Extracellular vesicles as an efficient nanoplatform for the delivery of therapeutics

ABSTRACT

Extracellular vesicles (EVs) are membrane-derived vesicles that are enriched with RNAs, proteins and other functional molecules. We exploit the unique physical properties of EVs as a promising and advantageous nanoplatform for the delivery of therapeutic drugs and genetic materials. Early successes in the discovery of various disease-related characteristics of EVs have driven a new wave of innovation in developing nanoscale drug-delivery systems (DDSs). Nevertheless, there are several issues that need to be considered during the development of these alternative DDSs, such as standardized isolation and preservation methods, efficient drug encapsulation, mechanisms of drug release and so on. In this mini-review, we summarize the current status and progress of EV-based DDSs as an efficient nanoplatform for therapeutics delivery, followed by a discussion on their challenges and future prospects for clinical translation and applications.

Introduction

Various drug-delivery systems (DDSs) have been extensively explored to improve the efficacy of drugs and reduce their side effects. The development of DDSs involves improving drug solubility,^1-4 activity,^5,6 bioavailability,^4,7-11 targeting¹² and dosing regimen¹³ as well as reducing toxicity.^14,15 Conventional DDSs, including liposomes, polymeric nanoparticles, inorganic nanoparticles, etc.,^16-18 are widely used. Despite considerable progress in developing advanced DDSs, efforts are urgently needed to develop a clinically adequate therapeutic delivery platform.¹⁹ Conventional nanoparticles are multipurpose and have great potential in therapeutic drug delivery applications, but they also have considerable defects (e.g., their xenobiotic origin), which often result in unexpected immune reactions and toxicity in organisms.^20-24 Inherent toxicity and side effects of drug nanocarriers are significant obstacles in the development of a high-performance and clinically safe nano-delivery platforms.^25-27

In light of the obstacles associated with current nano-DDSs, significant efforts have been made to determine revolutionary nanomaterials with biological origins. Very recently, scientists have been attracted by a group of cell-derived endogenous nanovesicles called extracellular vesicles (EVs).²⁸ EVs are membrane vesicles formed from the endosomal system that are released by nearly all types of cells to the extracellular space and thus play an important role in the intercellular communication.²⁹ The capability of EVs to transport molecules between cells indicates that they might serve as a natural DDS.³⁰ In recent years, EVs have been investigated as promising DDSs to target cells or tissues for nanomedicine.^31-33 Compared with conventional DDSs (e.g., liposome), EVs demonstrate attractive advantages and features such as natural biological effects, favorable pharmacokinetics and targeting specificities. However, therapeutic applications of EVs as DDSs are still in early stages of research, and further investigation are expected for their scalable isolation methods, high-efficient encapsulations as well as intrinsic cell targeting properties.

In this review, we summarize the progress in EV-based DDSs with emphasis on the challenges and hurdles in the development of EVs as DDSs. Although EV-based DDSs are not fully optimized for manufacturing scale-up and clinical translation, they provide alternative DDS models for delivering therapeutic drugs.

Biogenesis of EVs

Extracellular vesicles are membrane vesicles released from nearly all cell types in mammalian species, they display versatile physiological functions and are involved in the maintenance of homeostasis, and the regulation of signaling and intercellular communication between different cell types.²⁹ EVs play important roles in many pathological and physiological processes including inflammation,³⁴ angiogenesis,³⁵ immune response,³⁶ autophagy,³⁷ cell survival,^38,39 and cancer drug resistance.^40,41 Based on their morphology, formation pathway and content, EVs can be classified as exosomes, microvesicles (MVs) and apoptotic bodies.^42,43 Exosomes are the smallest membrane-bound vesicles with sizes varying from 40–100 nm,⁴⁴ and were first reported by Johnstone et al. as small vesicles released by reticulocytes.⁴⁵ They are produced from multivesicular bodies (MVBs) during endosomal maturation and are secreted via the fusion of MVBs with the cytomembrane. Exosomes are full of various biomolecules including mRNAs, miRNAs, lipid molecules (cholesterol, sphingomyelin, ceramide, etc.)^46,47 and proteins such as heat-shock proteins, glyceraldehyde-3-phosphate dehydrogenase (GAPDH),⁴⁸ endosomal sorting complex required for transport (ESCRT) machinery^42,43 and tetraspanin family molecules. In contrast, MVs also referred to as ectosomes, shedding vesicles or microparticles, and are directly formed from the cell membrane outward budding; they are more heterogeneous in diameter compared with exosomes (50–1000 nm).^49,50 Besides, microvesicles also contain a large number of biological molecules such as integrins, CD40 ligand, selectins and phosphatidylserine.^51,52 However, apoptotic bodies are significantly different from the other two types of EVs, as they are small vesicles formed in cells that suffered from programmed cell death (namely apoptosis) and exhibit a heterogeneous range of sizes and different morphologies (50–5000 nm).^53,54

In the intercellular environment, all three types of cellular vesicles—designated “early endosomes,” “late endosomes” and “recycling endosomes”—are successively formed during endosomal maturation. First, incoming cargos are generated from the internalized plasma membrane and then sorted into diverse intracellular destinations by early endosomes.⁵⁵ When early endosomes transform into late endosomes, intraluminal vesicles (ILVs) are formed. These late endosomes contain ILVs called MVBs. Subsequently, some MVBs fuse with lysosomes by degrading the cargos. Moreover, other MVBs fuse with the plasma membrane, resulting in the formation of exosomes in the extracellular space. Several studies have demonstrated that tetraspanins and endosomal-sorting complexes are essential to the formation of the intraluminal vesicles in the cell, and this special superfamily of membrane proteins noted mentioned is often used as exosome biomarkers.⁵⁶

Endosomal sorting complexes required for transport (ESCRT) work together with accessory proteins relating to the formation of ILVs, and exosome biogenesis relies on both ESCRT-dependent and ESCRT-independent mechanisms.⁵⁷ Recently, Colombo et al. analyzed the function of ESCRT components in EVs biogenesis using RNA interference. It was found that various ESCRT components such as vacuolar protein sorting-associated protein 4B and tumor susceptibility gene 101 were related to the composition, size and productivity of secreted EVs.⁵⁸ The existence of ESCRT-independent mechanisms have been corroborated by Stuffers et al., they demonstrated that CD63-positive EVs were secreted from cells that lack of four subunits of the ESCRT complex.⁵⁹

Exosome biogenesis occurs in MVBs, and MVs are formed by direct budding from the plasma membrane (Fig. 1). MVs are larger than exosomes and more heterogeneous in size and morphology.⁴³ The activation of MVs varies from cell to cell. For example, MVs are released by endothelial cells and circulating blood cells in response to complement, whereas monocytes, platelets and fibroblasts budding are released in response to bacterial cell wall components, thrombin, and stress relaxation, respectively.^60,61 As proteins express procedurally, the production of MVs that occurs throughout the cell cycle and under various culture conditions may not be consistent.⁴¹ MVs could reflect the antigenic content of the cells that they originate, providing a new strategy for natural vaccine delivery systems. However, exosomes are known to relate to intercellular communications and offer distinct advantages as highly effective drug carriers for future clinical translation.⁶¹

Figure 1.

Illustration of extracellular vesicles secreted from cells. Exosomes are formed from multivesicular bodies (MVBs) during endosomal maturation and are secreted by fusion of MVBs with the cytomembrane. Microvesicles (MVs) are directly releasing from the cell membrane outward budding.

Extracellular vesicles as novel DDSs

The structural characteristics of EVs are analogous to liposomes, which renders EVs attractive for drug delivery.⁶² Since liposomes are constructed of phospholipids, they are similar to the plasma membranes, and have been widely used for efficient drug delivery.⁶³ Thus far, several commercialized liposome-based products such as DaunoXome (a liposomal for the delivery of daunorubicin (DNR), approved for the management of advanced HIV-associated Kaposi's sarcoma), Myocet (a non-pegylated liposomal doxorubicin, approved for treatment of metastatic breast cancer) and Depocyt (a cytarabine liposome injection, approved for treatment of lymphomatous meningitis) have been put into the market.⁶⁴ Liposome research lays the foundation for investigations of the physicochemical characteristics, stability and drug loading of EVs.^65-67 Moreover, compared with liposomes, EVs are produced by the cells themselves, which make them more advantageous than liposomes in mimicking the cell membrane. These superior properties of EVs indicates the possibility of utilizing EVs from the body's own cells to deliver drugs, even across blood-brain barrier (BBB).⁶⁸

Extracellular vesicles for cancer treatments

More and more evidence has shown that EVs have splendid prospects in therapeutic delivery of small interfering RNAs and synthetic molecules.⁶⁹ Several studies have revealed the potential of EVs as therapeutic DDSs in various animal models of diseases.⁷⁰ Furthermore, EVs are being widely studied as anti-tumor DDSs due to their passive targeting ability to tumor tissues via enhanced permeation and retention effect.⁷¹ More importantly, EVs could be genetically engineered as targeted DDSs, offering a versatile platform for delivering drugs to specific targets with significantly promoted improved therapeutic effects. For example, Alvarez-Erviti et al. developed a DDS to deliver siRNA to the central nervous system using modified EVs from self-derived dendritic cells (DCs).⁶⁸ The DCs were isolated from mice and transfected with a plasmid encoding an exosomal membrane protein lysosome-associated membrane glycoprotein 2b (Lamp2b), genetically fused to a rabies viral glycoprotein (RVG), a peptide that binds to the acetylcholine receptor. The DC-derived EVs loaded with GAPDH siRNA exhibited specific brain-targeting gene knockdown, demonstrating the potential of EVs to act as targeted DDSs. As expected, both proteins and genes could be effectively delivered by EVs, which serve as cell-derived liposome-like nanoplatforms for the treatment of diseases such as cancer.^42-46 Interestingly, Bolukbasi et al. demonstrated that a zipcode-like 25-nt sequence promoted package miRNAs into EVs, demonstrating great potential for high-yielding EVs loaded with various RNAs.⁷⁷ Moreover, Gujrati et al. employed bacterial outer membrane vesicles (OMVs) to deliver siRNA for anticancer treatments. These results revealed that bacteria are a potential producer of biological nanovesicles for drug delivery.⁷⁸

In addition to delivering biomolecule-based drugs, EVs have also been exploited to deliver chemotherapeutic agents, with the goal of increasing their efficacy and reducing side effects. For example, EVs encapsulated with doxorubicin and curcumin both effectively inhibit the progression and deterioration of colon and mammary cancer.^79,80 These research results indicate that EVs have the capability to availably deliver chemotherapeutics drugs to suppress malignant tumors. Tian et al. engineered immature mouse DCs (imDCs) to express Lamp2b-iRGD peptide and used exosomes derived from these cells to deliver the chemotherapeutic drug doxorubicin to αv integrin-positive breast cancer cells in nude mice after i.v. injection.⁷⁹ These authors found that therapeutic exosomes caused less cardiac damage and more effectively inhibited tumor growth.Moreover, excellent delivery effects of therapeutic chemotherapeutic drugs in EVs have been validated with a variety of tumor models, including hepatocarcinoma⁸¹, prostate cancer⁸², lymphocytic leukemia⁸³ and pancreatic tumors.⁸⁴ Table 1 summarizes some recent studies using EVs as therapeutic delivery tools for cancer treatment.

Table 1.

Application of EVs as therapeutic delivery tools for cancer treatments.

