Monitoring of the toxic dinoflagellate Karenia brevis using gyroxanthin-based detection methods

The threat to human health and fisheries resources due to blooms of the toxic dinoflagellate Karenia brevis has lead to widespread public concern and calls for continuous monitoring of coastal waters for this organism. Here, a rapid and sensitive photopigment-based monitoring approach is described that incorporates refinements to standard filtration and analytical methods. This method uses the biomarker pigment gyroxanthin-diester contained in cells of some gymnodiniod species including K. brevis. Investigations of the retention efficiencies of five filter types for gyroxanthin from natural blooms of K. brevis showed no significant differences between GF/F, GF/C, 934-AH, GF/A or GF/D filters. Retention efficiencies were generally greater than 98% of cells added, indicating that the larger nominal pore size filters may be used safely for sample collection, reducing overall filtration times for large volumes of water. Simulated bloom experiments using cultures of K. brevis added to unfiltered water from Galveston Bay showed that retention of gyroxanthin on GF /D filters was significantly lower than on other filter types. There were significant interactions (p < 0.01) between filter type and cell density for the variables gyroxanthin, gyroxanthin chl a super(-1) and gyroxanthin cell super(-1), suggesting that the performance of the different filter types was dependent on cell density. Retention efficiencies for the simulated blooms ranged between > 99% of cells retained and < 30% of cells retained (greatest losses were for the GF/D filters). Combined results of natural and simulated blooms indicated that GF/C, 934-AH or GF/A filters gave the best retention efficiency with the fastest filtration times. Sample processing times were also improved by modifying the flow gradients in an existing HPLC protocol allowing the analysis of 106 samples in 24 h. The resulting protocol is suitable for incorporation into routine water quality monitoring programs, and would greatly facilitate the early detection and tracking of K. brevis blooms in coastal waters.