The Use of Adenovirus-Mediated Gene Transfer to Develop a Rat Model for Photoreceptor Degeneration

Abstract

purpose. To investigate the effects of recombinant adenovirus-mediated downregulation of cathepsin S (CatS) on the retinal pigment epithelium and/or neural retina in vivo.

methods. The expression of green fluorescent protein (gfp) after subretinal injection of a recombinant adenovirus, Ad.gfp, into rat eyes was first established by in vivo fundus fluorescence photography and fluorescence microscopy. The autofluorescent debris accumulation in Ad.CatSAS (recombinant adenovirus carrying the antisense CatS gene)-injected rat eyes was monitored by fluorescence microscopy, and the antisense CatS RNA expression was demonstrated by in situ hybridization. Changes in the retinal morphology were assessed by light microscopy.

results. The gfp expression was present in 30% to 90% of the injection area at 3 days and was absent 9 days after Ad.gfp injection. In Ad.CatSAS-injected eyes, the expression of antisense CatS RNA was demonstrated by in situ hybridization. Autofluorescent debris accumulation was significantly higher in Ad.CatSAS-injected eyes than in control eyes. The shortening of photoreceptor outer segments in Ad.CatSAS-injected eyes coincided with intense autofluorescent debris accumulation. The number of layers of photoreceptor cells decreased with time and were 11, 9, and 8 at 7, 14, and 28 days after Ad.CatSAS injection, respectively. In control eyes, the number of layers of photoreceptor cells (14) remained unchanged.

conclusions. These results demonstrate that recombinant adenovirus-mediated transient modulation of gene expression in retinal pigment epithelial (RPE) cells could induce changes in the retina, and, in spite of the low expression of endogenous CatS in RPE cells, this enzyme plays an important role in maintenance of normal retinal function.

Cysteine proteases, cathepsin B, H, L, and S, which are predominantly present in cells of mononuclear phagocytic origin are essential for the turnover of intracellular and extracellular proteins. The inhibition of cysteine protease activity in several organs, including the eye, has been shown to result in the accumulation of autofluorescent debris. ^{1 2} The accumulation of autofluorescent debris in the eye has been linked to several eye diseases, such as vitelliform macular dystrophy and, possibly, to age-related macular degeneration. Therefore, the development of models facilitating studies of autofluorescent debris accumulation in the eye is of great importance. Traditional peptide-based inhibitors have a short half-life and, thus, studies of their biologic effect on lysosomal enzyme inhibition require repeated injections. In this respect, recombinant adenovirus-mediated gene delivery, which can ensure transgene expression for an extended period, may offer an alternative technology for gene expression–related functional studies. 

There have been several reports demonstrating efficient transduction of retinal cells, particularly retinal pigment epithelial (RPE) cells, by recombinant adenoviruses. ³ Recently, our laboratory has successfully combined recombinant adenovirus-mediated gene delivery with antisense DNA technol-ogy by constructing a recombinant adenovirus carrying the cathepsin S (CatS) gene in antisense orientation (Ad.CatSAS). Ad.CatSAS transduces cultured RPE cells with high efficiency and downregulates endogenous CatS production in the transduced cells. ⁴ The present study was conducted to examine the feasibility of adenovirus-mediated gene delivery for the development of animal models. The transduction efficiency of this delivery system is high but transient. ^{3 5} Therefore, in this work, the longevity, intensity, and area of real-time transgene expression using Ad.gfp, a recombinant adenovirus construct carrying the green fluorescent protein (gfp) gene was first established before investigating if subretinal injection of Ad.CatSAS could induce any changes in the RPE cell layer or in the neural retina of the rat eye. 

Materials and Methods

Construction, Production, and Delivery of Recombinant Adenoviruses

Ad.gfp was constructed by first subcloning the XbaI/EcoRI fragment of the gfp gene from the plasmid pEgfp-N1 (Clonetech, Palo Alto, CA) into the XbaI/EcoRI restricted recombinant adenovirus vector, pCA13 (Microbix Biosystems, Toronto, Ontario, Canada). The DNA from the resultant plasmid, pCA13gfp, was then cotransfected with ClaI-digested Ad5 dl324 DNA into 293 cells, selected, propagated, and purified as described earlier for Ad.CatSAS and Ad.CatSS. ⁴ The titers used of Ad.gfp, Ad.CatSAS, and Ad.CatSS were 3 × 10¹⁰, 10¹¹, and 10¹¹ plaque-forming units per milliliter, respectively. Two microliters of each viral preparation or phosphate-buffered saline containing 10% glycerol (vehicle) was injected into the subretinal space of 6 week-old normal RCS rdy ⁺ rats. ⁶ All animal experimentations adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Assessment of gfp Gene Expression

The expression of gfp in Ad.gfp-injected eyes (n= 9) was followed by in vivo fundus fluorescence photography at 3, 9, and 17 days after injection, as previously described. ⁶ The size of the gfp-expressing area within the bleb created in each eye was assessed by two independent observers. After confirmation of gfp expression, the rat eyes were either used for wholemount preparation (n = 6) ⁶ or processed for plastic-embedding for histologic examination (n= 3). ⁷  

In Situ Hybridization

The sense and antisense digoxigenin (DIG)-labeled CatS riboprobes were prepared after linearization of the pGem11CatS plasmid ⁴ with NsiI and HindIII, respectively. The linearized plasmids were phenol and chloroform extracted and the sense and antisense probes were transcribed at 37°C for 2 hours with T7 or SP6 polymerase respectively, in the presence of DIG-11-uridine-triphosphate using the DIG RNA labeling kit (Boeh-ringer–Mannheim, Mannheim, Germany). The DIG-labeled riboprobes were then precipitated, and the efficiency of labeling and the sensitivity and specificity of the labeled riboprobes were checked using the protocol provided by the manufacturer. Ten-micrometer-thick cryosections of eyes enucleated 4 days after subretinal injection of vehicle or Ad.CatSAS were hybridized with sense and antisense CatS riboprobes, as previously described. ⁸ The hybridization temperature was 45°C and the posthybridization wash in 0.1× SSC/0.1% sodium dodecyl sulfate was performed at 50°C. The hybridized riboprobes were detected using the Nucleic Acid Detection Kit (Boeh-ringer–Mannheim) according to the manufacturer’s protocol. 

Evaluation of Retinal Morphology and Fluorescent Debris Accumulation after Subretinal Injection of Ad.CatSAS or Ad.CatSS

Eyes injected subretinally with Ad.CatSAS (n = 4) and Ad.CatSS (n = 4) were enucleated 7 days after injection and were immediately embedded in O.C.T. compound (Sakura Finetechnical, Tokyo, Japan). Sixty 10-μm thick serial sections of the injection area were prepared and examined by fluorescence microscopy. The intensity of the fluorescent signal in sections 1, 30, and 60 was analyzed by two independent observers and graded from 1 to 3. The retinal morphology, length of the photoreceptor outer segments (POSs), and number of layers of photoreceptor cells in Ad.CatSAS- and Ad.CatSS-injected eyes (n = 3) were assessed by light microscopic analysis of plastic or paraffin-embedded sections at 7, 14, and 28 days after injection. ⁷  

Results

Optimization of Transgene Delivery into the Retina

At 3 days after injection, red-free retinal photography of Ad.gfp-injected eyes (n = 9) demonstrated no changes in the retina other than a small bleed around the injection site (data not shown). The fundus remained normal throughout the course (28 days) of the study. In vivo fundus fluorescence photography demonstrated a strong expression of gfp at 3 days after injection (n = 9). The signal was intense (Fig. 1A ) and covered 90%, 80%, 75%, 70%, 60%, 50%, 40%, 30%, and 30%, respectively (mean ± SD of 58.3% ± 22.0%), of the area within the bleb. By 9 days after injection, the fluorescent signal was detected in only a few cells, and by 17 days after injection, no signal was detected (data not shown). Fluorescence microscopic examination of wholemount preparations of Ad.gfp-injected eyes (n = 6) showed that most of the gfp-expressing cells were hexagonal RPE cells (Fig. 1B) . No gfp signal was detected in the neural retina (data not shown), and not all the RPE cells within the bleb expressed gfp. The intensity of the gfp signal detected varied from weak (Fig. 1B , thin arrows) to strong (Fig. 1B , thick arrows). Histologic examination of Ad.gfp-injected eyes (n = 3) showed no signs of morphologic changes (Fig. 1C) . 

Expression of CatS Antisense RNA in the RPE Layer

Hybridization of cryosections of Ad.CatSAS-injected eyes at 4 days after injection with DIG-labeled sense CatS riboprobe resulted in a strong signal, seen as a black precipitate, (Fig. 2A , arrows) in the RPE cell layer in and around the injection sites. RNase treatment of serial sections of Ad.CatSAS-injected eyes before hybridization with the sense CatS riboprobe resulted in a significant decrease in the signal intensity (Fig. 2B) , confirming that the signal seen in Figure 2A was of RNA origin. The specificity of the sense CatS riboprobe was demonstrated by the absence of signal in control vehicle-injected eyes hybridized with the sense CatS riboprobe (Fig. 2C) and by the absence of signal in Ad.CatSAS-injected eyes hybridized with an antisense CatS riboprobe (Fig. 2D) . 

The Effect of Ad.CatSAS Delivery on Retinal Morphology

At 7 days after injection, gross morphologic examination of cryosections demonstrated some changes in the area corresponding to the bleb generated by the subretinal injection of Ad.CatSAS (Fig. 3A ). In comparison with control vehicle- (data not shown) or Ad.CatSS-injected eyes (Fig. 3B) , a significant shortening of photoreceptor layer was observed in Ad.CatSAS-injected eyes (Fig. 3A , double-headed arrow). Fluorescence microscopy of the same sections before hematoxylin and eosin staining revealed that in Ad.CatSAS-injected eyes, the morphologic changes closely correlated with the presence of an autofluorescent signal (Fig. 3C) . This fluorescent signal was granular in appearance (Fig. 3C , white arrows) and was significantly more intense than the evenly distributed autofluorescent signal observed in control vehicle-injected (data not shown) or Ad.CatSS-injected eyes (Fig. 3D , white arrows and Table 1 ). Both the strong and weak signals observed in Ad.CatSAS- and Ad.CatSS-injected eyes, respectively, were localized to areas corresponding to the RPE cell layers marked by the black arrows in Figures 3A and 3B . High-powered microscopic examination of plastic-embedded and paraffin-embedded sections revealed that the photoreceptor layers in Ad.CatSAS-injected eyes at 14 days after injection (Fig. 3E , double-headed arrow) and at 28 days after injection (Fig. 3F) , respectively, were shorter when compared with vehicle-injected (data not shown) or Ad.CatSS-injected eyes sampled at 14 days (Fig. 3G , double-headed arrow) and 28 days after injection (data not shown). The POSs in Ad.CatSAS-injected eyes were not only shorter (Fig. 3E) , but they were also disorganized. These changes were accompanied by a decrease in the number of layers of photoreceptor cells in the outer nuclear layer (Figs. 3E and 3F) to 11, 9, and 8 layers by 7, 14, and 28 days after injection, respectively when compared with the 14 layers present in both vehicle- and Ad.CatSS-injected eyes throughout the time course. The retina distant from the injection area, which was not effected by the Ad.CatSAS injection, remained normal, both in its morphologic and fluoromicroscopic appearance (data not shown). 

Discussion

The work presented in this study confirmed previous observations that subretinally injected recombinant adenovirus almost exclusively transduces RPE cells. This finding suggests that recombinant adenovirus can be used to target RPE cells without the application of an RPE-specific promoter. Although adenovirus-mediated transgene expression is transient, there are a variety of sometimes contradictory reports in the literature on the longevity of transgene expression. ^{3 5}  

To our knowledge, this study, in which fundus fluorescence photography developed for monitoring adeno-associated-virus–mediated gene delivery was used, ^{6 9} is the first to report the noninvasive, real-time monitoring of recombinant adenovirus–mediated gfp expression in vivo. Our study showed that gfp was expressed at high enough levels to be detected by in vivo fundus fluorescence photography from 3 to 9 days after injection. Because our goal was to identify the longevity of high level of transgene expression, the absence of an in vivo gfp signal was considered to be the limit of gfp expression. It was beyond the scope of this study to investigate whether the rapid decrease in signal intensity was caused by an immune response ¹⁰ or by the shutdown of the viral promoter. Although the expression of gfp was transient, the level of expression was high, and it was concluded that constant, high-level expression of biologically active products for up to a week renders this technology potentially suitable for the development of animal models. 

In Ad.CatSAS-injected eyes, the large amount of autofluorescent debris accumulating in the RPE cells was similar in appearance to the granular debris induced by cysteine protease inhibitors. ^{1 2} Although the debris induced by cysteine protease inhibitors does not have the same spectral characteristics as that of age-related-lipofuscin, it can follow the same route of compartmentalization in the RPE cells. ² Thus, animal models based on the forced accumulation of POS-derived debris can be relevant to the study of the processes occurring in the aging or diseased human eye. 

In a previous study, the accumulation of autofluorescent debris in animals injected with cysteine protease inhibitors was reported to be accompanied by some disorganization in the POS layer. ² Compared with these protease inhibitor studies, the neural retinal changes observed in Ad.CatSAS-injected animals were more pronounced. These changes included the shortening of the POSs and a decrease in the number of layers of photoreceptor cells in the outer nuclear layer, which were not observed in vehicle-, Ad.gfp-, or Ad.CatSS-injected animals. Therefore, it can be concluded that these changes were induced by Ad.CatSAS. Protease inhibitor uptake is not a cell-specific process, thus protein inhibition studies were unable to determine whether changes in the photoreceptor layer were due to inhibition of neural retinal cysteine protease activity or to the breakdown of RPE function. Because subretinally delivered recombinant adenoviruses specifically transduce RPE cells, ³ and the effect of the antisense RNA is limited to the transduced cells, it is thus suggested that the neural retinal changes observed are secondary effects of the downregulation of CatS activity in the RPE cell layer. From this study, a correlation between Ad.CatSAS delivery and photoreceptor degeneration was established. We propose that further studies be performed to elucidate the exact role of CatS, either by the use of a recombinant adeno-associated virus that can mediate long-term downregulation of CatS activity or by the generation of knockout mice. 

