Introduction

We have previously developed a series of mutation-independent strategies, which are in principle relevant to any of the 1500 dominant diseases known to exist in man^1,2,3. These involve silencing both mutant and wild-type transcripts from a given gene while simultaneously introducing suppression-resistant replacement genes, which potentially encode wild-type protein. Rhodopsin, peripherin, and collagen 1A1 transcripts have been silenced in this way using hammerhead ribozymes. More recently, we have focused on the use of RNA interference (RNAi) technology for transcript suppression⁴.

RNAi has emerged as a powerful method for gene silencing^5,6,7,8. Briefly, in mammalian systems, small interfering RNAs (siRNAs) are introduced directly into the cell, or processed in the cell from transcribed short hairpin RNAs (shRNA) by Dicer⁹, and then assembled into the RNA-induced silencing complex, where they are unwound, allowing the antisense strand to form a duplex with the target mRNA, which is then degraded by endonuclease digestion. Compared to ribozyme technology, RNAi is at least as potent^2,4, is less dependent on RNA secondary structure, and does not require any particular sequence motif. RNAi holds great potential for gene therapy with respect to suppression of infectious viral RNAs¹⁰, cancer¹¹, and mutant gene products in dominant diseases¹². Furthermore, it has in some cases been possible to suppress mutant transcripts while allowing normal expression of wild-type alleles^13,14. However, since siRNA allele specificity can be difficult to achieve^4,15,16 and because in many cases the cost of developing individual therapeutics for each mutation would be prohibitive, the use of mutation-independent methods outlined here will greatly facilitate development of therapies for dominant heterogeneous disorders such as autosomal dominant retinitis pigmentosa (adRP).

Retinitis pigmentosa, the most commonly inherited retinal degeneration, is a disease of the retina caused by degenerating photoreceptors. It is characterized by night blindness and loss of peripheral vision in the early stages, followed by later loss of central vision¹⁷. Mutations in the rod photoreceptor-specific gene encoding rhodopsin, the pigment responsible for phototransduction¹⁸, cause up to 25% of adRP¹⁹. Over 100 mutations in rhodopsin have been characterized (http://www.retina-international.org/sci-news/rhomut.htm), most of which are single point missense mutations causing protein misfolding. Misfolded proteins are thought to form aggregates, which are central to disease pathology in adRP as well as in many other dominant neurodegenerative disorders^20,21.

To date RNAi technology has been used to suppress VEGF and TGF- type II receptor in murine models of ocular disease^22,23. In addition, photoreceptor degeneration in a constitutively active TRP heterozygote has been rescued following mutation-independent suppression of TRP by siRNA in Drosophila²⁴, and neurodegeneration has been inhibited in a mouse model of dominant spinocerebellar ataxia following intracerebellar injection of AAV-expressing shRNA targeting ataxin-1²⁵. Furthermore, RNAi-mediated downregulation of EGFP and replacement by a codon-modified version in cell culture has been demonstrated by Kim and Rossi²⁶.

The rhodopsin gene exhibits a high degree of intragenic mutational heterogeneity and causes dominant RP, thus serving as a good model in which to develop mutation-independent approaches to gene therapy for dominant disease in general. We demonstrate here suppression of murine rhodopsin transcript by up to 90% with full concomitant expression of replacement transcript.

Results and discussion

Determination of the quantity of shRNA and siRNA required for optimal silencing of murine rhodopsin transcript

An initial study indicated that of five shRNA sequences targeting the murine rhodopsin gene, shMR3 was the most efficient (Supplementary Data Fig. 1 and Table 1). To determine the optimal concentration of either shMR3 or the synthesized siRNA equivalent, siMR3, required for maximal rhodopsin downregulation, we cotransfected increasing quantities of these suppressors with a murine rhodopsin-expressing plasmid into COS-7 cells and measured transcript levels by RT-PCR. Rhodopsin expression was standardized per unit -actin expression and expressed as a percentage of that for the appropriate nontargeting control. The resulting titration curves are shown in Fig. 1. For the particular conditions of the heterologous cell culture system used throughout this study (i.e., 5 10⁵ cells per well of a six-well plate, see Materials and Methods), we found that murine rhodopsin was suppressed to about 20% and that the minimum suppressor doses to achieve this 80% downregulation were 5 g of shMR3 and 0.25 g (19.5 pmol) of siMR3. To verify further that the molecular effects of 5 g of shMR3 were comparable to exogenous application of 0.25 g siMR3, we carried out ribonuclease protection assays. We end-labeled sense and antisense strands of siMR3 separately and used them to protect RNA isolated from cells transfected with shMR3, siMR3, and nontargeting controls. The results given in Fig. 2 indicate that transcription of 5 g of shMR3 (lanes 2 and 10) is indeed approximately equivalent to the addition of 0.25 g (19.25 pmol) of the synthesized siMR3 (lanes 5 and 14).

