Molecular mechanism of β-arrestin-biased agonism at seven-transmembrane receptors

Abstract

The concept of biased agonism has recently come to the fore with the realization that seven-transmembrane receptors (7TMRs, also known as G protein-coupled receptors, or GPCRs) activate complex signaling networks and can adopt multiple active conformations upon agonist binding. As a consequence, the "efficacy" of receptors, which was classically considered linear, is now recognized as pluridimensional. Biased agonists selectively stabilize only a subset of receptor conformations induced by the natural "unbiased" ligand, thus preferentially activating certain signaling mechanisms. Such agonists thus reveal the intriguing possibility that one can direct cellular signaling with unprecedented precision and specificity and support the notion that biased agonists may identify new classes of therapeutic agents that have fewer side effects. This review focuses on one particular class of biased ligands that has the ability to alter the balance between G protein-dependent and β-arrestin-dependent signal transduction.

Figure 1

Pluridimensionality of β-arrestin-dependent signaling at seven-transmembrane receptor (7TMRs). Some of the best-characterized β-arrestin-induced signaling mechanisms are schematically represented. They include RhoA-dependent stress fiber formation (14); inhibition of nuclear factor κB (NF-κB)-targeted gene expression through IκB stabilization (15, 16); protein phosphatase 2A (PP2A)-mediated dephosphorylation of Akt, which leads to the activation of glycogen synthase kinase 3 (GSK3) and dopaminergic behavior (17); extracellular signal-regulated kinase (ERK)-dependent induction of protein translation and antiapoptosis (18, 19); phosphatidylinositol 3-kinase (PI3K)-mediated phospholipase A2 (PLA2) induction and increased vasodilation through GPR109A activation (20); and Kif3A-dependent relocalization and activation of the protein Smoothened (Smo) in the primary cilium (21). Other abbreviations: AT_1A R, angiotensin type 1A receptor; β₁AR, β₁ adrenergic receptor; BAD, Bcl-2-associated death promoter; D2R, dopamine receptor D₂; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; EIF4E, eukaryotic translation initiation factor 4E; HB-EGF, heparin-binding EGF-like growth factor; MEK, mitogen-activated protein kinase/extracellular signal signal-regulated kinase kinase; MMP, matrix metalloproteinase; PGD2, prostaglandin D2; Ptc, Patched; ROCK, Rho-associated protein kinase; Shh, sonic hedgehog homolog.

Figure 2

Comparison of pluridimensional efficacies for different ligands. (a) Matrix of pairwise-determined biased factors βof one ligand for n dimensions. Biased factors βcan be calculated from dose-response curves as recently proposed (36). Values of βcalculated for each pair of readouts are represented in a heat map in which red corresponds to positive bias and blue corresponds to negative bias. Matrices for two different ligands can be statistically tested for balance versus bias in the pluridimensional efficacies. (b) For multiple ligand comparisons (e.g., data from high-throughput multiplexed screening), the matrix corresponding to each ligand can be linearized and subsequently analyzed by hierarchical clustering. Clusters of compounds with similar efficacies or particular “signatures” can be identified. Information about the structure of the intracellular signaling network can potentially be inferred from these analyses (i.e., determination of independent versus co-regulated cellular outcomes).

Figure 3

Multiple ligand-specific conformations of the β₂ adrenergic receptor (β₂AR). Cysteines and lysines were labeled in the purified β₂AR using isotope-coded N-ethylmaleimide and succinic anhydride, respectively. Mass spectrometry studies revealed that nine residues (four cysteines and five lysines) were suitable for quantifying site-specific conformational changes. (a) Schematic representation of the β₂AR in single-letter amino acid code sequences showing the studied cysteines (C, red ) and lysines (K, green). (b) Labeling factors (i.e., negative log of relaxation times), measured at each of the nine cysteines or lysines upon receptor binding with nine different ligands, were plotted in a heat map (47). Values were calculated for each labeled residue upon treatment with each of the ligands; red corresponds to positive values and blue corresponds to negative values, relative to vehicle-treated β₂AR. Relative activities of the nine ligands for G protein activation (G prot), β-arrestin recruitment (β-arr), and extracellular signal-regulated kinase (ERK) phosphorylation at the β₂AR are represented in the heat map to facilitate direct comparison (right panel ) (47). The ligands were classified as full agonists, partial agonists, or antagonists on the basis of their pharmacological profiles.

Figure 4

G protein–coupled receptor kinase (GRK)-mediated phosphorylation “bar code” at the C terminus of seven-transmembrane receptors (7TMRs) converts ligand-induced conformation of the receptor into selective β-arrestin intracellular functions. A general model for β-arrestin-biased ligand mechanism of action is proposed (78, 80, 82). GRK2/3 and GRK5/6 exert qualitatively different actions. GRK2/3 require Gβ for membrane recruitment and activation and phosphorylate specific serines and threonines in the receptor’s C-tail; the GRK2/3-mediated phosphorylation “bar code” then leads to desensitization and internalization upon β-arrestin recruitment. GRK5/6 do not require G proteins for their activation; the GRK5/6-dependent phosphorylation “bar code” then creates a signaling platform in which β-arrestin bridges the activated receptor to partners involved in intracellular signaling such as the mitogen-activated protein (MAP) kinase extracellular signal-regulated kinase (ERK) module. Other abbreviation: MEK, MAP kinase/extracellular signal-regulated kinase kinase.