Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2012 Jun 19;20(9):1750–1766. doi: 10.1038/mt.2012.109

Secondary Sphere Formation Enhances the Functionality of Cardiac Progenitor Cells

Hyun-Jai Cho 1,2, Ho-Jae Lee 2,3, Seock-Won Youn 2, Seok-Jin Koh 2, Joo-Yun Won 2, Yeon-Ju Chung 3, Hyun-Ju Cho 2, Chang-Hwan Yoon 1, Sae-Won Lee 2, Eun Ju Lee 2, Yoo-Wook Kwon 2, Hae-Young Lee 1, Sang Hun Lee 4, Won-Kyung Ho 4, Young-Bae Park 1,2, Hyo-Soo Kim 1,2,3,5,*
PMCID: PMC3437578  PMID: 22713697

Abstract

Loss of cardiomyocytes impairs cardiac function after myocardial infarction (MI). Recent studies suggest that cardiac stem/progenitor cells could repair the damaged heart. However, cardiac progenitor cells are difficult to maintain in terms of purity and multipotency when propagated in two-dimensional culture systems. Here, we investigated a new strategy that enhances potency and enriches progenitor cells. We applied the repeated sphere formation strategy (cardiac explant → primary cardiosphere (CS) formation → sphere-derived cells (SDCs) in adherent culture condition → secondary CS formation by three-dimensional culture). Cells in secondary CS showed higher differentiation potentials than SDCs. When transplanted into the infarcted myocardium, secondary CSs engrafted robustly, improved left ventricular (LV) dysfunction, and reduced infarct sizes more than SDCs did. In addition to the cardiovascular differentiation of transplanted secondary CSs, robust vascular endothelial growth factor (VEGF) synthesis and secretion enhanced neovascularization in the infarcted myocardium. Microarray pathway analysis and blocking experiments using E-selectin knock-out hearts, specific chemicals, and small interfering RNAs (siRNAs) for each pathway revealed that E-selectin was indispensable to sphere initiation and ERK/Sp1/VEGF autoparacrine loop was responsible for sphere maturation. These results provide a simple strategy for enhancing cellular potency for cardiac repair. Furthermore, this strategy may be implemented to other types of stem/progenitor cell-based therapy.

Introduction

Cell-based therapies have been investigated experimentally and clinically in the contexts of regenerating or repairing damaged hearts.1 Over the past decade, various types of extracardiac cells have been proposed as potential cell sources. However, the cardiovascular differentiation of extracardiac cells is the subject of considerable debate.2,3 Clinical trials, especially on the use of bone marrow-derived cells, have shown modest benefits in acute or chronic myocardial infarction (MI) patients.4,5 Thus, the quest for an optimal cell type continues.

Recent studies have raised the possibility that postnatal hearts possess resident stem/progenitor cells, which presumably are imprinted with cardiovascular fate as compared with extracardiac cells.6,7 Cardiac resident stem/progenitor cells have shown to differentiate into cardiovascular lineages, have regenerative potentials, and improve cardiac function when transplanted into ischemic hearts.8,9,10,11,12,13 However, c-kit (+) or sca-1 (+) cardiac stem/progenitor cells and side population cells are complicated to maintain when propagated in vitro for transplantation purposes. In contrast, the generation of cardiospheres (CSs) from cardiac explants is regarded to be relatively simple.10,14,15 But, due to insufficient cell numbers of direct outgrowing cells from explants and CSs for in vivo transplantation, in vitro expansion protocol, the cardiosphere-derived cell (CDCs) technology, was developed.14 However, CSs and CDCs are heterogeneous and contain stem/progenitor cells and fibroblast-like cells,15,16,17 even though the correlation between the therapeutic efficacy and the heterogeneity or homogeneity of transplanted cells is not clear.

Cell survival and engraftment after transplantation is also a key requirement for cardiac repair.18,19 Several studies have reported that cellular engraftment after transplantation into damaged tissues is limited, and that transplanted cells are susceptible in hostile ischemic environment and tend to disappear within a few days.20,21,22

Accordingly, a stable and reproducible strategy is demanded to acquire optimal cell populations in vitro while maintaining cellular potency to repair infarcted hearts, and to enhance cellular engraftment following transplantation in vivo to facilitate cell therapy. To meet these challenges, we investigated whether repeated sphere formation, that is, primary CS formation → sphere-derived cells (SDCs) → secondary CS formation by three-dimensional culture, could enhance the multipotency of cardiac stem/progenitor cells. And we studied if transplantation of secondary CSs enhances engraftment, it will consequently improve cardiac function after MI. We also investigated the molecular mechanisms responsible for sphere formation.

Results

Generation of primary CSs from cardiac explants

Hearts were harvested from C57BL/6 mice. Minced ventricular tissues were digested, and cultured. Three days after implanting cardiac explants on fibronectin (FN)-coated dishes, phase-bright cells were observed as reported previously.10,15 To generate primary CSs, cells were harvested around day 8 by using trypsin and reseeded on poly-D-lysine (PDL)-coated dishes. Three days later (day 11), these cells formed primary CSs. Floating CSs were reattached on FN-coated dishes (day 16), and adherent cells rapidly expanded from CSs (Figure 1a,b).10,14,15

Figure 1.

Figure 1

Open in a new tab

Generation of primary and secondary cardiospheres and their characteristics. (a) Timeline of primary CS, SDC, and secondary CS generation. Within 48 hours, secondary CSs were generated from SDCs. (b) Phase-contrast bright field images and step-by-step average cell numbers (n = 5). Bar: 500 µm. (c) The gene expressions of Oct4 and c-kit were measured by real-time PCR. (d) The protein expressions of Oct4 and c-kit were measured by western blots and quantified by densitometry. (e) Primary and secondary CSs were positive for alkaline phosphatase (ALP) staining, but SDCs were not. The protein expressions of Oct4 and c-kit were assessed by confocal imaging. Oct4 expression was verified using Oct4 promoter-driven GFP cells. Nuclear colocalization of Oct4 was also confirmed in the single, dissociated secondary CS cells (Supplementary Figure S1). Secondary CSs expressed Oct4 and c-kit more homogeneously and densely than primary CSs, whereas SDCs lacked expressions. TO-PRO-3 (nuclei). Bar: 50 µm. AU, arbitrary unit; CS, cardiosphere; FN, fibronectin; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; PDL, poly-D-lysine; SDC, sphere-derived cell.

These cells showed mesenchymal stem cell-like (or fibroblast-like) morphology. To analyze surface markers' expression, we cultured three different cell lines with the modified media which contained EGF, bFGF (basic fibroblast growth factor), cardiotrophin-1, B27, and 3.5% fetal bovine serum (FBS). As a historical control, we summarized the results from Marban group's various publications16,17 (Supplementary Figure S1). Furthermore, as an internal control of the experiment, we cultured cells with 10% FBS and without cytokine, which is similar media as reported previously.16,17 Majority of cells expressed mesenchymal lineage surface markers, but did not express hematopoietic cell-related markers23,24. Interestingly, there are some differences. The adherent cells cultured with cytokines and low concentration of FBS expressed CD31, CD90, CD106, and Sca-1 more abundantly, compared to a historical control from Marban's previous publications and an internal control of cells with high FBS without cytokine (Supplementary Figure S1). These results suggest that some modifications of culture such as the concentration of serum and cytokines could significantly change cellular characteristics and alter the study conclusion. Therefore, in this study, we refer to this adherent cell from CS as SDC rather than CDC which may indicate a specific cellular product currently under investigation in the clinical trial (CADUCEUS, NCT00893360, http://www.ClinicalTrials.gov). In the series of in vitro and in vivo experiments, we used SDCs cultured with cytokines and low concentration of FBS.

Next, we compared the expression of stemness-related genes of primary CS and SDC. Real-time PCR data showed that CS expressed Oct4 and c-kit, as has been previously reported.10,15 However, surprisingly, SDCs showed reduced Oct4 and c-kit gene expressions (Figure 1c). Alkaline phosphatase and immunofluorescent stainings showed consistent results (Figure 1e). To exclude artifacts of Oct4 staining, we generated CSs from the hearts of Oct4 promoter-driven enhanced green fluorescent protein (eGFP) mice, and confirmed that primary CSs expressed Oct4 but that SDCs did not (Figure 1e). Taken together, these results indicate that primary CS possesses multipotency, but that SDC, expanded by two-dimensional monolayer culture, has limited cellular potency.

Characteristics and differentiation potentials of secondary CSs

Primary CSs, obtained from outgrowing cells from explants, have far insufficient number of cells for in vivo transplantation. When we started with 5–10 pieces of cardiac tissues from a mouse, we obtained 1.43 × 105 ± 4.59 × 104 phase-bright outgrowing cells at day 8, and thereafter 130 ± 22 primary CSs were generated with these cells at day 14 (Figure 1b). Therefore, a relevant expansion protocol is necessary. To overcome the limitations of monolayer culture, we investigated a repeated sphere formation strategy to restore the expression of stemness-related genes. Spheres were generated from SDCs using a PDL-coated dish in the same manner used for primary CS generation. However, we found that the size of sphere by the PDL-based method was variable (50–200 µm).

The size and cell number variation of each sphere significantly affected gene expression profiles and also changed the yield of cellular product, batch-by-batch (H.S. Kim, unpublished data). We concluded that uniformly sized spheres are very important for reproducible in vitro and in vivo experiments. Accordingly, we applied hanging-drop technique to regulate cell numbers and size in each sphere, which is widely used for embryonic stem cell experiments to regulate cell numbers in spheres.25 In this regard, most of in vitro and in vivo secondary CS experiments in this study were performed using hanging-drop. We hung 1,000 cells in a drop, and 48 hours later observed uniformly sized spheres (100–200 µm). We call the subsequently formed CS from SDCs as “secondary CS or 2nd CS”. Using this repeated sphere formation strategy, we generated 5,850 ± 1,150 secondary CSs at day 24 (Figure 1b).

Compared to SDCs, secondary CSs recovered Oct4 and c-kit gene expressions (Figure 1c). Protein expressions were confirmed by western blots and quantified (Figure 1d). In addition, we reconfirmed the protein expression and nuclear colocalization of Oct4 in the single cell, dissociated secondary CS cells by high magnification confocal imaging (Supplementary Figure S2). Intriguingly, as compared with primary CSs, secondary CSs expressed Oct4 and c-kit more homogeneously and densely, suggesting that secondary sphere formation enhances multipotency (Figure 1d).

Next, we evaluated the differentiation potentials of secondary CS. First, we checked endothelial differentiation. Endothelial nitric oxide synthase promoter-driven eGFP reporter26 was transduced into SDC, then secondary CSs were generated and subsequently attached to dishes for differentiation. At the baseline before the induction of differentiation, there was no eGFP expression. After 7 days, a few SDCs were found to express eGFP (2.7 ± 0.7%). In contrast, many cells sprouted from secondary CSs and expressed eGFP (16.4 ± 4.2%, P = 0.02 compared with SDCs, N = 4, each group). Furthermore, these cells were also positive for isolectin B4 (ILB4), a marker of endothelial cells (Figure 2a). To verify endothelial differentiation, we performed the same experiment by vascular endothelial growth factor receptor 2 (VEGFR2) promoter-driven eGFP reporter. These findings indicate that secondary CSs have greater endothelial differentiation potential than SDCs have.

Figure 2.

Figure 2

Open in a new tab

Differentiation potentials of secondary cardiospheres. (a) Cells were transduced with eNOS promoter-driven eGFP or VEGFR2 promoter-driven eGFP using lentivirus. Seven days after differentiation induction, secondary CSs expressed GFP but SDCs scarcely expressed. GFP-expressing cells were confirmed to have differentiated into the endothelial lineage by isolectin B4 staining. (b) Cardiomyogenic differentiation was induced by coculture with rat cardiomyocytes. Cells were from the heart of a β-actin promoter-driven eGFP-expressing mouse to distinguish them from rat cardiomyocytes or transduced with α-MHC promoter-driven eGFP reporter to validate cardiac differentiation. Cx43 was densely expressed between secondary CSs and rat cardiomyocytes. Bar: 50 µm. (c) Calcium transients showed the synchronous contractions between secondary CSs from the eGFP-expressing mouse and wild-type rat cardiomyocytes. The ratio of fluorescence intensity, F340/F380, was indicated as calcium transient of each cell. α-MHC, α-myosin heavy chain; CS, cardiosphere; eGFP, enhanced green fluorescent protein; eNOS, endothelial nitric oxide synthase; SDC, sphere-derived cell; VEGFR2, vascular endothelial growth factor receptor 2.

Second, we induced cardiomyogenic differentiation. We cocultured SDCs and secondary CSs from β-actin promoter-driven eGFP-expressing mice with rat cardiomyocytes. When we induced the differentiation without coculture, both SDCs and secondary CSs did not contract spontaneously. After 5–7 days of coculture, cells sprouted from secondary CSs and showed beating activity (Supplementary Figure S3 and Supplementary Videos S1 and S2). Furthermore, connexin 43 (Cx43) was highly expressed between cells derived from secondary CSs and rat cardiomyocytes, whereas SDCs rarely contracted and few cells expressed Cx43 (Figure 2b, middle panel). Next, we performed the GFP (+) sorting experiment following coculture. After the enzymatic detachment-FACS-reattachment process, GFP (+) secondary CS-derived cells did not contract spontaneously; even rat cardiomyocytes contracted very weakly after sorting. In another set of experiments to quantify the differentiation, we transduced α-myosin heavy chain (α-MHC) promoter-driven eGFP to monitor cardiomyocyte-specific gene expression. We observed α-MHC promoter was significantly activated in the secondary CSs, compared with SDCs (6.3 ± 1.0% versus 2.8 ± 0.4%, P < 0.01, N = 4, each group) (Figure 2b, lower panel). We also checked electrophysiologic properties.27 Cytosolic calcium transients revealed synchronous contraction between secondary CSs and rat cardiomyocytes, indicating that they were functional cardiomyocytes (Figure 2c).

To summarize, these findings suggest that secondary CS possesses greater differentiation potential to endothelial cell and cardiomyocyte than SDC, and that cardiomyocytes from secondary CS make electrical coupling with authentic cardiomyocytes.

Secondary CS transplantation following MI promotes cardiac repair

To investigate the functional benefits of cell transplantation into ischemic myocardium, we injected SDCs (1 × 105), secondary CSs (1 × 105), or phosphate-buffered saline (PBS) intramyocardially, after left anterior descending artery ligation. Regarding secondary CSs, we created two groups: the secondary CS-single cell dissociated group (the 2nd-S group) and the secondary CS-hanging drop group (the 2nd-H group). To produce secondary CSs with a consistent size, we generated secondary CSs containing 1,000 cells using the hanging-drop method before transplantation.28 In the 2nd-H group, 100 secondary CSs were generated from 1 × 105 SDCs by hanging-drop for 48 hours. In the single cell (2nd-S) group, secondary CSs were dissociated into single cells using trypsin. To check the cell number of spherical form of secondary CSs, we dissociated secondary CSs at 6, 24, 48, and 72 hours after hanging-drop. When we dissociated spheres at 6 hours, the cell number was slightly decreased (5 × 104). The cell number was 1 × 105 at 24 hours, 7 × 104 cells at 48 hours, and 1.2 × 105 at 72 hours. We have speculated possibilities why the number did not increase exponentially after hanging-drop despite of the high proliferative potential of SDCs. The one possibility could be cell apoptosis and/or necrosis at the initial phase of sphere formation. Other possibility may be that the proliferative potential of SDCs could decrease transiently due to relative hypoxic environment of sphere structure compared with two-dimensional culture. In this study, 48 hours after hanging-drop of SDCs, we transplanted the spherical form of secondary CSs into the infarcted heart (2nd-H group). Therefore, we believe that the same number of cells was transplanted in each group.

An echocardiographic examination at 14 days after transplantation showed that left ventricular (LV) systolic function indexes, such as, fractional shortening and ejection fraction, were better in groups treated with either one of three types cell transplantation (SDC, 2nd-S, and 2nd-H groups) than with PBS (Figure 3a). In the short-term follow-up, there was no significant difference of global systolic function (fractional shortening and ejection fraction) between cell injection groups. However, importantly, in terms of LV remodeling, sphere transplantation (2nd-H group) was more effective than dissociated-single cell transplantation (2nd-S group) or SDCs in preventing LV dilatation of systolic and diastolic phases (P value <0.05, 2nd-H group versus SDC and 2nd-S groups). These data suggest that the transplantation of secondary CSs in spherical form promotes cardiac repair and ameliorates adverse cardiac remodeling as compared with the transplantation of single-dissociated cells from secondary CSs or SDCs. To examine pathologic changes after MI and cell transplantation, we checked fibrosis lengths and infarct thicknesses (Figure 3b). Masson's trichrome staining revealed that secondary CS transplantation significantly reduced infarct size and increased wall thickness in infarcted areas, indicating the preservation or regeneration of functional myocardium.

Figure 3.

Figure 3

Open in a new tab

Transplantation of secondary cardiospheres into ischemic myocardium. After MI, we transplanted three types of cells, namely, SDCs, singly dissociated secondary CS cells (2nd-S), secondary CS-hanging drop cells as spherical form (2nd-H). PBS was injected as a control. All types of cells were generated from heart explants of eGFP-expressing mice. (a) Echocardiographic parameters at 14 days after cell transplantation. Left ventricular end-systolic dimension (LVESD), LV end-diastolic dimension (LVEDD), LV fractional shortening (LVFS), and LV ejection fraction (LVEF) (n = 6 per group, *P < 0.01, PBS versus the two transplanted groups; **P < 0.05, 2nd-H group versus SDC and 2nd-S groups). (b) Masson's trichrome staining; 2nd-H transplantation markedly reduced infarct-related fibrosis length (*P < 0.01, 2nd-H group versus the other three groups); 2nd-H transplantation increased infarct thickness more than transplantation with SDCs or 2nd-S (*P < 0.05, PBS versus the SDC and 2nd-S groups; **P < 0.01, the 2nd-H group versus the SDC and 2nd-S groups). Bar: 500 µm. (c) Amplification plots of various sets of known number of female and male cells using Y chromosome-specific primers and probe (right panel). Standard curve to calculate unknown male cell numbers (left panel). (d) Transplantation of male cells to female infarcted heart. Real-time PCR for Y chromosome showed that cell transplantation as spherical form was superior cellular engraftment to single-dissociated cells from secondary CSs and SDCs (n = 4 per group, *P < 0.01, 2nd-H group versus the SDC or 2nd-S groups). (e) Transplantation of GFP-expressing cells to wild-type infarcted heart. Secondary CSs as spherical form (2nd-H group) enhanced engraftment at 7 days after transplantation (n = 6 per group, *P < 0.01, 2nd-H group versus the SDC or 2nd-S groups). No significanct difference was observed between the SDC and 2nd-S groups. Bar: 50 µm. (f) In the peri-infarct area at day 7 post-transplantation, engrafted SDCs and 2nd-S cells did not express cardiac transcription factor Nkx2.5. However, 2nd-H cells expressed Nkx2.5, and some Nkx2.5 (+) cells did not express GFP, suggesting that 2nd-H cells activated endogenous cardiac progenitor cells. Bar: 20 µm. (g) At day 14, transplanted GFP (+) secondary CSs differentiated into the cardiomyocyte (yellow arrow) and the endothelial cell (white arrow). Bar: 20 µm. α-SA, α-sarcomeric actinin; CS, cardiosphere; eGFP, enhanced green fluorescent protein; ILB4, isolectin B4; MI, myocardial infarction; PBS, phosphate-buffered saline; SDC, sphere-derived cell.

Collectively, these findings suggest that secondary CS transplantation induces favorable cardiac remodeling after MI.

Engraftment and differentiation of secondary CS in vivo

The physical incorporation or engraftment and cardiovascular differentiation of transplanted cells are major determinants for the effects of cell therapy.29

Physically, three-dimensional sphere formation and subsequent cell transplantation could be a very important variable to determine the degree of cellular engraftment. To address this point, we transplanted the same cells prepared in two different ways; spherical (2nd-H) and single-dissociated forms (2nd-S). To quantify the degree of cellular engraftment, we performed sex-mismatched cell transplantation experiments (male cells to female infarcted heart). We harvested tissue and extracted DNA immediately after cell injection and at day 3, day 7, and day 14. The engrafted male cells in the female infarcted heart were quantified by real-time PCR using Y chromosome-specific primers and probe as we previously described.30 By amplification plots using various sets of known number of female and male cells, we made a standard curve and then calculated unknown male cell numbers (Figure 3c). The number of cells immediately after injection was similar in three groups. At days 3 and 7, the transplantation of secondary CSs in spherical form revealed superior cellular engraftment to single-dissociated cells from secondary CSs and SDCs (Figure 3d, P < 0.01, 2nd-H group versus the 2nd-S and SDC groups). The number of engrafted cells declined rapidly by day 14. However, significantly better engraftment was still observed in the 2nd-H group (7.3 ± 1.8% of initially injected cells), compared with 2nd-S (3.5 ± 1.5%) and SDC (3.0 ± 0.6%) groups (Figure 3d).

To visualize the transplanted cells in ischemic myocardium, we prepared cells from heart explants of mice expressing the actin promoter-driven GFP. We harvested myocardial tissues at 7 days after cell transplantation. Engrafted cells were recognized by immunofluorescent staining against GFP and observed in the peri-infarct border zone. As has been previously reported19,20 and accordant with sex-mismatched cell transplantation data, only a few cells were observed in SDC group. In contrast, secondary CSs prepared by hanging-drop (2nd-H group) showed better engraftment at 7 days after transplantation (Figure 3e, P < 0.01, 2nd-H group versus the 2nd-S and SDC groups). These findings suggest that cell preparation methods used before transplantation significantly affect outcomes.

To evaluate the fate of engrafted cells at 7 and 14 days, we performed immunofluorescent staining with Nkx2.5 and Troponin T. In the group transplanted with spheres of secondary CS (2nd-H), transplanted cells found in the peri-infarct zone expressed Nkx2.5 at 7 days. But in the animals transplanted with SDCs or single-dissociated cells from secondary CS (2nd-S), the injected cells rarely expressed Nkx2.5 (Figure 3f). Interestingly, some Nkx2.5-expressing cells did not express GFP, which suggested that they were resident cardiac cells rather than transplanted ones. These findings suggest that exogenous secondary CS transplantation may also activate endogenous cardiac stem/progenitor cells to differentiate into cardiomyocytes.

At day 7, transplanted 2nd CSs (2nd-H) resided as agglomerates and expressed troponin-T. However, troponin-T expressing cells were round-shaped rather than elongated, striated cardiomyocyte-like cells. At day 14, GFP (+)/troponin-T (+) cells became elongated and striated (Supplementary Figure S4). We confirmed cardiomyogenic differentiation by immunofluorescent staining using antibodies against another structural protein of cardiomyocytes, α-sarcomeric actinin (α-SA) and GFP (Figure 3g, upper panel). The amount of differentiation was quantified by GFP (+) and α-SA area using morphometry; 2nd-H group showed that total infarcted area was 850,327.8 ± 285,843.9 µm2 and the recognized extent of GFP (+) area was 40,653.7 ± 6,376.6 µm2 which represented about 4.8% of the infarcted area, whereas SDC group showed that total infarcted area was 1,640,769.5 ± 361,007.2 µm2 and the extent of GFP (+) area was 19,259.1 ± 1,925.6 µm2 which represented about 1.2% of the infarcted area (P < 0.01 between 2nd-H and SDC group). In the GFP (+) area, α-SA (+) area was 3,457.3 ± 430.4 µm2 in the 2nd-H group and 406.4 ± 63.4 µm2 in the SDC group, which represent 8.5 and 2.1% of engrafted cells, respectively (P < 0.01 between 2nd-H and SDC group).

We also found that some GFP (+) cells formed capillary-like structures and they were positive for ILB4 (Figure 3g, lower panel). These findings indicate that transplanted secondary CSs underwent cardiomyogenic and vasculogenic differentiation.

Taken together, the transplantation of secondary CSs was found to be superior to the transplantation of SDCs, in terms of cellular engraftment and differentiation into cardiovascular lineages.

Secondary CS transplantation prevents adverse cardiac remodeling and achieves functional benefits in the long-term

Short-term (2 weeks) follow-up data demonstrated that secondary CS transplantation in spherical form attenuated negative ventricular remodeling as compared with other groups. However, previous study suggested that short-term therapeutic effects of cell transplantation may not sustain in the long run.31 Therefore, to examine long-term effects of secondary CS transplantation, we followed up the LV dimension and function at 10 weeks after MI and cell injection. Echocardiography showed that LV dimensions at both systole and diastole were significantly smaller and LV systolic function was greater in the 2nd-H group compared with the 2nd-S and SDC groups (Figure 4a). In accordance with the physiologic data, Masson's trichrome staining showed that LV wall was replaced by fibrotic scar tissue and LV dimension was markedly enlarged in the PBS group (Figure 4b). But, in 2nd-H group, LV thickness and dimension were preserved.

Figure 4.

Figure 4

Open in a new tab

Long-term benefits of secondary cardiosphere transplantation. Secondary CS transplantation in spherical form (2nd-H group) prevents detrimental cardiac remodeling and sustains beneficial effects in the long-term. (a) Echocardiographic parameters at 10 weeks (n = 5 per group, *P < 0.05, PBS versus all cell-transplanted groups; **P < 0.05, 2nd-H group versus SDC and 2nd-S groups). (b) Masson's trichrome staining (*P < 0.05, 2nd-H group versus the other three groups). Bar: 1,000 µm. 2nd-H, secondary CS-hanging drop cells as spherical form; 2nd-S, singly dissociated secondary CS cells; CS, cardiosphere; LVEDD, left ventricular end-diastolic dimension; LVEF, LV ejection fraction; LVESD, LV end-systolic dimension; LVFS, LV fractional shortening; PBS, phosphate-buffered saline; SDC, sphere-derived cell.

Taken all together, we conclude that secondary CS transplantation in spherical form prevents detrimental cardiac remodeling after MI in the short-term, sustains beneficial effects in the long-term, and subsequently achieves the improvement of systolic function.

The upregulation of multiple cytokines after sphere formation

In addition to the cardiovascular differentiations of transplanted cells, their paracrine effects are also mainly responsible for the benefits of cell transplantation.32 Thus, we screened for changes in humoral factors before and after sphere formation by microarray analysis. Six hours after sphere initiation, multiple angiovasculogenic factors were upregulated, and these upregulations were maintained for 48 hours (Figure 5a). We validated microarray results by real-time PCR. The mRNA expression of a representative angiogenic cytokine, VEGF, was markedly upregulated from 6 hours after sphere formation, and these expression levels were subsequently maintained (Figure 5b). (Relative expression adjusted by GAPDH (AU). VEGF mRNA (real-time PCR))

Figure 5.

Figure 5

Open in a new tab

The paracrine effects of secondary cardiospheres. (a) Microarray profiles of angiovasculogenic cytokines; 0 hour indicates SDCs before being introduced to secondary CS by a hanging-drop. (b) Real-time PCR of a representative cytokine, VEGF, of neovascularization. (c) The secretion of VEGF protein into culture supernatant after sphere formation as determined by ELISA (n = 4; *P < 0.01, 0 hour versus 24 or 48 hours after sphere formation). (d) Three days after MI and cell transplantation, tissue VEGF mRNA and protein expression was measured by real-time PCR and western blot (n = 3; *P < 0.01). (e) Colocalization with injected GFP (+) cells and VEGF expression at day 3. The GFP (+) transplanted cells were colocalized with VEGF (left panel) and also observed around native GFP(−) capillaries (right panel). (f) Capillary density at day 14. Secondary CS transplantation further enhanced neoangiogenesis (n = 4; *P < 0.05, PBS versus SDC; **P < 0.05, SDC versus 2nd CS). Isolectin B4 (ILB4) for capillary. Bar: 50 µm. α-SA, α-sarcomeric actinin; AU, arbitrary unit; bFGF, basic fibroblast growth factor; CS, cardiosphere; DAPI, 4′,6-diamidino-2-phenylindole; ELISA, enzyme-linked immunosorbent assay; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; IL, interleukin; MI, myocardial infarction; PBS, phosphate-buffered saline; SDC, sphere-derived cell; VEGF, vascular endothelial growth factor.

A secreted form of VEGF protein was upregulated 20-fold at 24 hours after sphere formation and further augmented at 72 hours in conditioned media from secondary CSs, indicating that secondary CSs secrete substantial amounts of angiogenic factors as compared with SDCs (Figure 5c).

Next, to check the in vivo implications, we examined angiogenic humoral effects following cell transplantation. Three days after MI and cell injection, we harvested peri-infarct tissues and measured VEGF mRNA and protein expression by real-time PCR and western blot. Secondary CS transplantation in spherical form (2nd-H) markedly increased tissue VEGF expression (Figure 5d). We also examined the tissue localization of VEGF expression following cell transplantation by immunofluorescence staining. At day 3, we observed robust engraftment of GFP (+) secondary CS-derived cells in the peri-infarct zone. These GFP (+) cells were colocalized with VEGF expression (Figure 5e, left panel). Furthermore, to check the link between VEGF secretion and in vivo capillary growth, we co-stained the tissue with an endothelial cell marker, ILB4. The GFP (+), VEGF (+) transplanted cells were observed around native GFP (−), ILB4 (+) capillaries, which could stimulate vessel growth (Figure 5e, right panel).

In the tissue section at day 14, as a consequence of cell transplantation, we evaluated the degree of neovascularization. The capillary density of the peri-infarct zone in the SDC transplanted group was greater than in the PBS-injected group, indicating that cell transplantation had an angiogenic effect (Figure 5f). The secondary CS transplantation, as compared with the SDC transplantation, showed an additional increase in capillary density, and this correlated well with the reduction of the adverse cardiac remodeling following ischemic injury.

These in vitro and in vivo findings suggest that secondary CS-mediated neovascularization significantly contributes to functional improvements after MI and cell transplantation.

E-selectin/ERK/VEGF autoparacrine signaling loop is responsible for sphere formation

We investigated the molecular mechanisms responsible for sphere formation. By microarray, we examined and normalized 25,697 probes (Supplementary Table S2). By one-way analysis of variance, we chose 5,948 genes which were differently expressed among three time points. Thereafter, we selected early responsive 591 genes which were upregulated in 6 hours compared with 0 and 48 hours (Supplementary Figure S5a). Among 591 genes at the initiation phase of sphere formation, we focused on cell–cell adhesion pathway because dynamic interactions through cell–cell contacts have been known to be important for stem cell maintenance.33 Using Gene Ontology database (http://www.geneontology.org), we selected 12 candidate genes (Supplementary Figure S5b), and we finally focused on E-selectin because the gene expression of E-selectin was found to be upregulated in primary CSs, whereas its expression was minimal in SDCs. During the generation of secondary CSs, E-selectin was induced as early as 3 hours after sphere initiation (Figure 6a, upper panel). We confirmed the protein expression of E-selectin by immunofluorescent 3D confocal imaging (Figure 6a, lower panel). In addition, high magnification image validated the surface expression of E-selectin (Supplementary Figure S6).

Figure 6.

Figure 6

Open in a new tab

Molecular mechanisms of sphere formation. (a) Primary CS and secondary CS highly expressed E-selectin, but SDC did not. Gene expressions were measured by RT-PCR and protein expressions were confirmed by immunofluorescent staining. (b) The cells from E-selectin KO cardiac explants were not able to generate CSs. CD44 (a ligand of E-selectin) is expressed in both KO and WT mice. Bar: 200 µm. (c) Western blots at 12 hours after sphere formation demonstrated that E-selectin expression was induced following sphere formation. Blocking ERK pathway (E) suppressed E-selectin expression, but blocking AKT (A) pathway did not. When ERK was blocked, Sp1 was downregulated. (d) After 24 hours, ERK inhibition significantly inhibited VEGF expression, indicating VEGF participates in the later stages of sphere formation (*P < 0.01, ERK block versus the WT and versus an AKT block). (e) Signal pathway blocking experiments demonstrated that the ERK pathway is responsible for sphere initiation and maturation. Average sphere areas were measured using pixel values (*P < 0.01, ERK block versus the WT; **P > 0.05, AKT block versus the WT). Bar: 200 µm. (f) Sphere maturation at 48 hours was markedly decreased by siRNA, targeting VEGF (*P < 0.01, control siRNA versus VEGF siRNA). Bar: 200 µm. AU, arbitrary unit; CS, cardiosphere; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; KO, knock-out; RT-PCR, reverse transcription-PCR; SDC, sphere-derived cell; siRNA, small interfering RNA; VEGF, vascular endothelial growth factor; WT, wild-type.

To address causal relationships and functional significance of E-selectin, we harvested hearts from E-selectin knock-out (KO) mice which were viable and fertile but impaired angiogenesis and postnatal vasculogenesis.34 We were not able to generate spheres (Figure 6b). Furthermore, when we knocked down E-selectin by siRNA in wild-type SDCs, sphere initiation was significantly retarded following hanging-drop (Supplementary Figure S7). Taken together, these data indicate that the induction of E-selectin expression is indispensable to sphere initiation.

We next analyzed the key signaling pathway responsible for sphere formation. Neurosphere studies have suggested that PI3K/AKT and MAPK/ERK signaling pathways are required for sphere maintenance and stem cell survival.35,36 When secondary CS was initiated, we found that the ERK pathway was activated whereas the AKT pathway was slightly suppressed (Figure 6c). Next, when the ERK pathway was blocked, E-selectin expression decreased and sphere formation was retarded, but AKT inhibition did not affect sphere formation (Figure 6c,e). In addition to measuring the average sphere area following each pathway blocking, we also counted the number of cells per sphere (Figure 6e). We hung drop 5,000 cells, and 48 hours later, we dissociated each sphere and counted cell numbers. Control group showed 5,572 ± 655 cells/sphere, whereas ERK-blocked group showed 525 ± 156 cells/sphere (P < 0.01). However, AKT-blocked group revealed no significant decrease of cell numbers, compared with control group (4,748 ± 582 cells/sphere, P > 0.05). Taken together, these data indicated that the ERK/E-selectin pathway is, at least in part, responsible for sphere initiation.

Interestingly, when we blocked the ERK pathway, Sp1 was subsequently downregulated (Figure 6c). Sp1 is a well-known transcription factor of VEGF.37,38 Thus, we checked VEGF expression, and found that ERK inhibition markedly downregulated VEGF expression 24 hours after sphere initiation (Figure 6d). Furthermore, when VEGF was blocked by siRNA, sphere maturation was remarkably decreased (Figure 6f). These findings indicate that an ERK-dependent Sp1 pathway plays a key role in VEGF synthesis and that VEGF autoparacrine loop is responsible for sphere maturation.

With the blocker concentration we have used to inhibit the certain signal, the blockade of ERK, VEGF, and AKT did not affect the survival and proliferation of secondary CS cells and attachment of secondary CSs (Supplementary Figure S8).

Collectively, these findings indicate that E-selectin upregulation via ERK activation is responsible for the initiation of sphere formation and that ERK-dependent Sp1 upregulation and VEGF secretion are required for the maturation of secondary CS (Figure 7).

Figure 7.

Figure 7

Open in a new tab

The concept of secondary cardiospheres for repairing damaged heart. ERK/E-selectin and ERK/Sp1/VEGF signaling cascades underlie the molecular mechanisms of sphere initiation and maturation. The transplantation of secondary CSs enhanced cellular engraftment, cardiovascular differentiation, and humoral factor secretion. CS, cardiosphere; SDC, sphere-derived cell; VEGF, vascular endothelial growth factor.

In addition, when we overexpressed E-selectin in SDCs before sphere formation, VEGF expression was augmented by more than threefolds following sphere initiation (Supplementary Figure S9), suggesting that the degree of E-selectin induction during sphere formation may be an important factor for beneficial effects of cell transplantation.

Discussion

Functional cardiac repair by cell-based therapy has been extensively investigated over the last decade.7,13 The optimal cell type required for cardiac repair should be available in sufficient numbers, engraft efficiently when transplanted into injured hearts, possess the multipotency to differentiate into cardiovascular lineages, and secrete cardioprotective humoral factors. Furthermore, optimal cell transplantation should sustain the beneficial effects and maintain functional improvement of the injured heart at long-term follow-up. In the present study, we hypothesized that repeated three-dimensional sphere formation would meet these requisites (that is, cardiac explants (5–10 pieces of tissues: total 15 mg) → migrating cells from explant → primary CS (about 100 spheres/35 mm PDL dish) formation → SDCs (5 × 106 cells/100 mm dish) in adherent culture condition → secondary CS formation by three-dimensional culture (more than 5,000 spheres)). We here demonstrate that; (i) primary CSs express Oct4 and c-kit, but when expanded in a two-dimensional culture system, SDCs lose these characteristics, (ii) using the three-dimensional sphere formation strategy, secondary CS can be generated from monolayer SDCs, re-gain the expressions of multipotency markers and effectively differentiate into cardiovascular lineages, (iii) secondary CS transplantation promotes cardiac repair after MI more efficiently than SDC transplantation by increasing cellular engraftment, cardiovascular differentiation, endogenous cardiac stem/progenitor cell activation, and neovascularization by humoral factor secretion, (iv) at long-term follow-up, secondary CS transplantation in spherical form prevents detrimental cardiac remodeling and sustains beneficial effects, (v) the ERK/E-selectin and the ERK/Sp1/VEGF signaling cascades underlie the molecular mechanisms of sphere initiation and maturation. A schematic concept of secondary CS for repairing the ischemic heart is provided in Figure 7.

Primary sphere formation and subsequent secondary sphere generation are a privilege of neural stem cells.39,40,41 Stem/progenitor cells in secondary neurospheres retain proliferation and differentiation potentials whereas cells in the attached condition readily lose these potentials.39 Consistent with neurosphere study, our results show the primary CSs express the multipotency markers, but that SDCs gradually lost these expressions, suggesting that cellular potency is likely to diminish when sufficient cell numbers are acquired using conventional two-dimensional expansion. Thus, we generated secondary CSs from SDCs using three-dimensional culture. Secondary CSs have restored Oct4 and c-kit expressions and exhibited potent differentiation potentials, indicating that repeated sphere formation may be useful for selecting stem/progenitor cells, eliminates differentiated cells, and provides a microenvironment mimicking in vivo niche.16

Cellular engraftment following transplantation is one of the most important factors of cell-based cardiac repair. Irrespective of the type of cells, we, like other researchers,20,21,22,30 observed that transplanted cells disappeared rapidly after being injected to the heart. In this study, we repeatedly observed that the number of engrafted cells declined rapidly by day 14 following transplantation. Therefore, we have tried to search a new cell preparation method to improve engraftment. When we compared secondary CSs with SDCs, we found that multiple adhesion molecules and cytokines were upregulated after sphere formation. Specifically, E-selectin and VEGF were strongly induced in secondary CSs, which have been reported to be required for cellular engraftment and survival.34 When transplanted in vivo, secondary CSs in spherical form showed substantially greater engraftment efficiency than SDCs and single-dissociated cells from secondary CSs. These findings strongly suggest that spheres have advantages provided by the three-dimensional microenvironment which maintains cell-to-cell interactions and aids engraftment.42 In addition to short-term benefits of cell transplantation,31 secondary CS transplantation in spherical form prevented detrimental cardiac remodeling and maintained functional improvement in the long-term.

Regarding the mechanisms responsible for the beneficial effects of cell-based therapy, humoral factors secreted by transplanted cells also contribute to the functional benefits.32 In this study, we monitored changes in humoral factors before and after sphere formation. Secondary CSs expressed and secreted 20-fold more VEGF than SDCs. Consistent with our in vitro findings, secondary CSs increased capillary density in vivo.

Interestingly, GFP (+) secondary CS transplantation increased the number of GFP (−)/Nkx2.5 (+)/Troponin T (−) cells in peri-infarct areas, indicating the activations of endogenous cardiac stem/progenitor cells. These findings indicate that secondary CS-mediated neovascularization and activation of endogenous regenerating cells in a paracrine manner significantly contributed to the beneficial effects of cell transplantation.

During the preparation of the manuscript, Marban group reported the consistent experimental findings with our results, which have demonstrated that the transplantation of cells in spherical form would be better than the transplantation in single-dissociated form and VEGF expression is higher in secondary CSs than in CDCs.16,43 Interestingly, VEGF expression level of primary CSs is similar to that of secondary CSs. However, primary CSs have an actual limitation to acquire sufficient cell numbers from a biopsy for in vivo cell transplantation albeit only enough to perform in vitro work such as PCR or staining. Therefore, we here have focused on CDCs (or SDCs in this study) and secondary CSs in further in vitro cellular and molecular experiments as well as in vivo transplantation.

The molecular mechanisms of sphere formation have remained elusive, although previous neurosphere studies have reported that PI3K/AKT and MAPK/ERK signaling pathways are involved in sphere maintenance and cell survival.35,36 In this study, we hypothesized that sphere formation could be divided into two stages (sphere initiation and maturation). During early stage, we presumed that adhesion molecules would play a critical role in cell–cell reaggregation. Surprisingly, E-selectin expression was found to be induced as early as 3 hours after sphere formation. Furthermore, explant-migrating cells from E-selectin KO mice were not able to initiate cell–cell reaggregation, and when we knocked down E-selectin by siRNA in wild-type SDCs, sphere initiation was significantly retarded, indicating that the upregulation of E-selectin is indispensable to sphere initiation. Regarding the upstream signaling pathway involved, when ERK pathway was blocked, E-selectin expression was decreased and sphere initiation was retarded. For sphere maturation, based on gene expression profiles at 48 hours after sphere formation, we hypothesized that autoparacrine pathways were responsible. When the ERK pathway was blocked, Sp1 was downregulated, and subsequently VEGF expression 24 hours after sphere initiation was markedly reduced. Furthermore, blocking VEGF with siRNA or antibody remarkably decreased sphere maturation, suggesting VEGF participates in the later stages of sphere formation. Together, we propose that ERK/E-selectin and ERK/Sp1/VEGF signaling cascades underlie sphere initiation and maturation.

In conclusion, we describe a simple but effective means of facilitating cardiac stem/progenitor cell-based therapy. Although the generation of human secondary CS from such as endomyocardial biopsy specimen of the patient and preclinical and clinical translations for repairing the damaged heart should be further tested, we here report the possibility of therapeutic benefits of secondary CS. Our findings demonstrate that secondary CS formation has three advantages for cardiac repair; (i) it increases multipotency and enriches stem/progenitor cells in vitro, (ii) it reinforces cell engraftment and differentiation when transplanted in vivo, and (iii) it augments paracrine humoral effects. Furthermore, this repeated sphere formation strategy may be applicable to other types of stem/progenitor cells for cell-based therapy.

Materials and Methods

Animals. Wild-type C57BL/6J, β-actin promoter-driven eGFP-expressing mice, Oct4-promoter–driven GFP (B6;CBA-Tg(Pou5f1-EGFP)2Mnn) mice, and E-selectin knock-out (B6;129S2-Seletm1Hyn/J) mice34 were purchased from Jackson Laboratory (Bar Harbor, ME). Balb/c athymic nude mice were from Orient Bio, Seongnam-si, Korea. All animal experiments were performed after receiving approval from the Institutional Animal Care and Use Committee at Seoul National University Hospital, Seoul, Korea.

Cell isolation and CS generation. Harvested hearts were chopped into pieces smaller than 1 mm2 using sterile instruments and digested three times for 5 minutes at 37 °C with collagenase type IV and trypsin 0.2% in shaking incubator. Explants were cultured on FN-coated dishes in basal media (IMDM, 10% FBS, 2 mmol/l L-glutamine, 0.1 mmol/l 2-mercaptoethanol, and antibiotics)10 for 7 days to propagate phase-bright cells. These were detached with 0.05% trypsin with minimal contamination of non-phase–bright cells. Cells were then cultured on PDL-coated dishes (BD, Franklin Lakes, NJ) in growth media (35% IMDM/65% DMEM/F12, 3.5% FBS, 2% B27, 20 ng/ml epidermal growth factor, 40 ng/ml bFGF, 4 nmol/l cardiotrophin-1, 50 nmol/l thrombin, 2 mmol/l L-glutamine, 0.1 mmol/l 2-mercaptoethanol, and antibiotics)10 and primary CSs formed within 5 days. Primary CSs were attached to 35 mm FN-coated dishes at 50–100/dish and subcultured at a ratio of 1:2–1:3 every 3–4 days to acquire sufficient numbers of monolayer SDCs, which were detached from dishes using trypsin and used to generate secondary CSs by culture for 48 hours on PDL dishes as performed for primary CSs or cultured using the hanging-drop method.25,28 Each drop contained ~1,000 cells. Briefly, 20 µl drop of fluid was attached on the lid of petridish. And bottom plates of the petridish contained with minimum 20 ml PBS in dish (100 mm) to avoid desiccation of drop. Lid was gently inverted on the bottom dish. All cells were cultured in the growth media except for explants culture in the basal media.

Analysis of gene and protein expressions, and signaling pathways

Microarray. Global gene expression analyses were performed using Mouse Whole-genome BeadChips (Illumina, San Diego, CA). Samples were prepared as described by the manufacturer. To detect different probes (differentially expressed genes) among XX, YY, and ZZ, one-way analysis of variance was used and Benjamin & Hochberg false discovery rate was used. Significance was accepted for false discovery rates of <5% and P value of <0.01. Cells were harvested and total RNA was extracted using TRIZOL (Life Technologies, Carlsbad, CA).

Reverse transcription-PCR. Conventional reverse transcription-PCR and quantitative real-time PCR were performed using ABI7200 (Applied Biosystems, Carlsbad, CA) as described previously.44 The primers used are listed in Supplementary Table S1. Total RNA was extracted using TRIZOL (Life Technologies).

Western blot. Protein extracts (20 µg per sample) from cells were separated by SDS-PAGE (Bio-Rad Laboratories, Hercules, CA) and electro-transferred.45 Membranes were probed with E-selectin (Santa Cruz Biotechnology, Santa Cruz, CA), total/phosphor-AKT (Cell Signaling), total/phosphor-ERK (Cell Signaling Technology, Boston, MA), Sp1 (Santa Cruz Biotechnology), and α-tubulin (Calbiochem, Darmstadt, Germany) antibodies.

Enzyme-linked immunosorbent assay. To check VEGF secretion, cells were cultured in cytokine-free, 0.5% FBS containing media (35% IMDM and 65% DMEM/F12) 24 hours before experiments. Supernatants were collected and enzyme-linked immunosorbent assay was performed according to the manufacturer's instructions (R&D Systems, Minneapolis, MN).

For signal blocking experiments, we applied U0126 (A.G.Scientific) to block ERK46 and LY294002 (A.G.Scientific, San Diego, CA) to block AKT.47 The siRNA for VEGF and a scrambled control were purchased from Santa Cruz Biotechnology.

In vitro differentiation

Preparation of lineage tracing vectors. To assess in vitro differentiation potentials, we applied lineage tracing lentiviral vectors before initiating differentiation. Self-inactivating lentiviral vectors containing the eGFP gene and a woodchuck post-transcriptional regulatory element (WPRE) were generated as previously described26,48 by transient transfection in 293T cells with packaging plasmid, pCMV8.91 and pMD.G for vesicular stomatitis virus–G protein (VSV-G) pseudotyping.49 eGFP expression was controlled using constitutively active promoter (spleen focus forming virus promoter: SFFVp), endothelial nitric oxide synthase promoter, VEGFR2 promoter, or α-myosin heavy chain promoter for lineage tracing. Lentiviral particles were collected every 24 hours for 3 days, and filtered through 0.22 µm filters. To transduce lentivirus into cells, cells were incubated in fresh growth media mixed with growth media containing lentiviral particles (1:1 volume ratio) for 24 hours. After three washes and a medium change, cells were cultured for another 24 hours, detached and subjected to differentiation assays.

Endothelial differentiation. Cells were plated on type IV collagen-coated dishes and cultured in endothelial growth media (EGM2-MV; Lonza, Basel, Switzerland) containing 5% FBS and additional VEGF (50 ng/ml) for 7 days. Endothelial differentiation was assessed by monitoring eGFP fluorescence from lineage tracing vectors, ILB4 staining.

Cardiomyogenic differentiation. Cells were plated on 1.5% gelatin-coated dishes and cocultured with neonatal rat cardiomyocytes in differentiation media (DMEM/F12, 20% FBS, 10 ng/ml BMP2, and 10 ng/ml TGF-β).27 Rat neonatal cardiomyocytes were prepared as described previously.50 Cardiomyogenic differentiation was assessed by connexin 43 staining or monitoring eGFP fluorescence under α-myosin heavy chain promoter. To identify electrophysiologic coupling and cellular excitability, cytosolic (Ca2+) response was measured using fluorescence imaging as described previously.27 To distinguish secondary CSs from rat cardiomyocytes, we used cells from the heart of a β-actin promoter-driven eGFP-expressing mouse. Cells were loaded by incubation with 2 µmol/l fura-2 AM (Sigma-Aldrich, St Louis, MO) plus 0.1 % Pluronic F-127 in normal tyrode solution for 15 minutes at room temperature. For fluorescence excitation, we used a polychromatic light source (xenon-lamp based, Polychrome-IV; TILL-Photonics, Martinsried, Germany), which was coupled into the epi-illumination port of an upright microscope (BX51WI; Olympus, Tokyo, Japan) via a quartz light guide and an UV condenser. Cell imaging was performed with a ×40 water immersion objective (NA 0.8, LUMPlanFl; Olympus) and an air-cooled slow-scan CCD camera (SensiCam, PCO, Kelheim, Germany). The monochromator and the CCD camera were controlled by a PC and ITC18, running a custom-made software programmed with Microsoft Visual C++ (version 6.0). Standard two-wavelength protocol was used for fluorescence imaging of cells. Images were taken at 10 Hz with double wavelength excitation at 340 nm (F340) and 380 nm (F380). The ratio of fluorescence intensity, F340/F380, was indicated as calcium transient of each cell. The normal tyrode solution contained (in mmol/l): NaCl 148, KCl 5, CaCl2 2, MgCl2 1, glucose 10, HEPES 10, adjusted to pH 7.4 with NaOH.

MI induction and cell transplantation. Mice were anesthetized with Zoletil (91 mg/kg intraperitoneally; Virbac, Carros, France) and Xylazine (11.65 mg/kg intraperitoneally; Bayer, Holliston, MA), intubated and artificially ventilated (Harvard Apparatus). Through left-sided thoracotomy, the heart was exposed and the left anterior descending artery was ligated using 8-0 polypropylene sutures and a dissecting microscope. The apex of the LV was observed for evidence of myocardial blanching and akinesia indicating interruption in coronary flow. Thereafter, 1 × 105 GFP cells in 50 µl of PBS were directly transplanted into peri-infarct areas and the apex (total three sites) using a 31G needle.20 The same volume of PBS was injected in an identical fashion as a control. The chest wall was closed after draining out all bloody fluid.

Physiologic assessments and histologic measurements. Two and ten weeks after MI and cell transplantation, LV size and function were evaluated by echocardiography using a 15.0-MHz ultraband linear transducer (Aplio XG; Toshiba, Tokyo, Japan). LV dimensions at end systole and end diastole were measured and fractional shortening and ejection fraction were calculated as previously described.20 After an echocardiographic examination, hearts were perfused retrogradely through the right carotid artery with PBS containing 4% paraformaldehyde. Tissues were embedded in OCT compound (Sakura Finetek, Torrance, CA) or paraffin. In 2 weeks short-term follow-up study, transverse tissue sections at the papillary muscle level of LV corresponding to M-mode echo were analyzed with Masson's trichrome staining, and in the 10 weeks long-term follow-up study, sagittal tissue sections were analyzed. Infarct size (the percentage of fibrosis length and infarct thicknesses) was measured using a digital image-analysis system (Image Pro, version 4.5; MediaCybernetics, Bethesda, MD). To count engrafted GFP cells, 10 randomly selected fields were examined per each section under fluorescent microscopy, and the number of positive cells was converted to numbers per square millimeter.20 The quantitative analysis was performed with single-blinded manner.

Quantification of cellular engraftment after transplantation. To quantify the degree of cellular engraftment, male cells were transplanted to female infarcted heart. Hearts were harvested immediately after cell injection at day 3, day 7, and day 14, and extracted DNA. The engrafted male cells were quantified by real-time PCR using Y chromosome-specific (sex-determining region of Y chromosome, Sry) primers and probe as we previously described.30 In brief, the known number of female and male cells was mixed and all genomic DNA was extracted. Based on amplification plots and a strand sample curve (Figure 3c), absolute number of male cells from unknown cell numbers was calculated.

Immunostaining and microscopic examinations. Immunofluorescent staining was performed as previously described with minor modifications.20,27 Cells or tissue sections were incubated with antibodies against GFP (Abcam, Cambridge, UK), Oct4 (Santa Cruz Biotechnology, BD), c-kit (BD), Connexin 43 (Santa Cruz Biotechnology), Troponin T (Santa Cruz Biotechnology), and Nkx2.5 (Abcam). DAPI, TO-PRO-3 or SYTOX blue (Invitrogen, Carlsbad, CA) were used for nuclear counterstaining. To identify capillaries, ILB4 (Vector Laboratories, Burlingame, CA) was used. Images were acquired using confocal microscopy (LSM710; Zeiss, Oberkochen, Germany). For quantification purposes, at least five randomly selected fields were examined and averaged. Numbers of positive cells are presented as number of cells per square millimeter.

Statistical analysis. All data are presented as means ± SEM. The Student's t-test or one-way analysis of variance with Bonferroni's correction were used for intergroup comparisons. SPSS version 12.0 (SPSS, Chicago, IL) was used throughout, and P values of <0.05 were considered to denote statistical significance.

SUPPLEMENTARY MATERIAL Figure S1. Flow cytometry analysis of SDCs. Figure S2. Oct4 expression of secondary CS. Figure S3 and Video S1 and S2. Cardiomyogenic differentiation of secondary CS. Figure S4. Cardiovascular differentiation of engrafted secondary CSs. Figure S5. Microarray analysis to dissect the molecular mechanisms responsible for sphere formation. Figure S6. E-selectin expression in secondary CSs. Figure S7. E-selectin knock-down in SDCs and secondary CS formation. Figure S8. With the blocker concentration we have used to inhibit the certain signal, the blockade of ERK, VEGF, and AKT did not affect the survival and proliferation of secondary CS cells and attachment of secondary CSs. Figure S9. E-selectin overexpression in SDCs and secondary CS formation. Table S1. Primer sets used RT-PCR. Table S2. Whole microarray dataset (25,697 genes) provided with Excel file.

Acknowledgments

This study was supported by a grant from the Innovative Research Institute for Cell Therapy (A062260) and National Research Foundation funded by the Korea Government (MEST) (2010-0020258), Republic of Korea. H.-S.K. is also a professor, Molecular Medicine & Biopharmaceutical Sciences, Seoul National University, sponsored by World Class University program of the Ministry of Education, Science & Technology, Republic of Korea. The authors declared no conflict of interest.

Supplementary Material

Figure S1.

Flow cytometry analysis of SDCs.

Click here for additional data file. (12MB, pdf)
Figure S2.

Oct4 expression of secondary CS.

Click here for additional data file. (12MB, pdf)
Figure S3 and Video S1 and S2.

Cardiomyogenic differentiation of secondary CS.

Click here for additional data file. (12MB, pdf)
Figure S4.

Cardiovascular differentiation of engrafted secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S5.

Microarray analysis to dissect the molecular mechanisms responsible for sphere formation.

Click here for additional data file. (12MB, pdf)
Figure S6.

E-selectin expression in secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S7.

E-selectin knock-down in SDCs and secondary CS formation.

Click here for additional data file. (12MB, pdf)
Figure S8.

With the blocker concentration we have used to inhibit the certain signal, the blockade of ERK, VEGF, and AKT did not affect the survival and proliferation of secondary CS cells and attachment of secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S9.

E-selectin overexpression in SDCs and secondary CS formation.

Click here for additional data file. (12MB, pdf)
Table S1.

Primer sets used RT-PCR.

Click here for additional data file. (58.5KB, doc)
Table S2.

Whole microarray dataset (25,697 genes) provided with Excel file.

Click here for additional data file. (11.9MB, xls)

REFERENCES

  1. Chavakis E, Koyanagi M., and, Dimmeler S. Enhancing the outcome of cell therapy for cardiac repair: progress from bench to bedside and back. Circulation. 2010;121:325–335. doi: 10.1161/CIRCULATIONAHA.109.901405. [DOI] [PubMed] [Google Scholar]
  2. Murry CE, Reinecke H., and, Pabon LM. Regeneration gaps: observations on stem cells and cardiac repair. J Am Coll Cardiol. 2006;47:1777–1785. doi: 10.1016/j.jacc.2006.02.002. [DOI] [PubMed] [Google Scholar]
  3. Reinecke H, Minami E, Zhu WZ., and, Laflamme MA. Cardiogenic differentiation and transdifferentiation of progenitor cells. Circ Res. 2008;103:1058–1071. doi: 10.1161/CIRCRESAHA.108.180588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Lipinski MJ, Biondi-Zoccai GG, Abbate A, Khianey R, Sheiban I, Bartunek J.et al. (2007Impact of intracoronary cell therapy on left ventricular function in the setting of acute myocardial infarction: a collaborative systematic review and meta-analysis of controlled clinical trials J Am Coll Cardiol 501761–1767. [DOI] [PubMed] [Google Scholar]
  5. Abdel-Latif A, Bolli R, Tleyjeh IM, Montori VM, Perin EC, Hornung CA.et al. (2007Adult bone marrow-derived cells for cardiac repair: a systematic review and meta-analysis Arch Intern Med 167989–997. [DOI] [PubMed] [Google Scholar]
  6. Kajstura J, Urbanek K, Perl S, Hosoda T, Zheng H, Ogórek B.et al. (2010Cardiomyogenesis in the adult human heart Circ Res 107305–315. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  7. Laflamme MA., and, Murry CE. Heart regeneration. Nature. 2011;473:326–335. doi: 10.1038/nature10147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, Chimenti S.et al. (2003Adult cardiac stem cells are multipotent and support myocardial regeneration Cell 114763–776. [DOI] [PubMed] [Google Scholar]
  9. Oh H, Bradfute SB, Gallardo TD, Nakamura T, Gaussin V, Mishina Y.et al. (2003Cardiac progenitor cells from adult myocardium: homing, differentiation, and fusion after infarction Proc Natl Acad Sci USA 10012313–12318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Messina E, De Angelis L, Frati G, Morrone S, Chimenti S, Fiordaliso F.et al. (2004Isolation and expansion of adult cardiac stem cells from human and murine heart Circ Res 95911–921. [DOI] [PubMed] [Google Scholar]
  11. Pfister O, Mouquet F, Jain M, Summer R, Helmes M, Fine A.et al. (2005CD31- but Not CD31+ cardiac side population cells exhibit functional cardiomyogenic differentiation Circ Res 9752–61. [DOI] [PubMed] [Google Scholar]
  12. Smits AM, van Laake LW, den Ouden K, Schreurs C, Szuhai K, van Echteld CJ.et al. (2009Human cardiomyocyte progenitor cell transplantation preserves long-term function of the infarcted mouse myocardium Cardiovasc Res 83527–535. [DOI] [PubMed] [Google Scholar]
  13. Leri A, Kajstura J., and, Anversa P. Role of cardiac stem cells in cardiac pathophysiology: a paradigm shift in human myocardial biology. Circ Res. 2011;109:941–961. doi: 10.1161/CIRCRESAHA.111.243154. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  14. Smith RR, Barile L, Cho HC, Leppo MK, Hare JM, Messina E.et al. (2007Regenerative potential of cardiosphere-derived cells expanded from percutaneous endomyocardial biopsy specimens Circulation 115896–908. [DOI] [PubMed] [Google Scholar]
  15. Davis DR, Zhang Y, Smith RR, Cheng K, Terrovitis J, Malliaras K.et al. (2009Validation of the cardiosphere method to culture cardiac progenitor cells from myocardial tissue PLoS ONE 4e7195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Li TS, Cheng K, Lee ST, Matsushita S, Davis D, Malliaras K.et al. (2010Cardiospheres recapitulate a niche-like microenvironment rich in stemness and cell-matrix interactions, rationalizing their enhanced functional potency for myocardial repair Stem Cells 282088–2098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Davis DR, Kizana E, Terrovitis J, Barth AS, Zhang Y, Smith RR.et al. (2010Isolation and expansion of functionally-competent cardiac progenitor cells directly from heart biopsies J Mol Cell Cardiol 49312–321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Terrovitis J, Lautamäki R, Bonios M, Fox J, Engles JM, Yu J.et al. (2009Noninvasive quantification and optimization of acute cell retention by in vivo positron emission tomography after intramyocardial cardiac-derived stem cell delivery J Am Coll Cardiol 541619–1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Terrovitis JV, Smith RR., and, Marbán E. Assessment and optimization of cell engraftment after transplantation into the heart. Circ Res. 2010;106:479–494. doi: 10.1161/CIRCRESAHA.109.208991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cho HJ, Lee N, Lee JY, Choi YJ, Ii M, Wecker A.et al. (2007Role of host tissues for sustained humoral effects after endothelial progenitor cell transplantation into the ischemic heart J Exp Med 2043257–3269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li Z, Lee A, Huang M, Chun H, Chung J, Chu P.et al. (2009Imaging survival and function of transplanted cardiac resident stem cells J Am Coll Cardiol 531229–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Laflamme MA, Chen KY, Naumova AV, Muskheli V, Fugate JA, Dupras SK.et al. (2007Cardiomyocytes derived from human embryonic stem cells in pro-survival factors enhance function of infarcted rat hearts Nat Biotechnol 251015–1024. [DOI] [PubMed] [Google Scholar]
  23. Shenje LT, Field LJ, Pritchard CA, Guerin CJ, Rubart M, Soonpaa MH.et al. (2008Lineage tracing of cardiac explant derived cells PLoS ONE 3e1929. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Andersen DC, Andersen P, Schneider M, Jensen HB., and, Sheikh SP. Murine “cardiospheres” are not a source of stem cells with cardiomyogenic potential. Stem Cells. 2009;27:1571–1581. doi: 10.1002/stem.72. [DOI] [PubMed] [Google Scholar]
  25. Dani C, Smith AG, Dessolin S, Leroy P, Staccini L, Villageois P.et al. (1997Differentiation of embryonic stem cells into adipocytes in vitro J Cell Sci 110 (Pt 11)1279–1285. [DOI] [PubMed] [Google Scholar]
  26. Yoon CH, Koyanagi M, Iekushi K, Seeger F, Urbich C, Zeiher AM.et al. (2010Mechanism of improved cardiac function after bone marrow mononuclear cell therapy: role of cardiovascular lineage commitment Circulation 1212001–2011. [DOI] [PubMed] [Google Scholar]
  27. Chang SA, Lee EJ, Kang HJ, Zhang SY, Kim JH, Li L.et al. (2008Impact of myocardial infarct proteins and oscillating pressure on the differentiation of mesenchymal stem cells: effect of acute myocardial infarction on stem cell differentiation Stem Cells 261901–1912. [DOI] [PubMed] [Google Scholar]
  28. Dang SM, Kyba M, Perlingeiro R, Daley GQ., and, Zandstra PW. Efficiency of embryoid body formation and hematopoietic development from embryonic stem cells in different culture systems. Biotechnol Bioeng. 2002;78:442–453. doi: 10.1002/bit.10220. [DOI] [PubMed] [Google Scholar]
  29. Ziebart T, Yoon CH, Trepels T, Wietelmann A, Braun T, Kiessling F.et al. (2008Sustained persistence of transplanted proangiogenic cells contributes to neovascularization and cardiac function after ischemia Circ Res 1031327–1334. [DOI] [PubMed] [Google Scholar]
  30. Lee DW, Lee TK, Cho IS, Park HE, Jin S, Cho HJ.et al. (2011Creation of myocardial fibrosis by transplantation of fibroblasts primed with survival factors Am J Physiol Heart Circ Physiol 301H1004–H1014. [DOI] [PubMed] [Google Scholar]
  31. van Laake LW, Passier R, Monshouwer-Kloots J, Verkleij AJ, Lips DJ, Freund C.et al. (2007Human embryonic stem cell-derived cardiomyocytes survive and mature in the mouse heart and transiently improve function after myocardial infarction Stem Cell Res 19–24. [DOI] [PubMed] [Google Scholar]
  32. Gnecchi M, Zhang Z, Ni A., and, Dzau VJ. Paracrine mechanisms in adult stem cell signaling and therapy. Circ Res. 2008;103:1204–1219. doi: 10.1161/CIRCRESAHA.108.176826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Peerani R., and, Zandstra PW. Enabling stem cell therapies through synthetic stem cell-niche engineering. J Clin Invest. 2010;120:60–70. doi: 10.1172/JCI41158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Oh IY, Yoon CH, Hur J, Kim JH, Kim TY, Lee CS.et al. (2007Involvement of E-selectin in recruitment of endothelial progenitor cells and angiogenesis in ischemic muscle Blood 1103891–3899. [DOI] [PubMed] [Google Scholar]
  35. Campos LS, Leone DP, Relvas JB, Brakebusch C, Fässler R, Suter U.et al. (2004Beta1 integrins activate a MAPK signalling pathway in neural stem cells that contributes to their maintenance Development 1313433–3444. [DOI] [PubMed] [Google Scholar]
  36. Ishii S, Okada Y, Kadoya T, Matsuzaki Y, Shimazaki T., and, Okano H. Stromal cell-secreted factors promote the survival of embryonic stem cell-derived early neural stem/progenitor cells via the activation of MAPK and PI3K-Akt pathways. J Neurosci Res. 2010;88:722–734. doi: 10.1002/jnr.22250. [DOI] [PubMed] [Google Scholar]
  37. Curry JM, Eubank TD, Roberts RD, Wang Y, Pore N, Maity A.et al. (2008M-CSF signals through the MAPK/ERK pathway via Sp1 to induce VEGF production and induces angiogenesis in vivo PLoS ONE 3e3405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ryuto M, Ono M, Izumi H, Yoshida S, Weich HA, Kohno K.et al. (1996Induction of vascular endothelial growth factor by tumor necrosis factor alpha in human glioma cells. Possible roles of SP-1 J Biol Chem 27128220–28228. [DOI] [PubMed] [Google Scholar]
  39. Galli R, Gritti A, Bonfanti L., and, Vescovi AL. Neural stem cells: an overview. Circ Res. 2003;92:598–608. doi: 10.1161/01.RES.0000065580.02404.F4. [DOI] [PubMed] [Google Scholar]
  40. Rietze RL, Valcanis H, Brooker GF, Thomas T, Voss AK., and, Bartlett PF. Purification of a pluripotent neural stem cell from the adult mouse brain. Nature. 2001;412:736–739. doi: 10.1038/35089085. [DOI] [PubMed] [Google Scholar]
  41. Ma BF, Liu XM, Xie XM, Zhang AX, Zhang JQ, Yu WH.et al. (2006Slower cycling of nestin-positive cells in neurosphere culture Neuroreport 17377–381. [DOI] [PubMed] [Google Scholar]
  42. Hattori F, Chen H, Yamashita H, Tohyama S, Satoh YS, Yuasa S.et al. (2010Nongenetic method for purifying stem cell-derived cardiomyocytes Nat Methods 761–66. [DOI] [PubMed] [Google Scholar]
  43. Chimenti I, Smith RR, Li TS, Gerstenblith G, Messina E, Giacomello A.et al. (2010Relative roles of direct regeneration versus paracrine effects of human cardiosphere-derived cells transplanted into infarcted mice Circ Res 106971–980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yoon CH, Hur J, Park KW, Kim JH, Lee CS, Oh IY.et al. (2005Synergistic neovascularization by mixed transplantation of early endothelial progenitor cells and late outgrowth endothelial cells: the role of angiogenic cytokines and matrix metalloproteinases Circulation 1121618–1627. [DOI] [PubMed] [Google Scholar]
  45. Cho HJ, Youn SW, Cheon SI, Kim TY, Hur J, Zhang SY.et al. (2005Regulation of endothelial cell and endothelial progenitor cell survival and vasculogenesis by integrin-linked kinase Arterioscler Thromb Vasc Biol 251154–1160. [DOI] [PubMed] [Google Scholar]
  46. Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS.et al. (1998Identification of a novel inhibitor of mitogen-activated protein kinase kinase J Biol Chem 27318623–18632. [DOI] [PubMed] [Google Scholar]
  47. Vlahos CJ, Matter WF, Hui KY., and, Brown RF. A specific inhibitor of phosphatidylinositol 3-kinase, 2-(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one (LY294002) J Biol Chem. 1994;269:5241–5248. [PubMed] [Google Scholar]
  48. Koyanagi M, Iwasaki M, Rupp S, Tedesco FS, Yoon CH, Boeckel JN.et al. (2010Sox2 transduction enhances cardiovascular repair capacity of blood-derived mesoangioblasts Circ Res 1061290–1302. [DOI] [PubMed] [Google Scholar]
  49. Scherr M, Battmer K, Blömer U, Schiedlmeier B, Ganser A, Grez M.et al. (2002Lentiviral gene transfer into peripheral blood-derived CD34+ NOD/SCID-repopulating cells Blood 99709–712. [DOI] [PubMed] [Google Scholar]
  50. Hahn JY, Cho HJ, Bae JW, Yuk HS, Kim KI, Park KW.et al. (2006Beta-catenin overexpression reduces myocardial infarct size through differential effects on cardiomyocytes and cardiac fibroblasts J Biol Chem 28130979–30989. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1.

Flow cytometry analysis of SDCs.

Click here for additional data file. (12MB, pdf)
Figure S2.

Oct4 expression of secondary CS.

Click here for additional data file. (12MB, pdf)
Figure S3 and Video S1 and S2.

Cardiomyogenic differentiation of secondary CS.

Click here for additional data file. (12MB, pdf)
Figure S4.

Cardiovascular differentiation of engrafted secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S5.

Microarray analysis to dissect the molecular mechanisms responsible for sphere formation.

Click here for additional data file. (12MB, pdf)
Figure S6.

E-selectin expression in secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S7.

E-selectin knock-down in SDCs and secondary CS formation.

Click here for additional data file. (12MB, pdf)
Figure S8.

With the blocker concentration we have used to inhibit the certain signal, the blockade of ERK, VEGF, and AKT did not affect the survival and proliferation of secondary CS cells and attachment of secondary CSs.

Click here for additional data file. (12MB, pdf)
Figure S9.

E-selectin overexpression in SDCs and secondary CS formation.

Click here for additional data file. (12MB, pdf)
Table S1.

Primer sets used RT-PCR.

Click here for additional data file. (58.5KB, doc)
Table S2.

Whole microarray dataset (25,697 genes) provided with Excel file.

Click here for additional data file. (11.9MB, xls)

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES