Confirmation of Mendelian Properties of Heterodimeric Fibrinogen Molecules in a Heterozygotic Dysfibrinogenemia, “Fibrinogen Amarillo,” Using GPRphoresis to Differentiate SemiFibrin Molecules from Fibrinogen and Fibrin

Thrombosis Research 101 (2001) 91 ± 99

REGULAR ARTICLE

Confirmation of Mendelian Properties of Heterodimeric Fibrinogen Molecules in a Heterozygotic Dysfibrinogenemia, ``Fibrinogen Amarillo,'' Using GPRphoresis to Differentiate SemiFibrin Molecules from Fibrinogen and Fibrin

John R. Shainoff1, Oscar D. Ratnoff2, Gary B. Smejkal1, Patricia M. DiBello1, William R. Welches3, Helmut Lill4, Olga V. Mitkevich1,5 and Phillip Periman6 1 Department of Chemistry Cleveland State University, 2351 Euclid Avenue, Cleveland, OH 44115-2406, USA 2 Hematology/Oncology Division, Department of Medicine, Case Western Reserve University, University Hospitals of Cleveland, Cleveland, OH 44106, USA 3Meridia South Pointe Hospital, Cleveland, OH 44122, USA 4 R&D Coagulation, Roche Diagnostics, D-82377 Penzberg, Germany 5Russian Cardiology Research Center, 3rd Cherepkoskaya Street, Moscow 121552, Russia 6Harrington Cancer Center, Amarillo, TX 79106, USA (Received 11 September 2000 by Editor Mosesson; accepted 11 October 2000)

Abstract The fibrinogen molecule consists of two sets of Aa, Bb, and g chains assembled into a bilateral disulfide linked (Aa, Bb, g)2 structure. Cleavage of the two A-fibrinopeptides (FPA, Aa1±16) from normal Aa chains with arginine at position 16 (RFPA) by thrombin or the venom enzyme atroxin transforms fibrinogen into self-aggregating fibrin monomers (a, Bb, g) 2 . Mutant Aa16R ! H fibrinopeptide (HFPA) cannot be cleaved from fibrinogen by atroxin. Many studies on heterozygous dysfibrinogenemias with this mutation suggested that incorporation of the mutant chains into the molecules was ordered in a manner yielding only (1) homodimeric normal (RFPARFPA) atroxin-coagulable molecules and (2) homodimeric abnormal (HFPAHFPA) atroxin-inAbbreviations: RFPA, normal fibrinopeptide A Aa1 ± 16R cleavable by atroxin; HFPA, mutant Aa16R!H fibrinopeptide not cleavable by atroxin. Corresponding author: John R. Shainoff, Ph. D., Department of Chemistry, Cleveland State University, 2351 Euclid Avenue, Cleveland, OH 44115-2406, USA. Tel: +1 (216) 687 2463; Fax: +1 (216) 687 9298; E-mail: .

coagulable molecules in equal quantities. Although heterodimeric molecules (RFPAHFPA) could not be found in studies on the intact protein, Meh et al. demonstrated their existence by showing that CNBr digests of fibrinogens from atroxin-treated Aa16R ! H heterozygotic dysfibrinogenemias consistently yielded N-terminal fragments (NDSKs) with partially resolved electrophoretic bands predominantly in between the NDSKs of fibrinogen and a-fibrin. An opportunity to confirm and better quantify the heterodimers arose with the recent development of a method (GPRphoresis) for identifying molecules lacking only one FPA, which is applied here in study of a newly presenting case of an Aa16R ! H dysfibrinogenemia, ``fibrinogen Amarillo.'' GPRphoresis uses electrophoretic shifts, staged with GPRP-NH2 to separate the self-aggregating fibrin monomers lacking both FPAs from weakly aggregating ``semifibrin'' molecules lacking one FPA An antifibrin a17± 23 antibody is used to measure and differentiate the semifibrin from fibrinogen with FPA fully intact. Applying GPRphoresis to atroxin digests of fibrinogen Amarillo clearly demonstrated

0049-3848/01/$ ± see front matter D 2001 Elsevier Science Ltd. All rights reserved. PII S0049-3848(00)00383-2

92

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

RFPARFPA, RFPAHFPA, and HFPAHFPA molecules in nearly perfect Mendelian 1:2:1 proportions. In turn, the high levels of the semifibrin in the terminal atroxin digests provide genetic phenotypic evidence supporting fidelity of the GPRphoresis method. D 2001 Elsevier Science Ltd. All rights reserved. Key Words: Fibrinogen; Dysfibrinogenemia; GPRphoresis; Semifibrin; Atroxin

T

he fibrinogen macromolecule consists of two sets of disulfide linked Aa, Bb, and g chains arranged bilaterally, and is converted to self-aggregating fibrin monomer upon cleavage of fibrinopeptide A (FPA) from the Aa chains [1±3] with resultant exposure of aggregation sites with the critical sequence GPR located penultimate to the fibrinopeptide [4]. The assembly of the polypeptide chains into the molecule is highly ordered, beginning with incorporation of g and Aa chains into Aag dyads, followed by rate-limiting formation of AaBbg triads, followed by pairing of the triads into the six-chain molecule [5,6]. The Aa chains appear to be linked to each other in a parallel orientation [7], but the g chains appear antiparallel [8]. Many studies on dysfibrinogenemias with Aa chain mutations have suggested that assembly of Aa chains into the macromolecule is either compartmentalized or governed by a segregating factor producing only homozygotic normal and abnormal molecules in approximately 1:1 proportions, since heterozygotic molecules were not found [7,9±16]. These inferences were based largely on measurements of self-aggregating vs. nonaggregating derivatives in solution-phase when the fibrinogen from heterozygotic propositi with either Aa16R ! H or Aa16R ! C substitutions were subjected to a thrombin-like venom enzyme (EC 3-4-21, atroxin from Bothrops atrox) that readily releases normal fibrinopeptide RFPA from fibrinogen but does not release the mutated peptide. On the other hand, solution-phase quantitations by Siebenlist et al. [17] raised a prospect that the heterodimers exist. Although the existence of heterodimers could not be demonstrated in studies on intact molecules, studies on CNBr digests by Meh et al. [18] provided convincing

evidence of their existence. N-terminal fragments (the NDSKs) of atroxin-treated specimens of fibrinogens with Aa16 substitutions, both Aa16R ! H and Aa16R ! C, were shown by SDS-PAGE to migrate in three partially resolved bands with the principal component in between the NDSKs of fibrinogen and afibrin, suggestive of the existence of heterodimeric and homodimeric molecules in a Mendelian 1:2:1::RFPARFPA:RFPAHFPA:HFPAHFPA pattern. A new incidence (``fibrinogen Amarillo'') of the fairly common heterozygotic dysfibrinogenemia partially expressing the Aa#16R ! H mutation was identified in our laboratory and the ``GPRphoresis'' method that we used to identify the fibrin intermediate lacking one of the two FPA that are released by thrombin provided a new tool to examine the production of semifibrin molecules among products of atroxin reactions with the mutant fibrinogen. The characterizations of this dysfibrinogenemia, as presented here confirm the existence of high levels of semifibrin in this type of dysfibrinogenemia, and demonstrate more clearly that the proportions of homo- and heterodimers conform with a Mendelian pattern of expression.

1. Materials and Methods Commercial sources of chemicals and biologics were: Peptide Synthesiz (3rd Cherepkoskaya Street, Moscow 121552, Russia; GlyProArgProAmide, GPRP-NH 2 ), Global Imports (Tampa, FL; Hispanagar D1 agarose), XC Corp hw w w . x c b e a d s . c o m i ( g l y o x y l a g a r o s e ) , Research Organics hwww.resorg.comi (buffer components), Sigma hwww.Sigma-Aldrich.comi (atroxin, Lubrol, Tween, MW-SDS-200 molecular weight markers), United States Biochemical hwww.USbiochem.comi (hemin, thimerosal), R&D Coagulation, Roche Diagnostics (D-82377 Penzberg, Germany; antifibrin a17±23 monoclonal antibody; mAb). Otherwise, chemicals were specified as reagent grade or better. 1.1. Protein and Antibody Preparations Fibrinogen was prepared from citrated human plasma by sequential precipitations with ethanol,

