Regeneration of the myocardium by transplantation of cardiomyocytes is an emerging therapeutic strategy. Human embryonic stem cells (HESC) form cardiomyocytes readily but until recently at low efficiency, so that preclinical studies on transplantation in animals are only just beginning. Here, we show the results of the first long-term (12 weeks) analysis of the fate of HESC-derived cardiomyocytes transplanted intramyocardially into healthy, immunocompromised (NOD-SCID) mice and in NOD-SCID mice that had undergone myocardial infarction (MI). Transplantation of mixed populations of differentiated HESC containing 20-25% cardiomyocytes in control mice resulted in rapid formation of grafts in which the cardiomyocytes became organized and matured over time and the noncardiomyocyte population was lost. Grafts also formed in mice that had undergone MI. Four weeks after transplantation and MI, this resulted in significant improvement in cardiac function measured by magnetic resonance imaging. However, at 12 weeks, this was not sustained despite graft survival. This suggested that graft size was still limiting despite maturation and organization of the transplanted cells. More generally, the results argued for requiring a minimum of 3 months follow-up in studies claiming to observe improved cardiac function, independent of whether HESC or other (adult) cell types are used for transplantation.

CMPCs differentiated into the same cell types in situ as can be obtained in vitro. This excludes the need for in vitro pre-differentiation, making CMPCs a promising source for (autologous) cell-based therapy.

We have developed a mouse severe combined immunodeficient (SCID) model of myocardial infarction based on permanent coronary artery occlusion that allows long-term functional analysis of engrafted human embryonic stem cell-derived cardiomyocytes, genetically marked with green fluorescent protein (GFP), in the mouse heart. We describe methods for delivery of dissociated cardiomyocytes to the left ventricle that minimize scar formation and visualization and validation of the identity of the engrafted cells using the GFP emission spectrum, and histological techniques compatible with GFP epifluorescence, for monitoring phenotypic changes in the grafts in vivo. In addition, we describe how magnetic resonance imaging can be adapted for use in mice to monitor cardiac function non-invasively and repeatedly. The model can be adapted to include multiple control or other cell populations. The procedure for a cohort of six mice can be completed in a maximum of 13 weeks, depending on follow-up, with 30 h of hands-on time.