G-protein-coupled Receptor Rhodopsin Regulates the Phosphorylation of Retinal Insulin Receptor^*

Insulin receptor (IR)² activation has been shown to rescue retinal neurons from apoptosis through a phosphoinositide 3-kinase (PI3K) cascade (1). Particular insights have come from observations that IR substrate-2 (principle substrate of the IR) knock-out mice lose up to 50% of their photoreceptors by 2 weeks of age, due to increased apoptosis (2). Very recently, we have shown that Ak2, another downstream target of the IR, is essential for photoreceptor survival and maintenance, since ablation of this protein in the retina resulted in stress-induced retinal degeneration (3). Specific deletion of bcl-xl (down-stream effector of Akt) from rod photoreceptor cells also results in stress-induced retinal degeneration (4). These studies clearly suggest that the IR pathway is important for photoreceptor survival and maintenance.

Many retinal degenerative diseases show an early loss of rod cells that is followed by loss of cone cells, and the pathological phenotype for this loss is apoptosis (5–7). The induction of cell death is a highly regulated process and can be suppressed by a variety of extracellular signals (8, 9). The ability of trophic factors to promote survival is mediated, at least in part, by PI3K (1, 10–12). Activation of PI3K is a critical step in signal transduction pathways triggered by a variety of extracellular stimuli (13). The lipid products of PI3K serve as second messengers to recruit specific phospholipid-binding proteins to the plasma membrane and control the activity and subcellular localization of a diverse array of signal transduction molecules (14).

In Drosophila, the IR serves an important function in guiding the development of retinal photoreceptor axons from the retina to the brain (15). The IR also influences the size and number of photoreceptors in Drosophila (16). In Caenorhabditis elegans, the IR regulates neuronal survival (17). The high degree of IR signaling homology among C. elegans, Drosophila, and humans suggests functional conservation of the IR in the mammalian retina. In humans, defects in IR signaling in the central nervous system are associated with Alzheimer disease (18–20). The lack of IR activation leads to neurodegeneration in the brain/neuron-specific IR knock-out mice (21).

The retina expresses PI3K, which is regulated through the light-induced tyrosine phosphorylation of the IR in vivo (22, 23). Light-induced activation of the retinal IR is independent of insulin secretion, and the light effect is localized to photoreceptor neurons (22). These results suggest that there exists a cross-talk between phototransduction and other signal transduction pathways. To dissect the stage of the phototransduction cascade that is coupled to the activation of the IR, we examined IR phosphorylation in three knock-out mouse models that are deficient in either phototransduction or the ability to form functional rhodopsin. In the present study, we demonstrate that light-induced tyrosine phosphorylation of the IR requires the photobleaching of rhodopsin but not transducin signaling. We also show that the light-dependent IR activation occurs in rod, but not in cone, photoreceptors. These studies suggest that rhodopsin can cross-talk with other signaling pathways in addition to the classical phototransduction cascade.

EXPERIMENTAL PROCEDURES

Materials—Polyclonal anti-Glut1, anti-IRβ, and monoclonal anti-PY-99 antibodies were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Polyclonal anti-insulin/IGF-1-like growth factor-1 receptor (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) phosphospecific antibody was obtained from BIOSOURCE (Camarillo, CA). The monoclonal anti-opsin antibody (Rho 4D2) was a gift from Dr. Robert Molday (University of British Columbia). Human insulin R (rDNA origin) was obtained from Lilly. [γ-³²P]ATP was from PerkinElmer Life Sciences. Anti-GST antibody was from Amersham Biosciences. Echelon Research laboratories Inc. (Salt Lake City, UT) provided d-myo-phosphatidylinositol-4,5-bisphosphate (PI-4,5-P₂). Wheat germ agglutinin (WGA)-Sepharose 4B beads were from Vector Laboratories (Burlingame, CA). All other reagents were of analytical grade from Sigma.

Animals—All animal work was in strict accordance with Ref. 78 and the Association for Research in Vision and Ophthalmology resolution on the use of animals in vision research. All protocols were approved by the institutional animal care and use committee of the University of Oklahoma Health Sciences Center and the Dean McGee Eye Institute. Transducin α subunit knock-out (Trα^-/-) mice (24) and opsin knock-out mice (25) were derived on a BALB/cx129/SvJ background at Tufts University. Mice lacking retinal pigment epithelium 65 (RPE65^-/-) were derived at the National Institutes of Health (Bethesda, MD) (26). Wild-type controls were obtained from breeding pairs established with C57BL/6-DBA F1s. A breeding colony of albino Sprague-Dawley rats is maintained in our vivarium in cyclic light (5 lux; 12 h on/12 h off). Experiments were carried out on both male and female rats (150–200 g).

