Photoreceptor Degeneration in Pro23His Transgenic Rats (Line 3) Involves Autophagic and Necroptotic Mechanisms

Abstract

Photoreceptor death contributes to 50% of irreversible vision loss in the western world. Pro23His (P23H) transgenic albino rat strains are widely used models for the most common rhodopsin gene mutation associated with the autosomal dominant form of retinitis pigmentosa. However, the mechanism(s) by which photoreceptor death occurs are not well understood and were the principal aim of this study. We first used electroretinogram recording and optical coherence tomography to confirm the time course of functional and structural loss. Electroretinogram analyses revealed significantly decreased rod photoreceptor (a-wave), bipolar cell (b-wave) and amacrine cell responses (oscillatory potentials) from P30 onward. The cone-mediated b-wave was also decreased from P30. TUNEL analysis showed extensive cell death at P18, with continued labeling detected until P30. Focused gene expression arrays indicated activation of, apoptosis, autophagy and necroptosis in whole retina from P14-18. However, analysis of mitochondrial permeability changes (ΔΨm) using JC-1 dye, combined with immunofluorescence markers for caspase-dependent (cleaved caspase-3) and caspase-independent (AIF) cell death pathways, indicated mitochondrial-mediated cell death was not a major contributor to photoreceptor death. By contrast, reverse-phase protein array data combined with RIPK3 and phospho-MLKL immunofluorescence indicated widespread necroptosis as the predominant mechanism of photoreceptor death. These findings highlight the complexity of mechanisms involved in photoreceptor death in the Pro23His rat model of degeneration and suggest therapies that target necroptosis should be considered for their potential to reduce photoreceptor death.

Keywords: apoptosis; autophagy; cell death; necroptosis; retinitis pigmentosa photoreceptor.

FIGURE 1

Fundus and OCT imaging images of P23H-3 and SD rat retinae. Representative fundus images SD (A) and P23H-3 (B) rats at P60. No morphological differences were apparent in fundus scans of P23H-3 rats at any age, compared to age-matched SD rats. Representative OCT images of the central retina in SD (C) and P23H-3 (D) rats at P60. Segmentation analyses of OCT images showing mean thicknesses of total retina (E), ONL (F), IPL (G), and GCL (H). Total thicknesses of P23H-3 retinae were significantly thinner than SD at P18, P60, and P90 (E). At P18 this was attributable to changes in the thickness of IPL (G) and GCL (H) but not the ONL (F). Changes in the ONL were only detected at P60 and P90 (F). All histograms indicate mean ± SEM, two-way ANOVA, Tukey’s post hoc test, *p < 0.05, **p < 0.001, and ****p < 0.0001, n ≥ 8 in each group. *: significant difference for genotype, #: significant difference for age. Abbreviations: ONL, outer nuclear layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. Scale bar (C,D) = 100 μm.

FIGURE 2

Alterations in rod responses of P23H-3 rats with age. (A–D) Representative rod responses in SD and P23H-3 rats at P18 (A), P30 (B), P60 (C), and P90 (D). (E–G) Box and whisker plots of rod ERG responses showing change in rod a-wave amplitudes (E), rod b-wave amplitudes (F), and rod OP amplitudes (G). Rod a-wave amplitudes were significantly reduced in the P23H-3 compared to the SD from P30 (E). The rod b-wave amplitudes were significantly attenuated in the P23H-3 rats from P30 onward (F). The rod a-wave (PIII) (E), rod b-wave (PII) (F), and summed rod OP (G) showed a progressive age-related reduction from P30 to P90 in both strains. All box and whisker plots show interquartile range (box), median (transverse line), mean (+) and 95% confidence intervals (error bars). (H) Relative percentage changes in rod a-wave, b-wave, and summed OP responses. At P30 and P60, the a-wave and b-wave responses were reduced in P23H-3 to a similar extent. However, the loss of summed OP amplitudes was significantly increased compared to those of b-wave. At P90, the extent of the b-wave loss was significantly less than a-wave and summed OP. Data were analyzed by two-way ANOVA with Tukey’s post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001, n ≥ 8 in each group. *, significant difference for genotype; #, significant difference for age.

FIGURE 3

Alterations in cone responses of P23H-3 rats with age. Representative modeled cone responses in SD and P23H-3 rats at P18 (A), P30 (B), P60 (C), and P90 (D). (E–H) Box and whisker plots of cone ERG responses in SD and P23H-3 rats showing change in cone b-wave amplitudes (E), cone b-wave time-to peak (F), cone OP amplitude amplitudes (G), and cone OP implicit times (H). Cone b-wave amplitudes were significantly lower in P23H-3 compared to age-matched SD controls at P30 and P60 (E). The cone b-wave timing and summed OP amplitudes were not significantly affected in P23H-3 at any age examined (F,G). OP implicit timing showed a significant increase in P23H-3 compared to SD only at P30 (H). All box and whisker plots show interquartile range (box), median (transverse line), mean (+), and 95% confidence intervals (error bars). Data were analyzed by two-way ANOVA, Tukey’s post hoc test, *p < 0.05, and ****p < 0.0001, n ≥ 8 in each group. *, significant difference for genotype; #, significant difference for age.

FIGURE 4

Photoreceptor cell death in P23H-3 rats with age. TUNEL (green) and nuclei (blue) staining in SD (A) and P23H-3 (B) retina at P18 (A,B). TUNEL positive cells (arrowheads) were detected predominantly in ONL of P23H-3 retinae across all ages but were rarely detected in the ONL of SD retinae. (C) Quantification of total retinal TUNEL⁺ cells at each age revealed significantly increased numbers of TUNEL⁺ cells in P23H-3 retina at P18, but not at P30 or P60. (D) Significant differences in the number of TUNEL⁺ cells across retinal eccentricity were detected in the P23H-3 retina at P30 with greater photoreceptor death in central and mid-peripheral than in peripheral regions of retina. (E) Quantification of mean ONL thickness showed significant decreases with age (#), with P23H-3 ONL significantly thinner than SD at P60, but not at P18 or P30. (F) Across central to peripheral eccentricity at P60 there were significant decreases in ONL in the central and mid-peripheral, but not peripheral regions. Data are expressed as means ± SEM, Two-way ANOVA, Tukey’s post hoc test, *p < 0.05, **p < 0.001, and ****p < 0.0001, n ≥ 5 in each group. ^∗: significant difference for genotype, ^#: significant difference for age. Abbreviations: ONL, outer nuclear layer; INL, inner nuclear layer. Scale bar = 20 μm.

FIGURE 5

Dysregulation of programmed cell death-related genes in P23H-3 retinae. Changes in programmed cell death related genes (fold regulation) in P23H-3 compared to age-matched SD rats at P14 (A) and P18 (B). Genes are clustered according to involvement in autophagy, apoptosis or necrosis pathways as defined by Qiagen. Data presented as mean ± SEM, Student’s t-test, *p < 0.05, **p < 0.01.

FIGURE 6

Mitochondrial membrane potential (Δψm) in P23H-3 and SD retinal cells at P18. Cell gating strategy for the analysis of retinal cells with different mitochondrial membrane potentials (A,E,I) was based on cell size and DAPI fluorescence. Three distinct subpopulations of cells were detected in all samples: small dead cells (B,F,J), small live cells (C,G,K), and large live cells (D,H,L). Representative plots for each of the three populations of cells show levels of polarized (red fluorescence) and depolarized (green fluorescence) mitochondria due to the JC1 dye, in CCCP-treated (B–D), SD (F–H), and P23H-3 (J–L) retinal cells. While no major fluorescence shifts could be detected for small live and large live cells between SD and P23H-3 samples, there appeared to be a slight shift from red- green for the small dead cells (Arrowheads; F,J). Quantification of red:green fluorescence intensity ratio in the three cell populations from P23H-3 and SD samples showed that small dead (M), but not small live (N) or large live cells (O), had a significant drop in red:green fluorescence intensity ratio (p < 0.0001; two-tailed unpaired t-test) in the P23H-3 retinae. Data presented as mean ± SEM, Student’s t-test, ***p < 0.001, n = 6 in each group.

FIGURE 7

Activated caspase-3 and AIF staining in P23H-3 retinae at P18. Confocal images of P18 retinal sections labeled with anti-cleaved caspase-3 (A, green) and AIF (B, green), showing only occasional cells were labeled. AIF is clearly detectable in the numerous mitochondria in the photoreceptor inner segments and also in the retinal pigmented epithelium and outer plexiform layer. Each of the labeled cells is shown at higher magnification in the three insets showing fluorescence due to DAPI (blue), the antibody (green) and the merged image. Abbreviations: ONL, outer nuclear layer; INL, inner nuclear layer. Scale bar = 20 μm (A,B); 2 μm (insets).

FIGURE 8

Autophagy markers (BECN1, ATG5, LC3B) in P23H-3 retinae at P18. Confocal images of P18 retinal sections labeled with BECN1 (A,B, green) and cytochrome oxidase (MTCO1; insets A,B, red), ATG5 (C,D, green) and MitoTracker (inset, D) and LC3B (E,F, green) in wild-type SD (A,C,E) and P23H-3 (B,D,F) retinae. Staining for BECN1 (B) and ATG5 (D) appeared to be mildly increased and more distinct in the P23H-3 retina. The yellow colored co-labeling with cytochrome oxidase (insets A,B, red) or MitoTracker (inset, D, red) indicates increased association of BECN1 and ATG5 with mitochondria in P23H inner segments. Labeling of phagosomes (arrowheads) in the inner segments with LC3B appeared to be increased in the P23H-3 (F) compared to SD retina (E). All sections were counter-stained with DAPI (blue) to label cell nuclei. Abbreviations: GCL, ganglion cell layer; INL: inner nuclear layer; IPL, inner plexiform layer; IS/OS, inner/outer segments; ONL, outer nuclear layer; OPL, outer plexiform layer. Scale bar: (A–F), 20 μm; insets, 10 μm.

FIGURE 9

RIPK3 and phospho-MLKL staining in P23H-3 retinae at P18. Confocal images of P18 retinal sections labeled with RIPK3 (A,B green) and phosphorylated (S345)-MLKL (C,D green) antibodies. No differences were noted in localization of the RIPK3 kinase in the wild-type SD (A) compared to P23H-3 (B) retina, with weak but specific staining observed in the inner segments of the photoreceptors (inset, B) as well as in the OPL, INL, IPL, and GCL. By contrast, there was a marked difference in the expression of phosphorylated MLKL, with dramatic induction in the outer segments of the P23H-3 retina (D), which was completely absent in the wild-type SD retina (C). The inset in panel (D), shows the distinct localization in the outer segments. Abbreviations: GCL, ganglion cell layer; INL: inner nuclear layer; IPL, inner plexiform layer; IS/OS, inner/outer segments; ONL, outer nuclear layer; OPL, outer plexiform layer; RPE, retinal pigmented epithelium. Scale bar = 20 μm.