Development of retinal gene therapy to treat dominantly inherited disease using a novel RNA-based silencing system

Genetic diseases are now the most common cause of untreatable blindness in young people. Gene therapy is a method of treating a disease by manipulating the genetic code. Recently the first ever gene therapy was approved for use in the NHS and this was for a rare inherited form of blindness. The purpose of this research is to develop another genetic treatment, but this time for a more common cause of blindness.

Genetic diseases which are described as 'dominantly inherited' pass from one generation to the next and are usually caused by a defect on one gene that makes a toxic protein. Retinitis pigmentosa (RP) is an incurable cause of genetic blindness in young people and it is often dominantly inherited. It is most commonly caused by mutations in the rhodopsin (RHO) gene which codes for the light sensitive pigment in the retina. Patients with only one copy of the RHO gene can see perfectly well, but if the second copy has a mutation in it that makes abnormal rhodopsin protein then this will accumulate in the light sensing cells (photoreceptors) and cause them to degenerate. This is a slow process over several years, but eventually when all the photoreceptors have gone the affected patient becomes completely blind. Sadly they also pass on the genetic mutation to their children who have a 50% chance of going blind from inheriting the same mutation.

Our proposed research involves using established gene therapy techniques to take advantage of a naturally occurring cell pathway that is used to inactivate genes. When a gene is read, the DNA is converted into RNA and this RNA is then chopped up into smaller fragments that make the code for a particular protein - otherwise known as messenger RNA. There are however smaller RNA fragments known as microRNAs which can bind to the messenger RNA and inactivate it. These microRNA molecules regulate gene expression - they are made in the cell nucleus by a complicated process that involves folding them into a loop before they can bind to the messenger RNA. In 2007 however it was discovered in David Bartel's lab at the Massachusetts Institute of Technology that some genes release RNA fragments that can spontaneously form microRNA loops without the complicated processing. These microRNA molecules are known as 'mirtrons'. Our proposed research involves using an inactivated virus (known as a viral vector) to deliver microRNA molecules derived from mirtrons directly into the photoreceptor cells with the aim of inactivating the mutant rhodopsin. We have designed the viral vector to be similar to the one recently approved by NICE in England because we know it is safe and effective. We have put two mirtrons in the viral vector, together with an extra normal copy of the RHO gene which has been modified slightly so that the mirtrons cannot inactivate it. Hence when the viral vector is injected into the retina, the mutant RHO gene is suppressed and the normal copy is boosted. We tested this in a mouse in our laboratory that has the same RHO mutation as human patients and we could delay the mouse retinal degeneration at one of the doses we tested. This experiment represents the first time that mirtron gene therapy has been successfully applied in a living animal and we are extremely excited about it, because it has huge potential to treat patients with dominantly inherited eye disease (and probably other diseases outside the eye).

Although we have written up the results for publication, we are keen to develop this as a treatment for patients and this is why we have applied for MRC DPFS funding. We have only tested one viral vector and although it worked, we are aware that the genetic code in the vector could be improved substantially to give an even better effect. We need to test the vector in another mouse model of human RP and one that contains the entire human RHO gene so that we can measure the effects and work out exactly where the RHO gene should be targeted and how many mirtrons we need.

Inherited retinal degenerations such as retinitis pigmentosa (RP) are the most common cause of untreatable blindness in the young. Recently great progress has been made with gene replacement therapy using adeno-associated viral (AAV) vectors following FDA approval of Luxturna to treat Leber congenital amaurosis, but therapies for more common dominantly inherited conditions remain an unmet need. Dominant mutations in the rhodopsin gene are overall the most common cause of RP and although rhodopsin is only expressed in rods, the unique photoreceptor architecture means that once rods are lost, cones also degenerate through secondary mechanisms, leading to complete blindness.

Gene silencing approaches using sequence-specific short hairpin (sh)RNA molecules are not new, but a key problem in translation to clinical trial is limiting off-target effects, particularly in cones which are essential for human vision but do not express mutant rhodopsin. The Pol III promoters (e.g. U6 or H1) that drive siRNA expression are not cell-specific. Hence following AAV gene therapy the shRNA would also be expressed in cones. This would increase the risk of off-target effects and activation of the interferon pathway caused by uncontrolled expression of double-stranded RNA.

To overcome this problem, we have for the last 4 years worked an interference pathway in which the RNA is released from an intron within the gene being expressed - known as a 'mirtron'. A key advantage is that RNA expression is regulated by the same rod-specific promoter and is not therefore induced in cones. We co-express a codon-optimised rhodopsin transgene that replaces the mutant rhodopsin knocked down by the mirtron. Preliminary work has identified suitable artificial mirtrons that are correctly spliced and which do not affect the codon-optimised gene. Furthermore mirtron-derived pri-miRNA transcripts do not require Drosha/DGCR8 and thus bypass a potential bottleneck in RNA processing.