Widespread intracoronary allogeneic cardiosphere-derived cell therapy with and without cyclosporine in reperfused myocardial infarction

Abstract

Allogeneic cardiosphere-derived cell (CDC) therapy has been demonstrated to improve myocardial function when administered to reperfused myocardial infarcts. We previously pretreated animals with low-dose cyclosporine immunosuppression to limit allogeneic CDC rejection, but whether it is necessary and, if so, can be initiated at the time of reperfusion remains uncertain. Closed-chest swine (n = 29 animals) were subjected to a 90-min left anterior descending (LAD) coronary artery occlusion. Using a three-way blinded design, we randomized two groups to receive global intracoronary infusions of 20 × 10⁶ CDCs 30 min after reperfusion. A third control group was treated with saline. One CDC group received cyclosporine 10 min before reperfusion (2.5 mg/kg iv and 100 mg/day po), whereas the other groups received placebos. After 1 mo, neither chronic infarct size relative to area at risk (saline control, 46.2 ± 4.0%; CDCs, 46.4 ± 2.1%; and CDCs + cyclosporine, 49.2 ± 3.1%; P = 0.79) nor ejection fraction (saline control, 51 ± 2%; CDCs, 51 ± 2%; and CDC + cyclosporine, 48 ± 2%; P = 0.42) were different among treatment groups. Multiple histological measures of cellular remodeling, myocyte proliferation, and apoptosis were also not different among treatment groups. In contrast to previous studies, we were unable to reproduce the cardioprotective effects demonstrated by allogeneic CDCs without cyclosporine. Furthermore, initiation of intravenous cyclosporine at the time of reperfusion followed by oral therapy was not sufficient to elicit the functional improvement observed in studies where cyclosporine was started 72 h before CDC therapy. This suggests that oral cyclosporine pretreatment may be necessary to effect cardiac repair with allogeneic CDCs.

NEW & NOTEWORTHY In a three-way blinded, randomized design, we determined whether allogeneic CDCs administered at reperfusion improved myocardial function and whether intravenous cyclosporine enhanced their efficacy. In contrast to prior studies using oral cyclosporine, CDCs with or without intravenous cyclosporine had no effect on function or infarct size. This indicates that CDCs may be most efficacious for treating chronic LV dysfunction where cyclosporine can be initiated at least 72 h before cell therapy.

INTRODUCTION

Cell-based therapies for ischemic heart disease have been extensively studied over approximately the past 20 years with the goal of identifying an effective treatment strategy to repair damaged myocardium and replace myocytes lost after injury. Cardiosphere-derived stem cells (CDCs) have emerged as a promising candidate approach in this area and autologous CDCs quickly entered clinical trials following supportive preclinical studies in large animal models of myocardial infarction (1, 2). Although interest in CDCs was initially fueled by their cardiomyogenic potential, more recent observations support the notion that these cells primarily exert therapeutic effects on endogenous myocytes via paracrine mechanisms including the release of exosomes (3, 4). This has motivated the use of allogeneic CDC therapy as a therapeutic platform that would be readily available and not require ex vivo expansion of autologous cells. Indeed, allogeneic CDCs have been shown to be safe and effective in preclinical animal models of acute and chronic ischemic heart disease (2, 5–10).

Like other forms of cell therapy, the translation of preclinical results with CDCs in animal models to humans with cardiovascular disease has been variable. Small early phase clinical studies have supported beneficial effects of autologous CDCs administered into the infarct-related artery (2) as well as widespread global administration of allogeneic CDCs in nonischemic cardiomyopathy (11). In contrast, recent larger studies employing allogeneic CDCs administered into the infarct-related artery after myocardial infarction have failed to demonstrate a reduction in infarct size or improvement in global function (12). The inability to translate preclinical studies of cell therapy could be related to rigor, limited sample size causing false-positive results, and/or potentiation of cardiac repair in healthy juvenile animals versus subjects with chronic disease. Alternatively, the inability to translate favorable effects could reflect several modifiable factors. First, the majority of adverse postinfarction left ventricular remodeling may have occurred before treatment with CDC infusion and administration at the time of reperfusion may more effectively limit myocyte loss and the development of left ventricular dysfunction. The lack of effect could also reflect the need to infuse CDCs to remote myocardium as well as the infarct to prevent deleterious postinfarction left ventricular (LV) remodeling (10). Finally, the therapeutic potential of allogeneic CDCs may be limited by more rapid clearance than autologous cells, which would reduce the duration of paracrine effects in vivo (5). Here, the duration of action could potentially be prolonged by using immunosuppression to inhibit CDC clearance.

To resolve the latter three issues, we performed the present study to assess whether cyclosporine immunosuppression enhanced the efficacy of global intracoronary CDCs administered immediately after reperfusion of an acute myocardial infarct. A three-way treatment design compared CDCs administered with or without cyclosporine immunosuppression to saline-treated controls. We adapted our previous approach where low-dose oral cyclosporine was started 72 h before infarction to a translationally relevant approach where an intravenous loading dose was administered immediately before reperfusion and followed by daily oral administration as in our previous study (10). To ensure rigor, we employed randomization and blinding in all aspects of the study, including treatment allocation, animal subject exclusion, and analysis.

METHODS

All procedures and protocols conformed to institutional guidelines for the care and use of animals in research and were approved by the University at Buffalo Institutional Animal Care and Use Committee. The data that support the findings of this study are available from the corresponding author on reasonable request. The animal studies were conducted in accordance with the ARRIVE (Animal Research: Reporting of in vivo Experiments) guidelines (13). Briefly, we employed rigorous methodology including randomization and blinding to treatment allocation as outlined in previous studies from our laboratory (14) and summarized in recent consensus statements (15, 16). All personnel involved in data collection and analysis (including hemodynamics, CT imaging, echocardiography, and histopathological measurements) were blinded to the treatment status of each animal. We biased sex toward female enrollment to allow cell retention to be assessed with sex-mismatched CDC delivery but without consideration of treatment. The studies were powered with the primary endpoint being a 5% difference in ejection fraction between CDC- and vehicle-treated animals at 1 mo. A sample size of 10 animals per group had 0.80 power at the P < 0.05 level significance.

Experimental Design

The closed-chest myocardial infarction protocol was similar to that previously described by our laboratory and the protocol summarized in Fig. 1 (14). Farm-bred pigs of either sex, 3–4 mo of age, were purchased from WBB Farm (Alden, NY). Following initial sedation with a telazol (100 mg/mL)-xylazine (100 mg/mL) mixture (0.04 mL/kg iv), anesthesia was maintained with a continuous infusion of propofol (5–10 mg/kg/h) through an 18-gauge ear vein catheter. All animals received prophylactic antibiotics (cefazolin; 1,000 mg iv). They were subsequently intubated and mechanically ventilated with oxygen at a respiratory rate of 15–20 breaths/min throughout the study. Electrocardiographic activity, arterial oxygen saturation (pulse oximetry), and body temperature were monitored throughout the study. A 6-Fr introducer was placed in a femoral vein and a 7-Fr introducer was placed in a femoral artery. Arterial blood gases were assessed throughout the study and maintained within physiological levels via adjustments in ventilation rate and/or volume. After heparinization (5,000 U iv), baseline heart rate and femoral arterial blood pressure were assessed. Baseline LV function was also assessed via parasternal, transthoracic two-dimensional (2-D) echocardiography (GE Vivid 7) and placement of an intraventricular pressure catheter (Millar) with the pig in the left lateral recumbent position. Measurements of ejection fraction and regional wall thickening were obtained using American Society of Echocardiography criteria (17). Following data collection and before left anterior descending (LAD) coronary artery occlusion, amiodarone (5 mg/kg) and lidocaine (1.5 mg/kg) boluses were administered followed by continuous infusions (amiodarone, 0.04 mg/kg/min; lidocaine, 0.05 mg/kg/min). They were stopped 15 min after reperfusion. Defibrillator pads were placed on the animal’s chest and connected to a LIFEPAK 15 defibrillator (Physio-Control). If ventricular fibrillation developed during the study, biphasic defibrillation starting at 200–250 J was administered and when needed, increased by 50–60 J until the maximum of 360 J was reached. Once successful, pressure was closely monitored to ensure appropriate recovery. If prolonged hypotension or refractory ventricular fibrillation developed, the pig was excluded from the study.

Figure 1.

Experimental overview. After baseline data collection including echocardiography and multidetector computed tomography (MDCT) imaging, swine were subjected to a 90-min occlusion of the left anterior descending (LAD) coronary artery in the closed-chest state. Contrast-enhanced MDCT was performed 5 min after the onset of LAD occlusion for assessment of the ischemic area at risk (AAR). Ten minutes before reperfusion, a bolus of cyclosporine (2.5 mg/kg) or saline (vehicle) was administered via intravenous infusion in blinded and randomized fashion. Thirty minutes after reperfusion, an intracoronary infusion of allogeneic CDCs (20 million) or saline was delivered globally to the heart via the left anterior descending (LAD) coronary artery, left circumflex artery (LCX), and right coronary artery (RCA). Serial echocardiography was performed to assess left ventricular (LV) function, and MDCT imaging was repeated 1 mo postreperfusion to assess LV ejection fraction and remodeling. After euthanasia, the heart was excised for postmortem assessment of infarct size and cellular remodeling. All study personnel were blinded to treatment allocation until completion of data collection and analysis. CDCs, cardiosphere-derived cells. n = number of animals.

After an additional bolus of heparin (5,000 U iv) a 6-Fr. guiding catheter (Cordis, Hockey Stick) was advanced into the left main coronary artery for angiography (Iohexol, 350 mg I/mL). We administered nitroglycerin (250 µg ic) and inserted a guide wire into the LAD artery. Subsequently, an appropriately sized balloon angioplasty catheter (Boston Scientific, Maverick, 2.5–3.25 mm × 6–12-mm long) was advanced into the LAD and positioned just distal to the second diagonal artery targeting an ischemic area at risk (AAR) of ∼20% of LV mass. The angioplasty balloon was transiently inflated to 6–9 atm (<10 s) to angiographically confirm that the diameter was appropriate to completely occlude the LAD. The animals were subsequently transported to the adjacent CT scanner (GE Discovery VCT PET/CT) to begin the infarct protocol. We inflated the balloon catheter to totally occlude the LAD for 90 min. Shortly after the onset of LAD occlusion, we assessed the ischemic risk area via contrast-enhanced MDCT. Pigs were subsequently transferred back to the adjacent physiology laboratory where we repeated coronary angiography to document that the LAD remained occluded. After 45 min, an additional heparin bolus (3,000 U) was administered to ensure continued anticoagulation.

Ten minutes before reperfusion, animals were randomly allocated to receive a bolus of cyclosporine (2.5 mg/kg iv in 20 mL of saline; NDC 0574–0866-10, Perrigo) or saline with all study personnel blinded to treatment allocation. Plasma samples were drawn to assess cyclosporine concentration. After 90 min of LAD occlusion, the balloon was deflated after angiographic confirmation of occlusion and the angioplasty balloon and guiding catheter were withdrawn. After 20 min of reperfusion, a pre-cell administration echocardiogram was performed. Then, with a 5-Fr. diagnostic catheter (sones or hockey stick), ∼20 million CDCs or saline was infused in divided doses into each of the three major coronary arteries. An echocardiogram was performed to assess LV function after which animals were weaned off anesthesia, extubated, and returned to the animal facility for recovery. Animals randomized to cyclosporine (Teva NDC 0093–5742-65 or Mayne Pharma NDC 51862-460-47) received 200 mg of the capsule contents mixed in peanut butter once a day, whereas all other animals received peanut butter alone to preserve blinding. Plasma samples were obtained to assess peak cyclosporine levels immediately before reperfusion, 1 h after reperfusion, and at the end of the study.

One month after reperfusion, animals returned to the laboratory for a terminal physiological study. Following sedation and anesthesia, catheters were placed to assess hemodynamics including LV pressure (Millar), and coronary angiography was performed to confirm patency of the LAD. We repeated assessment of LV function with both echocardiography and MDCT. At the end of the study, animals were deeply anesthetized with 5% isoflurane. The heart was arrested with an injection of KCl and immediately excised. The atria and right ventricles were removed, and the LV was subsequently sectioned into 8-mm-thick concentric rings parallel to the atrioventricular groove that were individually weighed and used for postmortem infarct size and histopathology measurements.

Allogeneic CDC Isolation and Ex Vivo CDC Expansion

We isolated CDCs as previously described by our laboratory (6, 9, 10). Atrial tissue samples were collected from three male donor pigs and placed in cold sterile phosphate-buffered saline (PBS). Tissue was minced into 1–2 mm pieces, washed several times in cold PBS, and covered with freshly prepared collagenase IV (Gibco No. 17104019, 1 mg/mL in DMEM:F12) for 1 h at 37°C. Before CDC culture, a six-well plate was coated with fibronectin (0.02 µg/mL sterile H₂O) for at least 2 h in a 37°C incubator. Cardiac explant media (CEM) was produced using 80% IMDM-20% FBS (Hyclone; Cat. No. SH3007003), glutamax (Gibco; Cat. No. 35050061), 0.1 mm β-mercaptoethanol (Gibco; Cat. No. 21985023), and penicillin-streptomycin (Gibco; Cat. No. 15140148). The fibronectin solution was aspirated from the six-well plate, and explant pieces were placed on the plate containing a minimal amount of CEM media, such that the pieces maintained contact with the plastic and did not float. The plates were incubated at 37°C, 5% CO₂, gently exchanging media without disturbing adhesions to the fibronectin every 2–3 days. After ∼10 days, when the plate was well grown and confluent, outgrowth cells were collected, first by saving two PBS washes and then by trypsinization (0.05%, Gibco; Cat. No. 25300054) until ∼50% of the cells were released. Floating cells were washed with CEM media and added to the collection to quench the trypsin. The remaining cells and explants were recovered with CEM and placed back in incubator for a second harvest 3–5 days later.

Collected cells were spun down at 180 g for 8 min and resuspended in CEM media. Trypan blue-stained cells were counted, and 200,000–250,000 cells were plated per well in a six-well low attachment plate (Corning; Cat. No. 3471). Cells clumped together to make multiple cardiospheres per well. Medium was exchanged every 2 days by transferring cardiospheres to conical tubes, allowing them to settle for 7–10 min and then aspirating and exchanging media before returning them to the plate. After 5 days, each well of cardiospheres was plated onto one to two wells of fibronectin-coated plates and rocked to ensure even distribution. Cardiospheres were incubated at 37°C, 5% CO₂, and fed every 2 days for 7–10 days until adhered cells reach 70–80% confluence as P0 CDCs. These CDCs were then trypsinized and passed or frozen in CDC freezing media (CEM/5% B27; Gibco; Cat. No. 17504044/10% DMSO) for later use.

To passage cells, CDCs grown to ∼80% confluence were trypsinized with 0.05% or 0.25% trypsin and transferred to conical tubes with a final concentration of at least 10% FBS to stop digestion. Tubes were centrifuged at 50 g for 7 min, and cells were resuspended in fresh CEM and plated on prepared fibronectin plates at a ratio of 1–8.

CDC Administration

Cells were cultured for three to six passages, at which time they were collected for characterization by flow cytometry and intracoronary infusion to recipient pigs. Approximately 100,000 cells in 50-µL PBS-1% BSA were incubated with 1–2 µL of fluorescently labeled antibody, 7-AAD viability stain (eBioscience; Cat. No. 00699350), or appropriate isotype control. The antibodies are listed in Supplemental Table S1 (all Supplemental material is available at https://doi.org/10.6084/m9.figshare.20341182). Before administration, cells were filtered through a 30-µm pore filter to circumvent administering cell aggregates (MACS preseparation filters, Miltenyi Biotec) and suspended in heparinized PBS (3,000 U heparin in 30 mL in total) for intracoronary infusion. This was distributed between three 10-mL syringes that were covered with transpore tape to obscure the contents and blind the investigative team. Because of a concern about an increased number of periprocedural hematomas at the vascular access site in other ongoing studies, a small number of the initial studies (3 control, 1 CDC, and 2 CDC + C) eliminated the heparin in the CDC infusion syringes. Although systemic anticoagulation was thought to be sufficient to prevent CDC clumping, three animals receiving CDCs without heparin developed small punctate triphenyl tetrazolium chloride (TTC) infarcts outside of the risk area. These three animals were excluded from the primary results before unblinding to treatment. A secondary analysis including them was performed to determine if this impacted any conclusions.

The syringes containing CDCs or saline control were infused into each of the three major coronary arteries over 10 min. The same sequence was used in each animal: the initial 10 mL was delivered into the LAD, followed by the second 10 mL into the circumflex artery LCX, and finally the third 10-mL syringe into the right coronary artery (RCA). Each syringe containing the study agent was followed by a saline flush. Angiograms of each vessel were performed before and after cell administration to exclude interval development of slow flow indicative of potential microvascular obstruction.

Pathological Determination of Infarct Size

All investigators and technicians remained blinded to treatment allocation for all measurements. Every other LV ring was weighed, placed in 1% triphenyl tetrazolium chloride (TTC) for 20 min to stain viable myocardium, and imaged with a digital scanner (HP Scanjet G4050). Infarct size was quantified by planimetry using ImageJ software (National Institutes of Health, NIH). The infarct areas from the apical and basal sides of each ring were averaged and multiplied by the ring weight to obtain infarct volume. For the rings not directly analyzed in TTC, the corresponding apical- and basal-area measurements of adjacent rings were used. The measurements of all rings were summed to obtain total infarct volume. The results represent the average of two blinded observers.

Myocyte Morphometry, Proliferation, and Apoptosis

An 8-mm ring near the midventricular level was used for all histopathological analyses. Samples corresponding to the normal remote region (posterior descending artery-supplied territory) were fixed in formalin and paraffin-embedded for histopathological analyses. Periodic acid-Schiff (PAS) glycogen-stained sections (5 μm) were used to quantify myocyte diameter and myocyte nuclear density in each region as previously described. Ten random fields were selected, and data were expressed as nuclei per tissue area (mm²).

Myocyte proliferation was quantified with Ki67 and myocyte apoptosis with TUNEL staining using paraffin-fixed tissue sections (5 μm) from the remote region as we have previously reported major changes in remote zone remodeling after global administration of CDCs immediately after infarction (10), as well as in models of hibernating myocardium. Myocyte nuclei were identified with cardiac troponin I (Invitrogen; Cat. No. 701585) and DAPI (Vectashield). Myocyte proliferation was determined using anti-K_i-67 (mouse monoclonal antibody, clone MIB-1, Dako; Cat. No. GA62661-2). Myocyte apoptosis was quantified using the In Situ Cell Death Detection Kit, Fluorescein (Roche) according to the manufacturer’s guidelines. Apoptotic cells were detected by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) and epifluorescence with a FITC filter. Samples used for proliferative (K_i-67⁺) or apoptosis (TUNEL⁺) measurements were each costained with cardiac troponin I antibody and colocalization was used to quantify proliferating or apoptotic cardiomyocytes. Nuclei that could not be definitively confirmed to be of myocyte origin were excluded. Images were acquired with a laser-confocal microscope (Zeiss LSM 700) and AxioImager equipped with ApoTome (Zeiss). The frequency of proliferating (K_i-67) and apoptotic (TUNEL) myocytes was expressed as cells per square millimeter, with normalization of these values to the number of positive cells per million myocytes as previously described (18).

Multidetector Computed Tomography

Contrast-enhanced multidetector computed tomography (MDCT) was performed with a 0.5 mm × 64 slice MDCT scanner (Discovery VCT, GE Healthcare). The following parameters were used for each scan: gantry rotation time, 400 ms; temporal resolution, 175 ms; slice thickness, 0.625 mm; spatial resolution, 0.97 × 0.97 mm (voxel size, 0.97 × 0.97 × 0.625 mm); helical pitch variable depending on heart rate (range, 0.20–0.26); tube voltage, 120 kV; and tube current, 600 mA.

After a scout scan and timing bolus with intravenous iohexol (350 mg I/mL; 10 mL, 4 mL/s), LV function was assessed via arterial phase ECG-gated CT following an intravenous bolus of iohexol (350 mg I/mL; 1 mL/kg at 4 mL/s). First-pass CT acquisitions were reconstructed in 10% phases from 5% to 95% throughout the entire R–R interval and reformatted at a 6-mm slice thickness in the short axis. Using ImageJ software, the endocardium and epicardium of each slice were manually traced at end diastole and end systole for calculation of LVEF, LV end-diastolic volume, LV end-systolic volume, LV stroke volume, and LV mass.

We assessed the ischemic area at risk (AAR) using an arterial phase ECG-gated CT performed 5 min after LAD occlusion. Following an intravenous bolus of iohexol (350 mg I/mL; 2 mL/kg at 4 mL/s), first-pass images were reconstructed at mid-diastole (75% of the R–R interval) and reformatted at 8-mm slice thickness in the short axis. The CT AAR was manually traced by two independent blinded investigators using ImageJ software and expressed as a percentage of LV mass. Results represent the average of the two measurements.

Assessment of Serum Cardiac Troponin I

Blood samples collected from a peripheral vein at baseline and 1 h postreperfusion were allowed to clot at room temperature for 40 min, centrifuged at 1,500 g for 15 min, aliquoted, and frozen for storage at −80°C. Serum was thawed once and cardiac troponin I (cTnI) was quantified in duplicate with a porcine-specific cTnI ELISA kit for serum (Life Diagnostics, CTNI-9-HS) according to the manufacturer’s instructions.

Quantification of Donor-Derived CDCs in the Myocardium at 1 Mo

Allogeneic male donor-derived cells were quantified in myocardial samples from CDC-treated female recipients (6 control, 8 CDCs, and 8 CDCs + cyclosporine) using genomic DNA and a real-time PCR-based approach adapted from previously published protocols (19). PCR primers were designed against porcine Y chromosome-specific SRY (forward, 5′- CTCACAGCCCATGAACATAACC-3′, reverse, 5′- GAAAGTCCCGGCTGTAAACC-3′). Porcine GPI (glucose-6-phosphate isomerase, forward, 5′- TTCAGGACGTTCAACTCAATAGG-3′, reverse, 5′- GGGACTATGACTGTCAGGTAAGG-3′) served as a positive control for the presence of porcine DNA. Targets were amplified using SsoFast EvaGreen Supermix (Bio-Rad) and CFX Real-Time PCR Detection System (Bio-Rad). Data were processed using Bio-Rad CFX Manager software and included melt-curve analysis for the verification of single PCR products for each reaction.

Statistical Analysis

Data are expressed as means ± SE. Differences between treatment groups were assessed by a one-way analysis of variance (ANOVA). When significant differences were detected, a post hoc Student’s t test was used for all pairwise comparisons (SPSS Statistics, Version 22; IBM). Post hoc tests were not adjusted for multiple comparisons. All other between-group differences were assessed using an unpaired Student’s t test. The Shapiro–Wilk test was used to determine whether data were normally distributed. Serum cTnI values were not normally distributed, thus logarithmic transformations were performed before statistical analysis. For all comparisons, P < 0.05 was considered significant.

RESULTS

A total of 41 studies were attempted with the following exclusions applied while blinded to treatment allocation. Three animals did not complete the initial infarct protocol: one secondary to hemodynamic instability from unusually severe hypotension (<40 mmHg) during occlusion (saline), a second because of refractory ventricular fibrillation, and another because of persistent partial LAD occlusion after deflating the angioplasty balloon. After the initial study and randomization, six animals were excluded because of death associated with cardiorespiratory complications. Finally, among animals that completed the final study (n = 32 animals), three were excluded because of small punctate TTC infarcts outside of the risk area (2 CDC and 1 CDC + cyclosporine). These were felt to be embolic and occurred in the initial studies where heparin was omitted from the CDC syringes. After these exclusions, 29 animals completed the protocol (n = 9 saline, n = 10 CDC, and n = 10 CDC + cyclosporine). Table 1 summarizes key characteristics of the animals receiving CDCs, CDCs + cyclosporine, or vehicle during the initial infarct study. Sex, weight, and hemodynamic parameters did not differ among treatment groups at any point in time (ANOVA). The rectal temperature remained constant throughout ischemia and reperfusion.

Table 1.

Baseline parameters at the time of myocardial infarction

	Vehicle	CDCs	CDCs + CsP
N	9	10	10
Exclusions, animals excluded/animals assigned	3/12	3/13	6/16
Male/female	3/6	1/9	2/8
Body weight, kg	42.7 ± 2.1	47.2 ± 1.8	43.0 ± 2.5
Heart rate, beats/min
Baseline	100 ± 8	92 ± 4	96 ± 4
45-min occlusion	67 ± 3	67 ± 3	69 ± 3
75-min occlusion	66 ± 3	65 ± 3	67 ± 3
15-min reperfusion	97 ± 7	96 ± 8	88 ± 6
60-min reperfusion	100 ± 4	100 ± 6	87 ± 6
Mean arterial pressure, mmHg
Baseline	106 ± 4	109 ± 6	112 ± 6
45-min occlusion	78 ± 4	82 ± 7	76 ± 4
75-min occlusion	76 ± 4	80 ± 7	75 ± 4
15-min reperfusion	72 ± 4	77 ± 5	69 ± 4
60-min reperfusion	90 ± 2	86 ± 5	80 ± 5
Body temperature, °C
Baseline	38.1 ± 0.3	38.0 ± 0.5	38.6 ± 0.2
45-min occlusion	38.5 ± 0.3	38.1 ± 0.2	38.7 ± 0.3
75-min occlusion	38.5 ± 0.3	38.1 ± 0.1	38.8 ± 0.2
90-min occlusion	38.5 ± 0.3	38.2 ± 0.1	38.7 ± 0.2
15-min reperfusion	38.5 ± 0.3	38.2 ± 0.1	38.7 ± 0.2
Plasma cardiac troponin, ng/mL
Baseline	0.042 ± 0.022	0.019 ± 0.009	0.014 ± 0.005
90-min occlusion	0.079 ± 0.021	0.092 ± 0.037	0.062 ± 0.016
60-min reperfusion	62.7 ± 9.8	63.5 ± 13.5	59.3 ± 9.9
Area at risk, by MDCT
Absolute grams	20.0 ± 1.0	23.6 ± 0.8	21.6 ± 1.1
%LV	22.2 ± 1.0	23.7 ± 0.7	24.7 ± 1.4

Values are represented as means ± SE; N, number of animals. All measures are not significantly different between groups by ANOVA. CDCs, cardiosphere-derived cells; CsP, cyclosporine treated; MDCT, multidetector CT; LV, left ventricle.

Characterization of CDC surface markers by flow cytometry at the time of injection (n = 14) was comparable with our previously reported studies (6, 9, 10) and did not vary between CDC-treated groups (Supplemental Table S1). Donor CDCs had negligible CD45 expression (0.1%), high levels of CD90 (95.7%), and CD105 (98.6%) and low levels of c-Kit (5.4%). In cyclosporine-treated animals, plasma levels measured 10 min before reperfusion reached 2,043 ± 141 ng/mL and remained elevated at 534 ± 58 ng/mL 60 min after reperfusion. Trough plasma levels measured 24 h after receiving oral cyclosporine were below the lower limit of detection of our assay (<30 ng/mL). Finally, persistent CDC retention was not detectable after 1 mo as assessed by qPCR for the Y chromosome in female recipients.

Effect of CDCs and CDCs + Cyclosporine on Myocardial Infarct Size

There were no differences in the ischemic AAR assessed with MDCT before reperfusion among treatment groups (saline, 22.2 ± 1.0%; CDCs, 23.7 ± 0.7%; and CDCs + cyclosporine, 24.7 ± 1.4%; P = 0.26; Fig. 2). Pathological infarct size by TTC staining was similar among all three treatment groups whether expressed as a percentage of AAR (saline, 46.2 ± 4.0%; CDCs, 46.4 ± 2.1%; and CDCs + cyclosporine, 49.2 ± 3.1%; P = 0.79) or as a percentage of LV mass (saline, 6.7 ± 0.7% versus CDCs, 6.7 ± 0.3% and CDCs + cyclosporine, 6.9 ± 0.6%; P = 0.95; Fig. 2). The similarity in infarct size was corroborated by serum cTnI levels measured 1 h after reperfusion, which also did not differ among groups (saline, 62.7 ± 9.8 ng/mL; CDCs, 63.5 ± 13 ng/mL; and CDCs + cyclosporine, 59.3 ± 9.9 ng/mL; P = 0.96). A secondary analysis performed with the three exclusions was the same (Supplemental Table S2). Thus, global intracoronary CDCs administered with or without cyclosporine started at the time of reperfusion had no effect on myocardial infarct size.

Figure 2.

Measures of area at risk and infarct size. The ischemic area at risk (AAR) was quantified by contrast-enhanced multidetector computed tomography (MDCT) imaging during the primary index occlusion and was similar between groups (A; ANOVA). Infarct size by pathology when normalized to left ventricular (LV) mass (B) or to CT AAR (C) was not different among groups (ANOVA). Solid black circles indicate data points for individual animals from each treatment group. Example images of MDCT derived AAR and pathological infarcts are illustrated above. All values are represented as means ± SE. CT AAR: vehicle, n = 9; CDCs, n = 7; and CDCs + cyclosporine, n = 7 animals. Pathology infarct size: vehicle, n = 9; CDCs, n = 10; and CDCs + cyclosporine, n = 10. CDCs, cardiosphere-derived cells.

Effect of CDCs and CDCs + Cyclosporine on LV Function and Hemodynamics

There were no differences in echocardiographic indices of regional or global function among treatment groups at any time point (Fig. 3 and Table 2). Regional LAD wall thickening fell from 63 ± 2% to 3 ± 1% 1 h after reperfusion (P < 0.0001) and was similar among treatment groups. Although regional LAD wall thickening increased in each group 1 mo after reperfusion (P < 0.05), it was similar among treatment groups (saline, 26 ± 8%; CDCs, 18 ± 4%; and CDCs + cyclosporine, 18 ± 5%; P = 0.56). Paired analysis of individual groups demonstrated a small increase in remote zone wall thickening in animals receiving CDC + cyclosporine (P < 0.05), but there were no significant changes in remote zone regional wall thickening among treatment groups at the end of the study (p-ns ANOVA). Echocardiography-derived LV ejection fraction values fell from 67 ± 1% to 51 ± 1% 1 h after reperfusion (P < 0.0001) with no differences among treatment groups. There were neither significant changes in ejection fraction in any group over time nor differences among treatment groups in the final study (saline, 51 ± 2%; CDCs, 48 ± 2%; and CDCs + cyclosporine, 51 ± 2%; P = 0.65; ANOVA). The echocardiographic estimates of EF were consistent with MDCT assessment of LV ejection fraction. A secondary analysis of the functional parameters with the addition of the three exclusions was the same (Supplemental Table S3).

Figure 3.

Echocardiographic measures of function across time. Cardiac function was assessed using 2-D echocardiography before ischemia, before and after cell treatment, and 1 mo after infarction. Ejection fraction (A) decreased similarly after ischemia-reperfusion (I/R) across all treatment groups and remained depressed 1 mo later. There was no significant difference among treatment groups (ANOVA). The wall thickening (WT) of the left anterior descending (LAD)-perfused myocardium (B) decreased after I/R and recovered slightly, albeit similarly in all groups after 1 mo (P < 0.05 vs. precell therapy, paired t test). The wall thickening of the remote noninfarcted myocardium (C) perfused by the posterior descending artery (PDA) was not different between groups at any timepoint before treatment but increased in the CDCs + cyclosporine group at 1 mo (P < 0.05 vs. precell therapy, paired t test). All values are represented as means ± SE. Vehicle, n = 9; CDCs, n = 10; and CDCs + cyclosporine, n = 10. CDCs, cardiosphere-derived cells; 2-D, two-dimensional. n = number of animals.

Table 2.

Echocardiographic indices of LV function at baseline and 1 mo in vehicle-, CDC-, and CDC + CsP-treated swine

	HR, beats/min	LAD WT, %	PDA WT, %	Indexed LVEDV, mL/kg	Indexed LVESV, mL/kg	LVEF, %
Baseline
Vehicle	99 ± 7	62 ± 5	129 ± 8	1.75 ± 0.11	0.59 ± 0.05	68 ± 1
CDCs	93 ± 5	62 ± 4	110 ± 9	1.72 ± 0.11	0.56 ± 0.07	68 ± 2
CDCs + CsP	96 ± 4	64 ± 3	121 ± 10	1.78 ± 0.05	0.61 ± 0.02	68 ± 1
20 min postreperfusion (precell)
Vehicle	89 ± 3	6 ± 3	86 ± 7	1.34 ± 0.11	0.69 ± 0.07	49 ± 2
CDCs	90 ± 6	2 ± 1	100 ± 12	1.28 ± 0.13	0.63 ± 0.08	51 ± 2
CDCs + CsP	81 ± 5	3 ± 2	94 ± 6	1.36 ± 0.11	0.70 ± 0.06	48 ± 2
1 h postreperfusion (postcell)
Vehicle	95 ± 4	5 ± 3	107 ± 6	1.45 ± 0.08*	0.73 ± 0.06	50 ± 3
CDCs	88 ± 5	3 ± 1	86 ± 7	1.33 ± 0.15	0.66 ± 0.10	52 ± 2
CDCs + CsP	85 ± 6	4 ± 2	99 ± 7	1.46 ± 0.09	0.73 ± 0.05	50 ± 2
1 mo postreperfusion
Vehicle	92 ± 7	26 ± 8*	100 ± 7	1.62 ± 0.12*	0.80 ± 0.08	51 ± 2
CDCs	79 ± 5	18 ± 5*	111 ± 10	1.79 ± 0.10*	0.94 ± 0.08*	48 ± 2
CDCs + CsP	86 ± 6	18 ± 4*	119 ± 11*	1.88 ± 0.10*	0.94 ± 0.09*	51 ± 2

Values are represented as means ± SE; n = 9 animals for vehicle; n = 10 for CDCs and CDCs + Cs. *P < 0.05 vs. 20 min postreperfusion. CDCs, cardiosphere-derived cells; CsP, cyclosporine treated; HR, heart rate; indexed LVEDV, indexed left ventricular (LV) end-diastolic volume; indexed LVESV, indexed LV end-systolic volume; LAD, region supplied by left anterior descending coronary artery; LV, left ventricle; LVEF, LV ejection fraction; PDA, remote regions supplied by the right coronary artery; WT, wall thickening.

Effect of CDCs and CDCs + Cyclosporine on Anatomic LV Remodeling by MDCT

The CT indices of LV mass and volume at baseline before infarction and 1 mo after reperfusion are summarized in Supplemental Table S3 and Fig. 4. Assessment of LV hemodynamics demonstrated no differences among treatment groups at baseline or at 1 mo after MI (Supplemental Table S4). Baseline EF was similar among groups as was EF 1 mo after reperfusion (saline, 49 ± 2%; CDCs, 46 ± 2%; and CDCs + cyclosporine; 47 ± 2%; P = 0.72; ANOVA). Body weight increased to a similar extent in all groups and, thus, absolute LV mass increased significantly secondary to growth. Despite this, there were no differences in indexed LV volumes among treatment groups at baseline or at 1 mo after infarction. When serial values were compared with baseline in individual groups, there were significant increases in indexed LV end-systolic volume in controls and those treated with CDCs, whereas it did not increase in animals receiving CDCs + cyclosporine. Although indexed LV mass tended to decrease slightly over the 1-mo timeframe in animals receiving CDCs + cyclosporine, there were no significant differences among treatment groups at the end of the study. A secondary analysis with the addition of the exclusions had the same results (Supplemental Table S3).

Figure 4.

Computed tomography (CT) measures of ejection fraction and left ventricular (LV) remodeling. Contrast-enhanced multidetector computed tomography (MDCT) imaging was performed at baseline and 1 mo to assess cardiac function and LV remodeling. Neither ejection fraction (A), indexed LV mass (B), end-diastolic volume (EDV; C), or end-systolic volume (ESV; D) was statistically different between treatment groups at baseline or 1 mo after acute MI (ANOVA). All values are represented as means ± SE. CT baseline: vehicle, n = 9; CDCs, n = 7; and CDCs + cyclosporine, n = 7. CT 1 mo: vehicle, n = 7; CDCs, n = 8; and CDCs + cyclosporine, n = 9. CDCs, cardiosphere-derived cells; LV, left ventricle; MI, myocardial infarction; LCX, left circumflex artery; RCA, right coronary artery. n = number of animals.

Effects of CDCs on Remote Zone Myocyte Cellular Remodeling

Postmortem histological samples were assessed for measures of remote zone myocyte cellular remodeling with the results summarized in Fig. 5. Myocyte nuclear density was no different in saline-treated or CDC-treated animals (saline, 798 ± 27; CDCs, 762 ± 44; and CDCs + cyclosporine, 821 ± 37 myocyte nuclei/mm²; P = 0.52). Likewise, myocyte diameter (saline, 17.9 ± 0.7; CDCs, 17.0 ± 0.5; and CDCs + cyclosporine, 18.0 ± 0.7 µm; P = 0.51; Fig. 5) was not different among treatment groups. Staining for the myocyte proliferative marker Ki67 showed an insignificant trend to increase when CDCs were administered with cyclosporine (saline, 1,304 ± 198; CDCs, 1,706 ± 486; and CDCs + cyclosporine, 2,484 ± 597 nuclei/10⁶ myocytes; P = 0.22; Fig. 5). Myocyte apoptosis rates by TUNEL staining were very low across all groups and not different (saline, 14 ± 7 versus CDC, 21 ± 11 versus CDC + cyclosporine, 9 ± 7 nuclei/10⁶ myocytes; P = 0.63; Fig. 5). Thus, allogeneic CDCs administered with or without cyclosporine had no impact on remote zone myocyte cellular remodeling.

Figure 5.

Remote histological measures of cellular remodeling, regeneration, and apoptosis. Cellular remodeling was assessed on paraffin-embedded, periodic acid Schiff (PAS)-stained sections processed from remote, noninfarcted myocardium. There were no changes observed with treatment in myocyte nuclear density (A) or cell diameter (B; ANOVA). In addition, immunostaining was performed on similar sections to investigate effects on cellular regeneration and apoptosis. Again, there were no differences in Ki67 (C) or TUNEL (D) staining, suggesting no significant effect on regeneration or apoptosis (ANOVA). All values are represented as means ± SE. Vehicle n = 9; CDCs, n = 10; CDCs + cyclosporine, n = 10. CDCs, cardiosphere-derived cells. n = number of animals.

DISCUSSION

There are two major findings from the present study that contrast with previous studies of allogeneic CDCs administered at the time of reperfusion from our laboratory (10) and others (7, 8). First, when allogeneic CDCs were administered to the entire heart immediately after reperfusion, they failed to reduce chronic infarct size or improve global or regional LV systolic function. There was also no effect on histological measures of myocyte survival and/or proliferation when measured 1 mo after myocardial infarction. The effects on function and infarct size are at odds with prior preclinical studies evaluating their effects after reperfusion but are compatible with recent human studies demonstrating the lack of effect of intracoronary allogeneic CDCs on LV function or infarct size when administered several weeks after myocardial infarction (12). Second, and somewhat surprisingly, the widespread administration of CDCs with intravenous cyclosporine immunosuppression initiated at the time of reperfusion did not reproduce the beneficial effects that we had previously demonstrated employing low-dose oral cyclosporine started 72 h before cell therapy (10). Collectively, these findings suggest that widespread infusion of allogeneic CDCs may only be a viable therapeutic approach when administered in the setting of chronic LV dysfunction (e.g., ischemic cardiomyopathy) where oral cyclosporine can be started well in advance of administering CDCs to the heart.

Comparison with Previous Studies Using Allogeneic CDCs Restricted to the Infarct-Related Artery

Prior studies have suggested that allogeneic CDCs are relatively immune privileged in a fashion similar to allogeneic mesenchymal stem cells, which has given rise to their potential use without the need to expand autologous cells ex vivo (20). This is relative rather than absolute since, in the absence of cyclosporine (9, 10), prior studies in large animals have failed to identify donor CDC retention at time points later than 1 mo after administration. Nevertheless, the allogeneic approach is attractive since it affords the opportunity to effect cardiac repair and prevent deleterious remodeling at the earliest phase after infarction by serving as an “off-the-shelf” intervention. Kanazawa et al. (7) used the stop-flow injection approach and reported reductions in myocardial infarct size (59% versus 80% of area at risk) 48 h after reperfusion following a 90-min LAD occlusion and termed the benefit “cellular postconditioning.” Nevertheless, despite reductions in infarct size and border zone apoptosis, ejection fraction did not increase. When assessed after 2 mo, a small reduction in infarct volume/area at risk persisted yet absolute infarct size was similar to controls despite a lower LV mass after CDCs (8). Although systolic LAD wall thickening increased over time in both groups, there were no differences in LV ejection fraction. A similar dissociation between a reduced infarct size and lack of significant increases in ejection fraction using the stop-flow technique and CDCs restricted to the LAD have been recently confirmed by our own group (10) as well as others (21). In the present study, global infusion of allogeneic CDCs under continuous flow had no effect on myocardial infarct size and was consistent with our previous findings employing CDCs with oral cyclosporine pretreatment (10). In contrast, Tseliou et al. (22) infused CDCs into a chronic infarct with as well as without stop-flow and found that the improvement in regional function and reduction in infarct size were comparable with both approaches. Taken together, these divergent results suggest that when CDCs are administered in the setting of an acute infarction, the stop-flow approach may be required to effect a reduction in infarct size. Whether this reflects improved initial CDC retention during the early healing phase or perhaps an impact of ischemic conditioning of CDCs to enhance in vivo viability will require further study.

Variable Effects of Immunosuppression on the Effects of Global Allogeneic CDC Infusion

Our previous studies of global allogeneic CDC infusion have all employed a regimen where animals are pretreated with low-dose cyclosporine for 3 days before CDC infusion. This was intended to prolong the duration of allogeneic CDC viability and, thus, the duration over which paracrine factors (including exosomes) could be released into the myocardium in vivo. Our previous results using low-dose cyclosporine pretreatment and allogeneic CDCs have been consistent among multiple swine models of ischemic heart disease. Results of these prior studies are compared with those of the present study where cyclosporine was initiated at the time of reperfusion in Fig. 6. Global infusion of allogeneic CDCs increased regional myocardial function in swine with chronic hibernating myocardium and a normal ejection fraction (6, 9). Global infusion of CDCs also improved ejection fraction with a reduction in myocardial fibrosis after coronary microembolization (23) and increased ejection fraction without a change in scar volume when administered during reperfusion after a myocardial infarction (10). A consistent observation in each of these animal models was that CDCs promoted favorable myocyte cellular remodeling in normally perfused myocardium remote from the region subjected to acute or chronic ischemia. This was characterized by a significant reduction in myocyte apoptosis and increases in myocyte nuclear proliferative markers such as Ki67 and phospho-histone-H₃. These opposing effects resulted in an increase in myocyte nuclear density and reduced myocyte cell size after CDCs as compared with vehicle controls. As a result, myocyte loss in regions remote from the ischemic region was reduced. Remote zone myocyte remodeling was absent following regional LAD CDC administration (10) and global CDC infusion without oral cyclosporine (22). Collectively, these observations support the notion that pretreatment with low-dose oral cyclosporine potentiates the reparative actions of CDCs when administered to the entire heart.

Figure 6.

Summary of the functional effects of widespread allogeneic CDCs in the present vs. prior studies from our laboratory. Intracoronary CDCs were administered 72 h after initiating cyclosporine in swine models of chronic left ventricular dysfunction [hibernating myocardium (6, 9) and chronic infarction from LAD microembolization (23)] and 30 min following acute reperfusion of an anterior myocardial infarction from an LAD occlusion (10). A: ejection fraction. B: ischemic zone wall thickening. C: remote zone wall thickening. There was functional improvement in terms of regional wall thickening and, when reduced below normal, increases in ejection fraction. In contrast, when intravenous cyclosporine was started immediately before reperfusion in the present study, there was no effect on regional function or ejection fraction. See main text for further discussion. CDCs, cardiosphere-derived cells.

Surprisingly, when cyclosporine was started in a translationally relevant approach using an intravenous loading dose immediately before reperfusion along with the same low-dose oral cyclosporine regimen that we used in these prior studies, the beneficial effects of allogeneic CDCs on systolic function and remote zone cellular remodeling were lost. The potential reasons that CDCs with cyclosporine initiated at reperfusion were ineffective are speculative, but several possibilities need to be considered. We based the intravenous dose of cyclosporine used on clinical studies of acute myocardial infarction (24, 25). Peak cyclosporine levels were high (>2,000) at the time of reperfusion and fell to the therapeutic range 60 min after reperfusion. It is possible that this high level resulted in toxicity to the CDCs, which is supported by the fact that we could not detect CDCs using PCR at the end of the study. This contrasts with their presence in prior studies employing oral cyclosporine pretreatment (6, 9). A second possibility is that intravenous cyclosporine alters the initial host immune response to acute myocardial infarction. Vagnozzi et al. (26) recently demonstrated that intramyocardial injection of bone marrow mononuclear cells after infarction creates a local inflammatory stimulus that attenuates postinfarction remodeling. These favorable effects could be blocked with cyclosporine immunosuppression or inhibiting macrophage polarization to a reparative phenotype (27). Finally, it is plausible that low-dose cyclosporine administered several days before infarction alters the T-cell immune response to myocardial injury in a fashion that favorably modifies the response to intracoronary allogeneic CDC therapy (28, 29). There are multiple immune mechanisms impacted by cyclosporine and low-dose cyclosporine has immunomodulatory actions that differ from the high doses used to prevent organ transplant rejection (30). Mechanistic studies in higher throughput murine models employing global arterial allogeneic CDC infusion and serial characterization of the T cell and macrophage response will be required to address these possibilities.

Relation to Prior Studies of Cyclosporine and Acute Infarct Size

Although our experiments were directed at studying the effect of CDCs in acute myocardial infarction, they also provide additional albeit indirect insight into the cardioprotective role of cyclosporine administered at the time of reperfusion. Since infarct size in animals with CDCs with or without cyclosporine was no different and similar to saline controls, our results also argue against an independent effect of cyclosporine on infarct size. The lack of effect on chronic infarct size is consonant with the lack of effect of cyclosporine on acute infarct size at doses of 2.5 mg/kg iv (31) and 10 mg/kg iv (32, 33) in swine by others. This contrasts with the brief report of Skyschally et al. (34) who found that an intermediate dose of cyclosporine (5 mg/kg iv) produced a small but significant reduction in infarct size during prolonged low-flow ischemia in open-chest swine. The lack of effect of cyclosporine in our study at the same dose of cyclosporine could relate to the absence of significant residual perfusion in our occlusion/reperfusion model and/or the chronic versus acute assessment of infarct size. Regardless, the absence of an effect on chronic infarct size is consonant with the negative Phase II human clinical trial of cyclosporine in reperfused ST elevation infarction (24).

Methodological Limitations

There are a number of methodological concerns that relate to the biological activity of our allogeneic donor CDCs. First, there is no standard accepted approach to characterize CDC bioactivity before their use in vivo. We used the same isolation and expansion procedures, employed three separate CDC donors, and performed flow cytometry on each to confirm that the cell surface markers were similar to what we have previously used in our four prior positive studies (Fig. 6). Nevertheless, we cannot definitively exclude the possibility that the bioactivity of these allogeneic donors was in some way different than those that we have isolated using the same techniques in prior studies. Second, we used atrial tissue as a source of CDCs, which produces a second variable that differs from our previous positive study (10), which employed CDCs isolated from ventricular tissue. Although prior studies from our laboratory and others have predominantly used ventricular biopsies to isolate CDCs, we have previously demonstrated the efficacy of allogeneic CDCs expanded from atrial tissue in chronic myocardial infarction from regional microembolization in swine pretreated with oral cyclosporine (23). In these studies, atrial CDCs increased regional function and ejection fraction (46.6% versus 36.0% in saline controls, P < 0.05) more prominently than in our studies with ventricular CDCs (summarized in Fig. 6). They also stimulated myocyte proliferative indices and reduced myocyte cellular hypertrophy in a fashion similar or greater than what we have found using CDCs isolated from ventricular tissue. Although suggestive of superior reparative properties, we do not have any head-to-head comparisons to strongly support our choice. Thus, although unlikely, we cannot exclude the possibility that atrial CDCs are not efficacious in acute myocardial infarction. Third, some studies have suggested an inverse relationship between the expression of CD90 and the in vivo functional effects of human and murine CDCs (35). Our present and prior studies in swine have consistently demonstrated high levels of CD90 (>90%). Despite this, they have elicited quite significant increases in regional and global function (Fig. 6), increases in myocyte nuclear density, and reductions in cellular hypertrophy. This suggests that there may be a species-specific difference in relation to the impact of CD90 on the therapeutic efficacy of CDCs. Although we cannot exclude the possibility that isolating the CD90 negative fraction may provide greater efficacy, we do not believe that this is a likely cause of our negative results. Finally, we did not characterize the immunosuppressive efficacy of intravenous cyclosporine, nor have we performed these studies in prior investigations using oral cyclosporine. Kanazawa et al. (8) found no donor antibodies or histological evidence of immune rejection in swine studied 2 mo after intracoronary CDCs in the absence of cyclosporine although CDCs were likely no longer present at this time point. Considering these negative data, we did not assess donor antibodies in our study. Thus, we cannot determine whether the absence of any CDC retention after 4 wk in this study versus the low rates of CDC retention seen at a similar time point with oral cyclosporine pretreatment in our prior studies reflects immune rejection, toxicity of the higher dose of intravenous cyclosporine on the CDCs, or another mechanism.

Conclusions

In summary, our study indicates that when administered at the time of reperfusion, widespread allogeneic intracoronary CDC infusion has no effect on myocyte proliferation or cellular hypertrophy and does not improve global function or reduce infarct size. This lack of effect was also found when CDC administration was preceded by cyclosporine initiated at the time of reperfusion. These findings indicate that infusion of intracoronary allogeneic CDCs may not be a viable therapeutic approach in the setting of an acute myocardial infarction. The contrast with the favorable results obtained when low-dose cyclosporine was initiated 72 h before allogeneic CDCs raises the possibility that cyclosporine treatment may be most useful for amplifying the effects of CDCs for the treatment of chronic left ventricular dysfunction. Further studies will be required to determine if oral cyclosporine pretreatment can overcome the recently reported lack of effect of allogeneic CDCs to patients with healed myocardial infarcts (12).

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author on reasonable request.

SUPPLEMENTAL DATA

Supplemental Tables S1–S4: https://doi.org/10.6084/m9.figshare.20341182.

GRANTS

This work was funded by National Institutes of Health Grants HL-055324, HL-061610, F32HL-114335, and UL1TR001412 (to J.M.C.); American Heart Association Grant 17SDG33660200 (to B.R.W.); U.S. Department of Veterans Affairs Grant 1IO1BX002659 (to J.M.C.); New York State Department of Health Grant NYSTEM CO24351 (to G.T.); and the Albert and Elizabeth Rekate Fund in Cardiovascular Medicine (to J.M.C.).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

G.T. and J.M.C.Jr. conceived and designed research; G.T., B.R.W., and R.F.Y. performed experiments; G.T., B.R.W., R.F.Y., and J.M.C.Jr. analyzed data; G.T., B.R.W., R.F.Y., and J.M.C.Jr. interpreted results of experiments; G.T., B.R.W., and J.M.C.Jr. prepared figures; G.T. and J.M.C.Jr. drafted manuscript; G.T., B.R.W., and J.M.C.Jr. edited and revised manuscript; G.T., B.R.W., R.F.Y., and J.M.C.Jr. approved final version of manuscript.