Cone Cell Survival and Downregulation of GCAP1 Protein in the Retinas of GC1 Knockout Mice

Abstract

purpose. To examine the spatial and temporal characteristics of cone cell survival and the expression of guanylate cyclase–activating proteins (GCAPs) in the guanylate cyclase (GC)-1 knockout (KO) mouse retina.

methods. Immunohistochemical analyses with peanut agglutinin and an antibody specific for cone transducin were used to examine cone cell survival in the GC1 KO retina at 4, 5, 9, 16, and 24 weeks of age. Immunohistochemical and Northern and Western blot analyses were used to examine the expression of GCAP1 and GCAP2 in 4- to 5-week-old mice.

results. The number of cone cells appeared normal throughout the superior and inferior retinal regions in 4- and 5-week-old GC1 KO mice but gradually decreased by 6 months. Cone cell loss was exacerbated in the inferior retinal region, with only 2% to 8% remaining by 6 months of age; however, 40% to 70% of the cone cells survived in the superior region at this age. GCAP1 and GCAP2 protein levels were downregulated in GC1 KO retinas at 4 weeks of age and GCAP1 immunostaining was absent from the photoreceptor outer segments.

conclusions. The results of this study show that the rate of cone cell loss in the GC1 KO mouse is comparable to that previously described in the GUCY1*B chicken and in humans with Leber congenital amaurosis (LCA)-1. The GCAP expression data, when combined with those of previous electrophysiological studies of the GC1 KO mouse retina, provide evidence that GC1-GCAP1 interactions are essential for cone cell function in mice and that GC2 and GCAP2 activities contribute to the rod cell response in the absence of GC1.

The ability of vertebrate cone and rod photoreceptors to transduce light into neurochemical signals is highly dependent on the regulation of cGMP and calcium levels within the outer segments of these cells. Over the course of a light-signaling event, cGMP levels fall and rise in the outer segments after the sequential activation of cGMP phosphodiesterase and guanylate cyclase (GC). ¹ The synthesis of cGMP by GC is tightly regulated by calcium through the action of guanylate cyclase activating proteins (GCAPs), calcium-binding proteins that stimulate GC activity under low free intracellular calcium conditions. ^{2 3 4} Two isoforms of retinal GC, retGC1 and retGC2 (or GCE and GCF), ^{5 6 7 8} and of GCAP, GCAP1 and GCAP2 ^{4 9 10} are expressed in the photoreceptor cells of several vertebrate species. A third GCAP, GCAP3, has recently been identified in human and zebrafish retina, but its role in phototransduction remains unclear. ^{3 11}  

Recent experimental and clinical studies support the hypothesis that GC1 and GCAP1 are major players in the phototransduction cascade and that both of these proteins are essential for normal cone photoreceptor function and survival. Studies of rod and cone cell function in the absence of these proteins illustrate their importance to normal photoreceptor function and the complexity of their interactions. The absence of GC1 in mouse ¹² and chicken ¹³ abolishes cone cell function in the retinas of these animals. On the contrary, rod cell function is differentially affected in these two species. In the GC1 knockout (KO) mouse, the amplitude of the rod response is reduced by 50% to 70%, ¹² whereas in the GUCY1*B chicken, no rod response is detected. ¹⁴ Analyses of GCAP1/GCAP2 KO mouse models reveal that GCAP1 is sufficient to maintain normal cone ¹⁵ and rod cell function. ¹⁶ In the absence of GCAP1, GCAP2 only partially restores rod function. ¹⁷ Biochemical studies of these proteins in vitro have shown that GCAP1 activates GC1 more efficiently than GC2 and that GCAP2 activates both GC1 and GC2 with similar efficiency. ^{3 18} The observation that GCAP1 protein levels are significantly reduced in the GC1 null cone-dominant retina of the GUCY1*B chicken is consistent with the biochemical studies and supports the notion that GC1 and GCAP1 are functionally coupled in vivo. Recent clinical studies also support a central role for GC1 in photoreceptor function. Mutations in the regions of the GC1 gene that encode the extracellular, dimerization, and catalytic domains of the enzyme, which significantly reduce or abolish its activity, have been linked to Leber congenital amaurosis (LCA)-1, ^{19 20 21} a disease characterized by severely diminished sight or blindness at birth. Mutations in the GC1 gene that fall within the dimerization domain ^{22 23 24 25 26} and mutations that affect the ability of GCAP1 to bind calcium and regulate GC activity ^{27 28 29} have been linked to cone–rod dystrophies. These experimental and clinical observations support the hypothesis that GC1-GCAP1 interactions play a central role in phototransduction. Furthermore, they suggest that cone phototransduction and survival are critically dependent on these proteins. 

