A positive genetic selection for transmembrane domain mutations in HRD1 underscores the importance of Hrd1 complex integrity during ERAD

Abstract

Misfolded proteins in the endoplasmic reticulum (ER) are retrotranslocated to the cytosol for ubiquitination and degradation by the proteasome. During this process, known as ER-associated degradation (ERAD), the ER-embedded Hrd1 ubiquitin ligase plays a central role in recognizing, ubiquitinating, and retrotranslocating scores of lumenal and integral membrane proteins. To better define the mechanisms underlying Hrd1 function in Saccharomyces cerevisiae, several model substrates have been developed. One substrate is Sec61–2, a temperature sensitive allele of the Sec61 translocation channel. Cells expressing Sec61–2 grow at 25 °C because the protein is stable, but sec61–2 yeast are inviable at 38 °C because the mutated protein is degraded in a Hrd1-dependent manner. Therefore, deleting HRD1 stabilizes Sec61–2 and hence sec61–2hrd1△ double mutants are viable at 38 °C. This unique phenotype allowed us to perform a non-biased screen for loss-of-function alleles in HRD1. Based on its importance in mediating substrate retrotranslocation, the screen was also developed to focus on mutations in sequences encoding Hrd1’s transmembrane-rich domain. Ultimately, a group of recessive mutations was identified in HRD1, including an ensemble of destabilizing mutations that resulted in the delivery of Hrd1 to the ERAD pathway. A more stable mutant resided in a buried transmembrane domain, yet the Hrd1 complex was disrupted in yeast expressing this mutant. Together, these data confirm the importance of Hrd1 complex integrity during ERAD, suggest that allosteric interactions between transmembrane domains regulate Hrd1 complex formation, and provide the field with new tools to define the dynamic interactions between ERAD components during substrate retrotranslocation.

Introduction

Misfolded or incompletely processed lumenal and transmembrane proteins in the endoplasmic reticulum (ER) are retrotranslocated to the cytosol, ubiquitinated, and degraded by the proteasome. This process is referred to as endoplasmic reticulum-associated degradation (ERAD) and is conserved in all eukaryotes (Preston and Brodsky 2017). In Saccharomyces cerevisiae, the Hrd1 complex is one of two central components of the ERAD machinery based on its ability to contribute to both the ubiquitination and retrotranslocation of soluble lumenal and some integral membrane proteins that display misfolded lesions in the lumen or lipid bilayer, respectively (Bagola et al. 2011; Carvalho et al. 2006; Denic et al. 2006; Finley et al. 2012; Gauss et al. 2006; Mehnert et al. 2014; Sato et al. 2009; Wangeline et al. 2017; Wu and Rapoport 2018; Xie and Ng 2010). The other complex contains Doa10, which like Hrd1, is an integral membrane ubiquitin ligase in the ER membrane. In contrast to Hrd1, however, Doa10 primarily recognizes membrane proteins with misfolded lesions facing the cytosol (Ravid et al. 2006; Vashist and Ng 2004).

Hrd1 possesses eight transmembrane domains in the N-terminal portion of the protein and a RING domain facing the cytosol at the C-terminus. Biochemical analysis revealed that the Hrd1 core complex is comprised of Hrd1, Usa1, Hrd3, and Der1 (Carvalho et al. 2006; Christianson et al. 2011; Denic et al. 2006; Gauss et al. 2006). Hrd1 associates on its lumenal side with Hrd3 and on its cytosolic side with Usa1, which serves as a linker between Hrd1 and Der1. Usa1 interacts through its N-terminal cytosolic region with Hrd1 and through its C-terminal cytosolic region with Der1 (Gardner et al. 2000; Horn et al. 2009; Knop et al. 1996). In addition to bridging the interaction between Hrd1 and Der1, Usa1 also facilitates Hrd1 oligomerization (Horn et al. 2009). When digitonin-solubilized microsomes are subjected to sucrose density gradient analysis, most Hrd1 migrates at a position corresponding to a molecular mass of > 500 kDa, which is much larger than the size of a complex if it were instead composed of monomers of Hrd1, Usa1, Hrd3, and Der1. Indeed, deletion of USA1 shifts Hrd1 to a lighter fraction, although deletion of DER1 has no effect, suggesting that Usa1 also facilitates Hrd1 oligomerization (Carvalho et al. 2006; Nakatsukasa et al. 2013). Nevertheless, structural analyses have revealed that the Hrd1 complex functions as a monomer and the active Hrd1 complex contains a single copy of Der1 and Hrd3. Moreover, a monomeric Hrd1-Usa1 fusion protein supports ERAD-like activity (Wu et al. 2020). To date, the role of Usa1-dependent Hrd1 oligomerization in ERAD is unclear, although Hrd1 autoubiquitination might be sufficient to trigger the conversion of inactive Hrd1 oligomers into active monomers (Baldridge and Rapoport 2016). Together, there is clearly a need for better tools—and ideally genetic mutations—in which Hrd1 dynamics can be examined, especially in vivo.

In contrast to uncertainty surrounding the regulatory/oligomeric properties of Hrd1, it is abundantly clear that the Hrd1 transmembrane region plays a crucial role in ERAD. This domain directly recognizes proteins with lesions in the transmembrane domain (Sato et al. 2009), and the lumenal loop(s) within the transmembrane domain associate with Hrd3 (Gardner et al. 2000; Wu et al. 2020). In addition, the transmembrane domains of Hrd1 and Der1 form two halfchannels with cytosolic and lumenal cavities, respectively, and lateral gates facing one another in a thinned membrane region (Wu et al. 2020). To define the cellular activity of these transmembrane domains in Hrd1, which likely form the retrotranslocation channel, it will again be vital to employ genetic methods.

To these ends, mutagenic PCR was used to introduce random mutations in the transmembrane regions of Hrd1. To screen for ERAD-deficiency, we then used strains expressing the thermosensitive Sec61–2 protein as the only copy of this essential component of the ER protein translocation channel (Biederer et al. 1996). While Sec61–2 is stable at 25 °C, it is degraded by the proteasome in a Hrd1-dependent manner at 38 °C. This unique feature of the sec61–2 mutant allowed us to screen for viable loss-of-function mutations in HRD1 that resided in the sequence encoding the transmembrane domain. Biochemical analyses of Hrd1 function in the isolated mutant strains indicated that unstable as well as a relatively stable mutant, L201P, disrupted the Hrd1 complex. Based on the Hrd1 structure in isolation and as a member of the Hrd1 complex, our data suggest that allosteric relays within the Hrd1 protein are transmitted to peripheral members of the Hrd1 complex, which function together to orchestrate the destruction of ERAD substrates.

