Author + information
- Received February 24, 2012
- Revision received April 11, 2012
- Accepted April 24, 2012
- Published online October 23, 2012.
- Hye Jin Hwang, MD⁎,†,
- Woochul Chang, PhD‡,
- Byeong-Wook Song, MS†,§,
- Heesang Song, PhD∥,
- Min-Ji Cha, MS†,§,
- Il-Kwon Kim, PhD†,§,
- Soyeon Lim, PhD¶,
- Eun Ju Choi, BS†,§,
- Onju Ham, BS†,§,
- Se-Yeon Lee, BS†,§,
- Jaemin Shim, MD⁎,†,
- Boyoung Joung, MD, PhD⁎,†,
- Hui-Nam Pak, MD, PhD⁎,†,
- Sung Soon Kim, MD, PhD⁎,†,
- Bum-Rak Choi, PhD#,
- Yangsoo Jang, MD, PhD⁎,†,§,⁎⁎,
- Moon-Hyoung Lee, MD, PhD⁎,†,⁎ ( and )
- Ki-Chul Hwang, PhD†,§,⁎⁎,⁎ ()
Objectives The purpose of this study was to evaluate the antiarrhythmic potential of mesenchymal stem cells (MSC) under a different environment.
Background Little is known about how environmental status affects antiarrhythmic potential of MSCs.
Methods To investigate the effect of paracrine factors secreted from MSCs under different circumstances on arrhythmogenicity in rats with myocardial infarction, we injected paracrine media (PM) secreted under hypoxic, normoxic conditions (hypoxic PM and normoxic PM), and MSC into the border zone of infarcted myocardium in rats.
Results We found that the injection of hypoxic PM, but not normoxic PM, markedly restored conduction velocities, suppressed focal activity, and prevented sudden arrhythmic deaths in rats. Underlying this electrophysiological alteration was a decrease in fibrosis, restoration of connexin 43, alleviation of Ca2+ overload, and recovery of Ca2+-regulatory ion channels and proteins, all of which is supported by proteomic data showing that several paracrine factors including basic fibroblast growth factor, insulinlike growth factor 1, hepatocyte growth factor, and EF-hand domain-containing 2 are potential mediators. When compared with PM, MSC injection did not reduce or prevent arrhythmogenicity, suggesting that the antiarrhythmic or proarrhythmic potential of MSC is mainly dependent on paracrine factors.
Conclusions A hypoxic or normoxic environment surrounding MSC affects the type and properties of the growth factors or cytokines, and these secreted molecules determine the characteristics of the electro-anatomical substrate of the surrounding myocardium.
Advances in stem cell biology provide a basis for potential cell therapy to cure myocardial infarction (MI). In spite of early success reports, there are inconsistencies regarding the therapeutic benefits of mesenchymal stem cells (MSC)–based therapy for MI (1). Cellular heterogeneities caused by stem cell injection may increase arrhythmogenesis (2,3), and growth factors and cytokines released from transplanted MSC may be beneficial (4,5). Although stem cell therapy has emerged as a promising option to repair impaired contractile function, confounding issues concerning the proarrhythmic and antiarrhythmic potentials of stem cell therapy have led to intense discussion.
The proarrhythmic potentials by MSC can be attributed to tissue heterogeneity by transplantation of unexcitable MSC (2), possible heterogeneous sympathetic nerve sprouting (6), teratoma formation, and spontaneous focal activity. The inconsistent results may be due to varying circumstances surrounding implantation sites. However, little is known about how environmental status affects the potential role of stem cells, particularly with respect to their ability to prevent arrhythmogenicity.
This investigation focused on whether paracrine factors released from MSC could modulate the electrophysiological properties of the border or ischemic zone, creating spontaneous focal activity through impaired intracellular calcium handling (7,8) or reentrant arrhythmia (9). The present study also sought to evaluate the direct effect of MSC by comparing MSC-engrafted rats with rats treated only with paracrine factors. We found that MSC produced beneficial antiarrhythmic paracrine factors under the hypoxic condition, providing new insights for potential role of environment to activate stem cells.
Preparation of paracrine media
Paracrine media (PM) was prepared as follows: 90% confluent second passage MSC were fed with serum-free Dulbecco Modified Eagle Medium. MSC (1 × 106cells) were cultured for 12 h under hypoxia or normoxia. The media of MSC cultured under hypoxia was defined as hypoxic PM and media cultured under normoxia as normoxic PM, and both were assumed to contain various paracrine molecules. Hypoxic PM and normoxic PM were collected. Hypoxic conditions were created by incubating cells at 37°C in an anaerobic system (Technomart Inc., Seoul, South Korea) with a 5% CO2, 5% H2, and 85% N2 atmosphere, and a chamber oxygen level of 0.5%. A colorimetric redox indicator solution (resazurin, 89%, Sigma-Aldrich, St. Louis, Missouri) was used to ensure that hypoxic conditions were maintained. For injection, conditioned medium from 1 × 106 cells was centrifuged to remove cell debris and concentrated by VIVASPIN6 (Vivascience Ltd., Scottsdale, Arizona). The medium was continuously centrifuged a 1,000× g for 30 min at 4°C using an ultracentrifuge.
MI induction and treatment of hypoxic PM, normoxic PM, and MSC
All animal experimental procedures were approved by the Committee for Care and Use of Laboratory Animals, Yonsei University College of Medicine, and performed in accordance with the Guidelines and Regulations for Animal Care. MI was produced in male Sprague-Dawley rats (250 to 300 g) by left anterior descending coronary artery ligation, as described. Briefly, after anesthesia with ketamine (100 mg/kg) and xylazine (5 mg/kg), hearts were exteriorized by opening the chest at the third and fourth ribs. After 1 h of occlusion, the infarcted heart was reperfused and sham (saline), hypoxic PM, normoxic PM, or MSC were injected at the border region. Hypoxic PM or normoxic PM of MSC (1 × 106 cells) was enriched to 100 μl and injected into the infarct border region in 3 different sites using a syringe with a 30-gauge needle. For MSC transplantation, cells (1 × 106 cells) were suspended in 10 μl of serum-free medium and injected into the border in the same way. Rats that expired during the procedure or immediately after cell implantation were excluded for sudden death mortality. For identification of arrhythmic death as the cause of death, rats were monitored by telemetry (Online Fig. 1) and autopsies were performed with these rats to confirm that the animals did not expire from cardiac rupture. Experimental groups were used for surface electrocardiography, telemetric monitoring, in vivo isoproterenol test, optical mapping, Millar catheterization, and morphologic analysis at 11 days after injection of hypoxic PM, normoxic PM, or MSC.
For optical mapping, hearts were excised, retrogradely perfused through the aorta with Tyrode solution, gassed with 95% O2 and 5% CO2, and stained with a voltage sensitive dye, di-4-ANEPPS (Invitrogen Corporation, Carlsbad, California). Hearts were placed in a chamber to maintain 37.0 ± 0.2°C, and 5 μmol/l blebbistatin was added to the perfusate to reduce movement artifacts. Excitation light was delivered by epi-illumination with 2 green light-emitting diode lamps (500 ± 30 nm, LL-50R30-G25, Optronix, Seoul, Korea). Fluorescence images from the anterior surface of the heart were recorded with a charge-coupled device camera (128 × 128 pixels, Dalsa Inc., Billerica, Massachusetts) and the field of view was 1.0 × 1.0 cm2 with a spatial resolution of 78 × 78 μm2. The sampling rate was 420 frames/s. The rhythm of optical recordings was continuously monitored by electrocardiograms obtained with widely spaced bipoles, 1 at the apex of the left ventricle and the other at the high lateral wall of the right ventricle, using a Biopac System (BIOPAC Systems Inc., Goleta, California). Fluorescence recordings were filtered with a Gaussian filter (3 × 3 pixels) in the spatial domain, and first-order derivatives (dF/dt) were calculated using a polynomial filter (third order, 13 points) in the temporal domain. Activation time points at each site were determined from (dF/dt)max, and isochronal maps of activation were generated. The duration of the action potential at 80% repolarization (APD80) was measured. Conduction velocity (CV) within the border or normal regions was measured under point stimulation at a cycle length of 280 ms for 20 beats. Local CV vectors were estimated from each pixel's 7 nearest neighbors in the activation time of its temporal wave. To improve the signal-to-noise ratio, the local CV was spatially filtered using a 7 × 7 nearest-neighbor Gaussian convolution kernel and the acute change in its magnitude was suppressed using log-transformation (10,11). A CV map of the anterior surface of the heart focused by charge-coupled device camera was derived from local CV. Border zones were identified from activation map where slow conduction occurred. Infarct zones were defined as the area where the voltage signals were near completely absent. Analyses were performed with software using IDL (version 6.4, Research Systems, Boulder, Colorado) and MATLAB (Mathworks, Inc., Natick, Massachusetts).
Data are expressed as mean ± SE for continuous variables and as proportions for categorical variables. Statistical analyses of more than 2 groups were done by 1-way analysis of variance for parametric variables. Bonferroni adjustment was used for multiple comparisons. Premature ventricular contractions (PVC) and CV with nonparametric distribution were analyzed by the Kruskal-Wallis test followed by post hoc testing via Mann-Whitney U test with Bonferroni adjustment for multiple testing. The chi-square test or the Fisher exact test for categorical variables was used. Survival was analyzed by the Kaplan-Meier method with the Breslow test. A p value of <0.05 was considered statistically significant. The SPSS statistical package (version 18.0, SPSS Inc., Chicago, Illinois) was used. Details were described in Online Appendix.
Cell culture, in vivo, and in vitro analyses
These methods are provided in detail in the Online Appendix.
Paracrine factors from MSC under hypoxic condition prevent sudden death in infarcted rats
The risk of sudden death is highest in the early phase of MI (12), so we evaluated survival rates over 11 days after injury and treatment of sham (n = 23), PM secreted from MSC under hypoxia (hypoxic PM group; n = 22), PM secreted under normoxia (normoxic PM group; n = 23), and MSC (n = 22). As shown in Figure 1A, Kaplan-Meier survival curves revealed a marked reduction in sudden deaths in hypoxic PM–treated rats compared with sham-injected rats (13% in hypoxic PM group vs. 57% in sham group, p < 0.01) or normoxic PM–treated rats (13% in hypoxic PM group vs. 39% in normoxic PM group, p < 0.05). This finding suggests that paracrine factors secreted from MSC might have different protective effects, especially with respect to arrhythmogenicity, depending on their environment. MSC-transplanted (32%, n = 22) or normoxic PM–treated rats (39%, n = 23) showed a slight decrease of death, but not significant, compared with sham-injected rats (p > 0.05).
Hypoxic PM enhances electrical stability
We studied animals using left ventricular catheterization (Online Table 1), electrocardiogram (Online Table 2), or telemetric monitoring. We found improved left ventricular contractile function and shorter QRS duration in hypoxic PM–treated rats than in sham-injected rats. In addition, PVC tended to be suppressed by hypoxic PM treatment (Online Fig. 2). We further examined the protective effect of paracrine factors with isoproterenol infusion, which facilitates both early afterdepolarization and delayed afterdepolarization (13). For 15 min after intraperitoneal injection of isoproterenol (2 mg/kg), sustained ventricular tachyarrhythmia (VT) was induced in 1 of 9 sham-injected rats, but not in the hypoxic PM–, normoxic PM–, and MSC-treated rats. Frequent PVC were induced in the sham-injected animal group, whereas PVC were significantly reduced in hypoxic PM–treated rats (Figs. 1B and 1C).
To further investigate vulnerability to arrhythmias of infarcted hearts injected with paracrine factors, we performed an ex vivo electrical vulnerability test using a burst-pacing protocol. No monomorphic or polymorphic VT was observed in noninfarcted control rats (n = 10). However, VT were induced in 81.8% of sham-injected rats (n = 11), indicating that localized MI markedly enhances induction of VT, allowing a systematic evaluation of how PM secreted from MSC affects electrical vulnerability. VT occurred in 16.7% of 12 hypoxic PM–treated rats (sham vs. hypoxic PM, p < 0.01), in 58.3% of 12 normoxic PM–treated rats (normoxic PM vs. hypoxic PM, p < 0.05), in 46.7% of 15 MSC-treated rats (MSCs vs. hypoxic or normoxic PM, p > 0.05) (Fig. 1D). VT vulnerability score was also different between hypoxic- and normoxic-PM treatments, data not shown. In an analysis of covariance adjusted for left ventricular ejection fraction, the strong association between VT vulnerability and PM treatments was noted (Online Fig. 3). This supplied evidence of a significant difference of antiarrhythmic role of PMs secreted from different environments.
Hypoxic PM improves conduction through border zone
We assessed the electrophysiological impact of paracrine factors on the infarcted myocardium by optical mapping using Langendorff perfusion at 7 to 11 days after injury and treatment. Wave propagation in sham-injected hearts was prominently blocked at the boundaries of the infarcted regions (Online Fig. 4). In contrast, action potentials initially circumventing the infarct region propagated transmurally in hypoxic PM–treated hearts (Online Fig. 4). Focal activity arising from the infarct border is a representative electrophysiological feature that triggers arrhythmia in MI (7,8). Spontaneous ectopic beats emanating from the border region at sinus rhythm were observed frequently in 55.6% of sham-injected rats (n = 9) (Fig. 2A). However, treatment with hypoxic PM markedly suppressed the ectopic focal beats from the border region (0%, n = 6), with normoxic PM treatment (33.3%, n = 6) resulting in moderate suppression. In MSC-treated hearts, the ectopic focal beats were also moderately suppressed (25%, n = 8), to similar extent compared with the beats of normoxic PM–treated hearts. Local CV was markedly depressed in the border zone of sham-injected hearts and moderately in normoxic PM–treated border, in contrast to the notable restoration of CV seen in hypoxic PM–treated hearts (Fig. 2B). Figure 2C presents the different regional distributions of CV in hypoxic PM- and normoxic PM–treated hearts. CV was also moderately restored in MSC-treated border zone to a similar extent compared with normoxic PM–treatment (Fig. 2B). In contrast, APD80 was significantly different between the MSC-transplanted border zones and normoxic PM–treated border zones (Fig. 2D, Online Fig. 5). To identify the electrophysiological alteration due to MSC implantation, we confirmed the presence of 4′,6-diamidino-2-phenylindole–labeled MSC at the injected sites (Fig. 2D).
Hypoxic PM restores conduction by a decrease in fibrosis and recovery of connexin 43
This series of experiments proposes that the enhanced conduction resulting from treatment with hypoxic PM could be attributed to the rescue of hypoxic cardiomyocytes from apoptosis and the subsequent decrease in fibrous tissue (14), which caused slow conduction by a zigzag course of activation, thus facilitating reentry (9). Deoxyuride-5′-triphosphate biotin nick end labeling–positive myocardial cells were significantly reduced by approximately 50% in hypoxic PM–treated hearts compared with sham-injected hearts (Fig. 3A). Fibrosis was significantly decreased in hypoxic PM–treated hearts and not noticeably decreased in normoxic PM–treated hearts (Fig. 3B). The degree of fibrosis in MSC-transplanted rats was similar to that in normoxic PM–treated rats (15.1 ± 2% in MSC-transplanted hearts vs. 15.3 ± 1.5% in normoxic PM–treated hearts, p > 0.05), implicating that the effects of paracrine action of MSC in this model might be similar to the effects of normoxic PM. Decreased CV in ischemia is also induced by alterations in gap junction conductance (15). Figure 3C showed that total connexin 43 (Cx43) signal in the border zone of a post-infarct heart was decreased and was not in the intercalated disk area, and dephosphorylated Cx43 was increased. Hypoxic PM ameliorated dramatically the distribution of total Cx43, whereas the disarray of Cx43 was partially restored by normoxic PM or MSC injection (Fig. 3C). Moreover, dephosphorylated Cx43 fluorescence intensity was moderately decreased with normoxic PM or MSC treatment but remarkably attenuated by hypoxic PM treatment (Fig. 3C). Total amount of Cx43 decreased in the sham-injected area and moderately decreased by the normoxic PM–treated border zone, but increased in the hypoxic PM–treated border zone (Fig. 3D).
Hypoxic PM modulates Ca2+-regulatory ion channels and proteins
Alteration of Ca2+-regulatory ion channels and proteins in the border zone contribute to triggering easily focal activity and may affect electrical conduction (16,17). We observed a marked decrease in messenger ribonucleic acid (mRNA) levels of L-type calcium channel (LTCC), sarcoplasmic/endoplasmic reticulum calcium adenosine triphosphatase 2a, Na+/K+ adenosine triphosphatase, and the calcium-binding proteins, calreticulin and calmodulin in the border zone, which was significantly restored in hypoxic PM–treated area and moderately in normoxic PM–treated area (Fig. 4A). The increased mRNA level of the Na+/Ca2+ exchanger in the border zone, of which increased activity lead to delayed afterdepolarization, was completely rescued by hypoxic PM and partially by normoxic PM (Fig. 4A). We did not assess the alteration of Ca2+-regulatory ion channels and proteins in MSC-transplanted border zone, because ion channels of MSC are known to be different (18).
Different action of hypoxic PM and normoxic PM in vitro study
For the in vitro study, cardiomyocytes (2 × 106) were subjected to 3 h of hypoxia followed by reperfusion and treatment of hypoxic or normoxic PM. We also evaluated survival and change of Cx43 and dephosphorylated Cx43 in ischemic cardiomyocytes with no treatment, treated by hypoxic or normoxic PM, which was consistent with the in vivo study results (Online Figs. 6 and 7).
Next, alterations in impaired intracellular calcium were examined, because focal activity is easily induced by Ca2+ overload in an ischemic setting. A significant increase in fluorescence intensity was seen in hypoxic cardiomyocytes, indicating Ca2+ overload. However, hypoxic PM decreased the fluorescence intensity by 65% compared with that for control subjects and produced a 2-fold decrease compared with that in subjects treated with normoxic PM (Fig. 4B). In addition, we observed mRNA expression levels of LTCC, sarcoplasmic/endoplasmic reticulum calcium adenosine triphosphatase 2a, Na+/K+ adenosine triphosphatase, calreticulin, calmodulin, and Na+/Ca2+ exchanger, which were consistent with those in the in vivo study (Online Fig. 8).
Different expression of paracrine factors secreted under hypoxic and normoxic environment
We then performed proteomic analysis to identify the paracrine factors that were responsible for the differences between hypoxic PM and normoxic PM. As shown in Online Table 3, the secreted levels of the growth factors basic fibroblast growth factor (bFGF), insulin-like growth factor (IGF)-1, and hepatocyte growth factor (HGF) differed between hypoxic PM and normoxic PM. We found that bFGF, which stimulates Cx43 (19), regulates intracellular Ca2+ and is related to antiapoptosis (20), which was 8-fold higher in hypoxic PM than in normoxic PM (Online Figs. 7 and 8). The level of IGF-1, which has been shown to be related to Cx43 expression (21), and HGF, antifibrotic and antiapoptotic factor (22,23) were elevated in hypoxic PM versus normoxic PM (Online Figs. 7 and 8). Moreover, EF-hand domain-containing 2, reticulocalbin 2, secreted modular calcium-binding protein 1, and secretogranin II, which are known to regulate calcium homeostasis (24–27), were detected only in hypoxic PM (Online Table 3).
The current study demonstrated that: 1) paracrine factors from MSC depend on the surrounding conditions, specifically whether it is under hypoxic or normoxic condition; 2) paracrine factors released under hypoxic condition have a strong antiarrhythmic effect and prevent sudden arrhythmic death; and 3) their antiarrhythmic effect is attributable to restoration of conduction and decreased ectopic focal activity, which are mediated by suppression of fibrotic response, recovery of gap junctions and calcium-regulatory proteins, and alleviation of calcium overload. To the best of our knowledge, this is the first experimental study to report on the effect of paracrine factors on the antiarrhythmic potential of MSC and their potential impact on preventing sudden deaths.
Stem cell therapies to cure MI have focused on differentiating stem cells or engineering stem cells to behave like cardiomyocytes. One of the major challenges facing stem cell therapy is to create tissue heterogeneity, which may potentially increase vulnerability to cardiac arrhythmias. This study provides an alternative therapeutic approach using paracrine factors, only released from the activated MSC.
Fibrotic tissue surrounding surviving cells in infarcted myocardium reduces the speed of impulse propagation, blocks conduction, and facilitates reentry (9). Moreover, fibrosis facilitates the ability to generate triggered activity by a mechanism of sink-source mismatches through coupling between cardiac myocytes and fibroblasts (28). This sequential process was weakened by paracrine growth factors or cytokines from hypoxic MSC, such as bFGF and HGF, thereby allowing a more synchronized electrical propagation and contraction in hypoxic PM–treated hearts. Even though reduced fibrosis and cell death may lead to improvement of contractile function and antiarrhythmic effect, the potency of antiarrhythmic effect of paracrine molecules secreted from MSC is likely to be more predominant (Online Fig. 3). Disruption of Cx43 coupling is another main cause of decreased CV in the border zone (15). We determined that the levels of IGF and bFGF secreted from MSC were significantly increased under hypoxia, and this was associated with increased Cx43 expression (19,21,29), leading to enhanced conduction. Although the current study did not demonstrate the direct association between IGF/bFGF and Cx43 expression, bFGF and IGF were reported to increase expression of Cx43 and suppress the ventricular arrhythmia in another study (18). We also found that increased intracellular Ca2+ concentration was rescued with hypoxic PM. An increase in Ca2+ ions in the cell leads to an overload of the sarcoplasmic reticulum, resulting in spontaneous Ca2+ ions leakage from the sarcoplasmic reticulum. Reduction in intracellular Ca2+ by hypoxic PM (Fig. 4B) is in line with reduction of PVC observed in the in vivo studies (Fig. 1C). Moreover, LTCC during decreased gap junction coupling plays an increasing role in propagation (16,17). In our study, a recovery of LTCC by hypoxic PM may contribute to the increased conduction at the border zone.
Despite the antiarrhythmic effects of paracrine factors from MSC, the inexcitability of MSC could be regarded as proarrhythmogenic. However, it is difficult to evaluate the direct proarrhythmic effect of cells transplanted into the myocardium, independent of the paracrine effect released from injected MSC. It is assumed that MSC implanted after 1-h ligation of the left anterior descending coronary artery would be more likely under normoxic condition over time. It may be possible to deduce the cellular effect of MSC in terms of arrhythmogenicity by comparing normoxic PM–treated rats and MSC-implanted rats. Our study demonstrated that the degree of fibrosis, expression level of Cx43 or Ca2+-related proteins was not different in MSC- and normoxic PM–treated hearts, suggesting that MSC implanted at the border zone after ligation and release might release paracrine molecules similar to components of normoxic PM. The current study suggests that the cellular properties of MSC appear to be neither as beneficial, nor as influential, for arrhythmogenicity as paracrine factors are. The action potential of the MSC-injected region was longer than that of the normoxic PM–injected site and similar to the normal zone of the MSC-engrafted heart. Recently, computer simulation studies showed that myocytes can compensate for the additional electrical load of coupled fibroblasts by increasing the sodium channel current, and a reflow of this additional charge from fibroblasts to the myocytes occurs during repolarization, consequently increasing APDs (30,31). The MSC-myocyte coupling, which prolongs the myocyte refractory period, may facilitate reentry by creating greater dispersion of repolarization. Nevertheless, this study showed that there was no significant difference in survival rate in MSC-engrafted rats versus normoxic PM–injected rats.
Several study limitations are apparent. First, although we identified several growth factors (IGF, bFGF, and HGF) and Ca2+ homeostasis-related proteins, it is possible that other factors are involved. The focus of the present study was the role of environment surrounding stem cells, rather than on the precise mechanism of paracrine factors on arrhythmogenicity. Further studies are needed to identify major factors linking molecular mechanisms of antiarrhythmic effect from MSC. Second, the monitoring of survival was limited to a short period (11 days), not qualifying as a long-term outcome of paracrine factors. This study described ventricular arrhythmia occurring in the early phase of MI, because a therapeutic strategy to target this period is lacking (32), whereas the risk of sudden arrest or death is highest in the early phase of MI (12).
Our study showed that in contrast to the negligible proarrhythmic potential of MSC, paracrine molecules secreted from MSC had strong antiarrhythmic potential, depending on their surrounding environment. It is likely that a hypoxic or normoxic environment surrounding MSC affects the type and properties of the growth factors or cytokines, and these secreted molecules determine the characteristics of the electro-anatomical substrate of the surrounding myocardium. Until now, previous studies have focused mainly on cell types and delivery routes. However, our results demonstrate that the environment of injected cells and their dynamic interaction should also be considered and further investigated to identify therapeutic potency and potential.
For expanded methods and supplemental figures and tables, please see the online version of this paper.
This research was supported by a Korea Science and Engineering Foundation (KOSEF) Grant funded by MOST (M1064102000106N410200110), grants from the Stem Cell Research Center of the 21st Century Frontier Research Program (SC-2150) and Basic Science Research Program through the National Research Foundation of Korea (7-2011-0267) funded by the Ministry of Education, Science, and Technology, and the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (A085136). There was no industry involvement in the study. The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Hwang, Chang, and Song contributed equally to this work.
- Abbreviations and Acronyms
- duration of the action potential at 80% repolarization
- basic fibroblast growth factor
- conduction velocity
- connexin 43
- hepatocyte growth factor
- insulinlike growth factor
- L-type calcium channel
- myocardial infarction
- messenger ribonucleic acid
- mesenchymal stem cell(s)
- paracrine media
- premature ventricular contraction(s)
- ventricular tachyarrhythmia
- Received February 24, 2012.
- Revision received April 11, 2012.
- Accepted April 24, 2012.
- American College of Cardiology Foundation
- Chang M.G.,
- Tung L.,
- Sekar R.B.,
- et al.
- Gnecchi M.,
- He H.,
- Noiseux N.,
- et al.
- Dangman K.H.,
- Dresdner K.P. Jr..,
- Zaim S.
- Xing D.,
- Martins J.B.
- de Bakker J.M.,
- van Capelle F.J.,
- Janse M.J.,
- et al.
- Efimov I.R.,
- Huang D.T.,
- Rendt J.M.,
- Salama G.
- Kanai A.,
- Salama G.
- De Ferrari G.M.,
- Viola M.C.,
- D'Amato E.,
- Antolini R.,
- Forti S.
- Fadok V.A.,
- Bratton D.L.,
- Konowal A.,
- Freed P.W.,
- Westcott J.Y.,
- Henson P.M.
- Beardslee M.A.,
- Lerner D.L.,
- Tadros P.N.,
- et al.
- Shaw R.M.,
- Rudy Y.
- Chang M.G.,
- Zhang Y.,
- Chang C.Y.,
- et al.
- Hahn J.Y.,
- Cho H.J.,
- Kang H.J.,
- et al.
- Merriam L.A.,
- Scornik F.S.,
- Parsons R.L.
- Vannahme C.,
- Smyth N.,
- Miosge N.,
- et al.
- Sato D.,
- Xie L.H.,
- Sovari A.A.,
- et al.
- Kardami E.,
- Stoski R.M.,
- Doble B.W.,
- Yamamoto T.,
- Hertzberg E.L.,
- Nagy J.I.