The nuclear envelope is punctured by nuclear pore complexes (NPCs) whose central channel is filled with a disordered mesh of FG-nucleoporins. Two transport regimes decide whether a gene-therapy nanoparticle ever reaches the genome:
Passive diffusion: works only for d < d_pass ≈ 5 nm (~40 kDa)
Facilitated (NLS): works for d up to d_max ≈ 39 nm, needs importin-β
Physical cutoff: d > d_max never crosses, any NLS count
For cargo in the facilitated window, translocation probability per pore encounter follows an avidity model — more nuclear-localization-signal (NLS) copies bind more importin-β molecules, increasing the odds of engaging the FG-mesh:
P_bind = 1 − exp(−k · N_NLS), k ≈ 0.55
P_ran = RanGTP_nuclear / (RanGTP_nuclear + K_m) (Hill-type gate)
P_cross = P_bind · P_ran (per pore-contact attempt)
- Cargo diameter — below ~5 nm, cargo diffuses through the pore freely on contact, exactly like a small protein or oligonucleotide. Above ~39 nm the channel is geometrically too narrow regardless of ligands — this is why bare plasmid DNA or bulky LNP-CRISPR ribonucleoprotein complexes stall outside the nucleus in non-dividing cells.
- NLS copies — each copy is a binding site for importin-β; avidity (not single-site affinity) drives whether the complex commits to translocation, which is why gene-therapy vectors are often engineered with multiple NLS peptides.
- RanGTP gradient — a nuclear-high / cytoplasmic-low RanGTP gradient (maintained by RCC1 and RanGAP) is what makes transport directional and irreversible: RanGTP inside the nucleus dissociates the importin–cargo complex, releasing cargo into the nucleoplasm. Collapse this gradient and even NLS-tagged cargo stalls mid-pore.
- Active NPCs — the number of pores embedded in the envelope; more pores mean more simultaneous contact opportunities, raising the population-level delivery rate without changing any single particle's odds.
Real-world relevance: this bottleneck is exactly why non-viral gene-therapy payloads (LNP-mRNA, CRISPR ribonucleoproteins, naked plasmids) reach the nucleus so inefficiently in post-mitotic cells such as neurons, cardiomyocytes and hepatocytes, where the envelope never breaks down during mitosis to give cargo a free pass — unlike dividing cells, where nuclear envelope breakdown briefly bypasses the NPC bottleneck entirely.