Tung-Ying Lu entered the lab early one Sunday morning to follow up on a first attempt at a new experiment. Shocked by what she saw, she picked up the phone to call her lab director, Lei Yang, assistant professor and director of the Stem Cell Core at the University of Pittsburgh. “Good news,” she reported, “I can see the construct beating.”
Lu and colleagues had taken on the daunting challenge of creating a new heart using human stem cells. When he initiated the project in 2010, Yang believed that the construct would be so big that the cardiomyocytes differentiated from stem cells would be incapable of generating the mechanical force necessary to contract the new heart. But they did—and on the first attempt.
“That was the most surprising and most exciting part of this study,” Yang said.
Lu recorded a video immediately. “Frankly, we were not so sure we could make another beating heart, and we were not so sure it could beat the next day,” explained Bo Lin, co-first author with Lu on the Nature Communications article reporting their study (1). This initial success had to be documented.
Lin’s doubts were well founded; making a heart from scratch is clearly no simple matter.
Yang hopes to one day create heart tissues for clinical therapy, especially for personalized medicine. From the beginning, he intended to use human cells to create a new heart. To date, tissue engineering with human heart cells remains largely unexplored, partially due to the scarcity of human cardiomyocytes.
“We couldn’t get enough cardiomyocytes,” he said. “That is why embryonic stem (ES) and induced pluripotent stem (iPS) cells are a very nice resource. Those cells can provide an unlimited number of cardiomyocytes or cardiovascular progenitor cells for tissue engineering.”
Yang is no stranger to working with stem cells. Prior to opening his lab at the University of Pittsburgh, he spent several years as a postdoctoral fellow in the laboratory of Gordon Keller at the Mount Sinai School of Medicine. Together, he and Keller identified one of the earliest cardiovascular progenitor cell types, multipotential cardiovascular progenitors (MCPs; 2), and worked out the culture methods to differentiate MCPs into cardiac, endothelial, and vascular smooth muscle cells—just the cells needed to create the heart.
Led by Lin, Yang's group first reprogrammed human dermal fibroblasts into iPS cells and then differentiated these into MCPs using Yang and Keller's culture methods (3). “Using these as a cell resource, we can generate personalized heart constructs. That's definitely very important for the future development of personalized medicine or for personalized therapy of heart disease,” Yang explained.
With MCP cells in hand, they turned their attention to the heart. But with the need for appropriate proportions of different cell types organized into the proper architecture and communicating together, growing heart cells in a dish falls quite short of creating a fully functional organ.
A number of synthetic and natural matrices have been developed for tissue engineering, but rather than using those materials to structure their heart construct, Yang's group decided to repurpose an existing scaffold: a mouse heart. Using a method called perfusion decellularization developed in 2008 by Doris Taylor (4), currently Director of Regenerative Medicine Research at the Texas Heart Institute, Lu and Lin secured a mouse heart to some tubing and began perfusing it with trypsin and detergents.
“By definition, perfusion decellularization is the removing of cells,” said Jeremy Song, a visiting research fellow at Massachusetts General Hospital who recently used the same approach to create a scaffold for a functional rat kidney (5). The result was a translucent, flaccid ghost of the former organ made up of the extracellular matrix (ECM) network.
While the process of removing cells from heart, kidney, and lung has been optimized, protocols for repopulating the organ are less refined and more problematic. In the case of Song’s rat kidney scaffold, he decided to deliver both endothelial and epithelial cells to generate endothelial barriers and resorption, secretion, and endocrine functions, respectively.
“The kidney has a ureter that is designed to basically go only one way, so getting cells into the epithelial compartment was the greatest challenge,” Song said, noting that there were many failed trials along the way.
While endothelial cells seeded the decellularized kidney smoothly by gravity perfusion, it took a vacuum system to pull the epithelial cells into the scaffold. “Obviously, if you apply a very strong vacuum, you would not only suck cells into the epithelial compartment, but you would destroy some of the matrix and have the matrix break down due to the pure force of the suction,” explained Song. “On the opposite side, if you used very little suction, you would deliver very few cells.”
After extensive trial and error, Song finally found the right balance that allowed seeding of the kidney without too much damage to the structure.
Yang’s group had a different problem when repopulating the heart. In their case, although gravity perfusion pushed cells into the heart, within days they washed back out again. They resolved this by seeding cells intermittently, with eight hour breaks between perfusion cycles, allowing the cells to settle and attach to the ECM.
In contrast with previous studies using decellularized organs and differentiated neonatal cells, Yang’s group had to control differentiation of the MCP cells within the scaffold. “When we cultured the MCP cells, we found that if we added different growth factors, we could induce these MCP cells to more cardiomyocytes with less endothelial cells, or we could induce these MCP cells to more endothelial cells with less cardiomyocytes.”
When designing the perfusion media for moving the cells into the heart, Yang had to determine what the resulting cell population should look like after differentiation within the scaffold and balance differentiation factors accordingly. A greater proportion of cardiomyocytes would increase chances of getting a beating construct, but the heart also needed vascular and smooth muscle cells to function. But once he poured these progenitor cells into the heart shaped bag and induced differentiation, how would the cells localize to their particular niches and establish relationships with adjacent cells?
“We found most of the cardiomyocytes in the ventricles of the decellularized mouse heart, and surprisingly, we found most of the endothelial cells dividing about the endocardium layer,” Yang reported. Without manipulation beyond the perfusion media carrying differentiation factors for MCPs, the cells migrated and differentiated in their precise niches. The majority of the progenitor cells spontaneously settled and matured exactly where they belonged, and this phenomenon wasn't specific to MCPs.
Song also found that the ECM scaffold provided several advantages to the construct. “What I thought was so surprising was that these specific cells, most of them, if not the majority of them, knew exactly where to sit down. When we looked at the regenerated kidney after biomimetic culture, we found that glomerular epithelial cells actually sat at the glomerulus, that loop of Henle epithelial cells sat at the loop of Henle,” noted Song. “There's something so elegant about the ECM that it actually tells the cells where to sit.”
Although the engineered kidneys produced urine, and the hearts made of human MCP cells in a mouse scaffold beat, there is still a long way to go before solid organs are manufactured and provided to patients on demand. Both groups are currently optimizing decellularization protocols and working with larger scaffolds from pigs and humans to scale up the organs. And studies will continue to determine the best cell types to use for this procedure and how to differentiate and culture them both in a dish and in organ scaffolds.
The heart, in particular, is especially challenging since it requires an electrical conduction system to coordinate beating throughout its different regions. “For kidney or liver organ regeneration, you just put certain cells into the decellularized kidney or liver... and can claim the regeneration of that organ. But for the heart, if the construct is not electrically synchronized or controlled, we can never say we successfully generated a functional heart,” Yang said.
He is quick to add that, despite the distance from achieving personalized organ transplants, there is much to be gained from the regenerated organs available today. For example, his lab is currently using their constructs to study cell migration and interaction during heart development. He pointed out that these hearts should also be useful for clinical testing or understanding the mechanisms of human heart disease. Song agreed, emphasizing the unique ability of decellularized organs to serve in studies of the extracellular matrix.
“So definitely, tissue engineering is not only an approach for engineering tissue. It can be also used to crosstalk with other fields of biomedical science,” Yang concluded.
(1) Lu TY, Lin B, Kim J, Sullivan M, Tobita K, Salama G, Yang L. Repopulation of decellularized mouse heart with human induced pluripotent stem cell-derived cardiovascular progenitor cells. Nat Commun. 2013 Aug 14;4:2307
(2) Yang L, Soonpaa MH, Adler ED, Roepke TK, Kattman SJ, Kennedy M, Henckaerts E, Bonham K, Abbott GW, Linden RM, Field LJ, Keller GM. Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population. Nature. 2008 May 22;453(7194):524-8.
(3) Lin B, Kim J, Li Y, Pan H, Carvajal-Vergara X, Salama G, Cheng T, Li Y, Lo CW, Yang L. High-purity enrichment of functional cardiovascular cells from human iPS cells. Cardiovasc Res. 2012 Aug 1;95(3):327-35.
(4) Ott HC, Matthiesen TS, Goh SK, Black LD, Kren SM, Netoff TI, Taylor DA. Perfusion-decellularized matrix: using nature's platform to engineer a bioartificial heart. Nat Med. 2008 Feb;14(2):213-21. doi: 10.1038/nm1684. Epub 2008 Jan 13.
(5) Song JJ, Guyette JP, Gilpin SE, Gonzalez G, Vacanti JP, Ott HC. Regeneration and experimental orthotopic transplantation of a bioengineered kidney. Nat Med. 2013 May;19(5):646-51.