Ribosomes are complex molecular machines that translate the sequence of bases in a messenger RNA (mRNA) transcript into a polypeptide that subsequently folds to form a protein. Each ribosome is composed of two major subunits: the small subunit reads the mRNA transcript, and the large subunit joins amino acids together to form the polypeptide. This process stops when the ribosome encounters a stop codon and releases the completed polypeptide.

It is critical that cells perform some form of quality control on the polypeptides as they are translated to prevent a build up of incomplete, incorrect or toxic proteins in cells. Problems can occur if a ribosome stalls while translating the mRNA transcript, or if the mRNA transcript is defective. For example, most mRNA transcripts contain a stop codon, but some do not, and these non-stop mRNA transcripts result in a non-stop polypeptide that remains tethered to the ribosome. It is important that the cell identifies and removes these faulty polypeptides so as to leave the ribosome free to translate other (non-faulty) mRNA transcripts. A regulatory protein called ubiquitin is responsible for marking and sending proteins that are faulty, or are no longer needed by the cell, to a molecular machine called the proteasome, where they are degraded by a process called proteolysis. In 2010 researchers identified Ltn1 as the enzyme that attaches ubiquitin to non-stop proteins in yeast.

Now, building on this work, Verma et al. identify additional proteins involved in this process. In particular, an ATPase enzyme called Cdc48 (known as p97 or VCP in human cells) and two co-factors—Ufd1 and Npl4—promote release of the ubiquitinated non-stop polypeptides from the ribosomes, thus committing the marked polypeptide to destruction by the proteasome. Verma et al. also show that the Cdc48-Ufd1-Npl4 complex is involved in other aspects of quality control of newly synthesized proteins within cells. Collectively these processes are known as ribosome-associated degradation.

Mutations of the gene that codes for human p97 can cause a number of diseases, including Paget's disease of the bone and frontotemporal dementia, so an improved understanding of ribosome-associated degradation could provide new insights into these diseases.

Maintaining protein homeostasis is key for the healthy lifespan of the organism (Koga et al., 2011) and cells have evolved sophisticated mechanisms to control the quality of newly synthesized proteins (Bukau et al., 2006; Smith et al., 2011). Dysfunction of these quality control (QC) pathways can tip the balance towards the accumulation of toxic proteins and manifestation of disease, most notably neurodegeneration (Powers et al., 2009; Hartl et al., 2011; Dennissen et al., 2012).

There are multiple pools of QC substrates in cells. A relatively well-characterized substrate pool arises from misfolding of nascent chains that have been properly translated. Depending upon the relative rate at which malfolded sequences that comprise degrons form and bind their Ub ligase receptors vs the rate at which translation is completed, ubiquitination and degradation of misfolding proteins may initiate on the ribosome. It has been shown, for instance, that the acetylated N-terminus recognized by the Ub ligase Doa10 (Hwang et al., 2010) is generated co-translationally (Gautschi et al., 2003; Polevoda et al., 2008), and that the Ub ligase Ubr1 can initiate degradation of an N-end rule substrate prior to completion of its translation (Turner and Varshavsky, 2000).

A second pool of QC substrates arises from failure to complete translation successfully. This can occur because the mRNA is defective or the ribosome stalls due to either irresolvable structures in the mRNA or interactions between the nascent peptide and the ribosome (Lu and Deutsch, 2008; Kramer et al., 2009; Ingolia et al., 2011). Defective mRNAs are recognized by a cytosolic mRNA surveillance machinery that mediates their destruction. There are at least three pathways: no-go decay (NGD) targets mRNAs with strong stalls in translation elongation leading to endonucleolytic cleavage of the mRNA, nonsense-mediated decay (NMD) targets messages containing premature termination codons (PTC), and non-stop decay (NSD) targets mRNAs that lack a stop codon. Within this broad scheme, there are many variations by which aberrant mRNAs are generated (Parker, 2012). Interestingly, recent data suggest an intimate coupling between QC of defective mRNAs and the nascent peptides they produce (Shoemaker and Green, 2012). For example, Upf1 is a positive effector of NMD that is also needed for proteasome-mediated degradation of the prematurely terminated products (Takahashi et al., 2008; Kuroha et al., 2009).

Quality-control of mRNAs that lack a stop codon and the proteins produced from them has been the focus of considerable recent attention. Translation of these ‘non-stop’ mRNAs yields a ribosome-tethered polypeptide that is targeted for ubiquitination and degradation (Ito-Harashima et al., 2007; Dimitrova et al., 2009; Bengtson and Joazeiro, 2010). Notably, the mRNA may be processed by multiple pathways (e.g., NSD and NGD) because absence of a stop codon can cause translation to stall either because the end of the message is encountered or polylysine encoded by the poly(A) tail interacts with the negatively charged exit tunnel of the ribosome. Non-stop mRNAs can arise naturally through premature termination or polyadenylation within ORFs, which are estimated to occur for up to 10% of all transcriptional events (Frischmeyer et al., 2002). Proteins encoded by these messages could potentially impose a considerable burden on QC pathways. Thus, disposal of proteins produced from non-stop messages is likely to be of general significance to understanding human diseases that result from defects in protein homeostasis.

Stalled ribosomes that trigger NGD/NSD are recognized by the Dom34–Hbs1 complex which is structurally related to the termination factors eRF1 and eRF3 but lacks the residues necessary for both stop codon recognition and hydrolysis of peptidyl-tRNA and instead binds to the empty A site in the stalled ribosome (Doma and Parker, 2006; Becker et al., 2011; Tsuboi et al., 2012). Dom34–Hbs1 works in concert with a member of the ABC family of ATPases, known as ABCE1 in mammalian cells (Pisareva et al., 2011) or Rli1 in budding yeast (Shoemaker et al., 2010; Shoemaker and Green, 2011), to disassemble stalled ribosomes for subsequent reuse. Additionally, the translational GTPase Ski7, which is closely related to Hbs1 (and thus to eRF3), has been proposed to recognize ribosomes stalled at the 3′ end of non-stop messages (van Hoof et al., 2002). However, no Dom34/eRF1-like molecule has been identified to date that can interact with Ski7 specifically so it remains unknown if it can promote recycling of ribosome subunits. Following dissociation of subunits, the mechanism by which the associated nascent peptide is extricated remains poorly understood. What has been shown so far is that the Ub ligase Ltn1 can bind the ribosome and ubiquitinate NSD substrates that contain a polylysine tract (Bengtson and Joazeiro, 2010). Ltn1 belongs to the RING domain family of E3 Ub ligases that ubiquitinate substrates targeted for degradation by the 26S proteasome (Deshaies and Joazeiro, 2009; Finley, 2009). However, the mechanism by which the nascent non-stop peptide is released from the ribosome and degraded remains unknown. Nevertheless, the underlying biochemistry is likely to be important for human health, because point mutations in mouse Ltn1 result in neurodegeneration and a null allele is lethal (Chu et al., 2009).

