence of recombinants by digestion with EcoRI and HindIII (GIBCO, BRL). One uL of minipreps DNA was mixed with 2.5 uL React 2 buffer (GIBCO, BRL), 10 U EcorI, 1 U HindIII and 19.5 uL sterile dde and incubated at 37 °C for 2 hours. The mixture was mixed with 2.5 uL 10X loading dye and loaded on a 0.8% submerged agarose gel along with a 1 Kb molecular weight marker (GIBCO, BRL). The gel was run at 75 volts for 1.5 hours in 1X 46 TBE buffer. The gel was stained with ethidium bromide and photographed as previously described. To check for the orientation of the insert within the plasmid the restriction enzyme PstI (GIBCO, BRL) was used. PstI cuts the vector once just outside the insert as well as within the insert, the reaction mixture consists of 1 uL of purified clones, 2.5 uL React 2 buffer, 20.5 uL sterile dde and 10 U PstI followed by incubation at 37 °C for 2 hours. 2.5 uL of 10X loading dye was added to the mixture and applied to 0.8% agarose gel along with 1 Kb molecular weight marker (GIBCO, BRL), the gel was run, stained and photographed as previously described. After obtaining the desired clones, glycerol stocks were prepared as follows: 5 mL of LB broth with ampicillin (0.05 mg/mL) was inoculated from the same colonies that were used to grow the culture for minipreps above. The culture was inubated at 37 °C overnight in a shaking incubator. 800 uL of this culture was transferred into microcentrifuge tube, mixed with 200 uL of sterile glycerol and stored at -70 °C Preparation of Large scale DNA A cesium chloride (CsCl) gradient centrifugation method (Sarnbrook et al 1992) was used to obtain clean DNA for sequencing. Two hundred and fifty mL of LB broth with arnpicillin were inoculated with the selected and identified clones and incubated overnight at 37°C in shaking incubator, the culture was centrifirged at 14,000 xg for 10 minutes at 4°C in a Sorvall centrifuge. The supernatant was discarded and the cells were resuspended in 12 mL resuspension solution (50 mM Tris-HCl (pH 7.5), 10 mM EDTA, 100 ug/mL Rnase A) lysed with 12 mL lysing solution (0.2 M NaOH, 1%SDS) and 47 neutralized with 12 mL neutralization solution (1.32 M potassium acetate, pH 4.8) followed by centrifugation at 14,000 xg for 15 minutes at 4 °C. DNA was precipitated from the clear supernatant by mixing with 0.6 volume isopropanol followed by centrifugation at 14,000 xg for 15 minutes at 4°C. The pellet was washed with 70% ethanol , dried and resuspended in 11.3 mL sterile TE buffer (pH 7.5). Cesium chloride was added (11.55 gm) and dissolved completely, 8 mg ethidium bromide (800 uL of 10 mg/mL)) was , added and the mixture was centrifuged at 10,000 xg for 5 minutes at 25°C to remove any cell debris which formed a layer at the surface of the tube. The clear solution was transferred to an uLtra centrifuge tube which was , sealed by heat. Supercoiled, nicked, and linear plasmid DNA were separated by ultra centrifugation for 16 hours at 55,000 rpm under vacuum. DNA bands were visualized by UV light. The bands of the supercoiled DNA was aspirated by a syringe (Figure 5). Ethidium bromide was extracted with water saturated butanol, excess CsCl was removed by dialysis against TE buffer. The DNA was , precipitated with 2.5 X volume absolute ethanol and 0.1 X voltune sodium acetate followed by incubation at -20°C for 30 minutes and centrifugation for 15 minutes at 4°C at 10,000 rpm. The pellet was washed by adding 5 mL 70% ethanol followed by centrigugation for 10 minutes at 4 °C at 3,000 xg. The pellet was dried in a speed vac and resuspended in sterile ddHZO. DNA was quantitated as described in the DNA isolation section. Purified DNA was rmr on a 0.8% agarose gel to confirm its purity prior to sequencing. 48 Protein Linear Plasmid DNA Nicked-Circular Plasmid DNA Supercoiled Plasmid DNA Figure 5. Purification of the supercoiled plasmid DNA Supercoiled plasmid DNA was prepared by ultracentrifugation using CsCl - ethedium bromide gradient centrifugation method. Arrows indicate the difl‘erent bands separated as a result of ultracentrifirgation. 4 9 Sequencing Sequencing reactions were performed by cyclic sequencing using a dye primer kit where four fluorescent labeled M13 primers, forward (S'TGTAAAACGACGGCCAGT- 3') and reverse (5'-CAGGAAACAGCTATGACC-3'), were used. These primers are compatible with the pUC19 sequence. Reactions were performed in four different tubes (A, C, G, T) according to the manufacturer’s recommendations (Perkin Elmer, Fostr City, CA). The A and C tubes contained 200 ng of cloned template and 4 uL of the T or C mix. The G or T tubes contained 400 ng of the cloned template and 8 uL of the G or T mix. Each mix contains one dideoxynucleotide, the other three nucleotides and one 5' end fluorescently labeled primer, Tris-HCl (pH 9.0), MgClz, thermal stable phosphatase, and AmpliTaq DNA Polymerase, FS (Perkin Elmer Cetus, Norwalk, CT). Reactions are subjected to 15 cycles of 96°C for 10 seconds, 55°C for 5 seconds, and 70°C for 60 seconds followed by another 15 cycles of 96°C for 10 seconds and 70°C for 60 seconds, using the GeneAMP 9600 thermal cycler (Perkin Elmer Cetus, Norwalk, CT). The contents of the four tubes were combined and the DNA was precipitated with 111 uL of a solution containing 70% ethanol and 0.5 M MgClz. The supernatant was aspirated, the pellet was dried in a vacuum centrifuge and mixed with 4 uL loading dye (5:1 deionized forrnamide : 50 mM EDTA). Samples were denatured at 95°C for 3 minutes in a heating block and loaded on 4.75% polyacrylamide (29:1 acrylamide : N, N’-methylene bisacrylanride) / 8.3 M urea sequencing gel (40 x 25 x 0.04 cm). Sequencing gels were run on a 373A automated sequencer (ABI/ Foster City, CA). Gels were run for 14 hours at 32 watt. Raw data generated by the 373A sequencer was analyzed and edited using the Sequencing Analysis Software (ABI/ Foster City, CA). 5 o Sequencing analysis Sequences generated by Sequencing Analysis software were anlyzed using the Genetic Computer Group (GCG) package (Madison, WI). For the third primer set, the 863 bp of interest was assembled using the Assemble program of GCG which makes new sequence constructs from pieces of existing sequences which were obtained from both directions (forward and reverse, where the reverse sequence was reversed using the reverse program of GCG). The Bestfit program which uses the local homology algorithm of Smith and Waterman (1981) to find the best segment of similarity between two sequences was used to compare all the sequences with each other. The Translate program (which creates a- peptide sequence by translating the specified nucleic acid sequence) of GCG was used to generate the peptide sequence. Pileup program (which creates a multiple sequence alignment from a group of related sequences using progressive, pairwise alignments) (Feng D., and Doolittle R. 1987) of GCG was used to align and compare all the generated peptides to each other as well as to the published sequence (Lopez et al., 1987) Results Polymerase Chain Reaction Different PCR conditions had been tested in order to optimize the conditions to amplify the desired fragments. These trials included changing the annealing temperatures (56 °C, 57 °C, 58 °C, 60 °C, 61 °C, 64.5 °C, 65 °C, 67 °C, 68 °C), using different MgCl2 concentrations (1.5 mM, 1 mM, 0.85 mM, 0.75 mM, 0.65 mM, 0.55 mM, 0.50 mM) and applying touch down PCR (Don et al., 1991). Some PCR conditions resulted in multiple bands in the desired lane along with the desired fragments while some conditions didn’t amplify anything at all. Also, another pair of primers DES (5'- CCACTACTGAACCAACCCCAAGCCC-3)'and DE8 (5'- TCTCAAGGTCCCCAAACCTC—3') was used to amplify the third fragment under different PCR conditions. These primers either amplified non-specifically or would not amplify at all. Primers DE6 and DE7 were selected to amplify this fragment, the final selected conditions are described in the Materials and Methods section. Afier optimizing the conditions, the first primer set (DEl and DE2) successfully amplified the expected 704 bp fragment from the mother, the propositus as well as the normal control and there was no amplification in the negative control (Figure 6). The second set of primers (DE3 and DE4) amplified the expected 664 bp fragment from the mother, the propositus and the control. No products were amplified in the negative 51 52 