Direct binding of visual arrestin to microtubules determines the differential subcellular localization of its splice variants in rod photoreceptors

Proper function of visual arrestin is indispensable for rapid signal shut-off in rod photoreceptors. Dramatic light-dependent changes in its subcellular localization are believed to play an important role in light adaptation of photoreceptor cells. Here we show that visual arrestin binds microtubules. The truncated splice variant of visual arrestin, p44, demonstrates dramatically higher affinity for microtubules than the full-length protein (p48). Enhanced microtubule binding of p44 underlies its earlier reported preferential localization to detergent-resistant membranes, where it is anchored via membrane-associated microtubules in a rhodopsin-independent fashion. Experiments with purified proteins demonstrate that arrestin interaction with microtubules is direct and does not require any additional protein partners. Most importantly, arrestin interactions with microtubules and light-activated phosphorylated rhodopsin are mutually exclusive, suggesting that microtubule interaction may play a role in keeping p44 arrestin away from rhodopsin in dark-adapted photoreceptors.     

The translocation of signaling molecules in dark adapting mammalian rod photoreceptor cells is dependent on the cytoskeleton

In vertebrate rod photoreceptor cells, arrestin and the visual G-protein transducin move between the inner segment and outer segment in response to changes in light. This stimulus dependent translocation of signalling molecules is assumed to participate in long term light adaptation of photoreceptors. So far the cellular basis for the transport mechanisms underlying these intracellular movements remains largely elusive. Here we investigated the dependency of these movements on actin filaments and the microtubule cytoskeleton of photoreceptor cells. Co-cultures of mouse retina and retinal pigment epithelium were incubated with drugs stabilizing and destabilizing the cytoskeleton. The actin and microtubule cytoskeleton and the light dependent distribution of signaling molecules were subsequently analyzed by light and electron microscopy. The application of cytoskeletal drugs differentially affected the cytoskeleton in photoreceptor compartments. During dark adaptation the depolymerization of microtubules as well as actin filaments disrupted the translocation of arrestin and transducin in rod photoreceptor cells. During light adaptation only the delivery of arrestin within the outer segment was impaired after destabilization of microtubules. Movements of transducin and arrestin required intact cytoskeletal elements in dark adapting cells. However, diffusion might be sufficient for the fast molecular movements observed as cells adapt to light. These findings indicate that different molecular translocation mechanisms are responsible for the dark and light associated translocations of arrestin and transducin in rod photoreceptor cells.     

Structure and function of the visual arrestin oligomer

A distinguishing feature of rod arrestin is its ability to form oligomers at physiological concentrations. Using visible light scattering, we show that rod arrestin forms tetramers in a cooperative manner in solution. To investigate the structure of the tetramer, a nitroxide side chain (R1) was introduced at 18 different positions. The effects of R1 on oligomer formation, EPR spectra, and inter-spin distance measurements all show that the structures of the solution and crystal tetramers are different. Inter-subunit distance measurements revealed that only arrestin monomer binds to light-activated phosphorhodopsin, whereas both monomer and tetramer bind microtubules, which may serve as a default arrestin partner in dark-adapted photoreceptors. Thus, the tetramer likely serves as a 'storage' form of arrestin, increasing the arrestin-binding capacity of microtubules while readily dissociating to supply active monomer when it is needed to quench rhodopsin signaling.     

Visual arrestin binding to microtubules involves a distinct conformational change

Recently we found that visual arrestin binds microtubules and that this interaction plays an important role in arrestin localization in photoreceptor cells. Here we use site-directed mutagenesis and spin labeling to explore the molecular mechanism of this novel regulatory interaction. The microtubule binding site maps to the concave sides of the two arrestin domains, overlapping with the rhodopsin binding site, which makes arrestin interactions with rhodopsin and microtubules mutually exclusive. Arrestin interaction with microtubules is enhanced by several "activating mutations" and involves multiple positive charges and hydrophobic elements. The comparable affinity of visual arrestin for microtubules and unpolymerized tubulin (K(D) > 40 mum and >65 mum, respectively) suggests that the arrestin binding site is largely localized on the individual alphabeta-dimer. The changes in the spin-spin interaction of a double-labeled arrestin indicate that the conformation of microtubule-bound arrestin differs from that of free arrestin in solution. In sharp contrast to rhodopsin, where tight binding requires an extended interdomain hinge, arrestin binding to microtubules is enhanced by deletions in this region, suggesting that in the process of microtubule binding the domains may move in the opposite direction. Thus, microtubule and rhodopsin binding induce different conformational changes in arrestin, suggesting that arrestin assumes three distinct conformations in the cell, likely with different functional properties.     

