Comparison of the physicochemical properties of fragment D derivatives of fibrinogen and fragment D-D of cross-linked fibrin

Biochimica et Biophysica Acta, 427 (1976) 1-14

© Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands BBA 37278 C O M P A R I S O N OF T H E P H Y S I C O C H E M I C A L PROPERTIES OF F R A G M E N T D DERIVATIVES OF F I B R I N O G E N A N D F R A G M E N T D-D OF CROSS-LINKED FIBRIN

VICTOR J. MARDER*, ANDREI Z. BUDZYNSKI and GRANT H. BARLOW Department of Medicine and the Specialized Center of Research in Thrombosis, Temple University Health Sciences Center, Philadelphia, Pa. and Abbott Laboratories, Scientific Division, North Chicago, IlL (U.S.A.)

(Received June 23rd, 1975)

SUMMARY The molecular weight of Fragment D derivatives obtained from plasmic digests of fibrinogen and cross-linked fibrin was determined by equilibrium sedimentation and compared with the summated molecular weight of their polypeptide chains observed after electrophoresis of reduced protein in sodium dodecyl sulfate polyacrylamide gels. The measured molecular weight of Fragment D (Stage 2) of fibrinogen is 103 500, which is compatible with a molecule containing only one each of the Aa (13 000), Bfl (43 000) and 7 (39 000) chain remnants. Fragment D-D of crosslinked fibrin has a molecular weight of 189 000, compatible with a molecule containing one isopeptide-bound y-y chain (80 000) and two each of Bfl (43 000) and Aa (13 000) chain remnants. The NH2-terminal amino acid residues of the Fragment D derivatives were measured quantitatively using a thioacetic-thioglycolic acid method, and molar quantities were calculated on the basis of the molecular weights determined by equilibrium sedimentation. Fragment D preparations obtained from Stage 2 and Stage 3 digests of fibrinogen have 3 mol of NHz-terminal amino acids per molecule, while Fragment D-D has seven. These data support the view that two Fragment D molecules, each of three polypeptide chains, are derived by plasmic degradation from each fibrinogen molecule, and that an isopeptide-bound, six chain Fragment D-D molecule is released from cross-linked fibrin by plasmin. Equilibrium sedimentation measurement of the molecular weights of Fragment X (Stage 1 and Stage 2) and Fragment Y are 265 000 and 148 000, respectively. These values are compatible with asymmetric cleavages of Fragment X to Fragments Y and D (Stage 2), and of Fragment Y to Fragments D (Stage 2) and E, and with a fibrinogen model in which the two halves are joined by disulfide bonds only in the amino-terminal regions.

* A preliminary report of this work was presented in April, 1975, at the 59th meeting of the Federation of American Societies for Experimental Biology, Atlantic City, N.J., U.S.A. * Requests for reprints should be addressed to Victor J. Marder, M. D., Temple University School of Medicine, 3400 North Broad Street, Philadelphia, Pa. 19140, U.S.A.

INTRODUCTION ~[he reconstruction of parent molecules from enzymatic or chemical cleavage products has been a useful tool in defining the structure of large proteins, such as myosin and immunoglobulin. Similarly, the physicochemical and immunologic characteristics of the plasmic degradation products and reduced polypeptide chains of fibrinogen have been applied to the structure of the parent molecule*. A molecular model consisting of three pairs of polypeptide chains (Aa, Bfl and ~,) held together by disulfide bonds has evolved, but many of the details regarding the organization and tertiary structure of the molecule are still controversial or speculative. One model considers that disulfide bonds link the amino-terminal parts of each half of the molecole together [3-7], and that asymmetric plasmic degradation [2, 8, 9] leads to the formation of one Fragment E derivative containing these disulfide links [10-12] and two Fragment D derivatives, one from each of the two carboxy-terminal portions. An alternative model considers the halves of fibrinogen to be linked by disulfide bonds in two regions, at the amino terminus and roughly midway between the amino and carboxy termini [13-15]. In this latter scheme, plasmin would liberate a single dimeric Fragment D molecule, that is, one which contains the carboxy-terminal regions of both halves of fibfinogen. These two models imply different physicochemical characteristics for the degradation products, most clearly reflected by the number of polypeptide chains postulated for the Fragment D derivatives of fibrinogen. The asymmetric scheme predicts Fragment D molecules with three chains; the dimeric scheme, Fragment D molecules with six chains. Since Factor XIll cross-links the carboxy-telminal ends of the y-chains of two diffelent fibrin molecules [16], plasmic degradation of such cross-linked fibrin produces a "D-dimer" [17, 18] or "doubleD" [19] molecule, which consists of two Fragment D molecules joined by Factor XllI-induced isopeptide bonds. These unique derivatives, here called Fragment D-D, are approximately twice the size of Fragment D and should contain the equivalence of twice the numbei of polypeptide chains in Fl agment D. Accordingly, the products of fibrinogen (Fragments X, Y, D and E) and cross-linked fibxin (Fragment D-D) have been isolated and their molecular weights determined by equilibrium sedimentation. The values for the Fragment D and D-D derivatives have been utilized in calculations of quantitative NHz-terminal amino acid residues and for comparison with molecular weight values derived from sodium dodecyl sulfate-polyacrylamide gel electrophoresis of reduced polypeptide chains of these fragments. MATERIALS AND METHODS

Preparation of fibrinogen degradation products. Human fibrinogen (Grade L, A.B. Kabi, Stockholm) dissolved in water is 9 5 ~ clottable after the addition of thrombin (bovine, Parke-Davis, Detroit, Michigan) and contains enough plasminogen * According to nomenclature agreed upon by the International Society on Thrombosis and Hemostasis, the chains of fibrinogen are named Aa, Bfl, and ~; those of fibrin, a, fl and ~ [1]. Th ~. stages of plasmic degradation of fibrinogen are defined according to Marder and colleagues [2]. A Stage 1 digest is characterized by the predominance of Fragment X; Stage 2 contains Fragments X, Y, D and E; and Stage 3 contains mainly Fragments D and E. Fragments X and D isolated from different stages of degradation are named accordingly, e.g. Fragment X (Stage 2), Fragment D (Stage 3), etc.

to cause extensive proteolydc degradation after the addition of streptokinase (Varidase ®, Lederle Laboratories, Pearl River, N.Y.). Technical details of proteolytic degradation products are described elsewhere [8] and depending upon the types of degradation products present, one can determine the degree of degradation. Degradation is terminated by the addition of soybean trypsin inhibitor (Worthington, Freehold, N.J.) or e-aminocaproic acid (Sigma, St. Louis, Mo.)[8]. Thrombin clottability of Fragment X obtained from Stage 1 digests is 30-80 ~ and that obtained from Stage 2 digests is 5-60 ~o. A mixture of Fragments D and E is separated from Stage 2 and Stage 3 digests by gel filtration on Sephadex G-200 gel columns, then subjected to block electrophoresis on Pevikon (C-870, Mercer, New York) to obtain purified fragments [20]. Preparation of Fragment D-D from Factor XllI-cross-linked fibrin. Factor XIII purified from citrated bovine plasma by the method of Loewy and colleagues [21] has a specific activity of approx. 300 units per mg of protein, as determined by the dansylcadaverine technique [22]. 1 g offibrinogen is reconstituted in 400 ml of 0.05 M Tris-HC1 buffer at pH 7.5, containing 0.1 M NaCI, and supplemented with 1 ml of purified bovine Factor XIII (3 mg/ml), 6.5 ml of 1 M CaC12 and 0.07 ml offl-mercaptoethanol (0.002 M final concentration). Clotting is induced by the addition of 0.5 ml of bovine thrombin (1000 N.I.H. units/ml) and the clot is incubated at room temperature for 18 h and then separated from the liquor by pressing through a fine nylon cloth. This fibrin preparation does not dissolve in 2 ~ acetic acid and yields less than 2 ~ of soluble protein in the supernatant fluid after overnight incubation. The clot is shredded into small pieces, suspended in 100 ml of 0.15 M Tris. HC1 buffer at pH 7.8 and incubated at 37 °C with gentle magnetic stirring. Human plasmin (10.8 C.T.A.* units/ml) (Michigan State Department of Health, Detroit, Mich.) is added as three 0.8-ml portions at the beginning of digestion and after delays of 6 and 12 h. After a lapse of 24 h, the small amount of particulate material is removed by centrifugation at 3000 x g for 15 min. Fragment D-D is purified from the digestion mixture by gel filtration on a Sephadex G-200 column (2.5 x 170 cm), using 3 M urea in 0.025 sodium citrate, adjusted to pH 5.5 with 5 ~ acetic acid. Polyaerylamide gel electrophoresis. Electrophoresis of non-reduced and reduced proteins are performed in 7~o polyacrylamide gels (0.5 x 7 cm) cont~tining 0.1 sodium dodecyl sulfate [23]. Reduction of disulfide bonds is achieved by incubation of 0.1 ml of protein (5 mg/ml) with 0.4 ml of a solution containing 9 M urea, 3 ~ sodium dodecyl sulfate and 3 ~ fl-mercaptoethanol for 24 h at 37 °C. Quantitative determination of NH2-terminal amino acids utilizes the thioacetyl-thioglycolic acid technique as described by Weinstein and Doolittle [24]. The NH2-terminal residue is thioacetylated and split from the polypeptide chain by trifluoroacetic acid, after which the thioacetyl group is removed by acid hydrolysis. The liberated free amino acid is recovered, then identified and quantitated using an automatic amino acid analyzer (Model 119, Beckman, Palo Alto, Calif.). The procedure is calibrated with three standards, human fibrinogen, bovine fibrinogen and albumin, which contain two alanine and two tyrosine, two glutamic acid and two tyrosine and one aspartic acid amino-terminal residues per mol, respectively. Between 20 and 50 nmol of Fragment D (Stage 2 and Stage 3 digests) and Fragment D-D are * Committee on ThrombolyticAgents (1969) Thromb. Diath. Haemorrh. 21, 259-272.

