Indian Journal of Experimental Biology Vol. 42, May 2004, pp. 452-460

Isolation of a haemorrhagic protein toxin (SA-HT) from the Indian venomous butterfish (Scatophagus argus, Linn) sting extract

S Karmakar, D C Muhuri, S C Dasgupta*, A K Nagchaudhuri** & A Gomes t

Laboratory of Tox inology and Experimental Pharmacodynamics, Department of Physiology, University of Calcutta, 92, Acharya Prafulla Chandra Road, Kolkata 700009, India

Received 9 April 2003 ; revised 29 January 2004

A haemorrhagic protein toxin (SA-HT) was isolated and purified from the spine extract of the Indian venomous butter­fish, S. argus Linn, by two step ion exchange chromatography. The toxin was homogeneous in native '1nd SDS-PAGE gel. SDS-molecular weight of the toxin was found to be 18.1 ± 0.09 kDa. SA-HT produced severe haemorrhage on stomach wall but devoid of cutaneous haemorrhage. UV, EDTA, trypsin, protease, cyproheptadine, indomethacin, acetylsalicylic acid and BW755C treatment significantly antagonized the haemorrhagic activity of SA-HT. The toxin produced dose and time de­pendent oedema on mice hind paw. which was significantly encountered by cyproheptadine, indomethacin and BW755C. SA-HT increased capillary permeability on guineapig dorsal flank . On isolated guineapig ileum. rat fundus and uterus. SA­HT produced slow contraction which was completely antagonised by prostaglandin blocker SC19220. On isolated rat duo­denum. SA-HT produced slow relaxation . SA-HT significr.ntly increased plasma plasmin. serum MDA level and decreased serum SOD level indicating the possible involvement of cyclooxygenase and lipooxygenase pathway.

Key words : Butterfish. Scatophagus argus. Venomous fish. Haemorrhagic toxin

IPC Code: Int. C17: AOI N 63/00

More than 200 species of marine fish among in excess of 22,000 species of fish are known or suspected to be venomous ' . The vast majority of these fish are non­migratory, slow moving and tend alive in shallow waters in protected habitats2

• Piscine envenomation, a common occurrence mainly amongst fishermen and divers3

, manifestated by intense pain at the site of en­venomation, marked increase in local vascular perme­ability, local tissue necrosis, muscular incoordination, paralysis, hypotension, respiratory and cardiac fail-

4·7 ure . Scatophagus argus (Linn.), locally known as "Py­

ratellilPyrachanda" is a venomous spotted butterfish available in waters of India, Sri Lanka, through the East Indies to China, Taiwan, the Philippines, Mela­nesia, Polynesia and Queensland (Australia). The sting of this fi sh among fisherman , fisheater produces several pathophysiological changes include intense

*Postgraduate Department of Zoology. Maulana Azad College. Kolkata 700 013. India **Department of Pharmaceutical Technology. Jadavpur Univer­sity. Kolkata 700 032. India tCorrespondent author Phone: 91-033-2244-4755 Fax : 91-033-2351-9755 & 91-033-2241-3222 Email: agomes@cllcc.ernet.in&gomesantony@hotmail.com

pain, severe swelling, redness, fever, throbbing sen­sation, thus keeping the poor fishermen out of their work for several days and lead to occupational health hazard8

• Virtually, there is no information available on the venom pharmacology, venom constituents and venom antagonists. No specific treatment is available against the sting of this fish, except some sympto­matic treatment. Muhuri et al.9 have reported the pharmacodynamic action of the S.argus venom in ex­perimental animals.

The present communication describes the isolation and purification of a haemorrhagic protein toxin (SA­HT) from the Indian venomous butterfish, Scatopha­gus argus, Linn sting extract. Some of the pharmacol­ogical properties of the toxin were examined with particular reference to haemorrhagic/oedema forming action and its antagonisms.

Materials and Methods Atropine sulphate' (Boehringer Ingelhelm, Ger­

many), histamine acid phosphate (BDH, UK), BW 755C (Burrows Wellcome, UK), acetylcholine chlo­ride, ammonium persulphate, ascorbic acid, glycine, serotonin cre~· tinine sulphate (E.Merck, Germany), SC-19220 (G.D. Se~;rle & Co., USA), acetyl salicylic acid, betalacto-globulin, bovine serum albumin, di-

KARMAKAR et al. : HAEMORRHAGIC PROTEIN TOXIN (SA-HT) FROM BUTTERFISH VENOM 453

ethyl amino ethyl cellulose, 5,5' -dithio-bis (2-nitrobenzoic acid, DTNB), Evan's blue, ovalbumin, riboflavin, RNAse, superoxide dismutase, thiobarbitu­ric acid, xanthine, xanthine oxidase, indomethacin, prostaglandin E2 (Sigma, USA), acrylamide, acetic acid, bis-acrylamide, coomassie blue, ethylene dia­mine tetraacetic acid, disodium salt, Folin phenol rea­gent, 2-mercaptoethanol, sodium dodecyl sulphate, TEMED, tris-hydroxymethyl amino methane (SRL, India), cyproheptadine (Merck Sharp Dhome, UK), mepyramine (May & Baker, UK), and methysergide bimalate (Sandoz, Switzerland) were used. Chemical and solvents otherwise not mentioned were of analyti­cal grade.

