WO2005079815A2 - Methods and reagents for treating autosomal dominant diseases of the eye - Google Patents

Abstract

The invention relates to treatment for an autosomal dominant disease of the eye such as retinitis pigmentosa or glaucoma comprising means to substantially suppress all alleles of a gene associated with the disease. Replacement of the gene is specifically disclaimed. Typically, the disease is a dominant hereditary disease in which the therapy involves suppressing both a diseased and a normal allele of the gene. The means for suppressing both alleles of the gene of interest typically comprises an siRNA molecule which is designed to silence both alleles of the gene of interest. Medicaments and kits for treating diseases are also described.

Description

METHODS AND REAGENTS FOR TREATING DISEASE

Field of the Invention

The invention relates to therapies and medicaments for the treatment of disease, particularly dominant hereditary" diseases, more particularly dominant hereditary diseases of the eye such as glaucoma and RPlO-type Retinitis pigmentosa (RP) .

Background Art

RP is an hereditary disease involving progressive death of the photoreceptor cells of the retina and is probably the most prevalent cause of registered visual handicap in those of working age in developed countries .

While RP is currently incurable, extensive progress has been made in understanding the genetic basis of the disease, to date mutations in approximately 30 genes having been implicated in disease pathology, although in any given patient or family usually only a single one of these genes is defective (for a summary of existing prior art in respect to disease etiology and development of therapeutics, see 1) .

Based upon knowledge of the cause of the disease, attempts are now being made to develop novel therapeutics (gene therapies) for RP. The therapeutic approach depends upon the genetic sub- type of the disease, in that RP may be inherited either in a dominant or in a recessive sense (X- linked forms of RP are also recessive) .

Glaucoma is a complex disease, which may involve degeneration of the trabecular meshwork and/or lamina cribrosa of the eye, resulting in aberrant function of drainage channels and/or degeneration of the optic nerve head. As a result, ganglion cells (the output neurons of the retina) die, resulting in narrowing of and/or loss of the visual fields, leading, if untreated, to severe visual handicap in a significant proportion of cases. The majority of cases of open-angle glaucoma involve increased intraocular pressure although a growing number of so-called normal pressure glaucomas are now being identified. In those cases where pressure build up is registered, pressure-reducing eye drops are often of substantial value in slowing down the progression of the disease. However, surgical intervention is sometimes required to alleviate intraocular pressure and some forms of open angle glaucoma become refractory to treatment . Open angle glaucoma affects up to 1 million persons within the British Isles at the present time. While most forms of the disease are ^λmultigenic' or ^λmultifactorial' , some forms of the diseases are inherited according to apparent mendelian ratios, ie, they are transmitted in an autosomal dominant sense. In some such forms of disease, mutations within the so-called myocilin gene have been identified (Stone et al, Science, 275, 1997, 668-670) . Moreover, in up to 4% of multifactorial forms of disease, similar mutations have been encountered. Thus, 40,000 persons, or more, within the British Isles, have a form of glaucoma caused my mutations within the myocilin gene. Since mice with a targeted disruption of the myocilin gene appear to be phenotypically normal, with no apparent eye symptoms, it is unlikely that the forms of glaucoma with mutations in the myocilin gene are caused by 'haplosufficiency' , rather, mutations within the gene have a dominant negative effect on phenotype (Kim et al, Molecular and Cellular Biology, 21, 2001, 7707-7713).

Humans have approximately 35,000-40,000 genes, each gene being duplicated (alternative forms of each gene are termed 'alleles'), one member of each pair being inherited from either parent. In recessive forms of disease each allele is non-functional . Therefore, therapies for this form of disease require 'gene replacement'. As an example of this approach, delivery of the missing gene to the retinas of the Briard dog, these animals having a form of retinal degeneration akin to a disease termed Leber congenital amaurosis in man (LCA) , a condition similar in many respects to RP, by injecting a virus (adeno-associated virus) carrying a functional version of the gene in question (termed RPE65) has been shown to partially restore visual function in these animals (2) . So far as we are aware, approval for phase 1 clinical trails in human subjects is currently being sought in the United States and possibly elsewhere. In another example, the retinas of the so-called Royal College of Surgeons (RCS) rat can also be partially 'rescued' using a similar approach (3) .

In contrast to the above (recessive disease) , in many so-called dominant forms of disease, both alleles of a given gene remain active. In this situation one allele produces a normal gene product (the allele is transcribed into a messenger RNA which is then translated into a protein product) and is inherited from the parent who does not suffer from disease symptoms. However, the other allele, inherited from a parent with the disease, has become altered (mutated) such that it produces an altered transcript, which in turn encodes an altered protein which exerts a toxic effect. The altered protein is toxic to the photoreceptor cells and such cells gradually die, resulting in a retinal degeneration.

A number of agents, for example, hammerhead ribozymes or siRNA molecules, have been developed which are capable of discriminating between mutant and normal transcripts and selectively destroying the former. However, in dominant RP, as in many other dominant diseases, many different mutations may occur within the same gene in different patients. Design constraints prevent hammerhead ribozymes, and to an appreciable extent, siRNA molecules, from being designed to target all such mutations and even if such design constraints could be lifted it would be commercially inviable to design and clinically validate individual therapeutic agents of this sort, given that in some genes well over one hundred different mutations have been encountered to date in specific forms of disease.

However, a means has been developed to circumvent this significant problem of 'intragenic heterogeneity' by developing suppressing agents such as those described above to act in a so-called 'mutation-independent.' manner (1, 4-7) . The agent simultaneously targets a selected site (an optimal site for cleavage) in transcript derived from both normal and mutant alleles, a replacement 'therapeutic' gene being introduced as part of the therapeutic process, which has been altered, for example at the third base (degenerate) site of any given codon such that transcript from it now escapes suppression but nevertheless encodes functional protein. Such approaches have been termed 'mutation-independent suppression-replacement' technologies and have the advantage that a single therapeutic agent (for example hammerhead ribozyme or siRNA molecule) can be used to target all mutations within a given gene (1, 4-7) .

A gene for one form of autosomal dominant RP was located in 1993 on the long arm of chromosome 7 (8) . By convention, the disease locus was termed RP10. Using modern methods involving analysis of the global transcriptional profiles of the retinas of normal mice and those with an engineered form of retinal degeneration (9) a gene was identified as a candidate for mutational screening in this form of disease. This lead to the identification of a mutation in the gene encoding inosine monophosphate dehydrogenase 1 (IMPDHl) in RP10 patients (10) . IMPDHl encodes the rate-limiting enzyme of the so- called de novo pathway of GTP biosynthesis . Already the number of mutations identified within the IMPDHl gene is expanding even though the gene was only very recently implicated in retinal disease.

Statements of Invention

In a first aspect, there is provided a method of treating an autosomal dominant disease of the eye, which method comprises a step of suppressing substantially all alleles of a gene associated with the disease, wherein a step of introducing a replacement gene is disclaimed.

In a second aspect, the invention relates to a medicament for the treatment of an autosomal dominant disease of the eye, the medicament comprising means for suppressing all alleles of a gene associated with the disease, either alone or in a vector, wherein medicaments including a replacement gene are disclaimed.

In a third aspect, the invention relates to a kit for use in the treatment of an autosomal dominant disease of the eye, the kit comprising means for suppressing all alleles of a gene associated with the disease, either alone or in a vector, wherein medicaments including a replacement gene are disclaimed.

In a fourth aspect, the invention relates to a use of means for suppressing all alleles of a gene associated with a disease, without a replacement for the endogenous gene, alone or in a vector, in the manufacture of a medicament for the treatment of an autosomal dominant disease of .the eye.

Suitably, the disease in one in which a mutation in one allele (the diseased allele) of the gene is implicated in the disease pathology. In such cases, the treatment involves suppressing both a normal and diseased allele of the gene. In other cases, the disease may be one in which there is no mutation in the gene implicated in the disease pathology, rather there is one or more extra copies of the gene present. When the disease in one in which there is a normal and a diseased allele of the gene, typically the means for suppressing both alleles of a gene of interest comprises one or more suppression effectors.

In this specification, the term "suppression" should be taken to mean silencing or reducing gene expression, generally in a sequence specific manner. Likewise, the term "suppression effectors" should be taken to include the nucleic acid or amino acid based moieties which are used to silence or reduce expression of the gene(s) of interest. In this specification the term "replacement gene" should be understood as meaning a gene, typically a normal allele thereof, which is used in so-called mutation- independent, suppression-replacement strategies to replace a diseased allele of the gene and which is expressed to produce a wild-type protein.

Suitably, the disease is of the type for which inactivation of both alleles of a gene associated with the disease has minimal negative effect on a diseased subject. Examples of such diseases include RPlO-type Retinitis pigmentosa and glaucoma.

Typically, the suppression step involves treatment with a suppression effector which is adapted to down-regulate transcripts of the endogenous gene. In a particularly preferred embodiment, the suppression effector is adapted to down-regulate transcripts from both a diseased allele and normal allele of the endogenous gene, and, ideally, transcripts possessing different mutations of a diseased allele of a given gene.

In one embodiment of the invention, the suppression effector is chosen from the group comprising: nucleic acids; peptide nucleic acids (PNA's); peptides; antibodies; ribozymes; hammerhead ribozymes; siRNA; and triple helix nucleotides . SiRNA and hammerhead ribozymes are particularly preferred.

In one specific embodiment the disease is RPlO-type Retinitis pigmentosa, and wherein the gene to be suppressed is IMPDHl. The sequence of a number of siRNA suppression effectors suitable for silencing IMPDHl gene are provided below.

For treatment of glaucoma or RP, the suppression step is carried out specifically in retinal tissue.

In the RP10 type RP embodiment, the method includes a further step of treating retinal tissue with a metabolite of IMPDHl such as XMP or GTP. In this specification, the term "metabolite of IMPDHl" should be taken to mean any compound lying downstream of the rate limiting enzyme (IMPDHl) of de novo GTP biosynthesis.

In a fifth aspect, the invention relates to a method for the treatment of Retinitis pigmentosa, typically RPlO-type Retinitis pigmentosa, comprising a step of delivering a metabolite of IMPDHl such as XMP or GTP to retinal tissue of an individual in need of treatment .

Ideally, the method further includes a step of delivering to the retinal tissue means for suppressing both a normal and diseased allele of IMPDHl .

Typically, the means of delivery of the metabolite to retinal tissue is selected from the group comprising: sub-retinal delivery; intra-ocular delivery; delivery via the venous system; intraperitoneal injection; iontophoresis; microdialysis; oral injestion; and eye drops.

In a sixth aspect, the invention relates to a medicament for the treatment of RPlO-type Retinitis - pigmentosa, comprising a metabolite of IMPDHl such as XMP or GTP.

In a seventh aspect, the invention relates to a metabolite of IMPDHl such as XMP or GTP in the manufacture of a medicament for the treatment of Retinitis pigmentosa, typically RPlO-type Retinitis pigmentosa.

In an alternative embodiment of the invention there is provided a therapy for those forms of glaucoma caused by mutations within the myocilin gene, the therapy comprising means for substantially complete ablation of transcripts derived from the myocilin gene alleles. Both normal and disease allele transcripts are substantially ablated.

The means for causing substantially complete ablation can act at the level of DNA, or transcript or protein.

The means could provide a therapeutic agent for glaucoma.

Suitably the means could be siRNA or shRNA.

Any system capable of delivery of such agents to the appropriate tissues of the eye could be used, for example, naked DNA, DNA complexed with an agent such as a liposome or a viral delivery agent. Any such viral agent could in principle be used, for example, adeno-associated virus (AAV) .

Glaucoma is one of a group of autosomal dominant diseases of the eye for which therapies based on the suppression only paradigm for RP10 described in detail herein could be developed.

Whereas the discussion herein concentrates on RP10 therapy the invention can be extended to relate to other autosomal dominant diseases of the eye.

siRNA The silencing effect of complementary double stranded RNA was first observed in 1990 in petunias by Richard Joergensen and termed cosuppression. RNA silencing was subsequently identified in C. elegans by Andrew Fire and colleagues, who coined the term RNA interference (RNAi) . RNAi has been shown to be effective in both mammalian cells and animals . An important feature of dsRNA or siRNA or RNAi is the double stranded nature of the RNA and the absence of large overhanging pieces of single stranded RNA, although dsRNA with small overhangs and with intervening loops of RNA has been shown to effect suppression of a target gene. The pathway for silencing gene expression involving long (>30 nucleotides) double stranded RNA molecules has been elucidated and is thought to work via the following steps (shown in Drosophila melanogaster) . Firstly, the long dsRNA is cleaved into siRNA approximately 21 nucleotides in length. This siRNA targets complimentary mRNA sequence, which is degraded. However, in mammals it has been found that long dsRNA triggers a non-specific response causing a decrease in all mRNA levels. This general suppression of protein synthesis is mediated by a dsRNA dependent protein kinase (PKR) . Elbashir et al . were able to specifically suppress target mRNA with 21 nucleotide siRNA duplexes. Notably, siRNA bypassed the non-specific pathway and allowed for gene-specific inhibition of expression. dsRNA can be delivered as synthesized RNA and or by using a vector to provide a supply of endogenously generated dsRNA. dsRNA may be locally or systemically delivered. Indeed functional siRNAs have been generated both in cells and in transgenic animals and have been delivered using a variety of vector systems including lentivirus.

