SOME CHEMICAL AND PHYSICAL PROPERTIES OF TWO TOXINS 

 FROM THE RED-TIDE ORGANISM, GYMNODINIUM BREVE '^ 



BY DEAN F. MARTIN = AND ASHIM B. CHATTERJEE * 



ABSTRACT 



A procedure is given for isolating and purifying two toxins 

 (substance I and II) from cultures or blooms of Gymnodin- 

 iutn breve. In vitro studies of acetylcholinesterase activity 

 show that the major toxin is not a cholinesterase inhibitor. 

 On the basis of the infrared data the properties of toxin 

 samples isolated from blooms of G. breve appear to be 

 identical to those from axenic cultures. 



Substance II is a light yellow, low-melting solid. Carbon, 

 hydrogen, and phosphorus are present; sulfur, chlorine, 

 bromine, and nitrogen are absent. The percentage com- 

 position is given for carbon, hydrogen, oxygen, and phos- 

 phorus; an empirical formula of CHicjOuP is indicated by 



the analytical data. A molecular weight of 650 was obtained 

 from a sample of substance II. 



Substance II was characterized by the absorption spectra 

 (ultraviolet, infrared), by the nuclear magnetic resonance 

 spectrum, and by the specific optical activity. These obser- 

 vations and the mass spectrogram provide useful structural 

 information. 



Substance I was characterized by the infrared spectrum 

 which indicated the presence of a carbonyl group. Not 

 enough of substance I was isolated for an elemental anal- 

 ysis. 



Evidence for a causal relation between blooms 

 of the Florida red-tide organism, Gymnodinium 

 hreve, and animal intoxication once rested on 

 seasonal and geographic correlations of these 

 events. The value of the correlation was limited 

 because of the presence of accompanying bac- 

 terial species with ichthyotoxic properties 

 (Bein, 1954). The relation was established by 

 Ray and Wilson (1957), who showed that cul- 

 tures of G. breve killed fish, and by McFarren, 

 Tanabe, Silva, Wilson, Campbell, and Lewis 

 (1965), who isolated a ciguateralike poison 

 from oysters and clams (taken from the area in 

 which a bloom occurred), from a bloom of G. 

 breve, and from laboratory cultures of G. breve. 



Before the present study, comparatively little 

 was known about the chemical and biochemical 

 properties of the toxin of G. breve. Starr 

 (1958) described attempts to characterize the 

 toxin through bioassays in which mullet and 

 guppies were used to compare the potency of 

 different toxic preparations. 



' Contribution No. 19 from the Marine Science Institute, Univer- 

 sity of South Florida, Tampa, Fla. 33620. 



-' Cnntriliution No. 62 from the Bureau of Commercial Fisheries 

 BioloRical Laboratory, St. PetersburR Beach, Fla. 33706. 



' Professor. Department of Chemistry and Marine Science Insti- 

 tute, University of South Florida, Tampa, Fla. 33620. 



' VisitinK Research Assistant Professor, Department of Chemistry. 

 University of South Florida, Tampa, Fla. 33620. 



Sasner (1965) reported that the toxin was an 

 endotoxic substance which was soluble in water 

 and ethyl alcohol, insoluble in chloroform, heat 

 labile, acid stable, and slowly dialyzable. The 

 toxic materials in his extracts affected all excit- 

 able membranes studied: sciatic nerve, sartorius 

 nerve-muscle preparations, and skin from the 

 leopard frog, Rana pipiens; the anterior byssus 

 retractor muscle of Californian mussel, Mytilus 

 calif ornianus; and the leg nerves of the spider 

 crab, Loxyrhynchiis sp. In all cases, the mem- 

 branes had been rendered unexcitable to 

 electrical stimulation. The toxin caused de- 

 polarization and complete loss of the resting 

 potential. According to Sasner (1965), the 

 physiological effects of G. breve toxin closely 

 paralleled those of a British gymnodinoid, 

 Gymnodinium veneficum (Abbott and Ballan- 

 tine, 1957). The toxin produced paralysis but 

 was different from a curare-type (synaptic 

 blocking agent) toxin, which is characteristic of 

 Gonyaulax catenella (Schantz, Lynch, Vayvada, 

 Matsumoto, and Rapoport, 1966). 



A number of toxigenic algae are known ; Shilo 

 (1967) included species of dinoflagellates, cer- 

 tain blue-green algae (e.g., Microcystis aerugi- 

 nosa) and several species of Chrysophyta (e.g., 

 Prymnesium parvum) . Before the present 

 study, however, few of the algal toxins seem to 



Published November 1970, 



FISHERY BULLETIN: VOL. 68, NO. 3 



433 



