Nuclear pores are massive protein assemblies that puncture the double membrane surrounding every cell’s nucleus, acting as the sole gateways for molecules moving between the nucleus and the cytoplasm. Each human cell has roughly a thousand of these structures, and every one is built from about 30 different proteins arranged into a barrel-shaped channel. Their job is both simple and extraordinarily demanding: let the right molecules through while keeping everything else out, and do it fast enough to handle millions of transport events per minute across the entire cell. How they manage this selective, high-speed trafficking involves a gel-like barrier, molecular address labels, and an energy gradient that tells cargo which direction to go.
Anatomy of the Pore
A nuclear pore complex, or NPC, is one of the largest molecular machines in any cell. Each one is assembled from proteins called nucleoporins, present in multiples of eight copies per pore, giving the whole structure an eight-fold rotational symmetry when viewed from above.1bioRxiv. Structure and mechanics of the human Nuclear Pore Complex basket Picture an octagonal barrel set into the nuclear envelope, with extensions reaching into both the cytoplasm and the nucleus.
The core scaffold consists of three stacked rings embedded in the nuclear membrane, plus a ring that sits in the space between the two membrane layers. Extending outward from the cytoplasmic face are flexible filaments, which help capture incoming cargo and recycling transport receptors.2Journal of Cell Biology. Ran-dependent docking of importin-β to RanBP2/Nup358 filaments is essential for protein import and cell viability On the nuclear side, a basket-like structure projects inward, and one of its key proteins, Tpr, plays a direct role in exporting messenger RNA.3bioRxiv. Distinct Basket Nucleoporins roles in Nuclear Pore Function and Gene Expression So the pore is not a simple hole; it is an asymmetric structure with distinct architecture on each side, tailored to the different jobs performed at each face.
The Gel That Decides Who Passes
The central channel of the NPC is not empty. It is packed with long, floppy protein tails called FG-nucleoporins (FG-Nups), named for the repeating phenylalanine-glycine sequences that stud their length. These tails are intrinsically disordered, meaning they do not fold into a fixed shape. Instead, they wave around and interact with one another, forming a mesh that behaves like a selective hydrogel.4PubMed. Phase separation of FG-nucleoporins in nuclear pore complexes Recent work has shown that purified FG-Nups can undergo liquid-liquid phase separation into droplets that reproduce the permeability properties of a real pore, which has given researchers a powerful way to study the barrier outside the cell.5PubMed Central. Phase separation of intrinsically disordered FG-Nups is driven by highly dynamic FG motifs
Small molecules and ions slip through this mesh freely by passive diffusion. The traditional textbook cutoff for passive entry was about 40 kilodaltons for a globular protein, with anything larger supposedly locked out. The real picture is messier. Experimental work has found that some proteins well above 60 kilodaltons can still diffuse through.6PubMed Central. The maximal size of protein to diffuse through the nuclear pore is larger than 60kDa Rather than a hard molecular-weight wall, the barrier is better understood as a soft limit: molecules with a diameter larger than about 5 nanometers face a steeply rising energy cost that makes passage increasingly unlikely.7PLoS ONE. Energetics of Transport through the Nuclear Pore Complex Shape, surface charge, and flexibility all matter, so a rigid spherical protein and a floppy one of the same mass can have very different odds of slipping through.
What makes the barrier genuinely selective, though, is how it treats transport receptors. These receptors have surfaces that interact with the FG repeats, essentially melting their way through the mesh. In an elegant demonstration, researchers designed an entirely synthetic FG-Nup from scratch, coated nanopores with it, and showed that a yeast transport receptor passed through almost as easily as through a bare pore, while a similarly sized non-transport protein was nearly completely blocked.8Nature Communications. A designer FG-Nup that reconstitutes the selective transport barrier of the nuclear pore complex The selectivity is baked into the chemistry of the FG mesh itself.