Comparisons of the biochemical and functional phenotypes of the rod-enriched (∼97% rods) retinas of GC1 KO mice and the cone-enriched (∼80% cones) retinas of GC1 null GUCY1*B chickens provide a unique opportunity to examine the effects of the absence of GC1 on rod and cone function. In this study, we used biochemical, molecular, and immunohistochemical techniques to examine GCAP expression in the GC1 KO mouse retina. We also characterized the spatial and temporal patterns of cone cell survival and death in animals ranging from 5 weeks to 6 months of age by monitoring the expression of the α-subunit of cone transducin (Gαt2). Our data, viewed in the context of the results of recent studies of GC1 and GCAP KO mice and of our previous studies of the GUCY1*B chicken retina, shed new light on the roles that GCAP1 plays in rod and cone phototransduction in mouse retina. In addition, the time course of cone cell loss in the GC1 KO was consistent with that observed in the GUCY1*B chicken, a result that serves as an important point of reference for assessing the efficacy of gene-based strategies for treatment of GC1-null photoreceptors. 

Methods

Mice

Homozygous male Gucy2e ^−/− mice ¹² were obtained from David Garbers (University of Texas Southwestern Medical Center, Dallas, TX). This KO line was originally generated on a B6129 (129x1Sv/C57BL6) background and has been sustained by heterozygote x heterozygote matings for more than 20 generations. To ensure that the breeding colony at the University of Florida was pathogen free, these mice were rederived by using a standard embryo-transfer protocol. Superovulated female C57BL6 mice were mated overnight to homozygous (Gucy2e ^−/−) male mice. Female mice with plugs were euthanatized at approximately 0.5 days after coitus, in accordance with the guidelines in the 2000 Report of the American Veterinary Medical Association (AVMA) Panel on Euthanasia. ³⁰ The fertilized embryos were isolated, cleaned, and cultured overnight in antibiotic medium, up to the two-cell stage, when they were implanted into previously prepared pseudopregnant recipients. The females were returned to the specific pathogen-free (SPF) animal housing facility and maintained until the litters were born and weaned. The egg recipients were then sent to Charles River Laboratories (Wilmington, MA) for a full analysis of their pathogen status. On return of the status report, the rederived GC1 KO mice were subsequently considered free of all known disease and were housed in an isolator unit for breeding in a normal 12-hour light–dark cycle. All animals were handled in accordance with the University of Florida College of Medicine’s policies on the treatment of laboratory animals and with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The genotypes of rederived GC1 KO mice were verified with PCR, as described in the next section and in the legend to Figure 1A . The absence of GC1 protein in the retinas of rederived GC1 KO mice was confirmed by immunohistochemical analyses (Fig. 1A) . 

PCR Genotyping

Progeny derived from the rederivation were genotyped with multiplex PCR, which permitted simultaneous amplification of the wild-type (WT) GC1 gene (Gucy2e) (5′-CAG ATG TGA TCT GCA ACG GAG; 5′-AGT CAG TCC CAT CCC TAT CAC) and the neomycin-resistant gene cassette, which was inserted into exon 5, disrupting the region encoding the transmembrane domain of GC1 (5′-TGC TCT GAT GCC GCC GTG TTC C; 5′-CGA TCT TTT CGC TTG GTG GTC GA). ¹² Genomic DNA was isolated from tail samples with a kit (DNeasy Qiagen, Valencia, CA), and 200 ng DNA was amplified (AmpliTaq Gold; Applied Biosystems, Foster City, CA) with the following cycle parameters: 95°C, 3 minutes (1 cycle); 95°C, 1 minute; 57°C, 1 minute; 72°C, 1 minute (30 cycles); 72°C, 10 minutes (1 cycle); and a soak at 4°C. Amplified products were resolved on 1.5% agarose gels. 

Immunohistochemistry and Microscopy

Eyes from dark-adapted GC1 KO (n = 3–4) and WT (n = 3–4) mice were fixed in 4% paraformaldehyde in PBS overnight at 4°C. The eyes were then rinsed in PBS, cryoprotected by soaking in a 30% sucrose (wt/vol)-PBS solution overnight at 4°C, and sectioned (16 μm) for immunohistochemical analyses. All analyses were performed using retinas from at least three different animals. GC1 was detected with a rabbit polyclonal antibody (pAb) raised against purified bovine GC1 (generously provided by Akio Yamazaki, Wayne State University, Detroit, MI). ¹³ GCAP1 labeling was performed with the UW14 rabbit pAb (generously provided by Krzysztof Palczewski, University of Washington, Seattle, WA), ^{31 32} an antibody that has been shown to be specific for mouse GCAP1. ⁹ For this study, the UW14 antibody was used at a dilution of 1:1000 in primary dilution buffer (0.3% Triton X-100 and 1% BSA in PBS). Peanut agglutinin (PNA) ³³ and a rabbit pAb specific for cone transducin α-subunit (Gαt2; catalog no. sc-390; Santa Cruz Biotechnology, Santa Cruz, CA) were used to identify cone cells. The specificity of the Gαt2 antibody was confirmed by preadsorption of the immune serum with purified Gαt2 peptide according to the protocol provided by the manufacturer before immunostaining, which neutralized Gαt2 labeling in the retina (data not shown). Western blot analysis was also used to confirm the specificity of the Gαt2 antibody, resulting in the detection of a single band at the expected size of ∼46 kDa. PNA tagged with the Alexa-594 fluorophore (Molecular Probes, Eugene, OR) was used at a final concentration of 25 μg/mL in PBS and anti-Gαt2 was used at a dilution of 1:500 in primary dilution buffer. The primary rabbit pAbs were visualized by labeling with a goat anti-rabbit IgG secondary antibody tagged with the Alexa-488 fluorophore diluted 1:500 in PBS (Molecular Probes). Sections were counterstained with 4′,6′-diamino-2-phenylindole (DAPI), mounted in an aqueous-based medium, and coverslipped. Confocal images were acquired with a confocal microscope (model 1024ES; Bio-Rad, Hercules, CA) and were processed with confocal-imaging software (Bio-Rad) and image-analysis software (Photoshop, ver. 7.0; Adobe Systems, Mountain View, CA). To produce the confocal images of the immunostained WT and GC1 KO retinal sections, 18 to 28 0.3-μm optical sections were acquired for each antibody label by using identical parameters and were then stacked. Image acquisition for each retinal section started and stopped in the z-planes one step above and one step below the detection of fluorescent signal from the immunolabeled cells, respectively. 

Western Blots

Retinas from WT and Gucy2e ^−/− mice were dissected from surrounding ocular tissues and prepared for SDS-PAGE analyses according to Laemmli. ³⁴ The concentration of protein in each sample was determined with a bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL). Total protein lysates were separated on 15% SDS-PAGE gels (Mini-Protean II gel system, Bio-Rad) with two retinas per lane, and transferred to polyvinylidene difluoride (PVDF) membrane with a Tris/glycine/methanol buffer. After transfer, the blots were stained with Ponceau red, and the gels were stained with Coomassie blue to assess the efficiency of protein transfer and the equivalency of protein in each sample. GCAP1 and GCAP2 proteins were detected by labeling with primary pAbs UW14 (1:1000) ³¹ and UW50 (1:1000) ¹⁰ (generously provided by Krzysztof Palczewski), respectively. A rabbit pAb specific for the rod transducin Gα subunit (Gαt1; catalog no. sc-389; Santa Cruz Biotechnology) at a dilution of 1:1000 was used as a control for protein loading. All antibodies were visualized with goat anti-rabbit IgG secondary antibody tagged with horseradish peroxidase (HRP) and a chemiluminescence detection kit (ECL; Amersham Biosciences, Piscataway, NJ). Processed blots were exposed to autoradiography film (Biomax; Eastman Kodak, Rochester, NY) and imaged (Gel Doc 1000 imaging system; Bio-Rad). Western blot analyses were repeated at least one time on protein samples from different animals for each blot. 

