A thrombin-like enzyme from timber rattlesnake venom

Biochimica et Btophysica Acta. 748 (1983) 236-244

236

Elsevier BBA 31727

A THROMBIN-LIKE ENZYME FROM TIMBER RATI'LESNAKE VENOM YU-YAN SHU *, JEFFREY B. MORAN and COLLIS R. GEREN **

Department of ChemistQ', Unwersi(v of Arkansas, Favetteville, A R 72 701 ( U.S. A.) (Received February 21st, 1983)

Key words: Snake venom: Procoagulant," Thrombin; Fihrinogen clea~,age; (Crotaht*" horridus horridt**')

The procoagulant component has been purified from timber rattlesnake (Crotalus horridus horridus) venom by DEAE-cellulose ion-exchange chromatography followed by affinity chromatography on immobilized p-aminobenzamidine and a final DEAE-Sepharose chromatography. As obtained, the procoagulant gave a single band of M, 29500 _-_+2000 on SDS-polyacrylamide gel electrophoresis whether or not the sample was reduced prior to electrophoresis. Schiff's stain indicated the presence of some carbohydrate. The procoagulant showed one predominant and four minor bands on discontinuous gel electrophoresis. All caused fibrinogen solutions to clot. Treatment of the preparation with neuraminidase did not cause the minor bands to comigrate with the major band. Amino acid analysis revealed the presence of eight half-cystines, all of which were present as cystines in the intact molecule. The procoagulant has 13 tryptophans per molecule and an extinction coefficient for a 1% solution at 280 nm of 26.3. This venom proeoagulant was found to induce clotting by catalyzing the hydrolysis of only the A fibrinopeptide from the Aa-chain of fibrinogen. It was not inhibited by protein trypsin inhibitors, N-ethylmaleimide or dithiothreitol, but it was inactivated by phenylmethylsulfonyl fluoride, indicating an active-center serine. The procoagulant catalyzed the release of negligible acid-soluble peptides from bovine serum albumin, casein and hemoglobin.

Introduction Many snake venoms contain proteolytic enzymes which exert profound effects on the blood coagulation process, either promoting or inhibiting clotting [1]. Such enzymes are useful, both as research tools for investigating fibrinogen and other elements of the normal clotting system, and for their possible therapeutic applications. A number of snake venom components have been described as thrombin-like procoagulants [2-9]. Like thrombin, most of these have also been

* Permanent address: Department of Biochemistry, Guanxi Medical College, Nanning, Guanxi, China. ** To whom correspondence should be addressed. Abbreviations: BAEE, benzoyl-L-arginine ethyl ester; TAME, tosyI-L-arginine methyl ester. 0167-4838/83/$03.00 © 1983 Elsevier Science Publishers B.V.

reported to have hydrolytic activity with arginine esters. Lundblad et al. [10] reported that thrombin, in addition to catalyzing the hydrolysis of the A and B fibrinopeptides from fibrinogen, also catalyzed the hydrolysis of benzoyl-L-arginine ethyl ester (BAEE), tosyl-L-arginine methyl ester (TAME), and other synthetic arginine esters. The venom thrombin-like enzymes have been found to catalyze the hydrolysis of only the A fibrinopeptide from fibrinogen. The exception has been the procoagulant fraction from the venom of the southern copperhead ( Agkistrodon contortrix contortrix) reported by Herzig et al. [11] and Shainoff and Dardik [12]. This venom component first preferentially catalyzed the hydrolysis of fibrinopeptide B from fibrinogen, followed by the release of A. Most of the venom thrombin-like enzymes have been described as carbohydrate-containing,

237 serine-active-center proteinases of approx. M r 30000-35000. Multiple charge forms have been reported for at least two of these venom enzymes [2,5]. The one exception to the size given above was that of 19500 reported by Bonilla [6] for a thrombin-like enzyme purified from timber rattlesnake ( Crotalus horridus horridus) venom. With this in mind, a more thorough examination of the thrombin-like enzyme from timber rattlesnake venom was undertaken. The results of this work are described in this report. Materials and Methods

