Gene Replacement Therapy Rescues Photoreceptor Degeneration in a Murine Model of Leber Congenital Amaurosis Lacking RPGRIP

Abstract

purpose. Retinitis pigmentosa GTPase regulator (RPGR) is a photoreceptor protein anchored in the connecting cilia by an RPGR-interacting protein (RPGRIP). Loss of RPGRIP causes Leber congenital amaurosis (LCA), a severe form of photoreceptor degeneration. The current study was an investigation of whether somatic gene replacement could rescue degenerating photoreceptors in a murine model of LCA due to a defect in RPGRIP.

methods. An RPGRIP expression cassette, driven by a mouse opsin promoter, was packaged into recombinant adeno-associated virus (AAV). The AAV vector was delivered into the right eyes of RPGRIP ^−/− mice by a single subretinal injection into the superior hemisphere. The left eyes received a saline injection as a control. Full-field electroretinograms (ERGs) were recorded from both eyes at 2, 3, 4, and 5 months after injection. After the final follow-up, retinas were analyzed by immunostaining or by light and electron microscopy.

results. Delivery of the AAV vector led to RPGRIP expression and restoration of normal RPGR localization at the connecting cilia. Photoreceptor preservation was evident by a thicker cell layer and well-developed outer segments in the treated eyes. Rescue was more pronounced in the superior hemisphere coincident with the site of delivery. Functional preservation was demonstrated by ERG.

conclusions. AAV-mediated RPGRIP gene replacement preserves photoreceptor structure and function in a mouse model of LCA, despite ongoing cell loss at the time of intervention. These results indicate that gene replacement therapy may be effective in patients with LCA due to a defect in RPGRIP and suggest that further preclinical development of gene therapy for this disorder is warranted.

Inherited retinal degenerations are a diverse group of conditions that result from mutations in any one of more than 100 different genes (http://www.sph.uth.tmc.edu/Retnet/ provided in the public domain by the University of Texas Houston Health Science Center, Houston, TX). ¹ Most of the disorders are progressive and result in death of rod and/or cone photoreceptor cells. Many of the causative genes have now been identified, providing an impetus to develop gene-based treatments. Furthermore, the eye has advantages as a target organ for gene therapy: It is easily accessible, allowing localized exposure to therapeutic agents with reduced risk of systemic effects, and noninvasive techniques are available to monitor the effects of treatments. ^{2 3 4} Because many of the retinal degenerations are caused by defects in genes that are expressed in either photoreceptors or the retinal pigment epithelium (RPE), these cells are the main targets for gene transfer. A wide range of viral vectors has been evaluated for use in the eye. To date, the most useful vectors for photoreceptor gene transfer appear to be those based on adeno-associated virus (AAV). After subretinal delivery, AAV vectors mediate efficient and long-term transgene expression in photoreceptor cells, as well as in the RPE. ^{5 6 7} The availability of such efficient vectors systems for transduction of the retina has enabled effective gene therapy strategies to be developed in several animal models of inherited retinal degeneration. ^{8 9 10 11 12}  

Leber congenital amaurosis (LCA) is a severe, early-onset form of photoreceptor degeneration involving both rods and cones. ^{13 14 15 16 17} The severity of these disorders makes them prime candidates for initial clinical trials involving gene replacement therapies. LCA is caused by mutations in at least eight genes, three of which are RPE-specific, the remainder being expressed in photoreceptors, ^{13 18} which include the gene for RPGRIP. RPGRIP was identified through its interaction with RPGR, ^{19 20 21} another essential photoreceptor protein encoded by the X-linked RP3 locus. ^{22 23} Both proteins localize to the connecting cilia of rods and cones. ^{21 24} Because RPGRIP associated stably with the ciliary axoneme and its localization remained unchanged in photoreceptors lacking RPGR, RPGRIP was proposed to be the primary resident, whereas RPGR depended on RPGRIP for its localization in the connecting cilia. ^{21 25} Further studies showed that RPGR level was unchanged in the RPGRIP-deficient (RPGRIP ^−/−) mice, but that the protein failed to localize in the connecting cilia, thereby confirming that RPGRIP tethers RPGR in the connecting cilia. ²⁶ Thus, RPGRIP is required for the normal localization and for the proposed function of RPGR in regulating protein trafficking across the connecting cilia. ²¹ Studies of the mutant mice also suggest that RPGRIP may additionally function in nascent disc morphogenesis, because outer segment disc formation was severely disrupted in this mutant. Consistent with the notion that RPGRIP both subserves RPGR function and has an additional role in photoreceptors, loss of RPGRIP in mice leads to a more severe disease than the loss of RPGR. ⁴ Photoreceptor degeneration is evident in RPGRIP ^−/− mice as early as postnatal day 15 and progresses to a substantial loss of most cells by 3 months of age. This course of disease in the RPGRIP ^−/− mutant is in line with the clinical manifestations of LCA in which patients have early onset of visual loss and nearly complete loss of vision by early adolescence. 

In general, it appears that RPE defects are easier to treat than photoreceptor defects and preclinical studies of gene therapy for LCA due to mutations in the RPE-specific gene RPE65 have been particularly successful. ^{9 11} Trials of gene therapy to treat this defect are likely to be the first to enter the clinic. In this study, we investigated the efficacy of gene replacement therapy for the treatment of a mouse model of LCA due to a primary defect in a photoreceptor-specific gene. We attempted to correct the retinal phenotype of the RPGRIP ^−/− mouse model of LCA with AAV2-mediated gene replacement of recombinant RPGRIP. We performed a single subretinal injection of an AAV vector in which RPGRIP expression is driven by a mouse opsin promoter, followed by morphologic and functional evaluation. We showed that AAV-mediated RPGRIP expression can restore the function of the photoreceptors and prolong their survival in this model of severe retinal degeneration. 

Materials and Methods

Recombinant AAV Construct and AAV Vector Production

The murine opsin promoter fragment (mOps), spanning nucleotides −218 to +17, relative to the transcription start site, was amplified by PCR from genomic DNA. The promoter was cloned into a parental plasmid AAV-CMV-GFP ²⁷ to generate the plasmid pD10/mOps-GFP. The RPGRIP cDNA encompassing the entire coding region was cloned into this construct replacing the GFP gene, to form pD10/mOps-RPGRIP. The two plasmids, pD10/mOps-GFP and pD10/mOps-RPGRIP, were packaged into AAV2 to generate two recombinant AAV2 viral vectors: AAV-mOps-GFP and AAV-mOps-RPGRIP (total length, 4850 bases). 

Recombinant AAV2 vectors were produced by a method described previously that used a replicating amplicon pHAV7.3, containing the rep and cap genes of AAV and the PS1 HSV helper virus. ²⁷ The pD10/mOps-RPGRIP or pD10/mOps-GFP plasmids and the pHAV7.3 amplicon were transfected (at 1:1 ratio) into BHK cells by incubation with a mixture of three components: peptide6 ((K₁₆)GACRRETAWACG), plasmid DNA, and lipofectin (Invitrogen-Gibco, Paisley, UK) in a weight ratio of 0.75:4:1 in serum-free medium (OptiMEM; Invitrogen-Gibco). BHK cells (10⁷ in a 15-cm dish) were transfected with the mixture containing a total of 60 μg of plasmid DNA for 4 hours and incubated with DISC-HSV (PS-1) as a helper virus. After 24 to 36 hours, at completion of the lytic cycle, the cells were collected, centrifuged and lysed by a repeated freeze–thaw process. AAV particles were purified using a heparin column purification protocol as described ²⁸ : DNA remaining in the lysate was degraded with 50 units of endonuclease per milliliter lysate (for 30 minutes at 37°C), and cell debris was removed by centrifugation. The lysate was treated with 0.5% deoxycholic acid for 30 minutes at 37°C, filtered through 5- and 0.8-μm syringe filters (SLSV R25 LS and SLAA 025 LS; Millipore, Bedford, MA) and applied to a heparin-agarose column (Sigma-Aldrich, Poole, UK) prewashed with phosphate-buffered saline with MgCl₂ (1 mM) and KCl (2.5 mM; PBS-MK). The column was washed with 10 mL PBS-MK+0.1 M NaCl, and the AAV particles were eluted with 6 mL PBS+0.4 M NaCl (the first 2 mL was discarded). The resultant AAV preparation was concentrated (Centricon 10 columns; Millipore), washed in PBS-MK, and concentrated again to a volume of approximately 100 μL. AAV particle titers were determined by dot-blot analysis of the vector DNA. 

Animals and Vector Delivery

The generation and analysis of RPGRIP ^−/− mice have been described previously. ⁴ The RPGRIP ^−/− mice used in this study were bred from sibling mating among RPGRIP homozygotes maintained at our institutional animal facility. At postnatal ages of P18 to P20, RPGRIP ^−/− mice (n =16) were placed under general anesthesia, and their pupils were dilated. With the aid of a dissecting microscope, a 0.5-mm incision was made through the cornea adjacent to the limbus and a 33-gauge blunt needle fitted to a syringe (Hamilton, Reno, NV) was passed just behind the lens and inserted into the subretinal space in the superior retina. All injections were made in a location approximately two thirds of the distance vertically from the optic disc to the ora serrata. A volume of ∼1 μL of AAV2-mOps-RPGRIP at a titer of 1 × 10¹²/mL was injected into the right eye of each animal. The same volume of normal saline was injected into the left eyes as the control. Fundus examinations immediately after the injection showed that slightly less than half of the retina at the injection site was detached, appearing as a bolus, confirming subretinal delivery. In the experiment to validate the reporter vector AAV-mOps-GFP, 2 μL of the vector at a particle titer of 1 × 10¹²/mL was injected subretinally. All experiments involving animals were approved by the institutional Animal Care and Use Committee and performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Histology, Immunofluorescence, and ERG

Light and electron microscopy was performed as previously described. ²¹ Subcellular localization of RPGR and RPGRIP was examined by immunofluorescence on freshly cut, frozen retinal sections as described. ^{21 26} Rhodopsin was detected by staining with the monoclonal antibody rho-1D4 (gift of Robert Molday, University of British Columbia, Vancouver, British Columbia, Canada). Mice were dark-adapted overnight for ERG recording. The dark-adapted, rod-dominated responses were elicited with 10-μs flashes of white light (4.3 log ft-L) presented in a Ganzfeld dome. ERGs were recorded from both eyes of RPGRIP ^−/− mutant mice at 2, 3, 4, and 5 months after injection to monitor retinal function over this time period. Mice were examined for any signs of media opacities before each recording, and only those with clear media were tested. For morphometric analyses of photoreceptor inner and outer segment length and outer nuclear layer thickness, measurements were made in the vertical meridian (superior to inferior) at five locations to each side of the optic nerve head separated by approximately 400 μm each. Measurements began at approximately 500 μm from the optic nerve head itself. 

Statistical Analysis

Commercial software (JMP, ver. 3.2; SAS Institute, Cary, NC) was used to compare most outcomes in treated versus untreated eyes by the paired t-test. The PROC MIXED feature of SAS (version 6.12) was used to compare outcomes in treated versus untreated eyes in analyses involving unbalanced data (i.e., some data available for one eye and not the other) or when adjusting for a within-subject covariate (i.e., some fellow eyes prepared by different histologic methods). 

Results

Validation of the Mouse Opsin Promoter by AAV-Mediated GFP Reporter Gene Expression

The presumed mouse opsin promoter fragment (mOps) encompassing nucleotides −218 to +17 upstream from the transcription start site has not been validated previously in an in vivo setting. To verify its transcriptional activity and cell specificity in the retina, a GFP reporter gene was constructed and was packaged into recombinant AAV2 to produce AAV-mOps-GFP. This vector was injected subretinally into the eyes of WT mice (n = 3). Six weeks after injection, the mice were killed, and the frozen retinal sections were examined by fluorescence microscopy. As shown in Figure 1 , GFP expression was detected only in photoreceptor cells and was completely absent in RPE, a tissue that is normally infected by rAAV-2 after subretinal injections. These data confirm that the region encompassing nucleotide −218 to +17 upstream from the transcription start site of the mouse opsin gene is an effective promoter in the context of AAV-mediated gene delivery to photoreceptor cells. 

Expression and Subcellular Localization of RPGRIP, and Its Binding Partner RPGR, after AAV-Mediated RPGRIP Gene Transfer

We examined whether subretinal injection of AAV-mOps-RPGRIP led to expression of RPGRIP and whether the resultant protein localized correctly to the connecting cilia. We addressed this question by immunofluorescence analysis of frozen retinal sections using an anti-RPGRIP antibody. Retinas of RPGRIP ^−/− mutants taken at 5 months after injection of AAV-mOps-RPGRIP stained positive with the RPGRIP antibody (n = 7; Fig. 2 , top), whereas saline-injected control retinas were completely free of any RPGRIP staining (n =5). No RPGRIP staining was observed in the RPE cell layer, consistent with the findings with the GFP reporter gene. The distribution of RPGRIP was more pronounced in the superior hemispheres coincident with the site of subretinal injection. The RPGRIP signal appeared as dots between the inner and outer segments corresponding to the location of the connecting cilia. These data show that the AAV-mOps-RPGRIP directed efficient gene transfer and tissue-specific expression of RPGRIP in the photoreceptors and that the subcellular localization of the recombinant RPGRIP was identical with that of the endogenous WT protein. 

RPGRIP normally anchors RPGR within the connecting cilia. On loss of RPGRIP, RPGR becomes diffusely distributed in the photoreceptor cells. ²⁶ We therefore tested whether expression of the recombinant RGPRIP restored the normal subcellular localization of RPGR. Immunofluorescence staining with an anti-RPGR antibody showed that RPGR was restored to its normal location in the connecting cilia in treated eyes (n = 7; Fig. 2 , bottom). As expected, restoration of normal RPGR localization was only seen in areas where recombinant RPGRIP expression was found. It was not seen in control retinas (n = 5). These results indicate that the RPGRIP protein appeared functional. 

Preservation of Photoreceptors in the RPGRIP^−/− Mutant by AAV-Mediated Gene Replacement

Previous studies of the course of degeneration in the RPGRIP ^−/− mutant showed that most of the photoreceptors were lost by 3 months of age. ²⁶ We therefore euthanatized all mice for histologic analyses at 5 months after injection after the final ERG recording. Eyes that had received AAV-mOps-RPGRIP were found to have significantly more rows of photoreceptors than control eyes injected with saline (Fig. 3A) . Photoreceptor inner and outer segments were also better organized and longer in the treated eyes than in the control fellow eyes. Photoreceptor outer segments in the treated eyes were well aligned with tightly packed disks shown by electron microscopy, whereas photoreceptors in the control eyes had either no outer segments or had very shortened and disorganized residual disc structures (Fig. 3B) . As another measure of rod photoreceptor rescue, we examined rhodopsin subcellular localization in photoreceptor cells. Rhodopsin was mislocalized in photoreceptor cell bodies and synapses in mice without RPGRIP. ²⁶ As shown in Figure 4 , rhodopsin was found in photoreceptor cell bodies in the control retina but was partitioned primarily in photoreceptor outer segments in the treated retinas. 

Photoreceptor rescue, as measured by outer nuclear layer and combined inner and outer segment thickness, was more prominent in the superior hemisphere (Fig. 5) . This area corresponded with the site of injection. On average, the outer nuclear layer thickness at the site of injection was significantly greater in treated eyes than in control eyes (P < 0.001; Fig. 6A ). Occasionally, treated eyes showed as many as seven rows of nuclei at the site of injection. A significant preservation of photoreceptors was also observed in the entire retina (P = 0.002; Fig. 6A ). In addition, the combined inner and outer segments were also significantly longer in the treated than in the control eyes (P < 0.001, Fig. 6B ). 

Rescue of Mutant Photoreceptor Function by AAV-Mediated Gene Replacement

As a measure of retinal function, dark-adapted, rod-dominated ERGs were recorded from both AAV-mOps-RPGRIP-treated and control eyes at 2, 3, 4, and 5 months after injection. Representative dark-adapted ERG tracings are shown (Fig. 7A) . Average ERG b-wave amplitudes began to diverge between the treated and control groups at 3 months after injection (Fig. 7B) . Treated eyes showed a significantly slower mean exponential rate of ERG b-wave amplitude decline (6.1%/month) than control eyes (21.9%/month) during this period (P = 0.01). As a result, mean b-wave amplitude at 5 months after injection (Fig. 7C)was significantly higher (P = 0.004) in the treated eyes (n = 8; geometric mean = 118 μV) than in the control eyes (n = 9; geometric mean = 66 μV). 

Discussion

The present study demonstrates the first photoreceptor rescue in an animal model of LCA with gene replacement therapy targeting photoreceptor cells. Previous studies have shown survival of photoreceptors in a form of retinal degeneration caused by mutations in the RPE, but in these diseases the photoreceptors are inherently healthy. Attempts to treat retinal degeneration due to photoreceptor defects have succeeded in showing some rescue of function, but have failed to prevent the continuing loss of photoreceptor cells. ²⁹  

Without treatment, RPGRIP ^−/− mutant mice undergo a rapid course of photoreceptor degeneration characterized by ongoing cell loss as early as postnatal day 15 and profound disruption of outer segment formation. ²⁶ Subretinal delivery of AAV-mOps-RPGRIP in the mutant resulted in RPGRIP expression to the connecting cilia of photoreceptors. Furthermore, expression of the recombinant RPGRIP restored the normal localization of RPGR in the connecting cilia whereas in control eyes RPGR remained diffusely distributed in photoreceptors. Analyses with light microscopy showed a much thicker photoreceptor nuclear layer and longer photoreceptor inner and outer segments in treated eyes. By electron microscopy, photoreceptor outer segments in the treated eyes were well organized with tightly packed disks resembling those of WT mice. In contrast, the control eyes either had no outer segments or had disorganized, residual outer segments. In addition, rhodopsin mislocalization characteristic of RPGRIP ^–/– photoreceptors was reversed in the treated retinas. Morphologic rescue of photoreceptor cells in the treated eyes fully overlapped with the area of injection and expression of transgenic RPGRIP in retinal sections. Functional analysis showed a 72% slower mean rate of decline in ERG amplitude over the entire follow-up period and a higher average ERG amplitude at final follow-up, indicating some functional preservation of the photoreceptors as well. These data demonstrate that delivery of AAV-mediated RPGRIP replacement therapy effectively reconstituted RPGRIP function in the recipient photoreceptors and ameliorated the course of photoreceptor degeneration in these mutant mice. 

RPGRIP is essential in both rod and cone photoreceptor cells. Because a rhodopsin promoter was used in the current study to drive RPGRIP expression from the AAV vector, it seems likely that RPGRIP expression would be limited to only rod photoreceptors. Indeed, antibody staining for blue and green cone opsins in a treated eye found only a few remaining cone outer segment remnants in areas with substantial rod photoreceptor rescue (data not shown), suggesting that rescue and hence expression was limited to rod photoreceptors. 

Although it was significantly better than in control eyes, the functional rescue of photoreceptors after AAV-RPGRIP treatment was incomplete. Five months after injection the mean ERG b-wave amplitude was only approximately 20% of that in wild-type mice at that age. There are several factors that are partially responsible for this result. First, a single subretinal injection was given in the superior hemisphere and not more than half of each retina was directly exposed to the injected vector. As a result, RPGRIP expression was highest and the rescue was most successful in the superior hemisphere. In the inferior hemisphere furthest removed from the site of injection, the retinal morphology was closer to that of control eyes. Second, the retinal disease had progressed for weeks before our treatment took effect. Degeneration of rod photoreceptors begins as early as 15 to 20 days after birth and progresses rapidly in RPGRIP ^−/− mice. ²⁶ Because the subretinal delivery of vector was performed at postnatal days 18 to 20 and, because AAV2 vectors usually give rise to transgene expression approximately 3 to 4 weeks after injection, the earliest age when the recombinant RPGRIP would be present in recipient photoreceptors is estimated to be approximately 1.5 months. In our own experience with this type of AAV vectors, maximum expression is usually observed at 3 months after injection. The observation that up to six to seven rows of photoreceptor nuclei remained at 5 months after injection in some treated retinas suggests that once recombinant RPGRIP was present, it led to a near complete cessation of the degenerative process. Furthermore, outer segments appeared to form once recombinant RPGRIP was expressed. Because in absence of RPGRIP there were few organized outer segments, introduction of the recombinant RPGRIP not only halted cell death but also caused the outer segments to form in the treated areas. 

In conclusion, we have demonstrated a significant rescue of the retinal phenotype with AAV mediated gene replacement therapy in an animal model of LCA without RPGRIP. Future studies should address, in addition to safety, the choice of promoters for expression in both rods and cones and use of multiple or repeated subretinal injections to cover the entire retina. A further improvement may be expected from the use of AAV2 pseudotyped with the AAV5 capsid, which has been shown to mediate a more efficient and more rapid onset of transgene expression in photoreceptor cells. ^{7 30 31} Because both human LCA patients with null RPGRIP alleles and RPGRIP ^−/− mice exhibit an early-onset, severe form of retinal degeneration, a similar treatment approach for these patients seems very compelling. Our results suggest that further preclinical development of gene replacement therapy for LCA due to defects in RPGRIP is warranted. 

 Supported by National Eye Institute Grants EY10581 and EY14188 and Core Grant for Vision Research EY14104, the Macular Vision Research Foundation, the Foundation Fighting Blindness, and the Sir Jules Thorn Charitable Trust.

 Submitted for publication March 23, 2005; revised April 26, 2005; accepted June 14, 2005.

 Disclosure: B.S. Pawlyk, None; A.J. Smith, None; P.K. Buch, None; M. Adamian, None; D.-H. Hong, None; M.A. Sandberg, None; R.R. Ali, None; T. Li, 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.

 

Validation of the mouse opsin promoter construct in the AAV2 vector. The GFP reporter vector, AAV-mOps-GFP, was delivered subretinally and GFP expression analyzed by fundus examination (left) and fluorescence microscopy (right). Delivery of the vector (2 μL) led to transduction of slightly more than half of the retinal area. Transduction efficiency in photoreceptors appeared high, and there was no leaking expression in the RPE or inner retina. Green, GFP; red, propidium iodide nuclear counterstain; RPE, retinal pigment epithelium; OS, outer segment; IS, inner segment; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GC, ganglion cell layer.

Figure 2.

 

Immunofluorescence staining for RPGRIP and RPGR proteins at 5 months after subretinal delivery of AAV-mOps-RPGRIP. Frozen, unfixed retinal sections from wild-type (WT) eyes and control and treated RPGRIP ^−/− eyes were stained with anti-RPGRIP or anti-RPGR antibodies (orange). Both RPGRIP and RPGR were concentrated in the connecting cilia. Sections were counterstained with Hoechst dye 33342 to highlight cell nuclei (blue). Abbreviations are defined in Figure 1 .

Figure 3.

 

Morphologic analyses of the treated and control RPGRIP ^−/− retinas at 5 months after injection. (A) Light micrographs of the superior hemisphere of a representative wild-type (WT) retina and control and treated RPGRIP ^−/− retinas. Right: higher magnification images of boxed areas in retinas shown on the left. A single row of photoreceptors remained in the injected areas of the control retina, whereas four to five rows of cells remain in the treated retina. (B) Electron microscopy of the same tissues samples as shown in (A). The treated retina shows well-organized outer segment discs resembling those in the WT. CC, connecting cilia.

Figure 4.

 

Rod photoreceptor rescue as indicated by the normal localization of rhodopsin in the outer segments. Rhodopsin (orange) normally localizes primarily in photoreceptor outer segments of the wild type (WT) but was mislocalized in cell bodies of the RPGRIP ^−/− control. In the treated retinas, rhodopsin showed outer segment localization similar to that of the WT. Sections were counterstained with Hoechst dye 33342 to highlight cell nuclei (blue). Abbreviations are defined in Figure 1 .

Figure 5.

 

Variable rescue of photoreceptor cells in different regions of the retina. Shown are morphometric measurements of outer nuclear layer (ONL) thickness (A) and combined outer and inner segment (OS/IS) lengths (B) along the vertical meridian of a representative pair of treated and control retinas.

Figure 6.

 

Summary of morphologic rescue in the treated eyes. (A) Comparison of the mean outer nuclear layer thickness of the treated and control eyes based on either the treatment area alone or on thicknesses averaged across the entire retina (n = 11). Treatment preserved an additional 3.1 rows at the treatment area (P < 0.001) and an additional 1.6 rows over the whole retina (P = 0.002). (B) Comparison of the mean inner and outer segment (IS/OS) combined lengths within the treatment area in the treated and control eyes (n = 11). Treatment, on average, led to a twofold increase in IS/OS length (P < 0.001). Data are expressed as the mean ± SEM.

Figure 7.

 

Functional rescue assessed by ERG analyses. (A) Representative ERG waveforms from treated and control RPGRIP ^−/− eyes. A wild-type (WT) waveform is shown for comparison. The treated retina shows a discernible a-wave and a higher b-wave than in the control retina. (B) The b-wave amplitudes (mean ± SEM) for treated versus fellow control eyes began to show apparent divergence 3 months after injection. (C) Rod ERG b-wave amplitude (Log_e) declined at a slower rate in treated eyes than in control eyes (n = 5, P = 0.01). Error bars, ±SE. Equivalent exponential slopes are 6.1%/month and 21.9%/month, respectively.