6. PALYTOXIN IN TWO SPECIES OF PHILIPPINE CRABS

Takeshi YASUMOTO and Daisuke YASUMURA Faculty of Agriculture, Tohoku University, Tsutsumi-dori, Amemiyamachi, Sendai, Miyagi 980, Japan

Yasushi OHIZUMI and Masami TAKAHASHI Mitsubishi-Kasei Institute for Life Science, 11 Minamiooya, Machida-shi, Tokyo 194, Japan

Angel C. ALCALA and Lawton C. ALCALA Department of Zoology, Silliman University, Dumaguete City 6501, The Philippines

ABSTRACT

Two species of xanthid crab Lophozozymus pictor and Demania a1ca1ai collected on southern Negros, Philippines, were found to be highly lethal by mouse assays. The toxin in both species was indistinguishable from pa1ytoxin, a highly lethal toxin of zoanthids Pa1ythoa spp., in chroma­ tographic properties, lethal potencies and in ultraviolet absorption spectra.

INTRODUCTION

In tropical Pacific areas widespread rumors exist regarding the occurrence of toxic crabs (KONOSU and HASHIMOTO, 1978). The species most frequently implicated in human intoxication is Zosimus aeneus in which occurrence of saxitoxin analogues (HASHIMOTO, 1979; YASUMOTO et a1., 1981; KOYAMA et a1., 1981; RAJ et a1., 1983) and, more recently, tetrodotoxin (YASUMURA et a1., in press) have been confirmed. However, toxic principles in other crabs implicated in poisoning incidences have remained unidentified. TEH and GARDINER (1974) first reported the presence of a potent toxin in Lophozo­ zymus pictor and suggested the toxin to be different from both saxitoxin and tetrodotoxin on the basis of dose-death time relation ships and gel permeation chromatographic properties. Incidence of human fatalities due to ingestion of this species was reported on Negros Island, Philippines (GONZALES and ALCALA, 1977; ALCALA, 1983). In the same area, a human fatality resulting from ingestion of another species Demania toxica also took place (ALCALA and

-45- HALSTEAD, 1970) and a related species Demania a1ca1ai was shown to be highly lethal (CARUMBANA et a1., 1976). We now report isolation of a toxin from specimens of L. pictor and D. a1ca1ai collected in Negros Island, Philippines. As resem­ blence between the crab toxin and palytoxin, a highly potent toxin of zoanthids, was recognized at the early stage of investigation, emphasis was laid to prove the identity of the two toxins.

MATERIALS AND METHODS

Materials Four specimens of Lophozozymus pictor and two specimens of Demania a1ca1ai were collected at Negros Island, Philippines, during the period from Nov. 1982 to May 1983. They were frozen immediately after catch and kept at -20°C until use for extraction. Reference palytoxin was generously donated by Professors D. UEMURA of Shizuoka University and Y. HIRATA of Meijo University.

Purification Pulverized crabs were extracted with twice volume (v/w) of methanol 5 times at room temperature and the combined extracts were evaporated to dryness. The residue was suspended in water, freed of lipids with diethyl ether, and the toxin was extracted with I-butanol. The toxic residue obtained after evaporation of I-but­ anol was successively treated on columns of TSK G3000S (Toyosoda, 2.8 x 5 cm), DEAE-Sephadex A-25 (Pharmacia, 2.8 x 40 cm) and CM­ Sephadex C-25 (Pharmacia, 1 x 20 cm) following the method for puri­ fication of palytoxin described by HIRATA et a1. (1979). Purifi­ cation on the DEAE-Sephadex and CM-Sephadex columns were repeated twice. Further purification of the toxin was carried out by high performance liquid chromatography (HPLC) on Develosil ODS-I0/20 column (Nomura Kagaku, 2 x 25 cm) with alternative use of the following two solvents: methanol-O.lN acetic acid (8:2) and aceto­ nitrile-0.05N acetic acid (5:5). Separation of the toxin was monitored by mouse assays and with a UV-spectromonitor (Japan Spectroscopic Co., UVIDEC-I00-II) at wavelength 263 nm. Comparison of the toxin with reference palytoxin by HPLC was carried out on an ERC ODS-1282 column (Erma Optical Works, 0.6 x 25 cm) with the aforementioned two solvents and on a TSK G3000SW column (Toyosoda, 0.75 x 60 cm) using 0.03N acetic acid as eluant.

Thin layer chromatography (TLC) TLC was carried out on Silica gel 60 and NH2F254s plates (Merck, precoated) with two different solvent systems: l-amyl­ alcohol-py.ridine-water( 7: 7: 6) and pyridine-water-l-butanol-acetic acid (10:12:15:13). The toxin on Silica gel 60 plates was detected by heating the plates after spraying sulfuric acid, and that on NH2- F254s plates by exposure to UV light (254 nm).

-46- Bioassay of toxin A bioassay method proposed by TEH and GARDINER (1974) for the toxin in L. pictor was employed. Groups of 3 male mice of ddY strain with body weight of 17 g were injected intraperitoneally with suitable volumes of sample solutions and the death times noted Toxin amounts in mouse unit (MU) was calculated from the equation Y = 225.19 X-0 •99 •

Spectra UV spctra were taken in aqueous solution with a Hitachi 124 spectrophotometer.