The siRNA may be blunt ended or may have overhangs at its 3' or 5 ' termini, preferably at both of its termini . The overhangs are preferably short in length, for example less than 30 nucleotides, preferably less than 20 nucleotides more preferably less than 10 nucleotides, even more preferably less than 5 nucleotides, most preferably less than 3 nucleotides in length. Typically, the overhangs are two nucleotides in length.

Typically the region of the siRNA sequence with sequence identity to the target gene is from 14 to 30 nucleotides in length, for example from 16 to 24 nucleotides, more preferably from 18 to 22 nucleotides, most preferably from 19 to 21 nucleotides in length.

Small Interference RNA (siRNA) Suppression Effectors for IMPDHl

ShRNAl Forward strand 5 ' GATCCCCAAGCTGGTGGGCATCGTCATTCAAGAGATGACGATGCCCACCA GCTTTTTTTGGAAA3 '

ShRNAl Reverse strand 5 'AGCTTTTCCAAAAAAAGCTGGTGGGCATCGTCAAGAGAACTTTGACGATG CCCACCAGCTTGGG3 ' ShRNA2 Forward strand 5 ' GATCCCCAAGTTTGAGAAGCGGACCATTCAAGAGATGGTCCGCTTCTCAA ACTTTTTTTGGAAA3 '

ShRNA2 Reverse 5 'AGCTTTTCCAAAAAAAGTTTGAGAAGCGGACCAAGAGAACTTTGGTCCGC TTCTCAAACTTGGG3 '

ShRNA3 Forward strand 5 ' GATCCCCGTTTGAGAAGCGGACCATGTTCAAGAGACATGGTCCGCTTCTC AAACTTTTTGGAAA3 '

ShRNA3 Reverse strand 5 'AGCTTTTCCAAAAAGTTTGAGAAGCGGACCATGAGAGAACTTCATGGTCC GCTTCTCAAACGGG3 '

ShKNA4 Forward strand 5 ' GATCCCCCTCACCTACAACGACTTCCTTCAAGAGAGGAAGTCGTTGTAGG TGAGTTTTTGGAAA3 '

ShRNA4 Reverse strand 5 ' AGCTTTTCCAAAAACTCACCTACAACGACTTCCAGAGAACTTGGAAGTCG TTGTAGGTGAGGGG3 '

ShRNA5 Forward strand 5 ' GATCCCCGTGCCCTACCTCATAGCAGTTCAAGAGACTGCTATGAGGTAGG GCACTTTTTGGAAA3 '

ShRNA5 Reverse strand 5 'AGCTTTTCCAAAAAGTGCCCTACCTCATAGCAGAGAGAACTTCTGCTATG AGGTAGGGCACGGG3 ' SiRNAl224 Forward strand 5 ' GCUGCCUAUCGUCAAUGAU3 ' SiRNAl224 Reverse strand 5 'AUCAUUGACGAUAGGCAGC3 ' ^• Si RNA 2048 Forward strand 5 'UGUACUCAGGAGAGCUCAA3 Si RNA 2048 Reverse strand 5 'UUGAGCUCUCCUGAGUACA3 ' ShMR3 Forward strand 5 ' GCCTGAGGTCAACAACGAA3 ' SiMR3 Reverse strand , 5 ' TTCGTTGTTGACCTCAGGC3 ' Recently, novel suppression technology such as RNA interference (RNAi) has been exploited as a reverse genetic tool to modulate gene expression and for the study of gene functionality in multiple organisms. RNAi is a multistep process and is activated by double-stranded RNA (dsRNA) that is homologous in sequence to the silenced gene. Initially, long triggering ds-RNAs are processed into small interfering RNAs (siRNA) of 20-25nt through the action of an RNase III endonuclease. The fragmented siRNAs associate with a multi-component nuclease and mediates the degradation of homologous RNA. A conventional DNA vector-based approach for the synthesis of short hairpin RNAs (shRNA) driven by Hi RNA promoter is used to silence IMPDHl in mammalian cultured cell lines. Five 21nt sequence motifs with a 9 nt stem loop targeting IMPDHl mRNA are sub- cloned into pSuper vector and acted as templates for the synthesis of shRNAs under the control of Hi promoter. Through RT-PCR analysis, a high level of endogenous IMPDHl transcript has been detected in certain mammalian cell lines (e.g. HEK293T and cos- 7) . Following this, the designated sh-RNA expressing vectors are transfected into mammalian cell lines and the degree of shRNA-induced silencing on endogenous IMPDHl transcripts are evaluated. Subsequently, optimal shRNAs are expressed transgenically in mice. shRNA-expressing mice are then crossed to mice expressing a mutant human IMPDHl transgene and/or a replacement siRNA resistant human IMPDHl transgene, and the resulting double /triple mutant progeny are evaluated for IMPDHl activity.

Design of siRNA Molecules

The techniques of designing siRNA molecules which are capable of silencing both alleles of a given gene will be well known to those skilled in the art (39, 40) .

Ribozymes A ribozyme can be designed to cleave a dsRNA molecule by designing specific ribozyme arms which bind to a particular RNA on either side of a consensus NUX site 5' and or 3 ' to the dsRNA sequence, where N is selected from the group consisting of C, U, G, A and X is selected from the group consisting of C, U or A. Thus any RNA sequence possessing NUX site is a potential target. However, other variables require consideration in designing a ribozyme, such as the two dimensional conformation of the dsRNA containing the nucleotides that are to be cleaved by the ribozyme (e.g., loops) and the accessibility of a ribozyme for its target. The utility of an individual ribozyme designed to target NUX site in an open loop structure of transcripts comprising the dsRNA will depend in part on the robustness of the RNA open loop structure. Robustness may be evaluated using an RNA-folding computer program such as RNAPlotFold. A robust loop refers to the occurrence of the loop for most or all of the plotfolds with different energy levels. Robustness of loop structures is evaluated over a broad energy profile, depending on the length of the sequence, according to art known parameters.

While various agents such as ribozymes that cleave RNA at site specific recognition sequences can be used to cleave dsRNA, the use of hammerhead ribozymes is preferred. Hammerhead ribozymes are small catalytic RNA enzymes that can elicit sequence specific cleavage of a target RNA transcript. Hammerhead ribozymes cleave RNAs at locations dictated by flanking regions that form complementary base pairs with the target RNA. The sole requirement is that the target RNA has the following sequence of two bases: 5'-UX-3' where X = A, C or U. The construction and production of hammerhead ribozymes is well known in the art.

Ribozymes for use in the present invention may also include RNA endoribonucleases (hereinafter "Cech- type ribozymes") such as the one which occurs naturally in Tetrahymena Thermophila (known as the IVS, or L-19 IVS RNA) and which has been extensively described by Thomas Cech and collaborators . The Cech-type ribozymes have an eight base pair active site which hybridizes to a target RNA sequence whereafter cleavage of the target RNA takes place. The invention encompasses those Cech-type ribozymes which target eight base-pair active site sequences that are present in a target allele. Hairpin, hammerhead, trans-splicing ribozymes and indeed any ribozyme could be used in the practice of the invention. In addition, any RNA inactivating or RNA cleaving agent which is capable of recognition of and/or binding to specific nucleotide sequences in a dsRNAi(e.g. splicesome-mediated RNA trans-splicing) is contemplated. Suppression agents of the invention also include minizymes, maxizymes and or any other suppression agent (s) able to cleave a target RNA in a sequence specific manner. The ribozymes can be composed of modified oligonucleotides (e.g., for improved stability, targeting, etc.). Modified oligonucleotides can be transfected into cells which express dsRNA.

A preferred method of delivery involves using a DNA construct "encoding" the ribozyme under the control of a strong tissue specific, cell specific or inducible promoter, so that transfected cells will produce sufficient quantities of the ribozyme to cleave the dsRNA. Because ribozymes unlike antisense molecules, are catalytic, a lower intracellular concentration of ribozymes may be required for efficient cleavage.

Vectors

Vectors of and for use in the invention may be viral, non-viral, an artificial chromosome or any vehicle for delivery of the dsRNA nucleotides. Exemplary viral vectors which may be used in the practice of the invention include those derived from adenovirus; adenoassociated virus; retroviral-C type such as MLV; lentivirus such as HIV or SIV; herpes simplex (HSV) ; and SV40. Exemplary, non-viral vectors which may be useful in the practice of the invention include bacterial vectors from Shigella flexneri, such as the S. flexneri which is deficient in cell-wall synthesis and requires diaminopimelicacid (DAP) for growth. In the absence of DAP, recombinant bacteria lyse in the host cytosol and release the plasmid. Cationic lipid mediated delivery of suppression effectors, soluble biodegradable polymer-based delivery, or electroporation/ ionthophoresis may also be used. Delivery may be in vivo or ex vivo to cells.

Nucleic acids encoding at least one dsRNA and / or ribozyme for suppression of gene expression may be provided in the same vector or in separate vectors . The dsRNA can be delivered as naked DNA, modified DNA, naked RNA or in a carrier vehicle or vector. Naked nucleic acids or nucleic acids in vectors can be delivered with lipids or other derivatives which aid gene delivery. Nucleotides may be modified to render them more stable, for example, resistant to cellular nucleases while still supporting RNaseH mediated degradation of RNA or with increased binding efficiencies .

In an embodiment, vector constructs can include more than one dsRNA nucleotide sequence, wherein each dsRNA may target either the same or different target genes or target nucleotide sequences.

Vectors encoding a tissue specific and or cell specific and or inducible dsRNA may be delivered alone or with one or more agent (s) to aid delivery of constructs and or nucleotides.

Pharmaceutical Compositions

Pharmaceutical compositions according to the present invention, and for use in accordance with the present invention may comprise, in addition to active ingredient, a pharmaceutically acceptable excipient, carrier, buffer stabiliser or other materials well known to those skilled in the art. Such materials should be non-toxic and should not interfere with the efficacy of the active ingredient. The precise nature of the carrier or other material will depend on the route of administration, which may be, for example, oral or intravenous or by other routes of injection.

The primary therapy for the RP10 form of retinitis pigmentosa described herein involves simultaneous ablation of both normal and mutated alleles of the IMPDHl gene, resulting in the conversion of a severe autosomal dominant condition into a very much milder form of disease, which, from data accrued in mice with a targeted disruption of the IMPDHl gene, may be akin to forms of night blindness. However, since IMPDHl-/- mice have no IMPDHl enzyme, supplies of nucleotides downstream of the block in the metabolic pathway induced by null mutations within the IMPDHl gene, for example, XMP, GMP, GDP or GTP, are limiting in the retinas of these animals. Since an adequate supply of such compounds is necessary for normal retinal function, the principle of a form of supplemental therapy for the RP10 form of RP presents itself. According to this process XMP, GMP, GDP or GTP or structurally and/or functionally related compounds are introduced into retinal tissues. Any method of introduction of such compounds into retinal tissues would be appropriate, for example, by natural diffusion following administration of eye drops, or by iontophoresis or electroporation following eye-drop administration or administration by intra-ocular or sub-retinal injection

Examples of the techniques, formulations and protocols mentioned and other techniques, formulations and protocols which may be used in accordance with the invention can be found in Remington's Pharmaceutical Sciences, 16th edition, Oslo, A. (ed), 1980.

The invention will be more clearly understood from the following description of some experiments, given by way of example only, with reference to the following Figures wherein

Figure 1 illustrates in situ hybridisation studies on mouse retinal cryosections

Figure 2 shows ERG analysis in Impdhl^-/" mice

Figure 3 shows light micrographs of retinal sections from Impdhl^{- "} mice.

Figure 4 shows ERG analysis of delayed dark-adaption in Impdh-/- mice

Figure 5 illustrates an analysis of wild-type and mutant IMPDHl proteins expressed in HEK293T cells Figure 6 shows fluorescent analysis of Hela cells expressing IMPDHl-GFP fusion proteins.

Figure 7 shows computer modelling of wild-type and mutant IMPDHl proteins

Figure 8 is a schematic illustration of IMPDHl Arg224Pro transgene

Figure 9 shows down-regulation of IMPDHl by shRNAs in HEK293T cells

Figure 10 shows down-regulation of IMPDHl by shRNA combinations in HEK293T cells

Figure 11 shows concentration-dependent down- regulation of IMPDHl expression by shRNA.

