Research progress in liver tissue engineering

1. Introduction

As we know, liver plays a crucial role in biliary secretion, detoxification, synthesis of proteins, glycogen storage. Liver diseases at end-stage will result to liver failure, which is lethal to patients. The definitive treatment for end-stage liver failure is liver transplantation. However, the demand for transplantation far exceeds the number of available donor organs. Organ transplantation is a highly successful method of treating organ failure, but its application is severely limited by an inadequate supply of suitable donor organs.

In 2013 in Australia and New Zealand, 1221 transplants were performed, but 1458 patients remained on transplant waiting lists at the end of the year. In China, there are about 300 thousand patients waiting for liver transplantation but only 45% of them have the chance to receive a liver transplantation. All over the world, more than 50% patients lost their chance of liver transplantation (http://columbiasurgery.org/liver/liver-transplant-waiting-list). What’s more, the patients have to take medicines all lifelong to control immune rejection.

Fortunately, a promising tissue-engineering/regenerative medicine approach for functional Organ replacement has emerged in recent years, liver tissue engineering, which tries to fabricate a new bioartificial organ with cells and scaffolds. Liver tissue engineering, which tries to generate livers in vitro, would allow an unlimited supply of organs without the need for immunosuppression and would revolutionize the treatment of organ failure. However, there are many obstacles to be overcome before the clinical application of this form of treatment. Three elements must be considered in tissue engineering: Cell source, scaffold and bioreactor, which can provide appropriate environment for cells in scaffold. They can give cells and scaffold a good biochemical and mechanical conditions.

In this concise review, we will review some recent progresses in liver tissue engineering in 4 aspects: potential cell sources, scaffolds, bioreactors and the first step application of bio-engineered artificial liver in animals.

2. Progress in cell source on liver tissue engineering

An ideal cell source would be one that can proliferate or self-renew as needed and yet give rise to the heterogeneous types of cells necessary to form a functional liver. Liver-derived cell types can be isolated, expanded in vitro, and used. However, the obstacle against their clinical application is the requirement of large number. They are not available from patients themselves and are not enough from other donors either [1]. Embryonic stem cells can be differentiated into virtually any cell lineage for tissue or organ engineering. Furthermore, ESCs are derived early after fertilization, they lack the epigenetic modifications, which means they don’t express major histocompatibility antigen (MHC) type I and type II and they can be used in allograft transplantation without immune rejection. While, the ethical debate surrounding their use is the main obstacle for their application. What’s more, the risk of teratogenic in ESCs application is not yet excluded [2,3]. Inducible pluripotent stem cells (iPS) are a type of pluripotent stem cells generated directly from adult somatic cells by introduction of four genes. iPSs hold great promise in the field of regenerative medicine because of their unlimited amplification and totipotential capacity. However, they are still in the stage of research and are not mature in their application [4].

At present, the most likely candidate to fulfill such demands is mesenchymal stem cell (MSC) derived from primary tissue or organ, like W-J MSC(Wharton’s jelly), BMSC (Bone marrow), amniotic fluid stem cell and adipose-derived stem cell (ADMSC). They are proved to be differentiated into hepatocyte-like cells [4]. Another potential cell source is induced hepatocyte by direct gene modification [5,6].

A functional bioartificial liver needs seeding cells with hepatocyte-like function, therefore, MSCs need to be differentiated before or afer their application. Three techniques reported to induce MSC into hepatocyte-like cells: co-culture with hepatocyte [7–9], MSC induced differentiation by conditional culture medium [2,5,10] or by gene modification [11–14]. Three co-culture methods, including direct, indirect and three dimensional (3D) co-culture, have been reported to induce successfully MSCs differentiation into hepatocytes.

