Nuclear entry of nonviral vectors

Purpose: Nonviral gene delivery is limited by multiple extracellular and intracellular barriers, but the nuclear envelope remains a major unresolved barrier, especially in nondividing cells. Plasmids must enter the nucleus to be transcribed, yet the mechanisms of plasmid nuclear import are much less understood than protein nuclear import. This review summarizes mechanisms of plasmid nuclear import in nondividing cells and discusses approaches to exploit or circumvent these processes to improve gene transfer.

Hypothesis: No formal experimental hypothesis. Central thesis: the nuclear envelope is a critical rate-limiting barrier for nonviral gene delivery; plasmids can enter nuclei of nondividing cells via the nuclear pore complex in a sequence-specific manner through DNA nuclear targeting sequences (DTSs), which bind transcription factors that bridge the DNA to importin machinery. Exploiting DTSs or NLS-based strategies can increase nuclear import and gene expression.

Aims: Summarize studies elucidating mechanisms of plasmid nuclear import in nondividing cells. - Discuss the nuclear envelope as a barrier to gene transfer. - Describe DNA nuclear targeting sequences (DTSs) and their mechanisms. - Review cell-specific DNA nuclear import. - Discuss protein factor requirements for DNA nuclear import. - Review approaches to increase DNA nuclear import, including NLS peptides, IBB, glucocorticoid receptor, and DTS incorporation. - Comment on in vivo DNA nuclear import and future directions for nonviral gene therapy.

Delivery system: Payloads: plasmid DNA, SV40 genome, pBR322, pUC19, pGL3-basic, GFP/luciferase reporter plasmids, linear DNA fragments, NLS-modified DNA, PNA-clamped DNA. - Vectors/carriers: naked DNA, cationic lipids/lipoplexes, polyplexes including polyethyleneimine (PEI) and polylysine, dendrimers, NLS-containing peptides/proteins, importin-binding domain (IBB), glucocorticoid receptor–dexamethasone conjugates, EBNA-1/OriP, TetR-NLS/tetO, PNA-NLS conjugates. - DTS elements: SV40 72-bp enhancer repeat, NF-κB binding sites, OriP, tet operator, smooth muscle gamma actin (SMGA) promoter, flk-1 promoter. - Routes/methods: cytoplasmic vs nuclear microinjection, transfection, electroporation, in vivo naked DNA injection, electroporation-mediated gene transfer to vasculature. - Target cells/tissues: mouse fibroblasts, TC7, HeLa, COS, primary human airway epithelial cells, rat myotubes, mouse skeletal muscle, rat mesenteric vasculature, Xenopus oocytes, zebrafish/shrimp embryos.

Approach: Review of preclinical literature; no primary experiments. Discusses in vitro microinjection, in situ hybridization, digitonin-permeabilized cells, synchronized cell transfection, and in vivo plasmid delivery in mouse muscle and rat vasculature. Models include dividing and nondividing mammalian cells, primary cells, and intact animal tissues. No clinical trials are discussed.

Key methods: Microinjection into cytoplasm vs nucleus followed by reporter gene expression. - In situ hybridization to detect plasmid DNA localization. - Quantitative PCR, Southern blot, electron microscopy, fluorescence-activated cell sorting for nuclear plasmid quantification. - Fluorescently labeled PNA clamps for DNA tracking. - Digitonin-permeabilized cell nuclear import assays with cytoplasmic/nuclear extracts, importins, RAN. - Synchronized cell transfection to assess cell-cycle dependence. - In vivo electroporation and in situ hybridization in muscle/vasculature.

Key results: Nuclear microinjection of pBR322 plasmids yielded expression in ~50–100% of mouse fibroblasts; cytoplasmic injection yielded no expression in any of 1000 injected cells. - Cytoplasmic injection of 1000–3000 plasmid copies gave <3% of the expression obtained by nuclear injection. - Nondividing primary human airway epithelial cells were only ~10% as likely to express gene product as dividing cells. - Synchronized cells transfected in G2 or G2–M expressed 50- to 3000-fold more gene product than cells transfected in G1. - Following lipoplex transfection, 2000–10,000 plasmids are delivered per cell; only 1–10% reach nuclei by 24–36 h, though one fluorescent-labeling study reported 30–60% nuclear localization, possibly overestimated. - In dividing cells, 30–100 times more plasmid injected into cytoplasm vs nucleus was needed for equivalent expression. - Cytoplasmic plasmid half-life: ~50–90 min in HeLa and COS cells; <2 h in another study; modeled degradation 30–1400 molecules/min, translating to ~5 h half-life. - SV40 DTS import detected within 40–60 min after cytoplasmic microinjection; up to 100% of PNA-labeled DNA detected in nucleus at longer times. - With DTS, 10–1000 copies injected into cytoplasm produced expression as early as 2–4 h postinjection; DTS-lacking plasmid showed no expression prior to cell division. - Even with a DTS, ~20 times more DNA injected into cytoplasm vs nucleus was needed for equivalent expression. - NF-κB binding sites increased gene expression 12-fold; tetO/TetR-NLS increased expression ~20-fold and nuclear localization 4-fold in growth-arrested cells. - SMGA DTS functioned in smooth muscle cells but not endothelial, epithelial, or fibroblast cells; flk-1 DTS functioned in endothelial cells. - In vivo, SV40 DTS increased gene expression ~20-fold in murine skeletal muscle and 40- to 200-fold in intact rat arteries and veins; DTS plasmids localized to nuclei and persisted at 24–48 h. - In mouse myotube in vivo, no expression was observed until ~100,000 DTS-lacking plasmids were injected, suggesting mass action can overcome DTS dependence.

Interpretation: The nuclear envelope is a major barrier for nonviral gene delivery, especially in nondividing cells. Sequence-specific DTSs can mediate plasmid nuclear import via transcription factor–importin bridging, and incorporating DTSs into plasmids can increase gene expression in vitro and in vivo. NLS-peptide approaches have shown variable success. The authors conclude that elucidating basic mechanisms of plasmid nuclear import will lead to more effective nonviral gene therapy approaches.

Limitations: Review article; no primary data. - Most DTS studies are in cultured cells; in vivo success is limited, mostly with SV40 DTS in muscle and vasculature. - NLS-containing peptide/protein approaches have shown inconsistent enhancement of nuclear import and expression. - No way to predict which sequences act as DTSs; identification requires brute-force testing. - DTS dependence is less absolute in vivo; high plasmid concentrations can overcome the barrier via mass action. - Cytoplasmic nucleases degrade DNA, competing with nuclear import. - Mechanisms of DNA nuclear import in intact cells remain incompletely understood. - Differences between cell types and between cultured cells and intact tissues limit translation. - No clinical trials or large-animal validation are discussed.