The mouse has become a desirable model for experiments designed to better understand normal and diseased retinal structure and function due to the ability to label specific cell classes and genetically modify retinal architecture. Advances in high-resolution, in vivo imaging technologies in the human eye have fueled increased interest in developing techniques to probe the living mouse eye on a microscopic spatial scale. Most studies that have acquired in vivo images of the mouse retina have used wide-field, macroscopic imaging techniques as the lateral resolution required to visualize individual cells and processes is limited by the eye’s optical aberrations. Recent experiments have demonstrated the use of adaptive optics to correct aberrations in the mouse eye and obtain high-resolution reflectance, fluorescence and 2-photon retinal images. However, the precise localization of microscopic structures imaged within this 3 dimensional, living tissue has proven challenging, with most studies determining positions in depth based on theoretical/empirical calculations and/or histological confirmation (which precludes longitudinal examination). Knowing the exact spatial location of cells imaged in vivo could have valuable consequences, including improved interpretation of the cell type being examined and characterization of changes in cellular structure and spatial location in disease.

In this paper, Zawadzki et al. present a clever technique that uses wide-field phase variance optical coherence tomography (pv-OCT) and scanning laser ophthalmoscope (SLO) images to axially localize cells fluorescently imaged in the same mouse eyes with an adaptive optics SLO (AOSLO). Given OCT’s inability to detect fluorescently labeled structures, the authors describe a custom-built AOSLO that can acquire images of single cells in the mouse eye in reflectance and fluorescence. After obtaining AOSLO reflectance and fluorescence images at the same focusing depths in the retina and axially registering AOSLO reflectance images of vasculature with volumetric pv-OCT maps of perfused vasculature at corresponding retinal locations, the authors show that it is possible to determine the exact axial position of each AOSLO reflectance and fluorescence image relative to a given retinal layer in the pv-OCT volume. Using this technique, the authors localize changes in the structural appearance of EGFP-expressing microglial cell bodies and processes (fluorescently imaged with an AOSLO) between different retinal layers. They also nicely differentiate whether punctate fluorescent structures seen in mice with fluorescently labeled cone photoreceptors were cone synaptic pedicles (more anterior) or cell bodies (more posterior), based on the focusing depth at which each structure was seen in the AOSLO image and the depth of its corresponding registered pv-OCT image in the retina.

In conclusion, Zawadzki et al. convincingly show that registration of high-resolution, high-magnification AOSLO images with wide-field pv-OCT and SLO images allows for effective depth localization and improved interpretation of fluorescently imaged cellular structures in the living mouse eye. While improvements are still required to precisely specify the position of fluorescent features in 3D relative to other structures in the retina, it will be exciting to follow future applications of this technique for localized examination of cellular features in normal and diseased rodent eyes. Such developments could open new avenues for the characterization of changes in cellular structure, movement and position with disease, the functional assessment of individual normal and diseased cells, and the targeted delivery of drug or laser-based therapeutic treatments to specific regions of the retina in normal mice and mouse models of disease.

---

---

 Add Comment

You must log in to add comments.