Proteolytic activities of cobra venoms based on inactivation of α2-macroglobulin

Biochimica et Biophysica Acta, 784 (1984) 97-101 Elsevier

97

BBA 31820

PROTEOLYTIC ACTIVITIES OF COBRA VENOMS BASED ON INACTIVATION OF a2-MACROGLOBULIN HERBERT J. EVANS and V. HOPE GUTHRIE Department of Biochemistry, Medical College of Virginia, Virginia Commonwealth University, Richmond VA 23298 (U.S.A.) (Received July 4th, 1983)

Key words: Proteolytic activity; a 2- Macroglobulin; Inactivation; Snake venom; (N. nigricollis)

The venoms of various cobra species showed a wide range of abilities to cleave hide powder azure, with Naja naja kaouthia and Ophiophagus hannah venoms showing the lowest activities and Naja nivea venom showing the greatest activity on this dye-linked substrate. The activities of the venoms on hide powder did not completely correlate with their ability to inactivate the a2-macrogiobulin of human serum, Incubation of 4 - 5 I~g of Naja nigdcollis venom per itl of serum for 30 min caused loss of 95% of the a2-macroglobulin activity of the serum. The inactivation was rapid, reaching 80% inactivation 5 min after mixing. This loss of a2-macroglobulin activity was used to quantitate the weak proteolytic activity of N. nigricoilis venom and a partially purified sample of the major fibrinogenolytic proteinase of the venom. The inactivation of a2-macroglobulin was also used to compare the proteinase activities of venoms from seven species or subspecies of cobra. Based on a2-macroglobulin inactivation, N. nigricollis had the highest proteinase activity among the tested venoms. The measurement of az-macroglobulin inactivation should provide a useful alternative to hide powder digestion for demonstration of weak proteolytic activities in venoms.

Introduction

Elapidae venoms have relatively little effect on casein and other protein substrates when compared with the proteolytic capacity of Crotalidae and Viperidae venoms [1-6]. A high proportion of the protein of cobra venoms, for example, comprises low molecular weight toxins and phospholipases, and only a small proportion of the venoms is devoted to enzymes with molecular weights over 20 000. Recent work in this laboratory demonstrated that Naja nigricollis venom contains one or more ZnE+-requiring metalloproteinases with the ability Abbreviations: SDS, sodium dodecyl sulfate; BAPNA, N-abenzoyl-DL-arginine p-nitroanilide-HCl; BAEE, N-a-benzoylh-arginine ethyl ester; TAME, N-a-p-tosyl-L-ar~nine methyl ester. 0167-4838/84/$03.00 © 1984 Elsevier Science Pubfishers B.V.

to cleave the Aa-chain of fibrinogen and the apolymer of fibrin [7]. Because of the interest in such enzymes for prevention or degradation of blood clots, we are isolating and studying these proteinases from cobra venoms. The proteinases have narrow specificity and cleave relatively few peptide bonds in their substrates. Their quantitation using any of the-common assays for proteinases based on generation of soluble peptides is difficult. We have routinely located fibrinogenolytic activity by observing cleavage of the Act-chain of fibrinogen after resolving the reduced incubation products on SDS-polyacrylamide gels, Because the electrophoretic method is slow, we have sought a faster and more quantitative method for determination of the activity. The present paper describes a method for quantitation of the proteinases of cobra venoms based on their ability to complex with and inactivate a2-macroglobulin.

98 This inactivation was used to compare the proteolytic activities of several cobra venoms. Materials and Methods Materials N. nigricollis (West African and East African) venoms were obtained from Miami Serpentarium. All other cobra venoms were obtained from Sigma Chemical Company. A partially purified sample of the major fibrinogenolytic proteinase from N. nigricollis venom was prepared by chromatography of the crude venom on Bio-Rex 70 resin (Evans, H.J., unpublished work). H u m a n serum used as a source of az-macroglobulin was prepared by recalcification of fresh citrated plasma by making it 8 m M in CaC12. The recalcified plasma was centrifuged 30 min after clot formation and the supernatant serum was divided into 2-ml aliquots and frozen at - 2 0 ° C until use. Trypsin, soybean trypsin inhibitor, BAPNA, hide powder azure and all other reagents were obtained from Sigma." Protein determination Protein concentrations were determined by the Bradford procedure [8] using Bio-Rad protein reagent: Bovine gamma-globulin was used to generate the standard curve. Measurement of trypsin-like activity Trypsin-like activity was determined by the ability of crude venoms to cleave BAPNA in incubations containing 270 #1 of 0.05 M Tris-HC1 ( p H 8.0) and 1.5 ml of 0.003 M BAPNA. After 30 rain at room temperature, the incubations were stopped by the addition of 250 #1 of 30% acetic acid and the absorbance was read at 410 nm against a blank prepared with venom omitted.

