Preparation and Properties of Derivatives of Bovine Factor X and ...

5 downloads 195 Views 5MB Size Report
Limited proteolysis of bovine blood coagulation Fac- tor X by chymotrypsin produces a derivative in which the light chain is cleaved between Tyr 44 and Lys 45.
Vol. 261, No. 9, Issue of March 25, pp. 4015-4023.1986 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1986 by The American Society of Biological Chemists, Inc.

Preparation and Properties of Derivatives of Bovine Factor X and Factor Xa from Which the y-Carboxyglutamic Acid Containing Domain Has been Removed* (Received for publication, September 30,1985)

Takashi MoritaS and Craig M. Jackson9 From the Department of Biological Chemistry. Divisionof Biology and Biomedical Science, Washington University School of Medicine, St. Louis, Missouri 63110

tion process is directly related to the Gla-dependent Ca2+Limited proteolysis of bovine blood coagulation Factor X by chymotrypsin produces derivative a in which mediated binding of these proteinsto negatively charged lipids the light chainis cleaved between Tyr44 and Lys45. in cell membranes (1-4). The failure to have such binding Two peptide products, residues 1-44 of the Factor X and theassociated enhancement of the rate of conversion of light chain and a modified zymogen, Factor X(-GD) the zymogens to active enzymes in proteins in which Gla have been isolated and characterized by sodium dode-formation has been blocked by vitamin K antagonists accy1 sulfate-polyacrylamidegel electrophoresis, elution counts for the anticoagulant effect of these d r u g s (4). In behavior on anion-exchange chromatography, amino addition to elimination of binding to phospholipids, when Gla acid composition, and by partial amino acid sequenceis absent from Factor X, the rate of Factor X activation by determination. Factor X(-GD) no longer contains the the enzyme from Viperu russellii venom is also markedly 12 y-carboxyglutamic acid residues of native the zymogen and thus serves as a model for investigation of diminished (5). This particular property has been exploited the properties conferred on FactorX by the presence for routine clinical assay during oral anticoagulant therapy of y-carboxyglutamic acid. Cleavage of Factor X at (6-8). The difficulty in obtaining acarboxy Factor X in suffiTyr 44 by chymotrypsinis inhibited byCa2+ and Mg2+cient quantities for chemical investigations led us to develop a proteolytically modified derivative of Factor X which could ions.FactorX(-GD) is activated by the coagulation factor activator of Vipera russellii venom, but at less serve as a model for further study of the role of Gla in the than 1%of the rate of activation of native Factor X. properties of this vitamin K-dependent protein. Specific seThe susceptibilityof Tyr 44 to chymotryptic cleavage lection of chymotrypsin as theprotease for Factor X modifiimplies that this residue is on the surface of the light cation was suggested from the work of Milstone et ul. (9) who chain of FactorX. observed that chxmotryptically modified Factor X still posFactor Xa(-GD)is indistinguishable from native Fac- sessed some, but not all of the functional properties of the tor Xa in its activity on Benzoyl-Ile-Glu-Gly-Arg-p-intact, native molecule. nitroanilide, on prothrombinalone, and on prothrombin plus Factor Va. In the presenceof phospholipid the MATERIALS AND METHODS~ rate ofprothrombin activation catalyzedbyFactor V. russellii venom was purchased from the Miami Serpentarium Xa(-GD) is the sameas in the absence of phospholipid.

and the Factor X- and Factor V-activating enzymes purified as described previously (18,191. QAE, G-25, G-50,and G-100 Sephadex and Sepharose 4B were from Pharmacia. Sodium b~ro[~H]hydride, 25 mCi, 422 mCi/mmol, was from the Radiochemical Center, AmerSince the initial discovery of y-carboxyglutamic acid in sham. [‘*C]Iodoacetamide, 10 mCi/mmol, was from New England prothrombin and other vitamin K-related proteins, investi- Nuclear. N“-p-Tosyl-L-lysine chloromethyl ketone-treated a-chymogation of the effects of this modified amino acid on the trypsin, dithiothreitol, sodium lauryl sulfate, soybean trypsin inhibipolyethylene glycol 20,000, N-acetylneuroaminic acid, iodoacetproperties of the proteins that contain it have been extensive. tor, amide, and acrylamide were from Sigma. Iodoacetamide was recrysReferences to the original reports can be found in Refs. 1-3. tallized from water before use. Heparin-Sepharose from Na heparin The principal function of Glal and Ca2+ions in the coagula- (160 USP units/mg Diosynth) and soybean trypsin inhibitor-Sepharose were prepared by coupling to cyanogen bromide-activated Seph* Preliminary accounts of this work have been reported at the 8th arose 4B (20). Guanidinium chloride, ultrapure, was from Heico. Bz(S-2222) and Bz-Ile-Glu(piperidine Steenbock Symposium, University of Wisconsin, Madison, WI, June Ile-Glu(-H/OMe)-Gly-Arg-pNa (S-2337) were from AB Kabi Peptide 10-13, 1979, and at the International Workshop on Regulation of amide)-Gly-Arg-p-nitroanilide Coagulation, University of Oklahoma, Norman, OK, September 4-8, 1979. The costs of publication of this article were defrayed in partby from which residues 1-44 of the light chain have been removed by the payment of page charges. This article must therefore be hereby chymotryptic protiolysis. Gla peptide is residue 1-44; NaDodSOr, marked “advertisement” in accordance with 18 U.S.C. Section 1734 sodium dodecyl sulfate; iPrzPF, diisopropylphosphofluoridate;iPrzP, solely to indicate this fact. the diisopropylphosphoryl group; pNa, p-nitroanilide. $Current address: Department of Biology, Faculty of Science, Portions of this paper (including part of “Materials and Methods,” Kyushu University, Fukuoka 812, Japan. part of “Discussion,” additional Tables I-IV, and additional Figs. 1To whom correspondence should be addressed American Red 3) are presented in miniprint at the end of this paper. Miniprint is Cross, Blood Services Research Laboratory, 100 Mack Ave., Box 351, easily read with the aid of a standard magnifying glass. Full size Detroit, MI 48232. photocopies.are available from the Journal of Biological Chemistry, The abbreviations used are: Gla, y-carboxyglutamic acid; RVV- 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. X-CP, the Factor X-, Factor IX-, and Protein C-activating enzyme 85M-3279, cite the authors, and include a check or money order for from the venom of V. russellii; HEPES, 4-(2-hydroxyethyl)-Z-pipera- $4.40 per set of photocopies. Full size photocopies are also included zineethanesulfonic acid; Bz, benzoyl; Boc, t-butoxycarbonyl; Factor in themicrofilm edition of the Journal that is available from Waverly X(-GD) and Factor Xa(-GD), Factor X and Factor Xa, respectively, Press.

