Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular analysis of the prothrombin gene in two patients Duke, Leslea Marie 1993

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

Item Metadata

Download

Media
831-ubc_1994-0085.pdf [ 1.46MB ]
Metadata
JSON: 831-1.0094716.json
JSON-LD: 831-1.0094716-ld.json
RDF/XML (Pretty): 831-1.0094716-rdf.xml
RDF/JSON: 831-1.0094716-rdf.json
Turtle: 831-1.0094716-turtle.txt
N-Triples: 831-1.0094716-rdf-ntriples.txt
Original Record: 831-1.0094716-source.json
Full Text
831-1.0094716-fulltext.txt
Citation
831-1.0094716.ris

Full Text

MOLECULAR ANALYSIS OF THE PROTHROMBIN GENE IN TWOPATIENTSbyLeslea Marie DukeB. Sc (Hon.), Queen’s University, 1991.A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FOR THE DEGREE OFMasters of ScienceinTHE FACULTY OF GRADUATE STUDIESDepartment of Biochemistry and Molecular BiologyWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIADecember 1993.© Leslea Marie Duke, 1993.In presenting this thesis in partial fulfillment of the requirements for anadvanced degree at the University of British Columbia, I agree that the libraryshall make it freely available for reference and study. I further agree thatpermission for extensive copying of this thesis for scholarly purposes may begranted by the Head of my Department or by his or her representatives. It isunderstood that copying or publication of this thesis for financial gain shallnot be allowed without my written permission.Department of Biochemistry and Molecular BiologyThe University of British Columbia2174 Health Sciences MallVancouver, CanadaV6T 1Z3Date: December 1993.2I,..11ABSTRACTA deficiency of prothrombin (Fil) is an extremely rare bleedingdisorder. Two patients were studied with a severe form of this disorderwhich is known as Hypoprothrombinemia. Each patient was identified asbeing the product of a consanguineous union, and therefore a singlehomozygous mutation would be expected. To characterize the deficiency, aclinical analysis was performed. Fil antigen and activity levels were providedfor each patient. Fli-Vancouver was assessed at 3% Fil activity and 3% Filantigen levels. Fll-Utrecht exhibited similar levels of 3% for both activityand antigen. The objective of this study was to analyze the HI gene in eachpatient, and to identify the molecular basis for the disease.Genomic DNA was provided for each patient and the exons of theprothrombin gene were amplified specifically by using the polymerase chainreaction. Primers were designed to include the intron/exon splice junctionsand approximately 100 bp 5’ to exon 1 in the amplification reactions. Tofacilitate directional cloning, restriction sites were incorporated into theprimers. In this way, each region could then be analyzed by DNA sequencing.The sequence analyzed in each patient was then compared to the reportedsequence of wild-type Fil gene (Degan et al., 1987). Several base changes wereobserved in both patients, and all were found to be homozygous.A single inserted ‘A’ was found in the region 5’ to exon 1, atnucleotide position -54 , in both Fil-Vancouver and Fll-Utrecht (numbering isaccording to the sequence of Degen et al., 1987). This insertion was laterfound in the wild-type sequence reported by Bancroft et al., (1990). It isevident therefore that this alteration is not the cause of the deficiency;however, it does provide the opportunity to focus on the promoter region as111a potential region for mutation. At nucleotide position 461, a single ‘T’ wasdeleted from the splice region 5’ to exon 2. This change was not expected tointerfere with the proper splicing of FIT, as the deleted base occurs in a seriesof 3 ‘T’ residues. The function of this base could be easily adapted by either ofthe flanking bases. This change was observed in both patients. A silentpolymorphism resulting from the point mutation of the codon CTA -> CTGwas found in exon 2 of Fil-Vancouver and FII-Utrecht. The leucine residueat codon 56 would be unaffected by this mutation. A polymorphism wasfound in the splice region 3’ to exon 6, which results from the point mutationof G -> A, at nucleotide position 4272. This mutation occurs in both patients,and was also reported by Iwahana et al., (1992). The most significant mutationfound was the result of a three bp deletion of the in-frame codon AAG, atnucleotide position 7485-7487. This mutation deletes a single lysine residue atthe codon position 301. This mutation was found in Fil-Vancouver, and wasfound to be homozygous. This deletion occurs in the activation region of FIT,in what would eventually be the A-chain of thrombin (Fila). This mutationis expected to have an effect on the 3-dimensional structure of thepolypeptide. A silent polymorphism was identified in exon 10 of bothpatients. This point mutation resulted from the substitution of ACA -> ACC,at nucleotide position 8903. This threonine at position 388 was unaffected bythis change.The molecular basis of the deficiency in FIT-Vancouver appears to bethe result of a single deleted lysine residue. This mutation is expected to haveglobal effects on the proper folding of the protein during synthesis, resultingin its degradation. A mutation resulting in the synthesis of an aberrantlyfolded protein is quite common among disorders where there is little or nogeneration of protein. The basis of the deficiency in FTT-Utrecht remainselusive. The observed changes observed in the FIT gene were not expected tohave a significant effect on the generation of the protein. The intron/exonjunctions and splice branch points all remain intact, therefore, the disorder isnot expected to be due to an error in splicing. It is possible that the deficiencyis in fact, due to a defect in transcription. It is common to find mutations inthe upstream promoter and enhancer regions which affect transcription suchthat minimal amounts of protein are generated. Further studies would benecessary to test the fidelity of the Fil promoter in this patient.ivVTABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS vLIST OF FIGURES viiiLIST OF TABLES ixLIST OF ABBREVIATIONS xACKNOWLEDGEMENTS xiiINTRODUCTION 11. BLOOD COAGULATION 1A. Overview of Hemostasis 2B. Blood Clotting Cascade 22. STRUCTURAL ORGANIZATION OF THE DOMAINS OF THEVITAMIN K DEPENDENT BLOOD CLOTHNG PROTEINS 63. PROTHROMBIN 10A. Gene Structure and cDNA 10B. Protein Structure 13C. Gene Structure Related to Protein Structureand Function 141. Signal and Propeptide Regions 142. Aromatic Stack 153. Kringle Domains 154. Activation Domain 155. Protease Domain 164. PROTHROMBIN ACTIVATION AND THE PROTHROMBINASECOMPLEX 165. VITAMIN K DEPENDENCE OF PROTHROMBIN 206. DEFICIENCIES OF PROTHROMBIN 21A. Hypoprothrombinemia 21B. Dysprothrombinemia 22C. Combined Hypo-Dysprothrombinemia 227. SCREENING TESTS FOR DISORDERS IN HEMOSTASIS 228. PREVIOUSLY REPORTED PROTHROMBIN DEFICIENCIES 24A. Hypoprothrombinemias 27B. Dysprothrombinemias 27viC. Compound Heterozygotes for Dysprothrombinemia andHypoprothrombinemia 319. OBJECTIVES OF THIS STUDY 33MATERIALS AND METHODS 341. REAGENTS 342. STRAINS, VECTORS AND MEDIA 34A. Vectors 34B. Bacterial Strains 35C. Media 353. OLIGONUCLEOTIDES 354. GENOMIC DNA 375. ISOLATION OF DNA 37A. Plasmid DNA 37B. Isolation of Single Stranded DNA 386. GEL ELECTROPHORESIS 38A. Agarose Gel Electrophoresis 38B. Denaturing Polyacrylamide Gels 397. POLYMERASE CHAIN REACTION 398. DNA SUBCLONING 41A. Purification of DNA Fragments 41B. Restriction Digestion and Blunt-ending of Purified PCRProducts 42C. Vectors Used for Ligations 42D. Ligation of PCR Fragments into Bluescript Vectors 439. TRANSFORMATION OF RECOMBINANTBLUESCRIPT INTO E. COLI 4310. DNA SEQUENCE ANALYSIS 44A. Single Stranded DNA Sequencing 44B. Double Stranded DNA Sequencing 46RESULTS AND DISCUSSION 471. SUMMARY OF CLINICAL STATUS OF PATIENTS 472. PCR AMPLIFICATION OF THE PROTHROMBINGENE IN EACH PATIENT 473. DNA SEQUENCE ANALYSIS OF THE PROTHROMBIN GENE INHYPOPROTHROMBINEMIC PATIENTS 48A. Sequence Changes Observed in FII-Utrecht 49B. Sequence Changes Observed in Fil-Vancouver 49DISCUSSION 561. NATURALLY OCCURING MUTATIONS FOR STRUCTURALAND FUNCTIONAL ANALYSIS OF PROTEINS 562. SEQUENCE ANALYSIS OF PROTHROMBINMUTATIONS 56A. Inserted ‘A’ in Putative Promoter Region 57B. A Deletion of a ‘T’ in the Splice Region 5’ to Exon 2 57C. Leucine Polymorphism in Exon 2 58D. Base Substitution in Splice Region 3’ to Exon 6.58E. Lysine Deletion in Exon 9 of Fil-Vancouver 59F. Threonine Polymorphism in Exon 10 593. PROPOSED EFFECTS OF OBSERVED MUTATIONS ONPROTHROMBIN 59A. Fli-Vancouver 59B. FII-Utrecht 604. FUTURE STUDIES 62REFERENCES 65viiviiiLIST OF FIGURESFigure 1: Diagramatic Representation of the Blood Clotting Cascade 3Figure 2: Common Features of Blood Clotting Proteins 7Figure 3: Organization of the Prothrombin Gene, mRNA, and Protein 11Figure 4: The Prothrombinase Complex 17Figure 5: Schematic Diagram of Products of Prothrombin Activation by FXaorFila 19Figure 6: PCR Strategy for the Amplification of the Exons of Prothrombin....40Figure 7: Sequence Changes Involving an Inserted ‘A’ in the PutativePromoter Region and a Deleted ‘T’ in the Splice Region 5’ toExon2 52Figure 8: Autoradiograms of Polymorphisms in Exon 2 and the Splice Region3’toExon6 53Figure 9: Deletion of the Codon ‘AAG’ in Exon 9 of FJJ-Vancouver 54Figure 10: Sequence Analysis of Threonine Polymorphism in Exon 10 55LIST OF TABLESTable 1: Summary of Reported Prothrombin Deficiencies 26Table 2: PCR Primers and Cloning Strategy 36Table 3: Oligonucleotide Primers For DNA Sequence Analysis 45Table 4: Summary of Sequence Analysis on Prothrombin DeficientPatients 50ixxLIST OF ABBREVIATIONSA AdenosineAmp ampicillinATP adenosine triphosphatebis N, N’-methylenebisacrylamidebME b-mercaptoethanolbp(s) base pair(s)C cytosineCi Curiecpm counts per minuteDEAE diethylaminoethyldH2O distilled waterDMSO dimethyl sulfoxideDNA deoxyribonucleic aciddNTP Deoxynucleotide triphosphateE. coli Escherichia coliEDTA ethylenediaminetetraacetic acidEGF epidermal growth factorEtBr ethidium bromideFli prothrombin (Factor II)G guanosineGla ‘y-carboxyl glutamic acidIPTG isopropyl-13-D-thiogalactopyranosidekDa kilo Daltonskbp(s) kilo base pair(s)LB Luria brothxint(s) nucleotide(s)OD optical densityPCR polymerase chain reactionPEG polyethylene glycolRNa se ribonucleaseSDS sodium dodecyl sulfateT thymineTEMED N,N,N’,N’-tetramethylethylenediamjneUV ultra-violetX-gal 5-bromo-4chloro-3-indo1y1--D-ga1actopyranosidexliACKNOWLEDGEMENTSMy most sincere thanks to Ross MacGillivray. Ross, you opened thedoor to a tremendous opportunity, and for that I am truly grateful. To all themembers of the MacGillivray lab, thanks for all your advice, assistance, andsupport. Each and everyone of you contributed to the completion of thisthesis. Thanks to Dr. David Banfield, and Tina Umelas for theircontributions to this project. My most sincere thanks to Alexis Maxwell.Alexis, thank you for reaching out and guiding not only my academicendeavors, but for making home seem not so far away. To Hung Vo, myfriend and lab-mate, although you have many talents, it is your generosityand compassion that I will remember the most. Thanks to Willie Pewarchukfor your endless knowledge of computers. To David Leggett, your advice andencouragement was only exceeded by your friendship, all of which I am trulygrateful. To my friends, I thank you for your diversions. Special thanks goesto Deb Sauve. Most of all, to my parents, I am grateful for your support andencouragement.1INTRODUCTION1. BLOOD COAGULATIONA. Overview of HemostasisHemostasis is the arrest of hemorrhage through the interplay andregulation of four systems: endothelium, platelets, coagulation, andfibrinolysis (Colman et al., 1987). Normal endothelium inhibits coagulationand adhesion of platelets on its surface by the expression of heparin-likesubstances, and the presence of protein complexes such as thrombinthrombomodulin. Disruption of the vascular endothelium, due to bloodvessel injury, precipitates activation of a number of systems designed tocontain blood flow and repair the damaged tissue. Exposure of thesubendothelium causes the accumulation and subsequent activation ofplatelets at the site of injury. Platelets form the initial line of defense againstvascular leakage forming a platelet plug. Once activated, platelets exposereceptors on their surface, such as the receptor to which fibrinogen (Fgn)binds. In larger vessels, simultaneous activation of the blood clotting cascadeparallels formation of the platelet plug. Although not completelyunderstood, the blood clotting cascade is a tightly regulated system ofhemostasis (Davie et al., 1992). This cascade involves a sequential series ofreactions involving the activation, by limited proteolysis, of plasma proteinzymogens (Davie et al., 1979). In the final step of this cascade, prothrombin(FIl) is converted to thrombin. The serine protease thrombin is then able toconvert the plasma soluble fibrinogen (Fgn) to fibrin. Fibrin forms a meshnetwork with activated platelets at the site of injury to halt