In vivo assays of docosahexaenoic acid biosynthesis in fish

A method was developed for measuring the synthesis of docosahexaenoic acid from linolenic acid in vivo. Rainbow trout (Oncorhynchus mykiss) were fed a pulse of diet containing deuterated (D5) (17,17,18,18,18)-18:3n-3 ethyl ester and 21:4n-6 ethyl ester and sampled at intervals post-dose. The appearance of D-5-22:6n-3 in different tissues was determined by gas chromatography-negative chemical ionisation mass spectrometry of the pentafluorobenzyl esters. In trout of circa 5 g weight acclimated onto a diet containing 11% vegetable oil, sufficient to meet the essential fatty acid requirement for n-3 and n-6 fatty acids, and 5% fishmeal, the whole body accumulation of D-5-22:6n-3 was linear over 7 days corresponding to a rate of 0.54 +/- 0.12 ug g fish(-1) mg D-5-18:3n-3 eaten(-1) day(-1). The amount of D-5-22:6n-3 peaked in liver at day 7, in the whole carcase at day 14 and in brain and eyes at day 24. Trout reared on a diet lacking fish oil and fish meal, and therefore completely lacking 20:5n-3 and 22:6n-3, gave a similar rate of 22:6n-3 synthesis whereas fish reared on a diet containing 11% fish oil gave a rate of approximately one tenth those fish reared on vegetable oil. Smaller trout (0.8 g) fed a diet containing no fish meal or fish oil gave a rate of 22:6n-3 synthesis almost 10 fold higher than that of 6 g fish. A further series of experiments showed that caeca as well as liver were a site of 22:6n-3 synthesis. At early time points (2 days post-feeding) over 5 times more D-5-22:6n-3 was found in caeca than in liver on a per mg protein basis and this decreased towards unity by day 15. The method is sensitive (threshold circa 1 pg fatty acid) and permits identification and quantitative analysis of all the pathway intermediates (D-5-18:4n-3, D-5-20:4n-3, D-5-20:5n-3, D-5-22:5n-3, D-5-24:5n-3, D-5-24:6n-3) if required.