Stringent requirement for HRD1, SEL1L, and OS-9/XTP3-B for disposal of ERAD-LS substrates

Abstract

Sophisticated quality control mechanisms prolong retention of protein-folding intermediates in the endoplasmic reticulum (ER) until maturation while sorting out terminally misfolded polypeptides for ER-associated degradation (ERAD). The presence of structural lesions in the luminal, transmembrane, or cytosolic domains determines the classification of misfolded polypeptides as ERAD-L, -M, or -C substrates and results in selection of distinct degradation pathways. In this study, we show that disposal of soluble (nontransmembrane) polypeptides with luminal lesions (ERAD-L(S) substrates) is strictly dependent on the E3 ubiquitin ligase HRD1, the associated cargo receptor SEL1L, and two interchangeable ERAD lectins, OS-9 and XTP3-B. These ERAD factors become dispensable for degradation of the same polypeptides when membrane tethered (ERAD-L(M) substrates). Our data reveal that, in contrast to budding yeast, tethering of mammalian ERAD-L substrates to the membrane changes selection of the degradation pathway.

Figure 1.

Schematic representation of the seven canonical ERAD substrates used in this study. BACE476, CD3-δ, NHK_BACE, and NHK_CD3δ are type I membrane proteins (ERAD-L_M substrates); BACE476Δ, CD3-δΔ, and NHK are the corresponding soluble ERAD-L_S substrates.

Figure 2.

Involvement of HRD1 in disposal of membrane-tethered and soluble BACE476. (A) Radiolabeled BACE476 was immunoisolated after the indicated chase times from wild-type MEFs (wt; lanes 1–3) or MEFs lacking HRD1 (Hrd1^−/−; lanes 4–6). Relevant bands were quantified and plotted. (B) Same as described in A for BACE476Δ. (C) Radiolabeled BACE476Δ was immunoisolated after the indicated chase times from cells lacking HRD1 transfected with an empty plasmid (mock; lanes 1–3), a plasmid for expression of wild-type HRD1 (lanes 4–6), or a plasmid for expression of inactive HRD1 (HRD1*; lanes 7–9). Relevant bands were quantified and plotted. Error bars represent standard deviation (n = 2). Molecular mass markers are shown on the left for all gels (given in kilodaltons).

Figure 3.

Involvement of HRD1 in disposal of membrane-tethered and soluble CD3-δ. (A) Radiolabeled CD3-δ was immunoisolated after the indicated chase times from wild-type MEFs (wt; lanes 1–3) or from MEFs lacking HRD1 (Hrd1^−/−; lanes 4–6). Relevant bands were quantified and plotted. (B) Same as described in A for CD3-δΔ. (C) Radiolabeled CD3-δΔ was immunoisolated after the indicated chase times from cells lacking HRD1 transfected with an empty plasmid (mock; lanes 1–3), a plasmid for expression of wild-type HRD1 (lanes 4–6), or a plasmid for expression of inactive HRD1 (HRD1*; lanes 7–9). Relevant bands were quantified and plotted. Molecular mass markers are shown on the left for all gels (given in kilodaltons).

Figure 4.

Consequences of HRD1 and GP78 down-regulation on degradation of membrane-tethered and soluble variants of BACE476 and CD3-δ. (A) The efficiency of siRNA-based HRD1 and GP78 down-regulations were checked by immunoblotting. Tubulin is a loading control. (B) Radiolabeled BACE476 was immunoisolated after the indicated chase times from cells expressing a scrambled siRNA (siSCR; lanes 1–3), an siRNA targeting HRD1 (siHRD1; lanes 4–6), GP78 (siGP78; lanes 7–9), or both HRD1 and GP78 (siHRD1/siGP78; lanes 10–12). Relevant bands were quantified and plotted. (C–E) Same as described in B for BACE476Δ, CD3-δ, and CD3-δΔ, respectively. Molecular mass markers are shown on the left for all gels (given in kilodaltons).

Figure 5.

Disposal of soluble misfolded polypeptides relies on SEL1L and both OS-9 + XTP3-B. (A) The efficiency of siRNA-based SEL1L, OS-9, and XTP3-B down-regulations were checked by immunoblotting. Tubulin is a loading control. (B) Radiolabeled BACE476 was immunoisolated after the indicated chase times from cells expressing a scrambled siRNA (siSCR; lanes 1–3), an siRNA targeting SEL1L (siSEL1L; lanes 4–6), OS-9 (siOS-9; lanes 7–9), XTP3-B (siXTP3-B; lanes 10–12), or both XTP3-B and OS-9 (siXTP3-B/siOS-9; lanes 13–15). Relevant bands were quantified and plotted. (C–E) Same as described in B for BACE476Δ, CD3-δ, and CD3-δΔ, respectively. Molecular mass markers are shown on the left for all gels (given in kilodaltons).

Figure 6.

Trapping of BACE476Δ by OS-9.1 upon inactivation of the HRD1 pathway. (A) BACE476Δ was immunoisolated from detergent extracts of cells incubated with a scrambled siRNA, and siRNA targeting HRD1, GP78, GP78, and HRD1 (lanes 1–4) and SEL1L, OS-9, XTP3-B, and XTP3-B + OS-9 (lanes 6–9, respectively). Proteins were separated in SDS polyacrylamide gels and transferred on PVDF. The membranes were blotted with antibodies recognizing endogenous GRP94, BiP, OS-9.1, and OS-9.2 and BACE476Δ as a loading control. (B) Same as described in A for BACE476. Arrowheads indicate the coprecipitated OS-9.1. TCE, total cell extract; IP, immunoprecipitation.

Figure 7.

Involvement of HRD1 and GP78 in disposal of soluble and membrane-tethered NHK variants. (A) Radiolabeled NHK was immunoisolated after the indicated chase times from cells expressing a scrambled siRNA (siSCR; lanes 1–3), an siRNA targeting HRD1 (siHRD1; lanes 4–6), GP78 (siGP78; lanes 7–9), or both HRD1 and GP78 (siHRD1/siGP78; lanes 10–12). Relevant bands were quantified and plotted. (B and C) Same as described in A for NHK_BACE and NHK_CD3δ, respectively. Molecular mass markers are shown on the left for all gels (given in kilodaltons).

Figure 8.

Structure and composition of N-linked glycans. (A) The Asn-linked core oligosaccharide is composed of two N-acetylglucosamines (squares), nine mannoses (circles; the cleavable α1,2-bonded mannoses are shown in dark green), and three glucoses (triangles). The linkages are indicated, and letters a–n are assigned. a–c define the three oligosaccharide branches. (B) Aberrant oligosaccharide transferred to nascent polypeptide chains in B3F7 cells (Moremen and Molinari, 2006).