A–D. Cross-sections of 8-week-old retinas stained with H&E show disrupted morphology similar to that seen at 4 weeks in Cbp CH and R-DCKO mice. Scale bar = 50 μm. E–G. ERG testing shows persistence of the functional impairment in R-DCKO retinas (panels F and G). Dark-adapted b-wave deficits in some p300 CH mice tested at this time are reflected in the slightly decreased average and broad error bars for this genotype (Panel F orange line). Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted a-waves (Panel E), b-waves (Panel F), and light-adapted b-waves (Panel G). H–K. Cross-sections of 12-week-old retinas stained with H&E. Morphologic abnormalities in Cbp CH retinas (panel J) have resolved, although whorls and rosettes are still seen in R-DCKO retinas (panel K). Scale bar = 50 μm. L–N. ERG testing at 12 weeks revealed decreases in function in both Cbp CH and P300 CH, and R-DCKO retinas have lost cone responses in addition to rod function. Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted a-waves (Panel L), b-waves (Panel M), and light-adapted b-waves (Panel N). Asterisks (*) indicate p<0.001 vs. Cre negative controls in post-hoc tests.

Loss of p300/CBP leads to chromatin decondensation and changes in distribution of histone marks in R-DCKO nuclei.A. Electron micrographs of nuclei in the ONL of P22 retinas. Compared to Cre neg control littermates, compound heterozygotes (p300 CH and Cbp CH) show slight increases in euchromatin (light areas within nuclei). In R-DCKO nuclei areas of euchromatin are greatly increased, and electron-dense heterochromatin appears reduced. B. Heterochromatin was quantified as a percentage of the total nuclear area in 50 nuclei from 10 micrographs for each genotype. Error bars = 1 SD. Differences from Cre neg values were significant at p<0.0001. C & D. Comparison of immunoreactivity patterns for repressive histone marks H3K9me3 (green in panel C, white in insets) and H3K27me3 (green in panel D, white in insets) in control (left image) and R-DCKO (right image) retinas confirm loss of the characteristic rod chromatin condensation pattern in R-DCKO outer retina cells. Anti-PKC-alpha (red) marks bipolar cells. E & F. Comparison of immunoreactivity patterns for acetylated histone H3 (AcH3, green in panel D) and H4 (AcH4, green in panel E) reveals the redistribution of these activation marks in R-DCKO cells, corresponding to loss of the characteristic peripheral rod euchromatin distribution pattern. DNA is counterstained with Draq-5 (red). Scale bars: cross-sections = 20 µm, insets = 10 µm. G. Western blots of acid-extracted retinal histones from 15-week-old Cre-negative (1) or R-DCKO (2) retinas. CB, Coomassie blue stained gel. AcH3, blot stained for acetylated histone H3; AcH4, blot stained for acetylated histone H4. H. Quantification of band fluorescence intensities for AcH3 levels relative to total H2B levels, and AcH4 levels relative to total H3 levels at P20 did not show significant differences between Cre neg and R-DCKO samples.

Ep300/Cbp conditional knockout in cones also disrupts cone structure, gene expression and function.Cone opsin-driven Cre (CCre) was used to knock out Ep300/Cbp in cone photoreceptors; morphology, cone gene expression/distribution, and ERG function were assessed at 6–7 weeks of age. A–D. Compared to Cre negative controls (Panel A; inset shows two presumptive cones), H&E staining of CCre conditional knockout retinas reveals no major abnormalities (panels B–D), but cells with large nuclei can be seen scattered throughout the outer retina in C-DCKO mice (Panel D arrowheads and high-magnification inset). E. Cone arrestin (red) and p300 (green) expression are decreased in these cells (arrowheads). F. S-opsin expression (red) is also decreased in these cells (arrowheads), which lack outer segments. G. Peanut agglutinin labelling (red) identifies the displaced, abnormal cells in the outer retina (arrowheads) as cones. Blue in Panels E–G is DAPI labelling of nuclei. H. Cone α-transducin (green) is decreased and mislocalized to the cell bodies. Draq5 (red) marks nuclei. I. ERGs performed on 6 week old CCre mice confirmed decreases in cone-driven responses in C-DCKO retinas (red lines). Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted and light-adapted b-waves. Asterisks (*) indicate significance differences (p<0.001) from Cre negative controls in post-hoc tests.

A. H&E-stained sagittal sections of retinas from mice representing each indicated genotype (see Table 1), at the indicated ages. Irregularities appear in the ONL of R-DCKO mice at P10, and in Cbp CH retinas at P14. Scale bar = 50 μm. B. Immunofluorescence staining for p300 (green) shows progressive loss of p300 from ONL cells of R-DCKO retinas between P7 and P14. Scale bar = 25 μm. C. TUNEL staining for cell death was performed on three retinas of each genotype at P10, P14, and P32. TUNEL positive cells were only increased relative to age-matched Cre negative controls in Cbp CH retinas at P32 (arrow). D. P32 Cbp CH retina showing TUNEL positive cells (black arrowheads). These are frequently seen near ONL irregularities. OS, outer segments. Scale bar = 50 μm. E. P32 R-DCKO retina containing one TUNEL+ cell (black arrowhead). R-DCKO retinas do not show increased cell death relative to Cre-negative littermates at any age examined.

A. Electron micrographs of nuclei in the ONL of P22 retinas. Compared to Cre neg control littermates, compound heterozygotes (p300 CH and Cbp CH) show slight increases in euchromatin (light areas within nuclei). In R-DCKO nuclei areas of euchromatin are greatly increased, and electron-dense heterochromatin appears reduced. B. Heterochromatin was quantified as a percentage of the total nuclear area in 50 nuclei from 10 micrographs for each genotype. Error bars = 1 SD. Differences from Cre neg values were significant at p<0.0001. C & D. Comparison of immunoreactivity patterns for repressive histone marks H3K9me3 (green in panel C, white in insets) and H3K27me3 (green in panel D, white in insets) in control (left image) and R-DCKO (right image) retinas confirm loss of the characteristic rod chromatin condensation pattern in R-DCKO outer retina cells. Anti-PKC-alpha (red) marks bipolar cells. E & F. Comparison of immunoreactivity patterns for acetylated histone H3 (AcH3, green in panel D) and H4 (AcH4, green in panel E) reveals the redistribution of these activation marks in R-DCKO cells, corresponding to loss of the characteristic peripheral rod euchromatin distribution pattern. DNA is counterstained with Draq-5 (red). Scale bars: cross-sections = 20 µm, insets = 10 µm. G. Western blots of acid-extracted retinal histones from 15-week-old Cre-negative (1) or R-DCKO (2) retinas. CB, Coomassie blue stained gel. AcH3, blot stained for acetylated histone H3; AcH4, blot stained for acetylated histone H4. H. Quantification of band fluorescence intensities for AcH3 levels relative to total H2B levels, and AcH4 levels relative to total H3 levels at P20 did not show significant differences between Cre neg and R-DCKO samples.

Knockout of both Ep300 and Cbp in rods disrupts photoreceptor architecture and function.A–G. Cross-sections of 4-week-old retinas of the indicated genotypes (see Table 1), stained with hematoxylin and eosin (H&E). H. Section from the same R-DCKO eye as in panel G, fluorescently labeled with anti-PKCα (green, for bipolar cells) and DAPI (red), to show the boundary between the outer and inner nuclear layers. Scale bar = 50 μm for all 8 panels. I–K. Immunofluorescent staining for p300 protein verified expression in all nuclei in Cre neg controls (I). Ep300 expression is lost in the outer nuclear layer (ONL) of p300KO (J) and R-DCKO (K) retinas. L. R-DCKO section stained for p300 (green) and cone arrestin (CARR, red), showing that the few remaining p300-positive cells in the outer retina are cones. Scale bar = 20 μm for all 4 panels. OS, outer segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GC, ganglion cell layer. M–O. Amplitudes of dark-adapted (“Dark”) and light-adapted (“Light”) flash electroretinograms (ERG) at 4 weeks of age. Flash intensities (log [CdSec/M2]) are indicated on the X-axis. Error bars indicate +/− 1SD of the mean amplitude for 6 animals of each genotype tested. Two-way repeated measures ANOVA showed significant interactions between genotype and log light level at p<0.0001 for dark-adapted a-waves (Panel M), b-waves (Panel N), and light-adapted b-waves (Panel O). Asterisks (*) indicate values significantly different (p<0.001) from Cre negative controls in post-hoc tests.

Discover a faster, simpler path to publishing in a high-quality journal. PLOS ONE promises fair, rigorous peer review, broad scope, and wide readership – a perfect fit for your research every time.

Compound heterozygotes show age-dependent phenotypes.A–D. Cross-sections of 8-week-old retinas stained with H&E show disrupted morphology similar to that seen at 4 weeks in Cbp CH and R-DCKO mice. Scale bar = 50 μm. E–G. ERG testing shows persistence of the functional impairment in R-DCKO retinas (panels F and G). Dark-adapted b-wave deficits in some p300 CH mice tested at this time are reflected in the slightly decreased average and broad error bars for this genotype (Panel F orange line). Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted a-waves (Panel E), b-waves (Panel F), and light-adapted b-waves (Panel G). H–K. Cross-sections of 12-week-old retinas stained with H&E. Morphologic abnormalities in Cbp CH retinas (panel J) have resolved, although whorls and rosettes are still seen in R-DCKO retinas (panel K). Scale bar = 50 μm. L–N. ERG testing at 12 weeks revealed decreases in function in both Cbp CH and P300 CH, and R-DCKO retinas have lost cone responses in addition to rod function. Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted a-waves (Panel L), b-waves (Panel M), and light-adapted b-waves (Panel N). Asterisks (*) indicate p<0.001 vs. Cre negative controls in post-hoc tests.

Expression of photoreceptor genes is decreased in R-DCKO retinas.A. Summarized microarray findings for R-DCKO vs. Cre neg retinas. Each gene was categorized by the cell process in which it functioned, and results for each category are represented as a percentage of all the down- or up-regulated genes (see Supplemental Tables 2 & 3 for details). B. Schematic distribution of the 62 down-regulated photoreceptor or phototransduction-related genes in R-DCKO microarrays (red), compared with the 247 retinal disease loci listed in RetNet (https://sph.uth.edu/RetNet/home.htm; green) and a list of 230 genes down-regulated in Crx−/− retinas compiled from published sources [51]–[53]. Numbers in overlapping areas indicate the numbers of genes affected in both/all three conditions. All overlapping genes are listed in Supplementary Tables S4 and S5. C–F. Expression of the indicated rod gene (C. Nrl; D. Crx; E. Rhodopsin (Rho); F. Rod Transducin (Gnat1)) was assessed by quantitative RT-PCR (qRT-PCR) at P14, and is expressed as percent of the level of Cre-negative littermate controls (% Cre neg). Protein localization was verified by immunohistochemistry (IHC) at P30. Scale bar = 20 µm for all images. Levels of acetylated histone H3 (AcH3) or H4 (AcH4) on the respective promoter was determined by quantitative chromatin immunoprecipitation (qChIP) at P14, and is expressed as the value from the immunoprecipitated sample divided by the value from the untreated “input” sample, multiplied by 100 (“IP/input”).

Development of conditional knockout phenotypes.A. H&E-stained sagittal sections of retinas from mice representing each indicated genotype (see Table 1), at the indicated ages. Irregularities appear in the ONL of R-DCKO mice at P10, and in Cbp CH retinas at P14. Scale bar = 50 μm. B. Immunofluorescence staining for p300 (green) shows progressive loss of p300 from ONL cells of R-DCKO retinas between P7 and P14. Scale bar = 25 μm. C. TUNEL staining for cell death was performed on three retinas of each genotype at P10, P14, and P32. TUNEL positive cells were only increased relative to age-matched Cre negative controls in Cbp CH retinas at P32 (arrow). D. P32 Cbp CH retina showing TUNEL positive cells (black arrowheads). These are frequently seen near ONL irregularities. OS, outer segments. Scale bar = 50 μm. E. P32 R-DCKO retina containing one TUNEL+ cell (black arrowhead). R-DCKO retinas do not show increased cell death relative to Cre-negative littermates at any age examined.

