CHARACTERIZING MOLECULAR PATHWAYS THAT RESCUE STALLED

RIBOSOMES AND DECAY PROBLEMATIC MESSENGER RNAS

by

Karole Nicole D’Orazio

A dissertation submitted to Johns Hopkins University in conformity with the

requirements for the degree of Doctor of Philosophy

Baltimore, Maryland

January 2020

Abstract

Translation through problematic sequences in mRNAs leads to ribosome collisions that trigger a collection of quality control events including ribosome rescue, degradation of the stalled nascent polypeptide via the Ribosome- mediated Quality control Complex (RQC), and targeting of the mRNA for decay

(No Go Decay or NGD). Using reverse genetic screens in yeast, we identify

Cue2 as the key endonuclease that is recruited to stalled ribosomes to promote

NGD. Following Cue2-mediated cleavage, ribosomes upstream of the cleavage site translate to the end of the truncated mRNA and are rescued by the

Dom34:Hbs1 complex. We also show that the putative helicase Slh1 reduces ribosome occupancy on intact problematic mRNAs and thereby reduces endonucleolytic cleavage by Cue2. The synergistic activities of Cue2 and Slh1 define two parallel pathways that allow cells to recognize and respond to ribosomes trapped on problematic mRNAs. From the same study in yeast, we also identify a set of novel candidate genes that act on NGD-substrates. In preliminary experiments we note that these factors are involved in many different processes involving translation elongation, deubiquitination processes, and mRNA decay and future work will further characterize these genes.

ii Rachel Green (Sponsor and reader), Professor, Department of Molecular Biology and Genetics, Johns Hopkins University School of Medicine

Brendan Cormack (Reader), Professor, Department of Molecular Biology and

Genetics, Johns Hopkins University School of Medicine

iii Preface

Acknowledgments

When I think about the people that gave me the chance to become a scientist, before I knew what being a scientist meant, I think of Dr. Ed Luk. He was a new professor at SUNY Stony Brook when I joined his lab and his patience, knowledge of biochemistry, mentorship skills, and incredible positivity and determination set the stones for my path to Johns Hopkins University. He helped me get into the graduate school of my dreams and has given me motivation all along the way.

It has been an incredible honor to train under the advisory of Dr. Rachel

Green in the Molecular Biology and Genetics Department at Johns Hopkins

University. Rachel’s passion for science and the enjoyment that she gets out of our work is a driving force for everyone in her lab. I want to thank her for sharing her brilliance with me and for sending me all over the world to carry out my experiments. She forced me to push the boundaries of my comfort zone, setting no limits on the type of science I could do or type of scientist I could be. She taught me to be fearless in both science and in life and I will be forever grateful for that. Rachel also promoted my work and me in a way that I never thought anyone would, giving me the confidence I needed to pursue a career in science and I will never forget that.

Thank you to my incredible lab members, past and present, who have taught me so much about how to be a scientist. Thank you specifically to Boris

Zinshteyn for teaching me to question everything, Laura Lessen for teaching me

iv mental strength, Daniel Goldman for teaching me patience, and Colin Wu for teaching me determination.

I am incredibly lucky to have such a dedicated and supportive thesis committee. Thank you to Dr. Scott Bailey for clearly having my back through every thesis committee meeting, Dr. Brendan Cormack for spending countless hours helping me with genetics and analyzing pathways, Dr. Carol Greider for treating me like her own lab member and Dr. Geraldine Seydoux for giving me a boost whenever I needed help with science or life decisions. They have all gone above and beyond to help me throughout my time here. Thank you to Dr. Grant

Brown as well for teaching me how to be a yeast geneticist and for trusting me to come to his lab and perform endless genetic screens.

My friends were there to go through the uncertainties and disappointments that come along with graduate school and to celebrate the successes that I never thought I would accomplish. Thank you to my friends in Baltimore: the ones who have been there since day one - Chris Cho, Kayarash Karimian, Joyce Lee, and

Chirag Vasavda; the ladies who lunch - Meiling May and Miriam Akeju; and the others I can always count on - Allison Daitch, Byron Ho, Maxime Cheve, and

Ryan McQuillen. Thank you to my friends on Long Island who bring so much happiness and balance to my life – Melissa Milhim, Rachael Patane, Nicole

Tollinchi, and Jessie Vitrano. Thank you to the Clarke and Zuniga families for supporting me like family.

v Thank you to my person, Jessica Zuniga. You help me through everything in life and I have no idea what I would do without you. My ride or die, number one

AP fan.

Most of all thank you to my parents Dominic and Carol D’Orazio. You provide me with never-ending support and the comfort that, wherever life takes me, I will always have my family by my side. Thank you to my extended family for providing me so much guidance, especially my cousin John for being my traveling partner and my Grandma Ann for being the strongest woman I know.

Thank you to my brothers and sisters and all of their children – Ann, Ryan,

Justin, Noah, Lucas and Avery Dumond; Max, Kristen, Max Jr, Anthony and

Alivia D’Orazio; Dominic, Michele, Nikki, and Isabel D’Orazio; Liz, Troy, Troy Jr.,

Dylan, and Evan Silva; Vito, Angela, Dominic and Angelena D’Orazio; Andrea

D’Orazio, Triny Alarcon, Liliana Alarcon, and Leo Alarcon; and Brianna D’Orazio

– for the love and encouragement you have given me and for being the most fun group of people to go through life with.

vi Dedication

I would like to dedicate this dissertation to Rosanna D’Orazio and our parents

Dominic and Carol D’Orazio, Colleen Clarke and her parents Kevin and Mary

Clarke, and Leo Alarcon and his parents Trinidad Alarcon and Andrea D’Orazio.

Thank you for motivating me to pursue a career research with the hopes of giving sick children and their families an easier road to travel.

vii Table of Contents

Abstract…………………………………………………………………………………...ii

Acknowledgments…………………………………………………………………..…..iv

List of Tables.……………………………………………………………………………xi

List of Figures…………………………………………………………………………..xii

Chapter I: Introduction……………………………………….………… ...... ……….1

1.1 Translation dependent control of cellular mRNA levels…….…………..2

1.2 Ribosome signaling decay of problematic mRNAs……………………..5

1.2.1 No Go Decay…………………….………………………………5

1.2.2 Non-Stop Decay…………………….…………………………..7

1.3 Thesis contributions…………………….…………………….…………….9

1.3.1 The endonuclease Cue2 cleaves mRNAs at stalled

ribosomes during No Go Decay……………………………….9

1.3.2 Deletion screens for factors involved in mRNA decay and

translation suppression in response to ribosome stalling…..9

Chapter II: The endonuclease Cue2 cleaves mRNAs at stalled ribosomes during

No Go Decay…………………….…………………….…………………….………..10

2.1 Introduction…………………….…………………….………………….…12

2.2 Results…………………….…………………….…………………….……16

2.2.1 Screening for factors involved in NGD……….……………….16

2.2.2 CUE2 domain structure and homology modeling……….…..18

2.2.3 Characterizing roles of CUE2 in NGD in vivo……….……….20

viii 2.2.4 Contribution of Cue2 to NGD is increased in specific genetic

backgrounds…..………………….……….……………………………21

2.2.5 Ribosome profiling provides high-resolution view of Cue2 cleavage sites…………………..……………………………………...23 2.2.6 In vitro reconstitution of Cue2 cleavage on isolated colliding ribosomes……….…………….……….……………………………….27

2.2.7 Ribosome profiling provides evidence that Slh1 inhibits

ribosome accumulation on problematic mRNAs……….……..……29

2.2.8 Genome-wide exploration of endogenous mRNA substrates

of Cue2 and Slh1……….…………………….…………..……………30

2.3 Discussion…………………….…………………….……………………..31

Chapter III: Deletion screens for factors involved in mRNA decay and translation suppression in response to ribosome stalling…………………….………………...75

3.1 Introduction…………………….…………………….………………….…76

3.2 Results…………………….…………………….…………………….……76

3.2.1 Deletion screens to identify factors involved in mRNA decay

and translation repression/activation on NGD substrates…………76

3.2.2 Temperature sensitive mutant screens……….………………78

3.2.3 Deletion screen candidates……….…………………….……..78

3.2.4 Transfer RNA modification genes are drastically enriched in

NGD screens……….…………………….……….……………….…..78

3.2.5 Translation repressor GIGYF2 homologs Smy2 and Syh1 act

on NGD mRNAs……….…………………….……….………………..79

3.2.6 Ribosome occupancy and mRNA decay……….………….…80

ix 3.2.7 Deubiquitinases involved in NGD……….………………….…82

3.3 Discussion…………………….…………………….……………………..83

Chapter IV: Materials and Methods…………………….…………………….….…117

Chapter V: Conclusion…………………….…………………….…………………..132

References…………………….…………………….……………………….……….135

Appendices…………………………………………………………………...135

Bibliography…..………………………………………………………………136

Biographical Statement…………………….…………………….……………….…151

x List of Tables

Table 1: OPT reporter OE screen……….…………………….……….……….……57

Table 2: NGD-CGA reporter OE screen……….…………………….……….……..63

Table 3: NGD-AAA reporter OE screen……….…………………….……….……..69

Table 4: OPT reporter deletion screen……….…………………….……….………93

Table 5: NGD-CGA reporter deletion screen……….…………………….………...98

Table 6: NGD-AAA reporter deletion screen……….…………………….……….103

Table 7: OPT reporter temperature sensitive screen……….…………………....109

Table 8: NGD-CGA reporter temperature sensitive screen……….…………….111

Table 9: NGD-AAA reporter temperature sensitive screen……….……………..114

xi List of Figures

Figure 1: Yeast overexpression screens identify a novel factor involved in

NGD……………………………………………………………………………………..37

Figure 2: Overexpression screens identify NGD-related factors. ……….……….39

Figure 3: Cue2 is the endonuclease what cleaves mRNA during

No Go Decay…………………………………………………………………………...40

Figure 4: Cue2 homology modeling and mutational analysis.……….……………42

Figure 5: Canonical decay by Xrn1 is the major contributor to No Go Decay…..45

Figure 6: Ribosome profiling analysis of NGD on reporter mRNAs in various genetic backgrounds.……….…………………………………………………..…….46

Figure 7: Comparison between monosome 21 nt RPFs and disome footprints……………………………………………………………………....48

Figure 8: 16, 21 and 28 nt RPFs on the NGD-CGA reporter from hel2∆dom34∆ski2∆…………………………………………………………………….49

Figure 9: In vitro cleavage of purified Cue2-SMR..……….…………………….….50

Figure 10: Cue2 targets prematurely polyadenylated mRNAs genome-wide for

NGD……………………………………………………………………………………..52

Figure 11: Cue2-dependent 16 nt RPFs on prematurely polyadenylated mRNAs………………………………………………………………………………….54

Figure 12: Multiple converging pathways at NGD substrates.……….…………...56

Figure 13: Yeast deletion screens reveal candidate genes involved in NGD…...86

Figure 14: Yeast temperature sensitive mutant screens reveal candidate genes involved in NGD.……….…………………….……….………………..….…………..88

xii Figure 15: Syh1 and Smy2 alter the decay of NGD substrates.…….……………90

Figure 16: Ribosome clearance factors alter the decay of NGD substrates.……91

Figure 17: Deubiquitinases are involved in regulating NGD mRNA levels………92

xiii Chapter I

Introduction

Messenger RNA (mRNA) serves as a molecular middleman between the hereditary information stored in DNA and the cellular activity carried out by proteins. In eukaryotes, RNA polymerase II transcribes DNA into RNA and the ribosome translates mRNA into protein. While many detailed mechanisms signaling transcription of mRNAs have been elucidated, mechanisms signaling mRNA decay remain vague. The decay of mRNA serves two purposes, regulating general mRNA levels to control protein expression and destroying problematic mRNA to avoid making toxic proteins. Messenger RNA half-lives vary from seconds to >30 minutes in yeast (Chan et al. 2018) and seconds to days in mammalian cells (Tani et al. 2012) proving mRNA decay plays a major regulatory role in determining the protein content of cells.

1 1.1 Translation dependent control of cellular mRNA levels

Eukaryotic mRNA contains a 5’ 7-methylguanosine cap, a 5’ leader sequence, an open reading frame, a 3’ untranslated region (UTR), and a poly-adenosine (poly-

A) tail. Decay of canonical mRNAs is thought to occur through poly-A tail shortening followed by 5’ decapping and exonucleolytic decay, but the mechanism by which an mRNA is signaled for deadenylation remains unknown

(Parker 2012). Recently, the codon content of the open reading frame of mRNA has been shown to strongly correlate with mRNA turnover, providing a potential mechanism for the ribosome to serve as an essential regulator of mRNA decay

(Presnyak et al. 2015).

In eukaryotes, translation initiates through the small (40S) subunit of the ribosome scanning along the 5’ leader of the mRNA until the start codon is recognized, upon which the large (60S) subunit binds and begins translation. The

80S ribosome then moves along the open reading frame, making peptide bonds at a rate dependent on the codons being translated. Finally, the ribosome recognizes the stop codon and terminates translation directly upstream of the 3’-

UTR and poly-A tail. The elongation rate of the ribosome depends on the ability of the incoming charged-tRNA to bind to the respective codon, the abundance of this charged tRNA in the cell, the ability of the ribosome to synthesize the peptide bond, and the ability of the ribosome to translocate down the mRNA and push the peptide further through the peptide exit tunnel.

Messenger RNA decay indirectly correlates with codon optimality, or how efficiently a particular codon is translated (Presnyak et al. 2015). Codon

2 optimality is determined by the tRNA Adaptive Index (tAI), which is a measure of the abundance of the charged-tRNA and the ability of the anticodon of the decoding tRNA to bind to the codon (dos Reis, Savva, and Wernisch 2004).

Although mRNAs with lower average codon optimality have not been shown to increase in ribosome occupancy, ribosomes are notably slower to decode lower optimality mRNAs as shown by run-off experiments (Presnyak et al. 2015). Slow decoding of an mRNA has been proposed to somehow signal decay.

Since this phenomenon was observed, a number of factors have been implicated in sensing slow ribosomes and triggering decay. Dhh1 is a decapping- associated decay factor suggested to be involved in signaling nonoptimal mRNA decay, but the mechanism was unknown (Radhakrishnan et al. 2016). Caf1 is a deadenylase shown to specifically deadenylate nonoptimal mRNAs, but the upstream signals for this were also unknown (Webster et al. 2018). Recently, the protein Not5, a member of the CCR4-NOT complex that is a major component of the cellular RNA decay machinery, was shown to bind in the tRNA exit site, or E site, of ribosomes exclusive of a tRNA in the E site (Buschauer et al. 2019).

Interestingly, it is plausible that slowly translating ribosomes contain an empty E site allowing for recognition by the decay factor Not5 and signaling of mRNA decay, although this pathway is currently understudied.

Extensive studies in yeast have identified a large collection of proteins involved in general mRNA decay, a number of which have been implicated in sensing codon optimality. Taken together, it seems clear that ribosome occupancy and activity is relevant to canonical mRNA degradation, but further

3 studies are required to determine the mechanism and extent to which this determines mRNA turnover.

4 1.2 Ribosome signaling decay of problematic mRNAs

When mRNA features essential for translation are disrupted, the ribosome becomes a molecular sensor that signals a cascade of events decaying the mRNA, degrading the nascent peptide product and rescuing stuck ribosomes

(Brandman and Hegde 2016). In the absence of these pathways, the cell can accumulate potentially toxic aberrant protein products and sequester ribosomes– essential, macromolecular complexes that are energetically exhaustive for the cell to make (Brandman and Hegde 2016). Disruptions in mRNA that stall ribosomes include RNA damage causing No Go Decay (NGD), premature polyadenylation leading to the absence of a stop codon and Non Stop Decay

(NSD), or nonsense mutations introducing an early stop codon and causing

Nonsense Mediated Decay (NMD) (Simms, Thomas, and Zaher 2017). These

RNA decay processes are associated with specific ribosome rescue and peptide degradation processes, all working together for efficient quality control.

1.2.1 No Go decay

Much like DNA damage, RNA damage takes the form of broken, crosslinked, or deaminated nucleotide sequences (Yan and Zaher 2019). If the ribosome begins to translate such an mRNA, then the damage will ultimately lead to a ribosome stall followed by a ribosome pileup (Yan and Zaher 2019). This stacking of ribosomes is a major signal of trouble and collided ribosomes have recently been shown to be a substrate for the E3 ligase Hel2 (Matsuo et al. 2017;

Simms, Yan, and Zaher 2017). The cryo-electron microscopy structure of the collided ribosome substrate reveals two such ribosomes with their 40S subunits

5 forming specific contacts (Ikeuchi et al. 2019; Juszkiewicz et al. 2018). The leading ribosome is in the unrotatated state, with an empty A-site and a peptidyl tRNA in the P-site, and the lagging ribosome is in the rotated state, with the peptidyl-tRNA in the A-P site and an uncharged tRNA in the P-E site. Through an unknown mechanism, Hel2 binds and ubiquitinates the 40S subunits of these collided ribosomes, signaling ribosome rescue, peptide degradation and mRNA cleavage (Matsuo et al. 2017; Simms, Yan, and Zaher 2017).

Ribosome rescue, or ribosome splitting on a damaged RNA occurs through the enzymatic function of the Ribosome Quality control Trigger (RQT) complex consisting of the putative RNA helicase Slh1 (Ski2-like helicase 1) and associated factors Cue3 and Rqt4 (Matsuo et al. 2017). In this reaction, splitting of the large and small subunits differs from translation termination in that the peptide-tRNA bond remains intact. Therefore, a large ribosomal subunit containing a peptidyl-tRNA is produced and must be processed by the cell to recycle the large subunit and degrade the peptide. The hydrolase Vms1 has recently been shown to cleave the CCA end of the peptidyl-tRNA, releasing the nascent peptide (Zurita Rendón et al. 2018), while the Ribosome Quality Control

Complex (RQC) ubiquitinates this nascent peptide, signaling it for degradation by the proteasome (Brandman et al. 2012; Shen et al. 2015).

Stalled ribosomes signal mRNA decay through a process called No Go Decay

(NGD) in which an endonuclease cleaves the mRNA near the site of the stalled ribosome (Doma and Parker 2006). Hel2 ubiquitination is known to be required for mRNA cleavage on stalls within an open reading frame and the identity of this

6 endonuclease has remained a pressing question in the field along with the signaling mechanism for cleavage (Ikeuchi et al. 2019).

1.2.2 Non-Stop Decay

Other types of problematic mRNAs can be produced from errors during mRNA production in the nucleus. During transcription, the polymerase is terminated by cleavage of the nascent mRNA transcript from the polymerase followed by polyadenylation by a poly-A polymerase (Mischo and Proudfoot

2013). If this process occurs prematurely, or if aberrant splicing occurs, the transcript could be polyadenylated within the open reading frame before a stop codon is transcribed. This results in an mRNA on which the ribosome translates into the poly-A tail, adding iterative lysines (encoded by AAA) to the nascent chain, creating an aberrant peptide.

In addition to making an incorrect peptide product, poly-A sequences themselves prove to be difficult for the ribosome to translate. Poly-A sequences are highly structured compared to other same-base oligos (Tang et al. 2019), likely inducing slow translation and causing ribosome collisions (Chandrasekaran et al. 2019; Tesina et al. 2019). Furthermore, the ribosome most easily frameshifts on poly-A sequences, possibly due to a combination of the unique poly-A structure and ribosome collisions (Chandrasekaran et al. 2019; Koutmou et al. 2015; Simms et al. 2018). Lastly, poly-A tails are coated by poly-A binding protein (PABP), which may affect translation elongation, though this has not yet been studied in depth.

7 Unlike NGD, decay of an mRNA without a stop codon, or Non-Stop Decay

(NSD), occurs through the major 3’ to 5’ exonucleolytic decay machinery known as the exosome, and the associated SKI complex (Frischmeyer et al. 2002; van

Hoof et al. 2002). The mechanism of recognition of a ribosome translating through the poly-A tail is unknown, but the SKI complex, consisting of proteins

Ski2, Ski3, Ski8 and Ski7, is an essential player in this process (van Hoof et al.

2002).

Ski2 is functionally and structurally similar to the RQT member Slh1, thus the

SKI complex may function similarly to the RQT complex to clear ribosomes from a poly-A tail. Dissimilarly, Ski2 and associated factors are bound by the exosome. Therefore, the SKI complex reaction likely doubles as a ribosome rescue and mRNA decay pathway. The method of degradation of the nascent peptide from a ribosome rescued by the SKI complex has not yet been elucidated, but it is possible this degradation occurs through Vms1 and the RQC as well.

8 1.3 Thesis contributions

1.3.1 The endonuclease Cue2 cleaves mRNAs at stalled ribosomes during No

Go Decay

Using reverse genetic screens in yeast we identify the long sought after endonuclease involved in No Go Decay (NGD). We find that this endonucleolytic pathway is a minor pathway in yeast and a major contributor to NGD is the canonical decay factor Xrn1. Furthermore, we touch upon the balance between ribosome rescue and mRNA decay, showing that in the absence of ribosome rescue, mRNA decay is drastically increased.

1.3.2 Deletion screens for factors involved in mRNA decay and translation

suppression in response to ribosome stalling

Through reporter deletion screens, we identified novel factors involved in NGD and translation suppression in yeast. We expand upon a complex interplay between ribosome clearance and mRNA decay, look at homologs of translation respressors identified in the screens, and show novel deubiquitinases that regulate mRNA levels.

9 Chapter II

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

Summary:

Translation of problematic sequences in mRNAs leads to ribosome collisions that trigger a series of quality control events including ribosome rescue, degradation of the stalled nascent polypeptide via the Ribosome-mediated

Quality control Complex (RQC), and targeting of the mRNA for decay (No Go

Decay or NGD). Previous studies provide strong evidence for the existence of an endonuclease involved in the process of NGD, though the identity of the endonuclease and the extent to which it contributes to mRNA decay remain unknown. Using a reverse genetic screen in yeast, we identify Cue2 as the conserved endonuclease that is recruited to stalled ribosomes to promote NGD.

Ribosome profiling and biochemistry provide strong evidence that Cue2 cleaves mRNA within the A site of the colliding ribosome. We demonstrate that NGD primarily proceeds via Xrn1-mediated exonucleolytic decay and Cue2-mediated endonucleolytic decay normally constitutes a secondary decay pathway. Finally, we show that the Cue2-dependent pathway becomes a major contributor to NGD in cells depleted of factors required for the resolution of stalled ribosome complexes (the RQT factors including Slh1). Together these results provide insights into how multiple decay processes converge to process problematic mRNAs in eukaryotic cells.

10 Credit: Experiments performed by KN D’Orazio unless otherwise stated here.

Figure 3A-3D and 4A-4F homology modeling performed by N Sinha. Figure 6-8,

9C, 10, and 11 ribosome profiling and sequencing experiments performed by CC

Wu.

11 Introduction

Translation is a highly regulated process in which ribosomes must initiate, elongate, and terminate accurately and efficiently to maintain optimal protein levels. Disruptions in the open reading frames of mRNAs cause ribosome stalling events, which trigger downstream quality control pathways that carry out ribosome rescue, nascent peptide degradation (via the Ribosome-mediated

Quality control Complex or RQC) and mRNA decay (No Go Decay or NGD)

(Brandman and Hegde 2016; Simms, Thomas, and Zaher 2017). Recent work in eukaryotes has revealed that ribosome collisions on problematic mRNAs create a unique interface on the aligned 40S subunits that serves as a substrate for E3 ubiquitin ligases such as Hel2 and Not4, and the RQC-trigger (RQT) complex, comprised of factors Slh1, Cue2 and Rqt4; together these factors are thought to trigger downstream quality control (Ferrin and Subramaniam 2017; Garzia et al.

2017; Ikeuchi et al. 2019; Juszkiewicz et al. 2018; Juszkiewicz and Hegde 2017;

Matsuo et al. 2017; Simms, Yan, and Zaher 2017; Sundaramoorthy et al. 2017).

A failure to process such colliding ribosomes and their associated proteotoxic nascent peptide products results in broad cellular distress, made evident by the strong conservation of these pathways (Balchin, Hayer-Hartl, and Hartl 2016).

Previous genetic screens and biochemistry have identified key factors involved in the recognition of stalling or colliding ribosomes and in targeting the nascent polypeptide to the RQC (Brandman et al. 2012; Kuroha et al. 2010; D. P.

Letzring et al. 2013). As these earliest screens focused on the identification of factors that stabilized reporter protein expression, factors involved in regulating

12 mRNA decay by NGD remain largely unknown. The hallmark of NGD

(Frischmeyer et al. 2002; van Hoof et al. 2002) is the presence of endonucleolytic cleavage events upstream of the ribosome stalling sequence as first detected by northern analysis (Doma and Parker 2006) and more recently by high resolution ribosome profiling or sequencing approaches (Arribere and Fire 2018; N.

Guydosh and Green 2017; N. R. Guydosh et al. 2017; Simms, Yan, and Zaher

2017). Importantly, these mRNA cut sites depend on ribosome collisions and the consequent polyubiquitination of the 40S subunit by the yeast protein Hel2

(Ikeuchi et al. 2019; Simms, Yan, and Zaher 2017). Multiple studies in the field collectively position NGD cleavage events within the vicinity of colliding ribosomes (Arribere and Fire 2018; N. Guydosh and Green 2017; N. R. Guydosh et al. 2017; Ibrahim et al. 2018; Ikeuchi et al. 2019; Ikeuchi and Inada 2016;

Simms et al. 2018; Simms, Yan, and Zaher 2017).

Interestingly, an endonuclease has been implicated in a related mRNA decay pathway, Nonsense Mediated Decay (NMD), in metazoans. During NMD, the ribosome translates to a premature stop codon (PTC), where an initial endonucleolytic cleavage event is carried out by a critical PIN-domain containing endonuclease, SMG6 (Glavan et al. 2006); sequencing experiments suggest that

SMG6 cleavage occurs in the A site of PTC-stalled ribosomes (Arribere and Fire

2018; Lykke-Andersen et al. 2014; Schmidt et al. 2015). In C. elegans, there is evidence that the initial SMG6-mediated endonucleolytic cleavage leads to iterated cleavages upstream of the PTC similar to those characterized for the

NGD pathway in yeast (Arribere and Fire 2018). The idea that the various decay

13 pathways may act synergistically is intriguing. Homologs of SMG6 (NMD4/EBS1 in yeast) (Dehecq et al. 2018) and other PIN domain-containing proteins, as well as other endonuclease folds, exist throughout Eukarya and anecdotally have been evaluated by the field as potential candidates for functioning in NGD, though there are no reports to indicate that any function in this capacity. The identity of the endonuclease responsible for cleavage in NGD remains unknown.

The presumed utility of endonucleolytic cleavage of a problematic mRNA is to provide access to the mRNA for the canonical exonucleolytic decay machinery that broadly regulates mRNA levels in the cell. In yeast, Xrn1 is the canonical 5’ to 3’ exonuclease which, after decapping of the mRNA, degrades mRNA from the 5’ end; importantly, Xrn1 normally functions co-translationally such that signals from elongating ribosomes might be relevant to its recruitment

(W. Hu et al. 2009; Pelechano, Wei, and Steinmetz 2015). Additionally, the exosome is the 3’ to 5’ exonuclease which,after deadenylation, degrades mRNAs from the 3’ end and is recruited by the SKI auxiliary complex consisting of

Ski2/Ski3/Ski8 and Ski7 (Halbach et al. 2013). While Xrn1-mediated degradation is thought to be the dominant pathway for most general decay in yeast (Anderson and Parker 1998), the exosome has been implicated as critical for many degradation events in the cell including those targeting prematurely polyadenylated mRNAs (these mRNAs are usually referred to as Non-Stop

Decay (NSD) targets) (Frischmeyer et al. 2002; van Hoof et al. 2002; Tsuboi et al. 2012). In metazoans, it is less clear what the relative contributions of Xrn1 and the exosome are to the degradation of normal cellular mRNAs. How the

14 endonucleolytic and canonical exonucleolytic decay pathways coordinate their actions on problematic mRNAs remains unknown.

Here we present a reverse genetic screen in S. cerevisiae that identifies

Cue2 as the primary endonuclease in NGD. Using ribosome profiling and biochemical assays, we show that Cue2 cleaves mRNAs in the A site of collided ribosomes, and that ribosomes which accumulate at these cleaved sites are rescued by the known ribosome rescue factor Dom34 (N. R. Guydosh and Green

2014; Shoemaker, Eyler, and Green 2010). We further show that stall-dependent endonucleolytic cleavage represents a relatively minor pathway contributing to the decay of the problematic mRNA reporter, while exonucleolytic processing by the canonical decay machinery, in particular Xrn1, plays the primary role. The

Cue2-mediated endonucleolytic cleavage activity is substantially increased in genetic backgrounds lacking the factor Slh1, a known component of the RQT complex (Matsuo et al. 2017), suggesting that the relative contribution of this pathway could increase in different environmental conditions. Our final model provides key insights into what happens in cells upon recognition of stalled ribosomes on problematic mRNAs, and reconciles how both endo- and exonucleolytic decay act synergistically to resolve these dead-end translation intermediates.

15 Results:

Screening for factors involved in NGD

To identify factors that impact the degradation of mRNAs targeted by

NGD, we developed a construct that directly reports on mRNA levels. Previous genetic screens in yeast (Brandman et al. 2012; Kuroha et al. 2010; D. P.

Letzring et al. 2013) were based on reporters containing a stalling motif in an open reading frame (ORF). As a result, these screens primarily revealed machinery involved in recognition of stalled ribosomes and in degradation of the nascent polypeptide, but missed factors involved in mRNA decay. In our reporters (GFP-2A-FLAG-HIS3), the protein output for the screen (GFP) is decoupled from the stalling motif positioned within HIS3 by a 2A self-cleaving peptide sequence (Di Santo, Aboulhouda, and Weinberg 2016; Sharma et al.

