Development of scaffold-free, human induced pluripotent stem cell derived, cardiac tissues matured under physiological levels of pressure and stretch

Introduction:: Historically, in vitro cardiovascular research relied on cardiomyocytes isolated from animal sources or non-contractile model cell lines. The ability to generate and differentiate human induced pluripotent stem cells (hiPSCs) into cardiomyocytes (hiPS-CMs) has resulted in a renewable source of cardiomyocytes for cardiovascular research. hiPS-CMs share electrophysiological and biochemical markers of adult cardiomyocytes but remain embryonic in phenotype limiting their translational applications. Utilizing hiPS-CMs, Human Cardiac Tissue Chips (CTCs) have emerged in recent years as a powerful model of the human heart, used to study cardiac development and disease and in testing for cardiotoxicity in drug development. These systems aim to recapitulate key aspects of in vivo tissue such as mechanical properties, 3D structure, and both cardiac and non-cardiac cells. Despite many advances, achieving a CTC with cardiomyocytes matured into an adult, physiologically representative phenotype, remains a challenge. In this work, we aim to improve upon the state-of-the-art CTCs by (1) generating tissues using exclusively three-dimensional (3D) culture techniques without dissociating cellsthus preserving inter-cellular junctions formed during differentiation. (2) Avoiding the use of non-physiological exogenous scaffolding or extracellular matrix (ECM) material, which may induce a fibrotic phenotype. (3) Generating stromal cells along with cardiomyocytes from a single iPSC source. (4) Maturing the tissues under physiologically relevant levels of hemodynamic pressure and stretch, a key element of development often omitted with other CTC systems.


Materials and Methods:: We generated iPSC derived cardiac tissues through fusion of cardiac organoids. Briefly, agarose microwells were generated by 3D printing a master mold, replica molding into PDMS, and replica molding into agarose. These agarose wells we placed into 12 well plates and seeded with iPSCs, generating iPSC spheroids approximately 300 um in diameter. Once formed, the spheroids were moved into 6 well plates and kept on a shaker throughout differentiation. Differentiation into cardiac organoids was done by a 3-step modulation of Wnt signaling. Spontaneous beating was observed by day 10 of differentiation. Meanwhile, CTC devices were fabricated from PDMS. Briefly, a frame was cast from 3D printed molds, and a thin membrane was spin coated onto a silicon wafer. Soft lithography techniques were used to generate the millimeter scale loading well, and all components were plasma bonded together. The well area was then coated with Matrigel. Beating organoids were seeded into the wells and allowed to fuse for approximately 7 days (Fig A). Once fused, tissues were stimulated in a custom pneumatic set up. A filling pressure of 10 mmHg caused the tissue, adhered to the thin membrane, to stretch across a rounded obstacle until a programmable pump applied 15 mmHg of pressure, forcing the membrane back to its relaxed position (Fig B). This cycle was repeated at 60 bpm for 48 hours. Following stimulation, the electrical activity of the tissue was characterized by optical mapping using voltage sensitive dyes imaged with a photodiode array.


Results, Conclusions, and Discussions:: Using the described 3D techniques and sequential Wnt modulation, we were able to generate spontaneously beating cardiac organoids. Using flow cytometry, we confirmed the presence of cardiomyocytes (Cardiac Troponin T+), fibroblasts (vimentin+), and endothelial cells (CD31+). These isogenic cells were generated in a single protocol from the same iPSCs, allowing creating of a true “patient on a chip” device. These organoids were successfully fused into a continuous millimeter scale structure (Fig A) which exhibited spontaneous, synchronized beating. The fused structure and device design allowed for further maturation under pressure and stretch. The action potentials of the tissues were recorded using voltage sensitive dyes (Fig C), which were traced to generate an optical map of action potential propagation (Fig D). We observed that stimulation under pressure and stretch resulted in increased conduction velocity (14 vs. 8.2 cm/s) and increased action potential durations (APD 50 of 422 +/- 30 vs 188 +/- 34 ms). Both measures indicate cardiomyocyte maturation resulting from stimulation. In this ongoing work, we are characterizing the effects of gradually increasing pressure and stretch levels on electrophysiology along with other structural and functional makers of cardiomyocyte maturity.


Acknowledgements (Optional): :

References (Optional): :