THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 250, No. 12, Issue of June 25, pp. 4678-4683, 1975

Printed in U.S.A.

Partial Chemical Characterization of Rat Fibrinogen* (Received for publication, September 3, 1974)

HESSEL BOUMA, III, AND GERALD M. FULLER

From the Department of Human Biological Chemistry and Genetics, The University of Texas Medical Branch, Galveston, Texas 77550

Rat fibrinogen has been purified and compared with bovine and human fibrinogen with respect to a number of chemical characteristics, including molecular size, charge distribution, NH,-terminal amino acids, total amino acid composition, and interspecies immunological cross-reactivity. Although human and bovine fibrinogen demonstrated three nonidentical polypeptide chains by sodium dodecyl sulfate gel separations and by CM-cellulose separations, rat fibrinogen Acu and B/l chains exhibited identical molecular weight sizes as well as identical charges. The presence of two nonidentical chains in these preparations was shown by qualitative NH,-terminal sequence analyses. The y chain of rat fibrinogen was also shown to be quite distinct from the y chains of human and bovine fibrinogen in its elevated content of cysteinyl and methionyl residues. Rat fibrinogen possesses the first reported blocked y chain NH,-terminal amino acid of any species. It is concluded that, although many chemical properties of rat fibrinogen are unique, the basic molecular structure has remained consistent when compared with that of fibrinogen from the vertebrates studied thus far. Moreover, the inducibility of this system, together with the partial chemical characterization of the fibrinogen molecule, provides important information for the use of rat fibrinogen as a model system in studying the biosynthesis and assembly of this complex molecule.

Many of the chemical features of fibrinogen from a number of vertebrate species have been characterized (l-3). It is known, for example, that fibrinogen from vertebrates is com- posed of three pairs of nonidentical polypeptide chains (Aa2, BP,, r2)’ which are bound together through a complex set of disulfide bridges. Fibrinogen has a mass of approximately 340,000 * 20,000 daltons. The physical properties of the molecule remain fairly consistent throughout those species that have been studied. It is not surprising that the bulk of the information concerning the physiochemical properties of fi- brinogen has been established with the use of either the human or bovine molecules, since the plasma from these sources is readily available in large quantities. More recently, interests of several laboratories have been directed toward gaining an understanding of the processes of assembly of this complex molecule as well as learning more about the control of its biosynthesis. The clottable molecule, fibrinogen, purified from the laboratory white rat, was characterized with respect to some of its chemical properties in an effort to establish a convenient laboratory model for the study of its biosynthesis and assembly. Several unique chemical features of this fibrino- gen are described in this report.

*This work was supported in part by Research Grants HL 16445, CA 14675, and HD 03321 from the National Institutes of Health.

’ The International Committee on Nomenclature of Blood Clotting Factors designates the fibrinogen chains with the fibrinopeptides still attached as Aoi and BP. The third polypeptide chain is designated y. Fibrin polypeptide chains are called 01, 8, and y respectively.

MATERIALS AND METHODS

Rat fibrinogen was purified from plasma obtained from adult rats which had been induced into a hyperfibrinogenemic state with a subdermal injection of 1.0 ml of commercial turpentine (4) 48 hours prior to exsanguination. Whole blood was withdrawn from the femoral artery into heparinized syringes and immediately chilled to 4”. Human fibrinogen was purified from the pooled outdated plasma obtained from the University of Texas Medical Branch blood bank. Bovine fibrinogen was purified beginning with commercially prepared Cohn Fraction 1 (Pentex; 90% clottable).

The concentration of plasminogen in the plasma was reduced by passing rat and human plasma over a lysine-coupled Sepharose 4B affinity column, according to the method of Deutsch and Mertz (5). Fibrinogen was precipitated from plasma with a 25% (w/v) ammonium sulfate fractionation. This crude fibrinogen preparation was redis- solved into 0.10 M sodium phosphate (pH 7.2)-0.15 M NaCl buffer, dialyzed against one change of the same buffer, and then re- precipitated with a 20% (w/v) ammonium sulfate fractionation. The resulting fibrinogen precipitate was redissolved in a minimum amount of the phosphate buffer, dialyzed against the same buffer, and then subjected to chromatography on Sephadex G-200 (50 x 80 cm). In a control experiment, 1 ml of 10 mM e-aminocaproic acid was added to the syringe prior to drawing 9 ml of blood. In addition, 1.0 mM c-aminocaproic acid was added to all buffers during the purification procedure. When e-aminocaproic acid was added, the step using the lysine-Sepharose chromatography was omitted. The commercial bo- vine fibrinogen was purified by a 20% (w/v) ammonium sulfate precipitation followed by (G-200 Sephadex) chromatography.

Isolation of Fibrinogen Chains-Rat and human fibrinogen samples were reduced and alkylated to the S-carbamylmethyl derivatives using iodoacetamide, essentially according to the procedure of McDonagh et al. (6), except that 1.64 mmol of 2-mercaptoethanol/g of fibrinogen were used for reduction. Alkylation was accomplished by using a molar excess of iodoacetamide 5 times that of the reducing agent, added

4678

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

4679

directly to the fibrinogen solution. The alkylated protein was dialyzed against five changes of 0.10 M ammonium bicarbonate at 4”, after which the protein was lyophilized and stored at -20“. Complete reduction and alkylation were verified by amino acid analyses which revealed an absence of half-cysteine.

The reduced and blocked polypeptide chains were then separated by CM-cellulose chromatography in 8.0 M urea, as described by McDonagh et al. (6), except that a simple two-chamber gradient was used consisting of 500 ml of starting buffer (50 mM sodium acetate, pH 4.8) and 500 ml of the final buffer (170 rnM sodium acetate, pH 5.3). Each protein region was pooled, dialyzed against several changes of 0.10 M ammonium bicarbonate, lyophilized, and stored at -20”.

Polyacrylanide Gel Electrophoresis-Polyacrylamide gel electro- phoresis carried out in sodium dodecyl sulfate was routinely used for the analytical determination of fibrinogen and fibrin components. Molecular weight estimations were performed on 10% acrylamide gels containing 0.1% of the anionic detergent Na dodecyl-SO ,, 2 according to the method of Weber and Osborn (7), using the reported molecular weights for pancreatic ribonuclease (13,700), immunoglobulin light chain (23,000) and heavy chain (50,000), and human fibrinogen Aa chain (73,000), B/3 chain (56,000) and y chain (47,000) as reference standards (8). The gels were scanned on a Gilford linear scanning device at 650 nm. Molecular weights were calculated from the gel scans, using pancreatic ribonuclease (13,700) as a migratory front. Five per cent polyacrylamide gels were also used for establishing the purity of the proteins and their constituents. The buffer system for these gel analyses as well as those used for molecular weight determinations was 0.05 M sodium phosphate buffer, pH 7.2, containing 0.1% Na dodecyl- SO,. Acrylamide gels were run with additional buffer systems: acid urea gels, pH 1.0; 8 M urea, sodium acetate gels, pH 5.8; 8 M urea, phosphate gels, pH 7.2; 8 M urea, 0.1% Na dodecyl-SO,, Tris-glycine gels, pH 8.3; and 8 M urea, 0.1% Na dodecyl-SO, phosphate gels, pH 7.2. The acrylamide gel concentration was 5% for these experiments.

Amino Acid Analyses-The total amino acid composition of rat and human Cm-fibrinogen as well as the isolated Cm-fibrinogen polypep- tide chains were determined from 1-mg samples hydrolyzed in 1.0 ml of 5.7 N HCl and 0.2 ml of 5% phenol. The hydrolysis tubes were flushed with nitrogen and evacuated prior to being sealed. Hydrolyses were performed at 107” in 24., 48., 72., and 96.hour time courses. Analyses were carried out using a JEOL automatic amino acid analyzer equipped with a two-column system and an integrator. The cysteine content of the reduced and alkylated whole proteins and polypeptide chains was determined as Cm-cysteine.

