Terminal Differentiation, Advanced Organotypic Maturation, and Modeling of Hypertrophic Growth in Engineered Heart TissueNovelty and Significance
Rationale: Cardiac tissue engineering should provide “realistic” in vitro heart muscle models and surrogate tissue for myocardial repair. For either application, engineered myocardium should display features of native myocardium, including terminal differentiation, organotypic maturation, and hypertrophic growth.
Objective: To test the hypothesis that 3D-engineered heart tissue (EHT) culture supports (1) terminal differentiation as well as (2) organotypic assembly and maturation of immature cardiomyocytes, and (3) constitutes a methodological platform to investigate mechanisms underlying hypertrophic growth.
Methods and Results: We generated EHTs from neonatal rat cardiomyocytes and compared morphological and molecular properties of EHT and native myocardium from fetal, neonatal, and adult rats. We made the following key observations: cardiomyocytes in EHT (1) gained a high level of binucleation in the absence of notable cytokinesis, (2) regained a rod-shape and anisotropic sarcomere organization, (3) demonstrated a fetal-to-adult gene expression pattern, and (4) responded to distinct hypertrophic stimuli with concentric or eccentric hypertrophy and reexpression of fetal genes. The process of terminal differentiation and maturation (culture days 7–12) was preceded by a tissue consolidation phase (culture days 0–7) with substantial cardiomyocyte apoptosis and dynamic extracellular matrix restructuring.
Conclusions: This study documents the propensity of immature cardiomyocytes to terminally differentiate and mature in EHT in a remarkably organotypic manner. It moreover provides the rationale for the utility of the EHT technology as a methodological bridge between 2D cell culture and animal models.
- cardiac myocytes
- caspase activation
- extracellular matrix
- tissue engineering
Different myocardial tissue engineering formats have been developed throughout the past decade.1 However, a low degree of cell maturation remains a key caveat in cardiac muscle engineering. A detailed understanding of “developmental” processes in tissue engineered myocardium probably is essential to guide tissue formation and maturation in vitro and to enhance the applicability of tissue engineered myocardium in substance screening, target validation, and tissue repair.
Normal heart muscle growth encompasses processes of terminal differentiation and maturation by hypertrophic growth, leading to the formation of binucleated and rod-shaped myocytes.2 Physiological maturation entails a characteristic shift in gene expression, including a reduction of transcripts encoding for fetal isoforms of myofibrillar proteins while the proportion of adult isoforms increases.3 Terminal differentiation, for example, withdrawal from the cell cycle, is another hallmark of advanced maturation already reached very early during development.4
Cardiomyocyte monolayer cultures show neither the distinct morphological (rod-shaped) nor the molecular (adult gene expression program) make-up of mature myocytes,5–7 probably as a consequence of the lack of a 3D growth environment and inappropriate biomechanical loading. Despite some evidence for advanced maturation in 3D tissue engineering models, it remains unclear whether cardiomyocytes in tissue engineered myocardium can, and if so to what extent, develop “physiologically” ex vivo.8,9
Three fundamentally different myocardial tissue engineering concepts are presently explored: (1) the classic approach involves cell seeding on preformed scaffold material10–14; (2) an alternative strategy is based on stacking cell sheets to generate multilayered muscle constructs15; (3) we have developed another method, taking advantage of the inherent capacity of immature cardiomyocytes to reassemble into spontaneously beating tissue if maintained at high density in a spatially defined culture environment under defined load.9,16,17 This cell entrapment method was further refined yielding engineered heart tissues (EHTs) with functional properties of native myocardium.9
In the present study, we demonstrate that EHT cultures can support terminal differentiation and tissue-like cardiomyocyte maturation. This finding is underscored by the similarity of morphological and molecular features of EHT- and postnatal heart-derived cardiomyocytes. Interestingly, the process of cardiomyocyte maturation and EHT-development showed 2 distinct phases: (1) a consolidation phase during culture days 0–7 with high levels of apoptotic cell death as well as extracellular matrix (ECM) degradation and (2) a maturation phase during culture days 7–12 with myocyte binucleation, formation of anisotropically organized sarcomeres in preferentially rod-shaped cardiomyocytes, a shift from fetal-skeletal to adult-cardiac actin transcript expression, and ECM build-up. Exposure to different hypertrophic stimuli during the maturation phase elicited distinct hypertrophic phenotypes, that is, concentric or eccentric hypertrophy.
EHTs were prepared from collagen type I, Matrigel, as well as neonatal rat heart cells (2.5×106) and cultured for 12 days.9
EHTs were labeled with 1 μCi/mL 3H-thymidine for 6 hours on the indicated culture days. DNA was prepared using standard procedures and subjected to liquid scintillation counting.
EHTs or hearts were digested with Liberase Blendzyme III in the presence of 30 mmol/L 2,3-butanedione monoxime at 37°C to prepare cardiomyocytes for morphological assessment by confocal laser scanning microscopy and flow cytometry.
One-Dimensional Electrophoresis and Immunoblotting
EHT protein was separated by SDS-PAGE. For detection of myosin heavy chain (MHC), isoforms gels were stained overnight with SYPRO Ruby. Blots were probed with monoclonal antibodies directed against indicated proteins and developed with ECL-plus.
Two-Dimensional Electrophoresis and Nanoflow Liquid Chromatography Tandem Mass Spectrometry
Protein extracts were separated by 2D electrophoresis. Protein spots were excised and enzymatically degraded. Peptides were separated by a nanoflow HPLC system on a reverse-phase column and applied to an LTQ ion-trap mass spectrometer.
3H-Phenylalanine and 3H-Proline Incorporation
EHTs were labeled with 1 μCi/mL 3H-phenylalanine or 3H-proline as indicated. Protein was precipitated in 10% ice-cold trichloroacetic acid at 4°C overnight and subjected to liquid scintillation counting.
35S-Cysteine/-Methionine Incorporation and Autoradiography
EHTs were labeled with 100 μCi/mL 35S-cysteine/-methionine for 18 hours on the indicated culture days and proteins were separated by SDS-PAGE. Gels were stained with Coomassie blue, immersed in Amplify Fluorographic solution (Amersham Biosciences), vacuum dried, and subjected to autoradiography.
An expanded Methods section can be found online at http://circres.ahajournals.org.
Construction of Spontaneously Contracting EHT
We generated spontaneously contracting EHTs from an initially liquid reconstitution mixture composed of enzymatically dispersed neonatal rat heart cells, collagen type I, and basement membrane proteins (Matrigel).9 Cell clusters within condensing EHTs started to beat within 2–3 culture days (Online Video I). On culture day 7, all EHTs contracted spontaneously and in unison (Online Video II). After 12 culture days, EHTs demonstrated a solid composition and vigorous contractions (Online Video III).
Hypertrophic Growth of Cardiomyocytes in EHT
The changes in EHT morphology and function prompted us to assess indices of cell proliferation and hypertrophy. We observed a marked reduction in cell number to 30% of the input cells between EHT culture day 0 (day of EHT construction) and day 12 (2.5×106 versus 0.8±0.03×106 cells/EHT; n=10;Figure 1A). Despite the cell loss, DNA content decreased only by 42% (45±3 versus 26±1 μg/EHT; n=17–20;Figure 1B), implying a relative increase of DNA content per cell (≈18 versus ≈32 pg DNA/cell on culture days 0 and 12, respectively). Further analysis of cardiomyocyte (α-actinin–positive; Online Figure I, A) and fibroblast (vimentin-positive; Online Figure I, B) quantity indicated that the initial cell loss was mainly the consequence of high cardiomyocyte loss while fibroblast content remained stable (Figure 1C). Cell cycle activity assessed by flow cytometry after 4′,6-diamidino-2-phenylindole (DAPI) labeling identified highest levels of DNA-synthesis in fibroblasts (17±1%) and cardiomyocytes (9±0.2%) on culture day 0 with lower but constant DNA synthesis levels during culture days 3–12 (fibroblasts: 7% to 10%; cardiomyocytes: 3% to 5%; n=4–8;Figure 1D). Interestingly, 3H-thymidine incorporation dropped markedly (P<0.05) during the first 24 hours of EHT culture to increase first slowly (until culture day 5) and then markedly (P<0.05) on culture days 7–12 (Figure 1E). The steep increase in DNA synthesis in the absence of increasing cell numbers appeared to be the consequence of (1) cardiomyocyte binucleation (37±2%; n=4 EHTs, 30–40 cells each;Figure 1F) and (2) enhanced nuclear DNA content (polyploidy) in a subset of cardiomyocytes (>2N: 14±1%, n=4) in advanced EHT cultures (Online Figure II). These properties, together with a high RNA/DNA ratio and 3H-phenylalanine incorporation (Online Figure III) appear to be signs of terminal differentiation and hypertrophic growth, suggesting advanced organotypic maturation in particular during late EHT culture.
Apoptotic Cell Death in Early EHT
The marked cell loss in EHT (Figure 1A) led us to investigate whether this was a consequence of apoptosis and represents a particular shortcoming of EHT versus conventional 2D cultures. Activated caspase-3, a surrogate for apoptosis, was especially high during early EHT culture (Figure 2A and Online Video IV). In agreement with this, high proapoptotic bax (Figure 2B) and low antiapoptotic Bcl-2 (Figure 2C) transcript abundances were observed, resulting in a markedly elevated bax/Bcl-2 ratio (Figure 2D). Notably, parallel cultures of EHT versus 2D showed similar levels of apoptosis (analyzed by flow cytometry), Trypan blue exclusion, and drop in cardiomyocyte number (Online Figure IV, A through C), collectively arguing against a unique apoptotic burden in EHT.
Lack of Evidence for Hypoxia in EHT
Hypoxia has been suggested as a limitation in tissue engineering and could have triggered apoptotic cell death. Surprisingly, we did not observe a regulation of highly sensitive hypoxia-response genes, for example, prolyl-4-hydroxylase domain isoforms 2 and 3 (prolyl-4-hydroxylase domain enzyme; PHD2/PHD3) mRNA and hypoxia-inducible factor-1α (HIF-1α) protein (Figure 3A through 3C). This led us to conclude that cells sensed normoxic conditions comparable to physiological tissue conditions throughout EHT culture and that elevated VEGF-A transcripts observed in late EHT culture (Figure 3D) were unrelated to hypoxia.
Sarcomere Maturation in EHT
A hallmark of maturation in terminally differentiated cardiomyocytes is the shift from a fetal to an adult gene expression program. This encompasses an increase in α-cardiac (α-cd) actin and a decrease in α-skeletal (α-sk) actin transcript concentration10,18 as well as a shift from β-(fetal/slow)-MHC to α-(adult/fast)-MHC in rodents.3,18 We could indeed observe an increase in α-cd actin and a parallel decrease in α-sk actin transcripts (Figure 4A and 4B), leading to an overall increase in total α-sarcomeric actin protein in late EHT-cultures (Figure 4C). In contrast, α-MHC transcript expression was unchanged, whereas β-MHC transcripts were elevated, resulting in a lower α-/β-MHC transcript ratio in EHT as compared with adult myocardium (7±2 versus 88±19-fold n=10/9;Figure 4D through 4F). Notably, also on protein level we identified a 4±0.8-fold (n=8) α-MHC excess in day 12 EHT (Figure 4F, inset). Direct comparison of atrial natriuretic peptide (ANP), α-sk actin, and α-cd actin transcript abundance in monolayer and EHT cultures documented that so called fetal genes (ANP, α-sk actin) were more abundant in monolayer as compared with EHT cultures (Online Figure V).
Dissection of the EHT Proteome
Increasing fluorescence intensity after α-actinin immunolabeling (in particular in day-12, EHT-derived cardiomyocytes; Online Figure I, A) provided additional evidence for tissue maturation. Subsequently, we performed proteome analyses to obtain a more comprehensive snapshot of the EHT proteome on culture days 0 and 12 (Figure 5A). The identity of a select set of proteins was confirmed by mass spectrometry (Online Table I). In agreement with the notion of advanced organotypic and in particular ventricular maturation in EHT, we observed a markedly enhanced level of the ventricular isoform of myosin light chain (MLC2v) protein per cardiomyocyte in day 12 versus day 0 EHTs (Figure 5B). Robust detection of tropomyosin isoforms, desmin, and M-type creatine kinase provided further evidence for the presence of cardiomyocytes with an advanced degree of maturation. Vimentin protein, characteristically expressed in fibroblasts, did not significantly change during EHT culture (Figure 5C), indicating phenotypic stability in this most abundant cellular constituent of EHT.
Structural Properties of EHT-Derived Cardiomyocytes
The shape and size of cardiomyocytes changed dramatically during EHT maturation from round and unstructured directly after isolation (mean diameter: 10±0.2 μm; volume: 570±32 μm3, n=60) to rod-shaped and clearly cross-striated with sarcomeres in registry in 12-day-old EHTs (mean diameter: 5.8±0.1 μm, length: 72±2 μm, volume: 2,040±120 μm3, n=113;Figure 6). Compared with cardiomyocytes from 12-day-old rats, EHT-derived cardiomyocytes acquired a similar length but were thinner and consequently less voluminous (mean diameter: 9.2±0.1 μm, length: 67±1 μm, volume: 4,724±154 μm3, n=3 hearts, 80–100 cells each;Figure 6). Myocytes from adult rats were clearly larger than EHT- and 12-day-old, rat heart–derived myocytes (mean diameter: 24.7±0.4 μm, length: 114±1 μm, volume: 57 102± 1735 μm3, n=3 hearts, 80–100 cells each;Figure 6). Mean diastolic sarcomere length was similar in all groups (EHT-derived: 1.85±0.04 μm [n=31]; day-12 rat heart: 1.83± 0.02 μm [n=28]; adult rat heart: 1.84±0.02 μm, [n=27]).
Organotypic Response of EHT to Hypertrophic Stimuli
We cultured EHTs during the maturation phase (culture days 7–12) in the presence of phenylephrine (20 μmol/L; PE) and angiotensin II (100 nmol/L; Ang) to assess their responsiveness to simulated neurohumoral overstimulation. Compared with standard medium conditions (EHT, day 12), we observed a concentric hypertrophy at the cellular level with a marked increase in cell width without major changes in cell length (Figure 6). During the course of these experiments, we also identified dramatic differences in responses to serum (10% in the culture medium) with “hypertrophy-inducing serum” (HIS), leading to a remarkable elongation of cardiomyocytes without major changes in cell width (Figure 6). These findings highlight the necessity for rigorous serum screens but also indicate the opportunity to use EHT for phenotypic screens to identify hypertrophy inducing secretomes and/or specific hypertrophic factors. Importantly, the phenotypic changes induced by PE/Ang and HIS were accompanied by distinct patterns of hypertrophic gene expression with particularly high ANP in HIS and low α/β-MHC ratio as well as high α-sk actin transcript abundance in PE/Ang treated EHTs (Figure 7).
Intense Matrix Restructuring During EHT Culture
At the time of casting, EHT contained 0.5±0.05 mg rat tail collagen, 1.1±0.02 mg extracellular basement membrane protein, 2.7±0.1 mg serum protein (in 210 μL DMEM with 20% horse serum and 4% chick embryo extract), and a cell suspension containing 2.2±0.1 mg proteins (2.5×106 cells in 377 μL DMEM with 10% fetal calf serum: cells, 0.5 mg; serum, 1.7 mg; n=4 in each group;Figure 8A). The nominal EHT protein content decreased from 6.4±0.1 mg (n=4) to 0.8±0.05 mg during culture (n=4;Figure 8A) despite elevated incorporation of 3H-phenylalanine (Online Figure III, B), 3H-proline (Figure 8B), and 35S-methionine/cysteine (Figure 8C). Experiments with the latter isotope mixture demonstrated pronounced incorporation especially in 40-kDa and >200-kDa proteins probably resembling actin (molecular weight: 43 kDa; see alsoFigure 4C) and myosin/collagen (molecular weight: 220/290 kDa). High 3H-proline incorporation (Figure 8B) and high collagen type I/III transcript levels (Figure 8D) at later stages of EHT culture provided further evidence for endogenous ECM synthesis during the EHT maturation phase (culture days 7–12). Sirius red staining documented thick collagen fibers (orange in polarized light) aligned along the major force axis in 12-day-old EHTs and thin collagen fibers (green in polarized light), which represent freshly synthesized collagen (Figure 8E). Additional evidence for de novo collagen matrix production stems from transmission electron microcopy, which identified cross-striated mature collagen (Figure 8F), being absent in the original collagen hydrogel. Cell loss (Figure 1A) and matrix disaggregation were apparently key factors for EHT protein loss, especially during EHT consolidation (culture days 0–7). Upregulation of matrix metalloproteinases (MMP-2 and MMP-14; Online Figure VI, A and B) and their tissue inhibitors (TIMP-1 and TIMP-2; Online Figure VI, C and D) supported the hypothesis of intense matrix restructuring. MMP-3 and MMP-13 could not be reliably detected in neonatal rat heart cells (Ct values: >40 [MMP-3] and >35 [MMP-13]; n=10/target), but were clearly present at EHT-culture day 3 (Ct values: 28 [MMP-3] and 25 [MMP-13]; n=10/target), supporting the general concept of strong MMP-based matrix remodeling, especially at early time points of EHT culture.
We investigated the hypothesis that immature rat cardiomyocytes undergo terminal differentiation and a process of advanced organotypic maturation in 3D EHT cultures and made the following key observations: (1) cardiomyocytes matured in EHT in a partially organotypic manner as indicated by the formation of a clearly anisotropic and cross striated rod-shaped cell morphology, abundant binucleation, and a fetal-to-adult actin isoform shift; (2) the ventricular MLC2 isoform was identified in EHT and strongly upregulated on protein level, providing further evidence for advanced ventricular maturation; (3) cardiomyocytes in EHTs responded to different hypertrophic stimuli with distinct morphological (concentric versus eccentric hypertrophy) and molecular (fetal gene expression) changes; (4) apoptosis in enzymatically isolated myocytes limited cell and especially cardiomyocyte survival in EHT; (5) EHT resembled normoxic tissue at all investigated time points; (6) matrix restructuring paralleled EHT-development and resulted in at least partial replacement of ECM constituents. Collectively, our data documents that the EHT culture format induces terminal differentiation and advanced maturation of initially immature cardiomyocytes to a “ventricle-like” phenotype in vitro. The process of EHT “development” can be classified as “EHT-consolidation” (culture days 0–7) followed by “EHT-maturation” (culture days 7–12). Demonstration of qualitatively different responses, for example, concentric versus eccentric hypertrophy, to distinct hypertrophic stimuli, for example, PE/Ang versus HIS, suggests that EHT can be exploited as a novel test-bed to dissect mechanisms of hypertrophic growth.
Using tissue-engineered myocardium as a model of myocardial development or in substance screening clearly depends on its close resemblance with bona fide heart muscle. Classical monolayer cultures display little structural, molecular, and also functional similarities with mature myocardium and do in general not respond reliably to hypertrophic stimuli, unless subject to inherently hostile serum starvation at low seeding density. Most notably, cardiomyocytes in monolayer cultures quickly lose their regular rod-shaped morphology and concomitantly reexpress fetal genes, indicating a molecular “resetting” to a prenatal state of development.5–7 Growth on patterned substrates may partially improve this condition and support anisotropic growth19,20; yet, morphological and molecular data documenting advanced maturation and formation of 3D tissue on a macroscopic scale are limited.
Particular morphological hallmarks of advanced cardiomyocyte maturation are a rod-shaped geometry and binucleation. Cardiomyocytes enzymatically isolated from EHTs were rod-shaped and abundantly binucleated (37%), resembling to some degree a state of maturity observed in cardiomyocytes from 10- to 12-day-old rat hearts.2,21,22 However, compared with heart-derived cardiomyocytes, EHT-derived cardiomyocytes were thinner (length/width ratio, 12:1 in EHT versus 7:1 in 12-day-old rat hearts and 5:1 in adult rat hearts). This difference in aspect ratio was at least in part “normalized” under PE and Ang stimulation (length/width ratio, 10:1;Figure 6) but also paralleled by a “mild” induction of ANP, all in the presence of “normal” serum. Interestingly, HIS caused marked cardiomyocyte elongation and ANP upregulation, responses which have been implicated in “pathological” hypertrophy. Although additional studies are warranted to establish phenotype “serome” relationships and identify distinct underlying mechanisms, we believe that our data provide compelling evidence for EHTs as a robust and nearly “physiological” in vitro system, which could be used, for example, to decipher the complex paracrine regulation of physiological versus pathological growth. In line with this notion, we could recently provide confirmatory evidence for the role of the MEK1-ERK1/2 pathway in concentric versus eccentric myocyte hypertrophy, conditions associated with pressure and volume overload, respectively, by making use of the EHT system.23
During physiological myocyte development, elongation of cardiomyocytes precedes parallel sarcomere assembly.22,24 Subsequently, concentric hypertrophy, being the morphological correlate of parallel sarcomeric assembly, represents a compensatory mechanism to adapt to increasing load. Similarly, multinucleation and polyploidy have been reported to be enhanced under increasing load.25 Accordingly, DNA synthesis (Figure 1D and 1E) was markedly elevated particularly after day 7 of EHT culture, that is, the time when EHTs were subjected to phasic stretch. Interestingly, enhanced DNA synthesis did not go along with an increase in myocyte or nonmyocyte cell number, suggesting load-induced karyokinesis, in the absence of palpable cytokinesis.
On the molecular level, the shift from skeletal to cardiac actin transcript expression (Figure 4A and 4B and Online Figure V) and the detection of elevated ventricular MLC2 in 12-day EHT (Figure 5B) provided further evidence for advanced organotypic maturation of cardiomyocytes in EHT. In apparent disagreement with this notion was the absence of the commonly reported massive β- to α-MHC transcript isoform shift. This may, however, be a consequence of the low (subphysiologic) endogenous beating frequency of EHT (≈2 Hz), making faster actin-myosin kinetics dispensable. Whether electric stimulation of EHT at near physiological frequencies (6 Hz) would facilitate a shift from the observed ≈7-fold α-MHC transcript excess in spontaneously beating EHTs to a ≈88-fold excess as observed in adult heart muscle (Figure 4F), needs further investigation. Interestingly, PE/Ang and HIS lowered the α-/β-MHC transcript ratio, as anticipated under hypertrophy-inducing conditions.
Abundant caspase-3 activation and elevated bax expression suggested that apoptosis limited cell survival in EHT. It is important to note that caspase activation does not always lead to fully executed apoptosis with nuclear fragmentation but is also involved in reversible myofilament breakdown after cell isolation.26 Induction of apoptosis during enzymatic cell isolation and cell loss during early culture are also commonly observed in monolayer cardiomyocyte cultures (Online Figure IV). This set of data collectively argues against the notion that the reported apoptosis represents a specific tissue engineering limitation. We could recently demonstrate that activation of prosurvival pathways such as the Akt pathway can protect cardiomyocytes in early EHT cultures from apoptosis.27
Hypoxia-induced apoptosis has been suggested as a main limitation for cell survival in tissue-engineered myocardium,28 and we initially interpreted VEGF-A transcript elevation as a sign of chronic hypoxia in particular in later stages of EHT culture. However, more comprehensive investigations of more sensitive biomarkers for acute (HIF-1α) and chronic (PHD2/3) hypoxia did not provide any evidence in support of EHT hypoxia during culture. We emphasize that cardiomyocytes are physiologically exposed to oxygen pressure below 40 mm Hg, which corresponds to <5% ambient oxygen,29 and that the provided oxygen supply (21% ambient oxygen) is apparently sufficient for normoxic EHT maintenance. Although the stimulus for enhanced VEGF-A expression has not been identified, one should consider that VEGF-A by itself may be cardioprotective30 and in fact may be an important prerequisite for the observed rapid vascularization of EHT grafts in vivo.31,32
Of particular interest for in vivo applications in regenerative medicine is also the apparent replacement of the original hydrogel by endogenously produced ECM. This remodeling process is on the one hand crucial for the formation of mechanically stable EHTs. On the other hand, it provides a perspective for the generation of nonimmunogenic “therapeutic” EHTs from autologous cells.
Taken together, cardiomyocyte maturation in EHT compares favorably to myocyte maturation in monolayer culture and does to some extent simulate physiological development in vivo. The observed differences in cardiomyocyte size and MHC isoform composition may be a consequence of “subphysiologic” loading and low intrinsic contraction frequency and thus may be interpreted as a “physiological” response to partially unphysiologic culture conditions. Interestingly, concentric and eccentric hypertrophic growth could be stimulated in EHT using simulated neurohumoral/serum activation. These data in particular suggest that EHT may constitute a unique model system to study mechanisms governing hypertrophic growth in cardiomyocytes. As a consequence of the observed differentiation and maturation inducing capacity, EHT cultures may also find a novel application as in vitro test-bed to define the fate of progenitor cells in a tissue-like context.
Sources of Funding
This study was supported by the German Research Foundation (DFG; Zi708/7–1, 8–1, 10–1, FOR604, and KFO155 to D.K., W.A.L., and W.H.Z.), the Federal Ministry for Education and Research (01GN 0520, 01GN0827, and 01GN0957 to W.H.Z.), the Deutsche Stiftung für Herzforschung (F29/03 to W.H.Z.), and the European Union (EU FP7 CARE-MI to W.H.Z.).
We thank M. Bauer and N. Feifel for designing qPCR primer-probe sets, B. Endress for excellent technical assistance, and C. Perske for helping with light microscopy of Sirius red stained samples.
In August 2011, the average time from submission to first decision for all original research papers submitted to Circulation Research was 13.5 days.
Non-standard Abbreviations and Acronyms
- atrial natriuretic peptide
- extracellular matrix
- engineered heart tissue
- myosin heavy chain
- myosin light chain, ventricular isoform
- matrix metalloprotease
- prolyl-4-hydroxylase domain enzyme
- tissue inhibitor of MMP
- vascular endothelial growth factor
- Received July 5, 2011.
- Revision received August 22, 2011.
- Accepted September 7, 2011.
- © 2011 American Heart Association, Inc.
- Eschenhagen T,
- Zimmermann WH
- Lompre AM,
- Nadal-Ginard B,
- Mahdavi V
- Weikert C,
- Eppenberger-Eberhardt M,
- Eppenberger HM
- Radisic M,
- Park H,
- Shing H,
- Consi T,
- Schoen FJ,
- Langer R,
- Freed LE,
- Vunjak-Novakovic G
- Zimmermann WH,
- Schneiderbanger K,
- Schubert P,
- Didie M,
- Munzel F,
- Heubach JF,
- Kostin S,
- Neuhuber WL,
- Eschenhagen T
- Shimizu T,
- Yamato M,
- Isoi Y,
- Akutsu T,
- Setomaru T,
- Abe K,
- Kikuchi A,
- Umezu M,
- Okano T
- Eschenhagen T,
- Fink C,
- Remmers U,
- Scholz H,
- Wattchow J,
- Weil J,
- Zimmermann W,
- Dohmen HH,
- Schafer H,
- Bishopric N,
- Wakatsuki T,
- Elson EL
- Feinberg AW,
- Feigel A,
- Shevkoplyas SS,
- Sheehy S,
- Whitesides GM,
- Parker KK
- Anversa P,
- Olivetti G,
- Loud AV
- Kehat I,
- Davis J,
- Tiburcy M,
- Accornero F,
- Saba-El-Leil MK,
- Maillet M,
- York AJ,
- Lorenz JN,
- Zimmermann WH,
- Meloche S,
- Molkentin JD
- Olivetti G,
- Anversa P,
- Loud AV
- Engelmann GL,
- Vitullo JC,
- Gerrity RG
- Communal C,
- Sumandea M,
- de Tombe P,
- Narula J,
- Solaro RJ,
- Hajjar RJ
- Das DK,
- Maulik N
- Zimmermann WH,
- Melnychenko I,
- Wasmeier G,
- Didie M,
- Naito H,
- Nixdorff U,
- Hess A,
- Budinsky L,
- Brune K,
- Michaelis B,
- Dhein S,
- Schwoerer A,
- Ehmke H,
- Eschenhagen T
Novelty and Significance
What Is Known?
Tissue-engineered myocardium may be used as in vitro tool for drug development and as a surrogate for heart muscle for in vivo applications in myocardial repair.
Cardiac myocytes dedifferentiate in culture, leading to a loss in organotypic cell morphology and reexpression of fetal genes.
The validity of classical monolayer cultures as in vitro platform for modeling of hypertrophic cardiomyocyte growth is limited.
What New Information Does This Article Contribute?
Engineered heart tissue (EHT) formation is a staged process comprising an early tissue consolidation phase with selection of the “fittest” myocytes and fibroblasts as well as comprehensive extracellular matrix (ECM) remodeling; this is followed by a phase of organotypic maturation.
Organotypic maturation of cardiomyocytes in EHT includes terminal differentiation, abundant binucleation, development of an essentially rod-shaped morphology, and a fetal-to-adult shift in gene expression pattern.
EHT may be useful in modeling of concentric versus eccentric cardiomyocyte hypertrophy.
Tissue engineering could potentially provide realistic heart muscle models and surrogate myocardium. However, cellular maturity in tissue-engineered myocardium has been sparsely documented. We show that a unique collagen hydrogel-based, cardiac tissue-engineering format, EHT, supports organotypic maturation in originally immature cardiomyocytes from neonatal rats. Our studies provide novel insight into the developmental properties of EHT, for example, hypertrophic growth under normoxic conditions and comprehensive ECM remodeling leading to replacement of the original hydrogel scaffold with ECM. The latter finding highlights the predicted but thus far undemonstrated capacity of the cardiac fibroblast to function as a key “engineer” in myocardial tissue engineering. We have established appropriate experimental conditions that differentially affect cardiomyocyte growth (ie, concentric hypertrophy under simulated neurohumoral activation and eccentric hypertrophy in the presence of hypertrophy-inducing serum). Collectively, the results of this study enhance the utility of the EHT technology as methodological bridge between classic 2D cell culture and animal models. It may represent a useful tool for identifying specific environmental cues that facilitate organotypic maturation of cardiomyocytes from human (stem) cell sources as well.