Vitamin K: lessons from the past

Vitamin K was discovered in the early 1930s by the Danish scientist Henrik Dam as a fat-soluble factor required for normal hemostasis [1]. In the next 50 years, clinicians, scientists and others believed that blood coagulation was the only physiological process regulated by vitamin K. During this period, it was discovered that vitamin K functions as a coenzyme during the post-translational carboxylation of glutamate residues into gammacarboxy glutamate (Gla) [2]. Two enzymatic activities were found to be associated with Gla formation: gammaglutamyl carboxylase, which is the enzyme responsible for Gla formation and the concomitant oxidation of vitamin K hydroquinone into vitamin K-epoxide [3], as well as the vitamin K-epoxide reductase enzyme complex by which the vitamin K-epoxide produced in the carboxylation reaction is re-converted into vitamin K hydroquinone [4]. It is amazing that it has taken more than half a century to realize and establish that this complex enzyme system is also necessary for the functioning of many other proteins not involved in hemostasis. The discovery of the first Gla-containing protein not involved in blood coagulation (bone Gla-protein, also known as osteocalcin) was a milestone in the changing of our concepts [5], but initially the importance of this discovery was recognized only by those involved in fundamental vitamin K research. It took more than two decades before it was broadly realized that vitamin K is not just for clotting anymore [6]. Following the initial discovery of osteocalcin, other Gla-proteins were discovered, including matrix Gla-protein (MGP, the most potent inhibitor of soft tissue calcification presently known [7]) and growth-arrest-sequence protein 6 (Gas6, a regulator of cell proliferation and apoptosis [8]). Surprisingly, the occurrence of Gla-proteins was not confined to mammals or vertebrates, but highly specialized Gla-proteins were also identified as components of cone snail neurotoxins [9]. This suggests that the vitamin K-dependent gammaglutamyl carboxylase developed early in evolution, and may be widely spread in nature. In humans, Gla-proteins have turned out to function as key regulators of important physiological processes including blood coagulation, soft tissue calcification (MGP), bone formation (osteocalcin), cell growth and apoptosis (Gas6). Detailed knowledge of the human genome has led to the identification of two new families of Gla-containing proteins of unknown function, the proline-rich Gla-proteins (PRGPs) and the trans-membrane Gla-proteins (TMGPs). In those cases in which we do know their function, the presence of Gla-residues in the respective Gla-proteins is essential for their activity [10]. In this issue Berkner and Runge describe the discovery and potential impact of the extrahepatic Gla-proteins in an excellent review [11]. After the discovery of so many different tissues requiring vitamin K, and so many exciting new Gla-proteins, one might expect that we have learned our lesson from the past. Unfortunately, this is not true. First, regulatory authorities produce guidelines for recommended daily vitaminK intake; in countries where such recommendations do exist, they are based on the human requirement for maintaining normal hemostasis. However, the liver (i.e. the place where the Gla-containing coagulation factors are synthesized) is extremely efficient in extracting vitamin K from the circulation. It has been demonstrated that in the healthy population, at vitamin K intakes which are adequate for full carboxylation of all blood coagulation factors, 20–30% of the circulating osteocalcin remains under-carboxylated [12]. In our experience based on thousands of subjects, complete carboxylation of circulating osteocalcin is extremely rare in non-supplemented (with vitaminK) subjects [12,13]. Under-carboxylation of osteocalcin was also demonstrated to be common in bone tissue [14]. New evidence based on conformation-specific antibodies has revealed that the same is true for MGP (Schurgers et al., unpublished data). The implication of these findings is that protection against cardiovascular calcification is probably suboptimal in most of the population, and that such protection can be improved by simple dietary measures, i.e. increasing our vitamin K intake. Consistent with these findings, several population-based studies suggest that low vitamin K intake is Correspondence: Cees Vermeer, Department of Biochemistry, University of Maastricht, PO Box 616, 6200 MD Maastricht, the Netherlands. Tel.: +31 43 388 1682; fax: +31 43 388 4160; e-mail: c.vermeer@ bioch.unimaas.nl Journal of Thrombosis and Haemostasis, 2: 2115–2117