Differentiation of bone marrow‐derived mesenchymal stem cells into hepatocyte‐like cells in an alginate scaffold

Abstract

Objectives:  Alginate scaffolds are the most frequently investigated biomaterials in tissue engineering. Tissue engineering techniques that generate liver tissue have become important for treatment of a number of liver diseases and recent studies indicate that bone marrow‐derived stem cells (BMSCs) can differentiate into hepatocyte‐like cells. The goal of the study described here, was to examine in vitro hepatic differentiation potential of BMSCs cultured in an alginate scaffold.

Materials and methods:  To investigate the potential of BMSCs to differentiate into hepatocyte‐like cells, we cultured BMSCs in alginate scaffolds in the presence of specific growth factors including hepatocyte growth factor, epidermal growth factor and fibroblast growth factor‐4.

Results:  We can demonstrate that alginate scaffolds are compatible for growth of BMSCs and when cultured in alginate scaffolds for several days they display several liver‐specific markers and functions. Specifically, they expressed genes encoding alpha‐foetoprotein, albumin (ALB), connexin 32 and CYP7A1. In addition, these BMSCs produced both ALB and urea, expressed cytokeratin‐18 (CK‐18) and were capable of glycogen storage. Percentage of CK‐18 positive cells, a marker of hepatocytes, was 56.7%.

Conclusions:  Our three‐dimensional alginate scaffolds were highly biocompatible with BMSCs. Furthermore, culturing induced their differentiation into hepatocyte‐like cells. Therefore, BMSCs cultured in alginate scaffolds may be applicable for hepatic tissue engineering.

Introduction

Currently, liver transplantation is a well‐established and successful procedure used to treat diseases that had lead to liver cirrhosis, and consequently liver failure during end‐stage disease. Despite the therapeutic potential, liver transplantation is limited because of a significant shortage of liver donors, high cost of the surgical procedure and need for life‐long immunosuppressive therapy, which comes with its own specific risks. As a result, hepatic tissue engineering, which involves use of cells and porous biomaterial scaffolds, would be an innovative way to construct an implantable liver.

Bone marrow contains a population of multipotent cells, known as bone marrow‐derived mesenchymal stem cells (BMSCs), which are capable of differentiating into a number of different cell types of mesoderm lineages including adipocytes, osteoblasts and other mesodermal cells (1, 2). Thus, BMSCs have been considered as an important source of stem cells for use in cell therapy and recently, they have been shown to be capable of differentiating into hepatocyte‐like cells (3). Study of the hepatic differentiation potential of BMSCs is important from the perspective of clinical application of these cells, as well as from a basic science point of view. There is urgent need for an adequate supply of human hepatocytes for transplantation and for use in artificial liver devices. In addition, understanding of the biology of BMSCs is important for basic research, such as into mechanisms that would lead to differentiation of hepatocyte‐like cells.

Differentiation of BMSCs into hepatocyte‐like cells in a three‐dimensional scaffold can improve and broaden application of BMSCs in hepatic tissue engineering. In the past, various scaffolds derived from synthetic and natural materials have been used as matrices for cells in various tissue engineering applications. These scaffolds not only provide the means for imparting a three‐dimensional structure, but also provide mechanical integrity for tissue construction (3, 4). Alginate is one of the most frequently investigated biomaterials used in tissue engineering. It is a linear, unbranched, polysaccharide composed of 1,4‐linked b‐d‐mannuronic acid (M‐block) and a‐l‐guluronic acid (G‐block); it is found in varying compositions and sequences. Gel formation is achieved through exchange of sodium ions from the G‐block, with divalent cations such as calcium; stacking of G‐blocks forms egg‐box structures (5). In addition, alginate is biocompatible, hydrophilic and biodegradable under normal physiological conditions. These unique properties make it an attractive candidate for use as matrix material in sustained release applications and for implantation and encapsulation of isolated cells (6, 7).

Previously, we have demonstrated that alginate scaffolds are biocompatible with hepatocytes (8) and here we use an alginate scaffold as matrix for differentiation of BMSCs into hepatocyte‐like cells. Furthermore, we evaluate the use of the scaffold as matrix for hepatic tissue engineering.

