Nuclear export signals: small domains with large impact Engelsma, D.H.

Hele tekst

(1)Nuclear export signals: small domains with large impact Engelsma, D.H.. Citation Engelsma, D. H. (2008, October 16). Nuclear export signals: small domains with large impact. Retrieved from https://hdl.handle.net/1887/13258 Version:. Corrected Publisher’s Version. License:. Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden. Downloaded from:. https://hdl.handle.net/1887/13258. Note: To cite this publication please use the final published version (if applicable)..

(2) Chapter 1. General Introduction. 9.

(3)

(4) Chapter 1 General Introduction. 1. Nuclear transport The word eukaryote derives from the Greek words ευ (good/true) and κάρυον (nut, nucleus), indica ng that a eukaryo c cell contains a “real” nucleus, opposed to prokaryotes that do not have a nucleus. The nucleus separates the genomic DNA from the cytoplasm by a double membrane, termed the nuclear envelope (NE), while prokaryotes have their DNA freely floa ng in the cytoplasm. One of the func ons of the NE may be to form an addi onal protec ve layer to shield the precious genomic informa on from environmental damage. Furthermore, the NE spa ally separates DNA transcrip on in the nucleus from protein transla on in the cytoplasm and as a consequence these two processes can be be er regulated independently. To allow macromolecular communica on between both compartments, the NE is penetrated by large protein complexes that form aqueous channels, termed nuclear pore complexes (NPCs). These pores allow diffusion of small molecules as well as facilitated transport of proteins, RNAs and ribonucleoproteins (RNPs). This facilitated transport is achieved by soluble transport receptors, termed karyopherins that shu le between both compartments, picking up cargo at one side of the NE and releasing it at the other side. These transport receptors are divided into impor ns, impor ng cargo into the nucleus and expor ns that transport cargo out of the nucleus (reviewed in (Weis, 2003)) (Figure 1). Nuclear transport is an extremely quick and efficient process. Each NPC has been calculated to transport up to one thousand molecules per second (Ribbeck and Gorlich, 2001; Smith et al., 2002). Proteins are translated in the cytoplasm and as a consequence proteins that func on in the nucleus need to transverse the NE. In contrast, nuclear export is needed for the regula on of nuclear proteins, like transcrip on factors, or for recycling of import adaptors (see Figure 1). However, the largest group of molecules that have to be exported from the nucleus are RNA molecules and protein/RNA complexes, termed ribonucleoproteins (RNP). Some karyopherins transport a plethora of cargos, while others are specialized in the transport of one specific molecule or group of molecules (reviewed in Gorlich, 1999 #174}). In brief, the direc on of transport is governed by the small Ras-like GTPase Ran, of which the concentra on of the GTP-bound form is high in the nucleus and low in the cytoplasm. This will be further discussed in 1.2. Expor ns bind their cargo only in the presence of RanGTP and therefore only in the nucleus, whereas impor ns bind their cargo in the absence of RanGTP and release their cargo upon RanGTP binding. A er dissoci-. a ng from the cargo in the opposite compartment, the empty karyopherin recycles back to the original compartment (Figure 1). Both import and export pathways have one very general transport receptor that transpor¬ts a large number of cargo. The impor n responsible for this general import is Impor n β (also termed karyopherin β) and the expor n is termed CRM1 or expor n 1 (see 1.3.3). Most of their cargo is recognized by the exposure of short signals, termed nuclear localiza on signals (NLS) and nuclear export signal (NES), respec vely (reviewed in (Weis, 2003).. 1.2 Ran system Ran is a member of the small Ras-like GTPases. Like all GTPases, Ran cycles between an ‘ac ve’ GTP-bound form and an ‘inac ve’ GDP-bound form and undergoes significant conforma onal changes upon hydrolysis (Scheffzek et al., 1995). Only in its ac ve GTP-bound form, can it bind transport receptors (Figure 2). RCC1 (regulator of chromosome condensa on 1) is Ran’s guanine nucleo de exchange factor (GEF), exchanging RanGDP to RanGTP. Ran possesses very weak GTPase ac vity, but this is s mulated by several orders of magnitude by its specific GTPase ac va ng protein, RanGAP1 (Klebe et al., 1995). Ran binding protein 1 (RanBP1) further boosts the hydrolysis of RanGTP. By binding RanGTP, RanBP1 s mulates a brief dissocia on of GTP, facilita ng hydrolysis by RanGAP. Perhaps more importantly, RanBP1 is essen al for dissocia ng RanGTP from nuclear transport receptors (Bischoff and Gorlich, 1997)(Figure 2). In interphase cells, RanGTP is present in a steep gradient over the nuclear envelope, being at a high concentra on in the nucleus and low in the cytoplasm (Kalab et al., 2002). This gradient is maintained by a separated localiza on of its cofactors. RCC1 is mainly nuclear, as it is bound to chroma n, whereas the RanGTP hydrolyzing factors RanGAP1 and RanBP1 are restricted to the cytoplasm (Izaurralde et al., 1997). Moreover, SUMOylated RanGAP is a ached to the cytoplasmic side of the NPC, which also has RanBP1-like domains (Wu et al., 1995). This will lead to an immediate GTP hydrolysis upon arrival of export complexes at the cytoplasmic face of the NPC. RanGDP in turn is recycled back into the nucleus by the import receptor NTF2 (nuclear transport factor 2) (Figure 2) (Ribbeck et al., 1998). But how does the RanGTP gradient regulate the direc on of transport? This is achieved by the Ran-binding proper es of karyopherins. Impor ns bind their cargo in the cytoplasm and are translocated to the nucleus where RanGTP binding to the Ran binding domain of impor ns will lead to dissocia on of cargo (Figure 1). The opposite holds true for expor ns, as the expor n-cargo interac on is rather weak and needs to be s mulated by RanGTP. 11.

(5) General Introduction. Figure 1. Receptor-mediated nucleocytoplasmic transport. (Le ) CRM1-mediated nuclear export. In the cytoplasm, the export receptor CRM1 binds to NES-containing proteins. This interac on is stabilized in Ran in its GTP-bound form. Subsequently, the trimeric complex translocates to the nucleoplasm, where the complex is dissociated upon GTP hydrolysis. CRM1 can recycle to the cytoplasm through hydrophobic interac ons with the NPC. Other expor ng-mediated transport occurs via the same principle. (Right) Impor n β mediated nuclear import. Impor n β imports NLS-mediated cargo to the nucleoplasm. This interac on is o en bridged by the import adaptor Impor n α. In the nucleus high RanGTP concentra ons are present and RanGTP binding to Impor n β causes release of both Impor n α and cargo. The stable Impor n β/RanGTP complex is recycled back to the cytoplasm, where RanGTP hydrolysis releases Impor n β. Impor n α cannot recycle to the cytoplasm by itself unless supported by the export receptor CAS and RanGTP to get back to the cytoplasm.. binding. Upon arrival in the cytoplasm, Ran will either be dissociated at the NPC or, as a backup mechanism, by soluble RanBP1 and RanGAP in the cytoplasm. Because RanGTP is high in the nucleus and low in the cytoplasm, import complexes can only form in the cytoplasm and export complexes only in the nucleus (reviewed in (Fornerod and Ohno, 2002)). Besides Ran’s regulatory func on in nuclear transport, it has also been implicated at several stages of mitosis. Ran is important for entry into mitosis (reviewed by (Sazer and Dasso, 2000)). It regulates the spindle forma on and it is required for proper func oning of the spindle checkpoint (Arnaoutov et al., 2005). This checkpoint ascertains whether chromosomes in metaphase are properly aligned before onset of anaphase. Finally, RanGTP regulates NPC assembly by recrui ng nucleoporins to chroma n, inser ng them into membranes and by s mula ng the assembly of NPC subcomplexes (Walther et al., 2003).. 1.3 Transport receptors Most nuclear transport receptors fall into the class of karyopherin β proteins, also termed impor n β-like proteins as impor n β was the first iden fied nuclear transport factor. Fourteen different karyopherins have. 12. been iden fied in yeast and at least twenty in humans (Mosammaparast et al., 2001). Some karyopherins transport a whole range of proteins, whereas others are specialized in the transport of only one molecule, or a specific subset of molecules. For example Impor n α/β imports a wide range of proteins, impor n 4, 5 and 9 import ribosomal proteins and histones, transpor n 1 imports RNA binding protein and histones and transpor n 2 imports HuR. CRM1 is a very generic receptor, expor ng a wide range of proteins, while expor n-t exports only tRNA, and expor n 6 is the receptor for profilin and ac n. Two karyopherins func on in two direc ons: vertebrate impor n 13 func ons as an impor n for Ubc9, RBM8 and Pax6, but also as an expor n for eIF-1A (Mosammaparast et al., 2001). Yeast Msn5 also mediates bidirec onal transport (Yoshida and Blobel, 2001). Karyopherins range between 95 and 145 kDa and possess a rather weak sequence homology (15-20% iden cal), being most iden cal in the N-terminal Ran binding or “CRIME” domain (for CRM1, Impor n β, etc) (Fornerod et al., 1997a; Gorlich et al., 1997). Structurally, karyopherins are almost en rely composed of 1920 HEAT (named a er the proteins where they were first iden fied in, Hun ngton, Elonga on Factor A and TOR) repeats, presumably rendering the protein the plas city to bind different cargos (Cook et al., 2007). Most kary-.

