Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 1.
Published in final edited form as: Exp Eye Res. 2022 Mar 1;218:109012. doi: 10.1016/j.exer.2022.109012

Long-term retinal protection by MEK inhibition in Pax6 haploinsufficiency mice

James D Cole 1,2,*, Kara M McHaney 1,*, Behnam Rabiee 3,4,*, Jingyi Gao 1, Carlos Rodriguez 1, David A Miller 5, Mingna Liu 1, Marta Grannonico 1, Pedro Norat 1, Hao F Zhang 5, Ali R Djalilian 3,†, Xiaorong Liu 1,6,7,†
PMCID: PMC9050935  NIHMSID: NIHMS1786632  PMID: 35245513

Abstract

Aniridia is a panocular condition characterized by impaired eye development and vision, which is mainly due to the haploinsufficiency of the paired-box-6 (PAX6) gene. Like what is seen in aniridia patients, Pax6-deficient mice Pax6Sey-Neu/+ exhibit a varied degree of ocular damage and impaired vision. Our previous studies showed that theses phenotypes were partially rescued by PD0325901, a mitogen-activated protein kinase kinase (MEK or MAP2K) inhibitor. In this study, we assessed the long-term efficacy of PD0325901 treatment in retinal health and visual behavior. At about one year after the postnatal treatment with PD0325901, Pax6Sey-Neu/+ mice showed robust improvements in retina size and visual acuity, and the elevated intraocular pressure (IOP) was also alleviated, compared to age-matched mice treated with vehicles only. Moreover, the Pax6Sey-Neu/+ eyes showed disorganized retinal ganglion cell (RGC) axon bundles and retinal layers, which we termed as hotspots. We found that the PD treatment reduced the number and size of hotspots in the Pax6Sey-Neu/+ retinas. Taken together, our results suggest that PD0325901 may serve as an efficacious intervention in protecting retina and visual function in aniridia-afflicted subjects.

Keywords: Aniridia, Pax6, MEK inhibitor, PD0325901, retinal damage, long-term neuroprotection

1. Introduction

Aniridia is a panocular developmental eye disorder that causes varied visual impairments in young patients (Glaser et al., 1994; Graw et al., 2005; Gregory-Evans et al., 2014; Hingorani et al., 2012; Lee et al., 2008; Nishina et al., 1999; Simpson and Price, 2002; Yokoi et al., 2016)(add your reivew). In about 90% of cases, aniridia is associated with mutations in the PAX6 gene, which encodes for a transcription factor necessary for regulating eye and brain development, as well as olfactory and pancreatic development (Cvekl et al., 2004; Glaser et al., 1994; Graw et al., 2005; Hingorani et al., 2012; Lima Cunha et al., 2019; Nishina et al., 1999; Samant et al., 2016; Wang et al., 2017b; Warren et al.; Yasue et al., 2017). Previous studies suggest that compensation for the loss of Pax6 protein dosage during early development could counteract the potential ocular and systemic abnormalities (Rabiee et al., 2020; Yasue et al., 2017). For example, we recently demonstrated that treatment of newborn Pax6-deficient mice with topical or systemic PD0325901 led to significantly increased corneal Pax6 levels, improved corneal morphology, and enhanced ocular function (Cole J D, 2021; Rabiee et al., 2020).

However, much remains to be characterized about the long-term protective effects this treatment has on retinal health and visual behavior. As retinal damage is one of the primary contributors to visual deficits in aniridia, it is crucial to monitor the retinal morphology and function both during and after PD0325901 treatment. Most, if not all, aniridia patients exhibit varying degrees of foveal hypoplasia, and about 10% of patients have optic nerve hypoplasia (Gregory-Evans et al., 2014; McCulley et al., 2005; Sannan et al., 2017), suggesting uneven or localized damage of varying severity among individual patients. Likewise, an increase in intraocular pressure (IOP) is present in about 50% to 85% of aniridia patients, but has not been examined in mice (Cole J D, 2021; Netland et al., 2011). The elevated IOP leads to diminished vision via the loss of RGCs as subjects age (Cole J D, 2021; Hingorani et al., 2012; Lee et al., 2008; Netland et al., 2011; Samant et al., 2016).

In this study, we characterized the long-term protective effects of PD0325901, a mitogen-activated protein kinase kinase inhibitor, on the retina and vision in adult Pax6Sey-Neu/+ mice. Previous studies have found that MAPK/ERK signaling regulates the transcription factors involved in early development, including Pax6 (de la Puente et al., 2016; Rabiee et al., 2020; Solberg et al., 2019). We thus sought to determine whether inhibiting MEK pathways via PD0325901 may compensate for the retinal damage induced by the reduced dosages of Pax6 in heterozygous mutants. At one year after the postnatal treatment of PD0325901, we measured IOP, size of the retina, and visual acuity. In addition, we characterized retinal damage in flat-mounted retinas and histological sections and found that topical application of PD0325901 from P5 to P30 provides robust protection against aniridic retinal damage.

2. Materials and Methods

2.1. Mice and PD0325901 treatments

Pax6Sey-Neu/+ mice with a Balb-c background were used in this study (Rabiee et al., 2020). These mice were bred by crossing heterozygous mutants with wildtype (WT), and pups were genotyped according to the published protocol (Rabiee et al., 2020). The animal protocols were approved by the Committee on the Ethics of Animal Experiments of the University of Illinois at Chicago and the University of Virginia. All procedures were conducted in compliance with the recommendations of the Association for Research in Vision and Ophthalmology (ARVO) and the National Institutes of Health (NIH) guidelines.

Topical administration of PD0325901 (abbreviated as PD) consisted of 1mM PD0325901, 2% dimethyl sulfoxide (DMSO), and 2% hydroxypropyl methylcellulose in phosphate buffered saline (PBS) (Rabiee et al., 2020). Topical treatment of PD or vehicle was applied once a day to both eyes for 5 days, and this schedule was repeated after a two-day break, starting at postnatal day (P) 5 up to P30 (LoRusso et al., 2010; Rabiee et al., 2020). Formulations were applied directly to the palpebral membrane prior to eye opening, which were able to penetrate this membrane and translocate to the eye (Rabiee et al., 2020). The last PD treatment was at P30, and all the following measurements described below were carried out at 10-14 months of age.

