Study protocols were approved by pertinent Institutional Ethics Committee, Institutional Review Board and Institutional Animal Care and Use Committee.
Ischemic heart disease patient cohort
Patients undergoing coronary artery bypass surgery provided informed consent. Information concerning age, gender, history, medications, perioperative laboratory evaluation and coronary assessment is presented as Supplemental Tables 1
Stem cell culture
Human bone marrow aspirates, obtained during coronary artery bypass surgery following sternotomy, yielded plastic-adherent hMSC with identity confirmed by Fluorescence-Activated Cell Sorting (FACS) using the CD34−
panel (Supplemental Figure 1
). Cells were cultured at 37°C in advanced-MEM supplemented with 5% human platelet lysate (14
), and stimulated with TGF-β1 (2.5 ng/ml), BMP-4 (5 ng/ml), Activin-A (5 ng/ml) (15
), FGF-2 (10 ng/ml) (6
), IL-6 (100 ng/ml) (17
), Factor IIa (hα-thrombin, 1 U/ml) (18
), IGF-1 (50 ng/ml) (6
) and/or retinoic acid (1 μM) (19
). Phospho-AKT inhibition was achieved with SR13668 (50 nM). Differentiation beyond the cardiopoietic state was carried-out with 1% platelet lysate in the presence of FGF-2 and hα-thrombin for 10 days. To scale-up production under GMP compliance, cell culture procedures were site-certified by independent regulatory review as meeting mandatory environmental and quality assurance criteria required to ensure traceability, purity, homogeneity and sterility for batch testing and release (14
Real-time quantitative polymerase chain reaction (qRT-PCR) was performed using a TaqMan PCR kit (Applied Biosystems) in triplicate. Threshold cycle (CT) values were determined using the 2−ΔΔCT method, normalized to human specific GAPDH (P/N 435,2662-0506003). Genes representative of cardiac transcriptional activity included Nkx2.5 (Hs00231763_m1), GATA4 (Hs00171403_m1), MEF2C (Hs00231149_m1), Tbx5 (Hs00361155_m1), FOG1 (Hs00542350_m1), MESP1 (Hs00251489_m1), GATA6 (Hs00232018_m1), and Flk-1 (Hs00911699_m1).
Protein and DNA probing
Immunostaining for cardiac transcription factors Mef2C (1:400, Cell Signaling Technologies), Nkx2.5 (1:150, Santa Cruz Biotechnology), Gata-4 (1:150, Santa Cruz Biotechnology), Tbx5 (1:5000, Abcam), Mesp-1 (1:250, Novus Bio), Fog-2 (1:100, Santa Cruz Biotechnology), along with Phospho-AKTSer473 (1:100, Cell Signaling Technologies), human-specific Troponin-I (1:100, Abcam), sarcomeric α-actinin (1:500, Sigma-Aldrich), MLC2v (1:500, Synaptic Systems), Sca-1 (1:100, R&D Systems), CD-31/PECAM-1 (1:500; Beckman Coulter), α-smooth muscle actin (1:500; Sigma), human Lamin A/C (1:50, Novacastra), Connexin-43 (1:100, Zymed Laboratories) and Ki67 (1:500, NeoMarkers) was performed following fixation in 3% paraformaldehyde and permeabilization with 1% Triton X-100. Human ALU-DNA probing (Biogenex) required hybridization (85°C, 10 min) and incubation at 37°C overnight followed by anti-Fluorescein GFP-labeled secondary detection. For genomic probing, human and murine DNA were labeled in Vysis® Spectrum Green or Orange d-UTP via nick translation followed by hybridization and denaturation (Abbott Molecular). Frozen tissue was fixed (3:1 methanol:acetic acid), and pretreated in 2xSSC for 2 min (73°C), placed 5 min in 1 M Tris, 0.5 M EDTA (90°C), 2 min in 4 mg Pepsin/L, 0.9% NaCl (37°C), and 30 s in 70% ETOH. Slides were denatured in 70% formamide (73°C), followed by a 70%/85%/100% ETOH series. Denatured genomic probes were applied to sections, hybridized overnight (37°C), and washed in 2xSSC/0.1% NP40 solution (73°C). Slides were counterstained with DAPI-containing mounting medium.
Images were obtained with LSM 510/LSM 700 laser scanning confocal or Apotome structured illumination microscopes (Carl Zeiss). Quantification of cytosolic and nuclear induction of Nkx2.5, Mef2C or GATA-4 was performed using Zeiss Axioplan and Metamorph software (Molecular Devices). Cell cycle activity was quantified using a hematocytometer. Cell structure was resolved by transmission electron microscopy (20
). Calcium transients were monitored in cells loaded with 5 μM of the calcium-selective probe fluo-4-acetoxymethyl ester (Molecular Probes) using a temperature controlled Zeiss LSM 510 or Apotome microscope with line-scan or phase images acquired during 1 Hz stimulation (15
Myocardial infarction was performed in 8–12 weeks old nude, immunocompromised mice (Harlan) (21
). Following intubation and ventilation under isoflurane anesthesia, proximal left anterior descending artery was ligated with 9-0 suture. Injury induced ST-changes on electrocardiography (MP150; Biopac) and akinetic regions on echocardiography (Sequoia 512, Siemens; Vevo2100, VisualSonics). Following a blinded design, 1-month post-infarction a total of 600,000 hMSC, suspended in 12.5 μl phosphate buffered saline, was injected under microscopic visualization in five epicardial sites (120,000 cells per site) in the anterior wall of the left ventricle. Saline-treated group underwent identical procedure. Left ventricular function and structure were serially followed by echocardiography at pre-infarction, 1-month after infarction (pre-cell therapy), 1-, 2-, 6-, 12-, and 20-month post-cell therapy. Ejection fraction (%) was calculated as [(LVVd-LVVs)/LVVd]x100, where LVVd is left ventricular end-diastolic volume (μl), and LVVs is left ventricular end-systolic volume (μl).
Data are presented as mean±SD unless otherwise indicated (JMP 8; SAS Institute). Paired group analysis was performed using Student’s t-test for each pair of samples without correction for type I error, and non-parametrically validated using Wilcoxon signed rank test (JMP). Kaplan-Meier analysis with log-rank testing was applied for survival evaluation. A p value <0.05 was considered significant.