Perfusion-decellularized matrix: using nature's platform to engineer a bioartificial heart

Harald C Ott, Thomas S Matthiesen, Saik-Kia Goh, Lauren D Black, Stefan M Kren, Theoden I Netoff & Doris A Taylor


Summary:

The Taylor group at the University of Minnesota published this paper in Nature Medicine in early 2008 describing their newly developed method for the growth of a bioartificial heart, using an existing heart as a starting point. This is done in vitro by first decellularizing the original heart, leaving behind an empty "scaffold" of extracellular matrix, or ECM. This shell of ECM is then seeded with fresh neonatal cardiac cells, and incubated in either a perfused bioreactor or a static 2-D culture. Upon characterization of the resulting heart tissue samples and whole organs, it was found that while full physiologic contractile capacity was not obtained, the new hearts were indeed contractile in response to electrical stimulus, demonstrating the potential viability of this technique for the generation of artificial hearts.

In perfusing the whole-organ heart, three different detergent solutions were used and the relative effectiveness of each was compared. Between solutions of sodium dodecyl sulfate (SDS), polyethylene glycol (PEG), and Triton X-100, the SDS detergent was the most effective at decellularizing the organs.

These decellularized hearts were characterized with a variety of techniques to confirm that they were indeed free of cellular materials. Basic H&E stains were done to visualize heart tissues after treatment with detergent, and qualitatively examined for cells. DAPI stains for nuclei were done to verify the absence of nuclei following detergent treatment, as well as similar stains for cellular actin and myosin to confirm full decellularization. The mechanical properties of the decellularized constructs was also measured; the tangential of the decellularized tissues was higher than that of the cadaveric rat tissues, and similar to that of raw elastin and collagen matrix, as expected.

Figure 1. Whole heart, decellularized with SDS

Recellularization of the decellularized hearts was done in a bioreactor capable of simlulating physiologic conditions of the heart by perfusion with an oxygenated culture media. Neonatal cardiac cells were seeded onto the initial construct through intramural injection. Endothelialization of the construct was done through the seeding of endothelial cells in a media perfusion. After 8 to 10 days of growth, these were compared to similarly reseeded hearts which had been grown without perfusion in a simple tissue-culture dish. Experiments were conducted on isolated tissue sections, as well as on a whole-heart scale.

With the sectioned samples, contraction could be elicited with electrical stimuli within the 9th day of growth, up to a frequency of 4 Hz. The perfused tissues, however, were capable of generating contractile forces some 300-800% that of the non-perfused tissues. Furthermore, the size of cardiac muscles generated in the non-perfused samples was limited to ~50 microns in diameter, while the perfused tissues were able to generate viable myofibers up to 1.1 mm thick.

Characterization of the whole heart focused on examining the ability of the recellularized hearts to generate physiologic contractile forces in the ventricles against an afterload. The hearts grown in perfused conditions were capable of generating 2.4 mmHg of contractile force in response to electrical stimulation of up to 4 Hz, with diminishing force as the frequency was increased.

Fig 2. Setup for characterization of whole heart reconstructs

Significance:

Previous studies had already demonstrated to viability of the decellularization technique in generating a matrix scaffold onto which cells can be seeded, but this group is the first to have gone beyond the tissue level and applied the technique to an entire organ. At the time of publication, the "viability" of the heart constructs generated was still limited - the 2.4 mmHg contractile force achieved is only ~2% of the capacity of a typical adult rat heart. However, the authors were optimistic in their conclusions, highlighting the fact that the heart constructs even had fuctional, competant valves. The fact that the cells in the regenerated organ were capable of depolarizing and repolarizing in response to electrical stimuli, generating measureable mechanichal contractions, demonstrates the potential for this technique as a method for growing replacement human organs for transplant.


- David Li