Extracellular vesicle‐coated nanoparticles

Synthetic nanoparticles have been used for a variety of theranostic applications to aid in the betterment of human health. These nanoparticles can provide platforms for targeted imaging and therapy of diseased tissues. The development of surface coatings for nanoparticles has enabled their selective uptake in tissues of interest, and the use of membrane‐derived nanoparticle coatings provides a particularly promising approach for the regulation of nanoparticle‐tissue interactions. Membranous extracellular vesicles (EVs) secreted by cells have been known to contain antigens, proteins, and other cell components on their surface that facilitate their uptake in target cells, enabling the transport of information from one cell to another. EV‐based nanoparticle coatings allow for the expansion of nanoparticle targeting from typical approaches that target individual antigens, to an approach that can simultaneously target many antigens for more efficient uptake within target cells. EV‐derived coatings also possess immune evasive properties that can lead to increased circulation time. In this mini‐review, we describe the various approaches and applications for EV coating of nanoparticles, a majority of which focus on cancer applications. We also provide an overview of commonly used EV sources for nanoparticle coating applications.

from one cell to another. In recent years, there has been growing interest in the use of EVs for diagnostic and therapeutic applications. A number of other review papers have described applications of exosomes and EVs in drug delivery, immunotherapy, and other biomedical applications, with a majority focusing on applications in cancer. [8][9][10][11][12][13][14][15][16][17][18][19] The presence of specific antigens from the source cells on the surfaces of EVs has typically been found to allow for specific uptake of EVs within that cell type. For example, cancer cell-derived exosomes have commonly been used to encapsulate drugs, thus increasing the delivery efficiency of these therapeutic moieties to cancer tumors.
At the same time, synthetic nanoparticles have long served as therapeutic and imaging agents, enabling targeted imaging and treatment of diseases such as cancer and atherosclerosis. [20][21][22] A number of publications describe the use of cell membrane-derived coatings to enable nanoparticle targeting. [23][24][25][26][27][28][29][30][31] The use of EV-derived coatings to facilitate nanoparticle uptake is a logical progression of membrane-derived coatings, and takes advantage of EV surface antigens for uptake within cells of interest. The goal of this mini-review is to explore examples of EV-coated nanoparticles and provide insight into their use. We begin by reviewing common EV sources and types of synthetic nanoparticles upon which EV coatings are generated. We then describe the major approaches to generate EV coatings on nanoparticle surfaces. Afterwards, we describe the many approaches that have taken advantage of EV-coated nanoparticles for cancer therapy. Lastly, we describe the applications of EV-coated nanoparticles for diseases aside from cancer, and provide a perspective on the future direction of research on EV-coated nanoparticles. An overview of common EV sources, coating methods, and applications are provided in Figure 1.

EXTRACELLULAR VESICLES AS VERSATILE COATINGS FOR NANOPARTICLES
EVs have emerged as versatile coatings for synthetic nanoparticles, enabling selective uptake in tissues of interest. In this section, we explore EV and nanoparticle compositions, coating methods, and applications.

Common EV sources and types of synthetic nanoparticles
Cancer cells and mesenchymal stem cells (MSCs) are the most commonly utilized exosome sources for nanoparticle coatings. Beyond these, EV-based nanoparticle coatings have also been derived from human umbilical vein endothelial cells (HUVECs), macrophages, and bacterial cells. To date, a majority of the nanoparticles coated with exosomes have been metallic in nature, including gold nanoparticles, iron oxide nanoparticles, (IONs) and gold-IONs (GIONs). However, there have been a few examples of poly(lactic-co-glycolic acid) (PLGA) nanoparticles and metal organic framework (MOF) nanoparticles coated with exosomes.