Getting Into the Nucleus
Large proteins that need to reach the nucleus carry a molecular address tag called a nuclear localization signal, or NLS. The classic version is a short stretch of positively charged amino acids embedded in the protein’s sequence.9PubMed Central. Types of nuclear localization signals and mechanisms of protein import into the nucleus This tag is recognized by a transport receptor complex in the cytoplasm. In the best-studied pathway, an adaptor protein called importin alpha grabs onto the NLS, and importin beta then latches onto importin alpha, creating a three-part complex ready for transit.10PubMed Central. Classical nuclear localization signals: definition, function, and interaction with importin alpha
Importin beta does the heavy lifting at the pore. Its surface interacts with the FG repeats lining the channel, allowing the entire complex to thread through the mesh. Once the complex reaches the nuclear side, a small protein called Ran (in its GTP-bound form) binds to importin beta and pries it away from the cargo. The cargo protein is released into the nucleus, and the import receptors are recycled back to the cytoplasm to pick up a new load.
Getting Out of the Nucleus
Export works by a mirror-image logic. Proteins destined for the cytoplasm carry a nuclear export signal, or NES, which is typically a short stretch rich in hydrophobic amino acids like leucine.11PubMed Central. Mechanistic insights from the recent structures of the CRM1 nuclear export complex and its disassembly intermediate The most versatile export receptor is CRM1 (also known as exportin 1), which binds a wide range of protein and ribonucleoprotein cargoes. Inside the nucleus, CRM1 grabs both the NES-bearing cargo and a molecule of Ran-GTP, forming a three-part export complex. This complex crosses the pore, and once it reaches the cytoplasm, Ran’s GTP is hydrolyzed to GDP, which causes the complex to fall apart and the cargo to be released.
The Ran Gradient Sets the Compass
Both import and export depend on the same small protein, Ran, but Ran exists in two states: bound to GTP or bound to GDP. The key to the whole system is that these two forms are not evenly distributed. Ran-GTP is concentrated inside the nucleus, while Ran-GDP dominates in the cytoplasm.12PubMed Central. RanGTPase: A Key Regulator of Nucleocytoplasmic Trafficking This asymmetry is maintained by two enzymes stuck in their respective compartments: one in the nucleus that loads GTP onto Ran, and one in the cytoplasm that triggers Ran to burn its GTP into GDP.
You can think of this gradient as a kind of transport potential energy. A high Ran-GTP concentration in the nucleus drives export by allowing export complexes to assemble there, while the low Ran-GTP in the cytoplasm drives import by ensuring import complexes fall apart only once they arrive in the nucleus.13Current Biology. Nucleocytoplasmic transport: Driving and directing transport It is an elegant trick: one gradient provides directionality for traffic moving in both directions simultaneously. And the dependence on Ran is so fundamental that when researchers artificially flooded the cytoplasm with Ran-GTP, they could actually reverse the direction of transport through the pore.14PubMed Central. The direction of transport through the nuclear pore can be inverted
RNA Takes a Different Route
Messenger RNA export does not use the Ran-GTP gradient the same way protein cargo does. Instead, mRNA relies on a dedicated receptor called NXF1, which assembles onto messenger ribonucleoprotein particles in the nucleus and guides them through the pore. On the cytoplasmic face, an RNA helicase called DBP5 strips NXF1 off the mRNA once it emerges, preventing it from sliding back. This stripping step is what makes the process irreversible and directional.15PubMed Central. Nuclear export factor RBM15 facilitates the access of DBP5 to mRNA Think of it like a ratchet: the mRNA can move forward through the pore, but once DBP5 acts, it cannot go backward.
This separation of RNA export from the standard Ran-dependent machinery makes biological sense. Messenger RNA molecules are enormous compared with most protein cargoes, and they need to be remodeled as they cross, shedding some associated proteins and picking up new ones. Having a dedicated pathway with its own quality-control steps ensures that only properly processed transcripts make it to the cytoplasm for translation.
Pores That Breathe
Nuclear pores are not rigid tunnels. High-resolution imaging of yeast cells showed that NPCs reversibly switch between a dilated and a constricted state. In actively growing cells, the pores tend to sit in a wider conformation, but when cells are starved of energy or placed under osmotic stress, the pores constrict.16PubMed. Nuclear pores dilate and constrict in cellulo The mechanism appears to be linked to nuclear envelope membrane tension rather than an active motor within the pore itself.
This matters more than it might sound. Modeling work suggests that even modest changes in pore diameter can alter transport rates exponentially, meaning a small constriction could slash molecular traffic by orders of magnitude.17PubMed Central. Role of pore dilation in molecular transport through the nuclear pore complex: Insights from polymer scaling theory The cell may use this as a volume knob, dialing transport up or down in response to metabolic conditions. Recent simulations have begun to describe the structural transitions between dilated and constricted states in detail, identifying low-frequency collective motions of the scaffold that smoothly interconvert the two conformations.18PubMed Central. Conformational Dynamics of the Nuclear Pore Complex: Insights from Elastic Network Models and Coarse-Grained Simulations
Beyond dilation, nucleoporins are subject to chemical modifications that tune their behavior. A sugar modification called O-GlcNAcylation helps stabilize several key nucleoporins. When the enzyme responsible for adding this sugar tag is removed, nucleoporins like Nup62 and Nup93 are degraded much faster than normal, and the pore’s selective barrier begins to break down.19Journal of Molecular Cell Biology. Post-translational O-GlcNAcylation is essential for nuclear pore integrity and maintenance of the pore selectivity filter Another modification, monoubiquitylation of the basket nucleoporin Nup60, is triggered by DNA damage and helps regulate the cell’s repair response.20PubMed Central. Posttranslational marks control architectural and functional plasticity of the nuclear pore complex basket The pore, in other words, is not just a passive gateway; it is a sensor and responder that the cell can remodel on the fly.
Building New Pores
Cells need to make new NPCs, both during cell division and during normal growth. These two situations use fundamentally different assembly strategies. After a cell divides and the nuclear envelope reforms around the separated chromosomes, pore components are inserted into the reassembling membrane in a rapid, large-scale process. Between divisions, when a growing cell needs more pores to handle increasing traffic, new NPCs are assembled into an intact, already-closed nuclear envelope, a much slower process that requires the membrane to be bent and fused from both sides.21PubMed Central. Mechanisms of nuclear pore complex assembly – two different ways of building one molecular machine The structural intermediates, the regulatory proteins involved, and the types of membrane bending are all different between these two assembly modes.
Jobs Beyond Transport
Nucleoporins have turned out to do far more than ferry cargo. Several of them interact directly with chromatin, helping to organize which genes are active and which are silenced. Some nucleoporins tether active genes near the nuclear periphery, positioning them close to pores for efficient mRNA export, while others help maintain the boundary between open and condensed chromatin regions.22PubMed Central. Functional interactions between nucleoporins and chromatin There is also growing evidence that nucleoporins contribute to genome stability, and that mutations or fusions involving nucleoporin genes show up in various cancers.23PubMed Central. Nucleoporins in Cancer: Functional Roles and Therapeutic Opportunities The NPC is increasingly understood as a hub that integrates transport, gene regulation, and genome maintenance.
What Happens When Pores Deteriorate
Not all nucleoporins turn over at the same rate. Some, like Nup153, are continuously replaced. But the scaffold nucleoporins that form the core rings can persist for the entire lifetime of a non-dividing cell, making them some of the longest-lived proteins in the body. In neurons and other long-lived post-mitotic cells, this means the same scaffold proteins can accumulate damage over decades. Research in aging rat neurons revealed that deteriorating NPCs become progressively leaky, allowing cytoplasmic proteins to seep into the nucleus where they do not belong. Oxidative damage to scaffold nucleoporins was particularly pronounced in old cells.24PubMed Central. Age-dependent deterioration of nuclear pore complexes causes a loss of nuclear integrity in postmitotic cells
This age-related leakiness connects to a broader story in neurodegeneration. Impaired nucleocytoplasmic transport has emerged as a contributing mechanism in amyotrophic lateral sclerosis (ALS) and other neurodegenerative diseases, where proteins that should be in one compartment end up mislocalized to the other.25PubMed Central. Lost in Transportation: Nucleocytoplasmic Transport Defects in ALS and Other Neurodegenerative Diseases Whether pore deterioration is a cause or a consequence of disease is still debated, but the correlation is striking enough that some researchers are exploring whether restoring normal transport could slow neurodegeneration.
Viruses That Hijack the Pore
Several viruses have evolved ways to sabotage nuclear pore function to benefit their own replication. Poliovirus and human rhinovirus encode proteases that specifically chop FG-nucleoporins like Nup62, removing the very FG-repeat domains that form the selective barrier.26Journal of Biological Chemistry. Specific Cleavage of the Nuclear Pore Complex Protein Nup62 by a Viral Protease The poliovirus 2A protease also cleaves Nup98 and Nup153, shutting down the export of host mRNAs, ribosomal RNAs, and small nuclear RNAs from the nucleus to the cytoplasm.27PubMed. RNA nuclear export is blocked by poliovirus 2A protease and is concomitant with nucleoporin cleavage The strategy is ruthlessly effective: by crippling the host cell’s ability to export its own mRNA, the virus ensures that the cell’s protein-making machinery is free to translate viral RNA instead.
HIV-1 takes a different approach. Its cone-shaped capsid, which encloses the viral genome, is physically large enough that it should be excluded by the pore. Yet the capsid interacts directly with FG-Nups in the central channel, and its shape, orientation, and flexibility all influence whether it can squeeze through. Modeling work has shown that the capsid maximizes contacts with the FG-Nups during translocation, and that flexibility in its protein lattice helps it adapt to the stresses of passage.28PubMed Central. HIV-1 capsid shape, orientation, and entropic elasticity regulate translocation into the nuclear pore complex Understanding this process in detail is a major goal for antiviral drug development.
Ancient Evolutionary Roots
Where did nuclear pores come from? A landmark structural analysis found that the protein folds making up key NPC scaffold components are strikingly similar to the protein coats that wrap transport vesicles inside cells. Both use the same architectural motifs: beta-propeller folds, alpha-solenoid folds, or a distinctive combination of both.29PubMed Central. Components of coated vesicles and nuclear pore complexes share a common molecular architecture The implication is that both the NPC and the vesicle-trafficking system descended from a shared ancestor, an early membrane-curving module that helped the first eukaryotic cells develop their internal compartments. The nuclear pore, in this view, is not an invention from scratch but a repurposed version of very ancient membrane-bending machinery.
Implications for Drug Design
Many anticancer drugs and gene therapies need to reach the nucleus to work. Getting past the nuclear pore is one of the hardest steps. A drug or nanoparticle circulating in the bloodstream must first enter the cell, escape from internal compartments, cross the cytoplasm, and then somehow pass the NPC’s selective barrier. Researchers are now designing nanoparticles decorated with nuclear localization signals or other peptides that mimic the cell’s own import machinery, essentially tricking the pore into letting the therapeutic cargo through.30PubMed Central. Strategies on the nuclear-targeted delivery of genes These “smart” vectors can dramatically improve how much of a gene therapy construct actually reaches its target DNA. The approach is still largely experimental, but the NPC has become one of the most important barriers that nanomedicine researchers are learning to negotiate.31Journal of Drug Delivery Science and Technology. Overcoming the barriers of nuclear-targeted drug delivery using nanomedicine-based strategies for enhanced anticancer therapy