State of the field: cellular and exosomal therapeutic approaches in vascular regeneration

Abstract

Pathologies of the vasculature including the microvasculature are often complex in nature, leading to loss of physiological homeostatic regulation of patency and adequate perfusion to match tissue metabolic demands. Microvascular dysfunction is a key underlying element in the majority of pathologies of failing organs and tissues. Contributing pathological factors to this dysfunction include oxidative stress, mitochondrial dysfunction, endoplasmic reticular (ER) stress, endothelial dysfunction, loss of angiogenic potential and vascular density, and greater senescence and apoptosis. In many clinical settings, current pharmacologic strategies use a single or narrow targeted approach to address symptoms of pathology rather than a comprehensive and multifaceted approach to address their root cause. To address this, efforts have been heavily focused on cellular therapies and cell-free therapies (e.g., exosomes) that can tackle the multifaceted etiology of vascular and microvascular dysfunction. In this review, we discuss 1) the state of the field in terms of common therapeutic cell population isolation techniques, their unique characteristics, and their advantages and disadvantages, 2) common molecular mechanisms of cell therapies to restore vascularization and/or vascular function, 3) arguments for and against allogeneic versus autologous applications of cell therapies, 4) emerging strategies to optimize and enhance cell therapies through priming and preconditioning, and, finally, 5) emerging strategies to bolster therapeutic effect. Relevant and recent clinical and animal studies using cellular therapies to restore vascular function or pathologic tissue health by way of improved vascularization are highlighted throughout these sections.

INTRODUCTION

Given the importance of the vasculature, especially the microvasculature in tissue health, it is not surprising that compromised or dysfunctional vessels contribute to many disease conditions. In numerous cases, microvascular repair or regeneration to restore microcirculatory health are necessary components to treat the disease. Instances of microvascular insufficiency, i.e., situations of reduced perfusion due to low vascular densities and/or compromised vasodynamics, contribute to or cause tissue ischemia, nonhealing ulcers, limited tissue metabolism and health, and even complications related to aging (1–8). Loss of microvessel integrity due to insult or disease results in microvascular collapse during tissue inflammation and repair, complicating healing (9, 10).

Microvascular dysfunction in the coronary circulation is often missed by angiography due to the small nature of the vessels but instead can be determined by the coronary flow reserve (CFR), a metric obtained via Doppler ultrasound (11–13). CFR is the ratio of hyperemic coronary flow velocity during an adenosine or dobutamine stress compared with resting coronary flow velocity, and a CFR ≤ 2.5 is considered coronary microvascular dysfunction (CMD) (11, 14). CMD can also be defined as endothelial dysfunction with constriction to acetylcholine, and <20% coronary dilation to nitroglycerin (14). CMD has been correlated to peripheral microvascular dysfunction as well as skin microvascular function with retinal microangiopathy and renal microvascular restrictive index, supporting the current discussion that skin microvascular function should be used as a screening indicator of retinal, renal, and possibly cardiac microvascular functions (15–17). Although skin microvascular function can be measured with a variety of non- to mildly invasive clinical assessments (18), more studies are needed to determine if skin microvascular function is an accurate indicator of target organ microvascular function.

Microvascular regeneration or repair is a multistage process involving progressive changes in microvessel phenotypes and microvascular networks. Importantly, repairs leading to effective perfusion involve not only deriving new microvessels and vessel segments, but the coordinated organization of these elements into a new or existing network that now meets the perfusion, metabolic, and functional demands of the tissue (19–21). Adaptation of the neovessels in the context of this network reorganization is a dynamic, multicell process involving neovessel pruning or remodeling into arterioles, capillaries, or venules (22). Critically important to vascularization outcomes is the establishment of vasoactive regulation of the new microvasculature to enable proper physiological responses to tissue function (1, 2, 23, 24).

Current pharmacological clinical approaches in microvascular pathologies are aimed at treating their symptomatic consequences. For example, in CMD, clinically indicated first line treatment includes β blockers to reduce inotropism to preserve oxygen demand and reduce chest pain as well as enhance vasorelaxation by blocking renin release (25). However, such an approach does not address the pathologic microvessels themselves or their regeneration. For pathologies such as myocardial infarction (MI) or peripheral artery disease, microvascular regeneration is still limited to the experimental realm (26–28). Recombinant vascular endothelial growth factor (VEGF) and other growth factor therapies for microvascular regeneration following MI, for instance, have completed the clinical trials, with the Vascular Endothelial Growth Factor in Ischemia for Vascular Angiogenesis, Euroinject One Trial, and Kuopio Angiogenesis trial reporting nonsignificant effects relative to placebo (19). Given the biological complexity intrinsic to microvascular regeneration, repair, and tissue vascularization, many cell-based approaches are showing greater promise in regenerating or improving microvascular form and function. Here, these cell therapies act mostly in a paracrine factor through secreted factors or exosomes [containing factors, microRNA (miRNA), etc.], or through direct cell interactions such as differentiation, or also through multiple cell interactions such as interactions of donor perivascular niche with endothelial cells (29–32).

There are numerous tissue sources for cells and cell systems capable of regenerating or repairing the microvasculature, each with distinct advantages and disadvantages. Equally varied are the molecular and physiological mechanisms by which these cells elicit a therapeutic benefit. Cellular therapies and cell-free therapies using exosomes or conditioned media have been known to promote angiogenesis, stabilize microvasculatures, and restore vasoactivity to improve vascular densities and perfusion (23, 24, 33). These effects can be accompanied by reduced oxidative stress, normalized mitochondrial and endoplasmic reticular (ER) function, and limited vascular cell senescence and apoptosis (34, 35). The significant potential of vascular and microvascular cell-based therapies are being realized in ongoing and recently completed clinical trials, building on current levels of understanding and highlighting future directions for investigation. The objective of this review is to integrate current progress along with the major issues that span stem cell and vascular research fields, with emphasis on microvascular regeneration where applicable.

STEM CELL SOURCES, UNIQUE CHARACTERISTICS, ADVANTAGES, AND DISADVANTAGES

Stem cells are defined by their ability to self-renew and differentiate. However, their potential to carry out those abilities is variable throughout development and through the lifespan (36). Since their discovery in the early 1960s (37), there has been a race for discovering sources of stem cells that can be used therapeutically, including for microvascular repair. In this section, we briefly detail each of the major sources of stem cells and provide a summary of their methods of isolation (also described in Fig. 1) as well as the unique characteristics, advantages, and disadvantages of each stem cell type (Table 1).

Figure 1.

Cell sources and isolation strategies. Locational sources and strategies for isolation and differentiation of stem cells including bone marrow-derived mesenchymal cells, adipose-derived cells and microvascular fragments, stem cells from birth derived tissues, and embryonic stem cells. Image created with BioRender.com. ADSCs, adipose derived stem cells; BM-MSCs, bone marrow mesenchymal stem cells; CXCR, C-X-C chemokine receptor; FGF, fibroblast growth factor; G-CSF, granulocyte colony stimulating factor; HGF, hepatocyte growth factor; MSCs, mesenchymal stem cells; OPN, osteopontin; SDF-1, stromal cell derived factor-1; SVF, stromal vascular fraction; VEGF, vascular endothelial growth factor.

Table 1.

Advantageous and disadvantageous qualities of stem cell sources

	Advantage	Disadvantage
Bone marrow	Enhanced engraftment and retention of other stem cells Greater transcriptomic heterogeneity (vs. adipose MSC)	Higher immunogenicity (vs. ADSC) Invasive harvesting procedure Lower cell yield (vs. ADSC)Enter senescence earlier (vs. ADSC)
Adipose	Ease of isolationHigher cell yield (vs. BM-MSC) Autologous potential Low immunogenicity (vs. BM-MSC) Higher immunosuppression capacity (vs. BM-MSC) Heterogeneous cell population	Lower transcriptomic heterogeneity (vs. BM-MSC) Poor survival/engraftmentInconsistency among donors/samples
Human umbilical	High concentration of cells Limited senescence during passaging	Genomic/chromosomal tests may be needed to rule out abnormalities
Embryonic	High plasticity Low immunogenicity	Ethical concernsRisk of teratoma formation
Amniotic	High plasticity High concentration of cells	Only available starting second trimester limiting autologous use.
Placental	Chorionic Villi MSC: Available first trimester Ideal for fetal therapeutic applications Limited senescence during passaging	Chorionic Villi MSC: Relatively poor engraftment/retention
Induced pluripotent	High plasticity Higher immunosuppression Multiple locations for harvest/minimally invasive	hiPSC-CM: Poor retention/engraftmentHigh costLong time to procure
Microvascular fragments	Quicker revascularization of target tissue Supplement to other stem cell sources	Lower cell fraction than SVF (macrophages, stem cells, etc.)

ADSCs, adipose derived stem cells; BM-MSCs, bone marrow mesenchymal stem cells; hiPSC-CM, human induced pluripotent stem cell; MSCs, mesenchymal stem cells; SVF, stromal vascular fraction.

Bone Marrow-Derived Stem Cells

The normal function of the bone marrow is to generate hemocytes for circulation in the blood. Beyond hemogenesis, bone marrow contains mesenchymal stem cells (BM-MSCs) that can differentiate into supporting cell types within the blood/bone niche, such as osteoblasts, chondrocytes, and myocytes (38). Once isolated, BM-MSCs can differentiate beyond their biological niche and follow lineages of endothelial cells, myocytes, hepatocytes, epithelial cells, or neural cells generated by coculturing with cytokines during in vitro expansion for cell therapy applications (39).

One of the first methods of isolation entails aspirating the bone marrow from the superior iliac crest under mild anesthesia (conscious sedation followed by local anesthetic) (40, 41). However, a more recent, less invasive method involves mobilizing the BM-MSCs pharmacologically with granulocyte colony-stimulating factor, then harvesting by a peripheral blood draw and isolation with fibrin microbead binding (42). Direct comparison of the concentration of mesenchymal stem cells (MSCs) between the two methods showed that bone aspirate yielded ∼5.25 million cells/mL versus a yield of ∼2.7 million cells/mL using the blood draw method (43). In addition, studies have shown that both methods are similar in terms of cell survival, development of graft versus host disease, and differentiation capacity in vitro, however, the peripheral blood draw method is favored for cell therapy/microvascular repair therapies due to a greater ease of harvest and less patient discomfort (44, 45).

Adipose-Derived Stem Cells

Like BM-MSCs, adipose-derived stem cells (ADSCs) can also serve as a source of multipotent MSCs to be used for cell therapy. Primary benefits relate to the ease in harvesting adipose tissue, usually via lipo-aspiration, a higher cell yield due to the presence of more stem cells in the fat than bone marrow, and less patient discomfort (46). ADSCs are constituents of the stromal vascular fraction (SVF)—a collection of mesenchymal and microvascular cells, including perivascular support cells (47). Enzymatic-based isolation methods to isolate the SVF use collagenase, centrifugation, and filtration, giving a much higher cell yield than mechanical disruption alone. This has led to the development of medical equipment that performs this automated process in the operating suite (48).

SVF contains a nonuniform mixture of cells that work together via directed differentiation and paracrine signaling pathways to enhance revascularization via angiogenesis (49, 50). The heterogeneous population of SVF is composed of MSCs (CD90), endothelial cells (CD31), pericytes (CD146), hematopoietic stem cells (CD45), macrophages (CD11b), B (CD19) and T (CD4) cells, dendritic cells (CD14), and endothelial progenitor cells (EPCs). SVF can be further refined to produce homogeneous proangiogenic cell populations such as adipose-derived mesenchymal stem cells (AD-MSCs) or EPCs (51, 52). Albeit, EPC therapy in-it-of-itself is understudied and is limited by ambiguous definition, overlapping markers, diverse culturing methods, and limited understanding of mechanisms of action (53). Alternatively, therapy with whole SVF is known to increase circulation of host regenerative cells such as EPCs (54, 55). Overall, SVF has been shown to contain a higher stromal cell population than bone marrow (56). In addition, the benefit of having multiple locations for harvesting and yielding more cell populations gives adipose tissue a greater autologous benefit (57). In summary, adipose tissue (specifically SVF) offers an arguably greater benefit for microvascular repair therapies because of its multiple locations available for harvesting, ease of harvest, and heterogeneous cell populations that can help facilitate physiologic vascular remodeling (57, 58).

Embryonic and Induced Pluripotent Stem Cells

Embryonic stem cells (ESCs) are derived from the inner mass of the blastocyst resulting in embryo destruction or from single blastomeres as an embryo-preserving technique that is similar to preimplantation genetic testing routinely used during assisted reproduction. Clinical grade human ESCs are typically derived from donated excess in vitro fertilization cryopreserved embryos and then stored long-term as cell lines. The use of ESCs has been a historically inflammatory subject, particularly in the United States where laws around ESC research vary widely by state. With the development of induced pluripotent stem cells (iPSCs), enthusiasm for ESC research has somewhat faded (59). iPSCs are adult somatic cells that have been induced to pluripotency using the Yamanaka transcription factors. Both ESCs and iPSCs are capable of self-renewal without differentiation or with differentiation into many different cell lineages and/or types (60, 61). Published protocols have demonstrated that iPSCs and ESCs can be coaxed into different lineages, including vascular endothelial cells (62), smooth muscle cells (63), and cardiomyocytes (64, 65).

