Light Threshold–Controlled Cone α-Transducin Translocation

Abstract

purpose. Light-induced translocation of rod α-transducin (rTα, GNAT1) has been recognized as one of the mechanisms for light adaptation in rods. However, cone α-transducin (cTα, GNAT2) has not been shown to have such light-dependent redistribution. To investigate potential reasons for the restriction of cTα to the cone outer segment, the authors established a transient transgenic strategy to express cone Tα within rod photoreceptor cells, and the location of the cone Tα within rods and cones was examined under different light conditions.

methods. Vector DNA that expresses cTα and green fluorescent protein (GFP) bicistronically under control of the cytomegalovirus (CMV) promoter was injected subretinally into the eyes of neonatal rats, and this was followed by electroporation. The localization of cTα in rods and cones under different light conditions was determined by immunofluorescent techniques.

results. Injection of the cDNA constructs resulted in the successful transient transfection of retinal cells. When cTα was exogenously expressed in rods, its localization paralleled that of endogenous rTα under light and dark conditions. Further experiments, with higher intensity light (7000 lux), demonstrated that endogenous cTα can also translocate in cone photoreceptor cells to the same extent it does in rods under 600 lux light.

conclusions. The authors successfully established an in vivo transient retinal transfection model. The demonstration of cTα translocation in rods indicates cTα is not inherently prevented from translocating. The novel observation of cTα translocation under high-intensity light suggests a light threshold regulates the redistribution of cTα possibly as a protective response against very bright light.

Vertebrate rod and cone photoreceptor cells have unique morphologic features that contribute to their detection of light. Although rods can detect light at the single photon level, they saturate and are inactivated at relatively low intensities of light. Cones are less sensitive and must adapt to a background illumination that can vary by more than 10 orders of magnitude ¹ and still be functional. Besides using different G-protein–coupled receptors and chromophores, rod and cone photoreceptor cells adapt to background light by a variety of molecular mechanisms, ^{2 3 4} which results in a light-dependent decrease in photoreceptor Ca²⁺ concentration that causes decreased phototransduction efficiency. A novel mechanism of adaptation of rods to light was proposed ^{5 6 7 8 9 10 11 12 13} whereby a massive translocation of some of the phototransduction signaling proteins occurs between the inner and outer segments of photoreceptor cells. 

Rod transducin, the second signal protein within the phototransduction pathway, was one of the first proteins discovered to undergo light-dependent translocation. Rod α-transducin (rTα) is located in the rod outer segment (OS) in the dark but redistributes and compartmentalizes into the inner segment (IS), cell bodies, and synaptic terminal (ST) on exposure to light. However, cone α-transducin (cTα) is compartmentalized in the cone outer segment and does not translocate under the same lighting conditions (600 lux) that are sufficient for complete translocation of rTα. ^{10 14} This inability to translocate can be attributed to the amino acid sequence and structural features of cTα or to the differences in the cellular components within cones and rods. To address this, we used in vivo transient transfection to generate transgenic rats that express cTα in rods. Our data demonstrate that in the absence of light, cTα compartmentalizes with rTα in the rod outer segment. However, in a light environment (600 lux), cTα does translocate to the inner compartments of the rods. Additional experiments led to the discovery that at very high intensity of light (7000 lux), endogenous cTα is no longer confined to the cone outer segment but also redistributes throughout the cone compartments. The similarity between the translocation of exogenous cTα in rods and endogenous cTα in cones indicates a common mechanism exits in rods and cones to regulate the translocation. Because cTα translocation occurs only in a high-intensity light environment, a light threshold is involved in the regulation of its translocation. The decrease in cTα in the cone outer segment also suggests that cones can modulate phototransduction in response to very bright light and that the function of this translocation is to provide protection for cones against light-induced damage. 

Materials and Methods

Animals

Pregnant Sprague–Dawley (SD) albino rats were bred and maintained in a regular cyclic light environment (12 hours on/12 hours off). The neonatal pups were injected subretinally and underwent electroporation between postnatal day (P)0 and P3 and were maintained for 3 to 4 weeks. Animals were cared for and handled in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and with the Institutional Animal Care and Use Committee (IACUC)–approved animal use protocols, which comply with the University of Oklahoma Faculty of Medicine guidelines for the use of animals in research. For light adaptation, SD rats (3–4 weeks old) were dark adapted for at least 12 hours before exposure to 600 lux or 7000 lux fluorescent white light for 1 hour or 10 minutes, respectively. For dark adaptation, rats were maintained in complete darkness for at least 12 hours before the experiment, and all procedures were performed under infrared illumination. 

DNA Constructs

The cDNA of mouse cone α-transducin (mcTα) was amplified from a λgt11 library ¹⁵ and was cloned into pEGFP-IRES2 using EcoRI. The primers were 5′-GAATTCAAATGGGGAGTGGCAT-3′ and 5′-GAATTCAGCATTAAAAGAGCCCACA-3′. The control vector pSIREN-RetroQ-Zsgreen was used to establish the in vivo transfection technique and to visualize the cell populations expressing Zoanthus sp green fluorescent protein (ZsGreen) under control of the cytomegalovirus (CMV) promoter. 

Western Blot

HEK-293 cells were maintained at 37°C with 5% CO₂ in Dulbecco modified Eagle medium (DMEM; Sigma, St. Louis, MO) supplemented with 3% calf serum, 4 mM l-glutamine, and 100 U/mL penicillin-streptomycin. Transfection followed the FuGene 6 (Roche Molecular Biochemicals, Indianapolis, IN) user protocol. The cells were harvested in the lysis buffer (62.5 mM Tris, pH 7.4) and were sonicated three times for 10 seconds each, followed by centrifugation in an Eppendorf microcentrifuge for 10 minutes at maximum speed. Soluble fractions of transfected and nontransfected cells were loaded onto SDS-PAGE. The cTα was identified using rabbit antiserum sc-390 (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA). 

Subretinal Injection and In Vivo Electroporation

The method used was originally described by Matsuda and Cepko. ¹⁶ Briefly, the neonatal rat pups were “cold” anesthetized by placement on ice. One microliter of a DNA solution (5 μg/μL) in H₂O was injected into the subretinal space of the right eye of each pup with the use of a Hamilton syringe with a 33-gauge, blunt-end needle inserted through an incision first made with a 30-gauge, bevel-point needle. The pups were then electroporated with tweezer-type electrodes (BTX, San Diego, CA) over the eye regions with five 100-V square pulses of 50-ms duration at 950-ms intervals (ECM830; BTX). The same amount of H₂O was injected into control pups. Each experiment was performed three times and involved the injection of at least 10 eyes; each figure is representative of the data obtained. 

Immunohistochemistry

Rats were killed by cervical dislocation, and the eyes were enucleated and fixed in 4% paraformaldehyde for 24 hours. After the cornea and lens were removed, the eyecups were further fixed overnight. Subsequently, the eyecups were cryoprotected with increasing concentrations of sucrose (10%, 20%, 30%) in PBS. Eyecups were embedded in a mold filled with optimal cutting temperature (OCT) compound, which was frozen after it was put over the liquid nitrogen bath. Sections (20 μm) were cut with a cryostat (CM 3050C; Leica, Wetzlar, Germany) and mounted on glass slides. 

Sections were blocked with 0.1 M Tris, pH 7.4, and 5% nonfat powdered milk for 2 hours at room temperature and were probed with sc-389 (rabbit anti–rod-Tα, 1:1000; Santa Cruz Biotechnology), sc-390 (rabbit anti–cone-Tα, 1:1000; Santa Cruz Biotechnology), or B-1075 (biotinylated peanut agglutinin, 10 μg; Vector Laboratories) at 4°C for 24 hours. Goat anti–rabbit antibody conjugated with dye (Alexa 594 or 488, 1:1000; Molecular Probes, Eugene, OR) and streptavidin conjugated with dye (Alexa 594, 1:1000; Molecular Probes) were used as secondary antibodies. Fluorescent signals were visualized with a 60× water immersion or a 100× oil immersion objective lens on a confocal laser scanning microscope (IX81-FV500; Olympus, Melville, NY) and were analyzed by software (FluoView; Olympus). 