Fragment l Frag-at 2 Molecular Wt Marker Propositus | Shah» 3054 —> 203‘ —> = 1636 —> W me —> -~ 704 664 Figure 6. PCR amplification of the first and the second fragments. Agarose gel electrophoresis (0.8%) of amplified PCR products using primers set 1 and 2, respectively. The first lane is one kb marker (GIBCO), second and sixth lanes are normal control DNA, third and seventh lanes are Propositus DNA, fourth and eighth lanes are mother DNA, and fifth and ninth lanes are no template blanks. Sizes of the molecular weight bands in base pairs are indicated at the left. 53 control (Figure 6). One Kb molecular weight marker (GIBCO) was used to estimate the sizes of the amplified products, sizes are indicated at the left. Figure 7 shows the amplification results of the third fragment (DE6 and DE7) . The expected size for this fragment was 775, but the amplification gave 2 different sizes. This was likely due to polymorphism resulting in a variable number of tandem repeats in this area of the gene. The control DNA amplified a fragment of 824 bp, whereas the propositus DNA amplified an 863 bp fragment. However, amplification of the mother’s DNA of the same region gave two fragments, one 824 bp, and the other 863 bp. No amplification was detected in the negative control. Heteroduplexing Purified PCR products were used for heteroduplexing. In case of heterozygosity, there will be a chance for the mutant allele to reanneal to the normal one. The area of mismatch will form a bulb~like structure which will alter its migration in the gel compared to normal DNA. Figure 8 shows results for heteroduplexing analysis on 12% polyacrylamide/ 10% urea for normal controls and the mother. As shown in Figure 8 no obvious differences in the migration pattern were detected between the mother and the normal control after staining the gel with silver nitrate. At the time of performing heteroduplexing analysis, no DNA was available from the propositus , so no data was obtained. This analysis was not sufficient to determine the area of the gene to be sequenced so SSCP was performed. 54 Single Stranded Conformation Polymorphism (SSCP) Different combinations of conditions were used to optimize the running conditions for this technique. These included the use of different gel sizes and thickness. Long (40 x20 x 0.025 cm), 12.5% acrylamide gels had been used, but these were very hard to handle during the staining procedure, so smaller gel sizes (20 x 8 x 0.1 cm and 8.6 x 8.5 x 0.1 cm) using 6% acrylamide with and without 5% glycerol were used. Different concentrations of TBE buffer (1X versus 2X TBE ) were also evaluated. In addition the gels were run at room temperature as well as at 4 °C, and different voltages (400 V, 200 V and 180 V) were used. Modifications were also made during the staining step such as 0.1% versus 0.2% silver nitrate and the presence or absence of Na,s,o,.5H,o in the developing solution which was used to reduce the staining of the background compared to the DNA bands. Some of these conditions resulted in dark gels which made it hard to interpret the data, but using 0.1% silver nitrate resulted in better colors. The optimum conditions that are used to run these gels are described in the Materials and Methods section. This technique is highly affected by the fragment size, with optimum fragment size being less than 300 bp (Ainsworth et al., 1994). Purified PCR products generated earlier were too big for this analysis, thus they were digested with different restriction enzymes to generate smaller size fragments. Digestion of the first fragment with Ban 1 resulted in 2 fragments with sizes of 382 bp and 322 bp. Digestion of the second fragment with Ban 1 enzyme resulted in two fragments with sizes of 392 bp and 272 bp (Figure 9). The third fragment was digested with Sst I and Pst I (Gibco, BRL), which should have resulted in 3 fragments of 346 bp, 235 bp and 203 bp. But as shown in 55 Molecular Wt. Marker Central Prep-attic Mather Ila-I Sm In hp 2036 ——>* 1636 ———> a 1018 —> .1 863 824 Figure 7. PCR amplification of the third fiagment. Agarose gel electrophoresis (0.8%), using primers set 3. First lane is one Kb marker (GIBCO), second lane is normal control, third lane is propositus, fourth lane is mother and fifth lane is no template blank. Sizes of the molecular weight marker are indicated at the left. Arrows at the right indicate the two sizes amplified due to the presence of two tandem repeats (824 bp) and three tandem repeats (863 bp). 56 Km Inga-u hon-A nape-I Nor-.1 COI'OI and I rather Figure 8. Heteroduplexing analysis gel. Silver Stained 12% acrylamide! 8% Urea gel which represents heteroduplexing analysis for two normal controls and the mother for the first, second and third fragments of the gene (Primers sets 1, 2 and 3 respectively). 57 3 Firm Wag-lent Second Frag-alt g — _ 5' . . i r i a r i g E 5 E i 5 E at 517 506 396 —> 344 —> <— 392 <— an (— 322 29: —> ‘ 1" no —> Figure 9. Restriction digest of the first and the second PCR fragments. The first and the second PCR fiagment were digested with BanI to generate smaller fragments that can be used for SSCP analysis. One kb marker was loaded in the first lane, control DNA in the second and fifth lanes, propositus DNA in the third and sixth lanes, and Mother DNA in the fourth and seventh lanes. Sizes of the molecular weight marker in base pairs are indicated at the lefi. 58 3 '5 S i I B E -. 5 s 7‘. a 3- s s 8 a s Iineinbp 1636 —> mu —> 517 50‘ :t 39: ——> 344 —-> 293 ——> no ———> 201 -—‘> Figure 10. Restriction digest of the third PCR fragment. The third PCR fragment generated using primers set 3 was digested with PstI and SstI to generate smaller fragment to be used in SSCP analysis was run on 8% acrylamide gel. One kb marker was loaded in the first lane, the control DNA in the second lane, the propositus in the third and the mother which gave an extra band is in the fourth lane. Sizes of the molecular weight marker fragments are indicated at the left. The top two arrows at the right indicate the differences in the restriction digest products between the normal control, the propositus and the mother due to the presence of VNTR in this fragment, where the normal control has 2 copies of the 39 bp tandem repeat, the propositus homozygous for 3 copies, whereas, the mother is heterozygous for the repeat with 2 and 3 copies. 59 Figure 10, the control PCR products gave 3 fragments of 385 bp, 235 bp and 203 bp. The mother PCR products gave four fragments with sizes of 424 bp, 385 bp, 235 bp and 203 bp and the propositus PCR products gave 3 fragments of 424, 235 and 203 bp. For SSCP analysis, the DNA is single stranded. The migration of the single stranded DNA in the gel will depend on its conformation. Single stranded DNA from normal control, heterozygote and mutant homozygote will show different migration patterns on the gel. Figure 11 shows a typical 6% polyacrylamide/ 5% glycerol gel (8.6 x 8.5 x 0.1 cm) which was stained using silver nitrate. No obvious differences between the control, the mother and the propositus were detected within the first (Figure 11, panel A) or second fragment (Figure 11, panel B) of the gene. However, SSCP gels of the third fragment demonstrated some differences in the migration patterns among the control, mother and propositus products as shown in Figure 12. These differences appear to be localized within two subsets (A+B) of the third fiagment. The arrow in panel A represents the absence of this band from the propositus lane. The arrows in panel B indicate two areas of differences among the three lanes Cloning The PCR products from the third fragment for the propositus and the mother were cloned into pUC19. Clones generated using propositus PCR products gave the expected size fi‘agment of 923 bp which represent the insert DNA (863 bp) and 60 bp from the vector. Whereas, clones generated fi'om the mother’s PCR products resulted in two subfamilies of clones. The first was 923 bp similar to that found in the propositus, and the second was approximately 880 bp. The difference is a result of one allele with 3 A B “I“ PCR W“ Second PCR mm 3 i .. i '3 v a i a i ‘3 é a 6 g s 6 a: S a Figure 11. SSCP analysis of the first and second fiagments. SSCP analysis of the first (A) and second (B) PCR fiagments. Silver stained 6% acrylamide/ 5% glycerol gels. No differences were detected between normal control, propositus and the mother. 