Improving the Sensitivity of P‐Glycoprotein to Drug‐Like Inhibitors in ATPase Assays and ESR Studies

Abstract

Multidrug resistance (MDR) in cancers is frequently associated with transmembrane efflux proteins including P‐glycoprotein (P‐gp), multidrug resistance associated protein (MRP1), and breast cancer resistance protein (BCRP). Over‐expression of these transporters in cancer cells increases efflux of therapeutic drugs rendering them ineffective. In order to re‐sensitize multidrug resistant cancers to chemotherapy, we have found inhibitors of P‐gp by in silico screening methods (Brewer et al, Mol. Pharmacol. 2014). To investigate the mechanism underlying the inhibition, we have employed biochemical assays and electron spin resonance spectroscopy (ESR), and have established nanodisc technologies that incorporate and stabilize P‐gp in native‐like phospholipid bilayer environments for our drug finding and mechanistic studies. In ATP hydrolysis assays using P‐gp in micelles, we found that the concentration of detergents used to solubilize P‐gp during purification had a significant impact on the sensitivity of P‐gp to the putative inhibitors. We hypothesized that partitioning of the compounds into empty micelles may decrease the effective concentration of the potential inhibitors. Carefully decreasing the amount of detergents used during purification resulted in significantly increased sensitivity of P‐gp to the inhibitors, while overall protein activity and stability were not affected. Similarly, excess lipids used during reconstitution of P‐gp into nanodisc may cause partitioning of inhibitors into empty nanodiscs or empty lipid vesicles, resulting in lowered sensitivity of P‐gp to the inhibitors. By optimizing the amount of lipids used during nanodisc assembly, we restored the sensitivity of P‐gp to the inhibitory compounds. Biophysical studies using ESR and a spin‐labeled analog of ATP showed that the computationally identified compounds inhibited P‐gp by binding to the ATP binding domain of P‐gp, which was consistent with our predictions by in silico screening method.Support or Funding InformationThis work is supported by NIH NIGMS [R15GM09477102] to JGW/PDV, SMU University Research Council, SMU Engaged Learning program, the SMU Center for Drug Discovery, Design and Delivery, the Communities Foundation of Texas, and a private gift from Ms. Suzy Ruff of Dallas, Texas.