Retinal Organ Cultures—Retinas were removed from Sprague-Dawley albino rats that were born and raised in dim cyclic light (5 lux; 12 h on/12 h off) and incubated at 37 °C in Dulbecco's modified Eagle's medium (Invitrogen) in the presence or absence of 1 μm insulin for 30 min. At the indicated times, retinas were snap-frozen in liquid nitrogen and stored at -80 °C until analyzed. The retinas were lysed in lysis buffer (1% Nonidet P-40, 20 mm HEPES (pH 7.4), and 2 mm EDTA) containing phosphatase inhibitors (100 mm NaF, 10 mm Na₄P₂O₇, 1 mm NaVO₃, and 1 mm molybdate) and protease inhibitors (10 μm leupeptin, 10 μg/ml aprotinin, and 1 mm PMSF) and kept on ice for 10 min followed by centrifugation at 4 °C for 20 min.

PI3K Assay—Enzyme assays were carried out as previously described (27). Briefly, assays were performed directly on IRβ immunoprecipitates of retinal lysates or rod outer segments (ROS) prepared from light- and dark-adapted retinas in 50 μl of reaction mixture containing 0.2 mg/ml PI-4,5-P₂, 50 μm ATP, 10 μCi of [γ-³²P]ATP, 5 mm MgCl₂, and 10 mm HEPES buffer (pH 7.5). The reactions were carried out for 30 min at room temperature and stopped by the addition of 100 μl of 1 n HCl followed by 200 μl of chloroform/methanol (1/1, v/v). Lipids were extracted and resolved on oxalate-coated TLC plates (silica gel 60) with a solvent system of 2-propanol, 2 m acetic acid (65/35, v/v). The plates were coated in 1% (w/v) potassium oxalate in 50% (v/v) methanol and then baked in an oven at 100 °C for 1 h prior to use. TLC plates were exposed to x-ray film overnight at -70 °C, and radioactive lipids were scraped and quantified by liquid scintillation counting.

GST Pull-down Experiments—Pull-down experiments were carried out as described (28) using 5 μg of GST-fusion proteins that had been adsorbed onto GST-Sepharose 4B matrix. Retina lysates (insulin-treated or untreated in vitro) or light- and dark-adapted retina lysates were incubated with GST/GST-p85 (N-SH2) fusion proteins (29) at 4 °C for 1.5 h with continuous mixing. The Sepharose beads were washed three times in 500 μl of wash buffer (50 mm HEPES (pH 7.4) containing 118 mm NaCl, 100 mm NaF, 2 mm NaVO₃, 0.1% (w/v) SDS, and 1% (v/v) Triton X-100) and centrifuged at 5,000 rpm for 30–60 s at 4 °C. Bound proteins were eluted by boiling in 2× SDS sample buffer 5 min prior to 10% SDS-PAGE. After SDS-PAGE, the gels were subjected to Western blot analysis with anti-IRβ antibody. To ensure an equal amount of fusion protein in each experiment, the blot was stripped and reprobed with anti-GST antibody.

Preparation of Osmotically Intact Rod Outer Segment Disks— Osmotically intact rod outer segment disks were prepared by Ficoll flotation (30). The ROS were prepared from frozen bovine retinas according to the method described (31). The ROS pellet was resuspended in 30 ml of Ficoll in distilled water, and the suspension was kept at 4% under nitrogen for at least 2 h to allow the ROS plasma membrane to burst. This suspension of ROS was divided between two small SW-28 centrifuge tubes and layered with cold water. After 2 h of centrifugation in a SW-20 rotor at 25,000 rpm, the intact disks were collected from 5% Ficoll water interface. The bottom pellet contained the ROS plasma membrane-enriched fraction.

Preparation of Sealed Bovine Rod Cell Outer Segments—ROS were prepared from bovine retinas on a continuous sucrose gradient (25–50%, w/v) as described (32). Dark-adapted ROS were prepared under dim red light from dark-adapted retinas. Bleached ROS were prepared by exposing dark-adapted retinas to room light (ceiling fluorescent bulbs, 700 lux) for 30 min at room temperature before preparation of ROS. Protein quantification was performed with BCA reagent (Pierce) following the manufacturer's instructions.

