- - -
- - -
Dépôt Institutionnel de l’Université libre de Bruxelles / Université libre de Bruxelles Institutional Repository
Thèse de doctorat/ PhD Thesis Citation APA:
Zhang, T. (2005). Characterization of the shuttling properties of RNA-binding TIA proteins (Unpublished doctoral dissertation). Université libre de Bruxelles, Faculté des sciences, Bruxelles.
Disponible à / Available at permalink : https://dipot.ulb.ac.be/dspace/bitstream/2013/210999/1/fa0af6d8-7027-42a9-a3a2-53aef6d73940.txt
(English version below)
Cette thèse de doctorat a été numérisée par l’Université libre de Bruxelles. L’auteur qui s’opposerait à sa mise en ligne dans DI-fusion est invité à prendre contact avec l’Université (di-fusion@ulb.be).
Dans le cas où une version électronique native de la thèse existe, l’Université ne peut garantir que la présente version numérisée soit identique à la version électronique native, ni qu’elle soit la version officielle définitive de la thèse.
DI-fusion, le Dépôt Institutionnel de l’Université libre de Bruxelles, recueille la production scientifique de l’Université, mise à disposition en libre accès autant que possible. Les œuvres accessibles dans DI-fusion sont protégées par la législation belge relative aux droits d'auteur et aux droits voisins. Toute personne peut, sans avoir à demander l’autorisation de l’auteur ou de l’ayant-droit, à des fins d’usage privé ou à des fins d’illustration de l’enseignement ou de recherche scientifique, dans la mesure justifiée par le but non lucratif poursuivi, lire, télécharger ou reproduire sur papier ou sur tout autre support, les articles ou des fragments d’autres œuvres, disponibles dans DI-fusion, pour autant que :
Le nom des auteurs, le titre et la référence bibliographique complète soient cités;
L’identifiant unique attribué aux métadonnées dans DI-fusion (permalink) soit indiqué;
Le contenu ne soit pas modifié.
L’œuvre ne peut être stockée dans une autre base de données dans le but d’y donner accès ; l’identifiant unique (permalink) indiqué ci-dessus doit toujours être utilisé pour donner accès à l’œuvre. Toute autre utilisation non mentionnée ci-dessus nécessite l’autorisation de l’auteur de l’œuvre ou de l’ayant droit.
--- English Version ---
This Ph.D. thesis has been digitized by Université libre de Bruxelles. The author who would disagree on its online availability in DI-fusion is invited to contact the University (di-fusion@ulb.be).
If a native electronic version of the thesis exists, the University can guarantee neither that the present digitized version is identical to the native electronic version, nor that it is the definitive official version of the thesis.
DI-fusion is the Institutional Repository of Université libre de Bruxelles; it collects the research output of the University, available on open access as much as possible. The works included in DI-fusion are protected by the Belgian legislation relating to authors’ rights and neighbouring rights.
Any user may, without prior permission from the authors or copyright owners, for private usage or for educational or scientific research purposes, to the extent justified by the non-profit activity, read, download or reproduce on paper or on any other media, the articles or fragments of other works, available in DI-fusion, provided:
The authors, title and full bibliographic details are credited in any copy;
The unique identifier (permalink) for the original metadata page in DI-fusion is indicated;
The content is not changed in any way.
It is not permitted to store the work in another database in order to provide access to it; the unique identifier (permalink) indicated above must always be used to provide access to the work. Any other use not mentioned above requires the authors’ or copyright owners’ permission.
DBM 00654
Université Libre de Bruxelles
~ ——--- Faculté des Sciences
Characterization of the shuttling properties of RNA- binding TIA proteins
Promoteurs
Prof. Georges HUEZ Prof Véronique KRUYS
Laboratoire de Chimie Biologique Département de Biologie Moléculaire
ULB - IBMM !
BIBLIOTHEQUE I
Thèse présentée en vue de l’obtention du grade de Docteur en Sciences
Tong ZHANG Novembre 2005
Binding TIA proteins
Promotuers
Prof. Georges HUEZ Prof. Véronique KRUYS
Laboratoire de Chimie Biologique Département de Biologie Moléculaire
Faculté des sciences Thèse présentée en vue
Université Libre de Bruxelles de Vobtention du grade
de Docteur en Sciences
(B^merciements
Je remercie le professeur Georges Huez de m 'avoir accueillie au sein de son laboratoire et de m 'avoir permis de réaliser ce travail. Merci pour votre soutien et votre aide.
Je voudrais dire un tout grand merci à Véro car sans elle ce travail n 'aurait jamais abouti !!
Merci de m 'avoir soutenue dans tous ces difficiles instants qui parcourent une thèse !!
Merci pour toutes ces bonnes idées qui ont fini par me mettre sur la bonne voie !!
Et puis aussi mille mercis pour toutes les heures que tu as passées à corriger cette thèse !!
Je voulais aussi te remercier pour toutes les petites conversations qu 'on a eues au cours de ces années. J'espère qu 'ily en aura encore bien d'autres...
Cyril, tout d'abord merci de m'avoir accueillie et de m'avoir fait découvrir ce merveilleux pays qu 'est la Belgique ...merci de ton soutien sur le difficile chemin de la thèse...
On aura eu des bons et des moins bons moments, mais ce que je retiendrai surtout c 'est ta grande disponibilité !! MERCI !!
Et puis il y a Julie, toi tu auras été là à la fin de ce long périple, tu m'auras supportée jusqu 'au bout !! Je voudrais te dire un grand merci pour tous les moments café-réconfort !!
Et puis aussi te dire merci pour ton amitié sans faille !! Merci pour tout et bonne chance pour la suite ma Juju !!
Sans oublier Delphine et Sabrina. Merci à vous pour toutes ces petites soirées très .sympathiques !! Merci à toi Sah pour tous tes bons conseils !! Ton calme et ta façon de gérer la vie m'a toujours impressionnée!! Merci pour tout et pour mes chats en particulier !! Merci à toi Delphine pour ta gentillesse et ton écoute sans égale !! A très bientôt les filles !!
Je voudrais aussi remercier ceux sans qui, les jours de grandes manips, j'aurais perdu pieds a Merci à vous de m'avoir prêté vos mains pendant ces nombreuses années et merci pour tous ces bons moments de complicité !! Merci à vous Coco, Patrick et Flo !!
Merci à Monique et Marylin pour leur grande aide !! Sans vous mes fins de mois auraient été difficiles !!
Enfin, merci à tous ceux qui ont vécu quelques temps à mes cotés dans ce laboratoire !!
Chacun, à votre manière, avez apporté quelque chose dans ma vie !! Merci !!
MERCI...
Abbreviations
3’-UTR: 3’ untranslated région Act D: actinomycin D
ARE: AU-rich element BSA: bovin sérum albumin cDNA: complementary DNA CHX: cycloheximide
CIRP: cold-inducing RNA-binding protein COUP-TF: chicken ovalbumin upstream promoter- transcription factors
Cox-2: cyclooxygenase-2
CPRG: chlorophenol red-P-Dgalactopyranoside CRM: Chromosome région maitenance 1 CTE: constitutive transport element CTL: cytotoxic T cell
dsRBD: double-stranded RNA binding domain eEF: eukaryotic élongation factor
EJC: exon-junction complex
ELAV: embryonic léthal abnormal vision EMSA: electrophoretic mobility shift assays ERK: extracellular signal-regulated kinase FAST: Fas activated serine/threonine FGFRF-2: fibroblast growth factor HIV: human immunodeficiency virus HMMP: human matrix metallinoproteinase HuR: human autoantigen R
ILF: interleukin enhancer binding factor LMB: leptomycin B
LPS: lipopolysaccharides MER2: meiotic recombination 2 msl: male spécifie léthal
NES: nuclear export signal NLS: nuclear localization signais
PABP: poly (A)-binding protein PCBP: poly (rC) RNA-binding protein
PHAX: phosphorylated adaptor for RNA export PIAS: protein inhibitor of activated STAT PML: promyelocytic leukaemia
RanBP: Ran-binding proteins
RanGAP:RanGTPase activating protein-1
RanGEF: Ran guanine nucléotide exchange factor Ras: rat sarcoma virus
REF: RNA and export factor binding proteins RNA-BPs : RNA-binding proteins
RNP: ribonucleoprotein RRM: RNA récognition motif Sev: sendai vims
SG: stress granule
SPR: signal récognition particle SR: serine/arginine rich
SUMO: small ubiquitin-related modifier TIA-1: T cell intracellular antigen-1 TIAR: TIA-1 related protein TNF: tumor necrosis factor TTP: tristetraprolin
TZF: tandem CCCH zinc finger
U1 snRNP: U1 small nuclear ribonucleoprotein UBA: ubiquitin associated
Ubc9: SUMO-conjugating enzyme U-rich: uridine-rich éléments UV: ultraviolets
NMR: nuclear magnetic résonance
NPC: nuclear pore complex
Abstract
Abstract
Post-transcriptional régulation of mRNA plays a very important rôle in eukaryotes. Indeed, after transcription, many events contribute to the formation of mature mRNAs. These include mRNA capping, polyadenylation, splicing (alternative splicing) and nuclear export. In the cytoplasm, the localization, the stability and translation of mRNAs can also contribute to the régulation of gene expression. Ail these mechanisms rely on non spécifie and spécifie interactions of proteins with RNA sequences. These proteins are identified as RNA-binding proteins and constitute a major class of proteins which control gene expression. Among them, we find TLAR and TIA-1 which are two closely related RNA-binding proteins containing three RNA récognition motifs (RRMs) followed by a 100 amino acid-long auxiliary région. These proteins are involved in several mechanisms of RNA metabolism, including alternative hnRNA splicing and régulation of mRNA translation. In contrast to germinal cells, in which TLAR and TIA-1 are mostly cytoplasmic, most somatic cells accumulate these proteins both in the nuclear and cytoplasm, the highest proportion being nuclear at the equilibrium.
In our study, we first characterized the subcellular localization of these proteins in somatic cells. We demonstrate that TLAR and TIA-1 continuously shuttle between the cytoplasm and the nucléus and belong to the class of shuttling RNA-BPs whose nuclear import is transcription-dependent. We identified RRM2 and the first half of the C-terminal région as important déterminants for the TLAR/TLA-1 nuclear accumulation. In contrast, the nuclear export of TIAR/TIA-1 is mediated by RRM3. Both RRMs contribute to TLAR/TIA-1 nuclear accumulation or export by their RNA-binding capacity. Indeed, whereas mutations of RRM2 most conserved RNP2 octa- or RNPl hexapeptides redistribute TLAR to the cytoplasm, similar modifications in RRM3 abolish TLAR nuclear export. We also show that TLAR/TLA-1 nuclear accumulation is Ran-GTP-dependent pathway, in contrast to its nuclear export which is unaffected by Ran-GTP déplétion and which is independent of the major CRMl exportin pathway. Our study demonstrates the importance of TLAR/TLA-1 RNA-binding domains for their subcellular localization and provides the first evidence for distinct functions of TLAR/TLA-1 second and third RRMs.
With the aim to elucidate TLAR fimetion and/or to identify the components of TLAR nuclear import
machinery, we perform a yeast two hybrid screening using TLAR as bait. Four proteins were
identified. Three of them are RNA-binding proteins which are Tristetraprolin (TTP), cold-inducing
RNA-binding protein (CLRP), poly (rC)-binding protein 1 (PCBPl). The fourth partner corresponds to
the SUMO-conjugating enzyme, Ubc9. The four interactions were confirmed by GST pull-down
experiments. As sumoylation intervenes in several cellular processes including nucleo-cytoplasmic
transport, we focused on the rôle of TLAR-Ubc9 interaction. We also investigated the sumoylation
status of TLAR protein.
1-1. The post-transcriptional régulation of gene expression in eukaryotes and the RNA-binding proteins. 1
1-2. The proteins TIA-1 and TIAR...2
1-2-1. Structure and genomic organization of TIAR and TIA-1...2
1-2-2. Tissue distribution and subcellular localization of TIAR and TIA-1...3
1-2-3. The RNA binding ability of TIAR and TIA-1...4
1- 3. The functions of TIAR and TIA-1 proteins... 4
1-3-1. TIAR and TIA-1 in apoptosis... 4
1-3-2. Rôle of TIAR, TIA-1 in mRNA alternative splicing...5
1-3-3. Rôle of TIAR and TIA-1 as ARE-binding proteins... 7
1- 3-4. Rôle of TIAR, TIA-1 in the formation of cytoplasmic stress granules... 8
2. The nucleo-cytoplasmic transport... 10
2- 1. Nuclear transport receptors... 10
2-2. The directionality of the nuclear-cytoplasmic transport—RanGTPase System... 11
2-3. The directionality of the RanGTPase System...12
2-4. The nuclear-cytoplasmic transport signais...13
2- 4-1. The import signais... 13
2-4-2. The export signais... 14
2-4-3. The shuttling signais...14
2-5. Structure and composition of the NPC... 14
2-6. Translocation through the NPC...15
2-7. The nuclear import mechanism...17
2-7-1. Importin-mediated nuclear import mechanism...17
2-7-1-1. The adaptors of importin-mediated nuclear import...17
2-7-1-2. The different characteristics of the importin-P family members in importin-mediated nuclear import and the cofactors of dissociation of import substrates from import receptors in the nucléus.. 18
2-7-2. Importin-independent nuclear import mechanism...19
2-7-3 The transcription-dependent and independent nuclear import of RNA-binding proteins... 20
2- 8. The Nuclear export mechanism... 21
2-8-1. Rôle of the cliromosome région maintenance 1...21
2-8-2. The other exportins... 22
2- 8-3. The nuclear export of RNAs... ... 23
2-8-3-1. The nuclear export of tRNAs... 23
2-8-3-2. The nuclear export mRNAs... 24
2-8-3-2-1. mRNA export receptors and adaptors...24
2-S-3-2-2. The coupling of nuclear export and mRNA splicing...25
2-8-3-2-3. Coupling of mRNA nuclear export to transcription...26
2-S-3-2-4. Exportin-dependent nuclear export of spécifie mRNA... 27
3. The sumoylation and its functions... 28
3- 1. The SUMO pathway... 28
3-2. SUMO isoforms and SUMO protein stmetme...29
3-3. Enzymes involved in SUMO ligation and release...30
3- 3-1. The SUMO-Activating Enzyme AoslAJba2...30
3-3-2. The SUMO-Conjugating Enzyme Ubc9... 31
3-3-3. SUMO ligases (E3)... 32
3-3-4. SUMO deconjugating enzymes...33
3-4. Substrates and functions of SUMO protein modification...33
Objective and Strategy...34
Results... 35
1. Nuclear-cytoplasmic distribution of TIAR and identification of sequence déterminants
mediating TIAR nuclear-cytoplasmic transport in somatic cells... 35
2. The mechanistn of nuclear-cytoplasmic transport of TIAR... 43
2-1. Nuclear export is independent from the CRMl nuclear export pathway... 43
2-2. TIAR nuclear import but not export is a Ran-GTP-dependent mechanism... 45
3. TIAR and TIA-1 nucleo-cytoplasmic shuttling rely on the same sequence déterminants.... 47
4. Identifîcation of interacting partners of TIAR protein...50
4-1. Yeast two hybrid screening...50
4-2. Vérification of the interactions by a P-galactosidase assay and GST-pull-down experiments... 51
4- 3. Identification of TIAR domains involved in the interaction with CIRT, Ubc9 and TTP...54
5. Rôle of the TIAR-Ubc9 interaction...57
5- 1. Analysis of TIAR subcellular localization upon mislocalization of Ubc9...57
5-2. Is TIAR sumoylated?... 58
Discussion and Perspectives... 62
1. The shuttling properties of TIAR and TIA-1...62
1-1. The nuclear import of TIAR and TIA-1... 62
1-2. The nuclear export of TIAR and TIA-1... 64
2. Identifîcation of TIAR-interacting proteins...65
Annexes... 69
Al. Materials and Methods... 69
Al-1. Materials... 69
Al-2. Methods... 71
A2. References... 75
A3. Articles... 94
Introduction
Introduction
Introduction
1. The characteristics of RNA-binding proteins containing RNA récognition motifs
1-1. The post-transcriptional régulation of gene expression in eukaryotes and the RNA- binding proteins
Post-transcriptional régulation of mRNA plays a very important rôle in eukaryotes. Indeed, after transcription, many events contribute to the formation of mature mRNAs. These include mRNA capping, polyadenylation, splicing (alternative splicing) and nuclear export (Bingham,P. M. et al. 1987; Bernstein, P. and J. Ross. 1989; Clawson, G.A. et al.1985;
Hinnebusch, A. G. 1988). In the cytoplasm, the localization, the stability and translation of mRNAs also contribute to the régulation of gene expression. Ail these mechanisms rely on non-specific and spécifie interactions of proteins with RNA sequences. These proteins are identified as RNA-binding proteins and constitute a major class of proteins controlling gene expression.
