Juvenile myelomonocytic leukemia (JMML) and chronic myelomonocytic leukemia (CMML) are mixed myelodysplastic syndromes (MDS)/myeloproliferative neoplasms (MPNs) that share similar clinical manifestations, but differ in prognosis and ages at diagnosis. Both are characterized by monocytosis and monocytic infiltration in vital organs. JMML is extremely aggressive, with death occurring within one year after diagnosis in children (<6 Yrs) without therapy, whereas CMML presents in adults with a more indolent course. A dysregulated RAS pathway is the key mechanism of JMML and CMML, but the number of non-synonymous mutations per tumor in JMML patients is ranked among the least of all human cancers. Previously, we reported that mice with somatic Pten deletion and germline mutant Nf1 (PtenΔ/ΔNf1wt/-) mimic human MPNs with features of a pediatric mixed MDS/MPNs, in terms of age of onset and organ infiltration with myelomonocytic cells, but lack a key feature of JMML with respect to GM-CSF hypersensitivity (Blood 2016; 127:1912). In order to simulate the molecular dynamics observed in JMML, we introduced Nf1 Loss-Of-Heterozygosity (Nf1LOH) by somatically deleting the second copy of Nf1 in mice on postnatal day 8 (PND8, equivalent to a full-term newborn age in humans).We generated mice with PtenfloxP/floxPNf1Fcr/floxPMx1-Cre+on a C57BL6/129 genetic background, and conditionally deleted Pten and the second copy of Nf1 myeloid-specifically by intraperitoneal injection of Poly(I:C). Mice with bi-allelic Pten deletion and Nf1LOH (PtenΔ/ΔNf1LOH) were born healthy but revealed signs of sickness in the 2nd week of life, and all died by age 4 weeks (equivalent to 1-3 years old in humans); whereas mice with heterozygous Pten deletion and Nf1LOH (Pten wt/ΔNf1LOH) did not show obvious signs of disease until age 2 months, and possessed a lifespan longer than 3 months. The natural lifespan of PtenΔ/ΔNf1LOH mice (n=8) was significantly shorter than littermates with PtenΔ/ΔNf1wt/Δ (n=18), Pten wt/ΔNf1LOH(n=10), or wild type (ptenwtNf1wt, hereafter referred asWT, n=14, p<0.001). We analyzed 13 PtenΔ/ΔNf1LOH mice at age PND18-21 along withtheir littermates. Substantial hepatosplenomegaly was observed in mice with Pten deletion and correlated with the loss of copy number of Pten and Nf1 inthe following order: PtenΔ/ΔNf1LOH> PtenΔ/ΔNf1wt/Δ > Pten wt/ΔNf1LOH> WT (median spleen weight: 228>180>134>56mg, respectively, p<0.001). Mice with PtenΔ/ΔNf1LOH(n=13) or Pten wt/ΔNf1LOH(n=12) had significantly elevated white blood cells (WBCs) and a lower hemoglobin than PtenΔ/ΔNf1wt/Δ (n=21) or WT littermates (n=15). HE stained tissue sections of formalin-fixed organs revealed increased cellularity in bone marrow (BM), and severe infiltration of mature myeloid cells in the spleens, livers, and lungs of mice with PtenΔ/ΔNf1LOH. Flow cytometry analysis (FACS) revealed that PtenΔ/ΔNf1LOHmice had significantly increased monocyte/macrophages and granulocytes in BM and peripheral blood (PB), but significant reduction in T-cells and B-cells in PB and spleen. A significant increase in Ter119+CD71high cells was also found in PtenΔ/ΔNf1LOH mice, indicating MDS or leukemia. Interestingly, splenic marginal zone B-cells were significantly lower exclusively in PtenΔ/ΔNf1LOHmice. Mice with Ptenwt/ΔNf1LOHhad the similar FACS profile with less disease severity at age PND18-21. They lived longer than 3 months without obvious signs of disease except increased WBCs and mild anemia. BM cells from PtenΔ/ΔNf1LOH mice at an age of 3 weeks were hypersensitive to GM-CSF, but Ptenwt/ΔNf1LOH mice did not display this until an age of 2 months. Based on the Bethesda proposals for classification of non-lymphoid hematopoietic neoplasms in mice, we have developed a mouse model of a mixed MDS/MPNs with features of JMML and CMML, in terms of age of onset and organ infiltration with myelomonocytic cells, and GM-CSF hypersensitivity.

In conclusion, our data demonstrate that bi-allelic Pten deletion and Nf1LOH at an early age results in JMML in mice, whereas single-allelic Pten deletion with Nf1LOH induces CMML. This is the first mouse model that differentially resembles pediatric and adult mixed MDS/MPNs, and closely mimics the true genetic and epigenetic dynamics in humans. It provides a novel tool for studying the mechanisms underlying pediatric and adult mixed MDS/MPNs.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution