P-glycoprotein is a membrane protein involved in the phenomenon of multidrug resistance. Its activity and transport function have been largely characterized by various biochemical studies and a low-resolution image has been obtained by electron microscopy. Obtaining a high-resolution structure is, however, still remote due to the inherent difficulties in the experimental determination of membrane protein structures. We present here a three-dimensional (3D) atomic model of P-glycoprotein in absence of ATP. This model was obtained using a combination of computational techniques including comparative modeling and rigid body dynamics simulations that embody all available cysteine disulfide crosslinking data characterizing the whole protein in absence of ATP. The model features rather well most of the experimental interresidue distances derived both in the transmembrane domains and in the nucleotide binding domains. The model is also in good agreement with electron microscopy data, particularly in terms of size and topology. It features a large cavity detected in the protein core into which seven ligands were successfully docked. Their predicted affinity correlates well with experimental values. Locations of docked ligands compare favorably with those suggested by cysteine-scanning data. The finding of different positions both for a single ligand and for different ligands corroborates the experimental evidence indicating the existence of multiple drug binding sites. The interactions identified between P-glycoprotein and the docked ligands reveal that different types of interactions such as H-bonds, π-π and cation-π interactions occur in agreement with a recently proposed pharmacophore model of P-glycoprotein ligands. Furthermore, the model also displays a lateral opening located in the transmembrane domains connecting the lipid bilayer to the central cavity. This feature supports rather well the commonly admitted mechanism of substrate uptake from the lipid bilayer. We propose that this 3D model may be an important tool to understand the structure-function relationship of P-glycoprotein. © 2006 Wiley-Liss, Inc.
CITATION STYLE
Vandevuer, S., Van Bambeke, F., Tulkens, P. M., & Prévost, M. (2006). Predicting the three-dimensional structure of human P-glycoprotein in absence of ATP by computational techniques embodying crosslinking data: Insight into the mechanism of ligand migration and binding sites. Proteins: Structure, Function and Genetics, 63(3), 466–478. https://doi.org/10.1002/prot.20892
Mendeley helps you to discover research relevant for your work.