Translation of mRNAs to produce proteins is a fundamental cellular process that supports the cell’s ability to carry out the basic enzymatic reactions needed for life. To prevent errors that arise during this complex process from compromising cellular metabolism, specialized molecular pathways have evolved to recognize and regulate problematic translation events (D’Orazio and Green, 2021a; Inada, 2017; Yan and Zaher, 2019). These mechanisms are coupled to RNA decay pathways that target problematic mRNAs and prevent continuing diversion of ribosomes toward unproductive translation. The set of factors involved in this crucial recognition of problems that arise during translation elongation and the mechanisms by which they exert their downstream effects on mRNA stability remain only partially characterized.

General mRNA decay in yeast is catalyzed primarily by mRNA decapping and 5’ to 3’ exonucleolytic degradation by Xrn1, while the 3’ to 5’ exonuclease (the exosome) is thought to play a role only under certain circumstances (Muhlrad et al., 1994). Recent foundational work in yeast, and subsequently in zebrafish and mammals, discovered that mRNA stability is correlated with its codon usage (Presnyak et al., 2015; Mishima and Tomari, 2016; Wu et al., 2019a): mRNAs enriched in non-optimal codons have short half-lives and are rapidly degraded by the cytoplasmic Ccr4-Not deadenylation complex and the decapping activator Dhh1 (Radhakrishnan et al., 2016; Webster et al., 2018; Sweet et al., 2012). Recent biochemical and structural evidence supports a model in which suboptimal codons in the ribosomal A site slow down translation elongation, allowing deacylated tRNA to diffuse away from the E site and enabling the critical adaptor protein Not5 to bind. Not5 binding in the vacant ribosomal E sites can recruit the Ccr4-Not complex to promote mRNA deadenylation, decapping, and decay (Buschauer et al., 2020). These observations provide molecular insight into codon-optimality-mediated decay (COMD) and support for the idea that this pathway represents an important determinant of general cellular mRNA half-lives.

In contrast to the normal slowing of translation that occurs transiently as ribosomes decode less optimal codons, the cell also possesses quality-control machinery to resolve more deleterious ribosomal stalls that can arise from chemical damage in the mRNA (truncation, depurination, nucleobase dimers, oxidative damage, etc.), difficult to unwind secondary structure, or incorrect nuclear mRNA processing events (D’Orazio and Green, 2021a). Adjacent pairs of specific rare codons can mimic these events and induce strong inhibition of translation elongation and associated mRNA decay in Saccharomyces cerevisiae (Gamble et al., 2016). For example, consecutive CGA codons induce terminal stalls and have been routinely included in reporter mRNAs to trigger an mRNA surveillance pathway referred to as no-go decay (NGD) (Tsuboi et al., 2012; Letzring et al., 2013; Tesina et al., 2020). On these problematic mRNAs, ribosomes stall on the CGA codons, leading to ribosomal collisions that promote small-subunit protein ubiquitination by the E3 ligase Hel2 (mammalian ZNF598) (Juszkiewicz et al., 2018, Matsuo et al., 2017, Saito et al., 2015, Sundaramoorthy et al., 2017), ribosomal clearance by the helicase Slh1 and ribosome quality control trigger (RQT) complex (Ikeuchi et al., 2018), and nascent peptide decay by the ribosome quality control (RQC) complex (Brandman et al., 2012). The accumulation of colliding ribosomes is thought to trigger decapping and Xrn1-mediated mRNA degradation, although the specific molecular players and interactions responsible for triggering this decay remain poorly defined (D’Orazio et al., 2019; Simms et al., 2018). Additionally, under conditions where the ribosome rescue machinery is compromised or overwhelmed, cleavage of reporter mRNAs by the endonuclease Cue2 and Dom34-mediated rescue provides an alternate route for mRNA degradation and ribosome rescue (Doma and Parker, 2006; D’Orazio et al., 2019; Glover et al., 2020). NGD, as a ‘quality control’ pathway, is thought to minimize the detrimental effects from damaged or problematic mRNAs in the cell and to reduce the overall impact of proteotoxic stress (Brandman et al., 2012; Ishimura et al., 2014; Martin et al., 2020).

Although NGD and COMD can be triggered by seemingly similar mRNA sequences to converge on Xrn1-mediated exonucleolytic decay of mRNAs (Pelechano et al., 2015), the extent to which these processes overlap in specificity and activity remains unclear, as do their complete sets of accessory factors. Moreover, the molecular states of the ribosome which define and activate these pathways have not been systematically compared. In this study, we address these questions through genetic screening and functional assays in the yeast S. cerevisiae. We use reporter mRNAs designed to trigger NGD or COMD and perform reporter-synthetic genetic array (R-SGA) screens to identify factors critical to these separate mRNA decay pathways. Importantly, we identify a critical role for Syh1 as a strong effector for decay of mRNAs with terminal stalls. We use flow cytometry and northern blotting combined with genetic perturbations to reveal the contributions of other NGD factors and COMD factors to translational repression and decay of the reporter mRNAs. Finally, we use ribosome profiling of NGD and COMD reporters to isolate the activities of the major players in these pathways and connect these activities to the molecular states of elongating ribosomes. These data provide a basis for understanding the unique contributions of each of these pathways to translation-coupled mRNA decay and contextualizes their effects in the larger cellular process of translation surveillance.

In this study, we use carefully designed reporter mRNAs to study translation-coupled mRNA decay pathways in S. cerevisiae. Using R-SGA screening with a reporter mRNA containing iterated CGA codons, we identified and validated a set of genes that contribute to no-go decay (NGD). Subsequent analysis allowed us to compare the mechanisms of the pathways that regulate decay of mRNAs with either highly problematic (NGD) or slowly decoded sequences (COMD). We find that an independent pathway of NGD is driven primarily by the actions of the GIGYF1/2-homologous proteins Syh1 and Smy2; in contrast, the Syh1/Smy2-dependent pathway has no discernible impact on the stability of non-optimal mRNA sequences. We show that the previously defined COMD factor Not5 contributes modestly to decay of the NGD reporter and very strongly to decay of non-optimally coded mRNAs. Finally, we connect these distinct molecular decay profiles with ribosome states using ribosome profiling, showing that the presence of colliding ribosomes (disomes) is correlated with targeting via NGD while the presence of slow ribosomes (monosomes with empty A and E sites) is correlated with targeting via COMD.

Our assays using the NGD reporters reveal the interplay between Hel2, its dependent NGD factors, and Syh1/Smy2 in responding to ribosome collisions. First, we found that combining deletion of SYH1 and its paralog SMY2 led to modest stabilization of NGD reporter mRNAs and deletion of CUE2 and SYH1 together led to very potent stabilization (Figure 2C). In light of previous work establishing the relationship between Hel2, Cue2, and Slh1 (D’Orazio et al., 2019), we interpret the mRNA stability data as follows: (1) in the wild-type strain, the NGD reporters are destabilized because colliding ribosomes lead to the recruitment of Syh1/Smy2 and elicit mRNA destabilization while Hel2-mediated ubiquitination licenses Slh1-mediated ribosome clearing and Cue2-mediated endonucleolytic cleavage (2) in the syh1∆smy2∆ strain, NGD reporter levels are only partially rescued because Cue2-mediated endonucleolytic decay can still trigger NGD; and (3) in the hel2∆syh1∆ and cue2∆syh1∆ strains, NGD reporters are strongly stabilized because both Syh1-dependent and Cue2-dependent decay pathways are impaired. According to this view, there are two distinct pathways for NGD that can compensate for each other in various conditions. This model gives new context to the pathways that respond to elongation stalls and emphasizes the importance of Syh1, and to a lesser extent Smy2, as distinct effectors of NGD. These data further support our previously discovered mechanism in which the ubiquitination activity of Hel2 triggers RQT-mediated ribosome rescue through Slh1, resorting to Cue2-dependent endonucleolytic cleavage when mechanisms to resolve ribosome collisions are overwhelmed. The synergistic activities of Syh1/Smy2- and Hel2-triggered decay form the basis of a robust cellular system for targeting problematic mRNAs for destruction (Figure 6).

Figure 6

Download asset Open asset

This interplay was further explored in our Ribo-seq experiments. For the minCGA reporter, we observed an ordered, periodic pattern of monosome and disome footprints upstream of the CGA region in the WT strain, and this pattern was substantially disrupted in the hel2∆ strain. Further, we observe a larger proportion of ‘escaping’ monosome footprints downstream of the CGA region in the hel2∆ strains. Given the role of Hel2 in recognizing collided ribosome and promoting clearance by RQT, we interpret these data to mean that the activity of Hel2 on collided ribosomes in part stabilizes the collided ribosome interface in such a way that recognition and clearance by Slh1 is facilitated (Meydan and Guydosh, 2020). These observations are distinct from the maintenance of wild-type-like periodicity when SYH1 is knocked out (Figure 5—figure supplement 1C).

The next set of questions focused on how Syh1 (and its homolog Smy2) is recruited to problematic mRNAs and how it triggers translational repression or mRNA decay. Recent work in mammalian cells investigating the mechanism of recruitment of GIGYF2 to NGD-targeted mRNAs yielded two competing models: one in which ZNF598 (a mammalian HEL2 homolog) acts to recruit GIGYF2 (Hickey et al., 2020) and one in which EDF1 acts to recruit GIGYF2 (Sinha et al., 2020; Juszkiewicz et al., 2020). We tested both models using our yeast reporter system. First, we observed strong stabilization of the GFP-CGA reporter mRNA levels when SYH1 is deleted in a hel2∆ background (comparing hel2∆ to hel2∆syh1∆) thus establishing that Hel2 function is not necessary for Syh1 function in yeast. Second, we did not observe the same strong rescue of minCGA reporter mRNA levels in a hel2∆mbf1∆ strain as we did in a hel2∆syh1∆ strain, suggesting that Syh1 function is also not dependent on Mbf1. Although Mbf1 does bind collided disomes in yeast (Sinha et al., 2020; Pochopien et al., 2021), these data suggest that Mbf1 has a limited impact on NGD per se in this system. A previous study that employed bioID mass spectrometry using Syh1 as bait (Opitz et al., 2017) raises the possibility that Syh1 interacts directly with Asc1, a ribosomal protein known to be important for NGD in yeast (Kuroha et al., 2010; Letzring et al., 2013; Brandman et al., 2012), and a top effector of NGD in our genetic screen (Figure 1A). Further work exploring the nature of the physical interactions between Syh1/Smy2 and ribosomal collisions is necessary to make stronger conclusions about the mechanism of Syh1/Smy2-mediated NGD.

We also considered a role for other factors in facilitating Syh1-mediated decay, including the potential eIF4E2 homolog, EAP1, which did not affect reporter levels (Figure 2—figure supplement 1F). Another candidate effector protein for Syh1 function is Dhh1 whose mammalian homolog DDX6 interacts with GIGYF1/2 and has been shown to facilitate translational repression (Peter et al., 2019; Weber et al., 2020). In yeast, however, there is scarce evidence for a Dhh1-Syh1/Smy2 complex (Ergüden, 2019) and the conserved DDX6 binding motif of GIGYF1/2 is absent in Syh1/Smy2 (Figure 2—figure supplement 1G). These data together raise the interesting possibility that Syh1 and Smy2 have a distinct mechanism of regulation in yeast that involves direct signaling of mRNA decay independent of translational repression. We note that the strong mRNA decay phenotype associated with Syh1/Smy2 function and the NGD reporters in yeast (here and in Hickey et al., 2020) is distinct from the translational repression phenotype associated with GIGYF2:4E2 function in mammalian systems (Morita et al., 2012; Peter et al., 2019).

