Structural and Functional Protection of Photoreceptors from MNU-Induced Retinal Degeneration by the X-Linked Inhibitor of Apoptosis

Abstract

purpose. To evaluate the neuroprotective effects of adenoassociated virus delivery of XIAP in N-methyl-N-nitrosourea (MNU)-induced retinal degeneration in Sprague-Dawley rats.

methods. Sprague-Dawley rats were injected subretinally with recombinant adenoassociated virus (rAAV) encoding either XIAP or green fluorescent protein (GFP; injection control). Six weeks after injection, the animals received an intraperitoneal injection of MNU, a DNA methylating agent, at a dose of 60 mg/kg. Electroretinograms (ERGs) were recorded at 0, 24, 48 and 72 hours and 1 week after MNU. The rats were killed after the ERG was performed and were perfused with 4% paraformaldehyde. Eyes were then enucleated and embedded for cryosectioning. Eye sections were analyzed by TUNEL and histologic techniques. Real-time PCR and Western analysis were performed to confirm the overexpression of XIAP in injected eyes.

results. Real-time PCR and Western analysis confirmed the overexpression of XIAP in virus-injected eyes in comparison to uninjected control eyes. At 24 hours after MNU injection, fewer cells had undergone apoptosis in the XIAP-treated eyes in comparison with GFP-injected or uninjected eyes. Hematoxylin and eosin staining revealed that the uninjected and GFP-injected photoreceptors were destroyed by 72 hours after injection of MNU, whereas the AAV-XIAP-injected eyes showed structural protection of the photoreceptors at all time points throughout the 1-week sampling period. ERGs showed functional protection up to 1 week after MNU injection in the AAV-XIAP–injected eye, whereas no response was observed in the control eye.

conclusions. The results suggest that XIAP is protective against this potent chemotoxic agent and holds promise as a therapeutic agent in gene therapy approaches to treating retinitis pigmentosa.

Retinitis pigmentosa (RP) is the most common inherited cause of blindness in the developed world, affecting approximately 1 in 3500 individuals globally. ^{1 2 3} It can be inherited in an autosomal dominant, autosomal recessive, or X-linked fashion. It is characterized by a progressive loss of rod and cone photoreceptors leading to night blindness and can be diagnosed by changes in the electroretinogram (ERG). A gradual loss of visual acuity and peripheral vision leading to “tunnel” vision accompanies these changes, followed by loss of the central visual field in most cases. Fundus examinations in patients with RP reveal retinas with bone spicule formation resulting from melanin pigment that has infiltrated the retina. ^{4 5 6} The first symptoms of RP usually appear in adolescence. Because the disease progresses slowly, individuals may not lose their central vision until the sixth decade of life. ⁷ To date, there is still no cure or effective therapy for the treatment of RP. Mutations in more than 55 genes have been implicated in causing RP and associated syndromes. ⁸ In most cases, regardless of the underlying mutation, the outcome of the disease is the same—photoreceptor cell death by apoptosis. 

Apoptosis is a tightly controlled mechanism within the body that removes injured or nonfunctional cells that are deemed beyond repair. In most cases, apoptotic pathways converge on a family of cysteine proteinases known as caspases. ⁹ These caspases destroy key cellular targets, ultimately leading to cell death. The Bcl-2 and the inhibitor of apoptosis (IAP) gene families encode proteins that regulate apoptotic pathways. ¹⁰ However, only the IAPs are known to repress terminal caspase effectors. ¹¹ The IAPs were initially found in baculoviruses followed by their discovery in Drosophila, Caenorhabditis elegans (nematode), Saccharomyces cerevisiae (yeast), and many vertebrates. ¹² Inclusion in the IAP family is based on the presence of at least one approximately 80-amino-acid motif called the baculoviral inhibitor of apoptosis repeat (BIR), which contains a conserved cysteine and histidine core sequence Cx₂Cx₆Wx₃Dx₅Hx₆C. ¹³ The number of BIR domains can vary between one and three among the IAPs, but they are always present in the N terminus of the protein. The BIR domains and their linker regions are critical for antiapoptotic activity, because they are directly responsible for the inhibition of distinct caspases. ^{14 15 16} Of all the IAPs, the X-linked inhibitor of apoptosis (XIAP) is the most potent. It inhibits apoptotic cell death by directly binding to and inhibiting caspase-9, ^{17 18} an initiator caspase, and caspase-3 and -7, the effector caspases. ^{19 20} Recent evidence has suggested that members of the IAP family, including XIAP, have E3 ligase activity and can ubiquinate the caspases they bind, thereby targeting caspases for proteosome degradation. E3 ligase activity is mediated by the carboxyl-terminal RING zinc finger domain. ^{21 22 23}  

XIAP has been shown to confer resistance to apoptosis in a variety of cell death models. In the four-vessel occlusion (4-VO) model of forebrain ischemia, it has been shown that adenovirus (Ad)-mediated overexpression of XIAP prevents both the production of catalytically active caspase-3 and the degeneration of CA1 neurons in the hippocampus. ²⁴ In the 6-hydroxydopamine model of Parkinson’s disease, recombinant Ad-XIAP injection into the striatum protects dopaminergic neurons both histologically and functionally. ^{25 26} In the eye, intravitreal gene delivery of XIAP using an adenoassociated virus (AAV) shows protection of optic nerve axons in a hypertensive rat model of glaucoma ²⁷ and Ad-XIAP has been shown to protect axotomized retinal ganglion cells from cell death. ²⁸  

In the present study, we examine the neuroprotective effects of XIAP in a chemotoxic model of retinal degeneration. N-methyl-N-nitrosourea (MNU), has been used extensively to study photoreceptor apoptosis in various animal models. ^{29 30 31 32} It is an alkylating agent that causes DNA methylation at the O⁶ position of guanine leading to the formation of the O⁶-methylguanine adduct (O⁶-meGua). This methylated nucleotide pairs with thymine instead of cytosine, leading to GC→AT transition mutations. ³³ The O⁶-meGua adduct can be detected in photoreceptor cell nuclei within 12 hours of an intraperitoneal (IP) injection of MNU, and has been shown to activate caspases. ^{32 34} Photoreceptor apoptosis peaks at 24 hours and continues through day 7. ³²  

We show in this study that XIAP can protect photoreceptors at both the structural and functional level against MNU-induced apoptosis. We believe that because XIAP promotes such potent protection in the MNU model, which is one of the most severe retinal degeneration models available, it holds great promise for the treatment of less severe and much more slowly progressing forms of retinal degeneration, such as RP. 

Methods

Animals

Forty-six 6-week-old, male Sprague-Dawley rats were purchased from Charles River Laboratories (Wilmington, MA). The rats were cared for in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and the guidelines of the University of Ottawa Animal Care Committee. Rats were reared under standard laboratory conditions (22 ± 2°C, 60% ± 10% relative humidity, and a 12-hour light–dark cycle) and had free access to food and water throughout the experiment. 

Construction of the rAAV Vectors Expressing XIAP or GFP

A cDNA expression construct encoding the full-length human XIAP open reading frame (ORF) with an N-terminal hemagglutinin (HA) tag was constructed in the pTRUF2 vector. The pTRUF2 vector contains the mouse opsin promoter (MOP500), which restricts expression to the photoreceptors. The resultant construct, mOp-HA-XIAP-rAAV, and mOp-gfp-rAAV (described elsewhere ³⁵ ) were used to package, purify, concentrate, and titer recombinant AAVs, as previously described. ^{36 37} The titers of rAAV serotype 2 vectors using the opsin promoter (rAAV-XIAP and rAAV-green fluorescent protein [GFP], respectively) were adjusted to 4 × 10¹² physical particles/mL. Ratios of physical-to-infectious particles were all lower than 100. For production of these vectors, a mini-Ad helper plasmid pDG ³⁸ was used to produce rAAV vectors with no detectable adenovirus or wild-type AAV contamination. rAAV vectors, purified using iodixanol gradient/heparin-affinity chromatography, were more than 99% pure, as judged by polyacrylamide silver-stained gel electrophoresis (data not shown). Contaminating helper adenovirus and wild-type AAV, assayed by serial dilution cytopathic effect or infectious center assay respectively, were below detection levels. 