Lyophilized timber rattlesnake venom was purchased from the Miami Serpentarium, Miami, FL. Reagents for SDS- and discontinuous polyacrylamide gel electrophoresis and P-150 were purchased from Bio-Rad Laboratories, Richmond, CA. Bovine type I fibrinogen (95% clottable), cross-linked hemoglobin electrophoresis standards, benzamidine, trypsin, soybean trypsin inhibitor, egg white trypsin inhibitor, a,-antitrypsin, iodoacetamide, bovine serum albumin, casein, hemoglobin, dithiothreitoi, N-ethylmaleimide, phenyimethylsulfonyl fluoride, tetranitromethane, type V neuraminidase, TAME and BAEE were purchased from Sigma Chemical Company, ST. Louis, MO. Bovine B grade fibrinogen (98% clottable), human thrombin (2500 NIH units/mg), D-phenylalanyi-L-prolyI-L-arginine chloromethyl ketone (PPACK) and hide powder azure were obtained from Calbiochem, La Jolla, CA. HPLC grade acetonitrile was obtained from Fisher Scientific, Memphis, TN. DEAE-cellulose was from Whatman, Inc., Clifton, N J, DEAE-Sepharose from pharmacia, Piscataway, N J, and p-aminobenzamidine agarose (6-aminocaproic acid spacer) from Pierce Chemical Co., Rockford, IL. The procoagulant containing fraction III a was purified from lyophilized timber rattlesnake venom by gradient elution from DEAE-cellulose as described by Sullivan et al. [13]. Clotting assays were routinely accomplished at 37°C. Each assay consisted of 300 ~1 of 5 m g / m l fibrinogen (95% clottable protein) in 20 mM Trisbuffered saline, pH 7.4, and 25 ~1 of sample. The fibrinogen was preincubated for 5 min at 37°C prior to addition of the sample. Each assay was

tilted every 15 s and clot formation was observed visually. TAME- and BAEE-hydrolyzing activities were determined spectrophotometrically by the methods of Hummel [14] and Schwert and Takenaka [15], respectively. The pH optimum for the BAEE-hydrolyzing activity was determined titrimetrically by the method of Smith and Parker [16] using a Radiometer Model 27 pH Meter equipped with Titrator II and Autoburette II accessories. Hide powder azure hydrolysis was examined by the method of Rinderknecht et al. [17], while hydrolysis of other protein substrates was done essentially by the method of Kunitz [18]. The effect of protein inhibitors (soybean trypsin inhibitor, egg white trypsin inhibitor, and a~-antitrypsin) and modification reagents (phenylmethylsulfonyl fluoride, N-ethylmaleimide etc.) on the olotting and BAEE-hydrolysis activities were accomplished by incubation of the venom fraction or component with the agent for 30 min at room temperature prior to initiation of assays. Metal dependency was examined by dialysis against 100 vol. of 100 mM Tris-HCl, pH 7.4, containing 10 mM EDTA. Spectrophotometric determinations were performed with a Gilford Model 252 up-date on a Beckman DU monochromator. Protein concentrations were estimated by absorbance at 280 nm (A280), assuming an extinction coefficient of 1.0 for a 1.0 m g / m l solution. When necessary, protein solutions were concentrated with CF-25 Centriflo Membrane Cones from Amicon Corporation, Lexington, MA. SDS-polyacrylamide gel electrophoresis was performed essentially by the method of Weber and Osborn [19]. Samples for examination of fibrinogen subunits were incubated with an equal volume of a solution of 4% mercaptoethanol, 4% SDS and 10 M urea for 24 h at room temperature prior to electrophoresis on 7.5% polyacrylamide gels. Discontinuous polyacrylamide gel electrophoresis was by the method of Ornstein [20] and Davis [21]. All electrophoresis was done using a Hoeffer DE 101 Electrophoresis Unit and a Buchler Model No. 3-1155 Power Supply. Both Coomassie brilliant blue and Schiff's stain [22] were used to visualize proteins. A Gilford 2410-S Linear Transport Accessory was used to obtain scans of electrophoresis gels.

