Lipid nanoparticles (LNPs) deliver therapeutic siRNA and mRNA into cells, but only ~1–2% of the siRNA internalized in the source study escapes from endosomes into the cytosol, and the mechanism of this escape has remained unresolved since Gilleron et al. (2013) directly visualized it. The open problem is to establish, at molecular and biophysical resolution, how ionizable-lipid LNPs transfer their nucleic acid cargo (siRNA in the source work) across the endosomal membrane, in which endosomal compartment and maturation state productive escape occurs, and by what molecular mechanism LNPs alter endosomal maturation (the hybrid early–late compartment with retarded maturation observed in the source work). Later literature deepened the measurements — escape from maturing endosomes or recycling tubules, galectin-detectable endosomal membrane damage, acidification-driven inverse-mesophase transitions, in vivo quantification of ~8% of an optimized mRNA-LNP particle population reaching the cytosol within 30 minutes, and simulation of a stalk–pore fusion pathway — but the compartment of escape, the cause of the low efficiency, and the escape mechanism itself remain without consensus, as the field's own 2024–2026 reviews state. Solving it would convert LNP design from empirical screening into mechanistic prediction and would clarify how escape trades off against endosomal-damage-triggered inflammation.

Background

Small interfering RNAs (siRNAs) are short double-stranded RNAs that silence genes through the RNA interference machinery, which acts in the cytosol; mRNA therapeutics likewise must reach the cytosolic translation machinery. The dominant clinical delivery vehicle for these RNAs is the lipid nanoparticle (LNP), a ~60–100 nm particle typically containing an ionizable lipid (positively charged only at acidic pH, allowing it to bind the negatively charged RNA and to interact with endosomal membranes after acidification), a helper phospholipid, cholesterol, and a PEGylated lipid that shields the particle in the bloodstream. After endocytosis, the particle is confined to membrane-bounded endosomes, and the therapeutic RNA can act only if it crosses the endosomal membrane into the cytosol — the step called endosomal escape. Endosomes mature along a regulated pathway: early endosomes marked by the small GTPase Rab5 and EEA1 (early endosome antigen 1) convert, via the Rab-conversion machinery (the SAND1/Mon1 complex that replaces Rab5 with Rab7), into late endosomes marked by Rab7 and LAMP1 (lysosome-associated membrane protein 1), which fuse with lysosomes where cargo is degraded.

Gilleron et al. (Nat Biotechnol 2013, doi:10.1038/nbt.2612) formulated traceable LNPs (the clinically benchmarked DLin-MC3-DMA ionizable lipid system) carrying siRNAs conjugated to 6-nm colloidal gold and combined quantitative fluorescence microscopy, electron microscopy and mathematical modeling to count escape events in cultured cells and mouse liver. They found that uptake proceeds by clathrin-mediated endocytosis and macropinocytosis with cell-type-dependent kinetics, that LNP exposure recruits the Rab5 effector Rabankyrin-5 and creates a hybrid compartment simultaneously positive for early (EEA1) and late (LAMP1) endosomal markers with delayed delivery to lysosomes, and — the central quantitative result — that only 1.34 ± 0.08% (HeLa cells) and 1.66 ± 0.07% (mouse hepatocytes in vivo) of internalized siRNA reaches the cytosol, during a limited time window when LNPs reside in that specific maturing compartment. The escape events were not accompanied by gross morphological signs of large-scale endosomal rupture, arguing against the classic proton-sponge hypothesis — the idea that buffered proton accumulation in the endosome causes osmotic swelling and bursting. The authors therefore proposed that escape occurs through local, lipid-mediated events such as exchange of ionizable LNP lipids into the endosomal membrane and formation of ion pairs between cationic (protonated) lipid headgroups and the anionic siRNA, whose efficacy depends on the lipid acid dissociation constant (pKapK_a); but the detailed mechanism, the origin of the hybrid compartment and the retarded maturation, and the determinants of cell-type variability (including the kinetics of PEG-lipid deshielding, i.e. loss of the short-tailed PEG-DMG (a polyethylene-glycol lipid with short C14 acyl chains) shield from the particle surface, and recruitment of serum proteins such as apolipoprotein E) remained unresolved, and the paper's own scope sentence calls for a better understanding of the mechanisms of siRNA cellular uptake, intracellular transport and endosomal release. A companion study in the same period showed that the picture is quantitatively dominated by opposing fluxes: Sahay et al. (Nat Biotechnol 2013, doi:10.1038/nbt.2614) found that ~70% of internalized siRNA is exocytosed again through late-endosome/lysosome recycling controlled by NPC1 (Niemann–Pick C1), so that productive cytosolic delivery is a small remainder of a large trafficking flux.

The subsequent decade produced sharp but mutually inconsistent observations. Wittrup et al. (Nat Biotechnol 2015, doi:10.1038/nbt.3298) visualized release of lipid-formulated siRNA from maturing endosomes within ~5–15 minutes of endocytosis and showed that cytosolic galectins — sugar-binding proteins that bind to the exposed inner face of damaged endosomal membranes — immediately recognize the damaged endosome and target it for autophagy. Du Rietz et al. (Nat Commun 2020, doi:10.1038/s41467-020-15300-1) developed the galectin-recruitment assay into a quantitative escape sensor. Paramasivam et al. (J Cell Biol 2021, doi:10.1083/jcb.202110137), using super-resolution microscopy of LNP-delivered mRNA, located escape events in endosomal recycling tubules rather than in the hybrid early–late compartment and showed that total uptake does not predict delivery. Structural work established that LNPs containing clinical ionizable lipids (SM-102, ALC-0315) undergo acidification-driven internal transitions to inverse bicontinuous cubic and hexagonal lyotropic liquid-crystalline mesophases — periodic nanostructures formed by self-assembled lipids in water — whose appearance correlates with transfection efficiency (Yu et al., JACS 2023, doi:10.1021/jacs.3c08729; Acc Chem Res 2025, doi:10.1021/acs.accounts.5c00522), supporting membrane-fusion-like release driven by non-bilayer lipid phase behavior. Omo-Lamai et al. (Nat Nanotechnol 2025, doi:10.1038/s41565-025-01974-5) showed that endosomal escape itself creates endosomal membrane holes: large irreparable holes recruit galectins and trigger inflammation, while a class of ionizable lipids creates smaller holes repaired by the ESCRT (endosomal sorting complex required for transport) membrane-repair pathway, decoupling expression from inflammation. Most recently, coarse-grained molecular dynamics simulations identified a stalk–pore fusion pathway in which transfer of ionizable lipids from the LNP into the endosomal membrane promotes nucleic-acid reorientation, stalk formation and pore opening (Singh et al., Nat Commun 2026, doi:10.1038/s41467-026-76810-y), and an in vivo assay based on immunoisolation of liver lysosomes in LysoTag mice combined with lysosomal barcoding quantified endosomal escape in vivo (~8% of an optimized branched-ionizable-lipid mRNA formulation, counted per administered particle, reaching the cytosol within 30 minutes) and found that loss of Rab7, a mediator of late-endosome maturation, increases escape (Jozić et al., Nat Biotechnol 2026, doi:10.1038/s41587-026-03022-6). Yet authoritative syntheses of this literature — Chatterjee & Peer (PNAS 2024, doi:10.1073/pnas.2307800120), Mrksich et al. (Adv Drug Deliv Rev 2024, doi:10.1016/j.addr.2024.115446), and Pei (ACS Nano 2025, doi:10.1021/acsnano.5c11721) — all conclude that there is still no consensus on the compartment of escape, the cause of the low efficiency, or the escape mechanism, and Pei's own vesicle-budding-and-collapse proposal and the stalk–pore simulation result are contested candidate models rather than established mechanisms.

Problem Statement

Establish the mechanism of lipid-nanoparticle-mediated endosomal escape of nucleic acids at molecular and biophysical resolution, as posed by the direct-visualization findings of Gilleron et al. (2013) for siRNA-loaded ionizable-lipid LNPs (the DLin-MC3-DMA system) in the systems they studied — cultured cells including HeLa cells, and mouse liver in vivo: determine (i) the biophysical process by which the siRNA cargo crosses the endosomal membrane — e.g. ion-pair formation and lipid exchange between protonated ionizable LNP lipids and the endosomal membrane, fusion-pore formation, vesicle budding-and-collapse, or another route — and which of these routes dominates, or whether several coexist, under which conditions, given that the source ultrastructural data showed no gross morphological signs of large-scale endosomal rupture (arguing against proton-sponge-type swelling and bursting, but not excluding the galectin-detectable nanoscale membrane damage that later work shows accompanies productive release) yet only ~1–2% of internalized siRNA succeeds; (ii) the identity and maturation state of the endosomal compartment in which productive escape occurs (the hybrid early–late compartment observed in the source work, the maturing endosomes reported by Wittrup et al. 2015, the recycling tubules reported by Paramasivam et al. 2021, or otherwise), and the molecular mechanism by which LNP lipids generate that compartment and retard its maturation, including which step of the Rab-conversion machinery (e.g. SAND1/Mon1) or endosomal lipid composition is perturbed; and (iii) from (i) and (ii), a mechanistic account sufficient to explain the low escape efficiency and the established dependencies of escape on ionizable-lipid structure and pKapK_a, formulation composition, endosomal maturation state, and cell type, so as to inform rational rather than purely empirical design of the next generation of delivery systems. Such an account should be consistent with — and, for the systems it claims to cover, should explain — the accumulated and partly contradictory later observations (galectin-detectable membrane holes, acidification-driven mesophase transitions, in vivo escape of ~8% of an optimized mRNA-LNP particle population, simulation-derived stalk–pore fusion pathways), while recognizing that these studies use unmatched cargoes, formulations, cell types, denominators and timing, so that unifying all of them across systems is not required of a single answer. Observations made for mRNA-loaded LNPs (for example the recycling-tubule escape reported by Paramasivam et al. 2021) are related evidence whose transferability to the siRNA case is itself unsettled — Chatterjee & Peer (2024) note that knowledge gained on siRNA delivery does not necessarily translate to mRNA payload release — so an answer need not establish the mRNA route to complete this problem. This statement also scopes out, as an adjacent uptake-side question, the determinants of cell-type-variable uptake kinetics (PEG-lipid deshielding rates, recruitment of serum components such as apolipoprotein E); because the source work's own scope sentence covers siRNA cellular uptake, intracellular transport and endosomal release, this exclusion is a scoping decision of the present statement rather than a claim about how the source separates its open items.