Preparation of sting extract-The sting extract was prepared after collection of live fish (S.argus) from Canning fish market (West Bengal). The dorsal pecto­ral and anal spines were cut, stored at O°C, immedi­ately transferred to the laboratory and stored at -20°C. The pooled spines were ground, homogenized with 0.9% NaCI/O.l M phosphate buffer (PH 7.2) and cen­trifuged at 10,000 rpm at 4°C for 30 min. The super­natant was used as crude venom/sting extract and stored at 4°C until further use. The sting extract was expressed in terms of protein equivalent and estimated by the method of Lowry et al. 10.

DEAE-cellulose column chromatography---ln step I, the sting extract/crude venom (20 mg) was absorbed onto a column (50x15 mm) of DEAE-cellulose equilibrated with 0.02 M phosphate buffer (PH 7.2) and with a bed volume of 10 ml. Elution was accom­plished at room temperature (28°C) at a flow rate of 20 ml hr- I with a stepwise gradient of NaCI (0.05,0.1 , 0.2, 0.3, 0.4, 0.5 and 1 M). Protein was estimated ac­cording to Lowry et al. 2

, using BSA as standard. The fraction was desalted by passing through a sephadex G-1O column (50 x 10 mm) and was subsequently lyophilised. Each fraction was tested for homogeneity and biological activity. In step II, the haemorrhagic protein fraction of the step I was subjected to rechro­matogram on DEAE-cellulose equilibrated with 0.02 M phosphate buffer (PH 7.2) and with a bed volume of 5 ml. Elution was accomplished at room tempera­ture (28°C) at a rate of 15 ml hr- I with a stepwise gra­dient of Nael (0.32, 0.35, 0.38, 0.4 and 0.5 M), dis­solved in 0.02 M phosphate buffer (PH 7.2).

Electrophoresis and determination of molecular weight-Polyacrylamide gel electrophoresis (PAGE) was carried out with the purified toxin on 7.5%, poly­acrylamide rod gels using tris glycine buffer 1 M (PH

8.3) according to the method of Davis ll . The gels were stained with 0.2% amido black 10 Band de­stained with 7% acetic acid. Sodium dodecyl sulphate (SDS-PAGE) was carried out according to the method of Weber and Osbornl2 on 7.5% acrylamide slab gel along with the marker protein, bovine serum albumin (68,000), ovalbumin (45,000), pepsin (35,000), beta­lactoglobulin (18,400) and RNAse (13,700). The gel was stained with 0.25% Coomassie Brilliant Blue R250 and destained with 7.5% acetic acid containing 5% methanol. The zones of protein bands were re­corded on the basis of the relative mobility to the marker protein of unknown molecular weight. The molecular weight of the unknown protein was deter­mined graphically.

Cutaneous haemorrhage-This was performed ac­cording to the method of Kondo et at. 13. Test material (in 0.1 ml) was injected on the shaved back of the mice intradermally. Same amount of 0.9% saline was injected in control animal. Similarly, a standard group was run with (5 f..lg in 0.1 ml) Vipera russelli russelli venom. After 24 hr, all the animals were sacrificed, skin was opened and observed for haemorrhagic le­sion. Control and experimental groups were compared against standard group which pwduced a haemor­rhagic lesion of 10 mm diameter 24 hr later.

Systemic haenwrrhage-SA-HT was injected into the stomach wall (Pyloric region) of the overnight fasted rats, under light ether anaesthesia. Normal sa­line was injected into the stomach wall of the control rats . After 2 hr, the animals were sacrificed, stomach was opened and observed for haemorrhagic lesions .

The systemic haemorrhagic activity of SA-HT was tried to antagonise with UV -exposure (2.90 x 10 15

quanta/second strength for 30 min of SA-HT), EDT A (incubation of 'SA-HT' with 10 rnM EDT A for 30 min at 37°C) treatment or enzyme digestion (incu­bated with trypsin/protease, 0.5 mg/ml of SA-HT at 37°C for 30 min) and compared with same dose of untreated SA-HT on rat stomach. In case of chemical antagonists treatment, cyproheptadine (10 mg/kg, sc), indomethacin (10 mg/kg, sc), acetyl salicylic acid (130 mg/kg, sc) and BW 755C (15 mg/kg, sc) were given 1 hr before the SA-HT i :ljection.

Paw oedema-Paw oedema was measured in Swiss albino male mice according to the method of Wang and Tengl4. In chemical antagonists assessment, the animals were divided into control and treated group. The treated group of animals were pretreated with cyproheptadine (10 mg/kg, sc), indomethacin 10

454 INDIAN J EXP BIOL, MAY 2004

mg/kg, sc) and BW 755C (15 mg/kg, sc) before 1 hr of SA-HT injection. The control group of animals received vehicle only.

Capillary permeability-This was tested on guinea pig by blue dye extravasation method of Kellet l5

. The dorsal region of guinea pigs were depilated with 5% sodium sulphide on the previous day. Test material (or normal saline for control) was injected intrader­mally. After 30 min, Evan's blue (60 mg/kg) was in­jected intravenously. After 30 min the animals were sacrificed and skin was removed. Two parameters were measured - (a) diameter of the dye extravasation areas (corresponding to the site of injection) "on the inner surface of the skin, (b) the amount of dye ex­tracted from the skin. For measuring the amount of dye, skin of the appropriate region was removed, sub­divided into several pieces and immersed in a solution containing 6 ml of 0.5% aqueous solution of sodium sulphate and 14 ml of sodium acetate with mild shaking for 24 hr. Thereafter the solution was centri­fuged (3000 rpm x 10 min) and the supernatant was read at 620 nm in a spectrophotometer l6

