.~4_-.__-_ “4.11. .3 .a‘: *1“— :;. '5: " ”4‘6: ;_:»...‘ ._. 4 ‘ “‘3 --.a..“‘ .3. Lg1‘) “Q3111“: ‘1‘? “1“,. " f, .15. up!” 1' 7'” 2: . 1 u?" 1 1 31:11..J ‘1‘411‘, 1 at”) ‘Lqflj’u \E ,‘fufitv‘f‘ ”“3 km pun- 1" ’33:“) @792 1st -‘.-...‘\.1 511113333 :1 .. .~. ‘21:“ 111.9 . 4 if. , 5‘11“, “I , _, .. . : . 3‘ . . .. 7’ ' "1' U M’( 11) Mgmyé‘ . . . ' . . - ”‘1‘ 1‘ . .n".:.;;.t..x.1., 5‘"? ‘(mg‘i .n« 1 . 1‘. L N1291- “Mg; .1" h ‘V1‘H H131 ~ "1335 33‘ ‘11 . *3... .1 1 v a. 233 . 1 31““‘11‘ “ «3:3 . ' ‘11.” Vl‘lV' : ‘. V .1: “- '. . \ I . Nu ‘ . ‘. "-11’ ‘: "!;1i§;t‘§fi‘ri§“‘ ..‘G .1 1 1 fih‘i‘alr‘f‘. :fi“ ‘V ‘ N 111“ .3’ ~ . U. .- J“ ahu‘gfi‘}. ‘_ .V ,id". 1' ' ht‘h“ ‘ 5:" _’ I. “I?" “-':.I ‘. ‘1: [‘3 7' 13,... 1. .11.. 1‘1V‘Lg‘fil“ ;‘ 2‘!‘%‘....13;;3;4. i‘ =3}3. - ' 25‘?“ “as? 2' u. v 471:“ s... .:« . 3-5;: *3 =7; Jag. 3:159; -— ‘2..- .. r;.z:.-.~.’ ' :3“:~"?§E+ :4. ‘7‘ «if! . - ....;- 3‘. . A ' '7 - ’7}; .. .2}... “Zn. : r70 A 1 1 3 L“ 3 ._ q . . 15‘ ‘3'; Lu. W ‘11:“? “I?“ ”3314' 1‘3"] 1'3. . - . - ‘1. _ 1%.“? 53:55”:- ‘E - *“éuic. “‘ g ‘ . ..1.‘:§}m .‘ . _ < . . ..<.v . 1 ‘ I . 7 ”J:: kl 13%;: 1 _. ‘ 35:" ! . "‘Rht‘. 1] “35133:” ““1“ “”< A. -v Era “ya-"fig. ':7.1-1 '3 5112“- a . fl 7.:- ... , . . ‘ . 1‘72: 7 ' . ‘5” _ ' ’33:: 1321' - 5E. ._ a; .22: ..‘z::_&"““‘:;,€=z';4=£ fl} .- ‘.'.. . i - .9. V' _; ~1§a > "L. t .._..." -xizrii“;: t4 5"" .. :11 ” ""111 i" -' 3“" 1:“ 1‘ ‘1 'I ‘ ' 1 . . “ ~ . . 3.1} 33‘. 11‘ .V‘ :7 1‘ « .1" .14., " , g .335" , 11—"; L, A... ._ . :__. v... =3 * 4. ”A - .- ”,3. ’22-‘31- 3::- - ->%L ‘.‘: .—— .._. . .‘7 . J.“ .._-.._... '4'“ n O 1.2.1...- . w- ‘13.“. 2.. __z. . P‘ . L“ 1‘? 1 , c “in“ J": . .1 .134 ‘- ‘1 I I; of .. «wry:- :27»- .51.. .1. .131} L E4 I r1: 1 3:...2"... » .32. .. . 35.}; 9.15.} ‘4‘ W .._. :23; .3: :3? waiw 1%; .42. $1.15.... a a“; .._»? ‘ ‘3' a"! L. ”:3 5.1“: ..v: 7:}:- a . - A LEE-u ,x €sz .3553. _ my? gt;— £2: . ‘1:- ~33?” *2. a... . Rigs—=1: . 2&4: uCaZu”“‘ . w - .‘ Jim: _ -: .— i- " ‘ I ,_ ‘ 5—31.: ‘9 .. M . ":1: ‘1. ft s3- - “343%.:- :ts‘P: JP? ‘5; .. - ‘F . A .3. .Ag?‘ .. £§~_ . ‘1‘1‘ 3“" +11), . 1 «V v 131 1 ‘11‘ ”3‘7“!“ W '2‘ “If! "V"I‘nf‘4u~k.‘:j;11 ‘ m1. .... 1. “V‘ .3331 mg? .‘1 .1621 ii ‘i‘1'".‘é‘-“5“§ 4.“ sniffct' ‘. 2‘ y . . -. ...‘. L ".7 p} "' 'I 1:32; 5.3%.?! 1‘72; 35:: um .. “31333233. - "“Wfic‘é 1.5. 1;"?‘5‘2‘é; 2:. ‘fi‘i‘gmw N52“ VI. .m $2.2“qu 11 Kg} Si. 1:53 41.3” salt: {It ‘1“. pg“: ht." ..‘g 5‘" h -. fix... LC ..u, 11‘3"; bx.p%uh, ‘ 5'1? “5‘ '1: . “5331". ‘11-’31 “ ... . 1:: 1 .4 7h“ Ih:‘\=: x 1-7:!“ “‘5 s . ,. 1.; -: - 4‘ ‘I-.(V_ L1 ‘ an... “ .fi - 1. "3.3 . «14-... 11W .. *131' '- 1 - ..V k Hu‘A‘ I i".‘§;‘ "i . > .r‘jF—‘r‘ v . 3%? “Iii“ . 30231:" ‘ 5‘73? ‘ . Jub— . m . _ “Sf - magnate? . . 592;...» fig: 2:" w :17»??? .2.” _ _w., 21} - 3 .w. r .f .5: ' .;. '53:. . ; '1; V L _ ' .5 ‘ {J .4 7 .. 3r“ . 11 L :4 a..- I‘ . ’ 355;: A ' ‘ _. ‘3‘ .2 . mxi‘Tfif TAN ||\\\|\\\\|\\\\|\l\WWII 3 1293 109 LIBRARY Michigan State A ' University ‘ This is to certify that the dissertation entitled APPARENT PROTHROMBINASE ACTIVITY OF HUMAN FACTOR X: A MECHANISM FOR THE FACTOR VIII INHIBITOR BYPASSING ACTIVITY OF PROTHROMBIN COMPLEX CONCENTRATES presented by Douglas William Estry has been accepted towards fulfillment of the requirements for Ph . D. degree in Pathology flaw/K 724% Major professor Garson Tishkoff Date May 7, 1985 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LlBRARlES RETURNING MATERIALS: Place in book drop to remove this checkout from your record; FINES will be charged if book is returned after the date stamped below. JUI26 '87 fi :37 K209 ~ War. ‘19.. fl APPARENT PROTHROMBINASE ACTIVITY OF HUMAN FACTOR X: A MECHANISM FOR THE FACTOR VIII INHIBITOR BYPASSING ACTIVITY OF PROTHROMBIN COMPLEX CONCENTRATES BY Douglas W. Estry A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pathology 1985 This dissertation is dedicated to my wife Suzanne and to my children. They endured my anxieties and frustrations and have continued to give me their love and support. ii ACKNOWLEDGMENTS I would like to extend my special thanks to Dr. Garson H. Tishkoff, director of my dissertation. He has provided not only the proper physical environment for conducting my research but he has also provided valuable scientific input coupled with a knowledge of the direct clinical applicability of this work. His distinguished background in research has made him an invaluable source of information and provided me with the necessary constructive insight important to a critical evaluation of any research project. I would also like to thank the members of my committee, Dr. '1‘. Bell, Dr. H. Bowman and Dr. K. Schwartz for their time, input and both technical and clinical expertise. Finally my thanks to Dr. Clarence Suelter for his technical advice and suggestions and to The Great Lakes Regional Red Cross Blood Center for their continued support of this work. iii TABLE OF CONTENTS LiSt Of Tables 0 O O O O O O O O O O O O O O O O O O O O O C I O O O C O O O O O O O O O O O O 0 Vi List Of Figures 0 O O O O O O O O O O O O O O O O O O O O O O 0 O O O O O O O O O O O O O O O O Vii Literature ReView O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O C O O O O O 0 Characterization of the Factor VIII Antibody ...... IntrOduCtion OI.OOOOOOOOOOOOOOOOOOOOO0.0.0.... InCidence O0..OOOOOOOOOOOOOOQOOOOOOOO0.0.0.... Antibody Characteristics ..................... Development of an Inhibitor .................. Kinetics of the Inhibitor Response ........... QGle-‘H H Inhibitor Assay Techniques ........................ ll Theraputic Implications ........................... 14 Immunosuppression ............................ 15 PlasmaphereSis .0...000.000.00.00.0.00.0000... 15 ImmunOtOIerance 0.0...OOOOOOOOOOOOOOCOOOOOOOOO 16 Porcine Factor VIII .......................... 17 Prothrombin Complex Concentrates .................. 18 Testing Procedures ........................... 19 Adverse Reactions ..OOOOOOOOOOOOOOOOOOOOOOO... 20 Hepatitis OOOOOOOOOOOOOOOOOOOO0.00.0.0.0. 20 Acquired Immunodeficiency Syndrome ...... 21 Thrombotic Complications ................ 22 Clincial Efficacy of Treatment with Prothrombin Complex Concentrates ........... 23 Factor VIII Bypassing Activity .................... 26 Prothrombinase Complex ............................ 30 VitaminK O...0.0.0.000000000000000.0.0.000... 31 Calcium Binding and Protein Conformation ..... 33 Factorv0.......00....OOOOOOOOOOOOOOOOOOOOOOO 37 Phospholipid: Surface Assembly ............... 38 Materials and "ethOds OOOOOOOOCOOO0.00.00.000.000.00.... 41 Chromatography .0.0.00000000000000000000000.0.0.... 41 iv Proteins .......... Factor II, X and IX Purification ............. Factor VII Purification ...................... Factor Xa .... Factor V ..... Antithrombin I Phospholipids ..... II and Fibrinogen ......OOCOOOOC Additional Reagents 00............OOOOOOOOOOOOOOOOO Clotting Factor and FEIBA Assay Procedures ........ Prothrombin Aetivation Assay ......OOOOOOOOOOOOOOOO ReSUItS ......OOOOOOOOOO ......OOOOOOOOCOOOOOO000......O. Protein Isolation and Characterization ............ Prothrombin Complex Proteins ................. Factor V ..... Factor VII ... Amino Terminal In Vitro Characteri Inhibitor Bypass FEIBA Assay .. Effect of AT I sequenCing ......OOOOOOOOIOOOOO zation of the Factor Eight ing ACtiVity ....OOOOOOOOOOOOOO. II on FEIBA ......OOOOOOOOOOOOOO Effect Of p-APMSF on FEIBA O O O O O O O O O O O O O O O O O O 0 Effect of Prothrombin Complex Proteins on FEIBA O O O O O O O O O O O O O O O O O O O O O O O O O Prothrombinase Assay .0.00......OOOOOOOOOOOOOOOOOOO Kinetics of the Prothrombinase Complex ....... Lipid Binding Inhibition of StUdy ......OOOOOOOOO0.0.0.0.... Prothrombinase Activity With p-APMSF 0............OOOOOOOOOOO0.0... Inhibition of with AT III DiSCUSSion ......OOOOOO. Appendix OOOOOOOOOOOOOOO List of References Prothrombinase Activity ............OOOOOOOOO...0.0.0.0000... 41 42 46 47 49 49 50 50 51 52 54 54 54 56 56 57 58 58 60 61 62 65 65 68 70 71 78 86 94 Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7 Table 8 Table 9 LIST OF IABLBS Structural Characteristics of Factor VIII Antibodies ................... New Oxford and Bethesda Antibody Assays ........ Protein Constituents of Factor IX Concentrates . Biological Probes for Assessing FEIBA.um"."~. Factor Concentrations @f Autoplex® and Immune FEIBA .....OOOOOOOOOOOOOOOOOOO Amino Terminal Sequence of Factor x ............ Effect of At III-Sepharose on Factor x and Factor Xa FEIBA ...................... Effect of p-APMSF on Factor X and FaCtor xa FEIBA .....OIOOOOOOOOOOOOO...... Effects of II, Ix and VII on the FEIBA Assay ... 5 12 19 20 27 58 61 63 64 Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure LIST OF FIGURES l Carboxylation of glutamic acid ............... N Structure of glutamic acid and y-carboxyglutamic acid .................. Prothrombin activation fragments ............. Prothrombin complex proteins ................. FaCtorvcoco0.0000000000000000000000000000coo O‘U‘lbw Corrective effect of factor Xa and Factor x on APTT of factor VIII Inhibitor Plasma . 7 Double recriprocal plot of the initial velocity as a function of the prothrombin concentration................ 8 Double reciprocal plot of the initial velocity as a function of the prothrombin concentration................ 9 Inhibition of factor Xa enzymatic activity by factor x zymogen ............ 10 The effect of the inhibitor p-APMSF on the time course prothrombin activation by factor Xa or factor X zymogen ........... 11 The effect of AT III-Sepharose on the time course of prothrombin activation by factor Xa or factor X zymogen ........... Vii 67 68 70 74 75 LITERATURE REVIEW CHARACTERIZATION OF THE FACTOR VIII ANTIBODY Introduction Antibodies to factor VIII have been demonstrated in two distinctly different patient populations: Inhibitors that develop in patients with inherited severe hemophilia A (factor VIII defeicency) and inhibitors that arise spontaneously in non-hemophilic patients such as, postpartum complications, certain immunologic disorders and in older individuals as a spontaneous complication. These clinical subgroups must be kept in mind when the antibody response to factor VIII is being evaluated. Although the antibody produced in both cases is usually of the immunoglobulin G class, there appears to be some difference in subclass type, light chain specificty and in the reaction kinetics of the antibodies produced. In hemophilia A second order inactivation kinetics are more common while more complex kinetics with incomplete inactivation is characteristic of antibodies developing in non-hemophilic patients. Both of these characteristics will be discussed in more detail. Three important considerations regarding antibodies to factor VIII include the incidence, properties and etiology of the factor VIII inhibitor. Because of the size and complexity of the factor VIII molecule there are a great number of potential antigenic determinants. For this reason the characteristics of the factor VIII inhibitors, although not unique, are extremely complex. As of yet, no definitive answer has been found that thoroughly explains the etiology and characteristics of the immune response to factor VIII. Incidence It is important, when discussing the incidence of factor VIII inhibitors in the hemophilic population, to remember that development of an antibody is the result of multiple immunizations with an isologous plasma protein component in response to the need for replacement therapy. Therefore, the incidence of antibody development is correlated with the total number of exposure days to factor VIII prior to its appearance. In light of this, various statistics on the incidence of factor VIII inhibitors in different age groups and between the "at risk“ population of severe hemophiliacs (< 1% factor VIII) and those with a mild deficiency (2% to 30% factor VIII) will be discussed. In 1975 Ruggeri reported an inhibitor incidence of 13% in the "at risk“ population of hemophiliacs (1). However, when broken down into the age groups 0-9, 10-24, and over 25 years of age the percentages for the same “at risk" population were 22%, 12% and 6% respectively; This suggested an increased risk at a younger age. If,however, these results were evaluated strictly on the basis of the number of exposure days to factor VIII prior to the development of an inhibitor I'predisposed" individuals would be selected out at a younger age and the value of 6% in the 25 and over age group would reflect a statistically biased figure. In a report on the incidence of inhibitors among hemophiliacs a value of 14.6% was given for the incidence of inhibitors in patients under 10 years of age as opposed to an overall value of’7}8% for the total group studied (2). However, in a large national study Biggs reported an incidence of only 4.1% for the same age group (3). (Again, it would be valuable to relate total number of exposure days prior to development of an inhibitor in both of these reports. Likewise, the incidence that has been reported- based on the difference between a severe and mild hemophiliac would be more meaningful if evaluated in terms of the total number of exposure days. Despite the variability in the reported incidences of inhibitors to factor VIII the majority of these inhibitors occur in severe hemophiliacs with a frequency of 5 to 18%. Antibody Characteristics Based on the molecular sieving characteristics on Sephadex gel, the fractionation pattern on DEAE cellulose, the ability to be precipitated the protein with high salt concentrations, the electrophoretic pattern and the ability to neutralize the inhibitor activity with antisera specfic for IgG, inhibitors to factor VIII have the same immunochemical properties as other immunoglobulins of the IgG class (4). Initial work on the antibody structure of inhibitors to factor VIII suggested that the apparent homologous antibodies were of monoclonal origin (5,6). They were identified as having a single kappa light chain specificity and a similiar homogeneity of the heavy chain, specfically IgG4. This heavy chain subclass is consistant with observation that the factor VIII antibodies do not fix complement. More recently a limited number of patients demonstrating a combined light chain population of both kappa and lambda and a mixture of predominately IgG3 and IgG4 heavy chain subclasses have been reported (7,8). This heterogeneous population of IgG antibodies appears to be more characteristic of cases of inhibitor development in non-hemophilia A patients although a broad generalization should be avoided. Table 1 is a summary of data compiled from several different studies on the structure of the factor VIII antibodies. Although the majority of the inhibitors are 196 in nature and exhibit restricted heterogeneity in heavy chain subclass and light chain type, IgM antibodies have been reported particularly in the non- hemophilic population (9). The majority of the antibodies produced in response to factor VIII are specific for the procoagulant portion of the factor VIII molecule and have no effect on other clotting factors (10). The inhibitors do not interfere with the other functional activities of the factor VIII complex. Factor VIII related antigen and bleeding times are equivalent to the values found in hemophiliacs without an inhibitor. There is normal platelet retention on glass beads and normal ristocetin aggregation and ristocetin cofactor activity. In addition, factor VIII inhibitors show a great deal of species specificity. For example, bovine and procine factor VIII are only partiallyinhibited by antibody to human factor VIII and as a result hemophiliacs with an inhibitor can be treated effectively with factor VIII from these animals (11). As would be expected, however, antibodies to the animal protein‘will develop and, as a result, antibodies with two specificities can then be demonstrated in the human serum. Table 1. Structural Characteristics of Factor VIII Antibodies. Un- IgG Light Chains Patients Number specified 1+4 3+4 4 k* 1* k+l Hemophilic 15 7 5 3 - 10 l 4 Nonhemophilic 14 ll 2 — l 6 2 6 Hemophilic or nonhemophilic 10 10 - - - 10 - - W k = kappa, l = lambda From: Mariani, Russo and Mandelli (12) Antibodies formed in response to factor VIII are non- precipitating. This may be due to the fact that the factor VIII:C is present in too low a concentration to form a visible precipitate and/or either the antigens are monovalent or the antigenic determinants lie far apart and are therefore not capable of forming a lattice to permit precipitation. The fact that VIII:C can be measured by a two-site immunoradiometric assay would rule against the possibility of a monovalent antigen. Finally, as is the case in the production of monoclonal antibodies, the more restricted the antibody response the less likely it is to form a precipitate. Development Of An Inhibitor A major question that remains unanswered pertains to the mechanism responsible for the development of a factor VIII inhibitor and why only 5 to 18% of the multitransfused hemophiliacs develop an antibody. Meyer pr0posed two hypotheses that might explain the genesis of factor VIII antibodies (13). The first is based on the genetic control of the general immune response and involves antigen recognition, macrophage processing and antibody production. It is known that hemophilia may occur as a result of either suppression of VIII:C production in which case there is a lack of cross reacting molecules (CRM-), or may result from production of a defective protein (CRM+). It was felt that antibody production would be confined to the CRM- group because they would be most likely to respond to the foreign protein. This explanation did not, however, agree with the observed statistics showing that 80% of hemophiliacs are CRM- and only approximately 10% of these develop an antibody. Alternatively» there may be allotypic forms of the factor VIII molecule. This particular explantion remains speculative because the assay systems are not sensitive enough, at present, to detect single amino acid variations that could account for the allotypic differences. Finally, Meyer suggested, in a second hypothesis, that all severe hemophiliacs show an immune response to factor VIII in the form of low titer and low affinity IgM antibodies that would not be detected. In most cases the immune reponse would be suppressed but in 5 to 18% the IgM response would switch to IgG synthesis. In 1977 Frommel and Allain, in order to determine whether a familial predisposition to become immune to factor VIII did or did not exist, presented their accumulated data on a number of family groups genetically disposed to factor VIII deficiency (14). Twenty five sibships were investigated with an overall incidence of factor VIII antibody of 46%. Of these 25 sibships, concordance for responsiveness was found in 19. Only 6 brother pairs demonstrated nonconcordance. This data suggested that genetic factors could be involved in determining immune responsiveness to factor VIII. If linkage between the immune response to factor VIII and the major histocompatibility complex (MHC) could be shown this would establish genetic control of the antibody response to factor VIII. Frommel et al. reported on HLA antigens and factor VIII based on the following assumption for linkage: If the ability to develop antibody to factor VIII is dependent on the inheritance of a dominant MHC-linked gene, then it would be expected that hemophilic siblings with antibody to factor VIII will share one or two HLA haplotypes, and that non-concordant sibs (responder - non-responder) will differ in one or both haplotypes, unless a cross-over had occurred during meiosis (15). They concluded that although the magnitude of the immune response is under genetic control there was insufficient evidence to support linkage with the NBC and, in addition, that immunity to factor VIII 'does not express preferential association with HLA specificities.