Transplantation of Nanoparticle Transfected Skeletal Myoblasts Overexpressing Vascular Endothelial Growth Factor-165 for Cardiac Repair
Background— We investigated the feasibility and efficacy of polyethylenimine (PEI) based human vascular endothelial growth factor-165 (hVEGF165) gene transfer into human skeletal myoblasts (HSM) for cell based delivery to the infarcted myocardium.
Methods and Results— Based on optimized transfection procedure using enhanced green fluorescent protein (pEGFP), HSM were transfected with plasmid-hVEGF165 (phVEGF165) carried by PEI (PEI- phVEGF165) nanoparticles. The transfected HSM were characterized for transfection and expression of hVEGF165 in vitro and transplanted into rat heart model of acute myocardial infarction (AMI): group-1=DMEM injection, group-2= HSM transplantation, group-3= PEI-phVEGF165–transfected HSM (PEI-phVEGF165 myoblast) transplantation. A total of 48 rats received cyclosporine injection from 3 days before and until 4 weeks after cell transplantation. Echocardiography was performed to assess the heart function. Animals were sacrificed for molecular and histological studies on the heart tissue at 4 weeks after treatment. Based on optimized transfection conditions, transfected HSM expressed hVEGF165 for 18 days with >90% cell viability in vitro. Apoptotic index was reduced in group-2 and group-3 as compared with group-1. Blood vessel density (×400) by immunostaining for PECAM-1 in group-3 was significantly higher (P=0.043 for both) as compared with group-1 and group-2 at 4 weeks. Regional blood flow (ml/min/g) in the left ventricular anterior wall was higher in group-3 (P=0.043 for both) as compared with group-1 and group-2. Improved ejection fraction was achieved in group-3 (58.44±4.92%) as compared with group-1 (P=0.004).
Conclusion— PEI nanoparticle mediated hVEGF165 gene transfer into HSM is feasible and safe. It may serve as a novel and efficient alternative for angiomyogenesis in cardiac repair.
Heart cell therapy using skeletal myoblast (SkM) transplantation limits postinfarction remodeling and improves global myocardial function.1–3 More recently, SkM transplantation is being combined with therapeutic angiogenesis by genetic modification of the donor cells to overexpress genes encoding for single or multiple growth factors.4,5 Though this combined therapeutic strategy holds great promise for the treatment of ischemic heart disease, the approach of transgene delivery into SkM needs to be optimized. Various viral vectors have demonstrated high transduction efficiency of therapeutic genes into SkM.4–7 Nevertheless, their use has demerits including immunogenicity and oncogenic potential which severely hinder their clinical application.8 Human clinical trials have shown that viral vector-based delivery of genes caused inflammatory reactions, formation of antiadenoviral antibody, transient fever, and increase of liver transaminase.9–11 Nonviral vector gene delivery approach provides a safer alternative to overcome these untoward effects of viral vectors. Use of plasmid DNA either alone or complexed with cationic liposomes/polymers are being assessed with encouraging results.12,13 Similarly, the use of polymer based nanoparticles confers several advantages including ease of preparation, purification and chemical modification as well as their enormous stability.13,14 PEI has been widely used for nonviral transfection of cells.14 PEI is cationic in nature and has strong DNA compaction capacity, effective DNA protection and with an intrinsic endosomolytic activity.14 All these properties of PEI contribute to its transfection efficacy.
The present study has been carried out to design PEI (molecular weight 25 kDa: PEI-25) nanoparticles and optimization of conditions for transfection of hVEGF165 gene into HSM with minimum cytotoxic effects. The genetically modified HSM were later used for heart cell therapy in an experimental rat heart model of AMI. We posit that the use of PEI nanoparticle for hVEGF165 transfection of HSM is a safe and efficient approach for repair of the infarcted heart. We anticipate that nanoparticle gene transfection will envision a new approach for gene therapy in cardiovascular research.
PEI-25 Complexation With Plasmid DNA
Plasmids carrying enhanced green fluorescence protein (pEGFP) and phVEGF165 were kindly provided by Dr Ratha Mahendran and A/Prof Ruowen Ge, National University of Singapore (NUS), respectively. Plasmid and PEI-25 solution (10 mmol/L) were diluted in 50 μL of 150 mmol/L NaCl separately. To determine optimum ratio between PEI and DNA, 15 to 24 equivalents of PEI nitrogen per DNA phosphate (N/P) were mixed using the following equation:
N/P= (V×10 mmol/L)/(QDNA×3); where V=volume (μL) of 10 mmol/L PEI and QDNA=quantity of DNA (μg) used per 1×105 HSM.
PEI-DNA complex was developed by mixing the respective saline solution containing DNA or PEI, and the mixture was vortexed gently followed by sedimentation for 10 minutes. PEI-DNA mixture was then used for transfection of HSM for 24 hours at 37°C.
Loading Efficiency of PEI Nanoparticles
The amount of encapsulated DNA in the PEI-25 nanoparticles was measured as the difference between the amount of plasmid DNA added to the nanoparticles and nonentrapped DNA. After complexation, the nanoparticle suspension was centrifuged for 15 minutes at 8×103rpm and the supernatant was checked for the unbound DNA concentration with Nanodrop ND-1000 Spectrophotometer.
Scanning Electron Microscopy
Scanning electron microscopy was used to image the shape and size of the PEI-DNA nanoparticles. The buffer containing nanoparticles was spread on specimen stub (Agar Aid Scientific) and air-dried. The sample was coated with gold with a current at 15 mA for 30 seconds using SCD-005 Sputter Coater (Bal-Tec). Samples were visualized using XL30 FEG Philips electron microscopy (FEI).
Particle Size and Zeta Potential Analysis
The size distribution and zeta potential of the PEI-DNA nanoparticles were determined with a Zetasizer Nano ZS-machine (Malvern Instruments) equipped with a 4mW, 633 nm Ne-He laser at 25°C and a fixed scattering angle of 90o.
Protection and Release Assay of DNA
For determination of polyplex resistance to nuclease digestion, 50 μL DNase buffer (10×; Promega) was mixed with 450 μL corresponding solution containing 10 μg naked plasmid DNA or PEI-DNA. After withdrawing 100 μL for the time-0, 2 U DNase-I (Promega) was added into the mixture and incubated at 37°C. Samples (100 μL) were collected at 30-minute time intervals until 2 hours after incubation. To inactivate DNase-I, all samples were treated with 10 μL of EDTA (20 mmol/L) for 10 minutes at 65°C. Finally, 10 μL heparin (4 i.u/μL) was added to each sample and incubated at 65°C for 24 hours to facilitate dissociation of DNA from nanoparticles. The samples were electrophoresed using 1% agarose gel for analysis on Gel Document System (Biorad)
Culture of HSM
HSM were purchased from Bioheart Inc. The cells were cultured and propagated in laminin coated 225 mm2 tissue culture flasks using Super-medium (Cell Transplants Singapore Pte Ltd). The purity and uniformity of HSM preparation was determined as described earlier.4
Optimization of PEI-pEGFP Transfection Into HSM
PEI-pEGFP complexes were formed with N/P equivalents from 15 to 24. HSM at a density of 1×105/well in 12-well plate were incubated with PEI-pEGFP nanoparticles. After 24 hours, transfection medium was replaced with fresh medium.
Transfection efficiency and cell viability were determined using Coulter flow cytometer (FACS; Epics Elite Esp) as described.15 Nontransfected and PEI-25 without DNA treated HSM were used for baseline setting of auto-fluorescence. Data were analyzed using WinMDI version 2.8 with gating at 1%.
In Vitro HSM Transfection With PEI-phVEGF165
HSM were transfected with PEI-phVEGF165 using the optimized transfection conditions based on FACS results. The gene transfection and expression efficiencies were analyzed by immunostaining, RT-PCR, and enzyme-linked immunosorbent assay (ELISA).
Immunostaining of HSM
HSM transfected with PEI-phVEGF165 were grown on chamber slides for 24 hours and immunostained for hVEGF165 expression as described earlier.4 The immunostained cells were visualized using Olympus BX41 (Olympus) equipped for epifluorescence microscopy and images were recorded using a digital camera with MagnaFire 2.1 software.
Quantitative RT-PCR Analysis for hVEGF165
Nontransfected, phVEGF165, and PEI-phVEGF165 transfected HSM samples on 1, 4, 8, and 18 days after transfection were collected to quantify the hVEGF165 gene expression. The following primers were used for hVEGF165 expression (303bp): sense ATG AAC TTT CTG CTG TCT TGG, anti-sense GTT GGA CTC CTC AGT GGG C; 18S was used as internal control and purchased from Ambion Inc (Catalog #1718). The isolation of total RNA and cDNA synthesis was carried out as described earlier.4 The QPCR thermal cycling program for 40 cycles was: 1 cycle of enzyme activation at 95°C for 15 minutes, denaturation at 95°C for 30 seconds, annealing at 58°C for 30 seconds and extension at 72°C for 30 seconds.
ELISA for hVEGF165
Transfected HSM (2×105) were grown in vitro and the cell supernatant samples were collected at regular time intervals of 2 days (from day-0 until day-18). For quantification of hVEGF165 expression, hVEGF specific Sandwich ELISA kit (Chemicon Inc) was used according to supplier’s instructions.
One day before cell transplantation, HSM were labeled with 4, 6-diamidino-2-phenylindole (DAPI) (Sigma, USA) 12 hours at 37°C and 5% CO2 incubator.
Experimental Animal Model and Cell Transplantation
All animals received human care in compliance with the Guide for the Care and Use of Laboratory Animals, NIH, USA and NUS. All animals were maintained by Animal Holding Unit of NUS.
Experimental AMI was induced in 48 young female Wistar rats (≈200g, 10 weeks of age) by permanent ligation of left anterior coronary artery ligation.16 Ten minutes later, DMEM without HSM (group-1) or containing 1×106 nontransfected HSM (group-2) or PEI-phVEGF165–transfected HSM (group-3) were intramyocardially injected into the infarct and periinfarct regions. All animals received cyclosporine (5 mg/kg/d) starting 3 days before and until 4 weeks after treatment. Sixteen animals were used for each group. Three animals per group were randomly sacrificed at 1 day and 1 week respectively. The remaining 10 animals in each group were sacrificed at 4 weeks after treatment.
Regional Blood Flow Studies
Rats harvested at 4 weeks (n=5 animal each group) after treatment received fluorescent microsphere injection to assess regional blood flow of left ventricle (LV). Under anesthesia, right or left femoral artery was isolated and prepared for blood sampling. One-milliliter solution containing 4×104 Fluospheres yellow-green polystyrene microspheres (Molecular Probes) was directly injected into LV while 4 mL blood sample was drawn. The microsphere from LV anterior wall and blood sample was recovered and regional blood flow was measured.4
In Situ Cell Death Assay
One day after cell transplantation, rat hearts (n=3 animal each group) were harvested to detect apoptotic cells in heart using In situ Cell death Detection Kit (Roche) as per supplier’s instruction. For apoptotic index, tissue sections were counter-stained with propidium iodide (PI) after TUNEL. Total number of cell nuclei and apoptotic nuclei were counted in four fields (×400) per slide, and the apoptotic index was calculated as the percentage of TUNEL+ apoptotic nuclei to total nuclei per field.
Tissue sections from hearts explanted at 1 week after cell transplantation were immunostained for hVEGF165 expression.4 The heart tissue harvested at 4 weeks after cell transplantation were immunostained for human skeletal myosin heavy chain (MY32, Sigma). Blood vessel density was measured at ×400 magnification in 8 microscopic fields/heart (n=5 animal each group) after double fluorescent immunostaining for PECAM-1 (Santa Cruz Biotechnology Inc) and smooth muscle actin (SMA; Sigma).
Heart Function Assessment
Echocardiography was performed by an investigator blinded to the therapeutic intervention on the animals using Aloka ultrasound machine (Aloka). From M-mode echocardiograms, measurements were obtained for LV-anterior wall thickness at end-diastole (LVAWTed), end-systole (LVAWTes), LV-internal diameters at end-diastole (LVIDed), and end-systole (LVIDes). LV-ejection fraction (LVEF) and fractional shortening (LVFS) were calculated as: LVEF=1−(LVIDes/LVIDed)2 and LVFS=1−(LVIDes/LVIDed). LV-anterior wall thickening percentage (LVAWTP) was calculated as (LVAWTes-LVAWTed)/LVAWTed.
All statistical analyses were performed using SPSS (version 10.0). Apoptotic index, blood vessel density, and regional blood flow data were presented as median (25% quartile, 75% quatile) and analyzed by Friedman and Wilcoxon nonparametric methods to test any difference between the groups. The rest data were presented as mean±SEM. Heart function between groups was analyzed by the method of analysis of variance (ANOVA) using Bonferroni test. All tests were performed with a significance level of 5%.
Statement of Responsibility
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Characterization of PEI-DNA Particles
The ultra structure of PEI-DNA complexes is shown in Figure 1A. The nanoparticle size and zeta potential showed concentration dependence (Figure 1B and 1C) and at higher N/P ratio, larger nanoparticle size and higher zeta potential were achieved. Loading efficiency of PEI-DNA nanoparticles was about 99% for all PEI-DNA ratios. PEI protected the encapsulated DNA from degradation for over 120 minutes as compared with the unprotected naked DNA which fully degraded after 30 minutes by DNase-I (Figure 1D).
Transfection of HSM With PEI-pEGFP
Maximum transfection of pEGFP (9.18±2.37%) into HSM was achieved at 21:1 N/P ratio which was associated with low cell toxicity (cell viability=92.48±0.68%) (Figure 2A). Increase in DNA from 2 μg to 5 μg/1×105 cells with N/P ratio at 21 also improved transfection efficiency (10.93±3.1%) at 3 μg DNA (Figure 2B). The optimum transfection conditions using PEI-25 were N/P ratio of 21 using 3 μg DNA/ 1×105 HSM and were used throughout the study (Figure 2C through 2E).
Characterization of PEI-phVEGF165–Transfected HSM
Immunostaining revealed hVEGF165 expressing HSMin vitro (Figure 3A and 3B). Transfection of HSM with phVEGF165 resulted in poor transfection efficiency (1.32±0.06 times as compared nontransfected HSM), whereas the gene expression of hVEGF165 from PEI-hVEFG165 transfected HSM increased 8.37±0.21 times at day-2, 7.72±0.27 times at day-4 and 2.84±0.22 times at day-8, whereas it was 1.69±0.04 times at day–18 (Figure 3C). ELISA showed that the transfected HSM secreted hVEGF165 for 18 days of observation (4.47±0.38 ng/mL) with peak level expression at day-2 after transfection (20.2±1.51 ng/mL; Figure 3D).
DAPI expressing HSM survived in the rat heart until 4 weeks after cell transplantation (Figure 4A and 4B). The surviving HSM developed into multinucleated myotubes which stained positively for human skeletal myosin heavy chain (Figure 4C through 4E). The PEI-phVEGF165 myoblast also expressed hVEGF165 at 1 week after transplantation (Figure 4F through 4H).
Apoptosis in infarcted area was reduced where HSM were transplanted (Figure 5C and 5D) as compared with DMEM injected group-1 (Figure 5A and 5B). This effect was further enhanced by hVEGF165 transfected HSM (Figure 5E and 5F). Though no significant difference (P=0.109) was achieved (because of the small animal number at 1 day after treatment), the percentage of TUNEL+ cells in the scar zone was highest : 68.78% (62.16%, 77.32%) in group-1, whereas they were 39.75% (33.58%, 41.66%) in group-2 and 25.73% (24.41%, 28.33%) in group-3 (Figure 5G).
Evidence for Angiogenesis
Blood vessel density based on PECAM-1 immunostaining (at 400× magnification) was highest in group-3: 21.40 (18.69, 23.45; P=0.043) as compared with group-1: 9 (8.38, 10.75) and group-2 11.71 (10.47, 13.03) at 4 weeks after treatment (Figure 6A through 6I). The blood vessel density based on SMA immunostaining also was highest in group-3 15.3 (13.9, 18.13; P=0.043) as compared with group-1 and group-2. Dual immunostaining for PECAM-1 and SMA showed that percentage of the mature blood vessels in group-3 was 75.86% (72.26%, 77.29%) which was similar to those of group-1: 73.68% (62.5%, 77.5%; P=0.5) and group-2: 75.45% (69.32%, 78.28%; P=0.893; Figure 6L).
Regional Blood Flow
The LV-anterior wall blood flow (ml/min/g) was significantly reduced in group-1: 0.74 (0.5, 0.87) as compared with group-2 1.32 (1.13, 1.64; P=0.043,) and group-3 2.3 (1.53, 2.84; P=0.043; Figure 7A).
Heart Function Studies
Echocardiography at 4 weeks showed that EF and FS in group-3 were 58.44±4.92% and 36.34±3.85%, respectively (Figure 7B and 7C). These were higher as compared with group-1 (41.67±4.77%, P=0.004 and 24.09±3.13%, P=0.004). Though EF (50.53±4.73%) and FS (30.2±3.26%) were improved in group-2, no significant difference was achieved as compared with group-1 (P=0.432 and P=0.465).
Though the LVIDed and LVIDes in all 3 animal groups increased, there was limited incremental tendency for LVIDed and LVIDes in group-2 (6.63±0.26 mm, P=1; 4.64±0.32 mm, P=0.973) and group-3 (6.48±0.23 mm, P=1; 4.04±0.26 mm, P=0.043) as compared with group-1 (6.74±0.41 mm; 5.18±0.50 mm; Figure 7D and 7E). LVAWTed and LVAWTes were best maintained in group-3 (1.76±0.27 mm; 2.72±0.37 mm; Figure 7F and 7G), followed by group-2 (1.69±0.44 mm and 2.41±0.75 mm) and group-1 which showed thinning of LVAWTed (1.64±0.33 mm) and LVAWTes (2.23±0.53 mm). LV-anterior wall thickening percentage was highest in group-3 (56±5.89%) followed by group-2 (43.06±10.6%) and group-1 was the lowest (35.61±6.49%) (Figure 7H).
This study shows that transfection of HSM with PEI-phVEGF165 nanoparticles improves their reparability of the injured heart. High molecular weight PEI (800 kDa) has been used for gene transfer and forms compact and stable PEI-DNA complexes.17 However, this was associated with remarkably reduced cell viability. This effect can be moderated with low molecular mass PEI (5–48 kDa), whereas higher polymer concentration (increased N/P ratio) will be required to achieve comparable efficacy.18–19 N/P ratio and zeta potential dramatically influence the efficiency of PEI mediated gene delivery.14,20 Previous studies have shown that 5 to 6 NH2 nitrogen moieties of PEI are protonated at physiological pH and only these positively charged NH2 groups ionically interact with the negatively charged DNA.20 Our data are consistent with these observations.
Besides other factors, transfection efficiency is influenced by particle size which is mainly moderated by the molecular weight of PEI and N/P ratio.21 PEI-DNA complexes formed between 40 to 50 nm resulted in poorer transfection efficiency as compared with larger particle (200 to 300 nm). We observed that larger nanoparticle size gave higher transfection efficiency. However, for balance between cell viability and transfection efficiency, we opted for N/P at 21 as optimum transfection condition. We observed that incubation of HSM with PEI-DNA in the presence of low serum (2%) gave poor cell survival. Hence, we opted for Super-medium that contained 10% fetal bovine serum as gene transfection medium to promote cell viability during transfection reaction.
Although 11% of HSM were transfected with PEI-hVEGF165, peak level expression (20.2 ng/mL) was observed on day-2 and a meager 4.47 ng/mL on day-18 after transfection. A comparison of the efficiency of PEI-25 with (n-[1-(2,3-Dioleoyloxy) propyl]-N,N,N-trimethylammonium methylsulfate) DOTAP for β-galactosidase gene delivery showed that when PEI-25 gave 48.3±7% A549 cells transfected and expressed β-galactosidase protein (1.07±0.4 μg/μL). Contrarily, DOTAP gave 49.3±2.7% A549 cells transfected and expressed β-galactosidase only up to 0.47±0.3 μg/μL.22 A previous study has reported that >95% transfected rat SkM expressed only 2.78±0.2 ng/mL hVEGF165 protein and was sufficient to initiate neovascularization.23 The initial hVEGF165 protein level in our study was 20.2 ng/mL from PEI-phVEGF165 transfected HSM and 12.7 ng/mL during the first four days after transfection.
Our previous study using adenoviral vector carrying hVEGF165 to transduce hSkM had resulted in more than 95% transduction efficiency.4,5 The peak level of VEGF protein (37±3 ng/mL) was reached at day-8 after transduction. Though the protein level in current study was lower than that, we believed that it was sufficient to initiate and maintain neovascularization. This was supported by the data of blood vessel density and LV regional wall blood flow in our study.
Consistent with our previous studies, transplantation of PEI-phVEGF165 HSM had best improved LV systolic function and prevented LV remodeling. The surviving HSM not only expressed hVEGF165 to increase neovascularization, but also expressed skeletal myosin heavy chain in the rat heart. It was interesting that not only PEI-phVEGF myoblasts transplantation, but also skeletal myoblasts transplantation reduced apoptotic cells in cell transplanted area (Figure 5). This effect may be related with the paracrine factors released from HSM, including VEGF, hepatocyte growth factor, and platelet-derived growth factor.4,24–25 Moreover, overexpression of hVEGF165 from transfected HSM was cytoprotective for cardiomyocytes in the ischemic myocardium in the acute stage, whereas it stimulated neovascularization and increased blood flow at later stage of infarction.
In summary, our study highlights the feasibility, safety, and efficacy of PEI-phVEGF165–transfected HSM transplantation for cardiac repair. Reduced cardiomyocyte apoptosis, improved wall thickness, increased neovascularization, and regional blood flow of the infarcted myocardium together resulted in improved heart function. The PEI-25 based nanoparticle gene delivery approach may have clinical relevance and open a new concept for nonviral angiogenic gene delivery for the treatment of ischemic heart disease.
We appreciate Dr YuTing Zhang’s help and suggestions in doing statistical analysis.
Sources of Funding
The project was funded by Singapore National Medical Research Council (NMRC) grant R-364-000-021-213.
Presented at the American Heart Association Scientific Sessions, Chicago, Ill, November 12–15, 2006.
Ye L, Haider KhH, Sim EKW. Adult Stem Cells for Cardiac Repair: A Choice between skeletal myoblasts and bone marrow stem cells. Exp Biol Med. 2006; 231: 8–19.
Su H, Lu R, Kan YW. Adeno-associated viral vector-mediated vascular endothelial growth factor gene transfer induces neovascular formation in ischemic heart. Proc Natl Acad Sci U S A. 2000; 97: 13801–13806.
Lee RJ, Springer ML, Blanco-Bose WE, Shaw R, Ursell PC, Blau HM. VEGF gene delivery to myocardium: deleterious effects of unregulated expression. Circulation. 2000; 102: 898–901.
Davis ME. Non-viral gene delivery systems. Biotechnol. 2002; 13: 128–131.
Jiang S, Haider HKh, Idris NM, Salim A, Ashraf M. Supportive interaction between cell survival signaling and angiocompetent factors enhances donor cell survival and promotes angiomyogenesis for cardiac repair. Circ Res. 2006; 99: 776–784.
Kunath K, von Harpe A, Fischer D, Petersen H, Bickel U, Voigt K, Kissel T. Low-molecular-weight polyethylenimine as a non-viral vector for DNA delivery: comparison of physicochemical properties, transfection efficiency and in vivo distribution with high-molecular-weight polyethylenimine. J Control Release. 2003; 89: 113–125.
Tatsumi R, Hattori A, Ikeuchi Y, Anderson JE, Allen RE. Release of hepatocyte growth factor from mechanically stretched skeletal muscle satellite cells and role of pH and nitric oxide. Mol Biol Cell. 2002; 13: 2909–2918.
Sejersen T, Betsholtz C, Sjolund M, Heldin CH, Westermark B, Thyberg J. Rat skeletal myoblasts and arterial smooth muscle cells express the gene for the A chain but not the gene for the B chain (c-sis) of platelet-derived growth factor (PDGF) and produce a PDGF-like protein. Proc Natl Acad Sci U S A. 1986; 83: 6844–6848.