Comp. Biochem. PhysioL Vol. 102B, No. 1, pp. 163-167, 1992 0305-0491/92 $5.00+0.00 Printed in Great Britain © 1992 Pergamon Press Ltd

FIBRINOLYSIS RELATING SUBSTANCES IN MARINE CREATURES*

HIROYUKI SUMI, NOBUYOSHI NAKAJIMA and HISASHI MIHARA~f

Department of Food Science and Nutrition, Okayama Prefectural College, Okayama, Japan; and tDepartment of Physiology, Miyazaki Medical College, Miyazaki, Japan

(Received 28 May 1991)

Abstract--1. Extracts with physiological saline solution were obtained from about 20 species of invertebrates and seaweed. Tosyl-L-Arg-MeOH hydrolysing and fibrin plate lytic activity were detected in the invertebrates Stichopus japonicus, Crassost gigas, Tapes japonica, and Kintai-gai as well as the seaweed Codiales codium.

2. These activities were all labile against heat (at 65°C for 1 hr). Except for the extract from Stichopus japonicus, lytic activities against fibrin plates with and without plasminogen were similar.

3. The extract from S. japonicus showed plasminogen activating potency as well as the existence of urokinase (UK) activity enhancing factor.

4. On the other hand, the extract of the seaweed Hizikia fusiformis showed a strong UK inhibiting activity.

5. A fraction of fibrinolytic enzyme was obtained from the extract of S. japonicus by absorption to the celite affinity chromatography. It was orally administered to rabbits at a dosage of 40 mg/kg/day.

6. Fibrinolytic activity was determined periodically on the eugloblin fraction of plasma samples collected from these animals.

7. As compared with the pretreatment value, the activity increased about 2 times (P < 0.01) and 3 times (P < 0.005) after 4 and 8 weeks, respectively, of the treatment.

8. After 8 weeks of treatment, the kidney of treated rabbits was extracted with 2 M KCI. The activity of tissue plasminogen activator (free-type TPA) was revealed to be enhanced significantly (P < 0.001) in the extracts.

9. The fibrinolytic enzyme increased in the blood was recognized by zymography to be mainly the UK type plasminogen activator with mol. wt of 53,000.

INTRODUCTION

Thrombosis has been revealed recently to induce a wide range of symptoms covering cerebral infarction, angina pectoris and senile dementia. Though there are therapeutic drugs for thrombosis, any prophylac- tic drug to dissolve the thrombus has yet to be discovered. During the last decade, we have investi- gated new 'oral fibrinolytic therapies', different from the conventional intravenous infusion therapies, via the activation of fibrinolytic enzymes from the living body to dissolve the thrombus or the enhancement of synthesis of the fibrinolytic enzymes in the cells (endothelium) of the living body (Sumi et al., 1980, 1983, 1985; Robbins et al., 1981; Sasaki et al., 1983, 1985; Toki et al., 1985) in addition to screening about 200 kinds of food in order to find sub- stances that show enhancement of fibrinolysis in the blood after oral administration. We discovered nattokinase from a traditional Japanese food, 'nat to ' (fermented soybeans) and reported its properties (Sumi, 1989, 1990; Sumi et al., 1987, 1989).

In this paper, the results obtained by the screening in marine creatures will be reported with the re- suits of administration to rabbits using a partly purified standard sample of part of the detected substances.

*Part of these data was presented at the International Society of Hematology 23rd Congress, Boston, 1990.

MATERIALS AND METHODS

The samples of marine creatures in this study were about 20 species of invertebrates and seaweed collected in the neighbouring waters of Miyazaki and Okayama prefec- tures or purchased in shops at Okayama City. The samples were rinsed with water and lyophilized. They were then homogenized in physiological saline solution to make a concentration of 0.05A).1 g dry wt/ml. Supernatant of the homogenate was collected after centrifugation at 3000g for l0 min.

Holothuris leucospitota and Codiales codium are represen- tatives of samples showing strong fibrinolytic activities (Fig. l). The decomposing activity against Tosyl-L-Arg- MeOH (TAMe) was determined by the radial diffusion assay plate method. The reactant system consisted of 0.05 M Tris buffer solution at pH 7.4 containing synthesized sub- strate TAMe at a fnal concentration of 1 mM, 0.75% of agar, and 1 mg/ml of BTB. To the reactant system, I0 #g of the sample was applied and allowed to react for 18 hr at 37°C before determination.

