2009.05.28 – Mutation in TET2 in Myeloid Cancers


The New England Journal of Medicine

Volume 360:2289-2301 May 28, 2009 Number 22

Mutation in TET2 in Myeloid Cancers

François Delhommeau, Pharm.D., Ph.D., Sabrina Dupont, Ph.D., Véronique Della Valle, Ph.D., Chloé James, M.D., Ph.D., Severine Trannoy, B.S., Aline Massé, Ph.D., Olivier Kosmider, Pharm.D., Ph.D., Jean-Pierre Le Couedic, B.S., Fabienne Robert, Ph.D., Antonio Alberdi, Ph.D., Yann Lécluse, B.S., Isabelle Plo, Ph.D., François J. Dreyfus, M.D., Christophe Marzac, M.D., Nicole Casadevall, M.D., Catherine Lacombe, M.D., Ph.D., Serge P. Romana, M.D., Ph.D., Philippe Dessen, M.D., Ph.D., Jean Soulier, M.D., Ph.D., Franck Viguié, M.D., Michaela Fontenay, M.D., Ph.D., William Vainchenker, M.D., Ph.D., and Olivier A. Bernard, Ph.D.


Background The myelodysplastic syndromes and myeloproliferative disorders are associated with deregulated production of myeloid cells. The mechanisms underlying these disorders are not well defined.

Methods We conducted a combination of molecular, cytogenetic, comparative-genomic-hybridization, and single-nucleotide–polymorphism analyses to identify a candidate tumor-suppressor gene common to patients with myelodysplastic syndromes, myeloproliferative disorders, and acute myeloid leukemia (AML). The coding sequence of this gene, TET2, was determined in 320 patients. We analyzed the consequences of deletions or mutations in TET2 with the use of in vitro clonal assays and transplantation of human tumor cells into mice.

Results We initially identified deletions or mutations in TET2 in three patients with myelodysplastic syndromes, in three of five patients with myeloproliferative disorders, in two patients with primary AML, and in one patient with secondary AML. We selected the six patients with myelodysplastic syndromes or AML because they carried acquired rearrangements on chromosome 4q24; we selected the five patients with myeloproliferative disorders because they carried a dominant clone in hematopoietic progenitor cells that was positive for the V617F mutation in the Janus kinase 2 (JAK2) gene. TET2 defects were observed in 15 of 81 patients with myelodysplastic syndromes (19%), in 24 of 198 patients with myeloproliferative disorders (12%) (with or without the JAK2 V617F mutation), in 5 of 21 patients with secondary AML (24%), and in 2 of 9 patients with chronic myelomonocytic leukemia (22%). TET2 defects were present in hematopoietic stem cells and preceded the JAK2 V617F mutation in the five samples from patients with myeloproliferative disorders that we analyzed.

Conclusions Somatic mutations in TET2 occur in about 15% of patients with various myeloid cancers.

The myelodysplastic syndromes and myeloproliferative disorders are clonal disorders that arise in hematopoietic stem cells,1 whereas acute myeloid leukemia (AML) can arise from stem cells or more mature progenitor cells.2 The main feature of the heterogeneous subtypes of the myelodysplastic syndromes is ineffective hematopoiesis that affects one or several lineages.3 In about a third of patients, the disorder progresses to AML. Some of the molecular abnormalities that are associated with specific subtypes of myelodysplastic syndromes4 or the transformation of myelodysplastic syndromes to AML5 have been identified, but the basis of these related syndromes remains poorly understood.6

The myeloproliferative disorders, especially polycythemia vera and essential thrombocythemia, are frequently associated with a somatic mutation, V617F, in the Janus kinase 2 (JAK2) gene, which results in a constitutively active Janus protein kinase.7,8,9 In mouse models, the presence of JAK2 V617F alone can cause myeloproliferative disorders but does not give hematopoietic stem cells a proliferative advantage.10,11,12 Recent evidence indicates that JAK2 V617F does not account for all the pathophysiological effects of the myeloproliferative disorders.13,14,15,16 For example, JAK2 V617F was absent in some patients in whom a JAK2 V617F–positive myeloproliferative disorder progressed to AML.17

We hypothesized that an early genetic event in a hematopoietic stem cell could be common to the myelodysplastic syndromes, myeloproliferative disorders, and AML. By using various genetic approaches, we identified a tumor-suppressor gene, TET2, that is deleted or mutated in patients with various myeloid cancers.



Table 1 in the Supplementary Appendix (available with the full text of this article at NEJM.org) lists the clinical features of the patients, whose conditions were diagnosed with the use of standard international criteria.18,19,20 Samples of bone marrow or blood were obtained from 320 patients (84 with myelodysplastic syndromes, 203 with myeloproliferative disorders, 2 with primary AML, 22 with secondary AML, and 9 with chronic myelomonocytic leukemia) after the patients provided written informed consent and the local research ethics committees (at Hôtel-Dieu and Cochin hospitals) gave approval.

