William Vainchenker had long suspected a rogue JAK2 tyrosine kinase to be the cause of the myeloproliferative disorder, polycythemia vera. But with very limited resources at his disposal at the Gustave Roussy Institute in Paris, he could not afford large-scale sequencing efforts. Working with only three patients, his team concentrated on tracking down JAK2 gene mutations. "We were lucky," says Vainchenker. "We found the same mutation in two of them."

The culprit turned out to be a single nucleotide mutation at codon 617, changing valine to phenylalanine in JAK2's autoinhibitory domain. "Finding the mutation was the hard part," said Vainchenker. "The rest was pretty straightforward." The team went on to show that the V617F mutation produced a constitutive JAK2 activation in erythroid progenitor cells, causing spontaneous STAT-mediated transcription even in the absence of erythropoietin. They also demonstrated that mice transfected with V617F developed eythrocytosis, the characteristic symptom of polycythemia vera. Although Vainchenker never wavered in his pursuit of JAK2, few believed he was on the right track, says Josef Prchal of the University of Utah School of Medicine in Salt Lake City. "I didn't believe it myself until I read his paper."

Vainchenker's paper was one of three published in the spring of 2005 that unveiled the V617F mutation, but the other work approached the problem from a completely different direction. Prchal and Robert Kralovics had observed that many patients with myeloproliferative disease had a loss of heterozygosity (LOH) in the tip of the short arm of chromosome 9. Taking a closer look at the 9pLOH region in 244 patients using microsatellite mapping and DNA sequencing, Kralovics and Radek Skoda's team at the University Hospital Basel in Switzerland found the same V617F mutation. Their study also showed that patients with 9pLOH were more likely to develop complications such as bone marrow fibrosis or thrombosis.

Gary Gilliland's team at the Harvard Medical School took a cue from the molecular characterization of another myeloproliferative disease, chronic myelogenous leukemia; they suspected that a mutant tyrosine kinase could be responsible for polycythemia vera as well. Using an Internet protocol, the team screened granulocyte DNA from 325 patients with polycythemia vera, essential thrombocythemia or myelofibrosis for mutations in the activation loops and autoinhibitory domains of 85 tyrosine kinases. The V617F allele again emerged as the common factor. "We weren't sure we had the right thing at first because we didn't think one mutation could account for three diseases," says Gilliland.

The discovery of V617F invites comparisons to that of BCR-ABL, the tyrosine kinase mutation that came to light in 1983 as the direct cause of chronic myelogenous leukemia. Unlike BCR-ABL, however, V617F's true role is unclear. "The JAK2 mutation is clearly the most important discovery of a gene in polycythemia vera," says Prchal. "But I don't think it is a disease-inducing mutation." Ayalew Tefferi of the Mayo Clinic in Rochester, Minn., agrees. "We know that in mice it contributes to the phenotype, that is, erythrocytosis," he says. "But is it the cause of the disease in humans?"

Gilliland acknowledges that V617F raises many questions. "Why is it that of all the different ways that nature could potentially think of to mutate JAK2 to activate it, the only mutation we ever see in these patients is V617F?" His research has shown that the mutation is an acquired one even in families with an inherited proclivity for myeloproliferative disorders, leading him to conclude that a germline mutation could predispose people to acquire V617F. However, as he points out, "what possible germline allele could predispose you to only one point mutation among all the possible ones that could occur in the genome?" Prchal, Gilliland, and other researchers are now on the hunt for this "predisposition" allele.

Irrespective of V617F's pathogenetic significance, most experts value it as a diagnostic tool. In their initial work, Skoda, Gilli-land, and Vainchenker discovered the mutation in 60% to 90% of patients with polycythemia. More sensitive methods and stricter diagnostic criteria have since found that most but not all patients with polycythemia carry the allele. However, as Tefferi points out, V617F is not 100% specific to polycythemia vera; 50% of patients with essential thrombocythemia and idiopathic myelofibrosis also carry it. "That is the unfortunate part, compared to the BCR-ABL mutation," he says. Further, as his team at the Mayo Clinic has found, some patients with myelodysplastic syndrome and certain other forms of blood cancer also have the mutation. Given this diversity of phenotypes, some researchers now suggest that the mutation might actually be a 100% specific clue to a new and distinct clonal myeloproliferative disorder; the different phenotypes might then just be its different stages. From this perspective, they argue, V617F-positive myeloproliferative diseases should be treated as distinct from their wild-type counterparts.

While the debate continues about V617F's true role, researchers are exploring the therapeutic avenues it has opened. Inspired by the success in treating chronic myelogenous leukemia with a small-molecule enzyme inhibitor, imatinib - now a billion-dollar drug marketed by Novartis - many laboratories are now screening inhibitors for the JAK2 V617F tyrosine kinase. "Forget if the mutation is 50% or 90% of the cause," says Tefferi. "If we can inhibit this enzyme, will it make a difference?"