bleeding. Prior torepair of the damaged endothelium is the removal of the blood clot by aprocess known as fibrinolysis. Plasminogen is a key component offibrinolysis. Both Fgn and fibrin maintain receptors for plasminogen and2plasmin. Plasminogen becomes integrated into the clot during its formation.Plasminogen is activated to plasmin which can then cleave fibrin polymersfor solublization of the clot. Fibrin dissolution products then impede furtherclot formation in two ways: fibrin products can bind to the Fgn sites onthrombin, preventing activation of more Fgn; in addition, the split productscan be inefficiently incorporated into the clot and destabilize it by impedingfibrin polymerization (Colman et aL, 1987). Coordination and regulation ofthese four components of hemostasis maintains the integrity of thecirculatory system.B. Blood Clotting CascadeThe blood clotting cascade involves the systematic activation andamplification of a series of reactions resulting in the formation of aninsoluble blood clot (Davie et al., 1991). This process is achieved by one of twoseparate but converging pathways as shown in Figure 1 (Davie et a!., 1991,MacFarlane, 1964.; Davie and Ratnoff, 1964). The extrinsic pathway of bloodcoagulation is initiated by the expression of tissue factor (TF). TF is amembrane protein which is exposed when there is a blood vessel injury. TFacts as a cofactor in FVII activation; however, the mechanism involved hasyet to be identified. In the presence of Ca2, the TF-FVIIa complex activatesFIX and FX. FXa assembles on the membrane surface through the cofactorFVa, and in the presence of Ca2. This assembly of factors is known as theprothrombinase complex, and its function is to activate prothrombin (Fil) tothrombin (Fila). FUa catalyses the conversion of fibrinogen (Fgn) to fibrin(Fn) by the proteolytic release of fibrinopeptides A and B. These Fnmonomers spontaneously polymerize to form a network. Fila also activatesthe transglutaminase FXffl to FXffla. The fibrin network is stabilized by3Figure 1. Diagramatic Representation of the Blood Clotting Cascade.The process of blood coagulation can be initiated by either the intrinsicor extrinsic paths. Both paths converge with the activation of FX. Themajor reactions of this complex and highly regulated system aredepicted above. Blood clotting factors (F) are identified by Romannumerals. Activated forms of proteins can be distinguished frominactive forms by the subscript ‘a’. Ca represents calcium ions. TFstands for tissue factor, and PL stands for phospholipids. Thecomponents of the prothrombinase complex are indicated by the box.4INTRINSIC PATHWAYFXI FXIaI ++CaFIX ‘FIXaFXEXTRINSIC PATHWAYFVIIa 4 Fy11ITFÔ++CaFXPROTHROMBINASECOMPLEX FXIII-Flla IIIFXIIIaFVIIIaCaPLFilFgn—* Fn5crosslinking of the y -chains of Fn by this transglutaminase. This generates astrong clot which is resistant to the sheer forces imposed within the vascularsystem.The intrinsic pathway originates with all of its components in theblood. Initially, FXII was thought to be the first protein to be activated in thisprocess; however, it is now thought that the primary role of FXII may be inthe inflammatory response. FM has been shown to be activated by thrombinin the presence of a negatively charged surface (Naito and Fujikawa). FXII canthen be activated by FXIa to FXIIa by simultaneous binding with highmolecular weight kininogen (HMWK) to a negatively charged surface. FXIIacan upregulate the activation of FM. FXIa activates FIX to FIXa, in thepresence of calcium ions (Ca2). In a membrane bound complex with FVIIIa,FIXa causes the activation of FX. At this point in the cascade the twopathways converge into the common path as described above. The intrinsicsystem is probably not physiologically relevant for initiation of bloodcoagulation; however, it may be involved in maintenance of the cascade onceit is activated by the extrinsic system.The process of blood coagulation is a complex, highly regulated system.Activation of each of the individual components is achieved by limitedproteolysis. Each individual component is regulated by a number of proteinsin a feedback mechanism. In addition, the entire pathway is also globallyregulated by inhibitors, anticoagulants, and by the blood clotting factorsthemselves. This representation, therefore, includes only the major reactionsinvolved in the formation of a blood clot.62. STRUCTURAL ORGANIZATION OF THE DOMAINS OF THE VITAMINK DEPENDENT BLOOD CLOHING PROTEINSProthrombin, as well as many of the blood clotting proteins, belongs toa family of proteins known as the serine proteases. This group of proteins ischaracterized by the catalytic triad which serves as the active site. This triad iscomposed of a serine, a histidine, and an aspartic acid. The serine proteaseshydrolyze peptide bonds specific for each individual enzyme. Trypsin is acommon serine protease that shares a high degree of sequence identity withthe blood clotting serine proteases. Thrombin has the same specificity forsmall substrates as does trypsin, but they differ in physiological substrates.This uniqueness of substrate specificity seems to be dependent on themolecular surface surrounding the active site (Furie and Furie, 1988).Overall, the blood clotting serine proteases are twice the size of trypsin butthe catalytic domains are quite similar (Furie and Furie, 1988). Differencesamong the enzymes are manifested predominantly by substrate specificity.The common structural features of the blood clotting proteins can beattributed to evolutionary events in the diversification of this family ofproteins. Common gene organization among the members suggests acommon ancestral gene diversified by evolutionary processes such as exonswapping, gene duplications, rearrangements, and homologous crossovers.Further diversification could evolve through individual mutations. Theblood clotting proteins represent a family of proteins with diversifiedfunctions, but common structural elements (Furie and Furie, 1988).Analysis of the domains of several of the blood clotting proteinsreveals extensive regions of sequence identity. Figure 2 depicts the common7Figure 2. Common Features of Blood Clotting Proteins.Many of the structural domains of the blood clotting proteins are conserved.The similarity can often be extrapolated to function. It is believed that regionsof identity evolved from a single ancestral gene. The structural regions areindicated in the key, and their function described in the text. The structuraldiagrams for prothrombin (Degen et al., 1983), FIX (Kurachi and Davie, 1982;Choo et al., 1982), FX (Leytus et al., 1984; Fung et al., 1985), FVII (Hagen et al.,1986), and Protein C (Foster and Davie, 1984; Beckmann et al., 1985, Long et al.,1984) taken from Furie and Furie (1988).8IIIIIIIIIIIIII///////Z PROTHROMBIN (FIT)..syYY1’ Q292.IIIIIIIIIIIIII’////////, FACTOR IX...YyYF1 —Q292 IIIIIIIBIIHI FACTOR XY Y V LQ2_ IIIIIIIIIIIIIIf/////// FACTOR VII9292 IIIIIIIIBIIB PROTEIN C (PC)I.SIGNALPEPTIDEPROPEPTIDEV V Y GLA DOMAIN11111111 ACTIVATIONPEPTIDEUKRINGLEAROMATIC STACK CATALYTICDOMAINEGF DOMAIN9structural domains of many of the blood clotting proteins. The C-terminaldomain is shared by all of the blood clotting factors shown. This regioncontains the conserved sequence Gly-Asp-Ser-Gly-Gly, which includes theactive site serine (Furie and Furie, 1988). This area shares amino acidsequence identity with the pancreatic serine proteases such as trypsin andchymotrypsin (MacGillivray et al., 1988). Prothrombin, FX, FIX, and ProteinC (PC), are all synthesized as preproproteins. All of these proteins haveextensive sequence identity in their signal and propeptide regions. Theamino termini of these proteins are characterized by the presence of Glaresidues (y-carboxyglutamic acid residues) that are involved in calcium-dependent membrane binding. Prothrombin contains two regions calledkringles (Magnusson et al., 1975). This sequence involves approximately 80amino acids, and three invariant disulfide bridges clustered in the center ofthe kringle. This structure is also found in FXII, plasminogen, andplasminogen activator. The exact function of these kringles has not beenprecisely determined; however, they are thought to be involved in protein-protein interactions and macromolecular assembly (Furie and Furie, 1988).The kringle structure is not present in FVII, FIX, FX, or PC, but in its place isan EGF-like domain. This region is composed of 53 amino acids and threedisulfide bonds, and functions to bind specific cell-surface receptors. Acommon feature to all of Fil, FVII, FX, FIX, and PC, is the aromatic amino acidstack. This region contains a conserved sequence of amino acids witharomatic side chains which interact in a ring cluster. This region is thoughtto play a role in membrane binding (Furie and Furie, 1988). The activationregions also share structural similarities in all of the described proteins. Thissubclass of proteins, therefore represents a case of evolutionary diversificationof common genes to carry out unique functions.103. PROTHROMBINA. Gene Structure and cDNAFigure 3 shows a diagramatic representation of the gene, the mRNA,and the Fli protein. The gene for prothrombin (Fil) has been localized tochromosome 11 at position llpll-q12 (Royle et al., 1987), and is 21kb in length(Furie et al., 1988; Degan et al., 1987). The entire prothrombin gene has beensequenced (Degan et al., 1987). The gene is composed of fourteen exons,separated by thirteen introns (Degan et al., 1987). Exons range in size from 25to 315 bp, whereas introns range from 84 to 9447 bp. Approximately ninetypercent of the gene involves intervening sequence. Of this, forty percent iscomposed of repetitive DNA sequences in a clustered fashion; 30 copies ofAlu and 2 of the Kpn family of repeats (Degan et al., 1987; Furie et al., 1988).The sequence of the splice junctions at the 5’ and 3’ of intervening sequencesis consistent with that found in other eukaryotic genes, in that they follow theGT-AG rule of Breathnach et al., (1978). The one exception to this is thepresence of a GC rather than a GT at the 5’ end of the intron spanning exons12 and 13 (Degan et al., 1987). The 5’ consensus sequences of TATA(TATAAA ) within 50 bp and CCAAT sequences (5’GGCCAATCT3’) within200 bp of the transcription start site which normally direct the properinitiation of transcription by RNA polymerase are absent in prothrombin(Chow et al., 1991, Bancroft et al., 1992). A lack of a TATA sequence maycorrespond to the heterogeneity of start sites (Bancroft et al., 1992). Majortranscription initiation sites have been mapped to 23 and 36 bp upstream ofthe initiator codon by Chow et a!., (1991); whereas Bancroft et al., (1992)11Figure 3. Organization of the Prothrombin Gene, mRNA, and Protein.A. The FIT gene is depicted in the 5’ to 3’ direction (Degen and Davie, 1987).The relative size of the exons are indicated by boxes, separated by interveningsequences (introns are not drawn to scale). The Fil gene is 20 kb in length.Refer to text for specific information on the FIT gene. B. The mRNA forhuman FIT is 2 kb in length. A leader sequence, shown in black, is followedby the entire coding region for FIT, followed by a 97 bp non-coding region, anda 27 bp poly(A) tract (Degen et al., 1983). C. The entire FIT protein structure isshown (Degen et al., 1983). Structural regions are labelled as shown, and arefarther described in the text.12A.B.C.KEY:1234567891011121314I I LH-AAA5 LEADER CODING REGIONI SIGNALPEPTIDE•IIIIIIIIIIy y y GLA DOMAINIIIIIU ACTIVATIONPEPTIDEHKRINGLEAROMATIC STACK CATALY[ICDOMAIN3’UTR13identified the sites between -37 and -31. A weak promoter has therefore beenproposed immediately upstream of the transcription start site (-1 to -435), andan enhancer has been identified between nucleotides -860 and -940. (Chow etal., 1991). Bancroft et a!., (1992) identified a positive cis-acting regulatoryregion between nucleotides -2969 to -797 bp. A region highly similar to theHNF-1 protein binding site has been identified at -888 to -876 (Bancroft et a!.,1992).The cDNA for prothrombin is 2005 bp in length (Furie and Furie., 1988;MacGihivray and Davie, 1984; Degan et al., 1983). The cDNA encodes a leadersequence of 36 amino acids, 579 amino acids of the mature protein, a stopcodon, an intervening sequence of 97 bases, and a 27 base poly(A) tail (Deganet al., 1983). The longest 5’ noncoding region preceding the initiatormethionine is 27 bp by MacGillivray et al. (1986); whereas Degan et al., (1989)reported it to be 26 bp.B. Protein StructurePlasma prothrombin is synthesized exclusively in the liver as a singlechain glycoprotein, of 72 000 MW. The entire protein sequence of Fil hasbeen described (Magnusson et al., 1975; Butkowski et al., 1977; Walz et al.,1977). The nascent protein is composed of three regions; a hydrophobic preregion of up to 34 amino acid residues, a basic pro-region of up to 9 residues,and the coding region of 579 residues. The pre-region is cleaved by a signalpeptidase on the lumenal side of the endoplasmic reticulum (ER). The proregion is partially split in the smooth ER, and completely in the Golgi(Jackson et al., 1987). The 579 residue mature protein contains 8%carbohydrate in 3 chains from asparagine residues 78, 100, and 373 (Degan and14Davie, 1987). Plasma levels of prothrombin circulate at about 100 jig/mi(Furie and Furie, 1988).C. Gene Structure Related to Protein Structure and FunctionThe exons of the prothrombin gene are organized such that eachencodes a particular structural domain of the protein. This structural domaincan then be related to the function of that region of the molecule, with respectto the overall function of the protein.1. Signal and Propeptide RegionsFIl is synthesized with an additional 43 residues at the amino terminusencoded by exon 1. Two arginine residues precede the N-terminal alaninefound in the mature protein. Typically, an Arg-Aia bond is not cleaved by asignal peptidase. This region represents a signal and a propeptide region. Byanalogy with FIX, cleavage by the signal peptidase is at approximately -18(Banfield et al., 1993). The pro-region is cleaved by a proprotein peptidase toreveal the mature protein. There is a large degree of sequence identity ofthese regions in all of the vitamin K dependent clotting proteins. Thepropeptide region contains the y-carboxylation recognition site (y-CRS). Tengiutamic acid residues, encoded by GAG, are post-translationally modified togamma-carboxylated glutamic acid residues (Gla) by a membrane boundvitamin K dependent carboxylase (Suttie et al., 1993). This Gla domain isthought to be involved in calcium dependent membrane binding of theprotein. Two or three Gla residues bind a single calcium ion, which forms anon-covalent intramolecular bridge between the polypeptide backbone (Furieand Furie, 1988). Membrane binding is then facilitated by direct binding15through the calcium ions (Mann et al., 1982), or by exposure of a secondarybinding site due to the tertiary folding of the protein (Borowski et al., 1986).2. Aromatic StackExon 3 encodes the aromatic amino acid stack. This region has aconserved sequence of Phe-Trp-X-X-Tyr. This region contains a series ofhydrophobic side chains which interact in a ring cluster. This region is foundon the surface of the protein and is thought to be involved in membranebinding (Park and Tulinsky, 1986; Furie and Furie, 1988).3. Kringle DomainsIn prothrombin, exons 4 through 7 encode 2 kringle domains(Magnusson et al.,1975). These structures are composed of approximately 80amino acids. Three disulfide bonds are generated from 6 invariant cysteineresidues (Magnusson et al.,1975). The sulfur atoms of the disulfide bonds areclustered in the center of the kringle (Park and Tulinsky, 1986) which isresponsible for its unique tertiary structure. The exact function of kringleshas not yet been positively elucidated, but it is speculated that they areinvolved in membrane binding as it has been shown that the second kringleof prothrombin binds FVa (Esmon and Jackson, 1974.). Kringle structureshave also been observed in tissue plasminogen activator, urokinase,plasminogen, and FXII (Furie and Furie, 1988).4. Activation DomainExons 8 and 9 encode the activation domain of Fil, which includes thetwo sites of FXa cleavage. FXa cleaves Fil at Arg273-Thr, and Arg322-Ile16(Roberts et al., 1981). This region also contains one of the two Fila cleavagesites, at Arg286-Thr. This region then becomes the A chain of thrombin.5. Protease DomainThis region is encoded by exons 10 through 14. The amino acids thatencode this catalytic triad are encoded on separate exons in FIT. Histidine ison exon 10, aspartate is on 11, and serine is found on exon 13 (Furie andFurie, 1988). Once activated by FXa, the protease region, comprising the activesite catalytic triad and the substrate binding pocket, resides in the B-chain ofFila. Crystallographic studies of chymotrypsin and trypsin revealed not onlya high degree of sequence identity between the proteins, but identity whichextended to the three dimensional structure and function of the polypeptidebackbone. This catalytic region contains the residues like Ser-195, His 57, Asp102, and the salt bridge like lie 16 to Asp 194 of chymotrypsin (numberingbased on chymotrypsinogen) (Magnusson et al., 1975). Models of Fila haveshown a substrate binding pocket with Asp522 at the base, to be similar toAsp189 in trypsin, and Ser189 of chymotrypsin (Magnusson et al., 1975).Entrance to the pocket is via two glycine residues (551 and 561), which issimilar to chymotrypsin. It has been speculated that the different molecularsurfaces surrounding the enzyme active sites may be responsible for therestricted substrate specificity of Fila compared to trypsin and chymotrypsin(Furie and Furie, 1988).3. PROTHROMBIN ACTIVATION AND THE PROTHROMBINASECOMPLEXA schematic diagram of the prothrombinase complex is shown inFigure 4. The activation of the proenzyme FIT involves the assembly of four17FliaFigure 4. The Prothrombinase ComplexSchematic diagram outlining the components of the prothrombinasecomplex on a phospholipid surface. Clotting factors (F) are labelled byRoman numerals. Gla residues, involved in phospholipid binding, areshown on FXa and FIT. Activation of FIT, results in the release of thethrombin portion of the molecule (Fila). The Pro region of Fli, including theGla residues remains associated with the membrane after cleavage.FXaRI++CadgI/Ill, i/I/li j! I/Il/i18additional components at the site of injury. The amplification processinvolved in the activation of various components of the blood clottingcascade, as well as the specific concentration of the required components to alocalized area, facilitates the rapid activation of Fil. Fli is present in lowamounts in plasma, but localization to the membrane allows concentration ofthis essential protein. The components required for Fli activation are FVa,FXa, calcium, and phospholipids. The sum of these components comprisethe prothrombinase complex which assembles on the activated plateletsurface. FV is activated by thrombin to FVa; this protein acts as a surfacereceptor on platelets and phospholipids, to which FXa and Fli attach.Phospholipids provide a membrane surface on which the complex canassemble. FXa and Fil bind to the membrane through their y-carboxyglutamicacid residues (Gla). This interaction is mediated by the presence of calciumions. Calcium is thought to either directly bind the proteins through the Glaresidues, or to cause a membrane binding site to be exposed through aconformational change in the protein.Prothrombin is synthesized as a single chain plasma glycoprotein.Activation of this proenzyme involves a number of site specific cleavages. Adiagram of the process of RI activation is shown in Figure 5. Based onproteolytic fragmentation, Fil can be divided into 3 domains (Mann et al.981). The prothrombin fragment 1 region is composed of residues 1-155.The prothrombin fragment 2 region spans residues 156-273. Theprethrombin region, which is the immediate precursor of thrombin, spansresidues 274-581. FXa cleaves FIT sequentially at two specific sites. The firstcleavage occurs at Arg273-Thr, releasing fragment 1.2 (pro region comprisingresidues 1-273 ) from the prethrombin 2 (Figure 5A). A second FXa cleavageoccurs at Arg322-Ile, splitting the prethrombin 2 region into the A (274-322)19I. rss1Prothrombin•‘I. IFragment 1.2I IFFragment 1.2Prethrombin 2Pretbrombin 2rrs1Prethrombin 1C. s—sAThrombinFigure 5. Schematic Diagram of Products of Prothrombin Activation by FXaor Fila.Prothrombin has two sites that are sequentially cleaved by FXa to yieldFragment 1.2 and the two chain thrombin molecule. Cleavage at the first siteyields Fragment 1.2 and Prethrombin 2 (A). Cleavage at the second site aloneyields Meizothrombin (B). Thrombin has two additional cleavage sites inprothrombin. Independent of FXa, Flia cleaves Fil to yield Fragment 1,Fragment 2, and Prethrombin 2. A second cleavage by Flia removes 13residues from the N-terminus of Prethrombin 2, or what will finally be the Achain of thrombin.FXa\Ia+ r4s1maA.rSS1Fragment 1 Prethrombin 1III I r55iORB.I-Fragment 1I II IFragment 1 Fragment 2MeizothrombinB20and B (323-581) chains of x-thrombin. In the absence of the first cleavage, thisproduct is called Meizothrombin (Figure 5B). Two additional cleavages of Filare recognized by thrombin. The first occurs between Arg155-Ser. Thiscleavage generates prothrombin fragment 1 and prethrombin 1. This bond isalso deaved in fragment 1.2, giving rise to fragment 1 and fragment 2. Asecond thrombin cleavage occurs at Arg256-Thr in both the A-chain ofthrombin, and in prethrombin 2. This cleavage shortens the N-terminus ofthe A chain by 13 residues. The cleavage of both FXa sites and the second Filasite generates thrombm (Figure 5C).5. VITAMIN K DEPENDENCE OF PROTHROMBINLike many of the blood clotting factors, synthesis of functionalprothrombin is dependent on the presence of vitamin K. Vitamin K isrequired by a membrane bound carboxylase, which is involved in the post-translational modification of Gla residues (Nelsestuen et al., 1974). Tenglutamic acid residues of prothrombin are modified in this way. Thiscarboxylation is important for the calcium binding properties of the molecule.The administration of vitamin K agonists such as Dicumarol leads toabnormal HI synthesis, as well as abnormal synthesis of other vitamin Kdependent proteins FIX and FX (Ferlund et al., 1975). This abnormal Fil is nolonger able to bind calcium, and subsequently is unable to participate incoagulation, although abnormal FII can be activated by non-physiologicalmeans to generate fully functional Fila (Ferlund et al., 1975). By allowingassociation with the membrane during macromolecular assembly, thisdependence on the Gla region serves to increase the specificity of Filactivation, and limits the reactions by dissociation of the active portion of themolecule from the membrane bound region.216. DEFICIENCIES OF PROTHROMBINProthrombin deficiencies can occur in two distinct forms:Hypoprothrombinemia and Dysprothrombinemia. Both of these disordersare extremely rare in the population. The two forms of the deficiency exhibitthe same clinical symptoms, but to varying degrees. Affected individualsexhibit bleeding whose severity varies with the level and activity ofprothrombin. Bleeding manifestations of bruising , menorrhagia,postpartum hemorrhage, and hemorrhage following surgery or trauma arecommon among affected patients. Diagnosis is made on the basis of analysisof the family history, functional Fil activity levels, and immunological levels.Acquired FIl deficiency can therefore be distinguished from congenital Fildeficiency . To eliminate the possibilities of combined defects, it is necessaryto assay for deficiencies of other coagulation factors. The presence of avitamin K deficiency, warfarin ingestion, or liver disease would result in aprothrombin deficiency, but other coagulation factor levels would be low aswell. In vitamin K deficiency and warfarin ingestion, the active levels of Filwould be decreased. In addition, immunological levels would beapproximately 50%, as the decarboxylated form of Fil is less stable (Roberts etal., 1981). Prothrombin deficiencies are generally treated by replacementtherapy with plasma, or prothrombin complex concentrates.A. HypoprothrombinemiaHypoprothrombinemia is classified as a true deficiency ofprothrombin. It is inherited in an autosomal recessive fashion.22Heterozygotes are asymptomatic with functional Fil levels of approximately50%. Homozygotes are symptomatic, with functional Fil activity levels of 2-25%. In hypoprothrombinemia, functional activity levels correlate withimmunological levels of the protein.B. DysprothrombinemiaDysprothrombinemia involves the synthesis of an abnormal Filmolecule which maintains antigenicity, but lacks normal activity. Theabnormal molecule is present, but has decreased or absent biological function.Again, heterozygotes are asymptomatic, with immunological levels of 100%,and activity levels of approximately 50%. Homozygotes have activity levelsof less than 50%.C. Combined Hypo-DysprothrombinemiaThe occurrence of a combined deficiency in a population is extremelyrare. This condition occurs when one allele for prothrombin encodes adysfunctional molecule resulting in a dysprothrombinemia, and the otherallele encodes a hypoprothrombinemic condition, or absence of a molecule.Immunological levels approach 50% due to the presence of the defective FITmolecule, whereas functional activity levels are found to be minimal.7. SCREENING TESTS FOR DISORDERS IN HEMOSTASISTo identify the source of a disorder in hemostasis, it is necessary toestablish in which of the four areas the deficiency or dysfunction lies. This isaccomplished by establishing the integrity of each of the four systems;vascular maintenance, platelet function, coagulation, and fibrinolysis(Colman et al., 1987). The bleeding time remains the most fundamental of23these tests. This measures the