Subretinal Injections

Rats were anesthetized by halothane gas inhalation for the entire injection procedure. The subretinal injections were performed as previously described. ³⁹ Briefly, the right eye of each rat was dilated fully with 1% tropicamide and 2.5% phenylephrine hydrochloride (Alcon Canada, Mississauga, Ontario, Canada). A drop of 0.5% proparacaine (Alcon Canada) was used as a topical anesthetic. The vibrissae of the rats were matted down with a lubricant (KY Jelly; Johnson & Johnson, New Brunswick, NJ) to obtain an unobstructed view of the eye. Hydroxypropyl methylcellulose (Gonak; Akorn, Buffalo, NY) was applied to the cornea to keep the eye hydrated and to obtain a clear view of the retina. The cornea was punctured with a 28-gauge needle approximately 1 mm from the dilated pupillary margin. A 33-gauge blunt needle (Hamilton, Reno, NV) was inserted through the corneal puncture, maneuvered around the lens displacing it medially, and advanced through the retina. A 2-μL volume of virus with fluorescein tracer was injected into the subretinal matrix of the eye. The 33-gauge blunt needle was slowly withdrawn and an anti-infective agent (Maxitrol; Alcon) was applied to the cornea to prevent infection. The subretinal injection induced a localized retinal detachment that was readily visualized, due to the presence of the fluorescein tracer. Thirty-six rats were injected subretinally with rAAV-XIAP, and 10 rats received rAAV-GFP. 

Experimental Procedure

MNU (Sigma, St. Louis, MO) was dissolved in physiological saline to a final concentration of 10 mg/mL immediately before use and kept on ice. Six weeks after unilateral subretinal injection of rAAV vectors, rats received a 60 mg/kg dose of MNU by IP injection. ERGs were recorded, and animals were sampled at 0, 24, 48, and 72 hours and at 1 week after MNU injection. Each time point had two animals injected with rAAV-GFP and six animals injected with rAAV-XIAP. 

Electroretinography

All rats were weighed and dark-adapted overnight before ERG analysis. Under safe-light conditions, rats were given an anesthetic cocktail consisting of 80 mg/kg ketamine and 6.4 mg/kg xylazine. Pupils were dilated with 1% tropicamide and a topical anesthetic (0.5% proparacaine) was applied. 

DTL microconductive fiber electrodes were placed on each eye. A gold minidisc reference electrode was moistened in saline and placed on the tongue. A ground needle electrode was placed subcutaneously in the tail. Hydroxypropyl methylcellulose was applied to both eyes to maintain corneal hydration and to assist in the appropriate positioning of the DTL electrode. The rat was then positioned inside a Ganzfeld (model GS2000; Nicolet Instruments, Madison, WI) facing the rear of the globe before ERG recording. ERGs were amplified 50,000 times and recorded (Pathfinder II; Nicolet Instruments) with a band-pass filter of 0.3 to 300 Hz. Twenty ERG traces were averaged over a luminance range of −3.0 to 1.4 log cd-s/m². ERGs were recorded at pretreatment baseline (time 0) and at 24, 48, and 72 hours and 1 week after MNU injection. 

Tissue Fixation and Processing

At each of the five time points, eight animals (two GFP-treated and six XIAP-treated) were transcardially perfused with 4% paraformaldehyde (PFA) for tissue fixation. The eyes were scored with a white-hot 18-gauge needle before their removal for orientation purposes during embedding. After enucleation, a portion of the cornea was removed, and the eyes were placed in 4% PFA overnight. Eyes were washed three times in PBS, the lenses were removed, and the eye cups were placed in 30% sucrose and left at 4°C until saturated. The eyes were then placed in a 1:1 mixture of 30% sucrose-ornithine carbamoyltransferase (OCT) compound and allowed to equilibrate in the mixture for 1 to 2 hours at 4°C. The left and right eye cups from each rat were placed in plastic base molds, filled with 1:1 OCT-sucrose, cornea side up, and aligned by their score mark. The mold was lowered carefully onto a Petri dish floating on liquid nitrogen. Once completely frozen, the molds were transferred to a −80°C freezer where they were stored until sectioned. Cryosections (16 μm) were prepared using a Shandon cryostat, air dried for 3 hours, and stored at −20°C with desiccant. 4′-6-diamidino-2-phenylindole (DAPI) and hematoxylin and eosin staining was performed according to standard protocols. 

TUNEL Staining

Cell death assays were performed on retinal sections, by using a peroxidase in situ detection kit according to the manufacturer’s instructions (Apotag Plus; Intergen, Purchase, NY). TUNEL sections were counterstained with eosin. 

mRNA Expression Levels

Three rats were killed 12 weeks after undergoing a subretinal injection of rAAV-XIAP. The eyes were enucleated and the retinas dissected. RNA was isolated with extraction reagent (Trizol; Gibco, Grand Island, NY) and further purified (RNeasy column; Qiagen, Valencia, CA) according to the manufacturer’s instructions. The purified RNA was quantitated and real-time RT-PCR analysis (Taqman; Applied Biosystems, Foster City, CA) was performed on XIAP-injected eyes and uninjected contralateral control eyes with XIAP-specific primers. ⁴⁰  

Protein Expression Levels

Three rats were killed 6 weeks after receiving a subretinal injection of rAAV-XIAP. The eyes were enucleated, and the retinas dissected. Each retina was homogenized in 200 μL RIPA buffer (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 1% NP-40, 0.25% Na-deoxycholate, 1 mM NaF, 1 mM Na₃VO₄, 1 mM phenylmethylsulfonyl fluoride [PMSF], and 10 μg/mL each of leupeptin and aprotinin). Thirty micrograms of protein from XIAP-injected retinas and uninjected contralateral controls was electrophoresed on a 10% polyacrylamide gel and transferred onto polyvinylidene difluoride (PVDF) membrane (Immobilon P; Millipore, Bedford, MA). Blots were probed with antibodies to RIAP3 (rat XIAP) or to the HA epitope tag (Roche Molecular Biochemicals, Indianapolis, IN). 

Results

XIAP Overexpression in the Retina

Real-time PCR analysis (Taqman; Applied Biosystems) with primers specific to human XIAP (but which also cross-hybridize to endogenous rat IAP) verified XIAP overexpression in the right rAAV-XIAP–injected eye in relation to endogenous IAP levels in the uninjected left eye (data not shown). GAPDH was used as an internal control to confirm the integrity of the mRNA. Western blot analysis confirmed that overexpression of XIAP also occurred at the protein level (Fig. 1) . The XIAP-injected retina showed high levels of HA-tagged protein, which was absent in the contralateral control (Fig. 1A) . Studies with the RIAP3 antibody (which binds with both the human XIAP transgene and the endogenous rat IAP) showed a clear overexpression in the XIAP-injected eye in comparison with the contralateral uninjected eye (Fig. 1B) . Given that the protein extract was made from the whole retina, and only a fraction of the retina was covered by the subretinal injection, the level of XIAP overexpression at the site of the injection would be even greater than the Western blot indicates. 

TUNEL Staining

To assess whether XIAP protects the photoreceptor cells from acute chemotoxic insult with MNU, TUNEL staining was performed. At 24 hours after MNU treatment, the rAAV-XIAP–injected eye showed significantly fewer TUNEL-positive (apoptotic) cells than the uninjected eye of the same animal (Fig. 2) . Eyes that received an rAAV-GFP subretinal injection had apoptotic profiles similar to those of uninjected controls (data not shown). 

Histologic Protection of Photoreceptors

Hematoxylin and eosin (H&E) and DAPI staining techniques were used to assess the degree of histologic protection by XIAP against MNU-induced photoreceptor apoptosis. Histologic loss of photoreceptors was not evident at 24 hours after MNU, although the rAAV-XIAP–injected eyes showed fewer TUNEL-positive cells. With both H&E and DAPI staining, the retinal morphologies of the rAAV-GFP–injected, rAAV-XIAP–injected and uninjected eyes were indistinguishable (Figs. 3A 3B 3C 4C 4D) . At 48 hours there was wide-spread destruction of the photoreceptors in the uninjected and GFP-injected eye, whereas the rAAV-XIAP–injected eye showed preservation of the outer nuclear layer (ONL; Figs. 3D 3E 3F ; Figs. 4E 4F ). The differences between AAV-XIAP–injected and control eyes were further accentuated at 72 hours after MNU, when there was virtually complete destruction of the ONL in the uninjected and rAAV-GFP–injected eyes, but continued protection from apoptosis in the rAAV-XIAP–injected eye (Figs. 3G 3H 3I ; Figs. 4G 4H ). At 1 week, the photoreceptors of the control eye were essentially destroyed, whereas rAAV-XIAP–treated retina still showed structural integrity of the photoreceptors at the injection site (Figs. 3J 3K 3L , Figs. 4I 4J ). Overall, XIAP-injected eyes showed histologic protection of the outer nuclear layer in 66% of the animals (four of six) at each of the time points tested. The greatest degree of protection was seen in the vicinity of the subretinal injection. Protection diminished and eventually disappeared, moving away from the injection site. 