4015

4016

Factor and X(-GO)Xa(-GD) Factor

Research, Molndal, Sweden (6). Boc-Val-Leu-Gly-Arg-pNawas a gift from Professor Sadaaki Iwanaga, Kyushu University, Fukuoka, Jawas from ICN Pharmaceupan. p-Nitrophenyl-p’-guanidinobenzoate ticals. Polyethylene glycol 6,000 was from J. T. Baker Chemical Co. Common organic and inorganic reagents and buffer salts were of analytical reagent grade or of the highest quality commercially available. Fibrinogen was prepared from bovine plasma by the procedure of Straughn and Wagner (21). Phospholipids and single bilayer vesicles were prepared as described previously (22). Isolation of Factor X and Separation of Factor X from Protein ZBovine Factor X was purified as described previously (10) and separated from Protein Z (11)on heparin-Sepharose (12). The Factor X was homogeneous on polyacrylamide gel electrophoresis in the absence (13) and in the presence of NaDodS04 (14). The specific activity in the one-stage clotting assay (15) was greater than 85 units/mgand after complete activation (16), 5000 f 500 units/mg relative to the Factor Xa reference preparation, ICTH PRP, Lot B (17). Reduction and Carboxyamidomethylation of Factor Xa(-GD)-BFactor Xa(-GD) (2.0 ml, 2.3 mg/ml) dissolved in 0.5 M Tris-HC1,6 M guanidine hydrochloride, 2 mM EDTA, pH 8.1, was purged with N2 to remove dissolved 0 2 . Disulfide reduction was obtained by addition of dithiothreitol to give a final concentration of 20 mM followed by incubation a t 50 “C for 3 h. Free sulfhydryl groups were reacted with [“C]iodoacetamide (28.9 pci), final concentration 150 pM in 0.5 N NaOH followed by nonradioactive iodoacetamide (120 pmol) in 0.5 N NaOH for completion of carboxyamidomethylation. Purification and Actiuation of Bovine Factor V-Factor V was prepared from bovine plasma by the procedure of Esmon (19) as modified in this laboratory. Modifications include substitution of DEAE-Sephacel for QAE-Sephadex, elimination of the calcium oxalate adsorption step, and inclusion of heparin-agarose chromatography after separationof the Factor Vfrom adsorbed plasma on DEAESephacel. The Factor V prepared by this method was homogeneous by NaDodS0,-polyacrylamide gel electrophoresis (5% acrylamide) using the Weber and Osborn (14) system and could be quantitatively converted from the precursor to theactivated form by either thombin or the activating enzyme or V. russellii venom (19). Factor V was activated using the V. russellii venom activator VCP by incubation of Factor V (0.1 mg/ml) with V-CP (0.01 mg/ml) for 20 min at 37 “cin 0.05 M Tris-HC1,O.l M NaCI, 5 mM CaC12,pH 7.5. Tritium Labeling of Factor X-Factor X was radiolabeled with tritium as described by Van Lenten and Ashwell (23) and modified by Silverberg et al. (24). The specific radioactivity of Factor X, (2.89 X lo6 cpm/Am) was indistinguishable from that of Factor X2 (2.85 X lo6 cpm/Am). Amino Acid Anulyses-Amino acid analyses were performed on 6 N HCI hydrolysates on a Durrum D-500 amino acid analyzer using ninhydrin detection (25). Amino Acid Sequence Determinution-Liquid-phase analyses were done in a Beckman 890 C Sequencer with the use of Polybrene and a 0.33 M Quadrol program (26). Liberated phenylthiohydantoins were identified by thin-layer chromatography on fluorescent-coated silica gel plates, developed in ethylene dichloride/glacial acetic acid (307, v/v) (27). Actiuity Assay-Factor X clotting activity was determined by the procedure of Bachmann et al. (15) using pooled normal plasma as a reference and defining the activity in 1 ml of this plasma as 1 unit. Activated Factor X was assayed using the same procedure, except that Russell’s viper venom was omitted from the reagents. One unit of Factor Xa activity was defined by the reference preparation, ICTH PRP, Lot B (17) available from the National Institute for Biological Standards and Control, Holly Hill, London. Thrombin activity was assayed as described by Fenton and Fasco (28) using as reference thrombin Lot J , U. S. Reference Thrombin provided by Dr. David Aronson, Bureau of Biologics, U. S. Food and Drug Administration, Bethesda, MD. Determination of Kinetic Parameters for Factor Xu and Xa(-GD) Hydrolysis of Peptidep-Nitroanilides-Active-site titrations of Factor Xa and Xa(-GD) were made using a Cary 219 spectrophotometer and a modification of the procedure of Chase and Shaw (29) in which 0.1 M NaCl, 0.1 M HEPES, pH 8.3, replaced the veronal buffer in the original procedure. For the peptide pNa hydrolysis rate measurements (30), stock enzyme solutions were diluted in 0.5 M NaCI, 0.01 M HEPES, 0.01 M Tris-HC1, 0.1% polyethylene glycol 6000, pH 7.8, into polypropylene titration vessels (Radiometer) which had been rinsed with 0.1%

-

I

-

E

8 E

2 -50 3 -

N

x

x

-0 MIN

FIG. 1. Time course of limited proteolysis of bovine Factor X by chymotrypsin. Bovine Factor XI (1.66 mg/ml) or Factor X2 (1.25 mg/ml) were incubated with a-chymotrypsin (2.6 pg/ml with X, and 2 pg/ml with X,) a t 22 “C in 0.05 M Tris-HC1, 0.1 M NaCl a t pH 7.5 for 50 min. NaDodS04-polyacrylamide electrophoresis gels (8%polyacrylamide) shown in the inset are from left to right: 1, the starting Factor X 2, after 15 min; and 3, after 45 min. Disulfide bonds were reduced prior to electrophoresis. One-stage clotting activity as a percentage of the initial value; 0, X,; 0, X2. polyethylene glycol 20,000 and oven-dried to eliminate adsorption of the enzyme to the surface. Reaction velocities were determined after adding an aliquot of the diluted enzyme to a thermostatted polystyrene cuvette (Walter Sarstedt, Princeton, NJ) which contained a Teflon-coated magnetic stirrer and the substrate solution in 0.1 M NaCl, 0.01 M HEPES, 0.01 M Tris-HC1, 0.1% polyethylene glycol 6000, pH 7.8. Therate of peptide p-nitroanilide hydrolysis was determined from the change in absorbance at 405nm using an extinction coefficient for p-nitroaniline of9920 M” cm” in this reaction buffer (31). Absorbance data were transferred directly from the spectrophotometer to a PDP 11/34a computer using a program developed in this laboratory. Initial velocities were estimated using a program3 based on application of the direct linear plot procedure to the integrated Michaelis-Menten equation (32). K, and V, were estimated using a computer-based version (33) of the direct linear plot method (34). Electrophoresis-Electrophoresis in polyacrylamide gels employed the procedures of Weber and Osborn (14) in the presence of NaDodSO, and of Davis (13) for disc electrophoresis under nondenaturing conditions. Specific acrylamide concentrations are given in the legends to the figures in which gel photographs are shown. Extinction Coefficients-Valuesforemployedwere: FactorXI and Xp, 12.4 (16, 35); Xa, 10.0; and X(-GD) andXa(-GD), 10.0. Radioactiuity Determinution-Concentrations of “C- and 3H-labeled protein were determined by liquid scintillation counting in a Packard Model 2425 counter at an efficiency of 64% and 30-33%, respectively. The liquid scintillation mixture, consisting of toluene (3 volumes), Triton X-100 (3 volumes), 1,4-bis[2-(5-phenyloxazolyl)] benzene (50 mg/liter), and 2,5-diphenyloxazole (5 g/liter) was from Research Products International, Mount Prospect, IL. RESULTS

Proteolytic Modification of Factor X by ChymotrypsinFactor X, in 0.05 M Tris-HC1, 0.1 M NaC1, pH 7.5, was incubated with chymotrypsin at 22 “C for 50 min. Chymotrypsin (previously freed of trypsinactivity by incubation with Tos-Lys-chloromethylketone) (36), was used at a ratio to Factor X of 1-700 (w/w). Disappearance of clotting activity was monitored by the Factor X one-stage clotting assay and proteolysis by NaDodS04-polyacrylamide gel electrophoresis. The time course for the disappearance of clotting activityfor Factors XI and X2 are independently shown in Fig. 1. NaDodS04-polyacrylamide electrophoresis gels are shown for Factor X (X, and X,) for three time points in the inset to Fig. 1. No differences were observed in thetime courses for either form of Factor X, either by clotting assay or by electrophoresis. C. M. Jackson and T. L. Carlisle, unpublished data.

Factor X(-GO) and Factor Xa(-GD)

