THE JOURNAL OF BIOLOGICAL CHEMISTRV Vol. 258, No. 20, Issue of October 25, pp. 12566-12573.1983 Printed in U.S.A.

Kallikrein-like Enzymes from Crotalus atrox Venom* (Received for publication, January 17, 1983)

Jon B. BjarnasonS, Amy Barishs, Greg S. DirenzoQ, Ronald Campbell§, and Jay W. Fox87 From the $Department of Chemistry, University of Iceland, Rekyauik, Iceland and the $Department of Microbiology, University of Virginia Medical School, Charlottesville, Virginia 22908

The symptoms which immediately follow envenom- ation by many crotalid snakes include hypotension, hypovolemia, hemoconcentration, and shock. We have isolated and characterized two proteases (E1 and EII) from the venom of Crotalus atrox which may be in- volved in the onset of these symptoms. E1 and E11 have molecular weights of 27,500 and 29,200 and isoelec- tric points of 4.7 and 4.3, respectively. Specific ester- olytic activities of E1 and E11 on N"-p-tosyl-L-arginine methyl ester are 51.5 pmol min"mg" and 48.1 pmol min"mg", respectively. Both enzymes are rather spe- cific in their substrate requirements in that neither was demonstrated to have any proteolytic activity against either of the oxidized chains of insulin, or glucagon. Neither enzyme was shown to have plasmin or fibrinolytic activity. Both enzymes are able to cleave a kininogen analog to release bradykinin. This proteo- lytic activity is inhibited by aprotinin and phenylmeth- anesulfonyl fluoride but not by ethylenediaminetet- raacetate. The enzymes are active upon the kallikrein substrates 52666 and 52302. The K , values of the enzymes with these substrates are similar to those reported for kallikrein. Structural similarity between the two enzymes was demonstrated by ultraviolet and circular dichroic spectroscopy, and amino acid analy- sis. Tryptic peptide mapping of the two native enzymes also suggested a large degree of structural similarity. Furthermore, sequence studies on the NHz-terminal regions of the enzymes indicate that they share a sig- nificant degree of sequence homology with porcine kallikrein and crotalase, a kallikrein-like enzyme from Crotalus adamanteus. The main physical differ- ence between the two kallikreins reported here ap- pears to be due to the carbohydrate moieties on the enzymes. At present the in vivo role of venom kalli- kreins in envenomation pathology is uncertain; how- ever, it is possible that they play an important part in giving rise to the initial symptoms of hypotension and shock.

The Western diamondback rattlesnake (Crotalus atrolc) is indigenous to a large area spanning the Southwestern United States. Upon envenomation, there are marked effects on the victim's cardiovascular system, respiratory system, somatic nerve system, and skeletal muscle (1). Since death is a rela- tively uncommon consequence of envenomation, much em-

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

IT Recipient of National Institutes of Health Grant GM 31289-01. To whom correspondence should be addressed.

phasis has been placed on research involving the dramatic local effects caused by envenomation such as hemorrhage, myonecrosis, inflammation, edema, and pain (2-7). Another area of focus by investigators has been on the systemic effects of crotalid envenomation. These effects play an important role in giving rise to hemorrhage, hypotension, coagulation, hemolysis, and hemoconcentration. All of these factors serve to produce the overall symptom of crotalid poisoning some- times called rattlesnake venom shock (8). It is the rapid onset of venom shock which probably plays a major role in prey immobilization and possibly death.

In the past, there have been reports of the hypotensive nature of certain crotalid venoms (9-11) and also of the isolation of kallikrein-like enzymes from viper and crotalid venoms (12-14). It has been proposed that these venom kallikreins along with other hypotensive factors in the venom serve in the production of venom shock (8). In this report, we discuss the isolation from the venom of C. atrox of two proteases with similar structural and functional properties as certain snake and mammalian kallikreins.

EXPERIMENTAL PROCEDURES AND RESULTS'

Amino Acid Composition and NHz-terminal Sequence Anal- ysis-Table I contains the amino acid composition of E1 and EII. As can be seen, the compositions of each protein are nearly identical, suggesting homology between the two pro- teins. The total number of residues/molecule for E1 and E11 based upon the estimated protein fraction of the molecular weights are 216 and 219 residues, respectively.

Fig. 1 shows the NH,-terminal sequences of E1 and E11 compared to another snake venom protease, crotalase (from Crotalus adamanteus venom), and porcine a chain kallikrein. In each case, the HPLC analysis of the first Edman degra- dation cycle on E1 and E11 showed the presence of only one PTH derivative (see Table VI in Miniprint). The NH,-ter- minal sequences of E1 and E11 are identical to each other up to residue number 21. A notable degree of sequence homology is also observed between EI, EII, crotalase, and kallikrein (15, 16).

Portions of this paper (including "Experimental Procedures," part of "Results," Figs. 2-13 and Tables 11-VI) are presented in miniprint at the end of this paper. The abbreviations used are: PTH, phenyl- thiohydantoin; TAME, N"-p-tosyl-L-arginine methyl ester; PMSF, phenylmethanesulfonyl fluoride; HLPC, high performance liquid chromatography; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate; TPCK, L-1-tosylamido-2-phenylethyl chlo- romethyl ketone. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 83 "133, cite the authors, and include a check or money order for $12.00 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

12566

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Kallikrein-like Enzymes from Crotalus atrox Venom 12567

TABLE I Amino acid composition of EI and EII

Values are the result of triplicate analyses at 24, 48, and 72 h hydrolysis time. Total residues/molecule based upon estimated pro- tein molecular weight for E1 and EII.

