Regenerative medicine

Heart redevelopment

Scientists report the conversion of one type of differentiated cell, the fibroblast, into another — the cardiomyocyte. This approach may find use in regenerative strategies for the repair of damaged hearts.

The human heart is made up of a number of cell types: these include beating muscle cells called cardiomyocytes; the vascular cells, which ensure a continuous blood supply to the heart muscle; and cardiac fibroblasts, which account for more than 50% of heart cells and have various roles, including some in heart repair. Researchers are currently directing considerable effort into coaxing stem cells through the series of complex developmental steps necessary to produce differentiated cardiomyocytes for repair of diseased or damaged hearts. In a paper published in Cell, Ieda et al.1 report an achievement in mice that offers an alternative route: direct reprogramming of both cardiac and skin fibroblasts into cardiomyocyte-like cells using three developmental transcription factors — Gata4, Mef2c and Tbx5.

The background to this experiment pre- dates the advent of regenerative medicine. The now-classical experiments2 in the 1960s showed that injecting a nucleus from a differentiated intestinal epithelial cell into a fertilized, but nucleus-free, frog egg gives rise to a normal adult frog. The differentiated nucleus was therefore reprogrammed by factors within the egg cytoplasm to a state of maximal developmental potential. The birth of Dolly the sheep in the 1990s through a similar approach3 proved that nuclear reprogramming is also possible in mammals.

These pioneering experiments made one point clear: differentiation from egg to adult cells occurs without permanent changes in the genome. However, they seemed to violate an axiom on which much of our understanding of developmental biology was based — that development involves progressive and stable restriction of the options available to an embryonic cell for lineage specialization4.

The ability of a single transcription factor to induce a profound shift in the identity of a differentiated cell was first demonstrated5 using the muscle-specification factor MyoD. This induced a complete redesign of the genome architecture — and biosynthetic and functional capacity — of a variety of differentiated cells. Although MyoD only partly converted some cells, and other cells were fully resistant to it, this study led to the idea of 'master regulatory genes'.

In the case of MyoD, conversion of cells to muscle-like cells occurred, in some cases, via an undifferentiated and proliferative progenitor state5. Since then, there have been many other examples of induced nuclear reprogramming, including conversion of legs to eyes in fruitflies6, fibroblasts to neurons7, pancreatic exocrine cells to functional β-cells8 and neural stem cells to monocytes9. Extraordinary among these examples is the groundbreaking finding that four transcription factors can convert both mouse and human fibroblasts, and a host of other differentiated cell types, into broadly potent stem cells that resemble embryonic stem (ES) cells10. This achievement sets the stage for in vitro production of patient-specific and disease-specific cells for therapies and analyses of disease mechanisms.

Although many of the transcription factors that guide heart development have been identified, no single factor seems to be a cardiac master regulator. But a recent study11 found that two transcription factors, Gata4 and Tbx5, in combination with Baf60c — a component of a protein complex that modifies DNA–protein fibres called chromatin — could induce differentiation of non-cardiac mesoderm from the embryo into beating cardiomyocytes. Such factors probably perform dual functions: the opening and rearrangement of chromatin, as has been shown for Gata4, and the induction of a self-maintaining program of gene expression that defines the differentiated state.

Ieda et al.1 tested different combinations of 14 genes that encode transcription factors with known roles in heart development for their ability to convert neonatal mouse fibroblasts in culture into cardiomyocytes. With introduction into fibroblasts of as few as three factors (Gata4, Tbx5 and Mef2c) — which include the two already known to be effective for reprogramming embryonic cells11 — some 25% of these cells expressed the cardiac-specific marker MHC and another marker of cardiomyocyte maturity, cTnT. However, these factors are involved in regulating a host of cardiac genes. It is therefore of paramount importance to determine whether the apparent cardiomyocyte traits that the authors observe reflect enforced expression of some cardiac genes or a truly self-maintaining cardiac program.

The authors1 note that their reprogrammed cells resembled cardiomyocytes in several ways: they were beating cells, had identifiable myofilaments containing the α-actinin protein, and showed Ca2+ oscillations and action potentials similar to adult ventricular cardiomyocytes. Moreover, the overall gene-expression pattern in these cells was broadly very different from that of cardiac fibroblasts and similar, although not identical, to that of neonatal cardiomyocytes. Modifications of histone proteins and of DNA associated with select genes suggested both broad changes in chromatin and establishment of new, stable gene-expression states.

The induced cardiomyocytes were stable for at least a week in culture, even after the expression of the three transcription factors was turned off. What's more, when the authors injected fibroblasts that had been reprogrammed in vitro into the heart muscle of live animals, the cells differentiated in situ to form small isolated cardiomyocyte-like cells with myofilaments. It is not known, however, whether these new inhabitants of the heart were electrically coupled.

The cell type that underwent 'transdifferentiation' in Ieda and co-workers' experiments is highly relevant to the mechanism of nuclear reprogramming that they report. Solid organs such as the heart contain a variety of cell populations, and, to isolate fibroblasts from smooth muscle and endothelial cells, the authors selected a subpopulation of neonatal cardiac fibroblasts that expressed the cell- surface protein Thy1. Such cells are likely to be highly immature and even stem-cell-like — in the adult, Thy1 marks mesenchymal stem-like cells and their early descendants, which are known to have the potential for differentiation into several other cell types. Nonetheless, the efficiency of conversion Ieda et al. report suggests that their starting material is not a rare stem-cell subpopulation. Whether adult cardiac fibroblasts can be similarly converted to cardiomyocytes should be determined, because it could be of relevance for the use of this strategy in the clinic.

Using genetic lineage tracing, the authors show that conversion into cardiomyocytes does not proceed via formation of an immature mesodermal or cardiac progenitor. However, given the known limitations of the tagging system they used (the marker genes of interest were expressed only transiently or at low levels), this finding would need confirmation using other experimental approaches.

More work is required before this approach1 sees use in heart repair. It is a fertile beginning, nonetheless, and adds significantly to the growing body of data suggesting that we might be able to short-circuit developmental programs and directly establish an entirely new and stable program for clinical use. As is the case for reprogramming of differentiated cells to an embryonic stem-cell-like state12, the mechanism underlying such transdifferentiation may involve obligatory changes in senescence genes, in epithelial status, in expression of regulatory noncoding RNAs such as micro-RNAs, and in chromatin modifiers. Beyond clinical potential, therefore, this paper opens many doors to explore the nature of reprogrammed cellular states.


  1. 1

    Ieda, M. et al. Cell 142, 375–386 (2010).

    Google Scholar 

  2. 2

    Gurdon, J. B. Annu. Rev. Cell Dev. Biol. 22, 1–22 (2006).

    Google Scholar 

  3. 3

    Campbell, K. H. S., McWhir, J., Ritchie, W. A. & Wilmut, I. Nature 380, 64–66 (1996).

    Google Scholar 

  4. 4

    Zhou, Q. & Melton, D. A. Cell Stem Cell 3, 382–388 (2008).

    Google Scholar 

  5. 5

    Weintraub, H. et al. Proc. Natl Acad. Sci. USA 86, 5434–5438 (1989).

    Google Scholar 

  6. 6

    Gehring, W. J. & Ikeo, K. Trends Genet. 15, 371–377 (1999).

    Google Scholar 

  7. 7

    Vierbuchen, T. et al. Nature 463, 1035–1041 (2010).

    Google Scholar 

  8. 8

    Zhou, Q., Brown, J., Kanarek, A., Rajagopal, J. & Melton, D. A. Nature 455, 627–632 (2008).

    Google Scholar 

  9. 9

    Forsberg, M. et al. Proc. Natl Acad. Sci. USA doi:10.1073/pnas.1009412107 (2010).

  10. 10

    Yamanaka, S. & Blau, H. M. Nature 465, 704–712 (2010).

    Google Scholar 

  11. 11

    Takeuchi, J. K. & Bruneau, B. G. Nature 459, 708–711 (2009).

    Google Scholar 

  12. 12

    Samavarchi-Tehrani, P. et al. Cell Stem Cell 7, 64–77 (2010).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Harvey, R. Heart redevelopment. Nature 467, 39–40 (2010).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing