COMPUTATIONALLY AIDED RATIONAL DESIGN, SYNTHESIS AND EVALUATION OF PFKFB3 LIGANDS FOR ATHEROSCLEROTIC PLAQUE STABILISATION

Several types of cells actively involved in atherosclerosis undergo a metabolic reprogramming that comprises an accelerated glycolytic flux to the detriment of mitochondrial oxidative phosphorylation. This phenomenon, known as the Warburg effect, allows cells to produce ample energy and biomass with negligible oxygen consumption, resulting in uncontrolled high-speed proliferation. This phenotypic dysfunction underlies both inflammation and angiogenesis, two processes that are foundation of the pathological behaviour of the plaque and promote its destabilisation. The preponderant kinase domain of inducible PFKFB3 enzyme catalyses the synthesis of F2,6P, which simultaneously promotes glycolysis and inhibits gluconeogenesis. PFKFB3 has been found to be one of the most abundant and overly expressed isoenzymes in the Warburg effect, suggesting its key role in the pathology. From a clinical perspective, PFKFB3 represents an emerging biological target, and its inhibition would be an innovative therapeutic strategy for the treatment of atherosclerotic lesions. Our study revolves around the identification of new PFKFB3 kinase inhibitors. The rational design was carried out with the aid of computational methods, based on the 3D-structures of both PFKFB3 protein and known active ligands. In particular, two classes of inhibitors were used as reference ligands for the development of two distinct virtual screening workflows that led to the identification of two generations of candidate inhibitors. The first hierarchical workflow consisted of a pharmacophore-based library filtration, followed by molecular docking and molecular dynamic simulations, whereas the second consisted of a ligand- optimisation process based on the design of a focussed synthesisable library of analogues that were submitted to molecular docking along with a concomitant similarity search for scaffold hopping. The first- and the second-generation candidate compounds, which were predicted to have good in silico affinity towards PFKFB3, were synthesised and biologically evaluated through kinase assay, providing further insights about these selected drug-like molecules and new structure-activity relationship information.