First author

Warfarin has long been a household name thanks to its uses as a drug and in controlling rodent pests. Its protein target is less well known — the enzyme VKOR, which, in mammals, catalyses the generation of vitamin K hydroquinone. This is an important component in the vitamin K cycle and is required to sustain blood coagulation. Until now, no one had been able to determine exactly how mammalian VKOR worked because nobody had succeeded in purifying sufficient quantities for structural studies. Weikai Li at Harvard Medical School in Boston, Massachusetts, and his colleagues have now discovered a stable bacterial version of VKOR that has allowed them to examine the enzyme's crystal structure (see page 507). Li tells Nature more.

Why did you do this study?

We wanted to understand how mammalian VKOR works because its function has important medical implications. Warfarin is a commonly used anticoagulant drug that works by inhibiting VKOR, but it has a narrow therapeutic window. Too large a dose can cause lethal bleeding; too low and it isn't effective. Genetic variations exist in VKOR among patients, which affect the enzyme's sensitivity to warfarin, and that's one of the reasons it is hard to get the dose right.

What challenges did you face?

First we had to screen a number of bacterial VKOR variants to find one that would be appropriate for structural studies. Then we had to isolate the protein. For that, we had to find the right detergent: one that could dissolve the membrane surrounding the protein without denaturing the protein. The next step was obtaining pure and regularly ordered crystals that would yield the best diffraction. That's an essential and often rate-limiting step in structure determination.

Are your findings of value to human health?

Some people have mutations in VKOR that make it resistant to warfarin. By viewing the protein's structure, we can see the locations of mutated amino acids. A better understanding of how these amino acids interact with the drug may make it possible to devise a safer anticoagulant that's easier to dose than warfarin. Safe, appropriate dosage is crucial — a blood clot in the leg of someone who's received a subtherapeutic dose might travel to the lung and kill the person.

Are there any other implications?

A colleague of ours discovered that the bacterium that causes tuberculosis uses a protein homologous to VKOR. It's a separate study, but our work may help his group to design new antibiotics for tuberculosis.