Genetic Analyses

A detailed description of the materials and methods used in these experiments is provided in the Supplementary Appendix. Briefly, bone marrow mononuclear cells, lymphocytes, or granulocytes were isolated and stored in liquid nitrogen. Cellular experiments were performed as described previously.10,11 Purified CD34+ cells (1×105 to 10×105 cells) were injected intravenously into mice that had undergone sublethal irradiation.14

Polymerase-chain-reaction (PCR) assays and direct sequencing reactions were performed with primers that are listed in Table 2 of the Supplementary Appendix. JAK2 mutational status was determined as described previously.10

In comparative-genomic-hybridization and single-nucleotide–polymorphism (SNP) analyses, paired DNA samples from malignant and nonmalignant cells were analyzed with the use of 244K oligonucleotide arrays (Agilent Technologies) and Affymetrix 250K SNP (Affymetrix) arrays.

No commercial entity was involved in the conduct of the study, the analysis or storage of the data, or the preparation of the manuscript. The authors vouch for the completeness and accuracy of the data and the analysis.


Deleted Region on Chromosome 4q24

We previously reported on six patients with myelodysplastic syndromes or AML who had acquired rearrangements on chromosome 4q24.21 In two of these patients, the chromosomal abnormality was found in both myeloid and lymphoid cells, indicating the involvement of a lymphomyeloid progenitor. In this study, we discovered a deleted region on chromosome 4q24 containing a single gene, TET2, in cells from the three patients with AML, by means of fluorescence in situ hybridization (Figure 1A). Bone marrow cells from the three patients with myelodysplastic syndromes (Patients MDS01, MDS02, and MDS03) had a similar deletion. Taken together, five patients had a heterozygous loss, and one had a deletion of both copies.

View larger version (39K):

[in this window]

Figure 1. Structural DNA Rearrangements and Mutations in TET2 on Chromosome 4q24 in Subgroups of Patients.

In Panel A, a minimally deleted region on chromosome 4q24 is seen in subgroups of patients with myelodysplastic syndromes and acute myeloid leukemia (AML). This region was defined with the use of RP11-1061M121, RP11-356L5, and RP11-16G16 bacterial artificial chromosomes (represented by horizontal lines with double arrows) in fluorescence in situ hybridization. The region encompasses a unique gene, TET2. In Panel B, acquired abnormalities on chromosome 4q24 are shown for three of five patients with myeloproliferative disorders who carried the JAK2 V617F mutation. The abnormalities were detected with the use of high-resolution single-nucleotide-polymorphism (SNP) and comparative-genomic-hybridization (CGH) arrays. The blue bars indicate the regions affected by loss of heterozygosity (LOH). Patient MPD05 had a smaller LOH because of an acquired 325-kb deletion. TET2 was the only open reading frame observed in this region. In Panel C, nucleotide sequences of DNA from bone marrow (BM) samples from patients with AML and myelodysplastic syndromes are compared with wild-type and nontumor samples from the same patients (when available). The presence of wild-type nucleotide sequences in the samples was attributed to residual normal hematopoietic cells. Matched samples consisted of bone marrow cells obtained after transplantation, during a period of clinical health (Patient nAML1); B type lymphoid cells transformed by Epstein–Barr virus (Patient nAML2); and peripheral-blood mononuclear cells stimulated by a phorbol ester (Patient MDS03). In Panel D, wild-type TET2 sequences are compared with TET2 sequences in tumor cells and matched mononuclear cells from Patients MPD01 and MPD04. The dashed vertical lines frame the nine-nucleotide deletion shown for Patient MPD04. The arrow indicates the presence of a small, residual peak of wild-type allele in mononuclear cells from Patient MPD01. In Panels C and D, asterisks indicate nucleotide changes. In Panel E, alignment of the conserved region spanning amino acids 1842 through 1921 of human TET2 protein is shown. Invariant amino acids (identical in all species) are marked by a star, with conserved amino acids indicated by a double dot under the alignment. An arrow points from the amino acids targeted by mutational events in myeloid cancers to the resulting amino acid. In Panel F, the TET protein family is shown with the location of acquired missense mutations. The arrowhead indicates the fusion point of mixed-lineage leukemia (MLL) protein on TET1. The conserved regions that are shared by all TET family proteins appear as shaded boxes. DM denotes Drosophila melanogaster, Fugu Takifugu rubripes, Hs Homo sapiens, and Mm Mus musculus.

