The Exozyme Model: a New Paradigm of Exosome Subunit Activity Revealed by Diverse and Distinct Substrate Specificities of Exosome Subunits in Vivo

THE EXOZYME MODEL: A NEW PARADIGM OF EXOSOME SUBUNIT ACTIVITY REVEALED BY DIVERSE AND DISTINCT SUBSTRATE SPECIFICITIES OF EXOSOME SUBUNITS IN VIVO

DANIEL LOUIS KISS

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Dissertation advisor: Erik D. Andrulis, Ph.D.

Department of Molecular Biology and Microbiology

CASE WESTERN RESERVE UNIVERSITY

May 2010

We hereby approve the thesis/dissertation of

______Daniel L. Kiss______

Candidate for the Ph.D.______degree*.

(signed)Jonatha Gott______

(Chair of the committee)

Donal Luse______

Alan Tartakoff______

Erik Andrulis______

(Thesis Advisor)

(date) 11/12/09______

*We also certify that written approval has been obtained for any

proprietary material contained therein.

This work is dedicated to my parents, Kamill and Margit Kiss, who came to this country from Hungary with nothing, yet were able to provide me and my siblings

with everything we needed. Their personal sacrifices over the many years did not go unnoticed or unappreciated. It is their examples that I try to emulate each

and every day.

Table of Contents Table of Contents ...... 4

List of Tables ...... 9

List of Figures ...... 10

List of Abbreviations ...... 13

Chapter 1: ...... 20

Introduction ...... 20

1.A. RNA and the importance of RNA turnover ...... 20

1.B. The core exosome complex paradigm ...... 24

1.B.1. The subunits of the exosome complex ...... 25

1.B.2. Arrangement of exosome subunits within a complex ...... 29

1.B.3. The exosome complex has nuclear and cytoplasmic cofactors ...... 32

1.B.4. Functions of the exosome complex ...... 33

1.B.4.a. 3’ end processing of stable RNAs ...... 33

1.B.4.b. RNA surveillance and turnover ...... 35

1.B.4.b.1. Nuclear surveillance of stable RNAs ...... 36

1.B.4.b.2. Nuclear surveillance of unstable non-coding RNAs ...... 37

1.B.4.b.3. Nuclear RNA surveillance of mRNAs ...... 38

1.B.4.b.4. Cytoplasmic mRNA surveillance ...... 39

1.B.4.b.5. Nonsense-mediated decay ...... 40

1.B.5. Mechanism of core exosome function ...... 41

1.B.6. Summary ...... 43

1.C. Cracks in the core exosome complex paradigm ...... 44

1.C.1. Archaeal crystal structures ...... 44

1.C.2. Distinct complexes are purified from yeast and fly ...... 46

1.C.3. Exosome subunits have distinct localization patterns ...... 50

1.C.4. Functional evidence ...... 51

1.C.4.a. Microarray studies show exosome subunits survey distinct RNAs

in vivo ...... 52

1.C.4.b. Targeted studies show exosome subunit mutants accumulate

distinct RNA intermediates in vivo ...... 56

1.C.5. Individual exosome subunits have complex independent functions in

vivo ...... 58

1.C.6. Some exosome subunits form sub-complexes in vivo ...... 60

1.C.7. Summary ...... 60

1.D. Hypothesis ...... 61

Chapter 2: ...... 64

Genome-wide Analysis Reveals Distinct Substrate Specificities of Rrp6, Dis3, and Core Exosome Subunits ...... 64

2.A. Abstract ...... 65

2.B. Introduction ...... 67

2.C. Materials and methods ...... 70

2.C.1. T7 primers and double-stranded RNA preparation...... 70

2.C.2. S2 cell tissue culture and dsRNA treatment...... 72

2.C.3. CuSO4 induction of exo-FLAG expressing cell lines and cell counting

...... 72

2.C.4. Western blotting...... 73

2.C.5. Microarray experiment design ...... 74

2.C.6. Microarray controls and statistics ...... 75

2.C.7. Grouping of pathways ...... 76

2.C.8. UTR analysis ...... 76

2.C.9. NMD analysis ...... 76

2.D. Results ...... 77

2.D.1. Exosome subunits can be depleted in S2 cells ...... 77

2.D.2. Depletion of select exosome subunits affects the stability of a subset

of other subunits ...... 77

2.D.3. Exosome subunits are not required for cell proliferation ...... 80

2.D.4. Gene expression arrays identify exosome subunit surveyed mRNAs 82

2.D.4.a. DesMAPs of individual subunits do not reveal a core set of

transcripts surveyed by exosome subunits ...... 84

2.D.4.b. Structurally similar subunits have distinct desMAPs ...... 93

2.D.6. The UTRs of affected genes are longer than transcriptomic average.

...... 101

2.D.6.a. Length of dsRNA is not a predictor of UTR length ...... 103

2.D.6.b. Certain subunits stabilize transcripts with longer UTRs ...... 107

2.D.6.d.1. UTR patterns in increased and decreased transcripts ...... 110

2.D.7. Certain subunits regulate the stability of mitochondrial transcripts . 116

2.D.8. NMD targeted transcripts are enriched in certain desMAPs ...... 125

2.E. Discussion ...... 132

Chapter 3: ...... 141

Levels of Heat Shock Protein 70 and 26 mRNAs upon Recovery from Heat

Shock are Differentially Affected by Exosome Subunit Depletion ...... 141

3.A. Abstract ...... 142

3.B. Introduction ...... 144

3.C. Methods and materials ...... 147

3.C.1. RNAi depletion coupled with heat shock & RNA harvesting ...... 147

3.C.2. Probe preparation for S1 nuclease protection assays ...... 148

3.C.3. S1 nuclease protection assay ...... 148

3.D. Results ...... 150

3.D.1. Detection of hsp70 mRNA via S1 nuclease protection assays...... 150

3.D.2. Tracking hsp70 mRNA levels upon recovery from heat shock...... 153

3.D.3. Depletion of exosome subunits has different effects on hsp70 mRNA

levels upon recovery from heat shock...... 153

3.D.3.a. Depletion of Rrp4 and Rrp40 causes continued accumulation of

hsp70 mRNA after recovery from heat shock ...... 156

3.D.4. Levels of hsp26 mRNA upon recovery from heat shock are altered by

exosome subunit depletion...... 161

3.E. Discussion ...... 165

3.F. Conclusions ...... 172

3.G. Acknowledgements ...... 173

Chapter 4: ...... 174

General discussion and future directions ...... 174

List of references ...... 179

List of Tables

Table 1. Types of RNA ...... 21

Table 2. Exosome subunits...... 27

Table 3. Distinct exosome complexes are purified from S. Cerevisiea...... 48

Table 4. T7 oligonucleotide primers for dsRNA synthesis...... 71

Table 5. Numbers of transcripts changed at least two-fold in each desMAP...... 87

Table 6. mRNAs affected in all 9 desMAPs...... 91

Table 7. Number of transcripts shared in all binary comparisons...... 94

Table 8. Lengths of dsRNA used...... 105

Table 9. Mitochondrial genes affected by exosome subunit depletion...... 118

Table 10. The per desMAP distribution of mRP transcripts...... 122

Table 11. Summary of mitochondrial transcripts in each desMAP...... 123

Table 12. Some NMD-surveyed transcripts are conserved among desMAPs. . 127

Table 13. Gene groups affected (per desMAP) for increased transcripts...... 130

Table 14. Grouped analysis of decreased transcripts...... 133

Table 15. Oligonucleotide probes for S1 nuclease protection assays ...... 149

List of Figures

Figure 1. mRNA regulation points ...... 23

Figure 2. Domain maps of Drosophila melanogaster exosome subunits ...... 28

Figure 3. Core exosome complex structure...... 31

Figure 4. Proposed mechanism for exosome complex function...... 42

Figure 5. Archaeal exosome structures...... 45

Figure 6. Microarray studies show exosome subunits affect distinct RNAs...... 53

Figure 7. Exosome subunits are effectively depleted using dsRNAs...... 78

Figure 8. Some exosome subunits are co-depleted when certain subunits are targeted with dsRNAs...... 79

Figure 9. Most exosome subunits are non-essential in D. melanogaster S2 cells.

...... 81

Figure 10. Overexpression of certain exosome subunits hinders cell proliferation.

...... 83

Figure 11. mRNAs affected by exosome subunit depletion are predominantly increased...... 85

Figure 12. desMAP size varies by subunit...... 86

Figure 13. The affected mRNAs in the desMAPs of different subunits vary in both identity and effect...... 89

Figure 14. Most mRNAs are found in few desMAPs...... 90

Figure 15. Weighted distribution of all transcripts in multiple desMAPs...... 92

Figure 16. Most of the transcripts in the desMAPs of S1 domain subunits are not shared between the three subunits...... 95

Figure 17. Transcripts present in all three S1 domain desMAPs are most likely to be in other exosome subunit desMAPs...... 97

Figure 18. Transcripts present in the desMAPs of RNase PH subunits are more conserved than those of S1 domain subunits...... 98

Figure 19. Transcripts present in all RNase PH-domain desMAPs are more likely to be shared with other exosome subunit desMAPs...... 100

Figure 20. The desMAPs of Dis3, Rrp6 and Rrp47 show little overlap...... 102

Figure 21. The UTRs of affected transcripts are longer that the transcriptome- wide average in D. melanogaster...... 104

Figure 22. dsRNA lengths do not predict UTR lengths...... 106

Figure 23. The UTR length distributions of affected transcripts favor longer UTRs.

...... 108

Figure 24. The UTR distributions of Rrp4's and Rrp6's desMAPs diverge from the transcriptomic average...... 109

Figure 25. The 5' UTR length distributions of increased transcripts ...... 111

Figure 26. The 5' UTR length distribution of decreased transcripts...... 112

Figure 27. The 3' UTR length distribution of increased transcripts...... 114

Figure 28. The 3' UTR length distribution of decreased transcripts...... 115

Figure 29. Mitochondrial transcripts are enriched in certain desMAPs...... 120

Figure 30. Most michondrial genes are not conserved between subunits...... 121

Figure 31. NMD targeted transcripts are enriched in certain desMAPs...... 126

Figure 32. Many mRNAs increased by exosome subunit depletion code for proteins working in similar pathways...... 131

Figure 33. mRNAs decreased by exosome subunit depletion have few epistatic interactions...... 134

Figure 34. Detection hsp70 mRNA via S1 nuclease protection assays...... 152

Figure 35. Tracking hsp70 mRNA levels in GFP dsRNA treated cells...... 154

Figure 36. Depleting Dis3, Rrp46, or Rrp6 has little or no effect on hsp70 mRNA levels upon recovery from heat shock...... 155

Figure 37. Depletion of either Mtr3 or Ski6 increases hsp70 mRNA levels...... 157

Figure 38. Depletion of Csl4 or Rrp47 affects hsp70 mRNA levels...... 158

Figure 39. Depletion of Rrp4 or Rrp40 causes hsp70 mRNA to increase after removal of heat stress...... 159

Figure 40. hsp70 mRNA continues to accumulate in Rrp4 and Rrp40 depleted cells...... 160

Figure 41. Tracking hsp26 mRNA levels in GFP dsRNA treated cells ...... 162

Figure 42. Depletion of certain exosome subunits yields different results when examining hsp70 and hsp26...... 164

Figure 43. hsp26 mRNA behaves similarly to hsp70 mRNA in Dis3, Rrp47, Rrp4, and Rrp40 depleted cells...... 166

List of Abbreviations mRNA Messenger RNA rRNA Ribosomal RNA tRNA Transfer RNA tmRNA Transfer-messenger RNA snRNA Small nuclear RNA snoRNA Small nucleolar RNA

SmY SmY RNA gRNA Guide RNA

RNase Ribonuclease aRNA Antisense RNA long ncRNA Long noncoding RNA miRNA Micro RNA piRNA Piwi-interacting RNA siRNA Small Interfering RNA tasiRNA Trans-acting siRNA rasiRNA Repeat associated siRNA

CUTs Cryptic unstable transcripts

UNTs Upstream noncoding transcripts

PROMPTs Promoter upstream transcripts

SUTs Stable unannotated transcripts sRNAs Short RNAs

PASRs Promoter-associated sRNAs

TASRs Termini-associated sRNAs lRNAs Long RNAs

Rrp Ribosomal RNA proccessing

Dis3 Defective in sister chromatid disjoining 3

Csl4 Cep 1 synthetic lethal 4

Mtr mRNA transport mutant mRP Mitochondrial ribosomal protein

Ski Superkiller min Minutes

RNAP RNA polymerase desMAP Depleted exosome subunit microarray profile dsRNA Double stranded RNA

Exo Exosome subunit

GFP Green flourescent protein

UTR Untranslated region

NMD Nonsense-mediated decay

NGD No-go decay

NSD Non-stop decay

NRD nonfunctional rRNA decay

Abbrev. Abbreviations

FC Fold change

P Present

A Absent

M Marginally Present

I Increased

D Decreased d Drosophila rp49 Ribosomal protein 49

Hsp70 Heat shock protein 70

Hsp26 Heat shock protein 26

TAP Tandem affinity purification

ARE AU rich element

RNA BP RNA binding protein

PCR Polymerase chain reaction

ARE BP AU rich element binding protein

TBST Tris buffered saline + Tween 20

SDS Sodium dodecyl sulfate

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

HRP Horseradish peroxidase

Oligo Oligonucleotide

GCOS Gene chip operating system

MIAME Minimum information about a microarray experiment

TRAMP Trf4, Air2, Mtr4 polyadenylation

Dob1 dependent on eIF4B polyA+ polyadenylated

TTP Tristetraprolin

KSRP KH-type splicing regulatory protein

Met tRNAi initiator methionine tRNA

FISH fluorescence in situ hybridization

RT-qPCR quantitative reverse transcriptase PCR mtn metallothionein

ETS external transcribed spacer

Acknowledgements

The research presented here is the direct result of years of hard work and, most importantly, strong collaborations both from within the Andrulis Lab and without. Special thanks are extended to current and former Lab members, especially Erik Andrulis, Megan Mamolen and Amy Graham. Their examples have made me a better and more thorough scientist. A special thanks to Abram

Stavitsky, since he always had time to listen and to share ideas and his perspective on (many) different issues.

This work would not have been completed without the unwavering support of my parents, Kamill and Margit Kiss, my brother, Kamill, his wife Nora, and my sister Anna. In addition to my direct family I can list a stable of friends that have been my second family for the better part of two or three decades. This extended clan includes the Walsh, Csiszar, Dicken, Hughes, Hewlett, Piazza, Szűle, and

Hargitai families. In addition to those friends, I need to thank the many new friends I made here at Case. Their timely encouragement provided necessary energy during several occasions when the outcome of this work was still uncertain.

The Exozyme Model: A New Paradigm of Exosome Subunit Activity Revealed by

Diverse and Distinct Substrate Specificities of Exosome Subunits In Vivo

Abstract

DANIEL LOUIS KISS

The RNA processing exosome was originally defined as an evolutionarily conserved multi-subunit complex of ribonucleases (RNases) responsible for the processing and/or turnover of stable RNAs and mRNAs. Results from several in vivo systems challenge the prevailing model of exosome complex function, which relies heavily on in vitro reconstructions of the complex. Further, the detailed mechanisms for how individual exosome subunits participate in each of these

RNA metabolic pathways remains unclear. Here, I use RNA interference to deplete most exosome subunits in Drosophila melanogaster S2 tissue culture cells. Surprisingly, neither depletion nor over-expression of most exosome subunits greatly hinders the proliferation of S2 cells. I assayed the effects of the depletions on global mRNA levels using gene expression microarrays.

Consistent with the RNase activities ascribed to the exosome complex, most affected mRNAs are increased. However, my results revealed that the pools of mRNAs affected by exosome subunit depletion vary significantly in both the number and identity of the transcripts affected. Notably, the altered mRNAs possess both 5’ and 3’ untranslated regions that are longer than the representative transcriptomic average. The data also show a subunit specific preference for certain mitochondrial-targeted transcripts and for transcripts

surveyed by the nonsense-mediated decay (NMD) pathway. In particular, a very

large portion of transcripts affected by Rrp6 depletion were NMD substrates,

suggesting that Rrp6 may have a core exosome-independent function in NMD.

Since exosome subunit depletion had different effects on one class of known

exosome substrates (NMD-surveyed mRNAs) in my microarray studies, I wanted

to determine if another class of mRNA would be affected in a similar manner. I

selected two heat shock-inducible transcripts, hsp70 and hsp26, which contain

AU-rich elements (AREs) for further study. Consistent with the microarray

results, I observed distinct effects on hsp70 and hsp26 mRNA levels when different subunits are depleted. In this work, I propose and test the validity of the exozyme model, which is a new model for exosome subunit complex assembly and function. Together, these studies provide compelling evidence for multiple, independent functions for individual exosome subunits and suggests a larger population of exosome subunit assemblages than heretofore anticipated.

Chapter 1:

Introduction

1.A. RNA and the importance of RNA turnover

The past decades have witnessed an expansion in both the types of RNA known to science, and the functions attributed to those RNA molecules (Table 1).

Some of those RNA species, such as tRNAs and rRNA have extraordinarily long half-lives, while others such as cryptic unstable transcripts (CUTs) have such short half-lives that they are only detectable when cellular processes have been disrupted (240). Both the diversity of the RNA species identified to date, and the many functions attributed to them indicate that RNAs play vital roles in many processes in vivo. In addition to the large variety of new RNAs that have been characterized, many different RNA regulatory mechanisms, collectively termed

RNA surveillance, have been identified (48, 73, 74). The abundance of RNA surveillance pathways indicates that the transcription of new RNAs is only one of the ways cells regulate the levels of different transcripts in vivo (73, 74).

Depending upon the mechanism, RNA surveillance pathways can either stabilize or destabilize the surveyed RNA (48, 73, 74). As the surveillance processes of mRNAs are the best characterized, the simplified ‘life cycle’ of an mRNA, and its’ surveillance points are detailed in Figure 1. The dashed lines in Figure 1 indicate that mRNAs which fail to undergo that step of processing, or are processed incorrectly, are targeted for degradation.

The surveillance points listed in Figure 1 occur via a host of different factors

Table 1. Types of RNA

Type of RNA Abbreviation Function Distribution References protein Messenger RNA mRNA coding All organisms (110) & many others Ribosomal RNA rRNA translation All organisms (166) & many others Transfer RNA tRNA translation All organisms (97, 98) & many others stalled Transfer- ribosome messenger RNA tmRNA rescue bacteria (59, 85) Small nuclear splicing and Eukaryotes and RNA snRNA other archaea (212) Small nucleolar RNA Eukaryotes and RNA snoRNA modification archaea (123) mRNA trans SmY RNA SmY splicing nematodes (112) mRNA Guide RNA gRNA modification mitochondria (3) tRNA Ribonuclease P RNase P maturation All organisms (167) Ribonuclease rRNA MRP RNase MRP proccessing Eukaryotes (239) RNA Y RNA proccessing Animals (171) Telomerase telomere RNA synthesis Eukaryotes (most) (16, 208, 248) mRNA Antisense RNA aRNA surveillance All organisms (23, 24) Long noncoding many and RNA long ncRNA unknown Eukaryotes (238) gene Micro RNA miRNA regulation Eukaryotes (most) (169, 188) Piwi-interacting transposon RNA piRNA defense animals (most) (99) Small Interfering gene RNA siRNA regulation Eukaryotes (most) (146, 249, 250) Trans-acting gene siRNA tasiRNA regulation plants (227, 228) Repeat associated transposon siRNA rasiRNA defense insects, others? (55, 135, 244) Cryptic unstable transcripts CUTs unknown yeasts, others? (240) Upstream noncoding transcripts UNTs unknown plants, others? (39) Promoter upstream transcripts PROMPTs unknown yeasts, others? (178) Stable unannotated transcripts SUTs unknown yeasts, others? (241)

Type of RNA Abbreviation Function Distribution References gene Short RNAs sRNAs regulation? humans, others? (117) Termini- associated gene sRNAs TASRs regulation? humans, others? (117) gene Long RNAs lRNAs regulation? humans, others? (117)

Figure 1. mRNA regulation points

and mechanisms in both the nucleus and the cytoplasm, but the surveyed RNAs

‘caught’ in each surveillance step are degraded. Those misprocessed, damaged, or otherwise defective RNAs are often (or exclusively, Figure 1 solid arrows) targeted for degradation by cellular ribonucleases (RNases). The two major

types of RNases are endonucleases, which cut RNAs internally, and

exonucleases, which degrade RNAs from either their 3’ or 5’ ends (168, 251).

Some RNases can possess both endo- and exo-ribonuclease activites (132, 192,

197). The RNA processing exosome complex has been identified as one of the cell’s predominant 3’  5’ exoribonucleases and has been directly linked to many

of the turnover directed RNA surveillance points listed in Figure 1.

In the following work, I describe the activities attributed to the exosome

complex and the establishment of the core exosome paradigm. I detail the

current understanding of exosome subunit and complex architecture and follow

by explaining the core exosome model, which describes the proposed

mechanism of exosome subunit function. I then discuss data that are

inconsistent with the predictions of the core exosome model. Based upon these

inconsistencies, I propose and test the validity of the exozyme model, an

alternative paradigm of exosome subunit complex structure and function.

1.B. The core exosome complex paradigm

The RNA processing exosome was initially characterized in Saccharomyces

cerevisiea through a genetic screen searching for strains defective in rRNA

processing by David Tollervey’s group (102, 153, 195). The complex was initially

24 described as a five polypeptide complex required for the 3’→5’ processing and degradation of rRNA precursors (102, 153, 195). Later work identified several additional subunits that comprise the complex, and showed that the larger complex metabolized many types of RNAs including mRNA, rRNA, snRNA, snoRNA, and tRNA precursors in both the nucleus and the cytoplasm (26, 102,

113, 153, 195). The presence of an exosome-like complex is evolutionarily conserved, with a hexameric bacterial analog in PNPase, a primitive nine subunit form present in some archaea, and a more diversified complex present in all eukaryotes studied to date (68, 124, 195, 243).

The current paradigm detailing the composition, assembly, and function of the eukaryotic exosome complex posits that a stoichiometric complex containing single copies of each subunit is responsible for all of its functions. In this model, called here the “core exosome” model, all functions and contributions of individual subunits to distinct RNA processing and turnover events occur only in the context of this stoichiometric complex. This paradigm was established as a fusion of genetic and biochemical evidence from several model systems including Saccharomyces cerevisiea, Sulfolobus solfataricus, Archaeoglobus fulgidus, Trypanosoma brucei, Drosophila melanogaster, and human tissue culture cells (4-6, 12, 25, 27, 28, 32, 41, 64, 65, 69, 75, 153). Here, I review the information that led to the establishment of the core exosome model (102, 155,

195, 221).

1.B.1. The subunits of the exosome complex

Depending upon the organism and the cellular compartment, the exosome

complex consists of four (archaea) to eleven (many eukaryotes) distinct protein

subunits (Table 2) (6, 12, 65, 69, 102). I first discuss the exosome complex of S.

cerevisiea as this organism is the primary model system used in the field;

however, I also point out differences between the exosome complexes of

organisms as they arise. Based on their RNA interaction domain structures,

exosome subunits can be grouped into four subsets (102). The largest of these

subunit groups is comprised of the subunits with recognizable RNase PH

domains and includes mRNA transport 3 (Mtr3), superkiller 6 (Ski6) also called

ribosomal RNA processing 41 (Rrp41, referred to as Ski6 hereafter), Rrp42,

Rrp43, Rrp45, and Rrp46 (Table 2, Figure 2). The next subunits, Rrp40, Rrp4,

and Cep1 synthetic lethal 4 (Csl4) all contain S1 RNA binding domains (Table 2,

Figure 2). Defective in sister chromatid disjoining 3 (Dis3) (also called Rrp44 and

Mtr17, referred to as Dis3 hereafter) is the only RNase II family protein in the

complex and is homologous to E.coli RNase R (Table 2, Figure 2) (153, 164).

These first three groups of subunits are both nuclear and cytoplasmic in all organisms studied. The final subunit, Rrp6 (PM-Scl-100 in humans) is homologous to E.coli RNase D and is nuclear restricted in yeast and trypanosomes, but both nuclear and cytoplasmic in human and fly tissue culture cells (6, 25, 86, 88, 133).

The D. melanogaster exosome complex contains most of the subunits found

in yeast and humans. However, at least two different forms of the complex have

Table 2. Exosome subunits. * denotes this subunit is present in cytoplasmic complex, but missing from nuclear complex, n/a = the organism lacks that gene, no = gene is present but does not associate with the exosome complex

Subunit Domain(s) Archaea Yeast Human Fly Mtr3 RNase PH n/a yes yes yes Ski6/Rrp41 RNase PH yes yes yes yes Rrp42 RNase PH yes yes yes yes Rrp43/Oip2 RNase PH n/a yes yes n/a Rrp45/PM-Scl-75 RNase PH n/a yes yes yes* Rrp46 RNase PH n/a yes yes yes Rrp40 S1 n/a yes yes yes Rrp4 S1 yes yes yes yes Csl4/Ski4 S1 yes yes yes yes RNase R + Dis3/Rrp44/Mtr17 PINC n/a yes no yes RNase D + Rrp6/PM-Scl-100 HRDC n/a yes yes yes (32, 69, (25, 61, (27, 41, (12, 75, References 124, 142) 153, 192) 178, 181) 87, 88)

Figure 2. Domain maps of Drosophila melanogaster exosome subunits Exosome subunits are grouped by domain structures (88, 145).

been purified (12, 75). These forms, presumably a cytoplasmic form and a

transcription complex associated nuclear form, differ by the exclusion of Rrp45

from the latter complex (12, 75). Neither complex isolated from D. melanogaster

S2 cells contains an equivalent of Rrp43; nor has an obvious homologue been

identified to date in D. melanogaster (12, 75). While Rrp45 is missing from the

nuclear complex in D. melanogaster other proteins such as the transcription

elongation factors dSpt5 and dSpt6 have been shown to co-purify with the

transcription associated complex in D. melanogaster (12).

1.B.2. Arrangement of exosome subunits within a complex

Tollervey and colleagues first reported the exosome as an entity (153) and validated earlier work from Yanagida and coworkers which showed that

Schizosaccharomyces pombe Dis3 was part of a larger complex (122, 153).

Work from the Tollervey group, later confirmed by others, also indicated that Dis3 dissociated from the remainder of the complex at higher salt concentrations (6,

61). Early two-hybrid screens from the Pruijn and Clayton laboratories showed that several human and trypanosome RNase PH-domain subunits interact (65,

182, 183). These observations supported a model in which the RNase PH

domain subunits assembled into a hexameric ring (65, 182). This arrangement

became widely accepted as the archaeal RNase PH domain subunits were

shown to assemble into such a structure (32, 68). The Clayton study also

reported that depletion of one exosome subunit occasioanlly resulted in the

destabilization of other subunits, a phenomenon they called co-depletion (65).

Co-depletion was proteasome-dependent indicating that the co-depleted subunits

29 were destabilized at the protein level (65). Co-depletion has also been observed in other model systems including fly and human tissue culture cells (215). It is thought that depletion of one subunit causes complex disassembly and therefore resulted in the destabilization and turnover of one or more subunits as evidenced by co-depletion (65, 137, 215).

Exosome complexes using recombinant subunits from two different archaeons and humans have been reconstructed in vitro and crystallized (32, 68,

137, 142). I discuss the structures of the eukaryotic complex here (Figure 3), however I will return to the structures of the archaeal complexes later (see section 1.C.) The crystal structures of the reconstructed human exosome complex (Figure 3) is comprised of a stoichiometric “core” consisting of three S1 domain subunits (Rrp40, Csl4, and Rrp4) and six RNase PH domain subunits

(Ski6, Rrp45, Rrp46, Rrp43, Mtr3, and Rrp42). As predicted by the 2-hybrid assays the RNase PH domain subunits form a hexameric ring, the ‘top’ face of which serves as a landing pad for a cap composed of the S1 domain subunits

(Figure 3) (65, 137, 182).

The nine subunit core of RNase PH and S1 domain subunits has been proposed to serve as a scaffold for Dis3 and Rrp6. These two subunits are thought to associate with the bottom face of the RNase PH ring, although a crystal structure of such a ten or eleven subunit complex has not been obtained

(137). Electron microscopy and bioinformatics have indicated a likely Dis3 docking port on the bottom face of the RNase PH subunit ring (Figure 3c) (8,

233). In addition, D. melanogaster Dis3 has also been shown to interact directly

Figure 3. Core exosome complex structure.

(A and B) X-ray crystal structures of an in vitro reconstructed human exosome complex (137). (C) The crystal structures from Figures 3a and 3b served as a basis for modeling in Dis3 attachment to the complex (21).

with Rrp6 (86). As no other exosome subunit interacts directly with Rrp6, Dis3

likely serves as an adapter between Rrp6 and the core exosome (86).

Depending upon the conditions, the reconstructed exosome core complexes had limited RNase activity in vitro that increased or decreased when the remaining two polypeptides were added (61, 137).

1.B.3. The exosome complex has nuclear and cytoplasmic cofactors

Several cofactors, or cofactor complexes, are required for the exosome

complex (often shortened to the exosome) to function properly in vivo. These

cofactors are thought to act as compartment specific substrate specificity modules for the exosome complex (13, 63, 104, 129, 222, 240). The exosome

complexes’ nuclear specific cofactors include dependent on eIF4B (Dob1)/Mtr4

(Mtr4 hereafter) and Lrp1/C1D/Rrp47 (Rrp47 hereafter) (52, 152). Rrp47 has

been shown to associate with Rrp6 and its functions are tightly correlated with

those of Rrp6 (65, 93, 156, 194). While most of Mtr4’s functions have recently

been ascribed to Trf- Air-Mtr4 polyadenylation (TRAMP) complexes, it is likely

that Mtr4 has TRAMP independent functions (13, 104, 129, 152, 220, 222, 240).

Two functionally distinct (see below) TRAMP complexes, TRAMP4 and

TRAMP5, have been characterized (63, 104, 189). TRAMP5 is composed of

Trf5, Air1, plus Mtr4, and its activities seem to be restricted to a subset of the

exosome’s functions (see below) (63, 104). TRAMP4 can consist of Trf4, Mtr4, plus either Air1 or Air2 and until recently was thought to share all of TRAMP5s

functions, but is also required for a larger variety of the exosome functions (see below) (63, 104, 114, 129, 189, 222, 240).

In the cytoplasm, Ski complexes are the primary cofactors of the exosome complex. Ski complexes consist of Ski2, Ski3, and Ski8, which assemble into four

subunit complexes containing single copies of Ski2 and Ski3, and two copies of

Ski8 (205, 234). An additional protein, Ski7, is also required for most cytoplasmic

functions of the exosome (234). It has become a common convention to mutate or use RNA interference (RNAi) to deplete one of these cofactors, in lieu of

actual exosome subunits, when assaying cytoplasmic functions of the exosome

complex (20).

1.B.4. Functions of the exosome complex

The exosome complex has been shown to be a processive, sequence

independent, 3’  5’ exoribonuclease (61, 137, 153). In the cell, its functions

can be divided into two broad categories, 3’ end processing of RNAs and RNA

surveillance coupled with turnover, both of which are dependent upon the 3’  5’

exoribonuclease activity of the exosome complex (102, 153, 172, 195, 213, 221).

I combine RNA surveillance and turnover since the recruitment of the exosome to

an RNA undergoing surveillance results in the degradation of that RNA (41, 49,

113, 114, 130, 133, 158). I will discuss both of these functions, beginning with

RNA 3’ end processing in detail below.

1.B.4.a. 3’ end processing of stable RNAs

The first of the above functions to be attributed to the complex was RNA processing, or the trimming of 3’ extensions on several different species of stable

RNA (153, 154, 217, 218). At first, Rrp4, then other exosome subunits, were shown to play a role in trimming the 3’ extensions of 5.8S rRNA in yeast, and then in other organisms (4-6, 10, 25, 40, 56, 127, 152-154, 245). Collectively, these works showed that all exosome subunits were involved in the initial steps of 5.8S rRNA processing. These early genetic studies were therefore vital in establishing the core exosome paradigm (4-6, 10, 25, 40, 56, 128, 153, 154,

245). Other rRNA processing functions such as the trimming of both the 35S and

27S rRNA precursors, to 32S and 25S, respectively, were then shown to be dependent upon the exosome complex and its cofactors (4, 6, 39, 56, 70, 94,

104, 127, 152, 203, 236, 245).

In addition to rRNAs, exosome subunits have been shown to be required for the proper 3’ end processing of small nuclear RNAs (snRNAs) and small nucleolar RNAs (snoRNAs) (4, 36, 51, 90, 100, 121, 149, 175, 201, 217, 218,

223). In these studies, yeast strains with mutated or depleted exosome subunits accumulated aberrant, 3’ extended U1, U4, and U5 snRNAs (4, 217, 218).

Several different types of snoRNAs, such as intron-derived (U18, U24, snR44), polycistronic (snR72, 73, 190, U14), and independently transcribed snoRNAs

(snR33 and snR40) were shown to have 3’ end processing defects when assayed by Northern blotting in yeast strains with mutated or depleted exosome subunits (4, 218). Some of these species were also shown to be polyadenylated

(polyA+), an observation which was not fully understood at the time (4, 218).

However, the purpose and origin of those polyA+ RNA species were explained

by the characterization of the two TRAMP complexes (63, 104, 196, 222).

As mentioned earlier, two functionally distinct TRAMP complexes, TRAMP4

and TRAMP5, work as cofactors for the nuclear exosome complex (63, 104, 189,

196, 222). TRAMP complexes appear to reside at the nexus of exosome

complex’ RNA processing and RNA surveillance/turnover functions in the

nucleus (4, 63, 104, 196, 218, 222). Through an undefined mechanism, TRAMP

complexes can survey RNAs and add short polyA tails to mark aberrant RNAs for subsequent degradation by the exosome (63, 104, 196, 222). These functions become apparent in exosome subunit mutant yeast strains as the polyA+ RNAs are not degraded and accumulate to detectable levels (4, 63, 104, 196, 218,

222). Until recently, TRAMP5’s activities seemed to be restricted to the

surveillance of rRNA, snRNA and snoRNA (63, 104, 189). TRAMP4 was thought

to share all of TRAMP5’s functions, but also be required for the surveillance of

tRNAs and the turnover of cryptic unstable transcripts (CUTs)(see below) (114,

129, 189, 222, 240). Recent microarray work resulting from collaboration

between the Gerber and Keller labs convincingly shows that the substrates of the

two TRAMP complexes have a considerably smaller overlap in substrates than

previously anticipated (189). Further, at least for TRAMP4, the poly-A

polymerase activity of the complex is not required for the surveillance of most

surveyed RNAs (189).

1.B.4.b. RNA surveillance and turnover

The RNA surveillance functions attributed to the exosome complex are diverse, and occur in both the nucleus and cytoplasm with the aid of the different cofactors mentioned above. The aforementioned TRAMP complexes help prepare nuclear RNAs for exosome-mediated degradation, while Ski complexes are thought to perform a similar function in the cytoplasm (10, 13, 29, 58, 63, 96,

104, 144, 165, 192, 196, 220, 222, 234). Here I will discuss the surveillance and/or turnover functions of the nuclear and cytoplasmic exosome complexes.

1.B.4.b.1. Nuclear surveillance of stable RNAs

Early experiments showed that exosome subunits were required for the turnover of the 5’ external transcribed spacer (ETS) which is an rRNA processing intermediate of the rRNA precursor (33, 57, 61, 203, 218, 220, 224). The turnover of this RNA fragment, which is generated by the first cleavage step in the processing of the 35S rRNA precursor, demonstrated that the exosome complex could fully degrade RNAs in addition to processing their 3’ ends (33, 57,

61, 203, 218, 220, 224). Indeed, certain subunits (Rrp4, Rrp6, Rrp43, Ski6, and

Dis3) were then linked to the turnover of 23S and 21S RNA which are miscleaved rRNA precursors (5, 70, 94, 104, 152, 236, 245). Collectively these rRNA surveillance activities which target and degrade misprocessed rRNAs have been termed nonfunctional rRNA decay (NRD) (46, 131, 204). As shown earlier with 3’ end RNA processing, other types of RNA are also degraded by the exosome complex.

The Anderson lab was the first to show a tRNA surveillance system involving exosome subunits, specifically Dis3 (113). Their work showed that both Dis3 and

Trf4, the TRAMP4 component, were required for the recognition and turnover of

Met hypomodified tRNAi (113). That study was the basis of work by several other

labs which led to the discovery of the TRAMP complexes, and the subsequent

realization that only Dis3, and no other exosome subunits, was required for the

Met surveillance and turnover of hypomodified tRNAi (113, 129, 196, 222, 240).

1.B.4.b.2. Nuclear surveillance of unstable non-coding RNAs

With the release of multiple transcriptomic studies in yeast, humans, mice, and rice it has become clear that much larger portions of the genome are transcribed than previously thought (118, 119, 160, 178, 229, 230, 241). The use of high-resolution microarrays in an rrp6∆ strain first linked exosome subunits to

the turnover of intergenic non-coding RNAs (ncRNA) (240). In that work, the

Jacquier lab observed that a diverse class of short, polyA+ RNAs was

transcribed from intergenic regions of the yeast genome (240). They named these transcripts cryptic unstable transcripts (CUTs), as they were inherently

unstable and only detectable when Rrp6 was deleted (240). The authors

confirmed that CUTs were transcribed by RNAP II, and subsequent work has

shown that some CUTs undergo several steps of processing (14, 210, 240).

Recently, Steinmetz and coworkers have shown that, at least in yeast, many

CUTs are generated from bidirectional promoters throughout the genome (160,

241). In plants, upstream noncoding transcripts (UNTs) and their human

counterparts, promoter upstream transcripts (PROMPTs) have been identified

with the use of RNAi targeting exosome subunits (39, 160, 178, 240). These two

RNA classes are distinct from CUTs since they are associated with the promoter regions of genes, and therefore are not truly intergenic transcripts (39, 178).

To date, most of these transcripts have no defined function; however, some

CUTs have been shown to be involved in gene regulation directly via the RNA or indirectly via chromatin structure and transcriptional regulation (15, 34). As

UNTs and PROMPTs are associated with the promoter regions of genes, it is

suspected that they possess gene regulation related functions (39, 178).

Another group of intergenic transcripts, termed stable unannotated transcripts

(SUTs), was also identified in rrp6∆ yeast (241). SUTs, unlike CUTs, UNTs, and

PROMPTs, are stable, though present at very low levels in normal cells, and

have no defined function (241).

1.B.4.b.3. Nuclear RNA surveillance of mRNAs

Exosome subunits have also been implicated in the degradation of poly-A+

RNAs in the nucleus (DRN), and other specialized forms of mRNA turnover (22,

49, 95, 102, 195). To characterize this process, the Sherman lab concentrated

mRNAs in the nucleus by using yeast strains with mutated nuclear pore

components (49, 127). A combination of fluorescence in situ hybridization

(FISH), biochemical, and microarray studies showed that Rrp6, Rrp47, Dis3, and

Ski6 were involved in the turnover of ‘normal’ (poly-A+) mRNAs in the nucleus

(49, 50, 127). In addition to the general turnover of mRNAs by DRN, other

specialized forms of nuclear RNA turnover have been attributed to the exosome

complex.

Exosome subunits have been linked to the surveillance of each mRNA

transcription and processing step (102, 103, 225, 232). Nearly all exosome

subunits have been shown to co-localize and co-purify with elongating RNA polymerase (RNAP) II on actively transcribing heat shock genes in D. melanogaster (12, 190). Through chromatin immunoprecipitation experiments coupled to microarray analysis, Rrp6 and Rrp47 were shown to co-localize at

many actively transcribed gene loci in yeast (93). Multiple exosome subunits

were also demonstrated to be required for the surveillance and turnover of mis-

spliced mRNAs in yeast (31, 49, 95, 134, 147, 151, 211, 247). Additional studies

from many labs confirmed that exosome subunits were vital for the turnover of

mRNAs with malformed 3’ ends (30, 60, 89, 93, 95, 121, 134, 211, 214, 223,

231, 235, 237). Studies mutating different nuclear export factors also linked

exosome subunits to the surveillance of mRNAs that could not be properly

exported (72, 77, 95, 115, 116, 134, 211, 231, 247).

1.B.4.b.4. Cytoplasmic mRNA surveillance

As in the nucleus, the exosome complex has been linked to the general

turnover of mRNAs in the cytoplasm (10, 13, 39, 45, 132, 137, 165, 191, 192,

215, 220). Not only does this general turnover mechanism apply to whole

mRNAs, it also applies to mRNA fragments including intron lariats whose

degradation requires cytoplasmic exosome cofactors (96). Exosome subunits

have also been linked to the degradation of normal mRNAs such as those with

specialized elements in their 3’ untranslated regions (UTR) (41, 83, 84).

Specifically, exosome subunits, in conjunction with cofactors, have been shown

to be important for the decay of c-fos and tumor necrosis factor (TNF) mRNAs in

vitro, each of which contains an AU-rich element (ARE) (41, 83, 84, 158). The

surveillance of other ARE-containing transcripts has also been linked to exosome

subunits (9, 17, 42, 76, 83, 84, 91, 111, 136, 158, 180, 202, 215, 242). A distinct

3’ UTR regulatory element located in the phosphoglycerate kinase mRNA of

trypanosomes modulates the stability of that mRNA in a developmentally

regulated manner (18, 45, 198, 199).

The above examples show that exosome subunits degrade normal

cytoplasmic mRNAs in a regulated manner; however, as in the nucleus, they

have also been linked to the turnover of multiple different types of aberrant

mRNAs. Exosome-surveyed aberrant mRNAs include those that lack initiation or

termination codons in processes called no-go decay (NGD) and non-stop decay

(NSD) respectively (46, 58, 148, 170, 192, 204). The exosome complex has also been linked to the turnover of RNAs targeted for decay by small interfering RNAs

(siRNAs) (106, 165).

1.B.4.b.5. Nonsense-mediated decay

Another well characterized cytoplasmic mRNA surveillance pathway shown to

involve multiple exosome subunits is the nonsense-mediated decay (NMD)

pathway (161, 162, 184). NMD is a translation-dependent mRNA surveillance

pathway that detects mRNAs with premature termination codons (PTCs) and

targets them for degradation (45, 78, 79, 132, 133, 156, 161, 162, 185, 187, 192,

209). The turnover of NMD targeted transcripts has been well characterized in

yeast and metazoan (human and fly) cells, and occurs by different mechanisms

depending upon the organism studied (78, 133, 156, 185, 187). In Drosophila,

PTC containing transcripts are recognized by the NMD machinery and cleaved in

the vicinity of PTC (78). The newly produced fragments have free 3’ and 5’ ends

which are then recognized by cytoplasmic RNases (78). The 5’ fragment of the

NMD targeted mRNA which is produced by the endonucleolytic cleavage, has a

free 3’ end which is then recognized and degraded in an Rrp4 and Csl4

dependent manner (78).

1.B.5. Mechanism of core exosome function

The current paradigm of exosome complex function, the core exosome model,

is built on the idea that all exosome subunits function in a stoichiometric complex.

The proposed mechanism detailed below is patterned upon mechanistic insights learned from studies involving the archaeal exosome. In archaea, crystal structures show that the three S1 domain subunits form a pore-containing cap

through which the RNA is threaded (Figure 4a) (138, 140-142). In eukaryotes,

the TRAMP and Ski complexes are thought to facilitate this step by unwinding

structured RNAs in the nucleus and cytoplasm, respectively (63, 101, 104, 129,

222, 240).

In plants and archaea the RNA then enters the cavity formed by the

hexameric RNase PH domain subunit to the active site of Rrp41 where the RNA

Figure 4. Proposed mechanism for exosome complex function.

Catroons showing the proposed mechanisms of (A) archaeal or (B) eukaryotic exosome complex function. Red stars are sites of RNase activity, black dashed lines represent RNAs, and the dashed red line represents to putative path of

RNA in the eukaryotic complex (195).

42 is degraded (Figure 4a) (38, 40, 138-142). However, in eukaryotes the RNA is thought to exit the RNase PH ring via an exit channel located at the base of the

RNase PH ring. The RNA is then bound by Dis3, which degrades the RNA in a

3’  5’ manner (Figure 4b) (61). The position of Rrp6 in the complex is still undetermined, however, some evidence indicates that Rrp6 would associate with

Dis3 (86). Several results, which will be discussed in detail below, indicate that

Dis3, Rrp6, and possibly other exosome subunits can degrade RNAs without other exosome subunits (4, 6, 33, 195, 196).

1.B.6. Summary

The exosome complex in eukaryotes is thought to be a stoichiometric complex containing single copies of three S1 domain subunits and six RNase PH domain subunits that assemble into a single complex. Compartment specific cofactors present RNAs to the complex for subsequent processing or degradation by helping to thread the RNA through the pores formed by the S1 and RNase PH domain subunits (Figure 4). In most eukaryotes, the complex’

RNase activity is thought to reside in two additional proteins, Dis3 and Rrp6, which are loosely associated with the complex (Figure 3c, 5b) (6, 61, 145).

Crystal structures of such eukaryotic complexes coupled with results showing that RNAi depletion of certain individual exosome subunits results in the destabilization of other subunits have solidified this model. Further, the co- depletion results have been interpreted to mean that knocking out one polypeptide within the exosome core is sufficient to destabilize the complex and

43 thereby disrupt the majority, if not all, of core exosome-mediated RNA metabolic functions. All subsequent experiments probing the functions of exosome subunits have been designed and interpreted with this paradigm in mind.

1.C. Cracks in the core exosome complex paradigm

Although the exosome core paradigm has been useful towards understanding how exosome subunits function in vitro, detailed evidence from many in vivo studies challenge this paradigm. I devote the following pages to summarize the structural, biochemical, cell biological, bioinformatic, and genetic evidence that contradict the predictions of the core exosome model. Further, I show that both exosome subunit subcomplexes and structurally distinct exosome complexes are present in vivo. Finally, I show that certain functions ascribed to the entire exosome complex can either be performed by a subset of exosome subunits, or in some cases, a single subunit.

1.C.1. Archaeal crystal structures

As mentioned above, some archaea possess a homologue of the eukaryotic exosome complex. Both Sulfolobus sulfaraticus and Archaeoglobus fulgidus are archaea that possess a simplified form of the exosome consisting of four subunits, two S1 domain containing proteins, Csl4 and Rrp4, plus two RNase PH domain containing proteins, Rrp41 and Rrp42 (Table 2). These subunits multimerize and assemble into distinct nine subunit complexes in vitro (Figure 5)

(32, 68, 140). All archaeal complexes have a hexameric ring of alternating Rrp41

Figure 5. Archaeal exosome structures. The (A) Rrp4-capped (orange) and (B) Csl4-capped (red) exosome complexes of

Archaeoglobus fulgidus (32). The hexameric ring of Rrp41 (blue) and Rrp42

(green) below the S1-domain subunit cap is shown in both panels

and Rrp42 subunits topped with a trimeric cap of S1 domain subunits (Figure 5)

(32, 137, 142). It is the composition of these caps which varies from complex to complex (Figure 5) (32). Two distinct archaeal exosome complexes with either homogenous Csl4 or Rrp4 caps were reconstructed in vitro and crystallized

(Figure 5) (32). Further, when all the subunits of the archaeal complex were co- expressed from polycistronic vectors in E. coli, the Hopfner group was able to purify exosome complexes of approximately the right size, suggesting that these subunits self-assemble into complexes (32). Attempts to resolve the crystal structures from these harvested complexes were unsuccessful as the crystals did not diffract (32). Searching for an explanation, they determined the ratios of the subunits in the purified complexes using tagged copies of either Csl4 or Rrp4.

The ratios shifted depending upon which subunit was tagged, but were consistent with exosome complexes containing caps composed of all possible combinations of Csl4 and Rrp4 (32). Further, when purified, those ‘mixed’ complexes were able to degrade short RNAs in vitro(32). As all four exosome subunits are co-expressed in archaea, these data imply that functional, but non- stoichiometric, exosome complexes exist in vivo (32). The presence of these non-stoichiometric complexes in archaea suggests that similar complexes are possible in eukaryotes.

1.C.2. Distinct complexes are purified from yeast and fly

The second set of evidence contradicting the core exosome model come from different studies which purified complexes from intact cells. In two of these

studies tandem affinity purification (TAP) tagged versions of the entire yeast

proteome were expressed from their endogenous promoters and the resulting

complexes were purified and their composition was determined by mass

spectrometry using similar automated procedures (80, 125). The results of these

two studies, and two prior smaller scale studies are summarized in Table 3 (80,

81, 125, 126). Although the procedures used to carry out these studies were very

similar, the results were substantially different (80, 81, 125, 126). Depending

upon which subunit was tagged, very different complexes were recovered (Table

3) (80, 81, 125, 126). For example, in the second Gavin study, between eight

and eleven exosome subunits were recovered regardless of which subunit was

tagged (80). Interestingly, Rrp46, an RNase PH domain subunit which,

according to the core exosome hypothesis, should be required to form an intact

hexameric ring; was the most commonly absent subunit from these purifications

(80). A second study by Krogan and coworkers used this methodology and

obtained more diverse complexes (Table 3) (125). In this second study, one

exosome subunit, Rrp41, failed to purify any additional exosome subunits (Table

3) (125). Others such as Rrp40, Rrp42, Csl4, and Rrp45 failed to purify even half of the remaining exosome subunits (Table 3) (125). Despite the similar approaches, there was little agreement between the two proteomic studies (80,

125). For example, in the Krogan study, Mtr3-TAP was the most successful in purifying the exosome complex, while Mtr3-TAP was one of the worst in the

Gavin study (Table 3) (80, 125). In both studies, known exosome cofactors and associated proteins, such as MPP6 and most Ski complex components are

Table 3. Distinct exosome complexes are purified from S. Cerevisiea.

Krogan et al. (2006) Gavin et al. (2006)

Tagged # of # of Missing Exosome Subunits Missing Subunits Subunit Extra Subunit Identified Subunit(s) Extra proteins Identified (s) proteins Ski7, Srp1, Ski6/Rrp41, Ssa2, Ura2, Rrp4 8 Rrp46, Mtr3 -- 11 -- Rrp5 MPP6 ,SRP1, SKI7, MYO3, KAP95, ILS1, Srp1, Tef2, RPB1, Ths1, Tub3, RPL40A, Vma2, SRP40, Cdc19, CRN1, RPB1, Gcn20, CST29, Gfa1, Grs1, RPA190, Kap95, Rrp4, Rrp43, SRP1, Mtr3, Rps0b, Dis3/Rrp44, YGR130C, Rrp42, Rps17a, Rrp6 6 Rrp46, Mtr3 DDI1 8 Rrp46 Rps1b Rrp4, Rrp6, Rrp40, Ski6/Rrp41, Rrp42, Rrp43, Dis3/Rrp44, Rrp45, Rrp46, YBR025C, Csl4, Dis3, OSM1, SKI7, Srp1, Ssa2, Rrp40 1 Mtr3 YDR101C 10 Csl4 Ssb1, Pdc1 Rrp4, Rrp6, Rrp40, Rrp42, Rrp43, Dis3/Rrp44, Ecm16, Rrp45, Rrp46, Imd4, Ski7, Rrp41 / Rrp47, Csl4, Srp1, Ssa2, Ski6 0 Dis3, Mtr3 SKI7 11 -- Utp10 Rrp4, Rrp6, Hhf1, Imd3, Rrp40, Pab1, Ski6/Rrp41, Rpl27a, Rrp42, Rrp43, Rps22b, Dis3/Rrp44, Ski7, Sro9, Rrp45, Rrp46, Ssa2, Ssb1, Rrp47, Dis3, FAB1, SRP1, Ura2, Utp20, Rrp42 1 Mtr3 SKI7, MPP6 11 -- Rrp5 LYS20, RPS26B, PHO81, SRB2, RRP5, SKI7, MYO3, Rrp4, RPS26B, Dis3/Rrp44, TRF4, MTR4, Rrp46, Rrp47, SKI2, SRP1, Rrp42, Ski7, Ssa2, Rrp43 6 Mtr3 PRP43 9 Rrp46 Rrp5

Krogan et al. (2006) Gavin et al. (2006)

Tagged # of # of Missing Exosome Subunits Missing Subunits Subunit Extra Subunit Identified Subunit(s) Extra proteins Identified (s) proteins Ecm16, Imd4, Pab1, Rpl20b, Rpl27b, Rrp4, Rrp6, Rrp5, Ski2, Rrp42, Ski3, Ski7, Rrp43, YBL055C, MPP6, Srp1, Ssa2, Mtr3, ENT3, SKI2, Ssb1, Ura2, Dis3/Rrp44 IMD3, RPA49, Utp10, , Rrp46, SKI7, RPA135, Utp20, Rrp45 4 Rrp47, SKI3, RPC82 11 -- Utp22, Vma2 Imd3, Pab1, DDI1, SSM4, Rpp0, Ski7, RSN1, MPP6 , Srp1, Ssa2, Rrp47, SKI7, SPS100, Ssb1, Ssb2, Rrp46 9 Mtr3 SRP1, IDI1 11 -- Utp22 Rrp4, Rrp6, Rrp42, Rrp43, Dis3/Rrp44 MPP6 , SKI7, , Rrp45, MYO3, TOP2, Rrp40, Ski7, Srp1, Rrp46, PKC1, UBA1, Rrp45, Ssa2, Ssb2, Csl4 3 Csl4, Mtr3 MRPL24 8 Rrp46 Yir035c

NUP170, MRC1, RPC53, GDH2, RPC11, ZIP1, POG1, YJL011C, RPC37, RPC31, MPP6 , SKI7, RPO31, RET1, RPB8, RPC40, RPO26, RPC82, Rrp4, NUP170, RPB5, Ski6/Rrp41, KAP95, LSM4, Dis3 / Rrp46, GSY2, SRP1, Rrp44 7 Mtr3 SKI2, TOP2 10 Rrp45 Ssa2, Ssb1 Imd4, Rrp5, Ski2, Ski3, Rrp4, Ski7, Ssa2, CBK1, MPP6 , Rrp40, Ssb2, Utp10, Mtr3 10 Rrp46 SKI7 8 Rrp46 Utp15, Vma2 Rrp4, Ski6, Rrp42, MPP6 , Q0055, Mtr3,Rrp43 YDR533C, LAP4, Rrp45, YML014W, Rrp46, YMR187C, Rrp47 4 Rrp47, ALG11 N/A N/A N/A

faithfully purified with multiple tagged subunits (Table 3) (80, 125). However, in

many cases proteins with no known role in exosome-mediated processes were also purified (Table 3) (80, 81, 125, 126). Presumably, these novel exosome subunit interacting proteins have functional, possibly in RNA processing and/or turnover, roles.

Exosome complex purification experiments using FLAG-tagged exosome subunits in D. melanogaster S2 tissue culture cells also yielded heterogeneous complexes (88). In that study, each exosome subunit was recovered above background levels; however, certain tagged subunits co-purified other subunits

with different efficiencies (88). For example, FLAG-tagged Dis3 and Rrp4 co- immunoprecipitated (co-IP) all subunits with a high specificity (88). Other FLAG-

tagged subunits such as Rrp40, Ski6, and Csl4 behaved differently, efficiently co-

IP-ing most, but not all exosome subunits. Still other subunits such as Mtr3,

Rrp42, and Rrp46 failed to co-IP the majority of the remaining subunits efficiently

(88). One possible explanation of these results is that the tagged subunits which

recover other exosome subunits poorly are present in only a subset of the

complexes present in the cell.

1.C.3. Exosome subunits have distinct localization patterns

Cell biological evidence from different organisms also contradicts the predictions of the core exosome model (88, 105). Since the core exosome model attributes all functions of each subunit to their roles within the complex it

also predicts that the subunits would co-localize. The data obtained by O'Shea

and co-workers in a genome-wide screen shows that GFP tagged exosome subunits do not have overlapping localization patterns in yeast (82, 105). Certain

subunits, such as Rrp6, Rrp40, Ski6, Rrp43, and Rrp45 were restricted to the

nucleus and nucleolus, while others like Mtr3, Rrp42, Rrp46, Dis3, and Csl4 also showed a cytoplasmic localization (105).

In D. melanogaster S2 tissue culture cells, both endogenous and FLAG- tagged exosome subunits have distinct localization patterns (88). These patterns were more diverse than those observed in yeast (88). In S2 cells, most exosome subunits are predominantly cytoplasmic (88). Rrp6 and Dis3 are exceptions, as they are predominantly, but not exclusively, nucleolar and nuclear respectively

(88). Further, of the cytoplasmic subunits Mtr3 and Rrp40, are predominantly associated with the plasma and/or nuclear membrane (88). Other subunits such as Rrp4, Rrp42, and Csl4 localize to cytoplasmic foci that vary in both size and number (88). These distinct localization profiles are conserved between different organisms and are inconsistent with the core exosome model which predicts that

the subunits would co-localize.

1.C.4. Functional evidence

Substantial experimental evidence from every exosome model system also

shows that there are different functional effects when different exosome subunits

are mutated or depleted. These observations challenge the core exosome

paradigm since that model is based on the assumption that loss of one subunit

destabilizes the entire complex and results in a total loss of exosome complex

activity. Below, I review the transcriptomic, genetic, and biochemical studies

showing that different exosome subunits likely have specialized functions.

1.C.4.a. Microarray studies show exosome subunits survey distinct RNAs

in vivo

Findings from transcriptomic studies in yeast and plants challenge the core

exosome model. As mentioned above, the core exosome model predicts that a

conserved set of RNAs would be affected when individual subunits are mutated

or depleted. This would be consistent with the transcriptomic profiles observed

when the components of other RNA processing or surveillance machineries were

depleted (92, 177, 185, 186). Such studies include a comprehensive test of

NMD factors in D. melanogaster S2 cells, and different mRNA export or

spliceosome component proteins, all of which yielded highly conserved sets of

mRNAs (92, 177, 185, 186). In contrast to those precedents with other RNA

surveillance systems, microarray studies in yeast and plants show that depletion or mutation of individual exosome subunits affects distinct sets of RNAs (Figure

6) (39, 100).

The first array study to examine multiple exosome subunits was performed by the Tollervey lab in yeast (100). They used RNA harvested from rrp41-1, rrp6∆

and rrp47∆ yeast strains in conjunction with gene expression arrays. Their

arrays showed that the affected mRNAs were predominantly increased with

Figure 6. Microarray studies show exosome subunits affect distinct RNAs. Microarray studies in (A and B) yeast and (C and D) plants show that mutation, deletion, or depletion of different exosome subunits stabilizes different RNAs (39,

100).

~82%, ~75%, and 67% of affected transcripts being increased in rrp6∆, rrp47∆, and rrp41-1 strains respectively (100). While most transcripts were increased,

they were not shared between the different subunits (Figure 6a and 6b) (100). rrp47∆ yeast shared the highest proportion (~47%) of its transcripts with both rrp6∆ and rrp41-1 strains, while the rrp6∆ strain shared the fewest (~21%)

(Figure 6a and 6b) (100). Further, in rrp6∆ yeast, nearly half the transcripts

were unique to that strain (Figure 6a and 6b) (100). The authors even posit that

their results could infer that Rrp6 has additional functions outside the exosome

complex (100).

To confirm and expand upon their results, the authors examined several

individual RNAs. First, they focused on NRD1 mRNA and they observed that this transcript accumulates to abnormally high levels at permissive temperature in all three yeast strains. However, upon shift to restrictive temperature, the transcript disappears from rrp41-1 cells, but is maintained at elevated levels in rrp6∆,

rrp47∆ strains (100). Tollervey and coworkers also observed that different

species of 3’ extended snoRNAs and snRNAs accumulated in the three mutant

strains (100).

Similar results were obtained in plants using whole genome tiling arrays (39).

In that study (Figures 6b and 6c), RNA was harvested from either csl4-2 mutant

(a null allele) plants or plants depleted of Rrp4 or Rrp41 using RNAi (39). The

tiling array results in plants are not directly comparable to the previous study as

tiling arrays don’t look at individual mRNAs, but rather look at the entire genome

in an incremental fashion (39, 100). The closest comparison that can be made is

54 by comparing gene regions on the tiling arrays to the gene expression results in yeast (39, 100). As with the yeast arrays, a large majority (~84%) of the gene regions in Rrp4 and Rrp41 depleted plants were increased (c.f. Figure 6c and

6d); however csl4-2 plants showed the opposite trend with only 32% of the transcripts being increased (Figure 6d) (39). The authors do not comment further on the nature of this large disparity between csl4-2 and Rrp4 or Rrp41 depleted plants (39).

As in yeast, there were substantial differences in the identities of the RNAs affected by exosome subunit depletion or mutation in plants (39). The largest difference was observed in csl4-2 cells, where ~84% of affected gene regions were unique to csl4-2 plants (Figures 6c and 6d) (39). The gene regions affected by Rrp4 and Rrp41 depletion are more similar, but in each case, nearly one-third of the detected RNAs are unique to one subunit (Figures 6c and 6d) (39).

Further, this trend held true for other types of RNA including rRNAs, miRNAs, tRNAs, snRNAs, snoRNAs and 7SL RNA (39). This study shows that the effects of depleting Rrp4 and Rrp41 in plants were similar, but not identical, while the

RNAs altered in the csl4-2 plants were radically different than the other two subunits (Figures 6c and 6d) (39).

Novel, non-coding (ncRNAs) transcribed from gene regions upstream of mRNA promoters were also identified by these tiling arrays (39). UNTs are similar to the more recently named PROMPTs which were identified with the use of tiling arrays using RNA from human cells depleted of Rrp46 (39, 178). Both

UNTs and PROMPTs were identified using similar methodologies in different

55 organisms and have very similar characteristics (39, 178). Both UNTs and

PROMPTs are transcribed from both DNA strands in the vicinity of gene promoters (39, 178). Like CUTs, these transcripts are inherently unstable as they are only detectable in RNase deficient cells (39, 178). Mining of the microarray data shows that UNTS are differentially affected by the depletion of

Rrp4 and Rrp41 (39). Likewise, quantitative reverse transcriptase PCR (RT- qPCR) experiments show that the levels of individual PROMPTs in human cells were differentially affected when different exosome subunits (Rrp40, Rrp46, Dis3, and Rrp6) were depleted in human cells (39, 178). Curiously, the authors use

Dis3 depletion to argue exosome complex specific effects despite the fact that

Dis3 does not associate with the human exosome complex (178). These array studies show that mutation or depletion of different exosome subunits has distinct effects on multiple types of RNAs (39, 178). Further, these differences are conserved among yeast, plants, and human cells (39, 100, 178).

