Targeting challenging proteins from integrin family

Utilizing Receptor.AI’s pocket detection workflow to target the ‘undruggable’ integrin family

~5M

compounds
screened

hit rate with
97 found

binding pockets
targeted

potency of 32 best compounds

Kd of one of the top compounds

Targets are heterodimeric receptors from the integrin family involved in protein-protein interactions (PPIs).
They are promising treatment targets for
‍immunological disorders, inflammation, and cancers.
At the time we started our exploration there were still
‍no binding sites for small molecules known.
Also, no potent small molecule binders were known.
Considering the above, targets are truly challenging, often even deemed undruggable by small molecules.

We predicted integrin conformational changes using integrin
crystal structures in AI-driven MD simulations.
Identified binding pockets within dynamic conformations by our validated proprietary tool AI-driven Pocket Extractor.
Screened approximately 5M compounds with proprietary AI algorithms to evaluate potential interactions.
Refined the selection using AI docking tool ArtiDock.
To validate top candidates’ ability to compete with a known ligand for binding we conducted in vitro ligand displacement assays.
To measure binding kinetics and affinity and provide detailed insights into the strength of the interactions we used SPR analysis.

Achieved improved hit rate and validation, identifying 97 promising hits out of ~1,200 compounds - resulting in a 12% hit rate.
Identified 4 binding pockets, including one cryptic.
Successfully targeted 3 pockets - Pocket I (41 hits), Pocket II
(40 hits), and Pocket III (16 hits).
Top 32 compounds with potency within the range of 1-10 μM.
SPR show the top hit has strong binding affinity (Kd of 0.35 μM).

*Pocket I visualization

*Pocket II visualization

*Pocket III visualization