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

glycine, and ammonium sulfate/e-aminocaproic acid, as described [19]. Human thrombin was the commercially supplied Fibrindex (Ortho Diagnostics, Raritan, NJ). The anti-FPA mAb was affinity purified using fibrinogen linked to glyoxyl agarose as described previously [20,21]. The anti-FPA binds both the free peptide and the intact segment at the N-terminus of the Aa chain of fibrinogen (Aa1±16). This mAb binds equivalently to fibrinogen containing two FPA per molecule and aprofibrin with one FPA per molecule, but not to fibrin containing no FPA [22]. The anti-FPA also binds equivalently with both normal RFPA and mutant HFPA. The antifibrin a17 ± 23 mAb (a generous gift from Boehringer Mannheim, now Roche Diagnostics) was affinity purified as described [23], and binds equivalently to fibrin lacking both FPA and to a-profibrin lacking only one FPA, but does not bind to fibrinogen with both FPA intact [22]. The binding of antifibrin a17 ± 23 mAb to fibrin and semifibrin occurs only after denaturation of the proteins [23], effected in electropherograms by fixing the protein with 5% trichloroacetic acid (15 min) followed by extensive washing with PBS±Tween. 1.2. Labeling of Antibodies and Normal Fibrinogen Radioiodinated (125I) fibrinogen, labeled by the iodine monochloride method of McFarlane [24], was used for studies with normal fibrinogen to enable us to precisely compare results on atroxin reactions with our preceding studies on thrombin reactions [22]. It also enabled us to visualize and quantify overall protein distributions on the PhosphorImager (Molecular Dynamics, Sunnyvale, CA) after drying immunostained gels. Amarillo fibrinogen was not radiolabeled, and separate electrophoretic gels were prepared for immunostaining and for general protein staining. The antifibrin a17±23 and anti-FPA antibodies were labeled with glutaraldehyde-treated horseradish peroxidase using a modification of the method of Boorsma [25], as described [26]. 1.3. Kinetic Studies Fibrinogen (3 mg/ml) was incubated with either thrombin (0.5 U/ml) or atroxin (0.3 mg/ml) for

93

the indicated times in Tris-buffered saline ± EDTA (TBSE: 7 mM Tris±HCl, 0.143 M NaCl, 0.2 mM EDTA) containing GPRP-NH2 (2 mM). The reactions were stopped by adding either phenylpropylalanine chloromethylketone (20 mM PPACK) for thrombin or p-nitrophenyl p0guanidino-benzoate (0.18 mM NPGB) for atroxin [27]. The reaction mixtures (0.5 ml samples) were analyzed for fibrinogen derivatives by GPRphoresis [22], commenced within an hour for specimens treated with NPGB. Fibrinopeptide release was measured by HPLC [28,29]. The use of NPGB interfered with HPLC determinations of the fibrinopeptides, and separate samples were prepared for those analyses. To deproteinize samples used for HPLC determinations of FPA, 0.1 ml portions of reaction mixtures were acidified with 4 ml 1 N HCl and transferred immediately to a boiling water bath for 3±5 min, then made alkaline with 6 ml 1 N NaOH, and subsequently ultrafiltered centrifugally with Ultrafree 10,000 NMWL filter units (Millipore, Bedford, MA). 1.4. GPRphoresis Gel-shifts induced with GPRP-NH2 admixed in the resolving gel were used to distinguish selfaggregating from weak and nonaggregating components in thrombin/fibrinogen and atroxin/ fibrinogen reaction mixtures. The equipment and procedure used were essentially as described [22]. The GPRP-NH2 in the buffer migrates cathodally leaving behind buffer free of it. On traversing the boundary between GPRP-NH2-rich and -poor buffer, fibrin molecules form clots in a sharply demarcated zone, while semifibrin and fibrinogen molecules continue migrating together, unaffected by the absence of GPRPNH2. The presence of semifibrin in the ``fibrinogen'' band was assessed by immunoprobing with antifibrin a17±23 antibody. Colloidal Coomassie violet R-250 [30] was used for general protein staining. Immunostaining was performed directly in the gel after (1) an overnight exposure of the gel to the peroxidase labeled antibody, followed by (2) removal of excess antibody with 1.4 vol. of wash-buffer suctioned through the gel on a gel dryer, and subsequently (3) localizing retained antibody

94

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

with diaminobenzidine and hydrogen peroxide as described [22,26].