Preparation of Rat Rod Outer Segments—ROS were prepared from rat retinas using a discontinuous sucrose gradient as previously described (22). Retinas were homogenized in 4.0 ml of ice-cold 47% sucrose solution containing 100 mm NaCl, 1 mm EDTA, 1 mm NaVO₃, 1 mm PMSF, and 10 mm Tris-HCl (pH 7.4) (buffer A). Retinal homogenates were transferred to 15-ml centrifuge tubes and sequentially overlaid with 3.0 ml of 42%, 3.0 ml of 37%, and 4.0 ml of 32% sucrose dissolved in buffer A. The gradients were spun at 82,000 × g for 1 h at 4°C. The 32/37% interfacial sucrose band containing ROS membranes was harvested and diluted with 10 mm Tris-HCl (pH 7.4) containing 100 mm NaCl and 1 mm EDTA and centrifuged at 27,000 × g for 30 min. The ROS pellets were resuspended in 10 mm Tris-HCl (pH 7.4) containing 100 mm NaCl and 1 mm EDTA and stored at -20 °C. All protein concentrations were determined by the BCA reagent following the manufacturer's instructions.

Immunoprecipitation—ROS membranes were solubilized for 30 min at 4 °C in a lysis buffer (1% Nonidet P-40, 20 mm HEPES (pH 7.4), and 2 mm EDTA) containing phosphatase inhibitors (100 mm NaF, 10 mm Na₄P₂O₇, 1 mm NaVO₃, and 1 mm molybdate) and protease inhibitors (10 μm leupeptin, 10 μg/ml aprotinin, and 1 mm PMSF) and kept on ice for 10 min. Insoluble material was removed by centrifugation at 17,000 × g for 20 min at 4 °C. Either ROS or retina lysates were precleared by incubation with 40 μl of protein A-Sepharose for 1 h at 4°C with mixing. The supernatant was incubated with anti-IRβ (4 μg) antibody overnight at 4 °C and subsequently with 40 μl of protein A-Sepharose for 2 h at 4 °C. Following centrifugation at 14,000 rpm for 1 min, immune complexes were washed three times with wash buffer (50 mm HEPES (pH 7.4) containing 118 mm NaCl, 100 mm NaF, 2 m NaVO₃, 0.1% (w/v) SDS, and 1% (v/v) Triton X-100). Immunoprecipitates were subjected to either immunoblot analysis with anti-PY-99 (1:1000), anti-IR (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) (1:1000), anti-opsin (1:10,000), or anti-IRβ (1:1000) antibodies, or measured directly for PI3K activity.

SDS-PAGE and Western Blotting—Proteins were resolved by 10% SDS-PAGE and transferred onto nitrocellulose membranes. The blots were washed two times for 10 min with TTBS (20 mm Tris-HCl (pH 7.4), 100 mm NaCl, and 0.1% Tween 20) and blocked with either 5% bovine serum albumin or nonfat dry milk powder (Bio-Rad) in TTBS for 1 h at room temperature. Blots were then incubated with anti-IRβ (1:1000), anti-PY-99 (1:1000), anti-Glut1 (1:1000), anti-IR (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) (1:1000), anti-opsin (1:10,000), or anti-GST (1:5000) antibodies overnight at 4 °C. Following primary antibody incubations, immunoblots were incubated with HRP-linked secondary antibodies (either anti-rabbit or anti-mouse) and developed by ECL according to the manufacturer's instructions. Densitometric analysis of immunoblots where indicated was performed using Kodak Image Station 4000R (Eastman Kodak Co.) in the linear range of detection. Absolute values were then normalized to total protein as indicated in the figure legends.

Statistical Methods—The data were expressed as the mean ± S.D. and compared by Student's t test for unpaired data. The critical level of significance was set at p < 0.05.

RESULTS

IRs Are Localized to Plasma Membrane—Bovine crude rod outer segments were subjected to Ficoll gradient centrifugation to isolate ROS disks (30). ROS, plasma membrane, and isolated disks were subjected to Western blot analysis with anti-IRβ, anti-Glut1, and anti-opsin antibodies. The results indicate that IR, Glut1, and opsin immunoreactivity was present in the ROS (Fig. 1). IR and the plasma membrane marker Glut1 immunoreactivity was enriched in the plasma membrane fraction of the ROS (Fig. 1). Opsin blots show the enrichment of opsin in the disk membranes (Fig. 1). These results suggest that IRs are localized to the plasma membrane.

