|Home | About | Journals | Submit | Contact Us | Français|
There are an increasing number of ubiquitin ligases (E3s) implicated in endoplasmic reticulum (ER)-associated degradation (ERAD) in mammals. The two for which the greatest amount of information exists are the RING finger proteins gp78 and Hrd1, which are the structural orthologs of the yeast ERAD E3 Hrd1p. We now report that Hrd1 targets gp78 for proteasomal degradation independent of the ubiquitin ligase activity of gp78. The significance of this is underscored by the diminished level of a gp78-specific substrate, Insig-1, when Hrd1 expression is decreased and gp78 levels are consequently increased. These finding demonstrate a previously unappreciated level of complexity of the ubiquitin system in ERAD and have potentially important ramifications for processes where gp78 is implicated including regulation of lipid metabolism, metastasis, cystic fibrosis and neurodegenerative disorders.
Endoplasmic reticulum (ER)-associated degradation (ERAD) is the primary mechanism by which cells dispose of misfolded, unassembled and highly regulated proteins from the secretory pathway. This mechanism of quality control is critical to cellular homeostasis. When accumulated misfolded protein exceeds the combined capacity of the ER to refold and degrade them, ER stress and the unfolded protein response (UPR) ensue. One outcome of this is restoration of homeostasis, but if this fails, programmed cell death is initiated [1, 2]. These processes, ERAD, ER stress and the UPR are increasingly associated with human diseases. Targeting of proteins for ubiquitination and proteasomal degradation is an essential aspect of ERAD, with specificity largely determined by polytopic ER-resident Really Interesting New Gene (RING) finger ubiquitin ligases (E3s). In yeast there are two such E3s, Hrd1p/Der3p and Doa10p, both of which are dependent on their RING fingers for function. By far the more thoroughly studied of these is Hrd1p. This RING finger protein functions in complex with a number of other proteins in targeting substrates for degradation .
In mammals there are two Hrd1p orthologs . Mammalian Hrd1, which is also referred to as Synoviolin or as HsHrd1 in humans, targets ERAD substrates that include the ‘orphan’ G-protein coupled receptor, Parkin-associated endothelin receptor-like receptor (Pael-R) for ubiquitin-mediated degradation . It is also implicated in rheumatoid arthritis as a consequence of inhibiting apoptosis, hence the moniker Synoviolin . The other Hrd1p ortholog is gp78, also known as the human tumor autocrine motility factor receptor or RNF45. gp78 was the first described mammalian ERAD E3 and has known targets that include subunits of the T cell antigen receptor and proteins involved in cholesterol metabolism including hydroxymethylglutaryl coenzyme A (HMGCoA) reductase, insulin-induced gene-1 (Insig-1) and apoplipoprotein B [3, 6-9]. Cellular targets also include proteins implicated in neurodegenerative disorders, emphysema and cystic fibrosis and there is in vitro evidence for targeting of a critical protein involved in drug metabolism [10-13]. gp78 is the first example of a prometastatic E3. This is due, at least in part, to its targeting of the metastasis suppressor Kangai1 (KAI1)/CD82 for degradation in sarcoma . gp78 differs from Hrd1 in its complex domain structure, which in addition to its RING finger, includes an ubiquitin-binding coupling of ubiquitin conjugation to ER degradation (CUE) domain and a specific binding site for its cognate ubiquitin-conjugating enzyme (E2) Ube2g2. This site is known as the Ube2g2 Binding Region (G2BR). All three of these regions are required for the cellular function of gp78 . In some cases gp78 functions together with other E3s in targeting substrates for degradation [10, 13].
There is little known about the regulation of gp78 levels in vivo. One mechanism to control its levels is through stability. In this regard, we have established that gp78 catalyzes its own RING finger-dependent ubiquitination in vitro and undergoes ubiquitination and proteasomal degradation in cells in a RING finger-dependent manner [6, 15]. Notably, however, gp78 is found in complex with Hrd1 in cells . We therefore evaluated the functional relationship between these two ERAD E3s. We now report that Hrd1 targets gp78 for ubiquitination and proteasomal degradation with downstream effects on proteins targeted for degradation by gp78.
Transfections were carried out in human epithelial kidney (HEK) 293 cells from the American Tissue Culture Collection (ATCC, Manassas, VA, USA) maintained according to standard procedures. Wild type (WT) and Synoviolin−/− Mouse embryonic fibroblasts (MEFs) were maintained as described . HEK293 cells were transfected using Lipofectamine (Invitrogen, Carlsbad, CA, USA). For most experiments cotransfected green fluorescent protein (GFP) was used as an indicator of equal transfection efficiency. For all experiments cells were harvested between 24 and 48 h after transfection. Protein stability was assessed by inhibiting new protein synthesis with cycloheximide (CHX) (Sigma-Aldrich, St. Louis, MO, USA) at 50-80 μg/ml. MG132 was from Calbiochem (La Jolla, CA, USA). Protease inhibitor cocktail tablets were from Roche.
Plasmids encoding WT and double RING mutant gp78 (gp78R2M), small hairpin RNAs (shRNA) for gp78, human Hrd1 and a negative control (CTL) shRNA have been described [6, 14] as have plasmids encoding human Hrd1, Pael-R-FLAG , Insig-1-Myc  and HA epitope-tagged ubiquitin . Polyclonal Hrd1, gp78 and Nedd4 antibodies have been described [6,14, 22]. Polyclonal GFP and β-actin antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibody to Flag (M2) was from Sigma-Aldrich. Antibody to HA was from Roche (Basel, Switzerland).
For assessment of ubiquitination, cells were lysed in 1% SDS in 10 mM Tris-HCl pH 8, 150 mM NaCl [Tris-buffered saline (TBS)] by boiling twice for 5 min with vigorous vortexing in between. Samples were then diluted with TBS containing 1.5% Triton X-100. After centrifugation at 14,000 × g for 10 min, supernatants were immunoprecipitated for 2 h at 4°C, washed 3 times in 20 mM Tris-HCl pH 8, 120 mM NaCl, 1 mM EDTA, 0.5% NP-40 and subjected to SDS-PAGE and immunoblotted as described . For all other experiments cells pellets were lysed in mammalian protein extraction reagent (M-PER; Pierce, Rockford, IL, USA) and post-nuclear supernatants resolved by SDS-PAGE and immunoblotted. Protease inhibitors were added to all cell lysates.
Total RNA was isolated using RNeasy Mini Kit (QIAGEN, Valencia, CA, USA). A 1 μg aliquot of total RNA was reverse-transcribed using Superscript II Kit (Invitrogen). Real-time PCR was carried out using SYBER Green Master Mix on an ABI 7500 (Applied Biosystems, Foster City, CA). Values for specific genes were normalized to the 18S ribosomal subunit. Error bars represent the standard deviation from the mean from two independent determinations. Primers sequences were as follows:
To examine a possible effect of Hrd1 on the level of gp78, HEK293 cells were transfected with expression plasmids encoding gp78 together with WT Hrd1. As shown (Fig. 1A, left), transfection of Hrd1 resulted in a marked decrease in gp78. A long exposure of the gp78 immunoblot (right) shows an increased in higher molecular weight immunoreactive species with Hrd1 coexpression, consistent with ubiquitination. The effect of overexpressed Hrd1 on gp78 was confirmed for endogenous gp78 (Fig. 1B), with the specificity of the decrease demonstrated by a lack of change in the level of an unrelated non-ER E3, Nedd4. This effect was post-transcriptional, as gp78 transcript levels were unaffected by Hrd1 transfection (Supplemental Material Fig. 1), and determined to be consistent with the known proteasomal degradation of gp78  using MG132, a peptide aldehyde proteasome inhibitor (Fig. 1C). To assess whether Hrd1 results in an increase in ubiquitination of endogenous gp78, cells were transfected with Hrd1 together with HA-ubiquitin and proteasome function was then inhibited for 4.0 h with MG132 prior to lysis of cells under denaturing conditions that disrupt protein-protein interactions. Co-expression of Hrd1 led to a pronounced increase in immunoprecipitated ubiquitinated gp78 (Fig. 1D, top panel, lane 3). These findings are collectively consistent with enforced expression of Hrd1 leading to ubiquitination and proteasomal degradation of gp78.
To assess the significance of this result, we next asked whether endogenous Hrd1 similarly regulates gp78. Transfection of cells with a previously characterized Hrd1 shRNA resulted in a striking increase in endogenous gp78 protein compared to a control shRNA (Fig. 1E, compare lanes 1 and 3). The decrease in Hrd1 mRNA with Hrd1 shRNA transfection was ~50% (Supplemental Material Fig. 2), consistent with the effect on protein level determined by immunoblotting in other experiments (e.g. Fig. 2A).
To determine whether there was a reciprocal effect of gp78 on endogenous Hrd1, cells were transfected with gp78 shRNA. As shown (Fig. 1F) the levels of Hrd1 did not change despite a dramatic decrease in gp78. This lack of effect of loss of gp78 expression was confirmed by cycloheximide chase; similarly overexpression of gp78 was without effect on endogenous Hrd1 (data not shown). Thus, while Hrd1 decreases gp78 levels there is no evidence of a converse effect of gp78 on Hrd1.
To directly assess whether the decrease in gp78 is a consequence of enhanced degradation, loss of endogenous gp78 was assessed by cycloheximide chase after knockdown of Hrd1 (Fig. 2A). As is evident, diminished Hrd1 expression results in stabilization of gp78.
