Biased ligand quantification in drug discovery: from theory to high throughput screening to identify new biased μ opioid receptor agonists.

Abstract

Biased GPCR ligands are able to engage with their target receptor in a manner that preferentially activates distinct downstream signalling and offers potential for next generation therapeutics. However, accurate quantification of ligand bias in vitro is complex, and current best practice is not amenable for testing large numbers of compound. We have therefore sought to apply ligand bias theory to an industrial scale screening campaign for the identification of new biased μ receptor agonists. μ receptor assays with appropriate dynamic range were developed for both Gαi -dependent signalling and β-arrestin2 recruitment. Δlog(Emax /EC50 ) analysis was validated as an alternative for the operational model of agonism in calculating pathway bias towards Gαi -dependent signalling. The analysis was applied to a high throughput screen to characterize the prevalence and nature of pathway bias among a diverse set of compounds with μ receptor agonist activity. A high throughput screening campaign yielded 440 hits with greater than 10-fold bias relative to DAMGO. To validate these results, we quantified pathway bias of a subset of hits using the operational model of agonism. The high degree of correlation across these biased hits confirmed that Δlog(Emax /EC50 ) was a suitable method for identifying genuine biased ligands within a large collection of diverse compounds. This work demonstrates that using Δlog(Emax /EC50 ), drug discovery can apply the concept of biased ligand quantification on a large scale and accelerate the deliberate discovery of novel therapeutics acting via this complex pharmacology.