Although NGD and COMD are both known to be co-translational mRNA decay processes, the distinct ribosome states that trigger these events have not been systematically explored. For example, it has not been clear whether ribosomes translating highly non-optimal CGA repeats (that efficiently trigger NGD) may also be recognized by Not5. Conversely, the possibility remained open that highly non-optimal reporter mRNAs cause ribosome collisions that activate NGD in addition to COMD.

Our analysis of the COMD reporters revealed responses only to factors previously implicated in this pathway (Buschauer et al., 2020). The half-lives of the minNONOPT reporter were not impacted by deletion of SYH1, HEL2, or HEL2/SYH1 (Figure 4B) and the ribosome profiling data revealed a relatively even distribution of monosomes and low abundance of disome peaks. Additionally, we observed an enrichment of 21 nt RPFs over the coding region representing slowly elongating ribosomes (Figure 5D). These data are broadly consistent with the previously reported Not5-driven mechanism being responsible for COMD with essentially no contribution from colliding ribosomes and NGD. By contrast, the minCGA reporter was strongly stabilized by deletion of factors implicated in NGD (Syh1/Smy2 and Hel2) and weakly stabilized by deletion of factors implicated in COMD (Not5) (Figure 4B). The ribosome profiling data revealed abundant collided disomes at the CGA repeat sequences, but also an enrichment of 21 nt monosome RPFs in the actual CGA repeats reflecting ribosomes struggling their way through these difficult to decode sequences (Figure 5B and E). These data rationalize both the strong contribution of the NGD factors and the more modest contribution of the COMD factors to minCGA reporter stability.

Our study provides evidence for generally non-overlapping targets of NGD and COMD and ribosome states correlated with each pathway. While the NGD machinery, under the control of Syh1/Smy2 and Hel2, responds to specific defects in elongation due to stalled and collided ribosomes, the COMD machinery, under the control of Not5, surveys the pool of translating ribosomes for mRNAs on which there is overall slow translation. We speculate that the COMD pathway is a general one that regulates overall mRNA stability, independent of ribosome dysfunction, while the NGD pathway evolved to deal with more acute environmental disturbances such as UV or oxidative damage (Yan et al., 2019; C. C.-C. Wu et al., 2020). Future studies will better characterize the molecular mechanisms of these pathways and will provide new foundations for an understanding of the homeostasis of cellular translation and mRNA decay.

Ribo-seq data is available in the NCBI Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) database with the accession GSE189404. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al. 2019) partner repository with the dataset identifier PXD030076. Code to process sequencing data is available at https://github.com/greenlabjhmi/2022_syh1/, (copy archived at swh:1:rev:2f26e680e06d11fd517105d52f19005f00b5c7af).

1. 1. Veltri AJ
   2. Green R

   (2021) NCBI Gene Expression Omnibus

   ID GSE189404. Distinct ribosome states trigger diverse mRNA quality control pathways.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?&acc=GSE189404
2. 1. Veltri AJ
   2. Green R

   (2021) PRIDE

   ID PXD030076. Distinct ribosome states trigger diverse mRNA quality control pathways.

   https://www.ebi.ac.uk/pride/archive?keyword=PXD030076

1. 1. Amberg DC
   2. Burke DJ
   3. Strathern JN

   (2006) Tandem Affinity Protein (TAP) Purification from Yeast

   Cold Spring Harbor Protocols 2006:db.

   https://doi.org/10.1101/pdb.prot4153

   • Google Scholar
2. 1. Brandman O
   2. Stewart-Ornstein J
   3. Wong D
   4. Larson A
   5. Williams CC
   6. Li G-W
   7. Zhou S
   8. King D
   9. Shen PS
   10. Weibezahn J
   11. Dunn JG
   12. Rouskin S
   13. Inada T
   14. Frost A
   15. Weissman JS

   (2012) A ribosome-bound quality control complex triggers degradation of nascent peptides and signals translation stress

   Cell 151:1042–1054.

   https://doi.org/10.1016/j.cell.2012.10.044

   • PubMed
   • Google Scholar
3. Software

   1. Brown JD
   2. Ryan MD

   (2010)

   Ribosome ‘Skipping’: ‘Stop-Carry On’ or ‘StopGo’ Translation

   Recoding: Expansion of Decoding Rules Enriches Gene Expression.
4. 1. Buschauer R
   2. Matsuo Y
   3. Sugiyama T
   4. Chen YH
   5. Alhusaini N
   6. Sweet T
   7. Ikeuchi K
   8. Cheng J
   9. Matsuki Y
   10. Nobuta R
   11. Gilmozzi A
   12. Berninghausen O
   13. Tesina P
   14. Becker T
   15. Coller J
   16. Inada T
   17. Beckmann R

   (2020) The Ccr4-Not complex monitors the translating ribosome for codon optimality

   Science 368:eaay6912.

   https://doi.org/10.1126/science.aay6912

   • PubMed
   • Google Scholar
5. 1. Cosentino GP
   2. Schmelzle T
   3. Haghighat A
   4. Helliwell SB
   5. Hall MN
   6. Sonenberg N

   (2000) Eap1p, a novel eukaryotic translation initiation factor 4E-associated protein in Saccharomyces cerevisiae

   Molecular and Cellular Biology 20:4604–4613.

   https://doi.org/10.1128/MCB.20.13.4604-4613.2000

   • PubMed
   • Google Scholar
6. 1. Dobin A
   2. Davis CA
   3. Schlesinger F
   4. Drenkow J
   5. Zaleski C
   6. Jha S
   7. Batut P
   8. Chaisson M
   9. Gingeras TR

   (2013) STAR: ultrafast universal RNA-seq aligner

   Bioinformatics 29:15–21.

   https://doi.org/10.1093/bioinformatics/bts635

   • PubMed
   • Google Scholar
7. 1. Doma MK
   2. Parker R

   (2006) Endonucleolytic cleavage of eukaryotic mRNAs with stalls in translation elongation

   Nature 440:561–564.

   https://doi.org/10.1038/nature04530

   • PubMed
   • Google Scholar
8. 1. D’Orazio KN
   2. Wu CC-C
   3. Sinha N
   4. Loll-Krippleber R
   5. Brown GW
   6. Green R

   (2019) The endonuclease Cue2 cleaves mRNAs at stalled ribosomes during No Go Decay

   eLife 8:e49117.

   https://doi.org/10.7554/eLife.49117

   • PubMed
   • Google Scholar
9. 1. D’Orazio KN
   2. Green R

   (2021a) Ribosome states signal RNA quality control

   Molecular Cell 81:1372–1383.

   https://doi.org/10.1016/j.molcel.2021.02.022

   • PubMed
   • Google Scholar
10. Preprint

    1. D’Orazio KN
    2. Lessen LN
    3. Veltri AJ
    4. Neiman Z
    5. Pacheco M
    6. Loll-Krippleber R
    7. Brown GW
    8. Green R

    (2021b) Genetic Screens Identify Connections between Ribosome Recycling and Nonsense Mediated Decay

    bioRxiv.

    https://doi.org/10.1101/2021.08.03.454884

    • Google Scholar
11. 1. Ergüden B

    (2019) Dhh1 is a member of the SESA network

    Yeast 36:99–105.

    https://doi.org/10.1002/yea.3363

    • PubMed
    • Google Scholar
12. 1. Fillingham J
    2. Kainth P
    3. Lambert J-P
    4. van Bakel H
    5. Tsui K
    6. Peña-Castillo L
    7. Nislow C
    8. Figeys D
    9. Hughes TR
    10. Greenblatt J
    11. Andrews BJ

    (2009) Two-color cell array screen reveals interdependent roles for histone chaperones and a chromatin boundary regulator in histone gene repression

    Molecular Cell 35:340–351.

    https://doi.org/10.1016/j.molcel.2009.06.023

    • PubMed
    • Google Scholar
13. 1. Gamble CE
    2. Brule CE
    3. Dean KM
    4. Fields S
    5. Grayhack EJ

    (2016) Adjacent codons act in concert to modulate translation efficiency in yeast

    Cell 166:679–690.

    https://doi.org/10.1016/j.cell.2016.05.070

    • PubMed
    • Google Scholar
14. 1. Giaever G
    2. Chu AM
    3. Ni L
    4. Connelly C
    5. Riles L
    6. Véronneau S
    7. Dow S
    8. Lucau-Danila A
    9. Anderson K
    10. André B
    11. Arkin AP
    12. Astromoff A
    13. El Bakkoury M
    14. Bangham R
    15. Benito R
    16. Brachat S
    17. Campanaro S
    18. Curtiss M
    19. Davis K
    20. Deutschbauer A
    21. Entian KD
    22. Flaherty P
    23. Foury F
    24. Garfinkel DJ
    25. Gerstein M
    26. Gotte D
    27. Güldener U
    28. Hegemann JH
    29. Hempel S
    30. Herman Z
    31. Jaramillo DF
    32. Kelly DE
    33. Kelly SL
    34. Kötter P
    35. LaBonte D
    36. Lamb DC
    37. Lan N
    38. Liang H
    39. Liao H
    40. Liu L
    41. Luo C
    42. Lussier M
    43. Mao R
    44. Menard P
    45. Ooi SL
    46. Revuelta JL
    47. Roberts CJ
    48. Rose M
    49. Ross-Macdonald P
    50. Scherens B
    51. Schimmack G
    52. Shafer B
    53. Shoemaker DD
    54. Sookhai-Mahadeo S
    55. Storms RK
    56. Strathern JN
    57. Valle G
    58. Voet M
    59. Volckaert G
    60. Wang C
    61. Ward TR
    62. Wilhelmy J
    63. Winzeler EA
    64. Yang Y
    65. Yen G
    66. Youngman E
    67. Yu K
    68. Bussey H
    69. Boeke JD
    70. Snyder M
    71. Philippsen P
    72. Davis RW
    73. Johnston M

    (2002) Functional profiling of the Saccharomyces cerevisiae genome

    Nature 418:387–391.

    https://doi.org/10.1038/nature00935

    • PubMed
    • Google Scholar
15. 1. Gietz RD
    2. Schiestl RH

    (2007) High-efficiency yeast transformation using the LiAc/SS carrier DNA/PEG method

    Nature Protocols 2:31–34.

    https://doi.org/10.1038/nprot.2007.13

    • PubMed
    • Google Scholar
16. 1. Glover ML
    2. Burroughs AM
    3. Monem PC
    4. Egelhofer TA
    5. Pule MN
    6. Aravind L
    7. Arribere JA

    (2020) NONU-1 Encodes a Conserved Endonuclease Required for MRNA Translation Surveillance

    Cell Reports 30:4321–4331.

    https://doi.org/10.1016/j.celrep.2020.03.023

    • PubMed
    • Google Scholar
17. Software

    1. GSEApy

    (2022) GSEApy

    Github.

