Kap-Centric control of nuclear pores based on promiscuous binding to FG nucleoporins

Nuclear pore complexes (NPCs) are remarkable molecular machines that perforate the nuclear envelope (NE) in eukaryotic cells and mediate the rapid bidirectional traffic of hundreds of proteins, ribonucleoproteins, and metabolites across the nuclear envelope. Their enormous structure is composed of multiple copies of 30 different proteins (Nups) that add up to 60 – 120 MDa of mass depending on the organism. Each NPC contains a 50 nm-diameter central channel through which only molecules smaller than ~40 kDa or ~5 nm in size can diffuse passively. The movement of larger molecules is impaired by a permeability barrier generated by ~200 partly intrinsically disordered phenylalanine-glycine (FG)-rich nucleoporins (FG Nups) that are tethered to the NPC transport channel surface. These FG Nups interact promiscuously with nuclear transport receptors (NTRs), such as karyopherins (Kaps; e.g. Kap-beta1) or NTF2, that mediate rapid trafficking of cargoes.
Given that the number of FG repeats per FG Nup also varies from 5 to ~50, NTR-FG Nup binding involves highly multivalent interactions, which are generally known to impart a strong avidity that enhances stability and specificity. However, this is paradoxical in the context of the NPC, because the high submicromolar Kap-beta1-FG domain binding affinities predict slow off rates (given a diffusion-limited on rate) that contradict the rapid (~5 ms) in vivo dwell time. As this implies, Kap-FG binding ought to be sufficiently strong to ensure selectivity but also weak enough to promote fast translocation through the NPC. Nonetheless, an explanation as to how promiscuous binding of FG Nups to NTRs is balanced against the mechanistic control of the FG domain barrier is still lacking.
The purpose of my work was to investigate FG Nup-NTR binding promiscuity and multivalency by measuring the interaction kinetics, binding affinity and in situ associated conformational changes in Nsp1p FG domains when binding NTF2 and Kap-beta1, both separately and together. Experimentally, this was achieved by using a novel surface plasmon resonance technique to correlate in situ mechanistic changes (molecular occupancy and conformational changes) with FG Nup-NTR binding.
The obtained results show that surface-tethered Nsp1p FG domains form molecular brushes at physiological conditions. Kap-beta1 binding provokes brush extension while partitioning into a fast and slow kinetic phase, where the latter may form an integral part of the FG domain barrier. In contrast, NTF2 binding to pristine Nsp1p layers induced collapse, but not under competing interactions from Kap-beta1. Therefore, promiscuous binding of NTF2 to Kap-beta1-preloaded Nsp1p attenuates NTF2 towards higher off rates and more transient interactions.
My work demonstrates that promiscuous binding of NTRs to FG Nups ought to influence nucleocytoplasmic transport. This depends on the concentration, size and binding strength of each NTR. Indeed, some form of hierarchy may exist between different NTRs such that their relative concentrations may impact NPC barrier function. This interpretation departs from the conventional view that the FG Nups alone form the NPC permeability barrier. Rather I conclude that concentrating NTRs in the NPC transport channel also contributes to generating crowding-based selective barrier function of the pore.