In Vivo CRISPR/Cas9 Gene Editing Corrects Retinal Dystrophy in the S334ter-3 Rat Model of Autosomal Dominant Retinitis Pigmentosa

Abstract

Reliable genome editing via Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)/Cas9 may provide a means to correct inherited diseases in patients. As proof of principle, we show that CRISPR/Cas9 can be used in vivo to selectively ablate the rhodopsin gene carrying the dominant S334ter mutation (Rho(S334)) in rats that model severe autosomal dominant retinitis pigmentosa. A single subretinal injection of guide RNA/Cas9 plasmid in combination with electroporation generated allele-specific disruption of Rho(S334), which prevented retinal degeneration and improved visual function.

Figure 1

Allele-specific targeting and disruption of Rho^S334 in vitro. (a) Schematic of px330 construct used. (b) gRNA^TRGT and predicted genomic DNA binding sites in Rho^S334 and Rho^WT. PAM (red underlined bases) and mismatches (red font) are shown. (c) Phase contrast photomicrograph of mCherry⁺ and mCherry⁻ MSCs^S334 3-days post-lipofection with gRNA constructs prior to fluorescence-activated cell sorting (FACS) isolation. (d) FACS gating strategy for mCherry⁺ MSC^S334 isolation is shown. mCherry⁺ cells represented 12% of the total population, with the brightest 3.4% mCherry⁺ MSCs^S334 selected for genomic DNA sequence analysis using PCR amplicons encompassing predicted Cas9 cleavage sites. (e–h) Genomic DNA sequencing results are shown with Phred quality scores (gray shading) at bottom. DNA disruption is shown from gRNA^TRGT-transfected MSCs^S334 downstream from the Cas9 cleavage site (blue highlight) in Rho^S334 (e), but not in Rho^WT (f). Genomic disruption was absent at the Rho^S334 locus using gRNA^CNTRL (g) or with no vector in untreated eyes (h). Bar = 400 µm. CMV, cytomegalovirus; MSC, mesenchymal stem cell; PAM, protospacer adjacent motif.

Figure 2

Allele-specific targeting and disruption of Rho^S334 in vivo. (a) Fast-Green DNA dye shows plasmid distribution following unilateral subretinal injection in S334ter-3 rats at P0. (b) Representative retinal flat-mount shows variable mCherry intensity and uneven distribution 4 days after injection. (c) Gating strategy for fluorescence-activated cell sorting isolation of enzymatically dissociated P4 retinal cells with mCherry fluorescence intensity at high (red, 0.3%) intermediate (purple, 0.6%), and no (blue, 97.9%) expression. Sanger sequencing of PCR-amplified genomic Rho loci from photoreceptors showed in vivo disruption of Rho^S334 (d), but not Rho^WT (e) using gRNA^TRGT. (f) Rho^S334 locus targeting schematic and deep sequencing reads shows insertions/deletions (indels) proximal to the predicted cleavage site (arrowhead) from gRNA^TRGT-expressing cells. PAM, protospacer adjacent motif.

Figure 3

Phenotypic rescue by Rho^S334-selective ablation. (a–e) Fluorescent confocal images of gRNA^TRGT-treated (a–c) and gRNA^CNTRL-treated (d,e) retinas at P33. (a) Montage image shows that mCherry distribution correlated with ONL rescue (DAPI, blue) and POS formation (RHO^WT C-terminal immunolabel, green). Inset: magnified image of outlined region shows preserved ONL with organized POS (arrowheads) adjoining degenerated ONL with diminished POS (arrows). (b,c) RHO N-terminal (b) and C-terminal (c) immunostaining in gRNA^TRGT-treated retinas was absent from PR cell bodies and localized to OS. Significant PR preservation in the ONL was observed from gRNA^TRGT treatment (b, bracketed areas). (d,e) RHO N-terminal (d) and C-terminal (e) immunostaining was absent in gRNA^CNTRL-treated retinas, which lacked POS formation and ONL contained one row of remaining PR nuclei (d, bracketed areas). DAPI, 49,69-diamidino-2-phenylindole; INL, inner nuclear layer; IS, inner segment; ONL, outer nuclear layer; OS, outer segment; POS, photoreceptor outer segment; PR, photoreceptor.

Figure 4

Rho^S334 ablation preserved cone morphology and second-order retinal neuron synapses. Fluorescence confocal microscopy images from gRNA^CNTRL-treated (a–d) or gRNA^TRGT-treated (e–j) eyes at P33. (a,b) Surviving PRs after gRNA^CNTRL treatment were non-Rho-expressing cones PRs (a,d, cone arrestin, green), which lacked typical morphological features observed in retinas rescued with gRNA^TRGT treatment (e,h). Individual channel images corresponding to bracketed areas in (a) and (e) show rescued PR nuclei (DAPI, (b) versus (f)) in gRNA vector transfected areas (mCherry, (c) versus (g)) with preserved cone morphology (i.e., pedicles and POS, (d) versus (h)). (i) Greater dendritic arborization of INL-resident rod-bipolar neurons (PKC-α, green) was evident at the OPL in mCherry⁺ areas following gRNA^TRGT treatment (left inset), in contrast to the adjacent degenerated area lacking mCherry⁺ (right inset). (j) Similarly, synaptophysin immunolabel (green) showed greater intensity in OPL regions in which PR nuclei preservation (DAPI) corresponded with mCherry⁺ expression (inset: arrowheads), in sharp contrast with the adjacent unprotected area to which gRNA^TRGT transfection did not extend (inset: arrows). DAPI, 49,69-diamidino-2-phenylindole; INL, inner nuclear layer; ONL, outer nuclear layer; OPL, outer plexiform layer; PKC-α, protein kinase C-alpha; PR, photoreceptor.

Figure 5

PR rescue by gRNA^TRGT treatment corresponded with vision rescue. (a) Fluorescent microscopy montage image shows mCherry⁺ reporter distribution (arrows) of gRNA^TRGT vector in a retinal flat-mount at P33 calculated at 29% of total retina area by NIH ImageJ analysis. (b) By retinal cross-section, mCherry⁺ regions from gRNA^TRGT treatment contained significantly more PR nuclei than the mCherry⁺ regions from gRNA^CNTRL treatment, or comparable regions from untreated control areas (gRNA^TRGT: 307 ± 82 PR nuclei/100 µm, N = 5 versus gRNA^CNTRL: 33 ± 3, **P ≤ 0.01, N = 3 versus Untreated: 27 ± 13, ^††P ≤ 0.01, N = 4). (c) Visual acuity (optokinetic response) was significantly higher from gRNA^TRGT treatment at P39, than from gRNA^CNTRL treatment (gRNA^TRGT: 0.185 ± 0.008 c/d, N = 5 versus gRNA^CNTRL: 0.121 ± 0.009 c/d, N = 4, ^††P ≤ 0.01). Visual acuity in gRNA^TRGT-treated eyes was significantly higher than in untreated contralateral eyes (Treated: 0.185 ± 0.008 versus Contralateral: 0.138 ± 0.006 c/d, N = 5, ***P ≤ 0.001). Visual acuity in eyes injected with gRNA^CNTRL was not different from that of contralateral noninjected eyes (Treated: 0.121 ± 0.009 versus Contralateral: 0.121 ± 0.012 c/d, N = 4, P = 0.763). (d) By using the fellow eyes of individual animals as internal controls, the higher visual acuity from gRNA^TRGT treatment represented a 35 ± 4.6% increase, compared to a 2.3 ± 0.7% decrease with gRNA^CNTRL injection (P < 0.01). c/d = cycles/degree. All values represent mean ± SEM. N.S., not significant; PR, photoreceptor.