Taxicar, VoL 30, No . 11 . pp. 1387-1397, 1992. Printed in Grat Britain.
C
0041-0101/92 ßI00 + .00 1992 Perpmoo Press Ltd
BROAD SUBSTRATE SPECIFICITY OF SNAKE VENOM FIBRINOLYTIC ENZYMES : POSSIBLE ROLE IN HAEMORRHAGE MASUG1 MARUYAMA, MAsAHIR0 SUGIKI, ETSUo YOSHIDA, KAZUHIIto SHmAYA HISAsm MniARA
and
Department of Physiology, Miyazaki Medical College, 5200 Kiham, Kiyotake-cho, Miyazaki-ken 889-16, Japan (Received 7 May
1992,
accepted
8
June
1992)
and H . MmARA . Broad substrate specificity of snake venom fibrinolytic enzymes : possible role in haemorrhage. Toxicon 30, 1387-1397, 1992.-We found previously that two fibrinolytic enzymes (jararafibrases I and II) purified from Bothrops jararaca venom displayed a haemorrhaggc activity . To elucidate the mechanisms involved and the role of the enzymatic activity in haemorrhage, the enzymatic properties of the purified enzymes were examined . The substrate specificity of the enzymes was determined using type I collagen, type IV collagen, gelatin, laminin and fibronectin as substrates . The enzymes degraded type IV collagen, gelatin, laminin and fibronectin into smaller fragments, but degraded type I collagen only partially in a non-specific manner. The specific activities of jararafibrase I for type IV collagen and gelatin were 172 t 5 units/mg protein and 1315 f 177 units/mg protein, respectively. The specific activities of jaraafibrase II for type IV collagen and gelatin were 9.2 f 0.6 units/mg protein and 143 f 15 units/mg protein, respectively. It was evident that the enzymes had rather broad substrate specificities and degraded basement membrane components including type IV collagen . The number of type IV collagen units of bacterial collagenase which gave the minimal haemorrhaggc dose was 191 .4, while the numbers of type IV collagenase units of jararafibrases I and II which gave the minimal haemorrhaggc dose were 1 .5 and 0.25, respectively. It is suggested that the broad substrate specificity of the enzymes is essential for inducing haemorrhage with a single enzyme . M . MARuyAMA, M . SuGml, E . YOSFMA, K . SHIMAYA
INTRODUCTION
haemorrhage and necrosis represent major complications following snake envenomation, especially in the case of Viperidae and Crotalidae snake envenomations. These snake venoms also provoke systemic coagulopathy with a decreased fibrinogen level (AuNG-KHIN, 1980; MARuyAMA et al., 1990). However, the pathophysiological mechanisms involved in the local haemorrhage induced by snake envenomation still remain unclear. Disruption of the capillary basement membrane has been observed in the haemorrhaggc area by histological techniques (OwNBY et al., 1990; KAMIGuTI et al., 1991). LocAL
1387
1388
M. MARUYAMA et al.
It has been reported that many of the purified haemorrhagic factors possess proteinase activities, especially fibrinolytic activity (iVIARKLAND, 1991). Local haemorrhage after snake envenomation is conceivably induced by disruption of the vascular basement membrane, occurring simultaneously with abnormal haemostatic systems. Enzymatic degradation of the vascular basement membrane is assumed to be a major cause of haemorrhage. However, it is unclear whether the degradation of the basement membrane is caused by a direct proteolytic effect of haemorrhagic factors in the snake venoms or by the victim's secondary activated proteinases following envenomation . We previously purified two fibrinolytic enzymes, which were suggested to be metalloproteinases, from Bothrops jararaca venom and found that each of them displayed a haemorrhagic activity (MARUYAMA et al., 1992). In the present study, we examined the activity of these fibrinolytio-haemorrhagic enzymes purified from Bothrops jararaca on various natural substrates, especially on basement membrane components, in order to clarify the direct effect of the enzymes on basement membrane degradation. MATERIALS AND METHODS Materials (p-Amidinophenyl)methanesulphonyl fluoride (APMSF) was purchased from Wako Pure Chemical Industries, Ltd, Osaka, Japan. Fibrinogen (bovine, 75% clottable) was purchased from Miles Inc., Kankakee, IL, U.S.A . ["C] Acetic anhydride was purchased from New England Nuclear Corp ., Boston, MA, U.S.A. Type I collagen (bovine achilles tendon), type IV collagen (human placenta), laminin (EHS mouse sarcoma) and bacterial collagenase (Type IA, Clostridhrn hirtolyticron) were purchased from Sigma Chemical Co ., St . Louis, MO, U.S .A . Fibronectin (human plasma) was purchased from Cosmo Bio Co., Ltd, Tokyo, Japan. Rat tail tendon type I oDllagen was purified by the method of SCHOR (1980). Venom source . A pool of lyophilized venom obtained from adult specimens of Bothrops jararaca maintained at the Laboratory of Herpetology, Instituto Butantan, was used . Methods Prrffrcation procedures. Jararafibrases I and II were purified as reported previously (MAAuyANA et al., 1992). The final products revealed a single protein band on analytical polyacrylamide gel electrophoresis (PAGE) and isoelectric focusing in agarose gel. The specific fibrinolytic activities of jararaflbreses I and II were 3.6 t 0.3 units/mg protein and 12.6 f 1 .2 units/mg protein, respectively . Protein concentration. Protein concentration was estimated by the method of LAwRY et al. (1951) employing bovine serum albumin as a standard . For determination of the type I and IV collagen concentrations, known amounts of type I and IV collagen were employed as standards. Assay offtbrinolytic activity. The flbrinolytic activity was measured using 0.6% bovine plasminogen-free fibrin plates (Moauaoro et al., 1991). For this purpose, 30 id of sample was placed on a fibrin plate and the lysis area was measured after incubation at 37°C for 18 hr. The specific activity was calculated from a standard curve for the lysis area obtained with planmin (Sigma Chemical Co .) on the plasminogen-free fibrin plates. One fibrinolytic activity unit was defined as being of the same magnitude as the lytic activity of 1 casein unit (CU) of standard planmin on a plasminogen-free fibrin plate. Assay of type l collagenase activity. The type I collagenase activity was measured according to the method of LnvDSLAD and Fur.rEt (1982) . Type I collagen (bovine achilles tendon) was labelled with "C by the method of Grsat.ow and McBrunr; (1975) . The specific radioactivity of the substrate was approximately 290,000 cpm/mg protein. Immediately before the assay, the '+C-labelled type I collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Trio-HCI buffer containing 10 mM CaCl 2 and 0.4 M NaCl, pH 7.5, and employed for the assay as the substrate solution . Then 100 id of substrate solution was mixed with 100 pl of sample . The resultant mixture was incubated at 35°C for 120 or 180min. The reaction was terminated by adding 100pl of 100 mM ethy ic acid (EDTA) containing 1.5 mg/ml of unlabelled rat tail tendon type I collagen . After incubation at 35°C for 30 min, 30D id of 1.4-dioxane/methanol (4 :1 by volume) was added, and the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 200 pl of the supernatant was mixed with 4ml of Aqueous Counting Scintillant II (ACS II, Amersham Japan, Tokyo, Japan) solution, and the radioactivity was counted in a liquid scintillation system (Aloka, LSC-700) . One type I collagenase unit was defined as the activity which degraded type I collagen at a rate of 1 pg/min at 35°C . Assay of type IV collggenase activity. The type IV collagenase activity was measured according to the method of L.raTA et al. (1981b). Type IV collagen (human placenta) waslabelled with "C by the method of Gmr.ow and
