Past research

CRIME domain proteins for CRM1, importin β, etcetera.

Many years ago, in the lab Gerard Grosveld (St. Jude's Children's Research Hospital, Memphis, USA), we identified proteins that interacted with the nucleoporin CAN/Nup214. The CAN or CAIN gene was originally identified as a chromosomal translocation target on 9q34 (next to c-Abl) in acute myeloid leukemia (von Lindern et al., 1992). One of the CAN/Nup214-interacting proteins was the human homologue of a yeast protein named CRM1, that we proposed to be a new transport factor. This proposal was based on experimental data, and on sequence homology with the import receptor importin β. We identified a group of related proteins with homology to both CRM1 and importin β that was proposed new family of transport factors (Fornerod et al., 1997a).

Interaction assay on RanGTP beads showing that CRM1 binding to the NES of HIV-1 Rev is RanGTP-dependent and leptomycin B sensitive. S, supernatant; P, pellet

Next, in the lab of Iain Mattaj (EMBL, Heidelberg, Germany), together with Mutsuhito Ohno, we found that CRM1 was an export receptor for "leucine rich" nuclear export signals. We showed that CRM1 binds directly to the export signal, and that this can be inhibited by the fungal cytotoxin leptomycin B (LMB). Leptomycin B was also shown to directly bind to CRM1. NES binding was only stable upon cooperative binding of the small GTPase Ran in its GTP-bound form. This suggested the mechanism with which the RanGTP gradient across the nuclear envelope directs transport of NES-containing (ribonucleo) proteins, and we proposed that this mechanism would be more generally applicable to nuclear export processes.

RanGTP imposes directionality to nuclear import (left) and export (right). Adapted from Fornerod et al., 1997b.

Indeed, we could show that a trimeric CRM1-NES-RanGTP complex is disassembled by Ran cofactors that are present at the cytoplasmic side of the NPC. With a quantitative CRM1/NES-cargo binding assay, we demonstrated that there are significant differences among natural NESs in affinity for CRM1, suggesting that the steady state nucleo-cytoplasmic distribution of a shuttling protein could be determined by the relative strength of its NES (Askjaer et al., 1999).

With Gert-Jan Arts, post-doc in Iain Mattaj's lab, we identified the nuclear export receptor for tRNAs, exportin-t, another member of the importin β family. Similar to CRM1, exportin-t bound directly to its tRNA cargo in a RanGTP-dependent manner (Arts et al., 1998).

Refocussing on the nuclear pore complex, we have extended the analysis of the CAN/CRM1 interaction, and demonstrated that the interaction dependents on RanGTP, suggesting a role of this interaction in export complex disassembly and CRM1 reimport (Askjaer et al., 1999).

To get a handle on the functional significance of nucleoporin/transport receptor interactions, we made use of the Xenopus nuclear reconstitution system. Egg extracts of the african clawed frog, Xenopus laevis, support efficient nuclear formation around demembrenated sperm chromatin in vitro. These nuclei contain numerous nuclear pore complexes, and actively import protein substrates.

Fluorescently labelled BSA-NLS imported into in vitro reconstituted nuclei.

When nucleoporin Nup153 is depleted from the extracts, prior to nuclear assembly, nuclei are formed that are deficient in importin β mediated nuclear import but not in transportin-mediated nuclear import, revealing different nucleoporin requirements for different nuclear import pathways (Walter et al., 2001).