Interactions of Metarhodopsin II : Arrestin Peptides Compete with Arrestin and Transducin

Arrestin blocks the interaction of rhodopsin with the G protein transducin (G(t)). To characterize the sites of arrestin that interact with rhodopsin, we have utilized a spectrophotometric peptide competition assay. It is based on the stabilization of the active intermediates metarhodopsin II (MII) and phosphorylated MII by G(t) and arrestin, respectively (extra MII monitor). The protocol involves native disc membranes and three sets of peptides 10-30 amino acids in length spanning the arrestin sequence. In the absence of arrestin, not one of the peptides by itself had an effect on the amount of MII formed. However, inhibition of arrestin-dependent extra MII was found for the peptides at residues 11-30 and 51-70 (IC(50) < 100 microm) and residues 231-260 (IC(50) < 200 microm). A similar pattern of inhibition by arrestin peptides was seen when arrestin was replaced by G(t) or the farnesylated G(t)gamma C-terminal peptide. Only arrestin-(11-30) inhibited MII.G(t) less (IC(50) = 300 microm) than phosphorylated MII.arrestin. We interpreted the data by competition of the arrestin peptides for interaction sites at rhodopsin, exposed in the MII conformation and specific for both arrestin and G(t). The arrestin sites are located in both the C- and N-terminal domains of the arrestin structure.


INTRODUCTION
G protein-coupled receptors (GPCRs) enable eukaryotic cells to respond to a large variety of extracellular signals including hormones, odorants, and light (1). One of the best studied GPCR-initiated signaling pathways is the visual cascade in retinal rods (2). It is initiated by the absorption of a photon in the visual receptor, rhodopsin, and subsequent isomerization of the 11-cis-retinal, which is covalently attached to The deactivation of rhodopsin starts with the binding of a rhodopsin kinase to photoactivated rhodopsin (4). Rapid phosphorylation of the receptor at C-terminal sites (5) increases its affinity for arrestin (6). Biochemical (7) and electrophysiological (8) evidence has been accumulated that visual arrestins deactivate the visual cascade by direct competition with the G-protein for the active receptor. However, the mechanism of interaction underlying the quench is still not well understood. UV-Vis spectroscopy has shown that, to bind arrestin, the light activation of rhodopsin must proceed up to the MII conformation, in which the Schiff base bond of the retinal to the apoprotein is still intact but deprotonated (9). Only after MII formation and phosphorylation of the receptor, arrestin interacts rapidly with the receptor. Although the catalytic activity of MII towards the G-protein is quenched, forms (see Ref. (6) and citations therein). A conformational difference in arrestin and/or rhodopsin on interaction was early suggested by the unusually high apparent activation energy of rhodopsin-arrestin interaction (9). Later studies have indeed shown that binding of arrestin to the active receptor protects arrestin against limited proteolysis (10) and Lysine acetylation (11), suggesting that arrestin bound rhodopsin adopts a conformation (A b ) which is different from that of free, inactive (A i ) arrestin.
Investigations employing protein engineering (12,13) and phosphorylated peptides (14) have provided evidence for a sequential mechanism, in which the contact of the negatively charged regions of phosphorylated rhodopsin (P-site) with the cationic region acts as a trigger, switching arrestin into its active conformation and allowing interaction with the rhodopsin binding sites exposed on photoactivation.
A highly cationic region near the center of the arrestin sequence, beginning with residue 163, was proposed to mediate the interaction with the P-site, thus enabling the contact with other interaction sites exposed in the MII state (M-sites) (10). Arg 175 (within the putative recognition site for the phosphorylated site(s) on rhodopsin) was identified as a key residue for arrestin to distinguish between phosphorylated and unphosphorylated rhodopsin (15,16). Recent structural assignments (17,18) have now identified this residue as part of a "polar core" (18), a central region in arrestin, localized between the N-and C-terminal domains of the molecule. It may act as a fulcrum for the conversion of inactive A i to active A b (19).
The conformational switch is additionally controlled by arrestin´s C-terminus (20,21). When it is lacking, as in a splice variant of arrestin, p 44 , the short arrestin binds both phosphorylated and nonphosphorylated forms of MII and even Cterminally truncated rhodopsin (22,23). Protein engineering (24) and spectroscopic data (25)  residues interact with positively charged residues on the N-terminal region, and that this interaction is broken upon binding to phosphorylated MII.
In the present study we attempt to identify specific sites of arrestin that become exposed by the conformational switch, leading to their interaction with the respective receptor sites in the MII conformation. The technique applied is based on the spectrophotometric MII stabilization assay (9) and on competition with peptides derived from arrestin and G t . Peptide competition has already been applied in previous studies, to map regions of interaction in the receptor-arrestin complex. Studies with peptides from rhodopsin's surface exposed sequences suggested a role of different loop structures in the interaction (26). Regarding the interactive domains in arrestin, Hargrave and coworkers have employed a library of peptides covering the entire sequence. Based on phage-display and PDE activation data, they localized one of the principal regions of interaction within residues 109-130 (27) and found indications of additional sites in their data.
We will indeed provide evidence that more than one (namely three) regions in the arrestin molecule are involved in receptor binding. Employing the sensitive MII stabilization assay, and peptide competition not only with arrestin but also with G t and Isolation of Bovine Rod Outer Segments (ROS). ROS were purified from fresh, dark-adapted bovine retinas obtained from a local slaughterhouse using the discontinuous sucrose gradient method described by Papermaster (28). Retinas were dissected, and ROS were isolated, under dim red illumination. All subsequent procedures were performed at 0-5 °C and the ROS were stored frozen at -80°C until use.
Preparation of Washed Membranes. Rhodopsin was prepared by removing the soluble and membrane-associated proteins from the disc membrane by repetitive washes with a low ionic strength buffer (29). All purification steps were performed under dim red illumination and the membrane suspension was stored at -80°C until use.
Preparation of Phosphorylated Opsin. Phosphorylated opsin was prepared from washed disc membranes as described previously by Wilden and Kühn (30). To remove retinaloxime from the membrane-bound phosphorylated opsin the membranes were treated with urea and fatty acid-free bovine serum albumin (31 washed four times with 10 mM BTP buffer, pH 7.5 containing 100 mM NaCl to remove excess 11-cis-retinal and stored at -80°C.

Preparation of arrestin.
Arrestin was purified from frozen dark-adapted bovine retinas as described by Heck et al. (33).

Preparation of transducin.
Transducin was isolated from frozen dark-adapted bovine retinas according to Heck and Hofmann (34).

Peptide synthesis, purification and characterization.
Peptides were synthesized using the Fmoc (N-(9-fluorenyl)methoxycarbonyl) strategy with HBTU activation ( Fastmoc -0.1 mmol small-scale cycles) on an ABI Model 433A peptide synthesizer. The peptides were purified by reverse phase high performance liquid chromatography (HPLC), lyophilized and stored at -20°C. Immediately before the experiments, the peptides were dissolved in deionized water to obtain stock solutions of 2 mM and pH was adjusted to 7.0 with NaOH. Farnesylation of G t -derived peptides was carried out as described (35).
Protein Determinations. The concentration of rhodopsin and phosphorylated rhodopsin was determined spectrophotometrically at 498 nm as previously described (23). Purified arrestin was determined spectrophotometrically at 278 nm, assuming a molar absorption coefficient of E 0.1% = 0.638 (36) and a molecular mass of 45,300 Da.
Purified transducin (G t ) concentration was determined using the Bradford method (37).   and apparent Hill coefficients (n), are listed in Table I. High values of n may be due to a tendency of the peptide to aggregate and do therefore not enter the calculation in the Appendix.

Light induced Interaction between Rhodopsin and Arrestin is
Transducin forms enhanced MII complexes with photoactivated nonphosphorylated rhodopsin. It is further known that arrestin inhibits the activity of photoactivated rhodopsin towards the G-protein (7). We therefore investigated whether arrestin peptides can inhibit the transducin dependent formation of extra MII.
As shown in Fig. 2B, the two peptides that compete with arrestin also compete with G t with similar relative efficiency, providing a control for direct interaction of the peptides with rhodopsin (see Discussion).

Arrestin Peptides Compete with G t γ-farnesyl but not G t α C-terminal
Peptides. The question was whether arrestin peptides interfere with G t -derived peptides in their interaction with photoexcited rhodopsin. One may expect that this can be tested by direct competition, because C-terminal peptides from G t stabilize the MII state like the holoprotein (41,42), but arrestin peptides do not (see above, Fig.   2). Fig. 3 shows for two examples that arrestin peptides can indeed competitively inhibit the formation of extra MII induced by G t -derived peptides. The effect is specifically seen with the G t -farnesyl peptide, indicating an overlap of the respective binding sites at rhodopsin. No such inhibition was measured for the interaction with the G t α C-terminal peptide (native and high affinity analog).
As is seen in Fig. 4, almost all the peptides that compete with arrestin do so with G t and the G t γ C-terminal peptide as well. An exception is the region near the N-

Peptide competition as a monitor of arrestin-receptor interaction sites.
The arrestin that bind to rhodopsin. Using a phage-display technique of arrestin fragments and G t binding and activation assays, these authors identified a stretch comprising residues 109-130 as a site involved in the interaction with rhodopsin. Because of the low affinity of this region (IC 50 of 1.1 mM) it was suggested that this portion of arrestin may be only one of several binding sites for rhodopsin (27). In the present study, using the fundamentally different extra MII assay, we find indeed peptide competition with similar IC 50 in this region (Table I)  The experiments were otherwise performed as in Fig. 1.
The IC 50 value represent the arrestin peptide concentration where the extra MII signal is inhibited to 50%.
Note: Peptide binding to R or R* led to very similar K D3 values (errors < 1%).