P-glycoprotein (MDR1, ABCB1)
P-glycoprotein (MDR1, ABCB1) – the prototypic polyspecific ABC-transporter
When in 1976 Juliano and Ling discovered that the main difference between a wild type tumour cell and a multidrug resistant tumour cell is a glycosylated transmembrane protein which reduces the accumulation of cytotoxic agents in the cell, this marked the beginning of a remarkable journey. A few years later it became evident that this protein named P-glycoprotein (P for permeability) is the molecular basis for multiple drug resistance in tumour therapy and that inhibitors of this protein may resensitize multidrug resistant tumours.
P-glycoprotein (P-gp) is a transmembrane, ATP-driven transport protein which transports a wide variety of xenobiotics across cellular membranes and actively pumps them out of cells. It is one of 48 members of the ABC-transporter (ATP binding cassette) family. Furthermore, its expression in almost all barrier tissues in the human body, such as the gut, the blood-brain barrier, the blood-placenta barrier, and important organs like the liver outlined its major role in drug-drug interactions. Accordingly, the US Food and Drug Administration and the European Medicines Agency recommend testing for every drug candidate for clinical drug interaction studies due to interaction with P-glycoprotein. For the medicinal chemist, designing in or designing out P-glycoprotein substrate and/or inhibitor properties became an important task in lead optimisation. Thus, anticancer drugs as well as drugs targeting the cns should not show substrate properties.
In contrary, if one wants to avoid cns side effects designing in P-glycoprotein substrate properties might be a good strategy, as is the case with the antihistaminic drug cetirizine. In addition, also for absorption of drugs P-glycoprotein plays a major role as outlined in the BCS classification system by Benett. Briefly, in case a compound shows low water solubility and low membrane permeability, outward transport by P-glycoprotein is the rate limiting step for absorption. In this case, co-administration of inhibitors of P-gp might dramatically influence the plasma concentration and tissue distribution of drugs. Therefore, computational models predicting whether a compound is a substrate or an inhibitor of P-gp allow to identify potential risks in the drug discovery process.
Building on the 25+ years of experience of the Pharmacoinformatic Research Group of Gerhard Ecker at the University of Vienna, Phenaris GmbH offers a broad range of computational models and services to predict the interaction of a set of compounds with P-glycoprotein. Core of these tools is the Transporter Model Space, which provides three different machine learning models for inhibitor prediction, all showing accuracies > 85%. For accessing the models, we offer very flexible licensing conditions, ranging from pay per use up to annual site licenses. More elaborated studies such as docking based classification or substrate predictions are available upon request.
For more information please contact us.