(6) Chapter 1. Figure 2. The Ran cycle. RanGTP is kept at high concentra on in the nucleus, because the Ran guanidine exchange factor RCC1 is only nuclear as its bound to chroma n. RCC1 will mainly replace GDP for GTP because of the high GTP/GDP ra o in the cell. Upon entrance in the cytoplasm RanGTP will immediately be hydrolysed by the combined ac on of the Ran GTPase ac va ng protein RanGAP and RanBP1. This can occur already at the NPC, because RanGAP is bound to Nup358, the cytoplasmic fibrils of the NPC, which also contains RanBP1-like domains. Alterna vely, hydrolysis can occur once export complexes have passed the NPC, because RanGAP and RanBP1 are also present as soluble proteins in the cytoplasm. Recycling of RanGDP to the nucleus, where GDP will be exchanged for GTP, requires binding to the RanGDP-specialyzed import recepter, NTF2.. an gen and is s ll the most widely used signal for heterologous nuclear targe ng (Kalderon/Roberts1984). NLScontaining cargo does not bind directly to impor n β, but is recognized by an adapter protein, impor n α (also termed karyopherin α) (see Figure 1). Impor n α interacts with impor n β via the amino-terminal impor n β binding domain (IBB) and this interac on stabilizes the impor n α/cargo interac on (E hymiadis et al., 1997; Rexach and Blobel, 1995). Impor n α does not contain a HEAT-repeat domain, but is composed of ten armadillo (ARM) repeats that are related to HEAT repeats. Impor n β can also bind its cargo directly via a nonclassical NLS (Moore et al., 1999; Palmeri and Malim, 1999; Truant et al., 1999). This could be explained by the similarity between the structure of impor n α IBB domain in complex with impor n β and the IBB domain in complex with an NLS (Con et al., 1998). However, two different cargoes that bind impor n β directly have been crystallized, PTHrP and SREBP-2 (Chubb et al., 2002; Lee et al., 2003) and although SREBP-2 binds at the same site as IBB, PTHrP binds impor n β at a different site. These different interac ons underscore the flexibility of transport receptors. The release of cargo occurs by binding RanGTP to the Ran binding domain of impor n β, followed by a conforma onal change and subsequent destabiliza on of the import complex. Empty impor n β recycles back to the cytoplasm for another round of import. Impor n α in turn, requires the binding of a specialized expor n, termed CAS (cellular apoptosis susceptability) that is solely responsible for the export of impor n α (see Figure 1) (Kutay et al., 1997). Impor n α/β are also responsible for the import of integral membrane proteins into the inner nuclear membrane via a signal resembling the classical import signal (Kiseleva et al., 1998). Another important func on of impor n α/β is the regula on of many processes during mitosis and NE assembly (Weis, 2003). 1.3.2 Transportin. opherins can bind cargo directly, although in some cases an adaptor is used (Fig. 1) providing an addi onal level of specificity and regula on. Another common feature of karyopherins is their ability to interact with the NPC, in par cular through hydrophobic interac ons with FGrepeats. Because of these abundant but rather weak interac ons karyopherins can translocate rapidly through the NPC, both in the absence and presence of cargo, exceeding the size limit of diffusion (Bayliss et al., 2000; Cook et al., 2007). 1.3.1 Importin α/β Impor n β (karyopherin β)-mediated import is defined as the “classical” import pathway that transports proteins containing nuclear localiza on signals (NLS) to the nucleus. The NLS is a short monopar te or bipar te sequence of basic residues, mainly consis ng of lysines and arginines. The first NLS iden fied was a monopar te sequence (PKKKRKV) in the simian virus 40 (SV0) large T. Transpor n 1, also termed karyopherin β2, shares ~24 % sequence homology with Impor n β and was first iden fied as the import receptor for hnRNP A1 by recognizing a 39 amino acid mo f termed the M9 signal (Pollard et al., 1996; Siomi and Dreyfuss, 1995). Subsequently, transpor n 1 has been shown to import several mRNA binding proteins, e.g. hnRNP D, TAP, JKTBP, HIV Rev, and c-FOS (Imasaki et al., 2007). Unlike classical NLS sequences, the import signals recognized by transpor n 1 are rather long and have only low sequence similarity. Recent muta onal analysis has pinned down the importance of the PY mo f, present in many of the transpor n 1 cargoes (Iijima et al., 2006; Suzuki et al., 2005). 1.3.3 CRM1 CRM1 (chromosome region maintenance-1) was originally iden fied in Schizosaccharomyces pombe, where cold-sensi ve mutants led to deformed chromosome. 13.

(7) General Introduction domains (Adachi and Yanagida, 1989). Later, simultaneously with CAS, it was the first iden fied export receptor that required RanGTP binding to s mulate binding of cargo (Fornerod et al., 1997a; Kutay et al., 1997). Hence it was renamed expor n 1 (Xpo1 in yeast (Stade et al., 1997)). Whereas CAS is a highly specific expor n for impor n α (Figure 1), CRM1 appeared to be a general expor n that recognizes proteins bearing a short hydrophobic sequence of ~ 10-12 amino acids, ini ally termed leucine-rich nuclear export signals (NES). The number of CRM1-mediated cargoes is s ll growing, including both shu ling transcrip on factors, small diffusible proteins and transla on factors that have to be excluded from the nucleus, import adaptors, and also several RNA molecules. Besides an important role in interphase cells, CRM1 also has an important func on in mitosis (see 1.3.3.3)(Arnaoutov et al., 2005). 1.3.3.1 CRM1 structure CRM1 is a 115 kDa HEAT-repeat protein. Only the carboxy-terminal third (HEAT repeat 14-19) has been crystallized and together with electron microscopy analysis and homology modeling, we now have a be er understanding of CRM1 at the structural level (Petosa et al., 2004). CRM1 is a superhelical structure composed of 19 HEAT repeat mo fs of which the first three comprise the CRIME domain. This domain, together with an acidic loop in HEAT repeat eight, is responsible for RanGTP binding (Fornerod et al., 1997a). A central conserved region encompassing HEAT repeat 8-11 is responsible for NES binding. Interes ngly, residues involved in NES binding (Leu525, Lys568 and Phe572) are located at the outer surface, which are also thought to be required for interac ons with the NPC. The large acidic loop in HEAT repeat eight has strong homology with transpor n and has been proposed to alter its conforma on upon RanGTP hydrolysis, resul ng in displacement of cargo (Petosa et al., 2004). 1.3.3.2 RanBP3 Without RanGTP the affinity of the CRM1-cargo interac on is very weak, in the mM range. RanGTP greatly s mulates this interac on, though the Kd is o en not lower than 0.5 μM (Askjaer et al., 1999). Because of this weak interac on, complex forma on is thought to be the ratelimi ng step in CRM1-mediated export. To aid CRM1 NES forma on another Ran binding protein, termed RanBP3 s mulates these interac ons in two ways. First, RanBP3 tethers CRM1 to RCC1 to ensure that RanGTP loading occurs in proximity of CRM1 molecules. Simultaneously, RanBP3 also s mulates RCC1’s exchange ac vity. Second, RanBP3 enhances the affinity of CRM1 for cargo by binding directly to CRM1 (Englmeier et al., 2001; Lindsay et al., 2001). Evidence from the modeled structure of CRM1 suggest that RanBP3 interacts with the acidic loop and it has been suggested that RanBP3 stabilizes the loop in an “open” conforma on to allow both cargo and RanGTP to. 14. bind simultaneously (Petosa et al., 2004). 1.3.3.3 CRM1 in mitosis Besides its role in transport, CRM1 may also be used as a plaorm to a ach NES-containing proteins to certain subcellular structures. For example, CRM1 a aches nucleophosmin to centrosomes and this interac on is important for proper centrosome duplica on (Wang et al., 2005). In certain cell types, CRM1 regulates proper chromosome segrega on by being localized at kinetochores from the end of prophase ll anaphase or telophase. The localiza on of CRM1 to kinetochores is required for a achment of Nup358 and RanGAP to these structures. This interac on can be abolished by leptomycin B (LMB), a specific CRM1 inhibitor (see below), sugges ng that CRM1 binds in a RanGTP- and cargo-dependent manner (Arnaoutov et al., 2005; Joseph et al., 2004). 1.3.4 Transport inhibition So far, CRM1 is the only karyopherin for which a highly specific inhibitor has been iden fied. This inhibitor, termed leptomycin B (LMB) is a small molecule, which was iden fied in Streptomyces and forms a covalent bond with cysteine 528 in human CRM1 or cysteine 529 in S. pombe. This cysteine is lacking in S. cerevisiae and is therefore completely resistant to LMB, emphasizing the specificity for CRM1. LMB irreversibly and immediately inhibits CRM1 and is ac ve in a low nM range (Fornerod et al., 1997a; Wolff et al., 1997). In addi on to LMB other CRM1 inhibitors have been iden fied. Ratjadone A resembles LMB, binding CRM1 at cysteine 528 and is equally potent and specific (Koster et al., 2003). More interes ng is the small compound PKF050-638, which is thought to interact with the same cysteine residue. However, unlike LMB and Ratjadone A, the inhibi on by PKF050-638 is reversible a er removal of the drug (Daelemans et al., 2002).. 1.4 Nuclear export signals In 1995 two studies simultaneously reported the existence of a leucine-rich nuclear export signal. One was iden fied in Human Immunodeficiency Virus type 1 (HIV-1) Rev, the export adaptor for unspliced viral mRNA and the other in the protein kinase A inhibitor (PKI) (Fischer et al., 1995; Wen et al., 1995). The export signals were termed leucine-rich nuclear export signals, because of the presence of four leucines that were regularly spaced by other residues, following the consensus sequence L-X2-L-X2-3-L-X-L. Subsequent descrip ons of NESs revealed 3 that leucines could be replaced by any other hydrophobic residue, although the two carboxy-terminal hydrophobic residues, which form the most important part of the NES, are o en leucines or isoleucines (la Cour et al., 2003). Muta on of one of these core residues abolishes export ac vity completely (Wen et al., 1995). The residues in between the hydrophobic residues are frequently either.

(8) Chapter 1 charged or small. Even though more than one hundred NESs have been iden fied, a more defined consensus sequence has not yet been described. In addi on, several NESs have been described that do not follow the general consensus sequence, though they all expose several hydrophobic residues. Based on this loose consensus, most proteins are predicted to contain one or more NESs. However, the majority will not be func onal, either because they are not exposed or simply because they do not bind CRM1. Ideally, to prove the export func on of an NES the following experiments should be performed. First, the localiza on of the protein should be studied after LMB treatment or a er mutagenesis of the NES. This however, does not prove a direct interac on with CRM1. One should be extra cau ous if the structure of the protein is unknown, because the predicted NES could be part of an internal hydrophobic interac on that may be disrupted upon muta on. This could lead to protein misfolding and a subsequent altered localiza on. Second, to demonstrate a direct interac on, in vitro binding assays should be performed. The interac on should be RanGTP dependent and should be abolished upon LMB addi on or by muta on of one or more hydrophobic residues. Third, the in vitro and in vivo experiments should also be carried out with the isolated NES.. least two of these NESs (found in β-ac n and IkBα). At least three out of four hydrophobic residues in the NES were shown to contact the proposed involved residues in CRM1 (Petosa et al., 2004). However the structure of an NES complexed with CRM1 awaits crystaliza on. 1.4.3 NES regulation. NESs have different affini es for CRM1, which results in different export capaci es. The affinity of NESs can be measured in vivo with an export assay based on the shuttling HIV-1 Rev protein, which is fused to a fluorescent reporter protein, like GFP. The Rev NES is mutated, but its NLS is intact, resul ng in nuclear localiza on. Inser on of an ac ve NES, strong enough to overrule the NLS, will shi the reporter protein to the cytoplasm (Henderson and Ele heriou, 2000). A widely used quan ta ve in vitro assay is the CRM1 RanGAP assay. Strong NESs require lower CRM1 and RanGTP concentra ons to form stable complexes. This assay measures the hydrolysis of RanGTP when it is not incorporated into export complexes. (Askjaer et al., 1999; Bischoff and Gorlich, 1997). These assays have proven very useful for the iden fica on of true NESs and are frequently used for the studies described in this thesis (see Chapter 2, 5 and 6).. NESs can be regulated by several mechanisms. A frequently adopted mechanism is NES inac va on by phosphoryla on in or near the NES sequence, directly lowering the affinity of the NES (Ikuta et al., 2004; Zhang and Xiong, 2001). Phosphoryla on can also lead to ac va on of the NES as seen for the transcrip on factor Nuclear Factor of Ac vated T-cells (NFAT). Ca2+ influx leads to ac va on of the phosphatase calcineurin, which in turn dephosphorylates thirteen phosphoserine residues that are required to mask the NES. As a consequence the NLS is exposed and NFAT will be imported into the nucleus where it can perform its func on as a transcrip on factor (Okamura et al., 2000). Third, inter- or intramolecular interac ons can result in NES masking upon phosphoryla on or other post-transla onal modifica ons of the protein. An example of intramolecular masking is shown for INI1 (integrase interactor 1), a member of SWI/SNF chroma n remodeling complex, where the C-terminal 66 amino acids prevent CRM1 from binding the NES. This masking can be released upon a signal-induced conforma onal change or degrada on of the C-terminus (Craig et al., 2002). An example of intermolecular masking will be given in Chapter 5. Finally, NES ac vity can also be influenced by reten on either in the nucleus, compromising the export capacity of NESs, or in the cytoplasm, driving the direc on of transport towards the cytoplasmic compartment. An example of nuclear reten on is shown for the glucocor coid receptor (GR), which contains a nuclear reten on signal in the hinge region that ac vely impedes its export (Carvalho et al., 2001). Finally, an interes ng example of NES regula on is seen in yeast. The yeast AP-1-like transcrip on factor (Yap1p) is responsible for the upregula on of genes in response to oxida ve stress. In an oxida ve environment, the NES is no longer recognized by CRM1, because of intramolecular disulfide bond forma on, resul ng in nuclear Yap1p (Kuge et al., 2001; Yan et al., 1998).. 1.4.2 Structure. 1.5 NPC. Muta onal analyses combined with the puta ve CRM1 structure have iden fied three important residues for NES interac on; two hydrophobic residues, Leu525 and Phe572 and one posi vely charged residue, Lys568. Besides a charged moiety, lysines also have a long flexible alipha c chain that may be used to interact with hydrophobic residues of NESs. Crystal structures of five known NES-containing proteins computa onally fi ed onto CRM1 revealed that the NESs lie (partly) within an α helix and have a strong structural similarity (Ringer et al., 1999). However, some doubt has been cast on at. 1.5.1 Composition. 1.4.1 Assays to study NESs. The NE of a human cell contains between 500-5000 NPCs, depending on the metabolic state of the cell. The yeast NPC is ~ 50 MDa and the human NPC is ~120 MDa (Reichelt et al., 1990). Strikingly, both NPCs are composed of a similar set of only 30 different proteins, termed nucleoporins (or Nups). Despite its bulky mass, the composi on of the NPC is simplified by an eight-fold rota onal symmetry. Recent modelling suggests that the core-NPC consists of eight spokes, which in turn consist of two highly. 15.

(9) General Introduction homologous columns (Alber et al., 2007). The NPC has both ver cal and horizontal symmetry, being only asymmetrical in its nuclear and cytoplasmic extremi es (Figure 3). The metazoan NPC has a diameter of ~ 145 nm and is ~ 80 nm in length. The inner diameter has been es mated to be ~ 44 nm (Akey and Radermacher, 1993), which is in accordance with the maximum cargo size of 25-39 nm, based on transport studies with coated gold par cles (Dworetzky and Feldherr, 1988; Pante and Kann, 2002). As seen with scanning electron microscopy, the NPC has a basket-like structure on the nuclear side and eight protruding filaments on the cytoplasmic side (Goldberg and Allen, 1992). Some nucleoporins are highly dynamic and have short residency me at the NPC (minutes), whereas membrane-bound and more internally located nucleoporins are very stably localized (reviewed in (Rabut et al., 2004; Tran and Wente, 2006). Recently, yeast nucleoporins have been classified into three groups, based on their structural domains. The first group consists of membrane-bound nucleoporins, containing transmembrane α-helices. This group is thought to form the outermost layer of the NPC, responsible for anchoring the NPC into the NE. The innermost layer consists of nucleoporins containing phenylalanine-glycine (FG) repeat domains (Figure 3) (Devos et al., 2006). These domains are very abundant, highly unstructured and unfolded and fill most of the NPC channel (Ribbeck and Gorlich, 2001; Rout et al., 2000) and references therein). Most of these nucleoporins contain coiled coils with which they are thought to interact with other NPC components. The last group, consis ng of approximately half of the nucleoporins has an α-solenoid, a β-propeller or a combina on of both domains. These nucleoporins form a structural scaffold between the membrane-bound nucleoporins and the FG-repeat-containing nucleoporins (Figure 3) (Devos et al., 2006). 1.5.2 Models of translocation through the NPC How the FG-repeats selec vely allow transport of substrates is s ll a ma er of debate. One model, termed “virtual ga ng” predicts that the unstructured filamentous FG-repeats fill the central channel, restric ng molecules larger than ~ 30-40 kDa from passing through. Receptor/ cargo complexes however, by virtue of interac ons with the FG-repeats can translocate through the NPC channel (Rout et al., 2000). This hypothesis is supported by a recent study where the composi on of the NPC was constructed based on computa onal modeling of biochemical interac on data. The inner layer of the NPC reveals an FG-repeat-containing cloud that thins towards the interior of the central channel, leaving an ~ 10 nm channel in the middle, which is consistent with the maximal diffusible size (Alber et al., 2007). The second model, termed “selec ve phase” predicts that FG-repeats do interact with each other (Ribbeck and Gorlich, 2002). Evidence for this model comes from a study in yeast where it was. 16. shown that most FG-repeat containing nucleoporins present at the core of the NPC possess a weak affinity for each other. The FG-repeats of the asymmetrically located nucleoporins do not interact with other FG-repeats, sugges ng that these nucleoporin either form repulsive bristles or s mulate more specific nuclear transport receptor interac ons (Patel et al., 2007). In other studies it has been shown that certain FG nucleoporins form a saturated hydrogel in vitro (Frey and Gorlich, 2007; Frey et al., 2006). The interac ng FG-filaments form a meshwork that func ons as a sieve, perming only small molecules to pass through. In this model receptor/cargo complexes will dissolve the meshwork locally, allowing the complex to slide through (Frey and Gorlich, 2007; Ribbeck and Gorlich, 2001; Weis, 2007). A third model predicts that affini es of karyopherins for FG nucleoporins increase towards the final compartment. This implies that direc on of nuclear transport is not solely governed by the Ran system, but is aided by interac ons at the NPC (Ben-Efraim and Gerace, 2001). However, experiments in yeast, where asymmetrically located nucleoporins were either deleted or swapped, showed no defects in nuclear transport (Strawn et al., 2004; Zeitler and Weis, 2004). Moreover, permeabilized cell assays revealed that the transport direc on of receptor/cargo complexes can be reversed by switching the RanGTP gradient, which also strongly argues against the affinity gradient model (Nachury and Weis, 1999). 1.5.3 Flexibility of the NPC The centre of the NPC is filled with an electron-dense structure as seen with electron microscopy, which is thought to be at least partly composed of transi ng cargo. Cryoelectron tomography in the slime mould Dictostelium analyzed this central plug in NPCs incubated with gold-labeled import complexes and in principle two different states could be iden fied, which are thought to represent different steps in transport. One state shows an NPC with a more internally located plug and cytoplasmic filaments that are unordered and variably protrude from the cytoplasmic side of the NPC core. In the second state, the central plug is located in the plane of the cytoplasmic filaments. The filaments are bent towards the central plug and are thought to be interac ng with cargo at this state. These results also demonstrate a more general conforma onal change in the two states (Beck et al., 2004; Beck et al., 2007). This flexibilty of the NPC has also been shown for the diameter of the central channel (Akey, 1989; Beck et al., 2004; Kiseleva et al., 1998). Recently, structural evidences for this flexibility has been obtained by analyzing two internally located nucleoporins, Nup58 and Nup45, whose α-helices interact with each other. These dimers form tetramers through hydrophilic interac ons. Interes ngly, these la er interac ons are variable, leading to sliding along the dimer surface, resul ng in a shi of ~ 11 Å and an increase of the en re NPC channel of ~ 30 Å. This sliding of nucleoporin may facilitate the passage the large cargo (Melcak et al., 2007)..

(10) Chapter 1 The diameter of the pore is also thought to be influenced by changes in Ca2+ influx. Several studies have shown either a structural change of the NPC, in par cular of the nuclear basket, or a func onal change, leading to changes in permeability and nuclear transport ((Erickson et al., 2006; Thorogate and Torok, 2007) and references therein).. lourgidis et al., 2006). Nup88, but not Nup214, interacts with Nup358 as shown by immunoprecipita on studies. Deple on of either Nup88 or Nup214, results in a strong reduc on of Nup358 from the NE. These results suggest that Nup88 forms a plaorm for a aching Nup358 to the NPC (Bernad et al., 2004). 1.6.1 Nup358. The dynamics of the NPC can also take place at the level of NPC density in the NE. For example, quickly dividing cells require a fast incorpora on of NPCs in the NE Alterna vely, incorpora on of specific nucleoporins may alter the func on of the NPC. A nucleoporin storage site could be useful to accomplish this rapid increase in NPC density. In the cytoplasm, stacks composed of double membranes are present, which are perforated by NPCs in both orienta ons. These organelles are termed annulate lamellae (AL) and have been proposed to serve as a storage site for nucleoporins. Interes ngly, besides the standard subset of nucleoporins to build up an NPC, transport factors associated with the NPC are also present at AL (Kessel, 1992).. 1.6 Nucleoporins The structural components that form the NPC are termed nucleoporins or nups and they are o en named a er their molecular weights. Even though the protein sequences of yeast and vertebrate nucleoporins are poorly conserved, more than eighty percent of the vertebrate nucleoporins have a structural counterpart in yeast. The vertebrate NPC differs however by having a subset of nucleoporins with more complex domains or modifica ons that are not present in yeast. For example, general structures present in mul ple vertebrate nucleoporins are the zinc-fingers, implicated in Ran binding (Nakielny et al., 1999; Wang et al., 2003) and in coordina ng NE breakdown (Higa et al., 2007; Nakielny and Dreyfuss, 1999; Prunuske et al., 2006). Tryptophan-alanine (WD) repeat-containing nucleoporins are more abundant in vertebrates and are thought to mediate the assembly of protein complexes (Cronshaw et al., 2002). Moreover, serines or threonines in vertebrate nucleoporins can be phosphorylated in a cell-cycle dependent manner or Olinked glycosylated, though li le is understood about the precise role of these modifica ons (Miller et al., 1999). The eminent nuclear basket structure at the nuclear side of the NPC is mainly composed of Nup153 and TPR (translocated promoter region), of which the la er has been mapped more exteriorly and is a ached to the pore via Nup153 (Figure 3) (Hase and Cordes, 2003; Krull et al., 2004). At the other side of the vertebrate NPC, three exclusively cytoplasmic nucleoporins are located: Nup358, Nup214 and Nup88. These nucleoporin will be discussed independently in the next sec on. Nup214 forms a dynamic subcomplex with Nup88 and their stability is directly dependent on their interac on (Bastos et al., 1997; Bernad et al., 2004; Fornerod et al., 1997b; Xy-. Nup358 is one of the few nucleoporins that is specific for metazoans. It was ini ally termed Ran binding protein 2 (RanBP2) when iden fied in a yeast two-hybrid screen for Ran binding proteins (Yokoyama et al., 1995). Nup358 is the largest nucleoporin, ~ 35 nm in length, and is present at the cytoplasmic side of the NPC, where it is the main cons tuent of the cytoplasmic fibrils (Delphin et al., 1997; Walther et al., 2002; Wilken et al., 1995; Wu et al., 1995; Yokoyama et al., 1995). It has mul ple domains for which increasing number of interac ng proteins and func ons have been iden fied. These domains are illustrated in Figure 7 of Chapter 2. The amino-terminus has been proposed to be the site of a achment to the NPC core, possibly via the predicted leucine zipper and a coiled-coil. Nup358 has four Ran binding domains (Wu et al., 1995) and a zinc-finger domain, which has been shown to coordinate NE breakdown (Prunuske et al., 2006). The internal repeat domain (IR1) a aches RanGAP (see 1.4) to the NPC via a SUMO moiety and also interacts with the SUMO E2 conjugase, Ubc9. Interes ngly, Nup358 itself has SUMO E3 ligase ac vity and together with Ubc9 it can SUMOylate proteins like p53 and IκBα. RanGAP however is not SUMOylated by Nup358 (Reverter and Lima, 2005). At the carboxy-terminus of Nup358 is a domain with homology to cyclophilin A, which has been shown to play a role in the ubiqui n-proteasome system (Yi et al., 2007). Nup358, together with other NPC components and transport factors (see 1.5.3.3) also plays a role during mitosis as the Nup358-RanGAP complex, when SUMOylated, is directly required for the attachment of microtubules to kinetochores (Joseph et al., 2004). Nup358 does not seem to be required for general import or export via the impor n α/β or CRM1 pathway, as deple on of Nup358 has no or only li le effect on these pathways (Bernad et al., 2004). Finally, Drosophila Nup358 has been shown to play a role in mRNA export. Deple on of Nup358 results in an mRNA export defect and in the cytoplasmic accumula on of the mRNA export factor NXF1/p15. Binding of the mRNA export factor NXF1/p15 to Nup358 inhibits leakage of the export factor to the cytoplasm, with the resultant increased concentra on of nuclear NXF1/p15 making mRNA export more efficient . However, Drosophila lacks RanBP1, so deple on of Nup358 here depletes the cell of all RanBP1-like ac vity, likely having general effects on many transport processes. As men oned before, Nup358 does not have a counterpart in yeast. However, a recent study by Stelter et al. described the presence of dynein light chain (Dyn2) at. 17.

(11) General Introduction. 1.6.2 Nup214. cell cycle arrest, reduc on in protein import and mRNA export defects (Boer et al., 1998). However, it may be hard to dis nguish the func ons of Nup88 and Nup214 as the stability of both proteins depend on each other (Bernad et al., 2004). Nup214/CAN also has an amino-terminal β-propeller structure and two central coiled-coils through which it interacts with Nup88 (Nehrbass and Blobel, 1996). Nup214 plays a role in mRNA export by a aching the RNA helicase Dbp5 (DEAD box protein 5) and the mRNA export factor GLE1 (Kohler and Hurt, 2007).. Nup214/CAN was first iden fied as an oncogenic fusion product causing acute myeloid leukemia (AML) when fused to the protein tyrosine kinase c-Abl. Because of this fusion with Abl, the protein was termed Cain (Can) (von Lindern et al., 1990). Shortly a er, Nup214/Can was also found in transloca ons with the chroma n protein DEK and with the PP2A inhibitor SET, causing AML and acute undifferen ated leukemia (AUL), respec vely (von Lindern et al., 1992a; von Lindern et al., 1992b). Nup98, a nucleoporin located at the nuclear side of the NPC, has also been implicated in mul ple forms of leukemia, by being fused to homeodomain transcrip on factors or histone methyl transferases. Interes ngly, all transloca ons retain the FG repeats of Nup214 or Nup98, sugges ng this domain is important for the oncogenic proper es of the fusion proteins (reviewed by {Kalverda, 2007 #1000} (Kasper et al., 1999). Deple on of Nup214 also results in. The high-affinity interac on of Nup214’s FG-repeat region with the export receptor CRM1 has been studied extensively, as demonstrated both in vitro and in vivo (Fornerod et al., 1996; Fornerod et al., 1997b; Xylourgidis et al., 2006). Because of this strong interac on, Nup214 has been speculated to play a role in preven ng the reimport of export complexes (Askjaer et al., 1999), or to be the site of complex dissocia on (Askjaer et al., 1999; Hutten and Kehlenbach, 2006; Kehlenbach et al., 1999). The role of Nup214 in export however, remains controversial. Hu en and Kehlenbach demonstrated a strong CRM1 export defect upon knockdown of Nup214 by RNAi (Hutten and Kehlenbach, 2006), whereas a study by Bernad et al. showed only a minor export defect (Bernad et al., 2004), similar to the defect observed in Nup358 knockdown. In Drosophila S2 cells, CRM1-mediated export is not compromised in Nup214 mutants that are unable. the yeast NPC. Five copies of Dyn2 have been shown to bind the dynein light chain interac ng domain (DID) of Nup159, the yeast counterpart of Nup214. This array of Dyn2 interac ng with Nup159 forms a rod-like structure of ~ 20 nm. Because the yeast NPC lacks Nup358, Dyn2 has been hypothesized to be the vertebrate counterpart of the cytoplasmic filaments and may play a role in organizing the FG-repeats of Nup159 (Stelter et al., 2007).. Figure 3. Structure of the Nucleopore complex The core of the NPC, which penetrates the NE, consists of transmembrane containing-nucleoporins (anchor-Nups). These nucleoporins are laced by scaffold nuceloporins (scaffold-Nups), containing β-propellors and α-helices, which in turn form a plaorm for FG-repeat containing nucleoporins (FG-Nups). The FG-repeats are unstructured filaments that fill the centre of the NPC, either partly or completely. These filaments form a hydrophobic meshwork required for the interac on with the transport receptors. On the cytoplasmic side three exclusively cytoplasmic nucleoporins are present: Nup358, Nup214 and Nup88. The Nup214/Nup88 subcomplex forms the anchor for the cytoplasmic fibrils, composed of Nup358. Via a SUMO moiety, RanGAP is a ached at Nup358, which also contains four RanBP1-like domains. The nuclear side of the NPC forms a basket-like structure which is composed of Nup153 and TPR, of which the la er is situated more exteriorly.. 18.

(12) Chapter 1 to bind CRM1. Instead Nup214 mutant larvae show an increase in CRM1-mediated export, sugges ng that the CRM1-Nup214 interac on is actually inhibi ng the CRM export pathway (Sabri et al., 2007). These discrepancies will be discussed in Chapter 7.. ported in a RanGTP-dependent manner and are further processed in the cytoplasm to single stranded RNA molecules. Subsequently, these RNAs are incorporated into the RISC complex and basepair with their target mRNAs, leading to transla onal inhibi on (reviewed in (Kohler and Hurt, 2007)).. 1.6.3 Nup88 1.7.2 rRNA Nup88 does not have an FG-rich repeat, but its yeast counterpart is predicted to have an amino-terminal β-propeller and a carboxy-terminally placed coiled coil (Alber et al., 2007). Whereas Nup214 and Nup358 have only eight copies at a each NPC, Nup88 is predicted to have 32 copies (Cronshaw and Matunis, 2004). Nup88 is also abundant in the cytoplasm, although the purpose of this excess has not yet been determined. Nup88 has been shown to be upregulated in many cancers and premalignant lesions, however in these studies Nup88 was predominantly detected in the nuclei of cells (Mar nez et al., 1999), which contrasts with its proposed regular cytoplasmic localiza on. This overexpression is correlated with tumorigenesis and aggressiveness of colorectal cancers (Emterling et al., 2003) and also with the metasta c poten al of melanomas (Zhang et al., 2002). Whether Nup88 func ons as an oncogenic protein or only follows an increased demand for nuclear transport is not understood at present.. 1.7 RNA export Export of RNA molecules is required for two reasons. First, RNA molecules are transcribed in the nucleus, but most of them exert their func on, or need to be translated, in the cytoplasm. Second, for nuclear RNAs like small nucleolar RNAs, certain matura on steps take place in the cytoplasm. The karyopherins responsible for this transport are some mes very specialized, e.g. expor n-t only exports tRNAs, whereas other RNAs, like rRNAs, snRNAs and some mRNAs, are exported by the generic export receptor CRM1. The export of the majority of messenger RNAs (mRNAs) is mechanis cally different, as the export factors involved do not resemble karyopherins (Kohler and Hurt, 2007). 1.7.1 tRNA and miRNA tRNAs are exported by expor n-t that, instead of recognizing linear sequences like classical transport signals, interacts with secondary and ter ary structural elements in the tRNA (Arts et al., 1998; Lipowsky et al., 1999). Expor n-5 is also able to export tRNA, and this receptor is also responsible for miRNA export (Lund et al., 2004). miRNAs are transcribed either from the intron of a coding gene or from a separate miRNA coding gene. The miRNA precursors form hairpin structures that are ex-. rRNAs are ribosomal RNA molecules that are incorporated into large protein complexes together construc ng the ribosomal subunits. Each ribosome is composed of a 40S and 60S subunit that together contain four rRNA molecules (28S/25S rRNA, 5.8S rRNA, 5S rRNA and 18S rRNA). Ribosomal biogenesis is rather complex and involves processing of rRNAs and associa on with ribosomal proteins to form ribonucleoproteins (RNP). These processes take place in the nucleus as well as in the cytoplasm. The final matura on of the 40S and 60S subunits takes place in the nucleus and their subsequent export into the cytoplasm depends on CRM1. For both subunits the CRM1 interac on is thought to be bridged by export adaptors (Zemp and Kutay, 2007). The export of the 40S subunit is not completely understood, although the involvement of CRM1 is evident (Gleizes et al., 2001; Moy and Silver, 2002). Several other factors have been implicated although it is unclear at present whether these are true ribosomal export adaptors (Oeffinger et al., 2004). CRM1 mediates the export of the 60S ribosomal subunit and this interac on is bridged by the export adaptor NMD3 (non-sense mediated mRNA decay protein 3) that bears a classical NES required for the CRM1 interac on (Thomas and Kutay, 2003). In the cytoplasm, CRM1 is released from NMD3 by RanGTP hydrolysis. Subsequently, the GTPase Lsg1 dissociates NMD3 from the 60S subunit (Hedges et al., 2005; West et al., 2005). In Chapter 6 a specific role for Nup214 in 60S ribosomal export will be described. 1.7.3 snRNA Small nuclear RNAs (snRNAs) are spliceosomal RNAs that are incorporated in the spliceosome and are required for proper splicing of RNAs. This splicing occurs in the nucleus, but the assembly of the snRNAs U1, U2, U4, U5 and U6 with a group of seven proteins known as Sm ribonucleoproteins, occurs in the cytoplasm, at least in higher eukaryotes. snRNAs are exported by CRM1, but similar to rRNA, adaptor proteins are required to bridge this interac on. Before being exported, snRNAs acquire a 5’ cap structure (Hamm et al., 1990), which is recognized by a protein termed CBC (cap binding complex) (Izaurralde et al., 1995). Subsequently phosphorylated PHAX (phosphorylated adaptor for RNA export) will bind the complex and recruit CRM1 via its NES. The PHAX/CRM1/RanGTP complex by itself is rather weak, but upon CBC binding this interac on is greatly enhanced. In the cytoplasm,. 19.

(13) General Introduction dephosphoryla on of PHAX, together with RanGTP hydrolysis, results in dissocia on of the complex (Ohno et al., 2000). In the cytoplasm snRNAs are assembled into mature snRNPs and imported back into the nucleus by impor n β. In this case, instead of impor n α, a higly specialized adaptor, termed snurpor n 1, is used (Huber et al., 1998; Palacios et al., 1997). Interes ngly, CRM1 has a dual func on in snRNP matura on. Besides being directly involved in the export of snRNAs, CRM1 also recycles snurpor n 1 back to the cytoplasm. Snurpor n 1 however, lacks a classical NES, but binds CRM1 through large domains, involving residues located throughout the protein. This finding is supported by muta on analysis of CRM1, in which none of the single, double or triple muta ons abolished Snurpor n binding completely. This result showed that unlike the classical NES binding site, snurpor n 1 binding requires mul ple residues spread out over CRM1 (Petosa et al., 2004). Another characteris c is the unusually high affinity for CRM1 (Kd ≈ 10 nM), which may be explained by the increase in avidity due to the mul ple interac ons. The affinity of small classical NESs is approximately 100-fold lower (Paraskeva et al., 1999). 1.7.4 mRNA export Export of general mRNA molecules is perhaps the most complex export pathway and involves a wide range of proteins and processes before an export-competent mRNP is formed. The major processes before mRNA export are 5’ capping, splicing of the introns, cleavage and polyadenyla on of the 3’ end. During these processes, mRNA is assembled into an mRNP par cle, which will be bound by proteins that serve as export adaptors, like YRA1 or ALY/REF (Lei et al., 2001; Zenklusen and Stutz, 2001; Zhou et al., 2000). The generic export receptor TAP-p15 (also termed NXF1-p15 and Mex67/Mtr2 in yeast) is recruited to the mRNP via these adaptors (Kohler and Hurt, 2007). TAP-p15 does not belong to the karyopherin family and has no RanGTP binding domain, although it does interact with FG-rich nucleoporins. The lack of RanGTP binding predicts the existence of a different dissocia on mechanism at the cytoplasmic site of the NPC (Kohler and Hurt, 2007). Remodelling of the mRNP par cle could form an irreversible ac on that prevents mRNPs from re-import. Studies in yeast suggest that this remodelling is established by the RNA helicase Dbp5, which forms a key player in mRNA export. It unwinds mRNA in an ATP-driven fashion and is a ached at the cytoplasmically oriented nucleoporin Nup159. Dpb5 is a shu ling protein that is also present in the nucleus where it is recruited to nascent mRNPs during transcrip on (Hodge et al., 1999; Schmi et al., 1999; Zhao et al., 2002). However, Dbp5 will not be ac ve in the nucleus as it has very low intrinsic helicase ac vity. At the NPC, this ac vity is enhanced ~600-fold by the combined ac on of Gle1 and InsP6, both present at the cytoplasmic side of the NPC (Alcazar-Roman et al., 2006; Weirich et. 20. al., 2006). The unwinding of the mRNA may lead to dissocia on of the mRNA-associated proteins, including its export factors (Cole and Scarcelli, 2006). Dbp5 has also been proposed to func on as a ratchet wheel, pulling the mRNP complexes out of the nucleus (Stewart, 2007). Although CRM1 is not the major mRNA export factor (discussed in (Fornerod and Ohno, 2002)), it is responsible for export of a specific subset of mRNAs. For example, export of AU-rich elements-containing mRNAs present in short-lived mRNAs, like mRNAs coding for interferons, cytokines, proto-oncogenes and growth factors is mediated by CRM1. These AU-rich elements are recognized by the RNA-binding protein HuR, a protein involved in mRNA stabiliza on, and subsequent binding of the NEScontaining proteins APRIL and pp32 recruits CRM1 to the complex (Gallouzi et al., 2001). Other examples of AREcontaining mRNAs exported via the CRM1 pathway are c-FOS, COX-2 and IFN-α (reviewed in (Hu en and Kehlenbach, 2007)).. 1.8 Viruses and nuclear transport Viruses o en hijack the host transport machinery to promote their replica on and infec vity (reviewed in (Fontoura et al., 2005)). Although some viruses replicate in the cytoplasm, viruses that replicate in the nucleus require the host’s import machinery to transport the viral genome into the nucleus and allow replica on and gene expression. Adenovirus, for example, disassembles at Nup214 at the NPC and ‘injects’ its DNA into the nucleus, a process aided by host cell factors (Trotman et al., 2001). Besides import, viruses o en use the host cell’s nuclear export machinery to export their mRNAs and/or genomes. A well-studied example is the export of the introncontaining transcripts of the retrovirus HIV-1. HIV-1 has one pro-viral transcript that encodes the informa on for nine proteins that are expressed from different mRNA molecules formed by alterna ve splicing. The mRNAs for the regulatory proteins Tat, Rev and Nef are fully spliced and exported via the canonical mRNA export pathway. The other six transcripts are either unspliced or par ally spliced. These transcripts would normally be retained in the nucleus, but the virus encodes a non-stuctural protein named Rev that allows nuclear export of these transcripts. Rev is a shu ling protein that specifically recognizes a ter ary structural element in these transcripts, termed the Rev response element (RRE). Rev mediates the export of these mRNAs via its leucine-rich NES, which was one of the first NESs iden fied (Fischer et al., 1995). At steady state, Rev predominantly localizes to nucleoli by interac ng with the nucleolar protein B23, where it is thought to inhibit cell growth by inhibi ng transport of (pre-) ribosomal proteins (Miyazaki et al., 1996). Whereas the HIV-1 genome harbours the informa on for Rev, which recognizes viral transcripts, simian type D.

(14) Chapter 1 retroviruses, like the Mazon-Pfizer monkey virus, require a host factor for the binding of their transcripts. Like in HIV-1, the export of unspliced mRNA molecules is mediated by a structural element, termed the cons tu ve transport element (CTE). The mRNA export factor TAP is responsible for recogni on and export of CTE-containing transcripts (Grüter et al., 1998). Interes ngly, the introncontaining TAP mRNA also contains a CTE and is transported by the TAP protein itself. Therefore, it is thought that TAP plays a more general role in the export of introncontaining mRNAs (Li et al., 2006). Viruses can also inhibit the host’s transport pathways in order to cause cytotoxicity to the host and/or to promote their own transport. An example of this is inhibi on of host mRNA export by the Vesicular Stoma s Virus (VSV) Matrix protein (M protein). M protein exerts this func on by inhibi ng Rae1, a mRNA export factor that is attached at the nuclear located nucleoporin Nup98. M protein binding is thought to interfere with mRNA export by compe ng for the Rae1-Nup98 interac on (Faria et al., 2005; von Kobbe et al., 2000). In Chapter 4, I will describe the inhibi on of the host mRNA export pathway by the parvovirus NS2 protein of MVM (see 1.8.1). Viruses may also adopt the host’s nuclear transport system to export themselves out of the nucleus. Although most nuclear assembled viruses are too big to pass the NPC, one excep on is the small parvovirus with a diameter of only 25 nm, which equals the size of a ribosome and is therefore small enough to translocate through NPCs. Larger viruses escape the nucleus either by cell lysis, or by using alterna ve ways to penetrate the NE. One elegant example is the nuclear egress of the large enveloped herpesvirus. Studies in cytomegalovirus, a member of the β-herpesvirus subfamily, revealed that two viral proteins, M53/p38 and M50/p35 that together form a docking site for the virus, govern the first step in nuclear egress. M50/p35 recruits protein kinase C, which phosphorylates the nuclear lamina causing its dissolu on (Muranyi et al., 2002). Subsequently, the herpesvirus translocates through the NE by an ini al envelopment at the inner nuclear membrane, followed by a re-envelopment upon fusion with the outer nuclear membrane. Upon arrival in the cytoplasm, the capsid acquires its final envelope by fusing with vesicles derived from the Golgi network (reviewed in (Me enleiter et al., 2006)). 1.8.1 Minute Virus of Mice Minute Virus of Mice (MVM) is the mouse species of the parvovirus genus. It is a small single stranded DNA virus, composed of an icosahedral capsid, en rely made up of the two viral proteins, VP1 and VP2, that surround the 5 kb single-stranded DNA molecule. Natural infec ons are asymptoma c (Kimsey et al., 1986; Rubio et al., 2005), but an immunosuppressive variant of MVM, termed MVMi, causes death in newborns and leukopenia in SCID mice (Ramirez et al., 1996; Segovia et al., 1999). Interest-. ingly, cells transformed by various mechanisms have an increased suscep bility for cytotoxicity caused by MVM (Cornelis et al., 1988a; Cornelis et al., 1988b; Mousset et al., 1986). The MVM life cycle starts with uptake of the virus at the cell membrane through receptor-mediated endocytosis. As nuclear assembly and DNA replica on take place in the nucleus, MVM should insert its DNA in the nucleus or alterna vely, the complete virus should enter the nucleus. For MVM, evidence for both mechanisms have been demonstrated. In lysosomes, the low pH causes structural rearrangements of the capsid, resul ng in the exposure of an N-terminal NLS in VP1, which is required for MVM import (Lombardo et al., 2002; Mani et al., 2006). Evidence for an alterna ve mechanism has been shown in Xenopus oocytes, where MVM injec ons led to small breaks in the NE through which MVM could enter the cell nucleus (Cohen and Pante, 2005). In the nucleus, the replica on of the ssDNA requires DNA polymerase and as a consequence MVM can only infect dividing cells. The viral genome contains the informa on for two structural proteins VP1 and VP2 and at least two non-structural proteins NS1 and NS2 (Cotmore et al., 1983). VP1 and VP2 form heterotrimers and VP2-only homotrimers in the cytoplasm and their import relies on the N-terminal NLS in VP1 and on a nuclear localiza on mo f present in both VP1 and VP2. In the nucleus these trimers assemble into icosahedral capsids composed of twenty trimeric subunits (Lombardo et al., 2002; Riolobos et al., 2006). MVM spreads by cell lysis, but it is able to escape the nucleus prior to cell lysis (Miller and Pintel, 2002). The mechanism of the la er is not precisely known, although this viral egress is mediated by an N-terminal pep de of VP2, termed 2Nt. This signal is exposed upon inser on of the replicated ssDNA into newly assembled capsids (Maroto et al., 2004). The two non-structural proteins NS1 (83 kDa) and NS2 (25kDa) are expressed early during infec on (Clemens and Pintel, 1988) and have a rela vely short half-life of less than one hour (Schoborg and Pintel, 1991). Together they are responsible for managing all processes required for viral replica on. NS1 is a nuclear phosphoprotein that is imported via a classical NLS, composed of at least two lysine-containing domains (Nuesch and Ta ersall, 1993). It is required for viral DNA replica on (Cotmore and Tattersall, 1987) and for causing cytotoxicity to the host cell (Caillet-Fauquet et al., 1990). NS2 has been shown to func on in mul ple processes like capsid assembly (Cotmore et al., 1997), viral messenger transla on (Naeger et al., 1993) DNA replica on (Choi et al., 2005), virus produc on (Naeger et al., 1990), cytotoxicity (Legrand et al., 1993) and capsid egress from the nucleus (Eichwald et al., 2002; Miller and Pintel, 2002) though its mode of ac on has not yet been unraveled. NS2 exists in a nonphosphorylated and a phosphorylated form, of which the la er is excluded from the nucleus (Cotmore and Tattersall, 1990). Thus far, only three interactors of NS2 are. 21.

(15) General Introduction known: Smn (Survival Motor Neuron), a protein mutated in the neurodegenera ve disease spinal muscular atrophy, the phosphothreonine and phosphoserine-binding protein 14-3-3 and CRM1, which is its most studied interac on partner. Whereas 14-3-3 predominantly binds the phosphorylated form of NS2 in the cytoplasm (Bodendorf et al., 1999; Brockhaus et al., 1996), CRM1 interacts with the non-phosphorylated form in both compartments (Bodendorf et al., 1999). CRM1 exports NS2 out of the nucleus by binding the NS2 NES (82-MTKKFGTLTI-91) (Askjaer et al., 1999; Eichwald et al., 2002). This NES has been used in many studies because of its potent binding to CRM1 (Askjaer et al., 1999; Petosa et al., 2004). The CRM1-NS2 interac on has been implicated in ssDNA replica on (Choi et al., 2005) and viral egress from the nucleus (Eichwald et al., 2002; Miller and Pintel, 2002). Early on in infec on, NS2 is mainly localized in the cytoplasm, while at later me points, NS2 accumulates in the nucleus. Furthermore, an enhancement of the CRM1NS2 interac on has been shown to improve viral fitness poin ng to a role for nuclear export pathways in MVM pathogenicity (Lopez-Bueno et al., 2004). In Chapter 3 of this thesis I describe a study on the NES of MVM NS2 and show that this NES has all the features of a supraphysiological NES. We further showed that this NES is required for parvoviral egress from the nucleus. Chapter 4 is a short report on another yet unknown func on of MVM NS2, the inhibi on of cellular mRNA. Thus, as in recent years much progress has been made in understanding the basics of nucleocytoplasmic transport, basic ques ons s ll needs to be elucidated. Major issues are the precise requirements of nuclear transport signals, whether different classes of cargo require different types of signals and which interac ons at the NPC are required for nucleocytoplasmic transport. In this thesis I address these ques ons using nuclear export signals and the transport receptor CRM1. I emphasize however, that the answers obtained may in general be applicable to many other nucleocytoplasmic transport pathways as well. References Adachi, Y. and Yanagida, M. (1989) Higher order chromosome structure is affected by cold-sensi ve muta ons in a Schizosaccharomyces pombe gene crm1+ which encodes a 115-kD protein preferen ally localized in the nucleus and its periphery. J Cell Biol, 108, 1195-1207. Akey, C.W. (1989) Interac ons and structure of the nuclear pore complex revealed by cryo-electron microscopy. J Cell Biol, 109, 955-970. Akey, C.W. and Radermacher, M. (1993) Architecture of the Xenopus nuclear pore complex revealed by three-dimensional cryoelectron microscopy. J Cell Biol, 122, 1-19. Alber, F., Dokudovskaya, S., Veenhoff, L.M., Zhang, W., Kipper, J., Devos, D., Suprapto, A., Karni-Schmidt, O., Williams, R., Chait, B.T., Sali, A. and Rout, M.P. (2007) The molecular architecture of. 22. the nuclear pore complex. Nature, 450, 695-701. Alcazar-Roman, A.R., Tran, E.J., Guo, S. and Wente, S.R. (2006) Inositol hexakisphosphate and Gle1 ac vate the DEAD-box protein Dbp5 for nuclear mRNA export. Nat Cell Biol, 8, 711-716. Arnaoutov, A., Azuma, Y., Ribbeck, K., Joseph, J., Boyarchuk, Y., Karpova, T., McNally, J. and Dasso, M. (2005) Crm1 is a mito c effector of Ran-GTP in soma c cells. Nat Cell Biol, 7, 626-632. Arts, G.J., Kuersten, S., Romby, P., Ehresmann, B. and Ma aj, I.W. (1998) The role of expor n-t in selec ve nuclear export of mature tRNAs. Embo J, 17, 7430-7441. Askjaer, P., Bachi, A., Wilm, M., Bischoff, F.R., Weeks, D.L., Ogniewski, V., Ohno, M., Niehrs, C., Kjems, J., Ma aj, I.W. and Fornerod, M. (1999) RanGTP-regulated interac ons of CRM1 with nucleoporins and a shu ling DEAD-box helicase. Mol Cell Biol, 19, 6276-6285. Bastos, R., Ribas-de-Pouplana, L., Enarson, M., Bodoor, K. and Burke, B. (1997) Nup84, a novel nucleoporin that is associated with CAN/Nup214 on the cytoplasmic face of the nuclear pore complex. J Cell Biol, 137, 989-1000. Bayliss, R., Li lewood, T. and Stewart, M. (2000) Structural basis for the interac on between FxFG nucleoporin repeats and impor n-beta in nuclear trafficking. Cell, 102, 99-108. Beck, M., Forster, F., Ecke, M., Plitzko, J.M., Melchior, F., Gerisch, G., Baumeister, W. and Medalia, O. (2004) Nuclear pore complex structure and dynamics revealed by cryoelectron tomography. Science, 306, 1387-1390. Beck, M., Lucic, V., Forster, F., Baumeister, W. and Medalia, O. (2007) Snapshots of nuclear pore complexes in ac on captured by cryo-electron tomography. Nature, 449, 611-615. Ben-Efraim, I. and Gerace, L. (2001) Gradient of Increasing Affinity of Impor n beta for Nucleoporins along the Pathway of Nuclear Import. J Cell Biol, 152, 411-418. Bernad, R., Van Der Velde, H., Fornerod, M. and Pickersgill, H. (2004) Nup358/RanBP2 A aches to the Nuclear Pore Complex via Associa on with Nup88 and Nup214/CAN and Plays a Suppor ng Role in CRM1-Mediated Nuclear Protein Export. Mol Cell Biol, 24, 2373-2384. Bischoff, F.R. and Gorlich, D. (1997) RanBP1 is crucial for the release of RanGTP from impor n beta-related nuclear transport factors. FEBS Le , 419, 249-254. Bodendorf, U., Cziepluch, C., Jauniaux, J.C., Rommelaere, J. and Salome, N. (1999) Nuclear export factor CRM1 interacts with nonstructural proteins NS2 from parvovirus minute virus of mice. J Virol, 73, 7769-7779. Boer, J., Bonten-Surtel, J. and Grosveld, G. (1998) Overexpression of the nucleoporin CAN/NUP214 induces growth arrest, nucleocytoplasmic transport defects, and apoptosis. Mol Cell Biol, 18, 1236-1247. Brockhaus, K., Plaza, S., Pintel, D.J., Rommelaere, J. and Salome, N. (1996) Nonstructural proteins NS2 of minute virus of mice associate in vivo with 14-3-3 protein family members. J Virol, 70, 7527-7534..

(16) Chapter 1 Caillet-Fauquet, P., Perros, M., Brandenburger, A., Spegelaere, P. and Rommelaere, J. (1990) Programmed killing of human cells by means of an inducible clone of parvoviral genes encoding nonstructural proteins. Embo J, 9, 2989-2995.. localiza on and turnover characteris cs. Virology, 177, 477-487. Craig, E., Zhang, Z.K., Davies, K.P. and Kalpana, G.V. (2002) A masked NES in INI1/hSNF5 mediates hCRM1-dependent nuclear export: implica ons for tumorigenesis. Embo J, 21, 31-42.. Carvalho, C., Pereira, H.M., Ferreira, J., Pina, C., Mendonca, D., Rosa, A.C. and Carmo-Fonseca, M. (2001) Chromosomal G-dark bands determine the spa al organiza on of centromeric heterochroma n in the nucleus. Mol Biol Cell, 12, 3563-3572.. Cronshaw, J.M., Krutchinsky, A.N., Zhang, W., Chait, B.T. and Matunis, M.J. (2002) Proteomic analysis of the mammalian nuclear pore complex. J Cell Biol, 158, 915-927.. Choi, E.Y., Newman, A.E., Burger, L. and Pintel, D. (2005) Replica on of minute virus of mice DNA is cri cally dependent on accumulated levels of NS2. J Virol, 79, 12375-12381.. Cronshaw, J.M. and Matunis, M.J. (2004) The nuclear pore complex: disease associa ons and func onal correla ons. Trends Endocrinol Metab, 15, 34-39.. Chubb, J.R., Boyle, S., Perry, P. and Bickmore, W.A. (2002) Chroma n mo on is constrained by associa on with nuclear compartments in human cells. Curr Biol, 12, 439-445.. Daelemans, D., Afonina, E., Nilsson, J., Werner, G., Kjems, J., De Clercq, E., Pavlakis, G.N. and Vandamme, A.M. (2002) A synthe c HIV-1 Rev inhibitor interfering with the CRM1-mediated nuclear export. Proc Natl Acad Sci U S A, 99, 14440-14445.. Clemens, K.E. and Pintel, D.J. (1988) The two transcrip on units of the autonomous parvovirus minute virus of mice are transcribed in a temporal order. J Virol, 62, 1448-1451. Cohen, S. and Pante, N. (2005) Pushing the envelope: microinjec on of Minute virus of mice into Xenopus oocytes causes damage to the nuclear envelope. J Gen Virol, 86, 3243-3252. Cole, C.N. and Scarcelli, J.J. (2006) Transport of messenger RNA from the nucleus to the cytoplasm. Curr Opin Cell Biol, 18, 299306. Con , E., Uy, M., Leighton, L., Blobel, G. and Kuriyan, J. (1998) Crystallographic analysis of the recogni on of a nuclear localiza on signal by the nuclear import factor karyopherin alpha. Cell, 94, 193-204. Cook, A., Bono, F., Jinek, M. and Con , E. (2007) Structural biology of nucleocytoplasmic transport. Annu Rev Biochem, 76, 647-671. Cornelis, J.J., Becquart, P., Duponchel, N., Salome, N., Avalosse, B.L., Namba, M. and Rommelaere, J. (1988a) Transforma on of human fibroblasts by ionizing radia on, a chemical carcinogen, or simian virus 40 correlates with an increase in suscep bility to the autonomous parvoviruses H-1 virus and minute virus of mice. J Virol, 62, 1679-1686. Cornelis, J.J., Spruyt, N., Spegelaere, P., Gue a, E., Darawshi, T., Cotmore, S.F., Tal, J. and Rommelaere, J. (1988b) Sensi za on of transformed rat fibroblasts to killing by parvovirus minute virus of mice correlates with an increase in viral gene expression. J Virol, 62, 3438-3444. Cotmore, S.F., D’Abramo, A.M., Jr., Carbonell, L.F., Bra on, J. and Ta ersall, P. (1997) The NS2 polypep de of parvovirus MVM is required for capsid assembly in murine cells. Virology, 231, 267280. Cotmore, S.F., Sturzenbecker, L.J. and Ta ersall, P. (1983) The autonomous parvovirus MVM encodes two nonstructural proteins in addi on to its capsid polypep des. Virology, 129, 333-343.. Delphin, C., Guan, T., Melchior, F. and Gerace, L. (1997) RanGTP targets p97 to RanBP2, a filamentous protein localized at the cytoplasmic periphery of the nuclear pore complex. Mol Biol Cell, 8, 2379-2390. Devos, D., Dokudovskaya, S., Williams, R., Alber, F., Eswar, N., Chait, B.T., Rout, M.P. and Sali, A. (2006) Simple fold composi on and modular architecture of the nuclear pore complex. Proc Natl Acad Sci U S A, 103, 2172-2177. Dworetzky, S.I. and Feldherr, C.M. (1988) Transloca on of RNAcoated gold par cles through the nuclear pores of oocytes. J Cell Biol, 106, 575-584. E hymiadis, A., Shao, H., Hubner, S. and Jans, D.A. (1997) Kine c characteriza on of the human re noblastoma protein bipar te nuclear localiza on sequence (NLS) in vivo and in vitro. A comparison with the SV40 large T-an gen NLS. J Biol Chem, 272, 22134-22139. Eichwald, V., Daeffler, L., Klein, M., Rommelaere, J. and Salome, N. (2002) The NS2 proteins of parvovirus minute virus of mice are required for efficient nuclear egress of progeny virions in mouse cells. J Virol, 76, 10307-10319. Emterling, A., Skoglund, J., Arbman, G., Schneider, J., Evertsson, S., Carstensen, J., Zhang, H. and Sun, X.F. (2003) Clinicopathological significance of Nup88 expression in pa ents with colorectal cancer. Oncology, 64, 361-369. Englmeier, L., Fornerod, M., Bischoff, F.R., Petosa, C., Ma aj, I.W. and Kutay, U. (2001) RanBP3 influences interac ons between CRM1 and its nuclear protein export substrates. EMBO Rep, 2, 926-932. Erickson, E.S., Mooren, O.L., Moore, D., Krogmeier, J.R. and Dunn, R.C. (2006) The role of nuclear envelope calcium in modifying nuclear pore complex structure. Can J Physiol Pharmacol, 84, 309-318.. Cotmore, S.F. and Ta ersall, P. (1987) The autonomously replicating parvoviruses of vertebrates. Adv Virus Res, 33, 91-174.. Faria, P.A., Chakraborty, P., Levay, A., Barber, G.N., Ezelle, H.J., Enninga, J., Arana, C., van Deursen, J. and Fontoura, B.M. (2005) VSV disrupts the Rae1/mrnp41 mRNA nuclear export pathway. Mol Cell, 17, 93-102.. Cotmore, S.F. and Ta ersall, P. (1990) Alternate splicing in a parvoviral nonstructural gene links a common amino-terminal sequence to downstream domains which confer radically different. Fischer, U., Huber, J., Boelens, W.C., Ma aj, I.W. and Luhrmann, R. (1995) The HIV-1 Rev ac va on domain is a nuclear export signal that accesses an export pathway used by specific cellular RNAs.. 23.

(17) General Introduction Cell, 82, 475-483.. Rpl10p and the cytoplasmic GTPase Lsg1p. Embo J, 24, 567-579.. Fontoura, B.M., Faria, P.A. and Nussenzveig, D.R. (2005) Viral interac ons with the nuclear transport machinery: discovering and disrup ng pathways. IUBMB Life, 57, 65-72.. Henderson, B.R. and Ele heriou, A. (2000) A comparison of the ac vity, sequence specificity, and CRM1-dependence of different nuclear export signals. Exp Cell Res, 256, 213-224.. Fornerod, M., Boer, J., van-Baal, S., Morreau, H. and Grosveld, G. (1996) Interac on of cellular proteins with the leukemia specific fusion proteins DEK-CAN and SET-CAN and their normal counterpart, the nucleoporin CAN. Oncogene, 13, 1801-1808.. Higa, M.M., Alam, S.L., Sundquist, W.I. and Ullman, K.S. (2007) Molecular characteriza on of the Ran-binding zinc finger domain of Nup153. J Biol Chem, 282, 17090-17100.. Fornerod, M. and Ohno, M. (2002) Expor n-mediated nuclear export of proteins and ribonucleoproteins. Results Probl Cell Differ, 35, 67-91. Fornerod, M., Ohno, M., Yoshida, M. and Ma aj, I.W. (1997a) CRM1 is an export receptor for leucine-rich nuclear export signals. Cell, 90, 1051-1060. Fornerod, M., van-Deursen, J., van-Baal, S., Reynolds, A., Davis, D., Mur , K.G., Fransen, J. and Grosveld, G. (1997b) The human homologue of yeast CRM1 is in a dynamic subcomplex with CAN/ Nup214 and a novel nuclear pore component Nup88. The EMBO Journal, 16, 807-816. Frey, S. and Gorlich, D. (2007) A saturated FG-repeat hydrogel can reproduce the permeability proper es of nuclear pore complexes. Cell, 130, 512-523. Frey, S., Richter, R.P. and Gorlich, D. (2006) FG-rich repeats of nuclear pore proteins form a three-dimensional meshwork with hydrogel-like proper es. Science, 314, 815-817. Gallouzi, I.E., Brennan, C.M. and Steitz, J.A. (2001) Protein ligands mediate the CRM1-dependent export of HuR in response to heat shock. Rna, 7, 1348-1361. Gleizes, P.E., Noaillac-Depeyre, J., Leger-Silvestre, I., Teulieres, F., Dauxois, J.Y., Pommet, D., Azum-Gelade, M.C. and Gas, N. (2001) Ultrastructural localiza on of rRNA shows defec ve nuclear export of preribosomes in mutants of the Nup82p complex. J Cell Biol, 155, 923-936. Goldberg, M.W. and Allen, T.D. (1992) High resolu on scanning electron microscopy of the nuclear envelope: demonstra on of a new, regular, fibrous lace a ached to the baskets of the nucleoplasmic face of the nuclear pores. J Cell Biol, 119, 1429-1440. Gorlich, D., Dabrowski, M., Bischoff, F.R., Kutay, U., Bork, P., Hartmann, E., Prehn, S. and Izaurralde, E. (1997) A novel class of RanGTP binding proteins. J Cell Biol, 138, 65-80. Grüter, P., Tabernero, C., von Kobbe, C., Schmi , C., Saavedra, C., Bachi, A., Wilm, M., Felber, B.K. and Izaurralde, E. (1998) TAP, the human homolog of Mex67p, mediates CTE-dependent RNA export from the nucleus. Molecular Cell, 1, 649-659. Hamm, J., Darzynkiewicz, E., Tahara, S.M. and Ma aj, I.W. (1990) The trimethylguanosine cap structure of U1 snRNA is a component of a bipar te nuclear targe ng signal. Cell, 62, 569-577.. Hodge, C.A., Colot, H.V., Stafford, P. and Cole, C.N. (1999) Rat8p/ Dbp5p is a shu ling transport factor that interacts with Rat7p/ Nup159p and Gle1p and suppresses the mRNA export defect of xpo1-1 cells. EMBO J, 18, 5778-5788. Huber, J., Cronshagen, U., Kadokura, M., Marshallsay, C., Wada, T., Sekine, M. and Luhrmann, R. (1998) Snurpor n1, an m3G-capspecific nuclear import receptor with a novel domain structure. EMBO, 17, 4114-4126. Hu en, S. and Kehlenbach, R.H. (2006) Nup214 is required for CRM1-dependent nuclear protein export in vivo. Mol Cell Biol, 26, 6772-6785. Hu en, S. and Kehlenbach, R.H. (2007) CRM1-mediated nuclear export: to the pore and beyond. Trends Cell Biol, 17, 193-201. Iijima, M., Suzuki, M., Tanabe, A., Nishimura, A. and Yamada, M. (2006) Two mo fs essen al for nuclear import of the hnRNP A1 nucleocytoplasmic shu ling sequence M9 core. FEBS Le , 580, 1365-1370. Ikuta, T., Kobayashi, Y. and Kawajiri, K. (2004) Phosphoryla on of nuclear localiza on signal inhibits the ligand-dependent nuclear import of aryl hydrocarbon receptor. Biochem Biophys Res Commun, 317, 545-550. Imasaki, T., Shimizu, T., Hashimoto, H., Hidaka, Y., Kose, S., Imamoto, N., Yamada, M. and Sato, M. (2007) Structural basis for substrate recogni on and dissocia on by human transpor n 1. Mol Cell, 28, 57-67. Izaurralde, E., Kutay, U., von Kobbe, C., Ma aj, I.W. and Gorlich, D. (1997) The asymmetric distribu on of the cons tuents of the Ran system is essen al for transport into and out of the nucleus. EMBO J, 16, 6535-6547. Izaurralde, E., Lewis, J., Gamberi, C., Jarmolowski, A., McGuigan, C. and Ma aj, I.W. (1995) A cap-binding protein complex mediating U snRNA export. Nature, 376, 709-712. Joseph, J., Liu, S.T., Jablonski, S.A., Yen, T.J. and Dasso, M. (2004) The RanGAP1-RanBP2 complex is essen al for microtubule-kinetochore interac ons in vivo. Curr Biol, 14, 611-617. Kalab, P., Weis, K. and Heald, R. (2002) Visualiza on of a Ran-GTP gradient in interphase and mito c Xenopus egg extracts. Science, 295, 2452-2456.. Hase, M.E. and Cordes, V.C. (2003) Direct interac on with nup153 mediates binding of tpr to the periphery of the nuclear pore complex. Mol Biol Cell, 14, 1923-1940.. Kasper, L.H., Brindle, P.K., Schnabel, C.A., Pritchard, C.E.J., Cleary, M.L. and van Deursen, J.M.A. (1999) CREB binding protein interacts with nucleoporin-specific FG repeats that ac vate transcrip on and mediate NUP98-HOXA9 oncogenicity. Mol Cell Biol, 19, 764 - 776.. Hedges, J., West, M. and Johnson, A.W. (2005) Release of the export adapter, Nmd3p, from the 60S ribosomal subunit requires. Kehlenbach, R.H., Dickmanns, A., Kehlenbach, A., Guan, T. and Gerace, L. (1999) A role for RanBP1 in the release of CRM1 from. 24.

No results found