1.C.4.b. Targeted studies show exosome subunit mutants accumulate distinct RNA intermediates in vivo

In the studies using multiple exosome subunits above, the authors state that the heterogeneous nature of the data sets were unexpected (39, 100). Indeed, the core exosome paradigm predicts the opposite result. To determine if the heterogeneous data sets were a byproduct of the broad nature of microarray experiments, I analyzed the primary data that helped to establish the core exosome paradigm. The majority of these experiments were Northern blots

using RNA harvested from temperature sensitive (ts) mutants of exosome subunits or from strains where exosome subunit expression was driven by galactose (GAL) inducible promoters (4-7, 31). To summarize the results briefly, regardless of the approach or RNA species tested, similar but not identical RNA processing intermediates accumulated in yeast strains with depleted or mutated

exosome subunits (4-7, 31, 35, 36, 39, 40, 47, 51, 53, 56, 61, 64, 70, 71, 94, 100,

101, 104, 120, 121, 128, 129, 132, 149, 153, 156, 175, 178, 201-203, 216-218,

220, 223, 236, 246, 247).

As mined from the works cited above, yeast strains or other cells deleted or

depleted of exosome subunits accumulate different RNAs or distinct processing intermediates of the same RNA depending upon which subunit is targeted. The

3’ end processing of 5.8S rRNA is the most thoroughly characterized function of

the exosome complex. Several studies have shown that exosome subunits are

required for the trimming of this rRNA precursor (4-6, 25, 31, 38, 40, 64).

However, in these studies, multiple distinct 3’ extended 5.8S rRNA processing

intermediates accumulate in different exosome subunit deletion or GAL depletion

strains (4-6, 25, 31, 38, 40, 64). The most striking difference occurs in rrp6∆ and

rrp47∆ yeast, which are the only two strains to accumulate a 5.8S+27 precursor

(4, 6, 31). Depletions of other subunits such as Ski6, Rrp42, and Rrp43 show

several subunit specific 3’ extended 5.8S rRNA species (4, 6, 31). Subsequent

studies examining the turnover of the 5’ ETS in yeast yield similar results (4-7).

Other stable RNAs such as snoRNAs (U14, 18, and U24) and snRNAs (U4 and

U5) also accumulate distinct processing intermediates in mutant or GAL depleted

yeast strains (4-6). Further, these distinct processing intermediates were

independently reproduced by multiple subsequent studies, both targeted and

transcriptomic, in many labs. (4-7, 31, 33, 35, 36, 39, 40, 47, 51, 53, 56, 61, 64,

70, 71, 94, 100, 101, 104, 120, 121, 128, 129, 132, 149, 153, 156, 175, 178, 201-

203, 216-218, 220, 223, 236, 246, 247). The studies listed above all challenge

the core exosome model, which states that loss of one subunit destabilizes the

complex, which then causes a total loss of RNase, and therefore RNA

processing, activity.

1.C.5. Individual exosome subunits have complex independent functions in

vivo

The first papers which identified all the subunits of the exosome complex were

among the first to show that exosome subunits had specialized functions (4-6,

25, 26, 31). As mentioned above, these, and other works all showed that a

distinct 3’ extended rRNA processing intermediate (5.8S +27 rRNA) accumulated

only in rrp6∆ and rrp47∆ yeast strains (4-6, 25, 26, 31, 33, 40, 64, 152, 153, 156,

245). Mutations and/or depletions of every other exosome subunit produced

multiple species of 5.8S rRNA with longer 3’ extensions (4-6, 25, 26, 31, 40, 64,

152, 153, 156, 245). The core exosome paradigm was modified to explain these

results with a two-step hand-over mechanism (175, 201, 215). In the first step,

the core exosome trims the 7S pre-rRNA to 5.8S +27nt precursor (175, 201,

215). Then the trimmed precursor is further processed by Rrp6, which works in conjunction with Rrp47 to trim most of the remaining nucleotides (175, 201, 215).

Finally, MPP6 replaces Rrp6 and Rrp47 to remove most of the remaining 3’

extension (150, 193, 194).

Recent work from the Butler lab used Rrp6 domain mutants to show Rrp6 can

remove this 3’ extension, and perform certain other RNA processing functions

even when it cannot physically associate with other exosome subunits (33). This

demonstrates that certain RNA processing functions attributed to the entire

nuclear exosome, via experiments in Rrp6∆ strains, may in fact be core exosome

complex independent and be performed by Rrp6 alone (33). Additional work in D.

melanogaster S2 cells has linked Rrp6 to cell cycle progression and shown that

those roles are independent of other exosome subunits (87). When compared to

other exosome subunits, Rrp46, Rrp6, and Dis3 depletion in human cells all have

different effects on the levels of certain PROMPTs (33, 178). Rrp6∆ yeast strains

also show distinct effects on global mRNA levels (100). These observations

make it likely that other functions attributed to the exosome complex via Rrp6,

may be carried out by Rrp6 in the absence of the remaining exosome subunits

(33, 100, 178).

Core exosome complex independent RNA surveillance functions are not

restricted to Rrp6 (196). For example, the surveillance and turnover of

Met hypomodified tRNAi , a function that was initially attributed to the entire

exosome complex, can be performed independently by Dis3 alone (113, 114,

129, 196, 222). Further, other exosome subunits are not involved in this process as purified Dis3 and TRAMP complexes were sufficient to perform this activity in vitro (196). These results all challenge the core exosome model by showing that

59 certain RNA processing and surveillance functions attributed to the entire exosome complex can be performed independently by individual exosome subunits.

1.C.6. Some exosome subunits form sub-complexes in vivo

Finally, new evidence shows that eukaryotic exosome subunits can form subcomplexes in vivo (86). Two exosome subunits, dRrp6 and dDis3, are known to associate in complexes with importin-α3 and these exosome subunit-importin sub-complexes were shown to be independent of other exosome subunits (86).

These data were obtained from cell biological and biochemical results showing that N-terminally truncated Dis3 does not associate with most other exosome subunits in S2 cells (86). In fact, the only subunit to interact (by co-IP) or co- localize with N-terminally truncated Dis3 was Rrp6 (86). The precise functional nature of this exosome subunit subcomplex has yet to be elucidated, but it may be involved in the proper nuclear and cytoplasmic localization of other exosome subunits (86). Several other possible exosome subunit subcomplexes have also been observed; however no attempt has been made to assess whether they retain functionality in vivo or in vitro (206, 207). Collectively these results, coupled with the complex purification results presented in section 1.C.2. imply that exosome subcomplexes may exist in vivo.

1.C.7. Summary

The results above demonstrate that, contrary to the core exosome model, individual exosome subunits contribute differently to multiple distinct RNA processing pathways in vivo. Indeed, even the early observations used to form the basis of the core exosome model do not bear out the model’s predictions.

When examined in detail, the published data in the field contradicts the predictions of the core exosome model in whole, or in part. The observations presented above pose an overall question: do exosome subunits function as core complexes, sub-complexes, independently, or a combination of all those possibilities?

1.D. Hypothesis

The core exosome paradigm cannot address the conflicts posed by the data presented above. Further, many of the core exosome model’s predictions have been partially or completely refuted by experimental observations. The data are consistent with a model where individual exosome subunits assemble into multiple distinct complexes in vivo. This new model of exosome subunit function, which I call the “exozyme hypothesis”, is rooted in several basic observations.

First, it is widely accepted the natural composition of the archaeal exosome varies from complex to complex (32, 66, 67, 138). Thus, archaea possess functional exosome complexes of mixed composition in vivo (32). Second, subsets of eukaryotic exosome subunits can independently perform RNA metabolic functions in vivo (33, 196). Third, exosome subunits do not always co- localize in vivo (88, 105). Finally, depletion of different exosome subunits shows

61 that they survey different RNAs and produce distinct RNA processing intermediates (37, 100, 178). These observations are true for all model systems including yeast, plants, archaea, and both fly and human tissue culture cells (12,

32, 39, 75, 86-88, 100, 105, 178). The exozyme hypothesis integrates these observations, all of which are at odds with the core exosome paradigm.

In the exozyme hypothesis, exosome subunits can assemble into and function as independent complexes termed exozymes. Exozymes can either assemble independent of the 9-subunit core exosome, be dissociated from the core, and/or be in a dynamic relationship with the core. Exozymes may be comprised of only exosome subunits or a mélange of exosome subunits, cofactors, and non- exosome polypeptides. An individual exozyme complex may function independently of other exozymes and the exosome core via different substrate specificities, or they may share RNA substrates or distinct steps of RNA processing.

In order to test the exozyme hypothesis directly, I carried out a nearly comprehensive gene expression microarray study by using RNAi to deplete exosome subunits in Drosophila S2 cells. In addition to the genome-wide approach, I focus on mRNAs known to be surveyed by exosome subunits, namely the ARE containing hsp70 and hsp26 mRNAs. Taken together, the published data in the field plus my transcriptomic and targeted studies indicate that individual exosome subunits contribute in a non-overlapping manner to distinct mRNA turnover pathways. These observations are inconsistent with the core exosome paradigm, however, they are consistent with either a model for

62 exosome subunit substrate specificity within a single exosome complex or, alternatively, the exozyme hypothesis.

Chapter 2:

Genome-wide Analysis Reveals Distinct Substrate Specificities of Rrp6,

Dis3, and Core Exosome Subunits

Daniel L. Kiss and Erik D. Andrulis

Department of Molecular Biology and Microbiology, Case Western Reserve

University School of Medicine, Cleveland, Ohio

2.A. Abstract

The RNA processing exosome complex was defined as an evolutionarily conserved multi-subunit complex of ribonucleases responsible for the processing and/or turnover many types of RNAs. The exosome complex is involved in the general and targeted surveillance of mRNAs in both the nucleus and the cytoplasm including nonsense-mediated decay (NMD) targets. The detailed roles of individual exosome subunits in these RNA metabolic pathways remains unclear. Here, I propose and test the exozyme model as a way to resolve the discrepancies between data yielded from in vitro reconstructions of the complex and those that arise from assaying exosome subunit function in vivo. To do this I use RNAi to deplete core exosome subunits, the exonucleases Rrp6 and Dis3, and Rrp47, an exosome cofactor, in Drosophila melanogaster S2 tissue culture cells. These studies show that only Rrp6 and Dis3 are required for continued cell proliferation. I also assay the effects on global mRNA levels using gene expression microarrays. As predicted by the RNase activities attributed to the exosome complex and certain subunits, affected mRNAs are predominantly increased. The affected mRNAs encode proteins that function in connected cellular pathways, suggesting strong interactions among and between exosome subunits. However, as predicted by the exozyme model, my results reveal substantial differences in the pools of affected mRNAs for each depleted exosome subunit microarray profile (desMAP). Both the size and the content of the desMAPs vary considerably from subunit to subunit. Particular classes of mRNAs are also affected differentially. For example, ~25% of the affected

transcripts in Rrp6 depleted cells are NMD substrates, while those transcripts

only represent between 1.5% and 6.3% of the remaining subunits’ desMAPs.

Nuclear encoded mitochondrial transcripts were also differentially enriched in

these experiments. Finally, the affected mRNAs possess 3’ and 5’ untranslated regions that are longer than the representative transcriptomic average. We

conclude that individual exosome subunits are largely functionally independent at

the level of individual transcripts, but are interdependent on a transcriptomic level. These findings are predicted by and lend support to the exozyme model of exosome subunit function.

2.B. Introduction

The RNA processing exosome was initially characterized as a five polypeptide

complex important for the 3’→5’ processing and degradation of rRNA (102, 153,

195). Subsequent work identified additional subunits that comprise the complex,

localized exosome subunits to both the nucleus and the cytoplasm, and showed

that the exosome complex also metabolized many types of RNAs including

mRNA, rRNA, snRNA, snoRNA, and tRNA precursors (4, 6, 25, 31, 102, 113,

153, 195). Exosome subunits have also been implicated in most RNA metabolic

functions including the degradation of poly-A+ RNAs in the nucleus (DRN), and

other specialized forms of nuclear mRNA turnover (22, 49, 95, 102, 195). The

complex is also involved in the destabilization of intergenic transcripts such as

CUTs (cryptic upstream transcripts), UNTs (upstream noncoding transcripts), and

their human equivalents PROMPTs (promoter upstream transcripts) (39, 160,

178, 240). In the cytoplasm, multiple exosome subunits have been shown to participate in degrading transcripts targeted by both the siRNA and nonsense

mediated decay (NMD) machineries, although the mechanisms by which the

surveyed transcripts are degraded varies by organism and system (78, 133).

Exosome subunits have also been linked to the degradation of mRNAs with

specialized elements in their 3’ untranslated regions (UTRs) (41). Specifically,

exosome subunits, in conjunction with cofactors, have been shown to be

important for the decay of c-fos and tumor necrosis factor (TNF) mRNAs, each of which contains an AU-rich element (ARE), in vitro (41). Despite great progress in

understanding the nature and scope of the RNA metabolic pathways and

67 functions of individual exosome subunits, we are only beginning to comprehend how these subunits assemble and function as active complexes in vivo.

Strides towards understanding exosome subunit assembly and complex architecture have been made with in vitro reconstructions of archaeal and eukaryotic exosome complexes. Archaeoglobus fulgidus and some other archaea possess a simplified form of the exosome consisting of four subunits: two S1 domain containing proteins, Csl4 and Rrp4, plus two RNase PH domain containing proteins, Rrp41 and Rrp42, which multimerize and assemble into distinct nine subunit complexes in vitro (32, 68, 140). All archaeal complexes have a hexameric ring of alternating RNase PH domain subunits capped with three S1 domain subunits (32, 137, 142). Notably, the existence of structurally distinct archaeal exosome complexes has been proven in vitro and complexes with mixed caps are thought to be prevalent in vivo (32, 66, 67, 138).

Stoichiometric eukaryotic exosomes from yeast and humans have also been crystallized (137). Both the yeast and human exosome complexes comprise a

“core” consisting of three S1 domain subunits (Csl4, Rrp4, and Rrp40) and six

RNase PH domain subunits (Ski6/Rrp41, Rrp42, Rrp43, Rrp45, Rrp46, and

Mtr3). The nine subunit core of RNase PH and S1 domain subunits has been proposed to serve as a scaffold for two additional polypeptides, Dis3 and Rrp6,

RNase II/R and RNase D homologs, respectively. The reconstructed core had limited RNase activity in vitro that increased when the remaining two polypeptides were added (21, 137). Although it was initially shown that multiple subunits within the complex were catalytically active, recent studies have argued

that this activity is predominantly, if not exclusively, found in Dis3 and Rrp6 (61,

137, 153, 219).

The current paradigm posits that a single stoichiometric species of exosome complex is responsible for all ascribed RNA metabolic functions (102, 195, 221).

In this model, called here the core exosome model, all functions and contributions of individual subunits to distinct RNA processing and turnover events occur only in the context of a stoichiometric complex. This idea is

supported by work in trypanosomes showing that certain subunits are co-

depleted when other subunits are targeted by RNAi thus presumably

destabilizing the core complex (65). However, biochemical, cell biological,

bioinformatic, and genetic evidence from recent work indicates that certain

proteins, especially Rrp6, and Dis3 function independently of other exosome

subunits (33, 87, 196) and form subcomplexes (86, 88).

The core exosome model predicts that depletion or mutation of individual

exosome subunits should yield similar results on exosome complex surveyed

RNAs. Exactly such results, a reproducible transcript set shared by all

components, were obtained when components of the mRNA export and NMD

machinery were depleted (92, 185, 186).

The results from a previous microarray study with RNA harvested from

rrp41-1, rrp6∆, and rrp47∆ yeast strains shows that many unique mRNAs are

stabilized in on a transcriptomic level (100). This observation was confirmed and

extended upon with tiling arrays using RNA harvested from depletions (Rrp4,

Rrp41) or an exosome subunit null mutant (csl4-2) in plants (39). The results of

both of these studies contradict the predictions of core exosome model, and

suggest that an alternative model of exosome subunit function is needed.

To better characterize the functions of individual exosome subunits in the

context of the core in vivo, we carried out a nearly comprehensive microarray

study by using RNAi to deplete exosome subunits in Drosophila S2 cells.

Roughly 80% of the affected transcripts were increased when compared to a

GFP dsRNA treated control. In general, the affected transcripts had long UTRs, and known exosome targets, including NMD transcripts, were enriched in our data set. At the level of individual transcripts, the experiments yielded distinct profiles when different subunits were depleted; however, the cellular pathways in which those altered transcripts function were similar. Our data show complexity for exosome subunit mediated RNA metabolic processes and suggest a more dynamic interplay between and among exosome subunits within the exosome complex or subcomplexes.

2.C. Materials and methods

2.C.1. T7 primers and double-stranded RNA preparation.

T7 primer sequences used to generate exosome subunit specific dsRNAs are

listed in Table 4 and dsRNA was prepared as described earlier (87, 88).

Generally, dsRNAs were between 400 and 700 base pairslong and targeted the

5’ portions of exosome subunit mRNAs. Briefly, in vitro transcription templates

were generated by PCR using exosome subunit-specific primers with T7

overhangs and transcribed using a MEGAscript T7 kit (Ambion, Austin, TX)

Table 4. T7 oligonucleotide primers for dsRNA synthesis.

Oligo name Sequence

T7/dMtr3F GAATTAATACGACTCACTATAGGGAGAGGTGAACCAGACGAAGAACACC T7/dMtr3R GAATTAATACGACTCACTATAGGGAGAGCAGCTTTACATAGGTGGCTGG T7/dSki6F GAATTAATACGACTCACTATAGGGAGAATGTCTGGTAAATACGATCTGC T7/dSki6R GAATTAATACGACTCACTATAGGGAGACAACTTTCGCAAAGTCAGAGGC T7/dRrp46F GAATTAATACGACTCACTATAGGGAGAGAGTGTAACTACCGACCAAAAGC T7/dRrp46R GAATTAATACGACTCACTATAGGGAGAGCACTCAATATCATTGAACTGGG T7/dRrp40F GAATTAATACGACTCACTATAGGGAGACACGGTGGTGGCTAGCAAGGCGG T7/dRrp40R GAATTAATACGACTCACTATAGGGAGAGGAATGGGCCTTCAACCATATCC T7/dRrp4F GAATTAATACGACTCACTATAGGGAGAGGTGGTGGCCCGCGTCAGCGAGG T7/dRrp4R GAATTAATACGACTCACTATAGGGAGAGGTGTCGTAAATCATCAGCTTGC T7/dCsl4F GAATTAATACGACTCACTATAGGGAGAGCCTGCCCGGGGAGCGACTCTGC T7/dCsl4R GAATTAATACGACTCACTATAGGGAGAGGCAGAACTTAGCCACTTTCCG T7/dDis3F GAATTAATACGACTCACTATAGGGAGAGGCCGTGGATGGCGACCTGG T7/dDis3R GAATTAATACGACTCACTATAGGGAGAGTAATTGTCCAGGGCATTTTCGG T7/dRrp6F GAATTAATACGACTCACTATAGGGAGACTAATGGTGGTTGACACGGTGG T7/dRrp6R GAATTAATACGACTCACTATAGGGAGACGATGTGAGGCTTATTATAACGC T7/dRrp47F GAATTAATACGACTCACTATAGGGAGAACAGCAGCATCGAACTCCTGGA T7/dRrp47R GAATTAATACGACTCACTATAGGGAGACAACTAAGATATCATATCGTCGG T7/dGFPF GAATTAATACGACTCACTATAGGGAGAGAATTAATACGACTCACTATAGGGAGA T7/dGFPR GAATTAATACGACTCACTATAGGGAGAATGTTATTTGTATAGTTCATCCATGC

according to the manufacturer’s instructions. The resulting RNA was purified by

LiCl precipitation (-20°C, overnight), resuspended, and annealed by incubation at

two temperatures (5 minutes 95°C, and 90 minutes at 70°C) followed by slow

cooling to room temperature (~2.5 hour cool time). The concentration of double-

stranded (dsRNA) was determined by UV spectrophotometry. Annealing was

verified by agarose gel electrophoresis. Only dsRNAs with ≥ 95% of the RNA in

a single band were used for treating cells. dsRNAs that did not meet that

requirement were re-annealed or discarded.

2.C.2. S2 cell tissue culture and dsRNA treatment.

D. melanogaster embryonic S2 cell lines were grown in Hyclone HyQ-CCM3

media (Logan, UT) at 27°C. 106 S2 cells were seeded into 25 cm2 flasks,

allowed to recover for 24 hours, treated with dsRNAs (30µg/ml) on days 0, 1, and

3, and resuspended by pipetting each time dsRNA was added. RNAi depletions

for microarray experiments were harvested on day 5 (87). A portion of each

culture was reserved for western blotting as described previously (88). For

extended dsRNA treatment and counting experiments, 106 Day 5 dsRNA treated

cells were seeded into fresh 25 cm2 flasks with or without continued dsRNA

treatment and counted on days 6, 8, and 10.

2.C.3. CuSO4 induction of exo-FLAG expressing cell lines and cell counting

106 S2 cells were seeded into 100 mm2 petri dishes, allowed to recover for 24 hours, and treated with 1.4 mMol CuSO4, on day 0 only. On days 0, 2, 4, 6, 8,

10, and 12, the culture volume was standardized by adding additional media,

cells were resuspended by pipetting, and counted by hemacytometer.

2.C.4. Western blotting

Efficiency of exosome subunit depletion was verified for each dsRNA

treatment by western blotting as follows. Two sets of westerns were run for each

experiment. The first set of westerns was used to determine loading and to

approximate depletion efficiency. Equal volumes of S2 cells boiled in 1x SDS gel

loading dye were loaded into 8% or 10% SDS-PAGE gels and run at 150 volts for

~2.5 hours. Samples were then electroblotted (~250 total volt hours, usually overnight at 20 volts) onto nitrocellulose membranes and blocked for at least one hour using 5% milk in TBST. Blots were cut into strips and probed with primary antibodies using different dilutions (1:2000 for most exosome subunits, 1:4,000

for Rrp4, 1:40,000 for Lamin Dm0, a gift of Dr. Paul Fisher (Stony Brook

University, Stony Brook, NY)) and rocked for ~4 hours at room temperature, or overnight at 4°C. Strips were rinsed, re-blocked, and HRP conjugated secondary antibodies, Jackson Immuno-Research, were added (1:10,000 dilution) for at least 2 hours at room temperature. Bands were visualized using an enhanced chemiluminescence kit (Amersham). The intensity of an internal control band

Lamin Dm0, or mono-clonal anti α-Tubulin, Sigma (1:2000 dilution) was used to verify the reproducibility of the Lamin control, was determined using Quantity

One software (Biorad) for each experimental (exosome subunit dsRNA) and control (Mock and GFP dsRNA treated) sample.

The second set of Westerns was carried out as above, but loading was

determined by normalizing the counts of Lamin to be loaded into each lane.

Final Lamin Dm0 and exosome subunit band intensities were calculated and %

depletion was calculated by comparing the normalized (targeted exosome

subunit signal normalized to Lamin Dm0 in that lane) signals from exosome

subunit depleted lanes to signals from both GFP and mock treated cells.

2.C.5. Microarray experiment design

Once depletion of the subunits was verified by western blotting, 50 µg

samples of total RNA from independent, duplicate RNAi depletions (RNAi

targeted protein reduced to <10% of normal levels as judged by western blotting)

were sent to the Gene Expression and Genotyping Core Facility, Case

Comprehensive Cancer Center (http://www.gegf.net/web/?page_id=2, Cleveland,

OH). Samples were column purified, prepared, and hybridized using Genechip

Drosophila_2 microarrays (Affymetrix), washed, and data acquired (GeneChip

Fluidics Station 450, GeneChip Scanner 3000, GeneChip Command Console

Software Version 1.1) according to Affymetrix guidelines. The resulting data

were analyzed using GCOS (Affymetrix), Microsoft Access, NetAffx (Affymetrix),

and Genespring GX 10 (Agilent) as described earlier (87). In compliance with

the Minimum Information About a Microarray Experiment (MIAME) standards, the raw and normalized data, plus a short summary of the experimental procedures discussed in this document have been deposited in NCBI's Gene Expression

Omnibus and are accessible through GEO Series accession number GSE17874

(62). (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE17874).

2.C.6. Microarray controls and statistics

Non-specific transcripts were limited in the following ways. First, RNA

harvested from GFP dsRNA treated cells was used as our baseline sample. This

was to account for RNAs whose expression was altered simply by dsRNA

treatment. Second, GCOS software (Affymetrix) assigned P-values to changed

transcripts based on fourteen probeset pairs per transcript per array (GFP and

exosome subunit depleted). Only transcripts with change P-values of <0.005

(increases) or >0.995 (decreases) were retained. Third, transcripts with fold changes below ±2 fold were excluded from further analysis. Finally, all arrays were performed using RNA harvested from independent duplicate exosome subunit depletions and compared against arrays using RNA harvested from independent duplicate GFP dsRNA treated controls. This regime yielded four total comparisons . For the purposes of this study, a reproducibly changed transcript’s expression is altered at least ±2 fold (in the same direction, with a P- value threshold of 0.005) in 3 or 4 independent exosome subunit depleted to

GFP array comparisons (4 total comparisons were made). All reproducibly changed (as outlined above) transcripts for each subunit were grouped into a depleted exosome subunit microarray profile (desMAP) Proportional Venn and

Euler diagrams were adapted from the following website:

http://www.cs.kent.ac.uk/people/staff/pjr/EulerVennCircles/EulerVennApplet.html

2.C.7. Grouping of pathways

Increased and decreased gene lists for each subunit were separately

imported into Pathway Studio 5.0 (Ariadne). Pathway Studio 5.0 (Ariadne) then

was directed to search for all connected and related proteins and biological

pathways. Significant associations (P-value <0.05) were grouped by similarity and tallied as percent of total significant pathways.

2.C.8. UTR analysis

All data for both 3’ and 5’ UTRs were downloaded from Flymine at

http://www.flymine.org/release-18.0/begin.do. Final 5’ and 3’ UTR lengths of

genes with multiple annotated UTRs were determined by averaging all the 5’ or

3’ UTRs reported for that particular gene. The 5’ and 3’ UTR lengths of affected

genes, both individually and in aggregate, were determined by matching gene

identifiers with the averaged Flymine UTR datasets.

2.C.9. NMD analysis

Many core Drosophila NMD targets been identified in a previously published

work (185). To standardize nomenclatures between two different arrays

(Drosgenome1, (185), and Drosophila_ 2, this study), probe and gene identifiers

of affected genes from each study were uploaded into the NetAffx database

(Affymetrix), http://www.affymetrix.com/analysis/index.affx. The 74 gene

identifiers that were present in both studies were compared to the desMAPs of

76 individual exosome subunits to evaluate which NMD targeted transcripts were present in each desMAP.

2.D. Results

2.D.1. Exosome subunits can be depleted in S2 cells

To determine the functions of individual exosome subunits we used dsRNAs to deplete them in Drosophila S2 tissue culture cells. Depletion was verified by

Western blotting and percent depletion was calculated by comparing to a loading standard, Lamin Dm0 (Figure 7). Depletion was very effective for most subunits with some depleted subunits approaching the detection level of the antibody

(Figure 7). Only cells in which subunits were verifiably depleted to below 10% of endogenous levels were used for further analysis. Drosophila lacks an obvious

Rrp43 homologue and Rrp42 and Rrp45 did not verifiably deplete to below 10%.

Hence, these subunits are not included in most areas of this study.

2.D.2. Depletion of select exosome subunits affects the stability of a subset of other subunits

Earlier work on Trypanosomes and human tissue culture cells has shown that depleting certain exosome subunits resulted in the partial proteasome-mediated destabilization of other exosome subunits (65, 215). To determine if exosome subunits behaved in a similar manner in the Drosophila complex, we examined the stability of a large number of exosome subunits following dsRNA treatment

(Figure 8a). Depletion of certain subunits resulted in one or more co-depleted

Figure 7. Exosome subunits are effectively depleted using dsRNAs. S2 cells were treated with exosome subunit specific dsRNAs (Exo) as described in methods. Effectiveness of dsRNA treatment was determined by western blotting and depletion was calculated by comparing bands to a loading standard

(Lamin Dm0). Representative blots for cell proliferation experiments are shown for each depleted exosome subunit.

Figure 8. Some exosome subunits are co-depleted when certain subunits are targeted with dsRNAs.

(A) Cells were depleted of exosome subunits and western blotted as earlier.

Blots were cut into strips and probed with different antibodies. A blot probed with

a Lamin Dm0 specific antibody serves as a loading control. (B-E) Schematics

depicting the co-depletion relationship between individual subunits.

79 subunit(s) being reduced to ~25 - 40% of normal levels as judged by Western blotting (Figure 8a). The co-depletions are summarized in schemetic form in

Figures 8b-8e. In general, depletion of RNAse PH domain subunits (Mtr3, Ski6, and Rrp46) caused the co-depletion of one or more additional subunits.

Depletion of Rrp42 or Ski6 co-depleted either 5 (Figure 8d) or three subunits

(Figure 8e). Depletion of Rrp46, Mtr3, or Rrp6 each resulted in the co-depletion of one additional subunit (Figure 8c). However, targeting the S1 domain subunits

(Rrp40, Rrp4, and Csl4) resulted in no co-depletion (Figure 8). Thus, we do not obtain a broad destabilization of subunits following depletion as might be expected for core complex disassembly (62).

2.D.3. Exosome subunits are not required for cell proliferation

It has been previously demonstrated that Rrp6 is required for cell proliferation whereas Rrp40 is not (87). I expanded these viability studies to include most of the remaining exosome subunits (Figure 9). As with Rrp40 depleted cells; most exosome subunit depleted cells continued to proliferate (Figure 9a). Only the depletion of Rrp6 or Dis3 slowed cell proliferation noticeably (Figure 9a).

Further, most cells continued to proliferate at near-normal rates even when they were followed for a longer time period in the presence or absence of continued dsRNA treatment (Figure 9b and 9c). Rrp4-depleted cell grew slowly when the dsRNA treatment was extended to 10 days, but these cells proliferated normally when the dsRNA treatment was discontinued after five days (cf. Figure 9b and

9c). These data indicate that most core exosome subunits, and hence, by

Figure 9. Most exosome subunits are non-essential in D. melanogaster S2 cells.

(A) During dsRNA treatment, cells were counted by hemocytometer at designated times. On day five, cells were diluted (to 106 cells/ml) and cultured in

the (B) presence or (C) absence of continued dsRNA treatment.

81 inference, the core exosome, are not essential in D. melanogaster S2 tissue culture cells (Figure 9).

Over-expression of certain exosome subunits also has little effect on cell proliferation (Figure 10). For example, over-expression of Ski6, Dis3, Rrp4,

Rrp47, and Rrp42 has only small effects on cell proliferation (Figure 10b), while over-expressing other exosome subunits including Mtr3, Csl4 and Rrp40 noticeably reduced cell proliferation (Figure 10). As with Rrp6 depleted cells

(Figure 9), S2 cells over-expressing Rrp6 showed a marked decrease in cell proliferation (Figure 10). Finally, Mtr3 over-expressing cells grew very poorly

(Figure 10). This was unexpected as Mtr3 depletion had no effect on cell proliferation (Figure 9). Rrp46, Rrp40, and Csl4 overexpressing cells also proliferated poorly, but the effect was not as pronounced as with Mtr3 and Rrp6 overexpressing cells (Figure 10). However, these results are difficult to interpret since most induced S2 cells, including the stable cell line expressing an empty

Metallothionein (mtn) expression vector failed to reach a density comparable to uninduced cells (Figure 10b). Presumably this slowed growth is caused by the addition of CuSO4 which is required to induce expression of the mtn-driven expression constructs.

2.D.4. Gene expression arrays identify exosome subunit surveyed mRNAs

To understand better the contributions of individual exosome subunits to exosome complex function in vivo, we performed microarray analyses using RNA isolated from exosome subunit-depleted (below 10% of normal levels) S2 cells after five days of dsRNA treatment. RNA isolated from cells depleted of the

Figure 10. Overexpression of certain exosome subunits hinders cell proliferation.

