Mutant dominant-negative rhodopsin∆I256 causes protein aggregates degraded via ERAD and prevents normal rhodopsin from proper membrane trafficking

Abstract

Dominant mutations in the rhodopsin gene (Rho) contribute to 25% of autosomal dominant retinitis pigmentosa (adRP), characterized by photoreceptor loss and progressive blindness. One such mutation, Rho ^∆I256 , carries a 3-bp deletion, resulting in the loss of one of two isoleucines at codons 255 and 256. Our investigation, using recombinant expression in HEK293 and COS-7 cells, revealed that Rho ^∆I256, akin to the known adRP mutation Rho ^P23H, induces the formation of rhodopsin protein (RHO) aggregates at the perinuclear region. Co-expression of Rho ^∆I256 or Rho ^P23H with wild-type Rho ^WT, mimicking the heterozygous genotype of adRP patients, demonstrated the dominant-negative effect, as all isoforms were retained in perinuclear aggregates, impeding membrane trafficking. In retinal explants from WT mice, mislocalization of labeled adRP isoforms at the outer nuclear layer was observed. Further analysis revealed that RHO^∆I256 aggregates are retained at the endoplasmic reticulum (ER), undergo ER-associated degradation (ERAD), and colocalize with the AAA-ATPase escort chaperone valosin-containing protein (VCP). These aggregates are polyubiquitinated and partially colocalized with the 20S proteasome subunit beta-5 (PSMB5). Pharmacological inhibition of proteasome- or VCP activity increased RHO^∆I256 aggregate size. In summary, RHO^∆I256 exhibits dominant pathogenicity by sequestering normal RHO^WT in ER aggregates, preventing its membrane trafficking and following the ERAD degradation.

Keywords: ADRP; ERAD; VCP; dominant-negative effect; proteasome; retinitis pigmentosa; rhodopsin.

FIGURE 1

RHO^∆I256 mislocalizes to the cytosol and forms high molecular weight aggregates in vitro. Fluorescence images of HEK293 (A–C) and COS-7 cells (D–F) transfected with Rho ^WT -EGFP, Rho ^P23H -EGFP, or Rho ^∆I256 -EGFP (green). RHO^WT-EGFP is predominantly localized at the plasma membrane, while RHO^∆I256-EGFP (white arrows in C, F) and RHO^P23H-EGFP (white arrows in B, E) accumulate in the cytosol. Scale bar: 20 μm. There was a significant increase in the mean fluorescence intensity (MFI) in cells transfected with Rho ^∆I256 -EGFP or Rho ^P23H -EGFP cells compared to Rho ^WT -EGFP (G–H). Immunoblot analysis revealed that RHO^∆I256-EGFP forms HMW aggregates in both soluble (I) and insoluble cell fractions (J). Data were expressed as mean ± SEM. Significance was calculated by Mann-Whitney-U-test. **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 2

Rho ^∆I256 -EGFP and Rho ^WT -mcherry co-expression causes intracellular retention of WT rhodopsin. Fluorescence microscopy detecting EGFP (A, C, E, G, I, K) and mcherry (B, D, F, H, J, L) in HEK293 and COS-7 cells co-transfected with Rho ^WT -mcherry and Rho ^∆I256 -EGFP or Rho ^P23H -EGFP plasmids. (A1, C1, E1, G1, I1, and K1) show the merged images of the EGFP and mcherry channels. Arrows indicate the colocalization of RHO^WT-mcherry with RHO^∆I256-EGFP or RHO^P23H-EGFP. Magnified view is shown in (A1', C1', E1', G1', I1', and K1'). Arrowheads mark the mislocalized RHO^WT retained in misfolded aggregates (D, F, J, L). Scale bar: 10 μm.

FIGURE 3

In C57BL/6J retinal explants, the trafficking of endogenous RHO^WT is affected by transfection with Rho ^∆I256 . Immunofluorescence pictures reveal the localization of total RHO (red, second column), including endogenously expressed RHO and EGFP-tagged exogenously expressed RHO in three experimental groups in C57BL/6J retinae transfected with Rho ^WT -EGFP/XPMag (first row), Rho ^P23H -EGFP/XPMag (second row), and Rho ^∆I256 -EGFP/XPMag (third row). Exogenously expressed RHO (green, first column) is indicated by EGFP fluorescence. Exogenous RHO^∆I256-EGFP is mislocalized in the ONL (white arrowheads, green channel) and leads to the mislocalization of endogenous RHO^WT (red, third row). Moreover, in Rho ^∆I256 -EGFP/XPMag transfected C57BL/6J mouse retinal explants, mislocalized RHO^∆I256-EGFP partially overlaps endogenous RHO^WT (white arrows, merged close-up), and the OS layer is thinner. In contrast, exogenous RHO^WT-EGFP correctly targets the OS layer and colocalizes with endogenously expressed RHO^WT in transfected C57BL/6J mouse retinal explants. Scale bar: 20 μm.