Figure 12 shows down-regulation of IMPDHl by siRNAs in HEK293T cells

Figure 13 shows epifluorescent microscopic pictures of HEK293T cells after 72hr of co-transfection with sil224 or si2048 and wild type or mutant EGFP-IMPDHl

Figure 14 shows quantification of IMPDHl protein expression in HEK293T cells after 72hrs of co- transfection with pwtIMPDHl-gfp/ pmutIMPDHl-gfp and sil224/ si2048. Figure 15 shows Western blot analysis of IMPDHl protein expression in Hela cells

Figure 16 shows ERG analysis of Impdh-/- mice following intravitreal injection of GTP

Figure 17 shows HPLC analysis of retinal purine nucleotides in Impdh-/- mice following subretinal injection of XMP.

MATERIALS AND METHODS

In situ hybridisations In si tu hybridisations involved the use of digoxigenin (DIG) -labelled riboprobes on frozen cryosections . Sense and antisense DIG-labelled probes were generated from PCR templates which had incorporated T7 and T3 promoters . Sections were fixed in paraformaldehyde post cutting, treated with active DEPC and hybridisation with both sense and antisense probes was allowed to proceed overnight at 58°C. Following stringent washes with SSC, incubation with an AP coupled, anti-DIG antibody was carried out. Binding of the probes was detected using NBT/BCIP solution. Sections were mounted and analysed using a Zeiss Axioplan 2 microscope.

Mouse electroretinography Mouse ERGs were recorded at 6 weeks of age (2 animals) , 5 months (6 animals) , 8 months (4 animals) 11 months (5 animals) and 13 months (3 animals) . Animals were dark-adapted overnight and prepared for electroretinography under dim red light. Pupillary dilatation was achieved by instillation of Cyclopentalate 1% and Phenylephrine HCl 10%. The subjects were anaesthetized by means of Ketamine (2.08 mg per 15 gram body weight) and Xylazine (0.21 mg per 15 gram body weight) injected intraperitoneally. Standardized flashes of light were presented to the mouse in a Ganzfeld bowl to ensure uniform retinal illumination. The ERG responses were recorded simultaneously from both eyes by means of contact lens electrodes (Medical Workshop, Groningin, Netherlands) using Amethocaine 1% as topical anaesthesia and Vidisic (Dr. Mann Pharma, Germany) as a conducting agent and to maintain corneal hydration. A reference electrode was positioned subcutaneously approximately 1 mm from the temporal canthus and the ground electrode placed subcutaneously anterior to the tail . The responses were analysed using RetiScan RetiPort electrophysiology equipment (Roland Consulting Gmbh) . The protocol used was based on that approved by the International Clinical Standards Committee for human electroretinography. Rod-isolated responses were recorded using a dim white flash (-25 dB maximal intensity) presented in the dark-adapted state. The maximal combined rod/cone response to the maximal intensity flash was then recorded. Following light adaptation for 10 minutes to a background illumination of 30 candelas per m2 presented in Ganzfeld bowl the cone-isolated responses were recorded to the maximal intensity flash (3 candelas per m2 per second) presented as a single flash and 10 Hz flickers, a-waves were measured from the baseline to the trough and b-waves from the baseline (in the case of rod-isolated responses) or from the a-wave trough.

ERG analysis of dark adaption kinetics C57BL6 wild-type mice were dark-adapted overnight and base-line electroretinograms were performed. To investigate the dark-adaptation kinetics of the C57BL6 wild-type animals some days later the mice were dark-adapted overnight then light adapted for 10 minutes. Following light adaptation the animals were placed in a totally dark room for 30 minutes. Following the 30 minutes dark-adaptation they were anaesthetised with Ketamine and Xylazine and an electroretinogram was performed. IMPDH-/- mice were dark-adapted overnight and base- line electroretinogram were performed. To investigate the dark-adaptation kinetics of the IMPDH-/- mice some days later the mice were dark- adapted overnight then light adapted for 10 minutes. Following light adaptation the animals were placed in a totally dark room for 30 minutes, 60 minutes, 90 minutes, 120 minutes, 150 minutes and 180 minutes. Following the 30 minutes dark-adaptation they were anaesthetised with Ketamine and Xylazine and an electroretinogram was performed.

Retinal histology Eyes were perfused overnight with a mixture of 2.5% glutaraldehyde and 2% paraformaldehyde on 0. IM phosphate buffer, pH7.4, and processed in Epon. Sections lμm thick were cut through the optic nerve head, along the vertical meridian of the eye, and were stained with toludine blue for light microscopy.

Expression of wild type and mutant His-tagged IMPDHl in E.coli Primers were designed to introduce Xhol sites at either end of the wild-type IMPDHl cDNA sequence in the pGEM vector (11). The 1.5kb cDNA was amplified by PCR with Pfu DNA polymerase (Stratagene) . IMPDHl cDNA inserts were then cloned into the Xhol site of the pET-15b vector (Novagen) which incorporates a His-tag at the N-terminus of the protein sequence. The single point mutations CGC to CCC (Arg224Pro) and GAC to AAC (Asp226Asn) were introduced using the Quikchange site-directed mutagenesis kit (Stratagene) . All inserts were sequenced completely using automated sequencing on an ABI 310 Genetic Analyzer (Perkin Elmer, Shelton, Connecticut) and Big Dye Terminator chemistries (Perkin Elmer) , to verify the presence of intended mutations and the absence of further unwanted mutations . Constructs were transformed into E. coli BL-21(DE3) for protein expression. Cells were grown at 37°C in 100 ml LB to an OD600 of 1.0 and induced with 400 (M IPTG. Induction of protein expression was carried out at 25°C because at higher temperatures only very low levels of IMPDHl protein, especially mutant protein, were detectable in soluble cell lysates . A lower induction temperature improves protein solubility and helps prevent inclusion-body formation. Cells were harvested 3-5 hours post-induction, resuspended in Ni-NTA bind buffer (Novagen) , treated with lysozyme and sonicated. Soluble cell lysates were enriched for His-tagged protein using Ni-NTA His- bind resin (Novagen) according to manufacturers instructions, and relatively pure samples of IMPDHl protein were isolated. Proteins samples were run on 10% SDS-polyacrylamide gels and stained with Coomassie Blue. Enzyme activity assay of recombinant IMPDHl proteins Protein concentration was determined using the Bio- Rad protein assay and bovine IgG as standard. The standard enzyme assay buffer contained 0.1 M Tris, 0.1 M KC1, 3 mM EDTA, 2 mM DTT pH 8.0, and the substrates 400 μM IMP and 400 uM NAD. The reaction was started by addition of 5-25 μg purified recombinant IMPDHl to buffer in 1 cm cuvettes. The initial rate of activity was measured in a temperature-controlled Cary 50 UV-Vis spectrophotometer by monitoring the absorbance increase at 340 nm for 1 min due to the formation of NADH from NAD+, and Cary WinUV software was used to measure reaction rates.

Expression of wild-type and mutant His-tagged IMPDHl in HEK 293T cells The wild-type and mutant IMPDHl cDNA sequences were sub-cloned from the pET15b vector by cutting with Xhol, inserts were blunt-ended and ligated into the blunted Acc651 site of the pcDNA3.1/HisC vector, which incorporates an N-terminus His-tag in the resulting protein (Invitrogen) . Human embryonic kidney 293 (HEK 293T) cells were cultured in DMEM with 10% fetal bovine serum at 37°C and an atmosphere of 5% C02. Cells were seeded at a density of 106 cells per 10 cm dish overnight and were transiently transfected using 20 μg of DNA and the calcium phosphate method. Cells were harvested by trypsinisation after 48 hours, washed twice with PBS and centrifuged gently to pellet. A protein isolation protocol for extraction of cell fractions was adapted from Deery et al . (12). Cell pellet from one dish of cells was resuspended in 1 ml Buffer A (10 mM HEPES pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.5 mM PMSF, 0.25 mM benzamidine) . Cells were incubated on ice for 10 mins then re-pelleted by spinning for 10 mins in a microfuge at 2000 rpm at 4°C. Cells were resuspended in 300 μl buffer A, and for each prep a 100 μl sample was removed for whole cell protein analysis. NaCl was added to this aliquot to a cone, of IM to lyse all membranes. To the remaining 200 μl of cell suspension detergent Nonidet P-40 was added to a concentration of 0.2%, the mixture was incubated on ice for 10 mins and the suspension was spun at 3500 rpm for 10 mins at 4°C. The supernatant which contains the soluble cytosolic proteins was removed. The pellet containing nuclei and insoluble components was resuspended in 200 μl Buffer B (5 mM HEPES pH 7.9 , 1.5 mM MgCl2 , 0.2 mM EDTA, 0.5 mM PMSF, 0.25 mM benzamidine) and NaCl was added to a final concentration of 1 M. After 45 minute incubation at 4°C, nuclear lysis occurs and viscosity was reduced by shearing DNA and passing 10 times through a 25 gauge needle. The suspension was then spun at 13000 rpm for 40 mins at 4°C. The supernatant containing soluble nuclear proteins was preserved. The final pellet was resuspended in 100 μl of 1% SDS to denature and solubilise remaining proteins. Protein concentrations were then measured using BioRad Protein assay kit with bovine IgG as standard. SDS-PAGE and western analysis of protein extracts Aliquots containing equal amounts of total protein (30 μg) were loaded onto 10% SDS-PAGE gels in duplicate. One gel was stained with Coomassie blue and the proteins on the second gel were transferred by electroblotting onto nitrocellulose membrane. Blots were blocked with 1% casein in TBS (Novagen) and then probed with an Anti His-tag monoclonal antibody (Novagen 200 ng/ml) . After washing the blots were probed with a horseradish peroxidase- conjugated anti-mouse IgG secondary antibody (Sigma) and signal was detected using SuperSignal West Pico Chemiluminescent Substrate (Pierce) according to manufacturers instructions.

Mammalian expression of wild-type and mutant GFP- tagged IMPDHl proteins. The IMPDHl wild-type and two mutant cDNA sequences (as described previously) , were subcloned into a pcDNA 3.1 vector that incorporates an EGFP protein tag at the N-terminus of the IMPDHl protein, and is under the control of a CMV promoter. Hela cells were grown in DMEM with 10% fetal bovine serum in 5% C0₂ at 37°C. 1.5 x 10⁵ cells were plated onto coated coverslips in 12-well plates. After 24 hours, cells were transfected with 1 μg plasmid DNA using Lipofectamine 2000 (Invitrogen) according to manufacturers instructions. Cells were fixed with 4% paraformaldehyde, washed, placed onto slides and stained with DAPI for nuclear staining. Slides were then mounted and visualized 24-72 hours post- transfection using a fluorescent microscope for both GFP and DAPI detection.

Protein modelling computational methods Homology Modelling. The IMPDH-2 crystal structure entries 1B30 (Homo sapiens) and 1JR1 (Cricetulus griseus) were downloaded from the PDB database ( http://www.rcsb.org). The manually edited structures for chain B of 1B30 and chain A of lJRl were read into the Insightll molecular modelling programme. The protein sequences were extracted and superposed using the structure alignment tool in the homology module. Any resultant alignment errors corrected manually. The sequence of the wild type IMPDH-1 protein was loaded into the homology module. Coordinates were assigned using those residues conserved between the wildtype sequence and 1B30. Where conservation was not present for wildtype and 1B30, but possible for 1JR1, the coordinates were assigned using the latter structure. The resultant model had coordinates assigned for 90.1 % of the sequence and was saved once null coordinate residues were pruned. The mutant structure was created by replacing Arg 224 with a proline residue in the Biopolymer module of Insightll. Both structures were subjected to the energy minimisation protocol detailed below before initiating molecular dynamics studies.

Energy Minimisations. For all crystals, hydrogens were added assuming a pH of 7.4 and standard aminoacid pKa values using the Biopolymer module in Insight 2000 (Accelrys) . The CFF forcefield was used throughout all the minimisations and subsequent MD simulations. No cut-offs were used to treat non- bonded interactions, while a fixed dielectric constant of 4.0 was used for computing electrostatic interactions. The protein structure was subjected to an energy minimisation, using the Discover 3 module in Insight 2000 (Accelrys) . This initial energy minimisation was carried out in a staged manner to prevent strong steric clashes from perturbing greatly the initial conformation. Van der Waals interactions were scaled down to 33% of their real value and a minimisation of 100 steps using steepest descents was performed. Van der Waals interactions were then scaled up to 66% of their real value and a second minimisation of 100 steepest descents steps was carried out. Van der Waals interactions were then scaled up to 100% of their real value and a third and final minimisation of 250 steepest descents steps was then conducted. Following this a Conjugate-Gradient minimisation of 500 steps was executed.