    https://github.com/zqfang/gseapy
18. 1. Guydosh NR
    2. Green R

    (2014) Dom34 rescues ribosomes in 3’ untranslated regions

    Cell 156:950–962.

    https://doi.org/10.1016/j.cell.2014.02.006

    • PubMed
    • Google Scholar
19. 1. Hendrick JL
    2. Wilson PG
    3. Edelman II
    4. Sandbaken MG
    5. Ursic D
    6. Culbertson MR

    (2001) Yeast frameshift suppressor mutations in the genes coding for transcription factor Mbf1p and ribosomal protein S3: evidence for autoregulation of S3 synthesis

    Genetics 157:1141–1158.

    https://doi.org/10.1093/genetics/157.3.1141

    • PubMed
    • Google Scholar
20. 1. Hickey KL
    2. Dickson K
    3. Cogan JZ
    4. Replogle JM
    5. Schoof M
    6. D’Orazio KN
    7. Sinha NK
    8. Hussmann JA
    9. Jost M
    10. Frost A
    11. Green R
    12. Weissman JS
    13. Kostova KK

    (2020) GIGYF2 and 4EHP Inhibit Translation Initiation of Defective Messenger RNAs to Assist Ribosome-Associated Quality Control

    Molecular Cell 79:950–962.

    https://doi.org/10.1016/j.molcel.2020.07.007

    • Google Scholar
21. 1. Ikeuchi K
    2. Izawa T
    3. Inada T

    (2018) Recent progress on the molecular mechanism of quality controls induced by ribosome stalling

    Frontiers in Genetics 9:743.

    https://doi.org/10.3389/fgene.2018.00743

    • PubMed
    • Google Scholar
22. 1. Ikeuchi K
    2. Tesina P
    3. Matsuo Y
    4. Sugiyama T
    5. Cheng J
    6. Saeki Y
    7. Tanaka K
    8. Becker T
    9. Beckmann R
    10. Inada T

    (2019) Collided ribosomes form a unique structural interface to induce Hel2-driven quality control pathways

    The EMBO Journal 38:e100276.

    https://doi.org/10.15252/embj.2018100276

    • PubMed
    • Google Scholar
23. 1. Inada T

    (2017) The Ribosome as a Platform for MRNA and Nascent Polypeptide Quality Control

    Trends in Biochemical Sciences 42:5–15.

    https://doi.org/10.1016/j.tibs.2016.09.005

    • PubMed
    • Google Scholar
24. 1. Ingolia NT
    2. Ghaemmaghami S
    3. Newman JRS
    4. Weissman JS

    (2009) Genome-wide analysis in vivo of translation with nucleotide resolution using ribosome profiling

    Science 324:218–223.

    https://doi.org/10.1126/science.1168978

    • PubMed
    • Google Scholar
25. 1. Ishimura R
    2. Nagy G
    3. Dotu I
    4. Zhou H
    5. Yang X-L
    6. Schimmel P
    7. Senju S
    8. Nishimura Y
    9. Chuang JH
    10. Ackerman SL

    (2014) RNA function. Ribosome stalling induced by mutation of a CNS-specific TRNA causes neurodegeneration

    Science 345:455–459.

    https://doi.org/10.1126/science.1249749

    • PubMed
    • Google Scholar
26. 1. Juszkiewicz S
    2. Chandrasekaran V
    3. Lin Z
    4. Kraatz S
    5. Ramakrishnan V
    6. Hegde RS

    (2018) ZNF598 Is a Quality Control Sensor of Collided Ribosomes

    Molecular Cell 72:469–481.

    https://doi.org/10.1016/j.molcel.2018.08.037

    • PubMed
    • Google Scholar
27. 1. Juszkiewicz S
    2. Slodkowicz G
    3. Lin Z
    4. Freire-Pritchett P
    5. Peak-Chew SY
    6. Hegde RS

    (2020) Ribosome collisions trigger cis-acting feedback inhibition of translation initiation

    eLife 9:e60038.

    https://doi.org/10.7554/eLife.60038

    • Google Scholar
28. 1. Koutmou KS
    2. Schuller AP
    3. Brunelle JL
    4. Radhakrishnan A
    5. Djuranovic S
    6. Green R

    (2015) Ribosomes slide on lysine-encoding homopolymeric A stretches

    eLife 4:e05534.

    https://doi.org/10.7554/eLife.05534

    • PubMed
    • Google Scholar
29. 1. Kuroha K
    2. Akamatsu M
    3. Dimitrova L
    4. Ito T
    5. Kato Y
    6. Shirahige K
    7. Inada T

    (2010) Receptor for activated C kinase 1 stimulates nascent polypeptide-dependent translation arrest

    EMBO Reports 11:956–961.

    https://doi.org/10.1038/embor.2010.169

    • PubMed
    • Google Scholar
30. 1. Letzring DP
    2. Dean KM
    3. Grayhack EJ

    (2010) Control of translation efficiency in yeast by codon-anticodon interactions

    RNA 16:2516–2528.

    https://doi.org/10.1261/rna.2411710

    • PubMed
    • Google Scholar
31. 1. Letzring DP
    2. Wolf AS
    3. Brule CE
    4. Grayhack EJ

    (2013) Translation of CGA codon repeats in yeast involves quality control components and ribosomal protein L1

    RNA 19:1208–1217.

    https://doi.org/10.1261/rna.039446.113

    • PubMed
    • Google Scholar
32. 1. Martin PB
    2. Kigoshi-Tansho Y
    3. Sher RB
    4. Ravenscroft G
    5. Stauffer JE
    6. Kumar R
    7. Yonashiro R
    8. Müller T
    9. Griffith C
    10. Allen W
    11. Pehlivan D
    12. Harel T
    13. Zenker M
    14. Howting D
    15. Schanze D
    16. Faqeih EA
    17. Almontashiri NAM
    18. Maroofian R
    19. Houlden H
    20. Mazaheri N
    21. Galehdari H
    22. Douglas G
    23. Posey JE
    24. Ryan M
    25. Lupski JR
    26. Laing NG
    27. Joazeiro CAP
    28. Cox GA

    (2020) NEMF mutations that impair ribosome-associated quality control are associated with neuromuscular disease

    Nature Communications 11:4625.

    https://doi.org/10.1038/s41467-020-18327-6

    • PubMed
    • Google Scholar
33. 1. Matsuo Y
    2. Ikeuchi K
    3. Saeki Y
    4. Iwasaki S
    5. Schmidt C
    6. Udagawa T
    7. Sato F
    8. Tsuchiya H
    9. Becker T
    10. Tanaka K
    11. Ingolia NT
    12. Beckmann R
    13. Inada T

    (2017) Ubiquitination of stalled ribosome triggers ribosome-associated quality control

    Nature Communications 8:159.

    https://doi.org/10.1038/s41467-017-00188-1

    • PubMed
    • Google Scholar
34. 1. McCusker JH

    (2017) Introducing MX Cassettes into Saccharomyces cerevisiae

    Cold Spring Harbor Protocols 2017:pdb.prot088104.

    https://doi.org/10.1101/pdb.prot088104

    • PubMed
    • Google Scholar
35. 1. McGlincy NJ
    2. Ingolia NT

    (2017) Transcriptome-wide measurement of translation by ribosome profiling

    Methods 126:112–129.

    https://doi.org/10.1016/j.ymeth.2017.05.028

    • PubMed
    • Google Scholar
36. 1. Meydan S
    2. Guydosh NR

    (2020) Disome and trisome profiling reveal genome-wide targets of ribosome quality control

    Molecular Cell 79:588–602.

    https://doi.org/10.1016/j.molcel.2020.06.010

    • PubMed
    • Google Scholar
37. 1. Mishima Y
    2. Tomari Y

    (2016) Codon Usage and 3’ UTR Length Determine Maternal MRNA Stability in Zebrafish

    Molecular Cell 61:874–885.

    https://doi.org/10.1016/j.molcel.2016.02.027

    • PubMed
    • Google Scholar
38. 1. Morita M
    2. Ler LW
    3. Fabian MR
    4. Siddiqui N
    5. Mullin M
    6. Henderson VC
    7. Alain T
    8. Fonseca BD
    9. Karashchuk G
    10. Bennett CF
    11. Kabuta T
    12. Higashi S
    13. Larsson O
    14. Topisirovic I
    15. Smith RJ
    16. Gingras A-C
    17. Sonenberg N

    (2012) A novel 4EHP-GIGYF2 translational repressor complex is essential for mammalian development

    Molecular and Cellular Biology 32:3585–3593.

    https://doi.org/10.1128/MCB.00455-12

    • PubMed
    • Google Scholar
39. 1. Muhlrad D
    2. Decker CJ
    3. Parker R

    (1994) Deadenylation of the unstable MRNA encoded by the yeast MFA2 gene leads to decapping followed by 5’-->3’ digestion of the transcript

    Genes & Development 8:855–866.

    https://doi.org/10.1101/gad.8.7.855

    • PubMed
    • Google Scholar
40. 1. Notredame C
    2. Higgins DG
    3. Heringa J

    (2000) T-Coffee: A novel method for fast and accurate multiple sequence alignment

    Journal of Molecular Biology 302:205–217.

    https://doi.org/10.1006/jmbi.2000.4042

    • PubMed
    • Google Scholar
41. 1. Opitz N
    2. Schmitt K
    3. Hofer-Pretz V
    4. Neumann B
    5. Krebber H
    6. Braus GH
    7. Valerius O

    (2017) Capturing the Asc1p/Receptor for Activated C Kinase 1 (RACK1) Microenvironment at the Head Region of the 40S Ribosome with Quantitative BioID in Yeast

    Molecular & Cellular Proteomics 16:2199–2218.

    https://doi.org/10.1074/mcp.M116.066654

    • PubMed
    • Google Scholar
42. 1. O’Connor PBF
    2. Andreev DE
    3. Baranov PV

    (2016) Comparative survey of the relative impact of MRNA features on local ribosome profiling read density

    Nature Communications 7:12915.

    https://doi.org/10.1038/ncomms12915

    • PubMed
    • Google Scholar
43. 1. Pelechano V
    2. Wei W
    3. Steinmetz LM

    (2015) Widespread Co-translational RNA Decay Reveals Ribosome Dynamics

    Cell 161:1400–1412.

    https://doi.org/10.1016/j.cell.2015.05.008

    • PubMed
    • Google Scholar
44. 1. Perez-Riverol Y
    2. Csordas A
    3. Bai J
    4. Bernal-Llinares M
    5. Hewapathirana S
    6. Kundu DJ
    7. Inuganti A
    8. Griss J
    9. Mayer G
    10. Eisenacher M
    11. Pérez E
    12. Uszkoreit J
    13. Pfeuffer J
    14. Sachsenberg T
    15. Yılmaz Ş
    16. Tiwary S
    17. Cox J
    18. Audain E
    19. Walzer M
    20. Jarnuczak AF
    21. Ternent T
    22. Brazma A
    23. Vizcaíno JA

    (2017) The PRIDE database and related tools and resources in 2019: improving support for quantification data

    Nucleic Acids Research 47:D442–D450.

    https://doi.org/10.1093/nar/gky1106

    • PubMed
    • Google Scholar
45. 1. Peter D
    2. Ruscica V
    3. Bawankar P
    4. Weber R
    5. Helms S
    6. Valkov E
    7. Igreja C
    8. Izaurralde E

