HPLC and GC/MS determination of deuterated vitamin K (phylloquinone) in human serum after ingestion of deuterium-labeled broccoli

J Nutr Biochem. 2002 Mar;13(3):168-174. doi: 10.1016/s0955-2863(01)00210-8.

Abstract

The ability to intrinsically label plant constituents with stable isotopes has the potential to advance the study of vitamin K-absorption and metabolism in humans. Broccoli, a primary food source of phylloquinone (VK-1), was grown hydroponically using 31 atom % deuterium oxide in order to label VK-1 within the food matrix. Deuterium-labeled broccoli (115 g; 168 &mgr;g VK-1) was fed to one male subject in a single serving. Multiple serum samples were drawn throughout the subsequent 24-hr period. Reversed-phase HPLC was used to extract and purify VK-1 in both broccoli and serum. Ion abundances of the deuterium-labeled and unlabeled (endogenous) VK-1 were determined using GC/MS in negative chemical ionization mode. No sample derivatization was required. Endogenous VK-1 produced isotopomers from m/z 450 to m/z 453. The labeled VK-1 isotopomers in the broccoli were from m/z 452 to m/z 467, with the most abundant isotopomer being m/z 458 (14.1% of total labeled VK-1). The GC/MS chromatograms from serum revealed both endogenous VK-1 and VK-1 derived from the deuterium-labeled broccoli. The profile of labeled VK-1 isotopomers in serum was identical to the VK-1 isotopomer profile in labeled broccoli, indicating that no deuterium was lost due to exchange either in the body or in sample preparation. At 4 hr following broccoli intake, there was an 81.1% enrichment of phylloquinone in serum; labeled VK-1 was no longer detectable in serum at 24 hr. Use of isotope labeled vegetables enables one to discriminate exogenous intake of VK-1 from endogenous pools and ultimately to determine bioavailability of VK-1 from foods.