Myeloproliferative neoplasms (MPN) are clonal hematopoietic stem cell (HSC) malignancies with increased expansion of myeloid lineages. JAK2 and MPL mutations are detected in MPN patients’ HSCs but their biological consequences appear to rather target the terminal myeloid differentiation than the early steps of hematopoiesis. Analysis of CD34+CD38− multipotent progenitors, CD34+CD38+ committed progenitors and mature cells, led us to identify two subsets of JAK2 V617F MPN at diagnosis with distinct kinetics of hematopoietic expansion. The first subset (85% of patients) is characterized by a late expansion of the malignant clone –i.e downstream the committed progenitor stage. In contrast, the second subset of patients (15%) has an early expansion of the clone, upstream the committed progenitor stage. The hallmark of this early expansion is a high percentage (>80%) of JAK2 V617F positive multipotent or committed progenitors, contrasting with low percentages (<50%) in other MPN patients (Dupont et al, Blood 2007).

We hypothesized that the second subset of patients had a pre-existent molecular defect able to promote the early expansion of the malignant clone. With high resolution SNP arrays (Affymetrix 500K) and CGH arrays (Agilent 244K), we compared malignant granulocyte or erythroblast DNA with paired non malignant lymphocyte DNA from five patients with such a phenotype. Three/5 harbored an acquired loss of heterozygosity (LOH) on the long arm of chromosome 4. In 2 patients, the LOH spanned a large region from the 4q22 band to the telomeric end, without copy number variation. In contrast, in the third patient the LOH was restricted to the 4q24 region and was due to a 325 kb microdeletion. This minimal candidate region contains only one single gene, TET2 (Ten-Eleven Translocation–2), which belongs to a family of three genes of unknown function. Sequencing of the coding region of TET2 in the two patients with large LOH revealed a mutation leading to a stop codon in exon 3 and a 9 nucleotide deletion in exon 6 resulting in the loss of 3 evolutionary-conserved residues. TET2 coding sequence was normal in the patient with the 325 kb deletion and in the samples from the 2 patients without 4q24 LOH. Analysis of lymphocyte DNA demonstrated that TET2 defects were acquired. To extend our results we sequenced TET2 in a series of 181 unselected JAK2 V617F MPN patients. We found TET2 deletions, frame shifts, stop codons or conserved amino-acid substitutions, in 25 MPN (3/10 primary myelofibrosis, 14/98 polycythemia vera, 8/73 essential thrombocythemia), resulting in an overall 14% frequency. In the majority of MPN patients our results suggest that the two copies are affected, indicating that TET2 behaves as a tumor suppressor gene. To determine whether TET2 inactivation was an early, pre-JAK2 V617F molecular event, we analyzed single clones grown from CD34+CD38− and CD34+CD38+ cells in five patients with TET2 mutations. We showed that TET2 defects target both multipotent and committed progenitors, some of them being TET2 mutated in the absence of JAK2 V617F. This indicates that TET2 inactivation is a pre-JAK2 V617F event in these five patients. Finally, TET2-mutated HSCs from MPN patients have an increased capacity to repopulate non obese diabetic-severe combined immunodeficient (NOD-SCID) mice, suggesting that TET2 regulates HSC properties.

In conclusion we have identified TET2 as a probable new tumor suppressor gene altered in 14% of MPN patients. This gene may have key functions in hematopoiesis and HSC biology.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution