K is for koagulation

Comparative genetic linkage studies in rats, mice and humans have finally identified a key component of vitamin K metabolism that is targeted by the commonest anticoagulant drugs in use today.

Vitamins have enormous medical and economic importance. For example, the discovery of vitamin K, essential for blood clotting, led to the prevention of countless deaths from bleeding and blood clots, and facilitated the control of devastating agricultural pests. These benefits accrued despite a limited understanding of vitamin K metabolism. Vitamin K must be enzymatically activated before it can do its job, but the proteins involved in this process have resisted identification for more than 60 years. At last real progress has been made, as Rost et al.1 and Li et al.2 describe in this issue.

The 1943 Nobel Prize in Physiology or Medicine went to Denmark's Henrik Dam and Edward Doisy of the United States for their characterization of vitamin K — so named because a lack of this vitamin causes a defect in blood koagulation (the Scandinavian spelling). The discovery had immediate medical benefits. Before the 1940s, vitamin K deficiency often caused fatal bleeding in infants, and this has been largely eliminated by giving vitamin K to pregnant women and newborns. Meanwhile, cows in the northern United States had been bleeding to death through eating mouldy sweet clover hay. In 1940, Karl Link identified the fungal product responsible as a vitamin K antagonist. He named a potent derivative of this antagonist ‘warfarin’, after the Wisconsin Alumni Research Foundation, to which he assigned the patent rights.

Vitamin K deficiency impairs blood clotting by preventing the carboxylation of essential glutamic acid residues in several blood-clotting proteins. The enzyme that usually catalyses this carboxylation reaction is vitamin-K-dependent carboxylase, which uses oxygen and a reduced form of vitamin K to add a molecule of carbon dioxide to glutamic acid, producing γ-carboxyglutamic acid (Fig. 1). The modification enables clotting factors to bind calcium ions, associate with membrane surfaces and clot blood.

Figure 1: The vitamin K cycle.


Vitamin-K-dependent carboxylase uses reduced vitamin K and oxygen to add a carbon dioxide molecule to the side chain of glutamic acid in certain blood-clotting proteins, producing γ-carboxyglutamic acid and vitamin K 2,3-epoxide. Vitamin K epoxide reductase regenerates reduced vitamin K for another cycle of catalysis. Warfarin inhibits the reductase, impairing the synthesis of clotting factors and causing bleeding. A deficiency in multiple clotting factors can also be caused by mutations in either vitamin-K-dependent carboxylase or epoxide reductase. Rost et al.1 and Li et al.2 have identified the gene that encodes vitamin K epoxide reductase. Above, Close-up of a blood clot (roughly × 1,500).

The other product of catalysis is vitamin K 2,3-epoxide, which is recycled to reduced vitamin K by vitamin K epoxide reductase (VKOR). VKOR activity is inhibited by warfarin. In large doses, warfarin causes bleeding and is an excellent rat poison. At just the right dose, however, it prevents the lethal blood clots that can occur in the lungs, hearts and brains of susceptible people (Fig. 1). The medical use of warfarin received a boost when US President Dwight Eisenhower was treated with it after a heart attack in 1955, and warfarin has been the most widely prescribed anticoagulant drug ever since. Some rats and mice have developed resistance to warfarin-like poisons, and some people — though very few — require extraordinarily large doses of warfarin to ‘thin’ their blood.

The gene that codes for vitamin-K-dependent carboxylase was cloned and characterized some time ago3. But VKOR has been purified only partially, and its gene has remained elusive. As Rost et al.1 and Li et al.2 report, a genetic approach has finally succeeded — thanks to a few key insights. First, previous work4 on rats with an inherited resistance to poisons had mapped the VKOR gene to rat chromosome 1, which is similar to mouse chromosome 7. Comparisons of these rodent chromosomes with the less-similar human gene map allowed VKOR to be traced to human chromosome 10, 12 or 16. Second, some humans with a hereditary deficiency in several vitamin-K-dependent clotting factors have defective VKOR activity, and this disorder had been mapped by Fregin and colleagues5 to human chromosome 16. By postulating that different VKOR mutations could cause either warfarin resistance or defective clotting factors, these authors5 succeeded in assigning the gene to the segment of human chromosome 16 that resembles the warfarin-resistance segment on the rodent chromosomes.

Rost et al. and Li et al. therefore focused on human chromosome 16. Rost et al.1 found that there were more than 100 genes in the relevant DNA segment. To winnow the list, they turned to patients and rats with inherited defects in VKOR activity. They found that mutations in one particular gene occurred in two families with defective vitamin-K-dependent clotting factors, in four families with hereditary warfarin resistance and in numerous rat strains with resistance to warfarin-like poisons. They also identified similar genes in fish, frogs and even mosquitoes (although not in fruitflies).

Rost et al. show that this candidate VKOR gene encodes a small transmembrane protein that is found in the endoplasmic reticulum, a subcellular compartment. Leaving open the possibility that this protein might be only one component of a larger complex, they name it vitamin K epoxide reductase complex subunit 1 (VKORC1). The protein produced from the normal gene shows VKOR activity that is inhibited by warfarin. As expected, the mutant protein encoded by the gene from patients with a clotting-factor deficiency is inactive. Surprisingly, though, the mutant proteins from warfarin-resistant people are fairly sensitive to warfarin when tested in cultured cells; it remains to be seen how VKORC1 mutations cause warfarin resistance in vivo.

Li et al.2 did not have the benefit of access to patients with VKOR mutations, but overcame this obstacle in an ingenious way. They first ruled out genes from the relevant section of chromosome 16 that code for proteins with known or predicted functions, and then selected the 13 remaining candidates that encode proteins containing potential transmembrane portions (given that VKOR seemed to be a membrane protein).

Their subsequent identification of the VKOR gene relied on the use of RNA interference. In this technique, double-stranded small interfering RNA molecules (siRNAs) are inserted into cells, where they target matching messenger RNAs for degradation. The result is that the gene encoding those messenger RNAs is effectively silenced. Li et al. designed siRNAs to silence each of the 13 candidate genes in human cells with a high level of VKOR activity. They found that targeting only one gene — identical to Rost and colleagues' VKORC1 gene — caused a marked reduction in VKOR activity. When expressed in insect cells, this gene produced active VKOR, which was inhibited by warfarin. As the authors point out, this siRNA strategy should be useful in other positional cloning projects for which a functional assay can be devised, even if the activity of interest depends on many gene products.

VKOR was previously thought to be a large multiprotein complex. But Li et al. found that expressing just one protein could confer VKOR activity on insect cells that lacked it. Although it seems a heavy burden for a small protein without obvious relationship to other reductases, this molecule alone may be responsible for recycling vitamin K.

Now that we have tracked down the VKOR gene, it should be possible to identify the chemical source of the electrons that actually reduce vitamin K epoxide in living animals. Although various sulphydryl compounds work in the laboratory, a role for other common reductase cofactors, containing nicotinamide or flavin, can now be tested. It will also be important to characterize the structure of this apparent warfarin target, perhaps leading to the identification of better vitamin K antagonists for treating blood clots or, possibly, for poisoning rats. Finally, we now know that deficiency in multiple vitamin-K-dependent clotting factors can be caused by mutations in either of two genes — that for vitamin K carboxylase or, as shown by Rost et al., for VKORC1. Yet this may not be an exhaustive list. If it is not, then linkage studies in affected families might quickly lead to the discovery of yet more components of the vitamin K cycle.


  1. 1

    Rost, S. et al. Nature 427, 537–541 (2004).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Li, T. et al. Nature 427, 541–544 (2004).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Wu, S. M., Cheung, W. F., Frazier, D. & Stafford, D. W. Science 254, 1634–1636 (1991).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Kohn, M. H. & Pelz, H. J. Blood 96, 1996–1998 (2000).

    CAS  PubMed  Google Scholar 

  5. 5

    Fregizn, A. et al. Blood 100, 3229–3232 (2002).

    Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Sadler, J. K is for koagulation. Nature 427, 493–494 (2004).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.