238

Release of fibrinopeptides from fibrinogen by both thrombin and the procoagulant from timber rattlesnake venom was assayed by the HPLC method of Higgins and Schafer [23]. Fibrinogen (98% clottable) was dissolved in 200 mM ammonium bicarbonate, pH 7.8, at a concentration of 2.5 mg/ml. Thrombin- and venom procoagulantinduced clots were formed using 8 ml of the fibrinogen solution and incubated 4 h at room temperature. The clots were then crushed with a glass stirring rod and subjected to 30 min of centrifugation at top speed in a clinical centrifuge. The supernatant from each clot was lyophilized, then redissolved in 300 ~1 of a solution containing 10% acetonitrile and 0.083 M ammonium phosphate, pH 3.1. Samples were applied to a 4.6 mm x 25 cm Altex Ultrasphere ODS C-18 column equilibrated with the 14% acetonitrile, 0.083 M ammonium phosphate, pH 3.1, solution. After 5 min of elution with the starting buffer, the acetonitrile concentration was increased to 19%. Chromatography was with an Isocratic Liquid Chromatograph, Model 330, and a Model l l 0 A pump from Altex Scientific, Inc., Berkeley, CA, at a flow rate of 1.0 m l / m i n . Detection was at 205 nm with a Holochrome Variable-Wavelength Detector from Gilson Medical Electronics, Inc., Middleton, WI. Amino acid analyses were accomplished with a microcomputer-controlled microbore amino acid analyzer with ninhydrin detection as described by Durham and Geren [24]. Hydrolysis of samples was done in duplicate with constant boiling HCI at I10°C for 24, 48 and 72 h [25] and all values reported are means. Only duplicate 24-h hydrolyses were done with the fibrinopeptides. Tryptophan was determined spectrophotometrically by the method of Edelhoch [26]. Total cysteine content estimation was accomplished with dithiothreitol and iodoacetamide as described by Gurd [27]. Reactions were in 50 mM Tris, pH 8.3, containing 8 M recrystallized urea. Samples were incubated with dithiothreitol for 1 h at room temperature, then sufficient recrystallized iodoacetamide was added to make the solution 0.25 M and incubation was continued for an additional 2 h at room temperature. The carboxymethylated protein was then dialyzed against deionized distilled water for 24 h at 4°C and subsequently hydrolyzed (24 h

only) and analyzed. Alkylation without reduction was performed in the same way, except for the omission of the dithiothreitol. In this manner, the amount of free cysteine residues was determined. Results

Fraction IIIa was obtained by DEAE-cellulose chromatography of whole timber rattlesnake venom as described earlier [13]. III a comprised 24 + 2% of the A28o protein of the whole venom (five different commercial preparations). As previously reported by Sullivan et al. [13], this fraction contained all of the BAEE-hydrolytic and procoaguiant activities of the whole venom. The procoagulant activity of Ill a was further purified by affinity chromatography on immobilized p-aminobenzamidine (Fig. 1). The procoagulant fractions were pooled, dialyzed against 20 mM Tris-HCI, pH 7.5, then applied to a DEAESepharose column. This chromatography is shown by Fig. 2. Table I summarizes the purification of this procoagulant. The apparent increase in total activity is most likely the result of choosing the

Fret Ion Number(2rnl) Fig. 1. Typical further separation of fraction Ill,, isolated from timber rattlesnake venom by affinity chromatography on immobilized p-aminobenzamidine. III a, 14 ml, 2.56 m g / m l , was dialyzed 2 h against I litre of 20 m M Tris. pH 7.5 at 4°C. A 4.5x1.1 cm column of p-aminobenzamidine agarose (6 C spacer) was equilibrated with 20 m M Tris, pH 7.5. The sample was applied to the column and the column wa,,; sequentially eluted with 20 m M Tris. pH 7.5 (indicated by the arrow a on the chromatograph), 20 m M Tris, (pH 7.5)/1 M NaCI (indicated by b), and 20 m M Tris, (pH 7.5)/500 m M bcnzamidine (indicated by c). The column was eluted at room temperature. The units for reciprocal clotting times are min i

239 .5

,

.4

//t

~,

.,

a

.."

!

,5

4 t 1,~

~"~,

IO 20 Fraction Number (2ml)

]2.0

,

I

30

Aa I 0

Fig. 2. Final purification of timber rattlesnake procoagulant on DEAE-Sepharose. The active fractions indicated in Fig. 1 were pooled and dialyzed overnight at 4°C against 100 vol. of 20 m M Tris, pH 7.5. The 4.5×1.1 cm column was equilibrated with 20 m M Tris, pH 7.5, and the sample was applied. The column was eluted with 20 m M Tris, pH 7.5, (indicated by the arrow on the chromatograph) followed by a gradient consisting of 20 ml of 20 m M Tris, pH 7.5, versus 20 ml of 20 m M Tris (pH 7.5)/500 m M NaCI. The salt gradient is indicated by the dashed line. The units for reciprocal clotting times are m i n - t.

amount of material that causes a clot between 45 s and 1 min as the basic unit for clotting activity. Obviously the recovery of activity with this procedure is good. Fig. 3 shows SDS- and discontinuous polyacrylamide gel separations of the purified procoagulant. The protein band on the SDS gel was also faintly stained with Schiff's stain, indicating its glycoprotein character. Reduced and non-reduced samples of the procoagulant migrated identically on SDS gels, indicating the presence of a single polypeptide chain. The apparent molecular weight of the procoagulant was 29 500 + 2000 (10 determinations) as compared to the migration of cross-linked hemoglobin standards. The discontinuous gel shows one major and four minor components. Incubation of 70 ~tg of the procoagulant with 10 /~g of neuraminidase did not cause any change in the pattern. The major band comprised at least 60% of the total protein in the purified procoagulant preparation. This value was determined by scanning the gel and quantitating the areas under the peaks. The purified procoagulant preparation was subjected to discontinuous electrophoresis on four gels (35 ~tg per gel). Two gels were stained and used as a pattern to allow sec-

TABLE I P U R I F I C A T I O N OF T H E P R O C O A G U L A N T F R O M TIMBER R A T T L E S N A K E V E N O M Total protein was estimated by 280 n m absorbance assuming an extinction coefficient of 1.0 for a 1.0 m g / m l solution of protein. A unit is defined as that amount of the indicated preparation that would cause a clot under the assay conditions described in Materials and Methods between 45 and 60 s after its addition to the fibrinogen solution. 5 of these units are approximately equal to 1 N I H unit. The apparent increase in total units is probably due to the imprecise definition of a unit rather than to any real increase in activity. Affinity chromatography was on p-aminobenzamidine-agarose (6 C spacer). Step

Total protein (mg)

#g protein/' unit

Total units

Whole venom DEAE-Cellulose Affinity chromatography DEAE-Sepharose

575 132 11 6.1

230 50 _ 2

2500 2 640

a

_ • 3035

a Much less activity was apparent after this step due to benzamidine inhibition. This will be described in more detail later.

""4

:i

.!1 Fig. 3. Electrophoretic gels of the purified timber rattlesnake procoagulant. From left to right, the first gel is a discontinuous separation of 20 ~g of the purified procoagulant. The second is an SDS ~ p a r a t i o n of 24 ~g of purified procoagulant, while the right gel contains cross-linked hemoglobin standards. The molecular weights of the standard are indicated with K representing 1000 daltons. Reduction of the procoagulant prior to SDS electrophoresis had no effect on its migration (not shown).

240

tioning of the other gels to separate the protein components. The acrylamide slices were crushed, then 300 #1 of 5 mg/ml fibrinogen solution were added. All of the sections with the exception of a control of crushed blank acrylamide caused the fibrinogen to clot within 10 min. A second set of gels were run and sectioned. This time the gel slices were crushed with 200 p.l of added 20 mM Tris, pH 7.4. The acrylamide was compacted by centrifugation and 50-#1 aliquots of the resulting supernatants were assayed for procoagulant activity with 300 #1 of fibrinogen solution. The supernatant from the major band caused a clot in 10 min while the others caused clots within 2 h. The supernatant from only acrylamide caused no clot. Apparently all of the proteins present in the preparation have procoagulant activity. Table II shows the amino acid composition of the purified procoagulant. Reduction and alkylation revealed eight half-cystines per M r 29500 molecule, while alkylation only revealed 0.54 free sulfydryls per molecule, suggesting that all eight

I c

I <

15

20 mlnJte5

25

Fig. 4. HPLC separations of clot supernatants. Procedures are described in Materials and Methods. The upper trace shows fibrinopeptides A and B released by thrombin, while the lower trace shows that only fibrinopeptide A is released by the timber rattlesnake procoagulant.

TABLE II A M I N O ACID COMPOSITION OF THE PROCOAGULANT ISOLATED FROM TIMBER RATTLESNAKE VENOM

The number of residues per molecule was based on an assumed molecular weight of 29500 (SDS gel clectrophoresis). Trypophan was determined by the method of Edelhoch [26]. Total half-cystine was determined by the reduction-carboxymethylation method of (iurd

[27]. Amino acid

Asx Thr

Ser (ilx

Pro Gly Ala Val

Met lie Leu

Tvr Phe His

Lys Arg Trp 1 / 2 Cys

Total

Hydrolysis

Mean

24 h

48 h

72 h

31.72 11.83 20.00 21.88 20.61 26.05 16.30 11.82 5.32 13.32 22.55 5.72 9.13 6.94 12.24 9.39

27.91 11.47 26.57 22.74 26.58 27.62 16.19 14.40 4.20 17.32 22.95 5.84 9.83 7.16 10.75 10.14 12 8

29.94 12.68 15.93 21.33 23.69 23.17 15.60 13.04 4.46 15.67 23.35 6.56 9.31 6.37 12.14 9.54

Nearest

54,

integer 29.86 11.99 17.50 21.98 23.63 25.61 16.07 13.08 4.66 15.44 22.95 6.(~, 9.42 6.82 11.71 9.69 12 8

30 12 18 22 24 26 16 13 5 15 23 6 9 7 12 10 12 8

3 450 1 212 1 566 1 838 2 328 1482 1 136 I 287 655 1 695 2 599 078 1 323 959 1 536 1 560 2232 824

268

29 660

241 exist as cystines. The relatively large n u m b e r of t r y p t o p h a n s per molecule is confirmed by this substance's extinction coefficient for a 1% solution at 280 n m of 26.3 and its ultraviolet a b s o r p t i o n spectrum, which c o n t a i n s a p r o n o u n c e d 280 peak with a shoulder at 294 n m indicative of tryptophan. N o glucosamine or galactosamine was observed in a n y of the a m i n o acid analyses. SDS gels were o b t a i n e d of fibrinogen that had been i n c u b a t e d with the purified timber rattlesnake v e n o m p r o c o a g u l a n t a n d c o m p a r e d to t h r o m b i n - i n c u b a t e d fibrinogen. The patterns of both were very similar. Both fibrinogens c o n t a i n e d a, fl and y subunits. Fig. 4 shows the H P L C

TABLE 111 COMPARISON OF AMINO ACID COMPOSITIONS OF FIBRINOPEPTIDES PRODUCED BY THE INCUBATION OF THE PURIFIED TIMBER RATTLESNAKE VENOM PROCOAGULANT AND THROMBIN WITH FIBRINOGEN TRSV is timber rattle snake venom. Clot liqueurs were separated from clots and fibrinopeptides purified by C-18 reversed-phase HPLC as described in Materials and Methods. The numbers in parentehses refer to values previously reported [31] that differ from those experimentally obtained. Amino acid

TRSV procoagulantinduced fibrino-

Thrombininduced fibrinopeptide A

Thrombininduced fibrinopeptide B

separation of clot liquors from both preparations. T h e t h r o m b i n catalyzed the release of both the A a n d B fibrinopeptides, while the p r o c o a g u l a n t caused only the release of the A peptide. This is consistent with other v e n o m procoagulants [2-4]. T a b l e III compares the a m i n o acid compositions of the v e n o m p r o c o a g u l a n t - p r o d u c e d A a n d the t h r o m b i n - p r o d u c e d A a n d B with literature values reported for these peptides. Over 95% of the BAEE-hydrolyzing activity of I l i A was separated from the p r o c o a g u l a n t d u r i n g its purification by ion-exchange a n d affinity chromatographies. The question r e m a i n s as to whether the r e m a i n i n g BAEE-hydrolytic activity is a property of the procoagulant, or a m i n o r c o n t a m i n a n t . I n c u b a t i o n of the purified p r o c o a g u l a n t with 17 m M t e t r a n i t r o m e t h a n e prior to BAEE-hydrolysis assays resulted in no change in the ester-hydrolyzing activity of the procoagulant while causing a 20% reduction in the rate of hydrolysis catalyzed by the separated BAEE-hydrolytic activity. This would suggest that two ester hydrolases were present, but not necessarily that the m i n o r BAEE hydrolase was the procoagulant. The BAEE-hydrolytic activity of the whole v e n o m was 32 interna-

1.,:1

. ~Z'.j0 0.O8

peptide A Asx Thr Ser Gly Pro Gly Ala Cys Val Met lie Leu Tyr Phe His Lys Arg

3 1 2 2 2 5 1 0 1 0 0 1 0 1 0 0 -- ~

3 1 2 2 2 5 1(0) 0 1 0 0 1 0 1 0 0 2

3(4) 1 1(0) 3 2(1) 3 2 0 1 0 0 1 1 1 0 0 -- "

a An accurate estimate of arginine was not obtained due to NH 3 contamination from the NH4HCO 3 buffer which was not totally removed by lyophilization.

o,F ..... \

.o.

y

o

~

"o

C . T . ,:.8 " • .o2

o ,ooWo~oo/• /

,2 ;,4

:2 i

o

~

~

6 P9

~o

Fig. 5. Inhibition of timber rattlesnake procoagulant by benzamidine. Clotting time assays were as described in Materials and Methods. Clotting time in minutes is represented by C.T., while Pro represents timber rattlesnake venom procoagulant. O, no benzamidine; ©, 100 mM benzamidine; A 300 mM benzamidine. The inset shows a secondary plot of slope versus benzamidine concentration. Benzamidineis a poor inhibitor as 230 mM was required to double clotting times.

242

tional u n i t s / m g of .4280 protein, while that of the purified procoagulant was 25 units/rag, so its specific esterase activity actually decreased during purification. The pH optimum for the procoagulant was 7.0. The pH optimum for the BAEE-hydrolyzing activity of III~ was 8.5. It was stable to short-term (0.5-2 h) incubation at pH 11.5, but activity was reduced at pH values below 5. Incubation of up to 82 ~g IIIa with 5 m g / m l solutions of bovine serum albumin, casein and hemoglobin for 1 h at 37°C in 20 mM Tris-buffered normal saline, pH 7.4, resulted in no apparent production of acid-soluble peptides. Denaturation of the proteins by heating at IO0°C for 15 min prior to incubation with IIla resulted in no increased susceptibility of bovine serum albumin and hemoglobin, and only a very small amount of

solubilized casein. Incubation of 82 /~g IIIa/ml with hide powder azure for 1 h at 37°C in 20 mM Tris-buffered saline, pH 7.4, resulted in no solubilized dye. IIIa at 100 ~ g / m l caused no hydrolysis of TAME. Incubation of 1.5 m g / m l III, in 200 mM sodium acetate, pH 6, at 37°C for 45 h resulted in no loss of procoagulant or BAEE-hydrolytic activities. From these results it was assumed that the procoagulant was devoid of those activities not found in III~. Also, the procoagulant (and BAEE hydrolase) was not destroyed by autoproteolysis. 30 min room temperature incubation with 20 mM N-ethylmaleimide had no effect upon the procoagulant activity. Likewise, 20 mM dithiothreitol had no effect. Extensive dialysis against either 100 mM Tris, pH 7.4, or 100 mM Tris (pH 7.4)/10 mM EDTA caused no change. Phenyl-

TABLE IV COMPARISON OF A M I N O ACID COMPOSITIONS OF TIMBER RATTLESNAKE PROCOAGULANTS WITH PROCOAGULANTS ISOLATED FROM OTHER VENOMS Compositions are expressed as residues per 1000 residues and residues per 100000 daltons so that direct comparisons could be made as the different procoagulants vary in molecular weight. TRV ProC is the procoagulant isolated from timber rattlesnake venom described in this work. TRV Pro(2 is the procoagulant isolated from timber rattlesnake venom de~ribed by Bonilla [6]. The composition was re-calculated from Bonilla's raw data as incorrect mean residue molecular weights were used in Ref. 6. Crotalase is the procoagulant isolated from Crotalus adamanteus venom. Arvin is the procoagulant isolated from Agkistrodon rhodostoma venom. Amino acid Asx Thr Set GIx Pro Gly Ala Cys Val Met lie Leu Tyr Phe His Lys Arg Trp Total Mr Ref.

Residues/lO00 residues TRV Pro(? 112 45 67 82 90 97 60 30 48 19 56 86 22 34 26 45 37 45 1001 29500 This work

TRV ProC

Residues/100000 daltons Crotalase

Arvin

Thrombin

156 46 87 104 40 75 58 35 69 12 40 58 46 40 23 46 29 35

116 52 59 86 82 75 41 52 64 7 67 79 26 49 34 41 45 22

140 37 65 61 65 84 51 23 61 23 75 65 23 37 37 47 89 14

108 50 54 112 54 81 46 23 62 16 43 93 39 39 19 70 66 23

999 19500 [6]

997 32700 [28]

997 33300 [29]

998 33700 [30]

TRV ProC

TRV ProC

Crotalase

Arvin

Thrombin

102 41 61 75 81 88 54 27 44 17 51 78 20 30 24 41 34 41

138 41 77 92 36 67 51 31 62 10 36 51 41 36 20 41 26 31

95 43 49 70 67 61 34 43 52 6 55 64 21 40 28 37 34 18

90 24 42 39 42 54 33 15 39 15 48 42 15 24 24 30 57 9

83 38 42 86 42 62 36 18 47 12 33 71 30 30 15 53 50 18

243 methylsulfonyl fluoride (0.5 mM) for 30 min at room temperature caused total loss of the procoagulant activity. An amount of the procoagulant that normally caused fibrinogen to clot in 1 rain did not cause a clot even when incubated with fibrinogen for 1 h after treatment with phenylmethylsulfonyl fluoride. (This treatment also caused the total loss of the venom's BAEE-hydrolytic activity.) 30 min incubation with 7 mM tetranitromethane increased clotting times induced by the procoagulant 6-10-fold. The procoagulant was stable in 5 M urea. Procoagulant activity did not decrease during incubation with trypsin (1 /~g trypsin/12 /~g procoagulant) for 2 h at room temperature. The procoagulant was not inhibited by incubation with soybean or eggwhite trypsin inhibitor, or al-antitrypsin in concentrations as high as 10 p.g of trypsin inhibitor//.tg of procoagulant in 20 mM Tris-buffered saline. (Trypsin was used as a control to verify that these reagents had antiproteinase activity.) Preincubation of the procoagulant with 1.3 mg h e p a r i n / m l prior to clotting time assays had no effect. 10 #M D-phenylalanyl-L-prolyI-L-arginine chloromethyi ketone totally inhibited thrombin that in the absence of inhibitor caused clotting in 1.5 min. This thrombin inhibitor had no effect on the venom procoagulant at concentrations as high as 160 ~M. Fig. 5 shows the effect of different benzamidine concentrations on the reciprocal of clotting times induced by different concentrations of the procoagulant. The inset on Fig. 5 plots slopes versus benzamidine concentration. The slope was reduced to 50% of its original value by 230 mM benzamidine.

Discussion The findings reported here do not agree with the earlier report by Bonilla [6] which described a procoagulant from timber rattlesnake venom that had a molecular weight of less than 20 000 and was not retained by DEAE-cellulose. Instead, we found a procoagulant that had an M r of 29 500, which is more in line with results reported for other venom procoagulants. We were also unable to find any conditions (pH 6-10) under which our procoagulant was not retained by DEAE-cellulose. Table IV compares the amino acid composition of our

procoagulant with that reported by Bonilla [6] and also with that reported for the procoagulant isolated from Crotalus adamanteus venom, Crotalase [28], the p r o c o a g u l a n t from Agkistrodon rhodostoma, Arvin [29], and thrombin [30]. These comparisons are based on residues per 100000 daltons as the procoagulants vary in molecular weight. The procoagulant described by Bonilla [6] is different and could be a fragment of the procoagulant described in the current report. Our procoagulant is most similar to Crotalase although it has much less half-cystine. Markland [4] has reported that pure Crotalase has a specific activity of 222 N I H units per mg. From Lundblad et al. [10], 1.0 NIH unit corresponds to that amount of procoagulant that will cause a clot in approx. 15 s (300/~1 assay). Using our defined unit (that amount of procoagulant that causes a clot between 45-60 s, 300 ~1 assay), a value of 500 u n i t s / m g was obtained for timber rattlesnake procoagulant. In comparing reciprocal clotting times, the NIH unit is approximately 4-times larger than the one used in this report, hence the timber rattlesnake procoagulant is only half as active per mg as Crotalase. The amount of procoagulant activity per gram of whole venom is 3-fold higher for Crotalus adamanteus venom than for that of Crotalus horridus horridus. The timber rattlesnake procoagulant is a serine-active-center proteinase which is quite stable. It, like Crotalase [4], shows several bands upon disc electrophoresis, but, unlike Crotalase, incubation with neuraminidase did not cause the minor bands to merge with the major band. All of the bands of the timber rattlesnake procoagulant separated by discontinuous electrophoresis did cause fibrinogen to clot. The procoagulant was fairly specific for fibrinogen as it had little apparent activity with other potential protein substrates. It only catalyzed the hydrolysis of fibrinopeptide A from fibrinogen.

Acknowledgements This work was supported by Public Health Service Grant G M 24173. C.R.G. is the recipient of Research Career Development Award NIH ES 00052.

244

References 1 Tu, A.T. (1977) Venoms: Chemistry and Molecular Biology, 329-350, John Wiley and Sons, New York 2 Nolan, C., Hall, L.S. and Barlow, G.H. (1976) Methods Enzymol. 45, 205-213 3 Stocker, K. and Barlow, G.H. (1976) Methods Enzymol. 45, 214-223 4 Markland, F.S.. Jr. (1976) Methods Enzymol. 45, 223-236 5 Bajwa, S.S. and Markland, F.S., Jr. (1979) Thromb. Res. 16, 11-23 6 Bonilla, C.A. (1975) Thromb. Res. 6, 151-1169 7 Magalhaes, A., De Oliveira, G.J. and Diniz, C.R. (1981) Toxicon 19, 279-294 8 Ouyang, C., Hong, J.-S. and Teng, C.M. (1971) Thromb. Diath. Haemorrh. 26, 224-234 9 0 u y a n g , C. and Yang, F.Y. (1974) Biochim. Biophys. Acta 351,354-363 10 Lundblad, R.L., Kingdom H.S. and Mann, K.G. (1976) Methods Enzymol. 45, 156-176 11 Her-zig, R.H., Ratnoff, O.D. and Shainoff, J.R. (1970) J. Lab. Clin. Med. 76, 451-465 12 Shainoff, J.R. and Dardik, B.N. (1979) Science 204, 200-204 23 Sullivan, J.A., Farr, E. and Geren, C.R. (1979) Toxicon 17, 269-277 14 Hummel, B.C.W. (1959) Can. J. Biochem. Physiol. 37, 1393-1399

15 Schwert, G.W. and Takenaka (1955) Biochim. Biophys. Acta 16, 570-575 16 Smith, E.L. and Parker, M.J. (1958) J. Biol. Chem. 233. 1387-1391 17 Rinderknecht, H., Geokas, M.C., Silverman, P. and Haverback, B.J. (1968) Clin. Chim Acta 21, 197-203 18 Kunitz, M. (19,16) J. Gen. Physiol. 29, 149-154 19 Weber. K. and Osborn, M. (1969) J. Biol. Chem. 244, 4406-4412 20 Ornstein, L. (1964) Ann. N.Y. Acad. Sci. 121,321-347 21 Davis, B.J. (1964) Ann. N.Y. Acad. Sci. 121,404-427 22 Zacharias, R.M., Zell, T.E., Morrison, J.H. and Woodlock, J.J. (1969) Anal. Biochem. 30, 148-152 23 Higgins, D.L. and Schafer, J.A. (1981) J. Biol. Chem. 256, 12013-12017 24 Durham, B. and Geren, C.R. (1981) Anal. Biochem. 116, 331-334 25 Spackman, DH., Stein, W.H. and Moore, S. (1958) Anal. Chem. 30, 1190-1206 26 Edelhoch, H. (1967) Biochemistry 6, 1948-1954 27 Gurd, F.R.N. (1967) Methods Enzymol. 11. 532-541 28 Markland, F.S. and Damus. P.S. (1971) J. Biol. Chem. 246, 6460-6473 29 Hatton, M.W.C. (1973) Biochem J. 131,799-807 30 Seegers, W.H., McCoy, L.. Kipper, R.K. and Murano, G. (1968) Arch. Biochem. Biophys. 128, 194-201 31 Laki, K. (1968) in Fibrinogen (Laki, K., ed.), p. 98, Marcel Dekker, New York