Dynamic in vivo quantification of rod photoreceptor degeneration using fluorescent reporter mouse models of retinitis pigmentosa.

ABSTRACT

Imaging techniques have revolutionised the assessment of retinal disease in humans and animal models. Here we describe a novel technique for the in vivo visualisation of rod photoreceptors which permits semiquantitative assessment of outer retinal degeneration, and validate this approach in two mouse models of retinitis pigmentosa (RP). Transgenic mice carrying an Nrl-EGFP allele and homozygous for either knock-out of rhodopsin (Nrl-EGFP, Rho-/-) or heterozygous for knock-in of P23H mutant rhodopsin (Nrl-EGFP, RhoP23H/+) were used in this study. These novel strains have green fluorescent rods which undergo a progressive degeneration. Fundus imaging was performed at three-weekly intervals by near infrared reflectance (NIR) and blue light autofluorescence (BAF) confocal scanning laser ophthalmoscopy (cSLO). Mean grey values (mGV), which quantify fluorescence levels within such images, were compared for degenerate and age-matched non-degenerate (Nrl-EGFP, Rho+/+) controls. Mean grey value significantly decreased over time in the Rho-/- and RhoP23H/+ groups but was maintained in Rho+/+ mice (P < 0.001, two-way ANOVA). This corresponded to outer nuclear layer (ONL) thinning as observed by histology. The mGV of superior retina was significantly greater than that of inferior retina in RhoP23H/+ (P = 0.0024) but not in age-matched Rho+/+ (P = 0.45) or Rho-/- (P = 0.65) mice reflecting histological findings. Focal loss of rods could be visualised and mapped in vivo with this technique following a toxic insult, with thinning of the ONL being confirmed in hypofluorescent regions by spectral domain ocular coherence tomography (OCT). Fluorescence labelling of rods permits in vivo characterisation of models of RP and may provide new insights into patterns of degeneration, or rescue effect after treatment. mGV can be used in such cases as a semiquantitative metric of ONL degeneration, and can be used to identify regional variations in photoreceptor loss.