of serum and 0.30 M Tris-HC1 (pH 8.0) in a final volume of 200 #1. After 10 min of preincubation, a2-macroglobulin was measured by a modification of the Ganrot procedure [10]. To the preincubation mixture, 50 #1 of 1.6 m g / m l trypsin was added. After 3 min, 50 #1 of 1.6 m g / m l soybean trypsin inhibitor and 1.20 ml of 0.003 M BAPNA were added and the reaction was timed for 15 min. The reaction was stopped with 200 #1 of 30% acetic acid and absorbance at 410 nm was read against a blank prepared in the same way but lacking venom protein and trypsin. Samples with venom protein omitted were used to determine the 100% value of ct2-macroglobulin activity. Controls with serum or soybean trypsin inhibitor omitted were routinely included to be certain that the trypsin inhibitor and trypsin were active. Individual samples were assayed in triplicate. One unit of a2-macroglobulin inactivating activity was defined as the amount of enzyme which reduced the absorbance at 410 nm by one absorbance unit under the conditions described above. Results

A wide range of responses are seen when the proteolytic activities of cobra venoms are compared by cleavage of hide powder azure (Fig. 1).

0.4

E

c tO O~ tO

0.2 u

.0 <

Measurement of proteinase activity The hide powder azure method [9] for measurement of proteinase activity was used directly, except the amount of substrate and volume were scaled down to 16 nag of substrate in 2 ml final volume of 0.05 M Tris-HCl (pH 8.0). The inactivation of the az-macroglobulin of serum was used to measure proteinase activity by adding various amounts of venom protein to preincubations in plastic test tubes containing 50 #1

0

10o

300 500 venom protein (#g) Fig. 1. Proteolytic activities of cobra venoms determined by hide powder azure cleavage.Absorbmtcewas measured after 30 min incubation of the indicated amount of venom protein and substrate in a final volume of 2 ml. x, N. nivea; e, N. nigricollis (W. African); O, N. nigricolli$ (E. African); v, N. nigricollis crawshawii; A, N. naja atra; A, N. naja kaouthia; I1, O. hannah.

99

The venoms of N. naja kaouthia and Ophiophagus hannah showed especially weak cleavage of the dye-linked substrate under our assay conditions, even at the highest venom concentrations tested. The weak proteolytic activities of these venoms in assays using dye-linked substrates can be partially explained by the small amounts of proteinases in the venoms, but another explanation could be the narrow specificity of the proteinases. Proteinases with narrow specificity would be expected to make fewer peptide-bond cleavages in the substrate and generate fewer soluble products. To circumvent this problem, we explored the possibility of quantitation by a method not dependent on extensive cleavage of the substrate, involving the interaction of the venom proteinases with a 2macroglobulin. Levels of a2-macroglobulin can be measured based on the ability of a2-macroglobulin to protect the esterolytic activity of trypsin from inhibition by soybean trypsin inhibitor [10]. The prior reactions of other proteinases with a2-macroglobulin would be expected to reduce the protective effect of a fixed amount of the macroglobulin, resulting in lower esterolytic activity.

To apply this system to the measurement of venom proteinases, it was first necessary to determine whether the venoms could act directly on the ester substrate, BAPNA, used to assay trypsin. Incubations of BAPNA with up to 140 # g / m l of the venoms for 30 min revealed insignificant trypsin-like activity, with a maximum generated absorbance of 0.01 unit at 410 nm. The initial demonstration of the interaction and inactivation of the a2-macroglobulin of serum was made using N. nigricollis (West African) venom (Fig. 2). The a2-macroglobulin of 50 #1 of serum was maximally inactivated by 200 #g of venom protein after a 30 min preincubation. When the preincubation time was varied, the inactivation of the macroglobulin was found to be very rapid (Fig. 3). Treatment of serum with two different amounts of venom protein showed that the interaction was essentially complete within 10 rain. The lower amount of venom protein did not continue to inactivate a2-macroglobulin a t the rapid initial rate, suggesting that the inactivation occurred by complex formation, although there was a continued slow drop in the macroglobulin

100

75

~:

50

0

8

25

+

i = '

'

t[ '

Venom p r o t e i n ( # g )

Fig. 2. Inactivation of serum az-macroglobulin by N. nigricollis (West A f r i c a n ) v e n o m . Serum was preincubated in a final volume of 200 #1 with the indicated a m o u n t s of crude venom for 30 rain before measurement of a2-macroglobulin activity. Values are the mean of three determinations (+S.D.). The 100% level of a2-macroglobulin activity, determined with no added venom,' had a m e a n absorbance of 0.527 at 410 nm.

0

10 20 30 60 preincubation time (rain) Fig. 3. Effect of preincubation time on inactivation of serum az-macroglobulin by N. nigricollis (West African) venom. Serum was preincubated with 1 ( O ) or 4 (O)/~g of venom protein per pl of serum in a final volume of 200 pl for the indicated time before measurement of a2-macroglobulin activity. Values are the mean of three determinations ( + S.D.). The 100% level of az-macroglobulin , determined with no added venom, had a m e a n absorbance of 0.544 at 410 nm.

100 TABLE I COMPARISON OF PROTEOLYTIC ACTIVITIES OF COBRA VENOMS BY INACTIVATION OF a2-MACROGLOBULIN AND HIDE POWDER DIGESTION

0.4

Incubations for a2-macroglobulin inactivation were performed as in Fig. 4. The data for hide powder digestion are from Fig. 1, given as A A / m g to allow comparison.

O

Cobra species

a2-Macroglobulin inactivation (AA/mg)

Hide powder digestion (AA/mg)

N. naja atra O. hannah N. naja kaouthia N. nivea N. nigricollis (West African) N. nigricollis crawshawii N. nigricollis (East African)

2.65 4.00 4.50 5.25

0.45 0.18 0.20 0.81

5.75

0.70

6.75

0.56

7.25

0.70

<

o

i

~o

. . . 40 . . Protein

~o

(~g)

Fig. 4. Comparison of inactivation of serum a2-macrogiobulin by crude venom and a partially purified proteinase. Serum was preincubated in a final volume of 200 /~1 with the indicated amounts of protein for 10 rain before measurement of a2-macrogiobulin activity. The change in (A) absorbance is the net decrease from the 100% control value (no venom protein). Values are the mean (+ S.D.) of four determinations (crude venom, $) or three determinations (proteinase, O).

activity for the full hour of the experiment. The advantage of using a:-macroglobulin inactivation as an index of proteolytic activity becomes apparent when the data are plotted as the difference between the absorbances generated by a2-macroglobulin protection of trypsin in the presence or absence of venom protein. The inactivation of a2-macroglobulin by crude N. nigricollis venom was compared with the inactivation by a partially purified sample of a fibrinogenolytic proteinase isolated from that venom (Fig. 4). The specific activity of the crude venom was 5.75 and the specific activity of the proteinase was 75, indicating that the proteinase had been purified about 13-fold. Proteinase measurement based on a2-macroglobulin inactivation thus allowed quantitation of this venom proteinase which generates only small amounts of soluble cleavage products. Several cobra venoms were compared for their ability to inactivate a2-macroglobulin in serum. The data from several plots, as in Fig. 4, were converted to the specific activity for each venom (Table I). In all instances, the sensitivity of this inactivation method is greater than that of the hide powder azure method. There is not complete corre-

lation between the two methods, since the two venoms with the lowest activities on hide powder, O. hannah and N. naja kaouthia, show intermediate activities with the other procedure. Discussion

Kress and Paroski showed that several venoms of the Crotalid, Viperid and Colubrid families inactivate al-proteinase inhibitor. When incubated with serum, the venoms cause loss of all trypsin and chymotrypsin inhibitory activity of the serum [11]. The Elapid venoms tested did not inactivate ~q-proteinase inhibitor [11], nor did they inactivate antithrombin III in another study [12]. The present work shows that the proteinases of Elapid venoms do, however, inactivate a2-macroglobulin, confirming that the interaction with t~2-macroglobulin is a more general property of venom proteinases. We have taken advantage of this complex formation to devise a method for quantitation of the weak proteolytic activities of cobra venoms. The method of a2-macroglobulin inactivation provides a non-kinetic measurement of proteinases which generate small amounts of soluble products with hide powder as the substrate. It should allow quantitation of these enzymes during their purification, and provides an alternative to the slower

I01

semi-quantitative assay using electrophoresis on SDS-polyacrylamide gels [7]. Although we have applied the method primarily to fibrinogenolytic proteinases, it obviously is not specific for these enzymes, since complex formation with az-macroglobulin appears to be a feature of most, if not all, endopeptidases [13]. The method should be generally applicable to measure low concentrations of proteinases when the samples are devoid of trypsin-like activity. Several non-Elapid venoms, on the other hand, readily lyse BAEE, TAME and BAPNA [1-3,14,15] and would not be amenable to the present method. There is general, but not complete, agreement between the results obtained in the comparison of cobra venoms by aE-macroglobulin inactivation and by hide powder azure cleavage (Table I). The incomplete correlation might be due to differences in the specificities of the proteinases, which would result in different efficiency of digestion of the hide power substrate. For example, N. naja atra venom may have the lowest overall proteinase content of the tested venoms, but a broader specificity of the proteinases is suggested by the results on hide powder. Also, the binding of proteinases to aE-macroglobulin may vary with the proteinase, and may be affected by other components of serum. Proteinase II isolated from Crotalus adamanteus, for example, could be released from its complex with a2-macroglobulin in the presence of serum or a high molecular weight proteinase fraction from the same venom [16]. Despite these cautions, the interaction with a 2macroglobulin offers the advantage of increased sensitivity over the hide powder digestion procedure, and the sensitivity of the method could be increased several-fold by replacing BAPNA with other trypsin substrates. We are now using this

procedure as the basis for assays of fibrinogenases during their purification from various cobra venom sources.

Acknowledgements We thank Drs. Judith Bond and: Joseph Liberti for reading, and Ms. Judy Watts for typing the manuscript. This work was supported by a grant from the Virginia Affiliate of the American Heart Association and by grant IN-105G from the American Cancer Society.

References 1 Henriques, O.B. and Evseeva, L. (1969) Toxicon 6, 205-209 2 Kocholaty, W.F., Ledford, E.B., Daly, J.G. and Billings, T.A. (1971) Toxicon 9, 131-138 3 Mebs, D. (1968) Hoppe-Seyler's Z. Physiol. Chem. 349, 1115-1125 4 Murata, Y., Satake, M. and Suzuki, T. (1963) J. Biochem. Tokyo 53, 431-437 50shima, G., Sato-Ohmori, T. and Suzuki, T. (1969) Toxicon 7, 229-233 6 Robertson, S.S.D., Steyn, K. and Delpierre, G.R. (1969) Toxicon 6, 243-245 7 Evans, H.J. (1981) Biochim. Biophys. Acta 660, 219-226 8 Bradford, M. (1976) Anal. Biochem. 72, 248-254 9 Rinderknecht, H., Geokas, M.C., Silverman, P. and Haverback, B.J. (1968) Clin. Chim. Acta 21, 197-203 10 Ganrot, P.O. (1966) Clin. Chim. Acta 14, 493-501 11 Kress, L.F. and Parosld, E.A. (1978) Biochem. Biophys. Res. Commun. 83, 649-656 12 Kress, L.F. and Catanese, J. (1980) Biochim. Biophys. Acta 615, 178-186 13 Barrett, A.J. and Starkey, P.M. (1973) Biochem. J. 133, 709-724 14 Tu, A.T., James, G.P. and Chua, A. (1965) Toxicon 3, 5-8 15 Geiger, R. and Kortmann, H. (1977) Toxicon 15, 257-259 16 Kress, L.F. and Kurecki, T. (1980) Biochim. Biophys. Acta 613, 469-475