2. Results 2.1. Fibrinogen Amarillo This dysfibrinogenemia was found in three siblings and their father who presented with a coagulation disorder typical of a heterozygotic Aa16R ! H mutation, exhibiting a delayed but full coagulability with thrombin, but only 50% coagulability with atroxin. The Aa16R ! H mutation was confirmed by HPLC of full thrombin digests of the fibrinogen that showed an abnormal FPA (Fig. 1) containing histidine (HFPA) instead of arginine (RFPA). The abnormal component gave a positive Pauli reaction for histidine and a negative Sakaguchi reaction for arginine. The mutation was also confirmed by amino acid sequencing. The abnormal HFPA was released in amounts essentially equal to the normal RFPA on extended incubations with thrombin (Fig. 1, lower panel), but was not released in reactions of the fibrinogen with atroxin that was known to release normal RFPA but not mutant HFPA Also, its release by thrombin was slow compared to the normal component of the total FPA, as anticipated [31]. 2.2. GPRphoresis As applied previously in studies on kinetics of the fibrinogen/fibrin conversion [22,32], GPRphoresis provides an accurate means for distinguishing fibrin monomer lacking two FPAs from intermediary/semifibrin molecules lacking only one FPA. The fibrin monomers comigrate with fibrinogen anodally until they reach the boundary between buffer that is rich in cathodally migrating GPRP-NH2 and trailing buffer lacking the peptide where the monomers precipitate in a sharp band as they traverse the boundary. Normal fibrinogen was converted fully to fibrin monomer by atroxin (Fig. 2, normal). Fibrinogen Amarillo underwent full, albeit slow, conversion to fibrin with thrombin, as indicated earlier by the fibrinopeptide analyses, and by GPRphoresis and clot formation (not shown), as anticipated for this type of dysfibrinogenemia [7,9 ± 18]. However,

Fig. 1. HPLC comparing peptides released in full thrombin-digests of normal (upper tracing) and Amarillo fibrinogen (lower tracing). The labeled peaks were verified by amino acid analyses and sequencing. The labeled components of normal FPA consisted of native phosphorylated FPA (AP), predominant dephosphorylated FPA (A), and minor N-terminal degraded FPA (AY). The resolvable mutant FPA counterparts, comprising half of the total in fibrinogen Amarillo are labeled AH and APH. Intact fibrinopeptide B is labeled B and derivatives arising from carboxypeptidase cleavage of R14 are labeled Bx and B0x, where B0 is an N-terminal degraded component produced in storage [28,29].

atroxin elicited only a 25% conversion of fibrinogen Amarillo to fibrin monomer in greatly prolonged reactions (Fig. 2, Amarillo). This 25% yield of fibrin that was indicated by GPRphoresis contrasted with estimates based on clot formation with atroxin that had suggested fibrin yields on the order of 50%, from heterozygotic Aa16R ! H dysfibrinogenemias, observed both currently in our laboratory and in previous studies by others [7,9±16], based on mass measurements of solution- and clot-phase products of end-stage digests with atroxin and thrombin. 2.3. Immunoprobing Electropherograms of atroxin/fibrinogen reaction mixtures were probed directly (in gel) with

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

95

formed with a Mendelian distribution (1 R FPA R FPA :2 R FPA H FPA :1 H FPA H FPA , where HFPAHFPA would not contribute to the immunostaining). The transient production of this semifibrin in the course of full conversion of fibrinogen Amarillo to fibrin by thrombin instead of atroxin reinforces this identification.

