Abstract

Core binding factor (CBF) is a heterodimeric transcription factor composed of RUNX1 (CBFa) and CBFb subunits which are essential for normal blood cell development. RUNX1 binds DNA and the affinity of this interaction is regulated by the binding of CBFb to the RUNX1 subunit. The RUNX1(AML1) gene is disrupted by a variety of chromosomal translocations, including t(8;21)(q22;q22), which produces the chimeric protein AML1-ETO associated with AML (M2 subtype). The AML1-ETO fusion protein is a dominant inhibitor of wildtype RUNX1-CBFb activity in vivo and functions as a transcriptional repressor of RUNX1-CBFb target genes, causing a blockage in normal hematopoietic development and predisposing for the development of leukemia. We have validated the importance of the RUNX1 Runt domain - CBFb interface as a potential therapeutic target by assessing the effect of mutations introduced into the Runt domain of AML1-ETO on AML1-ETO activity. Therefore, treatments targeting the AML1-ETO protein and blocking its interactions with CBFb are likely to be therapeutically beneficial. We initiated efforts to develop small molecule inhibitors of the RUNX1-CBFb interaction as possible therapeutics for the treatment of the associated leukemia. Based on the X-ray structures of the RUNX1 Runt domain in complex with CBFb and on mutagenesis data, the interaction areas for the virtual screening searches were defined. The computer program LUDI was utilized for virtual screening using CAP (Chemicals Available for Purchase) database of 70,000 commercially available compounds with drug-like properties. The resultant hits were subjected to further scoring and visual inspection of the potential interactions with the RUNX1 Runt domain to select compounds for experimental testing by FRET (Fluorescence Resonance Energy Transfer). Experimental screening resulted in identification of six initial lead compounds that inhibit the interaction of RUNX1 with CBFb. These compounds were further optimized using standard medicinal chemistry approaches to increase their affinity and determine the structure-activity relationship (SAR). This resulted in a number of compounds with low micromolar affinity which effectively block the heterodimerization of CBF. These compounds represent the first small molecule inhibitors targeting RUNX1 and inhibiting its interaction with CBFb. They represent a good starting point for the development of a therapeutically useful inhibitor. Several approaches are being explored to modify these compounds to achieve selectivity towards AML1-ETO versus wild type RUNX1.

Author notes

Corresponding author