CblQ367P Cooperates with Loss of Tet2 to Drive a Cnl-like Disease Via the NF-Kb Pathway

Recurrent mutations in genes encoding epigenetic regulators and signaling factors contribute to the development of myelodysplastic syndromes (MDS) and related myeloid neoplasms. Ten-Eleven-Translocation 2 (TET2) is an epigenetic regulator which is responsible for active DNA and RNA demethylation. Loss-of-function mutations of TET2 are recurrent events in MDS and myeloproliferative neoplasms (MPN). Casitas B-lineage Lymphoma (CBL) is an E3 ubiquitin ligase which ubiquitinates receptor tyrosine kinases and negatively regulates receptor tyrosine kinase signaling. Uniparental disomy of the long arm of chromosome 11 harboring mutant CBL allele is frequently observed in MDS, which functions as gain-of-function mutations that activate intracellular signaling. In MDS, CBL and TET2 mutations occur in 10% and 20% of patients respectively. Previous studies have shown that TET2 and CBL mutations significantly co-occur in MDS patients (Makishima et al. Nat Genet. 2017; Haferlach et al. Leukemia 2014).

To assess if Tet2 loss and Cbl mutation functionally cooperate to drive myeloid transformation, we crossed Tet2 conditional knockout mice (Mx1-Cre⁺ Tet2^flox/flox, Moran-Crusio et al. Cancer Cell 2011) and Cbl mutant mice (Mx1-Cre⁺ Cbl^Q367P, Nakata et al. Blood 2017) to generate mice with both disease alleles (Tet2^-/- Cbl^Q367P).

At 16-20 weeks post-recombination, Tet2^-/-Cbl^Q367P mice showed significant leukocytosis (mean at 20 weeks 64.4 x10³/μl , p < 0.02; WT 13.2 x10³/μl, Tet2^-/- 17.3 x10³/μl, Cbl^Q367P 32.5 x10³/μl) and expansion of CD11b⁺ Gr1⁺ mature neutrophils in both peripheral blood and bone marrow compared to mice with either disease allele alone. Subsequently, Tet2^-/-Cbl^Q367P mice developed massive splenomegaly and chronic neutrophilic leukemia (CNL)-like disease in vivo which was serially transplantable. This CNL-like disease further transformed into acute myeloid leukemia (AML) as evidenced by an increase in c-Kit⁺ compartment with blast cell-like morphology and death at six months after recombination. Importantly, the survival of Tet2^-/-Cbl^Q367P mice was significantly worse compared with WT mice or mice with either disease allele alone (median survival 207 days, p<0.03; WT, Tet2^-/-, and Cbl^Q367P >50 weeks) . CRISPR knockout of TET2 in CBL-mutant human AML cell line, MOLM13, also enhanced cell growth and expression of neutrophilic marker CD15. Together, these data suggest that TET2 and CBL mutations functionally cooperate to drive myeloid neoplasm.

Proteomic analyses of BM CD11b⁺ leukemia cells demonstrated that Tet2 loss and Cbl mutation cooperatively led to increases in activators of nuclear factor-kappa B (NFκB), including TGF-beta activated kinase 1 (Tak1), Inhibitor of NFκB kinase (IKK) and Inhibitor of NFκB (IκB). Western blot confirmed upregulated phosphorylation of IKK, IκB and downstream p65 in Tet2^-/-Cbl^Q367P cells compared to cells derived from WT or mice with either disease allele alone, indicating specific activation of NFκB signaling in Tet2^-/-Cbl^Q367P leukemia cells.

We next assessed whether the activated NFκB pathway is functional in Tet2^-/-Cbl^Q367P driven myeloid leukemogenesis. To this end, we performed in vitro colony forming unit assay with the clinical IKK and IκB inhibitors. Strikingly, Tet2^-/- Cbl^Q367P cells showed significantly higher sensitivity to IKK and IκB inhibitors compared to WT, Tet2^-/- or Cbl^Q367P cells in vitro. Moreover, in vivo treatment of Tet2^-/- Cbl^Q367P leukemic mice with IκB inhibitors reduced spleen size, significantly reduced disease burden in BM and spleen and improved overall survival compared to vehicle treatment (median survival 105 days vs. 144 days, p=0.04). Consistent with murine data, IκB inhibition significantly inhibited colony formation of TET2/CBL double-mutant human MDS/AML patient samples more potently than patient samples with either disease allele alone or neither of them.

Taken together, our model represents a significant step toward building high fidelity and genetically tractable models of CNL and provide insights into how epigenetic and signaling factor mutations cooperate in myeloid transformation and a rationale for mechanism-based therapy.