Affinity extraction combined with stable isotope dilution LC/MS for the determination of 5-methyltetrahydrofolate in human plasma.

The predominant circulating folate monoglutamate in human plasma (>90%), and thus the most significant folate for accurately diagnosing folate deficiency, is 5-methyltetrahydrofolic acid (5 MT). Folate deficiency is typically indicated when circulating folate levels are < or = 3 ng/mL. The quantitative determination of plasma folates in general, and of 5 MT in particular, is complicated by their naturally low levels (pg/mL to ng/mL), their instability, and their tendency to interconvert. Highly specific and sensitive analytical methods are needed to accurately quantify endogenous 5 MT in human plasma. A method that utilizes the specific high-affinity binding sites of bovine folate binding protein (FBP) and the selectivity and sensitivity of selected ion monitoring mode isotope-dilution liquid chromatography/mass spectrometry (LC/MS) to quantify plasma 5 MT has been developed. The method is based on the solid-phase affinity extraction (SPAE) of 5 MT and its stable isotopically labeled analogue ([13C(5)]5 MT) from plasma (1 mL) using FBP immobilized to polymeric beads. The excess high-affinity binding sites on the affinity columns enable quantitative extraction of 5 MT from plasma under optimized sample pH conditions. Additionally, it is demonstrated that plasma proteins do not hinder the determination of 5 MT; therefore, protein precipitation is not required before the affinity extraction step. Detection and quantification of the extracted 5 MT is provided by positive-ion mode LC/MS in which the protonated molecular ions [M+H](+) of the analyte and the internal standard are monitored. The method shows linearity over three orders of magnitude (0.04-40 ng/mL) and has limits of detection and quantification of 0.04 and 0.4 ng/mL, respectively. Calibration curves obtained by spiking 5 MT into plasma exhibited good linearity between 0 and 25 ng/mL and both the plasma calibration standards and the plasma samples were stable for at least 48 h at room temperature. The recovery (average +/- % RSD) of 5 MT spiked into plasma from 5 to 25 ng/mL was 98.0% +/- 1.6% (n = 15). 5 MT levels determined by SPAE-LC/MS compared to "total folate" levels determined by radioassay and microbiological assay were discordant. Reasons for the discordancy are theorized, but it is clear that there exists an urgent need for clinical reference materials containing certified folate levels. Copyright 2003 Elsevier Science (USA)