Substrate Specificities of Jararafibrases
1389
McBRiDE (1975). The specific radioactivity of the substrate was approximately 192,000 cpm/mg protein. Immediately before the assay, the "C-labelled type IV collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Tris--HCI buffer containing 10 mM Caa z and 0.4 M NaCl, pH 7.5, and employed for the assay as the substrate solution . Then 100 pl of substrate solution was mixed with 100 pl of sample and 200 id of 50 mM Tris-HCl buffer containing 5 mM CaCl2, 0.2 M NaCl, pH 7.5. The resultant mixture was incubated at 37°C for 60 or 180 min. The reaction was terminated by adding 125 pl of 10% trichloroacetic acid (TCA) containing 0.5% tannic acid, preceded by the addition of 100 pl of bovine serum albumin (0.2 mg/ml, GIBCO, Grand Island, NY, U.S.A.). After being allowed to stand in ice for 30 min, the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 400 id of the supernatant was mixed with 4 ml of ACS 11 solution, and the radioactivity was counted. One type IV collagenase unit was defined as the activity which degraded type IV collagen at a rate of I jug/min at 37°C. Assay ofgelathrase activity . The gelatinise activity was determined according to the method of MuRPrnr et al. (1980). The "C-labelled type I collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Tris-HCI buffer containing 10 mM CaCl2 and 0.4 M NaCl, pH 7.5 . The type I collagen was subsequently heat-denatured by incubation at 45°C for 15 min. Then 100 pl of substrate solution was mixed with 100 pl of sample solution and the resultant mixture was incubated at 37°C for 30 min. The reaction was terminated by adding 100 pl of 45% TCA. After being allowed to stand in ice for 10 min, the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 150 Id of the supernatant was mixed with 4ml of ACS II solution, and the radioactivity was counted. One gelatinise unit was defined as the activity which degraded gelatin at a rate of 1 ug/min at 37°C . Analysis by SDS-PAGE. First, 20 pl of substrate solution (1 mg/ml) was mixed with 5 id of enzyme solution containing 1 mM (p-amidinophenyl)methanesulphonyl fluoride (APMSF). The mixture was then incubated at 37°C for 4 hr in the case of laminin and fibronectin degradation analysis, at 35°C for 10 hr in type IV collagen degradation analysis, and at 35°C for 24 hr in type I collagen degradation analysis . The concentrations of the jaramfibrase I enzyme solutions were 95 #g/ml for laminin and fibronectin, and 16 pg/ml for type I and IV collagen . The concentrations of the jaramfibrase II enzyme solutions were 45 pg/ml for he minin and fibronectin, and 105 pg/ml for type I and IV collagen . The degradation patterns were analysed by SDS-PAGE according to the method of LAEmmr (1970), in the presence of 3 M urea . Staining was carried out with 0.2% Coomassie brilliant blue . Prestained protein mol. wt standards (GIBCO BRL, Gaithersburg, MD, U.S .A.) were employed for mol. wt estimation . Animal experiments. Female KUD Wistar rats, weighing 350-400 g, from the Experimental Animal Center at Miyazaki Medical College were used . The rats were kept at a controlled ambient temperature of 23 f 1°C with 50 t 10% relative humidity . Haemorrhagic activity was determined according to the method of TrmAKmN and RED (1983) . For this purpose, 50 Al of enzyme solution was injected intradermally into the rat dorsal skin under ether anaesthesia. The haemorrhaged areas were measured after 24 hr . The minimal haemorrhagec dose (MHD, 4rat) was defined as the sample amount which yielded a 1-cm diameter area of haemorrhage in the rat skin after 24 hr .
RESULTS
Type I collagen degradation
Jararafibrase I released radioactive peptides from "C-labelled type I collagen in a dose-dependent manner, as shown in Fig. IA. However, the linearity of the reaction was very poor . Jararafibrase II degraded type I collagen only partially (Fig. 1 B). As illustrated in Fig. 2, the control substrate solution after incubation at 35°C for 24 hr with 0.2 mM APMSF revealed a typical pattern of non-reduced type I collagen : two ß-chains (ß , ß,2), two a-chains (a,, a2) and very high mol. wt y-chain. No degradation was observed following incubation with jararafibrase II. On incubation with jararafibrase 1, the amount of ßchains was decreased. However, no further degradation of a-chains was observed . Gelatin degradation
Jararafibrases I and II degraded gelatin very strongly . The specific activities of jararafibrases I and II for gelatin were 1315 t 177 units/mg protein (n = 8) and 143 t 15 units/mg protein (n = 4), respectively (Table 1).
1390
M . MARUYAMA et ai.
3,000 -1
Z,
3,000 -
A
2 .000 -
2,000 -
1,000 -
1 .0()0-
B
O
0
0 0
50
1100
0
25
50
Enzyme concentration (jug/mi) FIG .
1.
CLEAVAGE OF RADIOLABEL L E, TYPE Each
point
I
oou.AGEN By jAaARAFiBRAsE
II (B).
represents the average value
of two
I (A) AND JARARAFIBRA.S E
identical experiments .
Type IV collagen degradation Jararafibrase I degraded type IV collagen very strongly and the specific type IV collagenase activity was 172 f 5 units/mg protein (n = 4). Jararafibrase II exerted a rather weak activity on type IV collagen, the specific activity being 9.2 f 0.6 units/mg protein (n = 4) (Table 1). As shown in Fig. 3, on SDS-PAGE, the control substrate solution incubated at 35°C for 10 hr with 0.2 mM APMSF revealed three main protein bands with apparent mol. wts of 220,000, 180,000 and 130,000. Following incubation with jararafibrase I, a band with an apparent mol. wt of 215,000 appeared just below the mol. wt 220,000 band . The protein band of mol. wt 180,000 disappeared, and a band with an apparent mol. wt of 175,000 was observed. The amount of mol. wt 130,000 band was decreased. Besides these changes, bands with apparent mol. wts of 110,000, 68,000, 61,000, 44,000 and 34,000 appeared . The 34,000 band seemed to be a main degradation product. After incubation with jararafibrase II, small amounts of mol. wt 215,000 and 175,000 bands were detected . The amount of 130,000 band was decreased. Besides these changes, bands with apparent mol. wts of 77,000, 68,000, 60,000, 34,000 and 24,000 were observed. The mol. wt 77,000, 68,000, 60,000 and 34,000 bands seemed to be the main degradation products. Collagenolytic, gelatinolytic andfibrinolytic activities of bacterial collagenase The specific activities of bacterial collagenase for type I collagen, type IN collagen, gelatin and fibrin were 11,104 f 363 units/mg (n = 3), 8860 f 240 units/mg (n = 3), 32,253 f 453 units/mg (n = 3) and 0.41 f 0.02 units/mg protein (n = 3), respectively (Table 1). No fibrinolytic activity was observed up to 125 ltg/ml of bacterial collagenase. Fibronectin degradation The fibronectin degradation by jararafibrase I, jararafibrase II and bacterial collagenase was investigated by SDS-PAGE (Fig. 4). The control substrate solution incubated at 37°C
Substrate Specificities of Jararafibrases
a MW
b
139 1
c
240,000 -1
117,000 76,000 -DO'
48,000 -028,000 19,000 16,000 FIG. 2. TYPE I COLLAGEN DEORADAnON BY JARARAFIBRA3E4 I AND II .
Twenty-microlitre aliquots of the substrate solutions (I mg/ml) were incubated without enzyme (lane a) and with 5 id of jararafibrase I (16 jig/ml, lane b), and jararafibrase II (105 ug/ml, lane c) at 35°C for 24 hr. All samples contained 0 .2 mM APMSF . SDS-PAGE was carried out with a 3% stacking gel and 7-12% gradient resolving gel under non-reducing conditions.
for 4 hr with 0.2 mM APMSF showed one main protein band with an apparent mol. wt of 270,000. Besides the main band, apparent mol. wt 250,000, 230,000, 225,000 protein bands and small amounts of apparent mol. wt 100,000, 63,000 and 33,000 protein bands were observed. After incubation with jararafibrase I, the four high mol. wt bands (270,000, 250,000, 230,000 and 225,000) disappeared, and several lower mol. wt degradation products were observed . The main degradation products had apparent mol. wts of 210,000, 195,000, 140,000, 44,000 and 31,000. Following incubation with jararafibrase II, three of the high mot. wt bands (270,000, 250,000 and 230,000) disappeared. The amount of 225,000 band was increased. Several lower mol. wt degradation products were observed. The apparent mol. wts of the main degradation products were 210,000, 195,000, 140,000, 66,000, 36,000, 31,000 and 28,000 . No detectable degradation of fibronectin by bacterial collagenase was detected up to 3.5 hg/ml enzyme solution, which contained
1392
M. MARUYAMA et ai . TABLE 1 .
PROTEINASE
ACTIVITM OF JARARAFIBRA.4E I, JARARAFIBRASE COr L AGENASE
II
AND BACTERIAL
Jararafibrase I
Jararafibrase II
Bacterial collagenase
3.6±0.3 (n = 3)
12.6±1 .2 (n = 3)
0.41±0.02 (n = 3)
ND
ND
11,104±363 (n = 3)
Specific activity for type IV collagen (units/mg protein)
172±5 (n = 4)
9.2±0.6 (n = 4)
8860±240 (n = 3)
Specific activity for gelatin (units/mg protein)
1315±177 (n - 8)
143±15 (n = 4)
32,253±453 (n = 3)
MHD (pg/mt)
8.7
27 .0
21 .6
Type IV collagenase units of MHD" (units/MHD)
1 .5
0.25
191.4
Specific activity for fibrin (units/mg protein) Specific activity for type I collagen (units/mg protein)
'The values were calculated from the average values of specific activity for type IV collagen . ND, not determined .
almost twice the amount of type IV collagenase activity (31.0 units/ml) compared to that ofjararafibrase I (16.3 units/ml) employed in the same assay. Degradation was observed with 35,ug/ml of bacterial collagenase. The degradation pattern with bacterial collagenase closely resembled that with jararafibrase II. The main degradation products had mol. wts of 225,000, 210,000, 195,000, 140,000, 66,000 and 30,000 .
degradation The laminin degradation by jararafibrase I, jararafibrase 11 and bacterial collagenase was investigated by SDS-PAGE (Fig. 5). The control substrate solution incubated at 37°C for 4 hr with 0.2 mM APMSF showed two main protein bands representing A-chain and B-chains (BI and B2) with apparent mol. wts of approximately 400,000 and 230,000, respectively. Protein bands with apparent mol. wts of 140,000, 125,000 and 120,000, which were assumed to represent entactin and its fragments, were also observed. Following incubation with jararafibrase I, the amount of A-chain was decreased. The protein bands of mol. wts 140,000 and 125,000 disappeared, and several lower mol . wt degradation products were observed. The main degradation products had mol. wts of 150,000, 80,000, 72,000, 67,000, 55,000, 41,000 and 30,000 . The degradation pattern obtained by incubation with jararafibrase II resembled that with jararafibrase I. The protein bands with mol. wts of 140,000 and 125,000 disappeared. The intensity of the A-chain was diminished, and the main degradation products had apparent mol. wts of 120,000, 72,000 and 55,000 . Similarly to the case of fibronectin, no degradation by bacterial collagenase was detected up to 3.5 ug/ml of enzyme solution, while degradation was observed with 35 jug/ml of bacterial collagenase. Laminin
Substrate Specificities of Jararafibrases
MW
a
b
1393
c
240,000 -> 117,000 76,000 48,000 4,-
28,000 -)P19,000 16,000 FIG . 3 .
Trw IV COLLAGEN DEGRADATION BY JARARAPIBRAS4
I AND II .
Twenty-microlitre aliquots of the substrate solutions (I mg/ml) were incubated without enzymes (lane a) and with 5 Al of jararafibrase I (16 ug/ml, lane b), and jararafibrase II (105 jug/ml, lane c) at 35°C for 10 hr. All samples contained 0.2 mM APMSF . SDS-PAGE was tamed out with a 3% stacking gel and 7-12% gradient resolving gel under reducing conditions.
Haemorrhagic activity As summarized in Table 1, the MHD values for jararafibrases I and II were 8 .7 ug/rat (1 .5 type IV collagenase units) and 27 .0 pg/rat (0.25 type IV collagenase units), respectively, while that for bacterial collagenase was 21 .6 yg/rat (191 .4 type N collagenase units) . DISCUSSION
The haemorrhaggc and necrotic activities of snake venoms are very important factors which affect the patient's prognosis, and may cause disabilities of the extremities as severe sequelae . However, the mechanisms involved in local haemorrhage by snake envenoma-
139 4
M . MARUYAMA et ai.
a Mw
b
c
d
e
f
g
h
240,000 ->
117,000 -)1,76,000 -0, 48,000
28,000 19,000 16,000 FiG. 4. FiaRONECriN DHGRADATIoN By JARARAirnmAsEs 1, 11 AND RACrERIAL COLLAGENASE. Twenty-microlitre aliquots of the substrate solutions (1 mg/ml) were incubated without enzyme (lane a) and with 5 pl of jararaflbrase I (95 Pg/m1, lane b), jararafibrase 11 (45 jig/ml, lane c), and bacterial collagenase (1 .3 pg/ml, lane f 3 .5 4ml, lane g ; 35 ug/ml, lane h) at 37°C for 4 hr. Jararafibrases I and 11 without substrate were applied to lanes d and e, respectively . SDS-PAGE was carried out with a 3% stacking gel and 7-12% gradient resolving gel under reducing conditions .
tion still remain unclear. Damaged vascular basement membranes and rupture of the endothelial cell membranes in the haemorrhagic area have been observed in histological studies (OwNBY et al., 1990 ; KAwGuT1 et al., 1991). There appear to be three possible mechanisms involved in basement membrane degradation: (1) that caused by direct proteolytic effects of haemorrhagic factors; (2) that caused by secondary activated proteinases after envenomation ; and (3) a combination of the above two mechanisms . In the present study, we examined the direct effects of fibrinolytic-haemorrhagic enzymes purified from Bothrops jararaca venom on the connective tissue matrix and basement membrane components. The enzymes degraded type I collagen only partially. On SDS-PAGE, the degradation products were splitted a-chain only, and no further degradation of the helical portion of the molecules was observed. It appeared that partial degradation of type I collagen occurred in a non-specific manner and degradation was limited to the non-helical region . On the other hand, jararafibrase I degraded type IV collagen very strongly, and jararafibrase II also degraded type IV collagen. The specific activities of jararafibrases I and II for type IV collagen were 172 f 5 units/mg protein and 9.2 t 0.6 units/mg
Substrate Specificities of Jararafibrases
a
b
C
d
e
1395
f
g
h
117,000 ~ 76,000 -10,
48 .000 -A-
28,000
-10-
19,000 16,000
-30.
