Retinal Plasticity: Functional Recovery after Bipolar Cell Loss in the Oxygen Induced Retinopathy Model

Abstract

The development of non-invasive live ocular imaging and electrophysiological test systems for rodent eyes provides new tools for not only averaged analysis of the entire retina but also the ability to see, test, and compare different subregions of the same retina. These new capabilities provide the possibility for more detailed examinations of local structural and functional relationships within a single eye and the ability to also follow changes longitudinally over time. We have developed protocols based around the Micron-III/IV retinal imaging camera system for combining fluorescent imaging of the neural retinal micro-vasculature by FA (fluorescein angiography), imaging of all neural retinal layers by SD-OCT (Spectral-Domain Ocular Coherence Tomography), and focal "spot" light-targeted electroretinography (Focal-ERG) to relate the local neurovascular unit structure to the inner (photoreceptor) and outer-retinal electrical response to light stimulation. For demonstration purposes we have used the popular mouse oxygen induced retinopathy (OIR) model, which causes radial central patches of retinal neuron loss mostly in zones away from and between the primary retinal arteries and veins. In this model, the loss of central microvasculature is induced developmentally in mouse litters exposed to 75% oxygen from age P7 to P11. Return to room air on P12, causes several days of retinal ischemia during which neurons, mostly of the inner retina, perish. Bipolar and ganglion cell death ends as neovascular growth revascularizes the central retina. This model provides for non-uniform retinal damage as well as gradual progression and resolution over time. The OIR model was used to generate regions of inner retinal neuron loss in B6.Cg-TgThy1-YFP mice. Using image-guided focal-ERG, the dark-adapted mixed rod-cone light response was compared using stimulation of small circular (0.27 mm diameter) target areas located in the central retinas of the same eyes (OIR and control). The same areas of the same retinas were followed over three ages after revascularization (P21, P28 and P42). ConclusionsCombined FA and SD-OCT imaging can provide local geographic specific information on retinal structural changes and be used to select different retinal areas within the same eye for testing of local light response. This analysis strategy can be employed for studies with rodent disease models that do not uniformly impact the entire retinal area. Combining these techniques would also be useful for testing gene and cell replacement therapies in retinal degeneration models where typically a small zone of the retina is treated. Both treated and untreated retinal zones within the eye can be followed non-invasively over many weeks. SUMMARYMouse models utilized for retinal disease research including retinal vascular models can display nonuniform changes over the entire retina. Damage or loss of retinal layers and retinal neurons due to hypoxia can impact some retinal areas while leaving adjacent regions unaltered. Combining vascular imaging by fluoresceine angiography, vascular imaging and retinal layer imaging by SD-OCT, and focal-ERG provides us with new tools to examine retinal structure-function relationships within a single retina.