Figure 1.

Titration curves from RT-PCR data comparing the relative efficiencies of shRNA (shMR3) and siRNA (siMR3) at targeting murine rhodopsin transcript expressed in COS-7 cells.

Full figure and legend (75K)

Figure 2.

Expression of shMR3 and siMR3 in COS-7 cells 24 h posttransfection as shown by RPAs using 5'-end-labeled sense (SE PROBE) and antisense (AS PROBE) strands of siMR3. Lanes 1 and 18, molecular weight marker; 2 and 10, 5 g shMR3; 3 and 11, 5 g shPER2; 4 and 12, 0.05 g (4 pmol) siMR3; 5 and 13, 0.25 g (20 pmol) siMR3; 6 and 14, 0.5 g (40 pmol) siMR3; 7 and 15, 2.5 g (200 pmol) siMR3; 8 and 16, 5 g (400 pmol) siMR3; 9 and 17, 5 g (400 pmol) siPER2. -Actin was used as an internal standard.

Full figure and legend (51K)

Silencing of endogenous murine rhodopsin in cultured retinal explants

The extent of endogenous murine rhodopsin suppression in 14-day-old cultured retinal explants, following delivery of the construct EGFP-shMR3 or EGFP-shNT (nontargeting control) at P0, is shown in Figs. 3 and 4. By culture day 14, the explants had differentiated and morphologically resembled normal retina in vivo, with clearly defined rod outer segments, outer nuclear layer, inner nuclear layer, and ganglion cell layer (Fig. 3A). The transfection efficiency of cells within the explants was approximately 20% as observed by EGFP-fluorescing cells, therefore visualization of the suppression of rhodopsin within the section is masked by the 80% untransfected cells. Immunostaining of trypsin-digested explants with rhodopsin antibodies showed that cells transfected with EGFP-shNT exhibit colocalization of rhodopsin protein and EGFP fluorescence, whereas those transfected with EGFP-shMR3 no longer express rhodopsin protein (Figs. 3E and 3F versus 3B and 3C). Following dissociation of the retinas and FACS, analysis of the EGFP-positive cells in isolation was possible. Fig. 4A shows that, as assessed by RT-PCR, rhodopsin transcript levels were reduced to 30% by EGFP-shMR3 compared to those obtained from explants treated with EGFP-shNT. Neither of the two EGFP-shRNA constructs had significant nonspecific effects on either the retinal gene -Pde or the housekeeping genes Eef2, Atp6, Ppia, and Pk3 (P < 0.0001). This 70% suppression of rhodopsin transcript expression demonstrates that shMR3 is effective at silencing endogenous retinal murine rhodopsin transcribed from the bona fide promoter in retina in addition to suppressing exogenously derived CMV-driven transcripts in heterologous cell systems. Endogenous rhodopsin protein levels in the cultured retinal explants were also suppressed following introduction of shMR3. We labeled 200 randomly chosen EGFP-positive cells from retinas electroporated with either EGFP-shMR3 or EGFP-shNT with rhodopsin antibodies and scored them for expression of both proteins. Fig. 4B shows that 85% of cells from control explants into which EGFP-shNT had been electroporated expressed rhodopsin protein, indicating that 85% of transfected cells within the cultured retinal explants were rod photoreceptors. Eighty-eight percent of cells from EGFP-shMR3-treated explants did not express rhodopsin, indicating that the endogenous rhodopsin protein had been suppressed in these rod photoreceptors. It is unclear why 12% of EGFP-positive cells continued to express rhodopsin while the other 88%, which received the same EGFP-shMR3 construct, did not. However, in real terms, it is likely that a reduction to 12% of mutant rhodopsin protein in the retina would represent a significant clinical amelioration, the extent of which would be determined by the nature of the mutation. In support of this hypothesis, a correlation between the severity of disease and the level of expression of mutant rhodopsin transcript has been observed for P23H²⁷ and P347S rhodopsin mutations²⁸, suggesting that reduction in levels of mutant rhodopsin may lead to less severe retinal degeneration.

Figure 3.

Immunostaining of electroporated retinal explants. Scale bars denote 20 m. (A) Scanning laser micrograph of a section through a 14-day cultured retinal explant electroporated with EGFP-shMR3 at P0 and stained with rhodopsin antibodies. ROS, rod outer segments; ONL, outer nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer. (B–G) Dissociated retinal cells immunostained with rhodopsin antibodies (B, E), showing EGFP fluorescence (C, F), and DAPI stained (D, G), electroporated with nontargeting EGFP-shNT (B–D) or targeting EGFP-shMR3 (E–G).

Full figure and legend (125K)

Figure 4.

Transcript and rhodopsin protein analyses of FACS-sorted dissociated retinal explant cells. (A) RT-PCR analysis of endogenous transcripts of rhodopsin, -Pde, Eef2, Atp6, Ppia, and Pk3 targeted with EGFP-shMR3. (B) Percentage of EGFP-positive cells expressing murine rhodopsin protein following electroporation with EGFP-shNT and EGFP-shMR3.

Full figure and legend (79K)

RNAi resistance of replacement genes in cell culture and mouse liver in vivo

The effects of the short hairpin RNA, shMR3, on the expression of wild-type murine rhodopsin cDNA (MR); replacement constructs MR1 and MR7 containing a single mismatch and seven mismatches, respectively, within the shMR3 target site of murine rhodopsin; human rhodopsin cDNA (HR); and MR plus MR7 combined as assessed by RT-PCR are shown in Fig. 5A (for alignments see Materials and Methods).

Figure 5.

RT-PCR analyses of suppression and replacement of murine rhodopsin transcript. (A) Rhodopsin expression following cotransfections of rhodopsin-expressing constructs MR, MR1, HR, MR7, and MR + MR7 with shMR3 and control shCOL4 in COS-7 cells. (B) Rhodopsin expression in murine liver following systemic delivery of MR or MR7 with control (siPER2) or targeting (siMR3) siRNA.

Full figure and legend (84K)

Replacement gene MR1 was suppressed by shMR3 to 41.8% giving a significant increase in rhodopsin expression compared to the wild-type gene, which had been suppressed to 22.4% (P = 0.002). However, this increase is modest considering that the mismatch was central, situated at position 10 from the 5' end of the target site and that similar mismatches have resulted in a far greater inhibition of RNAi activity^13,14,29. On the other hand it has been reported that at least two matches are often required^14,16. Nevertheless, human rhodopsin cDNA containing a 2-bp mismatch with respect to the murine form expressed rhodopsin at only 33.6% in the presence of shMR3, indicating that the potency of shMR3 for human rhodopsin transcript was even greater than that for MR1 despite the additional mismatch. Indeed, it has been postulated that during RNAi the overall stability of the RNA duplex formed between the antisense siRNA strand and the mRNA target site may determine the success of silencing rather than the absolute number of mismatches³⁰. It is most probable that the two HR mismatches, at positions 4 and 19 from the 5' end of the target sequence, had even less effect on the RNA duplex stability than the single central mismatch in MR1.

Since the parameters that define the design of an siRNA-resistant target are clearly complex, we mutated all 7 of the degenerate bases within the 21-bp rhodopsin target (N(19)TN) to generate construct MR7. Cotransfection with shMR3 gave a rhodopsin level of 89.6%, which differed significantly from that of the murine rhodopsin plus shMR3 (P < 0.00001) but not from that of MR7 cotransfected with control shCOL4 (P = 0.22). When both wild-type and MR7 murine rhodopsin transcripts were simultaneously expressed with shMR3, murine rhodopsin expression was observed to be 76%, indicating that suppression of wild-type mRNA was compensated for by expression of the siRNA-resistant replacement construct, MR7. (The level of 76% is somewhat higher than the approximately 56% ((89.6 + 22.4)/2) predicted from the individual target data. However, it was observed that MR7 expressed rhodopsin more efficiently than MR possibly due to the deletion of UTRs during its construction.) It is likely that in vivo, 76% of normal levels of rhodopsin would provide sufficient protein for functional vision. Indeed it has already been shown that rho^+/- mice with approximately 50% of normal rhodopsin levels appear to have near normal histology and ERG activity³¹.

Following high pressure/volume tail vein injection of siRNAs together with murine rhodopsin encoding cDNAs into recipient mice, we evaluated rhodopsin transcript expression levels in liver by RT-PCR. Ninety percent of transcript was suppressed when we co-injected siMR3 with MR cDNA compared to co-injection of the control siRNA, siPER2, with MR cDNA (Fig. 5B). In contrast, transcript expression levels of the replacement construct MR7 remained unchanged at about 100% regardless of whether the rhodopsin targeting siMR3 or the control siPER2 was co-injected. Thus, both the potency of siMR3 for the target and the resistance to suppression by replacement gene MR7 have been observed in an in vivo situation.

To summarize, we have described effective, high-level production of siRNA via transcription of an shRNA plasmid, a prerequisite first step in designing a gene therapy for which repeated administration would not be required. Although only transient expression of shMR3 is demonstrated herein, longer term expression should in principle be possible following subcloning into a viral vector. The siRNA appears to function well in retinal tissue and in mouse liver in vivo. However, for a wild-type expression profile within the retina, the replacement gene MR7 would have to be driven by a rod-specific promoter and tested following delivery into the retina. It is likely that many such constructs containing a variety of different rod-specific promoters, promoter lengths, 5' and 3' UTRs, heterologous introns, etc., would need to be tested to achieve a typical wild-type profile. Meanwhile, rod-specific siRNA silencing could in theory be achieved by generating a Cre-recombinase-inducible Pol III-driven shRNA as described by Tiscornia et al.³² or by modifying a Pol II rod-specific promoter to transcribe shRNA as has been done by Xia et al.³³ using the CMV promoter.

In conclusion, we report here the first example of mutation-independent suppression and replacement of a mammalian gene that can cause autosomal dominant disease in man. This strategy provides a means of circumventing the mutational heterogeneity found in many disorders while still correcting the primary genetic defect. We suggest that, in principle, this approach could be used for a gene therapy for adRP or, indeed, for any dominant disease in man.

Materials and methods

shRNAs and siRNAs

Table 1 (Supplementary Material) gives N(19) motifs from the siRNA target site sequences AA(N(19))TN from murine rhodopsin cDNA (Accession No. AK044333), murine peripherin cDNA (Accession No. NM_008938), human col1A1 cDNA (Accession No. NM_000088), and a nontargeting siRNA sequence (Qiagen–Xeragon Ltd., Crawley, UK), which were cloned after Brummelkamp¹⁹, to give shMR3, shPER2, shCOL4, and shNT, respectively. shCOL4 and shPER2 targeting human col1A1 and murine peripherin, respectively, shown to downregulate their respective targets by over 70% (personal communication from H. McMahon, Trinity College Dublin, 1/10/04, and unpublished results), were used as control shRNAs. siMR3, siPER2, and siLZ are synthetic siRNAs synthesized by Qiagen–Xeragon, using sequences outlined in Table 1 for shMR3, shPER2, and Escherichia coli -D-galactosidase (Accession No. NC_000913), respectively.

Rhodopsin cDNA constructs

A 1.6-kb fragment containing the murine rhodopsin cDNA sequence (1.1 kb) and 300 and 200 bp of 5' UTR and 3' UTR, respectively, was cloned into pCDNA3.1 (Invitrogen, Paisley, UK) to give construct MR. Replacement constructs MR1 and MR7 were derived from the 1.1-kb murine rhodopsin cDNA cloned into pCDNA3.1 by introducing a 1- or 7-bp substitution, respectively, at degenerative wobble sites over the target recognition site of shMR3 by PCR mutagenesis. The HR construct consists of a 101-bp 5' UTR, 1.1-kb cDNA, and 685-bp 3' UTR cloned into pCDNA3.1. Sequence of the antisense siRNA strand of shMR3 and the AA(N(19)UN target mRNA sequences were as follow, with mismatches in bold: (shMR3) 3'-UUCGGACUCCAGUUGUUGCUUAG-5', (MR) 5'-AAGCCUGAGGUCAACAACGAAUC-3', (MR1) 5'-AAGCCUGAGGUUAACAACGAAUC-3', (MR7) 5'-AAACCCGAAGUGAAUAAUGAGUC-3', and (HR) 5'-AAGCCGGAGGUCAACAACGAGUC-3'.