2.2. Retina Size, Intraocular Pressure (IOP), and Optomotor Tests

The widths of flat-mounted retinas were measured using the average of the widths of the superior-inferior and nasal-temporal leaflets of each retina. Mice were restrained using plastic sleeves without anesthesia, and the intraocular pressure (IOP) was measured using a TonoLab rebound tonometer (Chen et al., 2015; Cole J D, 2021; Feng et al., 2013a; Feng et al., 2016; Puyang et al., 2016; Yi et al., 2016). Optomotor response was tested using the PhenoSys qOMR system (PhenoSys, GmbH, Berlin, Germany). Because the wildtype mice with albino Balb-c background exhibited a severely reduced visual acuity (Rabiee et al., 2020; Yeritsyan et al., 2012), we counted the number of optomotor responses elicited at a single frequency, one known to induce robust responses in WT Balb-c mice (Rabiee et al., 2020). In brief, mice were adapted to the system for 5 minutes, then gratings were presented for 2 minutes at the frequency of 0.10 cycles/degree rotating clockwise and counterclockwise every 10 seconds (Rabiee et al., 2020; Rangarajan et al., 2011). The number of optomotor responses elicited during this time was tallied for each mouse. A mouse’s response was considered “robust” if it elicited four or more optomotor responses, “moderate” if it elicited a response one to three times, and “none” if the mouse never responded to the gratings (Rabiee et al., 2020).

Observers were blind to treatment/genotype groups before and during the experiments for all above procedures. Mouse genotype and treatment were revealed afterwards for data analysis purpose.

2.3. Immunohistochemistry and Confocal imaging

Mice were euthanized with Euthasol (15.6 mg/mL; Virbac, Greely, CO, USA) and perfused with 4% paraformaldehyde (PFA), then eyes were dissected and prepared for immunohistochemistry (Feng et al., 2016; Feng et al., 2013b; Miller et al., 2020). Retinal eye cups were cut into a 4-leaf clover shape for flat-mounting (see Fig 1C) or prepared for cryo-sectioning at 20μm thickness. Antibodies include anti-Tuj1 preconjugated with Alexa Fluor-488 (1:1000; BioLegend, San Diego, CA), anti-rbpms (1:250; Abcam, Waltham, MA), anti-choline acetyltransferase (ChAT,1:250; ThermoFisher Scientific, Waltham, MA), anti-Calbindin (1:1000, Sigma-Aldrich, St. Louis, MO), anti-glial fibrillary acidic protein (GFAP, 1:200; Abcam, Waltham, MA), anti-ionized calcium binding adaptor molecule 1 (Iba1, 1:250; Abcam, Waltham, MA), and 4′,6-diamidino-2-phenylindole (DAPI, VECTASHIELD®, Vector Laboratories)(Feng et al., 2016; Grannonico et al., 2021; Miller et al., 2020). Confocal images were performed using a Zeiss LSM 800 microscope (Carl Zeiss AG, Oberkochen, Germany)(Grannonico et al., 2021; Miller et al., 2020).

Fig. 1. Long-term protective effects by PD treatment on IOP and the size of the retina of Pax6Sey-Neu/+ mice.

Fig. 1.

Open in a new tab

(A) Timeline of experimental procedures. (B) IOP assessments of WT-Veh, WT-PD, Pax6-Veh, and Pax6-PD by a Tonolab tonometer (see inset). Pax6Sey-Neu/+ was abbreviated to Pax6, Vehicle group was abbreviated to Veh, and PD0325901-treated was abbreviated as PD. (C) Comparison of retinal flat-mount width as indicated in the inset. Sample numbers labeled at the bottom of each bar. ***: P<0.001; ****:P<0.0001 in one-way ANOVA post hoc Tukey’s test.

2.4. Quantification of hotspots, RGC axon bundle morphology, and retinal layer thickness

On flat-mount images, hotspots were defined as areas with disorganized RGC axon bundles. The hotspot was outlined based on local characteristics of the RGC axon bundles and the underlying ganglion cell layer (GCL), including: misaligned or disorganized RGC axon bundles comparing the surrounding axon bundles going straight towards to the optic nerve head (ONH), or that the bundles were displaced around a central bulge, and an abnormal density in the GCL. To quantify hotspot size, we used the contour measurement tool in Zeiss Zen to trace around regions with disorganized axon bundles and clumped somas in the GCL (also see Fig. 4, red outlines). The traced areas were cross-examined independently by two observers.

Fig. 4. PD treatment rescued the optic nerve head phenotype in Pax6Sey-Neu/+ mice.

Fig. 4.

Open in a new tab

(A) Confocal images of flat-mounted retinas immunostained by Tuj1-488 (green) for RGC axon bundles and DAPI (blue) in control, Pax6-Veh, and Pax6-PD mice. Yellow box shows the optic nerve head (ONH), where axon bundles converge to form the optic nerve as they leave the retina. Pax6-Veh exhibited axon bundle disruption while the PD treatment partially protected the normal phenotype. DAPI stain (blue) shows increased disorganization and displacement of somas around the ONH of Pax6-Veh, while PD treatment appears to have reduced this phenotype. Scale bar in top row = 500μm. Scale bar in bottom three rows = 100μm.

For all sagittal sections, ONH was used as a landmark. The superior end of the eye was also notched during eye dissection for orientation. Regions within the radius of 600μm from the ONH were used for quantification of layer thickness, with three measurements being made on each side of the ONH for a total of six per retina, and the average of the six measurements was used as the layer thickness. Similarly, hotspots in sagittal sections were determined based on the following criteria: abnormal layer thicknesses, and/or absences of soma stacking in the INL and ONL, wherein cells are distributed in disorganized fashion. For each retina, the thicknesses of normal and hotspot regions were measured separately.

To examine the overall changes in RGCs, the axon bundle number per retina and axon bundle width was measured using a custom MatLab program (Grannonico et al., 2021; Miller et al., 2020). In brief, we measured the number of axons bundles at 400μm radius from the ONH (Miller et al., 2020). Individual axon bundles were marked, and bundle width was automatically measured by the program according to the published procedure (Miller et al., 2020).

