Core binding factor (CBF) is a heterodimeric transcription factor composed of RUNX1 (CBFα) and CBFβ subunits which are essential for normal blood cell development. CBFβ functions to increase the DNA-binding of the RUNX1 subunit 20–40 fold and to protect the RUNX1 subunit against ubiqitination and proteasome degradation, making this protein-protein interaction critical for CBF function. Two of the most common translocations involving the subunits of CBF are the inv(16) and the t(8;21) which produce the chimeric proteins CBFβ-SMMHC and AML1-ETO, respectively, which are associated with the development of Acute Myeloid Leukemia (AML). The AML1-ETO fusion protein is a dominant inhibitor of wildtype RUNX1-CBFβ activity in vivo and causes a blockage in normal hematopoiesis, predisposing for the development of leukemia. The interaction between CBFβ and AML1-ETO is critical for its function, therefore treatments targeting AML1-ETO and blocking its interaction with CBFβ are highly likely to be therapeutically beneficial. The CBFβ-SMMHC fusion protein causes dysregulation of CBF function by means of anomalously tight binding to RUNX1. Since binding to RUNX1 is required for the dysfunction associated with CBFβ-SMMHC, this interaction represents an excellent target for inhibition as a potential therapeutic strategy. We have initiated efforts to develop small molecule inhibitors of the RUNX1-CBFβ interaction as possible therapeutics for the treatment of the associated leukemias. Both virtual screening searches, focused on the X-ray structures of RUNX1 Runt domain and CBFβ, and high-throughput screening of NCI (National Cancer Institute) and Maybridge fragment libraries were used to identify initial lead compounds interacting with these proteins and blocking heterodimerization of CBF. Compounds were tested experimentally by FRET (Fluorescence Resonance Energy Transfer) and ELISA for their inhibition of RUNX1-CBFβ interaction. This resulted in a number of initial lead compounds targeting either the Runt domain or CBFβ and inhibiting this protein-protein interaction. Based on the docking mode selected lead compounds were further optimized using medicinal chemistry approaches to increase their affinity and determine the structure-activity relationships (SAR). This resulted in several compounds with low micromolar affinity (IC50 < 10 μM) which effectively block the heterodimerization of CBF in vitro and in a cell-based assay. Interestingly, compounds targeting CBFβ bind to a site displaced from the binding interface for RUNX1 as shown by the NMR-based docking, i.e. these compounds function as allosteric inhibitors of this protein-protein interaction. The most potent compounds were tested either in the Kasumi-1 leukemia cell line harboring t(8;21) translocation or in the ME-1 cell line with inv(16), resulting in a blockage of proliferation, induction of apoptosis and differentiation of these cells. These compounds represent the first small molecule inhibitors targeting CBF and inhibiting this interaction. They represent good starting points for the development of therapeutically useful inhibitors. Several approaches are being explored to modify these compounds to achieve selectivity towards AML1-ETO or CBFβ-SMMHC oncoproteins versus wild type proteins.

Author notes

Disclosure: No relevant conflicts of interest to declare.