Fig. 2. General protein distributions in electropherograms showing the course of atroxin reactions with normal fibrinogen and fibrinogen Amarillo yielding the distinct fibrin monomer band ( f ) and the mixed fibrinogen (f)/semifibrin ( p) band. Fibrin monomer formation from fibrinogen Amarillo reached only onefourth of the essentially full conversion to monomer obtained with normal fibrinogen at 400 s, and extending reactions 20 times longer did not yield more product.

(1) anti-FPA mAb to assess the distribution of molecules with intact FPA and with (2) antia17 ± 23 to reveal components with GPR domains exposed by loss of FPA. With normal fibrinogen (Fig. 3), there was a transient appearance of product corresponding to the fibrin intermediate that stained with antifibrin a17± 23 and comigrated with fibrinogen. The intermediate rose maximally to 40% of the total protein in intermediary stages of reaction, but in the end-stages all but a trace 1.0% of the protein was converted to fibrin monomer (evident also from the general staining shown in Fig. 2). These results with atroxin were similar to studies on thrombin reactions [22], with the exception that the 40% maximal level of profibrin observed in the atroxin reactions was greater than the 20% level observed with thrombin. The 40% level is close to the theoretic 100*1/e = 37% that would be obtained upon independent release of the first and second FPA with equal rate constants [22]. With fibrinogen Amarillo, however, semifibrin molecules remained high after greatly prolonged incubation with atroxin. The intensity of immunostaining with antifibrin a17 ± 23 indicated that semifibrin comprised two-thirds of the protein stripped of normal FPA (Fig. 3, center panel). This 2:1 proportion of semifibrin/fibrin with atroxin-cleavable RFPA con-

Fig. 3. Immunostained electropherograms of atroxin reactions with normal and Amarillo fibrinogens probed with peroxidase-labeled antifibrin a17 ± 23 mAb and anti-FPA mAb. These antibodies bind equally to protein carrying either one or two of the respective epitopes. Protein staining with antifibrin a17 ± 23 corresponds to fibrin ( f ) and semifibrin ( p). Both the normal and Amarillo fibrinogen preparations contained a small amount of semifibrin prior to reaction, indicated in the 0-time sample not treated with atroxin. With normal fibrinogen, profibrin levels, revealed with antifibrin a17 ± 23 mAb (middle panel) rose transiently in intermediary stages of reaction and then fell to a trace level due to the virtually complete conversion to fibrin, as anticipated. With fibrinogen Amarillo, profibrin levels remained high because atroxin could not cleave mutant peptide retained by semifibrin derived from heterozygotic fibrinogen. The anti-FPA staining of the a/p band changed only slightly, because only one-fourth of the fibrinogen Amarillo, that with homozygous normal chains, was stripped of FPA and converted to fibrin, as shown in Fig. 2. There was a faint staining in the position of the fibrin band ( f ) by anti-FPA, which we attributed to some copolymerization of profibrin with the fibrin. Trace artifactual staining occurred at the origin (o) and at a junction between 1% and 3% agarose that was constructed for band sharpening, located just behind the f band.

96

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

The immunostaining with anti-FPA showed all of the FPA-containing protein migrating in the fibrinogen/semifibrin band from beginning to end of the reaction, except for traces that appeared in the fibrin band, due possibly to slight copolymerization of the semifibrin with the fibrin. The density of staining decreased only about 30%, and this was attributable to conversion of 25% of the protein to fibrin as shown earlier in Fig. 2 (Amarillo).

3. Discussion The results of this study corroborate more clearly the conclusions of Meh et al.[18] that heterodimers (RFPAH FPA) predominate over homodimeric forms of fibrinogen (either RFPARFPA or HFPAHFPA) in a heterozygous Aa16R ! H dysfibrinogenemia, ``fibrinogen Amarillo,'' in contrast to many studies [7,9± 16] of heterozygous Aa16R ! H dysfibrinogenemias that had indicated the converse. Further, the proportions of the heterodimers found in the analyses corresponded to levels expected from simple distributive (Mendelian) gene products for a heterozygote. The homozygous normal component that forms fibrin monomers with atroxin comprised only one-fourth of the total protein, while the heterozygotic gene product comprised one-half of the total, twothirds of the products reacting with antifibrin a17±23 mAb. The mistaken earlier inferences that heterodimers do not exist in this dysfibrinogenemia because semifibrin molecules could not be found resemble other mistakes, spanning over 40 years of searching [33] beginning with our own [34]. Presumably, this is largely because the semifibrin molecules are amphoteric, resembling fibrin in presence of a predominance of fibrin molecules but resembling fibrinogen in presence of a predominance of fibrinogen. The immunochemical quantitations of semifibrin comigrating with fibrinogen in these studies leave little doubt of their predominance in gene products in this dysfibrinogenemia. If the fibrinogen Amarillo contained only homodimeric normal and mutant molecules with no heterodimers, the homodimeric

normal molecules would transiently produce some semifibrin that would ultimately disappear due to their conversion to fibrin, and all of the immunostaining with antifibrin a17±23 would be localized in the fibrin band after the extended incubations as observed with normal fibrinogen. In our opinion, GPRphoresis coupled with immunoprobing with antifibrin a17±23 is the only certain way presently available to distinguish clearly between fibrin and semifibrin/ profibrin molecules. Further, the clear identification of high levels of semifibrin and low levels of fibrin in end-stage atroxin-digests of this heterozygotic dysfibrinogenemia adds genetic support for the method that we had used in preceding studies identifying a-profibrin as the first intermediate in the conversion of fibrinogen to fibrin [22]. The partial resolution of NDSKs from fibrinogen, semifibrin, and fibrin also enabled Meh et al. [35] to demonstrate the production of semifibrin molecules as intermediates in the fibrinogen/fibrin conversion; however, our analyses indicate that they may have overestimated the levels of the intermediate by a factor of two [22]. The lower levels found by us was interpreted as an indication that the rate constant for release of the second FPA, converting the intermediate to fibrin, is four times faster than the release of the first FPA, converting fibrinogen to the intermediate. An even faster ( > 10  ) relative rate for release of the second FPA, due possibly to some reseating of thrombin, was recently observed during fibrin formation from rabbit fibrinogen using either bovine or human thrombin [36]. Unlike the presence of Aa chain heterodimers in the dysfibrinogenemias, heterodimers were shown to be clearly absent in characterizations of the plasma phenotypes and biosynthetic intermediates of two variant forms of normal human fibrinogen with differing C-terminal domains, a principal component with molecular weight 340,000 arising from bypassed transcription of exon 6 and a minor 420,000-kDa component arising from transcription of exon 6 resulting in a C-terminal extension of the a chain [37±39]. The basis for that segregation is not known.

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

This work was supported in part by grants HL-60896 (JRS) and HL-01661 (ODR) from the National Heart, Lung, and Blood Institute, National Institutes of Health. A portion of the study was carried out at The Cleveland Clinic Foundation under grant HL16361 from the National Heart, Lung, and Blood Institute.

References 1. Lorand L. `Fibrino-peptide': new aspects of the fibrinogen±fibrin transformation. Nature 1951;167:992. 2. Bailey K, Bettelheim FR, Lorand L, Middlebrook WR. Action of thrombin in the clotting of fibrinogen. Nature 1951;167:233±5. 3. Kudryk BJ, Collen D, Woods KR, BlombaÈck B. Evidence for localization of polymerization sites in fibrinogen. J Biol Chem 1974;249: 3322±5. 4. Laudano AP, Doolittle RF. Synthetic peptide derivatives that bind to fibrinogen and prevent the polymerization of fibrin monomers. Proc Natl Acad Sci USA 1978; 75:3085±9. 5. Roy SN, Procyk R, Kudryk BJ, Redman CM. Assembly and secretion of recombinant human fibrinogen. J Biol Chem 1991;266: 4758±63. 6. Roy SN, Sun A, Redman CM. In vitro assembly of the component chains of fibrinogen requires reticulum factors. J Biol Chem 1996;271:24544±50. 7. Miyata T, Terukina S, Matsuda M, Kasamatsu A, Takeda Y, Murakami T, Iwanaga S. Fibrinogen Kawaguchi and Osaka: an amino acid substitution of Aa arginine-16 to cysteine which forms an extra interchain disulfide bridge between the two Aa chains. J Biochem (Tokyo) 1987;102:93±101. 8. Chen R, Doolittle RF. Identification of the polypeptide chains involved in the crosslinking of fibrin. Proc Natl Acad Sci USA 1969;63:420±7. 9. Galanakis DK, Henschen A, Keeling M, Kehl M, Dismore R, Peerschke EI. Fibrinogen Louisville: an Aa16 Arg ! His defect that forms no hybrid molecules in heterozygous individuals and inhibits aggregation

10.

11.

12.

13.

14.

15.

16.

17.

18.

97

of normal fibrin monomers. Ann NY Acad Sci 1983;408:644±8. Galanakis DK, Hultin M. Fibrinogen Stony Brook II: partial characterization of a heterozygously transmitted peptide A anomaly. Blood Coagulation Fibrinolysis 1990;1:567±70. Galanakis DK, Henschen A, Peerschke EI, Kehl M. Fibrinogen Stony Brook, a heterozygous A alpha 16Arg±Cys dysfibrinogenemia. Evaluation of diminished platelet aggregation support and of enhanced inhibition of fibrin assembly. J Clin Invest 1989;84:295±304. Kaudewitz H, Henschen A, Peerschke EIB, Kehl M. The molecular defect of the genetically abnormal fibrinogen Christchurch II. In: MuÈller-Berghaus G, Sheefers-Borchel U, Selmayr E, Henschen A, editors. Fibrinogen and its derivatives. Amsterdam: Elsevier, 1986. pp. 31±36. Pirkle H, Kaudewitz H, Henschen A, Theodor I, Simmons G. Substitution of Bb 14 arginine by cyst(e)ine in fibrinogen Seattle I. In: Lowe GDO, Douglas JT, Forbes CD, Henschen A, editors. Fibrinogen: biochemistry, physiology and clinical relevance. Amsterdam: Elsevier, 1987. pp. 49±52. Lee MH, Kaczmarek E, Chin DT, Oda A, McIntosh S, Bauer KA, Clyne LP, McDonagh J. Fibrinogen Ledyard (A alpha Arg16±Cys): biochemical and physiologic characterization. Blood 1991;78:1744±52. Matsuda M, Terukina S, Miyata T, Iwanaga S, Kasamatsu A, Takeda Y. (Substitution of A alpha arginine-16 by cysteine in two abnormal fibrinogens (Kawaguchi and Osaka I)), (in Japanese)Rinsho Byori (Jpn J Clin Pathol), 1987;73:161±9. Collect JP, Woodhead JL, Soria J, Soria C, Mirshahi M, Caen JP, Weisel JW. Fibrinogen Dusart: electron microscopy of molecules, fibers and clots, and viscoelastic properties of clots. Biophys J 1996;70:500±10. Siebenlist KR, Prchal JT, Mosesson MW. Fibrinogen Birmingham: a heterozygous dysfibrinogenemia (Aa 16 Arg ! His) containing heterodimeric molecules. Blood 1988;71:613±8. Meh DA, Siebenlist KR, Galanakis DK, Berg-

98

19. 20.

21.

22.

23.

24. 25.

26.

27.

28.

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

trom G, Mosesson MW. The dimeric Aa chain composition of dysfibrinogenemic molecules with mutations at Aa 16. Thromb Res 1995;78:531±9. Kanaide H, Shainoff JR. Cross-linking of fibrinogen and fibrin by fibrin-stabilizing factor. J Lab Clin Med 1975;85:574±97. Shainoff JR, Valenzuela R, Gonda SR, Caulfield M. Blot-screening of culture fluid immunoreactivities on glyoxyl agarose films. BioTechniques 1986;4:120±8. Valenzuela R, Shainoff JR, DiBello PM, Urbanic DA, Anderson JM, Matsueda GR, Kudryk BJ. Immunoelectrophoretic and immunohistochemical characterizations of fibrinogen derivatives in atherosclerotic aortic intimas and vascular prosthesis pseudo-intimas. Am J Pathol 1992;141:861±80. Shainoff JR, Smejkal GB, DiBello PM, Mitkevich OV, Levy PJ, Lill H, Dempfle CE. Isolation and characterization of the fibrinintermediate arising from cleavage of one fibrinopeptide A from fibrinogen. J Biol Chem 1996;271:24129±37. Dempfle CE, Dollman M, Lill H, Puzzovio D, Dessauer A, Heene DL. Binding of a new monoclonal antibody against N-terminal heptapeptide of fibrin alpha-chain to fibrin polymerization site ``a'': effects of fibrinogen and fibrinogen derivatives, and pretreatment of samples with NaSCN. Blood Coagulation Fibrinolysis 1993;4:79±86. McFarlane AS. In vivo behavior of I131-fibrinogen. J Clin Invest 1963;42:346±61. Boorsma DM. Preparation of HRP-labelled antibodies. In: Cuello AC, editor. Immunohistochemistry. New York: Wiley, 1983. pp. 87±100. Shainoff JR. Electrophoresis and direct immunoprobing on glyoxyl agarose and polyacrylamide composites. Adv Electrophor 1993;6:61±177. Janmey PA, Erdile L, Bale MD, Ferry JD. Kinetics of fibrin oligomer formation observed by electron microscopy. Biochemistry 1983;22: 4336±40. Ebert RF, Bell WR. Assay of human fibrinopeptides by high-performance liquid

29.

30.

31.

32.

33.

34. 35.

36.

37.

chromatography. Anal Biochem 1985;148: 70±8. Kehl M, Lottspiech F, Henschen A. Highperformance liquid chromatography of proteins as applied to fibrinogen chains. Hoppe-Seyler's Z Physiol Chem 1982;363: 1501±5. Patestos NP, Fauth M, Radola BJ. Fast and sensitive protein staining with colloidal acid violet 17 following isoelectric focusing in carrier ampholyte generated and immobilized pH gradients. Electrophoresis 1988;9: 488±96. Higgins DL, Shafer JA. Fibrinogen Petoskey, a dysfibrinogenemia characterized by replacement of arg-a alpha 16 by a histidyl residue, evidence for thrombin-catalyzed hydrolysis at a histidyl residue. J Biol Chem 1981;256:12013±7. Shainoff JR, Urbanic DA, DiBello PM, Valenzuela V. GPR-phoresis, a novel approach to determining fibrin monomer and other macromolecular derivatives of fibrinogen and fibrin in blood. Blood Coagulation Fibrinolysis 1993;4:87±92. Li X, Galanakis D, Gabriel DA. Transient intermediates in the thrombin activation of fibrinogen. Evidence for only the DesAA species. J Biol Chem 1996;271:11767±71. Shainoff JR, Page IH. Significance of cryoprofibrin in fibrinogen±fibrin conversion. J Exp Med 1962;116:687±707. Meh DA, Siebenlist KR, Bergtom G, Mosesson MW. Sequence of release of fibrinopeptide A from fibrinogen molecules by thrombin or atroxin. J Lab Clin Med 1995;125: 384±91. Shainoff JR, Smejkal GB, DiBello PM, Chase B, Mitkevich OV, Lill H. The fibrin intermediate, its place in the fibrinogen ± fibrin transformation. In: Mosesson MW, Nieuwenhuizen W, editors. New York: New York Academy of Sciences, 2000 (in press). Fu Y, Zhang JZ, Redman CM, Grieninger G. Formation of the human fibrinogen subclass fib420: disulfide bonds and glycosylation in its unique (alphaE chain) domains. Blood 1998;92:3302±8.

J.R. Shainoff et al./Thrombosis Research 101 (2001) 91±99

38. Applegate D, Steben LS, Hertzberg KM, Grieninger G. The alpha(E)C domain of human fibrinogen-420 is a stable and early plasmin cleavage product. Blood 2000;95: 2297±303.

99

39. Fu Y, Grieninger G. Fib420: a normal human variant of fibrinogen with two extended a chains. Proc Natl Acad Sci USA 1994;91: 2625±8.