Deletion of the p53 Target Gene PUMA Prevents Bone Marrow Failure in a Dyskeratosis Congenita Mouse Model

Dyskeratosis congenita (DC) belongs to the group of inherited bone marrow failure syndromes (IBMFS) and is characterized by premature telomere shortening caused by mutations in components of the telomerase or the shelterin complexes. The main cause of death in affected patients is hematopoietic failure, but there is also a 10-15% risk of malignant transformation into secondary myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Critically short telomeres activate a DNA damage response with p53-mediated cell cycle inhibition, senescence and/or apoptosis, the latter mediated primarily by PUMA, a BCL-2 family member belonging to the group of pro-apoptotic BH3-only proteins and transcriptionally regulated by p53. Activation of p53 and expression of its target genes are critical for the exhaustion of hematopoietic stem cells (HSCs) in DC patients. Inactivation of the DNA damage checkpoint could possibly mitigate hematopoietic failure but poses a significant risk of genomic instability and leukemia. Based on our earlier mouse model of secondary leukemia (Genes Dev, 24(15):1602-7), we hypothesized that selective inhibition of p53-mediated apoptosis - while all other p53-checkpoint-induced pathways remain active - could both delay hematopoietic failure and prevent malignant transformation.

We established a DC mouse model by serial transplantation of hematopoietic stem and progenitor cells (HSPCs) derived from generation 3 mTerc^-/- (G3 mTerc^-/-) mice lacking the RNA telomerase component. While 8-12 week old donor mice and primary recipients had only a mild hematopoietic phenotype, secondary recipients demonstrated severe lymphopenia. 41% of secondary recipients died within 50 days after transplantation, and flow cytometric and histological analysis revealed pancytopenia and bone marrow aplasia. The surviving secondary recipients were analyzed 16 weeks after transplantation and displayed severely reduced HSPC viability ex vivo. Aiming to inhibit HSPC apoptosis in vivo, we deleted Puma in G3 mTerc^-/- mice. PUMA deficiency significantly rescued bone marrow numbers, HSPC viability ex vivo (72% vs. 50% viable HSPCs, p=0.01) and hematopoietic output on a G3 mTerc^-/-background. Most importantly, death of secondary recipients was fully prevented in the absence of PUMA. This rescue is associated with significantly longer telomeres and reduced levels of γH2AX foci in G3 mTerc^-/-Puma^-/- donor and recipient HSPCs when compared to their PUMA proficient counterparts. Notably, no signs of myelodysplasia or leukemia were found in mice receiving serial transplantations of G3 mTerc^-/-Puma^-/- bone marrow.

Our data indicate that specific inhibition of the intrinsic apoptosis pathway is sufficient to restrain the death of HSPCs with critically short telomeres and ensure blood formation. The resulting reduction of proliferative pressure within the HSPC compartment preserves functional and genetic integrity of HSCs and leads to generally longer telomeres in the HSPC pool. We anticipate that prevention of bone marrow failure is sufficient to prevent outgrowth of (pre)malignant clones and transformation to secondary MDS and AML.