In direct co-culture, researchers mixed fetal hepatocytes with MSC labeled with Green Fluorescent Protein and cultured them in normal medium without growth factors, and 4 weeks later, they got hepatocyte-like cells with GFP [15]. Teong et al. [11] co-cultured MSCs with AML12 liver cell line in a volvox sphere capsule helped by a high voltage electrostatic field system. The encapsulated MSCs were able to differentiate into hepatocyte-like cells and express liver cell markers including albumin, alpha feto-protein and cytokeratin 18 on day 14.

Simon et al. [9] showed BMSCs could be differentiated into hepatocyte-like cells after 24 hours co-culture with CCl₄ injured hepatocytes. While, cells that were co-cultured with healthy hepatocytes did not present signs of differentiation. They thus analyzed microvesiscles in the in the supernatant of differentiated cells medium and found that BMSCs can be differentiated into hepatocytes with these microvesicles. Their results showed that MSCs were stimulated by factors secreted by CCl₄ injured hepatocytes. Their research indicates that MSCs can be differentiated by indirect co-culture or factors and cytokines.

In 2010, Li et al. tried an indirect co-culture on MSCs, they co-cultured MSCs with hepatocytes during 2 weeks, then they injected the MSCs into liver fibrosis rats and one month later, the fibrosis in rat models was evaluated by hematoxylin and eosin (H&E) and Masson’s trichrome (MT) stainings. H&E staining showed that fibrosis area reduced significantly in all MSC, co-cultured MSC, and growth-factor-treated MSC groups. Furthermore, in comparison with MSC group, fibrosis area diminished significantly in co-cultured MSC group and growth factor treated MSC group at weeks 1, 3, and 4. It is noteworthy to find that co-cultured MSCs were as effective as growth factor treated MSCs. RT-PCR and Western blot results showed co-cultured MSC exhibited strongest AFP (α-fetal protein) expression at week 3, but it largely decreased at week 4. On the other hand, CK18 (cytokeratine) and CK19 expression improved remarkably at week 4. These findings indicate that co-cultured MSCs were staying at hepatic progenitor cell stage until week 3, and then continuously differentiated into hepatocytes or bile-ductal epithelial cells at week 4. The authors thus concluded that co-culture with hepatocytes is a better method than use of growth factor supplements considering clinical application [6].

The most interest one is the 3D co-culture of MSC with hepatocytes. Zhong et al. [7] have mixed C3H10 MSC labeled with green fluorescent protein (GFP) and hepatic progenitor cells (HPCs) with a ratio of 1:1 in an AlgiMatrixTM 3D Culture System. After 5 days co-culture in vitro, a spherical tissue was formed, while spheroids were not formed in HP14-19 mono-culture group. And 30 days later, they expressed liver specific markers. The spherical tissue was subcutaneously implanted in athymic nude mice, and a liver-like tissue was obtained after 30 days culture, which expressed liver markers like AFP, CK-18 and albumin. Their results indicated that in co-culture environments, MSCs not only promoted HPCs’ differentiation into mature hepatocytes in vitro and in vivo, but also differentiated into functional hepatocyte-like cells themselves in the 3D co-culture environments. The 3D co-culture condition may be an efficient way to form well functional hepatocyte-like cells in liver tissue engineering and alleviate the problem of hepatocytes shortage.

Another method for MSC differentiation into hepatocytes is to induce them with conditional medium which has been reported by several laboratories [2,5,10]. Our team used a two-step conditional medium to induce MSC differentiation in vitro and 28 days later, MSCs changed to a paving stone like form, periodic acid-Schiff (PAS) stain showed they had the ability to accumulate glycogen and LDL (low density lipoprotein) swallow test showed that they had the function of hepatocyte. What’s more, they expressed AFP, ALB, Globulin and they secreted urea [3].

Although MSCs can be differentiated into hepatocyte-like cells in an average time of 3–4 weeks no matter by co-culture or by treating with growth factors, the efficiency of production of induced MSCs is low. Recently, researches have shown that over-expression of lineage-specific transcription factors can directly convert terminally differentiated cells into some other lineages [11–14].