tested, using 20 nmol of norleucine as an internal standard prior to each sample application on the amino acid analyzer. Equilibrium sedimentation. Samples are dialyzed against 0.3 M NaCI, 0.05 M Tris- HC1, 0.05 M e-aminocaproic acid at pH 7.0 for 18 h. Final protein concentration is between 2 and 3 mg/ml. Classical low speed, short column sedimentation equilibrium studies are performed in the analytical ultracentrifuge (Spinco Model E, Beckman, Palo Alto, Calif.). Runs are performed in various rotors and centerpieces ranging from An-D to An-E with speeds from 5200 to 10 000 rev./min at 20 °C. Schlieren optics are employed for all runs and the resultant concentration gradient in the cell is measured on photographic plates with a Gaertner microcomparator. Partial specific volume (f) is determined on a precision densimeter (H. Paar, Graz, Austria) using the mechanical oscillator technique [25], which is accurate to ~ 0.005. Some runs show heterogeneity at the meniscus, in which case the best line through the linear portion of the curve is utilized. Calculations of molecular weight are made as described by Schachman [26], in which 9 is the density of the solvent, ~ is the angular velocity of the centrifuge rotor in radians per second and R is the gas constant, according to the equation mol. wt. =

2 RT (1 -- ~9)~ 2

d In C dx z

RESULTS

Purified fibrinogen and fibrin derivatives Elution of a plasmic digest of cross-linked fibrin through a Sephadex G-200 gel column yields two major peaks (Fig. 1), acrylamide gel electrophoresis of which demonstrates that the first contains primarily Fragment D-D and the second, Frag-

AZBO

. . . . . . . . .

I 300

i

i

45O

600

EFFLUENT VOLUME (ml)

Fig. 1. Elution pattern of plasmic digest of Factor XIII-cross-linked fibrin clot through a Sephadex G-100 gel column. The composition of pooled fractions I and II is shown in Fig. 2.

D-D

- .....

E . . . . .

digest

I

17

Fig. 2. Sodium dodecyl sulfate-polyacrylamidegel electrophoresis of non-reduced plasmic lysate of cross-linked fibrin and fractions I and II obtained after Sephadex G-200 gel column elution of this material (Fig. 1). Electrophoresis in 7 ~ gels; sample size is 10/~g. Fraction I contains primarily (greater than 98 700)Fragment D-D, with trace amounts of heavier and lighter components; fraction II contains Fragment E (fibrin). ment E (Fig. 2). The Fragment D-D preparation obtained in this way is greater than 98 ~ pure, as determined by densitometric analysis. Fig. 3 shows the sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of non-reduced Fragment D-D and selected derivatives of fibrinogen, namely Fragments X (Stages 1 and 2), Y, and D (Stages 2 and 3). Fragment X (Stage 1) and (Stage 2) have essentially the same electrophoretic mobility, indicating approximately equal molecular weights, although the Stage 1 preparation contains a significant proportion of slightly heavier material. Three components are present in the heterogeneous Fragment D preparations. The Stage 2 material contains mostly the heaviest component; the Stage 3 material, mostly the smallest. The major component of the Fragment D-D preparation is slightly larger than Fragment Y. Figs. 4 and 5 show the sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of these fragments after reduction of disulfide bonds. Fragments X (Stage 1 and Stage 2) and Y have approximately the same polypeptide chain structure, except that only Fragment X (Stage 1) has a small amount of intact Bfl chain (58 000) and Aa chain (28 000) remnants, and Fragment X (Stage 2) and Fragment Y have decreasing amounts of intact ~, chain (47 000) and increasing amounts of the 40 000 chain remnant and light remnants of about 13 000 molecular weight. Fragment D (Stage 2 and Stage 3) and Fragment D-D all have the 43 000 and 13 000 dalton components, considered to be of Bfl and Aa chain origin, respectively [17, 27]. The

X(St,t

X(StI2)

Y

D-D

D(St.2) D(St.3)

Fig. 3. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of non-reduced plasmic derivatives of fibrinogen and cross-linked fibrin. With the exception of Fragment D-D obtained from cross-linked fibrin, all of the derivatives are obtained from fibrinogen digests.

58,000

_ _

51,000

. . . . .

47,000

. . . . .

40,000

. . . . .

26,000

. . . . .

13,000--

_

.

~

. . . .

~

. . . .

. . . .

X(St.1) Fig. 4. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of reduced Fragment X (Stage 1), Fragment X (Stage 2) and Fragment Y, obtained from plasmic digests of fibrinogen.