NH,-Terminal Analyses-Quantitative NH,-terminal analyses were performed by using thioacetylthioglycolic acid as a coupling reagent, as has been previously reported (9). Approximately 20 mg of thi- oacetylthioglycolic acid were added to 80 nmol of purified protein (fibrinogen or isolated polypeptide chains) dissolved in 1.0 ml of 6 M

guanidine HCl in 0.1 M Tris (pH 9.5) and 1.0 ml of pyridine. The mixture was incubated at 40” for 1 hour, maintaining the pH at 9.5 with 1.0 M NaOH. The protein was then washed three times using 6 ml of benzene for each wash and three times with 6 ml of 95% aqueous acetone. The resulting precipitate was taken up in 1 ml of water and freeze-dried. The sample was then incubated in 0.2 ml of trifluoroa- cetic acid at 40” for 20 min and extracted twice with 1.5 ml of 1,2-dichloroethane. The 1,2-dichloroethane extracts were transferred to a hydrolysis tube containing 0.2 ml of 5.7 N HCl and mixed thoroughly. The 1,2-dichloroethane was evaporated under a stream of nitrogen, an additional 0.2 ml of 5.7 N HCl was added, and the sample was hydrolyzed under reduced pressure at 130’ for 4 hours. Amino acid analysis was then carried out on an automatic amino acid analyzer.

Amino Acid Sequence Analyses-Edman Degradation-Purified rat fibrogen as well as isolated ALU-BP chains were analyzed in the Beck- man Automatic Sequencer model 890B which had been updated by the addition of an undercut cup and nitrogen flush mechanism. The proteins were sequenced using the dimethylbenzylamine procedure outlined by Hermodson et al. (10). The phenylthiohydantoins of the amino acids were initially identified using a gas-liquid chromato- graphic detection system and were confirmed by amino acid analysis. Seven degradative cycles were carried out on rat fibrinogen and iso- lated Acu-B/3 chains.

Cyanogen Bromide (CNBr) Cleauuge-Purified rat and human Cm-fibrinogen y chains were dissolved in 70% formic acid at a

‘The abbreviation used is: Na dodecyl-SO, sodium dodecyl sulfate.

concentration of 5 mg/ml (11). Each sample was treated with 30 mg/ml of CNBr for 4 hours at 30”. The reagents were subsequently removed by lyophilization after a 4.fold dilution with distilled water. The peptides generated in this cleavage were examined on 10% polyacrylamide 0.1% Na dodecyl-SO, gel electrophoresis.

Antisera-Adult rabbits were initially immunized with 3 mg of purified bovine fibrinogen, human fibrinogen, or rat fibrinogen emulsi- fied in Freunds’ complete adjuvant (Difco) injected subcutaneously. At lo- and 20-day intervals after the initial injection, additional subcutaneous injections were made of 3 mg of protein each, emulsified in an 8:l Freunds’ incomplete-complete adjuvant. The rabbits were bled at weekly intervals thereafter.

Monospecific Immunoglobulin Purification-Bovine, human, and rat fibrinogen-immunosorbent columns were prepared by covalently binding approximately 50 mg of purified fibrinogen from each species to 20 ml of CNBr-activated Sepharose 48 (Pharmacia) according to the procedures of Cuatrecasas et al. (12). These immunosorbent gels were equilibrated in borate-NaCl solution buffer (100 mM boric acid, 25 mM sodium borate, and 75 mM sodium chloride, pH 8.4).

An immunoglobulin fraction of each antiserum was initially pre- pared by subjecting the serum to a 50% (v/v) precipitation, using saturated ammonium sulfate. The crude immunoglobulin fraction was redissolved in borate-saline buffer and dialyzed in one change. This fraction was then slowly passed over its specific immunosorbent column. Nonspecific proteins were eluted from the columns by washing the columns with the borate-saline buffer until the A,,, of the eluate was less than 0.01. Monospecific antibodies bound to the column were subsequently eluted with 0.10 M glycine-HCI, pH 2.6. The eluted monospecific antibody fraction was dialyzed against the borate-saline buffer and stored at - 20”. Monospecificity and cross-reactivity be- tween antisera and antigens were established by Ouchterlony double immunodiffusion tests and by immunoelectrophoresis.

RESULTS

The purification of bovine, human, and rat fibrinogen by the procedures previously outlined resulted in essentially pure protein when analyzed on polyacrylamide gels. Purified hu- man, bovine, and rat fibrinogen was at least 93 to 95% clottable. In a control experiment a known plasmin inhibitor (13) was added to the syringe during blood collection. In addition, the inhibitor was maintained throughout the purifi- cation process in order to prevent any plasminolytic cleavage. Fibrinogen from hyperfibrinogenemic and normal controls as well as from plasmin-inhibited animals all showed identical chemical and physical properties. Induction to a hyperfi- brinogenemic state was performed only to improve the yield of rat fibrinogen and is accomplished through an incompletely understood acute phase reaction to various trauma states (14, 15).

An electrophoretic comparison of reduced bovine and rat fibrinogen and fibrin reveals a unique characteristic of the latter species (Fig. 1). All three component polypeptide chains were clearly separated in reduced bovine fibrinogen and fibrin, as well as that from other mammalian species observed so far. Reduced rat fibrinogen and fibrin, on the other hand, exhibited only two bands which could not be further resolved on a number of different gel systems. Since fibrin stabilization is mediated through the formation of a peptide bond between the c-amino group of lysine and the carboxamide group of gluta- mine on adjacent y chains, it is possible to identify the -y chain on Na dodecyl-SO, gel electrophoresis by its conversion to the heavier y-y dimer. The fastest migrating band of rat and bo- vine non-cross-linked fibrin is completely converted to the y-y dimer in cross-linked fibrin. In comparing bovine Aac chain with rat Au chain (contained in the rat Acu-BP band), it is shown that the latter migrates faster, reflecting a mass dif- ference of approximately 10,000 to 15,000 daltons. Quantita- tive comparison from gel scans of reduced rat fibrinogen re-

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

4680

FIG. 1. Comparative Na dodecylS0, polyacrylamide (5%) gel elec- trophoresis of bovine and rat fibrinogen and fibrin. A, bovine fibrino- gen; B, rat fibrinogen; C, reduced bovine fibrinogen; D, reduced rat fibrinogen. On gel scans, the Aol-Bfl:y ratio is 1.85:1. E, reduced cross-linked bovine fibrin; F, reduced cross-linked rat fibrin. Reduc- tion was accomplished by the addition of Z-mercaptoethanol to a 1% concentration. Cross-linked fibrin was prepared by clotting fresh plasma with topical thrombin (Parke, Davis), then washing and reducing the fibrin clot with 2-mercaptoethanol.

veals that there is 1.85 times as much stained material in the Ao(-B/3 band than in the y-band, suggesting an equimolar ratio of the three component chains of fibrinogen.

The molecular weights of rat fibrinogen chains were esti- mated by 10% acrylamide Na dodecyl-SO, gel electrophoresis (Fig. 2) using established molecular weight markers. Rat fibrinogen A&-BP chains had an estimated molecular weight of 51,700, and the y chain had an estimated molecular weight of 47,500. The summation of the component molecular weights of the peptide chain yields a total molecular weight of 302,000 for the whole rat fibrinogen molecule. No attempt was made to determine the molecular weight of whole fibrinogen on Na dodecyl-SO, gel electrophoresis directly, since molecular weights in this size range are generally inaccurate and quite variable according to carbohydrate content (16).

The preparative separation of reduced and alkylated human and rat fibrinogen on CM-cellulose is shown in Fig. 3. Gel electrophoresis of human fibrinogen resolves three peaks; however, rat fibrinogen is resolved into only two symmetrical peaks. Thus, both Aa: and BP chains appear to be very similar in size as well as in charge, as shown by Na dodecyl-SO, gel electrophoresis and in CM-cellulose chromatography, respec- tively. Fig. 4 shows a comparative electrophoretic analysis of human and rat fibrinogen before and after separation into individual chains.