2.5. Statistical analysis

Statistical tests were performed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA). Multiple samples were compared using one-way ANOVA followed by post-hoc Tukey multiple comparison tests. Chi-square test was used to compare multiple parameters in the optomotor assessments. Results were reported as mean±standard error of mean (SEM) except in the analysis of the axon bundle width distribution where standard deviation was used (SD).

3. Results

3.1. MEK inhibitor PD0325901 increases retina size and reduces elevated IOP of Pax6Sey-Neu/+ mice

Assessment of long-term protective effects by PD0325901 (abbreviated as PD) in Pax6Sey-Neu/+ mice (abbreviated as Pax6 mice) began at P200, about 6 months following topical treatments of PD and vehicle controls (Veh) from P5 to P30 (Fig. 1A). First, the PD treatment reduced the elevated IOP in the Pax6 mice (Fig. 1B). The wild type (WT) mice treated by vehicle (WT-Veh) had a mean IOP of 14.85±0.29 mmHg (N=12), similar to the WT-PD group (15.45±0.70 mmHg, N=12, p=0.91, One-Way ANOVA post hoc Tukey tests, same test applied below). The Pax6 treated with vehicle (Pax6-Veh) exhibited an elevated IOP with a mean of 18.79±0.46 mmHg (N=26, p<0.0001, Fig. 1B), which is consistent with the glaucomatous findings seen in classical aniridia. Application of PD reduced IOP to near WT levels: the mean IOP of the Pax6-PD (N=38) was 15.80±0.37 mmHg, a substantial improvement compared to Pax6-Veh (p<0.0001).

Next, we measured the diameter of the retina using flat-mounts (Fig. 1C). The WT-Veh had a mean width of 4.02±0.05 mm (N=5), similar to WT-PD (4.00±0.04 mm, N=5, p>0.99), again suggesting that the vehicle or PD treatment itself did not affect size of the retina. Because the application of the PD treatment on WT mice did not appear to have any adverse effects, maintaining a mean IOP and retinal width as in WT-Veh, we combined the WT-Veh and WT-PD into one control group (Ctrl) in subsequent assessments.

Importantly, the Pax6-Veh group exhibited a significantly smaller retina (3.17±0.11 mm, N=6, p<0.0001), and the PD treatment improved the phenotype of mutant mice (3.82±0.11 mm, N=6, p<0.0001). The Pax6-PD mice exhibited a width similar to Ctrl (p=0.21), suggesting that PD treatment protected against the microphthalmic phenotype in Pax6-Veh mice.

3.2. PD0325901 treatment reduces retinal damage in Pax6Sey-Neu/+ mice.

Confocal images of the cryo-sections of the retina with the ONH at the center were taken to examine the retinal layer morphology (Fig. 2). Our results showed that the overall retina morphology for the three groups appeared normal (Fig. 2A-C, left), except that some areas showed cell clumping and disorganized layer thickness, which we termed as “hotspots” (Fig. 2B-C, right). First, the retinal layer thicknesses of the ONL, outer plexiform layer (OPL) + INL, and inner plexiform layer (IPL) + ganglion cell layer (GCL) were measured separately as illustrated in Fig. 2A-C. In most areas the overall thickness of the retina was not altered and were thus labeled as “normal” areas. The total thickness (Fig 2D) of the control group’s retina was 197.16±3.01 μm (N=5), similar to Pax6-Veh’s normal area (209.64±8.41 μm, N=5), and Pax6-PD’s normal area (208.79±3.03 μm, N=5). The above three groups showed no statistical significance (Ctrl vs Pax6-Veh: p=0.61, Ctrl vs. Pax6-PD: p=0.68, Pax6-Veh vs Pax6-PD: p>0.99). Similarly, the individual layer structure in the normal area was also comparable. For example, the IPL+GCL thickness (Fig 2G) in the control group was 66.43±2.02 μm (N=5), Pax6-Veh was 69.28±4.94 μm (N=5) and Pax6-PD was 61.88±2.65 μm (N=5, Ctrl vs Pax6-Veh: p=0.99, Ctrl vs Pax6-PD: p=0.98, Pax6-Veh vs Pax6-PD: p=0.90).

Fig. 2. Quantification of retinal layer thickness suggests PD protects against retinal damage in Pax6Sey-Neu/+ mice.

Fig. 2.

Open in a new tab

(A-C) Confocal images of retinal sections stained with DAPI (blue) with optic nerve head (ONH) at the center. Superior end of the eye is one the left in all images. Yellow box shows examples of retinal layers. Red bracket indicates a characteristic hotspot (hs). Scale bars in A, B, C (left panels) = 500μm, (right panels) = 100μm. (D-H) Quantifications of layer thickness. WT-PD and WT-Veh were combined into one control group (labeled as Ctrl). *: P<0.05; **: P<0.01; ***: P<0.001; ****:P<0.0001 in one-way ANOVA test with post hoc Tukey test.

Within the Pax6-Veh hotspots, however, the total thickness significantly increased to 267.94±9.58¼m (N=5), a far cry from the total thicknesses of the Ctrl (p<0.0001), Pax6-Veh normal areas (p<0.0001), and Pax6-PD normal areas (p<0.0001). This aberrant thickness in Pax6-Veh hotspots appears to be the result of thickening in all sublayers. The mean thickness of the ONL increased from 61.54±2.00 μm in the Ctrl to 77.02±2.05 μm in Pax6-Veh hotspots (p=0.03, Fig. 2E), the mean thickness of the OPL+INL increased from 53.73±1.17 μm to 75.7±7.52μm (p=0.003, Fig 2F),), and the mean thickness of the IPL+GCL increased from 66.43±2.02 μm to 111.2±11.78 μm (p=0.0003, Fig 2G).

The mean total thickness of hotspot areas of Pax6-PD mice was 212.88±2.6 μm (N=5), not significantly different from Ctrl (p=0.39) or Pax6-Veh normal (p=0.99). The hotspot thickening was also reduced in most PD hotspot sublayers. The mean PD hotspot thickness was 63.21±3.0 μm in the ONL (PD hotspot vs Veh hotspot, p=0.053), 55.86±1.79 μm in the OPL+INL (PD hotspot vs Veh hotspot, p=0.008), and 80.91±2.42 μm in the IPL+GCL (PD hotspot vs Veh hotspot, p=0.015). Together these findings indicate that Pax6 mice exhibited localized neural damage with increased thicknesses in the ONL, OPL+INL, and IPL+GCL, and these phenotypes were partially alleviated by PD0325901 treatment.

