AI-ACCELERATED DRUG DISCOVERY

ATP synthase of Acinetobacter Baumannii

A novel approach to overcome multidrug resistance in Acinetobacter baumannii

AI revolutionizes the drug discovery process

Background

  • The ESKAPE bacteria (E. faecium, S. aureus, K. pneumoniae, A. baumannii, P. aeruginosa, and Enterobacter spp.) are responsible for threatening hospital infections.
  • Our goal is to develop inhibitors for A. baumannii ATP synthase, which would counter the drug resistance mechanism.

Methodology

  • The crystal structure of A. baumannii F0 complex of ATP synthase was used for virtual screening.
  • The binding pockets were deduced from literature data at the a/c subunit interfaces.  
  • There are “lagging” and “leading” pockets named by their relative position during the protein functioning cycle.
  • The Drug-Target Interaction (DTI) model is used to select top 10% of compounds from the pre-processed 3,8M compounds library using the smart consensus function type.
  • The top 50K compounds are subject to molecular docking with AI rescoring into each of the binding pockets independently.
  • 122 final hit candidates were selected.
  • Bacterial growth assay was used to test the potency of selected compounds.
  • The antimicrobial effect of inhibitors (IC50) was determined.
Project workflow
Project workflow

Results

  • 11 compounds out of 122 hit candidates were selected to start laboratory studies.
  • 2 compounds were validated as hits.
  • The compound R00439183 demonstrated the IC50 of 12.5 M whereas compound R00676319 demonstrated the IC50 of 4 M.
Inhibition of A. baumannii culture growth by R00439183 (left) and by R00676319 (right)
Inhibition of A. baumannii culture growth by R00439183 (left) and by R00676319 (right)
Dose-Response relationship for Compound 1, Compound 2 and Competitor Compound
Dose-Response relationship for Compound 1, Compound 2 and Competitor Compound