Both CXC chemokine receptor 4 (CXCR4) and atypical chemokine receptor 3 (ACKR3) are activated by the chemokine CXCL12 yet evoke distinct cellular responses. CXCR4 is a canonical G protein–coupled receptor (GPCR), whereas ACKR3 is intrinsically biased for arrestin. The molecular basis for this difference is not understood. Here, we describe cryo-EM structures of ACKR3 in complex with CXCL12, a more potent CXCL12 variant, and a small-molecule agonist. The bound chemokines adopt an unexpected pose relative to those established for CXCR4 and observed in other receptor-chemokine complexes. Along with functional studies, these structures provide insight into the ligand-binding promiscuity of ACKR3, why it fails to couple to G proteins, and its bias toward β-arrestin. The results lay the groundwork for understanding the physiological interplay of ACKR3 with other GPCRs.

G protein-coupled receptors (GPCRs), which regulate a vast number of eukaryotic processes, are desensitized by various mechanisms but, most importantly, by the GPCR kinases (GRKs). Ever since GRKs were first identified, investigators have sought to determine which structural features of GRKs are used to select for the agonist-bound states of GPCRs and how this binding event in turn enhances GRK catalytic activity. Despite a wealth of molecular information from high-resolution crystal structures of GRKs, the mechanisms driving activation have remained elusive, in part because the GRK N-terminus and active site tether region, previously proposed to serve as a receptor docking site and to be key to kinase domain closure, are often disordered or adopt inconsistent conformations. However, two recent studies have implicated other regions of GRKs as being involved in direct interactions with active GPCRs. Atomic resolution structures of GPCR–GRK complexes would help refine these models but are, so far, lacking. Here, we assess three distinct models for how GRKs recognize activated GPCRs, discuss limitations in the approaches used to generate them, and then experimentally test a hypothetical GPCR interaction site in GRK2 suggested by the two newest models.

The nine different membrane-anchored adenylyl cyclase isoforms (AC1-9) in mammals are stimulated by the heterotrimeric G protein Gα _s , but their response to Gβγ regulation is isoform-specific. For example, AC5 is conditionally activated by Gβγ. Here, we report cryo-EM structures of ligand-free AC5 in complex with Gβγ and of a dimeric form of AC5 that could be involved in its regulation. Gβγ binds to a coiled-coil domain that links the AC transmembrane region to its catalytic core as well as to a region (C _1b ) that is known to be a hub for isoform-specific regulation. We confirmed the Gβγ interaction with both purified proteins and cell-based assays. The interface with Gβγ involves AC5 residues that are subject to gain-of-function mutations in humans with familial dyskinesia, indicating that the observed interaction is important for motor function. A molecular mechanism wherein Gβγ either prevents dimerization of AC5 or allosterically modulates the coiled-coil domain, and hence the catalytic core, is proposed. Because our mechanistic understanding of how individual AC isoforms are uniquely regulated is limited, studies such as this may provide new avenues for isoform-specific drug development.