3.3. Immunohistochemistry shows disorganized retinal sublayer structure.

We next examined different cell types in the affected layers with various antibodies (Fig. 3). First, rbpms (red) is a marker to label most, if not all, RGCs in the mouse retina (Gao et al.; Rodriguez et al., 2014). The rbpms stains show highly disorganized cells in the GCL of the hotspots in Pax6-Veh, correlating with the findings in the flat-mounts shown in Fig 2. However, the rbpms+ density of the Ctrl group was 8.17±0.45 per 100 μm segment (N=6), not significantly different from the quantification of Pax6-Veh mice in both normal (8.23±0.49 per 100μm segment, N=6, p=0.99) and hotspot regions (6.70±0.40 per 100μm segment, N=4, p=0.32). The rbpms+ cell quantification for PD mice in both normal (7.24±0.69 per 100μm segment, N=5, p=0.69) and hotspot regions (6.88±0.42 per 100μm segment, N=5, p=0.39) were also not significantly different from the Ctrl group (Fig. 3).

Fig. 3. Disorganization of different types of retina cells in hotspots.

Fig. 3.

Open in a new tab

The left column shows normal retinal regions from a WT-Veh mouse as compared to the hotspots (hs) in Pax6-Veh (middle column) and Pax6-PD (right column). From top to bottom rows: DAPI, Rbpms, ChAT, Calbindin, GFAP, and Iba1, which were also labeled on the left corner. See details in the Results. Scale bar = 100μm.

We next examined the inner retina structure. Choline acetyltransferase (ChAT), stains for cholinergic amacrine cells, labeling somas in the INL and the processes that form a characteristic double bands in the IPL (Feng et al., 2013a; Feng et al., 2016; Feng et al., 2013b; Yasuhara et al., 2003). Calbindin is a calcium binding protein expressed in horizontal, amacrine cells, and some RGCs (Feng et al., 2013a; Feng et al., 2016; Feng et al., 2013b; Gao et al.; Puyang et al., 2016). ChAT staining suggests that cholinergic amacrine cells are largely unaffected between the groups, though the double-bands organization in the IPL was disrupted in the hotspot regions as indicated in Fig 3. Calbindin+ somas in the INL in the Pax6-Veh hotspot appeared increased, possibly contributing to the thickening seen in the INL.

Finally, we examined the glial cell populations in the retina. Glial fibrillary acidic protein (GFAP) is highly expressed in the end-feet processes of Müller glia cells, a population which controls maintenance and support throughout the retina (Gallego et al., 2012). The ionized calcium-binding adaptor molecule 1 (Iba1) is a protein expressed in retina microglia (Bosco et al., 2011; Puyang et al., 2016). The GFAP expression seemed upregulated in the somas and processes of the Müller glial cells in Pax6-Veh, and the number of Iba1+ microglia appeared upregulated in the inner retina as well. In the PD-treated mice, staining pattern of both GFAP and Iba1+ microglial cells returned to the baseline level as seen in the Ctrl group.

3.4. PD0325901 treatment partially protects the optic nerve head phenotype in Pax6Sey-Neu/+ mice.

We examined the ONH morphology of control, Pax6-Veh, and Pax6-PD mice via confocal imaging of flat-mounted retinas. Control mice (N=8) exhibited a well-organized ONH, with axon bundles converging to form the optic nerve at the center. By contrast, Pax6-Veh (N=5) showed a disorganized ONH, wherein bundles tangled and crossed, and the somas in the GCL showed pronounced clumping, like what was seen in hotspots in cryo-sections. The Pax6-PD group (N=6) showed considerable improvement in the ONH organization (Fig. 4, right column). Our results suggest that PD0325901 may protect axon bundle organization in the ONH.

3.5. PD0325901 treatment partially alleviates hotspot phenotype in the RNFL of Pax6Sey-Neu/+ mice.

To further characterize the localized RGC damage and the neuroprotection by PD, we analyzed the flat-mounted retina from all three groups by immunostaining for RGC axons using Tuj1, a monoclonal beta-tubulin antibody (Grannonico et al., 2021; Jiang et al., 2015; Miller et al., 2020). The overall morphology of RGC axon bundles was largely normal, as shown in Fig. 5. Again, we noticed damaged axon bundles that overlapped with noticeable clumps of somas displaced from underlying retinal layers (Fig. 5A). Hotspots were identified in the regions of the flat-mount within 600μm radius of the ONH. Regions which met our criteria (see Methods) were outlined and their areas were quantified. We counted the total number of hotspots in each retina. As shown in Fig. 5B, the mean number of hotspots per retina was significantly higher in Pax6-Veh mice, with a mean number of 4.2±0.6 per retina (N=5), compared to control (1.6±0.3, N=8, p=0.0002). Importantly, the mean number of hotspots was reduced in Pax6-PD (2.7±0.2, N=6, p=0.03).

Fig. 5. PD treatment partially alleviated the local retinal damage in Pax6Sey-Neu/+ mice.

Fig. 5.

Open in a new tab

(A) Confocal images of flat-mounted retinas immunostained by Tuj1-488 (green) and DAPI (blue). Pax6-Veh and Pax6-PD showed bundle disorganization (top) and displacement of somas from underlying layers (middle). Red outlines one hotspot in Pax6-Veh and Pax6-PD, respectively. Scale bars = 50μm (B) Quantification of the number of hotspots per retina. (C) Box plot of the total area of hotspots per retina. Mid-lines of boxes represent median, lower and upper bounds of the box represent 25th and 75th percentiles, respectively. Ends of whiskers represent the minimum and maximum values of the data set. *: P<0.05, **: P<0.01, ***:P<0.001 in one-way ANOVA post hoc Tukey’s test.

We next measured the area of each hotspot and calculated the total areas from all hotspots of each retina. As shown in Fig. 5C, the total area of hotspots in Pax6-Veh retina was (11.2±1.6) × 104 μm2/retina, significantly greater than control ((4.9±1.3) × 104 μm2/retina, p=0.01). Pax6-PD exhibited a reduced hotspot area (5.7±1.1) × 104 μm2/retina which was significantly smaller than the Pax6-Veh mice (p=0.04), but not significantly larger than those in control mice (p=0.89). Our results thus demonstrated that PD treatment reduced the overall number and size of hotspots.