•

Isolated smooth muscle-Isolated smooth muscles (Guinea pig ileum, rat fundal strip, rat uterus and rat duodenum) were suspended in oxygenated physio­logical salt solution and contractions were recorded

. h fl" I 17-19 WIt a ronta wntll1g ever

Mast cell degranulation-This was performed after modified method of Lagunoff and Benditt2o

• Perito­neum was collected from male albino Wistar rats . The omentum was place in separate watch glass in pres­ence of (0.5 ml) normal saline (0.9%, for control), 0.2% substance 48/80 (as standard) and two different doses (5, 10 Ilg) of test material. It was incubated at 37°C for 30 min. After incubation, the peritoneum from each group was put on clear slide in perfectly stretching condition and pass over flame for its fixa­tion. The slide was then treated with ethanol: chloro­form : acetic acid (6:3: 1) for 15 min. After down grading treatment with alcohol: water, the slides were stained with 0.1 % aqueous toluidine blue. The excess stain was washed with distilled water and after up­grading treatment with ethanol: water up to absolute ethanol, the slides were washed twice with xylol and mount with DPX. The mast cell degranulation was observed under the microscope and expressed in terms of % degranulation against control preparation.

Plasmin assay--Plasmin assay was carried out on the plasma of control and treated rats by the method of Friberger et al. 21. The substrate used for this assay

was S-2251 (H-D-Val-Leu-Lys-pNA), a synthetic chromogenic substrate. The method for the determi­nation of activity is based on the difference in absorb­ance (00) between the pNA formed and the original substrate. The rate of pNA formation, i.e.; the in­crease in absorbance per second at 405 nm is propor­tional to the enzymatic activity and is conveniently determined with a spectrophotometer.

MDA activity-Serum MDA level was measured according to the method of Chatterjee and Agarwaf2. In a test tube 75 III serum (contro!lexperimental), 10 III CaCh (0.68 mg/ml) and 10 III ascorbic acid (1.76 mg/dl) was added with 1 ml of phosphate buffer (0.1 M; pH 7.5), mixed and incubated for 1 hr at 37°C. After incubation, 20 III EDTA (0.5 mM), 2.5 ml TCA (20%)and 1 ml TBA (6.7 mg/ml) was added to the test tube, mixed and place in a boiling water bath for 10 min. The test tubes were cooled and the absorb­ance was measured at 532 nm against water blank.

SOD activity-The superoxide dismutase assay was carried out from the serum of control and fish venom (SA-HT) treated rats according to the method de­scribed by Beauchamp and Fridovich23. Three cuvets were marked as 'control', 'SOD control' and 'sam­pie'. 1 ml of assay buffer (50 mM Na2C03 + 10-4 M EDT A, pH 10.2) was taken in all the cuvets. Then 10 III of serum was added to the 'sample' and 40 III of SOD (50 mM, pH 10.2) was added to the SOD control cuvet. Reading (0 min) was taken of all the cuvets at 560 nm. Then Xanthine 25 III (0.87 mg/ml Na2C03 buffer) and Xanthine oxidase 15 III (50 mM, pH 10.2) was added to all the cuvets and mixed. Finally 3 III of NBT was added to all the cuvets and reading was taken at 1,2,3,6,7, 10 and 15 min at 560 nm.

All results were expressed as mean ± SE and the significance of the differences between means was determined by Student's 't' test. P values < 0.05 were considered significant.

Results Isolation and purification-The two-step purifica­

tion procedure was used for the isolation of homoge­neous haemorrhagic protein toxin (SA-HT) from S. argus sting extract. In first step the crude venom was resolved into seven major peaks on a DEAE-cellulose column (50 x 15 mm) (Fig. 1). Peak VI eluted with 0.4 M NaCI in phosphate buffer produced severe haemorrhage on the stomach wall of the rat by local injection but failed to produce cutaneous haemor­rhage. About 13.3 times purification and 10.05%

KARMAKAR et al. : HAEMORRHAGIC PROTEIN TOXIN (SA-HT) FROM BUTTERFISH VENOM 455

yield was achieved with this step of column chroma­tography. During homogeneity testing, the haemor­rhagic protein produced two bands on polyacrylamide gel electrophoresis after staining with coomassie blue and subjected to rechromatogram (step II).

In step II DEAE-cellulose column chromatography, the haemorrhagic fraction of the step I (Peak VI) re­solved into four protein peaks (Fig. 2). Among these four peaks, the peak III eluted with 0.38 M NaCl in phosphate buffer possess severe haemorrhagic activity on rat stomach and the other peaks were devoid of haemorrhagic activity. About 40 fold purification and 4.15% yield was achieved with this step (Table 1). The haemorrhagic fraction was provisionally named

DEAE-Cellulose column Chromatography (Step-I) )·0

09 ~ 00-3 <D <D

4: o o

0·2

0.05104 0·2104 0 ·4101 '104 t I 'OH .. " • a_3M' t 0-5M 1 •

NaCL gradient

45 Tube No.

Fig. I-Ion exchange chromatography of Scatophagus argus sting extract on DEAE-Cellulose column (Step I) [- Haemorrhagic fraction.]

as SA-HT (where SA=Scatophagus argus; HT = haemorrhagic toxin).

Homogeneity, SDS-molecular weight-Step-II re­chromatogram haemorrhagic fraction SA-HT was homogeneous as confirmed by native PAGE and SDS-PAGE. The SDS-molecular weight of the SA­HT was calculated to be 18.1 ± 0.09 kDa (Fig. 3).

Cutaneous haemorrhage-Cutaneous haemorrhage was compared against standard group, where Vipera russelli venom (5 Jlg in 0.1 ml, id, n=6) produced a mean diameter of 10.12 ± 0.30 mm haemorrhagic spot on dorsal flank of mice. In both the 0.9% saline (con­trol) and SA-HT (5 and 10 Jlg in 0.1 ml, id, n=6) in-

o <D <D «

0·15

8 0.07

DEAE-Cellulose Column Chromatography(Step-lI)

NoCI-gradient 0·32M a-35M 0'38M 0·4 M 0-5M

OIL-____ ~ ______ ~ ____ ~~A-

o 5 10 Tube No.

Fig. 2-Rechromatography (Step II) of peak VI ( 0.4 M NaC\). eluted from the first chromatogram on DEAE cellulose column. [- Haemorrhagic fraction]

Table I--Purification of SA-HT by DEAE cellulose ion exchange chromatography from spotted butterfish (S. argus) sting extract

Purification steps Total Protein MHD* Total No. Fold of Yield (mg) (J.lg) ofMHD purification (%)

S.argus Crude venom 20 200 100 100

DEAE-purified haemorrhagin Peak VI (1 5

' step chromatogram) 2.01 IS 133.3 13.3 10.05

DEAE-purified haemorrhagin Peak III (2nd step chromatogram) 0.83 5 160 40 4.15

*MHD = Minimum haemorrhagic dose (J.Lg)

456 INDIAN J EXP BIOL, MAY 2004

jected animals, no haemorrhagic lesions was observed (naked eye and under magnified lens).

Systemic haernorrhage-SA-HT at different doses (1, 2, 4, 5, 6 ).!g in 0.1 ml, n=6) produced haemor­rhage on the smooth muscle of the stomach wall of the rats by local injection. The mean diameter of haemorrhagic lesions for the given doses on stomach waJI was 0, 3.50 ± 0.18, 8.42 ± 0.12, 10.50 ± 0.20, 11.25 ± 0.55 mm respectively. The minimum haemor­rhagic dose (MHD), which produced haemorrhagic spot of about 10 mm for SA-HT was found to be 5 ).!g.

It was found that, the UV treatment caused signifi­cant reduction of the haemorrhagic activity of SA-HT by about 39.33 ± 0.12% (P < 0.001, n=7). Similarly, EDTA (10 mM) treatment of SA-HT significantly reduced its haemorrhagic activity about 34.72 ± 0.18% (P < 0.001 , n=7) as compar.ed to untreated same dose of SA-HT. It was found that proteolytic enzymes trypsin (0.5 mg/ml) and protease (0.5 mg/ml) treatment significantly reduced the haemor­rhagic activity of SA-HT about 73.80 ± 0.02% and 78.57 ± 0.10% (P < 0.001, n=7) respectively.

5·0r-------------------,

I- 4·5 ~ ...J o ~

<.:> o ...J

4·0

BOVINE SERUIot ALBUIotIN(68.000)

OVALBUIotINl,5.0 00)

PE PSIN 135.000)

1-________ ~{J-LACTOGLOBULINI18AOO) ·,A-HTl18.000)

• RNA 05.113.700)

OL-____ ~ ___ ~_~ ____ __' o 0·5 1·0 1·5

MOBILITY <em)

Fig. 3--Determination of SDS-PAGE molecular weight of SA­HT.

Pretreatment of rats with cyproheptadine, indo­methacin, acetyl salicylic acid and BW 755C showed 29.46 ± 0.18,64.95 ± 0.12,48.89 ± 0.08 and 76.14 ± 0.21 % (P < 0.001, n=7) protection against SA-HT induced haemorrhagic activity when compared to same dose SA-HT induced haemorrhage on untreated control rats.

Paw oedema--It was found that SA-HT at different doses (10, 20, 30, 40, 50 ).!g in 0.01 ml), produced dose dependent oedema (20, 35, 65, 100, 110% swel­ling) on mice left hind paw when compartd to 0.9% saline injected right hind paw control within 1 hr of observation. 100% swelling (oedema) was observed to occur with 40 ).!g of SA-HT. It was also observed that SA-HT produced time dependent oedema which reached its peak after 2 hr of injection of toxin. The oedema returned to normal level after 4 hr of obser­vation (Table 2).

Cyproheptadine, indomethacin and BW 755C pre­treatment significantly reduced the oedematous effect of SA-HT by about 34.18 ± 0.14, 26.43 ± 0.06 and 37.37 ± 0.16% (P < 0.001, n=6) respectively (Table 3).

Capillary permeability--SA-HT (7 ).!g) signifi­cantly increased cutaneous capillary permeability as compared with the control. The diameter of the area (mm) of 'bluening' (control=1.83 ± 0.40, SA­HT=12.83 ± 0.40*, n=6, P<O.OOI) as well as the amount ().!g/gm of tissue) of dye (control:::7 .00 ± 0.36, SA-HT=46.33 ± 0.55, n=6, *P<O.OOl) was signifi­cantly increased by SA-HT on guinea pig dorsal flank.

Isolated smooth muscle-On isolated guinea pig ileum, SA-HT (6 ).!g/ml) produced a slow contraction, which was unaffected by pre-treatment of atropine (10-7 g/ml), or mepyramine (10-7 glml) or methyser­gide (2 x 10-7 g/ml). However, SA-HT induced slow contraction was completely blocked by the pre­treatment with SC 19220 (3 x 10-5 glml).

On isolated rat duodenum, SA-HT (6 ).!g/ml) pro­duced a slow relaxation. On isolated rat fundus and uterus preparations, SA-HT (6 ).!g/ml) produced a

Table 2-Time response of SA-HT induced hind paw oedema in mice [Values are mean ± SE from 6 animals in each group]

SA-HT ()..lg)

40

30 60

39.66 + 0.42 63.50 + 0.42

% rise of swelling with time (min) 120 180 240

99.53 ± 0.23 45.33 ± 0.33 6.56 ± 0.49 .

KARMAKAR et al. : HAEMORRHAGIC PROTEIN TOXIN (SA-HT) FROM BUTTERFISH VENOM 457

Table 3--Effect of cyproheptadine, indomethacin and BW 755C on oedema produced by SA-HT [Values are mean ± SE, from 6 animals in eachgroup,p<O.OOI]

Group of animals

SA-HT (~g)

Dose of antagonist (mg/kg, sc)

% increase in paw diameter (2 hr)

Protections · (%)

I. Control (SA-HT only)

II. Treated

a. Cyproheptadine b. Indomethacin c. BW755C

P<O.OOI

40

40 40 40

10 10 5

slow contraction. In both the cases, this slow contrac­tion remained unaffected by pre-treatment of atropine and cyproheptadine, while it was completely antago­nized by the pre-treatment with SC 19220.

Mast cell degranulation-SA-HT (5 and 10 J!g) significantly degranulated peritoneum mast cells in dose dependent manner when compared to 0.9% sa­line treated control (n=6). The percentile mast cell degranulation in control was (7.33 ± 0.45%), SA-HT was (5 J!g dose, 35 .83 ± 1.30%, 10 J!g dose 63.16 ± 1.42%, p<O.OOl) and compound 48/80 was (82.83 ± 1.25%).

Action on plasma plasmin, serum MDA and SOD level-SA-HT (5 J!g) treatment significantly in­creased plasma plasmin level as compared to control animals. Plasmin level (unit/ml plasmin) was found 0.245 ± 0.01 in control normal rats where as SA-HT treated rats showed 0.536 ± 0.006 (n=6, P<O.OOI). Similarly, SA-HT (5 J!g) treatment significantly in­creased the serum MOA level after 2 hr of toxin treatment and serum MOA level returned towards normal after 4 hr. Serum MOA level (nM/ml) in con­trol was 3.52 ± 0.28 where after 2 hr of SA-HT (5 J!g) treatment 8.75 ± 0.61 * (n=6, *P<O.OOl) which re­turned to normal after 4 hr (3.43 ± 0.41) indicating increased lipid peroxidation. It was also found that, SA-HT (5 J!g) treatment decrease in SOD level of the serum of the treated group (inhibition of 00=18.2 ± 0.09%) at 3 min in comparison to the control group (inhibition of 00 = 45.50 ± 0.10%).

Discussion Fish envenomation is a major problem . among

fishermen and fisheater. Piscine venoms are mainly proteinaceous in nature and it is perhaps not surpris-

99.00 ± 0.36

65.16 ± 0.30* 72.83 ± 0.40* 61.00 ± 0.44*

34.18 ± 0.14 26.43 ± 0.06 37.37 ± 0.16

ing that all piscine venoms produce similar symptoms in envenomed humans, in particular extreme pain quite disproportionate to the size of the injury24.25. A variety of toxins have been isolated from the venoms of fishes like SNTX, Dracotoxin, VTX, Toxin-PC, TL Y, Trachinine, Nocitoxin26.32, having different pharmacological activitl3. The existing therapeutic management to combat fish envenomation are symp­tomatic treatments which include anti-inflammatory, analgesic drugs, to extend serum therapy. Venomous edible spotted butterfish Scatophagus argus sting ex­tract produced systemic haemorrhage, oedema in ex­perimental animals34. In the present communication, a haemorrhagic protein (SA-HT) was isolated from the sting extract of S.argus and purified by two steps ion exchange chromatography. The SDS-MW of the SA­HT was 18.1 ± 0.09 KDa. SA-HT produced haemor­rhage into the stomach wall by local injection, but failed to produce any significant haemorrhagic effect by intradermal injection (cutaneous haemorrhage). From the piscine venoms such type of haemorrhagin was not isolated till date but a systemic haemorrhagin (VRH-l) was isolated from th.; Vipera russelli russelli venom by subjecting the venom twice successively on CM-Sephadex C-50 column chromatography. Intra­dermal administration of this haemorrhagin in mice resulted in severe lung haemorrhage but produced little haemorrhage in skin35 .