interaction of platelets with the vessel wall.Along with establishing the integrity of the components of the coagulationsystem, this tests reflects the presence or absence of sufficient platelets, and theintegrity of the vascular wall with which they interact. A direct platelet countis definitive for a deficiency or dysfunction of platelets.A general estimate of problems within the coagulation system can beestablished by a number of assays. The activated partial thromboplastin time(APTT) establishes the integrity of the intrinsic pathway of coagulation. Thistest is a measure of the contact activation of FXII, FXI, FIX, and FVffl, as agroup. Less specifically this tests measures FV, Fil, FX, and Fgn. Excluded isFVII which is part of the extrinsic pathway. This tests gives prolonged timesrelative to normal in the presence of dysfunctional or deficient coagulationfactors. The prothrombin time (PT) measures the integrity of the intrinsicsystem, including FVII, FV, FX, Fli, and Fgn. The thrombin time is a measureof clot formation, or rather, the conversion of Fgn to fibrin. Thrombin is theinitiator and the formation of the clot is the endpoint.Specific factor assays are necessary if the above tests revealabnormalities in the system. The PT and PTT can be performed withpatients’ plasma, mixed with the plasma of a patient with a known deficiencyof a particular protein. The deficiency is identified by correction of anabnormal PT or Pfl by all but one of the plasmas with a known deficiency.Activity of specific factors can also be established by amidolytic assays.A chromogenic substrate, specific for the factor to be tested is used. As thesubstrate is cleaved, a chromogen is generated whose concentration can bemeasured by absorbance and then related to the level of activity of the actingfactor. This assay measures the activity of a particular protein, however, it is24not always indicative of the ability of the protein to function in the actualphysiological system.Various snake venoms have the ability to cleave clotting factors at thesame sites as some of their physiological activators. Echis carinatus (EC)venom contains a protease which splits the Arg32o-Ile bond of protbrombin.Taipan venom cleaves the same sites in prothrombin as the physiologicalactivator FXa (Huisse et al., 1986). Activation by this means, and evaluationof the reaction products establishes the integrity of these sites in the protein.Again, this method is not indicative of the protein’s ability to function in aphysiological system.Specific factor antibodies facilitate the detection of levels of clottingfactors present in individuals with a potential deficiency. Coagulation factorlevels are measured against levels established from a pooled population andsubsequent deficiencies are determined.8. PREVIOUSLY REPORTED PROTHROMBIN DEFICIENCESA deficiency of II can be classified into one of two forms. A completedeficiency of the protein is classified as a hypoprothrombinemic condition,Synthesis of an abnormal protein on the other hand, results in adysprothrombinemic condition. Each respective disorder may be present in ahetero or homozygous state. A variety of complex combinations are thenpossible. A complex heterozygous situation may result where one alleleencodes a hypoprothrombinemic condition, and the other allele encodes adysprothrombinemic condition. Two different dysprothrombins can occur ina single patient. It is also possible to see a double heterozygous conditionarising from the presence of two different hypoprothrombins; however, there25have been no reported cases as such. The resulting clinical symptoms havebeen discussed previously.The incidence of a prothrombin deficiency is quite rare in a population.There are 15 reported cases of dysprothrombinemias, 4 of which are complexheterozygotes for 2 different dysfunctional II molecules. Girolami (1971)summarized 21 known cases of congenital hypoprothrombinemias, of whichover 50% are consanguineous. Five cases exist as complex heterozygotes fordysprothrombinemia as well as hypoprothrombinemia. Since then, twomore cases of homozygous congenital hypoprothrombinemia have beenreported (Montgomery et al., 1978; Baudo et al., 1972). There has also beenone documented case of a deficiency of II, which also included a deficiency ofFVII, FIX, and FX. Although this condition resulted in a deficiency of II, it isneither a case of hypo- nor dysprothrombinemia, but rather a defect in eitherthe y—carboxylation mechanism within the hepatocyte, or a faulty vitamin Ktransport (Chung et al., 1979). This case does not represent a vitamin Kdeficiency, however, as administration of vitamin K does not alter thecondition. There have been several reports of II deficiency resulting from avitamin K deficiency, however, this does not involve hypo- ordysprothrombinemia as a sole cause. In order to establish a true Fil disorder,it is necessary to eliminate a vitamin K deficiency as a potential cause. Asummary of reported prothrombin variants, and proposed sites of mutationsis included in Table 1.Studying cases of reported Fil deficiencies is useful for providinginformation on the structural and functional aspects of the prothrombinmolecule. The analysis of previously reported defects may provide insightinto the cause of the deficiencies reported in this study.TABLE 1: SUMMARY OF REPORTED PROTHROMBIN DEFICIENCIES 26Fli VARIANT LOCATION OF MUTATION CONDI11ON GENOTYPEPoissy N/C Dys-Il HomozygousClamart Activation Region Dys-Il HeterozygousDenver N/C Dys-Il HomozygousBrussels Propeptide Region Dys-Il HeterozygousMadrid Activation Region DYS-Il HomozygousGainsville N/C Dys-IlHouston N/C Dys-IlCardeza Activation Region Dys-Il HeterozygousBarcelona Activation Region Dys-Il HomozygousMouse Activation Region Dys/Hypo-uI (2) Compound HeterozygousPadua N/C Dys-uI (2) Compound HeterozygousPerija N/C Dys-IlSalakta Substrate Binding Region Dys-ulHimi Protease Region Dys-lI (2) Compound HeterozygusQuick Fibrinogen Recognition Dys-lI (2) Compound HeterozygousSubstrate Binding PocketTokushima Protease Domain Dys/Hypo-Il (2) Compound HeterozygousKringle DomainMexico City Fragment 2 Region Dys/Hypo-lI (2) Compound HeterozygousHabana N/C Dys/Hypo-lI (2) Compound HeterozygousMetz Activation Region Dys/Hypo-lI (2) Compound HeterozygousSan Juan Activation Region Dys-ul (2) Compound HeterozygousFragment 1 RegionN/C = not characterizedDys = DysprothrombinemiaHypo = Hypoprothrombinemia27A. HypoprothrombinemiasGirolami et al., reported 3 cases of congenital hypoprothrombinemia, ofwhich 2 were related. All 3 individuals were homozygous for the condition,with Fil activity levels of 9-16% of normal. None of these FIT variants havebeen characterized as yet.B. DysprothrombinemiasThe presence of an abnormal molecule circulating in the blood hasfacilitated studies on structure and function, as the abnormal molecule can beisolated and characterized. Many of the dysprothrombinemias have beencharacterized at the protein level, and a few at the molecular level.Identification of the defect in these naturally occurring variants has allowedthe identification of structural features critical for the activation process,assembly of the prothrombinase complex, and substrate recognition.Of the prothrombin variants that have been identified, FIT-Brussels,FIT-Mexico City, and FIT-San Juan have mutations in the propeptide region ofthe molecule (Kahn et al., 1974; Valls-de-Ruiz et al., 1987; Roberts et al., 1987).FIT-San Juan and FIT-Mexico City are examples of complex heterozygotes andwill be described in a later section.Fil-Brussels has not been sufficiently characterized. This variant wasfound to have half the normal level of functional prothrombin bycoagulation assays, but normal levels by immunological assays. Based onabnormal electrophoresis patterns using antibodies specific for this region, thedefect was thought to be contained in the pro region of Fil. This patientexhibited hemorrhagic symptoms, such that the mutation causing the variantFIT would be expected to be significant.28Four of the dysprothrombinemias contain mutations in the activationregion of the protein. Fli-Clamart and Fli-Cardeza are heterozygous for thedisorder (Huisse et al., 1986; Shapiro et al., 1969); FIl-Madrid and FilBarcelona are homozygous (Diuguid et al., 1989; Rabiet et al., 1986). FITCardeza has prothrombin activity of 50% and antigen levels of 100% ofnormal. The defect in this prothrombin has been localized to theprethrombin 2 region of the protein. Fil-Cardeza was shown to be unable tobind FXa, and subsequently, the Arg320-lle bond cannot be cleaved (Shapiro etal., 1969).Fil-Clamart appears to have a defect in the FXa cleavage of the Arg320-ile bond of prothrombin. This bond is normally cleaved by Echis carinatus(EC) venom. This would indicate that the mutation is probably not at thisspecific site, but rather a site nearby which causes changes in secondarystructure, altering the accessibility of this bond by FXa, but not by E.C. venom.The cleavage at Arg271 seems affected at least in a secondary way as this bondcleavage is delayed relative to normal (Huisse et al., 1986).Fil-Barcelona exhibited severe hemorrhagic symptomsexemplified by low coagulant activity. Fil protein levels were found to benormal. This protein was found to have a defect in activation, indicated bythe absence of 1 of the 2 FXa cleavages. Amino acid analysis revealed amutation of Arg271 to Cys. In addition, the active site was titrated in order tosee what forms were present. It was shown that the second FXa cleavage isnecessary for the appearance of the active site. The first FXa cleavage (toremove the pro-region) is not essential for thrombin-like activity on smallsubstrates, but is necessary for clotting activity(Rabiet et al., 1986).FIT-Madrid was shown to have normal antigenic levels, but lowcoagulant activity. This patient was found to have a severe hemorrhagic29symptoms. The mutation was determined to be in the FXa-catalyzed cleavagesite between the profragment and thrombin regions as found in FIT-Barcelona(Diuguid et al., 1989).A case of dysprothrombinemia has been reported where the suspectedmutation resides in the substrate binding region. Fll-Salakta was reported tohave 100% HI antigenicity, but only 15-18% coagulant activity. The FXacleavage was found to be normal, but thrombin activity was abnormal.Interactions of Fll-Salakta, with compounds that bind to the primary bindingsite were found to be abnormal. Peptide sequence analysis revealed asubstitution of G1u466 with Ala. This mutation was thought to change theconformation around the substrate binding site containing Trp4s, which is ina unique surface loop on the molecule. This amino acid change wasspeculated to cause an alteration in this loop which in turn would reduce theaffinity for antithrombin Ill and fibrinogen (Miyata et al., 1992).A number of dysprothrombinemias have been reported but theirdefects have yet to be identified. FII-Gainsville was discovered in a pair ofidentical twins who exhibited Fil activity levels of 23-25% of normal (Smith etal., 1981). FIT-Houston exhibited half normal antigenic levels of Fil, andminimal Fil activity. Crossed immunoelectrophoresis may indicate thepresence of two forms of dysfunctional FIT, although this has yet to beconfirmed (Weinger et al., 1980). FIl-Perija, as well as FIT-Denver, and FITPoissey, are all examples of homozygous dysprothrombinemia (Ruiz-Saez etal., 1986; Montgomery et aL, 1980; Tapon-Bretaudiere et al., 1983). All threecases exhibit severe hemorrhagic symptoms, corresponding to minimalcoagulant levels. These variants have been studied by immunological andbiological assays only.30There are at least four reported cases of compound heterozygotes ofdysprothrombinemia: FII-Padua, Fli-Himi, Eli-Quick, and Fil San Juan(Girolami et al., 1974; Morishita et al., 1992; Henriksen et a!., 1986., Roberts etal., 1987).Fll-Padua represents an interesting case where Fil activity levels aredecreased in one and two stage assays, but normal by staphylocoagulase assays.The parents of this individual are not consanguineous. It is thereforesuspected that the structural abnormality in this dysprothrombinemia mayconsist of two separate populations of abnormal Fil, or a single populationthat is altered in such a way that it is half active and half inactive (Girolami eta!., 1974).Fil-Himi was identified as a complex heterozygote for twodysfunctional prothrombin molecules. Fil antigen levels were normal, butFII activity levels were only 10% of normal. Two point mutations wereidentified in the thrombin portion of the molecule. A mutation of a ‘T’ to a‘C’ at nucleotide position 8751, results in the substitution of a Thr for Met atcodon 337 (Thrombin Himi I). A second mutation of a ‘G’ to an ‘A’ atnucleotide position 8904 encodes His instead of Arg at codon 388 (ThrombinHimi II). It was established that Himi I was inherited from the father andHimi II from the mother (Girolami et al., 1974).Fil-Quick, upon activation, reveals two dysthrombins. The