You are here

Bioengineering Our Way towards Organ Replacement Therapy



Several recent papers have demonstrated how whole organs can be de-cellularised by perfusion with a series of detergents, leaving behind only the underlying extracellular matrix. The residual acellular (and importantly non immunogenic) environment maintains its natural architecture, structural and functional characteristics, and acts as an entirely natural scaffold into which healthy replacement cells can be re-seeded in order to regenerate the organ. In 2008, a replacement human trachea made in a similar way and seeded with autologous adult cells was successfully transplanted for the first time in a patient with severe bronhomalacia (Macchiarini et al.). Researchers are now tackling the problem of more complex whole organ reconstitution, an aim that will no doubt be aided by stem cell research.

Work done by Doris Taylor and colleagues in Minnesota demonstrated a method for the de-cellularisation of adult rat hearts which maintains intact valves and chamber geometry. It was hoped that subsequent seeding of these scaffolds with neonatal cardiac or aortic endothelial cells under simulated physiological conditions would stimulate cellular maturation (Ott et al.). Although macroscopic contractions were detected after four days, these reconstituted hearts were only able to generate pump function equivalent to 2% of normal adult levels. One major hurdle in the development of complex tissue engineered organs in vitro thus far is the very short distance over which oxygen and nutrients can diffuse, particularly within larger scaffolds. This limits the rate of neovascularisation to levels insufficient to prevent the onset of necrosis within the transplanted tissue, thus compromising cells positioned deep within the scaffold.

Improving on this approach, a recently published article in Nature Medicine by Korkut Uygun et al. (from the Center for Engineering in Medicine at Massachusetts General Hospital) describes successful de-cellularisation of rat liver combined with cellular reconstitution and transplantation of one lobe. They report good preservation of the liver microvasculature network on a functional level, enabling efficient re-cellularisation; 90% of seeded adult hepatocyte cells successfully engrafted into the liver lobe scaffold matrix, representing 50% of normal lobe cellularity (or 20% of global rat liver reconstitution, where 10% restoration is required for therapeutic functional improvement in a clinical setting). Moreover the group were able to recreate liver vessel vascular linings via introduction of endothelial cells, thus improving the survival of transfused hepatocytes. This is essential for transplanted hepatocyte survival, since this cell type has a particularly high oxygen demand, necessitating both arterial and venous blood supply to the liver in vivo. Following lobe transplantation alongside host liver in rat, minimal ischemia and good preservation of the deep 3D vascular network was observed over the 8 hour study period. Re-cellularised liver also supported some liver-specific functions in vivo including urea synthesis, cytochrome P450 expression and albumin secretion, albeit at reduced capacity (although the group reported the latter at approximately 15-25% below normal production levels), and did not test function in vivo for greater than 8 hours, or ex vivo for greater than 24 hours.

In a similar vein, another paper published recently in Science by Petersen et al. (from Yale University) utilised the architecture of de-cellularised rat lung (largely consisting of collagen) to demonstrate the possibility of re-perfusion of multiple cell types, in this case up to nine, followed by culture in a specialised biomimetic bioreactor. Vascular perfusion with culture medium resulted in enhanced endothelial cell adhesion and survival in reconstituted lungs, and air ventilation acting to simulate breathing resulted in enhanced survival of lung epithelium in distal alveoli. Additionally, some normal lung function was restored, including clearance of cell secretions from the airway tree, proliferation of type I alveolar and ciliated columnal epithelial cells, and production of developmental (but not adult) pro-surfactant proteins. Although reportedly difficult to culture lung cells in vitro, cells within reconstituted and vascularly perfused lungs replicated rapidly, rarely apoptosed and importantly, the residual matrix and vascular compartment niche orchestrated the specific and directed integration of cells within the scaffold. This impressive and unanticipated hierarchical cellular organisation thus highlights the importance of substrate cues from the de-cellularised matrix for the reformation of ‘normal’ anatomical architecture in reconstituted organs, and also the key importance of cue-responsive adult stem cells and precursor cells in the re-cellularised niche. Further analysis revealed normal cell surface marker expression, tight junctions (consistent with those required for critical blood air barrier function) and by day 4 a good distribution of small airway and type II alveolar epithelial cells. Consistent with a developmental paradigm for cell redistribution and growth in these engineered organs, the authors noted that aquaporin-5 expression, a marker of type I alveoli cells, was higher in lungs exposed to simulated breathing alongside de-novo formation of mesenchymal and airway epithelial progenitor basal cells derived from the transplanted inoculum, possibly generated from type I epithelial cells.

Moreover, upon transplantation into rats re-cellularised lung showed similar mechanical characteristics to native tissue (permitting ventilation at physiological pressures) and was able to perform basic gas exchange for short periods of time (45-120 minutes) as evidenced by visible changes in blood colour as a result of oxygenation. Haemoglobin was 100% saturated following oxygenation in both the native and engineered lung whilst CO2 decreased approximately 4 fold in engineered tissue. Although transplants were not assessed for in vivo function for longer than two hours, this study represents the first attempt to transplant bioengineered lung tissue into a living organism and is an important proof of principle that such technologies could one day be adapted to the clinic; of particular relevance in this organ system given that lung tissue regenerates so poorly. Encouragingly, the same group went further and performed similar experiments, with relative success, using portions of human cadaveric lung tissue repopulated with human epithelial carcinoma cells and endothelial cells derived from human cord blood endothelial progenitor cells.

Although both the liver and lung studies show that organs bio-engineered in this way can so far perform with limited functionality, they represent an important initial step and provide vital information toward the generation of fully functional organs in vitro. Furthermore it suggests that the use of this type of graft might be a viable strategy for future regenerative medicine purposes, particularly when combined with the potency of endothelial progenitor cells derived from cord blood. Also, it is not insignificant that this type of whole organ regeneration can make use of a huge resource of otherwise unsuitable donors (e.g. those who died from heart disease) and can be used not only in the potential creation of ‘replacement parts’ for those with organ failure, but also as a form of temporary support similar to, for example, kidney dialysis.

It is clear that many obstacles must be overcome before complex organ grafts can be transferred to the clinic. One such barrier is achieving an anatomically and functionally correct distribution of multiple cell types within the organ scaffold, a problem highlighted by the aberrant organisation of newly perfused cells within de-cellularised liver matrices as observed by Uygun et al. However, as Niklason’s group indicate from their studies in lung, a previously unappreciated level of stem/progenitor cell self-organisation may exist given the correct microenvironmental cues. Another issue is the inherent difficulty in connecting potentially delicate or underdeveloped bio-engineered organs with a normal vascular system operating within physiological parameters. In the case of re-cellularised lung, some bleeding into pulmonary airways and clotting upon transplantation was observed indicating inefficient barrier function and an insufficient re-covering with freshly perfused cells of basement membranes exposed during the de-cellularisation process. The authors suggest that this problem might be overcome given successive iterative improvements to the de-cellularisation protocol, aimed at preserving more intact matrix architecture, followed by longer in vitro culture periods to allow sufficient time for perfused cells to proliferate, migrate, differentiate and integrate more appropriately.

Finally, and crucially, these strategies for organ regeneration can only become viable for clinical purposes when suitable alternative sources of safe, scalable and patient compatible cells become available. In many cases, for example in the elderly or critically ill patients, autologous adult stem cells may be unavailable, grow poorly or be too damaged to utilise for re-cellularisation. It is hoped that human embryonic stem cells (hESCs) and their patient-specific induced pluripotent counterparts (iPSCs) can play a substantial role in providing various types of normal functional cell types for organ reconstruction. Indeed a huge variety of tissue-specific cell types derived from hESCs and iPSCs, including heart (Xu et al.), lung (Van Haute L. et al.) and liver (Lavon, N., et al.), have been reported in the literature. Substantial doubts remain, however, both in terms of the difficulty in driving the differentiation of these cells towards fully mature functional cells alongside the risk of tumour formation following transplantation. Notwithstanding, these studies contribute towards a sound scientific and technological foundation, which coupled with ongoing advances in stem cell based technologies might finally see the aspirations of the original “Immortalists” Lindberg and Carrel come to fruition.



Macchiarini, P. et al., Clinical transplantation of a tissue-engineered airway. Lancet 372, 2023–2030 (2008).

Ott, H.C. et al., Perfusion-decellularized matrix: using nature’s platform to engineer a bioartificial heart. Nat. Med. 14, 213–221 (2008).

Uygun K. et al., Organ reengineering through development of a transplantable recellularized liver graft using decellularized liver matrix. Nat. Med. Jul;16(7):814-20 (2010)

Petersen, T.H. et al., Tissue-Engineered Lungs for in Vivo Implantation Science. 24 June (2010)

Xu, C., et al., Characterization and enrichment of cardiomyocytes derived from human embryonic stem cells. Circ Res. 91(6): p. 501-8. (2002)

Van Haute L. et al., Generation of lung epithelial-like tissue from human embryonic stem cells. Respir Res. Nov 5;10:105. (2009)

Lavon, N., et al., Differentiation and isolation of hepatic-like cells from human embryonic stem cells. Differentiation. 72(5): p. 230-8. (2004)