The yeast ERAD-C ubiquitin ligase Doa10 recognizes an intramembrane degron

Abstract

Aberrant endoplasmic reticulum (ER) proteins are eliminated by ER-associated degradation (ERAD). This process involves protein retrotranslocation into the cytosol, ubiquitylation, and proteasomal degradation. ERAD substrates are classified into three categories based on the location of their degradation signal/degron: ERAD-L (lumen), ERAD-M (membrane), and ERAD-C (cytosol) substrates. In Saccharomyces cerevisiae, the membrane proteins Hrd1 and Doa10 are the predominant ERAD ubiquitin-protein ligases (E3s). The current notion is that ERAD-L and ERAD-M substrates are exclusively handled by Hrd1, whereas ERAD-C substrates are recognized by Doa10. In this paper, we identify the transmembrane (TM) protein Sec61 β-subunit homologue 2 (Sbh2) as a Doa10 substrate. Sbh2 is part of the trimeric Ssh1 complex involved in protein translocation. Unassembled Sbh2 is rapidly degraded in a Doa10-dependent manner. Intriguingly, the degron maps to the Sbh2 TM region. Thus, in contrast to the prevailing view, Doa10 (and presumably its human orthologue) has the capacity for recognizing intramembrane degrons, expanding its spectrum of substrates.

Figure 1.

Sbh2 is a Doa10 substrate and association with Ssh1 protects it from degradation. (A) Schematic of heterotrimeric yeast Sec61 and Ssh1 translocon complexes. The integral membrane protein Sss1, which is part of both complexes, is not depicted in the illustration. N, N terminus; C, C terminus. (B) Cycloheximide (chx) chase analysis of ectopically expressed (low-copy plasmid; MET25 promoter) HA-Sbh2 (in the presence of endogenous Sbh2) in WT, doa10Δ, hrd1Δ, and doa10Δ hrd1Δ cells. Pgk1 served as a loading control. The experiment shown is representative of n = 3 experiments. (right) Quantification of the gel on the left. HA-Sbh2 levels at t = 0 min were set to 100%. (C) Ssh1 protects Sbh2 from Doa10-dependent degradation. chx chase analysis of ectopically expressed HA-Sbh2 (as in B) in WT, ssh1Δ, doa10Δ ssh1Δ, and hrd1Δ ssh1Δ cells. Two different exposures of the anti-HA immunodetection are shown. The graph at right shows the mean degradation rates observed from three independent experiments. HA-Sbh2 levels at t = 0 h were set to 100%. Error bars represent ± SD. exp., exposure. (D) Degradation of unassembled Sbh2. chx chase analysis (time points t₁ = 0 h and t₂ = 5 h) of ectopically expressed HA-Sbh2 (as in B) in WT, sbh2Δ, sbh2Δ ssh1Δ, and sbh1Δ ssh1Δ cells. Relative protein levels listed below the blots were determined by quantification of pixel densities of HA-Sbh2 bands relative to those of Pgk1. HA-Sbh2 levels of WT cells at t₁ = 0 h were set to 100%. (E) HA-Sbh2 is an integral membrane protein in ssh1Δ cells. Subcellular fractionation of doa10Δ ssh1Δ cells expressing HA-Sbh2 from a plasmid (as in B). HA-Sbh2–expressing ssh1Δ doa10Δ cells and Doa10-13MYC–expressing cells were mixed at a 5:1 ratio before lysis. Lysates were treated with buffer alone or buffer containing 2.5 M urea, 0.1 M Na₂CO₃, pH 11.5, and 0.5 M NaCl or 1% Triton X-100 (TX-100) and 0.5 M NaCl, and divided into microsomal pellet (P) and supernatant (S) fraction by centrifugation. Fractions were examined by immunoblotting with appropriate antibodies. (F) Fluorescence microscopy of WT, ssh1Δ, ssh1Δ doa10Δ, and get2Δ cells overexpressing HA-mCherry-Sbh2 from a low-copy plasmid under the strong TEF2 promoter. Bars, 1 µm.

Figure 2.

Sbh2 degradation requires Doa10 E3 ligase activity and proceeds via the 26S proteasome. (A) Doa10 E3 ligase activity is required for Sbh2 degradation. chx chase assay of ectopically expressed (low-copy plasmid; MET25 promoter) HA-Sbh2 in WT, doa10Δ, and doa10-C39S cells (DF5 strain background). (B) Sbh2 is ubiquitylated in cells. In vivo ubiquitylation of HA-Sbh2: HA- or FLAG-tagged Sbh2 was ectopically expressed (low-copy plasmid; GPD promoter) in ssh1Δ and doa10Δ ssh1Δ cells together with MYC-ubiquitin. HA-Sbh2 was precipitated from the cell lysates with anti-HA agarose beads. Precipitates were analyzed by immunoblotting with anti-HA and anti-MYC. Asterisks indicate cross-reactive bands (IgG heavy and light chain, respectively) recognized by the secondary antibody (anti–rabbit peroxidase). IB, immunoblotting. (C) Proteasomal degradation of Sbh2. ssh1Δ pdr5Δ cells ectopically expressing HA-Sbh2 (low-copy plasmid; MET25 promoter) were grown to log phase (0 h time point), divided into two cultures, and treated for 3 h with either the proteasome inhibitor MG132 (50 µM) or the solvent DMSO. Samples were normalized for equal amounts of the stable protein Pgk1 before gel loading. (D) Ubc6 and Ubc7 are required for efficient Sbh2 degradation. Degradation of ectopically overexpressed HA-Sbh2 (low-copy plasmid; GPD promoter) in WT, ubc6Δ, ubc7Δ, and ubc6Δ ubc7Δ cells (DF5 strain background). chx chase analysis was performed as in Fig. 1 B. Relative protein levels listed below the blots were determined by quantification of pixel densities of HA-Sbh2 bands relative to those of Pgk1. 0 h time point was in each case set to 100%. (E) The AAA-ATPase Cdc48 is required for Sbh2 degradation. Degradation of ectopically overexpressed HA-Sbh2 (low-copy plasmid; GPD promoter) in WT and cdc48-3 cells (W303-1A strain background). Cells were grown to log phase at 25°C and shifted to the nonpermissive temperature of 37°C 30 min before addition of chx.

Figure 3.

Mapping of the degron within Sbh2. Analysis of Ura3-HA-Sbh2 fusion proteins. (A) Schematic depiction of Ura3-HA-Sbh2 and Ura3-HA-Sbh1 fusion proteins used in this study. Sequence regions derived from Sbh1 are in light blue, and Sbh2 derived sequences are depicted as double black lines. (B) The last 48 residues of Sbh2 promote degradation of a stable protein. Degradation of ectopically expressed Ura3-HA-Sbh2(aa 41–88) (low-copy plasmid; MET25 promoter) in WT, doa10Δ, ssh1Δ, and doa10Δ ssh1Δ cells. chx chase was performed as in Fig. 1 B. (C) The last 32 residues of Sbh2 promote degradation of a stable protein. Degradation of ectopically expressed Ura3-HA-Sbh2(aa 57–88) as in B. (D) Ura3-HA-Sbh2(aa 57–88) is an integral membrane protein in ssh1Δ cells. Subcellular fractionation of doa10Δ ssh1Δ cells expressing Ura3-HA-Sbh2(aa 57–88) as in Fig. 1 E. S, supernatant; P, pellet; TX-100, Triton X-100. (E) Suppression of growth defect of sbh1Δ sbh2Δ cells at a high temperature by HA-Sbh2 and Ura3-HA-Sbh2 fusion proteins. Cells were transformed with an empty vector (p413MET25) or a p413MET25-based plasmid encoding HA-Sbh2 or the indicated Ura3-HA-Sbh2 protein. Serial dilutions (sixfold) of cultures were spotted onto plates, and plates were incubated as indicated. Empty vector and HA-Sbh2 lanes are from one plate (one plate for 30°C and a second one for 38°C). Both Ura3-HA-Sbh2(aa 41–88) and Ura3-HA-Sbh2(aa 57–88) lanes are from plates that were incubated parallel to the empty vector and HA-Sbh2 plates. (F) The Ura3-HA-Sbh1(aa 50–82) fusion protein is stable. chx chase with ectopically expressed Ura3-HA-Sbh1(aa 50–82) as in B. (G) Sbh1 is a stable protein. chx chase with ectopically expressed HA-Sbh1 (as in B) in WT and doa10Δ cells.

Figure 4.

Investigation of Sbh1/Sbh2 chimeric proteins. Both Sbh2 TM and ER-luminal regions contribute to the degron. (A) Schematic representation of Sbh1, Sbh2, and Sbh1/Sbh2 chimeric proteins used in this study. Sbh1-derived regions are depicted in light blue, and Sbh2-derived regions are shown in black. The nomenclature for Sbh1/Sbh2 chimeric proteins is as follows: “Sbh-XYZ” designates a chimeric protein in which the number at position X indicates the source of the cytosolic domain (“1”: Sbh1; or “2”: Sbh2), the number at position Y indicates the source of the TM helix, and the number at position Z indicates the source of the ER-luminal domain, respectively. (right) Comparison of Sbh1, Sbh2, and Sbh1/Sbh2 chimera TA sequences. Residues depicted bold in red represent identical residues shared between Sbh1 and Sbh2; residues depicted in light blue are Sbh1-specific residues, and residues depicted in black indicate residues specific for Sbh2; the black horizontal line indicates the TM helix, and the asterisks indicate Sbh2 residues Ser61 and Ser68, respectively. N, N terminal; C, C terminal. (B) Degradation of HA–Sbh-122. chx chase as in Fig. 1 B. Note: There are different chase times for WT and ssh1Δ strains (30-min chase) and doa10Δ and doa10Δ ssh1Δ strains (120-min chase). (C) Degradation of HA-Sbh1, HA–Sbh-121, and HA–Sbh-112. All blots are from same experiment/gel but were cropped for clarity (dashed line). (D) Degradation of HA-Sbh-221. (E) Degradation of HA–Sbh-211.

Figure 5.

TM-residue Ser68 of the Sbh2 tail anchor is a part of the degron. (A) Sbh2 TA sequence in complex with Ssh1 TM helices 1 and 4. Top view of the interface of Sbh2 and Ssh1. Sbh2 residues 58–79 and Ssh1 residues 26–53 (TM1) and 148–179 (TM4) are shown. Sbh2 Ser61 and Ser68 are depicted in red in ball and stick mode. Picture was generated with PyMOL using the atomic coordinates from Protein Data Bank accession no. 2WWA (Becker et al., 2009). (B) Degradation of HA-tagged Sbh2 WT and S61P and S68A point mutants in ssh1Δ cells. A representative blot is shown. chx chase was performed as in Fig. 1 B. The graph at right shows the mean degradation rates observed from at least three independent experiments. Error bars represent ± SD (C and D) HA-Sbh2(S61P) and HA-Sbh2(S68A) are stable in doa10Δ ssh1Δ cells. chx chase analysis of HA-Sbh2(S61P) and HA-Sbh2(S68A) stability. (E) HA-Sbh2(S61P,S68A) is stable even in ssh1Δ cells. chx chase analysis of HA-Sbh2(S61P,S68A) stability. (F) Suppression of growth defect of sbh1Δ sbh2Δ cells at high temperature by mutant Sbh2 variants (S61P), (S68A), and (S61P,S68A), respectively. Growth assay as in Fig. 3 E. Empty vector and HA-Sbh2 lanes are from one plate. (G) Co-IP analysis of digitonin-solubilized microsomes to investigate the interaction between Sec61 β subunits and Sec61. WT or mutant HA-Sbh2 was ectopically expressed in doa10Δ ssh1Δ cells and was precipitated with anti-HA agarose beads. Precipitates were analyzed by immunoblotting with the indicated antibodies. The TA Ubc6 protein served as a negative control. (H) Co-IP analysis to investigate the interaction between Sec61 β subunits and Sec61. WT HA-Sbh1 or HA-Sbh2 or mutant HA-Sbh2 was ectopically expressed in SEC61-9MYC ssh1Δ cells (W303-1B strain background), and Sec61-9MYC was precipitated with an anti-MYC antibody (rabbit polyclonal). Precipitates were analyzed by immunoblotting with indicated antibodies. IB, immunoblot.

Figure 6.

Model for Sbh2 quantity control via a Doa10-dependent intramembrane degron. (A) The TA Sbh2 protein associates with the Ssh1 protein in the ER membrane (1). Together with a third integral membrane protein, Sss1 (not depicted), Ssh1, and Sbh2 form the heterotrimeric Ssh1 complex implicated in cotranslational protein translocation in S. cerevisiae (besides the heterotrimeric Sec61 complex). Assembled Sbh2 is stable/protected from degradation. In contrast, unassembled Sbh2 (e.g., in ssh1Δ cells or surplus Sbh2 in WT cells) is readily degraded via a Doa10-dependent ERAD pathway involving ubiquitylation (2), retrotranslocation, and proteasomal degradation of Sbh2 (3). Ub, ubiquitin. (B) Schematic depiction of the Doa10-dependent intramembrane (ERAD-M) degron of Sbh2. The degron encompasses the Sbh2 TM domain and the short ER-luminal domain. The serine residue at position 68 (S68) located within the Sbh2 TM helix is a crucial part of the degron. (C) Assignment of the two major S. cerevisiae ERAD E3 ligases to different ERAD substrate classes. The identification of a Doa10-dependent ERAD-M substrate, Sbh2, allows assigning Doa10 to the ERAD-M pathway.