mutationinvolved in Quick I was identified as a substitution of Arg382 with Cys.Arg382 has been shown to be critical in determining thrombin specificity forfibrinogen. Quick II was found to have some unusual properties. Quick IIwas unable to clot fibrinogen, nor could it bind artificial substrates orcompetitive inhibitors of thrombin. Quick II was identified as a substitutionof G1y558 by valine. This glycine residue has been shown to be highly31conserved in the chymotrypsin family of serine proteases where it forms partof the substrate binding pocket. This alteration provides evidence that thenot only does this glycine influence primary substrate specificity, but providesevidence that the catalytic activity of the serine proteases are influenced bystructural changes within the primary binding pocket (Henriksen et al., 1986).FIT-San Juan is a compound heterozygote, with FIT activity levels of20% and FIT antigen levels of 93% of normal. The two forms of dysfunctionalFIT in this patient have not been thoroughly characterized. Roberts et al.,(1987) have summarized was has been discovered so far. The mutationinvolved in San Juan I appears to be a defect in the activation region. Themutation associated with San Juan II appears to involve the gammacarboxylation region of prothrombin fragment 1.C. Compound Heterozygotes for Dysprothrombinemia andHypoprothrombinemiaThe presence of both conditions involving a deficiency in prothrombinin a single patient results in a compound heterozygous condition. Thisoccurs when an allele for dysprothrombinemia is inherited from one parent,and an allele for hypoprothrombinemia is inherited from the other. Patientswith this condition exhibit severe hemorrhagic symptoms. FIT antigen levelsare usually less than 50%, and FIT activity levels are minimal. There havebeen five reported cases of compound heterozygotes for DysHypoprothrombinemia: FIT-Mouse, FIT-Tokushima, FIT-Mexico City, FITHabana, and FII-Metz (Girolami et al., 1978; Twahana et al., 1992;Valls-de-Ruizet al., 1987; Rubio et al., 1983; Rabiet et al., 1979).FIT-Mouse has 50% FIT antigen, and 10% FIT activity. The cause ofeither of these conditions is uncertain. It is thought that the mutation residesin the FXa sensitive region (Girolami et al., 1978).32Fli-Tokushima has been characterized on the molecular level. Themutation involving the dysprothrombinemia was identified as a pointmutation of a ‘T’ to a ‘C’ at nucleotide position 9490. This results in asubstitution of Arg418 with tryptophan. The hypoprothrombinemiccondition resulted from a single base insertion in exon 6. A ‘T’ was insertedat nucleotide position 4177. This change alters the sequence from codon 114,resulting in a premature stop at codon 174 in exon 7. This stop occurs in thekringle 2 domain, preceding the thrombin portion of the molecule. Thedysprothrombinemic allele was inherited from the mother whereas thehypoprothrombinemic allele was inherited from the father (Iwahana et al.,1992).Fil-Mexico City has been studied at the protein level. Thedysfunctional prothrombin molecule is suspected to have a mutation infragment 2. The cause of the hypoprothrombinemic condition remainsuncertain; however, this is an acquired mutation in the patient as neitherparent was found to be heterozygous for the hypoprothrombinemic condition(Valls-de-Ruiz et al., 1987).Fil-Habana has been identified clinically as a complex heterozygote forboth types of prothrombin deficiency; however, no characterization of eithercondition has been reported. The mother was found to be heterozygous forthe dysprothrombinemia, and the father was heterozygous for thehypoprothrombinemic condition (Rubio et al., 1983).FII-Metz has been studied on the level of the protein. Thedysprothrombin was isolated and its activity towards natural and syntheticsubstrates was evaluated. Interactions of Fil substrates and inhibitors withFTI-Metz indicates that the structural defect involving thedysprothrombinemia resides in the thrombin portion of the molecule, more33specifically in the catalytic site, and not the specific interaction with substrate.The identity of the hypoprothrombinemic condition is undetermined (Rabietet al., 1979).9. OBJECTWES OF THIS STUDYTwo patients were presented with a severe bleeding disorder.Screening assays for hemostatic disorders were employed, as well as specificfactor activity and antigen assays. The patients were identified as havingsevere hypoprothrombinemia. As this disorder is extremely rare in thepopulation, it presented a unique opportunity to examine the molecular basisof this disease. DNA samples from the afflicted individuals were provided.Our goal was to identify causative mutations in the prothrombin gene ofthese patients, and speculate on the effects that these mutations would havecaused on the protein. The strategy employed involved the use of molecularbiology techniques such as the Polymerase Chain Reaction (PCR), cloning,and DNA sequencing in an attempt to elucidate discrepancies between thenormal prothrombin gene, and that present in the hypoprothrombinemicpatients. This study may facilitate not only an understanding of the structureand folding properties of the prothrombin molecule, but may provide insightinto the transcriptional processes of regulation.34MATERIALS AND METHODS1. REAGENTSYeast extract, bacto-agar, and bacto-tryptone were purchased fromDifco Laboratories. Agarose, acrylamide, bisacrylamide, urea , ammoniumpersulfate, and TEMED were purchased from Bio-Rad Laboratories. Phenolwas purchased from British Drug Houses Ltd. (BDH). DTT, EtBr, ME,DMSO, and RNase A were purchased from Sigma Chemical Co. cz-[thio-35S]-dATP was purchased from New England Nuclear. XGAL and IPTG werepurchased from 5 Prime -> 3Prime Inc. Deoxy and dideoxy-ribonucleotideswere obtained from Pharmacia. All other reagents were of reagent grade orhigher, and were obtained from Bio-Rad, BDH, Pharmacia or Sigma.Restriction endonucleases, T4 DNA Ligase, and T4 DNA Polymerasewere purchased from Bethseda Research Laboratories (BRL), BoehringerMannheim, or Pharmacia. Sequenase version 2.0 was purchased from UnitedStates Biochemical. Klenow-large fragment was obtained from BRL.Recombinant Taq polymerase (Amplitaq) was obtained from Perkin ElmerCetus. Kodax X-Omat and Kodax XAR film was used for autoradiography.2. STRAINS, VECTORS, AND MEDIAA. VectorsPCR amplified regions of the prothrombin gene were ligated into thepolycloning site of the vector Bluescript, obtained from Stratagene. Thisphagemid is 2.9 kb in length, and contains the Escherichia coli (E. coli )origin of replication, a gene for ampicillin resistance, the fi intragenic region,and part of the lac Z gene for cz-complementation with the lac Z gene of E.coli.35B. Bacterial StrainsThe hosts for transformations were the E. coli strain DH5cxF’. Thegenotype of these cells are: F,ndA1,hdR17(rk,mkj,uE44,thiManiatis et al., (1989). Theblue /white color selection was used for screening colonies on plates. Singlestranded DNA was generated by addition of the helper phage M13 vcs. Thishelper phage conveys kanamycin resistance to its E. coli host.C. MediaBacterial hosts were grown on LB plates (5g bacto-yeast extract, lOgNaC1, log bacto-tryptone, pH to 7.5, 15g agar, adjust volume to 1 litre,autoclave to sterilize) supplemented with 50 Ig/ml ampicillin, 25p.g/mlIPTG, and 5Opg/ml XGAL. IPTG and XGAL were made up indimethylformamide.3. OLIGONUCLEOTIDESOligonucleotides were designed for use in the Polymerase ChainReaction (PCR), in the amplification of specific exons of the prothrombingene. A summary of the oligonucleotide primers is shown in Table 2.Primers were designed to flank the target region, the forward primer wasdesigned to be complementary to the (-) strand, and the reverse primer to becomplementary to the (+) strand. Primers were designed to include the splicejunctions of each exon. In the case of exon 1, the primer also included thefirst 100 bases 5’ to the exon for amplification. Oligonucleotides were36T1T1zC.)0zzC) (‘IC)I-——--———--——--—--——I-)-I.-‘o\oo—1O’U..-——--———--——--—--——---00-—)-•——I-Cj——-aI‘-,t-5.-I-’I-—‘-•I .‘-.—aoc,.ccu.I.U.000—00— -00C)%—(CTh(J(Th((J((Th(r>>cr-c•(ThC)I(5rOrrrcc‘-rrmCrer !pHr !E(ThC)x-(crzc•->C)ra >.>-U.U.Qi.’.oqoo-ac-’.U.O ‘--ci.oU.-I--——--——--——--—--——--——---Cl)III‘iIIiFi’iIIIII I—I—IIIIIIiiI I-i()IidIIIIcDIfl‘‘‘‘‘‘HHH‘‘‘ mmIIIIZI37synthesized on an Applied Biosystems 391 DNA Synthesizer, and werepurified by reverse-phase chromatography on Sep-Pak C18 cartridges asdescribed by Atkinson and Smith (1984). The resulting primers wereresuspended in distilled water (dH2O), and stored at -20°C.4. Genomic DNADNA samples from hypoprothrombinemic patients were supplied forthe purposes of this investigation. Prothrombin Utrecht, was kindly donatedby Dr. H.K. Nieuwenhuis. Prothrombin activity and antigen levels weredetermined to be 3% of normal in both cases. Prothrombin Vancouver wasdonated by Dr. Ka-Wah Chan. Activity and antigen levels in this case wereeach found to be 3% of normal.5. ISOLATION OF DNAA. Plasmid DNAPlasmid DNA was obtained by a modification of the Alkaline Lysisprocedure of Maniatis et al., (1989). LB (5m1), containing ampicillin (50jig/mi) was innoculated with a single colony of bacteria and grown at 37°Covernight. The culture was centrifuged at 3000 rpm for 15 mm in a benchtopcentrifuge. The cell pellet was resuspended in 200 j.il of Glucose Buffer(50mM glucose, 25mM Tris-HCL pH 8.0, 10 mM EDTA). Suspensions wereincubated at room temperature for 5 mm, following addition of 400 p1Alkaline-SDS Buffer (0.2 N NaOH, 1% SDS).Cellular debris and genomic DNA was removed by the addition of 30 p1 of3M NaOAc pH 5.2, incubation at 4°C, and centrifugation in a microfuge for 10mm at room temperature. To 750 p1 of the supernatant, 450 jil of coldisopropanol was added, and incubated at 4°C for 5 mm. The plasmid DNA38was precipitated by centrifugation in a microfuge for 5 mm at roomtemperature. The resulting pellet was resuspended in 200 j.tl dH2O , followedby a short centrifugation to remove any particulate matter. The DNA wasreprecipitated by addition of 100 p1 of 7.5 M NaOAc pH 7.0, and 1 ml 95%ethanol. After centrifugation for 15 mm at room temperature, the pellet waswashed with 70% ethanol, dried, and finally resuspended in 35 p1 dH2O.B. Isolation of Single Stranded DNA.After a 4 hour incubation of a culture as described above, 200 p.1 of M13vcs helper phage were added to each culture. After a further hour at 37°C,kanamycin was added to a final concentration of 50 pg/mi and the culturewas left to incubate overnight. Following a 15 mm centrifugation at 3000rpm, the supernatant was incubated on ice for 30 mm, after the addition of 1ml 50% PEG: 7.5 M NaOAc pH 7.0 (1:1, v/v). The suspension wascentrifuged at 3000 rpm for 15 mm at 4°C. The resulting pellet wasresuspended in 300pi Tris-EDTA pH 8.0. The DNA was purified by extractionwith an equal volume of phenol:chloroform (1:1, v/v). DNA wasprecipitated by addition of 30 p13 M NaOAc pH 7.0, and 1 ml 95% ethanol,with centrifugation for 15 mm in a microfuge. The pellet was washed with70% ethanol, dried, and resuspended in 25 p.1 dH2O.6. Gel ElectrophoresisA. Agarose Gel Electrophoresis.DNA fragments, in the form of plasmid DNA, plasmid inserts, or PCRfragments were separated according to size by electrophoresis in agarose gels.The running buffer was lx TAE (32 mM Tris, 16 mM NaOAc, 0.8 mM EDTA,pH 7.2). EtBr was included in the gels at a concentration of lj.tg/ml, and DNA39was visualized by examination under ultraviolet light. DNA samples wereprepared in 1X Loading Dye (3% ficoll, 0.02% xylene cyanol, 0.02%bromophenol blue), and electrophoresed at 5-10 volts/cm, until an adequateseparation of the DNA had been achieved.B. Denaturing Polyacrylamide Gels.Sequence analysis of radiolabelled DNA fragments was facilitated byelectrophoresis of the DNA on 6% polyacrylamide gels at 60 watts constantpower for 2 to 6 hours. The gels were prepared by addition of 8.3 M urea andan appropriate volume of bis:acrylamide (38%:2%, w/v), made up in 1XTBE. The running buffer was also lx TBE (50 mM Tris base, 50 mM boric acid,1 mM EDTA). Polymerization of the gel was initiated by addition of 0.066%(w/v) ammonium persulfate and 0.024% TEMED. DNA was visualized byautoradiography by drying the gel on 3MM paper in a heated Bio-Rad geldrier for 1 hour, followed by exposure of the gel to Kodax X-Ray filmovernight.7. POLYMERASE CHAIN REACTION.The PCR was used for amplification of target sequences of DNA forsubsequent cloning. A summary of the PCR and cloning strategy is shown inTable 2. A schematic diagram of the oligonucleotide primers, and theirposition relative to the exons of FIT is shown in Figure 6. The method isdescribed by Saiki et al (1985), and was carried out by the use of a PerkinElmer Cetus DNA Thermocycler. Reactions contained a total volume of 50pi.Several different PCR buffers were used to maximize amplification ofvarious fragments. Buffer A ( 67 mM Tris pH 8.8, 16.6 mM (NH4)2S04, 10mM 2-mercaptoethanol, 1.0 mM