Supporting Information

Files in this Data Supplement:

Table 1. Antibodies for immunocytochemistry

Direct labeling corresponds to the utilization of fluorochrome-conjugated primary antibodies; indirect labeling reflects the use of non-conjugated primary antibody and fluorochrome-conjugated secondary antibody. Mixtures of fluorochrome-conjugated primary antibodies were utilized: *, Cocktail 1, §, Cocktail 2, †, Cocktail 3, and ‡, Cocktail 4. QD indicates direct labeling of primary antibodies with quantum dots (QD); indirect/QD indicates that both indirect labeling and direct labeling with QD were employed.

Table 2. Antibodies for FACS analysis

NK, natural killer; TF, transcription factor; PECAM-1, platelet endothelial cell adhesion molecule 1; VEGF-R2, vascular endothelial growth factor receptor 2.

Each sample was mixed with 15 ml of Platinum PCR BlueMix solution (Invitrogen) and 0.2 mM primer and subjected to PCR. For MLC2v, the rat and human primers were mixed, and the PCR was performed as follows: 94°C for 15 sec; 60 cycles of 94°C for 15 sec, 55°C for 30 sec, and 72°C for 30 sec; 72°C for 3 min. For Alu, the ARC-261r primer was employed in the first-round PCR: 94°C for 2 min; 40 cycles of 94°C for 15 sec, 55°C for 30 sec, and 72°C for 30 sec; 72°C for 2 min. For nested PCR, the first-round PCR product was diluted 1/200, and the ARC-17 and ARC-218r primers were used: 94°C for 2 min; 25 cycles of 94°C for 15 sec, 60°C for 30 sec, and 72°C for 20 sec; 72°C for 2 min. PCR products were separated by 2% agarose gel electrophoresis.

Table 5. Human primers and probes

Table 6. Magnitude of sampling

Number of cells counted in all experiments to evaluate the fraction of cells positive for c-kit, MDR1 and Sca-1-like proteins, and markers of commitment to myocyte, smooth muscle cell and endothelial cell lineages. ⁽²⁾ Number of cells counted in all experiments at all passages to assess the fraction of cells expressing the markers of stemness, lineage commitment, proliferation and senescence. ⁽³⁾ Number of nuclei in which telomeric length was measured ⁽⁴⁾ Number of clonogenic cells counted in all experiments to establish the fractions of undifferentiated and differentiated cells, including the magnitude of cell proliferation. ⁽⁵⁾ Aggregate area of regenerated myocardium in all animals measured morphometrically to calculate the volume of regenerated myocardium. ⁽⁶⁾ Aggregate area of tissue sections measured morphometrically to obtain the volume fractions of infarcted and surviving myocardium. ⁽⁷⁾ Total number of myocyte nuclei analyzed to acquire data necessary for the calculation of the density of myocyte nuclei per unit volume of myocardium. Parameters indicated under (6) and (7) were necessary for the computation of infarct size as the fraction of myocytes lost from the left ventricle (see Methods). ⁽⁸⁾ Aggregate area of regenerated myocardium in all animals evaluated morphometrically in order to calculate the volume fraction of myocytes and coronary vessels in the regenerated myocardium. ⁽⁹⁾ Number of newly formed myocytes analyzed by confocal microscopy for the calculation of the myocyte cell volume. ⁽¹⁰⁾ Number of profiles of coronary vessels counted in all animals for the calculation of length density of arterioles and capillaries. ⁽¹¹⁾ Number of nuclei in the regenerated myocardium assessed for the presence of human and mouse or rat chromosomes to exclude cell fusion.

SI Materials and Methods

We studied six normal human hearts obtained at autopsy from patients who died from causes other than cardiovascular diseases. Samples of the left ventricle were fixed in formalin and embedded in paraffin (1). These specimens were used to identify putative stem cell niches and the pattern of division of replicating hCSCs.

Discarded myocardial specimens were obtained from 88 patients who underwent cardiac surgery. We had no knowledge concerning the age, type of cardiac disease, and duration of the disease. Additionally, we had no information concerning the atrial or ventricular origin of the specimens. The size of the samples varied from ~20-100 mg. Two protocols were used for the isolation of hCSCs. The first consisted of the enzymatic dissociation of the samples of human myocardium in a solution containing collagenase from which c-kit^POS cells were sorted with rabbit anti-c-kit (Santa Cruz Biotechnology, Santa Cruz, CA) and magnetic immunobeads (Miltenyi, Auburn, CA) (2, 3). Sorted cells were plated immediately at low density (<1 cell per 50-100 mm²) to obtain multicellular clones derived from single founder cells or analyzed by FACS for cell characterization. With the second protocol, samples were minced and seeded onto the surface of uncoated Petri dishes (Corning, Corning, NY) containing F12 medium (Gibco, Grand Island, NY) supplemented with 5-10% FBS (Gibco) and insulin-selenium-transferrin mixture (Sigma, St. Louis, MO). Cells outgrown from the tissue specimens were sorted for c-kit (see above) and cultured (2, 3). Cell phenotype was defined by FACS and immunocytochemistry (2, 4, 5). For this purpose, the separated c-kit^POS-cells were fixed in 4% paraformaldehyde for 15 min at room temperature and tested for markers of cardiac, skeletal muscle, neural, and hematopoietic cell lineages (SI Table 1) to detect lineage negative (Lin^-)-hCSCs (1-3). Committed cells expressed the stem cell antigen and transcription factors and cytoplasmic proteins of cardiac cells. For immunocytochemistry, when possible, antibodies were directly labeled with fluorochromes (Molecular Probes, Eugene, OR) or quantum dots (Quantum Dot Corporation, Hayward, CA) to avoid cross-reactivity and autofluorescence (SI Table 1) (1, 3, 6, 7). Antibodies used for FACS analysis (2, 4, 5) are listed in SI Table 2.

Individual c-kit^POS-cells were seeded in single wells of Terasaki plates at a density of 0.25-0.5 cells per well (2, 3). Single-cell deposition was confirmed microscopically and wells containing more than one cell were excluded. After ~3 weeks, colonies developed and were expanded in F12 medium. Cell differentiation was induced by MEM containing 10% FBS and 10^-8 M dexamethasone (2, 3). In some experiments, BrdU (10 mM; Roche, Indianapolis, IN) was added three times a day for 5 days (2, 3). BrdU incorporation was determined by immunostaining with monoclonal antibody (Roche). Moreover, clones were fixed in 4% paraformaldehyde for immunocytochemistry. Cell phenotype was also defined by FACS.

To determine whether myocytes derived from clonogenic hCSCs were functionally competent, cells stained by DiI only were subjected to depolarizing pulses, 3 ms in duration, and twice-diastolic threshold in intensity, by platinum electrodes (8). Additionally, nonclonogenic hCSCs were infected with a lentivirus expressing EGFP under the control of the cytomegalovirus promoter (LentiV-CAG-GFP; kindly provided by Dr. E. Marban, Johns Hopkins University School of Medicine, Baltimore, MD). Subsequently, the EGFP-positive hCSCs were cocultured with neonatal rat cardiomyocytes to induce differentiation. Briefly, cultures of ~10⁵ neonatal myocytes were established in 35-mm dishes (9), and 3 days later, EGFP-positive hCSCs were plated. Electrical stimulation at 1 Hz was performed 10-15 days later. For measurements of calcium transients, cells were loaded with Rhod-2 (Molecular Probes), 1 mM, for 30 min before stimulation. Cells were examined by two-photon microscopy (BioRad MP2100, BioRad, Hercules, CA) working in a line-scan mode (10). Myocytes were scanned 50 times per sec along the laser line and contraction was recorded. The changes in width of the scan reflected the changes in dimension of the cell, i.e., shortening. EGFP and Rhod-2 were excited at 960 nm with mode-locked Ti:Sapphire femtosecond laser (Tsunami; Spectra-Physics, Mountain View, CA). The corresponding images were acquired at emission wavelengths of 515 (EGFP) and 600 nm (Rhod-2). By this approach, the green fluorescence of EGFP and the red fluorescence of Rhod-2 were detected simultaneously in the cells.

hCSCs were homogenized in CHAPS buffer and centrifuged at 4°C. Untreated and RNase-treated cell extracts were incubated with [g-³²P]ATP-end-labeled telomerase substrate (TS oligonucleotide: 5'-AATCCGT-CGAGCAGAGTT-3'), Taq polymerase and anchored reverse primer (3'-GCGCGC [CTTACC]₃CTAACC-5') for 30 min. Samples were exposed to 30 amplification cycles (11-13). PCR products were separated on 12% polyacrylamide gel. Telomerase-induced reaction generated a 6-bp ladder. The OD of the bands was normalized for PCR efficiency (11-13).

Telomere length in nuclei was evaluated by quantitative fluorescence in situ hybridization (Q-FISH) and confocal microscopy. A fluorescein isothiocyanate-peptide nucleic acid (FITC-PNA) probe was used (14). The fluorescent signals measured in lymphoma cells with short (L5178Y-S, 7 kbp) and long (L5178Y-R, 48 kbp) telomeres (kindly provided by Dr. M. A. Blasco, Spanish National Cancer Centre, Madrid, Spain) were used to compute absolute telomere length.

Myocardial infarction was produced in anesthetized female immunodeficient Scid mice (4, 5) and Fischer 344 rats (2) treated with a standard immunosuppressive regimen (15). Shortly after coronary occlusion or 5 days later, two injections of ~40,000 nonclonogenic or clonogenic hCSCs were made at the opposite sites of the border zone (2, 4). Animals were exposed to BrdU and killed from 5 to 21 days after infarction and cell implantation (2, 4, 5, 16). In a subset of animals, echocardiography was performed 2-3 days before measurements of left ventricular (LV) pressures and + and -dP/dt (2, 4, 5, 16). The heart was arrested in diastole and fixed by perfusion with formalin. In each heart, infarct size and the formation of human myocytes, arterioles, and capillaries was determined (2, 5, 16, 17). Samples were examined blindly (J.K.).

For infarct size, the number of myocytes in the LV of control and infarcted hearts was measured by employing a methodology well established in our laboratory (2, 5, 17). The quotient between the number of LV myocytes in sham-operated animals and the number of myocytes present in the infarcted LV gives the percentages of myocytes lost and remaining after infarction. The percentage of myocytes lost provides a quantitative measurement of infarct size, whereas the percentage of myocytes left correlates with ventricular function (7, 17).

For the detection of calcium transients in human myocytes and the surrounding myocardium, infarcted mice were used 2 weeks after coronary occlusion and the implantation of EGFP-positive hCSCs. The heart was excised, arrested in diastole with 30 mM KCl, and placed in a bath mounted on the stage of a two-photon microscope working in a line-scan mode (10). The heart was first perfused retrogradely through the aorta with an oxygenated Tyrode solution containing the calcium indicator Rhod-2, 10 mM. Subsequently, the heart was continuously perfused and superfused with oxygenated Tyrode. Cytochalasin D (50 mM) and acetylcholine (10 mM) were added to the perfusate to inhibit contraction and spontaneous activity (18). Electrical stimulation was accomplished by depolarizing pulses, 3 ms in duration. By this approach, the green fluorescence of EGFP and the red fluorescence of Rhod-2 were detected in human myocytes. Similar determinations were performed in EGFP-negative mouse myocytes labeled by Rhod-2 only.

Human cells were detected by in situ hybridization with a probe (Biogenex, San Ramon, CA) against the human-specific Alu repeat sequences (19). Similarly, human X-chromosomes, and mouse and rat X-chromosomes were identified as described (5, 7, 13, 16, 20, 21). DNA was extracted from tissue sections of the viable and infarcted LV of rats treated with human cells. PCR was conducted for human Alu and rat and human Myosin light chain 2v sequences (SI Table 3).

For real-time RT-PCR, the infarcted myocardium was obtained from rat hearts treated with clonogenic hCSCs and killed from 5 to 21 days after coronary ligation and cell implantation. Total RNA was purified by using RNeasy Mini kit and RNase-Free DNase Set (Qiagen, Valencia, CA), and in each sample, ~50-100 mg of RNA were collected. RNA was eluted in 100 ml of RNase-free water, and 50 mg of total RNA were used for poly(A)RNA selection by using MicroPoly(A)Purist (Ambion, Austin, TX). Nearly 1.5-2 mg of poly(A)RNA were recovered from each sample and treated with DNase I. After heat inactivation of DNase I, 500 ng of poly(A)RNA were used for reverse transcription (RT) into cDNA by using SuperScript III cDNA synthesis kit (Invitrogen, Carlsbad, CA); RNA was incubated with 5'-phosphorylated oligo(dT)₂₀ primer for 3.5 h at 50°C. Approximately 25 ng of poly(A)RNA were used for non-RT control reaction. Synthesized cDNA was treated with 2 units of RNase H (Ambion) at 37°C for 20 min.

Real-time RT-PCR analysis was performed on 7300 Real Time PCR System (Applied Biosystems, Valencia, CA) and run in duplicate by using 1/20th of the cDNA per reaction. Primers were designed from available human sequences by using the primer analysis software Primer Express v2.0 (Applied Biosystems); in our protocols, primers and probes did not cross-react with rat mRNA. This was avoided by applying distinct annealing temperatures in the detection of the various human transcripts (gene-specific primer temperature). Human myosin light-chain 2v (h-MLC2v), connexin 43 (h-Cx43), myosin heavy-chain 11 (h-Mhc 11), von Willebrand factor (h-vWf), and the housekeeping gene GAPDH (h-GAPDH) were analyzed because the sequences of these human genes were more readily distinguished from the corresponding rat genes (GenBank accession numbers; SI Table 4).

The PCR-reaction included 1 ml of template cDNA, 500 nM forward and reverse primers, and 100 nM probe conjugated with the fluorescent dye FAM in a total volume of 25 ml. Cycling conditions were as follows: 95°C for 10 min, followed by 50 cycles of amplification (95°C denaturation for 15 seconds, and 68°C or 70°C combined annealing/extension for 1 min). FAM signal was detected at the end of each cycle. Data were analyzed by using the Automatic Baseline of the Sequence Detection software (v. 1.2.2; Applied Biosystems), and the threshold was fixed at 0.01 manually for cycle threshold (Ct) determination. PCR efficiency was evaluated by using a standard curve of five serial dilution points; quantified values were normalized against the input determined by the housekeeping human gene GAPDH. After normalization, the samples were compared with clonogenic hCSCs identical to those injected in the animals. Untreated rat hearts were processed in the same manner and used as negative control (for human primers and probes, see SI Table 5).

Real-time PCR products were run on 2% agarose/1´ TBE gel. Amplified fragments were cut out, and DNA was extracted by using QIAquick Gel Extraction kit (Qiagen). DNA was eluted in 30 ml of 10 mM Tris buffer (pH 8.5), and amplified by Platinum Blue PCR Supermix (Invitrogen) in the presence of 260 nM of the forward and reverse primers used for real-time PCR. PCR reaction was carried out in Eppendorf Mastercycler. Cycling conditions were as follows: 94°C for 2 min, followed by 35 cycles of amplification (94°C denaturation for 20 seconds, 58°C annealing for 30 seconds, 72°C elongation for 20 seconds) with a final incubation at 72°C for 3 min. After purification using QIAquick PCR Purification kit (Qiagen), samples were submitted to the DNA Sequencing Facility at Cornell University (New York, NY) to obtain the DNA sequence.

A loxP-Cre-recombinase system was used to examine whether fusion events occurred between injected hCSCs and recipient cardiac cells in the infarcted mouse heart. hCSCs were infected with a lentivirus expressing a nuclear localized form of Cre-recombinase (lenti-Cre; ref. 22); ~60,000 hCSCs were incubated with 0.5 ml of culture medium containing 1.5 ´ 10⁶ pfu of lenti-Cre at 37°C for 1 min. Subsequently, 10 mg of polybrene was added and cells were incubated at 37°C for 1 h. Infection efficiency was established after fixation and staining for Cre-recombinase and was found to be ~90%. Nearly 60,000 lenti-Cre-infected hCSCs were injected in the infarcted heart of reporter mice in which EGFP expression was blocked by a loxP-flanked STOP fragment. Successful Cre excision would result in the appearance of EGFP in the fused mouse cells and their progeny. Mice were killed 10 days later and the presence of EGFP, Cre-recombinase, and Alu was determined by immunohistochemistry. The myocardium of transgenic mice in which EGFP was under the control of the cardiac specific a-MHC promoter was used as positive control.

The magnitude of sampling used in each in vitro and in vivo determination is listed in SI Table 6. In all cases, results are presented as mean ± SD. Statistical significance was determined by Wilcoxon rank sum test and Bonferroni method. All P values are two-sided, and P < 0.05 was considered to be significant.

1. Urbanek K, Torella D, Sheikh F, De Angelis A, Nurzynska D, Silvestri F, Beltrami CA, Bussani R, Beltrami AP, Quaini F, et al. (2005) Proc Natl Acad Sci USA 102:8692-8697.

2. Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, Chimenti S, Kasahara H, Rota M, Musso E, Urbanek K, et al. (2003) Cell 114:763-766.

3. Linke A, Muller P, Nurzynska D, Casarsa C, Torella D, Nascimbene A, Castaldo C, Cascapera S, Bohm M, Quaini F, et al. (2005) Proc Natl Acad Sci USA 102:8966-8971.

4. Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R, Nadal-Ginard B, Bodine DM, et al. (2001) Nature 420:701-705.

5. Urbanek K, Rota M, Cascapera S, Bearzi C, Nascimbene A, De Angelis A, Hosoda T, Chimenti S, Baker M, Limana F, et al. (2005) Circ Res 97:663-673.

6. Alivisatos P (2004) Nat Biotech 22:47-52.

7. Leri A, Kajstura J, Anversa P (2005) Physiol Rev 85:1373-1416.

8. Rota M, Boni A, Urbanek K, Padin-Iruegas ME, Kajstura TJ, Fiore G, Kubo H, Sonnenblick EH, Musso E, Houser SR, et al. (2005) Circ Res 97:1332-1341.

9. Kajstura J, Cheng W, Reiss K, Anversa P (1994) Exp Cell Res 215:273-283.

10. Plank DM, Sussman MA (2005) Methods Cell Sci 25:123-133.

11. Leri A, Barlucchi L, Limana F, Deptala A, Darzynkiewicz Z, Hintze TH, Kajstura J, Nadal-Ginard B, Anversa P (2001) Proc Natl Acad Sci USA 98:8626-8631.

12. Chimenti C, Kajstura J, Torella D, Urbanek K, Heleniak H, Colussi C, Di Meglio F, Nadal-Ginard B, Frustaci A, Leri A, et al. (2003) Circ Res 93:604-613.

13. Urbanek K, Quaini F, Tasca G, Torella D, Castaldo C, Nadal-Ginard B, Leri A, Kajstura J, Quaini E, Anversa P (2003) Proc Natl Acad Sci USA 100:10440-10445.

14. Leri A, Franco S, Zacheo A, Barlucchi L, Chimenti S, Limana F, Nadal-Ginard B, Kajstura J, Anversa P, Blasco MA (2003) EMBO J 22:131-139.

15. Zimmermann WH, Didie M, Wasmeier GH, Nixdorff U, Hess A, Melnychenko I, Boy O, Neuhuber WL, Weyand M, Eschenhagen T (2002) Circulation 106:I151-157.

16. Kajstura J, Rota M, Whang B, Cascapera S, Hosoda T, Bearzi C, Nurzynska D, Kasahara H, Zias E, Bonafe M, et al. (2005) Circ Res 96:127-137.

17. Anversa P, Olivetti G (2002) in Handbook of Physiology: The Cardiovascular System: The Heart, eds Page E, Fozzard HA, Solaro RJ (Oxford Univ Press, Oxford), pp 75-144.

18. Rubart M (2004) Circ Res 95:1154-1156.

19. Just L, Timmer M, Tinius J, Stahl F, Deiwick A, Nikkhah G, Bader A (2003) J Neurosci Methods 126:69-77.

20. Quaini F, Urbanek K, Beltrami AP, Finato N, Beltrami CA, Nadal-Ginard B, Kajstura J, Leri A, Anversa P (2002) N Engl J Med 346:5-15.

21. Dawn B, Stein AB, Urbanek K, Rota M, Whang B, Rastaldo R, Torella D, Tang XL, Rezazadeh A, Kajstura J, et al. (2005) Proc Natl Acad Sci USA 102:3766-3771.

22. Oka M, Chang LJ, Costantini F, Terada N (2005) Methods Mol Biol 329:273-281.