References

  1. Gilleron J, Querbes W, Zeigerer A, Borodovsky A, Marsico G, Schubert U, Manygoats K, Seifert S, Andree C, Stöter M, Epstein-Barash H, Zhang L, Koteliansky V, Fitzgerald K, Fava E, Bickle M, Kalaidzidis Y, Akinc A, Maier M, Zerial M. Image-based analysis of lipid nanoparticle–mediated siRNA delivery, intracellular trafficking and endosomal escape. Nature Biotechnology 31(7):638–646 (2013). doi:10.1038/nbt.2612, https://doi.org/10.1038/nbt.2612
  2. Sahay G, Querbes W, Alabi C, Eltoukhy A, Sarkar S, Zurenko C, Karagiannis E, Love K, Chen D, Zoncu R, Buganim Y, Schroeder A, Langer R, Anderson DG. Efficiency of siRNA delivery by lipid nanoparticles is limited by endocytic recycling. Nature Biotechnology 31(7):653–658 (2013). doi:10.1038/nbt.2614, https://doi.org/10.1038/nbt.2614
  3. Wittrup A, Ai A, Liu X, Hamar P, Trifonova R, Charisse K, Manoharan M, Kirchhausen T, Lieberman J. Visualizing lipid-formulated siRNA release from endosomes and target gene knockdown. Nature Biotechnology 33(8):870–876 (2015). doi:10.1038/nbt.3298, https://doi.org/10.1038/nbt.3298
  4. Paramasivam P, Franke C, Stöter M, Höijer A, Bartesaghi S, Sabirsh A, Lindfors L, Yanez Arteta M, Dahlén A, Bak A, Andersson S, Kalaidzidis Y, Bickle M, Zerial M. Endosomal escape of delivered mRNA from endosomal recycling tubules visualized at the nanoscale. Journal of Cell Biology 221(2):e202110137 (2021). doi:10.1083/jcb.202110137, https://doi.org/10.1083/jcb.202110137
  5. Du Rietz H, Hedlund H, Wilhelmson S, Nordenfelt P, Wittrup A. Imaging small molecule-induced endosomal escape of siRNA. Nature Communications 11:1809 (2020). doi:10.1038/s41467-020-15300-1, https://doi.org/10.1038/s41467-020-15300-1
  6. Majzoub RN, Chan C-L, Ewert KK, Silva BPF, Liang K-S, Safinya CR. Fluorescence microscopy colocalization of lipid–nucleic acid nanoparticles with wildtype and mutant Rab5-GFP: a platform for investigating early endosomal events. Biochimica et Biophysica Acta (BBA) – Biomembranes 1848(6):1308–1318 (2015). doi:10.1016/j.bbamem.2015.03.001, https://doi.org/10.1016/j.bbamem.2015.03.001
  7. Stewart MP, Langer R, Jensen KF. Intracellular Delivery by Membrane Disruption: Mechanisms, Strategies, and Concepts. Chemical Reviews 118(16):7409–7531 (2018). doi:10.1021/acs.chemrev.7b00678, https://doi.org/10.1021/acs.chemrev.7b00678
  8. Patel S, Kim J, Herrera M, Mukherjee A, Kabanov AV, Sahay G. Brief update on endocytosis of nanomedicines. Advanced Drug Delivery Reviews 144:90–111 (2019). doi:10.1016/j.addr.2019.08.004, https://doi.org/10.1016/j.addr.2019.08.004
  9. Paunovska K, Loughrey D, Dahlman JE. Drug delivery systems for RNA therapeutics. Nature Reviews Genetics 23(5):265–280 (2022). doi:10.1038/s41576-021-00439-4, https://doi.org/10.1038/s41576-021-00439-4
  10. Yu H, Iscaro J, Dyett B, Zhang Y, Seibt S, Martinez N, White J, Drummond CJ, Bozinovski S, Zhai J. Inverse Cubic and Hexagonal Mesophase Evolution within Ionizable Lipid Nanoparticles Correlates with mRNA Transfection in Macrophages. Journal of the American Chemical Society (2023). doi:10.1021/jacs.3c08729, https://doi.org/10.1021/jacs.3c08729
  11. Yu H, Dyett BP, Drummond CJ, Zhai J. Ionizable Lipid Nanoparticles for mRNA Delivery: Internal Self-Assembled Inverse Mesophase Structure and Endosomal Escape. Accounts of Chemical Research 58(20):3210–3222 (2025). doi:10.1021/acs.accounts.5c00522, https://doi.org/10.1021/acs.accounts.5c00522
  12. Chatterjee S, Kon E, Sharma P, Peer D. Endosomal escape: A bottleneck for LNP-mediated therapeutics. Proceedings of the National Academy of Sciences USA 121(11):e2307800120 (2024). doi:10.1073/pnas.2307800120, https://doi.org/10.1073/pnas.2307800120
  13. Mrksich K, Padilla MS, Mitchell MJ. Breaking the final barrier: Evolution of cationic and ionizable lipid structure in lipid nanoparticles to escape the endosome. Advanced Drug Delivery Reviews 214:115446 (2024). doi:10.1016/j.addr.2024.115446, https://doi.org/10.1016/j.addr.2024.115446
  14. Pei D. Endosomal Escape of Lipid Nanoparticles: A Perspective on the Literature Data. ACS Nano 19(47):40293–40303 (2025). doi:10.1021/acsnano.5c11721, https://doi.org/10.1021/acsnano.5c11721
  15. Omo-Lamai S, Wang Y, Patel MN, Milosavljevic A, Zuschlag D, Poddar S, Wu J, Wang L, Dong F, Espy C, Majumder A, Essien E-O, Shen M, Channer B, Papp TE, Tobin M, Maheshwari R, Jeong S, Patel S, Shah A, Murali S, Chase LS, Zamora ME, Arral ML, Marcos-Contreras OA, Myerson JW, Hunter CA, Discher D, Gaskill PJ, Tsourkas A, Muzykantov VR, Brodsky I, Shin S, Whitehead KA, Parhiz H, Katzen J, Miner JJ, Trauner D, Brenner JS. Limiting endosomal damage sensing reduces inflammation triggered by lipid nanoparticle endosomal escape. Nature Nanotechnology 20(9):1285–1297 (2025). doi:10.1038/s41565-025-01974-5, https://doi.org/10.1038/s41565-025-01974-5
  16. Shin J, Douglas CJ, Zhang S, Seath CP, Bao H. Targeting Recycling Endosomes to Potentiate mRNA Lipid Nanoparticles. Nano Letters 24(17):5104–5109 (2024). doi:10.1021/acs.nanolett.3c04415, https://doi.org/10.1021/acs.nanolett.3c04415
  17. Singh AP, Shibata K, Miyazaki Y, Shinoda W. pH-Sensitive Amino Lipid-Driven Pore Formation Enables Endosomal Escape of Lipid Nanoparticles. Nature Communications (2026). doi:10.1038/s41467-026-76810-y, https://doi.org/10.1038/s41467-026-76810-y
  18. Jozić A, Le Roux C, Kim J, Berchel M, Sahel DK, Bodi EK, Palumbo M, Vasudevan A, Murthy NTV, Eygeris Y, Gautam M, Bloom E, Barnes AP, Jaffrès P-A, Sahay G. In vivo endosomal escape assay identifies mechanisms for efficient hepatic LNP delivery. Nature Biotechnology (2026). doi:10.1038/s41587-026-03022-6, https://doi.org/10.1038/s41587-026-03022-6
  19. Bourguignon T, Sharma P, Hazan-Halevy I, Peer D. Lipid nanoparticles and endosomal escape: Characterizing the escape compartment and standardizing models. Materials Today 98:103436 (2026). doi:10.1016/j.mattod.2026.103436, https://doi.org/10.1016/j.mattod.2026.103436