Cell Transplantation Improves Ventricular Function After a Myocardial Infarction
A Preclinical Study of Human Unrestricted Somatic Stem Cells in a Porcine Model
Background— Cell transplantation offers the promise in the restoration of ventricular function after an extensive myocardial infarction, but the optimal cell type remains controversial. Human unrestricted somatic stem cells (USSCs) isolated from umbilical cord blood have great potential to differentiate into myogenic cells and induce angiogenesis. The present study evaluated the effect of USSCs on myocardial regeneration and improvement of heart function after myocardial infarction in a porcine model.
Method and Results— The distal left anterior descending artery of Yorkshire pigs (30 to 35 kg) was occluded by endovascular implantation of a coil. Four weeks after infarction, single-photon emission computed tomography technetium 99m sestamibi scans (MIBI) and echocardiography were performed. USSCs (100×106) or culture media were then directly injected into the infarcted region (n=8 per group). Pigs were immunosuppressed by daily administration of cyclosporin A. At 4 weeks after transplantation, MIBI and echocardiography were repeated and heart function was also assessed with a pressure-volume catheter. The infarcted myocardium and implanted cells were studied histologically. MIBI showed improved regional perfusion (P<0.05) and wall motion (P<0.05) of the infarct region in the transplant group compared with the control. Ejection fraction evaluated by both MIBI and echocardiography decreased in the control group but increased in the transplant group (P<0.01). Scar thickness of the transplant group was higher than the control. The grafted cells were detected 4 weeks after transplantation by both immunohistochemistry and in situ hybridization.
Conclusion— Engrafted USSCs were detected in the infarct region 4 weeks after cell transplantation, and the implanted cells improved regional and global function of the porcine heart after a myocardial infarction. This study suggests that the USSC implantation will be efficacious for cellular cardiomyoplasty.
After a myocardial infarction, some heart cells in the defect area are lost and others hibernate because they are underperfused. Implantation of healthy cells into the damaged myocardium offers the promise to replace the lost cells (myogenesis) and perfuse the hibernating cells (angiogenesis). Transplantation of cells, including cardiomyocytes,1–5 skeletal myoblasts,6 bone marrow cells,7–9 and smooth muscle cells,10 has been demonstrated to be of potential therapeutic value for the repair of damaged myocardium in animal models. The beneficial effect of cell transplantation on cardiac functions in preclinical studies has led to several phase I and II clinical trials.11–13
Although several cell types implanted into the infarct region have improved ventricular function after a myocardial infarction,4,6,7 the cell source might become an issue for a broad clinical application of cardiac cell therapy because expansion of autologous cells could be problematic. Recent evidence suggests that age and disease states affect the collection of sufficient healthy autologous bone marrow-derived cells for transplantation, which will decrease the ability of autologous cells to improve ventricular function after a myocardial infarction.14 Heeschen and colleagues15 also reported that bone marrow mononuclear cells from patients with ischemic heart disease have a significantly lower regeneration potential than cells from healthy controls. The availability of pluripotent cells, which can be implanted without manipulation or expansion in an allogenic setting, would permit rapid cardiac restoration after a myocardial infarction.
Mesenchymal stem cells (MSCs) isolated from adult bone marrow have shown a great potential for cell therapy because these cells possess pluripotent capabilities,16 proliferate rapidly, induce angiogenesis, and differentiate into myogenic cells.17–19 Recently, a pluripotent stem cell population with high proliferative potential, unrestricted somatic stem cell (USSCs), was isolated from human umbilical cord blood as described by Koegler and colleagues.20 USSCs have been suggested to be an earlier cell type than multipotent MSCs, possibly representing the precursor cells for MSCs as shown by comparing the differentiation potentials of USSCs20 and MSCs.21 In vitro and in vivo studies showed that the USSCs have the potential to differentiate into osteoblasts, chondrocytes, adipocytes, neurons, and cardiomyocytes.20 In a sheep model, the application of USSCs did not induce macroscopic or microscopic tumors in a long-term study after transplantation.20 Therefore, USSCs could be highly promising precursor cells for cardiac implantation after a myocardial infarction.
To explore the clinical application of human stem cells, we transplanted USSCs into porcine hearts after a myocardial infarction to investigate the effects of the implanted USSCs on regional perfusion, wall motion, and global heart function. We found that the transplanted USSCs survived in the infarct region. Regional myocardial perfusion and wall motion were improved and global heart function was preserved after USSC transplantation.
Adult female domestic Yorkshire swine (30 to 35 kg, Charles River, Quebec, Canada) were used in the study. All procedures were approved by the Animal Care Committee and Radiation Safety Committee of the Toronto General Hospital and were performed according to the Guide to the Care and Use of Experimental Animals of the Canadian Council on Animal Care and the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health.
Creation of the Myocardial Infarction
All surgical procedures were performed when pigs were under general anesthesia and electrocardiographic monitoring as we described previously.4,22 Animals were premedicated with ketamine (20 to 25 mg/kg, intramuscularly) and 4% to 5% isoflurane and oxygen. After intubation, anesthesia was maintained with 1.5% to 2.5% isoflurane and oxygen at 2 to 3 L/min.
Under the guide of fluoroscopy C-arm, a coronary guiding catheter was introduced into distal left anterior descending coronary artery (LAD) through the right carotid artery. A superselective infusion catheter was inserted into the LAD or its diagonal branch. A myocardial infarction was created by occlusion of the LAD or the branches using an intraluminal coil (Boston Scientific Corporation) and a piece of gelatin sponge. An ST-T change of the ECG and a follow-up coronary arteriography were recorded to confirm the occlusion of the vessel. The carotid artery was then repaired and the animals were recovered. The animals were treated with antibiotics (cefazolin, 1g), analgesics (nurmorphan 1 to 2 mg, buprenorphrine, 0.3 mg). A nitroglycerin ointment 2% (Nitrol, Paladin Labs Inc) was applied on the postauricular skin postoperatively. Brevibloc 0.1 mg · kg−1 · min−1 was used intravenously during the recovery period.
Human Unrestricted Somatic Stem Cell Preparation
USSCs were isolated, cultured, and prepared in Kourion Therapeutics AG as previously described.20 The USSCs were characterized by cell surface markers using flow cytometry with antibodies against CD13, CD14, CD29, CD31, CD33, CD 34, CD44, CD45, CD49e, CD56, human leukocyte antigen class II (Beckmann Coulter) and CD133 (Miltenyi Biotecy) according to the manufacturer’s protocol. The cells were transported to Toronto at 4°C and arrived on the day of cell transplantation.
The cells were washed with serum free Iscove’s Modified Dulbecco’s Medium (IMDM) twice by centrifugation, the supernatant was discarded, and the cells were resuspended in the serum free media. The number of cells was counted using a cell counter and 100×106 cells were prepared in 1 mL of serum free IMDM for transplantation. For evaluation of survival rate after transportation from Germany to Toronto, 100 μL of cell suspension were transferred into 2 cell culture dishes and cultured in IMDM containing 2% fetal bovine serum. Twenty-four hours later, the number of cells in each dish was counted. The data were presented as number of attached-growth cells/number of seeded cells × 100.
Four weeks after myocardial infarction, the animals were randomly assigned to transplantation and control groups (n=8 each). Under anesthesia, the heart was exposed through a left intercostal incision, as previously described. The infarcted area was visually identified and either 100×106 cells/mL or media (1 mL) was injected at 5 locations with a tuberculin syringe within 3 minutes. The animals recovered from the procedure and were treated with antibiotics and analgesics as described above. Three days before cell transplantation, the pigs in both groups were daily immunosuppressed with cyclosporine A (5 mg · kg−1 · d−1, Norvatis Pharmaceuticals) until the end of the study.
Cardiac Function Measurement
Echocardiography was performed before coil occlusion, before cell transplantation, and 4 weeks after transplantation using a SEQUOIA Echocadiography System (ACUSON Corporation,) with a 5-MHz transducer. The animals were sedated with the inhalation of isoflurane (2%) and placed at the left lateral decubitus position. Parasternal long- and short-axis views were obtained with both M-mode and 2-dimensional echo images. Left ventricle dimensions (end-diastolic diameter [LVEDD] and end-systolic diameter [LVESD]) were measured perpendicular to the long axis of the ventricle at the midchordal level. Fractional shortening (%) was calculated as (LVEDD−LVESD)/LVEDD × 100. Left ventricular ejection fraction (LVEF) (%) was calculated automatically by the echocardiography system by the equation:
where LVEDV=7.0×LVEDD3/(2.4+LVEDD) and LVESV=7.0 × LVESD3/(2.4+LVESD).
