ON HABU SNAKE VENOM 1. COMPARISON OF SEVERAL BIOLOGICAL ACTIVITIES OF FRESH AND DRIED HABU SNAKE VENOM

Japan. J. Microb., Vol. 3, No. 1, 1959 UDC: 612. 314. 019: 598. 126 STUDIES ON HABU SNAKE VENOM 1. COMPARISON OF SEVERAL BIOLOGICAL ACTIVITIES OF FRESH AND DRIED HABU SNAKE VENOM SUSUMU MITSUHASHI, HIROO MAENO, MASAYA KAWAKAMI AND HAJIME HASHIMOTO Department of Microbiology, School of Medicine, Gunma University, Maebashi, Japan (Director: Prof. S. Mitsuhashi) YOSHIO SAWAI, SHONOSUKE MIYAZAKI, MASAAKI MAKINO AND MASARU KOBAYASHI Laboratory of Biological Products, Institute for Infectious Diseases, University of Tokyo (Chief: Dr. Y. Sawai) TAKASHI OKONOGI AND KAZUO YAMAGUCHI Department of Pathology, School of Medicine, Gunma University, Maebashi, Japan (Director: Prof. S. Kawai) (Received December 28, 1958) Reliable information on the number of snakebite deaths is quite difficult to obtain. According to the approximate figures of the World Health Organization southeastern Asia has the highest rate, with 7,000 to 120,000 snakebite deaths each year in India and Burma. In Japan there are two dangerous species: Mamushi (Agkistrodon blomhoffli) and Habu (Trimeresurus flavoviridis). Habu is found in the southwest islands of Japan (Amami Islands, Okinawa, etc.). The cases of Habu snake bite in Amami Islands are about 300 per 150,000 habitants each year and death rate is 3 per cent1). The characteristic pathologic signs of Habu bite are hemorrhage and necrosis.2-7) The necrosis is severest in parts rich in muscular tissues, such as the leg, and in severe cases amputation is necessary. Heart weakness is characteristic in cases resulting in death. In 1908 Kitajima8) introduced serum therapy for Habu bite and the mortality decreased from 15 to 3 per cent. In spite of the effectiveness of serum therapy against death rate, the problem of local symptoms of Habu bite still remained unsolved. Hosoya and his coworkers9-23) reported many papers on serum therapy and the biological activities of Habu snake venom. Habu snake bites in the American Army in the Ryukyus Islands was reported in 1946.24) In 1957 we investigated snake bites in the Amami Islands and started studies of the therapy of Habu snake bite. Ohsaka25) reported the purification of Habu snake venom and stated that two hemolytic activities are present and that phosphodiesterase and 5'-nucleotidase activities can be separated from the venom toxicity. In this paper several biological activities of

96 MITSUHASHI et al. Vol. 3, No. 1 Habu snake venom and the comparison of fresh and dried venoms are described. MATERIALS AND METHOD A dried pool of venom was collected from species of Trimeresurus flavoviridis (Hallowell) caught on the Amami Islands from 1942 to 1958. Most of the samples were dried in a dessicator on calcium chloride or sulfuric acid by the usual method used in Amami Islands. The fresh materials were taken from the species of Trimeresurus flavoviridis (Hallowell) and transported by air from Amami Islands to Institute for Infectious Diseases, Tokyo. Part of the fresh samples were stored by lyophilizing. The venom solution from dried sample was prepared in physiological saline or in veronal-buffered saline of ph 7.4 immediately before use. The fresh sample is yellow, sticky and slight acid. Fluorescence is observed in dark by ultraviolet light. The upper part of the solution is yellow and the under part of the solution looks pale on standing in the cold. The solution turns yellow on shaking. Experimental animals. Six week-old mice of ddn strain, weighing 17 to 20 g, produced by Laboratory for Experimental Animals, Gunma University, were used throughout this study. Rabbits weighing about 3 Kg were used. The anti-serum for Habu venom was prepared in Institute for Infectious Diseases, Tokyo. Toxicity. Venom toxicity was determined by intravenous or intraperitoneal injection of serial dilutions of each sample into mice. Five mice were used for each dilution. The toxicity was expressed by the MLD (minimal lethal dose) for mice. Protein concentration. Protein was determined by the method described by Folin and Ciocalteu26) or by micro-kjeldahl method. Hemolysing activity. Hemolysing activity of Habu venom was determined by the method of Slotta and Szyska.27) All reagents were diluted with isotonic veronalbuffered saline of ph 7.4. 3.75 g of 5,5'-diethylbarbituric acid, 3.75 g of sodium 5,5'- diethylbarbiturate and 85 g of sodium chloride were dissolved in two liters of distilled water. A five fold dilution of this solution is isotonic. Sheep red cells were washed with isotonic veronal buffered saline by centrifugation and 1,000,000 red cells in one ml of veronal buffered saline were used for assay. A series of two-fold dilutions of venom samples was prepared in a volume of one ml. To each tube, one ml of sheep red cell suspension was added. The indirect hemolysin titer was determined by using isotonic veronal buffered saline containing 20 ƒêg (dry weight) ef egg-yolk for the dilution of all reagents. Two hours at 37 Ž or two hours at 37 Ž followed by standing overnight in the refrigerator was adopted for the incubation period. The hemolysin titer was expressed as the highest dilution of the venom solution showing complete hemolysis and as the 50 percent end point of complete hemolysis according to the method described by Slotta and Szyska.27) Hemorrhagic necrosis. Hemorrhagic necrosis was determined by injecting 0.1 ml series of 2 fold dilutions of venom into the dorsal skin of rabbit. The longest and shortest diameters of hemorrhagic necrosis were determined after 24 hours. The grade of necrosis was expressed as the area of necrosis with hemorrhage obtained by multiplying two diameters.

January, 1959 STUDIES ON HABU SNAKE VENOM 97 Proteinase activity.28-29) As a substrate for proteinase, two grams of casein was added to 75 ml M/15 Na2HPO4, dissolved by heating in a boiling water bath, and the volume made to 100 ml by adding distilled water. The ph of this solution was about 7.0. A mixture of 0.5 ml of venom solution and 0.5 ml of M/4 phosphate buffer (ph 7.5) was warmed to 35 Ž. The reaction was initiated by adding 1.0 ml of substrate at 35 Ž to the venom solution. Incubation for ten minutes at 35 Ž was followed by the rapid addition of 2.0 ml of 0.4 M trichloracetic acid. The quantity of split products remaining in solution was determined colorimetrically by means of the phenol reagent, and expressed as micromoles of tyrosine. RESULTS The amount of venom which a snake injects on one strike is not known. But the venom yield which could be extracted from a snake and the dry weight of venom recovered were investigated (Table 1). The average venom yield varied from 1.16 ml to 0.47 ml. This may reflect a difference in individual snakes as well as differences in feeding and maintaining snakes, and techniques of venom extraction. The dry weight of venom is about 300 mg/ml. Data for venom samples used Table 1. Yield and Dry weight of Habu snake venom. * Collected at the Health Center of Nase City, Amami Islands. ** Collected at the Institute for Infectious Diseases, University of Tokyo. Habu snakes were transported from Amami Islands to Tokyo and fed at the Institute for Infectious Diseases. Table 2. Venom samples used in these experiments. * Collected at the Health Center of Amami Islands. ** Collected at the Institute for Infectious Diseases, University of Tokyo.

98 MITSUHASHI et al. Vol. 3, No. 1 Table 3. The lethal action of Habu snake venom for mice. Dried venom was dissolved in physiological saline and 0.2 ml of graded dilutions was injected. Denominator indicates the total number of mice used at each dilution and numerator indicates the number of dead mice. Table 4. MLD of various samples of Habu snake venom for mice. in these experiments and conditions of preserving venom are shown in Table 2. The lethal activity of Habu snake venom for mice was determined. As shown in Table 3, 100 Đg of a dried pool of venom collected in 1942 was the MLD for mice by intraperitoneal injection. The MLD for mice of fresh venom extracted in 1958 was 57 to 75 Đg. A sample dried in a dessicator on sulfuric acid and kept in the dessicator for 16 years still had high lethal activity. Lyophilization appears to be a satisfactory method of preserving venom toxicity. The hemolytic activity of Habu snake venom was investigated. As shown in Table 5, direct hemolytic activity was weak and 5 mg/ml of venom was necessary to produce hemolysis. Five or 2.5 mg/ml of venom was necessary for direct hemolysis of the blood of these various animals tested (Table 6). But indirect hemolysis of Habu snake venom with addition of ovolecithin was marked and about 3 Đg/ml of venom was sufficient for complete hemolysis. The HD50 of fresh venom for sheep red cells

+++ January, 1959 STUDIES ON HABU SNAKE VENOM 99 Table 5. Direct hemolytic action of Habu snake venom. Sheep blood cell was used. : complete hemolysis, +++ }: indicates the grade of hemolysis *: after incubation of 2 hours at 37 Ž. * *: after 2 hours' incubation at 37 Ž and overnight at 4 Ž. Control: without venom. Table 6. Direct hemolytic action of Habu snake venom for the blood of various animals. Read after 2 hours' incubation at 37 Ž. Table 7. Indirect hemolysis of Habu snake venom following addition of ovolecithin. Sheep red cells were used. * Collected at the Health Center of Amami Islands. ** Collected at the Institute for Infectious Diseases, University of Tokyo. was 0.5 ƒêg. Dried pooled and lyophilized samples were also active. As stated above, localized symptoms are severe in Habu bites; necrosis is severest in the parts rich in muscular tissue. The determination of the amount of hemorrhage and necrosis caused by Habu snake venom is necessary to define the activity of the venom and to assay the effectiveness of anti-serum. The dorsal skin of a scalped rabbit was injected with graded

100 MITSUHASHI et al. Vol. 3, No. 1 Fig. 1. The necrotic activity of Habu snake venom. The dorsal area of scalped rabbit was injected intradermally with graded amounts of venom and the shortest and longest diameters of the neorotic area were measured after 24 hours. The square measure of hemorrhagic necrosis was obtained by multiplying the two diameters. dilutions of venom and the longest and shortest diameters of necrosis with hemorrhage were measured after 24 hours. The value obtained by multiplying the two diameters is the square measure of hemorrhagic necrosis. When these values were plotted against log2 of the quantity of venom administered, points lay on an exponential curve as shown in Fig. 1. The necrotic activity of venom was expressed as the amount of venom causing 100 mm2 of hemorrhagic necrosis. Calculated by Table 8. Proteinase activity of Habu snake venom. * Collected at the Health Center of A mami Islands and kept at the room temperature in dessicator on calcium chloride. ** Collected at the Institute for Inf ectious Diseases, University of Tokyo.

January, 1959 STUDIES ON HABU SNAKE VENOM 101 Fig. 1, the activity of fresh venom, the dried pool collected in 1942 and a dried pool collected in 1952 were about 50, 300 and 600ƒÊg respectively. From the clinical symptoms of Habu bites and the histopathological pictures seen in experimental animals given injections of Habu venom, it was thought that the toxic effects might be caused by a proteinase. It is known that most snake venoms contain proteinases; recently Rao et al30) and Hadidian31) reported that there is more than one proteolytic enzyme in India Cobra venom. When casein was used as the substrate for the enzyme, the proteinase activity of many samples of Habu venom tested were quite stable and the activity ranged between 0.68 and 1.13 u/mg. A dried pool collected in 1942 still has a high activity of 0.68 u/mg. DISCUSSION The dangerousness and the clinical symptoms of snake bite are difficult to categorize. They are influenced by the site of the bite, the amount of venom injected and the toxicity of venom. Of course the danger of a snake to man does not depend on the toxicity of venom alone. Extremely toxic venom does not in itself make a snake dangerous to man. Toxic venom becomes a menace only if the snake makes the poison in sufficient quantity, has an effective means of delivering it and is skillful and aggressive in doing so. His habitat is also of importance. From the facts that Habu snakes make the venom in large quantity, live near house to eat mice and are aggressive, Habu snakes belong to very dangerous group. The clinical symptoms of Habu bite are divided into two groups: general symptoms (nausea, vomiting, heart weakening, etc.) and local signs (hemorrhage, necrosis, swelling, pain etc.). The characteristic pathology of Habu bites are hemorrhage and necrosis with swelling at the site of the bite. The necrosis is severest in the parts rich in muscular tissue. From the histopathological picture seen in experimental animals given an injection of Habu venom, it was thought that there might be another factor causing hemorrahage and necrosis other than hemolytic and lethal activity. In this experiment, venom yield and dry weight of venom which were extracted from Habu snakes were measured. The average venom yield was I ml. Lower venom yields of 0.57 or 0.47 might have been caused by handling, technique of extraction, conditions of maintenance and feeding of snakes which had been transported from Amami Islands to Tokyo. The necrotic activity of Habu venom aside from its lethal and hemolytic activity was investigated, and the proteinase activity measured. The results were as follows: 1) Average venom yield was 1 ml and dry weight was about 300 mg/ml. 2) Direct hemolytic activity was weak. And direct hemolytic activity for the red cells of various animals was also studied. Five mg per ml of venom was necessary for the complete hemolysis. But indirect hemolysis of venom with ovolecithin was very active and HD50 of fresh venom for sheep red cells was 0.5ƒÊg/ml. Hemolytic activity was quite stable on standing at room temperature in dried conditions. 3) MLD for mice of fresh venom was about 70ƒÊg. The lethal activity was

102 MITSUHASHI et al. Vol. 3, No. 1 stable for keeping on standing in dried conditions. Dried pool of venom collected in 1942 and stored in dessicator at room temperature still has a high lethal activity of 100 MLD for mice. 4) The activity of venom to cause hemorrhage and necrosis was determined by measuring the square measure of necrosis of rabbit skin. The activity was expressed as the amount of venom to cause 100 mm2 of hemorrhagic necrosis. This activity of fresh venom, dried pool collected in 1942 and in 1952 were about 50, 300 and 600ƒÊg respectively. Habu snake venom contains hyaluronidase.32) Then it is thought that necrotic activity may be influenced not only by proteinase but also by spreading foctors such as hayluronidase. 5) The proteinase activity of Habu venom was determined. The enzymatic and histopathological studies of Habu venom proteinase will be reported in another papers. The authours are deeply indebted to the Ministry of Education for the financial aid to carry out these experiments. And the authors, particular thanks are due to Dr. A. Yamamoto, Institute for Infectious Diseases, Tokyo and to Japan Air Self Defence Force to the transport of Habu snakes from Amami Islands to Tokyo. LITERATURE 1) M. Sasa, Y. Tanaka, K.Semaru and D. Mori: Nisshin Igaku (Japanese), 43, 489, 1956 2) Y. Sawai: Tokyo Izi Shinshi, 74, 65, 1957. 3) S. Mitsuhashi, H. Maeno and Y. Hayashi: Kanto Branch Meeting (Tokyo) of Japan Bacteriological Association, November, 1957. 4) S. Mitsuhashi, H. Maeno, T. Okonogi and K. Yamaguchi: At the meeting of Institute for Infectious Diseases, University of Tokyo, March, 1958. 5) T. Okonogi, K. Yamaguchi and S. Mitsuhashi: Japan Pathological Association, April, 1958. 6) S. Mitsuhashi, H. Maeno,T. Okonogi et al: At the meeting of Institute for Infectious Diseases, University of Tokyo, November, 1958. 7) S. Mitsuhashi, H. Maeno, Y. Hayashi, T. Okonogi and K. Yamaguchi: Sogoigaku (Japanese), 15, 19, 1958. 8) T. Kitajima: J. Bact. (Japanese), 154, 541-556, 1908. 9) S. Hosoya, S. Yani and T. Tanaka: J. Exp. Med. (Japanese), 18, 1-20, 1934. 10) T. Tanaka: J. Exp. Med. (Japanese), 25, 89-113, 1941. 11) T. Tanaka: J. Exp. Med. (Japanese), 25, 159-177, 1941. 12) T. Tanaka and Y. Ohara: Taiwan Igaku (Japanese), 40, 1, 1941. 13) T. Tanaka and Y. Kuwashima: J. Exp. Med. (Japanese), 26, 1-2, 1942. 14) T. Tanaka: J. Exp. Med. (Japanese), 26 (2), 1-2, 1942. 15) T. Tanaka, T. Miyake and Y. Ohara: J. Exp. Med. (Japanese), 26 (5), 1-2, 1942. 16) Y. Kuwashima: Advance in Medicine, Vol. 1, Kyoritsu Co. Tokyo, 1942. 17) T. Tanaka and T. Miyake: J. Exp. Med. (Japanese), 27, 81, 1944. 18) H. Kubota: Nihon Igaku, No. 3388, 387-841, 1944. 19) T. Tanaka, H. Kubota and T. Miyake: Nihon Igaku, No. 338, 835-837, 1944. 20) N. Komatsu: J. Biochem. (Japanese), 22, 96-102, 1950. 21) S. Miyazaki and H. Sekihara: Showa Igakukai Zasshi, 7, 266, 1950. 22) T. Miyake: J. Bact. (Japanese), 7, 517-510, 1952. 23) Y. Kuwashima: Jap. J. Exp. Med., 23, 21-25, 1953, 23, 127-136, 1953, 23, 225-230,

January, 1959 STUDIES ON HABU SNAKE VENOM 103 1953, 23, 299-304, 1953, 23, 457-464, 1953, 23, 465-471, 1953. 24) W. B. Martin: The Bulletin of the US Army Medical Department, 5, 79, 1946. 25) A. Ohsaka: J. Biochem. (Japan), 45, 259, 1958. 26) O. Folin and V. Ciocalteu: J. Biol. Chem., 73, 627, 1927. 27) K. H. Slotta and G. Szyska: Ber., 71, B, 258, 1933. 28) H. Maeno and S. Mitsuhashi: Symposium on enzyme chemistry, Sapporo, 1958. 29) H. Maeno and S. Mitsuhashi: This journal, to be published. 30) Rao, Shanta and Tao, S. S.: Venoms, The American Association for the Advancement of Science, 1st Ed., p. 179, 1956. 31) Z. Hadidian: ibid., p. 205, 1956. 32) H. Maeno, M. Morimura and S. Mitsuhashi: unpublished observation.