| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online August 17, 2005 as doi:10.1096/fj.04-3269fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Center for Life Science and Technology, School of Fundamental Science and Technology, Keio University, Yokohama, Japan; and
Department of Pathophysiology, Cancer Research Institute, Sapporo Medical University School of Medicine, Sapporo, Japan
1 Correspondence: Center for Life Science and Technology, School of Fundamental Science and Technology, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, 223-8522, Japan. E-mail: rsudo{at}2000.jukuin.keio.ac.jp
SPECIFIC AIMS
In this study we developed a simple 3-dimensional (3D) culture method that mimics the structure of hepatic cords by stacking up 2-dimensional (2D) tissues of rat small hepatocytes (SHs), which are hepatic progenitor cells. To evaluate the organization of the cells in 3D stacked-up structures, we ascertained whether the cells adhered to each other, formed bile canaliculi (BC) in the attached portions, and increased their liver-specific functions.
PRINCIPAL FINDINGS
1. 3D stacked-up culture of SHs
For the reconstruction of 3D hepatic organoids it is important to organize the cells into a tissue resembling liver architecture. Although it seems clear that 3D culture of hepatocytes must maintain hepatic differentiated functions, there is no feasible method to reconstruct 3D tissues. We therefore developed a simple 3D culture technique. The schematic diagram shown in Fig. 1
summarizes 3D stacked-up structures produced by SHs and nonparenchymal cells (NPCs). In the present experiment SHs were isolated from adult rat livers. Pairs of microporous membranes were prepared and the cells were separately cultured on each membrane (Fig. 1A
). SHs could proliferate and form colonies on the membranes, and the colonies gradually developed over time in culture. NPCs such as stellate cells also proliferated and surrounded the colonies. Then the stellate cells translocated beneath the colonies and some cells infiltrated under the membranes through the micropores. After SHs were cultured on the membranes for 
20 days and allowed to develop, one membrane was inverted on top of the other to form an SH bilayer. Thus, the 3D stacked-up structures could be reconstructed.
|
2. Cell adhesion and bile canalicular formation in the 3D stacked-up structures
To investigate the organization of the cells after stacking, bile canalicular formation in the 3D stacked-up structures was examined using fluorescein diacetate (FD). FD is metabolized in hepatocytes, and becomes fluorescein when secreted into BC. FD treatment revealed that BC formation started 3 days after stacking. Short BC then formed between the cells in the 3D stacked-up structures. The BC gradually elongated with the culture time and developed into anastomosing networks within 2 wk (Fig. 1B
).
For the reconstruction of well-assembled 3D tissues, BC formation needs to occur in the attached portions between the cells of the upper and lower layers. We therefore labeled the cells of each layer with DiI and GFP to distinguish them in the 3D stacked-up structures. Fluorescent microscopy revealed that BC formed in the 3D stacked-up structures in which GFP-labeled cells were stacked onto the DiI-labeled cells, whereas no bile canalicular formation was identified in regions when both sides of the cells remained unattached. Although it is possible that the BC formation occurred in the region between the neighboring cells on each side, this was difficult to confirm using the labeling method described.
We then analyzed vertical sections of the 3D stacked-up structures using transmission electron microscopy. Perpendicular sections of the stacked cells showed that the hepatocytes of the upper and lower layers adhered to each other and formed a bilayer that was sandwiched between the membranes (Fig. 2
A, C). Between the cells of the upper and lower layers, a well-developed bile canaliculus was observed that formed a tubular structure (arrows, Fig. 2A
). As shown in the enlarged image in Fig. 2B
, the BC was 1 µm in diameter and possessed many microvilli. Cell-cell junctions, such as tight junctions and desmosomes, were observed around the tubular structures (Fig. 2B, D
).
|
3. Repolarization of the cells in the 3D stacked-up structures
Membrane polarity of SHs in the 3D stacked-up structures was also examined. Double fluorescent immunocytochemistry was carried out for actin and bile canalicular proteins, such as ectoATPase, multidrug resistance-associated protein 2 (MRP2), and 5'-nucleotidase (Fig. 3
). The results show that these proteins localized on the lumina of BC within the 3D stacked-up structures. Actin filaments were abundant on the outer surface of the tubular structures. These results indicate that bile canalicular proteins localized on the extracellular side of the apical membrane and that actin filaments accumulated beneath the apical membrane. Therefore, hepatocytes in the 3D stacked-up structures might acquire membrane polarity. Our findings indicate that the reconstructed BC were morphologically and functionally similar to BC in vivo.
|
4. Albumin secretions and gene expression of the hepatic differentiation markers
Hepatic differentiated functions of the cells were examined as well as morphogenesis after stacking. The albumin secretions produced by the cells before and after stacking were measured using an ELISA. When the cells were stacked up, the amount of albumin secreted into the medium rapidly increased within 2 days and remained high for >2 wk. By contrast, the level of albumin secretion gradually diminished in the cells that were cultured on a membrane without stacking. We next examined whether the cells within the 3D stacked-up structures could express other differentiation markers. Reverse transcription-polymerase chain reaction analysis revealed that the mRNA expression of hepatocyte nuclear factor 4 (HNF-4) and tyrosine aminotransferase (TAT) were stronger in stacked cells than those in cells not stacked, though both cells expressed albumin and MRP2 mRNAs. mRNA expression of a highly differentiated marker, tryptophan-2, 3-dioxygenase (TO) in stacked cells was much stronger than those in cells that were not stacked.
CONCLUSIONS AND SIGNIFICANCE
In the field of tissue engineering of the liver, it is still unclear how we can reconstruct a complicated tissue maintaining various functions and how hepatocytes can proliferate and maintain their differentiated functions in vitro. Although primary hepatocytes rapidly lose their liver-specific functions in vitro, SHs can proliferate on the membrane and differentiate into mature hepatocytes (MHs). SHs are therefore considered suitable for the reconstruction of functional tissues. Tissue engineering is needed to reconstruct well-organized 3D tissues. Although 3D culture methods, such as hepatocyte spheroids and cultures in 3D scaffolds, have been reported, the formation of the tissues depends on the spontaneous behavior of the cells. Hepatocytes in vivo construct hepatic cords; organization of the cells is achieved by the interaction of MHs, NPCs, and extracellular matrix (ECM) whereas spheroids are simple aggregates composed of MHs alone. Considering the complex structure of the liver, it might be difficult to reconstruct well-assembled tissue depending on the spontaneous behavior. Therefore, precise manipulation of the cells is necessary to facilitate tissue organization. In the present experiment we created a simple 3D culture method that could reconstruct 3D stacked-up tissues from 2D structures. Within these structures, the SHs could organize themselves into orchestrated tissues with liver-specific functions. These results suggest that the combination of SHs and the stepwise progression from 2D to 3D is important for the reconstruction of well-organized hepatic organoids.
The structure of the normal liver is composed of continuous stacked cell layers (Fig. 1B
). Hepatocytes form the hepatic cords, which are interconnected lines of cells. The strategy of the 3D stacked-up culture was to mimic these structures. In the present experiment we reconstructed 3D stacked-up structures in which the cells physiologically adhered and formed functional BC. These structures mimicked those of the hepatic cords. The cells could extend 3-dimensionally while maintaining both cell-cell and cell-ECM contacts. When SHs were sparsely inoculated onto collagen-coated dishes, they only attached onto the surface of the dish and did not maintain their cell-cell interactions. As a result, the cells extended 2-dimensionally and the cells became flat. By contrast, SHs in the 3D stacked-up structures could maintain both vertical and horizontal cell-cell interactions, as well as cell-ECM interactions. In vivo, hepatocytes seem to maintain their liver-specific function by forming 3D complex architectures interacting with NPCs and the ECM. Such microarchitectures might enhance the differentiation of SHs through maintenance of the cuboidal cell shape. In the present experiment, albumin secretion by the cells in the 3D stacked-up structures was enhanced compared with that in the cells on the membranes. In addition, the microarchitecture in the 3D stacked-up structures might also enhance the synthesis of liver-specific mRNAs, such as MRP2, HNF-4, TAT, and TO. Thus, the SHs achieved liver-specific functions by assembling into an appropriate configuration. The 3D stacked-up structures are not only similar to the structures of the hepatic lobules but also maintain their differentiated functions.
Liver transplantation is the only treatment for end-stage liver diseases, but donor organ shortages are a serious issue. Therefore, many researchers have attempted to restore liver function using bioartificial liver (BAL). Various cell types have been examined for use in the BAL reactor, but it is extremely difficult to maintain hepatocyte function in vitro. Therefore, tissue-engineered livers can be an alternative source for the BAL. In the present experiment, the cells were stacked onto those cultured on the membrane, which suggests the potential for multiple stacking; although only two membranes were stacked in the present experiment, three were successfully stacked in the preliminary study. Although further improvements will be necessary to achieve thicker complex tissues, the 3D stacked-up culture described here will be useful in the construction of tissue-engineered livers.
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-3269fje; doi: 10.1096/fj.04-3269fje
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