One example of a protocol to differentiate iPSCs into endothelial cells (iPSC-ECs) was demonstrated by Ikuno et al. (66) using VEGF and cyclic adenosine monophosphate (cAMP) after the iPSCs were directed down a mesodermal germ lineage. iPSC-ECs have been shown to be capable of forming microvascular structures in an in vitro three-dimensional (3-D) culture (60) and in vivo (67). iPSC-ECs and human umbilical vein endothelial cells (HUVECs) were compared for angiogenic potential in 3-D culture and in vitro (mice) to determine the regenerative potential of iPSC-ECs in their current state of therapeutic development (60). At 2 wk in 3-D culture, capillary sprouting was much more abundant in the HUVEC samples whereas the iPSC-ECs had significantly decreased total network length, fewer vessel branch points, and fewer segments formed compared with the HUVEC samples. Inferior angiogenic potential of iPSC versus HUVEC was attributed to lower matrix metalloproteinase-9 (MMP-9) expression and activity, highlighting the necessity for standardization of iPSC production (60). In vitro, both iPSC-EC and HUVEC implants formed vessels containing lumens and were perfused by the host following implantation, as measured by the presence of erythrocytes, at days 7 and 14 (67). By day 7, both groups demonstrated comparable perfusion of the matrices and by day 14 vessel densities were comparable between the groups. The iPSC-ECs had significantly less α-smooth muscle actin from fibroblast-derived pericyte-coated vessels than HUVECs, indicating lower vascular maturity. Further research is needed to overcome the barriers to iPSC-EC translation potential such as improving the maturity of these vessel-like structures as well as investigating whether modifying iPSC-EC MMP-9 expression or activity can boost angiogenic efficiency of iPSCs similar to HUVECs.

A more robust differentiation protocol has been developed by Wang et al. in 2020. Delivery of modified messenger RNA (mRNA) for the erythroblast transformation-specific variant transcription factor 2 during the intermediate mesodermal stage of differentiation allowed vigorous and reproducible differentiation of 13 human iPSC lines into iPSC-ECs with high efficiency, forming perfused vascular networks in vivo that were lined primarily with the human iPSC-ECs. Six-week-old NOD-SCID mice were subdermally implanted with 1 × 10⁶ iPSC-ECs in a collagen 1, fibrin, fibroblast growth factor (FGF), and erythropoietin matrix. After 7 days, the grafts were perfused and were phenotypically, transcriptionally, and functionally comparable with host endothelial cells (68). Effectiveness of iPSC-EC may be further enhanced by methods to facilitate maturation of cell function, as has been done in iPSC-derived cardiomyocytes by optimizing culture conditions such as extracellular matrix or 3-D culture conditions mimicking host environment. Indeed, growth of iPSC-ECs co-cultured with iPSC-derived spinal motor neurons in spinal cord organoid chips leads to increased maturation and transcription of vascular interaction pathways over growth in 96-well plates. Additional studies are necessary to determine appropriate maturation conditions. Overall, the continued development of iPSC-ECs is warranted, as it allows for opportunities to tailor treatment to individual patient-specific conditions, such as iPSC-EC microvascular grafts that produce factor VIII for hemophilia A (69).

One prevailing argument against the use of cellular therapeutics in a clinical setting is the risk of tumor or teratoma formation; however, the risk of tumorigenesis is not the same across all cell types (70, 71). Specifically, the high level of plasticity of ESCs and iPSCs makes them hard to control or predict the lineage after delivery (72, 73) and can result in the generation of teratomas following ESCs or iPSC administration (74). Nevertheless, studies have shown that even this can be combatted by directing ESC lineage in vitro before administration, which leads to reduced teratoma formation. For example, teratomas were not observed in over 200 animals transplanted with human ESC-derived cardiomyocytes when direct differentiation was used to establish a mesodermal or progenitor cardiomyocyte phenotype first (75). Moving forward, strict protocols such as this should be developed and followed to standardize a priming method for these cells that do not risk teratoma formation. Most obviously, the biggest ethical concern comes from the destruction of a viable human embryo in ESC research and clinical application (70, 76). In vitro fertilization produces multiple viable embryos that cannot all be implanted, and extras have been historically donated to research or stored in a biobank as an alternative to being discarded. Since the therapeutic use of ESC may not be viable due to ethical concerns, use of cell sources with similar properties to ESCs should be considered.

Umbilical, Amniotic, and Placental-Derived Stem Cells

Although inducing pluripotency poses a time and cost barrier and use of embryological stem cells is controversial, extra-fetal tissues present a multipotent-ready source of stem cells. These include tissues from the layers of the placental membrane after birth, amniotic fluid at different time points in utero, and the umbilical cord, making extra-fetal tissues relatively free of risks and ethical issues (77). From the umbilical cord, there are several MSC populations (umbilical cord blood-derived mesenchymal stem cell; UCB-MSCs) that can be isolated from the cord, cord lining, subamnion, or Whartons Jelly (78). Amniotic MSCs can be procured at any time during pregnancy, usually during diagnostic amniocentesis, or in some cases, therapeutic amnioreduction with the fluid undergoing centrifugation and culture (79). Placental MSCs are procured from whole organ during delivery or from diagnostic chorionic villus sampling specimens through digestion, centrifugation, and culture and have been shown to enhance angiogenesis in wound healing models after therapeutic delivery (79–81). The use of extra-fetal stem cells may be especially poignant for fetal interventions given their relation, however, their use is also applicable in the adult population considering some of their inherent advantages over adult stem cells (82).

Unique Regenerative Mechanisms, Advantages, and Disadvantages of Various Stem Cells

Although the proposed mechanisms of effect are similar across cell sources, there are a few nuances that distinguish the various stem cell sources. Bone marrow aspiration is an invasive procedure and it leads to a relatively lower cell yield compared with adipose. However, since BM-MSCs were one of the first cell therapy sources discovered they have been more heavily researched (83). On the contrary, ADSC yield is inconsistent among donors but provide a heterogeneous source of supporting cells and MSCs (84, 85). Despite inconsistencies in fat depot volumes and ADSC yields, MSCs from adipose tissue are ∼500 times more (∼5,000 cells/1 g fat) abundant than in BM-MSCs (57, 86). Furthermore, ADSCs enter senescence later than BM-MSCs, possess a greater proliferative capacity, and have superior protein secretion (FGF and insulin-like growth factor) and immunomodulatory effects (87). On the other hand, BM-MSCs show a preferential benefit in differentiating into osteogenic and chondrogenic lineages as well as benefits in proteins secreted [stromal cell derived factor-1 (SDF-1) and hepatocyte growth factor (HGF)] (88). This reveals the importance of selecting the appropriate source for a given clinical application.

To get a better understanding of the genomic landscape of adipose versus bone marrow MSCs, single-cell RNA sequencing data revealed these main differences: ADSCs showed higher immunosuppression capacity, lower transcriptomic heterogeneity, and lower immunogenicity than BM-MSCs, whereas BM-MSCs showed higher levels of metabolic activity, respiration, and oxygen consumption (89). Whether these differences impact effectiveness of microvascular regeneration should be investigated.

Despite the limited number of cells retrieved at the time of harvest, stem cells from bone marrow or adipose tissue can be expanded in culture but often undergo differentiation or enter senescence following 8 wk in culture (90). However, MSCs taken from birth-derived tissues, such as umbilical, placental, and Wharton’s Jelly-derived MSCs have a high proliferative capacity in culture and do not undergo senescence after prolonged passaging (86). Their collection is also noninvasive. One thing to note, though, is the health status of the fetus may not be known; therefore, genomic and chromosomal tests may need to be done to ensure no donor cell abnormalities (78).

Induced pluripotent stem cells have been shown to have similar plasticity to embryonic sources in terms of karyotype, phenotype, telomerase activity, and capacity for differentiation (70). However, they have also been shown to harbor epigenetic changes that can alter their differentiation capacity away from embryonic like cell types toward somatic cell types (91). In addition, iPSCs take a long time to expand and direct their differentiation, and thus are associated with a higher cost (92). Furthermore, iPSCs possess the ability to mobilize and engraft as well, but not along the same homing SDF-1/C-X-C motif chemokine receptor 4 (CXCR4) axis as BM- or AD-MSCs. iPSCs ability to engraft is mediated by integrin-β1 adhesion and requires lineage direction for targeting tissue in vascular therapeutics (93).

Angiogenic mechanisms have been shown to be different between cell sources and should be considered when choosing a source for cellular therapy. In fibrin-angiogenesis models, AD-MSCs use plasminogen activator-plasmin axis by endothelial cells for vessel invasion and elongation, with MMPs serving to regulate capillary diameter. Furthermore, they exhibit upregulation of angiogenic factors urokinase plasminogen activator, HGF, and tumor necrosis factor-α (TNFα) (94). This aspect of AD-MSC angiogenesis is more akin to fibroblast-mediated angiogenesis than BM-MSC-mediated angiogenesis. BM-MSCs execute angiogenesis in part through MMPs without involvement of the plasminogen activator-plasmin axis (95).

Paracrine angiogenic mechanism between cell sources is also known to vary and has been elegantly reviewed by Maacha et al. (96). To briefly summarize, MSC exosomes have been shown to carry proangiogenic miRNAs and differ between sources. MicroRNAs are small-size noncoding RNAs that negatively affect protein expression post-transcription and have recently garnered a lot of attention. Some proangiogenic miRNAs of AD-MSC exosomes include miRNA 148, 532-5p, 378, let-7f, 125a, 31, and 181b, with the latter three repressing antiangiogenic δ like canonical notch ligand 4, hypoxia inducing factor 1 (HIF-1), and transient receptor potential cation channel subfamily M member 7 expression (96). BM-MSC exosomes carry proangiogenic miRNAs 132, 494, 19a, 21a-5p, 210-3p, and 210 with the latter repressing antiangiogenic Efna3 expression in endothelial cells (96, 97). Finally, endometrial MSC exosomes have been shown to carry proangiogenic miRNA-21-5p. In terms of soluble factors, the total angiogenic potential is characterized as highest in Wharton’s Jelly, BM, and placental MSCs versus lower potential in AD- and UCB-MSCs (96). The AD-MSC secretome was shown to be high in VEGF/VEGF-D, insulin-like growth factor 1, interleukin (IL)-8, MMP-3, and MMP-9 and relatively lower in transforming growth factor-β (TGFβ-1), VEGF-A, HGF, bFGF, and angiopoietin-1. BM-MSCs were characterized as high in VEGF-D and lower relatively in VEGF, macrophage colony-stimulating factor, IL-1ra, SDF-1α, MCP-1, IGF-1, IL-8, MMP-3, and MMP-9. Amniotic MSC secretome was characterized as high in VEGF, TGFβ-1, VEGF-A, HGF, bFGF, and angiopoietin-1 (96). These differences may have functional impact for various therapeutic applications. When choosing sources of stem cells, attention should be paid to specific mechanisms of action pertinent to each cell type for the hypothesis in question, as well as the advantages and disadvantages of each cell type.

MECHANISMS OF ACTION FOR STEM CELL REGENERATION OF VASCULAR FUNCTION

Around the early 2000s, the use of stem cells as a cellular therapy for stimulating microvascular growth garnered excitement based on observations of apparent cell adoption of a vascular cell phenotype and mapping to vascular fate. Based on these initial findings, it was thought that stem cells mediated their therapeutic abilities through direct differentiation replacing the injured cells of the host tissue (98). Although direct cell interactions including differentiation are indeed supported (99), the relatively low rate of vascular integration and evidence of little to no incorporation (100) suggested that paracrine mechanism might play a more dominant role. Thus, the prevailing theory today is the paracrine hypothesis, which states that stem cells exert their regenerative effects through secretion of factors that enhance or regulate physiological and molecular processes. Evidence of this notion comes from the ability to use stem cell-conditioned media and yielding the same or similar therapeutic effects. This section explores the specific mechanisms, both direct and indirect (paracrine) of stem cell-mediated microvascular regeneration. Stem cell effects on angiogenesis and immunomodulatory effects will be described in each subsection as relevant and all information is summarized in Fig. 2. Effects of each stem cell source on microvascular regenerative parameters specifically are summarized in Table 2. In further support of expanded impacts, additional examples of stem cell effects on functional dilation, atherosclerosis, mitochondrial function, ER stress, DNA and lipid oxidation, and vascular aging are also provided. In these sections, the focus shifts to the specific stem cell types (e.g., mesenchymal) that dominate the literature. The lack of parallel studies across stem cell sources emphasizes the need for systematic comparisons of stem cell effects for optimizing therapy translation.

Figure 2.

Mechanisms of stem cell-mediated vascular functional regeneration. Mesenchymal stem cell therapies improve vascular health through direct (1) or paracrine (2) mechanisms to rejuvenate vasodilatory function (3), angiogenesis and vascular/capillary density (4 and 5), reduce calcification, smooth muscle proliferation, and thrombosis (6), reduce lipid peroxidation (7), attenuate oxidative stress and restore mitochondrial function (8), abrogate endoplasmic reticulum (ER) stress (9), DNA oxidation and fibrosis (10), apoptosis, and senescence (11). Image created with BioRender.com. eNOS, endothelial nitric oxide synthase; MSC, mesenchymal stem cell; MVB, multivesicular body; ROS, reactive oxygen species; VEGF, vascular endothelial growth factor; VSMC, vascular smooth muscle cell.

Table 2.

Microvascular regenerative effects of stem cell therapies

Study Model	Administrative Route	Pathologic State	Microvascular End Points	Additional Results and Notes	Reference(s)
Bone-marrow MSCs
Mice	Tail IV injection	Stroke	↑ CD31 expression (IHC)	↑ connexin43 gap junctions during revascularization	(101,102)
Rats	Injection to ischemic areaTail IV injectionIntra-arterialIntracavernous injection	MIMICerebral ischemiaErectile dysfunction	Rats treated with MSCs displayed: ↑ CD31 (IHC) ↑ VEGF expression analysis ↑ vascular density (capillary density) ↑ CD31 expression (IHC) ↑ vascular density (IHC) ↑ CD31 expression (IHC and Western blotting) ↑ VEGFA expression (western blotting) αSMA expression (IHC and western blotting)	↑Left ventricular function ↓infarct size ↑ wall thickness ↑ PDGF-β modulation ↓ oxidative stress ↓ inducible NOS ↑ endothelial NOS	(103) (104) (105) (106)
Rabbit	Saline injection into ischemic thigh muscle	Hindlimb ischemia	↑ angiographic score based on image analysis following angiography ↑ capillary density (histology) ↑ limb flow in response to ACh (flowmeter) Differentiation of MSCs into CD31 expressing cells (histology)	↑ Dilation to ACh ↑ Dilation to SNP	(107)
Human (phase I–II)	Trans endocardial injection	Nonischemic dilated cardiomyopathy	↑ endothelial progenitor colony forming units detected in peripheral blood within the allogeneic-MSC treated patients vs. autologous		(108)
Human (phase I–II) (M/F 20–80 yr old)	Intra-arterial	Severe PAD	↓ Transcutaneous pressure of oxygen	↓ rest pain ↓ ulcer size ↓major amputations ↑ pain-free walking distance	(109)
Human(phase I–II) (M/F 18–80 yr old)	Myocardial injection	Myocardial ischemia/ coronary heart disease	↓ CCS angina class Improved Angina Questionnaire scores	↓ hospital admissions ↑ Exercise capacity	(110)
Adipose-derived MSCs
Mice (exosome therapy)	Subcutaneous injection	Wound healing in diabetes	↑ vascular density (IHC)	↑upregulation of mmu_circ_0000250 miRNA in ADSC modified exosomes showed greater vascular density than ADSC exosomes alone	(111)
Rat	Intra-cardiac injection	MI	MSCs, relative to the no cell control, displayed: ↑ CD31 expression (IHC) ↑ microvascular density ↑ VEGF expression (enzyme Assay)	Study compares BM-MSCs, AMSCs, and umbilical cord blood derived (UCBMSCs), as well as their exosomes.	(112)
Swine	Renal artery injection	Renal artery stenosis	MSC treated groups showed: ↑ renal blood flow (Micro-CT) ↑ capillary density ↑ VEGF expression (protein Assay)	The study compared endothelial progenitor cells with adipose MSCs	(113)
iPSCs
Rat (exosome therapy)	Intramuscular injection	Hindlimb ischemia	• ↑ micro-vessels per aortic ring (aortic ring assay) • ↑ CD31 expression (IHC) • ↑ % blood perfusion • ↑ IGFBP-3, PTX3, ALIX (angiogenesis antibody array)	Study compared iVPCs exosome therapy with RAEC exosome therapy and showed greater vessel length in the iVPCs group.	(114)
Birth derived
Mice (human placental MSCs)	Injection of hydrogel	Hindlimb ischemia	MSC treated animals had: largest number of limb salvages $\uparrow$ blood restoration signal intensity (laser doppler perfusion imaging) ↑ CD31 and α-SMA expression (IHC)	No change in EF ↑muscle fibers ↓CD68 expression ↓ collagen (trichrome) largest number of limb salvages	(115)
Rat (endothelial like umbilical cord cells)	Tail vein injection	Aging and diabetes	↓ endothelial dysfunction ↑ CD31, CD34 expression ↑ NO Production ↓ in serum ET-1, ANG II, ROS, and VEGF compared with aged diabetes model	↓SBP, PWV, and PRU than diabetic control ↑ anti-inflammatory and anti-oxidant properties ↓SBP, PWV, and PRU than diabetic control ↓ IL-6, TNF-α, MDA	(116)
Rat (human amnion-derived MSCs)	Injection into the myocardium	Myocardial infarction	homing and expression of VEGF was characteristic of Amnion derived MSCs (IHC)	Heart function of MSC-treated groups were still significantly different than the sham model	(117)
Rat (Human placenta derived MSCs)	Implantation of PCL-MSC vascular graft	Hyperlipidemia	↑ CD31 and αSMA expression (IHC)	↓ NOS reducing ↓ calcification	(118)
Other
Mice (Aortic MSCs compared with BM MSCs)	IV injection	Whole thorax irradiation in lung	MSCs treated animals had: ↓ endothelial damage (flow cytometry, IHC)	Limited immune cell infiltration (flow cytometry)superoxide dismutase mimetic treatment limits endothelial loss of radiation damage	(119)
Mice (MSCs derived from kidneys)	Subcutaneous	Ischemic kidney	MSCs transplanted into Kidney displayed: • ↑ Lectin and CM-DiI dye (IHC)	• ↓ plasma creatinine • ↓TUNEL + nuclei • ↑Ki-67 nuclei	(120)

ACh, acetylcholine; EF, ejection fraction; F, female; IHC, immunohistochemistry; IV, intravenous, iVPCs, induced vascular progenitor cells, M, male; MI, myocardial infarction; MSC, mesenchymal stem cell; NOS, nitric oxide synthase; PAD, peripheral artery disease; PRU, renal artery resistance; PWV, pulse wave velocity; RAEC, rat aortic endothelial cells; SBP, systolic blood pressure; SMA, smooth muscle actin; SNP, sodium nitroprusside.

Direct Cell Interactions

There is evidence that MSCs are capable of direct differentiation into pericyte and endothelial-like cells (121, 122). Differentiation into the latter has been shown through mechanotransductive stimulation such as shear stress and MSC-endothelial extracellular matrix (ECM) protein interactions (99, 123–126). Differentiation also occurs in the presence of VEGF via Rho/ROCK and myocardin-related transcription factor A signaling and CYR61/CNN-1 gene upregulation (32). Nitric oxide released from nearby endothelial cells and S-nitrosoglutathione reductase also mediate differentiation into endothelial-like cells (127, 128). These differentiated cells are considered “like” endothelial cells in that they express a milieu of endothelial cell markers including VEGF receptors 1 and 2 and von Willebrand factor. Notably, differentiation is not complete. For example, as MSC-derived endothelial networks lack other endothelial markers such as platelet endothelial cell adhesion molecule 1 (129). It should also be noted that many of these studies are in vitro, so whether these conditions can cause endothelial-like differentiation in vivo should be determined.

Endothelial-like cells exhibit behavior akin to resident endothelial cells, namely, that they have angiogenic and vasculogenic potential to create new patent cell networks (116, 126, 130). Growth determination factor 11 and TGFβ receptor/ERK/EIF4E pathway further enhance endothelial-like differentiation and angiogenic potential (131). Autologous BM-MSC-derived endothelial-like cells have been used in 3-D bioprinting for the fabrication of carotid and femoral grafts that maintain patency with minimal inflammation when implanted in dogs (132).

These MSC differentiated endothelial-like cells can mediate recovery in the setting of vascular disorders (103, 115, 120). However, direct differentiation and engraftment does not always lead to recovery of organ function. For instance, amnion-derived MSCs differentiated to cardiomyocyte and endothelial-like cells in a rat model of heart failure, but this came without recovery of cardiac function or reduction in fibrosis (117). To note, functional vascular recovery is often achieved despite very low cell therapy engraftment and differentiation (107), leading to the era of the current paracrine hypothesis.

Stem cells can also associate directly with endothelial cells to mediate recovery through juxtacrine interactions (125, 133). Through gap junctions, such as connexin 43, miRNA, glucose, protein, and even mitochondria can be transferred between MSC and endothelial cells (134). In a mouse stroke model, the flux of glucose from endothelial cells to MSC allows for reduced VEGF uptake, angiogenic signaling, permeability, and inflammatory response as a protective recovery mechanism. This mechanism can be blocked by gap junction inhibition (101, 102). Following an ischemia/reperfusion-like injury in HUVECs, mitochondria from MSCs can be directly transferred from tunneling nanotubes rescuing oxygen consumption rate (135). Transfer of the aforementioned organelles is initiated upon MSC recognition of phosphatidylserines present on apoptotic-induced cultured endothelial cells. Other than direct transfer, stem cells can also release factors and mitochondria in a paracrine manner.

Indirect (Paracrine) Cell Interactions

The “paracrine hypothesis” suggests stem cells exert much of their beneficial effect on damaged tissue through the paracrine action of their secretome (i.e., exocytosed contents or vesicles) rather than engraftment of stem cells into damaged tissue (114). Using cell culture systems that prevent direct contact while allowing exchange of factors in the shared media, stem cells can exert regenerative, differentiation induction, and proangiogenic effects via paracrine mechanisms (136, 137). Much of the focus of the field has been on the potential of exosomes as a key therapeutic mechanism as well as an opportunity as a cell-free therapeutic alternative. To reflect this emphasis in the field, exosomes will be the focus in this section as well as highlighting nonexosome (e.g., conditioned media) paracrine effects when appropriate.

Exosomes are membrane-bound vesicles ranging from 30 to 150 nm released from cells that function in a paracrine manner transporting proteins, lipids, RNAs, microRNAs, mitochondria, and more between cells (30, 114, 138). Exosomes are formed through invaginations of the late endosome, where cytoplasmic contents are encapsulated in multivesicular bodies. Proteins from the ER and Golgi complex can be sorted into multivesicular bodies then released to the extracellular environment through fusion and exocytosis where exosomes often circulate to target cells via chemoattraction (139). Cellular recognition of exosomes occurs through either opsonized free floating exosomes, through adhesion, or via antigen recognition (140). Exosomes mediate their effects via intracellular signaling pathways that can be initiated after exosome soluble signaling or juxtacrine signaling. Genetic or protein content transfer can occur via fusion, macropinocytosis, or receptor-/raft-mediated endocytosis, albeit the latter two may also result in lysosomal degeneration as is the case with phagocytosis (139).

The utilization of MSC-derived exosomes circumvents the limited cell numbers as exosomes can be continuously harvested from media of cell cultures and have been shown to have efficacy as a vascular therapeutic. Antioxidant proteins can be released to have a direct restorative effect or miRNAs can indirectly influence host cell function through regulation of genetic expression (111). For example, in Yang et al.’s study, exosomal miRNA-181b-5p (181 b-Exos) increased protein expression of factors that promoted angiogenesis in brain microvascular endothelial cells after oxygen deprivation. Endothelial progenitor cells and AD-MSCs exosomes carry other proangiogenic miRNAs, such as miRNA-126, 296, 278, and 210 (141).

Exosomal therapy shows regenerative potential in vascular settings. Secrotomes isolated from human adipose mesenchymal stem cells increased vascularized granulation tissue and endothelial cell density of the dermis of in vivo mouse wounds compared with controls. Promisingly, injury sites in the group treated with secretomes expressed significantly higher levels of CD31 (a marker of vascular endothelial cells) (142). In vitro treatment with induced vascular progenitor cell (iVPC) exosomes resulted in increased vessel length and area of endothelial cells. In vivo, iVPC-Exo therapy significantly increased perfusion of an ischemic hindlimb model. The results suggest a similar angiogenic effect whether iVPCs or iVPC-Exo are administered, and that the effects of iVPCs are likely due to exosome secretion (114). A comparative study examined the effect of exosomes isolated from BM-MSCs, AD-MSCs, and UCB-MSCs for the treatment of MI in rats. All three types limited damage from induced MIs, stimulated angiogenesis via increased VEGF, bFGF, and HGF, and increased microvascular density, with AD-MSCs having the most significant effect (112). Signaling cascades initiated by the stem cell secretome increased the ability of fibroblasts, keratinocytes, and vascular endothelial cells to migrate. MSC exosomes also have applications for attenuating pathologies associated with vascular barrier dysfunction (143, 144).

Stem cell exosome-based therapy has many advantages compared with cellular therapy. Studies have shown that administration of stem cell-derived exosomes leads to similar outcomes compared with administration of the stem cells themselves (114). In some studies, exosomes even outperform their parent cells for microvascular regeneration (104). Without the stem cell, there is no evidence of tumorigenesis (114) and risk of immunogenicity (145) and cell emboli are greatly reduced. There are also practical advantages to noncell therapy including avoiding complex FDA guidance of cell therapy, simplified large-scale pharmaceutical production, and ease of storage compared with cells. Exosomes may also be used synthetically as drug/gene therapy delivery vehicles, enhancing bioavailability due to their size and have potential for homing to site of interest through targeting peptides on their membranes (146). The major disadvantages to exosomal therapy from a pharmaceutical standpoint is low yield of exosomal release from mammalian cells, cumbersome methodology for isolation, and ensuring batch to batch consistency in exosomal contents (146). Alternative methods that are potentially more amenable to large-scale exosome production are currently being investigated and need to be verified for clinical use in variety of stem cell sources and pathologies (147). Other methods such as changing culturing conditions are described by Phan et al. (138). Evaluation of surface markers is currently done for ensuring batch consistency (148), but this still does not fully ensure that inside contents are uniform throughout.