MgSO4)), Buffer B (10 mM Tris pH 8.5, 0.05%4012 14 28 26—.H6I I13 15 25 4273II—1 8 H 9 I I 10 H ii j I 12 I I 13 I—I 14 I4.4— 423 24Figure 6. PCR Strategyfor the Amplification of the Exons of ProthrombinThis schematic diagram illustrates the orientation of PCR primers with respectto the exons of human prothrombin. The boxes symbolize the relative size of eachexon, with the lines between representing introns. Forward and reverse primersare indicated by their orientation and number. Primer 29 (hPT29) is shown as anested primer between primers 26 and 27 for exon 7.41Tween-20, 0.05% Nonidet P-40, 1.0 mM MgC12), and the standard bufferrecommended by Perkin Elmer Cetus (500 mM KCL, 100 mM Tris-HCL pH 8.3,15 mM MgCl2, 0.1% gelatin (w/v)) were all employed. Reactions contained2pi genomic DNA (100 ng/p.l), 5 ul lox PCR buffer, 1 p.1 dNTP (10mM of eachof dATP, dCTP, dGTP, dTTP), 1 p.1 forward primer (lOOng/p.l), 1 p.1 reverseprimer (lOOng/p.l), and 1 p.1 Taq polymerase (5 units!p.1), with the volumemade up to 50 p.1 with dH2O. PCR reactions varied but followed the generalformat of denaturation at 94°C for 15 to 30 sec, annealing at 50-56°C for 15 to30 sec, and extension at 72°C for 30 to 60 sec, for a period of 25-30 cycles.Annealing temperatures were initially estimated by the equation:TA=[4°C(GC)+2°C(AT)] - 5°C. TA represents the final annealing temperaturefor a particular primer. The optimal annealing temperature was estimatedbased on the TA values from each set of primers.8. DNA SUBCLONINGA. Purification of DNA FragmentsPCR reactions containing a single product (evaluated by electrophoresison an agarose gel) were precipitated by addition of 0.5X volume 7.5M NaOAcpH 7.0, and 2)( volume 95% ethanol. After centrifugation, the pellet waswashed with 70% ethanol, dried, and resuspended in 20 p.1 dH2O. A singlePCR product was rescued from a non-specific amplification reaction by gelelution. The PCR reaction was made up in loading dye, and electrophoresedon a 1% agarose gel. The desired product was excised from the gel byremoving the block of agarose which contained it. The block was subjected toelectrophoresis within an dialysis bag with 300 p.1 TE pH 8.0 for 10 mm at 90volts. The DNA, having electrophoresed out of the gel and into solution wasremoved to an eppendorf tube where it was precipitated as described as above.42B. Restriction Digestion and Blunt-ending of Purified PCR ProductsFragments were digested with the appropriate restriction enzymes (asdirected by the manufacturer) according to the strategy for directional doningas outlined in Table 2. The digested product was precipitated as describedabove, and finally resuspended in 10 jil dH2O. In some cases, where PCRproducts were not directionally cloned, a blunt-ending strategy was employed.The enzyme used to do this was the Kienow fragment of E. coli DNAPolymerase 1 (Maniatis et al., 1989). Kienow fragment (1 p.1 of 5 units/ui), 3 p.1lox Kienow buffer (70mM Tris pH 7.4, 500 mM NaC1, 70 mM MgC12), and 20p.1 of purified PCR DNA were incubated at 37°C for 5 mm. After 1 p.1 of 1.25mM dNTP (1.25 mM with respect to each nucleotide) was added, and thesolution was incubated for 25 mm at 37°C. The enzyme was heat inactivatedfor 10 mm at 68°C. The product was precipitated as described above, thenresuspended in 10 p.1 dH2O.C. Vectors Used for LigationsThe vector Bluescript (Stratagene) was used for all ligations. Bluescript(BS) was cut with the appropriate enzymes for either directional, or blunt-endcloning. The cut vector was ethanol precipitated and resuspended in dH2O.The quantity of vector was estimated by electrophoresis on an agarose gel.Vectors containing both orientations of the polycloning site (BSKS/BSSK)were used for ligations. Incorporation of the PCR insert was evaluated byblue/white colour selection of resulting transformed colonies.43D. Ligation of PCR Fragments into Bluescript VectorsLigations were carried out using an appropriate ratio of 4:1 insert:vector(as judged by electrophoresis on an agarose gel). Reactions typically contained1 .tl of linearized vector, 3.0 i.l 5X Ligase Buffer (50 mM Tris pH 7.6, 10mMMgC12, 1mM ATP, 1mM DTT 5%PEG 8000 (w/v)), 1 tl T4 ligase (5 units/jil),and the DNA insert made up to a volume of 15 jil with dH2O. Ligations werecarried out for 3 to 12 hours at room temperature.9. TRANSFORMATION OF RECOMBINANT BLUESCRIPT INTO E CCLICompetent bacteria were obtained by growing a 50 ml culture ofDH5aF’ to an OD6oo of 0.5-0.6. Centrifugation of the culture at 5000Xg for 10mm resulted in a bacterial pellet which could be resuspended in ice cold 50mM CaC12. Incubation on ice for 30 mm followed by centrifugation generateda second bacterial pellet which was resuspended in 4 ml ice cold 50 mM CaC12,20% glycerol, and stored at -70°C. Cells were thawed immediately prior to use.The bacterial strain DH5cLF’ was used such that the generation of singlestranded DNA from M13 infected cells was an option.Ligations were used to transform bacteria by addition of 1-5 tl of theligation mixture to 50 .il competent cells, followed by incubation on ice for 30mm. The cells were heat shocked at 42°C for 2 mm, 1 ml of LB was added toeach reaction, and the reactions were incubated at 37°C with shaking for 20mm. Transformed bacteria were selected by plating 50-100 p.1 of the bacterialculture on LB agar plates supplemented with 100 jig/mi ampicillin, 25 jig/miIPTG, and 50 jig/mi XGAL. Plates were incubated at 37°C overnight. Coloniescontaining inserts were chosen for further DNA isolations.4410. DNA SEQUENCE ANALYSISA. Single Stranded DNA SequencingDNA sequence was determined by the chain termination method(Sanger et a!., 1977). Single stranded DNA was sequenced with T7 DNAPolymerase (Sequenase) using solutions described by the Sequenase protocol(3rd ed.) Single stranded template (7 p1) was mixed with 2 jil 5X AnnealingBuffer (200 mM Tris pH 7.5, 100mM MgC12, 250 mM NaC1), and 1 tl ofsequencing primer (100 ng/pi) (see Table 3). The reactions are incubated at68°C for 10 mm, and slowly cooled to room temperature over a period of atleast twenty minutes. To each sequencing reaction, the following was addedat room temperature: 1 p1 100 mM DTT, 0.5 p1 [a-35S]dATP (3000 Ci/mmol),2.5 p1 Labeling Mix (1.5 iM dCTP, 1.5 pMdTTP, 1.5 jiM dGTP), and 2 p1 T7Polymerase (diluted 1:15 with Enzyme Dilution Buffer (10 mM Tris-HCL pH7.5, 5 mM Dfl’, 0.5 mg/mi BSA)). The labeling reactions proceeded for 1.5-2mm before 3.5 p1 of the reaction mix was transferred to 2.5 p1 of eachtermination mix (80 pM with respect to each of ddATP, ddCTP, ddGTP,ddrrP, and 50 mM NaC1) pre-warmed to 42°C. Stop buffer dye (98%formamide, 20 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue)was added after a total of 5 mm incubation at 42°C. Samples were boiled for 3mm, before being loaded (2 p1) on a 6% polyacrylamide sequencing gel.Samples were stored at -20°C.rTl 0 0 z O 0 rn Cl, 0 I Q1B. Double Stranded DNA SequencingDouble stranded DNA sequencing was carried out as described abovewith the addition of a denaturation step. 15 p1 DNA was added to 9.6 p1dH2O, 3 p1 2N NaOH, and 2.4 p12.5 mM EDTA. Incubation at 37°C for 25 mmwas followed by ethanol precipitation by the addition of 0.1X volume 3 MNaOAc pH 5.2 and 2X volume cold 95% ethanol. The DNA was centrifugedin a microfuge, washed with 70% ethanol, and dried. The pellet wasresuspended in 7 p1 dH2O, and sequencing was carried on as above.4647RESULTS AND DISCUSSION1. SUMMARY OF CLINICAL STATUS OF PATIENTSDNA samples were provided from two patients for the purposes of thisstudy. Both patients exhibited severe bleeding problems. Patients were foundto have severe hypoprothrombinemia based on a clinical analysis.Prothrombin activity and antigen levels were analyzed for each individual.Fil-Vancouver activity and antigen levels were both assessed at 3% of normal.Fll-Utrecht had 3% activity and antigen levels. Plasma samples were notmade available for this study. In addition, family studies confirmed that eachpatient was the result of a consanguineous union.2. PCR AMPLIFICATION OF THE PROTHROMBIN GENE IN EACHPATIENTAll of the exons, including the 5’ and 3’ splice consensus regions weresuccessfully amplified by using the PCR. As all fragments were amplifiedand were of the expected size, it can be assumed that there were no grossdeletions or insertions within the exons. Exons were usually amplified inpairs, depending on their size and distance apart (see Figure 6). Restrictionsites were incorporated into each primer to allow for directional cloning ofinserts (refer Table 2).Exons 1 and 2, as a single fragment, were extremely difficult to amplify.Even at minimum annealing temperatures, several different products wereobtained. Isolation of the correct region was finally achieved by extractionfrom an agarose gel, of the region that contained DNA of the expected size.48This sample was then used as a template for a second PCR reaction. Thisallowed for the specific amplification of the correct region.In addition, exon 7 was a difficult piece to amplify. The first round ofthe PCR resulted in a smear of several products on an agarose gel. Tomaximize the generation of a specific product, a second PCR reaction wasperformed using the first PCR product as a template. A third primer wasdesigned such that it would anneal inside the region spanned by the first setof primers (nested primer), within the region spanning exon 7 (refer Figure6). The third primer (annealing 3’ to exon 7), was used in conjunction withthe forward primer (annealing 5’ to exon 7) used in the first round of thePCR. In this way, specific amplification of the exon VII target region wasmaximized.Exons 8 and 9 amplified as a single product quite successfully, but wereextremely difficult to done. Several attempts at directional cloning wereunsuccessful. The piece was finally cloned by using a blunt-end strategy.This strategy was successful, but was of low efficiency. Several ligationreactions were necessary in order to isolate a sufficient number oftransformants for sequencing.Exon 12 was another fragment which did not amplify with greatefficiency; however, cloning of this piece was very efficient.3. DNA SEQUENCE ANALYSIS OF THE PROTHROMBIN GENE INHYPOPROTHROMBINEMIC PATIENTSThe exons of the prothrombin gene, including the 5’ and 3’ splicejunctions, were sequenced in each patient. In addition, the first 100 bp 5’ toexon 1, including the putative promoter region was sequenced. Sequenceanalysis performed for each patient was compared to that of wild-type49prothrombin (Degan et al., 1987). Several unique polymorphisms, as well aspotentially significant base changes were identified in both patients. Theresults are summarized in Table 4.A. Sequence Changes Observed in Fil-UtretchAn insertion of a single ‘A’ was observed in the putative promoterregion 54 bp 5’to the transcription start site in exon 1. Nucleotidenumbering is according to the wildtype Fil sequence of Degan et al., (1987).In the splice consensus region 5’ to exon 2, there was a deletion of a ‘T’ atnucleotide position 461. Autoradiograms showing both of these changes areshown in Figure 7. A silent leucine polymorphism, resulting from a singlebase change from CTA->CTG was identified in exon 2 at position 554. Nochanges were observed in exons 3, 4, or 5, including their corresponding spliceregions. A single base change of a ‘G’ to an ‘A’ was observed in the consensussplice region 3’ to exon 6 at position 4272. No changes were found in exons 7through 9. A silent threonine polymorphism was observed in exon 10. Thisresults from a single base change from ACA to ACC at position 8903.Sequence analysis of exons 11 through 14 revealed no diversions from thepublished wildtype sequence. Compared to the normal prothrombin gene, allmutations found in this patient were homozygous. These results aresummarized in Table 4. Autoradiograms of the described sequences areshown in Figures 7,8, and 10.B. Sequence Changes Observed in Fil-VancouverAn insertion of a single ‘A’ was observed 54 bp 5’ to exon 1 as in theabove patient. This insertion occurs within the putative promoter region ofthe II gene. Similarly, a deletion of a ‘T’ was observed in the 5’ spliceTABLE4.SUMMARYOFSEQUENCEANALYSISONPROTHROMBINDEFICIENTPATIENTSPATIENTNUCLEOTIDE(bp*)CHANGECOMMENTSII-Utrecht-54InsertedAputativepromoterregion461DeletedTspliceregion5’toexon2554A->GsubstitutionLeucinepolymorphisminexon24272G->Asubstitutionpolymorphisminspliceregion3’toexon68903A->CsubstitutionThreoninepolymorphisminexon10Il-Vancouver-54InsertedAputativepromoterregion461DeletedTspliceregion5’toexon2554A->GsubstitutionLeucinepolymorphisminexon24272G->Asubstitutionpolymorphisminspliceregion3’toexon67485-74873bp(AAG)deletionLysinedeletioninexon98903A->CsubstitutionThreoninepolymorphisminexon10*NumberingbasedonsequenceofDegeneta!.,(1987).51consensus region of exon 2. A single base change of a ‘G’ to an ‘A’ wasobserved in the splice consensus sequence 3’ to exon 6, again as in theprevious patient. A 3 bp deletion of a AAG codon was observed in exon 