5 subunit of the Ccr4-Not complex: A central regulator of gene expression that integrates signals between the cytoplasm and the nucleus in eukaryotic cells

Cellular Signalling 25 (2013) 743–751

Contents lists available at SciVerse ScienceDirect

Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig

Review

The Not3/5 subunit of the Ccr4-Not complex: A central regulator of gene expression that integrates signals between the cytoplasm and the nucleus in eukaryotic cells Martine A. Collart a, c,⁎, Olesya O. Panasenko a, c, Sergey I. Nikolaev b, c a b c

Dpt of Microbiology and Molecular Medicine, Faculty of Medicine, University of Geneva, 1 Rue Michel Servet, 1211 Genève 4, Switzerland Dpt of Genetic Medicine and Development, Faculty of Medicine, University of Geneva, 1 Rue Michel Servet, 1211 Genève 4, Switzerland Institute of Genetics and Genomics Geneva, 1 Rue Michel Servet, 1211 Genève 4, Switzerland

a r t i c l e

i n f o

Article history: Received 11 December 2012 Accepted 21 December 2012 Available online 29 December 2012 Keywords: Ccr4-Not complex CNot3 Not5 Transcription mRNA degradation

a b s t r a c t The Ccr4-Not complex is a conserved multi-subunit complex in eukaryotes that carries 2 enzymatic activities: ubiquitination mediated by the Not4 RING E3 ligase and deadenylation mediated by the Ccr4 and Caf1 orthologs. This complex has been implicated in all aspects of the mRNA life cycle, from synthesis of mRNAs in the nucleus to their degradation in the cytoplasm. More recently the complex has also been implicated in many aspects of the life cycle of proteins, from quality control during synthesis of peptides, to assembly of protein complexes and protein degradation. Consistently, the Ccr4-Not complex is found both in the nucleus, where it is connected to transcribing ORFs, and in the cytoplasm, where it was revealed to be both associated with translating ribosomes and in RNA processing bodies. This functional and physical presence of the Ccr4-Not complex at all stages of gene expression raises the question of its fundamental role. This review will summarize recent evidence designing the Not3/5 module of the Ccr4-Not complex as a functional module involved in coordination of the regulation of gene expression between the nucleus and the cytoplasm. © 2012 Elsevier Inc. All rights reserved.

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Description of the Ccr4-Not complex . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1. Ccr4–Caf1 module . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2. Not4 module . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3. Not2–3/5 module . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. The Ccr4-Not complex: a platform that coordinates functions of its modules . . . 2.4.1. Role in mRNA degradation . . . . . . . . . . . . . . . . . . . . . . 2.4.2. Role in nuclear transcription . . . . . . . . . . . . . . . . . . . . . 2.4.3. Role in protein synthesis and degradation . . . . . . . . . . . . . . . 3. The CNot3 funcional module is essential for diverse cellular functions . . . . . . . . . 3.1. CNot3 as a module for cellular self-renewal . . . . . . . . . . . . . . . . . . 3.2. CNot3 in heart function . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. CNot3 as sensor for nutrient availability . . . . . . . . . . . . . . . . . . . . 4. CNot3 marks mRNAs at birth, for regulation from synthesis to translation and degradation Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . .

744 744 744 744 744 744 744 745 745 745 745 746 746 746 747 748 748 749 749

⁎ Corresponding author at: Dpt of Microbiology and Molecular Medicine, Faculty of Medicine, University of Geneva, 1 Rue Michel Servet, 1211 Genève 4, Switzerland. Tel.: +41 22 3795476; fax: +41 22 3795702. E-mail address: [email protected] (M.A. Collart). 0898-6568/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cellsig.2012.12.018

744

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751

1. Introduction Eukaryotic cells must continuously evolve to adapt to their environment. For unicellular organisms, changes in the environment are mostly related to nutrient levels, and the extent of different stresses, including temperature changes, heavy metals, oxidative stress and many more. But the role of the environment is much more complex for multicellular organisms. Indeed, a cell must additionally be able to integrate signals originating from the other cells of the organism, even from cells at distant sites. These signals arriving at the cell surface are of diverse types that include proteins, lipids and chemicals, and they will interact with the cell via receptors or transporters at the cell surface. Consequently, signaling cascades will be activated leading to a multitude of different modifications and adaptations within cells. Each signal leads to its specific characteristic cascade of events via its specific combination of second messengers. On one hand immediate responses are induced, such as fusion of secretory granules with the cell surface [1]. On the other hand, expression of specific genes is altered, either transiently to respond to a transient modification of the environment, or longer-lastingly to modify gene expression programs. Modification of gene expression in response to a signal involves not only activation of new genes, but also switching off the genetic program that was in place before. This shut down involves repression of the transcription of genes, and degradation of proteins and mRNAs that are no longer needed [2–4]. Transcription factors that coordinate activation of specific families of genes have been characterized and the regulation of their activity in response to signaling pathways is in many cases well defined [5]. Degradation of mRNAs and proteins has also been studied in response to signaling cascades [6,7]. However how the transcription responses are connected with the cytoplasmic events is much less understood. One recent study revealed that mRNA synthesis and degradation mutually feedback onto each other [8]. Another uncovered opposite evolutionary changes in transcription and mRNA degradation rates that were mechanistically coupled and generated by the same individual mutations [9]. A large fraction of these mutations mapped to the genes encoding the Rpb4/7 sub-complex of RNA polymerase II or the genes encoding the Ccr4, Caf1 and Not5 subunits of the Ccr4-Not complex [9]. This is consistent with studies implicating Rbp4/7 as well as the Ccr4-Not complex in both the transcription and mRNA degradation processes [10–13]. The Ccr4-Not complex is a conserved multi-subunit complex that regulates gene expression at multiple levels (for reviews see [14–16]). Besides its role in mRNA synthesis and degradation it has also been connected to protein quality control [17–20]. It is the ideal candidate for coordination of gene expression at all levels in cells, in response to modification of the environment. Many studies first in yeast, and more recently in higher eukaryotes, provide evidence for such a fundamental role of the Ccr4-Not complex in eukaryotic gene expression. In this review we will go over some of the recent studies that highlight such a role, and in particular shed light on a new functional module composed of Not3/5 and Not2. 2. Description of the Ccr4-Not complex 2.1. Composition The Ccr4-Not complex is evolutionarily conserved from yeast to human [21–25]. In Saccharomyces cerevisiae the core of the Ccr4-Not complex has a size of about 1 MDa and comprises 9 proteins (5 Not proteins, 3 Caf proteins and 1 Ccr4) [26,27]. The Not proteins (Not1, Not2, Not3, Not4 and Not5) obtained their name (Negative On TATA-less) from a selection suggesting that the NOT genes are important for repressing promoters lacking a canonical TATA box [28]. CCR4 (Carbon Catabolite Repression) was identified as a gene that positively regulates glucose-repressible enzymes [29]. The Caf proteins (Caf1, Caf40 and Caf130) derived their name from interaction with Ccr4 (Ccr4 associated factor) [30,31]. Other proteins, such as Caf4,

Caf16, Dhh1, Btt1 and several other less well-characterized proteins [32–35], associate with the core of the Ccr4-Not core in large complexes with the size of about 2 MDa (for review see [15]). In human cells there are two genes orthologous to yeast CAF1, CNOT7 and CNOT8 and 2 genes orthologous to CCR4, CNOT6 and CNOT6L [21,22] and for review see [11]. In contrast, there is only one gene, CNOT3, that is the ortholog of yeast NOT3 and NOT5, which have probably originated from a gene duplication event. They encode proteins with very similar N-termini, but have distinct C-termini (Fig. 1). Not5 has retained most of the essential functions and is quite critically important for yeast vegetative growth while deletion of Not3 has only very mild phenotypes [36]. The N-terminal domain of Not5 that is homologous to the N-terminal domain of Not3 is essential for growth at high temperature in the absence of Not3 [36]. But the critical importance of Not5 for vegetative growth really maps to its C-terminus (Fig. 1). 2.2. Structure The only structural data about the Ccr4-Not complex is known from baker's yeast. Not1 is the largest protein of the complex and is the major scaffolding protein to which the other Ccr4-Not subunits bind [26]. According to the electron microscopy data, the Ccr4-Not complex has an L-shape with two arms of 180 Å and 190 Å and a hinge domain in the middle [37]. The shorter arm is thought to be mainly formed by the Not proteins, whereas the longer, thinner arm, contains Caf40 and Caf130. Caf1 binds to the middle part of Not1 and brings Ccr4 to the scaffold [38,39]. Thus Ccr4 is positioned at a strategic place in the hinge-domain. Since biochemical and genetic data indicate that the Not-subunits interact with the C-terminus of Not1 it is likely that shorter arm corresponds to the C-terminus of Not1, and the longer arm to the N-terminus. The two-arm shape resolved by electron microscopy is very consistent with previous biochemical, molecular and genetic data (summarized in [11]). However, the complex is in fact heterogeneous, and an enrichment for L-shaped particles containing all Ccr4-Not subunits could be obtained only after chemical cross-linking of the purified complex [37]. 2.3. Function All the evidence points to the existence of different modules built on the Not1 scaffold, which perform different functions and contribute to regulate gene expression at different levels (Fig. 2). The existence of 2 enzymatic activities within the Ccr4-Not complex has clearly designated 2 functional modules (for review see [14]). 2.3.1. Ccr4–Caf1 module Caf1 and Ccr4 comprise the major eukaryotic deadenylase [40]. They perform the rate limiting deadenylation step in mRNA degradation. Ccr4 is a 3′exonuclease that has a DNA I-like EEP (endonuclease–exonuclease–phosphatase) domain [41,42]. Caf1 also has 3′exonuclease activity and contains an RNaseD/DEDD domain [43–45]. In S. cerevisiae Caf1 harbors a mutation in the active site [44] and does not contribute to the enzymatic activity, however already in Schizosaccharomyces pombe and further up in evolution, Caf1 contains the conserved catalytical residues [46] and is active. 2.3.2. Not4 module Not4 is a C4C4 RING E3 ligase, bearing in its RING two clusters of 4 Cys residues that coordinate two ions of Zn [47]. It not only directly ubiquitinates a variety of different substrates [17,19,20,48–50], but also exerts functions that require its association into the Ccr4-Not complex and its RING domain, but not its E3 ligase activity [18]. The role of Not4 appears to be regulatory. At the ribosome it plays a role in co-translational quality control, probably of both mRNA and protein (for review see Collart 2012, in press). Its function as an E3 ligase is

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751 3 26 37

71

124

177

119

195

215

745

384

558 560 C

yNot5 N 1

36

68

37

67

235 255 292

648

785 803 831 836

yNot3 N 1 C 128

155

233

314 395 397 425

457

618

655

CNot3 N 1

748 753 C

Pro-rich Gly/Serregion rich region

Pro-rich region

Domain super family: Not CCR4-Not complex component, N-terminal (IPR007207, PF04065) Domain super family: NOT2/NOT3/NOT5 (IPR007282, PF04153) Coiled-coil domain

Fig. 1. Schematic representation of known sequences of yeast Not5, yeast Not3 and human CNot3. The numbers correspond to the domain entries in the InterPro database (http:// www.ebi.ac.uk/interpro/) and the HMMPham database (http://pfam.sanger.ac.uk/) respectively.

important for this role. Not4 certainly also contributes to regulation of transcriptional responses, since it ubiquitinates and destabilizes stress-inducible transcription factors [49,50] or components that impact on the state of the chromatin [48]. Not4 also regulates DNA-binding of transcription factors [51] and interaction of transcription factors with co-activators [52]. Finally, Not4 is important for the assembly of a large multi-subunit complex, the proteasome, and for this function it needs its RING and Ccr4-Not interaction domains, but it does not require its E3 ligase activity [18].

2.3.3. Not2–3/5 module A third module is composed of Not3/5 and Not2. Many genetic experiments in yeast have shown that these proteins function together (for review see [15]), and we have data showing that recombinant Not2 and Not5 associate in a stable complex (Bukach et al., in preparation). The role of this module however, has been difficult to assess. In yeast, besides Not1 that is essential for viability, these 2 proteins are the most important for vegetative growth [14]. Many experiments have suggested that this module is closely connected to the transcription machinery [52–62] and to nuclear functions such as nuclear RNA surveillance and mRNA export [63–65]. Nevertheless, Not5 like other

Transcription and RNA targeting NOT5 NOT2 NOT3

NOT4 CAF40 CAF130

Protein quality control

NOT1 CAF1 CCR4 mRNA degradation

Fig. 2. Scheme of the Ccr4-Not complex indicating functional modules in different colors. Blue is for the Not2–3/5 module that associates with transcribing ORFs and/or mRNAs for Ccr4-Not complex recruitment; pink is for the Not4 module involved in protein quality control; yellow is for the Caf1–Ccr4 module involved in mRNA degradation. The Not1 scaffold is indicated in green. The Caf40 and Caf130 proteins whose exact function is not yet defined are in gray.

subunits of the Ccr4-Not complex is also present at translating ribosomes [19]. Interesting observations about this module of the Ccr4-Not complex have emerged from studies in higher eukaryotes [66–71]. They link this module both to coordination of transcriptional regulation and mRNA degradation (see below), while work in our group suggests the importance of this module for assembly of cellular complexes such as the proteasome (see below). 2.4. The Ccr4-Not complex: a platform that coordinates functions of its modules The association of all Ccr4-Not subunits within a complex is essential for cell viability, but it remains unclear whether the different subunits work only within the context of a complex, or whether the different functional modules also work outside of the complex. The observation that human CNot4 is not constitutively associated with the complex suggests that the modules may perform part of their function outside of the complex. However integration of the modules into the complex is at some point during their functional cycle essential, as will be outlined below. 2.4.1. Role in mRNA degradation The role of Ccr4 and Caf1 in mRNA deadenylation has been extensively described in many different reviews [72–75] and will not be covered here, except to remind the reader that mRNA deadenylation is the initial rate-limiting step in the major pathway of mRNA degradation. It is generally thought to be initiated by recruitment of the deadenylase to mRNA 3′UTRs via specific RNA binding proteins. Within the Ccr4-Not complex, only Ccr4 and Caf1 clearly contribute to the deadenylation process, but various different other subunits of the complex have been described as important for recruitment of the deadenylase to specific mRNAs (reviewed in [14]). Moreover, while in vitro Caf1 and Ccr4 are active deadenylases in the absence of Not1 [44,76,77], in vivo deadenylation requires Not1 [38,78]. These observations suggest that integration of Ccr4 and Caf1 within the Ccr4-Not complex may allow specific targeting and regulation of deadenylation. Hence, in cells deadenylation appears to be performed by Ccr4–Caf1, but in the context of the Ccr4-Not complex. Consistently formation of processing bodies, which contain factors required for translation repression, decapping and mRNA degradation (including Ccr4 and Caf1), requires deadenylation [79], and CNot1 as well as the deadenylase subunits [78]. 2.4.2. Role in nuclear transcription The contribution of the different subunits of the Ccr4-Not complex to transcription has not been well delimited (for reviews see [13,14,16]). Mostly the Not subunits of the complex have been connected to this

746

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751

function as mentioned above. The connections have been multiple. Not4 regulates stability of transcription factors or chromatin modifying enzymes [48–50]. Not4 also regulates interaction of transcription factors with co-factors [52] or DNA [51]. Ccr4-Not subunits cross-link to transcribing ORFs. This has been shown specifically not only for yeast Not2, Not5, Ccr4 and Caf1 [57,80] but also genome-wide to some extent for all subunits [81], though clearly Not2 and Not3 cross-link to the largest numbers of promoters. The human orthologs of Not1, Not2 and Caf1 also cross-link to promoters [82]. Not5 interacts with the general transcription factor TFIID [56–58,61] and Not1, Not2 and Not5 interact with the SAGA histone acetyltransferase ([35,53,83] and our unpublished observations). Not1 and Not5 also impact on the association of TFIID and SAGA with promoters [57,59,60,83]. Finally efficient transcription elongation requires the Ccr4-Not complex [84,85], probably because the complex helps RNA polymerase II to backtrack [86]. As can be seen by this enumeration, these studies have connected more particularly the Not subunits to transcription regulation, although the cross-linking of Ccr4 and Caf1 to transcribing ORFs suggests that as for mRNA degradation, even though the Not subunits may participate more directly to the transcription process, they are associated with the other subunits, at least in part, for this function. Hence, association of the Not subunits in the Ccr4-Not complex during transcriptional regulation is likely to play some role, possibly an essential one. 2.4.3. Role in protein synthesis and degradation Protein synthesis and degradation are strongly regulated in the cell by a mechanism called protein quality control that includes numerous chaperones and the proteasome [87,88]. A role of the Ccr4-Not complex, particularly the Not subunits, in protein quality control is suggested by the presence of the Ccr4-Not complex at translating ribosomes, the importance of Not2, Not4 and Not5 for a normal level of polysomes, and the accumulation of aggregated proteins in their absence [19]. Not4 ubiquitinates a small ribosomal protein Rps7A [19], and both subunits of a ribosome-associated chaperone called nascent polypeptideassociated complex, NAC [17]. This chaperone provides translation fidelity through the folding of newly synthesized proteins and protects them from interaction with inappropriate targets [89,90]. Its mono-ubiquitination by Not4 promotes its association with the ribosome and with the proteasome [20]. Hence, we have proposed that the Ccr4-Not complex, through its Not4 subunit, is involved in co-translational quality control of newly synthesized peptides (Collart 2012, in press). In addition, a direct effect of Not4 in substrate polyubiquitination and subsequent degradation was revealed by the identification of several targets for Not4 [48–50]. Subunits of the Ccr4-Not complex are also important for assembly of the proteasome [18]. This has been particularly well studied for the Not4 subunit, and the available evidence is that Not4 contributes to assembly of the proteasome through interaction with proteasome subunits and proteasome chaperones, such as Ecm29. A recent study reporting widespread co-translational assembly of protein complexes [91] determined that proteasome subunits of the regulatory particle interacted with mRNAs encoding other regulatory subunits and Ecm29, in a manner dependent upon intact polysomes [91]. This suggests that Ecm29 might interact with proteasome subunits co-translationally to support functional proteasome assembly. Hence, Not4 would participate in this process. Alternatively, Not4 might help assembly of the proteasome at the ribosome. Besides Not4, other subunits of the Ccr4-Not complex are important for integrity of the proteasome, and, in fact, Not2 and Not5 are even more essential for this than Not4 (unpublished observations). We believe that the role of Not4 is mostly regulatory while that of the Not2–3/5 module might be more fundamental. This is consistent with a more pronounced growth defect in yeast lacking Not2 or Not5 than in yeast lacking Not4. In fact, a very important and central role for the Not2–3/5 of the Ccr4-Not complex appears quite clearly from studies in higher eukaryotes, as will be outlined below.

3. The CNot3 funcional module is essential for diverse cellular functions The Ccr4-Not complex is essential in yeast, but each subunit is dispensable for viability to various extents, except the Not1 scaffold, which is absolutely essential. In the nematode Caenorhabditis elegans, deletion of all Not-orthologs tested gives rise to an early embryonic lethal phenotype [92,93]. This has indicated that the Not subunits of the Ccr4-Not complex are likely to be essential for embryonic development in higher eukaryotes. In contrast the deadenylase subunits, which have generally been expanded in eukaryotes other than S. cerevisiae, can be inactivated without leading to embryonic lethality. In worms, that of Caf1 but not of Ccr4 is embryonic lethal [92]. Similarly, in mouse the knockout of the CNot7 deadenylase (a Caf1 ortholog) leads to adult mice, with no overt problem except that they have a high bone mass phenotype [94] and the males are defective in spermatogenesis [95–97]. This specific phenotype is intriguing considering that in the adult mouse, CNot7 is one of the Ccr4-Not subunits that is ubiquitously expressed [98]. This is not the case for all subunits of the complex, and in particular not for CNot2 and CNot3 [98]. This has led to the idea that different forms of the Ccr4-Not complex function at different times and tissues and that the CNot2–CNot3 module might functionally regulate the Ccr4-Not complex. Like worms, mice lacking CNot3 die early in embryogenesis [99]. Blastocysts can develop and have a normal appearance. They also occur at Mendelian frequencies and express key markers of early embryonic differentiation at normal levels. However, while in epiblast cultures trophoblast cells from not3−/− spread and support outgrowths of the inner cell mass (ICM), the ICM cells themselves exhibit a severe outgrowth defect [99]. The importance of the Not subunits for embryonic development has hindered the generation of knockout animals to characterize their predominant role in the adult animal. However it is interesting to note that in humans, CNot3 was identified as a modifier of mutations in PRPF31 leading to retinitis pigmentosa with incomplete penetrance. In asymptomatic carriers CNot3 is expressed at low levels allowing higher expression of Prpf31 from the wild-type allele, and hence preventing retinal degeneration [67]. CNot3 is probably a direct negative regulator of PRPF31, since it associates with the PRPF31 promoter and its expression inversely correlates with that of PRPF31. CNot1, CNot2, and all of the deadenylase subunits affect cell proliferation and/or cell death [78,100–103], but expression of the different subunits of the Ccr4-Not complex does not change during the cell cycle. However CNot3, specifically, is modified at around mitosis. Its depletion promotes mitotic arrest, most likely by increasing stability of the mRNA encoding Mad1, a regulator of the spindle assembly checkpoint [68]. These different studies underlie the importance of CNot3 for various cellular functions. This critical role of the CNot3 module of the Ccr4-Not complex has been reinforced by global studies in cell lines or embryonic stem cells (ES cells), or finally analyses in heterozygous knockout mice, as will be outlined below.

3.1. CNot3 as a module for cellular self-renewal CNOT3 was identified in a genome-wide RNAi screen conducted in mouse ES cells to identify genes important for cells to self-renew [71]. It was the only gene encoding a Ccr4-Not subunit that came out of the screen. However, when all subunits of the Ccr4-Not complex were specifically tested, it could be shown that silencing of CNot1, CNot2 and CNot3, but not of the other components of the complex, had the same phenotype [66]. Consistently, the expression of these 3 subunits of the Ccr4-Not complex specifically, is high in the oocyte, fertilized egg and ES cells, but becomes significantly reduced upon differentiation of the pluripotent ES cells.

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751

The knockdown of CNot2 and CNot3 causes an increase in differentiated cells, mostly of the trophectoderm lineage. The morphological changes induced by silencing CNot1 are somewhat different from those caused by silencing of CNot2 and CNot3. Nevertheless in all cases the gene expression signatures are similar to those found during trophectoderm differentiation [66]. For the CNot3 knockdown, this was studied in detail [71] and it correlates with a transcriptional up-regulation of differentiation factors. It also correlates with a reduction of the pluripotency transcription factors Nanog, Sox2 and Oct4, which comprise the core transcription network that controls self-renewal of ES cells [104]. Reduction of CNot3 has no impact on the signaling pathways that are important for self-renewal or on the signaling pathways that can induce differentiation. It reduces ES cell proliferation and viability even when the ES cells are insulated from extrinsic differentiation signaling or when repression of pluripotent genes is inhibited. Hence, CNot3 regulates self-renewal independently of these signaling pathways and is more likely to play a role in the core self-renewal circuitry. The evidence is that CNot3 sustains self-renewal through specific regulation of genes involved in cell cycle progression and survival by being recruited to relevant genes [71]. Indeed a large number of promoter sites, close to transcription start sites, are bound by CNot3 in mouse ES cells. In many cases, these promoters are co-occupied by another factor important for self-renewal, Trim28 [105], and the sequences bound by the 2 proteins lie in close proximity and are enriched in specific chromatin marks (histone H3 lysine 4 and lysine 27 tri-methylation). They cluster with promoters bound by 2 other pluripotency transcription factors, c-Myc and Zfx, but are distinct from those of the core transcription network. The binding sites for the 4 factors lie in close proximity suggesting that they might act in a cooperative manner. They mostly define genes encoding transcription regulators involved in cell cycle and cell survival that are down-regulated during the 14 days of embryonic body formation. The knockdown of CNot1, CNot2 and CNot3 also increases the differentiation of human ES cells [66]. Hence, the role of these Not proteins in maintenance of self-renewal of ES cells is conserved. The enrichment for cancer genes supports the idea that the regulatory networks active in self-renewal in stem cells may be active in certain cancers. This idea led us to investigate a possible role of CNot3 in development of cancers. CNOT3 is indeed mutated in human cancers [106]. Furthermore, the pattern of mutations in CNOT3 may suggest its potential implication in tumorigenesis. Indeed 27 mutations in the coding sequence of CNot3 (which is 754 amino acids long) are found in the eight different types of cancers reported in the COSMIC databasev62 [106]. Strikingly, 9 mutations in CNot3 (33%) are recurrent with 6 mutations causing a glutamic acid (E) to lysine (K) amino acid change at position 20, and 3 mutations at position 70. This distribution of mutations is unlikely to be observed by chance (p value of 1.2 × 10−16, binomial test). 6 out of 9 recurrent mutations are specific to prostate cancer, 2 to hematopoietic and lymphoid tissues and 1 to the urinary tract. Both recurrently mutated amino acids (E20 and E70) are located in the Not3 N-terminal domain (see Fig. 1), which is extremely conserved down

E20K

747

to yeast (Fig. 3). Moreover glutamic acid at position 20 of human CNot3 remains unchanged in vertebrates and in yeasts (in Not3 and Not5). Glutamic acid at position 70 is conserved in Gnathostomata and is orthologous to aspartic acid found in lampreys and yeasts (again in Not3 and Not5). Similarity of negatively charged glutamic and aspartic acids additionally emphasizes the strength of evolutionary constraints over this residue in CNot3. On the other hand, in cancers negatively charged glutamic acids at positions 20 and 70 are mutated to positively charged lysines, which potentially may cause substantial conformational changes in the CNot3 protein. Recurrence of mutations at the same sites of the protein together with their heterozygous status is consistent with a “gain of function” type of mutations in cancers. This is also in line with frequent amplifications of the CNot3 locus in prostate and ovarian cancer cell lines (from the canSAR database: https://cansar.icr.ac.uk/cansar/). These data on the putative activating somatic alterations of CNot3 in cancer are complementary to the CNot3 loss of function phenotypes in Drosophila and mouse ([71] and see below). Both types of data point to the function of CNot3 in cell self-renewal and in the control of cell differentiation. 3.2. CNot3 in heart function A global RNA interference silencing screen of genes conserved between mammals and Drosophila melanogaster determined that the knockdown of several Ccr4-Not subunits, in particular CNot1, CNot3, CNot4 and to a lesser extent CNot2, reduces expression of a reporter construct specifically expressed in cardioblasts [99]. The strongest phenotype was found for CNot3, whose cardiac-specific knockdown resulted in dilated cardiomyopathy in Drosophila. Nevertheless similar abnormal heart structure and severely impaired cardiac function was observed in the CNot1 specific knockdown, as well as in the knockdown of Ubc4, an E2 enzyme known to function with the CNot4 E3 ligase [107]. The importance of CNot3 for heart function in the adult animal was evaluated in heterozygous CNot3 null mice. These mice express half the amount of CNot3, but normal amounts of the other subunits of the Ccr4-Not complex [70]. The haploinsufficiency showed no overt structural changes in the heart, but cardiac contractility was reduced, and this was due to an intrinsic impairment in heart function. Cardiac stress leads to exaggerated heart failure in not3 heterozygous mice, and expression of genes important for heart function was reduced. Whole hearts from the heterozygous mice bear reductions in specific chromatin marks (histone 3 lysine 4 tri-methylation and lysine 9 acetylation) that can be restored by treatment of hearts with a histone deacetylase inhibitor (valproic acid, VPA). Heart function, and all of these phenotypes, can be rescued by VPA. Thus, genes important for heart function might be under the control of the acetylation and methylation states of histone H3, themselves dependent upon the Ccr4-Not complex, in particular upon CNot3. These findings are consistent with studies in yeast,

E70K

H.sapiens P.troglodytes R.macaque M.musculus B.taurus L.africana M.eugenii O.anatinus A.carolinensis X.tropicalis D.rerio P.marinus S.cerevisiae_CNot5 S.cerevisiae_CNot3

Fig. 3. Amino acid alignment of the N-terminus of vertebrate CNot3 protein and S. cerevisiae Not3 and Not5 proteins. Red boxes correspond to amino acid positions recurrently mutated in cancers. Dots denote amino acids identical to human. Dashes denote missing data. S. cerevisiae is represented by the two paralogous proteins, Not3 and Not5.

