Intramyocardial Adipose-Derived Stem Cell Transplantation Increases Pericardial Fat with Recovery of Myocardial Function after Acute Myocardial Infarction

Intramyocardial injection of adipose-derived stem cells (ASC) with other cell types in acute myocardial infarction (AMI) animal models has consistently shown promising clinical regenerative capacities. We investigated the effects of intramyocardial injections of mouse ASC (mASC) with mouse endothelial cells (mEC) on left ventricular function and generation of pericardial fat in AMI rats. AMI rat models were created by ligating left anterior descending coronary artery and were randomly assigned into four groups: control (n = 10), mASC (n = 10), mEC (n = 10) and mASC+mEC (n = 10) via direct intramyocardial injections, and each rat received 1x106 cells around three peri-infarct areas. Echocardiography and cardiac positron emission tomography (PET) were compared at baseline and on 28 days after AMI. Changes in left ventricular ejection fraction measured by PET, increased significantly in mASC and mASC+mEC groups compared to mEC and control groups. Furthermore, significant decreases in fibrosis were confirmed after sacrifice on 28 days in mASC and mASC+mEC groups. Successful cell engraftment was confirmed by positive Y-Chromosome staining in the transplantation region. Pericardial fat increased significantly in mASC and mASC+mEC groups compared to control group, and pericardial fat was shown to originate from the AMI rat. mASC group expressed higher adiponectin and lower leptin levels in plasma than control group. In addition, pericardial fat from AMI rats demonstrated increased phospho-AMPK levels and reduced phospho-ACC levels. Intramyocardial mASC transplantation after AMI in rats increased pericardial fat, which might play a protective role in the recovery of myocardial function after ischemic myocardial damage.


Introduction
Ischemic cardiac diseases such as acute myocardial infarction (AMI) are a leading cause of death worldwide. Despite recent advances in clinical treatment for such diseases, the mortality rate associated with a failing heart remains very high. Inducing recovery of damaged myocardium has become an important therapeutic goal for treating AMI patients. Therefore, cardiac repair including regeneration of damaged left ventricular (LV) function through stem cellbased transplantation deserves much attention. Stem cell transplantation therapy has been used to treat failing hearts for more than a decade, and a large number of studies have provided profound evidence that implanted cells can participate in generation of new tissue to mechanically and biochemically support the damaged heart and thus enhance its function [1,2].
Several studies using various types of stem cells have shown promising results in halting myocardial damage and improving LV function [3,4]. Among many different stem cells, regenerative abilities of adipose-derived stem cells (ASC) are thought to be a practical cell source for cell-based transplantation due to ease of isolation from patients, less invasive collection, large quantities of autologous cells, multi-potency to differentiate into multi-lineages, developmental plasticity, and avoidance of ethical issues [5]. A number of preclinical studies have demonstrated treatment of damaged myocardium with ASC in combination with other cell types in order to maximize efficacy, and implanting ASC with endothelial cells (EC) provides efficient blood supply through formation of vasculature [6].
Increased pericardial fat is frequently observed in coronary artery disease (CAD) or in obese patients and has been implicated as an important risk factor of coronary atherosclerosis [7,8]. Excessive increase in pericardial fat layer is known to interfere with punctual contraction and relaxation of myocardium and is associated with deposition of excess cholesterol within underlying arteries. However, pericardial fat is also considered to exert local effects as a paracrine organ by secreting various adipokines, and low adiponectin levels in plasma has been associated with progression of CAD and higher incidence of AMI [9,10]. In this study, we investigated the effects of intramyocardial injections of ASC with EC and their effects on LV function after AMI in rats, in addition to the role of pericardial fat generated after ASC intramyocardial injections.

Animal care
All procedures were approved by the Korea University Institutional Ethics Committee for Animal Research (KUIACUC-2014-110). All animals were maintained in a specific pathogen-free (SPF) grade animal room at constant temperature of 22°C ± 1°C and 50% ± 5% humidity with 12-hour light/dark cycles. All researches complied with the guidelines from the US National Institutes of Health and Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes.

Cell sources
Mouse adipose-derived stem cells (mASC) were isolated from male ICR albino mice (Orient Experimental Animal Laboratory, Gyeonggi, Republic of Korea). The mice were given an intraperitoneal (IP) injection anesthesia with a mixture of ketamine (44 mg/kg; Yuhan) and xylazine hydrochloride (0.75 mg/kg; Rompun, Bayer), and euthanized. Mouse inguinal fat pads were dissected, minced, and digested with 0.1% Type I collagenase (#4197, Worthington Biochemical) for 45 mins at 37°C. The equal volume of culture medium containing 10% fetal bovine serum (FBS; #16000-044, Invitrogen) was added, followed by filtration through 100 μm nylon mesh (352360, BD). Then the samples were washed with PBS via centrifugation at 600 g, and the cells were suspended and incubated in 160 mM NH 4 Cl for 10 mins. The cells were plated in a 100-mm culture dish and maintained in Mesencult basal medium, supplemented with MSC stimulatory supplements (#05501, StemCell Technologies) and 100 unit/mL penicillin and 100 μg/mL streptomycin (P/S; #15140, Invitrogen). The cells were incubated at 37°C in a humidified incubator with 5% CO 2 . Culture media were changed daily and mASC were used at passage 3 to 5 for in vitro experiments and at passage 8 to 10 for in vivo experiments.
Mouse endothelial cells (mEC) isolated from the lungs of mice were purchased from Cell Biologics (#CD-1011L). Initially, 1 × 10 6 cells were plated in a 100-mm culture dish and cultured with Dulbecco's modified Eagle's medium-low glucose medium (DMEM-LG; #11885-092, Invitrogen), containing 10% FBS and P/S. Medium was changed every 2 days and mEC were used at passage 3 to 5 for in vitro experiments and at passage 9 to 11 for in vivo experiments.

AMI model and cell transplantation
Female Sprague-Dawley (SD) rats weighing 180 to 200 g were purchased from Orient Experimental Animal Laboratory (Gyeonggi, Republic of Korea). The SD rats were given an intraperitoneal (IP) injection anesthesia with a mixture of ketamine (60 mg/kg) and xylazine hydrochloride (7.5 mg/kg). An 18-gauge angiocatheter was utilized as an intubation tube throughout the procedure. The left coronary artery of the heart was ligated with a 6-0 silk suture, 5 mm from the left coronary atrial appendage. After confirmation of the presence of AMI, 1 × 10 6 of mASC, 1 × 10 6 of mEC, or 5 × 10 5 mASC + 5 × 10 5 mEC suspension in 100 μL of culture medium was injected at three peri-infarct areas. Culture medium in 100 μL was injected as a control treatment. After cell transplantation, the chest wall, muscle layers, and skin were closed with 3-0 silk sutures. All rats were continuously monitored after surgery. All rats were provided an IP injection of an analgesic agent, ketoprofen (5 mg/kg) after surgery and monitored at least twice a day for the first 7 days and at least once a day until 28 days after surgery. For sacrifice, rats were euthanized by IP injection with the mixture of ketamine (60 mg/kg) and xylazine hydrochloride (7.5 mg/kg).

Echocardiographic analysis
Echocardiography was performed with a commercially available echocardiography system (Vivid 7, GE Medical Systems) with a 10 MHz small linear array transducer (i13L, GE Medical Systems) [12]. Animals were anesthetized with a mixture of ketamine (60 mg/kg) via intramuscular (IM) injection. All measurements were taken over two consecutive cardiac cycles and averaged. All measurements were performed by an experienced cardiologist who was blinded to the study groups. Each value was the average of two measurements. The percentage of left ventricular fractional shortening (LFFS) was calculated using M-mode. Left ventricular ejection fraction (LVEF) was calculated as: LVEF (%) = (diastolic volume-systolic volume) / diastolic volume. End diastolic volumes and end-systolic volumes were calculated by manually drawing endocardial contours at end-diastolic and end-systolic phases in the apical two chambers view using the modified Simpson's rule. Moreover, wall motion score index (WMSI) was measured by echocardiography to assess functional improvement in segmental ventricular wall motion. Echocardiographic evaluations were repeated on 1 and 28 days.

Small animal positron emission tomography imaging
Cardiac positron emission tomography (PET) study was performed under anesthesia by 1-4% isoflurane inhalation (Forane, Choogwae) in oxygen with a small animal PET scanner (Inveon TM , Siemens Preclinical Solutions) during 20 mins after 1 hr uptake of 18 F-FDG. PET images were reconstructed using an ordered subset expectation maximization 2D algorithm. Rat PET cardiac images were reoriented to short-axis images and converted into DICOM format. Polar maps were generated by the QGS software (Cedars QGS 2008, Siemens) [13]. Myocardial infarct size was defined by the percent infarction area of the total LV myocardium. LVEF rate was calculated from gated PET images.

Histological examination after transplantation
At 28 days, rats were sacrificed and their hearts were fixed with 4% PFA and then embedded in paraffin wax for sectioning. Paraffin-embedded hearts were sectioned from the infarcted LV wall and the septum at 5 or 10 um thickness.
Masson's Trichrome (MT) staining was performed using the Trichrome stain kit (HT15) with following modifications: nuclei were stained with Celestine Blue solution followed by Gill's hematoxylin (GHS316) staining, and tissue was incubated for 1 hr in Bouin's solution (HT10132) before muscle staining with Biebrich scarlet-acid Fuchsin (HT151, Sigma). For fluorescence in situ hybridization (FISH), prepared tissue slides were dehydrated and incubated with a probe specific to whole mouse Y-Chromosome (Y-Chr; IDMR1056, Labs Biotechnology) for 10 hrs at 37°C in a humidified chamber. Tissues were then incubated with DAPI. Quantification was achieved using Image-Pro 7.0 software (Mediacybernetics).
Angiogenesis was assessed by staining thin sections with antibodies for smooth muscle αactin (α-SMA; A2547, Sigma). After immunostaining with antibodies for α-SMA, the sections were incubated with biotinylated anti-mouse IgG for 30 mins at RT, followed by ABC treatment (PK-6100). After color development with 3,3'-diaminobenzidine (DAB) substrate (SK-4100, both from Vector). The number of vessels on 30 sections (two fields per section, three sections per heart, n = 5 for each group) were counted α-SMA-positive vessels in the posterior wall, border, and infarct zone.
All digital images were obtained using a TE-FM Epi-fluorescence system attached to an Olympus BX61 inverted microscope. Images of sections were photographed at X400 magnification and analyzed by a trained blinded reader.
To evaluate FGF-21 levels secreted from mASC and mEC, 1. 5 X 10 5 of mASC and 1. 5 X 10 5 of mEC were plated on 6-well plates. After 24 hrs of culture, FGF-21 supernatant levels were determined by Mouse FGF-21 (MF2100, R&D Systems) Immunoassay kits. The ELISA reaction product was quantified by measuring absorbance at 450 nm and 540 nm using an ELISA reader (Spectra Max M2e), and data were analyzed using SoftMax Pro (both from Molecular Devices).

Quantification and statistical analysis
All statistical values were expressed as mean ± standard deviation (SD). Association was analyzed by Pearson correlation analysis for normal distribution or Spearman correlation analysis for non-normal distribution. Analysis of variance (ANOVA) was used to compare normally distributed data from all groups, and Ã p < 0.05 was considered to be statistically significant. Significant differences between means were determined by ANOVA followed by Student-Newman-Keuls test. All statistical analyses were performed using Sigma Stat 3.1 (Systat) and SAS 9.4 (SAS institute) software.

Evaluation of cardiac function after cell transplantation
Fifty-six rats were underwent AMI surgery. Among AMI induced rats, 10 rats for each group (control, mASC, mEC, and mASC+mEC) were assigned. In mortality and exclusion, 9 rats in control, mEC, and mASC+mEC groups (2, 4, and 3 rats, respectively) died during AMI surgery and post-operate care. Seven rats in control, mASC, mEC, and mASC+mEC groups (2, 1, 2, and 2 rats, respectively) were excluded due to an inordinately damage region (< 30% or > 60% of LVEF). Cardiac function was evaluated 28 days after transplantation of mASC and mEC into AMI induced rats (Fig 2A). Significant decreases in WMSI were observed at 28 days compared to 1 day in mASC and mASC+mEC groups (Fig 2B). In addition, mASC and mASC +mEC groups showed a significant increase in LVEF at 28 days (52.696 ± 1.849% and 50.520 ± 1.411%, respectively) compared to 1 day (44.505 ± 3.025% and 43.188 ± 3.743%, respectively). No statistically significant increase in LVEF was observed in control or mEC groups (Fig 2C) (S1A Fig).
Change of infarct size (%) was quantified based on images of PET scans taken at 1 and 28 days. Myocardial infarct size significantly decreased in mASC+mEC group compared to control and mEC groups. However, mASC group showed tendency of decreasing infarct size compared control group (Fig 2D and 2E). Increases in LVEF measured by PET were seen in mASC (9.1 ± 1.1%), mEC (5.0 ± 0.6%), and mASC+mEC groups (3.8 ± 2.4%) which were similar to echocardiography findings (Fig 2F). Cell transplantation therapies caused significant recovery of damaged myocardium compared to control group (1.1 ± 1.5%). Cross-sectional views from PET showed decreases in infarct size in mASC and mASC+mEC groups (S1B Fig). Comparison of fibrosis and α-SMA-positive vessel densities MT staining showed significant reduction in fibrosis after AMI in mASC (21.77 ± 4.62%) and mASC+mEC groups (14.74 ± 1.24%) compared to control group (27.43 ± 4.83%) (Fig 3A and  3B). Moreover, mASC which were obtained from male mice were detected in the recipient female rats by fluorescence staining for Y-Chr (Fig 3C).
α-SMA-positive vessels were counted at 3 areas: the posterior wall, border, and infarct zone. Significantly increases in the number of α-SMA-positive vessels were found in mASC (22.00 ± 1.70) and mASC+mEC groups (21.30 ± 2.41) compared to control (14.30 ± 2.71) in the border zone (Fig 3D and 3E). The numbers of α-SMA-positive vessels in mASC group were significantly increased compared to mEC group (18.60 ± 1.43) in the border zone ( Fig  Fig 1. Characterization (Fig 3E). These results indicate that intramyocardially injected mASC and mASC+mEC promote local angiogenesis in the border zone.

Increased pericardial fat after mASC transplantation
Pericardial fat observed at sacrifice was specifically located around the intramyocardial injection areas (Fig 4A), and MT staining confirmed adipose tissue in those areas (Fig 4B). Quantification of fat area based on H&E images demonstrated that mASC and mASC+mEC groups resulted in significant increases in pericardial fat compared to control (4.71-fold and 1.70-fold, respectively) or mEC groups (9.49-fold and 3.44-fold, respectively) ( Fig 4C). Pericardial fat was positively stained for FAT/CD36, which is a known positive antigen for fat ( Fig 4D). Pericardial fat weight was significantly higher in mASC and mASC+mEC groups (S2A Fig). However, there were no significant differences in body weight or heart weight of rats at sacrifice among four groups (S2B and S2C Fig). We found that pericardial fat weight showed significantly positive correlations with the parameters of cardiac function including LVEF and LVFS in control, mASC, mEC, and mASC+mEC groups (Fig 4E and S1 Table).
Flow cytometry with specific antibodies against fat surface antigens was used for identification of pericardial fat origin. Cells stained with mouse specific antibodies were not detected in pericardial fat (Fig 4F), but positive cells stained with rat specific antibodies such as CD44, CD54, CD106, and CD140b were found in pericardial fat (Fig 4G). FISH analysis did not identify a Y-Chr in pericardial fat tissue (data not shown). Thus, proliferated pericardial fat did not originate from the transplanted cells.

Analysis of adipokines after cell transplantation
A significant increase in plasma adiponectin levels between day 28 and 1 after AMI in mASC +mEC group (0.31 ± 0.021 ng/mL) was significantly higher than that of control group (0.20 ± 0.025 ng/mL) (Fig 5A). However, an increase in plasma adiponectin levels between day 28 and 1 after AMI in mASC group (0.31 ± 0.25 ng/mL) was not reached to the level of significance compared to that of control group (0.20 ± 0.02 ng/mL) may be due to a high variance (Fig 5A). Plasma adiponectin levels (1.607 ± 0.274 ng/mL) in the healthy rats were higher than plasma adiponectin levels (1.360 ± 0.214 pg/mL) at day 1 after AMI, but not significant. Significant increases in plasma leptin levels between day 28 and 1 after AMI in mASC (0.08 ± 0.003 pg/mL) and mASC+mEC group (0.09 ± 0.005 pg/mL) were found compared to that of control group (0.16 ± 0.025 pg/mL) (Fig 5B). Plasma leptin levels (0.178 ± 0.015 pg/mL) in the healthy rats were similar to plasma leptin levels (0.137 ± 0.016 pg/mL) at day 1 after AMI. Underlying mechanisms involved in the production of pericardial fat were further investigated. AMPK is an important molecule implicated in adipogenesis. pAMPK was significantly increased in mASC and mASC+mEC groups, and this increase in pAMPK levels are presumably influenced by various secreted adipokines. This active pAMPK is thought to inhibit ACC activity, thereby increasing pACC levels in mASC and mASC+mEC groups (Fig 5C). FGF-21, a master transcriptional regulator of adipogenesis, played a key role in insulin-sensitizing process in an adiponectin-dependent manner [15]. Therefore, we evaluated FGF-21 levels secreted from mASC and mEC by ELISA. Interestingly, we found that mASC abundantly secreted FGF-21 whereas mEC did not secrete FGF-21 ( Fig 5D).

Discussion
In a recent study, transplantation into cardiac disease models using ASC as well as mixtures of various cell sources has been attempted and yielded encouraging results [16]. ASC are well known for their superior capacity of vessel reparability and regeneration of myocardium after cell transplantation [17,18]. We further investigated the synergistic reparative capacity of mASC+mEC in this study, and intramyocardial injections of both mASC and mASC+mEC into infarcted myocardium evidently resulted in increased LV contractile function, supported by PET and echocardiography analysis. PET was performed in this study to evaluate an hearts. n = 8, 10, 10, and 8 rats in each group, *p < 0.05. (C) EF was determined from 2D echocardiography. n = 8, 10, 10, and 8 rats in each group, *p < 0.05. (D) Representative image of a polarmap using the 20-segment Cedar-Sinai method with 18 F-FDG PET. n = 4, 4, 4, and 5 rats in each group. (E) Changes in infarct size according to 18   objective quantitative assessment of LV function. However, the difference between echocardiography and PET results in mEC is due to excellent sensitivity and better spatial resolution of PET than echocardiography [19,20]. These findings suggest that mASC are more dominantly involved in repair processes and exert superior regenerative capacity in damaged myocardium than mEC, which is terminally differentiated cells. Improvements in LV function after intramyocardial injection of mASC are supported by the reduced fibrosis with increases in small vessels densities observed in this study on 28 days.
It is generally accepted that pericardial fat is highly linked to impairment of decreased cardiac function in humans [21,22]. Interestingly, pericardial fat increased prominently in mASC transplantation groups in the present study. Greater amounts of pericardial fat have generally been associated with increased risk for cardiovascular diseases [23,24]. Even so, compensatory increases in pericardial fat after intramyocardial ASC injections could play a different role, such as facilitating improvement of LV function as seen in this study. By supporting the energy needs of arterial walls and myocardium and secreting fatty acids that prevent lipotoxicity [25], pericardial fat near ischemic tissue could reduce the degree of hypertrophy of the heart and thus facilitate a protective role in ischemia. The positive effects of increased adiponectin levels have been reported as enhancement of cardiac function in anesthetized pigs [26], and treatment with exogenous adiponectin which significantly reduces the infarct size in myocardial ischemia and reperfusion [27,28]. Furthermore, high expression of adiponectin in patients with cardiac disease has been shown to provide protective local actions on adjacent coronary arteries [29].
In this study, transplantation of mASC not only increased the delta values of plasma adiponectin concentration, but also reduced the delta concentration in plasma leptin levels compared to control group. Our observation coincides with previous reports of correlations between plasma adipokine levels and heart function. A positive correlation between increases in plasma adiponectin levels and increases in hyperemic myocardial blood flow was found in morbidly obese individuals with surgical gastric bypass-induced weight loss [30]. Fontes-Carvalho et al. [31] also reported that patients with normal diastolic function had significantly lower leptin plasma levels compared to patients with diastolic dysfunction. In addition, increased leptin levels could result in hypertrophy of cardiomyocytes, increased incidence of AMI, aggravated atherosclerotic progression, or retarded recovery of damaged myocardium [32].
In terms of correlation between adipose tissue and cardiac function, Yuhki et al. [33] indicated that adiponectin secreted from adipose tissue activated AMPK signaling pathway, which increased energy supply resulting in preventing cell death in the ischemic heart, suggesting that adiponectin is a promising new therapy for AMI. Accordingly, it was necessary to study downstream events on cardiovascular system in response to changes of adiponectin and leptin levels secreted by pericardial fat. The changes in adiponectin and leptin levels were implicated by AMPK pathway and confirmed by Western blot results for AMPK and ACC. In this study, we observed that pAMPK levels in pericardial fat tissue were significantly increased as a result of mASC and mASC+mEC transplantation, indicating positive association with higher circulating adiponectin levels. In accordance with our observation, treatment of pioglitazone on obese rats has been shown to attenuate LV hypertrophy, fibrosis, and diastolic dysfunction by activating cardiac AMPK signaling as a result of the stimulation of adiponectin secretion [34]. Pei et al. [35] also reported that cardiac-derived adiponectin induced by long-term insulin treatment ameliorated myocardial ischemia/reperfusion injury in type 1 diabetic mice via myocardial AMPK activation. In addition, adiponectin improved cardiac function through antiapoptotic effects caused by up-regulation of AMPK in doxorubicin-induced cardiomyopathy in mice [36]. A number of studies have reported that AMPK phosphorylates both ACC1/2 on serine residues, resulting in inactivation of ACC for control of lipid metabolism [37,38]. Similarly, we also found that phosphorylation of ACC in pericardial fat tissues significantly increased in mASC and mASC+mEC groups, demonstrating that ACC is a downstream target of AMPK [39,40]. It seems that implantation of ASC results in increased pericardial fat that secretes higher levels of adiponectin. Higher adiponectin secretion activates pAMPK in the infarcted heart, and in turn activated ACC is inactivated via phosphorylation induced by pAMPK [41]. Regulation of fatty acid synthesis and oxidation via AMPK results in the enhancement in heart function [42].
To confirm that the engraftment was successfully introduced, the infarcted myocardium was subjected to fluorescence staining for Y-Chr. Although direct intramyocardial injection is known for its low retention rate and high death rates of transplanted cells, we detected several transplanted mASC by extensive FISH analysis within the peri-infarct areas on 28 days. Cell fusion between donor cells and transplanted cells was not observed in this study. Anti-rat specific antibody binding was detected in pericardial fat, while anti-mouse specific antibody binding was not, indicating that pericardial fat generated after cell transplantation originated from the host rat and was not proliferated tissue from the donor cells. These results suggest that injected mASC exert positive paracrine effects on the infarcted pericardium with increased production of a protective layer, ultimately resulting in improved myocardial function.
In this study, we found that mASC abundantly secreted FGF-21 whereas mEC did not secrete FGF-21 ( Fig 5D). It is quite possible that mASC-secreted FGF-21 could contribute to formation of pericardial fat after mASC injection, finally resulting in the improvement of cardiac function. In accordance with these results, increased pericardial fat weight via transplantation of mASC showed a significant positive association with the improvement in cardiac function such as LVEF and LVFS measured by echocardiography at 28 days (Fig 4E). The correlation of pericardial fat with body weight, heart weight, or deposition of visceral fat was examined, but these did not significantly affect each other in this study.
In our study, although xenograft transplantation was performed by implantation of mASC into AMI rat models, we considered immunosuppression overcome by following reasons. Xenograft transplantation of MSC including ASC facilitates homing of cells to the ischemic site and induces immune tolerance to suppress rejection response to xenograft transplantation [43,44]. However, recipient immune responses after ASC and EC transplantation into AMI models of immunodeficient animals should be further investigated. In the future, additional investigation about changes in various adipokines such as resistin, visfartin, chemerin, and apelin in plasma should be helpful.

Conclusions
In conclusion, this study demonstrated enhanced LV function upon intramyocardial injection of mASC after AMI in rats with increased pericardial fat. Increased circulating adiponectin levels together with decreased leptin levels after intramyocardial ASC injections suggest beneficial paracrine effects of secondary increase in pericardial fat in AMI models. Increased adiponectin secretion from pericardial fat with its association with enhanced LV function could provide new clinical implications of ASC transplantation in AMI. Comparisons between pericardial fat weights at 28 days after cell transplantation of all groups. Significantly increased pericardial fat weight was observed in mASC and mASC+mEC groups compared with control and mEC groups. A dotted line indicates the value of healthy rats (0.06 ± 0.01 g). n = 5 rats in each group, Ã p < 0.05. (B) No significant differences in heart weight were observed between any groups at 28 days after cell transplantation. A dotted line indicates the value of healthy rats (0.86 ± 0.04 g). n = 5 rats in each group, P-value not significant (NS). (C) Comparison of body weight between 0 and 28 days. No significant differences in body weight change were observed in any groups. A dotted line indicates the value of five healthy rats (208.9 ± 3.20 g and 247.5 ± 8.90 g). n = 10 rats in each group, NS. (TIF) S1 Table. Correlations of pericardial fat weight with cardiac function. (TIF)