Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 16.
Published in final edited form as: Nat Rev Bioeng. 2025 Mar 4;3(5):360–373. doi: 10.1038/s44222-025-00277-7

Extracellular vesicles for the delivery of gene therapy

Emilio Di Ianni 1, Wataru Obuchi 1,2, Koen Breyne 1,✉, Xandra O Breakefield 1,✉
PMCID: PMC12526208  NIHMSID: NIHMS2111508  PMID: 41104040

Abstract

Gene therapy has brought hope for the treatment of previously incurable diseases, such as genetic disorders, cancers and autoimmune diseases. However, gene therapy requires efficient delivery with cell and tissue specificity, which remains challenging owing to the limited targeting and cargo-loading capacity of viral delivery vehicles, as well as immunogenicity and toxicity concerns. Extracellular vesicles can be designed as non-viral carriers for gene therapy owing to their ability to deliver multiple cargo types, including transgenes, small encoding or non-coding RNA, DNA and functional proteins. Importantly, extracellular vesicles are immunologically neutral and can cross biological barriers. In this Review, we discuss the application of extracellular vesicles in gene therapy. We outline how the inherent content of extracellular vesicles can facilitate different gene-therapy approaches and examine the design of extracellular vesicles for the loading of gene-therapy tools, targeted delivery and cargo release. Finally, we survey clinical applications of extracellular vesicles and highlight important engineering and translational challenges.

Introduction

Extracellular vesicles, including exosomes, are membrane-enclosed, nanometre-sized particles released from all prokaryotic and eukaryotic cells. These vesicles can mediate cell-to-cell communication by conveying proteins, metabolites, glycomoieties, lipids, RNA and DNA through the extracellular space1. Extracellular vesicle populations reflect the phenotypic and genotypic state of the source cells, are omnipresent in biofluids, tissues and conditioned medium from cultured cells and are heterogeneous with respect to size, morphology, composition and function. They typically carry complete and incomplete macromolecules and may thus serve as a mechanism for cells to dispose of excess and unwanted components, which may also explain why extracellular vesicles are rapidly eliminated from the body within six hours2. The ability of extracellular vesicles to protect enclosed nucleic acids and other biological macromolecules from biodegradation, and to deliver them to recipient cells with minimal immune activation, makes them useful tools for delivering multiplex biomolecular information to cells3. However, their natural cargo might reduce their loading capacity, compared to liposomes4,5. In addition to the intraluminal macromolecules of extracellular vesicles, components such as glycoproteins and receptor ligands are displayed on their surfaces and influence the cells with which they come into contact6. Thereby, extracellular vesicles facilitate cellular communication across various body sites and can cross biological barriers, such as the blood–brain barrier7,8 and the placenta8. Extracellular vesicles can further be engineered to augment their native functions, for example, to address their limited native tissue and cell homing machinery7. Importantly, their targeting8 and cargo capacity can be modified to transform them into gene-therapy vector platforms9.

Gene therapies can be delivered by various vectors, including adeno-associated viruses (AAVs) and synthetic lipid nanoparticles (LNPs). The vector selection depends on the cargo type, target and vector limitations. The cargo capacity of AAVs is limited to single-stranded and double-stranded DNA (ssDNA and dsDNA, respectively), with a cargo limitation of 4.7 kb10. Furthermore, AAVs have high immunogenicity, limiting the possibility of repeated administrations11. Although AAV vectors are typically non-integrating, the risk of AAV genome insertion remains, especially if they are loaded with gene-editing tools. LNPs have been approved for clinical use, and they can be produced from synthetic materials at large scale and in homogeneous formulations. Their cargo can include large nucleic sequences, and they can be loaded with high efficiency. However, LNPs have little capacity to cross biological barriers and their formulation of a few lipid components limits engineering capabilities. In addition, LNPs have limited protein-packaging capacity and suffer from immunogenicity12.

In this Review, we discuss the engineering of extracellular vesicles for gene addition (knock-in, mRNA delivery and protein delivery), gene regulation (knockout, microRNAs (miRNAs), small-interfering RNAs (siRNAs), RNA interference (RNAi), CRISPR interference (CRISPRi), RNA editing and long non-coding RNAs (lncRNAs)) and gene editing (CRISPR–Cas9 and prime editing)12 (Fig. 1a). In particular, we outline how extracellular vesicles can be designed (Fig. 1b) to overcome the gene-therapy challenges associated with viral and other non-viral carriers6, including: the delivery of small molecules or drugs, protecting labile cargo (such as RNA) from degradation, biocompatibility, toxicity and immunogenicity, the crossing of biological barriers, clearance from blood, large-scale production, cell-specific targeting, cargo loading and cargo release13 (Fig. 1c). In addition, we highlight ongoing clinical trials and describe the remaining challenges in translating extracellular vesicles to the clinic for gene therapy.

Fig. 1 |. Extracellular vesicles as tools to deliver gene therapy.

Fig. 1 |

a, Different gene-therapy modalities. b, The trans-membrane proteins of extracellular vesicles can be fused with a tag to carry a therapeutic cargo, or the vesicles can be modified with reactive chemicals linked to a cargo. c, Extracellular vesicles can be loaded with cargo through endogenous (passive loading by cell) or exogenous (active loading) approaches, and they can also carry and release multiple cargos. Multiplex cargos can be added by co-transfecting cells with multiple vectors for different cargos linked to vesicle proteins, or by exogenously loading multiple cargos. The surface of extracellular vesicles can be further modified with anti-phagocytic signals to prevent recognition by immune cells and bypass liver cells to extend their circulation time. Surface modification with angiopep-2 or peptides allows crossing of biological barriers, such as the blood–brain barrier (BBB). Extracellular vesicles can also be engineered to contain targeting moieties to enable delivery of gene therapies to specific cells and tissues. ASO, antisense oligonucleotide; CD47, cluster of differentiation 47; gRNA, guide RNA; lncRNA, long non-coding RNA; siRNA, small interfering RNA; PEG, polyethylene glycol; RGD peptide, Arg–Gly–Asp peptide; RVG peptide, rabies virus glycoprotein peptide; TAT, trans-activator of transcription.

Extracellular vesicle cargo for gene therapy

Extracellular vesicles can contain different kinds of cargo, in particular, coding and non-coding RNAs, DNA and proteins, which can be exploited to alter the biological function of recipient cells (Table 1). Therefore, extracellular vesicles can be leveraged as non-viral vectors for gene therapy.

Table 1 |.

Cargo types and loading strategies for extracellular-vesicle-mediated gene therapy

Cargo Loading Target disease Ref.
Principle Strategy
Non-coding RNA miRNA Endogenous Passive loading by cell transfection Huntington’s disease (miR-124) 19
PD-L1 siRNA Exogenous Active loading by clicking peptide on vesicle surface (c(RGDyK)), and by mixing vesicles with cholesterol-conjugated siRNA GBM 145
sgRNA Exogenous Active loading through electroporation of extracellular vesicles Breast cancer 40
Coding RNA mRNA Endogenous Passive loading by cell transfection GBM PTEN 81
mRNA Endogenous Passive loading by cell transfection GBM yCD::UPRT 32
mRNA Endogenous Passive loading by cell transfection Breast cancer HChrR6 31
mRNA Endogenous Passive loading by cell transfection Atherosclerosis IL-10 cytokine 34
DNA pDNA encoding Cas9 sgRNA Exogenous Active loading by extracellular vesicle transfection Pancreatic cancer (KRAS-G12D) 74
Protein or multiplex CRISPR–CasRx (Cas13d) protein and complex of 3 gRNAs Endogenous Passive loading by cell transfection Acute lung injury (IL-6, TNF, IL-1β) 80

GBM, glioblastoma.

Non-coding RNA (20–25 nucleotides)

Small non-coding RNAs (sncRNAs) or fragments of coding RNAs and tRNAs are often found in extracellular vesicles14,15. In addition, extracellular vesicles can be enriched with therapeutic small-RNA species, such as siRNA and miRNA, to knock down targeted transcripts. For example, extracellular-vesicle-mediated delivery of siRNA can reduce BACE1 transcript and protein levels in neurons, microglia and oligodendrocytes after intravenous injection of these vesicles into wild-type mice, which is relevant to Alzheimer’s disease16. Similarly, siRNA delivered by extracellular vesicles into the mouse striatum enables knockdown of the huntingtin transcript (Htt) in cortical neurons17. In addition, intravenous injection of extracellular vesicles delivering miR-124, which has protective brain activity by regulating neurite outgrowth during neuronal differentiation, can alleviate dysfunction in cortical neural progenitors after ischaemic injury in a photothrombosis mouse model17,18. The same outcome can be achieved through striatal injection of the extracellular vesicles in a R6/2 transgenic mouse model of Huntington’s disease19. However, the precise and efficient delivery of sncRNAs to specific cells and tissues remains challenging in vivo. Nevertheless, extracellular vesicles can improve the stability and bioavailability of sncRNA20, and might allow it to cross biological barriers – a bottleneck in gene therapy21,22.

In addition to siRNAs and miRNAs, extracellular vesicles can also deliver non-coding RNA for therapeutic purposes. For example, lncRNAs are linear RNAs with a length of >200 nucleotides23 that can regulate transcription on several levels. Epistasis (that is, the interaction between genes) by lncRNAs is caused by an RNA–protein complex consisting of RNA polymerase II or transcription factors and lncRNA binding to DNA and regulating gene expression24. Extracellular vesicles can be a natural carrier of lncRNAs in various diseases, including cancer25,26 and Alzheimer disease27, and have therefore been investigated as carriers of therapeutic lncRNAs. For example, extracellular vesicles can be loaded with H19 lncRNAs through overexpression in Wharton–Jelly mesenchymal stromal or stem cells to deliver them to cardiomyocytes for heart-cell regeneration after myocardial infarction28,29. Similarly, circular RNA is involved in regulating gene expression and signal transduction in many biological processes and is thus explored in therapeutic extracellular vesicle applications23.

mRNA (up to about 3 kb)

Extracellular vesicles transfer RNAs between cells, a trait probably inherited from the horizontal transfer of retrotransposons using an RNA intermediate1,30. Importantly, extracellular vesicles can protect the stability of therapeutic mRNA and cross biological barriers. This characteristic can be harnessed to deliver mRNA for gene addition; for example, mRNA encoding the Escherichia coli enzyme HChrR6, which converts the prodrug 6-chloro-9-nitro-5-oxo-5H-benzo-(a)-phenoxazine (CNOB) into the cytotoxic drug 9-p amino-6-chloro-5H-benzo[a]phenoxazine-5-one (MCHB), can be delivered by extracellular vesicles to kill breast cancer cells in orthotopic BT474 xenografts in vivo31. Similarly, mRNA encoding yeast cytosine deaminase::uracil phosphoribosyl transferase (yCD::UPRT) can be loaded into extracellular vesicles that can be delivered into rat models of glioblastoma by repeated injection or ALZET osmotic pump infusion. In the tumour, the enzyme then converts the prodrug 5-fluorocytosine (5-FC) into the cytotoxic drug 5-uracil (5-FU)32.

