Intraocular Adenoviral Vector-Mediated Gene Transfer in Proliferative Retinopathies

Abstract

Purpose. The purpose of this study was to compare levels and patterns of expression of reporter genes achieved with an E1-deleted and partially E3-deleted type 5 adenoviral (Ad) vector after intravitreous or subretinal injections, or after intravitreous injections in mouse eyes with proliferative retinopathies.

Methods. Ad vectors containing reporter gene constructs were injected into the vitreous cavity or subretinal space of wild-type mice or mice with proliferative retinopathies, and quantitative comparisons were made of expression of transgenes.

Results. In normal eyes, peak Ad-mediated expression of luciferase, driven by a cytomegalovirus (CMV) promoter, occurred after injection of 10⁷ to 10⁸ viral particles and was 10 times greater after subretinal injections than after intravitreous injections. Intravitreous injections of Ad containing β-galactosidase (LacZ) expression constructs (AdLacZ.10) resulted in strong expression of LacZ in epithelial cells of the iris and ciliary body and focal expression in the retina. Subretinal injections of AdLacZ.10 resulted in strong expression in RPE cells. Expression of LacZ after intravitreous injection of AdLacZ.10 was significantly greater in mice with two types of proliferative retinopathy (ischemic retinopathy or transgenic mice with retina-specific expression of platelet-derived growth factor (PDGF)-BB or PDGF-AB) than littermate control animals. Cells within epiretinal membranes and activated Müller cells were preferentially transduced in eyes with proliferative retinopathy.

Conclusions. These data suggest that although higher intraocular expression levels can be achieved after subretinal injection of adenoviral vectors, intravitreous injections provide good transduction of cells lining the vitreous cavity. Compared with normal eyes, eyes with proliferative retinopathy showed increased transduction, which occurred preferentially in cells participating in the disease process. Intravitreous injection of adenoviral vectors containing appropriate expression constructs may provide a good strategy for acute treatment of proliferative retinopathies, such as diabetic retinopathy and proliferative vitreoretinopathy.

Delivery of therapeutic agents to the retina is complicated by the blood–retinal barrier. The delivery of proteins is particularly problematic and may require repeated intraocular injections, which are risky and inconvenient. Gene therapy is an appealing alternative, because a single intraocular injection of a vector containing an expression construct may result in sustained levels of a transgene product in the eye, with little or no systemic exposure. 

The choice of vector depends on many factors, including tropism of target cells, efficiency of transduction, size of the transgene, and needs regarding latency and duration of expression. Replication-defective adenoviral vectors have the advantages of large capacity (8 kb), short latency, good levels of expression in many settings, transduction of both dividing and nondividing cells, and relative ease of growth, allowing concentration to high titers. Previous studies have investigated intraocular injection of adenoviral vectors containing reporter gene constructs in wild-type animals and have demonstrated that intravitreous injections result in transduction of anterior segment tissues and some limited transduction of ganglion cells, whereas subretinal injections result in efficient transduction of retinal pigmented epithelial (RPE) cells, but relatively poor transduction of photoreceptors, except in neonatal mice. ^{1 2 3 4 5 6}  

Transduction in a diseased eye may differ from that in a normal eye. We sought to explore the possible role of adenoviral gene transfer as a treatment for proliferative retinopathies, and therefore we compared transfer of adenovirus-mediated reporter genes in normal mice and in mice with proliferative retinopathies. 

Materials and Methods

Production of Adenoviral Vectors Expressing Reporter Genes

Adenoviral vectors were constructed that express β-galactosidase (LacZ) or luciferase from a cytomegalovirus (CMV) immediate early promoter expression cassette. Similar adenoviral vectors containing a CMV expression cassette in the E1 region of the vector have been described. ⁷ The full-length open reading frame for β-galactosidase or luciferase was cloned into a pAdCMV shuttle plasmid that contains the expression cassette and the left end of the adenovirus genome. ⁷ Adenoviral vectors were generated by transfection of complementing cells, with the plasmid carrying the complete genome of the virus. ⁸ The production and quantification of these vectors have been described. ⁷  

