View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by ScholarSpace at University of Hawai'i at Manoa

Review of Clupeotoxism, an Often Fatal Illness from the Consumption of Clupeoid Fishes1

John E. Randall2

Abstract: Poisoning from eating clupeoid fishes such as sardines and herrings (Clupeidae) or anchovies (Engaulidae), termed clupeotoxism, is widespread in tropical and subtropical areas of the world but rare. A fatal case occurred in Kaua‘i in 1978 from the consumption of the Marquesan Sardine (Sardinella marquesensis). This species has been replaced in abundance in the Hawaiian Is- lands by another import, the Goldspot Sardine (Herklotsichthys quadrimaculatus). Onuma et al. (1999) obtained the head of a specimen of this sardine that caused a fatality in Madagascar and found that it contained palytoxin. Because bottom sediment was detected on the gills and in the esophagus, they concluded that the fish is a bottom-feeder, and the benthic dinoflagellate Ostreopsis siamensis, known to produce palytoxin, the toxic organism. The sediment on the gills was more likely the result of the fish being dragged over the substratum by a seine. The Goldspot Sardine feeds on zooplankton, not benthic organisms. Therefore, a pelagic dinoflagellate is the probable producer of palytoxin.

The consumption of certain tropical ma- eating a sardinelike fish known as Clupea rine fishes, even though well cooked, may re- thryssa (¼ the thread herring Opisthonems sult in severe illness and even death. Halstead oglinum) at what is now the Dominican Re- and Lively (1954) separated such poisonings public. Oldendorp (1777) claimed that sprat into four groups, ciguatera, tetraodon poi- (a general English common name for a clu- soning, gymnothorax (moray eel) poisoning, peid) is the most poisonous fish in the Virgin and scombroid (tuna) poisoning. In a review Islands, adding that it is toxic only in circum- of ciguatera, Randall (1958) pointed out that scribed areas and at certain times of the gymnothorax poisoning is severe ciguatera, year. Halstead (1967:610) listed 87 references and scombroid poisoning is usually food poi- dealing with clupeoid poisoning. He con- soning (unrefrigerated tunas and bonitos spoil cluded with the following statement, ‘‘It is quickly by bacterial action). He briefly men- noteworthy that to date no one has reported tioned other valid categories of toxemia from any experimental work on the toxicology, eating fishes, including one resulting from pharmacology, or chemistry of these vio- ingestion of ‘‘certain tropical clupeid fishes,’’ lently poisonous fishes. Apparently clupei- later called clupeoid poisoning or clupeo- form fishes, which are generally valuable food toxism. fishes, become toxic only at sporadic intervals. Halstead (1967) reviewed cases of clupeo- The biotoxicity of these fishes appears to be a toxism, the first by Desportes (1770) from completely unpredictable phenomenon.’’ Halstead (1978:492) gave the symptoms of clupeoid poisoning as follows: ‘‘The first in- dication of a biotoxication is the sharp metal- 1 Manuscript accepted 7 April 2004. lic taste which may be present immediately 2 Bishop Museum, 1525 Bernice Street, Honolulu, upon ingestion of the fish. This is soon fol- Hawai‘i 96817-2704 (e-mail: [email protected]). lowed by nausea, dryness of the mouth, vom- iting, malaise, abdominal pain, and diarrhea. Pacific Science (2005), vol. 59, no. 1:73–77 The gastro-intestinal upset may be accom- : 2005 by University of Hawai‘i Press panied by a feeble pulse, tachycardia, chills, All rights reserved cold clammy skin, vertigo, a drop in blood

73 74 PACIFIC SCIENCE . January 2005 pressure, cyanosis and other evidences of and 1959 during seven cruises of the research vascular collapse. Within a very short period vessel Hugh M. Smith. In late 1958 small of time, or concurrently, a variety of neu- individuals were found, indicating that the rological disturbances rapidly ensue such as species had become established. The sardine nervousness, dilated pupils, violent head- never became abundant enough to be of value aches, numbness, tingling, hypersalivation, as a baitfish, and it has not been reported in muscular cramps, respiratory distress, pro- recent years. gressive muscular paralysis, convulsions, A second sardine, the Goldspot Sardine coma and death. Death may occur in less than (Herklotsichthys quadrimaculatus), was first no- 15 minutes. Fergusen (1823) claimed that the ticed in Ka¯ne‘ohe Bay, O‘ahu, in 1975, be- poison was so rapid in its action that natives coming abundant in the bay the following have died while in the very act of eating the year. Figure 1 illustrates a specimen from yellow-billed sprat. . . .’’ American Samoa. Williams and Clarke (1983) Melton et al. (1984) reported on the death reported on its biology in O‘ahu, admitting of a fisherman on Kaua‘i from eating three that they first thought it was the Marquesan Marquesan Sardines (Sardinella marquesensis) Sardine. The Goldspot Sardine may have in 1978. I (second author of the paper) iden- entered Hawaiian waters in 1972 when the tified the fish from skeletal remains obtained tuna-fishing vessel Anela dumped baitfish from the family of the deceased. This species identified only as sardines on the return was introduced to O‘ahu from Nuku Hiva, cruise from Jaluit, Marshall Islands, where the Marquesas, as a tuna baitfish between 1955 bait had been seined (Randall 1987). The

