Molecular Vision 2005; 11:648-656 <http://www.molvis.org/molvis/v11/a77/> Received 8 March 2005 | Accepted 10 August 2005 | Published 29 August 2005

Download Reprint

Knockdown of wild-type mouse rhodopsin using an AAV vectored ribozyme as part of an RNA replacement approach

M. S. Gorbatyuk,^1,3 J.-J. Pang,^2,3 J. Thomas Jr.,¹ William W. Hauswirth,^2,3 Alfred S. Lewin^1,3
 
 

Departments of Molecular ¹Genetics and Microbiology and ²Ophthalmology and the ³Center for Vision Science, University of Florida, Gainesville, FL

Correspondence to: M. S. Gorbatyuk, University of Florida, Box 100266, Gainesville, FL, 32610; Phone: (352) 392-0673; FAX: (352) 392-3133; email: mari653@ufl.edu

Abstract

Purpose: To develop a hammerhead ribozymes (Rz) that might be exploited in a "digest and replace" gene therapy strategy for autosomal dominant retinitis pigmentosa (ADRP) caused by mutations in the gene for rhodopsin (RHO).

Methods: A ribozyme (Rz397) was designed to hybridize with an accessible region in rhodopsin mRNA. It was tested in vitro to determine the kinetics of cleavage of a target oligonucleotide. Following transfection of cultured cells, reduction of rhodopsin mRNA in response to Rz397 was measured RT-PCR. The gene for the ribozyme (Rz397) was inserted in an adeno-associated virus (AAV2) vector and packaged in AAV2 capsids. The virus was injected subretinally in the eyes of C57BL/6J (RHO^+/+) and rhodopsin knockout hemizygous (RHO^+/-) mice at postnatal days 6 (P6) and 30 (P30). Mice were analyzed by full-field electroretinography (ERG). The reduction of opsin protein was measured by western blot analysis and visualized by immunocytochemistry. Reduction of rhodopsin mRNA was assessed using in situ hybridization. Morphometric microscopy of fluorescent antibody-antigen complexes and autoradiography of retinas were used to quantify levels of rhodopsin protein and mRNA, respectively.

Results: Transient co-transfection of HEK 293 cells with a wild-type rhodopsin cDNA and Rz397 resulted in an approximately 60% reduction of RHO mRNA one day after transfection. RHO^+/--mice injected with AAV2-Rz397 at P6 showed a 50% reduction in b-wave amplitudes in injected eyes relative to saline injected contralateral eyes. However, injection of RHO^+/--animals at one month and of RHO^+/+-animals at either age had no impact on ERG. Nevertheless, we detected an 80% reduction of opsin protein in ribozyme-injected eyes of hemizygous mice (by western blot) and a 50% reduction in opsin content in RHO^+/+ mice (by morphometry). These reductions were confirmed by in situ hybridization.

Conclusions: AAV2-Rz397 led to significant (greater than or equal to 50%) reduction of rhodopsin mRNA and protein in mice. It affected ERG amplitudes only when injected in hemizygous RHO knockout pups. This RNA inhibitor may prove useful in treating animal models of ADRP as part of an RNA replacement approach.

Introduction

Rhodopsin is a prototype G-coupled receptor containing 7-transmembrane helical domains, but rhodopsin also contains an 11-cis retinal chromatophore covalently linked to lysine at position 296 [1]. Defects in G-coupled receptors are associated with a number of human diseases. Chief among these, in the case of rhodopsin, are retinitis pigmentosa and congenital stationary night blindness [2-4]. Retinitis pigmentosa is a progressive blinding disease that leads to the apoptotic death of rod photoreceptor cells and to the eventual death of cone photoreceptors. Retinitis pigmentosa (RP) may be inherited as an autosomal recessive, an autosomal dominant or as an X-linked trait. Over 100 mutations in the gene for rhodopsin lead to RP, and most of these cause autosomal dominant retinitis pigmentosa (ADRP). Animal models have been extremely useful in studying the physiological effects of RP-related mutations [5-11]. These models include both transgenic or knock-in animals and also mice, rats and dogs that carry spontaneous mutations similar to those that cause human RP.