80,000

.

.

.

43,000

. . . .

39,000

. . . .

32,000

------

26,000

. . . .

.

.

13,000 ------

Fig. 5. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of reduced Fragment D derivatives. Fragment D (Stage 2) and Fragment D (Stage 3) obtained from plasmic digests of fibrinogen; Fragment D-D (fibrin) obtained from plasmic digest of cross-linked fibrin. y chain remnants are of 39 000, 32 000 and 26 000 daltons, with the heavier one predominant in Fragment D (Stage 2) and the lighter ones in Fragment D (Stage 3). Fragment D-D has no representation of the 39 000, 32 000 and 26 000-dalton 7 chain remnants, but rather an 80 000-dalton component, considered to be an isopeptidebound dimer of the 7 chain remnants [17, 18].

Plasmic degradation of Fragment Y Timed digestion of Fragment Y (1 mg/ml) is initiated by the addition of plasmin (0.1 unit/ml final concentration), then inhibited with soybean trypsin inhibitor (0.1 mg/ml final concentration) after specific time intervals and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Fig. 7). Fragment Y is initially (10 min) converted to Fragment E and Fragment D (Stage 2), with ultimate conversion (30 and 180 min) of the latter to the more heterogeneous Fragment D (Stage 3) group of derivatives. The initial cleavage reaction releases little or no small peptide material, as reflected by the presence of only 2.1 ~ of non-trichloroacetic acid-precipitable material after 10 min of digestion.

Equilibrium sedimentation The molecular weights o f purified derivatives, as determined by equilibrium sedimentation data and direct measurement of partial specific volume, are shown in

st. 2

y

X

Y

0

10

30

TIME OF DIGESTION

180

|rain)

Fig. 6. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of Fragment Y and timed plasmic digests of Fragment Y. Digestions stopped at indicated times by the addition of soybean trypsin inhibitor (0.1 mg/ml final concentration). A Stage 2 plasmic digest of fibrinogen ("st. 2")

is included (left) for comparison with the electrophoretic positions of Fragments X, Y, D (Stage 2) and E. Table I. The observed molecular weights of Fragment X obtained from Stage 1 and Stage 2 digests are not significantly different (261 000 and 272 000, respectively). suggesting that the lower thrombin coagulability of the Stage 2 preparation may not be associated with the loss of peptide material. The average molecular weight of three separate runs of Fragment Y is 148 000 ~ 11 000. The mean of seven determinations of Fragment D (Stage 2) is 103 500 ± 11 000 as compaled with 79 000 daltons for the Fragment D preparation obtained fiom a Stage 3 digest. Fragment E purified from Stage 2 and Stage 3 digests has essentially the same molecular weight averaging 41 500 for four determinations. Fragment D-D has a molecular weight of 189 000 -~ 8500.

Quantitative amino-terminal amino acid analysis The type and quantity of the amino-terminal residues of the Fragment D derivatives purified from plasmic lysates of fibrinogen and cross-linked fibrin are shown in Table II. The number of tool of each residue is calculated according to the molecular weight values obtained by equilibrium sedimentation (Table I). Aspartic acid is a major residue in all three derivatives present at 2.1 mol/mol in Fragment D (Stage 2), 1.1 mol/mol in Fragment D (Stage 3) and 3.1 mol/mol in Fragment D-D. Valine and glutamic acid are present in Fragment D (Stage 3) (0.9 and 1.0 mol/mol, respectively) and in Fragment D-D (1.8 and 1 mol/mol, respectively), but not in

9 TABLE I MOLECULAR WEIGHT OF PLASMIC DERIVATIVES OF FIBRINOGEN AND CROSSLINKED FIBRIN AS DETERMINED BY EQUILIBRIUM SEDIMENTATION Fragment

X (Stage 1) X (Stage 2) Y D (Stage 2) D (Stage 3) E (Stage 2) E (Stage 3) D-D

Substrate

Fibrinogen Fibrinogen Fibrinogen Fibrinogen Fibrinogen Fibrinogen Fibrinogen Cross-linked fibrin

Partial specific volume (ml/g) 0.718 0.718 0.725 0.710 0.710 0.722 0.722 0.710

Calculated molecular weight No of runs

Mean

4 2 3 7 1 2 2 3

261 000 272 000 148 000 103 500* 79 000 40 000 43 000 189 000

* Standard deviation of these seven determinations is -- 11 000 daltons.

TABLE II QUANTITATIVE AMINO-TERMINAL AMINO ACIDS OF FRAGMENT D DERIVATIVES Fragment

D (Stage 2) D (Stage 3) D-D

Substrate

Fibrinogen Fibrinogen Cross-linked fibrin

NH2-terminal residue* Asp

Val

Ala

Glu

Gly

Total

2.1 1.1 3.1

-0.9 1.8

0.9 ---

-1.0 1.0

--1.0

3.0 3.0 6.9

* The data are presented as the number of mol of residue per tool of protein, using molecular weight values determined as in Table I of 103 500, 79 000 and 189 000 for Fragment D (Stage 2), Fragment D (Stage 3) and Fragment D-D, respectively.