9 atposition 7485 to 7487. This change deletes a single lysine residue. A silentthreonine polymorphism was observed in exon 10, resulting from a singlebase substitution of A -> C at position 8903. No other changes were observedin the exons or corresponding splice regions of the FIT gene in this patient.All observed mutations, and polymorphisms were found to be homozygous.These results are summarized in Table 4. Autoradiograms are shown inFigures 7-10.52CCACGT ACGT3’CATT4 TCC5’Figure 7. Sequence Changes Involving an Inserted ‘A’ in the PutativePromoter Region and a Deleted ‘T’ in the Splice Region 5’ to Exon 2.Sequence analysis using the chain termination method revealed two changesobserved in both FIT-Vancouver and FII-Utrecht. A single ‘A’ residue wasobserved 54 bp 5’ to the transition start site of exon 1 (numbering according toDegen et al., 1987) on the sense strand. This change is found in the sequencereported by Bancroft et a!., 1990. In the splice consensus region 5’ to exon 2, asingle ‘T’ has been deleted. This deletion occurs at nucleotide position 461 onthe sense strand. Base changes are indicated with an arrow.insertion deletion53C CT TA GACGTTAA4 GC3’ SpliceRegionFigure 8. Autoradiograms of Polymorphisms in Exon 2 and the Splice Region3’ to Exon 6.Sequence analysis revealed a new silent Leucine polymorphism in exon 2.This sequence change occurs at nucleotide position 554 on the sense strand,and involves a single base change in the codon CTA to CTG. Theautoradiogram is of the anti-sense strand. A second polymorphism wasfound at nucleotide position 4272, involving a single base change of a ‘G’ toan ‘A’. This polymorphism was previously reported by Iwahana et aL, 1992.The sequence shown here is of the anti-sense strand and therefore shows thecorresponding ‘T’ . Both of these changes were observed in Fli-Vancouverand FII-Utrecht.ACGTExon 2Leu 5654GCG1u300 ATGCATLys302 A TGCTASer303 C GGCACGTFil-VancouverGCGIu300A TGC___Lys301ATATGCATLys302 A TGCTASer303 C GGCFigure 9. Deletion of the Codon ‘AAG’ in Exon 9 of P11-Vancouver.Sequence analysis revealed the deletion of a 3 bp stretch, involving a singlein-frame codon AAG’. This sequence change occurs at nucleotide position7485 to 7487, and involves the deletion of a single lysine residue.Autoradiograms of the anti-sense strand of this mutation, as well as thenormal wild-type sequence are indicated with an arrow.wild-type55ACGTA —0 CThr 388Figure 10. Sequence Analysis of Threonine Polymorphism in Exon 10.Sequence analysis revealed a single base change of an ‘A’ to a ‘C’ at nucleotideposition 8903 on the sense strand. This silent Threonine polymorphism wasfound in both FIT-Vancouver and FII-Utrecht. The single base change isindicated with an arrow.56DISCUSSION1. NATURALLY OCCURRING MUTATIONS FOR STRUCTURAL ANDFUNCTIONAL ANALYSIS OF PROTEINSThe natural occurrence of individuals with Fil deficiencies hasfacilitated studies on the structure and function of the Fil molecule.Mutations which result in a hypoprothrombinemic condition allow analysisof structure and folding properties of the molecule. Generally, where little orno protein is being synthesized, the disruption tends to be in the transcriptionprocess, and therefore in the promoter or enhancer regions. Alternatively, amutation within the signal peptide may prevent the protein from beingexported from the cell. Improper folding of the protein, based on a change inthe primary structure, may result in the protein being recognized andhydrolyzed within the cell. Mutations which result in a dysprotlirombinemiccondition, are much more useful for studying functional aspects of theprotein. Identification of the structural defect that results in decreased oreliminated function can provide insight into the mechanism of action of theprotein.2. SEQUENCE ANALYSIS OF PROTHROMBIN MUTATIONSUpon sequence analysis of the prothrombin gene in both FilVancouver and Fll-Utrecht, a number of deviations were observed whencompared to the wild-type sequence. A number of silent polymorphismswere evident, as well as potentially significant mutations within theprothrombin gene. The wild-type sequence used for comparison, and relativenumbering of bases is that of Degan et al., (1987).57A. Inserted ‘A’ in Putative Promoter RegionThe first change was observed in what would be the putative promoterregion in both Fli-Vancouver, and Fll-Utrecht. At 54 bp before thetranscription start site (tss), there is an insertion of an additional base ‘A’.Although this base is absent in the sequence reported by Degan et a!., itis present in the 5’ sequence reported by Bancroft et al., (1990). With thisinformation, it appears obvious that this base has little effect on thehypoprothrombinemic patients condition; however, it opens up atremendous area of importance of study: the transcriptional control region.Typically, a condition which results in a lack of protein synthesis tends to bedue to a disruption in one of three areas: the transcription process, inefficientsecretion of the produced protein, or improper folding. If in fact a moreobvious mutation is not present in the exons, the promoter region remainsan important region of study and will be discussed in a later section.B. A Deletion of a ‘T in the Splice Region 5’ to Exon 2A change was observed in the prothrombin gene of both patientscompared to that of wild-type Fil. At nucleotide position 461, a single base ‘T’has been deleted. This change is expected to have negligible effects on thepatients condition based on two observations. Initially, this deletion occurs ina sequence involving three ‘T’ bases, or more specifically, the sequence5’CCTTTACAG3’. Even if this ‘T’ was important for splicing, either of theflanking ‘T’ bases could easily compensate. Secondly, the sequence of thesplice regions all observe the GT-AG rule of Breathnach et al., (1978), with theexception of the splice region immediately 3’ to exon 12. This region containsa GC rather than a GT. Studies in vitro have shown that genes containing58mutations in the GT sequence of splice regions are still functional, however,intermediates accumulate (Padgett et al., 1986). If this deleted ‘T’ caused analteration in splicing, you would still expect to see some normally splicedproduct, due to the other two ‘T’ bases, which would produce normal HI,presumably at levels higher than that observed with either of these patients.In addition, there would be the accumulation of incorrectly spliced mRNAwithin the cell. The second observation could only be confirmed by doingsome form of immunocytochemistry, in order to follow the mRNA throughthe cell. With this information, it would seem that this sequence change issimply a polymorphism, with no significant effect on generation of FIT ineither of these patients.C. Leucine Polymorphism in Exon 2A substitution of a G to an A, occurs at nucleotide position 554. Thismutation is a silent polymorphism of Leu56 in exon 2. This is a single basechange in the third position of the codon CTA. Because this change does notcause an alteration in the amino acid , it therefore has no effect on generationof protein. This polymorphism was observed in both Fil-Vancouver, and FITUtrecht.D. Base Substitution in Splice Region 3’ to Exon 6.A single base change of a ‘G’ to an ‘A’ was observed in the consensussplice region 3’ to exon 6 at position 4272. This was a homozygous mutationwhich was observed in both FIT-Vancouver and Fll-Utrecht. This is not anovel polymorphism as it was also reported by Iwahana et al., (1992).59E. Lysine Deletion in Exon 9 of Fli-VancouverA 3 bp deletion of the sequence AAG was observed in exon 9 of theprothrombin gene in FIT-Vancouver. This is an in-frame deletion whichresults in the deletion of the single amino acid Lys3ol. Although thismutation does not cause a global disruption in the amino acid sequence of Fil,it is expected to have a serious effect on the protein. As mentionedpreviously, mutations which cause little or no protein synthesis are expectedin one of three areas: the signal sequence, a mutation causing an alteration infolding, or in the absence of a more obvious mutation in an exon, a mutationinvolving transcriptional regulation. It would appear that this deletionwould cause a structural change in the protein on the tertiary level. Adeletion of an amino acid in one area, can cause global changes in the foldingof a protein.F. Threonine Polymorphism in Exon 10A silent polymorphism was observed in exon 10 of Thr389. This is theresult of a substitution of an ‘A’ with a ‘C’, in the codon ACA. Again, thissilent polymorphism was observed in both patients.3. PROPOSED EFFECTS OF OBSERVED MUTATIONS ON PROTHROMBINA. FIT-VancouverIt would seem that the most obvious explanation for ahypoprothrombinemic condition in this patient, would be the result of thedeleted lysine residue in exon 9. The addition sequence changes describedabove are not expected to have a significant effect on the protein, and aretherefore considered to be polymorphisms of the prothrombin gene. This60mutation at the nucleotide level is expected to cause global changes instructure of the protein, such that it is misfolded within the cell. A mis-folded protein is quickly degraded by the cell as a defense againstaccumulation of aberrant proteins. Although this seems the most obviousexplanation for the deficiency in this patient, it could be confirmed by one oftwo methods. It would be interesting to observe the aberrant protein as itpasses through the cell. This could indeed be accomplished byimmunocytochemistry. Fluorescent antibodies that are specific for Fli can beused to follow the transport of recombinant Fli-Vancouver through the cell.In this way, it would be obvious if the protein was being mis-shuttled in thecell, or trapped within any particular region. It would also be interesting totest the fidelity of the promoter region of Fli-Vancouver to eliminate thepossibility of a defect in transcriptional regulation, however, this will bediscussed further with respect to FII-Utrecht.B. FII-UtrechtFII-Utrecht presents an interesting case for a prothrombin deficiency.Analysis of the exons and the intron-exon splice regions revealed nosignificant changes which could be considered causative for this condition.Because this condition generates very little protein, it may be expected thatthe mutation lies somewhere in transcriptional regulation. There were nochanges in the signal sequence, or any that would cause aberrant folding;therefore, it could be that the mutation lies in the upstream promoter orenhancer regions. It is possible that there are global changes within theintrons such as the deletions of important regulatory regions that we areunaware of, or the generation of a cryptic splice site within an intron.Although the presence of a cryptic splice site would cause a severe disruption61in the protein, it would not account for all of the mRNA generated. Thefidelity of the intron-exon splice regions would allow for generation of at leastsome properly spliced mRNA, and we would therefore expect to see proteinlevels that are moderately higher than those observed in this patient. Thesequence changes observed in both this patient and in the EU-Vancouverpatient, were not expected to have a serious effect on the protein, and weretherefore considered to be insignificant.Very little is known about the transcriptional regulation of liverspecific genes. In addition, precious little is known about the regulatoryregions in prothrombin. It is speculated that the unidirectional transcriptionof the prothrombin gene is regulated by a proximal promoter, and regulatedby more distal enhancer elements (Chow et al., 1991). It is possible, therefore,that the mutation causing a FIT deficiency in FII-Utrecht is the result of adisruption in transcription. There are such reported cases of deficiencies indifferent proteins. Hemophilia B Leyden was identified by point mutationsin the 5’ flanking region of FIX at positions -20, -6, or +13, relative to thetranscription start site. The mutation at +13 has been shown to affect C/EBPbinding activity, and is therefore expected to play a role in expression(Crossley and Brownlee, 1990.)DNase footprint analysis has shown that a number of liver specificproteins bind to the upstream regulatory sequence URS of humanprothrombin (Chow et al. 1991). A consensus sequence for binding thetranscription factor HNF-1 has been identified, as well as a GC rich regionflanking the site. Mutations in the (URS) of prothrombin could potentiallyhave one of at least three effects. The mutation could disrupt the binding of apositively acting trancription factor. The mutation could introduce abinding site for a factor that would not normally bind, which could62competitively inhibit the binding of Fil-specific factors. Alternatively, themutation could cause a change in the secondary structure of the gene, byaltering the three dimensional stereospecific interactions of concurrentlybound transcription factors. All of these changes would disrupt the positiveeffects of the enhancer, and subsequently, the promoter would produce onlyminimal transcripts. This would be in accordance with the minimal proteinlevels observed in this patient.The existence of a GC rich region in the promoter region couldpotentially be a hot spot for mutations in this gene. Recurrent mutations inhaemophilia A as well as the predominance of reported mutations inarginine residues (due to the GC dinucleotides in this codon) ofprothrombin are indicative of this phenomenon. Deamination of methylatedcytosine to thymine results in CpG -> TpG.4. FUTURE STUDIESThere is a great deal about the hemostatic process that is poorlyunderstood including the complex regulation of the blood clotting cascadethat allows enhancement and attenuation of its response, and the specificmechanisms of activation of the individual proteins. Many of the strategiesemployed to elucidate the structural and functional relationships of theindividual proteins have involved the use of artificial systems. X raycystallographic data obtained from one protein is extrapolated to model asimilar protein or protein region. Many proteins are studied by examiningthe properties of degraded, or chemically modified forms. Although thismethod has its uses, it is limited by the extent of the modification, and allowsonly partially characterization of the degraded forms. The incidence ofnaturally occurring mutants within a population offers a tremendous63opportunity to study protein function. Determination of the defect facilitatesidentification of structural features which may be involved in the activationprocess, assembly, or substrate recognition.Recombinant DNA technology has greatly facilitated the study ofaltered forms of proteins. When recovery of a mutant protein is hampered byinsufficient plasma levels of the protein, the altered protein can be generatedin the laboratory. Recombinant Fil has been produced in mammalian tissueculture using baby hamster kidney (BHK) cells (Le Bonniec et al, 1991), andChinese hamster ovary (CHO) cells (Jorgensen et al., 1986). Insertion of theidentified mutation can be accomplished by site-directed mutagenesis usingthe PCR.The expression of mutant FIT in cases where there is little proteinsynthesized in vivo is not a useful approach in terms of functional studies.Alternatively, these cases present a unique opportunity to studytranscriptional regulation and or properties of folding and transport withinthe cell. Since little protein is expressed naturally, it would be difficult toproduce the same recombinant protein in tissue culture. Degradation wouldprobably occur before the protein could be secreted. Mutations found in theexons, or corresponding splice regions, such as FIT-Vancouver, would beexpected to cause a disruption in folding of the subsequent protein. It wouldbe interesting to follow such a protein through the cell usingimmunofluorescent antibodies to see if the protein, if not immediatelydegraded, is shuttled to a different part, or trapped in a certain region of thecell. Mutations found in promoter or enhancer regions, allow a study oftranscriptional regulation of prothrombin. Very little is known about of theliver specific expression of FIT, or of any of the liver specific genes. A strategyto study cis-acting regulatory elements can be facilitated by expression of64putative promoter and enhancer regions, ligated to reporter genes such aschioramphenicol acetyltransferase (CAT), or human growth hormone (HGH).Comparison of the relative expression levels of FIT-deficient promotersrelative to normal FIT may elucidate functional regions for transcription.This study could facilitate identification of binding proteins which may beessential for adequate expression of this protein, which in turn may beextrapolated to other liver specific proteins.There are also some obvious clinical benefits to studying FIT mutants.Diagnosis for inherited defects of prothrombin can be facilitated usingoligonucleotide primers as probes. Oligonucleotides can be designed carryingan observed mutation, which may be prevalent in the population. Byaltering hybridization and washing conditions, a normal FIT allele can bedistinguished from the mutant in a Southern blot analysis. This procedurewould be useful for carrier detection and pre-natal diagnosis.65REFERENCESAtkinson, T., Smith, M., (1984) Oligonucleotide Synthesis: a practicalapproach. M.J. Gaitland, Eds. pp 70-72 IRL Press, Oxford.Bancroft, J., McDowell, S., Degan, S.J. (1992) Biochemistry 31: 12469-12476.Bancroft, J.D., Schaefer, L.A., Degan, S., (1990) Gene 95: 253-260.Banfield, D.K., Irwin, D.M., Waltz, D.A., MacGillivray, R.T.A., (1993) J. Mo!.Evol. in press.Baudo, F., De Cataldo, F., Josso, F., Silvello, L., Acta Haemat. (1972) 47: 243-249.Bezeaud, A., Drouet, L., Soria, C., Guillin,M-C., (1984) Thrombos. Res. 34:507-518.Bezeaud, A., Elion, J., Guilin, M., (1988). Blood 71: 556-561.Bezeaud, A., Guillin, M.C., Olmeda, F., Quintana, M., Gomez, N., (1979)Thrombos. Res. 16: 47-58.Bode, W., Mayr, I., Baumann, U., Huber, R., Stone, S.T., Hofsteenge, J., (1989)EMBO J. 8: 3467-3475.Borowski, M., Furie, B.C., Bauminger, S., Furie, B., (1986) J. Biol. Chem. 261,14969-14975.Breathnach, R., Benoist, C., O’Hare, K., Cannon, F., Chambon, P., (1978) Proc.Nati. Acad. Sci. USA 75: 4853-4857.Butkowski, R.J., Elion, J., Downing, M.R., Mann, K.G., (1977) J. Biol. Chem.252: 4942-4957.Chow, B.K.C., Ting, V., Tufaro, F., MacGillivray, R.T.A., (1991) J. Biol. Chem.266: 18927-18933.Chung, K., Bezeaud, A., Goldsmith, J.C., McMillan, C., Menache, D., Roberts,R., (1979) Blood 53: 776-787.66Clark, B.A., Maslen, C.L., Sakai, L.Y., Dhalimi, M.Al., Lift, R., Lift, M., (1991)Nucleic Acids Res. 19: 4309.Colman, R., Marder, V., Salzman, E., Hirsh, J., Hemostasis and Thrombosis.Basic Principles and Clinical Practice: Overview of Hemostasis. 2nd ed.(1987). J.B. Lippincott Co. Philadelphia. pp. 3-17.Crossley, M., Browniee, G., (1990) Nature 345: 444-446.Davie, E.W., Fujikawa, K., Kisiel, W., (1991) Biochemistry 30: 10363-10370.Davie, E.W., Fujikawa, K., Kurachi, K., Kisiel, W., (1979) Adv. Enzymol. 48:277-318.Davie, E.W., Ratnoff, O.D., (1964) Science 145: 1310-1312.Degan, S., Davie, E., (1987) Biochemistry 26: 6165- 6177.Degan, S., Schaefer, L.A., Jamison, C.S., (1989) Thromb. Haemostas. 62 : 153.Degan, S.J., MacGillivray, R.T.A., Davie, E.W., (1983) Biochemistry 22: 2087-2097.Diuguid, D.L., Rabiet M., Furie, B.C., Furie, B., (1989) Blood 74: 193-200.Esmon, C.T., Jackson, C.M., (1974) J. Biol. Chem. 249: 7791-7797.Ferlund, P., Stenflo, J., (1965) J. Biol. Chem. 50: 6125-6133.Furie, B., Furie, B. C., (1988) Cell 53: 505-518.Girolami, A., Bareggi, G., Brunetti, A., Sticchi, A., (1974) J. Lab Clin. Med. 84:655-666.Girolami, A., Brit. J. Haematol. (1971), 21: 695-703.Girolami, A., Coccheri, S., Palareti, G., Poggi, M.,Burul, A., Cappellato, G.,(1978) Blood 52: 115-125.Girolami, A., Zanon, R., (1988) Amer. J. Hemat. 27 : 306-307.Henriksen, R., Owen, W., Nesheim, M., Mann, K. (1980) J. Clin Invest. 66:934-939.Henriksen, R., Owen, W.,J. Biol. Chem. (1986) 262: 4664-4669.67Huisse, M., Dreyfus, M., Guillin,M., (1986) Thrombos. Res. 44: 11-21Inomoto, T., Shirakami, A., Kawauchi, S., Shigekiyo, T., Saito, S., Miyoshi, K.,Morita, T., Iwanaga, S., (1987) Blood, 69: 565-569Iwahana, H. Yoshimoto, K., Itakura, M., (1992) Hum. Genet. 89 : 123-124.Iwahana, H., Yoshimoto, K., Shigekiyo, T., Shirakami, A., Saito, S., Itakura,M., (1992) Am. J. Hum. Genet. . 51: 1386-1395.Jackson, C.M., (1981) Hemostasis and Thrombosis: Mechanisms ofProthrombin Activation. Bloom and Thomas, Eds., ChurchillLivingston, Edinburgh. pp 135-147.Jackson, C.M., Nemerson, Y., (1980) Annu. Rev. Biochem. 49 : 765-811.Jorgensen, M.J., Cantor, A.B., Furie, B., Furie, B., (1986) J. Biol. Chem. 262:6729-6733.Josso, F., Beguin, S., Rio, Y., Haemostasis (1982) 12: 309-316.Josso, F., Monasterio DE Sanchez, J., Lavergne J.M., Menache, D., Soulier, J.P.(1971) Blood 38: 9-16.Kahn, M.J.P., Govaerts, A., (1974) Thrombos. Res. 5: 141-156.Kung, Wei-jen Hu., Tulinsky, A., (1980) J. Biol. Chem. 255: 10523-10525.Le Bonniec, B.F., MacGillivray, R.T.A., Esmon, C.T. (1991) J. Biol. Chem. 266:13796-13803.Mac Gillivray, R.T.A., Cool, D.E., Fung, M.R., Gumto, E.R., Koschinsky, M.L.,Van Oos, (1988 ) Genetic Engineering Principles and Methods. PlenumPress. New York and London. pp 265-331.Mac Gillivray, R.T.A., Irwin, D.M., (1986) Haemostasis 16: 227-238.MacFarlane, RG., (1964) Nature 202: 498-499.MacGillivray, R.T.A., Davie, E.W., (1984) Biochemistry 23 : 1626-1634.MacGillivray, R.T.A., Fung, M.R., Hay, C.W., Irwin, D.M., (1984) Bulletin ofthe Canadian Biochemical Society 21: 25-32.68MacGillivray, R.T.A., Irwin, D.M., Guinto, E.R., Stone, J.C., (1986) Ann. N.Y.Acad. Sci. U.S.A. 485: 73-79.Magnusson, S., Petersen, T.E., Sottrup-Jensen, L., Claeys, H., (1975) inProteases and Biological Control (Reich, E., Rifkin, D.B., & Shaw, E.,Eds) pp 123-149, Cold Spring Harbor Labs, Cold Spring Harbor, NY.Maniatis, T., (1989) Molecular Cloning. A Laboratory Manual. 2nd Ed. ColdSpring Harbor Laboratory Press, Cold Spring Harbor, N.Y., U.S.A.Maniatis, T., Goodbourn, S., Fischer, J.A., (1987) Science 236: 1237-1244.Mann, K.G., Elion, J., Butkowski, R.J., Downing, M., Nesheim, M.E., (1981)Meth. Enzymol. 80: 286-301.Mann, K.G., Nesheim, M.E. Tracey, P.B., Hibbard, L.S., Bloom, J.S., (1982)Biophys. J. 37: 106-107.McAlpine, P.J., Dickson, M., Guy, C., Wiens, A., Irwin, D.M., Macgillivray,R.T.A., (1991) Nucleic Acids Res. 19: 193.Miyata, T., Aruga, R., Umeyama, H., Bezeaud, A., Guillin, M., Iwanaga, S.,(1992) Biochemistry 31: 7457-7462Miyata, T., Morita, T., Inomoto, T., Kawauchi, S., Shirakami, A., Iwanaga, S.,Biochemistry (1987) 26: 1117-1122.Montgomery, R.R., Corrigan, J.J., Clarke, S., Johnson, J., (1980) Circulation 62(Suppl 3): 279.Montgomery, R.R., Otsuka, A., Hathaway, W.E. , (1978) Blood 51: 299-306.Morishita, E., Masanori, Saito, Hidesaku, A., Hiroshi, J., Uotani, C.,Kumabashiri, I., Yamazaki, M., Hachiya, H., Okamura, M., Matsuda, T.,(1991) Thrombos. Res. 62: 697-706.Morishita, E., Saito , M., Kumabashiri, I., Asakura, H., Matsuda, T.,Yamaguchi, K., (1992) Blood, 80: 2275-2280Morrison, S.A., Esnouf, M.P., Nature New Biol. 242: 92-94.69Mount, S. M. (1982) Nucleic Acids Research 10: 459-472.Naito, K., Fujikawa, K., (1991)1. Biol. Chem. 266: 7353-7358.Nelsestuen, G.L., Zytkovicz, T.H., Howard, J.B. (1974) J. Biol. Chem. 249:6347-6350.Owen, W.G., Esmon, C.T., Jackson, C.M., (1974) J. Biol. Chem. 249: 594-605.Padgett, R.A., Grabowski, P.J., Konarska, M.M., Seiler, S., Sharp, P.A., (1986)Annu. Rev. Biochem. 55: 1119-1150.Park, C.H., Tulinsky, A., (1986) Biochemistry 25: 3977-3982.Quick, A.J., Hussey, C.V., (1962) Lancet 1: 173-177.Quick, A.J., Pisciotta, A.V., Hussey, C.V., (1955) Arch. mt. Med. 95: 2-14.Rabiet, M., Elion, J., Benarous, R., Labie, D., Josso, F., (1979)Biochim. etBiophys. Acta 584: 6 6-75.Rabiet, M., Furie, B. C., Furie, B., (1986) J. Biol. Chem. 261: 15045-15048Rabiet, M.J., (1985) Thrombos. Haemost.as. 554: 46.Rabiet, M.J., Elion, J., Labie, D., Josso, F., (1979) FEBS Letters 108 : 287-291.Rabiet, M.J., Jandrot-Perrus, M., Boissel, J.P., Elion, J., Josso, F., (1984) Blood63: 927-934.Roberts et a!. Hemophilia and Hemostasis (1981). Alan R. Liss, Inc., 150 FifthAye, New York, NY. pp 85-102.Roberts, H.R., Foster, P.A. , Hemostasis and Thrombosis. Basic Principles andClinical Practice. 2nd ed. (1987). J.B. Lippincott Co. Philadelphia. pp.162-177.Rolye, N.J., Irwin, D.M., Koschinsky, M.L., MacGillivray, R.T.A., Hamerton,J.L., (1987) Som. Cell Mol. Genet. 13: 285-292.Rubio, R., Almagro, D., Cruz, A., Corral, J., (1983) Brit. J. Haematol., 54: 553-56070Ruiz-Saez, A., Luengo, J., Rodriguez, A., Ojeda, A., Gomez, 0., Acurero, Z.,(1986) Thrombos. Res. 44: 587-598Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A.,Arnheim, N., (1985) Science 230: 1350-1354.Sanger, F., Nicklen, S., Coulson, A.R., (1977) Proc. Nat!. Acad. Sci. U.S.A. 74:5463-5467.Seegers, W.H., (1979) Prog. Chem. Fibrinol.Thrombos. 4: 241-254.Shapiro, S., Martinez, J., Holburn, R., (1969) J. Clin. Invest. 48: 2251-2259Shapiro, S.S., Martinez, J., (1969) J. Clin. Invest. 48: 1292-1298.Smith, L.G., Coone, L.A.H., Kitchens, C.S., (1981) Amer. J. Hematol. 11: 223-231.Stanton, C., Taylor, R., Wallin, R., (1991) Biochem. J. 277: 59-65.Stroud, R.M., (A Family of Protein-Cutting Proteins)Suttie, J.W. (1973) Science 79: 192-194.Suttie, J.W., (1993) FASEB J. 7: 445-452.Tapon-Bretaudiere, J., Dumont, M-D., Fischer, A.-M., Bros, A., Chassevent, J.,Aufeuvre, J-P., (1983) Thrombos. Hemostas. 50: 250.Valls-de-Ruiz, M., Ruiz-Arguelles, A., Ruiz-Arguelles, G., Ambriz, R., (1987)Amer. J. Hematol. 24: 229-240Vetten, M., Ploos van Amstel, H.K., Reitsma, P.H., (1990) Nucleic Acids Res.18: 5917.Walz, D.A., Hewett-Emmett, D., Seegers, W.H., (1977) Proc. Nat!. Acad. Sci.U.S.A. 74: 1969-1972.Weinger, R., Rudy, C., Moake, J., Olson, J., Cimo, P., (1980) Blood 55: 811-816.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0094716/manifest

Comment

Related Items