Background
Small animal models of ischemic left ventricular (LV) dysfunction are important for the preclinical optimization of stem cell therapy. The aim of this study was to test the hypothesis that temporal changes in LV function and regional perfusion after cell therapy can be assessed in mice using echocardiographic imaging.
Methods
Wild-type mice ( n = 25) were studied 7 and 28 days after permanent ligation of the left anterior descending coronary artery. Animals were randomized to receive closed-chest ultrasound-guided intramyocardial delivery of saline (n = 13) or 5 × 10 5 multipotential adult progenitor cells (MAPCs; n = 12) on day 7. LV end-diastolic and end-systolic volumes, LV ejection fraction, and stroke volume were measured using high-frequency echocardiography. Multiplanar assessments of perfusion and defect area size were made using myocardial contrast echocardiography.
Results
Between days 7 and 28, MAPC-treated animals had 40% to 50% reductions in defect size ( P < .001) and 20% to 30% increases in total perfusion ( P < .01). Perfusion did not change in nontreated controls. Both LV end-diastolic and end-systolic volumes increased between days 7 and 28 in both groups, but LV end-systolic volume increased to a lesser degree in MAPC-treated compared with control mice (+4.2 ± 7.9 vs +19.2 ± 22.0 μL, P < .05). LV ejection fraction increased in the MAPC-treated mice and decreased in control mice (+3.0 ± 4.3% vs −5.6 ± 5.9%, P < .01). There was a significant linear relation between the change in LV ejection fraction and the change in either defect area size or total perfusion.
Conclusions
High-frequency echocardiography and myocardial contrast echocardiography in murine models of ischemic LV dysfunction can be used to assess the response to stem cell therapy and to characterize the relationship among spatial flow, ventricular function, and ventricular remodeling.
The number of patients in the United States with symptomatic ischemic left ventricular (LV) dysfunction who are not eligible for surgical or percutaneous revascularization therapy because of comorbidity or diffuse pattern of disease is growing. Proangiogenic stem cell therapy is a potential option for preserving myocardial viability and reducing debilitating angina and heart failure symptoms. Proangiogenic effects may also be important in stem cell–mediated myocardial regeneration. Although clinical trials with cell therapy for ischemic LV dysfunction have been completed, there are many unanswered questions with regard to the most appropriate cell type(s), route of administration, dose, and even mechanism of action. Accordingly, methods for the spatial and temporal characterization of LV function and regional perfusion are important for preclinical and clinical evaluation of these unsolved issues as well as defining the relationship between flow and functional recovery with cell therapy.
Small animal models of ischemic LV dysfunction have been developed and provide a relatively inexpensive and high-throughput platform to test new angiogenic or regenerative therapies. We hypothesized that quantitative myocardial contrast echocardiography (MCE) and two-dimensional and Doppler echocardiographic methods for evaluating LV function beyond fractional shortening could be used to assess the impact of stem cell therapy in a murine model of chronic ischemic LV dysfunction. To test this hypothesis, mice with chronic coronary artery occlusion and moderate to severely reduced LV systolic function were treated with intramyocardial injection of xenogenic multipotential adult progenitor cells (MAPCs), which have been shown to improve LV function after myocardial infarction and have also been shown to increase perfusion in ischemic muscle through proangiogenic paracrine effects.
Methods
Animal Preparation and Murine Model of Ischemic LV Dysfunction
The study was approved by the Animal Care and Use Committee at Oregon Health & Science University. Ischemic LV dysfunction was produced by permanent coronary occlusion in 25 C57Bl/6 mice (Jackson Laboratories, Bar Harbor, ME) aged 8 to 10 weeks for echocardiographic and histologic studies and in 4 athymic Crl:CD1- Foxn1 nu mice with spontaneous albinism (Charles River Laboratories, Wilmington, MA) for in vivo optical imaging experiments. Mice were anesthetized with inhaled isoflurane (1.0%–1.5%), intubated, and placed on positive pressure ventilation. A left lateral thoracotomy was performed using aseptic technique to expose the anterior myocardial wall, and a suture was placed around the mid left anterior descending coronary artery (LAD). The electrocardiogram was monitored to confirm myocardial injury, evidenced by ST-segment elevation. The chest wall was then closed with interrupted sutures, and the endotracheal tube was removed after voluntary respiration was confirmed. Mice were then recovered, and buprenorphine hydrochloride (0.2 mg/kg, intramuscular) was administered for analgesia. Ten nonischemic control C57Bl/6 mice were also studied at 8 to 10 weeks of age. For all subsequent imaging studies, mice were anesthetized with inhaled isoflurane (1.0%–1.5%), and heart rate ranged from 450 to 530 beats/min. For MCE, a catheter was placed in a jugular vein for the administration of microbubbles. All follow-up echocardiographic imaging was completed in <30 min.
Cell Therapy
Rat MAPCs (Athersys, Inc, Cleveland, OH) were isolated and expanded as previously described and stored in liquid nitrogen until the day of use. These cells are positive for CD29, CD49c, and CD90 and negative for CD34, CD45, and CD106. Cell viability on the day of use was assessed by trypan blue exclusion. On day 7 after LAD ligation, mice were anesthetized with inhaled isoflurane. Either MAPCs (5 × 10 5 suspended in 20 μL of saline; n = 12) or saline alone ( n = 13) was injected into the anterior-lateral aspect of the infarct border zone. Intramyocardial injection was performed percutaneously using a 30-gauge needle and high-frequency (40-MHz) ultrasound guidance (Vevo 770; VisualSonics, Inc, Toronto, Ontario, Canada) and a three-dimensional micropositioning system (VisualSonics, Inc). Intramyocardial delivery was confirmed by distortion of myocardial speckle and acute increase in wall thickness by real-time imaging during injection ( Video 1 ; available at www.onlinejase.com ). Animals were not studied if there was evidence of focal pericardial or mediastinal injection, evidenced by extracardiac fluid accumulation.
Cell Localization and Survival
In vivo optical imaging was performed to assess MAPC location and survival and function after injection in athymic Crl:CD1- Foxn1 nu mice ( n = 4). For these experiments, MAPCs were stably transfected with the firefly luciferase gene using a lentivirus vector. The number of cells injected 7 days after LAD ligation was increased to 1 × 10 6 to produce robust luciferase-generated photon activity through the chest wall. On days 1, 2, 3, and 7 after MAPC injection, mice were anesthetized with inhaled isoflurane and d -luciferin (150 μg/g, intraperitoneal) was administered. Optical imaging (IVIS Spectrum; Caliper Life Sciences, Hopkinton, MA) was performed 10 min after luciferin injection using medium binning. Luciferase activity was expressed as photons per second per square centimeter.
LV Function by Echocardiography
Echocardiography for LV function was performed in nonischemic control mice and in postinfarction mice on day 7 after coronary ligation (before MAPC or saline injection) and on day 28 (21 days after injection). High-frequency fundamental imaging (Vevo 770) was performed at 25 to 40 MHz depending on the echocardiographic data that were acquired. Mice were sedated with inhaled isoflurane (1.0%–1.5%). Images were obtained in the parasternal long-axis plane and three equally spaced parasternal short-axis planes 1 mm apart representing the basal, midpapillary, and apical levels. Image registration to ensure similar imaging planes for days 7 and 28 was based on both anatomic features (papillary muscles and trabecular features) and distance from the mitral valve plane. The LV cross-sectional area was measured at end-diastole and end-systole for each of the short-axis image acquisitions. LV end-diastolic volume (LVEDV) and LV end-systolic volume (LVESV) were calculated by summation of interpolated short-axis area measurements. LV ejection fraction (LVEF) was calculated as (LVEDV − LVESV)/LVEDV. Stroke volume was determined using two methods (1) the difference in LVESV and LVEDV and (2) the product of left ventricular outflow tract area and time-velocity integral on pulsed-wave Doppler. For the latter, left ventricular outflow tract pulsed-wave Doppler was obtained from a high parasternal long-axis view with heel-toe angulation to provide a similar angle of incidence between animals and within the range for automated angle correction.
Myocardial Perfusion Imaging
Regional myocardial perfusion and spatial extent of perfusion were assessed using MCE on the same days that LV function was assessed by two-dimensional echocardiography. Lipid-shelled decafluorobutane microbubbles were prepared by sonication of an aqueous lipid dispersion of polyoxyethylene-40-stearate and distearoyl phosphatidylcholine saturated with decafluorobutane gas. Microbubble concentration was measured by electrozone sensing (Multisizer III; Beckman Coulter, Brea, CA). Myocardial perfusion imaging was performed using a linear-array transducer (15L8) interfaced with an ultrasound system (Sequoia; Siemens Medical Solutions USA, Inc, Mountain View, CA). Short-axis images at the same midventricular and apical levels as for functional images were obtained, the position for which was guided by 14-MHz fundamental imaging and a micropositioning imaging stage. For perfusion imaging, a multipulse algorithm using phase inversion and amplitude modulation was used to detect the nonlinear fundamental component of the microbubble signal at a frequency of 7 MHz. Imaging was performed at a mechanical index of 0.16 and a dynamic range of 55 dB. Electrocardiographically triggered end-systolic frames were acquired during continuous intravenous infusion of microbubbles at 5 × 10 5 min −1 . Images were acquired for 15 cardiac cycles after a 5-frame destructive pulse sequence at a mechanical index of 0.97. Quantification of myocardial perfusion was performed from regions of interest placed over the entire short-axis area. Time-intensity data were fit to the function y = A (1 − e −β t ), where A is plateau intensity representing relative microvascular blood volume, β is the rate constant reflecting microvascular flux rate, and the product of A and β is an index of myocardial blood flow. Color-coded parametric images of microvascular blood volume (images at the completion of the replenishment sequence) were used to delineate the region devoid of perfusion. Interobserver variation of blinded measurements for the region devoid of perfusion was 7 ± 4%. The percentage change in nonperfused area from baseline to follow-up was calculated by averaging data for the midventricular and apical planes.
Microsphere Validation of Spatial Perfusion Assessment
Microsphere evaluation of regional perfusion on day 28 was performed in 8 mice undergoing LAD ligation ( n = 4 each for the MAPC and sham saline injection groups). After the completion of MCE, 8 × 10 5 fluorescent microspheres (15 μm Dye-Trak; Triton Technology, Inc, San Diego, CA) were injected into left ventricle using high-frequency ultrasound guidance. Several minutes later, hearts were excised, rinsed in phosphate-buffered saline, and sectioned into 1-mm-thick short-axis slices. Fluorescent epi-illumination microscopy (20× objective) with 460-nm to 500-nm excitation filters was performed. Digital planimetry was used to determine the region void of fluorescent microspheres on both the basal and apical surfaces of each slice, which were averaged and expressed as a ratio to the total LV short-axis area. Analysis was performed for the slices for which ventricular geometry matched the myocardial contrast echocardiographic imaging planes. Interobserver variation of blinded measurements was 6 ± 2%.
Statistical Analysis
Data were analyzed using SPSS version 16.0 (SPSS, Inc, Chicago, IL). All echocardiographic data were normally distributed and are expressed as mean ± SD unless otherwise stated. Temporal changes in luciferase data were analyzed using one-way analysis of variance and tests for nonlinear trends. Echocardiographic perfusion and function data were analyzed using paired Student’s t tests (two sided) for temporal differences within a treatment group and unpaired Student’s t tests for cohort-related differences at the different study intervals and Bonferroni correction for multiple comparisons when comparing the two treatment groups with nonischemic control animals. Correlations were made using least squares linear regression analysis. Differences in proportions for risk area size were analyzed using χ 2 tests. Differences were considered significant at P < .05.
Results
Intramyocardial Stem Cell Survival
On ultrasound-guided percutaneous injection of MAPCs, success of intramyocardial injection of cells into the anterolateral portion of the infarct was confirmed by mild tissue distortion during injection. In vivo optical imaging of luciferase-transfected MAPCs indicated that the cells remained primarily at the site of injection in the myocardium ( Figure 1 ). There was rapid loss either of cells or of cell function over the first week, manifest by a 500-fold reduction in luciferase signal.
Myocardial Perfusion
Myocardial perfusion was assessed using MCE in both apical and mid-LV short-axis locations that encompassed regions with akinesis on day 7 after LAD ligation in all animals. The apical location was midway between the midventricle and the true apex, which was not imaged because it was entirely akinetic in all animals and lacked any evidence of myocardial perfusion. On day 7 (before MAPC injection), the region that was void of myocardial perfusion was larger for the apical than the midventricular plane (26 ± 15% vs 12 ± 6%, P = .0001; Figure 2 A ). There was no difference in baseline defect size on day 7 between the MAPC-treated and sham-treated postinfarct groups. On day 28 (21 days after injection), there was a significant reduction in perfusion defect size only for MAPC-treated mice. In these mice, there was a similar 50% relative reduction in defect size for the apical and midventricular regions. The majority of MAPC-treated segments had reductions in perfusion defect size, which was not the case with sham-treated mice ( Figure 2 B). Validation of myocardial contrast echocardiographic defect size by fluorescent microsphere injection was performed in 8 mice on day 28 and demonstrated good agreement in the spatial extent of the defect size for the two techniques ( Figure 2 C).
Quantitative transmural myocardial perfusion for the entire apical and midventricular short-axis regions on day 7 in mice undergoing LAD ligation was reduced compared with nonischemic control mice ( Figure 3 ). On day 7, myocardial perfusion in the apical region was slightly lower for the MAPC-treated mice, although this did not reach statistical significance. Between days 7 and 28, myocardial perfusion increased significantly in the MAPC-treated mice and did not change in the nontreated mice. However, differences in perfusion between the treatment groups on day 28 did not reach statistical significance.