F r a g m e n t D (Stage 2). A l a n i n e is p r e s e n t only in F r a g m e n t D (Stage 2) a n d glycine only in F r a g m e n t D - D . The t o t a l o f all N H 2 - t e r m i n a l residues is 3 m o l / m o l in b o t h F r a g m e n t D derivatives derived f r o m fibrinogen, a n d slightly m o r e than d o u b l e this value (6.9 m o l / m o l ) in F r a g m e n t D - D . DISCUSSION It is generally a g r e e d t h a t the two halves o f the fibrinogen molecule are j o i n e d b y disulfide b o n d s at their a m i n o - t e r m i n a l p o r t i o n s [3-7] a n d t h a t this j u n c t i o n a l a r e a is l i b e r a t e d as F r a g m e n t E by p l a s m i n [10-12]. This f r a g m e n t is c o m p o s e d o f six p o l y p e p t i d e chains, three f r o m each h a l f o f fibrinogen, with an overall m o l e c u l a r weight o f 35 000-50 000 [8, 12, 28-31], d e p e n d i n g u p o n the technique a n d p r e p a r a t i o n utilized. F r a g m e n t D clearly originates f r o m m i d - c h a i n a n d c a r b o x y - t e r m i n a l p o r t i o n s o f fibrinogen [4, 5]. However, the structure o f F r a g m e n t D o b t a i n e d f r o m fibrinogen d e p e n d s u p o n w h e t h e r the c a r b o x y - t e r m i n a l areas o f each h a l f o f the molecule are j o i n e d t o g e t h e r by disulfide bonds. A c c o r d i n g to the a s y m m e t r i c scheme o f d e g r a d a t i o n [2, 8, 9], these F r a g m e n t D regions are n o t j o i n e d b y c o v a l e n t bonds,

10 and a molecule of three chains would lze liberated by plasmin from each of the two halves of fibrinogen. According to the dimeric hypothesis [13-15], the two Fragment D regions in fibrinogen are joined by disulfide bonds within the parent fibrinogen molecule, and Fragment D would consist of six chains, derived from both halves of fibrinogen. Both hypotheses consider that Fragment D exists in multiple molecular forms that are distinguished easiest by molecular weight, but all such derivatives would have the postulated three [4-7, 17-19, 27, 32] or six [13, 15] polypeptide chain structure, regardless of the size of the intact molecule. The clearcut difference of a three or six chain structure for Fragment D is assessed in this study by two approaches. First, the molecular weights of intact Fragment D and Fragment D-D as measured by equilibrium sedimentation are correlated with the size of the individual chains of the reduced proteins, as seen in the sodium dodecyl sulfate polyacrylamide gels after electrophoresis. Second, the molecular weights of the intact molecules are utilized in calculations of quantitative NH2-terminal analysis of the same preparations of Fragment D; the total amount of NH2-terminal residues reflects the number of polypeptide chains. Three Fragment D preparations are identified and tested, those from Stage 2 and 3 plasmic digests of fibrinogen and the third (Fragment D-D) from a plasmic lysate of Factor XIIl-cross-linked fibrin. Fragment D (Stage 2) has been found [17] to contain primarily a ~ remnant of molecular weight approx. 39 000, a Bfl remnant (43 000) and an Aa remnant (13 000) (Fig. 5). These bands correspond closely to those of the earliest Fragment D derivative (Fragment D1) noted by Mosesson and colleagues [13], who specify / 71 (42 000), / f14 (42 000) and / a/is (7000) components. Fragment D (Stage 3) has the same Bfl and Aa remnants as Fragment D (Stage 2), and differs from the latter by the predominance of 26 000 and 32 000 dalton ), remnants rather than that of 39 000 daltons. The Fragment D-D bands are the same as those of Fragment D for Bfl and Aa remnants; the distinguishing feature is an 80 000 7-), chain derivative, as has been shown by Pizzo and colleagues [17]. These results are summarized in Table lII, which also notes the observed molecular weight of the intact fragments. The models for Fragment D postulated by the asymmetric and dimeric schemes are drawn according to this information and are shown schematically in Fig. 7. The quantity of each polypeptide chain remnant that comprises a given molecule is restricted by the measured molecular weight of the respective molecule (Table I, Fig. 5). Thus, the molecular weight of Fragment D (Stage 2) of 103 500 allows a maximum of only one of the Aa, Bfl and ), chain remnants, the summated molecular weight of which is 95 000. Assuming two of each chain remnant (Fig. 7, "DI"), the molecular weight would total 190 000, a value which is not compatible with the observed molecular weight for the intact molecule. Similarly, Fragment D (Stage 3) allows for only three polypeptide chain remnants, and using the predominant chain remnant (26 000), the summated total for the three chains is 82 000, consistent with the observed 79 000 of the intact molecule. A six chain Fragment D (Stage 3) would not be compatible with the determined molecular weight of the intact molecule. Fragment D-D has a molecular weight of 189 000 as determined by equilibrium sedimentation. This derivative can be constructed in the way originally suggested [17], by summation of one ~,-~, chain (80 000) and two each of the Bfl (43 000) and Aa (13 000) chain remnants. Such a reconstruction is not compatible with the dimeric hypothesis [13-15], which postulates six chains for Fragment D obtained from

