Expression of PML-RARα by the Murine PML Locus Leads to Myeloid Self-Renewal, Clonal Expansion and Morphologic Promyelocytic Leukemia.

Acute promyelocytic leukemia (APL) is characterized by the t(15;17) translocation that leads to expression of a fusion protein, PML-RARα, and haploinsufficiency for both RARα and PML. We have generated a novel murine model of APL using homologous recombination to place a pathogenic human PML-RARα cDNA into the murine PML locus (mPML-PRflox). Expression of PML-RARα is initially prevented by stop codons within the loxP-flanked PGK-neo cassette. This new model recapitulates key elements of human APL lacking in other models; it provides PML locus appropriate regulation of PML-RARα expression, haploinsufficiency of PML, and somatic acquisition of the fusion protein.

We have exposed mPML-PRflox mice to a conditionally-active Cre transgene (ER-T2- Cre, which activates Cre only during Tamoxifen treatment), and a compartmentally restricted Cre transgene (Lysozyme M-Cre, which has low activity in early myeloid precursors and increasing activity with myeloid differentiation). Bone marrow and spleen cells doubly heterozygous (DH) for mPML-PRflox and an activated Cre allele express PML-RARα mRNA, display neutrophil POD disruption typical of PML-RARα activity, and exhibit a shift in myelopoiesis toward CFU-G formation. We found that expression of PML-RARα following transient ER-T2-Cre activation leads to myeloid self-renewal ex vivo and clonal expansion in vivo. Bone marrow cells from DH mPML-PRflox/ER-T2- Cre mice exposed to Tamoxifen could be serially replated in methylcellulose. With successive replating, the proportion of cells carrying a floxed PML-RARα allele increased from 30% in bone marrow cells to 95% following the third replating; DH mPML-PRflox/ ER-T2-Cre bone marrow cells that were not exposed to Tamoxifen, and bone marrow cells from wild type mice, could not be serially replated. Cells bearing a floxed PML-RARα allele also expanded in vivo. In DH mPML-PRflox/ER-T2-Cre mice, a single dose of Tamoxifen (4 mg) resulted in 5% of peripheral blood cells carrying a floxed PML-RARα allele on day 7, but this population expanded progressively to 40% on day 80 without further Tamoxifen exposure (n=5). Five doses of Tamoxifen (4 mg) lead to 20% peripheral blood cells caring a floxed PML-RARα allele on day 18, and this increased progressively to 80% on day 98 (n=4). Peripheral blood of DH mPML-PRflox/ER-T2-Cre mice unexposed to Tamoxifen carried undetectable or trace numbers of cells with a floxed PML-RARα allele on day 80 (n=3). In contrast, myeloid progenitors from DH mPMLPRflox/ Lysozyme M-Cre mice did not display self-renewal ex vivo or an expansion of floxed cells with successive methylcellulose replating. DH mPML-PRflox/Lysozyme M-Cre mice did develop promyelocytic leukemia with long latency (14 months) and low penetrance (7%, n=3), which could be transplanted to secondary recipients. These tumors displayed a high percentage of floxed PML-RARα alleles in peripheral blood, spleen and bone marrow cells (range 50 – 95%) and possessed 4.5 fold higher expression levels of PML-RARα mRNA than our previously characterized Cathepsin G PML-RARα knock-in tumors. Importantly, non-leukemic,18 month-old DH mPML-PRflox/Lysozyme M-Cre mice displayed little evidence of PML-RARα dependent clonal expansion. In these mice, peripheral blood cells and spleen cells maintained low levels of the floxed PML-RARα allele (10–20%), equivalent to 6 week-old mice (n=36).

These data suggest that PML-RARα expression by the murine PML locus leads directly to a myeloid self-renewal program and clonal expansion. Since Lysozyme M-Cre is expressed at low levels in early myeloid progenitors, the low penetrance of leukemia and rare clonal expansion in DH mPML-PRflox/Lysozyme M-Cre animals suggests that APL leukemogenesis may require PML-RARα expression in an early myeloid progenitor compartment, rather than a late compartment.