In order to determine the fibrinolytic activities, the method of Astrup and Miillertz (1952) using fibrin plates was employed. An artificial thrombus was prepared in a disk by coagulating 0.4% bovine fibrinogen using thrombin. The potency to dissolve the thrombus was determined. For the fibrinogen of the plate, a product on the market as it was (with plasminogen) and without plasminogen were used. Plasminogen was eliminated from the fibrinogen via treat- ment in a Lys-Sepharose column by the method of Deutsch and Mertz (1970). Urokinase (UK) was commercially obtained from Green Cross Co., Ltd. The euglobin fraction of plasma and samples was prepared by the method of

163

164 HIROYUKI SUMI et al.

(a) 1 ~

Fig. 1. Marine creatures indicating fibrinolytic activity: (a) H o l o t h u r a leucospi lota , (b) Codiales cod ium. Both collected at Nichinan Seashore in Miyazaki Prefecture in August 1989.

Milstone (1941). In this method, 30#1 of a sample was applied and allowed to react at 37°C for 18 hr.

Tissue plasminogen activator (TPA) was determined by the method of Mihara e t al. (1969). The kidney tissues were rinsed with a physiological solution of sodium chloride, minced, homogenized with 10% amount of 0.15 M KC1, left at 4°C for 60 min, and then centrifuged at 3000g for 10 min. Free-type TPA was determined on the obtained super- natant. Bound-type TPA was determined on the super- natant obtained after the following procedures: the precipitants obtained by the above-described procedures were homogenized again in the same amount of 2 M KC1 and then centrifuged. For the standard substance for TPA measurement, Melanoma TPA (Bio Pool Co., Ltd.) was

used. Fibrinolytic enzymes were subjected to celite affinity chromatography according to a previously reported method (Sumi et al., 1978). A column of celite activated with 4 M HC1 was used and absorbed enzymes were eluted with 0.1% liquid ammonium.

Molecular weight of the plasminogen activator in the plasma was determined with zymography according to the method of Tissot et al. (1982), combining the electrophoretic potencies in SDS-polyacrylamide and fibrin plate.

Aggregation of platelets was determined by the tube method (O'Brien, 1962) using an aggligometer (Evans Co., Ltd.). Collagen (Hormone Chemic) and thrombin (Green Cross Co., Ltd.) were used as aggregation-inducing sub- stances at the final concentration of 55.5#g/ml and 0.25 U/ml.

Table 1. TAMe hydrolysing activity

TAMe activity Scientific name (mm2/10 #1)

Anadara subcrenata Kintai-gai* Lot 1 44

Lot 2 18

Mytilus edulis 4 H. diversicolor aquatilis Butillus comutus 9 Crassostrea gigas 30 Tapes japonica 11 Cordiculina leana 6 Jelly fisht Holothuria leucosplota 37 Stichopus japonicus 106 Loligo pealici 3

Tangler Lot 1 Lot 2

Sargassum Eisenia bicyclis Undaria pinnatifida 8 Hizikia fusiformis Lot 1

Lot 2

Ecklonia cave Codiales codium 49 Prophyra yezoensis

*Japanese common name. tEnglish common name. The TAMe activity of each sample (10 #1) was indicated as the dissolved area (mm 2) of plate treated for 18 hr at 37°C.

Table 2. Fibrinolytic activity determined by plate method

+ Plasminogen - Plasminogen Scientific name (mm2/30/d) (mm2/30 #1)

Anadara subcrenata + + Kintai-gai* Lot 1 + +

Lot 2 + +

Mytilus edulis - 0 H. diversicolor aquatilis - + Butillus comutus 4 5 Crassostrea gigas 6 4 Tapes japonica - - Cordiculina leana Jelly fish~" Holothuria leucosplota 15 15 Stichopus japonicus 8 4 Loligo pealici 4 -

Tangler Lot 1 Lot 2

Sargassum - - Eisenia bicyclis - - Undaria pinnatifida + Hizikia fusiformis Lot 1

Lot 2 - -

Ecklonia cave - - Codiales codium 85 60 Prophyra yezoensis - --

*Japanese common name. tEnglish common name. The fibrinolysis of each sample (30 #1) was indicated as the dissolved area (mm 2) of plate with ( + ) or without ( - ) plasminogen.

Scientific name

Fibrinolysis relating substances in marine creatures

Table 3. Inhibitory activity for UK fibrinolysis

% Dissolving rate (control: 10 IU UK)

Anadara subcrenata 86 Kintai-gai* Lot 1 76

Lot 2 88

Mytilus edulis 91 H. diversicolor aquatilis 102 Butillus comutus 87 Crassostrea gigas 80 Tapes japonica 89 Cordiculina leana 84 Jelly fisht 87 Holothuria leucosplota 100 Stichopus japonicus 112 Loligo pealici 98

Tangler Lot 1 63 Lot 2 77

Sargassum 103 Eisenia bicyclis 86 Undaria pinnatifida 73 Hizikiafusiformis Lot 1 15

Lot 2 6

Ecklonia cave 71 Codiales codium 80 Prophyra yezoensis 70

*Japanese common name. l'English common name. Each sample was heated at 65°C for I hr and then mixed with the corresponding amount of UK (20 IU/ml). Following pre-incubation for 1 hr at room temperature, fibrinolysis was determined by the fibrin plate method. The activity is indicated as percentage setting the dissolved area of fibrin by UK without supplementation (at the final concentration of 10 IU/ml) × 100%.

RESULTS

Table 1 shows the results of determination of the TAMe hydrolysing activities of extracts from various creatures. The activities correspond to the non- specific trypsin-like enzymes of these extracts. Posi- tive results were obtained in invertebrates Stichopus japonicus (trepang), Crassostrea gigas, Tapes japon- ica, and kintai-gai as well as the seaweed Codiales codium. Under the same conditions, other seaweed species showed no activities.

165

NM

,~ Controt

' free' form

Control NM

' bound ' form

Fig. 3. TPA activity of the kidney tissue of rabbit orally treated with NM. Rabbits were orally administered NM at a dosage of 40 mg/kg/day twice a day for 8 weeks. Activities of 'free'- and 'bound'-type TPA were determined in the extract of their kidney tissues. Using melanoma TPA as the standard, the TPA activities are indicated as IU per g wet

weight of the tissue.

The results of fibrinolysis activity determined by the fibrin plate method are shown in Table 2. Though activities were detected in trepang and C. gigas, which had strong TAMe activities, no activity was detected in T. japonica and ony weak activities were found in Kintai-gai and Anadara subcrenata. The latter showed positive results in the determination of TAMe decomposing activity. Condiales codium was the only seaweed that showed positive fibrinolysis activity among the seaweed species tested. The pres- ence or absence of plasminogen in the plates did not affect the results except for trepang. The majority of marine creatures tested showed the activities ascribed to the enzymes which directly decomposed plasmin- like fibrin. However, they had no plasminogen acti- vator activities. These fibrinolytic enzymes were inactivated by heat treatment of the substrate at 65°C for 1 hr.

Table 3 shows the results of plate methods (with plasminogen) performed on the heat-inactivated sub- stances mixed with an equal amount of UK (20 U/ml). The resulting values were varied. The extract from trepang enhanced fibrin lytic activity of UK. In general, however, seaweed species showed

5o -Ant i-pLateLet ogg. (% dilution)

i 4O

E 2O

I0

• I,,

W, 20 ~

NH OH

PooLee i ~ =',

--3O0

o~

~oo hE

I I 40 60 80 I00

Tube No. (50mL)

Fig. 2. Celite column chromatography of extract of Stichops japonicus. Into 50 ml of distilled water, 0.63 g of standard NM was dissolved and loaded on a (:elite column (2.0 x 4.8 cm). After rinsing with water, fibrinolytic enzymes were eluted with 1% liquid ammonium. PIA is indicated by the inhibition of urokinase (IU). The inhibitory effect on the platelet aggregation with collagen is indicated as the percentage dilution of the eluant. Eluant included in the range indicated by the horizontal arrows was

pooled and lyophilized in order to be used as NM in the oral treatment experiment.

166 HIROYUKI SUM~ et al.

PLasma EugLobuLin

Before 4 8 Weeks Weeks Before 4 B

M.W.

5 x I0 4

5xlO 4

Fig. 4. Zymography of rabbit plasma. Fibrinolytic activities (plasminogen activator) of plasma (20 # 1) and eugloblin fraction of plasma collected before and after the treatment with NM. Numbers on the right side

indicate molecular weights (5 x 104) using UK as the standard.

inhibiting effects on the lytic activity of UK. The inhibitory effects were especially potent in Hizikia fusiformis, tangle, and Undaria pinnatifida. The in- hibitory substances were not dialysable in water and were heat-resistant (at 65°C for 1 hr), showing an instant inhibitory reaction with no need of pre- incubation. The effects on platelet aggregation were also investigated. The (heated) extract from trepang did not show complete inhibition against the aggre- gation of platelets by collagen. However, at a concen- tration of 0.4 mg (dry wt)/ml or more, lag time until aggregation was elongated. The (heated) extract from trepang inhibited the aggregation by thrombin com- pletely at a concentration of 2 mg/ml or more.

Figure 2 shows the results of celite affinity chroma- tography of extract from trepang with physiological saline solution (200 ml). The fraction absorbed in the column and eluted with 0.1% liquid ammonium showed fibrinolytic activity. The non-absorbed fraction showed inhibitory activities against fibrino- lysis. Between the two fractions, an active peak was detected to show inhibition against platelet aggregation.

As a trial, the fraction of trepang with fibrinolytic activity was collected from portions 25-95 (NM-003), eliminating the active portions with UK inhibition. The fraction was lyophylized and orally administered to rabbits. NM-003 was dissolved in 10 ml of physio- logical saline solution and administered using oral sounds at a dosage of 40 mg/kg body wt/day, twice a day for 8 weeks. As the result, the fibrinolytic activity of the eugloblin fraction in the plasma in- creased about 2.2 times (28.7 ___ 8.5 IU/ml) (P < 0.01) and about 2.9 times (38.8 +_ 17.0 IU/ml) (P < 0.001) after 4 and 8 weeks, respectively, of treatment as com- pared with the pretreatment value (13.3 + 5.2 IU/ml). Figure 3 shows the TPA activity determined on the kidney of rabbits treated for 8 weeks. There were no

differences in the 'bound'-type plasminogen activator eluted with 2 M KCL. The increased activity was detected in 'free'-type plasminogen activator eluted with 0.15 M KC1 (P < 0.001).

The molecular morphology of plasminogen acti- vator increased in plasma as shown in Fig. 4. When the plasma was used as a sample as it was or its eugloblin fraction was used, the window of fibrinolysis was detected at the UK position at a mol. wt of about 54,000. Its activity increased with the administration of NM-003.

DISCUSSION

In order to screen fibrinolytic enzymes applicable to oral fibrinolytic therapy, about 20 species of marine creatures were investigated in this study. Direct fibrinolytic activities were detected in invert- ebrates and seaweed. Among seaweed species, especially in Hizikia fusiformis, a substance heat- resistant (at 65°C for 1 hr) and non-dialysable in water was found to inhibit the fibrinolytic activity of UK. H. fusiformis is a traditional health food in Japan. In a certain district of Japan it is believed that H. fusiformis has a prophylactic effect on pyorrhea when used at tooth brushing. It is an interesting food. Though the number of treated rabits was small, the fibrinolytic fraction of trepang was confirmed to enhance the activity of plasminogen activator not only in the blood but also in the kidney tissue of rabbits after long-term oral administration (Fig. 3). The result of zymography (Fig. 4) indicates that the active substance had a UK-type activator with a molecular weight of about 54,000. The results were thus coincident to our previous test results of oral fibrinolytic therapies using UK or NK (Sumi, 1989, 1990; Sumi et al., 1980, 1983, 1985, 1987, 1989; Robbins et al., 1981; Sasaki et al., 1983, 1985; Toki

Fibrinolysis relating substances in marine creatures 167

et al., 1985). Trepang is a food frequently eaten raw as a relish with sake in Japan.

Considering the results of enhancement of the fibrinolytic activities in the blood induced by U K (Sumi et aL, 1980, 1983, 1985; Robbins et aL, 1981; Sasaki et aL, 1983, 1985; Toki et al., 1985), N K (Sumi, 1989, 1990; Sumi et al., 1987, 1989), or alcohol drinking (Sumi et aL, 1988), there may be a possibility of synergistic effects when these foods are eaten together. Screening on a further variety of samples as well as the investigation of obtained active substances will be undertaken in the future.

Acknowledgement--This study was supported in part by a grant from the Ministry of Education, Japan.

REFERENCES

Astrup T. and Mtillertz S. 0952) The fibrin plate method for estimating fibrinolytic activity. Archs Biochem. Biophys. 40, 346-351.

Deutsch D. G. and Mertz E. T. (1970) Plasminogen: puri- fication from human plasma by affinity chromatography. Science 170, 1095-1096.

Mihara H., Fujii T. and Okamoto S. (1969) Fibrinolytic activity of cerebro-spinal fluid and the development of artificial cerebral hematomes in dogs. Thromb. Diath. Haemorrth. 21, 294-303.

Milstone H. (1941) A factor in human blood which participates in streptococal fibrinolysis. J. lmmunol. 42, 109-116.

O'Brien J. R. (1962) Platelet aggregation Part II. Some results from a new method of study. J. din. Path. 15, 452-473.

Robbins K. C., Sumi H., Sasaki K. and Toki N. (1981) The transport of high molecular weight urokinase across the intestinal tract of dogs and human subjects. 2nd Int. Symp. on Urokinase, Abstract, p. 12. Geneva, Switzerland.

Sasaki K., Moriyama M., Sumi H., Toki N. and Robbins K. C. (1983) Intestinal transport of human t25I-high molecular weight urokinase in a dog model with a saph- enous vein thrombos. In Progress in Fibrinolysis, Vol. I, pp. 245-248. Churchill Livingstone, London.

Sasaki K., Moriyama M., Tanaka Y., Sumi H., Toki N. and Robbins K. C. (1985) Transport of 12SI-labcled human high molecular weight urokinase across the intestinal tract in a dog model with stimulation of synthesis and/or release of plasminogen activators. Blood 66, 69-75.

Sumi H. (1989) Fibrinolytic enzyme: property of natto- kinase and its application to oral fibrinolytic therapy. In "Functional" Food Materials: research and development of physiologically active substances derived from food, pp. 88-96. Industry and Technology Society, Tokyo.

Sumi H. (I 990) Manual o f Functional Food Material, Section 6, pp. 287-289. Nattokinase, CMC, Tokyo.

Sumi H., Toki N., Tanaka Y. and Tanaka T. (1978) Studies on human urinary enzymes and inhibitors: concentration method and characterization. J. Biochem. 83, 141-147.

Sumi H., Toki N., Sasaki K. and Robbins K. C. (1980) Oral administration of urokinase. Thrombos. Res. 20, 711-714.

Sumi H., Maruyama M., Yoneta T., Toki N. and Mihara H. (1983) Activation of plasma flbrinolysis after intra- rectal administration of high molecular weight urokinase and its derivative. Acta Haematol. 70, 289-295.

Sumi H., Seiki M., Morimoto N., Tsushima H., Maruyama M. and Mihara H. (1985) Plasma fibrinolysis after in- traduodenal administration of urokinase in rats. Enzyme 33, 121-127.

Sumi H., Hamada H., Tsushima H., Mihara H. and Muraki H. (1987) A novel fibronolytic enzyme (nattokinase) in the vegetable cheese Natto, a typical and popular soybean food of the Japanese diet. Experientia 43, 1110-1111.

Sumi H., Hamada H., Morimoto N. and Mihara H. (1988) Urokinase-like plasminogen activator increased in plasma after alcohol drinking. Alcohol & Alcoholism 23, 33--44.

Sumi H., Hamada H., Mihara H., Nakanishi K. and Muraki H. (1989) Fibrinolytic effect of the Japanese traditional food "Natto" (Nattokinase). Thrombosis & Haemostasis 62, 549.

Tissot J. D., Schneider P., Hauert J., Puegg M., Kruithof E. K. O. and Bachmann F. (1982) Isolation for human plasma of a plasminogen activator identical to urinary high molecular weight urokinase. J. din. Invest. 70, 1320-1323.

Toll N., Sumi H., Sasaki K., Boreisha T. and Robbins K. C. (1985) Transport of urokinase across the intestinal tract of normal human subjects with stimulation of synthesis and/or release of urokinase-type proteins. J. clin. Invest. 75, 1212-1220.

CBPB 102/I--L