Stem Cell Remediation of ER Stress

The ER is an important site for protein, lipid and sterol synthesis, protein trafficking and folding, and calcium storage and release. ER stress occurs when the capacity of the ER to fold proteins is outpaced by synthesis of proteins, due to physiologic or pathologic strain (149). In a pig study of renal artery stenosis and in porcine kidney tubular cell culture with induced ER stress (via thapsigargin treatment), AD-MSCs and EPCs were able to both restore renal blood flow and glomerular filtration rate (albeit AD-MSC were more effective) as well as similarly improve microvascular density in the inner and middle cortex (113). EPCs reduced oxidative stress (NADPH oxidase p67and p47 subunits) whereas AD-MSCs reduced inflammation (decreased TNFα and IL-1β, increased IL-10), caspase-3-mediated apoptosis, and ER stress via reduced C/EBP homologous protein (CHOP), GRP94, and Derlin-3. EPCs were also able to increase VEGF whereas both EPCs and AD-MSC reduced p67 and p47 and both decreased fibrosis. Interestingly, this mechanism for MSC repair seems to be contact dependent via a direct cell effect, as culturing with an insert plate to only allow exchange of culture media alone did not replicate these regenerative effects. MSCs can also reduce ER stress via blocking palmitic acid-mediated HUVEC endothelial-to-mesenchymal transition dependent on MSC secretion of stanniocalcin-1 (150).

Stem Cell Rejuvenation of Mitochondrial Function

One major element of stem cell-mediated vascular rejuvenation is antioxidation of host tissue (151). MSCs are highly resistant to oxidative stress since they constitutively express catalase, superoxide dismutase 1–3 (SOD1–3), glutathione peroxidase (GPx), SIRT 1, 3, and 6, and thioredoxin, heme oxygenase-1, as well as having abundant antioxidant molecule glutathione (GSH) and redox sensitive forkhead box O3 signaling (151–156). These antioxidative elements have been shown to mediate recovery from vascular injury and disease in part by reducing oxidative stress and its markers [nitrotyrosine, and 8-hydroxydeoxyguanosine (8-OHdG)] and increasing host endothelial cell/vascular smooth muscle cell (VSMC) antioxidative protein content and/or activity (34, 119, 157). For example, BM-MSCs upregulate aortic heme oxygenase-1 (HO-1) and catalase to reduce oxidative stress in mouse radiation-induced aortic injury, accompanied by reduced fibrosis, aortic thickening, apoptosis, and inflammation (TNFα and intercellular adhesion molecule 1) (158).

Some of these antioxidative factors have been shown to be secreted or trafficked in exosomes to vascular targets, including catalase, SOD1-3, GPx1-7, GSH, SIRT1-6, peroxiredoxin 1–6, and thioredoxin 1–2 (34, 119, 151, 159). Under hypoxic conditions of pulmonary arterial hypertension, MSC exosomes reduced reactive oxygen species (ROS) by increasing GSH/GSSH balance and also positively affected the citric acid cycle by increasing pyruvate dehydrogenase and glutamate dehydrogenase 1, restoring energy balance and oxygen consumption rate in VSMC (160). Furthermore, miRNA-132-3p delivered by MSC exosomes to endothelial cells reduces ROS by downregulating Ras p21 protein activator 1 and increasing Ras and PI3K expression (161). MSC exosomal-mediated reduction in ROS can also be accompanied by increasing endothelial- but reducing inducible-nitric oxide synthase, thus restoring nitric oxide bioavailability (106). Reduced cellular oxidative stress from antioxidative MSCs ameliorates endothelial lipid peroxidation and DNA oxidative damage and maintains mitochondrial DNA replication and stability (162–165). More details about the emerging understanding of the antioxidative role of stem cells are summarized by Stavely and Nurgali (151).

Healthy mitochondria exist in a balance between fission, fusion, biogenesis, and mitophagy, and deviation from these leads to increased oxidative stress, reduced respiration, and vascular function (4). Stem cell therapies have also been shown to mediate shifts in the mitochondrial bioenergetics, dynamics, and respiration. Zhu et al. (166) found that in HUVECs and rat aortas, high glucose insult led to excessive dynamin-related protein-1-mediated mitochondrial fragmentation, ROS production, reduced membrane potential and ATP production, blunted mitophagy leading to increased caspase-3 and Bax-mediated apoptosis. All these observations were reversed when cocultured or infused with BM-MSCs, including a return in mitochondrial morphology to a filamentous fused network. Notably, the authors found an increase in mitophagy mediators Pink1 and Parkin with MSC therapy, small inhibiting RNA (siRNA) inhibition of which ameliorated the restorative effects MSCs. In another study, diabetic endothelial dysfunction was treated with MSC-conditioned media (167). MSC-conditioned media reduced ROS, prevented apoptosis (increased Bcl-2, decreased Bax), increased migration and tubulogenesis in HUVECs subjected to high glucose, and improved vasodilation in thoracic aorta. These effects were mediated by increased expression of SIRT1 and mitochondrial biogenesis activator peroxisome proliferator-activated receptor (PPAR) γ coactivator 1 (PGC-1α), specifically through the phosphoinositide 3-kinase/protein kinase-B (Akt)/SIRT1 and SIRT1/AMP-activated protein kinase/PGC-1α axes.

Mitochondrial functional recovery from MSC therapy can also occur via the direct and paracrine mechanisms through donation of whole mitochondria, mitochondrial membrane fragments, or mitochondrial DNA through tunneling nanotubes, connexin 43 gap junctions, cell fusion, and exosomes/extracellular vesicles (EVs) (105, 135, 165, 168, 169). Apoptotic endothelial cells (HUVECs) exposed to oxygen/glucose deprivation and reoxygenation express phosphatidylserines that serve as a stem cell recognition site for the formation of nanotubules and nearly unidirectional transfer of mitochondria from stem cell to injured endothelial cell (135). In a rat model of stroke (middle cerebral artery occlusion/reperfusion), BM-MSCs were found to increase angiogenesis, motor function, microvascular basal and maximal oxygen consumption rate, and extracellular acidification rate and reduce infarct size and lactic acid production (105). Interestingly, vessel density in the peri-infarct area correlated with host cells that have received BM-MSC mitochondria. Restorative effects were not seen with nanotube inhibition with annexin V or latrunculin-A. Since these inhibitors do not reduce paracrine signaling, the BM-MSCs direct therapeutic mechanism was postulated to be through engraftment and subsequent direct mitochondrial transfer. In coronary artery VSMCs, MSCs nanotube-mediated mitochondrial trafficking is bidirectional and a signal for MSC proliferation (170). With hydrogen peroxide (H₂O₂) exposure, HUVECs transfer their mitochondria to MSCs where they are degraded via autophagy (171). This induces heme oxygenase-1 production and subsequent mitochondrial biogenesis (PGC-1α and mitochondrial transcription factor A) and mitochondrial donation from MSCs to HUVECs. HUVECs then exhibit increased mitofusion 1–2 and reduced apoptosis. The restorative response was dependent on mitochondria being “damaged” (high superoxide) and intact fission and fusion mechanisms. In lipopolysaccharide-induced lung injury model (pulmonary microvascular endothelial cells and in vivo), there is increased ROS with reduced mitochondrial membrane potential, basal and maximal oxygen consumption rate and ATP turnover, and diminished vascular barrier function. This was reversed by MSC-exosomes but not exosomes devoid of mitochondria (165).

Stem Cell Effect on Vascular Dilative Function, Aging, Atherosclerosis, and Hypertension

These antioxidative effects on mitochondrial biogenesis/dynamics/mitophagy confer functional benefits in terms of vasodilative function including that of microvessels. Indeed, it has been shown that mitochondrial state 3 and 4 respiratory function correlates with both acetylcholine and flow-mediated (but not nitric oxide mediated) vasodilation (172). Intravenous SVF delivery reversed aging-associated loss of β-adrenergic agonist (norepinephrine) dilation and improved dobutamine-induced coronary flow reserve, an indicator of coronary microvascular function in a rodent model (1). Myogenic responsiveness to intraluminal pressure changes is improved in mesenteric arteries via an H₂S-mediated mechanism after treatment with UCB-MSCs in mice with endothelial nitric oxide synthase (eNOS) knockout (173). There is improved endothelial-dependent (acetylcholine), but not -independent (nitric oxide) dilation with MSC therapy in rat thoracic aorta (high-glucose model) and during rabbit limb blood flow analysis (limb ischemia model) with improved perfusion (107). Interestingly, both acetylcholine and flow-mediated dilation operate via activation of eNOS and nitric oxide. In a rat model of brain death and ischemia reperfusion, BM-MSC conditioned media mediates a restoration of aortic vasodilatory sensitivity to acetylcholine, no change in maximal endothelium-independent response, and is less sensitive to constrictive phenylephrine (174). Furthermore, these aortas treated with conditioned media exhibited less neutrophilic infiltration and reduced caspase-3, 8, 9, 12, vascular cell adhesion molecule-1, and intercellular adhesion molecule-1 (CD54) expression, and nitrotyrosine (174). Interestingly, the regenerative effect of BM-MSCs on vasodilatory function to flow may be gender specific, as flow-mediated dilation was significantly restored in male but not female patients in the POSEIDEN-DCM study of dilated nonischemic cardiomyopathy (108).

Reduction of endothelium-dependent dilation in aging is associated with oxidative stress-induced senescence and short telomeres, reversed by antioxidant therapy (175). In HUVECs and in aged mice, gingival-derived MSC exosomes attenuate oxidative stress-induced senescence by reducing senescence-associated gene expression of B-galactosidase, p21, p53, phosphorylated γH2AX, and mTOR/pS6 signaling (176). Human iPSC exosomes reverse the effects of senescence (via high glucose insult) on HUVECs. Specifically, they reduce the proportion of senescent cells (senescence-associated B-galactosidase) to increase cell viability, and restore ability to form capillary-like structures (177). MSC exosomes reduce DNA damage, inflammation, fibrosis, and senescence-associated phenotype via ataxia-telangiectasia mutated/p53/p21 signaling in radiation-induced lung endothelial injury via miRNA-214-3p (178).

Aging-associated atherosclerosis and calcification represents another avenue in which stem cell therapy offers regenerative potential. In mouse aortic VSMCs with induced calcification by β-glycerophosphate, MSC-conditioned media reduces alkaline phosphatase activity, reduces intracellular calcium content, mRNA expression of Msx2, Runx2, osteocalcin, bone morphogenic protein-2, protein expression of bone morphogenic protein-2 and Runx2, reduced TNFα, IL-1β, IL-6, caspase-3, and increased Bcl-2/Bax ratio (179). In rat aortic VSMCs exposed to advanced glycation end products with induced calcification, BM-MSC exosomes calcification and ROS production are enhanced via miRNA-146a targeting of thioredoxin-interacting protein (180). In a rat model of hyperlipidemia, a synthetic vascular graft embedded with small EVs from human placenta-derived MSCs reduced calcification, thrombosis, and improved graft patency with M1–M2 macrophage phenotypic switch (118). These EVs contained VEGF, miRNA-126, and miRNA-145 (118). Rat aortic neointimal hyperplasia and vascular smooth muscle proliferation is inhibited by BM-MSC exosomal miRNA-125-b, which represses myo1e expression (181). These effects of stem cells on atherosclerosis and calcification are likely important for microvascular function as well, as coronary artery disease severity is correlated with microvascular dysfunction (182, 183), albeit microvascular disease may precede coronary artery disease (184).

MSC therapy can also mediate microvascular regeneration in the setting of increased hypertension and peripheral resistance. MSC therapy leads to a reduction in both circulating renin-angiotensin-aldosterone and the angiotensin type 1 receptor and increased angiotensin type 2 receptor expression in a chronic renal artery stenosis rat model (185). The reduction in sympathoexcitation reduces systolic blood pressure, albeit it does not lower systolic blood pressure to prestenotic levels. Most importantly, MSC therapy leads to recovery of the microvascular tree in the stenotic kidney (185) allowing for improved glomerular filtration rate and renal plasma flow (186). Human amniotic MSCs reduce portal pressure in portal hypertension alongside recovered microvascular function (187). Furthermore, there was reduced oxidative stress and inflammation and enhanced nitric oxide together culminating in enhanced liver function tests.

AUTOLOGOUS VERSUS ALLOGENEIC (SELF VS. SHELF)

There is current discussion about whether autologous or allogeneic sourcing is most appropriate for stem cell therapy and often referred to as “self or shelf.” Allogeneic cells can be more readily obtained, however, autologous cells better match host biologic and immunologic dynamics. If one were to imagine the “ideal stem cell donor candidate,” its’ properties would include a young person with no prior disease history, healthy lifestyle habits whose stem cells would be at no risk for causing rejection/immune response, maximum potential for engraftment, survival, proliferation, maximal, and consistent secretion of therapeutic factors. Likewise, the ideal stem cell recipient for most successful therapeutic benefit would be someone with minimal disease progression (preventative care) with limited comorbidities and healthy lifestyle habits. In reality, patients often have multiple comorbidities and progressive disease presentation. Similarly, donor stem cells, even from an ideal stem cell donor candidate, can still potentially elicit immune responses (188). How would one choose the universal stem cell donor that these allogeneic cellular therapeutics are based on? What would be the criteria? The dynamics of autologous versus allogeneic strategies are described in this section and compared in Fig. 3.

Figure 3.

Allogeneic vs. autologous stem cell therapeutic strategies in differing patient populations. The advantages and disadvantages of allogeneic (shelf) stem cell therapies are shown with attention paid to donor sex differences (A). For the autologous (self) strategy, effectiveness of stem cell therapies are shown in “ideal” recipients as well as various patient populations such as obesity, smoking, atherosclerosis, diabetes, and aging (B). Image created with BioRender.com. HGF, hepatocyte growth factor; SDF-1, stromal cell derived factor-1; VEGF, vascular endothelial growth factor.

Potentially, allogeneic stem cells could be produced in enormous quantities and readily available on a large pharmaceutical scale for clinical use, i.e., “off the shelf.” Issues of collection, aging, and disease states can be avoided with use of an ideal donor, who may be a young individual without chronic disease history and a healthy lifestyle (189). The challenges of allogeneic stem cell sources include immune reactions to the human leukocyte antigens (HLAs) on donor cells and infection transmission from donor to patient (189). On the other hand, MSCs are relatively hypoimmunogenic (190) and approximately half of international clinical trials from 2004–2018 use allogeneic MSCs as their source (191, 192). The mechanism behind why MSCs exhibit a hypoimmunogenic property is due to their capacity to be conditional antigen presenting cells in the early phase of an immune response (193, 194), then immunomodulate through secretion of soluble factors (195, 196). However, a newer area of research involves CRISPR-Cas9 gene editing to increase immune compatibility of allogeneic stem cells used for therapy (197, 198).

An additional consideration for allogeneic stem cells is sex as a biological variable (199, 200). SVF from male versus female mice exhibits slightly different cellular composition (201). Under oxidative stress conditions, SVF from male donors showcases some differences, including increased IL-33 (inflammatory) and FGF in SVF-HUVEC cocultures, whereas female SVF exhibited increased proangiogenic secretion of platelet-derived growth factor α and β and decreased IL-1β (202). Albeit, SVF antioxidant and VEGF gene and protein expression were similar between the sexes. These differences need to be further explored for any potential differences in therapeutic efficacy, in terms of cell efficacy as well as male to female or female to male stem cell delivery.

Pivoting to autologous stem cells, their use dismisses the immunological concerns of non-self-tissue. However, autologous stem cell use can be complicated by age and disease states that cause cellular dysfunction. Most logically, patients with gene-linked diseases may not receive any benefit via therapy with autologous cells. In addition, the vast number of cells necessary to isolate and expand ex vivo makes collection from autologous donors challenging, especially in states of disease or aging (189).

In a 2018 study by Redondo et al., BM-MSCs derived from patients with progressive multiple sclerosis (MS) were compared with MSCs from controls with the goal of determining whether the inflammatory conditions of MS compromise MSC function. The data showed that the MS-derived MSCs had decreased potential for expansion that was inversely related to the duration of progressive MS. In addition, the MS-derived MSC cells produced markers of early aging in vitro, including accelerated telomere shortening and senescence compared with control (203). In a 2021 study, MSCs obtained from the adipose of pigs with atherosclerotic renal artery stenosis displayed decreased levels of VEGF and capacity for angiogenesis, and increased rates of cellular senescence compared with control cells (204). Hyperglycemia can also influence MSC kinetics. MSCs isolated from abdominal fat of human diabetic donors between the age of 40 and 72 yr showed significantly fewer colony forming units, lower proliferation rates, and a lower number of cells displaying CD105+ (an angiogenic biomarker) compared with cells from nondiabetic age-matched controls (205). The results of these studies suggest that in patients with chronic disease states, autologous stem cells may not be the best option.

Age can also factor in, as adipose-derived SVF from old Fischer-344 rats (24 mo) has been compared with that of young rats (4 mo). In vivo, subcutaneous implants of age-dependent SVF-laden collagen constructs showed that old SVF resulted in decreased total and perfused vessels, and smaller, less mature vessels than those embedded with young SVF after 4 wk of implantation. They also found increased expression of thrombospondin-1, a protein correlated with decreased angiogenesis and migration in old SVF samples. Neutralization of thrombospondin-1 with antibodies before implantation ameliorated this effect (206). Another study revealed a negative correlation in yield of human SVF cells with age (19–71 yr old) in female donors (207).

Other donor characteristics that should be considered are lifestyle and modifiable factors. For example, viable SVF yield has a significant negative correlation to smoking level of the donor (208). ADSCs from smokers has exhibited reduced vasculogenic potential via increased angiostatic activin A expression, reduced ADSC migration and differentiation, reduced platelet-derived growth factor α and β, ANGPT1, ANGPT2, HGF, SDF-1, and increased Notch2, Notch3, and plasminogen activator inhibitor-1 (209). Increased physical activity of patients has also been associated with significantly lower viability of SVF without affecting the initial cell yield (208). Finally, ADSCs isolated from patients of a range of body mass index (BMI) have been compared. As BMI increases, there is an inverse correlation to differentiation potential, proliferation, and colony forming potential (199, 210). Further studies need to be done to ascertain more lifestyle, genetic, or naturally mitigating factors that can nullify a stem cell therapies beneficial affects to a vascular bed.

STRATEGIES TO IMPROVE STEM CELL ENGRAFTMENT AND FUNCTION

Stem cell therapies have been historically limited in their effectiveness by low engraftment and retention, and autologous potential is plagued by imperfections in stem cell function described previously. These described shortcomings of cell sources and therapies, either from autologous or allogeneic sources, give rise to the important discussion of whether these cells can be modified in ways to improve or exceed their initial potential. These limitations can potentially be ameliorated by priming or preconditioning the cells before therapeutic use with the goal of exposing exogenous cells with factors mimicking the potentially harsh environment in which they will be administered, thus improving cell engraftment and therapeutic robustness. Common strategies include hypoxic or ROS preconditioning, pharmacologic priming, gene editing, extracellular supportive methods, and co-therapies with microvascular fragments, all of which will be discussed in the following section and illustrated in Fig. 4.

Figure 4.

Priming, preconditioning, dosing, and administrative strategies to enhance stem cell potency and effectiveness. Optimal therapeutic performance by stem cells can be elicited via hypoxic/ reactive oxygen species (ROS) preconditioning and preconditioning with growth factors and cytokines (1), with genetic modification (2), with pharmacologic intervention (3), extracellular support such as hydrogels (4), or by utilization of novel dosing (5) or administration (6) strategies. Image created with BioRender.com. CXCR, C-X-C chemokine recptor; eNOS, endothelial nitric oxide synthase; HIF-1, hypoxia inducing factor 1; SDF-1, stromal cell derived factor-1.

Preconditioning with Hydrogen Peroxide

Repeated replication of stem cells causes senescence and damage via oxidative stress, similar to the aging of in vivo somatic cells. Reactive oxygen species, and reactive nitrogen species (RNS), are produced during cell expansion primarily via mitochondrial complexes, peroxisomes, and other cell processes. Although low acute levels of ROS are important for normal cellular functions, high chronic levels are capable of causing cellular harm and dysfunction (211). In addition, oxidative stress in the host environment can decrease stem cell engraftment (212).

For these reasons, methods to enhance stem cell resistance to oxidative stress, or conditioning them to be “accustomed” to oxidative stress has been an exciting area of research. One priming technique is preconditioning of stem cells with H₂O₂. For example, H₂O₂-preconditioned ADSC exosomes yielded improved microvascular density in rat skin flap model of ischemia reperfusion with reduced inflammation and apoptosis, increased flap survival, tube length, and blood perfusion units compared with unconditioned ADSC exosomes (213). Another study found that H₂O₂-preconditioning improved survival, migration, proliferation, tissue engraftment, wound closure, and microvascular density in mice with full thickness excisional wounds compared with unconditioned MSCs (214). The mechanism for this was upregulation of cyclin D1, SDF-1, and CXCR4/7 receptors as well as decreased Bax/Bcl-2 ratio, cleaved caspase 3/9, and inhibited p16 and glycogen synthase kinase 3β expression (214). Preconditioning also enhanced oxidative stress resistance, as preconditioned MSCs were able to resist increases in superoxide fluorescence and mitochondrial membrane depolarization during oxidative challenge compared with unconditioned cells (214).

Hypoxic Preconditioning

Hypoxic preconditioning is a common method to induce stem cell resistance to oxidative stress in preparation for therapeutic utilization in ischemic tissue. Hao et al. (215, 216) found that chorionic villi-derived MSCs subjected to hypoxic conditioning exhibited better survival and angiogenic capacity, including reduced caspase-3, activation of Akt signaling, increased VEGFA, HGF, and culminating in increased endothelial cell proliferation, migration, and tube formation. Mouse peripheral blood mononuclear cells subjected to hypoxic preconditioning show increased (compared with nonpreconditioned cells) antioxidant gene expression, lower ROS levels, and higher survivability in oxidative stress conditions and in ischemic hindlimb muscle tissue. This corresponded with greatly improved microvascular density and perfusion 28 days after therapy (217). Antioxidant and cytoprotective genes upregulated included heme oxygenase-1, autocrine motility factor, hexokinase-2, IL-1β, VEGF, and induced nitric oxide synthase (217). Hypoxia preconditioning of MSCs also can upregulate proteins supportive of the angiogenic process, fibronectin 1, E-cadherin, N-cadherin, and many integrins (218, 219). Hypoxia can also upregulate proangiogenic and cytoprotective miRNA expression, such as miRNA-675 (decrease HIF-1α negative regulators to increase VEGF) (220). Hypoxia preconditioned adipose MSCs have also been shown to reverse rat erectile dysfunction by upregulating VEGF and its receptor, angiotensin 1, basic FGF, brain-derived neurotrophic factor, glial cell-derived neurotrophic factor, SDF-1 and CXCR4, and neuronal nitric oxide synthase, culminating in increased angiogenesis (221). Furthermore, hypoxia preconditioned BM-MSCs release more VEGFA and SDF-1α than normoxic, becoming “endothelial-like” cells with angiogenic potential (222). VEGF is also produced more by hypoxia preconditioning of MSCs, enhancing angiogenesis when injected into rabbit molar pulp cavities (223).

Mechanisms behind hypoxia preconditioning increasing VEGF and angiogenic capacity could be wnt-dependent through protein kinase A signaling and HIF-1α-GRP78-Akt axis (31, 224, 225) or through PrP^C-dependent JAK2/STAT3 activation and inactivation of caspase-3 (226). One reason for increased survival after hypoxia preconditioning may be due to enhanced autophagy. In EPCs engrafted in ischemic limb, which exhibited greater microtubule-associated protein 1A/1B light chain 3 stained autophagic structures, restoration of blood perfusion, and enhanced arteriogenesis and angiogenesis after EPC transplant in rat ischemic abdominal wall muscle (227).

Cell-free therapeutic strategies also benefit from hypoxia preconditioning (219, 228). Human AD-MSCs derived EVs cultivated under hypoxic conditions demonstrated improved tube formation of targeted HUVECs compared with normoxic cultivation or supernatant (229). In human UCB-MSC exosomes, angiogenesis, proliferation, and migration are further enhanced by hypoxic preconditioning than nonhypoxic preconditioning in HUVEC cell culture and mouse femur fracture models (230). Meta-analysis has been done comparing hypoxic to normoxic preconditioned exosomes for cardiac repair after MI, but the same analysis has not been done for vascular parameters and would be useful to the field (231).

Finally, hypoxia preconditioning has benefits for tissue engineering. In 3-D bioprinted SVF-vascularized bone grafts, short-term hypoxic preconditioning facilitates microvascular growth with increased VEGFA and HIF-1α expression in vitro and in vivo and inosculation with host (mouse) vasculature (232). Hypoxic preconditioning of primary rat myoblasts caused downregulation of miRNA-1, miRNA-206, and angiopoietin-1 with upregulation of VEGFA with greater percentage volume of blood vessels in a Matrigel construct (233).

Priming with Growth Factors and Cytokines

Use of growth factors and cytokines can also have therapeutic benefit for regenerative stem cell therapy. Dental pulp MSCs secretion of VEGF and HGF as well as vasculogenic ability is enhanced by preconditioning with FGF (223). Priming ADSCs with TNFα has shown a multitude of changes: improved proliferation, cell motility and migration, increased adhesion to monocytes (albeit decreased adhesion to endothelial cells), increased mRNA of VEGF, IL-8, monocyte chemoattractant-1, intercellular adhesion molecule-1, increased ROS generation, and increased positive regulators of angiogenesis (Akt, small GTPase Rac1, ERK1/2, and p38 MAPK) (234). In mouse hindlimb ischemia, TNFα-primed ADSCs increased blood flow recovery, arteriole density, and reduced necrosis (234). The angiogenic properties of human lung mesenchymal stromal cells in wound healing and tube formation assays are enhanced by stimulation with IL-1β through NF-κB dependent-miRNA-433 targeting of Dickkopf Wnt signaling pathway inhibitor 1 and subsequent β-catenin upregulation (235).

Genetically Modified Stem Cells

The capacity for induction of angiogenesis by stem cells can be greatly enhanced by genetic modification. For instance, BM-MSCs overexpressing VEGF-A₁₆₅ leads to higher amounts of VEGF released, homing to hypoxic sites, enhanced angiogenesis, and perfusion over nonexpressed MSCs in limb ischemia models (236–238). Sometimes, a mixed strategy of preconditioning (i.e., hypoxic) and genetic overexpression is developed. Fierro and coworkers (239) explain in their MSC-VEGF formulation that their cells are thawed 48 h before injection to be preconditioned by hypoxia, allowing for 10% of cells to be retained one month after injection, as opposed to 1% without the hypoxic preconditioning.

Telomerase reverse transcriptase (TERT) and/or myocardin overexpressing adipose and bone-marrow MSCs not only increases viability and proliferation with reduced apoptosis, but also increases blood flow and arteriogenesis in mouse ischemic hind limb (240). Although the driving concept for TERT overexpression is to enhance stem cell function, TERT can be enhanced in recipient cells via exosomal mechanisms (241–243), and TERT itself is known to reduce oxidative stress and recover vascular function (244, 245). Therefore, whether stem cells are able to mediate vascular recovery via TERT donation or upregulation is warranted future inquiry.

Overexpression of HIF-1α via lentivirus in rat BM-MSCs leads to exosomes that transferred HIF-1α to hypoxic-injured HUVECs and improved their tube formation, increased VEGF, platelet-derived growth factor, and angiotensin 1 expression, preserved migration, and enhanced proliferation, all more efficiently than nonoverexpressed control exosomes (246). HIF-1α overexpression also leads to increased Jagged1 notch ligand and exosomal production carrying proangiogenic miRNA-15, miRNA-16, miRNA-17, miRNA-31, miRNA-126, miRNA-145, miRNA-320a, miRNA-424, with greater in vivo angiogenesis in mouse subcutaneous Matrigel plug than control nonoverexpressed exosomes (247).

Stem cells can also be made to overexpress miRNA to negatively regulate harmful processes that can overall enhance stem cell regenerative properties. Overexpressing miRNA-126-3p in human UCB-MSCs along with coculture with HUVECs (as well as exosomal treatment) resulted in increased tube formation, migration, and proliferation (248). In a rat tail vein arterialization model, human UCB-MSCs overexpressing miRNA-126-3p exhibited higher and accelerated re-endothelization and reduced inflammation (TNFα) and intimal hyperplasia, all showcasing the ability of miRNA overexpression for treatment of vein graft disease (248). As described in the study, microRNA-126-3p targeted HUVEC SPRED-1 and PIK3R2, negative regulators of VEGF, and then lowered their mRNA and protein concentration, alongside increased Akt and ERK1/2 phosphorylation, explaining their proangiogenic capabilities. On the contrary, some deleterious miRNAs could be targeted in stem cells to improve their function. For example, miRNA-142 is upregulated in aged mouse bone marrow MSC and increases ROS by disrupting pexophagy (autophagy of peroxisomes) via Epas1 (249).

Small interfering RNA (siRNA) can also be used for priming of stem cells. For instance, siRNA against caspase-3 in BM-MSCs allows for greater survival against H₂O₂-mediated apoptosis, showcasing the promise for such a strategy in ischemic and oxidative environments (250). In diabetic mice, BM-MSCs with siRNA targeted against NADPH oxidase 4 showed reduced oxidant levels, decreased adipogenic differentiation, decreased PPARγ, enhanced endothelial differentiation, greater tubular formation, and greater phosphorylated eNOS and nitric oxide bioavailability (251). Transplanted NADPH oxidase 4 siRNA diabetic MSCs into wild-type hindlimb ischemia showed blood flow recovery as well as increased collateral artery diameters and capillary density compared with non-siRNA diabetic MSCs (251).

Manipulation of stem cell EVs, exosomal content, or synthesis of simple synthetic EVs inspired by stem cell contents may be an advantageous strategy for the field, as major hurdles include heterogeneous composition of EVs with potentially deleterious miRNAs such as MSC-derived miR-130b-3p, which has been shown to promote lung cancer cell proliferation, migration, and invasion (252). Found naturally in dental pulp MSCs, microRNA-4732-3p has in vitro and in vivo antioxidant and angiogenic properties (i.e., improved tube formation, migration, and enhanced areal percentage of CD31). MicroRNA-4732-3p was incorporated into synthetic liposomes via electroporation and was delivered intramyocardially to infarcted rats, resulting in preserved cardiac function and reduced scar tissue (252). Therefore, a strategy where miRNAs are first studied and identified in natural exosomal therapeutics and then “mimicked” in synthetic liposomes has the advantages of homogeneity and upscaled copy number, potentially increasing dosage per delivery. Indeed, the synthetic miRNA-4732-3p liposomes outperformed control MSC EVs (252).

Genetically enhanced stem cells can exhibit antioxidative, antisenescent, and anti-inflammatory properties. PGC-1α overexpression enhances angiogenesis in diabetic conditions by protecting against oxidative conditions (253). Embryonic stem cells genetically edited to express active FOXO3 were derived into endothelial cells that led to vascular regeneration in a mouse ischemic injury (254). An intriguing strategy to increase stem cell therapeutic effectiveness relies not only on reducing donor stem cell ROS, but also ROS microenvironment of the host. Zhu et al. (255) used mouse BM-MSCs triple overexpressing SOD1 + glutathione peroxidase (intracellular antioxidants) + SOD3 (extracellular antioxidant), then transplanted into control, diabetic, or diabetic + N-acetylcysteine antioxidant treatment models of critical limb ischemia. Bone marrow MSC survival, proliferation, and migration was significantly enhanced by this treatment strategy, as was release of VEGF, FGF-2, HGF, and placenta growth factor-2, and these effects were mediated in part by loss of ROS (superoxide and hydrogen peroxide) induced advanced glycation end products (255). Survival of BM-MSCs was over 50% after 1–3 wk after injection versus 30%–35% with either overexpression or N-acetylcysteine alone versus 10% without overexpression/N-acetylcysteine treatment. This is significant as survival is typically around 1% for systemic injections and 10% for targeted injection (intracoronary or intramyocardial) (255). In addition, laser Doppler determined that this particular stem cell strategy enhanced blood flow recovery, increased capillary density, and reduced inflammatory infiltration in the ischemic limb (255). This study suggests that perhaps priming/preconditioning of the host themselves would be of benefit for stem cell therapeutic effectiveness.

Pharmacologic Preconditioning

Pharmacologic strategies can also be used to improve the effectiveness of stem cell treatment, often in combination with hypoxia/H₂O₂ preconditioning or with extracellular support. SIRT1 activation is often used in aging conditions due to its ability to enhance survival and respond to stress and inflammatory responses via targeting p53, FOXOs, and NF-κB. Liu et al. used SRT1720 SIRT1 activation combined with hypoxic and H₂O₂ strategy in aged rat and human BM-MSCs for therapeutic application in rat MI. Angiogenesis was greatly increased by this strategy as well as reduced stem cell apoptosis, fibrosis, scar size, and improved cardiac performance. The improved survival was due to upregulated Fas apoptosis inhibitory molecule, of which siRNA targeting ameliorated antiapoptosis (256). SIRT1 also has a role in diabetes, as diabetic rat BM early outgrowth cells exhibit reduced SIRT1 mRNA and reduced tube formation in vitro and neovascularization in vivo. There was also reduced secretion of proangiogenic ELR⁺ CXC chemokines CXCL1, CXCL3, and CXCL5 (257). SRT170 reversed these findings to improve angiogenesis in vitro (257).

ADSCs from aged equines with concomitant metabolic syndrome showcase enhanced ROS, senescence, ER stress, and mitochondrial dysfunction. Preconditioning with resveratrol (SIRT1 activator) and 5-azacytidine (inhibits DNA methylation) reduced ROS, enhanced SOD activity, promoted mitochondrial fusion over fission, restored mitochondrial membrane potential, restored elongated (vs. punctate) morphology, and restored mitophagy (increased PINK). Resveratrol and 5-azacytidine also reduced ER stress (reduced PERK), were anti-inflammatory (reduced TNFα, IL-6, and increased regulatory T-cells), decreased apoptosis, and increased proliferation of ADSCs, and microvesicles from these cells enhance vascularization in horses with suspensory ligament injury (258–260).

Other drugs to improve stem cell-mediated angiogenesis include 5-aza-dc, 2,4-dinitrophenol, and dimethyloxalylglycine (261). 5-Aza-2′-deoxycytidine increased transdifferentiation and endothelial marker expression and angiogenesis on Matrigel of BM stem cells (262). In BM-MSC, 2,4-dinitrophenol (electron transport chain blocker) and dimethyloxalylglycine (prolyl hydrolase inhibitor) increased atrial natriuretic peptide and VEGF, as well as HIF-1α, respectively promoting angiogenesis in rat MI models (263).

Pharmacologic preconditioning is also commonly used for antioxidative or anti-ischemic purposes. Trimetazine (anti-ischemic) lowers tissue demand for oxygen by lowering fatty acid oxidation and reducing ATP demand, making the host microenvironment more amenable for stem cell integration and therapeutic effect. Trimetazine preconditioned BM-MSCs were protected against ROS and had higher survival than nonpreconditioned, and exert enhanced neovascularization, capillary density, and cardiac function in ischemia reperfusion injury in part through enhanced HIF-1α and Bcl-2 (264, 265).

In an in vitro setting, BM-MSCs from diabetic rats preconditioned with antioxidant N-acetylcysteine showed reduced peroxisome proliferator-activated receptor γ, dichlorodihydrofluorescein and nitrotyrosine (reduced ROS), greater adiponectin, greater endothelial cell differentiation (percent CD31+), increased tubules, eNOS, and phosphorylated eNOS, and NO concentration compared with diabetic rat MSCs without preconditioning (251). Iron-chelator deferoxamine improved survival and total antioxidant capacity of human ADSCs secretome and increased VEGF receptor-α and angiopoietin 1 mRNA (266, 267). Mitotempol is a mitochondrial ROS scavenger that was used to precondition dysfunctional diabetic mouse ADSCs, which increased their survival, differentiation potential, migration, and proangiogenic qualities in diabetic mice with critical limb ischemia similar to control (nondiabetic) ADSCs (268). Preconditioning with antioxidant celastrol, in combination with an injectable hydrogel scaffold, led to increased rat and human BM-MSC survival, VEGFA and SDF-1α expression, and neovascular density postimplant into a scratched HUVEC wound closure culture model (269). This demonstrates the potential for a combination strategy with pharmacologic or other preconditioning strategies with extracellular support systems. For more pharmacologic strategies, refer to detailed reviews by Noronha et al. (261) and Miceli et al. (219).

Extracellular Support

It is known that the ability of endothelial cells to adapt to varied tissue environments and vascularize in an organotypic manner is lost after two-dimensional (2-D) culture (270). One strategy to increase engraftment is to administer the stem cells within an extracellular matrix designed to support the cells. For example, hydrogels are physiochemical mimics of the 3-D microenvironment that stem cells naturally inhabit, and therefore may be more encouraging of their natural development and therapeutic milieu. There are many types of hydrogel (natural: alginate, fibrin, collagen, chitosan, hyaluronic acid, etc. and synthetic: polyethylene glycol, methylcellulose etc.), each with advantages and limitations (271). Rat BM-MSCs encapsulated in n-isopropylacrylamide thermosensitive hydrogel further promotes angiogenesis and wound healing in diabetic foot ulcer (272). In a biomimetic pullulan-collagen hydrogel, murine BM-MSCs were seeded and produced greater angiogenic cytokines (VEGF) than in standard culture and showed enhanced viability, engraftment, and angiogenesis with accelerated healing faster than standard culture in murine excisional wound healing model (273). In rat critical limb ischemia, hydrogel with BM-MSCs greatly increased capillary density, number of blood vessels, and capillary to muscle fiber ratio compared with cell therapy or cell-free hydrogel alone (274). A detailed review of hydrogel applications, including advantages and disadvantages of hydrogel and cargo types for critical limb ischemia are provided by Xing et al. (271).

Exosomes can also be loaded into hydrogels. For example, Zhang et al. used UCB-MSC exosomes in hyaluronic acid hydrogel with nanohydroxyapatite poly-ε-caprolactone scaffolding to repair cranial bone defects in rat in part through enhanced vascular recovery. In vitro the exosome hydrogel contributed to enhanced proliferation, migration, and angiogeneic differentiation of endothelial progenitor cells, potentially via miRNA-21 upregulation of the NOTCH1/DLL₄ pathway (275). Hydrogel microenvironments can also influence stem cell characteristics. A newly designed polyethylene glycol maleate citrate and β-tricalcium phosphate hydrogel (with rat BM-MSCs) showed enhanced osteogenic and angiogenic properties in spinal fusion surgery, and the enhanced angiogenic properties were due to effects of the biomaterial in inducing MSC-exosomal angiogenic function (276). One advantage of using an exosome-laden hydrogel is the stable release of exosomes over time. Injectable chitosan-graft-aniline tetramer with exosomes were able to promote angiogenesis with greater tube formation, migration, and proliferation and accelerated diabetic wound healing while also promoting M2 macrophage phenotype (277).

In a 2018 study by Jeong et al., an injectable decellularized matrix (IDM) was combined with human ADSCs. They found that the addition of IDM reduced rates of anoikis (cell death due to loss of extracellular matrix) and upregulated mRNA expression of angiogenic factors in vitro. Conversely, mouse hindlimb ischemic tissue was used to compare IDM + human ADSCs treatment to ADSCs alone and found that engraftment was enhanced in the IDM + ADSCs group. Angiogenic paracrine expression and microvessel density were increased in the IDM + ADSCs group as well (278). A more recent study in 2021 found that gelatin microspheres were embedded with thymosin b4 (Tb4), an actin-sequestering protein known to support cardiac cells and promote vasculogenesis and cell engraftment. Tb4 has also been shown to promote cell survival in hypoxic conditions. Treatment with human iPSC-CMs + Tb4 into induced MIs in porcine increased stem cell engraftment and induced vasculogenesis compared with either treatment alone (279).

Spheroids are another strategy to generate a 3-D environment that encourages aggregation and cell-cell interaction that may better mimic natural conditions and self-assembling. The most common method is the hanging drop method, where MSC are contained in droplets of determined volume on lids carefully placed on culture plates containing media that prevents drop evaporation. MSC in spheroids exhibit further enhanced antioxidative, anti-inflammatory, and antiapoptotic properties such as reduced cleaved caspase-3, TNFα and enhanced catalase, HIF-1α, and superoxide dismutase, compared with 2-D monolayer culture (280). Meta-analysis of 28 studies of MSC in spheroids shows enhanced angiogenic potential from bone marrow, adipose, and UCB-MSC sources across nine of those studies (7 in vitro and 4 in vivo) till April 2019. In vitro, MSC spheroids lead to higher expression of VEGFA, VEGFB, HGF, fibronectin, laminin, fibrinogen, tissue inhibitor of MMP-1, MMP-2, TGFβ₁, FGFb, SDF-1α, HIF-1α, platelet-derived growth factor α and β, among others (280). However, in vivo models of wound healing (mice, AD-MSC), ischemic flap (mice, AD-MSC), renal ischemia-reperfusion (rat, AD-MSC), and hindlimb ischemia (mouse, UCB-MSC) demonstrate that MSC spheroids were associated with higher regenerative capacity for wound closure, cell engraftment, survival, differentiation into endothelial cells, secretion of VEGF, FGF2, intercellular adhesion molecule, vascular adhesion molecule, neuron-glial antigen 2, and HGF along with higher angiogenesis and vascularization versus 2-D culture (280). MSC spheroids can also be preconditioned with hypoxia or drugs such as DMOG, which further enhance oxidative resistance, survival, adhesion, CXC-chemokine receptor 4-homing to injury, and angiogenic potential (281). More comprehensive details on the role for extracellular support and biomaterials for enhanced stem cell contribution to angiogenesis are provided by Lee et al. (282) and Noronha et al. (261).

Microvessel Fragments and Heterogenous Cell Populations

Initial stem cell therapy studies used heterotypic cell populations versus homogeneous cell types. For example, experiments commonly used whole bone marrow versus specific MSC populations. Seemingly motivated by the cell therapy potential and a reductionist approach, the studies then detailed/focused on isolation and characterization of stem cell subtypes. The question remains whether a heterogeneous cell mixture is beneficial. The importance of this question is most provocatively highlighted by recent use of adipose-derived microvascular fragments and SVF. Both cell populations are derived from adipose tissue and SVF can be viewed as a more digested form of the same cell mixture. Importantly, both cell populations contain endothelial cells, smooth muscle cells, pericytes, immune cells, stem cells, and other stromal cells.

When it comes to improving engraftment to restore tissue and vascular function, one promising line of study is co-delivery of the cells of interest with native microvessel fragments (arterioles, venules, and capillaries) harvested from fat tissue. Transplanted microvessels have potential to improve engraftment and maturation of co-transplanted cells [e.g., human iPSC-cardiomyocytes (iPSC-CM)] by several mechanisms, including improved angiogenesis and secretion of proangiogenic factors, anastomoses with host vasculature, immunomodulation [promoting M1–M2 shift in macrophage phenotype (283)], and ability to adapt to varied host tissue environments. Of particular importance for microvessel fragments may be the “perivascular niche,” the idea that vasculature is made up of endothelial and smooth muscle cells, but also contains a microenvironment made of pericytes, immune cells, and regenerative cells (284, 285). These extravascular cells are necessary for native microvascular function, for example, pericytes secrete hepatocyte growth factor to promote vessel sprouting (286). This means that microvessels not only support cotransplanted stem cells; they may in fact contain their own regenerative milieu of cells that encourage angiogenesis and anastomoses, thereby increasing vascular density and perfusion. The emerging dynamics of microvascular fragment therapy is described in greater detail by Laschke et al. (287) and in Fig. 5.

Figure 5.

Microvascular fragment co-therapeutic strategies to enhance stem cell potency and effectiveness. The “perivascular niche” of microvascular fragments allows for effective therapeutic applications in myocardial infarction to reduce scar size, induce angiogenesis, inosculation, and perfusion, with improved cardiac performance. Furthermore, microvascular fragments optimize stem cell therapy when delivered as a co-therapeutic (287,288). Image created with BioRender.com. ESCs, embryonic stem cells; iPSC-CM, induced pluripotent stem cell cardiomyocyte like; MSCs, mesenchymal stem cells.

Stem and immune cells that are physically associated in the vascular wall or perivascular area of microvascular fragments such as endothelial progenitor cells and mesenchymal stromal cells are included when delivering microvascular fragments (287, 289, 290). Since these stem cells are delivered in their native microenvironment with microvascular fragments, they have higher proliferation rate and differentiation potential (291). Microvascular fragments are most easily isolated from adipose tissue via liposuction and limited enzymatic digestion (similar to adipose SVF albeit with a shorter digestion time).

Separately, stromal vascular fractions and microvascular fragments have historically been used for tissue engineered grafts. In a rat MI model, SVF on an implanted vicryl graft preserved microvascular perfusion and heart function due to SVF engraftment into coronary microvessels (23). Nunes et al. (292) found that SVF cells have angiogenic potential, incorporate into neovascular sites, and form perfused microvascular networks that anastomose with host microvascular fragments. These SVF-derived vascular networks also form a functional interface with parenchymal cells i.e., hepatocytes in their liver tissue mimic model. However, as mentioned earlier, isolation of SVF cells eliminates donor microvascular fragments and there is argument that utilization of microvascular fragments as a vascularization unit is more appropriate and efficacious.

Spater et al. (293) compared adipose SVF and microvascular fragments seeded on collagen-glycosaminoglycan scaffolds implanted into full thickness skin defect mouse dorsal skinfold chambers. Cellular composition (endothelial, perivascular, adipocytes, and stem cells) was comparable between SVF and microvascular fragments, albeit viability was higher in the latter with higher microvascular density, collagen count, and better incorporation into host tissue. Microvascular fragments/networks grafts themselves have been used in the setting of MI in mice to improve vascularization via inosculation and improved perfusion. These transplanted microvascular networks inosculate with host around day 7, complete with arterioles, venules, and capillaries containing erythrocytes culminating in reduced left ventricular infarct size and function (288). Considering the advantages of microvascular fragments, it is curious whether their supplementation of cell-based therapies would be of additive benefit.

There is emerging evidence that stem cell therapy can be enhanced by co-delivery with microvessel fragments that promote anastomosis with host vasculature, reducing ischemic conditions when administered alongside stem cell therapy (294). Whole microvessel inclusion is important because it allows quicker perfusion than waiting for them to assemble and mature. In a 2020 study by Sun et al., infarcts in rats were treated with microvessels plus human iPSC-CMs, iPSC-CMs alone, or iPSC-CMs cotransplanted with a suspension of dissociated endothelial cells via direct injection into the heart muscle. When compared with the iPSC-CMs only group, the microvessel plus iPSC-CMs group had significantly increased engraftment and stem cell survival as well as increased graft vascularity. The transplanted microvessels anastomosed with host vasculature and increased graft perfusion in the first week compared with iPSC-CM only group. The microvessels have been shown to secrete proangiogenic paracrine factors that may be the cause of earlier and more robust blood perfusion (294). In addition, cardiac function was improved only in the animals that received microvessels. Moreover, ready-made microvessel fragments allowed a sixfold increase in iPSC-CM survival, reduced scar size, reversed fractional shortening and ejection fraction loss, and twofold increase in vascular area and graft perfusion 5-days post-transplant (295).

In the context of stem cell transplantations in type 1 diabetes, Hiscox et al. (296) found that a engineered mouse islets with microvessel fragments increased islet stability. Very recent research by Aghazadeh et al. used microvessels cotransplanted with pancreatic progenitor cells into the subcutaneous space of the diabetic mice and results showed that a functional connection formed quickly (within 1 wk) leading to a significant decrease in pancreatic progenitor cell death. The progenitor cells developed into β cells and successfully reduced blood glucose levels ∼8 wk after transplantation into the subcutaneous space. This study also demonstrated that the implanted microvessels were better at connecting to host vasculature than single endothelial cells (29).

As adipose-derived microvessel fragments have been found to be rich in resident proangiogenic stromal and MSCs, a study more than a decade ago posed the intriguing question whether fragments from other tissue types would display similar angiogenesis and network formation (297). The investigators chose the microvessel fragments of a unique and highly differentiated tissue, brain cortex microvascular fragments (BMFs), to compare with rodent epididymal fat microvessel fragments (FMFs). They found that the BMFs, like the FMFs, could generate new microcirculation, indicating that the potential of microvessel fragments is not limited to FMFs and is probably independent of the tissue source. Notably, the BMFs formed vessel networks with higher vascular density sooner and with lower permeability (297).

These studies support that microvascular fragments provide an exciting new avenue for regenerative medicine. This is coupled with the ease of isolation and preservation advantages [can be cryopreserved without loss of function (298)]. Microvascular fragments themselves can also be primed, for example with erythropoietin or insulin-like growth factor 1 to increase angiogenic potential (299–301), or with other factors based on specific patient needs, such as hepatocyte growth factor for diabetic recipients (286). Just like stem cells, priming of microvascular fragments from autologous aged patients may be necessary, as vascularization function wanes with aging (302). More focus is necessary in the field on consistent microvascular fragment isolation protocols, specific cellular/fragment balance, advantages/disadvantages of autologous versus allogeneic microvascular fragment inclusion, priming conditions for microvascular fragment and stem cell co-therapy, and which combinations and in what proportions/doses are most efficient and complimentary with various stem cell therapies.

DOSING AND ADMINISTRATION STRATEGIES

What is the Ideal Dose?

Initially used to treat leukemia in the 1950s, there are currently 4,400 active clinical trials using cell-based therapeutics (303) with 900 focused on treating cardiovascular diseases (according to a search on February 2022 on grants.gov). One question being continually studied is the dosing strategy for cell-based therapies due to a deficit in the regulation and standardization of cell therapy doses. A few preclinical studies have begun testing varying doses of cells with studies ranging from administering 1 million to 450 million MSCs to improve cardiac function in a cardiovascular disease state in pigs and sheep (304). Three preclinical studies using a MI model in large animals treated with similar doses of mesenchymal cell therapies reported varying cardiac function results upon follow up. Hamamoto et al. (305) reported that 25 million allogeneic mesenchymal precursor cells directly injected into the epicardium during time of infarct reduced scar size, improved end diastolic, end systolic, and ejection fraction measures in sheep. However, Hashemi et al. (306) administered 24 million allogeneic MSCs via catheter to infarcted region 3 days post MI in pigs which improved only scar size. Both aforementioned results are inconsistent with a study done by Schuleri et al. (307) that showed administering 20 million autologous MSCs directly into the infarct resulted in no improvement in scar formation in a pig MI model. Although two of the studies demonstrated MSCs efficacy in improving cardiac function following MI at set end points, the therapeutic mechanism was not explored. A time-course (308) and 4-wk end point (309) in a rat MI model with intravenously (iv) injected 5 × 10⁶ allogeneic BM-MSCs was explored to examine the mechanistic effects of MSCs in the heart. Both studies showed that intravenously administered MSCs home to the infarcted area and increase capillary density via angiogenesis that correlated to improved heart function (308, 309). The varying results of favorable cardiac outcomes could be due to the degree of dose and method of injection, as well as the source of the cells being used (304).

Unlike pharmaceuticals that can be repeatedly administered as well as at varying concentrations, the use of cell-based therapies in clinical trials is primarily and historically been a one-time administration, therefore the long-term therapeutic potential of cell therapies needs to be examined. A clinical trial using a one-time administration of autologous bone marrow mononuclear cells to treat peripheral artery disease showed significant improvement in the ankle-branchial index and increased skin ulcer healing by 6 mo (109). To further examine long-term efficacy on the vasculature, a clinical trial used autologous BM-MSCs in improving coronary artery disease (110). At a 3-year follow-up, patients receiving the cell therapy reported reduced angina and reduced need for revascularization interventions (110). This data suggests that cell-based therapies are not only safe to use in patients long-term but can promote healing in tissue through angiogenesis. However, it is naive to think that one administration of cells is enough to provide life-long therapeutic remedy. Despite clinical trials reporting improvements long-term, there is little evidence that a majority of administered cells persist past a few months (310). Hong et al. (311) examined this phenomenon in mice after administration of 100,000 cardiac progenitor cells into the coronaries or the myocardium. At 35-days post-administration there were ∼1,000 cells left in the entirety of the heart (311). To mitigate this precipitous decline in cell retention, recent preclinical studies have been examining the efficacy of repeated dosing strategies for cell-based therapies. Guo et al. (312) administered 0, 1, 2, or 3 intraventricular doses, 14 days apart, of 1 million cardiac mesenchymal cells to 3-wk-old MI regions in mice. Mice receiving the three doses of MSCs had improved left ventricular ejection fraction following each administration (312). However, what if a higher dose was administered? It is important to remember that the continual administration of autologous stem-cell therapies is restricted by the quantity available for harvest from a patient as well as if the patient donating or receiving the cell therapy has any pathologies/conditions that would affect the cellular therapeutic. Another important consideration is the patient’s current medication intake, and how these can affect stem cell therapeutic potency and whether dosing or administration strategies should be altered based on patient medication (313). Finally, there is considerable variation in how proper dosing should be defined for exosome-based therapies, such as numbers of exosomes versus total exosomal protein or RNA content (97). It is crucial that accuracy and reproducibility be the driving force alongside intended therapeutic effect in how dosing of cell and cell-based therapies are determined.

How Should Cells Be Administered?

Although the initial consensus that stem cell therapies are safe and several studies mentioned in this review demonstrate therapeutic gains, the results can be variable, possibly due to the utilization of different routes of administration. The most common routes are topical, intravenous, intra-arterial, intramuscular, and intralesional (190). Intravenous administration is the most broadly studied due to the low risk of adverse events, repeated dosing if necessary, and is easier on the patient (310). An early study using a coronary ligation MI model in mice found that intravenous administration of isogenic MSCs 3-h post-ischemia resulted in a reduced infarct size, improved left ventricular function, and increased capillary density 4 wk later (309). These results were corroborated in a 12-wk end point (314). Together, these studies demonstrate that intravenous administration of a cell therapy results in increased angiogenesis and functional improvement to an ischemic injury in as little as 28 days that is sustained 12-wk postadministration. To investigate this repeated dosing via intravenous administration route, Hong et al. used a large animal MI model given a single 150 million dose or repeated (50 million/day for 3 days) doses of human ADSCs 28 days following the ischemic injury. Both treatment groups had improved ejection fraction, coronary flow reserve, and angiogenesis in the peri-infarct area. However, the single dose resulted in a significant decrease in the area of the perfusion defect (315). Shortly after studies demonstrating that intravenous administration of MSCs improved cardiac parameters after injury, in 2007 it was reported that the MSCs homed to the infarcted region of the myocardium (316). In a time-course study using a rat model of chronic MI, it was shown that 24 h following intravenous administration of 5 × 10⁶ MSCs ∼52% of injected cells was retained in the lungs while less than 1% were in the heart (317). However, no MSCs were detected in the lungs or heart at 30 days postinjection (317). Interestingly, despite the same method of administration, varying degrees of cell engraftment and biodistribution have been reported. For example, in an aged rat model, a tail vein injection of 6 million SVF cells resulted in ∼5% of SVF cells engrafting into the myocardium, with the largest portion being retained in the carotid artery (∼50% of total nucleated cells), ∼40% in the aorta, and ∼18% in the lung 28 days postadministration (24). Importantly, with the ever-growing evidence that intravenous administration of a cell therapy results in a majority of the cells getting trapped in the lungs, the possibility of adverse events need to be examined. However, it is important to note that extensive preclinical and clinical trials have shown the safety of stem cell therapies using varying sources (cell type and allogeneic vs. autologous), doses, and route of administration.

As discussed previously, the prevailing hypothesis for the therapeutic mechanism is the paracrine or secretory hypothesis due to repeated studies reporting low engraftment with a steady decline following administration. The biodistribution and the effect MSCs have on other organs leading to improvements in heart function and angiogenesis have been investigated. Luger et al. (318) examined intravenous delivery of human MSCs in immunocompetent mice with acute or chronic MI receiving the cell therapy 24 h or 4 wk postinjury, respectively. In the acute MI and chronic MI models, MSC retention in the heart was markedly low despite being measured shorty after administration. However, both groups showed improved left ventricular function as early as 1 wk following administration of the cell therapeutic. Interestingly, the acute MI model showed a reduction in the number of splenic natural killer cells. Normally following an MI, the natural killer cells residing in the spleen migrate to the site of injury producing a robust proinflammatory response with deleterious effects on healing and repair (319), but the supplementation of MSCs resulted in low splenic and cardiac natural killer cell numbers over the course of the study (10 wk) (318). MSCs also reduced the number of splenic and cardiac neutrophils (318) that can orchestrate a macrophage switch to a healing phenotype (320). This phenomenon was demonstrated by Dayan et al. (321) when left ventricular function was improved at 2 and 4 wk following an administration of human UC-MSCs 48 h after coronary artery ligation in MOD/SCID mice, which equated to a reduced number of macrophages in the myocardium with the remaining macrophages expressed as a pro-reparative and anti-inflammatory phenotype (M1 to M2). The following data have been corroborated using AD-MSCs (322). Mounting evidence suggests that despite a wide biodistribution from intravenous cell injection, there is a beneficial effect on organs with low engraftment (such as the heart), possibly due to cell therapy regulating the host’s immune cells via homing in lymphoid organs such as the spleen (310).

Many studies have been developing methods in conjunction with intravenous administration to increase cell engraftment/homing to the area of interest. A technique termed, “ultrasound targeted microbubble destruction” (UTMD), was developed in the early 2000s and uses a bioactive substance incorporated into microbubbles that can then be injected into a patient’s blood stream. Following injection, an ultrasound wand is placed over the area of interest where the sound waves burst the bubbles, thus promoting accumulation as a targeted therapeutic (323). Bone marrow mononuclear cells combined with UTMD in a rat model of ischemic hindlimb resulted in the mononuclear cells attaching to the endothelium and increased collateral vessel development (324). These results were mirrored in a rabbit MI model where UTMBD with BM-MSCs improved cardiac function in the infarcted heart, specifically ejection fraction and increased the collateral circulation (325). In more recent years, a modified version of UTMD has arisen using superparamagnetic nanoparticles (SPIONs) and external magnetic fields. The SPIONs have been added to AD-MSCs and intravenously injected to Wistar rats subjected to stroke (326). A strong magnet was placed next to the ischemic area for 30 min yielding a high initial recruitment of AD-MSCs to the brain, but researchers reported that the engrafted cells showed signs of death after 1 day (326). Contrary to these results, a study in a Wistar rat model of sciatic nerve damage demonstrated that intravenously injected AD-MSCs with SPIONs home to the injured nerve following a 24-h exposure to a magnetic field (327). At 7-days postinjection, 3.3% of administered cells still remained because cell viability and cell differentiation ability were unaffected by the SPIONs (327). The utilization of microbubbles or SPIONs represents a promising advancement in the application of cell therapies to increase homing while circumventing the need for direct injection/delivery. Albeit the intravenous administration of cell therapies has been widely discussed and used, in recent years researchers have been exploring other ways to deliver stem cells to patients.

The newest forms of cell administration, specifically for stromal cells isolated from adipose tissue, have been reported in solid, liquid, and gas phases. The solid phase of administration combines platelet poor plasma with adipose derived stromal cells to form gelatin seeded matrix that could be advantageous in dermal applications (328). As previously mentioned, both adipose-derived stem cells and adipose-derived stromal vascular fraction cells have been combined with collagen matrices and either injected (278) or implanted subdermally (293). In both cases, the utilization of a solid/semi-solid phase in conjunction with a cell therapy greatly increased the angiogenic potential, microvascular density, and cellular engraftment into the host tissue (278, 293). The stromal cells, typically in the liquid phase for most applications discussed and used in the clinic, are often injected into a vein or organ. More recently, stromal cells were used in combination with a micro-dermal roller aimed at increasing treatment penetration for improved restoration through the layers of the skin (328). In an animal model with a distal dermal flap developing necrosing, microneedeling with a dermal roller significantly increased the number of new vessels in the flap, improving viability (329). On human patients, microneedeling was combined with conditioned medium from human ESC cultures and resulted in improved skin pigmentation and decreased the look of wrinkles compared with microneedeling alone; however, the subdermal vascular effects of this therapy regime were not examined (330).

Finally, stromal cells in the liquid form have been combined with a nebulizer to deliver the therapeutic as an aerosolized treatment. The short- and long-term safety/efficacy of intranasal administration of MSCs was examined with human MSCs administered in repeating doses to juvenile immunodeficient mice, with no adverse events nor tumor formation reported. One hour postintranasal administration resulted in a majority of the cells accumulating in the lungs compared with the majority of the cells found in the stomach 1-day postadministration. Following up, cells could not be detected in the brain, lungs, liver, heart, spleen, kidney, or stomach 1-wk postadministration (331). The intranasal route of administration has the added benefit of bypassing the blood-brain barrier noninvasively, and for that reason it has been used to administer neural stem cells to the brain as a treatment for Parkinson’s disease (332, 333) and Alzheimer’s disease (334). The pace of scientific community in exploring the therapeutic mechanism behind cell and cell-based therapies combined with the novel developments in delivery and dosing strategies (Fig. 4) bring the use of stem cells, autologous or allogeneic, ever closer to a therapeutic for a wide range of disease and pathologic states.

FUTURE PERSPECTIVES

In consideration of designing an appropriate approach for the mediation of vascular and microvascular regeneration, this review highlighted critical questions necessary to move the field forward. What is the optimal therapeutic dose? How should cells be administered? What cells should be used? How should cells be conditioned, if at all? Is there an advantage to the use of homogenous or heterogeneous cell populations? Should cellular components (e.g., exosomes or secreted factors) be used in lieu of cells? How do various clinical situations and presentations effect how we answer these questions? These questions, while still completely unanswered, framed our exploration of the field’s current state and identification of future research directions. To provide quality evidence-based solutions to these questions, the field must embrace emerging trends in scientific rigor, reproducibility, and ethics. For instance, quality control standards are only beginning to be suggested in the field, and should be undertaken (335). This is especially true for cell-free studies using exosomes, as standardization is needed in terms of how we define dosage (based off of exosomal count vs. total protein content). Further comparative studies and discussion are needed to determine whether the advantages of exosomal therapy over stem cell therapies warrants such a refined approach (336). What potentially unique positive effects are we missing by removing the cellular component? Indeed, there is emerging discussion on the benefits of heterogeneity of using cell fractions or combined cell therapies versus one therapy with one isolated cell type, perhaps similar discussion is needed in the debate for cellular versus cell-free therapies (337–339). Finally, ensuring studies are representative of all facets of our population, including the inclusion of proper consideration of sex and gender as a biological variable (200) are necessary so that the answers we determine for these questions are translatable and equitable for all.

CONCLUSIONS

Cellular therapies represent a quickly evolving movement with exciting promise to overcome many limitations of current clinical standards. At the same time, the field is still young and concepts of optimal standardization in terms of which cells to use and in what situations, isolation practices, conditions for culture and expansion, ideal donor characteristics, dosing, and administration are not yet established. This review highlights the most recent trends in these areas as it relates to vascular regeneration in an effort to take stock of the current state of the field, in hopes that future approaches become more standardized. Isolation practices, unique characteristics, pros and cons of various cell types are provided in Fig. 1 and Table 1. Although the majority of the field has moved to accept the paracrine hypothesis as the prevailing theory for the majority of stem cell-mediated vascular regeneration, there is still relevance for the direct mechanism, such as differentiation and direct transfer of cellular contents which should not be wholly discounted. These mechanisms can improve vasodilative function, angiogenesis, vascular density, mitochondrial function while reducing calcification and smooth muscle proliferation, reduce lipid and DNA oxidation, ER stress fibrosis, senescence, and apoptosis according to Fig. 2, and microvascular recovery in particular is summarized in Table 2. There has been much debate about the merits of autologous versus allogeneic strategies, with advantages and disadvantages to both as highlighted in Fig. 3. The most feasible answer to the debate currently is that the use of autologous or allogeneic stem cells will likely depend on individual clinical characteristics and needs with each strategy having appropriate clinical situations. These situations should balance ease of using reproducible large-scale manufactured cell therapies versus cell therapies that can be more tailored to individual patient needs, all while keeping patient health status and goals of treatment in mind. The potential disadvantages of autologous therapy may be mitigated by advances in cell priming and preconditioning as well as alternative dosing and administration as highlighted in Figs. 4 and 5. In all, cellular therapeutics rejuvenate vascular dynamics in acute and chronic pathologies in animal and human models with wide opportunities for further optimization for individual pathologic states as well as donor and recipient health status.

GRANTS

This work was funded by Gheen’s Foundation (to A. J. LeBlanc), Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada Grant RGPIN 06621-2017 (to S. S. Nunes Vasconcelos), Canadian Institutes of Health Research Grant PJT153160 (to S. S. Nunes Vasconcelos), National Institute on Aging Grants AG053585 (to A. J. LeBlanc) and R01 AG049821 (to W. L. Murfee), Ministère du Développement économique, de la Création d'emplois et du Commerce, Ontario Ministry of Research and Innovation Grant ER17 13 420 149 (to S. S. Nunes Vasconcelos), and U.S. Department of Defense Grants W81XWH-19-RTRP-IDA and W81XWH-13-2-0057 (to A. J. LeBlanc).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.J.L. conceived and designed research; E.P.T. and G.R. analyzed data; E.P.T., V.S., G.R., D.B., and A.J.L. prepared figures; E.P.T., V.S., G.R., D.B., S.S.N., J.B.H., and W.L.M. drafted manuscript; E.P.T., V.S., G.R., D.B., S.S.N., J.B.H., W.L.M., and A.J.L. edited and revised manuscript; E.P.T., V.S., G.R., D.B., S.S.N., J.B.H., W.L.M., and A.J.L. approved final version of manuscript.