2012) (Figure 1A). Because GFP is released before the ribosome encounters the stalling sequence within the HIS3 ORF, its abundance directly reflects the reporter mRNA levels and translation efficiency independent of the downstream consequences of nascent peptide degradation. These reporters utilize a bidirectional galactose inducible promoter such that an RFP transcript is produced from the opposite strand and functions as an internal-control for measurement of general protein synthesis.

The screen utilized two different NGD constructs with stalling motifs inserted into the HIS3 gene: the first contains 12 CGA codons (NGD-CGA) which are decoded slowly (Kuroha et al. 2010; Daniel P. Letzring, Dean, and Grayhack

2010) by the low-copy ICG-tRNAArg and the second contains 12 AAA codons

16 (NGD-AAA) that mimic polyA tail and are known to trigger ribosome stalling and mRNA quality control in both yeast and mammalian systems (Arthur et al. 2015;

Frischmeyer et al. 2002; Garzia et al. 2017; N. Guydosh and Green 2017; van

Hoof et al. 2002; Ito-Harashima et al. 2007; Juszkiewicz and Hegde 2017;

Sundaramoorthy et al. 2017) (Figure 1A). The (CGA)12 and (AAA)12 inserts result in robust 3-fold and 2-fold decreases in the GFP/RFP ratio, respectively, compared to the no insert (optimal, OPT) control (Figure 1B). Similar changes are also seen in GFP levels by western blot and in full-length mRNA levels by northern blot (Figure 2A). As previously reported, deletion of the exosome auxiliary factor gene SKI2 stabilized the 5’ decay fragment resulting from endonucleolytic cleavage associated with NGD reporters (Figure 2A) (Doma and

Parker 2006). Finally, stalling during the synthesis of the FLAG-His3 fusion protein in the two stalling reporters (NGD-CGA and NGD-AAA) leads to degradation of the nascent peptide (Figure 2A, FLAG panel).

We used high-throughput reverse genetic screens and reporter-synthetic genetic array (R-SGA) methods (Fillingham et al. 2009; Tong et al. 2001) to evaluate the effects of overexpressing annotated genes on GFP expression. We began by crossing strains carrying the control (OPT) and the two different no-go decay reporters (NGD-CGA and NGD-AAA) described in Figure 1A into the S. cerevisiae overexpression library (Douglas et al. 2012; Giaever et al. 2002; Y. Hu et al. 2007). For each overexpression screen, we isolated diploid strains containing both the overexpression plasmid and our reporter. Selected strains were transferred to galactose-rich plates and the GFP and RFP levels were

17 evaluated by fluorimetry. We plotted the results from the screens individually, comparing Z-scores for the log2(GFP/RFP) signals from each NGD reporter strain to Z-scores from the corresponding strain carrying the OPT reporter

(Figure 1C-1D). Normalization with RFP intensity was used to eliminate non- specific factors that impact expression of both RFP and GFP.

The overexpression screen revealed a set of candidate genes that modulate GFP levels for the NGD reporters relative to the OPT reporter (Figure

1C-1D and Tables 1-3). Broadly, we see a stronger overlap in candidate overexpression genes among the NGD reporters than for either NGD reporter compared to the OPT reporter (Figure 2B). By far, the strongest outlier by Z- score causing reduced GFP expression for both NGD reporters, without affecting the OPT reporter, resulted from overexpression of the gene CUE2 (Figure 1C-

1D). Flow cytometry and northern analysis confirm that NGD-CGA reporter mRNA levels are substantially reduced upon CUE2 overexpression whereas the control RFP transcript is not affected (Figure 1E and 2C).

CUE2 domain structure and homology modeling

The domain structure of Cue2 reveals it to be a promising candidate for the missing endonuclease for NGD. CUE2 contains two conserved CUE

(coupling of ubiquitin to ER degradation) ubiquitin-binding domains at the N- terminus (Kang et al. 2003), followed by two putative ubiquitin-binding domains

(UBA* (ubiquitin-associated domain) and CUE* domain respectively), and an

SMR (small MutS-related) hydrolase domain at the C-terminus (Figure 3A). We performed alignments of the CUE and SMR domains of Cue2 using structure-

18 based alignment tools (Figure 4A, and 4B, respectively) and found putative homologs in various kingdoms of life (including the human NEDD4-binding protein 2, N4BP2). While the SMR domain of MutS family enzymes canonically function as DNA-nicking hydrolases (Fukui and Kuramitsu 2011), the SMR domain of SOT1 in plants exhibits RNA endonuclease activity (Zhou et al. 2017).

Additionally, SMR domains show structural similarity to bacterial RNase E (Fukui and Kuramitsu 2011). Alignments of the SMR domains of these and other proteins enabled us to identify conserved residues (Figure 3B), some of which are known to be critical for RNA endonuclease activity in the plant enzyme (Zhou et al. 2017).

Heuristic searches of the structurally defined SMR domain of the mammalian homolog of Cue2 (N4BP2) (Diercks et al. 2008) against known structures in the Protein Data Bank found it to be structurally homologous to the

C-terminal domain (CTD) of bacterial Initiation Factor 3 (Biou, Shu, and

Ramakrishnan 1995) (Figure 3B and Figure 4C-4D). During initiation in bacteria,

IF3 binds to the 30S pre-initiation complex (PIC) and helps position the initiator tRNA at the AUG start codon; the CTD of IF3 binds in close proximity to the P and A sites of the small subunit, closely approaching the mRNA channel

(Hussain et al. 2016). We aligned the Cue2-SMR homology model with the structure of the IF3-CTD on the ribosome and observed that conserved residues

D348, H350, and R402 are positioned along the mRNA channel in this model

(Figure 3C-3D and Figure 4E-4F) and thus represent potential residues critical

19 for endonucleolytic cleavage activity; additionally, R402 had previously been implicated in the RNA cleavage activity of other SMR domains (Zhou et al. 2017).

Characterizing roles of CUE2 in NGD in vivo

We performed several different experiments to ask if these conserved residues are necessary for Cue2 function. First, we generated mutations in HA- tagged Cue2 and performed flow cytometry on the NGD-CGA reporter under overexpression conditions for the different Cue2 variants. We find that individually mutating the conserved residues D348, H350, and R402 to alanine

(A) causes a modest increase in GFP expression relative to CUE2 WT, whereas the R402K mutation (which maintains a positively charged amino acid) has no discernible effect (Figure 3E). Mutating residues 348 through 350, which includes the conserved residues D348 and H350, nearly restores GFP reporter signal in this assay (Figure 3E). Importantly, each of these variants is expressed at similar levels (Figure 4G). These data suggest a potential role for residues D348, H350, and R402 in the SMR domain of Cue2 in reducing levels of problematic mRNAs.

We next asked if Cue2 is necessary for the previously documented endonucleolytic cleavage of the NGD reporter transcripts. In order to visualize the mRNA fragments resulting from cleavage, we deleted either the 3’ - 5’ mRNA decay auxiliary factor, SKI2, or the major 5’ - 3’ exonuclease, XRN1, to stabilize the 5’- and 3’-fragments, respectively (Doma and Parker 2006) (Figure 3F; lane 1 and 3G; lane 1). Upon deletion of CUE2 in the appropriate yeast background, we see that both of these decay intermediates disappear (Figure 3F; lane 2 and 3G; lane 2).

20 In order to further investigate the role of the specific amino acids proposed to be catalytically critical, we modified CUE2 at its endogenous chromosomal locus with an HA tag, to allow us to follow native CUE2 protein levels, and generated mutations at the endogenous CUE2 locus in a ski2Δ background. As a first test, we deleted the C-terminal portion of CUE2 comprising the SMR domain and revealed a complete loss of endonucleolytic cleavage (Figure 3H, lane 3). To further refine this analysis, we made combined mutations in CUE2 at D348 and

H350 and observed a complete abolishment of endonucleolytic cleavage activity.

Western analysis confirmed that protein expression levels from the CUE2 locus are equivalent for these different protein variants (Figure 4H). These data provide strong support for a hydrolytic role for Cue2 in the endonucleolytic cleavage of problematic mRNAs in the NGD pathway.

Contribution of Cue2 to NGD is increased in specific genetic backgrounds

Surprisingly, deletion of CUE2 in the WT, ski2Δ or xrn1Δ background does not demonstrably impact GFP or steady-state full-length mRNA levels (Figure

5A-5B, Figure 3F, or Figure 3G, respectively) suggesting that the endonucleolytic decay pathway is not responsible for the majority of the three-fold loss in mRNA levels observed for this NGD reporter (relative to OPT). However, we see that deleting XRN1 alone greatly restores NGD reporter mRNA levels while deleting

SKI2 alone has very little effect (Figure 5A-5B). These results together suggest that the majority of mRNA degradation observed for these NGD reporters is mediated by canonical decay pathways, and mostly by Xrn1.

21 A key insight into the regulation of decay by these distinct mRNA decay pathways (endo- and exonucleolytic) came from examining decay intermediates in strains lacking the factor Slh1. Slh1 is a member of the Ribosome Quality

Trigger complex that was previously implicated in regulating NGD (Ikeuchi et al.

2019; Matsuo et al. 2017). This large protein (~200 kDa) is homologous to helicases implicated in RNA decay and splicing including Ski2 and Brr2, respectively (Johnson and Jackson 2013). Slh1 contains two DEAD box motifs and mutations in them are known to disrupt targeting of the nascent peptide on problematic mRNAs to proteolytic decay by the RQC. Based on these observations, Slh1 has been proposed to directly or indirectly dissociate stalled ribosomes and to thereby target them for the RQC (Matsuo et al. 2017).

We found that in the ski2Δ slh1Δ and xrn1Δ slh1Δ strains, we see a sizeable build-up of endonucleolytically cleaved mRNA fragments (Figure 3F; lane 3 and 3G; lane 3). Consistent with the recent study by Ikeuchi et al. (Ikeuchi et al. 2019), we also see smearing of the signal representative of shorter 5’ and longer 3’ fragments due to additional cleavage events occurring upstream of the primary stall site in the absence of SLH1. Importantly, we also observe a robust decrease in full-length mRNA levels upon deletion of SLH1 (Figure 3F; lane 3 and 2G; lane 3). These data suggest a model wherein Slh1 activity somehow competes with Cue2-mediated endonucleolytic cleavage. To ask whether the increased endonucleolytic cleavage seen in the SLH1 deletion background is the result of Cue2 activity, we deleted CUE2 in the xrn1Δ slh1Δ background and observed a complete loss of the decay intermediate and a restoration of full-

22 length mRNA levels (Figure 3G; lane 4). These data provide strong evidence that the action of Slh1 negatively regulates the assembly of a potent Cue2 substrate though the molecular mechanism for this regulation is not defined.

Ribosome profiling provides high-resolution view of Cue2 cleavage sites

We next utilized ribosome profiling to assess the role and specificity of

Cue2 in endonucleolytic cleavage of problematic mRNAs. Ribosome profiling was performed in various genetic backgrounds where the NGD-CGA reporter construct was included to follow the process of NGD on a well-defined mRNA substrate. The approach benefits from our ability to distinguish three distinct sizes of ribosome-protected mRNA fragment (RPF) that correspond to three different states of the ribosome during translation. 16 nucleotide (nt) RPFs correspond to ribosomes that have translated to the truncated end of an mRNA in the cell, usually an mRNA that has been endo- or exonucleolytically processed; these 16 nt RFPs are enriched in a dom34Δ strain since these are established targets for Dom34-mediated ribosome rescue (Arribere and Fire

2018; N. Guydosh and Green 2017; N. R. Guydosh et al. 2017; N. R. Guydosh and Green 2014). 21 and 28 nt RPFs are derived from ribosomes on intact mRNA, with the two sizes corresponding to different conformational states of ribosomes in the elongation cycle (Lareau et al. 2014; Wu et al. 2019). In our modified ribosome profiling library preparation (Wu et al. 2019), 21 nt RPFs represent classical/decoding ribosomes waiting for the next aminoacyl-tRNA whereas the 28 nt RPFs primarily represent ribosomes in a rotated/pre- translocation state. Throughout this study, we performed ribosome profiling in a

23 ski2Δ background to prevent 3’-5’ exonucleolytic degradation by the exosome and enable detection of mRNA decay intermediates. We note that deleting SKI2 generally stabilizes prematurely polyadenylated and truncated mRNAs in the cell, as seen previously (van Hoof et al. 2002; Tsuboi et al. 2012); therefore use of this background likely enhanced Cue2 cleavage on these mRNA substrates.

In a first set of experiments, we compare the distributions of RPFs in ski2Δ strains to strains additionally lacking DOM34, or DOM34 and CUE2; in each case, we separately consider the different RPF sizes (16, 21 and 28) to determine the state of the ribosome on these sequences. In ski2Δ, we observe a clear accumulation of ribosome density at the (CGA)12 stall site (in the 21 nt RPF track) in our reporter (Figure 6A) consistent with ribosome stalling at this site.

Looking first within the actual (CGA)12 codon tract, we see a quadruplet of 21 nt

RPF peaks from the 2nd to the 5th CGA codon (middle panel). Because of difficulties in mapping reads to repetitive sequences such as (CGA)12, we also isolated RPFs from a minor (< 5% of total reads on reporter) population of stacked disome species (N. R. Guydosh and Green 2014) and show that although some ribosomes move into the subsequent CGA codons, the principal ribosome accumulation occurs between the 2nd and the 5th codon (Figure 7).

While 21 nt RFPs are enriched on CGA codons, 28 nt RPFs are not (Figure 6A, compare middle and bottom panels), indicating that most ribosomes found at the

CGA codons are waiting to decode the next aminoacyl-tRNA; these observations are consistent with the fact that CGA codons are poorly decoded (Daniel P.

Letzring, Dean, and Grayhack 2010). The 21 nt RPF quadruplet at the CGA

24 cluster represents the stalled “lead” ribosome in its classical (unrotated) conformation, which is consistent with recent cryoEM structures of collided ribosomes (Ikeuchi et al. 2019; Juszkiewicz et al. 2018). About 30 nts upstream of these lead ribosomes, we observe the accumulation of 28 nt RPFs that correspond to colliding ribosomes in a rotated state (Figure 6A, bottom panel), also in agreement with the ribosome states recently reported in cryo-EM structures of collided disomes. In the dom34Δ ski2Δ strain (where 16 nt RPFs are stabilized), we observe a strong quadruplet of 16 nt RPF peaks exactly 30 nts upstream of the quadruplet of 21 nt RPFs (Figure 6B), indicative of endonucleolytic cleavage events occurring upstream of the (CGA)12–stalled ribosomes. The precise 30 nt distance between the 16 nt RPFs and the downstream 21 nt RPFs is consistent with the measured distance between the A sites of two colliding ribosomes.

To address the role of Cue2 as the endonuclease, we examined the same collection of RPFs in the cue2Δ dom34Δ ski2Δ strain. In this background, the 16 nt RPFs indicative of endonucleolytic cleavage are dramatically reduced in abundance while the distributions of 21 and 28 nt RPFs are largely unaffected

(Figure 6C). These data argue that endonucleolytic cleavage occurs precisely in the A site of the collided, rotated ribosome and requires Cue2. A high-resolution view of the site of cleavage within the A site codon is found in the model in Figure

6G. These observations are broadly consistent with recent RACE analyses

(Ikeuchi et al. 2019) and with our structural homology modeling placing Cue2 in the decoding center of the ribosome (Figure 3C and 3D).

25 Based on our observations that deletion of Slh1 increases utilization of

Cue2 for processing of problematic mRNAs, we next examined the distribution of

RPFs on the NGD-CGA reporter in yeast strains lacking SLH1. In the Δslh1 ski2Δ strain, the pattern of RPF distributions is broadly similar to what we observed in the ski2Δ strain (Figure 6D compared to 6A): 21 nt RPFs accumulate on the 2nd through 5th CGA codons and 28 nt RPFs accumulate 30 nts behind the lead ribosome; we also see modest accumulation of RPFs downstream of the stalling sequence as previously reported (Figure 6D) (Sitron, Park, and

Brandman 2017). SLH1 deletion does not stabilize 16 nt RPFs, and therefore, likely does not act on cleaved Cue2-substrates (Figure 6D compared to 6A).

When we delete DOM34 in the slh1Δ ski2Δ background (slh1Δ dom34Δ ski2Δ), we see a dramatic accumulation of 16 nt RPF peaks (again, as a quadruplet, though with an altered relative distribution) behind the lead ribosomes (Figure 6E compared to 6B, top panels, note the larger y-axis scale in Figure 6E and 6F compared to 6B and 6C). These data are consistent with the substantial increase in cleavage we see via northern blot (Figure 3F and 3G) and previous findings that deletion of the RQT complex causes somewhat distinct cleavage events

(Ikeuchi et al. 2019). Importantly, deletion of CUE2 in the slh1Δ dom34Δ ski2Δ background leads to a near complete loss of the 16 nt RPFs (Figure 6F, top panel) and an enrichment of 21 and 28 nt RPFs.

Lastly, consistent with earlier studies showing that endonucleolytic cleavage events associated with NGD depend on the E3 ligase Hel2 (Ikeuchi et al. 2019), we show that deletion of HEL2 in both dom34Δ ski2Δ and slh1Δ

26 dom34Δ ski2Δ backgrounds leads to a complete loss of 16 nt RPFs (Figure 8A and 8B, respectively).

In vitro reconstitution of Cue2 cleavage on isolated colliding ribosomes

To test if Cue2 cleaves mRNA directly, we performed in vitro cleavage assays with the heterologously expressed SMR domain (and R402A mutant) of

Cue2 and purified colliding ribosomes. To isolate the colliding ribosomes, cell- wide ribosome collision events were induced using low dose cycloheximide treatment of growing yeast (Simms, Yan, and Zaher 2017). We both optimized the yield and ensured that the purified Cue2-SMR domain was the only source of

Cue2 in our experiment by isolating ribosomes from the slh1Δ dom34Δ cue2Δ ski2Δ strain; these cells also carried and expressed the NGD-CGA reporter.

Since collided ribosomes are resistant to general nucleases (Juszkiewicz et al.

2018), MNase digestion of lysates collapses elongating ribosomes from polysomes to monosomes but spares those with closely packed (nuclease resistant) ribosomes (Figure 9A). As anticipated, only cells treated with low doses of cycloheximide yielded a substantial population of nuclease resistant, or collided, ribosomes after MNase digestion (Figure 9A, black trace). We isolated nuclease resistant trisomes from a sucrose gradient to serve as the substrate for our in vitro reconstituted cleavage experiment.

We purified the isolated SMR domain of Cue2 based on our in vivo evidence that this domain represents the functional endonuclease portion of the protein. We added the purified SMR domain of Cue2 to the isolated nuclease resistant trisomes and resolved the products of the reaction on a sucrose

27 gradient that optimally resolves trisomes from monosomes and disomes. While we initially isolated trisomes, it is clear that the ‘untreated’ sample (black trace) contains trisomes, as well as disomes and monosomes; this may be the result of cross-contamination of those peaks during purification or the instability of the trisome complex. Nevertheless, on addition of Cue2-SMR (WT) (pink trace), we observe a substantial loss of trisomes and a corresponding increase in monosomes and disomes (Figure 9B), while on addition of the Cue2 SMR-

R402A mutant (orange trace), we observe a more modest decrease in trisomes

(Figure 9B). These results provide initial in vitro evidence that the SMR

(hydrolase) domain of Cue2 cleaves the mRNA within a stack of colliding ribosomes.

To extend these observations, we mapped more precisely the in vitro cleavage sites of the SMR domain by sequencing the ribosome-protected mRNA fragments from the untreated and SMR-treated samples. The strain used to prepare the cycloheximide-induced colliding ribosomes also expressed the NGD-

CGA reporter, so we anticipated that within our population of bulk colliding ribosomes would be a sub-population of colliding ribosomes on our reporter, also subject to cleavage by the SMR domain. We performed RNA-seq by standard means from the trisomes treated with or without the Cue2 SMR domain and aligned the 3’ ends of the RNA reads to the reporter sequence. In the sample treated with the wild type Cue2-SMR domain, we see a striking accumulation of

3’ fragment ends that maps precisely where the strong cleavage sites were reported (in the A site of the colliding ribosome) in the ribosome profiling data for

28 the same strain (slh1Δ dom34Δ sk2iΔ); reassuringly, these cleavage events are not seen in the sequencing data for the Cue2 SMR-R402A mutant or for the no treatment control (Figure 9C). The remarkable agreement between the in vivo- based ribosome profiling data and the in vitro-based cleavage data provides strong support for the identity of Cue2 as the endonuclease involved in NGD.

Ribosome profiling provides evidence that Slh1 inhibits ribosome accumulation on problematic mRNAs

Our analyses (northern blots and profiling) provide strong evidence that deletion of SLH1 in yeast results in increased levels of Cue2 cleavage on our

NGD-CGA reporter. Multiple molecular mechanisms could be responsible for this outcome. For example, a simple model might involve direct competition on the colliding ribosomes wherein Slh1 binding simply blocks Cue2 from binding to the same site. Another model could be that Slh1 functions to clear colliding ribosomes, dissociating large and small subunits from one another, and thus allowing the released peptidyl-tRNA:60S complex to be targeted for RQC

(Matsuo et al. 2017). Earlier studies have argued that Slh1 functions as a translational repressor though the dependence of RQC activity on Slh1 function argues against this model (Searfoss, Dever, and Wickner 2001).

We used ribosome profiling to directly assess ribosome occupancy on our problematic mRNA sequence as a function of Slh1 activity. As described above, when we compared 16 nt RPF accumulation on the NGD-CGA reporter in a ski2∆ or slh1∆ ski2∆ strain (Figure 6D) we saw no differential accumulation of these RPFs that might be indicative of Slh1 function on ribosomes trapped on

29 truncated mRNAs. However, when we compared RPFs on the uncleaved mRNAs (i.e. the 21 and 28 nt RPFs) in the cue2Δ dom34Δ ski2∆ and slh1Δ cue2Δ dom34Δ ski2∆ strains, we observed that SLH1 deletion results in a substantial build-up of RPFs both at the stalling site (21 nt RPFs) (compare middle panels, Figure 6C and 6F) and upstream (28 nt RPFs) of the stalling site

(compare bottom panels, Figure 6C and 6F and quantitation in 6H). These data were normalized to overall mRNA levels and reveal more than a 2-fold increase in ribosome occupancy when SLH1 is deleted. These data are consistent with the proposed model that Slh1 prevents the accumulation of colliding ribosomes

(Matsuo et al. 2017) thereby limiting the availability of Cue2 substrates.

Genome-wide exploration of endogenous mRNA substrates of Cue2 and Slh1

One possible endogenous target class for Cue2 is the set of prematurely polyadenylated mRNAs (where polyadenylation happens within the annotated

ORF) that are commonly generated by aberrant RNA processing events in the nucleus. We looked for such candidates in our data sets by first identifying genes that exhibit evidence of untemplated A’s within the monosome footprint population (i.e. ribosomes that are actively translating prematurely polyadenylated mRNAs) (Figure 10A, pink dots). These candidate genes nicely correlated with those previously identified from 3’ end sequencing approaches

(Ozsolak et al. 2010; Pelechano, Wei, and Steinmetz 2013) (e.g. see data for two specific mRNAs RNA14 and YAP1 in Figure 11A) and from ribosome profiling approaches exploring Dom34 function (N. Guydosh and Green 2017). We next identify Cue2-dependent 16 nt RPFs in both a ski2∆ dom34∆ (Figure 10A,

30 orange dots) as well as in a ski2∆ dom34∆ slh1∆ background (Figure 11B, orange dots) and see that they are substantially enriched in the prematurely polyadenylated mRNAs (Figure 10A and Figure 11B, red dots). We note that in the ski2∆ dom34∆ background a good fraction (21/55) of the prematurely polyadenylated genes are substantially reduced in 16 nt RPFs upon CUE2 deletion (Figure 10B). Moreover, we show that for well-defined mRNA targets

RNA14 and YAP1, the Cue2-dependent 16 nt RPFs are positioned as anticipated directly upstream of the previously characterized sites of premature polyadenylation (Figure 10C and Figure 11C) (N. Guydosh and Green 2017;

Pelechano, Wei, and Steinmetz 2013). These data support a role for Cue2- mediated endonucleolytic cleavage on such problematic transcripts. Although the

SKI complex and the exosome decay NSD substrates (Frischmeyer et al. 2002; van Hoof et al. 2002), Cue2-cleavage may be one mechanism for cells to initiate this decay.

Discussion

Cells have evolved a complex set of mechanisms to recognize and resolve ribosomes stalled on problematic mRNAs and to target the mRNAs for decay. The data presented here identify and characterize Cue2 as the endonuclease critical for NGD in yeast. We demonstrate using a wide range of in vivo and in vitro approaches that Cue2 is necessary and sufficient for cleavage of mRNAs loaded with stalled, colliding ribosomes. Structural homology modeling with the C-terminal domain of IF3 positions the catalytic SMR domain of Cue2 in the A site of the small ribosome subunit (Figure 3C and 3D, Figure 4D to 4F).

31 And, according to this model, the conserved amino acids in Cue2 that closely approach mRNA in the A site (Figure 3D, Figure 4B and 4C) are indeed critical for efficient endonuclease cleavage activity (Figure 3E and 3H). Although mutating any one of the residues D348, H350, or R402, did not completely abolish Cue2 activity, mutating a combination of these residues or removing the entire SMR domain from the protein had a very strong effect (Figure 3E and 3H).

Other studies have argued that the NGD endonuclease leaves a cyclic 3’ phosphate and a 5’ hydroxyl, products that are consistent with Cue2 being a metal independent endonuclease (Navickas et al. 2018). As metal independent

RNases typically depend on mechanisms where RNA is contorted to act upon itself, it is often difficult to identify a single essential catalytic residue.

Interestingly, N4BP2, the mammalian homolog of Cue2, contains a PNK domain; this kinase activity could be required for the further degradation of the mRNA through Xrn1, which depends on 5’ phosphate groups.

Strong support for the modeling of Cue2 in the A site came from the high- resolution ribosome profiling data establishing that Cue2 cleaves mRNAs precisely in the A site of the ‘colliding’ ribosome (Figure 6). We additionally find by profiling that the lead ribosome is stalled in an unrotated state (21 nt RPFs),

Arg unable to effectively accommodate the tRNAICG that should normally decode the

CGA codon in the A site in concordance with previous studies on problematic codon pairs (Wu et al. 2019). We also find that the lagging ribosomes are found in a pre-translocation or rotated state (28 nt RPFs). These different conformational states of the colliding ribosomes identified by ribosome profiling

32 correlate with those defined in recent cryoEM structures of colliding ribosomes

(Ikeuchi et al. 2019; Juszkiewicz et al. 2018). We note that our data and existing structures suggest that Cue2 requires partial displacement of the peptidyl-tRNA substrate positioned in the A site of the colliding ribosome in order to access the mRNA. Recent studies identify residues in Rps3 at the mRNA entry tunnel that are required for NGD cleavage consistent with the idea that Cue2 binds within the ribosome mRNA channel to promote cleavage (Simms et al. 2018).

Another structural feature of Cue2 is that it possesses two highly conserved CUE domains at the N-terminus, and up to two putative ubiquitin- binding domains prior to the proposed catalytic SMR domain. Endonucleolytic cleavage of problematic mRNAs was previously shown to depend on the function of the E3 ligase Hel2 which ubiquitinates several small subunit ribosome proteins

(Ikeuchi et al. 2019); our ribosome profiling data provides strong support for this model (Figure 8). We suspect that the ubiquitin-binding domains of Cue2 might recognize Hel2-ubiquitinated sites on the colliding ribosomes thus allowing recruitment of the SMR domain into the A site of the colliding ribosome. We can imagine that recognition of ubiquitin chains by Cue2 might involve multiple ubiquitinated sites on a single collided ribosome or those found on neighboring ribosomes. These conserved and putative ubiquitin-binding domains might also function within Cue2 itself to inhibit promiscuous activity of the endonuclease in the absence of an appropriate target, Indeed, this possibility prompted us to perform the in vitro cleavage experiments with the isolated SMR domain of Cue2.

33 In addition to the identification and characterization of the key endonuclease involved in NGD, our results help to clarify the relative contributions of decay pathways for problematic mRNAs. We show that canonical exonucleolytic processing of problematic mRNAs is the dominant pathway on our

NGD-CGA reporter, with the strongest contributions from Xrn1 rather than the exosome (Figure 5). Importantly, we find that the Cue2-mediated endonucleolytic pathway is activated in genetic backgrounds lacking the helicase Slh1 (a member of the previously identified RQT complex) (Figure 3F-3G). These ideas are brought together in the model in Figure 12 that outlines how multiple decay pathways converge to bring about the decay of problematic mRNAs. We anticipate that mRNAs with different problematic features might differentially depend on these pathways as suggested by previous studies (Tsuboi et al.

2012). Importantly, however, our identification of the endonuclease for NGD will allow subsequent studies to better disentangle the distinct contributions from multiple decay machineries in the cell. A challenge moving forward will be to identify how problematic mRNAs with colliding ribosomes signal to decapping factors and Xrn1 to initiate degradation (indicated with question marks in the model in Figure 12).

Our genetic insights also lead us to speculate that under conditions where cells are overwhelmed by colliding ribosome complexes, for example under stressful RNA-damaging conditions (Simms et al. 2014), that the RQT machinery might become limiting and that there will be an increased role for Cue2-mediated endonucleolytic processing. Previous genetic screens reveal increased sensitivity

34 of CUE2 deletion strains to ribosome-targeted compounds (Alamgir et al. 2010;

Mircus et al. 2015). More generally, the strong conservation of Cue2 throughout eukaryotes argues that this endonucleolytic processing pathway plays a fundamental role in biology.

Previous studies provide strong evidence that the activity of the RQT complex is critical in targeting nascent peptides on problematic mRNAs for degradation via the RQC (Matsuo et al. 2017; Sitron, Park, and Brandman 2017) through ribosome rescue events that generate a dissociated 60S-peptidyl-tRNA complex. Our ribosome profiling results indicate that ribosome occupancies on problematic mRNAs increase on deletion of SLH1 (Figure 6C, 6F, 6H) consistent with the possibility that Slh1 functions directly or indirectly to dissociate ribosomes. The increased CUE2-dependent cleavage (Figure 3F-3G and Figure

6B, 6E) that we observe in slh1∆ background suggests that an increase in the density of collided ribosomes makes an ideal Cue2 substrate.

Taken together, we suggest that Slh1 and Dom34 may both target collided ribosomes for the RQC, though their specificities may be somewhat distinct. For example, Slh1 may use its helicase function to target ribosomes on intact mRNAs with an accessible 3’ end, while Dom34 targets those ribosomes on truncated mRNAs that are inaccessible to Slh1 activity. In this view, a critical role of Cue2-mediated endonucleolytic cleavage is to provide another pathway for ribosome dissociation that can target nascent peptides for RQC. Previous studies provide strong support for the idea that Dom34 preferentially rescues ribosomes that are trapped on truncated mRNAs (N. R. Guydosh and Green

35 2014; Shoemaker, Eyler, and Green 2010) and in vitro studies have nicely connected this rescue activity of Dom34 to downstream RQC functions (Shao,

Von der Malsburg, and Hegde 2013). These two distinct rescue pathways, one on intact mRNA by the RQT complex (and Slh1) and the other on endonucleolytically cleaved mRNA by Dom34, might provide the cell with redundant means to deal with toxic mRNP intermediates.

Proteotoxic stress is a critical problem for all cells and elaborate quality control systems have evolved to minimize its effects as made clear by the exquisite sensitivity of neurons to defects in these pathways (Bengtson and

Joazeiro 2010; Ishimura et al. 2014). The synergistic contributions of exonucleases and endonucleases that we delineate here for targeting problematic mRNAs for decay are critical for ensuring that proteotoxic stress is minimized. Future studies will delineate the biological targets and conditions in which these systems function.

36

Figure 1: Yeast overexpression screens identify a novel factor involved in NGD.

(A) Schematic of reporters used in genetic screens. (B) Normalized reporter GFP levels. Violin plots show flow cytometry data from > 4000 WT cells containing the indicated reporter. (C-D) Plots of Z-scores overexpression screens comparing the NGD-CGA and NGD-AAA reporters to OPT. Z-scores reflecting the

37 significance of log2(GFP/RFP) values from each strain are plotted against each other for the two reporters. Dashed lines represent cutoffs at a Z-score greater than 2.5 or less than -2.5 for each reporter. Blue dots represent deletion strains that have a Z-score value outside the cut-off for the NGD reporters, but not the

OPT. Red dots represent strains of interest in this study. (E) Validation data for overexpression screen hit CUE2. Top, means for 3 individual flow cytometry experiments are plotted for the NGD-CGA reporter strain without CUE2 overexpression (left), and with CUE2 overexpression (right). Bottom, northern blots of steady state mRNA levels for the same strains.

38

Figure 2: Overexpression screens identify NGD-related factors. (A) Northern and western blot analysis for the indicated strains and reporters. (B) Venn diagram of top outliers from the overexpression screen. (C) Raw flow cytometry data from

>4000 cells from empty vector and CUE2 overexpression strains. This data is from one replicate of the triplicate in Figure 1E.

39

Figure 3: Cue2 is the endonuclease what cleaves mRNA during No Go Decay.

(A) Domain organization of S. cerevisiae Cue2, with the “NoSMR” mutant schematic below the WT CUE2. (B) Superimposition of homology model of Cue2

40 SMR domain (cyan) (templated on human N4BP2 SMR; PDB: 2VKC (Diercks et al. 2008)) on full-length IF3 (orange, PDB: 5LMQ (Hussain et al. 2016)).

Positions of Arg-159 of IF3-CTD and Arg-402 of Cue2-SMR are indicated. (C)

Putative positioning of Cue2-SMR in the context of IF3 and tRNAfMet bound to the small ribosome subunit. 30S PIC-2A (as described in Hussain et al., 2016

(Hussain et al. 2016)) is light gray; mRNA is dark gray; tRNAfMet is blue; IF3 is orange; Cue2-SMR is cyan. (D) Cue2-SMR homology model in the same orientation as seen in Fig. S2E, showing putative positions of Asp-348, His-350 and Arg-402 with 30S pre-initiation complex (PIC) with tRNAfMet density subtracted (PDB: 5LMQ, State 2A (Hussain et al. 2016)). (E) Mutation analysis of

Cue2 SMR domain. Log2(GFP/RFP) flow data with the indicated overexpression constructs are shown. P-values for the utant HA-CUE2 overexpression genes as compared to the wildtype HA-CUE2 overexpression gene are 0.007195, 0.02602,

0.03857, 0.7788, 0.002127 and 0.002869, respectively. (F-G) Northern blot analysis of the indicated strains, with full-length mRNA and the 5’ (F) and 3’ (G) fragments labeled; RFP probed for normalization. (H) Northern blot analysis of

NGD-CGA cleavage fragments in the ski2Δ strain with the indicated changes to the endogenous CUE2 locus; RFP probed for normalization.

41 Figure 4: Cue2 homology modeling and mutational analysis. (A) Sequence alignment of CUE domains of representative proteins. Conserved residues depicted in red on yellow background. Putative ubiquitin binding domains

42 indicated by asterisks. (B) Sequence alignment of SMR domains of representative proteins. Identical residues shown in white on red background; conserved residues shown in red on yellow background. (C) Structure-based sequence alignment of the SMR domain of Saccharomyces cerevisiae Cue2

(residues 344-440) with the C-terminal domain (CTD) of Thermus thermophilus translation initiation factor 3 (IF-3) (residues 82-171). Identical residues depicted in white on red background; conserved residues depicted in red on yellow background. (D) Superimposition of the CTD of IF3 (orange, PDB: 1TIG (Biou,

Shu, and Ramakrishnan 1995)) and a homology model of the SMR domain of

Cue2 (cyan) showing structural conservation between the two domains. (E)

Modeling of the Cue2-SMR homology model in the context of IF3 and tRNAfMet bound 30S pre-initiation complex (PIC) (PDB: 5LMQ, State 2A (Hussain et al.

2016) , also see Figure 2C-D). Light gray, 30S; dark gray, mRNA; blue, tRNAfMet; orange, IF3; cyan, Cue2-SMR. Vignettes show the position of Arg-159 of IF3 near the AUG-start codon at the P site, with the putative position of the catalytic

Arg-402 of Cue2-SMR in the vicinity of the mRNA, based on homology modeling.

(F) Alternate view of IF3-CTD (top panel) bound to 30S pre-initiation complex

(PIC) with the tRNAfMet density subtracted (PDB: 5LMQ, State 2A, (Hussain et al.

2016)), and N4BP2 SMR domain (bottom panel) showing putative positions of the corresponding Asp-1692, His-1694 and Arg-1731 residues (PDB: 2D9I

(Diercks et al. 2008)). (G) Western blot analysis of HA tagged WT and mutant

Cue2 overexpression levels for flow cytometry data in Figure 2E. (H) Western

43 blot analysis of HA tagged endogenous WT and mutant Cue2 levels for northern blot analysis in Figure 2H.

44

Figure 5: Canonical decay by Xrn1 is the major contributor to No Go Decay. (A)

Northern blot analysis of full length OPT and NGD-CGA reporter levels in the indicated strains. (B) Quantitated northern blot signal from full length GFP/RFP mRNA levels for the indicated strain, in triplicate.

45 Figure 6: Ribosome profiling analysis of NGD on reporter mRNAs in various genetic backgrounds. (A-F) 16, 21 and 28 nt RPFs mapped to NGD-CGA reporter in ski2∆ (A), slh1∆ ski2∆ (B), dom34∆ ski2∆) (C , cue2∆ dom34∆ ski2∆

46 (D), slh1∆ dom34∆ ski2∆ (E), and cue2∆ slh1∆ dom34∆ ski2∆ (F) strains with schematic depicting Cue2-mediated cleavage in the A sites of collided ribosomes. (G) Schematic of the precise Cue2 cleavage location on the mRNA, relative to collided ribosomes. (H) Comparison of the combined 21 and 28 nt ribosome occupancies on GFP, from 300 nt upstream of the (CGA)12 to the end of the (CGA)12 sequence, normalized to RFP for the indicated strains (n=2). p- value from Student’s t-test is indicated by asterisks. **, p< 0.01.

47

Figure 7: Comparison between monosome 21 nt RPFs and disome footprints.

Disome footprints show predominant ribosome stall site at the 2nd to 5th CGA codons.

48 Figure 8: 16, 21 and 28 nt RPFs on the NGD-CGA reporter from hel2∆ dom34∆ ski2∆ (A) and hel2∆ slh1∆ dom34∆ ski2∆ (B) strains.

49

Figure 9: In vitro cleavage of purified Cue2-SMR. (A) Sucrose gradients of undigested lysates from cells with low dose cycloheximide treatment (blue),

MNase digested lysates from cells with no drug treatment (gray), and MNase digested lysates from cells with low dose cycloheximide treatment (black). A260s are normalized to area under the curve, excluding mRNPs. (B) Sucrose gradient

50 of nuclease resistant trisomes, treated with no enzyme (black), the SMR domain of Cue2 (pink), or the SMR-R402A mutant domain (orange). (C) In vivo 16 nt

RPFs from the slh1∆ dom34∆ ski2∆ strain (cyan) compared to the 60-65 nt RPFs from the combined fractions (mono-, di-and trisomes) treated with no enzyme

(black), SMR domain (pink), or R402-SMR (orange) in vitro. Arrowheads indicate the positions where the in vitro coincide with the in vivo cleavages.

51 Figure 10: Cue2 targets prematurely polyadenylated mRNAs genome-wide for

NGD. (A) Ratio of 16 nt over 20-32 nt RPFs plotted in dom34∆ ski2∆ and cue2∆ dom34∆ ski2∆ strains. Genes in orange correspond to those with reproducibly decreased 16 nt RPFs upon CUE2 deletion. Pink dots indicate genes where premature polyadenylation was identified empirically from sequencing data.

NGD-CGA reporter, YAP1 and RNA14 are labeled and shown in green, blue and purple, respectively. (B) Overlap (red) between annotated prematurely polyadenylated genes (pink) and genes on which CUE2 deletion had a substantial effect on 16 nt RPFs (orange). (C) Examples of 16 nt RPFs mapped to genes, RNA14 (left) and YAP1 (right), in dom34∆ ski2∆ and cue2∆ dom34∆

52 ski2∆ strains with known premature polyadenylation sites indicated (grey traces, from Pelechano et al., 2013 (Pelechano, Wei, and Steinmetz 2013)).

53 Figure 11: Cue2-dependent 16 nt RPFs on prematurely polyadenylated mRNAs. (A) Examples of reads with untemplated A’s on RNA14 (top) and YAP1

(bottom). (B) Ratio of 16 nt over 20-32 nt RPFs plotted from slh1∆ dom34∆ ski2∆ and cue2∆ slh1∆ dom34∆ ski2∆ strains. Genes in orange indicate those with reproducibly decreased 16 nt RPFs upon CUE2 deletion. Pink dots indicate genes with premature polyadenylation identified empirically. NGD-CGA reporter,

YAP1 and RNA14 are labeled and shown in green, blue and purple, respectively

(C) Examples of 16 nt RPFs mapped to genes, RNA14 (left) and YAP1 (right) in

54 slh1∆ dom34∆ ski2∆ and cue2∆ slh1∆ dom34∆ ski2∆ strains with known premature polyadenylation sites (grey traces, from Pelechano et al. 2013).

55

Figure 12: Multiple converging pathways at NGD substrates. Proposed model for signaling mRNA decay of problematic messages in yeast. Ribosomes are grey; open reading frame is yellow; ‘xxx’ indicates a stalling sequence in the mRNA.

56 Table 1: OPT reporter OE screen

Med Mean SGD Symbol log(GFP/RFP) log(GFP/RFP) Z_LOG YLR341W SPO77 -0.33 -0.39 -5.68 YPL105C SYH1 -0.59 -0.60 -5.40 YNL091W NST1 -0.50 -0.47 -5.18 YNL188W KAR1 -0.28 -0.27 -5.14 YLR303W MET17 -0.22 -0.22 -4.97 YGL003C CDH1 -0.21 -0.23 -4.40 YJL129C TRK1 -0.37 -0.38 -4.24 YGR285C ZUO1 -0.24 -0.22 -4.16 YJL144W NA -0.23 -0.22 -4.09 YIL131C FKH1 -0.32 -0.32 -4.01 YIR013C GAT4 -0.20 -0.21 -3.98 YER123W YCK3 -0.38 -0.36 -3.89 YIL022W TIM44 -0.25 -0.25 -3.72 YDL186W NA -0.14 -0.16 -3.58 YPL248C GAL4 -0.42 -0.39 -3.52 YER165W PAB1 -0.15 -0.18 -3.31 YFL037W TUB2 -0.31 -0.30 -3.23 YML058W-A HUG1 -0.14 -0.17 -3.18 YPL089C RLM1 -0.15 -0.14 -3.09 YIL006W YIA6 -0.22 -0.21 -3.09 YBR222C PCS60 -0.22 -0.21 -3.08 YML017W PSP2 -0.21 -0.20 -3.07 YMR199W CLN1 -0.21 -0.20 -3.01 YDR356W SPC110 -0.19 -0.13 -2.99 YGR178C PBP1 -0.21 -0.21 -2.97 YHR082C KSP1 -0.33 -0.31 -2.93 YEL043W NA -0.24 -0.26 -2.93 YGL210W YPT32 -0.15 -0.15 -2.81 YNR060W FRE4 -0.32 -0.31 -2.80 YNL173C MDG1 -0.22 -0.19 -2.77 YOR208W PTP2 -0.33 -0.30 -2.74 YOR345C NA -0.11 -0.12 -2.74 YDR068W DOS2 -0.17 -0.15 -2.71 YBR301W PAU24 -0.11 -0.12 -2.70 YLR314C CDC3 -0.22 -0.18 -2.68 YBR143C SUP45 -0.19 -0.18 -2.66 YOR352W NA -0.37 -0.34 -2.66 YGL216W KIP3 -0.26 -0.18 -2.62

57 YOL128C YGK3 -0.18 -0.17 -2.62 YCL029C BIK1 -0.20 -0.17 -2.56 YHR120W MSH1 -0.18 -0.17 -2.56 YPL054W LEE1 -0.21 -0.19 -2.54 YBL107C MIC23 -0.28 -0.28 -2.50 YBR172C SMY2 -0.13 -0.11 -2.47 YEL031W SPF1 -0.16 -0.17 -2.46 YDR040C ENA1 -0.12 -0.11 -2.44 YGR106C VOA1 -0.18 -0.16 -2.43 YPL270W MDL2 -0.10 -0.13 -2.42 YEL014C NA -0.18 -0.16 -2.39 YDR471W RPL27B -0.20 -0.19 -2.38 YGR068C ART5 -0.25 -0.22 -2.37 YLR203C MSS51 -0.17 -0.16 -2.36 YLR259C HSP60 -0.15 -0.16 -2.35 YLL058W NA -0.22 -0.21 -2.35 YPR032W SRO7 -0.17 -0.16 -2.34 YFL016C MDJ1 -0.10 -0.10 -2.32 YIL105C SLM1 -0.08 -0.19 -2.32 YFL033C RIM15 -0.19 -0.20 -2.30 YOL020W TAT2 -0.19 -0.18 -2.30 YML007W YAP1 -0.15 -0.16 -2.29 YOL135C MED7 -0.12 -0.10 -2.28 YLR287C-A RPS30A -0.09 -0.10 -2.28 YJL047C RTT101 -0.23 -0.21 -2.28 YFL015C NA -0.09 -0.12 -2.25 YHL036W MUP3 -0.17 -0.15 -2.22 YLR227W-A NA -0.10 -0.10 -2.22 YPL189C-A COA2 -0.15 -0.15 -2.22 YHR143W-A RPC10 -0.12 -0.12 -2.21 YHR105W YPT35 -0.12 -0.12 -2.20 YGR039W NA -0.16 -0.15 -2.19 YOL125W TRM13 -0.27 -0.24 -2.19 YGR290W NA -0.07 -0.10 -2.18 YJL223C PAU1 -0.08 -0.10 -2.18 YGL073W HSF1 -0.19 -0.20 -2.18 YMR035W IMP2 -0.11 -0.12 -2.17 YHR113W APE4 -0.20 -0.19 -2.15 YER099C PRS2 -0.08 -0.09 -2.15 YAL034C FUN19 -0.21 -0.23 -2.15 YDR475C JIP4 -0.17 -0.14 -2.14 YJL154C VPS35 -0.21 -0.19 -2.14 YLR141W RRN5 -0.16 -0.14 -2.13

58 YPL160W CDC60 -0.29 -0.27 -2.13 YLL009C COX17 -0.20 -0.20 -2.13 YGR268C HUA1 -0.11 -0.11 -2.12 YPL237W SUI3 -0.15 -0.14 -2.12 YOR358W HAP5 -0.15 -0.16 -2.11 YNL100W AIM37 -0.07 -0.09 -2.11 YBL055C NA -0.11 -0.11 -2.10 YOL001W PHO80 -0.21 -0.19 -2.09 YBR212W NGR1 -0.16 -0.14 -2.08 YLR246W ERF2 -0.07 -0.09 -2.08 YDR032C PST2 -0.13 -0.11 -2.08 YKR091W SRL3 -0.15 -0.14 -2.05 YHR202W NA -0.15 -0.15 -2.05 YLR347C KAP95 -0.12 -0.14 -2.04 YJR100C NA -0.12 -0.14 -2.03 YOR147W MDM32 -0.21 -0.19 -2.03 YFR004W RPN11 -0.08 -0.11 -2.02 YOL027C MDM38 -0.18 -0.14 -2.01 YHR198C AIM18 -0.17 -0.18 -2.01 YEL024W RIP1 -0.08 -0.11 -2.01 YJR032W CPR7 -0.14 -0.14 -2.00 YMR298W LIP1 0.20 0.15 2.00 YDR229W IVY1 0.13 0.14 2.01 YHR066W SSF1 0.14 0.13 2.01 YHR119W SET1 0.15 0.14 2.01 YOR195W SLK19 0.18 0.18 2.02 YGL038C OCH1 0.11 0.15 2.02 YBR094W PBY1 0.22 0.26 2.03 YIL054W NA 0.22 0.20 2.04 YLR438W CAR2 0.40 0.17 2.06 YGL116W CDC20 0.26 0.26 2.06 YJR034W PET191 0.27 0.21 2.06 YMR256C COX7 0.26 0.21 2.08 YMR086C-A NA 0.21 0.21 2.08 YJL122W ALB1 0.15 0.15 2.09 YMR080C NAM7 0.18 0.19 2.09 YMR107W SPG4 0.22 0.21 2.09 YER130C COM2 0.13 0.16 2.09 YCL068C NA 0.11 0.11 2.11 YLR059C REX2 0.08 0.09 2.12 YDL183C NA 0.15 0.16 2.14 YFR022W ROG3 0.12 0.16 2.15 YBR066C NRG2 0.11 0.11 2.15

59 YPR123C NA 0.09 0.11 2.16 YLR265C NEJ1 0.11 0.12 2.16 YIL127C NA 0.16 0.14 2.17 YLR154W-F NA 0.27 0.24 2.17 YKL169C NA 0.07 0.09 2.17 YJL103C GSM1 0.24 0.24 2.18 YER068W MOT2 0.10 0.09 2.18 YMR071C TVP18 0.22 0.22 2.19 YDR096W GIS1 0.13 0.15 2.19 YNL137C NAM9 0.21 0.15 2.19 YNL298W CLA4 0.07 0.12 2.20 YOR008C SLG1 0.22 0.16 2.20 YBR058C UBP14 0.16 0.15 2.20 YPR200C ARR2 0.10 0.12 2.20 YHR162W MPC2 0.23 0.22 2.20 YKL069W NA 0.18 0.18 2.20 YIL051C MMF1 0.22 0.22 2.20 YGL056C SDS23 0.16 0.16 2.21 YKL172W EBP2 0.25 0.24 2.23 YGR220C MRPL9 0.09 0.12 2.25 YGL122C NAB2 0.11 0.10 2.25 YMR032W HOF1 0.06 0.10 2.25 YNL224C SQS1 0.12 0.16 2.27 YNL154C YCK2 0.17 0.15 2.27 YPL032C SVL3 0.27 0.25 2.27 YLR363C NMD4 0.24 0.23 2.28 YLR218C COA4 0.21 0.21 2.28 YPR008W HAA1 0.10 0.12 2.29 YOR331C NA 0.07 0.10 2.29 YIL074C SER33 0.17 0.16 2.30 YML081W TDA9 0.17 0.16 2.30 YBR160W CDC28 0.22 0.21 2.30 YOR273C TPO4 0.16 0.15 2.33 YDL224C WHI4 0.17 0.17 2.35 YDL198C GGC1 0.15 0.18 2.36 YOR338W NA 0.12 0.18 2.37 YIL151C ESL1 0.28 0.26 2.38 YNL180C RHO5 0.22 0.22 2.40 YNL167C SKO1 0.17 0.17 2.41 YER164W CHD1 0.19 0.22 2.43 YHR015W MIP6 0.10 0.11 2.43 YKL043W PHD1 0.11 0.13 2.43 YBR197C NA 0.25 0.24 2.43

60 YBR030W RKM3 0.23 0.22 2.44 YMR112C MED11 0.25 0.24 2.44 YLR301W HRI1 0.24 0.24 2.44 YFR043C IRC6 0.29 0.31 2.45 YHR115C DMA1 0.18 0.16 2.45 YMR286W MRPL33 0.29 0.24 2.46 YHR086W NAM8 0.14 0.17 2.46 YGL017W ATE1 0.27 0.27 2.46 YPL252C YAH1 0.21 0.20 2.51 YBL047C EDE1 0.18 0.18 2.51 YPL190C NAB3 0.19 0.17 2.52 YHL024W RIM4 0.10 0.11 2.55 YLR296W NA 0.26 0.25 2.56 YDR034C LYS14 0.33 0.33 2.56 YDR222W NA 0.10 0.14 2.57 YER148W SPT15 0.24 0.23 2.62 YIL045W PIG2 0.17 0.17 2.62 YLR266C PDR8 0.24 0.24 2.64 YPL011C TAF3 0.20 0.18 2.64 YMR187C NA 0.28 0.29 2.68 YCL048W SPS22 0.25 0.22 2.70 YMR258C ROY1 0.09 0.12 2.71 YNL189W SRP1 0.20 0.18 2.76 YLR228C ECM22 0.17 0.21 2.77 YDR464W SPP41 0.32 0.31 2.79 YCR065W HCM1 0.26 0.25 2.80 YBR208C DUR1,2 0.30 0.31 2.81 YHR203C RPS4B 0.10 0.12 2.82 YPL125W KAP120 0.12 0.12 2.85 YJR035W RAD26 0.27 0.26 2.86 YBR074W PFF1 0.26 0.26 2.90 YBR083W TEC1 0.22 0.20 2.92 YML042W CAT2 0.24 0.22 2.96 YIL147C SLN1 0.13 0.13 3.00 YGL071W AFT1 0.19 0.20 3.03 YPR129W SCD6 0.15 0.16 3.06 YJR035W RAD26 0.29 0.29 3.21 YER106W MAM1 0.17 0.18 3.43 YBR233W PBP2 0.29 0.31 3.46 YDR151C CTH1 0.20 0.19 3.52 YDR043C NRG1 0.21 0.19 3.57 YML117W NAB6 0.39 0.40 3.61 YOL124C TRM11 0.41 0.39 3.63

61 YMR111C NA 0.32 0.33 3.64 YMR182C RGM1 0.21 0.26 3.65 YMR177W MMT1 0.16 0.20 3.69 YLR199C PBA1 0.34 0.34 3.74 YOR032C HMS1 0.41 0.40 3.74 YNL020C ARK1 0.26 0.25 3.75 YOR359W VTS1 0.28 0.27 3.84 YLR220W CCC1 0.27 0.21 3.88 YGL207W SPT16 0.35 0.36 3.89 YLR082C SRL2 0.29 0.28 4.05 YGR116W SPT6 0.27 0.28 4.05 YPL128C TBF1 0.47 0.44 4.10 YDR306C NA 0.34 0.37 4.13 YKR096W ESL2 0.22 0.33 4.14 YOR166C SWT1 0.51 0.48 4.39 YBR250W SPO23 0.37 0.40 4.48 YHR177W NA 0.45 0.37 4.55 YHR046C INM1 0.47 0.50 4.67 YOR113W AZF1 0.43 0.42 4.74 YBR289W SNF5 0.42 0.42 4.76 YDL189W RBS1 0.35 0.34 4.96 YOL089C HAL9 0.64 0.64 5.03 YOR166C SWT1 0.59 0.59 5.38 YFR023W PES4 0.43 0.41 6.18

62 Table 2: NGD-CGA reporter OE screen SGD Symbol Avg_medlogratioNorm Avg_meanlogratioNorm Z_LOG YKL090W CUE2 -0.79 -0.79 -7.15 YBR172C SMY2 -0.67 -0.67 -5.98 YDR515W SLF1 -0.69 -0.68 -5.67 YHR206W SKN7 -0.42 -0.41 -5.61 YPL105C SYH1 -0.71 -0.69 -5.09 YGR271W SLH1 -0.43 -0.41 -4.87 YER165W PAB1 -0.63 -0.60 -4.71 YJR078W BNA2 -0.46 -0.43 -4.58 YOR345C NA -0.35 -0.34 -4.57 YDL067C COX9 -0.50 -0.47 -4.02 YLR073C RFU1 -0.50 -0.45 -3.85 YPL248C GAL4 -0.56 -0.51 -3.78 YDR040C ENA1 -0.42 -0.42 -3.69 YLR303W MET17 -0.40 -0.41 -3.61 YBR276C PPS1 -0.41 -0.40 -3.59 YML007W YAP1 -0.36 -0.33 -3.49 YDL186W NA -0.26 -0.25 -3.46 YPL270W MDL2 -0.28 -0.26 -3.46 YOR352W TFB6 -0.47 -0.46 -3.45 YOR352W NA -0.37 -0.36 -3.39 YIL055C NA -0.40 -0.41 -3.37 YKL109W HAP4 -0.33 -0.34 -3.35 YGL141W HUL5 -0.35 -0.34 -3.33 YIL119C RPI1 -0.26 -0.28 -3.32 YNL091W NST1 -0.36 -0.35 -3.22 YJL013C MAD3 -0.43 -0.43 -3.16 YKL165C MCD4 -0.27 -0.27 -3.15 YOR069W VPS5 -0.41 -0.39 -3.04 YER070W RNR1 -0.41 -0.39 -3.03 YOL130W ALR1 -0.24 -0.23 -3.00 YML130C ERO1 -0.22 -0.21 -2.90 YFR040W SAP155 -0.38 -0.37 -2.87 YBR198C TAF5 -0.21 -0.21 -2.86 YJL144W NA -0.34 -0.33 -2.86 YKR021W ALY1 -0.38 -0.36 -2.86 YKL028W TFA1 -0.36 -0.33 -2.85 YDR146C SWI5 -0.20 -0.21 -2.84 YDL066W IDP1 -0.29 -0.29 -2.81 YDL059C RAD59 -0.20 -0.21 -2.77 YMR034C NA -0.38 -0.35 -2.75

63 YFL015C NA -0.37 -0.35 -2.73 YLR227W-A NA -0.31 -0.30 -2.68 YKL018C-A NA -0.28 -0.27 -2.65 YOL008W COQ10 -0.32 -0.31 -2.63 YPL212C PUS1 -0.26 -0.26 -2.62 YLL058W NA -0.29 -0.26 -2.61 YLR259C HSP60 -0.19 -0.19 -2.59 YHR079C IRE1 -0.28 -0.29 -2.57 YAL010C MDM10 -0.26 -0.26 -2.54 YGL027C CWH41 -0.30 -0.29 -2.51 YBL097W BRN1 -0.21 -0.21 -2.49 YOR291W YPK9 -0.28 -0.27 -2.49 YPR048W TAH18 -0.29 -0.28 -2.49 YER039C-A NA -0.20 -0.18 -2.48 YDL111C RRP42 -0.32 -0.31 -2.48 YIL073C SPO22 -0.35 -0.31 -2.46 YLR141W RRN5 -0.22 -0.23 -2.45 YHR082C KSP1 -0.27 -0.25 -2.43 YLR246W ERF2 -0.19 -0.18 -2.42 YMR233W TRI1 -0.30 -0.28 -2.42 YML006C GIS4 -0.28 -0.26 -2.41 YBL076C ILS1 -0.33 -0.30 -2.41 YHR007C ERG11 -0.24 -0.23 -2.40 YGL049C TIF4632 -0.35 -0.32 -2.40 YGL023C PIB2 -0.30 -0.29 -2.39 YMR137C PSO2 -0.30 -0.30 -2.38 YBR175W SWD3 -0.15 -0.17 -2.37 YGR035W-A NA -0.29 -0.26 -2.36 YNL218W MGS1 -0.18 -0.17 -2.36 YAL009W SPO7 -0.17 -0.18 -2.34 YGR290W NA -0.18 -0.17 -2.34 YBR115C LYS2 -0.25 -0.25 -2.33 YPL218W SAR1 -0.29 -0.29 -2.32 YDR028C REG1 -0.21 -0.23 -2.27 YPL246C RBD2 -0.30 -0.28 -2.26 YOL020W TAT2 -0.28 -0.28 -2.26 YLR406C RPL31B -0.24 -0.23 -2.25 YLR057W MNL2 -0.16 -0.17 -2.24 YPL160W CDC60 -0.24 -0.24 -2.24 YGR268C HUA1 -0.18 -0.17 -2.24 YBR105C VID24 -0.17 -0.17 -2.24 YML079W NA -0.24 -0.25 -2.24 YDR068W DOS2 -0.26 -0.26 -2.23

64 YGR063C SPT4 -0.17 -0.16 -2.22 YLL042C ATG10 -0.31 -0.30 -2.22 YHR101C BIG1 -0.22 -0.21 -2.21 YER123W YCK3 -0.24 -0.24 -2.21 YLR427W MAG2 -0.26 -0.26 -2.20 YDL213C NOP6 -0.29 -0.28 -2.20 YDR160W SSY1 -0.24 -0.23 -2.20 YJR025C BNA1 -0.15 -0.16 -2.20 YCL037C SRO9 -0.23 -0.23 -2.19 YJL129C TRK1 -0.25 -0.22 -2.18 YGR239C PEX21 -0.29 -0.30 -2.18 YER144C UBP5 -0.30 -0.28 -2.17 YPL226W NEW1 -0.30 -0.28 -2.17 YDR255C RMD5 -0.27 -0.25 -2.16 YPL153C RAD53 -0.16 -0.16 -2.16 YPL122C TFB2 -0.20 -0.20 -2.15 YDL199C NA -0.18 -0.18 -2.13 YJL214W HXT8 -0.20 -0.22 -2.13 YNL100W AIM37 -0.14 -0.16 -2.13 YBR068C BAP2 -0.23 -0.23 -2.13 YML119W NA -0.22 -0.21 -2.12 YBR056W NA -0.17 -0.18 -2.11 YPL276W NA -0.24 -0.25 -2.09 YBR301W PAU24 -0.17 -0.15 -2.09 YGL022W STT3 -0.25 -0.26 -2.08 YGL172W NUP49 -0.25 -0.23 -2.07 YOR117W RPT5 -0.21 -0.21 -2.07 YOR311C DGK1 -0.20 -0.19 -2.07 YDR022C ATG31 -0.25 -0.24 -2.07 YFL033C RIM15 -0.18 -0.17 -2.06 YLR271W NA -0.22 -0.24 -2.05 YIL041W GVP36 -0.19 -0.17 -2.05 YLL033W IRC19 -0.29 -0.28 -2.04 YLR285C-A NA -0.14 -0.15 -2.04 YOR358W HAP5 -0.24 -0.24 -2.03 YBL011W SCT1 -0.23 -0.24 -2.02 YKL021C MAK11 -0.21 -0.21 -2.02 YJR091C JSN1 -0.23 -0.25 -2.02 YIL022W TIM44 -0.14 -0.15 -2.01 YFL041W FET5 -0.19 -0.19 -2.00 YMR246W FAA4 -0.19 -0.20 -2.00 YFL054C NA 0.22 0.24 2.01 YIL035C CKA1 0.28 0.28 2.01

65 YBL043W ECM13 0.24 0.23 2.01 YML065W ORC1 0.14 0.15 2.03 YDR124W NA 0.23 0.21 2.03 YHR153C SPO16 0.28 0.25 2.03 YMR071C TVP18 0.26 0.28 2.04 YDL183C NA 0.25 0.25 2.05 YOR106W VAM3 0.17 0.15 2.05 YGL179C TOS3 0.18 0.19 2.06 YDR364C CDC40 0.15 0.15 2.06 YOR260W GCD1 0.25 0.26 2.07 YER151C UBP3 0.22 0.24 2.10 YIL079C AIR1 0.25 0.24 2.10 YLR363C NMD4 0.29 0.29 2.10 YNL139C THO2 0.26 0.26 2.11 YOR370C MRS6 0.25 0.24 2.11 YOL031C SIL1 0.12 0.16 2.11 YHR015W MIP6 0.22 0.24 2.11 YBR250W SPO23 0.20 0.22 2.12 YOR082C NA 0.24 0.16 2.13 YOL124C TRM11 0.21 0.22 2.13 YGL096W TOS8 0.18 0.16 2.14 YBR025C OLA1 0.27 0.29 2.15 YBL092W RPL32 0.21 0.20 2.16 YIL019W FAF1 0.19 0.18 2.16 YPL012W RRP12 0.25 0.25 2.19 YLR059C REX2 0.19 0.16 2.20 YLR324W PEX30 0.25 0.28 2.21 YPR129W SCD6 0.17 0.17 2.22 YLR153C ACS2 0.23 0.25 2.23 YNR029C NA 0.27 0.24 2.25 YOR020C HSP10 0.20 0.17 2.26 YEL041W YEF1 0.30 0.29 2.27 YPL163C SVS1 0.31 0.31 2.27 YLR215C CDC123 0.31 0.26 2.27 YGL100W SEH1 0.20 0.17 2.28 YBR197C NA 0.32 0.31 2.28 YIL052C RPL34B 0.24 0.25 2.29 YJR090C GRR1 0.29 0.31 2.30 YCR033W SNT1 0.27 0.26 2.30 YIR018W YAP5 0.36 0.32 2.31 YGR014W MSB2 0.22 0.24 2.32 YOR211C MGM1 0.31 0.30 2.32 YDR247W VHS1 0.18 0.17 2.33

66 YNL180C RHO5 0.26 0.26 2.37 YLR224W NA 0.26 0.26 2.38 YOR008C SLG1 0.31 0.29 2.39 YBR030W RKM3 0.19 0.21 2.42 YBR094W PBY1 0.27 0.26 2.44 YJL198W PHO90 0.22 0.21 2.44 YPL214C THI6 0.18 0.18 2.48 YBR208C DUR1,2 0.33 0.34 2.50 YPL128C TBF1 0.23 0.26 2.53 YGL169W SUA5 0.30 0.32 2.53 YCR030C SYP1 0.25 0.29 2.54 YPL011C TAF3 0.31 0.32 2.54 YBR182C SMP1 0.31 0.30 2.56 YNL020C ARK1 0.24 0.24 2.57 YBR081C SPT7 0.28 0.29 2.58 YLR042C NA 0.19 0.20 2.58 YKL029C MAE1 0.19 0.22 2.59 YCR065W HCM1 0.27 0.26 2.61 YMR286W MRPL33 0.33 0.36 2.62 YMR012W CLU1 0.27 0.28 2.62 YNL298W CLA4 0.32 0.34 2.63 YBL047C EDE1 0.32 0.33 2.65 YBR083W TEC1 0.32 0.33 2.68 YOL158C ENB1 0.23 0.26 2.72 YNL224C SQS1 0.33 0.32 2.72 YPL095C EEB1 0.22 0.20 2.73 YPR120C CLB5 0.33 0.34 2.73 YPL159C PET20 0.23 0.23 2.74 YGL034C NA 0.24 0.28 2.74 YCL048W SPS22 0.32 0.35 2.74 YER164W CHD1 0.30 0.30 2.76 YHR203C RPS4B 0.32 0.31 2.79 YMR182C RGM1 0.27 0.32 2.80 YPR065W ROX1 0.26 0.26 2.80 YFR023W PES4 0.29 0.26 2.81 YLR199C PBA1 0.29 0.31 2.83 YMR298W LIP1 0.34 0.35 2.83 YLR045C STU2 0.32 0.29 2.85 YDR306C NA 0.25 0.25 2.91 YGL207W SPT16 0.35 0.32 2.93 YNL047C SLM2 0.23 0.22 2.94 YHR177W NA 0.36 0.37 2.94 YGL224C SDT1 0.22 0.22 2.95

67 YDR043C NRG1 0.35 0.35 3.01 YHR046C INM1 0.34 0.32 3.10 YNL196C SLZ1 0.31 0.32 3.11 YHR088W RPF1 0.21 0.23 3.12 YDL188C PPH22 0.32 0.32 3.14 YGR116W SPT6 0.41 0.39 3.17 YOR166C SWT1 0.44 0.43 3.20 YNL042W BOP3 0.41 0.43 3.22 YMR111C NA 0.33 0.34 3.31 YBL029W NA 0.29 0.26 3.36 YMR177W MMT1 0.44 0.44 3.44 YJR035W RAD26 0.39 0.36 3.60 YML032C RAD52 0.24 0.27 3.60 YOR359W VTS1 0.46 0.46 3.68 YOR032C HMS1 0.39 0.38 3.69 YJR035W RAD26 0.36 0.38 3.71 YLR136C TIS11 0.53 0.53 3.85 YDR096W GIS1 0.37 0.45 3.88 YDR034C LYS14 0.43 0.42 3.91 YHR115C DMA1 0.40 0.37 3.98 YDL224C WHI4 0.43 0.47 4.09 YER106W MAM1 0.32 0.31 4.13 YOR113W AZF1 0.37 0.36 4.22 YDR151C CTH1 0.55 0.50 4.27 YKR096W ESL2 0.58 0.59 4.70 YLR228C ECM22 0.56 0.58 4.76 YBR289W SNF5 0.41 0.41 4.83 YMR258C ROY1 0.37 0.37 4.94 YLR220W CCC1 0.63 0.62 5.34 YOL089C HAL9 0.74 0.73 6.83

68 Table 3: NGD-AAA reporter OE screen SGD Symbol Avg_medlogratioNorm Avg_meanlogratioNorm Z_LOG YKL090W CUE2 -0.78 -0.76 -6.85 YKL165C MCD4 -0.72 -0.73 -6.56 YLR068W FYV7 -0.76 -0.76 -6.50 YPL105C SYH1 -0.66 -0.65 -5.72 YHR206W SKN7 -0.58 -0.57 -5.66 YPL248C GAL4 -0.68 -0.62 -5.49 YDL067C COX9 -0.63 -0.60 -5.24 YLR073C RFU1 -0.62 -0.56 -4.92 YLL003W SFI1 -0.55 -0.53 -4.65 YGR271W SLH1 -0.50 -0.50 -4.52 YOR352W TFB6 -0.49 -0.50 -4.35 YDR515W SLF1 -0.52 -0.51 -4.14 YPL276W NA -0.47 -0.45 -4.06 YJR078W BNA2 -0.32 -0.30 -4.04 YNL091W NST1 -0.44 -0.45 -4.04 YPL226W NEW1 -0.38 -0.38 -3.99 YBR198C TAF5 -0.30 -0.30 -3.98 YCL029C BIK1 -0.42 -0.39 -3.90 YGL141W HUL5 -0.34 -0.39 -3.63 YDR098C GRX3 -0.35 -0.32 -3.49 YKR021W ALY1 -0.36 -0.33 -3.37 YML007W YAP1 -0.28 -0.25 -3.31 YJL092W SRS2 -0.31 -0.32 -3.26 YPL270W MDL2 -0.26 -0.27 -3.12 YER165W PAB1 -0.29 -0.28 -3.02 YDR028C REG1 -0.28 -0.32 -3.00 YDR255C RMD5 -0.31 -0.33 -2.98 YML130C ERO1 -0.23 -0.22 -2.95 YKL010C UFD4 -0.30 -0.29 -2.90 YOR345C NA -0.22 -0.22 -2.87 YHR082C KSP1 -0.34 -0.34 -2.86 YOR352W NA -0.32 -0.33 -2.86 YKL109W HAP4 -0.30 -0.30 -2.84 YBR222C PCS60 -0.28 -0.28 -2.83 YHR084W STE12 -0.28 -0.26 -2.81 YJL051W IRC8 -0.40 -0.35 -2.80 YOL020W TAT2 -0.25 -0.25 -2.79 YAL009W SPO7 -0.24 -0.24 -2.77 YOR311C DGK1 -0.21 -0.21 -2.73 YDL186W NA -0.19 -0.20 -2.72

69 YPL122C TFB2 -0.20 -0.20 -2.70 YCL065W NA -0.34 -0.31 -2.70 YIR001C SGN1 -0.25 -0.27 -2.70 YGR268C HUA1 -0.24 -0.23 -2.69 YLR259C HSP60 -0.25 -0.27 -2.67 YJR060W CBF1 -0.31 -0.29 -2.66 YGR035W-A NA -0.32 -0.29 -2.65 YIL022W TIM44 -0.28 -0.26 -2.63 YBR276C PPS1 -0.30 -0.29 -2.62 YBR115C LYS2 -0.31 -0.30 -2.62 YOR291W YPK9 -0.30 -0.29 -2.61 YER052C HOM3 -0.25 -0.25 -2.55 YJR120W NA -0.21 -0.19 -2.53 YDL111C RRP42 -0.24 -0.23 -2.52 YKR053C YSR3 -0.18 -0.19 -2.52 YDL135C RDI1 -0.26 -0.29 -2.49 YJR025C BNA1 -0.23 -0.25 -2.48 YOR087W YVC1 -0.32 -0.31 -2.48 YOL130W ALR1 -0.21 -0.21 -2.47 YPL160W CDC60 -0.25 -0.28 -2.44 YGL073W HSF1 -0.28 -0.27 -2.44 YBR105C VID24 -0.20 -0.21 -2.41 YMR137C PSO2 -0.23 -0.22 -2.41 YMR152W YIM1 -0.29 -0.28 -2.37 YKR038C KAE1 -0.24 -0.23 -2.36 YPR107C YTH1 -0.27 -0.26 -2.34 YBL031W SHE1 -0.28 -0.24 -2.34 YDL180W NA -0.23 -0.21 -2.33 YKL021C MAK11 -0.27 -0.27 -2.32 YLR207W HRD3 -0.26 -0.26 -2.30 YDR303C RSC3 -0.32 -0.28 -2.30 YGL027C CWH41 -0.30 -0.26 -2.30 YMR257C PET111 -0.18 -0.17 -2.29 YJR084W NA -0.22 -0.21 -2.27 YKL028W TFA1 -0.26 -0.26 -2.27 YLR271W NA -0.24 -0.25 -2.25 YMR233W TRI1 -0.27 -0.25 -2.24 YMR158W-B NA -0.16 -0.17 -2.22 YBR077C SLM4 -0.29 -0.27 -2.19 YGR133W PEX4 -0.26 -0.25 -2.18 YIL055C NA -0.26 -0.27 -2.18 YBR024W SCO2 -0.21 -0.20 -2.16 YFR048W RMD8 -0.25 -0.24 -2.16

70 YML079W NA -0.24 -0.24 -2.15 YKL024C URA6 -0.19 -0.21 -2.13 YMR034C NA -0.23 -0.20 -2.12 YGL222C EDC1 -0.18 -0.18 -2.11 YBR196C PGI1 -0.17 -0.16 -2.11 YER123W YCK3 -0.24 -0.23 -2.11 YFR040W SAP155 -0.27 -0.21 -2.10 YEL042W GDA1 -0.21 -0.19 -2.10 YDL043C PRP11 -0.20 -0.22 -2.09 YJR091C JSN1 -0.20 -0.19 -2.09 YBR049C REB1 -0.24 -0.24 -2.08 YNL127W FAR11 -0.19 -0.19 -2.07 YAL056W GPB2 -0.22 -0.22 -2.07 YDR100W TVP15 -0.17 -0.18 -2.05 YER078W-A NA -0.25 -0.23 -2.05 YDR510W SMT3 -0.24 -0.24 -2.04 YPL246C RBD2 -0.19 -0.18 -2.02 YLR026C SED5 -0.18 -0.18 -2.02 YHR032W ERC1 -0.18 -0.22 -2.02 YLR410W VIP1 -0.24 -0.24 -2.02 YGR031W NA -0.18 -0.15 -2.02 YOR069W VPS5 -0.20 -0.19 -2.01 YLL058W NA -0.26 -0.21 -2.00 YER184C NA 0.23 0.25 2.00 YNL079C TPM1 0.17 0.15 2.02 YPL159C PET20 0.23 0.23 2.02 YDR251W PAM1 0.20 0.20 2.03 YER016W BIM1 0.20 0.20 2.03 YHR153C SPO16 0.26 0.25 2.05 YBR156C SLI15 0.19 0.19 2.05 YNL020C ARK1 0.14 0.16 2.07 YOL128C YGK3 0.16 0.16 2.07 YDR464W SPP41 0.22 0.23 2.08 YGR081C SLX9 0.26 0.25 2.09 YOL103W ITR2 0.24 0.19 2.09 YPL013C MRPS16 0.15 0.16 2.10 YGL148W ARO2 0.27 0.26 2.10 YDR124W NA 0.27 0.25 2.10 YDL025C RTK1 0.28 0.26 2.10 YNL116W DMA2 0.22 0.23 2.10 YCR073W-A SOL2 0.25 0.24 2.13 YDR214W AHA1 0.24 0.24 2.14 YNL196C SLZ1 0.22 0.23 2.14

71 YOL161C PAU20 0.18 0.16 2.14 YGR014W MSB2 0.24 0.23 2.15 YNL078W NIS1 0.25 0.26 2.16 YOR166C SWT1 0.22 0.24 2.16 YMR241W YHM2 0.23 0.24 2.16 YPR129W SCD6 0.20 0.19 2.17 YGL100W SEH1 0.19 0.19 2.17 YPL060W MFM1 0.21 0.19 2.18 YDR075W PPH3 0.25 0.25 2.22 YML123C PHO84 0.20 0.21 2.22 YOR092W ECM3 0.24 0.22 2.22 YPR004C AIM45 0.24 0.21 2.22 YKL222C NA 0.20 0.22 2.23 YCR065W HCM1 0.25 0.24 2.25 YEL073C NA 0.27 0.26 2.25 YJR090C GRR1 0.25 0.24 2.25 YDR132C NA 0.22 0.24 2.25 YLR149C NA 0.23 0.24 2.25 YIL052C RPL34B 0.26 0.25 2.26 YNL224C SQS1 0.20 0.26 2.27 YLR152C NA 0.27 0.26 2.27 YDL121C NA 0.26 0.27 2.28 YJL117W PHO86 0.22 0.21 2.28 YDL188C PPH22 0.27 0.27 2.30 YDR482C CWC21 0.21 0.22 2.30 YHL024W RIM4 0.23 0.21 2.32 YBR173C UMP1 0.24 0.27 2.32 YPR065W ROX1 0.17 0.18 2.33 YBL043W ECM13 0.27 0.26 2.36 YGL169W SUA5 0.20 0.22 2.39 YNL180C RHO5 0.27 0.27 2.39 YLR224W NA 0.28 0.28 2.42 YGR116W SPT6 0.25 0.25 2.48 YOR193W PEX27 0.24 0.25 2.50 YOL136C PFK27 0.24 0.19 2.50 YBR056W-A NA 0.19 0.19 2.50 YGL224C SDT1 0.23 0.22 2.52 YJL115W ASF1 0.24 0.24 2.53 YER164W CHD1 0.24 0.28 2.55 YOR206W NOC2 0.30 0.29 2.56 YHR175W CTR2 0.31 0.31 2.56 YGL162W SUT1 0.29 0.32 2.58 YPL002C SNF8 0.22 0.19 2.58

72 YGR008C STF2 0.31 0.30 2.58 YKR083C DAD2 0.30 0.30 2.59 YPL011C TAF3 0.26 0.26 2.60 YNL108C NA 0.28 0.30 2.63 YGR249W MGA1 0.29 0.29 2.63 YLR042C NA 0.24 0.24 2.68 YLR181C VTA1 0.23 0.30 2.71 YBR289W SNF5 0.30 0.30 2.73 YLR215C CDC123 0.31 0.30 2.73 YMR204C INP1 0.34 0.31 2.75 YBR208C DUR1,2 0.29 0.32 2.80 YNL047C SLM2 0.23 0.25 2.81 YLR453C RIF2 0.21 0.21 2.84 YPL204W HRR25 0.29 0.27 2.90 YMR298W LIP1 0.34 0.36 2.90 YJR090C GRR1 0.31 0.33 2.93 YDL189W RBS1 0.32 0.29 2.97 YBL029W NA 0.30 0.26 2.99 YDR034C LYS14 0.35 0.34 3.00 YHR115C DMA1 0.23 0.23 3.03 YLR092W SUL2 0.29 0.28 3.04 YLR199C PBA1 0.31 0.35 3.12 YMR182C RGM1 0.36 0.36 3.13 YGL207W SPT16 0.37 0.35 3.17 YMR111C NA 0.31 0.35 3.20 YBR094W PBY1 0.36 0.37 3.25 YMR258C ROY1 0.22 0.25 3.27 YOR032C HMS1 0.42 0.40 3.32 YOL124C TRM11 0.38 0.40 3.38 YNR029C NA 0.40 0.39 3.38 YML032C RAD52 0.24 0.25 3.39 YER106W MAM1 0.29 0.30 3.39 YBR083W TEC1 0.31 0.34 3.41 YDR306C NA 0.42 0.39 3.53 YKL043W PHD1 0.38 0.41 3.65 YOR370C MRS6 0.44 0.42 3.67 YOR113W AZF1 0.44 0.44 3.90 YOR166C SWT1 0.44 0.44 3.93 YFR023W PES4 0.29 0.31 4.07 YBR250W SPO23 0.43 0.44 4.10 YHR086W NAM8 0.46 0.47 4.11 YMR177W MMT1 0.40 0.39 4.20 YOR359W VTS1 0.44 0.42 4.22

73 YLR273C PIG1 0.40 0.39 4.25 YLR220W CCC1 0.50 0.51 4.64 YDL224C WHI4 0.53 0.56 4.89 YOL089C HAL9 0.56 0.57 4.97 YLR228C ECM22 0.67 0.67 5.43 YDR151C CTH1 0.71 0.71 6.39 YKR096W ESL2 0.65 0.63 6.77 YMR001C CDC5 0.78 0.78 8.51

74 Chapter III

Yeast genetic screens for factors that act on translationally problematic mRNAs

Summary:

Ribosomes stalling within the open reading frame of mRNAs trigger an array of consequences including translation repression, mRNA decay, ribosome rescue, and peptide degradation. In yeast, deletion screens have been performed to identify factors that stabilize the nascent peptide, but not the other pathways signaled in addition to peptide degradation. We set up deletion and temperature sensitive mutant screens to identify novel factors involved in these processes. In these reverse genetic screens we found a number of novel and known factors that NGD-mRNAs in terms of their translation, decay and ribosome rescue. We briefly discuss follow up experiments being performed to further characterize the factors we identified.

Credit: Experiments performed by KN D’Orazio unless otherwise stated here.

Figure 15C performed by K Kostova.

75 Introduction:

When ribosomes are stalled internally on an mRNA, for example through

RNA damage that prevents decoding, ribosomes pileup and signal mRNA decay through two pathways: Xrn1-dependent canonical mRNA decay and Cue2- dependent endonucleolytic decay (D’Orazio et al. 2019). Ribosome rescue is also signaled and ribosomes are removed from the mRNA through two pathways: RQT dependent splitting via the helicase Slh1 and Dom34/Hbs1 dependent splitting through Cue2-dependent mRNA cleavage and the ATPase

RLI1 (D’Orazio et al. 2019; Matsuo et al. 2017; Shoemaker, Eyler, and Green

2010). These pathways to decay and rescue are intertwined and not mutually exclusive, but it is unclear how one action is chosen over another. We set out to identify novel factors involved in these pathways to find new insights into the regulation of mRNA decay (termed No Go Decay, NGD) and ribosome rescue at ribosome stalling sites.

Results:

Deletion screens to identify factors involved in mRNA decay and translation repression/activation on NGD substrates

With the reporters previously discussed in Chapter 2 of this thesis (Figure

1A), we used high-throughput reverse genetic screens and reporter-synthetic genetic array (R-SGA) methods (Fillingham et al. 2009; Tong et al. 2001) to evaluate the effects of deleting nonessential genes on NGD and ribosome accumulation on stalling reporters. We began by crossing strains carrying the control (OPT) and the two different no-go decay reporters (NGD-CGA and NGD-

76 AAA) into S. cerevisiae deletion libraries (Douglas et al. 2012; Giaever et al.

2002; Y. Hu et al. 2007). For each deletion screen, we used SGA technology to select for haploid strains with our reporter and the corresponding deletion.

Strains were transferred to galactose-rich plates and the GFP and RFP levels were evaluated by fluorimetry. We plotted the results from the screens individually, comparing Z-scores for the log2(GFP/RFP) signals from each NGD reporter strain to Z-scores from the corresponding strain carrying the OPT reporter (Figure 13A and 13B). Normalization with RFP intensity was used to eliminate non-specific factors that impact expression of both RFP and GFP.

The deletion screens revealed a set of candidate genes that strongly impact

GFP levels for the reporters (Figure 13A-13B and Table 5-6). We arbitrarily considered those hits which yielded a Z-score less than -2.5 or greater than 2.5 as candidate genes. We observed a strong overlap in the set of candidate genes for the NGD-CGA and NGD-AAA reporters, but little overlap between those genes affecting NGD and those affecting the OPT reporter suggesting the hits for the NGD reporters indeed are affecting NGD (Figure 13C). Gene Ontology (GO) analysis revealed that the mutants affecting NGD are enriched for specific biological processes related to RNA processing and protein modifications (most involving ubiquitin) (Figure 13D). Notably, NGD-CGA reporter gave a set of hits that seem to be more enriched in specific GO processes than the NGD-AAA reporter, likely because the NGD-CGA reporter has a stronger decay signal that allows for a broader range of altered GFP signal. Importantly, no gene ontology enrichment classes were found in the gene deletions affecting the OPT reporter,

77 suggesting that the optimal candidate genes are impacting the reporter in a non- specific manner.

Temperature sensitive mutant screens

In addition to the deletion library, we utilized the temperature sensitive library consisting of 787 temperature sensitive mutants, representing 497 genes, covering about 45% of known essential genes (Li et al. 2011). We put our OPT,

NGD-CGA, and NGD-AAA reporters into the temperature sensitive library using the same SGA technology used on the deletion screens (Figure 14A). We see a stronger overlap between the top hits from the two NGD reporters as compared to the NGD / OPT reporter overlap, suggesting some candidates from this screen indeed effect the NGD reporters uniquely due to their stalling sequences (Figure

14B). These top hits did not show specific gene ontology enrichment and follow up experiments are not being performed on this screen at this time.

Deletion screen candidates

Comprehensive follow-up flow cytometry experiments on candidate genes from the NGD-CGA deletion screens are currently being completed, however, experiments on select hits have been performed and will be detailed below.

Transfer RNA modification genes are drastically enriched in NGD screens

Transfer RNA modification genes were highly enriched in the NGD-CGA deletion screen and mildly enriched in the NGA-AAA deletion screen. These genes encode for factors that are part of or function together with the tRNA modifier complex Elongator, including ELP2, ELP3, ELP6, IKI3, URM1, UBA4,

NCS2, NCS6, and KTI12 specifically in the top ~25 deletion strains that cause an

78 increase in GFP in the NGD-CGA reporter (Tables 4-6). This screen reveals a link between these tRNA modifiers and NGD that has not yet been evaluated.

Speculatively, it is possible that elongation is affected by altering these tRNA modifiers, which may play a role in how often ribosomes collide. For example, if translation is slow due to poor decoding throughout the reporter, then ribosome collisions are less likely to happen at the induced stalling site.

Translation repressor GIGYF2 homologs Smy2 and Syh1 act on NGD mRNAs

Two GYF domain-containing proteins Syh1 and Smy2, mammalian homologs of which have previously shown to repress translation, were also identified specifically in the NGD screens (Figure 13A-13B, green). These are homologs of the translation repressor GIGYF2 in mammalian cells (Ash et al. 2010), but they have not yet been implicated in NGD or translation repression in yeast. We tested whether the GYF-protein translation repression role was functionally conserved in yeast. To do this, we made strains with deletions of SYH1, SMY2, or both SYH1 and SMY2 in the presence of our optimal and NGD-CGA reporters.

We compared the protein levels by flow cytometry and found that knockout of

SMY2 and SYH1 alone led to a partial increase in the GFP/RFP ratio (Figure

15A). Deleting both homologs led to a greater increase in GFP/RFP compared to the single deletes, suggesting that Smy2 and Syh1 have redundant function.

Surprisingly, the increased expression of the reporter was accompanied by mRNA stabilization as seen by both northern blot and qPCR (Figure 15B-15C).

This observation is consistent with the apparent lack of a yeast homolog of

4EHP, which is an essential part of the mechanism of GIGYF2 translation

79 inhibition in mammals (Hickey et al. 2019). Consistent with this, the yeast homologs lack the 4EHP binding domain found in human GIGYF2 (Hickey et al.

2019), but Smy2 has been shown to bind Eap1, a eIF4E-binding protein implicated in mRNA decapping and degradation (Ash et al. 2010; Hickey et al.

2019). These data suggest that the mechanism of action of these proteins differs between yeast and mammals, but the function to prevent aberrant protein products remains in tact.

Ribosome occupancy and mRNA decay

The RQT complex (Slh1, Cue2, and Rqt4) is a strong hit for both the NGD-

CGA and NGD-AAA reporters (Figure 13A-13B, red) and, as discussed in

Chapter 2, RQT is intricately tied together with NGD (D’Orazio et al. 2019).

Northern analysis of the major hit, slh1Δ, confirms that the full-length NGD mRNA is destabilized. In contrast, deletion of other ribosome rescue factors

Dom34 and Ski2 (factors critical for splitting ribosomes and recruiting the exosome, respectively) do not stabilize the NGD-CGA reporter transcript (Figure

16A-16B). With this data, we asked why does a lack of ribosome rescue through

SLH1-deletion lead to mRNA decay?

Our data from Chapter 2 show that in an xrn1Δ background, mRNA levels go down upon deletion of SLH1 due to an increase in Cue2 cleavage. However, we also showed that Xrn1 is the major contributor to NGD, thus when Xrn1 is present, Cue2 is not a major player for mRNA decay. As expected, upon deletion of Cue2 in the slh1Δ background, there is only a partial rescue of GFP levels when Xrn1 is present (Figure 16C). These data suggest that an increase in

80 ribosome occupancy (due to the SLH1 deletion) results in a stronger decay affect through Xrn1.

The RQC-mediating E3 ligase Hel2 is another major hit in our NGD deletion screens, but not the OPT screen. Previous ribosome quality control screens have identified HEL2 and showed that upon deletion of HEL2, ribosomes read through stalling sequences allowing for stabilization of the full-length protein (a phenotype also seen upon deletion of SLH1). However, as in the deletion of SLH1, our genetic screens report that NGD mRNA levels are drastically decreased in the absence of HEL2 (Figure 13A-13B). Follow up experiments show that indeed upon deletion of HEL2, GFP/RFP levels are down and GFP mRNA levels are down (Figure 16A-16D). We hypothesize that upon deletion of HEL2 or SLH1, ribosomes are not cleared, causing ribosome occupancy on the stalling reporter to increase, and the Xrn1 mediated decay pathway to also drastically increase.

We are currently asking when and how Xrn1 decay is signaled due to an increase in ribosome occupancy. One key insight into this mechanism is the relationship between nonoptimal reporters and mRNA decay. Nonoptimal mRNAs are decayed much more rapidly than their optimal counterparts

(Presnyak et al. 2015). Additionally, mRNA decay rates correlate with decreased codon optimality, so more optimal mRNAs are more likely to have a longer half- life. Decay of nonoptimal mRNAs is thought to occur through canonical Xrn1- dependent mRNA decay, similar to the primary mechanism of decay for our

NGD-CGA reporter (Presnyak et al. 2015; Radhakrishnan et al. 2016). Previous work has elucidated much of the canonical Xrn1 mediated decay pathway in

81 yeast, see review (Parker 2012). We hypothesize that NGD substrates go through the same decay pathway that nonoptimal mRNAs undergo and that ribosome clearing through Hel2/Slh1 infact prevents this decay from happening.

Therefore, when these ribosome rescue factors are deleted, the NGD substrate appears to the cell as an mRNA with very bad codon content, and the cell more rapidly decays it. Experiments are currently underway to test these hypotheses.

Deubiquitinases involved in NGD

The NGD screens revealed many factors associated with ubiquitination, a process known to be important for ribosome rescue and NGD at stall sites

(Ikeuchi et al. 2019; Matsuo et al. 2017). Namely, the candidate genes RUP1,

UBP6, and UBP3 are factors not previously shown to be involved in NGD, and are all interestingly related to deubiquitination (Figure 13A-13B, purple).

However, since UBP3 is also a major hit in the OPT reporter screen, this factor may have more general mRNA or protein effects. Other ubiquitin related factors

HEL2, CUE3, UBI4, and UBC4 have been previously linked to NGD (Ikeuchi et al. 2019).

Rup1 and Ubp6 are two interesting candidate genes that upon deletion cause a decrease in GFP signal. These factors are both relate to deubiqitination and have not yet been evaluated for a role in ribosome quality control. Rup1 is an accessory factor that is in complex with the deubiquitinase (DUB) Ubp2 (a factor that was not in the deletion screen) and the E3 ligase Rsp5 (an essential gene also not in the deletion screen). In addition to Rup1, we analyzed Ubp2 in follow up experiments, but not Rsp5 since it is essential. We deleted Rup1, Ubp2, and

82 Ubp6 from strains with our NGD-CGA reporter and see that GFP levels are indeed down in flow cytometry experiments and mRNA levels are down in northern blots (Figure 17A-17B). Furthermore, none of these hits recapitulate the readthrough phenotype seen in a HEL2-deletion strain nor do they have any effect on protein (Figure 14C). Further experiments need to be performed to investigate how these DUBs are acting on NGD substrates.

Note: BER1 is a gene that was also analyzed in Figure 17. Deleting BER1 causes an increase in GFP and the genetic interactors of Ber1 are significantly similar to Hel2, but we also do not know how Ber1 is acting here. Ber1 seems to have little to no effect on mRNA levels or protein output for NGD.

Discussion:

Our deletion screens revealed a number of hits that have not yet been evaluated for their role in No Go Decay, ribosome rescue, or translation activation/repression. The tRNA modifying genes that function in or with the

Elongator complex are particularly interesting because they are heavily enriched in the NGD screens. It is possible that translation of the stretch of CGA or AAA codons in the NGD reporters is substantially altered by deleting tRNA modifying enzymes, however, the tRNA that decodes CGA is not known to be modified by the Elongator complex (Karlsborn et al. 2014) so this is unlikely. Additionally, due to the directionality of the GFP upon deletion of these factors, it seems NGD substrates survive better without these tRNA modifying enzymes present. Future experiments can be done to elaborate on the relationship between tRNA modifications and NGD.

83 The relationship between ribosome occupancy and mRNA decay is an actively debated topic. There is evidence in the field supporting the hypothesis that ribosomes protect mRNAs from decay, whereas the data presented here and by others in the field suggest slowing or stacking ribosomes increase mRNA decay. Although these two hypotheses seem distinct, they are not mutually exclusive. It is possible that there are differences between slow ribosomes and stalled ribosomes. Furthermore, it is possible that there are differences amongst stalled ribosomes and the mechanism in which the ribosome stalls determines whether it is stabilizing to the mRNA or destabilizing. For example, if ribosomes are stalled using an elongation inhibitor, they might be poised in a rotated state, which may not be a signal to decay the mRNA. New work from the Beckmann

Inada and Coller labs suggests ribosomes with an empty E site may be the signal to decay the mRNA through recognition by Not5 (Buschauer et al. 2019). Our data suggests ribosomes stalled on CGA, with an empty A site and possibly an empty E site are a signal for decay. We also provide evidence that deleting factors that remove ribosomes from this mRNA causes an increase in decay, possibly because there is an increased likelihood of ribosomes with empty E sites. Additional experiments are currently being done to investigate these possibilities and identify the reasons for which the deleting ribosome rescue factors causes an increase in mRNA decay.

Ubiquitination is a major player in NGD and ribosome rescue, therefore, the ubiquitin related genes were particularly interesting as hits in our screens. Hel2 is known to ubiquitinate uS10, also known as RSP20 in yeast, however, whether

84 deubiquitination of uS10 occurs is still unknown. Data suggests ubiquitination of uS10 is not a degradation signal, therefore it is likely that deubiquitination does occur through an unknown DUB. Although we identify two DUBs not yet shown to be involved in NGD, neither had any effect on protein output from the CGA12 reporter as discussed above. However, the effect we would be looking for is an increase in nascent peptide degradation and since we cannot see any nascent peptide to begin with, we would not be able to see an increase in degradation.

The signal from a CGAx12 reporter to degrade the nascent peptide is extremely strong, so in order to be in the range in which we would see an effect from deleting the DUB, we should use a weaker degradation signal (possibly CGAx4).

It is clear that an array of regulatory pathways are involved in keeping translation functioning smoothly and properly. A multitude of pathways work together to resolve stalled ribosomes and prevent the problem from persisting. It seems that depending on the type of ribosomal stall, certain pathways take precedence over others. Future work to investigate how stalls differ molecularly to signal different pathways will elucidate major regulatory pathways in mRNA decay, ribosome rescue, and peptide degradation.

85

Figure 13: Yeast deletion screens reveal candidate genes involved in NGD. (A-

B) Plots of Z-scores from deletion screens from the (A) NGD-CGA reporter and the OPT reporter and (B) NGD-AAA reporter and the OPT reporter. Z-scores reflecting the significance of log2(GFP/RFP) values from each strain are plotted against each other for the two reporters. Dashed lines represent cutoffs at a Z- score greater than 2.5 or less than -2.5 for each reporter. Blue dots represent deletion strains that have a Z-score value outside the cut-off for the NGD-CGA

86 reporter, but not the OPT reporter. Colored dots represent strains of interest in this study for the NGD-CGA reporter. Note: more colored genes are hits in the

NGD-AAA screen as well, but the right plot is unlabeled. (C) Venn diagram of top outliers from the deletion screens. (D) GO term enrichment for candidate genes from deletion screens for the NGD-CGA (blue) and NGD-AAA (orange) reporters.

Note: OPT reporter did not produce any GO term enrichment.

87

Figure 14: Yeast temperature sensitive mutant screens reveal candidate genes involved in NGD. (A) Middle: Plots of Z-scores from temperature sensitive mutant screens from the (top) NGD-CGA reporter and the OPT reporter and (bottom)

NGD-AAA reporter and the OPT reporter. Z-scores reflecting the significance of

88 log2(GFP/RFP) values from each strain are plotted against each other for the two reporters. Dashed lines represent cutoffs at a Z-score greater than 2.5 or less than -2.5 for each reporter. Blue dots represent deletion strains that have a Z- score value outside the cut-off for the NGD-CGA reporter, but not the OPT reporter. The panel to the left is a list of all the blue dots from the top plot as they appear from highest to lowest Z-Score. The panel to the right is a list of all the blue dots from the bottom plot as they appear from highest to lowest Z-Score.

For both panels, green indicates an increase in GFP when the gene is depleted and red indicates a decrease in GFP when the gene is depleted. (B) Venn diagram of top outliers from the deletion screens.

89 Figure 15: Syh1 and Smy2 alter the decay of NGD substrates. (A) Box and whisker plots of flow cytometry data from 5000 cells for each given strain (x-axis, bottom label) and reporter (x-axis top label). (B) Northern blot for mRNA of the indicated reporter and indicated strain. (C) qPCR data for the indicated reporter and indicated strain (Hickey et al. 2019).

90 Figure 16: Ribosome clearance factors alter the decay of NGD substrates. (A)

Violin plots of flow cytometry data from 5000 cells for each given strain. (B) Top panel: Northern blot for mRNA of the indicated reporter and indicated strain, probed using the GFP RNA probe. Middle panel: Western blot for GFP protein levels. Bottom panel: Western blot for flag protein levels. (C) Violin plots of flow cytometry data from 5000 cells for each given strain. (D) Northern blot for mRNA of the indicated reporter and indicated strain, probed using the HIS3 RNA probe.

91 Figure 17: Deubiquitinases are involved in regulating NGD mRNA levels. (A)

Box and whisker plots of flow cytometry data from 5000 cells for each given strain and the NGD-CGA reporter. (B) Northern blot for mRNA of the NGD-CGA reporter and indicated strain, probed with the HIS3 RNA probe. (C) Western Blot analysis of the read-through protein product, probed using the Flag tag antibody.

92 Table 4: OPT reporter deletion screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG UBP3 YER151C -0.58 -0.57 -9.23 PTP2 YOR208W -1.20 -1.16 -8.87 RPS27B YHR021C -0.46 -0.38 -8.14 SIF2 YBR103W -0.29 -0.29 -7.24 MET8 YBR213W -0.33 -0.29 -7.18 RAM1 YDL090C -0.35 -0.27 -6.80 SNT1 YCR033W -0.27 -0.25 -6.19 RIM1 YCR028C-A -0.20 -0.24 -6.11 PXA2 YKL188C -0.68 -0.69 -6.07 PET122 YER153C -0.33 -0.34 -5.58 NA YGR021W -0.53 -0.50 -5.51 NA YOL013W-A -0.70 -0.54 -5.19 MRP8 YKL142W -0.61 -0.58 -5.11 SNC2 YOR327C -0.71 -0.66 -5.05 MET3 YJR010W -0.30 -0.24 -5.00 ALT1 YLR089C -0.80 -0.72 -4.96 NA YGR272C -0.64 -0.63 -4.86 RPS29A YLR388W -0.76 -0.70 -4.78 NA YOR012W -0.41 -0.49 -4.73 NUP53 YMR153W -0.46 -0.52 -4.73 CPR2 YHR057C -0.27 -0.22 -4.71 NA YGR001C -0.46 -0.42 -4.69 NA YMR155W -0.51 -0.52 -4.67 NA YNL200C -0.52 -0.49 -4.55 NA YNL057W -0.50 -0.49 -4.54 LYS7 YMR038C -0.57 -0.49 -4.45 NA YGL242C -0.39 -0.40 -4.44 GIC1 YHR061C -0.41 -0.21 -4.33 EGD2 YHR193C -0.29 -0.20 -4.29 GSP2 YOR185C -0.33 -0.43 -4.13 COY1 YKL179C -0.44 -0.47 -4.10 IMG2 YCR071C -0.15 -0.16 -4.01 MEP3 YPR138C -0.50 -0.51 -3.96 APP2 YMR192W -0.47 -0.43 -3.94 ATR1 YML116W -0.43 -0.43 -3.89 MAL32 YBR299W -0.49 -0.49 -3.76 SLM3 YDL033C -0.15 -0.15 -3.68 NA YLR247C -0.54 -0.53 -3.66 NA YNL013C -0.44 -0.39 -3.65

93 NA YHL029C -0.20 -0.17 -3.63 CMK2 YOL016C -0.33 -0.38 -3.61 NA YLR374C -0.52 -0.52 -3.57 CYT2 YKL087C -0.50 -0.45 -3.47 RFM1 YOR279C -0.47 -0.45 -3.44 DBR1 YKL149C -0.43 -0.39 -3.42 NA YNL100W -0.42 -0.37 -3.41 NA YLR283W -0.55 -0.50 -3.40 NA YFR012W -0.33 -0.31 -3.39 EMI1 YDR512C -0.20 -0.21 -3.32 GAL3 YDR009W -0.29 -0.24 -3.21 TIS11 YLR136C -0.51 -0.46 -3.15 MCK1 YNL307C -0.07 -0.34 -3.14 LCB4 YOR171C -0.37 -0.32 -3.11 BSC5 YNR069C -0.30 -0.32 -3.09 FYV4 YHR059W -0.14 -0.15 -3.08 YTA7 YGR270W -0.12 -0.15 -3.06 FAR11 YNL127W -0.43 -0.33 -3.06 DIP5 YPL265W -0.36 -0.40 -3.05 COX23 YHR116W -0.09 -0.15 -3.04 STP3 YLR375W -0.40 -0.44 -3.03 CGR1 YGL029W -0.42 -0.39 -3.03 NA YLR143W -0.54 -0.44 -3.02 BIO5 YNR056C -0.38 -0.31 -3.00 NA YGR273C -0.40 -0.39 -2.98 DAL5 YJR152W -0.33 -0.33 -2.94 NA YKR018C -0.28 -0.33 -2.89 IST1 YNL265C -0.26 -0.31 -2.88 HIS1 YER055C -0.16 -0.18 -2.87 UBP14 YBR058C -0.11 -0.12 -2.85 HST1 YOL068C -0.18 -0.30 -2.84 NA YMR041C -0.38 -0.31 -2.83 KEX2 YNL238W -0.33 -0.31 -2.83 RPS23A YGR118W -0.25 -0.25 -2.78 PAN3 YKL025C -0.35 -0.32 -2.77 RPL19B YBL027W -0.11 -0.11 -2.76 NA YMR158W-A -0.28 -0.30 -2.76 BPT1 YLL015W -0.33 -0.31 -2.70 MAK3 YPR051W -0.43 -0.35 -2.66 NA YLR364W -0.44 -0.39 -2.65 NA YGR122W -0.32 -0.34 -2.64 HIS7 YBR248C -0.08 -0.11 -2.64 APD1 YBR151W -0.08 -0.11 -2.64

94 SGF11 YPL047W -0.28 -0.35 -2.63 ZRC1 YMR243C -0.31 -0.29 -2.59 FRE7 YOL152W -0.14 -0.27 -2.58 RPS21B YJL136C -0.11 -0.12 -2.57 CTI6 YPL181W -0.32 -0.34 -2.56 MDM31 YHR194W -0.13 -0.12 -2.54 NAB6 YML117W -0.32 -0.28 -2.51 SRO9 YCL037C -0.12 -0.10 -2.50 MSC6 YOR354C -0.36 -0.33 -2.49 NA YNL143C -0.19 -0.27 -2.49 NA YNL320W -0.28 -0.27 -2.48 NA YKR070W -0.21 -0.28 -2.47 OST6 YML019W -0.32 -0.27 -2.47 NA YOR366W -0.32 -0.32 -2.46 NA YNR024W -0.25 -0.26 -2.46 LSM7 YNL147W -0.34 -0.32 -2.44 PTH1 YHR189W -0.10 -0.12 -2.43 ARG5,6 YER069W -0.27 -0.15 -2.40 RPL26A YLR344W -0.34 -0.35 -2.39 LEU4 YNL104C -0.28 -0.26 -2.37 VPS1 YKR001C -0.36 -0.27 -2.36 OMA1 YKR087C -0.34 -0.27 -2.36 SOY1 YBR194W -0.09 -0.09 -2.31 NA YDL119C -0.08 -0.09 -2.30 NA YPL014W -0.31 -0.30 -2.30 HMO1 YDR174W 0.07 -0.17 -2.29 MUP3 YHL036W -0.05 -0.11 -2.26 NMD4 YLR363C -0.25 -0.33 -2.26 ADH4 YGL256W -0.23 -0.20 -2.26 KEX1 YGL203C -0.25 -0.20 -2.26 NA YGL132W -0.20 -0.20 -2.25 NA YOR022C -0.22 -0.23 -2.23 MUM2 YBR057C -0.09 -0.09 -2.22 YSN1 YNR065C -0.32 -0.23 -2.22 PTC1 YDL006W -0.08 -0.09 -2.21 FUI1 YBL042C -0.07 -0.09 -2.21 SNF6 YHL025W -0.14 -0.11 -2.18 ADH1 YOL086C -0.22 -0.28 -2.18 NA YER078C -0.04 -0.13 -2.17 HOS4 YIL112W -0.10 -0.10 -2.15 DBP3 YGL078C -0.14 -0.19 -2.14 YAF9 YNL107W -0.20 -0.23 -2.13 SWC5 YBR231C -0.10 -0.09 -2.13

95 GRR1 YJR090C -0.07 -0.10 -2.11 PAT1 YCR077C -0.06 -0.09 -2.11 NA YBR209W -0.10 -0.09 -2.11 ARG80 YMR042W -0.24 -0.23 -2.09 VAC8 YEL013W -0.13 -0.13 -2.08 LSM1 YJL124C -0.06 -0.10 -2.07 NA YNR047W -0.29 -0.22 -2.06 PRR1 YKL116C -0.25 -0.23 -2.05 RIM101 YHL027W -0.14 -0.10 -2.04 NIF3 YGL221C -0.16 -0.18 -2.03 DSS4 YPR017C -0.23 -0.26 -2.02 NA YMR085W -0.16 -0.22 -2.02 PEX2 YJL210W -0.13 -0.10 -2.00 TPS3 YMR261C 0.18 0.21 2.01 DLS1 YJL065C 0.08 0.10 2.02 PHO80 YOL001W 0.22 0.21 2.04 SEC66 YBR171W 0.10 0.08 2.05 FMP36 YDR493W 0.23 0.26 2.06 RPE1 YJL121C 0.10 0.10 2.07 NA YDR157W 0.15 0.16 2.12 SUR2 YDR297W 0.17 0.16 2.13 MDH2 YOL126C 0.26 0.22 2.14 ECM15 YBL001C 0.09 0.09 2.14 MOD5 YOR274W 0.24 0.28 2.15 HHT1 YBR010W 0.10 0.09 2.15 SWI6 YLR182W 0.30 0.31 2.17 SGN1 YIR001C 0.10 0.10 2.21 NA YDL173W 0.07 0.09 2.24 IDH1 YNL037C 0.24 0.24 2.27 UBP16 YPL072W 0.24 0.29 2.31 MPH1 YIR002C 0.10 0.11 2.32 IPT1 YDR072C 0.17 0.17 2.33 RPA14 YDR156W 0.21 0.17 2.34 MRPL39 YML009c 0.36 0.26 2.40 SPO13 YHR014W 0.10 0.11 2.40 YAP1 YML007W 0.29 0.27 2.42 VAM7 YGL212W 0.20 0.22 2.48 NA YPL236C 0.34 0.32 2.48 GZF3 YJL110C 0.11 0.12 2.48 CDA2 YLR308W 0.25 0.31 2.48 SPO74 YGL170C 0.21 0.22 2.49 DAL80 YKR034W 0.27 0.28 2.49 FIT2 YOR382W 0.27 0.32 2.52

96 NA YEL033W 0.17 0.15 2.52 FTR1 YER145C 0.15 0.15 2.55 NA YOL053C-A 0.32 0.26 2.56 MRS4 YKR052C 0.32 0.29 2.60 NA YDR290W 0.34 0.34 2.68 MRPS17 YMR188C 0.35 0.30 2.70 SPH1 YLR313C 0.32 0.40 2.75 SKN1 YGR143W 0.28 0.25 2.81 PDR10 YOR328W 0.32 0.36 2.81 RPL36A YMR194W 0.35 0.32 2.90 ASK10 YGR097W 0.28 0.26 2.92 PCL1 YNL289W 0.32 0.31 2.95 NA YLR194C 0.47 0.43 2.96 NA YGL262W 0.25 0.28 3.12 UNG1 YML021C 0.39 0.34 3.13 NA YML090W 0.41 0.35 3.17 CPA1 YOR303W 0.41 0.41 3.20 CSN12 YJR084W 0.16 0.16 3.28 STV1 YMR054W 0.38 0.36 3.32 NA YMR172C-A 0.37 0.37 3.33 KSP1 YHR082C 0.15 0.15 3.38 RPN10 YHR200W 0.14 0.15 3.39 GIM3 YNL153C 0.40 0.37 3.51 MET18 YIL128W 0.09 0.17 3.59 NA YPR064W 0.46 0.46 3.60 NA YAL046C 0.53 0.51 4.05 ARO3 YDR035W 0.55 0.53 4.21 NA YNL195C 0.49 0.46 4.29 VPS30 YPL120W 0.55 0.56 4.35 RCE1 YMR274C 0.47 0.47 4.42 CCW14 YLR390W-A 0.63 0.66 4.53 NA YNL303W 0.56 0.53 5.01 NA YOL111C 0.51 0.52 5.05 SPT3 YDR392W 0.46 0.49 6.49

97 Table 5: NGD-CGA reporter deletion screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG SLH1 YGR271W -0.63 -0.61 -8.88 HEL2 YDR266C -1.18 -1.16 -7.98 RUP1 YOR138C -0.93 -0.92 -7.42 CUE3 YGL110C -1.01 -1.00 -7.31 UBI4 YLL039C -1.02 -1.00 -6.77 LSM1 YJL124C -0.54 -0.66 -6.70 RQT4 YKR023W -0.90 -0.91 -6.16 WHI3 YNL197C -0.80 -0.63 -5.85 RVS167 YDR388W -0.85 -0.83 -5.72 PAT1 YCR077C -0.82 -0.88 -5.70 UBP6 YFR010W -0.62 -0.64 -5.61 UBC4 YBR082C -0.38 -0.37 -5.33 SRN2 YLR119W -0.54 -0.54 -5.28 SIF2 YBR103W -0.59 -0.53 -5.18 PET122 YER153C -0.53 -0.57 -5.05 DST1 YGL043W -0.68 -0.67 -4.85 UBP3 YER151C -0.52 -0.54 -4.77 MUM2 YBR057C -0.44 -0.48 -4.73 UBC4 YBR082C -0.53 -0.47 -4.61 COX10 YPL172C -0.55 -0.60 -4.58 SWS2 YNL081C -0.47 -0.49 -4.53 COX23 YHR116W -0.47 -0.53 -4.49 YTA7 YGR270W -0.51 -0.52 -4.40 THR1 YHR025W -0.47 -0.51 -4.37 GAL10 YBR019C -0.41 -0.44 -4.28 DOA1 YKL213C -0.64 -0.63 -4.26 MFT1 YML062C -0.65 -0.60 -4.15 HTA1 YDR225W -0.59 -0.60 -4.14 CBC2 YPL178W -0.55 -0.53 -4.05 RIM1 YCR028C-A -0.54 -0.62 -4.00 VRP1 YLR337C -0.35 -0.40 -3.91 MNE1 YOR350C -0.43 -0.50 -3.85 SPO16 YHR153C -0.40 -0.45 -3.85 RAM1 YDL090C -0.14 -0.60 -3.85 KTR5 YNL029C -0.45 -0.41 -3.81 EAF7 YNL136W -0.40 -0.41 -3.76 VPS8 YAL002W -0.37 -0.38 -3.74 SHE4 YOR035C -0.46 -0.45 -3.66 LGE1 YPL055C -0.55 -0.47 -3.61

98 BCS1 YDR375C -0.48 -0.52 -3.56 NUP188 YML103C -0.52 -0.51 -3.54 NA YLR374C -0.38 -0.36 -3.52 FMP50 YKR027W -0.57 -0.52 -3.49 RTF1 YGL244W -0.49 -0.48 -3.49 LSM6 YDR378C -0.39 -0.51 -3.48 HOM3 YER052C -0.40 -0.40 -3.48 RMD11 YHL023C -0.37 -0.41 -3.47 DID2 YKR035W-A -0.52 -0.51 -3.44 PUF3 YLL013C -0.55 -0.51 -3.41 COX12 YLR038C -0.28 -0.35 -3.41 NA YGL042C -0.25 -0.24 -3.40 HMI1 YOL095C -0.42 -0.42 -3.39 NA YKR035C -0.50 -0.50 -3.35 NA YLR047C -0.32 -0.34 -3.31 TCO89 YPL180W -0.43 -0.43 -3.29 NPR2 YEL062W -0.32 -0.37 -3.26 CBT1 YKL208W -0.46 -0.48 -3.24 IMG2 YCR071C -0.46 -0.50 -3.21 MBA1 YBR185C -0.25 -0.32 -3.15 VID22 YLR373C -0.40 -0.32 -3.15 HOS4 YIL112W -0.39 -0.31 -3.13 EMI1 YDR512C -0.29 -0.35 -3.10 GRR1 YJR090C -0.34 -0.31 -3.09 NA YJL169W -0.34 -0.30 -3.02 NA YOR139C -0.41 -0.37 -3.02 NA YJR120W -0.41 -0.45 -3.01 COX17 YLL009C -0.39 -0.44 -2.97 CPR2 YHR057C -0.18 -0.35 -2.96 NA YJR079W -0.20 -0.28 -2.86 RIM101 YHL027W -0.29 -0.33 -2.83 ATP10 YLR393W -0.25 -0.29 -2.82 NA YER077C -0.22 -0.32 -2.82 THP2 YHR167W -0.22 -0.33 -2.79 NIP100 YPL174C -0.38 -0.37 -2.79 SPT2 YER161C -0.29 -0.31 -2.76 FCY2 YER056C -0.33 -0.31 -2.76 CYT1 YOR065W -0.34 -0.33 -2.69 MRM2 YGL136C -0.33 -0.37 -2.68 RCO1 YMR075W -0.38 -0.38 -2.67 HAP3 YBL021C -0.23 -0.28 -2.67 NA YLR252W -0.29 -0.27 -2.67 BRE5 YNR051C -0.30 -0.33 -2.63

99 RPS21B YJL136C -0.27 -0.26 -2.62 NA YNL165W -0.31 -0.28 -2.62 NA YOR111W -0.23 -0.32 -2.61 CSF1 YLR087C -0.28 -0.26 -2.57 VPS9 YML097C -0.34 -0.37 -2.56 ARC18 YLR370C -0.22 -0.26 -2.56 NA YNR020C -0.26 -0.32 -2.55 MMR1 YLR190W -0.24 -0.26 -2.54 GON5 YPL183W-A -0.29 -0.33 -2.52 IOC3 YFR013W -0.29 -0.34 -2.50 HSE1 YHL002W -0.26 -0.29 -2.50 NA YOR105W -0.31 -0.31 -2.47 CAF40 YNL288W -0.25 -0.27 -2.46 EST1 YLR233C -0.30 -0.25 -2.45 SOH1 YGL127C -0.38 -0.34 -2.43 LSM7 YNL147W -0.16 -0.17 -2.42 NA YER139C -0.30 -0.27 -2.42 BEM4 YPL161C -0.37 -0.31 -2.39 STE14 YDR410C -0.19 -0.35 -2.38 HIS5 YIL116W -0.23 -0.23 -2.37 SLM3 YDL033C -0.29 -0.37 -2.35 ACM1 YPL267W -0.16 -0.16 -2.31 PIN2 YOR104W -0.32 -0.29 -2.30 EFT1 YOR133W -0.28 -0.28 -2.27 FYV4 YHR059W -0.32 -0.27 -2.27 RGS2 YOR107W -0.25 -0.28 -2.27 NA YDR431W -0.17 -0.26 -2.26 TOF1 YNL273W -0.30 -0.25 -2.26 LCB4 YOR171C -0.28 -0.28 -2.25 NA YBL065W -0.21 -0.23 -2.24 IOC2 YLR095C -0.23 -0.23 -2.24 RPL35A YDL191W -0.36 -0.33 -2.23 NA YMR007W -0.36 -0.32 -2.23 MDM10 YAL010C -0.12 -0.23 -2.21 MRPL1 YDR116C -0.21 -0.32 -2.19 NA YIL067C -0.20 -0.22 -2.19 CSG2 YBR036C -0.22 -0.23 -2.18 MET3 YJR010W -0.14 -0.22 -2.18 SED1 YDR077W -0.32 -0.32 -2.17 NA YDR360W -0.30 -0.31 -2.14 RPL24A YGL031C -0.30 -0.30 -2.14 NA YAL037W -0.21 -0.22 -2.13 NA YLR455W -0.28 -0.30 -2.12

100 MET10 YFR030W -0.36 -0.29 -2.10 VID21 YDR359C -0.25 -0.31 -2.09 PMP3 YDR276C -0.31 -0.30 -2.06 SDC1 YDR469W -0.23 -0.23 -2.05 RPL9B YNL067W -0.22 -0.22 -2.05 SET2 YJL168C -0.16 -0.20 -2.03 VPS72 YDR485C -0.19 -0.23 -2.03 HTZ1 YOL012C -0.24 -0.25 -2.02 TAN1 YGL232W -0.29 -0.28 -2.02 RPS9B YBR189W -0.19 -0.21 -2.02 SET3 YKR029C -0.29 -0.30 -2.02 TAT1 YBR069C -0.26 -0.21 -2.01 NA YLR021W 0.22 0.20 2.00 NA YPL225W 0.27 0.26 2.02 RXT3 YDL076C 0.32 0.31 2.03 RER1 YCL001W 0.33 0.31 2.03 NA YCL006C 0.36 0.31 2.04 HUL5 YGL141W 0.26 0.28 2.04 YAP5 YIR018W 0.25 0.20 2.06 OAR1 YKL055C 0.32 0.30 2.07 SIR3 YLR442C 0.29 0.30 2.11 MMM1 YLL006W 0.26 0.31 2.11 PDA1 YER178W 0.15 0.24 2.12 NA YER119C-A 0.20 0.24 2.13 PIB2 YGL023C 0.25 0.29 2.14 NA YPL102C 0.27 0.28 2.15 SKI3 YPR189W 0.17 0.15 2.21 SKY1 YMR216C 0.33 0.31 2.21 NA YLR199C 0.17 0.22 2.21 ERV14 YGL054C 0.26 0.30 2.21 RCY1 YJL204C 0.17 0.22 2.22 SEM1 YDR363W-A 0.32 0.32 2.22 ETR1 YBR026C 0.26 0.22 2.23 LPD1 YFL018C 0.28 0.26 2.27 NA YOR291W 0.33 0.30 2.29 RPS0B YLR048W 0.22 0.23 2.30 GSF2 YML048W 0.26 0.33 2.32 PPM1 YDR435C 0.25 0.27 2.35 SNF1 YDR477W 0.18 0.16 2.38 CEM1 YER061C 0.21 0.27 2.39 RPL43A YPR043W 0.18 0.17 2.42 SPE2 YOL052C 0.28 0.29 2.42 NA YPR045C 0.18 0.17 2.45

101 KCC4 YCL024W 0.39 0.37 2.45 NA YDL183C 0.37 0.38 2.47 LYS20 YDL182W 0.39 0.39 2.54 MET28 YIR017C 0.28 0.25 2.59 TRF4 YOL115W 0.24 0.32 2.61 BAS1 YKR099W 0.34 0.39 2.64 RPN10 YHR200W 0.31 0.31 2.68 TEF2 YBR118W 0.20 0.19 2.73 PRE9 YGR135W 0.33 0.37 2.74 NA YBL104C 0.22 0.27 2.75 UBR1 YGR184C 0.31 0.32 2.75 SMY2 YBR172C 0.32 0.28 2.84 NA YNL226W 0.30 0.30 2.85 KTI12 YKL110C 0.40 0.43 2.90 NA YLR412W 0.34 0.29 2.94 RNH201 YNL072W 0.31 0.31 2.95 ELP4 YPL101W 0.42 0.38 2.95 GSH2 YOL049W 0.38 0.37 3.03 GAL3 YDR009W 0.37 0.44 3.08 ECM8 YBR076W 0.26 0.32 3.24 ARG3 YJL088W 0.20 0.32 3.25 NA YEL033W 0.40 0.40 3.52 NCS6 YGL211W 0.47 0.48 3.56 ELP3 YPL086C 0.48 0.49 3.77 ELP6 YMR312W 0.38 0.41 3.84 UBA4 YHR111W 0.46 0.47 4.06 NA YJL046W 0.39 0.41 4.13 HTD2 YHR067W 0.45 0.48 4.16 ASC1 YMR116C 0.58 0.60 4.18 NCS2 YNL119W 0.47 0.46 4.28 LIP5 YOR196C 0.56 0.53 4.33 ELP2 YGR200C 0.52 0.53 4.57 NA YNL120C 0.46 0.49 4.59 SGF73 YGL066W 0.56 0.63 4.60 URM1 YIL008W 0.55 0.56 4.87 IKI3 YLR384C 0.52 0.49 4.89 SPT3 YDR392W 0.76 0.82 5.68 SYH1 YPL105C 0.80 0.78 5.96

102 Table 6: NGD-AAA reporter deletion screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG CDC73 YLR418C -1.09 -1.06 -8.29 PAT1 YCR077C -0.69 -0.69 -6.19 SLH1 YGR271W -0.87 -0.80 -6.10 SRN2 YLR119W -0.39 -0.38 -5.91 SWS2 YNL081C -0.33 -0.32 -5.55 COX12 YLR038C -0.33 -0.34 -5.37 LSM1 YJL124C -0.41 -0.49 -5.35 COX17 YLL009C -0.35 -0.37 -5.34 SIF2 YBR103W -0.53 -0.50 -5.33 MFT1 YML062C -0.74 -0.68 -5.32 RIM1 YCR028C-A -0.62 -0.58 -5.21 RVS167 YDR388W -0.76 -0.70 -5.17 COX10 YPL172C -0.62 -0.61 -4.97 YLR338W YLR338W -0.33 -0.31 -4.85 RMD11 YHL023C -0.51 -0.49 -4.82 COX23 YHR116W -0.51 -0.49 -4.80 UBP3 YER151C -0.47 -0.53 -4.72 UBC4 YBR082C -0.66 -0.61 -4.64 ECM17 YJR137C -0.34 -0.32 -4.64 UBC4 YBR082C -0.47 -0.43 -4.61 PET122 YER153C -0.50 -0.51 -4.56 HOM3 YER052C -0.59 -0.51 -4.54 YGL042C YGL042C -0.60 -0.58 -4.45 MUM2 YBR057C -0.40 -0.41 -4.39 UBI4 YLL039C -0.30 -0.29 -4.31 DOA1 YKL213C -0.29 -0.28 -4.15 RUP1 YOR138C -0.29 -0.27 -4.13 BUD27 YFL023W -0.51 -0.46 -4.13 MLP2 YIL149C -0.33 -0.37 -4.10 LEO1 YOR123C -0.27 -0.27 -4.06 HTA1 YDR225W -0.55 -0.54 -4.03 MNE1 YOR350C -0.48 -0.49 -3.99 QCR2 YPR191W -0.46 -0.51 -3.90 NA YJR120W -0.28 -0.27 -3.90 NA YNR020C -0.32 -0.25 -3.86 YTA7 YGR270W -0.44 -0.39 -3.85 BCS1 YDR375C -0.51 -0.52 -3.83 NPR2 YEL062W -0.42 -0.43 -3.79 IMG2 YCR071C -0.38 -0.42 -3.78

103 VPS21 YOR089C -0.26 -0.25 -3.78 MET1 YKR069W -0.31 -0.26 -3.76 CBT1 YKL208W -0.24 -0.26 -3.76 LGE1 YPL055C -0.44 -0.46 -3.73 HEL2 YDR266C -0.55 -0.50 -3.69 TIR3 YIL011W -0.11 -0.37 -3.66 LYS1 YIR034C -0.21 -0.33 -3.58 PUF3 YLL013C -0.25 -0.24 -3.57 SNT1 YCR033W -0.44 -0.40 -3.57 NA YML013C-A -0.45 -0.45 -3.55 ARG3 YJL088W -0.16 -0.32 -3.51 NA YLR218C -0.21 -0.22 -3.42 STE4 YOR212W -0.38 -0.42 -3.38 SPO19 YPL130W -0.40 -0.42 -3.36 CAT5 YOR125C -0.35 -0.44 -3.35 HMI1 YOL095C -0.18 -0.22 -3.33 VPS8 YAL002W -0.33 -0.31 -3.29 GIC1 YHR061C -0.36 -0.33 -3.26 LSM6 YDR378C -0.38 -0.44 -3.26 ATP10 YLR393W -0.19 -0.21 -3.23 NA YNL165W -0.20 -0.18 -3.18 MRP49 YKL167C -0.24 -0.22 -3.16 BRE5 YNR051C -0.20 -0.20 -3.06 RQT4 YKR023W -0.21 -0.21 -3.06 THP2 YHR167W -0.26 -0.31 -3.06 UBX7 YBR273C -0.33 -0.34 -3.01 RTT109 YLL002W -0.21 -0.21 -3.00 CPR2 YHR057C -0.23 -0.30 -2.96 SLM3 YDL033C -0.28 -0.33 -2.95 DHH1 YDL160C -0.34 -0.33 -2.93 SWR1 YDR334W -0.39 -0.39 -2.91 YAK1 YJL141C -0.29 -0.27 -2.90 KTR1 YOR099W -0.20 -0.19 -2.89 LYS12 YIL094C -0.23 -0.37 -2.86 SIN3 YOL004W -0.20 -0.19 -2.86 NA YDR360W -0.42 -0.38 -2.84 NA YPL247C -0.29 -0.35 -2.79 MLH1 YMR167W -0.35 -0.36 -2.78 DBP7 YKR024C -0.18 -0.19 -2.77 PMS1 YNL082W -0.14 -0.16 -2.76 RAS1 YOR101W -0.19 -0.18 -2.76 HOR7 YMR251W-A -0.15 -0.16 -2.74 FMC1 YIL098C -0.29 -0.25 -2.74

104 CAN1 YEL063C -0.36 -0.30 -2.70 CYT2 YKL087C -0.30 -0.35 -2.70 COX5A YNL052W -0.15 -0.15 -2.70 MBA1 YBR185C -0.25 -0.25 -2.69 SPT2 YER161C -0.27 -0.30 -2.68 ROD1 YOR018W -0.23 -0.18 -2.67 UBP6 YFR010W -0.33 -0.30 -2.65 HSP82 YPL240C -0.33 -0.33 -2.64 LSM7 YNL147W -0.26 -0.34 -2.64 RPS21B YJL136C -0.21 -0.24 -2.62 NA YMR031W-A -0.28 -0.33 -2.61 TCO89 YPL180W -0.34 -0.32 -2.60 CSF1 YLR087C -0.18 -0.17 -2.58 SHE4 YOR035C -0.19 -0.17 -2.57 THR1 YHR025W -0.24 -0.26 -2.56 ARG5,6 YER069W -0.25 -0.29 -2.55 LEA1 YPL213W -0.35 -0.32 -2.55 QRI7 YDL104C -0.21 -0.28 -2.52 HOS4 YIL112W -0.31 -0.23 -2.48 VPS24 YKL041W -0.18 -0.17 -2.48 VID22 YLR373C -0.18 -0.16 -2.48 VPS9 YML097C -0.34 -0.32 -2.47 INO4 YOL108C -0.15 -0.16 -2.47 BRE1 YDL074C -0.35 -0.32 -2.47 CYT1 YOR065W -0.15 -0.16 -2.43 EMI1 YDR512C -0.32 -0.27 -2.41 NA YDL119C -0.34 -0.27 -2.41 NIP100 YPL174C -0.31 -0.30 -2.40 COQ2 YNR041C -0.28 -0.31 -2.38 NA YOR238W -0.25 -0.29 -2.36 CAF120 YNL278W -0.13 -0.13 -2.33 NA YMR007W -0.30 -0.30 -2.33 NA YLR374C -0.15 -0.15 -2.30 NA YDR090C -0.35 -0.31 -2.29 NA YLR173W -0.15 -0.15 -2.29 RIM101 YHL027W -0.28 -0.23 -2.29 RGS2 YOR107W -0.17 -0.15 -2.28 HIS5 YIL116W -0.15 -0.21 -2.26 NA YIL067C -0.17 -0.20 -2.23 RRN10 YBL025W -0.14 -0.21 -2.23 EST2 YLR318W -0.17 -0.14 -2.22 TEF4 YKL081W -0.14 -0.15 -2.20 NA YBL065W -0.20 -0.20 -2.19

105 TAT1 YBR069C -0.23 -0.20 -2.19 MMS22 YLR320W -0.16 -0.14 -2.15 MET2 YNL277W -0.11 -0.12 -2.15 BTS1 YPL069C -0.28 -0.27 -2.13 LYS4 YDR234W -0.22 -0.28 -2.10 RVS161 YCR009C -0.25 -0.23 -2.09 FCY2 YER056C -0.25 -0.24 -2.08 EAF7 YNL136W -0.13 -0.12 -2.05 NA YPR147C -0.31 -0.27 -2.05 NA YOR093C -0.15 -0.14 -2.05 IDH2 YOR136W -0.13 -0.14 -2.04 VAM3 YOR106W -0.16 -0.14 -2.04 NA YER078C -0.17 -0.23 -2.03 CSG2 YBR036C -0.16 -0.19 -2.01 RPA14 YDR156W 0.27 0.27 2.02 EAF6 YJR082C 0.17 0.18 2.02 COX14 YML129C 0.28 0.26 2.02 HIS7 YBR248C 0.30 0.23 2.05 HFA1 YMR207C 0.30 0.26 2.05 NA YDR048C 0.29 0.28 2.09 NPR1 YNL183C 0.12 0.12 2.09 IXR1 YKL032C 0.17 0.14 2.10 BCH1 YMR237W 0.19 0.27 2.11 RRD2 YPL152W 0.22 0.25 2.12 ATP11 YNL315C 0.12 0.12 2.14 UBR1 YGR184C 0.18 0.21 2.14 NA YJL144W 0.14 0.19 2.15 HCR1 YLR192C 0.17 0.14 2.16 DLS1 YJL065C 0.15 0.20 2.19 RCY1 YJL204C 0.25 0.20 2.22 ELP4 YPL101W 0.22 0.27 2.27 MCK1 YNL307C 0.14 0.13 2.28 NA YJL064W 0.25 0.21 2.31 MET8 YBR213W 0.22 0.22 2.31 SRB8 YCR081W 0.25 0.26 2.34 IKI3 YLR384C 0.14 0.15 2.34 SIR2 YDL042C 0.25 0.26 2.36 LIP5 YOR196C 0.13 0.15 2.38 NA YBL104C 0.21 0.23 2.39 PUF6 YDR496C 0.29 0.26 2.39 RPL37B YDR500C 0.36 0.27 2.40 TRF4 YOL115W 0.14 0.15 2.41 SKY1 YMR216C 0.32 0.31 2.43

106 URE2 YNL229C 0.13 0.14 2.44 EMI5 YOL071W 0.16 0.15 2.44 NA YPL144W 0.29 0.29 2.45 KEX2 YNL238W 0.15 0.14 2.48 STE20 YHL007C 0.25 0.25 2.49 AKL1 YBR059C 0.22 0.24 2.50 NA YDR157W 0.34 0.34 2.52 PGM2 YMR105C 0.30 0.32 2.55 MET28 YIR017C 0.27 0.23 2.57 CSR2 YPR030W 0.24 0.34 2.61 RPP1A YDL081C 0.32 0.29 2.61 NA YLR199C 0.18 0.16 2.62 ETR1 YBR026C 0.25 0.25 2.66 GAL4 YPL248C 0.26 0.32 2.68 LPD1 YFL018C 0.35 0.30 2.68 VPS75 YNL246W 0.17 0.15 2.70 GTR1 YML121W 0.35 0.34 2.72 NA YNL226W 0.17 0.16 2.77 RPL9B YNL067W 0.16 0.16 2.78 SMY2 YBR172C 0.28 0.26 2.79 RPS1B YML063W 0.32 0.36 2.84 STE2 YFL026W 0.35 0.32 2.89 RPN10 YHR200W 0.30 0.30 2.96 ECM8 YBR076W 0.25 0.28 2.98 NA YLL029W 0.22 0.20 2.98 SUR2 YDR297W 0.32 0.40 2.99 NA YDR514C 0.40 0.35 3.12 NA YEL033W 0.41 0.35 3.16 SDH4 YDR178W 0.46 0.42 3.17 SKI3 YPR189W 0.40 0.42 3.20 YMD8 YML038C 0.40 0.41 3.28 CEM1 YER061C 0.39 0.37 3.29 PDA1 YER178W 0.37 0.37 3.30 SKI7 YOR076C 0.25 0.22 3.55 NA YNL120C 0.20 0.20 3.57 NA YLR021W 0.24 0.24 3.80 NA YOR291W 0.47 0.46 3.82 SKI2 YLR398C 0.25 0.25 3.88 HTD2 YHR067W 0.44 0.40 3.95 NCS2 YNL119W 0.23 0.23 3.97 NA YBR204C 0.42 0.38 4.07 ELP3 YPL086C 0.50 0.49 4.07 ASC1 YMR116C 0.53 0.55 4.31

107 SYH1 YPL105C 0.54 0.52 4.32 UBA4 YHR111W 0.48 0.45 4.46 SPT3 YDR392W 0.57 0.61 4.53 ELP2 YGR200C 0.47 0.47 4.68 URM1 YIL008W 0.51 0.48 4.80 AIM22 YJL046W 0.51 0.52 5.75

108 Table 7: OPT reporter temperature sensitive screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG DFR1 YOR236W -3.76 -3.51 -12.36 PFY1 YOR122C -2.42 -2.40 -10.60 RNA1 YMR235C -0.98 -0.95 -5.41 SEC53 YFL045C -0.89 -0.87 -4.64 BRE5 YNR051C -0.70 -0.76 -4.30 DED1 YOR204W -0.79 -0.80 -4.28 SSL1 YLR005W -0.77 -0.74 -3.95 blank blank -0.62 -0.71 -3.75 HOM2 YDR158W -0.63 -0.64 -3.65 PAM16 YJL104W -0.83 -0.83 -3.61 SPT6 YGR116W -0.62 -0.62 -3.54 YHC1 YLR298C -0.66 -0.61 -3.49 SSL1 YLR005W -0.69 -0.60 -3.41 GUK1 YDR454C -0.67 -0.63 -3.35 GSP1 YLR293C -0.72 -0.75 -3.27 CDC36 YDL165W -0.55 -0.57 -3.25 TFB1 YDR311W -0.59 -0.58 -3.08 CLP1 YOR250C -0.51 -0.58 -3.06 PCP1 YGR101W -0.56 -0.56 -2.95 YKL088W YKL088W -0.61 -0.55 -2.92 ALA1 YOR335C -0.53 -0.55 -2.92 SSC1 YJR045C -0.81 -0.83 -2.87 TIM50 YPL063W -0.48 -0.49 -2.76 SGV1 YPR161C -0.67 -0.52 -2.73 ABF1 YKL112W -0.50 -0.62 -2.71 POB3 YML069W -0.48 -0.49 -2.60 PKC1 YBL105C -0.74 -0.75 -2.59 PRP43 YGL120C -0.70 -0.59 -2.58 ABD1 YBR236C -0.49 -0.49 -2.57 SGV1 YPR161C -0.37 -0.42 -2.36 MGE1 YOR232W -0.65 -0.68 -2.34 SGV1 YPR161C -0.42 -0.44 -2.32 CNS1 YBR155W -0.53 -0.53 -2.30 RPB2 YOR151C -0.54 -0.53 -2.27 TFB3 YDR460W -0.52 -0.39 -2.21 SPT14 YPL175W -0.37 -0.39 -2.17 CDC39 YCR093W -0.39 -0.41 -2.17 RAD5 YLR032W -0.61 -0.41 -2.17 DST1 YGL043W -0.59 -0.50 -2.15

109 NUP159 YIL115C -0.41 -0.41 -2.14 SEC23 YPR181C -0.42 -0.38 -2.14 CDC48 YDL126C -0.60 -0.49 -2.13 RAD3 YER171W -0.49 -0.49 -2.12 RNA1 YMR235C -0.67 -0.61 -2.11 MPS2 YGL075C -0.49 -0.48 -2.05 PRP18 YGR006W -0.43 -0.39 -2.05 GLE1 YDL207W -0.61 -0.59 -2.05 DED1 YOR204W -0.50 -0.47 -2.04 TFA2 YKR062W -0.63 -0.59 -2.04 SPN1 YPR133C -0.34 -0.36 -2.00 SPC97 YHR172W 0.33 0.36 2.01 SGF73 YGL066W 0.37 0.38 2.25 HRR25 YPL204W 0.41 0.44 2.58 SPT3 YDR392W 1.27 1.25 4.46

110 Table 8: NGD-CGA reporter temperature sensitive screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG FIP1 YJR093C -1.10 -0.93 -4.73 GLE1 YDL207W -1.23 -1.11 -4.65 SSL1 YLR005W -0.89 -0.85 -4.16 RPB5 YBR154C -1.00 -0.98 -4.11 PKC1 YBL105C -0.93 -0.93 -3.90 SPT6 YGR116W -0.85 -0.87 -3.87 TFA2 YKR062W -0.97 -0.92 -3.86 UBC4 YBR082C -0.82 -0.80 -3.57 TFB1 YDR311W -0.74 -0.70 -3.39 POB3 YML069W -0.68 -0.66 -3.22 PAT1 YCR077C -0.55 -0.65 -3.16 MAS6 YNR017W -0.73 -0.75 -3.12 SSC1 YJR045C -0.68 -0.75 -3.11 GUK1 YDR454C -0.74 -0.63 -3.09 SSL1 YLR005W -0.78 -0.69 -3.06 MSS4 YDR208W -0.50 -0.63 -3.05 LSM6 YDR378C -0.61 -0.62 -3.01 TIM50 YPL063W -0.66 -0.67 -2.99 DOA1 YKL213C -0.58 -0.58 -2.95 NSE5 YML023C -0.61 -0.60 -2.90 SUB2 YDL084W -0.65 -0.70 -2.89 MOB1 YIL106W -0.65 -0.69 -2.88 URA10 YMR271C -0.51 -0.64 -2.87 UTP15 YMR093W -0.63 -0.66 -2.75 SGV1 YPR161C -0.61 -0.61 -2.73 YHC1 YLR298C -0.69 -0.64 -2.65 CLF1 YLR117C -0.48 -0.52 -2.63 TAF7 YMR227C -0.52 -0.54 -2.63 ACT1 YFL039C -0.55 -0.52 -2.62 TFA1 YKL028W -0.57 -0.58 -2.61 PAM18 YLR008C -0.61 -0.63 -2.61 MGE1 YOR232W -0.58 -0.62 -2.58 DBP5 YOR046C -0.51 -0.53 -2.56 CDC36 YDL165W -0.56 -0.57 -2.53 YOR262W YOR262W -0.58 -0.52 -2.51 GSP1 YLR293C -0.47 -0.49 -2.49 UBA1 YKL210W -0.61 -0.60 -2.48 SPN1 YPR133C -0.50 -0.55 -2.44 SEC53 YFL045C -0.78 -0.50 -2.44

111 RNA1 YMR235C -0.61 -0.58 -2.42 ABF1 YKL112W -0.57 -0.57 -2.35 SYF1 YDR416W -0.52 -0.48 -2.34 SPC110 YDR356W -0.46 -0.44 -2.25 SPT16 YGL207W -0.45 -0.44 -2.25 PRP31 YGR091W -0.48 -0.44 -2.21 SGV1 YPR161C -0.42 -0.45 -2.20 BRE5 YNR051C -0.41 -0.48 -2.12 ABF1 YKL112W -0.35 -0.42 -2.10 NTR2 YKR022C -0.41 -0.41 -2.09 RPO21 YDL140C -0.46 -0.50 -2.08 SNU114 YKL173W -0.40 -0.40 -2.04 URM1 YIL008W 0.52 0.46 2.00 NIP7 YPL211W 0.42 0.44 2.00 RPP0 YLR340W 0.39 0.39 2.04 HYP2 YEL034W 0.50 0.48 2.05 RRS1 YOR294W 0.46 0.45 2.06 YHR020W YHR020W 0.40 0.41 2.07 RIA1 YNL163C 0.43 0.40 2.08 RPN6 YDL097C 0.43 0.41 2.09 UBA4 YHR111W 0.48 0.46 2.09 RPC34 YNR003C 0.42 0.42 2.10 LIP5 YOR196C 0.42 0.41 2.12 POL5 YEL055C 0.43 0.41 2.15 CDC123 YLR215C 0.47 0.43 2.15 GCD10 YNL062C 0.46 0.48 2.16 MAK5 YBR142W 0.50 0.48 2.20 MAK11 YKL021C 0.45 0.44 2.21 ELP4 YPL101W 0.45 0.45 2.28 AFG2 YLR397C 0.43 0.44 2.30 NMD3 YHR170W 0.52 0.51 2.32 RPL15A YLR029C 0.52 0.54 2.32 SGF73 YGL066W 0.51 0.51 2.32 HYP2 YEL034W 0.57 0.55 2.36 RPT4 YOR259C 0.48 0.47 2.38 DED81 YHR019C 0.58 0.57 2.45 POL5 YEL055C 0.52 0.49 2.48 NOP2 YNL061W 0.46 0.48 2.48 PUP2 YGR253C 0.51 0.50 2.51 NOC3 YLR002C 0.47 0.50 2.51 IKI3 YLR384C 0.51 0.50 2.52 NOP2 YNL061W 0.58 0.59 2.53 YLR022C YLR022C 0.49 0.49 2.53

112 MAK16 YAL025C 0.56 0.56 2.54 SUI2 YJR007W 0.55 0.58 2.64 PWP1 YLR196W 0.66 0.63 2.71 NOP2 YNL061W 0.55 0.54 2.81 HRR25 YPL204W 0.67 0.63 2.85 TIF6 YPR016C 0.60 0.59 2.96 MAK11 YKL021C 0.65 0.61 3.16 HTS1 YPR033C 0.73 0.73 3.77 SPT3 YDR392W 1.05 1.04 4.45

113 Table 9: NGD-AAA reporter temperature sensitive screen Med Mean Symbol SGD log(GFP/RFP) log(GFP/RFP) Z_LOG TFA2 YKR062W -1.17 -1.19 -5.12 SGV1 YPR161C -1.19 -1.21 -4.94 SSL1 YLR005W -0.92 -0.90 -4.42 FIP1 YJR093C -1.26 -1.03 -4.26 RAD3 YER171W -0.98 -0.94 -3.87 GLE1 YDL207W -0.85 -0.87 -3.75 SSL1 YLR005W -0.96 -0.88 -3.58 PKC1 YBL105C -0.80 -0.79 -3.39 YHC1 YLR298C -0.76 -0.78 -3.37 POB3 YML069W -0.69 -0.68 -3.34 SLU7 YDR088C -0.82 -0.79 -3.22 PAT1 YCR077C -0.62 -0.65 -3.20 MOB1 YIL106W -0.69 -0.72 -3.08 QCR2 YPR191W -0.75 -0.69 -2.97 RNA1 YMR235C -0.73 -0.66 -2.86 RSE1 YML049C -0.76 -0.70 -2.83 CDC28 YBR160W -0.51 -0.57 -2.81 UBC4 YBR082C -0.66 -0.68 -2.75 RPB5 YBR154C -0.64 -0.64 -2.74 RPO21 YDL140C -0.57 -0.64 -2.73 SGV1 YPR161C -0.51 -0.56 -2.73 MOT1 YPL082C -0.56 -0.54 -2.65 SSC1 YJR045C -0.62 -0.62 -2.64 CDC40 YDR364C -0.63 -0.64 -2.61 SPN1 YPR133C -0.59 -0.64 -2.59 NSE5 YML023C -0.51 -0.52 -2.52 PAM16 YJL104W -0.59 -0.61 -2.51 ABF1 YKL112W -0.59 -0.58 -2.51 SUB2 YDL084W -0.58 -0.58 -2.50 SPT6 YGR116W -0.59 -0.61 -2.49 TFA1 YKL028W -0.61 -0.61 -2.46 SUA7 YPR086W -0.55 -0.60 -2.45 MED6 YHR058C -0.53 -0.57 -2.44 BRE5 YNR051C -0.52 -0.58 -2.34 MGE1 YOR232W -0.57 -0.54 -2.31 CDC36 YDL165W -0.56 -0.57 -2.31 ACT1 YFL039C -0.64 -0.56 -2.30 LSM6 YDR378C -0.45 -0.47 -2.28 TAF7 YMR227C -0.40 -0.45 -2.20

114 UBA1 YKL210W -0.50 -0.51 -2.18 PAM18 YLR008C -0.49 -0.50 -2.16 MAS6 YNR017W -0.50 -0.50 -2.14 CEG1 YGL130W -0.49 -0.50 -2.14 CEG1 YGL130W -0.59 -0.52 -2.09 YHC1 YLR298C -0.52 -0.52 -2.08 ABF1 YKL112W -0.44 -0.51 -2.07 PRP19 YLL036C -0.52 -0.51 -2.06 PRP22 YER013W -0.50 -0.50 -2.03 YDR196C YDR196C -0.60 -0.50 -2.01 SUI2 YJR007W 0.47 0.47 2.00 ELP4 YPL101W 0.48 0.47 2.03 POL5 YEL055C 0.47 0.49 2.06 DBP6 YNR038W 0.53 0.49 2.07 RPP0 YLR340W 0.47 0.50 2.10 UBA4 YHR111W 0.51 0.50 2.11 GRC3 YLL035W 0.53 0.50 2.11 NOP2 YNL061W 0.46 0.44 2.15 ELP4 YPL101W 0.45 0.44 2.15 DCP2 YNL118C 0.51 0.52 2.16 BRX1 YOL077C 0.54 0.50 2.19 SKI8 YGL213C 0.45 0.44 2.19 CSL4 YNL232W 0.52 0.52 2.20 DBP9 YLR276C 0.50 0.53 2.20 NOP2 YNL061W 0.53 0.53 2.21 HYP2 YEL034W 0.53 0.51 2.22 NOP2 YNL061W 0.55 0.53 2.25 CDC123 YLR215C 0.51 0.46 2.26 IPI1 YHR085W 0.55 0.53 2.26 RRS1 YOR294W 0.57 0.53 2.27 CDC31 YOR257W 0.46 0.46 2.28 RPC11 YDR045C 0.52 0.52 2.28 NUG1 YER006W 0.53 0.55 2.29 SNU13 YEL026W 0.52 0.55 2.29 POL1 YNL102W 0.40 0.47 2.30 YTM1 YOR272W 0.48 0.47 2.32 SIK1 YLR197W 0.56 0.55 2.35 NOP2 YNL061W 0.56 0.57 2.38 YHR020W YHR020W 0.49 0.49 2.39 IKI3 YLR384C 0.49 0.49 2.43 AFG2 YLR397C 0.59 0.58 2.44 YAH1 YPL252C 0.47 0.50 2.45 MAK5 YBR142W 0.62 0.60 2.52

115 NOP2 YNL061W 0.61 0.60 2.62 SGF73 YGL066W 0.62 0.64 2.69 ARC19 YKL013C 0.55 0.55 2.71 SDA1 YGR245C 0.58 0.55 2.71 MAK11 YKL021C 0.65 0.65 2.72 RIA1 YNL163C 0.63 0.65 2.72 YLR022C YLR022C 0.60 0.67 2.79 PWP1 YLR196W 0.68 0.67 2.90 RRP1 YDR087C 0.77 0.69 2.90 NIP7 YPL211W 0.73 0.69 2.93 RPO31 YOR116C 0.71 0.70 2.94 HYP2 YEL034W 0.68 0.69 2.98 MAK11 YKL021C 0.62 0.61 3.02 DED81 YHR019C 0.72 0.73 3.17 TIF6 YPR016C 0.65 0.65 3.21 MAK16 YAL025C 0.83 0.77 3.25 POL5 YEL055C 0.64 0.66 3.25 HRR25 YPL204W 0.82 0.80 3.36 HTS1 YPR033C 0.80 0.82 3.44 SPT3 YDR392W 0.87 0.87 3.76 NOC3 YLR002C 0.77 0.78 3.84

116 Chapter IV

Materials and Methods

Plasmid construction

GFP reporters

The OPT reporter plasmid, or pKD065, was cloned in pECB1806 (Gamble et al.

2016) as follows. Briefly, the GAL1 promoter, a GFP-2A-FLAG, a partially non- optimal HIS3 and the ADH1 terminator were introduced in SpeI/SphI- digested pECB1806 using Gibson assembly leading to pKD064. The GAL1 promoter was amplified by PCR using primers KD235 and KD236 and the ADH1 terminator was PCR amplified using KD239 and KD240 with pECB1806 as template

(Methods-table supplement 1). All PCR products were cleaned using Zymo

Research DNA Clean and Concentrator Kit. The GFP-2A-FLAG and a partially non-optimal HIS3 sequence fragment was gene-synthesized by Integrated DNA

Technologies (IDT). The partially non-optimal FLAG-HIS3 in pKD064 was then replaced by a fully optimal sequence of FLAG-HIS3 from pJC797

(Radhakrishnan et al. 2016) to make pKD065 (Methods-table supplement 1).

The NGD-AAA and NGD-CGA reporter plasmids (pKD079 and pKD080, respectively) were cloned using pKD065. In brief, a (AAA)x12 or (CGA)x12 codon sequence was inserted 90 codons into the HIS3 gene of pKD065. The

PCR product from (AAA)12 primers (KD281 and KD280) or (CGA)12 primers

(KD283 and KD280) with pKD065 template was combined with the PCR product from primers KD287 and KD276 with pKD065 template and SalI/Sph1 digested

117 pKD065 and these products were Gibson assembled to produce pKD079 and pKD080 (Methods-table supplement 1).

HA-CUE2 overexpressing plasmid

The CUE2 overexpressing FLEX plasmid (pKD100) was rescued from the yeast

FLEX library (Y. Hu et al. 2007; Kainth et al. 2009). A gene block (gKD002) of truncated CUE2 (NoSMR) was inserted into BamHI/ SphI digested pKD100, to make pKD105. 5’ 3xHA-tagged CUE2, and NoSMR were PCR amplified using

KD414 and KD416 and Gibson assembled with BamHI/ SphI digested pKD100 to make pKD120 and pKD125, respectively (Methods-table supplement 1).

Site-directed mutagenesis

Using the standard protocol for the QuikChange Lightning Multi Site-Directed

Mutagenesis Kit from Agilent Technologies, the indicated mutations in the HA-

CUE2 construct (pKD120) were made to make pKD127, pKD129, pKD131, pKD133, and pKD145 (Methods-table supplement 1).

CRISPR plasmids

BplI digested plasmid pJH2972 (Anand, Memisoglu, and Haber n.d.) was used in a Gibson reaction with primers KD503, KD505, and KD509 to make plasmids pKD163, pKD165, and pKD169 respectively.

Cue2 E. coli expression plasmids

PCR products from KD457 and KD338 from plasmids pKD120 and pKD129 were

Gibson cloned into BamHI and XhoI digested pSMT3, to make pKD097 and pKD098, respectively.

118 Yeast Strains and growth conditions

Yeast strains used in this study are described in Methods-table supplement 1 and are all derivatives of BY4741 unless specified otherwise. Yeast strain construction was performed using standard lithium acetate transformations.

Reporters strains were constructed by integrating the various GFP-2A-FLAG-

HIS3 cassettes, from StuI digested pKD065, pKD079, and pKD080, into the

ADE2 locus of BY4741 (Methods-table supplement 1). For SGA experiments, query strains overexpression screens were constructed by introducing the GFP-

2A-FLAG-HIS3 cassettes from StuI digested pKD065, pKD079, or pKD080 at the

ADE2 locus in BY4741 (Methods-table supplement 1).

Deletion strains were constructed by inserting resistance cassettes from Longtine et al. 1998 (Longtine et al. 1998) into the designated loci and genotypes are listed in Methods-table supplement 1. Note: two different deletion strains were used for XRN1 deletions in this study. See Methods-table supplement 1 for genotypes.

HA tag insertions, point mutations, and the SMR deletion were made in yKD143 as described in Anand et al. 2017, using plasmids pKD163, pKD165, and pKD169 and homology directed repair templates.

Recipes for media used in this study are listed in Methods-table supplement 1.

For gal-induced growths, overnight cultures were grown in YPAGR media, or, for strains with plasmids, overnight cultures were grown in SC/A/GR/-Ura.

Overnights were diluted in the same media to an OD of 0.1 and harvested at an

OD of 0.4-0.5.

119 Flow cytometry:

Data Collection:

100 μl of log-phase cells were pelleted and washed once with 1 x PBS. Cells were then resuspended in 500 μl of 1 x PBS and 5000 cells were analyzed with a

Millipore guava easyCyte flow cytometer for GFP and RFP detection using 488 nm and 532 nm excitation lasers, respectively.

Data Analysis:

Cells were gated based on size, and random outliers were cut off from graphs for visual purposes (plot values were not changed upon removal of outliers). For all flow cytometry data, violin plots show the density of cells for log2(GFP/RFP) values for individual cells. Flow cytometry was done in triplicate, with each group of cells taken from individual growths. For triplicate plots, the average of each individual flow cytometry sample was taken and plotted. Machine settings remained constant between samples.

For statistical analysis of flow cytometry triplicates, a standard t-test was run in R and p-values are reported.

Northern Blots:

RNA isolation

25 mls of log-phase cells were pelleted and supernatant was poured off. Cells were resuspended in residual media, pelleted again, and flash frozen in liquid nitrogen. Cell pellet was resuspended in buffer with 8.4 mM EDTA, and 60 mM

NaOAc. 20% SDS was added to a final concentration of 1.5%. Cell solution was

120 warmed at 65°C for 2 minutes and added to acid phenol at 65°C. Phenol/cell solution was shaken at 1100 rpm at 65°C for 10-20 minutes with intermittent vortexing. Samples were put on ice for 5 minutes then spun at 16 krpm. The aqueous layer was removed and added to an equal volume of phenol, and samples were vortexed again. Samples were spun at 16 krpm again, the aqueous layer was removed and added to an equal volume of chloroform, and samples were vortexed. Lastly, samples were spun again at 16 krpm, the aqueous layer was removed and precipitated in NaOAc and isopropanol. The

RNA pellet was resuspended in 10 mM Tris-HCl, pH 8.0.

Northern blot

Between 5 and 10 μg of RNA were loaded into a 1.2% agarose, formaldehyde denaturing gel and run for 2-2.5 hours at 125 volts (for any given gel, the same amount of RNA was loaded for each sample, but this varied from gel to gel). The

RNA was vacuum transferred to a nitrocellulose (N+ Hbond, Amersham) membrane in 10 x SSC buffer. The RNA was then UV crosslinked to the membrane and placed in pre-hybridization buffer and rotated at 42°C. The indicated PNK-end labeled probe was added to the pre-hybridization buffer at

42°C after 30 minutes (Methods-table supplement 1). The membrane was probed overnight, rotating at 42°C. The membrane was washed 3 times in 2 x

SSC, 0.1% SDS for 20 minutes at 30°C, then exposed to a phosphoscreen. The phosphoscreen was scanned using a Typhoon FLA 9500.

End-labeled DNA probe

121 This indicated DNA-oligo listed in Methods-table supplement 1 was end labeled using gamma-ATP and the standard T4 Polynucleotide Kinase radioactive labeling protocol from NEB. The labeled oligo was purified using GE Healthcare illustra ProbeQuant G-50 Micro Columns.

Western Blots:

Protein Isolation

2 OD units of log-phase cells were pelleted and supernatant was poured off.

Cells were resuspended in residual media, pelleted again, and flash frozen in liquid nitrogen. Pellets were resuspended in 200 μl lysis buffer containing 20mM

Tris-HCl, pH 8.0, 140mM KCl, 5mM MgCl2, 1% triton, 1mM DTT, Roche cOmplete protease inhibitor tablet, and PMSF, pepstatin, and leupeptin protease inhibitors. The volume of cell solution was approximately doubled using acid washed glass beads and cells were mechanically lysed using bead beater 1 min on, 1 min off, for 3 cycles. 6x SDS loading dye was added to the lysis solution at

2x and samples were boiled for 5 min.

Western blot

Equal volumes of lysate were loaded for each sample onto 4-12% Criterion XT

Bis-Tris protein gels in 1x XT MES buffer. Protein was transferred to PVDF membrane via turbo blot. Membranes were then placed in 2.5% milk, 1x TBST blocking solution for 1 hour. Primary antibody was used in 1x TBST at either

1:50,000 for anti-eEF2 (Kerafast, rabbit), or 1:5000 for anti-PGK1 (Invitrogen, mouse), anti-HA (Roche, rat), anti-FLAG (Sigma, mouse), and anti-GFP (Takara,

122 mouse) and incubated on a rotator overnight at 4°C. Membranes were washed in

1x TBST 3 times, 10 minutes each. The corresponding HRP-conjugated secondary antibody was added to the membrane at 1:5000 in 1x TBST and incubated for 1-2 hours. Membranes were washed 3 times for 10 minutes each.

Pico solution (details) was added to membranes for approximately 3 minutes and then membranes were scanned using a G:BOX Chemi XX6 (Syngene) with varying exposure times.

Reporter-SGA screens

SGA procedure

SGA screens were performed using a Biomatrix Robot (S&P Robotics Inc.) with a few modifications (61). Briefly, yKD176, yKD177, and yKD178 query strains

(Methods) were crossed individually with the yeast nonessential deletion library

(34) and yKD131, yKD132, yKD133 query strains (Methods-table supplement 2) were crossed individually with the FLEX collection (32, 33). Both the deletion and

FLEX libraries were arrayed in a 1536-format containing 4 colonies for each deletion/FLEX strain. Because we found that our query strains had a slightly lower mating efficiency and growth defect, incubation times for every step of the

SGA protocol were prolonged by 50-75%. Mating and sporulation steps were performed on standard SGA media (61).

For the deletion screen, diploid strains were selected on DIP media and double haploid mutant strains were selected on HAP media listed in Methods-table supplement 2.

123 For the overexpression screen, diploid strains were selected on OEDIP media listed in Methods-table supplement 2. Sporulation and haploid double mutant selection steps were not performed in this screen.

To induce reporter expression and FLEX gene overexpression, cells were pinned again onto the same medium (haploid double mutant selection medium for the deletion screen or diploid selection medium for the FLEX screen) except that glucose was replaced by raffinose and galactose at a final concentration of 2%

(HAPGR or OEDIPGR media listed in Methods-table supplement 2). Cells were grown for 26-30 hours for the deletion screen and 40-46 hours for the overexpression screen before scanning on a Typhoon FLA9500 (GE Healthcare) fluorescence scanner equipped with 488 nm and 532 nm excitation lasers and

520/40 and 610/30 emission filters. Plates were also photographed using a robotic system developed by S&P Robotics Inc. in order to determine colony size.

Screen data analysis:

GFP and RFP fluorescent intensity data was collected using the microarray software, TIGR Spotfinder (Saeed AI, Sharov V, White J, Li J, Liang W, et al.

TM4: a free, open-source system for microarray data management and analysis.

Biotechniques. 2003;34:374–378). Colony size data was aquired using

SGATools (62) (http://sgatools.ccbr.utoronto.ca/). Subsequent data analysis was performed as previously described (53, 55). In brief, border strains and size outliers (<1500 or >6000 pixels) were eliminated, and median GFP and RFP values were taken for the remaining strains. Log2(mean GFP/mean RFP) values

124 were then calculated and LOESS normalized for each plate. Finally, Z-scores for each individual plate were calculated based on the LOESS normalized log2(mean

GFP/mean RFP). Strains for the AAA or CGA reporters with Z-scores greater than 2.5 or less than -2.5 were considered as hits if their Z-score in the OPT reporter was unaffected (i.e had a Z-score between -2.5 and 2.5). For subsequent experiments, hits were reconstructed in the background strain yKD133 as described above and all experiments were performed with the reconstructed strain.

Venn Diagrams:

Deletion or overexpression screen strains with a Z-score greater than 2.5 or less than -2.5 were considered candidate genes and analyzed for overlap using

BioVenn (http://www.biovenn.nl/) to produce the diagrams.

GO Analysis:

Yeastmine (https://yeastmine.yeastgenome.org/yeastmine/) was used to calculate GO term for Biological Processes enrichment using a p-value cutoff of

0.05 after Holm-Bonferoni correction. Categories that grouped identical sets of genes were condensed into one, and labeled with the most descriptive biological process.

Cue2-SMR prep

The SMR and the R402A-SMR constructs, pKD097 and pKD098, respectively, were expressed in RIPL BL21 E. coli strain using Kan resistance and inducing at

OD=0.4, at 18°C with IPTG at 0.5 mM overnight. Cells were harvested by

125 centrifugation and then flash frozen in liquid nitrogen. Pellets were then resuspended in protein prep lysis buffer [25 mM Tris-Cl pH 7.5, 500 mM KCl, 1 mM MgCl2, 5 mM bMe, 10% glycerol] + PMSF, leupeptin, pepstatin, and Roche cOmplete EDTA-free protease inhibitors.

Cells were lysed using the French press, 3 times, 1100 pressure units. Lysis solution was clarified at 20000xg for 30min. Lysate was filtered through a 0.2 um filter and run over HiTrap 5ml Ni-NTA column. Column was washed for 5 CV in lysis buffer with high salt (1 M KCl) and 20 mM imidazole. Sample was batch eluted using Ni elution buffer [25 mM Tris-Cl pH 7.5, 500 mM KCl, 1 mM MgCl2,

5 mM bMe, 10% glycerol, 500 mM imidazole]. Sample was diluted 10 fold into S column buffer [25 mM Tris-Cl pH 7.5, 200 mM KCl, 5 mM MgCl2, 5 mM bMe,

10% glycerol] and then run on a Resource S column. Sample was washed in S column buffer for 5 CV and then gradient eluted off of S column in S column buffer with 1 M salt. 6xHIS-Sumo tag was cleaved overnight using SUMO protease. And sample was run on an orthogonal Ni column to remove tag. Flow through was collected and concentrated to 40 uM SMR, and 20 uM SMR-R402A.

Isolation of nuclease resistant trisomes

Grow 1 liter of yKD307 in YPAGR to an OD 0.4. For cycloheximide treated cells, added 1 mg/L cycloheximide, grew for 30 minutes and filter harvested. For untreated cells, immediately filter harvested. Cell pellets were ground with 1 mL lysis buffer [10 mM potassium phosphate (pH 6.1), 140 mM KCl, 5 mM MgCl2,

1% Triton X-100, 0.1 mg/mL cycloheximide, 1mM DTT, Roche cOmplete EDTA- free protease inhibitor, leupeptin, PMSF, pepstatin] in a Spex 6870 freezer mill.

126 Cell lysates were clarified by centrifugation. CaCl2 was added to 200 OD units of clarified lysates to a final concentration of 2.5 mM and the indicated clarified lysates were treated with 20 ul of NEB MNase at 2e^6 gel units/ml for 30 minutes. 4 mM EGTA was added to quench the digestion. Samples were layered on a 15-45% sucrose gradient [10 mM potassium phosphate (pH 6.1), 140 mM

KCl, 5 mM MgCl2] and spun in SW28 rotor at max speed for 4 hours. Nuclease resistant trisome peak was isolated and diluted two fold in lysis buffer.

Cleavage of lysates with SMR and analysis

Protein prep buffer, the purified SMR domain of Cue2, and the R402A mutant of the SMR domain of Cue2 were added to separate aliquots of isolated nuclease resistant trisomes, such that the final concentration of enzyme (for samples with enzyme) = 3 µM. Samples were left at room temperature for 2 hours. Next, samples were run on a 15-45% sucrose gradient [10 mM potassium phosphate

(pH 6.1), 140 mM KCl, 5 mM MgCl2] and the A260 trace is reported. RPFs were extracted from pooled fractions (mono-, di-, and trisomes).

Sequence alignment

Information from NCBI conserved domain database (Marchler-Bauer et al. 2011), conserved domain architecture retrieval tool (Geer et al. 2002), available structures, and Phyre-based homology modeling (Kelley et al. 2015) was used to define domain boundaries. Structure-based multiple sequence alignments were carried out using Expresso (Armougom et al. 2006), and illustrated with ESPript

(Robert and Gouet 2014). Protein names are indicated, followed by the name of organism, and residues used for alignment. YEAST, Saccharomyces cerevisiae;

127 CANGA, Candida glabrata; ARATH, Arabidopsis thaliana; HUMAN, Homo sapiens; MACMU, Macaca mulatta; BOVIN, Bos taurus; THETH, Thermus thermophilus; BACSU, Bacillus subtilis; ECOLI, Escherichia coli.

Homology modeling

Homology model of the Cue2 SMR domain templated on human N4BP2 SMR domain (PDB: 2VKC (Diercks et al. 2008) was generated using Phyre 2.0 (Kelley et al. 2015) and SWISS-MODEL (Waterhouse et al. 2018). After identifying

N4BP2 as a Cue2 homolog, a heuristic DALI search was performed using the

N4BP2 SMR domain (PDB IDs: 2D9I and 2VKC) as query against structures in the Protein Data Bank (PDB). This exercise, along with templates identified using

Phyre 2.0 (Kelley et al. 2015) and SWISS-MODEL (Waterhouse et al. 2018), enabled us to identify structural conservation between the SMR domain of Cue2 and the C-terminal domain (CTD) of translational initiation factor 3 (IF3). This was further validated using structure-based sequence alignments as shown in

Figure 2-figure supplement 1B. Structural comparisons and fittings were performed using UCSF Chimera (Goddard, Huang, and Ferrin 2007; Pettersen et al. 2004). Superimposition of the CTD of IF3 (PDB: 1TIG (Biou, Shu, and

Ramakrishnan 1995)) and the Cue2-SMR homology model revealed structural conservation between the two domains (Figure 2-figure supplement 1C). This enabled superimposition of the Cue2 SMR in the context of full-length Thermus thermophilus IF3 (PDB: 5LMQ (Hussain et al. 2016)) (see Figure 2C). This fitting enabled overlay of the Cue2-SMR at the A/P-site in the context of IF3 and tRNAfMet bound 30S pre-initiation complex (PDB: 5LMQ, State 2A (Hussain et

128 al. 2016)) (see Figure 2D and Figure 2-figure supplement 1, D-E). The mRNA and the 30S ribosomal subunit are represented as surfaces to minimize over interpretation (Figure 2D and Figure 2-figure supplement 1, D-E).

Yeast growth conditions for ribosome profiling

Overnight seed cultures were grown in YPAD at 30°C. To induce NGD-CGA reporter expression, cells were collected by centrifugation, washed, and resuspended in YPAGR. Cells were then harvested at OD ~0.5 by fast filtration and flash frozen in liquid nitrogen.

Preparation of libraries for yeast ribosome footprints

Cell pellets were ground with 1 mL footprint lysis buffer [20 mM Tris-Cl (pH8.0),

140 mM KCl, 1.5 mM MgCl2, 1% Triton X-100, 0.1 mg/mL cycloheximide, 0.1 mg/mL tigecycline] in a Spex 6870 freezer mill. Lysed cell pellets were diluted to

15 mL in footprint lysis buffer and clarified by centrifugation. The supernatant was layered on a sucrose cushion [20 mM Tris-Cl (pH 8.0), 150 mM KCl, 5 mM

MgCl2, 0.5 mM DTT, 1M sucrose]. Polysomes were pelleted by centrifugation at

60,000 rpm for 106 min in a Type 70Ti rotor (Beckman Coulter). Ribosome pellets were gently resuspended in 800 µL footprint lysis buffer. 350 µg of isolated polysomes were treated with 500 units of RNaseI (Ambion) for 1 hr at

25˚C. Monosomes or disomes were isolated by sucrose gradients (10-50%).

RNA was extracted by hot acid phenol and then size-selected from 15% denaturing PAGE gels, cutting between 15-34 nt for monosome footprints, 40-80 nt for disome footprints, and 15-90 nt for in vitro Cue2 cleavage assays. Library construction was carried out as described (Wu et al. 2019). Libraries were

129 sequenced on a HiSeq2500 machine at facilities at the Johns Hopkins Institute of

Genetic Medicine.

Analysis of ribosome profiling data

The R64-1-1 S288C reference genome assembly (SacCer3) from the

Saccharomyces Genome Database Project was used for yeast genome alignment. For the rest of our libraries, 3’ adapter

(NNNNNNCACTCGGGCACCAAGGA) was trimmed, and 4 random nucleotides included in RT primer

(RNNNAGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGT

CGC/iSP18/TTCAGACGTGTGCTCTTCCGATCTGTCCTTGGTGCCCGAGTG) were removed from the 5’ end of reads with skewer (54). Trimmed reads were aligned to yeast ribosomal and non-coding RNA sequences using STAR (49) with ‘-- outFilterMismatchNoverLmax 0.3’. Unmapped reads were then mapped to genome using the following options ‘--outFilterIntronMotifs

RemoveNoncanonicalUnannotated --outFilterMultimapNmax 1 -- outFilterMismatchNoverLmax 0.1’. All other analyses were performed using software custom written in Python 2.7 and R 3.3.1.

For monosome footprints, the offset of the A site from the 5’ end of reads was calibrated using start codons of CDS (Wu et al. 2019). For 28 nt RPFs, offsets

(27:[16], 28:[16], 29:[17], 30:[17], 31:[17], 32:[17]) were used to infer the A sites of 27-32 nt reads. For 21 nt RPFs, offsets (20:[16], 21:[17], 22:[17]) were used.

For 16 nt RPFs, 3’ ends of 15- 17 nt RPFs were used to infer cleavage sites. For

60 nt disome footprints, offsets (57:[47],58:[47],59:[47],60:[47],61:[47],62:[47])

130 were used to infer the A sites of lead ribosomes. For 54 nt disome footprints, offsets (51:[47],52:[47],53:[47],54:[47]) were used. For 46 nt disome footprints,

3’ends of 44-48 nt RPFs were used to infer cleavage sites. For in vitro cleavage assays, 3’ends of 60-65 nt were used to infer Cue2 SMR cleavage sites.

Prematurely polyadenylated mRNAs were identified by monosome footprints (15-

34 nt) with the criteria of at least three RPFs of more than one untemplated A’s in the 3’end from both slh1∆ cue2∆ dom34∆ ski2∆ datasets. Cue2 target genes were identified using 15-17 nt RPFs that exhibited reproducible reduction upon

Cue2 deletion (adjusted p < 0.5, DESeq (Anders Simon and Huber Wolfgang

2010)).

131 Chapter V

Conclusion

Cells have many overlapping pathways acting simultaneously on ribosome- stalling substrates to prevent a buildup of toxic aberrant peptides. Throughout this thesis I focus on how mRNA decay is signaled by ribosome stalling and the interplay between a subset of known ribosome rescue factors and mRNA decay.

In this we have learned that the SMR domain-containing enzyme Cue2 cleaves

No Go Decay mRNAs independent of canonical mRNA decay. Furthermore, inhibiting release of ribosomes from a translation-stalling mRNA signals a stronger decay pathway through both Xrn1 and Cue2. Interestingly, ubiquitination by the E3 ligase Hel2 is necessary for Cue2 cleavage, but unnecessary for Xrn1 dependent decay. In fact, in a Hel2 knockout, mRNA decay is enhanced, likely through an accumulation of ribosomes followed by Xrn1-dependent decay, although we haven’t thoroughly tested this hypothesis yet.

Cue2 contains multiple ubiquitin binding domains, for which the ubiquitin- binding site has not yet been determined. Whether the role of Hel2-placed ubiquitin in the Cue2-cleavage process is direct recruitment of Cue2 or indirect upstream signaling of further processes that then recruit Cue2 is still unknown.

Correspondences with the Inada lab suggest that in a ribozyme, Cue2 can cleave independently of Hel2, but we have not yet tested this ourselves. The specific ubiquitin binding sites for Cue2 could be characterized further using yeast lysates

132 in vitro, which is currently being set up in the lab in collaboration with the

Beckmann group.

Cue2-dependent cleavage events become a major pathway when the ribosome rescue factor SLH1 is deleted. Interestingly, upon deletion of SLH1, we do not see truncated peptide products that would in theory accumulate upon translation of Cue2 mRNA cleavage products (data not shown). This is consistent with data from Matsuo et al, which shows stabilization of full length protein in the slh1Δ, but no truncated peptide in a stalling reporter (Matsuo et al. 2017). Ltn1 is the E3 ligase that signals degradation of CAT-tailed peptide products after RQT- dependent release. Deletion of this factor Ltn1 does not seem to stabilize the nascent peptides translated from a Cue2-cleaved mRNA, although profiling data suggests there are ribosomes that have translated these cleaved mRNAs (our negative data not shown, (Matsuo et al. 2017)). It is possible that the peptide products that are made from translation of truncated mRNAs are degraded in a manner independent of Ltn1. This truncated peptide product within the 60S subunit is exposed by the action of Dom34/Hbs1 ribosome splitting, however, the path for degradation of a peptide within a Dom34/Hbs1 split ribosome has not been shown in vivo. Future experiments analyzing ribozymes (substrates where the major pathway for rescue is through Dom34) and removing redundancies between overlapping pathways (SKI complex and RQT) will elucidate this mechanism.

Deubiquitination of ribosomal proteins is likely to occur at stalling reporters to counteract the known Hel2-catalyzed ubiquitination (Ubn) of 40S subunits.

133 However, deubiquitinases (DUBs) involved in these processes are currently unknown. From the NGD deletion screens performed here, we identify a subset of DUBs that are candidates for competing with Hel2-Ubn as well as removal of other ubiquitin (Ub) marks that are involved in these processes. It is possible that the Ub effect here is nonspecific (general cell-wide depletion of Ub), or specific

(precise Ub marks signal mRNA decay or affect mRNA decay factors).

Regardless, it will be interesting to learn how ubiquitin plays a role in mRNA decay through these DUBs.

Currently, we are studying many interesting new genes and pathways that require further characterization to learn their precise role in the cell’s quality control processes. The work done here begins to elucidate the complex interplay between ribosome rescue pathways, mRNA decay, and peptide degradation, while uncovering many new questions in the field of quality control.

134 References

Appendices

Extended Table 1: OPT reporter OE screen

Extended Table 2: NGD-CGA reporter OE screen

Extended Table 3: NGD-AAA reporter OE screen

Extended Table 4: OPT reporter deletion screen

Extended Table 5: NGD-CGA reporter deletion screen

Extended Table 6: NGD-AAA reporter deletion screen

Extended Table 7: OPT reporter temperature sensitive screen

Extended Table 8: NGD-CGA reporter temperature sensitive screen

Extended Table 9: NGD-AAA reporter temperature sensitive screen

Methods-table supplement 1

Methods-table supplement 2

135 Bibliography

Alamgir, Md et al. 2010. “Chemical-Genetic Profile Analysis of Five Inhibitory

Compounds in Yeast.” BMC chemical biology 10: 6.

http://www.ncbi.nlm.nih.gov/pubmed/20691087.

Anand, Ranjith, Gonen Memisoglu, and James Haber. “Cas9-Mediated Gene

Editing in Saccharomyces Cerevisiae.” Nature Research Protocol Exchange

Version 1([+https://doi.org/10.1038/protex.2017.021a+]).

Anders Simon, and Huber Wolfgang. 2010. “Differential Expression Analysis for

Sequence Count Data.” Genome Biology.

Anderson, J. S.J., and R P Parker. 1998. “The 3’ to 5’ Degradation of Yeast

MRNAs Is a General Mechanism for MRNA Turnover That Requires the

SKI2 DEVH Box Protein and 3’ to 5’ Exonucleases of the Exosome

Complex.” The EMBO journal 17(5): 1497–1506.

http://www.ncbi.nlm.nih.gov/pubmed/9482746.

Armougom, Fabrice et al. 2006. “Expresso: Automatic Incorporation of Structural

Information in Multiple Sequence Alignments Using 3D-Coffee.” Nucleic

acids research 34(Web Server issue): W604-8.

http://www.ncbi.nlm.nih.gov/pubmed/16845081.

Arribere, Joshua A, and Andrew Z Fire. 2018. “Nonsense MRNA Suppression via

Nonstop Decay.” eLIFE 7: 1–23.

Arthur, Laura et al. 2015. “Translational Control by Lysine-Encoding A-Rich

Sequences.” Science advances 1(6).

http://www.ncbi.nlm.nih.gov/pubmed/26322332.

136 Ash, Miriam-Rose et al. 2010. “Conserved Beta-Hairpin Recognition by the GYF

Domains of Smy2 and GIGYF2 in MRNA Surveillance and Vesicular

Transport Complexes.” Structure (London, England : 1993) 18(8): 944–54.

http://dx.doi.org/10.1016/j.str.2010.04.020.

Balchin, David, Manajit Hayer-Hartl, and F. Ulrich Hartl. 2016. “In Vivo Aspects of

Protein Folding and Quality Control.” Science (New York, N.Y.) 353(6294):

aac4354. http://www.ncbi.nlm.nih.gov/pubmed/27365453.

Bengtson, Mario H., and Claudio A P Joazeiro. 2010. “Role of a Ribosome-

Associated E3 Ubiquitin Ligase in Protein Quality Control.” Nature

467(7314): 470–73. http://dx.doi.org/10.1038/nature09371.

Biou, V, Fong Shu, and V Ramakrishnan. 1995. “X-Ray Crystallography Shows

That Translational Initiation Factor IF3 Consists of Two Compact Alpha/Beta

Domains Linked by an Alpha-Helix.” The EMBO journal 14(16): 4056–64.

http://www.ncbi.nlm.nih.gov/pubmed/7664745.

Brandman, Onn et al. 2012. “A Ribosome-Bound Quality Control Complex

Triggers Degradation of Nascent Peptides and Signals Translation Stress.”

Cell 151(5): 1042–54.

http://linkinghub.elsevier.com/retrieve/pii/S0092867412013396.

Brandman, Onn, and Ramanujan S Hegde. 2016. “Ribosome-Associated Protein

Quality Control.” Nature structural & molecular biology 23(1): 7–15.

http://www.ncbi.nlm.nih.gov/pubmed/26733220.

Buschauer, Robert et al. 2019. “The Ccr4-Not Complex Monitors the Translating

Ribosome for Codon Optimality.” bioRxiv: 854810.

137 https://www.biorxiv.org/content/10.1101/854810v1?ct=.

Chan, Leon Y. et al. 2018. “Non-Invasive Measurement of MRNA Decay Reveals

Translation Initiation as the Major Determinant of MRNA Stability.” eLife 7:

1–32.

Chandrasekaran, Viswanathan et al. 2019. “Mechanism of Ribosome Stalling

during Translation of a Poly(A) Tail.” Nature structural & molecular biology

26(12): 1132–40. http://www.nature.com/articles/s41594-019-0331-x.

D’Orazio, Karole N et al. 2019. “The Endonuclease Cue2 Cleaves MRNAs at

Stalled Ribosomes during No Go Decay.” eLife 8: 1–27.

http://www.ncbi.nlm.nih.gov/pubmed/31219035.

Dehecq, Marine et al. 2018. “Nonsense-Mediated MRNA Decay Involves Two

Distinct Upf1-Bound Complexes.” The EMBO journal 37(21): e99278.

http://www.ncbi.nlm.nih.gov/pubmed/30275269.

Diercks, Tammo et al. 2008. “Solution Structure and Characterization of the

DNA-Binding Activity of the B3BP-Smr Domain.” Journal of molecular

biology 383(5): 1156–70. http://www.ncbi.nlm.nih.gov/pubmed/18804481.

Doma, Meenakshi K., and Roy Parker. 2006. “Endonucleolytic Cleavage of

Eukaryotic MRNAs with Stalls in Translation Elongation.” Nature 440(7083):

561–64.

Douglas, Alison C. et al. 2012. “Functional Analysis with a Barcoder Yeast Gene

Overexpression System.” G3 (Bethesda, Md.) 2(10): 1279–89.

http://www.ncbi.nlm.nih.gov/pubmed/23050238.

Ferrin, Michael A, and Arvind R Subramaniam. 2017. “Kinetic Modeling Predicts

138 a Stimulatory Role for Ribosome Collisions at Elongation Stall Sites in

Bacteria.” eLife 6: 1–19. http://www.ncbi.nlm.nih.gov/pubmed/28498106.

Fillingham, Jeffrey et al. 2009. “Two-Color Cell Array Screen Reveals

Interdependent Roles for Histone Chaperones and a Chromatin Boundary

Regulator in Histone Gene Repression.” Molecular Cell 35(3): 340–51.

Frischmeyer, Pamela A et al. 2002. “An MRNA Surveillance Mechanism That

Eliminates Transcripts Lacking Termination Codons.” Science (New York,

N.Y.) 295(5563): 2258–61. http://www.ncbi.nlm.nih.gov/pubmed/11910109.

Fukui, Kenji, and Seiki Kuramitsu. 2011. “Structure and Function of the Small

MutS-Related Domain.” Molecular biology international 2011: 691735.

http://www.ncbi.nlm.nih.gov/pubmed/22091410.

Gamble, Caitlin E. et al. 2016. “Adjacent Codons Act in Concert to Modulate

Translation Efficiency in Yeast.” Cell 166(3): 679–90.

http://dx.doi.org/10.1016/j.cell.2016.05.070.

Garzia, Aitor et al. 2017. “The E3 Ubiquitin Ligase and RNA-Binding Protein

ZNF598 Orchestrates Ribosome Quality Control of Premature

Polyadenylated MRNAs.” Nature communications 8(May): 16056.

http://www.ncbi.nlm.nih.gov/pubmed/28685749.

Geer, Lewis Y., Michael Domrachev, David J. Lipman, and Stephen H. Bryant.

2002. “CDART: Protein Homology by Domain Architecture.” Genome

research 12(10): 1619–23.

http://genome.cshlp.org/content/12/10/1619.abstract.

Giaever, Guri et al. 2002. “Functional Profiling of the Saccharomyces Cerevisiae

139 Genome.” Nature 418(6896): 387–91.

http://www.ncbi.nlm.nih.gov/pubmed/12140549.

Glavan, Filip, Isabelle Behm-Ansmant, Elisa Izaurralde, and Elena Conti. 2006.

“Structures of the PIN Domains of SMG6 and SMG5 Reveal a Nuclease

within the MRNA Surveillance Complex.” The EMBO journal 25(21): 5117–

25. http://www.ncbi.nlm.nih.gov/pubmed/17053788.

Goddard, Thomas D., Conrad C. Huang, and Thomas E. Ferrin. 2007.

“Visualizing Density Maps with UCSF Chimera.” Journal of structural biology

157(1): 281–87. http://www.ncbi.nlm.nih.gov/pubmed/16963278.

Guydosh, Nicholas, and Rachel Green. 2017. “Translation of Poly(A) Tails Leads

to Precise MRNA Cleavage.” Rna 23: 749–61.

Guydosh, Nicholas R., and Rachel Green. 2014. “Dom34 Rescues Ribosomes in

3’ Untranslated Regions.” Cell 156(5): 950–62.

http://dx.doi.org/10.1016/j.cell.2014.02.006.

Guydosh, Nicholas R., Philipp Kimmig, Peter Walter, and Rachel Green. 2017.

“Regulated Ire1-Dependent MRNA Decay Requires No-Go MRNA

Degradation to Maintain Endoplasmic Reticulum Homeostasis in S. Pombe.”

eLife 6. http://www.ncbi.nlm.nih.gov/pubmed/28945192.

Halbach, Felix, Peter Reichelt, Michaela Rode, and Elena Conti. 2013. “The

Yeast Ski Complex: Crystal Structure and RNA Channeling to the Exosome

Complex.” Cell 154(4): 814–26. http://dx.doi.org/10.1016/j.cell.2013.07.017.

Hickey, Kelsey L et al. 2019. “GIGYF2 and 4EHP Inhibit Translation Initiation of

Defective Messenger RNAs to Assist Ribosome-Associated Quality Control.”

140 bioRxiv: 792994. http://biorxiv.org/content/early/2019/10/03/792994.abstract. van Hoof, Ambro, Pamela A Frischmeyer, Harry C Dietz, and Roy Parker. 2002.

“Exosome-Mediated Recognition and Degradation of MRNAs Lacking a

Termination Codon.” Science (New York, N.Y.) 295(5563): 2262–64.

http://www.ncbi.nlm.nih.gov/pubmed/11910110.

Hu, Wenqian et al. 2009. “Co-Translational MRNA Decay in Saccharomyces

Cerevisiae.” Nature 461(7261): 225–29.

http://dx.doi.org/10.1038/nature08265.

Hu, Yanhui et al. 2007. “Approaching a Complete Repository of Sequence-

Verified Protein-Encoding Clones for Saccharomyces Cerevisiae.” Genome

research 17(4): 536–43. http://www.ncbi.nlm.nih.gov/pubmed/17322287.

Hussain, Tanweer et al. 2016. “Large-Scale Movements of IF3 and TRNA during

Bacterial Translation Initiation.” Cell 167(1): 133–144.e13.

http://www.ncbi.nlm.nih.gov/pubmed/27662086.

Ibrahim, Fadia, Manolis Maragkakis, Panagiotis Alexiou, and Zissimos

Mourelatos. 2018. “Ribothrypsis, a Novel Process of Canonical MRNA

Decay, Mediates Ribosome-Phased MRNA Endonucleolysis.” Nature

Structural and Molecular Biology 25(4): 302–10.

http://dx.doi.org/10.1038/s41594-018-0042-8.

Ikeuchi, Ken et al. 2019. “Collided Ribosomes Form a Unique Structural Interface

to Induce Hel2-Driven Quality Control Pathways.” The EMBO journal 38(5):

e100276.

http://emboj.embopress.org/lookup/doi/10.15252/embj.2018100276.

141 Ikeuchi, Ken, and Toshifumi Inada. 2016. “Ribosome-Associated Asc1/RACK1 Is

Required for Endonucleolytic Cleavage Induced by Stalled Ribosome at the

3′ End of Nonstop MRNA.” Scientific Reports 6(June): 1–10.

http://dx.doi.org/10.1038/srep28234.

Ishimura, Ryuta et al. 2014. “RNA Function. Ribosome Stalling Induced by

Mutation of a CNS-Specific TRNA Causes Neurodegeneration.” Science

(New York, N.Y.) 345(6195): 455–59.

http://www.ncbi.nlm.nih.gov/pubmed/25061210.

Ito-Harashima, Sayoko, Kazushige Kuroha, Tsuyako Tatematsu, and Toshifumi

Inada. 2007. “Translation of the Poly(A) Tail Plays Crucial Roles in Nonstop

MRNA Surveillance via Translation Repression and Protein Destabilization

by Proteasome in Yeast.” Genes & development 21(5): 519–24.

http://www.ncbi.nlm.nih.gov/pubmed/17344413.

Johnson, Sean J., and Ryan N. Jackson. 2013. “Ski2-like RNA Helicase

Structures: Common Themes and Complex Assemblies.” RNA biology 10(1):

33–43. http://www.ncbi.nlm.nih.gov/pubmed/22995828.

Juszkiewicz, Szymon et al. 2018. “ZNF598 Is a Quality Control Sensor of

Collided Ribosomes.” Molecular Cell: 1–13.

https://linkinghub.elsevier.com/retrieve/pii/S109727651830697X.

Juszkiewicz, Szymon, and Ramanujan S. Hegde. 2017. “Initiation of Quality

Control during Poly(A) Translation Requires Site-Specific Ribosome

Ubiquitination.” Molecular Cell 65(4): 743–750.e4.

Kainth, Pinay et al. 2009. “Comprehensive Genetic Analysis of Transcription

142 Factor Pathways Using a Dual Reporter Gene System in Budding Yeast.”

Methods 48(3): 258–64. http://dx.doi.org/10.1016/j.ymeth.2009.02.015.

Kang, Richard S. et al. 2003. “Solution Structure of a CUE-Ubiquitin Complex

Reveals a Conserved Mode of Ubiquitin Binding.” Cell 113(5): 621–30.

http://www.ncbi.nlm.nih.gov/pubmed/12787503.

Karlsborn, Tony et al. 2014. “Elongator, a Conserved Complex Required for

Wobble Uridine Modifications in Eukaryotes.” RNA biology 11(12): 1519–28.

http://www.ncbi.nlm.nih.gov/pubmed/25607684.

Kelley, Lawrence A et al. 2015. “The Phyre2 Web Portal for Protein Modeling,

Prediction and Analysis.” Nature protocols 10(6): 845–58.

http://www.ncbi.nlm.nih.gov/pubmed/25950237.

Koutmou, Kristin S. et al. 2015. “Ribosomes Slide on Lysine-Encoding

Homopolymeric A Stretches.” eLife 2015(4): 1–18.

Kuroha, Kazushige et al. 2010. “Receptor for Activated C Kinase 1 Stimulates

Nascent Polypeptide-Dependent Translation Arrest.” EMBO reports 11(12):

956–61. http://dx.doi.org/10.1038/embor.2010.169.

Lareau, L F, D H Hite, G J Hogan, and P O Brown. 2014. “Distinct Stages of the

Translation Elongation Cycle Revealed by Sequencing Ribosome-Protected

MRNA Fragments.” Elife 3: e01257.

Letzring, D. P., A. S. Wolf, C. E. Brule, and E. J. Grayhack. 2013. “Translation of

CGA Codon Repeats in Yeast Involves Quality Control Components and

Ribosomal Protein L1.” Rna 19(9): 1208–17.

http://rnajournal.cshlp.org/cgi/doi/10.1261/rna.039446.113.

143 Letzring, Daniel P., Kimberly M. Dean, and Elizabeth J. Grayhack. 2010. “Control

of Translation Efficiency in Yeast by Codon-Anticodon Interactions.” RNA

(New York, N.Y.) 16(12): 2516–28.

http://www.ncbi.nlm.nih.gov/pubmed/20971810.

Li, Zhijian et al. 2011. “Systematic Exploration of Essential Yeast Gene Function

with Temperature-Sensitive Mutants.” Nature biotechnology 29(4): 361–67.

http://www.ncbi.nlm.nih.gov/pubmed/21441928.

Longtine, Mark S et al. 1998. “Additional Modules for Versatile and Economical

PCR-Based Gene Deletion and Modification in Saccharomyces Cerevisiae.”

Yeast (Chichester, England) 14(10): 953–61.

http://www.ncbi.nlm.nih.gov/pubmed/9717241.

Lykke-Andersen, Søren et al. 2014. “Human Nonsense-Mediated RNA Decay

Initiates Widely by Endonucleolysis and Targets SnoRNA Host Genes.”

Genes & development 28(22): 2498–2517.

http://www.ncbi.nlm.nih.gov/pubmed/25403180.

Marchler-Bauer, Aron et al. 2011. “CDD: A Conserved Domain Database for the

Functional Annotation of Proteins.” Nucleic acids research 39(Database

issue): D225-9. http://www.ncbi.nlm.nih.gov/pubmed/21109532.

Matsuo, Yoshitaka et al. 2017. “Ubiquitination of Stalled Ribosome Triggers

Ribosome-Associated Quality Control.” Nature communications 8(1): 159.

http://dx.doi.org/10.1038/s41467-017-00188-1.

Mircus, Gabriel et al. 2015. “Identification and Characterization of a Novel Family

of Selective Antifungal Compounds (CANBEFs) That Interfere with Fungal

144 Protein Synthesis.” Antimicrobial agents and chemotherapy 59(9): 5631–40.

http://www.ncbi.nlm.nih.gov/pubmed/26149982.

Mischo, Hannah E., and Nick J. Proudfoot. 2013. “Disengaging Polymerase:

Terminating RNA Polymerase II Transcription in Budding Yeast.” Biochimica

et biophysica acta 1829(1): 174–85.

http://dx.doi.org/10.1016/j.bbagrm.2012.10.003.

Navickas, Albertas et al. 2018.

“Https://Www.Biorxiv.Org/Content/10.1101/465633v1.” bioRxiv: 465633.

https://www.biorxiv.org/content/early/2018/11/08/465633.

Ozsolak, Fatih et al. 2010. “Comprehensive Polyadenylation Site Maps in Yeast

and Human Reveal Pervasive Alternative Polyadenylation.” Cell 143(6):

1018–29.

Parker, Roy. 2012. “RNA Degradation in Saccharomyces Cerevisae.” Genetics

191(3): 671–702. http://www.ncbi.nlm.nih.gov/pubmed/22785621.

Pelechano, Vicent, Wu Wei, and Lars M. Steinmetz. 2013. “Extensive

Transcriptional Heterogeneity Revealed by Isoform Profiling.” Nature

497(7447): 127–31. http://www.ncbi.nlm.nih.gov/pubmed/23615609.

Pelechano, Vicent, Wu Wei, and Lars M Steinmetz. 2015. “Widespread Co-

Translational RNA Decay Reveals Ribosome Dynamics.” Cell 161(6): 1400–

1412. http://www.ncbi.nlm.nih.gov/pubmed/26046441.

Pettersen, Eric F et al. 2004. “UCSF Chimera--a Visualization System for

Exploratory Research and Analysis.” Journal of computational chemistry

25(13): 1605–12. http://www.ncbi.nlm.nih.gov/pubmed/29788355.

145 Presnyak, Vladimir et al. 2015. “Codon Optimality Is a Major Determinant of

MRNA Stability.” Cell 160(6): 1111–24.

http://dx.doi.org/10.1016/j.cell.2015.02.029.

Radhakrishnan, Aditya et al. 2016. “The DEAD-Box Protein Dhh1p Couples

MRNA Decay and Translation by Monitoring Codon Optimality.” Cell 167(1):

122–132.e9. http://www.ncbi.nlm.nih.gov/pubmed/27641505. dos Reis, Mario, Renos Savva, and Lorenz Wernisch. 2004. “Solving the Riddle

of Codon Usage Preferences: A Test for Translational Selection.” Nucleic

acids research 32(17): 5036–44.

http://www.ncbi.nlm.nih.gov/pubmed/15448185.

Robert, Xavier, and Patrice Gouet. 2014. “Deciphering Key Features in Protein

Structures with the New ENDscript Server.” Nucleic acids research 42(Web

Server issue): W320-4. http://www.ncbi.nlm.nih.gov/pubmed/24753421.

Di Santo, Rachael, Soufiane Aboulhouda, and David E. Weinberg. 2016. “The

Fail-Safe Mechanism of Post-Transcriptional Silencing of Unspliced HAC1

MRNA.” eLife 5(OCTOBER2016).

Schmidt, Skye A. et al. 2015. “Identification of SMG6 Cleavage Sites and a

Preferred RNA Cleavage Motif by Global Analysis of Endogenous NMD

Targets in Human Cells.” Nucleic acids research 43(1): 309–23.

http://www.ncbi.nlm.nih.gov/pubmed/25429978.

Searfoss, A., T. E. Dever, and R. Wickner. 2001. “Linking the 3’ Poly(A) Tail to

the Subunit Joining Step of Translation Initiation: Relations of Pab1p,

Eukaryotic Translation Initiation Factor 5b (Fun12p), and Ski2p-Slh1p.”

146 Molecular and cellular biology 21(15): 4900–4908.

http://www.ncbi.nlm.nih.gov/pubmed/11438647.

Shao, Sichen, Karina Von der Malsburg, and Ramanujan S. Hegde. 2013.

“Listerin-Dependent Nascent Protein Ubiquitination Relies on Ribosome

Subunit Dissociation.” Molecular Cell 50(5): 637–48.

http://dx.doi.org/10.1016/j.molcel.2013.04.015.

Sharma, Pamila et al. 2012. “2A Peptides Provide Distinct Solutions to Driving

Stop-Carry on Translational Recoding.” Nucleic acids research 40(7): 3143–

51. http://www.ncbi.nlm.nih.gov/pubmed/22140113.

Shen, Peter S et al. 2015. “Protein Synthesis. Rqc2p and 60S Ribosomal

Subunits Mediate MRNA-Independent Elongation of Nascent Chains.”

Science (New York, N.Y.) 347(6217): 75–78.

http://www.ncbi.nlm.nih.gov/pubmed/25554787.

Shoemaker, Christopher J., Daniel E. Eyler, and Rachel Green. 2010.

“Dom34:Hbs1 Promotes Subunit Dissociation and Peptidyl-TRNA Drop-off to

Initiate No-Go Decay.” Science (New York, N.Y.) 330(6002): 369–72.

http://www.ncbi.nlm.nih.gov/pubmed/20947765.

Simms, Carrie L. et al. 2014. “An Active Role for the Ribosome in Determining

the Fate of Oxidized MRNA.” Cell reports 9(4): 1256–64.

http://dx.doi.org/10.1016/j.celrep.2014.10.042.

———. 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(11): e1007818.

147 http://www.ncbi.nlm.nih.gov/pubmed/30475795.

Simms, Carrie L., Liewei L. Yan, and Hani S. Zaher. 2017. “Ribosome Collision Is

Critical for Quality Control during No-Go Decay.” Molecular Cell 68(2): 361–

373.e5. https://doi.org/10.1016/j.molcel.2017.08.019.

Simms, Carrie L, Erica N Thomas, and Hani S Zaher. 2017. “Ribosome-Based

Quality Control of MRNA and Nascent Peptides.” Wiley interdisciplinary

reviews. RNA 8(1): 1–27. http://www.ncbi.nlm.nih.gov/pubmed/27193249.

Sitron, Cole S., Joseph H. Park, and Onn Brandman. 2017. “Asc1, Hel2, and

Slh1 Couple Translation Arrest to Nascent Chain Degradation.” Rna 23(5):

798–810. http://rnajournal.cshlp.org/lookup/doi/10.1261/rna.060897.117.

Sundaramoorthy, Elayanambi et al. 2017. “ZNF598 and RACK1 Regulate

Mammalian Ribosome-Associated Quality Control Function by Mediating

Regulatory 40S Ribosomal Ubiquitylation.” Molecular Cell 65(4): 751–

760.e4. http://dx.doi.org/10.1016/j.molcel.2016.12.026.

Tang, Terence T L, James A W Stowell, Chris H. Hill, and Lori A. Passmore.

2019. “The Intrinsic Structure of Poly(A) RNA Determines the Specificity of

Pan2 and Caf1 Deadenylases.” Nature structural & molecular biology 26(6):

433–42. http://dx.doi.org/10.1038/s41594-019-0227-9.

Tani, Hidenori et al. 2012. “Genome-Wide Determination of RNA Stability

Reveals Hundreds of Short-Lived Noncoding Transcripts in Mammals.”

Genome research 22(5): 947–56.

http://www.ncbi.nlm.nih.gov/pubmed/22369889.

Tesina, Petr et al. 2019. “Molecular Mechanism of Translational Stalling by

148 Inhibitory Codon Combinations and Poly(A) Tracts.” The EMBO journal:

e103365. http://biorxiv.org/content/early/2019/09/04/755652.abstract.

Tong, Amy Hin Yan et al. 2001. “Systematic Genetic Analysis with Ordered

Arrays of Yeast Deletion Mutants.” Science 294(December): 2364–68.

Tsuboi, Tatsuhisa et al. 2012. “Dom34: Hbs1 Plays a General Role in Quality-

Control Systems by Dissociation of a Stalled Ribosome at the 3’ End of

Aberrant MRNA.” Molecular Cell 46(4): 518–29.

http://dx.doi.org/10.1016/j.molcel.2012.03.013.

Waterhouse, Andrew et al. 2018. “SWISS-MODEL: Homology Modelling of

Protein Structures and Complexes.” Nucleic acids research 46(W1): W296–

303. http://www.ncbi.nlm.nih.gov/pubmed/29788355.

Webster, Michael W. et al. 2018. “MRNA Deadenylation Is Coupled to

Translation Rates by the Differential Activities of Ccr4-Not Nucleases.”

Molecular Cell 70(6): 1089–1100.e8.

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

Wu, Colin Chih-Chien, Boris Zinshteyn, Karen A Wehner, and Rachel Green.

2019. “High-Resolution Ribosome Profiling Defines Discrete Ribosome

Elongation States and Translational Regulation during Cellular Stress.”

Molecular cell 73(5): 959–970.e5.

http://www.ncbi.nlm.nih.gov/pubmed/30686592.

Yan, Liewei L., and Hani S. Zaher. 2019. “How Do Cells Cope with RNA Damage

and Its Consequences?” The Journal of biological chemistry 294(41):

15158–71. http://www.ncbi.nlm.nih.gov/pubmed/31439666.

149 Zhou, Wen et al. 2017. “PPR-SMR Protein SOT1 Has RNA Endonuclease

Activity.” Proceedings of the National Academy of Sciences of the United

States of America 114(8): E1554–63.

http://www.ncbi.nlm.nih.gov/pubmed/28167782.

Zurita Rendón, Olga et al. 2018. “Vms1p Is a Release Factor for the Ribosome-

Associated Quality Control Complex.” Nature communications 9(1): 2197.

http://dx.doi.org/10.1038/s41467-018-04564-3.

150 Biographical Statement

Karole N. D’Orazio

Department of Molecular Biology and Genetics Cell: (516) 640 6013 Email: [email protected] or [email protected]

Education

Johns Hopkins University School of Medicine 2014 – Present Ph.D. Candidate in Biochemistry, Cellular and Molecular Biology

Stony Brook University Master of Science in Biochemistry and Cellular Biology 2012 – 2013 Bachelor of Science in Biochemistry 2008 – 2012

Research Experience

Johns Hopkins University School of Medicine 2014 – Present Laboratory of Dr. Rachel Green - Developed reporters and successfully performed yeast genetic screens to identify novel factors involved in decay of problematic mRNAs - Characterized a novel factor that cleaves mRNAs at ribosome stall-inducing mRNA sequences and in vitro assayed endonucleolytic cleavage - Assessed an array of novel factors involved in decay of problematic mRNAs via molecular biology techniques and mass spectrometry

National Institute of Diabetes and Digestive and Kidney Diseases, NIH 2014 Laboratory of Dr. Elissa Lei - Studied the role of long noncoding RNAs in insulator complexes in Drosophila tissue

Stony Brook University 2012 – 2013 Laboratory of Dr. Ed Luk - Researched the nucleosome composition of the transcription start site in S. Cerevisiae to determine the mechanism of H2A.Z histone exchange

151 Publications (published/submitted)

D’Orazio KN, Wu CC-C, Sinha N, Loll-Krippleber R, Brown GW, Green R. (2019). The endonuclease Cue2 cleaves mRNAs at stalled ribosomes during No Go Decay. eLife, 8:e49117.

Hickey KL, Dickson K, Cogan JZ, Replogle JM, Schoof M, D’Orazio KN, Sinha NK, Frost A, Green R, Kostova KK, Weissman JS. GIGYF2 and 4EHP inhibit translation initiation of defective messenger RNAs to assist ribosome-associated quality control. In preparation. Mol. Cell. Biorxiv: doi: https://doi.org/10.1101/792994

Veltri AJ, D’Orazio KN, Green R. (2020). Make or break: the ribosome as a regulator of mRNA decay. Cell Res. doi:10.1038/s41422-019-0271-3

Awards/Honors/Fellowships

The Michael DiMaio, M.D. Research Prize for graduate student research 2019 Postbaccalaureate Intramural Research Training Award 2013 – 2014

Courses/Workshops

Cold Spring Harbor Laboratory August 2017 Yeast Genetics and Genomics - Learned fundamental aspects of yeast genetics and genomics and performed relevant experiments in a laboratory setting

Presentations Meeting/Abstracts

D’Orazio KN. (2019). The endonuclease Cue2 and the putative helicase Slh1 define parallel pathways to resolve stalled ribosomes on problematic mRNAs. Presented at Protein Synthesis and Translation Control EMBO Workshop.

D’Orazio KN. (2019). Quality control of eukaryotic mRNAs. Presented at PhD program (BCMB) recruitment events.

D’Orazio KN. (2018). A novel factor involved in mRNA cleavage at stacked ribosomes. Presented at HHMI Science Meeting. Poster.

D’Orazio KN. (2018). A novel factor involved in endonucleolytic cleavage at colliding ribosomes. Presented at the PhD program (BCMB) retreat.

152 D’Orazio KN. (2018). Identification of novel factors involved in mRNA surveillance. Presented at Cold Spring Harbor Translation Control. Poster.

Teaching and Mentoring Experience

Graduate Teaching - JHU Molecular Biology and Genomics Tutor 2018 – 2019 - JHU Molecular Biology and Genomics Lecturer 2015 – 2017 - JHU Molecular Biology and Genomics Teaching Assistant 2015 – 2017 Formed questions for exams and problem sets, graded assignments, held review sessions and office hours - Stony Brook University Undergraduate Biology Laboratory 2012 – 2013 Course Graduate Teaching Assistant - Stony Brook University Vice President of the Undergraduate 2010 – 2012 Biochemistry Society Volunteering - Mentored a student through Thread, a nonprofit 2014 – 2017 organization with the goal of supporting underachieving high school students

153