Exosome-Based CNS Delivery for Gene Therapy

📖 Wiki Page

technology1253 wordssynced 2026-04-02

Overview

Exosomes are small extracellular vesicles (30-150 nm) naturally secreted by most cell types, carrying protein, lipid, and nucleic acid cargo. They represent the body's own intercellular communication system and have evolved to efficiently deliver payloads between cells — including across biological barriers like the [blood-brain barrier](/entities/blood-brain-barrier) (BBB). For [neurodevelopmental epilepsies](/therapeutics/aav-gene-therapy-neurodevelopmental-epilepsy) (NDEs), exosome-based delivery offers a biologics approach that combines favorable safety properties with inherent CNS tropism[@herrmann2022].

Unlike lipid nanoparticles (LNPs), which are synthetically assembled, exosomes are biological particles with native membrane proteins and lipid compositions that can facilitate cell-type-specific targeting. This makes them attractive for delivering gene therapies to specific neuronal populations — particularly the GABAergic interneurons that are the primary therapeutic target in [Dravet syndrome](/diseases/dravet-syndrome) (via [SCN1A](/genes/scn1a)).

Biology of Exosomes

Biogenesis and Composition

Exosomes are generated through the endosomal pathway:

flowchart TD A["Cell membrane invagination"] --> B["Early endosome"] B --> C["Multi-vesicular body (MVB) formation"] C --> D["MVB fusion with cell membrane"] D --> E["Exosome release into extracellular space"] C --> F["Lysosomal degradation"] style A fill:#0a1929,color:#e0e0e0 style E fill:#0e2e10,color:#e0e0e0 style F fill:#3b1114,color:#e0e0e0

...

Overview

Biology of Exosomes

Biogenesis and Composition

Exosomes are generated through the endosomal pathway:

Mermaid diagram (expand to render)

Key exosome components:

Membrane: Lipid bilayer derived from the plasma membrane, enriched in cholesterol, sphingolipids, and ceramides
Surface proteins: Tetraspanins (CD9, CD63, CD81), integrins, MHC molecules — enabling cell-type targeting
Interior cargo: mRNA, miRNA, siRNA, proteins, CRISPR-Cas9 components, lipids

Natural CNS Tropism

Several cell-derived exosomes exhibit natural tropism for the brain:

Mesenchymal stromal cell (MSC) exosomes: Cross the BBB via unknown mechanisms, have shown broad CNS distribution in animal models
Neuron-derived exosomes: Contain neural-specific cargo and surface proteins that may enable re-targeting
Retrovirus-derived VLPs: Engineered viral-like particles combining viral fusogenicity with exosome features

The combination of small size (40-120 nm), flexible lipid composition, and specific surface protein signatures allows exosomes to navigate the neurovascular unit more effectively than larger particles[@matsumoto2020].

Advantages for NDE Gene Therapy

| Advantage | Description | NDE Relevance |
|-----------|-------------|----------------|
| BBB crossing | Natural transcytosis capability, especially MSC-derived | Critical for systemically administered NDE therapies |
| Cell-type specificity | Surface engineering can target specific neuronal subtypes | Enable targeting of GABAergic interneurons for SCN1A |
| Immunogenicity | Low pre-existing immunity, even with repeat dosing | Safe for pediatric NDE applications |
| Cargo versatility | Carries mRNA, ASO, siRNA, CRISPR, proteins | Versatile across NDE therapeutic modalities |
| Biologic safety | No viral genes, non-replicating, biodegradable | Safer than AAV for pediatric CNS delivery |
| Neuronal uptake | Naturally fusogenic with neuronal membranes | Efficient delivery to target neurons |

Engineering Strategies

Surface Modification

Exosome surface proteins can be engineered to enhance targeting:

Ligand Display: Express targeting ligands (antibodies, peptides, aptamers) on the exosome surface using engineered fusion proteins:

CD63-displayed peptides: For enhanced BBB penetration
Lamp2b chimeras: Fusion with targeting moieties
Anti-Transferrin Receptor (TfR) nanobodies: Enable RMT across BBB

Glycosylation targeting: Specific sugar residues on exosome surface proteins can bind brain endothelial lectins, facilitating transcytosis

Brain-targeting peptide display: Rabies virus-derived peptides (e.g., RVG fragment) displayed on exosome surface for neuronal targeting via acetylcholine receptor binding

Cargo Loading

| Method | Description | Efficiency |
|--------|-------------|------------|
| Electroporation | Electric field pulses create pores in exosome membrane | Good for nucleic acids, may damage membranes |
| Sonication | Acoustic cavitation temporarily permeabilizes membrane | Moderate efficiency |
| Extrusion | Force cargo through exosome membrane | Higher efficiency, may damage exosome |
| Lipid fusion | Fuse liposomes loaded with cargo to exosomes | Preserves integrity |
| Endogenous loading | Engineer donor cells to load cargo into exosomes | Most physiological |
| Click chemistry | Covalent attachment of cargo to exosome surface | Enables precise targeting |

Cell-Type Targeting for NDEs

For SCN1A/Dravet therapy, exosomes must preferentially target GABAergic interneurons:

Engineered exosomes displaying GABAergic-specific surface markers (e.g., parvalbumin-targeting peptides)
Exosomes derived from GABAergic neuron cultures
Dual-targeting strategies: BBB-crossing + neuronal subtype targeting

Clinical Applications in NDEs

Exosome-Encapsulated ASOs

For [Angelman syndrome](/diseases/angelman-syndrome) and UBE3A-targeting therapies, exosome delivery of GTX-102 or similar ASOs could improve CNS penetration and reduce off-target peripheral effects[@mohammad2023].

Potential advantages over naked ASO delivery:

10-50x higher brain exposure in animal models
Reduced peripheral organ accumulation (liver, kidney)
Extended duration of CNS residence

Exosome-mRNA for Dravet

Exosome-encapsulated SCN1A mRNA could provide transient expression of functional Nav1.1 channels in inhibitory interneurons, potentially bridging the gap before developmental damage becomes irreversible.

CRISPR-Cas9 Delivery

Exosomes can deliver CRISPR-Cas9 systems (mRNA + gRNA) for gene editing in CNS neurons. This is particularly relevant for:

SCN1A gene correction: Direct fixing of pathogenic variants
UBE3A activation: Epigenetic editing approaches
KCNQ2/CDKL5 correction: Engineering functional gene variants[@strempfl2024]

Manufacturing and Scale-Up

Challenges

Yield: Exosome production from cells is low (10^9-10^12 particles per 10^8 cells), making large-scale manufacturing difficult

Characterization: No standardized methods for exosome purity, potency, or identity

Storage stability: Exosomes are sensitive to freeze-thaw cycles and temperature fluctuations

Lot-to-lot variability: Biological production is inherently variable

Industrial Approaches

| Company/Group | Approach | Status |
|---------------|----------|--------|
| Caperna (Roche) | Engineered exosomes for CNS delivery | Preclinical |
| Evox Therapeutics | Engineered exosome platform | Preclinical |
| Codiak BioSciences | exoSTING and exoIL-12 programs | Clinical (oncology) |
| PureTech | Glyph technology (exosome targeting) | Research |
| Academic consortia | MSC exosome GMP production | GMP-grade available |

GMP Manufacturing

Clinical-grade exosome manufacturing follows a standardized pathway:

Cell banking: Master cell bank of exosome-producing cells (MSCs, engineered cells)

Upstream processing: Scalable bioreactor culture (hollow-fiber, stirred-tank)

Downstream purification: Tangential flow filtration, ion-exchange chromatography, ultrafiltration

Quality control: Particle size (NTA, DLS), protein markers (CD9, CD63, CD81), potency assays, sterility testing

Formulation: Buffer exchange, sterile filtration, vial filling

Current GMP exosome costs are high (~$10,000-50,000 per patient dose) but expected to decrease with improved manufacturing efficiency.

Comparison to Other Delivery Platforms

| Factor | Exosomes | LNP | AAV |
|--------|----------|-----|-----|
| BBB penetration | Moderate-high (engineered) | Low-moderate | Moderate |
| Cell-type specificity | High (surface engineering) | Moderate | Serotype-dependent |
| Immunogenicity | Very low | Low | Moderate-high |
| Manufacturing scale | Challenging | Scalable | Complex |
| Cargo capacity | Large (proteins, mRNA, CRISPR) | Large (mRNA, CRISPR) | Limited (~4.7kb) |
| Duration | Transient | Transient | Long-term |
| Redosing | Fully repeatable | Fully repeatable | Limited |
| Cost | High | Moderate | Very high |
| Clinical stage | Early (mostly preclinical) | Growing (CNS trials) | Mature (multiple approved) |

Key Open Questions

Can engineered exosomes achieve sufficient GABAergic interneuron targeting for meaningful SCN1A expression restoration in Dravet patients?

What is the minimal effective dose of exosome-delivered ASO for Angelman syndrome, and can it be manufactured at commercial scale?

How does exosome delivery compare to LNP delivery for CRISPR-Cas9 in CNS neurons in terms of editing efficiency and safety?

Can repeat dosing with exosomes maintain therapeutic effect without loss of efficacy due to anti-exosome antibodies?

What donor cell type (MSC, neural stem cell, iPSC-derived) produces exosomes with optimal CNS tropism for NDE applications?

Cross-Links

[Lipid Nanoparticle CNS Delivery](/technologies/lipid-nanoparticle-cns-delivery) — competing non-viral platform](/technologies)
[AAV Vectors](/technologies/aav-vectors) — viral delivery counterpart](/technologies)
[Gene Therapy for Neurodevelopmental Epilepsy](/technologies/gene-therapy-neurodevelopmental-epilepsy) — hub page](/technologies)
[Focused Ultrasound Neuromodulation](/technologies/focused-ultrasound-neuromodulation) — BBB opening for enhanced delivery](/technologies)
[SCN1A Gene](/genes/scn1a) — Dravet syndrome target](/genes)
[UBE3A Gene](/genes/ube3a) — Angelman syndrome target](/genes)
[Dravet Syndrome Foundation](/organizations/dravet-syndrome-foundation) — patient advocacy organization](/organizations)
[Angelman Syndrome Foundation](/organizations/angelman-syndrome-foundation) — patient advocacy organization

📖 View canonical wiki page →