Research ArticleGeneticsNeuroscienceOphthalmology Open Access | 10.1172/jci.insight.188710

Balasankara Reddy Kaipa,1,2 Ramesh Kasetti,3 Yogapriya Sundaresan,1,2 Linya Li,1,2 Sam Yacoub,3 J. Cameron Millar,3 William Cho,1,2 Dorota Skowronska-Krawczyk,1,2 Prabhavathi Maddineni,4 Krzysztof Palczewski,1,2,5,6 and Gulab S. Zode1,2

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Kaipa, B. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Kasetti, R. in: JCI | PubMed | Google Scholar |

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Sundaresan, Y. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Li, L. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Yacoub, S. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Millar, J. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Cho, W. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Skowronska-Krawczyk, D. in: JCI | PubMed | Google Scholar |

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Maddineni, P. in: JCI | PubMed | Google Scholar

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Palczewski, K. in: JCI | PubMed | Google Scholar |

1Gavin Herbert Eye Institute-Center for Translational Vision Research, Department of Ophthalmology, and

2Department of Physiology and Biophysics, University of California Irvine School of Medicine, Irvine, California, USA.

3Department of Pharmacology and Neuroscience, North Texas Eye Research Institute, University of North Texas Health Science Center at Fort Worth, Texas, USA.

4Department of Ophthalmology, School of Medicine, University of Missouri, Columbia, Missouri, USA.

5Department of Chemistry and

6Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California, USA.

Address correspondence to: Gulab S. Zode, 829 Health Sciences Rd., Irvine, California 92617, USA. Phone: 949.824.4366; Email: gzode@hs.uci.edu.

Find articles by Zode, G. in: JCI | PubMed | Google Scholar

Published January 21, 2025 - More info

Published in Volume 10, Issue 5 on March 10, 2025
JCI Insight. 2025;10(5):e188710. https://doi.org/10.1172/jci.insight.188710.
© 2025 Kaipa et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. View PDF Abstract

Elevation of intraocular pressure (IOP) due to trabecular meshwork (TM) dysfunction, leading to neurodegeneration, is the pathological hallmark of primary open-angle glaucoma (POAG). Impaired axonal transport is an early and critical feature of glaucomatous neurodegeneration. However, a robust mouse model that accurately replicates these human POAG features has been lacking. We report the development and characterization of a new Cre-inducible mouse model expressing a DsRed-tagged Y437H mutant of human myocilin (Tg.CreMYOCY437H). A single intravitreal injection of HAd5-Cre induced selective MYOC expression in the TM, causing TM dysfunction, reducing the outflow facility, and progressively elevating IOP in Tg.CreMYOCY437H mice. Sustained IOP elevation resulted in significant loss of retinal ganglion cells (RGCs) and progressive axonal degeneration in Cre-induced Tg.CreMYOCY437H mice. Notably, impaired anterograde axonal transport was observed at the optic nerve head before RGC degeneration, independent of age, indicating that impaired axonal transport contributes to RGC degeneration in Tg.CreMYOCY437H mice. In contrast, axonal transport remained intact in ocular hypertensive mice injected with microbeads, despite significant RGC loss. Our findings indicate that Cre-inducible Tg.CreMYOCY437H mice replicate all glaucoma phenotypes, providing an ideal model for studying early events of TM dysfunction and neuronal loss in POAG.

Graphical Abstract Introduction

Glaucoma is a group of multifactorial neurodegenerative diseases characterized by progressive optic neuropathy. It is the second leading cause of irreversible vision loss, affecting more than 70 million people worldwide (1, 2), and this number is estimated to increase to 112 million by the year 2040 (3). Primary open-angle glaucoma (POAG) is the most common form of glaucoma, accounting for approximately 70% of all cases (1). POAG is characterized by progressive loss of the soma and axons of retinal ganglion cells (RGCs), leading to irreversible vision loss (1, 2). Impaired axonal transport at the optic nerve head (ONH) is implicated as an early pathological event associated with glaucomatous neurodegeneration (4–6). Elevated intraocular pressure (IOP) is a significant risk factor for POAG and the only one that is treatable (7). The trabecular meshwork (TM), a molecular sieve-like structure, regulates IOP by constantly adjusting the resistance to aqueous humor (AH) outflow. In POAG, increased resistance to AH outflow elevates IOP, leading to neurodegeneration (8–10). This increase in outflow resistance is associated with TM dysfunction (11–15). Most current treatment strategies for POAG do not target the underlying pathology of the TM and RGCs, and vision loss continues to progress in some patients (16). To understand the pathological mechanisms of TM dysfunction/IOP elevation and glaucomatous neurodegeneration, there is an unmet need to develop a simple and reliable animal model that can faithfully replicate all features of POAG.

Currently, several mouse models of ocular hypertension (OHT) are utilized to study the pathophysiology of glaucomatous neurodegeneration (17, 18). These include DBA/2J mice and inducible mouse models that physically block AH outflow through the TM, leading to OHT and neuronal loss (19–24). While these models have provided important mechanistic insights, they do not allow the study of TM pathology, as they elevate IOP by physically blocking the outflow pathway (21–23). Thus, these models do not truly represent the human-POAG phenotype, where the AH outflow pathway is open. Importantly, these models destroy the TM, which can induce acute and sudden IOP-induced glaucomatous neurodegeneration. Moreover, these models are often technically challenging to replicate in laboratory settings, and they exhibit variable phenotypes and other confounding features such as ocular inflammation. Developing a mouse model that mimics a known genetic cause of human POAG represents an ideal strategy for understanding the pathophysiology of POAG.

Mutation of myocilin (MYOC) is the most common genetic cause of POAG and significantly contributes to juvenile-onset OAG (JOAG) (9, 25, 26). MYOC-associated JOAG is a more aggressive form of glaucoma, characterized by high IOP in young children and rapid progression to vision loss (9, 26). Mouse models with these genetic alterations in the MYOC gene mimicking human POAG have proved to be invaluable tools for understanding the pathogenesis of POAG and designing treatment strategies (27–29). Previously, we developed a transgenic mouse model (Tg-MYOCY437H) of myocilin POAG by random genomic insertion of the human mutant myocilin and demonstrated that Tg-MYOCY437H mice develop glaucoma phenotypes closely resembling those seen in patients with POAG with the Y437H MYOC mutation (29). However, Tg-MYOCY437H mice presented a few drawbacks, including (a) possible unknown gene mutations and multiple copies of the transgene due to random integration; (b) a mild phenotype on a pure C57BL/6J background, limiting the study of glaucomatous neurodegeneration; (c) possible silencing of the transgene upon subsequent breeding; and (d) lack of specific antibodies to detect myocilin in the mouse TM. To overcome these limitations, we utilized a TARGATT site-specific knock-in strategy (30) to generate transgenic mice expressing human mutant MYOC. This technology employs serine integrase, PhiC31 (ΦC31), to insert a single copy of a gene of interest into a preselected intergenic and transcriptionally active genomic locus (H11), which has been engineered with a docking site. This allows stable and single site-specific transgene integration. Since mutant MYOC is toxic to TM cells (29, 31), we exploited the Cre-lox system to develop an inducible mouse model in which mutant MYOC is expressed in the tissue of interest only upon Cre expression. Here, we report the development and characterization of a Cre-inducible transgenic mouse line expressing the DsRed-tagged Y437H mutant of human myocilin (Tg.CreMYOCY437H). We further utilized this model to investigate early events of glaucomatous TM dysfunction and neurodegeneration. In contrast to the microbead-occlusion (MB-occlusion) model of OHT, in which axonal transport remains intact despite significant loss of RGCs, we observed that sustained IOP elevation in the Tg.CreMYOCY437H mice significantly impairs axonal transport at the ONH prior to any loss of RGC somas and axons.

Results

Generation of an inducible mouse model of myocilin POAG. Using the TARGATT site-specific knock-in strategy, we developed Cre-inducible transgenic mice that express the DsRed-tagged Y437H-mutant of human MYOC (referred to as Tg.CreMYOCY437H). Under normal conditions, the mice do not express the human mutant-MYOC gene. Expression of Cre recombinase, however, leads to the removal of a Stop cassette and consequent expression of the DsRed-fused mutant MYOC (Figure 1A). A single transgene copy is inserted into a preselected intergenic and transcriptionally active genomic locus (H11) engineered with a docking site for stable and site-specific transgene integration. To confirm a site-specific knock-in of the transgene at the H11 site, we first performed PCR using primers specific to the integration site (Supplemental Figure 1A; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.188710DS1, Supplemental Information), which demonstrated a stable integration of the transgene in 1 founder line. These founder mice were further bred with C57BL/6J mice, and offspring were utilized for subsequent studies. For routine genotyping, primers specific to MYOC and DsRed region were utilized, which confirmed the presence of the transgene (Supplemental Figure 1B).

Figure 1

HAd5-Cre recombinase induces mutant MYOC in mouse TM. (A) Tg.CreMYOCY437H mice were engineered using a TARGATT gene knock-in strategy in which DsRed-tagged human Y437H-mutant MYOC was inserted in a transcriptionally active genomic locus (H11). A stop cassette prevents the expression of the mutant MYOC-DsRed fusion protein until Cre recombinase is introduced. Following Cre expression, the stop cassette is excised, allowing the expression of the mutant MYOC fused with DsRed only in targeted cells. (B) Representative slit-lamp images showing that no obvious ocular inflammation is associated with either HAd5-Empty– or HAd5-Cre–injected Tg.CreMYOCY437H mice (n = 6). Scale bars: 100 μm. (C) HAd5-Cre was injected intravitreally (2 × 107 pfu/eye) in mT/mG fluorescence-based reporter mice, and the conversion from tdTomato to GFP was examined 1 week after injection using confocal microscopy (n = 4). (D–F) Tg.CreMYOCY437H mice were injected intravitreally with HAd5-Empty or HAd5-Cre, and MYOC induction was examined in the TM using confocal imaging of DsRed protein in whole mount anterior segment (scale bars: 100 μm) (D), Western blot analysis of various ocular tissues with MYOC-antibody (E) showing the presence of human mutant myocilin in anterior-segment tissues of Cre-injected eyes (8 weeks after injection; AS, anterior segment; M, empty lane; CS, choroid and sclera); and confocal imaging of DsRed protein in the anterior-segment cross-section from Cre–- and Cre+-Tg.CreMYOCY437H mice (F) (n = 4). Scale bars: 50 μm.. (G) RNAScope analysis of MYOC (green) and Myoc (pink) transcripts in the anterior-segment cross-section of Cre–- and Cre+-Tg.CreMYOCY437H mice. The last panel shows DsRed-protein expression on the same slide. Note that DsRed fluorescence may appear to be less intense compared with other images, due to fluorescence quenching during sample processing (TM, trabecular meshwork; CB, ciliary body; SC, Schlemm’s canal). Arrows show the TM. Scale bars: 50 μm.

Helper Ad5-Cre induces mutant myocilin selectively in the TM. We have previously shown that intravitreal injection of Ad5 exhibits specific tropism in the mouse TM (32–35). Therefore, we utilized Ad5 expressing Cre recombinase to induce mutant myocilin in the TM of Tg.CreMYOCY437H mice (Supplemental Figure 2, A and B). Single intravitreal injection of Ad5-Empty or Ad5-Cre (2 × 107 pfu/eye) was performed in 3-month-old Tg.CreMYOCY437H mice. Analysis of anterior segment cross sections from the mouse eyes at 5 weeks after injection demonstrated robust MYOC-DsRed expression in the TM (Supplemental Figure 2A) of the Ad5-Cre–injected Tg.CreMYOCY437H mice. No MYOC-DsRed was detected in the corresponding sections from control mice, nor any other ocular tissues from the Cre-induced mice, including the retina. Western blot analysis of the iridocorneal angle, retina, and sclera clearly demonstrated that mutant myocilin was selectively induced in the TM tissue of Ad5-Cre–injected Tg.CreMYOCY437H mice, 5 weeks after the injection (Supplemental Figure 2B). Slit-lamp imaging revealed moderate ocular inflammation after Ad5-Cre injections (Supplemental Figure 2C). To reduce the ocular inflammation, we next utilized helper Ad5 (HAd5), which lacks most of the viral sequences except the cis-acting elements essential for viral replication and packaging. HAd5 is known to exhibit minimal immunogenicity while maintaining robust tropism for the tissue of interest (36). A single intravitreal injection of HAd5 expressing Cre or empty cassette was performed in 3-month-old Tg.CreMYOCY437H mice. Slit-lamp imaging demonstrated no visible signs of ocular inflammation in HAd5-Empty or HAd5-Cre–injected Tg.CreMYOCY437H mice (Figure 1B). First, we examined whether HAd5-expressing Cre exhibits functional activity in mouse TM, using fluorescence-reporter mT/mG mice. These mice express tdTomato in all tissues (red fluorescence); expression of Cre induces conversion of tdTomato to GFP (green fluorescence) (37). We performed intravitreal injection of HAd5-Empty or HAd5-Cre, and GFP/tdTomato expression was examined using confocal imaging of anterior-segment cross sections (Figure 1C and Supplemental Figure 3). Compared with empty-injected mT/mG mice, which only expressed tdTomato, Cre-injected mice exhibited GFP expression selectively in TM cells, and conversion efficiency was nearly 90% (Supplemental Figure 3). These data indicate that a single intravitreal injection of HAd5-Cre was highly efficient in transducing the mouse TM.

Next, we investigated whether HAd5-Cre induces mutant myocilin in the TM. A single intravitreal injection of HAd5-Cre or HAd5-Empty (2 × 107 pfu/eye) was performed in 2-month-old Tg.CreMYOCY437H mice, and MYOC-DsRed expression in various ocular tissues was evaluated (Figure 1, D–F). Whole mount anterior segment imaging demonstrated mutant-MYOC expression throughout the TM of Cre-injected Tg.CreMYOCY437H mice (Figure 1D). Western blot analysis of the anterior segment (AS), retina, and choroid-sclera (CS) tissue lysates demonstrated the presence of MYOC-DsRed protein in the AS but not in the retina or CS of Cre+Tg.CreMYOCY437H mice (Figure 1E). MYOC-DsRed protein was detected at approximately 75 kDa due to the DsRed tag on mutant MYOC (Figure 1E). We also detected higher molecular weight bands for MYOC, which likely represent heteromeric complexes of MYOC, as described previously (38, 39). Although endogenous myocilin was detected at 50 kDa in all ocular tissues, MYOC-DsRed (75 kDa) was only detected in the AS tissue lysates of Cre+Tg.CreMYOCY437H mice, and no mutant-MYOC protein was detected in the retina and CS of Cre+Tg.CreMYOCY437H mice. Notably, we observed that endogenous myocilin protein was increased in the lysates of iridocorneal-angle tissue of Cre-injected Tg.CreMYOCY437H mice compared with controls. Quantitative PCR (qPCR) analysis of anterior-segment tissues using primers specific to mutant MYOC confirmed the presence of MYOC mRNA in Cre-induced Tg.CreMYOCY437H mice (Supplemental Figure 4A). We also observed that expression of mutant MYOC induced a 3-fold increase in endogenous Myoc in the AS of Cre+Tg.CreMYOCY437H mice (Supplemental Figure 4B). Confocal imaging of anterior-segment cross-sections revealed a robust and selective induction of mutant MYOC in the TM of Tg.CreMYOCY437H mice (Figure 1F). Immunostaining for α-smooth muscle actin (α-SMA), which predominantly labels the TM and ciliary muscle, revealed a strong colocalization of mutant MYOC with α-SMA in the TM region of Cre-injected Tg.CreMYOCY437H mice (Supplemental Figure 5). These data indicate that HAd5-Cre selectively induces mutant MYOC in the TM of Tg.CreMYOCY437H mice.

Since mutant MYOC is driven by the CAG promoter, which can lead to overexpression of mutant MYOC, we next compared the mRNA transcript of mutant MYOC with endogenous Myoc in the TM region using RNA scope (Figure 1G and Supplemental Figure 6). These data also reveal that no mRNA transcript for mutant MYOC was detected in Cre–Tg.CreMYOCY437H mice, while abundant endogenous Myoc was detected in the TM region. Importantly, Cre+Tg.CreMYOCY437H mice displayed the presence of mutant MYOC transcript selectively in the TM region. Notably, total transcript measurements (per TM cell) revealed a significant upregulation of endogenous Myoc expression in the TM of Cre+Tg.CreMYOCY437H mice compared with Cre–Tg.CreMYOCY437H mice (Supplemental Figure 6). These data indicate that mutant MYOC expression induces endogenous Myoc in the TM. Moreover, the expression of the mutant MYOC transcript in Cre-injected eyes is similar to that of the Myoc in Cre–Tg.CreMYOCY437H mice. Together, these data establish that HAd5-Cre selectively induces mutant myocilin in the TM of Tg.CreMYOCY437H mice at a level similar to endogenous Myoc without causing ocular inflammation.

Expression of mutant MYOC reduces TM outflow and elevates IOP significantly in HAd5-Cre–injected Tg.CreMYOCY437H mice. Three-month-old Tg.CreMYOCY437H mice were injected intravitreally with HAd5-Empty or Cre, and IOPs were monitored weekly. Starting from 2 weeks after injection, Cre-treated Tg.CreMYOCY437H mice exhibited significantly higher and sustained IOP compared with empty-injected Tg.CreMYOCY437H mice (Figure 2A). Independent IOP measurements in conscious mice (without anesthesia) confirmed a pronounced and significant IOP elevation in HAd5-Cre–injected Tg.CreMYOCY437H mice (Supplemental Figure 7). Measurement of the outflow facility using the constant flow infusion method displayed a significantly reduced outflow facility 5 weeks after HAd5-Cre injection in Tg.CreMYOCY437H mice compared with the control group (11.75 nL/min/mmHg versus 22.32 nL/min/mmHg in Cre+ versus Cre–Tg.CreMYOCY437H mice) (Figure 2B). These findings demonstrate that HAd5-Cre induces significant and sustained IOP elevation due to reduced AH outflow facility from the TM in Tg.CreMYOCY437H mice.

Figure 2

Intravitreal administration of HAd5-Cre elevates IOP and decreases outflow facility in Tg.CreMYOCY437H mice. Three-month-old Tg.CreMYOCY437H mice received a single intravitreal injection of either HAd5-Empty or HAd5-Cre in both eyes. (A) Weekly IOP measurements demonstrated significant and sustained IOP elevation in Cre-injected Tg.CreMYOCY437H mice compared with HAd5-Empty–injected mice (n = 14 in Empty, and n = 18 in Cre-injected group; analyzed by 2-way ANOVA with multiple comparisons, ****P < 0.0001). (B) Outflow facility measurements showed a significant reduction in outflow facility 5 weeks post HAd5-Cre injection of Tg.CreMYOCY437H mice (n = 8) compared with HAd5-Empty–injected mice (n = 9) (unpaired t test, 2-tailed, mean ± SEM, **P < 0.0072).

Mutant myocilin induces ultrastructural and biochemical changes in the TM. In POAG, increased outflow resistance is associated with ultrastructural and biochemical changes in the TM, including increased extracellular matrix (ECM) deposition, actin-cytoskeletal changes, and induction of ER stress (11, 15, 40, 41). Next, we investigated whether mutant myocilin leads to morphological changes in the TM of Tg.CreMYOCY437H mice. H&E staining demonstrated open-angle and no noticeable morphological changes in the anterior-chamber structures of Cre+Tg.CreMYOCY437H mice, 5 weeks after injection (Supplemental Figure 8). We next performed transmission electron microscopy (TEM) to examine ultrastructural changes in the TM. Low-magnification TEM analysis demonstrated that the iridocorneal angle is open in both empty- and Cre-injected Tg.CreMYOCY437H mice (Figure 3A). Higher-magnification TEM images revealed loosely bound collagen fibers, ECM deposition, and disrupted TM integrity in Cre+Tg.CreMYOCY437H mice compared with Cre–Tg.CreMYOCY437H mice (Figure 3B). We further confirmed these findings using immunostaining. AS were immunostained with antibodies for fibronectin (FN) and actin (Supplemental Figure 9A). FN and actin significantly increased in the TM region of Cre-injected Tg.CreMYOCY437H mice (Supplemental Figure 9, B and C). Previous studies have shown that mutant-MYOC expression in the TM induces ER stress (29, 42–45). Therefore, we examined whether mutant-MYOC expression induces ER stress markers in the TM of Cre-induced Tg.CreMYOCY437H mice (Figure 3C). Immunostaining (Supplemental Figure 10) and Western blot analysis (Figure 3, C–E) demonstrated significantly increased ER stress markers, including GRP78, ATF4, and CHOP selectively in the TM of Cre-injected Tg.CreMYOCY437H mice. Together, these findings indicate that the expression of mutant myocilin induces ultrastructural and biochemical changes in the TM, leading to its dysfunction and IOP elevation in Tg.CreMYOCY437H mice.

Figure 3

Mutant-MYOC–induced ocular hypertension is associated with ultrastructural and biochemical changes in the TM. (A and B) Representative low-magnification (A) and high-magnification (scale bars: 20 μm) (B) TEM images of Tg.CreMYOCY437H mice 8 weeks after injection of HAd5-Empty or HAd5-Cre, showing the presence of loosely bound collagen fibers, ECM deposition, and loss of TM integrity in the juxtacanalicular-connective-tissue (JCT) region of Cre-injected Tg.CreMYOCY437H mice (n = 4 in each group) (TM, trabecular meshwork; CB, ciliary body; SC, Schlemm’s canal; CL, collagen fibers; ECM, extra cellular matrix). Scale bars: 1 μm. (C and D) Western blot and densitometric analyses showing that mutant MYOC induces ER stress in the anterior-segment tissue lysates of Cre-injected Tg.CreMYOCY437H mice (n = 3). (E) HAd5-Empty; AS, anterior segment. Two-way ANOVA with multiple comparisons (**P = 0.0053, *P = 0.0216).

Mutant-MYOC–induced sustained IOP elevation leads to functional and structural loss of RGCs. We next evaluated whether sustained IOP elevation induced by mutant MYOC is sufficient to cause functional and structural loss of RGCs. To assess the functional loss of RGCs, we performed pattern electroretinogram (PERG) 5, 10, and 15 weeks after injection (Figure 4, A–C). Representative PERG graphs and their analysis demonstrated no significant effect on PERG at 5 weeks, but significantly reduced PERG amplitudes and increased latencies were observed at 10 and 15 weeks after Cre injection of Tg.CreMYOCY437H mice (Figure 4, A–C). To determine the structural loss of RGCs, we next performed whole-mount retina staining with RNA binding protein with multiple splicing (RBPMS) antibody (Figure 4, D and E). As shown in representative RBPMS images, Cre-injected Tg.CreMYOCY437H mice exhibited moderate loss of RGCs in the peripheral retina, 15 weeks after injection (Figure 4D). RGC counting further confirmed significantly reduced RGCs in the periphery of Cre-injected Tg.CreMYOCY437H mice compared with controls at 15 weeks after injection (Figure 4E). Overall, there was a 33% loss of RGCs in the peripheral retina of Cre-injected Tg.CreMYOCY437H mice compared with controls. We did not observe RGC loss at 10 weeks after Cre injection in Tg.CreMYOCY437H mice (Supplemental Figure 11). These data indicate that sustained IOP elevation induced by mutant myocilin leads to functional and structural loss of RGCs in Tg.CreMYOCY437H mice.

Figure 4

Sustained IOP elevation leads to functional and structural loss of RGCs in Cre-injected Tg.CreMYOCY437H mice. Three- to 6-month-old Tg.CreMYOCY437H mice were injected intravitreally with HAd5-Empty or HAd5-Cre in both eyes, and IOP was monitored weekly to ensure IOP elevation. PERG was performed at 5, 10, and 15 weeks after treatment to assess the function of the RGCs. (A–C) A representative PERG graph (A) and its analysis (B and C) demonstrated significantly reduced PERG amplitude (B) and increased latency (C) starting from 10 weeks post Cre-injection, indicating functional loss of RGCs in Cre+-Tg.CreMYOCY437H mice (n = 6 in HAd5-Empty, and n = 6 in HAd5-Cre), 2-way ANOVA with multiple comparisons (***P = 0.0001, ****P < 0.0001). (D and E) RGC loss was further analyzed by staining the whole-mount retina with RBPMS antibody. A representative image of RBPMS staining of different regions of the retina (D) and its analyses (E) revealed a significant loss (33%) of RGCs in Tg.CreMYOCY437H mice 15 weeks after HAd5-Cre–injection compared with control mice injected with HAd5-Empty (n = 7 for HAd5-Empty, and n = 6 for HAd5-Cre). Two-way ANOVA with multiple comparisons (****P < 0.0001). Scale bars: 50 μm.

Mutant-MYOC–induced sustained-IOP elevation leads to optic nerve degeneration in Tg.CreMYOCY437H mice. We investigated whether sustained IOP elevation induced by mutant myocilin leads to optic nerve (ON) degeneration using paraphenylene diamine (PPD) staining (Figure 5A). Representative images of PPD-stained ON from Cre-injected Tg.CreMYOCY437H mice revealed ON degeneration, as evident from darkly stained axons, active gliosis, and glial scar formation. Approximately 20% and 45% axonal loss was observed in Cre-injected mice at 10 and 15 weeks, respectively, compared with empty-injected Tg.CreMYOCY437H mice (Figure 5B). To further confirm these findings, we performed GFAP immunostaining on retinal cross-sections from Cre-injected Tg.CreMYOCY437H mice (Supplemental Figure 12A). Immunostaining for GFAP revealed a prominent increase in GFAP reactivity in the ONH region, suggesting axonal injury. Axonal degeneration was associated with a decreased neuronal marker, Tuj1, and increased mitochondrial accumulation (TOM20) in the ONH region (Supplemental Figure 12B). These data establish that sustained-IOP elevation induces ON degeneration and gliosis in the ONH of Cre-injected Tg.CreMYOCY437H mice.

Figure 5

Sustained IOP elevation leads to optic nerve degeneration in Cre-injected Tg.CreMYOCY437H mice. Optic nerves were subjected to PPD staining to assess optic-nerve degeneration. (A) Representative images of PPD-stained optic nerves showing mild axonal degeneration as evident from darkly stained axons (yellow arrowhead) and the presence of glial scar formation (blue arrow) in Cre-injected Tg.CreMYOCY437H mice. Scale bars: 20 μm. (B) the mean axonal counts show a significant loss of ON axons (20% at 10 weeks and 45% at 15 weeks after injection in Cre-induced Tg.CreMYOCY437H mice (*P = 0.0351 for 10 weeks after injection, n = 5 for Empty, and n = 7 for Cre; ***P = 0.0001 for 15 weeks after injection, n = 6 for Empty, and n = 6 for Cre). Two-way ANOVA with multiple comparisons.

Impaired axonal transport at the ONH precedes neuronal loss in Tg.CreMYOCY437H mice. Since Tg.CreMYOCY437H mice exhibit well-defined timelines for RGC-axon loss; we further sought to understand the early pathogenic events preceding neuronal loss. Studies of RGC loss and ON degeneration suggested that axonal changes at the ONH may be the first site of damage in ocular hypertensive Tg.CreMYOCY437H mice. We, therefore, hypothesize that IOP-induced neurodegenerative changes in the ONH precede neuronal loss. Since Tg.CreMYOCY437H mice did not show significant structural loss of RGCs at 10 weeks after Cre injection (Supplemental Figure 11), we examined whether axonal dysfunction occurred at the ONH prior to RGC loss at 7 weeks after Cre injection. To test this, 15-month-old Tg.CreMYOCY437H mice were injected intracamerally with HAd5-Empty or Cre, and IOPs were monitored. IOP measurements confirmed a significant IOP elevation 6 weeks after Cre injection of Tg.CreMYOCY437H mice (Figure 6A). PERG, which measures the function of RGC soma, revealed no significant functional loss of RGC soma at 5 weeks after Cre injection of Tg.CreMYOCY437H mice (Figure 4, A–C). Whole-mount retinal staining with RBPMS revealed no significant loss of RGCs at 10 weeks after Cre injection of Tg.CreMYOCY437H mice (Supplemental Figure 11). Notably, visual evoked potential (VEP), which measures the postretinal function of the visual system — including ON and visual centers of the brain — demonstrated a significant loss of VEP amplitudes at 7 weeks, suggesting axonal dysfunction in Cre+Tg.CreMYOCY437H mice (Figure 6B). To understand whether axonal dysfunction occurs at early stages of neuronal loss due to sustained IOP, we examined the anterograde transport of fluorescently labeled cholera toxin B (CTB), as described previously (46). At 7 weeks after injection, HAd5-Empty or HAd5-Cre–injected Tg.CreMYOCY437H mice were intravitreally injected with green fluorescently tagged CTB dye (Figure 6C). Forty-eight hours after injections, anterograde transport of CTB through the ON and superior colliculus (SC) was monitored via fluorescence microscopy. As expected, CTB was transported to the SC in HAd5-Empty–injected Tg.CreMYOCY437H mice. However, CTB transport was completely blocked at the ONH, and no CTB was observed in the SC of Cre-injected Tg.CreMYOCY437H mice. Measurement of CTB fluorescence in the SC demonstrated a significant (~72%) loss of CTB in the SC of Cre-injected Tg.CreMYOCY437H mice (Figure 6D). Previous studies have shown that axonal transport deficits are age dependent (47). Since we observed significant axonal transport in 15-month-old Cre+Tg.CreMYOCY437H mice, we next explored whether axonal transport deficits are observed in younger Cre+Tg.CreMYOCY437H mice. Four-month-old Tg.CreMYOCY437H mice were intravitreally injected with HAd5-Empty or HAd5-Cre; at 7 weeks after injection, ocular hypertensive eyes were analyzed for CTB transport (Sup