Repeatability and validity of a fluorimetric HPLC method in the quantification of yessotoxin in blue mussels (Mytilus edulis) related to the mouse bioassay.

Repeatability and validity of a fluorimetric HPLC method in quantification of yessotoxin (YTX) in mussels related to the mouse bioassay was studied. Blue mussels (Mytilus edulis) from the Sognefjord, Norway were sampled from March to November, 1997, and October to December, 1998. A total of 75 samples were analysed for YTX by HPLC using 4-[2-(6,7-dimethoxy-4-methyl-3-oxo-3, 4-dihydroquinoxalinyl) ethyl]-1,2,4-triazoline-3, 5-dione (DMEQ-TAD) as a fluorimetric derivatization agent. Among these, 28 of the samples were analysed by HPLC in duplicate. All samples were analysed by the mouse bioassay using both chloroform and ether in the final step of extraction. The duplicate measurements using HPLC was found equal and the method repeatable (p<0.05). The absolute difference between the two measurements was found to increase with increasing level of measurements. This significant positive correlation (p<0.05) was mainly due to concentrations of YTX higher than 200 microg/100g mussel meat. However the precision of the results obtained was not found to be less in the upper level than in the lower level. Based on the internal correlation analysis including the mouse bioassay and the HPLC method a cut-off value of < or =10 microg YTX/5 g digestive gland was found preferable. The mouse bioassay of ether extracts often failed to detect high levels of YTX, and as demonstrated by the low kappa-values, the agreement between the mouse bioassay of ether extracts and the HPLC method was very weak. The HPLC method was found to give repeatable results and thereby found to be reliable. Consequently, the HPLC method seems to the method of choice for detection and quantification of YTX in mussels when compared with the mouse bioassay.