Viral mediated delivery of wild-type genes has proved successful for several animal models of recessive retinal degeneration, and clinical trials are planned for the recessive rod cell dystrophy Leber Congenital Amaurosis associated with mutations in the RP65 gene [12]. Dominant forms of RP are more refractory to treatment. Dominance is typically associated with toxicity of the mutant protein, so that the mutant gene must either be replaced or silenced. Fortunately, a variety of antisense strategies have been developed for silencing genes at the RNA level [13-17]. Such inhibitors may be employed in either an allele specific or an allele independent manner. Allele-specific inhibitors block the expression only of the defective mRNA and allow expression from the normal allele to support the function and survival of the rod cell. For example, in collaboration with the groups of LaVail and Flannery, we have used AAV2 to deliver ribozymes that cleave a mutant form of rhodopsin mRNA as an effective treatment in a rat model of ADRP [9,14]. Allele independent RNA inhibitors may be more generally useful, however [18-20]. Since many different DNA alterations in rhodopsin gene lead to ADRP, a set of more than 100 mutation-specific inhibitors (ribozymes, siRNA molecules, antisense oligonucleotides) might be required to provide comprehensive therapy for this disease. Allele-independent inhibitors, in contrast, are designed to cleave all forms of rhodopsin mRNA, mutant or wild-type. For therapy, they cannot be deployed independently but rather could be used in combination with a wild-type cDNA containing silent mutations that block base pairing with the antisense inhibitor.

We have been developing RNA-based inhibitors for rhodopsin, and in this report we describe the testing, in vitro and in vivo, of a ribozyme that leads to significant reduction of the expression of this G coupled receptor in rod photoreceptor cells. This work is intended to provide the first step toward an allele-independent approach for treating a variety of opsin mutations leading to ADRP.

Methods

Design and kinetic analysis of ribozymes

In order to develop ribozymes for use in existing animal models of ADRP, we tested several ribozymes that cleaved following GUC and CUC triplets in murine and canine rhodopsin (RHO) mRNA. Cleavage time course and multiple turnover analysis of hammerhead ribozymes (Rz) were performed as described in Fritz et al. [21]. In this paper, we describe the results for one of these (Rz397) that was able to cleave both dog (Genbank accession number X71380) and mouse (Genbank accession number BC013125) rhodopsin mRNA. An inactive Rz 397 was created by mutating a critical G at position 5 of the catalytic core of the ribozyme to C. Analysis of secondary structures of prospective ribozymes and target regions was performed by using Mfold 3.0 as provided online by The Bioinformatics Center at Rensselaer and Wadsworth. This analysis was performed on overlapping tiles of 50 nucleotides of rhodopsin mRNA and was used only to identify stable stem-loop structures that were consistently predicted. We also used Mfold to predict alternative folding of ribozymes that may interfere with recognition (hybridization) of the target sequence.

Vectors and cloning

DNA encoding the ribozyme was cloned in AAV2 vectors under the control CBA promoter for tissue culture experiments and under control of the mouse proximal opsin promoter for testing in animals. An internally cutting hairpin ribozyme was included within the primary transcript to release the hammerhead ribozyme from downstream sequences [14,22,23]. For testing ribozymes in tissue culture, the target plasmid consisted of the wild-type dog rhodopsin cDNA driven by an CMV promoter cloned in pcDNA3.1.Zeo⁺ vector (Invitrogen Corporation, Carlsbad, CA). For in vivo experiments recombinant AAV2 was produced by co-transfection with helper plasmids and purified by published techniques [24].

Tissue culture testing of ribozyme

Experiments were carried out on HEK 293 cells. Co-transfection of cells with plasmids containing target and ribozymes were performed according to the manufacturer's protocol for Lipofectamine 2000 (Invitrogen Corporation, Carlsbad, CA).

A molar ratio of 1:6 or 1:8 (target plasmid to Rz plasmid) was employed. In transfection experiments with Rz 397, 10 μg of total DNA was used for a 10 cm dish of HEK 293 cells at 70% confluence, and all cells were transfected with the same total amount of DNA, using vector DNA with an irrelevant ribozyme as carrier. Isolation of RNA from transfected cells was performed using a Sigma RNA isolation kit (Sigma-Aldrich, St. Louis, MO). Reduction of the target was determined by semi quantitative RT-PCR using reverse transcriptase (One Strand DNA kit, GE Healthcare, Piscataway, NJ) and an opsin-specific antisense primer: CGA TCC CAC ATG AGC ACT GC. Amplification of β-actin mRNA was used as an internal control for RNA recovery. The reverse transcription reaction for β-actin also employed a gene specific antisense primer: GTT TGA GAC CTT CAA CAC CCC. The PCR reaction to determine RHO mRNA levels used the sense primers for β-actin (ACT CCT GCT TGC TGA TCC AC) and (TGG CTT CAC CAC CAC CCT CT) for RHO. The amplification reaction contained 2 μl of the RT reaction, 200 μM of each dCTP, dGTP, and dTTP, 100 μM dATP, 2 mM MgCl₂ and 20 pM gene specific primers; α-[³²P] ATP was also added to master mixture (0.25 μCi/25 μl). Five μl of PCR product were run on a 1.5% agarose gel. The dried gel was analyzed using a Storm PhosphorImager (GE Healthcare, Piscataway, NJ) and ImageQuant software to determine the linear range of amplification. A preliminary control experiment was performed to determine the linear range of amplification of RHO and β-actin mRNAs. Products of reverse transcription of both RNAs were subjected to 16, 18, 20, 22, 24, and 26 cycles of amplification and band intensity was plotted as a function of cycle number. It was established that 20 cycles of amplification were sufficient to determine the quantity RHO mRNA and were within the linear range of amplification for both transcripts.