Huang et al. [14] successfully inducted functional hepatocyte-like (iHep) cells from mouse tail-tip fibroblasts by transduction of Gata4, Hnf1a and Foxa3, and inactivation of p19Arf. After 14 days culture, the cells showed characteristics of hepatocytes by expressing positively tight junction protein 1 (Tjp1) and E-cadherin at day 6. What’s more, albumin, Transthyretin (Ttr), transferrin (Trf) and CK18 were detected by gene analysis and confirmed by immunofluorescent staining in iHep cells at day 14. These inducted cells have also hepatocyte functions like glucogen storage confirmed by PAS staining, intake of acetylated low density lipoprotein labeled with the fluorescent probe 1, and indocyanine green (ICG) uptake. The authors compared also the global expression profiles among iHep cells, fibroblasts, and hepatocytes by gene chip, which showed that iHep cells had a gene map closer to hepatocytes than their original fibroblasts. Animal experience in Fah^-/-Rag2^-/- mice, mice defective in tyrosine metabolism require 2-(2-nitro-4-trifluoro-methylbenzyol)-1,3-cyclohexanedione (NTBC) supply for survival, showed that iHep cells improved the liver functions of NTBC-off Fah^-/-Rag2^-/- mice and even two months after transplantation, no tumor was found in these Fah^-/-Rag2^-/- mice liver.

Three years later, the same research team published their work in human skin fibroblasts [11]. This time, they transferred Foxa3, Hnf1a, Hnf4a genes into human fibroblasts and they successfully got iHep cells too, which had the ability of glycogen storage, LDL intake and ICG uptake. hiHeps were stained for ALB and a-1-antitrypsin (AAT) at 2 weeks after induction.

To our knowledge, there’s not yet report about induction MSC into hepatocytes by direct gene modification at present. It might be a promising method for seeding cell collection with the development of gene modification techniques.

3. Progress in scaffold on liver tissue engineering

As what mentioned before, liver tissue engineering need a proper scaffold to support and provide cells with proper microstructure. The scaffolds reported can be classified by structure as 2D and 3D scaffolds, by material as natural, like collagen and alginate, and biopolymer materials, like polylactic acid (PLA), poly(lactic-co-glycolic acid) (PLGA), and so on. In this review, we won’t get into the detail of the researches repeated on these traditional materials for a long time, but to introduce two brand new promising scaffolds published in recent years, decellulerized whole liver scafolds and poly (3-hydroxybutyrateco-3-hydroxyvalerate-co- 3-hydroxyhexanoate) PHBVHHx.

Decellularized liver keeps exactly the texture of the original organ. This natural structure can firstly provide a 3D poral matrix in favor of cultured cell proliferation, differentiation and function. Secondly, it consists primarily of type I collagen, glycosaminoglycans, fibronectin, laminin with exactly proportion as natural extracellular matrix, which are very important for keeping a diverse variety of growth factors. Thirdly, it keeps the blood vessels intact and this is helpful in oxygen supply and cell nutrition.

The liver decellularization is carried out by perfusing EDTA, SDS, Triton-X or other detergent and enzymatic products dissolved in phosphoric buffer solution (PBS) through liver portal vein. After a period of perfusion, all hepatocytes are expelled from liver, while, extracellular matrix were kept complete with blood and biliary vessels. The decellularized scaffolds were analyzed by immunohistochemistry, MT staining, scanning electronic microscope, and the DNA residue. The qualified matrix maintains the liver-specific proteins proportion of collagens I, IV, fibronectin and laminin. There should be less than 10% of DNA detected in the decellularized liver, which means almost all of cells have been driven away from the liver. The intact vascular system should be kept well and can be used in recellularization [16,17]. Furthermore, the biocompatibility and application potential of scaffolds were tested by recellularization with cells. Uygun et al. [17] have successfully recellularized this scaffold with rat hepatocytes. In our research,WJ-MSC were used for the recellularization, and a liver like new tissue is got one week later. HE and MT staining showed that it had a liver like microstructure with hepatic plate and hepatic lobule (not yet published). This technique has been applied successfully in rats [17–20], rabbits [21], porcine [22], and human being [23].

