Human mesenchymal stromal cells and derived extracellular vesicles: Translational strategies to increase their proangiogenic potential for the treatment of cardiovascular disease

Abstract Mesenchymal stromal cells (MSCs) offer great potential for the treatment of cardiovascular diseases (CVDs) such as myocardial infarction and heart failure. Studies have revealed that the efficacy of MSCs is mainly attributed to their capacity to secrete numerous trophic factors that promote angiogenesis, inhibit apoptosis, and modulate the immune response. There is growing evidence that MSC‐derived extracellular vesicles (EVs) containing a cargo of lipids, proteins, metabolites, and RNAs play a key role in this paracrine mechanism. In particular, encapsulated microRNAs have been identified as important positive regulators of angiogenesis in pathological settings of insufficient blood supply to the heart, thus opening a new path for the treatment of CVD. In the present review, we discuss the current knowledge related to the proangiogenic potential of MSCs and MSC‐derived EVs as well as methods to enhance their biological activities for improved cardiac tissue repair. Increasing our understanding of mechanisms supporting angiogenesis will help optimize future approaches to CVD intervention.


| INTRODUCTION
In both developed and developing countries, cardiovascular disease (CVD) is a major cause of morbidity and mortality, 1 and most importantly, ischemic heart disease such as acute myocardial infarction (MI) is a leading cause of heart failure. While obstruction to blood flow can be effectively treated by common surgical and catheter-based interventions, achieving cures for microvascular disease remains an elusive goal. The concept of promoting the perfusion of ischemic tissue through angiogenesis has been considered as a highly promising treatment strategy for CVD. Previous attempts to induce neocapillarization in ischemic tissue involved the targeted delivery of various proangiogenic growth factors and nucleic acids encoding them, as well as physical interventions to stimulate angiogenic processes. 2 However, as none of them proved to be sufficiently effective to reverse end-organ ischemia and prevent loss-of-function, other strategies had to be pursued. With the advent of cell therapies for nonhematological F I G U R E 1 Exosome biosynthesis.
(1) Early endosomes are formed by inward budding of the limiting membrane of cells. Surface proteins (orange triangles) may be incorporated into the early endosomal membrane. (2) Early endosomes undergo a maturation process to form late endosomes, in which the biogenesis of exosomes occurs by continuous invagination of the limiting membrane. (3) This particular type of late endosome, which ends up accumulating numerous small intraluminal vesicles with a diameter of 40 to 150 nm is called multivesicular body (MVB). During this process, cytosolic components (eg, miRNAs) are actively packed into the vesicles. In addition, communication with the Golgi apparatus through bidirectional vesicle exchange leads to the incorporation of tetraspanins (blue rectangles) into the membrane of the vesicles. (4) Besides that, cytosolic histone-bound DNA fragments can be transported to MVBs via the autophagosome pathway. (5) Finally, MVBs either fuse with the plasma membrane causing the release of their content into the extracellular environment, or fuse with lysosomes for degradation of their cargo. disorders in the 1990s, the idea of using viable cells to ameliorate or reverse tissue ischemia has rapidly gained traction. 3 Given their ease of isolation, robustness in culture, multilineage differentiation potential in vitro, and partially restricted immunogenicity, 4 mesenchymal stromal cells (MSCs) have been proposed as a promising tool for translational research in cardiology. In recent years, much work has been done to improve the functional properties of MSCs in terms of cell retention and survival of grafted cells, and to elicit their proangiogenic effects. For example, it has been hypothesized that the microenvironment of injured tissue is not conducive for cell engraftment and retention, and that the paracrine effect of transplanted MSCs lasts for only 24 to 48 hours. 5 To overcome these limitations, various scaffolds for cell transplantation were tested and showed promising results for the use in cardiac applications. 6,7 MSC transplantation to repair damage caused by MI and restore cardiac function has been demonstrated in both animal experiments and patients. [8][9][10][11] However, recent metaanalyses failed to show consistent improvement in infarct size or left ventricular function. 12,13 Consequently, the initial assumption that transplanted stem or progenitor cells support neovascularization by differentiation into endothelial cells was soon replaced by the notion of their predominantly paracrine function by producing and secreting small molecules responsible for proangiogenic effects, such as cytokines, chemokines, and growth factors. 14 Besides releasing a variety of soluble factors, MSCs have been shown to secrete extracellular vesicles (EVs) that are important mediators of cell-to-cell communication. 15 Among the known subtypes of EVs, endosome-derived exosomes carrying proteins, metabolites, lipids, and various RNAs have emerged as physiologically relevant components of the MSC secretome 16 (Figure 1). Earlier reports demonstrated that the paracrine activity of the MSC secretome has a therapeutic effect on a wide range of diseases and tissue injury in myocardium, kidney, liver, and lung. [17][18][19][20] The elucidation of paracrine effects thus not only improves our understanding of vascular pathologies, but also enhances the ability to facilitate neocapillarization (ie, endothelial sprouting) for regeneration purposes. In this article, we summarize ways to stimulate angiogenesis with the help of MSCs and their derived EVs, thereby enhancing tissue repair in a variety of pathologies associated with insufficient angiogenesis. We also present the latest advances in the identification of regulatory microRNAs (miRNAs) encapsulated in EVs and discuss their role in promoting angiogenesis.

| ROLE OF MSCs IN ANGIOGENESIS
The human body contains approximately 90 000 km of blood vessels that supply all cells and tissues with vital nutrients and oxygen needed for survival and proliferation. 21 The stimulation of new blood capillary vessel formation through the process of angiogenesis is an integral part of tissue growth and repair. It has been hypothesized that MSCs are part of the perivascular niche in various organs and play an important role in the orchestration of neocapillarization, 22,23 which has rapidly attracted considerable interest in the scientific community. In addition, due to their in vitro multipotent differentiation potential into mesenchymal lineages, including osteoblasts, chondrocytes, myocytes, and adipocytes, the idea was raised that they could also replenish lost tissue in vivo. 24 The therapeutic rationale for MSC treatment, for example, for acute MI patients, is to repair damaged heart tissue by cardiomyocyte differentiation and to provide growth factors to induce angiogenesis, to stimulate resident cardiac stem cell migration and commitment to cardiomyocytes. Most evidence suggests that the beneficial effects of MSCs are mainly caused by the secretion of a variety of bioactive paracrine factors. 25 Especially for bone marrowderived MSCs, numerous small molecules have been demonstrated to induce angiogenesis both in vitro and in vivo; key factors are summarized in Table 1.
Vascular endothelial growth factor (VEGF) and fibroblast growth factor-2 (FGF-2) are two of the most studied factors that regulate angiogenesis. 39 Given that elevated levels can induce cell proliferation and migration of endothelial cells, coordinated regulation of VEGF and FGF-2 expression is required to elicit the proangiogenic effects of MSCs. Another interesting proangiogenic protein is tumor necrosis factor alpha (TNF-alpha), as its effect on angiogenesis depends on the concentration and the duration of treatment. Therefore, it might have a dual role in angiogenesis: high doses of TNF-alpha were found to inhibit angiogenesis in mice in vivo, while low doses promoted it. 40 In addition to classical angiogenic factors, MSCs also secret EVs that carry a variety of biomolecules capable of regulating angiogenesis both in vitro and in vivo. 41,42 EVs were proposed as key agents in the modulation of angiogenesis 43 and have been shown to improve angiogenesis in a number of studies, including mouse and rat models of burn injuries, skin wounds, acute kidney injury, acute MI, and limb ischemia. 44

| ENHANCEMENT OF THE ANGIOGENIC POTENTIAL OF MSCs
MSCs can be obtained from a variety of tissues, such as bone marrow, adipose tissue, and umbilical cord tissue, with bone marrow being the most common stem cell source. 47  Overexpression of HIF-1 alpha also promotes incorporation of Jag-

