Vitamin K occurs in the natural world in several forms, including a plant form, phylloquinone (PK), and a bacterial form, menaquinones (MKs). In many species, including humans, PK is a minor constituent of hepatic vitamin K content, with most hepatic vitamin K content comprising long-chain MKs. Menaquinone-4 (MK-4) is ubiquitously present in extrahepatic tissues, with particularly high concentrations in the brain, kidney and pancreas of humans and rats. It has consistently been shown that PK is endogenously converted to MK-4 (refs 4-8). This occurs either directly within certain tissues or by interconversion to menadione (K(3)), followed by prenylation to MK-4 (refs 9-12). No previous study has sought to identify the human enzyme responsible for MK-4 biosynthesis. Previously we provided evidence for the conversion of PK and K(3) into MK-4 in mouse cerebra. However, the molecular mechanisms for these conversion reactions are unclear. Here we identify a human MK-4 biosynthetic enzyme. We screened the human genome database for prenylation enzymes and found UbiA prenyltransferase containing 1 (UBIAD1), a human homologue of Escherichia coli prenyltransferase menA. We found that short interfering RNA against the UBIAD1 gene inhibited the conversion of deuterium-labelled vitamin K derivatives into deuterium-labelled-MK-4 (MK-4-d(7)) in human cells. We confirmed that the UBIAD1 gene encodes an MK-4 biosynthetic enzyme through its expression and conversion of deuterium-labelled vitamin K derivatives into MK-4-d(7) in insect cells infected with UBIAD1 baculovirus. Converted MK-4-d(7) was chemically identified by (2)H-NMR analysis. MK-4 biosynthesis by UBIAD1 was not affected by the vitamin K antagonist warfarin. UBIAD1 was localized in endoplasmic reticulum and ubiquitously expressed in several tissues of mice. Our results show that UBIAD1 is a human MK-4 biosynthetic enzyme; this identification will permit more effective decisions to be made about vitamin K intake and bone health.

There are two forms of naturally occurring vitamin K, phylloquinone and the menaquinones. Phylloquinone (vitamin K 1 ) is a major type (>90%) of dietary vitamin K, but its concentrations in animal tissues are remarkably low compared with those of the menaquinones, especially menaquinone-4 (vitamin K 2 ), the major form (>90%) of vitamin K in tissues. Despite this great difference, the origin of tissue menaquinone-4 has yet to be exclusively defined. It is postulated that phylloquinone is converted into menaquinone-4 and accumulates in extrahepatic tissues. To clarify this, phylloquinone with a deuterium-labeled 2-methyl-1,4-naphthoquinone ring was given orally to mice, and cerebra were collected for D NMR and liquid chromatography-tandem mass spectrometry analyses. We identified the labeled menaquinone-4 that was converted from the given phylloquinone, and this conversion occurred following an oral or enteral administration, but not parenteral or intracerebroventricular administration. By the oral route, the phylloquinone with the deuterium-labeled side chain in addition to the labeled 2-methyl-1,4-naphthoquinone was clearly converted into a labeled menaquinone-4 with a non-deuterium-labeled side chain, implying that phylloquinone was converted into menaquinone-4 via integral side-chain removal. The conversion also occurred in cerebral slice cultures and primary cultures. Deuterium-labeled menadione was consistently converted into the labeled menaquinone-4 with all of the administration routes and the culture conditions tested. Our results suggest that cerebral menaquinone-4 originates from phylloquinone intake and that there are two routes of accumulation, one is the release of menadione from phylloquinone in the intestine followed by the prenylation of menadione into menaquinone-4 in tissues, and another is cleavage and prenylation within the cerebrum.

No abstract

We retrospectively analyzed the bone mineral density (BMD) of postmenopausal Japanese women taking an aromatase inhibitor (AI), exemestane, anastrozole or letrozole, and calculated the decrease rate constant of BMD in each individual to compare the influence of the three AIs on BMD. We also aimed to evaluate the preventive effect of bisphosphonates (BPs) on the AI-induced decrease in BMD. The decrease rate constant of BMD (k(e)) in each individual was determined as a slope of linear regression of the relationship between time and logarithm of BMD value in each patient during the AI therapy. To compensate for the age-related change in BMD level, we estimated the age-related decrease rate constant of BMD (ke,0) in healthy Japanese postmenopausal women from the literature. AIs decreased BMD with a ke value of -0.0329 yr(-1), which was 4.7-fold larger than the k(e,₀) value of -0.00699 yr(-1). No significant difference was detected in the influence on BMD among AIs. Co-administration of BP ameliorated the ke value to -0.0117 yr(-1), a value similar to k(e,₀). The influence of AIs on BMD was quantitatively evaluated by using the decrease rate constant of BMD (k(e)). The present study also suggests that BPs may be useful to prevent the decrease in BMD induced by AIs.

No abstract