Hematopoietic stem cell-derived exosomes have been found to express mRNAs of several pro-angiopoietic and anti-apoptotic factors such as insulin growth factor 1, vascular endothelial growth factor (VEGF), interleukin 8 and basic fibroblast growth factor [22]. These mRNAs have the ability to enhance endothelial cells’ proliferation and survival, exert anti-apoptotic effects and, thereby, stimulate tube formation [22]. Moreover, exosomes derived from a patient’s differentiated hematopoietic stem cells can be used for tissue-targeted cargo delivery through the expression of tissue-specific peptides. By loading miRNA or siRNA of the targeted gene, exosomes can selectively regulate gene expression (Figure 2) [23].

Mesenchymal stem cell (MSC)-derived exosomes have been widely investigated in a variety of disease states and have been shown to be capable of immunoregulation, regenerating tissue and promoting angiogenesis [24]. Employing the advantages of MSC-exosomes, they can be loaded with miRs and targeted to the desired site. Mardpour et al. [24] utilized hydrogel-mediated MSC-derived exosomes loaded in patients with chronic liver failure, managing to augment liver regeneration. An in vivo study, conducted by O’Brien et al. [25], used MSC-derived exosomes loaded with a potential tumor suppressor, miR-379, for breast cancer therapy. It was found that miR-379 indeed acted as a potent tumor suppressor in breast cancer, partly attributed to its interaction with COX-2.

In order to confer cell type targeting sensitivity, modification strategies of exosomes include genetic engineering and chemical modification [26,27,28]. In genetic engineering, the gene sequence of a guiding protein or polypeptide is fused with the preferred exosomal membrane protein. Via this approach, surface display of peptides and proteins is achieved, with, however, a limitation on targeting motifs that are genetically encodable. On the other hand, through chemical modification, a wide range of ligands, natural and synthetic, can be displayed via lipid assembly or conjugation reactions. The latter have the advantage to stably adjust exosomal surface proteins, but the complex exosomal surface may compromise the reaction efficiency [28]. Covalent modification may also risk the function of the exosome. Moreover, lipids and amphipathic molecules can become enclosed in the lipid bilayer of exosomes and allow their hydrophilic part to be displayed on the exterior, ultimately causing an increase in the toxicity of exosomes [28].

Proteomics data (present in databases such as ExoCarta and Vesiclepedia), studied to identify the components of exosomes, have shown that all EPH receptors and ephrinB proteins have been detected in exosomes purified from body fluids and normal cells as well as from a broad range of cancer cell types [29,30,31]. Nevertheless, the possibility of exploiting the EPH system for targeted delivery of therapeutic exosomes either through purification or genetic engineering has not yet been studied.

Targeted delivery systems represent the backbone of personalized medicine to surpass the toxic effects and the seemingly insurmountable obstacle of off-target implications. Emerging data have shed light to the paramount role of exosomes in disease and on their ability to carry proteins to distant tissue locations where they exert their properties [29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46]. Up until now, it was believed that for an EPH to interact with its ligand, a direct cell-to-cell interaction was required. However, Gong et al. [30] demonstrated in their studies that EPH receptors and ephrins were able to have a long-range intercellular communication and, paradoxically, still involve direct contact between two cell membranes [30]. This is possible through the integration of a member of the EPH/ephrin system in the membrane of the exosome. Following the release of the exosome, exosomal membrane-bound EPHs/ephrins can interact with their high affinity counterparts present on cell membranes of cells located at distant sites, a process that results in EPH–ephrin interaction without direct cell-to-cell membrane contact. This form of communication exploits the use of exosomes that are released by cells and are capable of travelling to distant sites via interstitial and other body fluids [8,9]. Therefore, taking into consideration the fact that EPH–ephrin signaling does not require direct cell contact [30], this makes the utilization of exosomal EPHs and ephrins paramount not just as a biomarker of disease progression, prognosis and carcinogenesis but also as a therapeutic approach for a wide variety of diseases.

Several studies have revealed an association between members of the EPH/ephrin system, as a component of cancer-derived exosomes and angiogenesis [42,43,44]. Sato et al. [44] studied the role of small extracellular vesicles (SEVs), in the size range of exosomes, derived from head and neck squamous cell carcinoma (HNSCC) in promoting tumor angiogenesis. Western blot analyses of exosomes and patient data analyses revealed that EPHB2 was overexpressed in HNSCC patients and was associated with poor patient prognosis and tumor angiogenesis. Moreover, functional experiments demonstrated that EPHB2 expression in exosomes modulated angiogenesis. Meanwhile, exosomal EPHB2 stimulated ephrin-B reverse signaling by inducing STAT3 phosphorylation [44]. Additionally, EPHA2 has been implicated in enhancing angiogenesis via proteomic analysis of lung tumor cell-derived exosomes, while an inhibition assay revealed that EPHA2 constitutes a major MAPK activator on exosomes [42]. More specifically, it was demonstrated that the direct communication among membrane protein (EPHA2) on exosomes and recipient cells resulted in stimulation of tumor endothelial cells [42]. These results highlight the fact that EPHA2 participates in angiogenesis as a ligand of the ephrin signaling pathway [42]. Furthermore, Wang H. et al. [43] attempted to provide an association on how oral squamous cell carcinoma (OSCC) affects the angiogenesis of human umbilical vein endothelial cells (HUVECs) via miR-210-3p expression and divulge the relationship between miR-210-3p, its target protein and the mechanism of angiogenesis regulation [43]. They observed that EPHA3 was the target gene of miR-210-3p and that its protein levels could influence the migration and proliferation of HUVECs. In addition, by measuring the levels of phosphorylated AKT in HUVECs, they demonstrated that when EPHA3 was downregulated, the AKT levels were elevated, while when EPHA3 was upregulated, the PI3/AKT pathway was suppressed. When concluding their results, Wang H. et al. proved that the exosomes secreted by OSCC cells could upregulate the expression of miR-210-3p while reducing EPHA3 expression in HUVECs and promoting tube formation via the activation of PI3/AKT signaling pathway [43].

Although cell senescence prevents the proliferation of cells at risk for neoplastic transformation, the altered secretome that senescent cells develop can promote and contribute to cancer cell proliferation. Takasugi et al. [46] demonstrated that sEVs carrying EPHA2 bind to ephrin-A1 that is overexpressed in a variety of cancer cells [8] and ultimately augments their proliferation via the EPHA2/ephrinA1 reverse signaling [46]. They further investigated the mechanism of enrichment of EPHA2 in sEVs, secreted from senescent cells, considering some post-translational modification of EPHA2 to be involved. They examined and concluded that tyrosine kinase phosphorylation of EPHA2 in senescent cells, resulting from oxidative inactivation of PTP1B phosphatase, is involved in its sEV sorting [46]. As a result, that reactive oxygen species (ROS)-regulated cargo sorting into sEVs could prove to be critical for the possibly detrimental growth-promoting effect of the senescent cells’ secretome [46].