A Unifying Mutation in Chronic Myeloproliferative Disorders
http://www.100md.com
《新英格兰医药杂志》
In 1892, Louis Vaquez of Paris described a patient with cyanotic polycythemia; the autopsy disclosed massive enlargement of the spleen and liver. In 1903, William Osler, then at Johns Hopkins Hospital, reported on four patients with polycythemia, two of whom had splenomegaly. He gave credit to Vaquez for the earlier description, and the disorder was later named Osler–Vaquez disease, though today it is usually referred to as polycythemia vera.
In 1951, William Dameshek drew attention to the relationships among polycythemia vera, idiopathic myelofibrosis, and essential thrombocythemia and proposed that these diseases, as well as chronic myeloid leukemia and erythroleukemia, should be grouped together in the general category of myeloproliferative syndromes. This proposal may have been regarded as suspect at the time, but this year it seems to have been fully vindicated by four research groups that independently discovered that most patients with polycythemia vera and some with myelofibrosis and essential thrombocythemia have an identical acquired point mutation in the Janus kinase 2 (JAK2) gene.
The four members of the Janus kinase (JAK) family, Janus kinases 1, 2, and 3 and tyrosine kinase 2 (JAK1, JAK2, JAK3, and TYK2), have slightly different functions.1 Each has a kinase domain (JAK homology 1, or JH1) and a catalytically inactive pseudokinase domain with an important regulatory function (JAK homology 2, or JH2). To some, the presence of these two similar domains in the protein, one active and the other inactive, was reminiscent of the Roman god Janus, who looked simultaneously in two directions — hence the name.
The JAK proteins function as intermediates between membrane receptors and signaling molecules. When particular cytokines or growth factors bind to their receptors on the cell surface, JAK proteins, which are kinases associated with the cytoplasmic regions of these receptors, become phosphorylated and thereby activated. This activation creates docking sites for downstream molecules, notably those of the STAT (signal transducer and activator of transcription) family (see diagram, Panel A). Activated STAT molecules enter the nucleus, where they act as transcription factors. JAK2 seems to be activated particularly when receptors bind to hematopoietic growth factors. Mutations in the drosophila homologue of the human JAK2 gene are known to cause a leukemia-like phenotype in affected flies, and rare cases of leukemia in humans are associated with TEL–JAK2 fusion genes.
Involvement of Janus Kinases in Cytokine Signal Transduction (Panel A)and Structural Map of Janus Kinase 2 (Panel B).
On ligand binding by receptors (Panel A), Janus kinase (JAK) proteins make contact with the cytoplasmic domain of the receptor, where they catalyze the phosphorylation of tyrosines. This action, in turn, recruits STAT (signal transducer and activator of transcription) molecules, which are phosphorylated on their SRC homology 2 (SH2) domains, dimerize, and translocate to the nucleus, where they act as transcription factors. The JAK homology 1 (JH1) domain (Panel B) is the active kinase domain of JAK2, and the JAK homology 2 (JH2) domain is a pseudokinase with autoinhibitory properties. JAK2 also carries a band 4.1(f)-ezrin-radixin-moesin (FERM) domain. P indicates a site of autophosphorylation. The position of the mutated V617F codon is indicated by the arrow.
In this issue of the Journal, Kralovics and colleagues in Basel, Switzerland, and Pavia, Italy (pages 1779–1790), report that 83 of 128 patients with polycythemia vera (65 percent) had a guanine-to-thymine mutation encoding a valine-to-phenylalanine substitution at position 617 (V617F) in the JH2, or autoinhibitory, domain of JAK2 (see diagram, Panel B). They found the same mutation in 57 percent of patients with myelofibrosis and in 23 percent of patients with essential thrombocythemia. The mutation was found only in hematopoietic cells and must therefore be an acquired somatic mutation.
Normally, one would say that such consistent findings still required independent confirmation, but such confirmation is already available from both sides of the Atlantic. A research group in Paris2 reported the same mutation in 40 of 45 patients with polycythemia vera and a minority of patients with myelofibrosis and essential thrombocythemia. A Boston group3 has reported that 121 of 164 patients with polycythemia vera had the V617F mutation in their granulocytes, as did 37 of 115 patients with essential thrombocythemia and 16 of 46 patients with myelofibrosis. And a group in Cambridge, United Kingdom,4 found the V617F mutation in 71 of 73 patients with polycythemia vera, 29 of 51 patients with essential thrombocythemia, and 8 of 16 patients with myelofibrosis. Though the proportions of patients with the mutation vary somewhat in the different series, there can be little doubt that the observation is real and likely to be of major importance.
Although these four articles all appeared within a period of just six weeks, the logic and preceding experimental work were essentially different in each case. The Swiss–Italian study was based on the prior observation that a minority of patients with polycythemia vera had a loss of heterozygosity in a restricted region of chromosome 9p that included the site of the JAK2 gene, and this led to a search for mutations in JAK2. The French study followed from the observation that the formation in vitro of the endogenous erythroid colonies that are characteristic of polycythemia vera could be inhibited by small molecules that block the action of JAK2. The Boston group, stimulated by the known involvement of activated tyrosine kinases in chronic myeloid leukemia, chronic myelomonocytic leukemia, and the hypereosinophilic syndrome, undertook a high-throughput DNA sequence analysis of the activation loops and autoinhibitory domains of 85 tyrosine kinases. The Cambridge group focused on the key role of JAK2 in signal transduction from multiple hematopoietic growth factor receptors and thus selected it as an attractive candidate for further study.
The remarkable finding of the V617F mutation by four independent research groups raises as many questions as it answers. How, for example, does this mutation contribute to the pathogenesis of these myeloproliferative disorders? It seems likely that it disrupts the JH2 inhibitory regulation of JAK2, leaving the enzyme constitutively active. The activated kinase may then bind to a receptor and recruit STATs in the total absence of hematopoietic growth factor or in the presence of only trace quantities of the factor. Such a mechanism fits nicely with the observation by the Cambridge group that all patients with polycythemia vera who had the mutation had cells in their blood that formed erythroid colonies in vitro in the absence of erythropoietin.4 The net result could be chronic low-grade but excessive stimulation of erythropoiesis, a characteristic of polycythemia vera.
Why are some patients with polycythemia vera heterozygous and others homozygous for the mutation? The evidence suggests that homozygosity for the mutant gene results from the mitotic recombination of chromatids bearing the mutation and not from a second mutation in a mutant heterozygous lineage. Either way, the homozygosity may be the result of two distinct steps, the first of which is enough to produce features of the disease.
And what of the patients with polycythemia vera, essential thrombocythemia, or myelofibrosis who apparently do not have JAK2 mutations? Can we expect to find another unifying mutation equivalent to this one? If so, will it be in the JAK–STAT pathway or independent of it?
Even if Dameshek was right to link these myeloproliferative disorders, how exactly could the same acquired mutation in the JAK2 gene cause three clinical entities that are more or less distinct? The simplest explanation is merely to postulate the existence of other genetic abnormalities that either are preexisting or occur after the acquisition of the JAK2 mutation. Other possibilities must also exist.
And what is the basis for the progression from a seemingly benign condition (polycythemia vera) to acute leukemia in some, but not all, cases? The question brings to mind the analogy with chronic myeloid leukemia, a disease associated with an activated tyrosine kinase, which also starts gently and progresses to an aggressive termination. Extending the analogy, one may speculate that a small molecule that inhibits normal and mutant JAK2 could prove clinically useful in these myeloproliferative disorders, as it has so impressively in chronic myelogenous leukemia. At the very least, this newly identified molecular lesion will form the basis of a new classification. But it seems likely that patients will also eventually derive substantial benefit from the discovery.
Source Information
Dr. Goldman is a Fogarty Scholar in the Hematology Branch of the National Heart, Lung, and Blood Institute, Bethesda, Md.
References
Yamaoka K, Saharinen P, Pesu M, Holt VE III, Silvennoinen O, O'Shea JJ. The Janus kinases (Jaks). Genome Biol 2004;5:253-253.
James C, Ugo V, Le Couedic J-P, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature (in press).
Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell (in press).
Baxter EJ, Scott LM, Campbell PJ, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005;365:1054-1061.(John M. Goldman, D.M.)
In 1951, William Dameshek drew attention to the relationships among polycythemia vera, idiopathic myelofibrosis, and essential thrombocythemia and proposed that these diseases, as well as chronic myeloid leukemia and erythroleukemia, should be grouped together in the general category of myeloproliferative syndromes. This proposal may have been regarded as suspect at the time, but this year it seems to have been fully vindicated by four research groups that independently discovered that most patients with polycythemia vera and some with myelofibrosis and essential thrombocythemia have an identical acquired point mutation in the Janus kinase 2 (JAK2) gene.
The four members of the Janus kinase (JAK) family, Janus kinases 1, 2, and 3 and tyrosine kinase 2 (JAK1, JAK2, JAK3, and TYK2), have slightly different functions.1 Each has a kinase domain (JAK homology 1, or JH1) and a catalytically inactive pseudokinase domain with an important regulatory function (JAK homology 2, or JH2). To some, the presence of these two similar domains in the protein, one active and the other inactive, was reminiscent of the Roman god Janus, who looked simultaneously in two directions — hence the name.
The JAK proteins function as intermediates between membrane receptors and signaling molecules. When particular cytokines or growth factors bind to their receptors on the cell surface, JAK proteins, which are kinases associated with the cytoplasmic regions of these receptors, become phosphorylated and thereby activated. This activation creates docking sites for downstream molecules, notably those of the STAT (signal transducer and activator of transcription) family (see diagram, Panel A). Activated STAT molecules enter the nucleus, where they act as transcription factors. JAK2 seems to be activated particularly when receptors bind to hematopoietic growth factors. Mutations in the drosophila homologue of the human JAK2 gene are known to cause a leukemia-like phenotype in affected flies, and rare cases of leukemia in humans are associated with TEL–JAK2 fusion genes.
Involvement of Janus Kinases in Cytokine Signal Transduction (Panel A)and Structural Map of Janus Kinase 2 (Panel B).
On ligand binding by receptors (Panel A), Janus kinase (JAK) proteins make contact with the cytoplasmic domain of the receptor, where they catalyze the phosphorylation of tyrosines. This action, in turn, recruits STAT (signal transducer and activator of transcription) molecules, which are phosphorylated on their SRC homology 2 (SH2) domains, dimerize, and translocate to the nucleus, where they act as transcription factors. The JAK homology 1 (JH1) domain (Panel B) is the active kinase domain of JAK2, and the JAK homology 2 (JH2) domain is a pseudokinase with autoinhibitory properties. JAK2 also carries a band 4.1(f)-ezrin-radixin-moesin (FERM) domain. P indicates a site of autophosphorylation. The position of the mutated V617F codon is indicated by the arrow.
In this issue of the Journal, Kralovics and colleagues in Basel, Switzerland, and Pavia, Italy (pages 1779–1790), report that 83 of 128 patients with polycythemia vera (65 percent) had a guanine-to-thymine mutation encoding a valine-to-phenylalanine substitution at position 617 (V617F) in the JH2, or autoinhibitory, domain of JAK2 (see diagram, Panel B). They found the same mutation in 57 percent of patients with myelofibrosis and in 23 percent of patients with essential thrombocythemia. The mutation was found only in hematopoietic cells and must therefore be an acquired somatic mutation.
Normally, one would say that such consistent findings still required independent confirmation, but such confirmation is already available from both sides of the Atlantic. A research group in Paris2 reported the same mutation in 40 of 45 patients with polycythemia vera and a minority of patients with myelofibrosis and essential thrombocythemia. A Boston group3 has reported that 121 of 164 patients with polycythemia vera had the V617F mutation in their granulocytes, as did 37 of 115 patients with essential thrombocythemia and 16 of 46 patients with myelofibrosis. And a group in Cambridge, United Kingdom,4 found the V617F mutation in 71 of 73 patients with polycythemia vera, 29 of 51 patients with essential thrombocythemia, and 8 of 16 patients with myelofibrosis. Though the proportions of patients with the mutation vary somewhat in the different series, there can be little doubt that the observation is real and likely to be of major importance.
Although these four articles all appeared within a period of just six weeks, the logic and preceding experimental work were essentially different in each case. The Swiss–Italian study was based on the prior observation that a minority of patients with polycythemia vera had a loss of heterozygosity in a restricted region of chromosome 9p that included the site of the JAK2 gene, and this led to a search for mutations in JAK2. The French study followed from the observation that the formation in vitro of the endogenous erythroid colonies that are characteristic of polycythemia vera could be inhibited by small molecules that block the action of JAK2. The Boston group, stimulated by the known involvement of activated tyrosine kinases in chronic myeloid leukemia, chronic myelomonocytic leukemia, and the hypereosinophilic syndrome, undertook a high-throughput DNA sequence analysis of the activation loops and autoinhibitory domains of 85 tyrosine kinases. The Cambridge group focused on the key role of JAK2 in signal transduction from multiple hematopoietic growth factor receptors and thus selected it as an attractive candidate for further study.
The remarkable finding of the V617F mutation by four independent research groups raises as many questions as it answers. How, for example, does this mutation contribute to the pathogenesis of these myeloproliferative disorders? It seems likely that it disrupts the JH2 inhibitory regulation of JAK2, leaving the enzyme constitutively active. The activated kinase may then bind to a receptor and recruit STATs in the total absence of hematopoietic growth factor or in the presence of only trace quantities of the factor. Such a mechanism fits nicely with the observation by the Cambridge group that all patients with polycythemia vera who had the mutation had cells in their blood that formed erythroid colonies in vitro in the absence of erythropoietin.4 The net result could be chronic low-grade but excessive stimulation of erythropoiesis, a characteristic of polycythemia vera.
Why are some patients with polycythemia vera heterozygous and others homozygous for the mutation? The evidence suggests that homozygosity for the mutant gene results from the mitotic recombination of chromatids bearing the mutation and not from a second mutation in a mutant heterozygous lineage. Either way, the homozygosity may be the result of two distinct steps, the first of which is enough to produce features of the disease.
And what of the patients with polycythemia vera, essential thrombocythemia, or myelofibrosis who apparently do not have JAK2 mutations? Can we expect to find another unifying mutation equivalent to this one? If so, will it be in the JAK–STAT pathway or independent of it?
Even if Dameshek was right to link these myeloproliferative disorders, how exactly could the same acquired mutation in the JAK2 gene cause three clinical entities that are more or less distinct? The simplest explanation is merely to postulate the existence of other genetic abnormalities that either are preexisting or occur after the acquisition of the JAK2 mutation. Other possibilities must also exist.
And what is the basis for the progression from a seemingly benign condition (polycythemia vera) to acute leukemia in some, but not all, cases? The question brings to mind the analogy with chronic myeloid leukemia, a disease associated with an activated tyrosine kinase, which also starts gently and progresses to an aggressive termination. Extending the analogy, one may speculate that a small molecule that inhibits normal and mutant JAK2 could prove clinically useful in these myeloproliferative disorders, as it has so impressively in chronic myelogenous leukemia. At the very least, this newly identified molecular lesion will form the basis of a new classification. But it seems likely that patients will also eventually derive substantial benefit from the discovery.
Source Information
Dr. Goldman is a Fogarty Scholar in the Hematology Branch of the National Heart, Lung, and Blood Institute, Bethesda, Md.
References
Yamaoka K, Saharinen P, Pesu M, Holt VE III, Silvennoinen O, O'Shea JJ. The Janus kinases (Jaks). Genome Biol 2004;5:253-253.
James C, Ugo V, Le Couedic J-P, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature (in press).
Levine RL, Wadleigh M, Cools J, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell (in press).
Baxter EJ, Scott LM, Campbell PJ, et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005;365:1054-1061.(John M. Goldman, D.M.)