We discovered involvement of the same region on chromosome 4q24 using a different method in patients with myeloproliferative disorders. Two subgroups of patients with myeloproliferative disorders and different profiles of JAK2 V617F clonal expansion have been described previously.10 The first subgroup had a low proportion of CD34+ cells bearing the JAK2 V617F mutation and a high proportion of granulocytes with the mutation, whereas the second subgroup had a high proportion of CD34+ cells bearing JAK2 V617F. We analyzed samples from five patients from the second subgroup (Patients MPD01 through MPD05) using comparative genomic hybridization and SNP arrays to compare DNA from cells that were presumed to be affected (granulocytes) with DNA from cells that were presumed to be normal (mononuclear cells or lymphocytes) (Figure 1B). In this subgroup, one patient with primary myelofibrosis (Patient MPD01) and one with polycythemia vera (Patient MPD04) had loss of heterozygosity (LOH) without copy-number modification in chromosome 4, ranging from q22 to the end of the long arm of the chromosome, which was not found in presumably normal cells.22 A third patient (Patient MPD05) had an acquired 325-kb deletion in the region on chromosome 4q24 containing TET2 as a single candidate gene. This region was normal in Patients MPD02 and MPD03.

Acquired Mutations in TET2

The TET2 gene comprises 11 exons spread over 150 kb and is widely expressed.23,24 In humans, the predicted TET2 protein belongs to a three-member family containing two highly conserved regions (Figure 1, and Fig. 1 and 2 in the Supplementary Appendix). TET2 has not previously been implicated in human cancers, but TET1 is fused to the mixed-lineage leukemia (MLL) gene in the chromosomal translocation t(10;11)(p12;q23), which occurs rarely in patients with acute leukemia.23,24 The function of the TET proteins is unknown, but the presence of the conserved carboxy-terminal domain in the MLL–TET1 fusion protein suggests its importance in cellular transformation.

The loss of both copies of TET2 in Patient MDS01 and the loss of one copy in eight other patients suggested that TET2 is a tumor-suppressor gene. We therefore searched for nucleotide sequence variation in the coding region of the remaining copy of TET2 in the eight patients with either a heterozygous deletion or a copy-neutral loss of heterozygosity. A comparison of these eight sequences with human sequences in the databases identified nucleotide changes in the remaining TET2 genes in six patients. One change caused a frame shift, two generated stop codons, one was an in-frame deletion, and two resulted in amino acid substitutions of extremely conserved residues (Table 1 and Figure 1C, 1D, and 1E, and Fig. 2 in the Supplementary Appendix). No variations were found in the coding sequence of TET2 in Patients sAML1 and MPD05. TET2 coding sequences were wild type in Patients MPD02 and MPD03, who retained both copies of TET2.

View this table:

[in this window]

Table 1. Clinical and Molecular Data for Patients with TET2 Mutations.

To investigate whether these variations were somatically acquired in myeloid cells, we studied DNA obtained from lymphocytes from four patients. In analyses of the coding sequences of TET2 in these DNA samples, we found no mutations in samples from Patients MDS03, MPD04, and nAML2 (Figure 1C and 1D) and no deletion in a sample from Patient MPD05 (data not shown). In addition, wild-type sequence was detected in mononuclear cells from Patient MPD01 (Figure 1D). In Patient nAML1, the wild-type sequence predominated, as compared with the mutated sequence, in a sample obtained when the patient was in remission. Together, these data indicate that the TET2 defects were somatically acquired.

To determine the prevalence of TET2 mutations in malignant myeloid disorders, we analyzed the TET2 coding sequence in 309 available samples from 81 patients with various subtypes of myelodysplastic syndromes, 21 patients with secondary AML, 9 patients with chronic myelomonocytic leukemia, 181 patients with JAK2 V617F–positive myeloproliferative disorders, and 17 patients with JAK2 V617F–negative myeloproliferative disorders. Sequence variants of TET2, resulting in conserved amino acid substitution, stop codons, or frame shifts, were detected in 45 of 309 patients (15%) (Table 1 and Figure 1E and 1F, and Fig. 3 in the Supplementary Appendix). In Patient MDS07, amplification of the 5′ region of exon 11 on PCR failed, suggesting an uncharacterized structural genomic rearrangement in this region. TET2 defects were found in 22 of 111 patients with various types of myelodysplastic syndromes, in 21 of 181 patients with myeloproliferative disorders associated with JAK2 V617F, in 1 of 6 patients with disease associated with a W515L/K mutation in the myeloproliferative leukemia virus oncogene (MPL), and in 2 of 11 patients who did not carry either an MPL W515L/K or JAK2 V617F mutation.

