Allele specific genome editing for the treatment of Retinitis Pigmentosa

Retinitis pigmentosa (RP) is a wide group of genetic disorders causing the progressive degeneration of photoreceptors in the retina, leading in the vast majority of cases to irreversible blindness early in life. The molecular genetics of RP is diverse, with approximately one-third of all cases caused by autosomal dominant mutations (adRP), which are particularly amenable for treatment using CRISPR-mediated allele-specific gene KO, when toxic gain of function mutant proteins are produced. The most commonly mutated gene in adRP is RHO, which encodes the photopigment rhodopsin, however more than 100 different mutations have been identified to-date (HGMD, www.hgmd.cf.ac.uk) and targeting strategies have been reported for only a handful of these so far. In this thesis we present a strategy to specifically knock-out the mutated RHO P347L (c.1040C>T) allele which is responsible for the development of a severe form of adRP. We employ both wt SpCas9 and previously reported high-fidelity SpCas9 variants to specifically target the RHO P347L mutation, while sparing the wt counterpart. We have tested our targeting strategy in vitro in HEK293T cells stably overexpressing the RHO P347L mutant allele, demonstrating that we can specifically edit the locus, but we were not able to achieve functional downregulation of mutant RHO with our high-fidelity Cas9 nucleases. To expand the population of patients eligible for potential treatment, we have also developed an allele-specific and mutation-independent editing approach exploiting our high-fidelity SpCas9 variants to specifically target RHO mutated alleles. Our high-fidelity variants can selectively distinguish alternative alleles of a high-frequency SNP located in the RHO gene to tag and destroy mutant rhodopsin, thereby stopping disease progression. Our strategy leverages on two basic genetic requirements, besides the knowledge of the mutation affecting RHO: i) heterozygosity for the selected SNP; ii) knowledge of the phase between the SNP’s alleles and the mutation in the rhodopsin gene in order to select which of the two alleles to target. We have tested our targeting strategy in vitro using custom-built cell lines expressing two different rhodopsin mutants (P23H and P347L) demonstrating allele-specific editing and specific downregulation of mutant RHO only when using our high-fidelity Cas9 nucleases. - 3/130 - In addition, we verified that no detrimental effects can be observed on the non-target wt RHO allele. Finally, we have determined the safety profile of our approach by measuring genome-wide off-targets of our best performing sgRNAs in combination with our high-fidelity nucleases by using GUIDE-seq and targeted deep-sequencing.