REEP6 deficiency leads to retinal degeneration through disruption of ER homeostasis and protein trafficking

Abstract

Retinitis pigmentosa (RP) is the most common form of inherited retinal dystrophy. We recently identified mutations in REEP6, which encodes the receptor expression enhancing protein 6, in several families with autosomal recessive RP. REEP6 is related to the REEP and Yop1p family of ER shaping proteins and potential receptor accessory proteins, but the role of REEP6 in the retina is unknown. Here we characterize the disease mechanisms associated with loss of REEP6 function using a Reep6 knockout mouse generated by CRISPR/Cas9 gene editing. In control mice REEP6 was localized to the inner segment and outer plexiform layer of rod photoreceptors. The Reep6-/- mice exhibited progressive photoreceptor degeneration from P20 onwards. Ultrastructural analyses at P20 by transmission electron microscopy and 3View serial block face scanning EM revealed an expansion of the distal ER in the Reep6-/- rods and an increase in their number of mitochondria. Electroretinograms revealed photoreceptor dysfunction preceded degeneration, suggesting potential defects in phototransduction. There was no effect on the traffic of rhodopsin, Rom1 or peripherin/rds; however, the retinal guanylate cyclases GC1 and GC2 were severely affected in the Reep6 knockout animals, with almost undetectable expression. These changes correlated with an increase in C/EBP homologous protein (CHOP) expression and the activation of caspase 12, suggesting that ER stress contributes to cell death. Collectively, these data suggest that REEP6 plays an essential role in maintaining cGMP homeostasis though facilitating the stability and/or trafficking of guanylate cyclases and maintaining ER and mitochondrial homeostasis.

Figure 1

Generation of Reep6 KO mice. (A) Schematic of generation of an insertion mutation in exon 4 of Reep6 using CRISPR/Cas9 gene editing. (B) Genotyping of Reep6 WT, heterozygous, and homozygous KO mice using Sanger sequencing. (C) Expression and localization of REEP6 in mouse retina. REEP6 (green) appears to co-localise with anti-KDEL (red) in the rod photoreceptor inner segment (IS). (D) REEP6 (red) expression was restricted to the inner segment (IS) and outer plexiform layer (OPL) of mouse retina shown at P20. No staining was detected in the Reep6 KO mice.

Figure 2

Reep6 KO mice exhibit photoreceptor degeneration. Histological analysis of retinal sections from Reep6^+/- and Reep6^-/- mice was performed at P20 (A) 2 months (C) and 4 months (E) of age. Reep6^+/- retina shows normal layered organization compared with the Reep6^-/- retina that shows progressive thinning of the ONL. Spidogram of the number of photoreceptor nuclei rows in the ONL of Reep6^-/- mice retinas at P20 (B) 2 months (D) and 4 months (F) of age confirms the progressive loss of photoreceptors is presented as means ±SEM (n = 3). SD-OCT analysis showed thinning of the overall retina progressively from 3 months (G), 4 months (H) to 5 months (I) and thinning of the ONL (arrow) in Reep6 KO mice compared with Reep6 heterozygous mice. Transmission EM showing ultrastructure of the photoreceptors at P20 (J) and 6 months of age (K). ONL, outer nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer; IS, inner segment; D, dorsal; V, ventral. Scale bars = (A) left: 40 µm, right: 20 µm, (C) left: 40 µm, right: 20 µm, (E) left: 40 µm, right: 20 µm, (J) 2 µm, (K) 20 µm.

Figure 3

Reep6^-/- mice exhibit defects in response to light. Quantitative evaluation of scotopic a-wave and b-wave for dark-adapted P20 (A) 2 month old (B) and 4 month old (C) Reep6^+/- (black line) and Reep6^-/- mice (grey line). Reep6^-/- mice exhibited decreased ERG responses at all time points. NS, not significant. * Significant at P< 0.05. **Significant at P< 0.01. ***Significant at P< 0.001 determined by Student t-test.

Figure 4

P20 Reep6 KO mice have an increase in distal rod IS ER volume and proliferation of mitochondria. (A) Serial block face SEM images from different regions of rod photoreceptor inner segments (IS) used to generate (B) full IS ER and mitochondrial models. Measurements from serial block face SEM data. (C) Indicates an increase in ER area in slices within the Reep6^-/- distal IS, further backed-up by (D) measurements from models generated of the distal IS of Reep6^+/+, Reep6^+/- and Reep6^-/- rods showing a significant increase in ER volume in Reep6^-/- (measurements made from models in Supplementary Material, Fig. 2). (E) There are more mitochondria in rod IS of Reep6^-/- compared to Reep6^+/- and Reep6^+/+ mice. (F) TEM images highlighting ER (blue) and mitochondria (red) in sections of the distal portion of Reep6^+/+, Reep6^+/- and Reep6^-/- IS. Measurements from TEM images of the distal IS portion of 3 Reep6^+/+, Reep6^+/- and Reep6^-/- eyes show (G) an increase in ER area and (H) mitochondrial proliferation in Reep6^-/- mice. Reep6^-/- rods (average = 8) compared to Reep6^+/- rods (average = 6) and Reep6^+/+ (average = 5) had a significant increase in mitochondrial count. (D) Data are mean ± SD (9 IS). *P < 0.05 determined by Mann-Whitney U test (E) Data are mean ± SD (12 IS). *P < 0.05, **P < 0.01, ****P < 0.0001 determined by Student t-test (G & H) Results are means (three eyes) ± SEM. **P < 0.1, ***P < 0.001, ****P < 0.0001 determined by Student t-test. Scale bars = (A, B) 1µm, (F) 500 nm.

Figure 5

Trafficking of Rhodopsin in Reep6 KO mice. (A) Localization of rhodopsin (Rho; red), is not affected in Reep6^-/- retina at P20. Nuclei are counter-stained with DAPI (blue), in 72 h dark-adapted P20 mice. Immuno-electron microscopy (EM) of rhodopsin using ID4 (B) and RETP1 (C) in Reep6 KO mouse retina at P21 reveals no change in localization. (D & E) Immuno- EM of rhodopsin at 4 months revealed no change in localization. Scale bar= (A) 10 µm, (B, C, D, E) 200nm.

Figure 6

Loss of REEP6 leads to mislocalization of GC1, GC2, and PDE6. (A, B) Immunofluorescence localization of outer segment proteins GC1 and GC2 (red) is significantly reduced/absent in Reep6^-/- dark-adapted retina at P20 (A, right). (C) PDE6A is mislocalized in Reep6^-/- mouse retina to the ONL and INL regions. (D) Localization of CNGB1 (red) is unaffected. Nuclei are counter-stained with DAPI (blue). ONL, outer nuclear layer; INL, inner nuclear layer; IS, inner segment; OS, outer segment. Scale bar= (A–D) 10 µm.

Figure 7

Reep6^-/- mice exhibit ER stress in the retina. (A) ER stress was evaluated by immunostaining with ER stress marker Caspase 12 in retinal sections from P20 Reep6^+/- control and Reep6^-/- mice. Caspase 12 (green) positive staining (arrowheads) was observed in the ONL of Reep6^-/- mice retina but absent from the Reep6^+/- control retina. (B) CCAAT-enhancer-binding protein homologous protein (CHOP) (red) immunostaining was performed to assay ER stress. CHOP positive punctate staining was observed in Reep6^-/- retina specifically in the IS region of the retina. Nuclei are counter-stained with DAPI (blue). Scale bar= (A & B) 50 µm.

Figure 8

Diagram illustrating the consequences of REEP6 loss in the retina. (A) Under normal dark conditions, the phototransduction pathway is not activated by light and the cGMP levels remain basal, the CNG channels are open, and the rods are depolarized. (B) In the presence of light, GC converts cGMP to GMP, resulting in decreased cGMP levels, and the CNG channels close. This causes hyperpolarization of the rods. The normal function of REEP6 is to traffic GC from the ER to the OS, where it localizes. (C) In the Reep6 KO mice, trafficking of GC from the ER to the OS is disrupted, resulting in failure of cGMP hydrolysis, and causing failure of CNG channels to open. This leads to prolonged hyperpolarization of the rods photoreceptors.