Tissue regeneration by stem cells is driven by the paracrine activity of shedding vesicles and exosomes, which deliver speciﬁc cargoes to the recipient cells. Proteins, RNA, cytokines and subsequent gene expression, orchestrate the regeneration process by improving the microenvironment to promote cell survival, controlling inﬂammation, repairing injury and enhancing the healing process. The action of microRNA is widely accepted as an essential driver of the regenerative process through its impact on multiple downstream biological pathways, and its ability to regulate the host immune response. Here, it is presented an overview of the recent potential uses of exosomes for regenerative medicine and tissue engineering. The conditions that aﬀect the production of exosomes in diﬀerent cell types are deliberated. This review also presents the current status of candidate exosomal microRNAs for potential therapeutic use in regenerative medicine, and in applications involving widely studied organs and tissues such as heart, lung, cartilage and bone.
Extracellular vesicles (EV) consist of exosomes, which are released upon fusion of the multivesicular body with the cell membrane, and microvesicles, which are released directly from the cell membrane. EV can mediate cell–cell communication and are involved in many processes, including immune signaling, angiogenesis, stress response, senescence, proliferation, and cell differentiation. The vast amount of processes that EV are involved in and the versatility of manner in which they can inﬂuence the behavior of recipient cells make EV an interesting source for both therapeutic and diagnostic applications. Successes in the ﬁelds of tumor biology and immunology sparked the exploration of the potential of EV in the ﬁeld of regenerative medicine.
Cell-free strategies for tissue regeneration
Tissue engineering strategies often involve the use of stem cells due to their regenerative/reparative potential. These eﬀects are still not fully understood but are generally attributed to stem cell trans-diﬀerentiation and cell fusion, paracrine eﬀects, mitochondrial transfer, and finally, to the release of extracellular vesicles (EVs) by stem cells. Recently, stem cell paracrine eﬀects are considered to be primarily responsible for the regenerative potential and increasing interest has focused on EVs and the bioactive molecules they release. Cell-free strategies using conditioned medium or EVs demonstrate that this may indeed be the case. Cell-free approaches may be advantageous when risk factors associated with stem cell use are considered. These can be intrinsic factors relating to cell origin, tumorigenic potential, diﬀerentiation and proliferation capacity, or extrinsic factors concerning cell handling, storage, transport conditions, among others. Furthermore, it is important to consider that stem cell yield decreases with donor age and that age also quantitatively impacts on performance. EVs, on the other hand, bypass a series of issues that arise with stem cell therapy. Moreover, EVs are easily stored and tested for optimal dosage and potency, all of which reduce the cost and time associated with stem cell expansion or collection from patients. Stem/progenitor cells transfer EVs containing functional regenerative signals to injured cells and organs. Examples of disease models used to demonstrate EV function in tissue regeneration include chronic wound, osteoarthritis, myocardial infarction, chronic kidney disease and lung diseases. Heterogenous compositions of proteins, lipids, and nucleic acids are selectively packaged by secreting stem cells in response to the microenvironment. The regenerative eﬀects of EVs are, at least in part, attributed to the transfer of specific protein and microRNA (miRNA) cargos. There is increasing interest in identifying the subpopulations of EVs that have maximum therapeutic potential and studies show that miRNAs are enriched in the biologically active exosome fraction, rather than in the microvesicle/shedding vesicle fraction.
Characterisation of exosomes and shedding vesicles
In 1967, shedding vesicles were described as platelet-dust by Peter Wolf. More than a decade later, Trams et al. found two populations of diﬀerently sized vesicles containing 5′-nucleotidase activity that were secreted by cells. These vesicles were derived from the plasma membrane present in cultures of normal and neoplastic cell lines. Pan et al. and Harding et al. studied the transferrin receptor during reticulocyte maturation and discovered that multivesicular endosomes (MVEs) contained bodies with a diameter of approximately 50 nm. Upon fusion of the MVEs with the plasma membrane, these vesicles were released from the cells and were later termed exosomes. These shedding vesicles (also called microvesicles or ectosomes) and exosomes are often collectively referred to as EVs. The secreted vesicles diﬀer depending on the cell type, origin and state. EVs mediate a series of cellular functions such as the transport of materials and intercellular communication. EVs are described as highly specialised messenger molecules, which can deliver biological signals. Increasing evidence proposes that these messages are highly dependent on specific conditions. For example, an environmental stressor or drug treatment can alter EV production, size profile and composition. Furthermore, EVs can carry ‘undesired’ messages that contribute to the spreading of diseases. EVs have diﬀerent origins; shedding vesicles (50 and 1000 nm) are formed at the cell surface via membrane budding, whereas exosomes (20–100 nm) originate from multivesicular bodies (MVBs) (250 to 1000 nm) . MVBs are either degraded or fused with the plasma membrane, releasing intraluminal vesicles (ILVs). Upon fusion and release of the vesicles inside the MVB cargo, these vesicles are called exosomes.
Production of shedding vesicles (SVs) involves traﬃcking of biomolecules towards plasma membrane regions that are enriched in ceramide and lipid rafts (cholesterol-rich microdomains). Ceramide-enhanced membrane curvature results in protrusion and budding, followed by stalk fission and subsequent detachment. The release of SVs may occur in resting cells, although this can be enhanced by several factors, including cell activation and exposure to proteins from activated complement cascades. Furthermore, cells subjected to irradiation, oxidative injury or under shear stress, show increased shedding. Membrane degradation and stimuli-driven increases in intracellular calcium are other impacting factors. Aside from calcium, phorbol ester activation of protein kinase C also enhances shedding in some cell types. Lipid raft-associated molecules tissue factor and flotillin and a high exposure of phosphatidylserine are typical markers of shedding vesicles. Depending on the cell type of origin, SVs may contain diﬀerent plasma membrane proteins. Specific markers may therefore be required to identify SVs depending on the cell type of origin. The generic marker annexin V is used to identify SVs, whereas CD45 is commonly used for leukocyte-derived SVs. For platelet-derived SVs, CD42b/CD31- and CD62P are used, while endothelial-derived SVs are characterised by CD31+/CD42−, CD62E, CD31+/CD42− and CD144. Apart from the already mentioned cholesterol and ceramide, sphingomyelin is also present in SVs. Additionally, SVs contain integrins, selectins, CD40 ligand and metalloproteinases. There are multiple databases that document molecular data of EVs, including shedding vesicles, such as Vesiclepedia and EVpedia. The miRandola 2017 database focuses on extracellular non-coding RNAs of EVs. Various RNA types including mRNA and miRNA are contained in SVs and these nucleic acids modulate functions of target cells. A recent study showed microvesicles from adiposederived stem cells promote angiogenesis via the delivery of microRNA-31 which, in turn inhibited anti-angiogenic gene HIF-1.
