The Arrestin–Rhodopsin Interface: Implications for the Structure of the Complex

Abstract

Abstract: : Purpose: Arrestin binding to phosphorylated light–activated rhodopsin (P–Rh*) plays a key role in the shut–off of visual signaling. To elucidate the mechanism of this interaction, we are identifying the elements of both proteins involved. Methods: For site–directed spin labeling, individual cysteines were introduced into both proteins on cysteine–less background. EPR spectra of free and interacting labeled proteins were used to determine the participation of labeled residues in the interaction. In parallel, mutant arrestins and chimeras between visual arrestin and arrestin2 were constructed and tested for their relative binding to their preferred target, rhodopsin and the m2 muscarinic cholinergic receptor, respectively, to identify residues responsible for their receptor specificity. Results: Both of these methods implicate elements on the concave sides of both domains in arrestin. The exchange of seven and ten residues, in the N– and C–domain, respectively, switches receptor preference. We have tested visual arrestin and arrestin2 mutants with spin–labels in 30 and 12 different positions, respectively. The mobility of 20 labels in visual and 8 labels in arrestin2 significantly changes upon receptor binding. The regions involved include a highly flexible loop (residues 68–78) in the center of arrestin molecule. Binding to P–Rh*of three different mutants of each arrestin spin–labeled in this loop results in a very dramatic decrease in mobility, suggesting its direct involvement in receptor binding. On the rhodopsin side, a spin label at 250 in the cavity between the transmembrane helices that opens due to helix movement upon light activation of rhodopsin is immobilized by arrestin binding. The mobility of labels in control positions shows that light–activated rhodopsin in complex with arrestin retains its characteristic "open" conformation, suggesting that arrestin occludes the label at 250 in the rhodopsin cavity directly. Conclusions: Multiple arrestin and rhodopsin residues participate in their interaction. Based on the localization of these residues and molecular modeling we hypothesize that the flexible loop of arrestin inserts into the cavity of light–activated rhodopsin. These data strongly support a model where one arrestin molecule binds one rhodopsin molecule.

Keywords: protein structure/function • photoreceptors • signal transduction