Progenitor Cell Mobilization and Recruitment: SDF-1, CXCR4, α4-integrin, and c-kit

Abstract

Progenitor cell retention and release are largely governed by the binding of stromal-cell-derived factor 1 (SDF-1) to CXC chemokine receptor 4 (CXCR4) and by α4-integrin signaling. Both of these pathways are dependent on c-kit activity: the mobilization of progenitor cells in response to either CXCR4 antagonism or α4-integrin blockade is impaired by the loss of c-kit kinase activity; and c-kit–kinase inactivation blocks the retention of CXCR4-positive progenitor cells in the bone marrow. SDF-1/CXCR4 and α4-integrin signaling are also crucial for the retention of progenitor cells in the ischemic region, which may explain, at least in part, why clinical trials of progenitor cell therapy have failed to display the efficacy observed in preclinical investigations. The lack of effectiveness is often attributed to poor retention of the transplanted cells and, to date, most of the trial protocols have mobilized cells with injections of granulocyte colony-stimulating factor (G-CSF), which activates extracellular proteases that irreversibly cleave cell-surface adhesion molecules, including α4-integrin and CXCR4. Thus, the retention of G-CSF-mobilized cells in the ischemic region may be impaired, and the mobilization of agents that reversibly disrupt SDF-1/CXCR4 binding, such as AMD3100, may improve patient response. Efforts to supplement SDF-1 levels in the ischemic region may also improve progenitor cell recruitment and the effectiveness of stem cell therapy.

I. Introduction

Over the last decade, a compelling body of evidence has accumulated to suggest that progenitor cells of bone marrow origin, such as endothelial pro-genitor cells (EPCs) and mesenchymal stem cells (MSCs), play a significant role in postnatal physiological and pathophysiological vasculogenesis^1–7 and could provide a promising new therapeutic approach for the treatment of ischemic disease.^8–15 These cells form the structural components of the new vasculature, mediate favorable cell–cell contacts, and release growth factors that contribute to vessel growth and protect against cell death in the ischemic tissue.^14,16,17 Furthermore, abnormally low levels of peripheral blood EPCs are closely associated with risk factors for cardiovascular disease, cardiovascular events, and mortality.^18,19

Currently, most clinical trials of cell therapy for the treatment of ischemic heart disease have used progenitor cells of bone marrow origin,^20–22 which are usually administered via intracoronary infusion or transplanted directly into the ischemic region. In general, the trials have found evidence of therapeutic benefit, but with only modest efficacy,^21–26 and the absence of more definitive results is often attributed to poor retention and survival of the transplanted cells.^21,22,27 Because increases in circulating progenitor cell levels are expected to enhance the number of cells recruited to the ischemic tissue,^28–31 techniques that promote progenitor cell mobilization are being rigorously investigated.^32–36 The effectiveness of this strategy has been demonstrated in numerous preclinical studies^{30,31,35–38} and has led to frequent investigations of progenitor-cell-mobilizing agents in early clinical trials.^{28,29,39–50} Granulocyte colony-stimulating factor (G-CSF) has been the most commonly used mobilizing agent, but the results from these trials have not met the expectations, despite substantial increases in peripheral blood progenitor cell counts.^{28,29,44,46,48,51,52} Thus, a better understanding of how progenitor cells interact with the microenvironment in the bone marrow and in the ischemic region could lead to the development of more effective cell-based therapies.

II. Progenitor Cell Mobilization

The mobilization of progenitor cells from bone marrow to the peripheral circulation is highly regulated under both normal physiological conditions and stress.^53,54 In adult bone tissue, progenitor cells are retained predominantly in specialized microenvironments near the endosteum (i.e., the osteoblast niche), where they interact with spindle-shaped, N-cadherin-expressing osteoblasts,^55,56 and in the sinusoids (i.e., the vascular niche), where they interact with SDF-1-expressing reticular cells.^57–59 Many different cell types, matrix proteins, and soluble factors cooperatively regulate the self-renewal, differentiation, and maintenance of progenitor cells^{55–57,60–65}; however, the bulk of experimental evidence suggests that progenitor cell retention and release are largely governed by two pathways, one of which is dependent on stromal-cell-derived factor 1 (SDF-1, also called CXC chemokine ligand 12 [CXCL12]) and the SDF-1 receptor CXC chemokine receptor 4 (CXCR4), and the other on α4β1-integrin (also called very late antigen-4 [VLA-4]).^{57,59,60,66–69} Initially, SDF-1/CXCR4 and α4β1-integrin signaling appear to proceed independently; for example, the α4β1-integrin antagonist Groβ can mobilize progenitor cells in mice transplanted with CXCR4-knockout bone marrow.⁷⁰ However, results from our recent studies suggest that c-kit, a receptor tyrosine kinase that binds stem cell factor (SCF), is an integral downstream component of both pathways.⁷¹

A. SDF-1/CXCR4

CXCR4 is a G protein-coupled receptor composed of 352 amino acids with seven transmembrane helices^72–74 and is broadly expressed by both mononuclear cells and progenitor cells in the bone marrow.^72–78 The ligand for CXCR4, SDF-1, is a secreted or membrane-bound protein that is abundantly expressed by osteoblasts, endothelial cells, and a subset of reticular cells in the osteoblast and vascular niches.^57,79–81 SDF-1/CXCR4 signaling induces the directional migration of cells and is involved in many biological processes, including cardiovascular organogenesis, hematopoiesis, immune response, and cancer metastasis. Interactions between SDF-1 and CXCR4 are crucial for maintaining populations of hematopoietic stem cells (HSCs) in adult animals,^{57,66,82–87} and mice that lack either SDF-1 or CXCR4 exhibit nearly identical phenotypes characterized by late gestational lethality and defects in bone marrow colonization, B-cell lymphopoiesis, blood vessel formation, and cardiac septum formation.^{83,85,88–90} Thus, the SDF-1/CXCR4 axis appears to have a fundamental role in both vasculogenesis and cardiogenesis.

The roles of SDF-1 and CXCR4 in bone marrow progenitor cell retention and release are well established.⁶⁶ Selective antagonism of CXCR4 with the pharmacological agent AMD3100 rapidly and potently mobilizes bone marrow progenitor cells in both animals and humans,^86,91–93 and systemically injected bone marrow progenitor cells accumulate predominantly in subdomains of bone marrow microvessels that are rich in SDF-1 expression.^68,94 Notably, both SDF-1 and CXCR4 expression are upregulated by relatively low oxygen tension (hypoxia) in discrete regions of the bone marrow and by the activation of hypoxia inducible factor 1 (HIF-1).^87,95–100

B. α4-integrin

Integrins are heterodimeric transmembrane receptors composed of non-covalently joined α and β subunits and have the remarkable ability to transmit both incoming and outgoing signals across the cell membrane.^101,102 Integrins usually induce signaling pathways by acting synergistically with growth-factor receptors to regulate cell shape, adhesion, migration, proliferation, and differentiation; but both can also function independently.^103,104 The α4-integrins, α4β1 and α4β7, bind to vascular cell adhesion molecule 1 (VCAM-1), which is expressed on the surface of endothelial and stromal cells^105–109 and to fibro-nectin in the extracellular matrix. These binding interactions are crucial for the adhesion of progenitor cells to the microenvironment and, consequently, to progenitor cell retention and recruitment.^61,64,65 The expression of α4-integrin is downregulated during progenitor cell mobilization,^110,111 and the cleavage of VCAM-1 and α4-integrin is a critical step during cytokine-induced bone marrow progenitor cell mobilization.¹¹² In adult mice, deletion of α4-integrin persistently alters the distribution of progenitor cells,^108,113,114 and antibody-mediated α4-integrin blockade mobilizes progenitor cells in both animals and humans.^62,115,116 The level of α4-integrin expression on mobilized peripheral blood progenitor cells is inversely correlated with bone marrow homing and predicts the rate of engraftment in patients who have received autologous progenitor cell transplantation.¹¹⁷ Furthermore, we have shown that the transient blockade of α4-integrin activity leads to higher peripheral blood EPC levels, greater EPC-mediated neovascularization, and less adverse cardiac remodeling after myocardial infarction, and that α4-integrin antibodies can release bone marrow EPCs from immobilized VCAM-1 or bone marrow stromal cells.³¹ Thus, α4-integrin antibodies appear to mobilize EPCs from the bone marrow by disrupting VCAM-1:α4-integrin binding.

C. c-kit

c-kit (also called CD117) is a type III receptor tyrosine kinase expressed predominantly in bone marrow stem/progenitor cells¹¹⁸ and has recently been identified as a marker for EPC and cardiac progenitor cell identity.^119–121 The ligand for c-kit, namely, SCF, is expressed in bone marrow endothelial cells and stromal cells as either a membrane-bound protein or a soluble one.^122,123 Dimers of SCF bind to c-kit, which triggers c-kit homodimerization and the phosphorylation of specific c-kit tyrosine residues.¹²⁴ The pattern of c-kit phosphorylation determines which signaling event is activated and can induce both positive and negative pathways.^124–126 SCF/c-kit signaling is essential for embryonic hematopoiesis,^127,128 and mutations that lead to the loss of c-kit (i.e., the W mutation), c-kit kinase activity (e.g., the W42 mutation), or SCF (the Sl mutation)¹²⁹ cause severe macrocytic anemia and death in utero or during the perinatal period. Notably, defects in c-kit activity are also associated with impaired vascular development and angiogenesis,^130–132 and cancer therapies that target c-kit are cardiotoxic.^133,134 c-kit also supports progenitor cell maintenance^135–137 and is a crucial component of cardiac regeneration.¹³⁸ After myocardial infarction, c-kit-positive bone marrow cells are recruited to the ischemic myocardium and facilitate cardiac repair by differentiating into cardiac cell lineages and by expressing angiogenic cytokines.^35,131

The mobilization of progenitor cells in response to α4-integrin blockade is markedly blunted in c-kit^W/W-V mutant mice, which are defective in c-kit kinase activity but have normal levels of c-kit expression and SCF binding at the cell surface.^{131,139–141} Thus, the kinase activity of c-kit appears to be crucial for progenitor cell mobilization, but the mechanism by which c-kit participates in the retention and release of progenitor cells is unclear. In the bone marrow, membrane-bound SCF can be cleaved by SDF-1 to form soluble SCF, which subsequently activates c-kit and leads to progenitor cell mobilization,⁵⁹ but the functional blockade of c-kit (with the c-kit–neutralizing antibody ACK2) has also been shown to mobilize bone marrow HSCs to the peripheral circulation and to enhance the engraftment of systemically injected donor bone marrow cells.¹⁴² Thus, both the activation and blockade of c-kit activity have been associated with progenitor cell mobilization.^59,142,143 Furthermore, c-kit is the only known receptor for SCF, but SCF is not always required for c-kit activity,^144–146 and neither SCF nor an SCF-neutralizing antibody are potent mobilizers,^147,148 so SCF-binding alone cannot adequately explain the role of c-kit in progenitor cell mobilization.^142,147,148

D. SDF-1/CXCR4–c-kit Signaling

The kinetics of bone marrow progenitor cell mobilization induced by a c-kit-neutralizing antibody (i.e., ACK2) and by antagonism of CXCR4 with the pharmacological CXCR4 antagonist AMD3100 are similar, so we investigated whether c-kit is involved in CXCR4-mediated bone marrow progenitor cell trafficking. Peripheral blood progenitor cell levels significantly increased, and bone marrow progenitor cell levels significantly declined after AMD3100 was injected into wild-type mice, but not after it was injected into c-kit kinase-defective (c-kit^W/W-V) mice.⁷¹ To determine which specific subpopulations of bone marrow cells were affected by the c-kit kinase deficiency, we developed a short-term, in vivo bone marrow clearance/repopulation assay. AMD3100 was administered to wild-type and c-kit^W/W-V mice, and then labeled bone marrow mononuclear cells were injected into the peripheral circulation and allowed to repopulate the bone marrow. Significantly fewer systemically administered CXCR4-expressing progenitor cells were observed in the bone marrow of c-kit^W/W-V mice than in the bone marrow of wild-type mice.

AMD3100 also failed to mobilize bone marrow progenitor cells that expressed a constitutively active c-kit kinase (c-kit^D816V) mutation. Thus, both the loss and the constitutive activation of c-kit kinase activity impaired AMD3100-induced bone marrow progenitor cell mobilization, which may seem contradictory. However, bone marrow progenitor cell levels were lower in c-kit kinase-defective mice than in wild-type mice before mobilization, and, after mobilization, systemically administered CXCR4-positive progenitor cells could not repopulate the bone marrow of c-kit kinase-defective mice. These observations suggest that c-kit kinase inactivation blocks the retention of CXCR4-positive progenitor cells and, consequently, that the cells susceptible to AMD3100-induced mobilization are (in effect) already mobilized.

In isolated bone marrow mononuclear cells, SDF-1/CXCR4 signaling upregulates, and the antagonism or genetic deletion of CXCR4 downregulates, c-kit phosphorylation. These results, as well as the lack of AMD3100-induced progenitor cell mobilization in mice with bone marrow cells that expressed a constitutively active c-kit mutant, suggest that CXCR4-mediated mobilization requires c-kit deactivation. Thus, AMD3100 and G-CSF appear to induce progenitor cell mobilization through fundamentally different mechanisms, because G-CSF-induced mobilization requires an increase in c-kit activation.⁵⁹ Furthermore, G-CSF-induced mobilization occurs 3–5 days after administration and is accompanied by an increase in the number of progenitor cells present in the perivascular niche,¹⁴⁹ whereas AMD3100-induced mobilization occurs within a few hours and, consequently, is unlikely to be preceded by the perivascular accumulation of progenitor cells. The two agents also appear to mobilize different subpopulations of progenitor cells, and more cells are mobilized when G-CSF and AMD3100 are combined than when G-CSF is administered alone.^150–155 Collectively, these observations would suggest that G-CSF and other slow-acting agents increase c-kit phosphorylation by upregulating SCF, which (by itself) has only a modest effect on mobilization but potently promotes progenitor cell proliferation; if so, G-CSF-induced mobilization may be delayed until an adequate surplus of progenitor cells is available for release to the peripheral blood.¹⁴⁹ Conversely, fast-acting agents, such as AMD3100, may mobilize progenitor cells directly by reducing c-kit phosphorylation in the perivascular niche.¹⁵⁶ This hypothesis is also supported by recent evidence that progenitor cells can be rapidly mobilized by the administration of a c-kit neutralizing antibody.¹⁴²

E. α4-integrin–c-kit Signaling

Because c-kit also appears to participate in progenitor cell mobilization through an α4-integrin-mediated mechanism,¹⁴⁰ and interactions between α4-integrin and VCAM-1 support the adhesion and retention of mononuclear cells in the bone marrow,³¹ we performed a series of in vitro experiments to determine whether the phosphorylation state of c-kit is altered by α4-integrin-mediated adhesion. Wild-type bone marrow mononuclear cells were applied to VCAM-1-coated or uncoated plates, allowed to adhere for 15 min, incubated with or without AMD3100 for another 15 min, and then c-kit phosphorylation at tyrosine 719 was evaluated. Phosphorylated c-kit levels were notably higher in adherent wild-type cells (i.e., cells from VCAM-1– coated plates) than in nonadherent wild-type cells (i.e., cells from uncoated plates), and treatment with an α4-integrin-blocking antibody reduced phosphorylated c-kit levels, whereas treatment with the CXCR4 ligand SDF-1 markedly increased c-kit phosphorylation. Furthermore, both SDF-1 and SCF induced c-kit phosphorylation, but c-kit levels were highest when the cells were incubated with both factors, and AMD3100 treatment suppressed SDF-1-induced, but not SCF-induced, c-kit phosphorylation. Collectively, these observations suggest that α4-integrin-mediated mononuclear cell adhesion is associated with an increase in phosphorylated c-kit levels, and that in adherent mononuclear cells, SDF-1 upregulates, and AMD3100 downregulates, c-kit phosphorylation. Thus, SDF-1- and SCF-induced c-kit activation may occur independently and regulate different cellular activities (Fig. 1).

FIG. 1.

c-kit is a common component of two signaling pathways that regulate progenitor cell trafficking. Progenitor cell retention and release are largely governed by two pathways, one of which is dependent on the binding of SDF-1 to CXCR4 and the other on α4-integrin/VCAM-1 binding. Both interactions lead to the phosphorylation of c-kit, which is crucial for the retention of progenitor cells in the bone marrow. AMD3100 disrupts the SDF-1/CXCR4 interaction, which reduces c-kit phosphorylation and mobilizes progenitor cells from the bone marrow. SCF also increases phosphorylated c-kit levels by binding directly to c-kit, but disruption of the SCF/c-kit interaction does not appear to induce progenitor cell mobilization. Furthermore, AMD3100 suppresses SDF-1-induced, but not SCF-induced, c-kit phosphorylation, and phosphorylated c-kit levels are higher when cells are incubated with both SDF-1 and SCF than with either individual factor. Thus, SDF-1 and SCF appear to regulate c-kit phosphorylation independently and likely coordinate different cellular activities.

III. Progenitor Cell Recruitment and Retention

Mobilized progenitor cells are recruited from the peripheral circulation to the ischemic region, where they become incorporated into the growing vasculature.^14,157 Several of the intermediate steps during progenitor cell recruitment, including chemotaxis, transendothelial migration, and adhesion to single layers of mature endothelial cells and integrin, are regulated by SDF-1/CXCR4 binding,^{30,99,158–160} but the mechanisms and downstream components of SDF-1/CXCR4 signaling at the injury site are poorly understood. SDF-1 expression is significantly elevated in the plasma of patients with acute myocardial infarction,¹⁶¹ and the expression of both CXCR4 and SDF-1 is elevated in ischemic myocardium,^158,162,163 whereas the blockade of SDF-1/CXCR4 signaling diminishes progenitor cell recruitment,^{99,138,162,164} and impairments in CXCR4 signaling contribute to the reduced angiogenic potency of EPCs from patients with coronary artery disease and related conditions such as aging and diabetes.^164–167 SDF-1 and CXCR4 may also participate in vascular remodeling by recruiting smooth muscle progenitor cells¹⁶⁸ and protect cardiomyocytes against ischemia/reperfusion damage by activating the antiapoptotic kinases Akt and extracellular-regulated kinase.¹⁶⁹

Hypoxia induces SDF-1 expression at the injury site,¹⁶³ where platelets are an important source of SDF-1 expression,^170,171 and we have recently shown that hypoxic preconditioning enhances the recruitment of cardiosphere-derived Lin-negative, c-kit-positive progenitor (CLK) cells (i.e., cardiac progenitor cells) by inducing CXCR4 expression.¹³⁸ CXCR4 expression is much lower in CLK cells than in bone marrow mononuclear cells under normoxic conditions, but increases significantly in response to hypoxia. The increase is accompanied by elevations in SDF-1 expression and preceded by the upregulation of HIF-1α, whereas the siRNA-mediated inactivation of HIF-1α abolishes CXCR4 upregulation. Hypoxic treatment also increased the migration of isolated CLK cells toward SDF-1 in a CXCR4-dependent manner, and hypoxic preconditioning was associated with a 2.5-fold increase in the recruitment of systemically injected CLK cells to the ischemic myocardium of mice after surgically induced myocardial infarction. The recruited cells expressed cardiac troponin I, von Willebrand factor, and smooth muscle actin, indicating that CLK cells can differentiate into cardiomyocytes, endothelial cells, and vascular smooth muscle cells, respectively.

IV. Therapeutic Implications

Both the release of progenitor cells from the bone marrow to the peripheral blood and the recruitment and retention of progenitor cells in ischemic tissue are regulated by interactions between SDF-1 and CXCR4.^{37,66,68,75,85,91,99,158,162,163,172–174} The interaction must be disrupted before progenitor cells can be mobilized from the bone marrow to the peripheral circulation and restored to enable retention of the mobilized cells in the ischemic tissue. Thus, the effectiveness of progenitor cell therapy is crucially dependent on how mobilization is induced and on the level of SDF-1 expression in the ischemic region at the time of cell administration. To date, G-CSF is the most frequently used mobilizing agent in clinical trials of progenitor cell therapy,^{28,29,44,46,48,51,52} but the efficacy results from many of these trials have been disappointing, perhaps because G-CSF mobilizes progenitor cells by activating extracellular proteases that irreversibly cleave cell-surface adhesion molecules, including α4-integrin, VCAM-1, and CXCR4.^92,175 Thus, the retention of G-CSF-mobilized cells in the ischemic region may be impaired, and mobilizing agents that reversibly disrupt SDF-1/CXCR4 binding, such as AMD3100, may improve the effectiveness of cell therapy³⁷ (Fig. 2).

FIG. 2.

Mechanisms of therapeutic progenitor cell mobilization. (A) Growth factors (e.g., G-CSF) mobilize progenitor cells by activating extracellular proteases that irreversibly cleave cell-surface adhesion molecules, including α4-integrin, c-kit, and CXCR4, which are crucial for progenitor cell retention in the ischemic region. (B) Receptor antagonists, such as the CXCR4 antagonist AMD3100 or the α4-integrin-blocking antibody Natalizumab, mobilize progenitor cells by reversibly blocking interactions that bind progenitor cells to the bone marrow substrate without cleaving the adhesion molecules. Thus, the use of reversible antagonists, rather than growth factors, for therapeutic progenitor cell mobilization may increase the number of mobilized cells retained in the ischemic region.

Cardiac SDF-1 expression is upregulated within minutes to an hour after myocardial infarction but declines 4–7 days later.¹⁵⁸ Thus, if progenitor cells are administered days after, or even years after (i.e., in patients with established coronary disease), the infarct event, retention of the administered cells is likely to be poor in the ischemic region. Thus, several approaches to increase SDF-1 levels in the ischemic region are currently being investigated. Locally delivered SDF-1 protein increased vascular growth in the injured limbs of mice after surgically induced hind-limb ischemia,¹⁷⁶ and SDF-1 also improved cardiac function after ischemic myocardial injury by increasing progenitor cell recruitment and angiogenesis and by reducing scar formation.¹⁷⁷ However, local SDF-1 delivery is limited by the rapid diffusion of the administered protein and by the activity of proteases in the inflammatory environment of the injury. Furthermore, the induction of SDF-1/CXCR4 signaling also stimulates the production of matrix metalloproteases (MMPs),^178–181 including MMP-2, which (in concert with several exopeptidases) cleaves SDF-1 to produce a neurotoxin that has been implicated in some forms of dementia.^163,182 To overcome this limitation, Segers et al. designed an SDF-1 variant that retains the chemotactic properties of the native molecule but is resistant to MMP-2 and exopeptidase cleavage. Nanofiber-mediated delivery of this construct, S-SDF-1 (also called S4V), promoted progenitor cell recruitment and improved cardiac function in a murine model of myocardial infarction.¹⁸³ Furthermore, sustained SDF-1 release has been achieved by covalently linking it to a polyethylene glycol fibrin patch.¹⁷⁴ When the patch was applied to the surface of infarcted mouse hearts, SDF-1 continued to be released for 28 days, and the treatment was associated with greater numbers of incorporated c-kit-positive cells (i.e., progenitor cells) and with improvements in left ventricular function.

Local SDF-1 expression has also been increased through the administration of genetically engineered MSCs.^184,185 Intravenous injections of either unmodified MSCs or MSCs that overexpressed SDF-1 to rats after acute myocardial infarction were associated with improvements in cardiac function, and the beneficial effects appeared to evolve primarily through the preservation of preexisting cardiomyocytes rather than the generation of new cardiomyocytes within the infarct zone.¹⁸⁴ Vascular density, cardiomyocyte survival, and cardiac myosin-positive area were greater in animals treated with the modified cells than in those treated with unmodified cells, and SDF-1 overexpression also increased the number of small cardiac myosin-expressing cells that had not differentiated into mature cardiac myocytes, but were capable of depolarizing and, consequently, may have contributed to improvements in contractile function.¹⁸⁵

V. Summary

Progenitor cell retention and release are largely governed by SDF-1/ CXCR4 and α4-integrin signaling. The initial steps of these two pathways appear to proceed independently, but both regulate c-kit phosphorylation,⁷¹ and the mobilization of progenitor cells in response to either CXCR4 antagonism or α4-integrin blockade is impaired by the loss of c-kit kinase activity.^{131,139–141} Furthermore, bone marrow progenitor cell levels are lower in c-kit kinase-defective mice than in wild-type mice before mobilization, and systemically administered CXCR4-positive progenitor cells cannot repopulate the bone marrow in the absence of c-kit kinase activity.⁷¹ Collectively, these observations suggest that c-kit-kinase inactivation blocks the retention of CXCR4-positive progenitor cells in the bone marrow, and that c-kit could function as a final common mediator of fundamental importance to the regulation of bone marrow progenitor cell trafficking.

SDF-1/CXCR4 and α4-integrin signaling are also crucial for the retention of progenitor cells in the ischemic region, which may explain, at least in part, why clinical trials of progenitor cell therapy have failed to display the efficacy observed in preclinical investigations. The lack of effectiveness is often attributed to poor retention of the transplanted cells,^21,22,27 and, to date, most of the trial protocols have mobilized cells via G-CSF administration, which activates extracellular proteases that irreversibly cleave cell-surface adhesion molecules, including α4-integrin and CXCR4.^92,175 Thus, the retention of G-CSF-mobilized cells in the ischemic region may be impaired, and mobilizing agents that reversibly disrupt SDF-1/CXCR4 binding, such as AMD3100, may improve patient response.³⁷ Efforts to supplement SDF-1 levels in the ischemic region may also improve progenitor cell recruitment and the effectiveness of stem cell therapy.