Development of capsaicin-derived prohibitin ligands to modulate the Aurora kinase A/PHB2 interaction and mitophagy in cancer cells

Abstract

Aurora kinase A/AURKA is a serine/threonine kinase overexpressed in a variety of solid and hematological malignancies. In the last decades, clinical trials aiming to counteract the overexpression of AURKA turned out to be largely unsuccessful. Meanwhile, recent discoveries pointed to new functions of AURKA at the subcellular level, including at mitochondria. At this location, AURKA induces organelle clearance by mitophagy acting in complex with the mitophagy mediator LC3, and its inner mitochondrial membrane receptor PHB2. The natural polyphenol xanthohumol was shown to act as a PHB2 ligand, altering the interaction between AURKA and PHB2 and restoring mitochondrial functions in cancer cells. However, its chemical nature prevents its broader use as an anticancer agent. Using Forsters Resonance Energy Transfer/Fluorescence Lifetime Imaging Microscopy (FRET/FLIM) in live breast cancer cells, we here explore the effects of alternative PHB ligands in altering the proximity between AURKA and PHB2. Among the already-available compounds, we found that the pungent natural product capsaicin partially alters the AURKA/PHB2 protein-protein proximity. We then synthesized 16 novel capsaicin analogs to enhance the effects of capsaicin. We found that replacing the long hydrophobic acyl moiety with a butyryl one increases the AURKA/PHB2 interaction. Among the capsaicin derivatives carrying this modification, we uncover that compounds 12 and 13 enhance the AURKA/PHB2 proximity. Molecular docking approaches corroborate FRET/FLIM data, and we visualize compounds 12 and 13 in complex with AURKA, PHB2 and LC3. We show that compounds 12 and 13 stabilize the AURKA/PHB2 interaction, and that they can bind to the inhibitory pocket of PHB2 and to the AURKA active site. Finally, we report that compound 13 specifically inhibits AURKA-dependent mitophagy, while leaving the activation of AURKA unaltered at centrosomes. Together, our data demonstrate that compound 13 is a promising PHB ligand acting on the AURKA/PHB2 interaction. Thanks to its specificity toward the mitochondrial roles of AURKA, it may provide the basis for the development of new anticancer drugs targeting the mitochondrial functions of AURKA.