Materials and methods

Strains and culturing conditions

Saccharomyces cerevisiae strain YWO0547 (MATα ade2–1, ura3, trp1–1, his3–11,15, leu2–3,112, can1–100, prc1–1, der3/hrd1::HIS3, sec61–2) and YWO0431 (MATα ade2–1, ura3, trp1–1, his3–11,15, leu2–3,112, can1–100, prc1–1, der3/hrd1::HIS3) were generous gifts from Dr. D.H. Wolf (Universität Stuttgart). KNY392 (MATa can1–100, leu2–3,-112, his3–11,-15, trp1–1, ura3–1, ade2–1, pdr5△::HPH, pep4△::LEU2, hrd1△::natMX) was constructed by transforming KNY140 (MATa can1–100, leu2–3,-112, his3–11,-15, trp1–1, ura3–1, ade2–1, pdr5△::HPH, pep4△::LEU2) (Nakatsukasa et al. 2013) with hrd1△::natMX cassette that was amplified by PCR from pFA6a-natMX6 (Euroscarf P30437) (Hentges et al. 2005) using primers OKN1307 and OKN1308. KNY147 (MATa can1–100, leu2–3,-112, his3–11,-15, trp1–1, ura3–1, ade2–1, pdr5△::HPH, pep4△::LEU2, doa10△::CgHIS) was constructed by transforming KNY140 with doa10△::CgHIS cassette that was amplified by PCR from p1804 (National BioResource Project, JAPAN) using primers OKN173 and OKN174. Strains deleted for UBC7 were constructed by transforming cells with a ubc7△::kanMX4 cassette that was amplified by PCR from pRS400 (Brachmann et al. 1998) using primers OKN117 and OKN118. Cells were grown in YPrich medium (YPD: 1% yeast extract, 1% peptone, and 2% glucose) or synthetic complete medium (0.67% yeast nitrogen base without amino acids, all standard amino acids, and 2% glucose). Appropriate amino acids were excluded from synthetic media to maintain plasmids.

Plasmids

Plasmids and oligonucleotide primers used in this study are listed in Tables 1 and 2, respectively. A plasmid encoding P_HRD1-HRD1–3xHA-T_ADH1 (pKN10)was constructed as follows. A fragment containing P_HRD1-HRD1 was amplified from YCpDER3 (a generous gift from Dr. D.H. Wolf, Universität Stuttgart) with primers OKN46 and OKN59. In parallel, a fragment containing 3xHA-T_ADH1 was amplified from pFA6a-3HA-KanMX6 (Bahler et al. 1998) with primers OKN60 and OKN47. These fragments were digested with EcoRI-ClaI and ClaI-XhoI, respectively, and inserted into the same sites of pRS314 (Sikorski and Hieter 1989) to create pKN10. A plasmid encoding CPY*-3HA (pKN12–22) was a generous gift from Dr. J. S. Weissman (University of California, San Francisco), and a plasmid encoding Pdr5*-HA (pKN44) was constructed as follows. A pRS313-based plasmid (PWO0892) encoding Pdr5*-HA, a generous gift from Dr. D.H. Wolf (Universität Stuttgart) (Plemper et al. 1998), was digested with SacI-XhoI. The resulting fragment was inserted into the same sites of pRS316 (Sikorski and Hieter 1989) to create pKN44. A plasmid encoding KWW (pSM101) was kindly provided by Dr. D.T.W. Ng (National University of Singapore), and a plasmid encoding Pca1 was kindly provided by Dr. Jaekwon Lee (Adle et al. 2009).

Table 1.

Plasmids used in this study

Name	Description	Source
pKN10	pRS314: PHRD1-HRD1–3xHA-TADH1	This study
pKN12–22	pRS316: CPY*-3HA	Dr. J. S. Weissman
pKN22	pRS304: PHRD1-HRD1–3xHA-TADH1 (L10S, V229A)	This study
pKN23	pRS304: PHRD1-HRD1–3xHA-TADH1 (L154P, C168R)	This study
pKN24	pRS304: PHRD1-HRD1–3xHA-TADH1 (L62R)	This study
pKN25	pRS304: PHRD1-HRD1–3xHA-TADH1 (I164T, L185P, L201P)	This study
pKN26	pRS304: PHRD1-HRD1–3xHA-TADH1	This study
pKN27	pRS304: self-ligated at PvuII sites	This study
pKN44	pRS316: Pdr5*-3HA	This study
pKN303	pRS304: PHRD1-HRD1-THRD1	This study
pKN305	pKN303: L62R	This study
pKN306	pKN303: L51R	This study
pKN309	pKN303: V229A	This study
pKN310	pKN303: L154P	This study
pKN311	pKN303: C168R	This study
pKN312	pKN303: I164T	This study
pKN313	pKN303: L185P	This study
pKN314	pKN303: L201P	This study
pSM101	Encoding KWW, URA3	Dr. D.T.W. Ng
pKN515	Encoding Pca1, URA3	Dr. Jaekwon. Lee

Table 2.

Oligonucleotide primers used in this study

Name	Sequence	Note
OKN46	GCGGAATTCGTATCCTCATTCTATCCATA	pKN10
OKN47	GCGCTCGAGGGTTGTTTATGTTCGGATGT	pKN10
OKN59	GCGATCGATATGCTGGATAAATTTATC	pKN10
OKN60	GCGATCGATCCCGGGTTAATTAACATC	pKN10
OKN75	ACACCTACAAAAGTACACTGT	Mutagenesis
OKN76	AAGAATCTTGTGAATACGTCAA	Mutagenesis
OKN117	ATGTCGAAAACCGCTCAGAAACGTCTCCTCAAGGAGCTTCAACAGTTAATAGATTGTACTGAGAGTGCAC	ubc7Δ::KanMX4
OKN118	TCAGAATCCTAATGATTTCAAAATGGATAACTTTACCTGTCTCTCAAATTCTGTGCGGTATTTCACACCG	ubc7Δ::KanMX4
OKN173	TAGCCAAGAGTACCACTAATTGAATCAAAGAGACTAGAAGTGTGAAAGTCGTTGTAAAACGACGGCCAGT	doa10Δ::CgHIS
OKN174	TATATGTAAATATGCTAGCATTCATTTTAAATGTAAGGAAGAAAACGCCTCAGGAAACAGCTATGACCAT	doa10Δ::CgHIS
OKN1286	ATCTTATTAAATTCTACCTTAAGATGGCAACTCCTAACGAAACTA	Introduction of L62R
OKN1287	TAGTTTCGTTAGGAGTTGCCATCTTAAGGTAGAATTTAATAAGAT	Introduction of L62R
OKN1288	GAAGGCTTCAATCTAATGGTTAGATCGATATTCATCTTATTAAAT	Introduction of L51R
OKN1289	ATTTAATAAGATGAATATCGATCTAACCATTAGATTGAAGCCTTC	Introduction of L51R
OKN1300	GCGGAATTCGCACTATAGCCGCACGTAA	pKN303
OKN1301	GCGCTCGAGACAAAAGGAGAGTCGATATATCTACATA	pKN303
OKN1307	ATGGTGCCAGAAAATAGAAGGAAACAGTTGGCAATTTTTGTAGTTGTCACCGGATCCCCGGGTTAATTAA	hrd1Δ::natMX cassette
OKN1308	CTAGATATGCTGGATAAATTTATCTGGTATGACAATTTTCTTGGCAATTTGAATTCGAGCTCGTTTAAAC	hrd1Δ::natMX cassette
OKN1323	TCTAATGAGAACAACCATATTGCTCATGGCGATCCTACAGATGAA	Introduction of V229A
OKN1324	TTCATCTGTAGGATCGCCATGAGCAATATGGTTGTTCTCATTAGA	Introduction of V229A
OKN1325	TTTAGTAGATTCTCATTTAACCCAGTACTATTGGCGGTTGTAGAC	Introduction of L154P
OKN1326	GTCTACAACCGCCAATAGTACTGGGTTAAATGAGAATCTACTAAA	Introduction of L154P
OKN1327	GACTACCAGATAATAACACGAAGAATCTCCTCCATATATACAAAC	Introduction of C168R
OKN1328	GTTTGTATATATGGAGGAGATTCTTCGTGTTATTATCTGGTAGTC	Introduction of C168R
OKN1329	TTGGCGGTTGTAGACTACCAGACTATAACACGATGCATCTCCTCC	Introduction of I164T
OKN1330	GGAGGAGATGCATCGTGTTATAGTCTGGTAGTCTACAACCGCCAA	Introduction of I164T
OKN1331	AGTGATATTGAATCCACATCCCCATACCTGATACAAGTAATGGAG	Introduction of L185P
OKN1332	CTCCATTACTTGTATCAGGTATGGGGATGTGGATTCAATATCACT	Introduction of L185P
OKN1333	ACCATGCTTTTGATTGATTTGCCAAATTTATTCCTACAGACTTGT	Introduction of L201P
OKN1334	ACAAGTCTGTAGGAATAAATTTGGCAAATCAATCAAAAGCATGGT	Introduction of L201P

Antibodies and western blot analysis

Anti-HA monoclonal antibody was purchased from MBL (Tokyo, Japan), and anti-Hrd1 and anti-Cdc48 polyclonal antibodies were described previously (Nakatsukasa et al. 2013). Anti-Pgk1 monoclonal antibody was purchased from Invitrogen.

Proteins were transferred from polyacrylamide gels to Immobilon-P (Millipore) using GENIE Electrophoretic Transfer device (Idea Scientific Company) in blotting buffer (25 mM Tris, 192 mM glycine, 10% methanol) at a constant current of 650 mA. The membranes were washed with TBS-T buffer (20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 0.1% Tween-20) and blocked with 3% skim milk in TBS-T buffer for 30 min. The membranes were incubated with primary antibodies in TBS-T buffer overnight at 4 °C and washed three times with TBS-T (10–60 min for each wash). The membranes were then incubated with secondary antibodies for ~60 min and washed three times with TBS-T. The membranes were incubated with Chemi-Lumi One (nacalai tesque) or Luminata Forte Western HRP substrate (Millipore) and exposed to X-ray film. The band intensities were quantified with ImageJ (NIH). Immunoblots were decorated with the indicated primary antibodies and appropriate HRP-conjugated anti-mouse or anti-rabbit IgG secondary antibodies (Sigma Aldrich).

Genetic screening and construction of hrd1 mutant strains

To screen for mutations that reside in the Hrd1 transmembrane domain, the DNA sequence corresponding to this region was amplified by modified error-prone PCR. The error-prone PCR reaction was assembled as follows: 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 7 mM MgCl₂, 0.3 mM MnCl₂, 0.2 mM dATP, 0.2 mM dGTP, 1 mM dCTP, 1 mM dTTP, 50 pmol of primers OKN75/OKN76, ~10 ng of pKN10, and 5 units of Taq polymerase (Promega GoTaq). Reaction mixtures containing 0.05 mM MnCl₂ or 0.15 mM MnCl₂ were also prepared. One hundred microlitre of each mixture was split into 10×10 μL reactions. Amongst this group, a total of 30 reaction mixtures was first heated to 94 °C for 5 min, followed by 30 cycles of PCR with cycle parameters being 94 °C for 30 s, 55 °C for 1 min, and 72 °C for 2 min. Then, the solution was further incubated at 72 °C for 1 min before storage at 4 °C. During these steps, a gapped pKN10 plasmid was prepared by digestion with SalI and StyI. The resultant DNA fragments from PCR were separated on agarose gels, purified, and cotransformed with the gapped version of pKN10 into the YWO0547 strain, which contains an integrated copy of Sec61–2 and lacks Hrd1.

After plating transformation mixtures onto synthetic complete medium lacking tryptophan and incubation for 2–3 days at room temperature, cells containing plasmids that were repaired by homologous recombination between the mutagenized transmembrane domain fragment and the gapped vector were selected. Transformants were subsequently screened for growth on the same medium at 26 °C and 38 °C. From a total of 2475 transformants screened, we isolated 54 clones that grew at the non-permissive temperature. These clones were restreaked onto new plates and we chose the top 17 candidates. To exclude frameshift or nonsense codon-containing mutants (Fig. S1), we then examined the steady state levels of the Hrd1–3HA mutants at 38 °C and 26 °C by western blot analysis for the 12 clones whose phenotypes were the strongest. As a result of this strategy, four candidates were considered for experimental analysis. Next, plasmids were rescued from the four yeast strains, and they were amplified in bacteria, cut with EcoRI and XhoI, and the resultant P_HRD1-HRD1–3xHA-T_ADH1 fragment was inserted into the same sites of pRS304 to create pKN22 (L10S, V229A), pKN23 (L154P, C168R), pKN24 (L62R), and pKN25 (I164T, L185P, L201P). To develop an isogenic control, the EcoRI-XhoI fragment containing wild-type P_HRD1-HRD1–3xHA-T_ADH1 prepared from pKN10 was inserted into the same sites of pRS304 to create pKN26 (“WT”). All of these plasmids were also cut with HindIII, which is present inside the TRP1 gene in pRS304, and the vector sequence was subsequently integrated into the trp1–1 locus of YWO0547. It should be noted that the HindIII site, which is also present between the EcoRI and XhoI sites in the multi-cloning site in pRS304, was removed by inserting the wild-type or mutant HRD1 gene at the EcoRI-XhoI sites. Finally, as a vector-only control, pRS304 was cut with PvuII and the multi-cloning site was removed. The resultant linear plasmid was self-ligated to create pKN27, which was then cut with HindIII and integrated into the trp1–1 locus in YWO0547. Precise integration of all engineered yeast was confirmed by PCR.

Hrd1 mutants with single amino acid substitutions was constructed as follows. The DNA fragment encoding P_HRD1-HRD1-T_HRD1 was amplified from genomic DNA by PCR with primers OKN1300/OKN1301. The resultant fragment was digested with EcoRI-XhoI and inserted into the same sites of pRS304 to make pKN303. The mutagenesis reaction (30 μL) was typically assembled as follows: 1× Q5 buffer (New England Biolabs), 1×GC-rich enhancer, dNTP (2 mM each), ~60 ng of pKN303, 0.3 μM of primers listed in Table 2, Q5 PCR enzyme. This reaction mixture was first heated to 94 °C for 2 min, followed by 16 cycles of reaction with cycle parameters being 94 °C for 30 s, 55 °C for 1 min, and 72 °C for 6 min. Then, the template pKN303 was specifically digested with DpnI at 37 °C for 2 h, before the mixture was transformed into E.coli competent cells (Mach1, Thermo Fisher Scientific). All DNA sequences were verified by sequence analysis (Nagoya University Center for Gene Research). The resultant plasmids (pKN303, 305, 306, 309, 310, 311, 312, 313, and 314) were cut with HindIII, which is present inside the TRP1 gene in pRS304, and the linearized plasmids were subsequently integrated into the trp1–1 locus of KNY392.

Yeast growth assays

The indicated yeast strains were suspended in 1.5 mL of sterile water (typically, ~0.5 OD₆₀₀ equivalent/mL), and tenfold serial dilutions were generated as follows. Two hundred microliters of the cell suspension were transferred to the first lane of a 96-well plate (EVERGREEN). One hundred and eighty microliters of sterile water were then placed in the second, third, and fourth lanes before 20 μL of the cell suspension in the first lane was transferred to the second lane (lane 2) and mixed by pipetting ten times. This step was repeated. Finally, a 2.5 μL aliquot of the cell suspension was spotted on agar plates, which were then incubated at the indicated temperature.

Measurements of ERAD substrate stability

The degradation of ERAD substrates was analyzed by cycloheximide chase as described previously (Nakatsukasa et al. 2015, 2018, 2019). In brief, yeast cells were grown to log-phase (OD₆₀₀ = 0.4–0.8), and cycloheximide was added at a final concentration of 150 μg/mL. At the indicated time points, cells were treated with 30 mM sodium azide and after centrifugation the resulting cell pellet was stored at −80 °C. The cell pellet was then suspended in 300 μL of 20% TCA (trichloroacetic acid) and was lysed by vigorous vortex mixing with glass beads for ~30 min using a multi-tube vortex mixer (TAITEC Max Mixer EVR-032) with occasional inversion of the tube to prevent the accumulation of cells at the bottom of the Eppendorf tube. The broken cell lysate was added to 600 μL of 5% TCA, and 650 μL of the suspension was transferred to a new tube. The beads were further washed by adding 400 μL of 5% TCA, and 450 μL of this second suspension was pooled with above. Total protein was then precipitated by centrifugation at 20,000g for 15 min at 4 °C. Finally, the precipitates were dissolved by vigorous vortex mixing in either KNTCASB (80 mM Tris–HCl, pH 7.5, 8 mM EDTA pH 8.0, 12.5% glycerol, 8 M urea, 4% SDS, 200 mM DTT, 0.8 mg/mL Tris, 0.1% BPB) or 2×Laemmli SDS-PAGE sample buffer supplemented with 100 mM Tris (please note that the second use of Tris is unbu?ered, which helps raise the low pH that results from residual TCA after extraction). Samples were heated at 55 °C for 15 min and cleared by centrifugation at 20,000×g for 1 min at room temperature before they were subjected to SDS-PAGE and western blotting (see above).

Sucrose density gradient analysis

Sucrose density gradient analysis of the Hrd1 complex was performed as described previously (Nakatsukasa et al. 2013) with minor modifications. Briefly, cells were grown to log phase (OD₆₀₀ < 1.0), and 30–50 OD₆₀₀ equivalents of cells were harvested. Cells were disrupted with glass beads in lysis buffer (20 mM HEPES, pH 7.4, 50 mM KOAc, 2 mM EDTA, 0.1 M sorbitol, 1 mM DTT, 20 μM MG132, and complete protease inhibitor cocktail [Roche]) by agitation on a Vortex mixer eight times for 30 s with 30-s intervals on ice between each cycle. The homogenate was collected and pooled with rinses of the beads with buffer 88 (20 mM HEPES, pH 7.4, 150 mM KOAc, 250 mM sorbitol, 5 mM MgOAc). After unbroken cells were removed by centrifugation at 300g for 5 min at 4 °C, the supernatant was centrifuged at 20,000g for 20 min at 4 °C. The pelleted membranes were then solubilized in solubilization buffer (30 mM Tris–Cl, pH 7.6, 150 mM NaCl, 2 mM MgCl₂, 5% glycerol, and complete protease inhibitor cocktail) supplemented with 1% digitonin (Wako, Osaka, Japan) on ice for 30 min. Next, the lysates were cleared by centrifugation at 20,000g for 10 min at 4 °C, and the supernatant (420 μL) was layered onto a 5–40% sucrose step gradient containing 0.5% digitonin. Centrifugation was performed at 45,000 rpm for 3.5 h at 4 °C in a P55ST2 rotor (himac CP100NX). Fractions (210 μL) were collected from the top before proteins were precipitated with TCA and dissolved in KNTCASB by incubation at 42 °C. SDS-PAGE and western blotting was performed as described above.

Results

Experimental design

Yeast strains carrying the sec61–2 mutant allele exhibit a temperature sensitive phenotype at 38 °C because the level of Sec61–2 decreases due to Hrd1- and Ubc7-dependent ERAD (Biederer et al. 1996; Flury et al. 2005). Ubc7 is the E2 ubiquitin-conjugating enzyme that pairs with Hrd3 (a cognate E3 ubiquitin ligase) to both prime and polyubiquitinate ERAD substrates (Lips et al. 2020). Hrd1 is thought to catalyze this event by stabilizing a transition state between Ubc7 and ubiquitin during the reaction (Cohen et al. 2015), and Ubc7 activation also requires its association with an ER resident receptor, Cue1 (Metzger et al. 2013). Consistent with these combined data, it was previously shown that deletion of HRD1 or UBC7 restores the amount of Sec61–2 and hence rescues the growth defect of these cells (Biederer et al. 1996; Flury et al. 2005) (Fig. 1A). Moreover, sec61–2 yeast expressing a mutant form of Hrd1 that impairs the RING domain also exhibit improved growth (Bordallo and Wolf 1999). As shown in Fig. 1C, we too found that expression of wild-type Hrd1 from a low copy plasmid led to a growth defect of sec61–2hrd1△ cells at 38 °C. This unusual phenotype allowed us to screen for novel loss-of-function hrd1 mutants by selecting sec61–2hrd1△ yeast that contained a mutagenized HRD1 plasmid and grew at 38 °C (Fig. 1A).

