Establishment of a Retroviral Mouse Model for inv(16)-Mediated Acute Myeloid Leukemia Suggests That the p14^ARF Tumor Suppressor Is a Transcriptional Target for Repression by the inv(16) Fusion Protein.

The inv(16) is one of the most frequent chromosomal translocations associated with acute myeloid leukemia. The chimeric gene encodes a fusion protein consisting of most of the RUNX1 co-factor core binding factor beta (CBFβ) fused to a smooth muscle myosin heavy chain, SMMHC. Given that RUNX1 and RUNX1-ETO regulate the p14^ARF promoter, we tested whether the inv(16) fusion protein (IFP) can also repress p14^ARF expression. In reporter assays, the IFP cooperated with RUNX1 to repress the p14^ARF promoter. To investigate whether the IFP can regulate endogenous ARF expression, we triggered the oncogene checkpoint and the expression of p19^ARF (the murine homologue) by expressing c-Myc in primary myeloid progenitor cells. After induction of c-Myc activity, p19^ARF levels increase 20–50 fold; however, the co-expression of the IFP and c-Myc effectively blunted the induction of p19^ARF. Next, we sought to take advantage of this information to establish a murine model of the inv(16). We reasoned that the IFP provides a growth advantage and perhaps allows mutations to accumulate by repressing p19^ARF. If so, then removing p19^ARF during the rapid growth of hematopoietic stem cells and progenitors during embryogenesis might allow the formation of the same mutations that would occur after expression of the IFP, but accelerate leukemogenesis upon IFP expression. Therefore, we used p19^ARF-null mice as bone marrow donors and used recombinant retroviruses to express the IFP together with a green fluorescent protein (GFP). Given that repression of the p19^ARF promoter is likely to be incomplete, we also tested whether removing one allele of p19^ARF might enhance the ability of the IFP to induce leukemia by further reducing the expression of p19^ARF. We found that expression of the IFP induced a myelomonocytic AML in 10 to 12 weeks when expressed in p19^ARF-null cells. In addition, the IFP cooperated with the removal of only a single allele of p19^ARF and the complete deletion of p19^ARF only modestly shortened the latency period. The leukemic blasts in all the cases showed large nuclei and many of them contained granules and/or vacuoles in the cytoplasm. In addition, the GFP+ blast cells that express the IFP did not express any of the standard lineage markers and the majority of these cells were ckit+/Sca1−. All these findings are consistent with a myeloblastic acute leukemia. We conclude that repression of p19^ARF by the IFP may stimulate formation of AML and that the IFP sets the stage for the rapid development of AML.