Postinfarct Cytokine Therapy Regenerates Cardiac Tissue and Improves Left Ventricular Function
http://www.100md.com
Buddhadeb Dawn, Yiru Guo, Arash Rezazade
参见附件。
the Institute of Molecular Cardiology (B.D., Y. Guo, A.R., A.B.S., G.H., S.T., J.V., Y. Gu, S.D.P., R.B.) and Institute of Cellular Therapeutics (Y.H., S.T.I.), University of Louisville, Kentucky
Cardiovascular Research Institute (J.K., P.A.), Department of Medicine, New York Medical College, Valhalla.
Abstract
We systematically investigated the comparative efficacy of three different cytokine regimens, administered after a reperfused myocardial infarction, in regenerating cardiac tissue and improving left ventricular (LV) function. Wild-type (WT) mice underwent a 30-minute coronary occlusion followed by reperfusion and received vehicle, granulocyte colony-stimulating factor (G-CSF)+Flt-3 ligand (FL), G-CSF+stem cell factor (SCF), or G-CSF alone starting 4 hours after reperfusion. In separate experiments, chimeric mice generated by reconstitution of radioablated WT mice with bone marrow from enhanced green fluorescent protein (EGFP) transgenic mice underwent identical protocols. Mice were euthanized 5 weeks later. Echocardiographically, LV function was improved in G-CSF+FL– and G-CSF+SCF–treated but not in G-CSF–treated mice, whereas LV end-diastolic dimensions were smaller in all three groups. Morphometrically, cytokine-treated hearts had smaller LV diameter and volume. Numerous EGFP-positive cardiomyocytes, capillaries, and arterioles were noted in the infarcted region in cytokine-treated chimeric mice treated with G-CSF+FL or G-CSF+SCF, but the numbers were much smaller in G-CSF–treated mice. G-CSF+FL therapy mobilized bone marrow–derived cells exhibiting increased expression of surface antigens (CD62L and CD11a) that facilitate homing. We conclude that postinfarct cytokine therapy with G-CSF+FL or G-CSF+SCF limits adverse LV remodeling and improves LV performance by promoting cardiac regeneration and probably also by exerting other beneficial actions unrelated to regeneration, and that G-CSF alone is less effective.
Key Words: myocardial regeneration cytokine stem cell ischemia/reperfusion bone marrow left ventricular function
Introduction
Despite significant advances in the treatment of acute myocardial infarction (MI), no therapies are currently available to restore dead myocardium. Although mobilization of bone marrow cells (BMCs) by cytokines has been suggested to regenerate cardiac tissue after MI and to produce functional improvement, the reports on the effects of cytokine treatment are conflicting.1–6 Furthermore, previous studies of cytokines have used models of permanent coronary ligation1–3,5 that do not reflect the fact that most patients with acute MI undergo coronary reperfusion. In addition, these studies have administered cytokines as a pretreatment1,2 or performed splenectomy1,2,5 or both,1,2 neither of which would be clinically feasible. Although various cytokines have been tested, the relative efficacy of different treatments or combinations of treatments has not been elucidated. Finally, the mechanism whereby cytokines improve left ventricular (LV) function and structure after MI remains obscure. It has been proposed that cytokines mobilize BMCs with subsequent homing to the infarcted tissue and transdifferentiation into cardiac lineage,1–3 but this hypothesis remains unproven. Thus, numerous unresolved issues persist regarding the use of cytokines as a strategy to achieve cardiac repair after MI.
Accordingly, in the present study, we examined the effect of three cytokines (granulocyte colony-stimulating factor [G-CSF], stem cell factor [SCF], and Flt-3 ligand [FL]) on cardiac repair. The specific goals were to determine: (1) whether cytokine therapy is efficacious in the setting of a reperfused MI; (2) whether it is efficacious when started after the MI and in the presence of an intact spleen; (3) which cytokine or cytokine combination, among G-CSF, FL, and SCF, is more effective in improving LV remodeling and function; (4) whether cytokine-mobilized BMCs home to the heart and transdifferentiate into the cardiac lineage; and (5) whether cytokines facilitate BMC homing to the injured myocardium by altering the expression profile of adhesion molecules in mobilized BMCs. A well-characterized murine model of temporary coronary occlusion followed by reperfusion (a setting designed to mimic the most common clinical setting of acute MI in humans) was used.7 A thorough assessment of LV function and anatomy was performed in conjunction with careful measurements of cardiac regeneration by confocal microscopy.
Materials and Methods
A detailed description of all materials and methods is provided in the online supplement, available at http://circres.ahajournals.org.
Generation and Characterization of EGFP Chimeric Mice
EGFP chimeric mice were generated8 by reconstituting radioablated adult male mice (B6,129 strain) with BMCs from syngeneic enhanced green fluorescent protein (EGFP) transgenic mice. On average, 85% of the peripheral blood nucleated cells were positive for EGFP in the recipients.
Experimental Protocol
Adult male wild-type (WT; B6,129 strain, 10 to 18 weeks of age) and EGFP chimeric mice were used. The overall experimental design is summarized in supplemental Figure I.
Myocardial Infarction
Anesthetized mice underwent a 30-minute occlusion of the left anterior descending coronary artery followed by reperfusion. Starting 4 hours after reperfusion, mice received daily subcutaneous injections of vehicle or cytokine as detailed in supplemental Figure I.
Echocardiography
Serial echocardiograms were obtained at baseline (2 days before coronary occlusion) and 2 and 35 days after reperfusion. Systolic and diastolic anatomical and functional parameters were calculated using standard methods.1
Morphometry and Histology
At the end of the follow-up period, the heart was perfusion fixed in 10% neutral buffered formalin,1,9 and the left ventricle was sectioned serially perpendicular to its longitudinal axis, processed, and embedded in paraffin. The infarct volume fraction was calculated by computerized planimetry (Image-Pro Plus; Media Cybernetics) of digital images of three Masson’s trichrome-stained serial LV sections taken at 0.5- to 1.0-mm intervals along the longitudinal axis.10 The midsection was used to measure LV diameter and volume.11,12 All anatomical parameters were corrected according to a uniform sarcomere length.13
Immunohistochemistry was performed in formalin-fixed 4-μm-thick histological sections. Cardiomyocytes were recognized by the presence of -sarcomeric actin and troponin T, endothelial cells by plaletet/endothelial cell adhesion molecule-1 (PECAM-1) and von Willebrand factor, and smooth muscle cells by -smooth muscle actin. Colocalization of cell-specific markers with EGFP was used to identify cells that originated from BMCs. All of the primary and secondary antibodies used in these studies are specified in supplemental Table I. A quantitative analysis of cardiomyocyte regeneration was performed in 3 to 7 EGFP chimeric mouse hearts in each of groups V through VIII.
Assessment of Cardiomyocyte Proliferation
The percentage of cardiomyocyte nuclei positive for 5-bromodeoxyuridine(BrdU) and Ki671 was determined at 14 days after reperfusion in 4 to 6 mice in each of groups IX through XII. In addition, the percentage of cardiomyocyte nuclei positive for Ki67 at 35 days after MI was determined in 6 mice in each of groups I through IV.
Flow Cytometry
After vehicle or cytokine treatment, peripheral blood was harvested from mice in groups XIII through XVI at serial time points, stained with antibodies, and lin–/Sca-1+/c-kit+ cells were counted from the lymphoid gate with a multiparameter, live sterile cell sorter.14,15 The flow cytometric procedure is summarized in supplemental Figure II. The expression of adhesion molecules on lin–/Sca-1+/c-kit+ cells from untreated murine bone marrow and peripheral blood of mice (groups XVII through XX) treated with identical dosages of vehicle or cytokines as groups I through IV, respectively, was also assessed by flow cytometry.
Statistical Analysis
Data are reported as mean±SEM. Morphometric and histologic data were analyzed with one-way ANOVA. Serial echocardiographic parameters were measured with a two-way (time and group) ANOVA followed by Student t tests with the Bonferroni correction.
Results
Exclusions
A total of 188 mice (162 WT and 26 EGFP chimeric) were used. One hundred fifty-six mice were assigned to the MI studies (groups I through XII) and 32 mice to the cellular mobilization (groups XIII through XX) studies. Twenty-nine mice died in the perioperative period. Nine mice were excluded from the study because of failure of the coronary occluder, leaving a total of 14, 19, 20, and 18 WT mice in groups I through IV, respectively, and 3, 7, 6, and 5 EGFP chimeric mice in groups V through VIII, respectively. No mouse was excluded from the cellular mobilization studies (groups XIII through XX; n=4 per group).
Myocardial Infarct Size
The impact of cytokine therapy on LV structural and functional parameters was assessed in groups I through IV. The infarct volume fraction was similar in the four groups (supplemental Figure III), indicating that the extent of myocardial cell death produced by the coronary occlusion was comparable. The infarct volume fraction measures the volume of scarred tissue expressed as a fraction of the entire left ventricle.10–12 Because the small islands of regenerated cardiomyocytes scattered throughout the scar area were included in the "infarct" region, in cytokine-treated hearts, the calculation of the infarct volume fraction was not affected by myocardial regeneration.
G-CSF+FL and G-CSF+SCF Therapy Improves LV Systolic Function
Before coronary occlusion (baseline), all parameters of LV function, measured by echocardiography, were similar in groups I through IV (Figure 1). At 48 hours after reperfusion, the degree of LV systolic functional impairment did not differ among groups (Figure 1), indicating that the injury sustained during ischemia/reperfusion was comparable. In vehicle-treated mice, there was further functional deterioration between 48 hours and 35 days after reperfusion (Figure 1). In contrast, in G-CSF+FL-treated (group II) and G-CSF+SCF-treated (group III) mice, LV systolic function remained unchanged or improved at 35 days compared with 48 hours (Figure 1). Mice in groups II and III exhibited significantly higher LV ejection fraction, fractional area change, and fractional shortening and lower LV end-systolic diameter compared with vehicle-treated mice (group I) (Figure 1). Compared with group I, myocardial function in the infarct region was greater in groups II and III (systolic thickening fraction in the infarct wall, 18.6±5.9%, 26.5±2.8%, 24.8±5.4%, and 20.9±6.6% in groups I, II, III, and IV, respectively), but the differences were not statistically significant. Compared with group I, myocardial function was also enhanced in the nonischemic LV region in groups II and III (systolic thickening fraction, 33.2±4.3%, 43.9±5.3%, and 54.4±3.2% [P<0.05 versus group I] and 35.1±4.1% in groups I, II, III, and IV, respectively). In mice treated with G-CSF alone (group IV), there was only a marginal improvement in parameters of LV systolic function (Figure 1E through 1G).
Cytokine Therapy Halts LV Remodeling
Both morphometric and echocardiographic analyses demonstrated that administration of cytokines attenuated adverse LV remodeling 5 weeks after infarction. Morphometrically, LV chamber diameter and LV chamber volume were lower in groups II, III, and IV compared with group I (Figure 2A and 2B; supplemental Figure IV). Echocardiographically, LV anatomical parameters were similar in groups I through IV, both before coronary occlusion (baseline) and 48 hours after reperfusion (Figure 2C and 2D). In vehicle-treated mice (group I), there was significant LV dilatation between 48 hours and 35 days after reperfusion (Figure 2), indicative of adverse LV remodeling. Compared with group I, both echocardiographic (Figure 2C and 2D) and morphometric (supplemental Figure IV) analyses showed smaller LV diameter and LV end-diastolic volume in cytokine-treated mice (groups II through IV). However, the reduction in LV dilatation in group IV was not statistically significant by echocardiography (Figure 2C and 2D), and the reduction in LV volume in group II was not statistically significant by morphometry (supplemental Figure IV). By morphometry, the infarct wall thickness improved in cytokine-treated (groups III and IV) mice (Figure 2E). Interestingly, mice that received G-CSF alone (group IV) also demonstrated a comparable improvement in morphometric parameters of LV remodeling (supplemental Figure IV) despite borderline functional improvement (Figure 1).
Cytokine Therapy Promotes Regeneration of Cardiomyocytes
In EGFP chimeric mice, the effects of cytokines on LV function and anatomy after infarction were directionally concordant with those in WT mice (supplemental Figures V and VI, respectively), despite the fact that LV size was smaller in the chimeric mice (supplemental Figure VI) compared with WT mice (Figure 2). Histologic quantitation of myocardial regeneration was performed in hearts (n=3 to 7 per group) from EGFP chimeric mice (groups V through VIII). Cardiomyocytes derived from BMCs were identified by concomitant positivity for -sarcomeric actin and EGFP (Figure 3). The extent of regeneration was assessed by measuring the area occupied by EGFP+ myocytes rather than their number; the former parameter more accurately reflects the functional impact of regeneration. In the infarcted region, EGFP+ regenerated cardiomyocytes occupied only 0.9±0.1% of the myocardial area in group I; in contrast, this fraction of reconstituted myocardium was much higher in groups II, III, and IV (14.9±1.5%, 10.8±1.9%, and 4.7±1.0% of the myocardial area, respectively). The area occupied by regenerated cardiomyocytes was similar in the infarct and in the border zone. In contrast, in the noninfarcted region, the area occupied by EGFP+ myocytes was uniformly low in all four groups (0.2±0.1%, 0.6±0.3%, 0.4±0.1%, and 0.4±0.1% of myocardial area in groups I, II, III, and IV, respectively; Figure 4). Newly formed myocytes expressed connexin 43 (Figure 5), indicating functional integration with adjacent myocytes. EGFP+ cells that were negative for cardiomyocyte markers constituted 1.2±0.6%, 3.9±0.6%, 3.5±0.6%, and 2.9±0.8% of the infarcted myocardial area and 0.7±0.1%, 1.9±0.4%, 0.6±0.1%, and 1.2±0.3% of the viable myocardial area in groups I, II, III, and IV, respectively (Figure 4). Thus, a fraction of the BMCs that homed to the infarcted myocardium differentiated into noncardiomyocyte cell types. In the ischemic zone, most EGFP+ cells were cardiomyocytes; in the nonischemic zone, the reverse was observed.
Cardiomyocyte Proliferation
The effect of cytokine administration on cardiomyocyte proliferation at 14 days after MI was assessed in groups IX through XII by examining BrdU incorporation (Figure 6A and 6C) in cardiomyocyte nuclei. In addition, 6 hearts from each of groups I through IV were examined for Ki67 expression at 35 days after MI (Figure 6B and 6D). Treatment with G-CSF+FL, G-CSF+SCF, or G-CSF resulted in a significant increase in BrdU+ and Ki67+ cardiomyocytes in the infarcted region, indicating increased numbers of cycling myocytes (Figure 6). The increase observed with G-CSF alone was significantly less than that seen with G-CSF+FL and G-CSF+SCF. In contrast, the number of cycling myocytes positive for Ki67 in the nonischemic zone at 35 days after MI was comparable in all groups (data not shown).
Regeneration of the Vasculature
Colocalization of EGFP positivity with -smooth muscle actin positivity in arteriolar smooth muscle cells (Figure 7A) and PECAM-1 positivity in endothelial cells (Figure 7B) were noted predominantly in the infarct area and, to a lesser extent, in the viable myocardium. Thus, EGFP+ BMCs differentiated not only into cardiomyocytes but also into vascular structures necessary to supply blood to the newly formed myocardium in the infarcted region.
Mobilization of Lin–/Sca-1+/C-kit+ BMCs
The efficacy of different cytokine regimens in mobilizing lin–/Sca-1+/c-kit+ BMCs was examined by flow cytometry in groups XIII through XVI. Compared with groups XIII, XV, and XVI, G-CSF+FL (group XIV) therapy resulted in mobilization of a significantly higher number of lin–/Sca-1+/c-kit+ BMCs (0.1±0.1, 28.8±3.9, 5.8±1.3, and 4.0±0.7 cells/μL at 5 days in groups XIII, XIV, XV, and XVI, respectively). The increase in lin–/Sca-1+/c-kit+ BMC levels in group XIV was apparent at 3 days of therapy and became progressively more pronounced at 5 and 10 days. G-CSF+SCF (group XV) and G-CSF alone (group XVI) resulted in minimal changes in peripheral blood levels of lin–/Sca-1+/c-kit+ cells.
Cytokine Therapy Enhances Expression of Surface Antigens That Favor Homing
In groups XVII through XX, peripheral blood cells were analyzed by flow cytometry for expression of adhesion molecules on mobilized lin–/Sca-1+/c-kit+ BMCs. As shown in supplemental Figure 7, all three treatments (G-CSF+FL, G-CSF+SCF, and G-CSF alone) resulted in increased expression of CD62P (P-selectin), an adhesion molecule that favors attachment and extravasation of BMCs,16 and decreased expression of CD49d, an adhesion molecule that plays a critical role in retaining cells within the bone marrow.17 Both of these changes would be expected to promote release of BMCs from bone marrow and their migration into the cardiac interstitium. However, there were significant differences in adhesion molecule profile between G-CSF+FL–treated mice on the one hand and G-CSF+SCF– and G-CSF–treated mice on the other hand. Specifically, G-CSF+SCF and G-CSF treatment reduced whereas G-CSF+FL treatment increased the expression of CD62L (L-selectin) and CD11a (Figure 8), two adhesion molecules that favor extravasation and homing of BMCs.16–18 Furthermore, the reduction in expression of CD49e, an adhesion molecule that promotes BMC attachment,17 was comparatively less with G-CSF+FL and G-CSF+SCF treatment vis-à-vis G-CSF alone (supplemental Figure VII). In conjunction with the quantitative analysis of the number of circulating lin–/Sca-1+/c-kit+ cells reported above, these results demonstrate significant differences among different cytokine regimens with respect to the modulation of the adhesion molecules responsible for homing.