To determine whether this effect is seen with disruption of the Hrd1 gene, endogenous gp78 stability was examined using mouse embryonic fibroblasts (MEFs) from Synoviolin−/− (murine hrd1−/−) and matching WT (Synoviolin+/+) MEFs . As with shRNA-mediated knockdown in HEK293 cells, the half-life of gp78 was significantly prolonged in Synoviolin−/− MEFs comparing to WT MEFs (Fig. 2B). These results demonstrate that the effect observed is not cell type-specific and that disruption of murine Hrd1 substantially stabilizes endogenous gp78.
gp78 has previously been shown to target itself for ubiquitination in a RING finger-dependent manner as well as cooperate with other E3s in degradation of heterologous substrates [6, 10, 13]. To assess whether the Hrd1-mediated degradation of gp78 requires gp78’s E3 activity, a characterized inactive double mutant of the gp78 RING finger was employed. Consistent with previous observations , expression of this mutant resulted in a high level of stable gp78 (Fig. 2C). Strikingly, however, when Hrd1 was co-expressed the level and stability of mutant gp78 were dramatically decreased. Thus, while gp78 clearly plays a role in its own stability in a RING finger-dependent manner, the capacity of Hrd1 to target it for degradation does not require gp78’s ubiquitin ligase activity.
Insig-1, a polytopic ER membrane protein that plays multiple roles in regulating cholesterol synthesis, is a known ERAD substrate of gp78 . In contrast Pael-R, which is implicated in Parkinson’s disease, is a Hrd1 substrate [Ref. 4 and Y.C. Tsai and A. M. Weissman unpublished observations]. When Hrd1 was knocked down endogenous gp78 increased, and consistent with the E3 specificities of the two substrates, Pael-R level increased (Fig. 3A, right) while Insig-1 dramatically decreased (Fig. 3A, left). To demonstrate whether the reduction in Insig-1 seen with Hrd1 knockdown was due to proteasomal degradation, transfected cells were treated with MG132 (Fig. 3B). Consistent with proteasomal degradation the decrease in Insig-1 seen with Hrd1 knockdown was abrogated by MG132. Taken together these results demonstrate that degradation of gp78 mediated by Hrd1 impacts a pool of gp78 that is active in the regulation of its substrates.
The critical role of ubiquitination in the regulation of myriad cellular proteins makes it likely that components of this system will regulate each other as well as themselves. We have previously provided evidence for this with the demonstration that the Nedd4 family of E3s can target Cbl proteins for ubiquitination and proteasomal degradation . Similarly, substrate recognition elements of multi-subunit E3 have been shown to be targets for heterologous ubiquitin ligases . We now add a new dimension to the regulation of ubiquitin ligases by heterologous E3s by establishing that the degradation of one ERAD RING finger E3, gp78, is mediated by another member of this family, Hrd1. This regulation is not simply a manifestation of acute manipulation of protein levels as Hrd1 knockout MEFs similarly demonstrate increased levels and stability of gp78.
There are now multiple examples of active ligases, including gp78 in at least one case , collaborating in the degradation of substrates. In contrast, the targeting of gp78 for ubiquitin-mediated degradation by Hrd1 demonstrated herein does not require the ligase activity of gp78. How Hrd1 targets gp78 remains to be determined. It has already been established that these two E3s exist in the same complex with components of putative retro-translocons . Whether the effect we observed is dependent on this interaction and whether this requires a direct interaction between these polytopic transmembrane proteins remains to be determined.
The most important implication of these findings has to do with regulation of substrates. The increase in endogenous gp78 seen with Hrd1 knockdown is reflected in a specific decrease in the regulator of cholesterol metabolism and gp78 substrate, Insig-1. Similar results have also been obtained for another gp78 substrate KAI1 (A. Shmueli, Y. C. Tsai and A. M. Weissman unpublished observations). Thus, while Hrd1 is not implicated in targeting of gp78 substrates such as Insig-1, KAI1 and others, our findings demonstrate that it clearly has the capacity to impact the fate of these proteins by destabilizing gp78. The findings presented in this report therefore have potentially important implications for a number of homeostatic processes as well as pathological conditions and add a new level of complexity to our understanding of ERAD.
The authors thank Toshihiro Nakajima for generously supplying Synoviolin MEFs; Dirk Bohmann, Robert M. Gemmill, Ryosuke Takahashi and Emmanuel J.H.J.Wiertz for providing critical plasmids; and Aaron Ciechanover, Zlatka Kostova, Jennifer Mariano and Jocelyn D. Weissman for invaluable discussions and comments. This research is supported by the Intramural Research Program of the US National Institutes of Health, National Cancer Institute, Center for Cancer Research and in part by a grant from the Michael J. Fox Foundation for Parkinson’s Research.
Note to the editor: All abbreviations are defined in their first use and the following are only used once in the manuscript: HMGCoA, CUE, G2BR, Ub.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.