Nuclear pores are massive protein channels that stud the surface of every cell’s nucleus, acting as the sole gateways between the nucleus (where DNA lives) and the cytoplasm (where most cellular work happens). Each human cell contains roughly 2,000 to 5,000 of these structures, collectively called nuclear pore complexes, or NPCs. They are among the largest molecular machines in biology, built from about 30 different proteins present in many copies each, and they handle all regulated traffic in and out of the nucleus. But recent research has revealed that NPCs do far more than shuttle molecules; they sense mechanical forces, help repair damaged DNA, and deteriorate in ways linked to aging and neurodegenerative disease.
How a Nuclear Pore Is Built
A nuclear pore complex sits in a hole where the two layers of the nuclear envelope fuse together. The structure is organized into four concentric rings stacked on top of one another: a cytoplasmic ring facing the cell’s interior, an inner ring and a luminal ring sitting within the plane of the nuclear membrane itself, and a nuclear ring facing the inside of the nucleus. Each ring displays an approximate eightfold rotational symmetry, meaning the whole assembly looks roughly the same if you rotate it by 45 degrees.1PubMed Central. Structure of cytoplasmic ring of nuclear pore complex by integrative cryo-EM and AlphaFold Flexible protein filaments extend from the cytoplasmic face, and a basket-like structure hangs off the nuclear side. The whole assembly is enormous by molecular standards, with an estimated mass around 120 million daltons in vertebrates.
Resolving the full atomic structure of the NPC has been one of the grand challenges of structural biology. Recent work has combined cryo-electron tomography, chemical crosslinking, and AI-assisted modeling to produce increasingly detailed maps of how the component proteins, called nucleoporins or “nups,” fit together.2PubMed Central. How the TREX-2 complex associates with the nuclear pore Still, much of the pore’s interior remains hard to pin down structurally, because the proteins that fill the central channel are intrinsically floppy. That floppiness turns out to be the whole point.
The Selective Barrier Inside the Channel
The central channel of the NPC is not an open hole. It is filled with a tangled mesh of proteins called FG-nucleoporins (FG-Nups), named for the repeating phenylalanine-glycine motifs that decorate their long, flexible tails. These proteins are intrinsically disordered, meaning they have no fixed three-dimensional shape. Instead, they wave around constantly, forming a gel-like barrier that blocks most large molecules from passing through while allowing small ones to slip past.3PubMed Central. Deciphering the intrinsically disordered characteristics of the FG-Nups through the lens of polymer physics
How exactly this barrier works has been debated for years, but a clearer picture is emerging. The FG motifs on these disordered proteins act as short-lived sticky patches that constantly make and break contact with one another, forming what researchers describe as a “percolated network.” This network behaves somewhat like a condensed phase, similar to how oil separates from water. In yeast, the FG-Nups located deeper inside the central channel (mostly the GLFG type) form this dense, dynamic mesh, while the FG-Nups at the entrance and exit of the channel (mostly the FxFG type) behave more like bristles on a brush, waving around and creating an additional barrier through sheer motion.4PubMed Central. Phase separation of intrinsically disordered FG-Nups is driven by highly dynamic FG motifs The result is a two-layered sieve: brush-like proteins at the gates and a dense gel in the middle.
What Gets Through Without Help and What Needs an Escort
Small molecules, ions, and proteins up to a certain size can drift through the pore by passive diffusion. The traditional textbook number for the cutoff was about 40 kilodaltons (roughly the size of a small enzyme), but experimental work has shown the limit is considerably more generous than that. Studies using artificial proteins of known sizes found that proteins well above 60 kilodaltons could still diffuse through the nuclear pore, suggesting that the barrier is not a hard wall but more of a gradient: the bigger you are, the slower you get through, until at some point you essentially cannot pass at all without assistance.5PubMed Central. The maximal size of protein to diffuse through the nuclear pore is larger than 60kDa
Larger cargo, and cargo that needs to move quickly or in a specific direction, relies on dedicated transport receptors. These receptor proteins (called importins when they carry cargo into the nucleus and exportins when they carry cargo out) have a special trick: their surfaces can interact directly with the FG repeats lining the pore channel. By hopping from one FG motif to the next, these receptors essentially dissolve into and through the barrier, dragging their cargo along for the ride.6PubMed. Structural and Functional Characterization of CRM1-Nup214 Interactions Reveals Multiple FG-Binding Sites Involved in Nuclear Export The cargo itself does not need to interact with the FG mesh; it just needs to be bound to a receptor that can.
How the Cell Gives Transport a Direction
The pore itself has no moving parts and does not push cargo in one direction. Directionality comes instead from a clever chemical trick involving a small protein called Ran. Ran exists in two forms: one bound to the energy molecule GTP and the other bound to GDP (the spent form of GTP). The cell maintains a steep gradient of these two forms across the nuclear envelope. Inside the nucleus, an enzyme loads GTP onto Ran, keeping Ran-GTP levels high. In the cytoplasm, a different enzyme stimulates Ran to burn its GTP into GDP, keeping Ran-GDP levels high there.7PubMed Central. RanGTPase: A Key Regulator of Nucleocytoplasmic Trafficking
This gradient drives directionality because transport receptors respond differently to the two forms of Ran. An importin carrying cargo into the nucleus will encounter high Ran-GTP once inside, which causes it to release its cargo. An exportin, by contrast, only grabs its cargo when Ran-GTP is present, so it loads up inside the nucleus and releases the cargo when it reaches the cytoplasm and Ran-GTP is converted to Ran-GDP. In both cases, the Ran gradient ensures that cargo is picked up on one side and dropped off on the other. The energy that maintains this whole system ultimately comes from GTP hydrolysis in the cytoplasm, which makes the disassembly of export complexes effectively irreversible and keeps the whole cycle running in one direction.8PubMed Central. The strategy for coupling the RanGTP gradient to nuclear protein export9PubMed. Mechanistic Insights from Structural Analyses of Ran-GTPase-Driven Nuclear Export of Proteins and RNAs
Getting Messenger RNA Out of the Nucleus
Protein transport is only half the story. Every messenger RNA molecule made in the nucleus must travel through a nuclear pore to reach the ribosomes in the cytoplasm where it will be translated into protein. RNA export uses some of the same general principles as protein transport, but it also has its own dedicated machinery. Messenger RNAs do not travel naked; they are coated in proteins, forming particles called messenger ribonucleoproteins (mRNPs). These particles are too large to diffuse passively and need active, receptor-mediated transport.
A critical step in mRNA export happens right at the cytoplasmic face of the pore. An enzyme called Dbp5, belonging to the DEAD-box family of proteins, sits at the cytoplasmic filaments and strips specific proteins off the mRNP as it emerges from the pore. This remodeling step is essential: by removing nuclear-side binding proteins from the mRNA, Dbp5 prevents the mRNA from sliding back through the pore, making export unidirectional.10PubMed. The DEAD-box protein Dbp5 controls mRNA export by triggering specific RNA:protein remodeling events The activity of Dbp5 itself is tightly regulated by pore components: a nucleoporin called Nup159 and a cofactor called Gle1 (activated by a small signaling molecule, inositol hexakisphosphate) work together to control when and how Dbp5 fires.11PubMed Central. The Dbp5 cycle at the nuclear pore complex during mRNA export II: nucleotide cycling and mRNP remodeling by Dbp5 are controlled by Nup159 and Gle1 The whole arrangement is a kind of molecular ratchet, ensuring that once an mRNA starts coming out, it keeps going.
Nuclear Pores as Mechanical Sensors
One of the more surprising recent findings is that nuclear pores are not rigid structures. They can physically expand and contract in response to mechanical forces on the nuclear envelope. When the nucleus is stretched, such as when a cell migrates through tight spaces or when it is placed in a low-salt solution that causes it to swell, the pore diameter increases. Under compression or in high-salt conditions, the pore constricts.12PubMed Central. On the nuclear pore complex and its emerging role in cellular mechanotransduction In one set of experiments using the organism Dictyostelium, pores dilated to about 77.5 nanometers in diameter under low-salt (swelling) conditions and constricted to about 67.8 nanometers under high-salt (shrinking) conditions.13Molecular Cell. What Are Nuclear Pores? Structure and Function – Section: Results
These changes may sound small, but modeling work suggests the relationship between pore size and transport rate is exponential: even modest widening can ramp up the flow of molecules by several orders of magnitude.14PubMed Central. Role of pore dilation in molecular transport through the nuclear pore complex: Insights from polymer scaling theory This means the NPC could act as a mechanosensor, translating physical forces on the cell into changes in gene expression by altering how freely transcription factors and signaling molecules move between the cytoplasm and nucleus. The idea is still being tested, but it opens an entirely new dimension of NPC function beyond simple gatekeeping.
