Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Long-term retinal cone rescue using a capsid mutant AAV8 vector in a mouse model of CNGA3-achromatopsia

  • Xufeng Dai,

    Roles Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Ying He,

    Roles Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Visualization

    Current address: Yucheng People's Hospital, Yucheng, Shandong, P.R. China

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Hua Zhang,

    Roles Funding acquisition, Resources, Supervision

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Yangyang Zhang,

    Roles Methodology, Resources

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Yan Liu,

    Roles Methodology, Resources

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Muran Wang,

    Roles Methodology, Resources

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Hao Chen ,

    Roles Supervision, Writing – review & editing

    chenhao823@mail.eye.ac.cn (HC); jpangoph@hotmail.com (JP)

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

  • Ji-jing Pang

    Roles Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing

    chenhao823@mail.eye.ac.cn (HC); jpangoph@hotmail.com (JP)

    Affiliation School of Ophthalmology and Optometry, The Eye Hospital, Wenzhou Medical University, Wenzhou, Zhejiang, P.R. China

Abstract

Adeno-associated virus (AAV) vectors are important gene delivery tools for the treatment of many recessively inherited retinal diseases. For example, a wild-type (WT) AAV5 vector can deliver a full-length Cnga3 (cyclic nucleotide-gated channel alpha-3) cDNA to target cells of the cone photoreceptor function loss 5 (cpfl5) mouse, a spontaneous animal model of achromatopsia with a Cnga3 mutation. Gene therapy restores cone-mediated function and blocks cone degeneration in the mice. However, since transgene expression delivered by an AAV vector shows relatively short-term effectiveness, this cannot be regarded as a very successful therapy. AAV2 and AAV8 vectors with capsid mutations have significantly enhanced transduction efficiency in retinas compared to WT AAV controls. In this study, AAV8 (Y447, 733F+T494V)-treated cpfl5 retinas showed greater preservation of short-term cone electroretinogram (ERG) responses than AAV8 (Y447, 733F)- or AAV2 (Y272, 444, 500, 730F+T491V)-mediated treatments. To explore the long-term rescue effect, AAV8 (Y447, 733F+T494V)-treated cpfl5 retinas were evaluated at 9 months following postnatal day 14 (P14) treatment. Rescued ERG responses in the cones of treated cpfl5 eyes decreased with increasing age, but still maintained more than 60% of the WT mouse responses at the oldest time point examined. Expression of CNGA3 and M/S-opsins was maintained in cone outer segments of the treated cpfl5 eyes and was equal to expression in age-matched WT retinas. Near-normal cone-mediated water maze behavior was observed in the treated cpfl5 mice. As these are the longest follow-up data reported thus far, AAV8 with capsid Y-F and T-V mutations may be one of the most effective AAV vectors for long-term treatment in a naturally occurring mouse model of CNGA3 achromatopsia.

Introduction

Achromatopsia, also known as rod monochromatism, is a relatively rare autosomal recessive retinal disorder characterized by cone photoreceptor dysfunction. Clinically, the disease is generally classified into complete (typical) and incomplete (atypical) forms [1]. Typical symptoms of complete achromatopsia are more severe than the incomplete form. They include seriously reduced visual acuity, nystagmus, photophobia, and color blindness [1]. With only rod-mediated vision, patients are extremely sensitive to light and have daylight blindness. To reduce photophobia, currently available medical care is to limit light exposure using dark glasses. With the development of adeno-associated virus (AAV) vectors as gene delivery tools for many recessively inherited retinal diseases, several promising gene therapy projects have been initiated [29]. Recent preclinical trials have made significant progress in providing effective treatment for achromatopsia. The first clinical trials of gene therapy are either underway or will be launched soon and they are expected to contribute important data on the safety and efficacy of these treatments [10].

Thus far, six genes have been implicated in achromatopsia-associated mutations [1115]: cyclic nucleotide-gated channel alpha-3 (Cnga3) [16,17], cyclic nucleotide-gated channel beta-3 (Cngb3) [18,19], guanine nucleotide binding protein alpha transduction active peptide 2 (Gnat2) [1,20], phosphodiesterase 6C (Pde6c) [21,22], Pde6h [12,13], and cyclic AMP-dependent activating transcription factor-6 alpha (Atf6) [14,15]. The proteins encoded by these genes play vital roles in the phototransduction cascade of cone photoreceptors.

The Cnga3 gene encodes a member of the cyclic nucleotide-gated ion channel protein family, which is critical for normal vision in cone photoreceptors [23]. As the first identified and second most common cause of achromatopsia, Cnga3 mutations account for approximately 25% of all cases [24,25]. A cone photoreceptor function loss 5 (cpfl5) mouse strain, with a naturally occurring Cnga3 mutation, was discovered at The Jackson Laboratory [11]. Due to a single nucleotide A to G transition at position 492 of exon 5, the deficient mice exhibit selective loss of cone-mediated electroretinogram (ERG) responses [11]. In addition, it has been shown that loss of CNGA3 results in impaired expression and trafficking of cone opsins [11,26].

New generations of viral vectors have made it possible to deliver functional genes to retinal cells [27]. Gene therapy, which can rescue visual function, has been used to treat achromatopsia in some animal models [28]. The most commonly used transgene vectors are those derived from AAVs. Early studies have shown short-term rescue [28,29]. In a later study using a wild-type (WT) AAV5 vector driven by the chicken beta actin (CBA) promoter, photopic ERG b-wave responses were maintained to an average of 80% of the WT mouse responses at 5 months following subretinal gene therapy [11]. However, a longer duration of rescue was not pursued in those studies. It cannot be considered a successful therapy if transgene expression delivered by an AAV vector shows only short-term effectiveness.

AAV vectors with different serotypes and capsid mutations have been developed. These include AAV2 and AAV8, with capsid surface-exposed tyrosine residues mutated to phenylalanine (Y-F) [11] and/or threonine mutated to valine (T-V) [30]. These mutations were shown to protect vector particles from proteasomal degradation [3034]. Designing new vectors with these mutations may be an effective way to improve longevity of transgene expression.

In this study, three AAV vectors with different capsid mutations were compared for treatment of cpfl5 mice. These vectors were AAV8 (Y447, 733F + T494V), AAV8 (Y447, 733F), and AAV2 (Y272, 444, 500, 730F + T491V). All vectors delivered the same Cnga3 cDNA, but had different Y-F and/or T-V mutations on the capsid. We report the 9-month preservation of cone structure and function using the AAV8 (Y447, 733F + T494V) vector to deliver treatment in a mouse model of achromatopsia.

Materials and methods

Animals

The congenic inbred strain of the cpfl5 mice and the isogenic WT C57BL/6J mice were acquired from The Jackson Laboratory (Bar Harbor, ME, USA). Mice were bred and maintained in the Animal Facilities of Wenzhou Medical University (Wenzhou, China). All animals were maintained on a cycle of 12 h of light and 12 h of dark, with free access to water and food. Animal experiments were approved by Wenzhou Medical University’s Institutional Animal Care and Use Committee (Permit Number: wydw2014-0072), and conducted according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

Construction of AAV vectors

The AAV8 (Y447, 733F + T494V) capsid is an AAV serotype 8 capsid with a double Y-to-F mutation at residue 447 and 733, accompanied by a single T-to-V mutation at residue 494. AAV8 (Y447, 733F + T494V) was used to package the vector DNA. For comparison, the vector DNA was also packaged in AAV8 (Y447, 733F) and AAV2 (Y272, 444, 500, 730F + T491V). IRBP/GNAT2 is a hybrid promoter consisting of a 277-bp GNAT2 promoter and a 214-bp interphotoreceptor retinoid binding protein (IRBP) enhancer. AAV vectors containing the IRBP/GNAT2 promoter exhibit cone photoreceptor-specific transgene expression [35]. Mouse Cnga3 cDNA was cloned under the IRBP/GNAT2 promoter to make an AAV-IRBP/GNAT2-Cnga3 construct [35]. All AAV vectors were constructed and purified at the University of Florida (Gainesville, FL, USA).

Vector delivery via subretinal injection

Subretinal injection was performed at postnatal day 14 (P14) [36] to achieve maximum rescue of cone photoreceptors. Cpfl5 mice were treated with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 (Group 1), AAV2 (Y272, 444, 500, 730F + T491V)-IRBP/GNAT2-Cnga3 (Group 2), or AAV8 (Y447, 733F)-IRBP/GNAT2-Cnga3 (Group 3). One microliter of each vector solution (1013 vector genomes per mL) was injected subretinally into one eye of each cpfl5 mouse. The other eye remained uninjected as a control. Subretinal injection was performed as described previously [36]. Each original AAV-vector solution (1E13 vg/ml) was diluted into 2E12 and 1E11 vg/ml, and the dilutions were also administered unilaterally by subretinal injection (1μl) on P14. A small amount of fluorescein (0.1 mg/mL final concentration) was routinely added to allow visualization of the AAV vector solution [36]. An injection was considered successful if blood vessels in the detached retina could be clearly seen with green dye underneath, suggesting that the AAV vector solution was in the subretinal space [36]. Mice were selected for further evaluation if they had minimal surgical complications and their initial detached retinal blebs (the area with vector solution underneath) covered more than 80% of the whole retina. We included at least six mice per group for statistical analysis.

ERG recordings

After overnight dark adaptation, mice were anesthetized with a solution of ketamine (70 mg/kg) and xylazine (5 mg/kg) under dim red light. Full-field ERGs were recorded under a standard Ganzfeld dome, which is controlled by a computer-based system (Roland Consult, Wiesbaden, Germany). White light-emitting diodes (LEDs, 450–780 nm) were used as stimulation and background light sources. Scotopic ERGs were recorded at 0 log cd-s/m2 stimulus intensity [37]. With an interstimulus interval of 30 seconds, five responses were recorded and averaged. After adapting to a steady background illumination (30 cd/m2) for 10 min, photopic ERGs were recorded with a white-light stimulus intensity of +1.0 log cd-s/m2 [38]. To increase the signal:noise ratio, 50 individual ERG responses were averaged to produce the final waveform. The flash duration was set at 2 ms and the band pass of the amplifiers at 1–500 Hz. Amplitudes and peak times of ERG responses were saved for further evaluation.

Immunohistochemistry

Mice were sacrificed by CO2 inhalation. Retinal sections were prepared as described in detail previously [11]. Briefly, eyes were enucleated and fixed immediately in 4% paraformaldehyde in 0.1 M phosphate buffer (pH = 7.4, 4°C) overnight. The cornea, lens, and vitreous were removed and the remaining eyecup was dehydrated in 30% sucrose for 4 h. After embedding in optimal cutting temperature medium (OCT; Sakura Finetek USA Inc., Torrance, CA, USA), samples were frozen in liquid nitrogen and cut on a cryostats. Cryosections (12-μm thick) were incubated overnight at 4°C with rabbit anti-mouse polyclonal CNGA3 primary antibody (1:200, bs-10772R; Bioss, Beijing, China). After three rinses with 0.1 M PBS, sections were incubated with goat anti-rabbit IgG conjugated to a Cy3 fluorochrome (1:400, AP187C; Merck Millipore, Darmstadt, Germany) for 2 h, followed by three rinses with 0.1 M PBS. Additionally, FITC-conjugated peanut agglutinin (PNA, 1:400; Vector Laboratories, Burlingame, CA, USA) was used to detect the interphotoreceptor matrix sheath, which surrounds the cone outer segments. Similarly, frozen sections were stained for M- or S-cone opsins [39]. Nuclei were stained with 4',6-diamidino-2- phenylindole (DAPI). Retinal whole-mounts were prepared and stained for M- or S-cone opsins as described previously [39]. Retinal cryosections and whole-mounts were mounted with coverslips and imaged by fluorescence microscopy.

Visually guided water maze behavioral test

The visually guided water maze behavioral test was performed as described previously [40] with only minor modifications. Briefly, 9 months after injection, the AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3-treated cpfl5 mice, together with age-matched untreated cpfl5 and WT mice, were initially trained to escape to a small platform positioned randomly in a water tank. The water tank had a diameter of 1.2 meters. Before formal tests, mouse pupils were dilated (1% atropine) and retinas were fully light-adapted (100 cd/m2 for 10 min) to prevent rod intrusion. During each test, a mouse was initially placed in the water tank from one of four equally spaced starting locations. The time taken to escape to the randomly positioned platform was recorded as the visually guided behavioral data. If a mouse could not escape to the platform within 60 seconds, it was guided to the platform and its escape time was recorded as 60 seconds. The water maze tests were performed in well-lit (18 lux) environments.

Statistical analysis

ERG data were presented as mean ± standard deviation (SD). SPSS 18.0 (IBM Corporation, Armonk, NY, USA) was used for statistical analysis. The data were checked by Shapiro-Wilk test for normality before applying any parametric test. Paired sample t-test or one-way ANOVA with least significant difference (LSD) post hoc test was used for data comparison between or among groups. A P-value of less than 0.05 was considered statistically significant.

Results

Rescue efficacy of AAV vectors with capsid Y-F and/or T-V mutations in cpfl5 ERGs

Cpfl5 mice exhibit selective loss of cone ERG responses, similar to the electroretinographic phenotype of complete achromatopsia patients with Cnga3 mutations. After treatment with one of the three AAV vectors at P14, eyes of cpfl5 mice were tested by scotopic and photopic ERGs (Fig 1A and 1B). ERG data in each group were normally distributed (Shapiro-Wilk test, n = 6, P > 0.05). The photopic b-wave amplitudes were significantly improved in the three treated cpfl5 groups (versus the untreated cpfl5 eyes, P < 0.001), to 42%∼84% of amplitudes in age-matched WT mice (P < 0.05). Among the three treated groups, Cnga3 was delivered by AAV8 (Y447, 733F + T494V), AAV2 (Y272, 444, 500, 730F + T491V), or AAV8 (Y447, 733F) vectors. At 1 month after treatment, early preservation of photopic ERG b-waves was statistically higher in AAV8 (Y447, 733F + T494V) (75 ± 13 μV) and AAV2 (Y272, 444, 500, 730F + T491V) (76 ± 14 μV) groups, compared with the AAV8 (Y447, 733F) group (56 ± 11 μV, P < 0.01; Fig 1C). Here, vectors with both Y-F and T-V mutations led to better ERG rescue than vectors with only a Y-F mutation. At 3 months after treatment, photopic amplitudes of AAV2 (Y272, 444, 500, 730F + T491V)- and AAV8 (Y447, 733F)-treated retinas decreased to 47 ± 9 μV and 44 ± 11 μV, respectively, while the average amplitude of the AAV8 (Y447, 733F + T494V) group continued to rise slightly to 86 ± 8 μV. According to these short-term efficacy data, the AAV8 vector with both Y-F and T-V mutations rescued the best cone ERG responses (Fig 1C). In addition, a dose escalation study for each vector had been performed. Cpfl5 retinas were analyzed by photopic ERGs at 1 month post-injection (Fig 1D). Among the three AAV mutants, a dose at 1E13 vg/ml produced the best improvement for b-wave amplitudes.

thumbnail
Fig 1. Comparisons of ERGs at two time-points following AAV vector treatment.

As a mouse model of achromatopsia, cpfl5 mice show selective loss of cone ERG responses. Scotopic (A) and photopic ERGs (B) were recorded at 1 and 3 months following treatment with AAV8 (Y447, 733F + T494V) (red), AAV2 (Y272, 444, 500, 730F + T491V) (green), or AAV8 (Y447, 733F) (blue). Recovered photopic ERG (cone-mediated) amplitudes were compared among groups (C). (D) Photopic-ERG dose response of the AAV-vector dilutions. The cpfl5 mice were treated in one eye at P14 and evaluated at 1 month after subretinal injection. Age-matched WT and untreated cpfl5 eyes were used as controls (n = 6 mice). P, postnatal day. *indicates P < 0.05, **indicates P < 0.01, ***indicates P < 0.001, NS = no statistical difference.

https://doi.org/10.1371/journal.pone.0188032.g001

Long-term (9 months) ERG rescue of AAV8 (Y447, 733F + T494V)-treated cpfl5 eyes

Among the three AAV vectors, AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 was selected for long-term evaluation. Thus, an ERG recording was repeated in the 9.5-month-old cpfl5 eyes. ERG data in each group were normally distributed (Shapiro-Wilk test, n = 6, P > 0.05). Untreated cpfl5 (566 ± 70 μV) and WT eyes (643 ± 64 μV) had similar scotopic b-wave amplitudes (n = 6, P > 0.05; Fig 2A and 2B). However, there was a decrease in scotopic responses in treated eyes (410 ± 54 μV) compared with untreated eyes (P < 0.01; Fig 2A and 2B). Rescue of photopic ERG responses was detected in the treated cpfl5 eyes, whereas ERG responses were nearly extinguished in the untreated cpfl5 eyes (Fig 2C). At 9 months post injection, photopic ERG b-waves in the treated cpfl5 maintained an average size of 68 ± 17 μV, about 65% of the WT level (101 ± 9 μV, n = 6, P < 0.001; Fig 2D). Furthermore, we analyzed implicit time of the rescued photopic b-wave. No significant difference was found between the treated cpfl5 (60 ± 5 ms) and WT eyes (56 ± 3 ms, n = 6, P > 0.05).

thumbnail
Fig 2. Long-term (9 months) electroretinographic assessment of treated cpfl5 eyes.

Cpfl5 eyes were treated with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 at P14. An ERG recording was repeated at 9 months following treatment. (A) Scotopic ERG elicited in treated cpfl5 eyes (red), compared to WT and untreated cpfl5 eyes (black). (B) Scotopic b-wave amplitudes elicited at 0 log cd-s/m2 intensity in the age-matched WT, treated, and untreated cpfl5 eyes (n = 6). (C) Photopic ERG elicited in treated cpfl5 eyes (red) compared to WT and untreated cpfl5 eyes (black). (D) Photopic b-wave amplitudes elicited at 1.0 log cd-s/m2 intensity in age-matched WT, treated, and untreated cpfl5 eyes (n = 6). P, postnatal day. **indicates P < 0.01, ***indicates P < 0.001, NS = no statistical difference.