Figure 1. Goldspot Sardine (Herklotsichthys quadrimaculatus), Tutuila, American Samoa. Review of Clupeotoxism . Randall 75 poisonous sardine in Kaua‘i was probably not previous incident of sprat poisoning in May H. quadrimaculatus, because a sample of eight was also preceded by flood weather in fish caught soon after the poisoning was March.’’ The Kaua‘i poisoning by the Mar- identified as Sardinella marquesensis,andH. quesan Sardine on 3 October followed a quadrimaculatus was not yet known from the period of heavy rain. island. Paralytic shellfish poisoning (PSP), which Melton et al. (1984:119) wrote, ‘‘Sardines, has resulted from eating mussels and other herrings, and anchovies feed on planktonic bivalve mollusks during the warm months in organisms, thus the source of the toxin is not temperate waters, has been attributed to benthic as it is in ciguatera. The toxic fishes planktonic dinoflagellates of the genus Go- are often caught in turbid, brackish areas in nyaulax. The symptomatology of PSP, how- contrast to the clear-water, coral-reef habitat ever, is clearly different from that of clupeoid typical of ciguatera. These clupeoid fishes are poisoning. Nevertheless, another pelagic di- the first to acquire the toxin, whereas those noflagellate of tropical and subtropical seas producing ciguatera in its most severe form should be expected as the responsible organ- are the large carnivorous species at the upper ism for the poisoning from eating clupeoid end of the food chain. The size of the fish fishes. causing clupeoid poisoning does not seem to It is surprising that Onuma et al. (1999) be related to the intensity of the toxemia, in concluded that the benthic dinoflagellate Os- sharp contract to the fishes that cause cigua- treopsis siamensis is the cause of clupeotoxism. tera. Also significant in this fatality is the fre- On 9 January 1994 in the Antalaha District of quent observation that victims who were the Madagascar, a woman of age 49 obtained four most ill or died were ones who had eaten sardines that were identified as Herklotsichthys viscera as well as the flesh of the fish.’’ They quadrimaculatus. She cut off the head of two then gave evidence of this from outbreaks in of the fish, cooked the body parts, and fed Fiji, Tonga (Banner and Helfrich 1964) and one to herself and one to her child. She no- Jamaica. They also noted that clupeoid poi- ticed an unusual bitter taste of the fish. A cat soning takes place during the warm months that had eaten the remains of the fish died in of the year, in contrast to ciguatera, which 15 min. The woman became seriously ill and may occur at any season. was taken to the hospital where she died 15 hr Melton et al. (1984:120) stated, ‘‘The after eating the sardine. The child did not probable source of clupeotoxin is a planktonic show any sign of poisoning. The head of the dinoflagellate. The first suggestion of this was toxic fish and that of the nontoxic one were made by D’Arras in 1877, according to Hal- retrieved from the kitchen, frozen, and later stead (1967), who reported the observation of placed in ethanol before being sent to the Father Montrouzier in New Caledonia that authors. Their analysis resulted in the identi- local sardines become poisonous because they fication of the causative agent as palytoxin, feed on a ‘green monad’ which discolored first isolated from the benthic colonial coe- areas of the sea seasonally at the region of lenterate Palythoa in Maui by Moore and Balade. These monads cause conjunctivitis, Scheuer (1971). Its structure was defined coryza and erythema in persons coming in by Moore and Bartolini (1981) and Uemura contact with them.’’ et al. (1981), and its pharmacology by Hirata The seasonality of clupeoid poisoning may et al. (1988). be related to the generally heavier rainfall Part of the esophagus of one of the sar- during summer months in tropical and sub- dines was attached to the head. Microscopic tropical localities. The World Health Orga- examination of the gills and esophagus re- nization report (1980:259) of the outbreak in vealed a large amount of bottom sediment. Jamaica on 4–5 July 1979 stated, ‘‘It is inter- Because of this, Onuma et al. (1999:64) esting to note that one of the most severe wrote, ‘‘It is highly likely therefore that the floodings in over a century occurred on 12 fish obtain the toxin from a benthic organism. June in the involved part of the island. The In view of the fact that Ostreopsis spp. densely 76 PACIFIC SCIENCE . January 2005 grow on the bottom in tropical seas (Faust, Literature Cited 1995) and produce palytoxin and its analogs, most probably the dinoflagellate is the source Alcala, A. C., L. C. Alcala, J. S. Yasumura, of the toxin in the sardine.’’ They added that and D. Yasumoto. 1988. Human fatality Alcala et al. (1988) reported a human fatality due to ingestion of the crab Demania rey- from ingesting the crab Demania reynaudii in naudii that contained a palytoxin-like tox- the Philippines that was shown to contain a in. Toxicon 26:105–107. palytoxin-like toxin. Banner, A. H., and P. Helfrich. 1964. The There is another explanation for the pres- distribution of ciguatera in the tropical ence of bottom sediment in the gills and the Pacific. Tech. Rep. Hawaii Mar. Lab. 3:1– esophagus of the two sardines from Madagas- 48. car. This schooling species occurs inshore Desportes, J. B. 1770. Histoire des maladies in protected waters and is generally taken de S. Domingue. Vol. 1: 108–110. by seines or throw nets. When the fish are Halstead, B. W. 1967. Poisonous and ven- dragged over the bottom, the gills and omous marine animals of the world. Vol. esophagus may become heavily laden with 2. Vertebrates. U.S. Government Printing sediment. Williams and Clarke (1983) re- Office, Washington, D.C. ported the usual depth of Herklotsichthys ———. 1978. Poisonous and venomous ma- quadrimaculatus during the day over reef flats rine animals of the world. Darwin Press, of Ka¯ne‘ohe Bay, O‘ahu, as 1–2 m. At night Princeton, New Jersey. the fish disperse from the shallows for their Halstead, B. W., and W. M. Lively. 1954. primary feeding. Their food consists of a Poisonous fishes and ichthyoscarcotoxism. wide variety of zooplankton, the smaller fish U.S. Armed Forces Med. J. 5 (2): 157–175. eating mainly small copepods. There was no Hirata, Y., D. Uemura, and Y. Ohizumi. evidence of feeding on phytoplankton, but of 1988. Chemistry and pharmacology of course copepods and other small zooplank- palytoxin. Pages 241–258 in Handbook of tonic animals are responsible for that major natural toxins. Vol. 3. Marine toxins and source of nutrition entering the food chain in venoms. Marcel Dekker, New York. the sea. A bottom-dwelling crab could obtain Melton, R. J., J. E. Randall, N. Fusetani, R. S. palytoxin by eating filter-feeding benthic in- Weiner, R. D. Couch, and J. K. Sims. vertebrates in its community. 1984. Fatal sardine poisoning. Hawaii Melton et al. (1984:120) wrote, ‘‘The peo- Med. J. 43 (4): 114–120, 124. ple of Hawaii do not often eat local clupeoid Moore, R. E., and G. J. Bartolini. 1981. fishes, and when they do, they are apt to Structure of palytoxin. J. Am. Chem. Soc. eviscerate them. Thus, the potential for out- 103:2491–2494. breaks in Hawaii may be lower than in less Moore, R. E., and P. J. Scheuer. 1971. developed insular regions of the world.’’ Palytoxin: A new marine toxin from a They concluded, ‘‘We also believe there is coelenterate. Science (Washington, D.C.) ample reason to inform the public of the 172:495–498. hazards associated with the consumption of Oldendorp, C. G. 1777. Geschichte der mis- clupeoid fishes in Hawaii.’’ sion der evangelischen bru¨der auf den car- Palytoxin used as a reference in the study aibischen inseln St. Thomas, St. Croix, by Onuma et al. (1999) was obtained from und St. Jean. Vol. 2, no. 6: 110–111. Wako Pure Chemical in Osaka, and anti- Onuma, Y., M. Stake, T. Ukena, J. Roux, S. palytoxin monoclonal antibody 73D3 (mAb- Chanteau, N. Rasolofornirina, M. Ratsi- 73D3) from Hawaii Biotech, Honolulu. maloto, H. Naoki, and T. Yasumoto. 1999. Onuma et al. (1999) recommended treating Identification of putative palytoxin as patients poisoned by palytoxin with vaso- the cause of clupeotoxism. Toxicon 37:55– dilatory agents such as papaverine or iso- 65. sorbide dinitrate, as suggested by Vick and Randall, J. E. 1958. A review of ciguatera, Wils (1990). tropical fish poisoning, with a tentative Review of Clupeotoxism . Randall 77

explanation of its cause. Bull. Mar. Sci. toxin. ACS (Am. Chem. Soc.) Symp. Ser. Gulf Caribb. 8 (3): 236–267. 418:241–254. ———. 1987. Introductions of marine fishes Williams, V. R., and T. A. Clarke. 1983. Re- to the Hawaiian Islands. Bull. Mar. Sci. 41 production, growth, and other aspects of (2): 490–502. the biology of the gold spot herring, Her- Uemura, D., K. Ueda, and Y. Hirata. 1981. klotsichthys quadrimaculatus (Clupeidae), a Further studies on palytoxin II structure recent introduction to Hawaii. Fish. Bull. of palytoxin. Tetrahedron Lett. 22:2781– 81 (3): 587–597. 2784. World Health Organization. 1980. Fish poi- Vick, J. A., and J. Wils. 1990. Pharmaco- soning. WHO Weekly Epidemiol. Rec. logical and toxicological studies of paly- 55:258–259.