The Role of Bone Marrow Mesenchymal Stem Cell Derived Extracellular Vesicles (MSC-EVs) in Normal and Abnormal Hematopoiesis and Their Therapeutic Potential

Mesenchymal stem cells (MSCs) represent a heterogeneous cellular population responsible for the support, maintenance, and regulation of normal hematopoietic stem cells (HSCs). In many hematological malignancies, however, MSCs are deregulated and may create an inhibitory microenvironment able to induce the disease initiation and/or progression. MSCs secrete soluble factors including extracellular vesicles (EVs), which may influence the bone marrow (BM) microenvironment via paracrine mechanisms. MSC-derived EVs (MSC-EVs) may even mimic the effects of MSCs from which they originate. Therefore, MSC-EVs contribute to the BM homeostasis but may also display multiple roles in the induction and maintenance of abnormal hematopoiesis. Compared to MSCs, MSC-EVs have been considered a more promising tool for therapeutic purposes including the prevention and treatment of Graft Versus Host Disease (GVHD) following allogenic HSC transplantation (HSCT). There are, however, still unanswered questions such as the molecular and cellular mechanisms associated with the supportive effect of MSC-EVs, the impact of the isolation, purification, large-scale production, storage conditions, MSC source, and donor characteristics on MSC-EV biological effects as well as the optimal dose and safety for clinical usage. This review summarizes the role of MSC-EVs in normal and malignant hematopoiesis and their potential contribution in treating GVHD.


Introduction
The bone marrow (BM) is a dynamic organ composed of hematopoietic stem cells (HSCs), their progeny, endothelial cells, and cells originating from mesenchymal stem cells (MSCs) such as osteoblasts and adipocytes. These cells, in association with the extracellular matrix, organize a specialized microenvironment regulating the formation of mature hematopoietic cells and their proper function.
MSCs reside throughout the body and are characterized by the capacity to self-renew and differentiate into several mesoderm lineages such as chondrocytes, osteocytes, and adipocytes in vivo [1,2]. Many reports have demonstrated that MSCs can differentiate into non-mesodermal

Extracellular Vesicles
EVs are a heterogeneous group of bilayer membrane structures that, according to their size, shape, biogenesis, and composition, are classified into two major categories known as exosomes and micro-vesicles (MVs) [26]. Exosomes are particles of endosomal origin with sizes ranging from 40 to 100 nm in diameter. They are generated from the internal budding of multivesicular bodies and released via exocytosis [27]. MVs are exosome-like vesicles with a size ranging from 50 to 1000 nm in diameter that are released by the budding of the cell membrane [28].
EVs are secreted by a variety of cell types such as MSCs, T cells, B cells, and dendritic cells, and can be isolated from all biological body fluids including serum, blood, breast milk, urine, and semen [29]. They are positive for CD29, CD44, CD73, and CD105 similar to MSCs, but they are also positive for CD81, CD82, CD63, CD53, CD9, and CD37 due to their endosomal origin [30]. EVs can be transported through blood and biological body fluids due to their size and interact with many target cells via surface receptors. Based on their cellular origin and biogenesis, EVs contain different proteins, soluble factors, and microRNAs. The biological effect of EVs is exerted in target cells via both an endocrine effect on distant cells and a paracrine effect on adjacent cells [31]. They are internalized via endocytosis, phagocytosis, pinocytosis, and membrane fusion [32]. Besides their content, EVs' biological effect depends on the functional and metabolic condition of the recipient cells [33].
The secretion of EVs from MSCs is influenced by inflammatory stimuli, stress, high levels of intracellular calcium, and acidic pH both in physiological and pathological conditions [34,35]. In this aspect, several studies have shown that MSC-EVs might be used as a cell-free therapy for cardiovascular, liver and kidney diseases, immunological disorders, wound healing, and tumor inhibition [27,36]. Moreover, EVs are safer as compared to their parental cells because they display low immunogenicity and no tumorigenicity. Additionally, they lack the risk for aneuploidy due to non-self-renewal ability [30]. The isolation, storage, and administration of EVs is much more cost-effective than MSCs.
EVs' ability to transport molecules and to target specific cell populations raises possibilities for their development as pharmaceutical vehicles and sources of diagnostic and prognostic markers. As mentioned above, it is well recognized that cells release two EV-subtypes with different mechanisms of biogenesis and organelle origin [26,27]. If functional differences between EV-subtypes do exist, highly purified vesicle populations are an absolute necessity for the development of EV-based therapies. Sample collection, isolation, and purification of EVs are linked concepts. Therefore, position papers and statements from the International Society for Extracellular Vesicles (ISEV) should be consulted for the most updated guidance on isolation and characterization [37,38]. Similar to the case for cellular therapeutics, EV production on an industrial scale must eventually occur in serum-free conditions to exclude xenogeneic components, and the cellular source must be considered carefully. Since removing serum may change the phenotypic and functional characteristics of cells and the EVs they produce, a culture change would necessitate confirmation that EVs' properties remain the same across media [39,40].
Current practices are to use either chemically-defined media or human platelet lysate (HPL) as a serum replacement. Chemically-defined media may allow a better control of production conditions, which are crucial for industrial-scale manufacturin. Although HPL is already used for the production of functionally efficient cellular therapeutics [41,42], it may contain unidentified pathogenic components that, theoretically, might be spread and hamper global up-scaling strategies [43].
Since EVs may reflect molecular expression patterns and functions of their parental cells, the optimal EV isolation method depends on the intended therapeutic use, route of administration, starting material (milk, plasma, urine, or cell culture), and desired end product (MVs, exosomes or total EVs) [44,45]. In view of the use of EVs as therapeutics agents, it is necessary to develop scalable, reproducible, and good manufacturing practice (GMP)-compliant manufacturing protocols in the context of appropriate regulatory frameworks. For example, the US Food and Drug Administration (FDA) (https://www.fda.gov/drugs) recommends that the route of administration, the acute and repeated dose, the local toxicity studies with histological evaluation, and the route-specific considerations should be taken into consideration when designing EV-based clinical trials [46].

