Mechanisms for Progenitor Cell-Mediated Repair for Ischemic Heart Injury

Abstract

Recent studies have shown that treatments involving injection of stem cells into animals with damaged cardiac tissue result in improved cardiac functionality. Clinical trials have reported conflicting results concerning the recellularization of post-infarct collagen scars. No clear mechanism has so far emerged to fully explain how injected stem cells, specifically the commonly used mesenchymal stem cells (MSC) and endothelial precursor cells (EPC), help heal a damaged heart. Clearly, these injected stem cells must survive and thrive in the hypoxic environment that results after injury for any significant repair to occur. Here we discuss how ischemic preconditioning may lead to increased tolerance of stem cells to these harsh conditions and increase their survival and clinical potential after injection. As injected cells must reach the site in numbers large enough for repair to be functionally significant, homing mechanisms involved in stem cell migration are also discussed. We review the mechanisms of action stem cells may employ once they arrive at their target destination. These possible mechanisms include that the injected stem cells (1) secrete growth factors, (2) differentiate into cardiomyocytes to recellularize damaged tissue and strengthen the post-infarct scar, (3) transdifferentiate the host cells into cardiomyocytes, and (4) induce neovascularization. Finally, we discuss that tissue engineering may provide a standardized platform technology to produce clinically applicable stem cell products with these desired mechanistic capacities.

INTRODUCTION

Although pluripotent stem cells have been used in therapies because their intrinsic capacity to differentiate into multiple lineages, there is a risk that these pluripotent stem cells implanted systemically or locally might promote teratoma formation and cause inflammation [1, 2]. Nevertheless, mesenchymal stem cells (MSC) and endothelial precursor cells (EPCs) injected intramyocardially have led to improved right ventricular systolic and diastolic function [3–5]. These stem cell-based cardiac therapies extracted from animal studies [6] (Table 1) offer potential treatments for myocardiac infarction as well as other heart diseases [7]. In practice, however, currently used methods have ignored the fact that MSCs are sensitive to the concurrent serum and O₂ deprivation to which they are exposed when transplanted in vivo in the hypoxic condition of the ischemic heart [8]. As such, majority of these transplanted MSCs are death so the clinical trials conducted showed no consistent results (Table 2) [4, 9, 10]. Limited understanding of the mechanisms involved in repair hinders the development of new methods of stem cell preparation for therapies. Here, we review the available data and discuss the possible mechanisms for beneficial transplantation of stem cells [11]. We have also explored the possible methods to manage the molecular signaling pathways responsible for stem cell migration, engraftment, and neovascularization upon the transplantation into an ischemic heart.

Table 1.

Selected Studies of Transplanted Stem Cells for Cardiac Repair

Cell Source/ Type	Donor/Host Animal	# of Cells Injected	Delivery Method	Cell Survival Time	% Survival	Phenotype	Regenerative Advantage Conferred Over Vehicle	Authors & Year
EPC	Pig	2.5×106	intracoronary infusion	4 weeks	20%	ND	sp-FGF1-modified EPCs induced neovascularization	Chen et al., 2009
MSC	Rat	6 ×106	injected at six injection sites	6 weeks	n=10, n= 9, n= 6	Endothelial,	Bcl-2 gene resulted in high-level VEGF expression in response to the hypoxic condition, improved LV remodeling and function	Li et al., 2007
MSC	Rat	5 × 10 6 in 0.1 ml	6 sites laterally and anterior of ischemic tissue	>6 weeks	ND	ND	Angiogenin enhanced MSCs promoted scar repopulation, angiogenesis, and prevented LV remodeling	Liu et al., 2008
MSC	Rat	3×106	Intramyocardial injection	3 months	-	Cardiomyocytes, smooth muscle	MSC co-overexpression of Ang-1 and Akt incorporate into scar tissue and increase neovascularization	Shujia et al., 2008
MSC	Rat	2.5 × 106	Tail vein injection	3 days	200 cells/ high-power field	Endothelial	CXCR4-MSCs showed increased homing ability into ischemic tissue and increased end diastolic pressure	Cheng et al., 2008
MSC	Rat	2 × 106	Tail Vein injection	4days to 5 weeks	130 cells/ mm2 60 cells/ mm2	None observed	SDF1-MSCs protect existing cardiomyocytes from apoptosis and improve electrical conduction in scar tisssue	Zhang (MING) et al., 2008
MSC	Rat	1×106	Aortic root	28 days	ND	ND	left and right ventricular pressure overload	Molina et al., 2008

Table 2.

