Development of a De Novo Protein Binder that Inhibits the Alpha Kinase eEF2K

Abstract

Elongation factor 2 kinase (eEF2K) is calmodulin activated and phosphorylates eEF2, a GTPase, that regulates global translation. When eEF2K phosphorylates eEF2, protein translation is halted. This process may be critical to studying how diseases like cancer dysregulate protein synthesis. eEF2K is an alpha kinase and not targeted by conventional kinase inhibitors. Traditional methods of structure-based drug design are incredibly time consuming and expensive, which may involve screening large libraries of small molecules. We have generated de novo small binder proteins ([~]10kDa) - using RFDiffusion and ProteinMPNN. One promising de novo binder protein we produced, CAM1 binds to a hydrophobic patch on the calmodulin binding domain of eEF2K with nanomolar affinity as determined by isothermal titration calorimetry. This binder, in vitro, significantly reduces eEF2 peptide phosphorylation, comparable to the gold-standard small molecule eEF2K inhibitor, A-484954. The predicted structure of CAM1 is a helical bundle which has been confirmed by circular dichroism spectroscopy. Impressively, CAM1 has a melting temperature >800C, and is produced recombinantly in bacteria, greater than 5 mg / culture liter. We have also determined that CAM1 transfection significantly reduces mammalian HeLa cell proliferation comparable to A-484954 treatment and inhibits the phosphorylation of eEF2. Our de novo binder, the first to our knowledge to inhibit an alpha kinase, and the first non-competitive eEF2K inhibitor, establishes an alternative method of targeting atypical kinase activity.