Extracellular vesicles also benefit from lower immunogenicity compared with other delivery platforms such as LNPs33, and they prevent the interaction of mRNA with receptors of the innate immune system, thereby enabling the translation of therapeutic mRNA, including mRNA encoding potent cytokines. For example, Il10 mRNA can be endogenously loaded into extracellular vesicles by transfection of human embryonic kidney 293 (HEK293T) cells for the treatment of atherosclerosis in an Apoe−/− mouse model34; here, after system administration, the Il10 mRNA-loaded vesicles accumulate in inflammatory macrophages, resulting in anti-inflammatory effects. This approach benefits from low immunogenicity, reflected in low expression of Il1b, Tnf and Il6 (ref. 34).

Therefore, extracellular vesicle-mediated delivery of therapeutic RNAs holds promise for the treatment of various diseases, including cancer, cardiovascular disorders and neurological conditions, but will require further optimization. In particular, extracellular-vesicle isolation and refining techniques are required to selectively isolate therapeutic extracellular vesicles in biofluids. This will improve the purity and consistency of extracellular-vesicle preparations, ensuring reproducibility and safety in therapeutic applications. Moreover, extracellular-vesicle engineering needs to be fine-tuned to improve the specificity of their delivery to target cells and to reduce off-target effects. This involves modifying extracellular-vesicle surface properties or engineering targeting moieties to improve precision in therapeutic delivery. Moreover, the loading capacity of extracellular vesicles needs to be increased, while ensuring cargo is localized in specific extracellular vesicle compartments, such as the lumen, membrane or surface, to ensure functional delivery and to avoid the inclusion of fragmented or non-functional cargos. These efforts aim to maximize the efficacy of extracellular-vesicle-based therapies while maintaining their structural integrity.

DNA

DNA delivery by extracellular vesicles for gene addition, inhibition or editing offers the advantage of greater stability than RNA. DNA expression is regulated by cell type-specific or cell status-dependent promoters, enabling the production of multiple RNAs and proteins. However, gene regulation through plasmid DNA (pDNA) transfer by extracellular vesicles might also stem from secondary transfer of pDNA-derived RNA or protein owing to endogenous loading of extracellular vesicles in transfected cells35. Extracellular vesicles can also be combined with liposomes to deliver CRISPRi-encoding pDNA for the knock down of genes, such as Runx2, which encodes a transcription factor that has a crucial part in skeletal development36. A similar approach, using extracellular vesicles to deliver CRISPR–Cas and pDNA encoding guide RNA (gRNA), allows the editing of CTNNB1 in vitro, which encodesthe important proto-oncogen β-catenin36.

Proteins and multiplex biomolecules

In addition to individual types of biomolecule, a diverse array of biomolecular components can be implemented in extracellular vesicles. For example, extracellular vesicles can be engineered to contain miR34a and the prodrug 5FC-converting cytosine deaminase protein to induce apoptosis in a glioblastoma cell line37. Moreover, extracellular vesicles can deliver protein-encoding mRNA as well as non-coding RNA38,39 for gene editing; for example, Cas9 mRNA and non-coding single guide RNA (sgRNA) can be co-loaded into extracellular vesicles to knock out the oncogene miR-125 in tumour cells as a potential treatment for acute myeloid leukemia and chemoresistant breast tumours40.

Extracellular vesicles can thus serve as non-viral vectors for the delivery of various cargos for gene therapy, including for gene replacement, gene editing, gene addition, immunotherapy and gene silencing, and to modify cells ex vivo and in vivo. Importantly, extracellular vesicles can facilitate the translation of RNA-based therapies, gene-editing tools, DNA-based therapies, protein therapeutics and immunomodulatory therapies.

Loading gene-therapy tools into extracellular vesicles

Extracellular vesicles are natural devices for the delivery of RNA, including coding and non-coding RNA1. They can carry and protect RNAs between 10 and 1,000 nucleotides41 by encapsulating them in a lipid bilayer with associated intraluminal RNA-binding proteins to protect them from degradation by RNases42. Therefore, extracellular vesicles have been particularly explored for the delivery of RNA. However, owing to their own RNA and protein cargo, additional loading capacity might be limited. Loading capacity might be increased by modifying their lipid membrane43. In addition, stabilized siRNAs can be loaded on both the inside5,44 and outside of extracellular vesicles45 for delivery.

Therapeutic cargo can be loaded into extracellular vesicles using recombinant and synthetic biology approaches to engineer source cells or to modify isolated extracellular vesicles7,46,47 (Fig. 2). In particular, extracellular vesicles can be loaded either endogenously, by genetically engineering the donor cell to favour the inclusion of specific cargo in extracellular vesicles, or exogenously, in which case cargo is loaded into isolated extracellular vesicles in vitro. For endogenous loading, cells are typically engineered by transfection or transduction to produce extracellular vesicles that contain specific molecules by fusing an extracellular-vesicle-targeting motif to the biomolecule of interest (active loading) or through overexpression of a biomolecule in the cytoplasm lacking an extracellular vesicle-targeting motif (passive). Exogenous modification can be achieved by transfection, sonication and electroporation. In addition, targeting moieties can be anchored through hydrophobic modifications using amphiphilic conjugates in the extracellular vesicle membrane. Furthermore, isolated extracellular vesicles can be combined with synthetic LNPs or modified to incorporate fusogenic components on their surface to enhance cargo deployment in recipient cells.

Fig. 2 |. Loading and delivery of gene-editing cargo.

Fig. 2 |

Extracellular vesicles can be loaded with cargo and their surface modified with targeting moieties and/or fusogenic components. These modifications can be exogenous or endogenous. Exogenous modification includes loading of cargo by sonication, electroporation or transfection reagents. In addition, amphiphilic conjugates can be anchored to the membrane of vesicles by hydrophobic insertion of lipid tails to display the therapeutic cargo and targeting moiety. Once isolated, extracellular vesicles can be combined with synthetic lipid nanoparticles to form hybrid delivery platforms. Endogenous modification can be achieved by engineering cells through transfection or transduction to actively or passively load specific cargo into extracellular vesicles. Extracellular vesicles can further incorporate fusogenic components on their surface to promote cargo deployment in recipient cells. RGD peptide, Arg–Gly–Asp peptide; RVG peptide, rabies virus glycoprotein peptide; TAT, trans-activator of transcription; VSV-G, vesicular stomatitis virus glycoprotein.

Endogenous loading

Transfection.

Cargo can be added to extracellular vesicles through transfection of source cells. For example, antisense oligonucleotides can be loaded into red blood cell-derived extracellular vesicles using a commercial transfection reagent, such as Lipofectamine48,49. Similarly, a phosphorothioated cytosine–phosphate–guanine (CpG) motif can be conjugated to antisense oligonucleotides, which are then rapidly internalized by neural stem cells through scavenger receptor-mediated endocytosis, without requiring transfection reagents, followed by their endosomal escape and subsequent passive loading into extracellular vesicles. Through a combination of scavenger receptor-mediated endocytosis and CpG-conjugated antisense oligonucleotides that target the STAT3 transcription factor, these extracellular vesicles show anti-tumour effects against glioma in mice50. This strategy might also be applied to other cells that can internalize CpG conjugates.

Active endogenous loading can also be achieved by employing extracellular vesicle proteins that are fused to RNA-binding domains, enabling loading of mRNA into extracellular vesicles after secretion in culture46,51. For example, the extracellular vesicle membrane protein tetraspanin CD63 can be fused to an archaebacterial-derived L7Ae peptide that interacts with RNAs encoding a C/D box52 – non-coding RNAs that guide methylation of ribosomal RNA for ribosome function52 – thereby allowing the recruitment and incorporation of mRNAs containing C/D boxes into budding extracellular vesicles. Similarly, the Gag homologue PEG10, encoded in mammalian genomes, can be used to load mRNA into extracellular vesicles53, resulting in virus-like particles. We note that extracellular vesicles containing viral-derived proteins, including Gag and VSV-G, are classified as virus-like particles rather than extracellular vesicles. Here, PEG10 is retrieved from long terminal repeat (LTR) retrotransposon sequences and its product secretes its own RNA into virus-like particles. This packaging mechanism can be exploited to load mRNA into extracellular vesicles through binding of the untranslated regions of PEG10 to the PEG10 protein. In addition, a sequence can be incorporated into the plasmid that acts as a kind of ‘postal’ (or zip) code: a signal on the RNA that binds to other proteins packaged into extracellular vesicles in cells. For example, extracellular vesicles can be enriched with mRNA encoding the enzyme HChrR6, which converts the prodrug CNOB into the cytotoxic drug MCHB31. Complemented by the exogenous addition of anti-HER2 fused to the lactadherin leader sequence, which binds to phosphatidylserine on the extracellular vesicle surface, these vesicles can be used for delivery of therapeutic mRNA against HER2-expressing breast cancer31.

RNA cargo can also be enriched in extracellular vesicles by the fusion of tetraspanin CD9, a membrane protein often found in extracellular vesicles, typically used as surface marker, to an RNA-binding protein that has affinity to the RNA of interest. For example, CD9 can be fused to the RNA-binding protein HuR, which interacts with miR-155 and regulates the expression of genes involved in cell proliferation, survival and immune signalling, thus allowing the delivery of this miRNA into cells in vitro and in vivo54.

To generate therapeutic glioblastoma-directed extracellular vesicles55, anti-EGFR monoclonal antibodies that target glioblastoma cells can be conjugated to the surface of extracellular vesicles that have been generated in HEK293F culture using DSPE-PEG-NHS (1,2-distearoyl-sn-glycero-3-phosphoethanolamine-poly(ethylene glycol)-N-hydroxysuccinimide) chemistry. In addition, the antiproliferative and pro-apoptotic drug verrucarin A can be spontaneously incorporated into these extracellular vesicles by co-incubation overnight. These drug-loaded antibody-tagged extracellular vesicles can penetrate the blood–brain barrier in a mouse model, targeting intracranial glioblastoma xenografts and delivering the drug intracellularly. However, it is yet unclear whether the addition of synthetic lipids, such as DSPE, to extracellular vesicles compromises their stability and whether the lipid dissociates.

We note that high doses of nucleic acid and extracellular vesicles are typically required to achieve therapeutic outcomes, which can lead to toxicity. However, dosing might be reduced by optimizing the loading efficiency and nucleic acid packaging55. For example, integrating siRNA sequences into a Dicer-independent RNA stem–loop, based on pre-miR-451 miRNA, which is enriched in small extracellular vesicles secreted by many cell types, reduces the expression of the genes targeted by the siRNA in the liver, intestine and kidney of mice. Therefore, small extracellular vesicles that efficiently package siRNA can substantially reduce the required therapeutic dose of the vesicles56.

Nanoporation.

In addition to small RNAs, mainly used for knockdown-based gene therapy, large mRNAs (up to 3 kb) can be loaded into extracellular vesicles57 by endogenous and exogenous approaches. However, adding larger mRNAs to extracellular vesicles is technically challenging and typically results in low concentrations of mRNA per vesicle41. Following pDNA transfection of cells that produce extracellular vesicles, cellular nanoporation using a focal and transient electrical stimulus promotes the release of extracellular vesicles carrying transcribed mRNA58. Compared to exogenous loading by bulk electroporation, cellular nanoporation results in the release of up to 50-fold more mRNA-loaded extracellular vesicles per cell, even from cells with low basal levels of extracellular vesicle secretion. Importantly, cells that take up exogenous synthetic mRNAs can endogenously load these into extracellular vesicles. Incorporation of synthetic mRNA by extracellular vesicles also occurs after uptake of mRNA-containing LNPs33. These mRNA-carrying extracellular vesicles contain recycled ionizable lipids and mRNA from endocytosed LNPs. After systemic injection in mice, their cellular uptake results in the production of recombinant protein in the heart, lung, liver and spleen.

Endogenous engineering of extracellular vesicles thus holds promise for targeted and efficient delivery of gene therapies. However, it remains limited by low cargo capacity59, low loading efficiency60, the technical complexity of obtaining uniform cargo loading61, and regulatory considerations with regard to product quality owing to the heterogeneous cargos derived from source cells61, as well as the low incorporation yields of small biologically active RNAs.

Exogenous loading

Electroporation and sonication.

To address the limitations of endogenous loading, exogenous loading methods, such as electroporation18 and sonication42,62, can be applied to incorporate large synthetic RNAs. However, electroporation can lead to RNA aggregation in the size range of RNA loaded into extracellular vesicles, making it difficult to distinguish RNA aggregates from extracellular vesicles. Sonication, by contrast, can affect the integrity of extracellular vesicles63. In particular, electroporation can be used to incorporate oligonucleotides, that is, unmodified or chemically modified single- or double-stranded DNA or RNA molecules, into extracellular vesicles. For example, a single-stranded antisense oligonucleotide targeting miR-210, an miRNA that can restore immune balance and prevent the development of psoriasis-like inflammation, can be loaded into extracellular vesicles isolated from interferon-γ (IFNγ)-stimulated mesenchymal stem cells (MSCs) by electroporation64. Intradermal administration of these extracellular vesicles can reduce psoriasis symptoms in mice.

Cas9 ribonucleoproteins can also be exogenously loaded into extracellular vesicles for gene editing65. For example, Cas9 and sgRNA complexes can be added to extracellular vesicles, isolated from hepatic stellate cells, followed by electroporation to form extracellular vesicle–Cas9–sgRNA complexes. These Cas9- and sgRNA-carrying extracellular vesicles enable cytosolic delivery of Cas9 in vitro, and, following cargo transfer into the nucleus, editing of target genes. However, although Cas9 can be efficiently loaded into extracellular vesicles by electroporation, loading large Cas9 proteins remains challenging owing to their large size. Therefore, the advantage of electroporation for Cas9 proteins and its effects on their functional activity remain to be investigated.

Hydrophobic modification.

Hydrophobic modification enables exogenous loading of small RNAs, for example, through cholesterol conjugation17. Extracellular vesicles modified with cholesterol show a homogeneous size distribution following isolation by size-exclusion chromatography and differential centrifugation17, with an average diameter of about 140 nm and a characteristic cup shape. Because of its hydrophobicity, conjugation of cholesterol enables the loading of stabilized therapeutic oligonucleotides, antisense oligonucleotides, antagomirs and aptamers45. The loading capacity of extracellular vesicles can be further improved using synthetic lipids; for example, siRNA can be loaded into vesicle–liposome hybrid nanoparticles by mixing extracellular vesicles, siRNA and a lipid thin film66.

Plasmid DNA transfer.

Loading extracellular vesicles with large genetic cargo (5–10 kb), such as pDNA, which contain promoters and enhancers in their sequence, remains challenging owing to their small size and limited loading capacity13,67–70. pDNA can be loaded into extracellular vesicles by direct incubation of pDNA with extracellular vesicles71, because pDNA can diffuse through the membrane of extracellular vesicles down a concentration gradient. However, passive integration of pDNA is less efficient than active loading by electroporation72. To improve the efficiency of pDNA transfer, pDNA can be complexed with lipids or proteins prior to incubation72.

Supercharged proteins.

Supercharged proteins, whose surface charge is modified by the incorporation of positive amino acids, can also be usd to load pDNA into extracellular vesicles – for example, to enable the expression of transgenes in the brains of mice73. This approach can also be applied to transport Cas-encoding pDNA into extracellular vesicles. For example, pDNA encoding CRISPR–Cas9 and sgRNA targeting KRAS with G12D mutation (KRAS-G12D) can both be exogenously incorporated into HEK293-cell-derived extracellular vesicles using transfection reagents, such as Exo-Fect74, to target oncogenic KRAS-G12D in subcutaneous and orthotopic pancreatic cancer mouse models.

Active loading.

RNA can also be actively incorporated into extracellular vesicles. In particular, extracellular vesicles derived from group O red blood cells are suitable for mRNA loading40. These extracellular vesicles are devoid of DNA or RNA and easy to scale up40. Importantly, because they do not contain A or B antigens on their surface, they can serve as universal donor extracellular vesicles. Because red blood cells have no nucleus and are therefore devoid of DNA, red blood cell-derived extracellular vesicles do not pose the risk of horizontal gene transfer. Therefore, red blood cell-derived extracellular vesicles offer an efficient and versatile platform for RNA drug delivery that is safe, scalable and applicable to various gene therapies. They can be used, for example, to deliver therapeutic RNAs, including both short RNAs (such as antisense oligonucleotides and gRNAs) and long RNAs (such as Cas9 mRNAs)40.

Extracellular vesicle–liposome hybrids

Genome editing by CRISPR–Cas9 technology requires the co-delivery of different components75, including Streptococcus pyogenes Cas9 (spCas9) and gRNA in suitable delivery vehicles. However, the size of these components makes it difficult to package them into extracellular vesicles, and spCas9 exceeds the DNA coding cassette capacity of AAV vectors. Therefore, hypercompact Cas species, such as Cas12 and Staphylococcus aureus Cas9 (saCas9), have been explored for AAV delivery76. Alternatively, Cas9 pDNA can be incorporated into extracellular vesicle–liposome hybrid particles by incubating a pDNA–liposome premix for 12 h at 37° C (ref. 36). The extracellular vesicle–liposome hybrids are then engineered by fusing extracellular vesicles with synthetic liposomes. Liposomes can encapsulate larger genetic cargo, such as Cas9 pDNA, more efficiently than can extracellular vesicles, which are limited by their natural cargo-carrying capacity77. We note that loading extracellular vesicles by fusion with synthetic liposomes might affect their integrity and introduce antigenic or toxic components, which might have to be considered in their clinical translation.

Ligand–receptor binding

The loading of large proteins into extracellular vesicles can be improved by exploiting ligand–receptor binding, for example, by applying light-induced protein heterodimerization78. CRY2 and CIBN heterodimerize in response to blue light (450–488 nm). This mechanism can be exploited for extracellular vesicle packaging, for example, by fusing CRY2 to Cas9 protein and CIBN to the tetraspanin CD9. By exposing cells that express these two proteins to blue light, CRY2 fused to Cas9 proteins can be actively loaded into extracellular vesicles that express CIBN fused to CD978. Extracellular vesicle loading can also be facilitated by fusing Cas to extracellular vesicle proteins, such as arrestin domain-containing protein 179. Similarly, CRISPR–CasRx (Cas13d) protein, an RNA-editing tool, can be engineered to bind to tissue plasminogen activator peptide ligand at the N terminus, which can then interact with the corresponding receptors on the membrane of the extracellular vesicle and facilitate its packaging into extracellular vesicles80.

In addition, nanobody-based strategies can be applied. For example, to load Cas9 into extracellular vesicles, green fluorescent protein (GFP) can be fused to the inner side of CD63, and Cas9 can be fused to a GFP-binding nanobody. Cas9 proteins are then selectively captured by the GFP nanobody and loaded into vesicles81. Moreover, N-myristoylation, an irreversible co-post-translational modification whereby a myristol group covalently binds to N-terminal glycine, can be exploited to incorporate myristoylated Cas9 into extracellular vesicles thanks to the association of myristoylated proteins with the vesicle membrane82.

Targeting extracellular vesicles

The delivery of extracellular vesicle-mediated gene therapy to specific cells and organs remains a translational challenge. Extracellular vesicles have intrinsic targeting properties upon in vivo administration owing to their lipid, protein and glycan composition that influences tropism toward specific organs or disease sites83. In particular, the surface composition of extracellular vesicles, including integrins and tetraspanins, contribute to their tropism77. However, directing extracellular vesicles to specific diseased tissues or cells typically requires additional engineering of their surfaces (Fig. 2).