Stable cell lines expressing tagged exosome subunits were (A) cultured normally or (B) induced with CuSO4 and counted over a twelve day time course as

explained in methods.

83 exosome subunits Mtr3, Ski6, Rrp46, Rrp40, Rrp4, Csl4, Dis3, Rrp6, and the exosome cofactor Rrp47 was assayed using Drosophila_2 microarrays

(Affymetrix). We limited non-specifically affected transcripts in four ways and defined each depleted exosome subunit microarray profile (desMAP) as containing only reproducibly changed transcripts as detailed extensively in sections 2.C.5. and 2.C.6. of Methods and Materials.

For all exosome depletions, 6117 transcripts (5043 increased and 1074 decreased) were identified as reproducibly changed at least two-fold (Table S3).

When transcripts present in multiple desMAPs were accounted for, a total of

2328 (1909, or 82%, increased and 419 decreased), out of the 6117, transcripts were identified (Figure 11, Table S3).

2.D.4.a. DesMAPs of individual subunits do not reveal a core set of transcripts surveyed by exosome subunits

Depletion of Mtr3, Ski6, Rrp4 and the exosome cofactor Rrp47 affected the largest number of (between 900 and 1200) transcripts (Figure 12, Table 5). In contrast, depletion of Rrp6, the subunit commonly used to define the nuclear functions of the exosome, altered the fewest (~400 transcripts) (Figure 12, Table

5). Only one desMAP, that of Dis3, contained more decreased than increased transcripts (Figure 12, Table 5). Depleting Mtr3, Ski6, Rrp4, and Rrp47 resulted in a very large pool of increased transcripts (greater than 87% of each profile was increased) and a comparatively small pool of decreased transcripts (Figure 12,

Table 5). Depletion of the remaining subunits showed an approximately 3:2 ratio

419

1909

Decreased Increased

Figure 11. mRNAs affected by exosome subunit depletion are predominantly increased.

The transcripts from all desMAPs were pooled and grouped by fold change.

1200

1000

800

600

400 Decreased 200 Increased Number of mRNAs 0

-200

-400 Mtr3 Ski6 Rrp46 Rrp40 Rrp4 Csl4 Dis3 Rrp6 Rrp47 Subunit Depleted

Figure 12. desMAP size varies by subunit.

The number of increased (red bars) and decreased transcripts (blue bars) are shown for each desMAP.

Table 5. Numbers of transcripts changed at least two-fold in each desMAP.

desMAP Increased Decreased Domain Mtr3 1123 63 RNase PH Ski6 1114 58 RNase PH Rrp46 268 164 RNase PH Rrp40 269 168 S1 Rrp4 885 1 S1 Csl4 319 153 S1 Dis3 209 261 RNase II/R Rrp6 131 98 RNase D Rrp47 725 110

of increased to decreased transcripts (Figure 12, Table 5). Interestingly, the

increased to decreased transcript ratio was not shared by subunits that had

similar domains (Figure 12, Table 5). For example, the desMAPs of the RNase

PH domain subunits Mtr3 and Ski6 were very different from the desMAP of the

remaining RNase PH domain subunit, Rrp46 (Figure 12, Table 5). The same

held true for the S1 domain subunits as the desMAP of Rrp4 was very different

from the desMAPs of Rrp40 and Csl4 (Figure 12, Table 5). Likewise, the overall

number of affected transcripts in each desMAP was also domain independent

(Figure 12, Table 5).

The heat map (Figure 13) depicts the desMAPs of all depleted exosome subunits. The changed transcripts vary greatly in number, identity, and the

degree of change among desMAPs; however, several desMAPs have many

transcripts in common (Figure 13, and Table S1-S7). These transcripts were

then grouped based upon the number of desMAPs in which they occurred

(Figure 14, Table S3). Over one third (~37%), or 875, of the affected transcripts

were found only in a single desMAP, with 1783 (~76%) being present in three or

fewer desMAPs (Figure 14, Table S3). Only 0.4% of unique transcripts (10 total,

all increased) were identified in all nine desMAPs (Figure 14, Table 6, S3). A

larger number of the transcripts were also present in seven (85) or eight (69)

desMAPs (Figure 14, Table S3). Even when the data were weighted to account

for transcripts that appeared in multiple desMAPs, only ~1.5% of transcripts were

present in all desMAPs, while over half (51%) were found in three or fewer

desMAPs (Figure 15, Table S3).

Figure 13. The affected mRNAs in the desMAPs of different subunits vary in both identity and effect.

A heat map showing changes in mRNA levels by subunit. Unchanged mRNAs

(fold change less than ±2) are shown in yellow, decreases (≤-2 fold change) are shown in shades of blue, smaller increases (2-3.5 fold changes) in shades of orange, and finally large increases (>3.5 fold change) in shades of red.

Figure 14. Most mRNAs are found in few desMAPs.

All affected mRNAs were grouped by the number of desMAPs in which they appeared.

Table 6. mRNAs affected in all 9 desMAPs. Abbreviations: FC, fold change; P, detected (present); I, increased

Number of Baseline RNAi Minimum Maximum Probe ID Subunits Detection Detection FC FC Change Description 1639232_s_at 9 P P 2.55 11.91 I CG11956 1638469_s_at 9 P P 2.59 12.51 I AY180918 1631007_at 9 P P 2.71 7.49 I CG5371 1623281_s_at 9 P P 2.83 11.31 I CG17949 1632430_at 9 P P 2.96 8.32 I CG6186 1624517_at 9 P P 3.52 7.27 I CG3132 1627000_s_at 9 P P 4.23 17.94 I CG6231 1635715_at 9 P P 4.23 11.50 I CG31611 1623296_at 9 P P 4.68 8.36 I CG32625 1639729_s_at 9 P P 5.29 54.26 I Z27119

1.47%

1 of 9

9.01% 2 of 9 14.29% 3 of 9 9.72% 4 of 9 15.22% 6.56% 5 of 9 6 of 9 10.12% 7 of 9

21.65% 8 of 9 11.95% 9 of 9

Figure 15. Weighted distribution of all transcripts in multiple desMAPs.

Transcript counts were weighted based upon the number of desMAPs to which an individual transcript belonged.

I next examined the degree of overlap between individual desMAPs. I

catalogued the transcripts that were present in each possible pair-wise desMAP comparison (increased transcripts, Table 7 (top) and decreased transcripts,

Table 7 (bottom)). All overlapping transcripts for each possible binary comparison are detailed in Table S4 and the sub-tables therein. The desMAPs of Mtr3 and Ski6 shared the most (~85%) transcripts of any two subunits, while the desMAPs of Rrp6 and Rrp4 had the fewest common (~16%) transcripts

(Figure 13, Table 7). Importantly, nearly a third of these ‘shared’ were altered in opposite directions in the desMAPs of Rrp6 and Rrp4, therefore the percentage of transcripts altered in the same direction was closer to ~11% (Table 7 and

Table S4). The desMAPs of Rrp6 and Dis3 are the two most divergent, as judged by the proportion of shared transcripts, from the desMAPs of other exosome subunits (Table 7 and Table S4). As observed in previous microarray studies, and predicted by the exozyme model, these data show a large and variation in the transcriptomic profiles of individual exosome subunits and cofactors (39, 100, 178).

2.D.4.b. Structurally similar subunits have distinct desMAPs

I previously observed that the domain structure of the depleted subunit was not a reliable predictor of desMAP size or the ratio of increased to decreased transcripts in a particular desMAP (Figure 12, Tables 5, 7). Although the sizes

of the desMAPs were very different, I wanted to determine the portion of the

desMAPs that were shared among subunits with similar domain structures.

When the transcripts in the desMAPs of the three S1 domain-containing subunits

Table 7. Number of transcripts shared in all binary comparisons.

Shared Increases Mtr3 Ski6 Rrp46 Rrp40 Rrp4 Csl4 Dis3 Rrp6 Rrp47 Mtr3 947 241 258 610 277 158 44 419 Ski6 49 241 251 594 260 161 44 321 Rrp46 53 51 208 206 143 99 36 132 Rrp40 51 49 126 220 149 95 23 136 Rrp4 1 1 0 1 180 111 26 228 Csl4 55 48 84 89 1 123 28 281 Dis3 56 51 111 95 1 102 38 137 Rrp6 20 22 39 34 0 32 52 33 Rrp47 47 40 59 60 1 87 71 24 Shared Decreases

Figure 16. Most of the transcripts in the desMAPs of S1 domain subunits are not shared between the three subunits.

Venn diagram showing the number of transcripts shared by the desMAPs of S1

domain subunits.

(Rrp40, Rrp4, and Csl4) were compared, only 10% of them were present in all

three desMAPs (Figure 16). Of the remaining transcripts affected by depleting

these three subunits, approximately 70% were present only in one S1 domain

desMAP, and just over 20% were present in two (Figure 16, and Table S5).

When comparing these subunits individually, they had desMAPs that were

between 80% (Rrp4 versus Csl4) and 50% (Csl4 versus Rrp40) different (Figure

16, and Tables 7 and S5).

Next, I wanted to determine if transcripts conserved among the S1 domain

subunits were more likely to be represented in the desMAPs of other exosome

subunits. To accomplish this, I cross-referenced the transcripts present in one, two, or three of the S1 domain subunit desMAPs with the pooled transcripts from all desMAPs after they had been ranked by the number of desMAPs in which they appeared (e.g. Tables 6, S4) (Figure 17). Predictably, the transcripts which

were present in only one S1 domain subunit desMAP (Figure 17a) were

clustered towards the left (non-conserved) end of the graph, while those present

in all three S1 domain desMAPs clustered with the conserved transcripts (Figure

17c).

I also compared the desMAPs yielded by depleting the RNase PH domain

containing exosome subunits (Mtr3, Ski6, and Rrp46) in the same manner. This

comparison yielded a different result than with the previous one with the S1

domain subunits. In this analysis, ~30% of transcripts were found in only one

RNase PH domain desMAP, 50% were present in two of three desMAPs, and

~19% were present in all three RNase PH subunit desMAPs (Figure 18 and

Figure 17. Transcripts present in all three S1 domain desMAPs are most likely to be in other exosome subunit desMAPs.

The transcripts (A) unique to a single, (B) present in two or (C) common to all three S1 domain subunit desMAPs are compared to the overall distribution of transcripts.

Figure 18. Transcripts present in the desMAPs of RNase PH subunits are more conserved than those of S1 domain subunits.

Venn diagram showing the number of transcripts shared by the desMAPs of S1

domain subunits.

Table S6). The vast majority of transcripts present in two desMAPs (718, or

~96%) were shared by Mtr3 and Ski6, with very few shared by either Rrp46 and

Mtr3 (17 transcripts) or Rrp46 and Ski6 (15 transcripts) (Figure 18). Nearly two- thirds of Rrp46’s desMAP (64%) was conserved with the desMAPs of both Mtr3 and Ski6 (Figure 18). As with the transcripts present in all three S1 domain subunit depletions, the RNAs present in all three RNase PH domain subunit depletions were much more likely to be present in additional desMAPs (Figure

19c). However, the RNAs present in two RNase PH desMAPs, ~96% of whichare in the desMAPs of Mtr3 and Ski6 cluster differently than those present in two S1 domain subunit desMAPs (c.f. Figures 17b and 19b). Whereas the transcripts present in two S1 domain subunit desMAPs are predominantly found in the desMAPs of three or more additional subunits (Figure 17b), the transcripts shared by the desMAPs of Mtr3 and Ski6 were unique to those two desMAPs

(~28%) or shared with only one additional (46%) exosome subunit (Figure 19b).

In general, although the transcripts stabilized by the depletion of S1 domain containing exosome subunits are less conserved within that group, they are more likely to be represented in the desMAPs of other, unrelated exosome subunits than those stabilized by RNase PH subunit depletion (Figures 17 and 19).

These results show that even the depletion of structurally similar subunits affects distinct sets of mRNAs (Figures 16 - 19).

I next examined the transcripts from the structurally distinct subunits (Dis3 and Rrp6) and an exosome cofactor (Rrp47). When compared to all the other desMAPs, the desMAPs of these three polypeptides were considerably more

Figure 19. Transcripts present in all RNase PH-domain desMAPs are more likely to be shared with other exosome subunit desMAPs.

The transcripts (A) unique to a single, (B) present in two or (C) common to all three RNase PH domain subunit desMAPs are compared to the overall distribution of transcripts.

100

distinct than those of the S1 domain and RNase PH domain subunits (Figure 20,

and Tables 5 and S5). The transcripts in the desMAPs of these three subunits

were almost evenly spread amongst the distribution of all other subunits (Figure

20). Examined in its entirety, the desMAP of each subunit was distinct, but the desMAPs of the RNase PH domain subunits, especially Mtr3 and Ski6, were more similar to one another than other exosome subunits (Figure 18, Table 7).

2.D.6. The UTRs of affected genes are longer than transcriptomic average.

I next wanted to determine if the mRNAs identified by my arrays shared

sequence and/or regulatory elements. Exosome subunits have previously been

linked to the surveillance of transcripts with specialized UTR elements, such as

the AREs of c-fos and TNF mRNAs (9, 41, 83, 84, 136, 180). This link between

exosome subunit mediated mRNA surveillance and UTR elements located in the

targeted mRNAs led me to reason that transcripts with longer UTRs, and hence a

higher likelihood of containing one or more UTR element(s), would be more

prevalent in the desMAPs. I obtained the transcriptome-wide 5’ and 3’ UTR

libraries (poly-A tail excluded) for D. melanogaster from Flymine

http://www.flymine.org/release-18.0/begin.do. To avoid arbitrarily selecting a

representative UTR length in cases where UTRs of multiple lengths were

annotated for a single transcript, the average of available UTR sizes was used.

In total, 304 and 290 of the 2328 affected transcripts in our desMAPs either

lacked annotated 5’ or 3’ UTRs respectively, or lacked UTRs entirely. These transcripts were excluded from further analysis.

101

Figure 20. The desMAPs of Dis3, Rrp6 and Rrp47 show little overlap.

Transcripts profiles of (A) Dis3 (B) Rrp6 and (C) Rrp47 are compared to the entire microarray data set.

102

As in other organisms, the transcriptome-wide average length of the D. melanogaster 5’ UTR, ~216 nt, was shorter than the average 3' UTR, ~365 nt,

(Figure 21). The average 5’ and 3’ UTR lengths of the genes affected by exosome subunit depletion were ~255 nt and ~415 nt respectively, which are significantly (P-value <0.0001 for both UTRs) longer than the transcriptome-wide average (Figure 21). Further, in the case of 3’ UTRs these differences were more substantial in decreased transcripts (522 average 3’ UTR length, P-value

<0.0001) than in increased transcripts (390 average, P-value = 0.0212; Figure

21). The same bias was not seen with 5’ UTRs as both increased and decreased transcripts had almost identical 5’ UTR lengths (Figure 21).

2.D.6.a. Length of dsRNA is not a predictor of UTR length

Several studies have indicated that RNAi treatments can result in the degradation of non-targeted mRNAs via off-target effects (107-109). To determine whether the dsRNAs used to deplete the exosome subunits could produce off-target effects I first cross-analyzed the lengths of the dsRNAs and the number of affected transcripts in each desMAP. While Mtr3 and Ski6 have both the longest dsRNAs, 751 and 735 bp respectively, and the largest desMAPs

(Figure 22a, Table 8) the dsRNAs used to deplete Rrp4 and Rrp47, the next two largest desMAPs by transcript number, used average length (Rrp4) or the shortest dsRNA (Rrp47) in this study (Figure 22a, Table 8). Further, of the 4 subunits with the largest number of affected transcripts a very large fraction, over

87% (>95% for Mtr3, Ski6, and Rrp4) are increased (Figure 12) where the

103

Figure 21. The UTRs of affected transcripts are longer that the transcriptome-wide average in D. melanogaster.

The average 5’ and 3’ UTR lengths of genes in each desMAP (affected) were compared to genome-wide average UTR lengths. Average 5’ and 3’ UTR lengths for increased and decreased RNAs are also represented.

104

Table 8. Lengths of dsRNA used.

dsRNA Subunit length (nt) Mtr3 751 Ski6 735 Rrp46 515 Rrp40 480 Rrp4 519 Csl4 555 Dis3 542 Rrp6 594 Rrp47 412

105

Figure 22. dsRNA lengths do not predict UTR lengths.

The length of the dsRNA used to deplete each exosome subunit was plotted opposite the number of (A) total or (B) decreased transcripts in each desMAP. A trendline shows the slight negative correlation of decreased transcripts and dsRNA length.

106

preference for longer 3’ UTRs was less pronounced than in decreased transcripts

(Figure 21). Since longer 3’ UTRs were more prevalent in decreased transcripts

(Figure 21) I next plotted the number of decreased transcripts per desMAP

versus the length of the dsRNA used in that depletion. There is a slight negative

(Figure 22b, trendline) bias against longer dsRNAs producing more decreased

transcripts. Collectively, these metrics suggest that the correlation between

exosome subunit depletion and increased UTR lengths is specific, and suggests

that they do not result from off-target effects.

2.D.6.b. Certain subunits stabilize transcripts with longer UTRs

Based upon the observation that the depletions favored transcripts with longer than average UTRs, I wanted to determine if UTRs of a particular length were favored in our desMAPs. The transcriptome-wide distribution of different UTR lengths was determined for both 5’ and 3’ UTRs (Figure 23, right Y- axes) and graphed opposite the distribution of UTR lengths of affected mRNAs in my microarray data set (Figure 23, left Y-axes). I performed similar, but more extensive comparisons for each individual desMAP. I obtained data for both increased and decreased transcripts for both the 5’ and 3’ UTRs of mRNAs in each desMAP (Figures 25-28). Upon examining the UTR distributions of

individual desMAPs, most resembled the entire study average for all desMAPs

(Figure 23). However, there were some large differences observed in certain

subunit distributions (Figures 25-28). The largest disparity between the average

107

Figure 23. The UTR length distributions of affected transcripts favor longer

UTRs.

The distribution of 5’ (A) and 3’ (B) UTR lengths in all desMAPs, left Y-axes, plotted opposite the transcriptome-wide distribution of UTR lengths, right Y-axes.

108

Figure 24. The UTR distributions of Rrp4's and Rrp6's desMAPs diverge from the transcriptomic average.

The 5’ and 3’ UTR length distributions of transcripts increased in the desMAPs of

(A, B) Rrp4 and (C, D) Rrp6, left Y-axes, plotted opposite the transcriptome-wide distribution of UTR lengths, right Y-axes.

109

of all desMAPs and individual desMAPs were observed in transcripts increased

in Rrp4 and Rrp6 depleted cells (Figure 24). Briefly, the distribution of mRNAs

increased in the desMAP of Rrp4 is skewed towards mRNAs with longer 5’ and

3’ UTRs (Figure 24a and b), while those increased in the desMAP of Rrp6 tend to contain shorter UTRs (Figure 24c and d). Upon further inspection, mRNAs with

5’ and 3’ UTRs less than 100 nt long are under-represented, when compared to the transcriptome-wide distribution, in the desMAP of Rrp4 (Figures 24a, b).

Likewise, mRNAs with UTRs longer than 100 nt are enriched in Rrp4’s desMAP

(Figures 24a, b). Almost the opposite observation is true for transcripts increased by Rrp6 depletion (Figures 24c, d). When examining the 5’ UTR length

distribution of transcripts increased in the desMAP of Rrp6, the distribution is

tilted to the left side of the graph showing that Rrp6 depletion stabilizes mRNAs

with shorter UTRs (Figures 24c). A similar, but considerably less pronounced

effect is observed when 3’ UTRs are examined (Figures 24d).

2.D.6.d.1. UTR patterns in increased and decreased transcripts

The UTR length distributions of the remaining desMAPs fall between the two

extremes exemplified by Rrp4 and Rrp6 and are graphed in Figures 25-28.

When examining the 5’ UTR distribution of increased transcripts, only the

desMAP of Rrp6 had an appreciably increased prevalence of mRNAs with UTRs

less than 100 nt (Figure 24c and 25h). Overall, mRNAs with 5’ UTRs of between

100-1000 nt are enriched to varying degrees in most desMAPs (Figures 25 and

26). Also, very long (e.g. 1500+ nt) 5’ UTRs are almost absent from the

desMAPs (Figures 25 and 26). The 5’ UTR distributions of decreased transcripts

110

Figure 25. The 5' UTR length distributions of increased transcripts

The 5’ UTR length distribution patterns of all increased transcripts in the desMAPs of (A) Mtr3, (B) Ski6, (C) Rrp46, (D) Rrp40, (E) Rrp4, (F) Csl4, (G)

Dis3, (H) Rrp6, and (I) Rrp47 are shown. The patterns of Rrp4 and Rrp6 from

Figure 25 are reproduced here to allow for easier comparisons among all

subunits.

111

Figure 26. The 5' UTR length distribution of decreased transcripts.

The 5’ UTR length distribution patterns of all decreased transcripts in the desMAPs of (A) Mtr3, (B) Ski6, (C) Rrp46, (D) Rrp40, (E) Csl4, (F) Dis3, (G)

Rrp6, and (H) Rrp47 are shown. The pattern of Rrp4 is excluded as that desMAP contained only one reproducibly decreased transcript.

112 show less variability from subunit to subunit than the distributions of increased transcripts (c.f. Figures 25 and 26). The desMAP of Rrp4 has been excluded from the analysis of decreased UTR profiles as this desMAP only contained one decreased mRNA (Figure 12, Table5). These distributions show that all desMAPs, with the exception of Rrp46 and Csl4, have a slightly increased occurrence of 5’ UTRs that are <100 nt in length (Figures 25 and 26). Other than this enrichment for the shortest 5’ UTRs observed with decreased transcripts in most desMAPs, these distributions show a general preference for longer UTRs

(Figure 26).

The pattern was more complicated when the 3’ UTRs of increased transcripts were examined (Figure 27). The UTR distribution of one subunit, Rrp47, was the only one which showed a marked increase, when compared to transcriptomic average, in transcripts with UTRs <100 nt (Figures 27i). The 3’ UTR distributions of both Dis3 and Csl4 also showed a small preference for these transcripts with short UTRs (Figures 27f and g). The predominant 3’ UTR distribution pattern for increased transcripts between all subunits showed that mRNAs whose UTR lengths were between 100-400 nt and 600-1,500 nt were more prevalent in the desMAPs (Figure 27). Also, increased mRNAs with 3’ UTR lengths over 1000 nt were less prevalent than predicted in the desMAPs of Csl4, Dis3, Rrp6, and

Rrp47 (Figures 27f-i).

When the 3’ UTR distribution of decreased transcripts was examined longer

UTRs (>750 nt) were greatly enriched in most profiles (Figure 28). Interestingly, this was especially true for the desMAPs of Csl4, Dis3, Rrp6, and Rrp47 (Figures

113

Figure 27. The 3' UTR length distribution of increased transcripts.

The 3’ UTR length distribution patterns of all increased transcripts in the desMAPs of (A) Mtr3, (B) Ski6, (C) Rrp46, (D) Rrp40, (E) Rrp4, (F) Csl4, (G)

Dis3, (H) Rrp6, and (I) Rrp47 are shown. The patterns of Rrp4 and Rrp6 from

Figure 4 are reproduced here to allow for easier comparisons among all subunits.

114

Figure 28. The 3' UTR length distribution of decreased transcripts.

The 3’ UTR length distribution patterns of all decreased transcripts in the desMAPs of (A) Mtr3, (B) Ski6, (C) Rrp46, (D) Rrp40, (E) Csl4, (F) Dis3, (G)

Rrp6, and (H) Rrp47 are shown. The pattern of Rrp4 was excluded as that desMAP contained only one reproducibly decreased transcript.

115

28e-h). Those are the same desMAPs which showed a preference against 3’

UTR lengths exceeding 1000 nt when increased transcripts were examined

(Figures 27f-i). The 3’ UTR distributions of several subunits (Mtr3, Ski6, Rrp46,

Csl4, and Rrp47) showed a very strong preference against mRNAs with 3’ UTR

lengths between 100 and 200 nt (Figures 28a-c, e, and h). mRNAs with 3’ UTRs

<100 nt were also enriched in the desMAPs of Mtr3, Rrp40, Csl4, and Dis3

(Figures 28a, d-f).

Collectively, these data show that the mRNAs identified in my microarray

study, both increased and decreased, have significantly longer UTRs than the

representative transcriptomic average (Figures 21, 23). There is no significant

difference in average 5’ UTR length between increased and decreased

transcripts (Figure 21). This trend is not reproduced when 3’ UTRs are examined

1 as decreased transcripts have UTRs that are roughly /3 longer than those of

increased transcripts (Figure 21). The UTR distributions of certain exosome subunits, especially Rrp4 and Rrp6, are also different from the distributions of the remaining exosome subunits (Figures 25-28). However, when compared to the

average transcriptomic distribution of UTR lengths, the general pattern for all

desMAPs showed a preference for longer UTRs, and mRNAs with shorter UTRs

were less prevalent in the desMAPs of most subunits (Figures 25-28). These

data can be interpreted to suggest that the UTRs of affected genes may contain

specialized regulatory elements.

2.D.7. Certain subunits regulate the stability of mitochondrial transcripts

116

Since two different proteomic screens have identified an exosome subunit,

Dis3, as a component of the mitochondrial proteome, I reasoned that Dis3, and/or other exosome subunits may be involved in the surveillance of mitochondrial transcripts (179, 200). My ability to assess this issue was limited as only nuclear encoded mitochondrial genes had probesets on the Drosophila_2 microarray. Using netaffyx, an online webtool provided by Affymetrix, I searched the affected mRNAs for gene names or descriptors which matched them to the mitochondria. In total, 209 mitochondrial genes were identified in the pooled desMAPs (Table 9). Most mitochondrial transcripts, 182 of 209 (87%), were increased with an average fold change of +3.88, while decreased transcripts were reduced an average of -3.47 fold (Figures 29, 30). Mitochondrial transcripts were present in every desMAP, but their number and identity varied. The desMAP of Rrp6 had the fewest with 6 mitochondrial transcripts, and Rrp47 having the most with 59 (Table 9, Figures 29, 30). In nearly all desMAPs, mitochondrial transcripts composed between 2% and 4% of affected transcripts

(Figure 29b). The only exception was the desMAP of Rrp47, where over 7% of the desMAP were mitochondrial transcripts (Figure 29b). As observed with the desMAPs, domain structure of the depleted subunit was not a predictor of whether mitochondrial transcripts would be found in a particular desMAP (Figure

29a). Also as noted earlier, domain structure was not tied to whether these mitochondrial transcripts were increased or decreased (Figure 30a). As three mitochondrial transcripts were present in all but one desMAP (Rrp6), and ten

117

Table 9. Mitochondrial genes affected by exosome subunit depletion.

# of desMAPs Gene Symbol Exosome Subunits 8 debcl all minus Rrp6 8 Cyp12a4 all minus Rrp6 8 CG1628 all minus Rrp6 7 Cyp12a5 Mtr3, Rrp40, Ski6, Rrp4, Rrp47, Rrp46, Dis3 7 CPTI Mtr3, Rrp40, Ski6, Csl4, Rrp47, Rrp46, Dis3 6 arg Rrp40, Rrp46, Rrp47, Csl4, Dis3, Rrp6 5 tud Mtr3, Rrp4, Ski6, Rrp40, Rrp46 5 Fmo-2 Mtr3, Rrp4, Ski6, Csl4, Rrp40 5 Dhc64C Mtr3, Rrp4, Ski6, Rrp40, Rrp46 5 dnk Mtr3, Rrp4, Ski6, Csl4, Rrp47 4 kdn Mtr3, Rrp40, Ski6, Rrp4 4 CG1746 Mtr3 Rrp47, Ski6, Dis3 4 mRpL15 Mtr3 Rrp47, Ski6, Dis3 4 CG5254 Rrp46, Rrp40, Rrp6, Dis3 3 Gs1 Mtr3, Rrp47, Rrp4 3 Gdh Mtr3, Ski6, Rrp4 3 Ptp61F Mtr3, Rrp47, Rrp4 3 Mdh Mtr3, Ski6, Rrp4 3 mRpS14 Mtr3, Ski6, Rrp47 3 mRpS10 Mtr3, Ski6, Dis3 3 mRpL54 Mtr3, Ski6, Rrp47 3 mRpS18B Mtr3, Ski6, Rrp47 3 mRpL12 Mtr3, Ski6, Rrp47 3 sesB Mtr3, Ski6, Rrp47 3 CG4598 Mtr3, Ski6, Rrp47 3 CG2249 Mtr3, Ski6, Rrp47 3 mRpS30 Mtr3, Ski6, Rrp47 3 Nmdmc Rrp46, Rrp6, Rrp40 3 CG1673 Csl4, Rrp6, Rrp47 3 Pepck Rrp46, Rrp6, Dis3 2 Cyp12c1 Mtr3, Ski6 2 Gpo-1 Rrp4, Rrp47 2 CG1041 Mtr3, Ski6 2 CG6782 Csl4, Rrp47 2 CG15093 Mtr3, Csl4 2 mRpL39 Dis3, Rrp47 2 CG14642 Mtr3, Ski6 2 mRpL36 Dis3, Rrp47 2 mRpS28 Csl4, Rrp47 2 CG12262 Mtr3, Rrp47 2 CG1516 Mtr3, Rrp47 2 CG4389 Ski6, Rrp47 2 Ptp61F Mtr3, Rrp47 2 Dhap-at Rrp40, Csl4 2 CG17510 Rrp40, Csl4 2 CG5646 Rrp46, Rrp40

118

# of desMAPs Gene Symbol Subunit # of desMAPs Gene Symbol Subunit 1 CG4589 Rrp4 1 l(2)35Di Rrp47 1 Dref Rrp4 1 mRpL42 Rrp47 1 CG4644 Rrp4 1 CG3011 Rrp47 1 CG5508 Rrp4 1 mRpL49 Rrp47 1 CG18324 Mtr3 1 mRpL4 Rrp47 1 mRpS11 Rrp4 1 Scs-fp Rrp47 1 mRpL2 Rrp47 1 scu Rrp47 1 Acon Ski6 1 CG11779 Rrp47 1 rpr Rrp47 1 mge Dis3 1 bonsai Ski6 1 mRpS31 Rrp47 1 CG3321 Ski6 1 Sod2 Rrp47 1 CG7181 Ski6 1 mRpS29 Rrp47 1 CG9603 Ski6 1 CG6020 Rrp47 1 tam Ski6 1 CG7943 Rrp47 1 l(1)G0255 Rrp47 1 CG13163 Rrp47 1 ECSIT Rrp47 1 Pdsw Rrp47 1 mRpS26 Rrp47 1 mRpL51 Rrp47 1 mRpL3 Ski6 1 mRpL27 Rrp47 1 CG9804 Rrp47 1 cype Rrp47 1 CG1140 Rrp47 1 mRpS9 Rrp47 1 CG8862 Rrp47 1 CG30493 Rrp40 1 CG9547 Rrp47 1 CG5059 Dis3 1 l(2)35Di Rrp47 1 Got1 Dis3 1 mRpL16 Rrp47 1 CG1516 Rrp6 1 blw Ski6 1 CG6661 Rrp40 1 yip2 Mtr3

119

Figure 29. Mitochondrial transcripts are enriched in certain desMAPs.

The (A) number and (B) percentage of nuclear encoded mitochondrial transcripts present in each desMAP.

120

Figure 30. Most michondrial genes are not conserved between subunits.

(A) Number of increased (red) and decreased (blue) mitochondrial transcripts per desMAP. (B) Conservation of mitochondrial transcripts between all desMAPs.

121

Table 10. The per desMAP distribution of mRP transcripts.

(I) # of Gene or Avg. Min. Max. desMAPs Symbol (D) FC FC FC Subunits 4 mRpL15 I 2.72 2.47 2.97 Mtr3, Ski6, Dis3, Rrp47 3 mRpS14 I 3.02 2.83 3.41 Mtr3, Ski6, Rrp47 3 mRpS10 I 2.85 2.25 3.38 Mtr3, Ski6, Dis3 3 mRpL54 I 2.84 2.30 3.43 Mtr3, Ski6, Rrp47 3 mRpS18B I 2.76 2.51 3.04 Mtr3, Ski6, Rrp47 3 mRpL12 I 2.65 2.30 3.04 Mtr3, Ski6, Rrp47 3 mRpS30 I 2.42 2.23 2.61 Mtr3, Ski6, Rrp47 2 mRpL39 I 2.78 2.59 2.97 Dis3, Rrp47 2 mRpL36 I 2.61 2.31 2.91 Dis3, Rrp47 2 mRpS28 I 2.55 2.28 2.81 Csl4, Rrp47 1 mRpS11 I 3.85 - - Rrp4 1 mRpL2 I 3.04 - - Rrp47

1 mRpS26 I 2.67 - - Rrp47

1 mRpL3 I 2.66 - - Ski6

1 mRpL16 I 2.58 - - Rrp47

1 mRpL42 I 2.49 - - Rrp47

1 mRpL49 I 2.49 - - Rrp47

1 mRpL4 I 2.47 - - Rrp47

1 mRpS31 I 2.36 - - Rrp47

1 mRpS29 I 2.32 - - Rrp47

1 mRpL51 I 2.18 - - Rrp47

1 mRpL27 I 2.10 - - Rrp47

1 mRpS9 I 2 - - Rrp47

122

Table 11. Summary of mitochondrial transcripts in each desMAP.

Total % of desMAP % of mito % of desMAP mito- = Total Small Large transcripts desMAP = Subunit size chondrial mitochondrial mRp Subunit Subunit = mRp mRp Mtr3 1186 34 2.87% 7 4 3 20.59% 0.62% Ski6 1172 36 3.07% 7 3 4 19.44% 0.63% Rrp46 432 12 2.78% 0 0 0 0 0 Rrp40 437 17 3.89% 0 0 0 0 0 Rrp4 886 19 2.14% 1 1 0 5.26% 0.11% Csl4 472 11 2.33% 1 1 0 10.00% 0.31% Dis3 470 15 3.13% 4 1 3 40.00% 1.84% Rrp6 229 6 2.62% 0 0 0 0 0 Rrp47 835 59 7.07% 20 8 12 35.09% 2.76%

123 transcripts were found in at least 5 desMAPs, several mitochondrial transcripts were affected by the depletion of multiple subunits (Figure 30b and Table 9).

Although a large proportion of Rrp47’s mitochondrial transcripts were only detected in that single desMAP, many transcripts were shared with other desMAPs (Table 9). Oddly, Rrp6 and Rrp47 show almost opposite desMAP profiles when these mitochondrial transcripts are considered (Figures 29a and

30a, plus Tables 9-11). All previous reports for Rrp6 and Rrp47 have shown similar effects on distinct substrates suggesting that these proteins are functionally linked (4-6, 93).

The opposite trend is observed with the mitochondrial transcripts surveyed by

Mtr3, Ski6, and Rrp47 (Table 9). In particular, these three subunits appear to co- survey a distinct class of nuclear encoded mitochondrial transcripts (Table 10).

The mitochondrial ribosomal protein (mRP) mRNAs were strikingly enriched and composed up to 40% of the mitochondrial transcripts of some profiles (Tables 10 and 11). These mRNAs, all of which are increased, encode protein components of the mitochondrial large or small ribosomal subunits. Unlike some other mitochondrial transcripts, these mRP mRNAs were restricted to a subset of desMAPs, and were not widely conserved (Table 10). Further, as with total mitochondrial transcripts, the desMAPs of Rrp6 and Rrp47 differed greatly with respect to mRP transcripts (Tables 10 and 11). While Rrp47 was implicated in the surveillance of 20 out of the 23 mRP genes identified in this study, mRP genes were completely absent from the desMAP of Rrp6 (Tables 10 and 11).

124

2.D.8. NMD targeted transcripts are enriched in certain desMAPs

My previous searches suggest that elements outside the coding sequence

(longer UTRs) may play a role in exosome-mediated mRNA surveillance.

Elements within the coding sequence are also known to serve this function. In this regard, I searched the microarray data to see if mRNAs with premature termination codons (i.e., NMD targets), whose turnover has been attributed to the exosome complex, were overrepresented in our data set (78, 185). I identified

74 (40%) of 185 mRNAs that are known Drosophila NMD targets in our pooled microarray data set (Table S7) (185). These NMD targets were then mapped to individual desMAPs (Figure 31). The number of NMD targeted transcripts in each desMAP did not directly correlate with the size of the desMAP (Figure 31a).

For example, Rrp6 had the smallest desMAP, by total number transcripts (Figure

12). However, the desMAP of Rrp6 was among the largest desMAPs containing the highest numbers of NMD surveyed transcripts (Figure 31a). Depletion of

Mtr3, Ski6, Rrp4, and Rrp6 stabilized the largest numbers of NMD surveyed transcripts, while depletion of Dis3, Csl4, and Rrp47 stabilized the fewest (Figure

31a). While the number of NMD targeted transcripts varied by desMAP, in general, NMD targets constituted between 1.5% and 6.3% of affected transcripts in most desMAPs (Figure 31a and 31b). The exception was the desMAP of

Rrp6, where 23% of the transcripts were NMD targets (Figure 31b).

Many of these transcripts were conserved among most or all desMAPs

(Table 12). As with the mitochondrial transcripts, depleting a subset of exosome subunits appear to co-regulate the levels of several NMD targeted mRNAs (Table

125

Figure 31. NMD targeted transcripts are enriched in certain desMAPs.

(A) The numbers of known NMD targeted transcripts in each desMAP. (B)

Prevalence of NMD transcripts, as percent of altered transcripts, in each desMAP.

126

Table 12. Some NMD-surveyed transcripts are conserved among desMAPs.

- Substitutes for minus in this table.

Gene Gene Symbol Count Found in desMAPs Symbol Count Found in desMAP(s) CG32477 9 All RhoGAP68F 4 Mtr3, Ski6, Rrp4, Rrp6 spir 9 All RhoGEF2 4 Mtr3, Ski6, Rrp4, Rrp6 CG31999 9 All Thd1 4 Mtr3, Ski6, Rrp4, Rrp6 CG10359 8 All - Dis3 kst 4 Mtr3, Ski6, Rrp4, Rrp6 CG10462 8 All - Dis3 tra 4 Mtr3, Ski6, Rrp4, Rrp6 Paip2 7 All - Dis3 and Csl4 Ugt86Da 4 Mtr3, Csl4, Rrp6, Rrp47 fax 7 All - Rrp46 and Rrp40 CG3262 3 Mtr3, Ski6, Rrp6 CG8771 7 All - Rrp46 and Rrp40 CG3532 3 Mtr3, Rrp4, Rrp6 CG3348 7 All - Rrp4 and Rrp47 cact 3 Mtr3, Ski6, Rrp6 Nup154 7 All - Dis3 and Rrp47 norpA 3 Mtr3, Ski6, Rrp6 fat2 6 All - Csl4, Dis3 and Rrp47 zip 3 Mtr3, Ski6, Rrp6 CG8552 6 All - Csl4, Dis3 and Rrp47 CG18596 3 Ski6, Rrp46, Rrp4 CG2093 6 All - Csl4, Dis3 and Rrp47 chico 3 Mtr3, Rrp4, Rrp6 trx 6 All - Csl4, Dis3 and Rrp47 msps 3 Mtr3, Rrp4, Rrp6 CG7843 6 All - Csl4, Dis3 and Rrp47 Klp31E 3 Mtr3, Rrp4, Rrp6 Dhc64C 6 All - Csl4, Dis3 and Rrp47 mus210 2 Ski6, Rrp4 poe 6 All - Csl4, Dis3 and Rrp47 pan 2 Mtr3, Rrp6 CG2926 6 All - Csl4, Dis3 and Rrp47 Mgat1 2 Mtr3, Rrp6 Mtr3, Ski6, Rrp4, Csl4, CG1146 5 Rrp6, Rrp47 CG30394 2 Mtr3, Rrp6 Mtr3, Ski6, Rrp40, Rrp4, CG12582 5 Rrp6 sda 2 Csl4, Rrp47 Rrp46, Rrp40, Csl4, Dis3, Mmp1 5 Rrp47 Ide 1 Rrp4 Mtr3, Ski6, Rrp46, Rrp4, TepIV 5 Rrp6 Fas1 1 Dis3 CG8793 4 Mtr3, Ski6, Rrp4, Rrp6 Aef1 1 Rrp4 brm 4 Mtr3, Ski6, Rrp4, Rrp6 CG10535 1 Rrp4 emb 4 Mtr3, Ski6, Rrp4, Rrp6 CG11970 1 Ski6 Dcr-2 4 Mtr3, Ski6, Rrp4, Rrp6 CG12177 1 Dis3 Ca-P60A 4 Mtr3, Ski6, Rrp4, Rrp6 CG12190 1 Rrp4 CG32176 4 Mtr3, Ski6, Rrp4, Rrp6 nct 1 Rrp4 CG4841 4 Mtr3, Ski6, Rrp4, Rrp6 CG9663 1 Rrp47 Hel89B 4 Mtr3, Ski6, Rrp4, Rrp6 nbs 1 Rrp4 CG31367 4 Mtr3, Ski6, Rrp4, Rrp6 CG9636 1 Rrp4 CG8092 4 Mtr3, Ski6, Rrp4, Rrp6 POSH 1 Rrp4 l(2)NC136 4 Mtr3, Ski6, Rrp4, Rrp6 CG32016 1 Rrp4 hyd 4 Mtr3, Ski6, Rrp4, Rrp6 MICAL-like 1 Rrp4 eIF5 4 Mtr3, Ski6, Rrp4, Rrp6 Moca-cyp 1 Rrp4 PRL-1 4 Mtr3, Ski6, Rrp4, Rrp6 CG3448 1 Rrp46 Ptp69D 4 Mtr3, Ski6, Rrp4, Rrp6 Cyp4e2 1 Dis3

127

12). This is readily apparent by the transcripts which are shared between the

desMAPs of Mtr3, Ski6, Rrp4, and Rrp6. As with the mitochondrial transcripts

above, the profiles of Rrp6 and Rrp47 differed strikingly when NMD transcripts

were considered (Figure 31 and Table 12). The desMAPs of these two proteins yielded nearly opposite results, with Rrp6 having the highest percentage of NMD targeted transcripts while Rrp47 had the smallest (Figure 31b). However, the large proportion of NMD-targeted mRNAs in the desMAP of Rrp6 could suggest

a core exosome independent role for Rrp6 in the surveillance of NMD-targeted transcripts (Figure 31b).

2.D.9. Genes increased by most exosome subunit depletions may be functionally linked

I wanted to determine if the proteins encoded by the mRNAs affected by exosome subunit depletion had roles in similar biological pathways. Therefore, I grouped the genes that were affected by each exosome subunit depletion based upon the functions and interactions of the proteins they encoded. Increased and decreased genes were grouped and analyzed separately. A P-value was assigned to the detected pathways depending upon the number of components within the individual pathway and their prevalence in our desMAPs. Specific and statistically significant pathways (P-value ≤ 0.05) were retained for further analysis; however, significant, but overly general pathways (e.g. nucleus, cytoplasm, and molecular function) were culled from further analysis. Related pathways were grouped into larger categories (e.g. mitosis and spindle were

128

grouped into cell cycle). Certain pathways were present in multiple groups (e.g.

protein ubiquitination was both protein modification and turnover), while other

pathways did not fit into any group (e.g. male courtship behavior). The prevalence of a group of pathways was calculated by determining the percentage

of total pathways that resided in each sub-group for that desMAP. These groupings were done for both increased and decreased transcripts. The results for increased transcripts are listed in (Table 13), and graphed in Figure 32.

When these groupings were plotted for the increased transcripts, the patterns of Rrp6 and Dis3 were different from the other exosome subunits (cf. Figure 32a

and 32b versus 32c- 32i) and the representative average of all desMAPs (Figure

32j). Cell cycle-related pathways were the most prominent in the desMAPs of

Dis3 and Rrp6, but the prevalence of other pathway groups were not shared

between the desMAPs of these two subunits (Figure 32a and 32b). When

compared to the whole study average (Figure 32j), many groups of transcripts

were underrepresented in Dis3 depleted cells (Figure 32a). Also, some common

groups such as transcription related, transport related, and signalling were

missing entirely from the desMAP of Rrp6 (c.f. Figure 32b and 32j). For

increased transcripts, the group patterns of the remaining exosome subunits

were very similar to one another (Figure 32c-i). For five of those subunits (Mtr3,

Ski6, Rrp46, Rrp40, and Rrp4) development- and cell cycle-related transcripts

were generally the most prominent classes; signaling-, transcription-, transport-,

and protein modification-related were also enriched to a similar extent (Figure

32c-g). These similar patterns were surprising since the desMAPs of these

129

Table 13. Gene groups affected (per desMAP) for increased transcripts.

Gene group Mtr3 Ski6 Rrp46 Rrp40 Rrp4 Csl4 Dis3 Rrp6 Rrp47 Amino Acid metab 0.64 0.35 0.23 0.72 1.66 0.26 1.44 0 2.28 carb related 1.80 1.28 0.69 1.44 1.42 0.78 0 0 0.98 Cell Cycle 9.85 8.85 7.57 7.66 9.94 15.32 17.69 12.84 14.31 cytoskeleton 4.87 4.77 4.59 4.31 5.44 3.64 1.08 1.35 3.90 Defense response 1.91 2.68 2.75 3.59 1.89 5.19 3.25 6.08 2.93 Development 12.18 14.55 16.28 16.99 16.21 7.01 2.17 2.70 10.57 Lipid metab 0.74 1.51 0.92 1.20 1.30 1.82 1.08 2.70 2.60 metabolism 1.48 1.75 0.92 1.67 2.49 1.30 1.08 0 2.60 Metal ion 1.38 1.75 1.83 2.15 1.30 1.56 2.53 5.41 0.81 Mitochondrial 1.80 1.98 2.06 2.15 1.18 1.04 2.17 0 1.63 Nucleic acid metab 3.28 4.77 1.83 0.72 5.09 1.04 2.89 8.11 0.33 nucleotide metab 2.97 3.84 4.59 4.55 3.43 6.49 3.97 0 4.88 Protein Modification 6.57 7.45 5.50 7.42 6.86 5.19 4.33 5.41 7.15 Protein turnover 2.75 2.91 4.36 4.55 3.43 2.34 0.72 3.38 3.74 Radical chem 0.95 1.86 1.15 0.24 1.89 2.60 0.36 0 0.65 Ribosomal 0.42 0.58 0 0 0.12 0 2.89 0 0.33 RNA metabolism 6.04 6.87 4.82 5.26 5.68 3.12 6.14 8.78 4.07 signaling 8.47 8.38 9.40 5.98 10.89 7.53 5.05 0 5.37 transcription related 6.04 7.22 10.09 11.24 6.98 5.45 3.25 0 5.85 transport 6.36 6.98 7.80 9.33 6.39 8.05 5.05 0 5.69

130

Figure 32. Many mRNAs increased by exosome subunit depletion code for proteins working in similar pathways.

The cellular processes affected by the transcripts increased in each desMAP were identified by using Pathway Studio 5.0 (Ariadne). Pathways were grouped

by similarity and patterns are shown as percentage of each group as part of the

whole profile. The patterns of (A) Dis3, (B) Rrp6, (C) Mtr3, (D) Ski6, (E) Rrp46,

(F) Rrp40, (G) Rrp4, (H) Csl4, (I) Rrp47, and the (J) average pattern for all

pathway groupings of increased transcripts are shown.

131 individual subunits were quite different both by number of transcripts (Table 5,

Figure 12) and by content (Figures 12 and 13, Table 7). Ribosomal and/or ribosome associated transcripts were also underrepresented in all desMAPs.

This absence is explained by their almost complete exclusion from the

Drosophila_2 microarray.

Grouping the increased transcripts yielded two general patterns, Dis3 and

Rrp6 on one hand, and the remaining exosome subunits on the other, of cell processes that were affected by exosome subunit depletion (Figure 32, Table

13). Performing a parallel analysis using the same procedures to group the decreased transcripts did not yield a similar result (Table 14, Figure 33). No consensus pattern is visible to the groups (Figure 33). As with the UTR analysis, the desMAP of Rrp4 was excluded from this analysis since it only contained a single decreased transcript (Figure 12). From these data, I surmise that although individual exosome subunits process and degrade distinct mRNAs, those mRNAs perform functionally interdependent roles in cellular pathways.

2.E. Discussion

In this study, I sought to identify and characterize the mRNAs that were surveyed by the subunits that comprise and associate with the exosome complex in vivo. The core exosome model predicts that the transcript profiles of all exosome subunits should be very similar; while the exozyme model predicts that the desMAPs would be distinct. Consistent with the predictions of the exozyme model, both here and in other studies, different transcriptomic profiles are

132

Table 14. Grouped analysis of decreased transcripts.

Gene group Mtr3 Ski6 Rrp46 Rrp40 Rrp4 Csl4 Dis3 Rrp6 Rrp47 Amino Acid metab 0 0 3.39 6.94 0 4.28 4.55 11.69 2.86 carb related 5.13 7.22 6.78 4.08 0 5.06 7.27 7.79 8.57 Cell Cycle 0 1.03 4.66 4.90 0 3.11 2.12 0 0.57 cytoskeleton 2.56 4.12 5.08 4.08 0 3.11 3.33 1.95 2.86 Defense response 0 2.06 2.12 1.63 0 1.56 5.76 1.95 2.29 Development 10.26 25.77 11.44 9.80 100.00 12.06 8.18 0 19.43 Lipid metab 0 3.09 2.97 4.49 0 3.50 1.52 0 2.29 metabolism 10.26 7.22 8.05 0 0 2.33 3.64 0 5.71 Metal ion 2.56 0 2.12 6.53 0 4.28 2.42 1.30 5.14 Mitochondrial 5.13 2.06 1.69 1.63 0 1.17 0.61 0 1.14 Nucleic acid metab 0 0 1.27 0.82 0 0.39 0.61 3.25 1.14 nucleotide metab 0 1.03 5.08 4.90 0 4.28 3.03 0 0 Protein Modification 0 5.15 5.51 4.08 0 9.34 4.85 4.55 5.14 Protein turnover 2.56 4.12 2.12 2.86 0 1.95 1.52 0.65 0.57 Radical chem 5.13 4.12 3.39 2.45 0 1.17 0 0 1.71 Ribosomal 0 0 0 0 0 0 0 0 0 RNA metabolism 0 1.03 2.54 1.22 0 0.78 2.12 1.30 1.14 signaling 17.95 19.59 14.41 12.65 0 19.46 8.18 0 14.29 transcription related 2.56 3.09 0.42 0.41 0 0.78 0.61 0 1.71 transport 17.95 11.34 11.44 10.61 0 5.84 5.15 0 7.43

133

Figure 33. mRNAs decreased by exosome subunit depletion have few epistatic interactions.

The cellular processes affected by the transcripts decreased in each desMAP were identified by using Pathway Studio 5.0 (Ariadne). Pathways were grouped by similarity and patterns are shown as percentage of each group as part of the whole profile. The patterns of (A) Dis3, (B) Rrp6, (C) Mtr3, (D) Ski6, (E) Rrp46,

(F) Rrp40, (G) Csl4, (H) Rrp47, and the (I) average pattern for all pathway groupings of decreased transcripts are shown. The group profile of Rrp4 was excluded as that desMAP contained only one reproducibly decreased transcript.

134 observed when exosome subunits are mutated or depleted (39, 100). My microarray data show that depletion of individual exosome subunits stabilizes distinct sets of mRNAs which vary greatly in both the size of the transcript pool and the identity of the transcripts affected. Validating their specificity, my microarrays identified 74 NMD targeted transcripts (P-value < 0.0001), which are known exosome subunit substrates in D. melanogaster (185). In general, the transcripts affected by exosome subunit depletion have larger than average 5’ and 3’ UTRs, and encoded proteins that were involved in similar biological pathways.

In contrast to work in yeast showing that exosome subunit genes are essential, our data show that depletion of most exosome subunits does not affect cell proliferation in Drosophila S2 cells (87). Only Rrp6 and Dis3 appear to be essential for cell proliferation in S2 Cells. Further, their roles in cell proliferation are likely to be both cell cycle related, and likely independent of other exosome subunits (87, 122, 203). Our data agree with earlier observations showing that depletion of one exosome subunit co-depletes one or more additional subunits

(65, 215). These partial (between 25 and 40% of the co-depleted subunit remains as assayed by western blotting) co-depletions make interpretation of the resulting data more complicated. However, closer examination of the desMAPs alleviates some of these concerns. The desMAPs of several co-depleted pairs of subunits, such as Mtr3 and Csl4 (where Mtr3 depletion elicits Csl4 co-depletion) and Rrp46 and Rrp40 (where Rrp46 depletion elicits Rrp40 co-depletion), are still substantially different. This suggests that the co-depleted subunits may not be

135

reduced below the threshold necessary to produce effects on mRNA turnover and processing specific to the co-depleted subunit. It is also possible that the

pools of exosome subunits remaining after depletion are sufficient to perform essential functions and thus allow the cells to proliferate. Nonetheless, the dissimilarity of desMAPs and the rigorous steps taken to exclude false positive transcripts coupled with the reproducibility of the data, all suggest that our findings are of sufficient quality to determine the roles of subunits in the turnover of distinct mRNAs, or classes of mRNAs. This stance is supported by the observation that dsRNA length does not directly correlate with desMAP size or complexity, suggesting that off-target effects of the dsRNAs are responsible for a minority of the transcripts in my desMAPs.

Previously published microarray studies show heterogeneity and little transcriptomic overlap when individual exosome subunits are mutated or depleted (39, 100). My microarray studies in Drosophila confirm and extend previous works in yeast and plants (39, 100). All microarray analyses of exosome subunit mutants or depletions show that individual exosome subunits survey different pools of RNAs (39, 100, 178). I find the similarities among the

available microarray studies noteworthy. First, consistent with this study, the

array profile of rrp6∆ yeast cells was the most divergent when compared to the

profiles of other subunits (100). Second, a small fraction of affected transcripts

were altered in the same manner in three yeast strains, and furthermore, many

transcripts were found in the profiles of only one or two of the three subunits

tested (100). Third, tiling microarray results from plants show certain mRNAs

136 were affected in a different manner depending on which subunit was depleted

(39). Finally, the overall number and location of gene regions affected by exosome subunit RNAi depletion in plants was dependent on which subunit was depleted (39). Together, these microarray studies from three organisms (S. cerevisiae, A. thaliana, and D. melanogaster) and genetic studies in yeast demonstrate that depleting or mutating individual subunits has different effects on the processing and/or turnover of individual RNAs.

In addition to similar effects at the transcriptomic level, my data are likewise consistent with previously observed effects for mRNA classes known to be exosome subunit substrates. For example, NMD targets are differentially represented in the profiles of exosome subunit depleted cells, especially that of

Rrp6. In agreement with this result, Rrp6 has been shown to be involved in the turnover of NMD targeted transcripts in human cells, and is both nuclear and cytoplasmic in both human and S2 cells (86, 88, 133). Rrp6 has recently been shown to trim rRNA precursors while not associating with other exosome subunits in yeast (33). These results provide further evidence that Rrp6 may function in a core-exosome independent manner.

If regulatory elements in the UTRs of mRNAs either target them directly to exosome subunits as Wilusz and coworkers have observed, or serve as landing pads for cellular factors which then recruit exosome subunits as Karin, Chen and coworkers showed, then one would predict that mRNAs with UTR regulatory elements would be exosome substrates (9, 41, 226). In agreement with this prediction, ARE containing mRNAs and the phosphoglycerate kinase mRNA of

137

trypanosomes, are known exosome substrates (18, 41, 44, 45, 198, 199, 226).

These longer UTRs also infer the presence of regulatory elements as the genes are under under constant selective pressure. Further, as longer UTRs have more bases, and therefore a greater likelihood of containing regulatory elements, they should be more prevalent in our desMAPs. This is precisely what the data show as the mRNAs in my desMAPs had longer than average 5’ and 3’ UTRs.

These putative UTR regulatory elements could be sequences that either act directly, via exosome subunit-RNA interactions, or indirectly, via RNA binding

protein-exosome subunit interactions, which target transcripts to certain exosome

subunits or to the exosome complex as a whole. Further, similar RNA sequence

elements can be present in other types of RNA, which may explain how the

exosome complex is recruited to and processes or degrades other classes of

RNAs. A detailed, bioinformatic analysis of the UTR sequences of mRNAs

identified by our microarrays could elucidate the identity of one or more of the

putative regulatory elements.

To date, the data from multiple microarray studies (including this one) and

associated experiments show that distinct RNAs are affected by exosome

subunit depletion or mutation (39, 100, 178). These data suggest that individual

exosome subunits have roles in RNA processing that are either independent of

the exosome complex, or that they may serve specific roles within the complex.

These results are at odds with genetic data suggesting that all exosome subunits

are required for complex integrity and the processing and/or turnover of RNAs,

including mRNAs (4-6, 65, 153). My grouped pathway data could help resolve

138

this paradox. While the individual transcripts affected by exosome subunit

depletion in the desMAPs are quite different, analysis of the biological functions of their encoded proteins shows that most of the transcripts are functionally

related. These data suggest epistatic interactions between and among different

exosome subunits and the mRNAs they survey. These epistatic interactions

explain how the surveillance of distinct RNAs by different subunits yields very

similar outcomes at the genetic and cellular level.

This transcriptomic study represents an important advance in our

understanding of the gene-specific and genome-wide roles of individual exosome

subunits in vivo. Moreover, I identify cellular pathways that are regulated by

distinct exosome subunits and provide compelling evidence for both overlapping

and independent functions for Rrp6, Dis3, Rrp47, and the core exosome

subunits. I also identify a subset of exosome subunits (Mtr3, Ski6, Rrp6, and

Rrp4) that may have specialized roles in the surveillance of nuclear-encoded

mitochondrial targeted mRNAs. These data show that individual, or subsets of,

exosome subunits have specialized mRNA surveillance functions are

inconsistent with the predictions of the core exosome model. However they

precisely match the predictions of the exozyme model. Future gene- or pathway-

directed analysis will yield additional insights into the structure and function of the

exosome core and its cofactors in living cells.

2.G. Acknowledgements

139

The authors would like to thank members of the Andrulis lab for discussions and

comments on the manuscript. The microarray analysis was aided by the Gene expression and Genotyping Facility of the Case Comprehensive Cancer Center, especially Patrick Leahy and Chunbiao Li. This work is supported by grant

GM072820 from the NIH (E.D.A). E.D.A. is a Mount Sinai Health Care

Foundation Scholar.

140

Chapter 3:

Levels of Heat Shock Protein 70 and 26 mRNAs upon Recovery from Heat

Shock are Differentially Affected by Exosome Subunit Depletion

Daniel L. Kiss and Erik D. Andrulis

Department of Molecular Biology and Microbiology, Case Western Reserve

University School of Medicine, Cleveland, Ohio

141

3.A. Abstract

The core exosome complex was originally defined as a multi-subunit complex of ribonucleases responsible for the processing and/or turnover of stable RNAs and mRNAs. The prevailing model of exosome complex function, the core exosome model, relies heavily on in vitro reconstructions of the complex and is challenged by data from several in vivo studies of exosome subunit function. I have proposed and begun to evaluate the exozyme model, an alternative model of exosome subunit function which posits that exosome subunits function and achieve substrate specificity by assembling into distinct subcomplexes. I used

RNAi to deplete most exosome subunits and an exosome cofactor in Drosophila melanogaster S2 tissue culture cells and assayed the effects on heat shock mRNA levels upon recovery from heat stress using S1 nuclease protection assays. Consistent with the microarray results presented above, and the predictions of the exozyme model, I find distinct effects on hsp70 mRNA levels when different subunits are depleted. Depletion of certain subunits (Mtr3, Ski6, and Csl4) resulted in a significant increase in the levels of hsp70 mRNA, while the depletion of other subunits had little (Rrp46, Rrp6) or no (Dis3) effect.

Depletion of Rrp40 or Rrp4 elicited an unexpected result as the accumulation of hsp70 mRNA was slowed, but continuous after the removal of heat stress. For most exosome subunit depletions, I observed similar results when I expanded these studies to include a second heat shock transcript, hsp26. The depletions of Rrp46 which significantly stabilized hsp26, but not hsp70 mRNA, plus both

Ski6 and Csl4 which stabilized hsp70 mRNA more than hsp26 mRNA were the exceptions. This study provides compelling evidence for multiple, independent

142

functions for individual exosome subunits and contradicts the predictions of the core exosome model of exosome subunit function. In agreement with the exozyme model, this study intimates a greater complexity of exosome subunit assemblages than heretofore anticipated.

143

3.B. Introduction

The core exosome paradigm posits that a single stoichiometric exosome

complex is responsible for all ascribed RNA metabolic functions. In this model,

all functions and contributions of individual exosome subunits to distinct RNA

processing and turnover events occur only in the context of a stoichiometric

complex. Further, the core exosome model predicts that knocking out one

polypeptide within the core is sufficient to disrupt the integrity of the complex and

thereby disrupt core exosome-mediated RNA metabolic functions. Both the

yeast and human exosome complexes are composed of a “core” consisting of

three S1 domain subunits (Csl4, Rrp4, and Rrp40) and six RNase PH domain

subunits (Ski6/Rrp41, Rrp42, Rrp43, Rrp45, Rrp46, and Mtr3). This nine subunit

core of RNase PH and S1 domain subunits has been proposed to serve as a

scaffold for two additional polypeptides, Dis3 and Rrp6, RNase II/R and RNase D

homologs respectively. Although it was initially shown that multiple subunits

within the complex were catalytically active, recent studies have argued that this

activity is predominantly, if not exclusively, found in Dis3 and Rrp6 (61, 137, 153,

219).

The exosome core paradigm has been constructed based on our

understanding how exosome subunits function in vitro; however, detailed

evidence from several studies challenges this paradigm in vivo. First, exosome

subunits were shown to have non-overlapping localization patterns in both yeast and Drosophila S2 cells (88, 105). Second, proteomic screens in yeast have

shown that exosome subunits interact with proteins with no known role in

144

exosome-mediated processes (80, 125). Third, Rrp6 has been shown to have

RNA processing and cell cycle functions that are independent of other exosome

subunits in yeast and Drosophila (33, 87). Fourth, two exosome subunits, dRrp6 and dDis3, are known to associate in complexes with importin-α3 that are independent of other exosome subunits (86). Fifth, studies have shown that distinct complexes lacking certain core subunits are possible in solution (206).

Finally, several other studies involving the cytology of individual exosome subunits, bioinformatic comparative analysis, yeast genetic studies, and viability studies also yield results that challenge the exosome core paradigm (4, 33, 39,

88, 105).

Findings from transcriptomic studies likewise challenge the exosome core model’s prediction that there would be a large degree of transcriptomic overlap when individual exosome subunits are depleted. This prediction is based on the

similar transcriptomic profiles observed when different NMD, mRNA export, or

spliceosome component proteins were depleted (92, 177, 185). Microarray

studies in plants and yeast show that depletion or mutation of individual exosome

subunits affects distinct sets of RNAs (39, 100). In yeast, the sets of transcripts

affected in an rrp6∆ strain were quite different from those in either a temperature

sensitive ski6-100 or an rrp47∆ strain (100). Further, whole genome tiling arrays

from Arabidopsis showed that rRNA, mRNA, and ncRNAs were affected in a

different manner when three exosome subunits (Ski6, Rrp4, and Csl4) were

depleted or deleted individually (39). Finally, RT-qPCR experiments used to

validate tiling array work from human tissue culture cells indicate that depletion of

145 different exosome subunits (Rrp40, Rrp46, Dis3, and Rrp6) has distinct effects on PROMPT levels (178). These results demonstrate that individual exosome subunits contribute differently to distinct RNA processing pathways in vivo and are not consistent with the exosome core model.

The conflicts posed by these and other reported data have led us to propose the exozyme hypothesis. In the exozyme hypothesis. exosome subunits assemble into and function as multiple independent complexes, termed exozymes, in vivo. Distinct exozymes can either assemble independently of the

9 subunit core complex and/or be in a dynamic relationship with the core exosome complex. Functional exozymes may be comprised of only exosome subunits or a combination of exosome subunits, cofactors, and non-exosome proteins. Individual exozymes may function independently of other exozymes and the exosome core via different direct (exozyme-RNA) or indirect (exozyme –

RNA-BP-RNA) substrate specificities. Alternatively, they may work cooperatively on certain RNA substrates or distinct steps of RNA processing.

In order to test the exozyme hypothesis directly, I have already carried out a nearly comprehensive microarray study by using RNAi to deplete exosome subunits in Drosophila S2 cells. Consistent with the exozyme hypothesis, the data yielded distinct transcript profiles when different subunits were depleted. To extend these findings, I used S1 nuclease protection assays targeting heat shock mRNAs upon recovery from heat shock in exosome subunit depleted cells to expand upon my microarray results. By focusing on a single transcript I sought to verify that the transcriptome-wide results observed via microarrays (Chapter 2)

146

could be reproduced in vivo with a known exosome substrate. Analysis of the

contributions of exosome subunits to the turnover of the ARE-containing mRNAs, hsp70 and hsp26, also demonstrated that knock-down of different exosome subunits had distinct effects on these transcripts. Taken together, these studies indicate that individual exosome subunits contribute in a non-overlapping manner to distinct mRNA turnover pathways. These data are consistent with either the exozyme hypothesis or, alternatively, a model for exosome subunit substrate specificity within a single core exosome complex.

3.C. Methods and materials

3.C.1. RNAi depletion coupled with heat shock & RNA harvesting

Cells were treated with dsRNAs as above (see section 2.C.2). On day 5 of dsRNA treatment, culture volumes were set to 4 ml with the addition of fresh

27°C media. A portion (0.5 ml) of the RNAi depleted cells was pelleted and lysed in either 1X SDS dye for western blotting to verify effective depletion, or TRIzol

Reagent (Invitrogen) according to the manufacturer’s instructions to serve as the pre-heat shock baseline (pre). The remaining cells were subjected to an instantaneous 37°C heat shock with the addition of and equal volume (3 ml) of

47°C media and incubated at 37°C for 30 minutes. The cells were returned to

27°C (T0) and 1 ml aliquots were removed, pelleted, and RNA was harvested at the indicated time points.

147

3.C.2. Probe preparation for S1 nuclease protection assays

The sequences of oligos used for S1 probes are listed in Table 15. Oligos

were ordered from Operon, resuspended in water and gel purified as follows.

500 µM of oligonucleotide were resolved using 12.5% denaturing poly-acrylamide

gels. The most intense bands were cut from the gel, gel slices were crushed,

and eluted by rotating overnight at 4°C using 600 µl of elution buffer (300 mM

NaOAC, 1 mM EDTA pH 8.0, and 0.1% SDS). Samples were centrifuged for 30

minutes at 4°C, and supernatants were split into two tubes, and oligos were

precipitated with the addition of three volumes of 100% EtOH +1 µl/ml glycogen

(20 mg/ml) and overnight incubation at -20°C. Samples were then centrifuged for

30 minutes at 4°C, pellets were resuspended in 50-100 µl of DEPC treated H2O

and purity of the oligos was verified using 10% or 12.5% denaturing poly-

acrylamide gels. The purified oligos were labeled with γ32P using T4

Polynucleotide Kinase (Promega) and unincorporated nucleotides were removed

using NucAwayTM spin columns (Ambion) according to the provider’s instructions.

3.C.3. S1 nuclease protection assay

10µg of total RNA was added to 30µl of hybridization buffer (400 mM NaCl, 40 mM PIPES pH 6.4, 1 mM EDTA pH 8.0) containing labeled probes. The samples were vortexed for 5-10 seconds, incubated at 94°C for five minutes, spun down, then incubated at 70°C. After four hours, samples were allowed to slow cool to room temperature (~10 minute cool time) and 300 µl of Nuclease buffer (4.5 mM

ZnSO4, 50 mM Sodium Acetate pH 4.2, 300 mM NaCl, 120 units/ml S1 nuclease

(Invitrogen)) was added. Following 20 minutes at 37°C the reaction was stopped

148

Table 15. Oligonucleotide probes for S1 nuclease protection assays

S1 Probe Sequence rp49 GCTTGGCGCGCTCGACAATCTCCTTGCGCTTCTTGGAGGAGACGCCGTGGGCG Actin 42A GTGTGATGCCAGATCTTCTCCATGTCGTCCCAGTTAGTCACGATACC Hsp70-5' GGTTACTTTTAATTGATTCACTTTAACTTGC Hsp70-Middle GATGGTCAGGATGGAGACATCGAAGGTG Hsp70-3' GTGGTCGAACTCCTCCTTCTC Hsp26 CCATGCACACCTGGAATCCATCCTTGCC

149

with the addition of 55µl of Stop buffer (4M Sodium Acetate, 100 mM EDTA pH

8.0, 0.33 mg/ml glycogen) plus 950µl of 95% ethanol, vortexed, and allowed to precipitate at -20°C overnight. Samples were centrifuged at maximum speed

(16,100G) for 15 minutes at 4°C, supernatants were discarded, pellets were

resuspended using formamide gel loading buffer, and separated using 10%

denaturing gels. The gels were dried and results were visualized by

autoradiography.

3.C.4. Quantification of results

The intensity of each band was quantified using ImageQuant software.

Probe signals within each lane were normalized to an internal loading control (a

probe for Actin 42A mRNA) and fold inductions were calculated by comparing

normalized probe signals from each post-heat shock time point to their pre-heat

shock counterparts. For the line graphs tracking mRNA levels as a function of

time, the fold induction at T0 was set to 100%. Fold inductions from subsequent

time points were then compared to the fold induction at T0 to determine the

percentage of signal remaining. Error bars reflect the SEM of 3 or 2 independent

experiments.

3.D. Results

3.D.1. Detection of hsp70 mRNA via S1 nuclease protection assays.

Since exosome subunit depletion differentially affects mRNAs targeted by

NMD, I wanted to determine if other mRNA surveillance pathways were affected

in a similar manner. Previous studies have linked exosome subunits to the

150

binding and/or turnover of ARE-containing transcripts such as TNF-α and C-fos

mRNAs (9, 41, 136). Co-IP experiments in addition to chromatin immuno-

precipitation and polytene immunoflourescence studies have also directly linked

exosome subunits to the surveillance of actively transcribing heat shock gene loci

(12, 190). As heat shock mRNAs are readily inducible, contain AREs, and have been linked to exosome complex mediated mRNA surveillance, it seemed logical to use heat shock mRNAs to further test the exozyme hypothesis. The exozyme hypothesis predicts that depleting individual exosome subunits will make distinct

contributions to heat shock transcript stability upon recovery from heat shock. To

test this prediction, we measured the levels of heat shock transcripts upon

recovery from heat stress in exosome subunit-depleted S2 cells.

To maximize chances of detecting hsp70 mRNA, I designed probes that

targeted the 5’, middle, and 3’ regions of the hsp70 transcript (Figure 34a). The

results shown in Figures 34b and 34c are from the initial pilot experiment to verify

both the specificity of my oligonucleotide probes and the inducibility of hsp70

mRNA. Unfortunately, the probes designed to target the ARE of hsp70 mRNA

failed to hybridize efficiently. In these early experiments, a probe targeting the

mRNA of D. melanogaster ribosomal protein 49 (rp49) was used as an internal

loading control (Figure 34b). This transcript had been established by many

studies as the standard mRNA loading control for RNA studies in S2 cells (163,

165, 173, 174). However, as the mRNAs encoding mitochondrial ribosomal

proteins were affected by exosome subunit depletion in S2 cells, I chose Actin

42A mRNA to replace rp49 as the reference transcript for all future studies. Actin

151

Figure 34. Detection hsp70 mRNA via S1 nuclease protection assays.

(A) Schematic showing approximate locations of hsp70 probes. (B) Sample S1

nuclease protection assay with non-heat shocked (NHS) or heat shocked (HS)

(30 minute instantaneous) cells. Rp49 mRNA is used as an internal loading

control, and a portion of undigested probe indicates probe sizes. (C) Average fold induction of hsp70 mRNA. Error bars are the ±SEM of five experiments.

152

42A was selected since my microarray profiles indicated that its levels and 3’ to

5’ signal ratios were very stable in all my dsRNA treatments.

3.D.2. Tracking hsp70 mRNA levels upon recovery from heat shock.

To determine both a baseline level of induction, and the approximate turnover time of hsp70 mRNA in dsRNA treated cells, I harvested RNA from GFP dsRNA treated S2 cells at selected time points upon recovery from an instantaneous 30 minute heat shock (Figure 35). The dsRNA treatment regimen, instantaneous heat shock, recovery, and RNA harvesting procedures are explained in detail in section 3.C.1. A representative gel showing one GFP dsRNA treated 6 hour time course is shown in Figure 35a. The raw data are normalized to an internal loading control (Actin 42A), and shown as fold induction (Figure 35b) or as a percent of the peak hsp70 mRNA levels remaining (Figure 35c). In GFP dsRNA-

treated cells, the half-life of hsp70 mRNA upon heat shock recovery is ~40

minutes (Figure 35c). Thinned lines which are copies of these GFP results will

be reproduced on all subsequent graphs to allow for quick and easy comparisons

of the data.

3.D.3. Depletion of exosome subunits has different effects on hsp70 mRNA

levels upon recovery from heat shock.

Consistent with our microarray observations, and the predictions of the

exozyme hypothesis, depletion of exosome subunits has different effects on the

half-life of hsp70 mRNA upon recovery from heat shock (Figures 36-39).

153

Figure 35. Tracking hsp70 mRNA levels in GFP dsRNA treated cells.

A) Representative gel and (B) graph showing fold induction of hsp70 mRNA. (C)

Graph showing turnover rate of hsp70 mRNA in GFP dsRNA treated cells.

Relative probe levels (hsp70 probe versus Actin42A) at T0 were set to 100% and the portion of transcript remaining was calculated by comparing subsequent time points to T0. Error bars reflect the ±SEM of 5 independent experiments.

154

Figure 36. Depleting Dis3, Rrp46, or Rrp6 has little or no effect on hsp70 mRNA levels upon recovery from heat shock.

Representative gels and graphs showing hsp70 mRNA turnover in (A) Dis3, (B)

Rrp46, or (C) Rrp6 dsRNA treated cells. Percent remaining values were calculated as in Figure 35. Error bars reflect the ±SEM of three (2 for Rrp6) independent experiments.

155

Depletion of Dis3 had essentially no effect as hsp70 mRNA levels were within the error observed in dsGFP-treated cells (Figure 36a). This is contrary to in vitro studies using purified yeast exosome complexes which suggest that Dis3 is the only exosome subunit possessing RNase activity (61, 132, 192, 197). Depletion of one RNase PH domain protein, Rrp46, also had a very small effect at earlier time points, but a more pronounced effect at two hours post recovery (Figure

36b). Depletion of Rrp6, supposedly the other active RNase in the exosome complex, resulted in a negligible effect on hsp70 mRNA levels at early time points (Figure 36c). However, Rrp6 was the only subunit whose depletion stabilized hsp70 mRNA at the very late time points (Figure 36c). Depleting Mtr3 and Ski6, two RNase PH domain subunits, resulted in a large stabilization of hsp70 mRNA at one and two hours post heat shock (Figure 37). Depletion of

Csl4, an S1 domain subunit, stabilized hsp70 mRNA in a similar manner to both

Mtr3 and Ski6 (Figure 38a). Knockdown of Rrp47, resulted in an apparent stabilization of only the 5’ portion of hsp70 mRNA (Figure 38b). As shown by the large error bars, the degree of change caused by Rrp47 knockdown did vary considerably between replicates (Figure 38b).

3.D.3.a. Depletion of Rrp4 and Rrp40 causes continued accumulation of hsp70 mRNA after recovery from heat shock

Finally, depleting either Rrp4 or Rrp40 yielded the most provocative, yet perplexing, results. In these two depletions, hsp70 transcripts did not accumulate significantly immediately following induction but, rather, continued to

156

Figure 37. Depletion of either Mtr3 or Ski6 increases hsp70 mRNA levels.

Representative gels and graphs showing hsp70 mRNA turnover in (A) Mtr3 or (B)

Ski6 dsRNA treated cells. Percent remaining values were calculated as in Figure

35. Error bars reflect the ±SEM of three (Ski6) or two (Mtr3) independent experiments.

157

Figure 38. Depletion of Csl4 or Rrp47 affects hsp70 mRNA levels.

Representative gels and graphs showing hsp70 mRNA turnover in (A) Csl4 or (B)

Rrp47 dsRNA treated cells. Percent remaining values were calculated as in

Figure 35. Error bars reflect the ±SEM of three (Csl4) or two (Rrp47) independent experiments.

158

Figure 39. Depletion of Rrp4 or Rrp40 causes hsp70 mRNA to increase after removal of heat stress.

Representative gels and graphs showing hsp70 mRNA turnover in (A) Rrp4 or

(B) Rrp40 dsRNA treated cells. Percent remaining values were calculated as in

Figure 35. Error bars reflect the ±SEM of three independent experiments.

159

Figure 40. hsp70 mRNA continues to accumulate in Rrp4 and Rrp40 depleted cells. hsp70 mRNA levels (as measured by fold induction) upon recovery from heat shock in exosome subunit depleted cells. The average fold induction at (A) 0 hours, (B) 1 hour, (C) 2 hours, (D) 4 hours, (E) 6 hours of three (two for Mtr3,

Rrp6, and Rrp47) independent RNAi/HS experiments.

160 increase after removal of heat stress (Figures 39 and 40). Unlike every other exosome subunit (and control sample) depletion, where hsp70 mRNA levels peaked within five minutes of heat stress removal (T0) (c.f. Figures 36-39), hsp70 mRNA peaked near the four or six hour time points of Rrp4 and Rrp40 depleted cells (Figures 39 and 40). The extent to which these transcripts accumulated (as shown by % remaining calculations) did fluctuate between independent replicates; hence the large error bars (Figure 39). However, the result that hsp70 mRNA levels continued to increase hours after heat stress removal when either Rrp4 or Rrp40 was depleted was invariant (Figures 39 and

40). Further, this lack of rapid hsp70 induction, followed by continued accumulation of hsp70 mRNA was only observed in Rrp4 and Rrp40 depleted cells (Figure 40). As suggested by Figure 37c, Rrp6 depleted cells did have increased levels of hsp70 mRNA at the later time points (Figure 40c-e).

However, unlike Rrp4 and Rrp40 depleted cells, the levels of hsp70 mRNA had an initial peak at T0, and then declined (Figure 40). Further experiments are required to determine if the increased prevalence of hsp70 mRNA results from continued transcription, delayed turnover, or a combination of those sources.

3.D.4. Levels of hsp26 mRNA upon recovery from heat shock are altered by exosome subunit depletion.

To determine whether exosome subunit depletion affected another transcript similarly, I assayed an additional heat shock transcript, hsp26. Previous studies have shown that, similar to hsp70 mRNA, hsp26 mRNA was also inducible with

161

Figure 41. Tracking hsp26 mRNA levels in GFP dsRNA treated cells

(A) Cartoon showing hsp26 mRNA and the position of the S1 probe.

Representative (B) gel and graphs showing hsp26 mRNA levels as (C) fold induction or (D) percent remaining in GFP dsRNA treated cells. Error bars are the ±SEM of 3 independent 30 minute instantaneous heat shock experiments.

162

short instantaneous heat shocks (11, 54). I used RNA harvested from the

previous exosome subunit dsRNA and heat shock treatments to assess hsp26 mRNA levels upon recovery from heat shock. Unlike hsp70 mRNA levels, which peaked very close to T0, hsp26 mRNA levels continued to increase for some time after return to 27°C; thus, the expression profiles of hsp26 and hsp70 mRNAs differ (cf. Figures 35 and 41). In these experiments, hsp26 mRNA levels appear to peak at 1 hour post heat shock (Figure 41); however, since there are no intermediate time points, we cannot exclude that the peak of hsp26 mRNA levels occurs between the zero and one hour time points. This difference between the rate and duration of transcription and/or turnover of these two heat shock transcripts makes a direct comparison of my hsp70 and hsp26 data difficult. However, relative comparisons such as the ones between exosome subunit depleted cells versus GFP dsRNA treated cells are still informative. The overall pattern showing that the depletion of individual exosome subunits has different effects in S2 cells is reproduced by these experiments. Further, these results with hsp26 mRNA support my microarray data presented in Chapter 2 and extend upon the hsp70 data presented above.

When compared to the effects on hsp70 mRNA levels, the effects of exosome subunit depletion on hsp26 levels are either different or similar depending upon the subunit depleted. For example, depletion of either Ski6 or Csl4 stabilized hsp70 significantly at early time points (Figures 37b and 38a). The stabilization effect observed with hsp26 mRNA upon depletion of these two subunits was

163

Figure 42. Depletion of certain exosome subunits yields different results when examining hsp70 and hsp26.

Representative gels and graphs showing hsp26 mRNA levels as percent remaining in (A) Ski6, (B) Rrp46, or (C) Csl4 depleted S2 cells. Error bars are the ±SEM of three (Ski6) or two (Rrp46, Csl4) independent experiments.

164

much smaller (when compared to GFP dsRNA treated cells) than that observed

with hsp70 mRNA (Figures 42a and 42c). Moreover, Rrp46 depletion had only a

small effect on hsp70 levels at 2 hours post heat shock, but depletion of this

subunit increased hsp26 mRNA levels (at the one and two hour time points),

relative to GFP dsRNA treated cells, more than any other exosome subunit

(Figure 42b).

Notably however, similar effects were observed for both heat shock

transcripts when most individual exosome subunits were depleted (Figures 36-39

and 43). This includes the negligible effects of Dis3 depletion which has an

hsp26 mRNA curve that is almost identical to GFP dsRNA treated cells (Figure

43a). Depletion of Rrp47 also stabilizes hsp26 mRNA in a slightly more pronounced manner than Dis3 (Figure 43b). The behavior of hsp26 mRNA directly mirrors that of hsp70 mRNA in Rrp4 or Rrp40 depleted cells (c.f. Figures

40 and 43). Specifically, the depletion of either Rrp4 or Rrp40 leads to a delay in hsp26 mRNA accumulation and turnover. These results demonstrate that depleting individual exosome subunits causes subunit specific effects on levels of hsp70 and hsp26 mRNAs upon recovery from heat shock.

3.E. Discussion

In this study I performed another in vivo test of the exozyme model. I assay the levels of two heat shock induced transcripts, hsp70 and hsp26, upon recovery from heat shock to assess the effects of depleting individual exosome subunits on the turnover of these ARE containing mRNAs. The exozyme model predicts that depletion of individual exosome subunits would have distinct effects

165

Figure 43. hsp26 mRNA behaves similarly to hsp70 mRNA in Dis3, Rrp47,

Rrp4, and Rrp40 depleted cells.

Representative gels and graphs showing hsp26 mRNA levels as percent remaining in (A) Dis36, (B) Rrp47, (C) Rrp4, or (D) Rrp40 depleted S2 cells.

Error bars are the ±SEM of three (A, B, C), or two (D) independent experiments.

166 on the turnover of RNA; while the core exosome model predicts that all exosome subunit depletions would have very similar effects. The different effects of individual exosome subunit depletion on hsp70 and hsp26 mRNA stability upon recovery from heat shock are inconsistent with the core exosome model yet provide further support for the exozyme model. Briefly, depending upon the transcript assayed, some exosome subunit depletions stabilized the mRNAs while others had little or no effect. For example, depletion of Dis3 had essentially no effect on the levels of either hsp70 or hsp26 mRNA, while depletion of either

Rrp4 or Rrp40 resulted in the continued accumulation of both hsp70 and hsp26 mRNAs. This observation is surprising as the primary RNase activity of the exosome complex is thought to reside in Dis3 (61). We must also add that depletion of Rrp4 or Rrp40 did not result in the co-depletion of any other exosome subunit. Although the depletion of most exosome subunits caused similar effects when the two heat shock transcripts were assayed, there were some differences.

As mentioned above, the depletion of some exosome subunits had different effects on hsp70 and hsp26 mRNAs. The largest difference occurred when comparing hsp70 and hsp26 mRNA levels in Rrp46 depleted cells. In those experiments, Rrp46 depletion elicited a large effect and caused an increase in hsp26 levels over time. Conversely, Rrp46 depletion produced only a minor effect on hsp70 mRNA over the same time period. The opposite observation was true for Ski6 and Csl4 depleted cells. In cells depleted of those two subunits, hsp26 mRNA was stabilized to a lesser extent than hsp70 mRNA.

167

There are several possible explanations of the differences observed between

transcripts and between exosome subunit depletions. The different effects may

be caused indirectly through ARE binding proteins (ARE-BPs) interacting with different binding sequences within the AREs of these two transcripts (9, 41, 76,

83, 143). It is possible that one or more ARE-BPs can bind differentially to the slightly different sequences in the AREs of the two heat shock transcripts. In agreement with this, several known ARE-BPs such as Tris-tetraproline (TTP), K

homology splicing regulatory Protein (KSRP), and BRF-1 have been linked to the

turnover of mRNAs containing different ARE sequences (9, 76, 83, 143, 180). In

this proposed model, these ARE-BPs may would then recruit or target these two mRNAs to different exosome subunits (9, 41, 76, 83, 143).

Another possibility is that certain exosome subunits directly interact with the

ARE containing mRNAs (9). Indeed, the human versions of Rrp45, Rrp43, and

Ski6 (all RNase PH domain containing subunits) have been shown interact with

ARE-like RNA sequences in vitro (9). The large apparent increase in hsp70

mRNA half-life I observed in Ski6 and Mtr3 depleted cells indicate that such a direct interaction may play a fundamental role in the surveillance of this transcript

in vivo (9). Curiously, not all RNase PH domain subunits behaved in a similar

manner, both when concerning only hsp70 mRNA and when the study was

broadened to include an additional transcript hsp26. For example, Ski6 depletion

stabilized hsp70 markedly, but had little effect on hsp26 levels. Depletion of

Rrp46 had exactly the opposite effect. Although the study by Wilusz and

coworkers did not test whether the exosome subunits bound the ARE-like RNA in

168

a sequence specific manner, such sequence specific interactions of the RNAse

PH domain subunits could help to explain both my observations with heat shock

transcripts and one facet of exosome complex substrate specificity (9).

Alternatively, the depleted subunits may not have direct role, as defined by

RNase activity of a particular subunit, in degrading hsp70 or hsp26 mRNA;

however, they may be critical components, structural or otherwise, of exozymes

that degrade those specific transcripts.

The most provocative result in the heat shock mRNA studies was observed

with the depletion of Rrp4 and Rrp40. In these cases, there was an apparent

delayed onset of transcription with heat shock transcript levels continuing to

elevate long after removal of heat stress. This prolonged transcriptional effect in

the absence of continuing transcriptional stimulus is, to the best of our

knowledge, unprecedented in any transcriptional study. One possibility is that

the timeframe of hsp70 and hsp26 induction has been stretched to a

considerably longer interval. Repeating the experiments with longer recovery

times (8, 12, 16, 20, and 24 hours) would confirm or refute this possibility.

Additional experiments, such as adding a transcriptional inhibitor like actinomycin

D to recovering cells, are required to verify that the accumulation of hsp70 and

hsp26 mRNA is dependent upon new transcription (54, 174). These experiments

are expected to show that the continued accumulation of hsp70 and hsp26

mRNA is dependent upon RNA Polymerase II (Pol II).

The next experiments would test to see if the transcription result is dependent

upon heat shock factor (HSF), a transcription factor known to drive heat shock

169

gene expression in response to heat stress (1, 157, 159, 176). First, we must

verify that HSF levels are normal in cells depleted of Rrp4 and Rrp40. Second,

we can use a time course of chromatin immunoprecipitation experiments to verify

that HSF is properly binding to the promoter regions of heat shock genes upon

heat stress. If HSF levels, promoter recruitment, or promoter occupancy are

different from untreated cells, then the transcriptional effect is most likely HSF

dependant. The extended transcription can be explained by an insufficient initial

pulse of hsp70 protein which operates in a feedback loop with HSF (1, 2, 157).

This would cause HSF to remain bound to heat shock loci upon recovery from heat stress, and hence lead to the continued mRNA accumulation observed in

Rrp4- and Rrp40-depleted cells (1, 2, 157). While such an effect explains the continued expression of the heat shock genes observed in Rrp4 and Rrp40 depleted cells, it does not explain how the transcripts are not properly induced.

Chromatin immunoprecipitation time courses assessing Pol II occupancy at heat shock loci can assess if polymerases escape the promoter area early and cause a ‘log-jam’ of slow polymerase molecules upon heat shock. Such a mechanism can explain both the slow, yet continued accumulation of both hsp70 and hsp26

mRNAs.

However, if Pol II and HSF binding and activity at the heat shock loci of Rrp4

and Rrp40 depleted cells are consistent with that of normal cells, then the

transcriptional effect could possibly be observed with other inducible promoters.

This possibility can be tested with time course experiments using additional

inducible endogenous genes or reporter constructs in exosome subunit depleted

170

cells. If the transcriptional effects are conserved between multiple inducible

genes, it is likely that Rrp4 and Rrp40 depletion elicits a general transcriptional

defect in S2 cells.

Several possible explanations can be derived from the relationship between exosome subunits and elongating transcriptional complexes on heat shock genes. Since Drosophila exosome subunits have been shown to interact with the

transcription elongation factors Spt5 and Spt6 on actively transcribing heat shock genes perhaps an exozyme containing Rrp4 and Rrp40 is required for efficient

elongation by Pol II transcription complexes (12, 190). Loss of either one of

these two proteins may inhibit Pol II elongation and hence lead to a muted initial

heat shock response in S2 cells. In this scenario, an insufficient hsp70/HSF

feedback loop (as outlined above) would cause the RNA to accumulate after heat

stress removal. The experiments outlined above (and others) are needed to

elucidate the enigmatic molecular mechanisms that govern the aberrant heat

shock response observed in Rrp4- and Rrp40-depleted cells.

Our hsp70 data directly contradict the conclusions of another group which claims that exosome subunits are not required for hsp70 mRNA turnover in S2 cells (20). In their study, Wahle and coworkers depleted only one cytoplasmic

exosome cofactor (Ski3) in S2 cells, assayed hsp70 levels, and drew conclusions

related to the function of the entire exosome complex (20). Although they

followed a common convention, their approach was incomplete for several

reasons (20). First, they only examined a cytoplasmic exosome cofactor and

hence all nuclear functions of exosome subunits were unaddressed (20).

171

Second, they did not perform any experiments targeting actual exosome subunits to verify that exosome subunits were involved (20). Lastly, the behavior of their hsp70 ARE point mutations were not consistent with the in vivo or in vitro

behavior of hsp70 mRNA truncations or reporters from earlier studies (20, 173,

174). In sum, this study from Wahle and coworkers was incomplete with respect

to the subunits of the exosome they assayed (none), and relied on reporters

which did not recapitulate the behavior, both in vitro and in vivo, of the

endogenous transcript they were attempting to assess (20, 173, 174). Therefore,

additional studies are required to understand the contributions of cytoplasmic and

nuclear exosome subunits to hsp70 mRNA turnover upon recovery from heat

shock.

3.F. Conclusions

Taken together, the results in this study show that depletion of individual

exosome subunits has distinct effects on the levels of two different ARE

containing mRNAs in vivo. While these observations directly contradict the core

exosome model which predicts exactly the opposite results, they precisely match

the predictions of the exozyme model. In the exozyme model, exosome subunits

assemble into distinct sub-complexes which process and/or degrade RNAs in

vivo. The substrate specificity of different exozyme complexes would be

determined by their composition of exosome subunits and their interaction with

other associated non-exosome proteins.

172

The different heat shock transcript specific effects observed with the depletion

of individual exosome subunits coupled with the observation of RNAse PH-

domain exosome subunits binding ARE-like mRNA sequences in vitro offers an explanation (9). This could suggest that direct (or indirect via RNA-BPs) sequence specific interactions occur between exosome subunits and the surveyed mRNAs. The experiments outlined earlier would begin to evaluate that possibility. These heat shock data, when coupled with my previous microarray observations establish a strong argument against the core exosome model and support exozyme model of exosome subunit function.

3.G. Acknowledgements

The authors would like to thank Dr. Paul Fisher for the Lamin antibody and members of the Andrulis lab for discussions. This work is supported by grant

GM072820 from the NIH (E.D.A). E.D.A. is a Mount Sinai Health Care

Foundation Scholar.

173

Chapter 4:

General discussion and future directions

By proposing the exozyme model, I suggest that eukaryotic exosome subunit assemblages have more in common with their archaeal forebears. In

Archaeoglobus fulgidus, at least two structurally distinct exosome complexes,

Csl4- or Rrp4-capped, are possible, and purifications from cells suggest the presence of exosomes with mixed Csl4/Rrp4 caps in archaea (32). The presence of exosome complexes with multiple cap configurations could explain how archaeal exosome complexes recognize distinct substrates in vivo and serves as the in vivo basis of the exozyme model. Indeed, the hypothesis that eukaryotic exosome subunits assemble into distinct exozyme complexes could explain several current unknowns in the field of exosome biology. Most importantly, the exozyme hypothesis, which is consistent with, and accomodates all published in vivo data, offers a mechanism by which different exozymes can recognize the many substrate types attributed to the exosome complex.

As my study only surveyed mRNAs, I will explain how this putative substrate recognition mechanism would work with that class of transcripts, and then expand it to include the remaining RNA types. The substrate recognition mechanism posited by the exozyme hypothesis relies upon different exosome subunit assemblages recognizing distinct RNA regulatory elements in a sequence specific manner. These regulatory elements could be targets individual exosome subunits, exozymes composed of a subset of exosome

174

subunits, or to an exozyme consisting of all exosome subunits. I envision three

major mechanisms for this exozyme-specific targeting of surveyed mRNAs.

The first mechanism would occur via protein-protein interactions with

sequence specific RNA binding proteins and exosome subunits. The prototype

for this is the interaction of several exosome subunits with KSRP and TTP, which

are required for the turnover of the ARE-containing c-fos and TNF mRNAs in vitro (41). In agreement with this proposed mechanism, several different ARE-

BPs have been identified and are known to interact with distinct ARE sequences

(41-43, 76, 83, 84). Proteomic studies in yeast have also identified many proteins that only interact with a limited subset of exosome subunits (80, 125).