ged1, a Notch ligand that increases angiogenesis, into MSC-derived
EVs, suggesting that an active HIF-1 alpha phenotype can be transmitted to surrounding cells. 55 In addition, hypoxia-preconditioned MSCs show a higher cell viability, enhanced proliferation potential, decreased production of reactive oxygen species, increased antioxidant glutathione production, and higher superoxide dismutase levels. 56 Other stress conditions that may be of interest for enhancing the angiogenic potential of MSCs include pH variation and calorie restriction. 57 However, given that modification of culture conditions is a rather indirect process for increasing the angiogenic activity of MSCs, as it affects not only one specific molecule but many factors, it may in turn lead to serious side effects. Apart from altering the overall culture environment, several growth factors have been shown to enhance the regenerative capacity of MSCs in vitro. For instance, pretreatment of MSCs with epidermal growth factor or transforming growth factor alpha increased the release of proangiogenic factors such as VEGF and hepatocyte growth factor, which play a central role in inducing angiogenesis and improving oxygen supply to ischemic tissues. 58,59 In addition, it has been shown that MSCs, when cultured on collagen-coated patches, are less fibrogenic and secrete more cardiotrophic factors. 60 Besides modulating culture conditions or using additives, the genetic modification of MSCs was also investigated. 61   that can be administered immediately were considered as a promising option for tissue repair. However, in clinical trials, the overall therapeutic effect was limited, similar to autologous MSCs. 12 Additionally, the use of viable cells still carries inherent risks such as microvasculature obstruction, immune rejection, and proarrhythmic side effects.
EVs derived from MSCs can overcome many of these concerns associated with the use of living cells, while having therapeutic effects similar to those achievable by the originating MSCs themselves. 81 In conclusion, rather than transplanting exogenous MSCs, MSC-derived EVs, even from allogeneic sources, offer a great alternative because they are nonproliferative, less immunogenic, and easier to store and deliver than MSCs. 82 However, as a prerequisite for application, it must be ensured that MSC-derived EVs can be produced in sufficient quantity and quality and that they are able to effectively mediate proregenerative and immunomodulatory effects of the parental cells.  Table 2 shows a selection of known miRNAs with proangiogenic activity.

| PROANGIOGENIC CHARACTERISTICS OF MSC-DERIVED EVS
For instance, with regard to miRNAs incorporated into MSCderived EVs, miRNA-21 activates the protein kinase B/extracellular signal-regulated kinase signaling pathway leading to the overproduction of VEGF. 104 MiR-126 exerts its activity by targeting phosphoinositide-3-kinase regulatory subunit 2 and sprouty-related EVH1 domain containing 1, two negative regulators of VEGF signaling. 122 MiR-130a is a strong positive regulator of angiogenesis because it targets, for example, the antiangiogenic factors growth arrest-specific homeobox and homeobox A5. 125 MiR-135b and miR-31 contribute to angiogenesis by accelerating HIF-1 alpha transcriptional activity via inhibition of factor-inhibiting hypoxia-inducible factor 1, an asparaginyl hydroxylase enzyme that suppresses HIF-1 alpha. 119,129 Likewise, miR-23a directly targets prolyl hydroxylase 1 and 2, leading to HIF-1 alpha stabilization. 109 However, despite the benefits of selective miRNAs for regenerative medicine approaches by inducing angiogenesis, there is a close relationship between vascularity and tumor expansion. 158 For example, the miRNA-1792 cluster has been shown to increase angiogenesis both in vitro and in vivo, and its predominance was observed in a variety of human cancers. 159 Similarly, plasma miRNA-21 levels have been described as a marker for various types of tumors, such as breast, colon, prostate, ovarian, pancreatic, and lung cancer. 160 In addition, since miRNAs do not perfectly complement their target mRNAs, they may target multiple genes whose protein products act on different signaling pathways and thus dysregulate several networks in tumor cells. 161 Given their pivotal role in carcinogenesis, off-target effects of miRNAs should be well characterized before evaluating their use in clinical settings.
Taken together, although aberrant angiogenesis may contribute to pathological conditions, including growth and dissemination of tumors, miRNA application is more welcome for the induction of angiogenesis than its repression. 162,163

| EVS AS VEHICLES FOR THE TARGETED DELIVERY OF PROANGIOGENIC MOLECULES
An EV-based delivery system for proangiogenic factors offers great benefits such as low toxicity, low immunogenicity, high blood circulation stability, biocompatibility, and biological barrier permeability.
Apart from the fact that stress situations, such as hypoxia, can alter the composition of EVs, the loading of EVs with proangiogenic factors is a more specific approach to facilitate angiogenesis. target the ischemic region of the heart, resulting in increased specificity and efficiency of EVs targeting the ischemic myocardium. 166 Another major challenge to the clinical application of EVs is that a high dose is required to improve angiogenesis to a physiologically relevant extent. 167 Although methods for isolating EVs are continuously being developed and optimized, the typical yield of an EV isolation can be less than 1 μg of total EV protein from 1 mL of culture medium, 168 while the therapeutic dose of EVs is normally in the range of 10 to 100 μg of protein in mouse models. 169 In turn, the effective dose in humans could be an order of magnitude or more to compensate for the rapid clearance of EVs from the body. In sum, EVs are promising carriers for proangiogenic molecules, and future efforts should investigate their specific delivery to target organs and the optimal dose.
Other important issues to be addressed are the precise mechanism of action of exogenously administered EVs in vivo, the appropriate time window for EV administration, and the route of administration that achieves maximum efficacy without side effects. 170

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.