The major families of RNA-binding proteins are classified into 6 groups according to their RNA-binding motifs (see table 1) (Chen Y. and Varani G. 2005).
The RRM-containing proteins constitute the most abundant and best characterized family of RNA-binding proteins (RBP) (Varani G. et al., 1998).
RRM are the loosely conserved RNA binding domains which hâve 21 conserved amino acid residues spread across an 80- to 90-amino-acid région, with the most conserved sequences being the RNP-1 octapetide and RNP-2 hexapeptide (Dreyfuss G. et al., 1988). Recently, the three-dimensional structure has been worked out for the RRM of UIA protein (Hoffinan D.W.et al., 1991; Nagai K. et al., 1990), and this structure seems to be conserved in other members of the RRM gene family (Kenan D. J. et al., 1991). Studies of several RRM- containing proteins hâve shown that this motif can confer the ability to bind single-stranded nucleic acids (Kumar A. et al., 1986).
The alignment of the amino acid sequence of 32 RRM revealed the very conserved residues mostly concentrated in the RNP2 and RNPl motifs, thereby allowing the identification of RNP2 and RNPl consensus sequences (see table 2).
The study of RRM RNA-binding affmity suggested that both conserved and non-conserved residues influence RNA binding. The conserved motifs alone generally may not be sufficient to confer a sequence-specific RNA-binding activity. The requirement of flanking sequences
1
Family Examples
Arginine-rich motif Tat, N protein and Rev All-helical proteins
aP protein domain
Rop, LU
RRM KH dsRBD
PAZ
UIA, PAPB, HuD Nova
RNase III
Argonaute RAZ domain Zinc-finger motifs TFIIIA, TISl Id and HIV-1 NC Multimeric motifs TRAP, Pumilio, Sm and Hfq
RNA-targeting enzyme RNase III endonuelease, pseudouridine (y) synthase
RNP -2 RNP -1
RNP CONSENSUS L47 F58 V44 G30N59 L 69 R30 G
s3 F53 A39 Fg3 V70
I30 I33 K22G25 K42 G44 I25
Table 2. The RRM consensus amino acid sequences for RNP-1 and RNP-2 derived from the
Introduction
for RNA binding suggests a function in the maintenance of the protein structure or in the contact with RNA.
RNA’s récognition motifs are diverse in sequence and secondary structure. The basic and solvent-exposed aromatic amino acids conserved in RNP-1 and RNP-2 présent a general complementary surface for interacting with single-stranded RNAs and with unpaired ribonucleotides within higher-order structures. Although sequence-specific RNA récognition may also involve non-conserved surface residues, a common RNA-binding surface shared among RJIM proteins could account for general RNA affinity.
RNA-binging motifs share a common tertiary structure as they are folded into a four-stranded, anti-parallel P-sheet packed against two a-helices in a Pi-ai-p
2-P
3-ct
2-P
4topology (Crowder S.
et al., 2001) (see figurel).
1-2. The proteins TIA-1 and TIAR.
T cell intracellular antigen-1 (TIA-1) was originally identified as a component of cytotoxic granules mediating apoptotic death of CTL (cytotoxic T cell) target cells. The molecular cloning of the cDNA encoding TIA-1 revealed that TIA-1 corresponds to a protein of 40kD belonging to the family of RNA-binding proteins (Tian Q. et al., 1991). TIA-1 related (TIAR) was identified by the molecular cloning of its cDNA by cross-hybridization with a TIA probe (Beck A. R. et al., 1996).
1-2-1. Structure and genomic organization of TIAR and TIA-1
In 1996, Beck cloned the cDNA of TIAR and compared the structure and the genomic organization of the human and murine RRM-type RNA binding proteins, TIA-1 and TIAR (Beck A. R. et al, 1996). This study revealed that the murine TIA-1 (mTIA-1) protein contains 386 amino acids and is 96% identical to the human TIA-1 (hTIA-1); The murine TIAR (mTIAR) protein contains 392 amino acids which are 99% identical to the human TIAR (hTIAR). Ail of them hâve three RRM domains and a 90 or 88 amino acid C-terminal auxiliary domain, respectively. mTIA-1 and mTIAR share 80% identity and présent the highest similarity (91%) in the RRM3 domain. The lowest although significant similarity (50%) is found in the auxiliary domain. Homologs of TIA-1 and TIAR are expressed in
2
Loop-3
Loop-1
Figure 1. Structure model of RRM, derived from NMR structure and consistent with the
crystal structure of the Ul-A (U1 snRNP-A) amino-terminal RRM. The most conserved
suface residues are indicated: aromatic residues at positions 13 and 54, phenylalanine at
position 56 and a basic residue at position 52. These residues hâve been implicated in
RNA binding and lie on two adjacent P-strands comprising the RNP-1 and RNP-2
éléments that are typical of most family members.
Introduction
Drosophila (Brand, S. and Bourbon, H, M.1993) and C.elegans (Wilson R. et al, 1994). It should be noticed that the auxiliary domains of mTIA-1 and mTIAR do not share any significant degree of similarity with other known proteins.
Both TIAR and TIA-1 are encoded as two isoforms. TIA-1 isoforms differ ffom each other by the inclusion or exclusion of an 11 amino acid-sequence which is located at the beginning of the second RRM domain, representing TIA-1 a and TIA-1 b isoforms respectively. In the case of TIAR, a 17 amino acid-sequence is added between the RNP-2 and RNP-1 motifs within RRM domain 1 of the protein in TIARa isoform (see figure 2). Both TIA-1 and TIAR isoforms are generated by alternative splicing events of the mRNA precursors.
The exon-intron organization of the mTIAR and mTIA-1 were established by screening a X,gtl 1 murine genomic DNA library with mTIAR or mTIA-1 cDNA probes containing the entire coding régions. The mTIAR RRM domains 1, 2 and 3 are encoded by exon 1-4, 5-7 and 8- 10, respectively. The c-terminal auxiliary domain is encoded by exons 11 and 12. The 5’- end of TIAR exon 3 encodes the altematively used 17 amino acid of TIARa peptide sequence and the 3’ end of exon 3 encodes amino acids 61-93 which are common to both mTIAR isoforms. In the case of mTIA-1, RRM domains 1, 2 and 3 are encoded by exon 1-4, 5-8 and 9- 11, respectively. The C-terminal auxiliary domain is encoded by exons 12 and 13. The exon 5 encodes the altematively used 11 amino acid of TIA-1 a peptide sequence.
1-2-2. Tissue distribution and subcellular localization of TIAR and TIA-1
mTIA-1 mRNA can be detected in ail tissues except the liver. mTIAR mRNA has a broad tissue distribution. At the protein level, the 40 and 42kDa isoforms of both mTIA-1 and mTIAR are predominantly expressed in brain, spleen and testis. mTIAR is also expressed in the liver and the lung. Both TIAR and TIA-1 proteins are only very weakly expressed or not at ail in heart, skeletal muscles and kidneys. The ratio of mTIA-la and TIA-lb isoforms is 1:1 in ail of the mTIA-1 expressing tissues. In contrast, TIAR short isoform (TIARb) is six-fold more expressed than the long isoform (TIARa) (Beck, A. R. et al, 1996).
In contrast to germinal cells such as oocytes, in which TIAR and TIA-1 are mostly cytoplasmic (Colegrove-Otero L.J. 2005), most somatic cells accumulate these proteins both in the nucléus and the cytoplasm, the highest proportion being nuclear at the equilibrium (Kedersha, N. L.1999).
3
TI A-la peptide
\7
RNP2 RNPl RNP2 RNPl RNP2 RNPl
^ RRM DI--- --- RRM D2--- --- RRMD3---Aux.domain »
TIARa peptide
mTIAR
Figure 2. Schematic structure of mTIA-1 and mTIAR. The relative length and position of the
alternatively used TIA-la and TIARa peptides are indicated, as are the RNP2 and RNPl consensus
motif sequences in each of the three RRM domains (D1-D3).
Introduction
1-2-3. The RNA binding ability of TIAR and TIA-1
What kind of RNA do TIAR and TIA-1 proteins bind? To address this question, Dember et al used in vitro sélection from pools of random RNA sequences to identify RNA sequences that TIAR and TIA-1 bind with high affînity (Dember, L. M. et al, 1996). Both proteins selected RNAs containing one or several short stretches of uridylate residues suggesting that these two proteins hâve similar RNA binding specificities. In vitro, the second RNA binding domain (RRM 2) of both TIAR and TIA-1 médiates the spécifie binding to uridylate-rich RNAs. The RRMl is devoid of RNA-binding activity and the RRM 3 binds RNA without known sequence specificity.
Although the RRM2 is both necessary and sufficient for this interaction, the affinity for the selected RNA increases when the second RNA binding domain of TIAR is expressed together with the first and the third RRM domains. The third RRM of both TIAR and TIA-1 is capable of affmity-precipitating a population of cellular RNAs in contrast to which RRM2 does not hâve this activity.
1-3. The functions of TIAR and TIA-1 proteins
1-3-1. TIAR and TIA-1 in apoptosis
The apoptotic fiinction of TIAR and TIA-1 was first shown for TIA-1 protein. In 1991, Tian et al introduced both purified natural TIA-1 or recombinant TlA-1 into digitonin- permeabilized thymocytes and observed that both proteins induced a dose-dependent appearance of nucleosome-sized DNA fragments characteristic of cells undergoing programmed cell death or apoptosis. This study also revealed that the carboxyl terminus of the proteins was sufficient for this function (Tian,Q. et al.,1991). One year later, the same group reported that TIAR also induced DNA fragmentation in digitonin-permeabilized target cells (Kawakami A. et al., 1992).
The apoptotic functions of TIAR and TIA-1 were fürther charactrized by Taupin and Tian in 1995, respectively. First, Taupin showed that the ubiquitously expressed, nuclear protein TIAR moves rapidly ffom the nucléus to the cytoplasm during Fas-mediated apoptotic cell death. This cytoplasmic translocation précédés the onset of DNA fragmentation (Taupin J. et al., 1995). Concomitantly, Tian reported that TIA-1 is phosphorylated by FAST kinase (Fas-
4
activated serine/threonine kinase) during Fas-mediated apoptosis suggesting that FAST kinase and TIA-1 are components of a molecular cascade that is activated during Fas-mediated apoptosis (Tian Q. et al., 1995).
The mechanism by which TIAR and TIA-1 proteins participate the apoptotic cell death was suggested by Li in 2004. He observed that the over-expression of FAST in Hela cells inhibits Fas- and UV-induced apoptosis. This antiapoptotic effect of FAST is regulated by the interaction with TIA-1 since a FAST mutant lacking TIA-1 binding domain does not inhibit apoptosis. However, the over-expression of recombinant TIA-1 counteracts the antiapoptotic effects of FAST. Since TIA-1 acts as a translational silencer (see below), Li hypothesized that FAST might exert its anti-apoptotic activity by preventing TIA-1-mediated silencing of mRNAs encoding inhibitors of apoptosis (Li W. et al., 2004).
Additionally, TIAR protein was also shown to promote virus-induced apoptosis (Iseni, F. et al., 2002). Iseni found that TIAR is involved in Sendai virus (SeV)-induced apoptosis by binding to the trader RNA of Sendai virus in Hela cells. Sendai virus leader (le) and trader (tr) RNAs are short transcripts generated during abortive antigenome and genome synthesis, respectively. The tr RNA binds to TIAR protein via its 5’ AU-rich sequence LnjLTUAAALTUlJU. TIAR overexpression coupled with SeV infection caused apoptosis in a greater fraction of the Hela cells than did the simple sum of the two, suggesting that SeV induces apoptosis via TIAR.
1-3-2. Rôle of TIAR, TIA-1 in mRNA alternative splicing
Alternative splicing constitûtes a major posttranscriptional mechanism to increase both
protein coding capacity and the regulatory flexibility of the eukaryotic genome. Many RNA-
binding proteins regulate alternative splicing events. The first due for the rôle of TIAR and
TIA-1 in mRNA splicing derived from the fact that NamSp, the most closely related protein to
TIAR and TIA-1 in yeast, was found to be specifically required for Merlp-activated splicing
of several precursor mRNAs, such as MER2 (meiotic recombination) (Engebrecht J. A. et al.,
1991), MER3 (Nakagawa T. and Ogawa H. 1999), and SPO70ATGR225W (putative activator
of meiotic anaphase promoting complex) (Davis C. A. et al., 2000) pre-mRNAs in yeast
(Spingola, M. and Ares, M. Jr. 2000). NamSp binds to intronic uridine-rich sequences
immédiately downstream from rather weak 5’ splice sites (ss) and promotes 5’ ss récognition
of these pre-mRNAs.
Introduction
Based on their phylogenetic relationship with NamSp, the involvement of TIA-1 and TIAR in splicing events was investigated. TIA-1 and TIAR were then found to act as modulators of alternative splicing of varions mRNA precursors. A TIA-1 binding element was found near an intron located at the 5 ’ untranslated région of Drosophila melanogaster gene male-specific- lethal 2 (msl-2) transcripts. In vitro experiments using nuclear extracts of Hela cells indicated that TIA-1 binding was necessary for the récognition of the 5’ss of this intron by U1 snRNP.
The intron splicing dépends on two features of its 5’ss: first, the fifth position of the intron does not correspond to the consensus nucléotide. Secondly, the TIA-1 binding requires the 11 uridine nucléotides downstream ffom 5’ss. The interaction between this U-rich sequence and TIA-1 promotes U1 snRNP recruitment. TIA-1 binding sequence overlaps with sex-lethal regulatory sequence, thereby competing with SXL binding. This compétition favours the formation of male-specific msl-2 transcripts (Fôrch, P. et al. 2000).
The alternative splicing of exon 6 of Fas gene promoted by TIA-1 relies on the same mechanism and results in the accumulation of transcripts that encode products inducing apoptosis, while preventing the accumulation of transcripts encoding an apoptotic inhibitor (Liu C.D. et al., 1994; Cascino I. et al., 1995).
Furthermore, TIAR and TIA-1 were shown to activate the fibroblast growth factor receptor 2 (FGFR-2) alternative exon K-SAM’s splicing. FGFR-2 alternative exon K-SAM is spliced ffom FGFR-2 pre-mRNA in a mutually exclusive, tissue spécifie fashion to generate recep tors with distinct specificity for growth factors (Yayon A. et al., 1992; Miki T. et al., 1992). The splicing requires several sequence éléments, including a uridine-rich sequence downstream ffom the 5’ ss of K-SAM following intron (Del gatto-Konezak, F. et al., 2000). Indeed, K- SAM following intron possesses weak 5’ss and K-SAM also contains an exonic splicing silencer (Del gatto-Konezak, F. et al., 1999). To overcome the activity of the silencer, three activating sequences are required in the downstream intron (Carstens, R. P.,et al., 1995; Del Gatto, F., and Breathnach, R.,1995; Del Gatto, F., et al., 1997). One of these activating sequence lASl, is a U-rich sequence which is located immediately downstream ffom the 5’
ss. The absence of lASl specifically reduces K-SAM exon splicing. TIA-1 binding to ISAl which is adjacent to the 5’ splice site enhances the K-SAM splicing of the 5’ss site in a U1 snRNP dependent-manner.
Finally, TIAR and TIA-1 were also shown to promote the alternative splicing of their own pre-mRNA (Le Guiner, C. et al., 2001).
TIAR and TIA-1 proteins constitute major pre-mRNA splicing regulators. It was therefore important to identify the domains of TIAR and TIA-1 required for pre-mRNA récognition.
6
Figur3. Schematic representatfon of TIA-l-mediated recruitment of U1 snRNP to weak 5’ ss foUowed by
uridine-rich sequences. 1,2, 3 indicate RRMl, 2,3 and Q is the auxilary domain.
Introduction
Fôrch (Fôrch, P. et al. 2002) revealed that for the mls-2 pre-mRNA récognition via U1 small nuclear ribonucleoprotein (U1 snRNP) splicing, the domain RRM 2 and the RRM 3 are sufficient for the pre-mRNA binding while the C-terminal domain contributes to the protein- protein interaction. The binding of individual RRM domains to the pre-mRNA was only détectable with RRM 2. The reconstituted protein which contains RRM 2 and RRM 3 bound the pre-mRNA with the same affinity as the full-length protein. Analysis of the activity of TIA-1 délétion mutants in promoting the recruitment of U1 snRNP to the msl-2 5’ ss région showed that neither the individual RRM nor the combination of RRM 2 and RRM3 promoted cross-linking of U1 snRNP to mls-2 5’ ss. However, the mutant containing the three RRMs significantly increased U1 snRNP binding. This suggested that RRMl domain of TIA-1 facilitâtes U1 snRNP recruitment. The observation that the protein containing the three RRM domains binds the U1 snRNP less efficiently than the full-length protein suggests that the C- terminal domain of TIA-1 also contributes to U1 snRNP recruitment. Co- immunoprecipitation experiments revealed a direct and spécifie interaction between the N- terminal région of the U1 snRNP-Ul-C and the C-terminal domain of TIA-1. This interaction indicates that TIA-1 binds to the 5’ss and is involved in a direct interaction with Ul-C to promote U1 snRNP recruitment (see figure 3).
1-3-3. Rôle of TIAR and TIA-1 as ARE-binding proteins
Genes encoding cytokines, lymphokines and proto-oncogenes are transiently expressed in response to extracellular stimuli and are regulated at multiple levels. Indeed, these genes are regulated at the transcriptional level, but also at post-transcriptional levels, such as mRNA stability and translation. A common feature of these genes is the presence of adenosine- uridine-rich éléments (ARE/AU-rich élément) located in their 3’ untranslated régions (3’UTR). These éléments are composed of successive or overlapping AUUUA pentamers (Shaw G. and Kamen R., 1986; Caput D. et al., 1986; Chen C. Y. and Shyu A. B., 1995; Ross J., 1995; Wilusz C. J. et al., 2001). Several genes encoding cytokines like Tumor Necrosis a (TNF) contain such éléments.
7
As mentioned above, although TIAR and TIA-1 mostly accumulate in the nucléus, a significant proportion of these proteins is found in the cytoplasm, suggesting cytoplasmic functions for these proteins.
ARE-mediated regulating mechanisms hâve been particularly well studied for Tumor Necrosis Factor a (TNF-a) mRNA. The translation of tumor necrosis factor a (TNF-a) mRNA is tightly regulated in monocytes/macrophages. In unstimulated cells, translation of TNF-a mRNA is repressed via the AU-rich éléments (ARE) in 3’ UTR of TNF-a mRNA.
Upon stimulating with lipopolysaccharides (LPS), this repression is overcome. With the aim to identify TNF ARE binding proteins, Gueydan C. et al performed EMSAs with rihoprobes corresponding to TNF-a mRNA 3’UTR and macrophage cytoplasmic extracts. These experiments revealed two protein complexes interacting with TNF a ARE sequence. In order to identify the proteins involved in both complexes, a macrophage cDNA expression library was screened with TNF-a mRNA 3’ UTR rihoprobes containing or not the ARE. This screening revealed TIAR as a TNF ARE binding protein (Gueydan C., t al., 1999).
In 2000, Piecyk et al. addressed the fünction of TIA-1 in the ARE-mediated TNF-a post- translational régulation by generating TIA-1 déficient mice. This study revealed that TIA-1 protein acts as a translational silencer that selectively régulâtes the expression of TNF-a.
Indeed, macrophages of TIA-1 déficient mice produced significantly more TNF in response to LPS as compared to wild-type cells. Since neither the accumulation nor the half-life of TNF-a transcript was modified in TIA-1-/- macrophages, the association of TNF-a mRNA to the translational machinery was compared in LPS-induced wild-type and TIA-1-déficient mecrophages. This experiment revealed an increase of TNF mRNA in polysomes in TIA-1- deficient cells, thereby revealing TIA-1 as a translational silencer of TNF mRNA (Piecyk M.
et al., 2000).
TIAR and TIA-1 were also shown to repress the translation of human matrix métal oproteinases-13 (hMMP-13) (Yu Q. et al., 2003), cyclooxygenase-2 (COX-2) (Dixon D. A. et al., 2000), and mammalian P
2-adrenergic receptor mRNAs (Kandasamy K. et al., 2005).
1-3-4. Rôle of TIAR, TIA-1 in the formation of cytoplasmic stress granules
The stress granules (SGs) are phase-dense particles produced by eukaryotic cells in response
to environmental stresses (Nover L. et al., 1983, 1989; Scharf K. D. et al., 1998; Kedersha N.
Introduction
L. et al., 1999). First described in plants, SGs can be induced by a wide variety of stresses such as oxidative stress, beat shock, UV radiation, osmotic shock, ER stress and viral infection.
In response to environmental stress, eukaryotic cells reduce their protein synthesis via post- transcriptional mechanism relying on the blockade of translation initiation. TIAR and TIA-1, were shown to participate to the formation of stress granules with other RNA-binding proteins, such as poly (A)-binding protein (PABP-I), human autoantigen R (HuR) (Gallouzi I.
E. et al., 2000) and tristetraprolin (TTP) (Kedersha N. et al., 2002). TIAR and TIA-I were first found to co-localize with poly (A)^ RNA at cytoplasmic foci that resembled SGs harbouring untranslated mRNAs in beat shocked plant cells (Nover L. et al., 1989, 1983;
Scharf K. D. et al., 1998). In 1999, Kedersha et al. showed that the assembly of TIAR and TIA-I-containing SGs is initiated by the phosphorylation of eIF-2a and that a TIA-1 mutant lacking the RRMs inhibited SG formation. These results suggest that TIAR and TIA-1 act downstream ffom the phosphorylation of eIF-2a to promote the séquestration of untranslated mRNAs at SGs.
SGs correspond to réservoirs of untranslated mRNAs which are in equilibrium with the polysomes. Indeed, while the formation of SGs is efficiently inhibited by emetine, a drug blocking translation élongation, puromycin which blocks translation intiniation favours the formation of SGs.
Recently, accumulating évidences revealed more and more clearly the mechanism and the flinctions of TIAR and TIA-1 in participating to the formation of SGs:
(1) TIAR and TIA-1 promote the assembly of eIF2/eIF5-deficient preinitiation complexes that are routed to SGs (Kedersha N. et al., 2002). The characterization of SG revealed additional SG components such as the small ribosomal subunits. Translation is normally initiated when the small ribosomal subunit and its associated initiation factors are recruited to a capped mRNA transcript to form a 48S complex. Hydrolysis of eIF2-associated GTP by eIF5 displaces the early initiation factors, allowing the binding of the large ribosomal subunit.
Repeated cycles of successful initiation convert an mRNA into a polysome. In stressed cells, activation of one or more eIF2a kinases results in the phosphoylation of eEF-2a (Kimball S. R.
2001; Williams D. D. 2001), which consequently inhibits eIF-2B, the GTP/GDP exchange factor that charges the eEF-2 temary complex (eIF2-GTP-tRNAi''^®*). Immunofluorescence microcopy to compare the subcellular localization of TIAR or TIA-1-containing SGs and individual components of the translation initiation apparatus in arsenite-stressed DU 145 cells
9
mammals number"
(■alternative names) •
Importin ^ NP.002256 proteins containing basic stretchessuchas: nbo- import PTAC97 )
somal protems; Jil V Rev ana iat, illLV Rex;
cyclin B. coie histones and otbers; histone H1 and tibosomal protems in conjunction with Importin 7 importin a
femily
NLS proteins impon
snurportin 1 U snRNPs impon
XRlPa replicanon protein A import
Transportin 1 (Kap^a. TRNl)
NP_002261 mRNA binding proteins, ribosomal proteins import Transportin 2
(Kap/SEb, TRN2)
NP_038461 overlapping function with transportin 1, suggested to be involved in niRNA export
import export?
Transportin SRI (TRN-SRl)
AAD38537 SR proteins import
Transportin SR2 (TRN-SR2)
NP_036602 SR proteins containing phosphorylated RS domains import
Importin 4a AAL55S22 ribosomal protein S3a import
Iraportin 4b NP_078934 ? import
Importin 5 (Kap^)
CAA70103 Ribosomal proteins, coie histones and other basic proteins
import Importin? NP_006382 ribosomal proteins, core histones and other basic
proteins; histone H1 and ribosomal proteins in conjunction with importin
p
import
Importin 8 NP_006381 SRP19 import
Importin 9 NP_060555 core histones with a preference for H2B, import of ribosomal proteins
import
Importin 11 AAF21936 UbcM2, ribosomal protein L12 import
Importin 13 NP_055467 UBC9,Y14
elElA
import export CRMl/Expor-
tin 1 pCPOl)
NP_003391 proteins with leucine-iich NES sequences, snurportin 1
export
PHAX NES-contaimng adaptor for U snRNAs export EUVRev NES-containing adaptor for unspliced HTV-RNAs export NMD3 potential NES-containing adaptor for pre-60S
ribosomal subunits
export
CAS NP_001307 importin or family members export
Exportin-t (Xpo-t)
NP_009166 tRNAs export
Exportin 4 NP_065252 elFSA export
Exportm 5 NP_065801 tRNA and perhaps other structured RNAs; ILF3 and eEFlA (RNA-mediated binding)
export
RanBPâ AAC14260 ? ?
RanBPie NP_055839 ? ?
? ?
Introduction
(prostate cancer cells) showed that TIAR and TIA-1 co-localize only with the eEF2/eIF5- deficient preinitiation complexes.
(2) The overexpression of TIA-IARRM, a RRM-truncated mutant sequestering endogenous TIA-1 and TIAR, prevents SG assembly and promotes the expression of cotransfected reporter genes in the stressed cells (Kedersha N. et al., 2000). In contrast, the overexpression of TIA-I represses the expression of cotransfected reporter genes in unstressed cells (Kedersha N. et al., 2000). These observations suggest that TIAR and TIA-I might regulate the equilibrium between polysomal mRNPs in both stressed and unstressed cells.
In conclusion, TIAR and TIA-I, like several other RNA-binding proteins, combine several biological fiinctions both in the nucléus and the cytoplasm.
2. The nuclear-cytoplasmic transport
In eukaryotes, different processes of gene expression occur in varions compartments, and complex mechanisms exist to transport macromolecules across these compartments. The transport of macromolecules between the nucléus and the cytoplasm is mediated by soluble transport receptors via the channel of the nuclear pore complexes (NPCs). The transport receptors are able to associate with components of the NPC and to bind to cargo molécules that need to be translocated across the pore. Cargo molécules are recognized via import or export targeting signais, which are referred to as nuclear localization signais (NLSs) or nuclear export signal (NESs) respectively (Gôrlich D. and Kutay U. 1999; Mattaj, I. W. and Englmeier L. 1998; Nakielny S. and Dreyfliss G. 1999; Weis K. 1998).
The cargoes of nuclear-cytoplasmic transport are either proteins or several kinds of RNAs (tRNA, rRNA, mRNA, snRNA and microRNA) (Wozniak R. W. et al., 1998; Stutz F. and Izaurralde E. 2003; Lund E. et al., 2004).
2-1. Nuclear transport receptors
Different classes of shuttling transport receptors médiate nucleocytoplasmic transport. These receptors are defmed by their ability to directly contact the NPC, thereby facilitating the passage of cargo molécules (see table 3).
10
BanGTP HanGDP
importin
impoli substrate dissociation
RanGTP
export substrate binding
Figure 4. Control of cargo binding to importins and exportins by Ran.
Exportin
Introduction
The largest class of nuclear transport receptors is the superfamily of importin P-related proteins which ail proteins share significant homology with the importin receptor protein, importin p. Members of this family hâve been classified as importins or exportins on the basis of the direction they carry their cargo (Strom A. C. and Weis K. 2001). A second group of transport receptors is NTF2, pi5 (a cofactor in messenger RNA export) and its homologues (Katahira J. et al., 1999). The third group of nuclear transport receptors includes proteins related to the mRNA export receptor TAP (Izaurralde E. 2002).
Importin P-related proteins account for the majority of cargos flow through the NPC. Fourteen members of this family hâve been identified in yeast and more than twenty members in higher eukaryotes. Most importin P-like receptors appear specialized in transporting their cargo in only one direction. A few of them can médiate both import and export (Mingot J.M. et al., 2001; Siomi M.C. et al., 1997; Gallouzi I. E. et al., 2001).
Ail the members of the importin P-related superfamily contain an N-terminal Ran-GTP binding domain as well as a C-terminal cargo binding domain. The association with Ran-GTP Controls the direction of the cargo transport. It was suggested that the binding of Ran-GTP to the N-terminal of importin p-related proteins induces a conformational change to the C- terminal cargo-binding domain of importins and exportins (Chook Y. M. et al., 2002).
2-2. The directionality of the nuclear-cytoplasmic transport—RanGTPase System
The directionality of nuclear transport relies on the asymmetrical distribution of Ran-GTP between the nucléus and cytoplasm. The maintenance of this asymmetry involves the spatial distribution of regulators that control Ran’s nucléotide State and its subcellular location.
The directionality of Ran-GTP to the transport reaction is described in the figure 4.
Uptake and release of cargo is restricted to the appropriate cellular compartment by the interaction of the receptor with Ran, a Ras-related GTPase that is associated either with GTP or GDP. The binding of Ran to GTP or GDP is regulated by several factors. The conversion of Ran-GDP into Ran-GTP requires the activity of the guanine nucléotide exchange factor (RanGEF in mammal, also known as RCCl in metazoans and Prp20p in yeast). In the nucléus, the Ran-GTP concentration is kept high by RanGEF; On the other hand, in the cytoplasm, the GTPase activity of Ran leading to the conversion of Ran-GTP into Ran-GDP
11
translocation through the NPC, RanGTP is converted to RanGDP with the aid of RanBPl/RanBP2
in the cytoplasm and dissociâtes from the Importin P family molécules. NTF2 carries RanGDP into
the nucléus. In the nucléus, the NTF2/Ran complex dissociâtes and the nucléotide exchange is
mediated by RCC 1 to generate RanGTP.
Introduction
is activated by the combined action of RanGTPase activating protein-1 (RanGAPl or Rnalp in yeast) (Bischoff F. R. et al., 1994, 1995) with the Ran-binding proteins 1 and 2 (RanBPl andRanBP2).
Importins bind a cargo in the cytoplasm in the absence of Ran-GTP. Upon arrivai in the nucléus, the importin-cargo complexes associate with Ran-GTP and this association induces a conformational change in the cargo-binding domain of the importin leading to the cargo release. The importin /Ran-GTP complex translocates back to the cytoplasm where RanGAPl dissociâtes Ran-GTP ffom the importin by promoting GTP hydrolysis into GDP with the assistance of RanBPl and/or RanBP2.
Exportins and their cargos form export complexes in the nucléus in the presence of Ran-GTP.
After export to the cytoplasm, the export complexes are disassembled by RanGAPl and RanBPl/RanBP2. The exportins reenter the nucléus on their own.
2-3. The directionality of the RanGTPase System
The recycling of Ran-GTP in the different cellular compartments relies on several regulatory factors (see figure 5).
—In the nucléus, RanGEF (Guanine nucléotide Exchange Factor), RCCl (RanGEF in metazoans) and Prp20p (RanGEF in yeast), stimulate the conversion of Ran-GDP to Ran- GTP. RanGEF’s localization is restricted to the nucléus (Ohtsubo M. et al., 1989; Nemergut M. E. et al., 2001).
—In the cytoplasm, RanGAPl (Ran-GTPase Activating Protein), Rnalp (yeast RanGAPl) activate the GTPase activity of Ran, thereby promoting the hydrolysis of Ran-GTP into Ran- GDP. RanGAPl is confmed to the cytoplasm (Matunis M. J. et al., 1996; Mahajan R. et al., 1997. Saitoh H. et al., 1998). It has been shown that Rnalp contains a nuclear export signal that binds to the export factor Xpolp, which rapidly exports any Rnalp that might enter the nucléus, thereby ensuring the concentration of Rnalp in the cytoplasm (Feng W. et al., 1995).
—The dissociation of Ran-GTP ffom importins (or exportins) in the cytoplasm is promoted by another family of Ran-binding proteins, namcd RanBPl and RanBP2 (known also as Nup358) (Coutavas E. et al., 1993; Beddow A. L. et al., 1995; Bischoff F. R. et al., 1995;
Ouspenski I. I. et al., 1995; Wu J. et al., 1995; Yokoyama N. et al., 1995). RanBPl and RanBP2/Nup358 stimulate the activity of RanGAPl by binding the Ran-GTP/importins complexes and rendering them more accessible to RanGAPl.
12
—In order to avoid the déplétion of nuclear Ran during the nuclear transport, Ran-GDP rapidly reenters into the nucléus via another transport factor NTF2 (Nuclear transport factor 2)( Ribbeck K. et al., 1998; Smith A., et al., 1998). NTF2 is unrelated to the importin P family and binds only to Ran-GDP. Upon entering the nucléus, Ran-GDP is dissociated ffom NTF2, and Ran’s GDP is replaced by GTP. The dissociation of NTF2 ffom Ran-GDP and the replacement of Ran-GDP to Ran-GTP in the nucléus are both fostered by RanGEF.
—The nuclear-cytoplasmic transport is highly dépendent on the cell energy State. For example, ATP déplétion induces a major drop of the nuclear ffee GTP concentration, thereby decreasing the nuclear concentration of Ran-GTP. Therefore, a cellular ATP déplétion induced hy the blockade of glycolysis and the respiratory chain results in the rapid inhibition of the Ran-dependent nucleo-cytoplasmic traffic (Schwoebel E. D. et al., 2002)..
2-4. The nuclear-cytoplasmic transport signais
Nuclear transport is a highly spécifie process that relies on the récognition of cargos by structural features présent within the molécules. These transport ‘signais’ can be recognized by transport receptors (either importins or exportins), directly or indirectly. Import and export signais are different.
2-4-1. The import signais
The most common motifs are the ‘classical’ nuclear localization signais (NLS). Three
categories of NLS can be distinguished. In the first class, the NLS is characterized by its
enrichment in basic residues. The prototype NLS is the one of in SV40 T-antigen and whose
sequence is PKKKRKVE (Kalderdon, D. et al., 1984). This type of NLS has been named
mono partite NLS in opposition to the bipartite. This second class corresponds to NLS
composed of two clusters of positively charged amino acid residues that are separated hy a
spacer of 10-12 amino acids. The prototype of this class of NLS is the one found in the
Xenopus laevis, nucleoplasmin whose sequence is KRPAATIGCAGQAKKKKLD (Robbins
J., et al 1991). A third type of NLS includes those resembling that of the yeast homeodomain
containing protein, Mata2 (Hall, M. N. et al., 1990) in which charged/polar residues are
interspersed with non-polar residues. This is also the case of the protooncogene c-myc NLS
Table 4, Shuttling signal sequences in nucleo-cytoplasmic transport
Shuttling signal Example substrates
Sequence Transport
receptor(s)
M9 domain hnRNPAl YNDFGNYNNOSSN Transportin
BIB domain rpL23a VHSHKKKKIRTSPTFRRP
KTI.R
Transportin, Imp5, Imp7 ImpP RS domain SR proteins Phosphorvlated RS Transportin
SR2
Leucine-rich NES
HIV Rev, PKI
Consensus: L-X
2-
3-
n..IM.F.M'> CRMl
KNS
sequence hnRNP K YDRRGRPGDRYDGMVG
FSAO ?
HNS
sequence HuR RRFGGPVHHQAQRFR
FSPM
Transportin 1 and transportin
2RS domain: serine/argininedomain SR protein: splicing repressor.
PKI: protein kinase inhibitor 1.
HNS: HuR nuclear cytoplasmic shuttling sequence.
hnRNP: heterogeneous nuclear ribonucleoprotein.
KNS: hnRNPK shuttling domain.
HuR: ELAV (embryonic léthal abnormal vision)-like protein
(PAAKRVKLD) in which proline and aspartic acid residues are at either side of the central basic cluster (Makkerh J. P. et al., 1996).
Ail three classes of NLS are believed to be recognized specifically by the nuclear transport receptor adaptors of the importin-a family (Efthymiadis A. et al., 1997; Smith H. M. et al., 1997; Briggs L. J. et al., 1998; Hu W. and Jans D. A., 1999; Hübner S. et al., 1999; Nadler S.G. étal., 1997).
2-4-2. The export signais
Nuclear export signal (NES) sequences are relatively short oligopeptides. They are enriched in leucine residues. A classical NES is composed of the peptide sequence X-R(
2.
4)-X-R
2-X-R- X, where X is leucine, isoleucine or valine and R is any amino acid (Hope T. J. et al., 1999;
Henderson B. R. and Eleftheriou A. 2000).
2-4-3. The shuttling signais
Some nuclear import signais hâve been shown to be involved in export activity as well.
Therefore, they are called shuttling signais. So far, such signais hâve been identified in few shuttling proteins (see table 4). It should be mentioned that these shuttling signais do not share obvions similarities, thereby suggesting distinct mechanisms of nuclear import and export.
2-5. Structure and composition of the NPC
The nuclear pore complex (NPC) is the only site of bidirectional exchange between the cytoplasm and the nucléus (Vasu S. K. and Forbes D. J., 2001). The three-dimensional structure of the NPC has been determined by the analysis of électron microscopie images and is illustrated in figure 6.
Recent mass spectrometry data revealed that NPC consists of about 30 different proteins
called nucleoporins in both yeast and vertebrates (Cronshaw J.M. et al., 2002). Many of them
are at a copy number of eight or multiples of eight (Fahrenkrog B. et al., 2001; Rout M. et al..
Tpr/ Mlp1/2
Nucleoplasm
Cytoplasm
Cytoplasmic filaments
Figure 6. The schematic représentation of the nuclear pore.
Viewed along the axis of its central channel, the NPC exhibits an octagonal symmetry. The main part of the pore complex forms a cylindrical structure, composed of spoke-ring complexes sandwiched between nuclear and cytoplasmic ring structures that are embedded in the nuclear envelope. The NPC is assymmetric. Attached to the cytoplasmic face, are flexible filaments protruding into the cytosol, while fibrils emanating on the nuclear side converge at their ends to form a cage-like structure referred to as the nuclear basket. (Fabre E. and huit E., 1997; Allen T.
D., et al., 2000; Fahrenkrog B, et al., 2001). Pore-associated filaments (dotted Unes) extend from the basket of the pore into the nucléus and contain the proteins MLPl/2 in yeast and Tpr in vertebrates (Paddy M. R. 1998).
« ^ / / I M
Nuclear basket
Central
transporter
2001). It can be estimated that a mammalian pore consists of a minimum number of 400 individual proteins.
NPCs from low and high eukaryotes are structurally and functionally similar. About two- thirds of the 30 different nucleoporins are conserved between yeast and mammals (Vasu S.K.
and Forbes D. J., 2001; Cronshaw J. M. et al., 2002). One important feature common of many nucleoporins from ail species is that they contain phenylalanine-glycine (FG) repeats, or FxFG and GLFG repeats, which are separated by hydrophilic linkers of variable sequence and length. These FG repeats provide binding sites for nuclear transport receptors such as importins, exportins, NTF2 (Bayliss R. et al.,2000; Bayliss R., et al.,2002). It bas been estimated that there are 1000 copies of such repeats within one NPC (Bayliss R., et al., 1999).
The high copy number of FG repeats was shown to favour the formation of the complex with nuclear transport receptors (Bayliss R., et al., 1999; Ribbeck K., et al., 2001).
FG repeats may function to concentrate carrier cargo complexes at the NPC entrance (Rout M. et al., 2000) or participate in a sequence of docking and undocking interactions when the carrier cargo complexes transit through NPCs (Allen N. P. et al., 2001; Radu A. et al., 1995;
Rexach M. et al., 1995; Stewart M. et al., 2001). Altematively, FG-repeats may form a meshwork within the central channel that is permeable only to molécules that interact with FG repeats (Ribbeck K. et al., 2001). Rout and Wente (Rout M. P. et al., 1994) found that the distribution of different FG-repeats subfamilies (FG, FxFG and GLFG) within NPCs is different (Rout M. P. et al., 2000; Stoffler D. et al., 1999). For example, in S. cerevisiae, there are 13 different FG-repeats. The GLFG-repeats are found on both sides of the NPC, some FG-repeats are exclusively on the cytoplasmic side, and some FxFG-repeats are exclusively on the nuclear side (Rout M. P. et al., 2000). FG-repeats show distinct binding préférences for different carriers (Allen N. P. et al., 2001), suggesting that potentially some may hâve specialized functions (Bayliss R. et al., 2000; Bayliss R. et al., 2002; lovine M. K.
et al., 1997; Shah S. et al., 1998; Allen N. P. et al., 2001; Ryan K. J. et al., 2000; Damelin M.
et al., 2000).
2-6. Translocation through the NPC
The NPC allows the passive diffusion of small molécules but at the same time restricts the
movement of macromolecules between the cytoplasmic and nuclear compartments. Diffusion
becomes inefficient if the molecular weight of the diffusive species approaches 20-40 kD. The
1. Brownian affinity gating
2. Sélective phase
3. Oily-spaghetti
Figure 7. Models of translocation through the NPC.
diffusion barrier is made by a single transport ‘gâte’ located in the central domain of the NPC.
Its estimated diameter is 10 nm (Fried H. and Kutay U. 2003).
NPC allows the rapid translocation of cargo having appropriate signais that interact with nuclear transport receptors and this translocation is energy-independent. These substrates can be as large as ribosomal particles and viral capsids (~35 nm diameter) (Panté N. and Kann M., 2002 ).
Three models bave been proposed to explain how transport receptor-nucleoporin interactions promote sélective transport (see figure 7) (Fried H. and Kutay U. 2003).
The ‘Brownian affmity gâte’ model (Rout M. P. et al., 2001) implies that a transport receptor/substrate complex initially binds to the cytoplasmic filament of the NPC. It has been postulated that there exists a high abundance of FxFG-containing binding sites on the cytoplasmic and nuclear extensions of the NPC based on positional mapping of individual yeast nucleoporins. The binding of the transport receptor/substrate complex in the proximity of the central channel of the NPC increases the likelihood of entrance of the complex into the interior of the NPC. Transfer through the channel would be by Brownian motion.
The ‘sélective phase’ model (Ribbeck K. et al., 2001) predicts the existence of a flexible meshwork of nucleoporins that interact with each other through their FxFG repeats. The meshwork is positioned centrally within the NPC and forms a permeability barrier for insert cargo. The interactions between nucleoporins in the mesh can be broken transiently by nuclear transport receptors, which can partition into the lattice by engaging into hydrophobie interactions with FxFG repeats. This model could explain transport of large cargo but the existence of such a meshwork has not yet been proven.
According to the ‘oily-spaghetti’ model (Macara LG. 2001), the NPC is an open structure with a central channel width of 10 nm. Flexible FxFG-containing nucleoporins coat the wall of this channel and provide binding sites for the passage of nuclear transport receptors.
Transport receptors can push aside the loose nucleoporin chains during passage. Transient association with FxFG repeats and random motion would achieve translocation.
In 2005, Peters R. suggested another model of translocation through the nuclear pore complex, named the reduction-of-dimensionality model (Peters R. 2005) (see figure 8).
Ail these models assume that translocation through the NPC occurs by random or facilitated
diffusion. As indicated before, the directionality of nuclear transport is the resuit of
Figure 8. Reduction-of-dimensionality Model of translocation through the nuclear pore
Cytoplasmic filaments
Threoding area SelecUvity filter
Nudear fjlaments FG surtace
A.The NPC is assumed to contain a large channel.The pore wall is Phenylalanine glycine (FG) motif, - i.e. - binding sites for — nuclear transport receptors. FG repeats ffom a surface (red) which extends onto filaments. A selectivity filter, consisting of a loose network of unfolded hydrophilic peptide chains, occupies the center of the channel.
Entranc© ---—i^—
________
\______________
FH amonts Central channel
_________ I_______________ L
Exit RIaments
B. The FG surface, although cohérent, has different sections. The filament section serves as antenna, collecting transport receptors and complexes ffom the aqueous phase. The section at the channel entrance threads transport receptors and complexes into the selectivity filter. The section distal of the selectivity filter leads transport receptors and complexes to the channel exiL
Small and large
neutral molécules
C. Neutal molécules, which do not bind to FG motifs, permeate the pore by diffusion in the aqueous phase. The selectivity filter resricts their passage to a tube of 8-lOnm in diameter.
Transport
complexes