Type of EVs	Therapeutic cargo	EV source	target tissue/cell	Outcome	Reference
MVs	mRNA and/or protein	HEK293T	schwannoma tumor	Effectively inhibit schwannoma tumor growth in vitro and in vivo	Mizrak72
exosomes	let-7a miRNA	HEK293T	breast tumors	Tumor been restrained observably	Ohno73
MVs	transforming growth factor β1(TGF-β1) siRNA	mouse fibroblast L929 cell	murine sarcomas	Significantly inhibited TGF-β1 expression and suppressed primary tumor growth	Zhang74
exosomes	miR-146b	MSC	glioma	Significantly reduced glioma xenograft growth in a rat model of primary brain tumor	Katakowski75
exosomes	miR-9	MSC	glioblastoma multiforme	Showed a potential role for MSCs in the functional delivery of synthetic anti-miR-9 to reverse the chemoresistance of GBM cells	Munoz76
exosomes	doxorubicin	immature mouse dendritic cells	breast cancer	Caused less cardiac damage, and more effectively in inhibition of tumor growth	Tian79
MVs	methotrexate (MTX) and cisplatin	H22 hepatoma cells	hepatocarcinoma	Inhibited the growth of subcutaneous hepatocarcinoma	Tang81
exosomes	poly and cyclophosphamide	DCs	L1210 tumour	Have a great capacity to resist tumor growth, increase survival time of mouse and stimulates DCs maturation	Guo83
MVs	paclitaxel	MSCs	pancreatic tumors	Effectively suppressed pancreatic tumors	Pascucci84
exosomes	tumor antigens	TS/A cells	breast cancer	induce potent CD8+ T-cell-dependent antitumor effects	Wolfers85
exosomes	CagA	CagA- expressing cells	gastric epithelial cells	Delivering the CagA to gastric epithelial cells	Shimoda86
exosomes	Survivin	melanoma	pancreatic carcinoma cells	Induced a significant increase in apoptotic cell death	Aspe87

Extracellular vesicles for other diseases

Recent studies have suggested that EVs are related to cardiovascular diseases and that EV levels of blood circulates may be associated with the onset and progression of disease severity.^88-91 Nevertheless, EVs also exhibit a cardioprotective effect.⁹² Chen et al. demonstrated protective benefits on the myocardium from serious ischemia-reperfusion injury by treating mice with EVs originating from cardiac progenitor cells.⁹³ Generally, sonic hedgehog (SHH) signaling is critical for neovascularization and angiogenesis, and SHH may be a therapeutic target in the vascular repair process. Fleury et al. demonstrated that SHH enriched EVs originating from T-lymphocytes and that SSH signaling could correct Ang II-induced hypertension and endothelial dysfunction in mouse models.^94,96

Cerebral inflammation is the defense reaction of an organism to brain injury. Inflammation is a pathological process of damage and resistance to damage. Macrophages play a significant role in the inflammatory response, producing a variety of cytokines and inflammatory mediators that are the internal mechanism of the inflammatory response and its development. A reasonable strategy to relief the disease is inhibiting the inflammatory factors and decreasing the number of macrophages, which would accordingly inhibit the inflammatory response. Nevertheless, the utilization of traditional medical remedies is restricted by the BBB. Zhuang et al. demonstrated that intranasal administration of curcumin-containing EVs efficaciously delivered curcumin to the brain. These results revealed that curcumin-containing EVs apparently suppressed brain diseases such as experimental autoimmune encephalitis (EAE), LPS-induced brain inflammation and brain tumors, providing a noninvasive treatment option for inflammatory brain disease.⁹⁷ In another study, Kalani et al. showed that curcumin-primed EVs possess the ability to alleviate endothelial dysfunction in a hyperhomocysteinemia mouse model.⁹⁸ Similarly, Sun et al. studied EVs as DDSs and found that EVs ensured the stability and concentration of curcumin during delivery, which was relevant to better bioavailability of drugs. The authors administered curcumin-loaded EVs to a mouse model of LPS-induced septic shock, and the results revealed that EVs significantly suppressed inflammation.⁹⁹ Recently, anti-inflammatory EVs derived from gene-modified DCs have attracted the attention of many researchers. Kim et al. expressed the IL-10 gene in DCs. These authors injected purified EVs secreted from the DCs into mouse models of collagen-induced arthritis. They detected inhibited inflammation progression. It has also been noted that EVs delivered either locally or systemically efficaciously downregulate the immunological reaction via the MHC II-dependent pathway. ¹⁰⁰