Molecular dynamics. A series of molecular dynamics (MD) simulations were carried out on the JRl wild type and mutant X-ray structures as well as on the Homology model wild type and mutant structures in order to efficiently explore conformational space. All MD simulations were conducted using the Discover 3 module in Insight 2000 (Accelrys) and the following conditions were applied. A canonical (N,V,T) ensemble was used and the NosE-Hoover thermostat (13) was applied to keep the temperature at 300 K. The Verlet velocity method (14) was employed to integrate the equations of motion, with a timestep of 0.5 fs, and the RATTLE algorithm was applied to constrain all bonds. The initial velocities were assigned randomly to match a Boltz ann distribution of the temperature. An initial equilibration period of 10,000 steps (5 ps) was allowed before proceeding to collection periods of 1,000,000 steps (500 ps) in each MD simulation. No other constraints were applied to the simulations and the whole protein structure was allowed to have full conformational flexibility. After the MD simulations were completed, the stored configurations were subjected to full energy minimisations. Initially, 500 steps of steepest descents were carried out, followed by a second minimisation stage using conjugate gradients with the Fletcher-Reeves method (15) up to a convergence gradient of 0.01 kcal mol-1 *-l.

Generation of the IMPDHl Arg22 Pro transgenic mouse A 5.7kb IMPDHl promoter fragment was obtained by long range PCR using mouse genomic BAC clone RPC123 as a template. Long range PCR was performed using a combination of Taq and Proofstart taq (Qiagen) with addition of Q solution to the PCR reaction mix. PCR cycle times were extended by 20s/cycle for the final 25 of a total of 35 cycles and incorporated a combined annealing and extension step of 68°C for 7 min. HindiII and BsaAI restriction sites were appended to the ends of forward and reverse primers respectively which were as follows; CGCAAGCTTAGTCTGGTACCAAGGGCAGA and TGAGCCCATCCTCGGGCACGTA. The reverse primer is located in exon 1 of the coding sequence. A single base change in the primer was introduced to give a BsaAI restriction site as is found in the human sequence at this location. The human and mouse sequences prior to this site are identical. A human IMPDHl coding sequence fragment was obtained by PCR from the pGEM clone containing human coding sequence, with an incorporated Arg224Pro mutation, as described above. The forward primer was designed to amplify a fragment with a start position at the BsaAI restriction site. The reverse primer was designed to provide a xhol site on the 3' end of the resulting PCR fragment . The promoter and coding sequence fragments were combined by blunt end ligation in the PcDNA3.1 vector which had been digested with HindiII and Xhol. The vector was prepared by caesium chloride gradient and the transgene released from the vector backbone by restriction digestion with HindiII and Sphl to include a poly adenylation sequence. Following digestion the 7kb transgene fragment was gel purified using the Qiaquick Gel- Extraction kit (Qiagen) . The 7kb transgenic construct was gel purified and resuspended in lx injection buffer (7.5mM Tris pH 7.4, 0.15mM EDTA) at a concentration of 1-2 ug/ml. Pronuclear injections of the DNA into C57B1/6J x CBA/J embryos are currently being carried out. Construction of small hairpin RNAs targeting IMPDHl using a DNA-vector based approach By following the guidelines provided by uww.oli5roensrii2e.co-. for the expression of short interfering RNAs from a vector system, five different sequences encoding shRNAs with 19 bases of homology to IMPDHl were synthesised as oligonucleotides (Sigma-Genosys Ltd) and annealed to give double-stranded DNA (ds-DNA) . The selection of the coding sequences for shRNAs was empirically determined using the search-engine supplied by ww . dharmacon . co / www.ambion.com, and analysed by BLAST search (www.ncbi.nlm.nih.gov/BLAST) to minimise sequence homology with other genes. The oligonucleotides were sub-cloned into pSuper vector (39) at the Hind III and Bgl II restriction sites. Recombinant pSuper vector was then transformed into SURE2 supercompetent cells (Stratagene^®) . The presence of positive clones was determined by PCR analysis using p-Bluescript primers and DNA sequencing (ABI PRISM) was used to confirm the correct insertion of shRNA encoding sequences into pSuper vectors, data not shown. Furthermore, two chemically synthesised siRNA duplexes (Dharmacon Research Inc.) specific for IMPDHl were selected using above electronic search engines and with reference to a RNAi selection protocol supplied by Swarup 2004 (16) . The sequences and locations of all the shRNAs and siRNAs are provided in the table below. 1 Table 1 : Sequences of IMPDHl specific shRNAs and 2 siRNAs 3

4 5 6 Cell culture and transfection assays Human embryonic kidney (HEK293T) and Hela cells (American Type Culture Collection) were cultured in DMEM (GIBCO/BRL) supplemented with 10% fetal calf serum (FCS) and streptomycin/penicillin in a 5% CO2 incubator at 37°C. For the transfection of shRNAs, HEK293T cells were transfected with a total of lug of plasmid DNA per well in 24-well plates at 50% confluency. Similarly at the time of transfection with chemically synthesised siRNAs, HEK293T cells were co-transfeeted with lug of plasmid DNA expressing wild type IMPDHl and 20pmol of siRNA per well in 2 -well plates at 50% confluency. For the inhibition of IMPDHl protein, HEK293T and Hela cells were plated onto cover slips coated with poly-L- lysine and were co-tranfected with 20pmol of siRNAs and lug of plasmids expressing wild type or mutant form of IMPDHl tagged with enhanced green fluorescent protein (EGFP) per well in 24-well plates. All the above transfections were carried out with Lipofectamine 2000 (Invitrogen) according to manufacturers instructions.

Determination of transfection efficiency HEK293T cells were transfected with 0.5ug-5ug of pCMVβ mammalian reporter vectors (GIBCOBRL) per well in 24-well plates. After 24 hours of transfection, cells were fixed with 25% glutaldehyde and stained with X-gal and β-gal staining solution at a final concentration of lmg/ml (PBS at pH7.2, 5mM potassium ferricyanide, 5mM potassium ferrocyanide, 2mM MgCl_2/ 0.02% NP-40 and 0.01% SDS) . Cells were then incubated for 24hr in a 37°C incubator, and the number of blue cells was counted under a light microscope to determine the transfection efficiency.

RT-PCR analysis Total RNA extraction was carried out using TRIZOL (MRC) according to the manufacturers protocol. In the case of co-tansfection, total RNA was purified with Dnasel (Promega) according to manufacturer's instructions. The level of endogenous IMPDHl in HEK293T cells was measured by SYBR Green real-time PCR comparative quantification technique, using endogenous IMPDH2 and β-actin as reference templates for standards. The level of down-regulation of IMPDHl was measured against a non-targeting shRNA (shMR3) using SYBR Green real-time PCR (ROCHE) relative quantification technique, standardised by the corresponding transcription binding protein (TBP) and β-actin endogenous controls. All the values of IMPDHl expression in different transfected HEK293T cells were analysed by pooled t-Test of ul- u2. P values <0.05 were considered significant. The primers used in the amplification of IMPDHl, IMPDH2 , TBP and β-actin (Sigma-Genosys) are listed below. Table 2 : Forward and reverse primers used for RT-PCR analysis .

Fluorescence analyses

The expression of EGFP in transfectants was examined by Zeiss epifluorescence fluorescent microscope after 72hrs of siRNA transfection. HEK293T cells were washed twice with PBS for 5 mins and then digested with IX trypsin. Trypsinised cells were then re-plated onto cover slips coated with poly-L- lysine, and were allowed to re-attach for 30mins. The cells were then washed twice with PBS and fixed with 4% paraformaldehyde for 5 mins. After fixing, cells were stained with DAPI solution (1:5000) for 5mins and the cover slips were transferred onto glass slides. Poly-aquamount solution was applied onto the cells and fresh coverslips was placed on top. The glass slides were then wrapped in tin foil to minimise exposure to light. The percentage IMPDHl protein suppression was calculated by correlating the number of positive green fluorescing cells in the non-targeting siRNA controls to that of the IMPDHl targeting siRNAs .

SDS-PAGE and western blot analysis of protein extracts to test efficiency of siRNAs For western blot analysis, 2ug of plasmids expressing wild type or mutant IMPDHl and lOOpmol of sil224/ si2048 were co-transfected into Hela cells per 6-well plates. Transfection was carried out using LF2000 (Invitrogen) according to manufacturer's instruction. After 72hr of transfection, the cells were harvested by trypsinisation, washed twice with PBS and centrifuged gently to pellet. The pellet was then resuspended in 500μl of ice-cold protein lysis buffer solution (50mM Tris pH7.5, 150mM NaCl, 1% NP- 40, 0.5πιM EDTA, 0.1% SDS, 1 protease inhibitor tablet Roche) . The cells were incubated on ice for lOmins and were passed through 21G gauge needles to reduce the viscosity of the protein solution. The cells were finally re-pelleted by mild centrifuging for 20mins at 4°C, and the supernatant containing soluble proteins were stored at -20°C. Protein concentrations were determined using BioRad Protein assay lit with bovine IgG as standard.

Aliqouts containing equal amounts of total protein (40ug) were resolved on 10% SDS-polyacrylamide gels in duplicate. One gel was stained with Commassie blue and the second gel was transferred onto nitrocellulose membrane. Blots were blocked with 5% skimmed milk and then probed with a polyclonal rabbit anti-IMPDHl primary antibody (provided by Dr Sara J. Browne, Human genetics centre, school of public health, University of Texas HSC, Houston, TX 77030, USA) . Subsequent to washing, the blots were probed with a horseradish peroxidase-conjugated anti rabbit IgG secondary antibody (Sigma) . The peroxidase-based detection was analysed with Chemiluminescence Reagent (Pierce) according to manufacturer's instructions.

Organotypic retinal cultures (retinal explants), in vitro electroporation and immunostaining Electroporation, maintenance and dissociation of explant cultures were carried out according to protocols described by Matsuda and Cepko, 2004 (17) . Retinae from newborn C57 mouse pups were dissected and electroporated with plasmids expressing either wild type or mutant IMPDHl tagged with GFP and sh-non- silencing RNA constructs. A total of 4 retinae for each construct are currently being electroporated in two independent experiments and cultured in vitro for 14 days. Retinal explants will be either fixed in PBS containing 4% paraformaldehyde and vibratome- sectioned (50-100 μm) or dissociated by trypsin digest, followed by single cell attachment onto poly-L-lysine coated coverslips and fixation as above. Cells will be counterstained with DAPI and visualized using either a fluorescence microscope (Axiophot; Zeiss Ltd, Herts, UK) or a laser-scanning microscope (LSM-510; Zeiss Ltd) .

Injection of guanine nucleotides into IMPDH-/- Mice Eyes 3 μls of PBS solution containing 150 μg of GTP or XMP was injected intravitreally or subretinally into the of the right eye of IMPDH-/- mice. The left eye was always an un-injected control.

ERG analysis to test for the effects of injected guanine nucleotides 48 hours following the injection into the right eye, after overnight dark-adaptation, the mouse was light adapted for 10 minutes, then dark adapted for 30 minutes, after which the animal was anaesthetised with Ketamine and Xylazine and an electroretinogram was performed.

IMPDH

Extraction of tissue samples for HPLC analysis. Retinas were homogenized in 200μl of 0.5 M Percloric Acid (PCA) . Extract was stored at -80°C until neutralization. The extract was thawed and centrifuged at 13000 rpm for 15 minutes at 4°C to precipitate insoluble macromolecules . The supernatant was neutralized by 4/5 volume 0.5M KOH plus 1/5 volume IM KH₂P0 , pH 7.5 and placed on ice for 10-5 minutes. The potassium perchlorate precipitate was removed by centrifugation 13000 rpm for 15 minutes at 4°C. The neutralized supernatant was stored at -20°C or analysed immediately. HPLC analysis Standard solutions of HPLC-grade ATP, AMP, ADP, GTP adenosine, guanosine, GDP, inosine, IMP, uric acid, xa thine, NAD and hypoxanthine were prepared in 0.1 M KH₂P0₄ pH6.5 at a concentration of 1 mM. Solutions were filtered through 0.22 μm filters. Buffers used for HPLC analysis were as follows; Buffer A (0.1 M KH₂P0₄, 8mMTBAs, pH6.0) and Buffer B (0.1 M KH₂P0₄, 8mM TBAs, 30% (v/v) CH₃CN, pH 6.0. Both buffers were filtered through 0.45μm filters and helium degassed. The HPLC system was equilibrated with 50% buffer A and 50% buffer B for 30 minutes. Gradient Conditions were as follows: 100% of buffer A for 2 minutes followed by sample injection, 0-10% buffer B for 1.5 minutes, 10% buffer B for 2 minutes, 10-20% buffer B for 1 minute, 20-40% buffer B for 5 minutes, 40-100% buffer B for 3 minutes, 100% buffer B for 5 minutes, 100-0% buffer B for 0.1 minute and 100% buffer A for 3.9 minutes. lOOμl of sample was injected and the flow rate was set to 1.5 ml/ in.

RESULTS

FIGURE LEGENDS

Figure 1. In situ hybridisation studies on mouse retinal cryosections Retinal sections with antisense and sense DIG- labelled riboprobes for (a) Impdhl, (b) Impdh2 and (c) Hprt are shown. The Impdh2 in situs are performed on CDl mouse retinas, which lack the pigment normally found in the RPE, while the Impdhl and Hprt in situs are performed on wild type pigmented retinas . Strong expression of Impdhl is detected in the photoreceptor neurons . Impdh2 expression appears to be restricted to the RPE, with possibly very low level staining in the photoreceptor outer segments . Analysis of Hprt reveals expression in the inner nuclear and ganglion cell layers along with a lower level of expression in the photoreceptors . (RPE) retinal pigment epithelium, (OS) outer segments, (ONL) outer nuclear layer, (OPL) outer plexiform layer, (INL) inner nuclear layer, (IPL) inner plexiform layer, (GCL) ganglion cell layer. 200X Magnification.

Figure 2. ERG analysis in Impdhl^"7" mice (a) Rod-isolated responses, (b) light-adapted cone responses to single flash and (c) maximal, dark- adapted, combined rod and cone responses were recorded in a 12-month C57 wild-type animal, and Impdhl^{" "} animals at 6 weeks of age, 5 months, 8 months, 11 months and 13 months. At 11 months of age all responses showed a significant reduction in a- and b- wave amplitudes and, in the case of the cone responses, significant delay in b wave timing. Further deterioration was noted by 13 months. In the case of the maximal, dark-adapted, combined rod and cone responses progressive reduction in a-wave amplitude was noted from 5 months of age indicating disturbance of phototransduction. (d) ERG analysis of combined rod and cone response in a 17-year old affected patient harbouring an Arg224Pro IMPDHl mutation shows completely diminished responses.

Figure 3. Light micrographs of retinal sections from Impdhl^-7" mice. Sections shown are (a) 6 month Impdhl^"7-, (b) 10 month Impdhl^"7", (c) 10 month wild-type C57/BL6. The outer nuclear layer thickness of Impdhl^"7" retina is virtually indistinguishable to that of wild-type up to 10 months of age.

Figure 4. ERG analysis of delayed dark-adaption in Impdh-/- mice Impdhl^"7" mice were dark-adapted overnight and then light adapted to 10 minutes. They were subsequently placed in a dark room and rod-isolated responses then recorded in animals following increasing time periods of dark -adaption. A rod isolated response is not measurable following 30 and 60 mins of dark adaptation. At 90 minutes a very small response was visible which increased in amplitude as dark adaptation proceeded. It took 180 minutes of dark- adaptation for the maximal response to the -25dB flash to be observed indicating that lack of retinal IMPDHl significantly impairs dark-adaptation.

Figure 5. Analysis of wild-type and mutant IMPDHl proteins expressed in HEK293T cells (a) SDS-PAGE with Coomassie Blue staining of protein extracts from cells transfected with pcDNA3.1/His constructs containing wild-type (WT) , Arg224Pro and Asp226Asn mutant IMPDHl sequences, and pcDNA vector alone (negative control) . Transfected cells were separated into whole cell extracts (WCE) , soluble cytosolic fraction, nuclear fraction and final pelleted fraction. M = Molecular mass marker, (b) Western blot of an identical gel with protein transfer onto nitrocellulose membrane and probed with Anti-His antibody and HRP-conjugated secondary antibody. M = 50kDa His-tag protein (positive control) .

Figure 6. Fluorescent analysis of Hela cells expressing IMPDHl-GFP fusion proteins The subcellular localisation of EGFP-IMPDHl proteins is shown for cells transfected with GFP construct only, GFP fused to Wild-type (WT) IMPDHl and GFP fused to two mutant forms of IMPDHl, containing the Arg224Pro and Asp226Asn substitutions. Cells were examined 24 and 48 hours post transfection. Arrows indicate aggresome formation.

Figure 7. Computer modelling of wild-type and mutant IMPDHl proteins Endpoint comparison of molecular dynamics simulation run solutions for wild-type (blue) and mutant (pink) homology models of human IMPDHl. (b) Model of IMPDHl mutant product (pink) exhibits significant predicted structural perturbation in region of Arg to Pro point mutation when compared to model of wild-type protein (blue) . The Arg to Pro mutation in the 'bud' or flanking region is highlighted in green in both models. Fig 8. Schematic Illustration of IMPDHl Arg224Pro transgene An IMPDHl transgene was generated by linking the mouse Impdhl promoter to the human coding sequence with an incorporated Arg224Pro substitution. This construct is being injected into fertilised mouse ova to provide a model of the human RP10 form of RP.

Figure 9. Down-regulation of IMPDHl by shRNAs in HEK293T cells HEK293T cells (24-well plates) were transfected with vectors (lug per well) expressing 19mer shRNAs for 24hr. The shRNAs are listed at the bottom of each bar. A control set was transfected with shRNAs directed at the rhodopsin gene (right bars) , and this was standardised to show 100% IMPDHl expression. The average of three independent experiments is shown; error bars indicate standard error.

Figure 10. Down-regulation of IMPDHl by shRNA combinations in HEK293T cells HEK293T cells were co-transfected with a combination of specific shRNAs and the % IMPDHl expression was measured by real-time PCR relative quantification method. ShRNA combination A included shRNA 1, 3, 4, 5. shRNA combination B included shRNA 1, 3, 4. shRNA combination C included shRNA 1 and 5. shRNA 2 was left out of this study because it showed poor suppression potential in the previous study. A control group was transfected with shRNAs directed at the rhodopsin gene (right bars) , and was standardised to show 100% IMPDHl expression. The average of two independent experiments is shown; error bars indicate standard error.

Figure 11. Concentration-dependent down-regulation of IMPDHl expression by shRNA. (A) HEK293T cells were transfected with different concentrations of pSuper vectors for 24hr. A control panel was transfected with the same corresponding concentrations of shRNAs directed at rhodopsin gene (right bars) . (B) Transfection efficiency was determined by lac-z staining of HEK293T. (C) HEK 293T cells showing Lac-z staining following transfection with pCMVβ. Transfected cells appear blue, whereas untransfected cells remain white. Photograph taken at 40X.

Figure 12. Down-regulation of IMPDHl by siRNAs in HEK293T cells HEK293T cells (24-well plates) were co-transfected with chemically synthesised siRNAs (20pmol per well) and plasmid expressing wild type IMPDHl (lμg per well) for 24hr. The percentage IMPDHl expression was measured by real time PCR relative quantification method. A control set was transfected with siMR3 directed at the rhodopsin gene (left bars) , and this was standardised to show 100% IMPDHl expression. The average of four independent experiments is shown; error bars indicate standard error. Figurel3. Epifluorescent microscopic pictures of HEK293T cells after 72hr of co-transfection with sil224 or si2048 and wild type or mutant EGFP-IMPDHl (A) HEK293T cells transfected with 20pmol of non- targeting siRNAs (siMR3) or sil224 or si2048 as a negative control for fluorescent signal. (B) HEK293T cells transfected with lug of EGFP expressing vector (pcDNA3.1-EGFP) and 20pmol of non-targeting siRNA (siMR3) and si2048 as a positive control. (C) HEK293T cells separately co-transfected with lug of wild type IMPDHl-EGFP expressing vector ( wtIMPDHl- gfp) and 20pmol of non-targeting siRNA (siMR3) or sil224/ si2048. (D) HEK293T cells separately co- transfected with lug of mutant IMPDHl-EGFP expressing vector (pmutlMPDHl-gfp) and 20pmol of siMR3 or sil224/ si2048. Lower panels in (C) and (D) illustrates DAPI staining of living cells.

Figure 14. Quantification of IMPDHl protein expression in HEK293T cells after 72hrs of co- transfection with pw IMPDHl-gfp/ pmutIMPDHl-gfp and sil224/ si2048. Green fluorescence was normalised to the control level, left bars (100%) . Data are averages of three independent experiments. Errors bars denote SDs (p<0.01) .

Figure 15. Western blot analysis of IMPDHl protein expression in Hela cells (A) SDS-PAGE with Coomassie Blue staining show equal loading of protein extracts from cells transfected with pwtlMPDHl (lane 3, positive control), pmutlMPDHl (lane 4, positive control), pwtIMPDHl+siMR3 (lane 5, positive control) and pwtIMPDHl+sil224 or si2048 (lanes 6 and 7) . (B) Western blot of an identical gel with protein transfer onto nitorcellulose membrane and IMPDHl protein expression was detected with polyclonal rabbit anti-IMPDHl primary antibody.

Figure 16. ERG analysis of Impdh-/- mice following intravitreal injection of GTP ERG analysis of rod-isolated responses was performed on the Impdh-/- mouse 48 hours after intravitreal injection of GTP and following 30 minutes of dark- adaption. As expected in the case of the Impdh-/- mice the right uninjected control eye shows no recordable rod response following the short dark- adaption phase. The left, injected eye however, generated a visible and measurable rod response. This may indicate that the injected GTP is providing the photoreceptors with the guanine nucleotides required for the recycling of light-bleached rhodopsin protein.

Figure 17. HPLC analysis of retinal purine nucleotides in Impdh-/- mice following subretinal injection of XMP HPLC analysis indicates that levels of GTP within the retinas of IMPDHl-/- mice are lower than in normal mice as assayed by HPLC analysis (A and B) . There is an increase in the peaks representing ADP, ATP and GTP nucleotides in the injected eye compared to the uninj ected control eye of the same animal (B and C) .

IMPDH and HPRT expression in the murine retina In order to assess levels of IMPDHl, IMPDH2 and HPRT in retinal tissues, in situ hybridisations were carried out using riboprobes specific for transcripts derived from these genes on mouse retinal cryosections . EST data obtained for the IMPDH2 gene (UniGene database; "http: //www.ncbi .nlm.nih.gov/entrez/query. fcgi?db=un igene ) indicate that this isoenzyme is expressed in retinal pigment epithelium (RPE) libraries, hence IMPDH2 in situ hybridisations were performed on retinal sections from CDl mice, which do not possess pigment in the RPE. The results of this analysis indicate that IMPDHl is preferentially and strongly expressed within photoreceptor cells (Fig. la) . Conversely, little or no IMPDH2 expression is detectable in photoreceptors and only low levels of HPRT transcript are observed (Fig. lb,c). These results thus indicate that IMPDHl is largely responsible for the production of guanine nucleotides within photoreceptor cells. In mice lacking the IMPDHl gene, it has previously been reported that there is no elevation in expression levels of IMPDH type 2 or HPRT transcripts detectable in response to a deficiency of IMPDH type 1, in a number of tissues examined (18) . It is very likely therefore, that in the case of a defective or absent IMPDHl protein in the retina, IMPDH2 and HPRT may not be able to substitute for the loss of IMPDHl activity. This is particularly relevant given that the retina is one of the most energy-demanding tissues of the body and that photoreceptor cells have a high requirement for GTP within the visual transduction cycle (19) .

Retinal structure and function in IMPDHl-/- mice A rational interpretation of the above data is that IMPDHl is important to normal photoreceptor function. Hence, we evaluated photoreceptor activity in the retinas of IMPDHl-/- mice using electro- retinography (ERG) . Rod-isolated responses (Fig. 2a) , light-adapted cone responses to single flash (Fig. 2b) and maximal, dark-adapted, combined rod and cone responses (Fig. 2c) were recorded at 6 weeks of age, 5 months, 8 months, 11 months and 13 months. At 11 months of age all responses showed a significant reduction in a- and b-wave amplitudes (Fig. 2a-c) and, in the case of the cone responses, significant delay in b-wave timing (Fig. 2b) . Further deterioration was noted by 13 months. In the case of the maximal, dark-adapted, combined rod and cone responses, progressive reduction in wave amplitudes was noted from 5 months of age indicating disturbance of phototransduction. This may be explained by disturbance in cGMP-dependent cation channel function which requires a constant turnover of cGMP, which must be synthesized from GTP. A reduction in ERG amplitude, brought about by membrane hyperpolarisation and closing of these channels, may be due to the disturbance in guanine biosynthesis consequent to lack of IMPDHl activity. It is notable that the electroretinographic disturbances observed in IMPDHl-/- mice appear to be much milder than those recorded in humans with dominant IMPDHl mutations, where generally all ERG responses are unrecordable by the end of the second decade. The response in an affected 17-year old male from a large RP10 family (8,10) to the maximal intensity flash presented in the dark adapted state, which normally elicits a mixed rod/cone response, indicates that there is no recordable photoreceptor activity (Fig. 2d) . The IMPDHl-/- mouse model at 4 months of age, which we estimate to be approximately equivalent to a teenage human subject in terms of relative lifespan, displays no detectable structural or functional degeneration of the retina. In addition, photoreceptor dysfunction in the IMPDHl-/- mouse is not the result of loss of photoreceptor cell viability, since the outer nuclear layer thickness of the retinas in these mice is similar to wild type retinas up to 10 months of age (Fig. 3a- c) , and even by 13 months of age there is only marginal loss of nuclei from the outer nuclear layer of the retina (data not shown) . Preliminary evidence suggests some degree of disorganization of photoreceptor outer segments in older animals which is akin to a mild RP-like retinopathy. Thus, IMPDHl does not appear to be essential for normal retinal development or early visual function. In the light of such observations, it appears unlikely that the autosomal dominant segregation pattern which is characteristic of the RP10 form of retinitis pigmentosa, is caused as a result of haploinsufficiency for the normal IMPDHl gene product. Rather, disease pathology is probably caused by a dominant-negative phenotypic effect exerted by mutant protein.