    (2019) Molecular basis for GIGYF–Me31B complex assembly in 4EHP-mediated translational repression

    Genes & Development 33:1355–1360.

    https://doi.org/10.1101/gad.329219.119

    • PubMed
    • Google Scholar
46. 1. Pochopien AA
    2. Beckert B
    3. Kasvandik S
    4. Berninghausen O
    5. Beckmann R
    6. Tenson T
    7. Wilson DN

    (2021) Structure of Gcn1 bound to stalled and colliding 80S ribosomes

    PNAS 118:e2022756118.

    https://doi.org/10.1073/pnas.2022756118

    • PubMed
    • Google Scholar
47. 1. Presnyak V
    2. Alhusaini N
    3. Chen Y-H
    4. Martin S
    5. Morris N
    6. Kline N
    7. Olson S
    8. Weinberg D
    9. Baker KE
    10. Graveley BR
    11. Coller J

    (2015) Codon optimality is a major determinant of MRNA stability

    Cell 160:1111–1124.

    https://doi.org/10.1016/j.cell.2015.02.029

    • PubMed
    • Google Scholar
48. 1. Radhakrishnan A
    2. Chen Y-H
    3. Martin S
    4. Alhusaini N
    5. Green R
    6. Coller J

    (2016) The DEAD-Box Protein Dhh1p Couples MRNA Decay and Translation by Monitoring Codon Optimality

    Cell 167:122–132.

    https://doi.org/10.1016/j.cell.2016.08.053

    • PubMed
    • Google Scholar
49. 1. Robert X
    2. Gouet P

    (2014) Deciphering key features in protein structures with the new ENDscript server

    Nucleic Acids Research 42:W320–W324.

    https://doi.org/10.1093/nar/gku316

    • PubMed
    • Google Scholar
50. 1. Saeed AI
    2. Sharov V
    3. White J
    4. Li J
    5. Liang W
    6. Bhagabati N
    7. Braisted J
    8. Klapa M
    9. Currier T
    10. Thiagarajan M
    11. Sturn A
    12. Snuffin M
    13. Rezantsev A
    14. Popov D
    15. Ryltsov A
    16. Kostukovich E
    17. Borisovsky I
    18. Liu Z
    19. Vinsavich A
    20. Trush V
    21. Quackenbush J

    (2003) TM4: A free, open-source system for microarray data management and analysis

    BioTechniques 34:374–378.

    https://doi.org/10.2144/03342mt01

    • PubMed
    • Google Scholar
51. 1. Saito K
    2. Horikawa W
    3. Ito K

    (2015) Inhibiting K63 polyubiquitination abolishes no-go type stalled translation surveillance in Saccharomyces cerevisiae

    PLOS Genetics 11:e1005197.

    https://doi.org/10.1371/journal.pgen.1005197

    • PubMed
    • Google Scholar
52. 1. Sezen B
    2. Seedorf M
    3. Schiebel E

    (2009) The SESA network links duplication of the yeast centrosome with the protein translation machinery

    Genes & Development 23:1559–1570.

    https://doi.org/10.1101/gad.524209

    • PubMed
    • Google Scholar
53. 1. Sharma P
    2. Yan F
    3. Doronina VA
    4. Escuin-Ordinas H
    5. Ryan MD
    6. Brown JD

    (2012) 2A peptides provide distinct solutions to driving stop-carry on translational recoding

    Nucleic Acids Research 40:3143–3151.

    https://doi.org/10.1093/nar/gkr1176

    • PubMed
    • Google Scholar
54. 1. Simms CL
    2. Kim KQ
    3. Yan LL
    4. Qiu JK
    5. Zaher HS

    (2018) Interactions between the mRNA and Rps3/uS3 at the entry tunnel of the ribosomal small subunit are important for no-go decay

    PLOS Genetics 14:e1007818.

    https://doi.org/10.1371/journal.pgen.1007818

    • PubMed
    • Google Scholar
55. 1. Sinha NK
    2. Ordureau A
    3. Best K
    4. Saba JA
    5. Zinshteyn B
    6. Sundaramoorthy E
    7. Fulzele A
    8. Garshott DM
    9. Denk T
    10. Thoms M
    11. Paulo JA
    12. Harper JW
    13. Bennett EJ
    14. Beckmann R
    15. Green R

    (2020) EDF1 coordinates cellular responses to ribosome collisions

    eLife 9:e58828.

    https://doi.org/10.7554/eLife.58828

    • PubMed
    • Google Scholar
56. 1. Sitron CS
    2. Park JH
    3. Brandman O

    (2017) Asc1, Hel2, and Slh1 couple translation arrest to nascent chain degradation

    RNA 23:798–810.

    https://doi.org/10.1261/rna.060897.117

    • PubMed
    • Google Scholar
57. 1. Subramanian A
    2. Tamayo P
    3. Mootha VK
    4. Mukherjee S
    5. Ebert BL
    6. Gillette MA
    7. Paulovich A
    8. Pomeroy SL
    9. Golub TR
    10. Lander ES
    11. Mesirov JP

    (2005) Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles

    PNAS 102:15545–15550.

    https://doi.org/10.1073/pnas.0506580102

    • PubMed
    • Google Scholar
58. 1. Sundaramoorthy E
    2. Leonard M
    3. Mak R
    4. Liao J
    5. Fulzele A
    6. Bennett EJ

    (2017) ZNF598 and RACK1 Regulate Mammalian Ribosome-Associated Quality Control Function by Mediating Regulatory 40S Ribosomal Ubiquitylation

    Molecular Cell 65:751–760.

    https://doi.org/10.1016/j.molcel.2016.12.026

    • PubMed
    • Google Scholar
59. 1. Sweet T
    2. Kovalak C
    3. Coller J

    (2012) The DEAD-box protein Dhh1 promotes decapping by slowing ribosome movement

    PLOS Biology 10:e1001342.

    https://doi.org/10.1371/journal.pbio.1001342

    • PubMed
    • Google Scholar
60. 1. Tesina P
    2. Lessen LN
    3. Buschauer R
    4. Cheng J
    5. Wu CC-C
    6. Berninghausen O
    7. Buskirk AR
    8. Becker T
    9. Beckmann R
    10. Green R

    (2020) Molecular mechanism of translational stalling by inhibitory codon combinations and poly(A) tracts

    The EMBO Journal 39:e103365.

    https://doi.org/10.15252/embj.2019103365

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

    (2012) Dom34:hbs1 plays a general role in quality-control systems by dissociation of a stalled ribosome at the 3’ end of aberrant MRNA

    Molecular Cell 46:518–529.

    https://doi.org/10.1016/j.molcel.2012.03.013

    • PubMed
    • Google Scholar
62. 1. Tyanova S
    2. Temu T
    3. Cox J

    (2016) The MaxQuant computational platform for mass spectrometry-based shotgun proteomics

    Nature Protocols 11:2301–2319.

    https://doi.org/10.1038/nprot.2016.136

    • PubMed
    • Google Scholar
63. Software

    1. Veltri AJ

    (2022) in-frame, version swh:1:rev:2f26e680e06d11fd517105d52f19005f00b5c7af

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:d7739f459817449dc41bb9cdc1b6d069a15e16d9;origin=https://github.com/greenlabjhmi/2022_syh1;visit=swh:1:snp:4f3b142d43a409e0b7dbaa15dca93148c5f9d306;anchor=swh:1:rev:2f26e680e06d11fd517105d52f19005f00b5c7af
64. 1. Wagih O
    2. Usaj M
    3. Baryshnikova A
    4. VanderSluis B
    5. Kuzmin E
    6. Costanzo M
    7. Myers CL
    8. Andrews BJ
    9. Boone CM
    10. Parts L

    (2013) SGAtools: one-stop analysis and visualization of array-based genetic interaction screens

    Nucleic Acids Research 41:W591–W596.

    https://doi.org/10.1093/nar/gkt400

    • PubMed
    • Google Scholar
65. 1. Wang J
    2. Zhou J
    3. Yang Q
    4. Grayhack EJ

    (2018) Multi-protein bridging factor 1(Mbf1), Rps3 and Asc1 prevent stalled ribosomes from frameshifting

    eLife 7:e39637.

    https://doi.org/10.7554/eLife.39637

    • PubMed
    • Google Scholar
66. 1. Weber R
    2. Chung MY
    3. Keskeny C
    4. Zinnall U
    5. Landthaler M
    6. Valkov E
    7. Izaurralde E
    8. Igreja C

    (2020) 4EHP and GIGYF1/2 Mediate Translation-Coupled Messenger RNA Decay

    Cell Reports 33:108262.

    https://doi.org/10.1016/j.celrep.2020.108262

    • PubMed
    • Google Scholar
67. 1. Webster MW
    2. Chen Y-H
    3. Stowell JAW
    4. Alhusaini N
    5. Sweet T
    6. Graveley BR
    7. Coller J
    8. Passmore LA

    (2018) MRNA Deadenylation Is Coupled to Translation Rates by the Differential Activities of Ccr4-Not Nucleases

    Molecular Cell 70:1089–1100.

    https://doi.org/10.1016/j.molcel.2018.05.033

    • Google Scholar
68. 1. Wu Q
    2. Medina SG
    3. Kushawah G
    4. DeVore ML
    5. Castellano LA
    6. Hand JM
    7. Wright M
    8. Bazzini AA

    (2019a) Translation affects MRNA stability in a codon-dependent manner in human cells

    eLife 8:e45396.

    https://doi.org/10.7554/eLife.45396

    • PubMed
    • Google Scholar
69. 1. Wu CCC
    2. Zinshteyn B
    3. Wehner KA
    4. Green R

    (2019b) High-resolution ribosome profiling defines discrete ribosome elongation states and translational regulation during cellular stress

    Molecular Cell 73:959–970.

    https://doi.org/10.1016/j.molcel.2018.12.009

    • PubMed
    • Google Scholar
70. 1. Wu CC-C
    2. Peterson A
    3. Zinshteyn B
    4. Regot S
    5. Green R

    (2020) Ribosome collisions trigger general stress responses to regulate cell fate

    Cell 182:404–416.

    https://doi.org/10.1016/j.cell.2020.06.006

    • PubMed
    • Google Scholar
71. 1. Xie Z
    2. Bailey A
    3. Kuleshov MV
    4. Clarke DJB
    5. Evangelista JE
    6. Jenkins SL
    7. Lachmann A
    8. Wojciechowicz ML
    9. Kropiwnicki E
    10. Jagodnik KM
    11. Jeon M
    12. Ma’ayan A

    (2021) Gene set knowledge discovery with enrichr

    Current Protocols 1:e90.

    https://doi.org/10.1002/cpz1.90

    • PubMed
    • Google Scholar
72. 1. Yan LL
    2. Simms CL
    3. McLoughlin F
    4. Vierstra RD
    5. Zaher HS

    (2019) Oxidation and alkylation stresses activate ribosome-quality control

    Nature Communications 10:5611.

    https://doi.org/10.1038/s41467-019-13579-3

    • PubMed
    • Google Scholar
73. 1. Yan LL
    2. Zaher HS

    (2019) How do cells cope with RNA damage and its consequences?