61 Central i ‘3 Mother Figure 12. SSCP analysis of the third fiagment. Digested PCR products were run on 6% acrylamide / 5% glycerol gel. Comparison between the control in the first lane, the propositus in the second lane and the mother in the third lane showed two main differences in subfragments A and B, respectively. Arrows indicate areas of differences. 62 VNTR and one with 2 VNTR. Five positive clones from the propositus, 2 positive clones from the mother’s first subfamily and 3 clones from the mother’s second subfamily were picked and glycerol stocks were prepared as described in the Materials and Methods. Large amounts of DNA were prepared by the CsCl gradient method and DNA was stored at - 20°C prior to sequencing. Sequencing Initially, manual sequencing of the third fragment from PCR products was performed. An average of 300 bp were obtained from each nm which was not sufficient to accurately determine the entire sequence. Therefore, automated sequencing from PCR products was performed. Initially, data generated using this approach were not conclusive and were hard to manipulate. Based on the PCR and restriction digest results for the mother we assumed that this could be due to the heterozygosity of the tandem repeat in this area of the gene. Thus cloning the PCR products into pUC 19 was determined to be necessary. A total of 5 clones from the propositus and 5 clones for the mother (3 clones containing 2 tandem repeats and 2 clones containing 3 tandem repeats) were used when sequencing the third fragment. Two hundred ng of cloned PCR products were used for cycle sequencing. This was performed using both M13 forward and reverse primers as described in Materials and Methods. Sequencing reactions were run on a 373 A automated sequencer (A31) and sequences were analyzed using sequencing analysis 2.1 software (ABI). Each sequencing reaction gave an average of 650 bp. The raw sequence data generated was analyzed by using the GCG program. Sequences for each clone have 63 been confirmed by sequencing using the M13 forward and reverse primers. The reverse and forward primers sequences were compared to each other. The full length sequences of the third fragment generated using the assemble program was compared to each other as well as to the published sequence (Lopez et a1 1987). In addition, the full length nucleic acid sequence was converted to the corresponding amino acid sequence. The generated peptide sequences were compared to each other and to the published sequence. Sequencing the mother’s DNA clones did not reveal any point mutations, insertions, or deletions and gave a correct Open reading frame when compared to the published sequence. However sequencing the propositus DNA clones showed one point mutation in two different clones where a C to T at position 1531 (Figure 13) changed the amino acid proline 481 into a serine as shown in Figure 14 (numbering according to Lopez et al., 1987). The sequence of the other 3 clones did not show any differences from the consensus sequence nucleic acid or amino acid. Sequences from the mother and the propositus clones confirmed the PCR and the restriction enzymes digestion results regarding the presence of a variable number of tandem repeat (VNTR)in this area. The mother’s DNA was confirmed to be heterozygote for the repeat with one allele having two copies of the 39 bp repeat while the other allele had 3 copies of the same 39 bp. Sequences from clones that have 2 copies and 3 copies of the VNTR showed a 100% match with the published sequence. Sequencing the propositus DNA supported the PCR and the restriction enzyme digestion results regarding the presence of 3 copies of the VNTR in both alleles. 64 Figure 13. Comparison of the third fragment sequences. Ten clones of the third fragment of GPIba gene for the mother and the propositus were sequenced and compared to the GPIba published sequence using the pileup and lineup program. Glycoprotein Ib alpha published sequence is represented by letter A, mother’s clones with 2 copies of the VNTR are designated B and C, whereas the clones with 3 copies of the repeat are designated D, E and F. Five clones for the propositus were designated G through K. Mismatches within the sequences are indicated by an asterisk. WQHCEOHJE’TUOIIIID Wc—JHCL‘O'TIFJOOEDII’ xcqeazzm'ntUCJtiwID 1 AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA AACTCCAAAA 51 CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA CCGTCCCGGA 101 CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC CCGACCACCC TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG TCCACTACTG GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA GCCCGCCCCA CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC CAGAGCCCAC 65 AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACCAACCCC AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA AACATGACCA CTCAGAGCCC CTCAGAGCCC CTCAGAGCCC CTCAGAGCCC CTCAGAGCCC CTCAGAGCCC CTCAGAGCCC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC AAGCCCGACC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC CCCTGGAGCC GCCCCCAGCC GCCCCCAGCC GCCCCCAGCC GCCCCCAGCC GCCCCCAGCC GCCCCCAGCC GCCCCCAGCC 50 ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC ACCTCAGAGC 100 CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CACTCCAAGC CGACCACCCC CGACCACCCC CGACCACCCC CGACCACCCC CGACCACCCC CGACCACCCC CGACCACCCC 4 in arm XQHIOWMOOWH’ WCAHDZGT'UMOOCDII’ XQHEQ'UFJOOUJID‘ GGAGCCCACC GGAGCCCACC GGAGCCCACC GGAGCCCACC GGAGCCCACC GGAGCCCACC GGAGCCCACC 201 CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC CAGAGCCCGC 251 ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA ATCGCCACAA TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG TCAGAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG CCCCAGCCCG GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT GCCCGACCAT 66 CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC CCCCCAGCCC ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG ACCACCCCGG CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT CCTGGTGTCT GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG GACCACCCCG AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC AGCCCACCCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GCCACAAGCC GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT GAGCCCACCT 250 AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC AATCCCGACC 300 TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC TGATCACTCC x xcq»4::mrnnrornnrw WQHIQWMUOUJJ’ 301 AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA AAAAAGCACA 351 CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC CCAAAAAAAC 401 CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC CTCCAAGGGC TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA TTTTTAACTA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA CATCCCTGAA ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG ATTTGGAGAG 67 CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CCACAAAACC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTTGATCAGC CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CTCCAGAAAT CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CGTATCACTC CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT CACCAAAGCT GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC GACCCTTTTC 350 TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA TTAGAATCCA 400 CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG CCGTGGGGTG 450 TCCACCCCGA TCCACCCCGA TCCACCCCGA TCCACCCCGA TCCACCCCGA TCCACCCCGA TCCACTCCGA TCCACTCCGA TCCACCCCGA TCCACCCCGA TCCACCCCGA '* XQHCEO'TJFIUOIIJIP memmmmoowv XQHIOWMUOGB’ 451 CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC CTTTTGCTGC 501 TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC TGCTCTTTGC 551 GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC GTGAAACCAC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCCTCCCCC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC CTCTGTGGTC AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA AGGCCCTGGA 68 TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA TGGGCTTCTA CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCATCCTGC CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA CTCTGGCCAA TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGTCTTGGGT TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG TGCTGAGCTG GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC GGTGCTGCTC 500 CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC CTCTTCTGGC 550 GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT GGTTGGGCAT 600 TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC TGACCACAGC PSCIHZEO'UE‘TUOIIJTP WQHZEQ'TJCUUOEDCU WQHZEQ'UFJUOUJCD' 601 CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC CACACAAACC 651 CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG CCCGGGCCTG 701 AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC AGCCTCTTCC ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG ACACACCTGG GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC GCTGCTCTTC TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG TGTGGGTACG 69 AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG AGCTGCAGAG CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT CTTCGAGGTT GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GCCTAATGGC GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA GGGACGGCAA CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGCTTCCCAC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC CGTGTGGGGC 650 GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC GTGACAGTGC 700 TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC TTTCCGCTCC 750 CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC CTCTAGTGGC PRC—JHCEO'TJEUUOEDID' KQHIOWMOOUJCU XQHCECJ'UE’JUOUJTP 751 AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG AGGAAGGAGG 801 GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG GCACAGTGAG 851 TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT TGGGGACCTT CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CCCTCAGCTC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC CATTAGGTAC 863 GAG GAG GAG GAG GAG GAG GAG GAG GAG GAG GAG 70 TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TGAGTCAGGG TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCTGGCCACA TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG TCGTGGTCAG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG GCCTCTGAGG 800 GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA GACCTGCTGA 850 GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT GTGGGAGGTT 71 Figure 14. Comparison of the third fragment peptide sequences. Ten clones of the third fragment of GPIba gene for the mother and the propositus were sequenced, translated and compared to the GPIba published sequence using the pileup and lineup program. Glycoprotein Iba published sequence is represented by letter A, mother’s clones with 2 copies of the VNTR are designated B and C, whereas the clones with 3 copies of the repeat are designated D, E and F. Five clones for the propositus were designated G through K. Mismatches within the peptide sequences are indicated by an asterisk. WQHIOWMUOUJII’ NQHIQ'TJWUOUJII’ NQHZBO'TJE’JUOIDZU 1 TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP TPKSTTEPTP ...SEPAPSP ...SEPAPSP ...SEPAPSP EPTSEPAFSP EPTSEPAPSP EPTSEPAPSP EPTSEPAPSP EPTSEPAPSP EPTSEPAPSP EPTSEPAPSP 101 KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP KSTFLTTTKP SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE SPTTSEPVPE ...... SEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA TTPEPTSEPA VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT VSLLESTKKT 72 PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PAPNMTTLBP PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PAPNMTTLEP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP PSPTTPEPTP IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL IPELDQPPKL TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT TPSPTTPEPT IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI IPTIATSPTI RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES RGVLQGHLES SEPAPSPTTP SEPAPSPTTP SEPAPSPTTP SEPAPSPTTP SEPAPSPTTP SEPAPSPTTP SEPAPSPTTP 100 LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP LVSATSLITP 150 SRNDPFLHPD SRNDPFLHPD SRNDPFLHPD SRNDPFLHPD SRNDPFLHPD SRNDPFLHPD SRNDPFLHSD SRNDPFLHSD SRNDPFLHPD SRNDPFLHPD SRNDPFLHPD * WQHEQWWUOUJCD' DSQHZEO'TJMOOCDP 7:Qiq:ncinrmtardurw 151 FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY FCCLLPLGFY 201 TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR TQTTHLELQR 251 GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG GRRPSALSQG VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA VLGLFWLLFA GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW GRQVTVPRAW RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS RGQDLLSTVS 73 SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW SVVLILLLSW LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT LLFLRGSLPT 279 IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* IRYSGHSL* VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD VGHVKPQALD FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR FRSSLFLWVR 200 SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA SGQGAALTTA 250 PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA PNGRVGPLVA Discussion Bernard-Soulier syndrome, an autosomal recessive disease, was first described in 1948 (Bernard et al., 1948), since then many cases have been studied and the diagnostic criteria of the disease had been established. The molecular halhnark of the disease is an absent or dysftmctional GPIb-IX-V complex (Nurden et al. 1981, Ware et a1. 1993) which serves as a receptor for the adhesive protein von Willebrand factor (va) (George et al., 1984). The familial cases of BSS that were studied at the molecular level showed that a defect in the genes encoding GPIba and GPIX could alter the conformation and prevent the formation of the complex (Clemetson et al., 1993). In this research a family with BSS was investigated. Consanguinity had been noticed as the parents are cousins (Figure 3). The mother expressed some clinical features characteristic of BSS including giant platelets and mild thrombocytopenia, whereas the propositus has episodically severe thrombocytopenia, bleeding episodes and her platelets were found to lack GPIba as documented by two dimensional gel electrophoresis (data not shown, personal communications with Dr.Mattson). Based on preliminary results, the assumption was made that the mother was a carrier heterozygote for the disease while the propositus was homozygote. Glycoprotein lba gene was amplified into three overlapping fragments using three sets of primers (Table 5). Amplifying the gene into three fragments enabled the 74 75 generation of small fragments suitable for heteroduplexing and single stranded conformation polymorphism analyses. The PCR products of the first and second fragments for the normal control, the propositus and the mother showed no gross rearrangements (insertion or deletion) within that part of the gene. However, results from the PCR products of the third fragment (Figure 7) showed differences in the sizes between the normal control, the mother and the propositus. It was believed that this was due to the presence of VNTR polymorphism that is present at the 3' end of the gene (Figure 2). Based on the calculated sizes of the PCR products, it was concluded that the mother was heterozygote for the 39 bp repeat, one allele having two copies of the repeat and the other allele having three copies of the repeat. However, amplification products from the propositus yielded one fragment of 860 bp which indicated homozygousity for a three copy repeat. Digestion of the third fragment with the restriction enzymes Pst I and Sstl (Figure 10) generated smaller fragments which further confirmed that differences in fragment sizes between the normal control, the mother and the propositus were due to a variable number of tandem repeats. The digest of the mother’s PCR products gave one subfragment that matched the result of the normal control and another subfragment that matched the result of the propositus. This confirmed that the mother is heterozygous for this repeat and the propositus is homozygous. 