Cell transfection and RNA isolation

COS-7 cells were cultured using standard procedures. All transfections were carried out in triplicate. Twenty-four hours prior to transfection, 5 10⁵ cells were plated in each well of a six-well plate and grown under standard conditions minus antibiotics. Cells were then rinsed with PBS and cotransfected with 1 g of a rhodopsin-expressing plasmid and 5 g of an shRNA-expressing plasmid or 5 g (approx 385 pmol) of siRNA using Lipofectamine 2000 as outlined by the manufacturer (Invitrogen, Paisley, UK) in a total volume of 1 ml per well. Identical control transfections were carried out with the exception that the appropriate control suppressor, shCOL4 or shPER2 for shRNA, or siLZ or siPER2 for siRNA, replaced those targeting rhodopsin. Twenty-four hours posttransfection total RNA was isolated from cells using Trizol (Invitrogen). RNA was then treated with RNase-free DNase (Promega, Madison, WI, USA) and then phenol, phenol–chloroform extracted, ethanol precipitated, and resuspended in 20 l RNase-free water.

Electroporation and gene silencing in cultured retinal explants

Restriction fragments from shMR3 and shNT containing the H1 promoter, shRNA sequences, and termination signal, were subcloned into pEGFP-1 (Clontech) into which the CMV promoter had been inserted immediately 5' of the EGFP cDNA, to give EGFP-shMR3 and EGFP-shNT, respectively. Electroporation, maintenance, and dissociation of explant cultures were carried out according to the protocols described by Matsuda and Cepko³⁴. Briefly, 12–14 retinas were dissected from newborn C57 mouse pups and electroporated with EGFP-shMR3 and EGFP-shNT, in two independent experiments. Electroporated retinas were cultured in vitro for 14 days and then were dissociated into single cells by digestion with trypsin. Retinal cell suspensions (4–6 retinas combined) were sieved through 50-m Filcon filters (Dakocytomation) and EGFP-positive cells were collected using a Beckman–Coulter Altra Fluorescence Activated Cell Sorter (Beckman–Coulter, Inc., Fullerton, CA, USA) at a rate of 1000 cells/s and 100,000–350,000 EGFP-positive cells per experiment were collected. For RT-PCR analysis, FACS-sorted retinal cells were concentrated through Ultrafree-MC filtration units (0.45-m Amicon/Millipore, Millipore Corp., Billerica, MA, USA) and were then disrupted in situ in the filter units using RNA lysis buffer (RNeasy Mini Kit; Qiagen). RNA was purified as outlined by Qiagen including an on-column DNase digestion step.

Immunostaining of retinal explants

For immunohistochemical analysis, retinal explants were fixed with phosphate-buffered saline containing 4% paraformaldehyde and Vibratome sections, 50–100 m in thickness, were incubated overnight at 4°C with mouse monoclonal antibodies against rhodopsin (rho4D2, 1:100, a kind gift from Dr. R. S. Molday, University of British Columbia, Vancouver, BC, Canada). Immunocytochemistry was performed with rho4D2 after dissociation of retinal explants by trypsin digestion and attachment of single cells on poly-L-lysine-coated coverslips. Primary antibodies were visualized with Alexa Fluor (reg) 568 goat anti-mouse antibodies (Molecular Probes, Inc., Eugene, OR, USA) and counterstained with 4',6-diamidine-2-phenylindole-dihydrochloride (DAPI; Sigma, Dublin, Ireland). Analysis of stained sections and cells was performed with a fluorescence microscope (Axiophot; Zeiss Ltd., UK) or a laser scanning microscope (LSM-510; Zeiss Ltd.). Percentages of EGFP-positive cells coexpressing rhodopsin were calculated following visual inspection of 200 cells per experiment from four fields of view taken at random.

Real-time RT-PCR analysis

RNA was analyzed by real-time RT-PCR using a Quantitect Sybr Green Kit as outlined by the manufacturer (Qiagen–Xeragon) on a LightCycler (Roche Diagnostics, Lewes, UK) under the following conditions: 50°C for 20 min; 95°C for 15 min; 35 cycles of 94°C for 15 s, 57°C for 20 s, 72°C for 10 s. HPLC-purified primers (Sigma–Genosys, Cambridge, UK) for the sequences amplified are given in Table 2 (Supplementary Material). cDNA fragments were amplified from rhodopsin and -actin (all cell culture experiments); rhodopsin, -Pde, Eef2, Atp6, Ppia, and Pk3 (cultured retinal explant experiment only); or rhodopsin and neomycin (mouse liver expression experiment) for each RNA sample a minimum of four times. For cell culture experiments rhodopsin expression was standardized to -actin expression. Standardization to other housekeeping genes such as Gapdh1, Tbp, and Impdh2 gave similar results (data not shown). For the mouse liver expression experiment rhodopsin was standardized against the coexpressed neomycin. For the retinal explant experiment, rhodopsin, -Pde, Eef2, Atp6, Ppia, and Pk3 expression was standardized to 18S rRNA expression. In all cases, the results were expressed as a percentage of those from the similarly standardized appropriate control experiment, obtained following transfection with a nontargeting control suppressor. The reciprocal values give percentage suppression of rhodopsin expression. Mean values, standard errors, and pooled t tests were calculated using Data Desk 6.0 PPC (Data Description, Inc., New York, NY, USA). Differences were deemed statistically significant at P < 0.05.

RNase protection assay (RPA)

RPAs were carried out using 10 g total cellular RNA extracted from 5 10⁵ cells transfected with either 5 g shRNA or 0.05–5 g siRNA and hybridized to approximately 300 fmol of 5'-end-labeled RNA oligo (Ambion RPA III and mirVana probe and marker kits). The RNA oligos were synthesized by Dharmacon, USA, using the equivalent RNA sequences from sh/siMR3 in Table 1 (Supplementary Material) and from -actin (XM_004814, nt 435–467). The RPAs were electrophoresed on a denaturing (8 M urea) 15% polyacrylamide gel, which was subsequently dried down and autoradiographed.

High-pressure/volume tail vein injection (adapted from^35,36,37)

129sv mice of weight 20–30 g were individually restrained inside a 60-ml volume plastic tube. The protruding tail was warmed for 5 min prior to injection under a 60-W lamp. The tail vein was visualized clearly by illumination from below. A mix containing 10 g of rhodopsin cDNA-expressing plasmid and 20 g of siRNA was made up to a volume of 0.1 body weight in ml with PBS and injected into the tail vein at a rate of approximately 1 ml/sec using a 26-gauge (26G 3/8) needle. Mice were sacrificed at 20 h postinjection and total liver RNA extracted as for COS-7 RNA, following homogenization of liver tissue. Rhodopsin transcript was quantified by RT-PCR (as described above) using neomycin as the reference gene. Control animals received siRNA targeting peripherin (siPER2) in addition to the rhodopsin plasmids.

Animal experiments were carried out according to the ARVO statement for the use of animals in ophthalmic and vision research.

References

1. Millington-Ward, S., et al. (1997). Stratagems in vitro for gene therapies directed to dominant mutations. Hum. Mol. Genet. 6: 1415–1426. | Article | PubMed | ChemPort |
2. Millington-Ward, S., et al. (2002). Validation in mesenchymal progenitor cells of a mutation-independent ex vivo approach to gene therapy for osteogenesis imperfecta. Hum. Mol. Genet. 11: 2201–2206. | Article | PubMed |
3. Farrar, G. J., Kenna, P. F. and Humphries, P. (2002). On the genetics of retinitis pigmentosa and on mutation-independent approaches to therapeutic intervention. EMBO J. 21: 857–864. | Article |
4. Millington-Ward, S., et al. (2004). RNAi of COL1A1 in mesenchymal progenitor cells. Eur. J. Hum. Genet. 12: 864–866. | Article | PubMed | ChemPort |
5. Novina, C. D. and Sharp, P. A. (2004). The RNAi revolution. Nature 430: 161–164. | Article | PubMed | ISI | ChemPort |
6. Sontheimer, E. J. and Carthew, R. W. (2004). Argonaute journeys into the heart of RISC. Science 305: 1409–1410. | Article | PubMed | ISI | ChemPort |
7. Caplen, N. J. (2004). Gene therapy progress and prospects. Downregulating gene expression: the impact of RNA interference. Gene Ther. 11: 1241–1248. | Article |
8. Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K. and Tuschl, T. (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in mammalian cell culture. Nature 411: 494–498. | Article | PubMed | ISI | ChemPort |
9. Brummelkamp, T. R., Bernards, R. and Agami, R. (2002). A system for stable expression of small interfering RNAs in mammalian cells. Science 296: 550–553. | Article | PubMed | ISI | ChemPort |
10. Giladi, H., Ketzinel-Gilad, M., Rivkin, L., Felig, Y., Nussbaum, O. and Galun, E. (2003). Small interfering RNA inhibits hepatitis B virus replication in mice. Mol. Ther. 8: 769–776. | Article | PubMed | ISI | ChemPort |
11. Duxbury, M. S., Matros, E., Ito, H., Zinner, M. J., Ashley, S. W. and Whang, E. E. (2004). Systemic siRNA-mediated gene silencing: a new approach to targeted therapy of cancer. Ann. Surg. 240: 667–674.
12. Xia, H., et al. (2004). RNAi suppresses polyglutamine-induced neurodegeneration in a model of spinocerebellar ataxia. Nat. Med. 10: 816–820. | Article | PubMed | ISI | ChemPort |
13. Ding, H., et al. (2003). Selective silencing by RNAi of a dominant allele that causes amyotrophic lateral sclerosis. Aging Cell 2: 209–217. | Article | PubMed | ISI | ChemPort |
14. Miller, V. M., et al. (2003). Allele-specific silencing of dominant disease genes. Proc. Natl. Acad. Sci. USA 100: 7195–7200. | Article | PubMed | ChemPort |
15. Hemmings-Mieszczak, M., Dorn, G., Natt, F. J., Hall, J. and Wishart, L. W. (2003). Independent combinatorial effect of antisense oligonucleotides and RNAi-mediated specific inhibition of the recombinant Rat P2X₃ receptor. Nucleic Acids Res. 31: 2117–2126. | Article | PubMed | ISI | ChemPort |
16. Martinez, L. A., et al. (2002). Synthetic small inhibiting RNAs: efficient tools to inactivate oncogenic mutations and restore p53 pathways. Proc. Natl. Acad. Sci. USA 99: 14849–14854. | Article | PubMed | ChemPort |
17. Kalloniatis, M. and Fletcher, E. L. (2004). Retinitis pigmentosa: understanding the clinical presentation, mechanisms and treatment options. Clin. Exp. Optometry 87: 65–80.
18. Hargrave, P. A. (2001). Rhodopsin structure, function, and topography. Invest. Ophthalmol. Visual Sci. 42: 3–9.
19. Wilson, J. H. and Wensel, T. G. (2003). The nature of dominant mutations of rhodopsin and implications for gene therapy. Mol. Neurobiol. 28: 149–158.
20. Saliba, R. S., Munro, P. M. G., Luthert, P. J. and Cheetham, M. E. (2002). The cellular fate of mutant rhodopsin: quality control, degradation and aggresome formation. J. Cell Sci. 115: 2907–2918. | PubMed | ISI | ChemPort |
21. Aherne, A., et al. (2004). On the molecular pathology of neurodegeneration in IMPDH1-based retinitis pigmentosa. Hum. Mol. Genet. 13: 641–650.
22. Reich, S. J., et al. (2003). Small interfering RNA (siRNA) targeting VEGF effectively inhibits ocular neovascularization in a mouse model. Mol. Vision 9: 210–216.
23. Hiroshi, N., et al. (2004). RNA interference targeting transforming growth factor- type II receptor suppresses ocular inflammation and fibrosis. Mol. Vision 10: 703–711.
24. Geng, C., Pellegrino, A., Bowman, J., Zhu, L. and Pak, W. L. (2004). Complete RNAi rescue of neuronal degeneration in a constitutively active Drosophila TRP channel mutant. Biochim. Biophys. Acta 1674: 91–97.
25. Xia, H., et al. (2004). RNAi suppresses polyglutamine-induced neurodegeneration in a model of spinocerebellar ataxia. Nat. Med. 10: 816–820. | Article | PubMed | ISI | ChemPort |
26. Kim, D.-H. and Rossi, J. J. (2003). Coupling of RNAi-mediated target down-regulation with gene replacement. Antisense Nucleic Acid Drug Dev. 13: 151–155. | Article | PubMed | ISI | ChemPort |
27. Olsson, J. E., et al. (1992). Transgenic mice with a rhodopsin mutation (Pro23His): a mouse model of autosomal dominant retinitis pigmentosa. Neuron 9: 815–830. | Article | PubMed | ISI | ChemPort |
28. Li, T., Snyder, W. K., Olsson, J. E. and Dryja, T. P. (1996). Transgenic mice carrying the dominant rhodopsin mutation P347S: evidence for defective vectorial transport of rhodopsin to the outer segments. Proc. Natl. Acad. Sci. USA 93: 14176–14181. | Article | PubMed | ChemPort |
29. Abdelgany, A., Wood, M. and Beeson, D. (2003). Allele-specific silencing of a pathogenic mutant acetylcholine receptor subunit by RNA interference. Hum. Mol. Genet. 12: 2637–2644. | Article | PubMed | ISI | ChemPort |
30. Saxena, S., Jonsson, Z. O. and Dutta, A. (2003). Small RNAs with imperfect match to endogenous mRNA repress translation: implications for off-target activity of siRNA in mammalian cells. J. Biol. Chem. 278: 44312–44319. | Article | PubMed | ISI | ChemPort |
31. Humphries, M. M., et al. (1997). Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat. Genet. 15: 216–219. | Article | PubMed | ISI | ChemPort |
32. Tiscornia, G., Tergaonkar, V., Galimi, F. and Verma, I. M. (2004). CRE-recombinase inducible RNA interference mediated by lentiviral vectors. Proc. Natl. Acad. Sci. USA 101: 7347–7351. | Article | PubMed | ChemPort |
33. Xia, H., Mao, Q., Paulson, H. L. and Davidson, B. L. (2002). SiRNA mediated gene silencing in vitro and in vivo. Nat. Biotechnol. 20: 1006–1010. | Article | PubMed | ISI | ChemPort |
34. Matsuda, T. and Cepko, C. L. (2004). Electroporation and RNA interference in the rodent retina in vivo and in vitro. Proc. Natl. Acad. Sci. USA 101: 16–22. | Article | PubMed | ChemPort |
35. Liu, F., Song, Y. K. and Liu, D. (1999). Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther. 6: 1258–1266. | Article | PubMed | ISI | ChemPort |
36. McCaffrey, A. P., Meuse, L., Pham, T.-T.T., Conklin, D. S., Hannon, G. J. and Kay, M. A. (2002). RNA interference in adult mice. Nature 418: 38–39. | Article | PubMed | ISI | ChemPort |
37. Lewis, D. L., Hagstrom, J. E., Loomis, A. G., Wolff, J. A. and Herweijer, H. (2002). Efficient delivery of siRNA for inhibition of gene expression in postnatal mice. Nat. Genet. 32: 107–108. | Article | PubMed | ISI | ChemPort |

Appendices

Appendix A

Supplementary Data

Supplementary data for this article can be found, in the online version, at doi:10.1016/j.ymthe.2005.03.028.

Acknowledgements

This work was supported by grants from The Wellcome Trust, 053333/Z/98/Z; The Health Research Board of Ireland, PRO262001; Fighting Blindness Ireland, T01014 and T01032; European Union 5th Framework Programme, HPRN CT200000098; Enterprise Ireland, F01313; Science Foundation Ireland, G20026; European Union Evi-GenoRet, LSHG-CT-2005-512036; and The British RP Society, GR545. The authors thank the following people for technical assistan Anne Cullen for confocal microscopy, Sarah Cannon for tissue culture, and Sylvie Mehigan and Caroline Woods for animal husbandry.