Abstract
Objective
Our objective was to determine whether autologous endothelial progenitor cells (EPCs) delivered into the pericardial space will migrate to and incorporate into ischemic myocardium in a porcine model.
Background
Use of EPCs to enhance neovascularization and preserve myocardial function in ischemic tissue is undergoing intense scrutiny as a potential therapy. Delivery into the pericardial sac may overcome some of the limitations of currently employed cell delivery techniques.
Methods
EPCs were immunopurified from peripheral blood of Yorkshire pigs by selecting for the CD31 surface antigen, and adherent cells were cultured for 3–5 days. After myocardial ischemia was induced in the left anterior descending (LAD) artery, either autologous DiI (1,1′-dioctadecyl-1-3,3,3′,3′-tetramethylindocarbocyanine perchlorate)-labeled EPCs ( n =10) or serum-free medium (SFM; n =8) was delivered into the pericardial space using a percutaneous transatrial approach. Animals were sacrificed on Day 7 or 21. Echocardiography was performed at baseline, during ischemia, and on Day 7 in six SFM group animals and six EPC group animals.
Results
On Day 7, EPCs were identified in the left ventricular (LV) anterior wall or anterior septum in all six EPC-treated animals (cell density of 626±122/mm 2 ). On Day 21, EPCs were identified in the LV anterior wall or anterior septum in three of four EPC-treated animals (cell density of 267±167/mm 2 ). These cells showed dual staining for DiI and Bandeiraea simplicifolia lectin I (a marker of both native and exogenous endothelial cells). At the Day 7 follow-up, echocardiography demonstrated that fractional shortening in the EPC-treated group was 30.6±3.4, compared with 22.6±2.8 in SFM controls ( P =.05).
Conclusions
EPCs can migrate from the pericardial space to incorporate exclusively into areas of ischemic myocardium and may have favorable effects on LV function.
1
Background
Use of endothelial progenitor cells (EPCs) for the treatment of ischemic heart disease has emerged as an area of intense investigation. These cells are known to migrate and incorporate into ischemic tissue and are associated with neovascularization and preservation of left ventricular (LV) function in animal models . A number of delivery methods have been utilized, such as intravenous , intracoronary, left ventricle intracavitary injection , direct myocardial injection , and coronary sinus delivery . Some of these approaches however rely on arterial perfusion for delivery of the cells, an approach likely limited by the poor perfusion of the ischemic tissue, while intramyocardial injection of stem cells has been associated with ventricular arrhythmias thought to be related to local myocardial heterogeneity generated by clusters of injected cells . Thus, the optimal and most efficient route of delivery remains under investigation.
We propose that intrapericardial delivery of autologous EPCs is a viable method for cell therapies targeting acute ischemic myocardium. The intrapericardial approach has been previously utilized for cardiac drug delivery and takes advantage of the pericardial space as a depot for targeted delivery, possibly increasing exposure time of EPCs to ischemic tissue . These principles were clearly demonstrated in the study by Waxman et al. , in which intrapericardial infusion of nitroglycerin achieved persistent coronary dilatation without systemic hypotension.
Using a swine model of balloon-induced myocardial ischemia, we tested the hypothesis that ex vivo cultured EPCs delivered into the pericardial space will incorporate and preferentially localize into areas of ischemic myocardium. Furthermore, we assessed whether this pericardial delivery of EPCs will preserve LV function following acute ischemia.
2
Materials and methods
The protocol was approved by the Tufts–New England Medical Center Animal Care and Use Committee and conformed to National Institutes of Health standards.
2.1
Cell isolation
Yorkshire pigs weighing between 28 and 35 kg were sedated with 2 ml of a KTX cocktail (50 mg/ml of ketamine, 50 mg/ml of xylazine, and 100 mg/ml of telazol), and 150 ml of peripheral blood was obtained from the femoral vein. Total peripheral blood mononuclear cells (MNCs) were isolated by density-gradient centrifugation. The EPC-enriched fraction was isolated from the total MNCs using the MACS magnetic bead system for selection of the CD31 (Miltenyi Biotec) surface antigen. The CD31 + MNCs were then resuspended in endothelial cell basal medium-2 (EBM-2, Clonetics) supplemented with EGM-2-MV SingleQuots (Clonetics) containing vascular endothelial growth factor (VEGF), basic fibroblast growth factor, insulin-like growth factor, epidermal growth factor, and 5% fetal bovine serum. The cells were incubated overnight on fibronectin-coated dishes, and the non-adherent fraction (the macrophage-rich portion) was removed after 24 h. The adherent population was maintained in the growth medium for 3–5 days. Prior to injection into the same pig, EPCs were labeled with fluorescent carbocyanine DiI (1,1′-dioctadecyl-1-3,3,3′,3′-tetramethylindocarbocyanine perchlorate)-labeled acetylated low-density lipoprotein (acLDL) (Molecular Probes). The cells were then resuspended in serum-free medium (SFM) and counted prior to injection into the animal.
2.2
Porcine ischemia model
The animals were sedated with KTX, intubated endotracheally, and anesthetized with isofluorane. Oxygenation was monitored with a pulse oximeter, and heart rate and rhythm were monitored from a surface electrocardiogram. Femoral arterial and venous access was established with standard percutaneous technique and 6-Fr sheaths inserted. A 6-Fr guiding catheter was advanced under fluoroscopy guidance to cannulate the left main coronary artery. All animals were given 325 mg of aspirin prior to the procedure and a heparin bolus after vascular access to maintain activated clotting times above 200 s. Myocardial ischemia was induced by inflating a coronary dilatation balloon in the mid portion of the LAD artery between the first and second diagonal branches. A 30-min balloon inflation was preceded by two preconditioning inflations lasting 2 and 5 min each. Prior to the 30-min balloon inflation, a 0.5-mg/kg lidocaine i.v. bolus was administered. Ischemia was confirmed by electrocardiographic changes (ST elevation) accompanied by anterior wall motion abnormality by two-dimensional echocardiography.
2.3
Intrapericardial delivery of autologous EPCs
Following reperfusion, animals received either autologous DiI-labeled EPCs ( n =10, mean=1.9±0.4×10 6 cells) or SFM ( n =8) into the pericardial space with the use of a percutaneous transatrial approach previously described . Briefly, a 6-Fr multipurpose guide catheter was advanced through the venous sheath under fluoroscopic guidance and positioned in the right atrial appendage. A small perforation was made in the right atrial appendage with a 0.014-in. guide wire inside a 0.038-in. infusion wire. Both wires were then advanced into the pericardial space. The inner wire was removed, leaving the 0.038-in. wire as an infusion port. The position of the infusion wire inside the pericardial space was confirmed by fluoroscopy and aspiration of pericardial fluid. Through the infusion wire, EPCs suspended in approximately 5 ml of SFM or 5 ml of SFM alone were injected into the pericardial space. The infusion wire was then removed from the pericardial space, and the guide catheter was withdrawn and the venous and arterial sheaths were removed. Three additional animals were administered autologous DiI-labeled EPCs intrapericardially but did not undergo coronary ischemia.
2.4
Follow-up and histological assessment of EPC incorporation
Animals were sacrificed on Day 7 ( n =15; 6 from the SFM group, 6 from the EPC group with ischemia, and 3 from the EPC group without ischemia) or Day 21 ( n =6; 2 from the SFM group and 4 from the EPC group with ischemia) ( Fig. 1 ). All animals were preanesthetized with KTX and anesthetized with isoflurane (inhalation). Following a coronary angiogram and a transthoracic echocardiographic study (described below), the animals were euthanized with intravenous potassium chloride. The heart and intact pericardium were exposed through a median sternotomy. A small incision was made in the parietal pericardium, and pericardial fluid was aspirated. The hearts were then explanted and perfused with heparinized saline through the cannulated ascending aorta. Additionally, two of the EPC-treated animals and two of the SFM control animals were perfused with a Bandeiraea simplicifolia lectin I (BS-I; an endothelial cell marker) solution following the saline perfusion. After the parietal pericardium was removed, the heart was sectioned in 3-mm transverse slices from apex to base. Sections of the LV anterior septum and anterior wall as well as the right ventricle and inferoposterior wall were obtained from each transverse slice. Tissue samples were also obtained from the liver and lung. Specimens were then mounted in optimal cutting temperature compound (O.C.T., TissueTek) and snap-frozen in liquid nitrogen. Sections (5 μm) were made, and tissue was analyzed under fluorescent microscopy for DiI and BS-I if applicable. Fluorescent cell density was performed by quantifying the number of fluorescent cells per square millimeter in a high-power field (20×) using the Scion image software (v. 4.0.2). All samples were overread by an observer who was blinded to SFM versus EPC infusion.
2.5
Assessment of LV function
Two-dimensional transthoracic echocardiographic images were acquired in the parasternal long and short axes. Regional wall motion and LV fractional shortening were measured before ischemia (baseline), at peak ischemia, and at the time of animal sacrifice. The physicians performing the echocardiographic studies and analysis were blinded to treatment.
2.6
Immunocytochemical staining
EPCs were isolated as detailed above and grown on fibronectin (Sigma)-coated coverslips for a period of either 2 or 5 days. At the end of the allotted growth period, cells were fixed with cold methanol–acetone (1:1, −80°C) following staining with fluorescent carbocyanine DiI-labeled acLDL (Molecular Probes). Coverslips were then incubated with mouse anti-human von Willebrand factor (vWF; DAKO Cytomation), followed by incubation with fluorescein-isothiocyanate-conjugated goat anti-mouse secondary antibodies (BD Biosciences). Nuclear counterstaining was done using DAPI (4′,6-diamidino-2-phenylindole; Innogenex). Coverslips were mounted, and the number of DiI-positive cells and that of vWF-positive cells were counted as subpopulations of DAPI-positive cells.
2.7
Immunohistochemical staining for capillary density
Sections of tissue (5 μm) from five of each of the Day 7 EPC-treated and SFM control animals were snap frozen in O.C.T. compound, mounted on slides, and fixed with acetone, and then they were treated with 3% peroxide to block endogenous peroxidase activity. Tissue was then incubated with mouse anti-human vWF (DAKO Cytomation) and goat anti-mouse secondary conjugated with horseradish peroxidase (Santa Cruz). A Vectastain ABC kit was used in conjunction with DAB (3,3-diaminobenzidine; Vector Labs) to stain vWF-positive areas brown. Capillaries were counted based upon a cross-sectional lumen of <10 μm and a transverse luminal length of <50 μm by an individual blinded to treatment. Capillary density was recorded as the average number of cells per square millimeter. The ratio of anterior–anteroseptal wall to inferoposterior wall was used in statistical tests to compensate for variances in individual animals.
2.8
Analysis of VEGF and hepatocyte growth factor in pericardial fluid
Samples of pericardial fluid taken from each animal on the day of the procedure and at the Day 7 follow-up were analyzed using a commercially available enzyme-linked immunosorbent assay (R&D Systems). All samples were run in duplicate, and optical density measurements taken at 450 nm were compared with a standard curve.
2.9
Statistical analysis
All result values are stated as the mean±SEM. Statistical significance was evaluated using the paired t test for the analysis of fractional shortening and the unpaired t test for capillary density. The Wilcoxon ranked-sum test was used to analyze data from pericardial fluid enzyme-linked immunosorbent assay. A value of P <.05 was considered statistically significant.
2
Materials and methods
The protocol was approved by the Tufts–New England Medical Center Animal Care and Use Committee and conformed to National Institutes of Health standards.
2.1
Cell isolation
Yorkshire pigs weighing between 28 and 35 kg were sedated with 2 ml of a KTX cocktail (50 mg/ml of ketamine, 50 mg/ml of xylazine, and 100 mg/ml of telazol), and 150 ml of peripheral blood was obtained from the femoral vein. Total peripheral blood mononuclear cells (MNCs) were isolated by density-gradient centrifugation. The EPC-enriched fraction was isolated from the total MNCs using the MACS magnetic bead system for selection of the CD31 (Miltenyi Biotec) surface antigen. The CD31 + MNCs were then resuspended in endothelial cell basal medium-2 (EBM-2, Clonetics) supplemented with EGM-2-MV SingleQuots (Clonetics) containing vascular endothelial growth factor (VEGF), basic fibroblast growth factor, insulin-like growth factor, epidermal growth factor, and 5% fetal bovine serum. The cells were incubated overnight on fibronectin-coated dishes, and the non-adherent fraction (the macrophage-rich portion) was removed after 24 h. The adherent population was maintained in the growth medium for 3–5 days. Prior to injection into the same pig, EPCs were labeled with fluorescent carbocyanine DiI (1,1′-dioctadecyl-1-3,3,3′,3′-tetramethylindocarbocyanine perchlorate)-labeled acetylated low-density lipoprotein (acLDL) (Molecular Probes). The cells were then resuspended in serum-free medium (SFM) and counted prior to injection into the animal.
2.2
Porcine ischemia model
The animals were sedated with KTX, intubated endotracheally, and anesthetized with isofluorane. Oxygenation was monitored with a pulse oximeter, and heart rate and rhythm were monitored from a surface electrocardiogram. Femoral arterial and venous access was established with standard percutaneous technique and 6-Fr sheaths inserted. A 6-Fr guiding catheter was advanced under fluoroscopy guidance to cannulate the left main coronary artery. All animals were given 325 mg of aspirin prior to the procedure and a heparin bolus after vascular access to maintain activated clotting times above 200 s. Myocardial ischemia was induced by inflating a coronary dilatation balloon in the mid portion of the LAD artery between the first and second diagonal branches. A 30-min balloon inflation was preceded by two preconditioning inflations lasting 2 and 5 min each. Prior to the 30-min balloon inflation, a 0.5-mg/kg lidocaine i.v. bolus was administered. Ischemia was confirmed by electrocardiographic changes (ST elevation) accompanied by anterior wall motion abnormality by two-dimensional echocardiography.
2.3
Intrapericardial delivery of autologous EPCs
Following reperfusion, animals received either autologous DiI-labeled EPCs ( n =10, mean=1.9±0.4×10 6 cells) or SFM ( n =8) into the pericardial space with the use of a percutaneous transatrial approach previously described . Briefly, a 6-Fr multipurpose guide catheter was advanced through the venous sheath under fluoroscopic guidance and positioned in the right atrial appendage. A small perforation was made in the right atrial appendage with a 0.014-in. guide wire inside a 0.038-in. infusion wire. Both wires were then advanced into the pericardial space. The inner wire was removed, leaving the 0.038-in. wire as an infusion port. The position of the infusion wire inside the pericardial space was confirmed by fluoroscopy and aspiration of pericardial fluid. Through the infusion wire, EPCs suspended in approximately 5 ml of SFM or 5 ml of SFM alone were injected into the pericardial space. The infusion wire was then removed from the pericardial space, and the guide catheter was withdrawn and the venous and arterial sheaths were removed. Three additional animals were administered autologous DiI-labeled EPCs intrapericardially but did not undergo coronary ischemia.
2.4
Follow-up and histological assessment of EPC incorporation
Animals were sacrificed on Day 7 ( n =15; 6 from the SFM group, 6 from the EPC group with ischemia, and 3 from the EPC group without ischemia) or Day 21 ( n =6; 2 from the SFM group and 4 from the EPC group with ischemia) ( Fig. 1 ). All animals were preanesthetized with KTX and anesthetized with isoflurane (inhalation). Following a coronary angiogram and a transthoracic echocardiographic study (described below), the animals were euthanized with intravenous potassium chloride. The heart and intact pericardium were exposed through a median sternotomy. A small incision was made in the parietal pericardium, and pericardial fluid was aspirated. The hearts were then explanted and perfused with heparinized saline through the cannulated ascending aorta. Additionally, two of the EPC-treated animals and two of the SFM control animals were perfused with a Bandeiraea simplicifolia lectin I (BS-I; an endothelial cell marker) solution following the saline perfusion. After the parietal pericardium was removed, the heart was sectioned in 3-mm transverse slices from apex to base. Sections of the LV anterior septum and anterior wall as well as the right ventricle and inferoposterior wall were obtained from each transverse slice. Tissue samples were also obtained from the liver and lung. Specimens were then mounted in optimal cutting temperature compound (O.C.T., TissueTek) and snap-frozen in liquid nitrogen. Sections (5 μm) were made, and tissue was analyzed under fluorescent microscopy for DiI and BS-I if applicable. Fluorescent cell density was performed by quantifying the number of fluorescent cells per square millimeter in a high-power field (20×) using the Scion image software (v. 4.0.2). All samples were overread by an observer who was blinded to SFM versus EPC infusion.