Clinical Trials Using Stem Cells for Treatment of Injured Cardiac Tissues to Improve Heart Function in Patients (Refer to http://clinicaltrials.gov/ct2/results?term=msc+heart)

Clinical Trials (gov ID)	Phase	Study Type	Target Heart Disease	Cell Type	Sponsor	Administration Route
NCT00587990	I/ II	Interventional	Left Ventricular Dysfunction	MSC	National Heart, Lung, and Blood Institute	Intramyocardial Injection
NCT00114452	I	Interventional	Myocardial Infarction	MSC	Osiris Therapeutics	Provacel™ (PUMP1)
NCT00768066	I/ II	Interventional	Heart Failure/ Left Ventricular Dysfunction	MSC	University of Miami	Transendocardial Injection
NCT00308633	-	Observational	Coronary Artery Disease	EPC	National Heart, Lung, and Blood Institute	-
NCT00053456	-	Observational	Coronary Arteriosclerosis	EPC	National Heart, Lung, and Blood Institute	-
NCT00644410	I/ II	Interventional	Congestive Heart Failure	MSC	Rigshospitalet, Denmark	Intramyocardial Injection

APOPTOTIC SIGNALS IN HEART INJURY

Understanding of some pathways involved in damaged tissue has begun to reveal how heart tissue is prompted to undergo apoptosis. Apoptosis triggered by hypoxia or injury is regulated either by direct targeting of mitochondria or TNFα signaling. Given the mitochondria’s primary role in cellular respiration for producing ATP, interruption of blood flow into tissues results in low oxygen concentration (i.e., hypoxia) within a cell and leads to stop ATP production. Subsequently, ATP-dependent physiological events are halted [12]. Energy conservation during hypoxia is partially achieved by a decrease in RNA Polymerase III (Pol III) activity, leading to reduce the macromolecule synthesis that is not immediately required for a cell. The Pol III recruitment involved in gene transcription is controlled by Pol III-specific transcription factor IIIB (TFIIIB) association with c-Myc, ERK, and RB. Under the hypoxic condition, TFIIIB dissociates from c-Myc and ERK, preventing Pol III from targeting its gene template while increasing its association with its negative regulator RB [13]. Furthermore, phospholipase C (PLC) and IP3 receptor bind to Na⁺/K⁺ ATPases to form a complex that regulates calcium release from the intra-cellular stores via the activation of IP3 receptors, leading to that only roughly 40% of a cell’s ATP is consumed [14, 15]. The central loop of the Na⁺/K⁺ ATPase-1 subunit is bound by PLC-1 while the N-terminus binds to IP3R2 and IP3R3 along with caveolin-1 and src, both of which respond to ouabain signals [15]. Cardiac glycosides like ouabain are normally used in high concentrations to inhibit the Na⁺/K⁺ pump, causing increased Na⁺ levels and preventing Ca⁺ extrusion through the Na⁺/Ca²⁺ pump as well as inducing hypertrophy. In lower doses, cardiac glycosides like ouabain have been reported to have positive inotropic effects via Erk1/2 activation [16]. Also, mitochondria regulate apoptosis by the activation of caspases. Metallothionein (MT), an anti-apoptotic compound, helps protect cells from oxidative cardiac injury by inhibiting a cell’s mitochondria from releasing cytochrome C, preventing caspase-3 activation and protecting it against ROS damage [17].

In addition, ischemia-associated cell death triggers immune responses that attract circulating cells including stem cells through secretion of various cytokines, such as SDF-1 (Fig. 1), interleukin-6 (IL-6) (Fig. 2), interleukin-1 (IL-1), and tumor necrosis factor-alpha (TNFα) (Fig. 2). TNFα signaling follows a pathway that activates caspase 8, resulting in apoptosis [18].

Fig. (1).

The PI3K/Akt signal pathway is crucial to SDF1/CXCR-mediated homing of endothelial progenitor cells (EPCs) after myocardiac infarction. Ischemia-associated cell death triggered by myocardiac infarction initiates an immune response by producing various cytokines, such as SDF-1. SDF-1 activates its corresponding receptor in circulating EPCs via activation of CXCR4. This binding in turn mediates the PI-3K/AKT signaling pathway that induces the progenitor cell migration toward myocardial infarcts to initiate the tissue repair by increasing CXCR4 and MMP-9 as well as cytoskeleton rearrangement. Either AMD3100, an inhibitor for CXCR4, or LY294002, an inhibitor for PI3K, can downregulate this signaling event.

Fig. (2).

Ischemia-associated cell death triggers an immune response that produces various cytokines, such as interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNFα). Cytokine-mediated signaling pathways modulate precursor cellular activity. IL-6 triggers preconditioning for precursor cell migration and cell protection while TNFα promotes apoptosis and may facilitate the survival and engraftment of precursor cells.

However, all of these are also known to be responsible for reperfusion injury [19, 20]. Reperfusion injury occurs when spillage of cellular components signals immune cells as well as endothelial cells to accumulate at a site where they release reactive oxygen species (ROS) and block small capillaries, resulting in cell death even after blood flow is restored [21]. The release of ROS after oxidative stress activates Ras. In addition, oxygen deprivation or stress has been shown to lead to hypertrophy which results after Erk ½ [22], MAPK [23, 24], and c-Jun-N terminal kinases (JNK) activation [25–29], while activation of p38-MAPKs leads to apoptosis [30, 31]. Signaling for repair, if not properly regulated or controlled, can lead to increased injury. Yet further research in this area is necessary to pinpoint exact mechanisms that can be targeted to prevent or alleviate damage caused by ischemia.

ISCHEMIC PRECONDITIONING FOR INDUCED TOLERANCE OF STEM CELLS

Acute cardiac ischemia results in a hypoxic microenvironment (i.e., low oxygen) which leads to cell death (apoptosis). This low oxygen environment makes it extremely difficult for the injured area to be functionally repaired and usually leads to formation of a collagen scar [32–34]. Consequently, for future stem cell treatments to be effective, the injected cells will need to be tolerant of hypoxic conditions [35–38]. Currently, many studies have focused on protecting injected cells against ischemia by increasing tolerance to hypoxic conditions through ischemic preconditioning of the heart [39, 40]. Ischemic preconditioning is achieved by stopping the heart’s blood supply several times for a few minutes [41]. During hypoxia, lack of ATP prevents ion pumps from working, resulting in a rise of intracellular calcium that accumulates inside mitochondria. When the tissue is reoxygenated, the rise of ATP, damaging to the electron transport chain, increases mitochondrial generation of reactive oxygen species (ROS). Thus, cardiac ischemic preconditioning activates signaling pathways that reduce calcium production [42]. Hypoxic tolerance also occurs naturally following ischemia because of the release of IL-6 and TNFα (Fig. 2). Cardioprotection from the functional and biochemical damage produced by ischemia can be induced by activators of cardiac ATP-sensitive K+ channels, a class of drugs that includes, in particular, aprikalim and nicorandil [43]. It is not, however, known whether these drugs can be used for preconditioning stem cells.

Recent clinical trials have sought to find an association between inflammatory signals, Matrix Metalloproteinases (MMPs), and heart injury. Serum levels of IL-6, MMP9 and high sensitivity C-reactive protein (hs-CRP) in 134 patients were measured [44] and it was found that the levels of these factors were higher in injured patients than controls. IL-6 is responsible for delayed preconditioning via JAK-STAT and conveys cell protection through PI3K/ Akt, nitric oxide synthase (iNOS) and Ca²⁺ handling [45] (Fig. 2).

Additionally, both IL-6 and TNFα activate TNFα receptor 1 [46] and p38MAPK [47]. Signaling via TNFα and its receptor 1 leads to apoptosis while forming a heterodimeric complex with TNFα receptor 2 results in NF-κB activation (Fig. 2). TNFα-receptor 2 is a protein that directly interacts with the TNFα receptor to form a heterodimeric complex with TRAF1 to activate MAPK8/JNK and NF-κB [41, 46]. Hydrogen sulfide preconditioning also protects the heart from ischemia and results in decrease in infarct site and improved contractile function. This is mediated by the activation of K_(ATP)/ PKC/ERK ½ and PI3k/ Akt pathways [48]. Preconditioning of MSCs used in transplantation, rather than the heart, could be performed at 1–3% O₂ concentrations. Maintaining MSCs in this hypoxic environment activates Akt which maintains viability and cell cycle rate, suggesting that preculturing MSC under hypoxic conditions prior to transplantation might improve their tissue regenerative potential by protecting them in the harsh infarct environment [49].

MESENCHYMAL AND ENDOTHELIAL STEM CELLS

Therapeutic injection of stem cells into a host requires accurate cell selection based on differentiation potential, relative ease of isolation, availability in large quantities, in vitro expansion, timing of injection (i.e., a therapeutic window) [50], site of injection, and long-term survival of implanted cells [11]. Based on these criteria, bone marrow-derived mesenchymal stem cells (BM-MSCs) and endothelial progenitor cells (EPCs) are the best candidates [51, 52] [53, 54]. MSCs and EPCs have demonstrated cardiomyogenic differentiation capacity in multiple in vitro and in vivo studies [55, 56]. However, these MSC and EPC therapies may not be compatible with current pharmacological treatments. For examples, treatment with the potassium channel blockers tetraethylammonium and clofilium inhibited MSC proliferation while treatment with 5-azacytidine caused profound changes in current density, which is the electrophysiological property required for functional cardiomyocytes [57–59]. These data suggest that the pharmacological intervention may interfere with stem cell therapies. A two-step treatment paradigm may be used to address the issue: First, the pharmacological treatment to stabilize the condition; second, the stem cell therapy to regenerate the damaged tissue.

