Facile identification of dual FLT3-Aurora A inhibitors: A computer-guided drug design approach

Computer-guided drug design is a powerful tool for drug discovery. Herein we disclose the use of this approach for the discovery of dual FMS-like receptor tyrosine kinase-3 (FLT3)-Aurora A inhibitors against cancer. An Aurora hit compound was selected as a starting point, from which 288 virtual molecules were screened. Subsequently, some of these were synthesized and evaluated for their capacity to inhibit FLT3 and Aurora kinase A. To further enhance FLT3 inhibition, structure-activity relationship studies of the lead compound were conducted through a simplification strategy and bioisosteric replacement, followed by the use of computer-guided drug design to prioritize molecules bearing a variety of different terminal groups in terms of favorable binding energy. Selected compounds were then synthesized, and their bioactivity was evaluated. Of these, one novel inhibitor was found to exhibit excellent inhibition of FLT3 and Aurora kinase A and exert a dramatic antiproliferative effect on MOLM-13 and MV4-11 cells, with an IC₅₀ value of 7 nM. Accordingly, it is considered a highly promising candidate for further development. In silico selection: We selected an Aurora hit compound as a starting point, followed by two consecutive computer-guided strategies to rapidly and efficiently modify the side chain and core. These efforts resulted in the identification of a potential FLT3-Aurora A inhibitor for further development to treat acute myeloid leukemia (AML).