https://doi.org/10.1371/journal.pone.0188032.g002

AAV-mediated CNGA3 expression in cpfl5 retinas

Mediated by the Y-F and/or T-V mutant AAV vectors, cone-specific transgene expression was driven by the IRBP/GNAT2 promoter [35]. For maximum rescue of cone photoreceptors, subretinal injections were performed at P14. At 3 and 9 months after treatment, CNGA3 expression was assayed by immunohistochemistry. At 3 months following subretinal injection, CNGA3 staining was detected primarily in the photoreceptor outer segment (OS) layer of cpfl5 retinas (Fig 3). Strong CNGA3 immunostaining was detected in the AAV8 (Y447, 733F + T494V)-treated group to a similar extent as in the age-matched WT retina. However, the immunostaining was relatively weaker in the AAV2 (Y272, 444, 500, 730F + T491V)- and AAV8 (Y447, 733F)-mediated groups. No CNGA3 expression was detected in the partner untreated retina from the same cpfl5 mouse (Fig 3). In the AAV8 (Y447, 733F + T494V)-mediated group, transgene expression was also detected at 9 months after treatment (Fig 4). These are the longest follow-up data for CNGA3 expression thus far.

thumbnail
Fig 3. Immuno-staining of retinal CNGA3 at 3 months post-injection.

CNGA3 expression (red) was detected primarily in the photoreceptor OS following treatment with AAV8 (Y447, 733F + T494V), AAV2 (Y272, 444, 500, 730F + T491V), or AAV8 (Y447, 733F). Age-matched WT and untreated cpfl5 retinas were used as controls. OS, outer segment; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer.

https://doi.org/10.1371/journal.pone.0188032.g003

thumbnail
Fig 4. Long-term retinal CNGA3 expression mediated by AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 treatment.

At 9 months following subretinal injection, retinal cryosections were immunostained with anti-CNGA3 antibody. CNGA3 expression (red) was detected primarily in the photoreceptor OS (arrows) in the AAV8 (Y447, 733F + T494V)-treated cpfl5 retinas and age-matched WT controls, but not in the untreated cpfl5 retinas. OS, outer segment; IS, inner segment; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole) (blue).

https://doi.org/10.1371/journal.pone.0188032.g004

Long-term rescue of cone opsins following AAV8 (Y447, 733F + T494V) treatment

CNGA3 deficiency has been shown to impair expression and localization of cone opsins, ultimately leading to cone photoreceptor death in Cnga3-/- and cpfl5 mice [11,41]. Compared to age-matched WT retinas, untreated cpfl5 retinas showed similar layers of photoreceptor nuclei and lengths of OS (Figs 5 and 6, right column). In untreated cpfl5 cones, loss of S-opsin proceeded more rapidly than M-opsin [11]. By 9.5-months of age, no retinal S-opsin was detected (Fig 6, bottom row) and there were only a few residual cones in the superior retina with mislocalization of M-opsin (Fig 5, bottom row). We used PNA to detect the interphotoreceptor matrix sheath, which surrounds the cone OS. In treated cpfl5 retinas, fluorescent microscopy at low magnification revealed preservation of M-opsin over most of the eyecup (Fig 5, second row). However, untreated retinas showed only some residual cones in which M-opsin had mislocalized to the inner segment, cone nuclei, and cone pedicles (Fig 5, bottom row). Double staining of M-opsin and cone-specific PNA (high magnification) suggests that the rescued M-opsin in treated retinas is located in the cone OS, consistent with that of WT controls (Fig 5, upper row). In addition, S-opsin was also in the cone OS of treated cpfl5 retinas (Fig 6). In summary, AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 treatment at P14 maintained normal expression and distribution of cone opsins. Gene therapy using the AAV vector effectively prevented cone degeneration for at least 9 months.

thumbnail
Fig 5. Long-term preservation of retinal M-opsin and CNGA3 after treatment with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3.

At 9 months after P14 treatment, cpfl5 retinal immunostaining revealed normal expression and the cone OS distribution (merge) of M-opsin, compared with WT controls. In untreated cpfl5 eyes, the inferior retinas had little M-opsin and cone specific PNA staining, whereas the superior retinas showed mislocalization of M-opsin within residue cones (arrows). In WT and treated cpfl5 retinas, M-opsin was located in the cone OS. Red: M-opsin or CNGA3 staining; Green: cone-specific PNA staining; Blue: nuclei staining with DAPI (4′,6-diamidino-2-phenylindole). P, postnatal day.

https://doi.org/10.1371/journal.pone.0188032.g005

thumbnail
Fig 6. Long-term preservation of retinal S-opsin after treatment with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3.

At 9 months after treatment of P14 mice, cpfl5 retinal immunostaining revealed normal expression (red) and the cone OS distribution (merge) of S-opsin compared with WT controls. S-opsin was not detected in untreated cpfl5 retinas. In WT and treated cpfl5 retinas, S-opsin was located in the cone OS. Red: S-cone opsin staining; Green: cone-specific PNA staining; Blue: nuclei staining with DAPI (4′,6-diamidino-2-phenylindole). P, postnatal day.

https://doi.org/10.1371/journal.pone.0188032.g006

Retinal whole-mounts from the treated cpfl5 eyes showed obviously preserved M-cone opsins and S-cone opsins compared with the contralateral untreated eyes (Fig 7A). At a high magnification (×40), the two cone opsins were counted in the same field of ventral nasal retina (Fig 7B and 7C). At 3 months after treatment, the average cone opsin counts were about 80% of WT levels (P < 0.01). However, M- and S-cone opsins decreased about 15% at 9 months following subretinal treatment. In the untreated cpfl5 retinal whole-mounts, little M-cone or S-cone opsins expression was detected within the ventral nasal retinas.

thumbnail
Fig 7. M-cone and S-cone opsins preservation in retinal whole mounts after treatment.

At 3 and 9 months after P14 treatment with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3, M-cone opsins (red) and S-cone opsins (white) were imaged (A) and counted (B, C) in the same field of ventral nasal retina (n = 6). WT and untreated cpfl5 mice were used as controls. D, dorsal; V, ventral; T, temporal; N, nasal. P, postnatal day; M, months. *indicates P < 0.05, **indicates P < 0.01, ***indicates P < 0.001, NS = no statistical difference.

https://doi.org/10.1371/journal.pone.0188032.g007

AAV8 (Y447, 733F + T494V) rescues cone-mediated water maze behavior in the cpfl5 mouse

To determine whether the observed electrophysiological, biochemical, and morphological preservation/restoration of the cpfl5 retina following AAV8 (Y447, 733F + T494V) vector treatment led to improvement in behavioral performance, we tested mice in a visually guided behavior task (Fig 8), as described previously [40]. During formal tests, mice pupils were dilated and retinas were fully light-adapted to prevent rod intrusion. The time taken to escape to the platform under well-lit condition was compared among groups. The WT mice with unilateral eyelid suture took 10 ± 6 seconds, treated cpfl5 mice took 16 ± 6 seconds, and untreated cpfl5 mice took 29 ± 11 seconds to reach the platform. The data in each group were normally distributed (Shapiro-Wilk test, n = 6, P > 0.05). Statistical analysis showed significant improvement in the treated cpfl5 mice compared to the untreated cpfl5 group (n = 6, P < 0.05). No statistical difference in performance was found between treated cpfl5 and WT mice (P > 0.05). When the treated eyes of cpfl5 mice were closed by suturing their eyelids, time to reach the platform increased from 16 ± 6 to 28 ± 12 seconds. No statistical difference in performance was found between the untreated and treated cpfl5 mice when the treated eyes were sutured (P > 0.05).

thumbnail
Fig 8. Cone-mediated visually guided behavioral test after treatment with AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3.

#Statistical analysis indicates a significant difference in performance (P < 0.01) in the WT mice with unilateral eyelid suture compared to untreated cpfl5 mice and the treated cpfl5 mice when the treated eye was sutured. *Statistical analysis indicates a significant difference in performance (P < 0.05) in the treated cpfl5 mice compared to untreated cpfl5 mice and treated cpfl5 mice with the treated eye sutured. No statistical difference in performance was found between the WT mice with unilateral eyelid suture and treated cpfl5 mice or between untreated cpfl5 mice and treated cpfl5 mice when the treated eye was sutured (n = 6).

https://doi.org/10.1371/journal.pone.0188032.g008

Discussion

AAVs are non-pathogenic, single-stranded, DNA-packaging dependoparvoviruses within the Parvoviridae family. One of the major applications of AAVs is as a gene therapy vector to treat monogenic recessive blindness. For example, AAV serotype 2 has been adopted for ongoing RPE65-Leber congenital amaurosis (LCA2) gene therapy clinical trials [42]. Transgene expression efficiency of AAV vectors can be affected by phosphorylation of capsid surface-exposed residues, the phosphorylation ultimately leads to capsid ubiquitination and proteasomal degradation of AAV particles [31,32,43]. AAV2 variants containing capsid surface-exposed Y-F and/or T-V mutations protect vector particles from proteasomal degradation, thus significantly increasing the efficiency of retinal transduction [30,33,34,44]. Inhibiting proteasomal degradation may improve longevity of transgene expression. To test this, we investigated AAV variants from serotypes 2 and 8, the latter of which was recently reported to have higher photoreceptor transduction efficiency than wild-type AAV2 and 5 [45,46]. We compared rescue differences between three cpfl5 groups, in which treatments were mediated by AAV8 (Y447, 733F + T494V), AAV8 (Y447, 733F), or AAV2 (Y272, 444, 500, 730F + T491V) vectors. At 1 month after treatment, photopic ERG responses were higher in AAV8 (Y447, 733F + T494V) and AAV2 (Y272, 444, 500, 730F + T491V) groups. Vectors with a combination of Y-F and T-V mutations seemed to lead to better ERG rescue than vectors with only a Y-F mutation. At 3 months after treatment, photopic responses of AAV2 (Y272, 444, 500, 730F + T491V)- and AAV8 (Y447, 733F)-treated retinas decreased, while the responses in those treated with AAV8 (Y447, 733F + T494V) were sustained or continued to rise slightly. Treated with the three AAV mutants, preservation of photopic-ERG responses was almost dose dependent across doses between 1E11 and 1E13 vg/ml. Among the original and diluted solutions, a dose at 1E13 produced the best improvement (Fig 1C and 1D). In summary, an AAV8 vector with both Y-F and T-V mutations gave the best preservation of cone ERG responses.

Besides design considerations to prevent proteasomal degradation, these results highlight the need for careful consideration of vector serotype in any therapeutic AAV vector platform. AAV8 vector can target a variety of retinal cells, including retinal pigment epithelial cells and photoreceptors, and its transfection is believed to be safe in mice, dogs and nonhuman primates [47]. Studies have showed that AAV8 has higher photoreceptor transduction efficiency than AAV2 and 5 [35]. Using WT or capsid mutant AAV8 vectors, gene therapy has successfully treated animal models of Leber’s congenital amaurosis, autosomal recessive retinitis pigmentosa and achromatopsia [35,45,46,48,49]. Compared to AAV2, AAV8 is able to achieve equivalent expression at lower dose [47,48]. In addition, transgene expression mediated by AAV8 is much sooner than AAV2 and 5 [45,47,48]. An AAV8 tyrosine-capsid mutant can confer more effective therapy than that of a standard AAV vector in an animal model with early-onset rapid retinal degeneration [45]. Besides AAV2, AAV8 can be another promising vector for human clinical gene therapy trials for choroideremia and even in the future for other retinal degenerative disorders [48]. Here, long-term Validation of capsid mutants AAV8 vector for CNGA3-achromatopsia has been demonstrated in a mouse model, but AAV vector performance in human patients could be different from mice. It is unlikely that concerns of long-term efficacy and safety can be sufficiently addressed in a small animal model. Thus, AAV8 or its mutants still need careful evaluation in large-animal models and/or non-human primates.

CNGA3-deficient retinas are characterized by early cone photoreceptor dysfunction, followed by rapid S-opsin loss, abnormal M-opsin distribution, and ultimately, cone cells death [11,41]. Three-week-old cpfl5 retinas exhibit normal M-opsin staining in the cone OS, but already show a decrease in S-opsin [11]. By 10 weeks, little S-opsin staining is detected and all M-opsin has mislocalized to the cone inner segment, nuclei, and pedicles [11]. As to the 9.5-month-old cpfl5 mice, we found a loss of the mislocalized M-opsin in the inferior retinas (Fig 5).

Here, gene therapy through subretinal injection was performed at P14 to achieve maximum rescue of cone photoreceptors. In mice younger than P14, especially within one week after birth, trans-corneal subretinal injection could cause severe damage to cornea, iris, lens, and retina because of the smaller eyeballs, underdeveloped cornea and lens, and difficulty in achieving excessive dilatation of the pupil, as described previously [36]. Additionally, it is difficult to detach a significant fraction of the mouse retina via subretinal injection prior to P14 [38]. We have previously found that subretinally injected vectors localized around the injection area in P10 mice and ultimate coverage of the vector solution was no more than one-third of the whole retina, an area much smaller than P14 injections [38].

AAV vectors containing the IRBP/GNAT2 promoter exhibit cone photoreceptor-specific transgene expression [35]. Using AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3, we obtained a more robust rescue than using AAV2 (Y272, 444, 500, 730F + T491V) or AAV8 (Y447, 733F) vectors in cpfl5 mouse, a spontaneous animal model of achromatopsia with a Cnga3 mutation. Photopic ERG responses in the mouse retina were rescued and maintained to an average of 2/3 of the WT mouse responses at 9 months after injection of AAV8 (Y447, 733F + T494V). Moreover, the rescued cone-ERG responses showed no implicit time delay. These data are from the longest follow-up period reported thus far. In our previous work, the long-term cone-ERG responses of a WT AAV5 vector treatment were significantly lower than those of the AAV8 triple-mutant vector treatment. The water maze visually guided behavioral test has been used in dark conditions to examine AAV-mediated rod function recovery in rd12 mice [40]. Here, we extended this application to test cone function in pupil-dilated and fully light-adapted cpfl5 mice. In addition to functional outcomes, our data show biochemical and structural preservation/restoration following AAV8 (Y447, 733F + T494V) vector treatment.

Cone-mediated ERG responses decreased about 20% from 3 to 9 months after treatment with the AAV8 (Y447, 733F + T494V) vector. In accordance with the deterioration of cone-ERG responses, the two cone opsins decreased about 15% at 9 months. This is likely due to inadequate expression of CNGA3 in some cone photoreceptors. Relative to 3 months following AAV8 (Y447, 733F + T494V) treatment, the immunostaining of rescued CNGA3 seemed to be lesser and weaker at 9 months (Figs 3 and 4). Additionally, the decrease upon treatment may be partly caused by a damaging effect of subretinal injection. In human retinas, cone density is very high in the central macula. Retinal detachment of the fovea, caused by subretinal injection, may cause more damage than benefit [30]. Rod-mediated ERG responses showed a reduction of b-wave amplitudes in treated compared with untreated eyes, similar as described previously [11]. The decrease upon treatment may be a consequence of injection-related damage.

Recently, successful gene therapies have been reported in CNGA3-deficient sheep, a large-animal model of achromatopsia [5052]. In these studies, AAV serotype 2 and 5 vectors carry Cnga3 gene under control of a human red/green cone opsin promoter. The results of a large-animal model provide an important data base for gene therapy in CNGA3-achromatopsia patients.

Here, we show that gene therapy mediated by AAV8 (Y447, 733F + T494V)-IRBP/GNAT2-Cnga3 can effectively rescue cone degeneration in the retinas of cpfl5 mice for at least 9 months. To our knowledge, this is the longest and most significant rescue reported so far in the mouse model of Cnga3-associated achromatopsia.

Acknowledgments

We thank Xiaogang Chen for helpful assistance and advice on cryosectioning.

References

  1. 1. Kohl S, Baumann B, Rosenberg T, Kellner U, Lorenz B, Vadalà M, et al. Mutations in the cone photoreceptor G-protein alpha-subunit gene GNAT2 in patients with achromatopsia. Am J Hum Genet. 2002; 71:422–425. pmid:12077706
  2. 2. Bainbridge JW, Smith AJ, Barker SS, Robbie S, Henderson R, Balaggan K, et al. Effect of gene therapy on visual function in Leber's congenital amaurosis. N Engl J Med. 2008; 358:2231–2239. pmid:18441371
  3. 3. Cideciyan AV, Aleman TS, Boye SL, Schwartz SB, Kaushal S, Roman AJ, et al. Human gene therapy for RPE65 isomerase deficiency activates the retinoid cycle of vision but with slow rod kinetics. Proc Natl Acad Sci U S A. 2008; 105:15112–15117. pmid:18809924
  4. 4. Maguire AM, High KA, Auricchio A, Wright JF, Pierce EA, Testa F, et al. Age-dependent effects of RPE65 gene therapy for Leber's congenital amaurosis: a phase 1 dose-escalation trial. Lancet. 2009; 374:1597–1605. pmid:19854499
  5. 5. Cideciyan AV, Hauswirth WW, Aleman TS, Kaushal S, Schwartz SB, Boye SL, et al. Human RPE65 gene therapy for Leber congenital amaurosis: persistence of early visual improvements and safety at 1 year. Hum Gene Ther. 2009; 20:999–1004. pmid:19583479
  6. 6. Simonelli F, Maguire AM, Testa F, Pierce EA, Mingozzi F, Bennicelli JL, et al. Gene therapy for Leber's congenital amaurosis is safe and effective through 1.5 years after vector administration. Mol Ther. 2010; 18:643–650. pmid:19953081
  7. 7. Jacobson SG, Cideciyan AV, Ratnakaram R, Heon E, Schwartz SB, Roman AJ, et al. Gene therapy for leber congenital amaurosis caused by RPE65 mutations: safety and efficacy in 15 children and adults followed up to 3 years. Arch Ophthalmol. 2012; 130:9–24. pmid:21911650
  8. 8. Bainbridge JW, Mehat MS, Sundaram V, Robbie SJ, Barker SE, Ripamonti C, et al. Long-term effect of gene therapy on Leber's congenital amaurosis. N Engl J Med. 2015; 372:1887–1897. pmid:25938638
  9. 9. MacLaren RE, Groppe M, Barnard AR, Cottriall CL, Tolmachova T, Seymour L, et al. Retinal gene therapy in patients with choroideremia: initial findings from a phase 1/2 clinical trial. Lancet. 2014; 383:1129–1137. pmid:24439297
  10. 10. Michalakis S, Schön C, Becirovic E, Biel M. Gene Therapy for Achromatopsia. J Gene Med. 2017; 19(3).
  11. 11. Pang JJ, Deng WT, Dai X, Lei B, Everhart D, Umino Y, et al. AAV-mediated cone rescue in a naturally occurring mouse model of CNGA3-achromatopsia. PLoS One. 2012; 7:e35250. pmid:22509403
  12. 12. Kohl S, Coppieters F, Meire F, Schaich S, Roosing S, Brennenstuhl C, et al. A nonsense mutation in PDE6H causes autosomal-recessive incomplete achromatopsia. Am J Hum Genet. 2012; 91:527–532. pmid:22901948
  13. 13. Brennenstuhl C, Tanimoto N, Burkard M, Wagner R, Bolz S, Trifunovic D, et al. Targeted ablation of the Pde6h gene in mice reveals cross-species differences in cone and rod phototransduction protein isoform inventory. J Biol Chem. 2015; 290:10242–10255. pmid:25739440
  14. 14. Kohl S, Zobor D, Chiang WC, Weisschuh N, Staller J, Gonzalez Menendez I, et al. Mutations in the unfolded protein response regulator ATF6 cause the cone dysfunction disorder achromatopsia. Nat Genet. 2015; 47:757–765. pmid:26029869
  15. 15. Ansar M, Santos-Cortez RL, Saqib MA, Zulfiqar F, Lee K, Ashraf NM, et al. Mutation of ATF6 causes autosomal recessive achromatopsia. Hum Genet. 2015; 134:941–950. pmid:26063662
  16. 16. Kohl S, Marx T, Giddings I, Jägle H, Jacobson SG, Apfelstedt-Sylla E, et al. Total colourblindness is caused by mutations in the gene encoding the alpha-subunit of the cone photoreceptor cGMP-gated cation channel. Nat Genet. 1998; 19:257–259. pmid:9662398
  17. 17. Kuniyoshi K, Muraki-Oda S, Ueyama H, Toyoda F, Sakuramoto H, Ogita H, et al. Novel mutations in the gene for α-subunit of retinal cone cyclic nucleotide-gated channels in a Japanese patient with congenital achromatopsia. Jpn J Ophthalmol. 2016; 60:187–197. pmid:27040408
  18. 18. Kohl S, Baumann B, Broghammer M, Jägle H, Sieving P, Kellner U, et al. Mutations in the CNGB3 gene encoding the beta-subunit of the cone photoreceptor cGMP-gated channel are responsible for achromatopsia (ACHM3) linked to chromosome 8q21. Hum Mol Genet. 2000; 9:2107–2116. pmid:10958649
  19. 19. Wawrocka A, Kohl S, Baumann B, Walczak-Sztulpa J, Wicher K, Skorczyk-Werner A, et al. Five novel CNGB3 gene mutations in Polish patients with achromatopsia. Mol Vis. 2014; 20:1732–1739. pmid:25558176
  20. 20. Ouechtati F, Merdassi A, Bouyacoub Y, Largueche L, Derouiche K, Ouragini H, et al. Clinical and genetic investigation of a large Tunisian family with complete achromatopsia: identification of a new nonsense mutation in GNAT2 gene. J Hum Genet. 2011; 56:22–28. pmid:21107338
  21. 21. Thiadens AA, den Hollander AI, Roosing S, Nabuurs SB, Zekveld-Vroon RC, Collin RW, et al. Homozygosity mapping reveals PDE6C mutations in patients with early-onset cone photoreceptor disorders. Am J Hum Genet. 2009; 85:240–247. pmid:19615668
  22. 22. Katagiri S, Hayashi T, Yoshitake K, Sergeev Y, Akahori M, Furuno M, et al. Congenital Achromatopsia and Macular Atrophy Caused by a Novel Recessive PDE6C Mutation (p.E591K). Ophthalmic Genet. 2015; 36:137–144. pmid:25605338
  23. 23. Wissinger B, Gamer D, Jägle H, Giorda R, Marx T, Mayer S, et al. CNGA3 mutations in hereditary cone photoreceptor disorders. Am J Hum Genet. 2001; 69:722–737. pmid:11536077
  24. 24. Kohl S, Varsanyi B, Antunes GA, Baumann B, Hoyng CB, Jägle H, et al. CNGB3 mutations account for 50% of all cases with autosomal recessive achromatopsia. Eur J Hum Genet. 2005; 13:302–308. pmid:15657609
  25. 25. Remmer MH, Rastogi N, Ranka MP, Ceisler EJ. Achromatopsia: a review. Curr Opin Ophthalmol. 2015; 26:333–340. pmid:26196097
  26. 26. Pang JJ, Lei L, Dai X, Shi W, Liu X, Dinculescu A, et al. AAV-mediated gene therapy in mouse models of recessive retinal degeneration. Curr Mol Med. 2012; 12:316–330. pmid:22300136
  27. 27. Pang JJ, Lauramore A, Deng WT, Li Q, Doyle TJ, Chiodo V, et al. Comparative analysis of in vivo and in vitro AAV vector transduction in the neonatal mouse retina: effects of serotype and site of administration. Vision Res. 2008; 48:377–385. pmid:17950399
  28. 28. Pang JJ, Alexander J, Lei B, Deng W, Zhang K, Li Q, et al. Achromatopsia as a potential candidate for gene therapy. Adv Exp Med Biol. 2010; 664:639–646. pmid:20238068
  29. 29. Michalakis S, Mühlfriedel R, Tanimoto N, Krishnamoorthy V, Koch S, Fischer MD, et al. Restoration of cone vision in the CNGA3-/- mouse model of congenital complete lack of cone photoreceptor function. Mol Ther. 2010; 18:2057–2063. pmid:20628362
  30. 30. Kay CN, Ryals RC, Aslanidi GV, Min SH, Ruan Q, Sun J, et al. Targeting photoreceptors via intravitreal delivery using novel, capsid-mutated AAV vectors. PLoS One. 2013; 8:e62097. pmid:23637972
  31. 31. Zhong L, Zhao W, Wu J, Li B, Zolotukhin S, Govindasamy L, et al. A dual role of EGFR protein tyrosine kinase signaling in ubiquitination of AAV2 capsids and viral second-strand DNA synthesis. Mol Ther. 2007; 15:1323–1330. pmid:17440440
  32. 32. Petrs-Silva H, Dinculescu A, Li Q, Min SH, Chiodo V, Pang JJ, et al. High-efficiency transduction of the mouse retina by tyrosine-mutant AAV serotype vectors. Mol Ther. 2009; 17:463–471. pmid:19066593
  33. 33. Aslanidi GV, Rivers AE, Ortiz L, Song L, Ling C, Govindasamy L, et al. Optimization of the capsid of recombinant adeno-associated virus 2 (AAV2) vectors: the final threshold? PLoS One. 2013; 8:e59142. pmid:23527116
  34. 34. Boye SL, Bennett A, Scalabrino ML, McCullough KT, Van Vliet K, Choudhury S, et al. Impact of Heparan Sulfate Binding on Transduction of Retina by Recombinant Adeno-Associated Virus Vectors.J Virol. 2016; 90:4215–4231. pmid:26865709
  35. 35. Du W, Tao Y, Deng WT, Zhu P, Li J, Dai X, et al. Vitreal delivery of AAV vectored Cnga3 restores cone function in CNGA3-/-/Nrl-/- mice, an all-cone model of CNGA3 achromatopsia. Hum Mol Genet. 2015; 24:3699–3707. pmid:25855802
  36. 36. Qi Y, Dai X, Zhang H, He Y, Zhang Y, Han J, et al. Trans-Corneal Subretinal Injection in Mice and Its Effect on the Function and Morphology of the Retina. PLoS One. 2015; 10:e0136523. pmid:26317758
  37. 37. Dai X, Zhang H, He Y, Qi Y, Chang B, Pang JJ. et al. The frequency-response electroretinogram distinguishes cone and abnormal rod function in rd12 mice. PLoS One. 2015; 10:e0117570. pmid:25706871
  38. 38. Dai X, Zhang H, Han J, He Y, Zhang Y, Qi Y, et al. Effects of Subretinal Gene Transfer at Different Time Points in a Mouse Model of Retinal Degeneration. PLoS One. 2016; 11:e0156542. pmid:27228218
  39. 39. Li X, Li W, Dai X, Kong F, Zheng Q, Zhou X, et al. Gene therapy rescues cone structure and function in the 3-month-old rd12 mouse: a model for midcourse RPE65 leber congenital amaurosis. Invest Ophthalmol Vis Sci. 2011; 52:7–15. pmid:21169527
  40. 40. Pang JJ, Chang B, Kumar A, Nusinowitz S, Noorwez SM, Li J, et al. Gene therapy restores vision-dependent behavior as well as retinal structure and function in a mouse model of RPE65 Leber congenital amaurosis. Mol Ther. 2006; 13:565–572. pmid:16223604
  41. 41. Michalakis S, Geiger H, Haverkamp S, Hofmann F, Gerstner A, Biel M. Impaired opsin targeting and cone photoreceptor migration in the retina of mice lacking the cyclic nucleotide-gated channel CNGA3. Invest Ophthalmol Vis Sci. 2005; 46:1516–1524. pmid:15790924
  42. 42. Schimmer J, Breazzano S. Investor Outlook: Significance of the Positive LCA2 Gene Therapy Phase III Results. Hum Gene Ther Clin Dev. 2015; 26:208–210. pmid:26684444
  43. 43. Zhong L, Li B, Mah CS, Govindasamy L, Agbandje-McKenna M, Cooper M, et al. Next generation of adeno-associated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proc Natl Acad Sci U S A. 2008; 105:7827–7832. pmid:18511559
  44. 44. Petrs-Silva H, Dinculescu A, Li Q, Deng WT, Pang JJ, Min SH, et al. Novel properties of tyrosine-mutant AAV2 vectors in the mouse retina. Mol Ther. 2011; 19:293–301. pmid:21045809
  45. 45. Pang JJ, Dai X, Boye SE, Barone I, Boye SL, Mao S, et al. Long-term retinal function and structure rescue using capsid mutant AAV8 vector in the rd10 mouse, a model of recessive retinitis pigmentosa. Mol Ther. 2011; 19:234–242. pmid:21139570
  46. 46. Dai X, Han J, Qi Y, Zhang H, Xiang L, Lv J, et al. AAV-mediated lysophosphatidylcholine acyltransferase 1 (Lpcat1) gene replacement therapy rescues retinal degeneration in rd11 mice. Invest Ophthalmol Vis Sci. 2014; 55:1724–1734. pmid:24557352
  47. 47. Vandenberghe LH, Bell P, Maguire AM, Cearley CN, Xiao R, Calcedo R, et al. Dosage thresholds for AAV2 and AAV8 photoreceptor gene therapy in monkey. Sci Transl Med. 2011; 3:88ra54. pmid:21697530
  48. 48. Black A, Vasireddy V, Chung DC, Maguire AM, Gaddameedi R, Tolmachova T, et al. Adeno-associated virus 8-mediated gene therapy for choroideremia: preclinical studies in in vitro and in vivo models. J Gene Med. 2014; 16:122–130. pmid:24962736
  49. 49. Deng WT, Dinculescu A, Li Q, Boye SL, Li J, Gorbatyuk MS, et al. Tyrosine-mutant AAV8 delivery of human MERTK provides long-term retinal preservation in RCS rats. Invest Ophthalmol Vis Sci. 2012; 53:1895–1904. pmid:22408006
  50. 50. Banin E, Gootwine E, Obolensky A, Ezra-Elia R, Ejzenberg A, Zelinger L, et al. Gene Augmentation Therapy Restores Retinal Function and Visual Behavior in a Sheep Model of CNGA3 Achromatopsia. Mol Ther. 2015; 23:1423–1433. pmid:26087757
  51. 51. Gootwine E, Abu-Siam M, Obolensky A, Rosov A, Honig H, Nitzan T, et al. Gene Augmentation Therapy for a Missense Substitution in the cGMP-Binding Domain of Ovine CNGA3 Gene Restores Vision in Day-Blind Sheep. Invest Ophthalmol Vis Sci. 2017; 58:1577–1584. pmid:28282490
  52. 52. Gootwine E, Ofri R, Banin E, Obolensky A, Averbukh E, Ezra-Elia R, et al. Safety and Efficacy Evaluation of rAAV2tYF-PR1.7-hCNGA3 Vector Delivered by Subretinal Injection in CNGA3 Mutant Achromatopsia Sheep. Hum Gene Ther Clin Dev. 2017; 28:96–107. pmid:28478700