600 J. Phycol. 39, 600–609 (2003) A NEW LARVAL FISH BIOASSAY FOR TESTING THE PATHOGENICITY OF PFIESTERIA SPP. (DINOPHYCEAE) 1 Vincent J. Lovko, Wolfgang K. Vogelbein, 2 Jeffrey D. Shields, Leonard W. Haas, and Kimberly S. Reece Virginia Institute of Marine Science, The College of William and Mary, Rt. 1208, Gloucester Point, Virginia 23062, USA Water quality, microbial contamination, prior fish health, and variable results have been major impedi- ments to identifying the cause and mechanism of fish mortality in standard aquarium-format Pfiesteria bioassays. Therefore, we developed a sensitive 96-h larval fish bioassay for assessing Pfiesteria spp. patho- genicity using six-well tissue culture plates and 7-day- old larval cyprinodontid fish. We used the assay to test pathogenicity of several clonal lines of Pfiesteria piscicida Steidinger and Burkholder and P. shumwayae Glasgow and Burkholder that had been cultured with algal prey for 2 to 36 months. The P. shumwayae cul- tures exhibited 80%–100% cumulative mortality in less than 96 h at initial zoospore densities of approx- imately 1000 cells·mL -1 . No fish mortalities occurred with P. piscicida at identical densities or in controls. In a dose-response assay, we demonstrated a strong positive correlation between dinospore density and fish mortality in a highly pathogenic culture of P. shum- wayae, generating a 96-h LD 50 of 108 zoospores·mL -1 . Additionally, we applied the assay to evaluate a 38-L P. shumwayae bioassay that was actively killing fish and compared results with those from exposures of juve- nile tilapia (Oreochromis niloticus) in a 500-mL assay system. Water from the fish-killing 38-L assay was fil- tered and centrifuged to produce fractions dominated by dinoflagellates, bacteria, or presumed ichthyotoxin (cell-free fraction). After 96 h, the larval fish assay ex- hibited 50%–100% cumulative mortality only in frac- tions containing dinoflagellates, with no mortalities occurring in the other fractions. The 500-mL bioassay with tilapia produced inconsistent results and demon- strated no clear correlation between mortality and treat- ment. The new larval fish bioassay was demonstrated as a highly effective method to verify and evaluate di- noflagellate pathogenicity. Key index words: bioassay; dinoflagellate; harmful al- gae; larval fish; Pfiesteria shumwayae Abbreviations: ASW, artificial seawater; BSL2, biohaz- ard safety lab level 2; BSL3, biohazard safety lab level 3; CCMP, Center for Culture of Marine Phytoplank- ton; DO, dissolved oxygen; VIMS, Virginia Institute of Marine Science Pfiesteria piscicida and Pfiesteria shumwayae, members of the “toxic Pfiesteria complex,” are considered respon- sible for acute fish kills, fish lesion events, and human disease in estuaries of the mid-Atlantic United States (Burkholder et al. 1992, 2001a,b, Burkholder and Glas- gow 1995, 1997b, Glasgow et al. 1995, 2001b, Noga et al. 1996). These heterotrophic dinoflagellates are re- ported to have a complex life history with multiple cyst, zoospore, and amoeboid stages and presumably have the ability to produce potent ichthyotoxins in re- sponse to fish or fresh fish excreta (Burkholder and Glasgow 1995, Burkholder et al. 1995, Steidinger et al. 1996, Marshall et al. 2000). The toxigenicity of Pfiesteria spp. is reported to de- pend on life history stage, functional type (Tox A, Tox B, Non-Inducible as described by Burkholder et al. 2001b), prey availability, and abiotic factors such as salinity, light, temperature, and nutrients (Burkholder et al. 1992, 1995, Burkholder and Glasgow 1995, 1997a,b, Marshall et al. 2000). Currently, no cell-specific diagnos- tic tests exist that can definitively determine the pres- ence of toxin-producing Pfiesteria. Fish bioassays, there- fore, are the only method currently available to detect toxic Pfiesteria strains (Burkholder et al. 2001a,c). Bio- assays used to culture and test potentially toxic strains of Pfiesteria spp. are typically conducted in aquaria or other vessels with volumes ranging from 2 to 40 L, although volumes as low as 300 mL have been used (Burkholder et al. 1995, 2001c, Lewitus et al. 1995). The vessels are inoculated with either environmental samples (water or sediment) or clonal dinoflagellate cultures, to- gether with fish, most commonly juvenile tilapia, Oreo- chromis spp. (Burkholder et al. 1995, Burkholder and Glasgow 1997a,b, Marshall et al. 2000, Glasgow et al. 2001b, Vogelbein et al. 2001). In bioassays inoculated with environmental sam- ples, mortalities typically begin in 4–12 weeks for “non- active” cells and 4–9 days for “active” cells collected at in-progress fish kills (Burkholder et al. 2001a,c, Glas- gow et al. 2001a). Fish death, along with confirmed presence of Pfiesteria zoospores at lethal densities (more than 300 cells·mL -1 ) constitutes a positive bioassay (Burkholder et al. 2001a,c, Glasgow et al. 2001a, Pfies- teria Interagency Coordinative Working Group). Sub- sequent bioassays, inoculated with cells isolated from the first set of positive bioassays, are conducted to ver- ify toxicity (Burkholder et al. 1995, 2001a,c, Burkholder and Glasgow 1997b, Marshall et al. 2000). A positive re- sult in the second set of bioassays confirms the culture as toxic (Burkholder et al. 2001a,c, Glasgow et al. 2001a, Marshall et al. 2000, Pfiesteria Interagency Coordinative Working Group). These bioassays are indirect and only 1 Received 19 July 2002. Accepted 13 February 2003. 2 Author for correspondence: e-mail wolf@vims.edu.