Impaired Dark Adaptation in IMPDH-/- Mice The IMPDH-/- mice showed no measurable rod-dominated response to the -25dB flash after 30 minutes or 60 minutes (Fig. 4) . At 90 minutes a very small response was visible which increased in amplitude as dark adaptation proceeded. It took 180 minutes of dark-adaptation for the maximal response to the - 25dB flash to be observed indicating that lack of retinal IMPDH 1 significantly impairs dark- adaptation. After 30 minutes of dark-adaptation the rod-dominated response to the -25dB light flash was very similar to that recorded after the base-line overnight dark-adaptation.

Functional analysis and mammalian expression of wild-type and mutant IMPDHl proteins. Two amino acid substitutions, Arg224Pro and Asp226Asn, originally identified in RP10 families (10,20), lie in the second CBS domain that forms part of a smaller flanking subdomain of the IMPDH enzyme (21) . This region lies adjacent to the catalytic domain and is not thought to be required for activity (22, 23). In order to assess possible effects of these two established IMPDHl mutations on the functional activity of the expressed enzyme, the wild-type and two mutant human IMPDHl proteins were expressed in a bacterial expression system with incorporation of a His-tag for purification purposes. Protein fractions were purified from soluble cell lysates and enzymatic activity was determined by measuring the absorbance increase caused by reduction of NAD+ (24) . Specific enzyme activities for wild-type, Arg224Pro and Asp226Asn IMPDHl enzymes were averaged at 1.22, 1.30 and 1.28 (mol min-lmg-1 respectively. Enzyme activities were measured on soluble protein only and all three enzymes showed similar levels of activity for up to seven days. Therefore, no significant differences in specific enzyme activity were detectable between the wild-type and mutant IMPDHl proteins . Calculated activities were also comparable to previously published reports of IMPDHl specific enzyme activity levels (11) . A dramatic reduction in enzyme activity would be expected if the disease-causing effects of the mutations were primarily attributable to diminished activity under physiological conditions. Mutant IMPDHl proteins were found to exhibit marked decreased solubility in comparison to wild-type when expressed at a temperature of 37JC in bacterial cells (data not shown) . In order to investigate this observation further, human embryonic kidney (HEK) 293T cells were transfected with mammalian expression constructs containing wild-type, Arg224Pro or Asp226Asn mutant IMPDHl sequences, with incorporation of a His-tag for protein detection purposes. After harvesting, cells were separated into total cell protein, soluble cytosolic, nuclear and insoluble pelleted fractions, and were subsequently analysed by SDS-PAGE electrophoresis (Fig. 5a) and western blotting (Fig. 5b) . Wild-type and mutant IMPDHl proteins were expressed at similarly high levels in whole cell extracts with a band clearly visible at 55kDa (Fig. 5a) . No protein band of this size was evident in cells transfected with a construct that did not contain an IMPDHl cDNA insert. Following analysis of cellular fractions obtained after further purification steps, wild-type IMPDHl protein was shown to localise in the soluble cytosolic fraction, while no mutant protein was detectable in this fraction. Both mutant proteins exhibited considerably decreased solubility and as a result almost all of the mutant protein localised in the final pellet, which was re-suspendable in strong denaturant. Accordingly, we hypothesise that such mutant IMPDHl proteins may be mis-folded, resulting in the formation of insoluble aggregates within the cell cytosol.

Mammalian expression of wild-type and mutant GFP- tagged IMPDHl proteins. Wild-type and mutant (Arg225Pro and Asp226Asn) GFP- IMPDHl fusion proteins were expressed in Hela cells and subcellular localization of the protein was determined using fluorescent microscopy. Cells were examined to look for evidence of protein aggregation brought about by misfolded mutant forms of the IMPDHl protein. All three GFP-tagged proteins showed GFP distribution throughout the cytosol of the cell when expressed, with little expression in the nucleus. There was also evidence of aggregated protein formation in the cytosol of cells transfected with both WT and mutant IMPDHl proteins. Greater levels of protein aggregation were observed however following expression of mutant IMPDHl proteins (Fig. 6) . In contrast, cells transfected with GFP construct alone showed GFP protein distribution throughout the cell cytosol and nuclei showing therefore that IMPDHl is normally expressed in the cytosol .

Computer protein modelling studies on mutant IMPDHl To further investigate the existence of possible tertiary structural perturbation in mutant IMPDHl we undertook a series of molecular modelling computational simulations for the Arg224Pro mutation. Proline residues located within (-helices have a tendency to distort the standard helical arrangement by causing the structure to kink about the position of the proline residue (25) . To test the hypothesis that a mutation from Arg to Pro could result in a mutant protein with perturbed structure correlating to impaired in vitro and in vivo function, structural homology models were constructed from both Arg-wild-type and Pro-mutant IMPDHl sequences, building from available crystal structures of human (21) and non-human (Cricetulus griseus). (22) IMPDH which exhibited high homology to our sequences of interest. The resultant wild-type and mutant structures were subjected to an extensive computational molecular dynamics simulation to challenge and refine the models (26) . We observed no significant deviation in structure for the wild-type protein post-dynamics (not shown) , but significant structural perturbation was noted in the mutant structure around the area of point mutation (Arg224 to Pro224) by the end of the simulation run. Figure 7a and Figure 7b illustrate the end-point results from molecular dynamics simulation on mutant (pink) and wild-type (blue) IMPDHl homology models. The Arg to Pro mutation in the 'bud' or flanking region is highlighted in green in both models . The perturbation of predicted tertiary structure is clearly visible as the mutated protein loses cohesion of structure in the 'bud' domain, distant from the principal catalytic domain. Such structural perturbation would undoubtedly disrupt the formation of the biological homotetramer subunit and correlates to the observed physical and functional data for the mutant protein.

Transgenic mouse

An IMPDHl transgene has been generated which carries the Arg224Pro substitution, previously identified as the disease-causing mutation in a large Spanish family with adRP. This transgene contains a mutated version of the human coding sequence of the gene, linked to the mouse IMPDHl promoter sequence. Control of expression of IMPDHl appears to be very complex, with several promoter regions which may be responsible for its particular pattern of expression in the retina and other tissues. The upstream region of the gene also possesses a number of non-coding exons, the role of which is unknown. For these reasons it was necessary to include over 5kb of promoter, in the hope of mimicking the natural expression characteristics of the native gene, which may be critical to its functioning (Fig. 8) . This transgene is currently being injected into the nuclei of fertilised mice ova to create a transgenic mouse as a model for RP10.

Suppression of IMPDHl by siRNAs siRNA suppression technology has been established as a standard method for studying gene function in mammalian cells. Although, in-vi tro synthesised siRNA is commonly used to elicit specific gene suppression in higher eukaryotes, the suppression effect is not only transient but also expensive. In comparison, the use of mammalian expression vectors to direct the synthesis of shRNAs is more cost- efficient and can result in long-term stable down- regulation of mRNA.

Initially, a high level of endogenous IMPDHl, IMPDH2 and β-actin RNA transcripts were detected in HEK 293T, using SYBR Green real-time PCR comparative quantification technique. Following this, IMPDHl expressing HEK293T cells were separately transfected with lug of each of the IMPDHl targeting shRNAs and the control (shMR3-Rhodopsin targeting) shRNA. Percentage transfection efficiency ranged from 75- 80%. The degree of down-regulation on IMPDHl expression was measured by real-time PCR relative quantification method with IMPDH2 or b-actin acting as the housekeeping gene. Figure 9 shows that all five shRNAs showed down- regulation of IMPDHl expression ranging from 16% to 51%. Various possibilities may account for the incomplete suppression of IMPDHl expression. First, the selection and accessibility of target sites in the substrate contributes significantly to the effectiveness of shRNA. However, currently there are no reliable methods for predicting ideal target sequences for siRNA. Secondly, a closer examination of the construction of the nine-nucleotide spacer arm linking the inverted repeats as constructed according to www. oligoengine . com revealed that the sequences were not complementary to each other, possibly creating a secondary loop structure which might have affect on the transcription process . Brummelkamp et al., (39) have demonstrated the importance of the size and nucleotide sequence of the loop on shRNA activity. The data above suggest that any of the 5 shRNAs transfected singly are only capable of inflicting partial inhibition on IMPDHl expression. In an attempt to increase the degree of suppression, two or more specific shRNAs targeting different regions of IMPDHl mRNA were co-transfected into HEK293T cells. With reference to Figure 10, higher inhibition was observed when cells were transfected with a combination of shRNAs. Up to 64% inhibition of IMPDHl expression was observed when cells were co-transfeeted by a combination of shRNA 1057 and 1969. The increase in inhibitory activity was not due to an increase of the concentration of shRNAs used because the total concentration was maintained at lug of RNA per well. In addition, the use of three or four shRNAs in combination yielded higher inhibitory activities than that of single shRNAs. Enhanced gene silencing with multiple shRNAs could possibly be due to a collaborative inhibition effect, whereby two or more shRNAs may increase the accessibility of the target mRNA for degradation.

A titration experiment was carried out to evaluate whether shRNA-mediated down-regulation of IMPDHl expression was dose-dependent, and whether a high concentration of plasmids would inflict cytotoxic effects on cells upon transfection. shRNA 2067 was transfected into HEK293T cells at concentrations ranging from 0.5ug to 5ug. With reference to figure 11, optimal IMPDHl inhibitory activity was achieved when cells were transfected with lug of plasmids. Further increasing the concentration of plasmids did not increase the suppression potential, indicating that maximal shRNA-mediated suppression had been achieved. In most studies, the maximum amount of plasmids used for transfection does not exceed l-2ug (24-well plate format) , as higher dosages might impose cytotoxic effects on cells. In this experiment, the percentage suppression from 5ug of plasmids (16%) was lower than that of lug of plasmids (61%) , indicating an inhibitory or toxicity effect at high plasmid concentration. Figure 11 illustrates that the calculated transfection. efficiency appears consistent for different concentrations of vectors transfected. As a result, there was no direct correlation observed between the concentration of shRNA-expressing vectors used and the number of cells transfected. For example, the transfection efficiencies for lug and 5ug of plasmids are similar (79% vs. 78%), but their gene silencing activity differed greatly (61% vs. 16%).

The subtle silencing effect elicited by shRNAs on IMPDHl have been reported by previously published studies, which illustrated that siRNAs targeting distinctive sites on the same mRNA elicited different silencing efficiency, and that silencing efficacy was highly dependent on the accessibility of target sites (27). Therefore two additional siRNAs, sil224 and si2048 (synthesised and purified by Dharmacon Inc) targeting different regions of both wild type and mutant form of IMPDHl were designed according to the eight criterion for RNAi selection described by Swarup (16) . HEK293T cells were separately co-transfected with 20pmol of either sil224 or si2048 and lug of plasmid expressing wild- type IMPDHl. The silencing effect on IMPDHl was measured by RT-PCR and standardised by the corresponding internal control β-actin. Transfection of sil224 and si2048 resulted in a potent reduction of greater than 70% of IMPDHl mRNA (fig. 12) . These data confirmed that both IMPDHl specific chemically synthesised siRNAs were able to knockdown IMPDHl expression by great extent.

By applying a rapid assay based on the use of GFP signals as an indicator of the degree of gene silencing, the potency of the two IMPDHl specific siRNAs were further examined. In the positive controls, lug of plasmids expressing wild type IMPDHl or mutant IMPDHl tagged with GFP was transfected into HeLa cells in combination with a non-targeting siRNA (siMR3), and similar level of GFP expression were observed in both cases (Fig. 13 C and D) . But when cells were treated with IMPDHl- specific siRNAs, the level of GFP expression was significantly reduced (Fig 13 C and D) . Since the level of GFP expression was directly correlated to the level of IMPDHl expression, percentage suppression was quantified by standardising the number of fluorescing cells in positive controls to show 100% IMPDHl expression. Figure 14 illustrates that cells transfected with IMPDHl specific siRNAs showed that up to 73% suppression of wild type IMPDHl and 87% suppression of mutant IMPDHl expression were achieved. These data provided a rapid, convenient and semi-quantitative readout for the efficacy of the two IMPDHl specific siRNAs .Western blot analysis was used to qualitate the silencing efficiency of sil224 and si2048 at protein level. Cells that have been transfected with either wild-type IMPDHl, mutant IMPDHl (Arg224Pro) or co-transfected with wild type IMPDHl and non- targeting siRNA (siMR3) showed normal IMPDHl gene expression (lanes 3,4, 5, Fig. 15B) . But when cells were treated with IMPDHl specific siRNAs, IMPDHl protein expression was not visible (lanes 6,7, Fig 15B) , whereas IMPDHl protein was not detectable in Hela cells (lane 2, Fig 15B) . Therefore the preliminary data shown here indicated efficient suppression of IMPDHl expression with siRNAs. The two functional siRNAs (sil224 and si2048) have been incorporated into DNA based RNA interference vectors for a more stable and long-term suppression effect. Currently the knock-down effect of IMPDHl expression in the mouse retina are being examined by the introduction of IMPDHl specific shRNAs and plasmids expressing either wild type or mutant IMPDHl tagged with GFP into mouse retinal explants via in-vi tro electroporation. The two functional siRNAs (sil224 and si2048) have been incorporated into DNA based RNA interference vectors for a more stable and long-term suppression effect. Currently the knock-down effect of IMPDHl expression in the mouse retina are being examined by the introduction of IMPDHl specific shRNAs and plasmids expressing either wild type or mutant IMPDHl tagged with GFP into mouse retinal explants via in-vi tro electroporation.

The effect of Intravitreal Injection of GTP into IMPDH-/- Mouse Eye Following intravitreal injection of GTP into the eyes of IMPDHl-/- mice a visibly greater electroretinographic response was seen in the injected right eye compared to the un-injected left eye (Fig. 16) . IMPDH-/- mice whose eyes were not injected with a GTP solution and who were subjected to an electroretinogram under the same conditions show no visible response to the same electroretinographic stimulus after 30 minutes of dark-adaptation. A supplemental therapy for RP10 based upon delivery to retinal tissues of such compounds is thus suggested.

Analysis of nucleotides in wild type and IMPDH1-/- mice, pre- and post-injection of XMP Levels of guanine nucleotides and related compounds in the retinas of wild type and IMPDHl-/- mice are shown in figure 17. It will be noted from this that levels of GTP within the retinas of IMPDHl-/- mice are lower than in normal mice as assayed by HPLC analysis (fig 17 A and B) . Analysis of the injected IMPDH-1/- eye indicates that the peaks representing ADP, ATP and GTP are larger in size than in the uninjected eye (fig 17 C) . Increased GTP levels suggests that the XMP is not only entering the retinal cells but is being converted into GTP which is required by the photoreceptors for number of functions. The presence of extra ATP might be explained by the fact that GTP is required for the production of ATP. Other concentrations and/or volumes of the therapeutic agent and/or other methods of delivery could, in principle, be used.

DISCUSSION

The RP10 gene was initially localized on chromosome 7q as a result of a systematic genetic linkage study mounted on DNA from a large Spanish family (8) . It is estimated, based on the number of pedigrees so far identified with the RP10 form of retinopathy, that this genetic sub-type of disease accounts for 3-5% of autosomal dominant cases of RP (28) . The in situ hybridisation data presented here clearly demonstrate high levels of IMPDHl expression within the photoreceptor cell layer of mouse retinas, in comparison to levels of transcript derived from the IMPDH2 and HPRT genes. The retina has the highest metabolic rate of any tissue of the body, photoreceptors having a particularly high requirement for GTP in visual transduction processes • (19) . Therefore, it is suggested that depletion of the GTP pool over time is the most likely explanation for the reduction in ERG amplitudes seen in the retinas of older IMPDHl-/- mice and is the cause of the mild retinopathy that such animals develop. In addition, the human and mouse IMPDHl sequences share 98% identity at the protein level which strongly suggests that IMPDHl function in the retina is highly conserved amongst species. The IMPDHl-/- mouse represents the first reported model carrying a targeted deletion of a ubiquitously- expressed dominant RP-associated gene and has been shown to display no detectable structural or functional degeneration of the retina up to 5 months of age. In contrast, previously reported murine models in which there is a targeted deletion of a dominant RP-associated gene, including the rhodopsin, rds-peripherin, Rpl and Crx knockout mouse models, have resulted in phenotypes in which there is significant degeneration of photoreceptor cells and in many cases, completely diminished ERG amplitudes by 5 months of age (29-332) . ERG analysis of the IMPDH-/- mice following dark adaptation indicate that they have significantly impaired dark adaptation kinetics. This, taken together with the very slow rate of retinal degeneration seen in these animals suggests that they have a disease akin to a mild retinal disorder seen in humans, commonly referred to as night-blindess . While it will not be possible to directly compare the molecular pathologies associated with autosomal dominant and recessive (represented by the IMPDHl-/- null mouse) forms of IMPDHl-based disease, unless human subjects with the latter form of disease are identified, the data presented nevertheless strongly support the concept that the retinopathy in human subjects is very much more severe than the disease state which has been shown to occur as a result of a targeted disruption of the IMPDHl gene in mice. The studies presented on mutant IMPDHl proteins show that there are no detectable differences in enzyme activity when compared to wild-type, but that a substantial difference in solubility between wild- type and mutant proteins is observable in both bacterial and mammalian cells. Such observations, along with distinct perturbations in protein structure predicted by computer modelling of mutant IMPDHl structure, point to a probable dominant- negative molecular pathology for RP10, as a result of alterations in the folding and solubility of the mutant protein. Following expression of GFP-IMPDHl fusion proteins in cells, there is also evidence to suggest that the mutant forms of protein may be prone to aggregation due to misfolding of the proteins . While over expression of wild-type GFP- IMPDHl proteins appear also to cause protein aggregation, this effect appears to be more pronounced when the mutant forms of protein are expressed. A higher percentage of cells containing mutant IMPDHl protein, compared to wild type, show evidence of aggresome formation in the cell cytosol . It is of note that similar differences in solubility between wild-type and mutant proteins have been demonstrated in relation to the RP-associated splicing factor PRPF31 (12) , and protein folding defects associated with rhodopsin mutations that lead to aggresome formation have also been described (33-35) . Hence, in RP10, mutant IMPDHl protein misfolding and subsequent aggregation may be a central event in the initiation of neuronal cell death, as it is in a range of other hereditary or multifactorial neurodegenerative diseases including Alzheimer disease, Parkinson disease and amylotrophic lateral sclerosis (36-38) .

Notwithstanding recent successes in restoration of visual function or preservation of the retina in autosomal recessive forms of degenerative retinopathy, including the Briard dog model of Leber congential amaurosis (2) and the RCS rat following AAV-mediated delivery of therapeutic genes to retinal tissues (3), the extensive intragenic heterogeneity associated with autosomal dominant forms of retinal degeneration is a major impediment to the development of successful and economically viable therapeutics. For example, in excess of 200 different mutations have been encountered within the rhodopsin and RDS-peripherin genes alone (Retina International Database "http: //www. retina-international . org/sci- news/rhomut .htm"

) . While the IMPDHl gene was implicated in disease etiology only within the last two years, the number of mutations identified within the gene in patients with RP is already rising (41) . Hence suppression strategies targeting individual mutations are unlikely to be viable in economic terms . While a series of so-called mutation-independent suppression-replacement strategies has been described, such approaches involve two major steps, the simultaneous suppression of transcripts derived from both normal and mutant alleles, together with the introduction of a replacement gene, altered, for example at third-base degenerate sites, such as to escape suppression but nevertheless encoding functional protein (4-6) . Insights gained into the nature of the molecular pathology of retinal degeneration in RP10, combined with studies on the autosomal recessive form of disease in the IMPDHl-/- mouse, suggest that RP10 may be a particularly attractive potential therapeutic target for two reasons. Firstly, only a single suppression step, involving simultaneous silencing of both normal and mutant alleles of the IMPDHl gene would, in principle, be sufficient to remove the major dominant negative pathological effect of mutant IMPDHl protein, leaving a situation in which photoreceptor neurons continue to survive and remain sufficiently functional such as to provide useful vision, possibly well into adult life. Secondly, while gene silencing using RNA interference is now an established technique in mammalian systems (42), discrimination between normal and mutant transcripts at the level of single nucleotides has so far met with only partial success (43) . In the case of the therapeutic approach outlined here, such constraints do not apply, since there is not a requirement to target single base changes . Moreover, since a gene replacement step will not be essential there will be no requirement to efficiently control the expression of a replacement gene. In respect to gene silencing it will be noted that a series of analyses reported herein indicate highly efficient ablation of shRNA-mediated transcripts derived from the IMPDHl gene. It is of note that intraocular injection of free nucleotides (for example XMP or GTP) followed by electroporation into retina, appears to result in increased levels of such comounds in retinal tissues as assayed by HPLC chromatography and an improvement in ERG rod response when the latter was tested in such a mouse. These data strengthen the concept that supplementation in the retina by delivery of such therapeutic compounds may have a beneficial effect. In principal delivery by any means would be acceptable, for example, electorphoresis or iontophoresis or by direct application in eye drop form . Thus those elements required to implement a therapeutic strategy in terms of sim suppression of IMPDHl transcripts and supplemental therapy based in nucleotide supplementation have been put in place.The invention is not limited to the embodiments hereinbefore described which may be varied in construction and detail without departing from the spirit of the invention.

