Inducible nonhuman primate models of retinal degeneration for testing end-stage therapies

Animal models for the evaluation of retinal stem cell therapies

Retinal degeneration (RD) diseases leading to severe vision loss can affect photoreceptors (PRs) that are responsible for phototransduction, or retinal pigmented epithelium (RPE) providing support for PRs. Human pluripotent stem cell (hPSC)-based therapies are a potential approach for restoration of retinal structure in patients with currently incurable RD diseases. Currently, there are two targeted hPSC therapeutics: PR rescue and PR replacement. PR rescue involves the transplantation of RPE or other neural progenitors into the subretinal space to slow down or prevent further RD. RPE transplantation plays a critical role in preserving photoreceptors by providing trophic support and maintaining retinal integrity, particularly in diseases like age-related macular degeneration (AMD). Advances in RPE transplantation methods, such as polarized monolayer cultures and scaffold-based approaches, have shown promise in enhancing graft survival and integration. However, limitations include inconsistent integration, variable neurotrophic factor secretion, and immune rejection risks in non-autologous transplants. In PR replacement, stem cell-derived photoreceptor-like cells or photoreceptor progenitors (PRP) obtained are transplanted into the eye. While PRPs are commonly obtained from retinal organoids (ROs), alternative sources, such as early differentiation stages or direct differentiation protocols, are also utilized to enhance the efficiency and scalability of PRP generation. Challenges include achieving proper integration, forming outer segments, rosette formation, and avoiding immune rejection or tumorigenicity. Various animal models that simulate human RD diseases are being used for establishing surgical feasibility, graft survival and visual functional recovery but fail to replicate clinical immune challenges. Rodent models lack macula-like structures and have limited reliability in detecting subtle functional changes, while larger animal models pose ethical, logistical, and financial challenges. Immunocompromised models have been developed for minimizing xenograft issues. Visual functional testing for efficacy includes optokinetic testing (OKN), electroretinography (ERG), and electrophysiological recordings from the retina and brain. These tests often fail to capture the complexity of human visual recovery, highlighting the need for advanced models and improved functional testing techniques. This review aims to aggregate current knowledge about approaches to stem cell transplantation, requirements of animal models chosen for validating vision benefits of transplantation studies, advantages of using specific disease models and their limitations. While promising strides have been made, addressing these limitations remains essential for translating stem cell-based therapies into clinical success.

Healing of ischemic injury in the retina

Neuro- and retinal degenerative diseases, including Alzheimer's, age-related macular degeneration, stroke, and central retinal artery occlusion, rob millions of their independence. Studying these diseases in human retinas has been hindered by the immediate loss of neuronal activity postmortem. While recent studies restored limited activity in postmortem CNS tissues, synchronized neuronal transmission >30 minutes postmortem remained elusive. Our study overcomes this barrier by reviving and sustaining light signal transmission in human retinas recovered up to four hours and stored 48 hours postmortem. We also establish infrared-based ex vivo imaging for precise sampling, a closed perfusion system for drug testing, and an ex vivo ischemia-reperfusion model in mouse and human retina. This platform enables testing of neuroprotective and neurotoxic effects of drugs targeting oxidative stress and glutamate excitotoxicity. Our advances question the irreversibility of ischemic injury, support preclinical vision restoration studies, offer new insights into treating ischemic CNS injuries, and pave the way for transplantation of human donor eyes.

Teaser

Reviving light signaling in postmortem human retinas challenges the irreversibility of ischemic injury and advances research to restore vision.

Safety and efficacy of CRISPR-mediated genome ablation of VEGFA as a treatment for choroidal neovascularization in nonhuman primate eyes

CRISPR-based genome editing enables permanent suppression of angiogenic factors such as vascular endothelial growth factor (VEGF) as a potential treatment for choroidal neovascularization (CNV)-a major cause of blindness in age-related macular degeneration. We previously designed adeno-associated viral (AAV) vectors with S. pyogenes Cas 9 (SpCas9) and guide RNAs (gRNAs) to target conserved sequences in VEGFA across mouse, rhesus macaque, and human, with successful suppression of VEGF and laser-induced CNV in mice. Here, we advanced the platform to nonhuman primates and found that subretinal AAV8-SpCas9 with gRNAs targeting VEGFA may reduce VEGF and CNV severity as compared with SpCas9 without gRNAs. However, all eyes that received AAV8-SpCas9 regardless of gRNA presence developed subfoveal deposits, concentric macular rings, and outer retinal disruption that worsened at higher dose. Immunohistochemistry showed subfoveal accumulation of retinal pigment epithelial cells, collagen, and vimentin, disrupted photoreceptor structure, and retinal glial and microglial activation. Subretinal AAV8-SpCas9 triggered aqueous elevations in CCL2, but minimal systemic humoral or cellular responses against AAV8, SpCas9, or GFP reporter. Our findings suggest that CRISPR-mediated VEGFA ablation in nonhuman primate eyes may suppress VEGF and CNV, but can also lead to unexpected subretinal fibrosis, photoreceptor damage, and retinal inflammation despite minimal systemic immune responses.

Long-Term Porcine Retina Explants as an Alternative to In Vivo Experimentation

Purpose

The porcine retina represents an optimal model system to study treatment approaches for inherited retinal dystrophies owing to close anatomical similarities to the human retina, including a cone enriched visual streak. The aim of this work was to establish a protocol to keep explants in culture for up to 28 days with good morphological preservation.

Methods

Two to four retina explants per eye were obtained from the central part of the retina and transferred onto a membrane insert with the photoreceptors facing down. Different medium compositions using Neurobasal-A medium containing 100 or 450 mg/dL glucose and combinations of fetal calf serum, B-27 with or without insulin and N-2 were tested. We developed a tissue quality score with robust markers for different retinal cell types (protein kinase C alpha, peanut agglutinin and 4',6-diamidino-2-phenylindol).

Results

Retinae were kept until 28 days with only little degradation. The best results were attained using Neurobasal-A medium containing 100 mg/dL glucose supplemented with B-27 containing insulin and N-2. For an easy preparation process, it is necessary to minimize transport time and keep the eyes on ice until dissected. Heat-mediated decontamination by the butcher has to be avoided.

Conclusions

Using a standardized protocol, porcine retina explants represent an easy to handle intermediate model between in vitro and in vivo experimentation. This model system is robustly reproducible and contributes to the implementation of the 3R principle to minimize animal experimentation.

Translational Relevance

This model can be used to test future therapeutic approaches for inherited retinal dystrophies.