Overall, we found TET2 defects in diverse myeloid disorders, with a prevalence of 15% (46 of 309 patients). Because most of such mutations are predicted to truncate the protein, they could result in partial or total loss of function of the TET2 protein. In 25 of 55 patients with TET2 defects, two different mutations were detected, which probably targeted both copies of TET2 (Table 1). This inference was confirmed in Patient MDS42 by subcloning and analysis of individual DNA molecules (Fig. 4 in the Supplementary Appendix). A single defect was observed in 30 of 55 samples, suggesting that haploinsufficiency of TET2 has a role in these cancers.

Early Progenitor Cells with TET2 Mutations

We sought TET2 defects in CD34+ cells, which include hematopoietic stem cells and hematopoietic progenitors, from four patients with myelodysplastic syndromes (Patients MDS03, MDS09, MDS28, and MDS35). In all four patients, a mutated TET2 sequence was detected (Figure 2, and Fig. 5 and 6 in the Supplementary Appendix). In one patient (Patient MDS03), CD34+ cells were analyzed in the first phase of refractory anemia with excess blasts (RAEB1, 5 to 9% blasts) and in the second phase (RAEB2, 10 to 19% blasts). The wild-type sequence was detected in the RAEB1 phase, but only a mutated sequence was found in the RAEB2 phase, suggesting the presence of expanded clones of progenitors bearing the mutation during disease progression (Figure 2A).

View larger version (34K):

[in this window]

Figure 2. TET2 Mutations in Fractionated CD34+ Cells from Patients with Myelodysplastic Syndromes.

In samples from patients with myelodysplastic syndromes, mutated TET2 sequence is observed in immature CD34+ cells and is associated with in vivo expansion of the mutated clone. In Panel A, sequencing chromatograms of sorted CD34+ cells from Patient MDS03 are shown in samples obtained during phases of refractory anemia with excess blasts of 5 to 9% (RAEB1) and with excess blasts of 10 to 19% (RAEB2). Nucleotide sequences that were observed in an unsorted bone marrow sample from the patients and in a wild-type control sample are shown for comparison purposes. Asterisks indicate the mutated nucleotide. In Panel B, polymerase-chain-reaction (PCR) assay followed by restriction-fragment–length polymorphism (RFLP) analysis shows DNA that was isolated from sorted CD34+CD38– and CD34+CD38+ cells from Patient MDS03 at the RAEB1 phase. Amplified fragments were digested with the use of the Taq endonuclease (Tas1) and underwent size fractionation with the use of agarose-gel electrophoresis. The proportion of TET2 that was mutated was evaluated by measuring the intensity of the mutated or wild-type signal relative to that of the signal generated by both alleles (wild type plus mutated), as compared with a control sample. MW denotes molecular weight (see Fig. 5 in the Supplementary Appendix for details). In Panel C, the results of PCR–RFLP analysis of TET2 that was directly performed from sorted CD34+CD38– and CD34+CD38+ cells from Patient MDS09 with the use of the Bacillus stearothermophilus endonuclease (BseL1) are shown. In Panel D, PCR-RFLP genotyping analysis with the use of BseL1 endonuclease shows sorted CD34+CD38– and CD34+CD38+ clones from Patient MDS09; the clones were grown at one cell per well. The proportion of mutated clones differs significantly between the two cell populations (P=0.01 by Fisher’s exact test). The histograms represent the fraction of clones with wild-type or mutated TET2. The absence of wild-type fragment in CD34+CD38+ clones is indicated by asterisks.

We fractionated the CD34+ cells from these four patients into CD34+CD38– populations (corresponding to stem cells and multipotent progenitors) and into CD34+CD38+ populations (corresponding to more mature progenitors). In Patient MDS03, mutations in TET2 were found in 16% of cells in the CD34+CD38– population and in 54% of cells in the CD34+CD38+ population (Figure 2B). The corresponding proportions for Patient MDS09 were 26% and 48%, respectively (Figure 2C). Further analysis was performed by seeding single hematopoietic progenitors from Patient MDS09 on a stromal-cell layer or in a methylcellulose medium. A TET2 mutation was identified in 8 of 32 clones (25%) derived from CD34+CD38– cells and in 18 of 30 clones (60%) derived from CD34+CD38+ cells (Figure 2D). Wild-type TET2 was not always detected in clones bearing a mutated TET2, suggesting its loss in a minority of the cells.

In Patients MDS28 and MDS35, the difference in the occurrence of TET2 mutations in CD34+CD38– and CD34+CD38+ cells was evaluated from the respective peak intensities and confirmed by subcloning and sequencing of bacterial clones (Fig. 6 in the Supplementary Appendix). In Patient MDS28, the mutated TET2 sequence was barely detectable in CD34+CD38– cells, whereas it represented one third of the TET2 sequences from the CD34+CD38+ cells. The data for these four patients with myelodysplastic syndromes indicate that TET2 mutations were present in a small number of immature CD34+CD38– cells and increased in the population of mature progenitors.

