OPTIMIZATION OF CHROMATOGRAPHIC CONDITIONS FOR THE DETERMINATION OF FOLATES IN FOODS AND BIOLOGICAL TISSUES FOR NUTRITIONAL AND CLINICAL-WORK

Recent studies implicate folate metabolism in the aetiology of heart disease, neural tube defects, malignant transformation and affective disorders. The paper reports a rapid, isocratic HPLC separation of 11 folylmonoglutamate compounds which should prove useful when adapted to the varied needs of analysts researching these and other specific areas. Also reported are the separation, UV spectra including lambda(max) values, fluorescence emission scans at an excitation wave-length of 295 nm and electrochemically derived hydrodynamic voltammograms with optimum oxidation voltages for p-aminobenzoylglutamate, tetrahydrofolate, 5-methyldihydrofolate, 5-methyltetrahydrofolate, 5-formyltetrahydrofolate, 5,10-methenyltetrahydrofolate, dihydrofolate, pteroylmonoglutamate and 5,10-methylenetetrahydrofolate. In particular, 5-methyltetrahydrofolate, the main food folate and form of the vitamin found in plasma, can be measured easily by eiectrochemical detection using a low and highly selective voltage of 450 mV. This reduced folate is also readily detected fluorimetrically using an excitation wavelength of 295 nm and measuring emission at 365 nm. Electrochemical and fluorimetric detection offer equal sensitivity for 5-methyltetrahydrofolate measurement (300 pg on column). No other folate studied could be measured down to this level using fluorimetric detection under the described conditions. At pH 3.5, folate coenzyme lambda(max) for UV detection varies between 267 and 300 nm with 5,10-methenyltetrahydrofolate giving maximum absorption at 355 nm. UV measurement of 5-methyltetrahydrofolate is approaching an order of magnitude less sensitive than the former methods of detection. However, for in vitro studies, particularly in the form of a photodiode array, UV detection is a particularly useful tool. For cerebrospinal fluid, plasma, erythrocyte or food measurement of 5CH(3)-H(4)PteGlu, electrochemical or fluorimetric detection is recommended; whilst for pharmacokinetic studies of plasma 5CHO-H(4)PteGlu during methotrexate rescue therapy, electrochemical or UV detection is most appropriate. For analysis of plasma PteGlu following supplementation, or in food stuffs, UV detection offers the best measurement technique. The information presented should help address the major problem of trace folate analysis by HPLC, that is the need to combine high sensitivity with optimum selectivity in studying complex matrices such as physiological fluids, tissue preparations and food samples.