Materials and methods

Isolation and cultivation of BMSCs

BMSCs were harvested from bone marrow of femurs and tibias of 2‐ to 3‐month‐old male Sprague–Dawley rats (obtained from the Laboratory Animal Unit of Sun Yat‐Sen University). To collect the bone marrow, a 21‐gauge needle was inserted into the shaft of the bone and it was then flushed with 30 ml of Dulbecco’s modified Eagle’s medium (DMEM). The cell suspension was centrifuged over a Ficoll step gradient of 1.077 g/ml density (Ficoll‐Histopaque 1077; Sigma, St Louis, MO, USA) at 1500 rpm (1800 g) for 10 min. Mononuclear cells at the gradient interface were collected and resuspended in DMEM supplemented with 10% foetal bovine serum (FBS). Isolated cells from each rat were seeded individually into 25 cm² flasks and cultured at 37 °C in 5% CO₂ for 3 days. After removing non‐adherent cells, adherent ones were grown to 90% confluence, harvested using 0.25% trypsin–ethylenediaminetetraacetic acid (EDTA) and seeded into 75 cm² flasks at a density of 1 × 10⁵ cells per flask for the first passage. To minimize variability, only cells from the sixth passage were used for all hepatic differentiation studies.

Preparation of alginate scaffolds

Alginate scaffolds, 18 mm in diameter and 2 mm thick, were fabricated with 2% alginate using a freeze‐dry technique and cross‐linked using calcium. Scaffolds were sterilized with low‐temperature sterilization.

Seeding of BMSCs into alginate scaffolds

Dried and sterilized alginate scaffolds were placed in 24‐well tissue culture plates. BMSCs (1 × 10⁵) in 100 μl culture medium were added to each scaffold. After 30 min, fresh DMEM supplemented with 10% FBS was added to each well. Cultures were incubated at 37 °C in 5% CO₂. Cell viability and proliferation were monitored using optical microscopy.

Observation of cells and scaffolds by scanning electron microscopy

Morphology of both the alginate scaffold and the cells cultured in the scaffold was observed using a JSM 5410LV instrument (JEOL, Tokyo, Japan). Scaffolds along with BMSCs were frozen to −18 °C and freeze‐dried. Samples were then cut into cross‐sections using a sharp scalpel. Cross‐sections were mounted on to an aluminium stub and coated with gold/palladium using a JEOL JFC‐110E ion sputter system.

Differentiation of BMSCs into hepatocyte‐like cells within the alginate scaffolds

To induce hepatic differentiation, BMSCs in the scaffolds were cultured in basal medium consisting of Iscove’s modified Dulbecco’s medium (IMDM; Gibco, Grand Island, NY, USA) supplemented with 10% FBS (Sigma), 50 mg/ml of an insulin–transferrin–selenium premix (ITS+, Becton Dickinson, San Jose, CA, USA), 0.5 mm dexamethasone (Sigma), 0.61 g/l of nicotinamide (Sigma), 0.2 mm ascorbic acid 2‐phosphate, 10 ng/ml epidermal growth factor (EGF), 20 ng/ml hepatocyte growth factor (HGF) and 10 ng/ml fibroblast growth factor‐4 (FGF‐4). Cells were fed every 2 days. After 7, 14, 21 and 28 days of culture, aliquots of their media were collected and stored at −80 °C to measure levels of urea and albumin (ALB).

Semi‐quantitative reverse transcriptase‐polymerase chain reaction for analysis of albumin, alpha‐foetoprotein, connexin 32 and CYP7A1 gene expression

RNA was extracted from the cells using an RNeasy kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s protocol. DNase I (Qiagen) digestion was performed to minimize contamination with genomic DNA. A Sensiscript RT kit (Qiagen) and oligo dT12–18 primers (Invitrogen, Carlsbad, CA, USA) were used to reverse transcribe the total RNA into cDNA. Semi‐quantitative PCR was performed using 0.2 mg of cDNA per 25 μl reaction volume using a one‐step reverse transcriptase‐polymerase chain reaction (RT‐PCR) kit (Qiagen). Final products were separated by electrophoresis on a 2% agarose gel and stained with ethidium bromide. Primers and product sizes are listed in Table 1.

Table 1.