Extracellular vesicle-based vaccines are potential candidates for infectious disease vaccines. Toxoplasmosis, caused by toxoplasma gondii, is widespread in humans and warm-blooded animals. Recently, Aline et al. demonstrated that EVs originating from DCs loaded with toxoplasma gondii antigens induced a potent protective immune reaction that inhibited toxoplasma gondii infection.¹⁰¹ Severe acute respiratory syndrome (SARS), also called atypical pneumonia, is characterized by progressive respiratory failure and death in approximately 10% of cases. ¹⁰² In 2007, Kuate et al. showed that EVs originating from HEK293T expressed SARS proteins and induced high levels of SARS-specific neutralizing antibodies titers in mice.¹⁰³ Moreover, after vaccinated patients with EV vaccines and adenoviral vector boosters, the serum-neutralizing antibody titer were even higher than that of recovering SARS patients.

Altogether, EVs are promising candidate drug carriers for the delivery of therapeutics (Table 1). Specifically, EVs can load with different drugs and maintain their nature biological properties during the encapsulation process for personalized medicine. The researchers mentioned above highlight the wider potential of EVs, beyond biomolecule-based drugs delivery, for the transfer of various other chemotherapeutic cargoes. In addition, EVs are superior biocompatibility with the lowest cytotoxicity, and more compatible with the host immune system than other nano-carriers.

Clinical trials: Progress and challenges

Various EVs have been investigated as therapeutic agents in clinical trials based on their superior performances in preclinical studies (Table 2).^104,105 Dendritic cell-derived EVs as anti-tumor therapies have been widely investigated in preclinical and clinical trials, and two trials entered phase I clinical trails. One trial focused on melanoma and the other trial focused on non-small cell lung cancer.^106,107 In addition, a phase II clinical trial was performed to treat non-small-cell lung carcinoma patients.¹⁰⁸ Feasibility and safety were demonstrated and the activation of natural killer (NK) cells was still detectable in later therapy. Following the same method, DC-derived EVs might be effective for treating other diseases such as angiocardiopathy, inflammation, nervous system diseases and infectious diseases. In a phase I clinical trial, EVs from ascites fluid were used for immunotherapy against colorectal cancer.¹⁰⁹ Preclinical EV-based antitumor studies have indicated that the therapeutic strategies are promising.

Table 2.

Therapeutic applications of EVs in clinical trials.

Country	EV source	Disease	Drug	phase
USA	dendritic cells	non-small cell lung cancer	MAGE peptides	I105
	glioma	malignant glioma	EVs + AS-ODN	I125
	fruit	colon cancer	curcumin	I114
	fruit	mucositis	curcumin	I115
France	dendritic cells	Metastatic melanoma	melanoma peptide antigens	I106
	dendritic cells	non-small cell lung cancer	IFN-γ, MAGE peptides	II108
UK	FetA modified strain 44/76	neningitis	vaccine	I113
	B:4:P1.7–2,4 strains	meningitis	rMenB vaccine, NadA/fHBP/NHBA	II112
China	ascites fluid	colorectal cancer	EVs, EVs + GM-CSF	I109
Egypt	umbilical cord-blood derived MSC	type I diabetes mellitus	EVs	I126
Germany	MSC	GVHD	EVs	I116
Norway	B:4:P1.7–2,4 strains	meningitis	vaccine	I110

IFN-γ: interferon gamma; NadA: Neisserial adhesin A; fHBP: factor H binding protein; NHBA: Neisserial heparin binding antigen; GM-CSF: granulocyte-macrophage colony-stimulating factor; MSC: Mesenchymal stem cell; GVHD: Graft-versus-host disease.