E1 El1

Residues y!::: CMC 10.1 10 Aspartic acid 27.2 27 Threonine" 11.6 12 Serine" 13.9 14 Glutamic acid 16.6 17 Proline 18.9 19 Glycine 19.6 20 Alanine 12.5 13 Valine 13.6 14 Methionine 2.5 3 Isoleucine 13.8 14 Leucine 14.3 14 Tyrosine 4.6 5 Phenylalanine 8.2 8 Histine 7.1 7 Lysine 13.5 14 Tryptophanb 1.7 2 Arginine 2.7 3

Total residues 216

Residues

9.7 10 27.0 27 11.5 27 14.2 14 17.7 18 19.0 19 21.7 22 13.7 14 14.6 15 2.6 3

13.7 14 14.6 15 5.3 5 9.3 9 4.8 5

13.3 13 1.3 1 3.1 3

219 ~

Corrected for decomposition by extrapolating to zero time. Determined spectrophotometrically.

Amino Terminal Sequence Comparison of E l and E l l Wi th S imi lar Proteases .I" NHZ-8-V-GGD-E-C-WI-WE-B-R-S-L-V-A-I-F- T-E-F-F-

FIG. 1. NHz-terminal sequence comparison of E1 and E11 with similar proteases. Identical sequences are boxed (see Mini- print Table VI).

DISCUSSION

The overall mechanism of the toxic action of rattlesnake poisoning is somewhat unclear due to the complex nature of crotalid venoms. In general, rattlesnake venoms are not strongly neurotoxic with the exceptions being venoms from Crotalus durissus terrificus and Crotalus scutulatus. This is in sharp contrast to the venoms from Hydrophiidae and Elapidae snakes which are extremely neurotoxic. The primary, initial, overt action of crotalid envenomation seems to be onset of venom shock symptoms such as hypotension and hypovole- mia, followed by rather considerable local tissue damage at the site of envenomation. To date, significant progress has been made in the understanding of the mechanism and factors involved in local tissue damage.

The initial transient hypotension that is common in rat- tlesnake envenomation is probably due, for the most part, to two types of factors present in the venom; one of the factors is the presence in some crotalid venoms of angiotensin-con- verting enzyme inhibitors. These interesting peptides act by inhibiting the conversion of angiotensin I to angiotensin I1 by the converting enzyme and thereby additionally serve to potentiate the pharmacological actions of bradykinin (31).

The other group of important factors involved in hypoten- sion is the kallikrein-like enzymes. Bradykinin has been dem- onstrated to be released by the proteolytic action of venom kallikreins on bradykininogen in plasma, intestine, uterus, and smooth muscle (31, 32). Additionally, bradykininogen levels have been shown to be decreased following rattlesnake envenomation (9). Prior to this investigation, there have been reports of at least two well characterized snake venom pro- teases which were identified as having kallikrein-like enzy- matic activity. One of these proteases was isolated from the venom of Vipera ammodytes ammodytes (12). This kallikrein was shown to be a glycoprotein of molecular weight 34,300. The protease had an isoelectric point of 7.2 and was six times as active as trypsin in releasing a kinin from plasma kinino- gen. As to whether the kallikrein released was lysyl-bradyki- nin or bradykinin was not discussed. Another kallikrein-like enzyme called crotalase, which was originally identified as a thrombin-like enzyme, has been isolated from the venom of C. adamanteus (13). Some of the distinguishing properties of this enzyme are molecular weight of 32,700, glycoprotein, serine esterase, and inhibition by specific chloromethylketone kallikrein inhibitors (13,33). This enzyme was shown to make up approximately 0.23% of the crude venom (13).

The two kallikreins (E1 and EII), as we now term the enzymes, isolated from C. atrox venom show some similarity to the enzymes mentioned above, particulariy crotalase. E1 and E11 are also very similar with regard to each other. The amino acid compositions and NH2-terminal amino acid se- quences of both enzymes are similar. The conformations of E1 and E11 as examined by UV and CD spectroscopy also appear similar to a degree. However, the spectroscopic studies did demonstrate some differences in the fine structure of the two enzymes' conformations. These similarities of native con- formations were further demonstrated by tryptic digestion studies on the two native enzymes. The HPLC elution profiles of the digestions were overall notably alike although at least one major difference was observed. As to whether this differ- ence is a result of a slight conformation difference due to different primary structures or different carbohydrate moie- ties between the two enzymes is at present unclear. However, this study has shown that there are both qualitative and quantitative differences in the carbohydrates present in these enzymes. The effects these may have had on the mapping and spectroscopic studies is uncertain.

The two enzymes share like specificity for releasing the same peptides in identical order from the KS-1, KS-2, and KS-3 substrates. However, when their activities were exam- ined with the chromogenic kallikrein substrates S2266 and S2302, some kinetic differences became evident. The relative reaction rates (normalized against trypsin) of E1 on both substrates were nearly identical whereas E11 demonstrated a higher rate with S2266. E1 had nearly identical K,,, values with both substrates; however, E11 had a larger K,,, with the glandular kallikrein substrate S2302 compared to the plasma kallikrein substrate S2266.

The C. atrox kallikreins seem to share some biochemical properties with the kallikreins from V. ammodytes ammodytes and C. adamanteus. The molecular weight of V. ammodytes kallikrein is 34,300 and crotalase is 33,000 compared to the molecular weights of 27,500 and 29,200 for E1 and EII, re- spectively. EI, EII, v. ammodytes kallikrein, and crotalase are all glycoproteins although they do not share exactly the same carbohydrate moieties. Both V. ammodytes kallikrein and crotalase are sialoglycoproteins whereas E1 does not contain sialic acid. EII, however, does contain sialic acid. With regards to the proteases' presence in crude venom, crotalase comprises approximately 0.23% by weight of the crude venoms compared

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

12568 Kallikrein-like Enzymes from Crotalus atrox Venom

to 0.13% and 0.26% for E1 and EII, respectively. Finally, EI, EII, crotalase, and porcine kallikrein do possess homologous amino acid sequences in the NH2-terminal region, and, as more sequence data are reported on both C. utrox kallikreins and crotalase, it is likely further sequence homology will be uncovered.

One very interesting difference between E1 and E11 and crotalase is the strong thrombin-like activity demonstrated by crotalase whereas no significant clotting ability was ob- served with either E1 or EII. In light of the many similar biochemical properties of EI, EII, and crotalase, far from insignificant is the observed amino acid sequence homology. The comparison of the complete primary structures of EI, EII, and crotalase may well lead to very interesting findings from the proteins’ structures which shed light on the dual enzymatic activities of crotalase and the mechanism of action of E1 and EII. Additionally, it is noteworthy that these two crotalid snakes are closely related taxonomically, and there- fore it is not surprising that the two snakes share similar toxins in their venoms. Consequently, in the future, it will also be interesting to examine what similarities and differ- ences the other toxins in the respective venoms possess and their effects in the overall pathological nature of the venoms.

Finally, the isolation of kallikrein enzymes from crotalid venoms further emphasizes that one of the important aspects of crotalid poisoning is the hypotensive venom shock which occurs almost immediately following envenomation. It is this aspect of envenomation which probably plays an important, immediate role in debilitating the victim, which is then fol- lowed by the local effects at the site of envenomation. Further investigation of these and other kallikrein-like enzymes in crotalid venoms may lead to a more complete understanding of the nature of crotalid envenomation.

REFERENCES

1. Hawgood, B. J. (1982) in Rattlesnake Venoms (Tu, A. T., ed) pp.

2. Ownby, C. L. (1982) in Rattlesnake Venoms (Tu, A. T., ed) pp.

3. Cameron, D. L., and Tu, A. T. (1977) Biochemistry 16 , 2546-

4. Bjarnason, J. B., and Tu, A. T. (1978) Biochemistry 17 , 3395-

5. Fox, J. W., Elzinga, M., and Tu, A. T. (1979) Biochemistry 18 ,

6. Ownby, C. L., Bjarnason, J . B., and Tu, A. T. (1978) Am. J .

7. Fabiano, R. J., and Tu, A. T. (1980) Biochemistry 2 0 , 21-27

121-162, Marcel Dekker, New York

163-209, Marcel Dekker, New York

2553

3404

678-684

Pathol. 93,201-210

8. Carlson, R. W., Schaeffer, R. C., Whigham, H., Michaels, S., Russell, F. E., and Weil, M. H. (1975) Am. J. Physiol. 229 , 1668-1674

9. Russell, F. E. (1965) Toxicon 2, 277-279 10. Sosa, B. P., Alagon, A. C., Possani, L. D., and Julia, J. Z. (1979)

Comp. Biochem. Physiol. B Comp. Biochem. 46,451-484 11. Rocha e Silva, M., Beraldo, W. T., and Rosenfeld, G. (1949) Am.

J. Physiol. 156, 261-272 12. Bailey, G. S., and Shipolini, R. A. (1976) Biochem. J. 153, 409-

414 13. Markland, F. S., and Damus, P. S. (1971) J. Biol. Chem. 246 ,

14. Viljoen, C. C., Meehan, C. M., and Botes, D. P. (1979) Toxicon 17, 145-154

15. Pirkle, H., Markland, F. S., Theodor, I., Baumgartner, R., Bajwa, S. S., and Kirakossian, H. (1981) Biochem. Biophys. Res. Com- mun. 99,715-721

16. Tschesche, H., Mair, G., Godec, G., Fieldler, F., Ehret, W., Hischauer, C., Lemon, M., Fritz, H., Schmidt-Kastner, G., and Kutzbach, C. (1979) Adv. Exp. Med. Biol. 120A, 245-260

17. Humrnel, B. C. W. (1959) Can. J . Biochem. Physiol. 37 , 1393- 1399

18. Tomana, M., Niedermeier, W., and Spivey, C. (1978) Anal. Biochem. 89, 110-118

19. Bradford, M. (1976) Anal. Biochem. 72 , 248-254 20. Laemmli, U. K. (1970) Nature (Lord.) 227, 680-685 21. Weber, K., Pringle, J. R., and Osborn, M. (1972) Methods En-

22. Winter, A,, Ek, K., and Anderson, U.-B. (1977) LKB Application

23. Moore, S., and Stein, W. H. (1963) Methods Enzymol. 6, 819-

24. Edelboch, H. (1967) Biochemistry 6 , 1948-1954 25. Schumacher, G. F. B., and Schill, W. B. (1972) Anal. Biochem.

26. Lundblad, R. L., Kingdon, H. S., and Mann, K. G. (1976) Methods Enzymol. 45,156-176

27. Claeson, G., Aurell, L., Friberger, P., Gustavsson, S., and Karls- son, G. (1978) Huemostasis 7, 62-68

28. Fox, J. W., Barish, A., Sydner, C. E., and Benzinger, R. (1982) Biochem. Biophys. Res. Commun. 106,265-269

29. Chen, Y.-H., Yang, J. T., and Martinez, H. M. (1972) Biochem- istry 11,4120-4131

30. Greenfield, N., and Fasman, G. D. (1969) Biochemistry 8,4108- 4116

31. Ferreira, S. H. (1966) in Hypotensive Peptides (Erdos, E. G., Back, N., and Sicateri, F., eds) pp. 356-367, Springer-Verlag, New York

32. Nustad, K., Johansen, L., Orstavik, T. B., and Pierce, J. V. (1980) in Enzymatic Release of Vasoactive Peptides (Gross, F., and Vogel, H. G., eds) pp. 89-100, Raven Press, New York

33. Markland, F. S., Kettner, C., Schiffman, S., Shaw, E., Bajwa, S. S., Reddy, K. N. N., Kirakossian, H., Patkos, G. B., Theodor, I., and Pirkle, H. (1982) Proc. Natl. Acud. Sci. U. S. A. 79,

6460-6473

zymol. 26,3-27

Note, No. 250, LKB-Produkter AB, Bromma, Sweden

831

48,9-26

1688-1692

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Kallikrein-like Enzymes from Crotalus atrox Venom 12569

sw1-t to

Kallikreil-like f m rnfalus Atrm v- J c h B B j m m . An, &ish, Grq S Dir-o and Jay W. Fm

naterlal - Lyaphillred crude venom was purchased from Nlaml Serpentallurn Laboratories. Diethy l l tminoethy l se l lu lose (DE-321 was purchased from whatman B10Chemlfa16, LTD. Sephadex G-75-40 and Pharmalytes vere Obtalned from

va8 purchased from Pierce Chemical Co. as were the quantitative RH-amxno

Pharmcxa Fzne Chemlcale. p-Aminobenzanidlne coupled to 4; beaded agarose

acid standards. Molecula~ weight StandardB. PIISP, fibrinogen, T M E , oxidized A chaln Of insulln, glucagon, and aprotinin vere from Sigma Chealcel Co. Chromogenic substrates 52266 and 82302 were from Helena Laboratories. OXldlZed lnsulln B chain was from SchYartz-Man". Neuraminidaee and trypsin

were from Beckman. Kallikrein model substrates KS-1, Ks-2, and Ks-3 vere the ITRTPCFJwere from Worthington Biochemicals. A 1 1 protein eequenclng chemicals

kind qlft Of PrOfe860r John M. Stewart. Polybrene was fcom Aldrlch Chemicaln. The u B m d a p a k C18 HPLC eoluons were from Waters. ~ 1 1 HPLC solvents were of HPLC grade and all other reagents and solvents wece of analytical grade.

I and TT formed at 2-4'C. except for the affinity chromatography which yo8 done at

- Step 1. A11 fractionation steps vere per-

room temperature. h o 109 prtions of Cmtalu #.KU venom were each dis- solved In distilled rater and eztensively dialyzed against I O M borate buffer, pB 9 . 0 , containing 0.111 NaC1 and 2 M CaC12. The first fractionation step vas on a column 12.5 x 35cm) of whatman DE-32 operating at a flow rate Of 85ml/h. The starting buffer vas the Bame as the dialysis buffer. The

buffer. The pooled fractions vere assayed for esterolytic activity on T M E column was eluted with a step gradient Of 0.1M and 0 . 4 1 NaCl in the starting

1171 and thoae WOla containing the greatest activity were Collected and lyo- phllzred.

Step 2. The combined A-1 pools from the tvo anion exchanger chromato-

NaCl and 2mn Cacl2. This sample was applied in three separate runs to a

graphies were dissolved ~n 75nl of 5mn Tris buffer, pH 8.5 contalnmq 0.1n

column of Sephadex G-75-40 15 x 9 5 c d equilibrated with the same buffer and operating at a flow rate Of 35ml/h.

lomn TTIS buffer, pH 8.5 containing 40mll NaCl and 2 M CaC12 and fractsonated Step 3. Pool 8-2 13 conblned pools of 8-2) Yo8 equilibrated against

by afflnity chromatography on a column 11.5 x 2OcmI of paminobenzamldlne 4% cross l m k e d agarose developed with Q gradient of the diseolutmn buffer and 0.111 acetic acid. The column flov rate was 30ml/h.

Step 4. The pooled and lyophilzzed fractions from Step 3 IC-11 Yere

equillbrated ulth lOm Tris buffer pH 8.5 contalnlng 4 0 m NaC1 and 2 m CaC12 and applied to a column 11.5 x 40cmI of DE-32. The column was developed with a l l n e a r salt gradlent 1" the starting buffer of 4 0 m t o 1OOm NaCl. total volume 2000ml. The column flow rate was 45ml/h.

to rechromatography as described in Step 4. Step 5. Fractions 0-2 and D-3 were each divided in half and sUb2ected

- 5 " M d method of Tomana et al. 1181. Protein was quantitated by the method Of

- Carbohydrates were assayed by the

Bradford 1191 . Allquots lcontaining 1Ovg protein) of stock solutions of the esterase were lyophilized then dissolved in 0.68N methanollc HC1 150~11. The reactlon tubes yere sealed and incubated at BO'% for 18h after which the two

drlde in ethylacetate vas added to each tube. The samples were then mleeted reactlon mixtures were dried under vacuum. 5 0 ~ 1 Of 6% trifluroacetlc anhy-

into a 5830A Xevlett-Packazd qas chroaatoqraph vlth electron Capture detec- tion device. The column packrng was 21 OV-105 on Chromosorb ISupelcoI. The resultant peaks were identified by a conparlson to a standard mixture of the follormq: mannose, galaCtoBe, fructose, N-acetyl glucosamine, sza11c a c ~ d , and fucose.

ts Of E 1 - Homo-

qenexty of the two proteases was aesertained by polyacrylamide gel electro-

The Tr1s PAGE Stacking gel was 2.5% CcOsS-llnked vlth a pH Of 7.65 and the phoremls both with and without SDS 120, 21) and also by lsoelectrlc focusing.

reso lv lnq gel was 7.5% cross-llnked wLth a pH Of 0.90. The Trls-PAGE-SDS stacklng gel was 4% cross-llnked at pH 7.12 and the resolvrng gel was 12.5% crass-lrnked at a pH Of 8 . 9 0 . The gel6 Were stalned with a Co)oma651e

blvelmethanol acetic acid solution. All polyacrylamlde gels were run Xlth lncieaslng Concentcations Of the proteases in order to detect the presence of any 611ght Impurities. Isoele~tr~c focY61ng was performed accordrnq to the procedure of Wlnter et a l . 1221 on a hOLlZOntl1 polyacrylanlde slab. A mrx- ture Of Pharaalytes was used Such that a pH gradient from 4 to 0 vas formed. Follovlng flxlng, the foCuSed proteins Yere visualized vlth a CoOmasSle blue s t a m . The pH gradient was deterrnmed by measuring the pH Of 5 x 5 m 511ces

focublng Of standard protelns vlth known isoelectrlc points.

from the gel. The effectiveness of the gradlent was conflrmrd by the

Ta demonstrate the presence of blallc acid ln E11 the protease 1 1 0 O ~ g I waa Incubated vlth 5049 of neuram~nldase 116Ulmgl ~n l m l of 5 m sodlurn acetate buffer pH 5.6. 0.5% NaCl, 0.1% CdCI2 for 4Oh at 37OC. Control d1ges- tlons of E1 ulth neuraminidase were also performed. Following the incubatron the samples Yere desalted by ultra-filtration and then SubIected to SDS-PAGE as demcrlbed above.

The amlno acid compositions Of the proteases were determlned by the method D f Moore and Steln 1231. 50ug samples of the proteases were dissolved In 50011 Of 5.7N HC1 contalnlng 0.3% phenol and 0.2% 8-mercaptoethanol. The

All hydrolyses yere performed in trlpllcdte for each hydLo1ysls tlme. tubes were evacuated, sealed, rhen hydrolyzed at llOoc for 24, 4 8 . and 72h.

Tryptophan was deternlned spectrophot~~et~lcally 1241.

f E 1

proteases were d16601Ved ln lml Of lomn TILS, pH 8.5, 2mM CaC12, at a concen- - For the dzgestions the

tration of Img/ml. Trypsin ITPLCKI vas added to the digestIan mlxture to give a ratlo 1w:wl Of 1:50 1trypsln:esterasel and the dlqestion m x t u r e m c u - bated at 37'C. Aliquot. of lOOul were taken at t m e d intervals and then acidtfied vzth glacial acetic acld to a pH Of 2.0. The allquots were then

developed wzth a two buffer system: A, 0.1% TFA ~n H20; 8, 0.11 TFA ln aceto-

in3ected Into a Y Bondapak C18 column 10.39 x 3Oclol. The column 1 8 6

nitrile. A llnear gradlent in buffer B was plodwed IAl%/minI wlth a Beckman 324 liquld Chcomatograph.

buffer, pH 0.5. 2mn CaC12, lOmM NaCl at a concentration Of lrnglrnl. The pep- - - The proteases yere each dissolved ~n 10mM Tr1s

of Imglml. The protease vas added to the substrate dlgestmn mxture so as

tide substrates were dissolved in lml of the same buffer at a concentratlo"

carried out at 37% vlth 1 0 0 ~ 1 allquots removed at tlrned mtervals and ana-

to make an enzyme to peptide Substrate ratlo 1 w : w l Of 1:50. InCUbatlonS were

lyred by HPLC on the C18 ~everse phase column system as described above. A11 digestions *ere done 1" duplicate a6 were the chromatographies. In these

acid analysls in order to ascertain the b l t e of peptide bond cleavage. The experiments eluted fractions were collected for acid hydmly616 and m l n o

gIu6agoni end the model kallikrein sUb6trateS, KS-1, lAc-Se~-LeU-Met-LyS-A~q- peptlde 6UbStrateS examined were: Oxldlzed A and B chalne Of bovlne ~ n s u l ~ n ;

PIO-P~O-Gly~Phe-Ser-Pra-Phe-Rrg-Ser-Val-Gl~-V~l-S~~-NH2l, KS-2 IAc-Sei-Leu- Met-Ly6-Arg-PrO-Pr~-GlY-NH21. and KS-3 IAc-Phe-Ser-Pro-Phe-Arg-Ser-Yal-Gln- Val-Ser-NH21.

RS-1 by the proteases vas also exammed. The digeetlon mlXtUre6 were pre- The sbillty Of certaln protease LnhlbLtorS to lnhlblt the c l e a v a g e of

pared ais above rlth the addltion of one of the follovlng mhlbitars: 1 O m n

EDTA, 101111 PMSF, or aprotlnin 117U/mlI. Aliquot8 110OUll were taken at 4h for analysis by HPLC for the presence Of dqestlon peptides.

. . . for fibrlnolytlc enzymatic activity according to the method of Schumacher and

- Both proteases were assayed

Schlll 125). The fibrin clot formation dCtlVltY of both enzymes vas aseayed accordinq to the method of Lundblad et a l . 1261.

f E 1 and - E 1

according to the method Of Claeson et al. (27). The two substrate= used I"

and E11 activity on two kallikrein chromqenrc substzates was assayed

the assay were s-2266 I D - V s l - L - L e ~ - L - A ~ g - p - " ~ t ~ ~ ~ " ~ l ~ d ~ l and 5-2302 ID-Pro-L- Phe-L-Arg-p-nltrOanll~d~l.

f EX and ELI - h ~ n o terrnznol primary stcuctuce ana-

both erper~ments the proteases were reduced and ca~boxylmldomethyloted 151 prior to application into the pinning cup. For a typlcal sequencer run 3m9

of POlyblene and 5nmioles Of glycylalanine Yere applied to the splnnlnq C U P

was applled followed by 1 cycle of the 0.11 Quadrol peptlde program followed by 3 cycles of Edman degradation. Ten Dmoles Of the proteln sample

18eckma.n-1211781 in vhlch phenylisothlocyanate was not introduced lnto the cup. Polloring this precycle, sequencing vas carrled Out and the phenylthm- hydantoins analyzed as previously described 1281.

lyses were performed on both E1 and EII with a Beckman 890C Sequencer. In

- The w spectra of the prOteln6 were obtalned at 25OC with a CarY Model 110 W - V I S spectrophoto- meter. The proteins Yere d16601Ved In 2 M Tris pH 7.9 containing 1OmM CaC12 and l O O M NaC1. A JASCO Model J-41C BpeCtrOpolarimeter was used to obtaln the circular diChrOlC spectra at 25%. The buffer used for the CD spectco6-

were used rhen measuring circular dichroxsm ~n the peptlde reglo". 5.0Omm copy was the same as in the w spectroscopy. Bath 2.00m and 0.505mm cells

cells yere used for measurements I" the aromatic reqlon.

BenvlLa

UMaLXm - The Initla1 DE-32 fractlonatlon Of the Crude venom gave an identical elutlon profile a16 earlier reported (see ref. 41. The a r g z n l n e

esterolytlc activity "06 confined to frDCtlOn A-1. The gel flltratlon chrom- atography proflle of pooled fraction A-1 on Sephadex G-75 1s seen 1" Flgule 2. The arginine esterolytic OCtlYity Vas Cunflned to fractlon 8-2. The elu- tlon praflle of fraction 8-2 on the p-ammabenramldlne Sepharose column 15

seen I" Flgure 3. The esterolytic actlvlty was Prlsarlly located I" fractlon C-1. The DE-32 a n m n exchange chromatography of peak C-1 1s seen ~n F ~ g u r e 4. Peaks 0-2 and 0 - 3 COntalned the highest ebtecOlYtiC dctlvlty. The DE-32 rechromatography of peaks 0-2 and 0-3 ace seen x" Flgures 5 and 6 respec-

C . - venom 51.3rnq and 26.1m9 of E1 and E11 respectively were Isolated. tlvely. The lsolatlon results are summarlzed ln Table 11. Fzom 209 of crude

The specrfLc activity on TAME for E1 and E11 was 51.1u/mg and 40.lu/mq respectively.

bohvdrate CDmWSltlOnS of E1 and EII - From SDS, and native PAGE the two pro- teases appeared homogeneous and thelr molecular welghts were estlmated to be

27,500 and 29,200 for E1 and E11 respectlveiy (Figure 71. From the 1 ~ o e I e c -

and 4.3 respectively. From SDS-PAGE analyh16 lncubatlon of E1 vlth neycamln-

trlC focuslng experiments the PI'S of E1 and E11 Y e ~ e determlned to be 4 . 1

dase did not appear to have any effect upon the mlecular welght of EI. HOW-

ever, follorlng mcubatxm wlth neurarnlnidase, E11 was Observed from SDS-PAGE to mlqrate at an apparent mOlecYlar weight of 20,600 IFlg. 0).

