Author + information
- Received October 24, 2011
- Revision received January 24, 2012
- Accepted February 18, 2012
- Published online June 5, 2012.
- Quanfu Xu, MD⁎,†,
- Florian H. Seeger, MD‡,
- Jessica Castillo, MD⁎,‡,
- Kazuma Iekushi, MD, PhD⁎,
- Reinier A. Boon, PhD⁎,
- Ruxandra Farcas, PhD⁎,
- Yosif Manavski, MSc⁎,
- Yi-Gang Li, MD†,
- Birgit Assmus, MD‡,
- Andreas M. Zeiher, MD‡ and
- Stefanie Dimmeler, PhD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Stefanie Dimmeler, Institute for Cardiovascular Regeneration, Centre of Molecular Medicine, Goethe University, Theodor Stern-Kai 7, 60590 Frankfurt, Germany
Objectives This study evaluated the regulation and function of micro-RNAs (miRs) in bone marrow–mononuclear cells (BMCs).
Background Although cell therapy with BMCs may represent a therapeutic option to treat patients with heart disease, the impaired functionality of patient-derived cells remains a major challenge. Small noncoding miRs post-transcriptionally control gene expression patterns and play crucial roles in modulating cell survival and function.
Methods Micro-RNAs were detected by miR profiling in BMCs isolated from healthy volunteers (n = 6) or from patients with myocardial infarction (n = 6), and the results were confirmed by polymerase chain reaction (PCR) in a larger cohort (n = 37). The function of selected miRs was determined by gain-of-function studies in vitro and by locked nuclear acid (LNA) modified inhibitors in vitro and in vivo.
Results We identified several miRs that are up-regulated in BMCs from patients with myocardial infarction compared with BMCs from healthy controls, including the pro-apoptotic and antiproliferative miR-34a and the hypoxia-controlled miR-210. Inhibition of miR-34 by LNA-34a significantly reduced miR-34a expression and blocked hydrogen peroxide–induced cell death of BMC in vitro, whereas overexpression of miR-34a reduced the survival of BMCs in vitro. Pre-treatment of BMCs with LNA-34a ex vivo significantly increased the therapeutic benefit of transplanted BMCs in mice after acute myocardial infarction (AMI).
Conclusions These results demonstrate that cardiovascular disease modulates the miR expression of BMCs in humans. Reducing the expression of the pro-apoptotic miR-34a improves the survival of BMCs in vitro and enhances the therapeutic benefit of cell therapy in mice after AMI. (BMC Registry, NCT00962364; Progenitor Cell Therapy in Dilative Cardiomyopathy, NCT00284713)
- acute myocardial infarction
- bone marrow mononuclear cells
- cell therapy
- locked nucleic acid
Cell therapy is a promising option to improve neovascularization and repair after ischemia. Transplantation of various adult and embryonic stem cells was shown to augment contractile function after acute myocardial infarction (AMI). Particularly, several subsets of bone marrow–derived cells (BMCs), including hematopoietic progenitor cells, endothelial progenitor cells, and mesenchymal stem cells, improved recovery after ischemia (1–4) most likely due to an effect on vascularization of the ischemic tissue (5). Based on these studies, clinical trials tested the effects of BMCs as well as selected subpopulations to treat patients with ischemic disease. Intracoronary infusion of BMCs in patients with AMI improved ejection fraction and prevented left ventricular remodeling in most, but not all, studies (6). However, the extent of improvement was variable between trials, and the benefit of cell therapy with BMCs particularly in chronically ill patients was modest (7). One reason underlying the modest improvement of cardiac function by cell therapy may relate to the use of the patients' own cells, which are partially compromised in function by the exposure to risk factors. Age, concomitant disease (e.g., diabetes), and heart failure were shown to impair the functionality of the patients' own cells (8–10). The mechanisms underlying the dysfunction are multiple and, for example, include the inhibition of endothelial nitric oxide synthase expression and dysregulation of the balance between nitric oxide and reactive oxygen species (11,12) and the activation of protein kinases (such as p38 mitogen-activated protein kinase) in diabetic patient-derived cells (13).
Micro-RNAs (miRs) are small noncoding RNAs that recently emerged as key epigenetic regulators that control self-renewal and differentiation of stem cells (14,15) and modulate the function of pro-angiogenic BMCs (16). Moreover, they play a crucial role particularly during pathophysiological stress conditions in the heart as well as in the vasculature (for review, see ). Micro-RNAs bind to up to hundreds of target genes and induce mRNA degradation or block translation of the targeted mRNA, thereby modulating the gene expression patterns.
We determined the regulation of the miR expression profile in BMCs isolated from patients with heart disease compared with healthy volunteers and elucidated the function of the dysregulated miRs. We demonstrated that several pro-apoptotic and senescence associated miRs, such as miR-34a (18,19) and let-7 family members (20), as well as the hypoxia-controlled miR-210 (21), are increased in BMCs isolated from patients with acute myocardial infarction or chronic heart failure. Inhibition of 1 of the dysregulated miRs, namely miR-34a, reduced cell death of BMCs in vitro and improved the functional capacity of BMCs to restore cardiac function in a mouse model of AMI in vivo.
Bone marrow aspirates were obtained from healthy volunteers without any evidence of coronary artery disease by history and physical examination, or from patients undergoing intracoronary infusion of BMCs for the treatment of AMI or ischemic cardiomyopathy (ICM) (time from last myocardial infarction >3 months), or heart failure without coronary artery disease (dilated cardiomyopathy [DCM]) within our clinical trials or an ongoing registry. The ethics review board of Goethe University (Frankfurt, Germany), approved the protocols, and the studies are registered with www.clinicaltrials.gov (NCT00962364 and NCT00284713). Written informed consent was obtained from each donor and patient.
BMCs were isolated by Ficoll density gradient centrifugation as previously described (7,22).
Total RNA was isolated with miRNeasy kits from Qiagen (Hilden, Germany). The expression of mature human miRs was determined by human miR microarrays (DNAvision, Charleroi, Belgium). Murine miRs isolated from BMCs of old (18 month) or young (10 weeks) mice were detected by the Agilent microarray (DNAvision). For confirmation of the results, real-time polymerase chain reaction (PCR) was performed using commercially available Taq man PCR primers on an Applied Biosystems StepOnePlus (Carlsbad, California). Expression was normalized to small nucleolar RNA U48. The relative expression is calculated using the formula 2–ΔCt.
Micro-RNA inhibition and overexpression
Micro-RNAs were inhibited by locked nucleic acid (LNA) modified antisense miRs (“in vivo LNA microRNA Inhibitors”; Exiqon, Vedbæk, Denmark) with the following sequences: LNA-Co: ACGTCTATACGCCCA; and LNA-34a: AGCTAAGACACTGCC. LNA inhibitors were incubated with BMCs without any transfection reagent. For overexpression, BMCs were transfected with pre-miR-34a (5 nM) or pre-miR-negative control (5 nM) (Ambion, Carlsbad, California) using the Nucleofector Device (Amaxa, Gaithersburg, Maryland).
After incubation of BMCs with LNA, a total of 1 × 106 BMCs were resuspended in 250 μl medium and placed in the upper chamber of a modified Boyden chamber filled with Matrigel (BioCoat invasion assay, 8-μm pore size; Becton Dickinson, Franklin Lakes, New Jersey). Then, the chamber was placed in a 24-well culture dish containing 500 μl x-vivo 10 medium. For some experiments, 100 ng/ml stromal cell–derived factor (SDF)-1 was added to the lower chamber. After 24 h of incubation at 37°C, transmigrated cells were counted. Invasion assays were run in duplicates.
BMCs (2 × 106) were incubated with the miR inhibitors or phosphate-buffered saline, and cell death was induced with 200 μM hydrogen peroxide for 12 hours. In the starvation-induced cell death study, 2 × 106 BMCs were incubated with the miR inhibitors or phosphate-buffered saline with or without 2% fetal calf serum (FCS) for 48 h. Dead cells were counted after incubation with Trypan blue. The percentage of dead cells was calculated by dividing the number of Trypan blue-positive cells by the number of total cells. Alternatively, pre-miR-34a was overexpressed, and cell death was quantified using the same method. Representative pictures of Trypan blue-positive cells were taken with a light microscope at 20× magnification.
Micro-RNA-34a promoter DNA methylation was determined using methylation specific primers after bisulfite conversion of DNA (23). In brief, bisulfite treatment of 500 ng gDNA was performed using the EpiTect Bisulfite Kit (Qiagen, Hilden, Germany), followed by the methylation-specific PCR (MSP) with the following sequences (MSP-34a-m: Forward primer, GGT TTT GGG TAG GCG CGT TTC; Reverse primer, TCC TCA TCC CCT TCA CCG CCG; MSP-34a-u: Forward primer, TTG GTT TTG GGT AGG TGT GTT TT; Reverse primer, AAT CCT CAT CCC CTT CAC CAC CA). Then, 10 and 20 μl of MSP products were separated by electrophoresis on 1.5% agarose gels, and results were visualized under ultraviolet light after staining with Medori Green.
For Western blot analysis, BMCs were lysed with radioimmunoprecipitation assay (RIPA) buffer (Sigma Munich, Germany) supplemented with protease inhibitor (Roche, Penzberg, Bavaria, Germany). After incubation for 15 min on ice, cell lysates were sonicated for 5 s on ice. Then, the cell lysates were digested with DNase (Qiagen) for 30 min on ice and then centrifuged for 15 min at 14,000g and 4°C. Protein content of the samples was determined according to the Bradford method. Proteins were loaded onto sodium dodecyl sulfate polyacrylamide gels and blotted onto polyvinylidene fluoride membranes (Millipore, Schwalbach, Germany).
Membranes were blocked in Tris-buffered saline-T containing 5% nonfat dry milk and incubated with specified primary antibodies against CDK4 (1/500; Santa Cruz, Heidelberg, Germany) and CCND1 (1/500; Santa Cruz) overnight at 4°C. Horseradish peroxidase (HRP)-conjugated secondary antibody was purchased from GE Healthcare (Munich Germany). Protein bands were visualized with electrochemiluminescent substrate reagent according to the manufacturer's protocol (GE Healthcare).
Acute myocardial infarction model
AMI was induced in male nu/nu mice (8 to 10 weeks old; Charles River, Sulzfeld, Germany) by permanent ligation of the left anterior coronary artery under mechanical ventilation and anesthesia with isoflurane and analgesia with bupivacaine (1 mg/kg, 0.25% bupivacaine 1 to 3 drops on the incision area) and carprofen (5 mg/kg subcutaneously) perioperatively and every 24 h for 3 days post-operatively. BMCs (2 × 106) were incubated with 500 nM LNA-34a or PBS for 24 hours. After washing, cells were re-suspended and 1 × 106 BMCs were injected in a total volume of 50 μl PBS intramuscularly at 4 points in the border zone distal to the ligation of the coronary artery immediately after induction of the myocardial infarction. Cardiac function was determined by high resolution echocardiography (Vevo770 small animal micro-ultrasound system, VisualSonics, Toronto, Ontario, Canada) at day 0 and day 14. Wall motion score index (WMSI) was analyzed using the 16-segment model based on 3-short axis views (4 apical, 6 middle, and 6 basal), with wall motion scored 1 for normal, 2 for hypokinetic, 3 for akinetic, 4 for dyskinetic, and 5 for aneurismal, according to the guidelines of the American Society of Echocardiography. WMSI was calculated as the ratio of the sum of wall motion scores over the total segments scored (24). Fractional shortening and ejection fraction were calculated based on the left ventricular end-systolic dimension and end-diastolic dimension obtained from the middle short-axis view in M-mode.
Distributions of categorical variables were tested by chi-square test or Fisher's exact test. Continuous variables are reported as mean ± SE, if not stated otherwise. All variables were tested for homogeneity of variance by Levènes test. If heterogeneity of variance was present, analysis of variance (ANOVA) was followed by Dunnett-T3 post hoc analysis. Otherwise, groups were compared using ANOVA followed by Bonferroni correction. Correlations were analyzed by Spearman test. The specific test used to analyze the data is indicated in the figure legends.
Experimental studies were analyzed using 1-way ANOVA (least significant difference test) for the comparison among 3 groups, and independent sample t-test for comparison between 2 groups. Probability values of <0.05 were considered significant and 2-sided tests were performed. Data are presented as means, and error bars depict the SEM.
Micro-RNA expression in patient-derived BMC
For analysis of the miR expression patterns, BMCs were isolated from healthy volunteers (n = 6) or patients with AMI or ischemic cardiomyopathy (ICM) (n = 6) (Online Table 1). The group of healthy volunteers was significantly younger, and did not take any medication (Online Table 1). Analysis of the miR microarray revealed that several miRs are regulated in BMCs isolated in all patients (n = 6) compared to those isolated from healthy volunteers (Online Tables 2 and 3, Fig. 1). Among others, the expression of the pro-apoptotic miR-34a and the hypoxia-regulated miR-210 were increased (Fig. 1B). Moreover, the expression of miR-1274a and miR-1274b, whose functions are unknown, were up-regulated (Fig. 1B).
To confirm the regulation of miR expression in patient-derived BMCs compared with healthy controls and to obtain further insights into the specific influence of the underlying disease on BMC miR expression, we determined the expression of selected miRs in a confirmation cohort of 11 healthy volunteers, 11 patients with ICM, 8 patients with AMI, and 7 patients with nonischemic heart failure (DCM) (Table 1). Consistent with the increased expression of miR-210 shown in the profiling, miR-210 was significantly increased in all patients (n = 26) compared with healthy controls (139 ± 8% increase; p = 0.012) (Online Fig. 1). The expression of miR-210 was also increased in BMCs isolated from patients with ICM, AMI, and DCM; however, the differences failed to achieve statistical significance (Fig. 2A). Micro-RNA-34a levels, which were among the most profoundly up-regulated in the profile, were significantly increased in all patients (238 ± 18%; p <0.001) and in patients with AMI, ICM, and DCM compared with healthy controls (Fig. 2B). Micro-RNA-1274a and miR-1274b were significantly increased in BMCs from patients compared with the expression in BMCs isolated from healthy volunteers (141 ± 12%; p = 0.043 and 149 ± 11%; p = 0.008, respectively). Although miR-1274a only showed a trend toward higher expression in patients with AMI and ICM (Fig. 2C), miR-1274b was significantly increased in AMI and ICM patient-derived BMCs compared with healthy controls (Fig. 2D).
Moreover, we determined the expression of additional miRs, such as members of the let-7 family, miR-1260, and miR-30b. Let-7b and let-7c were significantly increased in patient-derived BMCs compared with healthy controls (Online Fig. 1). Let-7b as well as let-7c were significantly increased in BMCs from patients with AMI compared with healthy controls (Figs. 2E and 2F), whereas only a trend towards increased expression was detected in ICM and DCM patient-derived BMCs. Micro-RNA-1260 was also increased comparing BMCs isolated from all patients with healthy controls (148 ± 12%; p = 0.019); however, no significant differences were detected between groups (Fig. 2G). Micro-RNA-30b, which regulates invasion (25), and miR-92a, which was previously shown to inhibit neovascularization after ischemia (26), were not regulated in BMCs isolated from patients with either AMI, ICM, or DCM compared with healthy controls (Fig. 2H) (data not shown).
Influence of age on miR levels in BMC
Because some of the up-regulated miRs were associated with aging (20,27), we determined the influence of age on the expression of miRs in BMCs. Micro-RNA-1274b, miR-34, miR-1260, and the members of the let-7 family showed a highly significant association with age (Figs. 3A to 3E). This association was also observed when only patients with cardiac disease were included in the analysis (data not shown). The age dependency of let-7b, let-7c, and miR-34a was confirmed in mice BMCs (Figs. 3G to 3I), suggesting that these miRs are influenced by age.
Regulation of miR-34a expression in BMC is independent on promoter DNA methylation
Because recent studies suggested that miR-34 expression is regulated by silencing DNA methylation of CpG islands (23,28), we evaluated the DNA methylation patterns of the miR-34a promoter in BMCs isolated from healthy controls and patients with congestive heart failure. Therefore, we used previously described MSPs (23), which were validated by using transformed cells. Consistent with previous studies, transformed cells showed high DNA methylation of the miR-34a promoter, which was reduced by treatment of the cells with the DNA methylation inhibitor azaC (Fig. 4A). However, no DNA methylation was detected in any of the BMC samples (Fig. 4B), suggesting that miR-34a expression is not regulated at the level of DNA methylation in BMCs.
Inhibition of miR-34a expression prevents cell death of BMCs
Because miR-34a is known to induce apoptosis in several cell types and was highly expressed in BMCs (Online Fig. 2A), we determined the influence of miR-34a inhibition on BMC survival. Because cholesterol-modified antagomirs directed against miR-34a only insufficiently repressed the miR (data not shown), we used LNA-modified anti-miRs (29), which dose-dependently inhibited miR-34a expression (Fig. 5A). LNA-34 most efficiently inhibited miR-34a (>80%) and miR-34b (∼50%), whereas the other family members, miR-34c and miR-449a, were only slightly reduced, and miR-449b was unchanged (Online Figs. 2A and 2B). LNA-34a significantly inhibited hydrogen peroxide and serum depletion-induced cell death (Figs. 5B and 5C, Online Fig. 3). In addition, a significant increase of SDF-1–induced migration was observed in LNA-34a treated BMCs (Fig. 5D). Consistent with a pro-apoptotic function of miR-34a, we demonstrated that overexpression of miR-34a in healthy volunteer-derived BMCs increased cell death (Figs. 6A and B).
Micro-RNA-34a is known to regulate various target genes that control apoptosis and cell cycle progression, such as Sirt1 (30), Bcl2, cyclins (CCN), and cyclin-dependent kinases (CDKs) (19). Therefore, we determined the regulation of miR-34a targets genes in BMCs. Overexpression of miR-34a in BMCs reduced the mRNA expression of Sirt1 by around 10% and repressed CDK4, CDK6, and the cyclins, CCN1 and CCN3, by 10% to 20%, whereas the expression of the cell cycle inhibitor p21 was not changed (Fig. 6C). The down-regulation of CDK4 and CCN1 by miR-34 was confirmed by Western blot (Fig. 6D).
Inhibition of miR-34a expression ex vivo improves the therapeutic benefit of cell therapy
Inhibition of BMC apoptosis by anti-miR-34a may provide a strategy to enhance the therapeutic benefit of BMC therapy. Therefore, we pre-treated human BMCs with LNA-34a, followed by injection in infarcted nude mice. Although the extent of functional impairment after infarction was similar in all 3 groups at baseline (Online Fig. 4), BMC injection increased cardiac function after 2 weeks of follow-up (Fig. 7). In the group of mice that had been injected with LNA-34a pre-treated BMCs, a significant further increase in functional recovery was observed (Fig. 7), suggesting that ex vivo inhibition of miR-34a improves the therapeutic effect of BMC cell therapy.
The data of the present study demonstrate that BMCs isolated from patients with AMI or heart failure show a distinct expression profile of miRs compared with healthy donor derived BMCs. Particularly, the pro-apoptotic miR-34a was significantly increased in patient-derived BMCs. Inhibition of miR-34a improved cell survival, whereas overexpression induced cell death in BMCs in vitro. Ex vivo pre-incubation of BMCs with a miR-34a inhibitor increased the capacity of the injected BMCs to augment cardiac function in mice after AMI. The miR-34 family is known for its function as a tumor suppressor (19,31). Ectopic overexpression of miR-34a induced apoptosis and cell cycle arrest in primary- and tumor-derived cell lines (31–33). Additionally, miR-34a is involved in oncogene-induced senescence (18) and is increased in senescent endothelial cells in vitro (27). These data are in accordance with our study showing that miR-34a regulates BMC survival in vitro and that miR-34a expression is associated with the age of the donor of the BMCs in humans and in mice. The mechanism underlying the increased miR-34a expression in BMCs of patients with heart disease or in aged mice is unclear. It is known that p53 induces the expression of the transcripts encoding miR-34 family members (31,34,35). Moreover, the expression of miR-34 can be increased independently of p53 (18) and is also regulated by DNA methylation of CpG islands of the miR-34 promoters (23,28,36). However, when we analyzed the DNA methylation of the promoter of miR-34a, no methylation was detected in any of the BMC samples, suggesting that age and disease either transcriptionally activate the miR-34a promoter or enhance the processing of miR-34a.
Consistent with previous studies (19,30), miR-34a reduced the expression of the anti-apoptotic proteins Sirt1 and Bcl2 and the cell cycle regulatory cyclins and CDKs in human BMCs. Of note, although the global gene expression pattern shown in the present study was consistent with the previous findings, the extent of target gene regulation was modest. This may in part be explained by the preferential inhibition of protein translation compared with direct RNA degradation by miRs as well as the inefficient overexpression of the precursor by electroporation in BMCs.
In addition to miR-34, several other miRs were dysregulated in patient-derived BMCs. Micro-RNA-210 was increased in the profile and in the confirmation study. Micro-RNA-210 is up-regulated by hypoxia in vitro and in plasma of critically ill patients with acute kidney injury (37). Because miR-210 suppresses mitochondrial functions (21) and mitochondrial function was shown to be essential for pro-angiogenic cell migration (38), we hypothesized that the increase in miR-210 levels might control BMC functions. However, blocking miR-210 neither affected mitochondrial membrane potential nor did it influence BMC survival and functions in vitro (Online Fig. 5).
The let-7 family members, let-7b and let-7c, were also significantly increased, particularly in BMCs from patients with AMI. Let-7b is known to reduce neural stem cell numbers and to suppress self-renewal by targeting the transcriptional regulator Hmga2 (20), and let-7c represses self-renewal genes in embryonic stem cells (39). Because let-7 expression was highly associated with age in the patient population and was increased in the BMCs isolated from old versus young mice, one may speculate that this miR may contribute to age-induced progenitor cell senescence. Interestingly, let-7 family members (let-7a/b/c/d/f) reduced the expression and release of vascular endothelial growth factor and interleukin-6 (40). Because the release of pro-angiogenic cytokines is considered a major mechanism underlying the BMC-mediated improvement of neovascularization and repair processes (41), the repression of these cytokines by augmented levels of let-7 family members may contribute to the impaired functionality of patient-derived BMCs.
Of note, the regulation of miRs in BMCs isolated from patients with AMI did not entirely reflect the miR expression patterns reported in a recent study demonstrating that miR-150 was significantly reduced in BMCs isolated from mice after AMI (42). Although miR-150 was slightly reduced in BMCs from patients with AMI (Online Fig. 6), the differences were not statistically significant. Other miRs, such as the let-7 family, which were also down-regulated in mice after AMI (42), were significantly increased in our patient cohort. One may speculate that the differences between our study and the miR expression patterns after AMI in mice, at least in part, might be related to the fact that patients (but not young healthy mice) are additionally exposed to risk factors, which might influence miR expression patterns in the bone marrow. However, our patient cohorts were rather small, and further confirmation in larger patient cohorts is warranted.
Together, our study provides novel insights into the dysregulated miR pattern of patient-derived BMCs. Inhibition of the elevated levels of the pro-apoptotic miR-34a ex vivo improved the functional benefit of transplanted BMCs in vivo and might comprise a strategy to enhance the effect of autologous cell therapy. In addition, the de-regulation of other miRs, particularly the senescence associated let-7 family members, may additionally impair the function of the cells and modulate gene expression patterns in BMCs from old and diseased patients.
The authors thank Tina Rasper for excellent technical assistance.
For supplemental tables and figures, please see the online version of this article.
The study was supported by the European framework program 7 (Endostem, #241440), the LOEWE Centre for Cell and Gene Therapy, the European Research Council (ERC grant Angiomir to Dr. Dimmeler) and the Deutsche Forschungsgemeinschaft (SFB834/B6 to Dr. Zeiher and Dr. Assmus). Dr. Zeiher has relationships with t2cure and Sanofi-Aventis. Dr. Dimmeler has relationships with Miragen and t2cure GmbH. All other authors have reported that they have no relationships relevant to the contents of this paper. Drs. Xu, Seeger, and Castillo contributed equally to this work.
- Abbreviations and Acronyms
- acute myocardial infarction
- analysis of variance
- bone marrow–derived mononuclear cell
- cyclin-dependent kinases
- dilated cardiomyopathy
- ischemic cardiomyopathy
- locked nuclear acid
- methylation-specific PCR
- polymerase chain reaction
- stromal cell–derived factor
- wall motion score index
- Received October 24, 2011.
- Revision received January 24, 2012.
- Accepted February 18, 2012.
- American College of Cardiology Foundation
- Dimmeler S.,
- Burchfield J.,
- Zeiher A.M.
- Losordo D.W.,
- Schatz R.A.,
- White C.J.,
- et al.
- Hare J.M.,
- Traverse J.H.,
- Henry T.D.,
- et al.
- Yoon C.H.,
- Koyanagi M.,
- Iekushi K.,
- et al.
- Lipinski M.J.,
- Biondi-Zoccai G.G.,
- Abbate A.,
- et al.
- Dimmeler S.,
- Leri A.
- Fadini G.P.,
- Boscaro E.,
- de Kreutzenberg S.,
- et al.
- Kissel C.K.,
- Lehmann R.,
- Assmus B.,
- et al.
- Thum T.,
- Fraccarollo D.,
- Schultheiss M.,
- et al.
- Boon R.A.,
- Urbich C.,
- Fischer A.,
- et al.
- Seeger F.H.,
- Haendeler J.,
- Walter D.H.,
- et al.
- Fleissner F.,
- Jazbutyte V.,
- Fiedler J.,
- et al.
- Assmus B.,
- Schachinger V.,
- Teupe C.,
- et al.
- Zhang Y.,
- Takagawa J.,
- Sievers R.E.,
- et al.
- Bonauer A.,
- Carmona G.,
- Iwasaki M.,
- et al.
- Yamakuchi M.,
- Ferlito M.,
- Lowenstein C.J.
- Lorenzen J.M.,
- Kielstein J.T.,
- Hafer C.,
- et al.
- Spyridopoulos I.,
- Fichtlscherer S.,
- Popp R.,
- et al.