The beat goes on: human heart muscle from pluripotent stem cells.

In 2003, C. Zandonella reviewed in Nature how much was achieved of the 1999 promise to grow a functioning heart in the dish in a decade and what the perspectives were at that time.1 Today, more than a decade after its public prediction, the heart in the dish is still an unfulfilled dream, but repairing injured hearts with engineered myocardial tissue patches is a viable and increasingly realistic perspective in regenerative cardiology. This is not so much because of the progress in tissue engineering techniques but rather because of the dramatic advances in stem cell biology. In 1998, the first human embryonic stem cells (hESC) were described,2 and 3 years later the generation of cardiac myocytes from hESC.3 Despite widespread ethical concerns and strict legal restrictions in many countries of the world including the US and Germany, hESC proved to be enormously important for several reasons. First, they helped to better understand the earliest steps of human development and human stem cell biology, which differ in critical aspects from those in the mouse. Second, human ESC provided, for the first time, an unlimited cell source with an undisputed capacity to differentiate into essentially all types of cells of the body, including cardiac myocytes. Their discovery boosted cardiac tissue engineering principally by answering the question as to where the several hundred millions of human cardiac myocytes may come from that are necessary to make large myocardial patches.4 Third, ironically, the ethical concerns against hESC work stimulated and in some cases initiated the search for pluripotent alternatives that would be devoid of the regulatory and ethical problems of hESC. This motivation, combined with the technical experience acquired with mouse and human ESC cultures, was an important driver for the progress in stem cell technology, including the nondestructive derivatization of pluripotent stem cells from single mouse blastomeres5 or human IVF embryos,6 mouse7 and human spermatogonial stem cells,8 or human parthenogenic blastocysts.9 Finally, the seminal discovery of methods to induce pluripotency in mouse and human somatic cells by introducing a combination of pluripotency factors would not have been possible without the longstanding experience both groups previously had with ESC.10–12 Human ESC, at least in principle, solved the issue of a renewable human cardiac myocyte source for cardiac regeneration and human-induced pluripotent stem (hiPS) cells that of a patient-derived autologous approach. It should be mentioned, however, that even autologous iPS may induce an immune response, likely due to abnormal gene expression.13 Another important bottleneck remained for years: the low efficiency of cardiac myocyte differentiation from pluripotent stem cells, which precluded the generation of real human heart muscles as well as efficient stem cell therapy. This problem was solved by studies that carefully deciphered the factors guiding the earliest steps of cardiac development under normal conditions and translated this knowledge into cell culture protocols.14–17 Thus, by using a multistep protocol with growth factors that subsequently drive mesodermal and then cardiac specification, cardiac myocyte differentiation rates went up from approximately 1%3 to 50% and more.17 By optimizing growth factor combinations and timing, cardiac myocyte differentiation rate can be increased even further and, importantly, the new protocols can be applied to essentially all pluripotent stem cell lines including hiPS.18,19 In light of these findings, one may ask why one should choose the cumbersome strategy of engineering a tissue patch in the dish instead of simply injecting stell cells or their derivatives into the heart. The arguments are severalfold. (i) The retention and survival rate of cells is consistently low when injected into the myocardium and even lower when cells are infused.20 When using potentially tumorigenic cells, such as ESC, the high wash-out rate after cell injection leads to cell deposition in the lung, liver, kidney, and spleen21 and constitutes a clear risk of systemic tumor formation as demonstrated in mice.22 (ii) There is not much evidence to support the intuitively attractive idea that the adult mammalian heart provides a “cardiogenic milieu” that will drive maturation and orientation of cardiac myocytes from pluripotent stem cells. Thus, it is unclear how massive numbers of immature cells that are required for cardiac repair shall form a well-organized tissue that functions in synchrony with the host myocardium and support its contractile function. Indeed, although survival of injected embryonic stem cell–derived cardiac myocytes has been reported, the efficiency of the formation of new, electrically coupled, and well-differentiated myocardium appears rather low.23–25 (iii) Injecting hundreds of millions of cells, even if they would be retained at the site of injection, carries the potential risk of inflammation. Such effect has been shown in studies with bone marrow–derived cells,26,27 but not in several others that reported instead anti-inflammatory effects after injection of mesenchymal28 or cardiosphere-derived cells.29 (iv) Importantly, propagating iPS and maybe any other pluripotent cell to very large numbers in the dish carries a substantial risk of mutations Original received May 23, 2011; revision received May 23, 2011; accepted May 31, 2011. From the Department of Experimental Pharmacology and Toxicology, University Medical Center Hamburg Eppendorf, Germany. Correspondence to Thomas Eschenhagen, Department of Experimental Pharmacology and Toxicology, University Medical Center Hamburg Eppendorf, Martinistraße 52, 20246 Hamburg, Germany. E-mail [email protected] (Circ Res. 2011;109:2-4.) © 2011 American Heart Association, Inc.