How Nuclear Pores Are Assembled and Maintained
Cells build new nuclear pores through at least two distinct pathways. One occurs at the end of cell division, when the nuclear envelope reforms around the separated chromosomes and pores must be rapidly inserted into the freshly assembled membrane. The other occurs during interphase, the long growth period between divisions, when the nucleus expands and additional pores are added to the expanding envelope. A systematic study using fluorescently tagged nucleoporins in living human cells revealed that these two assembly routes use strikingly different molecular strategies, even inverting the order in which major structural modules are added.15bioRxiv. A quantitative map of nuclear pore assembly reveals two distinct mechanisms
Once assembled, NPCs are not simply left alone. The cell has quality control systems that monitor pore integrity at multiple stages. Before nucleoporins even reach the nuclear envelope, cytoplasmic checks help enforce the correct protein-protein interactions and stoichiometry. After assembly, additional mechanisms survey for defective pores, and recent work has shown that damaged or aberrant NPCs can be targeted for destruction through a form of selective autophagy, a process dubbed “NPC-phagy,” where the cell’s recycling machinery engulfs and degrades individual pore complexes in the lysosome or vacuole.16PubMed Central. NPC-phagy: selective autophagy of the nuclear pore complexes17PubMed. Changing the guard-nuclear pore complex quality control
Nuclear Pore Breakdown in Aging and Neurodegeneration
In cells that divide regularly, old or damaged pore components get diluted out and replaced with fresh ones. But in cells that rarely divide, like neurons, NPCs can persist for the entire life of the cell. A landmark study found that some nucleoporins in rat neurons were among the longest-lived proteins ever measured in a mammal, with essentially no turnover over the animal’s lifespan. Critically, these long-lived pore components accumulated oxidative damage over time, leading to increased nuclear “leakiness,” meaning cytoplasmic proteins that should be excluded from the nucleus began seeping in.18PubMed Central. Age-dependent deterioration of nuclear pore complexes causes a loss of nuclear integrity in postmitotic cells
This deterioration has real consequences. In amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), particularly forms caused by the C9ORF72 repeat expansion mutation, researchers have found multiple breakdowns in nuclear transport. The toxic products of this mutation, including repeat RNA and dipeptide repeat proteins, interact with and disrupt several elements of the transport machinery, including importins, the Ran gradient, and the nucleoporins themselves.19PubMed Central. Emerging Connections between Nuclear Pore Complex Homeostasis and ALS Arginine-rich dipeptide repeat proteins, for instance, bind to and inhibit the import receptor importin-β, choking off the normal flow of proteins into the nucleus.20Neurotherapeutics. Nuclear Pore Dysfunction in Neurodegeneration – Section: ALS/FTD Separately, a protein called CHMP7, which normally helps remodel the nuclear envelope, has been found to accumulate abnormally in the nucleus in both sporadic and familial ALS, initiating a cascade of NPC injury and subsequent mislocalization of the RNA-binding protein TDP-43, a hallmark of ALS pathology.21PubMed Central. Nuclear accumulation of CHMP7 initiates nuclear pore complex injury and subsequent TDP-43 dysfunction in sporadic and familial ALS The emerging picture is that nuclear pore dysfunction is not just a bystander in neurodegeneration but an active contributor to disease progression.
How Viruses Exploit Nuclear Pores
Because DNA viruses typically need to get their genomes into the host nucleus to replicate, nuclear pores are prime targets for viral exploitation. Different viruses have evolved remarkably different strategies to do this. Herpes simplex virus type 1, for example, rides the cell’s microtubule highways to the nuclear surface, where its capsid docks directly to a specific nucleoporin called CAN/Nup214. A capsid protein called pUL25 acts as the interface between the virus and the pore, and once docked, the capsid injects its DNA through the pore channel without the capsid itself entering.22PubMed Central. Herpesvirus capsid association with the nuclear pore complex and viral DNA release involve the nucleoporin CAN/Nup214 and the capsid protein pUL25 The virus essentially treats the NPC as a docking station and hypodermic needle.
HIV-1 takes a more dramatic approach. Its cone-shaped capsid actually enters the nuclear pore channel, propelled by interactions with FG-nucleoporins. A 2025 study using advanced cryo-electron tomography showed that the capsid’s hexagonal lattice remains largely intact as it pushes into and through the central channel, and that the NPC scaffold rings frequently crack during the process. Computer simulations suggest the pore must widen beyond its resting diameter to accommodate the capsid, and the unique cone shape of HIV’s capsid helps initiate entry and force the rings apart.23PubMed. Passage of the HIV capsid cracks the nuclear pore This was a surprising finding: the virus does not sneak through the pore so much as break it open. Viral capsids larger than the pore’s resting diameter generally need to disassemble before or during passage, a common theme across DNA viruses.24PubMed Central. Virus strategies for passing the nuclear envelope barrier
Roles Beyond Transport
The NPC’s job description keeps expanding. One of the more surprising additions is a role in DNA repair and genome stability. When the cell suffers certain kinds of DNA damage, particularly breaks in repetitive sequences or in densely packed chromatin regions, the damaged sites physically relocate to the nuclear periphery and associate with nuclear pore complexes. At the pore, specific repair pathways are favored over others, and the pore environment seems to act as a quarantine zone that prevents dangerous recombination events between similar repetitive sequences.25PubMed. On the edge: how nuclear pore complexes rule genome stability Some nucleoporins even leave the pore and travel to damaged sites elsewhere in the nucleus, where they stabilize repair-associated chromatin states and restrain potentially harmful RNA intermediates.26PubMed Central. The nuclear pore complex as a spatial organizing hub for high-risk DNA lesions
NPCs also influence gene regulation directly. Certain genes are physically tethered to nuclear pores when they are active, and recent work has found that pores preferentially associate with “super-enhancer” regions, clusters of regulatory DNA that control the expression of key cell-identity genes.27eLife. NPC-DamID: a new method to probe NPC-genome interactions at the nuclear periphery – Section: NPCs associate with super-enhancer regions in different cell types The NPC is not just a doorway; it is part of the cell’s organizational map for deciding which genes get turned on and which stay silent.
Tuning the Gate Through Chemical Modifications
The cell can also adjust pore permeability without changing its physical diameter. When signaling pathways add phosphate groups to FG-nucleoporins, the chemical change alters how those proteins interact with transport receptors. Research has shown that phosphorylation of FG-Nups by the signaling enzyme ERK does not disrupt the overall architecture of the pore but directly inhibits the binding between FG-Nups and transport receptors, effectively raising the bar for active transport.28PubMed Central. Phosphorylation of nucleoporins: signal transduction-mediated regulation of their interaction with nuclear transport receptors This gives the cell a rapid, reversible way to regulate nuclear transport in response to external signals, stress, or growth cues, without needing to build or destroy any pores.
Not All Cells Have the Same Number of Pores
The density of nuclear pores on a cell’s nuclear envelope is not fixed across cell types. Cells with high metabolic and transcriptional activity tend to have more pores per unit area, while quiescent cells have fewer. This variation is not subtle. A study of human choroidal melanoma cells found highly significant differences in pore density among different cell morphologies within the same tumor: one cell type had the lowest density, another intermediate, and a third the highest.29PubMed. Differences in nuclear pore density among human choroidal melanoma cell types The pattern makes intuitive sense: a cell that is churning out large volumes of mRNA and importing large quantities of transcription factors needs more gates. Pore number per cell can range from a few hundred in a small, quiet lymphocyte to many thousands in a metabolically active oocyte or liver cell.
This variability is relevant beyond basic biology. Because pore density correlates with transcriptional activity, some researchers have explored whether differences in NPC composition or number could serve as markers for disease states, including certain cancers where nuclear transport is known to be disrupted. The idea is still in early stages, but it underscores how much the NPC reflects the broader state of the cell rather than being a static piece of infrastructure.