Timing of TET2 and JAK2 V617F Mutations

We analyzed hematopoietic progenitors from five patients who had myeloproliferative disorders with mutations in TET2 and JAK2. Immature CD34+CD38– cells were seeded at one cell per well in culture conditions that supported lymphomyeloid differentiation, whereas CD34+CD38+ cells were grown in erythroid–granulocytic methylcellulose assays. In all cells that were tested, both TET2 and JAK2 defects were present in clones containing lymphoid and myeloid cells together (Figure 3A and 3B). Overall, after testing of 232 clones from immature progenitors, the JAK2 V617F mutation was not observed in the absence of the TET2 mutation. These results demonstrate that in patients with myeloproliferative disorders (as in patients with myelodysplastic syndromes), the TET2 mutation is present in immature progenitors. In addition, the findings indicate that in the five patients with myeloproliferative disorders, TET2 defects preceded JAK2 mutations during the evolution of the disease. Of course, this finding does not formally rule out an alternative sequence of mutations in other patients with myeloproliferative disorders.

View larger version (48K):

[in this window]

Figure 3. TET2 and JAK2 Defects in Clones Containing Lymphoid and Myeloid Cells.

The TET2 mutation is a primary event in myeloproliferative disorders and is associated with in vivo expansion of the mutated clone. Sequence analyses of TET2 and JAK2 mutations in patients with myeloproliferative disorders were performed in CD34+CD38– cells grown in culture derived from clonal B cells, myeloid cells, and natural killer cells (BMNK) and in CD34+CD38+ cells in methylcellulose culture for differentiation of burst-forming unit–erythroid (BFU-E) or colony-forming unit–granulocyte–macrophage (CFU-GM) activity. In Panel A, the sequencing chromatograms show representative results from three clones per patient. Asterisks indicate the positions of nucleotide changes. For Patient MPD05, sequence analysis was performed with the use of a single-nucleotide polymorphism to detect the loss of heterozygosity (LOH) resulting from the deletion in TET2. In Panel B, the histograms show the fractions of immature progenitors (BMNK) and mature progenitors (BFU-E and CFU-GM) carrying defects in both TET2 and JAK2, clones with mutated TET2, and nonmutated clones. Panel C shows the fractions of immature progenitors (BMNK) and mature myeloid progenitors (BFU-E and CFU-GM) carrying TET2 defects in JAK2 wild-type progenitor cells from Patients MPD05 and MPD20. In Panels B and C, the numbers of analyzed clones are shown on the bars.

In Patients MPD01, MPD04, and MPD35, almost all the colonies derived from immature and mature progenitors carried a TET2 mutation, indicating the dominance of the TET2-mutated clone at early stages of hematopoiesis (Figure 3B). In Patients MPD05 and MPD20, most immature progenitor cells were wild type, whereas more mature progenitor cells had a mutated TET2. Within JAK2 wild-type progenitors from these two patients, the proportion of clones with TET2 defects in the immature population was 2 of 37 cells in Patient MPD05 and none of 34 cells in Patient MPD20; in the mature population, the proportion was 10 of 23 cells in Patient MPD05 and 9 of 54 cells in Patient MPD20 (Figure 3C). The proportions of mutated clones differed significantly between the two populations (P<0.001 for Patient MPD05 and P=0.01 for Patient MPD20, by Fisher’s exact test), indicating that the increase in cells with TET2 mutations does not require the presence of the JAK2 V617F mutation.

Studies of TET2 Mutation in Mice

To add to the evidence that the TET2 mutation occurs initially in hematopoietic stem cells, we grafted CD34+ cells from patients with polycythemia vera or myelofibrosis in nonobese diabetic mice with severe combined immunodeficiency (NOD–SCID) and depletion of natural killer cells, independently of the proportion of JAK2 V617F mutations.14 Purified CD34+ cells isolated from patients with myeloproliferative disorders and the JAK2 V617F mutation either with the TET2 mutation (Patients MPD01 and MPD04) or without the TET2 mutation (Patients MPD09, MPD11, and MPD27) were injected into NOD–SCID mice (Fig. 7 in the Supplementary Appendix).

The percentage of human cells from the three samples without a TET2 mutation decreased with time, whereas the percentage of human cells from the two TET2-mutated samples increased with time (Figure 4A). Furthermore, human hematopoietic reconstitution from TET2-mutated samples was skewed toward myeloid progenitors, at the expense of lymphoid progenitors, as judged from CD33 and CD19 antigen expression, unlike the predominantly lymphoid reconstitution observed with wild-type hematopoietic stem cells25 (Figure 4B).

View larger version (36K):

[in this window]

Figure 4. Repopulating Capacities of Stem Cells from Patients with Myeloproliferative Disorders, with and without TET2 Mutations, Transplanted into Mice.

Hematopoietic stem cells with the JAK2 V617F mutation from patients with myeloproliferative disorders with TET2 defects show enhanced repopulating capacities when the cells are grafted into nonobese diabetic mice with severe combined immunodeficiency (NOD–SCID). CD34+ cells from patients with myeloproliferative disorders were injected intravenously into NOD–SCID mice that had undergone sublethal irradiation. The percentage of human CD45+ cells in mouse bone marrow was monitored at 3, 6, 12, and 15 weeks after transplantation. As shown in Panel A, Patients MPD01 and MPD04 had TET2 defects, whereas Patients MPD09, MPD11, and MPD27 had normal TET2 and served as controls. Panel B shows the results of flow-cytometric analysis of human cells present in the bone marrow of NOD–SCID mice 15 weeks after transplantation with 3×105 CD34+ cells from Patients MPD04 and MPD09. The percentages of human CD45+ myeloid and lymphoid cells were determined with the use of anti–CD45-PC7, anti–CD33-APC, and anti–CD19-PE antibodies. Panel C shows the results of clonogenic methylcellulose assays performed on CD34+ cells before transplantation. The number of colony-forming units (D0 CFU) is shown. Fifteen weeks after transplantation, human cells were sorted from mouse bone marrow and analyzed in methylcellulose assays (W15 CFU) and long-term-culture–initiating cell (W15 LTC-IC) assays. JAK2 and TET2 sequence analyses were performed on progenitor-derived colonies. Bar graphs show the fraction of individual clones with both JAK2 and TET2 defects or with wild-type JAK2 and a TET2 defect. No clone with both wild-type JAK2 and wild-type TET2 was observed. Sequence traces show the presence of TET2 defects in pooled W15 CFU cells. The number of clones in each category is indicated on the bars. Asterisks indicate the positions of nucleotide changes.

Mouse bone marrow samples that contained human cells 15 weeks after transplantation were tested in vitro for their content of mature progenitors and long-term-culture–initiating cells, a surrogate assay for hematopoietic stem cells, and were analyzed for the presence of TET2 and JAK2 mutations. TET2 defects were found in all human long-term-culture–initiating cells and progenitors in the mice (Figure 4C).

Because only hematopoietic stem cells can sustain long-term hematopoietic reconstitution in NOD–SCID mice, these results are consistent with the occurrence of a TET2 mutation in a true hematopoietic stem cell. The proportion of progenitor cells carrying only the TET2 mutation increased at the expense of cells carrying both TET2 and JAK2 V617F mutations from baseline to 15 weeks after transplantation, which suggests that cells with a mutated TET2 can proliferate in vivo independently of the JAK2 V617F mutation (Figure 4C).


We report that deletions or mutations in TET2 are early events in some patients with myelodysplastic syndromes, myeloproliferative disorders, or secondary AML. The frequency of TET2 mutations in unselected patients was 19% (15 of 81 patients) with myelodysplastic syndromes, 12% (24 of 198 patients) with myeloproliferative disorders, 24% (5 of 21 patients) with secondary AML, and 22% (2 of 9 patients) with chronic myelomonocytic leukemia. The detection of acquired genetic defects targeting both TET2 copies in 24 of 55 patients indicates that TET2 is a bona fide tumor-suppressor gene in some myeloid cancers. In samples from patients with myeloproliferative disorders who had both TET2 and JAK2 mutations, TET2 mutations occurred first in the course of the disease. TET2 mutations were observed in various subtypes of myelodysplastic syndromes, in keeping with reported loss of heterozygosity and interstitial deletions on chromosome 4q24 in these subtypes.26,27 Similar events affecting chromosome 4q24 have been reported in patients with myeloproliferative disorders28 and those with relapsed AML.29

TET2 mutations were observed in primitive CD34+CD38– cells, and xenograft transplantation demonstrated the presence of mutant TET2 in hematopoietic stem cells in samples from patients with myeloproliferative disorders. In both myelodysplastic syndromes and myeloproliferative disorders, the TET2 mutation appeared to be associated with the amplification of the mutated clone at the early stages of hematopoietic differentiation.

TET2 defects were observed in both myelodysplastic syndromes and myeloproliferative disorders, which are two distinct myeloid diseases. The presence of clinical and biologic phenotypes characteristic of these diseases probably requires at least one additional contributing event, as exemplified by our finding that the JAK2 V617F or MPL W515L mutations, which are responsible for some biologic features of myeloproliferative disorders, were associated with TET2 mutations.

