• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • Other karyopherins besides CRM must


    Other karyopherins besides CRM1 must bind to FG-Nups in a similar fashion. However, at an atomic resolution, only the interaction of importin β with isolated FG motifs has been analyzed (Bayliss et al., 2000, Bayliss et al., 2002, Liu and Stewart, 2005). Despite similarities in the FG-binding pockets of CRM1 and other transport receptors, the export receptor has a particularly high affinity for Nup214 (Figure 1A). From the nucleoporin’s point of view, interactions with transport receptors have to fulfill two opposing functions: first, binding must be strong enough to discriminate between bona fide transport complexes (or empty transport receptors) and inert proteins, whose translocation through the pore should be obstructed. On the other hand, interactions at individual binding sites must be weak to allow release of transport complexes and their translocation within the time frame of milliseconds. Our results clearly show that there are many interaction sites between CRM1 and nucleoporins. Full-length Nup214 contains a total of 44 FG motifs, 32 of which are not present in the fragment that was used for crystallization. Hence, additional contacts between CRM1 and several of these FG sites are likely and may further contribute to a high-avidity interaction. Apart from a study that showed three interaction sites between a nucleoporin fragment and Kap95p (Liu and Stewart, 2005), multiple binding sites on transport receptors for FG motifs have so far only been simulated (Isgro and Schulten, 2005). Importantly, each of the multiple FG-binding pockets, which contact FG motifs on a rather linear, initially unstructured stretch of Proteinase K where of FG-Nups, is expected to contribute only weakly to the overall avidity of the complex. Our structure shows that intervening Nup sequences are hardly attached to the transport receptor and will therefore be flexible upon loosening a single FG contact. Rapid dissociation of single sites, followed by rebinding of the transport receptor to a close-by FG motif (possibly of another nucleoporin) should therefore be feasible. Such association/dissociation cycles should allow the transport complex to overcome the permeability barrier of the NPC. For CRM1 export complexes, GTP hydrolysis on Ran as promoted by cytoplasmic RanGAP ultimately leads to dissociation of the CRM1 export complex from a terminal binding site, e.g., at the cytoplasmic nucleoporin Nup214 (Kehlenbach et al., 1999). With respect to the kinetics of nuclear export, we cannot expect drastic changes upon manipulation of individual FG pockets within the CRM1 molecule, because the overall avidity of the transport receptor for nucleoporins in general will hardly be affected. Indeed, our double mutant CRM1 (D824K/W880A), which showed reduced binding to Nup214 fragments (Figures 5C and 5D), and even the triple mutant CRM1 (A156F/D824K/W880A) were as efficient in promoting nuclear export of GFP-NFAT as the wild-type protein (Figure 5E). The functional assay integrates possible interactions of CRM1 with full-length Nup214 and with all other FG nucleoporins, which may contribute to efficient passage of export complexes through the nuclear pore.
    Experimental Procedures
    Author Contributions
    Introduction Influenza virus, which belongs to family Orthomyxoviridae, replicates its single-stranded negative-sense genomic RNAs in the host nucleus. The viral genomic RNAs are then packed into viral ribonucleoproteins (vRNPs) with viral RNA-dependent RNA polymerase and nucleoproteins (NPs) and exported into cytoplasm for further viral particle assembly and budding at the plasma membrane (Eisfeld et al., 2015). vRNP nuclear export is mediated by the cellular chromosome region maintenance 1 (CRM1) pathway, with assistance from viral matrix protein 1 (M1) and nuclear export proteins (NEP or NS2) (Akarsu et al., 2003, Huang et al., 2013, O\'Neill et al., 1998). The vRNP–M1–NEP–CRM1–RanGTP complex is then exported through the nuclear pore to the cytoplasm, where the viral cargo is spontaneously released upon hydrolysis of RanGTP to RanGDP (Eisfeld et al., 2015).