Repurposing of cefpodoxime proxetil as potent neuroprotective agent through computational prediction and in vitro validation

Abstract

In recent reports, NR2B-NMDA receptor antagonists showed more research value because of its strong targeting ability and less side effects potential. In 2016, EVT-101 was reported to bind in an almost entirely new binding region of this target. Whether strikingly different binding modes can improve targeting and reduce side effects is worth studying. In our preliminary work, we explored the binding patterns of ifenprodil and EVT-101, found the key amino acids and summarized the pharmacophores, hoping to find such antagonists that target the two binding modes simultaneously. In this study, we developed a scalable virtual screening workflow in the FDA-approved drugs library to identify novel NR2B-NMDAR antagonists based on the combination of two pharmacophores. Cefpodoxime proxetil (5) was identified as the hit compound, and it was found for the first time that 5 might have neuroprotective activity as a NR2B-NMDAR antagonist. This result interested us to make further study, the ligand-receptor interactions modeled by molecular docking studies showed that the compound could perfectly merge both the pharmacophore characteristics of ifenprodil and EVT-101 at the binding cavity between the ATDs of GluN1 and GluN2B. The accuracy of molecular docking results and binding stability of ligand-receptor complexes were validated through 100 ns molecular dynamics simulation and binding free energy calculation. Afterwards, MTT assay (49.8%±0.1%, 5 μM) on NMDA injured SH-SY5Y cells and evidence of the effect on attenuating Ca²⁺ influx induced by NMDA were applied to validate the computational results, further investigation showed that 5 could suppress the NR2B upregulation induced by NMDA.