Molecular basis of the inhibitory effects of ATRA on osteoblast-induced MDS/AML

Osteoblasts are critical components of the hematopoietic stem cell (HSC) niche that regulate hematopoiesis. More recently, they have emerged as critical regulators of the development of hematological myeloid malignancies. We showed that a single activating mutation in ?-catenin signaling in osteoblasts is sufficient to lead to the development of MDS, rapidly progressing to AML in mice. The disease is transplantable and characterized by clonal evolution at the cytogenetic level. Activated ?-catenin signaling is present in osteoblasts of 38% of MDS and AML patients suggesting that this pathway may sustain dysplastic hematopoiesis and progression to MDS and AML in humans. Our initial observations support this indication and further suggest a novel means for treating this particular population of patients. In search of a potential FDA-approved compound with the ability to inhibit ?-catenin signaling we came across all-trans-retinoic acid (ATRA). ATRA is used in the treatment of acute promyelocytic leukemia where its mechanism of action relies on its ability to dissociate the NCOR-HDACL complex from RAR and allow DNA transcription and differentiation of the immature leukemic promyelocytes into mature granulocytes. However, reports from in vitro studies indicate that ATRA has another function: it inhibits ?-catenin functions. We have found that inhibition of ?-catenin signaling in 14 MDS/AML patients with active ?-catenin in their osteoblasts with ATRA improved their hematologic phenotype, stabilized disease status and inhibited ?-catenin activity. It also treated MDS and prevented AML in leukemic mice expressing constitutive active ?-catenin in osteoblasts. Based on these observations, we hypothesize that interrupting ?-catenin signaling in osteoblasts of mouse models and MDS/AML patients with active ?-catenin in their osteoblasts by pharmacological means will improve disease outcome. This may be achieved with ATRA, which may find a new use in the treatment of the portion of MDS/AML patients with activated ?-catenin in their osteoblasts. To test this hypothesis we will examine whether ATRA inhibits ?-catenin-induced AML on a mouse model of activated ?-catenin in osteoblasts; and whether this inhibition is independent of actions on HSCs. We will also dissect the molecular mechanism of ?-catenin inhibition by ATRA; and, verify the significance and specificity of ATRA inhibition in cytogenetically different types of human MDS and AML with activated ?-catenin in osteoblasts in vitro and in xenograft models we developed to examine interactions between human MDS/AML and stromal cells.