Overturning in vitro models depicting a solid FG-nucleoporin gel, this study proves the nuclear pore barrier is liquid in living cells—maintained by O-GlcNAcylation and importin-β—unifying decades of competing models of selective nuclear transport.
FG-nucleoporins maintain nanosecond-scale liquid-like mobility in live-cell nuclear pores through cooperative stabilization by O-GlcNAcylation and importin-β, which together prevent the liquid-to-solid transition observed in vitro, while importin-β enrichment at the NPC periphery creates a radial gradient that leaves extended FG domains in the central channel to form the permeability barrier and explains the peripheral cargo transport trajectories observed in recent high-resolution studies.
The nuclear pore complex (NPC) regulates the molecular traffic between nucleus and cytoplasm. Its permeability barrier is formed by intrinsically disordered proteins (IDPs) known as FG-nucleoporins (FG-NUPs), whose physical state has long been debated. Deciphering how FG-NUPs behave inside living cells is crucial for understanding how the NPC achieves selective and rapid transport. Here, by combining site-specific labelling with picosecond time-resolved fluorescence anisotropy, we reveal that FG domains exhibit nanosecond-scale, liquid-like mobility in live cells, yet undergo a liquid-to-solid transition in vitro. Experiments and coarse-grained molecular dynamics simulations further show that importin-β, a major nuclear transport receptor, and O-linked β-N-acetylglucosamine (O-GlcNAc), a key post-translational modification of FG-NUPs together stabilise the dynamic FG network. Finely balanced FG–FG interactions modulated by O-GlcNAcylation, along with FG–importin-β interactions, maintain liquidity and enrich importin-β near the NPC periphery, while extended FG domains remain in the central channel to form the transport barrier. These findings reconcile conflicting models of FG-NUP organisation and explain the recent observations in high-resolution MINFLUX studies of importin-β depletion from the pore centre in mammalian cells. Beyond resolving debates over FG-NUP behaviour, our study underscores the importance of studying IDPs in their cellular context, with broader implications for understanding IDP-related diseases, including viral infections, cancer, and neurodegenerative disorders.