11 ASYMMETRIC SCHEME

DIMERIC SCHEME

y

t I

B# Aa Aa

I

y

A= A=

BB

B#

FIBRINOGEN

FIBRINOGEN

42

"i~1

/,/

/~/r

/a, & /~,

D(STAGE 2)

/×

i~

,2

__

,2

Di

-"[-

(/r)2

(I)i)2

II

(/y, lz

II

(&lz

II

D-D

4

D I POLYMER

Fig. 7. Schematic representation of two models of fibrinogen and the Fragment D molecules postulat ed to derive from each. Fragment D (Stage 2) and Fragment D-D (left), as postulated by the asymmetric scheme of plasmic degradation and Fragment D1 and "Dx polymer" (right), according to the dimeric scheme. The labeling of Act, Bfl and 9) chain remnants is based upon cleavages of the amino-, carboxy- or both parts of a chain, in which case the symbol is written, for A a remnants, as/ct, a/, /a/, respectively [1]. The single vertical lines represent disulfide bonds; those drawn more heavily connect the two halves of fibrinogen together. The asymmetric scheme [2, 8, 9] (left) postulates one locus of such inter-half molecule disulfide bonds, located at the NH2-terminal area (at the left of the fibrinogen model) [4-7, 17, 27, 29, 33, 34, 38]. Since Fragment D derives from the carboxy-terminal regions of fibrinogen, two Fragment D molecules of three polypeptide chains each are liberated from each fibrinogen molecule, represented here as " D (Stage 2)". The observed molecular weight of the chains after reduction of disulfide bonds are noted, according to the observations in Fig. 5. Fragment D-D is the major derivative of a plasmin lysate of Factor XllI-cross-linked fibrin, and consists essentially of two Fragment D (Stage 2) molecules [17-19] held together by 7-~' isopeptide bonds [16], drawn as double vertical lines. After reduction, the ~,-~, chain has a molecular weight of 80 000 [17, 18] (Fig. 5). The dimeric scheme (right) postulates two loci between the two halves of fibrinogen [13-15], and the initial Fragment D derivative cleaved by plasmin, called "DI", retains one interhalf disulfide locus. This latter locus is the basis for the postulated six chain structure of Fragment D, in contrast with the three chain structure of Fragment D (Stage 2). The molecular weights of chains of Fragment D1 after disulfide reduction are as reported [13], and bear a close resemblance to the values noted for Fragment D (Stage 2) [17].

12 TABLE III COMPARISON OF THE TOTAL MOLECULAR WEIGHT OF INTACT FRAGMENT D MOLECULES WITH THAT OF THE POLYPEPTIDE CHAIN REMNANTS OBSERVED IN REDUCED SAMPLES Fragment

Substrate

Molecular weight* Intact Polypeptidechains molecule Aa Bfl y

D (Stage 2) D (Stage 3)

