Rapid, high performance method for the determination of vitamin K(1), menaquinone-4 and vitamin K(1) 2,3-epoxide in human serum and plasma using liquid chromatography-hybrid quadrupole linear ion trap mass spectrometry.

ABSTRACT

Unlike the other fat-soluble vitamins, vitamin K circulates in the human bloodstream at very low levels because of a low intake in the diet. Mammals have developed an efficient recycling system, known as vitamin K-epoxide cycle, which involve quinone, hydroquinone and epoxide forms of the vitamin. Phylloquinone (K(1)) is the main homologue, while menaquinone-4 (MK-4) is both a member of the vitamin K(2) family and metabolite of K(1) in extra-hepatic tissues. Notwithstanding the recent advances, many aspects of the complex vitamin K physiology still remain to be investigated. Therefore, there is a critical need to develop more reliable analytical methods for determining the vitamin K and its metabolites in biological fluids and tissues. Nevertheless, relatively low concentrations, unavailability of some authentic standards and occurrence of interfering lipids make this a challenging task. The method proposed in the present paper can directly and accurately estimate K(1), K(1) 2,3-epoxide (K(1)O), and MK-4 in human serum and plasma at concentrations in the ng/L-µg/L range, using labelled internal standards and a quadrupole linear ion trap instrument operated in multiple reaction monitoring (MRM) mode. High sensitivity was achieved by removing signal "endogenous suppressors" and making the composition of the non-aqueous mobile phase suitable to support the positive atmospheric pressure chemical ionization of the analytes. An excellent selectivity resulted from the combination of some factors the MRM acquisition, the adoption of an identification point system, an extraction optimized to remove most of the lipids and a tandem-C18 column-system necessary to separate isobaric interferences from analytes. The method was validated according to the Food and Drug Administration (FDA) guidelines and its accuracy was assessed by analysing 9 samples from the Vitamin K External Quality Assessment Scheme (KEQAS). Its feasibility in evaluating vitamin K status in human serum was also tested by monitoring a group of six healthy subjects and a group of six patients under oral anticoagulant therapy (OAT). Warfarinised patients did not show deficiency of K1 but levels comparable with those of healthy people and an accumulation of K1O up to 3.760µg/L. MK-4 was not detected in either of the two groups.