Animals and injection of AAV

All animal procedures were performed in accordance with NIH Guide for Care and Use of Laboratory Animals and the ARVO statement for the Use Animals in Ophthalmic and Vision Research. RHO^+/+ (C57BL/6J) and RHO^-/- (rhodopsin knockout) animals were purchased from Jackson Laboratories (Bar Harbor, MA). Mice were maintained in 12 h:12 h light/dark regime. Mice treated with AAV2 Rz397 were injected subretinally at postnatal day 6, postnatal day 16 or postnatal day 30 (P6 or P30). Adult animals were anesthetized by subcutaneous injection of ketamine/xylazine. Neonatal pups were anesthetized by hypothermia. The concentration of AAV2 Rz 397 (active and inactive) used for injections was 4x10¹² particles per ml. Adult animals received injection volumes of 1 μl and neonatal animals received 0.5 μl. Active ribozyme was injected only in the right eyes, and left eyes were either injected with saline or with a control virus as indicated in the figure legends. All injections were administrated in accordance the procedure described by Timmers et al. [25]. Evaluation of the extent of transduction was done by staining of RHO^+/- retinas injected at P₆ with an AAV vector expressing GFP.

Tissue preparation

Eyes were enucleated and fixed overnight in 4% of freshly made paraformaldehyde in phosphate buffered saline (PBS). Afterwards, eye cups were transferred to PBS to remove formaldehyde and submerged sequentially in solutions of 10%, 20%, and 30% sucrose for at least one h each.

Electroretinographic analysis (ERG)

At one and two months following injection, dark adapted animals (15 h) were analyzed by simultaneous full-field ERG. Following ketamine/xylazine anesthesia, electroretinographic responses were elicited with 10 μsec flashes of white light (100 mcds/m²). Responses from both eyes were recoded using a UTAS-E 2000 Visual Electrodiagnostic System (LKC, Gaithersberg, MD). For quantitative comparison of a- and b-wave amplitudes between right and left eyes, the results from individual mice were averaged and the means were compared statistically using Student's t-test for paired samples. Changes in ERG amplitudes were evaluated as absolute values of a- and b-wave amplitudes.

Retinal protein extraction and immunoblots

Two months after injection, mice were euthanized by carbon dioxide inhalation, and then eyes were removed. Retinas were dissected, and retinal extracts were prepared by sonication in Laemmli sample loading buffer (without bromphenol blue and without heating). Quantification of total retinal proteins was done by the Lowry method using a BioRad RC protein detection kit. For western blots, 10 μg of total protein was loaded on 12% SDS polyacrylamide gels run in Tris-glycine buffer. Proteins were electrophoretically transferred to nitrocellulose membranes. The blots were treated with a monoclonal antibody against bovine rhodopsin (B630) that was a gift of Dr. P. Hargrave. As a secondary antibody, we used alkaline phosphatase conjugated anti-mouse IgG. For detection of alkaline phosphatase we used the color reagent 5-bromo-4-chloro-3-indolyl phosphate (Zymed Laboratories Inc., South San Francisco, CA). Stained blots were scanned and analyzed for the intensity of the colored complex using BioRad Gel Analysis Software. Quantification of rhodopsin protein made use of β-actin as an internal control. Preliminary blots were performed to determine the range of linearity for detection of RHO and β-actin in retinal extracts.

Immunocytochemistry

For detection of rhodopsin in situ, eyes were fixed in 4% paraformaldehyde, washed in PBS and soaked in 30% sucrose. 10 μm frozen sections were then prepared and incubated with a 1:300 dilution of the primary antibody (B630). Antibody-antigen complexes were detected with Cy3-conjugated anti-mouse IgG antibodies (1:100). For counterstaining of cone photoreceptors, fluorescein labeled PNA (peanut agglutinin) was used. For detection of GFP in control injected retinas we used anti-GFP polyclonal antibody raised in rabbit (a gift of Dr. Paul Hargrave.) For micrographs, fluorescein labeled goat anti-rabbit IgG was used to develop the immune complexes. All images were made on a confocal microscope (BioRad MRC 1024). The visualization RHO was performed under 20x magnification. Differences between treated and control eyes was measured using a morphometric microscope (Zeiss Axiophot 2 with Sony DXC 970 camera) using the MCID software.