Fibrinogen Fibrinogen

103 500 79 000

13 000 13 000

43 000 43 000

D-D

Cross-linked fibrin

189 000

13 000

43 000

39 000 26 000 or 32 000 80 000

Total 1 each

2 each

95 000 82 000 or 88 000 96 000

190 000 164 000176 000 192 000"*

* Molecular weight values for the intact molecule are obtained from equilibrium sedimentation measurements; those for polypeptide chains, by analysis of sodium dodecyl sulfate-polyacrylamide gel electrophoretic results. ** This summated total considers the 80 000-dalton 7 chain remnant to be a double representation of ), chains, joined by Factor XIII-induced isopeptide bonds [17]. Two of the Act and Bfl chain remnants are added to this value to arrive at a sum total of 192 000. fibrinogen and which would predict a large polymer of Fragment D derived from cross-linked fibrin (Fig. 7). Just as the molecular weight values of the intact molecule are essential for the calculations shown in Table III, these are also utilized in the calculations of molar quantities of NH2-terminal residues of the derivatives. These data (Table II) indicate that the Fragment D preparations from fibrinogen have three NHz-terminal amino acid residues per molecule, and therefore, three polypeptide chains. Fragment D-D has just under seven residues per molecule, compatible with the structure shown in Fig. 7. The major amino-terminal derivatives that we observe for Fragment D (Stage 2) (Table II) are aspartic acid (about 2 mol/mol) and alanine (about 1 mol/mol), which agree with the results of Collen and colleagues [33] for their early Fragment D derivative, which has aspartic acid on the Aa and Bfl remnants and alanine on the y remnant. The major residues for Fragment D (Stage 3) are aspartic acid (about 1 tool/tool) and valine (1 mol/mol), an observation which agrees partially with the results of Furlan and colleagues [34] for a Fragment D of 83 000-94 000, in which valine, not aspartic acid, is on the Aa chain remnant and aspartic acid is retained on the Bfl remnant. Our observation of glutamic acid has not been made by others [33, 34], who note serine and/or glycine at the NH2 terminus of the ), chain remnant. The major residues of Fragment D-D are aspartic acid (about 3 mol/mol), valine (about 2 mol/mol), glycine (1 mol/mol) and glutamic acid (1 mol/mol). These data are compatible with a 9'-Y chain isopeptide-bound Fragment D-D molecule (Fig. 7) that has undergone limited cleavage of the A a and ), chains. The details of plasmic degradation of Fragment D-D are not yet known. The original description of the asymmetric cleavages of Fragments X and Y utilized molecular weight values obtained by Svedberg analysis, and these are corro-

13 borated by the studies presented here, which utilize the technique of equilibrium sedimentation (Table I). In addition, the molecular weight of Fragment X (Stage 2), Fragment D (Stage 2) and Fragment D-D have been determined by ultracentrifugal analysis. The molecular weight of purified Fragment X preparations of Stage 1 and Stage 2 digests are essentially the same (261 000 and 272 000, respectively), and are in good agreement with previous reports using Svedberg analysis and polyacrylamide gel electrophoretic techniques in the presence of sodium dodecyl sulfate [4, 7, 8, 27]. Similarily, the results for Fragment Y and Fragment D (Stage 3) agree well with previous values [2, 4-8, 17, 18, 27-34]. That for Fragment E is somewhat lower than previously reported by us [8], but within the range of values noted by others using a variety of techniques [4, 12, 28-32]. Fragment D (Stage 2) is significantly larger than Fragment D (Stage 3), compatible with the postulated cleavages and heterogeneity of the ~, chain remnants [17]. Considering that Fragment Y initially yields Fragment D (Stage 2) rather than Fragment D (Stage 3) upon cleavage by plasmin (Fig. 6), the asymmetric hypothesis is still entirely compatible with these newly determined molecular weight values. According to this scheme, Fragment X (Stage 2) (approx. 265 000 daltons) would still be cleaved into Fragment Y (148 000) and Fragment D (Stage 2) (103 500), and Fragment Y would be split into Fragment D (Stage 2) and Fragment E (41 500). Recent studies utilizing novel approaches to these problems have supported the asymmetric cleavage (2 D q- 1 E) hypothesis. Furlan et al. [34] purified the polypeptide chains of Fragment D derivatives by gel elution and identified the Bfl chain remnant by tryptic mapping. The molecular weight of this remnant is 45 000 in all Fragment D derivatives of 83 000 or more, contrary to the dimeric view [13] which postulates an early degradation of this chain. Donovan and Mihalyi [35] conclude from studies of endothermal transitions of fibrinogen and mixtures of Fragment D and E, measured by scanning calorimetry, that fibrinogen consists of two D and one E nodule. Takagi and Doolittle [36] show by analysis of short peptide sequences at the amino-terminal end of Fragments X, Y, D and E that the asymmetric cleavage scheme is reasonable and by sequencing of nine carboxy-terminal residues of the Bfl chains in fibrinogen and Fragment D [37] that a dimeric Fragment D with degraded fl chain remnants is unlikely. Tranqui-Pouit et al. [38] show that the electron microscopic patterns of coagulable derivatives oi fibrinogen are most compatible with only a single locus of disulfide bonds between the two halves of fibrinogen. Therefore, the data presented in this study as well as from independent approaches indicate that fibrinogen has disulfide bonds linking the two halves of the molecule at the NH2-terminal region, but not at the carboxy-terminal area, from which Fragment D originates. The data are not compatible with the existence of a dimeric, six chain Fragment D molecule derived from fibrinogen. The asymmetric cleavage hypothesis for plasmic degradation of fibrinogen is supported, as is the conclusion that two Fragment D molecules are produced from each parent fibrinogen molecule. ACKNOWLEDGEMENTS Supported by Grants No. 12148 and No. 14217 from the National Heart and Lung Institute, National Institutes of Health, Bethesda, Md, U.S.A.