5 . L.AMININ DEGRADA*noN BY JARARAFmRASE I, JARARAFIBRA38 II AND BACTERIAL COLLAGENASE. Twenty-microlitre aliquots of the substrate solutions (1 mg/ml) were incubated without enzyme (lane a) and with 5 id of jamrafibrase I (95 jug/ml, lane b), jaarafibmse II (45 pg/ml, lane c), and bacterial collagenase (1 .3 4ml, lane f, 3.5 Kg/ml, lane g; 35 pg/ml, lane h) at 37°C for 4 hr. Jararafibmses I and II without substrate were applied to lanes d and e, respectively, SDS-PAGE was carried out with a 3% stacking gel and 5-12% gradient resolving gel under reducing oonditions. nG .
protein, respectively. The enzymes also degraded gelatin, fibronectin, and laminin very strongly . Laminin and entactin are major glycoprotein components of the basement membrane, and play very important roles in cell attachment and basement membrane selfassembly (TIIYIPL., 1989; Yuxct-ENco and Sc13rr'rNY, 1990). The jararafibrase I and II degradation activities towards these extracellular matrix proteins were in good agreement with those of haemorrhagic enzymes purified from the Western diamondback rattlesnake (Crotaltis atrox) reported by BARAmovA et al. (1989). However, the respective degradation patterns were different. It has been reported that specific type IV collagenase alone was unable to destroy the basement membrane, without pretreatment with plasmin or trypsin (LIOTTA et al., 1981a; LAuG et al., 1983) . Degradation of other basement membrane components, such as glycoproteins, was suggested to be necessary to expose the type IV collagen to specific collagenase (LIOTTA et al., 1981a; LAuG et al., 1983 ; BoGENMANN and JomEs, 1983). Plasmin is thought to represent a pathophysiological enzyme which can initiate degradation of the basement membrane and connective tissue matrix. Such observations coincide with the results of the present in vivo experiments employing bacterial collagenase . The
13%
M . MARUYAMA et al.
enzyme was found to exert strong type I and type IV collagenase activities . However, it had rather weak proteolytic activities on fibrin, fibronectin and laminin, as compared to its collagenolytic activity. Fibrinolytic activity of bacterial collagenase was not observed up to 1108 type IV collagenase units/ml . No degradation of fibronectin and laminin was detected when 31 .0 type IV collagenase units of bacterial collagenase was employed (Figs 4 and 5). On the other hand, 16.0 type IV collagenase units of jararafibrase I and 0.4 type IV collagenase units of jararafibrase II readily degraded fibronectin, and laminin. The number of type IV collagenase units of bacterial collagenase which gave the MHD was 191 .4. It was confirmed that in order to produce haemorrhage by disintegrating the vascular basement membrane, specific type IV collagenase alone was not sufficient . On the other hand, the MHD ofjararafibrase I was 8.7,ug/rat, and it was calculated to be 1 .5 type IV collagenase units. Jararafibrase II induced haemorrhage, while it displayed a very weak type IV collagenase activity . The enzyme preferentially degraded fibrin, fibronectin and laminin. The number of type IV collagenase units ofjararafibrase II which gave the MHD (27.0 ug/rat) was a mere 0.25 units, which was far smaller than that of jararafibrase I. Jararafibrases I and II had a broad substrate specificity. Both jararafibrase I and jararafibrase II revealed a plasmin-like activity and degraded extracellular matrix glycoproteins, such as laminin, entactin and fibronectin, as well as fibrin. The enzymes also degraded gelatin and type IV collagen . It is assumed that the fibrinolytic properties of the enzymes may play an important role in the enhancement of haemorrhage, overcoming the victim's physiological haemostatic mechanisms . The results strongly suggest that each enzyme alone could degrade the basement membrane and the broad specificity of the enzymes is essential for provoking haemorrhage with a single enzyme . Nevertheless, the possibility does remain that haemorrhage could be enhanced by secondary activated enzymes produced from various proenzymes by haemorrhagc factors, especially from proenzymes or latent collagenolytic enzymes present in the victim's tissues.
REFERENCES AuNG-KrmN (1980) The problem of snake bites in Burma . Snake 12, 125-127 . BARAmovA, E. N., SKANNGN, J. D., BJARNASON, J . B. and Fox, J . W. (1989) Degradation of extracellular matrix proteins by hemorrhagic metalloproteinases . Arch . Blochem . Biophys. 275, 63-71 . BoaEnntANN, E. and JONES, P. A . (1983) Role of plasminogen in matrix breakdown by neoplastic cells. JNCI71, 1177-1182 . Gtsnow, M. T. and McBRme, B. C . (1975) A rapid sensitive collagenase assay . Analyt. Biochem. 68, 70-78 . KAwotm, A . S .,1~iEAtcsmN, R . D . G ., DEsIDND, H . and HurroN, R . A . (1991) Systemic haemorrhage in rats induced by a haemorrhagic friction from Bothrops jararaca venom. Toxicon 29, 1097-1105. LA©oaa, U . K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4 . Nature 227, 680-685. LAUG, W. E ., DEcLERM Y. A. and JONEs, P. A . (1983) Degradation of the subendothelial matrix by tumour cells. Cancer Res . 43, 1827-1834 . LmDBLAD, W . and FULLER, G . C. (1982) An improved assay of mammalian collagenase activity, and its use to determine hepatic extracelluhu matrix susceptibility to degradation . Clin. Chem. 28, 2134-2138 . LtorrA, L. A ., GoLDPARB, R. H., BRurroAoE, R., SEAL, G. P ., TERRANOVA, V. and GARBisA, S . (1981x) Effect of plasminogen activator (urokinase), planmin, and thrombin on glycoproteins and eollagenous components of basement membrane. Cancer Res . 41, 4629-4636 . LIarTA, L . A ., TRYGGVASDN, K., GAR=A, S ., ROBEY, P. G . and ABE, S . (19816) Partial purification and characterization of a neutral protease which cleaves type IV collagen . Biochemistry 20, 100-104 . LowRY, O . H., RosEBROUatt, N . J ., FARR, A . L. and RANDALL, R. J . (1951) Protein measurement with the Folin phenol reagent . J. biol. Chem . 193, 265-275 . MARKLAND, F. S., JR (1991) Inventory of a- and ß-fibrinogenases from snake venoms. Thromb . Haemost. 65, 438-443 .
Substrate Specificities of Jararafibrases
1397
MARUYAMA, M ., KmaGun, A . S ., CARD~, J . L. C., SANG-MAR77NS, I . S., CHUDZ~ A. M ., SANTORO, M . L ., Mo~, P., Tomy, S . C ., ANTONro, L . C., M~, H . and KELEN, E . M . A . (1990) Studies on blood coagulation and fibrinolysis in patients bitten by Bothrops jararaca (jararaca) . Thromb. Haemost. 63, 449 453 . MARUYAYA, M ., SUGM, M ., YOSHIDA, E., MmARA, H. and NAKAnmA, N. (1992) Purification and characterization of two fibrinolytic enzymes from Bothrops jararaca (jararaca) venom. Toxicon 30, 853-864 . Mommoro, N ., Suin, H ., TsustmmA, Y ., ETou, Y ., YOSHIDA, E. and MntARA, H . (1991) Activated fibrinolytic enzymes in the synovial fluid during acute arthritis induced by urate crystal injection in dogs . Blood Coag. Fibrin . 2, 609-616. MURPHY, G ., BRETZ, U ., BAGGIDrim, M . and REYNOLDS, J. J . (1980) The latent collagenase and gelatinase of human polymorphonuclear neutrophil leucocytes . Biochem. J. 192, 517-525. OwNay, C . L ., NtKA, T., ImAi, K . and SuGnrARA, H . (1990) Pathogenesis of hemorrhage induced by bilitoxin, a hemorrhagc toxin isolated from the venom of the common cantil (Agkistrodon bilineatus bilineatus). Toxicon 29,837-846 . SCHOR, S . L . (1980) Cell proliferation and migration on collagen substrate in vitro . J. Cell. Sci. 41, 159-175. TtEAKsroN, R. D . G . and REiD, H . A . (1983) Development of simple standard assay procedures' for the characterization of snake venoms . Bull. W.H.O . 61, 949-956 . TEL, R . (1989) Structure and biological activity of basement membrane proteins. Ear. J. Biochem . 180, 487-502 . YuxcrmNco, P . D. and ScHrrrNY, J. C. (1990) Molecular architecture of basement membranes. FASEB-J. 4, 1577-1590 .
C
0041-0101/92 ßI00 + .00 1992 Perpmoo Press Ltd
BROAD SUBSTRATE SPECIFICITY OF SNAKE VENOM FIBRINOLYTIC ENZYMES : POSSIBLE ROLE IN HAEMORRHAGE MASUG1 MARUYAMA, MAsAHIR0 SUGIKI, ETSUo YOSHIDA, KAZUHIIto SHmAYA HISAsm MniARA
and
Department of Physiology, Miyazaki Medical College, 5200 Kiham, Kiyotake-cho, Miyazaki-ken 889-16, Japan (Received 7 May
1992,
accepted
8
June
1992)
and H . MmARA . Broad substrate specificity of snake venom fibrinolytic enzymes : possible role in haemorrhage. Toxicon 30, 1387-1397, 1992.-We found previously that two fibrinolytic enzymes (jararafibrases I and II) purified from Bothrops jararaca venom displayed a haemorrhaggc activity . To elucidate the mechanisms involved and the role of the enzymatic activity in haemorrhage, the enzymatic properties of the purified enzymes were examined . The substrate specificity of the enzymes was determined using type I collagen, type IV collagen, gelatin, laminin and fibronectin as substrates . The enzymes degraded type IV collagen, gelatin, laminin and fibronectin into smaller fragments, but degraded type I collagen only partially in a non-specific manner. The specific activities of jararafibrase I for type IV collagen and gelatin were 172 t 5 units/mg protein and 1315 f 177 units/mg protein, respectively. The specific activities of jaraafibrase II for type IV collagen and gelatin were 9.2 f 0.6 units/mg protein and 143 f 15 units/mg protein, respectively. It was evident that the enzymes had rather broad substrate specificities and degraded basement membrane components including type IV collagen . The number of type IV collagen units of bacterial collagenase which gave the minimal haemorrhaggc dose was 191 .4, while the numbers of type IV collagenase units of jararafibrases I and II which gave the minimal haemorrhaggc dose were 1 .5 and 0.25, respectively. It is suggested that the broad substrate specificity of the enzymes is essential for inducing haemorrhage with a single enzyme . M . MARuyAMA, M . SuGml, E . YOSFMA, K . SHIMAYA
INTRODUCTION
haemorrhage and necrosis represent major complications following snake envenomation, especially in the case of Viperidae and Crotalidae snake envenomations. These snake venoms also provoke systemic coagulopathy with a decreased fibrinogen level (AuNG-KHIN, 1980; MARuyAMA et al., 1990). However, the pathophysiological mechanisms involved in the local haemorrhage induced by snake envenomation still remain unclear. Disruption of the capillary basement membrane has been observed in the haemorrhaggc area by histological techniques (OwNBY et al., 1990; KAMIGuTI et al., 1991). LocAL
1387
1388
M. MARUYAMA et al.
It has been reported that many of the purified haemorrhagic factors possess proteinase activities, especially fibrinolytic activity (iVIARKLAND, 1991). Local haemorrhage after snake envenomation is conceivably induced by disruption of the vascular basement membrane, occurring simultaneously with abnormal haemostatic systems. Enzymatic degradation of the vascular basement membrane is assumed to be a major cause of haemorrhage. However, it is unclear whether the degradation of the basement membrane is caused by a direct proteolytic effect of haemorrhagic factors in the snake venoms or by the victim's secondary activated proteinases following envenomation . We previously purified two fibrinolytic enzymes, which were suggested to be metalloproteinases, from Bothrops jararaca venom and found that each of them displayed a haemorrhagic activity (MARUYAMA et al., 1992). In the present study, we examined the activity of these fibrinolytio-haemorrhagic enzymes purified from Bothrops jararaca on various natural substrates, especially on basement membrane components, in order to clarify the direct effect of the enzymes on basement membrane degradation. MATERIALS AND METHODS Materials (p-Amidinophenyl)methanesulphonyl fluoride (APMSF) was purchased from Wako Pure Chemical Industries, Ltd, Osaka, Japan. Fibrinogen (bovine, 75% clottable) was purchased from Miles Inc., Kankakee, IL, U.S.A . ["C] Acetic anhydride was purchased from New England Nuclear Corp ., Boston, MA, U.S.A. Type I collagen (bovine achilles tendon), type IV collagen (human placenta), laminin (EHS mouse sarcoma) and bacterial collagenase (Type IA, Clostridhrn hirtolyticron) were purchased from Sigma Chemical Co ., St . Louis, MO, U.S .A . Fibronectin (human plasma) was purchased from Cosmo Bio Co., Ltd, Tokyo, Japan. Rat tail tendon type I oDllagen was purified by the method of SCHOR (1980). Venom source . A pool of lyophilized venom obtained from adult specimens of Bothrops jararaca maintained at the Laboratory of Herpetology, Instituto Butantan, was used . Methods Prrffrcation procedures. Jararafibrases I and II were purified as reported previously (MAAuyANA et al., 1992). The final products revealed a single protein band on analytical polyacrylamide gel electrophoresis (PAGE) and isoelectric focusing in agarose gel. The specific fibrinolytic activities of jararaflbreses I and II were 3.6 t 0.3 units/mg protein and 12.6 f 1 .2 units/mg protein, respectively . Protein concentration. Protein concentration was estimated by the method of LAwRY et al. (1951) employing bovine serum albumin as a standard . For determination of the type I and IV collagen concentrations, known amounts of type I and IV collagen were employed as standards. Assay offtbrinolytic activity. The flbrinolytic activity was measured using 0.6% bovine plasminogen-free fibrin plates (Moauaoro et al., 1991). For this purpose, 30 id of sample was placed on a fibrin plate and the lysis area was measured after incubation at 37°C for 18 hr. The specific activity was calculated from a standard curve for the lysis area obtained with planmin (Sigma Chemical Co .) on the plasminogen-free fibrin plates. One fibrinolytic activity unit was defined as being of the same magnitude as the lytic activity of 1 casein unit (CU) of standard planmin on a plasminogen-free fibrin plate. Assay of type l collagenase activity. The type I collagenase activity was measured according to the method of LnvDSLAD and Fur.rEt (1982) . Type I collagen (bovine achilles tendon) was labelled with "C by the method of Grsat.ow and McBrunr; (1975) . The specific radioactivity of the substrate was approximately 290,000 cpm/mg protein. Immediately before the assay, the '+C-labelled type I collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Trio-HCI buffer containing 10 mM CaCl 2 and 0.4 M NaCl, pH 7.5, and employed for the assay as the substrate solution . Then 100 id of substrate solution was mixed with 100 pl of sample . The resultant mixture was incubated at 35°C for 120 or 180min. The reaction was terminated by adding 100pl of 100 mM ethy ic acid (EDTA) containing 1.5 mg/ml of unlabelled rat tail tendon type I collagen . After incubation at 35°C for 30 min, 30D id of 1.4-dioxane/methanol (4 :1 by volume) was added, and the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 200 pl of the supernatant was mixed with 4ml of Aqueous Counting Scintillant II (ACS II, Amersham Japan, Tokyo, Japan) solution, and the radioactivity was counted in a liquid scintillation system (Aloka, LSC-700) . One type I collagenase unit was defined as the activity which degraded type I collagen at a rate of 1 pg/min at 35°C . Assay of type IV collggenase activity. The type IV collagenase activity was measured according to the method of L.raTA et al. (1981b). Type IV collagen (human placenta) waslabelled with "C by the method of Gmr.ow and
Substrate Specificities of Jararafibrases
1389
McBRiDE (1975). The specific radioactivity of the substrate was approximately 192,000 cpm/mg protein. Immediately before the assay, the "C-labelled type IV collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Tris--HCI buffer containing 10 mM Caa z and 0.4 M NaCl, pH 7.5, and employed for the assay as the substrate solution . Then 100 pl of substrate solution was mixed with 100 pl of sample and 200 id of 50 mM Tris-HCl buffer containing 5 mM CaCl2, 0.2 M NaCl, pH 7.5. The resultant mixture was incubated at 37°C for 60 or 180 min. The reaction was terminated by adding 125 pl of 10% trichloroacetic acid (TCA) containing 0.5% tannic acid, preceded by the addition of 100 pl of bovine serum albumin (0.2 mg/ml, GIBCO, Grand Island, NY, U.S.A.). After being allowed to stand in ice for 30 min, the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 400 id of the supernatant was mixed with 4 ml of ACS 11 solution, and the radioactivity was counted. One type IV collagenase unit was defined as the activity which degraded type IV collagen at a rate of I jug/min at 37°C. Assay ofgelathrase activity . The gelatinise activity was determined according to the method of MuRPrnr et al. (1980). The "C-labelled type I collagen (2 mg/ml, in 0.01% acetic acid) was mixed with the same volume of 100 mM Tris-HCI buffer containing 10 mM CaCl2 and 0.4 M NaCl, pH 7.5 . The type I collagen was subsequently heat-denatured by incubation at 45°C for 15 min. Then 100 pl of substrate solution was mixed with 100 pl of sample solution and the resultant mixture was incubated at 37°C for 30 min. The reaction was terminated by adding 100 pl of 45% TCA. After being allowed to stand in ice for 10 min, the mixture was centrifuged at 10,000 rpm for 15 min. Finally, 150 Id of the supernatant was mixed with 4ml of ACS II solution, and the radioactivity was counted. One gelatinise unit was defined as the activity which degraded gelatin at a rate of 1 ug/min at 37°C . Analysis by SDS-PAGE. First, 20 pl of substrate solution (1 mg/ml) was mixed with 5 id of enzyme solution containing 1 mM (p-amidinophenyl)methanesulphonyl fluoride (APMSF). The mixture was then incubated at 37°C for 4 hr in the case of laminin and fibronectin degradation analysis, at 35°C for 10 hr in type IV collagen degradation analysis, and at 35°C for 24 hr in type I collagen degradation analysis . The concentrations of the jaramfibrase I enzyme solutions were 95 #g/ml for laminin and fibronectin, and 16 pg/ml for type I and IV collagen . The concentrations of the jaramfibrase II enzyme solutions were 45 pg/ml for he minin and fibronectin, and 105 pg/ml for type I and IV collagen . The degradation patterns were analysed by SDS-PAGE according to the method of LAEmmr (1970), in the presence of 3 M urea . Staining was carried out with 0.2% Coomassie brilliant blue . Prestained protein mol. wt standards (GIBCO BRL, Gaithersburg, MD, U.S .A.) were employed for mol. wt estimation . Animal experiments. Female KUD Wistar rats, weighing 350-400 g, from the Experimental Animal Center at Miyazaki Medical College were used . The rats were kept at a controlled ambient temperature of 23 f 1°C with 50 t 10% relative humidity . Haemorrhagic activity was determined according to the method of TrmAKmN and RED (1983) . For this purpose, 50 Al of enzyme solution was injected intradermally into the rat dorsal skin under ether anaesthesia. The haemorrhaged areas were measured after 24 hr . The minimal haemorrhagec dose (MHD, 4rat) was defined as the sample amount which yielded a 1-cm diameter area of haemorrhage in the rat skin after 24 hr .
RESULTS
Type I collagen degradation
Jararafibrase I released radioactive peptides from "C-labelled type I collagen in a dose-dependent manner, as shown in Fig. IA. However, the linearity of the reaction was very poor . Jararafibrase II degraded type I collagen only partially (Fig. 1 B). As illustrated in Fig. 2, the control substrate solution after incubation at 35°C for 24 hr with 0.2 mM APMSF revealed a typical pattern of non-reduced type I collagen : two ß-chains (ß , ß,2), two a-chains (a,, a2) and very high mol. wt y-chain. No degradation was observed following incubation with jararafibrase II. On incubation with jararafibrase 1, the amount of ßchains was decreased. However, no further degradation of a-chains was observed . Gelatin degradation
Jararafibrases I and II degraded gelatin very strongly . The specific activities of jararafibrases I and II for gelatin were 1315 t 177 units/mg protein (n = 8) and 143 t 15 units/mg protein (n = 4), respectively (Table 1).
1390
M . MARUYAMA et ai.
3,000 -1
Z,
3,000 -
A
2 .000 -
2,000 -
1,000 -
1 .0()0-
B
O
0
0 0
50
1100
0
25
50
Enzyme concentration (jug/mi) FIG .
1.
CLEAVAGE OF RADIOLABEL L E, TYPE Each
point
I
oou.AGEN By jAaARAFiBRAsE
II (B).
represents the average value
of two
I (A) AND JARARAFIBRA.S E
identical experiments .
Type IV collagen degradation Jararafibrase I degraded type IV collagen very strongly and the specific type IV collagenase activity was 172 f 5 units/mg protein (n = 4). Jararafibrase II exerted a rather weak activity on type IV collagen, the specific activity being 9.2 f 0.6 units/mg protein (n = 4) (Table 1). As shown in Fig. 3, on SDS-PAGE, the control substrate solution incubated at 35°C for 10 hr with 0.2 mM APMSF revealed three main protein bands with apparent mol. wts of 220,000, 180,000 and 130,000. Following incubation with jararafibrase I, a band with an apparent mol. wt of 215,000 appeared just below the mol. wt 220,000 band . The protein band of mol. wt 180,000 disappeared, and a band with an apparent mol. wt of 175,000 was observed. The amount of mol. wt 130,000 band was decreased. Besides these changes, bands with apparent mol. wts of 110,000, 68,000, 61,000, 44,000 and 34,000 appeared . The 34,000 band seemed to be a main degradation product. After incubation with jararafibrase II, small amounts of mol. wt 215,000 and 175,000 bands were detected . The amount of 130,000 band was decreased. Besides these changes, bands with apparent mol. wts of 77,000, 68,000, 60,000, 34,000 and 24,000 were observed. The mol. wt 77,000, 68,000, 60,000 and 34,000 bands seemed to be the main degradation products. Collagenolytic, gelatinolytic andfibrinolytic activities of bacterial collagenase The specific activities of bacterial collagenase for type I collagen, type IN collagen, gelatin and fibrin were 11,104 f 363 units/mg (n = 3), 8860 f 240 units/mg (n = 3), 32,253 f 453 units/mg (n = 3) and 0.41 f 0.02 units/mg protein (n = 3), respectively (Table 1). No fibrinolytic activity was observed up to 125 ltg/ml of bacterial collagenase. Fibronectin degradation The fibronectin degradation by jararafibrase I, jararafibrase II and bacterial collagenase was investigated by SDS-PAGE (Fig. 4). The control substrate solution incubated at 37°C
Substrate Specificities of Jararafibrases
a MW
b
139 1
c
240,000 -1
117,000 76,000 -DO'
48,000 -028,000 19,000 16,000 FIG. 2. TYPE I COLLAGEN DEORADAnON BY JARARAFIBRA3E4 I AND II .