Our results support a role for wild-type TET2 in the control of balance between survival, growth, and differentiation in normal hematopoiesis. Elucidation of the biochemical pathway underlying the function of TET2 will require additional work and may help to define novel mechanisms of renewal and differentiation of hematopoietic stem cells.

Supported by grants from INSERM, La Ligue Nationale contre le Cancer, La Fondation de France, Institut National du Cancer, Canceropole Ile de France (Cellules Souches Cancéreuses and Apoptosis 2007), L’association Française contre les Myopathies, GIS-Institut des Maladies Rares, la Direction Régionale de la Recherche Clinique AP–HP, Ministère de la Recherche et de la Technologie (to Dr. Dupont), and the European Hematology Association (to Dr. Delhommeau).

No potential conflict of interest relevant to this article was reported.

We thank J. Prchal, S. Constantinescu, T. Mercher, F. Wendling, and E. Solary for their critical reading of the manuscript; D. Bouscary and I. Callebaut for their help and support; O. Beyne-Rauzy, C. Delacroix, C. Preudhomme, B. Quesnel, and E. Solary for the recruitment of patients with myelodysplastic syndromes; P.J. Weiller, L. Roy, J.-A. Ribeil, and E. Lippert for the recruitment of patients with myeloproliferative disorders; and I. Teyssandier, H. Bouamar, and C. Pierre-Eugene for the preparation of samples.
Source Information

From INSERM U790, Institut Gustave Roussy (F.D., S.D., C.J., A.M., J.-P.L.C., Y.L., I.P., N.C., P.D., W.V.) and Université Paris XI (F.D., S.D., A.M., J.-.P.L.C., N.C., P.D., W.V.), Villejuif; Hôpital St. Antoine, Paris (F.D., C.M., N.C.); Université Paris VI Pierre et Marie Curie, Paris (F.D., N.C.); INSERM E0210, Hôpital Necker, Paris (V.D.V., S.T., F.R., S.P.R., O.A.B.); Université René Descartes, Paris (V.D.V., O.K., F.J.D., C.L., S.P.R., F.V., M.F., O.A.B.); and Hopital Cochin, Paris; INSERM Unité 567, Centre National de la Recherche Scientifique UMR8104, Institut Cochin, Hôpital Cochin, Paris (O.K., F.J.D., C.L., M.F.); Assistance Publique–Hôpitaux de Paris, Laboratoire d’Hématologie (O.K., C.L., M.F., O.A.B.) and Unité Fonctionnelle d’Hématologie (F.J.D.), Hôpital Cochin, Paris; INSERM Unité 876, Université Bordeaux 2, Bordeaux (C.J.); and Université Paris Diderot, INSERM Unité MR944, Hôpital Saint-Louis, Paris (A.A., J.S.); — all in France.

Drs. Delhommeau, Dupont, Della Valle, and James contributed equally to this article.

Address reprint requests to Dr. Bernard at INSERM E0210, Hôpital Necker, 149 rue de Vaugirard, Paris 75015, or at olivier.bernard@inserm.fr.


