Existing cardiomyocytes generate cardiomyocytes at a low rate after birth in mice.

We generated two novel models for this study using the latest version of the sophisticated “mosaic analysis of double markers” (MADM) mouse model, in which asymmetric, indelible labeling of two daughter cells enables elucidation of precursor-progeny relationships. Unique to this system, only cells that arise from mitosis become achieve a specific label, so there is no ambiguity between newborn cells and cells that have undergone binucleation or DNA repair (a major limitation for cardiac assays that rely on proxies of division). The first model utilizes a cardiomyocyte-specific Cre (Myh6-CreERT2), and therefore only cardiomyocytes can undergo MADM recombination; only Myh6-expressing cardiomyocytes can be potential “parent” cells. In the second model (using ß-actin-CreER) any cell type can undergo MADM recombination, allowing us to elucidate contribution to postnatal cardiogenesis from putative progenitor cells. Using these novel models (this is the first demonstration of the elegant MADM system in the heart), we show that cardiomyocytes undergo symmetric division postnatally at a very low rate and we do not find any evidence of asymmetric division, self-renewal, or multi-lineage clones that would be characteristic of a progenitor cell. Further, we do not observe that a myocardial infarction injury increases the global rate of cardiomyocyte generation. Importantly, because MADM labeling can only occur in the setting of concomitant Myh6 expression, our data excludes possible dedifferentiation of cardiomyocytes to progenitor cells prior to division (one of the arguments potentially raised against the zebrafish studies that demonstrate a cardiomyocyte origin of post-injury regeneration using lineage tracing, which only relies on the historical expression of cardiomyocyte-associated genes).