Light-induced Activation of PI3K in Isolated Rod Outer Segments—We examined the light activation of the IR by immunoprecipitating it and measuring the associated PI3K activity (22). ROS prepared from dark-adapted bovine retinas were solubilized with 1% Triton X-100 followed by incubation with WGA-Sepharose 4B matrix, since WGA is known to interact with IRs (33). The unbound proteins were washed away and PI3K activity associated with WGA was measured. The results indicate a higher PI3K activity associated with bleached ROS (BROS) compared with dark-adapted ROS (DROS) (Fig. 2A), suggesting that light could activate the IR in vitro.

Insulin-induced Activation of PI3K in Isolated Rod Outer Segments—To test the effect of insulin on PI3K activity, bovine ROS isolated on continuous sucrose density gradients were incubated either in the presence or absence of 1 μm insulin for 30 min at room temperature. Following incubation, bovine ROS were solubilized with 1% Triton X-100 and then immunoprecipitated with anti-IRβ antibody, followed by measurement of the PI3K activity associated with anti-IRβ immunoprecipitates. The results indicate increased PI3K activity associated with insulin-stimulated bovine ROS (Fig. 2B) and suggest that functional IR is present in ROS.

Light Activation of the Retinal IR in Vivo—To determine if light has an effect on PI3K activity and phosphorylation of the IR, rats were dark-adapted overnight, and half were exposed to normal room light for 30 min. Retinas were quickly removed and homogenized, the lysates were subjected to ROS preparation or immunoprecipitated with anti-IRβ antibody, and PI3K activity was measured. ROS membranes that were prepared on discontinuous sucrose gradients showed greater PI3K activity in light-adapted rats compared with dark-adapted rats (Fig. 3, A and B). When retinal homogenates were immunoprecipitated with anti-IRβ antibody, PI3K activity was higher in immunoprecipitates from light-adapted compared with from dark-adapted rat retinas (Fig. 3, C and D). Collectively, these experiments suggest that the PI3K activity associated with the ROS membranes is activated through the IR.

Activation of IR by Insulin in Organ Cultures—The organ culture system has been successfully used to study protein phosphorylation and provides access to the retina for the addition of exogenous modulators of cellular function (29). IR phosphorylation was confirmed in anti-IRβ immunoprecipitates of insulin-treated and untreated retinas by probing lysates with anti-phosphotyrosine (PY-99) antibody. Increased phosphorylation was observed on Western blots of anti-IRβ insulin-treated immunoprecipitates, demonstrating the phosphorylation of IRβ in response to insulin (Fig. 4A). To ensure equal amounts of IRβ in immunoprecipitates, PY-99 blots were stripped and reprobed with anti-IRβ antibody (Fig. 4B). Higher PI3K activities were observed in the anti-IRβ immunoprecipitates of insulin-stimulated retinas (Fig. 4, C and D). These results further suggest that retinal IRs are functional and are sensitive to both light (Fig. 2) and insulin (Fig. 3).

Binding of IRβ to the p85 (N-SH2) Domain of PI3K—The addition of insulin to the retinal organ cultures resulted in activation of the IR as measured by receptor phosphorylation in a GST pull-down assay (29). The GST p85 (N-SH2) domain specifically pulled down the phosphorylated form of the IR (22, 34). In the presence of insulin, the p85 (N-SH2) domain of PI3K could pull down IRβ as detected on the Western blots probed with anti-IRβ antibody (Fig. 4E). To ensure equal amounts of fusion proteins in each pull-down, IRβ blots were stripped and reprobed with anti-GST antibody (Fig. 4F). These results suggest that the p85 subunit of PI3K specifically recognizes the phosphorylated form of the IR.

The Visual Transduction Cascade and IR Phosphorylation—We examined IR phosphorylation in mice lacking transducin (Trα^-/-) to determine if phototransduction events are required for IR phosphorylation. Trα^-/- mice were dark-adapted overnight, and half were exposed to normal room light for 30 min. Retinal lysates from light- and dark-adapted Trα^-/- mice were subjected to GST pull-down assays using the p85 (N-SH2) domain followed by Western blot analysis with anti-IRβ antibody. The results indicate that Trα^-/- mice still exhibit a light-dependent phosphorylation of IR (Fig. 5A). The proteins were normalized based on the opsin content, which was used as an internal control (Fig. 5B). To ensure equal amounts of fusion in each pull-down, IRβ blots were stripped and reprobed with anti-GST antibody (Fig. 5C). These results suggest that the visual transduction cascade is not necessary for light-dependent IR phosphorylation.

Photobleaching of Rhodopsin and IR Phosphorylation—To confirm that the light-induced IR phosphorylation is signaled through rhodopsin, we examined the phosphorylation of the IR in retinas from retinal pigment epithelium protein (RPE65) knock-out mice that are deficient in 11-cis-retinal, the chromophore for rhodopsin (26). These animals have opsin in their ROS but do not have a bleachable rhodopsin due to the absence of the chromophore. RPE65^-/- mice were dark-adapted overnight, and half were exposed to normal room light for 30 min. Retinal lysates from light- and dark-adapted RPE65^-/- mice were subjected to GST pull-down assays using the p85 (N-SH2) domain, followed by Western blot analysis with anti-IRβ antibody. The results indicate the absence of light-induced IR phosphorylation in RPE65^-/- mice (Fig. 6A). IR was recovered from retinal lysates stimulated with insulin in vitro (Fig. 6A, Control). Total IR content was examined with anti-IRβ antibody (Fig. 6B). The proteins were normalized based on the opsin content, which was used as an internal control (Fig. 6C). To ensure equal amount of fusion in each pull-down, the IRβ blot was stripped and reprobed with anti-GST antibody (Fig. 6D). These results suggest that photobleaching of rhodopsin is necessary for IR phosphorylation.

Insulin Activation of IR Phosphorylation in RPE65^-/-Mice— RPE65^-/- mouse retinas were stimulated with 1 μm insulin in organ cultures, after which the retinas were lysed and subjected to GST pull-down assays using the p85 (N-SH2) domain followed by Western blot analysis with anti-IRβ antibody. The results indicate the IR can be phosphorylated in response to insulin stimulation in RPE65^-/- retinas (Fig. 7A). Total IR content was examined with anti-IRβ antibody (Fig. 7B). The proteins were normalized based on the opsin content, which was used as an internal control (Fig. 7C). To ensure an equal amount of fusion in each pull-down, the IRβ blot was stripped and reprobed with anti-GST antibody (Fig. 7D). These results indicate the existence of a light-mediated IR pathway in the retina that is different from the known insulin-mediated pathway in non-neuronal tissues. Collectively, these experiments show that photobleaching of rhodopsin is necessary for light-mediated phosphorylation of the IR.

Light Activation of Tyrosine Phosphorylation in the Catalytic Loop of the IR—The catalytic loops within the tyrosine kinase domain of the IR contain a three-tyrosine (Tyr¹¹⁵⁸, Tyr¹¹⁶², and Tyr¹¹⁶³) motif (35, 36). It is generally believed that autophosphorylation within the activation loop proceeds in a processive manner initiated at the second tyrosine (position 1162), followed by phosphorylation at the first tyrosine (position 1158), and finally the last (position 1163), upon which the IR becomes fully active (35, 36). To determine whether light activates IR in the same manner as insulin, we immunoprecipitated the IR (Fig. 8) from retinal lysates that were prepared from light- and dark-adapted mice followed by Western blot analysis with phosphospecific anti-IR (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) antibody. The results indicate an increased phosphorylation in the light-adapted retinas compared with dark-adapted retinas (Fig. 8A). The blot was stripped and reprobed with anti-IRβ antibody to ensure an equal amount of protein in each line (Fig. 8B). LROS and DROS were solubilized with 1% Nonidet P-40, and the IR was immunoprecipitated followed by Western blot analysis with anti-IR (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) antibody. The results indicate the tyrosine phosphorylation in LROS (Fig. 8C). The blot was stripped and reprobed with anti-IRβ antibody to ensure an equal amount of protein in each line (Fig. 8D). These results suggest that light activates the IR phosphorylation in the catalytic loop within the tyrosine kinase domain.

Activation of IR Signaling through Light Stimulation of Rods—To determine whether the activation of the IR is signaled through light stimulation of rods, we examined the phosphorylation of the IR in retina lysates from opsin^-/- mice. Although these mice have the full complement of rod photoreceptor cells, they lack rod outer segments and scotopic electroretinographic signal (25). However, retinas of opsin^-/- mice exhibit normal photopic electroretinographic responses reflecting the existence of functional cone transduction pathways (25, 37). Therefore, these mice afforded the ideal system to test the activation of IR signaling through light stimulation of rods and to determine if any contributions are made by cones. To that end, C57Bl/6 control, heterozygous opsin^+/-, and homozygous opsin^-/- mice were dark-adapted overnight; half were exposed to normal room light for 30 min, and the rest were kept in the dark. Retinal lysates from light- and dark-adapted mice from the three groups were subjected to immunoprecipitation with anti-IRβ antibody followed by Western blot analysis with anti-IR (Tyr(P)¹¹⁵⁸/Tyr(P)¹¹⁶²/Tyr(P)¹¹⁶³) antibody. The blots were stripped and reprobed with anti-IRβ antibody. Densities were calculated from the respective immunoblots, and the results are expressed as phospho-IR/total IR. In control mice, we observed a significant increase in the IR phosphorylation from light-adapted retinas compared with dark-adapted retinas (Fig. 9A). In heterozygous opsin^+/- mice, the IR phosphorylation was higher but not statistically significant in light-adapted retinas compared with dark-adapted retinas (Fig. 9B), whereas in homozygous opsin^-/- mice, the light-dependent activation of the IR was lost (Fig. 9C). These results suggest that the light-dependent activation of the IR may be regulated through the rod transduction pathway but not the cone transduction pathway.

DISCUSSION

The experiments reported here show that IR phosphorylation requires the photobleaching of rhodopsin but does not require transducin signaling. These results also suggest that components downstream of transducin (PDE, Ca²⁺, cGMP, etc.) are not required for IR phosphorylation. Consistent with this idea, it has been shown that there is no difference in dark [Ca²⁺] between wild type and Trα^-/- mice (38). The Rpe65^-/- retinas fail to photobleach due to the absence of 11-cis-retinal (26), and thus retinal IR phosphorylation by light does not occur in these animals. However, stimulation of RPE65^-/- retinas with insulin results in the phosphorylation of the IR. These experiments clearly show the existence of a light-mediated IR pathway in the retina, which is different from the known insulin-mediated pathway in nonneuronal tissues. Our results also suggest that a cross-talk exists between phototransduction and other signal transduction pathways. This cross-talk phenomenon has been shown for other G-protein-coupled receptors (GPCRs), and many tyrosine kinase cascades are regulated by GPCRs (39, 40). Examples include mitogen-activated protein kinase cascade, extracellular-regulated kinases, and stress-activated protein kinases (40). Further, the binding of PYK2, a non-receptor protein-tyrosine kinase, to N-terminal domain-interacting receptors (Nir) is activated by GPCRs (41). The Nir proteins are the human homologs of the Drosophila retinal degeneration B protein (RdgB), a protein implicated in the visual transduction pathway in flies (41). These earlier studies along with the present study clearly suggest that photobleaching of rhodopsin may activate more than one signaling pathway.

Several studies have shown that retinal ROS contain intrinsic tyrosine kinase(s) that can be activated by light (42) to phosphorylate several ROS proteins (43, 44). It has been shown that light exposure in vivo activates Src and promotes its association with a complex containing bleached rhodopsin and arrestin (45). Retinal BIT (brain immunoglobulin-like molecules with a tyrosine-based activation motif) protein has been shown to be tyrosine-phosphorylated in vitro in a light-dependent manner (46). Evidence also indicates that the small G-protein Rac-1 may be regulated by rhodopsin in both Drosophila (47) and vertebrates (48). The α-subunit of the heterotrimeric G-protein, G₁₁α, does not participate in visual transduction, but the opsin-G₁₁-mediated signaling pathway is important for photic entrainment of the chicken pineal circadian clock (49). These studies clearly suggest that signaling proteins not directly involved in phototransduction could be activated upon illumination. The molecular mechanism behind the light activation of signaling proteins other than those involved in phototransduction (50–58) is not known.

Although no GTP-binding protein can substitute for transducin-mediated phototransduction in vertebrate rods (24), other G-proteins, such as Gα_q and Gα₁₁, appear to be present (59) and may be coupled to rhodopsin for other functions (49). IR phosphorylation in other G-protein knock-out/double knock-out (Trα and Gα_q or Gα₁₁) mouse models would give information about the importance of other G-proteins in signaling the IR.

The light-dependent IR phosphorylation in Trα^-/- mice could also be due to cone transducin. We addressed this issue using opsin^-/- mice; these mice do not have rod function and are used to study pure cone function (25, 37). Opsin gene disruption has been shown to affect the rod morphology and development (25). At 15 days, opsin^-/- retinas contained a normal number of rod nuclei with inner segments, but outer segments failed to develop (25). By 30 days, 10–15% and at 90 days, over 90% of rod nuclei are lost in these mice (25). We carried out our experiments in 3-week-old mice whose retinas still showed the presence of photoreceptor-specific proteins, such as transducin and arrestin (data not shown). We clearly showed that a light-dependent IR activation in these mice is absent, whereas light-dependent activation in wild type and heterozygote mice from the same litters was clearly observed. The results are almost identical to our previous studies on RD1 mutant mice, which also failed to activate the IR in the absence of photoreceptors (22). If cone transducin signals are involved, we would expect to see some light-dependent activation of the IR. Since none was found in these retinas, we conclude that the phenomenon we are describing occurs in rod photoreceptors. However, it may be possible that the light-dependent activation of the IR also occurs in cones but that the signal cannot be measured in rodent retinas, which contain very small numbers of cones relative to rods. The IR phosphorylation appears to be higher in heterozygous opsin^+/- mice, but the activation is not statistically significant between light- and dark-adapted retinas. The opsin content in the heterozygous mice has ∼50% of the normal amount of rhodopsin, and these mice are ∼50% less sensitive to light (25). It is tempting to speculate that full complementation of the IR activation might require >50% opsin.

The molecular mechanism behind the light-induced activation of the IR is not known. In this study, we observed the recognition of light-induced autophosphorylation/phosphorylation of IR by phosphospecific IR antibody, which specially recognizes the tyrosine phosphorylation in the catalytic loop of IR at tyrosines 1158, 1162, and 1163 (35, 36). These results suggest that light activates the tyrosine phosphorylation in the catalytic loop of IR.

It appears from our study that two types of IR activities exist in the retina, one activated through the photobleaching of rhodopsin and the other activated through ligands, such as insulin or IGF1 (60, 61). It has been shown previously that basal IR kinase activity in the retina was significantly greater than that of liver and remained constant between freely fed and fasted rats, suggesting that IR activation was not regulated by circulating insulin (60, 62). We have previously reported that light-induced tyrosine phosphorylation of the retinal IR is independent of insulin secretion (22). Our studies showing the localization of IR to the plasma membrane suggest that there must be soluble factor(s) connecting photobleaching of rhodopsin to the phosphorylation status of IR.

Consistent with this hypothesis, we have identified Grb14 (growth factor receptor-bound protein 14) as an IR-interacting protein (63, 64), and its intracellular localization is light-dependent (65). The Grb14 is a known inhibitor of protein-tyrosine phosphatase, PTP1b (66–69). We have also recently reported the expression of PTP1b in rod outer segments (64), and its levels are unchanged under light and dark adaptation.³ Based on these studies, we propose a working model (Fig. 10) in which the retinal IR autophosphorylates constitutively (60). In dark-adapted ROS, the IR is maintained in an inactive state by rapid dephosphorylation by PTP1b. Light activation of rhodopsin initiates the localization of Grb14 to ROS membranes, where it prevents the IR dephosphorylation by PTP1b. Light-activated IR subsequently associates with PI3K, a cell survival factor, activating the downstream survival pathway.

The IR pathway is known to be implicated both in neuroprotection of the retina (70) and in diabetic retinopathy (61). The functional consequences of light-dependent activation of the IR in photoreceptor cells are unknown. Activation of the IR signaling pathway has been shown to have complex physiological roles in a variety of cell types (71). In rods, IR activity is regulated by light, which suggests that its function could be related to rod-specific activities that are driven by light, such as shedding of photoreceptor tips (72), biogenesis of new ROS membranes through the addition of newly synthesized membranes at the base of the ROS (73), or light adaptation (74). Light activation could also initiate neuroprotective pathways (75–77). Further studies are required to understand the role of light-induced activation of IR in photoreceptor functions.