3.6. PD treatment partially preserves the number and size of RGC axon bundles in Pax6Sey-Neu/+ mice.

As Figure 3 showed that the rbpms+ RGC density was not significantly changed in Pax-Veh, we next took a different approach to quantify the size and number of RGC axon bundles to get an overall estimation of the changes in RGCs. We counted the total number of axon bundles at at the radius of 400μm from the ONH (Fig. 6). The number of RGC axon bundles in the Ctrl group was 81.7±5.5 (N=5), and it was changed to 98.7±11.6 in Pax6-Veh (N=5, p=0.33). The number of axon bundles in Pax6-PD was 71.0±5.6 (N=6), also not significantly different from the Ctrl (p=0.60) and the Pax6-Veh (p=0.06).

Fig. 6. PD treatment partially protected RGC axon bundles in Pax6Sey-Neu/+ mice.

Fig. 6.

Open in a new tab

(A) Schematic showing bundle assessment at 400μm radius from ONH. Yellow dots correspond to axon bundles. (B-C) Blue box shows the close-up of an individual axon bundle marked by a yellow dot. Purple box represents width measurement based on signal intensity in MatLab (see details in Methods). (D) Histogram of axon widths in control, Pax6-Veh, and Pax6-PD mice, means and standard deviations of each distribution indicated in top right.

The distribution of individual axon bundle width at 400μm from the ONH was plotted in Fig. 6D. Measurements of individual bundles showed the mean width of Pax6-Veh was 7.34±0.16 μm (N=9 retinas, n=440 bundles), ~15% less than the controls (8.68±0.19 μm, N=6 retinas, n=356 bundles, p<0.0001). PD treatment partially diminished the effects on RGC axon bundle size (Pax6-PD: 8.35±0.25μm, N=4 retinas, n=159 bundles), compared to the Pax6-Veh (p=0.0001). There was also a significant difference between the axon bundle widths of control and Pax6-PD (p=0.0004), suggesting a partial protection by PD treatment. Together these results suggest that axon bundles are smaller in the Pax6-Veh retina, but PD treatment protects against bundle thinning, reinforcing its efficacy as a viable aniridia intervention.

3.7. Long-term partial protection of visual acuity by PD0325901 treatment in Pax6Sey-Neu/+ mice.

Visual acuity was examined by the optomotor test as described in the Methods (Fig. 7). For control mice (N=10), six mice exhibited robust responses, and four exhibited a moderate response. By contrast, seven Pax6-Veh mice (N=9) exhibited no responses at all, and the remaining two only had moderate responses. Not surprisingly, Pax6-PD (N=10) showed predominantly (seven) robust responses, comparable to the control levels of acuity. Chi-square tests concluded that Pax6-Veh mice show a significantly different level of response from both control (p=0.0011) and Pax6-PD (p=0.003). The control and PD groups showed no significant difference in their results (p=0.42) suggesting that PD0325901 treatment partially protected against visual behavior deficits brought on by aniridic damage and maintains that acuity well into adulthood.

Fig. 7. Long-term protection of vision by PD treatment in Pax6Sey-Neu/+ mice.

Fig. 7.

Open in a new tab

Optomotor responses were counted for each mouse and their response was deemed robust if they responded four or more times, moderate if they responded one to three times, or none if they showed no optomotor response (see details in Methods). Pax6-PD showed predominately robust responses similar to control mice, while Pax6-Veh largely did not respond. **: P < 0.01 by Chi-squared test.

4. Discussion

4.1. PD0325901 provides robust protection against aniridia phenotypes

Our results indicate that treatment with the small molecule MEK inhibitor PD0325901 substantially protects against aniridic damage in Pax6Sey-Neu/+. This protection extends not only to the anterior eye segment, but also to the retina and vision into adulthood. Associated aniridic symptoms such as heightened IOP, microphthalmia, and visual deficits show significant reduction in the presence of PD treatments even six-months after drug application. Likewise, major retinal deficits such as hotspots, which could be a major contributor to visual impairment, are also reduced in both size and number with PD treatment. Abnormal thickening/wrinkling of retinal layers, though still present in Pax6-PD mice, is also remarkably decreased compared to hotspots in Pax6-Veh. While this protection is not perfect, it appears to have likewise partially improved axon bundle health and ONH morphology.

Although PD0325901’s protective effects in aniridia phenotypes are clear, there is much optimization needed before treatment can effectively mitigate all aniridia symptoms. Since Pax6 dosage is necessary at many critical developmental stages both pre- and postnatally (Cvekl et al., 2004; Gregory-Evans et al., 2014; Lalitha et al., 2020; Li and Lu, 2005; Rabiee et al., 2020; Remez et al., 2017; Warren et al., 1999; Yasue et al., 2017), potential intervention will likely prove more efficacious if also administered prior to birth. Not only will this remove early postnatal stressors on mothers and pups, but it will likely compensate for earlier damages instigated by haploinsufficiency. It is known that Pax6 is needed to facilitate differentiation and maturation of retinal progenitor cells (RPCs) in early and late-born retinal layer populations, reinforcing the necessity of earlier postnatal treatments (Remez et al., 2017). Our initial topical treatment may compensate for dosage required from P5 to P30, and this protection is demonstrably robust, but persisting damage may arise from complications not compensated either prenatally, or in other postnatal stages requiring Pax6.

4.2. MEK inhibition and treatment parameters

Previous studies have shown that MEK inhibition can indirectly influence Pax6 expression, and therefore may provide a novel avenue for Pax6 haploinsufficiency intervention (de la Puente et al., 2016; Li and Lu, 2005). This treatment has some potential pitfalls, however, as excessive MEK inhibition has been linked to retinal detachment and retinal vein occlusion (LoRusso et al., 2010). As an MEK inhibitor, PD0325901’s function highlights the significance of proliferation-related pathways in the healthy maintenance of ocular phenotypes. It has been employed as an effective anti-aging treatment and has also been shown to assuage myopathic features in mice afflicted with neurofibromatosis type 1 (Castillo-Quan et al., 2019; Summers et al., 2018). Such examples verify the safety and efficacy of PD0325901 at low doses, and combined with the results of this study, demonstrates its capability in maintaining developmental pathways.

In this way, PD0325901 is comparable to another popular aniridia intervention, the nonsense mutation suppressant known as ataluren. Ataluren is purported to target Pax6 mRNA transcripts at the translational level, and it has been shown to have similar protective effects by compensating for decreased Pax6 dosage (Gregory-Evans et al., 2014; Wang et al., 2017a). Like PD0325901, ataluren has also been shown to have deleterious effects at higher concentrations, such as lens hyperplasia (Wang et al., 2017a). While both provide ample protection against aniridic insults, continued research is needed to further assess both MEK inhibitors and ataluren to determine which is most viable for clinical applications.

4.3. Variations in hotspot phenotypes and possible links to visual deficits

Ongoing in vivo assessments of aniridia phenotypes are necessary to develop adequate diagnostic, prognostic, and therapeutic measures to mitigate the disease. This has proven difficult as abnormal aniridia phenotypes vary greatly between mice due to individual genetics (Cole J D, 2021; Netland et al., 2011). We found this to be particularly true of retinal hotspot damage, as the size, shape, and layer characteristics of damaged regions often varied greatly among individual mice. Our data suggests that hotspot size and number are certainly increased in aniridic mice compared to controls. As experimenters didn’t know the genotype or treatment groups when they analyze the imaging data, it is possible that the small number of hotspots found in control mice could be due to technical issues such as dissection-induced physical damage in the retina. We also believe the disrupted areas in flat-mounts were consistent with the swollen areas seen in cryo-sections, as DAPI stains reveal clusters of disorganized and displaced somas from underlying layers. In addition, the GCL+IPL layers proved the most obviously affected in mutant retinas. It appears that, while thickness of layers in hotspots is altered, the layers contributing to these abnormalities can vary. Our results suggest that all layers were affected to varied degree, and future work is thus needed to better understand the underlying mechanisms.

Our quantification of rbpms+ cells found that density of RGCs is not statistically significant between the three groups, though there appears to be a trend toward RGC density reduction in Pax6-Veh hotspots. These results coincide with other findings. For example, Hickmott et al. analyzed the total cell number in the GCLs of both WT and aniridic mice of three different backgrounds: C57Bl/6J, B6129F1, and 129S1/SvlmJ, and found no significant difference in GCL cell density between WT and aniridic mice, nor between any of the backgrounds (Hickmott et al., 2018).

Our results also suggest that other retinal cell populations might be affected (Fig. 3). ChAT stains seem largely normal in all three groups, but Calbindin may show a higher density of cells in the INL. This is likely the result of the increased clumping/thickening demonstrated in Fig. 2F. While both Ctrl and Pax6-PD maintained the expected Müller glia cells end-feet demarcation stained by GFAP, the Pax6-Veh mice showed more widespread staining in somas and processes. Iba1+ microglial cells might be upregulated in the Pax6-Veh, perhaps because of an inflammatory response to aniridic damage.

It is unknown whether local damage results in global acuity reduction. For one, it has been well-documented that RGC reduction, such as that commonly seen in glaucomatous eyes (i.e. those with heightened IOP), often accompanies a decline in visual acuity (Chen et al., 2015; Feng et al., 2013b; Hood, 2019; Medeiros et al., 2013; Puyang et al., 2016). It is however important to note that no strict causal relationship has been determined between local RGC loss and the global visual reduction, and some studies have even shown a reduction in the visual field prior to RGC population decline (Hood, 2019). In fact, multiple deficits in retinal functions were detected in aniridia mice. Gregory-Evans et al performed ERG analysis at P60 and found that the scotopic responses were nearly undetectable in aniridic mice, suggesting defects in the inner and outer retina (Gregory-Evans et al., 2014), consistent with our previous findings (Rabiee et al., 2020).

Perhaps more telling is aniridia’s disruptive effect on axon bundles, which is evident not only in hotspot regions (i.e. Fig 5) but also around the ONH (Fig. 4/6). Lower levels of Pax6 expression have been shown to impair axon fasciculation (Lalitha et al., 2020), consistent with the reduced width measures of bundles of Pax6-Veh mice (Fig. 6). More work is needed to understand exactly how axon disorganization/defasciculation contributes to visual deficit (Chen et al., 2015; Medeiros et al., 2013; Miller et al., 2020).

4.4. Conclusion

Our studies suggest that MEK inhibitors like PD0325901 provided long-term protection against aniridia-related damage in Pax6Sey-Neu/+ mice. Topical application from P5-P30 reliably increased retina size, reduced IOP elevation, and improved visual acuity. Likewise, isolated regions of retinal damage in different retinal layers saw significant reduction in clumping and disorganization when treated with PD0325901. Altogether, our results indicate that PD0325901 may serve as a reliable protective measure for patients with aniridia.

Highlights.

  • Pax6 gene mutation leads to localized retinal damage in mice.

  • PD0325901, a MEK inhibitor, reduced retinal damage and IOP elevation and improved visual acuity.

  • PD0325901 treatment provides long-term neuroprotection.

Acknowledgments:

This work was supported by NIH/NEI R01EY024349 (ARD), R01EY026286 and R01EY029121 (XL), and Vision For Tomorrow (ARD).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing interests: no conflict of interest.