 Primers used for PCR

Gene	Sequence	Products (bp)
ALB	Forward Primer: 5′‐AAGCTTCGCGACAACTAGGT‐3′	408
	Reverse Primer: 5′‐CCAGGCTTTGAAGGCTCTCTC T‐3′
AFP	Forward Primer: 5′‐GGCCGACATTTACATTGGACA C‐3′	287
	Reverse Primer: 5′‐GAGGCCAGAGAAATCAGCAGT G‐3′
Cox 32	Forward Primer: 5′‐TGG TCA AGT GTG AGG CCT TC‐3′	365
	Reverse Primer: 5′‐TGAGCATCGGTCGCTCTTCT‐3′
CYP7A1	Forward Primer: 5′‐CCTCCTGGCCTTCCTAAATC‐3′	351
	Reverse Primer: 5′‐GTCAAAGGTGGAGAGCGTGT‐3′

Western blot analysis

Western blot analyses were performed as described previously (9). After washing, membranes were blocked with 10% skimmed milk at room temperature for 1 h. Membranes were then incubated with mouse monoclonal antibody against alpha‐foetoprotein (AFP, 1:200 dilution; Zymed, South San Francisco, CA, USA), rabbit anti‐rat albumin polyclonal antibody (1:200 dilution; Dako, Carpinteria, CA, USA), and goat anti‐β‐actin polyclonal antibody (1:1000 dilution; Zymed) at 4 °C overnight. After washing, membranes were incubated with horseradish peroxidase‐conjugated anti‐rabbit, anti‐mouse and anti‐goat antibodies (1:5000 dilution; Zymed) at room temperature for 1 h. After washing, immunoreactive bands were detected using ECL chemiluminescence reagents (Amersham Bioscience, Arlington Heights, IL, USA). Density of bands was quantified using a laser densitometer (ATTO densitograph 4.0, Tokyo, Japan).

Urea assay

Concentration of urea in culture media was measured using a colorimetric assay (640‐1; Sigma) according to the manufacturer’s instructions. Dilutions of stock solution of urea (Sigma) were used to create a standard concentration curve. Media samples from three separate cultures were analysed in triplicate for each condition. Absorbance of basal medium was subtracted from absorbance of each test sample to obtain a final absorbance value to determine concentration of urea using the standard curve.

Albumin assay

Enzyme‐linked immunosorbent assay (ELISA) Quantitation kit (Bethyl Labs, Montgomery, TX, USA) was used to measure albumin levels in the culture medium. Samples from three separate cultures were analysed in triplicate for each condition. Absorbance of basal medium was subtracted from absorbance of each sample and albumin concentration was determined from a standard curve. Normal rat serum showed absorbance at or below the zero standard provided by the manufacturer. Final absorbance of test samples was obtained after subtracting zero absorbance. Final absorbance value obtained was then used to determine albumin concentration in serum, from a standard curve. All samples were run in duplicate.

Detection of stored glycogen and cytokeratin‐18 expression

To test for glycogen storage and cytokeratin‐18 (CK‐18) expression, slides were prepared from 10 mm slices of the scaffolds containing BMSCs, on day 28 of culture, using a frozen section method. Specimens were fixed in 95% ethanol overnight and the periodic acid–Schiff (PAS) method was used to detect presence of insoluble glycogen stored in cells. Samples were then oxidized with 1% periodic acid for 5 min, rinsed with distilled water for 5 min and treated with Schiff’s reagent for 10–15 min. After rinsing with distilled water for 5 min, the specimens were counterstained with Mayer’s haematoxylin for 30 s and observed under a light microscope. Samples used for detection of CK‐18 were fixed in 4% paraformaldehyde (Sigma‐Aldrich) for 4 min at room temperature. Sections were blocked in 1% BSA and 0.1% normal rabbit serum at room temperature for 1 h, followed by incubation in rabbit anti‐rat CK‐18 antibody (1:200 dilution; Dako) at 4 °C overnight. After washing, sections were incubated in secondary tetramethyl rhodamine isothiocyanate (TRITC)‐conjugated anti‐rabbit IgG antibody (1:200; Zymed) for 1 h at 25 °C, in the dark. After washing in phosphate‐buffered saline (PBS), specimens were analysed by fluorescence microscopy. In addition, sections were stained with Hoechst 33258 to identify all nucleate cells in the scaffold. Ten slides were chosen and five random visual fields per specimen were examined to evaluate CK‐18‐positive cells.

Results

Scaffold architecture and change in morphology of cells cultured in the scaffold

As a more convenient means of observation, scaffolds containing cells were examined by optical microscopy. It was found that the alginate scaffold was transparent, thereby allowing cell and scaffold structure to be observed directly and clearly. This transparent quality provided us with a convenient method for observing scaffolds and cells (Fig. 1a). On scanning electron microscopy (SEM) observation, scaffolds exhibited a highly porous structure with interconnected pores of 300–500 μm diameter (Fig. 1c). When seeded into the scaffold, BMSCs attached to its fibres, formed multicellular spheroids and proliferated. This morphology is different from that of cells cultured in dishes (Fig. 1b). These data indicated the structural properties of alginate scaffolds and their compatibility with BMSCs.

Figure 1.

 Morphological characteristics of BMSCs and the alginate scaffold. (a) Image of the alginate scaffold as observed by optical microscopy (100× magnification), fibres of the scaffold were observed clearly. (b) Image of BMSCs in the alginate scaffold as seen by optical microscopy (400× magnification). BMSCs formed multicellular spheroids that adhered to the scaffold fibres. (c) SEM images of morphology of the alginate scaffold and the cells.

RT‐PCR and Western blot analysis of hepatocyte‐specific markers

The aim of this study was to determine whether BMSCs could differentiation into hepatocyte‐like cells in the alginate scaffold. To detect presence of hepatocyte‐specific genes in differentiated cells, RNA had been harvested from the cells for RT‐PCR, at different time points (Fig. 2a). RT‐PCR analysis revealed that the differentiated cells expressed a subset of hepatocyte genes including AFP, ALB, connexin 32 (Cox 32) and CYP7A1. AFP, a marker of endodermal differentiation and an early foetal hepatocyte marker, was expressed in the cells after 14 days differentiation, but was not detected after 28 days in culture. In contrast, expression of ALB, a marker of mature hepatocytes, appeared after 14 days culture and its expression increased after 28 days in culture. Cox 32, a cell–cell communication factor, and CYP7A1, a cytochrome p450 (CYP) enzyme, were detected in cells that had undergone differentiation for 28 days. Thus, the hepatocyte‐specific gene expression pattern revealed that the BMSCs had differentiate into hepatocyte‐like cells in the alginate scaffold.

Figure 2.

 (a) Hepatocyte‐specific gene expression determined by RT‐PCR. On day 14 of differentiation, bands for AFP and ALB genes appeared in the BMSCs; whereas on day 28, the band for ALB strengthened but the one for AFP had disappeared. Cox32 and CYP7A1 genes were both expressed in the BMSCs after 28 days of culture. Glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) was used as a control. (b) Western blot analysis of hepatocyte‐specific expression of ALB and AFP proteins. Protein expression levels were normalized against β‐actin as internal loading control.

To further characterize hepatic differentiation, protein expression of ALB and AFP was determined by Western blot analysis (Fig. 2b). ALB was detected in cells that had been cultured for 14 and 28 days. AFP was present only in cells that had been cultured for up to 14 days. These data confirmed differentiation of the BMSCs into hepatocyte‐like cells, when cultured in the alginate scaffold.

ALB and urea synthesis

To test ALB and urea synthesis by the differentiated cells, presence of ALB and urea in the culture medium was measured during differentiation. Both ALB and urea were present and their levels gradually increased until day 21. After 21 days of culture, levels of ALB and urea in the medium remained stable for the remaining culture period (Fig. 3). These data further suggested that the BMSCs had undergone hepatic maturation when cultured in the alginate scaffold.

Figure 3.

 Urea and ALB production by BMSCs cultured in alginate scaffolds during differentiation. On days 7, 14, 21 and 28, culture medium was collected for analysis of ALB and urea contents. Two build‐up curves indicated increase in urea and ALB in the culture medium, respectively.

Glycogen production and expression of CK‐18

CK‐18 is typically expressed in mature hepatocytes. Thus, we used immunocytochemical analysis to identify levels of CK‐18 expression in differentiated BMSCs cells in the scaffold. Using fluorescence microscopy, we determined that nearly 56.7% of the cells in the scaffold were CK‐18 positive (Fig. 4a,b).

Figure 4.

