Vascular Endothelial Growth Factor Expressing Mesenchymal Stem Cells Improves Cardiac Function In Chronic Myocardial Infarction In Pigs

Transplantation of mesenchymal stem cells (MSCs) for myocardial reconstruction has shown promise in both animal models and human phase 1 clinical studies.1,2 Vascular endothelial growth factor (VEGF) is a strong therapeutic agent for treating ischaemia by inducing angiogenesis.3 The feasibility of ex vivo MSCs mediated gene transfer is documented.4,5 Matsumoto and colleagues6 have recently reported genetically engineered MSCs carrying VEGF165 delivery for revascularization in a model of acute myocardial infarction (MI). The promising data from our laboratory in both angiogenesis and MSCs transplantation in cunicular heart model of acute MI have prompted us to attempt the combined and simultaneous application of the two strategies. In the present study, we compared the effects of transplantation of MSCs, transfected with or without the VEGF gene, on cardiac function in pigs with chronic myocardial infarction. METHODS Animals The investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). For this study, thirty healthy, adult domestic pigs of either sex, weighing between 50 kg and 65 kg each (Research Centre of Laboratory Animals, Fourth Military Medical University, China) were used. MSCs isolation and culture Both donors and recipients were domestic pigs. Bone marrow was taken from breastbone. Bone marrow mononuclear cells were isolated using density gradient centrifugation (Nyco Prep 1.077 Animal; AXIS-SHIELD PoC, AS, Oslo, Norway). Then, the mononuclear cells were cultured in low glucose Dulbecco's Modified Eagle Medium containing 10% foetal bovine serum for MSC outgrowth. The nonadherent cells were removed by change of medium at 48 hours and every 4 days thereafter. VEGF gene transfer and preparation of cells MSCs were transfected with pcDNA3.1/h VEGF165 or empty vector using LipofectAMINE™ (Gibco Co, USA), and VEGF expression in cell culture was analysed as previously described.6 Two hours before cell transplantation, MSCs transfected with pcDNA3.1/h VEGF165 or empty vector were harvested and washed three to four times with heparinized saline. The MSC suspension was mixed with heparin, then filtered and prepared for implantation. The concentration of the final cell suspension was greater than 8×107 cells/ml. Preparation of animal models The pigs underwent transcatheter embolization of the distal portion of the bifurcation between the left anterior descending (LAD) coronary artery and the diagonal branch using a gelatine sponge to produce anteroapical MI.7 Each animal was randomly assigned into one of three groups: a) Combo group (n = 10), intramyocardial transplant of MSCs expressing VEGF; b) MSC group (n = 10), intramyocardial transplant of unmodified MSCs; and c) control group (n = 10), saline solution. Left ventricular (LV) electromechanical mapping and cells transplantation Four weeks after MI, NOGA nonfluoroscopic LV electromechanical mapping was performed to guide injections to the border zone surrounding the infarct. The NOGA system (Cordis, USA) of catheter based mapping and navigation has been previously described in detail.8,9 Briefly, it consists of a mapping catheter modified to integrate a retractable 27G needle for intramyocardial injection. The catheter "dead space" was 0.1 ml and flushed with 0.1 ml of cell suspension before endovascular insertion. The exact catheter tip location, orientation and the injection sites were indicated in real time on the LV map, and local electrical and location signal were traced to assure catheter stability and optimal endocardial contact. The needle was extended from 5 mm to 6 mm into the LV myocardium. Ischaemic myocardium was defined as a zone with unipolar voltage higher than an automatically determined cutoff and linear local shortening of <5%. This definition was consistent for all examinations throughout the study. Immediately after the ischaemic region was identified by NOGA mapping, 1 ml cell suspension or saline solution was injected into 10 sites within the ischaemic myocardium (100 μl to each site, 8×107 cells/ml) using the NOGA injection catheter. Physiological assessment of LV function and ischaemia Transthoracic echocardiography selective left coronary angiography and NOGA LV electromechanical mapping were performed before and 4 weeks after the injection. Echocardiographic left ventricular ejection fraction (LVEF), left ventricular fractional shortening (LVFS), and change in thickness of the LV wall (ΔT%) were quantified. Collateral flow to the LAD region was graded angiographically in a blinded fashion with the Rentrop scoring system.10 The area of ischaemia was quantified by NOGA mapping as previously described. Statistical analysis All values were expressed as mean ± standard deviation (SD). Student's paired t test was used to compare data before and after the treatment. ANOVA was performed to compare data in the three groups. Student Neuman Keuls test was performed to compare data among the three groups. A P<0.05 was considered statistically significant. RESULTS Mortality Eight pigs died during the experiment (2 in Combo, 2 in MSC and 4 in control group). Characterization of MSCsin vitro MSCs isolated by density gradient centrifugation were cultured for 5 to 7 days until individual colonies with fibroblast like morphology formed. The cells were spindle shaped, tightly attached to the culture dish and proliferated in the culture medium. Flow cytometric analyses showed that the third passage of MSCs was CD44-positive (95%) but CD34negative. VEGF expression of MSCsin vitro To determine whether cultured MSCs transfected with VEGF were able to release soluble VEGF protein, we collected conditioned medium from the MSCs and assayed for VEGF by ELISA. The VEGF levels secreted in the culture medium for the Combo group were twenty-fold higher than those in the control group [(2776.0 ± 200.7) vs (122.4 ± 10.9) pg/ml, P<0.01]. Transplanted VEGF-expressing MSCs attenuate ischaemic area Four weeks after MI, the ischaemic area in each pig's heart was determined using the NOGA mapping system. Before cell transplantation, the ischaemic areas (%) were (31.8 ± 7.3)% in the Combo group, (33.6±7.0)% in the MSC group, and (28.2 ± 5.7)% in the control group. Four weeks after treatment, the ischaemic area was significantly reduced in the Combo group [(-35.8 ± 17.9)%] compared with the other two groups [MSC, (—25.7 ±7.0) % and control, (3.1 ± 16.8) %; P<0.05 vs MSC, P<0.05 vs control] (Fig.).Fig.: Representative recordings of NOGA electromechanical mapping of the left ventricle immediately before and 4 weeks after cell transplantation in pigs with chronic myocardial ischaemia. The black dots in the pretreatment map indicate sites of gene transfer (A). The red area on the pretreatment linear local shortening map (top right) indicates an area of decreased wall motion in the lateral wall of the left ventricle. Four weeks after cell transplantation, this area of ischaemia improves significantly in a representative case from the Combo group (B) and moderately in a case from the MSC group (C), whereas no improvement is observed in a case from the control group (D).Transplanted VEGF-expressing MSCs enhance therapeutic neovascularization Collaterals were graded with Rentrop's classification (cardiac ischaemic score range 0 — 4). Rentrop scores of collateral development in the three groups were as follows: a) in the Combo group, Rentrop grade=1, 2 pigs; Rentrop grade=2, 5 pigs; Rentrop grade=3, 1 pig; b) in the MSC group, Rentrop grade=0, 1 pig; Rentrop grade=1, 4 pigs; Rentrop grade=2, 3 pigs; and c) in the control group, Rentrop grade=0, 2 pigs; Rentrop grade=1, 3 pigs; Rentrop grade=2, 1 pig. The Rentrop score after treatment was significantly higher in the Combo group than in the other two groups (P<0.05). The Rentrop score in the MSC group was similar to that in the control group. These data indicated that there was evidence of improved collateral formation in the Combo group. Transplanted VEGF-expressing MSCs augment LV function Echocardiographic LVEF, LVFS and ΔT% before treatment were similar in all groups. The improvement in LVEF, LVFS and ΔT% after treatment was significantly greater in the Combo and MSC groups than in the control group. However, the Combo group had a significantly higher LVEF than the MSC group (P<0.05, Table).Table: LVEF, LVFS, and ΔT% 4 weeks after treatment (mean ± SD, %)DISCUSSION Our results showed that transplanted VEGF expressing MSCs significantly improved chronic myocardial ischaemia as documented by NOGA mapping, improved collateral formation and resulted in a favourable trend in LV functional improvement compared with MSCs transplantation alone. In the present study, we used a percutaneous approach using gelatine sponge for controlled induction of anteroapical occlusive MI leading to chronic myocardial infarction in pigs. The method has low mortality and complication rates, precise control of the location of occlusion, is easily reproducible,11 and has a pathophysiology similar to human coronary artery disease. Percutaneous myocardial cell transplantation offers several advantages compared to the surgical approach. This strategy necessitates an easy to use and adequate device for in situ cell delivery that does not alter cell characteristics and allows precise targeting of definite myocardial areas. The small number of animals does not allow us to insure that the procedure would be successful in every case. However, our results clearly showed the feasibility of cell grafting by transendocardial route. The optimal graft cell number is that required to achieve the maximum attenuation of adverse remodelling or actual improvement in cardiac function after cell transplantation. In this study, we showed that the injection of 8 million MSCs results in reduced infarct size and improved function after MI. Shake et al12 demonstrated that a direct intramuscular injection of 60 million MSCs improved the function of porcine MI model. Davani et al13 showed that 1 million MSCs improve cardiac function in Lewis rats after MI. However, the overall functional result may be affected by various factors, such as grafting cell survival capacity and the condition of host myocardium. The site of introduction of the grafting cells is also an important consideration. In the present study, we transplanted MSCs to the border area of infarcts for salvaging the stunned or hibernating myocardium in this area. Investigating the effects of varying injection sites would provide new and interesting data on cell transplantation to the heart. The beneficial effects of ex vivo MSC mediated VEGF165 delivery are three-fold. First, the engrafted MSCs survive in the host myocardium and do repair myocardial damage. Second, VEGF transfected MSCs exert a cardioprotective effect through improved blood flow to the hibernating tissue at the periphery of the infarct. Finally, this approach would reduce the host inflammatory immune response that is a potential undesired effect of direct adenoviral vector administration, especially when autologous cells used as carriers of the exogenous angiogenic genes. Acknowledgement: We appreciate the excellent technical assistance of LI Li, LI Wei and ZHANG Rong-qing.