7 6 Localization of mutation(s) by heteroduplexing and Single Stranded Conformation Polymorphism Since no size differences , which might indicate gross insertion or deletion, had been detected using PCR amplification, heteroduplexing analysis, which helps to identify heterogeneous PCR products was used to screen and localize the mutation within one of the three PCR fragments. When the double stranded DNA is denatured and allowed to reanneal again, there is a chance for the mutated and the normal alleles to anneal. If this occurs, it alters the structure of double stranded DNA (heteroduplexes) retarding its gel migration when compared to the normal DNA (homoduplexes). Based on the clinical data presented it was hypothesized that the mother was heterozygous and therefore heteroduplexing analysis might identify the region of GPIba that has the defect. Unfortunately the data generated from heteroduplexing did not reveal any differences between the mother and the normal control. A possible reason why heteroduplexing did not resolve any differences could be due to the large size of the amplified fragment. Even if heteroduplexes were formed, the differences in migration patterns between homoduplexes and heteroduplexes might not have been resolved by heteroduplexing gels. Because the data generated by heteroduplexing analyses was not useful in detecting differences in any of the fragments, a different approach SSCP, was used. In this technique the DNA is denatured at 95 °C and cooled on ice immediately before it is loaded on the gel. The single stranded DNA takes certain conformations based on its sequence. The mutated sequence is detected as a change in mobility in polyacrylamide gel electrophoresis under non-denaturing conditions (Hayashi 1991). However, SSCP can 77 be affected by different parameters and conditions. Shorter fragments (less than 300 bp) are better suited for detection of mutations. As the detection of mutations depends on the conformation of the single stranded DNA. The physical environment of the gel will affect this detection. This includes temperature, percentage of acrylamide, the ionic strength of the buffer and glycerol concentration in the gel (Spinardi et al., 1991). Rising temperatures during the run may affect the reproducibility of the results. Therefore, efficient cooling is important. Hayashi 1991 , indicated that running a gel at room temperature (20 -26 °C) versus 4 °C can dramatically affect the mobility of the single- stranded DNA. Spinardi et a1. 1994, used a combination of different conditions to detect known mutations. They concluded that an increase in the ionic strength of the running buffer (TBE) over 1X, raising the glycerol concentration over 10% or increasing the polyacrylamide concentration above 6% did not result in any improvement. It has been found that a low concentration of glycerol (5- 10%) in the gel usually improves separation of the mutated single-stranded DNA. This could be due to the weak denaturing action of glycerol on nucleic acid, that it partially opening the folded structure of single-stranded DNA so that more surface area of the molecule is exposed. Therefore there is more chance for the acrylamide fibers to detect locally confined structural differences caused by mutation (Hayashi 1991). To obtain accurate data using SSCP and, since PCR products were bigger than the optimum 300 bp, digestion with restriction enzymes was used to obtain smaller fragments in an attempt to optimize the conditions for this technique. Digestion of the PCR products of the first and second fragments with Ban 1 gave fragments ranging in size from 272 - 392 bp which are suitable for SSCP analysis. Also restriction digest of PCR 78 products of the third fragment with Pstl and SstI gave fragments ranging in size from 203 bp - 424 bp which are also good for SSCP analysis. The completely digested PCR products were loaded on 6% polyacrylamide/ 5% glycerol gel which gave good resolution of the smaller fragments. The results that were obtained for PCR products for the first and second fragments did not show any differences in migration patterns between the control the mother and the propositus (Figure 11). These results might indicate that there is no mutation present in the first or second fragment of the gene. On the other hand, this may indicate that a mutation was not detected in this analysis because the conditions used were not the right combinations to detect mutations in this sequence. Results of SSCP analysis of the PCR products for the third fragment showed some significance differences in the migration patterns between the normal control the mother and the propositus. These differences have been localized to two subfragments within the 3' end of the gene. The first subfragment (A) in Figure 12, could be due to the presence of different numbers of the 39 bp repeats within this part of the gene where as the second subfragment (B) in Figure 12, is most probably due to a mutation which changed the migration pattern for the propositus in this region. These preliminary results led us to assume that the defect was within the 3' end of the gene. This part of the gene encodes the region of GPIba that associates with GPIbB, and has the transmembrane and cytoplasmic domain which helps in stabilizing the platelet plasma membrane by its association with the cytoskeleton through an actin binding protein. Although complete digestion of PCR products was examined before running the SSCP gels, any incomplete digestion that was not picked up by ethedium bromide staining will be detected by silver 79 staining because it is more sensitive. This could give false positive results and lead to misinterpretation of the results. To confirm that a mutation is indeed located within the 3' end of the gene and to identify the mutation we cloned and sequenced the 3' end of the gene for the mother and the propositus. Sequencing and Sequence Analysis Since the 3' end of GPIba gene is GC rich and has multiple copies of the 39 bp repeat, sequencing this region directly from PCR products was very difficult and false termination was detected within this fragment which made it impossible to interpret the data. Because of this the PCR product of this fragment was cloned into the pUC19 vector for sequencing. pUC19 is 2.8 Kb circular plasmid with a high copy number, it has the arnpicillin resistance gene which allows for selection. After transformation only colonies that contain the plasmid will grow on a media with ampicillin. This vector has a multiple cloning site within the Lac Z gene which is responsible for producing the N terminus (alpha peptide) of the enzyme B-galactosidase, where as the host strain provide the C- terrninus ( w- peptide) of this enzyme. X-gal is a colorless chemical substrate for B- galactosidase, when it is cleaved by the enzyme it gives a blue product. The insertion of a DNA fragment in the multiple cloning site will disrupt the alpha region of the lac-Z gene and it will not produce the alpha peptide of B-galactosidase. As a result the enzyme is not produced, X-gal is not cleaved and no blue color is developed. This characteristic of pUC19 allows for screening where white colonies indicate the presence of an insert. The purified PCR products of the third fragment were successfully cloned into pUC19 vector and purified using the CsCl gradient centrifugation method (Sambrook etal., 1992). 80 Since PCR primers can not differentially amplify maternal or paternal alleles, the amplified product represents a mixture of DNA from both the maternally and paternally inherited DNA. In an effort to determine the mutation responsible for the disease and the origin of that mutation, multiple clones have been sequenced. Five clones for the propositus DNA were successfully sequenced, two of which had a single point mutation, a C to a T, that changed the amino acid proline 481 to a serine. Sequences from the other F.— three clones demonstrated no mutations and the generated sequences completely matched _. the consensus sequence (Lopez et al., 1987). For the mother a total of 5 clones were i chosen for sequencing. Three of the 5 sequenced clones, as expected, had 3 copies of the i- 39 bp repeat and the remaining two had 2 copies of the repeat. No mutations were detected in any of the clones and the sequence had a 100% match with the published sequence (Lopez et al., 1987). The sequence data proved that the mother is heterozygous for the repeat, but since the propositus is homozygote for this repeat, we are mainly concerned about the copy of the gene that has 3 copies of the VNTR. Bernard-Soulier syndrome is inherited in an autosomal recessive manner. An affected individual should have inherited two defective alleles, one from each parent. From the heterozygous mother we expected to identify a mutation in one of the alleles, specifically, the one with the 3 copies of VNTR because as was demonstrated this is the allele that the propositus inherited from her mother. For this part of the gene, no mutation was identified in the mother’s DNA sequence indicating that the defect may reside in another part of the gene. The mutation that was identified in the propositus DNA sequence was not identified in the mother’s sequence which means that she inherited the allele that has the 81 mutation from her father. The propositus is homozygous, because she has all the clinical symptoms of the disease including bleeding episodes, so the mutation identified could not be the only cause of the disease. There are rare cases that seem to be due to a double heterozygous defect in GPIba chain (Ware et al., 1990), but other cases of BSS are produced by double heterozygote point mutation in the gene encoding GPIX (Clemetson et a1. 1993; Wright et al. 1993). Therefore, there is a possibility that another mutation a. could be present in the maternal allele that, when coupled with the paternally inherited . defect results in BSS.. The sequenced part of the maternal gene (3' end) did not 1 demonstrated any mutations. Therefore, the presence of a mutation in the 5' end of the 5 gene is possible. However this does not exclude that the propositus could be homozygous for another mutation (at the 5' end of the gene) which she inherited from both parents. As mentioned earlier, a defect in one of the genes encoding for any member of GP-lb-IX complex could prevent the assembly and cell surface expression of the complex (Clemetson et al., 1993). Although GPIba was not detected by two dimensional gel electrophoresis and showed low reactivity when treated with monoclonal antibodies, this does not exclude that a defect could be in one of the other genes of GPIbp, GPIX or GPV. On the other hand, a defect in the promoter for one of these genes could also lead to complete lack of expression of the designated glycoprotein without there being any problem in the translated region. This in turn could affect the formation and surface expression of the whole complex. In conclusion, the mother was found to be heterozygote for the 39 bp repeat, while the propositus is homozygote for this repeat. The point mutation identified here, Pro 481 to Ser, could not be the sole cause of the disease. Therefore, other parts of the gene and the GPIb—IX-V complex should be investigated. Future Research Since a point mutation had been identified, allele specific PCR can be used to type other family members for this specific mutation. This can be approached by H synthesizing two oligonucleotide primers, one primer (X) has this point mutation at its 3' end (5' CCAGAAATGACCCTTTTCTCCACT 3'), the other primer (Y) is a normal one re- (5' CCAGAAATGACCCTTITCTCCACC 3'). These two primers can be used with DE6 y primer (Table 5) to amplify a 440 bp fragments in two separate PCR reactions. DNA from a normal individual will amplify a fragment in the PCR that has primers Y and DE6 only, DNA from a person homozygote for this mutation will amplify a fragment in the PCR with primers X and DE6, while DNA from a person heterozygote for this mutation will amplify fragments from both PCR reactions using primers X / DE6 and Y / DE6. Since only the 3' end of GPIba gene was sequenced and a point mutation that could explain the propositus clinical symptoms of the disease was not found, it will be necessary to sequence the remaining part of the gene that is represented by the first two fragments. In case a mutation is identified, allele specific PCR can be used to screen other family members for this mutation using the approach just explained above. Also screening other members of the family can be approached by restriction fragment length polymorphism analysis (RF LP) if the mutation destroys a known restriction site or if it creates a new one. This approach does not apply for the mutation identified here because 82 83 this mutation did not destroy or create a restriction site. If no mutation (s) has been identified within the open reading frame of the gene, the 5' end which has the promoter sequence has to be investigated for mutations that might inactivate the promoter. After identifying the transcription starting site either by 81 nuclease mapping or by primer extension the promoter region can be sequenced using dye terminator sequencing reagent and the sequence analyzed by GCG and Genepro program to identify sequences within that region which might represent binding site for genes as transcription factors. After defining the promoter region, oligonucleotide primers should be designed i to specifically amplify the potential promoter, which will be cloned upstream of a reporter gene (either CAT or Luciferase) in such a way that the expression of the gene is driven by the promoter of interest. The promoter region fiom the mother, the propositus and a normal control should be cloned. All recombinants can be transfected into a eukaryotic cell line (Hela cells or NIH3 cells) by any transfection method. Cells can be harvested and protein purified and assayed for enzyme activity. Enzymatic activity for the mother, propositus and normal control can be compared and if comparable results are obtained, the possibilities that the defect causing the disease is present in the other components of the complex, GPIbB, GPIX or GPV should be investigated. Western blot analysis can be used to check for the presence or absence of these glycoproteins and , the molecular defect can be investigated in a manner similar to what was done in this project. LIST OF REFERENCES List of references Ainsworth J., D. I. Rodenhiser (1994). MW (A.J. Harwood Totowa, NJ) 31: pp 205-210 Alagille D., F. Josso, J. L. Binet, M. L. Blin (1964). La dystrophie thrombocytaire hemorragipare : Discussion nosologique. Nouv Rev Fr Hematol 4: 755 Andrews R., J. Fox (1991). Interaction of purified actin binding protein with the platelet membrane glycoprotein Ib-IX complex. J Biol Chem 266: 7144- 7147 Bernard J., J. P. Soulier (1948). Sur une nouvelle variete de dystrophie thrombocytaire hemorragipare congenitale. Sem Hop Paris 24: 3217 Bernard J. (1983). History of congenital hemorrhagic thrombocytopathic distrophy. Blood Cells 9: 179-193 Bizzaro N., G. Dianese (1988). Neonatal alloimmune amegakaryocytosis. Case report. Vox Sang 54: 112-114 Brass L.F., C. C. Shaller, E. J. Belmonte (1987). Inositpl 1,4,5, triphosphate induced granule secretion in platelets. Evidence that the activation of phospholipase C mediated by platelet thromboxane receptors involves 3 guanine nucleotide binding protein- dependent mechanism distinct from that of thrombin. J Clin Invest 79: 1269 Brown A. B. (1984). W (Lea & Febiger Philadelphia, PA) pp 179 - 186. Charo I. F., R. D. Feinman, T. C. Detwiler (1977). Interrelations of platelet aggregation and secretion. J Clin Invest 60: 866 Clemetson K. J. and J. Clemetson (1993). Molecular pathology of Bernard-Soulier Syndrome, Platelet-type von Willebrand’s disease and Glanzrnan’s thrombosthania. Beiter Infusionsther 31 : 168-173 Clemetson J. M., P. A. Kyrle, B. Brener, K. J. Clemetson (1994). Variant Bemard- Soulier syndrome associated with a homozygous mutation in the leucine-rich domain of 84 85 glycoprotein IX. Blood 84 (4): 1124-1131 Clemetson,K., J. McGregor, E. James, M. Dechavanne, and E. Luscher (1982). Characterization of the platelet Membrane glycoprotein abnormalities in Bernard Soulier Syndrome and comparison with normal by surface-labeling techniques and high- resolution two dimensional gel electrophoresis. J .Clin.Invet. 70:304-311 Coller B S. (1978). The effects of ristocetin and von Willebrand factor on platelet electrophoretic mobility. J Clin Invest 61: 1168-1175 Coller B. S. (1982). Effects of tertiary amine local anasthetics on von Willebrand factor- dependent platelet function: alteration of membrane reactivity and degradation of GPIb by a calcium-dependent protease(s). Blood 60: 731-743 Coller B. S., H. Gralnick (1977). Studies on the mechanism of ristocetin-induced platelet agglutination. Effects of structural modification of ristocetin and vancomycin. J Clin Invest 60: 302-312 Coller B. S. (1986). Activation affects access to the platelet receptor for adhesive glycoproteins. J Cell Biol 103: 451 Comveau M- D» G- A- Fritsma (1988). Haemestasrsandlhmmhosrsmthefihmeai Lalmatent. (J .B.Lippincott co.,Philadelphia, PA), pp 8, 14-17, 23-28 Cruz M. A., E. Peterson, S. M. Turci (1992). Functional analysis of a recombinant glycoprotein Ib A polypeptide which inhibits von Willebrand factor binding to the platelet glycoprotein Ib-IX complex and to Collagen. J Biol Chem 267: 1303-1309 Cullum C., D. P. Cooney, S. L. Schrier (1967). Familial Thrombocytopenic thrombocytopathy. Br J Haematol 13: 147 Don R. H., P. T. Cox, B. J. Wainwright, K. Baker, J. S. Mattick (1991). ‘Touchdown’ PCR to circumvent spurious priming during gene amplification. Nucleic Acids Research 19: 4008 Du, X., L. Beutler, C. Ruan, P. A. Castaldi, M. C. Berndt (1987). Glycoprotein lb and glycoprotein IX are fully complexed in the intact platelet membrane. Blood 69: 1524-1 527 Federici A., R. Bader, S. Pagani (1989). Binding of von Willebrand factor to glycoprotein lb and IIb-IIIa complex: affinity is related to multimeric size. Br J Haematol 73: 93-99 Feng D-F., R. F. Doolittle (1987). Progressive sequence alignmentas a prerequisite to correct phylogenetic trees. Journal of molecular evolution 25: 23 5-260 86 Finch, C., L. Miller, R. Handin, V. Lyle (1990). Evidence that an abnormality in the Glycoprotein Iba gene is not the cause of abnormal platelet function in a family with classic Bernard-Soulier disease. Blood 75:2357-2362. Fontento J. D., N. Tjandra, D. Bu, C. Ho, R. C. Montelaro, O. J. Finn (1993). Biophysical characterization of one-, two-, and three-tandem repeats on human mucin (muc-l) protein core. Cancer Res. 53: 5386-5394 fox J., M. Berndt (1989). Cyclic AMP-dependent phosphorylation of glycoprotein Ib inhibits collagen induced polymerization of actin in platelets. J Biol Chem 264: 9520- 9526 Fox J., L. Aggerbeck, M. Berndt (1988). Structure of the Glycoprotein Ib-IX complex from platelet membranes. J Biol Chem 263: 4882- 4890 Fox J., K. Boyles, M. Berndt (1988). Identification of a membrane skeleton in platelets. J Cell Biol 106: 1525 miner. van. a J“ urn- George J. N., A. T. Nurden, D. R. Phillips (1984). Molecular defects in interaction of platelets with the vessel wall. N Eng J Med 311: 1084 Gerald, R., T. Church, B. McMullen, and S. Williams (1990). Human platelet glycoprotein V: A surface leucine-rich glycoprotein related to adhesion. Biochemical and Biophysical Research Communications 170(1):153-161. Gerrard J. M., R. C. Carroll (1981). Stimulation of protein phosphorylation by arachidonic acid and endoperoxide analog. Prostaglandins 22: 81 Golden A., S. J. Brugge, S. J. Shattil (1990). Role of platelet glycoprotein lib-Illa in agonist-induced tyrosine phosphorylation of platelet proteins. J Cell Biol 111: 3117 Greensburg S., F. Divirgilio, T. H. Steinberg, S. C. Silverstein (1988). Extracellular nucleotides mediate Ca+2 fluxes in J 774 macrophages by two distinct mechanisms. J Biol Chem 263: 10337 Grottum K. A., N. O. Solum (1964). Congenital thrombocytopenia with giant platelets : A defect in the platelet membrane. Br J Haematol 16: 277-290 Grottum K. A., N. O. Solum (1969). Congenital thrombocytopenia with giant platelets: a defect in the platelet membrane. Br J Haematol 16: 277 Handin R. I., D. D. Wagner (1989). Molecular and cellular biology of von Willebrand factor. Prog Hemost Thromb 9: 233 87 Hartwig J., M. DeSisto (1991). The cytoskeleton of the resting human blood platelets: structure of the membrane skeleton and its attachment to actin filaments. J Cell Biol 112: 407-425 Hayashi, K. (1991). PCR-SSCP: A simple and sensitive method for detection of mutations in the genomic DNA. PCR Methods and Applications 1:34-38. He R., D. M. Reid, C. E. Jones, N. R. Shulman (1995). Extracellular epitopes of platelet glycoprotein Ib-a reactive with serum antibodies from patients with chronic idiopathic thrombocytopenic purpura. Blood 86: 3789-3 796 Hession C., C. Osborn, D. Goff (1990). Endothelial leukocyte adhesion molecule 1: Direct expression, clonong and firnctional interactions. Proc Natl Acad Sci 87: 1673 Hickey M. J., G. J. Roth (1993). Characterization of the gene encoding human platelet glycoprotein IX . J Biol Chem 268: 3438- 3443 Hickey, J. , S. Williams, and G. Roth (1989). Human platelet glycoprotein IX: An adhesive prototype of leucine-rich glycoproteins flank-center-flank structures. Proct. Natl. Acad. Sci (USA). 86:6773-6777. Hoffman R., E. Benz,D. J. Shattil, B. Furie,H. J. Cohen (1991). Wig W (Churchill Livingstone Inc. New York, NY), pp 1167 Ishida F., K. Furihata, K. Ishida, H. Kodaira, K. S.Han, L. Da-Zhuang, K. Kitano, K. Kiyosawa (1996). The largest isoforrn of platelet membrane glycoprotein Ib-a is commonly distributed in Eastern Asian populations. Thrombosis and Haemostasis 76(2): 245-247 Jenkins C., K. J. Clemetson, E. F. Luscher (1979). Studies on the mechanism of ristocetin-induced platelet agglutination: binding of ristocetin to platelets. J Lab Clin Med 93: 220-231 Judson P. A., D. J. Anstee, J. R. Clamp (1982). Isolation andcharacterization of the major oligosaccharide of human platelet membrane glycoprotein GPIb. Biochem J 205 :81-90 Kertz, K. W. Cullen, and V. Hedden (1994). Cycle Sequencing. PCR Methods and Applications 3:107-112. Korrel S. A. M., K. J. Clemetson, H. Van Halbeek. (1984). Structural studies on the O-linked carbohydrate chains of human platelet glycocalacin. Eur J Biochem 140: 571- 576 Kroll M. H., A. l. Schafer (1989). Biochemical mechanisms of platelet activation. 88 Blood 74: 1181 Kuijpers R., W. Ouwehand, P. Bleeker, D. Christie, A. Kr. Von dem Borne (1992). Localization of the platelet-specific HPA-2 (Ko) alloantigens on the N-terrninal globular fragment of platelet glycoprotein Ib alpha. Blood 79(1): 283-288 Kunishima S., H. Miura, H. Fukutani, H. Yoshida, K. Osumi, S. Kobayashi, R. Ohno, T. Naoe (1994). Bernard-Soulier syndrome Kagoshima: Ser 444—. stop mutation of glycoprotein (GP) Iba resulting in circulating truncated GPIb-a and surface expression of GPIb-B and GPIX. Blood 84(10): 3356-3362 Lanza, F., M. Morales, C. de—La-Salle, J. P. Gazenave, K. J. Clemetson, T. Shimimura, and D. R. Phillip (1993). Cloning and characterization of the gene encoding the human platelet glycoprotein V. A member of the leucine-rich glycoprotein fanrily cleaved during thrombin-induced platelet activation. JBC 268(28):20801-7 Lee, J. (1991). Alternative dideoxy sequencing of double stranded DNA by cyclic reaction using Taq polymerase. DNA and Cell Biology 10:67-73. Leren, T., K. Solberg, O.K. Rodningen, L. Ose, S. Tonstand, K. Berg (1993). .Evaluation of running conditions for SSCP analysis: Application of SSCP for detection of point mutation in the LDL receptor gene. PCR Methods and Applications 3: 1 59-162. Li C., S. E. Martin, G. J. Roth (1995). The genetic defect in two well-studied cases of Bernard-Soulier syndrome:a point mutation in the fifth leucine-rich repeat of platelet glycoprotein GPIba Blood 86(10): 3805-3814. Lopes, J. A. (1994). The platelet glycoprotein lb-IX complex. Blood Coagulation and Fibrinolysis. 5:97-119 Lopez, J. , D. Chung, K. Fujikawa, F. Hagen, T. Papayannopoulou, and G. Roth. (1987). Cloning of the alpha chain of human platelet glycoprotein lb: A transmembrane protein with homology to leucine -rich alpha2-glycoprotein. Proc. Nat. Acad. Sci (USA) 84:5615-5619 Lopez J. A., D. W. Chung, K. Fujkawa, F. S. Hagen, E. W. Davie, G. J. Roth (1988 b). Proc. Natl. Acad. Sci. USA 85: 2135-2139. Lopez A. J., B. Leung, C. Reynoldst, Q. Chester, and J. Fox (1992). Efficient plasma membrane expression of a functional platelet glycoprotein Ib-IX complex requires the presence of its three subunits. JBC 267( 18) 12851-12859 Lopez J. A., D. W. Chung, K. Fujikawa (1988). The alpha and beta chains of human platelet glycoprotein lb are both membrane proteins containing a leucine-rich amino acid sequence. Pro Natl Acad Sci (USA) 85: 2135-2139 89 Madoff M. A., S. Ebbe, M. Baldini (1964). Sialic acid of human blood platelets. J Clin Invest 43: 870-877 Maoke J., N. Turner, N. Stathopoulos (1988). Shear induced platelet aggregation can be mediated by vWF released from platelets, as well as by exogenous large or unusually large vWF multirners, requires adenosine diphosphate, and is resistant to aspirin. Blood 71: 1366- 1374 Marco L. D., M. Mazzucato, F. Fabris, D. Roia, P. Coser, A. Girolami (1990). Variant Bernard-Soulier syndrome type bolzano a congenital bleeding disorder due to a structural and functional abnormality of the platelet glycoprotein Ib-IX complex. J.Clin.Invest. 86: 25-30 Mattson J.C., D. M. Peterson, S. McCarron, N. Stathopoulos (1984). The Bemard- Soulier platelets: II. A comparative study of changes in platelet morphology and cytoskeletal architecture following contact activation. Scanning Electron Microscopy IV: 1941-1950 McGill M., G. Jamieson, J. Drouin, M. Cho, G. Rock (1984). Morphometric analysis of platelets in Bernard-Soulier Syndrome: size and configuration in patients and carriers. Thromb Haemost 52: 37 Miller, J., V. Lyle, and D. Cunningham (1992). Mutation of Leucine -57to phenylalanine in a platelet glycoprotein Iba leucine tandem repeat occuring in patients with an autosomal dominant variant of Bernard Soulier disease. Blood 79:439-446. Modderman, P. L. Admiral, A. Sonnenberg, and A. Von dem Borne (1992). Glycoproteins V and Ib-IX form a non covalent complex in the platelet membrane. JBC 267(1),364-369. Moroi, M., S. M. Jung, and N. Yoshida (1984). Genetic polymorphism of platelet glycoprotein Ib. Blood 64:622-629. Najean Y., N. Aradaillou, J. P. Cean (1963). Survival of radiochromium -1abeled platelets in thrombocytopenias. Blood 22: 718 Neame P. J., H. U. Choi, L. C. Rosenberg (1989). The primary structure of the core protein of the small, leucine -rich proteoglycan (PG 1) from bovine articular cartilage. J Biol Chem 264: 8653-8661 Nurden T. A. (1995). Polymorphism of human platelet membrane glycoproteins: structure and clinical significance. Thrombosis and Haemostasis 74(1): 345-351 Nurden A. T., J. P. Cean (1974). An abnormal platelet glycoprotein pattern in three cases of Glanzrnan’s thrombasthania. Br J Haematol 28: 233 9O Nurden A. T., J. P. Caen (1975). Specific roles for platelet surface glycoproteins in platelet function. Nature (Lond) 255: 720 Nurden, A. , D. Didry, and J-P. Rosa (1983). Molecular Defects of Platelets in Bernard Soulier Syndrome. Blood Cells 9:333-358. Nurden, A. , D. Dupuis, T. Kunicki, and J. Caen (1981). Analysis of Glycoprotein and Protein Composition of Bernard- Soulier Platelets by Single and Two Dimensional Sodium Dodecyl Sulfate-Polyacrylarnide Gel Electrophoresis. J. Clin. Invest. 67:1431- 1440. Okita, JR. ,D. Pidard, P. J. Newman, R. R. Montgomry, T. J. Kuinki (1985). On the association of glygoprotein lb and actin-binding protein in human platelets. J Cell biol. 100:317-321 Okita J., D. Pidard, P. J. Newman (1985). On the association of glycoprotein lb and actin- binding protein in human platelet. J Cell Biol 100: 317 Peerschke E. I. (1985). The platelet fibrinogen receptor. Semin Hematol 22: 241 Peerschke E. 1., M. B. Zucker, R. A. Grant, J. J. Egan, M. M. Johnson (1980). Cerrelation between fibrinogen binding to human platelets and platelet aggregability. Blood 55: 841 Peterson D. M., N. Stathopoulos, T. Giorgio (1987). Shear induced platelet aggregation requires von Willebrand factror and platelet membrane glycoprotein lb and IIb- Illa. Blood 69: 625-628 Phillips D.R., I. F. Charo, L. V. Parise, L. A. Fitzagerald (1988). The platelet membrane Glycoprotein lib-Illa complex. Blood 71: 831 Ramofi‘m C D Forbes (1984) MW 121W, (Grune & Stratton New York, NY) pp 1- 20 Sae-Tung, G., J-F. Dong, and J. A. Lopez.(1996). Biosynthetic defect in platelet glycoprotein IX mutants associated with Bernard-Soulier Syndrome. Blood 87 (4):1361- 1367. Saji H., E. Maruya, H. Fujii (1989). New platelet antigen, Sib', involved in platelet transfusion refractoriness in a Japanese man. Vox Sang 56: 283-287 Salle C. D. L., M-J. Baas, F. Lanza, A. Schwartz, D. Hanau (1995). A three-base deletion removing a leucine residue in a leucine residue in a leucine-rich repeat of a platelet glycoprotein lb or associated with a variant of Bernard-Soulier syndrome. British Journal of Haematology 89: 386-396 91 Sambrook J., T. Maniatis, E. Fritish, (1992). MW Manual. (Cold Spring Harbor Laboratory, New York, NY) pp 280 Shattil S.J., L. F. Brass (1987). Induction of fibrinogen receptor on human platelets by intracellular mediators. J Biol Chem 262: 992. Shogren R., T. A. Gerken, N. Jentoft (1989). Role of glycosylation on the conformation and chain dimensions of O-linked glycoproteins: light-scattering studies of ovine submaxillary mucin. Biochemistry 28: 5525-553 Simsek S., P. Noris, M. Lozano, M. Pico, A. E. G. Kr.Von Dem Borne, A. Ribera, D.Gallardo (1994c). Cy5209 Ser mutation in the platelet membrane glycoprotein lba gene is associated with Bernard-Soulier syndrome. British Journal of Haematology 88: 839-844 Simsek, S., L. G. Admiral, P. W. Modderman, C. E. Van der Schoot, and A. E. G. Kr. von der Borne (1994a). Identification of a homozygous single base pair deletion in the gene coding for the human platelet glycoprotein [be causing Bernard-Soulier syndrome. Thromb Haemost 72 (3):444-449. Simsek S., P. M. M. Bleeker, C. E. Van der Schoot, A. E. G. Kr.Von dem Borne (1994b). Association of a variable number of tandem repeats (VNTR) in glycoprotein Ibo: and HPA-2 alloantigens. Thrombosis and Haemostasis 72 (5): 757-761. Smith, A. (1991). DNA sequence analysis by primed synthesis. Methods in Enzymology 65: 560-581. Smith T. F., M. S. Waterman (1983). Comparison of bio-sequences. Advances in applied mathematics 2: 482-489. Snyder F. (1989). Biochemistry of platelet activating factor: a turique class of biologically active phospholipids. Proc Soc Exp Biol Med 190: 125 Sorrentino, R. ,C. Iannicola ,S. Costanzi, A. Chersi, and R. T031 (1991). Detection of complex alleles by direct analysis of DNA heteroduplexes. lmmunogenetics 33: 118-123. Tsuji T., S. Tsunehisa, Y. Watanabe (1983). The carbohydrate moiety of human platelet glycocalacin. The structure of the major SER/ THR-linked chain. J Biol Chem 285: 6335- 6339 Vicente V., R. A. Houghten, Z. M. Ruggeri (1990). Identification of a site in the A chain of platelet glycoprotein Ib that participate in von Willebrand factor binding. J Biol Chem 265: 274-280 92 Ware J., S. R. Russell, P. Marchese, M. Murata, M. Mazzucato, L. D. Marco (1993a). Point mutation in a leucine-rich repeat of platelet glycoprotein lb-a resulting in the Bernard-Soulier syndrome. J.Clin.Invest. 92: 1213-1220 Ware, J., SR. Russell, V. Vincente, RE.Scharf, A. Tomer, R. McMillan, Z. M. Rugerri (1990). Nonsense mutation in the glycoprotein Iba coding sequence associated with Bernard-Soulier Syndrome. Proc. Natl. Acad. Sci (USA) 87:2026-2030 Ware, J., S. Russel, P. Marchese, M. Mazzucato, L. De Marco (1993b). Point Mutation in a Leucine-rich Repeat of Platelet GP lba resulting in the Bernard Soulier Syndrome. J. Clin. Invest. 92: 1213-1218. weiss H. J., V. Turrito (1979). Prostacyclin (prostaglandin 12, PG 12) inhibits platelet adhesion and thrombus formation on subendothelium. Blood 53: 244- 250 Weiss H. J., V. T. Turrito, H. R. Baumgartuer (1978). Effect of shear rate on platelet interaction with subendothelirun in citrated and native blood. I. shear rate-dependent decrease of adhesion in von Willebrand’s disease and the Bernard-Soulier syndrome. J Lab Clin Med 92: 750- 764 weiss H., V. Turitto, H. Baumgartuer (1986). Platelet adhesion and thrombus formation on subendothelium in platelets deficient in glycoproteins lib-Illa, lb and storage granules. Blood 67: 322-327 Wencel-Drake J. D., B. Dahlback, M. H. Ginsberg (1986). Ultrastructural localization of coagulation factor V in human platelets. Blood 68: 244 Wenger R., A. Wicki, N. Kieffer (1989) the 5' flanking region and chromosomal localization of the gene encoding human platelet membrane glycoprotein Ib alpha gene. 85: 517-524 Wenger, R. , N. Kieffer, A. Wicki, and K. Clemetson (1988). Structure of human blood platelet membrane glycoprotein lba gene. Biochemical and Biophysical Research Communications. 156 (1):,389-395. White J., S. Burris, D. Hasegawa, M. Johnson (1984). Micropipette aspiration of human blood platelets, a defect in Bernard-Soulier Syndrome. Blood 63: 1249 Wicki, A. , and K. Clemetson.(1985) Structure and function of platelet membrane glycoprotein Ib and V effects of leukocyte elastase and other proteins on platelets response to von Willebrand factor and thrombin. Eur. J. Biochrn. 153: 1-11 Wrigt, S. K. Michaelides, D. J. D. Johnson, N. C. West, and E. G. D. Tuddenham. (1993) Double heterozygosity for mutations in the platelet glycoprotein IX gene in three siblings with Bernard-Soulier Syndrome. Blood 81: (9):2339-2347