Carbon-nanotube-embedded hydrogel sheets for engineering cardiac constructs and bioactuators

Abstract

We engineered functional cardiac patches by seeding neonatal rat cardiomyocytes onto carbon nanotube (CNT)-incorporated photo-cross-linkable gelatin methacrylate (GelMA) hydrogels. The resulting cardiac constructs showed excellent mechanical integrity and advanced electrophysiological functions. Specifically, myocardial tissues cultured on 50 μm thick CNT-GelMA showed 3 times higher spontaneous synchronous beating rates and 85% lower excitation threshold, compared to those cultured on pristine GelMA hydrogels. Our results indicate that the electrically conductive and nanofibrous networks formed by CNTs within a porous gelatin framework are the key characteristics of CNT-GelMA leading to improved cardiac cell adhesion, organization, and cell-cell coupling. Centimeter-scale patches were released from glass substrates to form 3D biohybrid actuators, which showed controllable linear cyclic contraction/extension, pumping, and swimming actuations. In addition, we demonstrate for the first time that cardiac tissues cultured on CNT-GelMA resist damage by a model cardiac inhibitor as well as a cytotoxic compound. Therefore, incorporation of CNTs into gelatin, and potentially other biomaterials, could be useful in creating multifunctional cardiac scaffolds for both therapeutic purposes and in vitro studies. These hybrid materials could also be used for neuron and other muscle cells to create tissue constructs with improved organization, electroactivity, and mechanical integrity.

Figure 1. Structural, physical and electrical characteristics of CNT-GelMA hydrogels

A) Schematic diagram illustrating the isolated heart conduction systems showing the purkinje which are located in the inner ventricular walls of the heart. Heart muscle with purkinje fiber networks on the surface of the heart muscle fibers. B) Preparation procedure of fractal-like CNT networks embedded in GelMA hydrogel. C) TEM image of GelMA coated CNTs. D) SEM images show porous surfaces of a 1 mg/ml CNT-GelMA thin film. Magnified image shows nanofibrous networks across and inside a porous structure. E) SEM images of pristine 5% GelMA. F) The elastic modulus of CNT-GelMA under compression at fully swollen state varies significantly with CNT concentration (*p < 0.05). G) Overall impedance of a 50 μm-thick hydrogel thin film decreased drastically with increased CNT concentrations.

Figure 2. Improved cardiac cell adhesion, maturation, and alignment on CNT-GelMA

A) Optical images of cardiomyocytes (day 1) revealed better cell retention and more homogeneous seeding on CNT-GelMA than on pristine GelMA. B) Percentage cell retention and C) viability one day after cell seeding on hydrogel surfaces showed strong dependence on CNT concentration (*p < 0.05). D) Normalized DNA quantities on day 3 and day 6 were not obviously affected by CNT concentration. E) Confocal images of cardiomyocytes after culturing for 5-days on pristine GelMA and 1 mg/ml CNT-GelMA revealed more uniform cell distribution and partial cell alignment on CNT-GelMA. F-actin and cell nuclei were labeled fluorescent green and blue respectively. Insets are corresponding FFT images. Higher-magnification images showed well-elongated cardiac cells and well-developed F-actin cross-striations (bottom right, white arrows) on CNT-GelMA, but not on pristine GelMA (bottom left). F) Alignment index derived from FFT images showed strong dependence on CNT concentrations (*p < 0.05). G) Stretching force resulted from strong cell-CNT interactions could affect cardiomyocyte organization and promote myotube striation.

Figure 3. Phenotype of cardiac cells on CNT-GelMA hydrogels

Immunostaining of sarcomeric α-actinin (green), nuclei (blue), and Cx-43 (red) revealed that cardiac tissues (8-day culture) on A) pristine GelMA and B) CNT-GelMA were phenotypically different. Partial uniaxial sarcomere alignment and interconnected sarcomeric structure with robust intercellular junctions were observed on CNT-GelMA. Immunostaining of Troponin I (green) and nuclei (blue) showed much less and more aggregated Troponin I presence on C) pristine GelMA than on D) CNT-GelMA. E) Quantification of α-actinin, Cx-43, Troponin I expression by western blot (*p < 0.05).

Figure 4. Improved mechanical integrity and advanced electrophysiological functions of cardiac tissues on CNT-GelMA

A) Spontaneous beating rates of cardiac tissues recorded from day 3 to day 9 on a daily basis. Phase contrast images showed cardiac tissues ruptured on B) pristine GelMA but intact on C) CNT-GelMA on day 5. D) Recording of synchronous beating signal of a tissue sample cultured on 1 mg/ml CNT-GelMA in response to applied external electric field at 0.5, 1, and 2 Hz. E) Excitation threshold of cardiac tissues on CNT-GelMA was 85% lower than that on pristine GelMA. F) SEM image shows morphology of cardiac cells cultured on CNT-GelMA. Red arrow: cytoplasmic prolongations adhered to CNT fibers; Yellow arrows: flat cell bodies.

Figure 5. CNT protected cardiac tissues against damages by heptanol and doxorubicin

Plots of spontaneous synchronous beating amplitude of cardiac tissues (5-day culture) on hydrogels with 0, 1, 3, 5 mg/ml CNTs over time in response to A) 4 mM heptanol and D) 300 μM doxorubicin. B) Time lapse before sporadic beating and stop beating induced by heptanol (*p < 0.05). C) Schematic illustration of alternative electric signal propagation through CNT networks after inhibition of direct intercellular gap junctions by heptanol. E) MTS data quantifying cell viability after 6 h exposure to doxorubicin (*p < 0.05). F) Proposed mechanisms for CNT to protect cardiac cells from damages caused by free oxygen radicals.

Figure 6. Engineered cardiac patches form free-standing 3D biohybrid actuators

A) Schematic drawing of tubular actuators (tightly and loosely rolled-up forms) and their corresponding beating directions (red arrow), along with optical images of two samples. Scale bar is 5mm. B) The displacement of the tubular actuator (yellow circled tip in a) over time under electrical stimulation (Square wave form, 1 V/cm, Frequency: 0.5 Hz – 3 Hz, 50ms pulse width). C) Spontaneous linear travelling of a triangular swimmer, as shown in optical images (at 0 and 7 secs) and the displacement vs. time plot. Ruler marking in b) is 1mm.