In 2014, a new material, called Poly PHBVHHx, was firstly used in tissue engineering, which is a member of polyhydroxyalkanoates (PHA) family,and drew intensive attention. PHA is produced in nature by bacteria fermentation of sugar or lipid, which gives this kind of material desirable biocompatibility and biodegradability. More than 150 different monomers can be combined within this family to give materials with extremely different properties. According to their composition, they have different ductile and more or less elastic, which gives the material proper mechanical strength as needed [24].

As the member of PHA family, PHBVHHx has the potential to be made as membrane or 3D. Scanning electronic microscope (SEM) shows that it has a honeycomb-like structure and displays different pore size. It can be prepared under different temperatures and therefore we will get different pore sizes. When it is prepared under −20°C, pore size will be 110–170 μm, under −80°C, 30–60 μm pore size, and −196°C contained, 20–40 μm, respectively. It has been planted with MSC into trachloride carbon-induced liver failure rat. 28 days after implantation, rat’s liver has a much better function than control group liver [25]. PHBVHHx may be a good potential material and maybe combined with Three-dimensional printing (3D printing).

3D printing is a different approach that marries biological and modern manufacturing technique, in which cells and materials are deposited precisely in a series of layers that gradually build up to form a tissue or potentially an organ. Although this is an exciting development, many effects need to be optimized and considered.

A typical process for 3D bioprinting includes: Imaging of the damaged tissue and its environment can be used to guide the design of bioprinted tissues. Biomimetic, tissue self-assembly and mini-tissue building blocks are designing approaches used singly and in combination. The choice of materials and cell source is essential and specific to the tissue form and function. Common materials include synthetic or natural polymers. Cell sources may be allogeneic or autologous. These components have to integrate with bioprinting systems such as inkjet, micro-extrusion or laser-assisted printers. Some tissues may require a period of maturation in a bioreactor before transplantation. So far, the following tissues have been printed in laboratories: aortic valve, vasculature, trachea, skin and cartilage [26–28].

4. Progress in bioreactor in liver tissue engineering

Up to now, there are mainly two sorts of bioreactors used in liver tissue engineering: parallel micro-fluidic platform and 3D perfusion system.

For parallel microfluidic platform, the device consists of several circular chambers, each accessible via a separate fluidic channel with an individually addressable inlet and outlet, realized by a computer controlling peristaltic pump. In circular chambers, scaffolds are fabricated by polydimethylsiloxane using conventional soft lithography and replica molding. Although it is a model a little far away from liver tissue engineering, it’s a very good model for toxic analysis [29].

The 3D scaffold perfusion system contents usually a circulating perfusion system consisting of a peristaltic pump, bubble trap, oxygenator and a computer for the monitoring process. The culture medium was continuously perfused through the portal vein of bioartificial liver, meanwhile, the concentration of oxygen in medium and pressure in portal vein were monitored continuously by a computer [30,31]. Buhler et al. applied the system to a decellularized porcine liver graft with human primary hepatocyte. After 48 hour perfusing culture, some areas of the liver had an orange colour in contrast to the pink medium. H&E staining of different regions showed positive nuclear staining of hepatocytes organized in three-dimensional cell complexes [22].

Our team designed a similar system with two pump to control the flow of culture medium and to keep a real circulation in decellularized whole liver scaffold with WJ-MSCs. After 14 days perfusing culture with conditional medium, we got a liver like “organ”. HE and MT staining showed that it had a liver microstructure with hepatic plate, vessels, and cell form (not yet published).

Liver tissue engineering is a promising approach to overcome the shortage of transplantation organs, nevertheless, there’s still a long way to go in seeding cell choice, scaffold fabrication and cultivation system in vitro.