    The Journal of Biological Chemistry 294:15158–15171.

    https://doi.org/10.1074/jbc.REV119.006513

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Distinct ribosome states trigger diverse mRNA quality control pathways" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Alan Hinnebusch as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by James Manley as the Senior Editor. The reviewers have opted to remain anonymous.

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

Essential revisions:

There was agreement among the referees and reviewing editor that an important strength of the paper involves (i) the genetic data comparing the effects of different mutations on the NGD and COMD reporters, indicating that NGD and COMD involve largely distinct factors, with the exception of Not5 that functions in both pathways; and (ii) the ribosome profiling data on these reporters revealing that the appearance of collided ribosomes at a discrete stall site applies only to the NGD reporter.

1) However, there was also agreement that these profiling data are not adequate to conclude that the different ribosomal states are necessarily the triggers for NGD or COMD, but merely provide correlative data consistent with this possibility and, as such, the title of the paper and last sentence of the Abstract should be modified to more accurately reflect what these data actually show.

2) There are important concerns about the second major conclusion, that Syh1 is the key effector of the NGD pathway in wild-type cells, and the claim that new mechanistic insights have been provided into Syh1 function. The genetic data are quite complex, with some ostensibly contradictory results involving Hel2; and the data seem to point instead to a model that Syh1 functions in only one of two pathways with extensively overlapping functions in NGD, the other involving Hel2 and Cue2, with the Syh1 pathway being only somewhat more important than the Cue2 pathway, such that mRNA decay is affected when Syh1 is deleted from otherwise WT cells. The authors should consider including this alternative interpretation of the results.

3) There are additional genetic experiments that are likely needed to support aspects of the final model in Figure 6, including that Xrn1 functions only in the Syh1 pathway, and that the Cue2 pathway requires Hel2, as described in Ref. #1's comments. These requests for additional work are viewed as being justified, not only to provide stronger support for the final model, but also because some of the key observations have been published previously by this group or other labs, including the involvement of Syh1 itself in NGD. As such, the genetic analysis of syh1 mutations should be as thorough as possible.

4) A related issue is whether Smy2 might have an important overlapping function with Syh1, which should be addressed by determining whether there is a strong NGD phenotype on deleting Smy2 in a double mutant lacking Syh1 (see Ref. #3 comments).

5) The referees also took issue with the claim that new mechanistic insight into Syh1 function has been provided, as it remains unclear whether Syh1 is actually recruited by collided ribosomes. This question could be addressed biochemically, or perhaps genetically by combining syh1 mutations (actually a hel2 syh1 mutation) with mutations in GCN1 or MBF1. If the authors elect not to carry out these experiments, they should soften their claims about providing a mechanistic understanding of Syh1 function and indicate in the Discussion that additional work is needed to determine whether Syh1 is actually recruited to collided ribosomes in the manner depicted in Figure 6.

6) Finally, it was felt that the results of the reporter screen of the yeast deletion collection conducted using the COMD reporter should be presented with the list of mutations that tested positive in that screen, and to compare it to the list of mutations similarly identified using the NGD reporter, which would increase the scientific value of the study.

Reviewer #1 (Recommendations for the authors):

– The authors wish to conclude that Syh1 represents the major pathway for NGD in WT cells, and that Cue2 functions only when the ability of Slh1 to diminish collided ribosomes is overwhelmed. I wasn't convinced of this interpretation as it seems equally possible that the two pathways are continuously functioning in a largely redundant manner, but with Syh1 making a moderate unique contribution even when Cue2 is present, while being able to fully compensate for loss of the Cue2 pathway in cells lacking either Cue2 or Hel2. On lines 350-352, they argue that "in the syh1∆ strain, NGD reporter levels are somewhat rescued because recruitment of Xrn1 is impaired, but Cue2-mediated endonucleolytic decay plays a larger role as Slh1 becomes overwhelmed". Why would Slh1 become overwhelmed in the absence of Syh1? In fact, deleting SLH1 reduces minCGA expression in cells lacking SYH1, so it doesn't seem to be overwhelmed.

– There were some additional mutants that should be analyzed to support their final model in Figure 6. The syh1 xrn1 double mutant should be examined to show that the minCGA reporter will be derepressed identically as in each of the syh1 or xrn1 single mutants, as predicted by the conclusion that Syh1 is essential for turnover of the CGA reporter by Xrn1. (In fact, it appears that the GFP-OPT reporter is derepressed to a greater extent by deletion of XRN1 vs deletion of SYH1, at odds with the model.) In addition, the derepression observed in the xrn1 single mutant should be less than observed in the syh1 cue2 double mutant, as the latter should be defective for both the Syh1/Xrn1 and Hel2/Cue2 pathways.

– Is there any independent evidence that Cue2 function requires Hel2? If not, then a hel2 cue2 syh1 triple is required, which should phenocopy the hel2 syh1 and cue2 syh1 double mutants in the manner predicted if Hel2 and Cue2 function, as proposed, in the same pathway.

– If it was not shown previously for the minCGA reporter, it is important to verify that deleting CUE2 will lead to a large derepression of this reporter in cells deleted of SLH1.

– The data in Figure 2 show that deleting SLH1 reduces minCGA mRNA levels in the absence of SYH1. Does this mean that increased ribosome collisions in cells lacking Slh1 also enhance turnover through the Cue2/Hel2 pathway? The authors should address this finding.

– Results from quantifying the 3' frag of the minCGA reporter seem to suggest different conclusions compared to those derived from data in Figure 2. Can the authors explain why?

– It seems important to show that the deletion of NOT5 or XRN1 will increase expression of the NONOPT GFP reporter as a positive control for the Figure 3B results.

– Figure 1D and Figure 2C: the authors should perform tests and indicate whether mean values differ statistically in cases where it is not obvious that the means differ between mutant and WT or between two different mutants, where such differences would be important for their final model.

– The authors' explanation of the impact of hel2Δ mutation on collided ribosomes on NGD reporters is hard to follow. On the one hand, hel2Δ seems to elevate NGD by impairing ubiquitylation of the collided ribosomes, thereby increasing collisions and enhancing NGD. On the other hand, hel2Δ is predicted to impair NGD mediated by the Cue2 endonuclease. They also conclude that WT Hel2 stabilizes disomes but facilitates their removal at collision sites-if they are stabilized by Hel2 why do they get removed? Can the authors attempt to resolve these apparent discrepancies in the Discussion?

Reviewer #2 (Recommendations for the authors):

Suggestions for improved or additional experiments, data or analyses, especially those directed at increasing the impact of the work and making it suitable for eLife.

The experimental and analytical work presented is very good and generally supports the authors conclusions. I am not completely convinced that these conclusions will have an impact suitable for eLife, but I think it is arguable that they could.

1. It seems that the SGA screens did not identify any new factors in NGD or COMD, though this is not completely clear in the manuscript. Can you please specifically state whether or not the study identified new factors involved in these pathways?

2. The positive hits for the COMD screen (nonopt) are not described at all. This is a major limitation. There are many points in the Figure 3C volcano plot that look like strong hits. A comprehensive report on the SGA results from nonopt vs opt reporters could help increase the impact.

3. Moreover, Venn diagrams of shared hits from the NGD and COMD SGA experiments would help the reader discern the extent to which the pathways are distinct.

Reviewer #3 (Recommendations for the authors):

The manuscript by Veltri et al. describes the use of a genome-wide screen of non-essential yeast genes to identify genes that modulate expression of a reporter with a strong translational stall [GFP-HIS3-(CGA)12]. These studies provide additional evidence that Syh1 and Smy2 proteins (homologs of mammalian GIGYF1/2 proteins) as well as the ribosomal protein Asc1 modulate mRNA decay of a reporter with a strong translational stall site; deletion of these genes results in increased expression of the reporter. Surprisingly in this screen, deletions of canonical No Go Decay (NGD) factors HEL2, SLH1, CUE3 and RQT4 result in reduced expression of the same reporter (the opposite of their expected effects). The validity of SYH1, HEL2 and SLH1 mutant effects are demonstrated by reconstruction of the deletions and demonstration that both protein and mRNA are affected as expected. To begin to understand the interaction of Syh1 with the NGD pathway and possibly the basis for its recruitment to the stalling reporter, the authors make double mutants between SYH1 and the NGD components. They find that Syh1 function in mRNA decay does not require function of the major NGD regulator Hel2, of the NGD endonuclease Cue2 or of the ribosome disassociation factor Slh1. In fact, double mutants in which SYH1 is deleted in conjunction with either HEL2 or CUE2 result in substantially increased mRNA from a HIS3-CGA stalling reporter compared to the single SYH1 deletion (similar to mRNA levels of HIS3 reporter without the stall site). By contrast, none of these factors affect mRNA stability of a HIS3 reporter in which translation elongation is likely uniformly slowed by suboptimal codons (HIS3-NONOPT). However, as expected, HIS3-NONOPT is a target of COMD (codon optimality mediated decay) and is stabilized by deletion of NOT5. In fact, the strong stalling reporter, HIS3-CGA, is also stabilized in the NOT5 deletion, in a manner quantitatively similar to its stabilization by the SYH1 deletion. Thus, mRNA stability of the strong stalling reporter is regulated by both NGD and COMD. To understand the molecular basis for recruitment of distinct decay factors, the authors investigate the ribosome states on these reporters using ribosome profiling. In both the OPT and NONOPT reporters, single ribosomes and disomes (collided ribosomes) are uniformly positioned across the coding sequence, while in the CGA reporter, ribosomes are absent from the region downstream of the CGA repeats and disomes are substantially increased and stacked at the CGA repeat (compared to the OPT reporter). In addition, ribosomes lacking tRNA from the A site are enriched on the NONOPT reporter and with the CGA codon repeat on the CGA reporter. The authors infer that the distinct ribosome signals of collided disomes or ribosomes with empty A sites are strong determinants of the factors that lead to mRNA decay.

The paper does not meet the claim in the abstract and body of the paper that Syh1 acts as the primary link to mRNA decay in NGD nor does it provide new mechanistic insight into role of Syh1 in NGD (line 23).

1. It is unclear if Syh1 primarily acts in the NGD pathway, basically it is unclear if the (CGA)12 reporters are primarily targeted by NGD since the deletions of the known NGD factors either have no effect or cause decreased stability of the (CGA)12 reporters. For instance, see line 155 "This focused analysis of protein and RNA reporter levels in these different strains validates results from the initial screen and indicates that the CGA reporter is being strongly regulated by canonical NGD machinery." However, the results with Hel2 and Slh1 are not the expected results of deleting factors with a key role in NGD (and moreover the HEL2 effects are not recapitulated with the HIS3 minireporter). The authors should address this discrepancy, at least by citing the evidence that this reporter is an NGD substrate for Cue2 (in D'Orazio et al).

2. The genetic epistasis further implies that Syh1 does not act as the primary nuclease in the NGD pathway. Figure 2- effects of syh1∆ on mRNA from HIS3-CGA12 reporter is less than a 2-fold effect, but double deletions syh1∆ hel2∆ and syh1∆ cue2∆ result in >7-fold increase in mRNA (close to reporter lacking CGA codons). Why conclude that Syh1 is the primary NGD factor? Genetics here indicates two redundant pathways-if one inactivates both, then the mRNA is stable. Primarily this result is correctly discussed in the Discussion, but is not really proven (see 3)

3. The choice to focus on only SYH1 (because its effects were stronger than SMY2) is questionable. It is possible to miss many scenarios in which one protein substitutes for the other or to miss the full spectrum of their importance. Since the most novel result in this manuscript is the involvement of Syh1 in NGD, it seems incumbent to test the homologous gene which also affects NGD. At least examine the effects of a double reconstructed mutant syh1∆ smy2∆ double deletions (with and without deletion of hel2) and show the effects of the single reconstructed smy2∆.

4. It is unclear what type of mechanistic insight is provided since the manuscript does not supply any information on the mechanism by which Syh1 is recruited to the target mRNA nor its molecular mode of action. The test of the two hypotheses about the recruitment are not complete. The deletion of MBF1 might activate the Cue2-dependent pathway, resulting in apparent low levels of GFP/RFP protein. However, there is no simple way to prove this point. If this were true, a double deletion of mbf1 and hel2 would completely stabilize the mRNA. This experiment or the one below are valuable because they provide some additional evidence that Syh1 is recruited by collided ribosomes.

5. The authors do not consider the possibility that Syh1 is recruited to collided ribosomes by Gcn1 or other components that bind collided ribosomes with Gcn1 (Gcn20, Rbg2, Gir2). Syh1 recruitment could be mediated by either the Gcn1 complex or Hel2 or both. Given the papers from the Guydosh and Zaher labs documenting the competition between the NGD and GAAC pathways (as well as the Gcn1 complex on collided ribosome Pochopien and Wilson paper), it seems incumbent upon the authors to test this possibility. This reviewer did note that GCN1, GCN20… deletions did not result in increased expression (high throughput data). It is possible that Syh1 can be recruited by either Gcn1 or Hel2 (or that the mutants in the deletion collection have acquired suppressors- a known problem with the collection.) Either of these points, might also be addressed with biochemical evidence that Syh1 recognizes collided ribosomes.

6. The paper provides insufficient evidence to meet the claim in the title and body to "define the molecular triggers that determine how distinct pathways target mRNAs for degradation in yeast (line 24)." There are multiple differences in the ribosome profiling of the different reporters, but no evidence of a causal relationship. One might consider simply revising the title and softening the claim here. The claim that ribosome collisions are key to Syh1 recruitment could be tested either by reducing ribosome collisions on a CGA reporter with different 5' UTRs (see Simms and Zaher 2019) or by reducing global initiation. The effects of the SYH1 deletion on these reporters could be measured as well as ribosome profiling- it is unclear if one would also need to reduce the number of CGA codons. (The authors might also look at the mass spec enrichment of two subunits of eIF3 in the pull down of Syh1).

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Distinct elongation stalls during translation trigger distinct pathways for mRNA degradation" for further consideration by eLife. Your revised article has been evaluated by James Manley (Senior Editor) and a Reviewing Editor.

The revised manuscript is much improved; however, there are a few remaining issues.

First, Figure 2 and its description still have some difficulties:

– Lines 203-205 and lines 398-399: these statements are not justified as the difference between WT and syh1 is not statistically significant (as shown in File S5). In fact, the syh1 mutation only has a strong effect on the minCGA reporter in cells lacking HEL2 or CUE2, underscoring the extensive functional redundancy between the Syh1 and Hel2/Cue pathways, at least for this reporter, which should be more fully acknowledged at these two locations as is done later on in the DISCUSSION. The Abstract would also be more accurate/informative if it stated explicitly the functional redundancy between the Syh1 and Hel2/Cue2 pathways.

– Lines 205-210 and lines 401-402: this description is not accurate as neither the syh1 nor smy2 mutations significantly rescued the mRNA level vs. WT, whereas the double mutation does. These findings don't justify a focus on Syh1 vs Smy2, nor exclusion of Smy2 from their model for NGD in WT cells. The greater importance of Syh1 vs. Smy2 emerges only in the next paragraph when the hel2 syh1 and hel2 smy2 double mutants are compared. Moreover, because the difference in mRNA level between hel2smy2 vs hel2 is significant, it should be acknowledged that Smy2 and Syh1 have overlapping functions with Syh1 playing merely the greater role. Including the latter point in the Abstract is also recommended.

– Lines 244-247: these data seem to be overinterpreted as there are no statistically significant differences in mRNA levels between the syh1mbf1 strain and either the syh1 or mbf1 single mutant strains. Thus, it would be safer to conclude that there is no evidence from these assays that Mbf1 is involved in either the Hel2/Cue2 or Syh1/Smy2 pathways.

– Overall, the Abstract and Discussion would be more accurate/informative if they conveyed the functional overlaps between Syh1 and Smy2, as well as the redundancy between the Syh1/Smy2 and Hel2/Cue2 pathways in NGD. The revised Discussion does do a good job of explaining this issue regarding Syh1 vs Hel2/Cue2, but it would be even better to acknowledge a significant, even if minor, contribution of Smy2 to the Syh1 pathway.

Second, it would be better to modify the title of the paper to replace the word "trigger" with "are associated (or linked) with", to fall in line with the wording changes made in the Abstract that softened this conclusion in the revised version of the paper.

Essential revisions:

There was agreement among the referees and reviewing editor that an important strength of the paper involves (i) the genetic data comparing the effects of different mutations on the NGD and COMD reporters, indicating that NGD and COMD involve largely distinct factors, with the exception of Not5 that functions in both pathways; and (ii) the ribosome profiling data on these reporters revealing that the appearance of collided ribosomes at a discrete stall site applies only to the NGD reporter.

1) However, there was also agreement that these profiling data are not adequate to conclude that the different ribosomal states are necessarily the triggers for NGD or COMD, but merely provide correlative data consistent with this possibility and, as such, the title of the paper and last sentence of the Abstract should be modified to more accurately reflect what these data actually show.

We agree that the ribosome profiling data do not directly demonstrate a connection between ribosome state and NGD / COMD and have altered the manuscript to reflect the correlative nature of our conclusions regarding ribosome states triggering mRNA decay.

2) There are important concerns about the second major conclusion, that Syh1 is the key effector of the NGD pathway in wild-type cells, and the claim that new mechanistic insights have been provided into Syh1 function. The genetic data are quite complex, with some ostensibly contradictory results involving Hel2; and the data seem to point instead to a model that Syh1 functions in only one of two pathways with extensively overlapping functions in NGD, the other involving Hel2 and Cue2, with the Syh1 pathway being only somewhat more important than the Cue2 pathway, such that mRNA decay is affected when Syh1 is deleted from otherwise WT cells. The authors should consider including this alternative interpretation of the results.

We agree that the claim that Syh1 is the major pathway in NGD was too strong. Our revised manuscript argues that the data can be explained by the presence of parallel NGD pathways dependent on either Syh1 or Hel2. We have altered the language in the results and Discussion sections to reflect this possibility and describe the parallel activities of Slh1/Cue2 in compensating for loss of Syh1.

3) There are additional genetic experiments that are likely needed to support aspects of the final model in Figure 6, including that Xrn1 functions only in the Syh1 pathway, and that the Cue2 pathway requires Hel2, as described in Ref. #1's comments. These requests for additional work are viewed as being justified, not only to provide stronger support for the final model, but also because some of the key observations have been published previously by this group or other labs, including the involvement of Syh1 itself in NGD. As such, the genetic analysis of syh1 mutations should be as thorough as possible.

The dependence of Cue2 function on Hel2 was previously established in D’Orazio et al. 2019 and Ikeuchi et al. 2019a and nicely reviewed in D’Orazio et al. 2021a. These studies show that Cue2-dependent endonuclease cleavages are lost when Hel2 is deleted. Furthermore, D’Orazio et al. 2019 established that Cue2 activity is increased in genetic backgrounds where Slh1 is deleted; we interpreted this to mean that Slh1 (known to function as a helicase) normally ‘rescues’ collided ribosomes, but that Cue2 acts as a failsafe mechanism that cleaves mRNAs when stalls are not resolved by Slh1. Of direct relevance to this manuscript, we suggest that the decrease of minCGA mRNA levels observed in the SLH1 deletion background is the consequence of increased cleavage by Cue2. Indeed, when CUE2 is knocked out in the slh1∆syh1 background (i.e. the triple mutant cue2∆slh1∆syh1∆) there is a strong rescue of reporter levels, highlighting the role of CUE2 in cleaving these mRNAs (Figure 2C). We have updated the manuscript to more explicitly describe the (previously established) dependence of Cue2 on Hel2 and to clarify our genetic argument based on the previous literature.

We agree with the reviewer that double knockouts of XRN1 and NGD factor genes are important experiments to establish the direct connection between Syh1- and Xrn1-mediated decay. In an effort to address these points, we constructed additional strains with the minOPT and minCGA reporters including xrn1∆, syh1∆xrn1∆, and hel2∆xrn1∆. While we were able to build these strains, we were concerned by their extremely impaired growth rate and thus their potential to pick up suppressor mutations over the course of genetic manipulation and reporter insertion. Indeed, we observed inconsistent control reporter levels between biological replicates in these strains and therefore are not able to make conclusions based on these experiments. We have altered the language throughout the manuscript to soften our claims about a direct connection between Syh1 activity and Xrn1-mediated RNA decay.

4) A related issue is whether Smy2 might have an important overlapping function with Syh1, which should be addressed by determining whether there is a strong NGD phenotype on deleting Smy2 in a double mutant lacking Syh1 (see Ref. #3 comments).

We agree that further justification of our choice to focus on SYH1 was warranted. We performed the suggested experiments including SMY2 knockouts and double mutants (now Figure 2—figure supplement 1B, shown here). The results of this experiment indicate that, while SMY2 knockout does increase minCGA reporter levels in a manner consistent with our previously published data (Hickey, et al. 2020), the strong rescue of minCGA reporter levels observed in the hel2∆syh1∆ strain is not observed in a hel2∆smy2∆ strain and the hel2∆syh1∆smy2∆ strain shows only a small increase compared to the hel2∆smy2∆ strain. Thus, while Smy2 does appear to contribute to decay in a similar fashion to Syh1, its contribution appears to be relatively minor, thus justifying our focus on Syh1.

5) The referees also took issue with the claim that new mechanistic insight into Syh1 function has been provided, as it remains unclear whether Syh1 is actually recruited by collided ribosomes. This question could be addressed biochemically, or perhaps genetically by combining syh1 mutations (actually a hel2 syh1 mutation) with mutations in GCN1 or MBF1. If the authors elect not to carry out these experiments, they should soften their claims about providing a mechanistic understanding of Syh1 function and indicate in the Discussion that additional work is needed to determine whether Syh1 is actually recruited to collided ribosomes in the manner depicted in Figure 6.

We agree with the reviewers. In order to more fully address this point, we performed an additional experiment (see Author response image 1) to assay recruitment of tagged Syh1 protein to collided ribosomes. We used an optimized concentration of the translation elongation inhibitor cycloheximide to induce global ribosome collisions. Then we performed sucrose gradient fractionation and immunoblotting to look for the recruitment of tagged Syh1 protein to the ribosome collisions. In this assay, we did not observe a substantial deep shift in the distribution of Syh1 in the samples with induced ribosome collisions. This may indicate that Syh1 is only transiently recruited to ribosome collisions or that it is not directly recruited to ribosome collisions. Further experiments are needed to determine the nature of potential physical interactions between the ribosome and Syh1 in the context of NGD. We have updated the discussion to indicate that further work is needed on this front.

To test the effects of Mbf1 on minCGA reporter levels in combination with Syh1, we performed additional genetic experiments, including the construction of MBF1 double knockouts with HEL2 and SYH1 (Figure S2E, shown below). If Syh1 is recruited to ribosome collisions by Mbf1 in yeast in a similar manner to the recruitment of GIGYF2 by EDF1 in mammalian cells, we expect that MBF1 deletion strains will behave similarly to SYH1 deletion strains, across genetic backgrounds. While the mbf1∆ strain does increase minCGA reporter mRNA levels to a similar modest extent as syh1∆, we note that the hel2∆mbf1∆ strain does not reveal the same strong increase in reporter levels as the hel2∆syh1∆ strain. In fact, hel2∆mbf1∆ minCGA mRNA levels are similar to those of mbf1∆ alone. Additionally, reporter levels are moderately increased in the syh1∆mbf1∆ strain compared to mbf1∆ or syh1∆ alone, suggesting an additive effect of these knockouts and potential parallel pathways (similar to the genetic interaction between SYH1 and HEL2). These results raise the interesting possibility that Mbf1 may actually function as part of the Hel2-dependent NGD pathway rather than in Syh1-dependent NGD. We have included these new data and a more careful discussion surrounding these ideas.

Author response image 1

Download asset Open asset

6) Finally, it was felt that the results of the reporter screen of the yeast deletion collection conducted using the COMD reporter should be presented with the list of mutations that tested positive in that screen, and to compare it to the list of mutations similarly identified using the NGD reporter, which would increase the scientific value of the study.

We have included further description of the top screen hits in the text and have included an additional sheet in supplementary table 1 describing the screen hits and the overlap.

Reviewer #1 (Recommendations for the authors):

– The authors wish to conclude that Syh1 represents the major pathway for NGD in WT cells, and that Cue2 functions only when the ability of Slh1 to diminish collided ribosomes is overwhelmed. I wasn't convinced of this interpretation as it seems equally possible that the two pathways are continuously functioning in a largely redundant manner, but with Syh1 making a moderate unique contribution even when Cue2 is present, while being able to fully compensate for loss of the Cue2 pathway in cells lacking either Cue2 or Hel2. On lines 350-352, they argue that "in the syh1∆ strain, NGD reporter levels are somewhat rescued because recruitment of Xrn1 is impaired, but Cue2-mediated endonucleolytic decay plays a larger role as Slh1 becomes overwhelmed". Why would Slh1 become overwhelmed in the absence of Syh1? In fact, deleting SLH1 reduces minCGA expression in cells lacking SYH1, so it doesn't seem to be overwhelmed.

See full response to editor above.

– There were some additional mutants that should be analyzed to support their final model in Figure 6. The syh1 xrn1 double mutant should be examined to show that the minCGA reporter will be derepressed identically as in each of the syh1 or xrn1 single mutants, as predicted by the conclusion that Syh1 is essential for turnover of the CGA reporter by Xrn1. (In fact, it appears that the GFP-OPT reporter is derepressed to a greater extent by deletion of XRN1 vs deletion of SYH1, at odds with the model.) In addition, the derepression observed in the xrn1 single mutant should be less than observed in the syh1 cue2 double mutant, as the latter should be defective for both the Syh1/Xrn1 and Hel2/Cue2 pathways.

See full response to editor above.

– Is there any independent evidence that Cue2 function requires Hel2? If not, then a hel2 cue2 syh1 triple is required, which should phenocopy the hel2 syh1 and cue2 syh1 double mutants in the manner predicted if Hel2 and Cue2 function, as proposed, in the same pathway.

– If it was not shown previously for the minCGA reporter, it is important to verify that deleting CUE2 will lead to a large derepression of this reporter in cells deleted of SLH1.

– The data in Figure 2 show that deleting SLH1 reduces minCGA mRNA levels in the absence of SYH1. Does this mean that increased ribosome collisions in cells lacking Slh1 also enhance turnover through the Cue2/Hel2 pathway? The authors should address this finding.

As discussed above, previous studies established that Cue2 is fully dependent on Hel2 for function and that in the absence of Slh1 that the role of Cue2 increases. These studies used similar reporters containing CGA12 repeats to demonstrate these effects (D’Orazio et al. 2019 and Ikeuchi et al. 2019a).

– Results from quantifying the 3' frag of the minCGA reporter seem to suggest different conclusions compared to those derived from data in Figure 2. Can the authors explain why?

We believe the model from D’Orazio et al. 2019 explains both the full length and 3’ fragment reporter data for the minCGA reporter. The 3’ reporter fragment can result from two different sources: (1) 5’ to 3’ degradation by Xrn1 that is blocked by stalled ribosomes near the stall site and (2) cleavage by Cue2 between collided ribosomes and slow clearing of ribosomes downstream of cleavage. Cue2 activity depends on Hel2 and normally contributes little to decay of the reporter, except in backgrounds where Slh1 (and thus ribosome clearing) is impaired. This explains why there is little accumulation of the 3’ fragment, except in backgrounds where Slh1 is deleted (slh1∆, slh1∆syh1∆) and thus where Cue2 plays a dominant role -- the bands in this case are a combination of Cue2 cleavage fragments and Xrn1-produced fragments. The triple knockout cue2∆slh1∆syh1∆ strain has reduced 3’ fragment accumulation compared to the slh1∆syh1∆ strain due to the loss of cleavage by Cue2, but the modest contribution of Xrn1 to 3’ fragment production is still present, as stalled ribosomes are not cleared. We have added additional clarification to these points in the manuscript to highlight the consistency of these results with the full length minCGA reporter results.

– It seems important to show that the deletion of NOT5 or XRN1 will increase expression of the NONOPT GFP reporter as a positive control for the Figure 3B results.

While we agree that this would be a clarifying detail to report, the reality is more complicated. Despite the stabilization of the GFP-NONOPT reporters that we (and others) show in various genetic backgrounds (NOT4, XRN1 and DHH1), the mRNA levels of these reporters are not increased in these same backgrounds. This has long been interpreted to reflect some sort of compensation by the cell, likely transcriptional, to maintain mRNA levels in an independent manner (Bregman, et al., Cell, 2011, DOI: 10.1016/j.cell.2011.12.005). This complexity was beyond the scope of this study, but likely explains the lack of interesting candidates to emerge in the GFP screen based on steady state expression levels of the GFP-NONOPT reporter. See Author response image 2.

Author response image 2

Download asset Open asset

– Figure 1D and Figure 2C: the authors should perform tests and indicate whether mean values differ statistically in cases where it is not obvious that the means differ between mutant and WT or between two different mutants, where such differences would be important for their final model.

We have included statistical tests for key data where indicated in the figures, and included full pairwise statistics in a supplemental table.

– The authors' explanation of the impact of hel2Δ mutation on collided ribosomes on NGD reporters is hard to follow. On the one hand, hel2Δ seems to elevate NGD by impairing ubiquitylation of the collided ribosomes, thereby increasing collisions and enhancing NGD. On the other hand, hel2Δ is predicted to impair NGD mediated by the Cue2 endonuclease. They also conclude that WT Hel2 stabilizes disomes but facilitates their removal at collision sites-if they are stabilized by Hel2 why do they get removed? Can the authors attempt to resolve these apparent discrepancies in the Discussion?

Based on the literature and our data, we suggest that the role of Hel2 is to stabilize the collided ribosome interface which is important for recruitment of downstream factors Slh1 and Cue2, explaining both the importance of Hel2 for promoting NGD and our interpretation that it stabilizes the disome structure. We suggest that in the HEL2 deletion strain there are decreased levels of the CGA reporter (in our R-SGA experiment and in the follow up reporter experiments) because of the compensatory effects of the Syh1-mediated NGD pathway. Additional language clarifying these views has been added to the discussion.

Reviewer #2 (Recommendations for the authors):

Suggestions for improved or additional experiments, data or analyses, especially those directed at increasing the impact of the work and making it suitable for eLife.

The experimental and analytical work presented is very good and generally supports the authors conclusions. I am not completely convinced that these conclusions will have an impact suitable for eLife, but I think it is arguable that they could.

1. It seems that the SGA screens did not identify any new factors in NGD or COMD, though this is not completely clear in the manuscript. Can you please specifically state whether or not the study identified new factors involved in these pathways?

2. The positive hits for the COMD screen (nonopt) are not described at all. This is a major limitation. There are many points in the Figure 3C volcano plot that look like strong hits. A comprehensive report on the SGA results from nonopt vs opt reporters could help increase the impact.

3. Moreover, Venn diagrams of shared hits from the NGD and COMD SGA experiments would help the reader discern the extent to which the pathways are distinct.

We have clarified in the Results section the high-confidence hits that emerged from our NGD screen and follow up validation. In particular, 14 of our top hits are now explicitly named in the text with a brief description of their function, and a full list of the 18 high-confidence hits from our validation are presented in an additional supplemental table. As the reviewer points out, many of these hits have previously been implicated in NGD, while the GIGY proteins are novel hits from this screen.

The COMD screen identified no strong hits that were validated, likely because even factors that might stabilize these mRNAs (such as Dhh1 and Not5) do not generally increase their steady state levels (discussed above). Nevertheless, we have added further analysis of the COMD screen, including GSEA analysis and presentation of the screen results in a supplementary table. We have also included additional Venn diagrams such as Figure 3—figure supplement 1 to help compare the CGA, COMD, and AAA reporter screens. There is very limited overlap between the top genes from the COMD screen with either the CGA or AAA screen hits.

Reviewer #3 (Recommendations for the authors):

The manuscript by Veltri et al. describes the use of a genome-wide screen of non-essential yeast genes to identify genes that modulate expression of a reporter with a strong translational stall [GFP-HIS3-(CGA)12]. These studies provide additional evidence that Syh1 and Smy2 proteins (homologs of mammalian GIGYF1/2 proteins) as well as the ribosomal protein Asc1 modulate mRNA decay of a reporter with a strong translational stall site; deletion of these genes results in increased expression of the reporter. Surprisingly in this screen, deletions of canonical No Go Decay (NGD) factors HEL2, SLH1, CUE3 and RQT4 result in reduced expression of the same reporter (the opposite of their expected effects). The validity of SYH1, HEL2 and SLH1 mutant effects are demonstrated by reconstruction of the deletions and demonstration that both protein and mRNA are affected as expected. To begin to understand the interaction of Syh1 with the NGD pathway and possibly the basis for its recruitment to the stalling reporter, the authors make double mutants between SYH1 and the NGD components. They find that Syh1 function in mRNA decay does not require function of the major NGD regulator Hel2, of the NGD endonuclease Cue2 or of the ribosome disassociation factor Slh1. In fact, double mutants in which SYH1 is deleted in conjunction with either HEL2 or CUE2 result in substantially increased mRNA from a HIS3-CGA stalling reporter compared to the single SYH1 deletion (similar to mRNA levels of HIS3 reporter without the stall site). By contrast, none of these factors affect mRNA stability of a HIS3 reporter in which translation elongation is likely uniformly slowed by suboptimal codons (HIS3-NONOPT). However, as expected, HIS3-NONOPT is a target of COMD (codon optimality mediated decay) and is stabilized by deletion of NOT5. In fact, the strong stalling reporter, HIS3-CGA, is also stabilized in the NOT5 deletion, in a manner quantitatively similar to its stabilization by the SYH1 deletion. Thus, mRNA stability of the strong stalling reporter is regulated by both NGD and COMD. To understand the molecular basis for recruitment of distinct decay factors, the authors investigate the ribosome states on these reporters using ribosome profiling. In both the OPT and NONOPT reporters, single ribosomes and disomes (collided ribosomes) are uniformly positioned across the coding sequence, while in the CGA reporter, ribosomes are absent from the region downstream of the CGA repeats and disomes are substantially increased and stacked at the CGA repeat (compared to the OPT reporter). In addition, ribosomes lacking tRNA from the A site are enriched on the NONOPT reporter and with the CGA codon repeat on the CGA reporter. The authors infer that the distinct ribosome signals of collided disomes or ribosomes with empty A sites are strong determinants of the factors that lead to mRNA decay.

The paper does not meet the claim in the abstract and body of the paper that Syh1 acts as the primary link to mRNA decay in NGD nor does it provide new mechanistic insight into role of Syh1 in NGD (line 23).

1. It is unclear if Syh1 primarily acts in the NGD pathway, basically it is unclear if the (CGA)12 reporters are primarily targeted by NGD since the deletions of the known NGD factors either have no effect or cause decreased stability of the (CGA)12 reporters. For instance, see line 155 "This focused analysis of protein and RNA reporter levels in these different strains validates results from the initial screen and indicates that the CGA reporter is being strongly regulated by canonical NGD machinery." However, the results with Hel2 and Slh1 are not the expected results of deleting factors with a key role in NGD (and moreover the HEL2 effects are not recapitulated with the HIS3 minireporter),. The authors should address this discrepancy, at least by citing the evidence that this reporter is an NGD substrate for Cue2 (in D'Orazio et al).

We have included additional references to evidence from D’Orazio et al. 2019 supporting the targeting of our CGA-repeat-containing reporters by NGD. Further, we have clarified in the text our interpretation that loss of HEL2 and SLH1 decrease CGA reporter levels because of the presence of multiple parallel decay pathways (i.e. SYH1-mediated decay and CUE2-mediated decay). We have also clarified our interpretation of the apparent discrepancy between the CGA and minCGA reporters in the hel2∆ strain. While loss of HEL2 has a reduced effect on the minCGA reporter, we believe this may be the result of different stalling contributions of the P2A-containing CGA reporter (the P2A sequence itself may cause ribosome stalling to some extent). Importantly, we do observe a decrease of minCGA reporter levels in the slh1∆ strain, consistent with the CGA reporter and our previous work demonstrating increased activity of the endonuclease Cue2 on NGD reporters in an slh1∆ strain (Figure 2C). Consistent with this interpretation, the levels of the 3’ fragment of the minCGA reporter (resulting from Cue2 cleavage and 5’ to 3’ decay) are greatly increased in the slh1∆ strain (Figure 2—figure supplement 1A).

2. The genetic epistasis further implies that Syh1 does not act as the primary nuclease in the NGD pathway. Figure 2- effects of syh1∆ on mRNA from HIS3-CGA12 reporter is less than a 2-fold effect, but double deletions syh1∆ hel2∆ and syh1∆ cue2∆ result in >7-fold increase in mRNA (close to reporter lacking CGA codons). Why conclude that Syh1 is the primary NGD factor? Genetics here indicates two redundant pathways-if one inactivates both, then the mRNA is stable. Primarily this result is correctly discussed in the Discussion, but is not really proven (see 3)

We agree with the reviewer’s argument and have altered our conclusions to reflect the redundant and compensatory nature of the Syh1 and Hel2/Cue2 mediated NGD pathways.

3. The choice to focus on only SYH1 (because its effects were stronger than SMY2) is questionable. It is possible to miss many scenarios in which one protein substitutes for the other or to miss the full spectrum of their importance. Since the most novel result in this manuscript is the involvement of Syh1 in NGD, it seems incumbent to test the homologous gene which also affects NGD. At least examine the effects of a double reconstructed mutant syh1∆ smy2∆ double deletions (with and without deletion of hel2) and show the effects of the single reconstructed smy2∆.

We performed an additional set of experiments with SMY2 knockouts to support our choice in focusing on Syh1. See full response to editor above.

4. It is unclear what type of mechanistic insight is provided since the manuscript does not supply any information on the mechanism by which Syh1 is recruited to the target mRNA nor its molecular mode of action. The test of the two hypotheses about the recruitment are not complete. The deletion of MBF1 might activate the Cue2-dependent pathway, resulting in apparent low levels of GFP/RFP protein. However, there is no simple way to prove this point. If this were true, a double deletion of mbf1 and hel2 would completely stabilize the mRNA. This experiment or the one below are valuable because they provide some additional evidence that Syh1 is recruited by collided ribosomes.

See full response to editor above.

5. The authors do not consider the possibility that Syh1 is recruited to collided ribosomes by Gcn1 or other components that bind collided ribosomes with Gcn1 (Gcn20, Rbg2, Gir2). Syh1 recruitment could be mediated by either the Gcn1 complex or Hel2 or both. Given the papers from the Guydosh and Zaher labs documenting the competition between the NGD and GAAC pathways(as well as the Gcn1 complex on collided ribosome Pochopien and Wilson paper), it seems incumbent upon the authors to test this possibility. This reviewer did note that GCN1, GCN20… deletions did not result in increased expression (high throughput data). It is possible that Syh1 can be recruited by either Gcn1 or Hel2 (or that the mutants in the deletion collection have acquired suppressors- a known problem with the collection.) Either of these points, might also be addressed with biochemical evidence that Syh1 recognizes collided ribosomes.

See full response to editor above.

6. The paper provides insufficient evidence to meet the claim in the title and body to "define the molecular triggers that determine how distinct pathways target mRNAs for degradation in yeast (line 24)." There are multiple differences in the ribosome profiling of the different reporters, but no evidence of a causal relationship. One might consider simply revising the title and softening the claim here. The claim that ribosome collisions are key to Syh1 recruitment could be tested either by reducing ribosome collisions on a CGA reporter with different 5' UTRs (see Simms and Zaher 2019) or by reducing global initiation. The effects of the SYH1 deletion on these reporters could be measured as well as ribosome profiling- it is unclear if one would also need to reduce the number of CGA codons. (The authors might also look at the mass spec enrichment of two subunits of eIF3 in the pull down of Syh1).

See full response to editor above.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

The revised manuscript is much improved; however, there are a few remaining issues.

First, Figure 2 and its description still have some difficulties:

– Lines 203-205 and lines 398-399: these statements are not justified as the difference between WT and syh1 is not statistically significant (as shown in File S5). In fact, the syh1 mutation only has a strong effect on the minCGA reporter in cells lacking HEL2 or CUE2, underscoring the extensive functional redundancy between the Syh1 and Hel2/Cue pathways, at least for this reporter, which should be more fully acknowledged at these two locations as is done later on in the DISCUSSION. The Abstract would also be more accurate/informative if it stated explicitly the functional redundancy between the Syh1 and Hel2/Cue2 pathways.

We have updated the language in the indicated sections to be in agreement with the statistical tests performed on the reporter levels. We have also included language in the abstract highlighting the redundancy between the Syh1/Smy2 pathway and the Hel2-mediated pathway.

– Lines 205-210 and lines 401-402: this description is not accurate as neither the syh1 nor smy2 mutations significantly rescued the mRNA level vs. WT, whereas the double mutation does. These findings don't justify a focus on Syh1 vs Smy2, nor exclusion of Smy2 from their model for NGD in WT cells. The greater importance of Syh1 vs. Smy2 emerges only in the next paragraph when the hel2 syh1 and hel2 smy2 double mutants are compared. Moreover, because the difference in mRNA level between hel2smy2 vs hel2 is significant, it should be acknowledged that Smy2 and Syh1 have overlapping functions with Syh1 playing merely the greater role. Including the latter point in the Abstract is also recommended.

We have made changes in these locations and in the abstract and discussion to more clearly recognize the contributions of Smy2 to NGD as indicated by our assays.

– Lines 244-247: these data seem to be overinterpreted as there are no statistically significant differences in mRNA levels between the syh1mbf1 strain and either the syh1 or mbf1 single mutant strains. Thus, it would be safer to conclude that there is no evidence from these assays that Mbf1 is involved in either the Hel2/Cue2 or Syh1/Smy2 pathways.

The conclusion that Mbf1 may be involved in the Hel2-mediated pathway has been removed to be in accordance with the statistical significance of the assays.

– Overall, the Abstract and Discussion would be more accurate/informative if they conveyed the functional overlaps between Syh1 and Smy2, as well as the redundancy between the Syh1/Smy2 and Hel2/Cue2 pathways in NGD. The revised Discussion does do a good job of explaining this issue regarding Syh1 vs Hel2/Cue2, but it would be even better to acknowledge a significant, even if minor, contribution of Smy2 to the Syh1 pathway.

As noted above, additional language has been added to the abstract to highlight the functional overlap between Syh1 and Smy2, as well as to emphasize the redundancy of the Syh1/Smy2-mediated and Hel2/Cue2-mediated pathways in NGD.

Second, it would be better to modify the title of the paper to replace the word "trigger" with "are associated (or linked) with", to fall in line with the wording changes made in the Abstract that softened this conclusion in the revised version of the paper.

The title has been updated to more correctly reflect the conclusions of our study.

Author details

1. Anthony J Veltri

   Department of Molecular Biology and Genetics, Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, United States

   Present address

   Versiti Blood Research Institute, Milwaukee, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Investigation, Visualization, Methodology, Writing – original draft, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7067-1796

2. Karole N D'Orazio

   Department of Molecular Biology and Genetics, Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, United States

   Present address

   Regeneron Pharmaceuticals, New York, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Writing – review and editing

   Competing interests

   is affiliated with Regeneron Pharmaceuticals

3. Laura N Lessen

   Department of Molecular Biology and Genetics, Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, United States

   Present address

   GlaxoSmithKline, Rockville, United States

   Contribution

   Data curation, Formal analysis, Investigation, Methodology, Writing – review and editing

   Competing interests

   is affiliated with GlaxoSmithKline

4. Raphael Loll-Krippleber

   Department of Biochemistry and Donnelly Centre, University of Toronto, Toronto, Canada

   Contribution

   Supervision, Investigation, Methodology, Project administration, Writing – review and editing

   Competing interests

   No competing interests declared

5. Grant W Brown

   Department of Biochemistry and Donnelly Centre, University of Toronto, Toronto, Canada

   Contribution

   Resources, Supervision, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9002-5003

6. Rachel Green

   Department of Molecular Biology and Genetics, Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, United States

   Contribution

   Conceptualization, Resources, Supervision, Funding acquisition, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   ragreen@jhmi.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9337-2003

• 2,262

  Page views

• 690

  Downloads

• 1

  Citations

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