The haemorrhagic action of SA-HT was reduced significantly by EDT A treatment of the toxin, which indicated the influence of metals on the haemorrhagic action of SA-HT. The protein structure of SA-HT was influenced by UV -ray and enzyme pre-treatment, thereby reducing the haemorrhagic activity. Cypro­heptadine (histamine and serotonin blocker), indo­methacin and acetyl salicylic acid (both are cyclooxy-

458 INDIAN J EX? BIOL, MAY 2004

genase pathway blocker) and BW 755C (cyclooxy­genase and lipooxygenase pathway blocker) pre­treatment of animals reduced the SA-HT haemor­rhagic effect. The oedematous effect of SA-HT was significantly antagonized by same chemical antago­nists . These findings suggest the involvement of his­tamine, serotonin, prostaglandins and leukotrienes with the haemorrhagic and proinflammatory actions of SA-HT. SA-HT significantly degranulated the mast cells, suggesting the histamine, serotonine, prosta­glandins and leukotrine mediated action of SA-HT. These liberators were present in the granular compo­nents of mast cells36.37. These mediators induced ac­tion of SA-HT was also supported by the SA-HT in­duced increased capillary permeability on guinea pig dorsal flank. The basic phospholipase A2 isolated from the Trimeresurus mucrosquamatus snake venom produced oedema by degranulating the mast cells. The oedema formation was inhibited by the pre­treatment with dexamethasone, indomethacin, and diphenhydramine. The purified PLA2 also resulted in an increase in vascular permeability which could be decreased by pre-treatment with three anti­inflammatory drugs38

• There was no report available on the fish venom haemorrhagins. So, the haemor­rhagins from different snake venoms were compared with SA-HT (Table 4). .

The slow contractile effect of SA-HT on isolated smooth muscle preparations was mediated through the prostaglandin receptors as the action of SA-HT was antagonised by prostaglandin receptor blocker SC 19220. It is unlikely that, being a protein, SA-HT may act directly through the PG receptors. It is probable that, SA-HT may cause the release of PG-like sub­stances, which ultimately produced the slow contrac­tion. The skin toxin of Arabian Gulf Catfish (Arius thalassillus, Ruppell) caused a release of prostaglan-

din (PGE2, TXB2 and 6-Keto-PG 1u) from the arterial specimens into the organ bath41. The venoms of the stonefish (Synanceja trachynis) and soldierfish (Gym­napistes marmoratus) also produced contractile re­sponses in isolated guineapig ileum, which were at­tributed to the venoms containing or releasing both acetylcholine and prostaglandins42,43 . The venoms of the Indian catfish Heteropneustes fossilis and Ploto­sus canius both produce contractile responses in a number of smooth muscle preparations, which are attributed to the venoms containing or releasing pros­taglandins44

,45. SA-HT produced slow relaxation on isolated rat duodenum-a kinin like effect. Similarly, Channa striatus skin ex~ract also produced relaxation of isolated rat duodenur(l,ls.

SA-HT significantly increased plasma plasmin and serum MDA level but reduced SOD level in experimental animal which enlight the mechanism of action of SA-HT induced haemorrhage. Previously it was observed that the haemorrhagic action of SA-HT was mediated through histamine, serotonine, prostaglandin and leukotrienes as their blocker antagonised the haemorrhagic action of SA-HT. SA­HT produced relaxation on the isolated rat duodenum, which indicated that SA-HT may have caused the release of kinin-like substances. SA-HT may have caused the release of these autacoids to produce haemorrhage. This haemorrhage formation was involved with the increased level of serum malondialdehyde (MDA) formation and decreased level of antioxidant scavenger, superoxide dismutase (SOD). These findings suggest the increased lipid peroxidation by SA-HT. During lipid peroxidation, the destruction of unsaturated fatty acids by the superoxide anions may have caused the excessive formation of prostaglandins and leukotrienes through the cyclooxygenase and lipooxygenase pathway.

Table 4--Comparison between SA-HT and snake venom haemorrhagins

Origin (naming)

Agkistrodoll acufus (ACI)

Crotalus afrox (Ht-d)

Scatophagus argus (SNHT)

Molecular weight (/-lg), (kDa)

24.5

24

18

MHD Inhibitors Reference

0.22 EDTA, 1,1O-phenanthroline and cysteine Nikai ef al.39

II EDTA, 1,1O-phenanthroline, amino acid Bjarnason & Tu40

hydroxamate

5 EDT A, cyproheptadine, indomethacin, acetyl Present study salicylic acid, BW 755C

KARMAKAR et al. : HAEMORRHAGIC PROTEIN TOXIN (SA-HT) FROM BUTTERFISH VENOM

These fatty acid metabolites, then may act on thevessel wall, cause endothelial destruction and producelocal oedema and haemorrhage. SA-HT inducedhaemorrhage was also involved with the increasedlevel of serum plasmin level. This plasmin then mayalso act on the blood vessel wall, stimulate increasedformation of kinin and may have further aggravatedthe situation. The ultimate result of this complexsystem will cause haemorrhage and bleeding in thestomach wall.

The above experimental data may serve as a basistowards the use of chemical antagonists in the man-agement of local effects in fish envenomation. Serumtherapy has been reported as an alternative treatmentusing an experimental antivenom. Preliminary studiesexamining the cross-reactivity of the venoms of fivevenomous fish (Pterois volitans, Pterois antennata,Pterois lunulata, Dendrocliirus zebra and Inimicusjaponicus) with antivenom raised against the venomof stonefish (Synanceja trachynis) found that the anti-venom neutralized both the lethal and haemolytic ef-fects of these venorns'". In another report, the anti-venom used was raised in rabbits showing its abilityto neutralize lethality, necrosis, nociception and oe-dema both by pre-incubating the fish venom (Thalas-sophryne nattereri) with anti venom before injectioninto mice or by independent injection of venom andantivenom'", It was also found that stonefish anti-venom effectively neutralises the ill vitro and ill vivoactivity of both Gymnapistes marmoratus venom andP li 49 I h . .. vo uans venom . n t e present commumcation,the action of SA-HT was tried to antagonise mainlywith chemical antagonists but development of antise-rum against SA-HT and their interaction could be aninteresting area of exploration though in contrast itwas reported that, studies using stonefish antivenomand N. robusta venom found no evidence of cross re-activit/G. Further work is warranted on the neutrali-zation of active principles of sting extract of butterflshby chemical antagonists as well as antiserum for bet-ter therapeutic management.

AcknowledgementThe authors are grateful to Late Prof. S.c. Lahiri,

for his encouragement and valuable suggestions. Thispaper is dedicated in his memory.

References1 Russell FE, Toxic effects of animals toxins in: Casarett and

Doull's toxicology - The basic science of poisons, edited byCD Klaasen (McGraw Hill, Sydney) 1996.

459

2 Maretic Z, Fish venoms. in Handbook of natural /oX;II.1' :

Marine toxins and venoms, edited by A T Tu (Marcel Dek-ker, New York) 1988,445.

3 Khoo H E, Hon W M, Lee S H & Yuen R, Effect of stonus-toxin (Lethal factor from Synanceja horrida venom) onplatelet aggregation, Toxicon, 33 (1995) 1033.

4 Kreger A S, Detection of a cytolytic toxin in the venom ofthe stone fish tSynanceja trachyniss, Toxicon, 29 (1991) 733.

5 Khoo H E, Yuen R, Poh C H & Tan C H, Biological activi-ties of Synanceja horrida (stonefish) venom, Narum! Toxins.1 (1992) 54.

6 Goudey-Perriere F & Perriere C, Pharmacological propertiesoffish venoms. C R Soc Bioi, 192 (1998) 503.

7 Sauviat M P, Meunier F A, Kreger A & Molgo .I, Effects oftrachyuilysin, a protein isolated from stonefish iSvnoncejatrachynisy venom, on frog atrial heart muscle, Toxicon, 38(2000) 945.

8 Muhuri D C, Studies Oil fish venom iScatopltagus argus),Ph.D. Thesis, Jadavpur University, Calcutta (1996).

9 Muhuri D, Karmakar S, Dasgupta S C, Naghchaudhuri A K& Gomes A, Pharmacological studies on the venomousspotted butterfish (Scatophagus argus) sting extract on ex-perimental animals. Indian J. Exp. Biol,42, (2004) (in press).

10 Lowry 0 H, Rosebrough N J, Farr A L & Randall R .T, Pro-tein measurement with the folin phenol reagents, J BiolChem, 193 (1951) 265.

11 Davis B J, Disc electrophoresis II. Method and application tohuman serum proteins, Allll NY Acad Sci, 121 (1964) 404.

12 Weber K & Osborn M, The reliability of molecular weightdetermination by dodecyl sulphate polyacrylamide gel elec-trophoresis, J Bioi Chem, 244 (1969) 4406.

13 Kondo H, Kondo S, Ikezawa H, Murata R & Ohsaka A,Studies on the quantitative method for the determination ofhaemorrhagic activity of Habu Snake Venom, Japan J MedSci Bioi, 13 (1960) 43.

14 Wang J P & Teng C M, Rat paw oedema induced by trimu-case-II, a kinin-releasing enzyme from. Trimeresurus 1Il1l-

crosquamatus venom, Eur J Pharmacal, 157 (1988) 61.15 Kellet D N, On the antiinflammatory activity of protamine

sulphate and of hexadimethine bromide, inhibitors of plasmakinin formation, Br J Pharmac, 24 (1965) 705.

16 Harada M, Takeuchi M, Fukao T & Katagiri K, A simplemethod for quantitative dye extravasated into the skin, JPharmac Pharmacol, 23 (1971) 218.

17 Vane J R, A sensitive method for the assay of 5-hydroxytryptamine, Br J Pharmac, 12 (1957) 344.

18 Dejalon P G, Bay Bayo J & Dejalon M G, Pharmacologicalexperiments 011 isolated preparations (Staff of the Depart-ment of Pharmacology, University of Edinburg, Edinburg)1945,92.

19 Hortan E W, Human urinary kinin excretion, HI' J Pharmac,14 (1959) 125.

20 Lagunoff D & Benditt P, Mast cell degranulation and hista-mine release observed in a new ill vitro system, J Exp Mal,112 (1960) 571.

21 Friberger P, Knos M, Gustavson S, Aurel L & Claeson G,Methods for determination of plasmin, antiplasmin andplasminogen by means of substrate S-2251. Haemostasis, 7(1978) 138.

460 INDIAN J-EXP BIOL, MAY 2004

22 Chatterjee S N & Agarwal S, Liposomes as membrane model for study of lipid peroxidation, Free Radical Biol Med, 4 (1980) 57.

23 Beauchamp C & Fridcivich I, Superoxide dismutase : Im­proved assays and an assay applicable to acrylamide gels, Anal Biochem, 44 (1971) 276.

24 Klaassen C D & Watkins J B, Casarett and Douli's Toxicol­ogy: The basic science o/poisons (McGraw-Hili, New York) 1999.

25 Sutherland S K & Tibballs J, Australian animals toxins : The creatures, their toxins and care 0/ the poisoned patient, Sec­ond edition (Oxford University Press, Melbourne) 2001.

26 Poh C H, Yuen R, Khoo H E, Chung M C M, Gwee M C E & Gopalakrishnakone P, Purification and partial characteri­zation of Stonustoxin (lethal factor) from Synallceja horrida venom, Comp Biochem Physiol, 99 (1991) 793.