Budding yeast Cdc48 (the human protein is known as p97 or valosin-containing protein [VCP]) is a member of the AAA (ATPases associated with diverse cellular activities) family of proteins. Cdc48 has intrinsic Ub-binding affinity that is greatly enhanced by the binding of UBX domain proteins or the Ufd1–Npl4 heterodimer to its N-terminus and Ufd2 to its C-terminus (Ye, 2006; Meyer et al., 2012). Cdc48/p97 is widely believed to function as a ‘segregase’ that uses the free energy released upon ATP hydrolysis to remodel ubiquitinated substrate complexes, usually, but not always, as a prelude to their degradation by the 26S proteasome (Jentsch and Rumpf, 2007). In the work reported here, we show that Cdc48–Ufd1–Npl4 promotes the release of tRNA-linked ubiquitinated nascent chains as well as aberrant non-stop and prematurely terminated polypeptides from the ribosome. Our findings establish a new function for Cdc48/p97 in RAD.

In an effort to systematically elucidate cellular functions of Cdc48 and its adaptors, mutants were subjected to chemical profiling by determining growth in the presence of various drugs, including translation inhibitors. All mutants were insensitive to anisomycin, an inhibitor that blocks the peptidyl transferase reaction on ribosomes (not shown). However, low concentrations of hygromycin B and paromomycin inhibited growth of multiple cdc48 mutants and the adaptor mutants npl4-1, ufd1, ubx1Δ, and ubx2Δ (Figure 1—figure supplement 1A,B and data not shown). Hygromycin B and paromomycin belong to the aminoglycoside class of antibiotics that bind close to the decoding center and affect translational fidelity (Carter et al., 2000; Kim and Craig, 2005). Deletion of LTN1 leads to hygromycin-sensitivity and since Ltn1 is involved in QC of NSD pathway substrates (Bengtson and Joazeiro, 2010), we explored the possibility that Cdc48–Ufd1–Npl4 and one or more Ubx proteins also function in this pathway.

To determine if Cdc48 contributes to protein degradation at the ribosome in unperturbed, exponentially growing cells, we purified ribosomes using a rapid, one-step affinity method designed to isolate 60S subunits, 80S monosomes and polysomes (Oeffinger et al., 2007; Halbeisen and Gerber, 2009; Halbeisen et al., 2009). Coomassie blue staining of purified ribosomes confirmed that Cdc48 activity was not required for their assembly (Figure 1, Figure 1—figure supplement 2A, lane 3), a conclusion that was further validated by sucrose gradient fractionation of ribosomes from wildtype, cdc48-3, and ufd1-2 whole cell lysates (Figure 1—figure supplement 2B). However, the mutants do appear to have a slightly lower ratio of polysomes:80S monosomes. Immunoblotting for Ub revealed strong accumulation of Ub conjugates on ribosomes affinity-purified from cdc48-3 cells (Figure 1A). To ascertain if the epitope tag on the ribosome subunit was exacerbating the effect of the cdc48-3 mutation, we isolated ribosomes from untagged cells by pelleting through a sucrose cushion (Halbeisen and Gerber, 2009). Again, Ub conjugates were observed to accumulate on ribosomes in cdc48-3 and npl4-1 mutants (Figure 1B). Because the proteolytic defects of npl4-1 are manifested at the semi-permissive temperature of 30°C (Bays et al., 2001), we carried out the experiment in Figure 1B at 30°C. The data indicate that cdc48-3 does not have to be shifted to the fully non-permissive temperature to observe Ub conjugate accumulation at the ribosome. Additionally, conjugate accumulation on cdc48-3 and npl4-1 ribosomes exceeded that observed for ribosomes isolated from cells treated with the proteasome inhibitor MG132, a result that will be further discussed below.

Figure 1 with 2 supplements see all

Download asset Open asset

To determine whether the ATPase activity of Cdc48 was required to prevent accumulation of ribosome-associated Ub conjugates, ribosomes were pelleted from cdc48-3 mutant cells expressing plasmid-borne wildtype CDC48, or a mutant (Q2) deficient in ATPase activity (Ye et al., 2003). Whereas CDC48 complemented the cdc48-3 phenotype, high levels of Ub conjugates were detected on ribosomes isolated from cells expressing the cdc48-Q2 mutant (Figure 1C). Consistent with the idea that Cdc48 ATPase functions at the ribosome to mediate release of ubiquitinated nascent peptides, both Cdc48 and Ufd1 were found to cofractionate with affinity-purified ribosomes isolated under stringent conditions (Figure 1D).

We next wished to ascertain if the Ub conjugates were derived from nascent polypeptides that were linked to tRNA. Magnetic beads containing bound ribosomes from wildtype or cdc48-3 cells were treated with TEV protease, which cleaves between Rpl18B and the TAP tag. The TEV cleavage was carried out in the presence or absence of puromycin. Puromycin causes premature termination of translation by binding in the A-site of the ribosome and reacting with peptidyl-tRNA in the P-site, yielding a peptide-puromycin conjugate (Nathans, 1964). Thus, we expected that puromycin would induce the release of peptidyl-tRNA from the ribosome as a peptide-puromycin conjugate. The supernatants were then bound to UBA Sepharose resin to enrich for Ub conjugates. The rationale behind this step was that ribosome-associated Ub conjugates that remained bound to the ribosome would bind poorly to UBA beads because ribosomes do not efficiently penetrate the pores of a Sepharose resin, whereas Ub conjugates released from the ribosome by puromycin would be able to bind more efficiently to UBA beads. Puromycin treatment of ribosomes from cdc48-3 but not wildtype cells released ubiquitinated proteins that were enriched on the UBA resin (Figure 2A, top panel). A similar result was obtained with ribosomes isolated from npl4-1 and ufd1-1 mutants (Figure 2—figure supplement 1A,B). The conjugates observed in Figure 2A were indeed linked to puromycin because they cross-reacted with an antibody specific for puromycin (Figure 2A, bottom panel). The same was true for Ub conjugates released from npl4-1 and ufd1-2 ribosomes (Figure 2—figure supplement 1C). The fact that no Ub immunoreactivity was recovered from mutant ribosomes in the absence of puromycin treatment validated our assumption that the UBA resin largely excluded mega-complexes. This experiment also suggests that at least some of the Ub conjugates that accumulated on cdc48 and npl4 ribosomes (Figure 1A,B) must be ubiquitinated nascent proteins linked to tRNA.

Figure 2 with 1 supplement see all

Download asset Open asset

To further confirm that the Ub conjugates that accumulated on ribosomes in cdc48 cells were ubiquitinated nascent peptides linked to tRNA we did three additional experiments. First, ribosomes affinity-purified from cdc48-3 were treated with RNAse A or not before incubation with puromycin followed by binding to the UBA resin. Immunoblotting for Ub revealed that both the control and RNAse-treated ribosomes contained Ub conjugates that bound the UBA beads (Figure 2B, top panel, lanes 2 and 3). However, ribosomes pre-treated with RNAse no longer yielded puromycin-reactive conjugates (Figure 2B, bottom panel). By contrast, pre-treating lysate with RNAse before affinity purification of ribosomes eliminated recovery of both Ub-, and puromycin immunoreactivity (Figure 2B, lane 4, top and bottom panels).

The second experiment that we did to confirm tRNA linkage was to treat affinity-purified ribosomes with CTAB. CTAB is an ionic detergent that disrupts protein structure and precipitates peptidyl-tRNA molecules selectively (Hobden and Cundliffe, 1978; Mariappan et al., 2010). In parallel with results obtained previously using puromycin, Ub conjugates were precipitated from cdc48-3 ribosomes in an RNAse-sensitive manner (Figure 2C).

As a final test of the idea that the Ub conjugates that accumulated on cdc48-3 ribosomes were attached to nascent chains, wildtype and mutant cells were pulsed with radioactive methionine for 90 s and then chased with cold methionine. Ribosomes were affinity-purified from cells sampled at different times of chase and treated with puromycin to release tRNA-linked nascent chains. This material was then fractionated on UBA resin to enrich for discharged Ub conjugates. Strikingly, ribosomes from both cdc48-3 and ufd1-2 cells contained far higher levels of radiolabeled high MW Ub conjugates than wildtype ribosomes, and a substantial portion of this signal persisted on ribosomes for 10–20 min (Figure 2D), which is considerably longer than the time required to synthesize a protein. Importantly, the low signal on wildtype ribosomes was not due to inefficient incorporation, because the WT inputs for the UBA fractionation step actually contained slightly more total pulse-labeled species (Figure 2—figure supplement 1D). Taken together, the experiments described thus far indicate that Cdc48 and its cofactors Ufd1 and Npl4 prevent accumulation of ubiquitinated, tRNA-linked nascent chains on the ribosome. For simplicity, we hereafter refer to these species as tUNPs (tRNA-linked, ubiquitinated nascent polypeptides). Notably, other Cdc48 cofactors including Ubx2, Ubx4, Ubx6, and Ufd2 did not appear to contribute to clearance of tUNPs from the ribosome (Figure 2—figure supplement 1C). These cofactors were tested because they had scored positively in the hygromycin-sensitivity assay whereas ubx3Δ, ubx5Δ, and ubx7Δ were hygromycin-insensitive, and therefore not pursued further. Hygromycin-sensitive ubx1Δ/shp1Δ was also not pursued further, because this mutant is extremely sick.

The findings summarized above raised the question: what is the composition of the tUNPs? One possibility is a class of aberrant proteins arising from translation of mRNAs that lack stop codons (‘non-stop mRNA’). Translation through the poly(A) tail of these mRNAs can result in a ribosome that stalls either because it reaches the end of the message, or the polylysine encoded by the poly(A) tail interacts with the negatively charged exit tunnel of the ribosome (Dimitrova et al., 2009). Prior work has demonstrated that the Ub ligase Ltn1 targets proteins translated from mRNAs that lack a stop codon or that contain polylysine stretches (Bengtson and Joazeiro, 2010). We therefore employed the puromycin release/UBA bead capture protocol to test whether deletion of LTN1 diminished the accumulation of tUNPs on ribosomes in cdc48-3 cells. We also tested the impact of ubr1Δ, because of prior reports that Ubr1 promotes co-translational ubiquitination (Turner and Varshavsky, 2000), and because ubr1Δ mutants, like ltn1Δ, are hygromycin-sensitive (not shown, but see below). The data in Figure 3A demonstrate that additional loss of LTN1 or UBR1 resulted in a marked decrease in the level of tUNPs on cdc48-3 ribosomes. This result encouraged us to evaluate the steady state levels of the Gfp-FLAG-His3Non-Stop (GFH^NS) fusion protein shown earlier to be an Ltn1 substrate at the ribosome that is degraded by the proteasome (Bengtson and Joazeiro, 2010). Amongst the Cdc48 pathway mutants examined, GFH^NS accumulated maximally in ufd1 mutants (similar to levels observed for ltn1Δ), and also exhibited significant accumulation in npl4-1 and cdc48-3 (Figure 3B). Cycloheximide chase analysis in ufd1-2 cells indicated that accumulation resulted from stabilization of the non-stop reporter (Figure 3C). As observed in earlier studies (Bengtson and Joazeiro, 2010), the same reporter containing a stop codon was not degraded and accumulated at a much higher level than the NS reporter (Figure 3C; amounts loaded were adjusted to yield signals of equal intensity at 0 min).

Figure 3 with 1 supplement see all

Download asset Open asset

If the Cdc48–Ufd1–Npl4 complex functions downstream of Ltn1 to mediate clearance of tUNPs from the ribosome, as suggested by the data in Figure 3A, we anticipated that ubiquitinated species of the non-stop reporter should accumulate upon inactivation of Cdc48 complex. To address this question, we prepared SDS lysates of wildtype and ufd1 mutant cells expressing GFH^NS, diluted the lysates with Triton X-100 to sequester the SDS in micelles, and immunoprecipitated with anti-FLAG. Immunoblotting for Ub demonstrated that GFH^NS accumulated in an ubiquitinated form in ufd1-1 and ufd1-2 but not in ltn1Δ mutants (Figure 3D).

The above data indicate that the non-stop, but not the stop codon-containing reporter was targeted for degradation by the combined actions of Ltn1 and Cdc48–Ufd1–Npl4. The role of the Cdc48 complex in reporter degradation appeared to be direct, because both Cdc48 and Ufd1 were co-immunoprecipitated with GFH^NS, but not GFH^Stop, even though GFH^Stop was more abundant (Figure 3E).

GFH^NS accumulates on 80S ribosomes in ltn1Δ mutants (Bengtson and Joazeiro, 2010). To evaluate if any of these molecules are linked to tRNA, we sedimented ribosomes rapidly through sucrose cushions and treated the pellet with RNAse A (or not). Strikingly, immunoblotting revealed an RNAse-sensitive species on ribosomes from ltn1Δ, as well as those from cdc48-3 and ufd1-2 mutants, but not from wildtype cells (Figure 3—figure supplement 1). The fraction of GFH^NS linked to tRNA was low. It is not clear whether this was due to hydrolysis of the tRNA linkage in vivo or in vitro.

For our subsequent experiments, we decided to focus on Protein A-NS (PrA^NS), a non-stop reporter utilized in earlier studies (Wilson et al., 2007; Bengtson and Joazeiro, 2010). Unlike GFH^NS, the RNAse-sensitive, tRNA-linked form of PrA^NS accumulated more efficiently and could be detected even in total cell extracts. We first examined the steady-state levels of PrA^NS in several mutants. Interestingly, levels of tRNA-linked PrA^NS in cdc48-3 and ltn1Δ cell lysates were equivalent to those detected in dom34Δ and ski7Δ, bona fide components of the NSD pathway (Figure 4A; equivalent loading confirmed by staining with Ponceau S and immunoblotting for tubulin, Figure 4—figure supplement 1A).

Figure 4 with 1 supplement see all

Download asset Open asset

To more precisely determine which subset of ribosomes contained tRNA-linked PrA^NS, we performed sucrose gradient fractionation. Shallow gradients (10–30%) were set up, enabling good resolution of the subunits, monosomes, and polysomes (Figure 4—figure supplement 1B). Immunoblotting for PrA^NS revealed that the RNAse-sensitive species peaked in the 80S fractions isolated from ltn1Δ mutants but was found in both the 60S and 80S-containing fractions from cdc48-3 cells (Figure 4B). Additionally, modified forms of slower mobility than the tRNA-linked species were detected in 60S ribosomes from cdc48-3 but not ltn1Δ, suggesting that these high MW species are Ltn1-dependent Ub conjugates. Only a small fraction of the PrA^NS detected on ribosomes migrated as Ub-conjugated species. This is consistent with what is typically observed for most Ub-dependent substrates upon inhibition of their degradation, and presumably reflects robust cytosolic deubiquitination activity. In addition, overproduction of PrA^NS from the GAL promoter may overwhelm Ltn1, which is inabundant (Bengtson and Joazeiro, 2010). Besides the modified forms of PrA^NS, unmodified protein was observed throughout the gradient and peaked in the ribosomal fractions. We do not know whether this latter form arose from hydrolysis of the peptidyl-tRNA linkage in cells or in vitro.

Our data thus far indicate that in cdc48, npl4, and ufd1 mutants, nascent proteins were targeted for degradation by Ltn1- and Ubr1-dependent ubiquitination but became stuck on 60S and 80S ribosomes as peptidyl-tRNAs, implying that Cdc48 activity promotes release of stalled, ubiquitinated peptides, which are then degraded by the proteasome. This hypothesis predicts that ribosomes isolated from cells in which proteasome activity is compromised should accumulate less tUNPs than cdc48 mutants. To test this prediction, we used the sucrose cushion method to isolate ribosomes from cdc48-3 and npl4-1 mutants, as well as pdr5Δ cells treated with or without the proteasome inhibitor MG132 (pdr5Δ enables cellular accumulation of this compound). The ribosome pellet was then treated with puromycin and incubated with UBA resin to enrich for tUNPs. Ribosomes isolated from cells with inhibited proteasomes contained far less tUNPs than ribosomes from cdc48-3 and npl4-1 cells (Figure 5A; the input lysates and ribosome pellets for this experiment are shown in Figure 1B). This result is in striking contrast to what we previously reported for proteasomes isolated from cdc48-3 and MG132-treated cells—both contained very high levels of accumulated Ub conjugates (Verma et al., 2011). A similar result was obtained using the cim3-1 mutant instead of MG132 treatment (Figure 5—figure supplement 1A; cim3-1 is a ts allele of the gene that encodes the proteasome subunit Rpt6).

Figure 5 with 1 supplement see all

Download asset Open asset

We next sought to evaluate the effect of proteasome deficiency on accumulation of a specific peptidyl-tRNA on the ribosome. To evaluate the subcellular distribution of the PrA^NS species that accumulate upon inhibition of Cdc48 or proteasome function, we sedimented extracts through sucrose cushions to generate ribosome pellets and post-ribosomal supernatants from wildtype, cdc48-3 and cim3-1 cells (Figure 5B). Notably, whereas most of the PrA^NS fractionated in the ribosome pellet in cdc48-3 cells (a substantial portion of which remained linked to tRNA), the opposite was true in cim3-1 cells, where the bulk of the PrA^NS was recovered in the supernatant fraction (Figure 5B). Comparison of extracts made by lysing cells directly in SDS (Figure 5—figure supplement 1B) or under native conditions (Figure 5B) revealed that the proportion of PrA^NS that accumulated as full length or truncated species in vivo and its susceptibility to proteolytic clipping in native cell extract differed for cdc48-3 and cim3-1 mutants. We do not know the basis for these differences.

Our observation that Ubr1 contributed to ubiquitination of tUNPs (Figure 3A) led us to wonder about the diversity of the QC mechanisms that underlie their formation. To determine if the role of Cdc48 is limited to the non-stop pathway or whether it plays a more global role in degradation of proteins encoded by aberrant messages, we also assessed the stability of endogenous phosphogluconate dehydrogenase truncated by a premature termination codon (Gnd1^PTC) in the middle of a folded domain at residue 368. Over-produced Gnd1^PTC was shown to be a substrate of the Ubr1 cytosolic quality control pathway in a prior study (Heck et al., 2010). The data in Figure 6A confirmed stabilization of Gnd1^PTC in ubr1Δ. Notably, cdc48-3 and ufd1-2 mutations also stabilized Gnd1^PTC, but ltn1Δ was without effect (equivalent loading confirmed in Figure 6—figure supplement 1). Gnd1^PTC was also stabilized in upf1Δ. Upf1 is a RING domain Ub ligase (Takahashi et al., 2008) that plays a crucial role in NMD and stimulates the degradation of aberrant PTC translation products (Kuroha et al., 2009). The dependence of Gnd1^PTC degradation on Cdc48 and Upf1 suggested that tRNA-linked Gnd1^PTC might become stuck on ribosomes, as observed for the non-stop reporters. To test this possibility we used the sucrose cushion method to isolate ribosomes from various mutants, and evaluated the ribosomal pellets for their content of Gnd1^PTC. Analysis of total cell lysates revealed strong accumulation of Gnd1^PTC in cdc48-3, ubr1Δ, and upf1Δ cells (Figure 6B). Fractionation of lysates prepared in an identical manner revealed an RNAse-sensitive species on ribosomes but not in post-ribosomal supernatants isolated from cdc48-3 mutants (Figure 6C) as well as ufd1-2, ubr1Δ and upf1Δ mutants (data not shown). In contrast, little tRNA-linked substrate was found in dom34Δ mutants (Figure 6C). Strikingly, the tRNA-linked species was not detected on ribosomes from ltn1Δ (data not shown).

Figure 6 with 1 supplement see all

Download asset Open asset

Whereas QC of defective mRNAs and recycling of ribosomes stalled on these templates has received considerable attention in prior studies, much less is known about what happens to the nascent polypeptides, some of which accumulate as peptidyl-tRNA. Some proteins encoded by non-stop mRNAs, including the reporter proteins used here (GFH^NS and PrA^NS), are degraded by the ubiquitin-proteasome system (UPS) (Wilson et al., 2007; Bengtson and Joazeiro, 2010). The Ub ligase Ltn1 binds 60S ribosomes and promotes ubiquitination of stalled GFH^NS (Bengtson and Joazeiro, 2010). Our data suggest that Cdc48–Ufd1–Npl4 acts downstream of Ltn1 to promote release of the non-stop peptide from the ribosome, so that it can be degraded by the proteasome. How Ltn1 and the Cdc48 complex does this remains unknown. Indeed, this remains unknown for any NSD or NGD pathway substrate; the Dom34-Hbs1 complex lacks the residues necessary for hydrolysis of the peptidyl-tRNA bond, and the two budding yeast peptidyl-tRNA hydrolases, Pth1 and Pth2, are non-essential mitochondrial proteins whose transmembrane configurations are currently unknown (Rosas-Sandoval et al., 2002). An interesting question for the future will be to determine whether Ltn1 and Cdc48–Ufd1–Npl4 act together with, or in a redundant pathway parallel to Ski7 or Dom34–Hbs1-Rli1.

We observed direct physical association of Cdc48 and Ufd1 with both affinity-purified ribosomes and the NSD pathway substrate GFH^NS. This interaction was highly specific, because Cdc48 and Ufd1 did not bind the same substrate when it was translated from a message that includes a stop codon. While this manuscript was in revision, it was reported that Cdc48, Ufd1, and Npl4 are components of the newly identified Ribosome Quality Control (RQC) complex, which also contains Ltn1, Rqc1 and Tae2 and associates with the 60S ribosomal subunit. The latter two proteins, like Ltn1 and Cdc48–Ufd1–Npl4, are highly conserved and have mammalian homologs (Brandman et al., 2012). Both Ltn1 and Rqc1 help link Cdc48–Ufd1–Npl4 to RQC. Rqc1 associates with and is required for the degradation of an Ltn1 substrate that contains an internal polyarginine sequence, but a functional role for Cdc48–Ufd1–Npl4 in RQC was not established by Brandman et al. (2012).

A key question that emerges from this work is, what is the signal that differentiates ribosomes engaged with NSD or other aberrant substrates from normally translating ribosomes? Ltn1 (Bengtson and Joazeiro, 2010) and RQC (Brandman et al., 2012) are reported to be predominantly bound to 60S ribosomal subunits, suggesting that they may act downstream of the initial recognition and splitting of a stalled ribosome. However, in ltn1Δ cells, GFH^NS accumulates on 80S monosomes (Bengtson and Joazeiro, 2010). Likewise, we report here that tRNA-linked PrA^NS appeared to be concentrated in monosomes in ltn1Δ. By contrast, tRNA-linked PrA^NS was spread more broadly across 60S and 80S fractions in extracts from cdc48-3 cells. The different fractionation behavior of tRNA-linked PrA^NS in ltn1Δ and cdc48 mutants may reflect leakiness of the cdc48 mutation (the experiment was done at the semi-permissive temperature of 30°C) or Cdc48-independent ribosome splitting. Detailed resolution of the exact sequence of events will require reconstitution of the overall process with defined components.

In addition to observing accumulation of engineered NSD pathway reporter proteins in Cdc48 pathway mutants, we also observed extensive accumulation of endogenous, tRNA-linked ubiquitinated nascent polypeptides (tUNPs) on ribosomes in these cells. These tUNPs could potentially arise from multiple sources including misfolded domains (Turner and Varshavsky, 2000), mRNAs whose transcription was prematurely terminated, mRNAs or nascent peptides containing sequences that cause ribosome pausing (Ingolia et al., 2011) or mRNAs that encode premature termination codons (e.g., Figure 6C). Thus it is perhaps not surprising that at least two Ub ligases (Ltn1 and Ubr1) contributed to their accumulation. In any event, we believe it is unlikely that there is a single pathway or mechanism for disposing of tUNPs. Even in the seemingly simple case of a ribosome that translates into the poly(A) tail of an mRNA that is prematurely polyadenylated within the coding sequence, the 3′-most ribosome may be stalled because it reached the end of the message, or because polylysine encoded by the poly(A) tail bound the exit channel of the ribosome (Ito-Harashima et al., 2007). Because stalls in translation elongation lead to endonucleolytic cleavage of the mRNA upstream of the 3′-most ribosome (Tsuboi et al., 2012), each trailing ribosome will present a unique substrate to the UPS, and may be processed by a different mechanism.

The intimacy with which protein biosynthesis is coupled to protein degradation is highlighted by the identification of the translasome, a supercomplex defined by mass spectrometric analysis of the translation initiation factor eIF3 and found to contain both ribosome and proteasome subunits as well as chaperones involved in protein QC (Sha et al., 2009). A tight coupling of these processes may be particularly important for neurons, which are post-mitotic and need to remain robust for the life of the organism (Segref and Hoppe, 2009). Indeed, homozygous lister mice deficient in Listerin (Ltn1) function exhibit severe neurological dysfunction (Chu et al., 2009), whereas mutations in human Ubr1 result in the Johanson–Blizzard syndrome characterized by, amongst other phenotypes, frequent mental retardation (Hwang et al., 2011). Mutations in p97 contribute to 1–2% of familial ALS (Johnson et al., 2010) and are the root cause of a syndrome, IBMPFD, that includes frontotemporal dementia (Watts et al., 2004). Defects in metabolism of protein aggregates (Manno et al., 2010), endosomal trafficking (Ritz et al., 2011) and autophagy (Ju et al., 2009; Tresse et al., 2010) have been previously suggested to contribute to the pathophysiology that results from mutation of p97. As the molecular mechanisms underlying these diseases are unraveled, it will be interesting to see if ribosome-associated degradation contributes to their etiology.

1. 1. Bays NW
   2. Wilhovsky SK
   3. Goradia A
   4. Hodgkiss-Harlow K
   5. Hampton RY

   (2001)

   HRD4/NPL4 is required for the proteasomal processing of ubiquitinated ER proteins

   Mol Biol Cell 12:4114–4128.

   • Google Scholar
2. 1. Becker T
   2. Armache JP
   3. Jarasch A
   4. Anger AM
   5. Villa E
   6. Sieber H

   et al. (2011)

   Structure of the no-go mRNA decay complex Dom34-Hbs1 bound to a stalled 80S ribosome

   Nat Struct Mol Biol 18:715–720.

   • Google Scholar
3. 1. Bengtson MH
   2. Joazeiro CA

   (2010)

   Role of a ribosome-associated E3 ubiquitin ligase in protein quality control

   Nature 467:470–473.

   • Google Scholar
4. 1. Brandman O
   2. Stewart-Ornstein J
   3. Wong D
   4. Larson A
   5. Williams CC
   6. Li GW

   et al. (2012)

   A ribosome-bound quality control complex triggers degradation of nascent peptides and signals translation stress

   Cell 151:1042–1054.

   • Google Scholar
5. 1. Bukau B
   2. Weissman J
   3. Horwich A

   (2006)

   Molecular chaperones and protein quality control

   Cell 125:443–451.

   • Google Scholar
6. 1. Carter AP
   2. Clemons WM
   3. Brodersen DE
   4. Morgan-Warren RJ
   5. Wimberly BT
   6. Ramakrishnan V

   (2000)

   Functional insights from the structure of the 30S ribosomal subunit and its interactions with antibiotics

   Nature 407:340–348.

   • Google Scholar
7. 1. Chu J
   2. Hong NA
   3. Masuda CA
   4. Jenkins BV
   5. Nelms KA
   6. Goodnow CC

   et al. (2009)

   A mouse forward genetics screen identifies LISTERIN as an E3 ubiquitin ligase involved in neurodegeneration

   Proc Natl Acad Sci USA 106:2097–2103.

   • Google Scholar
8. 1. Dennissen FJ
   2. Kholod N
   3. van Leeuwen FW

   (2012)

   The ubiquitin proteasome system in neurodegenerative diseases: culprit, accomplice or victim?

   Prog Neurobiol 96:190–207.

   • Google Scholar
9. 1. Deshaies RJ
   2. Joazeiro CA

   (2009)

   RING domain E3 ubiquitin ligases

   Annu Rev Biochem 78:399–434.

   • Google Scholar
10. 1. Dimitrova LN
    2. Kuroha K
    3. Tatematsu T
    4. Inada T

    (2009)

    Nascent peptide-dependent translation arrest leads to Not4p-mediated protein degradation by the proteasome

    J Biol Chem 284:10343–10352.

    • Google Scholar
11. 1. Doma MK
    2. Parker R

    (2006)

    Endonucleolytic cleavage of eukaryotic mRNAs with stalls in translation elongation

    Nature 440:561–564.

    • Google Scholar
12. 1. Finley D

    (2009)

    Recognition and processing of ubiquitin-protein conjugates by the proteasome

    Annu Rev Biochem 78:477–513.

    • Google Scholar
13. 1. Frischmeyer PA
    2. van Hoof A
    3. O'Donnell K
    4. Guerrerio AL
    5. Parker R
    6. Dietz HC

    (2002)

    An mRNA surveillance mechanism that eliminates transcripts lacking termination codons

    Science 295:2258–2261.

    • Google Scholar
14. 1. Gautschi M
    2. Just S
    3. Mun A
    4. Ross S
    5. Rucknagel P
    6. Dubaquie Y

    et al. (2003)

    The yeast N(alpha)-acetyltransferase NatA is quantitatively anchored to the ribosome and interacts with nascent polypeptides

    Mol Cell Biol 23:7403–7414.

    • Google Scholar
15. 1. Halbeisen RE
    2. Gerber AP

    (2009)

    Stress-dependent coordination of transcriptome and translatome in yeast

    PLoS Biol 7:e1000105.

    • Google Scholar
16. 1. Halbeisen RE
    2. Scherrer T
    3. Gerber AP

    (2009)

    Affinity purification of ribosomes to access the translatome

    Methods 48:306–310.

    • Google Scholar
17. 1. Hartl FU
    2. Bracher A
    3. Hayer-Hartl M

    (2011)

    Molecular chaperones in protein folding and proteostasis

    Nature 475:324–332.

    • Google Scholar
18. 1. Heck JW
    2. Cheung SK
    3. Hampton RY

    (2010)

    Cytoplasmic protein quality control degradation mediated by parallel actions of the E3 ubiquitin ligases Ubr1 and San1

    Proc Natl Acad Sci USA 107:1106–1111.

    • Google Scholar
19. 1. Hobden AN
    2. Cundliffe E

    (1978)

    Mode of action of alpha-sarcin and a novel assay of puromycin reaction

    Biochem J 170:57–61.

    • Google Scholar
20. 1. Hwang CS
    2. Shemorry A
    3. Varshavsky A

    (2010)

    N-terminal acetylation of cellular proteins creates specific degradation signals

    Science 327:973–977.

    • Google Scholar
21. 1. Hwang CS
    2. Sukalo M
    3. Batygin O
    4. Addor MC
    5. Brunner H
    6. Aytes AP

    et al. (2011)

    Ubiquitin ligases of the N-end rule pathway: assessment of mutations in UBR1 that cause the Johanson-Blizzard syndrome

    PLoS One 6:e24925.

    • Google Scholar
22. 1. Ingolia NT
    2. Lareau LF
    3. Weissman JS

    (2011)

    Ribosome profiling of mouse embryonic stem cells reveals the complexity and dynamics of mammalian proteomes

    Cell 147:789–802.

    • Google Scholar
23. 1. Ito-Harashima S
    2. Kuroha K
    3. Tatematsu T
    4. Inada T

    (2007)

    Translation of the poly(A) tail plays crucial roles in nonstop mRNA surveillance via translation repression and protein destabilization by proteasome in yeast

    Genes Dev 21:519–524.

    • Google Scholar
24. 1. Jentsch S
    2. Rumpf S

    (2007)

    Cdc48 (p97): a “molecular gearbox” in the ubiquitin pathway?

    Trends Biochem Sci 32:6–11.

    • Google Scholar
25. 1. Johnson JO
    2. Mandrioli J
    3. Benatar M
    4. Abramzon Y
    5. Van Deerlin VM
    6. Trojanowski JQ

    et al. (2010)

    Exome sequencing reveals VCP mutations as a cause of familial ALS

    Neuron 68:857–864.

    • Google Scholar
26. 1. Ju JS
    2. Fuentealba RA
    3. Miller SE
    4. Jackson E
    5. Piwnica-Worms D
    6. Baloh RH

    et al. (2009)

    Valosin-containing protein (VCP) is required for autophagy and is disrupted in VCP disease

    J Cell Biol 187:875–888.

    • Google Scholar
27. 1. Kim SY
    2. Craig EA

    (2005)

    Broad sensitivity of Saccharomyces cerevisiae lacking ribosome-associated chaperone ssb or zuo1 to cations, including aminoglycosides

    Eukaryot Cell 4:82–89.

    • Google Scholar
28. 1. Koga H
    2. Kaushik S
    3. Cuervo AM

    (2011)

    Protein homeostasis and aging: the importance of exquisite quality control

    Ageing Res Rev 10:205–215.

    • Google Scholar
29. 1. Kramer G
    2. Boehringer D
    3. Ban N
    4. Bukau B

    (2009)

    The ribosome as a platform for co-translational processing, folding and targeting of newly synthesized proteins

    Nat Struct Mol Biol 16:589–597.

    • Google Scholar
30. 1. Kuroha K
    2. Tatematsu T
    3. Inada T

    (2009)

    Upf1 stimulates degradation of the product derived from aberrant messenger RNA containing a specific nonsense mutation by the proteasome

    EMBO Rep 10:1265–1271.

    • Google Scholar
31. 1. Lu J
    2. Deutsch C

    (2008)

    Electrostatics in the ribosomal tunnel modulate chain elongation rates

    J Mol Biol 384:73–86.

    • Google Scholar
32. 1. Manno A
    2. Noguchi M
    3. Fukushi J
    4. Motohashi Y
    5. Kakizuka A

    (2010)

    Enhanced ATPase activities as a primary defect of mutant valosin-containing proteins that cause inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia

    Genes Cells 15:911–922.

    • Google Scholar
33. 1. Mariappan M
    2. Li XZ
    3. Stefanovic S
    4. Sharma A
    5. Mateja A
    6. Keenan RJ

    et al. (2010)

    A ribosome-associating factor chaperones tail-anchored membrane proteins

    Nature 466:1120–1138.

    • Google Scholar
34. 1. Meyer H
    2. Bug M
    3. Bremer S

    (2012)

    Emerging functions of the VCP/p97 AAA-ATPase in the ubiquitin system

    Nat Cell Biol 14:117–123.

    • Google Scholar
35. 1. Nathans D

    (1964)

    Puromycin inhibition of protein Synthesis: incorporation of puromycin into peptide chains

    Proc Natl Acad Sci USA 51:585–592.

    • Google Scholar
36. 1. Oeffinger M
    2. Wei KE
    3. Rogers R
    4. DeGrasse JA
    5. Chait BT
    6. Aitchison JD

    et al. (2007)

    Comprehensive analysis of diverse ribonucleoprotein complexes

    Nat Methods 4:951–956.

    • Google Scholar
37. 1. Parker R

    (2012)

    RNA degradation in Saccharomyces cerevisae

    Genetics 191:671–702.

    • Google Scholar
38. 1. Pisareva VP
    2. Skabkin MA
    3. Hellen CU
    4. Pestova TV
    5. Pisarev AV

    (2011)

    Dissociation by Pelota, Hbs1 and ABCE1 of mammalian vacant 80S ribosomes and stalled elongation complexes

    EMBO J 30:1804–1817.

    • Google Scholar
39. 1. Polevoda B
    2. Brown S
    3. Cardillo TS
    4. Rigby S
    5. Sherman F

    (2008)

    Yeast N(alpha)-terminal acetyltransferases are associated with ribosomes

    J Cell Biochem 103:492–508.

    • Google Scholar
40. 1. Powers ET
    2. Morimoto RI
    3. Dillin A
    4. Kelly JW
    5. Balch WE

    (2009)

    Biological and chemical approaches to diseases of proteostasis deficiency

    Annu Rev Biochem 78:959–991.

    • Google Scholar
41. 1. Ritz D
    2. Vuk M
    3. Kirchner P
    4. Bug M
    5. Schutz S
    6. Hayer A

    et al. (2011)

    Endolysosomal sorting of ubiquitylated caveolin-1 is regulated by VCP and UBXD1 and impaired by VCP disease mutations

    Nat Cell Biol 13:1116–1123.

    • Google Scholar
42. 1. Rosas-Sandoval G
    2. Ambrogelly A
    3. Rinehart J
    4. Wei D
    5. Cruz-Vera LR
    6. Graham DE

    et al. (2002)

    Orthologs of a novel archaeal and of the bacterial peptidyl-tRNA hydrolase are nonessential in yeast

    Proc Natl Acad Sci USA 99:16707–16712.

    • Google Scholar
43. 1. Segref A
    2. Hoppe T

    (2009)

    Think locally: control of ubiquitin-dependent protein degradation in neurons

    EMBO Rep 10:44–50.

    • Google Scholar
44. 1. Sha Z
    2. Brill LM
    3. Cabrera R
    4. Kleifeld O
    5. Scheliga JS
    6. Glickman MH

    et al. (2009)

    The eIF3 interactome reveals the translasome, a supercomplex linking protein synthesis and degradation machineries

    Mol Cell 36:141–152.

    • Google Scholar
45. 1. Shoemaker CJ
    2. Eyler DE
    3. Green R

    (2010)

    Dom34:Hbs1 promotes subunit dissociation and peptidyl-tRNA drop-off to initiate no-go decay

    Science 330:369–372.

    • Google Scholar
46. 1. Shoemaker CJ
    2. Green R

    (2011)

    Kinetic analysis reveals the ordered coupling of translation termination and ribosome recycling in yeast

    Proc Natl Acad Sci USA 108:E1392–E1398.

    • Google Scholar
47. 1. Shoemaker CJ
    2. Green R

    (2012)

    Translation drives mRNA quality control

    Nat Struct Mol Biol 19:594–601.

    • Google Scholar
48. 1. Smith MH
    2. Ploegh HL
    3. Weissman JS

    (2011)

    Road to ruin: targeting proteins for degradation in the endoplasmic reticulum

    Science 334:1086–1090.

    • Google Scholar
49. 1. Takahashi S
    2. Araki Y
    3. Ohya Y
    4. Sakuno T
    5. Hoshino S
    6. Kontani K

    et al. (2008)

    Upf1 potentially serves as a RING-related E3 ubiquitin ligase via its association with Upf3 in yeast

    RNA 14:1950–1958.

    • Google Scholar
50. 1. Tresse E
    2. Salomons FA
    3. Vesa J
    4. Bott LC
    5. Kimonis V
    6. Yao TP

    et al. (2010)

    VCP/p97 is essential for maturation of ubiquitin-containing autophagosomes and this function is impaired by mutations that cause IBMPFD

    Autophagy 6:217–227.

    • Google Scholar
51. 1. Tsuboi T
    2. Kuroha K
    3. Kudo K
    4. Makino S
    5. Inoue E
    6. Kashima I

    et al. (2012)

    Dom34:hbs1 plays a general role in quality-control systems by dissociation of a stalled ribosome at the 3′ end of aberrant mRNA

    Mol Cell 46:518–529.

    • Google Scholar
52. 1. Turner GC
    2. Varshavsky A

    (2000)

    Detecting and measuring cotranslational protein degradation in vivo

    Science 289:2117–2120.

    • Google Scholar
53. 1. van Hoof A
    2. Frischmeyer PA
    3. Dietz HC
    4. Parker R

    (2002)

    Exosome-mediated recognition and degradation of mRNAs lacking a termination codon

    Science 295:2262–2264.

    • Google Scholar
54. 1. Verma R
    2. Oania R
    3. Fang R
    4. Smith GT
    5. Deshaies RJ

    (2011)

    Cdc48/p97 mediates UV-dependent turnover of RNA Pol II

    Mol Cell 41:82–92.

    • Google Scholar
55. 1. Watts GD
    2. Wymer J
    3. Kovach MJ
    4. Mehta SG
    5. Mumm S
    6. Darvish D

    et al. (2004)

    Inclusion body myopathy associated with Paget disease of bone and frontotemporal dementia is caused by mutant valosin-containing protein

    Nat Genet 36:377–381.

    • Google Scholar
56. 1. Wilson MA
    2. Meaux S
    3. van Hoof A

    (2007)

    A genomic screen in yeast reveals novel aspects of nonstop mRNA metabolism

    Genetics 177:773–784.

    • Google Scholar
57. 1. Ye Y

    (2006)

    Diverse functions with a common regulator: ubiquitin takes command of an AAA ATPase

    J Struct Biol 156:29–40.

    • Google Scholar
58. 1. Ye Y
    2. Meyer HH
    3. Rapoport TA

    (2003)

    Function of the p97-Ufd1-Npl4 complex in retrotranslocation from the ER to the cytosol: dual recognition of nonubiquitinated polypeptide segments and polyubiquitin chains

    J Cell Biol 162:71–84.

    • Google Scholar

Author details

1. Rati Verma

   1. Division of Biology, California Institute of Technology, Pasadena, United States
   2. Howard Hughes Medical Institute, California Institute of Technology, Pasadena, United States

   Contribution

   RV, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   For correspondence

   verma@caltech.edu

   Competing interests

   No competing interests declared.

2. Robert S Oania

   1. Division of Biology, California Institute of Technology, Pasadena, United States
   2. Howard Hughes Medical Institute, California Institute of Technology, Pasadena, United States

   Contribution

   RSO, Acquisition of data

   Competing interests

   No competing interests declared.

3. Natalie J Kolawa

   Division of Biology, California Institute of Technology, Pasadena, United States

   Contribution

   NJK, Acquisition of data

   Competing interests

   No competing interests declared.

4. Raymond J Deshaies

   1. Division of Biology, California Institute of Technology, Pasadena, United States
   2. Howard Hughes Medical Institute, California Institute of Technology, Pasadena, United States

   Contribution

   RJD, Conception and design, Analysis and interpretation of data, Drafting or revising the article

   For correspondence

   deshaies@caltech.edu

   Competing interests

   RJD: Reviewing Editor, eLife; Founder, shareholder, and consultant of Cleave Biosciences, which is developing drugs that target enzymes involved in protein homeostasis.

• 3,354

  views

• 615

  downloads

• 201

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.