Fig. 1.

HG-associated AMDf and photoreceptor cell aging in 24-mo-old mice is prevented by changing to the LG diet. (A) Photoreceptor cell layer thickness, measured as the thickness of the ONL, is decreased, particularly in the superior hemisphere, in HG mice. (B) A boxplot of the retinal damage score, calculated based on the area under the ONL thickness plot, shows increased retinal damage score in HG mice. (C) The retinal damage score increases continually over time in HG mice and increases beginning at age 18 mo in LG mice, but photoreceptor cell loss in HGxoLG mice ceases after dietary change. Negative values indicate ONL layers thicker than the reference (i.e., the average in 24-mo-old LG mice) (D–I) Toluidine blue-stained sections through the retinas of LG and HGxoLG mice have normal architecture, whereas retinas of HG mice show the indicated lesions. Square brackets in F, H, and I indicate regions with a missing RPE monolayer or hypopigmentation as indicated. Asterisks in F and H indicate regions of RPE multilayering. The arrows in G point to a region of vacuolated RPE and the arrowhead points to a subretinal deposit. (J) Retinas stained for the rod bipolar marker PKCα (red) and horizontal cell marker Calbindin (green) show ectopic rod bipolar dendrites (yellow) that extend beyond the outer plexiform layer (OPL) into the ONL (arrows) more frequently in HG mice than in HGxoLG or LG mice. (K) The frequency of ectopic rod bipolar cell processes that are greater than 5 μm (solid bars) or 10 μm (hatched bars) was quantified and found to be greater in HG retinas. Abbreviations: INL, inner nuclear layer; IPL, inner plexiform layer; ONL, outer nuclear layer. (Scale bars: 100 μm in D–I; 10 μm in J.) *P < 0.05; **P < 0.01; ^#P < 0.001; error bars indicate SEM; sample size was n = 7 (LG and HGxoLG), n = 9 (HG) in A and B; n = 4 (LG), n = 5 (HG) for 10-mo-old and n = 3 for 18-mo-old mice in C; and n = 4 in J.