Biogenesis/formation and secretion of exosomes
Exosomes are formed by reverse-budding, where the vesicle contains cytosol. Their biogenesis initiates when cargo bound to the plasma membrane is endocytosed to form early endosomes. These early endosomes then mature into MVBs, which fuse with lysosomes to degrade their contents, or the MVBs release their contents as exosomes. The former process involves the labelling of membrane proteins for lysosomal destruction by ubiquitin. Recycling of the contents is prevented by the endosomal sorting complex for transport (ESCRT) machinery, specifically ESCRT-0, -I and -II, which presents ubiquitinbinding subunits that capture the ubiquitylated cargo. Protein sorting in MVBs involves segregation of the ubiquitinated proteins into membrane domains (lipid rafts) by ESCRT-0. These lipid rafts are rich in cholesterol and sphingomyelin, and contain lysobisphosphatidic acid (LBPA) and phosphatidylinositol-3-phosphate. Segregation is followed by endosomal invagination, with subsequent cargo sorting, and abscission of the vesicles through further intervention of the ESCRT machinery to form ILVs within the multivesicular endosomes (MVEs). MVEs deliver misfolded proteins in the plasma membrane, activated growth-factor, hormone and cytokine receptors to lysosomes. Alternatively, an ESCRT-independent mechanism leads to ILV formation. Alix (ALG-2-interacting protein X) binds directly with target molecules, where LBPA acts as a mediator, preventing degradation and inducing molecule sorting into ILVs. Furthermore, Alix supports endosomal membrane budding and abscission. Exosome secretion can occur in a constitutive or regulated manner, depending on the cell type of origin. In either case, MVBs move to the cell periphery, where they are tethered by tethering factors in conjunction with Rab GTPases. Fusion of the MVB membrane occurs through the vesicle-associated and acceptor membrane soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs), and synaptotagmin family members. As with shedding vesicles, environmental factors such as oxygen level, disease, mechanical stress or media composition modulate exosome release and exosome composition. For some cell types, increased amounts of exosomes are produced and released under hypoxic and oxidative stress conditions. Table 3 provides a few examples of conditions that impact on exosome production. Molecular composition. Cells release exosomes with distinct molecular and biological properties that mediate intercellular communication.28 The molecular composition of exosomes is dependent not only on the cell type of origin but also on the microenvironment. Microenvironment includes mechanical properties, topography and the presence of activating biochemical stimuli, which then regulate the protein cargo of the secreted exosomes.80 These proteins can be ubiquitous or cellspecific, and some are used as markers. Ubiquitous proteins that are commonly used as markers are tetraspanins (CD9, CD63, CD81 and CD82), heat shock proteins (HSP70 and HSP90), tumour susceptibility gene 101 (Tsg101) and Alix/syntenin. Other ubiquitous proteins include cytosolic proteins, flotillins, annexins, Rab proteins, molecules involved in signal transduction and metabolic enzymes. Caveolins, clathrin and transferrin receptors are also present and play an important role in the uptake of the vesicles by recipient cells. Specific proteins will depend on the cell type of origin . For example, exosomes derived from antigenpresenting cells contain antigen, present molecules such as major histo-compatibility complex (MHC) class I, MHC class II and CD1. Proteomic and genomic data of exosomes from diﬀerent sources are found in the previously mentioned databases (Vesiclepedia, EVpedia, and miRandola 2017) and also in ExoCarta 2012. The Urinary Exosome Protein Database focuses protein content of urinary exosomes. Cells transmit biological information to recipient injured cells through the biologically active lipids on EV membranes. The lipid contents of exosomes may also vary according to the cell type of origin. Exosomes are enveloped by a lipid bilayer that contains sphingomyelin, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, monosialotetrahex-osylganglioside (GM3) and phosphatidylinositol, which is similar in composition to the cell plasma membrane. Additionally, exosomes present cholesterol, ceramide, phospholipid phosphatidylserine and glyco-sphingolipids. Various signalling functions are contributed by bioactive lipids of stem cells including immunomodulation and anti-inflammatory activities and these have important roles in stem cell therapies. However, the particular lipids transported in EVs for specific regenerative medicine applications are yet to be identified. A breakthrough in the history of exosome research was in 2006–7 when the research groups of Ratajczak et al. and Valadi et al. reported that exosomes contained mRNA and miRNA, which could be delivered to recipient cells and subsequently have a functional role. Exosomal miRNAs are important modulators of gene expression and cause physiological changes in recipient cells. Exosomes contain RNA in a size range less than 700 nt. Additionally, mRNA fragments, long non-coding RNA, piwi-interacting RNA, ribosomal RNA and fragments of tRNA-, vault- and Y-RNA are also present. Furthermore, mitochondrial DNA, single-stranded DNA, double-stranded DNA and oncogene amplification products are present in exosomes. Exosomal transfer of miRNAs is now generally accepted as a mechanism for intercellular communication. Recently, specific miRNAs carried by exosomes were identified that could be potentially used in therapeutic applications. Two critical miRNAs that regulate inflammation, miR-155 and miR-146a, enhanced and reduced inflammatory gene expression respectively when delivered by exogenous exosomes in endotoxin-injury models. Furthermore, exosomes are used as disease biomarkers, as is the case for exosomal miRNA biomarkers miR-375, miR-21 and miR-574 in prostate cancer. The miRNAs within exosomes are not randomly selected for packaging but are selectively packaged by the secreting cell, possibly via a regulated mechanism. More studies are required to understand how secreting cells selectively choose and package diﬀerent cargos of RNA inside exosomes. Exosomal RNA profiles are reported to diﬀer from those of the parent cells but Guduric-Fuchs et al. showed that inducing overexpression of miR-146a in parent cells resulted in increased levels of miR-146a in the extracellular vesicles released from those parent cells. Since nearly half of the genes in human cells are regulated by miRNA, altered exosomal miRNA levels could determine outcomes of disease as well as therapy.
Shedding vesicles. Extraction methods for shedding vesicles are not as clearly defined as for exosomes. An initial centrifugation step is necessary to remove cells and stop the release of vesicles. The supernatant is then centrifuged at 10000–20000g and the pellet is analysed by flow cytometry. A subsequent ultracentrifugation at 100000g may be used for purification, yet this can lead to the formation of unwanted aggregates.
Exosomes. Several methods have been proposed for exosome isolation. The first method was developed by Johnstone et al. using ultracentrifugation (UC) to pellet the vesicles, and is still considered to be the “gold standard” for extraction. The method involves cell exclusion by centrifugation followed by re-centrifugation of the supernatant at 100000g for 90 minutes. Since then, several variations have been reported for diﬀerential ultracentrifugation. However, the viscosity of the solution impacts on the yield, resulting in poor exosome recovery from biofluids. Changes in sedimentation distance, temperature, time, or speed may impact on the yield and should therefore be studied. ExoQuick™ (EQ) isolation kits comprise a family of reagents that rely on exosome precipitation. EQ solution is added to the exosome-containing sample and refrigerated over night, after which the mixture is centrifuged at 1500g for 30 min for pellet retrieval. Other precipitation-based kits have also been developed such as Life Technologies™ Total Exosome Isolation Reagent, Exo-Spin™ and Invitrogen™ kits. In comparison with EQ, the average vesicle size from human serum samples was smaller and there was superior RNA recovery with the Life Technologies™ Total Exosome Isolation Reagent. Other precipitation-based methods (ExoSpin and Invitrogen kits) yielded higher BT474 cell-derived exosome content than density gradient method (PureExo®) and UC. Recently, a novel method has been proposed that exploits the presence of the negatively-charged lipid phosphatidylserine. Through titration with acetate, the surface charge of the cellfree supernatant is neutralized, resulting in exosome precipitation. Alvarez et al. analysed levels of mRNA, miRNA, and protein to determine which method was best for isolation of urinary exosomes. Whilst a modified EQ-based method led to a higher exosome yield and better quality mRNA and miRNA, UC-based methods resulted in higher protein purity. Similarly, for human serum samples, results showed similar exosomal miRNA profiles for the two methods, yet the miRNA yield was higher for EQ. Furthermore, miR-92a and miR-486-5p levels were statistically diﬀerent for both methods. The ultrafiltration or density gradient centrifugation method was developed with monocyte-derived dendritic cells. It involves clarification of the supernatant, followed by ultrafiltration with a molecular weight cut-oﬀ at 500 kDa. The concentrate is diafiltered and ultracentrifuged with an underlying density cushion of 30% sucrose/deuterium oxide. When compared with UC for isolation of human tongue cancer cell line-derived exosomes, density gradient centrifugation yielded uniformly distributed sized particles, as well as enriched exosome markers.
Yamada et al. studied combinations of UC followed by EQ or by density gradient centrifugation, in human milk exosome isolation. UC with EQ precipitation led to a rapid and increased exosome recovery, whereas UC with density gradient centrifugation led to samples with higher purity. Other exosome extraction methods include purification by high performance size exclusion liquid chromatography, size exclusion chromatography (SEC), and separation by sucrose density gradient due to their buoyant density of 1.23 to 1.16 g L−1.102 More recently, OptiPrepTM density gradient isolation method is progressively replacing the sucrose gradient method due to higher purity of the exosomes obtained. For rat and human blood plasma-derived exosomes, isolation by UC resulted in low yield and high sample contamination. In contrast, isolation by SEC led to purer samples, but a low vesicle yield. Quantification of exosomes derived from antigen-presenting cells is possible through immunoprecipitation technologies using antibody loaded magnetic cell beads. In this method, bead-exosome complexes are formed due to the immune-magnetic interaction between the beads and the exosomes containing human MHC class I molecules. Compared to the diﬀerential ultracentrifugation method, magnetic bead-based isolation of exosomes is designed to be fast, eﬃcient and selective by the use of antibodies against the common exosomal markers – tetraspanins C9 or CD81. For human colon carcinoma cells, a study showed that EpCAM immunoaﬃnity capture-enriched exosomes expressed exosome markers and associated proteins more highly than the UC and OptiPrep methods. Overall, comparisons of SEC, magnetic beads, UC and EQ, for isolation of exosomes derived from ascites of clinical adenocarcinoma samples, showed the highest sample purity and RNA concentration was obtained by the EQ method, followed closely by SEC. Similarly, EQ showed the highest exosomal protein yield, whilst ultracentrifugation had the lowest. Besides regenerative medicine applications, exosomes are important biomarkers of disease as they are conveniently found in abundance in many biological fluids including blood, urine and saliva. However, extraction of exosomes from viscous fluids is challenging for several traditional isolation methods. Novel microfluidic-based isolation methods result in higher yields and purity, when compared with the aforementioned methods. Microfluidic strategies based upon acoustic, electrophoretic, and viscoelastic mechanisms emerge as highly eﬀective approaches for isolation and fractionation of EV populations. These approaches reduce isolation time, reagent consumption, sample volume, and are more cost eﬀective than conventional methods. Another significant advantage of these methods is their compatibility with biofluids. As described below, exosomal mRNA and miRNA are key modulators in tissue regeneration, aﬀecting cellular behaviour and outcome through paracrine eﬀects. Thus, more important than the yield of each extraction method, is mRNA and miRNA purity and the quality of extracted exosomes, which should drive the choice for the method of isolation. Thus, ExoQuick™ and microfluidic strategies may be the best exosome isolation options for tissue regeneration applications due to a greater mRNA and miRNA purity. One of the previously stated advantages of microfluidics is the smaller sample size. Whilst this may indeed be an advantage in the case of disease screening or exosome characterisation, it may act as an impediment for tissue engineering strategies, where a greater number of exosomes are required per study. Furthermore, when considering clinical trials this problem increases dramatically. There is no consensus regarding the ideal method for exosome isolation and novel methods are likely to emerge that oﬀer improved yield and/or purity. The outcome of exosome isolation is dependent on the cell source from which the exosomes are derived, and caution is needed in choosing a method since diﬀerent extraction methods alter the exosomal protein and miRNA content. Thus, it is important to use a consistent isolation method when comparing exosome populations.
Roles of EVs in regenerative medicine
Overview. Tissue engineering is a promising option for the regeneration of tissues and this technology can be combined with stem cell based therapies since there is evidence that stem cells promote and regulate tissue regeneration. Current strategies use the cell and/or its by-products secreted into the culture medium, as the active biological agent(s). EVs are by-products that regulate intercellular communication, which is necessary for multicellular organisms to maintain their vital functions. These extracellular vesicles comprise the following sub-classes; apoptotic bodies, shedding vesicles and exosomes. EVs are a cell-free alternative to current stem cell therapies with advantages of lower immunogenic response and preservation of biochemical activity upon storage. Furthermore, employing EVs bypasses important safety concerns associated with the engraftment of viable replicating cells, which have the potential for long term pathological transformation. MSC-derived exosomes and shedding vesicles have been investigated for a range of in vitro and in vivo applications involving lung, liver, kidney, and colon injury, as well as myocardial infarction, skin burns and defects, and cerebral artery occlusion. Gatti et al. studied the eﬀect of human bone marrow mesenchymal stem cell (hBMSC)-derived exosomes on acute and chronic kidney injury induced by ischemia–reperfusion injury in a rat model, and concluded that exosome administration led to a reduction of apoptosis and enhanced the proliferation of tubular epithelial cells. A rat model of acute kidney injury induced by cisplatin was used by Zhou et al. to demonstrate the beneficial eﬀect of human umbilical cord MSC-derived exosomes (hUCMSC-Ex) in decreasing oxidative stress, inhibiting renal tubular apoptosis as well as reducing blood urea nitrogen and creatinine levels. Li et al. used a carbon tetrachloride-induced murine liver fibrosis model and determined that hUCMSC-Ex reduced fibrous capsules formation at the surface, and decreased hepatic inflammation. Similarly, Yih observed that exosomes isolated from human embryonic stem cell-derived human mesenchymal stem cells, limited the extent of injury using the same murine drug-induced liver injury model. Moreover, in injury models induced by Acetaminophen and hydrogen peroxide treatment of immortalized murine transforming growth factor alpha transgenic hepatocyte cells, exosomes increased cell viability in the Acetaminophen model, and were cytoprotective in the hydrogen peroxide model. In an induced stroke model, Xin et al. showed that rat bone marrow MSC-derived exosomes increased neurite remodelling, neurogenesis and angiogenesis in the ischemic boundary zone. Zhang et al. used a rat model for induced traumatic brain injury and verified that administration of exosomes increased angiogenesis and neurogenesis, and reduced neuroinflammation. Pascucci et al. studied the promotion of angiogenesis by horse adipose tissue MSCs-derived microvesicles (exosomes + shedding vesicles) through the scratch migration assay and rat aortic ring assay. Administration of microvesicles significantly increased cell migration after 24 hours, when compared to control medium (EndoGROTM basal medium, and 1:1 Dulbecco’s modified Eagle medium (DMEM) and endothelial basal medium (EBM) for the scratch assay and ring assay, respectively). Regarding the colon, Yang et al. showed protective eﬀects of rat bone marrow mesenchymal stem cell-derived microvesicles in the 2,4,6-trinitrobenzene sulfonic acidinduced colitis model. Microvesicles attenuated inflammatory activity, oxidative perturbations and reduced apoptosis in a dose-dependent manner. Exosomes also enhance MSC diﬀerentiation, as Takeda and Xu demonstrated the development of neuron-like morphology and the upregulation of neuronal markers in MSCs treated with neuronal cell line-derived exosomes. In a study reporting the potential of exosome therapy for invertebral disc degeneration, exosomes secreted by nucleus pulposus cells were found to promote diﬀerentiation of MSCs to a nucleus pulposus-like phenotype. The osteogenic potential of exosomes was demonstrated for the first time in vitro and in vivo when exosomes isolated from diﬀerentiated MSCs, induced lineage specific diﬀerentiation of naïve MSCs. The role of miRNA in tissue engineering and regenerative medicine is of interest because miRNAs influence a wide range of cell functions including diﬀerentiation and gene expression. Exosomal miRNA is thought to be the key factor in intracellular communication and regulation, with a preponderant role in the modulation of biological functions of acceptor cells. As proven by Valadi et al., exosomes can actively shuttle translatable RNAs between mast cells for the production of specific proteins. Since then, intensive investigations on the potential of miRNA in the repair and restoration of tissue function for regenerative therapies involving the heart, skin, lung, bone, cartilage, and neurons have been reported. miRNAs present a half-life between 28 and 220 h, which is significantly greater than typical mRNAs (10 h). Delivering mRNA is problematic because they are unstable and are rapidly degraded by nucleases and therefore viral vectors and lipid nanoparticles have been tested as delivery vehicles. Cells and extracellular vesicles are also attractive delivery vehicles for miRNA delivery, improving circulation stability even in the presence of ubiquitous ribonucleases (RNases). However, studies have shown that RNase pre-treatment of shedding vesicles led to an inhibition of the biological eﬀects. Additionally, exosomal miRNA has the advantage of being naturally stable enough to be a potential biomarker.
Taking into consideration the critical roles of MSCs and their products in tissue regeneration, MSC-derived exosomes are particularly promising candidates for developing cell-free therapies. In addition, exosomes released from immune cells (monocytes, leukocytes, granulocytes, and lymphocytes) are implicated in several fundamental biological processes, such as the recruitment of inflammatory cells, neovascularization, and coagulation. Thus, they are also of vital importance in ensuring the appropriate inflammatory reaction after injury, which would boost tissue repair and regeneration. Hence, existing evidence as to the potential uses of exosomes in promoting tissue repair and regeneration will be reviewed in the following section.
Neural regeneration. It has been established that exosomes could be used as biomarkers for brain injuries. A proof of concept study conducted by Ji et al. suggested that the serum exosomal miR-9 and miR-124 were promising biomarkers for diagnosing acute ischemic stroke (AIS) and evaluating the degree of damage caused by ischemic injury. Furthermore, the regenerative effects of exosomes on neurons and nerves have been reported. Frohlich et al. reported that exosomes derived from glutamate stimulated oligodendrocyte can promote survival in neurons deprived of oxygen and glucose. In addition, Xin et al. reported that exosomes extracted from multipotent mesenchymal stromal cells could deliver miRNA-133b to neural cells to boost neurite outgrowth, which was the first article revealing that communication occurs between MSCs and brain parenchymal cells. Takeda et al. found that treatment with exosomes derived from differentiating neuronal cells could induce neuronal differentiation in human MSCs. This work also suggested that delivery of miR-125b via exosomes might be the possible underlying mechanism. Recently, Zhang et al. had investigated the regenerative potential of exosomes derived from human bone marrow mesenchymal stem cells (hBMSCs) on traumatic brain injury (TBI) in rats. They found that compared with the negative control, endogenous angiogenesis and neurogenesis of rats with TBI systemically administered hBMSC-generated exosomes was enhanced, while neuroinflammation was attenuated. These results suggest that the exosomes released by hBMSCs significantly improve functional recovery in rats after TBI. The same group also found that native exosomes secreted by MSCs can promote axonal growth while tailored MSC-exosomes carrying elevated the miR-17-92 cluster could further boost this effect, since tailored exosomes can selectively deliver their cargo miRNAs to recipient neurons and activate their target signals. Furthermore, they also determined that neuronal internalization of MSC exosomes was accomplished mainly via the SNARE complex. El et al. also reported on the pro-regenerative effect of exosomes derived from human adipose-derived stem cells (hASCs) on HT22 neuronal cells post injury. They found that exosomes derived from the hASCs boosted neuronal survival and proliferation by increasing expression of PKCδII in HT22 cells. They also found that MALAT1, a long noncoding RNA in hASCs derived exosomes mediated splicing of PKCδII thereby increasing its expression. Additional research by this group indicated that the regenerative effect of hASCs derived exosomes could be further enhanced by insulin stimulation. More recently, it was reported by Mead et al. that exosomes isolated from BMSCs could significantly promote the survival of retinal ganglion cells (RGCs) and regeneration of their axons; these beneficial effects might be correlated with argonaute-2, a key miRNA effector molecule. Spinal cord injuries (SCIs) often result in permanent damage due to the failure of axonal regeneration. In the peripheral nervous system, axonal regeneration is mainly supported by Schwann cells (SCs). After nervous damage, SCs can dedifferentiate, proliferate and efficiently guide axons to their original target tissues. Lopez-Verrilli et al. reported that exosomes derived from dedifferentiated SCs could be specifically internalized by axons, markedly increasing axonal regeneration in vitro and enhancing regeneration after sciatic nerve injury in a Sprague Dawley (SD) rat model. Their research also indicated that SCs-derived exosomes promoted axonal regeneration by inhibiting activity of RhoA, a GTPase which could inhibit axonal elongation and promote growth cone collapse. But they didn`t mention the underlying molecular mechanism in article. Goncalves et al. found that the retinoic acid receptor β (RARβ) agonist could promote locomotor and sensory recovery in rat cervical avulsion models. Further mechanism research revealed that in RARβ-agonist-treated neurons, activity of PTEN (a major negative regulator of neuronal regeneration) was obviously decreased by cytoplasmic phosphorylation. Moreover, the exosomal secretion of RARβ-agonist-treated neurons also increased. After being taken up by astrocytes, these exosomes could reduce the proliferation of astrocytes and cause them to arrange around the regenerating axons, preventing scar formation. Finally, the neuronal and neuronal-glial regenerative effects of RARβ signaling result in axonal regeneration into the spinal cord.
Myocardial regeneration. The potential protective effects of exosomes have already been explored in a series of myocardial ischemia reperfusion injury models. Ibrahim et al. showed that exosomes isolated from cardiosphere-derived cells (CDCs) could inhibit apoptosis and promote the proliferation of cardiomyocytes when injected into mouse hearts suffering from ischemia injury. They also found that these beneficial effects were closely related to the enrichment of miR-146a in exosomes. Teng et al. reported that exosomes generated from BMSCs significantly enhanced tube formation of human umbilical vein endothelial cells and inhibited proliferation of T-cell in vitro. In addition, reduced infarct size and preserved cardiac function were also observed in SD rats with acute myocardial infarction due to enhanced neovascularization and suppressed inflammation response. Zhang et al. also found that preconditioning with MSC exosomes could boost the proliferation, migration, and angio-tube formation of cardiac stem cells (CSCs) in a dose-dependent manner. It was reported by Khan and colleagues that mouse embryonic stem cell-derived exosomes (mES-Ex) possessed the ability to promote endogenous repair and enhance cardiac function after myocardial infarction. They found that after mES-Ex were intramyocardially administered in mice at the time of myocardial infarction, both neovascularization and cardiomyocyte survival was enhanced, and myocardial fibrosis post infarction was evidently suppressed concurrent with enhanced c-kit(+) cardiac progenitor cells (CPCs) survival and proliferation. This research group also investigated the underlying mechanisms of these beneficial effects via microRNA array analysis. Their results indicated that the regenerative potential of mES-Ex was tied to the delivery of embryonic stem cell-specific miR-294 to CPCs, which could promote cell survival and proliferation of the latter. Zhao et al. showed that exosomes derived from human umbilical cord mesenchymal stem cells (hUCMSCs) possessed a protective effect in an acute myocardial infarction (AMI) rat model. They found that exosomes might improve cardiac systolic function by protecting myocardial cells from apoptosis and promoting angiogenesis. These beneficial effects were potentially associated with modulating the expression of members of the Bcl-2 family. Vicencio et al. also showed that a specific cardio-protective pathway, involving TLR4 and HSP27, could be activated by exosomes in plasma. Recently, Agarwal and co-workers evaluated the regenerative role of human CPCs (hCPCs)-derived exosomes in a rat myocardial ischemia reperfusion injury model. In their investigation, hCPCs obtained from children of different ages were isolated and cultured under hypoxic and normal conditions. Then, exosomes were isolated from the conditioned media and delivered to rats. Finally, their results indicated that exosomes released by neonate CPCs improved cardiac function by decreasing fibrosis and improving angiogenesis regardless of oxygen levels in culture conditions, whereas exosomes from older children could only gain reparative power when CPCs were subjected to hypoxic conditions. This is the first investigation demonstrating that donor age and hypoxia level can influence the therapeutic efficacy of human CPC-derived exosomes. Interestingly, Beltrami et al. found that the pericardial fluid (PF) also contained exosomes enriched with miRNAs co-expressed in the patient myocardium and vasculature versus peripheral plasma. Amazingly, the specific exosomes in the PF could improve the survival, proliferation, and networking of endothelial cells (ECs) cultured in vitro and restore the angiogenic capacity of ECs depleted of endogenous miRNA profiles. Most importantly, the PF exosomes could improve blood flow recovery and angiogenesis after ischemic injury in the mouse model. Further investigation suggested that PF exosomes might orchestrate the vascular repair process by delivering miRNA let-7b-5p to ECs.
Hepatic regeneration. Exosomes have already been used as specific biomarkers for hepatocyte damage and inflammation in acute liver injury. Momen-Heravi et al. reported that the exosomal levels in the plasma of patients with alcoholic hepatitis were obviously higher than those of the healthy population, as were the specific miRNA profiles in their exosomes. The regenerative potential of exosomes on liver has also been investigated recently. In acute liver injury, Nojima et al. found that hepatocyte-derived exosomes could promote the proliferation of hepatocytes in vitro and liver regeneration in vivo. Their research suggested that the underlying mechanism might involve exosomal transfer of neutral ceramidase and sphingosine kinase 2 (SK2) to target hepatocytes. Moreover, they also found that the levels of circulating exosomes with proliferative effects also increased after liver injury. Tan et al. also investigated the regenerative potential of MSCs-derived exosomes in a carbon tetrachloride (CCl4)-induced liver injury mouse model. They reported that CCl4-induced liver injury was notably attenuated by concurrent treatment with MSCs-exosomes, which might be achieved mainly through the activation of proliferative and regenerative responses. Recently, Yan and coworkers reported that systemic administration of hUCMSC-derived exosomes (hUCMSC-Ex) could effectively rescue mice from CCl4-induced liver failure; this protective effect was closely associated with hUCMSC-Ex-derived glutathione peroxidase1 (GPX1). The antioxidant and anti-apoptotic abilities of hUCMSC-Ex would diminish after knockdown of GPX1 in hUCMSCs. Nong et al. also evaluated the regenerative potential of exosomes derived from human-induced pluripotent stem cell-derived mesenchymal stromal cells (hiPSC-MSCs-Exo) during hepatic ischemia-reperfusion injury. Their results indicated that hiPSC-MSCs-Exo administration can alleviate warm hepatic ischemia-reperfusion injury by suppressing inflammatory responses, attenuating oxidative stress responses, and inhibiting cellular apoptosis. However, the molecular mechanism by which these effects occur was not further elucidated.
Renal regeneration. It was reported by Tomasoni et al. that exosomes released by hBMSCs could promote the proliferation of cisplatin-damaged proximal tubular epithelial cells via horizontal transfer of IGF-1 receptor (IGF-1R) mRNA. Zhou and colleagues also demonstrated that exosomes derived from hUCMSCs could alleviate acute kidney injuries induced by cisplatin in rats by suppressing renal oxidative stress and apoptosis, while increasing renal epithelial cell proliferation. Borges et al. found that tubular epithelial cells exposed to hypoxic conditions can produce exosomes enriched in transforming growth factor-β1 (TGF-β1) mRNA, which can activate fibroblasts to initiate the fibrotic repair response. This study suggested that TGF-β1 mRNA delivered by exosomes constituted a rapid response to initiate tissue regenerative responses after hypoxia injury. Their finding also enlightens the potential for exosome-targeted therapies to control tissue fibrosis. Burger et al. examined the therapeutic potential of human umbilical cord blood-derived endothelial colony-forming cells (ECFCs) and ECFC-derived exosomes in a mouse model of ischemic acute kidney injury (AKI). They found that intravenous administration of ECFCs can attenuate renal injuries in mice with ischemic AKI; while direct intravenous administration of ECFC-derived exosomes had the same effect. Recently, this group demonstrated that exosomes derived from ECFCs were enriched in miR-486-5p and delivery of ECFC-derived exosomes could reduce ischemic AKI via transfer of miR-486-5p targeting PTEN. In addition, Wang et al. found that MSCs that were engineered to overexpress miRNA-let7c could selectively localize to injured kidneys and up-regulate miR-let7c gene expression to attenuate kidney injury. The exosomes derived from these engineered MSCs were also able to selectively transfer miR-let7c to damaged kidney cells to achieve anti-fibrotic functions. Jiang et al. also investigated the therapeutic potential of exosomes from urine-derived stem cells (USCs-Exo) on kidney injury repair in a SD rat model. Their results suggested that streptozotocin-induced kidney injury could be alleviated by weekly intravenous tail injections of USCs-Exo, which could obviously inhibit podocyte apoptosis and promote vascular regeneration and cell survival. They also deduced from an USCs-Exo contents assay that the regenerative potential might be related to the enrichment of cytokines vascular endothelial growth factor (VEGF), TGF-β1, angiogenin, and bone morphogenetic protein 7 (BMP-7) in USCs-Exo.
Cutaneous regeneration. Angiogenesis is of crucial importance in various physiological processes including cutaneous wound healing and tissue regeneration. It has already been established that exosomes released from cancer cells can modify the tumor environment to enable the metastasis of cancer cells and promote angiogenesis, which was reviewed in other articles. This type of beneficial effect might also be found in exosomes derived from other sources, beyond cancer cells. Liang et al. found that exosomes released by human adipose-derived MSCs (adMSCs) can significantly promote endothelial cell angiogenesis in vitro and in vivo. Further investigation indicated that exosomes derived from adMSCs could transfer miR-125a to endothelial cells, resulting in the down-regulation of angiogenic inhibitor delta-like 4 (DLL4). Yuan et al. also demonstrated that exosomes extracted from human urine derived stem cells (hUSCs) could enhance skin wound healing by promoting angiogenesis in vitro and in vivo. However, they did not determine the underlying molecular mechanisms of this beneficial effect. Burn injury, one of the most common causes of cutaneous damage, could significantly intensify the inflammatory reaction, including increased tumor necrosis factor α (TNF-α) and interleukin-1β (IL-1β) levels, and decreased IL-10 levels. Li et al. found that administration of hUCMSC-exosomes could successfully reverse the burn-induced inflammatory reaction . Further research suggested that miR-181c in hUCMSC-exosomes weakened inflammation by down-regulating the TLR4 signaling pathway following burn injury, which could attenuate excessive inflammation and boost tissue repair. Zhao et al. investigated the regenerative potential of exosomes derived from human amniotic epithelial stem cells on the healing of full-thickness skin defects in rats. Exosomes were isolated and then different concentrations were subcutaneously injected around the wound site. Eventually, they found that exosomes released by human amniotic epithelial stem cells could promote the migration and proliferation of fibroblasts, accelerating healing of full-thickness skin defect in a dose-dependent manner. Zhang et al. also demonstrated that exosomes from human umbilical cord blood-derived endothelial progenitor cells (EPC-Exos) possessed robust pro-angiogenic and wound healing effects in a diabetic rat model. Microarray analyses indicated that exosomes markedly altered the expression of a series of genes involved in the Erk1/2 signaling pathway, and functional studies further confirmed that this signaling pathway was of vital importance during the exosome-induced angiogenic responses of endothelial cells. Recently, Guo and colleagues demonstrated that exosomes derived from platelet-rich plasma (PRP-Exos) can effectively induce the proliferation and migration of endothelial cells and fibroblasts to promote angiogenesis and re-epithelialization in chronic cutaneous wound healing processes, highlighting the healing of chronic ulcers. Pu and colleagues reported that the survival and capillary density of flaps subjected to ischemia-reperfusion injury were significantly enhanced via the injection of exosomes released by adipose-derived stem cells. Han et al. had also found that exosomes derived from corneal epithelial cells could fuse to keratocytes and induce myofibroblast transformation after a corneal wound occurred. Furthermore, corneal epithelial cell-derived exosomes can induce the proliferation of endothelial cells and aortic ring sprouting in vitro. Their results indicated that epithelial cell-derived exosomes might be involved in corneal wound healing and neovascularization processes, which may be applied as therapeutic interventions in the future. Exosomes could also orchestrate controlled cutaneous regeneration in a bipolar manner. Zhang et al. showed that exosomes derived from hUCMSCs could trigger the Wnt/β-catenin signaling pathway to repair damaged skin tissue during the early stages of deep second-degree burn healing, and they could also inhibit Wnt/β-catenin signaling through the induction of YAP phosphorylation to circumvent excessive skin cell expansion and collagen deposition after the remodeling phase.
Skeletal regeneration. Bone regeneration using MSCs and tissue engineering strategies is one of the most widely researched fields in regenerative medicine. Furuta and colleagues evaluated the therapeutic effects of exosomes isolated from MSC-conditioned medium (CM) in the fracture healing process in a CD9-/-mice model, which produces low levels of exosomes. Identical femur fractures were created in both test and control groups. Then mice in the test group were injected with exosomes, while controls were injected with exosome-free conditioned medium. The bone union rates were measured, and the results suggested that the exosomes in MSC-CM could accelerate the fracture healing process. Recently, Zhang et al. investigated the pro-osteogenic potential of hiPSC-MSC-Exos. They showed that the isolated exosomes could effectively stimulate the proliferation and osteogenic differentiation of bone marrow MSCs derived from ovariectomized rats in vitro and in vivo. These results also suggested that the therapeutic effects of hiPSC-MSC-Exos could be intensified by increasing their exosomal concentration. Bioinformatics analyses further confirmed that the PI3K/Akt signaling pathway was the principal regulator during the hiPSC-MSC-Exos-induced osteogenic differentiation of BMSCs.
Chondral regeneration. Zhang et al. showed the therapeutic effects of exosomes derived from human embryonic mesenchymal stem cells on cartilage repair. Osteochondral defects were created on bilateral trochlear grooves in a rat model. One defect was weekly intra-articularly injected with human embryonic MSC-derived exosomes for 12 weeks, and the contralateral defect was injected with PBS. Eventually, complete restorations of cartilage were observed in defects treated with exosomes, while only fibrous repair tissues were found in the control group. However, this research didn’t exclude the possibility that exosome injected into the trochlear groove might be transferred to the contralateral defect through blood circulation, thereby interfering with the outcome of control group. Recently, Zhu and co-workers evaluated the regenerative potential of exosomes secreted by human synovial membrane MSCs (SMMSC-Exos) and induced pluripotent stem cell-derived MSCs (iMSC-Exos) on osteoarthritis (OA), which is induced by failure of articular cartilage regeneration in a rat model. They found that both iMSC-Exos and SMMSC-Exos could attenuate OA by stimulating chondrocyte migration and proliferation, while the regenerative power of iMSC-Exos was stronger than that of SMMSC-Exos. This inspiring work also provided new perspectives for cell-free therapies for cartilage injury and osteoarthritis.
Muscular regeneration. Nakamura et al. showed that exosomes derived from MSCs could promote myogenesis and angiogenesis in vitro. They verified that MSC-derived exosomes promoted muscle regeneration in a mouse model of cardiotoxin-induced muscle injury, which might be mediated by miR-494. Recently, Choi and colleagues also found that exosomes derived from human skeletal myoblasts (hSkMs) during myotube differentiation could effectively induce a myogenesis response in hASCs. Moreover, a laceration mouse model verified that exosomes derived from differentiating hSkMs could accelerate skeletal muscle regeneration by reducing collagen deposition and increasing the number of regenerated myofibers in injured muscles.
Lung regeneration. In response to injury and disease, the lung is able to activate quiescent stem or progenitor cell lineages, as well as epithelial cell lineages, which can re-enter the cell cycle and thereby repopulate lost cells. Attempts to facilitate tissue regeneration of the lung using exogenous stem cells reveal that beneficial eﬀects are produced through paracrine actions regardless of the route of delivery. The lung is unique in that it permits not only an intravenous route for stem cell delivery but also a more direct delivery route by inhalation. Gupta et al. in 2007 carried out one of the first studies to show intratracheal delivery of MSCs is as eﬀective as the intravenous route in attenuating acute lung injury in mice. Subsequently, many studies investigated the intravenous and intratracheal routes of delivery of stem/progenitor cells but to date there is no consensus on the best route. Regardless of the route of delivery, most studies support a paracrine mechanism of action and it is widely accepted that stem cell-derived extracellular vesicles are promising paracrine factors that aid and accelerate lung regeneration. Microvesicles released by human MSCs were recently shown to be as eﬀective as the stem cell type of origin in alleviating lung inflammation caused by severe bacterial pneumonia, and in improving survival in a murine model. MSC exosomes also ameliorate chronic lung diseases including bronchopulmonary dysplasia, as recently reported. In a mouse model of hyperoxia-induced pulmonary hypertension and bronchopulmonary dysplasia, even a single dose of MSC exosomes restored lung architecture and led to significant long-term benefits in the lung function through immunomodulatory eﬀects. The mRNA sequencing of the lung revealed that the proinflammatory M1 state of macrophages was suppressed while the anti-inflammatory M2like state was augmented. Numerous studies showed not only specific miRNAs contained in EVs are crucial for lung repair in vivo, particularly for influenza, hypoxia-induced pulmonary hypertension, and ventilator-induced lung injury, but also specific mRNAs are eﬀective in endotoxin-induced acute lung injury, as detailed below. Tan et al. analysed the regulation of miRNAs during early and late stages of repair after influenza infection in a mouse model. The miRNAs that play major roles in priming pulmonary tissues for repair and regeneration following influenza were identified, and particularly miR-290, miR-21, let-7 and miR-200 appeared to be highly involved in the regeneration process. Beyond regulating regeneration, miR-21 and let-7 have further benefits such as anti-inflammatory properties. These properties are of significant benefit since many lung conditions that require regeneration have concomitant, uncontrolled inflammation. miR-21 is also present in microvesicles of important stem cell types such as murine embryonic stem cells and human bone marrow mesenchymal stem cells. Lee et al. provided strong evidence that exosomes prevented the activation of hypoxic signalling, which underlies pulmonary inflammation and pulmonary hypertension. In a model of hypoxia-induced pulmonary hypertension, mice treated with exosomes derived from umbilical cord-derived MSCs showed suppression of hypoxia-induced pulmonary inflammation and vascular remodelling, in a dose-dependent manner. In that study, exosomes derived from mouse lung fibroblasts were used as a control and showed no protective eﬀect against pulmonary hypertension, when compared with MSC-derived exosomes. The suppressive eﬀect of the exosomes derived from umbilical cord-derived MSCs was attributed to their enriched levels of miR-16 and miR-21. miR-16 reduces SERT expression, which is a critical protein for the resolution of pulmonary oedema. Although let7b miRNA levels were similar in exosomes derived from fibroblasts and MSC-derived exosomes, the latter contained levels of let7b pre-miRNA that were ten times higher. The reduction of IL-6 pro-inflammatory cytokine by the MSC-derived exosomes led to de-activation of the STAT3 gene, which is a key mediator of hypoxic, pro-inflammatory signalling associated with pulmonary hypertension. In a mouse model of endotoxin-induced acute lung injury, reduced inflammation and prevention of pulmonary oedema formation in the injured alveoli followed intratracheal administration of human bone marrow MSC-derived microvesicles. A major finding of this study was that the microvesicles contained the mRNA for the keratinocyte growth factor (KGF). KGF was previously shown to restore alveolar fluid clearance in an ex vivo model of acute lung injury. Transfer of the mRNA from the microvesicles to the injured alveolar epithelium, and subsequent expression of KGF, was thought to be a key factor in the repair process. miRNA expression studies in acute lung injury report that overexpression of several miRNAs including miR-146 may be directly related to the host response that regulates macrophage function and inflammatory cytokine expression. An important functional role of miR-146 was suppression of TNF-α, IL-6, and IL-1β expression in alveolar macrophages through the inhibition of IRAK-1 and TRAF-6 in a mouse model of ventilator-induced lung injury. A recent study found miR-146a was critical for the immunomodulatory eﬀects of exosomes derived from human umbilical cord MSCs in a mouse model of sepsis. Another study reported that miR-146a reduces inflammatory gene expression and inhibits endotoxin-induced inflammation in mice.93 miR-146a reduces microbial and mechanically induced inflammation in lung epithelia through the toll-like receptor signalling pathway. A recent review by Basu and Ludlow highlighted that the selection of RNA for recruitment into exosomes is highly precise and is regulated by proteins such as RNA binding complex ESCRT-II, which is present in the exosomal membrane. Interestingly, Song et al. showed that miR-146a could be selectively packaged into the exosomes of MSCs by pre-treating the cells with the proinflammatory cytokine IL-1β. Future studies that use exosomes to improve regeneration of the lung after injury should consider conditioning exosomes such that they are enriched in miR-146a.
Regeneration of other tissues and organs. The regenerative potential of exosomes has also been investigated in other tissues and organs such as in dental pulp tissue and pancreas. Huang and colleagues have evaluated the potential of exosomes derived from dental pulp cells cultured under odontogenic differentiation conditions to induce odontogenic differentiation in naive human dental pulp stem cells (DPSCs) and human bone marrow derived stromal cells (HMSCs) in vitro and in vivo. After being taken up by DPSCs and HMSCs, exosomes could trigger the p38 mitogen activated protein kinase (MAPK) pathway and increase the expression of genes required for odontogenic differentiation in a dose-dependent and saturable manner via the caveolar endocytic mechanism, boosting the regeneration of dental pulp-like tissue collectively. In addition, they found that exosomes could bind to matrix proteins such as type I collagen and fibronectin, which enabled exosomes to be tethered to biomaterials. Diabetes mellitus (DM) is a complex metabolic disease characterized by glucose overproduction and under-utilization, constituting one of the most important global health problems. Insulin, which functions to decrease blood glucose, is synthesized and secreted by pancreatic β-cells. Reductions in the amount and function of pancreatic β-cells result in relative or absolute insufficient insulin secretion, contributing to the pathophysiology underlying diabetes. Oh et al. investigated the regenerative potential of extracellular vesicles, including exosomes and microvesicles, derived from a murine pancreatic β-cell line. They found that bone marrow cells in diabetic immunodeficient mice can effectively differentiate into pancreatic β-cells when cultured with exosomes. This inspiring finding might provide an ideal solution to the dilemma that sources of functional islets for transplantation are particularly limited.
Exosomes as drug delivery vehicles
Exosomes are endogenous carriers that play a role in intercellular communication via receptor-mediated endocytosis. Exosomes comprise a bilayer lipid membrane with an aqueous core that includes proteins and genetic material, which makes it possible to load exosomes with hydrophilic and/or lipophilic drugs in vitro or in vivo. Their small size range is likely to contribute to their stability in the circulation, and they do not appear to be removed very rapidly by the circulation, or by passing through vessel walls. Furthermore, the ability of exosomes to target tissues through tissue-specific parent cell-derived markers, and their subsequent ability to cross the plasma membrane of cells, makes them attractive for directed delivery of the cargo. After encapsulation of the anti-inflammatory drug curcumin into exosomes, curcumin had increased solubility and stability in vitro, and increased bioavailability in vivo. An additional benefit was that encapsulated curcumin protected mice from lipopolysaccharide-induced septic shock. Since, curcumin is eﬀective in wound healing, its combination with exosomes, which have also been shown to promote functional skin regeneration, presents a unique opportunity to accelerate tissue regeneration beyond single delivery approaches. Another study showed that a doxorubicin-exosome delivery system resulted in inhibition of breast tumour growth in a nude mouse model. Exosomes also have the potential to deliver drugs across the blood–brain barrier (BBB). The encapsulation of the cancer drugs paclitaxel, doxorubicin and rhodamine into exosomes was performed and the eﬀectiveness of these exosomes for delivery across the BBB was demonstrated in a zebrafish (danio rerio) model. Another potential benefit of encapsulation was an increased cytotoxic eﬀect of paclitaxel in cells, which was shown after loading paclitaxel into prostate cancer cell-derived extracellular vesicles and loading them into autologous prostate cancer cells. Finally, catalase-loaded exosomes showed neuroprotective eﬀects in a mouse model, which demonstrated that exosomes are able to cross the BBB. Recently, exosomes conjugated with c(RGDyK) and loaded with curcumin were intravenously administered into an ischemic brain. Engineered exosomes targeted the lesion area and decreased both the inflammatory response and cellular apoptosis. The presence of mRNA and miRNA in exosomes suggests that these nucleic acids can be incorporated exogenously, transferred to recipient cells, and subsequently influence protein synthesis. Specific targeting can be performed by coupling the exosome with a biological recognition factor, which would interact with cell surface antigens or receptors. The first attempt at gene knockout using siRNA-loaded exosomes was by Alvarez-Erviti et al. This group used rabies virus glycoprotein (RVG)-targeted exosomes to deliver GAPDH siRNA to the brain, resulting in the knockout of the therapeutic target BACE1 in Alzheimer’s disease. Cooper et al. loaded brain-targeting RVG-exosomes with siRNA to alphasynuclein (α-Syn). Following exosome delivery, there were reduced α-Syn aggregates in the brains of mice, which is significant because α-Syn aggregates are associated with the neuropathological features of Parkinson’s disease. Mizrak et al. transfected HEK-293T cells with expression constructs that allowed the generation of secreted microvesicles containing protein–cytosine deaminase (CD) fused to uracil phosphoribosyltransferase (UPRT). The simultaneous injection of the prodrug 5-fluorocytosine, and microvesicles containing CD-UPRT-mRNA/protein, led to a regression in tumour size and decreased tumour growth. Transfecting the human embryonic kidney cell line 293 to produce GE11 peptide, which binds to epidermal growth factor receptor (EGFR) present in human tumours of epithelial origin, allowed Ohno et al. to isolate GE11-positive exosomes that target tumour cells. This group also transfected the cells with miRNA let-7a to obtain let-7a-containing exosomes, which targeted and inhibited tumour development in mice models. Finally, Katakowski et al. isolated exosomes from MSCs transfected with miR-146b. This miR reduces glioma cell motility and invasion, and reduces the expression of EGFR. Following delivery of exosomes loaded with miR-146b, there was a decrease in the growth of 9L glioma cells in vitro, and a reduction of 9L tumour volume in a rat model of a primary brain tumour.
Despite significant investment and progress in regenerative medicine in the last decade, the search for eﬀective approaches to regenerate functional tissues continues. Most synthetic and natural biomaterials including polymers, silk, keratin and ceramic that are used to support tissue regeneration unfortunately often fail to actively and dynamically aid formation of functional tissue structures. The introduction of stem cell therapy aimed to address this problem. Stem cells interact with body structures and respond to all biochemical and biomechanical cues leading to remodelling and regeneration of the tissues. To further enhance or guide cellular function, stem cell therapy is combined with the aforementioned biomaterials. However, in the recent years it becomes very clear that cells communicate and orchestrate tissue regeneration through the manufacture and deployment of paracrine biochemical signals in the form of EVs. Indeed, there is convincing evidence that stem cells without EVs do not yield any beneficial function. These discoveries gave rise to a new paradigm, where EVs are collected from stem cells (and other cells) and are used to actively regenerate tissues. The evidence suggests that EVs promote specific cellular responses through delivering biochemical cargos of RNAs, which are capable of reprogramming recipient cells. Currently, the focus is on noncoding RNAs, i.e. microRNAs, which promote the regeneration of skin, bone, heart, and lung tissue. With the growing library of the functional roles of exosomes carrying specific cargos of microRNAs, applications in regenerative medicine will continue to expand. However, the isolation and characterisation methods need to be standardised, but the field has yet to achieve consensus on these methods. When we achieve the level of matching the specific cargos to desired functional roles, EVs will eventually be a tool to communicate EVerything.
- Extracellular vesicles, exosomes and shedding vesicles in regenerative medicine – a new paradigm for tissue repair I. M. Bjørge, †a S. Y. Kim, †b,c J. F. Mano, a B. Kalionis d and W. Chrzanowski b,c. 14th November 2017, This journal is © The Royal Society of Chemistry 2018.
- Extracellular vesicles: potential roles in regenerative medicine, Olivier G. De Jong, BasW. M.Van Balkom, Raymond M. Schiffelers, CarlijnV. C. Bouten and Marianne C.Verhaar. 03 December 2014, Frontiers in Immunology.
- Hui Jing, Xiaomin He, Jinghao Zheng, Exosomes and regenerative medicine: state of the art and perspectives, Translational Research (2018), https://doi.org/10.1016/j.trsl.2018.01.005.