RNF185 is a novel E3 ligase of endoplasmic reticulum-associated degradation (ERAD) that targets cystic fibrosis transmembrane conductance regulator (CFTR)

Abstract

In the endoplasmic reticulum (ER), misfolded or improperly assembled proteins are exported to the cytoplasm and degraded by the ubiquitin-proteasome pathway through a process called ER-associated degradation (ERAD). ER-associated E3 ligases, which coordinate substrate recognition, export, and proteasome targeting, are key components of ERAD. Cystic fibrosis transmembrane conductance regulator (CFTR) is one ERAD substrate targeted to co-translational degradation by the E3 ligase RNF5/RMA1. RNF185 is a RING domain-containing polypeptide homologous to RNF5. We show that RNF185 controls the stability of CFTR and of the CFTRΔF508 mutant in a RING- and proteasome-dependent manner but does not control that of other classical ERAD model substrates. Reciprocally, its silencing stabilizes CFTR proteins. Turnover analyses indicate that, as RNF5, RNF185 targets CFTR to co-translational degradation. Importantly, however, simultaneous depletion of RNF5 and RNF185 profoundly blocks CFTRΔF508 degradation not only during translation but also after synthesis is complete. Our data thus identify RNF185 and RNF5 as a novel E3 ligase module that is central to the control of CFTR degradation.

Keywords: CFTR; ER Quality Control; ER-associated Degradation; Protein Degradation; Ubiquitin Ligase; Ubiquitination.

FIGURE 1.

RNF185 is an RNF5 homolog conserved in higher eukaryotes. Amino acid sequence alignment is shown for human (GI, 45708382) and mouse (GI, 15928691) RNF185 with their human (GI, 5902054), mouse (GI, 9507059), and Caenorhabditis elegans RNF5 (GI, 3874385) homologs. The two C-terminal membrane domains are underlined. The seven cysteine residues and the histidine residue constitutive of the RING domain are circled.

FIGURE 2.

RNF185 is a novel ubiquitously expressed E3 ligase. A, expression of RNF185 in mouse tissues. Total RNAs were purified from WT mouse tissues and were retrotranscribed for quantitative-PCR analysis using RNF185-specific primers. PPIA1 and 18 S RNA were used as references. Analysis was carried out on RNA samples extracted from tissues of three different mice. B, RNF185 can auto-ubiquitinate. Purified GST, GST-RNF185, and GST-RNF5 were incubated at 37 °C in the presence of ATP, ubiquitin (Ub), E1, and three different E2 enzymes. The reaction was next subjected to immunoblotting (IB) with anti-GST or anti-ubiquitin antibodies. C, schematic representation of the RNF185 constructs used in this study. RNF185 WT, wild-type RNF185; RNF185 ΔC, RNF185 with truncation of the most distal transmembrane domain; RNF185 RM, RNF185 with two punctal mutations in the RING domain; RNF185 ΔR, RNF185 mutant with total deletion of the RING domain. D, RNF185 ubiquitin ligase activity is dependent on the RING domain. GST-RNF185 WT and its RING mutant counterparts were processed as in B.

FIGURE 3.

RNF185 is an ER-localized E3 ligase. A, co-immunolocalization of RNF185 WT and RNF185 mutants with ER marker. HEK293 cells were co-transfected with the different RNF185 constructs and a plasmid expressing an ER-localized GFP. Very low doses (0.1 μg of DNA per well of a 24-well plate) of plasmid were used to prevent overexpression bias. Cells were fixed 24 h after transfection and processed for immunostaining using FLAG antibody. B, immunolocalization of endogenous RNF185. 24 h after transfection with ER-GFP plasmids, cells were fixed and processed for immunostaining using homemade RNF185 polyclonal antibody. C, RNF185 co-localization with a mitochondrial marker. RNF185 constructs were transfected in HEK293 cells. 24 h after transfection, mitochondria were visualized by the addition of MitoTracker to the culture medium before processing the cells for immunostaining using FLAG antibody.

FIGURE 4.

RNF185 interacts with ERAD components and is induced by UPR. A, RNF185 interacts with Derlin-1 and Erlin2. HEK293T cells were transfected with the control vector, FLAG-RNF185, or FLAG RNF5 (0.5 μg of plasmid per well of a 6-well plate). 24 h post-transfection, cells were lysed, and co-immunoprecipitation was performed using FLAG antibody. Immunoprecipitated (IP) proteins were loaded on reducing 14% SDS-PAGE and immunoblotted (IB) with antibodies against endogenous Derlin-1, endogenous Erlin2, or FLAG. B, RNF185 interacts with both enzymes of the Ubc6 family. HEK293T cells were transfected with the control vector or FLAG-RNF185 together with a plasmid expressing HA-UBE2J1 or myc-UBE2J2. Cells were then processed as in A. In this experiment, FLAG-RNF185 co-migrates with the antibody light chain as seen in the control lane (*). C, RNF185 expression is increased after tunicamycin treatment. HEK293 cells were treated with 2 μg/ml tunicamycin during the indicated times. Total RNAs were extracted and retrotranscribed. Q-PCR analysis was performed using RNF185-specific primers, and its expression levels were normalized to GAPDH levels (left panel). Results are shown as the mean of three independent experiments. Change in GRP78 expression was used as a control for UPR induction by tunicamycin (right panel).

FIGURE 5.

RNF185 induces the ubiquitin-proteasome-dependent degradation of CFTR proteins. A, RNF185 overexpression decreases the steady-state levels of WT CFTR and CFTRΔF508. Cells were co-transfected with control vector or increasing amounts of FLAG-RNF185 (0.5, 1, or 2 μg per well) and CFTR-HA or CFTRΔF508-HA. Low amount (0.1 μg per well) of a GFP-expressing plasmid was co-transfected in each condition, and the monitoring of GFP expression was used as a control for transfection efficiency. 24 h post-transfection, cells were lysed, and equal amounts of protein extracts were loaded on reducing SDS-PAGE for immunoblot (IB) with the indicated antibodies. GAPDH was used as a loading control. The core-glycosylated immature form and the mature-glycosylated form of CFTR are noted in B and C, respectively. Steady-state levels of CFTR and CFTRΔF508 were quantified using ImageJ software and were normalized to GAPDH and GFP levels. Results from three independent experiments have been plotted and are expressed as a percentage of the control (vector) condition. B, decrease in CFTR levels is dependent on RNF185 E3 ligase activity. Cells were co-transfected with control vector or vector expressing RNF185 WT, RNF185 RM, or RNF185 ΔR together with CFTR-HA or CFTRΔF508-HA. As in A, co-transfection with a GFP-expressing plasmid was used to monitor transfection efficiency. Cells were next processed as in A. C, RNF185 knockdown increases CFTR levels. Cells were co-transfected with CFTR-HA or CFTRΔF508-HA together with a control or a RNF185-specific siRNA. 48 h later, the cells were processed as described in A. RNF185 extinction was monitored after immunoprecipitation (IP) of the cellular extracts with anti-RNF185 antibody. Relative steady-state levels of CFTR proteins were quantified using ImageJ software. Results are expressed as a percentage of the control condition. D, RNF185 interacts with CFTR and CFTRΔF508. HEK293T cells were co-transfected with the indicated plasmids. HA-H3, a plasmid expressing HA-tagged histone H3, was used as a negative control for the immunoprecipitation (upper panel). Cells expressing WT RNF185 were treated with MG132 during 5 h before processing with the lysis. Co-immunoprecipitations were carried out with equal amounts of cell lysates using anti-HA antibody (upper panel) or anti-FLAG antibody (lower panel). The immunoprecipitates were next immunoblotted with the indicated antibodies. E, proteasome inhibition rescues RNF185-induced decrease in CFTR levels. HEK293 cells were co-transfected with the indicated plasmids and treated with ALLN or DMSO for 12 h, 24 h post-transfection. After cell lysis, the detergent-insoluble and -soluble fractions were subjected to immunoblot analysis with the indicated antibodies.

FIGURE 6.

Analysis of CFTRΔF508 degradation by cycloheximide chase. A, CHX chase analysis of CFTRΔF508 upon RNF185 expression. HEK293T cells were co-transfected with CFTRΔF508-HA together with a control vector or vector expressing RNF185 WT. GFP-expressing plasmid was co-transfected as a marker for transfection efficiency. 24 h later, protein extracts were prepared at the indicated time points after cycloheximide treatment (100 μg/ml) and loaded onto reducing SDS-PAGE. Immunoblotting (IB) was performed with the indicated antibodies. Relative changes in the half-life of CFTRΔF508 were quantified from three different experiments using ImageJ software and normalized to GAPDH and GFP levels. The obtained values were plotted against time. B, CHX chase analysis of CFTRΔF508 upon RNF185 knockdown. Cells stably expressing CFTRΔF508 were transfected with a control or an RNF185-directed siRNA. 48 h later, cells were treated with CHX and processed as in A. Down-regulation of RNF185 expression was controlled by Q-PCR (bottom panel in B).

FIGURE 7.

RNF185 targets CFTRΔF508 to co-translational degradation. A, measure of CFTRΔF508 labeling rates upon RNF185 overexpression. Cells were co-transfected with CFTRΔF508-HA together with RNF185 or the corresponding control vector. 24 h later, the cells were labeled with [³⁵S]Met/Cys radiolabeling mixture, and the synthesis of ³⁵S-labeled CFTRΔF508 protein was monitored over time by immunoprecipitating equal amounts of the labeled extracts with anti-HA antibody. ALLN or DMSO was added in the medium 90 min before labeling. Consistency of ³⁵S labeling between samples was controlled by loading the supernatants of the corresponding immunoprecipitation (IP) (depicted as lysates ³⁵S). Quantification of the experiment was performed using ImageJ software, and the intensity of labeled CFTR was normalized to the total amount of radioactivity initially present in the corresponding lysate. Results are expressed as a percentage of the vector condition quantified at 10 min in DMSO. RNF185 expression was confirmed by SDS-PAGE analysis (lower panel). IB, immunoblot. B, comparison of the experimental fitted curves (solid lines), accounting for the observed accumulation of ³⁵S-labeled CFTRΔF508 over time in the absence (blue line) and presence (red line) of RNF185, with the theoretical curve (dashed red line) predicting the accumulation of ³⁵S-labeled CFTRΔF508 if RNF185 only impacted the CFTR post-translational degradation rate. The experimental fitted curves were obtained as described under the “Appendix” and in supplemental Fig. S6. The theoretical RNF185 curve was obtained by setting equal rates of synthesis in the presence or absence of RNF185. C, measure of CFTRΔF508 labeling rates upon RNF185 knockdown. Cells stably expressing CFTRΔF508 were transfected using control or RNF185-directed siRNA. 48 h after transfection, the cells were labeled and processed as in A. Efficiency of RNF185 knockdown was controlled by Q-PCR analysis (right panel).

FIGURE 8.

Combined depletion of RNF185 and RNF5 synergistically blocks CFTRΔF508 degradation. A, analysis of CFTRΔF508 turnover upon combined RNF185 and RNF5 knockdown. HEK293 cells stably expressing a control shRNA or an shRNA sequence targeting RNF5 were co-transfected with CFTRΔF508-HA together with a control siRNA or a siRNA sequence targeting RNF185. 48 h later, the cells were treated with CHX for the indicated times and processed as in Fig. 6A. Immunoblotting (IB) following SDS-PAGE was performed using the indicated antibodies. Down-regulation of RNF185 expression was controlled by Q-PCR (right panel). B, relative changes in the half-life of CFTRΔF508 were quantified from three independent experiments using ImageJ software and normalized to GAPDH and GFP levels. The obtained values were plotted against time. Left panel is depicting relative values normalized to the control condition (control shRNA, control siRNA), where initial CFTR ΔF508 levels in this condition have been artificially normalized to 1. The right panel is depicting CFTRΔF508 intrinsic half-life after translation block, the initial time point for each condition being set at 100%.