Abstract
Purpose :
Retinal organoids are an ideal model for in vitro assessments of gene therapy vectors. Mirtron transgenes used for rhodopsin knockdown and replacement in vivo were packaged in AAV serotypes relevant for retinal organoid transduction to determine optimal conditions for achieving desired outcomes prior to assessment in patient-derived retinal organoids.
Methods :
A double mirtron transgene of RHOp.2xM.coRHO was packaged in AAV5 or AAV2 quad mutant with the triple mirtron transgene RHOp.3xM.coRHO packaged in AAV5. At differentiation day 150, 1E+10 or 5E+10 genome copies were applied to wild-type retinal organoids obtained from Newcells Biotech Ltd. At 4 weeks post-transduction, retinal organoids were processed for transcript assessments by TaqMan assay or immunostaining.
Results :
Native rhodopsin transcript levels were significantly knocked down to undetectable levels in retinal organoids treated with 1E+10 double mirtron AAV2 quad mutant (p<0.0001, n=5) and 5E+10 triple mirtron AAV5 (p<0.0001, n=5) vectors. Mirtron-resistant rhodopsin transcript levels were high in all retinal organoids that received mirtron vectors, with the highest levels detected in the AAV2 quad mutant group. Immunostaining revealed no apparent changes in retinal structure compared to untreated retinal organoids. Recoverin and cone opsin staining profiles were similar between treated and untreated retinal organoids. Similar levels of rhodopsin staining were observed between untreated organoids and those treated with double and triple mirtron AAV5.
Conclusions :
Complete knockdown of native rhodopsin transcripts was achieved in wild-type retinal organoids treated with the AAV5 triple mirtron vector whilst the equivalent double mirtron AAV5 vector did not achieve the same degree of knockdown. Replacement levels of mirtron-resistant rhodopsin were detected at transcript and protein level. Combined, the data suggest that the triple mirtron vector may offer the optimal combination of rhodopsin knockdown with concomitant replacement.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.