Development of a genetically modified hepatoma cell line with heat-inducible high liver function

Abstract

Hepatoma cells are a promising cell source for the construction of bioartificial liver (BAL) systems owing to their high proliferative capability. However, their low liver function compared with primary hepatocytes is a major problem. In a previous study, we established a genetically modified hepatoma cell line, Hepa/8F5, in which eight liver-enriched transcription factor (LETF) genes were transduced into mouse hepatoma Hepa1-6 cells using a drug-inducible transactivator system. These cells proliferate actively under normal culture conditions, meaning that large quantities can be prepared easily. When the overexpression of the LETFs is induced by the addition of an inducer drug, cell growth stops and cell morphology changes with concomitant high expression of liver functions. However, the liver functions largely depend on the presence of the inducer drug, which must be continuously added to maintain these enhanced functions. In the present study, we attempted to modify the method of induction of LETF overexpression in Hepa/8F5 cells to remove the requirement for continual drug addition. To this end, we constructed a system in which the artificial transactivator was transcribed and amplified under the control of a heat-shock protein promoter, and introduced the system into the genome of Hepa/8F5 cells. In our modified cell line, heat-triggered LETF expression was confirmed to induce high liver function. After drug-screening of transfected cells, we established a hepatoma cell line (Hepa/HS), which exhibited high, heat-inducible liver functions. The Hepa/HS cells may represent a new cell source for hepatic studies such as the construction of BAL systems.

Supplementary Information

The online version of this article (10.1007/s10616-021-00457-4) contains supplementary material, which is available to authorized users.

Introduction

The liver is the largest organ in the human body and is involved in numerous functions that are essential for survival. Liver transplantation is the most effective treatment for acute and fulminant liver failure, but donor shortages are a serious problem. The bioartificial liver (BAL) system comprises a bioreactor in which liver cells are embedded, and its utility as an alternative to donor liver transplantation has been investigated (García Martínez and Bendjelid 2018; Lee et al. 2016; Park and Lee 2005). The BAL system metabolizes xenobiotics, secretes bioactive substances, supports liver function from outside the body and may promote liver regeneration. The use of some BAL systems in clinical applications has been proposed, but these applications have not yet been fully explored, possibly because of the lack of suitable cells for the systems.

Although human primary hepatocytes are high functioning cells, they are difficult to culture and maintain their functions for long periods in vitro. Currently, pluripotent stem cells such as embryonic and induced pluripotent stem cells (ES and iPS cells, respectively) may be important sources for hepatic cells (Messina et al. 2020). However, these cells need to undergo differentiation to become functional hepatocytes. Recently, several small molecules have been reported to be able to convert rodent mature hepatocytes into hepatic progenitor cells (Katsuda et al. 2017). The generation of three-dimensional small hepatoblasts consisting of three progenitor cell (hepatocyte, vascular and mesenchymal) populations differentiated from iPS cells has also been reported (Takebe et al. 2017). However, stem/progenitor cell culture and differentiation are time-consuming and costly processes.

Hepatoma cells derived from cancerous liver may represent a source of cells for the BAL system due to their high proliferative capacity and ease of culture (Chamuleau et al. 2005; Hoekstra et al. 2011). However, the liver function of hepatoma cells is significantly reduced compared with primary hepatocytes. The introduction of a gene encoding glutamine synthetase has been reported to improve ammonia metabolism in HepG2 cells (Miyashita et al. 2000) and the introduction of genes encoding drug-metabolizing enzymes has also been reported in this cell line (Gómez-Lechón et al. 2017). However, as well as enhancing the specific function of hepatoma cells, enhancement of overall function is also necessary to be used in BAL systems. Liver-enriched transcription factor (LETF) plays a key role in liver development and maintenance of the hepatocyte phenotype through transcriptional regulation of liver-specific gene expression; overexpression of this factor has been suggested to enhance the hepatic functions of hepatoma cells (Gong et al. 2020). However, the creation of a liver phenotype through overexpression of LETF in hepatoma cells reduces the proliferative abilities of the cells; thus, the development of a system for controlling expression of LETF genes would be beneficial.

Tetracycline (Tet)-regulated expression systems have been widely used to regulate gene expression in vitro and in vivo (Gossen and Bujard 1992). In the Tet artificial transactivator system, the Tet repressor protein (tetR) from Escherichia coli or a reverse-mutated tetR (rtetR) is fused with a transcriptional activation domain (VP16) derived from the herpes simplex virus to form a tetracycline-regulated transactivator, Tet-transactivator (tTA) or reverse Tet-transactivator (rtTA), respectively, and the artificial transactivator is expressed in host cells. A synthetic promoter composed of a tetR-response element (TRE) sequence and a minimal cytomegalovirus (CMV) promoter region (PCMVmin) is placed upstream of the target gene. The transactivator recognizes the TRE sequence and induces high levels of target gene expression through binding to the sequence; the addition of Tet or its analog doxycycline (Dox) can be used to control the binding of the transactivator to the TRE sequence in order to regulate target gene expression.

The expression of heat shock protein 70B' (HSP70) is up-regulated in response to heat stress through the binding of heat shock factor (HSF) to a heat shock element (HSE) sequence (Cotto et al. 1996). Therefore, the HSP70 promoter has been used as a heat-inducible promoter. However, because the activity of the HSP70 promoter is not as strong as that of virus-derived promoters such as the CMV promoter, the artificial transactivator system has been incorporated to amplify its transcriptional activity (Ito et al. 2012; Yamaguchi et al. 2012, 2014).

We have previously established a genetically engineered mouse hepatoma cell line, Hepa/8F5, in which high liver function is induced by the overexpression of eight LETFs (Yamamoto et al. 2012, 2018). The Tet-transactivator system involving rtTA is used to control the expression of the eight LETFs, and liver function is induced only by the addition of the inducer drug, Dox. However, continuous addition is required to maintain liver function. The present study investigated the genetic modification of Hepa/8F5 cells to incorporate the HSP70 promoter for overexpression of LETF genes, with the aim of inducing liver function with heat shock. To evaluate the system, the heat shock response and liver function of the cells (Hepa/HS) were analyzed.

Materials and methods

Cell culture

Hepatoma (Hepa/8F5, Hepa/HS) cells were cultured in Dulbecco’s Modified Eagle’s Medium (high-glucose DMEM, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS) and antibiotics (0.1 mg/mL streptomycin sulfate and 100 U/mL potassium penicillin G) (Fujifilm Wako Pure Chemical, Osaka, Japan). The Hepa/8F5 cells (RCB4661) were obtained from Riken Bioresource Center (Tsukuba, Japan) and were usually cultured in 100-mm tissue culture dishes (Thermo Fisher Scientific, Rochester, NY, USA) at 37 °C in a 5% (v/v) CO₂ incubator.

Construction of transposon vector plasmid

A transposon vector based on the PiggyBac transposon was used to introduce the heat-inducible tTA expression/amplification system into Hepa/8F5 cells. The transposon vector plasmid (PB513B-1) was purchased from System Biosciences (Mountain View, CA, USA) and includes a puromycin resistance gene under the control of the elongation factor 1 alpha (EF1α) promoter to enable the selection of transduced cells. A DNA fragment encoding the HSP70B' promoter-tTA and TRE-PCMVmin-tTA expression units derived from our previous study (Yamaguchi et al. 2012) was synthesized (Genewiz Suzhou, Jiangsu, China) and ligated into the pUC57 plasmid to create pUC57/pHSP-tTA-TRE-tTA. For construction of the transposon vector, we ligated PB/pHSP-tTA-TRE-tTA-pEF1α-Puro^R, SpeI- and EcoRV-digested pUC/pHSP-tTA-TRE-tTA into SpeI- and NruI-digested PB513B-1.

Transfection and Hepa/HS clone selection

We seeded Hepa/8F5 cells into 100-mm tissue culture dishes. When the cell density reached 90% confluence, cells were transfected with the transposon vector (PB/pHSP-tTA-TRE-tTA-pEF1α-Puro^R) and PiggyBac transposase expression vector (System Biosciences) plasmids using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. After incubation at 37 °C for 6 h, the transfection mixture was removed and replaced with fresh medium. Cells were treated with 3 µg/mL puromycin (Thermo Fisher Scientific) for 10 days to prepare a transduced bulk of cells. Cell clones were established from the bulk by limiting dilution.

Heat treatment

Cells were seeded into 24-well tissue culture plates (Thermo Fisher Scientific) at a final density of 2.0 × 10⁴ cells/well in 1 mL medium per well. The next day (denoted day 0), plates were sealed and directly immersed in a calibrated water bath at 41, 43 or 45 °C for 0.5, 1 or 1.5 h. After heating in a water bath, plates were immediately returned to the incubator maintained at 37 °C and 5% CO₂. On day 5, we measured viable cell numbers and cytochrome P450 activity. Viable cell numbers were assessed by the Trypan Blue dye exclusion test. Cytochrome P450 activity was measured using a luminescence-based assay kit (P450-Glo, CYP3A4 assay with Luciferin-IPA; Promega, Madison, WI, USA) according to the manufacturer’s protocol.

Cell morphology observation

Cells were seeded in 6-well tissue culture plates (Thermo Fisher Scientific) at a final density of 1.0 × 10⁵ cells/well in 2 mL medium per well either with or without 0.1 µg/mL Dox (Sigma-Aldrich). The next day (denoted day 0), cells were exposed to heat treatment at 43 °C for 1 h. On day 5, cells were fixed in 4% paraformaldehyde for 15 min and permeabilized with phosphate-buffered saline (PBS) containing 0.1% Triton X-100 for 10 min. Samples were incubated with 4′,6-diamidino-2-phenylindole (DAPI, Sigma-Aldrich) for 15 min. After washing three times with PBS, cells were observed using a BZ-X810 fluorescence microscope (Keyence, Osaka, Japan).

Liver function analysis

Cells were seeded into 24-well tissue culture plates at a final density of 2.0 × 10⁴ cells/well in 1 mL medium per well with or without 0.1 µg/mL Dox. The next day (denoted day 0), cells were exposed to heat treatment at 43 °C for 1 h. The medium was changed every other day. On days 5, 7 and 9, the culture medium was collected for use in the albumin assay. Cells were trypsinized and viable cell number was assessed with the Trypan Blue dye exclusion test using a hemocytometer. Albumin concentration was measured by enzyme-linked immunosorbent assay (ELISA) using a commercially available kit (Mouse Albumin ELISA Quantitation Set; Bethyl, Montgomery, TX, USA) according to the manufacturer’s protocol and the rate of albumin secretion of the cells was calculated.

For the ammonia-removal assay, the medium was replaced with fresh medium (500 µL/well) containing 2 mM NH₄Cl (Fujifilm Wako Pure Chemical) on days 5, 7 and 9, and the cells were cultured for a further 6 h, after which the culture medium was collected. The concentration of ammonia in the medium was measured using a commercially available kit (Ammonia-Test; Wako, Fujifilm Wako Pure Chemical) according to the manufacturer’s protocol. The decrease in ammonia was calculated to determine the removal rate.

For the cytochrome P450 assay, the medium was replaced with fresh medium (300 µL/well) containing 3.0 µM luciferin substrate (Luciferin-IPA, Promega) on days 5, 7 and 9, and the cells were incubated for a further 1 h, after which the culture medium was collected. Luciferase (Luciferin Detection Reagent, Promega) was added to react with the luciferin that was generated by the CYP3A4 enzyme, and the luminescence intensity was measured using a Glo-max Luminometer (Promega). Cytochrome P450 (CYP3A4) activity was normalized to cell number.

Mitomycin C treatment

We seeded Hepa/HS cells into 24-well tissue culture plates at a final density of 2.0 × 10⁴ cells/well in 1 mL medium per well. The next day (denoted day 0), cells were heated at 43 °C for 1 h. The medium was changed every other day. On day 6, the medium was replaced with fresh medium containing 10 µg/mL mitomycin C (Nacalai Tesque, Kyoto, Japan). Cells were then incubated for 2 h, after which the medium was replaced with fresh medium. We analyzed viable cell numbers, albumin secretion rate and cytochrome P450 activity on day 9.

Statistical analysis

Data were compared using the student’s t-test, with P < 0.05 considered to indicate significant differences.

Results and discussion

Design of an expression system for liver-enriched transcription factors

In order for the BAL system to be suitable for clinical application, large quantities of cells with high liver functions must be able to be prepared at low cost. The functional expression and proliferative properties of hepatocytes are negatively correlated, meaning that it is difficult to obtain cells with both properties. The production of cells that are capable of switching between growth and functional expression under normal culture conditions is, therefore, very important for the development of a clinically useful BAL system. In a previous study, we established a transgenic mouse hepatoma cell line—Hepa/8F5—with enhanced typical liver functions such as albumin secretion, cytochrome P450 activity and ammonia removal (Yamamoto et al. 2012). These functions were induced through overexpression of eight LETF genes stimulated by the addition of Dox. Nine retroviral vectors containing tetracycline-inducible gene expression systems and LETF expression units were used for gene transduction into the genome of Hepa1-6 cells for inducible expression of LETF genes in Hepa/8F5 cells (Fig. S1a). One of the vectors conferred constitutive expression of rtTA, activated by the addition of Dox, while the others encoded eight LETF gene expression units under the control of an inducible promoter containing a binding site (TRE) for activated rtTA. Therefore, in the latter, induction and expression of LETFs occur in the presence of Dox, and Dox must be added continuously to maintain expression (Fig. S1b).

In the present study, we modified Hepa/8F5 cells to enable switching between proliferation and expression of proteins for liver function in response to heat by introducing a new heat-inducible gene expression/amplification system. To create a construct in which LETF gene expression was induced by heat treatment, we constructed a hybrid promoter system, which combines the tightly controlled heat inducibility of the HSP promoter with the high-level expression conferred by the Tet system and incorporates a transcriptional amplification system (Fig. 1a). When cells are exposed to heat shock, the HSP promoter is activated and tTA is expressed. The expressed tTA then binds to the TRE/PCMVmin promoter, which is downstream of the HSP promoter, and a transcriptional positive feedback loop is established, which further induces tTA expression. The high expression of tTA due to this feedback loop then induces high-level expression of the eight LETF genes under the control of the TRE/PCMVmin promoter. Therefore, this system leads to continuous expression of the LETF genes after the one-shot heat treatment (Fig. 1b).

Fig. 1.

Schematic diagram of the heat-inducible expression of liver-enriched transcription factor genes mediated by an artificial transactivator with transcriptional amplification. a Diagram of the structure of expression units contained within the transposon vector. b Induction of overexpression of liver-enriched transcription factor (LETF) in hepatoma (Hepa/HS) cells. (1) Heat activates the heat shock protein 70B' (HSP70B') promoter, (2) inducing tetracycline transactivator (tTA) expression under the control of the HSP70B'. (3) The tTA activates the tetracycline repressor protein-response element sequence and cytomegalovirus minimum promoter (TRE/P_CMVmin). (4–6) Expression of tTA under the control of the TRE/P_CMVmin promoter induces further expression of tTA, creating a transcriptional positive feedback loop. (7,8) Amplified expression of tTA also drives the expression of LETF genes under the control of TRE/P_CMVmin promoter. LETFx liver-enriched transcription factor gene, P_CMVmin CMV minimum promoter; P_EF1α EF1α promoter, P_HSP HSP70B' promoter, Puro^R Puromycin resistance gene; tTA Tet-responsive transactivator gene, TRE Tet-responsive element sequence

Determination of optimum heat treatment conditions

Figure 2a and b illustrate the number of viable cells and CYP3A4 activity, respectively, 5 days after heat treatment in each condition. Treatment at 41 °C did not significantly affect cell viability regardless of the treatment time, whereas treatment at 45 °C for 30 min caused a decrease in cell viability and cell death occurred after treatment for 1 h or more. The activity of CYP3A4 was significantly improved after treatment for any time at 43 °C, which was particularly noticeable after treatment for 1 h, and led to a 48-fold improvement. Therefore, the optimum condition to induce the expression of LETF was determined to be heat treatment at 43 °C for 1 h.

Fig. 2.

Determination of optimum heat treatment conditions. Cells were cultured at 37 °C (control) or exposed to heat treatment at 41, 43 or 45 °C for 0.5, 1 or 1.5 h. Five days after heat treatment, we measured a the number of viable cells and b and cytochrome P450 activity for each condition. Data are presented as means ± standard deviations (n = 3). *P < 0.05, **P < 0.01 vs. control by the Student’s t-test. C control (untreated), CYP3A cytochrome P450, N.D. not detected

Morphological observation of Hepa/HS cells

Transposon vectors insert a transgene randomly into the cell genome. Thus, transduced cells that are not cloned are considered to be a mixture of cells that will exhibit varying responses to heat and average expression levels. Therefore, we speculated that, after heat treatment, the cells with the highest levels of liver function could be isolated from cells into which the heat-inducible gene expression system had been introduced. Thirty-six clones were established by limiting dilution; Figure S2a and b illustrate the numbers of cells and CYP3A4 activity, respectively, at 5 days after heat treatment. Suppression of cell growth and enhancement of CYP3A4 activity were observed in many clones, but the level of enhancement differed between clones. Clone #13, in which CYP3A4 activity was particularly improved, was chosen for use in subsequent experiments and designated Hepa/HS.

The morphology and cell number of Hepa/8F5 and Hepa/HS cells with or without heat treatment are shown in Fig. 3a and b, respectively. Under normal conditions, Hepa/8F5 and Hepa/HS cells grew well, reaching 100% confluence on day 5 of culture with cell counts at this time of about 8 × 10⁵ and 1.4 × 10⁶ cells/well, respectively. The difference in cell number is a result of the different cell sizes. The specific growth rate of Hepa/HS cells before heat treatment was 0.034 h⁻¹, which was slightly higher than the specific growth rate of parental Hepa1-6 cells (0.024 h⁻¹). After heat treatment at 43 °C for 1 h, Hepa/8F5 cells were still proliferative, reaching 100% confluence on day 7 without any change in cell morphology. In contrast, heat treatment eliminated growth of Hepa/HS cells, with increased cell size and the appearance of binuclear cells observed, as has been reported in cultured primary hepatocytes (Bou-Nader et al. 2020). These results indicate that heat treatment induces overexpression of LETF in Hepa/HS cells, leading to the suppression of cell proliferation and hepatocyte-like morphological changes.

Fig. 3.

Microscopy images showing the morphology and growth of Hepa/8F5 and Hepa/HS cells. a Bright-field (upper) and fluorescence (lower) micrographs of Hepa/8F5 and Hepa/HS cells after 5 days of culture with no treatment (HS[−]/Dox[−]), with heat treatment (HS[+]/Dox[−]) or with the addition of doxycycline (HS[−]/Dox[+]). Expression of enhanced green fluorescent protein, which indicates the shapes of cells, and 4′,6-diamidino-2-phenylindole (DAPI) staining (blue) were merged in fluorescence micrographs. Scale bars = 100 µm. b Graphs of the total numbers of cells on days 5, 7 and 9 of culture with no treatment (circles), heat treatment (squares) or doxycycline addition (triangles). Data are presented as means ± standard deviations (n = 3). *P < 0.05, **P < 0.01 vs. untreated by Student’s t-test. Dox, doxycycline

Liver function analyses for hepatoma cells

In the Hepa/8F5 cells, albumin secretion improved following the addition of Dox, but not in response to heat treatment. In contrast, Hepa/HS cells exhibited increased albumin secretion following Dox addition, but also in response to heat treatment (Fig. 4a). In addition, the ammonia removal rate of Hepa/8F5 cells was negative under normal culture and heat treatment conditions (Fig. 4b), indicating that ammonia production may have been caused by amino acid metabolism. In contrast, ammonia in the medium of Hepa/HS cells was removed in response to Dox addition or heat treatment. In addition, the CYP3A4 of Hepa/HS cells was the same following heat treatment as for Dox addition (Fig. 4c). These results indicate that LETF overexpression in Hepa/HS cells can be induced by the introduction of a heat-inducible gene expression/amplification system.

Fig. 4.

Graphs showing the results of liver function analyses of Hepa/8F5 and Hepa/HS cells. We analyzed a the albumin secretion rate, b the ammonia removal rate and c cytochrome P450 activity on days 5, 7 and 9 of culture for untreated (circles), heat treated (squares) or doxycycline exposed cells (triangles). Data are presented as means ± standard deviations (n = 3). *P < 0.05, **P < 0.01 vs. untreated by Student’s t-test

In the liver, LETFs are highly expressed and complex regulatory mechanisms are involved in the development of liver function. The LETFs form a transcriptional network which regulates various target genes related to liver function as well as the expression of the LETFs themselves and mutual genes, which together create and maintain the liver phenotype. We previously reported quantitative measurement of the expression of hepatic-function-related genes such as ALB, TTR, AAT, TAT, G6P, Cps1, ApoAI, ApoB, ASGPR1, PPAR and GATA4 in Hepa/8F5 cells cultured in Dox-containing medium (Yamamoto et al. 2018). The expression of these genes was upregulated in Hepa/8F5 cells cultured with Dox compared with parental Hepa1-6 cells or Hepa/8F5 cells cultured without Dox. In the present study, Hepa/HS cells exhibited enhanced liver functions after heat treatment because expression tTA was induced and the protein binds to the TRE sequence to activate the same promoter as is activated by Dox, ultimately resulting in the induction of LETF gene expression. It is likely that the high liver function we observed in the present study was a result of these cells having a high degree of liver maturation.

The liver function of Hepa/HS cells was lower on day 9 of culture in those that were exposed to heat treatment compared with Dox. Mitomycin C treatment was performed to suppress proliferating cells with insufficient induction after heat treatment. Albumin secretion and CYP3A4 activity levels could be maintained for 9 days by initiating mitomycin C treatment on day 6 of culture (Fig. 5). These results suggest that transcriptional amplification of tTA due to heat treatment was not sufficient, and that some cells were unable to switch from the proliferative state to the liver-function-expressing state, meaning that proliferation continued gradually and overall liver function was decreased. Nevertheless, most cells in which tTA induction had worked well after heat treatment retained liver function even on day 9. Hepa/HS cells showed sufficient liver function on day 5 after heat treatment, and maintained high levels of liver function 2 weeks after heat treatment.

Fig. 5.

Bar graphs illustrating the effect of mitomycin C treatment on heat-induced liver function of hepatoma cells in prolonged culture. We measured a the number of cells, b the albumin secretion rate and c cytochrome P450 activity on day 9 of culture with (MC[+]) or without (MC[−]) mitomycin C treatment. Mitomycin C treatment was performed on day 6 after heat treatment. Data are presented as means ± standard deviations (n = 3). *P < 0.05, **P < 0.01 vs. untreated by Student’s t-test. CYP3A cytochrome P450, MC mitomycin C, RLU relative light units

It has been reported that the liver function of hepatoma cells can be improved and maintained for a long time through the use of three-dimensional culture on structures such as spheroids (Yamamoto et al. 2018; Sarkar et al. 2017; Fujii et al. 2020). We have previously improved the albumin secretion, ammonia removal and cytochrome P450 activity of Hepa/8F5 cells through three-dimensional culture using macroporous gelatin beads (Tonello et al. 2017). In this case, it was necessary to generate tissue formation by using cells treated with Dox; however, sufficient liver function could not be induced when Dox was added after the formation of the three-dimensional tissue-like structure. This is presumably because cells inside the three-dimensional tissue could not uptake sufficient amounts of Dox, and LETF expression was not induced in these cells. In Hepa/HS cells, liver function is induced by heat treatment; thus, effective induction of liver function will be possible in high-density three-dimensional culture, which is required in the bioreactors of BAL systems.

In conclusion, Hepa/HS cells proliferate easily under normal culture conditions and high liver function can be induced by heat treatment, which does not require the addition of expensive cytokines or low-molecular-weight drugs. Therefore, these cells may represent a useful cell source for BAL systems or drug screening.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This work was financially supported in part by Grants-in-Aid for Scientific Research (no. 26560216 and 20H00322) from the Japan Society for the Promotion of Science (JSPS).