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).
General Discussion 2
Identification of supraNESs
The work in this thesis started with a simple idea: can we reverse the direcon of nuclear transport receptors? In other words, can we turn an exporn into an imporn and vice versa? As explained in the introducon (Chap- ter 1), the direconality of nuclear transport is imposed by RanGTP: it releases cargo from imporns and helps to bind cargoes to exporns. The experiment was then simply to select from a random pepde library sequences that interact with an exporn without RanGTP and not with RanGTP, and, reversely, to select sequences that in- teract with an imporn in the presence of RanGTP and not without. As is oen the case in science, we found something else than we asked for: supraphysiological or supraNESs that bound in the absence of RanGTP and even beer in its presence.
In retrospect, the idenficaon of these NESs answered completely different quesons: why is the consensus se- quence of nuclear export signals so loose and their bind- ing to their receptor CRM1 so weak, even in the presence of RanGTP. Our data, presented in Chapter 2 suggests that NESs have to be weak in order to dissociate correctly from CRM1 and allow efficient recycling of CRM1. It also suggests that the haunt for an NES consensus sequence is likely to be fruitless. There might be one or more ‘blue- print’ NESs, from which the NESs must deviate to pre- vent themselves from being too strong. If more than one
‘blueprint’ sequences exist, it may indicate that NESs can interact with CRM1 at different sites. Further, I propose that idenficaon of other supraphysiological affinity in- teracons can be used to detect transion state interac-
ons in vivo that are normally too transient to detect.
SupraNESs in nature
A queson that intrigued us from the moment we had found a supraNES was whether these signals could be found in nature i.e. had also evolved outside the test tube.
Considering their likely cytotoxicity in interphase cells (Chapter 3) we hypothesized two potenal funcons for supraNESs: either in proteins funconing in metaphase, where nucleocytoplasmic transport is halted, or in viral proteins, which care less about damaging cells.
For the first possibility, a supraNES in a mitoc proteins, we considered Survivin. This protein was reported to be targeted to kinetochores (Skoufias et al., 2000), and con- tained a putave NES dependent on CRM1 (Knauer et al., 2006; Rodriguez et al., 2002). In addion, CRM1 was reported to be present at kinetochores and funcon in mitoc spindle formaon and chromosome segregaon (Arnaoutov et al., 2005).
A nuclear export signal in Survivin
Using purified proteins we tested the strength of the CRM1/Survivin interacon. Disappoinngly, instead of a supraphysiological binding, we found hardly any bind- ing at all. How could this be, as several authors had pub- lished on CRM1-mediated nuclear export of Survivin and even idenfied an NES (Colnaghi et al., 2006; Knauer et al., 2006; Knauer et al., 2007; Stauber et al., 2006)? A hint came from looking at the Survivin crystal structure (Chantalat et al., 2000; Muchmore et al., 2000; Verde- cia et al., 2000), which indicated that the idenfied NES makes part of the Survivin homodimerizaon interface.
As a first crude experiment, we repeated the in vitro binding experiments in the presence of moderate con- centraons of urea, which might break homodimeriza-
on and expose the NES. Indeed, in the presence of urea, CRM1 binding was greatly enhanced. This experiment paved the way for Chapter 5, where we show that Sur- vivin contains a (regular strength) NES that is modulated by homodimerizaon.
Our data also indicate that studies that concluded that nuclear export acvity is required for the mitoc func-
on of Survivin ((Knauer et al., 2006; Knauer et al., 2007;
Stauber et al., 2006) should be taken with a grain of salt. In these reports, mutaons were made in the Sur- vivin NES region, and mitoc funcon was found to be affected (Colnaghi et al., 2006). However, these point mutaons might not only affect the NES but also the di- merizaon of Survivin or its interacon with other pro- teins. Indeed, recently two reports have been published on the structure of Survivin in complex with two other chromosomal passenger proteins, Borealin and INCENP, that likely represents the form that is acve in mitosis.
These studies indicate that the Survivin homodimeriza-
on region is also involved in contacts with Borealin and INCENP (Bourhis et al., 2007; Jeyaprakash et al., 2007).
Furthermore we showed that point mutaons that were claimed to inhibit CRM1-mediated export (Knauer et al., 2006) were unaffected in CRM1 binding and CRM1-me- diated export. Our studies therefore imply that Survivin’s CRM1 interacon is irrelevant for Survivin’s mitoc func-
on, supporng work from Wheatley et al. (Colnaghi et al., 2006). Instead, mutaons in the NES region affect mitoc Survivin acvity by interfering with other mitoc binding partners.
A supraNES in MVM NS2
The second possibility of encountering supraNESs in a natural proteins was in to find them in virusses. The MVM NS2 NES was frequently used as a prototype NES by us and others, and we wanted to mutate this “regu- lar” NES into a supraphysiological one. Surprisingly, the NS2 NES without gain-of-funcon mutaons was already
General Discussion
100
General Discussion
localized at the NE, hinng to the presence of a supraNES in NS2. At the same me, a paper was published, describ- ing CRM1 relocalizaon by NS2 expression (Lopez-Bueno et al., 2004). As shown in Chapter 3, the NS2 NES pos- sesses all features of a supraNES. We showed that the supraphysiological nature of the NES is esenal for NS2 funcon in enhancing capsid nuclear export and viral fit- ness. However, several queson remain to be answered:
1. Is the the NE localization of NS2 relevant for its function?
SupraNESs localize at the NPC to Nup358 (Figure 4C Chapter 2 and Figure 1F Chapter 3). Strong knockdown of Nup358 however, never leads to a complete removal of supraNESs from the NE. This could be caused by an incomplete removal of the Nup358, but alternavely, NS2 may also be aached to other FG nucleoporins, e.g.
Nup214 (see below) or Nup62. Interesngly, EM in com- binaon with immunogold labeling to localize NS2, occa- sionally covered the enre inner NPC with gold parcles (unpublished observaons), consistent with this idea.
Alternavely, this may represent a “traffic jam” of export cargo, caused by lack of release from Nup358.
What is the funcon of the NE localizaon? To study the funcon of the NS2/NPC interacon, we have tested sev- eral possibilies. First, we studied if NS2 by locang at the NPC, could damage the NE or NPC integrity, which would favor MVM egress. However, no difference in diffusion speed across NPCs in NS2-expressing cells was seen and EM analysis of NS2-expressing cells did not reveal any al- teraon in pore size or NE connuaon (unpublished ob- servaons). Second, the NE interacon could be required for transport of MVM through the NPC. Careful analysis of MVM-infected cells, showed that NS2 colocalized with MVM and CRM1 in nuclear clusters, but these complexes were never observed at the NE (Figure 4C Chapter 3). This suggests that MVM/NS2/CRM1 complexes translocate through the pore rather quickly and that export-compe- tent virions are formed in the nucleus and not at the NE.
To rule out a funcon for NS2 at the NE, the effect of Nup358 depleon might be studied, but most likely, the Nup358 localizaon is an epiphenomenon.
2. Is CRM1 relocalization by NS2 relevant for its function?
The first observaon that NS2 could harbor a supraNES was reported in a study by Lopez-Bueno et al. showing a strong relocalizaon of CRM1 in MVM infected cells (Lopez-Bueno et al., 2004). Besides the strong nuclear reducon of CRM1, it was somemes increased at the NE and in nuclear clusters. As expected, NS2 has been found to localize at the same subcellular structures (Chapter 3). Therefore, the altered CRM1 localizaon is a consequence of NS2 localizaon, and also likely to be an epiphenomenon.
3. Is CRM1 inhibition by NS2 relevant for its func- tion?
Upon overexpression of NS2, CRM1 funcon is clearly
inhibited as demonstrated by NMD3 and NFκB localiza-
on (Figure 3, Chapter 3). This inhibion however is only observed in cells expressing high NS2 levels (Figure S2, Chapter 3). Interesngly, in cells infected with MVM a clear nuclear accumulaon of NFkB is observed (data not shown). This effect however, was also seen in cells expressing supraNES mutants and is therefore likely a stress response. To study to funconality of the CRM1 inhibion, a rescue experiment was performed. Expres- sion of the NS2 supraNES in trans could not rescue infec-
on with MVMNES22, a virus with a regular affinity NES in NS2. Even more, instead of an increase in infecvity, a significant decrease was observed (Figure 3D Chapter 3).
This suggests that an exogenously expressed supraNES competes with the NS2 supraNES, demonstrang that the supraNES only funcons within the context of NS2.
Altogether, it seems that CRM1 inhibion is not the ma- jor funcon of the supraNES, although it may contribute to its cytotoxicity or evasion of host immune response.
4. What is the relationship between the NS2 su- praNES and evolved NS2 mutations in immuno- supressed mice?
Lopez-Bueno et al. (2004) reported that mutaons in the region close to the NES resulted in an enhanced CRM1 interacon and infecvity. However, when tested in the RanGAP assay, these mutants did not demonstrate an increased affinity (our unpublished data). Mutaons at posion 150 to 153 of NS2, quite far away from the NES, showed a phenotype similar to mutaons imme- diately downstream of the NES (Lopez-Bueno et al., 2004). Because these mutaons have been shown to be responsible for 14-3-3 binding (J.M. Almendral personal communicaon), mutaons close the NES might also influence 14-3-3 binding. CRM1 and 14-3-3 cannot bind NS2 simultaneously (Bodendorf et al., 1999). Therefore, these mutants might decrease 14-3-3 binding and there- by enhance CRM1 binding.
Do supraNESs exist in other proteins?
Two cellular proteins have been described that, under certain condions, exhibit a much higher affinity for CRM1 or display other “supraphysiological” features. In addion several viral proteins are good candidates to contain supraNESs.
1. Snurportin
Snurporn binds CRM1 with an affinity similar to supra- NESs (see 3.7.2). However, unlike supraNESs, Snurporn uses a rather large domain of at least 159 residues. Be- cause of the high affinity, Snurporn/CRM1 heterodi- mers are rather stable, although this interacon is sll enhanced by RanGTP. Snurporn is normally not local- ized at the NPC (Paraskeva et al., 1999). However, in as- says using semi-permeabilized HeLa cells, a strong NPC localizaon of Snurporn could be observed, which may reveal an export intermediate as seen for supraNESs (Paraskeva et al., 1999).
2. NMD3
Another protein that has been reported to possess a high affinity for CRM1 is the 60S ribosomal subunit adapter NMD3. Wild-type NMD3 interacts with CRM1 via its C- terminal domain that contain two NES-like sequences.
The supraphysiological NMD3/CRM1 interacon is me- diated by a larger domain, encompassing the two NESs, and is smulated by RanGTP. Interesngly, in yeast, Nmd3p mutants that are deficient in ribosome binding, bind CRM1 with increased affinity and without RanGTP.
In vivo, the Nmd3p mutants are localized to the NPC, dependent on hydrophobic residues in the C-terminus (West et al., 2007). In vitro, wild-type NMD3 does not dis- play high affinity towards CRM1. However, addion of an N-terminal z-tag increases the affinity to supraphysiologi- cal levels (Thomas and Kutay), Chapter 2). Perhaps the fusion with the z-tag converts the affinity of the NMD3 to supraphysiological levels by forcing a conformaonal change in the protein. Structural analyses of NMD3 mu- tants in complex with CRM1 could give an explanaon for this affinity behavior.
3. Hepatitis B HBx
The X protein (HBx) of Hepas B is an oncogenic protein that causes genec instability by interfering with proper centrosome duplicaon (Forgues et al., 2001; Wen et al., 2008). HBx contains a bona fide NES (see Table 1) and has also been shown to interact with CRM1 in vitro. In- teresngly, HBx shows several features of containing a supraNES. 1) HBx colocalizes with CRM1 at the NE, 2) HBx expression relocalizes CRM1 to the cytoplasm and interacts with CRM1 in the cytoplasm, 3) HBx expression inhibits export of NFκB. The isolated NES of HBx how- ever, does not demonstrate these features, suggesng addional residues outside the NES sequence are also important (Forgues et al., 2001). As discussed above, during mitosis, CRM1 is localized at centrosomes, where it plays a role in the inhibion of centrosome replicaon.
During Hepas B infecon, HBx has been proposed to cause aneuploidy by inhibing CRM1, prevenng CRM1 to fulfill its inhibitory role in centrosome duplicaon (Forgues et al., 2003). It would be interesng to study if supraNESs in general cause aneuploidy, which could be tested by looking at the chromosome number upon over- expression of a supraNES.
4. Alphavirus CVEE
Venezuelan equine encephalis virus (VEEV) is a member of the alphaviruses and is an enveloped small virues of 70 nm and the nucleocapsid is 40 nm. It can affect all equine species and is transmied via mosquitos. In equines, VEEV can cause dead or central nervous system disorders.
In case of human infecons, flu-like symptoms occur and immunosuppressed people and very old or yould people can even die from it. One of the capsid proteins of the VEEV, C¬¬VEE has been shown to colocalize with NPCs at the NE and at annulate lamellae. This localizaon is medi- ated by an α-helix comprised of residues 34-51. Interest-
ingly this α-helix encompasses an NES-like sequence (see Table 1), which is responsible for its export and the NPC interacon. An NLS is located downstream of this α-helix and this enre region is responsible for the downregula-
on of cellular transcripon (Garmashova et al., 2007).
Recently, this region has also been shown to block NLS- and M9-containing cargo (Atasheva et al., 2008).
In conclusion, supraNESs appear to be present in dif- ferent viruses and likely have different specialized func-
ons. In the future more supraNESs are expected to be idenfied in viruses, and could be targets for therapeuc intervenon, for example by leptomycin B. In eukaryoc cells, addional proteins with supraphysiological affinity for CRM1 may also exist, but they are likely more difficult to idenfy as the strength must be regulated to shield its supraphysiological characteriscs.
Could supraNESs serve to export large cargo in general?
MVM nuclear export in mouse fibroblasts requires (1) CRM1, because egress is inhibited by LMB (Maroto et al., 2004; Miller and Pintel, 2002) (2) the NS2 supraNES, because cells infected with the NS2 NES22 mutant, harboring a regular NES NS2 show a strong decrease in capsid egress ((Eichwald et al., 2002) and Chapter 3) (3) the externalizaon of the N-terminal pepide of MVM VP2, 2Nt, which itself is dependent on virion maturaon (Maroto et al., 2004). 2Nt-conjugated BSA is exported in MVM infected cells in a leptomycin B-sensive man- ner (Maroto et al., 2004), indicang that 2Nt is neces- sary and sufficient for CRM1-mediated nuclear egress.
We have also shown that NS2 and CRM1 co-localize with mature MVM capsids in the nucleus (Chapter 3). The simplest model derived from these observaons is that 2Nt directly interacts with NS2, and that NS2 funcons as an nuclear export adaptor for CRM1. The size of an MVM parcle is on the border of what can be exported through the NPC (see Chapter 1). Large cargo, like MVM may need the high affinity interacon with CRM1 to de- crease the off-rate in the nucleus and at the NPC, allow- ing more me and strength to drag the capsid through the NPC. Interesngly, a similarity can be seen with ribo- somal export. First, the 60S subunit is similar in size (±25 nm). Second the 60S ribosomal subunit also requires an export adaptor, NMD3 that under certain circumstances exhibits supraphysiological characteriscs. These find- ings jusfy the speculaon that supraNESs could func-
on in exporng large cargo in general. Further reduc-
onist experiments are required to test this hypothesis.
We have aempted to obtain further experimental sup- port for the simplest adaptor model of MVM nuclear export by exogenously expressing GFP-2Nt together with NS2. However, no nuclear export of 2Nt-GFP was observed under these condions. Also we did not detect NS2 in pull down experiments with 2Nt pepde columns or found associaon of NS2 with isolated MVM capsids (unpublished data). Clearly, the nuclear export of MVM
102
General Discussion
is more complicated and future experiments should be directed towards idenficaon of addional factors that might bridge or influence the NS2/virion associaon.
Release me!
Upon RanGTP hydrolysis in the cytoplasm, supraNES cargo will exhibit a prolonged interacon with CRM1 as a results of the RanGTP-independent binding. This in- teracon is unfavorable, especially in a non-pathogenic situaon, because 1) it will block CRM1 and thereby impair many important export processes and 2) it may lead to re-import of the cargo/receptor complex. As a consequence supraNES-containing cargo should require a release mechanism in the cytoplasm. In the next sec-
on I will hypothesize on possible release mechanisms of supraphysiological cargo.
1. Parvovirus capsid
As described above, 14-3-3 and CRM1 cannot bind NS2 simultaneously. Moreover, CRM1 interacts with non- phosphorylated NS2 both in the nucleus and in the cy- toplasm (Bodendorf et al., 1999). 14-3-3 on the other hand, only binds phosporylated NS2, which is only pres- ent in the cytoplasm (Bodendorf et al., 1999; Brockhaus et al., 1996). Upon arrival at the cytoplasmic side of the NPC or entry of MVM/NS2/CRM1 complexes in the cyto- plasm, NS2 may be phosphorylated and subsequent 14- 3-3 binding may lead to dissociaon of the complex. This hypothesis is supported by the finding that NS2 14-3-3 mutants are stronger localized to the nuclear envelope, but also to the nucleus (our unpublished data), which could be explained by increased re-import. As discussed above, NS2 mutants selected in immunosuppressed mice are characterized by increased infecvity and CRM1 binding (Lopez-Bueno et al., 2004), and the laer may be caused by decreased compeon of 14-3-3. One should therefore also consider ming of nuclear accumulaon of NS2 in the increased infecvity of NS2 mutants. Per- haps, when NS2 enters the nucleus earlier during the vi- ral life cycle, all processes may be shied, resulng in an increased infecvity.
2. NMD3
Wild-type Nmd3 does not demonstrate features of a su- praNES, neither in vivo nor in vitro (West et al., 2007).
As discussed above, ribosome binding mutants of Nmd3 do demonstrate these features. Possibly, without ribo- somal binding, wild-type Nmd3p adopts a conformaon in which addional CRM1 interacng regions are shield- ed. 60S ribosomal subunit binding to Nmd3 may expose these regions and enhance the CRM1/Nmd3 interacon required for export of the large cargo (see above). The ri- bosomal protein Rpl10 and the cytoplasmic GTPase Lsg1 mediate the release of the 60S subunit in the cytoplasm (Hedges et al., 2005). Binding of these protein may smu- late a conformaonal change in Nmd3 that lowers the affinity for CRM1.
3. Snurportin 1
In the absence of RanGTP, Snurporn sll interacts with CRM1. In vitro binding assays however, demonstrate that m3G-cap addion completely abolish the CRM1 interac-
on. Furthermore, CRM1 and the m3G-cap are mutually exclusive for Snurporn binding (Paraskeva et al., 1999).
This suggests that dissociaon of CRM1/Snurporn com- plexes in the cytoplasm may be aided by mature m3G- cap-containing snRNP parcles.
In conclusion, data from the literature are consistent with the idea that endogenous proteins like NMD3 and Snurporn require efficient release mechanisms to dis- charge their interacon with CRM1. In viruses these re- lease mechanisms may be less efficient or even absent as cytotoxicity, by blocking CRM1 or obstrucng the NPC, may not be a disadvantage or even benefit the virus.
Other factors involved in transporting large cargo
Large cargo has been shown to slow down the transport rate in order to increase the efficiency (Ribbeck and Gor- lich, 2002). In the next paragraph I will outline several alternave or cooperave mechanisms that increase the transport efficiency of large cargo.
1. Multiple transport receptors
The binding of one transport receptor may not provide enough hydrophobicity to translocate through the hy- drophobic meshwork of the inner NPC. In yeast, besides Xpo-1, the mRNA export factor Mex67-Mtr2 and Arx1 are also responsible for transport of the 60S subunit.
Although Arx1 is not a classical transport factor, it can interact with FG-repeats of Nup100 and Nup116. It has been suggested that mulple transport receptors are required to aid the export of large cargo (Bradatsch et al., 2007; Yao et al., 2007). This in itself may suffice to explain the high affinity of CRM1 to Nmd3, as three transport receptors have to bind to the cargo similtane- ously. The same may be true for MVM. Mulple copies of CRM1 may cover the icosahedral capsid, which would facilitate its transport. Besides altering cargo-receptor properes, export of large cargo could also be facilitated by alteraons at the level of the NPC.
2. Sliding mechanisms within the NPC
As described in Chapter 1 (General Introducon), two abundant internally located nucleoporins, Nup58 and Nup45, can alter the pore size by intermolecular slid- ing of their α-helices. It has been suggested that other nucleoporins by virtue of their α-helices also exhibit similar sliding, leading to a further expansion of the pore diameter. Nup62 and Nup54 are closely associated with Nup58 and Nup45 and also contain α-helices that may allow sliding (Guan et al., 1995; Melcak et al., 2007). In- teresngly, the yeast homologues of Nup45/Nup58 and Nup62 (Nup49 and Nsp1), which also contain α-helices, are required for export of the large ribosomal subunit
(Hurt 1999 JCB). Two other coiled-coils-containing nucle- oporins are Nup214 and Nup88, which also interact with each other via these helices (Fornerod et al., 1997a).
In Chapter 6 we describe a role for the Nup214/Nup88 subcomplex in ribosomal export. By tesng deleon mu- tants we idenfied the coiled-coils of Nup214 to be ex- clusively responsible for this export defect. Interesngly, a region including the coiled-coils but excluding the FG repeat of Nup159, the yeast homologue of Nup214, was required for 60S preribosomal nuclear export in yeast (Gleizes et al., 2001), suggesng a conservaon of this mechanism throughout eukaryoc life.
Thus, together these data suggests that diameter of the NPC can be regulated by coiled-coil-containing nucleo- porins. This expansion might be required when large cargo like ribosomes is transported.
3. Calcium
The diameter of the pore is also thought to be influenced by changes in Ca2+ levels. Several studies have shown ei- ther a structural change of the NPC, in parcular of the nuclear basket, or a funconal change, leading to chang- es in permeability and nuclear transport ((Erickson et al., 2006; Thorogate and Torok, 2007) and references there- in). Experiments with thapsigargin, a drug that results in Ca2+ efflux from the ER into the cytosol, similar to iono- mycin, led to a decrease in nuclear transport. Interest- ingly, passive diffusion of molecules was decreased. This suggests that a decrease in pore diameter, rather than a decrease in specific receptor-nucleoporin interacons is responsible for this effect (Greber and Gerace, 1995). Us- ing GFP as a diffusible marker, no effect of calcium deple-
on was detected (Wei et al., 2003).
Two calcium sensors have been found to be localized at the NPC. The first one is glycoprotein 210 (gp210), an in- tegral membrane nucleoporin (anchor-Nup, see Figure 3 Chapter 1) that contains several calcium binding mofs.
Anbodies against the luminal part of gp210 resulted in a decrease in passive diffusion and protein import, sug- gesng this protein may regulate pore diameter (Erick- son et al., 2006; Greber and Gerace, 1992).
The second protein is Centrin-2, which is thought to in- teract with the scaffold nucleoporin complex, Nup170/
Nup106. Upregulaon of Centrin-2 results in a mRNA ex- port block. It has been hypothesized that this is caused by a local Ca2+ drop, leading to a decrease in pore diam- eter (Resendes et al., 2008). Finally, studies with atomic force microscopy in Xenopus oocytes revealed structural changes in central mass and diameter upon depleon of cisternal calcium stores (Erickson et al., 2006).
Although a role for calcium in nuclear transport has been demonstrated several mes, the precise mechanism has not yet been clarified. And further studies are needed to study how large cargo communicates with or influences calcium storages.
Controversy: the Role of Nup214 in nuclear ex- port
In mulple organisms, CRM1 has been shown to strongly interact with the FG-repeat region of Nup214 (Fornerod et al., 1997b; Huen and Kehlenbach, 2006; Xylourgidis et al., 2006; Zeitler and Weis, 2004). Mulple funcons have been proposed for this interacon. 1) Nup214 is the site of export complex dissociaon. Aer being ex- ported to the cytoplasmic side of the NPC, RanGAP and the RanBP1 like domains on the flexible Nup358 parci- pate in RanGTP hydrolysis. 2) Inhibion of re-import of CRM1-cargo complexes, to prevent NES-cargo from be- ing shuled back into nucleus before RanGTP hydroly- sis. 3) Smulated recycling of empty CRM1 (Huen and Kehlenbach, 2007). 4) Aenuaon of CRM1 export. In Drosophila, CRM1 has also been shown to bind to the FG-repeat of DNup214. Deleon of DNup214 results in a increase in NES-cargo export, while CRM1 accumulated in the nucleus. This suggests that the CRM1/Nup214 interacon funcons in negavely regulang export by aenuang recycling of CRM1 back to the nucleus (Xy- lourgidis et al., 2006).
In Chapter 6 we have studied the role of Nup214 in CRM1-mediated export and discovered that general NES-containing cargoes were not compromised upon Nup214/Nup88 knockdown. In a recent study by Kehlen- bach et al. (Huen and Kehlenbach, 2006) however, an export defect was observed for two independent car- goes tested, NFAT-GFP and Rev-GR-GFP. NFAT shules to the nucleus in response to calcium influx, which can be achieved by ionomycin addion. Subsequently, the ionomycin is washed out and export of NFAT is studied.
Upon Nup214 knockdown, a long-lasng NFAT-GFP ex- port block was observed. As described above, changes in calcium concentraon has been reported to influence the pore diameter. Therefore, the observed export block could be caused by the change in calcium levels. Export of the second cargo Rev-GR-GFP, was only mildly inhib- ited (Huen and Kehlenbach, 2006), and not inconsistent with minor effects seen in our data. Importantly, it re- mains to be shown whether the reported export defects are mediated by absence of the Nup214 FG repeats, as suggested, or by the coiled-coil domains. As described above and in Chapter 6, our observed NMD3-GFP export defect was not influenced by the FG-repeat region of Nup214, but by the coiled-coils. I favour the idea that the observed export defect by Huen et al. is a consequence of changes in pore diameter rather than a depleon of the Nup214/CRM1 interacon.
Conclusion
CRM1-mediated nuclear export influences a wide vari- ety of cellular processes. Considering this, the domains responsible for CRM1-mediated export are surprisingly small. In this thesis we have gone beyond the addion to the collecon of NESs and provided a deeper insight into the nature of the NES. Acquiring this fundamental knowledge has lead to insights into viral and preribo- somal nuclear export.
104
General Discussion
References
Arnaoutov, A., Azuma, Y., Ribbeck, K., Joseph, J., Boyarchuk, Y., Karpova, T., McNally, J. and Dasso, M. (2005) Crm1 is a mitoc ef- fector of Ran-GTP in somac cells. Nat Cell Biol, 7, 626-632.
Atasheva, S., Garmashova, N., Frolov, I. and Frolova, E. (2008) Venezuelan equine encephalis virus capsid protein inhibits nuclear import in Mammalian but not in mosquito cells. J Virol, 82, 4028-4041.
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.
Bourhis, E., Hymowitz, S.G. and Cochran, A.G. (2007) The mitoc regulator Survivin binds as a monomer to its funconal interactor Borealin. J Biol Chem, 282, 35018-35023.
Bradatsch, B., Katahira, J., Kowalinski, E., Bange, G., Yao, W., Se- kimoto, T., Baumgartel, V., Boese, G., Bassler, J., Wild, K., Peters, R., Yoneda, Y., Sinning, I. and Hurt, E. (2007) Arx1 funcons as an unorthodox nuclear export receptor for the 60S preribosomal subunit. Mol Cell, 27, 767-779.
Brockhaus, K., Plaza, S., Pintel, D.J., Rommelaere, J. and Salome, N. (1996) Nonstructural proteins NS2 of minute virus of mice as- sociate in vivo with 14-3-3 protein family members. J Virol, 70, 7527-7534.
Chantalat, L., Skoufias, D.A., Kleman, J.P., Jung, B., Dideberg, O.
and Margolis, R.L. (2000) Crystal structure of human survivin reveals a bow e-shaped dimer with two unusual alpha-helical extensions. Mol Cell, 6, 183-189.
Colnaghi, R., Connell, C.M., Barre, R.M. and Wheatley, S.P.
(2006) Separang the an-apoptoc and mitoc roles of survivin.
J Biol Chem, 281, 33450-33456.
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.
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.
Forgues, M., Difilippantonio, M.J., Linke, S.P., Ried, T., Nagashima, K., Feden, J., Valerie, K., Fukasawa, K. and Wang, X.W. (2003) In- volvement of Crm1 in hepas B virus X protein-induced aber- rant centriole replicaon and abnormal mitoc spindles. Mol Cell Biol, 23, 5282-5292.
Forgues, M., Marrogi, A.J., Spillare, E.A., Wu, C.G., Yang, Q., Yoshi- da, M. and Wang, X.W. (2001) Interacon of the hepas B virus X protein with the Crm1-dependent nuclear export pathway. J Biol Chem, 276, 22797-22803.
Fornerod, M., van Baal, S., Valenne, V., Shapiro, D.N. and Gros- veld, G. (1997a) Chromosomal localizaon of genes encoding CAN/Nup214-interacng proteins--human CRM1 localizes to 2p16, whereas Nup88 localizes to 17p13 and is physically linked to SF2p32. Genomics, 42, 538-540.
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.
Garmashova, N., Atasheva, S., Kang, W., Weaver, S.C., Frolova, E.
and Frolov, I. (2007) Analysis of Venezuelan equine encephalis virus capsid protein funcon in the inhibion of cellular tran- scripon. J Virol, 81, 13552-13565.
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 localizaon of rRNA shows defecve nuclear ex- port of preribosomes in mutants of the Nup82p complex. J Cell Biol, 155, 923-936.
Greber, U.F. and Gerace, L. (1992) Nuclear protein import is inhib- ited by an anbody to a lumenal epitope of a nuclear pore complex glycoprotein. J Cell Biol, 116, 15-30.
Greber, U.F. and Gerace, L. (1995) Depleon of calcium from the lumen of endoplasmic reculum reversibly inhibits passive diffu- sion and signal-mediated transport into the nucleus. J Cell Biol, 128, 5-14.
Guan, T., Muller, S., Klier, G., Pante, N., Blevi, J.M., Haner, M., Paschal, B., Aebi, U. and Gerace, L. (1995) Structural analysis of the p62 complex, an assembly of O-linked glycoproteins that lo- calizes near
the central gated channel of the nuclear pore complex. Mol Biol Cell, 6, 1591-1603.
Hedges, J., West, M. and Johnson, A.W. (2005) Release of the ex- port adapter, Nmd3p, from the 60S ribosomal subunit requires Rpl10p and the cytoplasmic GTPase Lsg1p. Embo J, 24, 567-579.
Huen, S. and Kehlenbach, R.H. (2006) Nup214 is required for CRM1-dependent nuclear protein export in vivo. Mol Cell Biol, 26, 6772-6785.
Huen, S. and Kehlenbach, R.H. (2007) CRM1-mediated nuclear export: to the pore and beyond. Trends Cell Biol, 17, 193-201.
Jeyaprakash, A.A., Klein, U.R., Lindner, D., Ebert, J., Nigg, E.A.
and Con, E. (2007) Structure of a Survivin-Borealin-INCENP core complex reveals how chromosomal passengers travel together.
Cell, 131, 271-285.
Knauer, S.K., Bier, C., Habtemichael, N. and Stauber, R.H. (2006) The Survivin-Crm1 interacon is essenal for chromosomal pas- senger complex localizaon and funcon. EMBO Rep, 7, 1259- 1265.
Knauer, S.K., Kramer, O.H., Knosel, T., Engels, K., Rodel, F., Ko- vacs, A.F., Dietmaier, W., Klein-Hitpass, L., Habtemichael, N., Sch- weitzer, A., Brieger, J., Rodel, C., Mann, W., Petersen, I., Heinzel, T. and Stauber, R.H. (2007) Nuclear export is essenal for the tumor-promong acvity of survivin. Faseb J, 21, 207-216.
Lopez-Bueno, A., Valle, N., Gallego, J.M., Perez, J. and Almendral, J.M. (2004) Enhanced cytoplasmic sequestraon of the nuclear export receptor CRM1 by NS2 mutaons developed in the host regulates parvovirus fitness. J Virol, 78, 10674-10684.
Maroto, B., Valle, N., Saffrich, R. and Almendral, J.M. (2004) Nuclear export of the nonenveloped parvovirus virion is directed
by an unordered protein signal exposed on the capsid surface. J Virol, 78, 10685-10694.
Melcak, I., Hoelz, A. and Blobel, G. (2007) Structure of Nup58/45 suggests flexible nuclear pore diameter by intermolecular sliding.
Science, 315, 1729-1732.
Miller, C.L. and Pintel, D.J. (2002) Interacon between parvovirus NS2 protein and nuclear export factor Crm1 is important for viral egress from the nucleus of murine cells. J Virol, 76, 3257-3266.
Muchmore, S.W., Chen, J., Jakob, C., Zakula, D., Matayoshi, E.D., Wu, W., Zhang, H., Li, F., Ng, S.C. and Aleri, D.C. (2000) Crystal structure and mutagenic analysis of the inhibitor-of-apoptosis protein survivin. Mol Cell, 6, 173-182.
Paraskeva, E., Izaurralde, E., Bischoff, F.R., Huber, J., Kutay, U., Hartmann, E., Lührmann, R. and Görlich, D. (1999) CRM1-medi- ated recycling of snurporn 1 to the cytoplasm. Journal of Cell Biology, 145, 255-264.
Resendes, K.K., Rasala, B.A. and Forbes, D.J. (2008) Centrin 2 lo- calizes to the vertebrate nuclear pore and plays a role in mRNA and protein export. Mol Cell Biol, 28, 1755-1769.
Ribbeck, K. and Gorlich, D. (2002) The permeability barrier of nuclear pore complexes appears to operate via hydrophobic ex- clusion. Embo J, 21, 2664-2671.
Rodriguez, J.A., Span, S.W., Ferreira, C.G., Kruyt, F.A. and Giacco- ne, G. (2002) CRM1-mediated nuclear export determines the cy- toplasmic localizaon of the anapoptoc protein Survivin. Exp Cell Res, 275, 44-53.
Skoufias, D.A., Mollinari, C., Lacroix, F.B. and Margolis, R.L. (2000) Human survivin is a kinetochore-associated passenger protein. J Cell Biol, 151, 1575-1582.
Stauber, R.H., Rabenhorst, U., Rekik, A., Engels, K., Bier, C. and Knauer, S.K. (2006) Nucleocytoplasmic shuling and the biologi- cal acvity of mouse survivin are regulated by an acve nuclear export signal. Traffic, 7, 1461-1472.
Thomas, F. and Kutay, U. (2003) Biogenesis and nuclear export of ribosomal subunits in higher eukaryotes depend on the CRM1 export pathway. J Cell Sci, 116, 2409-2419.
Thorogate, R. and Torok, K. (2007) Role of Ca2+ acvaon and bilobal structure of calmodulin in nuclear and nucleolar localiza-
on. Biochem J, 402, 71-80.
Verdecia, M.A., Huang, H., Dul, E., Kaiser, D.A., Hunter, T. and Noel, J.P. (2000) Structure of the human an-apoptoc protein survivin reveals a dimeric arrangement. Nat Struct Biol, 7, 602- 608.
Wei, X., Henke, V.G., Strübing, C., Brown, E.B. and Clapham, D.E.
(2003) Real-me imaging of nuclear permeaon by EGFP in single intact cells. Biophys J, 84, 1317-1327.
Wen, Y., Golubkov, V.S., Strongin, A.Y., Jiang, W. and Reed, J.C.
(2008) Interacon of hepas B viral oncoprotein with cellular target HBXIP dysregulates centrosome dynamics and mitoc spindle formaon. J Biol Chem, 283, 2793-2803.
West, M., Hedges, J.B., Lo, K.Y. and Johnson, A.W. (2007) Novel interacon of the 60S ribosomal subunit export adapter Nmd3 at
the nuclear pore complex. J Biol Chem, 282, 14028-14037.
Xylourgidis, N., Roth, P., Sabri, N., Tsarouhas, V. and Samakov- lis, C. (2006) The nucleoporin Nup214 sequesters CRM1 at the nuclear rim and modulates NFkappaB acvaon in Drosophila. J Cell Sci, 119, 4409-4419.
Yao, W., Roser, D., Kohler, A., Bradatsch, B., Bassler, J. and Hurt, E.
(2007) Nuclear export of ribosomal 60S subunits by the general mRNA export receptor Mex67-Mtr2. Mol Cell, 26, 51-62.
Zeitler, B. and Weis, K. (2004) The FG-repeat asymmetry of the nuclear pore complex is dispensable for bulk nucleocytoplasmic transport in vivo. J Cell Biol, 167, 583-590.