BM-MSCs secrete VEGF, IGF-1, EGF, angiopoeitin-1, stromal derived factor 1 (SDF-1), macrophage inflammatory protein-1 a/b, and erythropoietin in wound repair [60, 61]. Besides the commonly used EPCs, other EPC subpopulations have also been tested including early EPCs and late outgrowth endothelial cells (ECs). These cells were cultured with human fibroblasts to test whether their angiogenic properties were equivalent to standard EPCs [62]. Their ability to form tube like structures, integrate into existing networks, and undergoes paracrine-induced angiogenesis were assessed. It was found that only ECs were able to differentiate into tubules but not EPCs. EPCs did induce angiogenesis via paracrine secretion while ECs did not [62]. A recent study sought to optimize MSC expansion by replacing fetus calf serum (FCS) with platelet lysate (PL), since cells cultured with 10% FCS may pose a xenogeneic risk to the patient after transplantation [63]. MSCs were isolated via gravity sedimentation and grown in media supplemented with 5% PL fewer than 5% O₂ conditions. This method increased the yield and proliferation for expansion compared with standard protocols [63]. The functional studies showed that decreased levels of pro-apoptotic signals IL-1, IL-6, TNFα, bax, bak and p38 as well as matrix remodeling molecules such as MMP-3, MMP-6, MMP-9, TIMP-1 and TIMP-3, helped decrease inflammation [64–66].

HOMING MECHANISM BY WHICH INFLAMMATION SIGNALS FOR STEM CELL MIGRATION

As body injury-driven inflammation triggers the recruitment of immune cells and endogenous stem cells, to a site of injury for repair, thus, these cells have the potential benefit for clinical applications to treat heart diseases [67]. Multiple lines of evidence show that both MSCs and EPCs can migrate to damaged tissues [68–73]. It has been showed that the homing mechanism is based on their expression of the CXCR4 receptor that recognizes the chemokine stromal derived factor-1 (SDF-1), which is produced by an injured tissue (Fig. 1). The CXCR4 receptor is present intracellular and integrates into the plasma membrane upon activation of signal transduction in response to hypoxia, cytokines, and sheer forces [74–81]. Expression of CXCR4 in MSCs can be induced within a few hours by employing a cocktail of cytokines. Flt-3 ligand, SCF, IL-6, HGF and IL-3 all can upregulate cell surface and intracellular CXCR4 expression [82]. However, a recent study found that homing via the CXCR4 receptor and SDF-1 ligand decreases with age due to age-related decrease in cytokine/ chemokine expression. One obstacle in the future clinical application of systemically-injected MSCs is that long-term in vitro expansion of these cells decreases CXCR4 expression. Thus, cultured MSCs might not home into injured tissue in sufficient numbers to have a functional impact on damaged myocardium [83]. Tissue homing of MSCs can be increased by forced expression of CXCR4. However, this may also result in a loss of regulation of MSC migration because of a loss of CXCR4-internalization triggered “stop signal” [84]. Our data showed that both CXCR4 and PI-3K are essential for EPCs to home to areas of myocardial infarction because inhibitors of CXCR4 and PI3K reduced homing (Fig. 3).

Fig. (3).

In the rat myocardiac infarction model, neovascularization activity is induced. Endothelial precursor cells (EPCs) were implanted via tail vein injection 24 hours after coronary left anterior coronary artery ligation in rats. Vessel density is determined by staining with vWF 14 days after EPC injection (original magnification, 200×).

A. Immunofluorescence staining of vWF (red) shows the vessel density 14 days after EPC injection.

B. 14 days after EPC injection, EGFP (green) shows that EPCs home to the peri-infarct region with formation of lumen-like structures. vWF staining confirms that the lumen-like structures are new vessels.

C. Quantitative analysis of vessel density. Briefly, EPCs pretreated with AMD3100 (10 µg/ml) or LY294002 (20 µM) for 30 minutes. 5×10⁶ EPCs in 1ml EBM-2 or 1 ml EBM-2 were injected via tail vein 24 hours after left anterior coronary artery ligation in rats. Control vs EPCs vs AMD3100+EPCs vs LY294002+EPCs: 4.93±0.64 vs 14.47±1.37 vs 6.27±0.82 vs 6.10±1.02 n/mm². *p<0.05 vs control, #p<0.05 vs EPCs group. There was no statistical difference between the AMD3100+EPCs and LY294002+EPCs group.