27 Chhatwal I & Dreyer F, Isolation and characterization of dracotoxin from the venom of the greater weever fish Tra­chin us draco, ToxicOIl, 30 (1992b) 87.

28 Garnier P, Goudey-Perriere F, Breton P, Dewulf C, Petek F & Perri ere C, Enzymatic prope11ies of the stonefish (Synallceja verrucosa Bloch and Schneider, 1801) venom and purification of a lethal , hypotensive and cytolytic factor, ToxicOIl, 33 (1995) 143.

29 Auddy B & Gomes A, Indian catfish (Plotoslls callius, Ham­ilton) venom. Occurrence of lethal protein tox in (toxin-PC), Adv Exp Med Bioi, 391 (1996) 225.

30 Colasante C, Meunier F A, Kreger A S & Molgo J, Selective depletion of clear synaptic vesicles and enhanced quantal transmitter release at frog motor nerve endings produced by trachynilysin, a protein toxin isolated from stonefish (SYllanceja trachYllis) venom, Eur J Net/rosci, 8 (1996) 2149.

31 Perriere C, Le Gall G, Grosclaude J M. Garnier P, Dewulf C & Goudey-Perriere F, Storage influence on stonefish venom components activity, ToxicOII, 36 (1998) 1313.

32 Hahn S T & O'Connor J M (2000), An investigation of the biological activity of bullrout (Notesthes robusta) venom, ToxicOII, 38 (2000) 79.

33 Church J E & Hodson W C, The pharmacological activity of fish venoms, ToxicOII, 40 (2002) 1083.

34 Muhuri D, Karmakar S. Dasgupta S C, Nagchaudhuri A K & Gomes A, Lethal, oedema, haemorrhagic activity of spotted butterfish (Scatophagus argus, Linn) sting extract and its neutralization by chemical and biological antagonists. Toxi­con (Communicated 2003).

35, Chakrabarty D, Bhattacharyya D, Sarkar H S & Lahiri S C, Purification and partial characterization of a haemorrhagin (VRH-I) from Vipera russelli russelli venom, Toxieoll , 31 (1993) 1601.

36 Blood G D & Haegermark H, A study on morphological change and histamine release induced by compound 48/80 in rat peritoneal cell, Exp Cell Res, 40 (1965) 637.

37 Edward A M, Inflammation and anti-inflmmatories (John Wiley, New York) 1977, 1.

38 Chiu H F, Chen I J & Teng C M, Oedema formation and degranulation of mast cells by a basic phospholipase A2 puri­fied from Trimeresurus mucros-quamatus snake venom, Toxieon , 27 (1989) 115.

39 Nikai T, Ishisaki H, Tu AT & Sugihara H, Presence of zinc in proteolytic haemorrhagic toxin isolated from Agkistrodon acutus venom, Comp Biochelll Physiol, 72 (1982) 103.

40 Bjarnason J & Tu A T, Haemorrhagic toxins from western diamond back rattlesnake (c. atrox) venom: Isolation and characterization of five toxins and the role of zinc in haemor­rhagic toxin, Biochemistry, 17 (1978) 3395.

41 AI-Hassan J M, Thomson M, Ali M, Fayad S, Elkhawad A, Thulesius 0 & Criddle R S, Vasoconstrictor components in the Arabian Gulf Catfish (Arius thalassillus, Ruppell) protei­naceous skin secretion, ToxicOII , 24 (1986) 1009.

42 Hopkins B J, Hodgson W C & Sutherland S K, Pharmacol­ogical studies of stonefish (Synallceja /rachynis) venom, ToxicOIl, 32 (1994) 1197.

43 Hopkins B J, Hodgson W C & Sutherland S K, An in vitro pharmacological examination of venom from the soldierfish GYlllllapistes marinoratus, Toxicon, 35 (1997) 10 II.

44 Datta A, Gomes A, Sarangi B, Kar P K & Lahiri S C, Phar­macodynamic actions of crude venom of the Indian Catfish Heteropneustes fossilis, Indian J Med Res, 76 (1982) 892.

45 Auddy B, Alam M I & Gomes A, Pharmacological actions of the venom of the Indian catfish (Pl%sus canius Hamilton), Illdiall J Med Res, 99 (1994) 47.

46 Karmakar S, Pharmacological alld toxicological studies on snake head fish Channa sll'iatus, Ph.D. thesi s, Calcutta Uni­versity, Kolkata, 2003.

47 Shiomi H, Hosaka M, Fujita S, Yamanaka H & Kikuchi T, Venoms from six species of marine fish: lethal and haemo­lytic activities and their neutralization by commercial stone­fish anti venom, Mar Bioi, 103 (1989) 285.

48 Lopes-Ferreira M, Moura-da-Silva A M, Mota I & Takehara H A, Neutralization of Thalassophryne nattereri (niquim) fi sh venom by an experimental anti venom, Toxicon, 38 (2000) 1149.

49 Church J E & Hodgson W C, Stonefish (Synallceja spp) anti­venom neutralises the in vitro and in vivo cardiovascular ac­tivity of soldierfish (Gymnapistes marmora/us) venom, T-oxi­con, 39 (2001) 319.

50 Hahn S T & O'Connor J M, An investigation of the biologi­cal activity of bullrout (Notest/les robusta) venom, Toxicon, 38 (2000) 79.