MSC-EVs in Normal and Abnormal Hematopoiesis
The hematopoietic niche and its components construct a biological environment that maintains the homeostasis and responds to stress, damage, or disease conditions. These processes are mediated by chemokines, cytokines, growth factors, metabolites, multiple signaling pathways, and MSC-EVs [47]. EVs that target different cell types and regulate their fate are involved both in normal hematopoiesis and in hematological malignancies (Figure 1).

Roles of MSC-EVs in Normal Hematopoiesis
MSC-EVs may contribute to the activation of quiescent HSCs following different stimuli such as hemorrhage, changes in oxygen concentration, chemotherapy, and irradiation [48]. In vitro studies have demonstrated that MSC-MVs may enhance the proliferation of UC blood CD34 + HSCs to a lesser degree compared to their cellular counterparts by inducing the Wnt/β-catenin signaling pathway that increases the proliferation and inhibits HSC differentiation [7]. Moreover, miRNAs of MSC-EVs induce cell survival and proliferation and inhibit apoptosis or differentiation of all hematopoietic lineages [30]. In a co-culture system, the MSC-EVs miRNAs have been reported to increase the migration of CD34 + HSCs from peripheral blood (PB) to BM niche, via an increase in Cysteine-X-cysteine (CXC) motif chemokine receptor type 4 (CXCR4) expression. However, another study showed that MSC-EVs increase the differentiation of HSCs to myeloid progenitors by interacting with Toll-like receptor 4

Roles of MSC-EVs in Normal Hematopoiesis
MSC-EVs may contribute to the activation of quiescent HSCs following different stimuli such as hemorrhage, changes in oxygen concentration, chemotherapy, and irradiation [48]. In vitro studies have demonstrated that MSC-MVs may enhance the proliferation of UC blood CD34 + HSCs to a lesser degree compared to their cellular counterparts by inducing the Wnt/βcatenin signaling pathway that increases the proliferation and inhibits HSC differentiation [7]. Moreover, miRNAs of MSC-EVs induce cell survival and proliferation and inhibit apoptosis or differentiation of all hematopoietic lineages [30]. In a co-culture system, the MSC-EVs miRNAs have been reported to increase the migration of CD34 + HSCs from peripheral blood (PB) to BM niche, via an increase in Cysteine-X-cysteine (CXC) motif chemokine receptor type 4 (CXCR4) expression. However, another study showed that MSC-EVs increase the differentiation of HSCs to myeloid progenitors by interacting with Toll-like receptor 4 (TLR4) [49]. MSC-EVs were able to decrease radiation injury of murine HSCs by stimulating their proliferation in vitro [50].
MSC-EVs also have immunomodulatory effects akin to their parental cells. They affect the proliferation, polarization, maturation, and migration of macrophages via the production of chemokines, growth factors, and other signaling molecules [53]. MSC-EVs are also involved in MSC-EVs have both pro-angiogenic and anti-angiogenic properties. UC-derived MSCs secrete EVs that promote angiogenesis and MSC proliferation and migration by activating the Wnt/β-catenin signaling pathway [51]. Under hypoxic conditions, EVs highly express VEGF, VEGF Receptor 2 (VEGFR2), angiogenin, and IL6. However, MSC-EVs act on endothelial cells by transfering proangiogenic miR424, miR30c, and miR30b molecules [52].
MSC-EVs also have immunomodulatory effects akin to their parental cells. They affect the proliferation, polarization, maturation, and migration of macrophages via the production of chemokines, growth factors, and other signaling molecules [53]. MSC-EVs are also involved in the disturbance of macrophages 1/macrophages 2 (M1/M2) balance observed in macrophage populations in many pathological conditions by decreasing the expression of proinflammatory signals such as tumor necrosis factor (TNF)-α and IL6 from M1 macrophages and enhancing the expression of Arginase (Arg1) from M2 macrophages [54]. Moreover, the immunoregulatory activity of MSC-EVs is also exerted by impairing dendritic cell maturation, activating neutrophils, inhibiting NK cell proliferation, suppressing B and T cell proliferation, and elevating the number of T regulatory cells [55].
Specific effects of BM-derived MSC-EVs (BM-MSC-EVs) in the homeostasis and maintenance of the BM microenvironment are mediated through abundant miRNA secretion including miR143, miR10b, miR22, miR486, and miR21 [56,57]. More specifically, miR143 has immunomodulatory functions, mir10b and miR22 regulate the differentiation of MSCs, and miR486 and miR21 are involved in MSC proliferation and angiogenic activity [56] (Table 1). BM-MSCs-EVs reduced apoptosis, inhibited differentiation of target cells in vitro and increase engraftment of CD34 + umbilical cord blood cells in in vivo murine BM Differential gene regulation comprising small RNA-target genes in CD34 + cells exposed to BM-MSC-EVs compared with naive CD34 + cells. Reduced caspase 3/7 activity, down-regulation of MPL and ZFP36 and up-regulation of chemotactic factors (IL1b, CSF2, CCL3, GATA2, and CXCR4) were the main molecular mechanisms Additionally, exosomes of both sources altered phosphorylation of p38, p53, and JNK as well as prevented the inhibition of AKT pathway. Exosomes managed to inhibit reduction of Bcl-2 caused by Bortezomib

Roles of MSC-EVs in Hematological Malignancies
The BM microenvironment has been reported to affect the proliferative, self-renewal, and migratory properties of HSCs in a variety of hematological malignancies including myelodysplastic syndromes (MDS), multiple myeloma (MM), and acute myeloid leukemia (AML), among others, which leads to disease onset and/or progression [68]. MSCs in these diseases display altered properties such as growth defects, accelerated senescence, dysregulated osteogenic differentiation, genomic instability, and compromised capacity to support normal hematopoiesis [69][70][71][72]. Apart from the abnormal intrinsic properties, MSCs may influence the BM microenvironment via paracrine mechanisms by secreting soluble factors including EVs [73].
The role of MSC-EVs has been extensively studied in MM. It has been shown that MSC-EVs from healthy donors inhibit the growth of MM cells while their counterparts from MM patients promote the tumor growth [56]. MSC-EVs from MM patients express high levels of IL6, CCL2, and fibronectin and low levels of the tumor suppressor mir15a, which is capable of inhibiting MM cell growth but also to induce apoptosis, which maintains the disease in a stable state [59]. Another study has shown that MSC-derived exosomes activate the AKT pathway and inhibit the p38, p53, and c-Jun N-terminal kinase (JNK) pathway. Therefore, this promotes the survival and proliferation of MM cells [60]. Furthermore, the viability, proliferation, and migration of MM cells has been reported to decrease following treatment with BM-MSC-MVs from healthy donors via the activation of a mitogen-activated protein kinases (MAPK) pathway in contrast to MVs from MM patients [61]. It has also been reported that MSC-derived exosomes from old healthy donors display weaker immunomodulatory effects on MM cell lines in contrast to exosomes from younger donors due to the loss of young BM-MSC exosome-specific miRNAs. It is known that these miRNAs are related to cancer as the development of the majority of cancers is considered to be related to age [62]. Other studies suggest that MM EVs block the differentiation and mineralization of BM-MSCs [64,65].
With regard to AML, it has been reported that exosomes from AML cells contain miR-155, miR-375, and miR-150. These miRNAs regulate the secretion of cytokines and growth factors by BM micro-environment cells as well as the proliferation and migration of HSCs by decreasing CXCR4 expression and affecting the CXCR4/SDF-1 axis, which is fundamental for the retention and differentiation of HSCs in the BM [74]. Additionally, MSC exosomes from AML patients have been reported to protect leukemic cells carrying the fms like tyrosine kinase 3 (FLT3) internal tandem duplication from treatment with the AC220 specific FLT3 inhibitor [65]. Another study demonstrated that AML exosomes promote the leukemic cell survival and proliferation and suppress normal hematopoiesis [66].
Many studies have shown that BM-MSCs are involved in the generation of dysplastic hematopoietic cells, which contribute to disease initiation and evolution [67,75]. Pavlaki et al. have shown a decreased gene expression of the cyclin-dependent kinase inhibitors CDKN1A, CDKN2A, and CDKN2B in BM-MSCs in patients with MDS and an impaired capacity to promote the differentiation of CD34 + cells to myeloid and erythroid lineage. Moreover, BM-MSCs from MDS patients appeared to have a defective osteogenic and adipogenic capacity due to a downregulated canonical WNT (Wnt/β-catenin) expression and upregulated canonical WNT inhibitors [76]. BM-MSCs from MDS-patients are genetically unstable, but this did not affect their proliferative or survival capacity [77]. Another study has shown that there are no differences in the number, the differentiation potential, and the gene expression of proinflammatory or growth-promoting cytokines between the BM-MSCs from MDS patients and healthy donors [78]. However, little is known about the role of MSC-EVs in this disease. To the best of our knowledge, there is one study showing that MSC-MVs from MDS patients modify CD34 + cell properties, promote cell viability and clonogenic capacity, and alter their miRNA and gene expression [67].

MSC-EVs and Graft Versus Host Disease
Although targeted therapies have improved the prognosis of patients with hematological diseases, allogeneic hematopoietic stem cell transplantation (HSCT) remains the only curative therapy for patients with negative prognostic factors and/or refractory disease. GVHD occurs when immune cells transplanted from a genetically non-identical donor recognize recipient allo-antigens, which leads to their activation and, subsequently, organ damage [79].
Various immunosuppressants have been clinically applied for the prevention and treatment of GVHD [80]. However, a substantial proportion of HSCT recipients will likely develop this potentially life-threatening complication. The efficacy of standard primary therapy with corticosteroids is about 50%, and the complete response rate to secondary therapy with a variety of immunosuppressants is about 30% with a median overall survival of less than one year in steroid-refractory patients [81].
A series of clinical studies have examined the efficacy of systemic infusion of culture-expanded BM-MSCs for acute GVHD in allogeneic HSCT-treated patients and the results have shown overall responses ranging from 30% to 80% [82]. Attempts to improve the outcome of BM-MSC therapy have been based on the concept that this form of therapy is dependent on the number of infused cells that can successfully traffic to sites of damaged or diseased tissues. However, systemic administration of an increased number of cells has not augmented the therapeutic effects of BM-MSCs in GVHD [83,84]. In addition, in a study in which BM-MSCs were directly delivered into the gut via the mesenteric artery, the outcome was not more effective than a systemic injection [85]. These clinical results underscore the current hypothesis that the therapeutic effects of BM-MSCs, at least in acute GVHD, are attributed mainly to secreted immune-modulatory factors.
Kordelas, L. et al. showed, for the first time, that infusion of MSC-derived exosomes may significantly improve the symptoms of the steroid-resistant acute GVHD shortly after administration and without significant side effects [86]. Although an in vivo and ex vivo decrease of pro-inflammatory cytokines such as TNF-α, IL1β, and interferon (IFN)-γ was observed following exosomes' infusion, the biologic mechanisms by which BM-MSC-EVs exert their functions and effects remain unknown. Moreover, it has been reported that the therapeutic effect of BM-MSC-derived EVs to GVHD is associated with the inhibition of T cell induction and preservation of circulating naïve T cells [87].
It has also been shown that systemic infusion of human BM-MSC-EVs prolonged the survival of mice with acute GVHD and reduced the pathologic damage in multiple GVHD-targeted organs [88]. In EV-treated GVHD mice, CD4 + and CD8 + T cells were suppressed. Moreover, BM-MSC-EVs seemed to preserve CD4 + CD25 + Foxp3 + regulatory T cell populations. On the contrary, normal human dermal fibroblasts-derived EVs did not ameliorate the clinical or pathological characteristics of acute GVHD in mice, which suggests an immunoregulatory function unique to BM-MSC-EVs.
According to an array analysis, multiple soluble factors associated with the amelioration of GVHD are highly expressed in BM-MSC-EVs, such as CXCL12 [89]. A recent study showed that BM derived MSC-exosomes can effectively ameliorate chronic GVHD in mice by inhibiting the activation and infiltration of CD4 + T cells. Furthermore, MSC-exosomes exhibit immunomodulatory potential by inducing regulatory T cells and inhibiting Th17 cells. These immunosuppressive effects of MSCs-exosomes are mediated, at least in part, through IL-17α and IL-21 [90].
Until now, only one study has indicated that human UC-derived MSC-EVs (UC-MSC-EVs) represent an ideal alternative in the prophylaxis of acute GVHD in a mouse model of allogeneic HSCT by modulating immune responses. Recipients treated with UC-MSC-EVs had a significantly lower number of CD3 + CD8 + T cells, reduced serum levels of IL2, TNF-α, and IFN-γ, a higher ratio of CD3 + CD4 + /CD3 + CD8 + T cells, and higher serum levels of IL10. An in vitro experiment demonstrated that UC-MSC-EVs inhibited the mitogen-induced proliferation of splenocytes in a dose-dependent manner and the cytokine changes were similar to those observed in vivo [91].

Conclusions
A number of biologic functions and effects of MSCs has been shown to be mediated by their EV derivatives. MSC-EVs may influence the HSC and their micro-environment in normal and disease states by transferring their content and mediating anti-malignant or pro-malignant effects through largely unknown and still controversial mechanisms. Beyond the potential roles in the physiology and pathophysiology of hematopoiesis, several studies have also investigated the use of MSC-EVs as potential alternatives to MSCs for improving HSCs expansion and engraftment and for preventing GVHD following HSCT. It is anticipated that the MSC-EV usage may have potential advantages over their cellular counterparts in some aspects. For example, in contrast to MSCs, MSC-EV infusion will not be complicated by the possibility of being trapped in lung capillaries, which might reduce their approach in the damaged tissues/organs and the effectiveness of the treatment nor by the danger of malignant transformation associated with MSC infusion. However, there are still unanswered questions in the pre-clinical level such as the molecular and cellular mechanisms associated with the supportive action of MSC-EVs, the impact of the isolation protocols, the effect of MSCs' source and donor characteristics (i.e., gender and age) on EV biological characteristics and bioactive cargoes, and more [92,93]. Furthermore, the field is open for clinical studies investigating the potential changes of HSCs following MSC-EV infusion, the optimal dosage and safety of MSC-EVs, and the effect of the collection, purification, large-scale production, and storage on patients' outcomes.

Conflicts of Interest:
The authors declare no conflict of interest.