tiometrically at 38° C. with an automatic 

 titrator operated as a pH-stat unit. Enzyme 

 activity was expressed as micromoles of sodium 

 hydroxide added per minute (fig. 3) to a pH 7.5 

 end-point. 



The activity of the toxin was tested as fol- 

 lows. In a succeeding trial, the enzyme activity 

 was measured again, and after 2 minutes an 

 aqueous solution of the toxin (substance II 

 dispersed in water by means of the ultrasonic 

 macerator) was added. The determination 

 curves (fig. 3) indicate the addition of 0.2 ml. 

 or 0.4 ml. of toxin solution has no effect on 

 homogenate activity. The activity of the toxin 

 solution was verified: 0.2 ml. of solution {ca. 

 0.15 mg. substance II) added to 30 ml. of sea 

 water caused death of sheepshead minnows (5.5 

 cm., 4.1 g.) within 3 minutes. 



MASS SPECTRAL ANALYSIS 



The mass spectrum of substance II (bloom 

 sample) was obtained by using the following 

 conditions: ionizing voltage 70 e.v. pressure 1.5 

 XlO*^ mm. Hg., and temperature 50° C. (table 

 7). 



Table 7. — Mass-charge ratios (m/e) and relative 



abundances from mass spectral analysis of substance 

 II. 



Relative Relative 



m/e abundance m/e abundance 



27 11.0 151 S7.0 



31 30.9 153 13.5 



42 54.1 155 13.5 



45 80.0 157 13.5 



57 62.7 160 30.8 



59 77.5 240 14.9 



63 19.6 258 8.6 



67 22.1 266 6.3 



69 54.0 290 9.0 



72 60.6 301 6.3 



83 60.1 309 5.0 



86 56.5 310 5.0 



100 45.5 311 5.0 



102 24.7 319 5.0 



107 16.1 320 5.0 



121 17.2 321 5.0 



124 18.1 333 4.8 



128 24.8 334 4.8 



131 15.4 335 4.8 



138 15.1 348 4.8 



141 15.1 349 4.8 



143 15.1 1 350 4.8 



1 The ampliiied scale spectrum (100 times regular intensity) indi- 

 cates stable species ( with relative abundance of less than 1 ) at m/e 



of 375. 390, 450, 418. 440. 477, 508. Maximum relative abundance is 

 80. 



SOME CHEMICAL CHARACTERISTICS 

 AND STRUCTURAL FEATURES OF 

 THE TOXIN 



The purpose of the present study was to 

 isolate the toxins of Gymnodinium breve blooms 



and cultures and to determine the chemical and 

 physical properties of these toxins. During the 

 course of the study, sufficient information was 

 obtained from these properties to indicate some 

 structural features of the toxins. The topics 

 will be considered in order; a review of the 

 chemical characteristics of the toxin is neces- 

 sary to understand the structural characteris- 

 tics. Unfortunately, not enough substance I was 

 isolated to evaluate its chemical characteristics, 

 and most of the discussion must be confined to 

 substance II. 



CHEMICAL CHARACTERISTICS 



Several results of the present study are con- 

 sistent with those of previous investigators. 



First, several workers (McFarren et al., 

 1965; Sasner, 1965; Cummins, Stevens, Hunt- 

 ley, Hill, and Higgins, 1968) have also found 

 two (or more) toxins which differ in solubility 

 and other physical properties. In the present 

 study, two toxins — substances I and II — were 

 separated from Gymnodinium breve bloom wa- 

 ter. Both have been separated from cultures of 

 G. breve, although substance I was isolated only 

 in trace quantities and could easily have been 

 missed. 



Secondly, the major toxin (substance II) is 

 essentially an endotoxin. This determination 

 was made in the present study by dividing a 

 1-1. sample of G. breve culture into two por- 

 tions and extracting the medium before (por- 

 tion A) and after (portion B) the cells had 

 been removed by gravity filtration. Detectable 

 amounts of toxin could be isolated from portion 

 A but not from portion B. An alternative pos- 

 sibility, complete adsorption of the toxin in the 

 filter paper, has not been observed. 



Thirdly, the major toxin of G. breve may be 

 a neurotoxin, but it evidently does not inhibit 

 cholinesterase activity in vitro.  



Fourthly, the toxins (substance I or II) seem 

 to volatilize under reduced pressure, although 

 the toxin is not readily volatilized from aqueous 

 solution by a stream of nitrogen. 



The study provided the following results that 

 apparently have not been reported previously. 



1. The major toxin (substance 11) isolated 

 from a bloom of G. breve appears to be similar 

 to that isolated from cultures of G. breve. This 



440 



U.S. FISH AND WILDLIFE SERVICE 