D. Quantitavive analysis of EGFP-positive vessel density. There were no EGFP-positive vessels in the control group. EPCs vs AMD3100+EPCs vs LY294002+EPCs: 10.07±1.03 vs 4.13 ±0.73 vs 3.97±0.78 n/mm². #p<0.05 vs EPCs group. There was no statistical difference between the AMD3100+EPCs and LY294002+EPCs group.

(Reprinted from “Li et al., Stem cell engineering for treatment of heart diseases: Potentials and challenges, Cell Biology International, 33: 255–267, 2009,” with permission from the International Federation for Cell Biology and Elsevier -VAT #GB494627212 and license # 2306641097395)

Additionally, EPCs and MSCs have other migration mechanisms. Akt activation induces intercellular adhesion molecule 1 (ICAM-1) which is important in EPC homing to ischemic tissues [85]. Three stages believed to be involved in stem cell homing to an injury are endothelial cell adhesion, capillary incorporation, and transendothelial migration in which cells reach the extravascular space. Upregulation of the Akt gene results in increased levels of the effecter molecules VEGF and SDF-1 in EPCs [85]. Another receptor responsible for homing in EPCs is the leptin receptor. In a mouse model, EPCs cultured with leptin before injection into a host increased re-endothelialization, showing that leptin increases adhesive properties and homing potential of EPCs in vivo [86]. When the leptin receptor is internalized it increases STAT3 phosphorylation. It was reported that after seven days culture EPCs bound to vitronectin and fibronectin in a receptor-specific manner. These cells were also able to integrate into a monolayer of existing endothelial cells. Granulocyte colony stimulating factor (G-CSF) treatment was reported to increase MSC migration into peripheral blood. Treating both donor and recipient BM stem cells with G-CSF has been shown to improve cardiac function [87]. In this study, immunosuppressed mice pretreated with G-CSF showed increased cell survival in heart along with proliferation and angiogenesis. After 30 days of transplantation, re-endothelialization was observed [87]. Another study found that G-CSF alone failed to induce recruited MSCs to promote differentiation into cardiomyocytes or aid in repair [88]. These contradictory and inconclusive data are an indicator of why it is so difficult to pinpoint the exact mechanisms involved. Standardization of study protocols may lead to a set of consensus criteria for successful comparison of data between laboratories (e.g., EPCs by the European Society of Cardiology) [3].

SIGNALING EVENTS REGULATED BY CYTOKINES AND PARACRINES

Cytokines are signaling molecules involved in cellular communication. They are immunomodulating agents that act via autocrine, paracrine, and endocrine signals to activate the immune system. Chemokines are usually involved in chemotaxis while growth factors stimulate cell growth, proliferation and differentiation. Injured or dying cells release apoptotic signals that promote cell death in a nearby surrounding tissue. Secreted factors released by implanted MSCs, including VEGFα, IGF-1, EGF, keratinocyte growth factor, angiopoietin-1, stromal derived factor-1, macrophage inflammatory protein-1α/β (MIP-1α/β, erythropoietin, and PDGF, recruit macrophages and endothelial cells to a site of injury [60, 89]. Culturing MSCs in conditioned media from normal or hypoxic preconditioned cardiomyocytes revealed soluble paracrine factors involved in differentiation [90]. Further studies revealed that this activation of signaling via the Akt pathway provides paracrine protection for ischemic tissues. Indeed, MSCs overexpressing survival gene Akt release secreted frizzled related protein 2 (Sfrp2) that inhibits apoptotic signals received by Wnt signal ligands during injury due to post-hypoxia reoxygenation,

Sfrp2 exerts anti-apoptotic effect by antagonizing pro-apoptotic properties of specific Wnt ligands Wnt3a. It was shown that Sfrp2 binds directly to Wnt3a, leading to significantly inhibit the Wnt3a-activated caspase and transcriptional activities [91]. It seems contradictory, in the ischemic heart, Wnt signaling induces secreted frizzled related proteins (sFRPs), the endogenous Wnt-antagonists, leading to apoptosis [92]. Nevertheless, MSCs overexpressing both angiopoietin-1 (Ang-1) and Akt also provide long-term therapeutic benefits for preventing apoptosis in an ischemic heart after intramyocardial transplantation up to three months after initial transplantation [93]. In support of this observation, imaging showed myogenic differentiation, increased blood vessel density, and overall improved heart function. Jagged-1 activation of Notch signaling increases Wnt5a expression in EPCs, leading to cardiac differentiation [94].

Angiotensin II is responsible for the secretion of the cardioprotective factors IGF-1and HGF [95]. MSCs overexpressing angiotensin II induce VEGF production and secretion via the AT₁ receptor through Akt/ ERK ½ pathway [96]. Evidence suggests that paracrine signaling is in part responsible for the beneficial effects seen after initial transplantation. Medium from BM-MSCs overexpressing Akt (Akt-MSC) cells cultured under hypoxic conditions showed increased antiapoptotic molecules and triggered increased contraction in cultured rat cardiomyocytes [97]. The Akt-MSC-conditioned medium, containing VEGF, FGF-2, HGF, IGF-1, and TB4, injected into rat infarction models showed improved ventricular function. HGF, upon binding to its receptor c-met, regulates cell matrix interactions. A complex formed by paxillin, Raf, and MEK, in response to HGF, binds and activates ERK leading to epithelial morphogenesis [98]. Erythropoietin (EPO) has been shown to “attenuate” Ang II-caused hypertrophy by activating the PI3K/Akt-eNOS-NO pathway and downregulating TGF-β1 [99]. As downstream effectors, MEK1/2 (mitogen activated protein kinase) activates ERK1/2 (extracellular regulated kinase) on the TEY motif. The Raf-MEK-ERK complex is formed after TEY motif phosphorylation, binding of G-protein subunits, and dimerization of ERK1/2 which leads to Thr188 auto-phosphorylation and results in hyperthropic growth of cardiomyocytes [100].

Angiogenin is responsible for inducing vascularization in tissues. It acts as a tRNA-specific ribonuclease which binds to actin on the EC surface where it is endocytosed and transported into the nucleus for transcriptional regulation to promote invasiveness. Angiogenin-modified MSCs help improve viability in ischemic tissue [101]. The evidence showed that angiogenin overexpressing MSCs, six weeks after injection into infarcted rat heart, resulted in increased tolerance to low oxygen environments. These cells improved the left ventricular systolic and diastolic function through increased vasculogenesis. This is quite significant when compared to previous studies that showed almost complete cell death within four days of transplantation [101].

Many studies suggest that heart injury results in activation of the immune response, leading to the recruitment of immune cells to the injured site. For example, mononuclear cells (MNCs) are essential components of the immune system and can be used as a “Trojan Horse” for gene therapy [102]. MNCs overexpressing VEGF, when transplanted into peripheral circulation, have been implicated in increased angiogenesis and improved heart function in a pig model [103]. In this study, the human VEGF65 was transfected via a plasmid into MNCs isolated from pig peripheral blood, and these VEGF-expressed MNCs were injected into a pig with acute myocardial infarction (AMI) induced by occlusion of the mid portion of the left anterior descending coronary artery (LAD). They observed that the transplantation of these modified MNCs improved left ventricular (LV) function, collateral vessels, capillary density and increased wall thickness of the scar in the heart after AMI. These findings indicate that MNCs can be used as a gene therapy vehicle for the targeted therapy of ischemic heart disease.

Other important effectors including IL-6 and TNFα promote circulating stem cells to be recruited to a site of injury [104]. These factors may act on the Akt signaling that is important for cell growth and survival [105–108]. Current studies showed that stem cell survival rate after injection into ischemic cardiac tissue is critical to have directed clinical effects [109–114]. Since MAPKs are involved in both cell survival and apoptotic signaling, pretreatment for inhibition of MAPKs phosphorylation by specific inhibitors revealed the antiapoptotic role of MAPKs in stem cells and increased their survival in the injured tissue environment as nitric oxide (NO) induced apoptosis [115].

MSC AND EPC DIFFERENTIATION INTO CARDIAC TISSUE

Other possible mechanisms for the beneficial effects seen after stem cell injection in cardiac injury include the potential for MSCs and EPCs to differentiate into cardiomyocytes or transdifferentiate into cardiomyocytes [116–122]. Unfractioned bone marrow cells (BMC) mixed with either MSCs or EPCs can be injected to improve recruitment of progenitor cells into the damaged heart area [123]. Angiogenesis and survival of the transplanted cells near the scar are promoted by both EPCs and MSCs, while only MSCs invade the scar. Long-term co-culture of human MSCs with neonatal rat cardiomyocytes resulted in differentiated cells exhibiting a cardiomyogenic phenotype. It was observed that NKX2.5 and GATA4 transcription factors were translocated into the nucleus which coincided with sarcomeric actinin expression [124]. GLUT4 translocation to sarcolemma by activation of PI3K (stimulated by insulin) depended on activation of the PKC-zeta isoform but these levels were not above basal (only permissive, no strong correlation) [125]. Stable improvement in heart tissue for a period of eight months after implantation along with EPCs was observed with increased levels of VEGF-and bFGF-mediated angiogenesis [126].

Rat MSCs co-cultured with mature cardiomyocytes showed that some MSCs express cardiogenic markers under standard culturing methods. Under hypoxic conditions (less than 1% oxygen concentration), deprivation of glucose and serum further increased to 7% cardiac committed MSCs in the presence of difluoromethylornithine (DFMO), an inhibitor of polyamine synthesis and cell proliferation, while only 5% MSCs showed cardiac markers positive without DFMO [127]. Although pluripotent stem cells have been proposed for cardiac therapy, the propensity of these cells to differentiate into multiple lineages may lead to heterogeneous tumor formation, cardiogenic specification using TNFα, however, has been shown to lead to cardiac differentiation in vitro and in vivo without neoplastic outgrowth, suggesting that injecting cardiac progenitors instead of pluripotent stem cells are safest [128].

More importantly, co-cultures of EPCs and cardiomyocytes promote forming intercellular connections commonly referred to as nanotubes, sized from 50–800 nm with a length of 5–120 nm. These nanotubes are responsible for organelle and cytoplasmic molecule exchange and increased after longer culture times [129]. Different culturing protocols showed mixed results for transdifferentiation, where some studies showed evidence while others did not [130]. Transplanted MSCs and host myocardium are able to communicate via nanotubes in vitro and in vivo. These cells communicate by cytosolic molecule exchange as well as transfer of mitochondria resulting from fusion [122, 131–134]. Transfer of mitochondria from MSCs to myocardium, although not observed within 24 hours after transplantation in this study, might in part explain the beneficial effects seen after MSC transplantation to help repair damaged mitochondria [122]. TGFβ1 has been shown to promote cardiomyogenic differentiation in rat adipose-derived stromal cells when cultured for 2 weeks in vitro [135].

MOBILIZATION OF CARDIAC PROGENITORS

It had been widely accepted that the heart loses its ability to regenerate after birth, however, recent studies have shown the post-natal heart is, in fact, able to regenerate and heal itself [136]. Resident cardiac progenitor cells are capable of minor repair of the heart yet these are not available in sufficient quantities to repair large damaged areas [137–139]. Radioactive labeling (pulse-labeling) to measure rates of cell growth has been used in animals but not in humans, because of ethical concerns. However, researchers have found that radioactive material from nuclear weapon testifying from 1963, have essentially labeled the entire world population indirectly via atmospheric contamination. Thus, they measured carbon-14 levels in human heart cells and found that at age 25 about 1% of cells are replaced each year, decreasing to around half of one percent by age 75 [140]. It has been found in chick embryos that cardiomyocytes are able to re-enter the cell cycle and proliferate after differentiation. LIM homeodomain transcription factor Islet1 (Isl1) is present in cardiac progenitor cells. It was found that Isl1 is expressed in response to β-catenin regulation, leading to proliferation and survival [16]. Isl1-expressed precursors did not proliferate before becoming cardiomyocytes but did do so after, even when contractility was observed. It was observed that almost 83% of differentiated cardiomyocytes proliferated [141]. The existence of resident heart cells (cardiac stem cells), as opposed to cardiac progenitors, was assessed using cells obtained by digesting adult mouse heart. Rounded adherent cells were expanded and cultured 2–5 days after plating. Further culturing leads to self-beating patches that displayed many cardiomyocyte markers. Patch-clamp measurements showed that they were able to create spontaneous action potentials, a self-beating property due to HCN4 channels which lead to depolarization [142].