748

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751

which have shown that global histone acetylation [108] and trimethylation of histone H3 lysine 4 [109,110] are reduced in cells lacking Not4, and particularly Not5. They suggest that these subunits of the Ccr4-Not complex impact on chromatin marks, which in turn lead to alterations in gene expression through epigenetic reprogramming. An important role in heart function for the Ccr4-Not complex would be supported by the finding of mutations in the Ccr4-Not genes that correlate with heart disease. Indeed, in humans, the CNot1 locus was one of the 12 loci correlated with different cardiac defects in genome-wide association studies performed by 2 consortia using very stringent significance thresholds. Specific analysis of CNot3 revealed that there is a significant correlation between a common minor allele of CNot3 with a modification about 969 base pairs upstream of the transcription start site and an abnormal cardiac repolarization [99]. 3.3. CNot3 as sensor for nutrient availability Not3 heterozygous mice are born smaller than wild-type mice, and remain smaller throughout life [70]. All organs have a normal histology, but are smaller than organs in the wild-type mice, though at a similar ratio to body weight. The exceptions are liver and adipose tissue, which are particularly reduced. Hepatic lipids accumulate poorly and adipocytes in white and brown adipose tissues are smaller. Metabolic rate of the heterozygous animals is increased, and glucose and triglyceride blood levels are reduced under fasting and after feeding. Since insulin levels are normal, this means that the insulin sensitivity is increased, and this even on a high fat diet, or in genetically obese mice. In the liver of 12-week old heterozygous mice, only about 1% of the mRNAs are expressed with a more than 2-fold difference in heterozygous knockout mice compared to wild-type mice. Most of these mRNAs are involved in metabolic processes, in particular in lipid metabolism, with catabolism-related genes up-regulated and lipogenic genes down-regulated. For some mRNAs tested, up-regulation correlated with an increased length of their poly(A) tails, and the 3′UTR regions of these genes were necessary and sufficient to mediate the down-regulation of reporter constructs by CNot3. The evidence is that CNot3 mediates association of the deadenylase with the mRNA via this 3′UTR [70]. The level of CNot3 in the liver and adipose tissue consistently decreases after 24 h of fasting, and returns to control levels upon refeeding. Since CNot3 mRNA levels do not change, the regulation of CNot3 is post-transcriptional. In this context, it is interesting to note that the 2 orthologs of CNot3 in yeast, Not3 and Not5, have also been reported to decrease in response to nutrient limitation, and this occurs by a posttranscriptional mechanism [60]. The yeast orthologs become progressively hyperphosphorlyated and less abundant when yeast cells grow from high glucose to glucose depletion and then cease completely to grow because of nutrient depletion. Furthermore, in yeast cells the absence of Not5 also leads to significant changes in expression of metabolic genes [111]. These observations are consistent with a conserved response of Not3/5 to nutrient levels for regulation of metabolic genes. 4. CNot3 marks mRNAs at birth, for regulation from synthesis to translation and degradation CNot3 is essential for embryonic development. Consistently, CNot3 is important for maintenance of ES cell renewal. However, CNot3 also appears to be essential for tissue-specific functions in the fully developed animal. It is essential at least for cardiac and liver function, in this latter case in particular for lipid metabolism. Several interesting observations emerge from these different studies. First, the function for CNot3 defined in a given “system” is related to the expression program that is “on” in this system: in ES cells, the program for self-renewal, in cardiac cells, cardiac specific gene expression, and in liver cells, genes related to lipid metabolism for which this organ is specialized. In yeast, global analyses of gene expression under different

growth conditions have already revealed that the genes, which require the Ccr4-Not complex for appropriate expression, are different in different growth conditions [111]. Hence, the specificity of Ccr4-Not function at a given time in a specific cell is defined by the genetic program, which is on. Second, gene regulators acting at different levels seem to be able to mobilize CNot3. In the case of self-renewal of ES cells, CNot3 is reported to associate with DNA and coincide at loci that are bound by specific transcriptional regulators [71]. Hence, specific transcription factors at promoters may contribute to recruit CNot3 to DNA, and the targets of CNot3 appear to be defined at the transcription level. This is consistent with previous studies, which have reported association of Not2, Not3 or Not5 with transcribing ORFs [57,80–82]. In the case of the role of CNot3 in expression of cardiac genes [99], the phenotypes associated with a reduction of CNot3 correlate with specific chromatin marks. This tells us that the chromatin marks are directly connected to the impact of CNot3 on gene expression, be it transcriptional or post-transcriptional (since results are emerging to show that chromatin marks are also related to post-transcriptional gene regulation [112]). How the chromatin marks are “set” by CNot3 is not determined, but one can hypothesize that the initial event is also a specific recruitment of CNot3 by cardiacspecific transcription factors to genes expressed in the heart. In any event, this confirms that CNot3 targets are defined already in the nucleus. In contrast, in liver cells CNot3 is reported to recruit the deadenylase module of the Ccr4-Not complex to the 3′UTR of mRNAs [70]. CNot3dependent regulation of a reporter construct with a heterologous promoter suggests that the 3′UTR is sufficient for this regulation by CNot3 in liver cells. This would suggest that CNot3 action on this set of genes in the liver is uncoupled to their promoters. However, it is unclear whether one can conclude from a reporter construct transfected into cell lines for which the 3′UTR is sufficient for CNot3 regulation, that coordinate regulation of metabolic genes by CNot3 is really only strictly dependent upon the 3′UTR of their mRNAs. It could still be that at the level of a liver cell the specificity of CNot3 function for metabolic genes finds its origin at the recruitment of CNot3 to relevant genes by tissue-specific transcription factors. Defining exactly to which extent functional cross-talk between promoter and 3′UTR is relevant to coordinate gene regulation is certainly a challenge in such complex systems. To summarize, from these studies it appears that CNot3, and the Ccr4-Not complex, acts downstream of many different specific gene regulators. What then is the specificity of the regulation by the Ccr4-Not complex? Combining what we have learned from higher eukaryotes with what we know about the different functions of the Ccr4-Not complex in particular in yeast, we can suggest that the complex contributes to integrate the regulation of gene expression at all levels (Fig. 4). The presence of Not3/Not5 at active genes makes it possible to recruit Not4 (via the Not1 scaffold) to ubiquitinate and regulate levels of chromatin modifiers and transcription factors at these genes (Fig. 4). The outcome of this could be positive or negative regulation of transcription, consistently with the observation that CNot3 appears as a positive regulator in cardioblasts, but as a negative regulator in ES cells. Once mRNA synthesis is completed, Not3/5 (and/or other subunits of the Ccr4-Not complex) might be transferred from promoters to mRNAs. This will provide mRNAs with “marks” originating from the nucleus, which will follow them into the cytoplasm and also provide Ccr4-Not-dependent regulation of gene expression in the cytoplasm. Consistently, the Ccr4-Not complex is present in translating ribosomes [19], where it can then impact on translation and protein quality control. Here again, Not4 is likely to play an important regulatory role, by ubiquitinating proteins such as NAC and Rps7A. In polysomes the Ccr4-Not complex can also contribute to translation arrest and/or mRNA degradation if translation is arrested or encounters problems. Indeed, recent evidence has shown that Not4 contributes to co-translational quality control and that at least part of this effect is probably via mRNA degradation ([19,113,114] and for review see Collart 2012, in press) (Fig. 4).