14 REFERENCES 1 2 3 4 5 6 7 8 9 10 11

Blomb~ck, B. and Johnson, A. J. (1971) Thromb. Diath. Haemorrh. Suppl. 51, 251-256 Marder, V. J., Shulman, N. R. and Carroll, W. R. (1967) Trans .Assoc. Am. Phys. 53, 156-167 Blomb~ick, B. (1970) Symp. Zool. Soc. Lond. 27, 167-187 Pizzo, S. V., Schwartz, M. L., Hill, R. L. and McKee, P. A. (1972) J. Biol. Chem. 247, 636-645 Marder, V. J., Budzynski, A. Z. and James, H. L. (1972) J. Biol. Chem. 247, 4775-4781 Mills, D. A. (1972) Biochim. Biophys. Acta 263, 619-630 Furlan, M. and Beck, E. A. (1972) Biochim. Biophys. Acta 263, 631-644 Marder, V. J., Shulman, N. R. and Carroll, W. R. (1969) J. Biol. Chem. 244, 2111-2119 Marder, V. J. (1968) In Fibrinogen (Laki, K., ed.), pp. 339-358, Marcel Dekker, Inc., New York Marder, V. J. (1971) Scand. J. Haematol. Suppl. 13, 21-36 G~rdlund, B., Kowalska-Loth, B., Grond~hl, N. J. and Blomb~ick, B. (1972) Thromb. Res. 1, 371-388 12 Kowalska-Loth, B., G~,rdlund, B., Egberg, N. and Blomb~.ck, B. (1973) Thromb. Res. 2, 423-450 13 Mosesson, M. W., Finlayson, J. S. and Galanakis, D. K. (1973) J. Biol. Chem. 248, 7913-7929 14 Mosesson, M. W., Galanakis, D. K. and Finlayson, J. S. (1974) J. Biol. Chem. 249, 4656--4664 15 Mosesson, M. W. (1974) Semin. Thromb. Hemostasis 1, 63-84 16 Chen, R. and Doolittle, R. F. (1971) Biochemistry 10, 4486-4491 17 Pizzo, S. V., Taylor, Jr., L. M., Schwartz, M. L., Hill, R. L. and McKee, P. A. (1973) J. Biol. Chem. 248, 4584-4590 18 Gaffney, P. J. and Brasher, M. (1973) Biochim. Biophys. Acta 295, 308-313 19 Kope6, M., Teisseyre, E., Dudek-Wojciechowska, G., Kloczewiak, M., Pankiewicz, A. and Latallo, Z. S. (1973) Thromb. Res. 2, 283-291 20 Marder, V. J., James, H. L. and Sherry, S. (1969) Thromb. Diath. Haemorrh. 22, 234-239 21 Loewy, A. G., Dunathan, K., Kriel, R. and Wolfinger Jr., H. L. (1961) J. Biol. Chem. 236, 26252633 22 Lorand, L., Urayama, T., deKiewiet, J. W. C. and Nossel, H. L. (1969) J. Clin. Invest. 48, 10541064 23 Weber, K. and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 24 Weinstein, M. J. and Doolittle, R. F. (1972) Biochim. Biophys. Acta 258, 577-590 25 Kratky, O., Leopold, H. and Stabinger, H. (1973) Methods in Enzymology (Hirs, C. H. W. and Timashelf, S. N., eds.), Vol. 27, pp. 98-110, Academic Press, Inc., New York 26 Schachman, H. K. (1957) Methods in Enzymology (Colowick, S. P. and Kaplan, N. O., eds.), Vol. IV, p. 32, Academic Press, Inc., New York 27 Budzynski, A. Z., Marder, V. J. and Shainoff, J. R. (1974) J. Biol. Chem. 249, 2294-2302 28 Nussenzweig, V., Seligmann, M., Pelmont, U. and Grabar, P. (1961) Ann. Inst. Pasteur 100, 377-389 29 Furlan, M. and Beck, E. A. (1973) Biochim. Biophys. Acta 310, 205-216 30 Dudek, G. A., Kloczewiak, M., Budzynski, A. Z., Latallo, Z. S. and Kope6, M. (1970) Biochim. Biophys. Acta 214, 44-51 31 Nil6hn, J. E. (1967) Thromb. Diath. Haemorrh. 18, 89-100 32 Gaffney, P. J. and Dobos, P. (1971) FEBS Lett. 15, 13-16 33 Collen, D., Kudryk, B., Hessel, B. and Blomb~ick, B. (1975) J. Biol. Chem. 250, 5808-5817 34 Furlan, M., Kemp, G. and Beck, E. A. (1975) Biochim. Biophys. Acta 400, 95-111 35 Donovan, J. W. and Mihalyi, E. (1974) Proc. Natl. Acad. Sci. U.S. 71, 4125-4128 36 Takagi, T. and Doolittle, R. F. (1975) Biochem. 14, 940-946 37 Takagi, T. and Doolittle, R. F. (1975) Biochim. Biophys. Acta 386, 617-622 38 Tranqui-Pouit, L., Marder, V. J., Suscillon, M., Budzynski, A. Z. and Hudry-Clergeon, G. (1975) Biochim. Biophys. Acta 400, 189-199