Allele-specific gene-editing approach for vision loss restoration in RHO-associated retinitis pigmentosa
Curation statements for this article:-
Curated by eLife
**eLife assessment
**
This work provides a valuable allele-specific gene editing therapeutic approach to selectively target the human RHO-T17M mutation, one of the most frequent genetic causes of autosomal dominant retinitis pigmentosa. However, the current data are incomplete. Further validation of gene editing efficiency in rods at cellular level in vivo and use of Rho-T17M mice will strengthen the conclusion.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Mutant RHO is the most frequent genetic cause of autosomal dominant retinitis pigmentosa (adRP). Here, we developed an allele-specific gene-editing therapeutic drug to selectively target the human T17M RHO mutant allele while leaving the wild-type RHO allele intact for the first time. We identified a Staphylococcus aureus Cas9 (SaCas9) guide RNA that was highly active and specific to the human T17M RHO allele. In vitro experiments using HEK293T cells and patient-specific induced pluripotent stem cells (iPSCs) demonstrated active nuclease activity and high specificity. Subretinal delivery of a single adeno-associated virus serotype 2/8 packaging SaCas9 and single guide RNA (sgRNA) to the retinas of the RHO humanized mice showed that this therapeutic drug targeted the mutant allele selectively, thereby downregulating the mutant RHO mRNA expression. Administration of this therapeutic drug resulted in a long-term (up to 11 months after treatment) improvement of retinal function and preservation of photoreceptors in the heterozygous mutant humanized mice. Our study demonstrated a dose-dependent therapeutic effect in vivo . Unwanted off-target effects were not observed at the whole-genome sequencing level. Our study provides strong support for the further development of this effective therapeutic drug to treat RHO -T17M-associated adRP, also offers a generalizable framework for developing gene-editing medicine. Furthermore, our success in restoring the vision loss in the suffering RHO humanized mice verifies the feasibility of allele-specific CRISPR/Cas9-based medicines for other autosomal dominant inherited retinal dystrophies.
Article activity feed
-
-
**eLife assessment
**
This work provides a valuable allele-specific gene editing therapeutic approach to selectively target the human RHO-T17M mutation, one of the most frequent genetic causes of autosomal dominant retinitis pigmentosa. However, the current data are incomplete. Further validation of gene editing efficiency in rods at cellular level in vivo and use of Rho-T17M mice will strengthen the conclusion. -
Reviewer #1 (Public Review):
The authors attempted to delete a rhodopsin allele with single-nucleotide mutation seen in a Chinese subpopulation of autosomal dominant retinitis pigmentosa patients, (Rho-T17M). This was done in vitro and in vivo, while keeping the Rho wild type allele intact in vitro and in vivo using CRISPR-SaCAS9 guide RNA-specific approach, a previously established technique. In this study, solid in vitro data was presented showing that one of the tested guide RNAs was effective to specifically delete targeted the Rho-T17M sequence of synthetic DNA as well as in iPSCs from RP patients. However, the in vivo part of this study is incomplete. The issues are: 1. confusing choice of disease animal model (Rho-5m mice that carry 4 additional rhodopsin mutations other than the targeted T17M); 2. no proof of gene editing …
Reviewer #1 (Public Review):
The authors attempted to delete a rhodopsin allele with single-nucleotide mutation seen in a Chinese subpopulation of autosomal dominant retinitis pigmentosa patients, (Rho-T17M). This was done in vitro and in vivo, while keeping the Rho wild type allele intact in vitro and in vivo using CRISPR-SaCAS9 guide RNA-specific approach, a previously established technique. In this study, solid in vitro data was presented showing that one of the tested guide RNAs was effective to specifically delete targeted the Rho-T17M sequence of synthetic DNA as well as in iPSCs from RP patients. However, the in vivo part of this study is incomplete. The issues are: 1. confusing choice of disease animal model (Rho-5m mice that carry 4 additional rhodopsin mutations other than the targeted T17M); 2. no proof of gene editing efficiency at the cellular level of the targeted cell type (i.e. what percentage of rod photoreceptors lose the T17M disease mutation?); and 3) lack of evidence of therapeutic potential (i.e. is there any rescue of vision in the mouse disease model or any toxicity due to the vector itself?).
-
Reviewer #2 (Public Review):
The authors attempt to develop an allele-specific editing approach targeting RHO-T17M mutation for potential therapeutic use to treat the mutation associated with autosomal dominant retinitis pigmentosa.
The authors reported three sgRNAs for the RHO T17M allele for verification. It would be helpful to describe details of the discovery phase of these sgRNAs, including design, in silico predictions, inclusion criteria, off-target analysis, etc.
The authors claim that the targeted gene-editing efficiencies are dose-dependent. However, data were presented from only one mouse for the 5ｘ108 dose group (line 231-237), which might need more explanation.
With respect to Fig. 4C, the flat-mount retina is not representative. A better image of flat-mount of retina is preferred.
With respect to Fig. 6B & 6C, it seems that …
Reviewer #2 (Public Review):
The authors attempt to develop an allele-specific editing approach targeting RHO-T17M mutation for potential therapeutic use to treat the mutation associated with autosomal dominant retinitis pigmentosa.
The authors reported three sgRNAs for the RHO T17M allele for verification. It would be helpful to describe details of the discovery phase of these sgRNAs, including design, in silico predictions, inclusion criteria, off-target analysis, etc.
The authors claim that the targeted gene-editing efficiencies are dose-dependent. However, data were presented from only one mouse for the 5ｘ108 dose group (line 231-237), which might need more explanation.
With respect to Fig. 4C, the flat-mount retina is not representative. A better image of flat-mount of retina is preferred.
With respect to Fig. 6B & 6C, it seems that T17M protein and RHO-5m protein are likely detected in both cytoplasm and plasma membrane rather than being limited to the cytoplasm alone.
The therapeutic efficacy benefit should be supported by data of photoreceptor function and cell preservation after treatment. It is be better to include two more control groups, namely wild-type mice and untreated mutant mice, which may help evaluate improved response after treatment.
The mouse lines are confusing. Did the authors generate three lines of mice, including RHOwt/hum, Mut-RHOwt/hum, RHOhum/m-hum mice? Did the authors use the Rhohum/m-hum mice for verification of cutting efficiencies, whereas they use the other two lines of mice for rescue experiments? The authors should clarify.
Mut-RHOwt/hum mice have previously been reported to have fundus pigment abnormalities, so the fundus should be examined after rescue. The expression of Rho-5M mRNA was reduced in vitro. Was the expression of RHO mRNA also down regulated after rescue as well as in vitro? Did the subretinal injection of GFP spread to the whole retina? This can be determined with retinal flat mount or panretinal staining using GFP labeling. The authors showed that the cell numbers in the ONL were increased in the treatment group compared with the control group at 9 mpi. Were the other nuclear layers or plexiform layer also affected? Did the other retinal cells develop normally? Figure 8 showed retinal functions with AAV-based SaCas9/17-Sg2 in Mut-Rhowt/hum mice. ERG of Mut-Rhowt/hum mice without treatment are also needed.
The efficiency and safety of RHO T17M allele-specific editing in this paper are well supported by in vitro and in vivo experiments.
The fundamental basis of the study design should be clearly stated, ie which truncation variants in RHO cause disease or not. It is reported that truncation variants occurring before K296 are likely benign, which should be mentioned. This is the key starting point for this kind of study and is not limited to RHO. but as an allele-specific gene editing approach as a potential therapy for dominant mutations in any gene for which heterozygous loss-of-function is tolerated in the whole gene or in part of the gene (mostly at N-terminals). Apart from RHO, in fact, N-terminal truncating variants in several other IRD associated genes have been reported to be benign in heterozygotes, including CRX, TOPORS, RP1, etc. This study verified the efficiency and safety of this approach based on both patient derived iPSC and humanized animal models which are unique compared with other studies on RHO.
-
