Presence of atypical thrombopoietin receptor (MPL) mutations in triple-negative essential thrombocythemia patients.

Mutations in signaling molecules of the cytokine receptor axis play a central role in myeloproliferative neoplasm (MPN) pathogenesis. Polycythemia vera is mainly related to JAK2 mutations, whereas a wider mutational spectrum is detected in essential thrombocythemia (ET) with mutations in JAK2, the thrombopoietin (TPO) receptor (MPL), and the calreticulin (CALR) genes. Here, we studied the mutational profile of 17 ET patients negative for JAK2V617F, MPLW515K/L, and CALR mutations, using whole-exome sequencing and next-generation sequencing (NGS) targeted on JAK2 and MPL. We found several signaling mutations including JAK2V617F at very low allele frequency, 1 homozygous SH2B3 mutation, 1 MPLS505N, 1 MPLW515R, and 2 MPLS204P mutations. In the remaining patients, 4 presented a clonal and 7 a polyclonal hematopoiesis, suggesting that certain triple-negative ETs are not MPNs. NGS on 26 additional triple-negative ETs detected only 1 MPLY591N mutation. Functional studies on MPLS204P and MPLY591N revealed that they are weak gain-of-function mutants increasing MPL signaling and conferring either TPO hypersensitivity or independence to expressing cells, but with a low efficiency. Further studies should be performed to precisely determine the frequency of MPLS204 and MPLY591 mutants in a bigger cohort of MPN.

Different impact of calreticulin mutations on human hematopoiesis in myeloproliferative neoplasms.

Mutations of calreticulin (CALRm) define a subtype of myeloproliferative neoplasms (MPN). We studied the biological and genetic features of CALR-mutated essential thrombocythemia and myelofibrosis patients. In most cases, CALRm were found in granulocytes, monocytes, B and NK cells, but also in T cells. However, the type 1 CALRm spreads more easily than the type 2 CALRm in lymphoid cells. The CALRm were also associated with an early clonal dominance at the level of hematopoietic stem and progenitor cells (HSPC) with no significant increase during granulo/monocytic differentiation in most cases. Moreover, we found that half of type 2 CALRm patients harbors some homozygous progenitors. Those patients were associated with a higher clonal dominance during granulo/monocytic differentiation than patients with only heterozygous type 2 CALRm progenitors. When associated mutations were present, CALRm were the first genetic event suggesting that they are both the initiating and phenotypic event. In blood, type 1 CALRm led to a greater increased number of all types of progenitors compared with the type 2 CALRm. However, both types of CALRm induced an increase in megakaryocytic progenitors associated with a ruxolitinib-sensitive independent growth and with a mild constitutive signaling in megakaryocytes. At the transcriptional level, type 1 CALRm seems to deregulate more pathways than the type 2 CALRm in megakaryocytes. Altogether, our results show that CALRm modify both the HSPC and megakaryocyte biology with a stronger effect for type 1 than for type 2 CALRm.

TET2-mediated 5-hydroxymethylcytosine induces genetic instability and mutagenesis.

The family of Ten-Eleven Translocation (TET) proteins is implicated in the process of active DNA demethylation and thus in epigenetic regulation. TET 1, 2 and 3 proteins are oxygenases that can hydroxylate 5-methylcytosine (5-mC) into 5-hydroxymethylcytosine (5-hmC) and further oxidize 5-hmC into 5-formylcytosine (5-fC) and 5-carboxylcytosine (5-caC). The base excision repair (BER) pathway removes the resulting 5-fC and 5-caC bases paired with a guanine and replaces them with regular cytosine. The question arises whether active modification of 5-mC residues and their subsequent elimination could affect the genomic DNA stability. Here, we generated two inducible cell lines (Ba/F3-EPOR, and UT7) overexpressing wild-type or catalytically inactive human TET2 proteins. Wild-type TET2 induction resulted in an increased level of 5-hmC and a cell cycle defect in S phase associated with higher level of phosphorylated P53, chromosomal and centrosomal abnormalities. Furthermore, in a thymine-DNA glycosylase (Tdg) deficient context, the TET2-mediated increase of 5-hmC induces mutagenesis characterized by GC>AT transitions in CpG context suggesting a mutagenic potential of 5-hmC metabolites. Altogether, these data suggest that TET2 activity and the levels of 5-hmC and its derivatives should be tightly controlled to avoid genetic and chromosomal instabilities. Moreover, TET2-mediated active demethylation might be a very dangerous process if used to entirely demethylate the genome and might rather be used only at specific loci.

Myeloproliferative neoplasms: JAK2 signaling pathway as a central target for therapy.

The discovery of the JAK2V617F mutation followed by the discovery of other genetic abnormalities allowed important progress in the understanding of the pathogenesis and management of myeloproliferative neoplasms (MPN)s. Classical Breakpoint cluster region-Abelson (BCR-ABL)-negative neoplasms include 3 main disorders: essential thrombocythemia (ET), polycythemia vera (PV), and primary myelofibrosis (PMF). Genomic studies have shown that these disorders are more heterogeneous than previously thought with 3 main entities corresponding to different gene mutations: the JAK2 disorder, essentially due to JAK2V617F mutation, which includes nearly all PVs and a majority of ETs and PMFs with a continuum between these diseases and the myeloproliferative leukemia (MPL) and calreticulin (CALR) disorders, which include a fraction of ET and PMF. All of these mutations lead to a JAK2 constitutive activation. Murine models either with JAK2V617F or MPLW515L, but also with JAK2 or MPL germ line mutations found in hereditary thrombocytosis, have demonstrated that they are drivers of myeloproliferation. However, the myeloproliferative driver mutation is still unknown in approximately 15% of ET and PMF, but appears to also target the JAK/Signal Transducer and Activator of Transcription (STAT) pathway. However, other mutations in genes involved in epigenetics or splicing also can be present and can predate or follow mutations in signaling. They are involved either in clonal dominance or in phenotypic changes, more particularly in PMF. They can be associated with leukemic progression and might have an important prognostic value such as additional sex comb-like 1 mutations. Despite this heterogeneity, it is tempting to target JAK2 and its signaling for therapy. However in PMF, Adenosine Tri-Phosphate (ATP)-competitive JAK2 inhibitors have shown their interest, but also their important limitations. Thus, other approaches are required, which are discussed in this review.

Heterozygous and homozygous JAK2(V617F) states modeled by induced pluripotent stem cells from myeloproliferative neoplasm patients.

JAK2(V617F) is the predominant mutation in myeloproliferative neoplasms (MPN). Modeling MPN in a human context might be helpful for the screening of molecules targeting JAK2 and its intracellular signaling. We describe here the derivation of induced pluripotent stem (iPS) cell lines from 2 polycythemia vera patients carrying a heterozygous and a homozygous mutated JAK2(V617F), respectively. In the patient with homozygous JAK2(V617F), additional ASXL1 mutation and chromosome 20 allowed partial delineation of the clonal architecture and assignation of the cellular origin of the derived iPS cell lines. The marked difference in the response to erythropoietin (EPO) between homozygous and heterozygous cell lines correlated with the constitutive activation level of signaling pathways. Strikingly, heterozygous iPS cells showed thrombopoietin (TPO)-independent formation of megakaryocytic colonies, but not EPO-independent erythroid colony formation. JAK2, PI3K and HSP90 inhibitors were able to block spontaneous and EPO-induced growth of erythroid colonies from GPA(+)CD41(+) cells derived from iPS cells. Altogether, this study brings the proof of concept that iPS can be used for studying MPN pathogenesis, clonal architecture, and drug efficacy.