Coating methods
The coating methods for EVs on nanoparticle surfaces are similar to those of cell membranes on nanoparticle surfaces. 28 Four major approaches are used for the encapsulation of synthetic nanoparticles in EVs. The most common approach is the direct incubation of cells with medium containing the synthetic nanoparticles. The nanoparticles are then internalized by the cells and secreted within exosomes, taking advantage of the exosomal biogenesis pathway. The three remaining approaches involve the physical mixing of pre-collected exosomes with the synthetic nanoparticles. In one approach, the direct mixing of exosomes with synthetic nanoparticles leads to their encapsulation with time. In another approach, the nanoparticle-exosome mixture is sonicated, leading to the formation of an exosome coating on the nanoparticle surface. This approach has been especially popular in the coating of soft nanoparticles with exosomes. In the final approach, the nanoparticle-exosome mixture is extruded through porous membranes, leading to the formation of exosome-coated nanoparticles. We previously used electroporation to enhance the entry of nanoparticles into cell membranes. 32 Despite the years of work in this field, the ideal coating method remains a topic of debate, with some claiming that the use of the exosomal biogenesis pathway leads to low encapsulation efficiency. 33 In contrast, Sancho-Albero et al conducted a comprehensive evaluation of B16-F10 murine melanoma exosome coatings on hollow gold nanoparticles, finding that pre-incubation of the cells with the nanoparticles led to a higher loading efficiency than electroporation, passive loading by diffusion, thermal shock, sonication, or saponin-assisted loading. 34 However, the advantages and disadvantages of each coating method may depend on the specific nanoparticle type and intended application. Piffoux et al found that the methods by which iron oxide nanoparticles were loaded into HUVEC EVs had an effect on EV nanoparticle load, photosensitizer load, size, yield, purity, and phototoxicity. 35 For example, spontaneously-released EVs had higher yield but lower purity than EVs released under starvation. Thus, although a variety of reported methods exist for coating nanoparticles with EVs, the factors that make one method more favorable than another method F I G U R E 1 Overview of common EV sources, coating methods, and applications of exosome-coated nanoparticles are not standardized from one study to another. In order for any particular method to be proven as the most favorable, the same criteria (e.g., yield, purity, loading efficiency, and some other factors) must be used to evaluate multiple coating methods using different types of nanoparticles, EV sources, and coating methods. of cells with PEGylated HGNs took advantage of exosomal biogenesis and was the most efficient encapsulation method, resulting in 50% of exosomes being loaded with HGNs. 34 They then demonstrated that exosomeencapsulated HGNs were taken up by B16-F10 cells and could be used for photothermal therapy. Sancho-Albero et al have also demonstrated the assembly of palladium nanosheets directly within cancer cell exosomes, using them to deliver catalytic cargo directly to cancer cells. 38 Since breast cancer tumors often metastasize to the lungs, Roma-Rodrigues et al used exosome-coated nanoparticles to regulate the communication between breast cancer cells and bronchial-tracheal epithelial (BTE) cells. 39 MCF-7 breast cancer cells were incubated with gold nanoparticles functionalized with anti-RAB27A. This modulated the expression of RAB27A and decreased exosome secretion from the MCF-7 cells. When the exosomes derived from cells treated with gold nanoparticles functionalized with anti-RAB-27A were then exposed to BTE cells, they led to lower expression of C-Myc, an oncogene that is overexpressed in breast cancer. These findings indicate that exosome-coated nanoparticles can be transported from one cell type to another, and can be used not only for direct delivery of signals to cancer cells, but also potentially to prevent the spread of tumor cells to other organs.

Cancer-derived EVs
Other reports have expanded these investigations to in vivo use. Cheng et al utilized MOF NPs, loading them with proteins and then self-assembling the membranes from MDA-MB-231-derived EVs onto the NP surfaces. 40 They demonstrated that EV coatings reduced particle uptake by macrophages, and led to greater uptake by MDA-MB-231 cells compared to uptake by 293T human embryonic kidney cells, 3T3 mouse embryo fibroblasts, CAD mouse central nervous system-derived cells, MCF-7 human breast adenocarcinoma cells, and SH-SY5Y human neuroblastoma cells. After intravenous administration, coated MOF NPs had higher accumulation in orthotopic MDA-MB-231 xenograft tumors in mice than non-coated MOF NPs, and led to greater transduction of the loaded proteins. The therapeutic efficacy of loaded gelonin was also increased as a result of NP coating. In another approach that utilized breast cancer-derived exosomes to treat breast cancer, Bose et al sought to target miR-21, which is overexpressed in cancer cells and can lead to chemotherapy resistance. 41 They generated Cy5-anti-miR-21 packed EVs by exposing 4T1 cells to Cy5-anti-miR-21 in situ and then collecting the EVs. The Cy5-anti-miR-21 packed EVs were found to reduce resistance to doxorubicin. They then coated GIONs with the Cy5-anti-miR-21 packed EVs to allow for magnetic resonance (MR) contrast and photothermal properties. In vivo experiments in a syngeneic subcutaneous 4T1 tumor-bearing mouse model demonstrated that the particles accumulated at the tumor site, delivered anti-miR-21, and led to inhibited tumor growth and increased doxorubicin therapeutic efficacy.
Liu et al developed a microfluidic device that uses ultrasonication to coat NPs with membranes from EVs, membranes from cancer cells, or lipid membranes. 42 They isolated EVs and cell membranes from A549 human lung carcinoma cells, and coated PLGA nanoparticles with the membranes using the microfluidic ultrasonication method. When administered in vivo in A549 tumor-bearing mice, EV-coated PLGA NPs had higher circulation times than those coated with membranes from cancer cells, and were found to evade immune cells, thus leading to better delivery to tumor cells. The specific in vivo uptake of EV-coated PLGA NPs was also demonstrated by comparing their uptake in A549 tumors with their uptake in MDA-MB-231 tumors. Han et al expanded this work to use an aptamer coating layer on top of MDA-MB-231 EVcoated PLGA NPs, leading to enhanced tumor targeting. 43 In contrast to reports that emphasize the specific uptake of EV-coated nanoparticles within one target cell type, Yong et al demonstrated cross-reactivity between exosomes from different cancer types. 44 They loaded porous silica nanoparticles with doxorubicin (Dox-PSiNPs), and then coated the Dox-PSiNPs with exosomes from H22, Bel7402, and B16-F10 cancer cell lines by incubation with the cells and subsequent exocytosis (exosomal biogenesis pathway). They demonstrated that exosome coatings from tumor cells increased the nanoparticle uptake in cells, and that the coated particles were also taken up by cancer stem cells isolated from the cell lines. However, the exosome coatings were found to be cross-reactive between different cell types, with high uptake of H22 exosome-coated Dox-PSiNPs in B16-F10 cells and vice versa. This crossreactivity was further utilized in vivo, where H22-derived Dox-PSiNPs injected in 4T1 tumor-bearing mice had strong anticancer properties and led to decreased tumor volume and a reduction in cancer stem cells. Additionally, they demonstrated that particles coated with exosomes from HUVECs had less uptake in the H22 cells, and vice versa, which indicates a preferential uptake of cancer-derived exosomes in cancer cell lines.

Other types of EVs
In addition to using cancer cell-derived exosomes for cancer applications, other approaches have utilized exosomes derived from tumor-homing non-cancerous cells for cancer applications. Since MSCs have been known to have tumor-homing properties, Altanerova et al used exosomes from MSCs that expressed the yeast cytosine deaminase::uracil phosphoribosyl transferase suicide fusion gene in the presence of the prodrug 5fluorocytosine (5-FC). 45 Incubation of the MSCs with venofer IONs led to the secretion of exosomes containing iron oxide, which were then used to cause the death of PC3 and HeLa cells via alternating magnetic field-induced intracellular hyperthermia or by exposure to the 5-FC prodrug. Macrophages are another class of cells known to be recruited by tumors, and Xiong et al utilized RAW 264.7 macrophage-secreted exosomes to coat nanoparticles that were composed of a laurate-functionalized Pt(IV) prodrug, human serum albumin, and lecithin. 46 This led to prolonged blood circulation of the nanoparticles, enabled their uptake at the orthotopic 4T1 tumor site and metastatic breast cancer lung nodules, and allowed for administration of platinum-based chemotherapy. Piffoux et al incubated iron oxide nanoparticles with HUVECs, producing nanoparticle-loaded exosomes by starving the cells. 35 The coated nanoparticles were then used for photodynamic therapy of PC3 prostate cancer cells. They also evaluated the effects of exosome production method on the trade-off between EV nanoparticle load, photosensitizer load, size, yield, purity, and phototoxicity.

Other applications
Another application of EV-coated nanoparticles is neuroimaging. Betzer et al encapsulated glucose-coated GNPs with human MSC exosomes through direct incubation of the nanoparticles with the exosomes. 33 They demonstrated that nanoparticle size, incubation temperature, and the presence of a glucose coating on the nanoparticle surface played a role in successful encapsulation within exosomes. The gold nanoparticles provided CT contrast, allowing for visualization of exosome-encapsulated nanoparticle accumulation at the stroke region in C57BL/6 mice brains after acute striatal stroke. The colocalization of nanoparticles with exosomes was also confirmed using ex vivo fluorescence imaging of mice treated with double-labeled exosomes (exosome +GNPs+PKH26 dye). Despite these promising results, Qu et al raised concerns about the nanoparticle encapsulation method within the exosomes, proposing that the centrifugation step would not allow for separation of GNPs from exosome-coated GNPs. 47 In another neuroimaging application, Khongkow et al developed AuNPs coated with exosomes from transformed human embryonic kidney cells (HEK293T) transfected to express RVG-Lamp2b fusion proteins and target neuronal cells. 48 They demonstrated that these coated AuNPs adhered to brain cells under laminar flow. The transport of particles across the blood-brain-barrier was demonstrated in an in vitro model, as well as through in vivo bioluminescence imaging of mouse brains after intravenous administration of the particles. These studies demonstrated that RVG-targeted exosomes had higher accumulation in the brain than those not targeted with RVG, further cementing the role of cell and exosome surface proteins on the utility of secreted exosomes in nanoparticle targeting applications. In a unique approach, Gao et al coated PLGA nanoparticles with EVs from S. Aureus bacteria. 49 They found that the bacterial EV-coated nanoparticles had preferential uptake in macrophages exposed to that particular type of bacteria. Thus, PLGA nanoparticles coated with EVs from S. aureus had preferential uptake in macrophages exposed to S. aureus, while those coated with EVs from E. coli had preferential uptake in macrophages exposed to E. Coli. PLGA nanoparticles coated with EVs from S. Aureus had high uptake in vivo in tissues that had S. aureus infections, enabling the delivery of antibiotics to the site of infections.
Sancho-Albero et al further demonstrated that coating nanoparticles with exosomes from MSCs led to preferential uptake by MSCs compared to uptake in B16F1 cells, B16-F10 cells, and monocytes. 50 A summary of various EVcoated nanoparticles is provided in Table 1.

CONCLUSIONS
EV-coated nanoparticles represent a relatively new approach for targeted delivery of therapeutic and imaging agents to tissues of interest. In recent years, the field of EV-coated nanoparticles has greatly expanded, and many different coating approaches and nanoparticle types have been used for applications spanning cancer therapy, neuroimaging, and treatment of bacterial infections. Despite this progress, further studies must be conducted to clarify key areas of ambiguity within the field. The uncertainty over the best method for nanoparticle encapsulation within EVs remains a major obstacle to the widespread adoption of EV coating as a simple targeting approach for nanoparticles. The general disagreement within the field suggests that the ideal encapsulation method may vary based on nanoparticle type and cell type. Thus, extensive studies that evaluate multiple nanoparticle types (including "soft" and "hard" nanoparticles) and a variety of cell lines can be undertaken in order to provide a clear understanding of the parameters that make each type of encapsulation method ideal for some scenarios. Furthermore, the specific uptake of EV-coated nanoparticles within the cell line from which the EVs were derived must be further explored. Reports of cross-targeting between different cell types appear to be in conflict with some reports of specific uptake within the cells from which the EVs were derived. It is clear that the characteristics of each cell type govern the behavior of its EVs, and some similar properties may exist across different cell types (for example, multiple cancer cell lines) that allow for cross-reactivity between different cell types. These properties must be systematically studied, and clear benchmarks must be developed to indicate the cutoff between cross-reactivity and selective uptake. The use of EV coatings on nanoparticles will likely continue to grow with time, potentially one day taking the place of common targeting methods such as those using antibodies, peptides, or aptamers. One approach that will likely see increased application in upcoming years is the encapsulation of nanoparticles in EVs that have been engineered for increased surface expression of proteins and antigens of interest. [51][52][53] Additionally, the use of EV coatings on nanoparticle surfaces will likely be further expanded from focusing on typical well-known particles to being used as a strategy for targeting novel nanoparticles to tissues of interest. Furthermore, applications of EV-coated nanoparticles can be expanded from primarily cancer applications to other applications in which targeted delivery of therapeutic and imageable nanoparticles can be utilized. The immune-evasive and immune-activating properties of EVs may also be further exploited for use in immunomodulatory nanoparticle-based therapies. These properties may be especially important in the development of personalized EV-based therapies that can utilize patient-derived EVs. At the same time, the scalability of EV-based therapies must also be considered, as any therapy that depends on the isolation of EVs from patient samples will be limited by the number of EVs that are produced within, and isolated from, the tissues of interest. In addition, potential negative effects from long-term accumulation of EV-coated nanoparticles in vivo can be explored. Long-term studies must be conducted to ensure that the lengthened retention of EV-coated nanoparticles does not lead to negative consequences such as liver toxicity.

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.