Functional assembly of engineered myocardium by electrical stimulation of cardiac myocytes cultured on scaffolds

Abstract

The major challenge of tissue engineering is directing the cells to establish the physiological structure and function of the tissue being replaced across different hierarchical scales. To engineer myocardium, biophysical regulation of the cells needs to recapitulate multiple signals present in the native heart. We hypothesized that excitation–contraction coupling, critical for the development and function of a normal heart, determines the development and function of engineered myocardium. To induce synchronous contractions of cultured cardiac constructs, we applied electrical signals designed to mimic those in the native heart. Over only 8 days in vitro, electrical field stimulation induced cell alignment and coupling, increased the amplitude of synchronous construct contractions by a factor of 7, and resulted in a remarkable level of ultrastructural organization. Development of conductive and contractile properties of cardiac constructs was concurrent, with strong dependence on the initiation and duration of electrical stimulation.

Nearly 8 million people in the United States have suffered from myocardial infarction, with 800,000 new cases occurring each year (1). Conventional therapies are limited by the inability of myocardium to regenerate after injury and the shortage of organs for transplantation. One novel potential treatment involves the in vitro cultivation of cell-based cardiac grafts that can be surgically attached to the myocardium (2–9). However, cardiomyocytes cultured using conventional methods do not align and remain poorly differentiated (2). The application of cyclic mechanical stretch substantially improves cell differentiation and force of contraction, but the distribution of gap junctions remains unclear, and some hallmarks of cardiac development are missing, most critically the M bands and intercalated discs (9). Notably, orderly coupling between electrical pacing signals and macroscopic contractions, which is crucial for the development and function of native myocardium (10), has not been used in vitro. It occurred to us that applying electrical signals designed to induce synchronous construct contractions would enhance cell differentiation and functional assembly of engineered tissue by means of physiologically relevant mechanisms. We report here a biomimetic approach in which cardiac constructs cultured in vitro were induced to contract in response to a pulsatile electrical field. Electrical stimulation resulted in the progressive development of conductive and contractile properties in cardiac constructs over just 8 days of culture, with strong dependence on the initiation and duration of the applied electrical stimulation.

Materials and Methods

Construct Preparation. Neonatal rat ventricular myocytes (6 × 10⁶ cells) were seeded onto Ultrafoam collagen sponges (6 × 8 × 1.5 mm) using Matrigel (30 μl, Becton Dickinson) (11). Constructs were precultured without electrical stimulation for 1, 3, or 5 days in six-well plates (1 construct per well in 4 ml of culture medium in a 37°C/5% CO₂ humidified incubator) to allow the cells to attach to scaffolds and to recover from isolation.

Electrical Stimulation. After preculture, constructs were transferred to a glass chamber fitted with two 1/4-inch-diameter carbon rods (Ladd Research Industries, Burlington, VT) placed 1 cm apart and connected to a cardiac stimulator (Nihon Kohden, Tokyo) with platinum wires (Ladd Research Industries). Silicone spacers were used to create six wells between the two electrodes, containing one construct apiece. In the “stimulated” group, trains of electrical pulses (rectangular, 2 ms, 5 V/cm, 1 Hz) that are characteristic for native myocardium (12) and previously used for cell monolayers (13) were applied continuously for an additional 5 days. Constructs cultured without electrical stimulation under otherwise identical conditions served as “nonstimulated” controls. All chambers were mixed orbitally at 25 rpm to enhance mass transport at tissue surfaces (see Supporting Methods, which is published as supporting information on the PNAS web site).

Drug Studies. After 3 days of preculture, constructs were transferred to stimulation dishes, and one of the following drugs was applied: verapamil (50 μM), palmitoleic acid (500 μM), or LY294002 (50 μM) for an additional 5 days, with or without electrical stimulation. Unsupplemented culture medium served as a drug-free control. At the end of cultivation, contractile response was tested in the drug-free Tyrode's solution (see Supporting Methods).

Assessments. The progression of tissue assembly was assessed at various hierarchical levels: molecular [expression levels, amounts, and distributions of cardiac proteins: connexin-43 (Cx-43), cardiac troponin I (Tn-I), α and β isoforms of myosin heavy chain (MHC), α-actin, and β-integrin], cellular (cell number, viability, and metabolism), ultrastructural (morphology of cells and nuclei, development and volume fractions of sarcomeres, and development and frequency of gap junctions, intercalated discs, mitochondria, and microtubules), and functional [amplitude of contractions, excitation threshold (ET), maximum capture rate and transmembrane potentials]. Construct properties were compared to those measured for neonatal rat heart ventricles (see Supporting Methods for detailed methods).

Expression and presence of cardiac markers were determined by RT-PCR, Western blots, and immunohistochemistry. The ultrastructure was evaluated by morphometric analysis of transmission electron microscopy images. Contractile activity in response to electrical field stimulation was measured by video-microscopy (12). ET, the minimum voltage at which the entire construct was observed to beat, and the maximum capture rate, defined as the maximum pacing frequency for synchronous construct contractions, were determined. Amplitude of contraction was determined as the fractional change of the construct surface area. Electrical activity in constructs was recorded at the end of cultivation by using a platinum electrode positioned ≈2 mm away from the pair of stimulating electrodes (carbon rods).

Statistical Analysis. Differences between the groups were analyzed by using Tukey's test followed by one-way ANOVA for pairwise comparisons or Dunn's test. P < 0.05 was considered significant.

Results and Discussion

The application of electrical stimulation during construct cultivation markedly enhanced the contractile behavior (see Movies 1–3, which are published as supporting information on the PNAS web site). After 8 days of culture, the amplitude of contractions was 7-fold higher in stimulated than in nonstimulated constructs (Fig. 1A), a result of the progressive increase with the duration of culture (Fig. 1B). The ET decreased (Fig. 1C), and the maximum capture rate increased (Fig. 1D), both with time and because of electrical stimulation, suggesting functional coupling of the cells. The shape, amplitude (≈100 mV), and duration (≈200 ms) of the electrical activity recorded for cells in constructs stimulated during culture (Fig. 1E) were similar to action potentials reported for constructs that were mechanically stimulated during culture (9).

Fig. 1.

Functional properties of engineered cardiac constructs. (A) Contraction amplitude of paced constructs cultured for a total of 8 days, with or without electrical stimulation, shown as a fractional change in the construct surface area. (B) Contraction amplitude progressively increased with time. One representative contraction cycle is shown measured after 1, 2, 3, and 4 days of cultivation with electrical stimulation. (C and D) ET decreased (C) and maximum capture rate increased (D) significantly both with time in culture and because of electrical stimulation. (E) Electrical potentials recorded in the stimulated group after 8 days of culture resembled the action potentials measured in native heart ventricles. S, stimulus; R, response. (F) Protein levels determined by Western blots at the end of preculture (day 3) and for stimulated and nonstimulated constructs at the end of culture (day 8). CK-MM, creatine kinase-MM. (G) Relative protein amounts as determined from Western blots by using integrated pixel density analysis. * denotes statistically significant difference (P < 0.05; Tukey's post hoc test with one-way ANOVA, n = 5–10 samples per group and time point).

On the molecular level, electrical stimulation elevated the levels of MHC, Cx-43, creatine kinase-MM, and cardiac Tn-I (Fig. 1 F and G). Expression of genes for sarcomeric α-actin, α-MHC, β-MHC, Cx-43, and β-integrin was confirmed by RT-PCR. The absence of atrial natriuretic factor expression and comparable RNA and DNA contents suggested the lack of pathological cell hypertrophy (14) due to electrical stimulation (see Fig. 5, which is published as supporting information on the PNAS web site). Comparable levels of mRNA in all groups are consistent with translational regulation (15). With time in culture, the ratio of α-MHC (mature) and β-MHC (neonatal) isoforms decreased in nonstimulated and increased in stimulated constructs, suggesting that the maturation of cardiomyocytes depended both on culture duration and electrical stimulation.

Notably, improved contractile properties of electrically stimulated constructs were not reflected in any apparent differences in construct cellularity, cell damage, or cell metabolism (see Table 1, which is published as supporting information on the PNAS web site), but correlated instead with cell differentiation (Fig. 2). Viable cells were confined within an ≈100-μm thick peripheral layer (Fig. 2 A), the thickness of which corresponded to the penetration depth of oxygen (7). Myofibers aligned in the direction of the electrical field lines (Fig. 2C, arrow), possibly in an attempt to decrease the apparent ET in response to pacing (16). Stimulated constructs and neonatal ventricles contained high levels of cardiac Tn-I (Fig. 2B), sarcomeric α-actin (Fig. 2C), Cx-43 (Fig. 2D), α-MHC (Fig. 2E), and β-MHC (Fig. 2F) and contained elongated cells aligned in parallel (Fig. 2 E Insets and F Insets). In contrast, cells in nonstimulated constructs stayed round and expressed relatively low levels of cardiac markers. After 8 days, stimulated constructs exhibited markedly higher density of Cx-43 than either early (3-day) or nonstimulated constructs (Fig. 2D). Cross-striations characteristic for mature cardiac myocytes were detected in stimulated constructs and native ventricles, but not in nonstimulated constructs (Fig. 2 E Insets and F Insets).

Fig. 2.

Histomorphology and expression of cardiac proteins. Representative sections of constructs at the end of preculture (day 3) and culture (day 8, with or without electrical stimulation); sections of native ventricles are shown for comparison. (A) Hematoxylin/eosin (H and E). (B) Tn-I (brown). (C) Sarcomeric α-actin (brown). Arrows denote myotubes. (D) Cx-43 (green). (E) α-MHC (red). Blue, cell nuclei; red arrows, positive cells; blue arrows, negative cells. (F) β-MHC (green). Blue, cell nuclei; yellow arrows, positive cells; blue arrows, negative cells; yellow arrows in Insets, striations. (Scale bars: A–F, 50 μm; D–F Insets, 20 μm.)

After 8 days, stimulated constructs demonstrated a remarkable level of ultrastructural differentiation, comparable in several respects with that of native myocardium. Cells in stimulated constructs were aligned, elongated, and contained centrally positioned elongated nuclei, in contrast to round cells in nonstimulated constructs that had a high nucleus-to-cytoplasm ratio (Fig. 3A). Stimulated constructs and neonatal ventricles contained abundant mitochondria positioned between myofibrils, in contrast to nonstimulated constructs containing mitochondria scattered around the cytoplasm (Fig. 3B), and substantially larger amounts of glycogen. Electrical stimulation induced the development of long, well aligned registers of sarcomeres, containing compact and clearly visible M and Z lines and H, I, and A bands that closely resembled those in native myocardium (Fig. 3 B and C), representing a hallmark of maturing cardiomyocytes (10).

Fig. 3.

Ultrastructural organization. Shown are representative micrographs of stimulated and nonstimulated constructs after 8 days of cultivation, compared with neonatal rat ventricles. (A) Cell shape and orientation. (B) Overview of myofibrils. (C) Structure of a sarcomere. (D) Intercalated disk. (E) Gap junctions. (Insets) T-tubule. (F and G) Morphometric analysis. Volume fractions occupied by nuclei, sarcomeres, and mitochondria (F) and the frequency of membrane junctions (G). IC, intercalated. * denotes statistically significant differences between the groups (P < 0.05; Tukey's post hoc test with one-way ANOVA; n = 18–46 samples per group and time point). (Scale bars: A and B, 2 μm; C and D, 1 μm; E and Insets, 0.5 μm.)

Morphometric analysis documented that the volume fractions occupied by cell nuclei and mitochondria decreased and increased, respectively, in a statistically significant manner from nonstimulated to stimulated constructs and from stimulated constructs to neonatal heart ventricles (Fig. 3F). The volume fraction of sarcomeres in stimulated 8-day constructs was indistinguishable from that measured for neonatal ventricles in the present study (Fig. 3F) and a previous study (17); in contrast, nonstimulated constructs contained scattered and poorly organized sarcomeres (Fig. 3B). In stimulated constructs, intercalated discs were positioned between aligned Z lines (Fig. 3D) and were as frequent as in neonatal ventricles; gap junctions also were substantially better developed (Fig. 3E) and more frequent (Fig. 3G). T-tubules were detected in all samples (Fig. 3E Inset). Overall, the application of biomimetic electrical stimulation resulted in remarkably well developed ultrastructure over only 8 days of cultivation.

The effects of electrical stimulation depended strongly on the time of its initiation. After enzyme digestion, cells were transiently incapable of transmitting electrical signals and contracting in response to pacing. After 3 days of culture, cells started to reassemble their excitation–contraction coupling machinery, as evidenced by occasional localized contractions of small construct regions. The application of field stimulation initiated at this time induced cell elongation and orchestrated the assembly of myofibrils and gap junctions. Electrical stimulation initiated early (day 1) substantially decreased the amounts of Cx-43 and α-MHC (see Fig. 6, which is published as supporting information on the PNAS web site), and the constructs could not be induced to contract synchronously over 5 days of culture, a finding consistent with the low degree of capture in cardiomyocyte monolayers that were stimulated too early (13). Electrical stimulation initiated late (day 5) enabled the establishment of gap junctions but failed to enhance the functional organization of contractile apparatus, such that the contractile activity could be induced only in small discrete areas.

The collected experimental evidence is consistent with the progressive development of the conductive and contractile properties of electrically stimulated constructs depicted in Fig. 4. During isolation from the heart tissue, cells lose many of their surface channels and receptors (18), disassemble their myofibrillar equipment, and acquire round shape (9). Cx-43 present in the cells at day 3 (Figs. 1F and 2D) was localized only in the cytosol (Fig. 2D Inset), suggesting the absence of functional connexons. During preculture (Fig. 4, Phase 1), cells reassembled conductive and contractile proteins, with peak amounts at day 3. If applied early (day 1), electrical stimulation inhibited the accumulation of cardiac proteins and yielded poor contractile behavior. If applied late (day 5), electrical stimulation had less effect because of the reduced amounts of Cx-43 and contractile proteins available in the cells.

Fig. 4.

Concurrent and progressive development of conductive and contractile properties of cardiac constructs cultured in vitro. During phase 1 (preculture without electrical stimulation), cells accumulate and assemble conductive and contractile proteins lost or disorganized during isolation from heart tissue; electrical stimulation has an inhibitory role. During phase 2 (cultivation with electrical stimulation), constructs are cultured with the application of electrical stimulation. The cells oriented in the direction of electrical field will be the first ones to elongate and establish gap junctions with neighboring cells. As the contractions begin, they drive the organization of sarcomeres and thereby increase the contractile force in response to electrical stimuli. Electrical stimulation enhances the development of ultrastructural and contractile properties of the individual cells (arrow to the right) and increases the number of functionally coupled cells engaged in synchronous contractions of the constructs (arrow to the left). Reproduced with permission from illustrator Jennifer Fairman (Copyright 2004, Fairman Studios).

The application of electrical stimulation to cultured constructs (Fig. 4, Phase 2) induced hyperpolarization at the anode end of the cell and depolarization at the cathode end of the cell (16), such that the cells aligned with the electrical field lines were subjected to the largest voltage difference and were likely the first ones to generate action potentials and contract. Processes forming at the cells' ends (19) promoted the establishment of gap junctions, propagation of pacing signals, and generation of action potentials (Fig. 1E) that induced synchronous macroscopic contractions (Fig. 1 A) driving the organization of sarcomeres (Fig. 3) and thereby increasing construct responsiveness to pacing (Fig. 1B).

To confirm that functional gap junctions, cytoskeletal organization, and excitation–contraction coupling are all necessary for the development of contractile behavior (Fig. 4), constructs were cultured from day 3 to day 8 in the presence or absence of verapamil [L-type Ca²⁺channel blocker (13)], palmitoleic acid [blocker of gap junctions (20)], and LY294002 [blocker of the PI3K pathway implicated in the cytoskeletal rearrangement of field-stimulated endothelial cells (21)], with and without electrical stimulation.

During culture with verapamil, electrical stimulation maintained the amount of Cx-43 at levels comparable with stimulated drug-free controls (see Fig. 7, which is published as supporting information on the PNAS web site), suggesting that electrical stimulation, even without contractile activity, established and maintained functional gap junctions (Fig. 4). These constructs contained high levels of Tn-I, but the cells were neither aligned nor elongated, findings consistent with the lack of contractile activity. Upon verapamil removal, stimulated constructs responded to pacing, although with smaller amplitude and higher ET than stimulated drug-free controls, whereas the nonstimulated constructs could not be induced to contract (at amplitudes of ≤9.9 V/cm). Blockage of gap junctions and the PI3K-Akt pathway had irreversible adverse effects on the conductive and contractile properties, respectively. Constructs exposed to palmitoleic acid contained rounded cells that expressed Tn-I but not Cx-43 and could not be paced. Blockage of the PI3K pathway resulted in asynchronous localized contractions at high ETs (>8 V), consistent with poor organization of contractile proteins (Fig. 7). Overall, drug studies showed that long-term (5-day) blockage of either gap junctions or cytoskeletal arrangement had terminal adverse effects on construct development that could not be reversed by electrical stimulation, whereas the inhibition of Ca influx had only temporary effects that were reversed if constructs were stimulated during cultivation.

The enhanced organization of contractile apparatus by electrical stimulation in vitro may parallel some aspects of embryonic development, in which contractile proteins are first laid down within the cells before contractions begin and then organized into sarcomeres, resulting in increased contractile force (22). Notably, there was a positive feedback between electrical stimulation and the progressive development of contractile behavior (Fig. 1B) and tissue ultrastructure (Fig. 3). Induction and sensing of stretch appears to be crucial for the proper development and assembly of sarcomeres. In a previous study, cardiac constructs exposed to cyclic stretch assembled myofibrils with compact Z lines (9). In the present study, nonstimulated constructs had thick, diffuse Z lines (Fig. 3C) similar to those found in the mouse knockout for muscle LIM protein, a key Z-line protein involved in the sensing of stretch and signal propagation (23). Notably, induction of synchronous macroscopic contractions by electrical stimulation ensured that Ca handling was unaltered (24) and that the capacity and density of L-type Ca²⁺ channels were maintained (13).

In the context of this study, we found that suprathreshold electrical field stimulation induced remarkable enhancement of cell alignment and coupling, synchronous construct contractions, and ultrastructural organization over only 8 days of cultivation. Important areas of future work could include optimizing the conditions of electrical stimulation (frequency, field gradient, initiation, and duration of stimulation), quantifying the effect of electrical field stimulation on other cell types present in the native myocardium (endothelial cells and fibroblasts), and testing the biological function and remodeling of the cardiac grafts after implantation in injured myocardium.