RESULTS

In all the chromatographic experiments, the toxin in L. pictor and D. a1ca1ai were indistinguishable from palytoxin. Chromato­ grams of the crab toxin on columns of DEAE-Sepahdex and CM-Sephadex superimposed on those of palytoxin are shown in Figs. 1 and 2. Separation of the toxin from buffer salts are achievable on the TSK G3000S polystyrene gel column, which retained the toxin from aqueous solutions and released it with 75% ethanol in a similar manner described for palytoxin by HIRATA et a1. (1979). As shown in Fig. 3, the retention times for the crab toxin on an ERC ODS- 1282 column are identical with that of palytoxin with two different solvent systems. The crab toxin is also indistinguishable from palytoxin in gel permeation chromatography on a TSK G3000SW column (Fig. 3). Thin layer chromatography on two different types of plates using two solvent systems does not distinguish the toxin (Fig. 4). Like palytoxin the crab toxin exhibits absorption maxima at 233 and 263 nm in UV spectra. Molecular extinction coefficients at 263 nm were 23,570 for the toxin of L. pictor and 19,600 for the toxin of D. a1ca1ai, if we assume the crab toxin to have the same molecular weight (2,680 dalton) with palytoxin. These values are compatible with that reported for palytoxin (MOORE et a1., 1975). All these data unamimously support that the toxin isolated from L. pictor and D. a1ca1ai is palytoxin. The amounts of pure toxin isolated from L. pictor and D. a1ca1ai a1ca1ai were 1.0 and 0.8 mg, respectively. The lethal potency as measured by the mouse assay method of TEH and GARDINER (1974) was 0.5 pg/kg for individual toxin samples. The value is compatible with that measured on the reference palytoxin. The toxin contents of crabs and anatomical distribution of the toxin in the body are summarized in Table 1. All specimens tested ar~ lethal. One specimen of D. a1ca1ai contained as high as 1.2 x 10 MU of the toxin in the body. The toxin was found in all the tissue and organs but higher concentration was found in gills and the viscera consisting of the liver and gonad. Chelipeds, espe­ cially the flesh, were low in lethal potency. Eggs of L. pictor showed the highest toxin level in the body.

-47- DISCUSSION

The results obtained in this study strongly indicate that the toxic principle in L. pietor and D. alealai is palytoxin. Al­ though further confirmation by ordinary spectral analysis, e.g. H NMR spectrometry, was unattainable due to the extremely large molecular size of palytoxin, C129H223N3054, supportive data were obtained by OHIZUMI and TAKAHASHI (unpublished) who found pharma­ cological properties of the crab toxin to be identical with those of palytoxin previously reported by OHIZUMI and SHIBATA (1980). To our best knowledge the present study is the first to evidence the occurrence of palytoxin in crabs and to confirm its implica­ tion in human intoxication. It is not certain at present whether the toxin in crabs is of endogenous or exogenous origin. It is likely, however, that crabs accumulate the toxin by feeding on Palythoa spp., which were commonly seen in the sampling areas. In four specimens. the stomachs were empty and in two specimens only skeletons of unidentified crastacea and sponges were recognized in the stomach. Effort will be continued to test the occurrence of remnants of zoanthids in the stomach of crabs. The extremely high lethal potency of the crabs is worth to note and urges that people should be well informed of the potential danger of the crabs. The case report for a human fatality caused by ingestion of D. toxiea (ALCALA and HALSTEAD, 1970) suggests that the causative toxin was also palytoxin. Effort is being made to obtain specimens of D. toxiea for analysis.

REFERENCES

ALCALA, A.C. and HALSTEAD, B.W. (1970) Human fataly due to inges­ tion of the crab Demania sp. in the Philippines. Clin. Toxieol., 3, 609. ALCALA, A.C. (1983) Recent cases of crab, cone shell, and fish in­ toxication on southern Negros Island, Philippines, Toxieon, Suppl., 3, 1. CARUMBANA, E.E., ALCALA, A.C. and ORTEGA, E.P. (1976) Toxic marine crabs in southern Negros, Philippines. , 23, 265 GONZALES, R.B. and ALCALA, A.C. (1977) Fatalities from crab poi­ soning on Negros Island, Philippines. Toxieon, 15, 169. HASHIMOTO, Y. (1979) Marine Toxins and Other Bioactive Marine Metabolites. Tokyo: Japan Scientific Societies Press. HIRATA, Y., UEMURA, D., UEDA, K. and TAKANO, S. (1979) Several compounds from Palythoa tubereulosa (Coelenterata). Pure Appl. Chem. 51, 1875. MOORE, R.E., DIETRICH, R.F., HATTON, B., HIGA, T. and SCHEUER, P.J. (1975). The nature of the 263 chromophore in the palytoxins. J. Org. Chem., 40, 540.