‘ Kinetics Of The Inhibitor Response Two additional observations regarding the characteristics of the factor VIII antibody should be mentioned because of the effect they have on determining treatment and evaluating inhibitor assay results. The first of these bears directly on the antibody titer in response to immunologic challenge during administration of factor VIII concentrates. It has been noted that the population of inhibitor patients can be divided into high and low responder groups with an observed high/low ratio of approximately 4:1 (16). These groups are based on the degree of anamnestic response observed as a result of therapy with factor VIII. In the high responder group antibody is discovered after 2 to approximately 50 cumulative exposure days to factor VIII and antibody titer is greater than 5 Bethesda Units and may, upon challenge, go as high as 1000 to 2000 Bethesda Units. In the low responder group antibody does not usually appear until after 40 to 200 cumulative exposure days and anamnestic response is minimal and the antibody may disappear. It should be noted when discussing treatment that low responders can be given factor VIII although larger doses are required. The second characteristic of the factor VIII inhibitor relates to the inactivation kinetics mentioned previously in this section. Not only has a difference in the characteristic response of the factor VIII inhibitor to antigen challenge been noted but, in addition, two inactivation patterns representing two types of antibodies have been distinguished by kinetic analysis (17). Type I antibodies have a linear relationship*when the log of residual factor VIII activity is plotted vs. the concentration of the antibody. If the concentration of antibody is sufficient it will completely neutralize the factor VIII:C activity. Type II antibodies, on the other hand, show a nonlinear relationship when plotted under identical conditions and do not completely neutralize the factor VIII:C activity. Biggs et al. characterized the factor VIII response to both types of antibody (18,19). In the case of Type I reaction they concluded that.. antibodies which give more or less straight line concentration graphs when drawn on log/linear paper may be regarded for practical purposes as if they underwent an irreversible second-order reaction with factor VIII and that even when antibody is not present in excess the concentration graph can be used to give a relative measure of antibody concentration. In the case of Type II antibodies, although they were able to develop a mathematical model to fit the more complex kinetics of these antibodies, a specfic explanation for their mode of action was lacking. It was felt that one possible mechanism would involve an antibody factor VIII complex that retained activity; .An additional hypothesis on the action of Type II antibodies was suggested by Gawryl and Boyer (20). They suggested that the reduced factor VIII:C inhibition could be due to steric interference from the von Willebrand factor portion of the molecules They found that when factor VIII:C activity was seperated from the von Willebrand factor the inhibition kinetics were no longer complex but resembled those of a Type I reaction. Characteristically Type I antibodies show a fairly rapid inactivation of the VIII:C activity which reaches a plateau and upon further addition of factor VIII no additional neutralization is observed. However, type II antibodies demonstrate an initial rapid inactivation followed by a slowing that does not plateau. Upon addition of more factor VIII there is continued slow inactivation (18,19). The Type II reaction is most common in nonhemophiliacs and, according to Allain and Frommel (17), these antibodies have less affinity for factor VIII than those that arise in treated hemophiliacs. The implication in treating Type II individuals lies in the ability to express some factor VIII:C activity even in the presence of excess antibody (21-24). Finally, Croissant et al. demonstrated the presence of two distinctly different antibody populations in patients with autoantibodies to factor VIII indicating at least two different antigenic determinants that are recoginzed in vivo (25) O 10 INHIBITOR.ASSA! TECHNIQUES A number of assay techniques have been devised to quantitate the level of anti-factor VIII:C. These assays range from inhibition of factor VIII:C activity, as in the New Oxford and Bethesda assays, to immuno-radiometric assays (IRMA). Based on the characteristics of the inhibitors to factor VIII one could deduce that the variation in kinetics, which could be a reflection of a decreased affinity for the factor VIII molecule and/or different antibody populations to varying antigenic determinants, could very likely cause difficulty in determining antibody levels. In addition to the inhibitor kinetics, four additional methodological aspects of the inhibitor have been identified by Lechner and must be kept in mind when evaluating inhibitor assays (12). These include: 1) sensitivity; 2) compatibility; 3) reproducibility; and 4) the time course of inactivation. The time course of inactivation refers to the fact that in some patients the inactivation is very rapid, 30 minutes, while in most instances it takes several hours. The measurement of inhibitors by coagulant activity is presently performed by one of two different assay techniques. In 1975 the National Heart and Lung Institute sponsored a meeting to promote the standardization of the measurement of factor VIII inhibitors (26). The proposed assay (Bethesda assay) uses normal pooled citrated human plasma as a standard source of factor VIII and is most accurate in the range of 25% to 75% inhibition. Higher 11 inhibitor levels are diluted to fall in this range. However, more sensitive methods must be found to determine inhibitor levels lower than 25% but still of clinical significance (27). In this assay a test sample producing a residual factor VIII activity of 50% is considered to contain one Bethesda Unit (BU) of inhibitor/ml. The units are arbitrary and do not imply any particular level of factor VIII concentrate that, when infused, would neutralize the antibody. The second method is the New Oxford technique. The New Oxford method differs mainly in the use of factor VIII concentrate as a standardized source of factor VIII, as opposed to normal human plasma. Table 2 outlines both the New Oxford assay and the Bethesda method. Austen reported on a comparative study of the New Oxford technique vs. the Bethesda method (12). Eleven laboratories from different countries were involved in the study and the following conclusions were drawn: Table 2. New Oxford and Bethesda Antibody Assays New Oxford Bethesda Source of Factor VIII VIII Concentrate Plasma Amount of Factor VIII 0.5 to 1.0 U/ml 50% of plasma Incubation period (hrs) 4 2 Antibody Units 1 unit destroys 1 unit destroys 0.5 units of VIII 50% of factor VIII added From: Mariani, Russo and Mandelli (12) 12 1. There is excessive disagreement between laboratories in antibody measurement. 2. This between-laboratories error is the greatest component of the total errors. 3. Generally, laboratories ranked antibody samples in the same order. 4. Considerable improvement (30-40%) would be obtained if a standard antibody were available. 5. In antibody assay, particularly the Bethesda method, a fixed starting level of factor VIII should improve agreement. It should be kept in mind that these assays do not reflect the complex kinetic reactions seen in most patients with autoantibodies to factor VIII. It has been suggested that in the case where several dilutions yield a level of 50% of the control the least dilution giving a 50% value be used to calculate the results (12). Most recently, the results of an IRMA have been reported by Peak et al. and Furlong et al. (28,29). The results, although comparable to the Bethesda method yet more sensitive, vary somewhat based on the high titer anti-VIII antibody used as a standard. This is probably due to a slight variation in affinity as the dilution curves are parallel when percent inhibition is plotted versus the log dilution of antibody. 13 THERAPUTIC IHPLICATIONS The development of an inhibitor to factor VIII presents obvious therapeutic complications and although there tends to be some agreement on the particular course of treatment, there is still considerable disagreement on the best method of preventing an anamnestic response. The treatment protocols presently being used will be briefly discussed and a major emphasis will be placed on the use of the prothrombin complex concentrates (PCC) because of the central role that PCC's played in initiating this work. The recommended therapy varies depending on whether the patient is a high or low responder, whether the antibody developed as a result of an inherited deficiency of factor VIII or as an autoantibody, the severity and location of the bleed and the initial inhibitor titer. A number of clincians feel that, in all cases, the preferred method of treatment is factor VIII concentrate unless the antibody titer is sufficiently high to dictate against this. Additional courses of treatment include plasmapheresis to remove the inhibitor, factor VIII replacement with concomitant administration of immunosuppresants designed to minimize the immune response, use of nonhuman (procine or bovine) factor VIII and the use of either “non-activated“ PCC's or 'activated" PCC's both found to be capable of bypassing the need for factor VIII. 14 Immunosuppression Most often immunosuppressive therapy is used in the case of autoimmune inhibitors in non-hemophilic patients (30, 6, 31). Hultin et al. suggested that the major contributing factors to the effectiveness of immunosuppressive therapy are the titer and duration of inhibitor prior to initiation of therapy and the degree of factor VIII exposure between inhibitor appearance and attempted immunosuppression (32). Although there have been reported incidences of prevention or decrease in the anamnestic response when immunosuppressive therapy is used the majority of the cases of hemophilia A have not responded to immunosuppressive therapy (33,34). The use of immunosuppressive therapy must be evaluated carefully because of the long-term toxicities of these drugs (35). Plasmapheresis Plasmapheresis appears to be a viable option but only in a limited number of cases and only under certain circumstances. 'Considerations include: 1) material to be used in the exchange process, 2) the inhibitor titer and 3) the anamnestic response. Plasmapheresis appears most effective when the inhibitor titer is low and appears not to work when there is an anamnestic rise in the antibody titer. In addition complications associated with plasmapheresis have been anaphylactic reaction, headache, fever, vomiting and hepatitis and several cases of sudden death have been 15 reported (12). In addition to the conventional use of plasmapheresis antibodies have been removed by passing the plasma fraction over a sterilized protein A sepharose column prior to returning the plasma to the patient. This procedure selectively eliminates the IgG fraction and precludes the need to use fresh frozen plasma to maintain adequate hemostasis. Illnnotolerance In 1981 Brackmann and Egli reported the use of high doses of factor VIII concentrate in conjunction with activated factor IX concentrates to treat hemophiliacs with inhibitors (36). The intent of this procedure is to suppress the inhibitor production by inducing immunotolerance. Their protocol consisted of administration of 75-100 U/kg of factor VIII concentrate and 40-60 U/kg activated PCC's twice daily until the inhibitor level fell below 0.5 BU/ml. At this point, the therapy was altered to a regimen in which the same doses were alternated between twice a day and once a day, ‘When the inhibitor titer fell below'0.2'BU/ml each product was given only once daily until inhibitor levels were undetectable» The use of activated PCC's was then discontinued but factor VIII administration was sustained until the in vivo half-life of factor VIII was normal in the treated patients. Additional work using alternate protocols has been reported (37-39). Recently, White et al. reported the effective use of continuous factor VIII administration, without the use of activated PCC's, in 16 reducing a high-titer inhibitor (40). Most recently Aznar et al. reported on the continuous use of factor VIII concentrates and fluprednisolone to suppress the immune response. The corticoid was used as an immune suppressent (41). The main problem with these treatment regimens is the necessity for an extremely large amount of factor VIII. This becomes a drain on limited resources, and is an extremely costly procedure. White et a1. (40) reported the use of 5,695,230 U of factor VIII in one patient totaling approximately $626,500 in the 56 month course of treatment and this figure does not reflect the additional cost of the activated PCC's described in the original protocol of Brackmann and Gormsen (42). Porcine Factor VIII Because of the minimum degree of cross-reactivity between human and porcine factor VIII and because the porcine product has been highly purified to eliminate adverse reactions the product has been used used to treat high titer factor VIII inhibitor patients (43). Recently, Fatti and Mannucci reported on the use of a polyelectrolyte fractionated procine factor VIII to treat hemophilia A. Although there were problems with thrombocytopenia, resistance and anamnestic rises in antibody titer it was concluded that the use of the porcine product was a rational and effective therapeutic choice when one is faced with antibody titers above 10 U/ml and difficult clinical 17 situations (44). PROTHROHBIN COIPLEX CONCENTRATES In 1975 the results of a survey on the methods of preparation of Factor IX concetrates was reported by the Task Force on the Clinical Use of Factor IX Concentrates (45). Two major types of product were identified: 1) those containing mainly II, VII, IX, and X, and 2) those containing II, IX and X and being essentially free of factor VII. The products were prepared from a variety of starting material including: fraction I supernatant, cryoprecipitate supernatant, fraction III, fraction IV-l and precipitate P. In addition both citrate and EDTA have been used as the anticoagulant. The three absorbents most often used included DEAE-sephadex, DEAF-cellulose and tricalcium phosphate. The major difference in the absorbents relates to the content of the final product. Products prepared using tricalcium phosphate contained factor VII and those prepared with DEAE-sephadex or cellulose either did or did not depending on the isolation conditions. Generally, for any given product it was found that the potencies of II, IX and X were approximately equal and for products containing factor VII the ratio of coagulant/procoagulant proteins was usually between 20 and 30 U/ml. Additional information on techniques for the preparation of PCC's is available (46). Recently the use of "activated“ products has gained widespread interest and the two products most often used are Immuno FEIBM3>(Vienna, l8 Austria) and Autoplex3>(Hyland Laboratories). These particular products contain a greater amount of the activated form of the vitamin K dependent enzymes. Table 3 is a list of both the major protein constituents in the PCC's as well as minor contaminants that have been identified. Testing Procedures A number of assays have been devised to quantitate both the in vitro and in vivo potency of the PCCHL Although several of the in vitro assays correlate roughly with what can be expected to occur in vivo and allow the clinician to estimate a range of effective dosages, none of the in vivo assays correlate well with the observed clinical Table 3. Protein Constituents of Factor IX Concentrates Major Minor Misc. Factor II IIa Antithrombin III Factor X Xa Phospholipid Factor IX IXa Factor VII VIIa XII XI VIII V improvements seen as a result of the Factor Eigth Inhibitor Bypassing Activity (FEIBA). Table 4 lists a number of the assays that have been used as biological probes to assess FEIBA. For a comprehensive review of the assays available for 19 determining bypassing activity the reader is referred to Prouse et al., Prouse and Pepper, and Sar et al. (47-49). Table 4. Biological Probes for Assessing FEIBA* NAPTT* Rabbit Stasis Model** FEIBA Rabbit Non-Stasis Model+ TGt King Test Reca cification time Vermylen Test *Kingdon et a1. (50) **Wessler et al. (51) +Prowse and Williams (52) Adverse Reactions Two particular issues are of major concern as a result of administration of PCC's to either hemophilia B patients or hemophilia A individuals with an inhibitor. The first of these involves the risks of hepatitis and most recently the accompanying possibility of the transmission of the causitive agent of acquired immune deficiency syndrome (AIDS). The second relates to the observed thrombogenic complications associated with administration of PCC's. Hepatitis The incidence of reported cases of hepatitis among users of PCCVS is fairly high (53). Aronson reports thatOOOC while commercial products made in the USA have been most frequently involved in the published reports of hepatitis, it would seem that this is in fact a universal problem. A survey conducted by Hoofnagle et al. on the incidence of serologic markers indicating prior infection with hepatitis 20 B in patients being treated with PCC's revealed that the overwhelming majority were positive (54). Although even the most sensitive test cannot always detect 100% of the plasma units positive for hepatitis B antigen it is important that pooled plasma products be thoroughly tested. Hepatitis testing falls into three catagories that are differentiated based on the sensitivity of the assay. First generation testing includes Ouchterlony agarose double diffusion (AGD) which has a high degree of specificity but a low sensitivity. Second generation tests include counter immunoelectrophoresis (CEP), complement fixation (CF) and inhibition of passive hemagglutination (IPH). These tests are more rapid and 5 to 10 times as sensitive as AGD. It should be stressed that neither AGD, CEP or CF consistently detect HBsAG when present in relative low concentration. Third generation tests include radioimmunoassay (RIA), reversed passive hemagglutination (RPH) and the agglutination flocculation test (AFT). Testing with third generation tests provides approximately 100 to 1000 fold greater sensitivity than AGD. Acquired Immunodeficiency Syndrome Besides the obvious risk of hepatitis that is associated with receiving blood components from large batches of pooled plasma, there has been a great deal of concern recently over the possibility of contracting AIDS via the same mechanism. To date, there have been 21 approximately 60 cases of AIDS reported within the hemophilic population. A recent article by deShazo et al, found that the ratio of helper inducer lymphocytes (OKT4+) to suppressor-cytotoxic T-lymphocytes (OKT8+) was depressed in both hemophilia A and B patients receiving factor VIII concentrates and those receiving factor IX concentrates (55). In addition, it would be interesting to know what percent of the normal population, if any, have an abnormal OKT4+/OKT8+ ratio so that a more clear cut diagnostic implication could be associated with helper-suppressor ratios. Most recently a virus (human T-cell lymphotrophic retrovirus subgroup III), thought to be the causative agent of AIDS, has been isolated and a test has been developed to facilitate the wide spread screening of blood donors (56,57). In a population of Swiss patients studied by Schupbach et al. 100% prevalence of antibodies to HTLV-III was found in individuals with AIDS or pre-AIDS, a prevalence of 25% or greater in the known risk groups and no antibody- positive cases were found in 83 healthy donors (58). Thrombotic Complications Thrombotic complications have been reported as a result of administration of factor IX concentrates. These have included thrombophlebitis, localized arterial and venous thrombosis, disseminated intravascular coagulation (DIC) and pulmonary emboli (59-65). In addition, Gruppo et al. and Agrawal et al. have reported the deveIOpment of fatal myocardial infarction following the administration of 22 PCC's (66,67). Whether, or not the thrombogenic character of these concentrates is in anyway related to FEIBA has not been determined. This particular issue will be dealt with in more detail when the possible mechanisms of the bypass reactions are discussed. Clinical Efficacy Of Treatment With Prothrombin Complex Concentrates The majority of the reports on the efficacy of treatment with PCC's have been anecdotal in nature and a good deal have dealt mainly with the treatment of hemarthrosis. In addition variations in lot to lot potency of the complexes and intermanufacturer variations in producing the complex have made clinical assessment more difficult. Based on manipulations during the preparation of PCCHS there are two different products presently available for the treatment of hemophilia B and hemophilia A with a inhibitor to factor VIII. Broadly defined, these products fall into the classification of either a ”non-activated“ or "activated" PCC. The “non-activated" products are prepared according to the methodology previously cited in this review. ‘Until recently, however, the specific mechanism by which the “activated" products are prepared was not known. Information regarding the specific mechanism by which these products are manufactured, although protected by patent rights, is now available:(68,69). Mitru et al. (68) utilize two different methods for producing FEIBA in the PCCHL Method A involves adsorption of Effluent I from a Cohn 23 fractionation onto DEAE-Sephadexgg subsequent elution of factors II, VII, IX, and X and finally exposure to free Ca+2 concentrations in the range of 0.5 to 0.8 mM per liter of eluate. The activation procedure takes place usually for a period of about 24 hours at controlled pH and temperature. Method B is simply a modification of the elution characteristics used to remove the prothrombin complex from the DEAE-SephadexR. In this case all factor concentrations are kept at >10 U/ml and the eluate is simply maintained at a suitable pH and temperature during which time FEIBA is spontaneously generated. Both these products are reported to be substantially free of IIa (< 1.5 U/ml) and factor Xa with ratios of FEIBA to Na and Xa of at least 50:1 and 45:1 respectively. In an alternate method FEIBA is generated by first activating citrated plasma with a contact activator, such as kaolin, and then isolating factors II, VII, IX and X by chromatography on a basic ion exchanger such as DEAE (69). Because the efficacy of PCCWs for the treatment of inhibitors has been questioned, double blind studies have been conducted using both "activated" and "non-activated" products. The first of these studies was reported by Lusher et al. in 1980 (70). In this study Konyne and Proplex, both "non-activated” products, were tested for their effectiveness in treating acute hemarthrosis of the elbow, knee or ankle as opposed to an albumin placebo. Results were based on the examination of patients with both objective and 24 subjective criteria six hours after administration of either the placebo or prothrombin complex concentrate. Subjective data included the degree of joint pain. Objective data was based on joint circumference and the measured degree of mobility. Their findings indicated that the overall perceived effectiveness was 28.6% for the albumin placebo, 48.1% for Konyne and 53.2% for Proplex. There was a significant difference (P<0.0001) between the concentrates and the placebo but no statistical difference between the two concentrates. Subsequently, Sjamsoedin et al. reported on a double- blind clinical trial on the use of FEIBA, an “activated” product, as opposed to Prothromblex, a ll'non-activated" product (71). The study included mucocutaneous bleeding, joint bleeding and muscle bleeding and, the subjective and objective criteria used for assessment were similar to those used by Lusher et al. In this case, the Prothromblex was effective in 52% of the cases evaluated and the FEIBA product was found to be 64% effective. The value of 52% for the “non-activated" product agrees well with the work of Lusher et al. and although the "activated" product showed an additional 12% improvement, without more extensive comparisons it is difficult to say with certainty that there is a statistically significant difference between the I'activated" and "non-activated” product. Most recently Lusher reported on the trial results of Autoplex (“activated“) and Proplex (“non-activated") at the IXth International Congress on Thrombosis and Haemostasis 25 (72). In this particular trial no significant difference was found between the use of the 'activated' vs. "non-activated'l prothrombin complex concentrate. The fact that the prothrombin complex concentrates are effective has, therefore, been established. This refutes the observations that the products were losing their effectiveness (73-77). Whether or not the 'activated' product is equal to or better than the "non-activated'l product is difficult to ascertain with certainty. The major problem lies in the large variation between the clotting factor concentrations in each product. Table 5 shows the results of the factor assays on two lots each of Autoplex® and Immuno FEIBAon both Sephadex G-100 and dextran sulphate and demonstrated FEIBA in peaks from both columns but has not isolated or purified a specific protein or complex. Based on observations from Elsinger's work that FEIBA has a relatively high molecular weight on Sephadex G-100, Tishkoff and Hess (79) devised a number of experiments to try and demonstrate the existance of such a protein. Within the limits of the gel electrophoretic techniques used, they were unable to define a high molecular weight species having FEIBA. Barrowcliffe et al.(80) demonstrated the existance of a factor VIII-lipid complex in some PCC's and found that this complex was resistant to inhibition by the factor VIII antibodies. Although this could account for a minor portion of the bypassing activity the inconsistant amounts of factor VIII present do not correlate with the in vivo correction when 27 concentrates are injected nor the improved hemostatic function. Additional explanations for the bypassing activity that must be considered include the markedly elevated levels of coagulation zymogens following administration of PCC's and the presence of small amounts of activated clotting factors in these complexes. Although it is extremely difficult to rule out contamination with activated clotting factors as the causitive agent in the bypass mechanism a number of observations would suggest only a minor contribution. First, the presence of factor VIIa in PCC's could provide a factor X activating mechanism in conjunction with the released tissue factor. Although this would account for some improved hemostasis at the site of tissue injury or bleeding, it does not explain the shortening of the NAPTT or APTT seen in the in vitro test system when PCC's are evaluated for FEIBA. Second, small quantities of both IIa and Xa are known to contaminate PCC's. These enzymes could potentially contribute to the thrombogenic character and/or to the occasional DIC that has been described in the literature. In vitro data (78), however, suggests that FEIBA is not readily inhibited by PMSF and recent work on the in vivo inhibition of Xa and IIa by AT III indicates a very rapid inhibition within the human circulatory system (81L Finally, IXa as well as being inhibitable by AT III requires as a cofactor the presence of IIa activated factor VIII. In the absence of the antihemophilic factor the rate of activation of factor x by 28 the IXa complex would be approximately 30,000 fold less than it is in the presence of factor VIIIa. There are a number of articles in the literature documenting the existance of enzymatic activity in native zymogens (82-84). The most pertinent of these pertain to factor VII, chymotrypsinogen and trypsinogen. It has been shown by Nemerson et al. that the factor VII zymogen requires tissue factor for the activation of factor X. Because of the variable amounts of factor VII present in the PCC's and again because of the inability to explain the in vitro correction of the NAPTT or APTT via this mechanism the zymogen contribution by factor VII is questionable. It could be postulated that excessive amounts of either prothrombin or factor Ix could contribute to FEIBA. Factor IX however, like factor IXa, would require as a cofactor the presence of factor VIII. Finally, based on the kinetics of the prothrombinase complex the concentration of II in plasma has been shown to be approximately 5x the experimentally determined Km (85). This would indicate that although there is a large increase in the II concentration during PCC administration this zymogen increase would have little effect on the overall rate of the reaction. Therefore, because zero order kinetics already pertain, it is unlikely that increasing the II concentration would contribute significantly to enhanced hemostatic function. Work by Tishkoff (86) has demonstrated that a major contributing factor to the bypassing activity is factor X. 29 They found that factor x in combination with trace amounts of IIa could partially correct the APTT of a factor VIII deficient plasma with a high titer inhibitor. It was thought that the major contribution of IIa was as an activator of factor V and that factor X appeared to be the major catalytic agent. PROTHROMBINASE COHPLEX If the factor VIII bypassing activity of the PCC's is in anyway related to what is presently known about the classic mechanisms of coagulation then the most likely enzyme system to be involved would be the prothrombinase complex (IIase). Because a major portion of the work to be presented involves an understanding of the complex interactions of the components of the IIase complex a review of the mechanism of prothrombin activation will be presented. The components of the IIase complex include Ca+2, phospholipid, factor Va and factor Xa. A few of the questions that need to be answered include: What is the role of Ca+2 and what is the mechanism of phospholipid, Xa and factor Va binding? Does the composition of the phospholipid make a difference? What is the source of phospholipid? How do phospholipid and Va effect the rate of II cleavage? Does binding of Ca+2 induce conformational changes in II and factor X or Xa? What are the Km and Vmax for the complete complex? And, what is the order of binding of the individual components? 30 Vitamin K It has been established that vitamin K is essential for the normal production of several coagulant components. This critical vitamin K dependent, post-ribosomal, alteration results from the y-carboxylation of specific glutamyl residues in factor II, VII, IX, X, protein C and protein S (87-90). The carboxylation requires a reduced vitamin K, 02, C02 and a carboxylase complex which appears to include vitamin K epoxidase activity (91-93). A proposed mechanism suggested by Larson et al. is illustrated in Figure 1 (91). In addition the structure of glutamic acid and y-carboxyglutamic acid (GLA) are shown in Figure 2 (94). This vitamin K dependent modification of glutamic acid is necessary for calcium binding to occur and therefore for the physiologic activation of prothrombin and the lipid .binding characteristics of factor II, IX, and X. In addition, the GLA residues impart to these proteins their unique ability to be adsorbed to barium salts (95-99). The importance of the GLA residues can be demonstrated in humans and animals when vitamin K antagonists such as dicoumoral are used as anticoagulants. .Although the prothrombin molecule from either normal or dicoumoral treated individuals have identical amino acid compositions, the dicoumoral induced prothrombin will not bind Ca+2 and does not function normally (100—102). The defect has been shown to specifically involve the glutamic acid residues of the fragment 1 portion of the prothrombin molecule (103,104). The 31 KHZ o \J GLUflGLU(’) + 3* 2 U \(COZ [KHOOH] +_\ 320 no GLA Figure l. Carboxylation of Glutamic Acid. identification of the GLA residues of the vitamin K dependent factors provided a specific amino acid that could be implicated in defining the mechanism whereby the components of the IIase complex associate to form a functional enzyme unit. COOH HOOC COOH I \ / CH CH I 2 I $32 . $32 HZN— ca coon Hz ca coon Glutamic Acid y~carboxyglutamic acid Figure 2. Structure of Glutamic Acid and y-Carboxyglutamic Acid. Calcium Binding and Protein Conformation The majority of the work on the calcium binding characteristics of the vitamin K dependent proteins has been done using prothrombin as a model. However, there are many parallels with factor X as‘well.as the additional coagulant proteins. When analyzed by SDS electrophoresis human prothrombin has a molecular wight of approximately 72,000. 'The amino acid sequence of both human and bovine prothrombin are very 32 similar and there is extensive sequence homology among the proteinases of blood coagulation particularly as noted by the conservation of the charge relay system of the serine proteases and the nearby residues in the 3-dimensional structure (105). The activation intermediates of human prothrombin have been determined and are illustrated diagramatically in Figure 3. The Xa and IIa cleavage sites are included as well as alternate designations for the prothrombin fragments (106,107). As will be seen, an understanding of the various activation fragments of the prothrombin molecule will facilitate an understanding of the studies on Ca+2 binding, Va binding and conformational changes induced by Ca+2 and other divalent cations. As we have seen the functional characteristics of the vitamin K dependent proteins as either substrates or ultimately as proteolytic enzymes depends on their Ca+2 binding characteristics and subsequently their ability to bind a phospholipid surface. Papahadjopoulus and Hanahan were the first to demonstrate that Ca+2 ions were necessary for the physical integrity of the IIase cOmplex and that while they were not important for the binding of factor Va they were essential for the Xa mediated cleavage of the prothrombin molecule (108L. The calcium binding characteristics of prothrombin were more easily studied than those of the other vitamin K dependent proteins for two reasons: 1) prothrombin is present in large quantities and 33 IIa (Xa) Xa Xa I l I | _____________ l NH2 COOH Int. 3 Int. 4 Int. 2 Fragment 1-2 A B l V Fragment 1 Int. 3 23,000 Fragment 2 Int. 4 13,000 Prethrmobin 1 Int. 1 51,000 Prethrombin 2 Int. 2 41,000 Fragment 1-2 Int. 3-4 36,000 Figure 3. Prothrombin Activation Fragments. Prethrombin 2 or Int. 2 is the immediate precursor of a-thrombin which has A and B chains linked by a disulfide bond. is easily isolated and 2) unlike the other vitamin K dependent proteins the y-carboxyglutamic acid region of prothrombin is not linked to the rest of the molecule by disulfide bonds. Therefore, isolation of the Ca+2 binding region is facilitated. As previoulsy noted the fragment 1 or intermediate 3 portion of the prothrombin molecule has been shown to contain all of the Ca+2 binding sites (109- 111). The binding of Ca+2 to fragment 1 has been studied using a variety of methods including equilibrium dialysis, Hummel and Dreyer gel filtration, steady state dialysis and paramagnetic relaxation rates (112-117). A number of observations can be made based on these techniques. First, 34 there are approximately 6-10 calcium binding sites per prothrombin molecule with a dissociation constant of 6.3 x 10". Second, because of the relatively weak binding of Ca+2 to the prothrombin molecule studies can only be done at fairly high protein concentrations and greater effects from protein dimerization can be expected. In contrast, because of the relatively tight binding of manganese (Kd = 2.2 x 1075) and gadolinium (kd = 1.6 x 10'7) equilibrium studies can be done at much lower protein concentrations. This facilitates evaluation of the binding characteristics and provides a model to which the Ca+2 binding can be compared. Studies using trivalent lanthanide ions Mn(II) and Ca(II) have demonstrated two separate classes of binding sites for both factor X and prothrombin (112, 114, 118-120). In each case, two high affinity sites were identified and a number of low affinity sites were found. Finally, in addition to determining the number of metal binding sites and the affinity of those sites, most of the observations on the metal-binding characteristics have evidenced complex binding. Both positive cooperativety, a process similar to that observed with the 02 saturation curve of hemoglobin, and negative cooperativety, a process reflecting multiple classes of binding sites. If a process reflecting positive cooperativety is taking place then some mechanism, other than simple Ca+2 binding, must be looked for. Prendergast and Mann and Jackson, et al. observed a cation mediated self association of prothrombin fragment 1 in the presence of Ca+2 (116,117). This dimerization phenomenon may be of no 35 significance as protein concentrations that reflect physiologic levels fail to demonstrate complex binding characteristics and only when protein concentrations are much higher is positive cooperativity demonstrated. Alternatively, a change in the stoichiometry of the factor X or prothrombin molecule could account for an observation consistent with positive cooperativity. It has been observed that binding of several di and trivalent cations induces a spectral shift and flourescent quenching of specific tryptophan residues resulting from a conformational change in the molecule (116,121,122). Lewis et a1. and Keyt et al. have also isolated Ca+2 induced conformation specific antibodies to both prothrombin and factor X (123,124). Based on the flourescent quenching, immunochemical information, the presence of 2 high affinity Ca+2 binding sites and circular dichroic studies by Bloom and Mann (125) there is adequate data to support positive cooperation between the two high affinity Ca+2 binding sites resulting in an alteration in the tertiary structure of prothrombin or Factor X (126). Nelsestuen has shown that prothrombin must undergo the Ca+2 dependant conformational change prior to binding lipid and that an intact secondary and tertiary protein structure is required to form the phospholipid binding region (127). Physiologically, it is possible that prothrombin could circulate with the high affinity Ca+2 sites bound. This statement is based on the fact that half- maximal perturbation of the prothrombin molecule is seen at 36 a Ca+2 concentration of 0.22 mM (118) and physiologic levels of Ca+2 are in the range of 1 mM. Factor V Factor V is a high molecular weight glycoprotein that has been purified to homogeniety by several investigators (128-131). The single chain factor V, prior to involvment in the prothrombinase complex, is activated by human IIa or the factor V activator from Russells viper venom CRVV-V) to Va (132-134). Suzuke et al. demonstrated four fragments upon thrombin-catalyzed activation consisting of fragment D (105,000), fragment E (71,000), fragment Cl (150,000) and a dimer fragment 1-2 (71-74,000) (134). The biologically active fragments appeared to be fragment D noncovalently bound to fragment 1-2. In addition to activating factor V, thrombin will eventually proteolytically degrade the Va activity (126L Factor Va accelerates the rate of prothrombin activation. Nesheim et al. showed a 13,000 fold rate enhancement of the prothrombinase complex when the complete complex was compared to a Xa-Ca+2-phospholipid complex and a 278,000 fold enhancement of the complete complex over that of the enzyme, Xa alone as determined in an amidolytic assay (85). According to Rosing et al. the effect of Va is mainly to increase the Vmax as opposed to lipid which has its major effect on decreasing the Km (136). The binding of factor Va to prothrombin appears to be through the fragment 2 region of the prothrombin molecule and although there appears to be a single Ca+2 binding site that is necessary 37 for Va activity, the binding of factor Va to lipid is a Ca+2 independent process (137,138). Phospholipid: Surface Assembly In vivo, the classically defined mechanism for the assembly of the prothrombinase complex is via interaction with platelet surfaces, in particular platelet factor 3. Miletich et al. have demonstrated that factor V(Va) is responsible for the binding of factor Xa to the platelet surface (139,140). In addition to the Va receptor, phospholipid has been proposed to be involved although the exact mechanisms are unknown. Studies evaluating the total effect of the prothrombinase complex on the II——->IIa conversion have substituted phospholipid monolayers for platelet membranes. It has been reported that the platelet surfaces catalyze prothrombin activation 15x the rate of lipid monolayers. Nesheim et al. found, however, that under their experimental conditions there was no difference in the catalytic efficiencies of a system containing platelet lipid and Va and one containing added phospholipid and Va (85). They also concluded, as did Militich et al. (139) that factor Va comprises the Xa binding site on the phospholipid surface as well as on platelets. Both prothrombin and factor X bind phospholipid surfaces and a considerable amount of work has gone into defining a model lipid system (126,127,141).' The main feature appears to 38 be the need for an acidic phospholipid, of which phosphotidylserine is the most effective (142,143). The binding, as noted earlier, is Ca+2 dependent and requires a metal induced protein transition in prothrombin and factor X. Work by Lim et a1. indicated that prothrombin binds via the tip of fragment 1 and extends radially from the membrane surface (144). Factor X also appears to bind at one end and extend into solution. Mayer et a1. using surface pressure measurements concluded that binding of prothrombin resulted in only minor surface pertubations and therefore, there was not significant insertion of the protein into the monolayers (145%. In addition to the platelet or model lipid source providing ameans by which the prothrombinase complex is assembled, it may also provide a means of concentrating the components of the enzyme complex and increasing their local concentration. It is interesting to note that although lipid model studies have given a good indication about the type of interaction necessary to assemble the prothrombinase complex, the specific type of in vivo reactions are not clear. Nemerson and Furie speculated that it would be +2 and possible for any mechanism capable of binding Ca organizing the protein array to optimize coagulation (135). Therefore, platelet lipid does not necessarily provide the appropriate surface. Although the specific mechanism of assembly of this complex and the order in which it occurs is unkown, based on the data available one can postulate on possible mechanisms. 39 Kosow and Orthner, after a kinetic evaluation of the activation of human prothrombinase by Xa, Ca+2 and phospholipid, showed that phospholipid appeared to be the second reactant to bind to the enzyme (146). This is consistent with a mechanism in which two high affinity binding sites undergo a Ca+2 dependent cooperative binding process that results in a conformational change in factor X and II. This change, which is necessary for membrane binding, results in the exposure of, in the case of prothrombin, 6 additional Ca+2 binding sites and, in the case of factor X, as many as 16-20 Ca+2 binding sites, which are composed of y—carboxyglutamic acid residues and utilize the binding of Ca+2 to bridge the negative charged lipid surface to the negative charge of the GLA residues. Whether prothrombin or factor X(Xa) bind first is not known. The site on the lipid surface for binding Ca+2 would be the phosphate group but the specific platelet receptor has not been well defined. Along this line, it has been demonstrated that minimal prothrombinase activity exists with inactivated platelets but upon activation procoagulant activity increases significantly because of the internal location of the negatively charged phospholipids (147). Finally, factor Va appears to serve as the binding site for the X(Xa) molecule and to increase the maximum velocity of the reaction. Whether or not this alters the proteolytic activity of X(Xa) or provides a more favorable steric arrangement or both is not certain. 40 MATERIALS AND METHODS Chromatography Deae-Sephadex A-50, Sepharose 4B, QAE-Sephadex A-50 and Sephadex G—25 were the products of Pharmacia, Piscataway, NJL Ultrogel AcA—34 was purchased form LKB Ltd. Rockville, MD. Heparin—Sepharose (148), benzamindine-Sepharose (148), poly(homoarginine)-Sepharose (148), RVV-X-Sepharose (106) and AT III-Sepharose were prepared by coupling the ligand to Sepharose-4B activated with cyanogen bromide. Proteins Protein preparations were assessed for homogeneity by 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS—PAGE) according to the method of Weber and Osborn (149). The purity of factor V was determined on 5% gels and the standards used in both cases were carbonic anhydrase (M.W. 29,000), egg albumin (M.W. 45,000), bovine ablumin (M.W. 66,000), phosphorylase B (M.W. 97,400), B-glactosidase (M.W. 116,000) and myosin (M.W. 205,000). Protein samples were evaluated in both a reducing, 1% v/v 2— mercaptoethonal, and non-reducing system. Protein concentrations were estimated from the absorbance at 280 nm 41 using the following values for 3230 nm‘ factor X, 1.16; prothrombin, 1.38 (148): factor Xa, 0.96 (150): factor V, 0.89 (128): and AT III, 0.61 ml mg'lcm'l (151). The partially purified protein concentration of factor VII was estimated using the equation of Kalachar where protien concentration in mg/ml =*1.45D280 - O‘TDZ60 where D280 and D260 are optical densities at 280 nm and 260 nm respectively (152). Factors II, X and 1! Purification Factors II, X and IX were purified as described by DiScipio et al. IAll procedures, unless otherwise indicated, were performed at 4°C (148). Five liters of cryo-poor plasma anticoagulated with CPD-A.(Great Lakes Regional Red Cross Blood Services, Lansing, MI.) were pooled in a large plastic container. Heparin (50,000 units), 1.5 mM O-phenanthroline and 250 mg of polybrene were added and the plasma was stirred for 15 minutes. Two hundred ml of 1M barium chloride was then added with continuous stirring over a period of 30 minutes and the stirring was continued for an additional 15 minutes. The plasma was centrifuged at 8,000 xg for 15 minutes and the supernatant was discarded. The barium citrate pellet. was then washed twice with 1 liter each of 0.02 M imidazole buffer, pH 6.0, containing 0.2 M NaCl and 1 mM benzamidine. The adsorbed protein was eluted from the barium citrate pellet by homogenization in a siliconized Waring blendor with 1 liter of 0.2 M Mes-HCl, pH 5.85, conatinig 0.15 M sodium citrate, 10 mM benzamidine, 1000 units of heparin and 42 10 mg of soybean trypsin inhibitor (STI). The mixture was then stirred for 1 hour and the precipitate was removed by centrifugation at 8,000 xg for 20 minutes. The supernatant from the barium citrate eluate was brought to 10% saturation with solid, enzyme grade, ammonium sulfate by addition of 76 g of finely ground ammonium sulfate containing 1.5 ml of l M Tris base as a buffering agent. .After addition of the ammonium sulfate the solution was allowed to stir for an additional 15 minutes. The precipitate was removed by centrifugation at 8,000 xg for 15 minutes and discarded. The supernatant was brought to 40% saturation by the slow addition of 186 g of finely ground ammounium sulfate containing 9 ml of 1 M Tris base. The precipitate was again removed and discarded and the supernatant was brought to 70% saturation with 229 g of ammonium sulfate containing 10 ml of 1 M Tris base. The precipitate from this cut was collected by centrifugation at 8,000 xg for 30 minutes and then dissolved in 50 m1 of 0.02 M Tris-H3PO4, pH 5.85, containing 0.1 M EDTA, 2 mM benzamidine, and 0.02% sodium azide. This solution was then dialyzed overnight against 1 liter of the same buffer followed by two, one liter changes of 0.02 M Tris-H3PO4, pH 5.85, containing 2 mM benzamidine and 0.02% sodium azide. A fine precipitate developed during dialysis and was removed by centrifugation at 9,000 xg for 10 minutes. The protein was applied to a DEAE-Sephadex-ASO column (2u6 x 36 cm) equilibrated with 0.02 M Tris-H3PO4, pH 5.85, 43 2 mM benzamidine and 0.02% sodium azide. The column was washed with 250 ml of the same buffer plus 0.15 M NaCl and eluted with a gradient consisting of 250 m1 of 0.02 M Tris- H3PO4, pH 5.85, 2 mM benzamidine, 0.02% sodium azide and 0.15 M NaCl and 250 ml of the same buffer with 0.55 M NaCl. The column flow rate was 66 ml/hr and 3.5 m1 fractions were collected. The peak of protein containing the factor X, II, and IX activity was pooled and dialyzed against 3 one liter changes of 0.02 M Mes-Tris, pH 5.9, 2 mM benzamidine and 0.02% sodium azide. The protein obtained from the DEAE column was then applied to a benzamidine-Sepharose column (1.6 x 30 cm) equilbrated with 0.02 M Mes-Tris, pH 5.9, 2mM benzamidine, 0.02% sodium azide. The column was washed with the same buffer containing 0.3 M NaCl and then eluted with a linear gradient consisting of 250 ml of the above buffer with 0.3 M NaCl and 250 ml with 1.3 M NaCl. The flow rate was 24 ml/hr and 3.5 ml fractions were collected. The factor IX and X activity were seperated on this column. The factor X activity was pooled and immediately brought to 1 mM PMSF and applied to a poly(homoarginine)- Sepharose column (1.6 x 30 cm) equilibrated in 0.02 IM Mes-Tris, pH 5.9, 0.02% sodium azide. The column was washed with 250 m1 of 0.02 M Mes-Tris, pH 5.9, 0.02% sodium azide, 3.0 M NaCl and 2 x 10"4 M PMSF. The protein was then eluted with 250 ml of Mes-Tris buffer containing 2.4 M NaCl and 2.9 M quanidine hydrochloride and 11mM PMSF. The column was run at 16 ml/hr and 2.5 ml fractions were collected. 44 The factor X from this column was pooled and brought to 10 mM in benzamidine and the dialyzed against 1 liter of Mes- Tris buffer containing 2.4 M NaCl and 10 mM benzamidine. Following this the factor X was then dialyzed against two 1 liter changes of 0.02 M Mes-Tris, pH 5.9, 2 mM benzamidine and 0.02% sodium azide. The factor X was stored at -80°C. The factor IX activity from the benzamidine-Sepharose column was pooled and concentrated to approximately 60 m1 on an Amicon PM-lo membrane. The concentrate was then dialyzed overnight against 1 liter of 20 mM Mes-Tris, pH 5.9, 2 mM benzamidine and 0.02% sodium azide. Following this the sample was dialyzed an additional 8 hours against the same buffer containing 50 mM NaCl. At the end of this dialysis CaC12 was added to a final concentration of 2.5 mM and the sample was brought to lmM with 0.1 M PMSF. The sample was applied to a heparin-Sepharose column (1.6 x 13 cm) that had previously been layered with 2 cm of Sepharose— 4B and equilibrated with 20 mM Mes-Tris, pH 5.9, 2 mM benzamidine, 2.5 mM CaClz, 0.02% sodium azide and 50 mM NaCl. The flow rate was maintained at 20 ml/hr and 2.5 ml fractions were collected. After the sample application the column was washed with 60 m1 of equilibrating buffer. The protein was then eluted with a linear gradient consisting of 100 ml of equilibrating buffer containing 1 mM PMSF and 100 ml of the same Mes-Tris buffer containing 1.5 M NaCl and 1 mM PMSF. Following elution of prothrombin the gradient was stopped and the column was washed with 100 to 45 150 m1 of Mes-Tris buffer containing 0.3 M NaCl and 1 mM PMSF, The gradient was then reapplied and the factor IX activity eluted. The peak of prothrombin activity and factor IX activity were pooled seperatly, brought to 5.0 mM EDTA and dialyzed 5 hours against 1 liter of 20 mM Mes-Tris, pH 5.9, 2 mM benzamidine, 0.02% sodium azide and 50 mM EDTA and then overnight against 20 mM Mes-Tris, pH 5.9, 2 mM benzamidine, 0.02% sodium azide and 0.1 M NaCl. The samples were concentrated on a Amicon PM-lO membrane and stored at Factor VII Purification In the DiScipio procedure, factor VII eluted on the back half of the leading protein peak from the DEAE—Sephadex-ASO column. The factor VII activity was pooled and brought to 10 mM in benzamidine and further purified as follows: The purification of factor VII was a modification of the procedure of Broze and Majerus (153). The pooled factor VII activity, obtained as previously described, was dialyzed for 8 hours against two 1 liter changes of 0.05*M Tris HCl, pH 8.0, 0.15 M NaCl and 10 mM benzamidine. The sample was then applied to a QAE-Sephadex column (1.6 x 30 cm) equilibrated in the same buffer. The flow rate was maintained at 20 m1/hr and 2 ml fractions were collected. The column was then washed with 100 m1 of equilibrating buffer and the factor VII was eluted with a buffer containing 50 mM Tris-HCl, pH 8.0, 0.135 M NaCl, 0.005 M CaC12 and 10 mM benzamidine. Each fraction collected 46 contained 50 ul of 0.5 M NazEDTA, 2.0 M Tris-HCl, pH 8.0. The fractions containing factor VII activity were immediately pooled and concentrated to approximately 4 ml. The concentrate was dialyzed overnight against 20 mM Tris- HCl, pH 755, 0.15 M NaCl and 10 mM benzamidine. The factor VII activity was stored at -80°C and prior to use was desalted on Sephadex G-25 to remove the benzamidine. Factor Xa Human factor Xa was prepared by the method of Downing et al. (106) using the factor X activating enzyme from Russell's viper venom (RVV‘X) prepared according to the procedure of Kisiel et a1 (154L. The following modifications were made. Sephadex G—100 was substituted for Sephadex G- 150 and all isolation procedures were preformed at 4°C. After fractionation on QAE-Sephadex the fractions having :RVVHX activity were pooled and run on a Sephadex G-25 column equilibrated in 0.5 M NH4HCO3 and immediately lypholized. The lypholized samples were stored at -20°C. Factor X zymogen was activated by incubation with RVV‘ X-Sepharose in the presence of 6 mM CaC12 for 10 min at 37°C as follows: Sepharose-4B was activated in the conventional manner. After activation the gel was washed with 0.0168 M imidazole, pH 7.4, 0.145 M NaCl and then mixed with a volume of buffer equal to the packed gel volume containing 10 mg/ml of purified RVVeX (154%. After gentle stirring at 4°C overnight 47 the buffer-protein mixture was removed and the same volume of l M ethanolamine, pH 8.0, was added and stirring continued for an additional 2 hours at room temperature. The gel was then washed alternately with 0.1 M bicarbonate buffer, pH 9.5, 1 M NaCl and 0.1 M acetate buffer, pH 4.0, 1 M NaCl. The gel was then equilibrated with 0.0168 M imidazole, pH 734, 0.15 M NaCl. Prior to activation factor X was dialyzed against 0.0168 M imidazole, pH 7.4, 0.15 M NaCl. Both the factor X and the RVV‘X-Sepharose were brought to a final concentration of 6 mM CaC12 and approximately equal volumes ofJRVVeX-Sepharose and factor X were combined and incubated, with mixing, at 37°C. The reaction was stopped by addition of 0.2 M tetrasodium EDTA to a final concentration of 20 mM when the Xa activity reached a maximum. lThe Xa was then dialyzed against 20 mM Mes-Tris, pH 5.9, 2'mM benzamidine and 0.02% sodium azide and stored at -80°C. Prior to use the benzamidine was removed by dilution to give a final concentration of less than 4 x 10"6 M in the assay mixture. SDS-PAGE of factor Xa indicated that the preparation was homogenous. The active site concentration of factor Xa was determined by titration with p-nitrophenyerF-guanidinobenzoate (p-NPGB) according to the procedure of Smith employing an experimentally' determined molar extinction coefficient for p-nitrophenol of E402nm= 15,317 M'lcm'l (155). 48 Factor'V Factor V was isolated according to the method of Kane and Majerus (128). The following modifications were made: Fresh plasma was substituted for fresh frozen plasma as there appeared to be some activation products resulting from the use of the latter. In addition, PMSF was used in place of DFP as the active fluoride inhibitor. PMSF, was added immediately prior to the use of any buffer. The final product had a specific activity of 75 U/mg and showed a single molecular weight band on SDS-PAGE. Activation of factor V was carried out in the following manner. The protein was desalted on an AcA-34 column equilibrated with 20 mM Tris-HCl, pH 7.4, 0.15 M NaCl and 5 mM CaC12 to remove the benzamidine. Factor V (200 ug/ml) was incubated with a-thrombin (2 U/ml) for 5 min at 37°C in the same buffer. Thrombin was removed by adsorption onto benzamidine-Sepharose followed by filtration on a mini- column to remove the gel. Antithrombin III and Fibrinogen Purified AT III was generously provided by Dr. M. Wickerhauser (American Red Cross, Bethesda, MD.) and contained 8.1 mgfiml of AT III and was 95% pure by SDS-PAGE. Human fibrinogen was obtained from Kabi Group, Inc., Greenwich, CT. A stock solution was prepared by dissolving 1.5 g in 66 ml of distilled water and then dialyzing for 12 hours against 2.4:mM sodium citrate-50 mM sodium acetate- 49 0.15 M NaCl, pH 8.2. The protein was stored at -80°C. Phospholipids Phospholipid vesicles were prepared according to the procedure of Kosow and Orthner (146), and Nelsestuen and Lim (144). A chloroform suspension, prepared by mixing 50 mg of L-a-phosphatidylethanolamine and 22 mg of Folch Fraction III (Sigma Chemical Co., St. Louis, MOJ, was used routinely as a source of phopholipid. After sonication and centrifuga- tion, the phospholipid suspension was chromatographed on a Sepharose-4B column proviously primed with isotonic saline extract of rabbit brain followed by thorough washing and equilibration with 20 mM triethanolamine, pH 7.4. The molar concentration of lipid in the stock solution was determined by the method of Gomori (156). Dilutions of the stock solution were made in 50 mM triethanolamine, pH 7.4, 0.2 M NaCl. Addtitional Reagents Morpholinoethanesulfonic acid (Mes), Trizma base, quanidine hydrochloride, benzamidine, heparin sodium salt (grade 1, 164 units/mg), poly-L-lysine (M.W. 150,000- 300,000), p-aminobenzamidine hydrochloride, e-amino-N- caproic acid, l—cyclohexyl-3-(2-morpholinoethyl— carbodiimide, Ressull's viper venom, polyethylene glycol 6000 and p-nitrophenol were purchased from Sigma Chemical Company, St. Louis, MO. Triethanolamine hydrochloride, O-methylisourea hydrogen sulfate and cyanogen bromide were obtained from Aldrich Chemical Company, Milwaukee, WI. 50 p-NPGB was purchased from ICN Biochemicals, Cleveland, OH. All other reagents were of the highest purity available. p-(Amidinophenyl)methanesulfonyl flouride (p-APMSF) was a generous gift of Dr. David Bing, Boston, MA (157). Clotting Factor and FEIBA Assay Procedures Prothrombin, factors X, Xa, and IX were assayed by methods proviously described (158%. Factors V and Va were assayed by a modification of the one-stage procedure of Quick (159) where one unit of activity is defined as the amount of activity present in 1 ml of pooled normal plasma. Factor V deficient plasma, ACTINEL Factor IX deficient plasma and Activated Thromboplastin were obtained from American Dade, Miami, FL. FEIBA was monitored by the clot promoting activity of the test protein in the activated partial thrmoboplastin time (APTT) or nonactivated partial thromboplastin time (NAPTT) assay employing either factor VIII deficient plasma or factor VIII deficient plasma with an inhibitor as substrate. Inhibitor plasma was provided by Dr. John Penner, Michgian State University, East Lansing, MI. Individual APTT assays were performed as follows: The test sample (0.1 ml) and 0.1 ml of factor VIII deficient or inhibitor plasma were mixed together and warmed at 37°C for exactly 1 minute. ACTIIQ (0.1ml) was added and the incubation was continued for an additional 2 minutes. Recalcification was accomplished by the addition of 0.1 ml of prewarmed 0.02 M CaClz and the clotting time recorded. 51 The results are expressed in seconds as the average of duplicate determinations. NAPTT assays were performed as for the APTT except rabbit brain cephalin (Sigma Chemical Co., St. Louis, MO.) was substituted for ACTIN®. Prothrombin Activation Assay The assay system used was a modification of the procedure of Kosow and Orthner (146) where prothrombin activation in a steady state system is monitored by the generation of thrombin; thrombin peptidase activity was assayed on the peptide anilide S-2238 (H-D-Phe-Pip-Arg-p- nitroanilide HCl, Ortho Diagnostics, Raritan, NHL). The activation of prothrombin was carried out in disposable semimicro cuvettes at 30°C. The reaction mixture, in a total volume of 1 ml, consisted of variable concentrations of factor X (or factor Xa), prothrombin and phospholipid as specified for individual experiments under ”Results”. In addition, the reaction mixture contained 1 U/ml of factor Va, 2.5 mmol/ml CaC12 and 100 umol/ml o'f S-2238 in 50 mM triethanolamine, pH 7.4, 0.16 M NaCl buffer. The assay was performed as follows: Factor X or Xa, factor Va, CaC12 and phopholipid were preincubated for 2 minutes. At the end of this time S-2238 was added and the endogenous rate of activation was followed for 1 min. The reaction was initiated by the addition of prothrombin. The rate of release of p-nitroaniline was monitored by a DU-spectropho- tometer at 410 nm with Gilford adaptations and a chart 52 recorder. The spectrophotometric data were fitted to the equation A410 nm‘ at2/2 according to the procedure of Kosow and Orthner (146). The slope of this line, which is directly proportional to the rate of prothrombin activation, is con- verted to the concentration of thrombin formed per unit time using a molar extinction coefficient for p-nitroaniline ester of 9400 and an experimemtally determined first order rate constant of k = 55/8 for the cleavage of S-2238 by a-thrombin. This rate constant is in very close agreement with that determined by Kosow and Orthner (146). The endoge— nous rate due to spontaneous cleavage of the chromogenic substrate was subtracted from the absorbance readings prior to determination of the rate of prothrombin activation (See Appendix). 53 Protein Isolation and Characterization Prothrombin Complex Proteins Factors II, IX and X were purified to homogeneity from cryoprecipitated fresh frozen human plasma. The criteria for determining homogeneity consisted of the presence of a single band of protein on 7.5% SDS-gels, the absence of any detectable contamination with other clotting factors of the prothrombin complex and, in the case of factor X, amino terminal sequencing of the purified protein. Prothrombin and factor IX evidenced a single band on both reduced and non—reduced SDS-gels and factor X was a single band in a non-reduced system and demonstrated a heavy and light chain in a reducing system (Figure 4). The molecular weight of the purified prothrombin molecule was 72,000, factor IX was 58,000 and Factor X had a heavy chain molecular weight of 49,000 and a light chain molecular weight of 17,000. The only modification in the DiScipio procedure was the use of PMSF in the final stages of the factor X preparation and during heparin affinity chromatography. There was no detectable Xa activity in any of the isolation procedures (<0;001U/m1 when determined by comparison to a Xa standard). 54 It was necessary to carefully monitor the elution of factor X and IX from the benzamidine—Sepharose column. With multiple use, there was a decrease in the affinity of the benzamidine-sepharose for the proteins. As a result, adjustments in the salt gradient to lower ionic strengths was required to maintain adequate seperation of factor IX and X. - [III-I *\ ”.._,— '1’ AB CD EF Figure 4. Prothrombin complex proteins isolated and purified by the procedure of DiScipio. Proteins samples were run on 7.5% SDS—polyacrylamide gels in both a reducing buffer (2-mercaptoethanol) and a non—reducing buffer. A and B, prothrombin reduced and non-reduced; C and D, factor IX reduced and non—reduced: E and F, factor X reduced and non-reduced. 55 Factor‘V Factor V was isolated from fresh plasma as apposed to fresh frozen plasma. This change was made because of problems that arose with a precipitate in cryoprecipitated plasma that did not redissolve on thawing at 37°C. It is possible that limited activation of the plasma occured with the resulting formation of fibrin. The use of fresh frozen plasma also resulted in the coelution of several different contaminating bands along with the factor V. ‘When fresh plasma was used, factor V was isolated as a single band of protein with a molecular weight on 5% SDS-gels of approximately 300,000 (Figure 5). The specific acitivity was approximately 70 U/mg and could be activated 10-15 fold upon incubation with 2 U/ml of thrombin. Factor VII Factor VII was isolated as a partially purified protein by the modifications described in the methods section. With these modifications we obtained a 65,000 fold purification with a specific acitivity of approximately 900 U/mg. Although the factor VII was not purified to homogeniety there was no contamination with other components of the prothrombin complex, as determined by specfic factor assays, and it was therefore deemed suitable for in vitro coagulation studies that examined the effect of factor VII on FEIBA. 56 Amino Terminal Sequencing As part of the factor X characterization the Protein Sequencing Laboratory of the University of California, Davis, performed amino-terminal sequencing on the purified factor X. We were concerned about two particular issues. Figure 5. Purified factor V as determined by 5% SDS-PAGE (A). The molecular weight standard (B) is included for comparison and consists of the proteins described in the Materials and Methods section. One, we wanted the factor X protein characterized to the point that there was no question about the purity and identity of the molecule when compared to what had been 57 previously described in the literature. Two, because of the unusual characteristics we are ascribing to the factor X molecule we felt it was necessary to be sure that there was no alteration in the amino-terminal sequence that would be indicative of proteolytic alterations. The results obtained are shown in Table 6. Table 6. Amino Terminal Sequence of Factor X Heavy Chain - Ser Val Ala Gln Ala Thr Ser Light Chain - Ala Asn Ser Phe Leu Glu Met I I ‘ I . . I I l I “a: 1 I I I I I ‘ I.‘d‘ ‘3 "AI I‘I I‘.I Ili . I l . I In Vitro Characterization of the Factor Eight Inhibitor 'Bypassing Acitivity FEIBA Assay In order to determine the in vitro effectiveness of factor X in a clotting system the FEIBA assay was used. We performed the initial assays after both one and forty minute preincubations of factor X or Xa with either the factor VIII inhibitor plasma or factor VIII deficient plasma without an inhibitor. The rationale for these incubation times is based on the assumption that after a forty minute preincubation endogenous inhibitors (AT III) would have destroyed activity due to contamination with or, in some experiments, intentional addition of active enzyme components. Figure 6 graphically illustrates the coagula- tion results using the FEIBA assay with factor X as the test sample and factor Xa as the control. As can be seen, either 58 protein will correct the prolonged clotting time at one minute. However, after forty minutes of incubation the Xa activity has been eliminated and only the factor X zymogen evidences any ability to still correct the prolonged clot- ting time. The ability to correct the prolonged clotting time was seen in both the APTT and the NAPTT. 100 P/’—. Blank , Xa APTT (sec) l I J 1 20 40 Incubation time (min)- Figure 6. Corrective effect of factor Xa and factor X on APTT of factor VIII deficient inhibitor plasma. ‘The assay contained 0.05 units of Xa or 0.5 units of X, expressed in clotting units. 59 Effect of AT III on FEIBA In order to varify the assumption that endogenous AT III was in fact inhibiting Xa and, in addition, to eliminate any contamination of factor X with factor Xa, both factor X and Xa were preincubated with an excess of purified AT III. Experiments were conducted in two ways. 1) Factor X or Xa were incubated with an aliquot of AT III and 0.1 ml of this mixture was transferred to the FEIBA assay to determine if any bypassing activity remained when compared to controls that had been treated in an identical manner but without AT III. 2) Factor X or Xa was incubated with an equal amount (v/v) of AT III-Sepharose. The Sepharose was removed on a small column and the eluate was assayed for activity. In this case the controls were incubated with an equal amount of Sepharose that had been prepared without addition of AT III. This protocol was followed in order to determine if adding an excess of AT III had any effect on components, other than Xa, that are necessary for clot promotion in the FEIBA assay. The results of both approaches were essentially identical. Factor Xa activity was almost completly destroyed whereas the X activity was only slightly altered (Table 7). The small change in activity seen upon incubation of factor X with AT III was most likely due to inhibition of trace amounts of the active enzyme component, factor Xa. 60 Table 7. Effect of AT III-Sepharose on Factor X and Factor Xa FEIBA __._.1_._4_4#._r —a—a—l—l_‘_‘_n _.‘_4_._.‘_1_4_4_‘_._4 —‘-‘—1—n—4—4_1—l__l—l—n—A_A—A _4_4__‘_‘ ._.__‘_‘_‘_d_4_4—-—1 Reaction Mixture APTT NAPTT (sec) (sec) Buffer Blank 110 560 Factor X 68 151 Factor X + AT 111 74 196 Factor Xa 13 23 Factor Xa + AT III 72 114 _. —4 ‘—a_‘_4_4 ...-a-z __‘_._4 _..4 I_4_4_‘ _‘_a_._._._‘__n_4_J-.a_‘_4 _4_1_.J_.a _. .1-4—4 _.__._1 _‘__. Experiments were performed in the presence of equal volumes of AT 111- Sepharose (9 mg protein/ml) and either factor Xa or factor X zymogen. The reactants were incubated at 25°C for 60 min and the AT III- -Sepharose removed by filtration on a mini-column (1.4 x 6.5 cm). Following removal of AT III- Sepharose, the filtrate was assayed for FEIBA using factor VIII inhibitor plasma (45 Bethesda units) as substrate. The assay contained factor Xa (8.6 nmols) or factor X (0.27 umols). A control reaction mixture consisted of Sepharose that had been activated and treated in a manner identical to the AT III-Sepharose but without the addition of the ligand. Effect of p-APMSF on FEIBA (p-Amidinophenyl)methanesulfonyl fluoride has been shown to be a specific, irreversible inhibitor of serine proteases. It has a specificity for the postively charged side chains of lysine or arginine. It has also been shown to react with the diisopropylfluorophosphate (DFP) reactive site (157). We were able to demonstratate a complete loss of Xa activity in the FEIBA assay as a result of preincubation of Xa with a 600 M excess of p-APMSF. Although Xa activity was lost using the same experimental conditions there was little or no change in the Factor X coagulant activity (Table 8). Several problems exist, however, that could adversly affect the results of these experiments. First, it is possible that there might be a difference in the rate of inhibition of 61 purified factor Xa and the small amount of Xa that would theoretically be contaminiating the factor X product. Second, the half-life of p-APMSF is sufficiently short so that concern about the final concentration after a 1 hour incubation was.necessary. Finally, autocalalytic activation of the X molecule was a major concern during prolonged dialysis to remove excess inhibitor. For these reasons the following modifications were made in the experiment. In order to eliminate the effect of large variations in protein concentrations similiar molar concentrations of Factor X and Xa were used in the incubation mixture. Next, the concentration of p-APMSF was raised to 1 mM and an additional 1 mM was added after 30 minutes of incubation in order to counter any loss in activity due to the short half- life. The total incubation period was maintained at 1 hour. Finally, in.order to minimize autocalalytic activation, the inhibitor was removed on a Sephadex G-25 column and the factor X or Xa was assayed immediately for any change in activity. The results of these procedural modifications had little, if any, effect on the outcome of the FEIBA assay. Effect of Prothrombin Complex Proteins on FEIBA In order to determine if Factor II, VII or IX had FEIBA, assays were run at varying concentrations of these proteins. Each protein was tested individually and in combination with Factor X. Assays were performed according to the following protocol. 1) Both one minute and 40 minute 62 FEIBA assays were run on individual factors. 2) Both one minute and 40 minute FEIBA assays were performed on various Table 8. Effect of p-APMSF on Factor X and Factor Xa FEIBA Reaction Mixture APTT NAPTT (sec) (sec) Buffer Blank 104 462 Factor X 57 122 Factor X + p-APMSF 58 153 Factor Xa 27 44 Factor Xa + p-APMSF 84 291 Individual preparations were pretreated with a 600-mol excess of p-APMSF at 25°C for one hour and dialyzed overnight against 3 one-liter changes of 20 mM Mes-Tris, pH 6.0. 0.1 ml samples of factor X (l umollml) or factor Xa (11 nmol/ml) were assayed by APTT or NAPTT as described in the text using a factor VIII inhibitor plasma (45 Bethesda units) as substrate. factor X-factor II, VII or IX combinations. The rationale for this approach to the FEIBA assay was the same as previously described. If there was any enhanced activity we wanted to know if the acitivity could be blocked by naturally occuring inhibitors during a 40 minute incubation in substrate plasma. In all experiments the proteins were run over Sephadex G-25 prior to testing in order to remove the benzamidine in the storage buffer. Prothrombin alone or in combination with factor X had no effect on the FEIBA assay at concentrations that were as high as 10 times the normal plasma concentration. In addition, factor VII had no ability to enhance or inhibit the APTT at plasma concentrations that could be expected after administration of PCCTL There was, however, some ability to reduce the APTT at concentrations of factor VII that were 20 to 30 63 times normal. Factor IX alone, on the other hand, was able to signifigantly correct the APTT at 1 minute and in combination with factor X was able to enhance the factor X activity (Table 9). However, after 40 minutes of incubation, the factor IX acitivity was lost and the only apparent activity in the factor IX-X combination was that due to factor X. This is based on a comparison of the factor IX-X clotting time with that of factor X alone. These results suggest that the apparent activity was due to contamination with factor IXa and not to enhanced reactivity of factor IX with factor X. Table 9. Effects of II, IX and VII on the FEIBA Assay APTT (sec) 4 1 min 40 min Blank 96.8 Factor X 51 60 Factor IX 61.6 86 Factor VII 87.8 97.2 Factor II 95 Factor X + IX 37 51 Factor X + VII 53 63 Factor X + II 49 The FEIBA assay was performed according to the procedures outlined in the methods section. The following concentrations of each of the vitamin K dependent factors was used: factor X (0.4 U/ml), factor IX (0.4 U/ml), factor VII (0.6 U/ml) and factor II (0.67 umol). In the mixing experiments all factors were kept constant at the concentrations indicated. 64 Prothrombinase Assay It has been well established that certain criteria exist for the normal physiologic activation of the prothrom- bin molecule. These include Ca+2 , V(Va), phospholipid and Xa. We postulated that if factor X zymogen had intrinsic activity in the FEIBA assay then this activity would most likely be mediated via the components of the prothrombinase complex. In addition, if there was a difference in the kinetic constants of the X(Xa) mediated reactions then this would indicate a difference in the enzyme characteristics between factor X and Xa. ‘We were specfically'interested in showing the following characteristics; 1) that the Km of factor X for prothrombin was different than the Km of factor Xa for prothrombin. 2) When lipid was the limiting factor, factor X could compete with Xa for lipid binding sites. This would be reflected as a decrease in the rate of p- nitroanaline release, apparent inhibition. 3) The rate of release of p-nitroaniline, assumed to reflect activation of prothrombin, would not be affected when factor X was incu- bated with p-APMSF or AT-III but would be drastically decreased in the case of factor Xa. Kinetics of the Prothrombinase Complex In order to determine the KIn and Vmax for the factor X Ca+2 and Va concentrations and factor Xa system the lipid, were constant at 15 uM of lipid, 2.5 mM CaClZ and l U/ml of factor Va respectively. In all cases the concentration of 65 prothrombin was varied between values of 1/10 the predicted Km and 10 times the Km. The concentration of Xa was kept constant at 2-5 nM and the concentration of factor X was 0.5 uM. Factor Xa gave values of 0.281; 0.19 uM for the Km and 12.6 i 4.3 umol thrombin/min for the Vmax' Several difficulties arose in determining the kenitic constants for factor X. The initial values obtained, K m (L21 ¢;0.12 uM and Vma 11.6 i 3.5 nM thrombin/min, although X demonstrating a 1000 fold difference in the V , were not max significantly different when the Km values were compared with that for Xa (Figure 7). This could be due to contamination with Xa. To eliminate this possibility factor X was incubated with p-APMSF as previously described in this section. The control consisted of an aliquot of factor x incubated with methanol and an aliquot of factor Xa treated with both p-APMSF and methanol. After the samples had been applied to a Sephadex G-25 to remove excess inhibitor the proteins were immediately assayed for prothrombinase activity in order to determine kinetic constants. This particular procedure was followed to prevent autocatalytic activation of factor X which appeared to be an important Variable if prolonged dialysis was used to remove the inhibitor. The results of this experiment gave a Km for factor X cleavage of prothrombin of 2.6 x 10'8 M which was different by one order of magnitude from the Km for Xa cleavage of prothrombin (Figure 8). Incubation of the Xa control with p-APMSF again resulted in almost complete lose of activity and the factor X-methanol control demonstrated a 66 1.6.. 1.5.. a 1.4.- 1.3- 1.2I- '7 '7 A A 1.1- \ \ 1.0- a E 5 .o a 0.9- 8 s 2 ‘- 0.8)- : z I- I- . E 5 O'TF :: 3. v V >||< Ll, 0.6I- X 0.5- 0.4- O 0.3- 0.21-- e 0.1 l l , l l l l L I A; l l L l I -9 -6 -3 O 3 6 9 12 15 18 21 24 27 3O 33 36 (Prothrombin, 11M)" Figure 7. Double reciprocal plot of the initial velocity as a function of the prothrombin concentration. The assay mixture in a total volume of 1 ml consisted of factor Xa, 5.5 nmol/ml (O) or factor X, 0.5 umol/m1 (El), prothrombin 0.03 to 0.59 umol/ml and 15 umol/m1 of phospholipid. Calcium chloride, factor Va and S-2238 concentrations were as described under "Methods." The results shown are averages of 5 determinations for factor Xa and 3 determinations for factor X. 67 Km value very similiar to the results obtained for factor X that had not been preincubated with p—APMSF. mMThmmmmmmfl A I I I I l l I l l l I I l | | -33-3o-27-24-21-1s-15-12 -9 -6 -3 o 3 6 9 12 15 1s 21 24 21 so 33 as 39 42 45 4s 51 -1 (Prothrombin, 11M) Figure 8. Double reciprocal plot of the initial velocity as as a function of the prothrombin concetration. The assay conditions were identical to those described in figure 7 with the exception of the factor X concentration which was 0.35 umol/ml and the pretreatment of the factor X with p— APMSF followed by chromatography on Sephadex G—25. Factor X (D), Factor Xa (0). Lipid Binding Study In order to establish that factor X could in fact compete successfully with factor Xa for lipid binding sites we choose to conduct an expirement using limiting concentrations of phospholipid. In order to observe competitive binding between factor X and Xa rate limiting concentrations of phospholipid were necessary. A number of authors (141,142) have demonstrated the ability of native factor X to bind to the same kind of phospholipid surface as for prothrombin, factor Xa and factor Va. It appears 68 feasible, therefore, that factor X zymogen could compete with factor Xa for either substrate or accessory components ‘ and that under certain conditions where both protein species are present, the binding of zymogen to cofactors would be reflected in the pattern of steady state kinetics. Zur et al. have commented that in the combined presence of low activity zymogen and high activity enzyme and if an obligatory cofactor (i.e. phospholipid) is rate limiting, then the zymogen would act as an inhibitor (82). We carried out two sets of experiments. The first was conducted at low concentrations of phospholipid (0.05 umol/m1). As can be seen in figure 9, when Xa is held constant and factor X is varied there is a factor X dependent inhibition of the Xa rate reactions. The second experiment was conducted at a 10 fold higher concentration of phospholipid and the intent was to try and determine if the enzyme (Xa) and zymogen (X) activities were additive. Although there appeared to be a difference in the curve when compared to the low lipid concentration, it was not felt that the data conclusively demonstrated an additive effect. The primary reason for this lies in both the sensitivity of the assay system and the large difference in catalytic activity between factor X and Xa. In the chromogenic system the levels of factor X being used in this expirement do not show significant activity and are overshadowed by the Xa activity. 69 "M Thrombin . min" 1 I 1 I 1 I 1 1 1 I 1 1 1 1 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 Factor X ("W Figure 9. Inhibition of factor Xa enzymatic activity by factor X zymogen. The assay was performed in the presence of prothrombin (1.65 umol/ml), factor Xa (2.2 nmol/ml), factor X (0.0 to 0.59 umol/ml) and phospholipid (0.05 umol/ml). The other reactants were as described in the text. Inhibition of Prothrombinase Activity With p—APMSF Finally, we wanted to establish that the inhibition or lack of inhibition, of factor Xa and factor X, by p-APMSF and AT III was the same in the prothrombinase assay when compared to the FEIBA system. The initial experiments were conducted with a 600 mol excess of p-APMSF. However, because of the possibility of incomplete inhibition the inhibitor was increased to a final concentration of 2 mM. Both inhibitor concentrations gave similiar results. We observed inhibition of the initial reaction rate of factor Xa-prothrombinase activity, expressed as change in optical density in the chromogenic assay, equivalent to approximately 95% inhibition (Figure 10A). The same excess of 70 p-APMSF evidenced virtually no inhibition of the initial reaction rate of factor X zymogen (Figure 108). It is to be noted that following pretreatment of the native zymogen with p-APMSF, there was no significant lose of procoagulant activity when the zymogen was assayed in an RVVephospholipid activating system. 'This indicated that this particular inhibitor, at the concentrations being used, did not bind to the serine active site of the zymogen. Inhibition of Prothrombinase Activity with AT III Since factor Xa reacts with AT III with loss of activity, we compared the reaction of active enzyme and zymogen with the inhibitor. We employed experimental conditions whereby the inhibitor is removed from the assay system to avoid inhibition of the thrombin product.. This was accomplished by incubating factor Xa or zymogen with AT III-Sepharose following which the residual prothrombinase activity was determined. It is seen from Figure 11 that both activated factor X and zymogen lost enzymatic activity in the presence of AT III-Sepharose although the degree of activity lost by the zymogen was considerably less than the active enzyme. Quantitatively, the initial rates of reaction, expressed as change in optical density in the chromogenic assay, were inhibited 99% and 65% for factor Xa and zymogen, respectively, Several variables could affect the outcome of the experiment. 1) Contamination of the system with AT III leaching off the Sehparose. This 71 particular problem would be reflected most significantly in the factor X rate reaction because of the greatly decreased rate of prothrombin activation and the prolonged interval of time during which AT III could inactivate the IIa product. 2) Non-specifc adsorption of factor X to AT III-Sepharose would result in an apparent loss in activity. Although the major issue in this experiment was to demonstrate a difference in the inhibition of factor X and Xa, nonspecfic adsorption might be better demonstrated by first inactivating AT III with 1,2-cyclohexanedione and then attaching it to the Sepharose-4B. 'This control might help distinguish any effect not due to specfic binding of AT III to the active serine site. Two additional observations have been made that are based on differences in the factor X-Xa reaction. First, when assaying, in a FEIBA system, for the Xa or X mediated correction of the APTT a log-log plot of correction in seconds versus Xa activity results in a linear relationship. In contrast to this the same plot of factor X results in a curvilinear relationship with maximumIactivity at approximately 55 seconds. The activity appears to plateau at this level and increasing factor X concentrations do not result in any significant improvement in the APTT, It is possible that this difference is due to a much slower cofactor mediated actiVation of prothrombin. Specifically, IIa generation is considerably slower and the dependent activation of V‘to Va could very likely reflect this rate difference. The fact that Va generation plays a significant 72 role is demonstrated by the observation that trace amounts of IIa (concentrations not high enough to alter the FEIB activity by themselves) will potentiate the factor X mediated improvement in the APTT (86). The second observation is based on examining the time course of development of prothrombin activation products during both a factor X and factor Xa kinetic analysis. In order to monitor both the prothrombin cleavage products and the release of p-nitroaniline from the chromogenic substrate two identical reactions were run. The release of p- nitroaniline was monitored as described in the methods section under I'Prothrombin Activation Assay’. The prothrombin activation fragments were monitored by stopping the reaction at timed intervals by mixing 0.2 ml of the reaction mixture with 0.2 ml of 0.01 M phosphate buffer pH 7.2 containing 20 mM EDTA and 1% SDS. Each sample was then placed in a boiling water bath for one minute and 100 ul of this mixture was applied to 7.5% SDS-PAGE gels. Activation of prothrombin by factor Xa, in combination with lipid, Ca+2 and factor Va, resulted in the development of prothrombin intermediate 2, thrombin and a small amount of intermediate 1 within two minutes of the initiation of the reaction. The production of intermediate 2 as the major cleavage product is consistant with the presently accepted mechanism for cleavage of prothrombin by factor Xa. Small amounts of intermediate 1 would be expected in an in vitro system as a result of autocatalytic activation of prothrom- bin by thrombin. 73 0.18 P 016 - A 0.14 _ E 0.12 - C 8 0'10 — —— FactorXa v 7 000 - --- Factor Xa+ p-APMSF q . 0.06 - 4: 004 ”” J—r’ — 0.02 ..v’ ,”d’ A’Q’ I 1 1 1 I l 1 I 4_1 20 40 so so 100 120 140 160180 200 220 240 (sec)2 0.18 - 016 - B 0.14 - 0.12 - E g 0.10 - .. 2 0.08 - <1 0.06 - Factorx °-°4 ' --— FactorX+p—APMSF 0.02 l 1 l I I 1 I I 1 2 3 4 5 (min)2 Figure 10. The effect of the inhibitor p-APMSF on the time course of prothrombin activation by factor Xa (A) or factor X zymogen (B). Individual preparations were pretreated with a 600- mol excess of p-APMSF at 25°C for one hour and dialyzed overnight against 3 one—liter changes of 20 mM Mes—Tris, pH 6.0. The assay was performed in the presence of factor Xa (5.5 nmol/ml or factor X (0.7 umol/m1) and prothrombin (0.67 umol/ml). All other conditions are as given in the legend to figure 7. 74 I 0'1 2 Factor Xa 0.10 a ----— Factor Xa+ATIII A A410nm 0.06 * I 0.04 b a, I 0.02 I- a llllllLllxlllIlllLllw 5 1015 20 25 3° 35 4O 45 50 55 50 (mac)2 x 103 0.24 '- 0.22 " B P 0.20 "' p Factorx -—- Factorx +ATIII u 0.18 - / 0.10 - / 0.14 " / A A41Onm 0.10 - // / 0.08 P / A: 0.06 >- / 0.04 " / 0.02 —- / W l IIL1111111111111_1_J 1 8101214 1618 20 22 24 26 l l / l 1 l I 2 4 6 (min)2 Figure 11. The effect of AT III-Sepharose on the time course of prothrombin activation by factor Xa (A) of factor X zymogen (B). Experiments were performed in the presence of equal volumes of AT III Sepharose (9 mg protein/ml) and either factor Xa or factor X zymogen. The reactants were incubated at 25°C for 60 min and the AT III-Sepharose removed by filtration on a mini-column (1.4 x 6.5 cm). The final reaction mixture contained factor Xa “L6 nmol/ml), or factor X zymogen (0.27 umol/ml), prothrombin.(0.l6 umol/ml) and phospholipid (15.0 umol/ml), All other conditions are as described in the text. A control reaction mixture consisted of Sepharose that had been activated and treated in a manner identical to the AT III-Sepharose that had been activated and treated in a manner identical to the AT III- Sepharose but without the addition of the ligand. 75 When factor X was substituted for Xa in the prothrombinase complex the cleavage products that developed were markedly different in both the time course of development and the significance of the major band. In this case, after 10 minutes of incubation of the reaction mixture the only apparent change in the prothrombin molecule was the presence of a very faint band corresponding to intermediate 1. After 20 minutes of incubation there was still very little change in the prothrombin band but traces of interme- diate 2 and IIa were present. The trace of intermediate 1 seen after 10 minutes of incubation of prothrombin with factor X is most likely due to contamination of the system with IIa. .Although IIa is removed by adsorption onto benzamidine-Sepharose following activation of factor V it is likely that trace amounts are still present and this could account for the presence of the intermediate 1 band. The lack of an identifiable IIa band after prolonged incubation with factor X could be related to the lack of sensitivity of the PAGE technique when trying to detect small amounts of proteolytic activation products. .Alternatively, a » stoichiometric alteration in the prothrombin molecule could occur as a result of binding to the proteins of the prothrombinase complex resulting in an enhanced ability to cleave the chromogenic substrate and/or fibrinogen. 'The second possibility is similiar to the mechanism of the reported activation of prothrombin by staphylocoagulase or streptokinases activation of plasminogen (160-162). It would seem more likely, however, that the decreased rate of 76 reaction is due to the slow generation of small amounts of IIa. 77 DISCUSSION This work was initiated in an attempt to find a plausable explanation for the ability of the prothrombin complex concentrates to cause hemostasis in hemophiliacs with an inhibitor to factor VIII. The specific approach to the problem was based on the observation of Tishkoff that factor X and/or factor X in conjunction with IIa would correct the APTT or NAPTT of factor VIII deficient plasma that contained an inhibitor to factor VIII (86). The data presented in this paper supports the contention that factor X zymogen is the mediator of the coagulant activity observed in an intrinsic clotting assay system and can bypass the factor VIII inhibitor in vitro. It was speculated that the coagulant activity of native factor X was probably not attributable to the presence of trace contamination with active enzyme since preincubation of the zymogen with factor VIII inhibitor plasma, which contains a normal amount of AT III activity, resulted in incomplete inhibition of coagulant activity. The present study expands on this observation to include in-depth studies of the direct effect of AT III and a synthetic, highly specific, protease inhibitor, p-APMSF, on both the coagulant activity and steady state kinetics of factor X. In addition, the factor X molecule is critically examined for any possible alterations or cross contamination 78 with Xa. The effect of additional coagulant factors, II, VII and IX are evaluated, the V and Km for factor X is max determined and the activation fragments of II are evaluated for both a factor X and Xa catalyzed reaction. Enzymatic activity has been reported previously for other blood clotting zymogens such as factor VII and factor XII, although some investigators have questioned a physiolo- gic role for these relatively low-activity zymogens (163-164). Noteworthy in this regard are the observations of Kerr et al. on the existence of a preformed catalytic apparatus in the factor X zymogen as determined by the incorporation of specfic organic flouride inhibitors (165). In additition, recent work by Fair and Bahnak has demonstrated the existance of a single chain form of factor X that is produced by human hepatoma cells and then cleaved to the two chain form that is most often identified in plasma (166). In light of this observation, it may be necessary to reevaluate what is presently thought of as the mechanism of factor X activation. It would appear that not only is the cleavage of a small molecular weight peptide from the amino terminal end of the heavy chain required for factor X activation but, in addition, the factor X is first cleaved to a heavy and light chain structure. Therefore, factor X may exist in native plasma as what could be thought of as an intermediate form between the native, single chain, zymogen and the active enzyme form, Xa. This finding coupled with the fact that factor X is the only protein of the prothrombin complex that exists as a heavy and light 79 chain structure might suggest a rather unique function for this intermediate form of the factor X molecule. If the intermediate form of the factor X molecule is capable of catalytic activity it would be important to understand what conditions must exist for this activity to be expressed and what physiologic significance this activity has. Several different possibilities were considered to explain the in vitro proteolytic activity of the factor X zymogen. 1) Factor X zymogen is contaminated with factor Xa; 2) The zymogen enzymatic activity is attributable to an, as yet, unidentified intermediate of factor X present in low concentration: 3) Factor Xa catalyzes a feedback reaction on the substrate factor X, particularly in the presence of phospholipid and calcium; 4) Factor X zymogen has intrinsic enzymatic activity. To evaluate the first mechanism we used specfic, active site directed, inhibitors known to affect factor Xa activity. As can be seen from Figure 10 and Table 8 p-APMSF had no significant effect on either enzymatic rate reactions or on the coagulation assays. It should be noted that special precautions were taken to eliminate possible cross contamination of the zymogen preparations with active enzyme and also to prevent autocatalytic activation of factor X. We also controlled for the short half-life of p-APMSF as well as the effect of protein concentration on the rate of inhibition. We did not observe any change in the factor X activity when determined by the RVV’activating assay. 80 Although it has been shown that a decrease in the factor X activity occurs after incubation with some fluoride inhibitors (165) it was felt that because of the variation in the molecular structure of the inhibitor used in these experiments (p-APMSF) no significant comparisons could be drawn between our results and those using methanesulfonyl fluoride as the inhibitor. The AT III data shown in Figure 11 suggests one of two possibilities. Either the apparent slow rate of inhibition of the zymogen as compared with the active enzyme is due to a different kind of zymogen-antithrombin III interaction or, non-specifc adsorption of factor X to the Sepharose-AT III had occured. The latter possibility does not seem as likely as protein concentration before and after adsorption with AT III were almost identical., Antithrombin III was used in our study at much higher concentrations than those that are known to inhibit serine proteases of the coagulation system. Elsinger (78) has presented data that heparin plus AT III does not inhibit FEIBA under conditions where factor Xa is totally inhibited, and similiar findings have been reported by Barrowcliffe et al. (80). As another possible explanation for the apparent prothrombinase activity of factor X zymogen we attempted to identify trace component derivaties of the factor X zymogen that might account for the enzymatic activity. Recognizing that the limitations of analytic assay by SDS-PAGE are such that we were unable to visualize protein contaminants present in less than a 1% concentration, we found no 81 alteration in the factor X molecule on either normal gels or gels that had been loaded with excessive amounts of protein. In addition, the factor X was analyzed by amino-terminal sequencing to determine if any alteration in the protein had occurred as a result of our protein isolation technique. The results were consistant with the published literature and did not indicate any amino terminal alteration of factor x (148). Beta factor X would be a suitable candidate for an intermediate, enzymatically active; form of factor X and is formed by the cleavage of a small glycopeptide from the carboxyl-terminal end of the heavy chain of factor X by factor Xa. Although we have not sucessfully identified a zymogen intermediate, no zymogen of the coagulation system is known to be activated by COOH-terminal cleavage. A factor Xa feedback mechanism, occurring during the course of the reaction, could play a role in either the coagulation system or the steady state kinetics. Van Dieijen et al. (167) studied the factor Xa feedback mechanism of purified bovine proteins under optimal conditions. They found no extra factor Xa generated when Xa was incubated with factor X, calcium ions and phospholipid. Likewise, factor Xa and factor X were not seen on polyacrylamide gels in the presence of sodium dodecyl sulfate after 30 minutes of incubation. Noteworthy in this regard, Morrison and Jesty (168) did not observe release of activated peptide by brief treatment of tritiated factor X by Xa and moreover, Mertens and Bertina have shown that the action of factor Xa 82 on human factor X generates an enzymatically inactive derivative (169). Initial experiments that were conducted to differentiate the kinetic activity of factor X and Xa showed a 1000 fold variation in the maximum velocity but only a two to three fold difference in the Km. we had expected that if the factor X molecule was different in its enzymatic activity from Xa there should be a significant difference in the Km for the enzyme and zymogen. It was hoped that this difference would be at least one order of magnitude. The fact that the Km's were so close suggested the possibility of trace contamination with Xa or autocatalytic byproducts of factor X that could develop during the preincubation of the zymogen with p-APMSF and the subsequent dialysis to remove the inhibitor. In order to test this the mixture of inhibitor and factor X or Xa were applied to a Sepharose G— 25 column immediately following the preincubation to remove the inhibitor. It was felt that by eliminating the lengthy dialysis prior to testing, the possibility of autocalalytic acitvation or cleavage of factor X could by minimized. The Lineweaver-Burke analysis of data from this experiment gave a Km that was approximately one order of magnitude different than the enzyme Xa. It should be noted, however, that the Km varied in a direction that was opposite to what would have been predicted. Rather than the zymogen form requiring more substrate to reach half-maximum velocity it required less. This variation could be accounted for based on differences in the stoichiometry of the proteins involved in 83 the prothrombinase complex. Because factor X has, at this point, not undergone amino-terminal cleavage to the factor Xa molecule steric interference could limit the amount of prothrombin capable of binding to the complex. This in turn would be reflected in an apparent decrease in the Km for the prothrombin substrate. The expression of zymogen prothrombinase activity may have physiologic implications from our observation that the zymogen competes with factor Xa for phospholipid binding sites (Figure 9). This data is also consistent with our observation that phospholipid is obligatory for apparent zymogen enzymatic activity and, in addition, explains why increasing concentrations of factor X will inhibit the factor Xa clotting activity (data not shown). As can be seen in figure 10 as the concentration of factor X was increased there was a decrease in the rate of the enzymatic reaction with factor Xa that reached a maximum at approximately 0.36 umol of factor X. Further increases in factor X did not result in a decrease in the reaction rate. We have observed a total absence of factor X peptidase activity toward the factor Xa chromogenic substrate S-2222 within the sensitivity of our experiments (data not shown). Notwithstanding, we propose that apparent zymogen enzymatic activity demonstrated under experimental conditions in this paper reflects the enhanced reaction rate of the prothrombinase complex that may result from concentration on the phospholipid vesicle surface of the enzyme (factor X) 84 and substrate (prothrombin). We conclude from the present studies that the apparent enzymatic activity of the zymogen can initiate clotting under conditions where plasma levels of factor X are markedly elevated as occurs postinfusion in patients treated with PCC. 85 APPENDIX APPENDIX 1 The kinetic evaluation of the activation of prothrombin was based on an assay system developed by Kosow et al. (160,170). This assay is a continuous spectrophotometric assay and monitors the release of p-nitroaniline from the thrombin specific substrate S-2238. The Lineweaver-Burke plots that were obtained from this assay system were linear, indicative of Michaelis-Menten kinetics. The molar extinction coefficient (a) of p-nitroanilide is 9400 M"lcm'1 at 410 nm. This is based on the fact that a 1M solution of p-nitroaniline in a cuvette with a 1 cm light path will give an absorbance at 410 nm of 9400. Therefore, the absorbance of a solution of p-nitroaniline at 410 nm divided by 9400 gives the molarity of p-nitroaniline. The first-order rate constant (k) for the hydrolysis of S-2238 by IIa is the moles of p-nitroaniline released per mole of thrombin per unit time. This reaction is a simple single enzyme substrate reaction of the form: k1 k2 E+S fiES >P+E where E is enzyme, S is substrate, ES is an intermediate enzyme substrate complex and P relates to the production of product. k1, k_1 and k2 are rate constants and describe how the concentration of the enzyme-substrate will change with time. Therefore, the rate at which ES will be formed is described by kllEJIS], the rate of formation of product is described as kzlES] and finally, the rate of conversion of ES back to E and S would equal k_IIES]. An equation can be written that will define the change in [ES] as a result of the change in time. d[ES]/dt = kl[E][S] - k_1[ES] - kzlES] (1) In order to derive the Michaelis-Menten equation a term that describes total enzyme concentration [Et] is required. This is expressed as: [Et] = [E] + [ES] (2) which when used to express [E] can be be rearranged to: E = [Et] — [ES] (3) If this value is substituted into equation (1) the rate of change in [ES] expressed as a function of t becomes: 87 d[ES]/dt = k1([Et] - [ES])[S] - k_1[ES] - k2[ES] (4) Of d[ESJ/dt = k1([Et] - [ES])[S] - (k_1 + k2)[ES] <5) If the steady state assumption of Briggs and Haldane (171) is used then the rate of change of [E] and [ES] with time is .zero in comparison to the rate of change of either [S] or [P] and therefore: d[ES]/dt = 0 (6) and, k1([Et] - [ES])[S] - (k_1 + k2)[ES] = 0 (7) expanding this eaquation gives: kllEt][S] ‘ k1[ES][S] ' k_1[ES] “ kzlES] = 0 (8) k1[Et][S] = (k1[S] + k_1 + k2)[ES] <9) and solving for [ES] results in: kIIEtJIS] [ES] = (10) k1[S] + k_1 + k2 dividing by kl leaves Et[S] [ES] = (11) [S] + k_l + k2/kl From the first rate constant discussed it is known that v = k2[ES] therefore, if you multiple through by kzthe expression becomes: k [E ][S] v = 2 t (12) [S] + k_1 + k2/k1 88 The expression in the denominator that contains only rate constants is referred to as the Km and in the special instance in which the [S] is equal to the Km the reaction velocity is at half maximum. If the substrate concentrations were very large relative to the Km, then: and at very high substrate concentrations the maximal velocity Vmax is approximated. Therefore: vmax = kzlEt] <14) if this expression is substituted into equation (12) it becomes: V [S] max v = (15) Km + [S] This is the Michaelis-Menten equation and describes a rectangular hyperbola. Although this equation allows for the determination of the Vmax and Km for a particular enzyme substrate reaction it is more difficult to use for analysis. Linear transformations have been developed and the most common of these in the Lineweaver-Burke or double-reciprocal plot. + K /V l/[S] (l6) l/V = 1/Vmax m max Where l/v is plotted vs. l/[S] and the y intercept is equal to 1/Vmax and the x intercept is -1/Km. This particular plot was choosen as it is most often used in enzymology 89 although it may not be as statistically valid as other linear transformations. Based on these equations, the first order rate constant for IIa cleavage was determined using a constant enzyme concentration of 5.8 nmol/ml of IIa and a substrate concentration that varied form 1.6 x 10'4 mol/ml to 2.1 x 10‘5‘mol/ml. The Vma obtained from a double x reciprocal plot of this data was then fitted into the equation: (A410/sec)/9400 (17) 75' II mmoles IIa The reaction mixture in this case was in a total volume of l ml and duplicates the buffer system used in the prothrombinase assay. Factor IIa catalyzes the cleavage of H-D-Phe-Pip-Arg-pNA (S-2238) and this cleavage is monitored at 410 nm. The production of IIa is catalyzed by factor Xa, a serine protease, and this activation is markedly accelerated by the addition of factor va, phospholipid, and Ca+2 to from what is referred to as the prothrombinase complex. When S-2238 is added to the reaction mixture their is a parabolic increase in absorbancy with time that is indicative of a constant rate of IIa production. This curve conforms to the equation: x = vot + 1/2at2 (18) where x is absorbancy, V0 is the velocity at time zero and in this system correlates with endogenous activity or activity not due to specific enzyme addition, t is time and 90 a is acceleration. This equation is used to describe motion with constant acceleration. When t = 0 the instantaneous speed v is V0 and since acceleration is assumed to be constant v will increase in proportion to the time. If we were considering a straight-line speed-time graph the average acceleration would equal Av/At. If this was applied to an entire time internal to to t, which would correspond to v o and v respectively, the average acceleration would be replaced with a constant value for acceleration where: v - vo a =--—-- (19) t or v = vo + at (20) This equation, although describing a straight speed-time graph, does not describe the parabolic curves obtained during the activation of prothrombin. In this particular instance because the rate of increase of speed is uniform the average speed over an interval of this curve is (v0 4» v)/2. The change in absorbancy (x) or the distance covered is expressed then as the average speed multiplied by the elapsed time. x =-———————-t (21) If equation 20 is substituted for v: vo + (vo + at) x = t (22) 2 91 or v x = vot + 1/2at2 (23) This equation conforms to the equation for a straight line, y = a + bx, and a1 lowsthe determination of acceleration or change in absorbancy in a given time. If you subtract the endogenous activity of the mixture (vot) from the absorbancy (x) and plot this value vs. t2 you obtain a straight line with a slope (b) equal to a/2. This equation is valid only as long as IIa is being produced at a constant rate. Therefore, the reactions were followed for only short periods of time and only data having a correlation coefficient of greater then 0.995, as determined from the plots of abosrbancy vs. t2, was used. The slope of the line obtained from this plot, which equals a/2, is proportional to the rate of IIa formation. This value was then used to determine the velocity of the reaction via the equation: (M)/t = a/(ek/Z) (24) where a is acceleration, e:is the extinction coefficient of p-nitroaniline and k is the first order rate constant for the cleavage of S-2238 by IIa. The concentrations of phospholipid, Va, Ca+2 and enzyme or zymogen were kept constant as described in the "Materials and Methods" section and the concentration of II was varied. The velocity of the reaction was determined at each concentration of II and the Km and Vmax for either enzyme or 92 zymogen was obtained from a double reciprocal plot of the velocities and substrate concentrations. 93 LIST OF REFERENCES 10. 11. LIST OF REFERENCES Ruggeri, Z.M. Natural History of 32 Factor VIII Inhibitors in Haemophiliacs. Workshop on Inhibitors of Factor VIII and IX, Vienna, Austria, 1976. Allain, JJL, Frommel, D. Patterns of Immune Response in Classical Haemophilia. Workshop on Inhibitors of Factor VIII and Ix, Vienna, Austria, 1976. Biggs, R. Haemophilia Treatment in the United Kingdom from 1969 to 1974, Br. J. Haematol., 35:487, 1977. Shapiro, S.S. Antibodies to Blood Coagulation Factors. Clinics Haematol., 8:207, 1979. Andersen, B.R., Terry, W.D. Gamma G4 Globulin Antibody Causing Inhibition of Clotting Factor VIII. Nature, 212:174-175, 1968. Robboy, S.J., Lewis, E.J., Schur, P.H., Colman, R.W. Circulating Anticoagulants to Factor VIII. Am J. Med., 49:742-752, 1970. Kavanagh, M.L., Wood, C.N., Davidson, J.F. The Immunochemical Characterization of Human Antibodies to Factor VIII Isolated by Immuno-Affinity Chromatography. Thrombos. Haemostas., 45:60, 1981. Lavergne, J-M, Meyer, D., Reisner, H. Characterization of Human Anit-factor VIII Antibodies Purifired yby Immuno Comlex Formation. Blood, 48:931, 1976. McKelvey, E.M., Kwan, H.C. An IgM Circulating Anticoagulant with Factor VIII Inhibitory Activity. Ann. Int. Med., 77:575, 1972. Yang, HJL, Kuzur. M. Procoagulant Specificity of Factor VIII Inhibitor. Br. J. Haematol., 37:429-433, 1977. Kernoff, PJLA. Procine Factor VIII: Preparation and Use in Treatment of Inhibitor Patients. In Hoyer. an. (ed): "Factor VIII Inhibitors", Alan R. Liss, Inc., p.207, 1984. 94 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. Mariani, G., Russo, M.A., Mandelli, F. (eds): "Activated Prothrombin Complex Concentrates: Managing Hemophilia with Factor VIII Inhibitor", Praeger, 1982. Meyer, D. Specificity and Structure of Factor VIII and Factor IX Antibodies in Haemophiliacs. In Seligsohn, U., Rimon, A., Horoszowski, H. (eds): “Haemophilia”, Alan R. Liss, Inc., p. 69, 1981. Fronunel, D., Allain, J.P. Genetic Predisposition to Develop Factor VIII Anitbody in Classic Hemophilia. Clin. Immunol. Immunopathol., 8:34-38, 1977. Frommel, D., Allain J.P., Saint-Paul, E., Noel, E., Mannucci, F.P., Pannicucci. F., Blomback, M., Prou- Wartelle, O., Muller, J.Y. Thromb. Haemostas., 46:687- 689, 1981. Meyer, D. Specificity and Structure of Factor VIII and Factor IX Antibodies in Haemophiliacs. In Seligsohn, U., Timon, A., Horoszowski. H. (eds): "Haemophilia", Alan R. Liss, Inc., p.69, 1981. Allain, JJA, Frommel, D. Antibodies to Factor VIII specificity and Kinetics of Iso- and Hetero-antibodies in Hemophilia A. Blood, 44:313-322, 1974. Biggs, R., Austen, D.E.G., Denson, K.W.E., Rizza, C.R., Borrett, R. The Mode of Action of Antibodies Which Destroy Factor VIII. Br. J. Haematol., 23:125-135, 1972. Biggs, R., Austen, D.E.G., Denson, K.W.E., Borrett, R., Rizza, (LR. The Mode of Action of Antibodies Which Destroy Factor VIII. Br. J. Haematol., 23:137-155, 1972. Gawryl, DLS., Hoyer, IuW. Inactivation of Factor VIII Coagulant by Two Different Types of Human Antibodies. Blood, 60:1103-1109, 1982. Shapiro, SAL The Immunologic Character of Acquired Inhibitors of Antihemophilic Globulin (Factor VIII) and the Kinetics of Their Interaction with Factor VIII. J. Clin. Invest., 46:147-156, 1967. Green, D. Spontaneous Inhibitors of Factor VIII. Br. J. Haematol., 15:57-75, 1968. Shapiro, SAL, Hultin, M. Acquired Inhibitors to the Blood Coagulation Factors. Semin. Thromb. Hemostas., 1:336-385, 1975. Allain, J.P., Gaillandre, A., Frommel, D. Acquired Haemophilia: Functional Study of Antibodies to Factor VIII. Thrombos. Haemostas., 45:285-289, 1981. 95 25. 26. 27. 28. 29. 300 31. 32. 33. 34. 35. Croissant, M.P., Zuzel, M., Allain, J.P. Heterogeneity of Autoantibodies to Factor VIII: Differences in Specificity for Apparently Distinct Antigenic Determinants of Factor VIII Coagulant Protein Blood, 62:133-140, 1963. Kasper, C. A More Uniform Measurement of Factor VIII Inhibitors. Thrombos. Diathes. Haemorrh., 34:869-872, 1975. Kasper, C., Pool, J.G. Measurement of Mild Factor VIII Inhibitors in Bethesda Units. Thrombos. Diathes. Haemorrh., 34:875-876, 1975. Peake, I.R., Bloom, A.L., Giddings, J.C., Ludlam, C.A. An Immunoradiometric Assay for Procaogulant Factor VIII Antigen: Results in Haemophilia, von Willebrandfia Disease and Fetal Plasma and Serum. Br. J. Haematur 42:269-281, 1979. Furlong, R.A., Peake, I.R., Bloom, A.L. Factor VIII Clotting Anitgen (VIIICAg) in Haemophilia Measured by Two Immunoradiometric Assays (IRMA) using Different Anitbodies and the Measurement of Inhibitors to Procoagulant Facotr VIII (VIIIC) by IRMA. Br. J. Haemat., 48:643-650,.l981. Green, D. Suppression of a Antibody to Factor VIII by a Combination of Factor VIII and Cyclophosphamide. Rizza, C.R., Edhcumbe, J.O.P., Pitney, W.R., Child, CLA. The Treatment of Patients Having Spontaneously Occurring Anitbodies to Antihaemophilic Factor (Factor VIII). Thrombos. Disthes. Haemorrh., 28:120-128, 1972. Hultin, M.B., Shapiro, S.S., Bowman, H.S., Gill, F.M., Andrews, A.T., Martinez, J., Eyster, M.B., Sherwood, VLC. Immunosuppressive Therapy of Factor VIII Inhibitors. Blood, 48:95-108, 1976. Dormandy, K.M. Immunosuppression in the Treatment of Hemophiliacs with Anitbodies to Factor VIII. Preceeding of the IXth Congress of the World Federation of Hemophilia (Istanbul), Excerpta Medica, 225-239, 1975. Hultin, M. Management of Inhibitors by Immunosuppression. In Seligsohn, U., Rimon, A., Horoszowski, H. (eds): "Haemophilia", Alan R. Liss, Inc., p.107, 1981. Schein, P.S., Winokur, S.H., Immunosuppressive and Cytotoxic Chemotherapy: Long Term Complications. Ann. Int. Med., 82:84-95, 1975. 96 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. Brackmann, ELH., Egli, H. Treatment of Haemophilia Patients with Inhibitors. Seligsohn, EL, Rimon, A., Horoszowski, H. (edsJ: “Haemophilia", Alan R. Liss, Inc., p. 113, 1981. Stenbjerg, S.,.Jorgensen, J. Low Dose Factor VIII in the Elimination of Inhibitors. Presented at the XIV Congess of the World Federation of Hemophilia, San Jose, CR, 1981 (abstr. SI-6). Gormsen, J., Dalsgaard-Nielsen, J. High doses of Factor VIII in Haemophiliacs with Inhibitors, High Responders. Presented at the XIV Congress of the World Federation of Hemophilia, SanJose, CR, 1981 (abstr I-9). Lechner, K., Korninger, C., Niesser, H., Thaler, E., Francesconi, M. Suppression of Hemophilic Antibodies by Continuous Treatment With Factor VIII Concentrates and Prothrombin Complex Preparation. Presented at the 18th Congress of the International Society of Hematology, Montreal, Canada, 1980 (abstr 1047). White, G.C., Taylor, R.E., Blatt, P.M., Roberts, H.R. Treatment of a High Titer Anti-Factor-VIII Antibody by Continuous Factor VIII Administration: Report of a Case. Blood, 62:141-145, 1983. Azner, J.A., Jorquera, J.I., Peiro, A., Garcia, 1. The Importance of Corticoids Added to Continued Treatment with Factor VIII Concentrates in the Suppression of Inhibitors in Haemophilia A.. Thromb. Haemostas., 51:217-221, 1984. Brackmann, HJL, Gormsen, J. Massive Factor VIII Infusion in Haemophilic Patents with Factor VIII Inhibitor, High Responder. Lancet, ii:933, 1977. Mayne, E.E., Madden, M., Crothers, I.S., Ingles, T. Highly Purified Procine Factor VIII in Haemophilia A with Inhibitors to Factor VIII. Br. Med. J., 282:318, 1981. Gatti, In, Mannucci, PuM. Use of Porcine Factor VIII in the Management of Seventeen Patients with Factor VIII Antibodies. Thromb. Haemostas., 51:379-384, 1984. Manache, D. Factor IX Concentrates. Thrombos. Diathes. Haemorrh., 33:600-605, 1975. Chandra, S., Wickerhauser, M. Large Scale Preparation of nonthrombogenic Prothrombin Complex. Thromb. Res., 12:571-582, 1978. 97 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. Prowse,C.V” Chirnside,A.,Elton,ILA.: In Vitro Thrombogenicity Tests of Factor IX Concentrates. l. A Survey of Available Assays. Thromb. Haemostas” 42:1355-1367, 1979. Prowse, C” Pepper,S.S. In Vitro Tests of the Potential Thrombogenicity of Factor IX Concentrates: Inhibition and Characterisation Studies of NAPTT, TGtso and PF3 Moieties. Thromb. Res., 20:491-498, 1980. Sas, G., Owens, R.E., Smith, J.K., Middleton, 8., Cash, J;D. In Vitro Spontaneous Thrombin Generation in Human Factor-IX Concentrates. Br. J. Haemat., 31:25—35, 1975. Kingdon,H.Sq Lundblad,R.L” Veltkamp,J.J.Aronson, DJ“ Potentially Thrombogenic Materials in Factor IX Concentrates. Thromb. Diath. Haemorrh., 33:617-631, 1975. Wessler, S” Reimer, S.M.,Sheps, M.C. Biologic Assay of a Thrombosis Inducing Activity in Human Serum. J. Appl. Physiol., 14:943, 1959. Prowse, C.V., Williams, A.E. A Comparison of the In Vitro and In Vivo Thrombogenic Activity of Factor IX Concentrates Using Stasis (Wessler) and Non-Stasis Rabbit Models. Thromb. Haemostas., 44:81—86, 1980. Aronson, D.L. Factor IX Complex. Semin. Thromb. HemOSt., 6:28-43, 1979. Hoofnagle, JJL, Aronson, D., Roberts, H. Serologic Evidence for Hepatitis B Virus Infection in Patients with Hemophilia B. Thromb. Diathes. Haemorrh., 33:606- 609, 1975. deShazo, RJL, Andews, WJL, Nordberg, J” Newton, J” Daul, C., Bozelka, B. An Immunologic Evaluation of Hemophiliac Patients and Their Wives. Ann. Int. Med., 99:159-164, 1983. Schupbach, J., Popovic, M., Gilden, R.V., Gonda, M.A., Sarngadharan, MAL, Gallo, RJL, Serological Analysis of a Subgroup of Human T-lymphotropic Retroviruses (HTLV-III) Associated With AIDS. Science, 224:503-505, 1984. Sarngadharan, M.G., Popvic, M., Bruch, L., Schupbach, J., Gallo, R.C. Antibodies Reactive with a Human T- lymphotropic Retrovirus (HTLV-III) in the Serum of Patients with AIDS. Science, 224:506—508, 1984. 98 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. Schupbach, J., Haller, O., Vogt, M., Luthy, R., Joller, H., Oelz, O., Popovic, M., Sarngadharan, M.G., Gallo, R.C. Antibodies To HTLV-III in Swiss Patients With AIDS and Pre-Aids and in Groups at Risk for Aids. N. Engl. J. Med., 312:265-270, 1985. Kasper, C.K. Postoperative Thromboses in Hemophilia B. N. Engl. J. Med., 289:160, 1973. Steinberg, M.H., Dreiling, B.J. Vascular Lesions in Hemophilia B. N. Engl. J. Med., 289:592, 1973. Edson, J.R. Prothrombin Complex Concentrates and Thromboses. N. Engl. J. Med., 290:403, 1974. Marches, S.Z., Burney, R. Prothrombin Complex Concentrates and Thromboses. N. Engl. J. Med., 290:403-404, 1974. ' Menache, d., Roberts, H.R. Summary Report and Recommendations of the Task Force Members and Consultants. Thrombos. Diathes. Haemorrh., 33:645-647, 1975. Kasper. C.K., Thromboembolic Complications. Thrombos. Diathes. Haemorrh., 33:640-644, 1975. Cash, J.D., Dalton, R.G., Middleton, S., Smith, J.K. Studies on the Thrombogenicity of Scottish Factor IX Concentrates in Dogs. Thrombos. Diathes. Haemorrh., 33:632- 639, 1975. Gruppo, R., Bone, K., Donaldson, V. Fetal Myocardial Necrosis Associated with Prothrombin Complex Concentrate Therapy in Hemophilia A. N. Engl. J. Med., 309:242, 1983. Agrawal, B., Zelkowitx, L., Hletko, P. Acute Myocardial Infarction in a Young Hemophilic Patient During Therapy with Factor IX Concentrate and Epsilon Amino Caproic Acid. J. Ped., 98:931-933, 1981. Mitra, G., Coan, M.B., Wada, S. Blood-Coagulation Promoting Products and Methods of Preparing Them. U.S. Patent I 4,391, 746, 1983. Eibl, J., Schwarz, 0., Elsinger, F. Method of Producing a Blood-Coagulation-Promoting Preparation from Human Plasma. U.S. Patent $4,160,025, 1979. Lusher, J.M., Sandor, 8.8., Palascak, J.E., Rao, A.V., Levine, P.H., Blatt, P.M. Efficacy of Prothrombin- Complex Concentrates in Hemophiliacs with Antibodies to Factor VIII. N. Engl. J. Med., 303:421-425, 1980. 99 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. Sjamsoedin, ILJ.M., Heignen, L., Mauser-Bunschoten, Eda, van Geijlseijk, JJL, van Houwelingen, H. van Asten, P., Sixma, J.J. The Affect of Activated Prothrombin Complex Concentrate (FEIBA) on Joint and Muscle Bleeding in Patients with Hemophilia A and Antibodies to Factor VIII: A Double-Blind Clinical Trial. N. Engl. J. Med., 305:717-721, 1981. Lusher, J.M., Blatt, P.M., Penner, J.A., Aledort, L.M., Levine, P.H., White G.C., Warrier, A.I., Whitehurst, D.A. Autoplex vs Proplex: A Controlled, Double Blind Study of Effectiveness in Acute Hemarthrosis in Hemophiliacs with Inhibitors to Factor VIII. Blood Blatt, P.M., Menache, D., Roberts, H.R. A Survey of the Effectiveness of Prothrombin Complex Concentrates in Controlling Hemorrhage in Patients with Hemophilia and Anti-factor VIII Antibodies. Thromb. Haemost., In Press. Pollock, A., Lewis, M.J. Factor-VIII Inhibitor Bypassing Activity. Lancet, 2:43-44, 1976. Mannucci, ItM., Bader, P., Fuggeri, Z.M. Concentrates of Clotting-Factor Ix. Lancet, 1:41, 1976. Parry, D.M., Bloom, A. Failure of Factor VIII Inhibitor Bypassing Activity (Feiba) to Secure Haemostasis in Haemophilic Patients with Antibodies. J. Clin. Pathol., 31:1102-1105, 1978. Blatt, P.M., White, G.C. II, McMillan, C.W., Roberts, ILR. Treatment of Anti—factor VIII Antibodies. Thromb. Haemost., 38:514-523, 1977. Elsinger, F. FEIBA Immuno: A Preparation With Factor Eight Inhibitor Bypassing Activity. Presentation, XIIth Congress of The World Federation of Hemophilia, Vienna, Austria, 1977. Tishkoff, G.H., Hess, K. Personal communication. Barrowcliffe, T.W., Kemball-Cook, G., Gray, E. Binding to Phospholipid Protects Factor VIII from Inactivation by Human Antibodies. J. Lab. Clin. Med., 101:34-43, 1983. Bauer, K.A. The Rapid Inhibition of Thrombin and Factor Xa Within the Circulatory System of Humans. Clin. Res., 31:534A, 1983. Zur, M., Radcliffe, R.D., Oberdick, J., Nemerson, Y. The Dual role of Factor VII in Blood Coagulation. J. Biol. Chem., 257:5623-5631, 1982. 100 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. Morgan, P.H., Robinson, N.G., Walsh, K.A., Neurath, H. Inactivation of Bovine Trypsinogen and Chymotrypsinogen by Diisopropylphosphorofluoridate. Proc. Nat. Acad. Sci. USA, 69:3312—3316, 1972. Morgan,P.Hu Walsh,K.A” Neurath,H. Inactivation of Trypsinogen by Methane Sulfonyl Fluoride. FEBS Letters, 41:108—110, 1974. Nesheim, M.E., Taswell, J.E., Mann, K.G. The Contribution of Bovine Factor V and Factor Va to the Activity of Prothrombinase. J. Biol. Chem., 254:10952- 10962, 1979. Tishkoff, G.H. Prothrombin Complex Concentrates, Thromboses, and Factor VIII Inhibitors: Evidence for a Coagulant Formed by the Interaction of Factors X and II. Thromb. Diathes. Haemorrh., 34:589, 1975. Stenflo,J.,Fernlund,P.,Egan,W.,Roepstorff,P. Vitamin K Dependent Modifications of Glutamic Acid Residues in Prothromin. Proc. Nat.‘Acad. Sci. USA, 71:2730-2733, 1974. Suttie, Juw. Control of Prothrombin and Factor VII Biosynthesis by Vitamin K. Arch. Biochem. Biophys” 118:166—171, 1967. Bell, ILG., Matschiner, J}T. Synthesis and Destruction of Prothrombin in the Rat. Arch. Biochem. Biophys” 135:152-159, 1969. Suttie, J.W. The Effect of Cycloheximide Administration on Vitamin K-Stimulated Prothrombin Formation. Arch. Biochem. Biophys., 141:571-578, 1970. Larson, A.E., Friedman, P.A., Suttie, J.W. Vitamin K- dependent Carboxylase. J. Biol. Chem., 256:11032-11035, 1981. Wallin, R., Hutson, 8. Vitamin K-dependent Carboxylation. J. Biol. Chem., 257:1583-1586, 1982. Girardot, J-M. Vitamin K-depentent Carboxylase. J. Biol. Chem., 257:15008-15011, 1982. Brozovic, M. Oral Anticoagulants, Vitamin K and Prothrombin Complex Concentrates. Br. J. Haemat” 32:9-12, 1976. Stenflo, J., Ganrot, P—O. Vitamin K and the Biosnythesis of Prothrombin. J. Biol Chem., 247:8160— 8166, 1972. 101 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. Stenflo, J. Vitamin K and the Biosynthesis of Prothrombin. J. Biol. Chem., 247:8167-8175, 1972. Nelsestuen, G.L., Suttie, J.W. Mode of Action of Vitamin K. Calcium Binding Properties of Bovine Prothrombin. Biochemistry, 11:4961-4964, 1972. Nelsestuen, G.L., Suttie, J.W. The Purification and Properties of an Abnormal Prothrombin Protein Produced by Dicumarol-treated Cows. J. Biol. Chem., 247:8176- 8182, 1972. Esmon, C.T., Suttie, J.W., Jackson, C.M. The Functional Significance of Vitamin K Action. J. Biol. Chem., 250:4095-4099, 1975. Stenflo, J. Vitamin K and the Biosynthesis of Prothrombin. J. Biol. Chem., 249:5527-5535, 1974. Hemker, H.C., Muller, A.D. Kinetic Aspects of the Interaction of Blood-Clotting Enzymes. Thrombos. Diathes. Haemorrh., 20:78-87, 1968. Hemker, H.C., Veltkamp,‘J.J., Hensen, A., Loeliger, ELA. Nature of Prothrombin Biosynthesis: Preprothrombinaemia in Vitamin K-deficiency. Nature, 9:589-590, 1963. Stenflo, J. Vitamin K and the Biosynthesis of Prothrombin. J. Biol. Chem., 249:5527-5535, 1974. Nelsestuen, G.L., Zytkovicz, T.H., Howard, J.B. The Mode of Action of Vitamin K. J. Biol. Chem., 249:6347- 6350,1974. Jackson, CLM., Nemerson, Y. Blood Coagulation. Ann. Rev. Biochemistry, 49:765-811, 1980. Downing, M.R., Butkowski, R.J., Clark, M.M., Mann, K.G. Human Prothrombin Activation. J. Biol. Chem”, 250:8897- 8906,1975. Esnouf, MQP. The Prothrombin-Converting Complex. Biochem. Soc. Trans., 570th Meeting Cardiff, 5:1244- 1247, 1977. Paphadj0poulos, D., Hanahan, B.J., Observations on the Interaction of Phospholipids and Certain Clotting Factors in Prothrombin Activator Formation. Biochim. Biophys. ACTA, 90:436-439, 1964. 102 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. Benson, B.J., Kisiel, W., Hanahan, D.J. Calcium Binding and Other Characteristics of Bovine Factor II and Its Activation Intermediates. Biochim. Biophys. ACTA, 329:81-87, 1973. Gitel, S.N., Owen, W.G., Esmon, C.T., Jackson, C.M. A Polypeptide Region of Bovine Prothrombin Specific for Binding to Phospholipids. Proc. Nat. Acad. Sci. USA, 70:1344-1348, 1973. Nelsestuen, G.L., Suttie, J.W. The Mode of Action of Vitamin K. Isolation of a Peptide Containing the Vitamin K-Dependent Portion of Prothrombin. Proc. Nat. Bajaj, S.P., Nowak, T., Castellino,F.J. Interaction of Manganese with Bovine Prothrombin and Its Thrombin- Mediated Cleavage Products. J. Biol. Chem., 251:6294-6299, 1976. Bajaj, S.P., Butkowski, R.J., Mann, K.G. Prothrombin Fragments. J. Biol. Chem., 250:2150-2156, 1975. Furie, H.C., Mann, K.G., Furie, B. Substitution of Lanthanide Ions for Calcium Ions in the Activation of Bovine Prothrombin by Activated Factor X. J. Biol. Chem., 251:3235-3241, 1976. Benson, B.J., Hanahan, D.J., Structural Studies on Bovine-Prothrombin. Isolation and Partial Characterization of the Ca+ Binding and Carbohydrate Containing Peptides of the Neterminus Region. Biochemistry, 14:3265-3277, 1975. Prendergast, F.G., Mann, K.G. Differentiation of Metal Ion-Induced Transitions of Prothrombin Fragment 1. J. Biol. Chem., 252:840-850, 1977. Jackson, C.M., Brenckle, G.M. Divalent Ion Binding to Bovine Prothrombin Fragment l and its Consequences. In: The Regulation of Coagulation. ed. Mann and Taylor, Elsevier North Holland Inc., 11, 1980. Furie, BJL, Furie, B. Interaction of Lanthanide Ions with Bovine Factor x and Their Use in the Affinity Chromatography of the Venom Coagulant Protein of Vipera russelli. J. Biol. Chem., 250:601-608, 1975. Henriksen, R.A., Jackson, C.M. Cooperative Calcium Binding By the Phospholipid Binding Region of Bovine Prothrombin: A Requirement for Intact Disulfide Bridges. Arch. Biochem. Biophys., 170:149-159, 1975. 103 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. Bajaj, S.P., Byrne, R., Nowak, T., Castellino, F.J. Interaction of Manganese with Bovine Factor x. J. Biol. Chem., 252:4758-4761, 1977. Jackson, C.M., Peng, C., Brenckle, G.M., Jones, A., Stenflo, J. Multiple Modes of Association in Bovine Prothormbin and Its Proteolysis Products. J. Biol. Chem., 254:5020-5026, 1979. Tarvers, R.C., Noyes, C.M., Roberts, H.R., Lundblad, RJL Influence of Metal Ions on Prothrombin Self- association. J. Biol. Chem., 257:10708-10714, 1982. Lewis, R.M., Furie, B.C., Furie, B. Conformation Specfic Monoclonal Antibodies Directed Against The Calcium-Stabilized Structure of Human Prothrombin. Biochemistry, 22:948-954, 1983. Keyt, B., Furie, B.C., Furie, B. Structural Transitions in Bovine Factor x Associated with Metal Binding and Zymogen Activation. J. Biol. Chem”, 257:8687-8695, 1982. Bloom, J.W., Mann K.G. Metal Ion Induced Conformational Transitions of Prothrombin and Prothrombin Fragment 1. Biochemistry, 17:4430-4438, 1978. Nelsestuen, GJL, Broderius, M., Martin, G. Role of y-Carboxyglutamic Acid. J. Biol. Chem”, 251:6886-6893, 1976. Nelsestuen, GJL Role of y-Carboxyglutamic Acid. {L Biol. Chem., 251:5648-5656, 1976. Kane, W.H., Majerus, P.W. Purification and Characterization of Human Coagulation Factor V. J. Biol. Chem., 256:1002-1007, 1981. Nesheim, M.B., Myrmel, K.H., Hibbard, L., Mann, K.G. Isolation and Characterization of Single Chain Bovine Factor V. J. Biol. Chem., 254:508-517, 1979. Barton, P.G., Hanahan. D.J. The Preparation and Properties of a Stable Factor V from Bovine Plasma. Biochim. Biophys. ACTA, 133:506-518, 1967. Chulkova, In, Hernandez, G. The Preparation and Properties of Factor V from Bovine Plamsa. Clinica Chimica ACTA, 62:21-28, 1975. Suzuke, K., Dahlback, B., Stenflo, J. Thrombin- catalyzed Activation of Human Coagulation Factor V. J. Biol. Chem., 257:6556-6564, 1982. 104 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. The Subunit Structure of Thrombin-activated J. Biol. Chem., 254:964-973, 1979. Esmon, C.T. Factor V. Nesheim, M.B., Mann, K.G. Thrombin-catalyzed Activation of Single Chain Bovine Factor V. J. Chem., 254:1326-1334, 1979. Biol. Nemerson, Y., Furie, B. Zymogens and Cofactors of Blood Coagulation. CRC Critical Reviews of Biochemistry, 45-85, 1980. Rosing, J., Tans, G., Govers-Reimslag, J.W.P., Zwaal, R.F.A., Hemker, H.C. The Role of Phospholipids and Factor Va in the Prothrombinase Complex. J. Biol. Chem., 255:274-283, 1980. Guinto, B.R., Esmon, C.T. Formation of a CAlxium- binding Site on Bovine Activated Factor V Following Recombination of the Isolated Subunits. J. Biol. Chem., 257:10038-10043, 1981. Bloom, J.W., Nesheim, M.B., Mann, K.G. Phospholipid- Binding Properties of Bovine Factor V and Factor Va. Biochemistry, 18:4419-4425, 1979. Miletich, J.P., Jackson, C.M., Majerus, P.W. Properties of the Factor Xa Binding Site on Human Platelets. J. Biol. Chem., 253:6908-6916, 1978. Mitetich, J.P., Majerus, D.W., Majerus, P.W. Patents with Congenital Factor V Deficiency have Decreased Factor Xa Binding Sites on Their Platelets. J. Clin. Invest., 62:824-831, 1978. Subbaiah, P.V., Vajwa, S.S., Smith, C.M., Hanahan, D.J. Interactions of the Components of the Prothrombinase Complex. Biochim. Biophys. ACTA., 444:131-146, 1976. Nelsestuen, G.L., Lim, T.K. Equilibria Involved in Prothrombin and Blood-Clotting Factor x-Membrane Binding. Biochemistry, 16:4164-4171, 1977. Nelsestuen, G.L., Broderius, M. Interaction of Prothrombin and Blood-Clotting Factor X with Membranes of Varying Composition. Biochemistry, 16:4172-4181, 1977. Lim, T.K., Bloomfield, V.A., Nelsestuen, G.L. Structure of the Prothrombin - and Blood Clotting Factor x - Membrane Complexes. Biochemistry, 16:4177- 4181, 1977. Mayer, L.D., Nelsestuen, G.L., Brockman, H.L. Prothrombin Association with Phospholipid Monolayers. Biochemistry, 22:316-321, 1983. 105 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. Kosow, D.P., Orthner, C.L. Kinetics of the Activation of Human Prothrombin by Human Coagulation Factor Xa. Initial Rate Studies in the Presence of Ca... and Phospholipid. J. Biol. Chem., 254:9448-9452, 1979. van Rijn, J., van Dieijen, G., Rosing, J., Bevers, E., Hemker, H.C., Zwall, R.F.A. Activity of Human Platelets in Prothrombin and Factor X Activation Induced by Inonphore A23187. Thromb. Haemost., 50: 29, 1983. DiScipio, R.G., Hermodson, M.A., Yates, S.G., Davie, ELW. A Comparison of Human Prothrombin, Factor IX (Christmas Factor), Factor X (Stuart Factor), and Protein S. Biochemistry, 16:698-706, 1977. Weber, K., Osborn, M. The Reliability of Molecular Weight Determinations by Dodecyl Sulfate-polyacrylamide Gel Electrophoresis. J. Biol. Chem., 244:4406-4412, 1969. Jesty, J., Esnouf, M.P. The Preparation of Activated Factor X and Its Action on Prothrombin. Biochem. J., 131:791-799, 1973. Kress, L.F. Catanese, J.J. Identification of the Cleavage Sites Resulting from Enzymatic Inactivation of Human Antithrimbin III by Crgtaln§_adamanteus proteinase in the Presence and absence of Heparin. Biochemistry, 20:7432-7438, 1981. Kalckar, H.M. Differential Spectrophotometry of Purine Compounds By Means of Specific Enzymes. J. Biol. Chem., 167:461-475, 1947. Broze, G.J., Majerus, P.W. Purification and Properties of Human Coagulation Factor VII. J. Biol. Chem”, 255:1242-1247, 1980. Kisiel, W., Hermodson, M.A., Davie, E.W. Factor X Activating Enzyme from.Russell's Viper Venom: Isolation and Characterization. Biochemistry, 15:4901-4906, 1976. Smith, RJL Tfltration of Activated Bovine Factor X. J. 31010 Chemo] 259:2418’2423' 19730 Gomori, G. A Modification of the Colorimetric Phosphorus Determination for use with the Photoelectric Colorimeter. J. Lab. Clin. Med., 27:955-960, 1942. Laura, R., Robinson, D.J., Bing, D.H. (P-amidinophenyl)- methanesulfonyl flouride, an Irreversible Inhibitor of Serine Proteases. Biochemistry, 19:4859-4864, 1980. 106 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. Vician, L., Tishkoff, G.H. Purification of Human Blood Clotting Factor X by Blue Dextran Agarose Affinity Chromatography. Biochim Biophys ACTA., 434:199-208, 1976. Quick, AuJ. The Hemorrhagic Diseases. Springfield, Ill. Charles C. Thomas, Publ., pp. 312-318, 1942. Kosow, D.P. Kinetic Mechanism of the Activation of Human Plasminogen by Streptokinase. Biochemistry, 14:4459-4464, 1975. Hendrin, H., Lindhout, T., Mertens, K., Engels, W., Hemker, HAL Activation of Human Prothrombin by Stoichiometric Levels of Staphylocoagulase. CL Biol. Chem. 258:3637-3644, 1983. Gonzalez-Gronow, M., Sifring, GJL, Jr., Castellino, FnJ. Mechanism of Activation of Human Plasminogen by the Activator Complex, Streptokinase-Plasmin. J. Biol. Chem., 253:1090-1094, 1978 Kurachi. K., Fugikawa, K., Davie, E.W. Mechanism of Activation of Bovine Factor XI by Factor XII and XIIa. Biochemistry, 19:1330-1338, 1980. Silverberg, M., Kaplan, AMP. Enzymatic Activities of Activated and Zymogen Forms of Human Hageman Factor (Factor XII). Blood, 60:64-70, 1982. Kerr, M.A., Grahn, D.T., Walsh, K.A., Neurath, H. Activation of Bovine Factor X (Stuart Factor)-Analogy with Pancreatic Zymogen-Enzyme Systems. Biochemistry, 17:2645-2648, 1978. Fair, D.S., Bahnak, B.R. Human Hepatoma Cells Secrete Single Chain Factor X, Prothrombin and Antithrombin III. ZBlood, 64:194-204, 1984. Van Dieijen, G., Tans, G., Rosing, J., Hemker, H.C. The Role of Phospholipid and Factor VIIIa in the Activation of Bovine Factor X. J. Biol. Chem”, 256:3433-3442, 1981. Morriosn, £LA., Jesty, J. Tissue-factor-dependent Activation of Tritium Labeled Factor IX and Factor X in Human Plasma. Blood, 63:1338-1347, 1984. Mertens, K., Bertina, R.M. Pathways in the Activation of Human Coagulation Factor X. Biochem. J., 185:647- 658, 1980. Kosow, D.P., Furie, B., Forastieri, H. Activation of Factor X: Kinetic Properties of the Reaction. Thromb. Res., 4:219-227, 1974. 107 171. Briggs, G.E., Haldane, J.,B.,S. A Note on the Kinetics of Enzyme Action. Biochem. J., 19:338-339, 1925. 108 — ' lll‘Jli "la .Jkltllll'ilfli..l{).l‘-' El l.| I.‘ .-|.l"|n- I II - I '1' I l4“ 1. HICHIGQN STA E UNIV IBRARIES ll INN/WM! Ill/ITIIIIHIW HIM/Iii!!!” ”WM 31293105987980