4017

to be in the "breakthrough" volume of the column (Fractions 15-25). Four major and three minor products, in addition to chymotrypsin were eluted from the column. Tentative identification of the products in each of the peaks was madeusing NaDodS04-polyacrylamide gel electrophoresis and definitive identification from amino acid compositions and other properties. The second peak from the QAE-Sephadex column (Fractions 26-40) contains derivatives of Factor X with a completely modified light chain and with a limited cleavage in the heavy chain. The next two peaks contain exclusively Factor X,(-GD) (Fractions 43-55) and Factor XZ(-GD) (Fractions 56-65). The last peak (Fractions 75-85) contains the B tl "Gla peptide" (residues 1-44, light chain). A small trailing shoulder (Fractions 86-93), was observed following the Gla peptide and was subsequently shown to be a further cleaved form of the Gla peptide (residues 1-41). Traces of Factor X that by clotting assay were indistinguishable from the native zymogen prior to incubation were found in Fractions 65-75, Fig. 2B, and Fractions 74-86, Fig. 2C. These arethe expected elution positions for native Factors X1 and Xz, respectively (12). Factor X clotting activity was undetectable in the two peaks of Factor X(-GD), Fractions43-65, as anticipatedfrom the results of the experiment shown in Fig. 1 and the known requirement for y-carboxyglutamic acid residues for expression of biological activity in such one-stage clotting assays FRACTION NUMBER (37). FIG. 2. Isolation of the products of limited chymotryptic In addition to theelectrophoretic migration positions relaproteolysis of Factor X. A, bovine Factor X1 and Xz (94 mg, 3.1 tive to those of the native Factor X chains andthe molecular mg/ml) in 0.05 M Tris-HC1,O.l M NaCl, pH 7.5, were incubated with weight estimates from NaDodS04-polyacrylamidegel electrochymotrypsin(135 pg) at 22 "C. After30 min, 0.3 ml of 10 mM iPrzPF was added and inactivation of chymotrypsin allowedto occur for 20 phoresis, amino acid compositions were determined to unmin. Column chromatography wason QAE-Sephadex A-50 (1.4 X 36 ambiguously identify the major products, X(-GD) andthe Gla cm) in 0.05 M Tris-HC1,O.lM NaC1, pH 7.5, at 4 "C. A linear gradient peptide. Amino acid compositions and the expected values from 0.1 to 0.45 M NaCl in 0.05 M Tris-HC1, pH7.5, 250 ml/chamber based on the amino acid sequence of Factor X1 are given in at a flow rate of 15 ml/h was used to elute the proteolysis products. Table I of the Supplementary Material. The identity of the Fraction volumeswere 6 ml. 0,absorbance at 280 nm; X, conductivity. B, bovine Factor X, (16.6 mg, 1.66 mg/ml) in 0.05 M Tris-HC1, 0.1 M last peak, excluding the trailing shoulder, was confirmed by NaCl, pH 7.5, was incubated with chymotrypsin (25 pg) at 22 "Cfor determination of its amino-terminal sequence, Ala-Asn-Ser, identical to the known sequence of the light chain of Factor 30 min. iPrzPF treatmentandcolumnchromatographyonQAESephadex A-50 was the same as in A. 0, absorbance at 280 nm; 0, x1 (38). one-stage clotting activity; X, conductivity. C, bovine factor X, (12.5 Cleavage of Factor X(-GD) by the Protease of V. russellii mg, 1.25 mg/ml)in 0.05 M Tris-HC1, 0.1 M NaC1, pH 7.5, was Venom-Cleavageof FactorX(-GD) to form the product incubated with chymotrypsin (20 pg) at 22 "C for30 min. iPrzPF treatment and column chromatographyon QAE-Sephadex A-50 was corresponding to Factor Xawas investigated using the enzyme the same as in A. 0, absorbance at 280 nm; 0, one-stage clotting of V. russellii venom (RVV-X-CP). Because Ca2+ ions are activity; x, conductivity. required for the activation of native Factor X and acarboxy Factor X (5), CaClzwas included inthe reaction buffer. Analysis of the reaction products by NaDodS04-polyacrylExamination of the NaDodSO4-polyacry1amideelectropho- amide gel electrophoresis indicated that Factor X(-GD) was resis gels indicated that thelight chain of Factor X had been cleaved by RVV-X-CP to yield to the two predicted heavy quantitatively modified with a change in M, from 15,000 to chain products, i.e. the heavy chain from which only the 10,000. A minor component with mobility slightly greater activation peptide had been cleaved and theheavy chain from than that of the native heavy chain was also observed, sug- which both the activation peptide and the carboxyl-terminal gesting limited modification of the heavy chain as well. This peptide, Factor X (Fragment 4) (39, 40) had been cleaved. component was estimated to comprise less than 10% of the These products were indistinguishable by NaDodS0,-polytotal heavy chain mass on the basis of relative intensity of acrylamide gel electrophoresis from those produced from nathe stainedbands. Incubation of Factor X with chymotrypsin tive Factor X, Fig. 3A. No cleavage was observed if EDTA at higher temperatures, i.e. 37-40 "C, or for prolonged times was present in the reaction buffer. produced several additional products. Development of enzymatic activity concommitant with proPreparation and Characterization of Factor X(-GD),-The teolytic cleavage by RVV-X-CP was monitored by following products of chymotrypsin cleavage of Factor X (a mixture of the production of Boc-Leu-Gly-Arg-pNa hydrolase activity, X1 and XZ),Fig. 2 A , Factor X1 alone, Fig. 2B, and Factor Xz Fig. 3B. The final product was found to have a specific activity alone, Fig. 2C, were isolated by chromatography on QAE- nearly equal to thatof Factor Xa produced from native Factor Sephadex A-50. Chymotrypsin was inactivated with iPrzPF X. Similar to the activation of acarboxy Factor X (5), actiprior to chromatography and its elution position established vation of Factor X(-GD)occurs much more slowly than native Description of the products of proteolytic modification of Factor Factor X. Preliminary estimatesindicate that Factor X(-GD) X is greatlysimplified by using theirfinaldesignations prior to is activated at less than 1%the rate of native Factor X. Isolation of Factor Xu(-GD)-Activated Factor X(-GD) was presentation of the evidencefor such designations andthus has been done throughoutthis paper. produced from Factor X(-GD)(the mixture of Factors XI and

Factor X(-GO) and Factor Xu(-GD)

4018

TABLEI Comparison of Factor Xu and Factor Xaf-GD) hydrolysisof peptide p-nitroanilides Values in parentheses below the median estimate for each kinetic parameter are the 68% confidence intervals (34). Factor Xa(-CD)

Factor Xa Substrate Km

kcat s-l

Bz-Ile-Glu-Gly-Arg-pNan 146.0

189.0

175.0 (166.0-223.0)

I

(135.0-143.0)

(142.0-151.0) (89.1-102.0) (PiDb) Solution composition: 0.01 M Tris-HCI, 0.01 M HEPES, 0.1 M NaCl, pH 7.8, 25.0 "C. Pip, piperidine amide in the peptide substrate Bz-Ile-Glu(Pip)-Gly-Arg-pNa.

(151.0-164.0)

86.7 (83.3-93.2)

I

B

A

L

1

2

3

kcat s-'

(181.0-197.0)

(H OMe) 145.0 Bz-Ile-Glu-Gly-Arg-pNa"

96.5

Km

PM

PM

"

4

4

min

FIG.3. Activation of Factor X and Factor X(-GD) by RVVX-CP activator. Native Factor X (0.36 mg/ml) (0)or Factor X (-GD) (0.36 mg/ml) (0)in 0.05 M Tris-HC1, pH 8.0, was activated a t 22 "C with the RVV-X-CP venom activating enzyme (1.9 pg/ml) in the presence of 5 mM CaC12. Factor X(-GD) (A) was incubated with the RVV-X-CP in the presence of 0.5 mM EDTA. NaDodS04-polyacrylamide gels are (from left to right): 1, Factor X, 2, Factor X (-GD);3, the activation mixtureof native Xand X-CP(after 10 rnin); and 4, the activation mixture of X(-GD) and X-CP after 120 min. Disulfides were reduced using P-mercaptoethanol.

X2), as described above and the activatedFactor X(-GD) separated from the two activation peptides (12) by chromatography on QAE-Sephadex A-50, Fig. 4A. Factor Xa(-GD) was identified from the peptide-pNa hydrolase activity and by NaDodS0,-polyacrylamide gel electrophoresis. The activated Factor Xa(-GD) coeluted with both RVV-X-CP and Factor X(Fragment 4) because of the diminished net negative charge on the Factor X(-GD) thatresults from removal of the Gla peptide. Separation of Factor Xa(-GD) from RVV-X-CP and Fragment 4 was achieved by gel filtration on Sephadex G-100,Fig.4B. The RVV-X-CP was eluted near the void volume and Fragment 4 inthe internalvolume, well separated from Factor Xa(-GD).RVV-X-CP was identified onthe basis of its known elution position from previous work (12), and Fragment 4from its amino acid composition (data notshown). Comparison of the Activities of Factor Xu and Factor Xu(-GD)-Factor Xa interacts with 3 macromolecules and Ca2+ions during prothrombin activation (see Refs. 1 and 3). Of these interactions, those involving Ca2+and phospholipid are predicted to be altered orabsent inFactor Xa(-GD) because of the known involvement of Gla in these interactions (4,37,44,45).However, interaction with low molecular weight peptide substrates is suggested to be similar, if not identical from the data of Fig. 3. Each of the interactions that can be separately investigated, viz. number of active sites/molecule, peptide pNa hydrolase activity, prothrombin proteolytic ac-

(138.0-173.0)

tivity, enhancement of prothrombin proteolytic activity by Factor Va, and enhancement of prothrombin proteolytic activity inthe presence of Ca2+and phospholipid were examined. This was done by making specific comparison between Factor Xa and Factor Xa(-GD), similarly to previous studies on the activation of prothrombin and itsproteolytic derivatives (4649). Active Protease Concentration Determination by fHJiPr2PF Incorporation-Comparison of the number of active sites/ Factor Xa andFactor Xa(-GD) was made by determining the relative stoichiometry of incorporation of [3H]iPr2PFinto the active enzymes. A mixture (1.25 Azso/ml, 34 mg) containing equal amounts of Factor Xa and Factor Xa(-GD) in 0.05 M Tris-HC1,O.l M NaC1, pH 8.0, was incubated with [3H]iPr2PF (9 p ~ at)22 "C for 30 min and then at37 "C for 15 min. After this preincubation with [3H]iPr2PF, 1M iPr2PFwas added to give a final concentration of 10 mM and incubated at 37 "C for 90 min. The one-stage clotting activity of active Factor Xa after thisperiod indicated that 87% of the native Factor Xa had been inactivated. The remaining active Factor Xa and Factor Xa(-GD) wereremoved from iPr2P-Factor Xa and iPr2P-FactorXa(-GD) on acolumn (1.5 x 5.5 cm) of soybean trypsin inhibitor-Sepharose (47-49). iPr2P-FactorXaand iPr2P-Factor Xa(-GD)from the breakthrough fractions were then separated on a column (0.9 x 29 cm) of QAE-Sephadex A-50. Each protein was subsequently gel filtered on acolumn (2.5 X 84 cm) of Sephadex G-100. The specific radioactivities of the two active site-blocked species were nearly the same, 5.46 X lo9 cpm/mmol for Factor Xa and 5.15 X lo9 cpm/ mmol for Factor Xa(-GD). Given the assumed E;:,,,of 10.0 for the extinction coefficients for these two proteins, these values are very similar. If a correction to the extinction coefficients is made that takes intoaccount the relative number of Trp residues in these two proteases, 5 in @-FactorXa and 4 in 6-Factor Xa(-GD), then theFactor Xa(-GD)used in this experiment is calculated to be 80%as reactive with iPrzPF as Factor Xa. Hydrolysis of Bz-Ile-Glu(X)-Gly-Arg-pNa Substrates by Factor Xu and Factor Xu(-GD)-Factor Xa andFactor Xa(-GD) were compared with respect to their ability to hydrolyze two tetrapeptide pNa substrates, Bz-Ile-Glu(50% 0Me)-Gly-ArgpNa and Bz-Ile-Glu(piperidine amide)-Gly-Arg-pNa. These substrates possess the amino acid sequence immediately preceding the two peptides bondsin prothrombin that arecleaved by Factor Xa (50). Active protease concentrationswere determined by p-nitrophenyl-p'-guanidinobenzoate titration (29) as described under "Materialsand Methods." Initial velocities were determined as afunction of substrate concentration between 15 and 250 p ~ The . kinetic parameters were calcu-

Factor X(-GO) andFactor Xa(-GD)

4019

~~

FRACTION NUMBER

FRACTION NUMBER

FIG. 4. A, separation of Xa(-GD) from the activation peptides on QAE A-50. Factor X(-GD) from a mixture of X, and Xz (74 mg, 2.3 mg/ml) in 0.05 M Tris-HC1,0.15 M NaC1, pH 7.5, 5 mM CaClZwas incubatid with 1.6 mg of RVV-X-CP for 2 h a t 37 "C. The activation mixture was applied to 1.4 X 36-cm column of QAE-Sephadex A-50 equilibrated in 0.05 M Tris-HC1, 0.15 NaC1, pH 7.5. The activation products were eluted with a linear gradient in NaCl, 0.15 to 0.45 M, 250 ml/chamber a t a flow rate of 20 ml/h. Fraction volumes were 7.1 ml. 0, absorbance at 280 nm; 0, Boc-Leu-Gly-Arg-pNa hydrolase actiirity; X , conductivity. The arrow indicates the point of beginning the elution gradient. B , separation of Xa(-GD) from Factor X (Fragment 4) on Sephadex G-100. Fractions 9-24 from QAE-Sephadex A-50 were pooled, lyophilized, and then dissolved in 5 ml of water. This sample was applied to a column (3 X 87 cm) of Sephadex G-100. Elution was performed a t 4 "C with 0.05 M Tris-HC1,O.l M NaC1, pH 7.5, at a flow rate of 19 ml/h. Fraction volumes were 6.2 ml. 0, absorbance at 280 nm; 0, Boc-Leu-Gly-Arg-pNa hydrolase activity. Bars mark the fractiqns taken for subsequent composition determinations.

lated using the nonparametric method of Cornish-Bowden (32) as described under "Materials and Methods." No significant differences between Factor Xa and Factor Xa(-GD) were observed for either K,,, or kcat with either substrate (Table I) indicating that thetwo active proteases are indistinguishable in their peptide hydrolytic activities. Conversion of Prothrombin to Thrombin-A comparison of the ability of FactorXaandFactorXa(-GD) to convert prothrombin to thrombin was investigated as described previously in studiesof prothrombin activationin thislaboratory (46-49). The reaction progress curves are shown in Fig. 5; solution composition and reactant concentrations are presented in the legends to thefigure. Factor Xa and Factor Xa (-GD) were indistinguishable in their ability to activate prothrombin in solution, Fig. 5A. When phospholipid was added to accelerate thrombin formation,no enhancement of the rate was observed with Factor Xa(-GD), inmarked contrast to the situation with native FactorXa. When Factor Va was added, but in the absence of phospholipid, again Factor Xa and Factor Xa(-GD) were indistinguishable in the rate at which they catalyzed thrombin formation, Fig. 5B. It should be noted that the Factor Xa and Factor Xa(-GD) concentrations employed in these experiments are adjusted to compensate for the 80% active Xa(-GD) concentration determinedby iPrzPF incorporation. Effects of Divalent Cations on Cleavage of Factor X by Chymotrypsin-An effect of divalent cations onthe cleavage of Factor X by chymotrypsin may be expected because of the known binding of divalent cations to the Gla residues of the vitamin K-dependent proteins (1-3). The dependence of the proteolysis of Factor X on the CaC12 concentration was investigated, Fig. 6. Maximal inhibition was observed at Ca2+ concentrationsin excess of 2 mM andtherateat these concentrations was found to be approximately 5% of the rate observed in theabsence of CaC12. The Ca2+concentration for half-maximal inhibition was found to be 0.6 mM, the same value observed for half-maximal saturation of bovine Factor X with Ca2+ions by direct binding studies (44, 45). Mg2+ also

FIG.5. Comparison of Xa and Xa-(GD) action on prothrombin. A, effect of phospholipid. Prothrombin (4.9 pM); in the presence or absence of phospholipid (equimolar dioleoyl phosphatidylcholine and phosphatidylglycerol) single bilayer vesicles (0.3 mM phosphate), 10 mM CaCl2, and either Xa (0.16 PM) or Xa(-GD) (0.20 pM) were incubated at room temperature, 22-23 "C in 0.05 M Tris-HC1, 0.1 M NaC1, pH 7.5. Samples were taken at the times shown on the ordinate and assayed for thrombin clotting activity. Specific mixtures are: Xa, CaClz (O), Xa, phospholipid, CaCl, (0),Xa(-GD), CaClz (A),and Xa(-GD), phospholipid, CaClz (A).B , effect of Factor Va. Prothrom, Va (0.03 p M ) , and 10 mM CaCl2,and either Xa bin (4.9 p ~ )Factor (0.08 p ~ or) Xa(-GD) (0.102 p ~ were ) incubated and the reaction mixtures sampled as in A. Specific mixtures are: Xa, CaC12 (O),Xa, Factor Va, CaClz (U),Xa(-GD), CaCh (A),and Xa(-GD),Factor Va, CaCl, (m).

inhibited proteolysis by chymotrypsin and was equally effective as Ca2+ at 1 mM, inhibiting the rate of proteolysis by approximately 60%. Stabilization of Factor X a by Ca2' during Storage-Loss of one-stage clotting activity without loss of either amino acid ester or peptide pNa hydrolase activity has been observed during long term storage of Factor Xa in solution at 4 "C. NaDodS04-polyacrylamide gel electrophoresis suggested that this loss in activity was the result of modification of the light