These proteins are among the prime candidates for these putative additional

substrate specific exozyme cofactors (80, 125). For an example from archaeal

complexes, homo-trimeric Csl4 caps would only interact with a subset of RNA-

BPs, while homo-trimeric Rrp4 caps would interact with a distinct group of RNA-

BPs. In agreement with the predictions of this putative mechanism, my results

show that the transcript profiles of S1 domain subunit depleted cells have a

smaller degree of overlap than those from cells depleted of RNase PH domain

subunits.

The second mechanism could rely on direct interactions of the mRNA

regulatory sequence(s) with the RNA binding (or other undefined) domains of

one or more exosome subunit(s). Rrp40, Rrp4, and Csl4, all of which contain S1

domains and form a “cap” on the complex, would be among the likeliest

candidates for these interacting subunits. One could speculate that in archaea, a

175

direct RNA -S1 cap-dependent substrate recognition mechanism already exists.

This mechanism could occur as outlined above, but not require additional proteins to serve as molecular bridges between the surveyed RNAs and exozyme complexes. Other exosome subunits, such as the RNase PH domain subunits, may also play a role in directly RNA binding in vivo. Wilusz and

colleagues showed that three human RNase PH domain subunits, Rrp45, Rrp43,

and Ski6, can bind ARE-like RNA sequences in a sequence specific manner in

vitro (9). This result shows that multiple exosome subunits can directly interact

with RNAs in a sequence specific manner, and raises the possibility that such

interactions may help contribute to the substrate specificity of exosome subunits

(9).

Finally, the third mechanism achieves substrate specificity by requiring an

additional layer of complexity. This last putative mechanism entails specifically

expressed mi- or siRNAs which base-pair with the putative UTR regulatory

element(s) and target those mRNAs for exosome subunit mediated turnover.

The expression of these putative mi- or siRNAs can be regulated in a tissue

specific or other manner. As exosome subunits, specifically Csl4 and Rrp4, have

been linked to the turnover of siRNA targeted transcripts, this putative RNA

regulatory scheme is also grounded in experimental observations (165). It is also

likely that these three proposed mechanisms could work both independently and

in concert, thus providing an exquisitely precise mechanism to regulate the levels

of multiple transcripts using the same conserved machinery (exosome subunits).

Similar RNA sequence elements are likely present in other types of RNA, thereby

176

helping to explain how the exozyme complexes are recruited to and process or degrade other classes of RNAs. Based on many reported observations from both transcriptome-wide and targeted studies, I speculate that the subunit- specific differences seen with mRNA stability will extend to other types of RNA substrates (4-6, 39, 100, 178).

A high throughput bioinformatics screen of the affected mRNA sequences would likely identify these putative regulatory RNA sequence elements if they were present. Sequence analysis of the UTRs of affected mRNAs would be the most logical starting point of this work. This idea is supported by previous observations showing that exosome subunits are involved in the surveillance of

UTR elements located in the C-fos, TNF-α, and phosphoglycerate kinase mRNAs

(18, 19, 41, 42). Also, my UTR data (presented in Chapter 2) strongly suggests that the UTR regions of exosome subunit surveyed mRNAs are important. I predict that several types of regulatory sequences would emerge from such an analysis. Such sequences could be RNA secondary structures such as stem- loops, which serve either indirectly as RNA-BP binding sites, or directly, as exosome subunit docking sites. Other unstructured sequences could also work in a similar fashion. It is also likely that these sequences will not be completely mutually exclusive, as multiple exosome subunits, or exosome subunit cofactors, could interact with the same RNA regulatory site in either a competitive or cooperative manner. The resulting sequences can be verified by adding them to the 3’UTR regions of reporter constructs. Alternatively, an RNA-targeted

177 chromatin immunoprecipitation regimen using antibodies to known RNA-BPs could help to identify novel RNA regulatory motifs and/or sequences.

The widespread and significant subunit-specific differences in mRNA turnover in this and previous studies challenge the core exosome model of exosome subunit function in vivo. Indeed, using both genome-wide and gene-targeted approaches, I demonstrate that knockdown of different exosome subunits has different effects on mRNA stability in vivo. Based on many reported observations from both transcriptome-wide and targeted studies, I speculate that the subunit- specific differences I show with mRNA stability will extend to other types of RNA substrates in Drosophila melanogaster (4-6, 39, 100, 178). Evidence compiled from tiling arrays in Arabidopsis thaliana that shows different effects of exosome subunit depletion on multiple RNA types strongly supports this idea (39). Other incomplete results from human cells give credence to this idea by showing that

PROMPT levels are also differentially affected in exosome subunit depleted cells

(178). I propose that the in vivo RNA metabolic functions of exosome subunits are best explained by the exozyme model.

178

List of references

1. Abravaya, K., M. P. Myers, S. P. Murphy, and R. I. Morimoto. 1992. The human heat shock protein hsp70 interacts with HSF, the transcription factor that regulates heat shock gene expression. Genes Dev 6:1153-64. 2. Abravaya, K., B. Phillips, and R. I. Morimoto. 1991. Attenuation of the heat shock response in HeLa cells is mediated by the release of bound heat shock transcription factor and is modulated by changes in growth and in heat shock temperatures. Genes Dev 5:2117-27. 3. Alfonzo, J. D., O. Thiemann, and L. Simpson. 1997. The mechanism of U insertion/deletion RNA editing in kinetoplastid mitochondria. Nucleic Acids Res 25:3751-9. 4. Allmang, C., J. Kufel, G. Chanfreau, P. Mitchell, E. Petfalski, and D. Tollervey. 1999. Functions of the exosome in rRNA, snoRNA and snRNA synthesis. Embo J 18:5399-410. 5. Allmang, C., P. Mitchell, E. Petfalski, and D. Tollervey. 2000. Degradation of ribosomal RNA precursors by the exosome. Nucleic Acids Res 28:1684-91. 6. Allmang, C., E. Petfalski, A. Podtelejnikov, M. Mann, D. Tollervey, and P. Mitchell. 1999. The yeast exosome and human PM-Scl are related complexes of 3' --> 5' exonucleases. Genes Dev 13:2148-58. 7. Allmang, C., and D. Tollervey. 1998. The role of the 3' external transcribed spacer in yeast pre-rRNA processing. J Mol Biol 278:67-78. 8. Aloy, P., F. D. Ciccarelli, C. Leutwein, A. C. Gavin, G. Superti-Furga, P. Bork, B. Bottcher, and R. B. Russell. 2002. A complex prediction: three-dimensional model of the yeast exosome. EMBO Rep 3:628-35. 9. Anderson, J. R., D. Mukherjee, K. Muthukumaraswamy, K. C. Moraes, C. J. Wilusz, and J. Wilusz. 2006. Sequence-specific RNA binding mediated by the RNase PH domain of components of the exosome. RNA 12:1810-6. 10. Anderson, J. S., and R. P. Parker. 1998. The 3' to 5' degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3' to 5' exonucleases of the exosome complex. Embo J 17:1497-506. 11. Andrulis, E. D., E. Guzman, P. Doring, J. Werner, and J. T. Lis. 2000. High-resolution localization of Drosophila Spt5 and Spt6 at heat shock genes in vivo: roles in promoter proximal pausing and transcription elongation. Genes Dev 14:2635-49. 12. Andrulis, E. D., J. Werner, A. Nazarian, H. Erdjument-Bromage, P. Tempst, and J. T. Lis. 2002. The RNA processing exosome is linked to elongating RNA polymerase II in Drosophila. Nature 420:837-41. 13. Araki, Y., S. Takahashi, T. Kobayashi, H. Kajiho, S. Hoshino, and T. Katada. 2001. Ski7p G protein interacts with the exosome and the Ski complex for 3'-to-5' mRNA decay in yeast. Embo J 20:4684-93. 14. Arigo, J. T., D. E. Eyler, K. L. Carroll, and J. L. Corden. 2006. Termination of cryptic unstable transcripts is directed by yeast RNA-binding proteins Nrd1 and Nab3. Mol Cell 23:841-51. 15. Berretta, J., M. Pinskaya, and A. Morillon. 2008. A cryptic unstable transcript mediates transcriptional trans-silencing of the Ty1 retrotransposon in S. cerevisiae. Genes Dev 22:615-26. 16. Blackburn, E. H., C. W. Greider, and J. W. Szostak. 2006. Telomeres and telomerase: the path from maize, Tetrahymena and yeast to human cancer and aging. Nat Med 12:1133-8.

179

17. Blackshear, P. J. 2002. Tristetraprolin and other CCCH tandem zinc-finger proteins in the regulation of mRNA turnover. Biochem Soc Trans 30:945-52. 18. Blattner, J., and C. E. Clayton. 1995. The 3'-untranslated regions from the Trypanosoma brucei phosphoglycerate kinase-encoding genes mediate developmental regulation. Gene 162:153-6. 19. Blattner, J., S. Helfert, P. Michels, and C. Clayton. 1998. Compartmentation of phosphoglycerate kinase in Trypanosoma brucei plays a critical role in parasite energy metabolism. Proc Natl Acad Sci U S A 95:11596-600. 20. Bonisch, C., C. Temme, B. Moritz, and E. Wahle. 2007. Degradation of hsp70 and other mRNAs in Drosophila via the 5' 3' pathway and its regulation by heat shock. J Biol Chem 282:21818-28. 21. Bonneau, F., J. Basquin, J. Ebert, E. Lorentzen, and E. Conti. 2009. The yeast exosome functions as a macromolecular cage to channel RNA substrates for degradation. Cell 139:547-59. 22. Bousquet-Antonelli, C., C. Presutti, and D. Tollervey. 2000. Identification of a regulated pathway for nuclear pre-mRNA turnover. Cell 102:765-75. 23. Brantl, S. 2002. Antisense-RNA regulation and RNA interference. Biochim Biophys Acta 1575:15-25. 24. Brantl, S. 2007. Regulatory mechanisms employed by cis-encoded antisense RNAs. Curr Opin Microbiol 10:102-9. 25. Briggs, M. W., K. T. Burkard, and J. S. Butler. 1998. Rrp6p, the yeast homologue of the human PM-Scl 100-kDa autoantigen, is essential for efficient 5.8 S rRNA 3' end formation. J Biol Chem 273:13255-63. 26. Brouwer, R., C. Allmang, R. Raijmakers, Y. van Aarssen, W. V. Egberts, E. Petfalski, W. J. van Venrooij, D. Tollervey, and G. J. Pruijn. 2001. Three novel components of the human exosome. J Biol Chem 276:6177-84. 27. Brouwer, R., G. J. Pruijn, and W. J. van Venrooij. 2001. The human exosome: an autoantigenic complex of exoribonucleases in myositis and scleroderma. Arthritis Res 3:102-6. 28. Brouwer, R., W. T. Vree Egberts, G. J. Hengstman, R. Raijmakers, B. G. van Engelen, H. P. Seelig, M. Renz, R. Mierau, E. Genth, G. J. Pruijn, and W. J. van Venrooij. 2002. Autoantibodies directed to novel components of the PM/Scl complex, the human exosome. Arthritis Res 4:134-8. 29. Brown, J. T., X. Bai, and A. W. Johnson. 2000. The yeast antiviral proteins Ski2p, Ski3p, and Ski8p exist as a complex in vivo. Rna 6:449-57. 30. Brune, C., S. E. Munchel, N. Fischer, A. V. Podtelejnikov, and K. Weis. 2005. Yeast poly(A)-binding protein Pab1 shuttles between the nucleus and the cytoplasm and functions in mRNA export. RNA 11:517-31. 31. Burkard, K. T., and J. S. Butler. 2000. A nuclear 3'-5' exonuclease involved in mRNA degradation interacts with Poly(A) polymerase and the hnRNA protein Npl3p. Mol Cell Biol 20:604-16. 32. Buttner, K., K. Wenig, and K. P. Hopfner. 2005. Structural framework for the mechanism of archaeal exosomes in RNA processing. Mol Cell 20:461-71. 33. Callahan, K. P., and J. S. Butler. 2008. Evidence for core exosome independent function of the nuclear exoribonuclease Rrp6p. Nucleic Acids Res 36:6645-55. 34. Camblong, J., N. Iglesias, C. Fickentscher, G. Dieppois, and F. Stutz. 2007. Antisense RNA stabilization induces transcriptional gene silencing via histone deacetylation in S. cerevisiae. Cell 131:706-17.

180

35. Canavan, R., and U. Bond. 2007. Deletion of the nuclear exosome component RRP6 leads to continued accumulation of the histone mRNA HTB1 in S-phase of the cell cycle in Saccharomyces cerevisiae. Nucleic Acids Res 35:6268-79. 36. Carneiro, T., C. Carvalho, J. Braga, J. Rino, L. Milligan, D. Tollervey, and M. Carmo- Fonseca. 2007. Depletion of the yeast nuclear exosome subunit Rrp6 results in accumulation of polyadenylated RNAs in a discrete domain within the nucleolus. Mol Cell Biol 27:4157-65. 37. Chedin, S., A. Laferte, T. Hoang, D. L. Lafontaine, M. Riva, and C. Carles. 2007. Is ribosome synthesis controlled by pol I transcription? Cell Cycle 6:11-5. 38. Chekanova, J. A., J. A. Dutko, I. S. Mian, and D. A. Belostotsky. 2002. Arabidopsis thaliana exosome subunit AtRrp4p is a hydrolytic 3'-->5' exonuclease containing S1 and KH RNA-binding domains. Nucleic Acids Res 30:695-700. 39. Chekanova, J. A., B. D. Gregory, S. V. Reverdatto, H. Chen, R. Kumar, T. Hooker, J. Yazaki, P. Li, N. Skiba, Q. Peng, J. Alonso, V. Brukhin, U. Grossniklaus, J. R. Ecker, and D. A. Belostotsky. 2007. Genome-wide high-resolution mapping of exosome substrates reveals hidden features in the Arabidopsis transcriptome. Cell 131:1340-53. 40. Chekanova, J. A., R. J. Shaw, M. A. Wills, and D. A. Belostotsky. 2000. Poly(A) tail- dependent exonuclease AtRrp41p from Arabidopsis thaliana rescues 5.8 S rRNA processing and mRNA decay defects of the yeast ski6 mutant and is found in an exosome-sized complex in plant and yeast cells. J Biol Chem 275:33158-66. 41. Chen, C. Y., R. Gherzi, S. E. Ong, E. L. Chan, R. Raijmakers, G. J. Pruijn, G. Stoecklin, C. Moroni, M. Mann, and M. Karin. 2001. AU binding proteins recruit the exosome to degrade ARE-containing mRNAs. Cell 107:451-64. 42. Chou, C. F., A. Mulky, S. Maitra, W. J. Lin, R. Gherzi, J. Kappes, and C. Y. Chen. 2006. Tethering KSRP, a decay-promoting AU-rich element-binding protein, to mRNAs elicits mRNA decay. Mol Cell Biol 26:3695-706. 43. Ciais, D., M. T. Bohnsack, and D. Tollervey. 2008. The mRNA encoding the yeast ARE- binding protein Cth2 is generated by a novel 3' processing pathway. Nucleic Acids Res 36:3075-84. 44. Clayton, C., A. Schwede, M. Stewart, A. Robles, C. Benz, J. Po, M. Wurst, R. Queiroz, and S. Archer. 2008. Control of mRNA degradation in trypanosomes. Biochem Soc Trans 36:520-1. 45. Colasante, C., A. Robles, C. H. Li, A. Schwede, C. Benz, F. Voncken, D. L. Guilbride, and C. Clayton. 2007. Regulated expression of glycosomal phosphoglycerate kinase in Trypanosoma brucei. Mol Biochem Parasitol 151:193-204. 46. Cole, S. E., F. J. LaRiviere, C. N. Merrikh, and M. J. Moore. 2009. A convergence of rRNA and mRNA quality control pathways revealed by mechanistic analysis of nonfunctional rRNA decay. Mol Cell 34:440-50. 47. Cristodero, M., and C. E. Clayton. 2007. Trypanosome MTR4 is involved in rRNA processing. Nucleic Acids Res 35:7023-30. 48. Culbertson, M. R. 1999. RNA surveillance. Unforeseen consequences for gene expression, inherited genetic disorders and cancer. Trends Genet 15:74-80. 49. Das, B., J. S. Butler, and F. Sherman. 2003. Degradation of normal mRNA in the nucleus of Saccharomyces cerevisiae. Mol Cell Biol 23:5502-15. 50. Das, B., S. Das, and F. Sherman. 2006. Mutant LYS2 mRNAs retained and degraded in the nucleus of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 103:10871-6.

181

51. Davis, C. A., and M. Ares, Jr. 2006. Accumulation of unstable promoter-associated transcripts upon loss of the nuclear exosome subunit Rrp6p in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 103:3262-7. 52. de la Cruz, J., D. Kressler, D. Tollervey, and P. Linder. 1998. Dob1p (Mtr4p) is a putative ATP-dependent RNA helicase required for the 3' end formation of 5.8S rRNA in Saccharomyces cerevisiae. EMBO J 17:1128-40. 53. De Marchis, M. L., A. Giorgi, M. E. Schinina, I. Bozzoni, and A. Fatica. 2005. Rrp15p, a novel component of pre-ribosomal particles required for 60S ribosome subunit maturation. RNA 11:495-502. 54. Dellavalle, R. P., R. Petersen, and S. Lindquist. 1994. Preferential deadenylation of Hsp70 mRNA plays a key role in regulating Hsp70 expression in Drosophila melanogaster. Mol Cell Biol 14:3646-59. 55. Desset, S., N. Buchon, C. Meignin, M. Coiffet, and C. Vaury. 2008. In Drosophila melanogaster the COM locus directs the somatic silencing of two retrotransposons through both Piwi-dependent and -independent pathways. PLoS ONE 3:e1526. 56. Dez, C., J. Houseley, and D. Tollervey. 2006. Surveillance of nuclear-restricted pre- ribosomes within a subnucleolar region of Saccharomyces cerevisiae. Embo J 25:1534- 46. 57. Dichtl, B., and D. Tollervey. 1997. Pop3p is essential for the activity of the RNase MRP and RNase P ribonucleoproteins in vivo. EMBO J 16:417-29. 58. Doma, M. K., and R. Parker. 2006. Endonucleolytic cleavage of eukaryotic mRNAs with stalls in translation elongation. Nature 440:561-4. 59. Dulebohn, D., J. Choy, T. Sundermeier, N. Okan, and A. W. Karzai. 2007. Trans- translation: the tmRNA-mediated surveillance mechanism for ribosome rescue, directed protein degradation, and nonstop mRNA decay. Biochemistry 46:4681-93. 60. Dunn, E. F., C. M. Hammell, C. A. Hodge, and C. N. Cole. 2005. Yeast poly(A)-binding protein, Pab1, and PAN, a poly(A) nuclease complex recruited by Pab1, connect mRNA biogenesis to export. Genes Dev 19:90-103. 61. Dziembowski, A., E. Lorentzen, E. Conti, and B. Seraphin. 2007. A single subunit, Dis3, is essentially responsible for yeast exosome core activity. Nat Struct Mol Biol 14:15-22. 62. Edgar, R., M. Domrachev, and A. E. Lash. 2002. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30:207-10. 63. Egecioglu, D. E., A. K. Henras, and G. F. Chanfreau. 2006. Contributions of Trf4p- and Trf5p-dependent polyadenylation to the processing and degradative functions of the yeast nuclear exosome. RNA 12:26-32. 64. Estevez, A. M., T. Kempf, and C. Clayton. 2001. The exosome of Trypanosoma brucei. Embo J 20:3831-9. 65. Estevez, A. M., B. Lehner, C. M. Sanderson, T. Ruppert, and C. Clayton. 2003. The roles of intersubunit interactions in exosome stability. J Biol Chem 278:34943-51. 66. Evguenieva-Hackenberg, E., V. Roppelt, P. Finsterseifer, and G. Klug. 2008. Rrp4 and Csl4 are needed for efficient degradation but not for polyadenylation of synthetic and natural RNA by the archaeal exosome. Biochemistry 47:13158-68. 67. Evguenieva-Hackenberg, E., S. Wagner, and G. Klug. 2008. In vivo and in vitro studies of RNA degrading activities in Archaea. Methods Enzymol 447:381-416. 68. Evguenieva-Hackenberg, E., P. Walter, E. Hochleitner, F. Lottspeich, and G. Klug. 2003. An exosome-like complex in Sulfolobus solfataricus. EMBO Rep 4:889-93. 69. Evguenieva-Hackenburg, E., P. Walter, E. Hochleitner, F. Lottspeich, and G. Klug. 2003. An exosome-like complex in Sulfolobus solfataricus. EMBO Rep 4:889-93.

182

70. Fang, F., J. Hoskins, and J. S. Butler. 2004. 5-fluorouracil enhances exosome-dependent accumulation of polyadenylated rRNAs. Mol Cell Biol 24:10766-76. 71. Fang, F., S. Phillips, and J. S. Butler. 2005. Rat1p and Rai1p function with the nuclear exosome in the processing and degradation of rRNA precursors. RNA 11:1571-8. 72. Farny, N. G., J. A. Hurt, and P. A. Silver. 2008. Definition of global and transcript- specific mRNA export pathways in metazoans. Genes Dev 22:66-78. 73. Fasken, M. B., and A. H. Corbett. 2009. Mechanisms of nuclear mRNA quality control. RNA Biol 6:237-41. 74. Fasken, M. B., and A. H. Corbett. 2005. Process or perish: quality control in mRNA biogenesis. Nat Struct Mol Biol 12:482-8. 75. Forler, D., T. Kocher, M. Rode, M. Gentzel, E. Izaurralde, and M. Wilm. 2003. An efficient protein complex purification method for functional proteomics in higher eukaryotes. Nat Biotechnol 21:89-92. 76. Franks, T. M., and J. Lykke-Andersen. 2007. TTP and BRF proteins nucleate processing body formation to silence mRNAs with AU-rich elements. Genes Dev 21:719-35. 77. Galy, V., O. Gadal, M. Fromont-Racine, A. Romano, A. Jacquier, and U. Nehrbass. 2004. Nuclear retention of unspliced mRNAs in yeast is mediated by perinuclear Mlp1. Cell 116:63-73. 78. Gatfield, D., and E. Izaurralde. 2004. Nonsense-mediated messenger RNA decay is initiated by endonucleolytic cleavage in Drosophila. Nature 429:575-8. 79. Gatfield, D., L. Unterholzner, F. D. Ciccarelli, P. Bork, and E. Izaurralde. 2003. Nonsense-mediated mRNA decay in Drosophila: at the intersection of the yeast and mammalian pathways. Embo J 22:3960-70. 80. Gavin, A. C., P. Aloy, P. Grandi, R. Krause, M. Boesche, M. Marzioch, C. Rau, L. J. Jensen, S. Bastuck, B. Dumpelfeld, A. Edelmann, M. A. Heurtier, V. Hoffman, C. Hoefert, K. Klein, M. Hudak, A. M. Michon, M. Schelder, M. Schirle, M. Remor, T. Rudi, S. Hooper, A. Bauer, T. Bouwmeester, G. Casari, G. Drewes, G. Neubauer, J. M. Rick, B. Kuster, P. Bork, R. B. Russell, and G. Superti-Furga. 2006. Proteome survey reveals modularity of the yeast cell machinery. Nature 440:631-6. 81. Gavin, A. C., M. Bosche, R. Krause, P. Grandi, M. Marzioch, A. Bauer, J. Schultz, J. M. Rick, A. M. Michon, C. M. Cruciat, M. Remor, C. Hofert, M. Schelder, M. Brajenovic, H. Ruffner, A. Merino, K. Klein, M. Hudak, D. Dickson, T. Rudi, V. Gnau, A. Bauch, S. Bastuck, B. Huhse, C. Leutwein, M. A. Heurtier, R. R. Copley, A. Edelmann, E. Querfurth, V. Rybin, G. Drewes, M. Raida, T. Bouwmeester, P. Bork, B. Seraphin, B. Kuster, G. Neubauer, and G. Superti-Furga. 2002. Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature 415:141-7. 82. Ghaemmaghami, S., W. K. Huh, K. Bower, R. W. Howson, A. Belle, N. Dephoure, E. K. O'Shea, and J. S. Weissman. 2003. Global analysis of protein expression in yeast. Nature 425:737-41. 83. Gherzi, R., K. Y. Lee, P. Briata, D. Wegmuller, C. Moroni, M. Karin, and C. Y. Chen. 2004. A KH domain RNA binding protein, KSRP, promotes ARE-directed mRNA turnover by recruiting the degradation machinery. Mol Cell 14:571-83. 84. Gherzi, R., M. Trabucchi, M. Ponassi, T. Ruggiero, G. Corte, C. Moroni, C. Y. Chen, K. S. Khabar, J. S. Andersen, and P. Briata. 2006. The RNA-binding protein KSRP promotes decay of beta-catenin mRNA and is inactivated by PI3K-AKT signaling. PLoS Biol 5:e5. 85. Gillet, R., and B. Felden. 2001. Emerging views on tmRNA-mediated protein tagging and ribosome rescue. Mol Microbiol 42:879-85.

183

86. Graham, A. C., S. M. Davis, and E. D. Andrulis. 2009. Interdependent nucleocytoplasmic trafficking and interactions of Dis3 with Rrp6, the core exosome, and importin-alpha3. Traffic. 87. Graham, A. C., D. L. Kiss, and E. D. Andrulis. 2009. Core exosome-independent roles for Rrp6 in cell cycle progression. Mol Biol Cell 20:2242-53. 88. Graham, A. C., D. L. Kiss, and E. D. Andrulis. 2006. Differential distribution of exosome subunits at the nuclear lamina and in cytoplasmic foci. Mol Biol Cell 17:1399-409. 89. Grosshans, H., K. Deinert, E. Hurt, and G. Simos. 2001. Biogenesis of the signal recognition particle (SRP) involves import of SRP proteins into the nucleolus, assembly with the SRP-RNA, and Xpo1p-mediated export. J Cell Biol 153:745-62. 90. Grzechnik, P., and J. Kufel. 2008. Polyadenylation linked to transcription termination directs the processing of snoRNA precursors in yeast. Mol Cell 32:247-58. 91. Haile, S., A. M. Estevez, and C. Clayton. 2003. A role for the exosome in the in vivo degradation of unstable mRNAs. Rna 9:1491-501. 92. Herold, A., L. Teixeira, and E. Izaurralde. 2003. Genome-wide analysis of nuclear mRNA export pathways in Drosophila. EMBO J 22:2472-83. 93. Hieronymus, H., M. C. Yu, and P. A. Silver. 2004. Genome-wide mRNA surveillance is coupled to mRNA export. Genes Dev 18:2652-62. 94. Hiley, S. L., T. Babak, and T. R. Hughes. 2005. Global analysis of yeast RNA processing identifies new targets of RNase III and uncovers a link between tRNA 5' end processing and tRNA splicing. Nucleic Acids Res 33:3048-56. 95. Hilleren, P., T. McCarthy, M. Rosbash, R. Parker, and T. H. Jensen. 2001. Quality control of mRNA 3'-end processing is linked to the nuclear exosome. Nature 413:538- 42. 96. Hilleren, P. J., and R. Parker. 2003. Cytoplasmic degradation of splice-defective pre- mRNAs and intermediates. Mol Cell 12:1453-65. 97. Holley, R. W., J. Apgar, G. A. Everett, J. T. Madison, M. Marquisee, S. H. Merrill, J. R. Penswick, and A. Zamir. 1965. Structure of a Ribonucleic Acid. Science 147:1462-5. 98. Holley, R. W., G. A. Everett, J. T. Madison, and A. Zamir. 1965. Nucleotide Sequences in the Yeast Alanine Transfer Ribonucleic Acid. J Biol Chem 240:2122-8. 99. Horwich, M. D., C. Li, C. Matranga, V. Vagin, G. Farley, P. Wang, and P. D. Zamore. 2007. The Drosophila RNA methyltransferase, DmHen1, modifies germline piRNAs and single-stranded siRNAs in RISC. Curr Biol 17:1265-72. 100. Houalla, R., F. Devaux, A. Fatica, J. Kufel, D. Barrass, C. Torchet, and D. Tollervey. 2006. Microarray detection of novel nuclear RNA substrates for the exosome. Yeast 23:439- 54. 101. Houseley, J., K. Kotovic, A. El Hage, and D. Tollervey. 2007. Trf4 targets ncRNAs from telomeric and rDNA spacer regions and functions in rDNA copy number control. EMBO J 26:4996-5006. 102. Houseley, J., J. LaCava, and D. Tollervey. 2006. RNA-quality control by the exosome. Nat Rev Mol Cell Biol 7:529-39. 103. Houseley, J., and D. Tollervey. 2008. The nuclear RNA surveillance machinery: the link between ncRNAs and genome structure in budding yeast? Biochim Biophys Acta 1779:239-46. 104. Houseley, J., and D. Tollervey. 2006. Yeast Trf5p is a nuclear poly(A) polymerase. EMBO Rep 7:205-11.

184

105. Huh, W. K., J. V. Falvo, L. C. Gerke, A. S. Carroll, R. W. Howson, J. S. Weissman, and E. K. O'Shea. 2003. Global analysis of protein localization in budding yeast. Nature 425:686-91. 106. Ibrahim, F., J. Rohr, W. J. Jeong, J. Hesson, and H. Cerutti. 2006. Untemplated oligoadenylation promotes degradation of RISC-cleaved transcripts. Science 314:1893. 107. Jackson, A. L., S. R. Bartz, J. Schelter, S. V. Kobayashi, J. Burchard, M. Mao, B. Li, G. Cavet, and P. S. Linsley. 2003. Expression profiling reveals off-target gene regulation by RNAi. Nat Biotechnol 21:635-7. 108. Jackson, A. L., J. Burchard, J. Schelter, B. N. Chau, M. Cleary, L. Lim, and P. S. Linsley. 2006. Widespread siRNA "off-target" transcript silencing mediated by seed region sequence complementarity. Rna 12:1179-87. 109. Jackson, A. L., and P. S. Linsley. 2004. Noise amidst the silence: off-target effects of siRNAs? Trends Genet 20:521-4. 110. Jacob, F., and J. Monod. 1961. Genetic regulatory mechanisms in the synthesis of proteins. J Mol Biol 3:318-56. 111. Jing, Q., S. Huang, S. Guth, T. Zarubin, A. Motoyama, J. Chen, F. Di Padova, S. C. Lin, H. Gram, and J. Han. 2005. Involvement of microRNA in AU-rich element-mediated mRNA instability. Cell 120:623-34. 112. Jones, T. A., W. Otto, M. Marz, S. R. Eddy, and P. F. Stadler. 2009. A survey of nematode SmY RNAs. RNA Biol 6:5-8. 113. Kadaba, S., A. Krueger, T. Trice, A. M. Krecic, A. G. Hinnebusch, and J. Anderson. 2004. Nuclear surveillance and degradation of hypomodified initiator tRNAMet in S. cerevisiae. Genes Dev 18:1227-40. 114. Kadaba, S., X. Wang, and J. T. Anderson. 2006. Nuclear RNA surveillance in Saccharomyces cerevisiae: Trf4p-dependent polyadenylation of nascent hypomethylated tRNA and an aberrant form of 5S rRNA. RNA 12:508-21. 115. Kadowaki, T., S. Chen, M. Hitomi, E. Jacobs, C. Kumagai, S. Liang, R. Schneiter, D. Singleton, J. Wisniewska, and A. M. Tartakoff. 1994. Isolation and characterization of Saccharomyces cerevisiae mRNA transport-defective (mtr) mutants. J Cell Biol 126:649-59. 116. Kadowaki, T., R. Schneiter, M. Hitomi, and A. M. Tartakoff. 1995. Mutations in nucleolar proteins lead to nucleolar accumulation of polyA+ RNA in Saccharomyces cerevisiae. Mol Biol Cell 6:1103-10. 117. Kapranov, P., J. Cheng, S. Dike, D. A. Nix, R. Duttagupta, A. T. Willingham, P. F. Stadler, J. Hertel, J. Hackermuller, I. L. Hofacker, I. Bell, E. Cheung, J. Drenkow, E. Dumais, S. Patel, G. Helt, M. Ganesh, S. Ghosh, A. Piccolboni, V. Sementchenko, H. Tammana, and T. R. Gingeras. 2007. RNA maps reveal new RNA classes and a possible function for pervasive transcription. Science 316:1484-8. 118. Kawaji, H., T. Kasukawa, S. Fukuda, S. Katayama, C. Kai, J. Kawai, P. Carninci, and Y. Hayashizaki. 2006. CAGE Basic/Analysis Databases: the CAGE resource for comprehensive promoter analysis. Nucleic Acids Res 34:D632-6. 119. Kawaji, H., J. Severin, M. Lizio, A. Waterhouse, S. Katayama, K. M. Irvine, D. A. Hume, A. R. Forrest, H. Suzuki, P. Carninci, Y. Hayashizaki, and C. O. Daub. 2009. The FANTOM web resource: from mammalian transcriptional landscape to its dynamic regulation. Genome Biol 10:R40. 120. Kent, T., Y. R. Lapik, and D. G. Pestov. 2009. The 5' external transcribed spacer in mouse ribosomal RNA contains two cleavage sites. RNA 15:14-20.

