Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jul 1.
Published in final edited form as: Biotechnol Prog. 2008 Sep-Oct;24(5):1132–1141. doi: 10.1002/btpr.41

Augmentation of EB-Directed Hepatocyte-Specific Function via Collagen Sandwich and SNAP

Eric I Novik 1, Jeffery Barminko 2, Tim J Maguire 3, Nripen Sharma 4, Eric J Wallenstein 5, Rene S Schloss 6, Martin L Yarmush 7
PMCID: PMC4487523  NIHMSID: NIHMS702211  PMID: 19194923

Abstract

The development of implantable engineered liver tissue constructs and ex vivo hepatocyte-based therapeutic devices are limited by an inadequate hepatocyte cell source. In our previous studies, embryoid body (EB)-mediated stem cell differentiation spontaneously yielded populations of hepatocyte lineage cells expressing mature hepatocyte markers such as albumin (ALB) and cytokeratin-18 (CK18). However, these cultures neither yielded a homogenous hepatocyte lineage population nor exhibited detoxification function typical of a more mature hepatocyte lineage cell. In this study, secondary culture configurations were used to study the effects of collagen sandwich culture and oncostatin-M (OSM) or S-nitroso-N-acetylpenicillamine (SNAP) supplementation of EB-derived hepatocyte-lineage cell function. Quantitative immunofluorescence and secreted protein analyses were used to provide insights into the long-term maintenance and augmentation of existing functions. The results of these studies suggest that SNAP, independent of the collagen supplementation, maintained the highest levels of ALB expression, however, mature liver-specific CK18 was only expressed in the presence of gel sandwich culture supplemented with SNAP. In addition, albumin secretion and cytochrome P450 detoxification studies indicated that this condition was the best for the augmentation of hepatocyte-like function. Maintenance and augmentation of hepatocyte-like cells isolated from heterogeneous EB cell populations will be a critical step in generating large numbers of functional differentiated cells for therapeutic use.

Keywords: ES cells, hepatocytes, collagen sandwich, SNAP, cytochrome P450

Introduction

Acute liver failure affects hundreds of thousands of people per year around the globe and in many cases is resolved with an orthotopic liver transplant. Because of the shortage of donor organs, many patients will die while waiting for a donor organ to be available. Extracorporeal liver assist devices (LAD) could help to bridge patients for transplantantation; however, this technology is limited by a lack of an adequate hepatocyte cell source (Tilles et al., 2002a,b). Pluri-potent embryonic stem (ES) cells represent a promising renewable cell source to generate hepatocyte lineage cells, which have been incorporated into implantable engineered tissue constructs (Soto-Gutierrez et al., 2006) and ex vivo cell-based therapeutic devices such as LADs (Cho et al., 2008). However, the current differentiation techniques have not yet generated the large and functionally sustainable cell masses, which would be required to make such therapies clinically available.

ES differentiation into hepatocyte lineage cells, using a variety of differentiation platforms such as monolayer (Sharma et al., 2006), encapsulation (Maguire et al., 2006), and embryoid body (EB) mediated (Hamazaki et al., 2001; Heo et al., 2006; Kumashiro et al., 2005b), has been previously described by many investigators. Of these, EB-mediated differentiation, which mimics in vivo embryogenesis, has been characterized most completely. For example, following exogenous growth factor supplementation and coculture with nonparenchymal liver cell lines, investigators have demonstrated that EB-mediated differentiation yields up to a 70% albumin (ALB)-positive population, which expresses a variety of liver lineage genes and metabolizes lidocane and diazepam (Soto-Gutierrez et al., 2007).

In addition, in vitro aggregation of murine ES cells initiates the formation of EBs, which has been shown to facilitate spontaneous differentiation in the absence of growth factor and extracellular matrix supplementation, resulting in liver lineage cells characterized by 80% ALB expression as well as mature hepatocyte genes such as cytochrome P450-detoxifying enzymes (CYP450) (Novik et al., 2006; Tsutsui et al., 2006). However, despite the large number of studies reporting ES-hepatocyte lineage differentiation, there have been few reports of CYP450-related detoxification and drug metabolism, which may be required for successful use of these cells for therapeutic treatment. Furthermore, maintenance of differentiated function and/or increased cell mass after the initial differentiation are critical steps for use of ES-generated in vitro liver lineage cells in a LAD or drug discovery studies. Finally, the selection of a homogenous cell population is essential to prevent undesirable cellular interactions and functional expression. However, these issues have been largely ignored. Previous studies investigated the effects of oncostatin-M (OSM) and nitric oxide (NO) donors on fetal liver hepatocytes and have shown that their supplementation maintains long-term structure and function as well as inducing further differentiation (Ehashi et al., 2005; Iwai et al., 2002). In addition, studies have shown the potential of collagen sandwich cultures to augment hepatocyte populations from within the heterogeneous populations (Depreter et al., 2000).

In this study, we have evaluated the effects of collagen sandwich culture and S-nitroso-N-acetylpenicillamine (SNAP), a NO donor, or OSM supplementation on maintenance of previously reported EB mediated spontaneously differentiated hepatocyte-lineage cell function. We have assessed expression and secretion of ALB within cells secondarily cultured from the Day 17 EBs. In addition, using s-methylcholanthreene induction, we have assessed not only maintenance but also augmentation of function in the form of CYP450-mediated detoxification. These studies identified a secondary culture condition, which maintained liver-like function initially observed after 17 days of spontaneous EB-mediated differentiation, and furthermore, promoted the detoxification functions of Cyp450 enzymes.

Materials and Methods

Cell culture

All cell cultures were incubated in a humidified 37°C and 5% CO2 environment. The ES cell line D3 (ATCC, Manassas, VA) was maintained in an undifferentiated state in T-75 gelatin-coated flasks (Biocoat, BD-Biosciences, Bedford, MA) in Knockout Dulbecco's modified Eagle's medium (Gibco, Grand Island, NY) containing 15% knockout serum (Gibco), 4 mM l-glutamine (Gibco), 100 U/mL penicillin (Gibco), 100 U/mL streptomycin (Gibco), 10 μg/mL gentamicin (Gibco), 1,000 U/mL ESGRO™ (Chemicon, Temecula, CA), 0.1 mM 2-mercaptoethanol (Sigma-Aldrich, St. Louis, MO). ESGRO contains leukemia inhibitory factor (LIF), which prevents ES cell differentiation. Every 2 days, media was aspirated and replaced with fresh media. Cultures were split and passaged every 6 days, following media aspiration and washing with 6 mL of phosphate-buffered solution (PBS) (Gibco). Cells were detached following incubation with 3 mL of trypsin (Gibco) for 3 min, resulting in a single cell suspension, and subsequently, the addition of 12 mL of knockout DMEM. Cells were then replated in gelatin-coated T-75 flasks at a density of 1 × 106 cells/mL. Staining with Oct4, a recognized stem cell marker, demonstrated that the cells remained undifferentiated over the period used to accomplish these studies. 100% Oct4 staining was observed at all passages (data not shown).

To induce differentiation, cells were suspended in Iscove's modified Dulbecco's medium (Gibco) containing 20% fetal bovine serum (Gibco), 4 mM l-glutamine (Gibco), 100 U/ mL penicillin, 100 U/mL streptomycin (Gibco), and 10 μg/mL gentamicin (Gibco). EBs were formed and cultured for 2 days using the hanging-drop method (1 × 103 ES cells per 30 μL drop). The hanging drops where transferred to suspension culture in 100-mm Petri dishes and cultured for an additional 2 days. The EBs were then plated, one EB per well, in 6-well tissue culture polystyrene plates (BD-Biosciences) for an additional 14 days. For secondary culture, Day-17 EB cells were detached following incubation with 5 mL of trypsin (Gibco) for 3 min, resulting in a single cell suspension, and subsequently, the addition of IMDM media. Cells from Day-17 EBs were used because it has been observed that hepatocyte function is greatest on Day 17 (data not shown). Cells were then replated in 6-well tissue culture polystyrene (BD-Biosciences) at an initial seeding density of 5 × 104 Day 17 cells per well for further analysis. Culture medium was changed every 48 h. When OSM and SNAP were supplemented, 10 ng/mL OSM and 250 μM SNAP were added to the culture medium. When collagen sandwich culture was used, rat tail type I collagen (BD-Biosciences) gels were prepared by distributing 350 μL of collagen gel solution (3 parts 1.33× DMEM, pH 7.4, and 1 part collagen solution at 4 mg/mL, chilled on ice and mixed immediately before use) evenly over one well of a 6-well plate (BD-Biosciences) and incubated at 37°C for at least 1 h before use. A 5 × 105 cells were seeded in 2 mL of IMDM media on Day 0 and an additional 350 μL of collagen gel solution was distributed over the cells after 1 day of culture. Therefore, the second layer of collagen is added on Day 1 of secondary culture protocol. One hour of incubation at 37°C was allowed for gelation, and attachment of the second gel layer before the medium was replaced. Culture medium was changed every 48 h.

The Hepa 1-6 cell line (ATCC, Manassas, VA) was maintained in Dulbecco's modified Eagle's medium (Gibco) containing 10% fetal bovine serum (Gibco), 100 U/mL penicillin (Gibco), 100 U/mL streptomycin (Gibco), and 4 mM l-glutamine (Gibco). Hepa 1-6 cells were grown on tissue culture-treated T-75 flasks (Falcon, BD Biosciences, San Jose, CA). Hepa 1-6 cells were used as positive controls for each of the following assays.

On evaluation days 4, 6, 8, and 10 days in secondary culture, cells were replated into 12-well plates. Media samples were collected after 24 h of culture at 37°C and 5% CO2. The cells were then washed in PBS (Gibco) and fixed in 4% paraformaldehyde (Sigma-Aldrich) in PBS for 15 min at room temperature. Cells in collagen sandwich culture were dissociated with 0.5 mL of 0.1% collagenase (Sigma-Aldrich) in PBS for 30 min at 37°C before replating into 12-well plates.

In situ indirect immunofluorescent cytokeratin-18 and intracellular albumin analysis

After 24 h in culture and fixing with 4% paraformalde-hyde, the cells were washed for 10 min in cold PBS and fixed in 4% paraformaldehyde (Sigma-Aldrich) in PBS for 15 min at room temperature. The cells were washed twice for 10 min in cold PBS and then twice for 10 min in cold saponine (SAP)/PBS membrane permeabilization buffer containing 1% bovine serum albumin (BSA) (Sigma-Aldrich), 0.5% SAP (Sigma-Aldrich), and 0.1% sodium azide (Sigma-Aldrich). To detect intracellular ALB, the cells were subsequently incubated for 30 min at 4°C in a SAP solution containing rabbit anti-mouse ALB antibody (150 μg/mL) (MP Biomedicals, Irvine, CA) or normal rabbit serum (150 μg/mL) (MP Biomedicals) as an isotype control, washed twice for 10 min in cold SAP buffer, and then treated for 30 min at 4°C with the secondary antibody, FITC-conjugated donkey anti-rabbit, diluted 1:500 (Jackson Immuno Labs, Westgrove, PA). To detect cytokeratin-18 (CK18), which is produced in mature hepatocytes and a few other mature cell types, we incubated the cells for 30 min at 4°C in a SAP solution containing rabbit anti-moue CK18 antibody (IgG1) (1:50 dilution) (Santa Cruz Biotechnology) or the IgG1 fraction of normal rabbit serum (1:100 dilution) (Santa Cruz Biotechnology) as an isotype control, and then treated for 30 min at 4°C with the secondary antibody, FITC-conjugated goat anti-rabbit, diluted 1:200 (Jackson Immuno Labs, Westgrove, PA). For both stains, cells were then washed once with cold SAP buffer and once with cold PBS. Fluorescent images were acquired using a computer-interfaced inverted Olympus IX70 microscope. Specimens were excited using a 515-nm filter. Fluorescent intensity values were determined for each cell using Olympus Microsuite. Experimental intensity values for each cell were calculated after subtracting the average intensity of the isotype control.

Sandwich ELISA for detection of albumin secretion

To detect the secreted ALB within the media supernatants obtained on each of the analysis days, we used a commercially available mouse ALB ELISA kit (Bethyl Laboratories, #E90-134). A standard curve was generated by creating serial dilutions of an ALB standard from 7.8 to 10,000 ng/mL. Absorbance readings were obtained using a Biorad (Hercules, CA) Model 680 plate reader with a 450-nm emission filter. ALB values were normalized to the cell number recorded on the day of media sample collection.

Urea secretion

Media samples were collected on all analysis days. Urea synthesis was assayed using a commercially available kit (StanBio, Boerne, TX). A standard curve was generated by creating serial dilutions of a urea standard from 0 to 300 mg/mL. Absorbance readings were obtained using a Biorad (Hercules, CA) Model 680 plate reader with a 585-nm emission filter. Urea values were normalized to the cell number recorded on the day of media sample collection.

Measurement of cytochrome P450 activity

On evaluation days 4, 6, 8, and 10 days in secondary culture, cells were replated into 12-well plates. 3-Methylcholanthrene was used at a concentration of 2 μM (Sigma–Aldrich) for 48 h before the addition of resorufin as an inducer of cyto-chrome P450 activities. Cytochrome P450-dependent benzyloxyresorufin o-dealkylase activity (BROD, PROD, EROD, and MROD) was measured using resorufin substrates namely pentoxy-, benzyloxy-, ethoxy-, and methoxyresorufin from a Resorufin Sampler Kit (Invitorgen, Carlsbad, CA). The incubation mixture contained resorufin substrates (pentoxy-, ethoxy-, or methoxyresorufin, final concentration 5 mM) and dicumarol (80 mM) in phenol red free Earle's balanced salt solution (EBSS) (Gibco). The prepared solutions were pre-heated to 37°C before the incubation with cells. The 12-well plates were washed with 2 mL of EBSS (37°C) and further incubated with 2 mL of EBSS at 37°C for 5–7 min to remove the residual medium. Following removal of EBSS, the incubation mixture was added (2 mL per well) and the dishes were incubated at 37°C in a 5% CO2 incubator. At various time points (5, 10, 15, 20, 25 min) following incubation, 100 μL of the mixture was transferred into a 96-well plate. The fluorescence of the plate was measured using a fluorescence plate reader (DTX880, Beckman Coltour, Fullerton, CA, ext. 530 nm and emis. 590 nm) at the end of 25-min incubation. A standard curve of resorufin fluorescence was constructed using concentrations ranging from 1 to 1,000 nmol in EBSS. A linear curve was obtained with an r2 of 0.99. The constructed standard curve was used to convert the fluorescence values obtained from the plate reader to nano-moles of resorufin. Rate of formation of resorufin, as calculated from the early linear increase in the fluorescence curve, was defined as cytochrome P450 activity and expressed as nmol/min.

Statistical analysis of functional assays

Each data point represents the mean of three experiments (each with three biological replicates), and the error bars represent the standard deviation of the mean. Statistical significance was determined using the Student's t-test for unpaired data. Differences were considered significant when the probability was less then or equal to 0.05.

Results

Dynamic studies of secondarily cultured EB-derived hepatocyte lineage cells

Our previous studies have demonstrated that spontaneous EB-mediated differentiation of ES cells yields a population of cells displaying hepatocyte-specific characteristics such as ALB and CK18 expression. However, these cultures did not yield a homogenous hepatocyte lineage population. In addition, these cells did not exhibit detoxification function typical of a more mature hepatocyte lineage cell. Therefore, studies were initiated to determine whether maintenance and augmentation of hepatocyte-like function could be induced in a secondary culture configuration. Hepatocyte lineage maintenance was initially assessed by examining the dynamics of cell growth following removal of cells from their primary EB culture and replating into tissue culture polystyrene. A 5 × 104 cells from Day-17 EB cultures were replated into one well of a 6-well plate and evaluated on days 4, 6, 8 and 10 days post-replating. Cell number increased rapidly and confluence was reached at Day 6. Therefore, cells were replated into tertiary culture at 5 × 104 cells per well and continued to proliferate for the next 4 days (Figure 1A).

Figure 1. Nonsupplemented secondary culture characterization.

Figure 1

Open in a new tab

(A) Cell number was assessed by counting total cells dissociated following trypsinization. Cells were plated into tertiary culture at 5 × 104 cells per well at Day 6. (B) Time course of the percentage of cells expressing albumin in polystyrene secondary culture. Each data point represents the percentage of cells with an intensity reading above 0. The average of three experiments is presented. All values were statistically signifi-cant when compared with the ES control. (C) Time course of the percentage of cells expressing CK18 in polystyrene secondary culture. Each data point represents the percentage of cells with a normalized intensity reading above 0. The average of three experiments is presented. All values were statistically significant when compared with the ES control.

Next, experiments were designed to evaluate the maintenance of function seen in EB-generated hepatocyte lineage cells by assessing in situ intracellular ALB and CK18 expression. Secondary and tertiary cultures were initiated as outlined earlier, and ALB and CK18 expression were qualitatively assessed on 4, 6, 8, and 10 days post-replating using indirect immunofluorescence with either primary anti-ALB/CK18 antibody or an immunoglobulin control serum and subsequently fluorescently labeled secondary antibody. Images were captured using digital microscopy to determine the percent of ALB and CK18 expressing cells within the cultures. The Day-17 EB-generated cells were 80% ALB positive (Figure 1B) and 60% CK18 positive (Figure 1C). As depicted in Figure 1B, ALB expression was maintained for 6 days at ~40% in secondary culture, however, expression was not maintained past 6 days in tertiary culture. CK18 expression was maintained at minimal expression levels for 4 days in secondary culture but was absent on subsequent days (Figure 1C). In secondary polystyrene culture, EB-derived cells proliferated rapidly but could not sustain ALB expression. Therefore, we explored the addition of soluble factors on proliferation and maintenance of function.

OSM and SNAP supplementation

To investigate the effect of soluble factors previously shown to regulate hepatic function, replated cells were supplemented with either OSM or SNAP (Ehashi et al., 2005; Iwai et al., 2002). Cell numbers in the OSM-supplemented condition were similar to that of the unsupplemented cultures and they increased dramatically. However, cells exposed to SNAP were generally characterized by slower growth rates. Because of the rapid growth seen in the unsupplemented and OSM cultures, at Day 6, cells were replated into tertiary culture at 5 × 104 cells per well and continued to proliferate for the next 4 days (Figure 2A). ALB expression was maintained in the OSM-supplemented cultures for up to 8 days in secondary culture. The cells supplemented with SNAP also maintained some ALB expression up to 8 days in tertiary culture but at a lower level. There was no significant expression following 10 days in secondary culture in any condition (Figure 2B). Urea secretion was greatest at Day 8 in the OSM-supplemented condition; however, some secretion was detected at all experimental time points. The Day-17 EB hepatocyte-like cells exhibited a urea secretion rate of 50 μg/106 cells/day (Figure 2C). A summary of the hepatocyte-like functions tested is summarized in Table 1. CK18 expression as well as other hepatocyte functions such as ALB secretion, glycogen storage, and CYP450-mediated detoxification was not detected at any level in the OSM- or SNAP-supplemented cultures. Although addition of soluble factors maintained ALB expression for up to 8 days in the secondary culture, most hepatocyte functions were not maintained at any significant level and others were totally absent.

Figure 2. OSM and SNAP supplemented secondary culture characterization.

Figure 2

Open in a new tab

(A) Cell number was assessed by counting total cells dissociated following trypsinization. Cells from the three conditions were plated into tertiary culture at 5 × 104 cells per well at Day 6. (B) Time course of the percentage of cells expressing albumin in polystyrene secondary culture. Each data point represents the percentage of cells with an intensity reading above 0. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from other conditions on that day. (C) Time course of urea secretion rates in the supplemented conditions. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from other conditions on that day.

Table 1.

Function Summary for Supplemented Cultures

NS SNAP OSM
Intracellular ALB + + +
CK18 + – –
Urea + + +
ALB secretion – – –
CYP450 – – –
Glycogen – – –
Open in a new tab

(+) represents at least one time point of significant expression or secretion function in the three conditions; (–) represents functions which were absent on all experimental days.

Collagen sandwich culture

To determine whether we could further augment and/or maintain the function of the hepatocyte-like cells isolated from the Day-17 EB culture, collagen sandwich culture, a system which has been well studied for maintenance of mature hepatocyte function (Dunn et al., 1989), was utilized alone (GEL) and in conjunction with OSM (GOSM) and SNAP (GSNAP) supplementation. Cells cultured in a sandwich configuration were characterized by a slower rate of proliferation when compared with polystyrene culture. Cells in the GEL and GOSM conditions reached maximum growth at Day 6 and cells in GSNAP by Day 8. Because of the low-proliferation rates in sandwich culture, the cells did not reach absolute confluence and no tertiary culture was employed (Figure 3). As depicted in Figure 4, ALB expression was detected in all conditions for 6 days in secondary culture; however, expression was maintained at 10 days post-replating, only in the GSNAP and GOSM conditions (Figure 4A). Although, the combination of collagen gel cultures and SNAP on Day 10 of secondary cell culture yielded an eightfold increase in ALB positive cell population (Figure 4B) relative to the control, the cell number was fivefold lower (Figure 3). Even though the proliferation rate of SNAP-treated cells was significantly lower, cellular function better resembled that of an adult hepatocytes population which does not proliferate in vitro. A similar effect was evident in the expression of CK18 where some expression was detected in the nonsupplemented, non-GEL (NS) condition for 4 days but it was expressed only in the GSNAP and GOSM conditions after 10 days (Figure 5A). However, while ~45% of those expressed CK18 in the GSNAP condition only ~20% were positive in the GOSM condition (Figure 5B). The GEL, GOSM, and GSNAP conditions stored glycogen 10 days into secondary culture. The NS conditions did not significantly stain for glycogen (data not shown).

Figure 3. Collagen sandwich secondary culture characterization.

Figure 3

Open in a new tab

Cell number was assessed by counting total cells dissociated following trypsinization. Cell number in all GEL conditions was assessed by counting the total number of cells dissociated following collagenase digestion and trypsonization. Asterisk (*) indicates time point at which the NS cells were passed to 5 × 104 cells per well.

Figure 4. Intracellular albumin characterization.

Figure 4

Open in a new tab

(A) Fluorescence and bright field images of intracellular albumin on Day 10 of replated collagen sandwich cultures supplemented with SNAP, OSM, and polystyrene cultures without supplementation. (B) Time course of the percentage of cells expressing albumin in polystyrene secondary culture. Each data point represents the percentage of cells with an intensity reading above 0. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from NS and GEL conditions on that day.

Figure 5. Intracellular CK18 characterization.

Figure 5

Open in a new tab

(A) Fluorescence and bright field images of intracellular CK18 on Day 10 of replated collagen sandwich cultures supplemented with SNAP, OSM, and polystyrene cultures without supplementation. (B) Time course of the percentage of cells expressing CK18 in the sandwich culture conditions. Each data point represents the percentage of cells with an intensity reading above 0. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from all other conditions on that day.

Urea and ALB secretion, vital liver functions, were used to asses mature hepatocyte-specific differentiated function. A dynamic profile of ALB secretion was established using ELISA analysis. Although at 4 days in secondary culture there was an initial induction of ALB secretion in both the GOSM and GSNAP conditions, rates were significantly higher in the GSNAP condition on subsequent days when compared with all other conditions. In addition, secretion was maintained at 60 ηg/106 cells/day after 10 days in secondary culture (Figure 6A). The Day-17 EB-derived cells did not secrete ALB (data not shown) and had a urea secretion rate of 50 μg/106 cells/day. In secondary culture, all conditions maintained some urea secretion. However, the 25 μg/106 cells/day observed in the GSNAP condition was significantly higher than any other condition at Day 10 (Figure 6B).

Figure 6. Albumin and urea secretion rates in sandwich culture.

Figure 6

Open in a new tab

(A) Time course of ALB secretion rates in the sandwich culture conditions. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from all other conditions on that day. (B) Time course of urea secretion rates in the sandwich culture conditions. The average of three experiments is presented. All values were statistically significant when compared with the ES control. Asterisk (*) indicates statistically significant differences (P < 0.05) from all other conditions on that day.

At the end of the culture period, cells cultured in all conditions were characterized by a variety of cell morphologies. In all NG cultures, cells displaying elongated nonparenchymal morphology were present. In all double gel conditions, cells were assembled in random densely packed groupings and exhibited tightly packed morphologies. However, in the GSNAP condition, there was a second morphology that was characterized by more than 95% of cells in groups of round or square cells in a nonconfluent, loosely connected environment (Figure 7).

Figure 7. Cellular morphologies.

Figure 7

Open in a new tab

(A) ×10 magnification, phase contrast image of cells in the GEL condition. Cells with similar morphologies were observed in all Gel conditions. (B) ×20 magnification, phase contrast image of cells described in A. (C) ×10 magnification, phase contrast image of GSNAP cells in a nonconfluent, loosely connected environment. (D) ×20 magnification, phase contrast image of GSNAP cells were greater than 95% of cells are in groups of round or square cells indicated by white arrows. (E) ×20 magnification, phase contrast image of NG cells. Cells with similar morphologies were observed in all non-GEL conditions.

Cytochrome P450 detoxification

Based on urea, ALB, CK18, and morphological analysis, the GSNAP and GOSM conditions were superior to other conditions in maintaining hepatocyte-specific functions, with GSNAP being somewhat more effective than GOSM. To determine whether GSNAP or GOSM was more efficient at yielding a hepatocyte lineage population, detoxification, an essential function imperative in defining a usable replacement hepatocyte cell population, was assessed. Cytochrome P450 enzymes play a key role in detoxifying xenobiotics and were used in these studies to assess hepatocyte function. These studies monitored the expression and stabilization of BROD and methoxyresorufin o-dealkylase following induction with 3-methylcholanthrene for 48 h in Day-17 EB-derived cells and for 10 days in secondary GSNAP and GOSM cultures. BROD and MROD activity can be determined from the enzymatic conversion of resorufin. This activity detected via increasing concentration of resorufin was only apparent after 10 days in secondary GSNAP culture (Figure 8A). The rate of production was similar to that of the Hepa 1-6 control (Figure 8B). This activity was not seen in GOSM cultures.

Figure 8. Cytochrome P450 detoxification.

Figure 8

Open in a new tab

All graphs represent cells that have been in secondary culture for 10 days. Hepa 1-6 were used as a positive control. (A) BROD activity is measured every 5 min via metabolism of methoxyresorufin to resorufin. Increases in resorufin concentration indicate activity. (B) Averaged rates of production of MROD and BROD based on total cell number.

Discussion

The development of implantable engineered liver tissue constructs and ex vivo hepatocyte-based therapeutic devices is limited by an inadequate hepatocyte cell source. Differentiated pluripotent ES cells have been used to alleviate the cell source limitation problem but their utility is contingent upon generating the large number of cells and sustaining function for extended periods of time. In this study, we used previously identified hepatocyte lineage cells (Novik et al., 2006) to evaluate the effects of OSM, SNAP, and collagen sandwich culture on maintenance and augmentation of differentiated function already observed after EB-mediated differentiation. These studies indicate that maintenance of function, characterized by intracellular ALB and CK18 expression as well as urea secretion, was maintained in the presence of SNAP and OSM when used in conjunction with collagen sandwich culture. In addition, while ALB secretion and Cyp450 detoxification were not seen in the starting population, they were induced after 10 days in the GSNAP cultures.

Although others have reported long-term maintenance of hepatocyte-like function from ES sources differentiated in vivo (Teratani et al., 2005a,b) there have been no reports of maintaining EB-derived hepatocyte-like function after the primary differentiation is complete. In fact, most studies do not explore the function after the initial differentiation protocol. A significant problem associated with ex vivo adult hepatocyte culture is the rapid loss of differentiated function and morphology (Koide et al., 1989; Nahmias et al., 2007). In our studies, we saw a similar effect in that intracellular ALB and CK18 expression was significantly reduced when cultured under standard tissue culture conditions and passed several times into secondary and tertiary culture. Additionally, because of the heterogeneity of the population cells that are more likely to proliferate may be outgrowing the hepatocytes and masking their function. It is possible that a combination of these factors is contributing to loss of hepatocyte function observed in standard tissue culture conditions. To better maintain adult hepatocyte function in vitro, investigators have shown that when cultured in collagen sandwiches, hepatocytes maintain ALB secretion and cell morphology for up to 42 days (Dunn et al., 1991). Furthermore, it has been shown that the collagen sandwich configuration can select against other liver nonparenchymal cell types such as stellate cells in favor of hepatocytes, indicating a potential selectivity for hepatocytes within a heterogeneous population (Depreter et al., 2000). In fact, cells in the NG-NS conditions exhibit morphology similar to that of elongated nonparencymal cells such as stellate cells. The collagen sandwich configuration also provides a simple means of culturing the ES-derived hepatocytes when compared with techniques that require several costly growth factors and growth surfaces. In these studies there was a marked difference in cellular morphologies observed in the collagen sandwich culture; however, collagen sandwich culture alone could not maintain ALB expression past the 6 days in secondary culture and had limited CK18 expression.

Interestingly, when SNAP or OSM was added to the sandwich culture, results were significantly altered. SNAP has been shown to upregulate mitochondrial and differentiated function in hepatocyte-like cells derived from ES cells (Sharma and Yarmush, submitted), and OSM has been identified as a key morphogen in the transition of fetal to adult hepatocytes (Kamiya et al., 2006) as well as the adult livers regeneration in response to injury (Znoyko et al., 2005). Many investigators have measured ALB expression both using immunocytochemistry and RT-PCR to help identify hepatocyte lineage cells in a variety of differentiation protocols (Choi et al., 2002; Kumashiro et al., 2005a; Miyashita et al., 2002). Recently, ES cells transfected with green fluorescent protein (GFP) reporter gene regulated by ALB enhancer/promoter have been used to identify and isolate hepatocyte lineage cells during differentiation (Heo et al., 2006; Soto-Gutierrez et al., 2006; Teratani et al., 2005b; Yamamoto et al., 2003). In these studies there have been reports of up to 70% ALB positive cells. However, as others have previously shown that ALB may also be expressed in the visceral endoderm as well as fetal tissue, it alone cannot be used to confidently identify hepatocyte lineage cells (Asahina et al., 2004). The combination of sandwich culture with SNAP supplementation not only maintained ALB expression at all time points but also resulted in over 80% ALB positive cells indicating a relatively homogeneous population about 4 weeks after differentiation induction. In addition to ALB expression, CK18 expression was maintained at 45% up to 10 days in secondary culture. Although this is lower than the 60% positive population seen immediately following EB differentiation, the GSNAP culture condition was the only CK18 expressing condition. A similar trend was seen with urea secretion where after 10 days in secondary culture, urea secretion in the GSNAP condition was not as high as the Day-17 EB culture but was significantly greater than any other culture condition.

In addition to maintaining ALB, CK18, and urea secretion, the GSNAP condition also induced ALB secretion not seen at any significant level in Day-17 EB culture. ALB secretion from ES-derived hepatocyte lineage cells has been reported previously (Gouon-Evans et al., 2006; Maguire et al., 2007; Soto-Gutierrez et al., 2006; Teratani et al., 2005a; Tsutsui et al., 2006), and in the current studies it is first detected at Day 4 at 120 ηg/106 cells/day and decreases to about 60 ηg/106 cells/ day. Although this is significantly lower than the levels of secretion seen in the Hepa 1-6 control, it is significantly higher than any other experimental condition evaluated here and similar to previously reported ES-derived hepatocyte-like secretion level.

In addition to the secreting function, detoxification was also detected via CYP450 metabolism in the GSNAP condition. Xenobiotic metabolism has been well characterized in primary hepatocyte systems (Behnia et al., 2000; Roy et al., 2001), and although there have been reports of induction of CYP450 mRNA in ES-derived hepatocyte-like cells, there have been few reports on detoxification, a function which would be critical for use of these cells in a LAD (Asahina et al., 2004; Soto-Gutierrez et al., 2006; Tsutsui et al., 2006). Here, we used 3-MC for 48 h to induce CYP450 activities and observed that both BROD and MROD detoxification was observed at a level similar to the Hepa 1-6 mouse hepatocyte carcinoma cell line.

Interestingly, in these analyses, most reported functions were seen 10 days in the GSNAP condition. This allows us to maintain and augment function of spontaneously derived EB-mediated hepatocyte-like cells. A possible mechanism may be that the collagen sandwich is selected for the hepatocyte-lineage cells while SNAP is acting to drive the cells to a more mature phenotype. In addition to maintenance, we were able to show an increase in cell number from 5 × 104 Day-17 cells to 1 × 106 cells 10 days into secondary culture while still maintaining 80% ALB expression. Future experiments may be done to determine how long these cells can retain function following termination of SNAP supplementation. In addition, we may also explore the duration and time of initial exposure to the SNAP, which could uncover methods to further increase cell mass while sustaining function. The fact that Day-17 EB cells can proliferate in the GSNAP condition, while maintaining their hepatocyte-like characteristics, brings an added value to generate the large mass of cells required for use in a LAD. Nevertheless, our data indicate a combination of maintenance and augmentation of hepatocyte-specific functions in conjunction with an increase in cell mass in the GSNAP condition for up to 4 weeks postdifferentiation induction.

Contributor Information

Eric I. Novik, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854

Jeffery Barminko, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854.

Tim J. Maguire, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854

Nripen Sharma, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854.

Eric J. Wallenstein, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854

Rene S. Schloss, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854

Martin L. Yarmush, Dept. of Biomedical Engineering, Rutgers University, Piscataway, NJ 08854

Literature Cited

  1. Asahina K, Fujimori H, Shimizu-Saito K, Kumashiro Y, Okamura K, Tanaka Y, Teramoto K, Arii S, Teraoka H. Expression of the liver-specific gene Cyp7a1 reveals hepatic differentiation in embryoid bodies derived from mouse embryonic stem cells. Genes Cells. 2004;9:1297–1308. doi: 10.1111/j.1365-2443.2004.00809.x. [DOI] [PubMed] [Google Scholar]
  2. Behnia K, Bhatia S, Jastromb N, Balis U, Sullivan S, Yarmush M, Toner M. Xenobiotic metabolism by cultured primary porcine hepatocytes. Tissue Eng. 2000;6:467–479. doi: 10.1089/107632700750022125. [DOI] [PubMed] [Google Scholar]
  3. Cho C, Park E, Suganuma K, Nahmias Y, Tilles A, Berthiaume F, Martin L, Yarmush M. Homogeneous differentiation of hepatocyte-like cells from embryonic stem cells: applications for the treatment of liver failure. FASEB J. 2008;22:898–909. doi: 10.1096/fj.06-7764com. [DOI] [PubMed] [Google Scholar]
  4. Choi D, Oh HJ, Chang UJ, Koo SK, Jiang JX, Hwang SY, Lee JD, Yeoh GC, Shin HS, Lee JS, Oh B. In vivo differentiation of mouse embryonic stem cells into hepatocytes. Cell Transplant. 2002;11:359–368. [PubMed] [Google Scholar]
  5. Depreter M, Tytgat T, Beken S, Espeel M, De Smet K, Rogiers V, Roels F. Effects of extracellular matrix on the expression of peroxisomes in primary rat hepatocyte cultures. J Hepatol. 2000;32:381–391. doi: 10.1016/s0168-8278(00)80387-6. [DOI] [PubMed] [Google Scholar]
  6. Dunn JC, Tompkins RG, Yarmush ML. Long-term in vitro function of adult hepatocytes in a collagen sandwich configuration. Biotechnol Prog. 1991;7:237–245. doi: 10.1021/bp00009a007. [DOI] [PubMed] [Google Scholar]
  7. Dunn JC, Yarmush ML, Koebe HG, Tompkins RG. Hepatocyte function and extracellular matrix geometry: long-term culture in a sandwich configuration. FASEB J. 1989;3:174–177. doi: 10.1096/fasebj.3.2.2914628. [DOI] [PubMed] [Google Scholar]
  8. Ehashi T, Miyoshi H, Ohshima N. Oncostatin M stimulates proliferation and functions of mouse fetal liver cells in three-dimensional cultures. J Cell Physiol. 2005;202:698–706. doi: 10.1002/jcp.20167. [DOI] [PubMed] [Google Scholar]
  9. Gouon-Evans V, Boussemart L, Gadue P, Nierhoff D, Koehler CI, Kubo A, Shafritz DA, Keller G. BMP-4 is required for hepatic specification of mouse embryonic stem cell-derived definitive endoderm. Nat Biotechnol. 2006;24:1402–1411. doi: 10.1038/nbt1258. [DOI] [PubMed] [Google Scholar]
  10. Hamazaki T, Iiboshi Y, Oka M, Papst PJ, Meacham AM, Zon LI, Terada N. Hepatic maturation in differentiating embryonic stem cells in vitro. FEBS Lett. 2001;497:15–19. doi: 10.1016/s0014-5793(01)02423-1. [DOI] [PubMed] [Google Scholar]
  11. Heo J, Factor VM, Uren T, Takahama Y, Lee JS, Major M, Fein-stone SM, Thorgeirsson SS. Hepatic precursors derived from murine embryonic stem cells contribute to regeneration of injured liver. Hepatology. 2006;44:1478–1486. doi: 10.1002/hep.21441. [DOI] [PubMed] [Google Scholar]
  12. Iwai M, Tanaka S, Mori T, Harada Y, Muramatsu A, Morikawa T, Kashima K, Fushiki S. Investigation of parenchymal cell differentiation in organotypic slice culture of mouse fetal liver under administration of sodium butyrate. Cell Biol Toxicol. 2002;18:147–156. doi: 10.1023/a:1015500219572. [DOI] [PubMed] [Google Scholar]
  13. Kamiya A, Gonzalez FJ, Nakauchi H. Identification and differentiation of hepatic stem cells during liver development. Front Biosci. 2006;11:1302–1310. doi: 10.2741/1884. [DOI] [PubMed] [Google Scholar]
  14. Koide N, Shinji T, Tanabe T, Asano K, Kawaguchi M, Sakaguchi K, Koide Y, Mori M, Tsuji T. Continued high albumin production by multicellular spheroids of adult rat hepatocytes formed in the presence of liver-derived proteoglycans. Biochem Biophys Res Commun. 1989;161:385–391. doi: 10.1016/0006-291x(89)91609-4. [DOI] [PubMed] [Google Scholar]
  15. Kumashiro Y, Asahina K, Ozeki R, Shimizu-Saito K, Tanaka Y, Kida Y, Inoue K, Kaneko M, Sato T, Teramoto K, Arii S, Teraoka H. Enrichment of hepatocytes differentiated from mouse embryonic stem cells as a transplantable source. Transplantation. 2005a;79:550–557. doi: 10.1097/01.tp.0000153637.44069.c6. [DOI] [PubMed] [Google Scholar]
  16. Kumashiro Y, Teramoto K, Shimizu-Saito K, Asahina K, Teraoka H, Arii S. Isolation of hepatocyte-like cells from mouse embryoid body cells. Transplant Proc. 2005b;37:299–300. doi: 10.1016/j.transproceed.2005.01.036. [DOI] [PubMed] [Google Scholar]
  17. Maguire T, Davidovich AE, Wallenstein EJ, Novik E, Sharma N, Pedersen H, Androulakis IP, Schloss R, Yarmush M. Control of hepatic differentiation via cellular aggregation in an alginate microenvironment. Biotechnol Bioeng. 2007;98:631–644. doi: 10.1002/bit.21435. [DOI] [PubMed] [Google Scholar]
  18. Maguire T, Novik E, Schloss R, Yarmush M. Alginate-PLL microencapsulation: effect on the differentiation of embryonic stem cells into hepatocytes. Biotechnol Bioeng. 2006;93:581–591. doi: 10.1002/bit.20748. [DOI] [PubMed] [Google Scholar]
  19. Miyashita H, Suzuki A, Fukao K, Nakauchi H, Taniguchi H. Evidence for hepatocyte differentiation from embryonic stem cells in vitro. Cell Transplant. 2002;11:429–434. [PubMed] [Google Scholar]
  20. Nahmias Y, Berthiaume F, Yarmush ML. Integration of technologies for hepatic tissue engineering. Adv Biochem Eng Biotechnol. 2007;103:309–329. doi: 10.1007/10_029. [DOI] [PubMed] [Google Scholar]
  21. Novik EI, Maguire TJ, Orlova K, Schloss RS, Yarmush ML. Embryoid body-mediated differentiation of mouse embryonic stem cells along a hepatocyte lineage: insights from gene expression profiles. Tissue Eng. 2006;12:1515–1525. doi: 10.1089/ten.2006.12.1515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Roy P, Washizu J, Tilles AW, Yarmush ML, Toner M. Effect of flow on the detoxification function of rat hepatocytes in a bioartificial liver reactor. Cell Transplant. 2001;10:609–614. [PubMed] [Google Scholar]
  23. Sharma NS, Shikhanovich R, Schloss R, Yarmush ML. Sodium butyrate-treated embryonic stem cells yield hepatocyte-like cells expressing a glycolytic phenotype. Biotechnol Bioeng. 2006;94:1053–1063. doi: 10.1002/bit.20936. [DOI] [PubMed] [Google Scholar]
  24. Soto-Gutierrez A, Kobayashi N, Rivas-Carrillo JD, Navarro-Alvarez N, Zhao D, Okitsu T, Noguchi H, Basma H, Tabata Y, Chen Y, Tanaka K, Narushima M, Miki A, Ueda T, Jun HS, Yoon JW, Lebkowski J, Tanaka N, Fox IJ. Reversal of mouse hepatic failure using an implanted liver-assist device containing ES cell-derived hepatocytes. Nat Biotechnol. 2006;24:1412–1419. doi: 10.1038/nbt1257. [DOI] [PubMed] [Google Scholar]
  25. Soto-Gutierrez A, Navarro-Alvarez N, Zhao D, Rivas-Carrillo JD, Lebkowski J, Tanaka N, Fox IJ, Kobayashi N. Differentiation of mouse embryonic stem cells to hepatocyte-like cells by co-culture with human liver nonparenchymal cell lines. Nat Protoc. 2007;2:347–356. doi: 10.1038/nprot.2007.18. [DOI] [PubMed] [Google Scholar]
  26. Teratani T, Quinn G, Yamamoto Y, Sato T, Yamanokuchi H, Asari A, Ochiya T. Long-term maintenance of liver-specific functions in cultured ES cell-derived hepatocytes with hyaluronan sponge. Cell Transplant. 2005a;14:629–635. doi: 10.3727/000000005783982611. [DOI] [PubMed] [Google Scholar]
  27. Teratani T, Yamamoto H, Aoyagi K, Sasaki H, Asari A, Quinn G, Terada M, Ochiya T. Direct hepatic fate specification from mouse embryonic stem cells. Hepatology. 2005b;41:836–846. doi: 10.1002/hep.20629. [DOI] [PubMed] [Google Scholar]
  28. Tilles AW, Berthiaume F, Yarmush ML, Tompkins RG, Toner M. Bioengineering of liver assist devices. J Hepatobiliary Pancreat Surg. 2002a;9:686–696. doi: 10.1007/s005340200095. [DOI] [PubMed] [Google Scholar]
  29. Tilles AW, Berthiaume F, Yarmush ML, Toner M. Critical issues in bioartificial liver development. Technol Health Care. 2002b;10:177–186. [PubMed] [Google Scholar]
  30. Tsutsui M, Ogawa S, Inada Y, Tomioka E, Kamiyoshi A, Tanaka S, Kishida T, Nishiyama M, Murakami M, Kuroda J, Hashikura Y, Miyagawa S, Satoh F, Shibata N, Tagawa Y. Characterization of cytochrome P450 expression in murine embryonic stem cell-derived hepatic tissue system. Drug Metab Dispos. 2006;34:696–701. doi: 10.1124/dmd.105.007674. [DOI] [PubMed] [Google Scholar]
  31. Yamamoto H, Quinn G, Asari A, Yamanokuchi H, Teratani T, Terada M, Ochiya T. Differentiation of embryonic stem cells into hepatocytes: biological functions and therapeutic application. Hepatology. 2003;37:983–993. doi: 10.1053/jhep.2003.50202. [DOI] [PubMed] [Google Scholar]
  32. Znoyko I, Sohara N, Spicer SS, Trojanowska M, Reuben A. Expression of oncostatin M and its receptors in normal and cirrhotic human liver. J Hepatol. 2005;43:893–900. doi: 10.1016/j.jhep.2005.04.020. [DOI] [PubMed] [Google Scholar]

RESOURCES