Intrinsic versus engineered targeting

The intrinsic tumour-tissue-targeting abilities of tumour-derived extracellular vesicles can be exploited to exogenously deliver anti-cancer drugs in a cancer-cell-specific manner. For example, extracellular vesicles derived from a human ovarian cancer cell line (SKOV3) selectively accumulate in ovarian cancer cells in xenograft mouse tumour models after intravenous administration, probably owing to their cell tropism84. These extracellular vesicles can be applied to deliver CRISPR–Cas9 to ovarian cancer cells and suppress the expression of poly (ADP-ribose) polymerase-1 (PARP-1), resulting in the induction of cell apoptosis84.

To increase the tropism of cancer-derived extracellular vesicles, they can be engineered by surface modification with targeting moieties, including antibodies or peptides (Figs. 1c, 2), albeit with potentially undesired consequences that might compromise safety75. Higher tropism might have beneficial therapeutic effects, but whether other biomolecules transported by cancer-derived extracellular vesicles can cause unintended effects remains unclear. In particular, we note that extracellular vesicles originating from cancer cells have been associated with increasing cancer-cell proliferation and survival85, anti-tumour inflammation86 as well as increasing invasive and metastatic behaviour of cancer cells87. Similar issues might arise for extracellular vesicles isolated from immortalized cell lines, such as HEK293T cells, which express the oncoprotein SV40 T antigen88. Therefore, the effects of extracellular vesicles derived from cancer or immortalized cell lines on different cell types need to be determined88.

In addition, whether native extracellular vesicles are equipped with protein and ligands to selectively interact with specific cell types or tissues remains unknown89. The degree of innate tropism might be limited to organs of the reticuloendothelial system, such as the liver, spleen, lungs and kidney, which might be the predominant sites of extracellular vesicle accumulation upon exogenous administration2. Furthermore, targeting efficiency evaluation often relies on vesicles labelled with chemical dyes, which could introduce bias or artefacts. Alternatively, peptides, antibodies and proteins can be added to extracellular vesicles to promote cell- and organ-specific tropism.

Cell-specific targeting

Extracellular vesicle surfaces can be decorated to improve their circulation half-life90, homing ability91 and accumulation in desired body niches83. In particular, extracellular vesicles can be targeted by functionalizing their surface with cell-specific peptides7. For example, the mouse brain can be targeted by intravascular delivery of extracellular vesicles derived from dendritic cells engineered to express the membrane-associated protein LAMP2B, fused to the rabies virus glycoprotein (RVG) peptide that binds to nicotinic acetylcholine receptors16,92. In addition to the central nervous system, peripheral nervous system and neuromuscular synapses, the nicotinic acetylcholine receptor is also present in other tissues, including the kidney. Accordingly, these tissues can be targeted with extracellular vesicles derived from skeletal muscle satellite cells68 engineered to express the RVG peptide93. Similarly, systemic administration of bone marrow MSC-derived RVG-modified extracellular vesicles, loaded with miR-124, into mice stimulates cortical neural progenitors to assume neuronal identity and protect the brain against ischaemic injury by stimulating cortical neurogenesis18. Therefore, RVG-modified extracellular vesicles might be therapeutically applied for the targeted delivery of gene drugs to the brain.

Extracellular vesicles can also be functionalized with a T7 peptide that binds to the transferrin receptor, typically overexpressed on tumour cells. Here, the peptide can be added through endogenous incorporation using a lentivirus encoding T7-Lamp2b94. These T7-peptide-modified extracellular vesicles can deliver an siRNA to silence galectin-9 (a tumour immunosuppressor that inhibits TLR7), to repress the growth of glioblastoma cells in syngeneic mice and to activate anti-tumour immune responses. Moreover, extracellular vesicles can be modified with angiopep-2, a specific LRP1 ligand, to aid their blood–brain barrier penetration and to deliver synthetic siRNA against STAT3 specifically to glioblastoma cells. STAT3 mediates the transcription of downstream genes during glioblastoma progression42. The synthetic siRNA can be loaded into vesicles by ultrasonication, and the resulting targeted vesicles show high blood stability, efficient cellular uptake, anti-glioblastoma effects in vitro and inhibition of the proliferation of orthotopic human glioblastoma xenografts in nude mice with limited side effects, increasing the survival of the animals42.

Extracellular vesicles can also be designed for exogenous mRNA delivery to HER2-bearing breast cancer cells. For example, extracellular vesicles displaying an anti-HER2 single-chain fragment variable (scFv) antibody by plasmid transfection of HEK293 cells target HER2-expressing breast cancer cells. The vesicles can further be loaded with HChrR6-encoding mRNA and used in conjunction with CNOB to kill HER2-expressing cells, resulting in near-complete growth arrest of implanted orthotopic HER2+ breast cancer tumours in mice31. Functionalization of extracellular vesicles also enables receptor-specific delivery of gene-based enzyme prodrug therapies to breast cancer cells as well as other targets, including brain and prostate cancer cells31.

CD163+ macrophage-derived extracellular vesicles, decorated with an RGD (Arg–Gly–Asp) peptide that enables adhesion to fibronectin in the extracellular matrix, can deliver TGFβ to neovascular endothelial cells at the site of spinal cord injury, thereby improving neurological function in a mouse model of spinal cord injury95; here, the RGD peptide is added endogenously to the surface of the vesicles using a green circularly permuted (gc)-GFP-RGD-tag-Lamp2-expressing vector93. In addition, high-throughput screening (phage display) has been applied to identify peptides that bind to skeletal muscle. These peptides can be fused to CD63 on the surface of extracellular vesicles to enable targeted delivery of antisense oligonucleotides and induce dystrophin expression in a mouse model of Duchenne’s muscular dystrophy96. Similarly, peptide-mediated targeting can be achieved by modifying extracellular vesicles with angiopep-2 and the tTAT peptide, a cell-penetrating peptide that can enter the nucleus through receptor- and transporter-independent mechanisms. These vesicles can penetrate the blood–brain barrier and deliver a Cas9 sgRNA complex to knock out glutathione synthase and increase the radiosensitivity of glioblastoma in the mouse brain91. This approach provides high gene-editing efficiency in glioblastoma (up to 67.2%) with negligible off-target gene editing.

However, peptides might be degraded on the surface of extracellular vesicles during biogenesis and secretion. Moreover, peptides fused to surface-exposed proteins are susceptible to proteolytic degradation in the extracellular space97 and, therefore, may not be correctly displayed on the surface of extracellular vesicles. This might be partly addressed by adding engineered glycosylated motifs to the fusion peptides to prevent degradation91,98.

Antibody and protein targeting

Antibodies can be applied to functionalize extracellular vesicles with receptor- or surface-protein-specific targeting moieties. For example, to target and treat glioblastoma, MSCs can be engineered through plasmid transfection to produce extracellular vesicles that present anti-EGFR variant3 (vIII) antibody on their surface and that incorporate the cytosine deaminase enzyme, which enables activation of the prodrug 5-fluorocytosine (5-FC) together with the apoptosis-inducing agent miR-34 (ref. 37). Similarly, extracellular vesicles derived from epithelial cells can be designed with a CD19-targeting chimeric antigen receptor (CAR) to provide selective tropism to CD19+ B cell malignancies and deliver the CRISPR–Cas9 system. These CD19–CAR-extracellular vesicles accumulate in malignant B cells, where they release the CRISPR–Cas9 system to knock down the MYC oncogene in vitro and in vivo99. In addition, extracellular vesicles derived from HEK293FT cells (a version of HEK293 cells that are genetically modified by the Simian virus 40, SV40) for enhanced transfection efficiency and expression of transfected genes) can be engineered to present anti-CD2 fused to their PDGFR transmembrane domain, thereby possessing both targeting and fusogenic domains. These vesicles can deliver Cas9 sgRNA in vitro to primary human CD4+ T cells97 to knock out the gene encoding C–X–C chemokine co-receptor type 4, a clinically relevant target for the HIV. Extracellular vesicle CD64 molecules can also be exploited as adaptors for targeting ligands; for example, this approach can be applied to add anti-CD71, a membrane glycoprotein, and anti-programmed cell death-ligand 1 (PD-L1) antibodies100 to extracellular vesicles to target glioblastoma cells and generate anti-tumour immune activity in mouse models.

Targeting moieties can also be added using cross-linking reactions to covalently bind macromolecules, small molecules, carbohydrates, polysaccharides and peptides to the extracellular vesicle surface101–104. For example, extracellular vesicles can be isolated from Raw 264.7 cells105 and loaded with super-paramagnetic iron oxide nanoparticles (SPIONs) and curcumin. The vesicles can further be surface-functionalized with a neuropilin-1-targeted peptide (RGERPPR), a specific ligand for neuropilin-1, which is overexpressed on glioma cells105, using click chemistry that involves a cycloaddition reaction of sulfonyl azide. Administered to glioma cells in vitro and in orthotopic glioma mouse models, these vesicles can cross the blood–brain barrier and target glioma with imaging and therapeutic agents105. Similarly, modification of extracellular vesicles with azide functional groups allows the conjugation of anti-CD47 antibodies through reaction with dibenzocyclooctyne. These vesicles target 4T1 breast tumours expressing the CD47 receptor, thereby increasing their uptake into tumour sites106. Extracellular vesicles that target hepatocytes with high expression levels of asialoglycoprotein receptor 1 (ASGR1) can be endogenously generated through transfection of human primary hepatocytes with a pDNA expression cassette for the chloride intracellular channel 1 (CLIC1) and HaloTag proteins107. The CLIC1-HaloTag enables the addition of N-acetylgalactosamine (GalNAc) to the surface of the vesicles. ASGR1 interacts with GalNAc, resulting in the rapid internalization of the receptor–ligand complex. Thus, the CLIC1-HaloTag allows the addition of multiple targeting ligands on the surface of extracellular vesicles.

Deployment of gene therapy by extracellular vesicles

The efficiency of extracellular vesicle-mediated gene therapy is determined by the mode of entry of both the vesicles and the cargo into the destinate cells. Cells employ multiple clearance systems, including autophagy, lysosomal degradation108 and immune responses109, to protect themselves from external stimuli. Overcoming these clearance mechanisms remains a key bottleneck for extracellular-vesicle-mediated gene therapy, and lessons might be learned from approaches applying viral vectors and synthetic nanoparticles, including endosomal escape mechanisms to avoid lysosomal degradation or inhibition of autophagy to prevent intracellular clearance. Extracellular vesicles can therefore be engineered to enhance their stability (for example, through modification with PEG) or to enable their endosomal escape (for example, through modification with pH-sensitive peptides). Depending on the type of cargo (synthetic molecules, DNA, RNA, protein, or a combination of biomolecules) (Fig. 3a), specific cell-entry machinery must be incorporated into extracellular vesicles. Moreover, specific cell types might require distinct tools, because cargo deposition typically exploits cell-type-specific properties, such as receptor-mediated mechanisms or low-pH conditions during endocytic uptake110.

Fig. 3 |. Cargo deployment.

Fig. 3 |

a, Extracellular vesicles can carry different types of RNA and plasmid DNA (pDNA), which serves as promoter. Examples include: single guide RNA (sgRNA) and Cas9 complexes for gene editing, short interfering RNA (siRNA), antisense oligonucleotides and microRNAs (miRNAs) for mRNA silencing, and circular RNA (circRNA) and long noncoding RNA (lncRNA) for transcriptional regulation. The vesicles enter the cytosol from the extracellular space through endosomal uptake or by fusing with the cell membrane. The cargo needs to reach the nucleus for gene addition, gene editing and transcriptional regulation, whereas RNAs for translation or mRNA silencing are functional in the cytoplasmic environment. AAA, b, The efficiency of cargo delivery can be improved by adding viral fusogens to the surface of extracellular vesicles or by combining them with synthetic liposomes.

Fusogenic proteins

Fusogenic proteins – glycoproteins that facilitate the fusion of cell membranes, such as vesicular stomatitis virus glycoprotein (VSV-G) – can be adopted from viral vectors and presented on the surface of extracellular vesicles (Fig. 3b), for example, to deliver recombinant Cre and CRISPR–Cas9 ribonucleoproteins111. In addition, the baculoviral fusogenic glycoprotein gp64 can be incorporated into extracellular vesicles to deliver cargo, such as Texas Red-labelled dextran, to cells112. However, the immunogenic properties of viral proteins might impede the possibility for multiple treatments and broad cell tropism. Therefore, non-viral fusogenic proteins have also been explored.

Human endogenous retroviral envelope proteins, that is, abundant sequences that persist within the human genome as remnants of ancient retroviral infections, such as syncitin-1, can be added to extracellular vesicles, for example, to deliver diphtheria toxin to cancer cells113. Similarly, murine syncitin-A-mediated transfer of cre mRNA by extracellular vesicles has proved effective in fibroblasts53. However, syncytin has to interact with specific receptors on recipient cells to allow fusion and might thus work better for nanotubule transfer than for extracellular vesicles114 owing to its ability to promote direct cell–cell connections and enhance membrane fusion, which is crucial for nanotubule formation and cargo transfer. Alternatively, cargo delivery might be facilitated by exploiting other endosomal escape routes; for example, following extracellular vesicle uptake, endosomal compartments can be lysed through destabilization of osmotic pressure; or cargo delivery might be facilitated through transient transfection of donor cells with pDNA or cell-penetrating peptide complexes115.

Other uptake mechanisms

Viruses depend on cellular uptake to deliver their genetic material into cells for replication, and can employ mechanisms, such as pH sensors, to detect their uptake and ensure that they enter an environment conducive to their propagation. Therefore, many viral fusogenic proteins have evolved to act in pH-sensitive compartments, such as endocytic compartments. To avoid pH conditions in endocytic compartments that may compromise the integrity of the cargo, extracellular vesicle membranes can be modified with connexin43 (ref. 116), which can generate gap junctions that form channels between the vesicle lumen and the recipient cell cytoplasm, for example, to deliver miRNAs117. Similarly, cellular uptake of red blood cell-derived extracellular vesicles does not require fusion, but rather occurs through endolysosomal escape, as demonstrated by blocking uptake with bafilomycin A1, an inhibitor of endosomal acidification, and 4-bromobenzaldehyde N-(2,6-dimethylphenyl) semi-carbazone, a blocker of endosome maturation. Such red blood cell-derived vesicles can further be coated with monoclonal antibodies targeting C–X–C chemokine receptor type 4 (CXCR4) to deliver antisense oligonucleotides into a xenograft model of acute myeloid leukaemia118.

Modifications for intracellular cargo release

Endogenous loading into extracellular vesicles requires encoded signatures that target cargo to the secretory pathway and/or for membrane association, which can complicate cargo release in other cell compartments. To allow cargo release in specific compartments, linkers can be incorporated between the loading signatures in vesicle-generating cells and organelle-targeting signatures. For example, to release cargo from vesicle membranes into the vesicle lumen, pH-dependent self-cleavage sites, such as intein, can be implemented119 to separate cargo from the extracellular membrane in response to pH shifts and enrich free cargo proteins inside the vesicle119. Alternatively, optically reversible protein–protein interactions can be exploited; for example, cargo can be attached to the vesicles under blue light and detached after vesicle biogenesis by turning off the blue light120. Importantly, pDNA-transfection reagents, used to generate extracellular-vesicle-producing cells, can affect the outcomes of extracellular vesicle-mediated delivery and uptake, because the transfection agent might be incorporated in the vesicle outer membrane layer121. Synthetic lipids can be implemented to aid cargo release; for example, liposome–extracellular vesicle hybrids that incorporate fusogenic lipids can promote Ca2+-mediated and polyethylene glycol 8000 (PEG8000)-mediated fusion to deliver drugs, such as doxorubicin, to recipient cells122.

In addition, nuclear export signals can guide base editors to the extracellular-vesicle-secretion pathway. Upon vesicle formation, Gag proteases become active in the cytoplasm and can cleave an encoded Friend murine leukemia virus protease linker, thereby exposing nuclear localization signals to improve editing efficiencies in recipient cells123. This method is essential for delivering large cargos, such as prime editors. Incorporating multiple cleavage sites in the protein tether further promotes editing, owing to more efficient deployment of the prime editor in the recipient cell nucleus124.

Therefore, tools to overcome cellular barriers and navigate cargo to the appropriate subcellular compartment must be incorporated into extracellular vesicles.

Clinical applications

Extracellular vesicles have been preclinically and clinically tested for a variety of diseases125, including vaccination, cardiovascular disease, cancer therapy, immune modulation, regeneration and drug delivery126 by exploring different endogenous (native) cargos, including proteins, DNA, RNA, lipids, glycans, small RNAs and miRNA127. In addition, various administration routes have been explored, depending on the tissue, including intraperitoneal, topical, subcutaneous and intravenous administration. Moreover, the intranasal delivery of extracellular vesicles has been preclinically tested128.

In the first clinical trial of extracellular vesicles, vesicles derived from dendritic cells, loaded with major histocompatibility complex (MHC) class I peptides and tumour antigens, have been tested for cancer vaccination7,129. In addition, extracellular vesicles derived from MSCs130 are being explored in clinical trials owing to the regenerative benefit of MSCs and their released extracellular vesicles. MSCs are typically derived from human adipose tissue or umbilical cords. However, their properties vary depending on the tissue of origin, donor, mode of isolation and culture conditions. Therefore, there is a need to standardize these cells as sources of extracellular vesicles; for example, clonal immortalized cell lines can be used131. In addition, other cell types, such as red blood cells132, neural stem cells133, natural killer cells134 and endothelial cells135 are being tested as sources for extracellular vesicles. Importantly, if available, potency tests should be carried out to validate the action of extracellular vesicles136.

MSC-derived extracellular vesicles have been tested for psoriasis (NCT05523011), diabetes mellitus type I (NCT02138331), respiratory stress, especially associated with COVID-19 (NCT04493242, NCT05354141, NCT05787288 and NCT04798716), healing of macular holes (NCT03437759), blood vessel repair after aortic surgery (NCT04356300), osteoarthritis (NCT05060107), Alzheimer’s disease (NCT04388982), acute ischaemic stroke (NCT03384433), skin burns (NCT05078385), Gaucher lysosomal storage disease (NCT05843552) and as an inhalation in controls to screen for potential toxic effects (NCT04313647). In addition, MSC-derived extracellular vesicles loaded with miR-124 can ameliorate brain injury by promoting neurogenesis137, and have entered clinical trials for treating ischaemic stroke (NCT03384433). Moreover, in an ongoing clinical trial, MSC-derived extracellular vesicles loaded with KRAS-G12D siRNA are tested for the treatment of metastatic pancreatic cancer with the KRAS-G12D driver mutation (NCT03608631).

To extend on-site delivery of extracellular vesicles derived from MSCs, MSCs can be tethered to fibrous polyester materials138, core–shell hyaluronic acid microneedles or hydrogels98,139. Such approaches can further be combined with bioactive glass stimulation140. In addition, stents (short narrow metal or plastic tubes) can be designed to release extracellular vesicles141. To release extracellular vesicles in specific compartments when needed, they can be engineered with cleavable or breakable bonds tethered to a matrix that is implanted at the target site. For example, stents can be designed with reactive oxygen species (ROS)-responsive linkers that break in the presence of ROS, thereby releasing miRNA-containing extracellular vesicles in conditions with high ROS levels, such as ischaemia, myocardial infarction and renal ischaemia injury141. Alternatively, stable and controlled extracellular vesicle release can be achieved with hydrogel scaffolds, such as nano-hydroxyapatite and poly-ε-caprolactone hydrogels, which can release around 1012 extracellular vesicles per millilitre over a >20-day period142.

Outlook

Extracellular vesicles can be engineered to serve as non-viral platforms for the delivery of gene-editing tools. However, engineering challenges remain to be overcome for the clinical application of extracellular vesicles for gene therapy (Box 1). In addition to increasing the cargo-loading capacity, targeting moieties, fusion molecules or pore-forming proteins need to be incorporated in the extracellular vesicle surface to effectively deposit gene-therapy compounds inside specific cells. Moreover, signals need to be added to cargo molecules to ensure their escape from the degradative endosomal pathway and their delivery to the designated cell compartment. Furthermore, to establish extracellular vesicles as a gene-therapy platform, dosing needs to be reduced, off-target tropism must be limited and efficiency should be improved. Importantly, cargo-loading efficiency should be assessed and reported to determine the cargo copy numbers per vesicle.

Box 1 |. Translational considerations.

Extracellular vesicles can deliver proteins and oligonucleotides for gene therapy; however, various obstacles remain to be overcome for their clinical translation as gene-therapy platforms.

  • Targeting moieties, fusion molecules or pore-forming proteins need to be incorporated on the extracellular vesicle surface to deposit gene-therapy compounds into desired cells.

  • Extracellular-vesicle dosage needs to be reduced, off-target tropism should be prevented and efficiency must be optimized.

  • Various types of extracellular vesicle from different cell sources and with different inherent cargos should be tested to establish their susceptibility to click-chemistry-based surface modification.

  • The mechanisms of extracellular-vesicle uptake, subcellular fate and cargo deployment need to be thoroughly investigated.

  • Homogeneous extracellular vesicle products need to be manufactured in a standardized way, following good laboratory practice.

  • Extracellular vesicle production in bioreactors remains costly and suffers from biologic variability; therapeutic extracellular-vesicle production in vivo could reduce both costs and the dependence on variable cell sources.

The ability of innate extracellular vesicles to interact with specific cells and tissues remains to be fully evaluated, although their innate tropism appears to be limited77. Chemical conjugation, particularly bio-orthogonal chemistry, enables the modification of extracellular vesicles with receptor-specific ligands. Therefore, extracellular vesicles from different cell sources and with different cargo types should be tested for their chemical modification potential. Importantly, the mechanisms of extracellular vesicle uptake, subcellular fate and cargo deployment remain to be fully investigated. In addition, approaches are required to promote fusion with cell organelle membranes, uptake into the nucleus and functional cargo release within cells.

With regard to clinical translation, the heterogeneous nature, low yield and inconsistent preparation of extracellular vesicles can complicate good manufacturing practices and reproducibility. Therefore, consistent large-scale isolation and purification protocols should be developed to establish scaled-up production of standardized products143 and quality control, in addition to optimizing storage conditions, targeting capability and loading capacity, as well as cytosolic and nuclear delivery. Importantly, possible immunogenic side effects of surface-displayed fusogens should be investigated, and whether the delivery of immortalizing genes might lead to the transformation of normal cells or increase the tumorigenicity of cancer cells should be tested. Furthermore, the mode of delivery needs to be optimized because extracellular vesicles have a short half-life in vivo2,144.

Finally, the production of extracellular vesicles remains costly, compared to the production of synthetic delivery platforms, owing to the requirement of large bioreactors for high-yield vesicle production from source cells. The cost challenge could be addressed in part by producing therapeutic extracellular vesicles in vivo, using the body as an extracellular-vesicle factory.

Key points.

  • The diverse content of extracellular vesicles allows their application as vectors for gene therapy, including for gene addition, gene editing, mRNA, silencing and gene transcription regulation.

  • Extracellular vesicles can be loaded with gene-therapy tools using endogenous approaches, by engineering cells to secrete extracellular vesicles loaded with specific cargo, or using exogenous methods to directly incorporate cargo into vesicles.

  • Extracellular vesicles can be engineered to display targeting moieties that enable the delivery of gene-therapy tools to specific cells.

  • Fusogenic proteins or components of viral cell entry can be displayed on the surface of extracellular vesicles to control cargo deployment in the intracellular space.

  • Extracellular vesicles as carriers of protein, DNA or RNA are being tested in clinical trials for a variety of diseases.

Acknowledgements

We thank S. McDavitt for skilled assistance. X.O.B. is funded by National Cancer Instititue (NCI) grant CA232103 and National Institute of Neurological Disorders and Stroke (NINDS) grant NS122163. X.O.B., E.D.I. and K.B. are funded by Ionis Pharmaceuticals. K.B. is funded by National Institutes of Health (NIH) grant K22 CA2802019-01. W.O. was funded by Pharmacokinetics Research Laboratories and Daiichi Sankyo.

Footnotes

Competing interests

The authors declare no conflicts of interest.

References

  • 1.O’Brien K, Breyne K, Ughetto S, Laurent LC & Breakefield XO RNA delivery by extracellular vesicles in mammalian cells and its applications. Nat. Rev. Mol. Cell Biol 21, 585–606 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lai CP et al. Dynamic biodistribution of extracellular vesicles in vivo using a multimodal imaging reporter. ACS Nano 8, 483–494 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Herrmann IK, Wood MJA & Fuhrmann G Extracellular vesicles as a next-generation drug delivery platform. Nat. Nanotechnol 16, 748–759 (2021). [DOI] [PubMed] [Google Scholar]
  • 4.Fu S, Wang Y, Xia X & Zheng JC Exosome engineering: current progress in cargo loading and targeted delivery. NanoImpact 20, 100261 (2020). [Google Scholar]
  • 5.Galsgaard Malle M et al. Programmable RNA loading of extracellular vesicles with toehold-release purification. J. Am. Chem. Soc 146, 12410–12422 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Cheng L & Hill AF Therapeutically harnessing extracellular vesicles. Nat. Rev. Drug. Discov 21, 379–399 (2022). [DOI] [PubMed] [Google Scholar]
  • 7.Nieland L, Mahjoum S, Grandell E, Breyne K & Breakefield XO Engineered EVs designed to target diseases of the CNS. J. Control. Rel 356, 493–506 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Macedo-Pereira A, Martins C, Lima J & Sarmento B Digging the intercellular crosstalk via extracellular vesicles: may exosomes be the drug delivery solution for target glioblastoma? J. Control. Rel 358, 98–115 (2023). [DOI] [PubMed] [Google Scholar]
  • 9.Duan L et al. Exosome-mediated delivery of gene vectors for gene therapy. Nanoscale 13, 1387–1397 (2021). [DOI] [PubMed] [Google Scholar]
  • 10.Flotte TR Size does matter: overcoming the adeno-associated virus packaging limit. Respir. Res 1, 16–18 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cao D et al. Innate immune sensing of adeno-associated virus vectors. Hum. Gene Ther 35, 451–463 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Leandro K et al. Exploring the potential of cell-derived vesicles for transient delivery of gene editing payloads. Adv. Drug Deliv. Rev 211, 115346 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Huang X et al. Current and prospective strategies for advancing the targeted delivery of CRISPR/Cas system via extracellular vesicles. J. Nanobiotechnol 21, 184 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Nolte’T Hoen ENM et al. Deep sequencing of RNA from immune cell-derived vesicles uncovers the selective incorporation of small non-coding RNA biotypes with potential regulatory functions. Nucleic Acids Res. 40, 9272–9285 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wei Z et al. Coding and noncoding landscape of extracellular RNA released by human glioma stem cells. Nat. Commun 8, 1145 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Alvarez-Erviti L et al. Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes. Nat. Biotechnol 29, 341–345 (2011). [DOI] [PubMed] [Google Scholar]
  • 17.Didiot MC et al. Exosome-mediated delivery of hydrophobically modified siRNA for huntingtin mRNA silencing. Mol. Ther 24, 1836–1847 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Yang J, Zhang X, Chen X, Wang L & Yang G Exosome mediated delivery of miR-124 promotes neurogenesis after ischemia. Mol. Ther. Nucleic Acids 7, 278–287 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Lee S-T et al. Exosome-based delivery of miR-124 in a Huntington’s disease model. J. Mov. Disord 10, 45–52 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Wang J, Lu Z, Wientjes MG & Au JL-S Delivery of siRNA therapeutics: barriers and carriers. AAPS J. 12, 492–503 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Winkle M, El-Daly SM, Fabbri M & Calin GA Noncoding RNA therapeutics – challenges and potential solutions. Nat. Rev. Drug. Discov 20, 629–651 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Judge AD et al. Sequence-dependent stimulation of the mammalian innate immune response by synthetic siRNA. Nat. Biotechnol 23, 457–462 (2005). [DOI] [PubMed] [Google Scholar]
  • 23.Heo JI & Ryu J Exosomal noncoding RNA: a potential therapy for retinal vascular diseases. Mol. Ther. Nucleic Acids 35, 102128 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Wang M et al. A comprehensive overview of exosome lncRNAs: emerging biomarkers and potential therapeutics in endometriosis. Front. Endocrinol 14, 1199569 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Zhang LJ et al. Crosstalk among long non-coding RNA, tumor-associated macrophages and small extracellular vesicles in tumorigenesis and dissemination. Front. Oncol 12, 1008856 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Abramowicz A & Story MD The long and short of it: the emerging roles of non-coding RNA in small extracellular vesicles. Cancers 12, 1445 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Canseco-Rodriguez A et al. Long non-coding RNAs, extracellular vesicles and inflammation in Alzheimer’s disease. Int. J. Mol. Sci 23, 13171 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Zhu D et al. Intrapericardial long non-coding RNA–Tcf21 antisense RNA inducing demethylation administration promotes cardiac repair. Eur. Heart J 44, 1748–1760 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Vilaça A et al. Extracellular vesicle transfer of lncRNA H19 splice variants to cardiac cells. Mol. Ther. Nucleic Acids 35, 102233 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kawamura Y, Sanchez Calle A, Yamamoto Y, Sato TA & Ochiya T Extracellular vesicles mediate the horizontal transfer of an active LINE-1 retrotransposon. J. Extracell. Vesicles 8, 1643214 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Wang JH et al. Anti-HER2 scFv-directed extracellular vesicle-mediated mRNA-based gene delivery inhibits growth of HER2-positive human breast tumor xenografts by prodrug activation. Mol. Cancer Ther 17, 1133–1142 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Tibensky M et al. Gene-directed enzyme/prodrug therapy of rat brain tumor mediated by human mesenchymal stem cell suicide gene extracellular vesicles in vitro and in vivo. Cancers 14, 735 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Maugeri M et al. Linkage between endosomal escape of LNP-mRNA and loading into EVs for transport to other cells. Nat. Commun 10, 4333 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Bu T et al. Exosome-mediated delivery of inflammation-responsive Il-10 mRNA for controlled atherosclerosis treatment. Theranostics 11, 9988–10000 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Nordin JZ Transfection reagents affect extracellular vesicle cargo transfer to recipient cells: the importance of appropriate controls in EV research. J. Extracell Vesicles 11, e12227 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Lin Y et al. Exosome-liposome hybrid nanoparticles deliver CRISPR/Cas9 system in MSCs. Adv. Sci 5, 1700611 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rahmani R et al. Engineered anti-EGFRvIII targeted exosomes induce apoptosis in glioblastoma multiforme. J. Drug. Target 31, 310–319 (2023). [DOI] [PubMed] [Google Scholar]
  • 38.Wang Q et al. ARMMs as a versatile platform for intracellular delivery of macromolecules. Nat. Commun 9, 960 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chen R et al. Friend or foe? Evidence indicates endogenous exosomes can deliver functional gRNA and Cas9 protein. Small 15, e1902686 (2019). [DOI] [PubMed] [Google Scholar]
  • 40.Usman WM et al. Efficient RNA drug delivery using red blood cell extracellular vesicles. Nat. Commun 9, 2359 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.O’Grady T et al. Sorting and packaging of RNA into extracellular vesicles shape intracellular transcript levels. BMC Biol. 20, 72 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Liang SF, Zuo FF, Yin BC & Ye BC Delivery of siRNA based on engineered exosomes for glioblastoma therapy by targeting STAT3. Biomater. Sci 10, 1582–1590 (2022). [DOI] [PubMed] [Google Scholar]
  • 43.Peruzzi JA et al. Enhancing extracellular vesicle cargo loading and functional delivery by engineering protein–lipid interactions. Nat. Commun 15, 5618 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Yang Z et al. Improved extracellular vesicle-based mRNA delivery for familial hypercholesterolemia treatment. Theranostics 13, 3467–3479 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.O’Loughlin AJ et al. Functional delivery of lipid-conjugated siRNA by extracellular vesicles. Mol. Ther 25, 1580–1587 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Bost JP et al. Delivery of oligonucleotide therapeutics: chemical modifications, lipid nanoparticles, and extracellular vesicles. ACS Nano 15, 13993–14021 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Amiri A et al. Exosomes as bio-inspired nanocarriers for RNA delivery: preparation and applications. J. Transl. Med 20, 125 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Chen H et al. CD33-targeting extracellular vesicles deliver antisense oligonucleotides against FLT3-ITD and miR-125b for specific treatment of acute myeloid leukaemia. Cell Prolif. 55, e13255 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Jayasinghe MK et al. Red blood cell-derived extracellular vesicles display endogenous antiviral effects and enhance the efficacy of antiviral oligonucleotide therapy. ACS Nano 17, 21639–21661 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Adamus T et al. Glioma-targeted delivery of exosome-encapsulated antisense oligonucleotides using neural stem cells. Mol. Ther. Nucleic Acids 27, 611–620 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Hung ME & Leonard JN A platform for actively loading cargo RNA to elucidate limiting steps in EV-mediated delivery. J. Extracell. Vesicles 5, 10.3402/jev.v5.31027 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Shrivastava S et al. Exosome-mediated stable epigenetic repression of HIV-1. Nat. Commun 12, 5541 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Segel M et al. Mammalian retrovirus-like protein PEG10 packages its own mRNA and can be pseudotyped for mRNA delivery. Science 373, 882–889 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Li Z et al. In vitro and in vivo RNA inhibition by CD9-HuR functionalized exosomes encapsulated with miRNA or CRISPR/dCas9. Nano Lett. 19, 19–28 (2019). [DOI] [PubMed] [Google Scholar]
  • 55.Chen K et al. Targeted extracellular vesicles delivered verrucarin A to treat glioblastoma. Biomedicines 10, 130 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Reshke R et al. Reduction of the therapeutic dose of silencing RNA by packaging it in extracellular vesicles via a pre-microRNA backbone. Nat. Biomed. Eng 4, 52–68 (2020). [DOI] [PubMed] [Google Scholar]
  • 57.Somiya M & Kuroda S Verification of extracellular vesicle-mediated functional mRNA delivery via RNA editing. Preprint at bioRxiv 10.1101/2022.01.25.477620 (2022). [DOI] [Google Scholar]
  • 58.Yang Z et al. Large-scale generation of functional mRNA-encapsulating exosomes via cellular nanoporation. Nat. Biomed. Eng 4, 69–83 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Yeo RWY et al. Mesenchymal stem cell: an efficient mass producer of exosomes for drug delivery. Adv. Drug. Deliv. Rev 65, 336–341 (2013). [DOI] [PubMed] [Google Scholar]
  • 60.Batrakova EV & Kim MS Using exosomes, naturally-equipped nanocarriers, for drug delivery. J. Control. Rel 219, 396–405 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Witwer KW et al. Standardization of sample collection, isolation and analysis methods in extracellular vesicle research. J. Extracell. Vesicles 2, 10.3402/jev.v2i0.20360 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Lamichhane TN et al. Oncogene knockdown via active loading of small RNAs into extracellular vesicles by sonication. Cell Mol. Bioeng 9, 315–324 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Kooijmans SAA et al. Electroporation-induced siRNA precipitation obscures the efficiency of siRNA loading into extracellular vesicles. J. Control. Rel 172, 229–238 (2013). [DOI] [PubMed] [Google Scholar]
  • 64.Zhang W et al. Small extracellular vesicles derived from MSCs have immunomodulatory effects to enhance delivery of ASO-210 for psoriasis treatment. Front. Cell Dev. Biol 10, 842813 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Wan T et al. Exosome-mediated delivery of Cas9 ribonucleoprotein complexes for tissue-specific gene therapy of liver diseases. Sci. Adv 8, eabp9435 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Evers MJW et al. Functional siRNA delivery by extracellular vesicle-liposome hybrid nanoparticles. Adv. Healthc. Mater 11, e2101202 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Berggreen AH, Petersen JL, Lin L, Benabdellah K & Luo Y CRISPR delivery with extracellular vesicles: promises and challenges. J. Extracell. Biol 2, e111 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Zhu X et al. The CRISPR/Cas9 system delivered by extracellular vesicles. Pharmaceutics 15, 984 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Lu Y, Godbout K, Lamothe G & Tremblay JP CRISPR–Cas9 delivery strategies with engineered extracellular vesicles. Mol. Ther. Nucleic Acids 34, 102040 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Liang Y et al. Cell-derived extracellular vesicles for CRISPR/Cas9 delivery: engineering strategies for cargo packaging and loading. Biomater. Sci 10, 4095–4106 (2022). [DOI] [PubMed] [Google Scholar]
  • 71.Kang M et al. Placental extracellular vesicles can be loaded with plasmid DNA. Mol. Pharm 20, 1898–1913 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Roerig J & Schulz-Siegmund M Standardization approaches for extracellular vesicle loading with oligonucleotides and biologics. Small 19, 2301763 (2023). [DOI] [PubMed] [Google Scholar]
  • 73.Breyne K et al. Exogenous loading of extracellular vesicles, virus-like particles, and lentiviral vectors with supercharged proteins. Commun. Biol 5, 485 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.McAndrews KM et al. Exosome-mediated delivery of CRISPR/Cas9 for targeting of oncogenic KrasG12D in pancreatic cancer. Life Sci. Alliance 4, e202000875 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Wang D, Zhang F & Gao G CRISPR-based therapeutic genome editing: strategies and in vivo delivery by AAV vectors. Cell 181, 136–150 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Kabadi AM et al. AAV-based CRISPR–Cas9 genome editing: challenges and engineering opportunities. Curr. Opin. Biomed. Eng 29, 100517 (2024). [Google Scholar]
  • 77.Choi W, Park DJ & Eliceiri BP Defining tropism and activity of natural and engineered extracellular vesicles. Front. Immunol 15, 1363185 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Osteikoetxea X et al. Engineered Cas9 extracellular vesicles as a novel gene editing tool. J. Extracell. Vesicles 11, e12225 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.Chen Z, Qiao Z, Wirth CR, Park H-R & Lu Q Arrestin domain-containing protein 1-mediated microvesicles (ARMMs) protect against cadmium-induced neurotoxicity. Extracell. Vesicle 2, 100027 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80.Li T et al. Engineered extracellular vesicle-delivered CRISPR/CasRx as a novel RNA editing tool. Adv. Sci 10, e2206517 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.Ye Y et al. An engineered exosome for delivering sgRNA:Cas9 ribonucleoprotein complex and genome editing in recipient cells. Biomater. Sci 8, 2966–2976 (2020). [DOI] [PubMed] [Google Scholar]
  • 82.Whitley JA et al. Encapsulating Cas9 into extracellular vesicles by protein myristoylation. J. Extracell. Vesicles 11, e12196 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 83.van der Koog L, Gandek TB & Nagelkerke A Liposomes and extracellular vesicles as drug delivery systems: a comparison of composition, pharmacokinetics, and functionalization. Adv. Healthc. Mater 11, e2100639 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.Kim SM et al. Cancer-derived exosomes as a delivery platform of CRISPR/Cas9 confer cancer cell tropism-dependent targeting. J. Control. Rel 266, 8–16 (2017). [DOI] [PubMed] [Google Scholar]
  • 85.Becker A et al. Extracellular vesicles in cancer: cell-to-cell mediators of metastasis. Cancer Cell 30, 836 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86.Smirnova O et al. Direct and cell-mediated EV–ECM interplay. Acta Biomater. 186, 63–84 (2024). [DOI] [PubMed] [Google Scholar]
  • 87.Kalluri R & McAndrews KM The role of extracellular vesicles in cancer. Cell 186, 1610–1626 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 88.Kavsan VM, Iershov AV & Balynska OV Immortalized cells and one oncogene in malignant transformation: old insights on new explanation. BMC Cell Biol. 12, 23 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 89.Wiklander OPB et al. Extracellular vesicle in vivo biodistribution is determined by cell source, route of administration and targeting. J. Extracell. Vesicles 4, 26316 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 90.Liang X et al. Extracellular vesicles engineered to bind albumin demonstrate extended circulation time and lymph node accumulation in mouse models. J. Extracell. Vesicles 11, e12248 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 91.Liu X et al. Engineered extracellular vesicle-delivered CRISPR/Cas9 for radiotherapy sensitization of glioblastoma. ACS Nano 17, 16432–16447 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 92.Wang Q, Cheng S, Qin F, Fu A & Fu C Application progress of RVG peptides to facilitate the delivery of therapeutic agents into the central nervous system. RSC Adv. 11, 8505–8515 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 93.D’Souza SE, Ginsberg MH & Plow EF Arginyl–glycyl–aspartic acid (RGD): a cell adhesion motif. Trends Biochem. Sci 16, 246–250 (1991). [DOI] [PubMed] [Google Scholar]
  • 94.Li C, Guan N & Liu F T7 peptide-decorated exosome-based nanocarrier system for delivery of Galectin-9 siRNA to stimulate macrophage repolarization in glioblastoma. J. Neurooncol 162, 93–108 (2023). [DOI] [PubMed] [Google Scholar]
  • 95.Peng W et al. Targeted delivery of CD163+ macrophage-derived small extracellular vesicles via RGD peptides promote vascular regeneration and stabilization after spinal cord injury. J. Control. Rel 361, 750–765 (2023). [DOI] [PubMed] [Google Scholar]
  • 96.Gao X et al. Anchor peptide captures, targets, and loads exosomes of diverse origins for diagnostics and therapy. Sci. Transl. Med 10, eaat0195 (2018). [DOI] [PubMed] [Google Scholar]
  • 97.Stranford DM et al. Genetically encoding multiple functionalities into extracellular vesicles for the targeted delivery of biologics to T cells. Nat. Biomed. Eng 8, 397–414 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 98.Ju Y, Hu Y, Yang P, Xie X & Fang B Extracellular vesicle-loaded hydrogels for tissue repair and regeneration. Mater. Today Bio 18, 100522 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 99.Xu Q et al. Tropism-facilitated delivery of CRISPR/Cas9 system with chimeric antigen receptor-extracellular vesicles against B-cell malignancies. J. Control. Rel 326, 455–467 (2020). [DOI] [PubMed] [Google Scholar]
  • 100.Dong S et al. Adaptive design of mRNA-loaded extracellular vesicles for targeted immunotherapy of cancer. Nat. Commun 14, 6610 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 101.Bhatta R et al. Metabolic tagging of extracellular vesicles and development of enhanced extracellular vesicle based cancer vaccines. Nat. Commun 14, 8047 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 102.Zheng W et al. Surface display of functional moieties on extracellular vesicles using lipid anchors. J. Control. Rel 357, 630–640 (2023). [DOI] [PubMed] [Google Scholar]
  • 103.Lim GT et al. Bioorthogonally surface-edited extracellular vesicles based on metabolic glycoengineering for CD44-mediated targeting of inflammatory diseases. J. Extracell. Vesicles 10, e12077 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 104.Smyth T et al. Surface functionalization of exosomes using click chemistry. Bioconjug. Chem 25, 1777–1784 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 105.Jia G et al. NRP-1 targeted and cargo-loaded exosomes facilitate simultaneous imaging and therapy of glioma in vitro and in vivo. Biomaterials 178, 302–316 (2018). [DOI] [PubMed] [Google Scholar]
  • 106.Nie W et al. Responsive exosome nano-bioconjugates for synergistic cancer therapy. Angew. Chem. Int. Ed. Engl 59, 2018–2022 (2020). [DOI] [PubMed] [Google Scholar]
  • 107.Ivanova A et al. Creating designer engineered extracellular vesicles for diverse ligand display, target recognition, and controlled protein loading and delivery. Adv. Sci 10, e2304389 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 108.Mulcahy LA, Pink RC & Carter DRF Routes and mechanisms of extracellular vesicle uptake. J. Extracell. Vesicles 3, 10.3402/jev.v3.24641 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 109.Robbins PD & Morelli AE Regulation of immune responses by extracellular vesicles. Nat. Rev. Immunol 14, 195–208 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 110.Ribovski L, Joshi BS, Gao J & Zuhorn IS Breaking free: endocytosis and endosomal escape of extracellular vesicles. Extracell. Vesicles Circ. Nucl. Acids 4, 283–305 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 111.Ilahibaks NF et al. TOP-EVs: technology of protein delivery through extracellular vesicles is a versatile platform for intracellular protein delivery. J. Control. Rel 355, 579–592 (2023). [DOI] [PubMed] [Google Scholar]
  • 112.Ishikawa R, Yoshida S, Sawada SI, Sasaki Y & Akiyoshi K Fusogenic hybrid extracellular vesicles with PD-1 membrane proteins for the cytosolic delivery of cargos. Cancers 14, 2635 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 113.Bui S, Dancourt J & Lavieu G Virus-free method to control and enhance extracellular vesicle cargo loading and delivery. ACS Appl. Bio Mater 6, 1081–1091 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 114.Zhang C & Schekman R Syncytin-mediated open-ended membrane tubular connections facilitate the intercellular transfer of cargos including Cas9 protein. eLife 12, e84391 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 115.Nebogatova J et al. A method for using cell-penetrating peptides for loading plasmid DNA into secreted extracellular vesicles. Biomolecules 13, 1751 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 116.Varela-Eirín M et al. Extracellular vesicles enriched in connexin 43 promote a senescent phenotype in bone and synovial cells contributing to osteoarthritis progression. Cell. Death Dis 13, 681 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 117.Martins-Marques T et al. Cx43-mediated sorting of miRNAs into extracellular vesicles. EMBO Rep. 23, e54312 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 118.Pham TT et al. Endosomal escape of nucleic acids from extracellular vesicles mediates functional therapeutic delivery. Pharmacol. Res 188, 106665 (2023). [DOI] [PubMed] [Google Scholar]
  • 119.Liang X et al. Multimodal engineering of extracellular vesicles for efficient intracellular protein delivery. Preprint at bioRxiv 10.1101/2023.04.30.535834 (2024). [DOI] [Google Scholar]
  • 120.Yim N et al. Exosome engineering for efficient intracellular delivery of soluble proteins using optically reversible protein–protein interaction module. Nat. Commun 7, 12277 (2016) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 121.McConnell RE et al. Transfection reagent artefact likely accounts for some reports of extracellular vesicle function. J. Extracell. Vesicles 11, e12253 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 122.Geng T et al. A simple approach to re-engineering small extracellular vesicles to circumvent endosome entrapment. Int. J. Pharm 626, 122153 (2022). [DOI] [PubMed] [Google Scholar]
  • 123.Banskota S et al. Engineered virus-like particles for efficient in vivo delivery of therapeutic proteins. Cell 185, 250–265.e16 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 124.An M et al. Engineered virus-like particles for transient delivery of prime editor ribonucleoprotein complexes in vivo. Nat. Biotechnol 42, 1526–1537 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 125.Toh WS et al. A report on the International Society for Cell & Gene Therapy 2022 Scientific Signature Series ‘Therapeutic advances with native and engineered human extracellular vesicles’. Cytotherapy 25, 810–814 (2023). [DOI] [PubMed] [Google Scholar]
  • 126.Lener T et al. Applying extracellular vesicles based therapeutics in clinical trials – an ISEV position paper. J. Extracell. Vesicles 4, 30087 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 127.Zeng EZ, Chen I, Chen X & Yuan X Exosomal microRNAs as novel cell-free therapeutics in tissue engineering and regenerative medicine. Biomedicines 10, 2485 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 128.Peng H et al. Intranasal administration of self-oriented nanocarriers based on therapeutic exosomes for synergistic treatment of Parkinson’s disease. ACS Nano 16, 869–884 (2022). [DOI] [PubMed] [Google Scholar]
  • 129.Hsu DH et al. Exosomes as a tumor vaccine: enhancing potency through direct loading of antigenic peptides. J. Immunother 26, 440–450 (2003). [DOI] [PubMed] [Google Scholar]
  • 130.Witwer KW et al. Defining mesenchymal stromal cell (MSC)-derived small extracellular vesicles for therapeutic applications. J. Extracell. Vesicles 8, 1609206 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 131.Chen TS et al. Enabling a robust scalable manufacturing process for therapeutic exosomes through oncogenic immortalization of human ESC-derived MSCs. J. Transl. Med 9, 47 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 132.de Oliveira GP et al. Human red blood cells release microvesicles with distinct sizes and protein composition that alter neutrophil phagocytosis. J. Extracel. Biol 2, e107 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 133.Yoon Y et al. Implantation of the clinical-grade human neural stem cell line, CTX0E03, rescues the behavioral and pathological deficits in the quinolinic acid-lesioned rodent model of Huntington’s disease. Stem Cell 38, 936–947 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 134.Matchett EC & Kornbluth J Extracellular vesicles derived from immortalized human natural killer cell line NK3.3 as a novel therapeutic for multiple myeloma. Front. Immunol 14, 1265101 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 135.Xiong Q, Lu F, Xie X & Zhou W Hypoxia-induced endothelial cell-derived exosome stimulates vascular smooth muscle cell proliferation and migration. Biomed. Res 44, 245–255 (2023). [DOI] [PubMed] [Google Scholar]
  • 136.Zhang B et al. Topical application of mesenchymal stem cell exosomes alleviates the imiquimod induced psoriasis-like inflammation. Int. J. Mol. Sci 22, 720 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 137.Dehghani L et al. Safety of intraparenchymal injection of allogenic placenta mesenchymal stem cells derived exosome in patients undergoing decompressive craniectomy following malignant middle cerebral artery infarct, a pilot randomized clinical trial. Int. J. Prev. Med 13, 7 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 138.Su N et al. Mesenchymal stromal exosome-functionalized scaffolds induce innate and adaptive immunomodulatory responses toward tissue repair. Sci. Adv 7, eabf7207 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 139.Ma W et al. Polydopamine decorated microneedles with Fe-MSC-derived nanovesicles encapsulation for wound healing. Adv. Sci 9, e2103317 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 140.Xu H et al. Bioactive glass-elicited stem cell-derived extracellular vesicles regulate M2 macrophage polarization and angiogenesis to improve tendon regeneration and functional recovery. Biomaterials 294, 121998 (2023). [DOI] [PubMed] [Google Scholar]
  • 141.Hu S et al. Exosome-eluting stents for vascular healing after ischaemic injury. Nat. Biomed. Eng 5, 1174–1188 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 142.Zhang Y et al. Umbilical mesenchymal stem cell-derived exosome-encapsulated hydrogels accelerate bone repair by enhancing angiogenesis. ACS Appl. Mater. Interf 13, 18472–18487 (2021). [DOI] [PubMed] [Google Scholar]
  • 143.Sun M et al. Extracellular vesicles: a new star for gene drug delivery. Int. J. Nanomed 19, 2241 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 144.Guo Q et al. One-minute iodine isotope labeling technology enables noninvasive tracking and quantification of extracellular vesicles in tumor lesions and intact animals. Mol. Pharm 20, 3672–3682 (2023). [DOI] [PubMed] [Google Scholar]
  • 145.Tian T et al. Immune checkpoint inhibition in GBM primed with radiation by engineered extracellular vesicles. ACS Nano 16, 1940–1953 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES