ARTICLE DOI: 10.1038/s41467-018-04564-3 OPEN Vms1p is a release factor for the ribosome- associated quality control complex 1,2 2 3,4 2 1,2 Olga Zurita Rendón , Eric K. Fredrickson , Conor J. Howard , Jonathan Van Vranken , Sarah Fogarty , 5 2,3,4 3,6 2 2 Neal D. Tolley , Raghav Kalia , Beatriz A. Osuna , Peter S. Shen , Christopher P. Hill , 2,3,4,7 1,2 Adam Frost & Jared Rutter Eukaryotic cells employ the ribosome-associated quality control complex (RQC) to maintain homeostasis despite defects that cause ribosomes to stall. The RQC comprises the E3 ubi- quitin ligase Ltn1p, the ATPase Cdc48p, Rqc1p, and Rqc2p. Upon ribosome stalling and splitting, the RQC assembles on the 60S species containing unreleased peptidyl-tRNA (60S: peptidyl–tRNA). Ltn1p and Rqc1p facilitate ubiquitination of the incomplete nascent chain, marking it for degradation. Rqc2p stabilizes Ltn1p on the 60S and recruits charged tRNAs to the 60S to catalyze elongation of the nascent protein with carboxy-terminal alanine and threonine extensions (CAT tails). By mobilizing the nascent chain, CAT tailing can expose lysine residues that are hidden in the exit tunnel, thereby supporting efﬁcient ubiquitination. If the ubiquitin–proteasome system is overwhelmed or unavailable, CAT-tailed nascent chains can aggregate in the cytosol or within organelles like mitochondria. Here we identify Vms1p as a tRNA hydrolase that releases stalled polypeptides engaged by the RQC. 1 2 Howard Hughes Medical Institute, Chevy Chase, MD 20815-6789, USA. Department of Biochemistry, University of Utah School of Medicine, Salt Lake 3 4 City, UT 84112, USA. Department of Biochemistry and Biophysics, University of California, San Francisco, San Francisco, CA 94158, USA. California Institute for Quantitative Biomedical Research, San Francisco, CA 94158, USA. Molecular Medicine Program, University of Utah, Salt Lake City, UT 84112, 6 7 USA. Department of Microbiology and Immunology, University of California, San Francisco, San Francisco, CA 94158, USA. Chan Zuckerberg Biohub, San Francisco, CA 94158, USA. These authors contributed equally: Olga Zurita Rendón, Eric K. Fredrickson, Conor J. Howard. Correspondence and requests for materials should be addressed to A.F. (email: firstname.lastname@example.org) or to J.R. (email: email@example.com) NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 efects that impair the decoding of a messenger RNA was sufﬁcient to reverse the lethality of the vms1Δ mutant in 7,27 (mRNA) can cause ribosomes to stall and distort gene CHX (Fig. 1a). By contrast, deletion of the no-go decay Dexpression. To prevent the accumulation of potentially components DOM34 or SKI7 had no effect. These data suggest toxic aberrant polypeptides, and to maintain ribosome avail- that CHX causes accumulation of an RQC product that becomes ability, eukaryotic cells employ surveillance and clearance toxic in the absence of Vms1p. mechanisms, including the evolutionarily conserved ribosome- The RQC assembles on a failed 60S subunit to both ubiquitinate associated quality control complex (RQC). The RQC is composed and elongate nascent polypeptides with CAT tails. To determine of the E3 ubiquitin ligase Ltn1p, the ATPase Cdc48p and its whether one or both of these activities generate the putative toxic ubiquitin-binding cofactors, and the poorly characterized pro- RQC product, we eliminated them separately by expressing a D98Y 11,14,21 teins Rqc1p and Rqc2p—whose homologs are Listerin, VCP/p97, CAT-tailing-defective Rqc2p mutant (RQC2 ) or a 1–3 W1542E 17,28 TCF25, and NEMF, respectively, in mammals . variant of Ltn1p deﬁcient in ubiquitination (LTN1 ) . Upon recognition of certain types of stalls, the 60S and 40S CAT tailing by Rqc2p was dispensable, while ubiquitination by ribosomal subunits are split by the eRF1 homolog Dom34p, the Ltn1p was essential for conferring CHX sensitivity on a vms1Δ 4–10 GTPase Hbs1p, and the ATPase Rli1p . The RQC then mutant strain—indicating that the toxic species in this context recognizes and assembles upon the 60S ribosomal subunit that depends on ubiquitination rather than CAT-tailing (Fig. 1b). contains an incomplete polypeptide linked to a transfer RNA Presumably because of its proposed role in nascent chain 11–14 17 (tRNA) molecule (60S:peptidyl–tRNA) . Ltn1p ubiquiti- ubiquitination , and similar to RQC2 and LTN1 (Fig. 1b), nates the nascent chain and marks it for proteasomal degra- plasmid expression of wild-type (WT) RQC1 was sufﬁcient to dation, a process facilitated by Rqc1p through unclear reverse the rescue of vms1Δ cycloheximide sensitivity conferred by 15–17 mechanisms . Rqc2p recognizes and binds to the exposed RQC1 deletion (Supplementary Fig. 1a). 11,12,14 tRNA and stabilizes Ltn1p on the 60S . Rqc2p also In light of the association between Vms1 activity and recruits charged tRNAs to the 60S subunit to catalyze elonga- mitochondrial stress, we examined the ability of these single tion of the nascent protein with Carboxy-terminal Alanine and and double mutant strains to grow in glycerol medium, which Threonine extensions (CAT tails) via a 40S-independent requires mitochondrial respiration. Interestingly, deletion of 11,17 mechanism that is distinct from canonical translation . LTN1—but not RQC1, RQC2,or DOM34—strongly impaired This non-templated elongation of the nascent chain can glycerol growth of vms1Δ cells (Supplementary Fig. 1b). Similarly, mobilize potential ubiquitination sites, including lysine residues ski7Δ vms1Δ double mutant cells also exhibited partially impaired that arehiddeninthe ribosomeexittunnel, enhancingthe glycerol growth (Supplementary Fig. 1b). These data indicate a efﬁciency of ubiquitination and the capacity of the RQC to speciﬁc relationship between Vms1p, RQC function, and 17,18 protect cells from stochastic translation failures . mitochondrial homeostasis, which is consistent with a recent Nascent chain ubiquitination promotes extraction and pro- report that stalled polypeptides that cannot be ubiquitinated by teasomal degradation of the aberrant polypeptide in a Cdc48- Ltn1p accumulate within and compromise mitochondria— dependent process . If the ubiquitin–proteasome system is over- underscoring the reversal of the genetic interaction we observed whelmed or unavailable, CAT tailing enables nascent chains to between LTN1 and VMS1 on glucose vs. glycerol . form aggregates in the cytosol or within organelles like the These genetic interactions prompted us to determine whether 19–21,22 mitochondria . The critical unanswered question we Vms1p physically interacts with members of the RQC, as has 1,3,22 investigated here concerns the identity of the hydrolase that been reported previously . As expected, isolation of Rqc2p- releases ubiquitinated and CAT-tailed nascent chains from the HA co-immunoprecipitated Vms1p-V5. In contrast, Rqc1p-HA aberrant 60S:peptidyl–tRNA for extraction and degradation by or Ltn1p-HA showed minimal Vms1p interaction (Fig. 1c and the proteasome. Supplementary Fig. 1c-d). Consistently, both Rqc2p and Vms1p Like the canonical RQC components, Vms1p is conserved co-migrated with the 60S ribosome subunit during sucrose throughout Eukarya and promotes protein quality control in gradient sedimentation following CHX treatment (Fig. 1d, diverse settings through its interaction with Cdc48/p97. In Sac- Supplementary Fig. 1e). Co-migration of Vms1p with the 60S charomyces cerevisiae, Vms1p localizes to mitochondria in ribosome was not affected by either deletion of RQC2 or by 23,24 response to mitochondrial stress or damage . Mutants lacking expression of WT or a CAT-tailing defective, D98Y, mutant of Vms1p are sensitive to rapamycin, which impairs ribosomal Rqc2p (Supplementary Fig. 1e). Similarly, neither deletion nor 25,26 protein synthesis , although the mechanism for this sensitivity overexpression of Vms1p from the strong GAL1 promoter in is unknown. In this work, we show that Vms1p interacts galactose medium had any effect on the co-migration of Rqc2p genetically and physically with the RQC components, harbors an with the 60S ribosomal subunit (Supplementary Fig. 1e, f). eRF1 release factor-like domain, and is required for peptidyl- tRNA hydrolysis and clearance of RQC-engaged nascent chains. This role of Vms1p in promoting proteostasis and ribosome Vms1p is required for resolving RQC substrates. The genetic recycling by resolving 60S:peptidyl-tRNA species expands upon and physical interactions described above motivated us to eval- the previously characterized function of Vms1p as a Cdc48/p97 uate whether RQC substrates accumulate in VMS1 mutant cells. binding protein and raises interesting new questions regarding We utilized a well-characterized mRNA that encodes FLAG- the potential intersection of protein and organelle homeostasis. tagged green ﬂuorescent protein (GFP) followed by a hammer- Rz 6,11 head ribozyme that self-cleaves in vivo (FLAG-GFP , Fig. 2a) to generate a truncated mRNA encoding FLAG-GFP without a Results stop codon or poly-A tail. Translation of this mRNA triggers Vms1 physically and genetically interacts with the RQC. ribosome stalling and targeting to the RQC. Deletion of SKI7 Because vms1Δ cells are sensitive to rapamycin, we tested whether enhances expression of GFP by inhibiting degradation of the other perturbations to translation elicit a similar phenotype. We cleaved mRNA . We conﬁrmed that deletion of RQC1, RQC2, Rz found that the vms1Δ strain is sensitive to the protein synthesis and LTN1 each lead to accumulation of FLAG-GFP , whereas inhibitor cycloheximide (CHX), as are other RQC mutants in a the nascent chain failed to accumulate in the ski7Δ single mutant sensitizing SKI mutant background . Surprisingly, deletion of (Fig. 2a–c, Supplementary Fig. 2a). Loss of Vms1p also led to Rz any one of the RQC system components RQC1, RQC2,or LTN1, accumulation of FLAG-GFP to a level similar to that observed 2 NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 ARTICLE ab Glucose Glucose+CHX Glucose Glucose+CHX WT WT+EV vms1Δ vms1Δ+EV ski7Δ rqc2Δvms1Δ+EV ski7Δ vms1Δ rqc2Δvms1Δ+Rqc2p rqc1Δ rqc2Δvms1Δ+Rqc2p (D98Y) rqc1Δ vms1Δ ltn1Δ vms1Δ+ EV rqc1Δ ski7Δ ltn1Δ vms1Δ+ Ltn1p rqc2Δ ltn1Δ vms1Δ+ Ltn1p (W1542E) rqc2Δ vms1Δ rqc2Δ ski7Δ ltn1Δ ltn1Δ vms1Δ ltn1Δ ski7Δ dom34Δ dom34Δ vms1Δ dom34Δ ski7Δ IB: V5 (Vms1p) Gradient top 40S 60S 80S Polysomes IB: HA (Rqc2p) 150 IB: HA (Rqc2p) IB: V5 (Vms1p) 75 IB: V5 (Vms1p) 150 IB: HA (Rqc2p) IB: Rpl3p Fig. 1 Vms1 physically and genetically interacts with the RQC. a, b Serial dilutions of indicated strains were spotted on media containing glucose or glucose supplemented with cycloheximide (CHX). EV empty vector. c Immunoprecipitation using anti-HA antibody in the strains rqc2Δ vms1Δ expressing Rqc2p and Vms1p-V5 (control) or Rqc2p-HA and Vms1p-V5. Immunoblotting of HA and V5 were used to identify Rqc2p and Vms1p, respectively. d Polysome proﬁle of the rqc2Δ vms1Δ strain expressing Rqc2p-HA and Vms1p-V5 treated with CHX prior to fractionation using sucrose density centrifugation. The sedimentation of ribosomal particles was inferred from the A proﬁle (40S, 60S, 80S, and polysomes) and the distribution of the 60S subunit was conﬁrmed by immunoblotting of the ribosomal subunit, Rpl3p. Immunoblotting of HA and V5 was used to detect Rqc2p and Vms1p, respectively for the core RQC components (Fig. 2a–c, Supplementary Fig. 2b). pattern (Fig. 2d), consistent with the model that CAT tailing Combination of VMS1 deletion with the deletion of RQC1, RQC2, mediates intra-mitochondrial aggregation of polypeptides that and LTN1 had no additive effect on GFP accumulation (Fig. 2b). stall during co-translational import . Together, these data Immunoblot analysis showed similar results for the single, dou- demonstrate that Vms1p is required for the degradation of ble, and triple mutant strains, in which RQC2-dependent, high- substrates derived from truncated mRNAs, whether they are molecular-weight aggregates are also apparent (Fig. 2c, Supple- destined for a membranous organelle or the cytosol. 19–21 mentary Fig. 2b) . Loss of DOM34 led to decreased accu- mulation of GFP ﬂuorescence, even in the vms1Δ strain, consistent with the upstream role of Dom34p in ribosome split- Vms1p is structurally homologous to tRNA hydrolases. ting and suggestive of alternative pathways for degrading nascent Understanding of how Vms1p facilitates the clearance of stalled chains when the Dom34p/Hbs1p subunit splitting activity is translation products was guided by our recent crystal structure unavailable (Fig. 2a–c). Interestingly, accumulation of FLAG- determination of a portion of S. cerevisiae Vms1p (Fig. 3a, b) . Rz GFP in vms1Δ mutant cells occurs despite lower mRNA This structure includes the highly conserved central region of abundance (Supplementary Fig. 2c). Vms1p, which we named the mitochondrial targeting domain Rz In addition to the FLAG-GFP construct, which generates a (MTD) because it is necessary and sufﬁcient for mitochondrial cytosolic RQC substrate, we also examined RQC activity on localization . This localization activity requires a hydrophobic fumarase, which is encoded by the FUM1 gene and co- groove along the region of the MTD where the LRS interacts and 30 31 translationally imported into the mitochondria . As with direct binding to ergosterol peroxide . Intriguingly, the Vms1p Rz Rz FLAG-GFP , ﬂuorescence from the Fum1-FLAG-GFP con- MTD structure resembles structures of the catalytic domain of struct, expressed from the native FUM1 promoter, was also eukaryotic peptide chain release factor subunit 1 (eRF1), as well maintained at a low level in the ski7Δ mutant strain (Fig. 2d). as Dom34p and RNaseE, which both resemble tRNA hydro- 32–35 Deletion of VMS1, RQC1, RQC2,or LTN1 each led to profound lases (Fig. 3b, Supplementary Fig. 3a). The only region of the accumulation of GFP ﬂuorescence, almost all of which colocalized Vms1p MTD that diverges substantially from the release factor with mitochondria-targeted red ﬂuorescence protein (mtRFP). fold is the face of the MTD that mediates ergosterol peroxide The vms1Δ, rqc1Δ, and ltn1Δ mutants, which retain Rqc2p and binding and mitochondrial localization (Fig. 3b). CAT-tailing activity, all exhibited Fum1-GFP aggregates within or Sequence alignment showed that although Vms1p lacks a near mitochondria, comparable to recent observations of other strict GGQ motif characteristic of eRF1p, it does possess mitochondria-destined nascent chains (Fig. 2d). The rqc2Δ an invariant glutamine that can align with the catalytic glutamine mutant exhibited a more uniformly mitochondrial localization of eRF1 (Fig. 3c). In yeast Vms1p, this glutamine residue is NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications 3 | | | Rqc2, Vms1-V5 Rqc2-HA, Vms1-V5 IP:HA Elutes Inputs ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 a b 5′ PromGPD 3′ 3xFLAG GFP Rz FLAG-GFP **** c d 5′ PromFUM1 3′ FUM1 3xFLAG GFP Rz FLAG-GFP IB:Flag IB:Pgk1p Rz Fig. 2 Vms1p is required for resolving RQC substrates. a Fluorescence microscopy analysis of the indicated strains expressing the FLAG-GFP construct under the GPD promoter and the mitochondrial marker, mtRFP. b Flow cytometry quantiﬁcations of FLAG-GFP accumulation in the indicated strains. **** Median GFP intensity values are plotted (n = 3, mean ± s.e.m. P < 0.0001, The p value was calculated using unpaired Student’s t-test). c Immunoblot Rz analysis of indicated strains expressing the FLAG-GFP construct. Immunoblotting of Flag was used to detect the accumulation of the stalled construct. Rz Pgk1p was used as loading control. d Fluorescence microscopy analysis of the indicated strains expressing the Fum1-FLAG-GFP construct expressed from the FUM1 endogenous promoter and the mitochondrial marker, mtRFP embedded within a GGSQ motif that is reminiscent of the sensitivity (Fig. 3f). In contrast, mutation of the ‘GxxQ’ residues eRF1 catalytic GGQ, while in other species the conservation other G292 and Q295 and the highly conserved R288 residue conferred than the initial glycine and glutamine is less apparent (Fig. 3c, d). strong loss-of-function phenotypes (Fig. 3e). Deletion of S294 to The Vms1p MTD lacks similarity to the non-catalytic eRF1 convert the GGSQ of S. cerevisiae Vms1p into a GGQ motif, as in domain 1, which discriminates stop codons from sense codons . eRF1, also abrogated VMS1 function (Fig. 3e). While all of these −1 This is consistent with Vms1p functioning in stop codon- ‘GxxQ’‘mutants failed to confer resistance to 200 ng ml CHX, independent tRNA hydrolysis within a 60S, rather than 80S, only the R288A and G292A/G293A mutants were inactive at the −1 ribosome. lowest (100 ng ml ) concentration of CHX tested (Fig. 3e). These observations inspired us to determine whether Vms1p Interestingly, both of these mutants also failed to rescue glycerol enables the extraction of failed translation products from the growth in an ltn1Δ vms1Δ double mutant, whereas wild-type stalled 60S by hydrolyzing the ester bond anchoring them to VMS1 and the other mutants did rescue growth (Supplementary tRNA. We asked which residues and regions are required for the Fig. 3b). The R288A, G292A/G293A, and Q295L mutants also Rz genetic functions of VMS1.We ﬁrst tested an HbϕT motif just N exhibited enhanced accumulation of FLAG-GFP in the ski7Δ terminal to the conserved ‘GxxQ’ motif that mediates ribosome background similar to the vms1Δ mutant (Supplementary Fig. 3d). interactions of eRF1 (Fig. 3c, d). Vms1p mutants of these Importantly, the Vms1p mutants interact normally, if not more residues, H279A, H283A, R284A, and T286A, were indistinguish- strongly, with Rqc2p based on co-immunoprecipitation experi- able from WT, while the Y285A mutant exhibited a partial CHX ments (Supplementary Fig. 3e). In light of these observations, we 4 NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications | | | ski7Δ ski7Δvms1Δ ski7Δrqc2Δ ski7Δltn1Δ ski7Δdom34Δ ski7Δ ski7Δvms1Δ ski7Δrqc1Δ ski7Δrqc2Δ ski7Δltn1Δ ski7Δdom34Δ ski7Δrqc1Δ Merge mtRFP GFP DIC ski7Δ ski7Δvms1Δ ski7Δrqc1Δ ski7Δrqc1Δvms1Δ ski7Δrqc2Δ ski7Δrqc2Δvms1Δ ski7Δltn1Δ ski7Δltn1Δvms1Δ ski7Δdom34Δ ski7Δdom34Δvms1Δ Merge mtRFP Fum1-GFP DIC GFP median intensity ski7Δ ski7Δvms1Δ ski7Δrqc1Δ ski7Δrqc1Δvms1Δ ski7Δrqc2Δ ski7Δrqc2Δvms1Δ ski7Δltn1Δ ski7Δltn1Δvms1Δ ski7Δdom34Δ ski7Δdom34Δvms1Δ NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 ARTICLE 188 417 VIM 1 ZnF eRFL/MTD 632 LRS AnkR CC Vms1p eRFL eRF1p catalytic domain Overlay N C 170 180 190 200 eRF1p H. sapiens eRF1p S. cerevisiae Vms1p S. cerevisiae 280 290 300 310 Vms1p S. cerevisiae Vms1p S. pombe Vms1p C. elegans Vms1p D. discoideum Vms1p S. purpuratus Vms1p M. musculus Vms1p C. familiaris Vms1p P. troglodytes Vms1p H. sapiens Glucose+ Glucose+ Glucose+ Glucose+ Glucose CHX200 CHX100 Glucose CHX200 CHX100 +EV +EV +Vms1p-GFP +EV +Vms1p (R288A) +Vms1p-V5 +Vms1p (K290A) +Vms1p (H279A) +Vms1p (Q291L) +Vms1p (H283A) +Vms1p (R284A) +Vms1p (G292A) +Vms1p (Y285A) +Vms1p (G293A) +Vms1p (T286A) +Vms1p (G292/293A) +Vms1p (ΔS294) +Vms1p (Q295L) +Vms1p (D299A) Fig. 3 Vms1p is structurally homologous to tRNA hydrolases. a Domain structure of Vms1p. LRS leucine-rich sequence, ZnF zinc ﬁnger, MTD/eRFL mitochondrial targeting domain/eRF1-like, AnkR ankryin repeat, CC coil–coil, VIM VCP-interacting motif. Residues 188–417 represent the MTD/eRFL 31 35 boundaries. b Structural alignment of Vms1p (left, 5WHG ) and eRF1p (middle, 3JAHii , residues 144–280). Dashed lines indicate connections made by residues that are not resolved in the Vms1p crystal structure. The GGQ (red) loop of eRF1p is ordered in the ribosome-bound structure shown here. c Sequence alignment of Vms1p and eRF1p. White letters with gray, black, or red background indicates similarity, identity, or GxxQ residues, respectively. d Sequence alignment of Vms1p orthologs across the GxxQ region. Coloring as in c. e, f Serial dilutions of indicated strains were spotted on media containing glucose or glucose supplemented with cycloheximide (CHX). EV empty vector NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications 5 | | | vms1Δ vms1Δ WT **** **** ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 hereafter refer to the MTD as the MTD/eRFL domain, where chain and peptidyl-tRNA conjugates. We performed a similar eRFL refers to eRF1-like. experiment in the ski7Δ and ski7Δ vms1Δ mutant strains and In addition to these loop residues, the ability of Vms1p to found that in the ski7Δ background the deletion of VMS1 con- confer complete CHX resistance in both vms1Δ and ski7Δ vms1Δ ferred a much more obvious stabilization of the peptidyl-tRNA also required the p97/valosin-containing protein (VCP)-interact- species and qualitatively delayed release of the polypeptides ing motif (VIM), which mediates interaction with Cdc48p/VCP/ (Fig. 4c, d). In this ski7Δ background, deletion of RQC2 conferred p97 (Supplementary Fig. 3c). Interestingly, the VIM is not a modest stabilization of the peptidyl-tRNA conjugate and had required for growth of the ltn1Δ vms1Δ double mutant on little effect on the vms1Δ mutant (Supplementary Fig. 4a). We glycerol (Supplementary Fig. 3c), which indicates that mitochon- next puriﬁed full-length and C-terminally truncated (1–417) S. drial homeostasis can be maintained even without Cdc48p cerevisiae Vms1p and found that each of these proteins drama- binding. tically accelerated the production of the released polypeptide in a dose-dependent manner in WT, rqc2Δ and vms1Δ extracts (Fig. 4e, Supplementary Fig. 4b). Importantly, the 1–417 trun- Vms1p exhibits tRNA hydrolase activity towards RQC sub- cation mutant lacks the C-terminal VIM domain and is unable to strates. To directly test whether Vms1p catalyzes peptidyl-tRNA interact with Cdc48p (Supplementary Fig. 3e). We therefore hydrolysis, we utilized our recently described S. cerevisiae in vitro conclude that while the Vms1p–Cdc48p interaction is important translation (ScIVT) system to monitor the synthesis and fate of a for CHX resistance and other RQC-related functions that are robust stalling reporter and its peptidyl-tRNA intermediate . relevant to the VMS1 genetic interactions, Vms1p association RQC-intact extracts translate this reporter, split the stalled 80S with Cdc48p is dispensable for peptidyl-tRNA hydrolysis. ribosome into constituent 60S and 40S subunits, elongate the We next tested release factor activity of the MTD/eRFL domain nascent chain with a CAT tail, and ubiquitinate exposed lysine structure-based mutants described above. The Q295L, G292A, residues. These extracts also hydrolyze the peptidyl-tRNA ester and ΔS294 mutants also exhibited strongly impaired release bond to generate the released polypeptide (Fig. 4a). We factor activity (Fig. 4f and Supplementary Fig. 4c). Consistent observed that extracts prepared from vms1Δ mutant cells also with the stronger growth phenotypes we observed on different concentrations of CHX and on glycerol, R288A and G292A/ produced peptidyl-tRNA conjugates, but loss of the peptidyl- tRNA species and