Yong Ju Kim1, Klaus Peter Hofmann1,2, Oliver P. Ernst3, Patrick Scheerer4,*, Hui-Woog Choe1,5,* & Martha E. Sommer1,*
1Institut für Medizinische Physik und Biophysik (CC2), Charité-Universitätsmedizin Berlin, Charitéplatz 1, D-10117 Berlin, Germany. 2Zentrum für Biophysik und Bioinformatik, Humboldt-Universität zu Berlin, Invalidenstrasse 42, D-10115 Berlin, Germany. 3Departments of Biochemistry and Molecular Genetics, University of Toronto, 1 King’s College Circle, Toronto, Ontario M5S 1A8, Canada. 4Institut für Medizinische Physik und Biophysik (CC2), AG Protein X-ray Crystallography, Charité-Universitätsmedizin Berlin, Charite´platz 1, D-10117 Berlin, Germany. 5Department of Chemistry, College of Natural Science, Chonbuk National University, 561-756 Chonju, South Korea.
*Correspondence to: Martha E. Sommer or Hui-Woog Choe or Patrick Scheerer
Arrestins interact with G-protein-coupled receptors (GPCRs) to block interaction with G proteins1, 2 and initiate G-protein-independent signalling3. Arrestins have a bi-lobed structure that is stabilized by a long carboxy-terminal tail (C-tail), and displacement of the C-tail by receptor-attached phosphates activates arrestins for binding active GPCRs4. Structures of the inactive state of arrestin are available5, 6, but it is not known how C-tail displacement activates arrestin for receptor coupling. Here we present a 3.0Å crystal structure of the bovine arrestin-1 splice variant p44, in which the activation step is mimicked by C-tail truncation. The structure of this pre-activated arrestin is profoundly different from the basal state and gives insight into the activation mechanism. p44 displays breakage of the central polar core and other interlobe hydrogen-bond networks, leading to a ~21° rotation of the two lobes as compared to basal arrestin-1. Rearrangements in key receptor-binding loops in the central crest region include the finger loop7, 8, 9, loop 139 (refs 8, 10, 11) and the sequence Asp296-Asn305 (or gate loop), here identified as controlling the polar core. We verified the role of these conformational alterations in arrestin activation and receptor binding by site-directed fluorescence spectroscopy. The data indicate a mechanism for arrestin activation in which C-tail displacement releases critical central-crest loops from restricted to extended receptor-interacting conformations. In parallel, increased flexibility between the two lobes facilitates a proper fitting of arrestin to the active receptor surface. Our results provide a snapshot of an arrestin ready to bind the active receptor, and give an insight into the role of naturally occurring truncated arrestins in the visual system.