Functional Protection of Photoreceptors against MNU

ERG recordings were taken to ascertain whether functional protection correlated with histologic rescue at designated times after MNU treatment. At 24 hours after MNU, both the XIAP-treated and the control eye showed reductions in both the a- and b-wave amplitudes. This trend continued at the 48- and 72-hour time points (data not shown). However, at 1 week after MNU, the left eye showed an extinguished ERG, whereas the rAAV-XIAP–treated eye showed a diminished but recordable ERG response in two of the four animals that showed morphologic protection (Fig. 5) . These animals had b-wave amplitudes of 10.2 and 26.0 μV at an intensity of 0.25 cd-s/m². The average normal b-wave amplitude at this intensity is 206.4 ± 81.2 μV. Thus, the 1-week ERG in the rAAV-XIAP–treated eyes retained 5% to 15% of the b-wave amplitude in comparison with the pre-MNU baseline. 

Discussion

In the present study, AAV was used to deliver XIAP to the photoreceptors of the retina. AAV was chosen over adenovirus or lentivirus for several reasons. Adenoviral vectors have immune response and persistence problems (see reviews in Refs. ^{41 42 43} ). The new-generation “gutted” adenoviruses have fewer of these problems, but tissue tropism is a concern. Mûller cells are the primary cell transduced on intravitreal delivery, ^{42 43} and RPE cells are efficiently transduced by subretinal injection. ⁴³ This is problematic if ganglion cells or photoreceptors are the desired target. Lentiviral vectors show tremendous promise, because they can infect a variety of cell types and they can integrate in post-mitotic cells. ^{41 43} However, the possibility of low-level infectious wild-type human immunodeficiency virus (HIV) contaminants remains a safety concern. AAV vectors, at present, appear to be the most promising for ocular gene therapy. Although rAAV has a limited packaging capacity (∼4 kb), it is minimally toxic to photoreceptors, even at high dosages and initiates only a minimal immune response. ^{42 44} Expression is long lived, persisting for at least 1 to 2 years after delivery. ⁴⁵  

Our results show that rAAV-XIAP conferred significant protection to the photoreceptor layer of the retina. No such protection was seen in rAAV-GFP–injected eyes and the uninjected control. The XIAP protective effect was first apparent at 24 hours after MNU, where a much larger number of cells were undergoing apoptosis in the left, uninjected eye (and in the GFP injected eye) in relation to the rAAV-XIAP–injected right eye. Histologic protection was seen in four of six animals up to 1 week after MNU, when the experiment was terminated. Most significantly, a good proportion (two of four) of the eyes that showed morphologic protection also showed functional protection, as evidenced by ERG. It is important to note that there is only one report in the literature ²⁹ showing any type of protection after MNU treatment. In this study, intravitreal delivery of a caspase-3 inhibitor at 0 and 10 hours after MNU was used to obtain some degree of structural protection. A double injection of the inhibitor was delivered, and still no functional protection was shown. Our study with XIAP represents the first and only evidence of functional protection of photoreceptors after treatment with MNU. The degree of functional protection was somewhat limited, and there are three possible explanations for this. First, the 2-μL injection of the virus covered, at best, 10% to 20% of the rat retina ³⁵ ; consequently, only a fraction of the photoreceptors were efficiently transduced with the rAAV vector. This was confirmed by histologic data that show regions of morphologic protection adjacent to unprotected areas that lie outside the reach of the virus. The full-field ERG records the response of the whole retina, and thus the responses of fully protected regions were considerably diluted by most of the retina, which was not covered by the injection and underwent cell death. Clearly, full-field ERG significantly underestimates the degree of localized functional protection provided by rAAV-XIAP. Because our studies showed that XIAP-injected eyes retained 5% to 15% of pre-MNU b-wave amplitudes and we estimate that at best only 20% of the retina was covered by the subretinal injection, this emphasizes the strong protective effect of XIAP in transfected photoreceptors. Second, because XIAP overexpression cannot easily be tested in a retina that is severely stressed by MNU, we cannot be sure that all the subretinal injections were equally effective in overexpressing XIAP. Subtle differences in the injection procedure from one animal to the next could potentially result in differences in level of XIAP expression and in significant differences in morphologic and functional protection. Third, and perhaps most important, MNU is a powerful alkylating agent ^{31 32} and is expected to cause severe damage to the DNA, even before the cell initiates apoptotic signals. It is thus remarkable that any functional protection was observed at all. 

Given that XIAP is able to confer functional protection of the photoreceptors in an acute model of retinal degeneration, it holds promise as a therapeutic agent in slower retinal degenerations. More than 150 genes, a large number of which are expressed in the photoreceptors, are implicated in causing retinal dystrophies when mutated (Daiger S, Rossiter B, Greengerg J, Christoffels A, Hide W, ARVO Abstract 1352, 1998). Mutations impede the cell’s ability to function optimally and cause an inability to repair damage induced by extrinsic (intense light) ^{46 47} or intrinsic (genetic) factors. ^{48 49 50} For RP alone, at least 32 different genetic loci have been identified, and 21 genes have been cloned (summarized in the Retinal Information Network [RetNet]; http://www.sph.uth.tmc.edu/retnet/home.htm, provided by the University of Texas Houston Health Science Center). Every chromosome has a least one RP locus. ⁸ The end point, regardless of the mutation involved, is the death of the photoreceptors by apoptosis. 

Given all the different genes that can cause RP, targeting individual mutations for gene therapy strategies may not be a practical approach. The results from the MNU model suggest that XIAP may allow the broad protection of photoreceptors in many forms of RP, regardless of the initial disease-causing mutation. Moreover, the long time course and the progressive nature of many retinal degenerations implies that even if intervention is implemented after diagnosis, there is still hope that antiapoptotic strategies can effect a significant delay, if not a full halt, in disease progression. In RP, all the photoreceptors in an affected individual have the mutation, yet they function normally for many years before they begin to die. Photoreceptor cell death continues to occur slowly over a long period. It is possible that if the apoptotic threshold in the diseased photoreceptor cells could be raised, the cells could survive and function for longer periods. 

 Supported by Grant T-GT-0999-0024 from the Foundation Fighting Blindness, a Canadian Genetic Diseases Network strategic grant, and the Ottawa Hospital Foundation.

 Submitted for publication July 17, 2002; revised November 19, 2002; accepted November 25, 2002.

 Disclosure: D. Petrin, None; A. Baker, None; S.G. Coupland, None; P. Liston (P); M. Narang, None; K. Damji, None; B. Leonard, None; V.A. Chiodo, None; A. Timmers, None; W. Hauswirth, None; R.G. Korneluk, Aegera Therapeutics (F, I, C, P); C. Tsilfidis, Aegera Therapeutics (C)

 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.

 

Western blot confirming XIAP overexpression in the right eye of a rat given a subretinal injection of rAAV-XIAP. Evidence for the presence of the transgene is found in (A), which was probed with an antibody to the HA tag on the transgene. The level of overexpression is shown in (B), which was probed with an antibody to riap3. This antibody reacts with endogenous rat IAP as well as the XIAP transgene. Arrows: bands of interest.

Figure 2.

 

TUNEL staining at 24 hours after MNU treatment. Note the fewer TUNEL-positive cells in the outer nuclear layer (ONL) of the XIAP-injected right eye (B) in comparison with the uninjected left eye (A).

Figure 3.

 

Retinal sections at selected time points after MNU injection. No histologic differences were apparent at 24 hours after MNU (A, B, C). By 48 hours after MNU, the XIAP-injected outer nuclear layer (F) had many more photoreceptor nuclei than in the GFP-injected eye (D) or uninjected control (E). Differences were much more pronounced at 72 hours (compare I with G, H). At 1 week, XIAP-injected eye showed preserved morphology of the outer nuclear layer (L), but this layer had completely degenerated in the GFP-injected (J) and uninjected control (K).

Figure 4.

 

DAPI staining of XIAP-injected (B, D, F, H, J) and uninjected (A, C, E, G, I) retinas. Morphologic integrity was preserved in the XIAP-treated outer nuclear layer (ONL) up to 1 week after MNU treatment. The ONL was destroyed in uninjected retinas at 72 hours and 1 week (G and I, respectively).

Figure 5.

 

Representative full-field flash ERGs from a control rat (two upper traces) before MNU injection. The lower two traces are ERGs from a XIAP-treated right eye (OD) and untreated left eye (OS) of an experimental subject at 7 days after MNU treatment. The intensity of the flash in all four traces was 0.25 cd-s/m². The XIAP-treated eye showed a recordable ERG with greatly reduced a- and b-wave amplitudes. The unprotected eye showed an extinguished and nonrecordable ERG trace.

The authors thank Martine St-Jean and Sandra Hurley for expert technical assistance.