REFERENCES 1. 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. 2. Acland, G.M., Aguirre, G.D., Ray, J., Zhang, Q., Aleman, T.S., Cideciyan, A.V., Pearce- Kelling, S.E., Anand, V., Zeng, Y., Maguire, A.M. et al . (2001) Gene therapy restores vision in a canine model of childhood blindness. Nat. Genet., 28, 92-95. 3. Smith, A.J., Schlichtenbrede, F.C., Tschernutter, M. , Bainbridge, J.W., Thrasher, A.J. and Ali, R.R. (2003) AAV-Mediated gene transfer slows photoreceptor loss in the RCS rat model of retinitis pigmentosa. Mol. Ther., 8, 188-195. 4. Millington-Ward, S., O'Neill, B., Tuohy, G., Al-Jandal, N. , Kiang, A.S., Kenna, P.F., Palfi, A., Hayden, P., Mansergh, F., Kennan, A., Humphries, P. and Farrar, G.J. (1997) Strategems in vitro for gene therapies directed to dominant mutations. Hum. Mol. Genet., 6, 1415-1426. 5. O'Neill, B., Millington-Ward, S., O'Reilly, M. , Tuohy, G., Kiang, A.S., Kenna, P.F., Humphries, P. and Farrar, G.J. (2000) Ribozyme-based therapeutic approaches for autosomal dominant retinitis pigmentosa. Invest. Ophthalmol . Vis. Sci., 41, 2863-2869. 6. Millington-Ward, S., O'Neill, B., Kiang, A.S., Humphries, P., Kenna, P.F. and Farrar, G.J. (1999) A mutation-independent therapeutic strategem for osteogenesis imperfecta. Antisense Nucleic Acid Drug. Dev., 9, 537-542. 7. Millington-Ward, S., Allers, C, Tuohy, G. , Conget, P., Allen, D., McMahon, H.P., Kenna, P.F., Humphries, P. and Farrar, G.J. (2002) Validation in mesenchymal progenitor cells of a mutation-independent ex vivo approach to gene therapy for osteogenesis imperfecta. Hum. Mol. Genet., 15, 2201-6. 8. Jordan, S.A., Farrar, G.J., Kenna, P., Humphries, M.M. , Shells, D.M. , Kumar-Singh, R. , Sharp, E.M., Soriano, N. , Ayuso, C, Benitez, J. et al. (1993) Localization of an autosomal dominant retinitis pigmentosa gene to chromosome 7q. Nat. Genet., 4, 54-58. 9. Kennan A., Aherne A., Browne SJ., Daiger SP., Farrar GJ., Kenna PF. And Humphries P. (2003) On the role of IMPDHl in Retinal Degeneration. In LaVail, M.M. , Hollyfield, J.G. and Anderson, R.E. (eds) , Retinal Degenerations, Mechanisms and Experimental Therapy, Kluwer Academic, New York, pp. 1-12. 10. Kennan, A., Aherne, A., Palfi, A., Humphries, M. , McKee, A., Stitt, A., Simpson, D.A. , Demtroder, K., Orntoft, T., Ayuso, C. et al. (2002) Identification of an IMPDHl mutation in autosomal dominant retinitis pigmentosa (RP10) revealed following comparative microarray analysis of transcripts derived from retinas of wild-type and Rho(-/-) mice. Hum. Mol. Genet., 11, 547-557. 11. Hager, P.W., Collart, F.R., Huberman, E. and Mitchell, B.S. (1995) Recombinant human inosine monophosphate dehydrogenase type I and type II proteins. Purification and characterization of inhibitor binding. Biochem. Pharmacol., 49, 1323-1329. 12. Deery, E.C., Vithana, E.N., Newbold, R.J., Gallon, V.A. , Bhattacharya, S.S., Warren, M.J., Hunt, D.M. and Wilkie, S.E. (2002) Disease mechanism for retinitis pigmentosa (RPll) caused by mutations in the splicing factor gene PRPF31. Hum. Mol. Genet., 11, 3209-3219. 13. NosE, S. (1991) Constant temperature molecular-dynamics methods. Prog. Theoret. Phys. Supp, 103, 1-46. 14. Swope, W.C., Andersen, H.C., Berens, P.H. and Wilson, K.R. (1982) A computer simulation method for the calculation of equilibrium constants for the formation of physical clusters of molecules: Application to small water clusters. J. Chem. Phys . 76, 637-649. 15. Fletcher, R. (1980). Practical Methods of Optimization, Vol. 1, Unconstrained Optimization John Wiley & Sons U.S.A. 16. Swarup G. (2004) . How to design a highly effective siRNA. J Biosci. 29:129-31. 17. Matsuda T., Cepko C.L. (2004) Electroporation and RNA interference in the rodent retina in vivo and in vitro. Proc Natl Acad Sci U S A. 101:16-22. 18. Gu, J.J., Tolin, A.K. , Jain, J., Huang, H., Santiago, L. and Mitchell, B.S. (2003) Targeted Disruption of the Inosine 5'- Monophosphate Dehydrogenase Type I Gene in Mice. Mol. Cell Biol., 23, 6702-6712. 19. La Cour, M. (2002) The retinal pigment epithelium. In Kaufman, P.L., Aim, A. (eds) , Adleris Physiology of the Eye. 10th Ed. Mosby, St. Louis, pp. 348-357. 20. Bowne, S.J., Sullivan, L.S., Blanton, S.H., Cepko, C.L., Blackshaw, S., Birch, D.G. , Hughbanks-Wheaton, D., Heckenlively, J.R. and Daiger, S.P. (2002) Mutations in the inosine monophosphate dehydrogenase 1 gene (IMPDHl) cause the RP10 form of autosomal dominant retinitis pigmentosa. Hum. Mol. Genet., 11, 559-568. 21. Colby, T.D., Vanderveen, K. , Strickler, M.D., Markham, G.D. and Goldstein, B.M. (1999) Crystal structure of human type II inosine monophosphate dehydrogenase: implications for ligand binding and drug design. Proc. Natl. Acad. Sci. U S A, 96, 3531-3536. 22. Sintchak, M.D., Fleming, M.A. , Futer, 0., Raybuck, S.A., Chambers, S.P., Caron, P.R., Murcko, M.A. and Wilson, K.P. (1996) Structure and mechanism of inosine monophosphate dehydrogenase in complex with the immunosuppressant mycophenolic acid. Cell, 85, 921-930. 23. Sintchak, M.D. and Nimmesgern, E. (2000) The structure of inosine 5 ' -monophosphate dehydrogenase and the design of novel inhibitors. Immunopharmacology, 47, 163-184. 24. Carr, S.F., Papp, E., Wu, J.C. and Natsumeda, Y. (1993) Characterization of human type I and type II IMP dehydrogenases . J. Biol. Chem., 268, 27286-27290. 25. Chang, D.K., Cheng, S.F., Trivedi, V.D. and Lin, K.L. (1999) Proline affects oligomerization of a coiled coil by inducing a kink in a long helix. J. Struct. Biol., 128, 270-279. 26. Flohil, J.A., Vriend, G. and Berendsen, H.J. (2002) Completion and refinement of 3-D homology models with restricted molecular dynamics: application to targets 47, 58, and 111 in the CASP modeling competition and posterior analysis. Proteins, 48, 593-604. 27. Holen T, Amarzguioui M, Wiiger MT, Babaie E, Prydz H. (2002) Positional effects of short interfering RNAs targeting the human coagulation trigger Tissue Factor. Nucleic Acids Res . 30 : 1757-66 . 28. Daiger, S.P. Sullivan, L.S., Bowne, S.J., Kennan, A., Humphries P., Birch, D.G., Heckenlively, J.R. and the RPl consortium. (2003) Identification of the RPl and RP10 (IMPDHl) genes causing autosomal dominant RP. In LaVail, M.M., Hollyfield, J.G. and Anderson, R.E. (eds.), Retinal Degenerations, Kluwer Academic, New York, pp. 1-12. 29. Humphries, M.M. , Rancourt, D., Farrar, G.J., Kenna, P., Hazel, M. , Bush, R.A., Sieving, P.A., Sheils, D.M. , McNally, N. , Creighton, P. et al . (1997) Retinopathy induced in mice by targeted disruption of the rhodopsin gene. Nat. Genet., 15, 216-219. 30. Sanyal, S., De Ruiter, A. and Hawkins, R.K. (1980) Development and degeneration of retina in rds mutant mice: light microscopy. J. Comp. Neurol., 194, 193-207. 31. Gao, J., Cheon, K. , Nusinowitz, S., Liu, Q., Bei, D., Atkins, K. , Azi i , A., Daiger, S.P., Farber, D.B., Heckenlively, J.R. et al . (2002) Progressive photoreceptor degeneration, outer segment dysplasia, and rhodopsin mislocalization in mice with targeted disruption of the retinitis pigmentosa-1 (Rpl) gene. Proc. Natl. Acad. Sci. U S , 99, 5698- 5703. 32. Furukawa, T., Morrow, E.M., Li, T., Davis, F.C. and Cepko, C.L. (1999) Retinopathy and attenuated circadian entrainment in Crx- deficient mice. Nat. Genet., 23, 466-470. 33. Liu, X., Garriga, P. and Khorana, H.G. (1996) Structure and function in rhodopsin: correct folding and misfolding in two point mutants in the intradiscal domain of rhodopsin identified in retinitis pigmentosa. Proc. Natl. Acad. Sci. U S A, 93, 4554-4559. 34. Saliba, R.S., Munro, P.M., 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. 35. Illing, M.E., Rajan, R.S., Bence, N.F. and Kopito, R.R. (2002) A rhodopsin mutant linked to autosomal dominant retinitis pigmentosa is prone to aggregate and interacts with the ubiquitin proteasome system. J. Biol. Chem., 277, 34150-34160. 36. Sherman, M.Y. and Goldberg, A.L. (2001) Cellular defenses against unfolded proteins : a cell biologist thinks about neurodegenerative diseases. Neuron, 29, 15-32. 37. Kopito, R.R. (2000) Aggresomes, inclusion bodies and protein aggregation. Trends Cell. Biol., 10, 524-530. 38. Bence, N.F., Sampat, R.M. and Kopito, R.R. (2001) Impairment of the ubiquitin-proteasome system by protein aggregation. Science, 292, 1552-1555. 39. Brummelkamp R.T., Bernards, R. and Agami, R. (2002) A system for stable expression of short interfering RNAs in mammalian cells. Science, 296, 550-553. 40. Elbashir, M.S., Martinez, J. , Patkaniowska, A., Lendeckel, W. and Tuschl, T. (2001) Functional anatomy of siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate. EMBO J., 20, No. 23, 6877-6888. 41. Bowne, S.J., Gire, A., Sullivan, L.S., Heckenlively, J.R., Birch, D.G., Hughbanks- Wheaton, D., Lewis, R.A. and Daiger, S.P. (2003) Frequency and spectrum of IMPDHl mutations associated with autosomal dominant retinitis pigmentosa. Invest. Ophthalmol. Vis. Sci. 44, E-Abstract 2307. 42. McManus, M.T. and Sharp, P.A. (2002) Gene silencing in mammals by small interfering RNAs. Nat. Rev. Genet., 3, 737-747. 43. Miller, V.M. , Xia, H., Marrs, G.L., Gouvion, CM., Lee, G. , Davidson, B.L. and Paulson, H.L. (2003) Allele-specific silencing of dominant disease genes. Proc. Natl. Acad. Sci. U S , 100, 7195-7200.

Claims

Claims

1. A method of treating an autosomal dominant disease of the eye, which method comprises a step of suppressing substantially all alleles of a gene associated with the disease, wherein a step of introducing a replacement gene is disclaimed.

2. A medicament for the treatment of an autosomal dominant disease of the eye, the medicament comprising means for suppressing all alleles of a gene associated with the disease, wherein medicaments including a replacement gene are disclaimed.

3. A kit for use in the treatment of an autosomal dominant disease of the eye, the kit comprising means for suppressing all alleles of a gene associated with the disease, either alone or in a vector, wherein medicaments including a replacement gene are disclaimed.

4. Use of means for suppressing all alleles of a gene associated with a disease, without a replacement for the endogenous gene, alone or in a vector, in the manufacture of a medicament for the treatment of an autosomal dominant disease of the ey .

5. A medicament, use or kit as claimed in claims 2, 3 or 4 wherein the means for suppressing both alleles of a gene of interest comprises one or more suppression effectors.

6. A medicament, use or kit as claimed in claim 5 wherein the suppression effectors are capable of silencing or reducing expression of the gene and are chose from the group consisting of nucleic acids; peptide nucleic acids (PNA's); peptides; antibodies; ribozymes; hammerhead ribozymes; siRNA; and triple helix nucleotides .

7. A method, medicament, use or kit as claimed in any of the preceding claims wherein the disease is Retinitis pigmentosa or glancoma.

8. A method, medicament, use or kit as claimed in any of the preceding claims wherein the disease is RPlO-type Retinitis pigmentosa, and the gene to be suppressed is IMPDHl.

9. A method as claimed in claims 1, 7 or 8 wherein the suppression step is carried out in retinal tissue.

10. A method as claimed in claims 1, 7, 8 or 9 comprising a further step of treating retinal tissue with a metabolite of IMPDHl.

11. A method for the treatment of Retinitis pigmentosa, comprising a step of delivering a metabolite of IMPDHl to retinal tissue of an individual in need of treatment.

12. A method as claimed in claim 11 further comprising a step of delivering to the retinal tissue means for suppressing both a normal and diseased allele of IMPDHl.

13. A method as claimed in any of claims 1, 7, 8, 9, 10, 11 or 12 wherein the means of delivery is selected from the group comprising: sub-retinal delivery; intra-ocular delivery; delivery via the venous system; intraperitoneal injection; iontophoresis; microdialysis; oral injestion; and eye drops .

14. A medicament for the treatment of RPlO-type Retinitis pigmentosa, comprising a metabolite of IMPDHl such as XMP or GTP.

15. Use of a metabolite of IMPDHl in the manufacture of a medicament for the treatment of Retinitis pigmentosa.

16. A therapy for glaucoma caused by mutations within the myocilin gene, the therapy comprising means for substantial complete ablation of transcripts derived from the myocilin gene alleles.

17. A therapy as claimed in claim 16 wherein the means for causing substantially complete ablation acts at the level of DNA, transcript or protein.

18. Small Interference RNA (siRNA) Suppression Effectors for IMPDHl chosen from the following sequences

ShRNAl Forward strand 5 ' GATCCCCAAGCTGGTGGGCATCGTCATTCAAGAGATGACGATGCCCACCA GCTTTTTTTGGAAA3 '

ShRNAl Reverse strand 5 'AGCTTTTCCAAAAAAAGCTGGTGGGCATCGTCAAGAGAACTTTGACGATG CCCACCAGCTTGGG3 '

ShRNA2 Forward strand 5 'GATCCCCAAGTTTGAGAAGCGGACCATTCAAGAGATGGTCCGCTTCTCAA ACTTTTTTTGGAAA3 '

ShRNA2 Reverse 5 'AGCTTTTCCAAAAAAAGTTTGAGAAGCGGACCAAGAGAACTTTGGTCCGC TTCTCAAACTTGGG3 '

ShRNA3 Forward strand 5 ' GATCCCCGTTTGAGAAGCGGACCATGTTCAAGAGACATGGTCCGCTTCTC AAACTTTTTGGAAA3 '

ShRNA3 Reverse strand 5 'AGCTTTTCCAAAAAGTTTGAGAAGCGGACCATGAGAGAACTTCATGGTCC GCTTCTCAAACGGG3 '

ShRNA4 Forward strand 5 ' GATCCCCCTCACCTACAACGACTTCCTTCAAGAGAGGAAGTCGTTGTAGG TGAGTTTTTGGAAA3 ' ShRNA4 Reverse strand 5 'AGCTTTTCCAAAAACTCACCTACAACGACTTCCAGAGAACTTGGAAGTCG TTGTAGGTGAGGGG3 '

ShRNA5 Forward strand 5 'GATCCCCGTGCCCTACCTCATAGCAGTTCAAGAGACTGCTATGAGGTAGG GCACTTTTTGGAAA3 '

ShRNA5 Reverse strand 5 'AGCTTTTCCAAAAAGTGCCCTACCTCATAGCAGAGAGAACTTCTGCTATG AGGTAGGGCACGGG3 '

SiRNA1224 Forward strand 5 'GCUGCCUAUCGUCAAUGAU3 '

SiRNA1224 Reverse strand 5 'AUCAUUGACGAUAGGCAGC3 '

Si RNA 2048 Forward strand 5 'UGUACUCAGGAGAGCUCAA3

Si RNA 2048 Reverse strand 5 'UUGAGCUCUCCUGAGUACA3 '

ShMR3 Forward strand 5 'GCCTGAGGTCAACAACGAA3 '

SiMR3 Reverse strand 5 ' TTCGTTGTTGACCTCAGGC3 '