Intraocular Injection of Vector Constructs

Intravitreous or subretinal injections of vector constructs were performed as previously described ⁹ in C57BL/6 mice, C57BL/6 mice with oxygen-induced ischemic retinopathy, ¹⁰ or transgenic mice that express platelet-derived growth factor (PDGF)-A, ¹¹ PDGF-B, ¹² both PDGF-A and -B, or littermate control animals. Intraocular injections were performed with a Harvard pump microinjection apparatus and pulled-glass micropipettes. Each micropipette was calibrated to deliver, on depression of a foot switch, 1 μL vehicle containing viral particles. The mice were anesthetized, pupils were dilated, and under a dissecting microscope, the sharpened tip of the micropipette was passed through the sclera just behind the limbus into the vitreous cavity, and the foot switch was depressed. Subretinal injections were performed with the aid of a condensing lens system on the dissecting microscope, which allowed visualization of the retina during the injection. The pipette tip was passed through the sclera posterior to the limbus and was positioned just above the retina. Depression of the foot switch caused the jet of injection fluid to penetrate the retina. The blebs, which are areas of retinal detachment caused by the subretinal injection of vector, were uniform in size. 

Mouse Model of Oxygen-Induced Ischemic Retinopathy

Ischemic retinopathy was produced in C57BL/6 mice by a method described by Smith et al. ¹⁰ Seven-day-old (postnatal day [P]7) mice and their mothers were placed in an airtight incubator and exposed to an atmosphere of 75% ± 3% oxygen for 5 days. Incubator temperature was maintained at 23 ± 2°C, and oxygen was continuously monitored with an oxygen controller (PROOX model 110; Reming Bioinstruments Co., Redfield, NY). At P12 or P16, mice were given an intravitreous injection of 10⁸ particles of AdLacZ.10 and then killed on P16 or P21, respectively. Four or five days after injection was selected as the time point for examination of expression of transgenes, because it is within the period of peak expression. Eyes were rapidly removed and histochemically stained for LacZ, as described later. 

Transgenic Mice with Increased Expression of PDGF in Photoreceptors

Transgenic mice with increased expression of PDGF-A or -B in photoreceptors have been described. ^{11 12} PDGF-B transgenic mice show development of spontaneous traction retinal detachment and model aspects of proliferative vitreoretinopathy (PVR). Mice with increased expression of both PDGF-A and -B were generated by mating PDGF-A mice with PDGF-B mice. Transgenic mice and littermate control animals received an intraocular injection of vector on P5, P7, or P12 and were killed on P8, P12, or P17, respectively. The eyes were used for wholemounts or sections as described later. 

Luciferase Assay

Three days after intravitreous and subretinal injections of several doses (0, 10⁵, 10⁶, 10⁷, 10⁸, 5 × 10⁸, and 10⁹) of AdLuc.10, mice were killed, and eyes were enucleated. To assess the time course of expression after intravitreous injection, 5 × 10⁸ viral particles were injected, and eyes were enucleated at 3, 7, 14, and 30 days. Immediately after enucleation, eyes were snap frozen and stored at −80°C. Eyes were ground with a cold mortar and pestle on dry ice and lysed (Reporter Lysis Buffer; Promega, Madison, WI). Resultant lysates were analyzed with a luciferase assay system according to the manufacturer’s protocol (Promega). The total protein concentration was determined to normalize the measurement of luciferase expression based on a Bradford dye binding procedure with a protein assay (Bio-Rad, Hercules, CA). 

Histochemical Examination and Image Analysis

For histochemical analysis, eyes were fixed in 0.5% glutaraldehyde in phosphate buffer solution (PBS) for 1 hour and rinsed twice in 25% sucrose in PBS. Specimens were incubated in 25% sucrose in PBS overnight and embedded in optimal cutting temperature (OCT) compound (Bayer Diagnostics, Tarrytown, NY). Ten-micrometer frozen sections were rinsed in PBS and reacted overnight with 1 mg/mL 5-bromo-4-chloro-3-indolyl galactopyranoside (X-gal; Sigma, St. Louis, MO) in a solution containing 5 mM K₃Fe(CN)₆, 5 mM K₄Fe(CN)₆-³H₂O, and 1 mM MgCl₂ in PBS. Sections were postfixed for 15 minutes in 0.5% glutaraldehyde in PBS and washed in PBS. Some sections were also histochemically stained with biotinylated Griffonia simplicifolia lectin B4 (GSA; Vector Laboratories, Burlingame, CA), which selectively binds to vascular cells. Slides were incubated in methanol-H₂O₂ for 10 minutes at 4°C, washed with 0.05 M Tris-buffered saline (TBS; pH 7.6), and incubated for 30 minutes in 10% normal porcine serum. Slides were incubated for 2 hours at room temperature with biotinylated GSA, and, after they were rinsed with 0.05 M TBS, they were incubated with avidin coupled to peroxidase (Vector Laboratories) for 30 minutes at room temperature. After being washed for 10 minutes with 0.05 M TBS, slides were incubated with diaminobenzidine (Research Genetics, Huntsville, AL) to give a brown reaction product. 

For retinal wholemounts, eyes were fixed for 30 minutes in cold 4% paraformaldehyde in PBS and rinsed five times for 10 minutes in PBS. Whole eyes were incubated overnight in 1 mg/mL X-gal in a solution containing 5 mM K₃Fe(CN)₆, 5 mM K₄Fe(CN)₆-³H₂O, and 1 mM MgCl₂ in PBS. Eyes were postfixed for 30 minutes and wholemounted. To perform quantitative assessments, retinas were examined with a microscope (Axioskop; Carl Zeiss, Oberkochen, Germany), and images were digitized using a 3-charge-coupled device (CCD) color video camera and a frame grabber. Image-analysis software (Image-Pro Plus; Media Cybernetics, Silver Spring, MD) was used to delineate X-gal–stained areas of the retina, and the total area of staining was measured. 

Results

Adenoviral Vector–Mediated Expression of Reporter Genes in Retina and RPE of Wild-Type Mice

Several doses of AdLuc.10 were injected into the vitreous cavity or subretinal space of normal adult C57BL/6 mice. Three days after injection, mice were killed, and whole-eye homogenates were assayed for luciferase activity and protein content. Figure 1A shows relative luciferase units per milligram protein for the various vector doses. Expression plateaued at approximately 10⁷ to 10⁸ viral particles with intravitreous and subretinal injections, and maximal expression was roughly 1 log unit higher with subretinal injections. To assess the time course of expression after intravitreous injection, 5 × 10⁸ viral particles were injected, and assays were performed at 3, 7, 14, and 30 days. Expression was high at 3 days, decreased to low levels by day 14, and was close to baseline by 30 days (Fig. 1B) . 

The localization of expression of reporter genes was investigated with AdLacZ.10. Adult C57BL/6 mice were given an intravitreous injection of 5 × 10⁸ viral particles and retinal wholemounts were stained for LacZ at 3, 5, 7, 14, or 30 days after injection. The area of LacZ staining was quantified by image analysis. Expression of LacZ in the retina peaked on day 5 and then decreased by roughly 50% by day 30 (Fig. 2A) . LacZ staining of retinal wholemounts was performed 5 days after intravitreous injection of several different doses of viral particles. The area of LacZ-stained retina was highest after injection of 10⁹ particles, the highest dose used (Fig. 2B) . 

Five days after intravitreous injection of 5 × 10⁸ viral particles, there was strong staining for LacZ in the anterior segment, including the corneal endothelium, the trabecular meshwork, and the pigmented epithelium of the iris (Figs. 3A 3B) . There was also prominent staining in the ciliary epithelium, but only scattered focal staining in the retina (Fig. 3C) . Cross sections showed that the focal areas of staining were probably due to transduction of Müller cells and possibly some other cells (Figs. 3D 3E) . There was increased staining of cells in and around the optic nerve head (Fig. 3D) . 

Five days after subretinal injection of 10⁷ viral particles, LacZ staining was examined in wholemounts and cross sections. There was strong staining in the RPE (Fig. 3F) and little staining in the retina (Fig. 3G) , which again appeared to be due primarily to transduction of Müller cells. Choroidal flatmounts showed that staining was intense in all RPE cells within the region of the bleb from the subretinal injection, with little staining in regions where the retina had remain attached (Fig. 3H) . High-power views showed intense staining of all RPE cells throughout the region of the bleb (Fig. 3I) . 

Adenoviral Vector–Mediated Expression in a Model of PVR

In rho/PDGF-B mice, there is extensive proliferation of glial cells, pericytes, and RPE cells. ¹² These mice show spontaneous development of epiretinal membranes and traction retinal detachment and mimic aspects of PVR. In rho/PDGF-A mice, there is mild proliferation of glial cells, but not other cell types. They provide a model of mild nonvascular proliferative retinopathy, in which epiretinal membranes occur, but there is no traction retinal detachment. ¹¹  

At P7, wild-type and transgenic mice were given an intravitreous injection of 5 × 10⁸ particles of the vector containing the β-galactosidase expression cassette. At P12, wholemounts showed prominent staining of the optic nerve and mild focal staining throughout the retina in wild-type mice (Fig. 3J) . In rho/PDGF-A mice, there was increased LacZ staining in and around the optic nerve and increased focal staining of the retina (Fig. 3K) . Rho/PDGF-B mice showed a marked increase in focal staining throughout the retina (Fig. 3L) , as did mice that carried both PDGF-A and PDGF-B transgenes, which have a phenotype similar to that of mice that carry only a PDGF-B transgene (Fig. 3M) . Cross sections of PDGF-B or PDGF-AB mice showed strong staining in epiretinal membranes (Fig. 3N , arrowheads) and staining of linear structures within the retina (arrows) that probably represent Müller cells. Image analysis confirmed that the area of LacZ staining in the retina was significantly greater in rho/PDGF-B and -AB mice compared with that in wild-type or rho/PDGF-A mice (Fig. 4) . When mice were given an intravitreous injection of 5 × 10⁸ particles of the vector at P5 and then killed at P8, before there was much proliferation in the retina, there was no significant difference in the total area of LacZ staining among the four types of mouse (Fig. 4) . In cross sections, all four types of mouse at P8 showed more intense staining localized around the optic nerve (Fig. 3O) compared with older mice. In mice given an intravitreous injection of 5 × 10⁸ particles of vector at P7 or P12 and killed at P12 or P17, respectively, the area of LacZ staining was significantly greater in PDGF-B and -AB mice than in wild-type littermates (Fig. 4) . LacZ staining was particularly strong in epiretinal membranes (Figs. 3N 3P) . 

Adenoviral Vector-Mediated Expression in a Model of Ischemic Proliferative Retinopathy

Mice with oxygen-induced ischemic retinopathy mimic many aspects of ischemic retinopathy that occurs in human diseases, such as retinopathy of prematurity or proliferative diabetic retinopathy. ¹⁰ Control C57BL/6 mice or mice with ischemic retinopathy were given intravitreous injection of 5 × 10⁸ particles of AdLacZ.10 on P12 or P16 and killed on P17 or P21, respectively. At both time points, there was much greater LacZ staining in the retina of mice with ischemic retinopathy (Figs. 5C 5D 5G 5H 5I 5J 5K 5L) compared with the corresponding control (Figs. 5A 5B 5E 5F) . The expression was particularly strong in vascular tissue at the optic nerve (Fig. 5C , arrows), and in neovascularization on the surface of the retina (Figs. 5D 5H 5I 5J 5K 5L , arrowheads). At the P21 time point, there was prominent LacZ staining of Müller cells in some areas of ischemic retinas (Fig. 5H 5I 5J 5K 5L , arrows). 

Discussion

In this study, we have shown that in wild-type untreated mice, intravitreous injection of reporter genes packaged in E1-deleted and partially E3-deleted type 5 adenoviral vectors resulted in transduction of cells predominantly in the iris, cornea, and ciliary body, with only sporadic transduction of retinal cells. Subretinal injections resulted in good transduction of RPE cells and little transduction of cells in the overlying retina. In both cases, expression peaked within days and decreased to low levels within a month. These results are consistent with previously reported results in studies in which similar adenoviral vectors were used in wild type animals, although with some adenoviral vectors, greater transduction of Müller cells has been seen. ^{1 2 3 4 5 6} Differences in serotype or other characteristics of recombinant adenoviral vectors can result in alterations in tropism, and it is likely that small differences in the vectors used in our study compared with those used in some others account for the difference in transduction of Müller cells in wild-type animals. 

In two models of proliferative retinopathy, there was more adenoviral vector- mediated reporter gene expression than in normal mice. In a transgenic model in which epiretinal membranes caused traction retinal detachment, as is the case in humans with PVR, there was prominent expression of transgenes in the epiretinal membranes. In a model of proliferative retinopathy in which retinal nonperfusion and neovascularization occurs, similar to that in proliferative diabetic retinopathy, there was prominent expression of transgenes in neovascularization on the surface of the retina, in the optic nerve, and in Müller glia cells within the retina. Therefore, in proliferative retinopathies, adenoviral vector–mediated gene transfer is not only increased compared with that in normal retinas, but the greater expression occurs in pathologic tissue. Although adenoviral vectors can transduce nondividing cells, it is clear that this type 5 E1-deleted, E3-deleted adenoviral vector provides enhanced transduction of proliferating cells. When cells enter the cell cycle, the proteins expressed on the cell surfaces can be altered. Apparently, that alteration resulted in enhanced transduction of the adenoviral vector we used. This may provide a means through which expression of antiangiogenic or antiproliferative gene products are preferentially directed to sites where they are most needed. Müller cells also showed increased adenoviral-mediated transduction in the retinas of mice with proliferative retinopathies. Müller cells become activated after retinal detachment or other insults and increase their expression of glial fibrillary–associated protein. Activated Müller cells migrate and proliferate and participate in proliferative disorders, and therefore increased expression of an inhibitory protein by Müller cells may decrease their participation in the proliferative process and thereby have a beneficial effect. 

Toxicity consisting primarily of inflammation has been demonstrated after injection of some adenoviral vectors. There is one report of toxicity to the retina and RPE after subretinal injection of E1- and partial E3-deleted adenoviral vectors in rats. ¹³ Before use in humans, appropriate safety studies must be performed. All vectors have advantages and disadvantages. A key point of our study is that adenoviral vectors showed a previously unrecognized benefit for the treatment of proliferative retinopathies, because they preferentially transduced proliferative tissue. This is important information for investigators interested in ocular gene therapy. Because of this benefit, it may be worth the effort to further modify adenoviral vectors to decrease their inflammatory effects. Another potential benefit of the enhanced transduction of proliferating tissue is that such tissue is part of the pathologic process and is unwanted, so that any vector-mediated toxicity to those tissues could be beneficial. 

Recently, it has been demonstrated that adenoassociated viral (AAV) vectors result in long-term expression of transgenes in photoreceptors and, to a lesser extent, in RPE after subretinal injections. ¹⁴ These characteristics are advantageous for gene-replacement therapy in photoreceptors or RPE, or for long-term delivery of agents, such as neurotrophic factors. ¹⁵ However, AAV vectors have a long latency period of 4 to 6 weeks, which is a problem in treating acute disease. Proliferative retinopathies can progress quickly and often require rapid intervention. The risk of retinal detachment caused by PVR is greatest in the first 4 to 6 weeks after retinal reattachment surgery. The rapid onset of expression of transgenes and the preferential expression in epiretinal membranes and activated Müller cells after intravitreous injection of adenoviral vectors are ideal for this situation, and the time course of expression may be well suited for the time course of the disease process. 

In patients with proliferative diabetic retinopathy, clear media and an attached retina are needed to perform scatter photocoagulation, and when photocoagulation can be performed, it usually takes 3 weeks for it to achieve maximum effect. Antiangiogenic agents packaged in adenoviral vectors may be a useful way to achieve a beneficial effect in several situations—for instance, in patients with neovascularization and vitreous hemorrhage that prevents scatter photocoagulation. The rapid onset of expression of transgenes and targeting of expression to the optic nerve, neovascular tissue, and activated Müller cells provides a potential way to treat neovascularization, allowing time for the vitreous hemorrhage to clear and scatter photocoagulation to be performed. Severe neovascularization and traction retinal detachment present another particularly difficult treatment dilemma. If some regression of neovascularization can be achieved with adenoviral vector–mediated gene therapy, there is likely to be less bleeding during surgery, and therefore an improved surgical prognosis. Patients with severe neovascularization and macular edema often need aggressive scatter photocoagulation which exacerbates macular edema. Some regression of neovascularization with adenoviral vector–mediated gene therapy would allow more gradual photocoagulation which could result in less exacerbation of macular edema. 

Recently, we have demonstrated that systemic administration of adenoviral vectors containing an expression construct for endostatin or intraocular injection of adenoviral vectors containing an expression construct for pigment epithelium–derived factor inhibit intraocular neovascularization. ^{9 16} The present study demonstrates that despite the transient expression profile achieved with adenoviral vectors, they may be well suited to deliver antiangiogenic agents in patients with ischemic proliferative retinopathies or antiproliferative agents in patients with PVR. 

 Supported by grants from the Michael Panitch Fund; Foundation Fighting Blindness; Grants EY05951, EY12609, K08 EY13420 (PG), and P30EY1765, from the National Eye Institute; the Juvenile Diabetes Research Foundation (PG); Knights Templar (PG); a Lew R. Wasserman Merit Award (PAC) and unrestricted funds from Research to Prevent Blindness; and the Steinbach Foundation. PAC is the George S. and Dolores Dore Eccles Professor of Ophthalmology.

 Submitted for publication July 13, 2001; revised December 27, 2001; accepted January 15, 2002.

 Commercial relationships policy: E (VG, DM, LW); N ( all others).

 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.

 

Effect of dose and site of injection of adenoviral vectors on luciferase expression in mouse eyes. Each data point represents the mean ± SEM of eight experimental results. (A) Luciferase expression peaked 3 days after intravitreous or subretinal injection of 10⁷ to 10⁸ AdLuc.10. Subretinal injections resulted in roughly 10 times higher peak expression than intravitreous injections. (B) Time course of expression after intravitreous injection of 5 × 10⁸ particles of AdLuc.10. Expression was high at 3 days and decreased to low levels by 30 days.

Figure 2.

 

Expression of LacZ in the retina after intravitreous injection of AdLacZ.10. Each data point represents the mean ± SEM of 10 experimental results. (A) Time course of expression of LacZ in the retina after intravitreous injection of 5 × 10⁸ particles of AdLacZ.10. The area of LacZ staining was measured by image analysis in retinal wholemounts at various times after injection. Expression peaked on day 5 and decreased by roughly 50% by day 30. (B) The area of LacZ staining in retinal wholemounts was measured by image analysis 5 days after intravitreous injection of several different doses of AdLacZ.10. The area of LacZ staining was greatest after injection of 10⁹ particles, the highest dose examined.

Figure 3.

 

Staining for LacZ 5 days after intraocular injection of AdLacZ.10 in adult wild-type mice and at various times after injection in transgenic mice with increased expression of PDGF in the retina. Mice were given an intravitreous injection of 5 × 10⁸ (A–E, J–P) or a subretinal injection of 10⁷ (F–I) AdLacZ.10 particles. After 5 days, or as otherwise indicated, ocular sections and retinal and choroidal wholemounts were stained for LacZ. (A) Albino BALB/c mice showed staining in the corneal endothelium, the trabecular meshwork, the iris pigmented epithelium, and the ciliary body (bar, 200 μm). (B) Flatmounts of iris from albino BALB/c mice showed diffuse dark staining throughout the entire iris. (C.) Retinal wholemounts from adult C57BL/6 mice showed scattered focal staining throughout the retina, with more intense staining at the optic nerve. (D, E) Ocular sections from adult C57BL/6 mice showed prominent staining in and around the optic nerve and focal staining of cells in the inner nuclear layer. The stained cells within the retina often showed vertical extensions between the internal limiting membrane and the external limiting membrane, suggestive of Müller cells. (F) After subretinal injection in C57BL/6 mice, there was intense staining of the RPE. On this section, development was longer than necessary resulting in spread of reaction product into the retina. (G) With shorter development times, focal staining was seen in retinal cells, with a morphology characteristic of Müller cells. (H) Choroidal flatmounts from adult albino BALB/c mice showed strong expression of LacZ in RPE cells within the region where the bleb was located and essentially no expression in the region where the retina had remained attached. A high-power view of this region (I) shows the heavily stained hexagonal RPE cells. (J) After intravitreous injection of vector, littermates of rho/PDGF-B transgenic mice that did not carry a transgene showed sparse focal LacZ staining in the, retina with greater staining at the optic nerve, similar to that seen in C57BL/6 or BALB/c mice. (K) Sections from Rho/PDGF-A transgenic mice showed focal staining similar to that seen in sections from wild-type animals. (L) After intravitreous injection of vector, Rho/PDGF-B transgenic retinas showed extensive expression of LacZ throughout, as did retinas in rho/PDGF-AB double-transgenic mice (M), which have a phenotype similar to that of rho/PDGF-B mice. (N) Ocular sections from rho/PDGF-B transgenic mice given an intravitreous injection of vector on P7 and killed on P12 showed expression of LacZ in epiretinal membranes (arrowheads), in linear structures within the retina consistent with Müller cells (arrows), at the optic disc, and in the hyaloid vessels. (O) Sections from Rho/PDGF-B transgenic and littermate control animals given an intravitreous injection of vector at P5 and killed at P8 showed intense staining in the optic nerve and hyaloid vessels, but no staining in the retina. (P) Rho/PDGF-B transgenic mice given an intravitreous injection of vector on P12 and killed at P17 showed strong expression of LacZ within the epiretinal membranes (arrowheads) that were causing retinal detachments. Scale bars, (A, D) 200 μm; (B, N–P) 400 μm; (C, G, J–M) 800 μm; (E, F, H, I) 100 μm.

Figure 4.

 

Intravitreous injection of AdLacZ.10 resulted in significantly greater expression of LacZ in the retina of mice with proliferative retinopathy than in littermate control animals. Rho/PDGF-A, -B, or -AB transgenic and littermate control animals were given an intravitreous injection of 5 × 10⁸ particles of AdLacZ.10 on P5, P7, or P12 and killed on P8, P12, or P17, respectively. Retinal wholemounts were stained for LacZ, and the total area of LacZ staining in each retina was measured on computer by image analysis. Each bar represents the mean ± SEM from between 4 and 30 experimental values. *P < 0.05 by unpaired t-test for difference from wild-type littermate control animals.

Figure 5.

 

Staining for LacZ after intraocular injection of AdLacZ.10 in mice with oxygen-induced ischemic retinopathy. Mice with ischemic retinopathy (C, D, G–L) and control animals (A, B, E, F) were injected with 5 × 10⁸ particles of AdLacZ.10 at P12 or P16 and killed at P17 (A–D) or P21 (E–L), respectively. Ocular sections were stained with G. simplicifolia lectin (brown), which selectively labels vascular cells, and for LacZ. In control mice, LacZ staining was limited to the optic nerve head. In mice with ischemic retinopathy, there was much greater LacZ staining in the retina than in corresponding control mice. Expression of LacZ was most prominent in vascular tissue on the surface of the optic nerve (C, arrows) or on the surface of the retina (D, G–L, arrowheads). There was also prominent staining of Müller cells (H–L, arrows). Scale bars, (A, C, E, G, K) 400 μm; (B, D, F, H, J, L) 100 μm; (I) 200 μm.