F r m the gas Chmrnatographlc carbohydrate analyses both E1 and EII were

shown to contaln fucose, mannose, galactose. and N-acetylqlucosamlne. 1n addltlon to the above carbohydrates, E11 also C o n t a m s s ~ a l l c acld [Table

E1 and EII, the molecular weights Of the protein portions of the enzymes were

1111. Follovlng the ldentlflcatlon and quantitatlOn of the carbohydrate8 I"

estlmated to be 24,400 and 24.000 respectively.

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

12570 Kallikrein-like Enzymes from Crotalus atrox Venom

I STEP 2

5 I E-2

.40

. E - 30 2

- a

- 20 s "

- 10

20 80 140 200 260 320 380 440 FRACTION NUMBER

STEP 3 6 .

5 .

c-2

a 3 i'l c-l

2

FRACTION NUMBER

Pig. 5. Rechromatoqrsphy of peak 0-2 on DE-32. Condltlonb same as FIG. 4

E 0.

OD

4 0.

STEP 5 b

PIg. 6. Rcehmmatography of peak 0-3 on DE-32. Condltrona same as Fig. I .

Table I1

Ruif lsnl iom of E1 a d E11 fr- E. - Cr& V e n a

101.1 101.1 Spsifis Eeeo*cv of Fold h i f i a t i o n R o 1 . i ~ k t i r i l y k t i r i l y k l l r i l 9 Based on k t l r i t y

Fr.ctio. (-1 (hits1 ( Q n i l d q l (S of l n i l i * l ) I'.insl TAN3

Cr& .e- 20.000 24.800 1.1 l W . 0 1 .o

k-I 6.009 22.631 3.8 91.3 3 .O

0-2 1.160 17.136 13.6 69.1 11 .o

e1 319 7.146 22.4 28.8 18.1

D-2 68.4 3.176 47.9 13.2 38.6

M 41.0 2.IM 46.8 8.5 37.7

EI 11.3 2.621 51.1 10.6 41.2

e n 26.1 1.215 48.1 5.1 38.8

CV El Ell CV

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Kallikrein-like Enzymes from Crotalus atrox Venom 12571

S A B C D E F - - " ".

Table 111. Carbohydrate Analysie Of El and E l l '

____~

QUEOSe 1.8 2 5.8

nannose 4.9 5 1.9

GalaEtO*e 3.2 3 11.7

N-Ac-Glucosamine 6.3 6 1.6

sialrc Acid 0 0 0 . 5

'Determined according to the method of Tomana et a l . 11918).

4 2

a N

0.1

0

0 IO 20 30 40

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

12572 Kallikrein-like Enzymes from Crotalus atrox Venom The kininogen analog IS-1 Ya6 cleaved at two site8 by E 1 and E l l . The

following is the description of El digestion Of Is-1. At Oh msubation t m e the one malor peak at approximately 43 mi" vas identified by amino acld analysis as the intact KS-1 peptide (Figure 11 and Table VI. At 30 mln the intact KS-1 16 no longer present> however, four new peaks appeared. From amino acid analY616 of the four digentlon peptides, identifications weze made (Table VI. Peak 1 was the XS-1 terminal peptide AC-Sel-Le"-net-Ly6-A~9-P~~- PTO-Cly-Ser-PrD-Phe-Arg- resulting from the bond cleavage between -Arg13-

the first cleavage site l-Se~-V~l-Gln-V~1-Ser-NH21. In the 60 and 120 mi" Ser14-. Peak 2 was identified a6 the carboxyl terminal peptide fragment from

digeBt10ns one can see that the peak 1 peptlde 18 being dlgested to yield two new fragments located at Peaks 3 and 4 . Ammo acid analyaie identlfied peak 4 as the C-terminal peptide ~ - A T ~ - P ~ O - P ~ O - G ~ Y - P ~ ~ - S ~ ~ - P C O - P ~ ~ - ~ ~ T ~ ) from the peak 1 peptide resulting from the bond cleavage at Lys4-Arg5. Peak 3 Y L ~ B

identlfied ae the amino terminal fragment of the peak 1 peptide. NO further dzgestron was detected after 120 min. Both enzymes appeared to Cleave the two peptlde bonds Of Is-1 in the same order and with nearly ldentlcal rates.

4

3 10 20 30 40 50 TIME (min)

Fig. 11. HPLC of tlned digestions Of kininogen analog 15-1 by EII. Peptides

were analyzed by HPLC under conditions &dentleal to theee deserlbed ~n Fig. 9. Description Of peaks given I" text.

Val net I l e

1/2 cys 1.712)

0.8111 0.7(1) 1.912) 0.6111

Le" 1.111) 0.9111 0.9111 TYC Phe TI0

1.712) 2.0(21 1.8(21

H G LY Q

AZ9 1.OIll 0.9111 1.812)

0.111) 1.9121 1.812)

ReSldues Total

13 5 4 9 18

%ydrOlyS16 time Of 24-h. Integers in parentheses are expected theoretical values. Where no values are shown, the amino acid was not detected or present at a" amount too low to quantztate.

elther E1 or EII. EDTA did not affect the digestion of xs-I by El 01 €11.

with tryPBln, plasmin, or the Crude venom in the fibrmolysrs assays. A ~ B O

Neither E l nor E 1 1 demonstrated any fibrinolytlc activity when conpared

both enzyme8 were unable to produce typical fibrln Clot6 in the thrombln assay. However, after extended periods of time 0 4 h ) very small. atyplcal Clots were sometimes Observed. The rnechaniem of Productton of the atyplcal clots 16 unknown, therefore the thrombu-like actlvlty of E1 and Ell was con- sidered to be negllglble.

Apratinin and PMSF Were both able to inhiblt all cleavage Of kS-1 by

f E1 and appear t o be identical ~n peak positions. mtensltles, and flne structure

- The W Spectra Of both El and e 1 1

Ifig. not presented) The CD spectra for El and E11 do appear to have some r n ~ n o r dlffecences.

222nm with correspondtng mean residue velght eliptxcltes of -6811 and -5909. In the Peptide region (figure 121. the El spectrum has mlnlma at 211 and

219nm wlth mean resldue weight ellptlcltes of -6205 and -6176 respectively.

In the case Of E11 a minimum is observed at 212nm and a broad mlnimurn at

Both spectra appear LO represent protelna ulth predominantly O-helLcal struc- ture 129, 301. HDVeVer, a8 apparent from their CD spectra, the conformatlans of both protelns do differ to some small extent.

WAVELENGTH (NM)

fig. 12. Circular dichroism spectra of E l and EII In the Peptide Ceqlon. Buffer used was 2mn Tria. pH 7.9, lOmR CaC12, 0.1M NaCl. Spectra were recorded at 25OC. Rean residue elllptlclties lBmCy) a re based on mean resldue ueiqhts of 107 and 106 for El and E11 respectively.

The CD spectra In the aromatic reglan of E1 and E11 IFlgUre 131 are to a large degree a180 similar. In both spectra of the arornatlc r e g ~ o n three rnlnma are observed. In the El spectra the three malor n m m a are tound at 268, 214, and 280nn with molar elipticitles of -63,000, -60,500, and -57,300. The mnlrnurn at 268nm 1s likely due to an absorbance by phenylalanine: the

broad, shallow, negatlve band centered around 305nm 1s llkely the ieSYlt of 274nm and 28On0i peaks are probably due to tryosine. tryptophan, 01 both. The

dlsulfldea present LD the protelns. Overall, the appearance of the aromatic

molar e l l p ~ ~ c l t ~ e s being very similar. However, as I" the case of the pep- CD spectlum for E l l 16 qurte slmllar to that of EI, Ylth mlnlrna P051t10n and

tlde r e g ~ o n spectra. some minor flne Structural dlfferences are apparent. In both Instances the nature Of these differences are unknown. However the pro- teases, an observed vlth CD and W SpeCtrOBcopy, do appear to share very

the enzymes and suggests the posslblllty of ~soenrymes. similar conformations. This again p o m t s to the overall similarity between

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

Kallikrein-like Enzymes from Crotalus atrox Venom 12573

W A V E L E N G T H ( n m )

Fig. 13. Circular dichroism 6peCtra of El and E11 in the aromatic region. Buffer and temperatures same as in Pig. 11. U o h r ellipticities

and E l 1 respectively. leM) are based upon mo1ecula.r weights of 21,500 and 29,200 for E1

Table V I Edmo De8radatton of Amtoo Tcr.in.1 ue8tom of e l and €11'

cycle No.

m-A* E1

Y z c i d (nsol) €11 El €11

1 811 B i ,

2 Y.1 3

4

7.5 6.8

Y.1 7 . I 6.7

6.2

6.0

5.6

5.6

1 C Y - C Y , -CY* 5.9 5 .2 8 A*n *so

9 I l r I IC

5.9 S . I

5.8 5.4 10 A,n A m

11

5.1

Glo

s.l

Cl"

12 5 .6

E,* 81,

5 .2

5 .o 13 *'I 1.1 1.4

5.0

14 Ser SLT 4.0 l.Q

15

16

LC.

V.1

4.9

I7 AI.

4.5

18 I IC

4.0

4 . 2 4 . 1

19

20 Vi 1

Phr Phc 3.8 3 .l

21

3.8

SCr me 2 .2

3.6

3.1

12 Thr Tbr 1 .l 1.1

23 GI. GI" 3.1 2 . 8

GIY GIY 7 .O

G ~ Y G ~ Y 6.9

5 ASP ASP 5.9

6 Glu Glu 6.1

LIIl 5 . I

V.1 4.1

AI . 4 . 2

IlC

V.1

2 4 Phe Phe 2.7

2s n e Phr 1 . I

2 . I

1.2

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from

J B Bjarnason, A Barish, G S Direnzo, R Campbell and J W FoxKallikrein-like enzymes from Crotalus atrox venom.

1983, 258:12566-12573.J. Biol. Chem. 

  http://www.jbc.org/content/258/20/12566Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

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

  http://www.jbc.org/content/258/20/12566.full.html#ref-list-1

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

by guest on Decem

ber 25, 2020http://w

ww

.jbc.org/D

ownloaded from