How ligands achieve biased signaling at opioid receptors

Abstract ID 27497 Poster Board 317 Some drugs targeting the κ-opioid or μ-opioid receptors induce fewer hazardous side effects than others. This is likely because these drugs preferentially stimulate G protein signaling over arrestin signaling, a phenomenon known as biased signaling. Ligands with more finely tuned biased signaling profiles are highly sought after in order to further decrease side effects while increasing efficacy, but the design of such ligands has proven challenging because the molecular mechanism of biased signaling has remained unclear. To determine this mechanism, we used molecular dynamics simulations to identify differences between receptor conformations favored by ligands with distinct signaling profiles, and then validated the results experimentally. In two parallel studies, we compared the relatively safe opioids nalfurafine, targeting the κ-opioid receptor, and mitragynine pseudoindoxyl, targeting the μ-opioid receptor, to ligands known to induce severe side effects at each respective receptor. We found that, in both receptors, ligands achieve biased signaling by selecting among distinct receptor conformations. In particular, differently biased agonists differentially favor three active receptor conformations: a conformation that couples effectively to both G proteins and arrestins, a conformation that couples preferentially to G proteins, and a conformation that couples preferentially to arrestins. These conformations differ most notably in the orientation of the intracellular part of transmembrane helix 7, far from the ligand binding pocket but close to the binding surfaces of G proteins and arrestins. We identified several protein–ligand interactions that act together to determine the observed conformational ensemble. We validated our computationally derived mechanism using mutagenesis experiments and functional assays. Our work not only illuminates the structural basis for biased signaling at opioid receptors and related GPCRs, but also promises to guide the design of ligands with desired signaling profiles. An award of computer time was provided by the INCITE program. This research used resources of the Oak Ridge Leadership Computing Facility, which is a U.S. Department of Energy Office of Science User Facility supported under contract DE-AC05-00OR22725.