4020

Factor X(-GD) and FactorXu(-GD)

characterized and described in theSupplementary Material? Factor X(-GD) hasbeen compared with acarboxy Factor X with respect to activation by RVV-X-CP and therequirement for Ca2+for this reaction. Activation of Factor X(-GD) appears to be similar to apolipoprotein A-I (53) in thatthere is no requirement for added Ca"; in contrastto acarboxy Factor X (5). In the case of both acarboxy Factor X (5) and Factor X(-GD), however, no activation was observed in thepresence of EDTA. With the possible exception of the Ca" requirement, the qualitative characteristics of the activation process are similar insofar as they may be compared for both forms of Factor X, the native protein without Gla and the proteoFIG. 6. The inhibitory effect of calcium ions on the limited lytically generated derivative. proteolysis of Xa by a-chymotrypsin. Factor Xa (0.456 mg/ml) The activity of Factor Xa(-GD) has been compared with in 0.05 M Tris-HCI, 0.1 M NaCl, pH7.5, and CaClz concentrations as native Factor Xa in thehydrolysis of peptide pNa substrates given on the abscissa, was incubated with a-chymotrypsin (1 pg/ml) and prothrombin. The kinetic parameters for the hydrolysis at room temperature.Activity (X) by clotting activity assay expressed of the peptide pNa substrates are the same, and when phosas the change in the percentage of the initial value (2500 units/ml). pholipid is omitted from the prothrombin activation mixture, chain.5 Based on the inhibition of chymotryptic proteolysis the activity of Factor Xa(-GD) isindistinguishable from that by divalent cations, the effect of CaClz on the stability of of native Factor Xa in both the presence and absence of Factor Xa during storage at 4 "C was examined by the one- Factor Va. Although the limitation of the experiments with stage clotting assay. In the absence of CaCL, the activity of prothrombin as substrate to a single substrate concentration Factor Xa decreased to 20% of the initial level after 4.5 does not permit the conclusion that Factor Xa(-GD) and months in 0.05 M Tris-HC1, 0.1 M NaC1, pH 7.5, whereas a Factor Xa are quantitatively identical under all conditions, sample stored in the presence of 5 mM CaCL remained fully they are certainly similar. Quantitatively identical behavior active. Separation of the products generated during this long under all conditions cannot beexpected because of the reducterm storage in the absence of CaClz confirmed that inacti- tion in net negative charge that occurs upon conversion of vation as assessed in the clotting assay was the result of Factor X to Factor X(-GD) and is clearly reflected in the proteolytic cleavage of the "Gla domain" from the light chain chromatographic behavior on the anion exchanger, QAESephadex. The qualitative difference between Factor Xa and of activated FactorX. Factor Xa(-GD) in the presence of phospholipid is precisely as predicted from previous work on the relationship for proDISCUSSION thrombin between Gla,Ca2+ ion binding, and binding to Isolation of bovine Factor X that contains Glu rather than phospholipid (4, 44-49, 54). The observation that activation Gla residues (acarboxy Factor X) from cattle administered of prothrombin in the presence of Factor Va is indistinguishvitamin K antagonists has resulted in the demonstration of able whether the Gla domain is present or absent in Factor the same relationship among Ca" ion binding by Factor Xa, Xa indicates that in theabsence of phospholipid this portion Ca2+-mediatedbinding of Factor Xa to phospholipid bilayer of the light chain of Factor Xa is notdirectly involved in the vesicles, and phospholipid enhancement of the rate of pro- association of Factor Va and Factor Xa. thrombin activation (52) that hadbeen established earlier for Calcium and Mg2+ ion inhibition of the proteolysis of Factor prothrombin (46) and acarboxyprothrombin activation (4). In X by chymotrypsin and by a protease of unknown origin in addition to these observations, acarboxy Factor X was also stored preparations of Factor Xa suggest that net charge on observed to be activated much more slowly by RVV-X-CP the Factor X or Xa substratemarkedly influences the attack than native (Gla containing) Factor X ( 5 ) .The obvious prac- of proteases on peptide bonds in the vicinity of the Gla tical significance of this latter observation has already led to domain. This observation, although not yet understood in its exploitation in assay procedures for monitoring oral anti- terms of the actual process of proteolysis of the Factor X, coagulant therapy (6-8). The difflculty in obtainingacarboxy provides a useful method for stabilizing Factor X and Xa Factor X in sufficient quantity for more extensive investiga- during storage. tions of the function of Gla and the Gla domain of this The rapidity of the cleavage of the Tyr 44-Lys 45 peptide particular vitamin K-related protein led us to develop the bond in FactorX by chymotrypsin indicates that thispeptide procedure described here for preparing proteolytic derivatives, bond must be on the surface of the protein molecule and ina Factor X(-GD) and Factor Xa(-GD); derivatives that mighf state that is readily accessible to the protease. It is on the possess the same functional properties asacarboxy Factor X basis of these observations that the region of the Factor X and acarboxy Factor Xa. light chain consisting of residues 1-44 has been designated Limited proteolysis of Factor X by chymotrypsin under the Gla domain. Tyr residues are found at this position (44controlled conditions produces a derivative that is missing 45) in all of the other vitamin K-related proteins for which only the first 44 amino acid residues of the light chain of the sequence data areavailable, viz. prothrombin (50,631, Factor Factor X molecule in greater than 90%yield. This derivative IX (64), and Protein C (65) and, thus, similar chymotrypsin has been established to be exclusively this product by Na- susceptible regions are likely to exist. This has been demonDodS04-polyacrylamide gel electrophoresis, amino acid com- strated for prothrombin Fragment 1 (66) by us and since the position analysis, and partial amino acid sequence determi- preliminary report of this work for human prothrombin by nation. Because of possible effects of further cleaved deriva- others (67). In the case of Factor X and prothrombin Fragtives that could confound interpretation of experiments that employ these derivatives of Factor X and Factor Xa, the The descriptionof the characterization of all the minor products products of the further proteolysis have been isolated and is given later in "Characterization of Minor Products of Factor X C.-W. Peng and C. M. Jackson, unpublished observations.

Proteolytic Modification,"in the Supplementary Material. T. Morita andC. M. Jackson, unpublished observations.

'

4021

Factor X(-GO) a:nd Factor Xa(-GD) ment 1, it is interesting to speculate that the region around the accessible Tyr might act as a "hinge" and permit movement of the Gla domain of these molecules uponCa" binding in a way that perturbs thenearby Trp (50,.69)or in prothrombin, the Trp residues of the "kringle domain" (50) and, thus, account for the various spectral shifts observed byabsorption, circular dichroism, and fluorescence spectroscopy whenCa2+ or other divalent cations are bound (68-73). Acknowledgments-We wouldlike to thankItsuko Moritaand Julie A. Hall for their technical assistance, Dr. M. J. Lindhout for the gift of the Factor V, and Debra Hodak and Veronica Watkins for their assistance in preparation of the manuscript. The comments of Dr. Luis Glaser on this manuscript are also gratefully acknowledged. REFERENCES 1. Suttie, J. W., and Jackson, C. M. (1977) Physiol. Reu. 5 7 , l - 7 0 2. Stenflo, J., and Suttie, J. W. (1977) Annu. Reu. Biochem. 4 6 , 157-172 3. Jackson, C. M., and Nemerson, Y. (1980) Annu. Reu. Biochem. 4 9 , 765811 4. Esmon, C. T., Suttie, J. W., and Jackson,C. M. (1975) J. BioL Chem. 2 5 0 , 4095-4099 ..- - .. --

5. Lindhout, M. J., Kop-Klaassen, B.H.M., and Hemker, H. C. (1978) Biochim. Bjophys. Acta 533,327-341 6. Aurell, L., Fnberger, P., Karlsson, G., and Claeson, G. (1977) Thromb. Res. 11,595-609 7. vanWijk, E. M., Kahle, L. H., and tenCate, J. W. (1980) Clin. Chem. 2 6 , 885-890 8. Ciavarella, N., Coccheri, S., Gensini, G.F., Hassan H. J., Mannucci, P. M., Manotti, C., Margstakler, E., Mariani, G., Orlindo, M., Palareti, G., Petronelli, M., Pogliani, E., Ponari, O., Prisco, D., Reealcati, P., Rossi, E., Salvitti, C., and Tripodi, A. (1980) Thromb. Res. 19,493-502 9. Milstone, J. H., Oulianoff, H., Saxton, T. R., Milstone, V. K., and Brown, H. F. (1971) Yule J. Biol. Med. 43,223-235 10. Henriksen, R. A., and Jackson, C. M. (1975) Semin. Thromb. Hemostasis 1,284-309 11. Petersen, T. E., Th@gersen,H. C., Sottrup-Jensen, L., Magnusson, S., and Jornvall, H. (1980) FEBS Lett. 114,278-282 12. Morita, T. and Jackson, C. M. (1986) J. Bwl. Chem. 261,4008-4014 13. Davis, B. d. (1964) A n n N. Y. Acad. Sei. 121,404-427 14. Weber, K., and Osborn, M. (1969) J.Biol. Chem. 244,4406-4412 15. Bachmann, F., Ducked, F., and Koller, F. (1958) Thromb. Diath. Haemrrh. 2,24-38 16. Jackson, C. M. (1967) Ph.D. dissertation University of Washington 17. Jackson, C.M., Henriksen, R. A., Peig, C.-W., and Yin,E. T. (1976) Thromb. Diath. Haemrrh. 35,479-482 18. Esmon, C. T., and Jackson, C. M. (1973) Thromb. Res. 2,509-524 19. Esmon, C. T. (1974) Ph.D. dissertation, Washington University 20. March, S. C., Parikh, I., and Cuatrecasas, P. (1974) Anal. Biochem. 6 0 , 149-152 21. Straughn, W., and Wagner, R. H. (1966) Thromb. Diath. Haemorrh. 1 6 , 198-206 22. Dombrose, F. A., Gitel, S. N., Zawalich, K., and Jackson, C. M. (1979) J . Biol. Chem. 254,5027-5040 23. Van Lenten, L., and Ashwell, G. (1971) J. BioZ. Chem. 246,1889-1894 24. Silverberg, S. A., Nemerson, Y., and Zur, M. (1977) J. Biol. Chem. 2 5 2 , 8481-8488 ~ " . 25. Operation and Maintenance Manual Durrum Amino Acid Analyzer Model D-500 (1972) Durrum Instrument Co., Palo Alto, CA 26. Edman, P., and Henschen, A. (1975) in Protein Sequence Determination (Needleman, S. B., ed) pp. 232-279, Springer-Verlag, Heidelberg 27. Easley, C. W., Zegers, B.J. M., and DeVijlder, M. (1969) Biochim. Biophys. Acta 174,211-213 28. Fenton, J. W., 11, and Fasco, M. J. (1974) Thromb. Res. 4 , 809-817 29. Chase, T.,and Shaw, E. (1967) Biochem. Biophys. Res. Commun. 29,508. ~

51 A

30. Loitenberg, R., Hall, J. A., Blinder, M., Binder, E. P., and Jackson, C. M. (1983) Biochim. Biophys. Acta 742,539-557

31. Lottenberg, R., and Jackson, C.M. (1982) Biochim. Biophys. Acta 7 4 2 , 558-564 32. Cornish-Bowden, A. (1975) Biochem. J. 149,305-312 33. Henderson, P. J. F. (1978) in Techniques in Protein and Enzyme Biochemistry, B113, pp. 1-43, Elsevier/North-HollandBiomedical Press, Amsterdam 34. Cornish-Bowden, A., Porter, W. R., and Trager, W. F. (1978) J. Theor. Biol. 74,163-175 35. Jackson, C. M., Johnson, T. F., and Hanahan,D. J. (1968) Biochemistry 7 , 4492-4505 ""

""

36. Shaw, E. (1967) Methods EnzymoL 11,677-686 37. Lindhout, M. J., Kop-Klaassen, B. H. M., Kop, J. M. M., and Hemker, H. C. (1978) Biochlm. Bmphys. Acta 533,302-317 38. Enfield, D. L., Encsson, L. H., Walsh, K. A,, and Neurath, H. (1975) Proc. Natl. Acad. Scz. U. S.A. 72,16-19 39. Jesty J. Spencer A.K. Nakashima Y. Nemerson, Y., and Konigsberg, W.'(1675)J.Biol. ,Chek. 250,449;-4$04 40. Fujikawa K Titan1 K. and Davie,E.W. (1975) Proc. Natl. Acad. Sci. U. S. A: 72,3559-5369 41. Jackson, C. M. (1972) Biochemistry 11,4873-4882 42. Fujikawa, K.,Legaz,M.E., and Davie, E. W. (1972) Biochemistry 11, 4882-4891 43. Titani, K., hjikawa, K., Enfield, D. L., Ericsson L. H., Walsh, K., and Neurath, H. (1975) Proc. Natl. Acad. Sei. U. S. A. 72, 308Z73086 44. Henriksen, R. A., and Jackson, C. M. (1975) Arch. Biochem. Blophys. 170, 149-159 45. Lindhout; M. J., and Hemker, H. C. (1978) Biochim. Biophys. Acta 5 3 3 , 318-32fi ~~. 46. Gitel, S. N., Owen, W. G., Esmon, C. T., and Jackson, C. M. (1973) Proc. Natl. Acad. Scz. U. S. A. 70,1344-1348 47. Esmon, C. T., Owen, W. G., and Jackson, C. M. (1974) J. Biol. Chem. 249, "~

779R7807 ._".

48. Esmon, C. T., and Jackson, C. M. (1974) J. Biol. Chem. 249,7791-7797 49. Esmon, C . T., and Jackson, C. M. (1974) J. Biol. Chem. 249,7782-7790 50. Magnusson S. Petersen, T., Sottrup-Jensen L., and Claeys,H. (1975) Cold Sprr'ng HarborConf. Cell Proliferation 123-149 51. Mizuochi, T. Yamashita, K. Fujikawa, K., Titani, K., and Kobata, A. (1980) J . i w l . Chem. 2 5 5 , h526-3531 52. Lindhout, M. J. (1977) Ph.D. dissertation, Rijksuniversiteit Limburg, Maastricht Holland 53. Amphlett, G. W., Byme, R., and Castelliio, F. J. (1982) Biochemistry 21, 125-132 54. Bajaj, S. P., Butkowski, R. J., and Mann, K. G. (1975) J. Biol. Chem. 250, 2150-215fi 55. Lundblad, R. L. Noyes C. M Mann, K. G., and Kingdon, H. S. (1979) J. Biol. Chem. 2 k 4 , 8 5 k 3 5 2 ; 56. Jesty, J. (1979) J. Biol. Chem. 254,1044-1049 57. Mertens, K.,and Bertina, R. M. (1980) Biochem. J. 185,647-658 58. Dombrose, F. A., and Seegers, W. H. (1973) Thromb. Res. 3,737-743 59. Teng, C.-M., and Seegers, W. H. (1981) Thromb. Res. 22,213-220 60. Aronson, D. L., and Mustafa, A. J. (1971) Proc. Sac. Exp. Biol. Med. 137, 1262-1266 61. Paphadjopoulos, D., Yin, E. T., and Hanahan,D. J. (1964) Biochemistry 3, 1931-1939 62. Dahlback, B., and Stenflo, J. (1978) Biochemistry 17,4938-4945 63. Walz, D. A., Hewett-Emmett, D., and Seegers, W. H. (1977) Proc. Natl. Acad. Sci. U. S. A. 74,1969-1972 64. Katayama K. Ericsson, L. H., Enfield, D. L. Walsh K. A., Neurath H., Davie, 6. W., and Titani, K. (1979) Proc. katl. Ae'ad. Sci. U. S. A.'76, 4990-4994 65. Fernlund, P. Stenflo J. and Tufvesson, A. (1978) Proc. Natl. Acad. Sei. U:S. A. 76,5889-689; 66. Carhsle, T. L., Monta, T., and Jackson, C.M. (1980) in Vitamin K Meetobolsm and Vitamin K Dependent Proteins (Suttie, J. W., ed) pp. 58-61, University Park Press, Baltimore 67. Dode,C., Rabiet, M.-J., Bertrand, O., LaBie, D., and Elion, J. (1980) Biochem. Biophys. Res. Commun. 94,660-666 68. Bjork, I., and Stenflo, J. (1973) FEBS 32,343-346 69. Benavous, R., and Gacon, G. (1980) Biochim. Biophys. Acta 622,179-188 70. Prendergast F. G., and Mann, K. G. (1977) J. BWL Chem. 252 840-850 71. Brenckle, G:M., Peng, C. W., and Jackson, C:M. (1980) in bitamin K Meetobaiism and Vzlamzn K Dependent Proterns (Suttie, J. W.,ed) pp. 54-57, University Park Press, Baltimore 72. Jackson, C . M. (1980) in Vitamin K Metabolism and Vitamin K Dependent Protem (Suttie, J. W., ed) pp. 16-27, University Park Press, Baltimore 73. Esnouf, M. P., Israel, E. A., Pluck, N. D., and Williams, R. J. P. (1979) Thromb. Haemstasis 4 2 , 9 5 (abstr.)

i,

Lett.

Continued on next page.

Factor X(-GO) andFactor Xa(-GD) the Heavv

€4zppleaentary Haterial to Preparation and Properties of Bavine Factor X and Factor Xa from Which the y-CarbOxyglutamiC Acid Containing Domain Ifas Been R-ed Takashi Ilorita and Craig X. Jackson

A l l products of limited chymotryptic hydrolysis of Factor X have been identified from theiramino acid compositions as well as their chromatographic and NaCadSO electrophoretic behavior. Minor proteolysis products formed in additiont% the principal X(-GD) described in the main body of this manuscriDt were also characterized because of the DOsSihle future imDortance of this aerivative inkinetic studies of Factor Xa-action.

Factor x~(-GD) disulfide bonds were reduced using dithiothreitol under denat r ng conditions and the free sulfhydryl groups carbaxyamidomethylated with lac iodoacetamide to facilitate detection and identification of the separated chains. The two chains were separated by chromatography in 10% BOAC on sephadex 6-100, Figure 1. Homogeneity of the isolated chains was assessed by NaDodS04 polyacrylamide gel electrophoresis. Because of the large difference In cys content of the two polypeptide chains Of Factor X (41,42), definitive identification of the elution positions of the two chains from Factor x~(-GD) was obtained directly flthout further data. This identification was confirmed from the ratio of C radioactivity to absorbance at 280 nm. Figure 1 Separation of the Aeavy and Light Chains of Factor Xa(-GD)

Methods and Materials

After carboxyamidomethylation, as described in the text, the reaction mixture was applied to a column (1.5 x 91 cm) of Sephadex G-100 to separate each chain and excess reagent. Elution was performed '4 atC using 10% acetic acid. Fraction volupgs were 2.4 ml. The radioactivity ( C cpm) distribution in the first peak (heavy chain), second peak (light chain I-GDII . . . and third peak (not shown, excess reagent) were 3.6,6.9 and 89.5%. respetbively. Absorbance at 280 nm, ( 0 ) ; C Radioactivity. ( 0 ) (cpm).

XI-Gnl Factor X(-GD) and the Gla domain peptide isolated from the experiment shown in Figure 2A (main text) were hydrolyzed in 6N HC1 and their amino acid compositions determined (25), Table I. Based on the established residue specificity of chymotrypsin and the amino acid compcsition of the Gla domain peptide, particularly the single tyrosylresidue, this product was identified a s a r i s i n g f r a m r e s i d u e s l - 4 4 o f t h e Factor X light chain

Acid

Amino

X(-GD)

Gla peptide (1-44)

a

h

envnd 37.0 25.8 29.0 45.0 15.9 38.9 27.4

.mlud

37

4.1 1.3 3.5 12.7

30 28 45 18 40 27 22 24 5 12 26 9 18 12 21 23

e 24.1 4.9 6.2 24.4 8.2 16.9 10.9 19.2 21.0 f

d Calculated

C

Calcylated

Definitive identification of the two isolated chains and establishment of the sites of cleavage by RVV-X-CP in the heavy chain and CbymOtrypSin in the light chain were made by amino acid composition determination and analysis of the aminoterminal amino acidsequences. The compositions and the predicted oompcsitions from theknown amino acid sequence are given in Table 11. The amino acid sequence of the heavy chain of Factor Xal-GD) was (Ile~~Leu]-Val-Gly-Gly, aspredicted from the aminoterminal sequence of FaCtor Xa (43). and the sequence of the light chain, Lys-Asp-Gly, the Sequence for residues 45-47 of the light chain of Factor X (38). The Ile/Leu ambiguity at position 1 of the heavy chain is due to limited resolution during chromatographic identification of these apolar amino acids. AS no ambiguity in identification results, no further attempt to distinguish these residues was made.

4 1 3

14

0

0

1.9 4.0

1 4 2 2

e

1.6 0 0

Identification of the products present in the second peak eluted from the PAE Sephadex column of Figure ZA. (main text) and the shoulder on the trailing edge of the last peak required further investigation of the proteolysis of Factor x by chymotrypsin and determination of the amino acidcompositions of these products.

4.0 0.6 2.6 0.3 1.6 2.0 f

TO simplify identification of the products Present in the rjecond peak. these products were isolated from Factor X(-GD) that had been B labeled. These Factor XI-OD) derivatives were then incubated with RVV-X-CP and the . c i i ; a ; i & p r a d k t i isolated by gel filtration on sephadex G-100. Of the two a Determined by acid hydrolysis for 24 hours as described in Methods. products isolated the first possessed NaDodS04 polyacrylamide gel Number of residues calculatedon thebasis of 37Aspand45 Glu major residues r.ctroohnretic characteiistics indistinauishable from those of Factor x?.(" _ . " . c _ _ " . . .... per mol. GD) and was qat further characterized. +he second product, which contained most of the E label from the zymogen was bydrolyze$and its amino acid h Calculated fTOm the published sequence (38,431. composition determined, Table 111. On ;he basis of the H label which is in the sialic acid residues of the oligosaccharide chain linked to AS" 3 6 c Determined by acid hydrolysis of the Gla-peptide (fractions 80-86, (43.51) and its amino acid composition. this product wasconcluded to be Figure 2A.) as described in Methods. chain of FactorX. The principal derived from residues 34-51 of the heavy minor product present in the second peak of Figure 2 4 (main text) was d Calculated from the published sequence ( 3 8 ) . concluded tobe Factorx(-GD) from which the first 33 resrdues of the heavy chain had been cleaved. In separate experiments it was established that e present but not quantitatively determined. activation of this particular PKOteOlytiC derivativeof Factor X by RVV-X-CF occurs extremely slowly. even more slowly than observed fOKX(-GD) Factor f Present as determined from absorbance at 280 nm; not quantitated. suggesting that RVV-X-CP recognition of Factor X may be influenced hy both the aminoterminal portionof the heavy chain, as well as the aminoterminus of the light chain. 6

~

The amino acid composition of Factor Xa(-GD) from the chromatography illustrated in Figure 40 (main text) was determined a s described in the preceding paragraph for Factor X(-GD), and is given in Table 11. Comparison of the experimental data with the compositions predicted.from the amino acid sequence further substantiate the identification of the products that was given in the mainbody of the text.