Results

Establishment of In Vivo Transient Transfection

Our previous data indicated that cTα remains compartmentalized in the cone outer segment under the same lighting conditions that are sufficient for rTα to translocate from the rod outer segment to the rod inner segment. ¹⁰ To determine whether the inability of cTα to translocate is imposed by the cone intracellular environment or is inherent in its amino acid sequence, we set out to express cTα in rods. To initially test the in vivo transfection technique, ¹⁶ the pSIREN-RetroQ-ZsGreen vector (1 μL, 5 μg/μL) was injected subretinally into P0 to P3 rat pups, followed by five square-wave pulses generated through the tweezer-type electrodes over the eyelids. The pups were then kept with the mother until P21. The expression of ZsGreen, driven by the CMV promoter, can be detected by fluorescence microscopy using an excitation of 496 nm and an emission of 506 nm. Green fluorescence was easily visualized at P14, when the rats opened their eyes, until P60, the longest time tested. Representative images (Fig. 1)were taken from rats at P21. Figures 1a and 1bare from living animals that had been anesthetized and whose pupils were dilated with 10% phenylephrine HCl. Compared with the control eye that had no injection, the eye that was subretinally injected with pSIREN-RetroQ-ZsGreen showed bright green fluorescent signals. Figures 1c 1d to 1ewere from the eyecups that had been fixed with 4% paraformaldehyde and whose corneas and lenses had been removed. The eyecup in Figure 1cwas from the vehicle-injected control animal and showed no green fluorescent signal, whereas ZsGreen expression was obvious in the eyecups shown in Figures 1d and 1e . The transfection area was variable and could be restricted (Fig. 1d)or pan-retinal (Fig. 1e) , depending on each injection. The distribution of ZsGreen-positive cells in the retina was demonstrated using cryostat retinal cross-sections (Fig. 1f)viewed through a fluorescence microscope (20×). Most of the transfected cells were within the outer nuclear layer (ONL), and most (see higher 40× magnification inset) were identified by morphology as rods. Immunocytochemistry with anti-rTα (see Figs. 3d 3e 3f ) confirmed this. 

Light-Dependent Redistribution of Exogenous cTα in Rod Photoreceptor Cells

We chose to establish the in vivo transfection technique using the pSIREN-RetroQ-ZsGreen vector because Zs-green is much brighter than enhanced green fluorescent protein (EGFP) and is more easily observed in the eye of the living animal. However, pSIREN-RetroQ-ZsGreen is designed to express shRNA for knocking down specific mRNAs and did not lend itself to being easily engineered for expression of a protein, in this case cTα. Therefore, we switched to pEGFP-IRES2 to express EGFP and cTα under control of the same promoter, CMV, used by Matsuda and Cepko. ¹⁶ A cDNA vector (pEGFP-IRES2-mcTα) was then constructed using the mouse cTα sequence inserted after the CMV promoter such that cTα and EGFP were expressed from a bicistronic mRNA as separate proteins. To verify that cTα was actually expressed, human embryonic kidney (HEK)-293 cells were transfected, and analysis of homogenates on Western blots using antiserum against cTα demonstrated (Fig. 2A)immunologically reactive cTα of the correct molecular weight. 

The vector was transfected into the neonatal rat eyes (P0-P3) and at P21, after dark or light adaptation, the animals were humanely killed and the eyes were processed as described in Materials and Methods. Sections were immunostained with anti-cTα, and representative confocal images are shown in Figures 2 and 3 . Because the transfected cells that express cTα also express EGFP, they are easily visualized. Red fluorescence was used to visualize the subcellular localization of transfected cTα in the dark-adapted rods (Fig. 2Bb , arrowheads) and showed compartmentalized localization of cTα within the rod outer segments analogous to the localization of endogenous cTα (Fig. 2Bb , arrows) in the cone outer segment in the dark-adapted retinas. However, after exposure to 600 lux light for 1 hour, the exogenous cTα in rods redistributed throughout the cells, including the inner segments, cell bodies, and synaptic terminals, without compartmentation (Fig. 3b) . The distribution of exogenous cTα throughout a single rod cell is seen in Figures 3g to 3i(arrowheads), whereas an adjacent nontransfected cone revealed endogenous cTα localized within the cone OS (Fig. 3h , arrow). 

To test whether cTα expression in rods simply interfered with the translocation process in general, the location of rTα, was determined in the sections from the transfected retinas. The data show endogenous rTα (stained red) in the transfected cells (green) localized to the OS in the dark-adapted retina (Figs. 2Be 2Bf)and in the IS/ST in the light-adapted retina (Figs. 3e 3f) , identical with its compartmentation in nontransfected cells. 

High-Intensity Light-Induced Endogenous cTα Redistribution in Cones

How is it that cTα can translocate when expressed in rods but not when it is endogenously expressed in cones? One possible explanation is that the mechanism for cTα translocation might have a light intensity threshold that is much higher in cones than in rods, which are easily saturated with light of relatively low intensity. It is known that the translocation of rTα occurs when a low-level light-intensity threshold is exceeded. ^{9 17} Although 600 lux light is sufficient for rods to translocate endogenous rTα and exogenous cTα, much higher light intensity might be needed to initiate the translocation of endogenous cTα in cones. To test this, nontransfected rats were exposed to light of more than 10-fold higher intensity (7000 lux) than our normal light conditions. To avoid damage to the retina, the animals were only exposed for 10 minutes. Peanut agglutinin (PNA, stained red) was used to visualize the cone photoreceptor cell population, and the location of endogenous cTα was detected using anti–cTα serum. Our data (Figs. 4a 4b 4c)show that cTα is compartmentalized in the cone OS in dark-adapted retinas but that under high-intensity light, cTα (stained green) partially translocates to the cone IS/ST (Figs. 4g 4h 4i) . This pattern of localization throughout the cone cell is similar to that of cTα expressed in rods under 600 lux light for 1 hour (see Fig. 3 ). 

To rule out the possibility that the cells were “sick” or “dying” because of the bright light, three different tests were performed. First, no “terminal transferase dUTP nick end labeling” (TUNEL)–positive cells (an indicator of apoptosis) were found in parallel sections from the same retinas (data not shown). Second, the compartmentation of rTα and rod arrestin was examined in parallel sections and was found to be essentially identical to that which normally occurs with 600 lux light (Figs. 4j 4k 4l) . A third test demonstrated (Fig. 5)that cTα can translocate and compartmentalize back to the outer segments of cones after dark adaptation after they are exposed to 7000 lux light for 10 minutes. 

Discussion

Although many reports have demonstrated rTα and rod and cone arrestin ^{18 19} translocation in the photoreceptor cells, the light-induced translocation of cTα has not been shown. We transfected rat rods with cTα and report, for the first time, that cTα is able to translocate in rods under normal room lighting. Our data also demonstrate that endogenous cTα can translocate from OS to IS/ST in cones under high-intensity light and can translocate back to the OS after dark adaptation. 

The ability of cTα to translocate only under high-intensity light suggests that a light-intensity threshold exists for the translocation of cTα in cones. Such light intensity threshold–controlled translocation has already been demonstrated in rods for rTα ⁹ and rod arrestin. ^{17 20} The much higher light intensity needed to initiate the translocation of cTα in cones most likely results because cones do not saturate with light. ^{21 22} At increased light intensity, cones actually reduce the number of the functional opsin molecules as one of the mechanisms to achieve light adaptation. ¹ At high-intensity light, cones work with as few as 1% of their opsin molecules functional. Given this, we do not think cTα translocation is dependent on the number of opsin molecules activated but rather that some other modulator(s) exists that participates only at high light intensity. We think the mechanisms by which cTα translocates in rods and cones are similar but that the light intensity threshold in rods is lower because they saturate with light at lower intensities. 

The massive translocation of proteins in rod photoreceptor cells is considered an important rod adaptation mechanism. For cones, however, which have an enormous capacity for light adaptation, the translocation of cTα may not be as significant. For example, cones can reduce hyperpolarization in response to light by increasing cyclic guanosine monophosphate (cGMP) synthesis, cyclic nucleotide gated channel (CNG) activity, the phosphorylation of visual pigment molecules, ^{1 14} and the redistribution of cone arrestin. ^{18 19 23} They also can decrease photoisomerization and reduce the ability of the phosphorylated bleached and unbleached visual pigment molecules to activate cTα. ^{24 25} Although cTα translocation could be an additional adaptive mechanism, we think it is a protective mechanism. The high-intensity light produces a continuously activated cTα and increased phototransduction, which may overload the cone photoreceptor cells. The translocation of some of the cTα from the OS to the IS would reduce the stress and the activated signaling pathway in the cells while leaving a sufficient amount of cTα in the OS for phototransduction. This interpretation would be consistent with data from the GC1 knockout mouse. In this animal model, cone cell cGMP levels are significantly reduced and CNG channels are closed, analogous to maximum activation of phototransduction by light, even though phototransduction does not actually occur in this animal. Similarly, endogenous cTα distributes throughout the cell, ^{26 27} as one might expect if it was exposed to maximum phototransduction signaling. 

Our results show that exogenously expressed cTα partially redistributes in rods under the same lighting environment that results in the redistribution of rTα in rods. These data support the conclusion that the amino acid sequence of cTα does not inherently prevent it from being translocated. Our data also demonstrate that cTα can translocate in cones to the same extent it does when expressed in rods if the animal is exposed to high-intensity light (7000 lux) rather than to room lighting (600 lux). Rods are fully saturated by 600 lux light, but cones are not. Therefore, we think the determining factor(s) resides in the intracellular environment rather than in any special characteristics of the cTα itself (in other words, when in Rome, do as the Romans do). The failure of cTα to completely translocate in rods at 600 lux and in cones at 7000 lux suggests that some restriction is associated with protein–protein interactions, dependent on the amino acid sequence of cTα. Additional studies to identify such proteins are under way. 

 Supported by Core Grant EY12190 from the National Eye Institute, Grant P20 RR017703 from the Centers of Biomedical Research Excellence (COBRE) program of the National Center for Research Resources, general funds from the Presbyterian Health Foundation, and an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology.

 Submitted for publication February 1, 2007; revised March 17, 2007; accepted May 17, 2007.

 Disclosure: J. Chen, None; M. Wu, None; S.A. Sezate, None; J.F. McGinnis, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Representative images from transient transgenic animals. Rats were injected with pSIREN-RetroQ-ZsGreen, underwent electroporation at P0, and were kept until P21, as described. (a, b) Eyes from living animals viewed at 20×. (c–e) Eyecups. (f, inset) Epifluorescence microscopy images of retina cross-sections show the distribution of the transfected cells. Scale bars, 10 μm.

Figure 2.

 

Exogenously expressed cTα localizes to the OS in dark-adapted rods. (A) pEGFP-IRES2-cTα vector expresses (Western blot) immunoreactive cTα of the correct size in HEK293 cells. (B) In vivo transfection results in (Ba) EGFP expression in rods. Anti-cTα demonstrates (Bb) localization of cTα in the OS of rods (arrowheads) and cones (arrows) in a dark-adapted retina. The merged image (Bc) clearly shows the transfected rods. A control transfected retina (Bd) immunostained with anti-rTα (Be, Bf) shows rTα compartmentalized in the ROS, as previously demonstrated. Scale bar, 10 μm.

Figure 3.

 

cTα exogenously expressed in rods partially translocates to the rod IS and ST regions in the light (600 lux). Single confocal optical sections, 0.2-μm thick, are shown (a–f) along with high-magnification images (g–i). Transfection results in (a) EGFP expression in rods. Staining with anti-cTα demonstrates (b) localization of cTα in the OS and inner portions of rods (arrows) in a light-adapted retina. The merged image (c) clearly shows the colocalization of cTα in green-transfected rods. Staining for rTα shows the green-transfected cells (d, f) and the localization of endogenous rTα (e, f) in the IS/ST. A single transfected rod cell (g) is shown to have exogenously expressed cTα distributed (arrowheads) throughout its length (h, i). The single arrow (h, i) indicates a nontransfected cone cell in which endogenous cTα is compartmentalized in the COS. Scale bars: (a–f) 10 μm; (g–i) 5 μm.

Figure 4.

 

Endogenous cTα translocates in response to 7000 lux high-intensity light. Rats were first dark adapted overnight and then humanely killed under infrared light or exposed to 600 lux light for 1 hour or 7000 lux light for 10 minutes. (a–c) Images from the dark-adapted retinas. (d–f) Images from 600 lux light-adapted retinas. (g–l) Images from 7000 lux light-adapted retinas. PNA (red) was used as a cone-specific marker. (e) At low-intensity light (600 lux), endogenous cTα (green) localized in the cone OS. (h) Exposure to 7000 lux light results in cTα partially translocating from cone OS to IS/ST. (j, l) Proper localizations of rod arrestin (red) and rod Tα (green; k, l) within the (7000 lux) retinas are shown. Scale bars, 10 μm.

Figure 5.

 

High-intensity, 7000 lux light does not irreversibly inactivate the mechanism for translocation of cTα. Cones retain the ability to translocate and compartmentalize cTα back into the cone OS. (a, c) Cone sheaths (red) are identified by lectin (PNA) staining. (b, c) Anti-cTα (green) staining shows endogenous cTα localized to the cone OS. Scale bar, 10 μm.

The authors thank Mark F. Dittmar for help with the care and use of the animals.