Fig. 1.

A genetic screen to detect randomly generated mutations in the Hrd1 transmembrane domain. A Experimental design to screen for mutations in the transmembrane domain mutant of Hrd1. B The positions of mutations isolated in this study. C Growth analysis of sec61–2hrd1△ cells expressing the isolated hrd1 mutant alleles at 30 °C and 38 °C. Equal amounts of cells was subjected to the serial dilution analysis as described in the “Materials and methods”

Isolation of ERAD-defective HRD1 mutant alleles

Recent structural and biochemical studies have suggested that the portion of Hrd1 containing the eight transmembrane (TM) domains contributes to the retrotranslocation of lumenal substrates during ERAD (Wu et al. 2020). To better define the role of this region and the dynamic nature of the multi-protein Hrd1 complex in vivo, we sought HRD1 mutant alleles that encode amino acid substitutions residing in the region enriched for TM domains. To this end, the DNA sequence encoding this region was mutagenized by error-prone PCR, the fragments were mixed with gapped low copy plasmids, and the gapped vector and the mutagenized fragments were co-transformed into sec61–2hrd1△ yeast cells (see “Materials and methods”). From this protocol, we obtained 2475 transformants at 25 °C and their growth was tested at 38 °C. From this total of 2475 transformants screened, we isolated 54 clones that initially exhibited the strongest growth phenotype at the non-permissive temperature. These clones were then restreaked onto new plates and a smaller subset (17 candidates) was chosen based on improved growth. To better focus our efforts, we then selected 12 clones whose phenotypes were the most prominent. Next, we examined the steady state levels of these 12 Hrd1–3HA mutants at 38 °C and 26 °C by western blot analysis. This was undertaken to exclude frameshift or nonsense codon-containing mutants (Fig. S1). As a result of this strategy, four candidates were considered further for experimental analysis (labeled #18, 26, 44, and 45 in Fig. S1). Plasmids were subsequently isolated from these clones and the DNA sequences corresponding to the TM region were analyzed. Ultimately, we detected the following mutations: L10S/V229A (#18), L154P/C168R (#26), L62R (#44), and I164T/L185P/L201P (#45) (Fig. 1B).

For subsequent analyses and to exclude artefacts arising from extra-chromosomal Hrd1 expression, the mutant genes were amplified by PCR, subcloned into the pRS304 integrating vector (Sikorski and Hieter 1989), and reintroduced into the trp1–1 locus in sec61–2hrd1△ cells (YWO0547). We confirmed that sec61–2hrd1△ yeast expressing the integrated Hrd1 mutants grew at 38 °C, in contrast to the sec61–2hrd1△ strain expressing wild-type Hrd1, which failed to propagate (Fig. 1C). In turn, sec61–2hrd1△ yeast lacking Hrd1 (“vector”) exhibited robust growth, consistent with published results and the fact that Sec61–2 is stabilized in the absence of Hrd1 expression (Biederer et al. 1996; Flury et al. 2005).

Novel loss-of-function Hrd1 mutants exhibit slowed degradation of substrates with lumenal and integral membrane lesions

The data presented above are consistent with the Hrd1 mutants exhibiting a loss-of-function phenotype. To confirm this hypothesis, we first examined the half-lives of known Hrd1-dependent ERAD substrates displaying lesions either within the lumen or within the membrane-spanning region. One well-established Hrd1-dependent substrate, CPY*, is a soluble ERAD substrate due to the presence of a missense mutation in an otherwise vacuole-targeted protease (Finger et al. 1993). Therefore, hrd1△ cells expressing the isolated Hrd1 mutants were transformed with a plasmid encoding CPY*, and these were next subjected to a cycloheximide chase analysis. As shown in Fig. 2A, CPY* degradation was significantly slowed in cells expressing the Hrd1 mutants (lanes 1–20), whereas CPY* was degraded rapidly in cells expressing wild-type Hrd1 (lanes 21–24). We next measured the stability of a membrane-embedded model substrate, Pdr5*, which contains 12 transmembrane domains and misfolded lesions residing within these domains (Egner et al. 1998), as well as KWW, which contains a single spanning membrane domain fused to a misfolded domain that resides within the lumen (Vashist and Ng 2004). Consistent with both substrates exhibiting Hrd1-dependent ERAD in some prior studies (Plemper et al. 1998; Vashist and Ng 2004), Pdr5* and KWW were also stabilized in hrd1△ cells expressing the new Hrd1 mutants or that contained the vector control relative to the wild-type (“WT”) control (Fig. 2B, C). It is important to note that the slowed migration of KWW in the blots, especially when the protein is stabilized, is consistent with outer chain glycosylation as a result of O-linked glycosylation (Vashist et al. 2001; Vashist and Ng 2004). It is also important to note that other hrd1 mutants with different altered substitutions within the transmembrane domain can exhibit variable effects on the degradation of Sec61–2 and Pdr5*, in contrast to the hrd1 mutants analyzed here (see (Sato et al. 2009) and “Discussion”).

Fig. 2.

The degradation of model ERAD substrates is slowed in cells expressing the isolated Hrd1 mutants. Degradation of A CPY*-3HA, B Pdr5*-3HA, and C KWW-3HA was assessed by cycloheximide chase in cells containing a vector control or expressing the indicated Hrd1 mutants or wild-type Hrd1. A western blot probed with anti-HA antibody to detect the indicated ERAD substrates and Pgk1 (as a loading control is shown). Note that the migration of KWW is slowed during the chase due to O-linked glycosylation (see text)

As described above, Hrd1 forms a dimer or oligomer in the ER membrane complex. However, the role of Hrd1 oligomerization during ERAD is incompletely understood (Wu et al. 2020). Therefore, we wondered if the isolated Hrd1 mutants exhibit a dominant phenotype when they coexist with wild-type Hrd1 at the ER membrane. If so, the protein might permanently capture wild-type Hrd1 subunits, thus freezing the complex in an inactive state. To test this possibility, we first crossed hrd1△ haploid cells (Matα) expressing CPY* with hrd1△ cells (Mata). As anticipated, hrd1△/hrd1△ diploid cells (Matα/a) lacked Hrd1 and were unable to support CPY* degradation (Fig. 3A, lanes 1–4). On the contrary, HRD1/hrd1△ diploid cells degraded CPY* proficiently when compared with the degradation rate in haploid yeast (compare Figs. 2, 3A, lanes 5–8). Next, diploid cells were constructed by crossing HRD1 (Mata) cells with a Matα strain expressing the L10S/V229A, L154P/C168R, L62R, and I164T/L185P/L201P Hrd1 mutants. In these diploid yeast, CPY* was degraded to the same extent as in HRD1/HRD1 diploids (Fig. 3B, lanes 9–24), indicating that the isolated Hrd1 mutants are recessive.

Fig. 3.

The Hrd1 mutations are recessive. A hrd1△/hrd1△ or hrd1△/HRD1–3HA diploid cells expressing CPY*-3HA were subjected to A cycloheximide chase analysis. CPY*-3HA and Hrd1–3HA were detected by western blotting with anti-HA antibody. Cdc48 served as a loading control. B Diploid cells expressing wild-type Hrd1, wild-type and mutant forms of Hrd1–3HA, or CPY*-3HA were subjected to a cycloheximide chase analysis. CPY*-3HA and Hrd1–3HA were detected by western blotting with anti-HA antibody. Cdc48 served as a loading control

The isolated Hrd1 mutants are themselves targeted for ERAD

During the diploid analysis above, we noticed that the mutated forms of Hrd1 appeared to be short-lived and were degraded with a half-life of < 40 min (see for example “Hrd1–3HA”, Fig. 3B). Previous studies suggested that Hrd3 stabilizes Hrd1 through intramembrane interaction (Gardner et al. 2000). In the absence of Hrd3, Hrd1 is self-ubiquitinated and degraded in an ER-associated ubiquitin conjugating enzyme (i.e., Ubc7)-dependent manner (Carroll and Hampton 2010). Therefore, we suspected that some or all of the Hrd1 mutants might be unable to interact with Hrd3 and thus were consequently degraded with the help of Ubc7. Alternatively, the mutants might be conformationally defective to such an extent that they themselves become ERAD substrates and are consequently degraded in a Ubc7-dependent manner. Consistent with either of these models, all of the mutants were stabilized when UBC7 was deleted (Fig. 4). If indeed the Hrd1 mutants are radically misfolded and targeted for ERAD, it will be important in future experiments to determine whether the Hrd1 mutants, in either the monomeric or dimer form (Horn et al. 2009; Wu et al. 2020), are auto-ubiquitinated either in cis or in trans (Baldridge and Rapoport 2016; Carroll and Hampton 2010; Peterson et al. 2019), or are potentially modified by other ubiquitin ligases.

Fig. 4.

The degradation of Hrd1 mutants is Ubc7-dependent. UBC7 or ubc7△ cells expressing wild-type or mutant forms of Hrd1 were subjected to the cycloheximide chase analysis. CPY* was detected by western blotting with anti-HA antibody. Pgk1 served as a loading control

Analysis of single point mutants of Hrd1

To more specifically define the effects—if any—of the hrd1 mutant alleles that contained multiple amino acid substitutions, we introduced single point mutations into the gene encoding wild-type Hrd1. In addition, the 3xHA tag was removed, and the mutants were expressed from the chromosomal TRP1 locus in KNY392, a strain derived from W303–1a (see “Materials and methods”). Therefore, the constructed mutants included: L62R, which is predicted to be located in TM2; L154P, which is located in TM5; I164T, which is located in TM5; V229A, which is located in the loop between TM6 and TM7; C168R, which is located in TM5; L185P, which is located in TM6; and L201P, which is located in TM6 (Figs. 1B, 5). Although it was not identified in the screen, we also constructed an L51R allele because this position is expected to reside on the same side of the second alpha-helix as L62 (Fig. 5A, B). It is noteworthy that, with one exception, each of the randomly generated mutant alleles reside within TM domains.

Fig. 5.

The positions of isolated and generated HRD1 mutant alleles mapped onto the Hrd1 cryo-EM structure. A The positions of L51, L62, L154, and L201 in Hrd1 were mapped onto the dimeric Hrd1 structure reported by Schoebel et al. (2017). Side-chains are displayed by space-filling. B The positions of L51, L62, L154, and L201 in Hrd1 were mapped onto the monomeric Hrd1 (green)-Usa1 (yellow)-Der1 (light blue)-Hrd3 (magenta) structure reported by Wu et al. (2020). C An enlarged view from a different angle of the position of L62, L154, and L201 in Hrd1 in the monomeric Hrd1-Usa1-Der1-Hrd3 structure. D A ball-and-stick model depicting the L201 residue

Next, the expression of each integrated single mutation was analyzed by western blotting with anti-Hrd1 antibodies. With the exception of L51R, L62R, and L154P, the steady-state levels of the other mutant proteins were comparable to the amount of Hrd1 in a bona fide wild-type strain (Fig. 6A). Due to its position relative to L51R (Fig. 5A, B), it is perhaps not surprising that L62R was also expressed poorly compared to wild-type Hrd1. This result suggests that radical changes in the chemical/structural characteristics of amino acids in the second alpha-helix of Hrd1 are destabilizing.

Fig. 6.

Characterization of single point mutations in HRD1. A Steadystate levels of the indicated Hrd1 mutants were analyzed by western blotting with anti-Hrd1 antibodies. Pgk1 served as a loading control, and wild-type Hrd1 expression and yeast containing a vector control (vector) in the hrd1△ background served as a control. B CPY* degradation in cells expressing the indicated Hrd1 mutants was assessed by cycloheximide chase analysis followed by western blotting with anti-HA antibody. Pgk1 served as a loading control, and wild-type Hrd1 expression and yeast containing a vector control (vector) in the hrd1△ background served as a control. C Growth analysis of sec61–2hrd1△ cells expressing wild-type Hrd1, Hrd1L201P, or containing a vector control at 38 °C. Equal amount of cells were subjected to a serial dilution analysis as in Fig. 1. Plates were incubated at 30 °C and 38 °C. D CPY* degradation in cells expressing the isolated Hrd1 mutants was assessed by cycloheximide chase analysis at 30 °C. CPY* and Hrd1 were detected by western blotting with anti-HA antibody and anti-Hrd1 antibody, respectively. Pgk1 and Coomassie Brilliant Blue (CBB) staining of the membrane served as loading control. Quantified results are plotted in an accompanying graph (please see Fig. S2A, B). A representative result of three independent experiments is shown in this panel. E Pca1 degradation in doa10△ cells (KNY147) or in hrd1△ cells containing a vector control or expressing wild-type, L62R, or L201P Hrd1 was assessed by cycloheximide chase analysis. Pca1 and Hrd1 were detected by western blotting with anti-HA antibody and anti-Hrd1 antibody, respectively. CBB staining of the membrane served as a loading control

To determine which of the single amino acid substitutions exhibit a loss-of-function phenotype, we analyzed the rate of CPY* degradation in cells expressing the single mutant alleles. As shown in Fig. 6B (see Fig. S2A for full analyses and statistically validated results), CPY* was degraded proficiently in cells expressing I164T (TM5), C168R (TM5), L185P (TM6), and V229A (the loop between TM6 and TM7). In contrast, the unstable mutants [L62R (in TM2), L51R (in TM2), and L154P (in TM5)], were unable to support CPY* degradation. The one exception to this trend was L201P (TM6) (Fig. 5A–D). Although a cycloheximide chase analysis revealed that full length L201P was relatively unstable (Fig. 6D, E, see Hrd1 blots for L201P), as noted above the steady state levels of L201P were reproducibly comparable to wild-type Hrd1 (Fig. 6A, D, E; also see Fig. S2B for full analyses and statistically validated results). Nonetheless, CPY* degradation was slower in this mutant strain, thus indicating that expression of the L201P mutant leads to an ERAD defect (Fig. 6B, D). One explanation for why L201P was unstable, yet the steady-state levels appeared relatively similar to wild-type Hrd1, is that the mutant—consistent with its ERAD defect—induces the unfolded protein response (UPR). Indeed, Hrd1 is a known UPR target (Friedlander et al. 2000). Moreover, consistent with a loss of function phenotype for this allele, we also confirmed the ability of L201P—but not wild-type Hrd1—to fully restore viability to sec61–2hrd1△ yeast incubated at 38 °C (Fig. 6C). Importantly, in L201P as well as the other mutants, CPY* turnover was unchanged at 26 °C and 30 °C, indicating that the isolated Hrd1 mutants are not themselves thermosensitive (Fig. S2C). Finally, as a control, we demonstrated that the degradation of Pca1, which is a cadmium transporting P-type ATPase whose proteasome-dependent degradation is exclusively dependent on Doa10 (Adle et al. 2009), was unaffected in the L62R Hrd1 or L201P Hrd1-expressing mutant strains (Fig. 6E). The degradation of Ste6*, which is also primarily Doa10-dependent, was similarly unaffected (data not shown). This result establishes the specificity of the L62R and L201P alleles.

The L201P mutation in Hrd1 disrupts the Hrd1 complex

The integrity of the Hrd1 complex is essential to support the ERAD of Hrd1-dependent substrates, and previous studies have established that this Hrd1 “core” complex contains Hrd1, Hrd3, Usa1, and Der1 (Carvalho et al. 2006; Christianson et al. 2011; Denic et al. 2006; Gauss et al. 2006; Nakatsukasa et al. 2013). To investigate how TM domain mutations affect the formation of the Hrd1 complex, we performed a sucrose density gradient analysis. In contrast to co-immunoprecipitation assays to detect complex integrity, in which the efficiency of prey protein (in this case, Hrd3, Usa1, or Der1) capture also reflects the different amounts of the bait protein (i.e., Hrd1), a sucrose density gradient assay isolates protein complexes independent of this variable. Moreover, analysis of the sucrose density gradient fractions by SDS-PAGE is unaffected by the possibility that epitopes required for immunoprecipitation of a single protein might be buried or partially occluded within a complex. Our prior use of this technology allowed for an analysis of the oligomeric state of Hrd1 complex when the degradation was blocked in the absence of Cdc48/p97 activity (Nakatsukasa et al. 2013).

To this end, membrane fractions from select strains were solubilized with digitonin and resolved through a 5–40% continuous sucrose gradient by ultracentrifugation. Aliquots were collected, and the distribution of Hrd1 was analyzed by western blotting (Fig. 7). As a control, we first established that the Hrd1 peak (fraction 10) shifted to a lighter fraction (fractions 8–9 and 7–8, respectively) when Hrd3 or Usa1 was absent, consistent with a stoichiometric interaction between Hrd1, Hrd3, and Usa1 (Carvalho et al. 2006; Nakatsukasa et al. 2013). In hrd3△usa1△ double mutant cells, the peak appeared at even lighter fractions (fractions 5–6). When we examined the distribution of the residual Hrd1 in cells expressing the unstable L62R mutant, the Hrd1 peak appeared at fractions 8–9, which overlaps with the peak when Hrd3 is absent. This result again implies that L62R Hrd1 is degraded due to reduced Hrd3 binding, and/or that the radical change in its structure—which leads to destabilization—has secondary effects on complex integrity. Regardless, these data are consistent with a model in which a fraction of the stoichiometric Hrd1 complex is disrupted. Interestingly, in cells expressing the more stable L201P mutant, the complex also showed a broader distribution, one that is additionally in lighter fractions (fractions 4–13). These results again suggest that an alteration of L201 to a highly disruptive helix-breaking Pro residue disables the ability of Hrd1 to associate with other members of the complex.

Fig. 7.

Sucrose density gradient analysis of the Hrd1L62R and Hrd1L201P complexes. A Sucrose density gradient analysis of the Hrd1 core complex in wild-type or the indicated mutant yeast strains. Membrane fractions were solubilized with 1% digitonin and subjected to sucrose density gradient analysis. Fractions were collected and analyzed by western blotting with anti-Hrd1 antibody. Note that the membranes were exposed to film for different lengths of time to detect the protein. B The quantification of the results in (A) is shown. The distributions of Hrd1L62R and Hrd1L201P were compared with those of wild-type (WT) Hrd1 in wild-type, hrd3△, usa1△, and hrd3△usa1△ cells

Discussion

The HRD1 gene (also known as DER3) was initially isolated in screens that stabilized an ER membrane-resident enzyme, hydroxymethyl-CoA reductase, whose stability is metabolically controlled. The gene was also isolated in a screen for cells unable to degrade a model misfolded soluble protein that is selected for ERAD within the ER lumen (Bordallo et al. 1998; Hampton et al. 1996). Subsequent studies have established the central role of Hrd1 in facilitating the selection, retrotranslocation, and ubiquitination of numerous ERAD substrates. Even though Hrd1 function optimally requires a cohort of associated factors that make-up the Hrd1 complex, which includes Der1, Hrd3, and Usa1 (Bagola et al. 2011; Carvalho et al. 2006; Denic et al. 2006; Finley et al. 2012; Gauss et al. 2006; Mehnert et al. 2014; Sato et al. 2009; Wangeline et al. 2017; Wu and Rapoport 2018; Xie and Ng 2010), over-expression of Hrd1 is sufficient to restore degradation when other ERAD-requiring components are absent and has been shown to exhibit channel activity when reconstituted into lipid membranes (Baldridge and Rapoport 2016; Betegon and Brodsky 2020; Vasic et al. 2020). Structural analyses of Hrd1 in association with its partners have also revealed unprecedented insight into the organization of the Hrd1 complex, and models for how an ERAD substrate is liberated from the ER have been presented (Wu and Rapoport 2021). Yet, these biochemical analyses have lacked unbiased complementary genetic analyses that might yield new details about the structural/functional organization of the complex, uncover key residues that play critical roles in stabilizing or engaging Hrd1 partners, and/or provide the field with genetic tools that themselves can be used as a starting point for additional screens.

By performing an unbiased, PCR-based screen, we now report on a collection of mutations—residing in the N-terminal region of Hrd1 containing eight TM spans—that exhibit a range of defects in Hrd1 stability, ERAD, and Hrd1 complex integrity. After excluding truncation mutants or those whose expression was below detection, the remaining mutants were exclusively recessive loss-of-function alleles. We then focused our efforts on a group that included: (1) those that were unstable and were themselves targeted for ERAD, and—as shown for at least for one mutant (L62R)—led to a partial dissolution of the Hrd1 complex. This most likely occurred because the complex normally appears to exist in equimolar stoichiometry (Schoebel et al. 2017; Wu et al. 2020). (2) A mutant allele whose levels were comparable to wild-type Hrd1 but that similarly led to a partial disruption of the Hrd1 complex (L201P). This suggests that the L201P phenotype does not arise from it behaving as a hypomorph.

Based on a structural model of Hrd1 in isolation and within the Hrd1 complex, several hypotheses can be made about the underlying molecular defects of the L62R and L201P alleles (depicted in Fig. 5). Unexpectedly, the residues do not appear to face another Hrd1 monomer within the dimer, not do they seem to form appreciable interactions with other proteins in the Hrd1 complex. Consistent with this view, site-specific photocrosslinking at position L201 (in TM6) failed to uncover interactions with Der1 (Wu et al. 2020). More generally, TM6 is thought to line the translocation funnel and is situated ~5 amino acids away from the cytosolic face of the ER membrane (Schoebel et al. 2017). Thus, the ability of the L201P allele to destabilize the complex may derive from a major allosteric change in Hrd1, one that is translated to regions of the protein that interact with other Hrd1 complex members. It is noteworthy that L201 resides near the center of TM6, and we suggest that the helix-breaking properties of the Pro residue trigger this allosteric change. In contrast, the destabilizing L62R allele, situated in TM2, might result in new side chain interactions due to the presence of a novel charged Arg side chain. These illegitimate interactions might more globally destabilize the protein, potentially even leading to a change in the Hrd1 monomer–dimer transition. In order to firmly establish to what extent decreasing Hrd1 abundance correlates with the substrate stabilization, it will be important to analyze the stability of metabolically labeled substrate (i.e., pulse chase experiment of CPY*) in the absence of cycloheximide. In such an experiment, Hrd1-L201P would likely remain at a similar abundance as wild type Hrd1 throughout a time course.

Prior work from Hampton and colleagues indicated that the transmembrane domain of Hrd1 contains a code that is able to differentiate amongst different ERAD-membrane (“ERAD-M”) substrates, i.e., substrates with misfolded lesions that reside within or at the lipid bilayer. For example, the conversion of three hydrophilic amino acids within Hrd1 transmembrane helices to Ala results in a protein, 3A-Hrd1, that exhibited unique effects on the degradation of several proteins that were defined as ERAD-M substrates (Sato et al. 2009). Strikingly, another mutant, L61A, slowed the degradation of Sec61–2 but had no effect on Pdr5* turnover. In addition, in contrast to our work, none of the Hrd1 transmembrane domain mutants examined in the prior study affected the degradation of ERAD-lumenal (“ERADL”) substrates, such as CPY*. Interestingly, the deletion of USA1, which helps bridge components within the Hrd1 complex, slows the degradation of Sec61–2 but can also exhibit variable effects on other ERAD-M substrates (Carroll and Hampton 2010; Carvalho et al. 2006; Horn et al. 2009). In this case, however, the phenomenon most likely arises from the fact that the usa1△ strain results in a Hrd1 hypomorph. Finally, recent structural studies strongly suggest that a distinct component of the Hrd1 complex, Der1, plays a direct role in substrate recognition and retrotranslocation (Wu et al. 2020). Germane to this study, Sec61–2 degradation also requires Hrd3, which binds Hrd1 (Vashistha et al. 2016). Together, additional studies are clearly needed to define how diverse ERAD-M substrates are selected by Hrd1 alone and in combination with other members of the Hrd1 complex. To this end, the mutants generated in this study might serve as useful tools.

Yeast Hrd1 and the closest human homolog, HRD1/SYVN1, are 17.4% identical and 28.8% similar. Interestingly, L62 and L201 are identical between these proteins. Yeast Hrd1 and gp78/AMFR, another ER-embedded ubiquitin ligase that contributes to ERAD (Fang et al. 2001), are 16.0% identical and 28.5% similar. However, L62 and L201 are not identical in gp78/AMFR, although L62 is similar (Val). In the future, it will be interesting to define whether the results presented here can be recapitulated in human cell lines containing knock-in versions of L62R and L201P.

Finally, based on the results presented in this study, we propose that the continued use of the sec61–2 allele will allow for screens of small molecules that might modulate Hrd1 function; i.e., small molecules that restore the stability of the Sec61–2 protein in a Hrd1-dependent manner via a high throughput screen at elevated temperatures. Such a molecule would become a useful tool to further dissect the Hrd1 pathway under physiological conditions and potentially slow the degradation of one of the many metastable ERAD substrates in humans that, when prematurely, degraded, leads to catastrophic human diseases (Guerriero and Brodsky 2012; Needham et al. 2019). Indeed, drugs that facilitate the folding and slow the ERAD of the F508del allele of the cystic fibrosis transmembrane conductance regulator (CFTR) are clinically proven to radically reduce the symptoms associated with this lethal disease (Veit et al. 2020). We also propose that extragenic suppressor screens can now be performed that restore the ability of the L201P and L62R alleles to re-assemble the Hrd1 complex and support ERAD. This analysis might better define the dynamic control of Hrd1 during the cycle of substrate identification, retrotranslocation, ubiquitination, and release. Moreover, through the use of different ERAD substrates whose stabilizing will result in a screenable phenotype, we predict that a similar approach as that reported here can be used to isolate new mutations in Doa10, which like Hrd1 functions as a “retrotranslocase” (Schmidt et al. 2020).