appearance of the released translation product G293A mutants had no hydrolysis activity, even at 10-fold higher were slower than in WT extracts (Fig. 4a, b). This is somewhat concentration than the concentration at which WT Vms1p obscured by the fact that the vms1Δ mutant has lower overall catalyzed complete tRNA release (Fig. 4f). translation, which leads to a decreased amount of the free nascent a c ski7Δ vms1Δski7Δ WT vms1Δ Time (min): Time (min) : 15 30 45 60 15 30 45 60 15 30 45 60 15 30 45 60 50 50 Peptidyl–tRNA Peptidyl–tRNA 30 30 25 Released peptide Released peptide b d 100 100 vms1Δski7Δ vms1Δ 25 25 WT ski7Δ 15 30 45 60 15 30 45 60 Time (min) Time (min) e f vms1Δ t =15 t =30 vms1Δ G292A t =15 t =30 WT R288A G293A Q295L FL 1–417 Vms1p: –– Vms1p: 1× 1/10 1× 1/10 1× 1/10 1× 1/10 – – 50 50 Peptidyl–tRNA Peptidyl–tRNA 30 30 Released peptide 25 Released peptide 25 Fig. 4 Vms1p exhibits tRNA hydrolase activity towards RQC substrates. a Time courses of S. cerevisiae in vitro translation (ScIVT) reactions prepared with a truncated mRNA (lacking a stop codon). Extract genotypes are indicated above. Peptides that have been CAT-tailed and released are denoted with a cat **** tail icon. b Quantiﬁcation of peptidyl-tRNA species in a. Mean ± s.e.m.; n = 6. P < 0.0001. The p value was calculated using a two-way ANOVA. c Time **** courses of ScIVT reactions prepared as in a. d Quantiﬁcation of peptidyl-tRNA species in c. Mean ± s.e.m., n = 8. P < 0.0001. The p value was calculated using a two-way ANOVA. e ScIVT reactions prepared as in a with a vms1Δ extract. At t = 15, buffer (−) or pure protein was added. Slopes indicate a titration series of decreasing protein concentrations (see Methods). FL full length Vms1, 1–417 N terminus through eRFL domain. f ScIVT reactions prepared as in a with a vms1Δ extract. At t = 15, buffer, WT (1–417) protein, or mutant (1–417) protein was added 6 NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications | | | Relative peptidyl–tRNA (%) Relative peptidyl–tRNA (%) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 ARTICLE Discussion sodium dodecyl sulfate (SDS), 0.02% bromophenol blue). Source data can be found in Supplementary Fig. 5a. Our data have identiﬁed a key constituent of the RQC pathway in Eukarya: a tRNA hydrolase that liberates failed polypeptides Polysome proﬁling. Yeast cultures were grown to OD ~1, cycloheximide was from the aberrant 60S:peptidyl-tRNA species that accumulate 600 −1 added to a ﬁnal concentration of 0.05 mg ml , and cells were harvested by cen- when ribosomes stall. Without this activity, translation products trifugation 5 min later. Cell pellets were washed in buffer A (20 mM Tris-HCl pH remain anchored in 60S ribosomes, which therefore cannot be −1 7.4, 50 mM KCl, 10 mM MgCl , 1 mM dithiothreitol (DTT), 100 μgml cyclo- recycled for future use. The dual functions of the Vms1p MTD/ heximide, 1 × RNAsecure (Ambion), and 1× yeast protease inhibitor (Sigma)). Pellets were weighed and resuspended in 1.3 volumes of Buffer A. An equal volume eRFL identiﬁed here as an RQC release factor and previously as a of glass beads was added and suspensions were vortexed for 30 s for a total of 8 targeting domain in mitochondrial stress responses portend times interspersed with 1 min of incubation on ice. Following centrifugation at exciting future work at the intersection of proteostasis and 3000 × g for 5 min, supernatant was centrifuged at 11,300 × g for 2 min at 4 °C, after organelle homeostasis. We have previously reported that Vms1p which supernatant was centrifuged at 11,300 × g for 10 min. Protein extracts were localizes to mitochondria under conditions of mitochondrial overlaid onto a linear sucrose gradient of 15–50% and centrifuged at 234,600 × g for 90 min. The gradients were passed through a continuous-ﬂow chamber and damage or cellular stressors, including rapamycin treatment, by monitored at 254 nm with an ultraviolet absorbance detector (ISCO UA-6) to 22,23,31 binding to the oxidized sterol ergosterol peroxide .The obtain ribosomal proﬁles. Fractions (16) were collected, resuspended in 2× MTD/eRFL domain is necessary and sufﬁcient for this localiza- Laemmli sample buffer supplemented with 2.5% beta-mercaptoethanol, and ana- tion, which is mediated by a direct interaction between a face of lyzed by western blotting. Source data can be found in Supplementary Fig. 5b. the MTD/eRFL domain that should remain exposed even when the domain is in the ‘A-site’ of the 60S and the catalytic GGSQ SDS-PAGE. Whole-cell extracts were prepared from 3 to 5 ODs of cells at loop is presumably reaching into the peptidyl transferase center OD ~1.5 by solubilization in 250 µl of 2 M LiAc, incubated for 8 min on ice to catalyze hydrolysis of the peptidyl-tRNA ester bond. While the followed by centrifugation at 0.9 × g for 5 min at 4 °C. The pellet was resuspended relationship between mitochondrial localization and RQC- in 250 µl of 0.4 M NaOH and incubated on ice for 8 min followed by centrifugation coupled release factor activity remains unclear, it is intriguing at 16,000 × g for 3 min. Next, the pellet was resuspended in 1× Laemmli buffer with 2.5% beta-mercaptoethanol, boiled for 5 min, and centrifuged at 0.9 × g for 1 min. to speculate that this is indicative of a role for Vms1p—and the Supernatants were collected and loaded onto acrylamide/bisacrylamide (37.5:1) RQC as a whole—in the response to mitochondrial stress. gels. Subsequent immunoblotting was done with the indicated antibodies: HA Consistent with this possibility, ltn1Δ vms1Δ and ski7Δ vms1Δ (PRB-101C-200), V5 (ab9116), FLAG (F7425), Pgk1: (ab113687), and Rpl3 double mutant cells exhibit impaired glycerol growth, which (scRPL3). Source data can be found in Supplementary Fig. 6. correlates with impaired mitochondrial respiration . The diseases associated with even subtle distortions in pro- Fluorescence microscopy. WT (BY4741) or derived mutant strains were trans- tein quality control underscore the importance of the activity of formed with a plasmid expressing mitochondria-targeted (ATPase subunit, Su9) RFP, mtRFP, and plasmids expressing Flag His -GFP-Rz or FUM1-Flag His - Vms1pasareleasefactorfor theRQC.Among many other 3 6 3 6 GFP-Rz under the GPD or native FUM1 promoter, respectively. The cells were examples, hypomorphic mutations in the RQC-associated ubi- grown to mid-log phase and imaged using the Axio Observer Z1 imaging system quitin E3 ligase, Listerin (Ltn1p in yeast), lead to profound (Carl Zeiss). Digital ﬂuorescence and differential interference contrast images were neurodegeneration in mice . As we show, Vms1p protects cells acquired using a monochrome digital camera (AxioCam MRm) and analyzed using the Zen 2 software (Carl Zeiss). from inadequate Ltn1p activity by releasing CAT tailed nascent chains from stalled 60S ribosomes. In so doing, Vms1p rescues Fluorescence-assisted cell sorting. GFP-expressing strains and untransformed these ribosome subunits for future use and defends proteostasis control were grown to OD ~1 and pelleted by centrifugation at 100 × g for 5 min. from translation products whose accumulation could otherwise Cell pellets were washed once in 1× phosphate-buffered saline (PBS) buffer, cause disease. We therefore propose that Vms1p is required for resuspended in 1 ml of 1× PBS, and analyzed using the BDFACSCanto Analyzer the resolution of peptidyl-tRNA conjugates of stalled nascent (488 laser and optical ﬁlter FITC). A total of 30,000 events were measured and the chains in the cytosol as well as those destined for organelles like median values of three independent biological replicates were analyzed by one-way analysis of variance (ANOVA; Bonferroni correction analysis) with a conﬁdence the mitochondria, where it might mediate a particular role in interval of <0.05 using the statistics software: Graphpad Prism 6. Additionally, protecting mitochondria from proteostasis challenges. ski7Δ, ski7Δvms1Δ, and strains in Supplementary Fig. 3 were compared via unpaired Student's t-test (two-tailed) conﬁdence interval value set to p < 0.05. Error bars represent standard error of the mean. Methods Yeast strains and growth conditions. S. cerevisiae BY4741 (MATa, his3 leu2 met15 ura3) was used as the wild-type strain. Each mutant was generated in diploid Protein expression and puriﬁcation. For the His tagged proteins, constructs cells using a standard PCR-based homologous recombination method. The geno- were transformed into JRY1734 (pep4::HIS3 prb1::LEU2 bar1::HISG lys2::GAL1/ types of all strains used in this study are listed in Supplementary Table 1. Yeast 10-GAL4) and grown in synthetic media lacking Uracil with 3% glycerol and 2% transformations were performed by the standard TE/LiAc method and transformed ethanol. When the OD reached ~0.5, 0.5% galactose was added to the cultures, cells were recovered and grown in synthetic complete glucose medium lacking the which were grown for another 6 h before harvesting by centrifugation, washing of appropriate amino acid(s) for selection. The medium used included YPA and the pellet with sterile H2O, and ﬂash freezing in liquid nitrogen. Cells were lysed synthetic minimal medium supplemented with 2% glucose or 3% glycerol. using a pulverizer (SPEX SamplePrep 6870), and the lysed powder was thoroughly −1 −1 Cycloheximide was added at a ﬁnal concentration of 100 ng ml or 200 ng ml resuspended in lysis buffer (20 mM Tris pH 8.0, 300 mM NaCl, 5% glycerol) when indicated. supplemented with protease inhibitors (aprotinin, leupeptin, pepstatin A, and All plasmid constructs were generated by PCR and cloned into the yeast PMSF) (Sigma). The resuspended lysate was clariﬁed by centrifugation and added expression vectors pRS413, pRS14 or pRS416 as indicated in Supplementary to Ni-NTA resin (Qiagen #30250) for 1 hour, washed with 10 CV of lysis buffer, 10 Table 2. CV of lysis buffer with 40 mM imidazole, and eluted with lysis buffer made up with Growth assays were performed using synthetic minimal media supplemented 250 mM imidazole. Eluted protein was dialyzed into IVT-compatible buffer (20 with the appropriate amino acids and indicated carbon source. For plate-based mM HEPES-KOH pH 7.4, 150 mM KOAc, 5% glycerol, 2 mM DTT) and growth assays, overnight cultures were back-diluted to equivalent optical densities concentrated. (ODs) and spotted as 10-fold serial dilutions. Cells were grown at 30 °C. S. cerevisiae in vitro translation . Preparation of in vitro translation extracts, VMS1 Immunoprecipitations.p -VMS1-V5 (or VMS1 mutant) was co-expressed mRNA, and in vitro translation reactions was performed as previously described with an endogenous promoter-His -HA tagged RQC component (RQC1, RQC2, . Brieﬂy, S. cerevisiae strains were cryo-lysed and cell debris was cleared by 6 2 LTN1) in the cognate double mutant strain. Approximately 50 ODs were harvested sequential centrifugation before dialysis into fresh lysis buffer. mRNAs were gen- in log phase and resuspended in IP buffer (20 mM Tris pH 7.4, 50 mM NaCl, 0.2% erated by run-off transcription from PCR-ampliﬁed templates of 3xHA- Triton X-100), vortexed 10 × 1 min, clariﬁed via centrifugation, and added to anti- NanoLuciferase to produce transcripts lacking a stop codon and 3ʹ-untranslated HA magnetic beads (Thermo Scientiﬁc #88836). After 4 h of incubation, beads region (truncated quality control substrate). Transcription products were capped were pelleted via magnet and washed 4× with 1 ml of IP buffer. Proteins were and extracted prior to freezing for use in ScIVTs. For ScIVT reactions, extracts eluted with 50 µl of 2× Laemmli buffer (20% glycerol, 125 mM Tris-HCl pH 6.8, 4% were ﬁrst treated with MNase to remove endogenous mRNAs and then NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications 7 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 supplemented with 480 ng mRNA to initiate translation. Reaction aliquots were 13. Shao, S. & Hegde, R. S. Reconstitution of a minimal ribosome-associated sampled at indicated time points by quenching in 2× Laemmli Sample Buffer. ubiquitination pathway with puriﬁed factors. Mol. Cell 55, 880–890 Proteins were separated by SDS-PAGE, and hemagglutinin (HA)-tagged transla- (2014). tion products were visualized by immunoblotting (Roche 3F10). To quantify 14. Lyumkis, D. et al. Structural basis for translational surveillance by the large release, the abundance of peptidyl-tRNA was measured with Fiji (https://imagej. ribosomal subunit-associated protein quality control complex. Proc. Natl. net/Fiji) and normalized as percentage of the initial 15 min time point. Mean values Acad. Sci. USA 111, 15981–15986 (2014). of at least 6 technical replicates were analyzed and plotted in Prism (GraphPad 15. Shao, S., von der Malsburg, K. & Hegde, R. S. Listerin-dependent nascent software). P-values were calculated using a two-way ANOVA. Error bars represent protein ubiquitination relies on ribosome subunit dissociation. Mol. Cell 50, standard error of the mean. The ramps of Fig. 4e represent a decreasing titration 637–648 (2013). series of 4.2 µM, 0.42 µM, 0.21 µM, and 0.105 µM ﬁnal protein concentrations. In 16. Bengtson, M. H. & Joazeiro, C. A. Role of a ribosome-associated E3 ubiquitin Fig. 4f and Supplementary Fig. 4c, 1× and 1/10 refer to ﬁnal protein concentrations ligase in protein quality control. Nature 467, 470–473 (2010). of 4.2 µM and 0.42 µM, respectively. Source data can be found in Supplementary 17. Osuna, B. A., Howard, C. J., Kc, S., Frost, A. & Weinberg, D. E. In vitro Fig. 7. analysis of RQC activities provides insights into the mechanism and function of CAT tailing. eLife 6, e27949 (2017). 18. Kostova, K. K. et al. CAT-tailing as a fail-safe mechanism for efﬁcient Quantitative RT-PCR. RNA was puriﬁed from 40 ml of yeast cultures grown to OD ~1. Pelleted cells were washed once with water and resuspended in 700 μlof degradation of stalled nascent polypeptides. Science 357, 414–417 Trizol reagent (Ambion). An equal volume of glass beads was added and sus- (2017). pensions were vortexed for 30 s intervened with 1 min rest intervals. Next, the 19. Choe, Y. J. et al. Failure of RQC machinery causes protein aggregation and manufacture’s protocol from the Direct-zol kit (Zymo research: R2050-11-330) was proteotoxic stress. Nature 531, 191–195 (2016). followed. Complementary DNA (cDNA) was obtained from 0.5 μg of puriﬁed RNA 20. Defenouillere, Q., Zhang, E., Namane, A., Mouaikel, J., Jacquier, A. & using the High-capacity cDNA Reverse Transcription kit (4368814) from Applied Fromont-Racine, M. Rqc1 and Ltn1 prevent C-terminal alanine-threonine tail Biosystems. Quantitative PCR was performed using the LightCycler 480 SYBR (CAT-tail)-induced protein aggregation by efﬁcient recruitment of Cdc48 on Green I Master (04707516001) from Roche and using a FLAG-HIS and Actin stalled 60S subunits. J. Biol. Chem. 291, 12245–12253 (2016). nd primer pairs. Quantitative PCR analysis was done by Absolute Quantiﬁcation/2 21. Yonashiro, R. et al. The Rqc2/Tae2 subunit of the ribosome-associated quality derivative of three independent biological replicates, each performed in triplicate. control (RQC) complex marks ribosome-stalled nascent polypeptide chains The statistical analysis of mRNA transcript abundance was done after normal- for aggregation. eLife 5, e11794 (2016). ization with Actin. The statistics software Graphpad Prism 6 was used to perform 22. Izawa, T., Park, S. H., Zhao, L., Hartl, F. U. & Neupert, W. Cytosolic protein Student's t-test (unpaired two-tail) with a conﬁdence interval value of p < 0.05. Vms1 links ribosome quality control to mitochondrial and cellular Error bars represent standard error of the mean. homeostasis. Cell 171, 890–903.e18 (2017). 23. Heo, J. M. et al. A stress-responsive system for mitochondrial protein Data availability. The authors declare that the data supporting the ﬁndings of this degradation. Mol. Cell 40, 465–480 (2010). study are available within the paper and its supplementary information ﬁles. 24. Heo, J. M., Nielson, J. R., Dephoure, N., Gygi, S. P. & Rutter, J. Intramolecular Further relevant data on the genes studied in this manuscript (VMS1: YDR049W, interactions control Vms1 translocation to damaged mitochondria. Mol. Biol. RQC1: YDR333C, RQC2: YPL009C, LTN1: YMR2476, SKI7: YOR076C, DOM34: Cell. 24, 1263–1273 (2013). YNL01W) can be found at: https://www.yeastgenome.org. 25. Jefferies, H. B., Reinhard, C., Kozma, S. C. & Thomas, G. Rapamycin selectively represses translation of the “polypyrimidine tract” mRNA family. Proc. Natl. Acad. Sci. USA 91, 4441–4445 (1994). Received: 13 April 2018 Accepted: 3 May 2018 26. Jefferies, H. B., Fumagalli, S., Dennis, P. B., Reinhard, C., Pearson, R. B. & Thomas, G. Rapamycin suppresses 5’TOP mRNA translation through inhibition of p70s6k. EMBO J. 16, 3693–3704 (1997). 27. Shoemaker, C. J. & Green, R. Translation drives mRNA quality control. Nat. Struct. Mol. Biol. 19, 594–601 (2012). 28. Yang J., Hao X., Cao X., Liu B. & Nystrom T. Spatial sequestration and References detoxiﬁcation of Huntingtin by the ribosome quality control complex. eLife 5, 1. Brandman, O. et al. A ribosome-bound quality control complex triggers e11792 (2016). degradation of nascent peptides and signals translation stress. Cell 151, 29. Meaux, S. & Van Hoof, A. Yeast transcripts cleaved by an internal ribozyme 1042–1054 (2012). provide new insight into the role of the cap and poly(A) tail in translation and 2. Verma, R., Oania, R. S., Kolawa, N. J. & Deshaies, R. J. Cdc48/p97 promotes mRNA decay. RNA 12, 1323–1337 (2006). degradation of aberrant nascent polypeptides bound to the ribosome. eLife 2, 30. Yogev, O., Karniely, S. & Pines, O. Translation-coupled translocation of yeast e00308 (2013). fumarase into mitochondria in vivo. J. Biol. Chem. 282, 29222–29229 (2007). 3. Defenouillere, Q. et al. Cdc48-associated complex bound to 60S particles is 31. Nielson, J. R. et al. Sterol oxidation mediates stress-responsive Vms1 required for the clearance of aberrant translation products. Proc. Natl. Acad. translocation to mitochondria. Mol. Cell 68, 673–685 (2017). e676. Sci. USA 110, 5046–5051 (2013). 32. Holm, L. & Rosenstrom, P. Dali server: conservation mapping in 3D. Nucleic 4. van den Elzen, A. M. et al. Dissection of Dom34-Hbs1 reveals independent Acids Res. 38, W545–W549 (2010). functions in two RNA quality control pathways. Nat. Struct. Mol. Biol. 17, 33. Song, H. et al. The crystal structure of human eukaryotic release factor eRF1-- 1446–1452 (2010). mechanism of stop codon recognition and peptidyl-tRNA hydrolysis. Cell 100, 5. van den Elzen, A. M., Schuller, A., Green, R. & Seraphin, B. Dom34-Hbs1 311–321 (2000). mediated dissociation of inactive 80S ribosomes promotes restart of 34. Graille, M., Chaillet, M. & van Tilbeurgh, H. Structure of yeast Dom34: a translation after stress. EMBO J. 33, 265–276 (2014). protein related to translation termination factor Erf1 and involved in No-Go 6. Kobayashi, K. et al. Structural basis for mRNA surveillance by archaeal Pelota decay. J. Biol. Chem. 283, 7145–7154 (2008). and GTP-bound EF1alpha complex. Proc. Natl. Acad. Sci. USA 107, 35. Brown, A., Shao, S., Murray, J., Hegde, R. S. & Ramakrishnan, V. Structural 17575–17579 (2010). basis for stop codon recognition in eukaryotes. Nature 524, 493–496 7. Tsuboi, T. et al. Dom34:hbs1 plays a general role in quality-control systems by (2015). dissociation of a stalled ribosome at the 3’ end of aberrant mRNA. Mol. Cell 36. Chu, J. et al. A mouse forward genetics screen identiﬁes LISTERIN as an E3 46, 518–529 (2012). ubiquitin ligase involved in neurodegeneration. Proc. Natl. Acad. Sci. USA 106, 8. Shoemaker, C. J., Eyler, D. E. & Green, R. Dom34:Hbs1 promotes subunit 2097–2103 (2009). dissociation and peptidyl-tRNA drop-off to initiate no-go decay. Science 330, 369–372 (2010). 9. Shoemaker, C. J. & Green, R. Kinetic analysis reveals the ordered coupling of translation termination and ribosome recycling in yeast. Proc. Natl. Acad. Sci. Acknowledgements USA 108, E1392–E1398 (2011). This work was supported by a Faculty Scholar grant from the Howard Hughes Medical 10. Chen, L. et al. Structure of the Dom34-Hbs1 complex and implications for no- Institute (to A.F.), the Searle Scholars Program (to A.F.), NIH grant GM115129 (to C.P. go decay. Nat. Struct. Mol. Biol. 17, 1233–1240 (2010). H. and J.R.), a grant from the Nora Eccles Treadwell Foundation (to J.R. and C.P.H.), 11. Shen, P. S. et al. Protein synthesis. Rqc2p and 60S ribosomal subunits mediate NIH grant 1DP2GM110772-01 (to A.F.), training grants 17POST33670814 (to O.Z.R.), mRNA-independent elongation of nascent chains. Science 347,75–78 (2015). T32HL007576, AHA 14POST20380216, and T32DK007115 (to E.K.F.), a Hillblom 12. Shao, S., Brown, A., Santhanam, B. & Hegde, R. S. Structure and assembly Graduate Research fellowship (to C.J.H.), a Heyman Discovery fellowship (to B.A.O), pathway of the ribosome quality control complex. Mol. Cell 57, 433–444 and the Howard Hughes Medical Institute (to J.R.). A.F. is a Chan Zuckerberg Biohub (2015). investigator. This work was supported by the University of Utah Flow Cytometry Facility in addition to the National Cancer Institute through Award Number 5P30CA042014-24. 8 NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04564-3 ARTICLE Author contributions Open Access This article is licensed under a Creative Commons O.Z.R., E.K.F., C.J.H., C.P.H., J.R, and A.F. designed the study and wrote the manuscript. Attribution 4.0 International License, which permits use, sharing, N.D.T. and J.V.V. ran the polysome assays. O.Z.R., E.K.F., and C.J.H. collected the data. J. adaptation, distribution and reproduction in any medium or format, as long as you give V.V. and S.F. generated plasmid constructs and yeast strains. C.P.H. helped with appropriate credit to the original author(s) and the source, provide a link to the Creative structural analyses. R.K. performed structural homology modeling and alignments. B.A. Commons license, and indicate if changes were made. The images or other third party O. helped with the IVT assays. P.S.S. helped with the co-immunoprecipitation experi- material in this article are included in the article’s Creative Commons license, unless ments. All authors commented and approved of the ﬁnal manuscript. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from Additional information the copyright holder. To view a copy of this license, visit http://creativecommons.org/ Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- licenses/by/4.0/. 018-04564-3. Competing interests: The authors declare no competing interests. © The Author(s) 2018 Reprints and permission information is available online at http://npg.nature.com/ reprintsandpermissions/ Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. NATURE COMMUNICATIONS (2018) 9:2197 DOI: 10.1038/s41467-018-04564-3 www.nature.com/naturecommunications 9 | | |
Nature Communications – Springer Journals
Published: Jun 6, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera