Mobilized bone marrow cells repair the infarcted heart, improving function and survival

Abstract

Attempts to repair myocardial infarcts by transplanting cardiomyocytes or skeletal myoblasts have failed to reconstitute healthy myocardium and coronary vessels integrated structurally and functionally with the remaining viable portion of the ventricular wall. The recently discovered growth and transdifferentiation potential of primitive bone marrow cells (BMC) prompted us, in an earlier study, to inject in the border zone of acute infarcts Lin(-) c-kit(POS) BMC from syngeneic animals. These BMC differentiated into myocytes and vascular structures, ameliorating the function of the infarcted heart. Two critical determinants seem to be required for the transdifferentiation of primitive BMC: tissue damage and a high level of pluripotent cells. On this basis, we hypothesized here that BMC, mobilized by stem cell factor and granulocyte-colony stimulating factor, would home to the infarcted region, replicate, differentiate, and ultimately promote myocardial repair. We report that, in the presence of an acute myocardial infarct, cytokine-mediated translocation of BMC resulted in a significant degree of tissue regeneration 27 days later. Cytokine-induced cardiac repair decreased mortality by 68%, infarct size by 40%, cavitary dilation by 26%, and diastolic stress by 70%. Ejection fraction progressively increased and hemodynamics significantly improved as a consequence of the formation of 15 x 10(6) new myocytes with arterioles and capillaries connected with the circulation of the unaffected ventricle. In conclusion, mobilization of primitive BMC by cytokines might offer a noninvasive therapeutic strategy for the regeneration of the myocardium lost as a result of ischemic heart disease and, perhaps, other forms of cardiac pathology.

Figures

Figure 1

Mortality and myocardial regeneration. (A) Cytokine-treated infarcted mice, n = 15; untreated infarcted mice, n = 52; log-rank test, P < 0.0001. (B) Large infarct (MI) in a cytokine-treated mouse; forming myocardium (arrowheads) at higher magnification (adjacent panel). (C) MI in a nontreated mouse. Healing comprises the entire infarct (arrowheads). Scarring at higher magnification (adjacent panel). Red = cardiac myosin; yellow-green = propidium iodide (PI) labeling of nuclei; blue-magenta = collagen types I and III. (B and C, ×20; Insets, ×80.)

Figure 2

Myocardial regeneration. (A) Remaining viable (Re), lost (Lo), and newly formed (Fo) myocardium in LVFW at 27 days in MI and MI-C; SO, myocardium without infarct. (B) Cellular hypertrophy in spared myocardium. (C) M, EC, and SMC labeled by BrdUrd and Ki67; n = 11. * and **, P < 0.05 vs. M and EC. (D and E) Volume, number (n = 11), and class distribution (bucket size, 100 μm3; n = 4,400) of M within the formed myocardium. (F–H) Arterioles with TER-119-labeled erythrocyte membrane (green fluorescence); blue fluorescence = propidium iodide (PI) staining of nuclei; red fluorescence = α-smooth muscle actin in SMC. F, ×800; G and H, ×1,200.

Figure 3

Markers of differentiating cardiac cells. (A–F) Labeling of M by nestin (A, yellow), desmin (B, red), and connexin 43 (C, green); red fluorescence = cardiac myosin (A and C). (D and E) Yellow-green fluorescence reflects labeling of EC by flk-1 (arrows, D) and VE-cadherin (arrows, E); red fluorescence = factor VIII in EC (D and E). (F) Green fluorescence labeling of SMC cytoplasms by flk-1; endothelial lining is also labeled by flk-1; red fluorescence = α-smooth muscle actin; blue fluorescence = propidium iodide (PI) labeling of nuclei. (A and E, ×1,200; B and F ×800; C, ×1,400; D, ×1,800.)

Figure 4

MI, cardiac anatomy, and function. (A–C) LV dimensions at time of death, 27 days after surgery; SO (n = 9), nontreated infarcted (MI, n = 9), and cytokine-treated infarcted (MI-C, n = 10). (D) EF by echocardiography (SO, n = 9; MI, n = 9; and MI-C, n = 9). (E–M) M-mode echocardiograms of SO (E–G), MI (H–J), and MI-C (K–M); newly formed contracting myocardium (arrows). Detailed echocardiograms are shown in Fig. 8. (N) Wall stress, SO (n = 9); MI (n = 8); and MI-C (n = 9). Results are mean ± SD. * and **, P < 0.05 vs. SO and MI, respectively.