~~

~~~~~

TABLE I11 Compositions of the "Activation Peptide" from Degraded 3H Factor X (-GO) and the "Gla Peptide" (des residues 42-44) from Chymotryptic Hydrolysis of the Gla Peptide (Residues 1-44)

Li&lhi.chainIL=eLL Amino acid

. . p f E a € i . L x ~ ~ I t s ~ a n d " BminnBEidromoosltlcn Xa(-GD) Light chain

8-Xa (-GD) Residues a

ASP Thr ser Glu PI0

GlY Ala CmCys Val Met Ile Leu TYr Phe xis LYS Arg TIP

22

a

29.6 24.4 23.1 37.5 10.7 37.1 21.0 d 17.1 5.6 0.4 17.7 7.9 17.2 9.4 20.4 21.0

10.8 5.4 8.8 13.5 2.0 14.2 3.1

21 5 11 17 8 18 9 21 21 4

e

2.9

b

C

l l r v n d 11 5 8 13 2 13 2 13

18.6 18.5 9.9 24.3 7.9 19.0 16.3 17.2 5.4 7.3 13.4 5.1 12.2 5.7 15.3 15.0 e

0

3 0

1.9 3.2 2.0 5.3 3.5 5.1

2 3 2 5 3 5

6.0 0

6 0

19 20 12 24

from the published sequence. 143).

Calculated from the published sequence. (38).

d

Not determined.

e

Present, but not quantitated.

4Met; e

18 5 9 14 6

13

Found

7 3 7 7 3 3 5 0 0 0

T v

1.25e 0.07 0.14 1.00 0.10

AT9

0.2 0.19 2.27

Phe His LYS

9

a

Calculated for Residues 345d"

0

Ile Le"

8

13 18

2

0

1

1 6

0

1

3.9 1.1 2.0 14.0 0

0

1 4 2 2

f

2

1.8

0

0 0 4.3 0.2 2.8

0

4 1 2 14

1.2 3.5

0

1 7 1

c

Calculated

0 0

4 0

3

2

0

0

0

0

0

0

2 1

1.2 2.0 f

1 2 1

0

0

0

T v

0

2 0

Hexosamine

f

3-4

2 1 9

0

g

3-4

9

6

16 15 4

Determined after acid hydrolysis for 24 hours as described in Methods, thebasis Arg the numbers of residues per mol are calculated on of the content of the particular component.

b Calculated c

30 25 20 37 10 36 20

a l l r v n d -

Residues a

Found

1.55 0.93 1.27 2.48 1.16 2.34 1.07

Xa(-GD) Aeavy chain

Residues brc

EPvnd"

b

a

TABLE I1

a

b c d e

f

g

Determined after acid hydrolysis for 24 hours as described in Methods. Numbec of residuesare calculated on the basisof 1 Leu in the activation peptide and 2 Arg in the Gla peptide. calculated from the published Sequence, (43). Calculated from the published sequence, ( 3 8 ) . Not determined. The presence of a Val-Val sequence results in systematically low values after only 24 hours of hydrolysis. Present, but not quantitated. Data are from Hizuochi et al. (51).

Factor X(-GO) andFactor Xa(-GD)

4023

The product present in the trailing shoulder of the last peak of Figure 2A (main text) appearedlikelytobeafurther cleaved"G1apeptide"fromits chromatographic elution, its amino acid composition notably the absence of Tyr (Table III).and the existence of a potential chymotrypsin susceptible site, Trp 41 in the "Gla peptide" sequence. TO confirm this identification, isolated "Gla peptide. was incubated with chymotrypsin and the products isolated by chrbmitagraphy on Sephadex G-50, Fiiure 2.- The larger of the two products was identified by its amino acid Composition a~ residues 1-41 of the liaht chain and the smaller DrodUCt a s the triDeDtide, Ser-Lvs-Tvr (data not shown). In order to obtain qiantitative cleavage; both an iticreised concentration of chymotrypsin and a higher temperature, 37'C. were required. If a 2-fold increase in rate is assumed to result from the increase in temperature. then on this basis and the increased concentration of chymotrypsin used it may be estimated that the Tyr 44-Lys 45 peptide bond is cleaved approximately 100 times faster than the Trp 41-Ser 42 bond. Figure 2 Chymotryptic conversion of the Gla m m a i n peptide (residues 144) to two peptides; Residues 1-41 and Residues 42-44. Gla domain peptide, 4.2 absorbance units (280 nm) in 0 1 n (NR RCO was incubated withi o 0 ua of4c#vmo~ trypsin at 37' C-for 2 (ours. * T h e proteolysis products were isolated by chromatography at'4 C on 1.a5 x 95 cm Sephadex G-50 (Superfine), column equilibrated i n 0.1 n (Nu4) RCO Fraction volumes were 2.0 mf. dbsorbance at 280 nm, (I); 230 nm, ( B ) .

Figure 3 Demonstration of a Proteolytically Degraded Form of FaCtoK X6

Pro GlY Ala 1/2 cys Val net Ile Leu

Tyr

Phe His LVS

8 9 10 5 4 11 7 4

7.7 8.3 7.1 5.6 3.8 9.5 5.4 E

i

6.5 2.5 3.4 5.3 3.8 6.1 1.2 9.0 4.9

3 4 5 4

Factor Xal-GD) from OAE Sephadex chromatography was chromatographed on Sephadex G-100 in 10% acetic acid as described in the Legend to Figure. Nadodecylsulfatepolyacrylamide electrophoresis gels I88 acrylamide) from the starting sample Gthat was applied to the Sephadex 100 column, (1) without disulfide reduction, 12) after disulfide r e duction, (3) the isolated nondisulfide linked degradation product, without disulfide reduction. Discussion

6

Although limited proteolysis of Factor X by chymotrypsin under con1 trolled conditions produces a derivative that is missing only the first 44 9 amino acid residues of the liqht chain in - very vield. hiah . further nroteo5 lysis can occur and thus pilot experiments to ensure that such additional c 2 degradation is minimal are recommended. Because of possible effects of further cleaved derivatives that could confound interpretation of experiments that employ these derivatives of Factor X and Factor Xa, the products of the a Determined by acid hydrolysis fOK 24 hours as described i n Methods. further proteolysis have been isolated and characterized. These products are: Number of residues calculated on thebasis of 10 lysines per mole. a derivativeof Factor X from which the 33 first amino acid residues of the heavy chain have been cleaved in addition to the 44 residues of the light b Calculated from the published sequence 143) for residues 187-290. chain, a form of the "Gla peptide" that contains only 41 amino acid residues and the tripeptide, Ser-Lys-Tyr, residues 42-44. Because cleavage of the Trp c Present but not quantitatively determined. 41-Ser 42 peptide bond occurs at only of the 1% rateof cleavage of the Tyr 44-Lys 45 peptide bond, heterogeneity in Factor XC-GD) is not observed. All of the additional products are easily separated fromXI-GD) Factor and the of a ev" of F production "Gla peptide" a8 shown in Figure 2 (main text) and thus successful U of the HeavV chain of the derivative Factor X(-GD) assured. Identification of a degraded form of at near Arg 186 indicates Factor Xal-GD) in which cleavage has occurred or Factor Xal-GD) was Dreoared from Factor X I-GDl bv incubation with RVVthat an accessible peptide bond exists i n Factor Xa similar to that which X-CP as shown in Figure (kain text). Although initiaily only the two bands upon cleavage gives rise to y-thrombin in human thrombin (55). Products associated with the- heavy chain of Factor Xa-and the light chain from which the %la domain" has been cleaved are visible in NaDodSO polyacrylamide similar to those observed for this degraded farm of Factor Xa have been reported to be formed from Factor Xa-Antithrombin 111 complexes (56), and in electroDhoresis aels. after ChromatoaraDhv on OAE Seohadex %-SO and G-100 to human Factor X (57). Several reports of proteolytically cleaved forms of remove 'the actiGation peptide and thb iVV-X-CP, small amount of a new (58-601. The relationships of species with a mol of Wt approximately 25,000 was detected in the NaDodSO Factor X and Factor Xa have been published those products to the products described in this report cannotbe u n m gels, Figure 3. On the basisof the proposed location of the disulfide bridgd biguously determined. Early work on large scale purification of bovine Factor between the heavy and light chains of Factor X (38,43), a product formed by X 116). Factor X activation by trypsin (61). modified form of human Factor Xa expected to have a mol cleavage of Factor Xa at or near Arg 180 or Argis 186 (60). as well as a recent report on bovine Factor Xa binding to bovine wt of 25,000 and thus depending on its concentration, might be detectable in platelets ( 6 2 ) describe species of Factor xa that from their properties were NamdS04 electrophoresis gels at the position of the observed minor compoprobably forms of Factor Xa from which the "Gla domain" had been cleaved. nent. A band corresponding to the other product of this proposed cleavage, a product with a mol Wt of 12,000, was not observed in the gels, presumably because of the poor staining characteristics (see below). UpOn reduction of the disulfide bridges, N a m d S O A Dolvacrvlamide electrophoresis qels of this degraded Factor Xi(-GD) showea a broad;ned band i n th; light chain region, consistent with the presence of the light chain of Factor Xa(-GD), M r 11,000; DlUS the two ComDonents derived from the heavv chain. Mr 14.000 Droduct corresponding to iesidues 52-180 and nr 12,000 -fragment, residues 'i80-290 or 186-290. Gel filtration of the mixture of Factor Xa(-GD) and the degraded material on Sephadex 6-100 in 10% ROAc resolved the mixture into two companents, the smaller of which had a mol wt by NaDods04 polyacrylamide gel of electrophoresis of 12,000. Determination of the amino acid Composition this component indicated that this polypeptide was derivedfrom the carboxyterminal end of the Factor Xal-GD) heavy chain, Table IV. From the location of the disulfide bridgesin Factor X 138,431, and the composition data. the most likely cleavage site is b e t w e e n A r g 1 8 6 a n d L e u187. No attemptwas made to isolate the polypeptidecorresponding to residues 52-186.

_ _

a-

~

~~

"

~

"

~