Homology modeling and structural analysis of human P-glycoprotein

Efflux mechanism and pathway of verapamil pumping by human P-glycoprotein

Multidrug resistance (MDR) caused by overexpressed permeability-glycoprotein (P-gp) in cancer cells is the main barrier for the cure of cancers. P-gp can pump many chemotherapeutic drugs, which is a viable target to overcome P-gp-mediated MDR by efficient inhibitors of P-gp. However, limited understanding of the efflux mechanism by human P-gp hinders the development of efficient inhibitors. Herein, the transport of a P-gp inhibitor, verapamil, by human P-gp has been investigated using targeted molecular dynamics simulations and energetics analysis based on our previous research on the transport of a drug (doxorubicin). The energetics analysis identifies that the driving forces for the transport of verapamil are electrostatic repulsions contributed by the positively charged residues in the initial stage and then hydrophobic interactions contributed by the important residues in the later stage. This scenario is generally consistent with that in the transport of doxorubicin. However, the positively charged residues and the important residues for the transport of verapamil are incompletely consistent with the relative residues for the transport of doxorubicin. Moreover, the binding free energy contributions of the positively charged residues for the transport of verapamil are generally higher than them for the transport of doxorubicin, while the important residues constitute significantly different binding free energy compositions in the transports of the two substrates. Consequently, the pathway for the transport of verapamil is identified, which shares only two residues (F336 and M986) with the pathway of doxorubicin. This may imply the weak competitiveness of verapamil with doxorubicin in the substrate efflux. Taken together, this work provided new insights into the efflux mechanisms by human P-gp and would be beneficial in the design of potent P-gp inhibitors.

In vitro and in silico evaluation of P-glycoprotein inhibition through 99m Tc-methoxyisobutylisonitrile uptake

P-glycoprotein (P-gp) is a multidrug resistance (MDR) transporter with unknown structural details. This macromolecule is normally responsible for extruding xenobiotics from normal cells. Overexpression of P-gp in tumor cells is a major obstacle in cancer chemotherapy. In this study, human 3D model of P-gp was built by homology modeling based on mouse P-gp crystallographic structure and stabilized through 1 ns molecular dynamics (MD) simulation. Stabilized human P-gp structure was used for flexible docking of 80 drugs into the putative active site of P-gp. Accordingly, digoxin, itraconazole, risperidone, ketoconazole, prazosin, verapamil, cyclosporine A, and ranitidine were selected for further in vitro assay. Subsequently, cell-based P-gp inhibition assay was performed on Caco-2 cells while ^99m Tc-methoxyisobutylisonitrile (MIBI) was used as a P-gp efflux substrate for calculating IC50 values. Results of the ^99m Tc-MIBI uptake in drug-treated Caco-2 cells were in agreement with the previously reported activities. This study for the first time described the relation between molecular dynamics and flexible docking with cellular experiments using ^99m Tc-MIBI radiotracer for evaluation of potencies of P-gp inhibitors. Finally, results showed that our radiotracer-cell-based assay is an accurate and fast screening tool for detecting P-gp inhibitors and non-inhibitors in drug development process.

Large-scale classification of P-glycoprotein inhibitors using SMILES-based descriptors

P-glycoprotein (Pgp) inhibition has been considered as an effective strategy towards combating multidrug-resistant cancers. Owing to the substrate promiscuity of Pgp, the classification of its interacting ligands is not an easy task and is an ongoing issue of debate. Chemical structures can be represented by the simplified molecular input line entry system (SMILES) in the form of linear string of symbols. In this study, the SMILES notations of 2254 Pgp inhibitors including 1341 active, and 913 inactive compounds were used for the construction of a SMILE-based classification model using CORrelation And Logic (CORAL) software. The model provided an acceptable predictive performance as observed from statistical parameters consisting of accuracy, sensitivity and specificity that afforded values greater than 70% and MCC value greater than 0.6 for training, calibration and validation sets. In addition, the CORAL method highlighted chemical features that may contribute to increased and decreased Pgp inhibitory activities. This study highlights the potential of CORAL software for rapid screening of prospective compounds from a large chemical space and provides information that could aid in the design and development of potential Pgp inhibitors.

Homology modelling of human P-glycoprotein

P-glycoprotein (P-gp) is an ATP-binding cassette transporter that exports a huge range of compounds out of cells and is thus one of the key proteins in conferring multi-drug resistance in cancer. Understanding how it achieves such a broad specificity and the series of conformational changes that allow export to occur form major, on-going, research objectives around the world. Much of our knowledge to date has been derived from mutagenesis and assay data. However, in recent years, there has also been great progress in structural biology and although the structure of human P-gp has not yet been solved, there are now a handful of related structures on which homology models can be built to aid in the interpretation of the vast amount of experimental data that currently exists. Many models for P-gp have been built with this aim, but the situation is complicated by the apparent flexibility of the system and by the fact that although many potential templates exist, there is large variation in the conformational state in which they have been crystallized. In this review, we summarize how homology modelling has been used in the past, how models are typically selected and finally illustrate how MD simulations can be used as a means to give more confidence about models that have been generated via this approach.

Classification of P-glycoprotein-interacting compounds using machine learning methods

P-glycoprotein (Pgp) is a drug transporter that plays important roles in multidrug resistance and drug pharmacokinetics. The inhibition of Pgp has become a notable strategy for combating multidrug-resistant cancers and improving therapeutic outcomes. However, the polyspecific nature of Pgp, together with inconsistent results in experimental assays, renders the determination of endpoints for Pgp-interacting compounds a great challenge. In this study, the classification of a large set of 2,477 Pgp-interacting compounds (i.e., 1341 inhibitors, 913 non-inhibitors, 197 substrates and 26 non-substrates) was performed using several machine learning methods (i.e., decision tree induction, artificial neural network modelling and support vector machine) as a function of their physicochemical properties. The models provided good predictive performance, producing MCC values in the range of 0.739-1 for internal cross-validation and 0.665-1 for external validation. The study provided simple and interpretable models for important properties that influence the activity of Pgp-interacting compounds, which are potentially beneficial for screening and rational design of Pgp inhibitors that are of clinical importance.

Synthesis of new steroidal inhibitors of P-glycoprotein-mediated multidrug resistance and biological evaluation on K562/R7 erythroleukemia cells

A simple route for improving the potency of progesterone as a modulator of P-gp-mediated multidrug resistance was established by esterification or etherification of hydroxylated 5α/β-pregnane-3,20-dione or 5β-cholan-3-one precursors. X-ray crystallography of representative 7α-, 11α-, and 17α-(2'R/S)-O-tetrahydropyranyl ether diastereoisomers revealed different combinations of axial-equatorial configurations of the anomeric oxygen. Substantial stimulation of accumulation and chemosensitization was observed on K562/R7 erythroleukemia cells resistant to doxorubicin, especially using 7α,11α-O-disubstituted derivatives of 5α/β-pregnane-3,20-dione, among which the 5β-H-7α-benzoyloxy-11α-(2'R)-O-tetrahydropyranyl ether 22a revealed promising properties (accumulation index 2.9, IC50 0.5 μM versus 1.2 and 10.6 μM for progesterone), slightly overcoming those of verapamil and cyclosporin A. Several 7α,12α-O-disubstituted derivatives of 5β-cholan-3-one proved even more active, especially the 7α-O-methoxymethyl-12α-benzoate 56 (accumulation index 3.8, IC50 0.2 μM). The panel of modulating effects from different O-substitutions at a same position suggests a structural influence of the substituent completing a simple protection against stimulating effects of hydroxyl groups on P-gp-mediated transport.

On the origin of large flexibility of P-glycoprotein in the inward-facing state

P-glycoprotein (Pgp) is one of the most biomedically relevant transporters in the ATP binding cassette (ABC) superfamily due to its involvement in developing multidrug resistance in cancer cells. Employing molecular dynamics simulations and double electron-electron resonance spectroscopy, we have investigated the structural dynamics of membrane-bound Pgp in the inward-facing state and found that Pgp adopts an unexpectedly wide range of conformations, highlighted by the degree of separation between the two nucleotide-binding domains (NBDs). The distance between the two NBDs in the equilibrium simulations covers a range of at least 20 Å, including, both, more open and more closed NBD configurations than the crystal structure. The double electron-electron resonance measurements on spin-labeled Pgp mutants also show wide distributions covering both longer and shorter distances than those observed in the crystal structure. Based on structural and sequence analyses, we propose that the transmembrane domains of Pgp might be more flexible than other structurally known ABC exporters. The structural flexibility of Pgp demonstrated here is not only in close agreement with, but also helps rationalize, the reported high NBD fluctuations in several ABC exporters and possibly represents a fundamental difference in the transport mechanism between ABC exporters and ABC importers. In addition, during the simulations we have captured partial entrance of a lipid molecule from the bilayer into the lumen of Pgp, reaching the putative drug binding site. The location of the protruding lipid suggests a putative pathway for direct drug recruitment from the membrane.