Critical Role of Stat5 in the Maintenance of Leukemic Stem Cells in Ptpn11-Associated Juvenile Myelomonocytic Leukemia

Effective therapeutic interventions for juvenile myelomonocytic leukemia (JMML), a fatal childhood malignancy, are lacking. Relapse is the most frequent cause of treatment failure, most likely due to the persistence of leukemic stem cells (LSCs), a small population of self-renewing precursor cells that give rise to the bulk of tumor cells. This reservoir of tumor cells are responsible for long-term maintenance of leukemia growth, and are also a major source of drug resistance. Clearly, a novel approach focused on the unique properties of LSCs is needed. However, it remains a critical challenge how such cells may be eradicated. JMML is known to be caused by genetic mutations in cell signaling proteins involved in the Ras pathway, among which the phosphatase Ptpn11 (Shp2) is most frequently mutated. Ptpn11 mutations (heterozygous) cause greatly increased catalytic activity. We have recently created conditional knock-in mice with the Ptpn11^E76K mutation, the most common Ptpn11 mutation found in JMML. Induction of the Ptpn11^E76K/+ mutation in these mice (Ptpn11^E76K/+/Mx1-Cre⁺) leads to JMML-like myeloid malignancy with full penetrance by aberrant activation of stem cells and myeloid progenitors. In an effort to understand the biological properties of LSCs in JMML, we studied leukemic hematopoietic stem cells (Lin^-Sca-1⁺c-Kit⁺Flk2^-CD150⁺CD48^- cells)(referred to as LSCs since they reproduce the same disease in sub-lethally-irradiated transplants) in this mouse model. We found that Stat5 was hyper-activated in LSCs in Ptpn11^E76K/+/Mx1-Cre⁺ mice in the chronic phase. Stat5 hyperactivation is likely to be mediated through Jak2 kinase as Jak2 is highly activated in Shp2 E76K-expressing cells. More importantly, these LSCs appear to rely on hyperactivation of Stat5 for maintenance and self-renewal because deletion of Stat5ab in Ptpn11^E76K/+/Mx1-Cre⁺/Stat5ab^fl/fl double mutant mice resulted in massive cell death in LSCs while neither Stat5ab knockout nor Ptpn11^E76K/+ knock-in alone mice had these stem cell phenotypes. Apoptotic cells in the stem cell population were 6.38±1.42, 4.06±0.51, 10.52±6.88, and 24.50±10.27% in Ptpn11^+/+/Mx1-Cre⁺, Ptpn11^E76K/+/Mx1-Cre⁺, Stat5ab^fl/fl/Mx1-Cre⁺, and Ptpn11^E76K/+/Mx1-Cre⁺/Stat5ab^fl/fl mice, respectively. Consequently, the stem cell pool in Ptpn11^E76K/+/Mx1-Cre⁺/Stat5ab^fl/fl double mutants was drastically decreased. Numbers of stem cells per femur were 3.36±1.51, 0.94±0.63, 1.15±0.39, and 0.17±0.20 (x10³) in Ptpn11^+/+/Mx1-Cre⁺, Ptpn11^E76K/+/Mx1-Cre⁺, Stat5ab^fl/fl/Mx1-Cre⁺, and Ptpn11^E76K/+/Mx1-Cre⁺/Stat5ab^fl/fl mice, respectively. Ptpn11^E76K/+/Mx1-Cre⁺/Stat5ab^fl/fl mice died of pan cytopenia within 4-8 weeks of Stat5 deletion while none of other groups of mice did. Repopulation capabilities of the double mutant stem cells were essentially lost in sub-lethally-irradiated recipient animals. Further mechanistic investigations revealed that tyrosine phosphorylation levels (indicative of activity) of Stat3, one of the substrates of the Shp2 phosphatase, were decreased by ~7.50 fold in Ptpn11^E76K/+ LSCs as compared to Ptpn11^+/+ control cells due to the enhanced dephosphorylation by the hyperactive Shp2 E76K mutant. Thus, diminished Stat3 activity sensitized Ptpn11^E76K/+ LSCs to Stat5 depletion-induced cell death. Taken together, this study suggests a crucial role of Stat5 in the maintenance of LSCs in Ptpn11-associated JMML. The synthetic lethality induced by loss of Stat5 in Ptpn11-mutated JMML raises the possibility that clinically-used inhibitors of upstream Jak2 kinase may be effective in eradicating LSCs in this particular subtype of JMML.