Original Paper Accepted after revision: March 31, 2014 Published online: August 21, 2014

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

A P19 and P19CL6 Cell-Based Complementary Approach to Determine Paracrine Effects in Cardiac Tissue Engineering Tilo Dehne a, b Xavier Adam a, b Eva-Maria Materne a, b Mareike Carola Reimann a, b Jan Philipp Krüger a, b Sophie Van Linthout a, c Carsten Tschöpe a, c Marion Haag a, b Michael Sittinger a, b Jochen Ringe a, b  

 

 

 

 

 

 

 

 

 

a Berlin-Brandenburg Center for Regenerative Therapies, b Tissue Engineering Laboratory, Department of Rheumatology and Clinical Immunology, and c Department of Cardiology and Pneumology, Charité-Universitätsmedizin Berlin, Berlin, Germany  

 

 

Key Words Cell-based test system · Heart regeneration · Paracrine stimulation · Trophic effect · Human cardiac-derived adherent proliferating cells

Abstract The negligible self-repair potential of the myocardium has led to cell-based tissue engineering approaches to restore heart function. There is more and more consensus that, in addition to cell development, paracrine effects in particular play a pivotal role in the repair of heart tissue. Here, we present two complementary murine P19 and P19CL6 embryonic carcinoma cell-based in vitro test approaches to study the potential of repair cells and the factors secreted by these cells to induce cardiomyogenesis. P19 cells were 3-dimensionally cultured in hanging drops and P19CL6 cells in a monolayer. Both systems, capable of inducible differentiation towards the cardiomyogenic lineage shown by the appearance of beating cells, the expression of connexin 43 and cardiac troponins T and I, were used to test the cardiomyogenesis-inducing potential of human cardiac-derived adherent proliferating (CardAP) cells, which are candidates for heart repair. CardAP cells in coculture as well as CardAP cellconditioned medium initiated beating in P19 cells, depending on the cell composition and concentration of the medi-

© 2014 S. Karger AG, Basel 1422–6405/14/1991–0024$39.50/0 E-Mail [email protected] www.karger.com/cto

um. CardAP cell-dependent beating was not observed in P19CL6 cultures, but connexin 43 and cardiac troponin formation as well as expression of GATA-binding protein 4 indicated the dose-dependent stimulatory cardiomyogenic effect of human CardAP cells. In summary, in different ways, P19 and P19CL6 cells have shown their capability to detect paracrine effects of human CardAP cells. In a complementary approach, they could be beneficial for determining the stimulatory cardiomyogenic potential of candidate cardiacrepair cells in vitro. © 2014 S. Karger AG, Basel

Abbreviations used in this paper

3D CardAP Cx43 DMSO ELFA GAPDH GATA4 GJA1 MSC MEF2C MYL2 NKX2-5 TNNT2

three-dimensional cardiac-derived adherent proliferating connexin 43 dimethyl sulfoxide enzyme-linked fluorescent assay glyceraldehyde-3-phosphate dehydrogenase GATA-binding protein 4 gap junction protein, alpha 1 mesenchymal stem cell myocyte enhancer factor 2C myosin, light-chain 2 NK2 homeobox 5 troponin T type 2

Tilo Dehne Tissue Engineering Laboratory and Berlin-Brandenburg Center for Regenerative Therapies, Department of Rheumatology and Clinical Immunology Charité-Universitätsmedizin Berlin, Charitéplatz 1, DE–10117 Berlin (Germany) E-Mail tilo.dehne @ charite.de

Introduction

Paracrine interactions are one key mechanism in cellbased therapies focusing on the regenerative treatment of the damaged myocardium [Moerkamp and Goumans, 2012], but only a few studies have been conducted addressing paracrine factors and mechanisms of action. After the cultivation of rat cardiac stem cells or cardiomyocytes with conditioned MSC culture supernatants, a positive paracrine effect on cardiomyogenesis was observed [Nakanishi et al., 2008; Angoulvant et al., 2011]. In a more general approach, Ranganath et al. [2012] reviewed the MSC secretome as a potential therapeutic tool for the treatment of ischemic heart disease. However, an easy to handle and functional in vitro test, that can be used to determine the potential of human cells to induce cardiomyogenesis via paracrine stimulation, has not yet been published. In our study, we evaluated the potential of the cell lines, P19 and P19CL6, to screen and evaluate the potential of cells, factors secreted by cells or other substances to induce cardiomyogenesis. First, we examined the differentiation efficiency of both cell lines under standard conditions to ensure reproducibility and select the screening criteria applying immunocytochemistry, enzyme-linked fluorescent assays (ELFAs) and qPCR. Secondly, we used the inducible properties of both cell lines to test the potential of human cardiac-derived adherent proliferating (CardAP) cells to induce cardiomyogenesis. CardAP cells are fibroblast-like cells that can be isolated from biopsies of the right-ventricle side of the interventricular septum [Haag et al., 2010]. Human CardAP cells reduce murine acute Coxsackievirus B3-induced myocarditis and increase heart function after intravenous injection [Miteva et al., 2011]. However, they act in a paracrine fashion and do not differentiate into cardiomyocytes in standard assays [Haag et al., 2010; Miteva et al., 2011]. Here, the P19 and P19CL6 cell lines were applied to test their potential to induce cardiomyogenesis. Direct cell-cell interactions were tested in cocultures with P19 or P19CL6 cells as well as indirect interactions of P19 and P19CL6 cultures with growth medium conditioned with CardAP cell culture supernatants. To consider the dose-dependent effects, different proportions of CardAP cells or concentrations of conditioned medium were tested.

According to the World Health Organization, cardiac diseases remain the first cause of death in the world [Mathers et al., 2008]. Diseases that go along with a loss of myocardium are often fatal due to the lack of sufficient self-healing capacity of this tissue. Tissue engineering focuses on the cell-based regenerative treatment of myocardium to restore heart function [Carvalho and de Carvalho, 2010]. Since adult cardiomyocytes have a minimal capacity to reenter the cell cycle or proliferate and/or divide in vivo or in vitro, a wide range of other cells from cardiac tissue (e.g. cardiac stem cells) and other tissues [e.g. bone marrow hematopoietic stem cells, mesenchymal stem cells (MSC), epithelial progenitor cells and skeletal myoblasts] have been applied in clinical trials to treat damaged myocardium [Choi et al., 2011]. Although not clearly related to clinical relevance, in some trials, the heart function was found significantly improved and therefore potentially beneficial for patients in incurable stages. However, significant differentiation into cardiomyocytes or paracrine induction of resident cardiac cells into the cardiomyogenic lineage could not be convincingly proven [Choi et al., 2011]. So far, cardiomyogenesis has been investigated using selected model cell lines including mouse and human embryonic stem cells or induced pluripotent progenitor cells [Moretti et al., 2010; Yoshida and Yamanaka, 2011; Jiang et al., 2013], mouse progenitor cells [Keiichi, 2002] and mouse embryonic carcinoma cells [Habara-Ohkubo, 1996]. P19 embryonic carcinoma cells, derived from a teratocarcinoma induced in CH3/HC mice, can be efficiently differentiated into cardiomyocytes, and are thus one of the most frequently used cell lines in ex vivo models [van der Heyden and Defize, 2003]. They are pluripotent, remain in an undifferentiated state when kept in the exponential growth phase and form beating aggregates after stimulation with dimethyl sulfoxide (DMSO) [Habara-Ohkubo, 1996; Jasmin et al., 2010]. To increase the differentiation rate, the P19 subline, P19CL6, was established. P19CL6 yields a higher differentiation efficiency of beating cells [Ohtsu et al., 2005; Mueller et al., 2010]. Until recently, most cell-based therapy strategies were based on the assumption that an implanted cell differentiates into a repair tissue cell and forms new tissue [Ringe et al., 2012]. Today, it is known that the potential of cells to secrete regenerative factors also plays a crucial role. Thus, one emerging tissue engineering strategy is to use cells as delivery tools for paracrine factors that promote tissue repair [Maltais et al., 2010; Mirotsou et al., 2011].

Cell Lines and Cell Cultivation P19 murine embryonic carcinoma cells were purchased from Cell Lines Services (Eppelheim, Germany). For culture expansion, they were seeded at a density of 1.5 × 104 cells/cm2 and expanded

Determining Paracrine Effects for Cardiac Cell Therapy

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

Material and Methods

25

Fig. 1. Overview on research conditions in coculture and with CardAP cell-conditioned media.

in alpha-medium (Biochrom, Berlin, Germany) supplemented with 10% fetal bovine serum (Thermo Scientific Hyclone, Bonn, Germany), 1% penicillin/streptomycin solution and 4 mM L-glutamine (all Biochrom). P19CL6 murine embryonic carcinoma cells were purchased from RIKEN BioResource Center (Tsukuba, Japan), seeded at a density of 1.5 × 104 cells/cm2 and grown in alpha-medium supplemented with 7.5% fetal bovine serum, 2.5% newborn calf serum (Biochrom), 1% penicillin/streptomycin solution and 4 mM L-glutamine. Cell culture was performed under standard conditions (in an incubator at 37 ° C and 5% CO2) and after culture expansion, the cells were cryopreserved until further use.  

 

Isolation and Cultivation of CardAP Cells Human CardAP cells were obtained from cardiac biopsies of 6 patients undergoing an endomyocardial biopsy procedure by means of the femoral vein approach under biplane fluoroscopic control to evaluate unexplained left-ventricular dysfunction. Biopsies were taken from the right-ventricle side of the interventricular septum and cells were isolated and culture expanded as described previously [Haag et al., 2010]. Briefly, biopsies were washed with PBS (Biochrom) and outgrowth culture was performed in Iscove’s medium (Biochrom) supplemented with 10% human allogeneic

26

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

serum (German Red Cross, Berlin) and 1% penicillin/streptomycin solution. Outgrowing cells were harvested by treatment with 0.05% trypsin/0.02% ethylenediaminetetraacetic acid (Biochrom) and subcultured until passage 3. The culture medium was made of equal parts of Iscove’s medium, DMEM and Ham’s F12 medium (all Biochrom), supplemented with 5% human allogeneic serum, 1% penicillin/streptomycin solution, 20 ng/ml human basic fibroblast growth factor and 10 ng/ml human epidermal growth factor (both PeproTech, Hamburg, Germany). When reaching 90% confluence, CardAP cells were trypsinized and replated at a density of 6 × 103 cells/cm2. After passage 3, CardAP cells were cryopreserved. The donation of cardiac tissue was approved by the ethical committee of the Charité-Universitätsmedizin Berlin (No. 225-07) and was performed after written consent by the patient. Chemical Induction of Cardiomyogenesis in Cell Lines The standard procedure to induce cardiomyogenesis of P19 cells is based on stimulation with DMSO [McBurney, 1993]. In brief, culture-expanded P19 cells were thawed and prepared as a suspension. Aliquots of 10 μl containing 1 × 102 P19 cells were placed on the surface of bacteriological-grade Petri dishes [Becton Dickinson (BD) Falcon, Heidelberg, Germany]. Inverting the dish resulted in hanging drops facilitating the formation of three-di-

Dehne et al.

mensional (3D) cell aggregates within 2 days. Aggregates were harvested, transferred to ultralow attachment plates (BD Falcon), and cardiomyogenesis was then induced with 1% DMSO (Sigma-Aldrich, Germany) for 3 days. For the outgrowth of cells to form beating bodies, aggregates were transferred to tissue culture grade dishes (BD Falcon). No more induction was required during the outgrowth phase. According to our own preliminary results and to already-published data to stimulate cardiomyogenesis of P19CL6 cells [Mueller et al., 2010], the cells were thawed, seeded on tissue culture grade 24-well plates (BD Falcon) at a density of 1 × 104 cells/cm2, grown to confluence and pretreated with 10 μM azacitidine (Sigma-Aldrich) to increase sensitivity. Subsequently, cardiomyogenesis/cell beating was induced by the addition of 1% DMSO to the P19CL6 cell culture medium for 3 weeks.

10, 100, 1,000%). Cocultures were maintained in nonconditioned P19 or P19CL6 medium. To generate conditioned medium, CardAP cells were thawed, suspended in CardAP cell culture medium, seeded at a density of 6 × 103 cells/cm2 in tissue culture flasks and allowed to attach for 24 h. Since P19 and P19CL6 cells do not grow/beat in CardAP culture medium, the CardAP medium was then replaced by P19 or P19CL6 cell expansion medium for conditioning. To generate two different loaded conditioned media, the first complete medium change was performed after 24 h (first sample). The medium was then replaced by new medium, the cultures were incubated for 72 h and the medium was collected (second sample). Conditioned media samples were portioned and stored at –80 ° C until use. Dilutions of conditioned media were made with P19 or P19CL6 culture media (% conditioned media: 33, 50, 100%).

Immunocytochemical Analysis of Cardiomyogenesis To analyze the formation of cardiac markers, connexin 43 (gap junction protein, alpha 1, 43 kDa) and cardiac troponin T [TNNT2; troponin T type 2 (cardiac)], the cardiomyogenic development of P19 and P19CL6 cells was performed as described above. On each day of sampling, the medium was removed, the cells were washed with PBS, fixed with 4% buffered formaldehyde (Herbeta Arzneimittel, Berlin, Germany), permeabilized with 0.2% Triton X100 (Sigma-Aldrich), and blocked for 1 h with 10% serum (according to the host of the secondary antibody, donkey or goat). Cells were labeled with connexin 43 (polyclonal rabbit; Santa Cruz Biotechnology, Heidelberg, Germany) and TNNT2 (polyclonal goat; Sigma-Aldrich) antibodies, which were visualized using Cy3-labeled anti-rabbit (polyclonal goat; Abcam, Berlin, Germany) or antigoat (polyclonal donkey; Dianova, Hamburg, Germany) secondary antibodies, depending on the host of the primary antibody. Nuclei were stained with bisbenzimide H33268 (Sigma-Aldrich). Murine heart tissue sections served as positive controls. For the negative control, sections were incubated using only the secondary antibody.

Enzyme-Linked Fluorescent Assay To indicate cardiomyogenic differentiation of P19CL6 cells, the VIDAS troponin I ultra assay for the diagnosis of myocardial infarction (bioMérieux Clinical Diagnostics, Nürtingen, Germany) was used. This diagnostic kit is an enzyme-linked immunosorbent assay-based automated clinical test to measure the cardiac troponin I level, and has a solid phase receptor in the pipetting system and all necessary reagents in the reagent cartridge. For detection, cell layers in 24-well plates were detached and homogenized by repeated pipetting in culture medium (1 ml) and stored at –80 ° C until measurement. Two hundred microliters of thawed specimens were transferred to cuvettes containing alkaline phosphatase-conjugated anti-troponin I, which cleaves the substrate 4-methylumbelliferyl phosphate, resulting in 4-methylumbelliferone, a fluorescent dye quantified at 450 nm.

Cocultures and Conditioned Media for the Induction of Cardiomyogenesis To evaluate the cardiomyogenic effect of CardAP cells on the P19 (n = 3 patients) or P19CL6 (n = 6 patients) cells, cocultures (direct cell-cell interactions) and cultures with conditioned medium (indirect cell-cell interactions) were set up (fig. 1). To generate cocultures consisting of CardAP and P19 cells, for each cell type, a separate cell suspension was prepared and both suspensions were then mixed in different proportions to evaluate a dose-response relationship. In general, hanging drops (10-μl volume, n = 40 per CardAP cell donor) contained a fixed total number of 1 × 102 cells, but a varying number of CardAP and P19 cells (% CardAP/total cells: 25, 50 and 90%). Cocultures of CardAP and P19CL6 cells were prepared as monolayer cultures (n = 5 wells of a 24-well plate per CardAP donor). Two alternative approaches were performed. For both settings, a cell suspension of CardAP cells was prepared and added to azacitidine-pretreated P19CL6 monolayer cultures. To evaluate a dose-response relationship, in the first setting, the total cell number was kept constant (1 × 104 cells/cm2) and the number of CardAP and P19CL6 cells was varied (% CardAP cells/total cells: 10, 50 and 90%). In the second setting, the P19CL6 cell number was kept constant (1 × 104 cells/cm2) and the number of CardAP cells was varied (% CardAP/P19CL6 cells:

Determining Paracrine Effects for Cardiac Cell Therapy

 

 

 

 

Real-Time RT-PCR Analysis of Cardiomyogenesis To show cardiomyogenesis of P19CL6 cells on the molecular level, RNA was isolated using the RNeasy Mini kit (Qiagen, Hilden, Germany) and reverse-transcribed with the iScript cDNA synthesis kit (BioRad, Munich, Germany). Quantitative PCR was performed in triplicate on optical plates on a Mastercycler ep realplex (Eppendorf, Hamburg, Germany) using expression assays for TaqMan probes and primer sets (Applied Biosystems, Darmstadt, Germany; the order No. is noted in parentheses): connex – in 43 (Mm00439105_m1), GATA-binding protein 4 (GATA4, Mm00484689_m1), myocyte enhancer factor 2C (MEF2C, Mm01340842_m1), slow cardiac myosin regulatory light-chain 2 (MYL2, Mm00440384_m1), NK2 homeobox 5 (NKX2-5, Mm00657783_m1) and TNNT2 (Mm00441922_m1). The expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Mm99999915_g1) was used to normalize samples. Relative quantification of marker gene expression was performed applying the efficiency – corrected ΔΔCt method and is given as fold change compared to control samples [Pfaffl, 2001]. Statistical Analysis The significance level was obtained from a pairwise group comparison of log2-transformed values applying the t test statistics of the SigmaStat 3.5 software (Statcon, Witzenhausen, Germany). If the normality or equal variance test was not passed, the Wilcoxon rank-sum test was conducted. We considered p < 0.05 as significant.

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

27

A

C

50 μm

250 μm

D

B

DMSO

200 μm

ctr

250 μm

Fig. 2. Functionality of P19 cell cultures. Phase-contrast microscopy of monolayer cultures during expansion (A), aggregates in suspension before attachment to tissue culture surface (B) and aggregates attached to culture surface (C, D). Aggregates were gener-

ated in hanging drops. Once transferred to tissue culture surface, cells started to grow out of the cell body when induced with DMSO (C). Within 2–3 days of induction, the first aggregates started to contract (see online suppl. video 1). Noninduced controls (ctr) attached to the cell culture surface, but cells did not spread the sur-

Results

Cardiomyogenic Differentiation Efficiency of P19 and P19CL6 Cells Pluripotent P19 embryonic carcinoma cells were expanded in monolayer culture and showed the typical heterogenic morphology of these cells (fig. 2A). For cardiomyogenesis, suspensions with 1 × 102 P19 cells were prepared as hanging drops. After 2 days, regular cell bodies with a diameter of about 150 μm were visible (fig. 2B). Subsequently, 3D cell aggregates were treated with DMSO, a known inductor of cardiomyogenesis [McBurney, 1993] and plated in tissue culture dishes. Within 2–3 additional days, cells grew out of the aggregates (fig. 2C) and the first areas with beating cardiomyocytes could be observed (see online suppl. video 1; for all online suppl. material, see www.karger.com/doi/10.1159/000362540). In contrast, even after 10 days, unstimulated control cultures did not show any outgrowth of cells or beating activity (fig.  2D). Furthermore, immunocytochemical staining of DMSO-treated cultures revealed the formation of connexin 43 (fig. 2E) and TNNT2 (fig. 2G). Connexin 43 represents the main cardiac connexin [Verheule 28

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

E

Cx43 F

Cx43

DMSO

50 μm

ctr

50 μm

G

cTnT H

cTnT

DMSO

50 μm

ctr

50 μm

rounding and no beating activity was observed (D). E–H Detection of cardiac markers by fluorescence microscopy. Nuclei of cells were stained with bisbenzimide (blue) and heart typical proteins connexin 43 (Cx43; green; E, F) and TNNT2 (cTnT; red; G, H) were analyzed. DMSO-induced aggregate cultures (E) showed intense stain of connexin 43 compared to control cultures (F). TNNT2 was detected in fiber-like structures in induced cultures (G), whereas control cultures (H) showed only a weak baseline expression.

et al., 1997], and TNNT2 is one of three regulatory proteins (the others are troponin C and troponin I) forming the troponin protein complex, which is integral to cardiac muscle contraction [Ohtsuki and Morimoto, 2008]. TNNT2 appeared in fiber-like structures, whereas unstimulated control cultures exhibited only a weak baseline expression (fig. 2H). Connexin 43 could not be detected in the control cultures (fig. 2F). The P19 cell approach is time-consuming due to having to generate hanging drops and handle formed aggregates until transfer to the tissue culture grade dishes, so P19CL6 cells capable of differentiating in monolayer culture were also tested [Mueller et al., 2010]. During expansion, the monolayers exhibited a polygonal cell morphology (fig. 3A). After DMSO stimulation, these cells formed condensed areas, which henceforth developed to aggregated structures (fig.  3B, black arrows). Beating mostly originated from aggregates but also from involved adjacent cell areas (see online suppl. video 2). Thus, a clear allocation of beating to respective aggregates was not possible. Unstimulated P19CL6 cultures formed no condensed structures and showed no beating. Moreover, immunocytochemical staining of control cultures revealed Dehne et al.

A

E 0.8

B

F 0.8 ctr

Troponin I (μg/l)

50 μm

500 μm

D

C

–1.1

2.6

7.5

0.6 Troponin I (μg/l)

500 μm

50 μm

0.6

DMSO

0.4

0.2

50 μm

50 μm

0

G

10 1

13.1±1.6

8.0±3.4

cTnT 1.2±1.5

1.1±1.2

*

40.2±2.8

6.3±1.2

*

*

0.1

0.4

0.2

ctr

Day 7 Day 14 Day 21

200 μm

ctr DMSO

0.01 0.001 0.0001

*

0

H

1,000

Gene expression fold change DMSO vs. ctr

200 μm

GAPDH expression (%)

Cx43

4.3±1.2

Day 7

Day 14

DMSO

Day 21

100

10

1 GATA4

MYL2

MEF2C

GJA1

TNNT2

NKX2-5

GATA4

MYL2

TNNT2 NKX2-5

Fig. 3. Functionality of P19CL6 cell cultures. Phase contrast microscopy of monolayer cultures in expansion medium (A) and cultures induced with DMSO for 21 days (B). Induced cultures developed aggregated structures (black arrow), which served as the starting point of contractions extending to adjacent areas (B, see online suppl. video 2). C, D Detection of cardiac markers by fluorescence microscopy. Nuclei of cells were stained with bisbenzimide (blue) and heart typical proteins connexin 43 (Cx43; green; C) or TNNT2 (cTnT; red; D) were analyzed. DMSO-induced cultures exhibited an intense stain for connexin 43 especially in condensed areas, whereas control cultures showed a weak detection of connexin 43 in single cells. E, F Cardiac troponin I (cTnI) detected by ELFA in lysed P19CL6 cells. The highest amounts of cardiac troponin I were detected in DMSO-induced cultures after 21 days.

In control cultures, the expression remained constant over time. F Reproducibly (n = 3), the amount of cardiac troponin I after 21 days was 4.5-fold higher compared to noninduced control (ctr) cultures. Gene expression of cardiac marker genes in noninduced and induced P19CL6 culture was analyzed by RT-PCR after 21 days and given as fold change (FC: DMSO vs. ctr; G, H). Analysis of GATA4, MYL2, TNNT2 and NKX2-5 expression demonstrated a distinctive increase (FC >6) after induction if stimulated with DMSO. G MEF2C and connexin 43 (GJA1) were not differentially expressed. H Except for MYL2, for all genes, the highest differences were observed on day 14 or 21. MYL2 expression varied over time with the lowest difference on day 14. Numbers above brackets = fold change vs. control. * p < 0.05.

only a very weak formation of connexin 43 and TNNT2, whereas in the condensed and contracting areas of DMSO-treated cultures, intense stainings of both cardiac marker proteins were observed by fluorescence microscopy (fig. 3C, D; blue: bisbenzimide-stained cell nucleus).

Since the quantification of the differentiation efficiency (counting of beating and nonbeating aggregates) of P19CL6 cells was not possible as it had been for P19 cells, we also analyzed the progression of cardiac troponin I formation using an ELFA and applied PCR analysis of cardiac marker gene expression. After 7 days, no increase

Determining Paracrine Effects for Cardiac Cell Therapy

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

29

of cardiac troponin I was detected in DMSO-induced P19CL6 cells compared to controls, but on day 14, a 2.6fold induction was observed that increased to 7.5-fold after 21 days (fig.  3E, one culture). For three cultures (fig. 3F), a significant difference between stimulated and nonstimulated P19CL6 cultures was found on day 21 with a fold change of 4.5 on average (fig. 3F). The expression of cardiac marker genes, such as the transcription factors GATA4 and NKX2-5 as well as TNNT2 (a component of the troponin complex regulating cardiac muscle contraction) was significantly increased after 21 days of standard stimulation compared to nonstimulated control cultures (fig. 3G). GATA4 and NKX2-5, which are known to interact with each other, are early markers of cardiac progenitor cells and are essential for heart formation, but they cannot initiate cardiomyogenesis [Durocher et al., 1997]. The expression of MYL2 was changed more than 8.0-fold, but this was not significant because of the high variation observed. However, MEF2C and connexin 43 were not found to have been differentially expressed. Furthermore, gene expression of GATA4, MYL2, TNNT2 and NKX2-5 was measured over time (fig. 3H). Except for MYL2, for all genes, the highest difference between DMSO-induced culture and control culture was observed on day 14 or day 21. Expression of MYL2 varied over time with the lowest difference (fold change) on day 14. To summarize, the results clearly show reproducibly that both P19 and P19CL6 cells have a high differentiation efficiency that can be quantified either by counting the proportion of beating aggregates (P19 cells) or by measuring marker expression by ELFA and/or PCR. Human CardAP Cells Induced Cardiomyogenesis in Cell Lines As reported previously, CardAP cells do not differentiate into cardiomyocytes in standard assays [Haag et al., 2010]. Here, for the first time, we analyzed whether these CardAP cells can induce cardiomyogenesis of other cell types. To do this, we used both P19 and P19CL6 cells. Direct cell-cell interactions were investigated in cocultures with P19 (hanging drops) or P19CL6 (monolayer) cells maintained in nonconditioned P19 or P19CL6 medium. For indirect interactions, both cell types were fed with P19 or P19CL6 growth medium conditioned by CardAP cells. First, the cocultures of CardAP and P19 cells were tested (in three approaches). In about 4% of all plated cocultures consisting of 90% CardAP and 10% P19 cells, beating areas were observed (fig. 4A). A mixture of 25 or 50% CardAP and 75 or 50% P19 cells did not result in beating 30

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

aggregates. Compared to the DMSO-treated cultures, in which 78% of the aggregates were beating, this 4% was low, but nevertheless indicated a cardiomyogenic effect (see online suppl. video 3). In this study, we did not determine parameters like beating frequency and did not compare our data with in vivo data. By antibody staining, connexin 43 was detected in cultures with 50 and 90% CardAP cells (fig.  4A, B) and TNNT2 in cultures with 50% CardAP cells (fig. 4A, D). Second, P19 cells were treated with conditioned media derived from CardAP cultures, which were incubated for 24 and 72 h in P19 medium (fig. 1). Medium conditioned for 24 h (33% conditioned and 66% fresh medium) stimulated up to 30% of the P19 aggregates to contract, and the use of 100% for 24-hour conditioned medium up to 3% (fig.  4A; online suppl. video 4). However, medium conditioned for 72 h did not stimulate any beating. In line, only the combination 24-hour/33% conditioned medium resulted in beating structures that were positive for both cardiac markers connexin 43 and TNNT2 (fig. 4A, C, E). Interestingly, whereas in the positive (DMSO-treated; fig. 2) and negative control cultures (untreated; fig. 2) beating was correlated with the detection of these two marker proteins, in the CardAP cell setups, this was only observed for the 24-hour/33% conditioned medium combination (fig. 4). For instance, the combination of 72 h and 33% did not result in beating cells but in connexin 43- and TNNT2-positive cells. Third, coculture of CardAP and P19CL6 cells (n = 6) was performed with two alternative approaches in which CardAP cells were added to azacitidine-treated P19CL6 monolayers. In the first setting, the total cell number was kept constant and the number of CardAP and P19CL6 cells was varied (% CardAP/total cells: 10, 50 and 90%). In the second setting, the P19CL6 cell number was kept constant and the number of CardAP cells was varied (10, 100, 1,000% related to P19CL6 cell number). However, in contrast to the P19 approaches, no beating cells could be observed. Nevertheless, and independent of the setting and the proportion of both cell types, ELFA-based measurement of cardiac troponin I revealed an increase of this marker protein of between 1.2- and 1.4-fold (fig. 5; compared to unstimulated control cultures). Moreover, connexin 43 immunocytochemistry demonstrated a culture-dependent varying level of this protein. Intense staining, comparable with the intensity in DMSO-treated cultures (fig. 5C), was observed in the cocultures with a high proportion of CardAP cells (90, 1,000%; fig. 5D, E). In all cocultures with a low proportion of CardAP cells, a weak connexin 43 staining was detected (fig. 5A), which Dehne et al.

A

Conditioned medium

Coculture 24 h

Controls 72 h

25%

50%

90%

33%

50%

100%

33%

50%

100%

Neg.

DMSO

0±0

0±0

4±2

30±10

0±0

3±2

0±0

0±0

0±0

0±0

78±23

Beating activity (% total)

Immunofluorescence cTnT

–

+

–

+

–

+

+

–

–

–

+

Cx43

–

+

+

++

++

–

+

–

+

–

++

B

Cx43

cc 50%

50 μm

C

Cx43

cm 33%

50 μm

D

cc 50%

E

cTnT

50 μm

cTnT

cm 33%

50 μm

Fig. 4. A Test of CardAP cells stimulation potential with P19 cells. Different proportions of CardAP cells and P19 cells (25–90%) were tested as well as different dilutions (33, 50% in P19 media, 100% = pure) of media conditioned for 24 and 72 h. The maximum number of beating aggregates was determined after 10 days of induction and was related to the total number of present aggregates (≥40) given as a percentage. The same cultures were analyzed for

the expression of TNNT2 (cTnT) and connexin 43 (Cx43) detected with varying degrees: intense (++), weak (+) or no (–) stain. Consistent with the beating activity, intense stainings of both markers appeared in cultures induced with 24-hour/33% conditioned media (cm 33%; C, D). Connexin 43 (B) and TNNT2 (D) were also detected in the cocultures (cc), although no beating activity was observed.

was comparable to the unstimulated control cultures (fig.  5B). In the DMSO-stimulated and unstimulated P19CL6 control cultures, GATA4 was selected as a good marker in our setting. Consistent with the connexin 43 and cardiac troponin I results, GATA4 gene expression analysis revealed the highest changes in the cocultures with the highest proportion of CardAP cells. The increase was found to have changed significantly by up to 5.8-fold (the approach with a fixed number of total cells) and 8.1fold (the approach with a fixed number of P19CL6 cells) compared to the nonstimulated cultures (fig. 5F). In the cultures with the same number of CardAP cells and P19CL6 cells, a significant 4.2-fold higher expression of GATA4 was also observed (fig. 5F). Fourth, P19CL6 cells were treated with conditioned media derived from CardAP cultures which had been incubated for 24 and 72 h in P19CL6 medium (fig. 1). As in the cocultures, no beating areas were observed. However, cardiac troponin I production was increased to between 1.1- and 7.4-fold. This was consistent with the observation from the DMSO-stimulated control cultures. Interestingly, the highest increase (7.4-fold) was found in the P19CL6 cultures stimulated with the combination 24-

hour/33% conditioned medium (33% conditioned and 66% fresh medium; fig.  5A). This combination also resulted in the highest amount of beating aggregates in P19 cultures. Furthermore, in all P19CL6 cultures stimulated with conditioned medium, a weak but positive connexin 43 staining was observed (fig.  5A). In contrast to these immunostaining results, GATA4 gene expression was more induced by 24- or 72-hour/100% conditioned medium than by the 24- or 72-hour/33% combination (fig.  5F). The fold changes ranged from 4.0 to 4.2 in P19CL6 cultures fed with the 24-hour conditioned medium and from 5.2 to 8.1 in the 72-hour conditioned medium. Except for the 24-hour/33% approach, all media specimens revealed a significant change in GATA4 expression (fig. 5F). To exclude that GATA4 expression is biased by the presence of cells in cocultures or consumed media (autocrine effects), coculture of azacitidine-pretreated P19CL6 and nonpretreated P19CL6 cells as well as stimulation of azacitidine-pretreated P19CL6 cells with P19CL6-conditioned media was performed. Neither the addition of P19CL6 cells to the cocultures nor the stimulation with P19CL6-conditioned media led to the induction of

Determining Paracrine Effects for Cardiac Cell Therapy

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

31

Conditioned media

Coculture

A

Fixed total cell number Fixed P19CL6 cell number 10%

Controls

24 h

72 h

90%

10%

100% 1,000%

33%

50%

100%

33%

50%

100%

ctr

1.2 ±0

1.4 ±0.1

1.3 ±0.2

1.3 ±0.1

7.4 ±2.0

1.1 ±0

1.4 ±0.2

6.0 ±1.4

1.4 ±0.1

1.1 ±0.1

1.0 ±0.1

3.3 ±0.7

+

++

+

+

+

+

+

+

+

+

++

50%

DMSO

Troponin I protein expression Fold change vs. ctr

1.2 ±0.1

1.4 ±0.1

Immunofluorescence Cx43

+

ctr

B

+

++ DMSO

C

D

25 μm

Cx43

100 μm

Cx43

Cx43

100 μm

2.3 1.0

*

*

10.0

10%

3.8

8.1

*

5.8

100 μm

4.2

*

*

4.0

3.8

4.2

33%

50%

100%

*

5.3

*

5.9

8.1

33%

50%

100%

*

*

5.6

1.8 50%

90%

fixed total cell number

10%

100%

1,000%

fixed P19CL6 cell number

24 h

Coculture

Fig. 5. Test of CardAP cell stimulation potential with P19CL6 cells. A Different proportions of CardAP cells and P19CL6 cells were

tested as well as different mixtures (33, 50, 100% = pure) of media conditioned for 24 and 72 h. Cocultures were prepared with either a fixed total cell number (% CardAP cell/ total cells; 10, 50; 90%) or a fixed total P19CL6 number (10–1,000%, CardAP cells/P19CL6 cells). The cardiac marker troponin I was detected by ELFA and connexin 43 (Cx43) by immunocytochemistry. Troponin content was related to nonstimulated cultures (ctr) given as fold change. Connexin 43 (Cx43) was detected with varying degrees: intense (++), weak (+) or no (–) stain. B–E Fluorescence microscopy images of connexin 43 antibody staining. Control cultures revealed a

GATA4 expression (see online suppl. figure 1). To conclude, even though CardAP cells did not induce beating in P19CL6 cultures, gene and protein data were in line with the values measured in the DMSO-stimulated beating cultures. This indicates cardiomyogenic stimulation caused by human CardAP cells. 32

25 μm

100.0 GATA4 expression ratio vs. control

F

100 μm

1,000%

25 μm

25 μm

Cx43

E

90%

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

DMSO

72 h Conditioned medium

weak intensity and random distribution of the staining (B), whereas DMSO (C), and high CardAP cell proportions (90%, 1,000%; D; E) facilitated an intense expression of connexin 43, that was found predominantly in condensed cell areas. GATA4 gene expression in P19CL6 was analyzed by real-time RT-PCR after 21 days in coculture with different mixtures of CardAP cells and in culture with different mixtures of conditioned media from CardAP cells (F, n = 6). The addition of CardAP cells or conditioned medium increased the GATA4 expression. DMSO served as a positive control. Expression is given as fold change related to the expression of noninduced P19CL6 cultures (controls). Numbers above black bars = average fold change. * t test vs. control cultures p < 0.05.

Discussion

There is more and more consensus that, in addition to cell engraftment and the development of cells into repair tissue, paracrine effects particularly play a pivotal role in cell-based heart tissue repair and regeneration [Maltais et Dehne et al.

al., 2010; Mirotsou et al., 2011]. In this study, we described two approaches based on the P19 and P19CL6 cell lines that allow the detection of the cardiomyogenic stimulatory potential of cells and the factors secreted by the cells. We examined the differentiation of both cell lines into the cardiomyogenic lineage after chemical stimulation with DMSO, and thus ensured their functionality and identity. This was demonstrated on the gene (GATA4 expression) and protein (troponin I, TNNT2 and connexin 43 formation) levels, and by the microscopic observation of contracting cells. The inducible cardiomyogenic properties of both cell lines were used to analyze the paracrine potential of heart-derived CardAP cells in a coculture setup (direct cell-cell interactions) and a setup with conditioned medium (indirect cell-cell interactions). Here, P19 cells were successfully initiated to beat by CardAP cells in hanging-drop cocultures with 90% CardAP and 10% P19 cells as well as in hanging-drop cultures by indirect stimulation with medium conditioned for 24 h by CardAP cells (33% conditioned and 66% fresh medium). In the P19CL6 cell monolayer cultures, which were easier to handle, no beating cells could be observed. However, formation of cardiac troponin I and connexin 43 and a dosedependent effect of CardAP cells and their conditioned media on GATA4 expression revealed cardiomyogenesis. In summary, complementing one another, both P19 and P19CL6 cells are suitable for detecting paracrine effects and can be a useful component of in vitro test assays as initially applied in this study. In the context of heart tissue repair and regeneration, it is of interest that, in these approaches at least, human CardAP cells could induce cardiomyogenesis in a paracrine manner. So far, in vivo disease models like myocardial infarction or myocarditis were applied to identify paracrine effects [Miteva et al., 2011; Li et al., 2012b], but the elucidation of underlying mechanisms remains difficult because of the complex interplay between host, injury and transplanted cells, and the control of cell presence that is lacking at defect sites. Thus, knowledge of a stimulatory effect of a particular factor is often based on in vitro approaches. For example, the secretome of stimulated cells (e.g. with tumor necrosis factor alpha, interferon beta and lipopolysaccharides) was screened for paracrine factors (e.g. interleukins 6 and 8, monocyte chemoattractant protein 1 and vascular endothelial growth factor alpha) and its stimulatory effect was investigated applying bioactivity assays (monocyte migration, impact on migration of human MSC and enhanced survival of transplanted cells) [Li et al., 2012b; Ranganath et al., 2012]. These assays study the effect of secreted factors on cells, but do not spe-

cifically focus on their cardiomyogenic potential. More comprehensive in vitro approaches including all secreted factors make use of cardiomyogenic indicator cells, like cardiac progenitor cells or cardiomyocytes, to detect the stimulatory effects of conditioned media [Nakanishi et al., 2008; Angoulvant et al., 2011]. The limited availability of such stem cells and progenitor cells and their spontaneous beating activity has hampered an application for extensive screening and investigation so far. The use of induced pluripotent progenitor cells and their cardiomyogenic differentiated descendants might overcome the limited cell availability and opens up the perspective of personalized testing. So far, however, the generation and maintenance of induced pluripotent progenitor cells is costly and often results in mixed populations of terminal differentiated cells. These cells show spontaneous differentiation, and there is great variability of cells derived from patients of differing ages and according to sex [Brenner and Franz, 2011; Eschenhagen and Blankenberg, 2013; Mordwinkin et al., 2013; Xu and Sun, 2013]. For more than 20 years, cell lines such as P19 and P19CL6 are applied for the study of cardiac differentiation [van der Heyden and Defize, 2003]. The culturing costs are low and only a screening of sera supporting efficient cardiomyogenesis is necessary before starting [Wilton and Skerjanc, 1999]. P19 and P19CL6 cells do not show spontaneous differentiation and can be reliably induced towards a cardiomyocyte phenotype, making them competent for the detection of paracrine factors. A drawback of cell lines such as P19 and P19CL6 is that outcomes are not readily translatable to clinical practice unless correlated with clinical data. Alternatively, the use of adult stem cells such as MSC instead can enhance the clinical relevance, but standardization with donor-derived cells remains difficult. P19 cells and their derivate P19CL6 cells have already been used to identify the cardiomyogenic potential of single substances and to elucidate signaling pathways. Chen and Reese [2011] made use of P19 cells to study the retinol-signaling pathway and Danalache et al. [2010] found a novel cardiomyogenic peptide with its help. Sadek et al. [2008] used P19CL6 cells to identify small molecules involved in cardiac fate by screening a chemical library for activators of the signature gene NKX2-5. Jasmin et al. [2010] investigated the effect of cardiogenol C, and Fathi et al. [2009] described the cardiac differentiation capacity of oxytocin using P19CL6 cells. Li et al. [2012a] and Kami et al. [2008] studied the function of novel proteins that regulate cardiomyocyte differentiation of P19CL6 cells induced by DMSO. In this study, for the first time, we have analyzed the cardiomyo-

Determining Paracrine Effects for Cardiac Cell Therapy

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

33

genic potential of a cardiac-derived cell population using P19 and P19CL6 cells. CardAP cells, as used in this study, are a fibroblast-like cell type, isolated from biopsies from the right-ventricle side of the interventricular septum of patients, in order to evaluate unexplained left-ventricular dysfunction [Haag et al., 2010]. Although this general diagnosis includes a lot of unknown parameters such as inflammation and oxidative stress stimuli, human CardAP cells have already reproducibly demonstrated a reduction of murine acute Coxsackievirus B3-induced myocarditis and increased heart function after intravenous injection [Miteva et al., 2011]. Furthermore, Haag et al. [2013] showed that these cells were capable of inducing endothelial structures in a tube formation assay and secreting high amounts of the vascular endothelial growth factor and interleukin 8, which can mediate a proangiogenic effect of CardAP cell in a paracrine manner. These properties made CardAP cells an ideal candidate for testing whether cardiomyogenic stimulation potential can be detected using P19 and P19CL6 cells. In this study, we initially characterized the development of P19 and P19CL6 cells after stimulation with DMSO. As shown by Paquin et al. [2002], DMSO upregulates the expression of the oxytocin receptor and the early cardiac differentiation markers like GATA4, both of which are essential for the propagation of cardiac differentiation [Uchida et al., 2007]. For both cell types, we used the beating activity and the formation of connexin 43, TNNT2 and cardiac troponin I to document cardiac differentiation. Coordinated contraction is an ultimate feature of cardiomyocytes, requiring the propagation of an electrical excitation by connexins, for example, and the generation of a contractile force through proteins integral to the cardiac muscle contraction involving the troponin complex [Ohtsuki and Morimoto, 2008; Zhang and Shaw, 2012]. Nonstimulated P19 or P19CL6 cell lines exhibited a weak baseline expression of TNNT2 or troponin I, and P19CL6 also exhibited a weak expression of connexin 43, both indicating a differentiation, but the expression of these markers was considerable lower and did not result in any spontaneous beating when compared to induced samples. The analysis of GATA4 expression facilitated the discrimination of induced and noninduced cultures in P19CL6 cells. Since the fate of a stem cell like a P19 or P19CL6 cell is not only determined by its niche and secreted products, but also by its local microenvironment including the surrounding cells [Baraniak and McDevitt, 2010], we used an experimental setup, which allows analyzing paracrine 34

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

stimulation via direct cell-cell interactions in cocultures and indirect interactions by secreted factors in conditioned medium. We found that the cardiomyogenic effect of a conditioned medium of CardAP cells was more pronounced than the effect of direct cell-cell contact. Compared to cultures stimulated with conditioned medium, cocultures of CardAP and P19 cells displayed a lower beating activity and a lower expression of the cardiac markers connexin 43 and TNNT2. Furthermore, we detected a lower troponin I level in cocultures of CardAP and P19CL6 cells compared to cultures treated with CardAP conditioned media. It was not the aim to study the effect of CardAP cells on electrophysiological parameters in beating P19 or P19CL6 embryonic carcinoma cell-derived cardiomyocytes in detail. However, we do not expect beating parameters to reflect physiological characteristics. It was also not the objective of this feasibility-oriented study to identify particular factors secreted by CardAP cells during paracrine stimulation. Nevertheless, both cell lines allow such investigations in a complementary manner. Gene expression analysis of several cardiac markers in DMSO-stimulated P19CL6 cells revealed a significant induction of GATA4, TNNT2 and NKX2-5 after 21 days of culture. The results revealed GATA4 to be a robust and sensitive marker, and its induction is correlated with the induction of other cardiac markers, such as TNNT2 and NKX2-5. In accordance with already-described data on differentiating cardiomyocytes [Hu et al., 2010], GATA4 expression was chosen for high-throughput screening using P19CL6 cells. Screening of different compositions of cocultures revealed a dose-dependent effect of CardAP cells on GATA4 expression, which differed significantly from nonstimulated cell line cultures and was maximal at the highest CardAP cell numbers. These results were consistent with the occurrence of beating in P19 cells and the detection of increased connexin 43 and TNNT2 and cardiac troponin I in P19 and P19CL6 cells, respectively. At low concentrations, conditioned medium of CardAP cells also stimulated beating in P19 cells and the expression of connexin 43 and troponin I in both cell types. GATA4 gene expression in P19CL6 cells was significantly induced and increased further with the increasing concentration of conditioned medium. To further validate GATA4 as an adequate marker for screenings, a control experiment with cocultures of expanded P19CL6 cells as well as P19CL6 cells as prepared in the described method was performed to exclude autocrine effects. GATA4 expression was found to not have been affected in coculture. Likewise, Dehne et al.

conditioned medium from P19CL6 cells did not reveal any inductive effect on GATA4 expression. Furthermore, we checked the specificity of the GATA4 PCR assay: murine RNA of GATA4 was not detected in pure CardAP cell cultures or in human heart tissue (data not shown). These results suggest that the documented cardiomyogenesis-inducing paracrine effects were not simply related to the addition of any cells or conditioned media (autocrine effects), but rather the consequence of the addition of CardAP cells or their secreted factors. Although the expression of GATA4 as the sole marker is not sufficient to reliably detect cardiomyogenesis, this process was suitable for demonstrating the dose-dependent effects of cells and conditioned media. Comparing the results obtained in the P19 and P19CL6 cell culture methods, we observed a different effect of CardAP cells on the two cell lines. Although at a low degree, the P19 cells were induced to contractions, but no beating was observed in the P19CL6 cells. However, increased formation of the cardiac marker proteins connexin 43 and troponin I and gene expression of GATA4 compared to control cultures were shown. Furthermore, formation of cardiomyocytes is not necessarily connected with contraction [Desplantez et al., 2007]. Based on the results, P19 cells seem to be more efficiently stimulated into beating cells by CardAP cells. On the other hand, the hanging-drop culture approach employed for the P19 cells is more complex and has more limitations in the experimental design, because the cells have to be prepared in a 3D arrangement. The P19CL6 cell approach is more suitable for high-throughput analysis, because it is performed in monolayer culture in multiwell plate format and the design of the experiment is more flexible with

regard to the choice of possible coculture combinations (fixed total cell number and fixed P19CL6 cell number). As shown here, by combining the positive characteristics of P19 (sensitivity and beating) and P19CL6 cells (high throughput), the effective dosage as well as a deeper analysis of the paracrine stimulation potential can be studied in a complementary approach. In conclusion, we have shown the potential of two cellbased approaches for studying the paracrine stimulation of cardiomyogenesis that can be applied complementarily, because P19 cells offer sensitivity and evidence of beating while P19CL6 allow high throughput due to easier handling. Cell beating, the formation of connexin 43, TNNT2 and cardiac troponin I and the expression of GATA4 were revealed as useful read-out markers. Applying these cell lines and markers, we were able to detect, for the first time, the paracrine cardiomyogenic effect of human CardAP cells (a cell type which is discussed in the context of treating damaged myocardium [Haag et al., 2010]). The observed effects were dependent on the applied cell number and the concentration of conditioned medium, which should be considered if other cell types are tested for their cardiomyogenic paracrine stimulation potential by employing P19 or P19CL6 cells.

Acknowledgements The authors gratefully thank Barbara Walewska and Günter Krüger for excellent technical support and Hannah Louise John for carefully reading the manuscript and for useful comments. The study was supported by the Berlin-Brandenburg Center for Regenerative Therapies (Bundesministerium für Bildung und Forschung, grant No. 1315848A).

References Angoulvant, D., F. Ivanes, R. Ferrera, P.G. Matthews, S. Nataf, M. Ovize (2011) Mesenchymal stem cell conditioned media attenuates in vitro and ex vivo myocardial reperfusion injury. J Heart Lung Transplant 30: 95–102. Baraniak, P.R., T.C. McDevitt (2010) Stem cell paracrine actions and tissue regeneration. Regener Med 5: 121–143. Brenner, C., W.M. Franz (2011) The use of stem cells for the repair of cardiac tissue in ischemic heart disease. Expert Rev Med Devices 8: 209–225. Carvalho, A.B., A.C. de Carvalho (2010) Heart regeneration: past, present and future. World J Cardiol 2: 107–111.

Determining Paracrine Effects for Cardiac Cell Therapy

Chen, Y., D.H. Reese (2011) The retinol signaling pathway in mouse pluripotent P19 cells. J Cell Biochem 112: 2865–2872. Choi, Y.H., T. Saric, B. Nasseri, S. Huhn, S. Van Linthout, R. Hetzer, C. Tschope, C. Stamm (2011) Cardiac cell therapies: the next generation. Cardiovasc Ther 29: 2–16. Danalache, B.A., J. Gutkowska, M.J. Slusarz, I. Berezowska, M. Jankowski (2010) OxytocinGly-Lys-Arg: a novel cardiomyogenic peptide. PloS One 5: e13643. Desplantez, T., E. Dupont, N.J. Severs, R. Weingart (2007) Gap junction channels and cardiac impulse propagation. J Membr Biol 218(1–3): 13–28.

Durocher, D., F. Charron, R. Warren, R.J. Schwartz, M. Nemer (1997) The cardiac transcription factors NKX2-5 and GATA-4 are mutual cofactors. EMBO J 16: 5687–5696. Eschenhagen, T., S. Blankenberg (2013) Personalized therapy in cardiology: biomarkers, pharmacogenetics and therapy of monogenic diseases. Internist 54: 147–152. Fathi, F., S. Murasawa, S. Hasegawa, T. Asahara, A.J. Kermani, S.J. Mowla (2009) Cardiac differentiation of P19CL6 cells by oxytocin. Int J Cardiol 134: 75–81. Haag, M., J. Ritterhof, A. Dimura, K. Miteva, S. Van Linthout, C. Tschope, J. Ringe, M. Sittinger (2013) Pro-angiogenic effect of endomyocardial biopsy-derived cells for cardiac differentiation. Curr Tissue Eng 2: 154–159.

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

35

Haag, M., S. Van Linthout, S.E. Schroder, U. Freymann, J. Ringe, C. Tschope, M. Sittinger (2010) Endomyocardial biopsy derived adherent proliferating cells – a potential cell source for cardiac tissue engineering. J Cell Biochem 109: 564–575. Habara-Ohkubo, A. (1996) Differentiation of beating cardiac muscle cells from a derivative of P19 embryonal carcinoma cells. Cell Struct Funct 21: 101–110. Hu, D.L., F.K. Chen, Y.Q. Liu, Y.H. Sheng, R. Yang, X.Q. Kong, K.J. Cao, H.T. Gu, L.M. Qian (2010) GATA-4 promotes the differentiation of P19 cells into cardiac myocytes. Int J Mol Med 26: 365–372. Jasmin, D.C. Spray, A.C. Campos de Carvalho, R. Mendez-Otero (2010) Chemical induction of cardiac differentiation in P19 embryonal carcinoma stem cells. Stem Cells Dev 19: 403– 412. Jiang, B., H. Dong, Q. Li, Y. Yu, Z. Zhang, Y. Zhang, G. Wang, Z. Zhang (2013) Differentiation of reprogrammed mouse cardiac fibroblasts into functional cardiomyocytes. Cell Biochem Biophys 66: 309–318. Kami, D., I. Shiojima, H. Makino, K. Matsumoto, Y. Takahashi, R. Ishii, A.T. Naito, M. Toyoda, H. Saito, M. Watanabe, I. Komuro, A. Umezawa (2008) Gremlin enhances the determined path to cardiomyogenesis. PloS One 3: e2407. Keiichi, F. (2002) Reprogramming of bone marrow mesenchymal stem cells into cardiomyocytes. C R Biol 325: 1027–1038. Li, L., H. Gong, H. Yu, X. Liu, Q. Liu, G. Yan, Y. Zhang, H. Lu, Y. Zou, P. Yang (2012a) Knockdown of nucleosome assembly protein 1-like 1 promotes dimethyl sulfoxide-induced differentiation of P19CL6 cells into cardiomyocytes. J Cell Biochem 113: 3788–3796. Li, T.S., K. Cheng, K. Malliaras, R.R. Smith, Y. Zhang, B. Sun, N. Matsushita, A. Blusztajn, J. Terrovitis, H. Kusuoka, L. Marban, E. Marban (2012b) Direct comparison of different stem cell types and subpopulations reveals superior paracrine potency and myocardial repair efficacy with cardiosphere-derived cells. J Am Coll Cardiol 59: 942–953. Maltais, S., J.P. Tremblay, L.P. Perrault, H.Q. Ly (2010) The paracrine effect: pivotal mechanism in cell-based cardiac repair. J Cardiovasc Transl Res 3: 652–662.

36

Mathers, C., J.T. Boerma, D. Ma Fat (2008) The global burden of disease: 2004 update. Geneva, World Health Organization. McBurney, M.W. (1993) P19 embryonal carcinoma cells. Int J Develop Biol 37: 135–140. Mirotsou, M., T.M. Jayawardena, J. Schmeckpeper, M. Gnecchi, V.J. Dzau (2011) Paracrine mechanisms of stem cell reparative and regenerative actions in the heart. J Mol Cell Cardiol 50: 280–289. Miteva, K., M. Haag, J. Peng, K. Savvatis, P.M. Becher, M. Seifert, K. Warstat, D. Westermann, J. Ringe, M. Sittinger, H.P. Schultheiss, C. Tschope, S. Van Linthout (2011) Human cardiac-derived adherent proliferating cells reduce murine acute Coxsackievirus B3-induced myocarditis. PloS One 6: e28513. Moerkamp, A.T., M.J. Goumans (2012) Cardiac regeneration: stem cells and beyond. Curr Med Chem 19: 5993–6002. Mordwinkin, N.M., P.W. Burridge, J.C. Wu (2013) A review of human pluripotent stem cell-derived cardiomyocytes for highthroughput drug discovery, cardiotoxicity screening, and publication standards. J Cardiovasc Transl Res 6: 22–30. Moretti, A., M. Bellin, C.B. Jung, T.M. Thies, Y. Takashima, A. Bernshausen, M. Schiemann, S. Fischer, S. Moosmang, A.G. Smith, J.T. Lam, K.L. Laugwitz (2010) Mouse and human induced pluripotent stem cells as a source for multipotent Isl1+ cardiovascular progenitors. FASEB J 24: 700–711. Mueller, I., R. Kobayashi, T. Nakajima, M. Ishii, K. Ogawa (2010) Effective and steady differentiation of a clonal derivative of P19CL6 embryonal carcinoma cell line into beating cardiomyocytes. J Biomed Biotechnol 2010: 380561. Nakanishi, C., M. Yamagishi, K. Yamahara, I. Hagino, H. Mori, Y. Sawa, T. Yagihara, S. Kitamura, N. Nagaya (2008) Activation of cardiac progenitor cells through paracrine effects of mesenchymal stem cells. Biochem Biophys Res Commun 374: 11–16. Ohtsu, Y., K. Johkura, K. Ito, T. Akashima, K. Asanuma, N. Ogiwara, T. Oka, I. Komuro, K. Sasaki, J. Amano (2005) Stimulation of P19CL6 with multiple reagents induces pulsating particles in vivo. Curr Med Res Opin 21: 795–803. Ohtsuki, I., S. Morimoto (2008) Troponin: regulatory function and disorders. Biochem Biophys Res Comm 369: 62–73.

Cells Tissues Organs 2014;199:24–36 DOI: 10.1159/000362540

Paquin, J., B.A. Danalache, M. Jankowski, S.M. McCann, J. Gutkowska (2002) Oxytocin induces differentiation of P19 embryonic stem cells to cardiomyocytes. Proc Natl Acad Sci USA 99: 9550–9555. Pfaffl, M.W. (2001) A new mathematical model for relative quantification in real-time RTPCR. Nucleic Acids Res 29: e45. Ranganath, S.H., O. Levy, M.S. Inamdar, J.M. Karp (2012) Harnessing the mesenchymal stem cell secretome for the treatment of cardiovascular disease. Cell Stem Cell 10: 244– 258. Ringe, J., G.R. Burmester, M. Sittinger (2012) Regenerative medicine in rheumatic diseaseprogress in tissue engineering. Nat Rev Rheumatol 8: 493–498. Sadek, H., B. Hannack, E. Choe, J. Wang, S. Latif, M.G. Garry, D.J. Garry, J. Longgood, D.E. Frantz, E.N. Olson, J. Hsieh, J.W. Schneider (2008) Cardiogenic small molecules that enhance myocardial repair by stem cells. Proc Natl Acad Sci USA 105: 6063–6068. Uchida, S., S. Fuke, T. Tsukahara (2007) Upregulations of GATA4 and oxytocin receptor are important in cardiomyocyte differentiation processes of P19CL6 cells. J Cell Biochem 100: 629–641. van der Heyden, M.A., L.H. Defize (2003) Twenty-one years of P19 cells: what an embryonal carcinoma cell line taught us about cardiomyocyte differentiation. Cardiovasc Res 58: 292–302. Verheule, S., M.J. van Kempen, P.H. te Welscher, B.R. Kwak, H.J. Jongsma (1997) Characterization of gap junction channels in adult rabbit atrial and ventricular myocardium. Circ Res 80: 673–681. Wilton, S., I. Skerjanc (1999) Factors in serum regulate muscle development in P19 cells. In vitro Cell Develop Biol Anim 35: 175–177. Xu, X.Q., W. Sun (2013) Perspective from the heart: the potential of human pluripotent stem cell-derived cardiomyocytes. J Cell Biochem 114: 39–46. Yoshida, Y., S. Yamanaka (2011) iPS cells: a source of cardiac regeneration. J Mol Cell Cardiol 50: 327–332. Zhang, S.S., R.M. Shaw (2012) Multilayered regulation of cardiac ion channels. Biochim Biophys Acta.

Dehne et al.

Copyright: S. Karger AG, Basel 2014. Reproduced with the permission of S. Karger AG, Basel. Further reproduction or distribution (electronic or otherwise) is prohibited without permission from the copyright holder.