Controlled regulation of ion channels is essential for regeneration of cardiac pacemakers. Hyperpolarization-activated cyclic nucleotide-gated channels (HCN 1–4) and T-type calcium channels are responsible for the pacemaker property of human embryonic stem cell (hESC)-derived cardiomyocytes [143]. Data from mouse embryos suggest that all three layers present in the heart, cardiomyocytes, endothelial, and smooth muscle, originate from a Flk1 (a.k.a., KDR, VEGF-R2) positive progenitor. To test whether this is also the case in humans, they studied hESC differentiation. After hESCs were cultured with a cocktail of many growth factors, embryoid bodies were made which resulted in KDR low/c-kit-(CD117) negative population that displayed cardiac, endothelial and vascular smooth muscle properties in vitro and in vivo [144]. Over 50% of these progenitors differentiated into beating cardiomyocytes in monolayer culture. Evidence supports a common precursor responsible for all cardiac lineages [145]. A cardiac-specific Nkx2.5 subpopulation isolated from mouse embryos showed that about 28% of these cells express c-kit, which in turn give rise to both cardiomyocytes and smooth muscle cells [146]. The second heart field formed during cardiac development is responsible for the supply of most cardiac progenitors that express Nkx2–5, which is a homeodomain factor that is a repressor of Bmp2/ Smad1 signaling [147]. Smad1-dependent negative feedback loop is responsible for cardiac differentiation [148]. The cardiomyogenic potential of cardiac stem cells and bone marrow-derived stem cells is associated with c-kit expression, leading to the possibility that these cells may be part of the same reservoir, connected by the systemic circulation. Cardiomyocytes differentiated from bone marrow stem cells through the use of VEGF and FGFβ, express Flk1 and FGFR1, which are observed as parallel differentiation of ES cells into cardiomyocytes [89]. Resident cardiac stem cells are believed to have the ability to migrate to a site of injury via the SCF/ c-kit signaling [149–155]. This is specifically via phosphorylated p38 MAPK to induce SCF-mediated migration of cardiac stem cells.

TISSUE ENGINEERING

Research has rapidly moved towards three-dimensional (3-D) tissue engineering of left ventricular allograft, aortic valve allografts, and whole hearts, in mouse, rat, pig, and bovine models [156–167]. A cardiac patch made from MSCs shows the formation of neomuscle and neomicrovessels. Microvessels are essential, since they allow MSCs to undergo cell division or recruit other cells to make the patch thicker than it was originally [95]. Cell sheets are commonly made using temperature-responsive plates which allows for the intact removal from culture plates that contain cardiomyocyte cell sheets vascularized by microvesslesmicrovessels. These have been shown to have an improvement on the heart via secretion of paracrine factors and by providing mechanical support [168]. Decellularized heart scaffolds offer an exciting alternative for those whose heart injury is too severe for stem cell injection [157]. A bioartificial heart can be made using a donor heart since ECM without cells does not appear to elicit an immune response. These donor hearts would be treated with detergents that remove all cells but leave the ECM structure intact [169–172]. MSCs and EPCs seeded into the scaffold are able to recellularize it within 4 days, generate pump function at day 8, and undergo coronary perfusion by day 28 [173, 174]. A scaffold, serving to promote stem cell differentiation, is dependent, in part, on physical stimuli as well as soluble growth factors. Many studies have shown that the microenvironment significantly shapes cell differentiation [175]. In vivo, poor viability at the transplant site often hinders the therapeutic potential of MSCs. To overcome this barrier, cationized polysaccharide spermine-dextran is used as a carrier for protection of these cells from damage [168].

Tissue engineering may play a major role for future clinical applications because the heterogeneity of stem cells remains a main issue, especially for pluripotent stem cells. A clinically-applicable population of stem cells should be capable of being expanded and differentiated to result in a sufficient number of cells with consistent quality (e.g., a “pure” population of precursors). Cardiac precursors be derived from the embryoid bodies which are heterogeneous aggregates of cells, but these contain limited numbers of cells, making large-scale production difficult. It was reported that a degree of homogeneity of a stem cell product was obtained by using a microprinting strategy and a bioreactor methodology, with a large yield of cardiac precursors being obtained when applying hypoxic conditions [176]. All of these experiments suggest that an engineer design method can be used to develop a specific stem cell product with a desired specification for personalized treatment of ischemic heart injury.

CONCLUSION

MSCs and EPCs after systemic injection or local injection to a site of injury may be involved in direct tissue replacement or tissue protection by release of cytokines and chemokines, or both. This may strengthen the post-infarct scar or mobilize resident cardiac stem cells to recellularize the damaged tissue. Studies show that key effectors and modulators are responsible for some of the beneficial effects, including IL-6 and TNFα, which direct circulating stem cells to go to the site of injury. Akt signaling, which is important for early heart development and growth, is activated after transplantation. Currently, stem cell survival rates after injection into ischemic cardiac tissue are not high enough to give rise to functional improvement. Engineered tissue could serve as a platform to test the effectiveness of stem cell treatment by mimicking the injured heart microenvironment. It may be advantageous to engineer these cells to have stronger homing mechanisms by upregulating their responsiveness to cytokines and chemokines, allowing them to reach injured tissue more effectively and in higher numbers. Because the exact mechanisms by which injected stem cells repair an injury site are still unclear, it is essential to employ a standardized platform technology like tissue engineering to understand the mechanisms of stem cell repair in different preclinical and clinical models [177].