CRISPR Repair Reveals Causative Mutation in a Preclinical Model of Retinitis Pigmentosa

Abstract

Massive parallel sequencing enables identification of numerous genetic variants in mutant organisms, but determining pathogenicity of any one mutation can be daunting. The most commonly studied preclinical model of retinitis pigmentosa called the "rodless" (rd1) mouse is homozygous for two mutations: a nonsense point mutation (Y347X) and an intronic insertion of a leukemia virus (Xmv-28). Distinguishing which mutation causes retinal degeneration is still under debate nearly a century after the discovery of this model organism. Here, we performed gene editing using the CRISPR/Cas9 system and demonstrated that the Y347X mutation is the causative variant of disease. Genome editing in the first generation produced animals that were mosaic for the corrected allele but still showed neurofunction preservation despite low repair frequencies. Furthermore, second-generation CRISPR-repaired mice showed an even more robust rescue and amelioration of the disease. This predicts excellent outcomes for gene editing in diseased human tissue, as Pde6b, the mutated gene in rd1 mice, has an orthologous intron-exon relationship comparable with the human PDE6B gene. Not only do these findings resolve the debate surrounding the source of neurodegeneration in the rd1 model, but they also provide the first example of homology-directed recombination-mediated gene correction in the visual system.

Figure 1

CRISPR/Cas9-mediated repair of Y347X was achieved in a mosaic fashion in 2 of 11 Founders (F₀). (a) sgRNA targets exon 7 of the Pde6b locus (yellow box, top). Donor template (blue box, middle) encoding the wild-type allele corrects Y347X through homology-directed recombination. Donor template was made resistant to Cas9 by two modifications: (i) NdeI site and (ii) four wobble base pairs. Primers amplify an 845-bp fragment. (b) Surveyor assay reveals that 7 of 11 F₀ mice have CRISPR-induced DSBs. NdeI digestion detects correction of Y347X in F₀3 and F₀5. In F₀8 and F₀7, INDELs were incorporated alongside the ssODN. (c) Pde6b was repaired in 35.7% of F₀3 and 18.8% of F₀5 somatic cells. Allelic heterogeneity due to nonhomologous end joining was apparent as small deletions (1–10 bp) in F₀3 and F₀7, and large INDELs in F₀8, were detected. (Red letters: Y347X mutation; blue letters: wobble nucleotides carried by donor template; yellow letters: PAM sequence; brown letters: INDELs).

Figure 2

CRISPR-mediated repair of Y347X in Pde6b restored neurophysiology. (a) Uniform coloration of the fundus of an age-matched wild-type (WT) control. “Salt and pepper” retinopathy in the dorsal fundus of the rd1 mutant as well as vascular attenuation were observed (ON, optic nerve; arrows point to vessels; arrow head points out an imaging artifact). CRISPR-corrected F₀3 displays a fundus similar to that of WT, with preservation of vasculature. (b) Electroretinogram (ERG) (μV) revealed neurophysiological rescue in 1-month-old gene-corrected mice. Rd1 mice (red) had extinguished mixed maximum rod–cone, rod-specific, and cone ERG responses. In CRISPR-repaired F₀3 (dark blue) and F₀5 (light blue) mice, maximum, rod, and cone ERG responses were comparable with WT (black) and significantly heightened compared with the rd1 mutant. (c) Greater magnitude represents improved neurophysiology. Maximum a- and b-wave values obtained at 1 and 2 months in rd1 (red), F₀3 (dark blue), F₀5 (light blue), and WT (black) mice. Rd1 consistently displayed significantly diminished a-waves, which are photoreceptor mediated, and b-waves, which are inner retina mediated. F₀3 OS originally registered as nearly equivalent to the WT, but gradually decreased over time. a- and b-wave magnitudes in F₀3 OD and F₀5 OS/OD were poorer than that of WT but still greater than that of the rd1 mutant.

Figure 3

CRISPR-mediated repair of Y347X in Pde6b^rd1 restored retinal structure. (a) Optical coherence tomography data at 3 months from wild-type (WT), rd1, F₀3, and F₀5 mice reflected partial preservation of retinal layers in the CRISPR-rescued mice compared with the rd1 mutant. The outer nuclear layer (ONL) was not detectable in rd1. Yellow bar indicates the ONL thickness, which was normalized to WT and measured to be 76% and 50% in F₀3 and F₀5, respectively. (b) Hematoxylin–eosin-stained retinal section from WT, rd1, and F₀3 mice at 3 months revealed that retinal layers were preserved in F₀3 and were comparable with that of WT, whereas the rd1 mutant layers had collapsed and loss the photoreceptor (ONL) layer. INL, inner nuclear layer; RGC, retinal ganglion cell; RPE, retinal-pigmented epithelium.

Figure 4

Xmv-28 viral insertion was retained but benign in gene-repaired mice. (a) Restriction fragment polymorphism (RFLP) revealed CRISPR-mediated correction in 24–37% of cells comprising organs in F₀3, and 11–30% in F₀5. Data show that the CRISPR-mediated correction occurred at preimplantation, since repaired and uncorrected alleles were uniformly distributed in endoderm, mesoderm, and ectoderm organ lineages. (b) RFLP demonstrates that F₀3 backcrossed with the Rd1 mutant transmitted the corrected Y347X allele to 3 of 10 offspring (F₁1, 2, and 3). (c) Mixed maximal rod–cone electroretinogram (ERG) recordings from F₁ at P30. F₁1 (blue) registered a- and b-wave amplitudes similar to that of the wild-type (WT) control (black); rd1 (red) registered a flat ERG. (d) Optical coherence tomography compared retinal layer thicknesses in WT and F₁1. The latter's outer nuclear layer (ONL) was 90% of WT thickness, confirming rescue was transmitted to F₁. (e) Schematic representation of the Xmv-28 insertion (yellow box). CRISPR repairs the Y347X mutation (red), restoring it to the WT sequence (blue) while leaving the insertion intact. Dark blue and orange arrows represent primers used to screen the 5′ and 3′ ends of the viral insertion. Primer 10 targeted Pde6b exon 1, while JS610 targeted the Xmv-28 env gene, together amplifying the 5′ end of the insertion. Primer 81 targeted intron 1 of Pde6b, while primer 80 targeted the Xmv-28 gag gene, together amplifying the 3′ end of the insertion. (f) PCR confirmed that Xmv-28 insertion was retained in CRISPR/Cas9-repaired rd1 mice. This suggests that the Y347X mutation, rather than the viral insertion, causes the disease phenotype.