Abstract
Transplantation of stem cells, including mesenchymal stromal cells (MSCs), improves the recovery of cardiac function after myocardial infarction (MI) in experimental studies using animal models and in patients. However, the improvement of cardiac function following MSC transplantation remains suboptimal in both preclinical and clinical studies. Understanding the mechanism of cell therapy may improve its therapeutic outcomes, but the mode of action mediating stem cell promotion of cardiac repair is complex and not fully understood. Recent studies suggest that the immunomodulatory effects of MSCs on the macrophage M1/M2 subtype transition allow the transplanted stem cells to inhibit inflammation-induced injury and promote cardiac repair in acute MI. However, equally compelling evidence shows that there is poor survival and minimal graft persistence of transplanted MSCs within the infarcted heart tissues, negating the view that graft survival per se is required for the observed high rate and long duration of the transition from proinflammatory M1 to reparative M2 macrophages in the infarcted myocardium. Therefore, we raised a novel hypothesis that the therapeutic effects of MSC transplantation for acute MI depends not primarily on the grafted cells in infarct myocardium, but that MSCs migrating to and being lodged in the extracardiac organs, demonstrating good graft survival and persistence, may render the therapeutic effects in MI. More specifically, MSC transplantation promotes the transition from M1 to M2 in extracardiac organs, such as spleen and bone marrow, and therapeutic effects are conferred to the infarcted myocardium via paracrine effects. In MSC transplantation, the conversion from proinflammatory M1 to anti-inflammatory M2 monocytes may occur remotely from the heart and may serve as one of the major pathways in regulating the dual effects of inflammation. This hypothesis, if proven valid, may represent an important new mechanism of action to be considered for the future of MSC transplantation in the treatment of MI.
MSC Transplantation Balances PRO- and Anti-Inflammatory Responses in Myocardial Infarction
Acute myocardial infarction (MI) is a leading cause of morbidity and mortality worldwide. Advances in therapy have led to a decline in mortality during the acute phase, but this decrease in mortality is paralleled by an increase in the incidence of chronic heart failure in patients surviving with significant residual myocardial damage (20). Transplantation of stem cells, including mesenchymal stromal cells (MSCs), improves the recovery of heart function after MI in experimental studies and in patients (7). MSCs have the ability to differentiate into various lineages including chondrocytes, adipocytes, and osteocytes. MSCs prevent deleterious remodeling and promote recovery when introduced into the infarcted heart (23). When implanted in a healthy myocardium, MSCs can be allogeneically derived, delivered systematically, and differentiate into a cardiomyocyte-like phenotype (23). Benefits of the transplantation of MSCs include the potential to supply growth factors and cytokines to repair tissue (23), multipotent differentiation capability (35), and a tendency to avoid immune rejection. All of these factors and the potential of MSCs to repair damaged cardiovascular tissue have made these cells a strong cell-based therapeutic agent.
Stem cell therapy has been proven to provide therapeutic effects via paracrine mechanisms rather than cell replacement (7,37). Despite robust functional recovery observed after cell transplantation, the frequency of engraftment and differentiation is poor (39), suggesting an alternative mechanism of action that does not involve graft survival per se. MSCs secrete paracrine factors, including cytokines and growth factors, which can suppress the immune system, enhance angiogenesis, and stimulate differentiation of tissue-specific stem cells (35). Published data have demonstrated that MSCs can exert paracrine cardioprotection in the surviving myocardium via regulating the double-edged sword effects of inflammation secondary to MI (12,13,31). Following MSC transplantation, the transition from proinflammatory M1 to anti-inflammatory M2 monocytes has the potential to occur at sites distal from the heart, regulating the dual effects of inflammation, which would suggest that the migration of MSCs to extracardiac organs may render therapeutic effects in MI.
During MI, both ischemia-induced and ischemia/reperfusion (I/R) injury-induced cell death trigger a strong inflammatory reaction (30,32), which results in an influx of neutrophils, subsequently followed by infiltration of monocytes and lymphocytes. The immune response further damages the ischemic myocardium, but also initiates the wound-healing process and the formation of proper scar tissue (32). The balance between inflammatory injury and repair needs proper regulation in order to prevent excessive inflammation or inadequate stimulation of repair. To this end, maintaining the balance between pro- and anti-inflammatory responses to MI may exert therapeutic benefits. Merely targeting the proinflammatory end of the spectrum and neglecting the anti-inflammatory signaling pathway may not be optimal. Indeed, general suppression of the immune system using cortisone or nonsteroidal anti-inflammatory drugs has been shown to produce detrimental effects on overall survival in both animal models and clinical studies (11,16).
Accumulating evidence from laboratory studies has shown that monocytes and macrophages are central mediators of the inflammatory response, contributing to both the initiation and the inhibition of inflammation (12,13). Early postinfarction remodeling is characterized by a biphasic cellular response [reviewed in Frangogiannis (11)]. In the post-MI heart, initially an infiltration of classically activated (proinflammatory M1) macrophages triggers local inflammation by secretion of proinflammatory cytokines, such as interleukin-1β (IL-1β), IL-6, tumor necrosis factor-α (TNF-α), and interferon-γ (IFN-γ) (11), subsequently producing cytotoxic injury to the surviving cardiomyocytes. Interestingly during this degenerative process, these same macrophages upregulate matrix metalloproteinases (MMPs) that degrade the extracellular matrix (ECM) of the infarcted heart, making it easier to clear dead tissue from the infarcted area. By 5 days after MI, the macrophage cell phenotype has switched to the alternatively activated (anti-inflammatory M2) macrophage (11), now exerting anti-inflammatory actions, thereby mitigating the inflammatory response and initiating myocardial repair.
Recently, MSCs have emerged as potent modulators of the immune system. Treatment of post-MI inflammation using MSCs could provide a new approach of modulating the immune response, shifting the balance toward the reparative phase and reducing inflammation. The interaction of MSCs with several cells of the immune system has been widely documented, implicating a key role for MSCs in suppressing white blood cells or inducing them toward specific anti-inflammatory phenotypes (31). Of note, MSC transplantation significantly increased the percentage of reparative M2 macrophages (F4/80+CD206+) in the infarcted myocardium compared with saline-treated hearts at 3 and 4 days after MI (4). Moreover, the MSC-induced macrophage cytokine secretion displayed a profile relevant to healing and repair of the infarcted heart (4). Interestingly, early macrophage depletion partially reduced the favorable effects of MSC transplantation (4). In parallel, some of the favorable effects of MSCs are mediated by modulation of macrophage phenotype and function toward an anti-inflammatory response in the infarcted myocardium (6,17,19). Altogether, these studies support the notion that MSCs could affect the kinetics of macrophage subsets and exert paracrine cardioprotective effects in the surviving myocardium via increasing the percentage of reparative (M2) macrophages and regulating large amounts of anti-inflammatory and proinflammatory factors in the infarcted heart. These findings indicate that targeting macrophages with MSCs represents a novel approach to ameliorating tissue injury in the setting of MI.
Despite extensive evidence demonstrating the effects of MSCs on macrophages in the infarcted heart and their potential to regulate the inflammatory response, our knowledge on the role of MSC–macrophage interactions in the modulation of inflammation and promotion of cardiac repair remains limited. A number of experimental studies have documented that there is a poor survival of stem cells in the infarcted myocardium after transplantation (28,32). In a rat model of MI, the majority of the stem cells were trapped in the lung following systemic intravenous infusion (3). Less than 1% of the infused stem cells were detected in the injured heart. Even following direct intracardiac injection, stem cell retention in the heart declines rapidly, with only 10% present after 4 h and approximately 1% 24 h after injection (14,21). No long-term engraftment and subsequent vascular differentiation have been reported. It is noteworthy that a study using quantitative assays for human DNA and mRNA found that intravenously infused human multipotent stromal cells (hMSCs) after MI can enhance cardiac repair without significant engraftment (19). In our previous study, we systemically delivered human umbilical cord blood cells (HUCBCs) coinfused with a BBB permeabilizer (mannitol) into a stroke rat model and found that no HUCBCs were detected in the brains of animals (5). However, HUCBC–mannitol treatment significantly increased brain levels of neurotrophic factors, which correlated positively with reduced cerebral infarcts and improved behavioral outcomes. These findings demonstrated that infarct repair and functional improvements were more frequently observed with the poor survival of stem cells. The limited recruitment and retention of stem cells at the site of damage is generally insufficient for effective restoration of the tissue. Most of the beneficial effects may be explained by paracrine secretions arising from “acute” cell-to-cell contacts that have multiple effects including modulation of inflammatory or immune reactions. Due to the fact that therapeutic effects were observed despite the limited number of stem cells retained in injured hearts, we speculated that the high rate and duration (for 10–14 days after MI) (12) of the transition from proinflammatory M1 to reparative M2 macrophages in the infarcted myocardium was not primarily due to cell engraftment.
Splenic Migration of MSCs Contributes to Heart Repair
MSCs have been reported to migrate to bone marrow and the spleen after systemic or intracardiac transplantation in MI animal models (2,19). Our previous study demonstrated that multiple doses of human cord blood mononuclear progenitor cells (HCBMCs) increased the amount of transplanted cells that homed to and were retained in the infarct area and substantially improved or preserved cardiac functions after MI (38). We also found that the grafted cells migrated to the spleen, liver, and bone marrow and survived up to 8 weeks after MI (unpublished data). Tracking studies have shown that the majority of MSCs immediately localize to the lungs after intravenous infusion, and they tend to disappear from the lungs within hours, migrate preferentially to sites of injury and also to other tissues such as the liver, spleen, and bone marrow (2,19). Of interest, even after direct intracardiac injection of MSCs, some transplanted cells migrated to the spleen and liver and survived 4 weeks after transplantation (25). These studies clearly show that MSCs home to extracardiac organs, and these organs may contribute to the therapeutic outcomes of cell therapy.
At the early phase of inflammation after MI, infiltrating monocytes in infarcted myocardium are derived mainly from peripheral blood (25,36). Within days after injury, monocytes were recruited at a high rate but resided only for an average of 20 h in the infarcted heart (34). This monocyte recruitment was fostered by extramedullary monocytopoiesis in the spleen and bone marrow (34). Surgical removal of the spleen several days after coronary ligation substantially reduced the number of monocytes recruited to the infarct and led to impaired wound healing and heart failure (26). The biphasic cellular response of inflammation in response to injury, similar to that observed in myocardial infarction, was also found in peripheral blood and the spleen (25,26). A clinical study described a biphasic monocyte pattern in the blood of patients after MI, with inflammatory CD14+CD163 monocytes dominating first, then with anti-inflammatory CD14–CD163 monocytes ensuing at 3 days post-MI (34). In addition, monocytes mainly express M1 subtype (Ly-6C high) at 1–3 days, but display M2 subsets (Ly-6C low) at 3–5 days in red medulla of the spleen after MI in a mouse model (26,36). Another study investigating the effect of two types of MSCs, human BM-derived MSCs and human umbilical cord perivascular cells, in an experimental MI model of the immune-deficient NOD/SCID mouse revealed that MSCs increased both the proportion of M2 macrophages in infarcted myocardium and M2 monocytes in the peripheral blood (8).
Taken together, we propose the hypothesis that the therapeutic effects of MSC transplantation for MI may not solely depend on the grafted cells surviving in the infarcted myocardium, but that a complementary or even an alternative mechanism of action may entail stem cells migrating to the extracardiac organs and from there exerting their therapeutic effects (i.e., modulation of the inflammatory response) (Fig. 1). MSC transplantation promotes not only the conversion of macrophages of the M1 subtype to M2 in infarcted myocardium but also the transition of monocyte subtype M1 to M2 in the extracardiac organs, such as the spleen and bone marrow, altogether achieving a “regenerative” inflammatory response via paracrine effects. MSC transplantation may thus accomplish the transition of proinflammatory M1 to anti-inflammatory M2 monocytes remotely from the infarcted heart, posing as a novel regenerative pathway in regulating the dual effects of inflammation caused by MI. To our knowledge, no current studies have explored this innovative approach of cell therapy for MI. This hypothesis, if proven to be valid, may represent an important breakthrough in understanding MSC-mediated therapeutic actions in cardiac protection and repair for MI. This present hypothesis warrants in-depth investigations into MSC transplantation for both preclinical models of MI and patients with MI.

Transplantation of stem cells to repair the infarcted myocardium. Cell replacement of dead or dying cells in the infarcted heart represents a direct pathway of affording a therapeutic outcome in MI. Alternatively, migration of stem cells to the spleen, whereupon the transplanted cells sequester the splenic inflammatory response, may similarly exert therapeutic effects on the inflamed heart through an “extracardiac” indirect pathway.
Therapeutic Effects of Stem Cell Therapy Mediated by MSC Splenic Migration
The hypothesis can be validated by following key steps. First, that the grafted cells migrate to and survive in the extracardiac organs, such as the spleen, liver, and bone marrow, can be tested using an animal model of MI, which is created by ligation of the left anterior descending artery (LAD). At acute and subacute periods after MI, the animals are intravenously injected with MSCs, which can be obtained from HUCBCs, labeled with green fluorescent protein (GFP). At different maturation periods after cell transplantation, human DNA and mRNA levels, as well as macrophages in infarcted myocardium, and monocytes of the M1/M2 subtypes in peripheral blood, spleen, and bone marrow at different time points after cell transplantation, can be measured by flow cytometry and assessed by immunohistochemistry. We envision that stem cells can migrate to the extracardiac organs, coupled with their ability to transition monocytes from subtype M1 to M2. It is critical for this hypothesis to be tested, as studies may reveal whether the grafted cells and the transition of the monocyte subtype from M1 to M2 in extracardiac organs and anti-inflammatory cytokines in circulation is closely related to inhibition of inflammatory injury and improvement of cardiac repair after MSC transplantation.
Finally, the mechanisms mediating the promotion of the monocyte subtype transition from M1 to M2 may reveal how MSCs modulate macrophage and monocyte polarization. The ability of MSCs to modulate macrophage function relies on both cell contact-dependent mechanisms and paracrine effects through the release of soluble factors (9,10,31). MSCs may also modulate macrophage function through the generation of regulatory T cells (15). In the setting of inflammation after MI, the inflammatory signals recognized as damage-associated molecular patterns (DAMPs) are important for attraction and activation of various components of the immune system (22). We speculate that soluble DAMPs, such as heat shock proteins, from infarcted myocardium enter into circulation, possibly bind with toll-like receptors (TLRs) including TLR3, and other receptors on MSC membranes, and trigger the immunomodulatory properties of MSCs that have migrated to extracardiac organs. This hypothesis can be tested in an in vitro study, whereby MSCs and peripheral blood monocytes are cocultured and treated with plasma from experimental MI animals, then exposed to a TLR3- or heat shock protein-neutralizing antibody. This study should reveal whether MSCs trigger the conversion of monocytes from subtype M1 to M2 possibly through TLR3–ligand interaction with soluble DAMPs in the plasma from the experimental MI animal. In the end, these proposed studies will test the hypothesis that the therapeutic effects of stem cell therapy are mediated by MSCs migrating to the extracardiac organs and from these remote areas regulate inflammation of the infarcted heart via paracrine effects.
Splenic Function in MI Mimics the Inflammatory Response and its Role in Cell Therapy for Stroke and TBI
Several lines of investigation from noncardiac fields, especially those focused on brain disorders, have documented the migration of stem cells to the spleen. Following injury, splenic pathology resembling that seen after MI accompanies brain alterations (1). By the same token, when cell therapy is employed to treat the injured brain, the key role of the spleen in modulating myocardial infarction, especially the inflammatory response, similarly persists in stroke and traumatic brain injury (TBI). We speculate that therapeutic effects do not primarily depend on engraftment of cells in the infarct myocardium, but instead the cells migrate to the spleen where the transplanted cells exert their anti-inflammatory effects in promoting therapeutic effects for MI. In parallel to MI observations, anti-inflammatory effects of grafted stem cells were found in intravenously transplanted bone marrow stromal cells (hBMSCs) in rats exposed to experimental stroke. Within the first hour of transplantation, hBMSCs migrated preferentially to the spleen over the brain (1). This splenic migration potential of the cells likely induced a splenic anti-inflammatory response and possibly triggered therapeutic benefits against stroke, reminiscent of the cell therapy effects in MI.
Preferential migration to the spleen has also been detected when evaluating the effects of cell therapy in TBI (27). Intravenously transplanted human adipose-derived stem cells (hADSCs) preferentially migrated to the spleen rather than the injured brain within 1 h after transplantation, which again lends support to the notion of an indirect pathway of cell therapy for repair and improving motor and cognitive behaviors in TBI (27). Of note, one recent observation demonstrated that MSC treatment in TBI acted as remote “bioreactors” via stimulation of lung macrophages and augmentation of regulatory T-cell production in the spleen, leading to systemic increases of circulating anti-inflammatory cytokines and alteration of the local/regional milieu of the central nervous system (33). The altered intracerebral microenvironment may prompt the modulation of the resident microglia population to acquire a reparative phenotype, characterized by an increase in the ratio of M2 (anti-inflammatory) macrophages to M1 (proinflammatory) macrophages, and it is this effect that may primarily accounts for the observed neuroprotection in the brain, a regenerative medicine process that may similarly apply to MI.
Stem cell migration to the spleen after MI, stroke, and TBI may act as an anti-inflammatory agent, dampening the overall systemic inflammation that may exacerbate the primary injury. Cell therapy is able to induce beneficial effects from the periphery, remote from the site of injury, which offers an indirect pathway for regenerative medicine. Moreover, the observed splenic response seen in the heart and the brain suggest overlapping etiologies between diseases associated with these two organs. Indeed, we recently demonstrated a cross talk of cell death pathways between the brain and the heart. When the brain becomes injured (such as in stroke or TBI), it can no longer control the rhythm and rate of the heart appropriately, causing the patient to present cardiac symptoms in addition to neurological dysfunction. New findings suggest that the indirect pathway of ischemic stroke regulating cardiac cell death plays an important role in cardiac failure (18). Soluble toxic factors are likely secreted from damaged or dead neuronal cells during the early stages of brain injury and eventually reach the heart via these cell death signals, indicating a strong correlation between brain insult and cardiac arrest. Altogether these studies suggest that transplanted cells serve as anti-inflammatory mediators through either cell-to-cell interaction or bystander effects (i.e., stem cell secretion of therapeutic molecules). The latter mechanism of anti-inflammatory factor secretion by stem cells and the subsequent abrogation of the splenic inflammatory response to injury are shown in our series of studies to mediate MI, stroke, and TBI.
Similarities between Effect of MSC Therapy in the Injured Brain and Heart Ischemia
MSCs have a great potential for therapeutic effects in myocardial infarction as well as injuries to the brain, such as stroke and TBI. Interestingly, MSC therapy may exert common therapeutic actions, including increased plasticity, immunomodulation, and anti-inflammatory effects in both brain and heart ischemia (24). When an injury occurs in the brain or the heart, the immune system becomes modified, causing an inflammatory response to the site of injury. In stroke, TBI, and MI, transplanted MSCs have the ability to reduce levels of proinflammatory cytokines and increase anti-inflammatory cytokines, which can promote tissue repair (40). Transplantation of MSCs (although the grafted cells subsequently migrate to the spleen) can attenuate inflammation and accelerate vascularization in MI (24) and create a favorable environment for regeneration in stroke (29). In TBI specifically, the insult-induced inflammation appears to be a key factor in secondary brain damage, suggesting that anti-inflammatory or immunoregulatory responses from MSCs could provide effective treatments in sequestering this progressive disease pathology (40). These overlapping therapeutic effects of MSCs between the heart and brain suggest that by providing an anti-inflammatory response for the injured organ, MSCs may promote tissue repair via an immunomodulation- and anti-inflammatory-based mechanism, thereby supporting a novel therapeutic modality for the damaged heart and brain.
Conclusion
A novel cell therapy mechanism is described here, highlighting the concept that stem cells migrate not primarily to the injured organs, but specifically to the spleen, following transplantation into the injured heart or the brain. This indirect pathway of inducing therapeutic outcomes by cell transplantation likely triggers a myriad of regenerative factors that initiate in the spleen and culminate in the repair of both the heart and the brain. Such a critical role of the spleen as the primary area targeted by transplanted stem cells implicates a robust function of the stem cells to serve as anti-inflammatory agents and sequester the inflammation-induced secondary cell death inherent in MI, stroke, and TBI. This novel cell therapy property directed at limiting the splenic inflammatory response stands as a new approach in regenerative medicine.
Footnotes
Acknowledgment
The authors declare no conflicts of interest.
