Regenerative medicine aims at the functional restoration of a damaged, malfunctioning, or missing tissue. There are three main approaches in regenerative medicine. The first approach is cell-based therapies, where cells are administered to restore a tissue either directly or through paracrine functions. The second approach is often referred to as classical tissue engineering, and consists of the combined use of cells and a bio-degradable scaffold to form a tissue. Lastly, much progress has been made in material-based approaches, which rely on bio-degradable materials, often functionalized with cellular functions.

The first development in replacing malfunctioning tissues was by transplanting organs, tissues, or cells. Over the course of the last century vast improvements were made in the field of transplantation, starting with bone and cornea transplants at the beginning of the twentieth century, followed by the first kidney transplantation in the 1950s (1–3). As transplantation techniques for other organs developed over the following decades, the limiting factor for these procedures shifted from technical limitations to the supply of suitable organs and tissues. Besides shortage in supply, organ and tissue transplantation have another major drawback: the risk of immune rejection and the required chronic immunosuppression treatment.

In response to these issues, research focused on strategies that allow functional restoration of damaged tissues by cell-free approaches or approaches using autologous cell and tissue sources. Embracing the rapid developments in technology and our understanding of biological processes, the field of regenerative medicine is focusing on a wide array of techniques and approaches to restore tissue function. Suitable approaches depend on the function and environment of the newly generated tissue. For instance, in the replacement of insulin-producing cells in patients with type-1 diabetes, there is little need for load-bearing structures, but rather for structures mimicking the extracellular matrix (ECM) like hydrogels, to retain and stimulate insulin-producing cells (4). Heart valve replacements on the other hand require materials that are able to withstand large forces in addition to high flexibility (5), but due to their direct contact with a patients’ circulation also require the use of materials with high hemocompatibility and low immunogenicity. Utilizing autologous stem-, progenitor-, and tissue-specific cells to restore damaged tissues may bypass the problem of immunogenic responses against these implants. Following recent insights that the structural contribution of stem cells to regenerated tissues is limited, and that rather the stimulation of local healing processes plays an important role (6–9), research has increasingly focused on the paracrine hypothesis, investigating the stimulating factors released by these stem- and progenitor cells, including growth factors, cytokines, and extracellular vesicles (EV). At the same time, major breakthroughs in the field of EV have uncovered roles for EV in many processes including angiogenesis, regulation of immune responses, and ECM remodeling (10–13), which may be of specific interest for tissue engineering. Here, we review the recent developments in regenerative medicine and EV research, and discuss potential therapeutic applications of EV in restoring function in damaged tissues.

One of the earliest applications of cell therapy was the administration of cells for the reconstitution of blood or bone marrow (14, 15). As a result of developments during the last decades, including improved techniques in both transient and permanent regulation of gene expression, methods of cell isolation and propagation, and improved protocols to regulate differentiation of cells, cell therapies currently play a prominent role in the field of regenerative medicine (16). Cell therapies can directly aid repair by forming new functional tissues, or support tissue repair through paracrine mechanisms, for instance by secreting growth factors, immunomodulatory molecules, and EV. Examples of direct tissue formation by cell therapy are the use of autologous epithelial cells to repair cornea injuries (17), expansion, and transplantation of chondrocytes in cartilage repair (18), or the administration of endothelial colony-forming cells (ECFC) in a murine hind limb ischemia model to increase neovascularization (19). In these studies cell populations were isolated, expanded ex vivo, and re-introduced at the site of injury to generate new, functional tissues. The ex vivo expansion step allows the use of only limited amounts of tissue and the proper characterization of isolated cells. Adverse effects as dedifferentiation and induction of senescence are great challenges adhered to this approach (20). For instance, in vitro passaging of mesenchymal stem cells (MSC) results in cell enlargement, differentiation, and decrease in proliferation within 10 passages (21), and causes a strong response to micro-environment stiffness, affecting cell morphology, and function (22). Progenitor cells from aged or diseased donors show decreased proliferation, prevalence, as well as functionality (23–25). Despite these challenges, promising results have been achieved, for instance in treatment of patients with severe autoimmune diseases with hematopoetic stem cell transplantation (26).

It has become increasingly apparent that a more supporting role, employed by secretion products of stem and progenitor cells is responsible for many of the observed effects of stem cell therapies (6–9). These paracrine factors secreted by stem- and progenitor cells, like growth factors and cytokines, are of major interest to discover new therapeutics that stimulate local tissue regeneration for the use in tissue engineering as well [reviewed in Ref. (27, 28)].

Repair of damaged tissue requires not only the presence of cells capable of restoring the damaged structure, but requires a microenvironment that promotes appropriate tissue regeneration as well. In addition, cells need to be guided to form a structure of the appropriate size and shape, and in many cases (for instance in bone or cartilage repair, as well as in cardiovascular substitutes), require structural support. In a healthy tissue, the ECM plays a key role in guiding and regulating these processes, whereas in damaged tissue, the ECM is often absent, damaged, or functionally impaired. To address this problem and allow in situ regeneration, structures that (temporarily) provide the requirements for cell retention and tissue regeneration are employed and are referred to as scaffolds.

Scaffolds can either be of natural origin, such as decellularized ECM or modified elastin- or collagen gels, or of synthetic origin, such as synthetic hydrogels or porous polymer scaffolds. Using decellularized ECM from xenogenic or allogenic donors provides scaffolds that are most similar to the natural extracellular environment. Use of decellularized matrices is a promising technique, which yields biocompatible scaffolds with appropriate physical and biological properties. Many ECM components, as well as growth factors, are often conserved and can aid in proper regeneration of functional tissues (29). To decrease the risk of immune responses against antigens in these scaffolds, as well as the potential transfer of pathogens, a combination of enzymatic, physical, and chemical treatments is used to remove cellular components from the tissue (29). Decellularized matrices have been used for tissue engineering of several tissues, including heart valves (30), vascular grafts (31), and trachea (32). However, use of decellularized matrices has several disadvantages. Acquiring and isolating of appropriate tissues, followed by decellularization protocols, can be a relatively time-consuming and expensive procedure, and incomplete decellularization or antigen removal can result in immune reactions against grafts (33). Cell seeding of decellularized matrices can be technically challenging due to structural dimensions and porosity. Furthermore, control over the exact content of the matrices is limited due to donor variation, and despite pretreatment still there exists the risk of transfer of pathogens. In order to create scaffolds in a safe, reproducible, affordable, and controlled manner, extensive research is ongoing on the production of artificial porous scaffolds, exploring various production techniques and materials (34).

Artificial porous scaffolds should meet specific requirements to allow homing of appropriate cell populations. Ideally, a synthetic scaffold temporarily provides the required support and micro-environment, is bio-degradable and eventually replaced by autologous ECM. For cells to be able to migrate or be seeded in the scaffold and allow an environment with proper supply of nutrients, a porous structure is required (35). There are several techniques to generate porous scaffolds, including solvent casting, forming emulsions before polymerization, gas foaming, as well as binding of polymeric fibers by chemical treatment or heating (36–39). Using these techniques in generating scaffolds with consistent porosity in complex shapes, containing areas of varying thickness and materials, is technically challenging. Currently, the most commonly used technique in generating porous synthetic scaffold is electrospinning, which allows the generation of constructs with complex geometry, consisting of combinations of fiber types in both mixed and layered patterns (40). Bio-degradable polymers used in electrospinning include poly(ε-caprolactone) (PCL), poly(glycolic acid) (PGA), poly(hydroxy alkanoate) (PHA), and poly(lactic acid) (PLA). Mixing fiber types in specific patterns allows modulation of degradability, strength, and biological activity of scaffolds (41).

Electrospun scaffolds can be pre-seeded with autologous cells, which may be re-programmed, differentiated, and expanded in vitro, and can then be directly implanted or incubated in a bioreactor until the electrospun meshwork is fully degraded and replaced with ECM (5). Alternatively, scaffolds can be implanted without pre-seeding, allowing in situ recruitment of autologous cells and circumventing the expensive, time-consuming, and challenging process of cell isolation and expansion in vitro. Incorporation of bio-active molecules into the scaffold may be used to recruit proper cell populations, modulate the immune response, or guide cells to differentiate (Figure 1). For instance, ECM-derived peptides like the integrin recognition site peptide Arg-Gly-Asp (RGD) enhance cell adhesion and cell viability in scaffolds (42, 43), whereas coating with type I collagen-mimetic peptide enhances the migration, proliferation, and osteogenic differentiation of MSC (44). Scaffolds can also be designed to release peptides, proteins, or cytokines during degradation, or by coating fibers with a mixture of these bio-active molecules in a bio-degradable substance like fibrin or gelatin. For example, gradual release of vascular endothelial growth factor (VEGF) – a hypoxia-regulated growth factor that plays a key role in angiogenesis – and platelet-derived growth factor (PDGF) promoted endothelialization and smooth muscle cell ingrowth in electrospun scaffolds (45). Release of stromal cell-derived factor (SDF)-1α – a chemokine that is up-regulated in tissue damage and hypoxia, attracts hematopoietic stem cells, and induces endothelial progenitor cell (EPC) recruitment – by electrospun poly(lactic-co-glycolic acid) (PLGA) scaffolds reduced mast cell degranulation, and increased angiogenesis and decreased fibrosis (46). Coating of interposition grafts with SDF-1α combined with the ECM component fibronectin (47), or treatment with VEGF (48) has been reported to enhance graft endothelialization.

Many of the bio-active compounds used in these approaches act as paracrine factors in natural healing processes, or on the secretome of stem- or progenitor cell populations that induce local tissue regeneration in vivo (27). EV constitute a part of the secretome that also play an important role in local induction of tissue regeneration. For example, cardioprotective effects of conditioned medium from MSC in ischemia/reperfusion injury were shown to be mainly mediated by EV (49). Given the previous successes of paracrine factors in tissue engineering, these mediators of intercellular communication could also be of interest in the field of regenerative medicine.

Extracellular vesicles are lipid membrane vesicles, containing a variety of RNA species (including mRNAs, miRNAs), soluble (cytosolic) proteins, and transmembrane proteins presented in the appropriate, and functional orientation (50–52). EV play a role in many processes, including intercellular communication, recycling of membrane proteins and lipids, immune modulation, senescence, angiogenesis, and cellular proliferation and differentiation (10, 13, 52–56). Cells release several types of vesicles with different physiological properties, content, and function, as a result of their different mechanisms of generation, and include exosomes, microvesicles, and apoptotic bodies (57). In the EV research community, a full consensus in terminology and classification of vesicles is yet to be achieved (58). In the past, vesicle nomenclature was mainly based on the tissue of their origin. More recently, the field has started to shift toward a terminology that focuses rather on the mechanisms of generation of these vesicles. Vesicles in the first category, exosomes, originate in multivesicular bodies (MVB) (Figure 2, left). When MVB fuse with the plasma membrane, the intraluminal vesicles are released from the cell and are from thereon referred to as exosomes. Exosomes are reported to be between 40 and 150 nm in size, with a density ranging from 1.09 to 1.18 g/ml. The most common markers used are tetraspanins such as CD9, CD63, CD81, and CD82, lipid raft markers Flotillin-1 and -2, as well as Alix and Tsg101. Other markers that are used are heat shock proteins, MHC molecules, various components of the ESCRT complex and proteins of the Rab protein family (50, 51, 59–61). Microvesicles are shed directly from the plasma membrane and can be a lot larger than exosomes (50–1000 nm) (62). There is, however, an overlap in size between these two populations. Microvesicles also contain mRNAs and miRNAs, as well as soluble and transmembrane proteins. Like exosomes, microvesicles are able to transfer functional genomic and proteomic content to target cells (63, 64). Apoptotic bodies originate at the cell membrane as cells undergo apoptosis. Even though these vesicles are of interest in biomarker research, and have been shown to have effects on other cells, research on these vesicles in intercellular communication is limited (65–67). Furthermore, vesicular cell-derived microparticles with biological functions have been described (68–70). However, most descriptions of microparticles are heterogeneous with regard to the isolated biomaterials or refer to characteristics of non-cell-derived compounds, and depending on the protocols used these microparticles may contain exosomes, microvesicles, apoptotic bodies, or varying combinations of these vesicle populations. Generally, the term EV is used when discussing exosomes or microvesicles, or a combination of these vesicle populations, depending on isolation techniques. However, due to the technical limitations of current isolation techniques, samples may occasionally also contain apoptotic bodies and protein aggregates.

The first report of a cellular function of exosomes was the shedding of the transferrin receptor by maturing reticulocytes (55, 71). Pan and Johnstone showed that removal of this receptor from the cell membrane occurred through endocytosis, followed by formation of intraluminal vesicles (forming the MVB), which were released when the MVB fused with the cell membrane. After this discovery, it was believed that the exosome pathway was mainly involved in cell homeostasis, by secreting cellular waste (72). Not until a study of Raposo et al. for the first time showed an immunological role for exosomes, the stimulation of CD4⁺ T-cells by EBV-transformed B-cells in an antigen specific manner, did researchers begin to explore additional functions (12). Primarily being studied in the context of immunology, exosomes were increasingly considered potential mediators for intercellular communication. However, it was only after the discovery that exosomes are able to transfer functional mRNAs and miRNAs from one cell to another, that the field gained its full momentum (52, 73). Microvesicles have also been reported to transfer functional mRNAs and miRNAs to cells (66, 67).

Extracellular vesicles can communicate with target cells through several mechanisms (Figure 2, right). Firstly, transmembrane proteins on the EV membrane can interact with receptors on the cell membrane. These receptor–ligand interactions can then activate signaling cascades to affect target cells. EV can also fuse with their target cells to release their cargo, either by direct fusion with the cell membrane or by endocytosis, after which mRNAs, miRNAs, and proteins are released into the cytosol. Fusion of EV with target cells can either occur directly at cell membrane, or after endocytosis. After fusion, mRNAs transferred by EVs can be translated in to protein, and delivered miRNAs inhibit mRNA translation and affect cellular processes. The cargo and function of EV depends on their producing cells, and it has been shown that also cellular stress affects EV content, suggesting that intercellular communication through EV is a dynamic system, adapting its “message” depending on the condition of the producing cells (50–52, 74).

Extracellular vesicles are able to affect cell phenotype, recruitment, proliferation, and differentiation in a paracrine manner. These paracrine effects of EV have a potential benefit in regenerative medicine. EV can be incorporated in regenerative therapies, for example by (co-)injection, mixing with hydrogels, or coating scaffolds with EV using fibrin gels or specific linkers (Figure 3). Here, we will discuss the role of EV in essential processes in regenerative medicine: cell viability, immune responses, ECM interaction, and angiogenesis.

Even though, the existence of EV was discovered decades ago, interest in their role as paracrine factor was only relatively recently sparked. Much remains unknown about the pathways that determine the content of EV, and many tissue-specific functions of EV remain to be uncovered. Future studies will provide new insights in EV function and biogenesis, and reveal the roles of proteins and miRNAs in EV function. EV are important components of the secretome involved in intercellular communication, of which content and function can change depending on the conditions of the vesicle producing cells (74, 91, 102–104). Therefore, changes in EV content upon stimulation of producing cells with conditions relevant in development, tissue regeneration, and wound repair may reveal new pathways and insights in intercellular signaling that play key roles in these conditions. Altogether, these qualities make EV an interesting target for the potential discovery of new therapeutics in regenerative medicine.

Over the past decades, it has been shown that EV play a regulatory role, and have modulatory potential, in many biological processes. EV show great potential for therapeutics, biomarker research, and even alternatives to stem-cell-based therapies which rely on paracrine effects. These new approaches have great potential for the support of endogenous repair, including enhancements of existing regenerative medicine approaches. This potential merits further research in the potential of EV, as well the study of new techniques to produce and utilize engineered EV.

Raymond M. Schiffelers is the CSO of Excytex, a company developing tools for extracellular vesicle research. The other co-authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.