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751

To summarize, these findings allow us to propose a model for an integrated function for the yeast Ccr4-Not complex, whereby the Not2–3/ 5 module in particular may act early on in the nucleus to target the Ccr4-Not complex to specific loci, probably marked by specific transcription factors. By being present at promoters, Not2–3/5 is then able to recruit the Ccr4-Not complex, including the Not4 E3 ligase. In turn, Not4 has the ability to ubiquitinate transcription factors or factors that act on chromatin modifications to module the transcription status of these loci (Fig. 4). Obviously, while we propose here a predominant role for Not2–3/5 for initial recruitment to promoters, it could be that other subunits of the Ccr4-Not complex have this role in some conditions. This is compatible with the observation that there are indeed promoters to which other subunits of the Ccr4-Not complex predominantly cross-link [81]. Once recruited, the Ccr4-Not complex can participate in productive elongation during the transcription elongation process [86]. In case of abortive transcription, the Ccr4-Not can contribute to nuclear RNA surveillance, consistently with its described putative role in this process [63–65]. When the mRNAs produced at these loci are completed, one or more modules of the Ccr4-Not complex might be transferred from the promoter to the mRNA. This could occur prior to, or after, the association of specific RNA binding proteins to the mRNA 3′UTRs. Finally, the marked mRNAs will be exported to the cytoplasm for translation by the ribosome. We find the Ccr4-Not complex at translating ribosomes [19], maybe at the ribosomes translating the mRNAs, which were “marked” via Not2–Not3/5 in the nucleus. The mRNAs marked this way may need to be coordinately degraded or stored in response to extracellular signals, or might encode subunits of the same complex, and the Ccr4-Not complex would play an essential role in this coordination. It has long been known that perfect complementation of the knockout of endogenous genes often requires the endogenous promoter and 3′UTR, and our model for Ccr4-Not function would account for this long-standing observation. Obviously we are still far from having

749

the data that can fully establish this model, but it is a model that can explain and integrate a great number of observations made so far, and which will be exciting to explore in the future. Acknowledgments We thank Zoltan Villanyi and Virginie Ribaud for a critical reading of the manuscript. This work was supported by grant 31003A_135794 from the Swiss National Science awarded to MAC. References [1] R.D. Burgoyne, A. Morgan, Physiological Reviews (2003) 581–632. [2] N.L. Garneau, J. Wilusz, C.J. Wilusz, Nature Reviews. Molecular Cell Biology (2007) 113–126. [3] R.J. Jackson, C.U. Hellen, T.V. Pestova, Nature Reviews. Molecular Cell Biology (2010) 113–127. [4] D. Finley, H.D. Ulrich, T. Sommer, P. Kaiser, Genetics (2012) 319–360. [5] F. Spitz, E.E. Furlong, Nature Reviews. Genetics (2012) 613–626. [6] Y.C. Tsai, A.M. Weissman, Genes & Cancer (2010) 764–778. [7] J. Shim, M. Karin, Molecules and Cells (2002) 323–331. [8] M. Sun, B. Schwalb, D. Schulz, N. Pirkl, S. Etzold, L. Lariviere, K.C. Maier, M. Seizl, A. Tresch, P. Cramer, Genome Research (2012) 1350–1359. [9] M. Dori-Bachash, E. Shema, I. Tirosh, PLoS Biology (2011) e1001106. [10] V. Goler-Baron, M. Selitrennik, O. Barkai, G. Haimovich, R. Lotan, M. Choder, Genes & Development (2008) 2022–2027. [11] M.A. Collart, H.T. Timmers, Progress in Nucleic Acid Research and Molecular Biology (2004) 289–322. [12] C.L. Denis, J. Chen, Progress in Nucleic Acid Research and Molecular Biology (2003) 221–250. [13] J.C. Reese, Biochimica et Biophysica Acta (2012), http://dx.doi.org/10.1016/j. bbagrm.2012.09.001 (pii: S1874-9399(12)00160-5.). [14] M.A. Collart, O.O. Panasenko, Gene (2012) 42–53. [15] M.A. Collart, Gene (2003) 1–16. [16] J.E. Miller, J.C. Reese, Critical Reviews in Biochemistry and Molecular Biology (2012) 315–333. [17] O. Panasenko, E. Landrieux, M. Feuermann, A. Finka, N. Paquet, M.A. Collart, The Journal of Biological Chemistry (2006) 31389–31398.

Pol II TF

TF Pol II

Not4 Not3/5

Not3/5

TF Pol II

Not3/5 AAAAA Not3/5

NUCLEUS

A

CYTOPLASM

A

AAAAA

AAA

Not3/5

Caf1/Ccr4

AAAAA

AAAAA

Not1 Not3/5

Not1 Not3/5

Not4

Fig. 4. Model for the implication of the different modules of the yeast Ccr4-Not complex in control of gene expression. 1. In the nucleus the Not2–Not3/5 module is associated with loci marked by specific transcription factors and can act to recruit the subunits of the Ccr4-Not complex, in particular Not4, to the loci. 2. By being recruited, the Not4 E3 ligase has the ability to ubiquitinate transcription factors or factors that act on chromatin marks to module the expression status of these loci. 3. When the mRNAs produced at these loci are completed, the Not2–Not3/5 module might be transferred from the promoter to the mRNA 3′UTRs. Not2–Not3/5 might also be able to interact with 3′UTRs independently of the promoter. 4. The “marked” mRNAs will be exported to the cytoplasm for translation by the ribosome. 5. The Ccr4-Not complex is present in translating ribosomes, probably those translating the “marked” mRNAs. 6. Not4 contributes to co-translational protein quality control and might recruit the proteasome in case of aberrant peptides. 7. The Caf1/Ccr4 module deadenylates mRNA after translation arrest or termination.

750 [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34]

[35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]

[52] [53] [54] [55] [56] [57] [58] [59]

[60] [61] [62] [63] [64]

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751 O.O. Panasenko, M.A. Collart, Molecular and Cellular Biology (2011) 1610–1623. O.O. Panasenko, M.A. Collart, Molecular Microbiology (2012) 640–653. O.O. Panasenko, F.P. David, M.A. Collart, Genetics (2009) 447–460. T.K. Albert, M. Lemaire, N.L. van Berkum, R. Gentz, M.A. Collart, H.T. Timmers, Nucleic Acids Research (2000) 809–817. N.C. Lau, A. Kolkman, F.M. van Schaik, K.W. Mulder, W.W. Pijnappel, A.J. Heck, H.T. Timmers, The Biochemical Journal (2009) 443–453. A.P. Morel, S. Sentis, C. Bianchin, M. Le Romancer, L. Jonard, M.C. Rostan, R. Rimokh, L. Corbo, Journal of Cell Science (2003) 2929–2936. C. Temme, S. Zaessinger, S. Meyer, M. Simonelig, E. Wahle, The EMBO Journal (2004) 2862–2871. C. Temme, L. Zhang, E. Kremmer, C. Ihling, A. Chartier, A. Sinz, M. Simonelig, E. Wahle, RNA (2010) 1356–1370. L. Maillet, C. Tu, Y.K. Hong, E.O. Shuster, M.A. Collart, Journal of Molecular Biology (2000) 131–143. Y. Bai, C. Salvadore, Y.C. Chiang, M.A. Collart, H.Y. Liu, C.L. Denis, Molecular and Cellular Biology (1999) 6642–6651. M.A. Collart, K. Struhl, Genes & Development (1994) 525–537. C.L. Denis, Genetics (1984) 833–844. M.P. Draper, C. Salvadore, C.L. Denis, Molecular and Cellular Biology (1995) 3487–3495. J. Chen, J. Rappsilber, Y.C. Chiang, P. Russell, M. Mann, C.L. Denis, Journal of Molecular Biology (2001) 683–694. H. Hata, H. Mitsui, H. Liu, Y. Bai, C.L. Denis, Y. Shimizu, A. Sakai, Genetics (1998) 571–579. L. Maillet, M.A. Collart, The Journal of Biological Chemistry (2002) 2835–2842. H.Y. Liu, Y.C. Chiang, J. Pan, J. Chen, C. Salvadore, D.C. Audino, V. Badarinarayana, V. Palaniswamy, B. Anderson, C.L. Denis, The Journal of Biological Chemistry (2001) 7541–7548. Y. Cui, D.B. Ramnarain, Y.C. Chiang, L.H. Ding, J.S. McMahon, C.L. Denis, Molecular Genetics and Genomics: MGG (2008) 323–337. U. Oberholzer, M.A. Collart, Gene (1998) 61–69. F. Nasertorabi, C. Batisse, M. Diepholz, D. Suck, B. Bottcher, FEBS Letters (2011) 2182–2186. J. Basquin, V.V. Roudko, M. Rode, C. Basquin, B. Seraphin, E. Conti, Molecular Cell (2012) 207–218. A.P. Petit, L. Wohlbold, P. Bawankar, E. Huntzinger, S. Schmidt, E. Izaurralde, O. Weichenrieder, Nucleic Acids Research (2012) 11058–11072. M. Tucker, M.A. Valencia-Sanchez, R.R. Staples, J. Chen, C.L. Denis, R. Parker, Cell (2001) 377–386. M.P. Draper, H.Y. Liu, A.H. Nelsbach, S.P. Mosley, C.L. Denis, Molecular and Cellular Biology (1994) 4522–4531. H. Wang, M. Morita, X. Yang, T. Suzuki, W. Yang, J. Wang, K. Ito, Q. Wang, C. Zhao, M. Bartlam, T. Yamamoto, Z. Rao, The EMBO Journal (2010) 2566–2576. C. Bianchin, F. Mauxion, S. Sentis, B. Seraphin, L. Corbo, RNA (2005) 487–494. S. Thore, F. Mauxion, B. Seraphin, D. Suck, EMBO Reports (2003) 1150–1155. M.C. Daugeron, F. Mauxion, B. Seraphin, Nucleic Acids Research (2001) 2448–2455. M.J. Moser, W.R. Holley, A. Chatterjee, I.S. Mian, Nucleic Acids Research (1997) 5110–5118. H. Hanzawa, M.J. de Ruwe, T.K. Albert, P.C. van Der Vliet, H.T. Timmers, R. Boelens, The Journal of Biological Chemistry (2001) 10185–10190. D.P. Mersman, H.N. Du, I.M. Fingerman, P.F. South, S.D. Briggs, Genes & Development (2009) 951–962. K.F. Cooper, M.S. Scarnati, E. Krasley, M.J. Mallory, C. Jin, M.J. Law, R. Strich, Journal of Cell Science (2012) 1015–1026. K. Gulshan, B. Thommandru, W.S. Moye-Rowley, The Journal of Biological Chemistry (2012) 26796–26805. J. Gronholm, M. Kaustio, H. Myllymaki, J. Kallio, J. Saarikettu, J. Kronhamn, S. Valanne, O. Silvennoinen, M. Ramet, FASEB Journal: Official Publication of the Federation of American Societies for Experimental Biology (2012) 1239–1250. E. Lenssen, N. Azzouz, A. Michel, E. Landrieux, M.A. Collart, Eukaryotic Cell (2007) 2251–2259. J.D. Benson, M. Benson, P.M. Howley, K. Struhl, The EMBO Journal (1998) 6714–6722. S. Jayne, C.G. Zwartjes, F.M. van Schaik, H.T. Timmers, The Biochemical Journal (2006) 461–467. C.G. Zwartjes, S. Jayne, D.L. van den Berg, H.T. Timmers, The Journal of Biological Chemistry (2004) 10848–10854. V. Badarinarayana, Y.C. Chiang, C.L. Denis, Genetics (2000) 1045–1054. C. Deluen, N. James, L. Maillet, M. Molinete, G. Theiler, M. Lemaire, N. Paquet, M.A. Collart, Molecular and Cellular Biology (2002) 6735–6749. M. Lemaire, M.A. Collart, The Journal of Biological Chemistry (2000) 26925–26934. E. Lenssen, N. James, I. Pedruzzi, F. Dubouloz, E. Cameroni, R. Bisig, L. Maillet, M. Werner, J. Roosen, K. Petrovic, J. Winderickx, M.A. Collart, C. De Virgilio, Molecular and Cellular Biology (2005) 488–498. E. Lenssen, U. Oberholzer, J. Labarre, C. De Virgilio, M.A. Collart, Molecular Microbiology (2002) 1023–1037. S.L. Sanders, J. Jennings, A. Canutescu, A.J. Link, P.A. Weil, Molecular and Cellular Biology (2002) 4723–4738. T. Aoki, N. Okada, T. Wakamatsu, T.A. Tamura, Biochemical and Biophysical Research Communications (2002) 1097–1103. N. Azzouz, O.O. Panasenko, G. Colau, M.A. Collart, PloS One (2009) e6760. S.C. Kerr, N. Azzouz, S.M. Fuchs, M.A. Collart, B.D. Strahl, A.H. Corbett, R.N. Laribee, PloS One (2011) e18302.

[65] J. Assenholt, J. Mouaikel, C. Saguez, M. Rougemaille, D. Libri, T.H. Jensen, RNA (2011) 1788–1794. [66] X. Zheng, R. Dumitru, B.L. Lackford, J.M. Freudenberg, A.P. Singh, T.K. Archer, R. Jothi, G. Hu, Stem Cells (2012) 910–922. [67] G. Venturini, A.M. Rose, A.Z. Shah, S.S. Bhattacharya, C. Rivolta, PLoS Genetics (2012) e1003040. [68] A. Takahashi, C. Kikuguchi, M. Morita, T. Shimodaira, N. Tokai-Nishizumi, K. Yokoyama, M. Ohsugi, T. Suzuki, T. Yamamoto, Biochemical and Biophysical Research Communications (2012) 268–273. [69] R. Diaz-Pena, A.M. Aransay, B. Suarez-Alvarez, J. Bruges-Armas, N. Rodriguez-Ezpeleta, M. Regueiro, F.M. Pimentel-Santos, J. Mulero, A. Sanchez, E. Collantes, R. Queiro, J. Ballina, H. Alves, C. Lopez-Larrea, Annals of the Rheumatic Diseases (2012) 714–717. [70] M. Morita, Y. Oike, T. Nagashima, T. Kadomatsu, M. Tabata, T. Suzuki, T. Nakamura, N. Yoshida, M. Okada, T. Yamamoto, The EMBO Journal (2011) 4678–4691. [71] G. Hu, J. Kim, Q. Xu, Y. Leng, S.H. Orkin, S.J. Elledge, Genes & Development (2009) 837–848. [72] C.Y. Chen, A.B. Shyu, Wiley Interdisciplinary Reviews. RNA (2011) 167–183. [73] M. Bartlam, T. Yamamoto, Protein & Cell (2010) 443–452. [74] K. Wiederhold, L.A. Passmore, Biochemical Society Transactions (2010) 1531–1536. [75] R. Doidge, S. Mittal, A. Aslam, G.S. Winkler, Biochemical Society Transactions (2012) 896–901. [76] L.B. Clark, P. Viswanathan, G. Quigley, Y.C. Chiang, J.S. McMahon, G. Yao, J. Chen, A. Nelsbach, C.L. Denis, The Journal of Biological Chemistry (2004) 13616–13623. [77] A.T. Jonstrup, K.R. Andersen, L.B. Van, D.E. Brodersen, Nucleic Acids Research (2007) 3153–3164. [78] K. Ito, A. Takahashi, M. Morita, T. Suzuki, T. Yamamoto, Protein & Cell (2011) 755–763. [79] D. Zheng, N. Ezzeddine, C.Y. Chen, W. Zhu, X. He, A.B. Shyu, The Journal of Cell Biology (2008) 89–101. [80] M.J. Swanson, H. Qiu, L. Sumibcay, A. Krueger, S.J. Kim, K. Natarajan, S. Yoon, A.G. Hinnebusch, Molecular and Cellular Biology (2003) 2800–2820. [81] B.J. Venters, S. Wachi, T.N. Mavrich, B.E. Andersen, P. Jena, A.J. Sinnamon, P. Jain, N.S. Rolleri, C. Jiang, C. Hemeryck-Walsh, B.F. Pugh, Molecular Cell (2011) 480–492. [82] G.S. Winkler, K.W. Mulder, V.J. Bardwell, E. Kalkhoven, H.T. Timmers, The EMBO Journal (2006) 3089–3099. [83] N. James, E. Landrieux, M.A. Collart, Genetics (2007) 123–135. [84] C.L. Denis, Y.C. Chiang, Y. Cui, J. Chen, Genetics (2001) 627–634. [85] H. Gaillard, C. Tous, J. Botet, C. Gonzalez-Aguilera, M.J. Quintero, L. Viladevall, M.L. Garcia-Rubio, A. Rodriguez-Gil, A. Marin, J. Arino, J.L. Revuelta, S. Chavez, A. Aguilera, PLoS Genetics (2009) e1000364. [86] J.A. Kruk, A. Dutta, J. Fu, D.S. Gilmour, J.C. Reese, Genes & Development (2011) 581–593. [87] F.U. Hartl, M. Hayer-Hartl, Nature Structural & Molecular Biology (2009) 574–581. [88] B. Chen, M. Retzlaff, T. Roos, J. Frydman, Cold Spring Harbor Perspectives in Biology (2011) a004374. [89] B. Lauring, H. Sakai, G. Kreibich, M. Wiedmann, Proceedings of the National Academy of Sciences of the United States of America (1995) 5411–5415. [90] S. Wang, H. Sakai, M. Wiedmann, The Journal of Cell Biology (1995) 519–528. [91] C.D. Duncan, J. Mata, PLoS Genetics (2011) e1002398. [92] I. Maeda, A. Sugimoto, Tanpakushitsu Kakusan Koso. Protein, Nucleic Acid, Enzyme (2001) 2432–2435. [93] R.S. Kamath, A.G. Fraser, Y. Dong, G. Poulin, R. Durbin, M. Gotta, A. Kanapin, N. Le Bot, S. Moreno, M. Sohrmann, D.P. Welchman, P. Zipperlen, J. Ahringer, Nature (2003) 231–237. [94] K. Washio-Oikawa, T. Nakamura, M. Usui, M. Yoneda, Y. Ezura, I. Ishikawa, K. Nakashima, T. Noda, T. Yamamoto, M. Noda, Journal of Bone and Mineral Research: The Official Journal of the American Society for Bone and Mineral Research (2007) 1217–1223. [95] C. Berthet, A.M. Morera, M.J. Asensio, M.A. Chauvin, A.P. Morel, F. Dijoud, J.P. Magaud, P. Durand, J.P. Rouault, Molecular and Cellular Biology (2004) 5808–5820. [96] T. Ogawa, C. Ito, T. Nakamura, Y. Tamura, T. Yamamoto, T. Noda, Y. Kubota, K. Toshimori, Archives of Histology and Cytology (2004) 307–314. [97] T. Nakamura, R. Yao, T. Ogawa, T. Suzuki, C. Ito, N. Tsunekawa, K. Inoue, R. Ajima, T. Miyasaka, Y. Yoshida, A. Ogura, K. Toshimori, T. Noce, T. Yamamoto, T. Noda, Nature Genetics (2004) 528–533. [98] C. Chen, K. Ito, A. Takahashi, G. Wang, T. Suzuki, T. Nakazawa, T. Yamamoto, K. Yokoyama, Biochemical and Biophysical Research Communications (2011) 360–364. [99] G.G. Neely, K. Kuba, A. Cammarato, K. Isobe, S. Amann, L. Zhang, M. Murata, L. Elmen, V. Gupta, S. Arora, R. Sarangi, D. Dan, S. Fujisawa, T. Usami, C.P. Xia, A.C. Keene, N.N. Alayari, H. Yamakawa, U. Elling, C. Berger, M. Novatchkova, R. Koglgruber, K. Fukuda, H. Nishina, M. Isobe, J.A. Pospisilik, Y. Imai, A. Pfeufer, A.A. Hicks, P.P. Pramstaller, S. Subramaniam, A. Kimura, K. Ocorr, R. Bodmer, J.M. Penninger, Cell (2010) 142–153. [100] K. Ito, T. Inoue, K. Yokoyama, M. Morita, T. Suzuki, T. Yamamoto, Genes to Cells: Devoted to Molecular & Cellular Mechanisms (2011) 368–379. [101] A. Aslam, S. Mittal, F. Koch, J.C. Andrau, G.S. Winkler, Molecular Biology of the Cell (2009) 3840–3850. [102] S. Mittal, A. Aslam, R. Doidge, R. Medica, G.S. Winkler, Molecular Biology of the Cell (2011) 748–758. [103] M. Morita, T. Suzuki, T. Nakamura, K. Yokoyama, T. Miyasaka, T. Yamamoto, Molecular and Cellular Biology (2007) 4980–4990.

M.A. Collart et al. / Cellular Signalling 25 (2013) 743–751 [104] J. Wang, S. Rao, J. Chu, X. Shen, D.N. Levasseur, T.W. Theunissen, S.H. Orkin, Nature (2006) 364–368. [105] T.G. Fazzio, J.T. Huff, B. Panning, Cell (2008) 162–174. [106] S.A. Forbes, N. Bindal, S. Bamford, C. Cole, C.Y. Kok, D. Beare, M. Jia, R. Shepherd, K. Leung, A. Menzies, J.W. Teague, P.J. Campbell, M.R. Stratton, P.A. Futreal, Nucleic Acids Research (2011) D945–D950. [107] T.K. Albert, H. Hanzawa, Y.I. Legtenberg, M.J. de Ruwe, F.A. van den Heuvel, M.A. Collart, R. Boelens, H.T. Timmers, The EMBO Journal (2002) 355–364. [108] W. Peng, C. Togawa, K. Zhang, S.K. Kurdistani, Genetics (2008) 277–289. [109] R.N. Laribee, Y. Shibata, D.P. Mersman, S.R. Collins, P. Kemmeren, A. Roguev, J.S. Weissman, S.D. Briggs, N.J. Krogan, B.D. Strahl, Proceedings of the National Academy of Sciences of the United States of America (2007) 5836–5841.

751

[110] K.W. Mulder, A.B. Brenkman, A. Inagaki, N.J. van den Broek, H.T. Timmers, Nucleic Acids Research (2007) 2428–2439. [111] N. Azzouz, O.O. Panasenko, C. Deluen, J. Hsieh, G. Theiler, M.A. Collart, RNA (2009) 377–383. [112] L.I. Gomez Acuna, A. Fiszbein, M. Allo, I.E. Schor, A.R. Kornblihtt, Wiley Interdisciplinary Reviews. RNA (2012), http://dx.doi.org/10.1002/wrna.1142. Epub 2012 Oct 16. [113] L.N. Dimitrova, K. Kuroha, T. Tatematsu, T. Inada, The Journal of Biological Chemistry (2009) 10343–10352. [114] M.H. Bengtson, C.A. Joazeiro, Nature (2010) 470–473.