The transient downregulation of CatS activity in RPE cells in a rat model has been shown to induce the accumulation of autofluorescent debris and to compromise retinal morphology. This implies that CatS may play an important role in the maintenance of normal retinal function. Furthermore, these results demonstrate that recombinant adenovirus-mediated transient gene expression in RPE cells can be used to induce changes in the retina and therefore to provide an understanding of the role of certain genes in the maintenance of normal retinal function. 

 Supported by the National Health and Medical Research Association of Australia.

 Submitted for publication December 29 1999; revised May 6 and July 9, 1999; accepted July 14, 1999.

 Commercial relationships policy: N.

 

Figure 1.

 

Characterization of gfp expression in Ad.gfp-injected eyes. (A) Fundus fluorescence photographic image showing gfp expression in approximately 60% of the original retinal bleb at 3 days after injection. Arrow: injection site. (B) Fluorescence microscopic image showing hexagonal RPE cells expressing gfp in the wholemounted RPE-choroid-sclera of the same eye as shown in (A). Thin arrows: weak gfp signal; thick arrows: strong gfp signal. (C) Light micrograph of an Ad.gfp-injected eye with normal retinal morphology at 7 days after injection. Black arrows: position of RPE layer. Original magnification: (A) ×12; (B) ×50; (C) ×100.

Figure 2.

 

Light micrographs of cryosections of injected eyes enucleated at 4 days after injection. (A) Demonstration of antisense CatS RNA expression in Ad.CatSAS-injected eye hybridized with sense CatS riboprobe. (B) An RNase-treated section of the eye used in (A) hybridized with sense CatS riboprobe. Note decrease in hybridization signal intensity. (C) Section from vehicle-injected eye hybridized to sense CatS riboprobe. Note absence of hybridization signal. (D) Hybridization of a section from an Ad.CatSAS-injected eye with antisense CatS riboprobe. Note absence of hybridization signal. Arrows: (A) Hybridization signal in RPE cell layer; (B, C, and D) RPE cell layer. Magnification, ×100.

Figure 3.

 

Light micrograph (A, B) and fluorescence micrograph (C, D) of recombinant adenovirus–injected eyes at 7 days after injection. (A) Cryosection of an Ad.CatSAS-injected eye showing photoreceptor loss at 7 days after injection. (B) Cryosection of an Ad.CatSS-injected eye at 7 days after injection. (A, B) The double-headed arrows denote the thickness of the photoreceptor layer and arrows mark position of RPE cell layer. (C) Fluorescence microscopy of the same cryosection used in (A) before hematoxylin and eosin staining showing presence of strong, granular fluorescent material (arrows) in the layer corresponding to the RPE cell layer (A, arrows). (D) Fluorescence microscopy of the cryosection shown in (B) before hematoxylin and eosin staining detecting a weak and evenly distributed autofluorescent signal in the layer corresponding to the RPE cell layer (B, arrows). (E) Plastic-embedded section of an Ad.CatSAS-injected eye at 14 days after injection. (F) Paraffin-embedded section of Ad.CatSAS-injected eye at 28 days after injection. (G). Plastic-embedded section of Ad.CatSS-injected eye at 14 days after injection. Thickness of the photoreceptor layer in (E) and (G) is marked by the double-headed arrow. Arrows: RPE cell layer. Magnification, (A through D) ×50; (E, F, and G) ×100. RD, retinal detachment; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer.

Table 1.

 

Intensity of Autofluorescent Signal in Ad.CatSAS-, Ad.CatSS-, and Vehicle-Injected Rat Eyes

Compound	Animal	Cryosection Number
				1	30	60
Ad.CatSAS	1	+	+++	++
	2	++	++	+
	3	++	++	++
	4	++	++	+++
Ad.CatSS	1	−	ND	ND
	2	+	+	+
	3	++	+	+
	4	+	−	−
Vehicle	1	−	+	−
	2	−	−	−

 Signal intensity: − no signal, + weak, ++ medium, +++ strong. ND, not done.

The authors thank Louise Kemp, Katrina Spilsbury, and Meaghan Yu for help in preparing the manuscript, and Meditech Research Ltd. for their support.