RELATING PHOTOSYNTHETIC PIGMENTS AND IN-VIVO OPTICAL-DENSITY SPECTRA TO IRRADIANCE FOR THE FLORIDA RED-TIDE DINOFLAGELLATE GYMNODINIUM BREVE

The lipophilic pigment content and composition and in vivo optical den sity (OD) spectra for batch cultures of the Florida (USA) red-tide din oflagellate Gymnodinium breve Davis were analyzed after exposure to ir radiance treatments representative of an irradiance gradient that cell s might experience in a coastal water column. Decreases in total chlor ophyll (chl) content (attributable to decreases in chls a, c(1)/c(2) a nd c(3)) and total carotenoid content (attributable to decreases in to tal fucoxanthin and 19'-acylofucoxanthins, and gyroxanthin-diester) co incided with exposure to increased irradiance. The relative abundances of chi pigments remained constant whereas decreases in the relative a bundances of total fucoxanthin and 19'-acylofucoxanthins, and beta-psi -carotene and an increase in the relative abundance of diadinoxanthin coincided with exposure to increased irradiance. Because gyroxanthin-d iester has been observed only in a limited number of toxic dinoflagell ates (of which G. breve is the only warm-water taxon), was consistentl y quantifiable throughout various irradiance-induced physiological sta tes, and had a distinctive elution position and absorption maxima in t he chromatographic eluent, this carotenoid may serve as a 'biomarker' for G. breve within Florida coastal waters. An analysis of variance, p erformed on fourth-derivative plots derived from normalized OD spectra , identified portions of the OD spectra which differed among irradianc e treatments. Designated wavelengths coincided with wave crests in the fourth-derivative plots representative of absorption maxima for total chi c (460 to 468, 585 to 590 nm) and total fucoxanthin and 19'-acylo fucoxanthins, and diadinoxanthin (490 to 496 nm). Stepwise discriminan t analysis identified a set of 6 wavelengths (403, 541, 546, 509, 673 and 663 nm) which optimally classified OD spectra for populations amon g irradiance treatments. This lack of correspondence between wavelengt hs identified by the 2 statistical techniques was not unexpected; four th-derivative analysis identified portions of the OD spectra where dif ferences in the sharpness of curvature (as produced by absorption maxi ma of component pigments) occurred among treatments whereas discrimina nt analysis identified portions of the spectra which could be used to classify populations among treatments and did not necessarily have to correspond to absorption maxima. Such detectable differences in the bi o-optical 'signatures' of G. breve indicate that previous light histor y, as manifested in the pigment component, could be incorporated into future pigment-based monitoring applications to allow for the detectio n and/or physiological characterization of problematic taxa prior to b loom status.