In situ hybridization

Hybridization with a [³³P]-labeled RNA 700 bp probe to murine rhodopsin was performed using the mRNA Locator Kit and the In Situ Hybridization Kit from Ambion, (Ambion, Inc., Austin, TX) on 10 μm frozen sections from eyes fixed in 4% paraformaldehyde. The sections were thaw mounted onto ProbeOn slides (Fisher Scientific, Pittsburgh, PA) and then stored at -20 °C until processing. For quantitative analysis by in situ hybridization, animals were selected randomly from among the treated mice. RHO mRNA quantification was performed both on X-ray film and on emulsion-dipped slides. Examination of slides was done under bright and dark-field illumination using the Zeiss Axiophot 2 microscope. Exposure levels in autoradiograms were measured on a PC equipped with a Quick Capture frame grabber card (Data Translation, Marlboro, MA), a Northern Light precision illuminator (Imaging Research, St. Catherine, Ontario, Canada) and a Dage MTI CCD72 series camera (Dage MTI, Michigan City, IN) equipped with a Nikon 55 mm lens. To adjust for possible defects in the illumination or optical pathway, an image of the empty illumination screen was taken through a neutral filter and used for background shading correction. The gray levels of eight ¹⁴C plastic standards (GE Healthcare) provided a calibration curve. ¹⁴C-plastic standards were exposed to β-max film (BioMax MR Scientific Imaging Film, Kodak) along with the slide-mounted sections [26].

Results

Design and testing of Rz397

Ribozyme Rz397 was designed to cleave at homologous sites following a GUC triplet in the mouse (at position 333) and dog rhodopsin mRNA (at position 397, Figure 1A). We chose a target present in the mRNA from both species, so that the ribozyme could be tested first in mice before moving to a larger animal model of RP. There are numerous (>25) identical ribozyme cleavage sites (GUC, CUC, UCC) in dog and mouse rhodopsin mRNA, but Rz397 targets the sequence GUCUU, which is an especially susceptible site. Of the four ribozymes we tested, this was the most active on the rhodopsin mRNA from both species. A reaction time course was conducted in 5 mM MgCl₂ and with tenfold excess target RNA oligonucleotide compared to ribozyme. Cleavage reached a plateau by 10 min, and 15% of the target was cleaved by 2 min (Figure 1B). We chose this interval for multiple turnover analysis, measuring the fraction of RNA cleaved as a function of substrate concentration (Figure 1C). Kinetic parameters (K_M and k_cat) were determined using the best fit straight line for a Linneweaver-Burke double reciprocal plot (data not shown). K_M was determined to be 910 nM and k_cat was 1.29 min^-1, making this ribozyme compare favorably with others we have tested in animals [27-30].

Ribozyme reduces RHO mRNA in cultured cells

To test Rz397 on full length mRNA in vitro, co-transfection experiments were performed using a plasmid expressing dog RHO mRNA from the CMV immediate early promoter and a second plasmid expressing the ribozyme. Following co-transfection of HEK 293 cells with ribozyme and rhodopsin plasmids at 1:6 and 1:8 M ratios, we measured more than 60% reduction in the level of RHO mRNA normalized to endogenous β-actin (p<0.0004 and p<0.005, respectively; Figure 2). The fact that there was no further decrease in RHO mRNA with increasing ribozyme, suggests that we had saturated the available cleavage sites at sixfold excess ribozyme.

Ribozyme attenuation of the light response in mice

Once it was clear that Rz397 led to cleavage at accessible sites in RHO mRNA, ribozyme DNA oligonucleotides were cloned in an AAV2 plasmid vector capable of driving expression in the eye. The ribozyme was cloned in pMOPS HP, which contains the mouse opsin proximal promoter driving a cassette that contains an SV40 intron and polyadenylation signal and an internally cutting hairpin ribozyme to release the active hammerhead from the primary transcript [14,22,23]. The vector contains AAV2 terminal repeats, and the expression cassette was packaged in AAV2 capsids. To test the ribozyme in vivo, 0.5 μl of rAAV (4x10⁹ genome particles) was injected in the right eyes of mice (N=6) that were hemizygous for a disrupted rhodopsin allele (RHO^+/-) [31] or into the right eyes C57BL/6 mice (RHO^+/+). The left eyes were injected either with an inactive ribozyme or with saline solution. Control injection of AAV vector containing GFP was used to evaluate transduction area in injected eyes. Typically, 60-70% of the retina was transduced (Figure 3).

