Substrate-induced conformational changes in the nucleotide-binding domains of lipid bilayer–associated P-glycoprotein during ATP hydrolysis

Abstract

P-glycoprotein (Pgp) is an efflux pump important in multidrug resistance of cancer cells and in determining drug pharmacokinetics. Pgp is a prototype ATP-binding cassette transporter with two nucleotide-binding domains (NBDs) that bind and hydrolyze ATP. Conformational changes at the NBDs (the Pgp engines) lead to changes across Pgp transmembrane domains that result in substrate translocation. According to current alternating access models (substrate-binding pocket accessible only to one side of the membrane at a time), binding of ATP promotes NBD dimerization, resulting in external accessibility of the drug-binding site (outward-facing, closed NBD conformation), and ATP hydrolysis leads to dissociation of the NBDs with the subsequent return of the accessibility of the binding site to the cytoplasmic side (inward-facing, open NBD conformation). However, previous work has not investigated these events under near-physiological conditions in a lipid bilayer and in the presence of transport substrate. Here, we used luminescence resonance energy transfer (LRET) to measure the distances between the two Pgp NBDs. Pgp was labeled with LRET probes, reconstituted in lipid nanodiscs, and the distance between the NBDs was measured at 37 °C. In the presence of verapamil, a substrate that activates ATP hydrolysis, the NBDs of Pgp reconstituted in nanodiscs were never far apart during the hydrolysis cycle, and we never observed the NBD-NBD distances of tens of Å that have previously been reported. However, we found two main conformations that coexist in a dynamic equilibrium under all conditions studied. Our observations highlight the importance of performing studies of efflux pumps under near-physiological conditions, in a lipid bilayer, at 37 °C, and during substrate-stimulated hydrolysis.

Replacing the eleven native tryptophans by directed evolution produces an active P-glycoprotein with site-specific, non-conservative substitutions

“…The specific verapamil-stimulated ATPase activity (Vmax) was high at around 6 μmol/min/mg, and was similar for all three Pgp variants, assuring that removal of the eight TM Trps or all eleven Trps did not affect the hydrolysis rates. The Km(MgATP) of WT and WL-Pgp were in the mM range as previously reported 47,84 , and were somewhat higher than in W(3Cyto) (p < 0.05). The most notable differences were seen in the 'basal' ATPase www.nature.com/scientificreports www.nature.com/scientificreports/ activity of W(3Cyto) and WL-Pgp in the absence or at low verapamil, valinomycin and FK506 concentrations that were significantly higher than for WT Pgp (p < 0.05; Table 2, Fig.…”

“…cerevisiae cells) was a little lower than previously reported (20 mg/200 g cells) 47,84 , as expected due to the additional affinity purification step. The yields of W(3Cyto) and WL-Pgp (~10 mg/200 g cells and 6-8 mg/200 g cells, respectively) were somewhat lower than for WT.…”

“…WL-Pgp proteins can be purified from S. cerevisiae fermentor cultures using His 6 and dual StrepII-tags with excellent yield and purity (see methods) 47,79 . A single Pgp protein band was apparent on SDS-PAGE gels and no contaminating proteins were detected even with the highly sensitive fluorescent stain SyproRuby affirming that the double-affinity purified proteins were very pure (>95%, Fig.…”

“…Fourteen-liter cultures were grown in a New Brunswick BioFlow IV fermentor to a maximal A 600 of 5 to 6, which produced 200-250 g of cells. Microsomal membrane preparation, Pgp solubilization in 0.6% DDM in buffers containing 500 mM NaCl, and purification based on the affinity of the Pgp His 6 tag for Ni-NTA were performed as described 47,84,111 . Pgp proteins were subjected to an additional purification step on StrepTactin superflow resin (Qiagen, Valencia, CA) in Buffer A (50 mM Tris pH 8, 10% glycerol, 500 mM NaCl, and 0.05% DDM) with 1 mM DTT, and eluted by competition with 4 mM desthiobiotin 47 .…”

“…Nevertheless, more work is needed to determine the physical locations of those drug binding sites in Pgp and the spatial interactions of therapeutic drugs and inhibitors. Similarly, the mechanism by which ATP binding and hydrolysis achieve the conformational changes necessary to expel drugs from the cell is only partially understood [47][48][49][50] .…”