1. Biochemistry and Chemical Biology
Download icon

The GTPase Nog1 co-ordinates the assembly, maturation and quality control of distant ribosomal functional centers

Research Article
  • Cited 4
  • Views 1,280
  • Annotations
Cite this article as: eLife 2020;9:e52474 doi: 10.7554/eLife.52474

Abstract

Eukaryotic ribosome precursors acquire translation competence in the cytoplasm through stepwise release of bound assembly factors, and proofreading of their functional centers. In case of the pre-60S, these steps include removal of placeholders Rlp24, Arx1 and Mrt4 that prevent premature loading of the ribosomal protein eL24, the protein-folding machinery at the polypeptide exit tunnel (PET), and the ribosomal stalk, respectively. Here, we reveal that sequential ATPase and GTPase activities license release factors Rei1 and Yvh1 to trigger Arx1 and Mrt4 removal. Drg1-ATPase activity removes Rlp24 from the GTPase Nog1 on the pre-60S; consequently, the C-terminal tail of Nog1 is extracted from the PET. These events enable Rei1 to probe PET integrity and catalyze Arx1 release. Concomitantly, Nog1 eviction from the pre-60S permits peptidyl transferase center maturation, and allows Yvh1 to mediate Mrt4 release for stalk assembly. Thus, Nog1 co-ordinates the assembly, maturation and quality control of distant functional centers during ribosome formation.

Introduction

Error-free translation of the genetic code by the ribosome is critical for proteome homeostasis and cellular function. This essential task necessitates that only correctly assembled ribosomal subunits are committed for translation. Eukaryotic ribosome assembly is initiated by production of pre-rRNA in the nucleolus that is driven by RNA polymerase I (Peña et al., 2017; Kressler et al., 2017). The emerging pre-rRNA associates with small subunit (40S)-specific r-proteins, snoRNAs, U3snoRNP, numerous U3 proteins (UTPs) and assembly factors to form a pre-40S. Endonucleolytic cleavage releases the earliest pre-40S, permitting the remaining growing pre-rRNA to recruit large subunit (60S)-specific r-proteins and assembly factors to form a pre-60S. On their way through the nucleoplasm, pre-ribosomal particles interact with >200 assembly factors, including >40 energy-consuming AAA-ATPases, ABC-ATPases, GTPases and ATP-dependent RNA helicases. Elucidating their order of action and the mechanisms that co-ordinate their activities during pre-ribosome maturation is an important challenge.

Export-competent pre-ribosomes are transported through nuclear pore complexes (NPCs) into the cytoplasm, where they undergo maturation and proofreading before acquiring translation competence (Nerurkar et al., 2015). These steps include the release of assembly factors, transport receptors, pre-rRNA processing steps and the incorporation of remaining r-proteins that are critical for ribosome function. Cytoplasmic maturation of an exported pre-60S is initiated by the AAA-ATPase Drg1, which directly binds to and evicts the placeholder ribosomal-like protein Rlp24 from the pre-60S (Pertschy et al., 2007; Kappel et al., 2012; Kressler et al., 2012). This step is a prerequisite for the release of two GTPases (Nog1 and Nug1), the assembly factor Nsa2, the subunit anti-association factor Tif6, the ribosomal-like protein Mrt4 and the export factors Mex67-Mtr2, Bud20, Arx1 and Nmd3 (Loibl et al., 2014; Zisser et al., 2018). Following Rlp24 release, the 60S maturation pathway, via a yet unknown mechanism, bifurcates to proofread the polypeptide exit tunnel (PET) and to assemble the ribosomal stalk (Lo et al., 2010). PET maturation is accomplished through the release of Arx1, an assembly factor that covers the tunnel and prevents the premature loading of the protein-folding chaperone machinery (Bradatsch et al., 2012; Greber et al., 2012).

Arx1 release from the pre-60S requires the cytoplasmic zinc-finger protein Rei1, the DnaJ domain-containing protein Jjj1 and Ssa1/Ssa2 (Hsp70) ATPase activity (Meyer et al., 2007; Meyer et al., 2010). Cryo-electron microscopy (cryo-EM) studies have revealed that Rei1 inserts its C-terminal tail (Rei1-CTT) into the PET (Greber et al., 2016). Failure to insert the Rei1-CTT into the tunnel impairs Arx1 release, and blocks subsequent pre-60S maturation (Greber et al., 2016). Cryo-EM analyses of a nuclear pre-60S revealed that the C-terminal tail of the GTPase Nog1 (Nog1-CTT) intertwines around Rlp24, makes contact with Arx1 and inserts its end into the PET (Wu et al., 2016). For Rei1-CTT to gain access to the PET, the Nog1-CTT end must be extracted. How these events are orchestrated during cytoplasmic maturation is unknown.

Another functional center whose assembly is completed in the cytoplasm is the 60S ribosomal stalk. The stalk is built from a single copy of ribosomal proteins uL10 (Rpp0) and two heterodimers of P1 (Rpp1) and P2 (Rpp2), and plays an essential role by recruiting and activating translation elongation factors. On a mature 60S subunit, the stalk is anchored through the interaction of uL10 with rRNA and uL11 (Rpl12) (Ben-Shem et al., 2011). During nuclear 60S assembly, the ribosomal-like protein Mrt4 functions as a placeholder for uL10 (Lo et al., 2009; Kemmler et al., 2009). Mrt4 removal from the pre-60S is triggered by the release factor Yvh1 (Kemmler et al., 2009; Lo et al., 2009). uL10 can be loaded onto the pre-60S to assemble the stalk only after Mrt4 is released. How stalk assembly is coordinated with other cytoplasmic maturation steps remains unclear.

Removal of the placeholders Rlp24, Arx1 and Mrt4 from the pre-60S is an essential prerequisite for PET maturation and quality control, polypeptidyl transferase center (PTC) maturation, stalk assembly and the incorporation of the remaining r-proteins. How these spatially distant events on the pre-60S are orchestrated remains unknown. By employing quantitative mass spectrometry, genetic and cell-biological approaches, we reveal that the ATPase Drg1 and the GTPase Nog1 license these events, and ensure that only correctly assembled 60S subunits enter translation.

Results

The nuclear localized GTPase Nog1 co-enriches with a late pre-60S that contains the export factors Nmd3, Bud20, Mex67-Mtr2 and Arx1 (Jensen et al., 2003; Kallstrom et al., 2003). In wild-type (WT) cells, Nog1 is not detected on a cytoplasmic pre-60S purified through tandem-affinity purification (TAP) of Lsg1 (Kressler et al., 2008; Altvater et al., 2012). However, Nog1 mislocalizes to the cytoplasm, and accumulates on Lsg1-TAP particles upon impairment of the cytoplasmic AAA-ATPase Drg1 or upon treating yeast cells with the Drg1-inhibitor diazaborine (DIA) (Pertschy et al., 2007; Loibl et al., 2014). These data show that Nog1 travels to the cytoplasm, where it is released in a Drg1-dependent manner and rapidly recycled back to the nucleus. How and exactly when Nog1 is released from the pre-60S in the cytoplasm is unclear.

Nog1DN accumulates on a cytoplasmic pre-60S particle

The Nog1 G-domain exhibits characteristic G1–G5 motifs (Figure 1A, upper panel) suggesting that GTP-binding or hydrolysis of Nog1, like that of other GTPases such as Nug2/Nog2 and Lsg1, might regulate interactions between Nog1 and the pre-60S particle (Hedges et al., 2005; Matsuo et al., 2014). Dominant-negative mutations have been described within the G1 motif of Lsg1(K349N/R/T) (Hedges et al., 2005) and the G3 motif of Nug2 (G369A) (Bourne et al., 1991; Hedges et al., 2005; Matsuo et al., 2014) that impair the release of these GTPases from a pre-60S. A G224A mutation in the G3 motif of human Nog1, which presumably blocks GTP hydrolysis, was shown to be dominant negative in mammalian cells, and induced pre-rRNA processing and assembly defects (Lapik et al., 2007).

Nog1DN accumulates on a cytoplasmic pre-60S.

(A) Domain organization and location of Nog1 on a pre-60S (PDB-3JCT). NTD: N-terminal domain; CTT: C-terminal tail. The black arrow points to the G223A mutation in the G3 motif within the G-domain of Nog1. Selected assembly factors (Mrt4, Nsa2, Tif6, Rlp24, Bud20 and Arx1) that are bound to the pre-60S are shown in color. (B) NOG1-G223A (NOG1DN) is dominant negative. Wild-type cells were transformed with a plasmid encoding NOG1 or NOG1DN under control of the galactose-inducible GAL1 promoter and spotted in 10-fold dilutions on glucose- or galactose-containing media. Plates were incubated at 30°C for 2–4 days. (C) Nog1DN accumulates on a cytoplasmic pre-60S. Early nucleolar (Ssf1-TAP) and cytoplasmic (Lsg1-TAP) particles were isolated from either Nog1- or Nog1DN-expressing cells. Co-enrichment of Nog1 was investigated by Western blotting. Whole-cell extracts (WCEs) depict Nog1 protein levels in the indicated strains. (D) Nog1DN-GFP accumulates in the cytoplasm. Wild-type cells expressing either Nog1-GFP or Nog1DN-GFP under control of GAL1 were grown in raffinose-containing synthetic medium to early log phase and then supplemented with 2% galactose. After 30 min incubation, cells were washed, incubated in YPD for 3 hr and then visualized by fluorescence microscopy. Scale bar = 5 µm. Quantification is listed in Supplementary file 3. WCEs depict Nog1 protein levels in the indicated strains.

We investigated whether the G-domain contributes to Nog1 release from the pre-60S in the cytoplasm. To this end, the orthologous mutation in the G3 motif (G223A) of yeast Nog1 (Figure 1A), hereafter termed Nog1DN, was transformed into a Nog1 shuffle strain, wherein the NOG1 gene was disrupted but the viability of the yeast cells was maintained through a centromeric plasmid containing a WT copy of NOG1. We did not obtain transformants for the Nog1DN mutant. To demonstrate dominant-negative behavior, we placed the Nog1DN mutant under the control of an inducible GAL1 promoter, and transformed this plasmid into WT yeast cells. On glucose-containing medium, where Nog1DN expression is repressed, the resulting transformants grew similar to WT. By contrast, expression of Nog1DN in galactose-containing medium was lethal to yeast cells (Figure 1B), confirming the dominant-negative behavior of the G223A mutation.

Nog1 is recruited to the pre-60S in the nucleolus (Kressler et al., 2008; Altvater et al., 2012), and is released from the particle in the cytoplasm (Pertschy et al., 2007; Lo et al., 2010; Altvater et al., 2012). We investigated whether the Nog1DN mutant was released from the pre-60S in the cytoplasm. For this, we isolated the Lsg1-TAP particle after inducing expression of either Nog1 or the Nog1DN mutant allele for 2.5 hr (Figure 1C). Western analyses revealed that Nog1DN mutant protein, but not Nog1, accumulated on the Lsg1-TAP particle (Figure 1C). Whole cell extracts (WCE) revealed similar Nog1 and Nog1DN protein levels (Figure 1C), suggesting that Nog1DN co-enrichment with Lsg1-TAP is not due to altered expression of the mutant protein. Moreover, the Nog1DN-GFP fusion showed an increase in cytoplasmic signal, supporting the notion that Nog1DN release from the pre-60S in the cytoplasm is impaired (Figure 1D). Although a nuclear signal of Nog1DN-GFP is observed in these cells, this mutant did not efficiently co-enrich with Ssf1-TAP under the same conditions (Figure 1C), possibly owing to blockage of downstream cytoplasmic maturation steps that indirectly impair early assembly steps (see later). We conclude that a functional G-domain is essential to evict Nog1 from the pre-60S in the cytoplasm.

Nog1DN impairs cytoplasmic maturation of the pre-60S particle

We investigated the consequences of impaired Nog1DN release on the composition of the cytoplasmic Lsg1-TAP particle by Sequential Window Acquisition of all THeoretical fragment ion spectra mass spectrometry, also termed SWATH-MS. SWATH-MS is a mass spectrometry approach that combines data-independent acquisition with a peptide-centric data query strategy (Gillet et al., 2012). In contrast to selected reaction monitoring mass spectrometry (SRM-MS) (Picotti and Aebersold, 2012), SWATH-MS can be extended to the analysis of any peptide and protein of interest post-acquisition, while maintaining optimal consistency of quantification in pull-down samples (Collins et al., 2013; Lambert et al., 2013).

We interrogated quantitatively the protein composition of four well-characterized pre-60S particles representing different maturation stages (Nissan et al., 2002): Ssf1-TAP, an early nucleolar particle; Rix1-TAP, a nucleoplasmic particle; Arx1-TAP, a particle loaded with nuclear export factors; and Lsg1-TAP, an exclusively cytoplasmic pre-60S. The data were analyzed using OpenSWATH software (Röst et al., 2014), and accuracy was compared with that of SRM-MS based analyses (Altvater et al., 2012). We found that the proteomic heat map obtained using SWATH-MS was in agreement with that generated through SRM-MS (Altvater et al., 2012) and Western analyses (Figure 2—figure supplement 1). In contrast to SRM-MS, SWATH-MS permitted the quantitation of the approximate residence time of nearly all assembly factors along the 60S maturation pathway (Figure 2).

Figure 2 with 1 supplement see all
Association of assembly factors with pre-60S at different maturation stages revealed by SWATH-MS.

Pre-60S particles representing nucleolar (Ssf1-TAP), nuclear (Rix1-TAP), nuclear to cytoplasmic (Arx1-TAP) and cytoplasmic (Lsg1-TAP) stages were analyzed by SWATH-MS. A heat map was generated on the basis of the average of three independent biological replicates and depicts the relative and individual enrichment of assembly factors with maturing pre-60S particles. The acquired SWATH data were normalized on the basis of the average intensities of three depicted 60S r-proteins, and the intensity of each factor was scaled to the highest intensity of that factor in the selected condition represented in the heat map. Maximum enrichment is depicted in purple and minimum enrichment in gold. Assembly factors with similar maximum enrichment profiles were then manually grouped into clusters on the basis of the available cryo-EM data (nucleolar pre-60S: PDB-6C0F and PDB-6ELZ; nuclear pre-60S: PDB-3JCT and PDB-5FL8; cytoplasmic pre-60S: PDB-5H4P, PDB-5T62, PDB-5APN, PDB-6RZZ and PDB-6N8N). ¥ depicts factors whose protein levels were analyzed by Western blotting (Figure 2—figure supplement 1).

We correlated the protein heat map with reported pre-ribosome cryo-EM structures (Sanghai et al., 2018; Kater et al., 2017; Wu et al., 2016; Barrio-Garcia et al., 2016; Ma et al., 2017; Malyutin et al., 2017; Greber et al., 2016). These analyses allowed organization of assembly factors into different clusters on a maturing pre-60S. For example, the early nucleolar Ssf1-Rrp14-Rrp15-Mak11 cluster, the nuclear ITS2 factors Nop15-Rlp7-Cic1-Nop7, the 5S RNP-associated Rrs1-Rpf2, the Rix1-Rea1 machinery and the Drg1-dependent factors Rlp24-Nog1-Nsa2-Bud20 appear to undergo coordinated and grouped release. Our analyses also revealed temporal association of assembly factors for which structural information in the context of the pre-60S is lacking. These factors include nucleolar proteins and the uncharacterized nuclear-localized Tma16 (Figure 2). We conclude that SWATH-MS is a reliable tool to quantify the protein contents of pre-ribosomal particles.

Next, we quantified and compared the protein contents of genetically trapped cytoplasmic pre-60S particles isolated from yeast cells expressing either the dominant-negative Drg1-E617Q (Lsg1-TAP:Drg1DN) (Altvater et al., 2012) or the Nog1DN (Lsg1-TAP:Nog1DN) mutant. We focused on those assembly factors that are known to participate in pre-60S cytoplasmic maturation (Figure 3A). In agreement with previous studies (Pertschy et al., 2007; Lo et al., 2010; Altvater et al., 2012), a Drg1DN:Lsg1-TAP particle accumulated Rlp24, Bud20, Nog1, Mrt4, Nmd3 and Tif6 (Figure 3A and Figure 3B). In addition, the cytoplasmic chaperone Sqt1, which loads uL16 onto the pre-60S (Pausch et al., 2015), and the uncharacterized ribosome-associated factor Tma16 (Fleischer et al., 2006) accumulated on the Drg1DN:Lsg1-TAP particle (Figure 3B). Like the Drg1DN:Lsg1-TAP particle, the Lsg1-TAP:Nog1DN particle accumulated Mrt4, Sqt1, Tma16, Nmd3 and Tif6 (Figure 3A and Figure 3B). However, unlike the Drg1DN:Lsg1-TAP particle, Rlp24 and its interaction partner Bud20 did not accumulate on the Lsg1-TAP:Nog1DN particle (Figure 3A and Figure 3B). Both Drg1DN and Nog1DN-trapped Lsg1-TAP particles failed to recruit Yvh1. Although Rei1 recruitment to the Lsg1-TAP:Drg1DN particle was impaired, Nog1DN-trapped particles recruited Rei1 (Figure 3C). By contrast, Reh1, which functionally overlaps with Rei1 (Parnell and Bass, 2009), accumulated on both Drg1DN and Nog1DN-trapped Lsg1-TAP particles (Figure 3B). The relative co-enrichments of assembly factors on the Lsg1-TAP:Nog1DN particle monitored by SWATH-MS were in agreement with Western analyses (Figure 3C). These data suggest that Nog1DN-expressing cells are impaired in a subset of events along the 60S cytoplasmic maturation pathway.

Nog1DN impairs cytoplasmic maturation of a pre-60S.

(A) Current model for the cytoplasmic maturation pathway of a pre-60S. A cytoplasmic pre-60S at different maturation stages is represented as a cartoon. Energy-consuming assembly factors are indicated with ATP or GTP. Events that are impaired by Drg1DN and Nog1DN expression are highlighted in blue and red, respectively. (B) SWATH-MS analysis of Lsg1-TAP pre-60S in Drg1DN- and Nog1DN-expressing cells. Assembly factors accumulating on the pre-60S particle in a Drg1- or Nog1-dependent manner were clustered on the basis of their increased enrichment. (C) Western analyses of selected pre-60S assembly factors involved in cytoplasmic maturation. Nog1DN-trapped Lsg1-TAP particles were subjected to Western analysis using the indicated antibodies. CBP: calmodulin-binding peptide present in the Lsg1 TAP-tag.

Nog1DN does not hinder initiation of cytoplasmic maturation

Cryo-EM studies have revealed that different domains of the assembly factor Nog1 (NTD, G-domain and CTT) meander from the peptidyl transferase center (PTC) to the PET, and contact assembly factors on the pre-60S (Wu et al., 2016). The Nog1-NTD interacts with the assembly factor Nsa2, whereas the Nog1-G domain contacts the ribosomal-like protein Mrt4, and the Nog1-CTT intertwines around Rlp24 and contacts Arx1 (Figure 1A). Although Nog1, Nsa2, Arx1, Bud20, Mrt4 and Rlp24 are already recruited to the pre-60S during nucleolar/nucleoplasmic maturation, this cluster of assembly factors (Figure 1A) is evicted only in the cytoplasm (Figure 3A) (Pertschy et al., 2007; Lo et al., 2010; Altvater et al., 2012).

To dissect the impact of impaired Nog1DN release from the pre-60S on the cytoplasmic maturation pathway, we employed a cell-biological approach. Release of the ribosomal-like-protein Rlp24 from the pre-60S initiates the cytoplasmic maturation cascade (Figure 3A) (Pertschy et al., 2007; Lo et al., 2010). On the pre-60S, Rlp24 directly contacts Bud20, and then intertwines around the Nog1-CTT, forming a Nog1-CTT:Rlp24 complex (Figure 4A). The last 40 residues of the Rlp24 C-terminal tail (155–190), which remain unresolved in cryo-EM structures (Wu et al., 2016; Zhou et al., 2019), recruit and activate the AAA-ATPase Drg1 to extract Rlp24 from the pre-60S (Pertschy et al., 2007; Lo et al., 2010; Kappel et al., 2012; Altvater et al., 2012). Given the intimate interactions between Nog1-CTT and Rlp24 on the pre-60S (Wu et al., 2016) (Figure 4A), we wondered whether Nog1DN hinders Drg1-mediated Rlp24 release. As judged by immunofluorescence, Rlp24-TAP did not mislocalize to the cytoplasm upon Nog1DN expression (Figure 4B), suggesting that Nog1DN accumulation on a cytoplasmic pre-60S did not interfere with the release and recycling of Rlp24 back to the nucleus. These data are consistent with those from SWATH-MS analyses, which show that Rlp24 does not accumulate on the Lsg1-TAP:Nog1DN particle (Figure 3B). This finding is in agreement with previous studies which showed that Drg1-ATPase activity is necessary and sufficient to release Rlp24 from the pre-60S (Kappel et al., 2012). The presence of Nog1DN on the pre-60S did not interfere with recruitment of r-protein eL24 (yeast Rpl24), as judged by Western analyses (Figure 3C). Cell-biological studies indicate that the release of the Rlp24-interacting assembly factor Bud20 from the pre-60S is also not impaired in Nog1DN-expressing cells (Figure 4C). We conclude that Nog1DN does not disturb initiation of the pre-60S cytoplasmic maturation cascade.

Nog1DN impairs stalk assembly and terminal maturation steps.

(A) Cryo-EM structure (PDB-3JCT) of a Nog1-containing pre-60S particle, depicting the interaction between Rlp24 and Nog1. (B) A RLP24-TAP strain expressing NOG1 or NOG1DN under the control of GAL1 was grown in raffinose-containing synthetic medium to early log phase, then supplemented with 2% galactose, and further grown for 3 hr. Cells were then fixed and prepared for indirect immunofluorescence using an anti-Protein A antibody. Cells were visualized by fluorescence microscopy. Scale bar = 5 µm. Quantification is listed in Supplementary file 3. (C) The indicated GFP-tagged strains expressing NOG1 or NOG1DN under control of GAL1 were grown as in panel (B) and then visualized by fluorescence microscopy. Scale bar = 5 µm. Quantification of the data is listed in Supplementary Table 2. (D) Yvh1 clashes with the Nsa2:Nog1-G domain. Cryo-EM structures of nuclear Nog1-containing (PDB-3JCT) and late cytoplasmic Yvh1-containing (PDB-6RZZ) pre-60S particles were superimposed. Release of the Nsa2:Nog1-G domain complex allows recruitment of Yvh1 to the pre-60S.

While the Lsg1-TAP:Drg1DN particle accumulated Rlp24, Drg1DN, Nog1 and another GTPase Nug1 (Altvater et al., 2012), the Lsg1-TAP:Nog1DN particle accumulated only Nog1DN, and not Rlp24, Drg1 or Nug1 (Figure 3B and Figure 4C). On the basis of these data, we conclude that Nog1DN eviction from the pre-60S occurs after Drg1-mediated Rlp24 release.

Nog1DN interferes with ribosomal stalk assembly

Following Drg1-mediated Rlp24 release from the pre-60S, the cytoplasmic maturation pathway bifurcates to: (1) assemble the ribosomal stalk, and (2) mature the PET and proofread its integrity (Figure 3A) (Lo et al., 2010). The ribosomal stalk is assembled in the cytoplasm from the acidic r-proteins uL10 (yeast P0) and P1/P2 heterodimers. Mrt4 functions as a nuclear placeholder for uL10 and joins the pre-60S in the nucleolus (Lo et al., 2010; Kemmler et al., 2009). Mrt4 release in the cytoplasm requires Yvh1, an assembly factor that maximally co-enriches with Lsg1-TAP (Kemmler et al., 2009; Lo et al., 2010). SWATH-MS and Western analyses revealed that Lsg1-TAP:Nog1DN failed to recruit Yvh1, and accumulated Mrt4 (Figure 3B and Figure 3C). In agreement with these data, we found that Nog1DN-expressing cells mislocalized Mrt4-GFP into the cytoplasm (Figure 4C). Accordingly, Western analyses showed that uL10 recruitment to Lsg1-TAP:Nog1DN was impaired (Figure 3C). Cryo-EM studies suggest that the Yvh1-binding site on the pre-60S clashes with the Nog1-G domain (Figure 4D) (Zhou et al., 2019; Kargas et al., 2019), providing an explanation as to why Nog1 eviction is critical for stalk assembly. Given that Drg1-ATPase activity is required to evict Nog1, these structural data explain why Drg1DN-expressing cells are impaired in Mrt4 release, and consequently stalk assembly (Altvater et al., 2012; Lo et al., 2010). We conclude that stalk assembly requires Drg1-ATPase activity and a functional Nog1-GTPase domain.

Nog1DN does not impair PET maturation

We investigated the impact of Nog1DN expression on PET maturation and quality control (Figure 3A). During nuclear assembly of the pre-60S, the aminopeptidase fold of Arx1 covers the PET (Bradatsch et al., 2012; Greber et al., 2012). Arx1 eviction from the pre-60S is triggered in the cytoplasm by Rei1 (Meyer et al., 2010). Rei1 is recruited to the pre-60S through its N-terminal domain (Lebreton et al., 2006), whereas its C-terminal tail-end (Rei-CTT) is inserted into the PET (Greber et al., 2016). Rei1 recruitment to the pre-60S requires Drg1-mediated Rlp24 release (Figure 5A) (Lo et al., 2010; Altvater et al., 2012). We found that Nog1DN expression did not interfere with Rei1 recruitment (Figure 3B and Figure 3C). However, co-expression of Nog1DN and Drg1DN impaired Rei1 recruitment to the pre-60S (Figure 5A). Accordingly, Nog1DN-expressing cells when treated with the Drg1-inhibitor diazaborine (DIA) mislocalized Arx1-GFP to the cytoplasm in these cells (Figure 5B). Cryo-EM studies indicate that the Rei1-binding site on the pre-60S clashes with the Nog1-CTT:Rlp24 complex (Figure 5C) (Zhou et al., 2019), providing an explanation as to why PET maturation is inhibited in Drg1DN-expressing (or DIA-treated) cells. All of these data show that Rlp24 extraction from the Nog1-CTT is critical to recruit Rei1 to the pre-60S in order to initiate PET maturation.

Nog1DN does not impair PET maturation and quality control.

(A) Drg1DN impairs Yvh1 and Rei1 recruitment, whereas Nog1DN impairs only Yvh1 and not Rei1 recruitment. Lsg1-TAP was co-transformed with plasmids each expressing copper-inducible Drg1 or Drg1DN and galactose-inducible Nog1 or Nog1DN, respectively. Cells were grown in raffinose-containing synthetic medium to early log phase and then supplemented with 0.5 mM copper sulphate and 2% galactose. After 3 hr, cells were lysed, and the Lsg1-TAP particle was isolated and subjected to Western analyses using indicated antibodies. (B) Indicated strains expressing Nog1DN under the control of GAL1 were treated with DMSO or diazaborine to block Drg1 activity and visualized by fluorescence microscopy as in Figure 4. Scale bar = 5 µm. Quantification of the data is listed in Supplementary Table 2. (C) The Rei1-binding site on the pre-60S clashes with the Nog1-CTT:Rlp24 complex. Cryo-EM structures of Nog1-containing (PDB-3JCT) pre-60S and a Rei1-containing (PDB-6RZZ) pre-60S were superimposed. Release of Rlp24 allows recruitment of Rei1 to the pre-60S. (D) Arx1 mislocalizes to the cytoplasm in a rei1-TAP mutant expressing NOG1DN. The ARX1-GFP rei1Δ strain was transformed with either REI1- or rei1-TAP-encoding plasmids under control of their natural promoter. Subsequently, the resultant strains were transformed with plasmids encoding galactose-inducible NOG1 or NOG1DN, respectively. The transformants were grown in raffinose-containing medium to early log phase and then supplemented with 2% galactose. After 3 hr, cells were visualized by fluorescence microscopy. Scale bar = 5 µm. Quantifications are listed in Supplementary file 3.

Nog1DN-expressing cells did not mislocalize Arx1-GFP into the cytoplasm (Figure 5D), suggesting that Nog1DN presence on the pre-60S does not interfere with the ability of Rei1-CTT to probe the PET. Failure to insert Rei1-CTT into the PET by attaching a bulky domain (rei1-TAP) blocks Arx1 release and progression of cytoplasmic maturation, suggesting that PET proofreading represents a quality control check point (Greber et al., 2016). Consistent with this, Nog1DN expression in a rei1-TAP mutant (in which Rei1-CTT function in probing the PET is compromised) mislocalized Arx1-GFP to the cytoplasm (Figure 5D). Before Rei1-CTT can probe PET integrity, the terminal end of Nog1-CTT needs to be extracted from the PET. Given that Rei1 recruitment and PET probing is not impaired in Nog1DN-expressing cells, we suggest that Nog1DN-CTT has been removed from the PET, and this does not require a functional Nog1-G3 domain. We conclude that the presence of Nog1DN on the pre-60S does not hinder Rei1 recruitment or the insertion of the Rei1-CTT into the PET.

The Nog1-CTT:Rlp24 complex negatively regulates PET maturation

Nog1-CTT intertwines around the helical segments of Rlp24, then contacts Arx1 and finally inserts its terminal end into the PET (Figure 1A). We investigated which regions of Nog1-CTT contribute to pre-60S cytoplasmic maturation (Figure 6A and Figure 6B). We found that a Nog1 mutant lacking the entire CTT, including the Rlp24-interacting region, Nog11–426, was lethal in yeast (Figure 6B). Overexpression of a Nog1 mutant lacking the Rlp24-interacting region, Nog1∆427–536, severely impaired the growth of yeast in a dominant-negative manner (Figure 6C). However, unlike the Nog1DN mutant, Nog1∆427–536-expressing cells did not mislocalize Mrt4-GFP and Tif6-GFP to the cytoplasm (Figure 6C). Likewise, a Rlp24 mutant lacking Nog1-interacting helices, Rlp24∆91–105, also impaired yeast growth in a dominant-negative manner (Figure 6A and Figure 6C). Expression of the Rlp241–146 mutant (Figure 6A) that lacks the Drg1-binding platform, and therefore cannot be extracted from pre-60S, mislocalized Arx1-GFP, Mrt4-GFP and Tif6-GFP to the cytoplasm (Figure 6C) (Lo et al., 2010). However, Rlp24∆91–105-expressing cells did not mislocalize these factors to the cytoplasm (Figure 6C). Given that Nog1 and Rlp24 rely on each other for their recruitment to the pre-60S (Saveanu et al., 2003), the toxicity of Nog1∆427–536 and Rlp24∆91–105 mutants probably reflects a failure to form a Nog1-CTT:Rlp24 complex on the pre-60S during early nuclear maturation. In contrast to the lethal Nog11–426 mutant, yeast strains expressing Nog1-CTT truncation mutants containing the Rlp24-interacting region but lacking the rest of the CTT complemented the lethality of the nog1∆ strain, but induced impaired growth at 20°C and 25°C (Figure 6A and Figure 6B). In agreement with these observations, growth of the Nog1-GFP strain was severely impaired at these temperatures (Figure 6B). We monitored PET maturation (Arx1-GFP), stalk assembly (Mrt4-GFP) and Tif6-GFP localization after treating the Nog11–479 truncation mutant (which retains the Rlp24-interaction platform) with DIA, which impairs Drg1-mediated Rlp24 release. We found that Arx1-GFP mislocalized to the cytoplasm in these cells (Figure 6D), emphasizing the importance of Rlp24 release and of the disassembly of the Nog1-CTT:Rlp24 complex for PET maturation. As expected, Mrt4-GFP and Tif6-GFP also mislocalized to the cytoplasm (Figure 6D), consistent with the idea that Nog1 release from the pre-60S requires Rlp24 removal. In conjunction with cryo-EM studies (Figure 5C), these data support the idea that the Nog1-CTT:Rlp24 complex negatively regulates PET maturation.

The Nog1-CTT:Rlp24 complex negatively regulates PET maturation.

(A) Domain organization of Nog1 and Rlp24 showing interaction surfaces, and sites for truncation analyses. (B) Growth analysis of Nog1 truncations. Left panel: a Nog1 shuffle strain was transformed with the indicated constructs and spotted in 10-fold dilutions 5′-fluoroorotic acid (5-FOA) (SD) plates and incubated for 2–6 days at the indicated temperatures. Right panel: a nog1Δ shuffle strain was transformed with the indicated plasmids. After shuffling out the URA3-NOG1 plasmid on FOA-containing plates, strains were spotted in 10-fold serial dilutions on YPD plates and incubated for 2–4 days at the indicated temperatures. (C) Rlp24:Nog1 interactions contribute to early nucleolar/nuclear maturation. Left panel: BY wild-type cells were transformed with the indicated plasmids under control of the galactose-inducible GAL1 promoter and spotted in 10-fold dilutions on glucose- or galactose-containing media. Plates were incubated at 30°C for 2–4 days. Right panel: Mrt4-GFP, Arx1-GFP and Tif6-GFP strains were transformed with the indicated plasmids under control of the GAL1 promoter. Cells were grown in raffinose-containing synthetic medium to early log phase and then supplemented with 2% galactose. After 3 hr, cells were visualized by fluorescence microscopy. Scale bar = 5 µm. Quantifications are listed in Supplementary file 3. (D) Drg1-mediated release of Rlp24 from Nog1 is required for PET assembly. Nog1 shuffle strains containing GFP-tagged versions of assembly factors were transformed with Nog1 and Nog11–479 encoding plasmids. After shuffling out URA3-NOG1 plasmid on 5-FOA, the resultant strains were grown in YPD and were treated with DMSO or diazaborine (DIA) for 30 mins to inhibit Drg1-ATPase activity. The GFP tagged assembly factor was visualized by fluorescence microscopy. Scale bar = 5 µm. Quantifications are listed in Supplementary file 3.

Nog1DN blocks the terminal cytoplasmic maturation steps

The terminal steps during cytoplasmic maturation of a pre-60S involve sequential release of the nuclear export signal (NES)-containing Crm1 adaptor Nmd3 and the 60S anti-association factor Tif6 (Weis et al., 2015; Ma et al., 2017; Kargas et al., 2019). Nog1DN expression accumulated Nmd3 and Tif6 on the pre-60S as determined by SWATH-MS and Western analyses (Figure 3B and Figure 3C). To monitor Nmd3 accumulation on a cytoplasmic pre-60S by cell-biological means, we employed the nuclear-localized Nmd33A-GFP mutant fusion, which harbors mutations in its NES that decrease its nuclear export rate (Hedges et al., 2005). As expected, at steady state, Nmd33A-GFP and Tif6-GFP localize to the nucleus in cells expressing WT Nog1 (Figure 4C). However, in Nog1DN-expressing cells, Nmd33A-GFP and Tif6-GFP mislocalized to the cytoplasm (Figure 4C). We suggest that the presence of Nog1DN on the pre-60S impairs Nmd3 and Tif6 release.

Like Nog1DN-expressing cells, yvh1Δ cells mislocalize both Mrt4-GFP and Tif6-GFP to the cytoplasm, impairing stalk assembly and progression of the cytoplasmic maturation pathway (Kemmler et al., 2009; Lo et al., 2010). Given that Yvh1 recruitment to the pre-60S is impaired in Nog1DN-expressing cells, we investigated whether the dominant-negative behavior of Nog1DN is due to the failure to release Mrt4. To test this, we employed a dominant gain-of-function allele MRT4G68E that bypasses Yvh1-mediated Mrt4 release from the pre-60S (Figure 7A and Figure 7B) and relocalizes Tif6-GFP to the nucleus (Figure 7A) (Kemmler et al., 2009). We found that Nog1DN expression is toxic in yvh1∆MRT4G68E cells (Figure 7B). Although Mrt4G68E-GFP relocalized to the nucleus, Tif6-GFP still mislocalized to the cytoplasm in these cells (Figure 7A), showing that the presence of Mrt4 on the pre-60S is not the cause of the dominant-negative behavior of Nog1DN. Consistent with the sequential release of Nmd3 and Tif6 from the pre-60S, Western analyses show that Nmd3 accumulates on the Lsg1-TAP particle isolated from yvh1∆MRT4G68E cells expressing Nog1DN (Figure 7C). Notably, the r-protein uL10 is recruited to this Lsg1-TAP particle (Figure 7C), indicating that Nog1DN does not interfere with stalk assembly in yvh1∆MRT4G68E cells. In light of a recent cryo-EM study, these data suggest that Nog1DN presence on the pre-60S prevents the rearrangement of rRNA helix H89, and sterically hinders the incorporation of uL16 into its RNA-binding site (Figure 7D), a critical step that triggers downstream events that drive a pre-60S toward translation competence (Kargas et al., 2019; Zhou et al., 2019).

Nog1DN blocks the terminal cytoplasmic maturation steps.

(A) Nog1DN impairs recycling of Mrt4 and Tif6. The mrt4Δyvh1Δ strain was co-transformed with plasmids encoding Mrt4-GFP or Mrt4G68E-GFP and Yvh1 or galactose-inducible Nog1DN, respectively. To monitor Tif6 localization, the strain was co-transformed with plasmids encoding Tif6-GFP, Mrt4 or Mrt4G68E-GFP and Yvh1 or galactose-inducible Nog1DN, respectively. Cells were grown in raffinose-containing synthetic medium to early log phase and then supplemented with 2% galactose. After 3 hr, cells were visualized by fluorescence microscopy. Scale bar = 5 µm. Quantifications are listed in Supplementary file 3. (B) Mrt4G68E bypasses the need for YVH1, but does not rescue lethality associated with Nog1DN expression. The depicted strains were spotted in 10-fold dilutions on galactose-containing media and incubated at 30°C for 2–4 days. (C) Nog1DN impairs recruitment of Yvh1, but not release of the Mrt4-G68E gain-of-function mutant. A Lsg1-TAP mrt4Δ strain was co-transformed with plasmids encoding Mrt4 or Mrt4G68E and empty vector or galactose-inducible Nog1DN, respectively. Cells were grown in raffinose-containing synthetic medium to early log phase and then supplemented with 2% galactose. After 3 hr, cells were lysed, purified through Lsg1-TAP and subjected to western analyses using the indicated antibodies. (D) Nog1 release permits uL16 recruitment. Superimposition of the cryo-EM structures of a nuclear Nog1-containing pre-60S particle (PDB-3JCT) and uL16 on a mature 60S subunit (PDB-6R84) show how the Nog1-NTD and Nog1 G-domain prevent recruitment of uL16 to the pre-60S.

Discussion

During its journey from the nucleolus to the cytoplasm, and concomitant progression towards the mature subunit, a pre-60S interacts with diverse energy-consuming enzymes. However, the precise order of recruitment and eviction, as well as the mechanisms that co-ordinate the different energy-consuming steps remain unknown. Here, we reveal that sequential Drg1-ATPase and Nog1-GTPase activities license early cytoplasmic maturation events that eventually drive a pre-60S towards translation competence.

Nog1 eviction requires a functional G-domain and Drg1-ATPase activity

The GTPase Nog1 travels into the cytoplasm with the pre-60S, where it is released in a Drg1-ATPase manner and recycled back into the nucleus (Pertschy et al., 2007; Lo et al., 2010). Our finding that the dominant negative Nog1DN (Nog1G223A) mutant accumulates on Lsg1-TAP (Figure 1C) supports the notion that an impaired G-domain does not hinder recruitment to the assembling pre-60S particle. Instead, like other GTPases involved in the 60S maturation pathway (Hedges et al., 2005; Matsuo et al., 2014), an impaired G3 domain blocks Nog1 release from the pre-60S. We and others (Lapik et al., 2007) have been unable to detect GTPase activity of recombinant Nog1 in vitro. It could be that GTP hydrolysis of Nog1, like that of its bacterial homolog ObgE, is triggered only in the context of the pre-ribosome (Feng et al., 2014). Nog1 is a predicted potassium-selective cation-dependent GTPase (also referred to as Group I CD-GTPase) (Ash et al., 2012). Group I CD-GTPases can be identified by the presence of two conserved asparagine residues in the G1 motif (Ash et al., 2012). The first of the two conserved asparagine residues (AsnK) is coordinated to the potassium ion, whereas the second asparagine (AsnSwI) facilitates the Switch I structure. In yeast Nog1, AsnK is present (N177), however, the asparagine AsnSwI is substituted with an arginine (R185). This raises the possibility that the second asparagine (or equivalent) might be provided by a yet unknown activator. The equivalent G223A mutation in the G3 motif reported here is also dominant negative in the Ras GTPase (Ford et al., 2005). Structural studies of this mutant revealed that the switch region adopts an open conformation that prevents GTP from inducing an active confirmation (Ford et al., 2005), which sequesters Ras GEFs into non-productive complexes leading to the depletion of their intracellular pool (Ford et al., 2005). It would be interesting to determine the molecular basis that underlies the dominant-negative nature of Nog1DN, and to determine whether it is similar to that for the RasG60A mutant.

Nog1, whose GTPase domain is docked at the P0 stalk base at a position similar to that in bacterial ObgE, is remarkably unique with its CTT, which wraps around half the 60S subunit, passing by the PTC and entering the PET (Wu et al., 2016) (Figure 1A). Covering a total distance of >250 Å, the CTT contacts Rlp24, Bud20, Tif6 and Arx1 (Figure 1A). Nog1-CTT intertwines with the C-terminal helices of Rlp24 forming a Nog1-CTT:Rlp24 complex on the pre-60S (Figure 4A). Intriguingly, despite these intimate interactions, we found that Rlp24 and Nog1DN are independently released from the pre-60S. Previous work showed that Drg1 grips Rlp24 via its C-terminus, and then triggers its ATP-dependent release from the pre-60S (Kappel et al., 2012). Drg1DN expression prevents the release of both Rlp24 and Nog1 from the pre-60S, suggesting that Nog1 eviction also requires Drg1-ATPase activity (Pertschy et al., 2007; Altvater et al., 2012). A recent cryo-EM study identified an early cytoplasmic pre-60S state in which Nog1-CTT was inserted into the PET, but the Nog1-NTD and Nog1-G domains were displaced from the pre-60S (Zhou et al., 2019). In this particle, Nog1-CTT intertwines around Rlp24, providing an explanation as to why Nog1 remains bound to the pre-60S. We suggest that Nog1 eviction from the pre-60S requires a functional Nog1 G-domain and Drg1-mediated extraction of Rlp24.

Another factor that makes contact with Rlp24 is the export factor Bud20 (Altvater et al., 2012; Bassler et al., 2012). It seems likely that Bud20 release from the pre-60S is a consequence of Rlp24 release. Consistent with this, Drg1DN expression interferes with both Rlp24 and Bud20 release and recycling from the pre-60S. Given that localization of Rlp24 and Bud20 is not altered upon Nog1DN expression (Figure 4C), we suggest that Bud20 release from the pre-60S occurs prior to bifurcation of the cytoplasmic maturation pathway.

The Nog1-CTT:Rlp24 complex gates PET quality control

Drg1DN expression impairs Rei1 recruitment to the pre-60S and consequently Arx1 release (Altvater et al., 2012). Drg1-mediated Rlp24 extraction from the Nog1-CTT:Rlp24 complex permits Rei1 recruitment to the pre-60S, and subsequently allows the Rei1-CTT to probe PET integrity for Arx1 release. Rei1 recruitment to the pre-60S is unaffected in Nog1DN-expressing cells. Given that PET maturation is successfully accomplished in Nog1DN-expressing cells, we suggest that Nog1DN-CTT has been extracted from the PET as a consequence of Rlp24-extraction from the pre-60S. Thus, Nog1 eviction from the pre-60S is not required to accomplish PET maturation.

Recent cryo-EM studies revealed that the cytoplasm-localized assembly factor Reh1, which functionally overlaps with Rei1 (Parnell and Bass, 2009; Greber et al., 2016) also inserts its C-terminal tail into the PET (Ma et al., 2017; Kargas et al., 2019). Our SWATH-MS data indicate that Reh1 accumulates on the Lsg1-TAP:Nog1DN particle (Figure 3B), supporting the notion that Reh1 release occurs after Nog1 release from the pre-60S. These data are consistent with the idea that the PET is under continuous surveillance through sequential interactions with Nog1-CTT, Rei1-CTT and Reh1-CTT.

Nog1 gates stalk assembly and PTC maturation

Stalk assembly requires prior release of Mrt4 that is facilitated by the release factor Yvh1 (Lo et al., 2009; Kemmler et al., 2009). Nog1DN expression interferes with Yvh1 recruitment to the pre-60S particle, and therefore impairs Mrt4 release (Figure 4C and Figure 4D). Cryo-EM studies have revealed that the presence of Nog1 prevents the pre-60S from recruiting Yvh1, as both factors localize close to the stalk base and obstruct the Yvh1-binding site (Figure 4D) (Kargas et al., 2019; Zhou et al., 2019). Given that Nog1 release is impaired in Drg1DN-expressing cells (Altvater et al., 2012), these data provide an explanation as to why stalk assembly is impaired in these cells.

While Mrt4G68E efficiently bypassed the need for Yvh1-mediated release, this gain-of-function allele did not rescue the dominant-negative phenotype of Nog1DN (Figure 7B). These cells were still impaired in Nmd3 and Tif6 release from the pre-60S (Figure 7A and Figure 7C). Retention of Nog1DN on the pre-60S prevents rearrangement of rRNA helix H89, and sterically hinders uL16 incorporation. Stable accommodation of uL16 into the pre-60S is a critical event to re-orient rRNA helices H89 and H38, that permits a conformational rearrangement of the Nmd3 C-terminal domain (Malyutin et al., 2017; Zhou et al., 2019; Kargas et al., 2019). This rearrangement triggers the GTPase Lsg1 to evict Nmd3, and subsequently licenses Efl1/Sdo1 to mediate Tif6 release. Notably, the dedicated chaperone Sqt1 that guides incorporation of uL16 into the pre-60S accumulates on the Lsg1-TAP:Nog1DN particle (Figure 3B). In light of these data and work from the Kressler laboratory (Pausch et al., 2015), it is tempting to speculate that a pre-formed Sqt1:uL16 complex docks on the pre-60S. Nog1 eviction permits transfer of uL16 into its RNA-binding site, and release of its dedicated chaperone Sqt1. We suggest that Nog1 release from a pre-60S licenses stalk assembly and downstream maturation steps that are critical to acquire translation competence.

SWATH-MS analyses allowed the identification of assembly factors whose release or recruitment appears to be dependent on the release of the GTPase Nog1 from cytoplasmic pre-60S particles. These include the assembly factors Sqt1, Mex67, Ecm1, Tma16 and Arb1, for which structural information in the context of a pre-60S is lacking. In contrast to Sqt1, Mex67, Ecm1 and Tma16, which accumulated on the Lsg1-TAP Nog1DN particle, we found that recruitment of the ABC-ATPase Arb1 was impaired (Figure 3B). Arb1 has been implicated in both 40S and 60S subunit assembly (Dong et al., 2005). A recent cryo-EM structure of a translationally stalled-mature 60S subunit shows that Arb1 docks in a region of rRNA that overlaps with the binding site of Nmd3 on the pre-60S (Su et al., 2019). We speculate that Nog1 release, stable uL16 incorporation and Nmd3 release precede Arb1 recruitment to the pre-60S.

Nog1 co-ordinates the assembly, maturation and quality control of distant ribosomal centers

Previously, the Johnson laboratory had ordered cytoplasmic maturation events on the pre-60S into a coherent pathway (Lo et al., 2010). In that pioneering study, AAA-ATPase Drg1 activity was proposed to initiate two parallel branches of the pathway, namely PET maturation with quality control and ribosome stalk assembly (Figure 3A). The two branches converge to initiate Nmd3 and Tif6 release prior to acquiring translational competence. Our data suggest that Nog1 plays a pivotal role in co-ordinating bifurcation and convergence of the 60S cytoplasmic maturation pathway. Cryo-EM structures and the data presented here permit the refinement of early events that drive the cytoplasmic pre-60S maturation (Figure 8). Upon arrival in the cytoplasm, the pre-60S recruits Drg1 via the C-terminal region of Rlp24. This interaction stimulates Drg1 ATPase activity resulting in Rlp24 removal, and as a consequence, the terminal end of Nog1-CTT is extracted from the PET (State I to State II). Removal of Rlp24 permits Rei1 recruitment, and allows Rei1-CTT to probe PET integrity in order to trigger Arx1 release through the J-domain protein Jjj1 and ATPases Ssa1/2 (State III) (Meyer et al., 2007; Greber et al., 2016). Release of Rei1 then permits the assembly factor Reh1 to probe the PET (State IV) (Ma et al., 2017; Kargas et al., 2019). Nog1 eviction from the pre-60S requires the extraction of Rlp24 from Nog1-CTT by the ATPase Drg1 and GTP hydrolysis within the Nog1-G domain. This event unleashes a sequence of events in different regions of the pre-60S: (1) it permits recruitment of the release factor Yvh1 to the pre-60S in order to mediate Mrt4 release and initiate stalk assembly (State IV to State V); (2) it permits loading of r-proteins eL40 and uL16 into their rRNA-binding sites (State V), (Zhou et al., 2019; Kargas et al., 2019); and (3) it re-orientates rRNA helices H89 and H38 to promote a conformational rearrangement of the Nmd3 C-terminal domain (Malyutin et al., 2017; Zhou et al., 2019; Kargas et al., 2019), leading to Nmd3 eviction and thus completing PTC maturation (State V to State VI). We therefore propose that Nog1 serves as a hub that coordinates spatially distant maturation events to ensure only a correctly assembled 60S subunit enters translation.

A revised model for cytoplasmic maturation of the pre-60S.

Drg1 initiates the 60S cytoplasmic maturation pathway by binding to and releasing Rlp24 from the pre-60S bound Rlp24:Nog1-CTT complex in an ATP-dependent manner (State I to State II). Removal of Rlp24 also releases its interacting partner Bud20, and consequently extracts the Nog1 C-terminal tail out of the polypeptide exit tunnel (PET) (State II). Extraction of Nog1-CTT from the PET, permits Rei1-CTT to probe PET integrity (State III). Successful Rei1-CTT insertion into the PET triggers ATP-dependent release of Arx1 by Ssa1/2 and Jjj1, and concurrent Rei1 eviction (State III to State IV). This permits Reh1-CTT insertion into the PET (State IV). Nog1 release permits Yvh1 recruitment to evict Mrt4 and to initiate stalk assembly (State IV to State V). Nog1 eviction also permits eL40 and uL16 incorporation, and Nsa2 release. Stable uL16 accommodation into the pre-60S leads to Lsg1 mediated Nmd3 release, and thereby completes PET maturation. Nmd3 release triggers Reh1 release, and also licenses the GTPase Efl1 and co-factor Sdo1 to evict Tif6 (State V to State VI), thus driving a pre-60S towards translation competence.

Finally, Nsa2 and Reh1 accumulate on Lsg1-TAP in Drg1DN- as well as Nog1DN- expressing cells (Altvater et al., 2012; Figure 3C). These data are consistent with the idea that Nog1 eviction from the pre-60S precedes Nsa2 and Reh1 release. However, the precise mechanisms that drive Nsa2 and Reh1 release remain unknown.

Nmd3 release licenses Efl1/Sdo1 to mediate Tif6 release (State VI) (Kargas et al., 2019). Although Lsg1 appears to be required only during a specific step of the cytoplasmic maturation pathway, it seems that it can be recruited to a newly exported pre-60S. Lsg1 recruitment to the pre-60S does not seem to depend on the initial cytoplasmic maturation events, such as Rlp24 extraction by Drg1 or Nog1 release, as we efficiently isolated Drg1DN- and Nog1DN-trapped particles via Lsg1-TAP (Altvater et al., 2012) (Figure 3B). These purifications suggest that the GTPases Nog1 and Lsg1 can bind simultaneously to cytoplasmic pre-60S particles.

Co-ordination between energy-consuming enzymes ?

Strikingly, the release of the GTPase Nug2 from the pre-60S in the nucleus was reported to require its own GTPase activity, and the ATPase activity of the AAA-ATPase Rea1 (Matsuo et al., 2014). Nug2 binds the pre-60S at a site, which partially overlaps with the binding site for the NES-containing adaptor Nmd3 (Sengupta et al., 2010). Only after Nug2 release can Nmd3 be recruited to the pre-60S, a critical step that drives pre-60S nuclear export. In this case, coupled ATPase and GTPase activities, together with components of this assembly factor cluster (Figure 2), form a nuclear checkpoint that prevents a pre-60S from prematurely acquiring export competence. Our data suggest that Nog1 release, like Nug2 release, from the pre-60S presumably requires its own GTPase activity, and Drg1 ATPase activity. Failure to release Nog1 blocks a specific branch of the cytoplasmic maturation pathway, and consequently PTC formation.

In contrast to their prokaryotic counterparts, formation of the eukaryotic ribosome requires the concerted efforts of >200 assembly factors and >40 ATPases, GTPases, ATP-dependent RNA helicases and kinases that interact with pre-ribosomes at distinct maturation steps. Our workflow of combining genetic trapping with SWATH-MS provides a powerful tool to uncover the intricate coordination between energy-consuming enzymes to release assembly factor clusters sequentially from maturing pre-ribosomes, and thereby reveal checkpoints during ribosome formation.

Materials and methods

Yeast strains and plasmids

Request a detailed protocol

The Saccharomyces cerevisiae strains used in this study are listed in Supplementary file 1. Genomic disruptions, C-terminal tagging and promoter switches at genomic loci were performed according to established protocols (Janke et al., 2004; Longtine et al., 1998; Puig et al., 2001).

The plasmids used in this study are listed in Supplementary file 2. Details of plasmid construction will be provided upon request. All recombinant DNA techniques were performed according to established procedures using Escherichia coli XL1 blue cells for cloning and plasmid propagation. Mutations in NOG1 and RLP24 were generated using the QuikChange site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA, USA). All cloned DNA fragments and mutagenized plasmids were verified by sequencing.

Fluorescence microscopy

Request a detailed protocol

For assessing the localization of PGAL1-Nog1G223A-GFP, BY4741 cells were transformed with YEP351-Nog1-GFP and YEP351gal-Nog1G223A-GFP and grown in SR medium until OD600 = 0.4–0.6, then induced for 30 min with 2% galactose. Cells were then washed once in YPD and incubated in YPD for 3 hr before imaging. For assessing the localization of assembly factors, endogenously GFP-tagged strains were transformed with PGAL1 plasmids and were grown in SR medium until OD600 = 0.2–0.4, and then induced with 2% galactose for 3 hr. For Nmd33A-GFP visualization, a Nmd3 shuffle strain was transformed with Nmd33A-GFP (NMD3I493A L497A L500A) and transformants were applied on FOA plates to shuffle out the wild-type NMD3. The resulting Nmd33A-GFP strain was transformed with PGAL1 plasmids and grown the same as the endogenously GFP-tagged strains. For experiments with diazaborine, cells were incubated 1 hr prior to imaging with either DMSO or with 370 µM diazaborine (dissolved in DMSO; M. Peter, ETH Zürich, Switzerland) in 1 ml of cell culture for 1 hr at 30°C on shaker. When cells were ready to be harvested, the pellet was washed once with H2O. 3 μl of cells were transferred on a microscopy slide (VWR) covered with a glass slip (18 × 18 mm no 1, VWR).

Rlp24-TAP localization was assayed by indirect immunofluorescence. The Rlp24-TAP strain was transformed with pGAL1-containing plasmids, and cultures were grown in the appropriate conditions to OD600 = 0.4–1. After adding formaldehyde (final concentration 4%) to the cultures, cells were fixed for 30 min at 30°C. Cells were then centrifuged, and incubated with 1 ml of 0.1M KPi (pH 6.4) with 3.7% formaldehyde for 15 min at 30°C. Cells were then washed twice with 0.1M KPi (pH 6.4) and once in spheroplasting buffer (0.1 M KPi [pH7.4], 1.2 M sorbitol, 0.5 mM MgCl2). The pellet was resuspended in 200 µl spheroplasting buffer and was either stored at −20°C or directly processed for spheroplasting. For spheroplasting, 2 µl of 1M DTT was added to the 200 µl cells and incubated for 15 min at 30°C. Zymolyase 100T was added (final concentration 50 µg/ml) to the 200 µl and incubated for 10 min. Cells were quickly checked under the microscope to see whether spheroplasts were formed (spheroplasts appear black), and if necessary incubated for a longer time, but no longer than 20 min. Then, the spheroplasts were centrifuged for 2 min at 2000 rpm, washed and resuspended in spheroplasting buffer. 20 µl of poly-L-lysine (0.1% [w/v], Sigma-Aldrich) was applied per well on slides (8-well, 6-mm Menzel-Gläser Diagnostika, Brauschweig, Germany), incubated for 5 min, washed three times with dH2O, aspirated and air-dried. Approximately 40 µl of spheroplasts were added onto a lysine-coated slide well. Non-adhering cells were removed after 30 s and the slide was incubated in an ice-cold methanol bath for 6 min before it was transferred to an ice-cold acetone bath for 10 s. The slide was air-dried and processed with 30 µl BSA/PBS (1x PBS, 1% w/v BSA) for 30 min in the dark at room temperature (RT) in a humid chamber. The blocking solution was removed, and primary antibody (anti-CBP 1:1000; Thermo Scientific, Rockford, IL, USA) in BSA/PBS was added to the wells and incubated for at least an hour (up to overnight) at RT in the dark. Wells were washed three times with BSA/PBS and incubated with secondary antibody, AlexaFluor568 coupled anti-rabbit (Molecular Probes, Inc, Eugene, OR, US), at RT in the dark. After washing, cells were incubated for 30 s with DAPI (1 µg/ml in BSA/PBS; Sigma-Aldrich). Wells were washed three times with BSA/PBS and dried before the slides were mounted with Mowiol (Calbiochem, San Diego, CA, USA).

Fluorescence signal was examined using a Leica DM6000B microscope fitted with a 63 × 1.25 NA 1.30–0.60 NA oil immersion lens (HCX PL Fluotar; Leica). Pictures were acquired with a digital camera (ORCA-ER; Hamamatsu Photonics) and with Openlab software (Perkin Elmer) or with Leica LAS software. Representative sections were selected using ImageJ software and processed in AdobePhotoshop. Corresponding differential interference contrast (DIC) pictures were taken for each fluorescence image. All cell-biological studies were performed on at least three different occasions and in triplicates; >90% of cells showed the reported phenotypes in a sample size of >1000 cells. A summary of the quantification of all of the cell-biological data is listed in Supplementary file 3.

Biochemical analyses

Request a detailed protocol

Whole-cell extracts were prepared by alkaline lysis of yeast cells (Kemmler et al., 2009). Tandem affinity purifications (TAP) of pre-ribosomal particles were carried out as previously described (Faza et al., 2012; Altvater et al., 2014). Calmodulin-eluates were separated on NuPAGE 4–12% Bis-Tris gradient gels (Invitrogen, Carlsbad, CA, USA) and visualized by either silver staining or western analyses using indicated antibodies. All biochemical purifications were performed on at least three different occasions and in triplicates. To analyze the samples by SWATH-MS, protein TEV eluates were precipitated using trichloroacetic acid (TCA), washed once in cold acetone, air dried and processed further (see paragraphs on SWATH-MS analyses).

Western analyses were performed as previously described (Kemmler et al., 2009). The following antibodies were used: α-Mex67 (1:5000; C Dargemont, Institut Jacques Monod, Paris, France), α-Nmd3 (1:5000; A Johnson, University of Texas at Austin, Austin, TX, USA), α-Nog1 (1:1000; M Fromont-Racine, Institut Pasteur, Paris, France), α-Nop7 (1:2000; B Stillman, Cold Spring Harbor Laboratory, New York, NY, USA), α-Nug1 (1:1000; this study), France), α-Tif6 (1:2000; GenWay Biotech, San Diego, CA, USA), α-Yvh1 (1:4000; Altvater et al., 2012), α-Rpl1 (1:10,000; F Lacroute, Centre de Génétique Moléculaire du CNRS, Gif-sur-Yvette, France), α-Rpl3 (1:5000; J Warner, Albert Einstein College of Medicine, Bronx, NY, USA), α-Rpl35 (uL29) (1:4000; Altvater et al., 2012), α-Gsp1 (1:3000; rabbit; Fischer et al., 2015), α-Nsa2 (1:2000; M. Fromont-Racine, Institut Pasteur, Paris, France), α-GFP (1:2000; Roche); and α-FLAG (1:2000; Sigma). The secondary HRP-conjugated α-rabbit and α-mouse antibodies (Sigma-Aldrich, St. Louis, MO, USA) were used at 1:1000–1:5000 dilutions. Protein signals were visualized using an Immun-Star HRP chemiluminescence kit (Bio-Rad Laboratories, Hercules, CA, USA) and captured by Fuji Super RX X-ray films (Fujifilm, Tokyo, Japan) or using ImageQuant LAS 4000 (GE Healthcare).

Sample preparation for the SWATH-MS analyses

Request a detailed protocol

The proteins were solubilized in a denaturing buffer containing 8 M urea and 0.1 M NH4HCO3. They were then reduced with 12 mM DTT at 37°C for 30 min and alkylated with 40 mM iodacetamide at room temperature in the dark for 30 min. The samples were then diluted with 0.1 M NH4HCO3 to reach a final urea concentration of 1M and digested with sequencing-grade porcine trypsin (Promega, 1:100 trypsin:protein). Digestion was stopped by adding formic acid (FA) to a final concentration of 1% (pH ~2). Peptides were desalted using macro spin columns (Nest group) according to the following procedure: cartridges were wetted with one volume (350 µl) 100% methanol, washed with two volumes of 80% acetonitrile and 0.1% FA, and equilibrated with three volumes of 0.1% FA. The acidified peptides were loaded twice on the cartridge, washed with three volumes of 0.1% FA and eluted with two volumes of 50% acetonitrile and 0.1% FA. Peptides were dried in a speedvac concentrator and resolubilized in 10 µl of 0.1% FA. The samples were then transferred to an mass spectroscopy (MS) vial and spiked with 1:20 (v:v) of iRT peptides (Escher et al., 2012).

Mass spectrometry data acquisition

Request a detailed protocol

1 µg of peptides were injected on a 5600 TripleTof mass spectrometer (ABSciex, Concord, Ontario, Canada) interfaced with an Eksigent NanoLC Ultra 1D Plus system (Eksigent, Dublin, CA, USA). The peptides were separated on a 75-µm-diameter, 20-cm long New Objective emitter packed with Magic C18 AQ 3 µm resin (Michrom BioResources) and eluted at 300 nl/min with a linear gradient of 5–35% Buffer A for 120 min (Buffer A: 2% acetonitrile, 0.1% FA; Buffer B: 98% acetonitrile, 0.1% FA). MS data acquisition was performed in either data-dependent acquisition (DDA, top20, with 20 s dynamic exclusion) or data-independent acquisition (DIA) SWATH-MS mode (32 fixed precursor isolation windows of 25 Da width [+1 Da overlap] each acquired for 100 ms plus one MS1 scan acquired for 250 ms) as described in Gillet et al. (2012). The mass ranges recorded were 360–1460 m/z for MS1 and 50–2000 m/z for MS2. For either mode, the collision energy was set to 0.0625 × m/z – 6.5 with a 15 eV collision energy spread regardless of the precursor charge state.

SWATH-MS assay library generation

Request a detailed protocol

The DDA data recorded as described above were used to generate an assay library essentially as described (Schubert et al., 2015). In short, the raw DDA files were converted to mzXML using the qtofpeakpicker component of msconvert (ProteoWizard v 3.0.9987). The converted files were searched with Comet (2014.02 rev. 0) and Mascot (version 2.5.1) using the yeast SGD database (release 13.01.2015), which contained 6713 proteins plus one protein entry for the concatenated sequence of the iRT peptides and as many decoy protein entries generated by pseudo-reversing the tryptic peptide sequences. The search parameters were as follows: + /– 25 ppm tolerance for MS1 and MS2, fixed cysteine carbamidomethylation, variable methionine oxidation, semi-tryptic and up to two missed cleavages per peptide allowed. The Comet and Mascot search results were further processed using peptideProphet (Keller et al., 2002) and aggregated using iProphet (Shteynberg et al., 2011) (TPP v4.7 rev 0). The search results were filtered for an iProphet cutoff of 0.877603, corresponding to a 1% protein false discovery rate (FDR) estimated by MAYU (Reiter et al., 2009). The search results contained 22,961 peptides matching a set of 2610 proteins. The consensus spectral library was generated using SpectraST (Lam et al., 2008), and the assay library thereof was exported using the spectrast2tsv.py script (Schubert et al., 2015) with the following parameters: six highest intensity fragments (of charge 1+ or 2+) per peptide, within the mass range 350–2000 m/z and excluding the fragments within the precursor isolation window of the corresponding swath. The final library contains assays for 23,960 peptide precursors (including 21,472 proteotypic precursors covering 2209 unique proteins). The assay library was exported to a TraML format with shuffled decoys appended as described (Schubert et al., 2015).

SWATH-MS data analysis

Request a detailed protocol

The SWATH-MS data were extracted with the above-mentioned assay library through the iPortal interface with openSWATH (Röst et al., 2014) (openMS 2.1.0), pyProphet (Teleman et al., 2015) and TRIC alignment (Röst et al., 2016) using the parameters described in Navarro et al. (2016). The SWATH identification results were further filtered to keep all the peptide assays with m-score below 0.01 for the protein entries with at least one peptide with an m-score below 0.00000208508 (corresponding to a protein FDR of 1%). A set of ribosomal proteins (RPL28, RPL14B, RPL14A, RPL3, RPL19A/RPL19B, RPL35A/RPL35B, RPL12A/RPL12B, RPL27A/RPL27B, RPL2B/RPL2A, RPL14A/RPL14B, RPL7B/RPL7A, and RPL1A/RPL1B) was used to normalize (mean-center) the data as described in Altvater et al. (2012). Only the proteotypic peptides were then kept, as well as the assays identified in the three triplicates for at least one AP-MS condition. Finally, proteins with fewer than two peptides were filtered out. The missing values for each peptide assay were imputed using a random value between 0.7- and 0.9-fold the lowest intensity of that peptide assay throughout the dataset. All of the peptide assay values were then summed to a protein intensity value and a protein intensity mean for each AP-MS condition. For the general protein heatmaps, each protein intensity was normalized to that of the highest protein value for the reported set. For the bar plots, the fold changes were calculated for each protein relative to that of the AP-MS condition with the highest protein intensity, while the standard deviations were calculated for each condition.

References

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
  16. 16
  17. 17
  18. 18
  19. 19
  20. 20
  21. 21
  22. 22
  23. 23
  24. 24
  25. 25
  26. 26
  27. 27
  28. 28
  29. 29
  30. 30
  31. 31
  32. 32
  33. 33
  34. 34
  35. 35
  36. 36
  37. 37
  38. 38
  39. 39
  40. 40
  41. 41
  42. 42
  43. 43
  44. 44
  45. 45
  46. 46
  47. 47
  48. 48
  49. 49
  50. 50
  51. 51
  52. 52
  53. 53
  54. 54
  55. 55
  56. 56
  57. 57
  58. 58
  59. 59
  60. 60
  61. 61
  62. 62
  63. 63
  64. 64
  65. 65
  66. 66
  67. 67

Decision letter

  1. Robin E Stanley
    Reviewing Editor; National Institutes of Health, United States
  2. James L Manley
    Senior Editor; Columbia University, United States
  3. Arlen Johnson
    Reviewer; University of Texas, Austin, United States
  4. Alan John Warren
    Reviewer; Cambridge Institute for Medical Research, United Kingdom

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The production of ribosomes relies on a fundamental and intricate assembly platform that is dependent on numerous assembly factors. This manuscript finely details the role of the GTPase Nog1 in late cytoplasmic stages of maturation of the large ribosomal subunit. The authors utilized SWATH-MS to delineate the temporal association of assembly factors with pre-60S particles. The mass spectrometry results compliment recent cryo-EM structures of pre-60S particles but also reveal insight into several assembly factors not yet visualized on pre-60S particles. Through additional genetic and cell-based assays the authors conclude that Nog1 directs both pre-60S maturation and quality control in the cytoplasm. This work lays the foundation for defining the molecular mechanisms of interplay between the various energy-consuming assembly factors required for ribosome production.

Decision letter after peer review:

[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The first decision letter after peer review is shown below.]

Thank you for submitting your work entitled "The GTPase Nog1 couples polypeptide exit tunnel quality control with ribosomal stalk assembly" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Arlen Johnson (Reviewer #2).

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

While there was much in the work that all three reviewers found commendable, two of them were adamant that it was imperative that additional experiments be conducted to truncate the Nog1-CTT and determine whether the outcomes support the predictions of your coupling model, of preventing Rei1 probing of PET integrity and blocking Arx1 release. However, because the Nog1-CTT is dispensable, there is a strong possibility that the CTT truncations would not have the predicted effects, in which event, a major conclusion of the work (described in the Abstract) would be unsupported. It was also felt that you should examine whether or not the MRT4-G68E mutant allows stalk assembly, and also probe the importance of the Rlp24-Nog1 interface; although neither of these deficiencies was judged to be as serious as the lack of information about the consequences of truncating the Nog1-CTT. In the event you choose to perform additional requested experiments, and also revise the paper extensively to address other criticisms, an effort would be made to have the revised manuscript, representing a new submission, reviewed by the same three reviewers who examined the current version of the work.

Reviewer:

This paper uses a combination of mass spectrometry (MS), genetics, molecular and cell biology to provide evidence regarding the biochemical functions of the ATPase Drg1 and GTPase Nog1 in late steps of 60S ribosome biogenesis. They employ dominant-negative (DN) alleles of the respective genes and examine the consequences of inactivating each protein on the association of biogenesis factors with different affinity-purified pre-60S complexes, or nucleolar vs. cytoplasmic localization of shuttling biogenesis factors. In this way they show that the NOG1-DN product, altered in the G-domain to impair GTP hydrolysis, accumulates on late pre-60Ss found in the cytoplasm harboring Lsg1 and is depleted from early nucleolar complexes containing Ssf1. They present results using SWATH-S MS analysis providing quantitative estimates of the occupancies of various biogenesis factors in the Ssf1, Lsg1, and also Rix1 and Arx1 containing complexes, purified using TAP-tagged forms of each protein. Applying the same analysis to the Lsg1 complex in WT vs. DRG1-DN or NOG1-DN cells leads them to conclude that ATP hydrolysis by Drg1 is required for its own dissociation and also that of Rlp24, Bud20, Nog1, Mrt4, Nmd3 and Tif6 from the Lsg1-bound pre-60S particles, and that GTP hydrolysis by Nog1 is required for release of Mrt14, Nmd3, and Tif6 in addition to itself, but not for release of Drg1, Rlp24 and Bud20. In addition, association of Yvh1 with this complex requires the activities of both Drg1 and Nog1. These conclusions for NOG1-DN were supported by imaging analysis of GFP-tagged versions of Rpl24, Bud20, Nug1, Mrt4, Nmd3 (a variant lacking its NES) and Tif6, leading them to conclude that Nog1 GTPase is not required to initiate cytoplasmic maturation promoted by Drg1 (release of Rlp24/Bud20) but is required for subsequent steps dependent on Yvh1 of Mrt4 release and assembly of the 60S stalk and the terminal steps of Nmd3 and Tif6 release. Other results indicate that Nog1 GTPase activity is not required for maturation of the peptide exit tunnel (PET), involving insertion of the Rei1 CTT into the tunnel and release of Aix1

1) This is a difficult paper to read, partly because of the complexity of the bifurcating processing pathways, but also because the authors don't always do a good job of reminding the reader of the rationale behind experiments or what results of individual experiments indicate about the pathway. They also sometimes selectively cite results described in the figures and on a few occasions describe results in misleading ways. More care should be taken to fully explain the rationale of each experiment, describe all of the relevant results, and draw an interim conclusion before going on to the next question. It could also be helpful to cite the final model in Figure 7 in the Results section as they go along, integrating each result in the final model rather than waiting until the Discussion. There are also numerous grammatical mistakes.

2) It seems important to confirm the MS data for DRG1-DN by Western analysis of the same set of proteins analyzed in Figure 3C, plus Nog1, to support key postulates of the final model.

3) It's also unclear why the authors have not imaged the panel of GFP-tagged proteins analyzed in Figure 4 for the NOG1-DN in the strain expressing the DRG1-DN allele, to confirm that localization of Bud20, Nug1, and Arx1 would be altered in addition to that of Mrt4, Nmd3 and Tif6 by inactivating Drg1. Some of this is accomplished later with the orthogonal approach of inhibiting Drg1 with diazaborine, and perhaps some of it was published previously, and if so, should be cited at the appropriate place in the Results.

4) There is a noteworthy discrepancy between the MS and Western data for Arx1 that should be discussed/resolved. Shouldn't both DN alleles lead to accumulation of Arx1 in the Lsg1-bound intermediate, whereas none is evident in the MS data of Figure 3B?

Reviewer #2:

This manuscript proposes that the GTPase Nog1, involved in ribosome assembly, couples quality control of the polypeptide exit tunnel (PET) with assembly of the P0 stalk. Nog1 is an unusual protein in that its N-terminus disrupts the ribosomal A site while its C-terminus wraps around the pre-60S particle and inserts into the PET. The authors use a dominant negative mutation in Nog1 that likely blocks GTP hydrolysis, although this is not shown. The major conclusion of this work is that the dominant negative mutant cannot be released from the ribosome and blocks the recruitment of Yvh1, preventing assembly of the P0 stalk but does not block events at the polypeptide exit tunnel. The data are consistent with the model in which the AAA-ATPase Drg1 initiates cytoplasmic maturation of the pre-60S by removing Rlp24, and probably the CTT of Nog1 at the same time. This is followed by release of Nog1 from the A site, providing linkage between events at the PET and P0 stalk. However, such coupling is not rigorously established. In addition, the function of Nog1 "probing" the exit tunnel seems dispensable because GFP fusion and C-terminal truncations of Nog1, which would preclude probing, are functional. This leaves open the question of the functional significance of Nog1 insertion into the PET. Lastly, the insertion of Nog1 into the A site disrupts the PTC, the catalytic center of the ribosome. The authors should also suggest that Nog1 couples probing of the PET with completion of the catalytic center.

1) The authors propose that Nog1 couples events probing the PET with stalk assembly. Additional experiments should be done to better establish this coupling. The C-terminus of Nog1 is dispensable. The authors should test if nog1∆CTT uncouples these events. Does nog1∆CTT allow stalk assembly in the presence of DRG1-DN?

2) "pre-60S particle requires successful completion of PET maturation/quality control and ribosomal stalk assembly" True but it also requires completion of the PTC! That's the all-important catalytic center.

3) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”. The authors show that Mrt4G68E localizes properly in the presence of Nog1DN but does not rescue the inviability. But does it allow assembly of the P0 stalk?

Reviewer #3:

The manuscript by Klingauf-Merurkar et al. nicely clarifies the details of several steps of cytoplasmic maturation of the large ribosomal subunit. Ribosome assembly is a very complicated pathway that is dependent upon numerous energy consuming enzymes such as the AAA-ATPase Drg1 and the GTPase Nog1. The essential GTPase Nog1 binds to the pre-60S in a very unusual way with its C-terminus inserting into the PET. Nog1 also intertwines with Rlp24, the substrate for Drg1. To study the roles of Nog1 and Drg1 Klingauf-Merurkar et al. used dominant negative mutants of both enzymes, previously established by others. The authors isolated well characterized pre-60S particles by TAP purification methods and carried out SWATH-S mass spectrometry. This mass spectrometry approach agrees with earlier studies but allows for quantification and the identification of ribosome assembly factors not characterized in recent cryo-EM structures. SWATH-S mass spectrometry of pre-60S particles isolated with the Drg1 and Nog1 dominant negative mutants led to the observation that the Nog1 mutant did not accumulate Rlp24 and Bud20, which both accumulate with the Drg1 mutant. The authors confirm their mass spectrometry results with TAP-purifications coupled to western blots and localization studies. They conclude that the activities of Nog1 and Drg1 are stepwise. ATP hydrolysis by Drg1 releases Rlp24 and Bud20 from pre-60S particles, as well as extracts the Nog1 CTT from the PET. Next GTP hydrolysis by Nog1 stimulates its removal from the pre-60S allowing later acting assembly factors to bind.

1) The authors conclude that ATP hydrolysis by Drg1 releases the Nog1 CTT from the PET but they provide no direct evidence for this. This is inferred from accumulation of later stage assembly factors only with the Nog1-DM mutant. The authors should test this hypothesis by carrying out experiments with a series of Nog1- CTT truncations.

2) The authors suggest that Rlp24 might serve as an adapter for Drg1 to extract the Nog1 CTT from the PET but there is no evidence to suggest that Nog1 is a direct substrate of Drg1. Previous studies have shown that the C-terminus of Rlp24 stimulates Drg1 ATP hydrolysis, strongly suggesting that Rlp24 is the substrate for remodeling by Drg1 not Nog1. Based on the cryo-EM structure by Wu et al. it seems a more likely scenario that remodeling/unfolding of Rlp24 by Drg1 leads to the removal of the Nog1 CTT from the PET. The authors should further address this by probing the Rlp24-Nog1 interface to disrupt the known binding interfaces in both proteins.

3) There is no mention in the manuscript of Ribosomal Protein L24. Rlp24 acts as a placeholder for L24 on pre-60S particles. If Rpl24 is released by Drg1 prior to release of Nog1, it seems possible that L24 could associate with Nog1 prior to its release. Do the authors detect L24 in their mass spectrometry analysis?

4) A novel finding in the manuscript is the identification of previously uncharacterized assembly factors such as Tma16 that accumulate with the Drg1 and Nog1 mutants. The authors should further explore the function of Tma16 and its links to the activities of Drg1 and Nog1.

[Editors’ note: what now follows is the decision letter after the authors submitted for further consideration.]

Thank you for submitting your article "The GTPase Nog1 co-ordinates assembly, maturation and quality control of distant ribosomal functional centers" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by James Manley as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Arlen Johnson (Reviewer #2); Alan John Warren (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This is an interesting study that combines SWATH-mass spectrometry with genetic and cell biological assays to study the role of the ATPase Drg1 and the GTPase Nog1 in late-stage cytoplasmic steps of 60S ribosomal subunit maturation. All three reviewers commended the authors for addressing the major concerns of the initial submission with the inclusion of key additional experiments. However, all three reviewers feel that revisions are needed to the figures and text to make the manuscript accessible to a general audience. The text requires a number of updates/corrections to take account of recently published work and citations need to be added appropriately within the text. Reviewers also raised concerns about the interpretation of some of the data. Based on our joint discussion we invite you to submit a suitably revised version of your manuscript in which you address the points raised below.

Essential revisions:

1) Data/Results Interpretation:

1a) This manuscript is missing any sort of quantification of the immunofluorescence data, which is present in 5 figures. According to the Materials and methods section each experiment was performed on three different occasions and in triplicate with a sample size of >1000 cells. The authors should quantify the number of cells with nuclear or cytoplasmic accumulation and present this data in a supplementary table.

1b) In Figure 3 the authors compare the SWATH-MS results with western blot analysis only for Nog1-DN. Antibodies may not be available for all the proteins shown by mass spectrometry but it's unclear why the authors do not show the mass spectrometry data for the ribosomal proteins that they analyze by western blot (uL10, eL24, and uL29). Comparing panels B and C would also be much easier if the western blots were shown in the same order as the mass spectrometry.

1c) Did the authors identify eL40 in the SWATH-MS data as this ribosomal protein is clearly critical for late cytoplasmic 60S maturation (Fernández-Pevida et al., 2012). eL40 should also be included in the "revised model for cytoplasmic maturation of the 60S ribosomal subunit".

1d) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: the authors reference Weis et al., 2015 to support "tightly coupled co-release of Nmd3 and Tif6". It would be more accurate to reference work from the Johnson lab in this regard (Lo et al., 2011, Malyutin et al., 2017; Patchett et al., 2017). We suggest that current cryo-EM data instead support the sequential removal of Nmd3 and Tif6 (Weis et al., 2015; Ma et al., 2017; Kargas et al., 2019). Please amend the text (and Figure 8, which is ambiguous in terms of Nmd3/Tif6 removal).

1e) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: "co-release of Nmd3 and Tif6…" This statement is not consistent with recent cryo-EM data (Kargas et al) as discussed above. Please update/discuss.

1f) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: the authors reference Weis et al., 2015 to support "tightly coupled co-release of Nmd3 and Tif6". It would be more accurate to reference work from the Johnson lab in this regard (Lo et al., 2011, Malyutin et al., 2017; Patchett et al., 2017). We suggest that current cryo-EM data instead support the sequential removal of Nmd3 and Tif6 (Weis et al., 2015; Ma et al., 2017; Kargas et al., 2019). Please amend the text (and Figure 8, which is ambiguous in terms of Nmd3/Tif6 removal).

1g) "Given the tightly coupled release of Nmd3 and Tif6, as expected…". Why is it necessary to invoke "tight coupling" here?

1h) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: A key problem caused by the retention of Nog1DN is that it prevents rearrangement of H89, providing a steric barrier to the recruitment of ribosomal protein uL16. The uL16 protein reorientates rRNA helices H89 and H38 to promote conformational rearrangement of the Nmd3 C-terminus, not the N-terminus (Kargas et al., 2019). The Nmd3 N-terminus remains fixed to Tif6 throughout the late steps of maturation while the C-terminal domain undergoes re-arrangement following the reorientation of H38 (Kargas et al., 2019). Please update this section.

2) The Discussion is long and somewhat repetitive with the Results. Please add appropriate subheadings to make this easier to follow. The Discussion must address the recent findings highlighted below and the proposed "revised model" should be clarified either to include all the other known cytoplasmic assembly factors or to explain clearly why specific factors have been excluded.

2a) In Figure 1C, the Western blot indicates that wild type Nog1 binds to the Lsg1-TAP particle, at least when Nog1 is over-expressed from the GAL promoter. Figure 2 also suggests that at least a subset of Lsg1-TAP particles bind Nog1, while the genetic experiments in Figure 3B clearly show that Nog1 accumulates on the Lsg1-TAP particle either when Drg1 function is impaired or in the presence of the Nog1DN mutation. These data suggest that Nog1 and Lsg1 likely bind simultaneously to a subset of native cytoplasmic pre-60S particles. This point should be discussed and if appropriate, Figure 8 updated.

2b) Using cryo-EM, Zhou et al., 2019, identify an early cytoplasmic pre-60S particle in which Bud20 is bound, the Nog1 CTT is in the PET, but the N-terminal and G-domains of Nog1 have apparently been displaced from the ribosome. Please discuss in the text how the SWATH-MS and genetic data presented in the current manuscript are consistent with the cryo-EM maps by Zhou et al?

2c) The SWATH-MS data in Figures 2 and 3 are consistent with recent cryo-EM data (Ma et al., 2017; Kargas et al., 2019) that reveal sequential recruitment of Rei1 and Reh1 into the PET. However, Reh1 is not included in the "revised model" in Figure 8. Please discuss/amend.

2d) In Figure 3B, the left hand panel- consistent with the cryo-EM structures from Kargas et al., 2019, another possible interpretation of the SWATH-MS data might be that like Tif6 and Nmd3, Reh1 is simply not recycled from late pre-60S particles in the Nog1DN or Drg1-deficient cells. It therefore seems unnecessary to invoke the recruitment of Reh1 and Rei1 through distinct binding sites. Please discuss. Perhaps it would be more accurate for the authors to indicate explicitly that they are proposing a revised model for the earlier steps of cytoplasmic pre-60S maturation.

2e) Recent work (Su et al., 2019) has identified the structure and function of Vms1 and Arb1 in RQC and mitochondrial proteome homeostasis. Please comment how the findings on Arb1 distribution in pre-60S particles in the current manuscript might relate to the identified function of Arb1 in RQC and update the text accordingly. It is inaccurate to state "structural information in the context of the ribosome is lacking".

2f) The Discussion should be updated to take account of recent cryo-EM data (Kargas et al., 2019) which reveals that in fact the N-terminus of Nmd3 remains fixed to Tif6 during the later steps of 60S maturation-it is the C-terminal domain that undergoes conformational rearrangement upon uL16 binding and H38 rearrangement.

2g) Figure 8A: Given that this is a "revised model for cytoplasmic maturation of the 60S ribosomal subunit", Lsg1, Reh1 and eL40 should also be included. In the figure legend, the statement "Nog1 triggers its own release together with Nsa2" does not appear to have been tested experimentally in the paper. Also, "we suggest that co-release of Nog1 and Nsa2…" Please provide evidence to support these statements or modify the text.

2h) Figure 8B: The typos in the legend should be corrected (H89, not H89) and the meaning of the text clarified. What rearrangement of Nmd3 is being referred to here? In the central panel, why is Rei1-CTT and not Reh1-CTT indicated? uL16 is recruited after the exchange of Rei1 for Reh1 (Kargas et al., 2019). Is this figure really necessary? It does not seem to directly relate to the work described in the manuscript and is not particularly informative.

https://doi.org/10.7554/eLife.52474.sa1

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

Reviewer #1:

[…] 1) This is a difficult paper to read, partly because of the complexity of the bifurcating processing pathways, but also because the authors don't always do a good job of reminding the reader of the rationale behind experiments or what results of individual experiments indicate about the pathway. They also sometimes selectively cite results described in the figures and on a few occasions describe results in misleading ways. More care should be taken to fully explain the rationale of each experiment, describe all of the relevant results, and draw an interim conclusion before going on to the next question. It could also be helpful to cite the final model in Figure 7 in the Results section as they go along, integrating each result in the final model rather than waiting until the Discussion. There are also numerous grammatical mistakes.

We thank the reviewer for his/her constructive comments. In the revised version of the manuscript, we now better explain how different domains of the GTPase Nog1 regulate PET maturation, PTC formation and stalk assembly.

2) It seems important to confirm the MS data for DRG1-DN by Western analysis of the same set of proteins analyzed in Figure 3C, plus Nog1, to support key postulates of the final model.

Western analyses validating accumulation of specific assembly factors, including Nog1, on the Lsg1-TAP particle in DRG1-DN expressing cells has been reported by several labs (Pertschy et al., 2007; Lo et al., 2010), including ours (Altvater et al., 2012).

3) It's also unclear why the authors have not imaged the panel of GFP-tagged proteins analyzed in Figure 4 for the NOG1-DN in the strain expressing the DRG1-DN allele, to confirm that localization of Bud20, Nug1, and Arx1 would be altered in addition to that of Mrt4, Nmd3 and Tif6 by inactivating Drg1. Some of this is accomplished later with the orthogonal approach of inhibiting Drg1 with diazaborine, and perhaps some of it was published previously, and if so, should be cited at the appropriate place in the Results.

Several laboratories, including ours, have reported mislocalization of Bud20, Nug1 and Arx1 in DRG1-DN expressing cells (or diazaborine treatment) into the cytoplasm (Pertschy et al., 2007; Lo et al., 2010; Altvater et al., Loibl et al., 2014). We cite the relevant references at appropriate places.

Here, we have localized Bud20, Nug1 and Arx1 in NOG1-DN expressing cells treated with diazaborine (Figure 5D). As expected, Bud20, Nug1 and Arx1 mislocalize into the cytoplasm only after treating NOG1-DN expressing cells with diazaborine.

4) There is a noteworthy discrepancy between the MS and Western data for Arx1 that should be discussed/resolved. Shouldn't both DN alleles lead to accumulation of Arx1 in the Lsg1-bound intermediate, whereas none is evident in the MS data of Figure 3B?

Arx1 bound to the pre-60S, mislocalises to the cytoplasm in DRG1-DN expressing cells (Altvater et al., 2012.; Lo et al., 2010). This is because recruitment of the Arx1-release factor, Rei1, to the pre-60S in the cytoplasm is impaired in DRG1-DN expressing cells (Altvater et al., 2012; Lo et al., 2010). Cryo-EM studies show that the Rei1-binding site on the pre-60S clashes with the Rlp24:Nog1-CTT complex (Figure 5C), providing an explanation as to why Drg1-mediated Rlp24 release is critical to recruit Rei1 for PET maturation (Arx1 release).

In contrast, recruitment of Rei1 to the pre-60S is unaffected in NOG1-DN expressing cells (Figure 3B, 3C and Figure 5A). Moreover, Arx1 release from the pre-60S is also not impaired (Figure 4C). This means that Rei1-CTT has successfully probed the PET in these cells. Failure to insert the Rei1-CTT into PET by attaching a TAP tag blocks Arx1 release, and mislocalizes Arx1 to the cytoplasm (Greber et al., 2016; Figure 5D). This mislocalisation is also observed in the rei1-TAP cells expressing NOG1-DN (Figure 5D) supporting the idea that Rei1 probes PET integrity in NOG1-DN expressing cells, and that Nog1-DN-CTT has to be removed from the PET.

Arx1 MS discrepancy: We have observed a similar discrepancy in Arx1 quantification in previous SRM-MS analyses (Altvater et al., 2012). The reasons for this discrepancy remain unclear. To resolve this, we have employed Arx1-GFP location as a reporter for PET maturation.

Reviewer #2:

This manuscript proposes that the GTPase Nog1, involved in ribosome assembly, couples quality control of the polypeptide exit tunnel (PET) with assembly of the P0 stalk. Nog1 is an unusual protein in that its N-terminus disrupts the ribosomal A site while its C-terminus wraps around the pre-60S particle and inserts into the PET. The authors use a dominant negative mutation in Nog1 that likely blocks GTP hydrolysis, although this is not shown. The major conclusion of this work is that the dominant negative mutant cannot be released from the ribosome and blocks the recruitment of Yvh1, preventing assembly of the P0 stalk but does not block events at the polypeptide exit tunnel. The data are consistent with the model in which the AAA-ATPase Drg1 initiates cytoplasmic maturation of the pre-60S by removing Rlp24, and probably the CTT of Nog1 at the same time. This is followed by release of Nog1 from the A site, providing linkage between events at the PET and P0 stalk. However, such coupling is not rigorously established. In addition, the function of Nog1 "probing" the exit tunnel seems dispensable because GFP fusion and C-terminal truncations of Nog1, which would preclude probing, are functional. This leaves open the question of the functional significance of Nog1 insertion into the PET. Lastly, the insertion of Nog1 into the A site disrupts the PTC, the catalytic center of the ribosome. The authors should also suggest that Nog1 couples probing of the PET with completion of the catalytic center.

1) The authors propose that Nog1 couples events probing the PET with stalk assembly. Additional experiments should be done to better establish this coupling. The C-terminus of Nog1 is dispensable. The authors should test if nog1∆CTT uncouples these events. Does nog1∆CTT allow stalk assembly in the presence of DRG1-DN?

In addition to the N-terminal domain (NTD) and a G-domain, Nog1 contains a long Cterminal tail (CTT). The Nog1-CTT intertwines around the ribosomal like protein Rlp24, then makes contacts with Arx1, and finally inserts its terminal end into the PET (Figure 1A). We have analysed the phenotypes of a series of Nog1-CTT truncation mutants. Removal of the part of the CTT (Nog11-536 and Nog11-479) that is inserted into the PET is not lethal, but renders a cold sensitive phenotype (Figure 6B). Likewise, attachment of a large tag GFP tag to Nog1CTT is not lethal, but renders a growth defect, stronger than the CTT truncations (Figure 6B). Sucrose density gradient analyses of lysates derived from the Nog1-GFP strain show the presence of halfmers that characterise 40S and 60S subunit imbalance (Author response image 1A). Removal of the entire Nog1-CTT that includes the region that intertwines with Rlp24 (Nog11-426) is lethal in yeast (Figure 6B).

Author response image 1
Functional analyses of yeast mutants expressing Nog1-GFP and Nog1-FLAG.

(A) Polysome profile analysis of Nog1-GFP. Wild-type and Nog1-GFP strains were grown in YPD at 30°C to mid-log-phase. Cell extracts were prepared after cycloheximide treatment and subjected to sedimentation centrifugation on 7-50% sucrose density gradients (Altvater et al., 2012). Halfmers are indicated with asterisks. (B) Left panel: Work flow depicting sequential ProteinA (Lsg1-TAP) and FLAG purification (Nog1DN-FLAG) from the indicated strains. Middle panel: Nog1DN-FLAG tail remains accessible for purification. Lsg1-TAP strain was transformed with either vector or Nog1DNFLAG under GAL1 promoter. Subsequently, either vector or with Drg1DN under GAL1 promoter were transformed. The resultant strains were grown in raffinose-containing synthetic medium and induced with 2% galactose for 3 hours. After sequential ProteinA and FLAG purifications, the eluates were analyzed by Western blotting using indicated antibodies. Right panel: Nog1-FLAG strain is growth impaired at 20 and 25 degrees. A nog1Δ shuffle strain was transformed with plasmids encoding either Nog1 or Nog1-FLAG plasmids. After shuffling out the URA3-NOG1 plasmid on FOA-containing plates, the resultant strains were spotted in 10-fold serial dilutions on YPD plates and incubated for 2-4 days at the indicated temperatures.

We apologize to the reviewer for not clearly explaining how different domains of Nog1 coordinate PET maturation and stalk assembly. PET maturation is impaired in DRG1-DN expressing cells. This is because the Rei1-binding site on the pre-60S clashes with the Rlp24:Nog1-CTT complex (Figure 5C). Drg1-driven Rlp24 release from the pre-60S (consequently Nog1-CTT) is critical for Rei1 recruitment (Lo et al., 2010; Altvater et al., 2012). In addition, Rei1 also needs to insert its CTT end into the PET to probe its integrity, to permit Arx1 release. For this to occur, Nog1-CTT has to be removed from the PET. Given that an impaired Nog1-G domain does not interfere with PET maturation implies that Rei1 has inserted its CTT into the PET. NOG1-DN expression in rei1-TAP strain impairs PET maturation supporting the notion that Rei1-CTT probes the PET in these cells, again implying Nog1-CTT is out (Figure 5D).

As requested, we investigated PET maturation (Arx1-GFP) after expressing DRG1-DN in a Nog1-CTT truncation mutant, Nog11-479, that still retains its Rlp24 binding site (Figure 6D). In light of cryo-EM data (Figure 5C), as expected, Arx1-GFP mislocalises to the cytoplasm in the Nog1-truncation mutant expressing DRG1-DN (Figure 6D). The Yvh1 binding site on the pre-60S clashes with the Nog1-G domain, providing an explanation as to why NOG1-DN expressing cells are impaired in stalk assembly (Figure 4D).

It is the negative regulation by the Nog1-CTT:Rlp24 complex in PET maturation (Rei1 recruitment) and stalk assembly (Yvh1 recruitment) that led to the proposal that Nog1 couples assembly and quality control of distant ribosomal functional centres.

2) "pre-60S particle requires successful completion of PET maturation/quality control and ribosomal stalk assembly" True but it also requires completion of the PTC! That's the all-important catalytic center.

We thank the reviewer for pointing this out. We now stated this in the Abstract and incorporated this important point in the Discussion.

3) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”. The authors show that Mrt4G68E localizes properly in the presence of Nog1DN but does not rescue the inviability. But does it allow assembly of the P0 stalk?

We have performed Western analyses to investigate this point. The ribosomal stalk protein P0 (uL10) is recruited to the 60S pre-ribosome upon expression of Nog1-DN in the MRT4G68E strain as judged by Western analyses (Figure 7C). Yes, it does allow assembly of the P0 stalk.

Reviewer #3:

[…] 1) The authors conclude that ATP hydrolysis by Drg1 releases the Nog1 CTT from the PET but they provide no direct evidence for this. This is inferred from accumulation of later stage assembly factors only with the Nog1-DM mutant. The authors should test this hypothesis by carrying out experiments with a series of Nog1- CTT truncations.

We apologize to the reviewer for not clearly explaining how Nog1 regulates PET maturation and stalk assembly.

In addition to the N-terminal domain (NTD) and a G-domain, Nog1 exhibits an extensive C-terminal tail (CTT). The CTT: (1) intertwines around the ribosomal like protein Rlp24 (2) contacts the assembly factor Arx1, before (3) inserting its terminal end into the PET (Figure 1A). We constructed Nog1-CTT truncation mutants analysed their phenotypes in yeast (Figure 6B). Removal of the part of the CTT that is inserted into the PET (Nog11-536 and Nog11479) is not lethal, but renders a cold sensitive phenotype (Figure 6B). Likewise, attachment of a large tag GFP tag is not lethal, but renders a growth defect even stronger than the Nog1CTT truncation mutants (Figure 6B). Further truncation of the Nog1-CTT that includes the region that intertwines with Rlp24 (Nog11-426) is lethal (Figure 6A, 6B).

PET maturation is impaired in DRG1-DN expressing cells (Pertschy et al., 2007; Lo et al., 2010; Altvater et al., 2012). This is because Rei1 recruitment to the pre-60S is impaired in these cells (Lo et al., 2010; Altvater et al., 2012). Cryo-EM studies indicate that the Rei1 binding site on the pre-60S clashes with the Rlp24:Nog1-CTT complex (Figure 5C) providing an explanation as to why Drg1-driven Rlp24 release from the pre-60S (and consequently Nog1-CTT) is critical for Rei1 recruitment.

It is also critical for Rei1 to insert its CTT into the PET, to permit Arx1 release and complete PET maturation (Greber et al., 2016). This can only occur if Nog1-CTT has been extracted from the PET. Based on two different observations that: (1) an impaired Nog1G domain does not interfere with PET maturation and (2) NOG1-DN expression only in the rei1-TAP mutant (where Rei1 is unable to insert its tail into tunnel) impairs PET maturation (Figure 5D), we propose that Rei1 has inserted its CTT into the PET, and therefore infer that NOG1-DN-CTT has been removed from the PET.

We investigated PET maturation (Arx1-GFP) in the Nog11-479 truncation mutant expressing DRG1-DN. The Nog11-479 truncation mutant retains the region that intertwines with Rlp24. In agreement with cryo-EM studies (Figure 5C), Arx1-GFP mislocalises to the cytoplasm in Nog11-479 cells treated with DIA (Figure 6D) suggesting that removal of Rlp24 from the pre-60S (and consequently Nog1-CTT) is critical for Rei1 recruitment.

The Yvh1 binding site clashes with the Nog1-G domain, thus providing structural basis as to why NOG1-DN expressing cells are impaired in stalk assembly (Figure 4D). It is the negative regulation of PET maturation and stalk assembly through the Nog1-CTT:Rlp24 complex and Nog1-G domain, respectively, that led to the proposal that Nog1 couples PET maturation and stalk assembly.

Our inference that Drg1-ATPase activity releases Nog1-CTT terminal end from the PET is based on a combination of biochemical and cell-biological observations: DRG1-DN blocks PET maturation by not recruiting the Arx1-release factor Rei1. Although NOG1-DN remains on the cytoplasmic pre-60S, it does not block Rei1 recruitment, as well as insertion of Rei1-CTT into the PET. Consequently, Arx1 release and recycling is not impaired. If NOG1DN-CTT had remained inserted inside the PET, then it would have blocked Arx1-release. As the reviewer correctly points out the likely explanation is that: Drg1 mediated removal of Rlp24 from the Nog1-CTT:Rlp24 complex leads to the removal of Nog1-CTT from the PET, thus permitting Rei1 recruitment and PET probing for tunnel maturation.

By employing tandem Lsg1-TAP followed by Nog1DN-CTT-FLAG purifications (work flow in Author response image 1B). we have attempted to directly demonstrate that the Nog1DN-CTT-FLAG is not inserted into the tunnel. Unfortunately, Nog1-FLAG fusion construct is not fully functional and renders a cold sensitive phenotype to yeast in a manner very similar to the one seen for the viable Nog1-CTT truncations, suggesting that the FLAGtagged tail might not be inserted into the PET. Consistent with this, we found that Nog1DNCTT-FLAG tail is accessible for affinity purification, that is it is not inserted in the PET upon Drg1DN expression (Lane 3). Note that the assembly factor Bud20 that interacts with Rlp24 coenriches with this particle, showing that Drg1DN expression blocked initiation of the cytoplasmic maturation pathway in these cells.

2) The authors suggest that Rlp24 might serve as an adapter for Drg1 to extract the Nog1 CTT from the PET but there is no evidence to suggest that Nog1 is a direct substrate of Drg1. Previous studies have shown that the C-terminus of Rlp24 stimulates Drg1 ATP hydrolysis, strongly suggesting that Rlp24 is the substrate for remodeling by Drg1 not Nog1. Based on the cryo-EM structure by Wu et al. it seems a more likely scenario that remodeling/unfolding of Rlp24 by Drg1 leads to the removal of the Nog1 CTT from the PET. The authors should further address this by probing the Rlp24-Nog1 interface to disrupt the known binding interfaces in both proteins.

Nog1 is not a direct substrate of Drg1; if this were the case it would have been completely removed from the pre-60S, without the need for the GTPase activity. Rlp24 directly binds to Drg1, stimulates its ATPase activity for release from the pre-60S. Given that parts of the Nog1-CTT intertwine around Rlp24 on the pre-60S, the reviewer is correct to point out that the extraction of Rlp24 by Drg1 is likely to lead Nog1-CTT removal from the PET.

The Drg1 “adaptor” suggestion put forth in the Discussion, stems from the model AAAATPase, Cdc48/p97. This AAA-ATPase employs different adaptor proteins to remodel different protein complexes and impact diverse cellular processes. Although only Rlp24 is a direct substrate of Drg1, we proposed that it serves as an adaptor through interactions with Nog1-CTT to indirectly remove the CTT-tail end from the PET.

Removal of the Nog1-CTT region that binds the Rlp24 (Nog11-426 mutant) is lethal in yeast. Overexpression of a Nog1 construct lacking the Rlp24 binding region, Nog1∆427-536, is dominant negative, but unlike NOG1-DN, it does not impair the 60S cytoplasmic maturation pathway (Figure 6B, 6C). Given that Nog1 and Rlp24 rely on each for their recruitment to the pre-60S (Saveanu et al., 2003), the toxicity of the Nog1∆427-536 very likely reflects a failure to form a functional Nog1-CTT:Rlp24 complex on the pre-60S during early nucleolar/nuclear maturation.

3) There is no mention in the manuscript of Ribosomal Protein L24. Rlp24 acts as a placeholder for L24 on pre-60S particles. If Rpl24 is released by Drg1 prior to release of Nog1, it seems possible that L24 could associate with Nog1 prior to its release. Do the authors detect L24 in their mass spectrometry analysis?

The release and recycling of Rlp24 from the 60S pre-ribosome is not impaired in NOG1-DN expressing cells. As requested, we have analysed recruitment of eL24 to the pre60S in NOG1-DN expressing cells by Western blotting. eL24 is recruited to the pre-60S in these cells (Figure 3C).

4) A novel finding in the manuscript is the identification of previously uncharacterized assembly factors such as Tma16 that accumulate with the Drg1 and Nog1 mutants. The authors should further explore the function of Tma16 and its links to the activities of Drg1 and Nog1.

By employing Western analyses, we confirmed the SWATH-MS data that Tma16 accumulates on cytoplasmic 60S pre-ribosomes in NOG1-DN expressing cells (Figure 3C). To investigate the precise location of Tma16 on the pre-60S and its functional contribution of to the cytoplasmic maturation pathway is an important issue. In our opinion, Tma16 analyses go beyond the main message of this study that sequential Drg1-ATPase and Nog1-GTPase activities license PET maturation, PTC formation and ribosomal stalk assembly.

[Editors' note: the author responses to the re-review follow.]

Summary:

This is an interesting study that combines SWATH-mass spectrometry with genetic and cell biological assays to study the role of the ATPase Drg1 and the GTPase Nog1 in late-stage cytoplasmic steps of 60S ribosomal subunit maturation. All three reviewers commended the authors for addressing the major concerns of the initial submission with the inclusion of key additional experiments. However, all three reviewers feel that revisions are needed to the figures and text to make the manuscript accessible to a general audience. The text requires a number of updates/corrections to take account of recently published work and citations need to be added appropriately within the text. Reviewers also raised concerns about the interpretation of some of the data. Based on our joint discussion we invite you to submit a suitably revised version of your manuscript in which you address the points raised below.

We thank the reviewers for their helpful comments, and especially pointing out incorrect statements regarding the 60S cytoplasmic maturation pathway.

Essential revisions:

1) Data/Results Interpretation:

1a) This manuscript is missing any sort of quantification of the immunofluorescence data, which is present in 5 figures. According to the Materials and methods section each experiment was performed on three different occasions and in triplicate with a sample size of >1000 cells. The authors should quantify the number of cells with nuclear or cytoplasmic accumulation and present this data in a supplementary table.

Each cell-biological experiment was performed at least three times, and >90% of the cells show the reported phenotypes. The images were quantified as described previously in Lo et al., 2010: the average of the fraction of cells that show stronger nuclear fluorescence as compared the cytoplasm was quantified. We list the quantification for every image in Supplementary file 3.

1b) In Figure 3 the authors compare the SWATH-MS results with western blot analysis only for Nog1-DN. Antibodies may not be available for all the proteins shown by mass spectrometry but it's unclear why the authors do not show the mass spectrometry data for the ribosomal proteins that they analyze by western blot (uL10, eL24, and uL29). Comparing panels B and C would also be much easier if the western blots were shown in the same order as the mass spectrometry.

1c) Did the authors identify eL40 in the SWATH-MS data as this ribosomal protein is clearly critical for late cytoplasmic 60S maturation (Fernández-Pevida et al., 2012). eL40 should also be included in the "revised model for cytoplasmic maturation of the 60S ribosomal subunit".

Nearly all ribosomal proteins are small, Lys/Arg-rich proteins. During sample preparation ribosomal proteins are broken down by trypsin into rather short peptides, as compared to larger assembly factors. In general, detecting and quantitating these tryptic peptides by mass spectrometry is challenging. Ribosomal proteins eL3, eL14 and u15 can be reliably quantified by SWATH-MS, and hence we have used them as “loading controls”. Further, we have validated the proteomic data by Western blotting and cell-biological approaches.

We have reorganized panels B and C to better compare the Western analyses and SWATH-MS data.

We were unable to reliably detect eL40 by SWATH-MS in WT Lsg1-TAP. As requested, we have included eL40 and uL16 in model for 60S maturation pathway in the text (subsection “Nog1 co-ordinates assembly, maturation and quality control of distant ribosomal centres”) and Figure 8.

1d) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: the authors reference Weis et al., 2015 to support "tightly coupled co-release of Nmd3 and Tif6". It would be more accurate to reference work from the Johnson lab in this regard (Lo et al., 2011, Malyutin et al., 2017; Patchett et al., 2017). We suggest that current cryo-EM data instead support the sequential removal of Nmd3 and Tif6 (Weis et al., 2015; Ma et al., 2017; Kargas et al., 2019). Please amend the text (and Figure 8, which is ambiguous in terms of Nmd3/Tif6 removal).

1e) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: "co-release of Nmd3 and Tif6…" This statement is not consistent with recent cryo-EM data (Kargas et al) as discussed above. Please update/discuss.

1f) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: the authors reference Weis et al., 2015 to support "tightly coupled co-release of Nmd3 and Tif6". It would be more accurate to reference work from the Johnson lab in this regard (Lo et al., 2011, Malyutin et al., 2017; Patchett et al., 2017). We suggest that current cryo-EM data instead support the sequential removal of Nmd3 and Tif6 (Weis et al., 2015; Ma et al., 2017; Kargas et al., 2019). Please amend the text (and Figure 8, which is ambiguous in terms of Nmd3/Tif6 removal).

1g) "Given the tightly coupled release of Nmd3 and Tif6, as expected…". Why is it necessary to invoke "tight coupling" here?

We thank the reviewers for pointing out our misconception regarding Nmd3 and Tif6 release from the pre-60S. Recent cryo-EM work (Kargas et al., 2019) indeed supports the model that Nmd3 and Tif6 are sequentially released. We have removed/amended incorrect statements and updated the model (Discussion) and in Figure 8.

1h) Subsection “Nog1DN blocks the terminal cytoplasmic maturation steps”: A key problem caused by the retention of Nog1DN is that it prevents rearrangement of H89, providing a steric barrier to the recruitment of ribosomal protein uL16. The uL16 protein reorientates rRNA helices H89 and H38 to promote conformational rearrangement of the Nmd3 C-terminus, not the N-terminus (Kargas et al., 2019). The Nmd3 N-terminus remains fixed to Tif6 throughout the late steps of maturation while the C-terminal domain undergoes re-arrangement following the reorientation of H38 (Kargas et al., 2019). Please update this section.

We thank the reviewers for pointing out our misconception regarding the order of uL16 incorporation, rearrangements in rRNA and Nmd3 after Nog1 release. We have corrected this in the text (Discussion).

2) The Discussion is long and somewhat repetitive with the Results. Please add appropriate subheadings to make this easier to follow. The Discussion must address the recent findings highlighted below and the proposed "revised model" should be clarified either to include all the other known cytoplasmic assembly factors or to explain clearly why specific factors have been excluded.

We have removed the repetitive sections, and shortened the Discussion. We now provide subheadings that highlight our key findings. The Discussion is focused on how different domains of Nog1 gate PET maturation/quality control, PTC maturation, and late cytoplasmic maturation steps (Discussion).

2a) In Figure 1C, the Western blot indicates that wild type Nog1 binds to the Lsg1-TAP particle, at least when Nog1 is over-expressed from the GAL promoter. Figure 2 also suggests that at least a subset of Lsg1-TAP particles bind Nog1, while the genetic experiments in Figure 3B clearly show that Nog1 accumulates on the Lsg1-TAP particle either when Drg1 function is impaired or in the presence of the Nog1DN mutation. These data suggest that Nog1 and Lsg1 likely bind simultaneously to a subset of native cytoplasmic pre-60S particles. This point should be discussed and if appropriate, Figure 8 updated.

In comparison to WT Nog1, Nog1DN accumulates on the Lsg1-TAP particle (Figure 1C, right panel). Further, WT Nog1 accumulates on the Lsg1-TAP particle in Drg1DN expressing cells (Altvater et al., 2012). These data indicate that Lsg1 and Nog1 can bind simultaneously to the pre-60S in the cytoplasm. This is now discussed (subsection “Nog1 co-ordinates assembly, maturation and quality control of distant ribosomal centres”), and Figure 8 is updated accordingly.

2b) Using cryo-EM, Zhou et al., 2019, identify an early cytoplasmic pre-60S particle in which Bud20 is bound, the Nog1 CTT is in the PET, but the N-terminal and G-domains of Nog1 have apparently been displaced from the ribosome. Please discuss in the text how the SWATH-MS and genetic data presented in the current manuscript are consistent with the cryo-EM maps by Zhou et al?

We thank the reviewers for bringing this point to our notice. In the reported early pre-60S particle (Zhou et al., 2019) the terminal end of the Nog1-CTT is in the PET, and the N-terminal and G-domains are displaced from the pre-ribosome. However, Nog1 remains attached to the pre-60S via Nog1-CTT that intertwines around Rlp24. These data are consistent with the idea that Nog1 release from the pre-60S requires both Drg1-ATPase activity (through Rlp24 release) and a functional Nog1 G-domain (presumably GTPase activity). This is discussed in the text (subsection “Nog1 eviction requires a functional G-domain and Drg1-ATPase activity”).

2c) The SWATH-MS data in Figures 2 and 3 are consistent with recent cryo-EM data (Ma et al., 2017; Kargas et al., 2019) that reveal sequential recruitment of Rei1 and Reh1 into the PET. However, Reh1 is not included in the "revised model" in Figure 8. Please discuss/amend.

2d) In Figure 3B, the left hand panel- consistent with the cryo-EM structures from Kargas et al., 2019, another possible interpretation of the SWATH-MS data might be that like Tif6 and Nmd3, Reh1 is simply not recycled from late pre-60S particles in the Nog1DN or Drg1-deficient cells. It therefore seems unnecessary to invoke the recruitment of Reh1 and Rei1 through distinct binding sites. Please discuss. Perhaps it would be more accurate for the authors to indicate explicitly that they are proposing a revised model for the earlier steps of cytoplasmic pre-60S maturation.

The reviewers are correct that the SWATH-MS data support the model that Rei1 and Reh1 are sequentially recruited to the cytoplasmic pre-60S, and that Reh1 recycling is impaired in Nog1DN expressing cells. This is now discussed in the text (subsection “The Nog1-CTT:Rlp24 complex gates PET quality control”) and included in Figure 8. We now explicitly state/discuss how our work and recent cryo-EM studies permit refinement of early steps that drive cytoplasmic pre-60S maturation (subsection “Nog1 co-ordinates assembly, maturation and quality control of distant ribosomal centres”).

2e) Recent work (Su et al., 2019) has identified the structure and function of Vms1 and Arb1 in RQC and mitochondrial proteome homeostasis. Please comment how the findings on Arb1 distribution in pre-60S particles in the current manuscript might relate to the identified function of Arb1 in RQC and update the text accordingly. It is inaccurate to state "structural information in the context of the ribosome is lacking".

We politely disagree with this comment: We had stated that “structural information in the context of the pre-ribosome is lacking". This statement is accurate since we do not know where Sqt1, Ecm1, Mex67, Tma16 and Arb1 are located on a pre-60S. For clarity, we have now changed “pre-ribosome” to pre-60S.

Previously, the Hinnebusch lab had shown that Arb1 contributes to the assembly of both 40S and 60S subunit (Dong et al., 2005). A recent cryo-EM structure of a translationally stalled-mature 60S subunit from the Neupert/Beckmann labs perhaps provides a clue of its location on a pre-60S. In this structure, Arb1 docks in the E-site region of 25S rRNA, which is available only when Nmd3 is released from the pre-60S. This is consistent with SWATH-MS data that Arb1 recruitment to the Lsg1-TAP:Nog1DN particle is impaired. Based on all these data we speculate that Arb1 is incorporated into the pre-60S only after stable uL16 recruitment and Nmd3 release. We have incorporated these comments into the Discussion.

Author response image 2
Arb1 location on a mature 60S clashes with Nmd3 binding site on a pre-60S.

Cryo-EM structures of Nmd3-containing (PDB-6RZZ, Kargas et al., 2019) pre-60S and Arb1-containing (PDB-6R84, Su et al., 2019) mature 60S subunit were superimposed. Surprisingly, Arb1 binds to a region of rRNA, which overlaps with the binding site of Nmd3 on the pre-60S. We speculate that Nmd3 release allows Arb1 recruitment to the pre-60S during final cytoplasmic maturation.

2f) The Discussion should be updated to take account of recent cryo-EM data (Kargas et al., 2019) which reveals that in fact the N-terminus of Nmd3 remains fixed to Tif6 during the later steps of 60S maturation-it is the C-terminal domain that undergoes conformational rearrangement upon uL16 binding and H38 rearrangement.

We thank the reviewers for correcting our misconception(s) regarding uL16 incorporation, rearrangements in H89 rRNA and Nmd3 that drive cytoplasmic pre-60S maturation. The text has been altered accordingly (subsection “Nog1 gates stalk assembly and PTC maturation”).

2g) Figure 8A: Given that this is a "revised model for cytoplasmic maturation of the 60S ribosomal subunit", Lsg1, Reh1 and eL40 should also be included. In the figure legend, the statement "Nog1 triggers its own release together with Nsa2" does not appear to have been tested experimentally in the paper. Also, "we suggest that co-release of Nog1 and Nsa2…" Please provide evidence to support these statements or modify the text.

We have updated Figure 8A by including Lsg1, Reh1, eL40 and uL16.

Nsa2 accumulates on Lsg1-TAP in Drg1DN as well as Nog1DN expressing cells (Altvater et al., 2012; Figure 3C). These data are consistent with the idea that Nog1 eviction from the pre-60S precedes Nsa2 release. We have modified the text accordingly (subsection “Nog1 co-ordinates assembly, maturation and quality control of distant ribosomal centres”).

2h) Figure 8B: The typos in the legend should be corrected (H89, not H89) and the meaning of the text clarified. What rearrangement of Nmd3 is being referred to here? In the central panel, why is Rei1-CTT and not Reh1-CTT indicated? uL16 is recruited after the exchange of Rei1 for Reh1 (Kargas et al., 2019). Is this figure really necessary? It does not seem to directly relate to the work described in the manuscript and is not particularly informative.

We agree with the reviewers that Figure 8B does not provide insights into how Nog1DN impairs early cytoplasmic maturation of the pre-60S. We therefore have removed Figure 8B from the manuscript, and updated Figure 8, as requested, by including Lsg1, Reh1, eL40 and uL16.

https://doi.org/10.7554/eLife.52474.sa2

Article and author information

Author details

  1. Purnima Klingauf-Nerurkar

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Formal analysis, Investigation, Methodology
    Contributed equally with
    Ludovic C Gillet
    Competing interests
    No competing interests declared
  2. Ludovic C Gillet

    Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Data curation, Formal analysis, Methodology
    Contributed equally with
    Purnima Klingauf-Nerurkar
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1001-3265
  3. Daniela Portugal-Calisto

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Formal analysis
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1591-5812
  4. Michaela Oborská-Oplová

    1. Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    2. Institute of Biochemistry, ETH Zurich, Zurich, Switzerland
    Contribution
    Formal analysis, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0976-4341
  5. Martin Jäger

    Institute of Biochemistry, ETH Zurich, Zurich, Switzerland
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared
  6. Olga T Schubert

    Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland
    Present address
    Department of Human Genetics, University of California, Los Angeles, Los Angeles, United States
    Contribution
    Conceptualization, Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2613-0714
  7. Agnese Pisano

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Formal analysis
    Competing interests
    No competing interests declared
  8. Cohue Peña

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Formal analysis, Methodology
    Competing interests
    No competing interests declared
  9. Sanjana Rao

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Formal analysis, Methodology
    Competing interests
    No competing interests declared
  10. Martin Altvater

    Institute of Biochemistry, ETH Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Formal analysis, Methodology
    Competing interests
    No competing interests declared
  11. Yiming Chang

    Institute of Biochemistry, ETH Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared
  12. Ruedi Aebersold

    1. Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    2. Institute of Molecular Systems Biology, ETH Zurich, Zurich, Switzerland
    Present address
    Faculty of Science, University of Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Resources, Supervision, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
  13. Vikram G Panse

    Institute of Medical Microbiology, University of Zurich, Zurich, Switzerland
    Contribution
    Conceptualization, Formal analysis, Supervision, Investigation, Project administration
    For correspondence
    vpanse@imm.uzh.ch
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7950-5746

Funding

Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung

  • Vikram G Panse

H2020 European Research Council (EURIBIO)

  • Vikram G Panse

Novartis StiftungfürMedizinisch-Biologische Forschung

  • Vikram G Panse

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank M Peter, C Dargemont, A Johnson, M Fromont-Racine, B Stillman, F Lacroute, and J Warner for generously sharing yeast strains and antibodies. This work is dedicated to the memory of Dr. Cohue Peña, who unexpectedly passed away. We thank all members of the Panse laboratory for enthusiastic discussions. We thank the services of the Center for Microscopy and Image Analysis, University of Zurich for providing and maintaining the imaging equipment.

Senior Editor

  1. James L Manley, Columbia University, United States

Reviewing Editor

  1. Robin E Stanley, National Institutes of Health, United States

Reviewers

  1. Arlen Johnson, University of Texas, Austin, United States
  2. Alan John Warren, Cambridge Institute for Medical Research, United Kingdom

Publication history

  1. Received: October 5, 2019
  2. Accepted: December 20, 2019
  3. Accepted Manuscript published: January 7, 2020 (version 1)
  4. Version of Record published: January 17, 2020 (version 2)

Copyright

© 2020, Klingauf-Nerurkar et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,280
    Page views
  • 184
    Downloads
  • 4
    Citations

Article citation count generated by polling the highest count across the following sources: PubMed Central, Crossref, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Biochemistry and Chemical Biology
    2. Cell Biology
    Philip T McGilvray et al.
    Research Article Updated

    Membrane proteins with multiple transmembrane domains play critical roles in cell physiology, but little is known about the machinery coordinating their biogenesis at the endoplasmic reticulum. Here we describe a ~ 360 kDa ribosome-associated complex comprising the core Sec61 channel and five accessory factors: TMCO1, CCDC47 and the Nicalin-TMEM147-NOMO complex. Cryo-electron microscopy reveals a large assembly at the ribosome exit tunnel organized around a central membrane cavity. Similar to protein-conducting channels that facilitate movement of transmembrane segments, cytosolic and luminal funnels in TMCO1 and TMEM147, respectively, suggest routes into the central membrane cavity. High-throughput mRNA sequencing shows selective translocon engagement with hundreds of different multi-pass membrane proteins. Consistent with a role in multi-pass membrane protein biogenesis, cells lacking different accessory components show reduced levels of one such client, the glutamate transporter EAAT1. These results identify a new human translocon and provide a molecular framework for understanding its role in multi-pass membrane protein biogenesis.

    1. Biochemistry and Chemical Biology
    2. Structural Biology and Molecular Biophysics
    Marie E Sweet et al.
    Research Article

    KdpFABC is an ATP-dependent K+ pump that ensures bacterial survival in K+-deficient environments. Whereas transcriptional activation of kdpFABC expression is well studied, a mechanism for down regulation when K+ levels are restored has not been described. Here we show that KdpFABC is inhibited when cells return to a K+-rich environment. The mechanism of inhibition involves phosphorylation of Ser162 on KdpB, which can be reversed in vitro by treatment with serine phosphatase. Mutating Ser162 to Alanine produces constitutive activity, whereas the phosphomimetic Ser162Asp mutation inactivates the pump. Analyses of the transport cycle show that serine phosphorylation abolishes the K+-dependence of ATP hydrolysis and blocks the catalytic cycle after formation of the aspartyl phosphate intermediate (E1~P). This regulatory mechanism is unique amongst P-type pumps and this study furthers our understanding of how bacteria control potassium homeostasis to maintain cell volume and osmotic potential.