FIGURE 4

RHO^∆I256 aggregates colocalize with ER and VCP. Immunofluorescence staining shows the localization of transfected Rho-EGFP (green, first column), the ER marker Calnexin (red, second column), and endogenous VCP (blue, third column) in HEK293 (A–I, A1, D1, G1) or COS-7 cells (J–R, J1, M1, P1). Colocalization is revealed in merged pictures (fourth column), and higher magnification pictures from merged insets are shown in column 5. RHO^WT is mainly localized to the plasma membrane (A and J). RHO^∆I256 aggregates localize with VCP to the ER (black arrows, D1’, M1’) and partially colocalize only with VCP (cyan arrows, D1’, M1’). RHO^P23H also accumulates in ER and VCP (black arrows, G1’, P1’). Scale bar: 10 μm.

FIGURE 5

Misfolded RHO^∆I256 is ubiquitinated. (A–L) Immunofluorescence of EGFP-tagged RHO^WT, RHO^∆I256, or RHO^P23H (green, first column) and Ub (red, second column) in HEK293 (A–F, A1, A1’-E1, E1’) and COS-7 cells (G–L, G1, G1’-K1, K1’). Colocalization is shown in merged pictures (third column). Ub is recruited into the aggregates pool, which is revealed as yellow dots in merged pictures in Rho ^∆I256 (black arrows, E1’, K1’) or Rho ^P23H transfected cells (black arrows, C1’, I1’). Scale bar: 10 μm; (M). Western blot results show enhanced smear Ub bands in cell lysates of Rho ^∆I256 or Rho ^P23H compared to Rho ^WT -transfected HEK293 cells. Ub expression is also present in untransfected control group. β-actin is used as housekeeping marker. (N). Quantification of Ub expression. Data are expressed as mean ± SEM. Statistical significance was determined by one-way ANOVA followed by Bonferroni’s multiple comparison test (*p < 0.05). Three independent experiments were included in the data analysis.

FIGURE 6

Rho^∆I256 is targeted to the proteasome system. Co-staining of EGFP-tagged RHO^∆I256, RHO^WT, or RHO^P23H (green, first column) with PSMB5 (red, second column) in HEK293 (A–E1’) and COS-7 cells (G–K1’). Colocalization is shown in merged pictures (third column). PSMB5 distributed in the cytosol and nucleus in Rho ^WT -EGFP transfected cells with no overlapping signal. In Rho ^∆I256 -EGFP transfected cells, PSMB5 partially accumulated in the cytosol (white arrows in F and L) and colocalized with RHO^∆I256 aggregates (black arrows in E1’ and K1’). Similar results were also seen in Rho ^P23H -EGFP transfected cells (white arrows in D and J; black arrows in C1’, I1’). Scale bar: 10 μm.

FIGURE 7

Proteasome and VCP inhibition increases HMW aggregates load in Rho ^∆I256 -transfected HEK293 cells. Western blot results for RHO in HEK293 cells after transfection with Rho ^∆I256 -EGFP (A), Rho ^P23H -EGFP (B), or Rho ^WT -EGFP (C) followed by treatment with DMSO, 5 µM MG132 or 2 µM ML240 for 6 h. The gray value changes in HMW rhodopsin aggregates bands above 180 kDa following ML132 or ML240 treatment relative to DMSO vehicle treatment is quantified and presented in (A1, B1, and C1). β-actin serves as housekeeping protein control. Data were expressed as mean ± SEM. Significance was calculated by a one-way ANOVA and Bonferroni’s multiple comparisons test. *p < 0.05. Three independent experiments are included in this data analysis.

FIGURE 8

Graphical abstract. Schematic illustration of the dominant-negative mechanism of RHO^∆I256 on RHO^WT and the degradation of these mutant proteins in the ERAD machinery (By FigDraw, ID: TWOIA64640). Misfolded RHO gets withheld in the ER and traps RHO^WT (a). Molecular chaperones, such as calnexin, calreticulin, and HSP70, identify misfolded proteins (b) and transport them to the ER retrotranslocation site, where the polypeptides are partially removed from the ER (c) and polyubiquitinated (d). Ubiquitination and VCP complex formation attract ubiquitin fusion degradation 1 (UFD1) and nuclear-protein localization 4 (NPL4) proteins to the retrotranslocon, which initiates the exit of the retrotranslocon from ER to cytoplasm (d). The complex retrotranslocates further into the cytoplasm (step d to e), where VCP-ATPase activity unfolds the protein (e). The unfolded VCP substrate is delivered to the proteasome for full degradation (f) (Vembar and Brodsky, 2008). Ubiquitin, VCP, and other cytosolic chaperones are then recycled.