The mechanisms and function of regulated ire1-dependent decay during endoplasmic reticulum stress

The endoplasmic reticulum (ER) is a dynamic organelle that is responsible for the folding and quality control of proteins within the endomembrane system. Both physiological and pathological conditions can result in accumulation of misfolded proteins within the ER, a situation termed ER stress, which results in cell death if not alleviated. Perturbations in ER function result in activation of three ER transmembrane proteins (Ire1, Perk, and Atf6) that are primarily responsible for facilitating the unfolded protein response (UPR). Activation of the UPR initially increases ER capacity to offset the surge in misfolded protein; however, during irremediable stress, the UPR activates pro-apoptotic pathways presumably to prevent the cytotoxic consequences of secreting misfolded proteins. Ire1 is an endoribonuclease that is responsible for the unconventional splicing of an intron from the transcription factor, Xbp1. However, Ire1 is also responsible for the direct degradation of a number of mRNAs, a process termed regulated Ire1-dependent decay (RIDD). In mammals, long-term activation of Ire1 results in nonspecific cleavage of ER-localized mRNAs and subsequent cell death. However, at early time points a limited number of mRNAs are prioritized to the RIDD pathway and are degraded relatively rapidly. In the work presented here, I address the questions of (1) how specific mRNAs are prioritized for degradation? And (2) what is the function mRNA degradation during acute of ER stress? I have found that specific nucleotide sequence and structural motifs are used to target mRNAs to the RIDD pathway in both fly and mammalian cells. Furthermore, I show that inhibiting translation of these motifs is also essential for RIDD targeting. Lastly, I show Ire1-dependent effects on lysosome accumulation during ER stress; this may enhance prosurvival signaling of the UPR. These data provide insight into the mechanisms of Ire1 function as well as a model for how the RIDD pathway may function in both the prosurvival and pro-apoptotic pathways of the UPR.

THE MECHANISMS AND FUNCTION OF REGULATED IRE1-DEPENDENT DECAY DURING ENDOPLASMIC RETICULUM STRESS by Kristin A. Moore A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Biology The University of Utah May 2016 Copyright © Kristin A. Moore 2016 All Rights Reserved The U n i v e r s i t y of U tah Gr a du a t e School STATEMENT OF DISSERTATION APPROVAL The dissertation of Kristin A. Moore has been approved by the following supervisory committee members: Julie Hollien Brenda Bass Markus Babst David Gard Mark Metzstein Chair Member Member Member Member 11/20/2015 Date Approved 11/20/2015 Date Approved 11/20/2015 Date Approved 11/20/2015 Date Approved 11/20/2015 Date Approved and by Denise Dearing Chair/Dean of the Department/College/School o f ____________________Biology and by David B. Kieda, Dean of The Graduate School. ABSTRACT The endoplasmic reticulum (ER) is a dynamic organelle that is responsible for the folding and quality control of proteins within the endomembrane system. Both physiological and pathological conditions can result in accumulation of misfolded proteins within the ER, a situation termed ER stress, which results in cell death if not alleviated. Perturbations in ER function result in activation of three ER transmembrane proteins (Ire1, Perk, and Atf6) that are primarily responsible for facilitating the unfolded protein response (UPR). Activation of the UPR initially increases ER capacity to offset the surge in misfolded protein; however, during irremediable stress, the UPR activates pro-apoptotic pathways presumably to prevent the cytotoxic consequences of secreting misfolded proteins. Ire1 is an endoribonuclease that is responsible for the unconventional splicing of an intron from the transcription factor, Xbp1. However, Ire1 is also responsible for the direct degradation of a number of mRNAs, a process termed regulated Ire1-dependent decay (RIDD). In mammals, long-term activation of Ire1 results in nonspecific cleavage of ER-localized mRNAs and subsequent cell death. However, at early time points a limited number of mRNAs are prioritized to the RIDD pathway and are degraded relatively rapidly. In the work presented here, I address the questions of (1) how specific mRNAs are prioritized for degradation? And (2) what is the function mRNA degradation during acute of ER stress? I have found that specific nucleotide sequence and structural motifs are used to target mRNAs to the RIDD pathway in both fly and mammalian cells. Furthermore, I show that inhibiting translation of these motifs is also essential for RIDD targeting. Lastly, I show Ire1-dependent effects on lysosome accumulation during ER stress; this may enhance prosurvival signaling of the UPR. These data provide insight into the mechanisms of Ire1 function as well as a model for how the RIDD pathway may function in both the prosurvival and pro-apoptotic pathways of the UPR. iv To my parents, Ann and Mike, and my sister, Katy for their blind support and consistent encouragement TABLE OF CONTENTS ABSTRACT.............................................................................................................. iii LIST OF FIGURES..................................................................................................viii LIST OF TABLES.....................................................................................................x CHAPTERS 1. INTRODUCTION....................................................................................................1 Overview..........................................................................................................2 ER Function.....................................................................................................3 ER Stress and the Unfolded Protein Response........................................... 6 Ire1 Structure and Function..........................................................................10 Hac1/Xbp1 Splicing.......................................................................................13 Regulated Ire1 Dependent Decay (RIDD).................................................. 14 Models of RIDD Function in Mammals........................................................16 Concluding Remarks.....................................................................................17 References.................................................................................................... 19 2. REGULATION OF SUMO mRNA DURING ENDOPLASMIC RETICULUM STRESS........................................................................................ 26 Abstract..........................................................................................................27 Introduction....................................................................................................27 Results...........................................................................................................28 Discussion.....................................................................................................31 Materials and Methods.................................................................................32 References....................................................................................................34 3. IRE1-MEDIATED DECAY IN MAMMALIAN CELLS RELIES ON mRNA SEQUENCE, STRUCTURE, AND TRANSLATIONAL STATUS................................................................................................................35 Abstract..........................................................................................................36 Introduction........................................ ,..........................................................36 Results........................................................................................................... 37 Discussion...................................................................................................43 Materials and Methods...............................................................................45 References..................................................................................................46 Supplemental Materials..............................................................................48 4. LYSOSOME ACCUMULATION DURING ER STRESS.................................51 Introduction..................................................................................................52 Results.........................................................................................................54 Discussion...................................................................................................56 Future Directions........................................................................................ 60 Materials and Methods...............................................................................61 References..................................................................................................64 5. SUMMARY AND CONCLUSION......................................................................67 Introduction..................................................................................................68 Noncanonical RIDD Targeting in Drosophila............................................69 Mechanisms of RIDD Targeting in Mammalian Cells.............................. 70 Differences Between RIDD in FIies and Mammals..................................72 A Possible Role for RIDD of Blosl............................................................75 Concluding Remarks..................................................................................76 References..................................................................................................77 vii LIST OF FIGURES Figure 1.1 The endomembrane system.......................................................................... 4 1.2 Translation of endomembrane proteins at the ER....................................... 5 1.3 Mediators of the UPR......................................................................................9 1.4 Ire1 conformations before and after activation........................................... 12 2.1 Sumo mRNA is a non-canonical RIDD target............................................ 29 2.2 A stem loop sequence in the sumo mRNA is important for RIDD targeting.........................................................................................................30 2.3 Sumo mRNA is not strongly affected by ER stress in mammalian cells................................................................................................................ 31 2.4 RIDD of sumo is dependent on Perk.......................................................... 32 3.1 The RIDD pathway varies across mammalian cell lines............................38 3.2 An Xbp1-like stem loop is necessary for RIDD and sufficient to induce GFP mRNA degradation in mammalian cells during ER stress............................................................................................................. 39 3.3 Xbp1-like SLs are not sufficient to target endogenous mRNAs to RIDD...............................................................................................................40 3.4 Translation attenuation mediated by Perk is important for RIDD..............41 3.5 RIDD relies on the translational status of target mRNAs...........................42 3.6 Translation attenuation of Xbp1-like SLs is required for RIDD..................43 3.7 RIDD target summary and model................................................................44 3.S1 (A-B) We transfected MC3T3-E1 cells with two different siRNAs targeting the Perk transcript.........................................................................49 4.1 Lysosomes accumulate in the perinuclear region with ER stress............ 55 4.2 Lysosome accumulation during ER stress is Ire1-dependent and Xbp1 -independent.........................................................................................57 4.3 Model of Blosl-dependent lysosome accumulation during ER stress..............................................................................................................59 5.1 Models of RIDD function in Drosophila....................................................... 71 5.2 Ire1 sequence conservation in flies and mammals....................................74 ix LIST OF TABLES Table 2.1 Primers used for qPCR.................................................................................33 3.1 Primers used for qPCR and Xbp1 splicing................................................. 45 CHAPTER 1 INTRODUCTION Overview Changes to cellular environments, such as the introduction of pathogens or addition of nutrients to a system, require cells not only to process the signals they receive from their environment, but also respond to them appropriately. Many of these responses require secretion of a vast number of proteins. Often, the inability to produce the needed secretion products results in disease states such as diabetes or pathogen infection. The endoplasmic reticulum (ER) is the entry point for proteins of the secretory pathway, and is responsible for the folding, processing, and quality control of secreted proteins. It is also a highly dynamic organelle that changes its capacity to meet the folding needs of the cell. These changes are often induced through accumulation of unfolded proteins within the ER, which overwhelm the capacity of the organelle to fold proteins efficiently, a situation termed ER stress. Markers of ER stress have been observed in various pathogenic states including cancer and neurological disease (Oakes and Papa, 2015), as well as during physiological processes such as differentiation (Iwakoshi et al., 2003). Thus the ability to understand and manipulate how cells respond to ER stress is thought to provide therapeutic insights into both developmental disorders as well as many diseases. I sought to determine the mechanism and function of an mRNA degradation pathway induced by ER stress. I first characterize the requirements for mRNA targeting to this pathway in cells from both flies and mammals. I also describe a role for a protein previously thought to function in an entirely separate pathway. Finally, I provide evidence in support of a functional model for how 2 3 degradation of a specific mRNA may contribute to the general response to ER stress. ER Function Spatial and functional compartmentalization of metabolic and biosynthetic processes is one of the primary distinctions of eukaryotic cells. The endomembrane system consists of the membrane bound organelles that create these compartments. These organelles and vesicles are responsible for transporting proteins from the ER to the plasma membrane (the secretory pathway), as well as for protein degradation and resource recycling, and are essential for cell function (Figure 1.1). In mammalian cells one-third of all proteins are predicted to transit through or remain within the endomembrane system, and in specialized secretory cell types, such as antibody producing plasma cells, hundreds of thousands of proteins are secreted each minute (King and Corley, 1989); therefore, efficiency and maintenance of this system is an essential task. The ER is a membrane bound organelle responsible for the folding, processing, and quality control of the vast majority of proteins that are part of the endomembrane system. It is also the primary source of lipids for almost all membrane bound organelles within the cell. Thus, it plays an important role in both the structure and function of all other membrane bound organelles. mRNAs that encode endomembrane proteins are translated on ribosomes that dock on the cytosolic side of the ER and translate proteins into the lumen of the ER through the Sec61 channel or translocon (Figure 1.2). These nascent protein chains are then folded into their native conformations with the assistance of ER- 4 Extra cellular space Figure 1.1 The endomembrane system. The endomembrane system is comprised of the nuclear envelope, ER, Golgi apparatus, lysosomes, vesicles, peroxisomes (not pictured), endosomes, and the plasma membrane. Proteins that either remain within the endomembrane system or are secreted out of the cell are translated on ribosomes at the rough ER. Mature proteins exit the ER in vesicles, the majority of which then fuse with the Golgi apparatus, where proteins are further sorted and modified. Proteins leave the Golgi in vesicles destined for other organelles of the endomembrane system. Proteins of the endomembrane system are often cycled through the endosome where damaged proteins are recognized and sent to the lysosome for degradation. (Arrows denote a possible pathway of a transmembrane protein that functions at the plasma membrane.) 