Transmembrane E3 ligase RNF183 mediates ER stress-induced apoptosis by degrading Bcl-xL

Abstract

The accumulation of misfolded proteins in the endoplasmic reticulum (ER) causes ER stress and triggers the unfolded protein response (UPR). Failure to resolve ER stress leads to apoptotic cell death via a yet unclear mechanism. Here, we show that RNF183, a membrane-spanning RING finger protein, localizes to the ER and exhibits classic E3 ligase activities. Sustained ER stress induced by different treatments increases RNF183 protein levels posttranscriptionally in an IRE1α-dependent manner. Activated IRE1 reduces the level of miR-7, which increases the stability of RNF183 transcripts. In addition, overexpression of RNF183 leads to increased apoptosis and its depletion alleviates ER stress-induced apoptosis. Furthermore, RNF183 interacts with Bcl-xL, an antiapoptotic member of the Bcl-2 family, and polyubiquitinates Bcl-xL for degradation. Thus, RNF183 plays an important role in executing programmed cell death upon prolonged ER stress, likely by inducing apoptosis through Bcl-xL.

Keywords: Bcl-xL; RING finger protein; apoptosis; endoplasmic reticulum stress; unfolded protein response.

Fig. 1.

RNF183 is an ER-localized membrane protein that regulates apoptosis. (A) HeLa cells were transfected with GFP and GFP-tagged RNF183 as indicated. Apoptosis was measured by Annexin-V-PE/7-AAD staining and analyzed by FACS. (Top) Percentages of Annexin-V-positive cells. Error bars indicate the SEM. (Bottom) Immunoblotting (IB) of GFP and GFP-RNF183. β-Tubulin was used as a loading control. (B) Schematic representation of the predicted domain structure of RNF183. Amino acid numbers are shown. (C) HeLa cells expressing GFP-tagged RNF183 were fixed, permeabilized, and immunostained with anti-calreticulin, GM130, and cytochrome C antibodies. (Scale bar: 10 μm.) (D) Cell fractionation of KI-A6-HeLa cells (see SI Appendix, Fig. S2 for details). The total lysates (TL) were separated on a 30% Percoll gradient and the fractions subjected to IB. Calnexin, ER marker; Tim23, mitochondrial marker; GM130, Golgi marker. (E) The TL of KI-A6-HeLa cells were separated into soluble (S) and pelleted membrane (P) fractions and then immunoblotted for GFP or the cytosolic protein β-tubulin. The total membrane (TM) fractions were alkaline extracted and S and P fractions immunoblotted for GFP and the soluble protein BiP. (F) The lysates of KI-A6-HeLa cells were digested with proteinase K in the presence or absence of Triton X-100, followed by IB. Membrane-spanning ER protein calnexin was used for comparison. (G) Topological diagram of RNF183. MW, molecular weight (in all figures).

Fig. 2.

RNF183 is induced by prolonged ER stress. (A) KI-A6-HeLa cells were treated with 5 μM thapsigargin (Tg) for varying duration. The levels of GFP-RNF183, BiP, and PARP were analyzed by immunoblotting (IB). β-Tubulin was used as a loading control. The relative levels of RNF183 were quantified using the Gel-Pro analyzer software. (B) As in A, but with 10 μg/mL tunicamycin (Tm). (C and D) KI-A6-HeLa cells were transfected with control siRNA (NC) or two siRNAs against RNF183 (100 nM each, si1 and si2) and treated with Tg (C) or Tm (D) for another 36 h. The apoptosis ratio was determined by Annexin V-FITC/PI staining. Quantified data represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, t test. Levels of GFP-RNF183 were determined by IB using GFP antibodies, with GAPDH as a loading control. (E) As in C and D, but treated with Tg or Tm for another 24 h. (F) Quantitative RT-PCR analysis of RNF183 in HeLa cells after treatment with vehicle, 10 μg/mL Tm at varying time points. RNA expression levels are fold change of the target gene normalized to GAPDH. Graphs depict means ± SD of triplicates.

Fig. 3.

RNF is regulated by the IRE1 pathway. (A) KI-A6 cells were treated with Tm (10 μg/mL) or IRE1 inhibitor 4μ8C (50 μM) as indicated. The levels of RNF183, BiP, and PARP were analyzed by IB. β-Tubulin was used a loading control. Unspliced XBP1 (XBP1u) and spliced XBP1 (XBP1s) were amplified by RT-PCR and separated on an agarose gel. RNF183 protein level quantification was performed using the Gel-Pro analyzer software. (B) As in A, but with Tg (5 μM). (C) KI-A6-HeLa cells were treated with Tg (5 μM) for varying duration. The levels of GFP-RNF183 were analyzed by immunoblotting (IB) with β-tubulin being a loading control. XBP1u and XBP1s were amplified by RT-PCR and separated on an agarose gel. RNF183 protein level quantification was performed using the Gel-Pro analyzer software. (D) KI-A6-HeLa cells were treated with DMSO or Tg (5 μM) for 24 h, during which 4μ8C (50 μM) was added for indicated duration toward the end of Tg treatment. The levels of GFP-RNF183 were analyzed by IB. GAPDH was used a loading control. RNF183 protein level quantification was performed using the Gel-Pro analyzer software.

Fig. 4.

RNF183 is negatively regulated by miR-7. (A) Several miR mimics were transfected into KI-A6-HeLa cells for 24 h, and IB of GFP-RNF183 was performed. GAPDH was used as a loading control. RNF183 protein level quantification was performed using the Gel-Pro analyzer software. (B) Nonspecific miR control (NC), miR-7, and miR-17 inhibitors were transfected into KI-A6-HeLa cells for 24 h. Levels of GFP-RNF183 were analyzed by IB using GFP antibodies, with GAPDH as a loading control. RNF183 protein level quantification was performed using the Gel-Pro analyzer software. (C) KI-A6-HeLa cells were treated with Tg (5 μM) for indicated time. Mature miR-7 expression levels were determined by qPCR. Quantified data represent mean ± SEM. *P < 0.05, **P < 0.01, t test. (D) Flp-in-293 Myc-RNF183 cells were treated with DMSO, Tg, or 4μ8C as indicated. Mature miR-7 expression levels were determined by qPCR. Quantified data represent mean ± SEM. **P < 0.01; ns, no significant difference; t test. (E) NC, miR-7, or miR17 mimics was transfected into KI-A6-HeLa cells for 24 h. Relative RNF183 mRNA levels were determined by qPCR. Quantified data represent mean ± SEM. **P < 0.01; ns, no significant difference; t test. (F) KI-A6-HeLa cells were treated with DMSO, Tg, or 4μ8C as indicated. Relative RNF183 mRNA levels were determined by qPCR. Quantified data represent mean ± SEM. *P < 0.05, **P < 0.01; ns, no significant difference; t test. (G) Flp-in-293 Myc-RNF183 and Myc-RNF183-3′-UTR cells were treated with/without tetracycline (Tet), or with/without Tm as indicated. RNF183 transcript levels were analyzed by qPCR. (Top) Relative RNF183 mRNA levels. (Bottom) IB of Myc-RNF183; β-tubulin was used as a loading control. Quantified data represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, t test.

Fig. 5.

RNF183 is an E3 ligase. (A) Sequence alignment of RING finger domains. The conserved C₃HC₄ motif is highlighted in bold with a blue background. The red asterisks (*) indicate the two mutated cysteines in RNF183. Hs, Homo sapiens; Mm, Mus musculus. (B) GFP-RNF183 and Ub-HA were cotransfected into HeLa cells. RNF183 was precipitated using anti-GFP antibodies and its ubiquitination analyzed by immunoblotting using the indicated antibodies. (C) GFP or GFP-tagged RNF183 were transfected into HeLa cells at the indicated concentrations for 24 h and total ubiquitinated proteins analyzed with anti-Ub antibodies. (D) Purified recombinant GST-Myc-cytRNF183 (residues 1–161) or RING finger mutant (CS, C13S/C15S) were incubated at 37 °C for 1 h in the presence or absence of E1 (UBA1), E2 (UBCU5C), Ub, and ATP. Samples were subjected to immunoblotting using anti-Myc and anti-Ub antibodies.

Fig. 6.

RNF183 interacts with Bcl-xL. (A) GFP-RNF183 and Bcl-xL-2xHA were cotransfected into HeLa cells and the cell lysates precipitated using anti-GFP or anti-HA antibodies. Immunoblotting was carried out using the indicated antibodies. (B) As in A, for reciprocal IP. (C) Lysates from HeLa or KI-A6-HeLa cells were immunoprecipitated by anti-GFP antibodies and immunoblotted with anti-Bcl-xL and GFP antibodies. (D) Recombinant GST-Myc or GST-Myc-RNF183 was incubated with His-HA-Bcl-xL. The mixtures were precipitated with glutathione beads, followed by IB with anti-Myc and HA antibodies. (E) As in D, but with the indicated constructs. (F) As in A, but with the indicated constructs.

Fig. 7.

RNF183 ubiquitinates Bcl-xL for degradation. (A) HeLa cells were cotransfected with Ub-myc, Bcl-xL-2xHA, and GFP-RNF183 as indicated. The Top shows the schematic diagram of RNF183 constructs used. The red asterisks (*) indicate the mutated amino acids in the RNF183 mutants. Cell lysates were precipitated with anti-HA antibody and IB carried out with the indicated antibodies. (B) Purified recombinant His-HA-Bcl-xL or His-HA-Bcl-2 was incubated with GST-Myc-RNF183 for the in vitro ubiquitination assay as described above. (C) KI-A6-HeLa cells were treated with Tg or Tm at the indicated concentrations. Endogenous levels of Bcl-xL, RNF183, PARP, and β-tubulin were analyzed by IB. Relative Bcl-xL expression levels are shown. (D) KI-A6-HeLa cells were transfected with indicated siRNAs, followed by Tg or Tm treatment for 24 h. Cell lysates were subjected to IB with Bcl-xL, GFP, PARP, and β-tubulin antibodies. Relative Bcl-xL protein level is shown. Relative Bcl-xL expression levels are shown. (E) KI-A6-HeLa cells were treated with Tm and proteasome inhibitor MG132 as indicated. Endogenous Bcl-xL was precipitated by anti-Bcl-xL antibodies under denaturing condition; the ubiquitination levels of Bcl-xL were analyzed by anti-Ub IB. Endogenous levels of Bcl-xL, RNF183, BiP, PARP, and β-tubulin were also analyzed by IB. Relative ubiquitination and Bcl-xL levels are shown. (F) As in E, but with treatment of control siRNA or siRNAs against RNF183.