Northern Blot Analysis

Retinas were dissected from the left and right eyes of WT and Gucy2e ^−/− mice and placed in screw-top tubes, frozen in liquid nitrogen, and stored at −70°C until further processing. Total RNA was extracted from the pooled retinas taken from each animal (RNeasy kit; Qiagen). Northern blot analyses were prepared as previously described. ³⁵ Four lanes were loaded with retinal RNA from two WT and two Gucy2e ^−/− mice, each lane containing 8 μg RNA. The blots were probed consecutively with ³²P-labeled cDNA probes specific for GCAP1 (mGCAP1 cut with HindIII and BamHI) and GCAP2 (bGCAP2 cut with EcoRI; clones generously provided by Wolfgang Baehr, University of Utah, Salt Lake City, UT) that were labeled with a kit (Random Primed DNA Probe Synthesis and Removal; Ambion, Austin, TX). 

Results

Temporal and Spatial Analyses of Cone Cells

In the only previous histologic study of GC1 KO mouse retinas, ¹² cone cells were indirectly labeled with PNA, a lectin that binds to the extracellular matrix comprising the cone cell sheath. ³³ To label cone cells directly, we chose to use an antibody specific for cone Gαt2, a member of the G-protein signaling complex in the cone phototransduction pathway. We first examined PNA binding and Gαt2 labeling in central and peripheral retinal regions of 4- and 5-week-old WT and GC1 KO mice (Fig. 1B) . At 4 weeks of age, PNA lectin binding and Gαt2 labeling were colocalized in the photoreceptor layer and, consistent with the previous report, ¹² no gross differences were observed in the number or distribution of cone cells in the WT and GC1 KO retinas (data not shown). At 5 weeks of age, the distribution and number of PNA and anti-Gαt2–labeled cones in GC1 KO retinas were essentially identical with that observed at 4 weeks (Fig. 1B) . This result sharply contrasted with the previous characterization of retinal morphology in these mice, which concluded that cone cell loss in the retinas of these animals was rapid and nearly complete by 5 weeks of age. ¹² These conflicting results prompted us to examine PNA labeling patterns in the retinas of the founder GC1 KO mice that we had received before moving forward with our analyses. Comparisons of the retinas of the 8-week-old founder GC1 KO mice with those of age-matched rederived GC1 KO mice revealed that there was a small decrease in the number of cones within 50 to 100 μm of the optic nerve head in both retinas, compared with WT, and that the number and distribution of cones outside this region were identical in the two retinas (Fig. 1C) . These results, together with our confirmation that the genotypes of our rederived mice are identical with those of the founder mice, argue against the possibility that our findings are a consequence of the rederivation process. 

Examination of the retinas of the rederived GC1 KO mice at 9 weeks of age revealed that cone cell loss in both the superior and inferior peripheral retinal regions was greater than that observed in the central regions, with the exception of the area immediately adjacent to the optic nerve head, which contained very few to no surviving cone cells at this age. Between 9 weeks and 4 months of age, cell loss in the inferior regions of the retina exceeded that observed in the superior regions of the retina (Figs. 2A 2B) . By 6 months of age, very few cone cells were detected in the inferior regions of the retina, whereas nearly 50% of the cone cells remained in the superior regions (Figs. 2A 2B) . No evidence of rod cell degeneration was observed at any of the ages examined, a result consistent with the previous analyses of these animals. ¹² These data show that photoreceptor degeneration in the GC1 KO mouse retina is limited to the cone cell population and that the specific loss of cone cells is progressive, with the greatest loss occurring between 9 and 16 weeks of age and that this loss is exacerbated in the inferior regions of the retina. We examined 20 to 25 sections of the retinas of at least three different GC1 KO mice from different litters for each age group and obtained identical results with each analysis. 

While conducting our analyses of the time course of cone loss in GC1 KO retinas, we noted that the relative levels of Gαt2 protein and its subcellular distribution within the photoreceptors of the retinas of GC1 KO mice were markedly different from WT at all ages. Examination of individual cone cells in the retinas of these mice revealed that the subcellular distribution of Gαt2 protein in these cells was qualitatively different from that observed in WT dark-adapted retina (Fig. 2C) . Unlike the Gαt2 labeling of WT cone cells, which was primarily restricted to the cone outer segment, Gαt2 labeling of GC1 KO cones was prominent in the inner segments and synaptic pedicles. 

Expression and Subcellular Distribution Patterns of GCAPs

We examined GCAP1 and GCAP2 expression in the retinas of 4-week-old GC1 KO mice. Western blot analyses of retinas of 4-week-old GC1 KO mice showed that the levels of GCAP1 and GCAP2 protein in the retinas of these mice were reduced by approximately 60%, compared with the levels found in the retinas of age-matched WT mice (Fig. 3A) . Northern blot analyses revealed that the levels of both GCAP1 and GCAP2 mRNA in GC1 KO retinas were not significantly different from those observed in WT retinas (Fig. 3A) , a result that suggests that the reduced levels of GCAP1 and GCAP2 protein in GC1 KO retinas are not due to altered transcription of the genes encoding these proteins. Immunohistochemical analyses of WT and GC1 KO retinas were consistent with the Western blot data. No GCAP1 protein was detected in the outer and inner segments of the rod and cone cells in GC1 KO retina; however, GCAP1 protein levels in the synaptic terminals of these cells remained unchanged relative to WT (Fig. 3B , top panels). The level of GCAP2 immunostaining was consistent with the overall decrease in GCAP2 protein level in the GC1 KO retina; however, unlike GCAP1, the distribution of GCAP2 protein within the outer and inner segments and synaptic regions of the GC1 KO photoreceptor cells was similar to WT (Fig. 3B) . 

Discussion

In this study, we examined the spatial and temporal characteristics of cone cell loss in the retinas of GC1 KO mice ranging in age from 1 to 6 months by monitoring the expression of Gαt2, a cone-specific marker. We found that cone cell loss was significantly higher in inferior retinal regions compared to superior regions. The rate of cone loss was greatest between 9 and 16 weeks of age and by 6 months of age, nearly all the inferior cones had degenerated. In addition to the finding that cone cell degeneration occurs in a more gradual and regionally dependent manner than previously reported, ¹² we also found that the level of GCAP1 protein was significantly reduced in the GC1 KO retina before the loss of cone cells. The GCAP1 present in these retinas was localized to the photoreceptor synaptic terminals. The level of GCAP2 protein was also reduced, but in contrast to GCAP1, was present in the photoreceptor outer and inner segments. 

In the GUCY1*B chicken, the only other animal model of LCA1, ¹³ the retina is fully developed at hatching, and the first signs of photoreceptor disease are observed between 1 to 2 weeks after hatching. ¹⁴ Photoreceptor cell loss in the GUCY1*B retina progresses from central to peripheral retina, and the greatest cell loss occurs after 9 weeks of age. Our analyses of the GC1 KO mouse showed that the greatest number of cone cells degenerated between 2 to 4 months of age in this animal model and that a significant number of cone cells survived in the superior regions, even at 6 months of age. The reason for cone preservation in the superior versus inferior retina is unclear. In a previous characterization of this model, cone loss was reported to be nearly complete in 5-week-old animals. ¹² We were able to replicate the PNA staining pattern reported by these investigators in 4-week-old animals, but not in 5-week-old or older animals, and we found no discrepancies in PNA labeling between the founder mice and our rederived mice. Because the morphologic analyses in the previous study were focused on regions immediately adjacent to the optic nerve head, we believe that the different conclusions are simply a reflection of the extent of analyses undertaken by our respective groups. It is noteworthy that our analyses of the GUCY1*B chicken and GC1 KO mouse, together with a recent report that showed that rod and cone photoreceptor cells were detected in the retinas of an 11.5-year-old patient with LCA1, ³⁶ suggest that the temporal characteristics of photoreceptor cell loss in the GC-null animal models and in humans with LCA1 are similar. 

Our observation that GCAP1 and GCAP2 protein levels are reduced in GC1 KO mouse retinas is consistent with our previous analyses of these proteins in GUCY1*B chicken retinas. ³⁷ In those studies, we found that the absence of GC1 resulted in an 80% to 90% decrease in levels of cGMP in the photoreceptors of 1- to 2-day-old dark-adapted GUCY1*B retinas compared with control retinas. ³⁷ This observation, together with the observation that GCAP1 exhibits increased susceptibility to proteolytic degradation in the absence of calcium, ³⁸ prompted us to hypothesize that the reduction in GCAP1 protein levels observed in GUCY1*B retina was due to increased proteolytic degradation of this protein that resulted from a decrease in intracellular [Ca²⁺] caused by chronic closure of the cGMP-gated channels. This scenario is also probable in the GC1 KO mouse retina. 

Although the overall downregulation of GCAPs in the GC1 KO retina was expected, it was intriguing that the only GCAP protein detected in the outer segment region of the photoreceptor cells was GCAP2. This observation and the results of recent analyses of GCAP1/GCAP2 double-KO mice provide new clues about the possible physiological roles that GCAP1, GCAP2, GC1, and GC2 play in photoreceptor function. First, the absence of GCAP1 immunostaining in the photoreceptor outer segments of GC1 KO retinas points to a possible role for GCAP2 in regulating GC2 activity in rod photoreceptors. The hypothesis that GC2 and GCAP2 underlie rod function in the GC1 KO mouse retina is indirectly supported by the recent observation that the kinetics of rod photoreceptor responses to light stimulation in the retinas of GCAP1/GCAP2^−/− mice expressing bovine GCAP2 are very similar to those measured in rods in the GC1 KO mouse. ¹⁷ The additional observation that expression of GCAP1 in the GCAP1/GCAP2 KO mice restores normal flash response kinetics in both rod ^{16 17} and cone ¹⁵ photoreceptor cells reinforces the view that GCAP1 is essential for cone function and that it alone is capable of maintaining normal rod function. Second, the observation that rod photoreceptor cells remain functional and do not degenerate in the GC1 KO mouse retina ¹² is consistent with the hypothesis that GC2, which has also been localized to photoreceptor outer segments, ³⁹ is capable of supporting rod phototransduction and survival in the absence of GC1. Because cone photoreceptors cannot function in the absence of GC1 and degenerate in both the GC1 KO mouse and GUCY1*B retina, it appears that these cells do not possess a “redundant” mechanism for cGMP synthesis that can sustain function and survival. This scenario holds true for cone cells in GUCY1*B; however, unlike the rods in the GC1 KO mouse retina, the absence of GC1 in the GUCY1*B retina adversely affects rod function and survival. Functional and histologic studies of human LCA1 retinas show that rod cell function is also compromised in them. ³⁶ The “bystander” hypothesis that has recently been proposed as a possible explanation for the observed secondary loss of rod or cone cells in genetic retinal diseases could explain these observations. ⁴⁰ If cone cells are the primary targets in LCA1, then rod cell death in chicken and human retinas could reflect greater connectivity by gap junctions between cones and rods in these retinas. The relatively uniform, central-to-peripheral loss of rods across the GUCY1*B chicken retina, ¹⁴ whose photoreceptor population is ∼80% cones, is consistent with the predictions of the “bystander effect” hypothesis. Based on these observations, rod loss in human LCA1 retina would be predicted to be most severe in the cone-enriched macular region. 

Finally, we noticed that the subcellular distribution of Gαt2 is altered in cone photoreceptors in GC1 KO retinas. Unlike WT retinas in which Gαt2 was restricted to the cone outer segments, Gαt2 immunoreactivity in GC1 KO retinas was also observed in the inner segments and synaptic regions of these cells. There are now several reports that show that rod transducin subunits and rod and cone arrestin undergo light-driven subcellular translocation. ^{41 42 43 44} Light-driven translocation of cone transducin has not yet been documented; however, the subcellular distribution pattern of cone Gαt2 that we observed in the dark-adapted GC1 KO retina is consistent with the pattern that would be predicted for this protein in retinas exposed to light. Based on observations made in studies of the GUCY1*B chicken, we have proposed that low cGMP levels resulting from the absence of GC1 lead to chronic hyperpolarization of the photoreceptor cells and mimic constant light exposure. ¹³ The distribution of Gαt2 in the dark-adapted cone cells of the GC1 KO retina is consistent with this hypothesis. 

In summary, the results of this study show that the retinal disease phenotype of the GC1 KO mouse is comparable to the phenotypes previously described in the GUCY1*B chicken and in humans with LCA1. These data pave the way to further comparative analyses of the rod-dominant mouse retina and the cone-dominant chicken retina, the results of which will provide clearer insight into the cellular and biochemical consequences of GC1 null mutations than could be obtained from studies of either model alone. At this point, it is clear from our analyses of the GC1 KO mouse and the GUCY1*B chicken, and from previous studies of mice expressing GCAP1 alone, ¹⁵ that GC1-GCAP1 interactions are critical for normal cone cell function. Our current studies of the effects of viral vector-mediated GC1 expression on cone and rod function in the retinas of these animals will help to clarify the role that GC1 plays in photoreceptor function. 

 Supported by National Eye Institute Grant EY11388, Training Grant EY07132, and Core Grant EY08571 to the Department of Ophthalmology, University of Florida College of Medicine; and a Research Grant from the University of Florida McKnight Brain Institute.

 Submitted for publication April 6, 2004; revised April 30, 2004; accepted May 4, 2004.

 Disclosure: J.E. Coleman, None; Y. Zhang, None; G.A.J. Brown, None; S.L. Semple-Rowland, 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.

 

Cone cell survival in 5-week-old GC1 KO retinas. (A) Top: genotyping of founder GC1 KO mice (F), rederived GC1 KO mice (RO), and progeny (F1 x F1) of rederived GC1 heterozygous (+/−) and KO (−/−) mice (blank contains no genomic DNA template). Amplification of the WT gene yielded a product of 404 bp and amplification of the neo gene cassette in the GC1 KO yielded a product of 309 bp (123, 123-bp DNA ladder; 1kb, 1-kb DNA ladder). Bottom: cross sections of retinas from 5-week-old WT and GC1 KO mice labeled with anti-GC1 (red) and counterstained with DAPI (blue). (B) Left: DAPI-stained section through the vertical meridian of a 5-week-old GC1 KO mouse eye labeled to show the areas within the superior (S) and inferior (I) regions that were examined throughout this study. Right: Distribution of cone cells in central-superior (S1) and central-inferior (I1) regions of WT (+/+) and GC1 KO (−/−) retinas. Sections were labeled with PNA (red), anti-Gαt2 (green), and counterstained with DAPI (blue) to highlight the cell layers. The images in the right column (+DAPI) are overlays of all three images. (C) Comparison of cone cells present in region S2 in founder and rederived GC1 KO mouse retinas. ONH, optic nerve head; ONL, outer nuclear layer; OPL, outer plexiform layer; IS, inner segment; OS, outer segment.

Figure 2.

 

Spatiotemporal analyses of cone cell survival in 4- and 6-month-old GC1 KO retina. (A) Regional comparison of cone cells in WT and GC1 KO retinas. Sections were labeled with anti-Gαt2 (green) and counterstained with DAPI (blue). Images are representative of sections from one animal in each age group. Identical staining results were observed in all retinas examined (n = 3). (B) Regional distribution of cone cell survival in GC1 KO retinas. The percentage of cones surviving is expressed relative to the number of WT cone cells. Estimates of the number of cone cells in each area were obtained by counting the number of Gαt2-positive photoreceptor cells present in two retinas from two different animals at each age. Cone cell survival in the WT retina is shown as 100%. (C) High-magnification view (×1000) of dark-adapted WT and GC1 KO cone cells labeled with anti-Gαt2. Note the increased Gαt2 immunoreactivity in the inner segment (arrows) and synaptic regions ( ) of GC1 KO cones. OPL, outer plexiform layer; ONL, outer nuclear layer; IS, inner segments; OS, outer segments.

Figure 3.

 

Analyses of GCAP1 and GCAP2 expression in GC1 KO retinas. (A) Western and Northern blot analyses showing relative levels of GCAP1 and GCAP2 protein and mRNA in whole retina homogenates. GCAP1 and GCAP2 protein was detected with the pABs UW14 and UW50, respectively. Equivalent protein loading was confirmed by reprobing the blots with an antibody specific to rod Gαt1. GCAP1 and GCAP2 mRNAs were detected with the corresponding ³²P-labeled cDNA probes. Equal loading was confirmed by staining the Northern blot analysis with methylene blue to visualize 18S rRNA. (B) Comparison of GCAP1 (UW14) and GCAP2 (UW50) immunolabeling and PNA labeling in retinal sections from 5-week-old WT and GC1 KO mice. Images are stacked confocal images, except in the far right image, which was obtained with UV fluorescence microscopy. The close-up of WT retina stained with anti-Gαt2 and PNA illustrates the relatively high levels of GCAP1 expressed in cone cells. OPL, outer plexiform layer; ONL, outer nuclear layer; IS, inner segments; OS, outer segments. Scale bars, 25 μm.

The authors thank Kim Howes and Rebecca Ellis for technical advice, Kathy Laughlin and James Grigg for excellent technical assistance, and Wolfgang Baehr for a critical reading of the manuscript.