  1. Tiu R, Gondek L, O’Keefe C, Maciejewski JP. Clonality of the stem cell compartment during evolution of myelodysplastic syndromes and other bone marrow failure syndromes. Leukemia 2007;21:1648-1657.
  2. Huntly BJ, Shigematsu H, Deguchi K, et al. MOZ-TIF2, but not BCR-ABL, confers properties of leukemic stem cells to committed murine hematopoietic progenitors. Cancer Cell 2004;6:587-596.
  3. Corey SJ, Minden MD, Barber DL, Kantarjian H, Wang JC, Schimmer AD. Myelodysplastic syndromes: the complexity of stem-cell diseases. Nat Rev Cancer 2007;7:118-129.
  4. Ebert BL, Pretz J, Bosco J, et al. Identification of RPS14 as a 5q- syndrome gene by RNA interference screen. Nature 2008;451:335-339.
  5. Braun T, Carvalho G, Coquelle A, et al. NF-kappaB constitutes a potential therapeutic target in high-risk myelodysplastic syndrome. Blood 2006;107:1156-1165.
  6. Nimer SD. Myelodysplastic syndromes. Blood 2008;111:4841-4851.
  7. Morgan KJ, Gilliland DG. A role for JAK2 mutations in myeloproliferative diseases. Annu Rev Med 2008;59:213-222.
  8. Delhommeau F, Pisani DF, James C, Casadevall N, Constantinescu S, Vainchenker W. Oncogenic mechanisms in myeloproliferative disorders. Cell Mol Life Sci 2006;63:2939-2953.
  9. Campbell PJ, Green AR. The myeloproliferative disorders. N Engl J Med 2006;355:2452-2466.
  10. Dupont S, Massé A, James C, et al. The JAK2 617V>F mutation triggers erythropoietin hypersensitivity and terminal erythroid amplification in primary cells from patients with polycythemia vera. Blood 2007;110:1013-1021.
  11. Villeval JL, James C, Pisani DF, Casadevall N, Vainchenker W. New insights into the pathogenesis of JAK2 V617F-positive myeloproliferative disorders and consequences for the management of patients. Semin Thromb Hemost 2006;32:341-351.
  12. Jamieson CH, Gotlib J, Durocher JA, et al. The JAK2 V617F mutation occurs in hematopoietic stem cells in polycythemia vera and predisposes toward erythroid differentiation. Proc Natl Acad Sci U S A 2006;103:6224-6229.
  13. Kralovics R, Teo SS, Li S, et al. Acquisition of the V617F mutation of JAK2 is a late genetic event in a subset of patients with myeloproliferative disorders. Blood 2006;108:1377-1380.
  14. James C, Mazurier F, Dupont S, et al. The hematopoietic stem cell compartment of JAK2V617F-positive myeloproliferative disorders is a reflection of disease heterogeneity. Blood 2008;112:2429-2438.
  15. Bellanné-Chantelot C, Chaumarel I, Labopin M, et al. Genetic and clinical implications of the Val617Phe JAK2 mutation in 72 families with myeloproliferative disorders. Blood 2006;108:346-352.
  16. Nussenzveig RH, Swierczek SI, Jelinek J, et al. Polycythemia vera is not initiated by JAK2V617F mutation. Exp Hematol 2007;35:32-38.
  17. Campbell PJ, Baxter EJ, Beer PA, et al. Mutation of JAK2 in the myeloproliferative disorders: timing, clonality studies, cytogenetic associations, and role in leukemic transformation. Blood 2006;108:3548-3555.
  18. Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002;100:2292-2302.
  19. Bennett JM, Catovsky D, Daniel MT, et al. Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French-American-British Cooperative Group. Ann Intern Med 1985;103:620-625.
  20. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997;89:2079-2088. [Erratum, Blood 1998;91:1100.]
  21. Viguié F, Aboura A, Bouscary D, et al. Common 4q24 deletion in four cases of hematopoietic malignancy: early stem cell involvement? Leukemia 2005;19:1411-1415.
  22. Kralovics R, Guan Y, Prchal JT. Acquired uniparental disomy of chromosome 9p is a frequent stem cell defect in polycythemia vera. Exp Hematol 2002;30:229-236.
  23. Ono R, Taki T, Taketani T, Taniwaki M, Kobayashi H, Hayashi Y. LCX, leukemia-associated protein with a CXXC domain, is fused to MLL in acute myeloid leukemia with trilineage dysplasia having t(10;11)(q22;q23). Cancer Res 2002;62:4075-4080.
  24. Lorsbach RB, Moore J, Mathew S, Raimondi SC, Mukatira ST, Downing JR. TET1, a member of a novel protein family, is fused to MLL in acute myeloid leukemia containing the t(10;11)(q22;q23). Leukemia 2003;17:637-641.
  25. Robert-Richard E, Ged C, Ortet J, et al. Human cell engraftment after busulfan or irradiation conditioning of NOD/SCID mice. Haematologica 2006;91:1384-1384.
  26. Mohamedali A, Gäken J, Twine NA, et al. Prevalence and prognostic significance of allelic imbalance by single nucleotide polymorphism analysis in low risk myelodysplastic syndromes. Blood 2007;110:3365-3373.
  27. Gondek LP, Tiu R, O’Keefe CL, Sekeres MA, Theil KS, Maciejewski JP. Chromosomal lesions and uniparental disomy detected by SNP arrays in MDS, MDS/MPD, and MDS-derived AML. Blood 2008;111:1534-1542.
  28. Gondek LP, Dunbar AJ, Szpurka H, McDevitt MA, Maciejewski JP. SNP array karyotyping allows for the detection of uniparental disomy and cryptic chromosomal abnormalities in MDS/MPD-U and MPD. PLoS ONE 2007;2:e1225-e1225.
  29. Raghavan M, Smith LL, Lillington DM, et al. Segmental uniparental disomy is a commonly acquired genetic event in relapsed acute myeloid leukemia. Blood 2008;112:814-821.

This article has been cited by other articles:

  • Levine, R. L., Carroll, M. (2009). A Common Genetic Mechanism in Malignant Bone Marrow Diseases. NEJM 360: 2355-2357 [Full Text]
  • (2009). Novel Mutations in Myeloid Malignancies. JWatch Oncology and Hematology 2009: 1-1 [Full Text]