Quantitative NH,-terminal analyses of rat fibrinogen, fibrin, and the isolated fibrinogen chains were performed using two different coupling reagents and procedures. The consistent finding in all experiments was the presence of 4 mol of alanine/mol of fibrinogen; 1 mol of alanine/mol (M, 52,000) of Acu-B/3 chains; and 4 mol of glycine/mol of fibrin. Neither isolated y chains nor fibrinogen or fibrin revealed the presence of any detectable NH,-terminal group for the y chain polypep- tide. To our knowledge, this constitutes the first report of such a blocked NH, terminus of the y chain of the fibrinogen molecule of any species.

To verify that two different polypeptides were present in the rat Acu-BP band observed on Na dodecyl-SO, gel electro- phoresis and CM-cellulose chromatography, purified Am-BP chains were subjected to NH,-terminal amino acid sequence

LoGlO M.W.

RELATIVE MIGRATION

FIG. 2. Molecular weight estimations of rat fibrinogen polypeptide chains, by Na dodecylS0, polyacrylamide (10%) gel electrophoresis. Mobility relative to bovine pancreatic ribonuclease was plotted against the known molecular weights of human fibrinogen polypeptide chains and the immunoglobulin chains, according to the method of Weber and Osborn (7).

170 [r

.6(

A 20 30 40 50 60

FRACTION NUMBER

170 -

0” 140 -

f ;

u” - P 110 . 5 . E 80-

50 -

B 20 30 40 50 60

J

FRACTION NUMBER

FIG. 3. Carboxymethylcellulose chromatographic separation of the reduced and alkylated subunit polypeptide chains of (A) human Cm-fibrinogen and (B) rat Cm-fibrinogen. The separations were carried out at room temperature in 8 M urea with a linear gradient (---) consisting of 500 ml of 50 mM sodium acetate (pH 4.8) initial buffer and 500 ml of 170 mM sodium acetate (pH 5.3) final buffer. The identity of each pooled protein fraction was established by Na dodecylS0, polyacrylamide (5%) gel electrophoresis (Fig. 4).

analyses. Seven cycles were performed; each cycle yielded two amino acids the identity of which was in agreement with the reported sequences (17) for the A and B fibrinopeptides of rat fibrinogen. These results confirm the presence of two noni- dentical chains in the preparations.

The separated polypeptide chains of rat Cm-fibrinogen were generally quite similar in amino acid composition to the

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

4681

FE. 4. Comparative Na dodecyl- SO, polyacrylamide (5%) gel electro- phoresis of CM-cellulose separated human and rat Cm-fibrinogen polypep- tide chains. A, human fibrinogen; B, re- duced human fibrinogen; C, D, and E, human Cm-fibrinogen ALY chains, Bb chains, and y chains, respectively; F, rat fibrinogen; G, reduced rat fibrinogen; H, and I, rat Cm-fibrinogen Aa-BB chains and y chains, respectively; J, bo- vine fibrinogen; K, reduced bovine fi- brinogen.

+

Fro 5. Comparative Na dodecylS0, polyacrylamide (10%) gel electrophoresis of cyanogen bromide-cleaved rat and human Cm- fibrinogen y chains. A, rat Cm-fibrinogen y chain; B, CNBr-cleaved rat Cm-fibrinogen y chain; C, CNBr-cleaved human Cm-fibrinogen y chain; D, human Cm-fibrinogen y chain.

reported compositions of human (see Table I) and bovine fibrinogen chains (U-20). The values for serine and threonine were derived after extrapolation of the time course of hy- drolyses back to zero times, while those values for valine, leucine, and isoleucine were averaged from the 72- to 96-hour hydrolyses only. All other amino acid values are averages from all hydrolysates. Two features of the rat fibrinogen molecule are noticeably different from human fibrinogen. First, pub- lished data for human fibrinogen suggest the presence of only 36 disulfide bridges representing both inter- and intrachain disulfide bonds. The amino acid compositions reported here, however, suggest that there may be as many as 42 disulfide

bridges in both human and rat fibrinogen. Second, the comparative amino acid composition of the chains of fibrino- gen indicates the presence of twice as many cysteine and methionine residues in the rat y chain as reported in the human y chain. The difference in methionine content is supported by comparative electrophoretic analysis of rat and human cyanogen bromide cleavage products of rat and human y chains (Fig. 5). Although the bands are insufficiently distinct to quantitate, it is apparent that extensive differences between the rat and human y chains exist, with the rat protein exhibiting significantly more bands in agreement with its elevated methionine content. Particular care had been taken during reduction and alkylation of rat and human fibrinogen to avoid the oxidation of methionine residues. Hence the number of cyanogen bromide-cleaved polypeptide chains significantly reflects the number of methionine residues present within the y chain.

Immunologically, bovine, human, and rat fibrinogen all behaved as classical antigens. When monospecific antibodies directed against bovine, human, and rat fibrinogen were examined by double immunodiffusion gels using plasma from these species, there were varying degrees of immunological cross-reactivity (Fig. 6). All interspecies cross-reactivity tests demonstrated incomplete immunological identity except for the rat fibrinogen antibodies, which did not cross-react with bovine fibrinogen.

DISCUSSION

As a result of shortened Acu chains, the molecular weight cf’ rat fibrinogen is less than that reported for other fibrinogens. The 10% decrease in molecular weight is not particularly striking if one considers that the molecular weight of fibrinogen is generally accepted to be 340,000 f 20,000. It does, however, represent one of the smallest molecular weight sizes reported, although Mosesson et al. (20) have demonstrated certain size and solubility variations in human fibrinogen. Others have demonstrated that the Arw chain of fibrinogen is most suscepti- ble to proteolytic cleavage by plasmin (22). The difference in molecular size of the ACY chain in the rat could be the result of plasmin cleavage. This possibility has been offered to explain the co-electrophoresis of Acu and BP chains observed in avian systems (23). However, in the experiments presented here, considerable care was taken to prevent such cleavage. Al- though limited plasmin cleavage cannot be conclusively ruled

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

4682

TABLE I Comparative amino acid composition of human and rat fibrinogen and their polypeptide chains

Data are expressed as residues per molecule.

Amino Acid Humana RatC Human” Humanb RatC

Lysine

Histidine Arginine Cm-cyskine

Aspartic Acid

Threonine Serine

Glutamic Acid Proline

G lycine Alanine

Half Cystine Valine

Methionine

lsoleucine

Leucine Tyrosine Phenylalanine

77

22 59 --

134

63

84

120 52

107 54

26 50

27

41 61

33 33

76+ 3 33 31

1%3 9 8

50+ 3 18 8

28+ 1 -- --

115+3 51 50 56+ 2 22 21

78+ 7 22 22

lOE5 4.4 38 55f 6 18 11 94” 2 35 33 39?2 22 24 --me 8 6 41+2 18 16 26rt 1 7 7 36+ 3 18 22

55+ 2 24 24 34?1 17 16

21+2 14 16

31+3 214 204+ 8

10+3 58 53+ 7 13+2 126 136-t 4

15+2 86 82+ 2 56+ 3 342 345* 15

31+3 174 1762 8

3%3 234 2262 8 54+2 326 332+ 9

25+4 160 160 (d)

43+ 1 274 2585 14 24+ 2 126 123+ 6 ---- ---

1=1 108

17?1 86

18+2 108 2&2 162 20+ 2 108 1941 80

ibrinogen A -BB chains T Fibri ,gen y c in

RatC

!(Aa-B8 +y)

Whole F rinoaen ib -“_

Rate HumanC Human”

-----

105+ 7

72f 5

112+ 6

164+ 7 101+ 4

965 4

I

215?26 215 62f 4 59

132+14 155

85+ 5 ---

348+21 359

1935 8 176

25& 5 202 35E13 329

1502 5 139

28% 5 288 1532 11 143 ----- 72

12% 5 144

72+ 9 64

9pIll 117 170+ 8 175 103 4 101 104* 3 99

aTaken from McKee et al. ( 18 ) , sulfitolyzed fibrinogen.

bTaken from Mills and Liener ( 19 ), sulfitolyzed fibrinogen.

CCarboxymethylated fibrinogen. Molecular weights are: rat fibrinogen, 302,000; Au-B8 chains, 103,000; y chains,

d 47,500; human fibrinogen, 330,000. Proline residues for whole rat fibrinogen molecule are based on the sum of the residues from the individual chains.

FIG. 6. Immunological cross-reactivities of monospecific bovine, anti-bovine fibrinogen (A) versus bovine fibrinogen (I, 4), human human, and rat fibrinogen antibodies with interspecies plasma. A, plasma (2), rat plasma (3), bovine plasma (5), and ovalbumin (6). C, anti-human fibrinogen (x) versus human fibrinogen (1, 4), bovine anti-rat fibrinogen (A) uersus rat fibrinogen (I, 4), bovine plasma (2), plasma (2), rat plasma (3), human plasma (5), and ovalbumin (6). B, human plasma (3), rat plasma (5), and ovalbumin (6).

out, we consider it unlikely, since all fibrinogen samples purified from many unstimulated as well as stimulated rats demonstrate the same physical and chemical properties. In addition, those experiments in which t-aminocaproic acid was present throughout purification showed the same size and charge properties as the controls. Moreover, antibodies to purified rat fibrinogen behaved with complete immunological identity to the fibrinogen in fresh rat plasma.

A number of investigations have been directed toward NH,-terminal analyses of fibrinogen (24). One of the most consistent findings has been variations in the NH,-terminal residues of Aa and BP chains from species to species as well as

considerable heterogeneity of the A and B fibrinopeptides. Tyrosine is the NH,-terminal residue of the y chain in all vertebrates thus far examined with the exception of the two perissodactyls, tapir and rhinoceros, which have leucine and arginine end groups, respectively, 3 and the lamprey, which has a nonacetylated serine NH,-terminal (25). By inference, one would predict the y chain NH,-terminal of the rat to be either an acetylated serine or theonine, or a pyrollidone carboxylic acid. Although serine could represent a single base mutational change from tyrosine, both arginine and leucine in the two

3P. O’Neill and R. F. Doolittle, personal communication.

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

perissodactyls represent at least double base changes or a base deletion or addition. This suggests that the NH,-terminal region of the y chain is not as highly conserved as the thrombin cleavage site of Arg-Gly and theoretically makes the other end groups equally likely. Such a suggestion also has implications on possible modes of fibrinogen biosynthesis and assembly, since one theory (1) proposes the existence of a single polypep- tide chain, profibrinogen molecule, in which the conserved NH,-terminal of the y chain appeared to be indicative of a proteolytic cleavage site resulting in the free y chain polypep- tide.

In addition to the charge and size homogeneity observed in Acu-BP chains of rat fibrinogen, in contrast to bovine and hu- man Aa-BP chains, the comparative amino acid compositions of the y chains demonstrate that the rat fibrinogen y chain is involved in significantly more disulfide bonding than is its human and bovine counterpart. The methionine con- tents of the y chains also differ, as verified by cyanogen bromide cleavages, and clearly indicate that the rat fibrinogen molecule is distinctly different from those of other vertebrate species studied so far.

Antibodies directed against rat fibrinogen derived from three different rabbits all showed the same unique immunological patterns. The cross-reactivity results may be explained in part from the realization that the lagomorph and rodent are more closely related phylogenetically than is the lagomorph to either the primates or artiodactyls. Thus, to respond immunologi- cally to rat fibrinogen, the rabbit capitalizes on minor but unique differences which evidently either are not present or are concealed in the bovine molecule, but are present in the human molecule.

The information provided in this report characterizes the rat fibrinogen molecule in regard to a number of chemical features which appear to be unique. More recently, several laboratories have been directing their attention toward understanding how fibrinogen is assembled and how its biosynthesis is controlled. The laboratory rat represents a useful model for studying how fibrinogen synthesis and assembly are regulated. The identifi- cation of these unique molecular features as well as the ability to induce fibrinogen synthesis should provide important infor- mation for those interested in fibrinogen assembly and biosyn- thesis using this animal model.

