King's College London

Research portal

In Silico and in Vitro Screening for P-Glycoprotein Interaction with Tenofovir, Darunavir, and Dapivirine: An Antiretroviral Drug Combination for Topical Prevention of Colorectal HIV Transmission

Research output: Contribution to journalArticle

Original languageEnglish
Pages (from-to)2660-2669
Number of pages10
JournalMolecular Pharmaceutics
Issue number8
Early online date20 Jul 2017
Publication statusPublished - 7 Aug 2017

King's Authors


The aim of the study was to use in silico and in vitro techniques to evaluate whether a triple formulation of antiretroviral drugs (tenofovir, darunavir, and dapivirine) interacted with P-glycoprotein (P-gp) or exhibited any other permeability-altering drug–drug interactions in the colorectal mucosa. Potential drug interactions with P-gp were screened initially using molecular docking, followed by molecular dynamics simulations to analyze the identified drug–transporter interaction more mechanistically. The transport of tenofovir, darunavir, and dapivirine was investigated in the Caco-2 cell models and colorectal tissue, and their apparent permeability coefficient (Papp), efflux ratio (ER), and the effect of transporter inhibitors were evaluated. In silico, dapivirine and darunavir showed strong affinity for P-gp with similar free energy of binding; dapivirine exhibiting a ΔGPB value −38.24 kcal/mol, darunavir a ΔGPB value −36.84 kcal/mol. The rank order of permeability of the compounds in vitro was tenofovir < darunavir < dapivirine. The Papp for tenofovir in Caco-2 cell monolayers was 0.10 ± 0.02 × 10–6 cm/s, ER = 1. For dapivirine, Papp was 32.2 ± 3.7 × 10–6 cm/s, but the ER = 1.3 was lower than anticipated based on the in silico findings. Neither tenofovir nor dapivirine transport was influenced by P-gp inhibitors. The absorptive permeability of darunavir (Papp = 6.4 ± 0.9 × 10–6 cm/s) was concentration dependent with ER = 6.3, which was reduced by verapamil to 1.2. Administration of the drugs in combination did not alter their permeability compared to administration as single agents. In conclusion, in silico modeling, cell culture, and tissue-based assays showed that tenofovir does not interact with P-gp and is poorly permeable, consistent with a paracellular transport mechanism. In silico modeling predicted that darunavir and dapivirine were P-gp substrates, but only darunavir showed P-gp-dependent permeability in the biological models, illustrating that in silico modeling requires experimental validation. When administered in combination, the disposition of the proposed triple-therapy antiretroviral drugs in the colorectal mucosa will depend on their distinctly different permeability, but was not interdependent.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454