Discussion
Most basic and clinical studies of cardiac regeneration have focused on cell-based therapies,4 and relatively less is known regarding the role of cytokines in this process. The present investigation provides a number of important new findings. Our results demonstrate that: (1) cytokine therapy with a combination of G-CSF+FL or G-CSF+SCF started after acute MI, halts or attenuates adverse LV remodeling, and improves LV systolic function; (2) cytokine therapy results in mobilization of bone marrow stem cells (lin–/Sca-1+/c-kit+), which home to the injured myocardium and transdifferentiate into the cardiac lineage (cardiomyocytes, vascular smooth muscle cells, endothelial cells); (3) the combination of G-CSF and FL is superior to G-CSF+SCF or G-CSF alone with respect to its effects on LV remodeling, LV function, and mobilization of lin– /Sca-1+/c-kit+ BMCs into the peripheral blood; (4) all three cytokine regimens upregulated the expression of CD62P and downregulated the expression of CD49d on BMCs (CD62P [P-selectin] promotes the extravasation of BMCs, whereas CD49d is important for the retention of these cells in the marrow16,17); and (5) in contrast to G-CSF+SCF and G-CSF alone, G-CSF+FL treatment increases the expression of adhesion molecules (CD62L and CD11a) that favor homing of BMCs.16–18 CD62L (L-selectin) and the 2 integrin leukocyte function antigen-1 (CD11a) initiate and facilitate the attachment of peripheral blood cells to the endothelium, thereby enhancing the extravasation of these cells.17,18 Thus, the present investigation not only establishes the usefulness of cytokine therapy for therapeutic cardiac regeneration but also provides mechanistic insights into this phenomenon and identifies the superiority of a specific cytokine combination (G-CSF+FL). These results have important implications for the role of cytokine-based strategies in effecting cardiac repair after acute MI.
Previous studies have shown that cytokine therapies initiated before1,2 or after3 MI result in improved LV remodeling and function. However, to our knowledge, this is the first study to demonstrate that cytokine administration results in transdifferentiation of mobilized BMCs into the cardiac lineage and in expression of adhesion molecules that favor homing of BMCs to the injured myocardium. Furthermore, this is the first investigation to systematically compare the efficacy of G-CSF, FL, and SCF and their combination on postinfarct LV remodeling and function, and to identify the combination of G-CSF and FL as the most effective treatment.
A number of methodological considerations are in order. In contrast to previous studies, which examined models of permanent coronary occlusion,1 we elected to use a transient (30-minute) coronary occlusion followed by reperfusion. This model has greater clinical relevance because most patients with acute MI undergo either spontaneous or therapeutic coronary recanalization. We chose to initiate cytokine treatment after the MI because pretreatment (as used in previous studies1,2) would not be clinically feasible. For similar reasons, we elected not to perform a splenectomy (which was performed in previous investigations1,2). The doses of cytokines used in this study were selected because they are clinically relevant. The large group sizes (14 to 20 mice per group in groups I through IV) were deemed to be important to achieve robust conclusions. In an effort to perform a rigorous assessment of global as well as regional LV function, we used multiple echocardiographic parameters, including fractional shortening (parasternal short-axis M-mode images), fractional area change (parasternal short-axis 2-D images), ejection fraction, and systolic wall thickening. Treatment with G-CSF+FL or G-CSF+SCF consistently improved all of these parameters, providing solid evidence for the functional benefit of these therapies. This concept is further corroborated by the fact that two independent techniques (echocardiography and morphometry, performed by blinded investigators who were unaware of the data obtained with the other technique) demonstrated that G-CSF+FL and G-CSF+SCF were effective in mitigating the progressive LV chamber dilatation that is characteristic of adverse LV remodeling and ischemic cardiomyopathy.19 The data obtained by these two methods were generally concordant. However, in Figure 2E, we present the morphometric infarct wall thickness data because the measurement of the average integrated wall thickness by morphometry may be inherently more accurate.
We chose to examine G-CSF, SCF, and FL because these cytokines are used clinically (G-CSF) or have been well characterized with respect to BMC mobilization.15,20 Comparison of different cytokines or combinations thereof is important because these proteins are known to preferentially mobilize different subsets of BMCs.15,21 FL was not studied alone because, unlike G-CSF, it does not expand lin–/Sca-1+/c-kit+ cells.15 SCF was not studied as single therapy because it has low efficacy in mobilizing BMCs when used alone.20 Our results indicate that the combination of G-CSF and FL produced the most robust results with respect to myocyte regeneration, LV remodeling, LV function, and mobilization of lin–/Sca-1+/c-kit+ cells. G-CSF+SCF also resulted in significant beneficial effects, albeit to a lesser extent. The most plausible explanation for the superiority of G-CSF+FL is that this combination resulted in mobilization of much higher numbers of lin–/Sca-1+/c-kit+ cells into the peripheral blood, which would be expected to enhance the homing and engraftment of BMCs in the damaged myocardium. In addition to these quantitative differences, our flow cytometric analysis reveals qualitative differences; that is, the lin–/Sca-1+/c-kit+ cells mobilized by G-CSF+FL expressed higher levels of CD62L and CD11a vis-à-vis the cells mobilized by G-CSF+SCF (Figure 8). The expression of these adhesion molecules may facilitate the egress of primitive BMCs from the circulation into injured tissues that express chemoattractants.22
Having established the functional and structural benefits of cytokine therapy, we used EGFP chimeric mice to investigate the mechanism for these salubrious actions. Our results demonstrate that after administration of G-CSF+FL or G-CSF+SCF, mobilized BMCs home to the infarcted myocardium and differentiate into cardiomyocytes, smooth muscle cells, and endothelial cells. We found that cardiomyocyte regeneration was considerably greater with G-CSF+FL or G-CSF+SCF versus G-CSF alone (Figure 4). These findings were independently validated by the measurement of cycling (ie, BrdU+ and Ki67+) cardiomyocytes in non-EGFP chimeric mice (Figure 6).
The mechanism(s) whereby cytokines produce beneficial effects on LV dimensions and function after MI is likely complex. Although the association of greater mobilization of lin–/Sca-1+/c-kit+ cells with greater improvement in LV remodeling and function after G-CSF+FL treatment suggests that these cells may play an important role in cardiac regeneration, our results do not prove that the beneficial effects of cytokines were attributable to this subset of BMCs. The BMCs responsible for cardiac regeneration may be cell types other than the lin-/Sca-1+/c-kit+ cells (eg, multipotent progenitors, endothelial progenitors, etc.). It is also unknown whether, within the lin-/Sca-1+/c-kit+ subset of BMCs, the transdifferentiation was attributable to hematopoietic or nonhematopoietic cells. We recently identified a CXCR4+/Sca-1+/lin–/CD45– nonhematopoietic subfraction of BMCs that appear to be committed to differentiation into cardiac lineages and are mobilized after MI in mice as well as in humans.22 It is possible that these cells may account for the salubrious actions of G-CSF+FL and G-CSF+SCF noted in this study. Finally, mobilization of resident cardiac stem cells by cytokines could be an additional or alternative mechanism.23 In support of this hypothesis, we noted small cardiomyocytes in the infarct area of cytokine-treated hearts that were not positive for EGFP (Figure 3F, arrowheads), suggesting that cytokine treatment may have induced migration, differentiation, and possibly proliferation of resident cardiac progenitor cells, which may have contributed to cardiac repair.
Harada et al6 recently concluded that administration of G-CSF ameliorates LV remodeling and dysfunction by inhibiting apoptosis, not by promoting cardiac regeneration. Our results differ because they document transdifferentiation of BMCs into cardiac lineage (Figures 3, 4, and 7; predominantly in the infarct area). Nevertheless, we also noted functional improvement in the noninfarcted myocardium, where the extent of myocyte regeneration was minimal (Figure 4). This enhancement in contractility in the noninfarcted region is likely attributable to nonregenerative actions of cytokines (eg, inhibition of apoptosis, favorable changes in the interstitium, etc.), suggesting that multiple mechanisms contributed to the overall improvement in LV function after G-SCF+FL and G-SCG+SCF therapy. Although G-CSF alone induced little cardiomyocyte regeneration, it had favorable effects on LV remodeling (Figure 2C and 2D; supplemental Figure IV); however, G-CSF alone did not improve LV function significantly (Figure 1). As shown in Figure 1, measurements of LV systolic function were greater in group IV than in vehicle-treated mice, but the differences were not statistically significant. Despite the large sample sizes (n=15 to 19 per group), this could be attributable to a type II error. Alternatively, it is possible that other nonregenerative beneficial effects of G-CSF on the myocardium resulted in an improvement in LV anatomy which, however, did not result in improved function because of the less efficient regeneration of cardiomyocytes with G-CSF therapy.
In conclusion, we demonstrated that a clinically relevant protocol for cytokine therapy exerts profound beneficial effects on postinfarction cardiomyopathy. Specifically, administration of G-CSF+FL or G-CSF+SCF, started after a reperfused acute MI, results in regeneration of cardiomyocytes and coronary vessels by mobilized BMCs and in improvement in LV remodeling and function via myocyte regeneration and probably via other beneficial actions of cytokines. We also demonstrated that the combination of G-CSF+FL is superior to G-CSF+SCF or G-CSF alone. These results have significant therapeutic implications. Recent studies with G-CSF have concluded that this cytokine improves LV function and structure in patients with acute MI.24,25 Our findings suggest that the combination of G-CSF and FL should have even greater efficacy in the clinical arena, providing a rationale for testing G-CSF+FL in larger species and in patients.
Acknowledgments
This study was supported in part by National Institutes of Health grants HL-72410, HL-55757, HL-68088, HL-70897, HL-76794, HL-78825, HL-63442, and DK-069766, national American Heart Association (AHA) scientist development grant 0130146N, and AHA fellowship grants 0265087B and 0325372B.
Footnotes
This manuscript was sent to Stephen F. Vatner, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Original received October 21, 2005; revision received February 27, 2006; accepted March 9, 2006.
References
Orlic D, Kajstura J, Chimenti S, Limana F, Jakoniuk I, Quaini F, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Mobilized bone marrow cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci U S A. 2001; 98: 10344–10349.
Fukuhara S, Tomita S, Nakatani T, Ohtsu Y, Ishida M, Yutani C, Kitamura S. G-CSF promotes bone marrow cells to migrate into infarcted mice heart, and differentiate into cardiomyocytes. Cell Transplant. 2004; 13: 741–748. [Order article via Infotrieve]
Kawada H, Fujita J, Kinjo K, Matsuzaki Y, Tsuma M, Miyatake H, Muguruma Y, Tsuboi K, Itabashi Y, Ikeda Y, Ogawa S, Okano H, Hotta T, Ando K, Fukuda K. Nonhematopoietic mesenchymal stem cells can be mobilized and differentiate into cardiomyocytes after myocardial infarction. Blood. 2004; 104: 3581–3587.
Wollert KC, Drexler H. Clinical applications of stem cells for the heart. Circ Res. 2005; 96: 151–163.
Deten A, Volz HC, Clamors S, Leiblein S, Briest W, Marx G, Zimmer HG. Hematopoietic stem cells do not repair the infarcted mouse heart. Cardiovasc Res. 2005; 65: 52–63. [Order article via Infotrieve]
Harada M, Qin Y, Takano H, Minamino T, Zou Y, Toko H, Ohtsuka M, Matsuura K, Sano M, Nishi JI, Iwanaga K, Akazawa H, Kunieda T, Zhu W, Hasegawa H, Kunisada K, Nagai T, Nakaya H, Yamauchi-Takihara K, Komuro I. G-CSF prevents cardiac remodeling after myocardial infarction by activating the Jak-Stat pathway in cardiomyocytes. Nat Med. 2005; 11: 305–311. [Order article via Infotrieve]
Guo Y, Wu WJ, Qiu Y, Tang XL, Yang Z, Bolli R. Demonstration of an early and a late phase of ischemic preconditioning in mice. Am J Physiol. 1998; 275: H1375–H1387.
Colson YL, Wren SM, Schuchert MJ, Patrene KD, Johnson PC, Boggs SS, Ildstad ST. A nonlethal conditioning approach to achieve durable multilineage mixed chimerism and tolerance across major, minor, and hematopoietic histocompatibility barriers. J Immunol. 1995; 155: 4179–4188.
Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Bone marrow cells regenerate infarcted myocardium. Nature. 2001; 410: 701–705. [Order article via Infotrieve]
Anversa P, Beghi C, Kikkawa Y, Olivetti G. Myocardial infarction in rats. Infarct size, myocyte hypertrophy, and capillary growth. Circ Res. 1986; 58: 26–37.
Li Q, Li B, Wang X, Leri A, Jana KP, Liu Y, Kajstura J, Baserga R, Anversa P. Overexpression of insulin-like growth factor-1 in mice protects from myocyte death after infarction, attenuating ventricular dilation, wall stress, and cardiac hypertrophy. J Clin Invest. 1997; 100: 1991–1999.
Li B, Setoguchi M, Wang X, Andreoli AM, Leri A, Malhotra A, Kajstura J, Anversa P. Insulin-like growth factor-1 attenuates the detrimental impact of nonocclusive coronary artery constriction on the heart. Circ Res. 1999; 84: 1007–1019.
Anversa P, Olivetti G. The cardiovascular system. In: Page E, Fozzard HA, Solaro RJ, eds. The Heart. 1st ed. New York, NY: Oxford University Press; 2002: 75–144.
Orfao A, Ruiz-Arguelles A. General concepts about cell sorting techniques. Clin Biochem. 1996; 29: 5–9. [Order article via Infotrieve]
Neipp M, Zorina T, Domenick MA, Exner BG, Ildstad ST. Effect of FLT3 ligand and granulocyte colony-stimulating factor on expansion and mobilization of facilitating cells and hematopoietic stem cells in mice: kinetics and repopulating potential. Blood. 1998; 92: 3177–3188.
Hope SA, Meredith IT. Cellular adhesion molecules and cardiovascular disease. Part I. Their expression and role in atherogenesis. Intern Med J. 2003; 33: 380–386. [Order article via Infotrieve]
Kronenwett R, Martin S, Haas R. The role of cytokines and adhesion molecules for mobilization of peripheral blood stem cells. Stem Cells. 2000; 18: 320–330.
Weber C. Novel mechanistic concepts for the control of leukocyte transmigration: specialization of integrins, chemokines, and junctional molecules. J Mol Med. 2003; 81: 4–19. [Order article via Infotrieve]
Pfeffer MA, Pfeffer JM, Lamas GA. Development and prevention of congestive heart failure following myocardial infarction. Circulation. 1993; 87: IV120–125. [Order article via Infotrieve]
Molineux G, Migdalska A, Szmitkowski M, Zsebo K, Dexter TM. The effects on hematopoiesis of recombinant stem cell factor (ligand for c-kit) administered in vivo to mice either alone or in combination with granulocyte colony-stimulating factor. Blood. 1991; 78: 961–966.
Hess DA, Levac KD, Karanu FN, Rosu-Myles M, White MJ, Gallacher L, Murdoch B, Keeney M, Ottowski P, Foley R, Chin-Yee I, Bhatia M. Functional analysis of human hematopoietic repopulating cells mobilized with granulocyte colony-stimulating factor alone versus granulocyte colony-stimulating factor in combination with stem cell factor. Blood. 2002; 100: 869–878.
Kucia M, Dawn B, Hunt G, Guo Y, Wysoczynski M, Majka M, Ratajczak J, Rezzoug F, Ildstad ST, Bolli R, Ratajczak MZ. Cells expressing early cardiac markers reside in the bone marrow and are mobilized into the peripheral blood following myocardial infarction. Circ Res. 2004; 95: 1191–1199.
Linke A, Muller P, Nurzynska D, Casarsa C, Torella D, Nascimbene A, Castaldo C, Cascapera S, Bohm M, Quaini F, Urbanek K, Leri A, Hintze TH, Kajstura J, Anversa P. Stem cells in the dog heart are self-renewing, clonogenic, and multipotent and regenerate infarcted myocardium, improving cardiac function. Proc Natl Acad Sci U S A. 2005; 102: 8966–8971.
Na S, Cho H, Chung J, Hahn J, Lee H, Kang H, Cho Y, Koo B, Kim Y, Youn T, Sohn D, Oh B, Lee M, Park Y, Choi Y, Kim H. Comparison of two methods of peripheral blood stem cell transplantation in patients with myocardial infarction: intracoronary infusion of mobilized peripheral blood stem cell vs. mobilization only [Magic cell investigator: myocardial regeneration and angiogenesis in MI with G-CSF mobilization and intra-coronary cell infusion]. J Am Coll Cardiol. 2005; 45 (suppl A): 4A(abstract).
Leone A, Galiuto L, Liuzzo G, Giordano A, Calcagni L, Garramone B, Giannico M, Cirillo F, Biasucci LM, Rebuzzi A, Crea F. Granulocyte-colony stimulating factor in the acute myocardial infarction (the Rigenera Study). J Am Coll Cardiol. 2005; 45 (suppl A): 238A(abstract).
the Institute of Molecular Cardiology (B.D., Y. Guo, A.R., A.B.S., G.H., S.T., J.V., Y. Gu, S.D.P., R.B.) and Institute of Cellular Therapeutics (Y.H., S.T.I.), University of Louisville, Kentucky
Cardiovascular Research Institute (J.K., P.A.), Department of Medicine, New York Medical College, Valhalla.
Abstract
We systematically investigated the comparative efficacy of three different cytokine regimens, administered after a reperfused myocardial infarction, in regenerating cardiac tissue and improving left ventricular (LV) function. Wild-type (WT) mice underwent a 30-minute coronary occlusion followed by reperfusion and received vehicle, granulocyte colony-stimulating factor (G-CSF)+Flt-3 ligand (FL), G-CSF+stem cell factor (SCF), or G-CSF alone starting 4 hours after reperfusion. In separate experiments, chimeric mice generated by reconstitution of radioablated WT mice with bone marrow from enhanced green fluorescent protein (EGFP) transgenic mice underwent identical protocols. Mice were euthanized 5 weeks later. Echocardiographically, LV function was improved in G-CSF+FL– and G-CSF+SCF–treated but not in G-CSF–treated mice, whereas LV end-diastolic dimensions were smaller in all three groups. Morphometrically, cytokine-treated hearts had smaller LV diameter and volume. Numerous EGFP-positive cardiomyocytes, capillaries, and arterioles were noted in the infarcted region in cytokine-treated chimeric mice treated with G-CSF+FL or G-CSF+SCF, but the numbers were much smaller in G-CSF–treated mice. G-CSF+FL therapy mobilized bone marrow–derived cells exhibiting increased expression of surface antigens (CD62L and CD11a) that facilitate homing. We conclude that postinfarct cytokine therapy with G-CSF+FL or G-CSF+SCF limits adverse LV remodeling and improves LV performance by promoting cardiac regeneration and probably also by exerting other beneficial actions unrelated to regeneration, and that G-CSF alone is less effective.
Key Words: myocardial regeneration cytokine stem cell ischemia/reperfusion bone marrow left ventricular function
Introduction
Despite significant advances in the treatment of acute myocardial infarction (MI), no therapies are currently available to restore dead myocardium. Although mobilization of bone marrow cells (BMCs) by cytokines has been suggested to regenerate cardiac tissue after MI and to produce functional improvement, the reports on the effects of cytokine treatment are conflicting.1–6 Furthermore, previous studies of cytokines have used models of permanent coronary ligation1–3,5 that do not reflect the fact that most patients with acute MI undergo coronary reperfusion. In addition, these studies have administered cytokines as a pretreatment1,2 or performed splenectomy1,2,5 or both,1,2 neither of which would be clinically feasible. Although various cytokines have been tested, the relative efficacy of different treatments or combinations of treatments has not been elucidated. Finally, the mechanism whereby cytokines improve left ventricular (LV) function and structure after MI remains obscure. It has been proposed that cytokines mobilize BMCs with subsequent homing to the infarcted tissue and transdifferentiation into cardiac lineage,1–3 but this hypothesis remains unproven. Thus, numerous unresolved issues persist regarding the use of cytokines as a strategy to achieve cardiac repair after MI.
Accordingly, in the present study, we examined the effect of three cytokines (granulocyte colony-stimulating factor [G-CSF], stem cell factor [SCF], and Flt-3 ligand [FL]) on cardiac repair. The specific goals were to determine: (1) whether cytokine therapy is efficacious in the setting of a reperfused MI; (2) whether it is efficacious when started after the MI and in the presence of an intact spleen; (3) which cytokine or cytokine combination, among G-CSF, FL, and SCF, is more effective in improving LV remodeling and function; (4) whether cytokine-mobilized BMCs home to the heart and transdifferentiate into the cardiac lineage; and (5) whether cytokines facilitate BMC homing to the injured myocardium by altering the expression profile of adhesion molecules in mobilized BMCs. A well-characterized murine model of temporary coronary occlusion followed by reperfusion (a setting designed to mimic the most common clinical setting of acute MI in humans) was used.7 A thorough assessment of LV function and anatomy was performed in conjunction with careful measurements of cardiac regeneration by confocal microscopy.
Materials and Methods
A detailed description of all materials and methods is provided in the online supplement, available at http://circres.ahajournals.org.
Generation and Characterization of EGFP Chimeric Mice
EGFP chimeric mice were generated8 by reconstituting radioablated adult male mice (B6,129 strain) with BMCs from syngeneic enhanced green fluorescent protein (EGFP) transgenic mice. On average, 85% of the peripheral blood nucleated cells were positive for EGFP in the recipients.
Experimental Protocol
Adult male wild-type (WT; B6,129 strain, 10 to 18 weeks of age) and EGFP chimeric mice were used. The overall experimental design is summarized in supplemental Figure I.
Myocardial Infarction
Anesthetized mice underwent a 30-minute occlusion of the left anterior descending coronary artery followed by reperfusion. Starting 4 hours after reperfusion, mice received daily subcutaneous injections of vehicle or cytokine as detailed in supplemental Figure I.
Echocardiography
Serial echocardiograms were obtained at baseline (2 days before coronary occlusion) and 2 and 35 days after reperfusion. Systolic and diastolic anatomical and functional parameters were calculated using standard methods.1
Morphometry and Histology
At the end of the follow-up period, the heart was perfusion fixed in 10% neutral buffered formalin,1,9 and the left ventricle was sectioned serially perpendicular to its longitudinal axis, processed, and embedded in paraffin. The infarct volume fraction was calculated by computerized planimetry (Image-Pro Plus; Media Cybernetics) of digital images of three Masson’s trichrome-stained serial LV sections taken at 0.5- to 1.0-mm intervals along the longitudinal axis.10 The midsection was used to measure LV diameter and volume.11,12 All anatomical parameters were corrected according to a uniform sarcomere length.13
Immunohistochemistry was performed in formalin-fixed 4-μm-thick histological sections. Cardiomyocytes were recognized by the presence of -sarcomeric actin and troponin T, endothelial cells by plaletet/endothelial cell adhesion molecule-1 (PECAM-1) and von Willebrand factor, and smooth muscle cells by -smooth muscle actin. Colocalization of cell-specific markers with EGFP was used to identify cells that originated from BMCs. All of the primary and secondary antibodies used in these studies are specified in supplemental Table I. A quantitative analysis of cardiomyocyte regeneration was performed in 3 to 7 EGFP chimeric mouse hearts in each of groups V through VIII.
Assessment of Cardiomyocyte Proliferation
The percentage of cardiomyocyte nuclei positive for 5-bromodeoxyuridine(BrdU) and Ki671 was determined at 14 days after reperfusion in 4 to 6 mice in each of groups IX through XII. In addition, the percentage of cardiomyocyte nuclei positive for Ki67 at 35 days after MI was determined in 6 mice in each of groups I through IV.
Flow Cytometry
After vehicle or cytokine treatment, peripheral blood was harvested from mice in groups XIII through XVI at serial time points, stained with antibodies, and lin–/Sca-1+/c-kit+ cells were counted from the lymphoid gate with a multiparameter, live sterile cell sorter.14,15 The flow cytometric procedure is summarized in supplemental Figure II. The expression of adhesion molecules on lin–/Sca-1+/c-kit+ cells from untreated murine bone marrow and peripheral blood of mice (groups XVII through XX) treated with identical dosages of vehicle or cytokines as groups I through IV, respectively, was also assessed by flow cytometry.
Statistical Analysis
Data are reported as mean±SEM. Morphometric and histologic data were analyzed with one-way ANOVA. Serial echocardiographic parameters were measured with a two-way (time and group) ANOVA followed by Student t tests with the Bonferroni correction.
Results
Exclusions
A total of 188 mice (162 WT and 26 EGFP chimeric) were used. One hundred fifty-six mice were assigned to the MI studies (groups I through XII) and 32 mice to the cellular mobilization (groups XIII through XX) studies. Twenty-nine mice died in the perioperative period. Nine mice were excluded from the study because of failure of the coronary occluder, leaving a total of 14, 19, 20, and 18 WT mice in groups I through IV, respectively, and 3, 7, 6, and 5 EGFP chimeric mice in groups V through VIII, respectively. No mouse was excluded from the cellular mobilization studies (groups XIII through XX; n=4 per group).
Myocardial Infarct Size
The impact of cytokine therapy on LV structural and functional parameters was assessed in groups I through IV. The infarct volume fraction was similar in the four groups (supplemental Figure III), indicating that the extent of myocardial cell death produced by the coronary occlusion was comparable. The infarct volume fraction measures the volume of scarred tissue expressed as a fraction of the entire left ventricle.10–12 Because the small islands of regenerated cardiomyocytes scattered throughout the scar area were included in the "infarct" region, in cytokine-treated hearts, the calculation of the infarct volume fraction was not affected by myocardial regeneration.
G-CSF+FL and G-CSF+SCF Therapy Improves LV Systolic Function
Before coronary occlusion (baseline), all parameters of LV function, measured by echocardiography, were similar in groups I through IV (Figure 1). At 48 hours after reperfusion, the degree of LV systolic functional impairment did not differ among groups (Figure 1), indicating that the injury sustained during ischemia/reperfusion was comparable. In vehicle-treated mice, there was further functional deterioration between 48 hours and 35 days after reperfusion (Figure 1). In contrast, in G-CSF+FL-treated (group II) and G-CSF+SCF-treated (group III) mice, LV systolic function remained unchanged or improved at 35 days compared with 48 hours (Figure 1). Mice in groups II and III exhibited significantly higher LV ejection fraction, fractional area change, and fractional shortening and lower LV end-systolic diameter compared with vehicle-treated mice (group I) (Figure 1). Compared with group I, myocardial function in the infarct region was greater in groups II and III (systolic thickening fraction in the infarct wall, 18.6±5.9%, 26.5±2.8%, 24.8±5.4%, and 20.9±6.6% in groups I, II, III, and IV, respectively), but the differences were not statistically significant. Compared with group I, myocardial function was also enhanced in the nonischemic LV region in groups II and III (systolic thickening fraction, 33.2±4.3%, 43.9±5.3%, and 54.4±3.2% [P<0.05 versus group I] and 35.1±4.1% in groups I, II, III, and IV, respectively). In mice treated with G-CSF alone (group IV), there was only a marginal improvement in parameters of LV systolic function (Figure 1E through 1G).
Cytokine Therapy Halts LV Remodeling
Both morphometric and echocardiographic analyses demonstrated that administration of cytokines attenuated adverse LV remodeling 5 weeks after infarction. Morphometrically, LV chamber diameter and LV chamber volume were lower in groups II, III, and IV compared with group I (Figure 2A and 2B; supplemental Figure IV). Echocardiographically, LV anatomical parameters were similar in groups I through IV, both before coronary occlusion (baseline) and 48 hours after reperfusion (Figure 2C and 2D). In vehicle-treated mice (group I), there was significant LV dilatation between 48 hours and 35 days after reperfusion (Figure 2), indicative of adverse LV remodeling. Compared with group I, both echocardiographic (Figure 2C and 2D) and morphometric (supplemental Figure IV) analyses showed smaller LV diameter and LV end-diastolic volume in cytokine-treated mice (groups II through IV). However, the reduction in LV dilatation in group IV was not statistically significant by echocardiography (Figure 2C and 2D), and the reduction in LV volume in group II was not statistically significant by morphometry (supplemental Figure IV). By morphometry, the infarct wall thickness improved in cytokine-treated (groups III and IV) mice (Figure 2E). Interestingly, mice that received G-CSF alone (group IV) also demonstrated a comparable improvement in morphometric parameters of LV remodeling (supplemental Figure IV) despite borderline functional improvement (Figure 1).
Cytokine Therapy Promotes Regeneration of Cardiomyocytes
In EGFP chimeric mice, the effects of cytokines on LV function and anatomy after infarction were directionally concordant with those in WT mice (supplemental Figures V and VI, respectively), despite the fact that LV size was smaller in the chimeric mice (supplemental Figure VI) compared with WT mice (Figure 2). Histologic quantitation of myocardial regeneration was performed in hearts (n=3 to 7 per group) from EGFP chimeric mice (groups V through VIII). Cardiomyocytes derived from BMCs were identified by concomitant positivity for -sarcomeric actin and EGFP (Figure 3). The extent of regeneration was assessed by measuring the area occupied by EGFP+ myocytes rather than their number; the former parameter more accurately reflects the functional impact of regeneration. In the infarcted region, EGFP+ regenerated cardiomyocytes occupied only 0.9±0.1% of the myocardial area in group I; in contrast, this fraction of reconstituted myocardium was much higher in groups II, III, and IV (14.9±1.5%, 10.8±1.9%, and 4.7±1.0% of the myocardial area, respectively). The area occupied by regenerated cardiomyocytes was similar in the infarct and in the border zone. In contrast, in the noninfarcted region, the area occupied by EGFP+ myocytes was uniformly low in all four groups (0.2±0.1%, 0.6±0.3%, 0.4±0.1%, and 0.4±0.1% of myocardial area in groups I, II, III, and IV, respectively; Figure 4). Newly formed myocytes expressed connexin 43 (Figure 5), indicating functional integration with adjacent myocytes. EGFP+ cells that were negative for cardiomyocyte markers constituted 1.2±0.6%, 3.9±0.6%, 3.5±0.6%, and 2.9±0.8% of the infarcted myocardial area and 0.7±0.1%, 1.9±0.4%, 0.6±0.1%, and 1.2±0.3% of the viable myocardial area in groups I, II, III, and IV, respectively (Figure 4). Thus, a fraction of the BMCs that homed to the infarcted myocardium differentiated into noncardiomyocyte cell types. In the ischemic zone, most EGFP+ cells were cardiomyocytes; in the nonischemic zone, the reverse was observed.
Cardiomyocyte Proliferation
The effect of cytokine administration on cardiomyocyte proliferation at 14 days after MI was assessed in groups IX through XII by examining BrdU incorporation (Figure 6A and 6C) in cardiomyocyte nuclei. In addition, 6 hearts from each of groups I through IV were examined for Ki67 expression at 35 days after MI (Figure 6B and 6D). Treatment with G-CSF+FL, G-CSF+SCF, or G-CSF resulted in a significant increase in BrdU+ and Ki67+ cardiomyocytes in the infarcted region, indicating increased numbers of cycling myocytes (Figure 6). The increase observed with G-CSF alone was significantly less than that seen with G-CSF+FL and G-CSF+SCF. In contrast, the number of cycling myocytes positive for Ki67 in the nonischemic zone at 35 days after MI was comparable in all groups (data not shown).
Regeneration of the Vasculature
Colocalization of EGFP positivity with -smooth muscle actin positivity in arteriolar smooth muscle cells (Figure 7A) and PECAM-1 positivity in endothelial cells (Figure 7B) were noted predominantly in the infarct area and, to a lesser extent, in the viable myocardium. Thus, EGFP+ BMCs differentiated not only into cardiomyocytes but also into vascular structures necessary to supply blood to the newly formed myocardium in the infarcted region.
Mobilization of Lin–/Sca-1+/C-kit+ BMCs
The efficacy of different cytokine regimens in mobilizing lin–/Sca-1+/c-kit+ BMCs was examined by flow cytometry in groups XIII through XVI. Compared with groups XIII, XV, and XVI, G-CSF+FL (group XIV) therapy resulted in mobilization of a significantly higher number of lin–/Sca-1+/c-kit+ BMCs (0.1±0.1, 28.8±3.9, 5.8±1.3, and 4.0±0.7 cells/μL at 5 days in groups XIII, XIV, XV, and XVI, respectively). The increase in lin–/Sca-1+/c-kit+ BMC levels in group XIV was apparent at 3 days of therapy and became progressively more pronounced at 5 and 10 days. G-CSF+SCF (group XV) and G-CSF alone (group XVI) resulted in minimal changes in peripheral blood levels of lin–/Sca-1+/c-kit+ cells.
Cytokine Therapy Enhances Expression of Surface Antigens That Favor Homing
In groups XVII through XX, peripheral blood cells were analyzed by flow cytometry for expression of adhesion molecules on mobilized lin–/Sca-1+/c-kit+ BMCs. As shown in supplemental Figure 7, all three treatments (G-CSF+FL, G-CSF+SCF, and G-CSF alone) resulted in increased expression of CD62P (P-selectin), an adhesion molecule that favors attachment and extravasation of BMCs,16 and decreased expression of CD49d, an adhesion molecule that plays a critical role in retaining cells within the bone marrow.17 Both of these changes would be expected to promote release of BMCs from bone marrow and their migration into the cardiac interstitium. However, there were significant differences in adhesion molecule profile between G-CSF+FL–treated mice on the one hand and G-CSF+SCF– and G-CSF–treated mice on the other hand. Specifically, G-CSF+SCF and G-CSF treatment reduced whereas G-CSF+FL treatment increased the expression of CD62L (L-selectin) and CD11a (Figure 8), two adhesion molecules that favor extravasation and homing of BMCs.16–18 Furthermore, the reduction in expression of CD49e, an adhesion molecule that promotes BMC attachment,17 was comparatively less with G-CSF+FL and G-CSF+SCF treatment vis-à-vis G-CSF alone (supplemental Figure VII). In conjunction with the quantitative analysis of the number of circulating lin–/Sca-1+/c-kit+ cells reported above, these results demonstrate significant differences among different cytokine regimens with respect to the modulation of the adhesion molecules responsible for homing.
Discussion
Most basic and clinical studies of cardiac regeneration have focused on cell-based therapies,4 and relatively less is known regarding the role of cytokines in this process. The present investigation provides a number of important new findings. Our results demonstrate that: (1) cytokine therapy with a combination of G-CSF+FL or G-CSF+SCF started after acute MI, halts or attenuates adverse LV remodeling, and improves LV systolic function; (2) cytokine therapy results in mobilization of bone marrow stem cells (lin–/Sca-1+/c-kit+), which home to the injured myocardium and transdifferentiate into the cardiac lineage (cardiomyocytes, vascular smooth muscle cells, endothelial cells); (3) the combination of G-CSF and FL is superior to G-CSF+SCF or G-CSF alone with respect to its effects on LV remodeling, LV function, and mobilization of lin– /Sca-1+/c-kit+ BMCs into the peripheral blood; (4) all three cytokine regimens upregulated the expression of CD62P and downregulated the expression of CD49d on BMCs (CD62P [P-selectin] promotes the extravasation of BMCs, whereas CD49d is important for the retention of these cells in the marrow16,17); and (5) in contrast to G-CSF+SCF and G-CSF alone, G-CSF+FL treatment increases the expression of adhesion molecules (CD62L and CD11a) that favor homing of BMCs.16–18 CD62L (L-selectin) and the 2 integrin leukocyte function antigen-1 (CD11a) initiate and facilitate the attachment of peripheral blood cells to the endothelium, thereby enhancing the extravasation of these cells.17,18 Thus, the present investigation not only establishes the usefulness of cytokine therapy for therapeutic cardiac regeneration but also provides mechanistic insights into this phenomenon and identifies the superiority of a specific cytokine combination (G-CSF+FL). These results have important implications for the role of cytokine-based strategies in effecting cardiac repair after acute MI.
Previous studies have shown that cytokine therapies initiated before1,2 or after3 MI result in improved LV remodeling and function. However, to our knowledge, this is the first study to demonstrate that cytokine administration results in transdifferentiation of mobilized BMCs into the cardiac lineage and in expression of adhesion molecules that favor homing of BMCs to the injured myocardium. Furthermore, this is the first investigation to systematically compare the efficacy of G-CSF, FL, and SCF and their combination on postinfarct LV remodeling and function, and to identify the combination of G-CSF and FL as the most effective treatment.
A number of methodological considerations are in order. In contrast to previous studies, which examined models of permanent coronary occlusion,1 we elected to use a transient (30-minute) coronary occlusion followed by reperfusion. This model has greater clinical relevance because most patients with acute MI undergo either spontaneous or therapeutic coronary recanalization. We chose to initiate cytokine treatment after the MI because pretreatment (as used in previous studies1,2) would not be clinically feasible. For similar reasons, we elected not to perform a splenectomy (which was performed in previous investigations1,2). The doses of cytokines used in this study were selected because they are clinically relevant. The large group sizes (14 to 20 mice per group in groups I through IV) were deemed to be important to achieve robust conclusions. In an effort to perform a rigorous assessment of global as well as regional LV function, we used multiple echocardiographic parameters, including fractional shortening (parasternal short-axis M-mode images), fractional area change (parasternal short-axis 2-D images), ejection fraction, and systolic wall thickening. Treatment with G-CSF+FL or G-CSF+SCF consistently improved all of these parameters, providing solid evidence for the functional benefit of these therapies. This concept is further corroborated by the fact that two independent techniques (echocardiography and morphometry, performed by blinded investigators who were unaware of the data obtained with the other technique) demonstrated that G-CSF+FL and G-CSF+SCF were effective in mitigating the progressive LV chamber dilatation that is characteristic of adverse LV remodeling and ischemic cardiomyopathy.19 The data obtained by these two methods were generally concordant. However, in Figure 2E, we present the morphometric infarct wall thickness data because the measurement of the average integrated wall thickness by morphometry may be inherently more accurate.
We chose to examine G-CSF, SCF, and FL because these cytokines are used clinically (G-CSF) or have been well characterized with respect to BMC mobilization.15,20 Comparison of different cytokines or combinations thereof is important because these proteins are known to preferentially mobilize different subsets of BMCs.15,21 FL was not studied alone because, unlike G-CSF, it does not expand lin–/Sca-1+/c-kit+ cells.15 SCF was not studied as single therapy because it has low efficacy in mobilizing BMCs when used alone.20 Our results indicate that the combination of G-CSF and FL produced the most robust results with respect to myocyte regeneration, LV remodeling, LV function, and mobilization of lin–/Sca-1+/c-kit+ cells. G-CSF+SCF also resulted in significant beneficial effects, albeit to a lesser extent. The most plausible explanation for the superiority of G-CSF+FL is that this combination resulted in mobilization of much higher numbers of lin–/Sca-1+/c-kit+ cells into the peripheral blood, which would be expected to enhance the homing and engraftment of BMCs in the damaged myocardium. In addition to these quantitative differences, our flow cytometric analysis reveals qualitative differences; that is, the lin–/Sca-1+/c-kit+ cells mobilized by G-CSF+FL expressed higher levels of CD62L and CD11a vis-à-vis the cells mobilized by G-CSF+SCF (Figure 8). The expression of these adhesion molecules may facilitate the egress of primitive BMCs from the circulation into injured tissues that express chemoattractants.22
Having established the functional and structural benefits of cytokine therapy, we used EGFP chimeric mice to investigate the mechanism for these salubrious actions. Our results demonstrate that after administration of G-CSF+FL or G-CSF+SCF, mobilized BMCs home to the infarcted myocardium and differentiate into cardiomyocytes, smooth muscle cells, and endothelial cells. We found that cardiomyocyte regeneration was considerably greater with G-CSF+FL or G-CSF+SCF versus G-CSF alone (Figure 4). These findings were independently validated by the measurement of cycling (ie, BrdU+ and Ki67+) cardiomyocytes in non-EGFP chimeric mice (Figure 6).
The mechanism(s) whereby cytokines produce beneficial effects on LV dimensions and function after MI is likely complex. Although the association of greater mobilization of lin–/Sca-1+/c-kit+ cells with greater improvement in LV remodeling and function after G-CSF+FL treatment suggests that these cells may play an important role in cardiac regeneration, our results do not prove that the beneficial effects of cytokines were attributable to this subset of BMCs. The BMCs responsible for cardiac regeneration may be cell types other than the lin-/Sca-1+/c-kit+ cells (eg, multipotent progenitors, endothelial progenitors, etc.). It is also unknown whether, within the lin-/Sca-1+/c-kit+ subset of BMCs, the transdifferentiation was attributable to hematopoietic or nonhematopoietic cells. We recently identified a CXCR4+/Sca-1+/lin–/CD45– nonhematopoietic subfraction of BMCs that appear to be committed to differentiation into cardiac lineages and are mobilized after MI in mice as well as in humans.22 It is possible that these cells may account for the salubrious actions of G-CSF+FL and G-CSF+SCF noted in this study. Finally, mobilization of resident cardiac stem cells by cytokines could be an additional or alternative mechanism.23 In support of this hypothesis, we noted small cardiomyocytes in the infarct area of cytokine-treated hearts that were not positive for EGFP (Figure 3F, arrowheads), suggesting that cytokine treatment may have induced migration, differentiation, and possibly proliferation of resident cardiac progenitor cells, which may have contributed to cardiac repair.
Harada et al6 recently concluded that administration of G-CSF ameliorates LV remodeling and dysfunction by inhibiting apoptosis, not by promoting cardiac regeneration. Our results differ because they document transdifferentiation of BMCs into cardiac lineage (Figures 3, 4, and 7; predominantly in the infarct area). Nevertheless, we also noted functional improvement in the noninfarcted myocardium, where the extent of myocyte regeneration was minimal (Figure 4). This enhancement in contractility in the noninfarcted region is likely attributable to nonregenerative actions of cytokines (eg, inhibition of apoptosis, favorable changes in the interstitium, etc.), suggesting that multiple mechanisms contributed to the overall improvement in LV function after G-SCF+FL and G-SCG+SCF therapy. Although G-CSF alone induced little cardiomyocyte regeneration, it had favorable effects on LV remodeling (Figure 2C and 2D; supplemental Figure IV); however, G-CSF alone did not improve LV function significantly (Figure 1). As shown in Figure 1, measurements of LV systolic function were greater in group IV than in vehicle-treated mice, but the differences were not statistically significant. Despite the large sample sizes (n=15 to 19 per group), this could be attributable to a type II error. Alternatively, it is possible that other nonregenerative beneficial effects of G-CSF on the myocardium resulted in an improvement in LV anatomy which, however, did not result in improved function because of the less efficient regeneration of cardiomyocytes with G-CSF therapy.
In conclusion, we demonstrated that a clinically relevant protocol for cytokine therapy exerts profound beneficial effects on postinfarction cardiomyopathy. Specifically, administration of G-CSF+FL or G-CSF+SCF, started after a reperfused acute MI, results in regeneration of cardiomyocytes and coronary vessels by mobilized BMCs and in improvement in LV remodeling and function via myocyte regeneration and probably via other beneficial actions of cytokines. We also demonstrated that the combination of G-CSF+FL is superior to G-CSF+SCF or G-CSF alone. These results have significant therapeutic implications. Recent studies with G-CSF have concluded that this cytokine improves LV function and structure in patients with acute MI.24,25 Our findings suggest that the combination of G-CSF and FL should have even greater efficacy in the clinical arena, providing a rationale for testing G-CSF+FL in larger species and in patients.
Acknowledgments
This study was supported in part by National Institutes of Health grants HL-72410, HL-55757, HL-68088, HL-70897, HL-76794, HL-78825, HL-63442, and DK-069766, national American Heart Association (AHA) scientist development grant 0130146N, and AHA fellowship grants 0265087B and 0325372B.
Footnotes
This manuscript was sent to Stephen F. Vatner, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Original received October 21, 2005; revision received February 27, 2006; accepted March 9, 2006.
References
Orlic D, Kajstura J, Chimenti S, Limana F, Jakoniuk I, Quaini F, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Mobilized bone marrow cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci U S A. 2001; 98: 10344–10349.
Fukuhara S, Tomita S, Nakatani T, Ohtsu Y, Ishida M, Yutani C, Kitamura S. G-CSF promotes bone marrow cells to migrate into infarcted mice heart, and differentiate into cardiomyocytes. Cell Transplant. 2004; 13: 741–748. [Order article via Infotrieve]
Kawada H, Fujita J, Kinjo K, Matsuzaki Y, Tsuma M, Miyatake H, Muguruma Y, Tsuboi K, Itabashi Y, Ikeda Y, Ogawa S, Okano H, Hotta T, Ando K, Fukuda K. Nonhematopoietic mesenchymal stem cells can be mobilized and differentiate into cardiomyocytes after myocardial infarction. Blood. 2004; 104: 3581–3587.
Wollert KC, Drexler H. Clinical applications of stem cells for the heart. Circ Res. 2005; 96: 151–163.
Deten A, Volz HC, Clamors S, Leiblein S, Briest W, Marx G, Zimmer HG. Hematopoietic stem cells do not repair the infarcted mouse heart. Cardiovasc Res. 2005; 65: 52–63. [Order article via Infotrieve]
Harada M, Qin Y, Takano H, Minamino T, Zou Y, Toko H, Ohtsuka M, Matsuura K, Sano M, Nishi JI, Iwanaga K, Akazawa H, Kunieda T, Zhu W, Hasegawa H, Kunisada K, Nagai T, Nakaya H, Yamauchi-Takihara K, Komuro I. G-CSF prevents cardiac remodeling after myocardial infarction by activating the Jak-Stat pathway in cardiomyocytes. Nat Med. 2005; 11: 305–311. [Order article via Infotrieve]
Guo Y, Wu WJ, Qiu Y, Tang XL, Yang Z, Bolli R. Demonstration of an early and a late phase of ischemic preconditioning in mice. Am J Physiol. 1998; 275: H1375–H1387.
Colson YL, Wren SM, Schuchert MJ, Patrene KD, Johnson PC, Boggs SS, Ildstad ST. A nonlethal conditioning approach to achieve durable multilineage mixed chimerism and tolerance across major, minor, and hematopoietic histocompatibility barriers. J Immunol. 1995; 155: 4179–4188.
Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Bone marrow cells regenerate infarcted myocardium. Nature. 2001; 410: 701–705. [Order article via Infotrieve]
Anversa P, Beghi C, Kikkawa Y, Olivetti G. Myocardial infarction in rats. Infarct size, myocyte hypertrophy, and capillary growth. Circ Res. 1986; 58: 26–37.
Li Q, Li B, Wang X, Leri A, Jana KP, Liu Y, Kajstura J, Baserga R, Anversa P. Overexpression of insulin-like growth factor-1 in mice protects from myocyte death after infarction, attenuating ventricular dilation, wall stress, and cardiac hypertrophy. J Clin Invest. 1997; 100: 1991–1999.
Li B, Setoguchi M, Wang X, Andreoli AM, Leri A, Malhotra A, Kajstura J, Anversa P. Insulin-like growth factor-1 attenuates the detrimental impact of nonocclusive coronary artery constriction on the heart. Circ Res. 1999; 84: 1007–1019.
Anversa P, Olivetti G. The cardiovascular system. In: Page E, Fozzard HA, Solaro RJ, eds. The Heart. 1st ed. New York, NY: Oxford University Press; 2002: 75–144.
Orfao A, Ruiz-Arguelles A. General concepts about cell sorting techniques. Clin Biochem. 1996; 29: 5–9. [Order article via Infotrieve]
Neipp M, Zorina T, Domenick MA, Exner BG, Ildstad ST. Effect of FLT3 ligand and granulocyte colony-stimulating factor on expansion and mobilization of facilitating cells and hematopoietic stem cells in mice: kinetics and repopulating potential. Blood. 1998; 92: 3177–3188.
Hope SA, Meredith IT. Cellular adhesion molecules and cardiovascular disease. Part I. Their expression and role in atherogenesis. Intern Med J. 2003; 33: 380–386. [Order article via Infotrieve]
Kronenwett R, Martin S, Haas R. The role of cytokines and adhesion molecules for mobilization of peripheral blood stem cells. Stem Cells. 2000; 18: 320–330.
Weber C. Novel mechanistic concepts for the control of leukocyte transmigration: specialization of integrins, chemokines, and junctional molecules. J Mol Med. 2003; 81: 4–19. [Order article via Infotrieve]
Pfeffer MA, Pfeffer JM, Lamas GA. Development and prevention of congestive heart failure following myocardial infarction. Circulation. 1993; 87: IV120–125. [Order article via Infotrieve]
Molineux G, Migdalska A, Szmitkowski M, Zsebo K, Dexter TM. The effects on hematopoiesis of recombinant stem cell factor (ligand for c-kit) administered in vivo to mice either alone or in combination with granulocyte colony-stimulating factor. Blood. 1991; 78: 961–966.
Hess DA, Levac KD, Karanu FN, Rosu-Myles M, White MJ, Gallacher L, Murdoch B, Keeney M, Ottowski P, Foley R, Chin-Yee I, Bhatia M. Functional analysis of human hematopoietic repopulating cells mobilized with granulocyte colony-stimulating factor alone versus granulocyte colony-stimulating factor in combination with stem cell factor. Blood. 2002; 100: 869–878.
Kucia M, Dawn B, Hunt G, Guo Y, Wysoczynski M, Majka M, Ratajczak J, Rezzoug F, Ildstad ST, Bolli R, Ratajczak MZ. Cells expressing early cardiac markers reside in the bone marrow and are mobilized into the peripheral blood following myocardial infarction. Circ Res. 2004; 95: 1191–1199.
Linke A, Muller P, Nurzynska D, Casarsa C, Torella D, Nascimbene A, Castaldo C, Cascapera S, Bohm M, Quaini F, Urbanek K, Leri A, Hintze TH, Kajstura J, Anversa P. Stem cells in the dog heart are self-renewing, clonogenic, and multipotent and regenerate infarcted myocardium, improving cardiac function. Proc Natl Acad Sci U S A. 2005; 102: 8966–8971.
Na S, Cho H, Chung J, Hahn J, Lee H, Kang H, Cho Y, Koo B, Kim Y, Youn T, Sohn D, Oh B, Lee M, Park Y, Choi Y, Kim H. Comparison of two methods of peripheral blood stem cell transplantation in patients with myocardial infarction: intracoronary infusion of mobilized peripheral blood stem cell vs. mobilization only [Magic cell investigator: myocardial regeneration and angiogenesis in MI with G-CSF mobilization and intra-coronary cell infusion]. J Am Coll Cardiol. 2005; 45 (suppl A): 4A(abstract).
Leone A, Galiuto L, Liuzzo G, Giordano A, Calcagni L, Garramone B, Giannico M, Cirillo F, Biasucci LM, Rebuzzi A, Crea F. Granulocyte-colony stimulating factor in the acute myocardial infarction (the Rigenera Study). J Am Coll Cardiol. 2005; 45 (suppl A): 238A(abstract).
您现在查看是摘要介绍页，详见ORG附件。