Portal de Periódicos da CAPES

Bioengineering the artificial liver with non‐hepatic cells: Where are we headed?

2009; Wiley; Volume: 24; Issue: 2 Linguagem: Inglês

10.1111/j.1440-1746.2008.05711.x

1440-1746

Yock Young Dan,

Organ Transplantation Techniques and Outcomes

Human hepatocytes have been in great demand for a wide variety of applications. Clinically, they could potentially be useful when transplanted directly into patients with liver failure or genetic disorders,1 or when employed in bioartificial liver devices for liver dialysis,2 especially where liver transplant is not available or possible. In addition, large numbers of human hepatocytes are needed for toxicology studies for drug development in the pharmaceutical industry. The major hurdles that have limited the use of human hepatocytes, however, are the lack of a continual and reliable source of such cells and the technical difficulty in maintaining their differentiated hepatocytic functions in vitro for significant periods of time.3 Following liver injury, the human hepatocyte possesses tremendous intrinsic ability to proliferate and regenerate itself in vivo.4 However, despite the rapid advances in the understanding of liver anatomy and physiology, liver biologists have failed to recreate the microenvironment of the hepatic sinusoids in vitro. Part of this reason, to date, is because it has not been possible to proliferate human hepatocytes reliably in culture or prevent them from dedifferentiating and losing their hepatocytic functions. As a result of these limitations, there have been significant efforts in identifying and utilizing alternative sources of cells that can be expanded easily in culture, and subsequently manipulated to give rise to hepatocytes, either by directed differentiation (hepatocyte linage stem/progenitor cells) or transdifferentiation (stem cells from other lineages).5 Of all the candidate cells that have been reported to have this potential, mesenchymal stem cells (MSC) or mesenchymal lineage stem cells have come through as the strongest contender. These cells have been isolated from bone marrow,6 umbilical cord blood,7 umbilical cord matrix,8 or adipose tissue.9 They have been shown, separately, to be able to transdifferentiate into hepatocyte-like cells with specific hepatocellular functions, using both in vitro as well as in vivo experiments. As MSC are readily available, easily expanded, and are not encumbered by the ethical concerns of embryonic or fetal stem cells, they are highly attractive as a hepatocyte cell source. The efficiency of transdifferentiation, however, has been variable. The other strategy taken by liver biologists has been to optimize culture environments to closely simulate the microenvironment of the liver so as to allow optimal growth of hepatocytes or hepatocyte-like cells in culture.10 The fact that in vitro hepatocytes do not survive well and lose their hepatocytic functions in culture3 is not surprising. The 2-D culture plate, where cells spread out in a monolayer, is a far cry from the liver's 3-D hepatic sinusoids in which hepatocytes align in polar fashion, sharing canaliculi with neighboring hepatocytes on their apical sides and interacting with the basement membrane and sinusoidal endothelium on their basal surfaces. Hepatocytes do indeed grow better and better maintain their function when the culture environment is modified to simulate the physiological anatomy. Simple sandwich collagen gels11 and spheroid cultures12 provide contact on both ends, and give liver cells a sense of polarity that enhances their functions. Roller bottles with coculture techniques simulate flow within the system and allow cell–cell interaction, improving hepatocytic differentiation.13 Selective substrates with collagen or matrigel have been reported to enhance the cell–matrix interaction and are prohepatocytic.14 In more recent years, bioengineers have spun out numerous 3-D-fabricated devices using a myriad of nanofibers, hollow fibers, and gels.10 These structures combine concepts of the 3-D lattice scaffold structure with variable tension, customized substrate for enhanced cell–matrix binding, and coculture to allow cell–cell interaction, all for the purpose of augmenting hepatocytic cell functions and survival. In this issue of the Journal of Gastroenterology and Hepatology, Kazemnejad et al.15 report the combination of using both strategies: MSC transdifferentiation in a prohepatocytic 3-D nanofibrous scaffold. Their aim was to achieve more efficient transdifferentiation of hepatocytes and better expression of their differentiation markers. Using bioengineering advances in polymer chemistry, they spun nanofibers into a 3-D scaffold from poly εcaprolactone incorporated with collagen and polyethersulfone. Such 3-D scaffolds are biocompatible, mimic the structure and biological functions of a native extracellular matrix, and have better mechanical strength and less variability when compared to natural polymer scaffolds. Using MSC from bone marrow and comparing this culture technique with conventional 2-D cultures, the authors found almost twofold more albumin- and α-fetoprotein (AFP)-positive cells in the 3-D culture scaffolds. Albumin and urea production were also enhanced approximately twofold. The net transdifferentiation efficiency was just under 50%. The authors surmised that the nanofibers incorporated with collagen and customizable pore size allowed for better cell adhesion and proliferation, as well as the slow diffusion of micronutrients critical for successful transdifferentiation and the maintenance of hepatic functions. This report adds to the increasing literature with respect to the reproducibility of MSC transdifferentiation into hepatocyte-like cells. In addition, it supports the superiority of bioengineered 3-D culture systems over conventional 2-D culture systems in supporting transdifferentiation and hepatocyte growth and function. However, two critical questions still beg answers: (i) can MSC reliably produce cells that are sufficiently useful for the intended purpose of hepatocytes? and (ii) how exactly do 3-D culture systems augment the transdifferentiation efficiency and viability of hepatocytes in culture? As reports of successful transdifferentiation of various mesenchymal lineage cells into hepatocytes accumulate, the weight of evidence suggests that this is a true reproducible phenomenon rather than an artefactual gimmick.5 Lending credence to the practicality of this approach is that even intrinsic liver progenitor cells—fetal liver as well as ‘oval cells’ in animal liver injury models—have been shown to carry mixed endodermal and mesenchymal markers. The mesenchymal–epithelial transition appears to be a characteristic feature of hepatocyte differentiation from lineage progenitor cells.16 If this was a physiological phenomenon, it is then not surprising that mesenchymal cells can be readily converted to hepatocytes. After all, the fetal liver originates from both the endoderm and mesenchyme in close interaction and transition during the fetal hematopoietic stage. If we accept this as a natural physiological pathway, then the emphasis shifts to how we can optimize the efficiency of transdifferentiation and the quality of hepatocyte-like cells. On the other hand, skeptics continue to raise the bar in their demand of rigorous proof of complete meaningful trandifferentiation.17 While individual or even groups of liver genes, message RNA, and proteins can be turned on with appropriate growth factors and culture stimuli, do hepatocyte-like cells have all the characteristics of a mature human hepatocytes? Neither albumin, AFP, or Cyp3A4 alone are specific for hepatocytes.17 Switching on the albumin gene in a cell with a transcription factor promoter does not make the cell a hepatocyte. Semantics aside, if the aim is to use hepatocytes for clinicotherapeutic applications, then the gold standard test would be whether these derived hepatocyte-like cells are able to show stable and reproducible functions that are comparable to normal adult hepatocytes for the specific application intended. For cellular transplant and bioartificial liver dialysis, hepatocytes will need to demonstrate ammonia and bilirubin metabolism, as well as clotting factor VII and albumin production equivalent to levels of fresh hepatocytes. For the purposes of drug metabolism evaluation, useful cytochrome P450 functional activity and inducibility to a degree similar to fresh adult hepatocytes will be the threshold requirement. Critical to our aims of using these cells outside the liver is the need to engineer an optimal in vitro environment that is prohepatocytic and that will augment hepatocyte derivation efficiently and maintain hepatocytes reliably. Of all the innumerable culture systems reported, which is the best 3-D culture system? Traditional culture models have evolved along a trial-and-error system of using what is available and reasonable and comparing it to the Petri dish. The next phase of research will need to focus on determining the critical elements in the culture system that are optimal for maintaining a hepatocytic phenotype. Is it just an issue of polarity that allows 3-D cell–matrix contact? That is, perhaps a collagen sandwich gel may perform similarly well compared to an expensive electrospun nanofiber scaffold. Is it the physical scaffold meshwork support that is critical, a concept analogous to marine biologists using tyres in the sea to encourage coral growth regardless of material? Which of the natural polymers, such as collagen, laminin, and fibronectin, are most optimal for maintaining the hepatocyte phenotype and how do they dictate the cell functions? What is the role of integrin contact signaling, or is it merely a paracrine effect of the growth factors diffusing through the porous surfaces on multiple sides of the cell? The potential factors are innumerable, ranging from oxygen tension, flow mechanics, and elasticity of lattice support. Carefully designed studies that study each of these factors by focusing on their specific role in accumulative fashion are now needed. Advances in bioengineering now allow us to customize culture models to replicate the liver microenvironment to support hepatocytes in vitro.10 Work on induced pluripotent cells and induced adult cells highlight the fact that a cell is but an expression of its phenotype, and can be in turn determined by the selective expression of a few genes.18 Understanding the controllers of hepatocytic genes and the ability to turn on the necessary ones by just microenvironment manipulation in the culture system will surely shorten the translation gap to use hepatocytes to clinical benefit. Will the day come when liver patients can be cured by fresh, newly-grown hepatocytes derived from their bone marrow mesenchymal cells? The challenge for the scientific community is not to come out with more combination systems, each better than the Petri dish, but a collective effort to decipher the secrets of the microenvironment through bioengineering models. In that light, Kazemnejad et al.'s15 report puts into perspective the dream and the long road ahead.