Extreme retinal remodeling triggered by light damage: implications for age related macular degeneration

PURPOSE: Our objective was to comprehensively assess the nature and chronology of neural remodeling in retinal degenerations triggered by light-induced retinal damage (LIRD) in adult albino rodents. Our primary hypothesis is that all complete photoreceptor degenerations devolve to extensive remodeling. An hypothesis emergent from data analysis is that the LIRD model closely mimics late-stage atrophic age relared macular degeneration (AMD). METHODS: Sprague-Dawley (SD) rats received intense light exposures of varied durations and survival times ranging from 0 to 240 days. Remodeling was visualized by computational molecular phenotyping (CMP) of a small molecule library: 4-aminobutyrate (γ), arginine (R), aspartate (D), glutamate (E), glutamine (Q), glutathione (J), glycine (G), and taurine (τ). This library was augmented by probes for key proteins such as rod opsin, cone opsin and cellular retinal binding protein (CRALBP). Quantitative CMP was used to profile 160 eyes from 86 animals in over 6,000 sections. RESULTS: The onset of remodeling in LIRD retinas is rapid, with immediate signs of metabolic stress in photoreceptors, the retinal pigmented epithelium (RPE), the choriocapillaris, and Müller cells. In particular, anomalous elevated aspartate levels appear to be an early stress marker in photoreceptors. After the stress phase, LIRD progresses to focal photoreceptor degeneration within 14 days and extensive remodeling by 60 days. RPE and choriocapillaris losses parallel Müller cell distal seal formation, with progressive neuronal migration, microneuroma evolution, fluid channel formation, and slow neuronal death. The remaining retina in advanced light damage can be classified as survivor, light damage (LD), or decimated zones where massive Müller cell and neuronal emigration into the choroid leaves a retina depleted of neurons and Müller cells. These zones and their transitions closely resemble human geographic atrophy. Across these zones, Müller cells manifest extreme changes in the definitive Müller cell τQE signature, as well as CRALBP and arginine signals. CONCLUSIONS: LIRD retinas manifest remodeling patterns of genetic retinal degeneration models, but involve no developmental complexities, and are ultimately more aggressive, devastating the remaining neural retina. The decimation of the neural retina via cell emigration through the perforated retina-choroid interface is a serious denouement. If focal remodeling in LIRD accurately profiles late stage atrophic age-related macular degenerations, it augurs poorly for simple molecular interventions. Indeed, the LIRD profile in the SD rat manifests more similarities to advanced human atrophic AMD than most genetically or immunologically induced murine models of AMD.