A. Summarized microarray findings for R-DCKO vs. Cre neg retinas. Each gene was categorized by the cell process in which it functioned, and results for each category are represented as a percentage of all the down- or up-regulated genes (see Supplemental Tables 2 & 3 for details). B. Schematic distribution of the 62 down-regulated photoreceptor or phototransduction-related genes in R-DCKO microarrays (red), compared with the 247 retinal disease loci listed in RetNet (https://sph.uth.edu/RetNet/home.htm; green) and a list of 230 genes down-regulated in Crx−/− retinas compiled from published sources [51]–[53]. Numbers in overlapping areas indicate the numbers of genes affected in both/all three conditions. All overlapping genes are listed in Supplementary Tables S4 and S5. C–F. Expression of the indicated rod gene (C. Nrl; D. Crx; E. Rhodopsin (Rho); F. Rod Transducin (Gnat1)) was assessed by quantitative RT-PCR (qRT-PCR) at P14, and is expressed as percent of the level of Cre-negative littermate controls (% Cre neg). Protein localization was verified by immunohistochemistry (IHC) at P30. Scale bar = 20 µm for all images. Levels of acetylated histone H3 (AcH3) or H4 (AcH4) on the respective promoter was determined by quantitative chromatin immunoprecipitation (qChIP) at P14, and is expressed as the value from the immunoprecipitated sample divided by the value from the untreated “input” sample, multiplied by 100 (“IP/input”).

A–G. Cross-sections of 4-week-old retinas of the indicated genotypes (see Table 1), stained with hematoxylin and eosin (H&E). H. Section from the same R-DCKO eye as in panel G, fluorescently labeled with anti-PKCα (green, for bipolar cells) and DAPI (red), to show the boundary between the outer and inner nuclear layers. Scale bar = 50 μm for all 8 panels. I–K. Immunofluorescent staining for p300 protein verified expression in all nuclei in Cre neg controls (I). Ep300 expression is lost in the outer nuclear layer (ONL) of p300KO (J) and R-DCKO (K) retinas. L. R-DCKO section stained for p300 (green) and cone arrestin (CARR, red), showing that the few remaining p300-positive cells in the outer retina are cones. Scale bar = 20 μm for all 4 panels. OS, outer segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GC, ganglion cell layer. M–O. Amplitudes of dark-adapted (“Dark”) and light-adapted (“Light”) flash electroretinograms (ERG) at 4 weeks of age. Flash intensities (log [CdSec/M2]) are indicated on the X-axis. Error bars indicate +/− 1SD of the mean amplitude for 6 animals of each genotype tested. Two-way repeated measures ANOVA showed significant interactions between genotype and log light level at p<0.0001 for dark-adapted a-waves (Panel M), b-waves (Panel N), and light-adapted b-waves (Panel O). Asterisks (*) indicate values significantly different (p<0.001) from Cre negative controls in post-hoc tests.

Cone opsin-driven Cre (CCre) was used to knock out Ep300/Cbp in cone photoreceptors; morphology, cone gene expression/distribution, and ERG function were assessed at 6–7 weeks of age. A–D. Compared to Cre negative controls (Panel A; inset shows two presumptive cones), H&E staining of CCre conditional knockout retinas reveals no major abnormalities (panels B–D), but cells with large nuclei can be seen scattered throughout the outer retina in C-DCKO mice (Panel D arrowheads and high-magnification inset). E. Cone arrestin (red) and p300 (green) expression are decreased in these cells (arrowheads). F. S-opsin expression (red) is also decreased in these cells (arrowheads), which lack outer segments. G. Peanut agglutinin labelling (red) identifies the displaced, abnormal cells in the outer retina (arrowheads) as cones. Blue in Panels E–G is DAPI labelling of nuclei. H. Cone α-transducin (green) is decreased and mislocalized to the cell bodies. Draq5 (red) marks nuclei. I. ERGs performed on 6 week old CCre mice confirmed decreases in cone-driven responses in C-DCKO retinas (red lines). Two-way repeated measures ANOVA indicated significance at p<0.0001 for dark-adapted and light-adapted b-waves. Asterisks (*) indicate significance differences (p<0.001) from Cre negative controls in post-hoc tests.