Transition of arrestin into the active receptor-binding state requires an extended interdomain hinge

Arrestins selectively bind to the phosphorylated activated form of G protein-coupled receptors, thereby blocking further G protein activation. Structurally, arrestins consist of two domains topologically connected by a 12-residue long loop, which we term the "hinge" region. Both domains contain receptor-binding elements. The relative size and shape of arrestin and rhodopsin suggest that dramatic changes in arrestin conformation are required to bring all of its receptor-binding elements in contact with the cytoplasmic surface of the receptor. Here we use the visual arrestin/rhodopsin system to test the hypothesis that the transition of arrestin into its active receptor-binding state involves a movement of the two domains relative to each other that might be limited by the length of the hinge. We have introduced three insertions and 24 deletions in the hinge region and measured the binding of all of these mutants to light-activated phosphorylated (P-Rh*), dark phosphorylated (P-Rh), dark unphosphorylated (Rh), and light-activated unphosphorylated rhodopsin (Rh*). The addition of 1-3 extra residues to the hinge has no effect on arrestin function. In contrast, sequential elimination of 1-8 residues results in a progressive decrease in P-Rh* binding without changing arrestin selectivity for P-Rh*. These results suggest that there is a minimum length of the hinge region necessary for high affinity binding, consistent with the idea that the two domains move relative to each other in the process of arrestin transition into its active receptor-binding state. The same length of the hinge is also necessary for the binding of "constitutively active" arrestin mutants to P-Rh*, dark P-Rh, and Rh*, suggesting that the active (receptor-bound) arrestin conformation is essentially the same in both wild type and mutant forms.     

Regulation of arrestin binding by rhodopsin phosphorylation level

Arrestins ensure the timely termination of receptor signaling. The role of rhodopsin phosphorylation in visual arrestin binding was established more than 20 years ago, but the effects of the number of receptor-attached phosphates on this interaction remain controversial. Here we use purified rhodopsin fractions with carefully quantified content of individual phosphorylated rhodopsin species to elucidate the impact of phosphorylation level on arrestin interaction with three biologically relevant functional forms of rhodopsin: light-activated and dark phosphorhodopsin and phospho-opsin. We found that a single receptor-attached phosphate does not facilitate arrestin binding, two are necessary to induce high affinity interaction, and three phosphates fully activate arrestin. Higher phosphorylation levels do not increase the stability of arrestin complex with light-activated rhodopsin but enhance its binding to the dark phosphorhodopsin and phospho-opsin. The complex of arrestin with hyperphosphorylated light-activated rhodopsin is less sensitive to high salt and appears to release retinal faster. These data suggest that arrestin likely quenches rhodopsin signaling after the third phosphate is added by rhodopsin kinase. The complex of arrestin with heavily phosphorylated rhodopsin, which appears to form in certain disease states, has distinct characteristics that may contribute to the phenotype of these visual disorders.     

Cell-free expression of visual arrestin. Truncation mutagenesis identifies multiple domains involved in rhodopsin interaction.

Visual arrestin plays an important role in regulating light responsiveness via its ability to specifically bind to the phosphorylated and light-activated form of rhodopsin. To further characterize rhodopsin/arrestin interactions we have utilized a rabbit reticulocyte lysate translation system to synthesize bovine visual arrestin. The translated arrestin (404 amino acids) was demonstrated to be fully functional in terms of its ability to specifically recognize and bind to phosphorylated light-activated rhodopsin (P-Rh*). Competitive binding studies revealed that the in vitro synthesized arrestin and purified bovine visual arrestin had comparable affinities for P-Rh*. In an effort to assess the functional role of different regions of the arrestin molecule, two truncated arrestin mutants were produced by cutting within the open reading frame of the bovine arrestin cDNA with selective restriction enzymes. In vitro translation of the transcribed truncated mRNAs resulted in the production of arrestins truncated from the carboxyl terminus. The ability of each of the mutant arrestins to bind to dark (Rh), light-activated (Rh*), dark phosphorylated (P-Rh), and light-activated phosphorylated rhodopsin were then compared. Arrestin lacking 39 carboxyl-terminal residues binds specifically not only to P-Rh* but also to Rh* and P-Rh. This suggests that the carboxyl-terminal domain of arrestin plays an important regulatory role in ensuring strict arrestin binding selectivity to P-Rh*. Arrestin that has only the first 191 amino-terminal residues predominately discriminates the phosphorylation state of the rhodopsin; however, it also retains some binding specificity for the activation state. These results suggest that the amino-terminal half of arrestin contains key rhodopsin recognition sites responsible for interaction with both the phosphorylated and light-activated forms of rhodopsin.     

Mechanism of light-induced translocation of arrestin and transducin in photoreceptors: interaction-restricted diffusion

Many signaling proteins change their location within cells in response to external stimuli. In photoreceptors, this phenomenon is remarkably robust. The G protein of rod photoreceptors and rod transducin concentrates in the outer segments (OS) of these neurons in darkness. Within approximately 30 minutes after illumination, rod transducin redistributes throughout all of the outer and inner compartments of the cell. Visual arrestin concurrently relocalises from the inner compartments to become sequestered primarily within the OS. In the past several years, the question of whether these proteins are actively moved by molecular motors or whether they are redistributed by simple diffusion has been extensively debated. This review focuses on the most essential works in the area and concludes that the basic principle driving this protein movement is diffusion. The directionality and light dependence of this movement is achieved by the interactions of arrestin and transducin with their spatially restricted binding partners.     

Arrestin and its splice variant Arr1-370A (p44). Mechanism and biological role of their interaction with rhodopsin

Deactivation of G-protein-coupled receptors relies on a timely blockade by arrestin. However, under dim light conditions, virtually all arrestin is in the rod inner segment, and the splice variant p(44) (Arr(1-370A)) is the stop protein responsible for receptor deactivation. Using size exclusion chromatography and biophysical assays for membrane-bound protein-protein interaction, membrane binding, and G-protein activation, we have investigated the interactions of Arr(1-370A) and proteolytically truncated Arr(3-367) with rhodopsin. We find that these short arrestins do not only interact with the phosphorylated active receptor but also with inactive phosphorylated rhodopsin or opsin in membranes or solution. Because of the latter interaction they are not soluble (like arrestin) but membrane-bound in the dark. Upon photoexcitation, Arr(3-367) and Arr(1-370A) interact with prephosphorylated rhodopsin faster than arrestin and start to quench G(t) activation on a subsecond time scale. The data indicate that in the course of rhodopsin deactivation, Arr(1-370A) is handed over from inactive to active phosphorylated rhodopsin. This mechanism could provide a new aspect of receptor shutoff in the single photon operating range of the rod cell.     

Metarhodopsin control by arrestin,light-filtering screening pigments,and visual pigment turnover in invertebrate microvillar photoreceptors

The visual pigments of most invertebrate photoreceptors have two thermostable photo-interconvertible states, the ground state rhodopsin and photo-activated metarhodopsin, which triggers the phototransduction cascade until it binds arrestin. The ratio of the two states in photoequilibrium is determined by their absorbance spectra and the effective spectral distribution of illumination. Calculations indicate that metarhodopsin levels in fly photoreceptors are maintained below ~35% in normal diurnal environments, due to the combination of a blue-green rhodopsin, an orange-absorbing metarhodopsin and red transparent screening pigments. Slow metarhodopsin degradation and rhodopsin regeneration processes further subserve visual pigment maintenance. In most insect eyes, where the majority of photoreceptors have green-absorbing rhodopsins and blue-absorbing metarhodopsins, natural illuminants are predicted to create metarhodopsin levels greater than 60% at high intensities. However, fast metarhodopsin decay and rhodopsin regeneration also play an important role in controlling metarhodopsin in green receptors, resulting in a high rhodopsin content at low light intensities and a reduced overall visual pigment content in bright light. A simple model for the visual pigment–arrestin cycle is used to illustrate the dependence of the visual pigment population states on light intensity, arrestin levels and pigment turnover.     

Phosphorylated rhodopsin and heparin induce similar conformational changes in arrestin

Photoactivated rhodopsin is quenched upon its phosphorylation in the reaction catalyzed by rhodopsin kinase and the subsequent binding of a regulatory protein, arrestin. We have found that heparin and other polyanions compete with photoactivated, phosphorylated rhodopsin to bind arrestin (48-kDa protein, S-antigen). This is shown (a) by the suppression of stabilized metarhodopsin II; (b) by changes in the digestion of arrestin in the presence of heparin; and (c) by the restoration of arrestin-quenched phosphodiesterase activity. When bound to arrestin, heparin also mimics phosphorylated rhodopsin by similarly exposing arrestin to limited proteolysis. We conclude that heparin and rhodopsin have similar means of binding to arrestin, and we propose a cationic region of arrestin (beginning with Lys163 of the bovine sequence) as the interaction site. In agreement with previous kinetic data we interpret the results in terms of a binding conformation of arrestin which is stabilized by rhodopsin or heparin and is open to proteolytic attack.     

Purification and characterization of bovine cone arrestin (cArr)

To elucidate the quenching mechanism of phototransduction in vertebrate cone photoreceptors, a cDNA clone encoding cone specific arrestin (cArr) was isolated from a bovine retinal cDNA library using a human cArr cDNA probe. Affinity-purified anti-peptide antibody specific to cArr was prepared. Immunohistochemical staining displayed specific labeling of cArr in cone photoreceptors and immunoblotting identified a 46 kDa protein band. We purified cArr from bovine retinas by sequential column chromatography using DEAE-cellulose, gel filtration and mono Q columns. Binding studies revealed no binding of cArr to rhodopsin regardless of whether it was bleached and/or phosphorylated. cArr also failed to bind to heparin-Sepharose under conditions which rod arrestin (rArr) bound to the column. The present data suggest that cArr may play a role in the quenching of phototransduction in cone photoreceptors and that its activity therein is different to that of rArr.     

Progressive reduction of its expression in rods reveals two pools of arrestin-1 in the outer segment with different roles in photoresponse recovery

Light-induced rhodopsin signaling is turned off with sub-second kinetics by rhodopsin phosphorylation followed by arrestin-1 binding. To test the availability of the arrestin-1 pool in dark-adapted outer segment (OS) for rhodopsin shutoff, we measured photoresponse recovery rates of mice with arrestin-1 content in the OS of 2.5%, 5%, 60%, and 100% of wild type (WT) level by two-flash ERG with the first (desensitizing) flash at 160, 400, 1000, and 2500 photons/rod. The time of half recovery (t(half)) in WT retinas increases with the intensity of the initial flash, becoming ～2.5-fold longer upon activation of 2500 than after 160 rhodopsins/rod. Mice with 60% and even 5% of WT arrestin-1 level recovered at WT rates. In contrast, the mice with 2.5% of WT arrestin-1 had a dramatically slower recovery than the other three lines, with the t(half) increasing ～28 fold between 160 and 2500 rhodopsins/rod. Even after the dimmest flash, the rate of recovery of rods with 2.5% of normal arrestin-1 was two times slower than in other lines, indicating that arrestin-1 level in the OS between 100% and 5% of WT is sufficient for rapid recovery, whereas with lower arrestin-1 the rate of recovery dramatically decreases with increased light intensity. Thus, the OS has two distinct pools of arrestin-1: cytoplasmic and a separate pool comprising ～2.5% that is not immediately available for rhodopsin quenching. The observed delay suggests that this pool is localized at the periphery, so that its diffusion across the OS rate-limits the recovery. The line with very low arrestin-1 expression is the first where rhodopsin inactivation was made rate-limiting by arrestin manipulation.     

The putative brain photoperiodic photoreceptors in the vetch aphid,Megoura viciae

In an attempt to identify the brain photoreceptors that mediate the photoperiodic response of the vetch aphid, Megoura viciae, we utilised immunocytochemical techniques and employed 20 antibodies directed against invertebrate and vertebrate opsins and phototransduction proteins. A sub-set of these antibodies (to Drosophila rhodopsin 1: RH1-1; vertebrate cone opsins: COS-1; CERN-874; CERN-933; vertebrate rod opsin: CERN-901; vertebrate arrestin: AB-Arr; vertebrate transducin+arrestin+rhodopsin kinase+cGMP phosphodiesterase: CERN-911; and vertebrate cellular retinoid binding protein: CRALBP) consistently labelled an anterior ventral neuropile region of the protocerebrum. These anatomical findings, coupled with previous localised illumination and micro-lesion studies, provide strong evidence that this region of the aphid brain houses the photoperiodic photoreceptors. The present study also confirms that the medial (Group I) neurosecretory cells are not the photoperiodic photoreceptors.     

Localization of caveolin-1 and c-src in mature and differentiating photoreceptors: raft proteins co-distribute with rhodopsin during development

Numerous biochemical and morphological studies have provided insight into the distribution pattern of caveolin-1 and the presence of membrane rafts in the vertebrate retina. To date however, studies have not addressed the localization profile of raft specific proteins during development. Therefore the purpose of our studies was to follow the localization pattern of caveolin-1, phospho-caveolin-1 and c-src in the developing retina and compare it to that observed in adults. Specific antibodies were used to visualize the distribution of caveolin-1, c-src, a kinase phosphorylating caveolin-1, and phospho-caveolin-1. The labeling pattern of this scaffolded complex was compared to those of rhodopsin and rhodopsin kinase. Samples were analyzed at various time points during postnatal development and compared to adult retinas. The immunocytochemical studies were complemented with immunoblots and immunoprecipitation studies. In the mature retina caveolin-1 and c-src localized mainly to the cell body and IS of photoreceptors, with only very weakly labeled OS. In contrast, phospho-caveolin-1 was only detectable in the OS of photoreceptors. During development we followed the expression and distribution profile of these proteins in a temporal sequence with special attention to the period when OS formation is most robust. Double labeling immunocytochemistry and immunoprecipitation showed rhodopsin to colocalize and co-immunoprecipitate with caveolin-1 and c-src. Individual punctate structures between the outer limiting membrane and the outer plexiform layer were seen at P10 to be labeled by both rhodopsin and caveolin-1 as well as by rhodopsin and c-src, respectively. These studies suggest that membrane raft specific proteins are co-distributed during development, thereby pointing to a role for such complexes in OS formation. In addition, the presence of small punctate structures containing caveolin-1, c-src and rhodopsin raise the possibility that these proteins may transport together to OS during development and that caveolin-1 exists predominantly in a phosphorylated form in the OS.     

Frog rod outer segments with attached inner segment ellipsoids as an in vitro model for photoreceptors on the retina

Purified suspensions of frog rod outer segments still attached to the mitochondria-rich inner segment portion of the receptor cell (OS-IS) can be obtained in quantities (0.1 mg/retina) sufficient for chemical analysis. In oxygenated glucose-bicarbonate Ringer's medium with added Percoll, they display normal dark currents, light sensitivity, and photocurrent kinetics for several hours. Two millimolar cytoplasmic levels of ATP and GTP are maintained, fivefold higher than in isolated OS. The levels are not altered by abolition of the dark current with ouabain. Nucleoside triphosphates are more effectively buffered than in isolated OS, and their levels remain constant during changes in external calcium levels. 32Pi is incorporated into endogenous ATP and GTP pools twice as efficiently as in isolated OS, and is used in the phosphorylation of rhodopsin. OS-IS take up and release 45Ca++ by Na+-, Ca++-, and IBMX-sensitive mechanisms. Illumination causes release of 45Ca++, which confirms retinal studies by other groups using Ca++-sensitive electrodes. Thus, OS-IS suspensions model the behavior of photoreceptors still attached to the living retina. Their availability permits the simultaneous assay and correlation of electrophysiological and chemical changes occurring during excitation and adaptation.     

Protein Kinase C Activity and Light Sensitivity of Single Amphibian Rods

Biochemical experiments by others have indicated that protein kinase C activity is present in the rod outer segment, with potential or demonstrated targets including rhodopsin, transducin, cGMP-phosphodiesterase (PDE), guanylate cyclase, and arrestin, all of which are components of the phototransduction cascade. In particular, PKC phosphorylations of rhodopsin and the inhibitory subunit of PDE (PDE _γ) have been studied in some detail, and suggested to have roles in downregulating the sensitivity of rod photoreceptors to light during illumination. We have examined this question under physiological conditions by recording from a single, dissociated salamander rod with a suction pipette while exposing its outer segment to the PKC activators phorbol-12-myristate,13-acetate (PMA) or phorbol-12,13-dibutyrate (PDBu), or to the PKC-inhibitor GF109203X. No significant effect of any of these agents on rod sensitivity was detected, whether in the absence or presence of a background light, or after a low bleach. These results suggest that PKC probably does not produce any acute downregulation of rod sensitivity as a mechanism of light adaptation, at least for isolated amphibian rods.     

Subspecies of arrestin from bovine retina. Equal functional binding to photoexcited rhodopsin but various isoelectric focusing phenotypes in individuals

Arrestin (also named 48-kDa protein or S-antigen) binds to photoexcited and phosphorylated rhodopsin and thereby prevents activation of cGMP phosphodiesterase (EC 3.1.4.35) by transducin in retinal rods. We report here that retinal arrestin consists of several subspecies (isoelectric points between pH 5.5-6.2), which can be separated by FPLC anion-exchange chromatography and by FPLC chromatofocusing resulting in highly enriched individual subspecies. The entire heterogeneity pattern of arrestin is present in rod outer segments, independently of whether arrestin orginated from the outer or mostly from the inner segment of rod cells. The different subspecies show a similar binding behavior to photoexcited rhodopsin phosphorylated to various degrees and they quench the cGMP phosphodiesterase activity equally well. In the presence of rod outer segment membranes, arrestin is phosphorylated light-dependently by protein kinase C (0.2 mol phosphate/mol arrestin). This implies that the heterogeneity of arrestin is not primarily due to phosphorylation. Arrestin from different individuals exists as four isoelectric focusing patterns which occur with remarkably different frequencies in calf and cattle. The complexity of the IEF pattern does not increase with aging. Distinct subspecies of arrestin may reflect differences in their primary structure, or may result from differentially regulated post-translational modifications in individuals.     

Transducin activation state controls its light-dependent translocation in rod photoreceptors

Light-dependent redistribution of transducin between the rod outer segments (OS) and other photoreceptor compartments including the inner segments (IS) and synaptic terminals (ST) is recognized as a critical contributing factor to light and dark adaptation. The mechanisms of light-induced transducin translocation to the IS/ST and its return to the OS during dark adaptation are not well understood. We have probed these mechanisms by examining light-dependent localizations of the transducin-alpha subunit (Gtalpha)in mice lacking the photoreceptor GAP-protein RGS9, or expressing the GTPase-deficient mutant GtalphaQ200L. An illumination threshold for the Gtalpha movement out of the OS is lower in the RGS9 knockout mice, indicating that the fast inactivation of transducin in the wild-type mice limits its translocation to the IS/ST. Transgenic GtalphaQ200L mice have significantly diminished levels of proteins involved in cGMP metabolism in rods, most notably the PDE6 catalytic subunits, and severely reduced sensitivity to light. Similarly to the native Gtalpha, the GtalphaQ200L mutant is localized to the IS/ST compartment in light-adapted transgenic mice. However, the return of GtalphaQ200L to the OS during dark adaptation is markedly slower than normal. Thus, the light-dependent translocations of transducin are controlled by the GTP-hydrolysis on Gtalpha, and apparently, do not require Gtalpha interaction with RGS9 and PDE6.     

Light-dependent association of Src with photoreceptor rod outer segment membrane proteins in vivo.

In vivo light exposure results in tyrosine phosphorylation of several rod outer segment (ROS) proteins (Ghalayini, A. J., Guo, X. X., Koutz, C. A, and Anderson, R. E. (1998) Exp. Eye Res. 66, 817-821). We now report the presence of Src in ROS and its increased association with bleached ROS membranes. Immunoprecipitation with anti-phosphotyrosine revealed that tyrosine kinase activity recovered from light-adapted ROS membranes was twice that recovered from dark-adapted ROS. Other experiments revealed the presence of both bleached rhodopsin and arrestin in immunoprecipitates of LROS, suggesting the formation of a multimeric complex containing Src, arrestin, and bleached rhodopsin. Additionally, when immobilized Src homology domains 2 and 3 (SH2 and SH3, respectively) were used to study the association of Src with ROS membranes, only bleached opsin and arrestin were found to associate with the SH2 domain of Src. These data strongly suggest that Src through its SH2 domain interacts with bleached rhodopsin and arrestin either directly or indirectly. Similar results were also obtained when dark-adapted and light-adapted retinas were used instead of ROS membranes. Our data strongly suggest that light exposure in vivo activates Src and promotes its association through its SH2 domain with a complex containing bleached rhodopsin and arrestin. A hypothesis for the functional significance of this phenomenon is presented.