TABLE 1.
Summary of models that can be used in the pre-clinical assessment of optogenetic gene therapies.
| Model | Gene | Structural changes (P = postnatal day) | Functional changes | Strengths/Limitations | Model references | Optogenetic related studies |
| Murine models | ||||||
| Naturally occurring | ||||||
| Wild-type | e.g., C57BL/6 | Normal. | Normal. | Small ocular size with similar retinal structure and function compared to the primate retina but differing in size and inner limiting membrane thickness. | Shupe et al. (2006), Mohan et al. (2012) | Multiple studies (see main text). |
| Rd1 | Pde6b Y347* | Rod photoreceptor degeneration begins at P8–P10 with complete loss by P21. Cone degeneration begins around P35 with continual loss up to P50, at which point a single row remains for up to 8 months of age. | Complete loss of scotopic ERG by P21 with loss of photopic ERG by P50. | Early onset, severe degeneration providing a model of late stage human retinal degeneration. Good for proof-of-principle assessments and for strategies aimed at end-stage disease. Offers a limited window of intervention for approaches that might be applied at earlier stages of disease and targeting residual cone photoreceptor cells. | Pittler et al. (1993), Farber et al. (1994), Hackam et al. (2004) | Multiple studies (see main text). |
| Rd2/rds | Prph2 Large intron insertion leading to absence of protein | Photoreceptors lack outer segments at P7 and slow degeneration begins at P14 with complete loss of periphery at 9 months and central retina at 12 months. | ERG responses reduced but detectable with continual age-related decline and abolished by 12 months of age. | Slow, progressive rate of degeneration reflective of PRPH2 retinitis pigmentosa in humans, useful for optogenetic therapy safety and efficacy assessments in early stage retinal degeneration. | Reuter and Sanyal (1984), Jansen and Sanyal (1984), Travis and Hepler (1993), Ma et al. (1995) | Not to date. |
| Rd6 | Mfrp Deletion of splice donor site and exon skipping | Subretinal deposits appear around P50. Photoreceptor degeneration occurs progressively with age with significant thinning at 7 months of age and complete loss at 24 months. | Abnormal rod and cone ERG responses are detected from P28 and show slow degeneration over time with absence of responses by P490. | Slow degeneration with limited changes to retinal structure and function with both rod and cones similarly affected. Long-term assessment of optogenetic therapy applied at an early stage of disease would require extensive aging to determine efficacy over time. | Hawes et al. (2000), Kameya et al. (2002) | Not to date. |
| Rd7 | Nr2e3 Deletion of exons 4 and 5 | Rosette formation of the outer nuclear layer begins by P13 but these resolve over time and are not present at 16 months of age. Outer nuclear layer and outer segments show reduced thickness compared to controls. | Rod and cone responses are normal at P30 and 5 months of age. Responses reduce to ∼50% compared to controls at 16 months of age. | Slow progressive photoreceptor degeneration with unusual early structural phenotype and aged mice showing only mild retinal degeneration. | Akhmedov et al. (2000) | Not to date. |
| Rd8 | Crb1 c.3481delC | Shortened inner and outer segments by P14. Subretinal spots appear that represent retinal disorganization and dysplasia apparent at P35. | ERG responses are attenuated compared to controls but not significantly so and are stable up to 12 months of age at which point progressive loss in function occurs. | Unusual structural changes, including to the inner layer, not reflective of a traditional retinitis pigmentosa phenotype and with limited functional defects. | Chang et al. (2002), Mehalow et al. (2003), Aleman et al. (2011) | Not to date. |
| Rd9 | Rpgr 32bp duplication in ORF15 | Normal structure up to 8 weeks of age with detectable differences in outer retinal thickness up to 12 months of age. | Normal rod and cone responses at 1 month of age with gradual age-related decline, responses detectable at 24 months of age. | Slow degeneration with limited changes to retinal function. The rate of degeneration is likely too slow to be appropriate for optogenetic strategies and is not characteristic of human late stage retinal degeneration. | Thompson et al. (2012) | Not to date. |
| Rd10 | Pde6b R560C | Rod photoreceptor degeneration begins at P18 with complete loss by P35 with subsequent cone loss. By P45 only a single layer of cone cells remain. | Reduced but measurable scotopic ERG and good photopic ERG at P14-P28 with complete loss by P60. | With an early onset but mild rate of degeneration this model is more reflective of human disease and useful for long-term assessment of safety and efficacy. | Chang et al. (2002), Gargini et al. (2007) | Doroudchi et al. (2011), van Wyk et al. (2017) |
| Rd12 | Rpe65 c.130C > T | Normal appearance up to P40 with outer nuclear layer structure maintained to 3 months of age, at which point clear loss of outer segments occurs plus deposits in the RPE. | Rod ERG response significantly reduced by P21 and barely detectable at 8 months of age whilst cone responses are maintained. | Slow rate of structural changes yet early onset functional changes. This model offers the potential to target residual cones in a retina reflecting earlier stages of disease. The causative gene is important in RPE function and causes LCA therefore offers a different model in which to test optogenetic therapy. | Pang et al. (2005) | Not to date. |
| Rd16 | Cep290 Deletion of exons 35–39 | Reduced outer nuclear layer thickness at P19 with only a single layer of cone cells by P45. | Reduced rod and cone functions at P18 with absent signals by P30. | Early onset with a rate of degeneration that falls between rd1 and rd10, this offers the benefits of a good window of opportunity for intervention plus reduced aging requirements for assessments of efficacy. | Chang et al. (2006) | Doroudchi et al. (2011) |
| (Nyxnob) | Nyx 85bp deletion | Model of congenital stationary night blindness with a non-degenerate retina. | Normal a-wave but absent b-wave. | Of limited use for optogenetic studies due to abnormal bipolar cell function. | Pardue et al. (1998), Gregg et al. (2003) | Scalabrino et al. (2015) |
| Transgenic | ||||||
| Cpfl1/Rho–/– | Cone photoreceptor function loss 1/rhodopsin double-knockout | Normal outer nuclear layer thickness at 3 weeks with ONL degenerating to one row of cell bodies by 10 to 12 weeks. | Loss of scotopic and photopic responses by week 12. | Good model reflecting retinitis pigmentosa through loss of photoreceptor cells whilst maintaining the inner retina and providing a good window for intervention. | Santos-Ferreira et al. (2016) | Santos-Ferreira et al. (2016) |
| Opn4–/–, Gnat1–/–, Cnga3–/– | Opn4, Gnat, Cnga3 triple-knockout (TKO) | Normal retinal structure is maintained but functional loss occurs. | Response to flash ERG is achieved, considered to be due to stimulated rod opsin. | Mice lack optomotor responses and pupillary constriction to light yet maintain photoreceptor cells. This is an unusual model that displays abnormal function but not degeneration. | Hattar et al. (2003), Allen et al. (2010) | Lu et al. (2020) |
| Rho–/– | Rho Knockout | Normal outer nuclear layer thickness but an absence of outer segments at P24. Thinning begins at P30 and by P90 only a single row of cone cells remain with no outer segments. | Reduced rod responses by P24. Cone ERG maintained to P47 after which degeneration occurs. | Retinal degeneration with cone cell survival and function allowing for a big window of intervention with optogenetic therapy. | Humphries et al. (1997), Toda et al. (1999), Hassall et al. (2020) | Not to date. |
| Nrl–/– | Nrl Knockout | Absence of rods from birth with abnormal outer nuclear layer structure at P35. Surviving cones display reduced outer segment thickness and irregular outer nuclear layer stacking. | Absence of rod ERG function by P35. Cone responses enhanced. | Unusual structural changes to the outer nuclear layer and despite cone cell survival, the phenotype is not characteristic of typical retinitis pigmentosa. | Mears et al. (2001), Daniele et al. (2005) | Not to date. |
| P23H/+ | Rho Knock-in of human P23H | Reduced outer nuclear layer thickness due to loss of rods at P63 with further loss to only 2–4 rows of nuclei at P112. Cone nuclei counts equivalent to age-matched control mice. | Scotopic ERG strongly reduced at P41 and barely detectable by P170. Photopic responses normal up to P40, mildly reduced at P70 and severely reduced by P170. | Slow, progressive degeneration, reflective of human RHO P23H retinitis pigmentosa and useful for assessment of safety and efficacy of optogenetic therapy in early stage retinal degeneration. | Sakami et al. (2011) | Not to date. |
| Dp71–/– | Knockout of the Duchenne muscular dystrophy small protein Dp71 | Blood-retinal barrier permeability. | Slightly reduced ERG a-wave compared to WTs. | Of limited use for optogenetic strategies but useful comparisons for vector transduction studies in the presence of altered barriers. | Dalloz et al. (2003) | Vacca et al. (2013) |
| Large animal models | ||||||
| Canine | e.g., Briard dog with a naturally occurring 4bp deletion in Rpe65 | Progressive loss of photoreceptors over many years. | Reduced ERG. | Useful as a naturally occurring larger model of retinal degeneration. | Narfström et al. (1989), Wrigstad et al. (1992) | Ameline et al. (2017) |
| Non-human primates | Marmoset, Macaque | Normal. | Normal. | Useful for in vivo assessment of safety of optogenetic strategies with post-mortem evaluation of efficacy in the presence of pharmacological blockade of native photoreception. | Picaud et al. (2019) | Ivanova et al. (2010), Dalkara et al. (2013), Sengupta et al. (2016), Chaffiol et al. (2017), McGregor et al. (2020) |
| Human-based organoids and explants | ||||||
| Human-derived organoids | Can be derived from iPSCs of normal controls or patients with inherited retinal disease. | Absence of outer segments when derived from normal controls. | Able to generate light-sensitive responses. | Being human-derived is advantageous to confirm efficacy of vectors in human cells but these take time to develop and convert from iPSCs (up to 200 days) and to then achieve evidence of transduction (up to 50 days), but functional outputs are possible. | Zhong et al. (2014), Reichman et al. (2017), Chichagova et al. (2019) | Gonzalez-Cordero et al. (2017), Khabou et al. (2018), Garita-Hernandez et al. (2020), Lane et al. (2020) |
| Human retinal explants | May be damage or scarred depending on the donor origin. | Restricted to donor availability, which may be of limited sample quality. Difficult to achieve any functional or quantitative output. | Orlans et al. (2018) | De Silva et al. (2016), van Wyk et al. (2017), Khabou et al. (2018) | ||
The asterisk indicates a premature stop codon/termination of the protein occurs. Broadly, the assessments include optimizations of vector and transgene biology and their transduction efficiency in target retinal cells, functional assessments of optogenetic molecules, as well as general safety of optogenetic gene therapy. These considerations will in turn determine the choice of a study model, or indeed assessments across multiple models within or across species.