185

121. Kim, M., L. Vasiljeva, O. J. Rando, A. Zhelkovsky, C. Moore, and S. Buratowski. 2006. Distinct pathways for snoRNA and mRNA termination. Mol Cell 24:723-34. 122. Kinoshita, N., M. Goebl, and M. Yanagida. 1991. The fission yeast dis3+ gene encodes a 110-kDa essential protein implicated in mitotic control. Mol Cell Biol 11:5839-47. 123. Kiss, T. 2001. Small nucleolar RNA-guided post-transcriptional modification of cellular RNAs. Embo J 20:3617-22. 124. Koonin, E. V., Y. I. Wolf, and L. Aravind. 2001. Prediction of the archaeal exosome and its connections with the proteasome and the translation and transcription machineries by a comparative-genomic approach. Genome Res 11:240-52. 125. Krogan, N. J., G. Cagney, H. Yu, G. Zhong, X. Guo, A. Ignatchenko, J. Li, S. Pu, N. Datta, A. P. Tikuisis, T. Punna, J. M. Peregrin-Alvarez, M. Shales, X. Zhang, M. Davey, M. D. Robinson, A. Paccanaro, J. E. Bray, A. Sheung, B. Beattie, D. P. Richards, V. Canadien, A. Lalev, F. Mena, P. Wong, A. Starostine, M. M. Canete, J. Vlasblom, S. Wu, C. Orsi, S. R. Collins, S. Chandran, R. Haw, J. J. Rilstone, K. Gandi, N. J. Thompson, G. Musso, P. St Onge, S. Ghanny, M. H. Lam, G. Butland, A. M. Altaf-Ul, S. Kanaya, A. Shilatifard, E. O'Shea, J. S. Weissman, C. J. Ingles, T. R. Hughes, J. Parkinson, M. Gerstein, S. J. Wodak, A. Emili, and J. F. Greenblatt. 2006. Global landscape of protein complexes in the yeast Saccharomyces cerevisiae. Nature 440:637-43. 126. Krogan, N. J., W. T. Peng, G. Cagney, M. D. Robinson, R. Haw, G. Zhong, X. Guo, X. Zhang, V. Canadien, D. P. Richards, B. K. Beattie, A. Lalev, W. Zhang, A. P. Davierwala, S. Mnaimneh, A. Starostine, A. P. Tikuisis, J. Grigull, N. Datta, J. E. Bray, T. R. Hughes, A. Emili, and J. F. Greenblatt. 2004. High-definition macromolecular composition of yeast RNA-processing complexes. Mol Cell 13:225-39. 127. Kuai, L., B. Das, and F. Sherman. 2005. A nuclear degradation pathway controls the abundance of normal mRNAs in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 102:13962-7. 128. Kuai, L., F. Fang, J. S. Butler, and F. Sherman. 2004. Polyadenylation of rRNA in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 101:8581-6. 129. LaCava, J., J. Houseley, C. Saveanu, E. Petfalski, E. Thompson, A. Jacquier, and D. Tollervey. 2005. RNA degradation by the exosome is promoted by a nuclear polyadenylation complex. Cell 121:713-24. 130. Lange, H., S. Holec, V. Cognat, L. Pieuchot, M. Le Ret, J. Canaday, and D. Gagliardi. 2008. Degradation of a polyadenylated rRNA maturation by-product involves one of the three RRP6-like proteins in Arabidopsis thaliana. Mol Cell Biol 28:3038-44. 131. LaRiviere, F. J., S. E. Cole, D. J. Ferullo, and M. J. Moore. 2006. A late-acting quality control process for mature eukaryotic rRNAs. Mol Cell 24:619-26. 132. Lebreton, A., R. Tomecki, A. Dziembowski, and B. Seraphin. 2008. Endonucleolytic RNA cleavage by a eukaryotic exosome. Nature 456:993-6. 133. Lejeune, F., X. Li, and L. E. Maquat. 2003. Nonsense-mediated mRNA decay in mammalian cells involves decapping, deadenylating, and exonucleolytic activities. Mol Cell 12:675-87. 134. Libri, D., K. Dower, J. Boulay, R. Thomsen, M. Rosbash, and T. H. Jensen. 2002. Interactions between mRNA export commitment, 3'-end quality control, and nuclear degradation. Mol Cell Biol 22:8254-66. 135. Lin, H., and H. Yin. 2008. A novel epigenetic mechanism in Drosophila somatic cells mediated by Piwi and piRNAs. Cold Spring Harb Symp Quant Biol 73:273-81.

186

136. Lin, W. J., A. Duffy, and C. Y. Chen. 2007. Localization of AU-rich element-containing mRNA in cytoplasmic granules containing exosome subunits. J Biol Chem 282:19958- 68. 137. Liu, Q., J. C. Greimann, and C. D. Lima. 2006. Reconstitution, activities, and structure of the eukaryotic RNA exosome. Cell 127:1223-37. 138. Lorentzen, E., J. Basquin, and E. Conti. 2008. Structural organization of the RNA- degrading exosome. Curr Opin Struct Biol 18:709-13. 139. Lorentzen, E., and E. Conti. 2008. Expression, reconstitution, and structure of an archaeal RNA degrading exosome. Methods Enzymol 447:417-35. 140. Lorentzen, E., and E. Conti. 2005. Structural basis of 3' end RNA recognition and exoribonucleolytic cleavage by an exosome RNase PH core. Mol Cell 20:473-81. 141. Lorentzen, E., A. Dziembowski, D. Lindner, B. Seraphin, and E. Conti. 2007. RNA channelling by the archaeal exosome. EMBO Rep 8:470-6. 142. Lorentzen, E., P. Walter, S. Fribourg, E. Evguenieva-Hackenberg, G. Klug, and E. Conti. 2005. The archaeal exosome core is a hexameric ring structure with three catalytic subunits. Nat Struct Mol Biol 12:575-81. 143. Lykke-Andersen, J., and E. Wagner. 2005. Recruitment and activation of mRNA decay enzymes by two ARE-mediated decay activation domains in the proteins TTP and BRF- 1. Genes Dev 19:351-61. 144. Madrona, A. Y., and D. K. Wilson. 2004. The structure of Ski8p, a protein regulating mRNA degradation: Implications for WD protein structure. Protein Sci 13:1557-65. 145. Mamolen, M., and E. D. Andrulis. 2009. Characterization of the Drosophilamelanogaster Dis3 ribonuclease. Biochem Biophys Res Commun. 146. McManus, M. T., and P. A. Sharp. 2002. Gene silencing in mammals by small interfering RNAs. Nat Rev Genet 3:737-47. 147. McPheeters, D. S., N. Cremona, S. Sunder, H. M. Chen, N. Averbeck, J. Leatherwood, and J. A. Wise. 2009. A complex gene regulatory mechanism that operates at the nexus of multiple RNA processing decisions. Nat Struct Mol Biol. 148. Meaux, S., and A. Van Hoof. 2006. Yeast transcripts cleaved by an internal ribozyme provide new insight into the role of the cap and poly(A) tail in translation and mRNA decay. RNA 12:1323-37. 149. Midtgaard, S. F., J. Assenholt, A. T. Jonstrup, L. B. Van, T. H. Jensen, and D. E. Brodersen. 2006. Structure of the nuclear exosome component Rrp6p reveals an interplay between the active site and the HRDC domain. Proc Natl Acad Sci U S A 103:11898-903. 150. Milligan, L., L. Decourty, C. Saveanu, J. Rappsilber, H. Ceulemans, A. Jacquier, and D. Tollervey. 2008. A yeast exosome cofactor, Mpp6, functions in RNA surveillance and in the degradation of noncoding RNA transcripts. Mol Cell Biol 28:5446-57. 151. Milligan, L., C. Torchet, C. Allmang, T. Shipman, and D. Tollervey. 2005. A nuclear surveillance pathway for mRNAs with defective polyadenylation. Mol Cell Biol 25:9996-10004. 152. Mitchell, P., E. Petfalski, R. Houalla, A. Podtelejnikov, M. Mann, and D. Tollervey. 2003. Rrp47p is an exosome-associated protein required for the 3' processing of stable RNAs. Mol Cell Biol 23:6982-92. 153. Mitchell, P., E. Petfalski, A. Shevchenko, M. Mann, and D. Tollervey. 1997. The exosome: a conserved eukaryotic RNA processing complex containing multiple 3'-->5' exoribonucleases. Cell 91:457-66.

187

154. Mitchell, P., E. Petfalski, and D. Tollervey. 1996. The 3' end of yeast 5.8S rRNA is generated by an exonuclease processing mechanism. Genes Dev 10:502-13. 155. Mitchell, P., and D. Tollervey. 2000. Musing on the structural organization of the exosome complex. Nat Struct Biol 7:843-6. 156. Mitchell, P., and D. Tollervey. 2003. An NMD pathway in yeast involving accelerated deadenylation and exosome-mediated 3'-->5' degradation. Mol Cell 11:1405-13. 157. Morimoto, R. I., K. D. Sarge, and K. Abravaya. 1992. Transcriptional regulation of heat shock genes. A paradigm for inducible genomic responses. J Biol Chem 267:21987-90. 158. Mukherjee, D., M. Gao, J. P. O'Connor, R. Raijmakers, G. Pruijn, C. S. Lutz, and J. Wilusz. 2002. The mammalian exosome mediates the efficient degradation of mRNAs that contain AU-rich elements. Embo J 21:165-74. 159. Nakai, A., Y. Kawazoe, M. Tanabe, K. Nagata, and R. I. Morimoto. 1995. The DNA- binding properties of two heat shock factors, HSF1 and HSF3, are induced in the avian erythroblast cell line HD6. Mol Cell Biol 15:5268-78. 160. Neil, H., C. Malabat, Y. d'Aubenton-Carafa, Z. Xu, L. M. Steinmetz, and A. Jacquier. 2009. Widespread bidirectional promoters are the major source of cryptic transcripts in yeast. Nature. 161. Neu-Yilik, G., N. H. Gehring, R. Thermann, U. Frede, M. W. Hentze, and A. E. Kulozik. 2001. Splicing and 3' end formation in the definition of nonsense-mediated decay- competent human beta-globin mRNPs. Embo J 20:532-40. 162. Neu-Yilik, G., and A. E. Kulozik. 2008. NMD: multitasking between mRNA surveillance and modulation of gene expression. Adv Genet 62:185-243. 163. O'Connell, P. O., and M. Rosbash. 1984. Sequence, structure, and codon preference of the Drosophila ribosomal protein 49 gene. Nucleic Acids Res 12:5495-513. 164. Ohkura, H., Y. Adachi, N. Kinoshita, O. Niwa, T. Toda, and M. Yanagida. 1988. Cold- sensitive and caffeine-supersensitive mutants of the Schizosaccharomyces pombe dis genes implicated in sister chromatid separation during mitosis. Embo J 7:1465-73. 165. Orban, T. I., and E. Izaurralde. 2005. Decay of mRNAs targeted by RISC requires XRN1, the Ski complex, and the exosome. Rna 11:459-69. 166. Palade, G. E. 1955. A small particulate component of the cytoplasm. J Biophys Biochem Cytol 1:59-68. 167. Pannucci, J. A., E. S. Haas, T. A. Hall, J. K. Harris, and J. W. Brown. 1999. RNase P RNAs from some Archaea are catalytically active. Proc Natl Acad Sci U S A 96:7803-8. 168. Parker, R., and H. Song. 2004. The enzymes and control of eukaryotic mRNA turnover. Nat Struct Mol Biol 11:121-7. 169. Pasquinelli, A. E., and G. Ruvkun. 2002. Control of developmental timing by micrornas and their targets. Annu Rev Cell Dev Biol 18:495-513. 170. Passos, D. O., M. K. Doma, C. J. Shoemaker, D. Muhlrad, R. Green, J. Weissman, J. Hollien, and R. Parker. 2009. Analysis of Dom34 and its function in no-go decay. Mol Biol Cell 20:3025-32. 171. Perreault, J., J. P. Perreault, and G. Boire. 2007. Ro-associated Y RNAs in metazoans: evolution and diversification. Mol Biol Evol 24:1678-89. 172. Perumal, K., and R. Reddy. 2002. The 3' end formation in small RNAs. Gene Expr 10:59- 78. 173. Petersen, R., and S. Lindquist. 1988. The Drosophila hsp70 message is rapidly degraded at normal temperatures and stabilized by heat shock. Gene 72:161-8. 174. Petersen, R. B., and S. Lindquist. 1989. Regulation of HSP70 synthesis by messenger RNA degradation. Cell Regul 1:135-49.

188

175. Phillips, S., and J. S. Butler. 2003. Contribution of domain structure to the RNA 3' end processing and degradation functions of the nuclear exosome subunit Rrp6p. RNA 9:1098-107. 176. Pirkkala, L., P. Nykanen, and L. Sistonen. 2001. Roles of the heat shock transcription factors in regulation of the heat shock response and beyond. FASEB J 15:1118-31. 177. Pleiss, J. A., G. B. Whitworth, M. Bergkessel, and C. Guthrie. 2007. Transcript specificity in yeast pre-mRNA splicing revealed by mutations in core spliceosomal components. PLoS Biol 5:e90. 178. Preker, P., J. Nielsen, S. Kammler, S. Lykke-Andersen, M. S. Christensen, C. K. Mapendano, M. H. Schierup, and T. H. Jensen. 2008. RNA exosome depletion reveals transcription upstream of active human promoters. Science 322:1851-4. 179. Prokisch, H., C. Scharfe, D. G. Camp, 2nd, W. Xiao, L. David, C. Andreoli, M. E. Monroe, R. J. Moore, M. A. Gritsenko, C. Kozany, K. K. Hixson, H. M. Mottaz, H. Zischka, M. Ueffing, Z. S. Herman, R. W. Davis, T. Meitinger, P. J. Oefner, R. D. Smith, and L. M. Steinmetz. 2004. Integrative analysis of the mitochondrial proteome in yeast. PLoS Biol 2:e160. 180. Prouteau, M., M. C. Daugeron, and B. Seraphin. 2008. Regulation of ARE transcript 3' end processing by the yeast Cth2 mRNA decay factor. Embo J 27:2966-76. 181. Raijmakers, R., W. V. Egberts, W. J. van Venrooij, and G. J. Pruijn. 2003. The association of the human PM/Scl-75 autoantigen with the exosome is dependent on a newly identified N terminus. J Biol Chem 278:30698-704. 182. Raijmakers, R., W. V. Egberts, W. J. van Venrooij, and G. J. Pruijn. 2002. Protein- protein interactions between human exosome components support the assembly of RNase PH-type subunits into a six-membered PNPase-like ring. J Mol Biol 323:653-63. 183. Raijmakers, R., Y. E. Noordman, W. J. van Venrooij, and G. J. Pruijn. 2002. Protein- protein interactions of hCsl4p with other human exosome subunits. J Mol Biol 315:809-18. 184. Rebbapragada, I., and J. Lykke-Andersen. 2009. Execution of nonsense-mediated mRNA decay: what defines a substrate? Curr Opin Cell Biol 21:394-402. 185. Rehwinkel, J., I. Letunic, J. Raes, P. Bork, and E. Izaurralde. 2005. Nonsense-mediated mRNA decay factors act in concert to regulate common mRNA targets. RNA 11:1530- 44. 186. Rehwinkel, J., P. Natalin, A. Stark, J. Brennecke, S. M. Cohen, and E. Izaurralde. 2006. Genome-wide analysis of mRNAs regulated by Drosha and Argonaute proteins in Drosophila melanogaster. Mol Cell Biol 26:2965-75. 187. Rehwinkel, J., J. Raes, and E. Izaurralde. 2006. Nonsense-mediated mRNA decay: Target genes and functional diversification of effectors. Trends Biochem Sci 31:639-46. 188. Reinhart, B. J., F. J. Slack, M. Basson, A. E. Pasquinelli, J. C. Bettinger, A. E. Rougvie, H. R. Horvitz, and G. Ruvkun. 2000. The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans. Nature 403:901-6. 189. San Paolo, S., S. Vanacova, L. Schenk, T. Scherrer, D. Blank, W. Keller, and A. P. Gerber. 2009. Distinct roles of non-canonical poly(A) polymerases in RNA metabolism. PLoS Genet 5:e1000555. 190. Saunders, A., J. Werner, E. D. Andrulis, T. Nakayama, S. Hirose, D. Reinberg, and J. T. Lis. 2003. Tracking FACT and the RNA polymerase II elongation complex through chromatin in vivo. Science 301:1094-6. 191. Schaeffer, D., S. Meaux, A. Clark, and A. van Hoof. 2008. Determining in vivo activity of the yeast cytoplasmic exosome. Methods Enzymol 448:227-39.

189

192. Schaeffer, D., B. Tsanova, A. Barbas, F. P. Reis, E. G. Dastidar, M. Sanchez-Rotunno, C. M. Arraiano, and A. van Hoof. 2009. The exosome contains domains with specific endoribonuclease, exoribonuclease and cytoplasmic mRNA decay activities. Nat Struct Mol Biol 16:56-62. 193. Schilders, G., R. Raijmakers, J. M. Raats, and G. J. Pruijn. 2005. MPP6 is an exosome- associated RNA-binding protein involved in 5.8S rRNA maturation. Nucleic Acids Res 33:6795-804. 194. Schilders, G., E. van Dijk, and G. J. Pruijn. 2007. C1D and hMtr4p associate with the human exosome subunit PM/Scl-100 and are involved in pre-rRNA processing. Nucleic Acids Res 35:2564-72. 195. Schmid, M., and T. H. Jensen. 2008. The exosome: a multipurpose RNA-decay machine. Trends Biochem Sci 33:501-10. 196. Schneider, C., J. T. Anderson, and D. Tollervey. 2007. The exosome subunit Rrp44 plays a direct role in RNA substrate recognition. Mol Cell 27:324-31. 197. Schneider, C., E. Leung, J. Brown, and D. Tollervey. 2009. The N-terminal PIN domain of the exosome subunit Rrp44 harbors endonuclease activity and tethers Rrp44 to the yeast core exosome. Nucleic Acids Res. 198. Schwede, A., L. Ellis, J. Luther, M. Carrington, G. Stoecklin, and C. Clayton. 2008. A role for Caf1 in mRNA deadenylation and decay in trypanosomes and human cells. Nucleic Acids Res 36:3374-88. 199. Schwede, A., T. Manful, B. A. Jha, C. Helbig, N. Bercovich, M. Stewart, and C. Clayton. 2009. The role of deadenylation in the degradation of unstable mRNAs in trypanosomes. Nucleic Acids Res. 200. Sickmann, A., J. Reinders, Y. Wagner, C. Joppich, R. Zahedi, H. E. Meyer, B. Schonfisch, I. Perschil, A. Chacinska, B. Guiard, P. Rehling, N. Pfanner, and C. Meisinger. 2003. The proteome of Saccharomyces cerevisiae mitochondria. Proc Natl Acad Sci U S A 100:13207-12. 201. Stead, J. A., J. L. Costello, M. J. Livingstone, and P. Mitchell. 2007. The PMC2NT domain of the catalytic exosome subunit Rrp6p provides the interface for binding with its cofactor Rrp47p, a nucleic acid-binding protein. Nucleic Acids Res 35:5556-67. 202. Stoecklin, G., T. Mayo, and P. Anderson. 2006. ARE-mRNA degradation requires the 5'- 3' decay pathway. EMBO Rep 7:72-7. 203. Suzuki, N., E. Noguchi, N. Nakashima, M. Oki, T. Ohba, A. Tartakoff, M. Ohishi, and T. Nishimoto. 2001. The Saccharomyces cerevisiae small GTPase, Gsp1p/Ran, is involved in 3' processing of 7S-to-5.8S rRNA and in degradation of the excised 5'-A0 fragment of 35S pre-rRNA, both of which are carried out by the exosome. Genetics 158:613-25. 204. Swisher, K. D., and R. Parker. 2009. Related mechanisms for mRNA and rRNA quality control. Mol Cell 34:401-2. 205. Synowsky, S. A., and A. J. Heck. 2008. The yeast Ski complex is a hetero-tetramer. Protein Sci 17:119-25. 206. Synowsky, S. A., R. H. van den Heuvel, S. Mohammed, P. W. Pijnappel, and A. J. Heck. 2006. Probing genuine strong interactions and post-translational modifications in the heterogeneous yeast exosome protein complex. Mol Cell Proteomics. 207. Synowsky, S. A., M. van Wijk, R. Raijmakers, and A. J. Heck. 2009. Comparative multiplexed mass spectrometric analyses of endogenously expressed yeast nuclear and cytoplasmic exosomes. J Mol Biol 385:1300-13. 208. Szostak, J. W. 1989. Telomerase: the beginning of the ends. Nature 337:303-4.

190

209. Takahashi, S., Y. Araki, T. Sakuno, and T. Katada. 2003. Interaction between Ski7p and Upf1p is required for nonsense-mediated 3'-to-5' mRNA decay in yeast. Embo J 22:3951-9. 210. Thiebaut, M., E. Kisseleva-Romanova, M. Rougemaille, J. Boulay, and D. Libri. 2006. Transcription termination and nuclear degradation of cryptic unstable transcripts: a role for the nrd1-nab3 pathway in genome surveillance. Mol Cell 23:853-64. 211. Thomsen, R., D. Libri, J. Boulay, M. Rosbash, and T. H. Jensen. 2003. Localization of nuclear retained mRNAs in Saccharomyces cerevisiae. Rna 9:1049-57. 212. Thore, S., C. Mayer, C. Sauter, S. Weeks, and D. Suck. 2003. Crystal structures of the Pyrococcus abyssi Sm core and its complex with RNA. Common features of RNA binding in archaea and eukarya. J Biol Chem 278:1239-47. 213. Tollervey, D., and J. F. Caceres. 2000. RNA processing marches on. Cell 103:703-9. 214. Torchet, C., C. Bousquet-Antonelli, L. Milligan, E. Thompson, J. Kufel, and D. Tollervey. 2002. Processing of 3'-extended read-through transcripts by the exosome can generate functional mRNAs. Mol Cell 9:1285-96. 215. van Dijk, E. L., G. Schilders, and G. J. Pruijn. 2007. Human cell growth requires a functional cytoplasmic exosome, which is involved in various mRNA decay pathways. RNA 13:1027-35. 216. van Hoof, A., P. A. Frischmeyer, H. C. Dietz, and R. Parker. 2002. Exosome-mediated recognition and degradation of mRNAs lacking a termination codon. Science 295:2262- 4. 217. van Hoof, A., P. Lennertz, and R. Parker. 2000. Three conserved members of the RNase D family have unique and overlapping functions in the processing of 5S, 5.8S, U4, U5, RNase MRP and RNase P RNAs in yeast. EMBO J 19:1357-65. 218. van Hoof, A., P. Lennertz, and R. Parker. 2000. Yeast exosome mutants accumulate 3'- extended polyadenylated forms of U4 small nuclear RNA and small nucleolar RNAs. Mol Cell Biol 20:441-52. 219. van Hoof, A., and R. Parker. 1999. The exosome: a proteasome for RNA? Cell 99:347- 50. 220. van Hoof, A., R. R. Staples, R. E. Baker, and R. Parker. 2000. Function of the ski4p (Csl4p) and Ski7p proteins in 3'-to-5' degradation of mRNA. Mol Cell Biol 20:8230-43. 221. Vanacova, S., and R. Stefl. 2007. The exosome and RNA quality control in the nucleus. EMBO Rep 8:651-7. 222. Vanacova, S., J. Wolf, G. Martin, D. Blank, S. Dettwiler, A. Friedlein, H. Langen, G. Keith, and W. Keller. 2005. A new yeast poly(A) polymerase complex involved in RNA quality control. PLoS Biol 3:e189. 223. Vasiljeva, L., and S. Buratowski. 2006. Nrd1 interacts with the nuclear exosome for 3' processing of RNA polymerase II transcripts. Mol Cell 21:239-48. 224. Vasiljeva, L., M. Kim, N. Terzi, L. M. Soares, and S. Buratowski. 2008. Transcription termination and RNA degradation contribute to silencing of RNA polymerase II transcription within heterochromatin. Mol Cell 29:313-23. 225. Vasudevan, S., and S. W. Peltz. 2003. Nuclear mRNA surveillance. Curr Opin Cell Biol 15:332-7. 226. Vasudevan, S., and S. W. Peltz. 2001. Regulated ARE-mediated mRNA decay in Saccharomyces cerevisiae. Mol Cell 7:1191-200. 227. Vaucheret, H. 2006. Post-transcriptional small RNA pathways in plants: mechanisms and regulations. Genes Dev 20:759-71.

191

228. Vaucheret, H., C. Beclin, and M. Fagard. 2001. Post-transcriptional gene silencing in plants. J Cell Sci 114:3083-91. 229. Velculescu, V. E., B. Vogelstein, and K. W. Kinzler. 2000. Analysing uncharted transcriptomes with SAGE. Trends Genet 16:423-5. 230. Velculescu, V. E., L. Zhang, W. Zhou, J. Vogelstein, M. A. Basrai, D. E. Bassett, Jr., P. Hieter, B. Vogelstein, and K. W. Kinzler. 1997. Characterization of the yeast transcriptome. Cell 88:243-51. 231. Vinciguerra, P., N. Iglesias, J. Camblong, D. Zenklusen, and F. Stutz. 2005. Perinuclear Mlp proteins downregulate gene expression in response to a defect in mRNA export. EMBO J 24:813-23. 232. Vinciguerra, P., and F. Stutz. 2004. mRNA export: an assembly line from genes to nuclear pores. Curr Opin Cell Biol 16:285-92. 233. Wang, H. W., J. Wang, F. Ding, K. Callahan, M. A. Bratkowski, J. S. Butler, E. Nogales, and A. Ke. 2007. Architecture of the yeast Rrp44 exosome complex suggests routes of RNA recruitment for 3' end processing. Proc Natl Acad Sci U S A 104:16844-9. 234. Wang, L., M. S. Lewis, and A. W. Johnson. 2005. Domain interactions within the Ski2/3/8 complex and between the Ski complex and Ski7p. RNA 11:1291-302. 235. Wang, S. W., A. L. Stevenson, S. E. Kearsey, S. Watt, and J. Bahler. 2008. Global role for polyadenylation-assisted nuclear RNA degradation in posttranscriptional gene silencing. Mol Cell Biol 28:656-65. 236. Wery, M., S. Ruidant, S. Schillewaert, N. Lepore, and D. L. Lafontaine. 2009. The nuclear poly(A) polymerase and Exosome cofactor Trf5 is recruited cotranscriptionally to nucleolar surveillance. RNA. 237. West, S., N. Gromak, C. J. Norbury, and N. J. Proudfoot. 2006. Adenylation and exosome-mediated degradation of cotranscriptionally cleaved pre-messenger RNA in human cells. Mol Cell 21:437-43. 238. Wilusz, J. E., H. Sunwoo, and D. L. Spector. 2009. Long noncoding RNAs: functional surprises from the RNA world. Genes Dev 23:1494-504. 239. Woodhams, M. D., P. F. Stadler, D. Penny, and L. J. Collins. 2007. RNase MRP and the RNA processing cascade in the eukaryotic ancestor. BMC Evol Biol 7 Suppl 1:S13. 240. Wyers, F., M. Rougemaille, G. Badis, J. C. Rousselle, M. E. Dufour, J. Boulay, B. Regnault, F. Devaux, A. Namane, B. Seraphin, D. Libri, and A. Jacquier. 2005. Cryptic pol II transcripts are degraded by a nuclear quality control pathway involving a new poly(A) polymerase. Cell 121:725-37. 241. Xu, Z., W. Wei, J. Gagneur, F. Perocchi, S. Clauder-Munster, J. Camblong, E. Guffanti, F. Stutz, W. Huber, and L. M. Steinmetz. 2009. Bidirectional promoters generate pervasive transcription in yeast. Nature 457:1033-7. 242. Yamasaki, S., G. Stoecklin, N. Kedersha, M. Simarro, and P. Anderson. 2007. T-cell intracellular antigen-1 (TIA-1)-induced translational silencing promotes the decay of selected mRNAs. J Biol Chem 282:30070-7. 243. Yehudai-Resheff, S., V. Portnoy, S. Yogev, N. Adir, and G. Schuster. 2003. Domain analysis of the chloroplast polynucleotide phosphorylase reveals discrete functions in RNA degradation, polyadenylation, and sequence homology with exosome proteins. Plant Cell 15:2003-19. 244. Yin, H., and H. Lin. 2007. An epigenetic activation role of Piwi and a Piwi-associated piRNA in Drosophila melanogaster. Nature 450:304-8. 245. Zanchin, N. I., and D. S. Goldfarb. 1999. The exosome subunit Rrp43p is required for the efficient maturation of 5.8S, 18S and 25S rRNA. Nucleic Acids Res 27:1283-8.

192

246. Zanchin, N. I., and D. S. Goldfarb. 1999. Nip7p interacts with Nop8p, an essential nucleolar protein required for 60S ribosome biogenesis, and the exosome subunit Rrp43p. Mol Cell Biol 19:1518-25. 247. Zenklusen, D., P. Vinciguerra, J. C. Wyss, and F. Stutz. 2002. Stable mRNP formation and export require cotranscriptional recruitment of the mRNA export factors Yra1p and Sub2p by Hpr1p. Mol Cell Biol 22:8241-53. 248. Zhu, J., H. Wang, J. M. Bishop, and E. H. Blackburn. 1999. Telomerase extends the lifespan of virus-transformed human cells without net telomere lengthening. Proc Natl Acad Sci U S A 96:3723-8. 249. Zilberman, D., and S. Henikoff. 2005. Epigenetic inheritance in Arabidopsis: selective silence. Curr Opin Genet Dev 15:557-62. 250. Zilberman, D., and S. Henikoff. 2004. Silencing of transposons in plant genomes: kick them when they're down. Genome Biol 5:249. 251. Zuo, Y., and M. P. Deutscher. 2001. Exoribonuclease superfamilies: structural analysis and phylogenetic distribution. Nucleic Acids Res 29:1017-26.

193