Using full-field electroretinography (ERG), we measured both the maximum a- and b-wave amplitudes at one and two months after injection and noticed a significant drop in both amplitudes by 2 months (Figure 4). In treated eyes of RHO^+/- mice injected at P6, we recorded a reduction of 60% in b-wave amplitude (p<0.037) compared to the inactive ribozyme, and a reduction of 53% (p<0.0004) relative to uninfected eye. Control animals injected in the right eyes with saline solution and not injected in the left eyes show a slight difference in a- and b-wave amplitudes between right and lefts eyes but this difference was not significant. Reduction in ERG response was also modest (17%) in RHO^+/+ mice injected at P30 and was not significant (p<0.06). Results for a-wave amplitudes were quite similar, though the variability was greater. To be sure that the decrease of b-wave amplitude in hemizygous mice was not restricted to neonates, RHO^+/- animals were also injected at P16 with active and inactive ribozyme and analyzed by ERG 2 months later. In this group of animals (n=7) we also observed a 60% reduction in maximum b-wave amplitude (p<0.01).

Reduction of rhodopsin protein

Protein was extracted from the retinas of three of animals treated with AAV-ribozyme and separated on SDS polyacrylamide gels for analysis by western blot. Rz397 caused an 80% (p<0.019) reduction in immune reactive opsin protein in RHO^+/- mice (Figure 5A). RHO^+/+ animals also exhibited reduction in rhodopsin protein (50%) though that difference was not significant (p<0.12; Figure 5B). While the reduction in protein level was roughly analogous to the decrease ERG response in RHO^+/- animals, the reduction of opsin protein was much greater than the loss of b-wave signal in RHO^+/+ mice, suggesting that the residual 50% of rhodopsin in ribozyme treated animals sustained >80% of the light response.

Detection of opsin reduction by immunocytochemistry and in situ hybridization

Because there was an apparent inconsistency in RHO^+/+ mice between the attenuation in b-wave amplitude, and the reduction in opsin protein, we sought to confirm the protein reductions by an independent technique. To visualize the levels of opsin protein in ribozyme treated eyes, lightly fixed frozen sections of treated and control eyes were decorated with a monoclonal antibody (B630) to bovine rhodopsin (a gift of Dr. Paul Hargrave), and immune complexes were detected using Cy3-labled anti-mouse IgG antibody. Cones were counterstained using PNA, and sections were visualized by fluorescence microscopy (Figure 6A). Morphometric analysis indicated a 50% reduction in opsin staining in RHO^+/+ mice (right panel). Nevertheless, phase contrast images of these same sections indicated no significant thinning of the outer nuclear layer (ONL) or shortening of the rod outer segments. Consequently, it appears that mice can sustain a loss of 50% of their rhodopsin production without inducing rod cell apoptosis or shortening of outer segments. This result is consistent with the survival of rods in RHO^+/- hemizygotes [31-33].

To confirm the effect of Rz397 in mice at the RNA level, we used in situ hybridization. Frozen sections were hybridized with a 700 nucleotide [³³P]-labeled probe derived from the murine rhodopsin cDNA. Sections were mounted on slides and dipped in photographic emulsion for visualization and exposed to X-ray film for scanning and quantitation (Figure 6B). This experiment revealed a 60±6% reduction in rhodopsin mRNA in RHO^+/+ animals relative to controls injected with saline (p<0.05).

AAV ribozyme treatment causes thinning of the ONL in RHO^+/- mice

While loss of approximately half of the rhodopsin in wild-type (RHO^+/+) mice did not have significant effect on the survival of rod photoreceptor cells, treating RHO^+/- hemizygotes with AAV Rz397 led to a over 30% reduction in the ONL thickness compared to eyes of the same animals treated with the inactive Rz397 (p<0.028; Figure 7). This result is associated with a 60% reduction in ERG amplitudes observed in these eyes, which is probably a direct result of the death of rod photoreceptor cells.

Discussion

Since site specific ribozymes have been used for the treatment of a rat model of ADRP, one might question the necessity of pursuing a more complicated mutation-independent approach. The rationale for the digest and replace strategy is twofold: First, this method should overcome the considerable problem posed by allelic heterogeneity among dominant mutations affecting rhodopsin. If successful, this approach should permit one or two gene therapy instruments to treat all ADRP cases mapping to RHO. Not all RHO mutations are associated with ribozyme cleavage sites, and single base changes may not be sufficient for the design of allele-specific siRNAs. Therefore, the allele-independent approach may be the only way to use this technology for certain ADRP mutations. The second issue is pragmatic. A gene therapy tool that could be used to treat all or most ADRP arising from RHO mutations is more likely to achieve financial support and government approval than is a set of vector-ribozyme combinations, each tailored to a small group of patients.

Mice hemizygous for a rhodopsin disruption have a very mild visual phenotype. Consequently, for an allele independent knockdown approach such as ours, it is important to determine how much further rhodopsin levels can be diminished without damaging photoreceptor function or survival. Below this level, ribozymes may promote the death of infected cells. The residual component of the toxic mutant rhodopsin may contribute to the susceptibility to apoptosis. AAV delivery of ribozyme Rz397 led to an 80% reduction of rhodopsin protein in mice hemizygous for a disruption of the rhodopsin gene (RHO^+/-, Figure 5). The residual rhodopsin is responsible for maintaining 50% of the normal b-wave response (Figure 4). Rhodopsin is a major structural protein in the rod outer segment, and hemizygous animals start out with shorter and thinner outer segments, in addition to a 43% reduction in rhodopsin [34]. Consequently, our AAV delivered ribozyme reduces rhodopsin content in wild-type animals to approximately that present in the RHO^+/- hemizygotes, while ribozyme treatment of the hemizygotes reduces expression even further. It remains to be determined whether this extent of inhibition will be sufficient to block the dominant negative effects of ADRP mutations affecting rhodopsin.

A second crucial variable in an allele-independent knockdown strategy is the extent of the retina transduced by AAV in subretinal injections. Using GFP as a marker protein, we estimate that our efficiency of rod cell transduction is 60-70% (Figure 3). This implies that reduction of gene expression in infected cells is nearly 100% in RHO^+/- mice and over 60% in RHO^+/+ animals. Because 30-40% of the retina was not transduced with AAV, our electroretinogram measurements, which record the response from the entire retina must necessarily provide an overestimate of the impairment in light response in infected photoreceptor cells. Since transduction was less than complete, it might be argued that our approach will be unable to rescue peripheral vision in RP patients. On the other hand, if RNA replacement can be used to protect rods in the perimacular region of the retina, this rescue might sustain central vision through trophic support of the cones. Finally, since we cannot be sure to infect all of the rod photoreceptors in a single injection, an RNA replacement strategy such as that we envision should deliver the ribozyme and the replacement gene in the same recombinant AAV. Otherwise, some cells may receive the knockdown agent (ribozyme) without the rescue gene (ribozyme-resistant rhodopsin), while other cells may overexpress rhodopsin without limiting the production of the endogenous protein.

Another barrier that must be overcome in the RNA replacement approach is creating a ribozyme resistant rhodopsin cDNA to incorporate in AAV vectors bearing the ribozyme. Generating a ribozyme resistant transgene is trivial and is accomplished through one or two silent mutations in the rhodopsin cDNA. For Rz397, for example, the target sequence CAUGGUCUUCGG is changed to UAUGGUGUUUGG, and cleavage experiments in vitro indicate that these changes render the RNA insensitive to cleavage by the ribozyme (data not shown). Furthermore, for ribozymes (as for siRNA) the issue of off-target cleavage must be considered. Ribozymes, have a short target sequence (10-11 nucleotides are required for cleavage) that can be expected to occur in other mRNAs. We routinely screen EST databases for the occurrence of such targets, and we try to assure safety by the local and limited volume of injection and by restricting expression to rod photoreceptor cells using the opsin promoter. Before moving from mouse to man, it will be important to examine possible off-target silencing effect of ribozymes, possibly using DNA microarray analysis.

Before applying a cut and replace strategy for rhodopsin mRNA, we must be able to regulate the expression of the replacement cDNA. Overexpression of wild-type rhodopsin can be toxic to photoreceptors [11,35]. Consequently, the replacement gene must be controlled either by a cell type specific promoter or by a regulated promoter to prevent this. Despite these major obstacles in the path, we believe that RNA inhibitors such as ribozymes will play a role in overcoming the allelic heterogeneity associated with ADRP. Our earlier results with allele-specific ribozymes indicate that such inhibitors can rescue rats from retinal degeneration even without complete inhibition of the expression of the mutant P23H transgene. Therefore, the substantial knockdown of rhodopsin mRNA and protein we have observed in mice using an allele-independent RNA inhibitor suggests that we have taken the first step toward an RNA replacement therapy for ADRP.

Acknowledgements

The authors thank Dr. Aguirre for canine rhodopsin cDNA, Leah Lewin for technical assistance and Verline Justilien for help with injections and the NEI (R01 EY11596), the Steinbach Fund and the Foundation Fighting Blindness for research support.

References

1. Menon ST, Han M, Sakmar TP. Rhodopsin: structural basis of molecular physiology. Physiol Rev 2001; 81:1659-88.

2. Cideciyan AV, Hood DC, Huang Y, Banin E, Li ZY, Stone EM, Milam AH, Jacobson SG. Disease sequence from mutant rhodopsin allele to rod and cone photoreceptor degeneration in man. Proc Natl Acad Sci U S A 1998; 95:7103-8.

3. Garriga P, Manyosa J. The eye photoreceptor protein rhodopsin. Structural implications for retinal disease. FEBS Lett 2002; 528:17-22.

4. Sung CH, Davenport CM, Nathans J. Rhodopsin mutations responsible for autosomal dominant retinitis pigmentosa. Clustering of functional classes along the polypeptide chain. J Biol Chem 1993; 268:26645-9.

5. Aguirre G. Genes and diseases in man and models. Prog Brain Res 2001; 131:663-78.

6. Kijas JW, Cideciyan AV, Aleman TS, Pianta MJ, Pearce-Kelling SE, Miller BJ, Jacobson SG, Aguirre GD, Acland GM. Naturally occurring rhodopsin mutation in the dog causes retinal dysfunction and degeneration mimicking human dominant retinitis pigmentosa. Proc Natl Acad Sci U S A 2002; 99:6328-33.

7. Kijas JW, Miller BJ, Pearce-Kelling SE, Aguirre GD, Acland GM. Canine models of ocular disease: outcross breedings define a dominant disorder present in the English mastiff and bull mastiff dog breeds. J Hered 2003 Jan-Feb; 94:27-30.

8. Kukekova AV, Aguirre GD, Acland GM. Cloning and characterization of canine SHARP1 and its evaluation as a positional candidate for canine early retinal degeneration (erd). Gene 2003; 312:335-43.

9. LaVail MM, Yasumura D, Matthes MT, Drenser KA, Flannery JG, Lewin AS, Hauswirth WW. Ribozyme rescue of photoreceptor cells in P23H transgenic rats: long-term survival and late-stage therapy. Proc Natl Acad Sci U S A 2000; 97:11488-93.

10. Machida S, Kondo M, Jamison JA, Khan NW, Kononen LT, Sugawara T, Bush RA, Sieving PA. P23H rhodopsin transgenic rat: correlation of retinal function with histopathology. Invest Ophthalmol Vis Sci 2000; 41:3200-9.

11. Olsson JE, Gordon JW, Pawlyk BS, Roof D, Hayes A, Molday RS, Mukai S, Cowley GS, Berson EL, Dryja TP. Transgenic mice with a rhodopsin mutation (Pro23His): a mouse model of autosomal dominant retinitis pigmentosa. Neuron 1992; 9:815-30.

12. Acland GM, Aguirre GD, Ray J, Zhang Q, Aleman TS, Cideciyan AV, Pearce-Kelling SE, Anand V, Zeng Y, Maguire AM, Jacobson SG, Hauswirth WW, Bennett J. Gene therapy restores vision in a canine model of childhood blindness. Nat Genet 2001; 28:92-5.

13. Kim B, Tang Q, Biswas PS, Xu J, Schiffelers RM, Xie FY, Ansari AM, Scaria PV, Woodle MC, Lu P, Rouse BT. Inhibition of ocular angiogenesis by siRNA targeting vascular endothelial growth factor pathway genes: therapeutic strategy for herpetic stromal keratitis. Am J Pathol 2004; 165:2177-85.

14. Lewin AS, Drenser KA, Hauswirth WW, Nishikawa S, Yasumura D, Flannery JG, LaVail MM. Ribozyme rescue of photoreceptor cells in a transgenic rat model of autosomal dominant retinitis pigmentosa. Nat Med 1998; 4:967-71. Erratum in: Nat Med 1998; 4:1081.

15. Nakamura H, Siddiqui SS, Shen X, Malik AB, Pulido JS, Kumar NM, Yue BY. RNA interference targeting transforming growth factor-beta type II receptor suppresses ocular inflammation and fibrosis. Mol Vis 2004; 10:703-11 <http://www.molvis.org/molvis/v10/a85/>.

16. Tolentino MJ, Brucker AJ, Fosnot J, Ying GS, Wu IH, Malik G, Wan S, Reich SJ. Intravitreal injection of vascular endothelial growth factor small interfering RNA inhibits growth and leakage in a nonhuman primate, laser-induced model of choroidal neovascularization. Retina 2004; 24:660.

17. Tolentino MJ, Brucker AJ, Fosnot J, Ying GS, Wu IH, Malik G, Wan S, Reich SJ. Intravitreal injection of vascular endothelial growth factor small interfering RNA inhibits growth and leakage in a nonhuman primate, laser-induced model of choroidal neovascularization. Retina 2004; 24:132-8.

18. Millington-Ward S, O'Neill B, Tuohy G, Al-Jandal N, Kiang AS, Kenna PF, Palfi A, Hayden P, Mansergh F, Kennan A, Humphries P, Farrar GJ. Strategems in vitro for gene therapies directed to dominant mutations. Hum Mol Genet 1997; 6:1415-26.

19. O'Neill B, Millington-Ward S, O'Reilly M, Tuohy G, Kiang AS, Kenna PF, Humphries P, Farrar GJ. Ribozyme-based therapeutic approaches for autosomal dominant retinitis pigmentosa. Invest Ophthalmol Vis Sci 2000; 41:2863-9.

20. Sullivan JM, Pietras KM, Shin BJ, Misasi JN. Hammerhead ribozymes designed to cleave all human rod opsin mRNAs which cause autosomal dominant retinitis pigmentosa. Mol Vis 2002; 8:102-13 <http://www.molvis.org/molvis/v8/a14/>.

21. Fritz JJ, Gorbatyuk M, Lewin AS, Hauswirth WW. Design and validation of therapeutic hammerhead ribozymes for autosomal dominant diseases. Methods Mol Biol 2004; 252:221-36.

22. Altschuler M, Tritz R, Hampel A. A method for generating transcripts with defined 5' and 3' termini by autolytic processing. Gene 1992; 122:85-90.

23. Shaw LC, Whalen PO, Drenser KA, Yan W, Hauswirth WW, Lewin AS. Ribozymes in treatment of inherited retinal disease. Methods Enzymol 2000; 316:761-76.

24. Zolotukhin S, Potter M, Zolotukhin I, Sakai Y, Loiler S, Fraites TJ Jr, Chiodo VA, Phillipsberg T, Muzyczka N, Hauswirth WW, Flotte TR, Byrne BJ, Snyder RO. Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors. Methods 2002; 28:158-67.

25. Timmers AM, Zhang H, Squitieri A, Gonzalez-Pola C. Subretinal injections in rodent eyes: effects on electrophysiology and histology of rat retina. Mol Vis 2001; 7:131-7 <http://www.molvis.org/molvis/v7/a19/>.

26. Hirata H, Cadet JL. p53-knockout mice are protected against the long-term effects of methamphetamine on dopaminergic terminals and cell bodies. J Neurochem 1997; 69:780-90.

27. Afzal A, Shaw LC, Caballero S, Spoerri PE, Lewin AS, Zeng D, Belardinelli L, Grant MB. Reduction in preretinal neovascularization by ribozymes that cleave the A2B adenosine receptor mRNA. Circ Res 2003; 93:500-6.

28. Drenser KA, Timmers AM, Hauswirth WW, Lewin AS. Ribozyme-targeted destruction of RNA associated with autosomal-dominant retinitis pigmentosa. Invest Ophthalmol Vis Sci 1998; 39:681-9.

29. Qi X, Lewin AS, Hauswirth WW, Guy J. Optic neuropathy induced by reductions in mitochondrial superoxide dismutase. Invest Ophthalmol Vis Sci 2003; 44:1088-96.

30. Shaw LC, Skold A, Wong F, Petters R, Hauswirth WW, Lewin AS. An allele-specific hammerhead ribozyme gene therapy for a porcine model of autosomal dominant retinitis pigmentosa. Mol Vis 2001; 7:6-13 <http://www.molvis.org/molvis/v7/a2/>.

31. Lem J, Krasnoperova NV, Calvert PD, Kosaras B, Cameron DA, Nicolo M, Makino CL, Sidman RL. Morphological, physiological, and biochemical changes in rhodopsin knockout mice. Proc Natl Acad Sci U S A 1999; 96:736-41.

32. Humphries MM, Rancourt D, Farrar GJ, Kenna P, Hazel M, Bush RA, Sieving PA, Sheils DM, McNally N, Creighton P, Erven A, Boros A, Gulya K, Capecchi MR, Humphries P. Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat Genet 1997; 15:216-9.

33. Liang Y, Fotiadis D, Filipek S, Saperstein DA, Palczewski K, Engel A. Organization of the G protein-coupled receptors rhodopsin and opsin in native membranes. J Biol Chem 2003; 278:21655-62.

34. Liang Y, Fotiadis D, Maeda T, Maeda A, Modzelewska A, Filipek S, Saperstein DA, Engel A, Palczewski K. Rhodopsin signaling and organization in heterozygote rhodopsin knockout mice. J Biol Chem 2004; 279:48189-96.

35. Tan E, Wang Q, Quiambao AB, Xu X, Qtaishat NM, Peachey NS, Lem J, Fliesler SJ, Pepperberg DR, Naash MI, Al-Ubaidi MR. The relationship between opsin overexpression and photoreceptor degeneration. Invest Ophthalmol Vis Sci 2001; 42:589-600.