• From Sigma Chemical Company (Poole, UK).. From Boehringer Mannheim UK (Lewes, UK) All other chemicals from BDH Limited (Poole, UK) based on colour formation by the Trinder reaction. P This assay lacks the sensitivity necessary for application to dilute lipoprotein fractions. We have therefore developed a micromethod, capable of measuring sub-nanomolar quantities of TAG, based on the formation of a coloured formazan by reduction of 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (INT). 4 Initially we tried to develop a two-stage assay in which free glycerol was allowed to react, then lipase added and the increment in glycerol measured. However, it was not possible to obtain complete reaction under these conditions. The method therefore involves separate estimations, one of free glycerol (without lipase) and one of total glycerol (with lipase present).

The final concentrations in the reagent mixture are as shown in Table 1. The INT is made up as a 12mmol/L solution in distilled water and is added just before use, as the mixture turns pink on standing. We run the method onthe Instrumentation Laboratories Multistat III Microcentrifugal Analyzer (IL (UK) Ltd, Warrington, UK). The loader settings are the same for both free and total glycerol on plasma: sample volume 4JLL with a 36JLL wash; reagent volume 100JLL with a 10JLL wash. The reference cuvette contains water instead of sample. The assays are run with glycerol standards of 1mmol/Land 2mmol/L. The wavelength is 500nm and the run time is 10min. The temperature is 30 C for free glycerol, but 37 C for total glycerol in order to avoid longer run times.(Free glycerol can also be measured at 37 C, but in practice the time taken for the Multistat to reach that temperature outweighs any shortening of run-time.) The TAG concentration is calculated from the difference between total and free glycerol. For lipoprotein fractions the sample volume can be increased to 30JLL, with standards of 250 and 500JLmol/L.