Outer membrane vesicles have also been studied widely as anti-bacterial vaccines due to they are secreted naturally by bacteria.^110,111 Outer membrane vesicle-based vaccines significantly suppressed bacterial infection in a phase I trial.^110,112 A phase II trial revealed that OMV combined with a recombinant vaccine displayed a better immunogenicity than recombinant vaccine alone.¹¹² Another phase I trial demonstrated that OMV vaccines from a FetA modified strain conferred protective effect against N. meningitides.¹¹³ The phase I clinical trial results showed an excellent immune response and mild side effects induced by OMV-based vaccines. These findings indicate that further exploration is essential before using OMVs as vaccines or DDSs.

Plant-based exosomes (plexosomes) used for DDSs are characterized by more advantages than conventional DDSs due to their being minimally toxic and low in immunogenicity. They are additionally unlimited in source. In two phase I clinical trials, plexosomes were loaded with curcumin for treating mucositis and colon cancers, respectively.^114,115 Results showed the possibility to avoided contact between chemoradiation and oral mucosa in the treatment of head and neck cancer. Inspired by the fact that mesenchymal stem cells (MSCs) possess an immunological suppression ability against specific immune and nonspecific immune responses, a phase I clinical trial has been conducted to explore the inhibitive effect of MSC-based EVs for graft-versus-host disease (GVHD).¹¹⁶

However, it should be noted that many issues still need to be addressed to bridge laboratory EVs experimentation with practical clinical settings. For instance, it is very important to select a suitable producer cell type. Besides, the consistency of the quality and quantity of EV production upon scale-up is required for developing an EV-based product, some mammalian primary cells have been widely investigated but are not suitable for large-scale production due to their low EV yield.¹¹⁷ Therefore, it is urgent to find other alternative EV sources. Isolation techniques are key issues that need to be improved, along with the lack of reliable isolation methods, which hinder the translation of EVs into clinical applications.¹¹⁸ At the present, ultracentrifugation is the most commonly used method to isolate EVs, but it is difficult to avoid undesirable co-isolation of contaminants.¹¹⁹ Not only the scheme but the technology requires a breakthrough in non-destructive, contamination-free isolation methods that are characterized by a short processing time. In addition, the optimization of storage conditions of EVs is also of great importance. Such work involves the selection of isotonic buffers, storage temperatures and container materials. However, there are no standard EV storage conditions thus far. The methods for characterization of EVs have much room for improvement, and conventional methods—including flow cytometry, fluorescence microscopy, nanoparticle tracking analysis and transmission electron microscopy—have many limitations.¹²⁰ Techniques need to be developed to define the degree of heterogeneity of scaled-up EV preparations and the acceptable limits on this variability that does not compromise the safety, efficacy and stability of the product.  Deeper fundamental research about the biological and pharmacological functions of EVs is also essential. Dose, immunization route, immune response, cytotoxicity and tumorigenic effects all require intensive study for therapeutic applications. Another major problem for EV translational applications is the lack of an existing legislation for regulating EV-based therapies; it necessary that both adequate infrastructure and quality management systems be developed. Before using EVs in clinics, it is necessary to standardize their isolation, storage and characterization and establish criteria for a quality-control system.

One possible solution is EV-mimetic nanovesicles. Lunavat et al. have generated EV-mimetic vesicles via serial extrusions of cells through filters, and this production method resulted in a 100-fold increase in the yield of naturally produced extracellular vesicles.¹²¹ In addition, EV-mimetic nanovesicles loaded with siRNA could effectively inhibit the expression of targeted genes. A similar method to generate EV-mimetic nanovesicles was applied by Jang et al.⁸⁰ Drug-loaded EV-mimetic nanovesicles were produced by serial extrusion in the presence of doxorubicin. The results revealed that nanovesicles were effectively accumulated within tumor-inhibiting cancer cell growth after intravenous injection in mice, and no side effects were observed. These studies demonstrated that EV-mimetic nanovesicles have a capacity for RNA and chemotherapeutics delivery. Viral antigen-loaded EV-mimetic nanovesicles (i.e., virus-mimetic vesicles), have been recently developed in a bioinspired manner for vaccines by maintaining the natural conformation of epitopes.¹²² Viral antigen-loaded EV-mimetic nanovesicles can resemble natural viruses in morphology and immunogenicity, and may result in a high level of antibody titers in response to the corresponding antigen.^122-124 Such EV-mimetic nanovesicles have significant potential to deliver therapeutics with specific ligands on the surface for targeted drug delivery and therapy.

Conclusions and future perspectives

Medical experts are constantly seeking to develop novel DDSs and improve the targeting and bioavailability of drugs. These professionals are also concerned with reducing drug toxicity and improving therapeutic efficacy. Advantages such as a strong packing capacity, a long half-life time, minimal undesirable immunogenicity and limited side effect are requirements for a perfect drug-delivery platform. In the past decade, thanks to persistent endeavors by researchers, EVs have been exploited as ideal drug-delivery platforms and have been widely used in preclinical studies and clinical trials to efficiently transport biological substances and chemotherapeutics to desired targets. The outstanding advantages of EVs in drug transmission as a DDS lie primarily in their biological origin, which is associated with better biocompatibility with organism tissue. In addition, many studies have shown that EVs as drug carriers possess excellent performance at improving drug stability, prolonging blood circulation time, reducing toxicity to healthy tissues and increasing tumor targeting and tumor inhibition. Utilizing autologous EVs for personalized nanomedicines is a promising therapy.

Although EVs play an important role and have exhibited promise in preclinical studies, there are several issues that need to be considered. These challenges include: the detailed mechanisms of the formation and release of drug remain unarticulated, distinct criteria for classification and nomenclature have not been established, and suitable and standardized isolation, separation, refinement and preservation methods are still urgently needed. In addition, regulative biodistribution to accumulate EVs at desired sites still necessitates further research. Of course, challenges also include production scale-up and characterization of the purified EV product, and the toxicological and ADME profile of the EV product also needs to be defined. Furthermore, EVs from aberrant cells or pathogens may carry tumorigenic and pathogenic potential, this concern may be laid to rest by selection of an appropriate benign cell type as source of EVs. In summary, EVs-based DDSs open up new avenues for the treatment of various diseases. However, it is imperative to fully understand the basic principles involved in the EVs. The future of EV-based DDSs mainly depends on cooperation among biologists, physicists, nanomaterial scientists and clinical specialists. With continuous efforts by multidisciplinary strategies, the use of such nanoplatforms will shed new light on the delivery of therapeutics to cancers and various other diseases.

Abbreviations

ADME

absorption, distribution, metabolism, excretion

BBB

blood-brain barrier

DDSs

drug-delivery systems

EAE

experimental autoimmune encephalitis

ESCRT

endosomal sorting complex required for transport

EVs

extracellular vesicles

fHBP

factor H binding protein

GAPDH

glyceraldehyde-3-phosphate dehydrogenase

GM-CSF

granulocyte-macrophage colony-stimulating factor

GVHD

graft-versus-host disease

IFN-γ

interferon gamma

ILVs

intraluminal vesicles

imDCs

immature mouse DCs

Lamp2b

lysosome-associated membrane glycoprotein 2b

MHC

major histocompatibility complex

MSCs

mesenchymal stem cells

MVBs

multivesicular bodies

MVs

microvesicles

NadA

neisserial adhesin A

NHBA

neisserial heparin binding antigen

NK

natural killer (NK) cells

OMVs

outer membrane vesicles

plexosomes

plant-based exosomes

RVG

rabies viral glycoprotein

SARS

severe acute respiratory syndrome

SHH

sonic hedgehog

Disclosure of potential conflicts of interest

All authors declare that they have no conflict of interest in relation to this work.

Funding

This study was supported by the Major State Basic Research Development Program of China (973 Program) (Grant Nos. 2017YFA0205201, 2014CB744503, and 2013CB733802), the National Natural Science Foundation of China (NSFC) (Grant Nos. 81422023, 51273165, and U1505221), the Fundamental Research Funds for the Central Universities, China (Grant No. 20720160065 and 20720150141), the International Postdoctoral Exchange Fellowship Program, and the Program for New Century Excellent Talents in University (NCET-13–0502).