Human embryonic stem cell-derived cardiomyocytes (hESC-CMs) show considerable promise for regenerating injured hearts, and we therefore tested their capacity to stably engraft in a translationally relevant preclinical model, the infarcted pig heart. Transplantation of immature hESC-CMs resulted in substantial myocardial implants within the infarct scar that matured over time, formed vascular networks with the host, and evoked minimal cellular rejection. While arrhythmias were rare in infarcted pigs receiving vehicle alone, hESC-CM recipients experienced frequent monomorphic ventricular tachycardia before reverting back to normal sinus rhythm by 4 weeks post transplantation. Electroanatomical mapping and pacing studies implicated focal mechanisms, rather than macro-reentry, for these graft-related tachyarrhythmias as evidenced by an abnormal centrifugal pattern with earliest electrical activation in histologically confirmed graft tissue. These findings demonstrate the suitability of the pig model for the preclinical development of a hESC-based cardiac therapy and provide new insights into the mechanistic basis of electrical instability following hESC-CM transplantation.
Introduction
Following myocardial infarction (MI), necrotic cardiomyocytes are replaced with non-contractile scar tissue, often initiating heart failure. Currently available treatment options for post-MI heart failure include drugs that slow disease progression but do not reverse damage, mechanical circulatory support with complications including thrombosis, infection and the need for an external power supply, and cardiac transplantation limited by the inadequate supply of donor hearts. This situation has driven intense recent interest in the development of alternative cell-based approaches to achieve cardiac repair. The transplantation of various adult stem cell types has been reported to improve left ventricular (LV) contractile function, but beneficial effects appear to be modest and attributable to indirect mechanisms rather than the generation of new cardiomyocytes (
). By comparison, cardiomyocytes derived from pluripotent stem cells (PSCs) show stable engraftment in multiple MI models, repopulating the infarct scar with electromechanically-integrated new muscle (
,
,
,
,
,
,
). In an initial proof-of-concept study, our group showed that the transplantation of human embryonic stem cell-derived cardiomyocytes (hESC-CMs) in a rat MI model mediates the partial remuscularization of the infarct scar and has beneficial effects on regional and global LV contractile function (
). Later, we used a guinea pig MI model and a fluorescent graft-autonomous reporter of graft activation to show that hESC-CM grafts are capable of electromechanical integration and synchronous activation with host myocardium during systole (
).
There have also been more recent efforts to test hESC-CMs and related PSC derivatives in large-animal MI models. Primate ESC-derived multipotent cardiovascular progenitors have been shown to differentiate into multiple cardiac lineages including ventricular myocytes following allotransplantation into infarcted non-human primates (
). The Murry laboratory described the successful engraftment of committed cardiomyocytes from hESCs in the infarcted hearts of small macaques (
). In the latter study, the authors observed an impressive degree of remuscularization following hESC-CM transplantation, as well as histological evidence of graft cardiomyocyte maturation over time. However, hESC-CM recipients exhibited transient, non-lethal ventricular tachyarrhythmias (VTs) that were not observed in infarcted monkeys receiving vehicle alone. More recently,
described qualitatively similar results following the allotransplantation of primate induced PSC-derived cardiomyocytes (iPSC-CMs) in cynomolgus monkeys.
While the field has learned a tremendous amount from the work in the preceding animal models, we reasoned that efforts to develop and translate a safe, effective PSC-based cell therapy would greatly benefit from additional preclinical testing in the infarcted pig heart. The limitations of rodent MI models are widely recognized, and even the aforementioned transplantation work in non-human primate models involved relatively small species with substantially different cardiac structure and physiology from humans.
Macaca nemestrina and
Macaca fascicularis typically have body weights of ∼8 kg and ∼3 kg, respectively; and both species exhibit sinus heart rates that exceed those of humans (
Macaca nemestrina, ∼120 beats per minute [bpm];
Macaca fascicularis, ∼165 bpm; humans, ∼60–100 bpm) (
,
,
,
). Heart rate is a particularly relevant parameter to consider when evaluating electrophysiological outcomes following cell transplantation because recipient species with rapid rates could affect the electromechanical function of implanted human cardiomyocytes and/or mask graft-related arrhythmias that would otherwise occur in slower-rated humans. By comparison, the pig heart's weight-to-body ratio is nearly identical to that of an adult human (∼5 g/kg) (
). The pig heart also has a cardiac structure, sinus rate (∼90 bpm), and contractile function that closely resemble that of an adult human (
). Relative to the primate, the pig also provides significantly greater throughput and reduced experimental costs, and its larger size facilitates better imaging and makes it more amenable to interventions used in adult humans (e.g., catheter-based electroanatomical mapping [EAM]). Given these practical considerations, the pig has been routinely used for the late preclinical testing of novel cardiac interventions.
With this in mind, we hypothesized that hESC-CM transplantation into the infarcted hearts of suitably immunosuppressed pigs would result in their stable engraftment and the partial remuscularization of the infarct scar with outcomes comparable with that seen in other smaller, preclinical models. While this was a feasibility study with primarily histological endpoints, we also examined the functional consequences of hESC-CM transplantation, including LV dimensions and contractile function using cardiac magnetic resonance imaging (MRI), as well as electrophysiological behavior by telemetric electrocardiography (ECG) monitoring and catheter-based EAM and pacing studies.