References

  1. Bosco A, Steele MR, Vetter ML, 2011. Early microglia activation in a mouse model ofchronic glaucoma. J Comp Neurol 519, 599–620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Castillo-Quan JI, Tain LS, Kinghorn KJ, Li L, Grönke S, Hinze Y, Blackwell TK, Bjedov I, Partridge L, 2019. A triple drug combination targeting components of the nutrient-sensing network maximizes longevity. Proceedings of the National Academy of Sciences 116, 20817–20819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Chen H, Zhao Y, Liu M, Feng L, Puyang Z, Yi J, Liang P, Zhang HF, Cang J, Troy JB, Liu X, 2015. Progressive degeneration of retinal and superior collicular functions in mice with sustained ocular hypertension. Investigative ophthalmology & visual science 56, 1971–1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cole JD, R. C, Norat P, Gao J, Provencio I, Netland P, Liu X, 2021. Neural damage and neuroprotection with glaucoma development in aniridia Current Neurobiology. [PMC free article] [PubMed] [Google Scholar]
  5. Cvekl A, Yang Y, Chauhan BK, Cveklova K, 2004. Regulation of gene expression by Pax6 in ocular cells: a case of tissue-preferred expression of crystallins in lens. Int J Dev Biol 48, 829–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. de la Puente P, Muz B, Jin A, Azab F, Luderer M, Salama NN, Azab AK, 2016. MEK inhibitor, TAK-733 reduces proliferation, affects cell cycle and apoptosis, and synergizes with other targeted therapies in multiple myeloma. Blood Cancer Journal 6, e399–e399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Feng L, Chen H, Suyeoka G, Liu X, 2013a. A laser-induced mouse model of chronic ocular hypertension to characterize visual defects. J Vis Exp, 50440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Feng L, Chen H, Yi J, Troy JB, Zhang HF, Liu X, 2016. Long-Term Protection of Retinal Ganglion Cells and Visual Function by Brain-Derived Neurotrophic Factor in Mice With Ocular Hypertension. Investigative ophthalmology & visual science 57, 3793–3802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Feng L, Zhao Y, Yoshida M, Chen H, Yang JF, Kim TS, Cang J, Troy JB, Liu X, 2013b. Sustained ocular hypertension induces dendritic degeneration of mouse retinal ganglion cells that depends on cell type and location. Invest Ophthalmol Vis Sci 54, 1106–1117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gallego BI, Salazar JJ, de Hoz R, Rojas B, Ramírez AI, Salinas-Navarro M, Ortín-Martínez A, Valiente-Soriano FJ, Avilés-Trigueros M, Villegas-Perez MP, Vidal-Sanz M, Triviño A, Ramírez JM, 2012. IOP induces upregulation of GFAP and MHC-II and microglia reactivity in mice retina contralateral to experimental glaucoma. Journal of Neuroinflammation 9, 92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gao J, Griner EM, Liu M, Moy J, Provencio I, Liu X, Differential effects of experimental glaucoma on intrinsically photosensitive retinal ganglion cells in mice. Journal of Comparative Neurology 10.1002/cne.25293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Glaser T, Jepeal L, Edwards JG, Young SR, Favor J, Maas RL, 1994. PAX6 gene dosage effect in a family with congenital cataracts, aniridia, anophthalmia and central nervous system defects. Nature Genetics 7, 463–471. [DOI] [PubMed] [Google Scholar]
  13. Grannonico M, Miller DA, Liu M, Norat P, Deppmann CD, Netland PA, Zhang HF, Liu X, 2021. Global and Regional Damages in Retinal Ganglion Cell Axon Bundles Monitored Non-Invasively by Visible-Light Optical Coherence Tomography Fibergraphy. The Journal of Neuroscience 41, 10179–10193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Graw J, Löster J, Puk O, Münster D, Haubst N, Soewarto D, Fuchs H, Meyer B, Nürnberg P, Pretsch W, Selby P, Favor J, Wolf E, de Angelis MH, 2005. Three Novel Pax6 Alleles in the Mouse Leading to the Same Small-Eye Phenotype Caused by Different Consequences at Target Promoters. Investigative Ophthalmology & Visual Science 46, 4671–4683. [DOI] [PubMed] [Google Scholar]
  15. Gregory-Evans CY, Wang X, Wasan KM, Zhao J, Metcalfe AL, Gregory-Evans K, 2014. Postnatal manipulation of Pax6 dosage reverses congenital tissue malformation defects. J Clin Invest 124, 111–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hickmott JW, Gunawardane U, Jensen K, Korecki AJ, Simpson EM, 2018. Epistasis between Pax6Sey and genetic background reinforces the value of defined hybrid mouse models for therapeutic trials. Gene Therapy 25, 524–537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hingorani M, Hanson I, van Heyningen V, 2012. Aniridia. European Journal of Human Genetics 20, 1011–1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hood DC, 2019. Does Retinal Ganglion Cell Loss Precede Visual Field Loss in Glaucoma? J Glaucoma 28, 945–951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Jiang S-M, Zeng L-P, Zeng J-H, Tang L, Chen X-M, Wei X, 2015. β-III-Tubulin: a reliable marker for retinal ganglion cell labeling in experimental models of glaucoma. Int J Ophthalmol 8, 643–652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lalitha S, Basu B, Surya S, Meera V, Riya PA, Parvathy S, Das AV, Sivakumar KC, Nelson-Sathi S, James J, 2020. Pax6 modulates intra-retinal axon guidance and fasciculation of retinal ganglion cells during retinogenesis. Scientific Reports 10, 16075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lee H, Khan R, O’Keefe M, 2008. Aniridia: current pathology and management. Acta Ophthalmologica 86, 708–715. [DOI] [PubMed] [Google Scholar]
  22. Li T, Lu L, 2005. Epidermal Growth Factor-induced Proliferation Requires Down-regulation of Pax6 in Corneal Epithelial Cells *. Journal of Biological Chemistry 280, 12988–12995. [DOI] [PubMed] [Google Scholar]
  23. Lima Cunha D, Arno G, Corton M, Moosajee M, 2019. The Spectrum of PAX6 Mutations and Genotype-Phenotype Correlations in the Eye. Genes (Basel) 10, 1050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. LoRusso PM, Krishnamurthi SS, Rinehart JJ, Nabell LM, Malburg L, Chapman PB, DePrimo SE, Bentivegna S, Wilner KD, Tan W, Ricart AD, 2010. Phase I Pharmacokinetic and Pharmacodynamic Study of the Oral MAPK/ERK Kinase Inhibitor PD-0325901 in Patients with Advanced Cancers. Clinical Cancer Research 16, 1924–1937. [DOI] [PubMed] [Google Scholar]
  25. McCulley TJ, Mayer K, Dahr SS, Simpson J, Holland EJ, 2005. Aniridia and optic nervehypoplasia. Eye (Lond) 19, 762–764. [DOI] [PubMed] [Google Scholar]
  26. Medeiros FA, Lisboa R, Weinreb RN, Liebmann JM, Girkin C, Zangwill LM, 2013. Retinal ganglion cell count estimates associated with early development of visual field defects in glaucoma. Ophthalmology 120, 736–744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Miller DA, Grannonico M, Liu M, Kuranov RV, Netland PA, Liu X, Zhang H, 2020. Visible-light optical coherence tomography fibergram for monitoring retinal ganglion cell axon bundles in vivo. Investigative Ophthalmology & Visual Science 61, 2529–2529. [Google Scholar]
  28. Netland PA, Scott ML, Boyle JWIV, Lauderdale JD, 2011. Ocular and systemic findings in a survey of aniridia subjects. Journal of American Association for Pediatric Ophthalmology and Strabismus {JAAPOS} 15, 562–566. [DOI] [PubMed] [Google Scholar]
  29. Nishina S, Kohsaka S, Yamaguchi Y, Handa H, Kawakami A, Fujisawa H, Azuma N, 1999. PAX6 expression in the developing human eye. Br J Ophthalmol 83, 723–727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Puyang Z, Feng L, Chen H, Liang P, Troy JB, Liu X, 2016. Retinal Ganglion Cell Loss is Delayed Following Optic Nerve Crush in NLRP3 Knockout Mice. Scientific reports 6, 20998–20998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rabiee B, Anwar KN, Shen X, Putra I, Liu M, Jung R, Afsharkhamseh N, Rosenblatt MI, Fishman GA, Liu X, Ghassemi M, Djalilian AR, 2020. Gene dosagemanipulation alleviates manifestations of hereditary PAX6 haploinsufficiency in mice. Science Translational Medicine 12, eaaz4894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rangarajan KV, Lawhn-Heath C, Feng L, Kim TS, Cang J, Liu X, 2011. Detection of Visual Deficits in Aging DBA/2J Mice by Two Behavioral Assays. Current Eye Research 36, 481–491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Remez LA, Onishi A, Menuchin-Lasowski Y, Biran A, Blackshaw S, Wahlin KJ, Zack DJ, Ashery-Padan R, 2017. Pax6 is essential for the generation of late-born retinal neurons and for inhibition of photoreceptor-fate during late stages of retinogenesis. Dev Biol 432, 140–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rodriguez AR, de Sevilla Müller LP, Brecha NC, 2014. The RNA binding protein RBPMS is a selective marker of ganglion cells in the mammalian retina. J Comp Neurol 522, 1411–1443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Samant M, Chauhan BK, Lathrop KL, Nischal KK, 2016. Congenital aniridia: etiology, manifestations and management. Expert Rev Ophthalmol 11, 135–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Sannan NS, Gregory-Evans CY, Lyons CJ, Lehman AM, Langlois S, Warner SJ, Zakrzewski H, Gregory-Evans K, 2017. Correlation of novel PAX6 gene abnormalities in aniridia and clinical presentation. Can J Ophthalmol 52, 570–577. [DOI] [PubMed] [Google Scholar]
  37. Simpson TI, Price DJ, 2002. Pax6; A pleiotropic player in development. BioEssays 24, 1041–1051. [DOI] [PubMed] [Google Scholar]
  38. Solberg NT, Melheim M, Strand MF, Olsen PA, Krauss S, 2019. MEK Inhibition Induces Canonical WNT Signaling through YAP in KRAS Mutated HCT-15 Cells, and a Cancer Preventive FOXO3/FOXM1 Ratio in Combination with TNKS Inhibition. Cancers (Basel) 11, 164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Summers MA, Vasiljevski ER, Mikulec K, Peacock L, Little DG, Schindeler A, 2018. Developmental dosing with a MEK inhibitor (PD0325901) rescues myopathic features of the muscle-specific but not limb-specific Nf1 knockout mouse. Molecular Genetics and Metabolism 123, 518–525. [DOI] [PubMed] [Google Scholar]
  40. Wang X, Gregory-Evans K, Wasan KM, Sivak O, Shan X, Gregory-Evans CY, 2017a. Efficacy of Postnatal In Vivo Nonsense Suppression Therapy in a Pax6 Mouse Model of Aniridia. Mol Ther Nucleic Acids 7, 417–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wang X, Shan X, Gregory-Evans CY, 2017b. A mouse model of aniridia reveals the in vivo downstream targets of Pax6 driving iris and ciliary body development in the eye. Biochim Biophys Acta Mol Basis Dis 1863, 60–67. [DOI] [PubMed] [Google Scholar]
  42. Warren N, Caric D, Pratt T, Clausen JA, Asavaritikrai P, Mason JO, Hill RE, Price DJ, 1999. The Transcription Factor, Pax6, is Required for Cell Proliferation and Differentiation in the Developing Cerebral Cortex. Cerebral Cortex 9, 627–635. [DOI] [PubMed] [Google Scholar]
  43. Yasue A, Kono H, Habuta M, Bando T, Sato K, Inoue J, Oyadomari S, Noji S, Tanaka E, Ohuchi H, 2017. Relationship between somatic mosaicism of Pax6 mutation and variable developmental eye abnormalities—an analysis of CRISPR genome-edited mouse embryos. Scientific Reports 7, 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yasuhara O, Tooyama I, Aimi Y, Bellier J-P, Hisano T, Matsuo A, Park M, Kimura H, 2003. Demonstration of cholinergic ganglion cells in rat retina: expression of an alternative splice variant of choline acetyltransferase. J Neurosci 23, 2872–2881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Yeritsyan N, Lehmann K, Puk O, Graw J, Löwel S, 2012. Visual capabilities and cortical maps in BALB/c mice. European Journal of Neuroscience 36, 2801–2811. [DOI] [PubMed] [Google Scholar]
  46. Yi J, Puyang Z, Feng L, Duan L, Liang P, Backman V, Liu X, Zhang HF, 2016. Optical Detection of Early Damage in Retinal Ganglion Cells in a Mouse Model of Partial Optic Nerve Crush Injury. Investigative Ophthalmology & Visual Science 57, 5665–5671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Yokoi T, Nishina S, Fukami M, Ogata T, Hosono K, Hotta Y, Azuma N, 2016. Genotype–phenotype correlation of PAX6 gene mutations in aniridia. Human Genome Variation 3, 15052. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES