Rational use of in vitro P-glycoprotein assays in drug discovery

P-glycoprotein (Pgp) affects the absorption, distribution, and clearance of a variety of compounds. Thus, identification of compounds that are Pgp sub strates can aid drug candidate selection and optimization. Our goal was to evaluate three assays used to determine whether compounds are Pgp substrate s. Sixty-six compounds were tested in monolayer efflux, ATPase, and calcein -AM assays. Assay results yielded two categories of compounds. Category I ( n = 35) exhibited concordance across the assays. Category II (n = 31) revea led differences among the assays that related to the apparent permeability (P-app) of the compounds. Within category II, two groups were discerned bas ed on the absence (group IIA, n = 10, nontransported substrates) or presenc e (group IIB, n = 21, transported substrates) of monolayer efflux. Detectio n of efflux (group IIB) was associated with compounds having low/moderate P -app values (mean = 16.6 nm/s), whereas inability to detect efflux (group I IA) was associated with compounds having high P-app values (mean = 535 nm/s ). The calcein-AM and ATPase assays revealed Pgp interactions for highly pe rmeable group IIA compounds but were less responsive than monolayer efflux for low/moderate P-app compounds of group IIB. All assays detected substrat es across a broad range of P-app, but the efflux assay was more prone to fa il at high P-app whereas the calcein-AM and ATPase assays were more prone t o fail at low P-app. When P-app is low, efflux is a greater factor in the d isposition of Pgp substrates. The efflux assay is more reliable at low/mode rate P-app and is the method of choice for evaluating drug candidates despi te low throughput and reliance on liquid chromatography with tandem mass sp ectrometry.