Twenty-microlitre aliquots of the substrate solutions (I mg/ml) were incubated without enzyme (lane a) and with 5 id of jararafibrase I (16 jig/ml, lane b), and jararafibrase II (105 ug/ml, lane c) at 35°C for 24 hr. All samples contained 0 .2 mM APMSF . SDS-PAGE was carried out with a 3% stacking gel and 7-12% gradient resolving gel under non-reducing conditions.
for 4 hr with 0.2 mM APMSF showed one main protein band with an apparent mol. wt of 270,000. Besides the main band, apparent mol. wt 250,000, 230,000, 225,000 protein bands and small amounts of apparent mol. wt 100,000, 63,000 and 33,000 protein bands were observed. After incubation with jararafibrase I, the four high mol. wt bands (270,000, 250,000, 230,000 and 225,000) disappeared, and several lower mol. wt degradation products were observed . The main degradation products had apparent mol. wts of 210,000, 195,000, 140,000, 44,000 and 31,000. Following incubation with jararafibrase II, three of the high mot. wt bands (270,000, 250,000 and 230,000) disappeared. The amount of 225,000 band was increased. Several lower mol. wt degradation products were observed. The apparent mol. wts of the main degradation products were 210,000, 195,000, 140,000, 66,000, 36,000, 31,000 and 28,000 . No detectable degradation of fibronectin by bacterial collagenase was detected up to 3.5 hg/ml enzyme solution, which contained
1392
M. MARUYAMA et ai . TABLE 1 .
PROTEINASE
ACTIVITM OF JARARAFIBRA.4E I, JARARAFIBRASE COr L AGENASE
II
AND BACTERIAL
Jararafibrase I
Jararafibrase II
Bacterial collagenase
3.6±0.3 (n = 3)
12.6±1 .2 (n = 3)
0.41±0.02 (n = 3)
ND
ND
11,104±363 (n = 3)
Specific activity for type IV collagen (units/mg protein)
172±5 (n = 4)
9.2±0.6 (n = 4)
8860±240 (n = 3)
Specific activity for gelatin (units/mg protein)
1315±177 (n - 8)
143±15 (n = 4)
32,253±453 (n = 3)
MHD (pg/mt)
8.7
27 .0
21 .6
Type IV collagenase units of MHD" (units/MHD)
1 .5
0.25
191.4
Specific activity for fibrin (units/mg protein) Specific activity for type I collagen (units/mg protein)
'The values were calculated from the average values of specific activity for type IV collagen . ND, not determined .
almost twice the amount of type IV collagenase activity (31.0 units/ml) compared to that ofjararafibrase I (16.3 units/ml) employed in the same assay. Degradation was observed with 35,ug/ml of bacterial collagenase. The degradation pattern with bacterial collagenase closely resembled that with jararafibrase II. The main degradation products had mol. wts of 225,000, 210,000, 195,000, 140,000, 66,000 and 30,000 .
degradation The laminin degradation by jararafibrase I, jararafibrase 11 and bacterial collagenase was investigated by SDS-PAGE (Fig. 5). The control substrate solution incubated at 37°C for 4 hr with 0.2 mM APMSF showed two main protein bands representing A-chain and B-chains (BI and B2) with apparent mol. wts of approximately 400,000 and 230,000, respectively. Protein bands with apparent mol. wts of 140,000, 125,000 and 120,000, which were assumed to represent entactin and its fragments, were also observed. Following incubation with jararafibrase I, the amount of A-chain was decreased. The protein bands of mol. wts 140,000 and 125,000 disappeared, and several lower mol . wt degradation products were observed. The main degradation products had mol. wts of 150,000, 80,000, 72,000, 67,000, 55,000, 41,000 and 30,000 . The degradation pattern obtained by incubation with jararafibrase II resembled that with jararafibrase I. The protein bands with mol. wts of 140,000 and 125,000 disappeared. The intensity of the A-chain was diminished, and the main degradation products had apparent mol. wts of 120,000, 72,000 and 55,000 . Similarly to the case of fibronectin, no degradation by bacterial collagenase was detected up to 3.5 ug/ml of enzyme solution, while degradation was observed with 35 jug/ml of bacterial collagenase. Laminin
Substrate Specificities of Jararafibrases
MW
a
b
1393
c
240,000 -> 117,000 76,000 48,000 4,-
28,000 -)P19,000 16,000 FIG . 3 .
Trw IV COLLAGEN DEGRADATION BY JARARAPIBRAS4
I AND II .
Twenty-microlitre aliquots of the substrate solutions (I mg/ml) were incubated without enzymes (lane a) and with 5 Al of jararafibrase I (16 ug/ml, lane b), and jararafibrase II (105 jug/ml, lane c) at 35°C for 10 hr. All samples contained 0.2 mM APMSF . SDS-PAGE was tamed out with a 3% stacking gel and 7-12% gradient resolving gel under reducing conditions.
Haemorrhagic activity As summarized in Table 1, the MHD values for jararafibrases I and II were 8 .7 ug/rat (1 .5 type IV collagenase units) and 27 .0 pg/rat (0.25 type IV collagenase units), respectively, while that for bacterial collagenase was 21 .6 yg/rat (191 .4 type N collagenase units) . DISCUSSION
The haemorrhaggc and necrotic activities of snake venoms are very important factors which affect the patient's prognosis, and may cause disabilities of the extremities as severe sequelae . However, the mechanisms involved in local haemorrhage by snake envenoma-
139 4
M . MARUYAMA et ai.
a Mw
b
c
d
e
f
g
h
240,000 ->
117,000 -)1,76,000 -0, 48,000
28,000 19,000 16,000 FiG. 4. FiaRONECriN DHGRADATIoN By JARARAirnmAsEs 1, 11 AND RACrERIAL COLLAGENASE. Twenty-microlitre aliquots of the substrate solutions (1 mg/ml) were incubated without enzyme (lane a) and with 5 pl of jararaflbrase I (95 Pg/m1, lane b), jararafibrase 11 (45 jig/ml, lane c), and bacterial collagenase (1 .3 pg/ml, lane f 3 .5 4ml, lane g ; 35 ug/ml, lane h) at 37°C for 4 hr. Jararafibrases I and 11 without substrate were applied to lanes d and e, respectively . SDS-PAGE was carried out with a 3% stacking gel and 7-12% gradient resolving gel under reducing conditions .
tion still remain unclear. Damaged vascular basement membranes and rupture of the endothelial cell membranes in the haemorrhagic area have been observed in histological studies (OwNBY et al., 1990 ; KAwGuT1 et al., 1991). There appear to be three possible mechanisms involved in basement membrane degradation: (1) that caused by direct proteolytic effects of haemorrhagic factors; (2) that caused by secondary activated proteinases after envenomation ; and (3) a combination of the above two mechanisms . In the present study, we examined the direct effects of fibrinolytic-haemorrhagic enzymes purified from Bothrops jararaca venom on the connective tissue matrix and basement membrane components. The enzymes degraded type I collagen only partially. On SDS-PAGE, the degradation products were splitted a-chain only, and no further degradation of the helical portion of the molecules was observed. It appeared that partial degradation of type I collagen occurred in a non-specific manner and degradation was limited to the non-helical region . On the other hand, jararafibrase I degraded type IV collagen very strongly, and jararafibrase II also degraded type IV collagen. The specific activities of jararafibrases I and II for type IV collagen were 172 f 5 units/mg protein and 9.2 t 0.6 units/mg
Substrate Specificities of Jararafibrases
a
b
C
d
e
1395
f
g
h
117,000 ~ 76,000 -10,
48 .000 -A-
28,000
-10-
19,000 16,000
-30.
5 . L.AMININ DEGRADA*noN BY JARARAFmRASE I, JARARAFIBRA38 II AND BACTERIAL COLLAGENASE. Twenty-microlitre aliquots of the substrate solutions (1 mg/ml) were incubated without enzyme (lane a) and with 5 id of jamrafibrase I (95 jug/ml, lane b), jaarafibmse II (45 pg/ml, lane c), and bacterial collagenase (1 .3 4ml, lane f, 3.5 Kg/ml, lane g; 35 pg/ml, lane h) at 37°C for 4 hr. Jararafibmses I and II without substrate were applied to lanes d and e, respectively, SDS-PAGE was carried out with a 3% stacking gel and 5-12% gradient resolving gel under reducing oonditions. nG .
protein, respectively. The enzymes also degraded gelatin, fibronectin, and laminin very strongly . Laminin and entactin are major glycoprotein components of the basement membrane, and play very important roles in cell attachment and basement membrane selfassembly (TIIYIPL., 1989; Yuxct-ENco and Sc13rr'rNY, 1990). The jararafibrase I and II degradation activities towards these extracellular matrix proteins were in good agreement with those of haemorrhagic enzymes purified from the Western diamondback rattlesnake (Crotaltis atrox) reported by BARAmovA et al. (1989). However, the respective degradation patterns were different. It has been reported that specific type IV collagenase alone was unable to destroy the basement membrane, without pretreatment with plasmin or trypsin (LIOTTA et al., 1981a; LAuG et al., 1983) . Degradation of other basement membrane components, such as glycoproteins, was suggested to be necessary to expose the type IV collagen to specific collagenase (LIOTTA et al., 1981a; LAuG et al., 1983 ; BoGENMANN and JomEs, 1983). Plasmin is thought to represent a pathophysiological enzyme which can initiate degradation of the basement membrane and connective tissue matrix. Such observations coincide with the results of the present in vivo experiments employing bacterial collagenase . The
13%
M . MARUYAMA et al.
enzyme was found to exert strong type I and type IV collagenase activities . However, it had rather weak proteolytic activities on fibrin, fibronectin and laminin, as compared to its collagenolytic activity. Fibrinolytic activity of bacterial collagenase was not observed up to 1108 type IV collagenase units/ml . No degradation of fibronectin and laminin was detected when 31 .0 type IV collagenase units of bacterial collagenase was employed (Figs 4 and 5). On the other hand, 16.0 type IV collagenase units of jararafibrase I and 0.4 type IV collagenase units of jararafibrase II readily degraded fibronectin, and laminin. The number of type IV collagenase units of bacterial collagenase which gave the MHD was 191 .4. It was confirmed that in order to produce haemorrhage by disintegrating the vascular basement membrane, specific type IV collagenase alone was not sufficient . On the other hand, the MHD ofjararafibrase I was 8.7,ug/rat, and it was calculated to be 1 .5 type IV collagenase units. Jararafibrase II induced haemorrhage, while it displayed a very weak type IV collagenase activity . The enzyme preferentially degraded fibrin, fibronectin and laminin. The number of type IV collagenase units ofjararafibrase II which gave the MHD (27.0 ug/rat) was a mere 0.25 units, which was far smaller than that of jararafibrase I. Jararafibrases I and II had a broad substrate specificity. Both jararafibrase I and jararafibrase II revealed a plasmin-like activity and degraded extracellular matrix glycoproteins, such as laminin, entactin and fibronectin, as well as fibrin. The enzymes also degraded gelatin and type IV collagen . It is assumed that the fibrinolytic properties of the enzymes may play an important role in the enhancement of haemorrhage, overcoming the victim's physiological haemostatic mechanisms . The results strongly suggest that each enzyme alone could degrade the basement membrane and the broad specificity of the enzymes is essential for provoking haemorrhage with a single enzyme . Nevertheless, the possibility does remain that haemorrhage could be enhanced by secondary activated enzymes produced from various proenzymes by haemorrhagc factors, especially from proenzymes or latent collagenolytic enzymes present in the victim's tissues.
REFERENCES AuNG-KrmN (1980) The problem of snake bites in Burma . Snake 12, 125-127 . BARAmovA, E. N., SKANNGN, J. D., BJARNASON, J . B. and Fox, J . W. (1989) Degradation of extracellular matrix proteins by hemorrhagic metalloproteinases . Arch . Blochem . Biophys. 275, 63-71 . BoaEnntANN, E. and JONES, P. A . (1983) Role of plasminogen in matrix breakdown by neoplastic cells. JNCI71, 1177-1182 . Gtsnow, M. T. and McBRme, B. C . (1975) A rapid sensitive collagenase assay . Analyt. Biochem. 68, 70-78 . KAwotm, A . S .,1~iEAtcsmN, R . D . G ., DEsIDND, H . and HurroN, R . A . (1991) Systemic haemorrhage in rats induced by a haemorrhagic friction from Bothrops jararaca venom. Toxicon 29, 1097-1105. LA©oaa, U . K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4 . Nature 227, 680-685. LAUG, W. E ., DEcLERM Y. A. and JONEs, P. A . (1983) Degradation of the subendothelial matrix by tumour cells. Cancer Res . 43, 1827-1834 . LmDBLAD, W . and FULLER, G . C. (1982) An improved assay of mammalian collagenase activity, and its use to determine hepatic extracelluhu matrix susceptibility to degradation . Clin. Chem. 28, 2134-2138 . LtorrA, L. A ., GoLDPARB, R. H., BRurroAoE, R., SEAL, G. P ., TERRANOVA, V. and GARBisA, S . (1981x) Effect of plasminogen activator (urokinase), planmin, and thrombin on glycoproteins and eollagenous components of basement membrane. Cancer Res . 41, 4629-4636 . LIarTA, L . A ., TRYGGVASDN, K., GAR=A, S ., ROBEY, P. G . and ABE, S . (19816) Partial purification and characterization of a neutral protease which cleaves type IV collagen . Biochemistry 20, 100-104 . LowRY, O . H., RosEBROUatt, N . J ., FARR, A . L. and RANDALL, R. J . (1951) Protein measurement with the Folin phenol reagent . J. biol. Chem . 193, 265-275 . MARKLAND, F. S., JR (1991) Inventory of a- and ß-fibrinogenases from snake venoms. Thromb . Haemost. 65, 438-443 .
Substrate Specificities of Jararafibrases
1397
MARUYAMA, M ., KmaGun, A . S ., CARD~, J . L. C., SANG-MAR77NS, I . S., CHUDZ~ A. M ., SANTORO, M . L ., Mo~, P., Tomy, S . C ., ANTONro, L . C., M~, H . and KELEN, E . M . A . (1990) Studies on blood coagulation and fibrinolysis in patients bitten by Bothrops jararaca (jararaca) . Thromb. Haemost. 63, 449 453 . MARUYAYA, M ., SUGM, M ., YOSHIDA, E., MmARA, H. and NAKAnmA, N. (1992) Purification and characterization of two fibrinolytic enzymes from Bothrops jararaca (jararaca) venom. Toxicon 30, 853-864 . Mommoro, N ., Suin, H ., TsustmmA, Y ., ETou, Y ., YOSHIDA, E. and MntARA, H . (1991) Activated fibrinolytic enzymes in the synovial fluid during acute arthritis induced by urate crystal injection in dogs . Blood Coag. Fibrin . 2, 609-616. MURPHY, G ., BRETZ, U ., BAGGIDrim, M . and REYNOLDS, J. J . (1980) The latent collagenase and gelatinase of human polymorphonuclear neutrophil leucocytes . Biochem. J. 192, 517-525. OwNay, C . L ., NtKA, T., ImAi, K . and SuGnrARA, H . (1990) Pathogenesis of hemorrhage induced by bilitoxin, a hemorrhagc toxin isolated from the venom of the common cantil (Agkistrodon bilineatus bilineatus). Toxicon 29,837-846 . SCHOR, S . L . (1980) Cell proliferation and migration on collagen substrate in vitro . J. Cell. Sci. 41, 159-175. TtEAKsroN, R. D . G . and REiD, H . A . (1983) Development of simple standard assay procedures' for the characterization of snake venoms . Bull. W.H.O . 61, 949-956 . TEL, R . (1989) Structure and biological activity of basement membrane proteins. Ear. J. Biochem . 180, 487-502 . YuxcrmNco, P . D. and ScHrrrNY, J. C. (1990) Molecular architecture of basement membranes. FASEB-J. 4, 1577-1590 .