Technetium 99m-Sestamibi Single Photon Emission Tomography
Technetium 99m-sestamibi single photon emission tomography (99mTc-SPECT MIBI) measurement was performed 3 days before cell transplantation and at 4 week after cell transplantation. After anesthesia with ketamine, 18 to 20 mCi of 99mTc-MIBI solution (E.I. du Pont de Nemours & Co) was administered intravenously 30 minutes before ECG-gated 99mTc-SPECT MIBI scan acquisition was performed with Autospect software and quantified by Cedars-Sinai Quantitative Gated SPECT software.
The data analysts independently reviewed each 99mTc-SPECT MIBI scan in a blinded fashion, performing a semiquantitative assessment of perfusion and wall motion in each segment in cross-sectional views at the apical (segments 1 to 6) and midventricular (segments 7 to 12) levels and in a vertical long-axis view (segments 19 and 20) as we described previously.4,22 Segments at the basal region (segments 13 to 18) were excluded because they were normal regions. The perfusion was graded from 0% to 100%, indicating no perfusion to normal perfusion. The wall motion was graded from 0 to 10, indicating dyskinesis, varying levels of hypokinesis, up to normal wall motion. The LVED and LVES volumes were measured with a count-based technique, and LVEF was calculated by the computer program.
Evaluation of Ventricular Function by Pressure-Volume Catheter
At end of the study (4 weeks after cell transplantation), ventricular function in the control (n=4) and transplantion (n=4) animals was evaluated by inserting conductance and Millar catheters (Millar Instruments into left ventricle through the apex.4,22 Pressure and volume data were collected with the Conduct-PC software (CardioDynamics). Three baseline values were obtained and 3 more values were then recorded after clamping of the superior and inferior vena cavae until the left ventricle had emptied, as indicated on the volume tracing. Parallel conductance was evaluated after the injection of hypertonic saline solution into the pulmonary artery, and the volume measurements were corrected for the parallel conductance. Preload recruitable stroke work and end-systolic pressure-volume relation were calculated as we described previously4,22
Morphological and Histological Studies
After completion of all measurements, the hearts were arrested with potassium chloride and rapidly excised. The coronary arteries were then perfused with 100 mL 10% formaldehyde, and the heart was fixed in diastole position with an intraventricular pressure of 30 mm Hg in formaldehyde solution. The fixed heart was sliced into 5-mm thick slices and photographed. Cross-sectional areas of the ventricular muscle, scar area, and scar thickness were measured with computed planimetry as we described previously.4 A cube of tissue (5 mm) from the center of the infarct zone was embedded in paraffin and cut into 10-μm sections. Serial cutting sections were used for immunohistochemical studies to localize the transplanted cells, infiltrated cells, and contractile protein positive cells.
For immunofluorescent staining, formalin-fixed paraffin embedded tissue sections were serially rehydrated in 100%, 95%, and 70% ethanol after deparaffinization with xylene. The antigen retrieval procedures were then used. Slides were microwaved in 10 mmol/L citrate buffer (pH 6) 3 times for 2 minutes each and washed in phosphate buffered saline (PBS) 3 times for 5 minutes each. Tissue sections were incubated in a blocking solution containing 2% goat serum and 5% bovine serum albumin in PBS to reduce nonspecific binding. Double-labeled tissue sections were incubated with goat anti-myosin heavy chain (1:200, Santa Cruz) and mouse anti-human mitochondria (1:100, Chemicon International) overnight at 4°C. Sections were then incubated for 1 hour at room temperature with a Cy3 conjugated donkey anti-mouse immunoglobulin G and an alexa fluor 647 conjugated chicken anti-goat immunoglobulin G. Nuclei were counterstained with Hoechst 33342 (2.5 μg/mL in PBS; Molecular Probes). Slides were mounted using an anti-fading aqueous mounting medium (Dakocytomation). Slides were then visualized using an Olympus Fluoview 1000 confocal microscope. Background levels of autofluorescence on unstained sections and those incubated with secondary antibodies only were used as controls to set the laser power and gain.
For immunohistochemical studies, endogenous peroxidase and biotin activities in the rehydrated sections were blocked with 3% aqueous hydrogen peroxide and Avidin/Biotin blocking kit for 15 minutes. The sample was treated for 10 minutes with 10% normal serum from the species where the secondary antibodies were generated. To retrieve antigens, the slides were pre-incubated with either pepsin (for CD3) or trypsin (for macrophage) or heated in a microwave (for mitochondria, troponin I, and myosin heavy chain) for 3 minutes in 10 mmol/L citrate buffer (pH 6.0). After washing, the sample was incubated overnight with the antibodies against human mitochondria (1:100; Chemicon), myosin heavy chain (1:400, Santa Cruz), troponin-I (1:800, Covance), CD3 (1:50; Dako), or macrophages (1:50; Neomarker) at room temperature in a moist chamber. After rinsing well with PBS, the sample was incubated for 30 minutes at room temperature with a species-specific biotinylated secondary linking antibody and then washed well with PBS. The sample was incubated with labeling reagent (Ultra Streptavidin-Horseradish Peroxidase Complex) at room temperature for 30 minutes, washed in PBS, developed with a freshly prepared working solution of Novared (Vector Laboratories) for 5 to 10 minutes, and rinsed in running tap water. The sample was counterstained with Hematoxyline. After rinsing, the sample was dehydrated through alcohols, cleared in xylene, and mounted in Permount (Fisher).
For in situ hybridization, deparaffinized samples were digested with a pepsin solution (Zytomed) for 40 minutes at 37°C. To stop the proteolysis reaction, the samples were covered with ethanol (100%), washed with distilled water, and dried again. For DNA hybridization, the samples were preincubated to denature DNA at 83°C for 10 minutes and then incubated with biotinylated DNA probe targeting specific human Alu-repeat DNA sequence (Texogene) for 1 hour at 37°C. After washing, the samples were rehydrated with Tris-buffered saline 0.05 mol/L and incubated with 3% aqueous hydrogen peroxide and an Avidin/Biotin blocking kit to inactivate endogenous peroxidase and to block biotin activities for 15 minutes. After washing with Tris-buffered saline, samples were incubated for 1 hour with rabbit antibody (1:50 dilution; Dako). To visualize bound rabbit antibodies, samples were incubated with a Dako EnVision kit according to manufacturer’s protocol and developed with freshly prepared DAB solution. Samples were counterstained with Methylene Green and mounted as described above.
For angiogenesis analysis, the myocardial biopsies were sectioned and immunohistochemically stained for factor VIII and α-smooth muscle actin. The number of large blood vessels (both factor VIII and smooth muscle actin positive) and capillaries (factor VIII positive only) were counted in a blinded fashion and compared between the groups.
Data are presented as mean±standard error. Statistical analysis was carried out with the SPSS software package (SPSS Inc) and the data were subjected to Student’s t test and ANCOVA.
Unrestricted Somatic Stem Cells
The USSCs used in the study were negative for CD14, CD31, CD33, CD34, CD45, CD56, CD133 and human leukocyte antigen class II and positive for CD13, CD29, CD44, and CD49e. After transportation from Germany to Toronto, the survival rate of USSCs before implantation was 93.3±15.4%.
The effect of USSCs on cardiac function at 4 weeks after cell transplantation was evaluated by echocardiography, 99mTc-SPECT MIBI scan, and pressure-volume measurements.
Four weeks after myocardial infarction, cardiac ejection fraction decreased significantly from 61±0.9% (baseline) to 54±1.4% (before transplantation) (P=0.004). The LVEF improved in cell transplant group at 4 weeks after transplantation, whereas it decreased in the media control group (P=0.02) (Figure 1A and 1B). The fractional shortening in the transplantation group was also significantly better (P=0.02) than in the controls (Figure 1C). LVEDD in control group increased 0.37±0.13 cm, but it only increased 0.12±0.1 cm in the transplantion group over 4 weeks after transplantation. The increase of LVED volume in the control group was 3 times greater than that of the transplantation group (9.8±8.5 versus 27.2±9.3 cm3, respectively). However, there was no statistical difference.
The LVEF in the transplantation group increased significantly after cell transplantation, whereas a decrease was observed in the control group (P=0.03, Figure 2A). The end-diastolic volume increased significantly in control group (P=0.02) compared with the transplantation group (Figure 2B).
Pressure-Volume Catheter Measurement
Figure 3 shows that preload recruitable stroke work was greater in the pigs receiving cell transplantation than in the control group (P=0.09). End-systolic elastance was also greater (P=0.07) in the transplantation group than in the controls.
Myocardial Perfusion and Wall Motion
Myocardial perfusion and wall motion were evaluated before and 4 weeks after cell transplantation by 99mTc-SPECT MIBI scan. Figure 4A shows the perfusion of each segment at the apical and midventricular levels before and after cell transplantation. The overall changes in myocardial perfusion were significantly improved (P=0.02) in the transplantation group compared with the changes in the control group (Figure 4B). In most animals, the infarcted regions were segments 19 and 20 (apex region) and the USSCs were transplanted into these regions. The myocardial perfusion rates in these regions were significantly different (P=0.007) in the transplantation and control groups (control, –2.4±1.8; transplantation, 4.6±1.6). Figure 5 shows the regional wall motion of the transplantation and control groups. The wall motion in the transplantation group was significantly greater (P=0.004) than in the control group (Figure 5B). The myocardial wall motions in segments 19 and 20 were different (P=0.057) in the transplantation and control groups (control, 0.47±0.60; transplantation, 1.66±0.47).
Morphological and Histological Studies
Figure 6 shows scar size, scar thickness, and ventricular volumes of the hearts that received USSCs or culture media. The scar size was smaller (P=0.01) and scar thickness was greater (P=0.05) in the transplantation group than the control group. However, there was no difference in left ventricular volume between the groups (P=0.21).
Myocardial tissues from the infarct regions of both the transplantation and control groups were evaluated histologically (Figure 7). To identify the transplanted USSCs, we used independent methods: Antibodies against human mitochondria protein (Figure 7A and 7F) and DNA probes targeting human Alu-repeat DNA sequence (Figure 8), which are species-specific and do not cross-react with porcine mitochondria or chromosome DNA, respectively (data not shown). Both methods showed positive cells in the myocardial scar tissue of the USSC transplantation group (Figures 7A and 8⇓A) but not in the control group (Figures 7F and 8⇓B). Antibodies against myosin heavy chain (Figure 7B and 7G) and troponin-I (Figure 7C and 7H) demonstrated some positive cells in the scar tissue from the transplantation group (Figure 7B and 7C) but not in the scar area of the control group (Figure 7G and 7H). In some engrafted cells, colocalization of human cell markers and cardiomyocyte-specific markers (myosin heavy chain [Figure 9A through 9C] and troponin I [Figure 9 D through 9G]) were observed. Antibodies against CD3 (Figure 7D and 7I) and macrophages (Figure 7E and 7J) showed CD3- positive cells and macrophages in the infarcted region of the transplantation group (Figure 7D and 7E) but not the control group (Figure 7I and 7J).
Blood vessels were found in the myocardial scar tissue in both the transplantation and control groups (Figure 10A and 10B), and total number of blood vessels in the cell transplantation group was significantly greater (P=0.02) than that of the control group (Figure 10C). Density of large blood vessels in transplantation group was 17.3±2.6 vessels/0.2 mm2, which was greater (P=0.05) than the 12.0±0.7 vessels/0.2 mm2 of the control group. Capillary density of the transplantation group was also higher (P=0.02) than that of the control group (26.8±2.5 versus 17.8±1.6 vessels/0.2 mm2, respectively).
Although cardiomyocyte regeneration has recently been reported in the myocardium after infarction, it cannot prevent the injured myocardium from being replaced by nonfunctional fibrous tissue, which may contribute to the onset and progression of ventricular dysfunction. The high morbidity of congestive heart failure and the shortage of donor hearts for transplantation demand a search for new approaches to prevent heart failure after a myocardial infarction. Recently, a number of studies have demonstrated that implantation of cells into damaged myocardia could regenerate the infarct, and the accumulated evidence suggests that cell transplantation may prevent heart failure.1–9
When the cell transplantation was proposed, the implanted muscle cells were believed to improve ventricular function by synchronous contraction with the host myocardium. None of the recent studies, however, has demonstrated that the implanted cells beat synchronously with the recipient heart. In addition, cardiac function has been shown to improve after the transplantation of cardiomyocytes as well as noncontractile cells, such as smooth muscle cells.10 Recently, a number of studies demonstrated that stem cells, such as bone marrow mononuclear cells or bone marrow-derived mesenchymal stem cells, could be used to regenerate damaged myocardia.8,22,23 In the present study, we evaluated unrestricted somatic stem cells as a cell source for myocardial regeneration.
Human umbilical cord blood contains USSCs, a population of pluripotent stem cells. USSCs are different from mesenchymal stem cells, however, and may be less mature and thus have a higher proliferative potential, an extended life span, and longer telomeres. In culture, these cells grows adherently and can be expanded up to 1015 cells without losing pluripotency, as they can differentiate into osteoblasts, chondroblasts, adipocytes, hematopoietic cells, and neural cells.20 The differentiation of USSCs along mesodermal and endodermal pathways was demonstrated in animal models. In addition, USSCs are abundantly available, can be routinely harvested without risk to the donor, and are seldom infected with agents, which gives them a definite advantage for the development of cell therapeutics in regenerative medicine. Umbilical cord stem cells are routinely frozen and therefore can be readily available for transplantation into an infarcted region without having to wait for expansion in a cell culture facility. Hence, the USSCs offer the promise of an “off the shelf” therapy for cardiac regeneration.
In the present study, we demonstrated that human USSCs transplanted into the infarcted myocardium in a porcine model survived in the myocardial scar tissue and prevented scar thinning and ventricular dilatation. The implanted cells also increased regional perfusion and wall motion and improved cardiac function. A number of cell clusters were observed in the myocardial scar tissue of transplanted area in the cell transplantation group. Many of the cells in the cell islands stained positive for human mitochondrial proteins and for human Alu-repeat chromosomal DNA, which suggested that they were the implanted cells derived from human USSCs. Serial section technique of the histological samples allowed us to identify expression of the cardiomyocyte-specific markers (myosin heavy chain and troponin I) of these implanted USSCs. In addition, costaining of human mitochondria and troponin I revealed the expression of cardiomyocyte markers in some engrafted human USSCs. Similar results were also observed in our previous study, which demonstrated that some of the cultured stem cells converted to myogenic progenitor cells within the infarct region and could contribute to preservation of ventricular function.22 These additional muscle-like cells increased scar thickness and prevented scar expansion, which might be a possible mechanism to improve cardiac function by USSCs. Supporting this hypothesis, echocardiography and 99mTc-SPECT MIBI scans showed that the LVED volume in the transplantation group was significantly smaller than that in the control group. According to Laplace Law, the transplanted heart should have better cardiac function.
The beneficial effect of implanted USSCs on global cardiac function might also be explained by an improvement in myocardial perfusion. The cell transplantation group had better regional perfusion, especially within the infarcted region where the cells were directly injected. The engrafted cells may provide progenitor cells, which participate in the formation of new blood vessels,24 or they may release vascular growth factors to increase regional perfusion by either induction of new vessel formation or vessel growth. The regional wall motion in the transplantation group was also better preserved than in the control group, especially within the transplanted area. Both improved regional perfusion and increased regional wall motion may have contributed to the improvement of global cardiac function.
A major limitation of current study and future clinical trials using USSCs to regenerate damaged myocardium is immunorejection. Although we have identified implanted cells within the transplanted area, the animals were immunosuppressed during the study. Immunohistochemical staining showed T-cells (CD3-positive cells) and macrophages in the vicinity of the transplanted cells. Future studies will be required to evaluate the long-term beneficial effect of USSCs in animal models with and without immunosuppression. The survival rate of implanted cells in the transplanted area and distribution of implanted cells in other organs should be evaluated. Finally, a dose-response relationship will need to be established between the number of implanted USSCs and the recovery of cardiac function after a myocardial infarction.
In conclusion, unrestricted somatic stem cells from human umbilical cord blood were successfully transplanted into myocardial scar tissue. The transplanted USSCs engrafted in a cluster-like pattern, expressed muscle-specific antigens, increased myocardial perfusion, prevented ventricular remodeling, and improved ventricular function. These data suggest that USSCs could be a promising candidate for the cellular cardiomyoplasty in the future.
We thank Dr Massimo Cimini for his help on the confocal analysis. The study was supported by Kourion Therapeutics AG, Germany.
↵*The first 2 authors contributed equally to this work.
Koh GY, Klung MG, Soonpaa MH, Field LJ. Long-term survival of AT1 cardiomyocyte grafts in syngeneic myocardium. Am J Physiol. 1993; 264: H1727–H1733.
Leor J, Patterson M, Quinones MJ, Kedes LH, Kloner RA. Transplantation of fetal myocardial tissue into the infarcted myocardium of rat: a potential method for repair of infarcted myocardium. Circulation. 1996; 94 (suppl II): II-332–II-336.
Tomita S, Li RK, Weisel RD, Mickle DA, Kim EJ, Sakai T, Jia ZQ. Autologous transplantation of bone marrow cells improves damaged heart function. Circulation. 1999; 100 (suppl II): II-247–II-256.
Perin EC, Dohmann H, Borojevic R, Silva SA, Sousa AL, Mesquita CT, Rossi MI, Carvalho AC, Dutra HS, Dohmann HJ, Silva GV, Belem L, Vivacqua R, Rangel FO, Esporcatte R, Geng YJ, Vaughn WK, Assad JA, Mesquita ET, Willerson JT. Transendocardial, autologous bone marrow cell transplantation for severe, chronic ischemic heart failure. Circulation. 2003; 107: 2294–2302.
Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach C, Fichtner S, Korte T, Hornig B, Messinger D, Arseniev L, Hertenstein B, Ganser A, Drexler H. Intracoronary autologous bone-marrow cell transplantation after myocardial infarction: the BOOST randomized controlled clinical trial. Lancet. 2004; 364: 141–148.
Heeschen C, Lehmann R, Honold J, Assmus B, Aicher A, Walter DH, Martin H, Zeiher AM, Dimmeler S. Profoundly reduced neovascularization capacity of bone marrow mononuclear cells derived from patients with chronic ischemic heart disease. Circulation. 2004; 109: 1615–1622.
Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, Moorman MA, Simonetti DW, Craig S, Marshak DR. Multilineage potential of adult human mesenchymal stem cells. Science. 1999; 284: 143–146.
Wakitani S, Saito T, Caplan AI. Myogenic cells derived from rat bone marrow mesenchymal stem cells exposed to 5-azacytidine. Musc Nerve. 1995; 18: 1417–1426.
Ferrari G, Cusella-DeAngelis G, Coletta M, Paolucci E, Stornaiuolo A, Cossu G, Mavilio F. Muscle regeneration by bone marrow derived myogenic progenitors. Science. 1998; 279: 1528–1530.
Kogler G, Sensken S, Airey JA, Trapp T, Muschen M, Feldhahn N, Liedtke S, Sorg RV, Fischer J, Rosenbaum C, Greschat S, Knipper A, Bender J, Degistirici O, Gao J, Caplan AI, Colletti EJ, Almeida-Porada G, Muller HW, Zanjani E, Wernet P. A. new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential. J Exp Med. 2004; 200: 123–135.
Airey JA, Almeida-Porada EJ, Colletti CD, Porada CD, Chamberlain J, Movsesian M, Sutko JL, Zanjani ED. Human mesenchymal stem cells form Purkinje fibers in fetal sheep heart. Circulation. 2004; 109: 1401–1407.
Asahara T, Masuda H, Takahashi T, Kalka C, Pastore C, Silver M, Kearne M, Magner M, Isner JM. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization. Cardiovasc Res. 1999; 85: 221–228.