-~- KONOSU, S. and HASHIMOTO, Y. (1978) Venoms of crastacea and merostomata. In: Handbook of Experimental Pharmacology, Vol. 48, p. 13, "Arthropod Venoms" (BETTINI, S. Ed.). Berlin, Heidelberg, New York: Springer-Verlag. KOYAMA, K., NOGUCHI, T., UEDA, Y. and HASHIMOTO, K. (1981) Occur­ rence of neosaxitoxin and other paralytic shellfish poisonings in toxic crabs belonging to the family Xanthidae. Bull. Jpn. Soc. Sci. Fish., 47, 965. OHIZUMI, Y. and SHIBATA, S. (1980) Mechanism of the excitatory action of palytoxin and N-acetylpalytoxin in the isolated guinea­ pig vas deferens. J. Pharmc~ exp. Ther., 214, 209. RAJ, U., HAQ, H., OSHIMA, Y. and YASUMOTO, T. (1983) The occur­ rence of paralytic shellfish toxins in two species of xanthid crab from Suva barrier reef, Fiji islands. Toxicon, 21, 547. TEH, Y.F. and GARDINER, J.E. (1974) Partial purification of Lopho­ zozymus pictor toxin. Toxicon, 12, 603. YASUMOTO, T., OSHIMA, Y. and KONTA, T. (1981) Analysis of paraly­ tic shellfish toxins of xanthid crabs in Okinawa. Bull. Jpn. Soc. Sci. Fish., 47, 957. YASUMURA,D., OSHIMA, Y., YASUMOTO, T., ALCALA, A.C. and ALCALA, L.C. (in press) Screening of Philippine crabs for tetrodotoxin and paralytic shellfish poisons. Toxicon, in press.

-49- Palytoxin ~ !:..pietor toxin--o--

2000 DEAE Sephadex A-25 (2.8 em I D X 40 em) Phosphate buffer pH 6.9

0

=> :E 0 >-. +-' ''- U ~, x I , CM Sephadex C-25 (1.0em ID X20 em) 0 Phospate buffer pH I- 500 4.6 \ " "\ \ \ \ \ \ \ \ 0..._ --0..., ...... ""0...... o

o 20 40 60 80 100 ( ml )

Fig. 1. Comparison between palytoxin and L. pictor toxin on DEAE-Sephadex A-25 and CM-Sephadex C-25 columns.

-50- !:..pietor toxin O.alealai toxin --0-- OEAE Sephadex A-25 (l. 8 em ID X40 em) Phosphate buffer pH 6.9

100,000 100

0 0 ::::> :::E 0 10 20 (.ml ) ~ .j.J 'r- U >< CM Sephadex C-25(1.0 em 10 X20 em) 0 I- Phosphate buffer pH 4.6

50

o 20 40 60 80 100 ( m1 )

Fig. 2. Comparison between L. pictor toxin and D. alcalai toxin on DEAE-Sephadex A-25 and CM-Sephadex C-25 columns.

-51- - :" -1": -:-. ~::---t ~'::=i:=:: "::. .=~-~:": --+--"';---H 1-4-\-+----, r---i---HI =-.- ---,--- -_ ... -

1-="_.

Column: ERC-ODS (6.0mm ID X250 mm) Column: ERe-ODS (6.0 mm ID X 250mm) Solvent: MeOH-O.l N AcOH (8:2) Solvent: CH3CN-0.OSN AcOH (5:5) at 0.9 ml/min. at O. 9 m~l Imi n.

Column: TSK-gel G3000SW (7.5mm IDX600mm) Solvent: 0.03N AcOH at 0.9 ml/min.

Fig. 3. Comparison between palytoxin and crab toxin by high performance liquid chromatography.

-52- Silica gel 60 Rf 1.0

0

0 Q 0• 0.5 0 0 Q e 0

0 C3 8

o • • • • • • • • e· • • ABC ABC ABC A 8 C Solvent I Solvent ]I Solvent I Solvent ]I Fig. 4. Comparison between crab toxin and palytoxin by thin layer chromatography. Sample: palytoxin (A), toxin from L. pictor (B) and toxin from D. alcalai (C). Solvent: I 1-amylalcohol-pyridine-water (7:7:6); ]I pyridine-water-acetic acid (10:12:15:13). Detection: sulfuric acid for Silica gel 60 plates and UV lamp (254 nm) for NH F s plates. 2 254

Table 1. Anatomical distribution of the toxin in two crabs

Specimens Lethality of the organs (MU/g)

Date and place Sex Weight Carapace Cheliped Liver and Gills Eggs of capture (g) cuticle flesh gonad

Lophozozymus pictor Nov. 18, 1982 M 145 300 220 94 6000 2400 Zamboanguita Dec. 10, 1982 F 218 1400 170 50 6000 3000 Zamboanguita May 18, 1983 M 165 2000 400 100 10000 2200 San Jose Sep. 18, 1983 F 120 160 40 20 200 180 700 Dumaguete

Demania alcalai July 25, 1983 F 103 500 160 80 5400 2400 Dumaguete July 25, 1983 F 129 500 400 240 4000 16000 Dumaguete

-53-