4683

Aclznowoledgments-We thank our colleagues Dr. Alex Ku- rosky and Mr. Billy Touchstone for expert help in the sequence analyses. We also appreciate the skillful secretarial help of Miss Shirley Sidoti.

1. 2. 3. 4.

5. 6.

7. 8.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Doolittle, R. F. (1973) Adu. Protein Chem. 27, l-109 Mihalyi, E. (1970) Thromb. Diath. Huemorrh. (suppl) 39, 43-61 Doolittle, R. F. (1970) Thromb. Diath. Haemorh. (suppl) 39,25-42 Weimer, H. E., and Humelbaugh, C. (1967) Can. J. Physiol.

Pharmacol. 45, 241-247 Deutsch, D. G., and Mertz, E. T. (1970) Science 170, 1095-1096 McDonaeh. R. P.. McDonash. J.. and Blomback. B. (1972) Proc.

N&l. &ad. SC;. ill. S. A.-69, 3648-3652 Weber, K., and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 Mills, D., and Karpatkin, S. (1970) Biochem. Biophys. Res.

Commun. 40, 206-211 Takagi, T., and Doolittle, R. F. (1974) Biochemistry 13, 750-756 Hermodson, M. A., Ericsson, L. H., Titani, K., Neurath, H., and

Walsh, K. A. (1972) Biochemistry 11, 4493-4501 Gollwitzer. R.. Timol. R.. Becker. U.. and Fruthmaver. H. (1972)

Eur. J. kochem.‘28, 447-506 ’ “,

Cuatrecasas, P., Wilcheck, M., and Anfinsen, C. B. (1968) PFOC.

Natl. Acad. Sci. U. S. A. 61, 636-643 Celander. D. R.. Naschke. M. D.. and Guest. M. M. (1961) Tex.

Rep. Biol. Med. 19, 50-b9 Neuhaus, O., Balagno, H. F., and Chandler, A. M. (1966) Am. J.

Physiol. 211, 151-156 Koj, A. (1970) in Energy Metabolism in Trauma (Porter, R. and

Knight, J., eds) pp. 79-88, J and A Churchill, London Segrest, J. P., Jackson, R. L., Andrews, E. P., and Marchesi, V. T.

(1971) Biochem. Biophys. Res. Commun. 44, 390-395 Blomblck, B., Blomblck, M., and Grondahl, N. J. (1965) Acta

Chem. Stand. 19, 1789-1790 McKee, P. A., Rogers, L. A., Marler, E., and Hill, R. L. (1966)

Arch. Biochem. Biophys. 116, 271-279 Mills, A., and Liener, I. (1969) Arch. Biochem. Biophys. 130,

629-635 Cartwright, T., and Kekwick, R. G. 0. (1971) Biochim. Biophys.

Acta 236.550-562 Mosesson, M. W., Alkjaersig, N., Sweet, B., and Sherry, S. (1967)

Biochemistry 6, 3279-3287 Pizzo, S. V., Schwartz, M. L., Hill, R. L., and McKee, P. A. (1973)

J. Biol. Chem. 248,4574-4583 Schwartz, M. L., Pizzo, S. V., Sullivan, J. B., Hill, R. L., and

McKee, P. A. (1973) Thromb. Diath. Haemorrh. 29, 313-338 Dayhoff, M. 0. (1972) in Atlas of Protein Sequence and Structure

Vol. 5, pp. D-87-97, National Biomedical Research Founda- tion, Silver Spring, Md.

25. Doolittle, R. F. (1965) Biochem. J. 94,735-741

REFERENCES

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from

H Bouma, 3rd and F M FullerPartial chemical characterization of rat fibrinogen.

1975, 250:4678-4683.J. Biol. Chem. 

  http://www.jbc.org/content/250/12/4678Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/250/12/4678.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 8, 2019

http://ww

w.jbc.org/

Dow

nloaded from