Hearty slices to plan for future health.

Clinical trials have shown that heart function can only be moderately improved by injection of stem cells. Gains to date are most likely the result of increased wall stiffness and angiogenesis rather than generation of new, beatingmyocytes. Poor results could be due to the number and kind of cells used, the timing andmethod of delivery, lack of engraftment in an inhospitable local environment, general immune responses, and many more variables. Thus, although the current approach has been responsive to public pressure to rush to translate new ideas to a cure, the lack of a sound foundation in the basics makes the failure of stem cell therapy to date unsurprising. Working with whole animals is a most difficult way to learn because somuch is beyond the control of the experimentalist and assays are limited. It is clear that experimental platforms are required to provide the rational groundwork before future cell therapy trials for cardiovascular disease are carried out. It is a pleasure to find a simple new platform of an in vitro beating heart model for long-term assessment of experimental therapeutics in this issue by Monville’s group. Ironically, some of the earliest ex vivo methods used explanted slices of organs in culture, which persisted with great success for the brain but were abandoned for the heart because physiological function lasted only a few hours. Surprisingly, 800 mm-thick heart slices grown on semi-porous membranes with an air–medium interface survive 3 months, which is far better than previous immersion methods. The thickness is puzzling as beating heart cells consume energy at a very high rate and are exquisitely sensitive to the lack of oxygen, relying heavily on oxidative metabolism to produce the ATP for the motor proteins and ion pumps. Isolated bundles and tissue-engineered constructs show cell death 150–200 mm below the surface, thought to be due to poor diffusion. As oxygen at 20% of air dissolves very poorly in water, it is hard to imagine how cells live deep in the heart slice because the vessels are useless without circulating blood. However, proof is provided by live nuclei seen in many figures and by the quantitative analysis of live human stem cells as a function of depth (see Supplementary material online, Figure S3). 6 Anatomy is science, too. Unfortunately, the field of heart stem cell science has been populated by sloppy numerical analyses of how many of which kind of cells are really present. Results have not been reproduced from one lab to another (see Murry et al. for a discussion of such issues). The first application of this heart slice platform is the rat ventricular tissue for the analysis of human embryonic stem cell engraftment into the myocardium in vitro. Both pharmacological interventions and cell therapeutics were characterized at the tissue organization level. Quantitative immunohistochemistry, gene expression, and mechanical function fully documented the conclusions. Assessment of various cell populations along with their respective engraftment capabilities is now possible. The novelty of this system is that the complex natural base provides a scaffold with live cell cues, scaffold matrix, and microstructure, so that this might be better than other models to date. There are several in vitro models with progressive complexities that have been used for stem cell research in addition to blind injection into the living animal. The simplest is the growing of stem cells on microwells, which can control both topography and growth factors but not physiological functions. Next, cardiac differentiation markers have been expressed in endothelial progenitor cells, human amniotic mesenchymal cells, and bone marrowderived mesenchymal stem cells after co-culture with a feeder layer of primary cardiomyocytes with spontaneous beating but unknown chemical milieu. A key point here is that although some cardiac differentiation markers were present in the stem cells, contraction was not. Tissue-engineered models in three dimensions are advantageous in the application of external forces to mimic the mechanically dynamic heart. However, integration of stem cells might be hampered because the constructs are covered with a sheath that might be difficult to penetrate. Finally, an animal heart was completely decellularized, leaving the extracellular matrix and vascular architecture for repopulation by endothelial stem cells to make vessels and rat neonatal cardiac myocytes. Electrical stimulation, paced contraction, and pumping for several weeks provide the proof of principle for future studies with numerous potential cardiac progenitor populations. However, despite the merits of a natural scaffold, good circulation, and The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology. * Corresponding author. Tel: þ1 312 413 0407; fax: þ1 312 996 6312. E-mail address: [email protected]