 BMSCs differentiated in the alginate scaffold expressed CK‐18 and were able to store glycogen. (a) Representative immunofluorescence image of differentiated BMSCs using anti‐CK‐18 antibody (400× magnification). (b) Hoechst staining identified all cells present in the scaffold. Of all of them, 56.7% were CK‐18 positive. (c) Representative image of PAS staining of BMSCs indicated ability of differentiated BMSCs to store glycogen (400× magnification).

Next, we used PAS staining to demonstrate the ability of these differentiated BMSCs to store glycogen. Using the PAS technique, glycogen storage was indicated by the presence of its cytoplasmic deposits (pink). As shown in Fig. 4c, such cytoplasmic deposits were observed in the BMSCs that had been cultured in the alginate scaffold. Taken together, all these data indicate that the BMSC‐derived hepatocyte‐like cells possessed some characteristics of normal hepatocytes.

Discussion

Tissue engineering is an interdisciplinary field that includes medicine, life sciences, and engineering to the in vitro development of biological tissues. To date, to treat patients with fulminant hepatic failure, various extracorporeal bioartificial liver (BAL) systems have been developed. Performance of a BAL system depends on the functional activity of the hepatocytes that are immobilized in the bioreactor system. BMSCs have been hypothesized to be the most promising seed cells for such use in hepatic tissue engineering. Our hypothesis was that a three‐dimensional environment would enhance cell–cell and cell–matrix interactions, couple appropriate chemical cues from growth factors and aid in differentiation of BMSCs into hepatocyte‐like cells. In the present study, we used an alginate scaffold as a three‐dimensional matrix to differentiate them into hepatocyte‐like cells. In addition, we evaluated whether the scaffold was an appropriate bioreactor device and whether BMSCs could be used as seed cells for hepatic tissue engineering, due to their ability to differentiation into hepatocyte‐like cells.

Initially, we examined the general characteristics of the alginate scaffold and compatibility of the scaffold with the BMSCs. According to Pham et al. (10), optimal tissue development requires infiltration of seed cells into the scaffold, which requires the scaffold to have a macroporous structure with interconnected pore diameters of at least 10 μm. SEM analysis revealed that our alginate scaffolds were highly porous structures with interconnected pores, between 300 and 500 μm in diameter, thus providing enough space for the cells. Importantly, well interconnected pores may aid in exchange of gases and nutrients in the scaffold. In addition, microscopic observation provides a convenient and direct method for observing cellular morphology and proliferation within the scaffold. Microscopically, the alginate scaffold was highly transparent, as shown in Fig. 1a. Further microscopic observation indicated that it consisted of a criss‐cross network of fibrous structures, consistent with our SEM results. The scaffold was found to be an appropriate hydrophilic matrix in which the cells could infiltrate through fluid medium, not only on the surface, but also into centres of construct. The highly porous, transparent and hydrophilic nature maked this alginate scaffold an excellent matrix for the BMSCs.

Using both light microscopy and SEM, we found that cells inside the alginate scaffolds formed multicellular aggregates and adhered to it. We also found that single cells gave rise to cell aggregates, as shown in Fig. 1b. Aggregated multicellular spheroids have been shown to promote extensive cell–cell contact and formation of gap and tight junctions (11). The spherical cells also had morphology and ultrastructure similar to cells found in vivo. Increased E‐cadherin‐mediated cell adhesion between cultured hepatocytes has been shown to result in increased induction of liver‐specific functions (12). These results indicate that our alginate scaffolds were compatible with the culture of our BMSCs.

After seeding primary cells into the scaffold, we added hepatocyte‐inducing medium, which resulted in gradual induction of hepatic function in the cells. The hepatocyte‐inducing medium used contained several cell factors known to play important roles in hepatocyte differentiation, including HGF and FGF. HGF, originally identified and cloned as a potent hepatocyte mitogen, induces mitogenic and morphogenic activities in a wide variety of cells that express c‐Met, a transmembrane HGF receptor with an intracellular tyrosine kinase domain (13, 14). Moreover, HGF plays an essential role as an endocrine or paracrine factor in development and regeneration of the liver (15, 16). HGF mRNA and HGF activity markedly increase in liver after various liver injuries including hepatitis, ischaemia, physical trauma and partial hepatectomy. Mimicking the injured liver microenvironment, Wang et al. induced isolated BMSCs to differentiate into hepatocyte‐like cells with HGF, in vitro, at a concentration (50 ng/ml), similar to that used in the present study (17). All biological responses induced by HGF are elicited by binding to its receptor. Oh et al. (18) detected expression of c‐Met mRNA by RT‐PCR in freshly isolated bone marrow cells in culture, and found that HGF efficiently induced their differentiation into albumin‐expressing hepatocyte‐like cells, at concentrations ranging from 0.5 to 5 mg/ml. HGF and c‐Met may mediate important steps in organogenesis in vitro and play a crucial role in placental and foetal hepatic development in vivo. Interestingly, the FGF family contains pleiotropic growth factors that control cell proliferation, migration and differentiation and after liver injury, it has been reported that FGF production markedly increases. This is likely to play an important role in repairing liver injury and FGF may stimulate repair of wounded hepatocyte monolayers by enhancing both hepatocyte proliferation and migration.

During differentiation, cells cultured in alginate scaffolds maintained rounded shape, aggregated to form multicellular spheroids and gradually adopted hepatocyte characteristics, as observed by their expression of liver‐associated genes and their ability to acquire hepatocyte functions. Liver‐associated gene expression during ontogeny is characterized by on–off switches such as serum AFP to ALB switches (19). Shortly after birth, expression of AFP protein in the liver significantly decreases and marked reduction in AFP mRNA levels occurs. Production of AFP is also associated with normal hepatocyte division. ALB, the most abundant protein synthesized by hepatocytes, is initially expressed in foetal rat liver on embryonic day 11.5 and continues to be expressed throughout adulthood. To determine whether our BMSCs cultured in the alginate scaffold had differentiated into hepatocyte‐like cells, we extracted total RNA from the cells after 14 and 28 days differentiation, and examined expression levels of AFP and ALB mRNA, by RT‐PCR. We found that AFP mRNA was expressed on day 14 day of differentiation, but was not detected after 28 days of culture. ALB mRNA continued to be expressed and increased up to 28 days of culture; induction of Cox 32 was also observed in the cells on day 28. Cox 32 is the predominant gap junction protein expressed by hepatocytes. Intracellular communication via gap junctions plays an important role in regulating cell survival, apoptosis, differentiation, proliferation and tumorigenesis. In the present study, Cox 32 was shown to be expressed in later stages of differentiation, thus providing further evidence that the BMSCs had differentiated into hepatocyte‐like cells. In addition, our results show that hepatic differentiation was initiated by the BMSCs in response to the culture medium (Fig. 3).

To confirm differentiation of BMSCs into hepatocyte‐like cells, we performed several functional tests to measure synthesis of ALB and urea, storage of glycogen, and expression of CK‐18 and CYP7A1. As shown in Fig. 2, production of ALB and urea in the culture medium gradually increased during differentiation. While urea production is a characteristic of hepatocyte activity, kidney tubular epithelial cells also produce it (20). However, ALB production is a specific metabolic marker of hepatocytes (20). These results indicate that BMSCs in the scaffold not only expressed ALB gene, but also had the ability to secrete ALB into the extracellular matrix. Furthermore, glycogen production is an additional metabolic function specific to hepatocytes. Using the PAS method, we observed that our BMSCs cultured in the alginate scaffold displayed the capacity to store glycogen on day 28 of differentiation. CK‐18 is expressed during hepatocyte development and is a marker of mature hepatocytes. On day 28 of culture, we detected presence of CK‐18‐positive cells in the scaffold. According to our results, 56.7% of cells in the scaffolds were expressing CK‐18, which may represent a differentiation rate for the BMSCs. CYPs constitute a superfamily of haeme proteins that play important roles in detoxification of numerous xenobiotics, as well as endogenous compounds, such as steroids, fatty acids, prostaglandins and leucotrienes. Particular CYPs are expressed at low levels in extrahepatic tissues such as the intestine, lungs and kidneys; however, the liver represents a major site of CYP‐mediated oxidative metabolism. Thus expression of CYP7A1 is yet another indicator of hepatocyte‐specific function.

In summary, presence of three‐dimensional spheroidal cultures is a promising indication of differentiation of BMSCs into hepatocyte‐like cells. The ultimate goal of hepatic tissue engineering is to create a complete, implantable and functional liver (21). Whether the combination of matrix and BMSCs described here will lead to creation of implantable organs, requires further research. However, we anticipate that differentiation of BMSCs in alginate scaffolds may become a very useful tool for hepatic tissue engineering.