5 Figure 1.2 Translation of endomembrane proteins at the ER. Proteins containing ER signal sequences or transmembrane domains are co-translationally localized to the ER. Ribosomes dock on the cytosolic side the ER and nascent protein chains are translated through the translocon into the lumen of the ER. During this process the mRNA encoding the protein is stably associated with the cytoplasmic side of the ER through its interactions with the ribosome. This association may be transient or long-term depending on the number of ER-localized ribosomes associated with the mRNA. resident chaperones. Immature proteins within the ER are also modified through additions such as glycans and disulfide bonds. Once proteins have reached their native conformation, they are sorted and transported out of the ER. To avoid the cytotoxic effects of releasing misfolded proteins, such as aggregation, ER chaperones recognize and sequester proteins that remain terminally misfolded. These proteins are exported back to the cytosol via the ER-associated degradation (ERAD) pathway, where they are degraded in a proteasome-dependent manner (Ruggiano et al., 2014). ER Stress and the Unfolded Protein Response The amount of incoming, unfolded proteins is normally balanced with the capacity of the ER to fold and modify them; however, both physiological and pathological events can result in dramatic and rapid increases in the load of proteins entering the ER. ER stress occurs when the number of misfolded proteins within the lumen of the ER overwhelms the ability of the organelle to fold and modify them efficiently. Chemicals that disrupt folding or ER function as well as mutations within hard to fold proteins can also induce ER stress. In metazoans, there are three sensors of ER stress: Ire1, Perk, and Atf6. In mammals, there are 2 homologs of both Ire1 and Atf6, noted as -alpha and - beta. These sensors are ER transmembrane proteins that transmit information about the environment of the ER lumen to the cytosol, resulting in various mechanisms that alleviate the folding burden on the ER (Moore and Hollien, 2012). Collectively, these pathways are termed the unfolded protein response (UPR). While the UPR is often considered a response to pathological 6 accumulation of misfolded proteins, Ire1-alpha and double Atf6-alpha and -beta knockout animals result in embryonic lethality (Urano et al., 2000; Yamamoto et al., 2007). Perk knockout animals are viable, but display severe bone and pancreatic defects, indicating a role for the UPR in development (Zhang et al., 2002). Furthermore, the differences in knockout phenotypes suggest that different branches of the UPR have varied levels of importance depending on cell type and protein-folding burden. Ire1 is an endoribonuclease that is conserved in eukaryotes. While yeast and flies possess a single Ire1 homolog, mammals possess two homologs of Ire1. Ire1a is ubiquitously expressed (Tirasophon et al., 1998), while Ire1p is confined to intestinal cells (Bertolotti et al., 2001). Plants also have two homologs of Ire1, Ire1A and Ire1B (Koizumi et al., 2001; Noh et al., 2002). Upon activation by ER stress, Ire1 oligomerizes resulting in trans-autophosphorylation and activation of its RNase domain (Kimata et al., 2007; Aragon et al., 2009; Korennykh et al., 2009; Li et al., 2010). In Saccharomyces cerevisiae, active Ire1 unconventionally splices an intron from the Hac1 mRNA in the cytosol (Cox and Walter, 1996; Mori et al., 1996), resulting in de-repression of Hac1 translation and a frameshift in the resulting transcript leads to activation of the Hac1 protein (Mori et al., 2000; Ruegsegger et al., 2001). The Hac1-spliced (Hac1s) protein is an active transcription factor that induces large transcriptional changes that increase ER function and capacity (Travers et al., 2000). In metazoans, a similar pathway is initiated when active Ire1 splices an intron from Xbp1, the Hac1 homolog (Shen et al., 2001; Yoshida et al., 2001; Calfon et al., 2002). In both yeast and mammalian cells, Ire1 also functions in activities beyond Hac1/Xbp1 splicing, 7 such as regulated Ire1 dependent degradation (RIDD - see following sections) (Hollien and Weissman, 2006; Han et al., 2009; Hollien et al., 2009; Kimmig et al., 2012; Tam et al., 2014). Additionally, in mammalian cells, active Ire1 interacts with the C-Jun N-terminal kinase (Jnk) activating proteins Traf2, Ask1, and Aip-1 (Urano et al., 2000; Nishitoh et al., 2002; Luo et al., 2008). These interactions result in activation of Jnk and its downstream signaling (Urano et al., 2000). Beyond Ire1, metazoans also rely on Perk and Atf6 for complete activation of the UPR (Moore and Hollien, 2012). Perk is an eIF2a kinase that dimerizes upon induction of ER stress, leading to trans-autophosphorylation and activation of its kinase domain (Korennykh and Walter, 2012). Once activated, Perk phosphorylates eIF2a, resulting in a decrease in general protein production through diminished regeneration of the active tRNA-met complex (Shi et al., 1998; Harding et al., 1999; Jackson et al., 2010). Interestingly, a subset of mRNAs with upstream open reading frames (uORFs), such as Atf4, is specifically translated under these conditions. Atf4 is another transcription factor that regulates many genes involved in the secretory system (Harding et al., 2000a). The third UPR transducer is Atf6 (Haze et al., 1999; Wang et al., 2000; Yamamoto et al., 2010). Unlike Ire1 and Perk, under unstressed conditions Atf6 forms homodimers and is thought to depolymerize upon induction of ER stress (Nadanaka et al., 2007). It then travels to the Golgi apparatus where site 1 and site 2 proteases cleave its cytosolic domain from the luminal domain (Ye et al., 2000). The cytosolic portion of Atf6 is a third transcription factor, which works in conjunction with Xbp1 and Atf4 to increase the capacity of the ER (Figure 1.3) (Yamamoto et al., 2007). 8 9 Figure 1.3 Mediators of the UPR. Summary of the three main signaling branches of the UPR. The sensors Ire1, Perk, and Atf6 detect changes in endoplasmic reticulum (ER) homeostasis and activate bZip transcription factors through unconventional mRNA splicing, translational upregulation, and intermembrane proteolysis, respectively. This transcriptional regulation, in parallel with Ire1- mediated mRNA decay and Perk-mediated translational regulation, restores and enhances ER function. Abbreviation: BiP, binding immunoglobulin protein. Reprinted with permission of Annual Review of Genetics, 46, Moore, K. A. & Hollien, J., The Unfolded Protein Response in Secretory Cell Function, 165-83, Copyright 2012. A large amount of energy goes into enhancing ER function and capacity; however, unresolvable ER stress shifts UPR signaling towards apoptosis (Shore et al., 2011). Increases in pro-apoptotic signaling through Jnk, and transcription of genes encoding pro-apoptotic proteins such as Puma, Noxa, Bim, and Chop all result from activation of various UPR mediators (Zinszner et al., 1998; Li et al., 2006; Puthalakath et al., 2007; Upton et al., 2008). These pathways are thought to converge on the intrinsic mitochondrial apoptotic pathways, and under prolonged ER stress, cleavage of apoptotic mediators such as Caspase-3 and -7 is observed (Gupta et al., 2010). While pro-apoptotic signaling is induced by all three of the UPR mediators, loss of any branch of UPR signaling results in increased rate of cell death indicating that the primary function of the UPR is to alleviate the ER stress, and that apoptotic signaling is activated only when ER stress is irremediable (Harding et al., 2000b; Lee et al., 2003; Yamamoto et al., 2010). However, the factors that signal the transition from prosurvival to pro-apoptotic processes are not well understood. Ire1 Structure and Function The research described in this dissertation primarily focuses on Ire1 and its ability to cleave mRNAs. The N-terminus of Ire1 resides within the ER lumen and is responsible for sensing ER stress, most likely through direct interaction with unfolded proteins (Gardner and Walter, 2011). The luminal domain is connected to its cytosolic kinase and RNase domains via a Type I transmembrane domain and a flexible linker domain. The kinase domain shares homology with the well-studied CDK2 serine/threonine class of kinases, while the 10 RNase domain appears to be unique (Korennykh and Walter, 2012). Ire1 resides mostly as a monomer in a complex with the Hsp70 chaperone, BiP, during unstressed conditions (Bertolotti et al., 2000). BiP binds to a juxta-transmembrane region of the Ire1 luminal domain that is not required for activation of Ire1 (Kimata et al., 2004). Upon accumulation of misfolded proteins, BiP is titrated off the luminal domain of Ire1, and there is increasing evidence that misfolded proteins bind directly to a groove within the Ire1 luminal domain resulting in dimerization (Credle et al., 2005; Gardner and Walter, 2011). Initial Ire1 dimers form in the face-to-face configuration, which results in exchange of kinase activation loops, while the RNase domains remain distant from each other (Figure 1.4B) (Joshi et al., 2015). In current models of Ire1 activation the initial "face-to-face" dimer is considered an early intermediate stage in Ire1 activation, while the back-to-back dimer is the active conformation of Ire1 (Walter and Ron, 2011; Maly and Papa, 2014). The face-to-face dimer allows for trans-autophosphorylation of the kinase activation loops, resulting in a conformational change into a new back-to-back dimer (Figure 1.4C) (Lee et al., 2008; Korennykh et al., 2011a). In the back-to-back conformation the kinase activation loops face away from each other while the RNase domains form a protein interface allowing for activation (Lee et al., 2008; Joshi et al., 2015). Back-to-back dimers are then thought to oligomerize into clusters of upwards of 8 monomers (Aragon et al., 2009; Li et al., 2010). Increases in oligomerization of Ire1 result in stabilization of the Helix-loop element (HLE) of the RNase domain (Korennykh et al., 2009), and in vitro studies have shown that increasing oligomerization correlates with increased RNase 11 12 A Ire1 monomer Luminal domain Transmembrane and linker domain Kinase domain Kinase activation loop RNase domain B Ire1 face-to-face dimer (inactive for RNA cleavage) C Ire1 back-to-back dimer (active for RNA cleavage) Active RNase domain Figure 1.4 Ire1 conformations before and after activation. (A) A representation of an Ire1 monomer. (B) The kinase and RNase domains of an Ire1 dimer in the face-to-face conformation. The activation loops of the kinase domains are predicted to interact each other resulting in ATP binding and transphosphorylation. The RNase domains do not interact in this conformation; consequently RNase activity is low. (C) The kinase and RNase domains of an Ire1 dimer in the back-to-back conformation. Binding to the ATP binding pocket leads to a conformational change in the Ire1 dimer resulting in the back to back configuration in which the RNase domains interact forming a functional RNA binding pocket and high RNase activity. This conformation also allows for higher order oligomerization through interactions with the newly accessible interfaces. activity, indicating that high order oligomers of active Ire1 are important for RNase activity (Korennykh et al., 2011a). Interestingly, while over 20 phosphorylated residues have been observed on active Ire1, the functionality of the kinase domain is unnecessary for RNase activation. Rather, nucleotide binding to the kinase pocket results in the conformational changes that allow for RNase activation (Korennykh et al., 2009). Incidentally many Ire1 kinase inhibiters strongly induce its RNase activity (Joshi et al., 2015). The catalytic activity of the RNase domain relies on acid-base catalysis. Proton transfer from a histidine (residue 1061 in yeast, 910 in humans), which acts as a general acid and a tyrosine (residue 1043 in yeast, 892 in humans), which acts as a general base, results in sequence-specific catalytic cleavage of mRNA (Dong et al., 2001; Korennykh et al., 2011b). Based on structural data, it appears that, although two active catalytic sites exist within the Ire1 back-to-back dimer, the size of the RNA binding pocket is too small to allow for cleavage of more than one RNA substrate at a time (Korennykh et al., 2009). However, in order for the RNA to dock appropriately in the RNase active site stabilization derived from HLE domains of both Ire1 molecules within the dimer is required. Hac1/Xbp1 Splicing Cleavage of Hac1/Xbp1 is extremely sequence and structure specific. Cleavage occurs 3' of a guanosine nucleotide at the third position within the loops of a conserved dual stem loop structure within the coding region of the transcript (Sidrauski and Walter, 1997; Calfon et al., 2002). Furthermore, mutations that disrupt base pairing within the stems, or either loop, also result in 13 cleavage inhibition. The cleavage results in a 2' 3' cyclic phosphate group, a 5' hydroxyl, and loss of an intron from the transcript (Gonzalez et al., 1999). The resulting 5' and 3' ends of the cleavage site are then ligated together by tRNA ligase in yeast, and unknown and possibly redundant ligases in metazoans. In addition to sequence and structure specificity of the stem loops, the secondary structure of the mRNA surrounding the dual stem loop region is also important for Ire1 cleavage (Sidrauski et al., 1996). Mutations that result in loss of secondary structure upstream and downstream of the stem loop structures also inhibit cleavage in mammalian cells (Calfon et al., 2002). Regulated Ire1 Dependent Decay (RIDD) RIDD was originally discovered in Drosophila S2 cells, where a large number of ER-localized mRNAs are degraded in an Ire1-dependent, Xbp1- independent manner (Hollien and Weissman, 2006). While the biochemical mechanisms of RNA cleavage by the RIDD pathway have not been determined, similar to Xbp1 splicing, RIDD is inhibited by mutations that block access to the Ire1 RNase active site and compounds that target the nuclease activity of Ire1 (Cross et al., 2012). To date, no mutations have been found that distinguish between the splicing reaction of Xbp1 and RIDD target cleavage in mammalian systems; however, evidence of differences between the two pathways do exist. For example, Xbp1 splicing happens on a much faster time scale than RIDD, with ~90% of Xbp1 transcripts being spliced within 10 minutes after the induction of ER stress, while maximum degradation through RIDD often takes several hours under the same conditions (unpublished observations). Furthermore, artificial 14 activation of an Ire1 mutant that binds to an ATP-analog results in only Xbp1 splicing. However, this mutant was fully competent to perform RIDD upon addition of ER stress and the ATP-analog, indicating that additional requirements beyond activation of Ire1's RNase domain must exist for RIDD (Han et al., 2009; Hollien et al., 2009). Work to elucidate RIDD mechanisms and function initially focused on the cell biology of the pathway in Drosophila cells. Gaddam et al. (2013) showed that ER-localization is both necessary and sufficient to target an mRNA to RIDD in Drosophila cells. Because the vast majority of mRNAs that encode proteins of the transmembrane system reside at the ER, they are disproportionately sensitive to RIDD (Hollien and Weissman, 2006). These results led to the hypothesis that RIDD may result in an initial downregulation of proteins entering the ER, which would then sensitize the ER to the major increases in transcription that support ER function and capacity. Interestingly, Smt3, the SUMO homolog in flies, is also degraded through RIDD even though its mRNA is not ER-localized, indicating that degradation of specific mRNAs may also have functional consequences for the cell. Furthermore, degradation of a specific mRNA (FatP) has been shown to be essential for normal fly eye development confirming that in addition to a possible broad function of mRNA degradation, loss of specific mRNAs is also a functional consequence of RIDD in flies (Coelho et al., 2013). RIDD has also been observed both in vivo and in vitro in mammalian systems. mRNA degradation through RIDD has been observed with chemical induction of ER stress, Ire1 overexpression, and Ire1 hyperactivation resulting from depletion of Xbp1 (Han et al., 2009; Hollien et al., 2009; Hur et al., 2012). 15 When Ire1 is either overexpressed or hyperactivated for extended periods of time, mammalian Ire1 cleaves numerous mRNAs that encode proteins of the secretory system, similar to patterns observed in Drosophila cells. However, it is clear that during acute instances of ER stress induction (i.e., chemical stressors) RIDD is much more specific than in S2 cells with only a subset of ER-localized mRNAs being targeted to the RIDD pathway (Hollien et al., 2009; So et al., 2012), indicating that specific mRNAs are prioritized to RIDD even though mammalian Ire1 is capable of cleaving mRNAs with less specificity. Additionally, because of the variability between mRNAs expressed in different systems and the massive remodeling of transcription that occurs during the UPR, compiling a comprehensive list of RIDD targets has remained challenging. However, of the confirmed RIDD targets in mammalian systems some similarities have been observed. Unlike S2 cells, in which unique Ire1 cleavage sequences do not seem to exist (Hollien and Weissman, 2006), mammalian RIDD targets contain Xbp1- like stem loops (Gaddam et al., 2013). Mutation of the predicted cleavage site within the loop sequence of two transcripts inhibited cleavage by recombinant human Ire1 in vitro, indicating that cleavage by mammalian Ire1 may be more dependent on target sequence than cleavage by Drosophila Ire1 (Hur et al., 2012; So et al., 2012). Models of RIDD Function in Mammals In mammals, there are two nonmutually exclusive models for RIDD function. Either, nonspecific degradation of numerous RNAs, or specific degradation of a few mRNAs affects the cell's ability to respond appropriately to 16 stressful situations. Furthermore, localized degradation of mRNAs, possibly at contact sites between the ER and mitochondria where Ire1 has been observed (Mori et al., 2013), results in reduction of proteins that are locally important. In this scenario, the local effect of RIDD may be large, but the overall changes in mRNA levels would be minimal and go largely unobserved when measuring total mRNA levels. These alternative models of RIDD function arise from the observation that Ire1 can degrade broad classes of mRNAs as observed with Ire1 overexpression or hyperactivation in mammalian cells, but under acute instances of stress, only a relatively small subset of mRNAs are degraded to a measurable degree (Han et al., 2009; Hollien et al., 2009; So et al., 2012). Furthermore, it has not been determined whether the RIDD pathway contributes to cell survival or cell death. Evidence from a variety of techniques and systems exists to support both possibilities (Han et al., 2009; Cross et al., 2012; So et al., 2012; Upton et al., 2012). In an effort to better understand the function of the RIDD pathway we have taken a mechanistic approach to determine how mRNAs are targeted to the RIDD pathway, which will guide our future studies towards finding the biologically relevant targets of the RIDD pathway and its physiological function. Concluding Remarks The endoplasmic reticulum is a dynamic organelle that is remodeled to fit the needs of the cell. To ensure efficient protein processing the environment of the ER is under constant surveillance. The mediators of the UPR (Ire1, Perk, and Atf6) counteract perturbations to this system by changing the translational and 17 transcriptional landscape of the cell. These changes result in increases in ER size and folding ability. Modification of ER capacity is essential for responding to pathological insults as well as the general secretory needs of different cell types; however, under conditions of irremediable stress the mediators of the UPR induce apoptotic pathways to avoid the cytotoxic effects of misfolded protein release. The factors that tip the balance from cell survival to cell death signaling during ER stress are not well understood. The RIDD pathway has been hypothesized to play a role in both pro­survival and prodeath pathways; however, its physiological function remains obscure. I have used molecular biology techniques and chemical induction of ER stress to address the question of how mRNAs are prioritized to the RIDD pathway and whether their specific degradation has physiological effects on the cell. Here I describe the requirements for RIDD targeting of both a noncanonical RIDD target in flies as well as the mechanisms employed by the RIDD pathway in mammalian cells (Chapters 2 and 3). Based on the information generated through these studies I propose a model for how RIDD signaling may function in both prosurvival and prodeath pathways depending on the type and duration of stress induced. Finally, based on our understanding of RIDD targeting I have hypothesized a role for degradation of a single mRNA, resulting in a more efficient response to ER stress (Chapter 4). 18 19 References Aragon, T., van Anken, E., Pincus, D., Serafimova, I.M., Korennykh, A.V., Rubio, C.A., and Walter, P. (2009). Messenger RNA targeting to endoplasmic reticulum stress signalling sites. Nature 457, 736-740. Bertolotti, A., Wang, X., Novoa, I., Jungreis, R., Schlessinger, K., Cho, J.H., West, A.B., and Ron, D. (2001). Increased sensitivity to dextran sodium sulfate colitis in IRE1beta-deficient mice. J Clin Invest 107, 585-593. Bertolotti, A., Zhang, Y., Hendershot, L.M., Harding, H.P., and Ron, D. (2000). Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat Cell Biol 2, 326-332. Calfon, M., Zeng, H., Urano, F., Till, J.H., Hubbard, S.R., Harding, H.P., Clark, S.G., and Ron, D. (2002). IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA. Nature 415, 92-96. Coelho, D.S., Cairrao, F., Zeng, X., Pires, E., Coelho, A.V., Ron, D., Ryoo, H.D., and Domingos, P.M. (2013). Xbp1-independent Ire1 signaling is required for photoreceptor differentiation and rhabdomere morphogenesis in Drosophila. Cell Rep 5, 791-801. Cox, J.S., and Walter, P. (1996). A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell 87, 391-404. Credle, J.J., Finer-Moore, J.S., Papa, F.R., Stroud, R.M., and Walter, P. (2005). On the mechanism of sensing unfolded protein in the endoplasmic reticulum. Proc Natl Acad Sci U S A 102, 18773-18784. Cross, B.C., Bond, P.J., Sadowski, P.G., Jha, B.K., Zak, J., Goodman, J.M., Silverman, R.H., Neubert, T.A., Baxendale, I.R., Ron, D., and Harding, H.P. (2012). The molecular basis for selective inhibition of unconventional mRNA splicing by an IRE1-binding small molecule. Proc Natl Acad Sci U S A 109, E869- 878. Dong, B., Niwa, M., Walter, P., and Silverman, R.H. (2001). Basis for regulated RNA cleavage by functional analysis of RNase L and Ire1p. Rna 7, 361-373. Gaddam, D., Stevens, N., and Hollien, J. (2013). Comparison of mRNA localization and regulation during endoplasmic reticulum stress in Drosophila cells. Mol Biol Cell 24, 14-20. Gardner, B.M., and Walter, P. (2011). Unfolded proteins are Ire1-activating ligands that directly induce the unfolded protein response. Science 333, 1891­1894. 20 Gonzalez, T.N., Sidrauski, C., Dorfler, S., and Walter, P. (1999). Mechanism of non-spliceosomal mRNA splicing in the unfolded protein response pathway. Embo j 18, 3119-3132. Gupta, S., Cuffe, L., Szegezdi, E., Logue, S.E., Neary, C., Healy, S., and Samali, A. (2010). Mechanisms of ER stress-mediated mitochondrial membrane permeabilization. Int J Cell Biol 2010, 170215. Han, D., Lerner, A.G., Vande Walle, L., Upton, J.P., Xu, W., Hagen, A., Backes, B.J., Oakes, S.A., and Papa, F.R. (2009). IRE1alpha kinase activation modes control alternate endoribonuclease outputs to determine divergent cell fates. Cell 138, 562-575. Harding, H.P., Novoa, I., Zhang, Y., Zeng, H., Wek, R., Schapira, M., and Ron, D. (2000a). Regulated translation initiation controls stress-induced gene expression in mammalian cells. Mol Cell 6, 1099-1108. Harding, H.P., Zhang, Y., Bertolotti, A., Zeng, H., and Ron, D. (2000b). Perk is essential for translational regulation and cell survival during the unfolded protein response. Mol Cell 5, 897-904. Harding, H.P., Zhang, Y., and Ron, D. (1999). Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase. Nature 397, 271-274. Haze, K., Yoshida, H., Yanagi, H., Yura, T., and Mori, K. (1999). Mammalian transcription factor ATF6 is synthesized as a transmembrane protein and activated by proteolysis in response to endoplasmic reticulum stress. Mol Biol Cell 10, 3787-3799. Hollien, J., Lin, J.H., Li, H., Stevens, N., Walter, P., and Weissman, J.S. (2009). Regulated Ire1-dependent decay of messenger RNAs in mammalian cells. J Cell Biol 186, 323-331. Hollien, J., and Weissman, J.S. (2006). Decay of endoplasmic reticulum-localized mRNAs during the unfolded protein response. Science 313, 104-107. Hur, K.Y., So, J.S., Ruda, V., Frank-Kamenetsky, M., Fitzgerald, K., Koteliansky, V., Iwawaki, T., Glimcher, L.H., and Lee, A.H. (2012). IRE1alpha activation protects mice against acetaminophen-induced hepatotoxicity. J Exp Med 209, 307-318. Iwakoshi, N.N., Lee, A.H., Vallabhajosyula, P., Otipoby, K.L., Rajewsky, K., and Glimcher, L.H. (2003). Plasma cell differentiation and the unfolded protein response intersect at the transcription factor XBP-1. Nat Immunol 4, 321-329. 21 Jackson, R.J., Hellen, C.U., and Pestova, T.V. (2010). The mechanism of eukaryotic translation initiation and principles of its regulation. Nat Rev Mol Cell Biol 11, 113-127. Joshi, A., Newbatt, Y., McAndrew, P.C., Stubbs, M., Burke, R., Richards, M.W., Bhatia, C., Caldwell, J.J., McHardy, T., Collins, I., and Bayliss, R. (2015). Molecular mechanisms of human IRE1 activation through dimerization and ligand binding. Oncotarget 6, 13019-13035. Kimata, Y., Oikawa, D., Shimizu, Y., Ishiwata-Kimata, Y., and Kohno, K. (2004). A role for BiP as an adjustor for the endoplasmic reticulum stress-sensing protein Ire1. J Cell Biol 167, 445-456. Kimata, Y., Ishiwata-Kimata, Y., Ito, T., Hirata, A., Suzuki, T., Oikawa, D., Takeuchi, M., and Kohno, K. (2007). Two regulatory steps of ER-stress sensor Ire1 involving its cluster formation and interaction with unfolded proteins. J Cell Biol 179, 75-86. Kimmig, P., Diaz, M., Zheng, J., Williams, C.C., Lang, A., Aragon, T., Li, H., and Walter, P. (2012). The unfolded protein response in fission yeast modulates stability of select mRNAs to maintain protein homeostasis. In: Elife, vol. 1, United States, e00048. King, L.B., and Corley, R.B. (1989). Characterization of a presecretory phase in B-cell differentiation. Proc Natl Acad Sci U S A 86, 2814-2818. Koizumi, N., Martinez, I.M., Kimata, Y., Kohno, K., Sano, H., and Chrispeels, M.J. (2001). Molecular characterization of two Arabidopsis Ire1 homologs, endoplasmic reticulum-located transmembrane protein kinases. Plant Physiol 127, 949-962. Korennykh, A., and Walter, P. (2012). Structural basis of the unfolded protein response. Annu Rev Cell Dev Biol 28, 251-277. Korennykh, A.V., Egea, P.F., Korostelev, A.A., Finer-Moore, J., Stroud, R.M., Zhang, C., Shokat, K.M., and Walter, P. (2011a). Cofactor-mediated conformational control in the bifunctional kinase/RNase Ire1. BMC Biol 9, 48. Korennykh, A.V., Egea, P.F., Korostelev, A.A., Finer-Moore, J., Zhang, C., Shokat, K.M., Stroud, R.M., and Walter, P. (2009). The unfolded protein response signals through high-order assembly of Ire1. Nature 457, 687-693. Korennykh, A.V., Korostelev, A.A., Egea, P.F., Finer-Moore, J., Stroud, R.M., Zhang, C., Shokat, K.M., and Walter, P. (2011b). Structural and functional basis for RNA cleavage by Ire1. BMC Biol 9, 47. 22 Lee, A.H., Iwakoshi, N.N., Anderson, K.C., and Glimcher, L.H. (2003). Proteasome inhibitors disrupt the unfolded protein response in myeloma cells. Proc Natl Acad Sci U S A 100, 9946-9951. Lee, K.P., Dey, M., Neculai, D., Cao, C., Dever, T.E., and Sicheri, F. (2008). Structure of the dual enzyme Ire1 reveals the basis for catalysis and regulation in nonconventional RNA splicing. Cell 132, 89-100. Li, H., Korennykh, A.V., Behrman, S.L., and Walter, P. (2010). Mammalian endoplasmic reticulum stress sensor IRE1 signals by dynamic clustering. Proc Natl Acad Sci U S A 107, 16113-16118. Li, J., Lee, B., and Lee, A.S. (2006). Endoplasmic reticulum stress-induced apoptosis: multiple pathways and activation of p53-up-regulated modulator of apoptosis (PUMA) and NOXA by p53. J Biol Chem 281, 7260-7270. Luo, D., He, Y., Zhang, H., Yu, L., Chen, H., Xu, Z., Tang, S., Urano, F., and Min, W. (2008). AIP1 is critical in transducing IRE1-mediated endoplasmic reticulum stress response. J Biol Chem 283, 11905-11912. Maly, D.J., and Papa, F.R. (2014). Druggable sensors of the unfolded protein response. Nat Chem Biol 10, 892-901. Moore, K.A., and Hollien, J. (2012). The unfolded protein response in secretory cell function. Annu Rev Genet 46, 165-183. Mori, K., Kawahara, T., Yoshida, H., Yanagi, H., and Yura, T. (1996). Signalling from endoplasmic reticulum to nucleus: transcription factor with a basic-leucine zipper motif is required for the unfolded protein-response pathway. Genes Cells 1, 803-817. Mori, K., Ogawa, N., Kawahara, T., Yanagi, H., and Yura, T. (2000). mRNA splicing-mediated C-terminal replacement of transcription factor Hac1p is required for efficient activation of the unfolded protein response. Proc Natl Acad Sci U S A 97, 4660-4665. Mori, T., Hayashi, T., Hayashi, E., and Su, T.P. (2013). Sigma-1 receptor chaperone at the ER-mitochondrion interface mediates the mitochondrion-ER-nucleus signaling for cellular survival. PLoS One 8, e76941. Nadanaka, S., Okada, T., Yoshida, H., and Mori, K. (2007). Role of disulfide bridges formed in the luminal domain of ATF6 in sensing endoplasmic reticulum stress. Mol Cell Biol 27, 1027-1043. Nishitoh, H., Matsuzawa, A., Tobiume, K., Saegusa, K., Takeda, K., Inoue, K., Hori, S., Kakizuka, A., and Ichijo, H. (2002). ASK1 is essential for endoplasmic 23 reticulum stress-induced neuronal cell death triggered by expanded polyglutamine repeats. Genes Dev 16, 1345-1355. Noh, S.J., Kwon, C.S., and Chung, W.I. (2002). Characterization of two homologs of Ire1p, a kinase/endoribonuclease in yeast, in Arabidopsis thaliana. Biochim Biophys Acta 1575, 130-134. Oakes, S.A., and Papa, F.R. (2015). The role of endoplasmic reticulum stress in human pathology. Annu Rev Pathol 10, 173-194. Puthalakath, H., O'Reilly, L.A., Gunn, P., Lee, L., Kelly, P.N., Huntington, N.D., Hughes, P.D., Michalak, E.M., McKimm-Breschkin, J., Motoyama, N., Gotoh, T., Akira, S., Bouillet, P., and Strasser, A. (2007). ER stress triggers apoptosis by activating BH3-only protein Bim. Cell 129, 1337-1349. Ruegsegger, U., Leber, J.H., and Walter, P. (2001). Block of HAC1 mRNA translation by long-range base pairing is released by cytoplasmic splicing upon induction of the unfolded protein response. Cell 107, 103-114. Ruggiano, A., Foresti, O., and Carvalho, P. (2014). Quality control: ER-associated degradation: protein quality control and beyond. J Cell Biol 204, 869­879. Shen, X., Ellis, R.E., Lee, K., Liu, C.Y., Yang, K., Solomon, A., Yoshida, H., Morimoto, R., Kurnit, D.M., Mori, K., and Kaufman, R.J. (2001). Complementary signaling pathways regulate the unfolded protein response and are required for C. elegans development. Cell 107, 893-903. Shi, Y., Vattem, K.M., Sood, R., An, J., Liang, J., Stramm, L., and Wek, R.C. (1998). Identification and characterization of pancreatic eukaryotic initiation factor 2 alpha-subunit kinase, PEK, involved in translational control. Mol Cell Biol 18, 7499-7509. Shore, G.C., Papa, F.R., and Oakes, S.A. (2011). Signaling cell death from the endoplasmic reticulum stress response. Curr Opin Cell Biol 23, 143-149. Sidrauski, C., Cox, J.S., and Walter, P. (1996). tRNA ligase is required for regulated mRNA splicing in the unfolded protein response. Cell 87, 405-413. Sidrauski, C., and Walter, P. (1997). The transmembrane kinase Ire1p is a site-specific endonuclease that initiates mRNA splicing in the unfolded protein response. Cell 90, 1031-1039. So, J.S., Hur, K.Y., Tarrio, M., Ruda, V., Frank-Kamenetsky, M., Fitzgerald, K., Koteliansky, V., Lichtman, A.H., Iwawaki, T., Glimcher, L.H., and Lee, A.H. (2012). Silencing of lipid metabolism genes through IRE1alpha-mediated mRNA decay lowers plasma lipids in mice. Cell Metab 16, 487-499. 24 Tam, A.B., Koong, A.C., and Niwa, M. (2014). Ire1 has distinct catalytic mechanisms for XBP1/HAC1 splicing and RIDD. Cell Rep 9, 850-858. Tirasophon, W., Welihinda, A.A., and Kaufman, R.J. (1998). A stress response pathway from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein kinase/endoribonuclease (Ire1p) in mammalian cells. Genes Dev 12, 1812-1824. Travers, K.J., Patil, C.K., Wodicka, L., Lockhart, D.J., Weissman, J.S., and Walter, P. (2000). Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell 101, 249-258. Upton, J.P., Austgen, K., Nishino, M., Coakley, K.M., Hagen, A., Han, D., Papa, F.R., and Oakes, S.A. (2008). Caspase-2 cleavage of BID is a critical apoptotic signal downstream of endoplasmic reticulum stress. Mol Cell Biol 28, 3943-3951. Upton, J.P., Wang, L., Han, D., Wang, E.S., Huskey, N.E., Lim, L., Truitt, M., McManus, M.T., Ruggero, D., Goga, A., Papa, F.R., and Oakes, S.A. (2012). IRE1a cleaves select microRNAs during eR stress to derepress translation of proapoptotic Caspase-2. Science 338, 818-822. Urano, F., Wang, X., Bertolotti, A., Zhang, Y., Chung, P., Harding, H.P., and Ron, D. (2000). Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 287, 664-666. Walter, P., and Ron, D. (2011). The unfolded protein response: from stress pathway to homeostatic regulation. Science 334, 1081-1086. Wang, Y., Shen, J., Arenzana, N., Tirasophon, W., Kaufman, R.J., and Prywes, R. (2000). Activation of ATF6 and an AtF6 DNA binding site by the endoplasmic reticulum stress response. J Biol Chem 275, 27013-27020. Yamamoto, K., Sato, T., Matsui, T., Sato, M., Okada, T., Yoshida, H., Harada, A., and Mori, K. (2007). Transcriptional induction of mammalian ER quality control proteins is mediated by single or combined action of ATF6alpha and XBP1. Dev Cell 13, 365-376. Yamamoto, K., Takahara, K., Oyadomari, S., Okada, T., Sato, T., Harada, A., and Mori, K. (2010). Induction of liver steatosis and lipid droplet formation in ATF6alpha-knockout mice burdened with pharmacological endoplasmic reticulum stress. Mol Biol Cell 21, 2975-2986. Ye, J., Rawson, R.B., Komuro, R., Chen, X., Dave, U.P., Prywes, R., Brown, M.S., and Goldstein, J.L. (2000). ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Mol Cell 6, 1355­1364. 25 Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., and Mori, K. (2001). XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107, 881-891. Zhang, P., McGrath, B., Li, S., Frank, A., Zambito, F., Reinert, J., Gannon, M., Ma, K., McNaughton, K., and Cavener, D.R. (2002). The PERK eukaryotic initiation factor 2 alpha kinase is required for the development of the skeletal system, postnatal growth, and the function and viability of the pancreas. Mol Cell Biol 22, 3864-3874. Zinszner, H., Kuroda, M., Wang, X., Batchvarova, N., Lightfoot, R.T., Remotti, H., Stevens, J.L., and Ron, D. (1998). CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev 12, 982-995. CHAPTER 2 REGULATION OF SUMO mRNA DURING ENDOPLASMIC RETICULUM STRESS Reprinted with permission from PlosOne, 8(9), Moore, K. A., Plant, J. J., Gaddam, G., Craft, J., & Hollien, J., Regulation of Sumo mRNA During ER Stress, E75723, Copyright 2013 27 OPEN 3 ACCESS Freely available online ©PLOS | Regulation of Sumo mRNA during Endoplasmic Reticulum Stress Kristin A. Moore, Joshua J. Plant, Deepika Gaddam, Jonathan Craft, Julie Hollien* Department o f Biology and th e Center fo r Cell and Genome Science, University o f Utah, Salt Lake City, Utah, United States o f America Abstract The unfolded protein response (UPR) is a collection o f pathways th a t maintains the protein secretory pathway during the many physiological and pathological conditions th a t cause stress in the endoplasmic reticulum (ER). The UPR is mediated in part by Ire1, an ER transmembrane kinase and endoribonuclease th a t is activated wh en misfolded proteins accumulate in the ER. Ire1's nuclease initiates th e cytosolic splicing o f th e mRNA encoding X-box b inding protein (Xbp1), a p ote nt transcription facto r th a t the n upregulates genes responsible fo r restoring ER fun ctio n. This same nuclease is responsible fo r the degradation o f many oth e r mRNAs th a t are localized to the ER, th ro u g h Regulated Ire1 Dependent Decay (RIDD). Here we show th a t Smt3, a homolog o fsm a ll u biq u itin -like modifier (sumo), is a non-canonical RIDDtarget in Drosophila S2 cells. Unlike oth e r RIDD targets, th e sumo transcript does n ot stably associate w ith th e ER membrane, b u t instead relies on an Xbp1-like stem loop and a second UPR mediator, Perk, fo r its degradation d uring stress. Citation: Moore KA, Plant JJ, Gaddam D, Craft J, Hollien J (2013) Regulation o f Sumo mRNA during Endoplasmic Reticulum Stress. PLoS ONE 8(9): e75723. doi:10.1371/journal.pone.0075723 Editor: Linda M. Hendershot, St. Jude Children's Hospital, United States o f America Received June 18, 2013; Accepted August 16, 2013; Published September 18, 2013 Copyright: © 2013 Moore et al. This is an open-access article distributed under th e terms o f th e Creative Commons Attrib utio n License, which permits unrestricted use, distribution, and reproduction in any medium, provided th e original author and source are credited. Funding: This w ork was funded by NIH R00 (GM081255) to JH and an NSF GK-12 award (0841233) to KM. The funders had no role in study design, data collection and analysis, decision to publish, or preparation o f th e manuscript. Competing Interests: The authors have declared th a t no competing interests exist. * E-mail: [e-mail] Introduction T h e flux o f p ro te in s th ro u g h th e sec reto ry p a thw a y varies extensively am o n g cell types a n d diffe rent p a th o lo g ica l a n d physiological conditions. As d em a n d fo r sec reted p ro te in s changes, so do th e systems w ith in th e en d o p la sm ic re tic u lum (ER) th a t are responsible for p ro te in folding a n d processing. E R stress results wh e n ac cum u la tio n o f u n fo ld ed p ro te in s overcomes th e folding cap ac ity o f th e E R . In m e ta z o an s, this situa tion is sensed by th re e m a in classes o f E R tra n sm em b ra n e p ro tein s- Ire1, Perk, a n d Atf6- which to g e th e r m ed ia te th e n um e ro u s ch an g es in gen e expression th a t define th e u n fo ld ed p ro te in re sponse (UPR) [1,2]. T h is re sponse is essential for n o rm a l d ev e lo pm e n t in m am m als a n d is th o u g h t to im p a c t several diseases, in clu d in g diabe te s, ca n ce r, a n d n eu ro d eg en e ra tiv e d isorders [3]. T h e U P R has b ro a d effects on tran sc rip tio n , tran sla tio n , a n d m R N A d ec ay d u rin g E R stress. T ra n s la tio n a l re g u la tio n is m ed ia te d largely b y Perk, w h ich dimerize s d u rin g E R stress a n d is a c tiv a ted th ro u g h au to p h o sp h o ry la tio n [4,5]. P erk p h o sp h o ry - lates th e tran sla tio n in itia tio n fa cto r e IF 2 a , th e re b y in h ib itin g cap-d e p e n d e n t tran s la tio n o f mo st tran sc rip ts [6,7]. H owever, tra n ­scripts c o n ta in in g u p s tre am o p en re a d in g frames (uORFs), such as th e basic-leucine z ip p e r (b-zip) tran sc rip tio n fa c to r Atf4, are selectively tran s la te d in these conditions a n d th u s th eir expression increases d u rin g E R stress [8]. Ire1 , a second m e d ia to r o f th e U PR , oligomerizes d u rin g stress, le a d in g to ac tiv a tio n o f its cytosolic kinase a n d en d o rib o n u c le a se d om ain s [9,10,11]. Ire1 specifically cleaves th e m R N A en c o d in g X -b o x b in d in g p ro te in (Xbp1), directly le a d in g to th e cytosolic splicing a n d tran sla tio n o f this b -z ip tran sc rip tio n fa c to r [12,13]. A lo n g w ith Atf4 a n d Atf6 (a th ird b -z ip tran sc rip tio n fa cto r a c tiva ted b y proteolysis d u rin g E R stress [14]), X b p 1 tran sc rip tio n a lly u p regula tes m an y genes en c o d in g ER-specific p ro te in folding c h a p e ro n e s a n d o th e r p ro te in s th a t fu n c tio n in th e sec retory p a thw a y [15,16]. Ire1 is also necessary fo r cleavage o f m an y o th e r mRNAs, in itia tin g th eir d eg rad a tio n th ro u g h R eg u la ted Ire1 D e p e n d e n t D e c ay (RIDD) [17,18,19]. A lth o u g h m u ch is kn own a b o u t th e m e c h an ism o f Xb p 1 splicing, th e fe ature s o f m RN As th a t identify th em as R ID D targe ts hav e b e e n m o re elusive. In Drosophila melanogaster cells, loca liz ation to th e E R m em b ra n e ap p e a rs to b e th e m a jo r fa c to r in ta rg e tin g m RN A s to this p a thw a y ; E R -ta rg e tin g signals a re b o th nec essary a n d sufficient for d eg rad a tio n b y R ID D [17,20], an d th e re is a s tro n g c o rre la tio n b e tw e en th e e x te n t o f m em b ra n e association o f a given m R N A a n d its d eg rad a tio n by R ID D d u rin g E R stress [20]. Conversely, cleavage site specificity does n o t a p p e a r to b e im p o rta n t for R ID D ta rg e tin g in Drosophila [20]. Based on gen e ontology classifications, R ID D targe ts in mammals a n d S. pombe a re en ric h e d fo r m RN A s e n c o d in g sec reto ry proteins, a n d th e re fo re a re p re sum e d to b e loca liz ed to th e E R [18,19,21]. H owever, R N A loca liz ation does n o t a p p e a r to fully a c c o u n t for th e specificity o f R ID D in these organisms, suggesting th a t th e re a re o th e r ta rg e tin g re q u irem en ts. T h e se re q u irem en ts m ay include specific sequences such as th e stem lo o p structure s th a t define th e cleavage sites in Xb p 1 a n d a re also e n ric h ed in m am m a lian R ID D targe ts [18,19,22]. Interestingly smt3, th e D. melanogaster h om o lo g o f sumo, was identified in m ic ro a rra y ex p e rim en ts as a p o te n tia l R ID D ta rg e t [17], despite lacking an y re co g n izab le seq u e n ce elements th a t would ta rg e t it to th e ER . T h is ob serv atio n led us to hypothesize th a t th e sumo tra n s c rip t m ay rely o n d iffe ren t mechanisms for d eg rad a tio n com p a re d to th e m ajo rity o f R ID D targe ts in flies. PLOS ONE | www.plosone.org 1 September 2013 | Volume 8 | Issue 9 | e75723 28 Regulation of Sumo mRNA during ER Stress H e re we d em o n s tra te th a t th e m R N A e n c o d in g sumo is a non-c a n o n ic a l R ID D ta rg e t a n d dep e n d s on b o th an X b p l-lik e stem loop s tru c tu re a n d P e rk for its d eg rad a tio n d u rin g E R stress. Results The mRNA encoding sumo is a non-canonical RIDD target W e previously o b served b y m ic ro a rra y th a t th e re la tive am o u n t o f th e sumo (smt3, CG4 4 9 4 ) tra n sc rip t dec rea ses d u rin g E R stress in D. melanogaster S2 cells, in an Ir e 1 -d e p e n d e n t b u t X b p 1 - in d e p e n d e n t m a n n e r [17]. W e co nfirmed this re su lt h e re by q u an tita tiv e re a l-tim e P C R (qPCR) (Figure 1A-B). D e p le tio n o f eith e r Ire1 o r Xb p 1 b y R N A i in h ib ite d th e u p re g u la tio n o f BiP, a m a jo r E R c h a p e ro n e , d u rin g E R stress (Figure 1A). However, dep le tio n o f Ire1 b u t n o t X b p 1 blocked th e d ow n reg u latio n o f sumo m R N A (Figure 1B). T o test w h e th e r this d ec rea se was th e re su lt o f m R N A decay, we tre a te d S2 cells with ac tin om y c in D (1 mg/mL) to block tran sc rip tio n a n d collected samples over time in th e p re sen c e a n d absence o f d ith io th re ito l (D T T , 2 mM), a re d u c in g a g e n t th a t strongly induce s E R stress. T u n ic am y c in a n d th apsigargin, two o th e r s tro n g in d u ce rs o f E R stress in m am m a lian cells, do n o t efficiently ac tiv a te Ire1 in S2 cells [17], th u s D T T was used to ac tiva te E R stress p athwa y s in th e following expe riments. S umo m R N A levels we re stable in a c tin om y c in -tre a te d cells over six h o u rs, b u t significantly d e c re a sed o ver time d u rin g E R stress (Figure 1C). T h e re fo re , sumo is a R ID D target. W h ile E R loca liz ation a p p e a rs to b e nec essary a n d sufficient to ta rg e t m RN A s to R ID D in S2 cells [17,20], sumo co n ta in s n e ith e r a signal sequence n o r a tra n sm em b ra n e d om a in , a n d thus its m R N A c a n n o t localize to th e E R b y c o n v e n tio n a l mechanisms. T o d e te rm in e ex p e rim en tally w h e th e r this m R N A is localized to th e E R th ro u g h an a lte rn a tiv e p a thw a y , we used a previously-desc ribed d e te rg e n t fra c tio n a tio n m e th o d [20,23] to sep a rate m em b ra n e -b o u n d vs. cytosolic m RN A s from S2 cells. As p re d ic ted from its seq u e n ce a n d th e kn own cy to so lic /n u c le a r fun c tio n s o f th e sumo p ro te in , sumo m R N A was highly en ric h ed in th e cytosolic fraction, alo n g with th e m R N A e n c o d in g a c tin (Figure 1D). Its fra c tio n a tio n b e h a v io r d id n o t ch a n g e with E R stress (Figure 1D), a lth o u g h as we h av e previously fo u n d , th e m em b ra n e -a sso c ia ted m R N A sp a rc b e c am e m o re d ig ito n in -e x tra c ta b le d u rin g E R stress, p e rh a p s d u e to th e c o n c u rre n t a tte n u a tio n o f tran sla tio n [20]. Interestingly, X b p 1 m RN A , w h ich is cleaved b y Ire1 d u rin g stress, also d id n o t strongly fra c tio n a te w ith th e m em b ra n e (Figure 1D), suggesting th a t strong, stable association with th e E R is n o t absolutely re q u ired for cleavage b y Ire1. T o fu rth e r test a possible ro le for E R loca liz ation in th e d eg rad a tio n o f sumo m RN A , we tre a te d S2 cells with p u rom y c in (35 mM), a tran sla tio n elo n g a tio n in h ib ito r th a t releases mRN As from ribosomes a n d disrupts th e E R loca liz ation o f m RN As th a t rely on tra n s la tio n -d e p e n d e n t m ech an isms o fm em b r a n e targe ting. D e g rad a tio n o f th e sumo tra n s c rip t d u rin g E R stress was n o t significantly affected b y p u rom y c in tre a tm e n t (Figure 1E). In contra st, o th e r R ID D targ e ts (sparc a n d T sp 4 2E e ) we re no longer d eg rad ed in th e p re sen c e o f p u rom y c in , m o st likely b ec au se th e m RN As we re no lo n g e r assoc ia ted with th e ER . T h e s e results suggest th a t rib o som e -d e p en d e n t m em b ra n e loca liz ation is n o t necessary fo r R ID D ta rg e tin g o f sumo m RN A. An Xbp1 -like stem loop is necessary and sufficient for targeting sumo mRNA to RIDD T o ex am in e th e cis elements in th e sumo tra n sc rip t im p o rta n t for its d eg rad a tio n d u rin g E R stress, we used re p o rte r plasmids expressing th e co d in g sequence o f sumo u n d e r th e co n tro l o f th e co p p e r-in d u c ib le m e tallo th io n e in p rom o te r. Afte r in d u c in g ex­p re ssion o f re p o rte r m RN A s in S2 cells with C u S O 4, we removed th e tran sc rip tio n a l in d u c e r a n d m o n ito re d m R N A d eg rad a tio n in th e pre sen c e a n d ab se n c e o f E R stress. A lth o u g h re g u la tio n o f loca liz ation, tran sla tio n , a n d d eg rad a tio n o f m RN A s often relies o n seq u e n ce elements w ith in th e 3 'U T R , we fo u n d th a t re p la c in g th e sumo 3 'U T R w ith th a t o f sp arc (an ER -lo ca liz ed R ID D target) o r G a p d h 1 (a cytosolic m R N A un affec ted b y E R stress) did n o t affect its ta rg e tin g to R ID D (Figure 2A). F u rth e r sequence analysis, h owever, revealed th a t th e D. melanogaster sumo tran sc rip t co n ta in s a p re d ic te d stem lo o p n e a r th e e n d o f its co d in g sequence th a t b e a rs a striking similarity to th e X b p 1 stem loop sequences th a t a re cleaved b y Ire1 (Figure 2B). De le tio n o f th e 27 nucleotides su rro u n d in g this s tru c tu re abolished E R s tress-d ep en d e n t d eg rad a tio n o f th e sumo m R N A re p o rte r (Figure 2C). T o p ro b e this sequence m o re specifically, we m a d e p o in t m u tan ts w ith in th e loop a n d th e stem. M u ta tio n o f an y o f th e 4 conse rved bases with in th e 7 -m em b e r lo o p [24] b lo ck e d ER s tress-d ep en d e n t d eg rad a tio n o f sumo m RN A , whe rea s m u ta tio n o f a no n -c o n serv e d b a se in th e loop h a d no effect (Figure 2C). Likewise, m u ta tio n o f 3 nucleotide s w ith in th e p re d ic te d stem s tru c tu re also blocked d eg rad a tio n (Figure 2C). T o d e te rm in e w h e th e r th e Xb p 1 -lik e stem lo o p w ith in th e sumo tra n sc rip t is sufficient for targ e tin g an m R N A to R ID D , we used a re p o rte r p lasm id e n c o d in g GF P. T h e G F P m R N A a lo n e is n o t a R ID D ta rg e t ([20] a n d F igure 2D, E). H owever, ad d itio n o f th e 27 nucleotide s su rro u n d in g th e sumo stem lo o p to th e 3 ' e n d o f th e G F P c o d in g seq u e n ce led to an E R s tress-d ep en d e n t incre ase in th e d eg rad a tio n o f G F P m R N A (Figure 2D). W e th e n tested w h e th e r this d eg rad a tio n was Ire1 d e p e n d e n t b y d e p le tin g Ire1 th ro u g h R NAi. D e g ra d a tio n o f th e G F P m R N A alone was un affec ted by Ire1 deple tion, whe rea s th e e n h a n c e d d eg rad a tio n o f th e G F P -stem lo o p m R N A seen d u rin g E R stress was in h ib ite d by Ire1 dep le tio n (Figure 2E). T h u s , d eg rad a tio n o f th e GF P-stem lo o p m R N A occurs th ro u g h R ID D . Sumo is not a strong RIDD target in mammalian cells T o d e te rm in e w h e th e r re g u la tio n o f sumo m R N A b y R ID D is conse rved, we sea rch e d fo r Xb p 1 -lik e stem loops in sumo tran sc rip ts o f o th e r organisms, using th e c rite ria th a t an Ire1 site m u s t co n ta in a stem lo o p with a t le a st 5 b asep airs in th e stem an d exactly 7 nucleotides in th e lo o p , a n d m u s t co n ta in th e four conse rved loop nucleotides d ep ic te d in F igure 2B. T h e Ire1 site was n o t widely conserved; even with in Drosophila, we fo u n d Ire1 sites in th e sumo tran sc rip ts o f only 2 o f th e 11 species we ex am in e d , n am e ly D. sechellia a n d D. simulans, th e closest relatives to D. melanogaster (Figure 3A). W e d id n o t u n co v e r an y p re d ic ted Ire1 sites in th e sumo tran sc rip ts for h um an s , X . laevis, o r C. elegans. De spite this lack o fg e n e ra l conse rvation, we did fin d an Ire1 site in a mo u se sumo tran sc rip t. M ic e possess th re e sumo genes, in c o n tra s t to D. melanogaster, which h a s only o ne. W h ile n e ith e r sumo1 n o r sumo3 co n ta in s a p re d ic te d Ire1 site, sumo2 h a s a stem lo o p a t exactly th e same position, re la tive to th e co d in g sequence, as th e o n e in D. melanogaster sumo. T h e lo o p a n d first fo u r basepairs o f th e stem a re perfec tly conse rved b e tw e en these two transc ripts (Figure 3A). T o d e te rm in e w h e th e r sumo2 is d ow n reg u lated d u rin g ER stress in m ouse cells, we tre a te d m ouse p re o ste o b la st M C 3T 3 -E 1 cells with D T T (2 mM , 4 hrs) a n d m e a su red m R N A levels for th e m ouse sumo homologs b y q P C R (Figure 3B). T h e se cells robustly d eg ra d e th e R ID D ta rg e t Blos1 in re sponse to E R stress. Sumo2 displayed a v e ry we ak d ow n reg u latio n (p-value fo r u n tre a te d vs. D T T - tr e a te d = 0.08). D e p le tio n o f Ire1 b y RNAi blocked th e d eg rad a tio n o f Blos1, a n d a p p e a re d to also affect sumo2 down- PLOS ONE | www.plosone.org 2 September 2013 | Volume 8 | Issue 9 | e75723 29 Regulation of Sumo mRNA during ER Stress actin spare Xbp1 sumo Figure 1. Sumo mRNA is a non-canonical RIDD target. For all panels, we measured relative RNA abundance by qPCR; shown are the averages and SDs of 3-4 independent experiments. Except for the fractionation in panel D, we normalized all mRNA abundance measurements to the housekeeping control Rpl19. A-B. Relative mRNA levels of BiP (panel A) and sumo (panel B) in mock-treated and Ire1- or Xbp1-depleted Drosophila S2 cells incubated in the absence and presence ofER stress (2 mM DTT, 4.5 hours). Xbp1 transcript levels in the Xbp1 RNAi-treated cells were 13.5% + /- 0.9% ofth e levels in control cells, as measured by qPCR. C. Timecourse ofsumo mRNA levels in S2 cells treated with or w ithout actinomycin D (1 mg/ mL) to block transcription and DTT (2 mM) to induce ER stress. D. Fraction membrane (membrane/total) for mRNAs from S2 cells treated with and without DTT (2 mM, 20 min). We separated cytosolic and membrane RNAs using detergent extraction (see Materials and Methods). E. Relative mRNA levels in cells treated with or without 35 mM puromycin (added 10 min prior to stress) and DTT (2 mM, 4 hrs). doi:10.1371/journal.pone.0075723.g001 regula tion; however th e overall effect was we ak a n d d id n o t pass th e s ta n d a rd p -v a lu e cu to ff fo r statistical significance, using a p a ire d t-test (p-value for d tt/u n tr e a te d , co n tro l vs Ire1 RNAi = 0.2). T o a c c o u n t for p o te n tia l v a ria tio n in Ire1 site p re ference s a n d test th e possibility o f sumo re g u la tio n in h um a n cells, we re p e a te d th e ab o v e ex p e rim en ts in H E K 2 9 3 cells (Figure 3C). T re a tm e n t with D T T (2 mM , 4 hrs) re su lted in a small b u t significant dec rea se in sumo2 m RN A levels (p-value = 0.02). T h is effect was n o t Ire1 -d e p en d en t, consistent with th e lack o f p re d ic te d Ire1 sites in h um a n sumo transc ripts. Levels o f sumo1 a n d 3 rem a in ed u n ch a n g ed in b o th M C 3T 3 -E 1 a n d H E K 2 9 3 cell types. T h e se results suggest th a t while sumo is d ow n reg u la ted in m am m als d u rin g E R stress, th e effect is small a n d th e m ech an isms re g u la tin g sumo levels v a ry b e tw e en organisms. RIDD of the sumo transcript is dependent on Perk D u rin g E R stress, P e rk ac tiv a tio n a n d p h o sp h o ry la tio n o f e IF 2 a re su lt in a n a tte n u a tio n o f tran sla tio n , w h ich c a n affect m RN A stability [25,26]. T o test w h e th e r P e rk is im p o rta n t for d eg rad a tio n o f sumo m RN A , we d ep le te d P e rk by R N A i a n d m e a su red th e re la tive a b u n d a n c e o f en d o g en o u s sumo m RN A in th e pre sen c e a n d ab se n ce o f E R stress. Strikingly, d eg rad a tio n o f th e sumo tra n sc rip t d u rin g E R stress was completely abolished in th e ab se n ce o f P erk (Figure 4C). S umo m R N A levels in th e absence o f stress we re n o t affected b y P erk dep le tio n (levels in P erk R N A i/ PLOS ONE | www.plosone.org 3 September 2013 | Volume 8 | Issue 9 | e75723 30 Regulation of Sumo mRNA during ER Stress • 3'UTR: sumo DTT: - i 0 spare Gapdhl to _2 <z O ' 1Bi f B g Ca U G C C U A A U G C X b p1 j A 5'- U AAGCUGA U- 3' GCC U G C C U A C G C G C G GC a A G 250-C A ' C U G J,,,u C G 5'- A U- 3' GAC GFP GFP+SL GFP GFP+SL DTT: Ire1 RNAi: - ° -4 Figure 2. A stem loop sequence in the sumo mRNA is important for RIDD targeting. For panels A, C-E: plasmids expressing reporter mRNAs under the control of a copper-inducible promoter were stably transfected into S2 cells. After inducing expression, we removed the copper to stop transcription of reporter mRNAs, incubated cells in the presence and absence o f ER stress (2 mM DTT, 5 hrs), and collected RNA samples. Relative RNA abundance was measured by qPCR and normalized to Rpl19. Shown are the averages and SDs of 3 (panels A, C) or 2 (panels D-E) independent experiments. A. Reporters expressing the coding sequence of sumo followed by various 3'UTRs. We normalized RNA levels to a control sample collected immediately before washing out the copper; thus RNA measurements reflect the amount o f degradation after 5 hrs without copper. B. RNA sequences ofsumo and Xbp1 from D. melanogaster, surrounding the stem loop structures discussed here. Highlighted in red arethe loop nucleotides conserved in Xbp1 across species. Numbering in the sumo mRNA is relative to the translation start site. C. Reporters containing the sumo coding sequence and 3' UTR, with various mutations. DSL= deletion of nucleotides 244-270 in the coding sequence ofsumo; stem mut.= C257G/G259C/ G260C. D. Reporters expressing GFP with the Gapdhl 3'UTR, with and without the stemloop sequence of sumo (nt 244-270). E. Degradation of reporters from D in untreated cells and those depleted of Ire1. doi:10.1371/journal.pone.0075723.g002 co n tro l cells = 0.8 4 , p = 0.5). F u rth e rm o re , X b p l splicing (Figure 4B) a n d d eg rad a tio n o f R ID D targe ts C G 3 9 8 4 , H y d r2 , a n d sp a rc (Figure 4C) we re large ly unaffec ted, alth o u g h d eg ra d a ­tion o f C G 6 6 5 0 , a n o th e r R ID D ta rg e t, was p a rtia lly inhibited. W e ste rn b lo t analysis o f th e p h o sp h o ry la tio n o f e IF 2 a confirmed th e efficient k nockdow n o f P e rk (Figure 4A), whose m R N A levels w e re re d u c ed to 39% + / - 7% com p a re d to controls, as mea su red b y q P C R . T h e se d a ta suggest th a t Ir e 1 -d e p e n d e n t d eg rad a tio n o f sumo m R N A is p a rtic u la rly sensitive to P erk activity. PLOS ONE | www.plosone.org 4 September 2013 | Volume 8 | Issue 9 | e75723 31 Regulation of Sumo mRNA during ER Stress melanogaster subgroup " D. m e la n o g a s te r D. s e c h e l l i a _ D. s im u la n s D. y a k u b a D. e r e e t a D. a n a n a s s a e D. p seudoo b s c u r a D. p e r s im l l l i s D. w i l l i s t o n i D. v i r i l l i s D. m o ja v e n s is D. g r im sh aw i D. m e la n o g a s te r H. m u s c u lu s (sumo2) H. s a p ie n s (sumo2) X . l a e v i s (sumo2) c. e le g a n s GT T TAC CAGCAGCAGAC T GGTGGCGC T C CATAA GTTTACCAG CTGGTGGCGCTCCTTAA GTTIACCAG CTGGTGGCGCTCCTTAA GTTTACCAG C GG GGCGCTCTATAA GTTTACCAG C GG GGCGCTCTATAA GTATACCA CAGCAGAC GGTGGAGCTGTCTAT GTGTACCAG C GG GGTTTCTATTAA GTGTACCAG C GG GGTTTCTATTAA GTGTA CA CAGCA AC GG GGAGGCAGCATC GTGTACCAG C GG GGTGTTTATTAA GTGTACCA CAGCAGAC GG GGTGTTTATTAA GTTTACCAGCAGCAGAC GG GGCGGCGGCGGC GTTTACCAG CTGGTGGCGCTCCATAA GTGT CCAG CTGG GGTGTCTACTAA GTGT CCA CAGCAGAC GG GGTGTCTACTGA GTTT CAGCAGCAGACTGGTGGATCCTACTAA GTCTACCA AGCAG GG GGATTCTAG Figure 3. Sumo mRNA is not strongly affected by ER stress in mammalian cells. A. Conservation of the Ire1 site and surrounding region in sumo transcripts across species. The stem loop is indicated abovethe sequences and the region aligning with the loop from D. melanogaster sumo is shown in red. Deviations from the D. melanogaster sequence in the stem loop region are shown in blue. While most species have a conserved loop sequence, perfect basepairing in the stem is present only in the fly and mouse sequences. B-C. We either mock-treated (control) or used siRNA to deplete Ire1a from MC3T3-E1 mouse osteoblasticfibroblasts (panel B) or Hek293 human kidney cells (panel C). We then compared RNA levels in the presence and absence of DTT (2 mM, 4.5 hrs), by qPCR. Blos1, a RIDD target in mouse and humans, is shown as a control. Except for Blos1, the differences in mRNA levels between control and Ire1 siRNA-treated cells were not statistically significant. Shown are the averages and SDs for 2-3 independent experiments. doi:10.1371/journal.pone.0075723.g003 Discussion E R stress occurs in m a n y physiological a n d path o lo g ica l conditions, a n d th e re sponse to ac cum u la tio n o f misfolded p ro te in s can d e te rm in e cell fa te . W h ile m u ch is kn own a b o u t th e initia tion a n d d ow n s tre am effects o f tran sc rip tio n a l re g u la tio n o f mRN As d u rin g th e U P R , th e fe atu re s th a t ta rg e t m RN As to th e R ID D p a thw a y a re less well u n d e rsto o d . W e previously fo u n d th a t in Drosophila S2 cells, E R loca liz ation is b o th nec essary a n d sufficient for ta rg e tin g m RN As to R ID D [17,20], while re co g n izab le Ire1 cleavage sites a re n o t p re d ic to rs o f R ID D ta rg e tin g [20]. F u rth e rm o re , previous mutagene sis ex p e rim en ts fo u n d a distinc t lack o f specific sequence elements affecting d eg rad a tio n , o th e r th a n E R -ta rg e tin g signal sequences [17]. H e re we d em o n s tra te th a t th e m R N A e n c o d in g sumo is a n exc eption to these rules. A lth o u g h th e sumo tra n s c rip t is d e g ra d e d b y R ID D , it is n o t stably associated with m em b ra n e s . D e g rad a tio n strongly d ep e n d s on a specific cis elem e n t in th e sumo cod in g sequence, com p rised o f a stem lo o p s tru c tu re v e ry similar to th e conse rved Ire1 re cognition sites in X b p 1 , a n d mutagene sis o f th e conse rved bases with in this stem lo o p inhibits d eg rad a tio n o f th e tran sc rip t. T h is p arallels mutagenesis ex p e rim en ts showing th a t these same conse rved bases w ith in th e Hac1 (the X b p 1 h om o lo g in yeast) a n d X b p 1 stem loops a re im p o rta n t fo r cleavage a n d splicing [27,28,29]. T h is d istin c t ta rg e tin g m e c h an ism suggests th a t down reg u latio n o f sumo serves a diffe rent fu n c tio n d u rin g E R stress, com p a re d to th e d eg rad a tio n o f o th e r R ID D targets, w h ich m ay relieve stress by re d u c in g th e p ro te in folding lo a d o n th e E R . T h e sumo p ro te in covalently modifies m a n y ta rg e t p ro te in s , often affecting p ro te in loca liz ation a n d activity [30]. Interestingly, th e spliced version o f m ouse X b p 1 c a n b e SUM O y la ted , le a d in g to a d ec rea se in tran sc rip tio n a l ac tiva tion o f ta rg e t genes [31,32]. T h u s d u rin g E R stress, d eg rad a tio n o f sumo m R N A m a y en h a n c e U P R signaling. H owever, th e stem loop s tru c tu re shown h e re to b e critical for R ID D o f sumo does n o t a p p e a r to b e widely conse rved b ey o n d D. melanogaster, a n d sumo m R N A is only very weakly down -reg u lated in th e m am m a lian cells we hav e tested. S umo re g u la tio n is highly sensitive to P erk, as its d eg rad a tio n is complete ly ab o lish ed w h e n P e rk is d ep le ted . T h is is in co n tra s t to c a n o n ic a l Drosophila R ID D targets, w h ich a re only mildly sensitive to P erk dep le tio n (Figure 4C a n d [17]). T h e m ech an isms th a t m ed ia te P e rk 's effect on sumo ta rg e tin g to R ID D a re u ncle ar. In te ra c tio n s b e tw e en tran sla tio n a l reg u la tio n a n d th e R ID D PLOS ONE | www.plosone.org 5 September 2013 | Volume 8 | Issue 9 | e75723 32 Regulation of Sumo mRNA during ER Stress Xbp1-u- Xbp1-s- Figure 4. RIDD of sumo is dependent on Perk. We used RNAi to deplete S2 cells of Perk. A. Western blot showing the levels of phosphorylated and total eIF2a. B. Agarose gel showing the relative levels ofunspliced and spliced Xbp1, amplified by reverse-transcription- PCR using primers surrounding the splice site. C. Relative mRNA levels in ER stress treated vs untreated cells, determined by qPCR. Panels A-B show representative data, panel C is the average and SD of 3 independent experiments. doi:10.1371/journal.pone.0075723.g004 p a thw a y a re n o t u n p re c e d e n te d , as p ro te c tio n from tran sla tio n a l a tte n u a tio n is o n e way b y w h ich m RN A s a t th e E R m em b ra n e can escape d eg rad a tio n b y R ID D [20]. I t is possible th a t tran sla tio n o f th e sumo tran sc rip t is especially a tte n u a te d wh e n P e rk is activated, o r th a t its d eg rad a tio n is especially re lia n t on this atten u a tio n , p e rh a p s fa cilita tin g th e fo rm a tio n o f th e sumo stem loop stru c tu re o r allowing Ire1 g re a te r access to th e m R N A th ro u g h ribosome deple tion. It is also possible th a t tran s la tio n a tte n u a tio n is generally re q u ired for R ID D , b u t sumo is u n iq u ely un affec ted by P e rk - in d e p e n d e n t m ech an isms o f a tte n u a tio n th a t o c c u r d u rin g ER stress in S2 cells [33]. Beyond a tte n u a tio n o f g en e ra l tran sla tio n , P e rk -d ep en d e n t e IF 2 a p h o sp h o ry la tio n also en h a n ce s tran sla tio n o f ce rta in m RN As such as Atf4 a n d G a d d 3 4 [8,34]. It is unlikely th a t such a p ro te in is m e d ia tin g P e rk 's effect on sumo d eg rad a tio n , as sumo m RN A is still d e g ra d e d d u rin g E R stress wh e n tran sla tio n is in h ib ite d (Figure 1E). Because we h av e n o t specifically ex am in e d tran sc rip tio n o f sumo in P erk -d ep leted cells, it is also possible th a t P erk kno ck d ow n in d irec tly affects sumo tran sc rip tio n by an unkn own mech an ism. O v e rall, we p ro p o se th a t w hile E R loca liz ation is a key fa cto r in targ e tin g mo st m RN A s for R ID D in D. melanogaster cells, stable m em b ra n e association can b e o v ercom e b y th e pre sen c e o f a specific Ire1 re co g n itio n site co u p le d with tran s la tio n a l a tten u a tio n via Perk. A lth o u g h this ap p e a rs to b e an ex c ep tio n to th e gene ral R ID D targ e tin g rules in flies, this m e c h an ism m ay b e m o re p re v a len t in o th e r organisms. R ID D targe ts in all systems studied so fa r a re en ric h ed fo r m RN As th a t a re p re d ic te d to loca liz e to th e E R [17,18,19,35], b u t m am m a lian R ID D targe ts a re also e n ric h ed for m RN A s c o n ta in in g X b p l-lik e stem loops [18,19,22]. In te re s t­ingly, while p re fe rre d cleavage sites o f several R ID D targe ts have b e e n d e te rm in e d in b o th D. melanogaster a n d S. pombe, mutagenesis o f these sites results in cleavage a t a ltern ativ e sites allowing for d eg rad a tio n to still o c c u r [17,21]. In co n tra st, mutagene sis o f residues im p o rta n t fo r cleavage o f at lea st two m am m a lia n R ID D targe ts inhibits th e ir d eg rad a tio n in vitro [22,36]. T h e s e correla tions suggest th a t while sumo re g u la tio n b y R ID D does n o t a p p e a r to b e widely conse rved, th e targ e tin g m e c h an ism exemplified by sumo m ay b e m o re gene rally applicable to R ID D in o th e r organisms, including mammals. Materials and Methods Cell culture, ER stress induction, and RNAi W e cu ltu re d Drosophila S2 cells (Invitrogen) a t ro om tem p e ra tu re in S c h n e id e r's m ed ia (Invitrogen) su p p lem en ted with 10% fetal b o v in e serum a n d antibiotics. Unless o therwise in d icate d , we in d u c e d E R stress for 5 ho u rs with 2 m M D T T . For R N A i ex p e rim en ts, we amplified regions o f th e Ire1 (CG4583), Perk, a n d X b p 1 co d in g sequences (CDS) from S2 cell cD NA using p rim e rs co n ta in in g T 7 R N A p o lymera se sites on th e 5 ' ends. W e th e n synthesized d sR N A from these tem p lates using th e M eg a scrip t T 7 kit (Ambion). W e in cu b a te d S2 cells with d sR N A in serum free m ed ia fo r 45 minute s, re p la ce d th e serum, a n d allowed cells to re co v er for 4 days. W e th e n re p e a te d th e d sR N A tre a tm e n t a n d subjected cells to E R stress o n e day following th e second d sR N A tre a tm e n t. W e cu ltu re d M C 3T 3 -E 1 (ATCC) a n d He k 2 9 3 (from A.V. M a r ic q lab) cells following A T C C guidelines in M E M a a n d DM EM m ed ia (Invitrogen), respectively, su p p lem en ted with 10% fe ta l bo v in e se rum a n d antibiotics. F o r Ire1 k nockdow n ex p e ri­m en ts we used organism-specific Ire1 siRN A (Qiagen) an d followed In v itro g en RN A im ax guidelines for tran sfectio n o f siRNA. W e subjected cells to E R stress 4 8 -7 2 h o u rs after transfection, w h e n cells we re a p p ro x im a te ly 80% confluent, an d collected R N A afte r 4 h ours. Quantitative real-time PCR For all R N A analyses, we isola ted to ta l R N A using T riz o l re a g e n t (Invitrogen), a n d synthesized cD N A using 2 mg o f total R N A as a tem p la te , T 1 8 as a p rim e r, a n d M -M uLV reverse tran sc rip ta se (NEB). W e m e a su red re la tive m RN A a b u n d a n c e by re a l time q u an tita tiv e P C R using a Mastercy cle r ep re alplex (Eppendorf) w ith SYBR G re e n as th e flu o re scen t dye. We m e a su red e a c h sam p le in trip lic ate a n d n o rm a liz e d to th e rib o som al p ro te in R p l1 9 m RN A . T o co n tro l fo r co n tam in a tin g p lasm id o r g en om ic DNA we also m e a su red samples to w h ich no reverse tran sc rip ta se was ad d e d . T h e p rim e rs used for q P C R are given in T a b le 1. Digitonin fractionation T o sep a ra te m em b ra n e a n d cytosolic m RN A s we used a m e th o d dev e lo p ed by S tephens a n d N ic c h itta [23] a n d modified in o u r pre v io u s studies [20]. Briefly, we in c u b a te d S2 cells with o r w ith o u t D T T (2 mM , 20 minutes), a d d e d cycloheximide (35 mM) fo r 10 min, a n d collected cells by cen trifu g atio n . W e th en re su sp en d ed cells in cytosol buffer (150 m M K O A c , 20 mM H e p e s p H 7.5, 2.5 m M Mg(OAc)2 , 200 U /m L R N a s eO U T , 35 mM cycloheximide) c o n ta in in g 1 m g /m L d igitonin (15 min o n ice). W e th e n c entrifuged th e lysates (800 xg, 5 m in a t 4 C) an d collected th e s u p e rn a ta n t as th e cytosolic fractio n . W e re suspended th e p ellet in cytosol buffer with 1% T rito n X -1 0 0 (15 m in on ice), c en trifu g ed as above a n d collected th e s u p e rn a ta n t as th e m em b ra n e -b o u n d fractio n . W e m e a su red th e a b u n d a n c e o f specific RNAs in ea ch fra c tio n b y q P C R a n d c a lcu lated th e PLOS ONE | www.plosone.org September 2013 | Volume 6 8 | Issue 9 | e75723 33 Regulation of Sumo mRNA during ER Stress T a b le 1. Primers used fo r qPCR. Gene Name Primer1 Primer 2 Dm sumo (smt3) TTTGTTATTTACGCACACAGACG GTCTGACGAAAAGAAGGGAGG Dm Ribosomal Protein L19 (Rpl 19) AGGTCGGACTGCTTAGTGACC CGCAAGCTTATCAAGGATGG Dm Act5C (actin) ATGTGTGACGAAGAAGTTGCT GAAGCACTTGCGGTGCACAAT Dm sparc AAAATGGGCTGTGTCCTAACC TGCAGCACAATCTACTCAATCC Dm Xbp1 GGCCATCAACGAGTCACTGCT TGTGTCCACCTGTTGTATACC Dm Tsp42Ee AACAACGTGCGTAACT ACAAGC TTCCAAATTTAAATCTTTCCCG Dm CG3984 CTACTGTTGTTCCTGGTACCCC CTGGTTGCTCAGTAACACTTGG Dm Hydr2 CGCATACACGACTATTTAACGC TTTGGTTTCTCTTTGATTTCCG Dm CG6650 ACAATGGGACAGGCAAAGAC GGTGACATTCGTTTCCGAGT Dm sumo reporters CAGTGCAACTAAAGGGGGGATC TTTGTTATTTACGCACACAGACG, or TCCGTCGCGGCCGCTTATGGAGCGCCACCAGTCTGCT GFP reporters CCTGAAGTTCATCTGCACCA TGCTCAGGTAGTGGTTGTCG Mm Rpl19 CTGATCAAGGATGGGCTGAT GCCGCTATGTACAGACACGA Mm and Hs sumo1 GGAGGCAAAACCTTCAACTG CCCCGTTTGTTCCTGATAAA Mm sumo2 GGGAGCCTGCTACTTTACTCC TCCATCTCATGTCAACCAGAA Mm sumo3 GATGGCTCGGTGGTACAGTT TGTCCTCATCCTCCATCTCC Mm and Hs Blos1 CAAGGAGCTGCAGGAGAAGA GCCTGGTTGAAGTTCTCCAC Hs Rpl 19 ATGTATCACAGCCTGTACCTG TTCTTG GTCTCTTCCTCCTTG Hs sumo2 AGCTGAGGAGACTCCGGCGCTCGC AGTAGACACCTCCCGTCTGC Hs sumo3 AGAATGACCACATCAACC AGTAGACACCTCCCGTCTGC Dm=Drosophila melanogaster. Mm = Mus musculus. Hs=Homo sapiens. doi:10.1371/journal.pone.0075723.t001 fractio n m em b ra n e as th e a b u n d a n c e o f a p a rtic u la r m R N A in th e m em b ra n e -b o u n d fra c tio n div id ed b y th e sum o f th a t m R N A 's a b u n d a n c e in th e cytosolic a n d m em b ra n e -b o u n d fractions. Plasmids and reporter RNA analyses F o r sumo re p o rte rs, we amplified th e sumo (smt3, CG4494) C D S from S2 cell cD N A a n d su b clo n ed into an expression v ec tor co n ta in in g th e co p p e r-in d u c ib le D. melanogaster m etallothione in p rom o te r d esc rib ed previously [17]. T o ex am in e th e effects o f th e 3 'U T R , we sepa rate ly amplified th e 3 'U T R s o f sumo, sparc (CG6378), a n d G ap d h 1 (CG12055) from S2 cD N A a n d subcloned into th e sumo expression v e c to r ju s t d ow n s tre am o f th e C D S . We in tro d u c e d m u ta tio n s into th e sumo v e c to r co n ta in in g th e sumo 3 'U T R fo r F igure 2C using PC R -b a se d mutagene sis. F o r G F P re p o rte rs, we used a previously-de scribed EG F P re p o rte r in th e co p p e r-in d u c ib le expression v e c to r [20], a n d re p la c e d th e v ec tor SV40 3 'U T R with th a t f r om G a p d h 1 . T o in tro d u c e th e sumo SL, we a d d e d th e 30 n u cle o tid e sequence from th e 3 ' e n d o f th e sumo C D S (including th e stop codon) in -fram e to th e 3 ' e n d o f th e G F P CDS. W e g en e ra te d stable, poly clo n al cell lines b y co tran sfec tin g o u r expression plasmids (1.8 mg) with a p u rom y c in re sistanc e plasmid (0.2 mg) using Cellfectin II (Invitrogen) a n d selecting for re sistant cells. T o m o n ito r d ec ay o f m RN A s expressed from re p o rte r construc ts, we tre a te d cells with C u S O 4 (200 mM overnight) to in d u c e expression, collected ‘‘time 0 '' R N A samples, th e n wa shed cells to rem o v e th e C u SO 4. W e h av e previously shown this p ro c e d u re to b e effective in b locking tran sc rip tio n o f th e re p o rte r m RN A , such th a t su b seq u e n t m e a su rem en ts reflect dec ay rates [20]. W e th e n eith e r left cells u n tre a te d o r a d d e d D T T (2 mM) a n d collected R N A samples afte r 5 h o u rs. R N A ab u n d a n c e m e a su rem en ts we re n o rm a liz e d to th e level o f R N A in th e C u S O 4-tre a ted cells. Western blot analysis W e wa shed cells in PBS be fo re lysing in 1x R IPA buffer (25 m M T ris, p H 7.6, 150 m M N aC l, 1% N P -4 0 , 1% Na - d eoxycholate, a n d 0 .1% SDS) with p ro te a se in h ib ito rs (T h e rm o scientific). W e resolved p ro te in on N u P a g e Bis-Tris 4-12% gels (Invitrogen), tran s fe rred th em to nitroce llulose m em b ra n e s an d p ro b e d for to ta l e IF 2 a (abcam, 1:500) o r S e r5 1 -P e IF 2 a (abcam 26197, 1:1000) followed by a sec o n d ary IR D y e 8 0 0CW an tib o d y (Licor 926-32210, 1:10000). W e visualized immu n o b lo ts using a Licor Odyssey imager. Xbp1 splicing assay Using S2 cD N A as a tem p la te we assayed X b p 1 splicing th ro u g h P C R analysis o f a fra gm e n t o f th e X b p 1 tran sc rip t en c om p assin g th e 23 n u cle o tid e splice site. W e resolved th e spliced a n d unspliced p ro d u c ts using a 2% aga rose gel. Prime rs for this assay a re shown in T a b le 1. Acknowledgments We thank Alex Eldredge for help with cloning, the A.V. Maricq lab for providing Hek293 cells, and the Hollien lab for discussions. Author Contributions Conceived and designed the experiments: KAMJJP D G JC JH . Performed the experiments: KAMJJP D G JC JH . Analyzed the data: KAMJJP DG JC JH . Wrote the paper: KAM JH . PLOS ONE | www.plosone.org 7 September 2013 | Volume 8 | Issue 9 | e75723 34 Regulation o f Sumo mRNA during ER Stress References 1. M oore KA, Hollien J (2012) T he unfolded protein response in secretory cell function. A nnu Rev G enet 46: 165 183. 2. W alter P, R on D (2011) T he unfolded protein response: from stress pathw ay to hom eostatic regulation. Science 334: 1081 1086. 3. W ang S, K aufm an R J (2012) T h e im pact of the unfolded protein response on h um an disease. J Cell Biol 197: 857 867. 4. H arding H P, Z hang Y, R on D (1999) Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase. N ature 397: 271 274. 5. Shi Y, V attem KM , Sood R , An J , Liang J, et al. (1998) Identification and characterization o f pancreatic eukaryotic initiation factor 2 alpha-subunit kinase, PEK , involved in translational control. M ol Cell Biol 18: 7499 7509. 6. Sood R, Porter A C, M a K, Q uilliam LA, Wek R C (2000) Pancreatic eukaryotic initiation factor-2alpha kinase (PEK) hom ologues in hum ans, Drosophila m elanogaster and C aenorhabditis elegans that m ediate translational control in response to endoplasmic reticulum stress. B iochem J 346 Pt 2: 281 293. 7. H arding H P, Zhang Y, Bertolotti A, Zeng H , R on D (2000) Perk is essential for translational regulation and cell survival during the unfolded protein response. M ol C ell 5: 897 904. 8. H arding H P, Novoa I, Z hang Y, Zeng H , Wek R, et al. (2000) R egulated translation initiation controls stress-induced gene expression in m am m alian cells. M ol Cell. U nited States. pp. 1099 1108. 9. Credle JJ, Finer-Moore Js, Papa FR , Stroud RM , W alter P (2005) O n the m echanism ofsensing unfolded protein in the endoplasmic reticulum. Proc Natl Acad Sci U S A 102: 18773 18784. 10. C ox JS , Sham u C E, W alter P (1993) T ranscriptional induction of genes encoding endoplasmic reticulum resident proteins requires a transm em brane protein kinase. Cell 73: 1197 1206. 11. M ori K, M a W, G ething M J, Sam brook J (1993) A transm em brane protein with a cdc2+/C D C 28-related kinase activity is required for signaling from the E R to the nucleus. Cell 74: 743 756. 12. Calfon M , Zeng H, U rano F, Till JH , H ubbard SR, et al. (2002) IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 m RN A . N ature 415: 92 96. 13. Yoshida H , M atsui T , Y am am oto A, O kada T , M o ri K (2001) XBP1 m R N A is induced by AT F6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell. U nited States. pp. 881 891. 14. H aze K, Yoshida H , Y anagi H, Y ura T , M o ri K (1999) M am m alian transcription factor ATF6 is synthesized as a transm em brane protein and activated by proteolysis in response to endoplasmic reticulum stress. M o l Biol Cell 10: 3787 3799. 15. Lee A H, Iwakoshi N N , Glim cher LH (2003) XBP-1 regulates a subset of endoplasmic reticulum resident chaperone genes in the unfolded protein response. M ol Cell Biol 23: 7448 7459. 16. Lu PD, H arding H P, R on D (2004) Translation reinitiation at alternative open reading frames regulates gene expression in an integrated stress response. J Cell Biol. U nited States. pp. 27 33. 17. Hollien J, W eissman JS (2006) Decay o f endoplasmic reticulum-localized m RN As during the unfolded protein response. Science 313: 104 107. 18. Hollien J, Lin JH , Li H, Stevens N, W alter P, et al. (2009) R egulated Ire1- dependent decay of messenger RNAs in m am m alian cells. J C ell Biol 186: 323 331. 19. H an D , Lerner AG, V ande Walle L, U pton JP , X u W , et al. (2009) IR E 1alpha kinase activation m odes control alternate endoribonuclease outputs to determine divergent cell fates. Cell. U nited States. pp. 562 575. 20. G addam D , Stevens N , Hollien J (2013) Com parison ofm R N A localization and regulation during endoplasmic reticulum stress in Drosophila cells. M o l Biol Cell 24: 14 20. 21. Kim mig P, Diaz M , Zheng J, Williams C C , Lang A, et al. (2012) T he unfolded protein response in fission yeast m odulates stability ofselect m RN As to m aintain protein homeostasis. elife 1: e00048. 22. H ur KY, So JS , R uda V, Frank-Kamenetsky M , Fitzgerald K, et al. (2012) IR E 1alpha activation protects mice against acetam inophen-induced hepatotox-icity. J Exp M ed. U nited States. pp. 307 318. 23. Stephens SB, D odd R D , Lerner RS, Pyhtila BM, Nicchitta C V (2008) Analysis of m RN A partitioning betw een the cytosol and endoplasm ic reticulum compartments of m amm alian cells. M ethods M ol Biol 419: 197 214. 24. Calfon M , Zeng H, U rano F, Till JH , H ubbard SR, et al. (2002) IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 m RN A. N ature 415: 92 96. 25. Kaw ai T , Fan J , M azan-M am czarz K, Gorospe M (2004) Global m RN A stabilization preferentially linked to translational repression during the endoplasmic reticulum stress response. M o l Cell Biol. U nited States. pp. 6773 6787. 26. M ajum der M , H uang C , Snider M D , K om ar AA, T anaka J , et al. (2012) A novel feedback loop regulates the response to endoplasmic reticulum stress via the cooperation o f cytoplasmic splicing and m RN A translation. M ol Cell Biol. United States. pp. 992 1003. 27. K aw ahara T , Y anagi H, Y ura T , M o ri K (1998) U nconventional splicing of H A C 1 /E R N 4 m R N A required for the unfolded protein response. Sequence-specific and non-sequential cleavage of the splice sites. J Biol C hem 273: 1802 1807. 28. Iwakoshi N N , Lee A H, Vallabhajosyula P, O tipoby K L, Rajewsky K, et al. (2003) Plasma cell differentiation and the unfolded protein response intersect at the transcription factor XBP-1. N at Im m unol 4: 321 329. 29. G onzalez T N , Sidrauski C, Dorfler S, Walter P (1999) M echanism of non-spliceosomal m RN A splicing in the unfolded protein response pathw ay. Em bo j 18: 3119 3132. 30. Flotho A, Melchior F (2013) Sumoylation: a regulatory protein m odification in health and disease. A nnu Rev B iochem 82: 357 385. 31. C hen H, Q i L (2010) SUM O modification regulates the transcriptional activity ofX B P1. B iochem J 429: 95 102. 32. Jian g Z, Fan Q , Z hang Z, Zou Y, C ai R, et al. (2012) SENP1 deficiency promotes ER stress-induced apoptosis by increasing XBP1 SUM O ylation. Cell Cycle 11: 1118 1122. 33. G arreyJL , Lee YY, Au H H , B ush ellM ,Jan E(2010) H o stand viral translational m echanism s during cricket paralysis virus infection. J V irol 84: 1124 1138. 34. Lee YY, Cevallos R C , Ja n E (2009) An upstream open reading frame regulates translation o f G A D D 34 during cellular stresses that induce eIF2alpha phosphorylation. J Biol C hem 284: 6661 6673. 35. Kim mig P, Diaz M , Z h en gJ, Williams C C , Lang A, et al. (2012) T he unfolded protein response in fission yeast m odulates stability ofselect m RN As to m aintain protein homeostasis. Elife. U nited States. pp. e00048. 36. So JS , H ur KY, T a rrio M , R uda V, Frank-Kamenetsky M , et al. (2012) Silencing oflipid metabolism genes through IRE 1alpha-m ediated m R N A decay lowers plasma lipids in mice. Cell M etab 16: 487 499. PLOS ONE | www.plosone.org 8 September 2013 | Volume 8 | Issue 9 | e75723 CHAPTER 3 IRE1-MEDIATED DECAY IN MAMMALIAN CELLS RELIES ON mRNA SEQUENCE, STRUCTURE, AND TRANSLATIONAL STATUS Reprinted with permission from Molecular Biology of the Cell, 26(16), Moore, K.A., & Hollien, J., Ire1-mediated Decay of mRNA Relies on Sequence, Structure, and Translational Status, 2873-84, Copyright 2015 36 Irel-mediated decay in mammalian cells relies on mRNA sequence, structure, and translational status Kristin Moore and Julie Hollien Department of Biology and Center for Cell and Genome Science, University of Utah, Salt Lake City, UT 84112 MBoC I ARTICLE ABSTRACT Endoplasmic reticulum (ER) stress occurs when misfolded proteins overwhelm the capacity of the ER, resulting in activation o f the unfolded protein response (UPR). Ire1, an ER transmembrane nuclease and conserved transducer of the UPR, cleaves the mRNA encod­ing the transcription factor Xbp1 at a dual stem-loop (SL) structure, leading to Xbp1 splicing and activation. Ire1 also cleaves other mRNAs localized to the ER membrane through regu­lated Ire1-dependent decay (RIDD). We find that during acute ER stress in mammalian cells, Xbp1-like SLs within the target mRNAs are necessary for RIDD. Furthermore, depletion of Perk, a UPR transducer that attenuates translation during ER stress, inhibits RIDD in a sub­strate- specific manner. Artificially blocking translation of the SL region of target mRNAs fully restores RIDD in cells depleted of Perk, suggesting that ribosomes disrupt SL formation and/ or Ire1 binding. This coordination between Perk and Ire1 may serve to spatially and tempo­rally regulate RIDD. Monitoring Editor Gia Voeltz University of Colorado, Boulder Received: Feb 11, 2015 Revised: Jun 9, 2015 Accepted: Jun 16, 2015 INTRODUCTION The endoplasmic reticulum (ER) is the entry point for proteins tar­geted to the secretory pathway. Secreted proteins are translated from mRNAs localized to the cytosolic face of the ER membrane and enter the ER as nascent chains that are folded and modified before exiting the organelle. The flux of proteins through the ER varies ex­tensively among cell types and environments. Changes in this flux can result in ER stress, an imbalance between the load o f unfolded proteins entering the ER and the capacity of the organelle to fold and modify them efficiently. In metazoans, ER stress activates three ER transmembrane proteins: inositol-requiring 1 (Ire1), PKR-like en­doplasmic reticulum kinase (Perk), and activating transcription factor 6 (Atf6), which coordinate a signaling network known as the unfolded This a rtic le was p u b lis h e d o n lin e a h ea d o f p r in t in M BoC in Press (h ttp ://w w w .m o lb io lc e ll.o rg /c g i/d o i/1 0 .1 0 9 1 /m b c .E 1 5 -0 2 -0 0 7 4 ) on J u n e 24, 2015. A d d re s s co rre s p o n d e n c e to : J u lie H o llie n (h o llie n@ b io lo g y .u ta h .e d u ). A b b re v ia tio n s used: ER, e n d o p la sm ic re ticu lum ; Ire1, in o s ito l- re q u irin g en zyme 1; Perk, PKR-like e n d o p la sm ic re ticu lum kinase; qPCR, q u a n tita tiv e re a l-tim e PCR; RIDD, re g u la te d Ire 1 -d e p e n d e n t de cay ; SL, s tem lo o p ; UPR, u n fo ld e d p ro te in response; X b p 1 , X -b o x b in d in g p ro te in 1. © 2015 M o o re an d H o llie n . This a rtic le is d is tr ib u te d by The Am e rica n Society for Cell B io lo g y u n d e r license from th e author(s). Two m on th s a fte r p u b lic a tio n it is a va ilab le to th e p u b lic u n d e r an A ttrib u tio n -N o n c om m e rc ia l-S h a re A lik e 3.0 Un­p o r te d Cre a tive C om m o n s License (h ttp ://c re a tiv e c om m o n s .o rg /lic e n s e s /b y -nc-sa/3.0). "A SCB®," "T h e Am e rica n So c iety fo r Cell Bio lo g y® ," an d "M o le c u la r B io lo g y o f th e Cell®" are re g is te re d trad em ark s o f T he Am e rica n S o c iety fo r C ell Bio lo gy. protein response (UPR; Walter and Ron, 2011). Although ER stress results from a variety of pathological conditions, loss o f individual UPR sensors also affects normal development and physiology in several model organisms (Moore and Hollien, 2012). Perk directly phosphorylates eukaryotic translation initiation fac­tor 2 a (eIF2a), which leads to the attenuation of translation initia­tion and limits the protein-folding load on the ER (Harding et al., 1999). This phosphorylation event also leads to translational up-regulation of certain proteins, including activating transcription fac­tor 4 (Atf4) (Harding et al., 2000). Concurrently, Ire1 oligomerizes in response to ER stress, activating its cytosolic nuclease domain (Li et al., 2010), and cleaves the mRNA encoding X-box binding pro­tein 1 (Xbp1). This cleavage occurs at two specific sites in a dual stem-loop (SL) structure (Yoshida et al., 2001; Calfon et al., 2002). The resulting 5' and 3' fragments are then ligated, forming a spliced transcript encoding the active transcription factor, which, together with other UPR transcription factors, up-regulates numerous genes that increase the capacity of the secretory pathway (Travers et al., 2000; Harding et al., 2003). Ire1 is also responsible for the cleavage of other ER-localized mRNAs, leading to their degradation through regulated Ire1-de-pendent decay (RIDD; Hollien and Weissman, 2006; Hollien et al., 2009). RIDD was originally observed in Drosophila melanogaster S2 cells, where a large number o f mRNAs associated with the ER are degraded during ER stress (Hollien and Weissman, 2006). RIDD is important for Drosophila eye development, confirming a Volume 26 August 15, 2015 Supplemental Material can be found at: http://www.molbiolcell.oig/content/suppl/2015/06/21/mbc.E15-02-0074v1.DC1.html 2873 37 physiological role for this pathway in vivo (Coelho et al., 2013). In S2 cells, ER localization of an mRNA is both necessary and sufficient for its degradation by RIDD (Gaddam et al., 2013). However, excep­tions to this rule exist. For example, the Drosophila transcript en­coding small ubiquitin-modifier (Sumo) is targeted to RIDD despite localizing to the cytosol. This mRNA requires an Xbp1-like SL in its coding region to be degraded by Ire1 (Moore et al., 2013). In addi­tion, for unknown reasons, RIDD of Sumo requires Perk (Moore et al., 2013), even though Perk depletion does not appear to gener­ally affect RIDD in S2 cells (Hollien and Weissman, 2006). RIDD also occurs in mammalian cells (Han et al., 2009; Hollien et al., 2009). Activation of Ire1 through overexpression in cultured cells or tissue-specific Xbp1 mutations in mice, which result in hyper­activation of Ire1, induces broad cleavage of ER- localized mRNAs (Han et al., 2009; So e t al., 2012). However, during chemical induc­tion o f ER stress in both mammalian cell culture and mice, the mag­nitude of degradation and number o f mRNAs targeted to the path­way are more limited than in S2 cells (Hollien et al., 2009; So et al., 2012). This restriction o f RIDD substrates suggests a dependence on additional factors or sequence elements beyond mRNA localiza­tion to the ER. One likely requirement is the presence an Xbp1-like SL within the target mRNA sequence. These SLs are more prevalent in mammalian RIDD targets than in those o f D. melanogaster (Gaddam et al., 2013). Furthermore, mutation of a conserved gua­nine (G) within the loop blocks mRNA cleavage by human Ire1 in vitro (Hur et al., 2012) and also affects the regulation of at least one RIDD target in human cells (Bright et al., 2015). In this study, we investigate the mechanism and substrate selectiv­ity of RIDD during acute, chemically induced ER stress in mammalian cells and describe an unexpected role for Perk in the RIDD pathway. RESULTS RIDD targeting in different cell types Previous work in mammalian cells suggested that the extent o f deg­radation of RIDD targets in the absence o f Ire1 overexpression is fairly small, on the order of twofold (Hollien et al., 2009). We first asked whether this result was cell-line dependent. We treated sev­eral different mammalian cell lines with chemical inducers o f ER stress: dithiothreitol (DTT), which blocks disulfide bonding; thapsi-gargin (Tg), which depletes ER calcium reserves; and tunicamycin (Tm), an inhibitor o f N-linked glycosylation. We then used quantita­tive real-time PCR (qPCR) to measure the stress-dependent changes in the relative levels o f mRNAs that were previously identified as RIDD targets in mouse fibroblasts (Hollien et al., 2009; Figure 1, A and B). Note that because our qPCR expression data are inherently ratiometric, we use a log 2 scale throughout this article, meaning that a unit of 1.0 refers to a twofold change in expression. Xbp1 splicing was nearly complete in all stress conditions tested (Figure 1C). However, the extent o f RIDD targeting varied among individual mRNAs and among the different cell types (Figure 1, A and B). In both human cell lines tested (Hek293 and Hep G2), Blos1 was degraded during ER stress, but other mouse RIDD targets were either not degraded (Scara3) or not expressed to detectable levels (Col6a1 and Hgsnat; Figure 1B). O f note, the mouse Scara3 transcript contains an Xbp1 -like SL, but the human transcript does not. We observed the most robust RIDD in MC3T3-E1 cells, a pre­osteoblast cell line derived from mouse calvaria (Kodama et al., 1981), and therefore used these cells for further study. Using small interfering RNA (siRNA)-mediated silencing, we verified that the down-regulation o f RIDD targets was Ire1 dependent and Xbp1 in­dependent (Figure 1, D-G). Xbp1-like stem loops are necessary for RIDD in mammalian cells To test the importance of Xbp1 -like SLs for RIDD in a cellular con­text, we used a reporter-based approach. We created plasmids ex­pressing the coding sequences (CDSs) o f the mouse RIDD targets Hgsnat and Blos1 with vector-derived 5' and 3' untranslated regions (UTRs) and stably transfected them into MC3T3-E1 cells. After treat­ment o f cells with or without DTT (2 mM, 4 h), we measured the rela­tive abundance o f the reporter mRNAs by qPCR, using primers that spanned the CDSs and reporter UTRs and therefore did not amplify the endogenous transcripts. As expected, the mRNAs expressed from both of these plasmids were down-regulated during ER stress (Figure 2, B and C), indicating that the CDS is sufficient for RIDD of these transcripts. The CDSs o f Blos1 and Hgsnat contain Xbp1 -like SLs (Figure 2A), as defined by a seven-nucleotide (nt) loop with the four con­served residues essential for Xbp1 splicing (Calfon et al., 2002) and a stem o f at least four consecutive base pairs (allowing for AU, GC, and GU pairs). To test whether these sites were important for RIDD, we mutated the putative Ire1 cleavage site G to cytosine (C) and measured reporter degradation. For Blos1, this mutation, as well as mutation of a second conserved loop residue, completely ablated degradation (Figure 2C). For Hgsnat, mutation o f the putative cleavage site in one of the two SLs (Hgsnat SL #1) blocked RIDD (Figure 2B), whereas the corresponding mutation in a second SL (Hgsnat SL #2) did not affect its degradation during ER stress (Figure 2B). The stem o f Hgsnat SL #2 is shorter and has fewer GC pairs than Hgsnat SL #1 (Figure 2A), suggesting that the stability of the stem is important for RIDD. To test this, we made mutations that disrupted the base-pairing o f the Xbp1-like SL o f our Blos1 reporter. These mutations blocked RIDD targeting (Figure 2D). Restoring base-pairing within the putative stem region with complementary mutations that preserved the GC content o f the SL restored RIDD. However, mutations that replaced GC pairs with AU pairs prevented RIDD (Figure 2D). Together these results indicate that both the se­quence and stability o f Xbp1 -like SLs are important for RIDD in mouse cells, as suggested previously for human cells (Bright et al., 2015). To ensure that reporter expression levels did not influence RIDD, we measured the level of overexpression o f Blos1 mRNA in our re­porter cell lines. Total Blos1 mRNA abundance was measured by qPCR using primers that annealed within the CDS of the Blos1 tran­script and therefore amplified both endogenous and reporter mRNAs. The overexpression of the Blos1 reporter mRNAs varied from -4 - to 32-fold above endogenous Blos1 levels, which were measured using a control cell line transfected with green fluorescent protein (GFP). However, there was no correlation between reporter expression level and degradation during ER stress. Furthermore, we created two independent cell lines that expressed WT Blos1 to dif­ferent levels (4- vs. 32-fold overexpression) and observed no differ­ence in the extent of the reporter mRNA degradation during stress (Figure 2E). An Xbp1-like stem loop is sufficient to target GFP mRNA to RIDD To determine whether an Xbp1 -like SL is sufficient to induce deg­radation o f a transcript not normally targeted to the RIDD path­way, we used reporters expressing either GFP or an ER-targeted GFP (ssGFP) containing the signal sequence from Drosophila Hsp70-3. In S2 cells, this ssGFP mRNA reporter (but not the cyto­solic GFP mRNA) is degraded by RIDD (Gaddam et al., 2013). Similarly, rat cells overexpressing Ire1 degrade an ER-targeted 2874 I K. Moore and J. Hollien Molecular Biology o f th e Cell 38 A 1 MC3T3-E1 DTT Tg IH i ■ Bl1osl -2 ^ T t Scara3 1 { Col6a1 - 3 Hgsnat C MC3T3-E1 Hek293 Unt DTT Tg Unt DTT Tm % Spliced: 3 86 74 24 73 72 SD: 2 2 6 6 2 8 MC3T3-E1 siRNA: Neg MC3T3-E1 siRNA: Neg Xbp1 -4 1 1 -1 -2 Hek293 HepG2 Min6 DTT Tm DTT DTT Tg siRNA: Neg Ire1 DTT: - % Spliced: 3 89 SD: 1 2 G >o ur=y ccC 50 I FIGURE 1: The RIDD pathway varies across mammalian cell lines. For all abundance measurements, mRNA was reverse transcribed and measured by qPCR and data were normalized to the housekeeping control mRNA Rpl19. The legend in A applies to bar graphs in A, B, D, and F. (A) Relative mRNA levels o f RIDD targets in mouse MC3T3-E1 cells treated with either DTT (2 mM) or Tg (2 pM) for 4 h to induce ER stress. (B) Relative mRNA levels of Blosl (black) and Scara3 (gray) in the indicated cell lines treated with DTT (2 mM), Tm (2.5 pg/ml), or Tg (2 pM) for 4 h. Note that Scara3 was not expressed strongly enough in Min6 cells to measure mRNA levels. (C) Samples from A and B were amplified by PCR using primers surrounding the Xbp1 splice sites. Shown are representative agarose gels with the spliced and unspliced products and averages and SDs of the percentage spliced Xbp1 for three independent experiments. (D) Relative mRNA levels of RIDD targets in MC3T3-E1 cells transfected with either Neg (negative control) or Ire1 siRNAs and then treated with or without DTT (2 mM, 4 h). (E) Xbp1 splicing in samples from D. (F, G) Relative mRNA levels of RIDD targets (F) and Xbp1 (G) in MC3T3-E1 cells transfected with Neg or Xbp1 siRNAs and then treated with or without DTT (2 mM, 4 h). Shown in all panels are the averages and SDs from two (Hek293 cells, Tm treatment) or three (all other panels) independent experiments. Ut, untreated. To confirm that reporter mRNAs were correctly localized, we used detergent frac­tionation to separate membrane-associated versus cytosolic mRNAs, as described previ­ously (Stephens et al., 2008; Gaddam et al., 2013). As expected, ssGFP mRNA fraction­ated predominately with the membrane, along with a membrane-bound control, BiP In contrast, GFP mRNA fractionated pre­dominantly with the cytosol, similarly to the control glyceraldehyde 3-phosphate dehy­drogenase (Gapdh; Figure 2G). Addition of the Blos1 SL to the 3' UTR of either GFP reporter mRNA (GFP-SLUTR or ssGFP-SLutr) resulted in its Ire1-dependent degradation during ER stress (Figure 2, F and H), indicating that an Xbp1-like SL is sufficient to target GFP mRNA to RIDD. Addition of the SL also resulted in a partial shift of GFP mRNA localization toward the membrane fraction (Figure 2G), suggesting that the SL alone may mediate membrane association. Xbp1-like SLs do not predict RIDD targets generally On the basis of these results, we hypothe­sized that endogenous mRNAs with Xbp1- like SLs would be RIDD targets. Previous work in mammalian cells has not led to a comprehensive list o f RIDD targets, in part because transcription is highly regulated during ER stress and complicates the global analysis o f mRNA degradation. Therefore we carried out a limited test of our hypoth­esis by blocking transcription in MC3T3-E1 cells using actinomycin D (2 pg/ml) and then measuring the relative degradation of sev­eral mRNAs in the presence and absence of DTT (1 mM, 4 h). We chose mRNAs that met the following criteria: 1) they were expressed in MC3T3-E1 cells (Nabavi et al., 2012), 2) they were associated with Gene Ontology terms indicating ER, Golgi, lysosome, plasma membrane, or extracellular localiza­tion of the encoded protein, and 3) they contained strong Xbp1-like SLs with at least three GC base pairs in the stem. Surpris­ingly, none o f the 10 mRNAs we measured was degraded more strongly during ER stress (Figure 3). These results indicate that although the presence of an Xbp1-like SL is sufficient to target GFP mRNA to RIDD, it is not generally sufficient to target endoge­nous mRNAs to RIDD and additional target­ing features must exist. GFP mRNA (Han et al., 2009). However, in MC3T3-E1 cells, nei­ther GFP nor ssGFP transcripts were down-regulated during ER stress (Figure 2F), supporting the idea that mRNA membrane as­sociation is not sufficient for RIDD in mammalian cells during acute ER stress. Perk-mediated attenuation of translation is important for RIDD Previously we determined that the noncanonical Drosophila RIDD target Sumo relies on both a SL and the presence of Perk to be de­graded during ER stress (Moore et al., 2013). To determine whether B 0 D E re1 F 100 0 Volume 26 August 15, 2015 Ire1-mediated decay in mammalian cells I 2875 39 GCA A G 717C;G^ C C U A GC GC GC A U GCA U G 11111111C G UU UG CG CG U A mHgsnat mHgsnat SL #1 SL #2 GCA U G 335588C G Cu GCA C G 476 U A GCA CG C 502CU AC AU CG GC CG CG AU AU GC GC CG G A U A GC GC CG __UA mBlos1 mXbp1 C D z% p•- E a °w £<0 03 i5 357 365 GC GG A U A U CG CC GC GG 354 368 ^ GA A U C U GA Cell Line: WT1 WT2 Ex Level: Low High I -2 |Gapdh Bip GFP GFP Reporter: & H siRNA: Neg Ire1 Neg Ire1 FIGURE 2: An Xbp1-like stem loop is necessary for RIDD and sufficient to induce GFP mRNA degradation in mammalian cells during ER stress. (A) RNA SLs from mouse Hgsnat, Blos1, and Xbp1. Red lettering indicates Xbp1 loop residues conserved across species, and arrows indicate putative Ire1 cleavage sites. Numbering is relative to mRNA translation start sites. (B-F) We stably transfected MC3T3-E1 cells with plasmids expressing reporter mRNAs, incubated cells with or without DTT (2 mM, 4 h), and measured relative abundances of the mRNA reporters by qPCR relative to the housekeeping control Rpl19. (B, C) Reporters expressing the mHgsnat (B) or mBlos1 (C) coding sequences (CDSs) with and without mutations in the Xbp1-like loop. (D) Reporters expressing the Blos1 CDS with and without mutations in the stem region of the Xbp1-like SL. Blue lettering indicates mutated residues. (E) Changes in mRNA abundance for the WT Blos1 reporter in two independent cell lines (WT1 and WT2) after DTT treatment. The cell lines differ only in their levels of reporter expression (Ex), either low (4-fold above endogenous levels) or high (32-fold above endogenous levels). (F) Reporters expressing GFP or an ER-targeted GFP (ssGFP) with and without the mBlos1 SL inserted 15 nt downstream of the stop codon. (G) Fraction membrane (membrane/total) of mRNAs from MC3T3-E1 cells stably expressing different GFP reporters measured by digitonin fractionation followed by qPCR. (H) We depleted Ire1 from stably transfected cells and then measured reporter mRNA levels as in B-F. Shown are averages and SDs from three or more independent experiments. *p < 0.05, two-tailed unpaired t test. Ut, untreated. E F 2876 I K. Moore and J. Hollien Molecular Biology o f th e Cell 40 mRNA:Tmem183a Caml STT3 Y ifla G a ln tll Cog2 Gorasp2Slc20a2 Plscr3 Sdc4 Blosl localization ER ER/Golgi Golgi Golgi Golgi of protein: membrane ER PM PM ECM Lysosome Xbp1-like SL location: 3'UTR CDS CDS 3'UTR CDS CDS 3'UTR 3'UTR 3'UTR 3'UTR CDS FIGURE 3: Xbp1-like SLs are not sufficient to target endogenous mRNAs to RIDD. MC3T3-E1 cells were treated with 1 mM DTT, 2 pg/ml actinomycin D (Act), or both for 4 h. We then measured relative mRNA levels of noted transcripts by qPCR. Transcripts were chosen based on their predicted localization to the ER (based on Gene Ontology term analysis) and the presence of Xbp1-like SLs, defined as 1) a seven-membered loop with the four conserved residues (as in Figure 2A), and 2) a stem of at least 5 base pairs including three GC pairs. The verified RIDD target Blos1 was also measured as a control. Shown are averages and SDs from two independent experiments. ECM, extracellular matrix; PM, plasma membrane; Ut, untreated. Perk plays a role in the mammalian RIDD pathway, we transfected MC3T3-E1 cells with either a negative control (Neg) siRNA or a com­bination of four siRNAs targeting Perk and then induced ER stress with either DTT or Tg. Depletion o f Perk strongly inhibited RIDD of both Blos1 and Col6a1 and partially inhibited RIDD o f Scara3 (Figure 4, A-C). RIDD o f Hgsnat, however, was not affected by Perk knock­down (Figure 4B; see next section). We saw similar effects when two distinct Perk siRNAs were transfected individually (Supplemental Figure S1, A and B). Finally, Perk knockdown also inhibited RIDD of Blosl in Hek293 cells (Figure 4, D and E), indicating a conserved effect across species. In addition to phosphorylating eIF2a and thereby attenuating translation initiation, Perk also phosphorylates other targets, includ­ing Nrf2 (Cullinan et al., 2003) and diacylglycerol (Bobrovnikova- Marjon et al., 2012). To determine which aspect of Perk signaling is important for RIDD, we used integrated stress response inhibitor (ISRIB), a chemical that blocks translation attenuation during ER stress but does not affect the phosphorylation of eIF2a or other Perk targets (Sidrauski et al., 2013). ISRIB significantly inhibited RIDD (Figure 4F and Supplemental Figure S1C). Therefore Perk's ability to attenuate translation during ER stress is important for RIDD. Accord­ingly, artificially attenuating translation with the initiation inhibitor harringtonine fully restored RIDD in cells depleted o f Perk (Figure 4G). Knockdown of Perk also resulted in a 25-40% decrease in Xbp1 splicing in response to ER stress (Figure 4H), an effect noted previ­ously (Majumder et al., 2012). As with RIDD, inhibition o f translation initiation by harringtonine fully restored Xbp1 splicing. Harringtonine did not cause a general increase in Ire1 activity, as harringtonine treatment alone actually led to a reduction in constitutive Xbp1 splic­ing in unstressed cells (Figure 4H). Overall these results indicate that attenuating translation initiation during ER stress allows for more ef­ficient RIDD and Xbp1 splicing. RIDD relies on the translational status of target mRNAs There are two general possibilities for why Perk-mediated transla­tion attenuation is important for RIDD: either halting translation allows for depletion of an unstable factor that is important for RIDD, or depletion of ribosomes from the RIDD target mRNA allows it to be degraded. The fact that Perk knockdown had varying effects on different mRNAs suggests that translation attenuation o f the RIDD target itself is of primary importance. In support o f this model, we noted that Hgsnat, the RIDD target that was insensitive to Perk de­pletion, has two large clusters o f rare codons near the 5' end o f the transcript, which may act to constitutively reduce translation and al­low for Hgsnat mRNA degradation during ER stress, regardless of Perk activity. Rare codon clusters were not found in the 5' regions of Perk-sensitive RIDD targets (see later discussion of Figure 7A). To test directly whether the translational status of mRNA targets is important for RIDD, we asked whether limiting translation o f Perk-sensitive RIDD targets caused them to become Perk insensitive. We introduced translation-stalling SLs (Vattem and Wek, 2004) 6 nt up­stream o f the translation start site within the 5' UTRs of two RIDD reporters, one expressing the Blos1 CDS (as in Figure 2C) and one expressing ssGFP with the Blos1 SL inserted in the CDS, 68 nt up­stream o f the stop codon (ssGFP-SLCDS). We then stably transfected these reporters into MC3T3-E1 cells and tested for RIDD as de­scribed. In Neg siRNA-treated cells, these reporter mRNAs were degraded similarly to their wild-type counterparts. However, unlike the wild-type reporters, degradation o f the translationally stalled re­porters was unaffected by depletion of Perk (Figure 5, B and C). We conclude that attenuating translation o f the target itself is important for degradation by RIDD. Ribosome binding to an mRNA may limit Ire1's access, thus in­hibiting cleavage and subsequent degradation of the mRNA. To test this idea we used cycloheximide (Chx), a translation elongation in­hibitor that stalls ribosomes along mRNAs without releasing them. Chx significantly inhibited RIDD o f both Blos1 and Col6a1 but not Scara3 (Figure 5D), correlating with the relative sensitivities o f these mRNAs to Perk depletion. These results indicate that attenuating translation initiation and essentially reducing the number o f ribo­somes on an mRNA enhances RIDD, whereas blocking translation elongation by locking ribosomes on an mRNA inhibits RIDD. Volume 26 August 15, 2015 Ire1-mediated decay in mammalian cells I 2877 41 A MC3T3-E1 B MC3T3-E1 C MC3T3-E1 a) " 100 siRNA: 0 Neg Perk Q_ d)> ' 0TO) cr ta 50 z< a: £ siRNA: 0 siRNA: Neg Perk £ o D HEK293 E HEK293 Blosl Col6a1 Scara3 Hgsnat F * n.s. o 100 - siRNA: Neg Perk H I H ■ \ 'J 50 l E £ BP H i * -1 I cc cc o H H siRNA: ^ = O .Q -J TO G siRNA Har DTT Neg Perk <Z tao OCff -2 -4 1----------T ISRIB DTT: <Z T<OD a: 2 E i.1 I'T T T P n.s. * + o * * * H siRNA Har DTT: Neg Perk + - + - + + + - + - + + % spliced: 26 16 73 78 29 18 57 77 SD: 5 1 2 2 1 2 4 2 FIGURE 4: Translation attenuation mediated by Perk is important for RIDD. (A-C) We transfected MC3T3-E1 cells with Neg or Perk siRNAs and then incubated them with and without 1 mM DTT (B) or 2 pM Tg (C) for 4 h. We then measured the percentage of Perk mRNA remaining (A) and RIDD target mRNA levels (B, C). The legend in B applies to bar graphs in B, C, F, and G. Asterisks represent significant differences between Neg and Perk siRNA-treated samples. (D, E) Perk (D) or human Blosl (E) mRNA measured from Neg or Perk siRNA-treated Hek293 cells with or without DTT (2 mM, 4 h). (F) Blosl (black bars) and Col6a1 (gray bars) mRNA levels in MC3T3-E1 cells treated with 500 nM ISRIB, 1 mM DTT, or both for 4 h. (G, H) Blos1 (black bars) and Col6a1 (gray bars) mRNA levels (G) and Xbp1 splicing (H) from control or Perk-depleted MC3T3-E1 cells treated with 1 pM harringtonine (Har), 1 mM DTT, or both for 4 h. All mRNA levels were determined by qPCR. Shown are averages and SDs from at least three independent experiments. *p < 0.05, two-tailed paired t test. Ut, untreated. 2878 I K. Moore and J. Hollien Molecular Biology o f th e Cell 42 A Stop codon C S= R□ I H ^QQ CTCD -2 o> C ^O £3 -3 f t Xbp1-like SL Blosl O 5' translation stalling SL (TS) TS-Blos1 B < ZD P£ a Q_ <D L_ O o £ i l SL . siRNA: J I----- ■ Neg siRNA: ■ Neg Perk Perk Col6a1 D . Blos1 Scara3 $ ® 1 i ----------------;---------------------------- 0 p i f m I J- cB"0 -3 * DTT I----- siRNA: - § DTT + Chx 3 *<