Development of a High-Throughput Human HepG 2 Dual Luciferase Assay for Detection of Metabolically Activated Hepatotoxicants and Genotoxicants

Abstract

Hepatic toxicity remains a major concern for drug failure; therefore, a thorough examination of chemically induced liver toxicity is essential for a robust safety evaluation. Current hypotheses suggest that the metabolic activation of a drug to a reactive intermediate is an important process. In this article, we describe a new high-throughput GADD45β reporter assay developed for assessing potential liver toxicity. Most importantly, this assay utilizes a human cell line and incorporates metabolic activation and thus provides significant advantage over other comparable assays used to determine hepatotoxicity. Our assay has low compound requirement and relies upon 2 reporter genes cotransfected into the HepG₂ cells. The gene encoding Renilla luciferase is fused to the CMV promoter and provides a control for cell numbers. The firefly luciferase gene is fused to the GADD45β promoter and used to report an increase in DNA damage. A dual luciferase assay is performed by measuring the firefly and Renilla luciferase activities in the same sample. Results are expressed as the ratio of the 2 luciferase activities; increases over the control are interpreted as evidence of stress responses. This mammalian dual luciferase reporter has been characterized with, and without, metabolic activation using positive and negative control agents. Our data demonstrate that this assay provides for an assessment of potential toxic metabolites, is adaptable to a high-throughput platform, and yields data that accurately and reproducibly detect hepatotoxicants.

Sensitive, reproducible, and stable human hepatic cell lines that reflect human hepatocyte metabolism and toxicity are of significant value in the early high-throughput screening armamentarium. Such systems provide for the early detection of potential toxicity in drug discovery and facilitate prioritization of compounds for animal studies. ^1,2 Prediction of drug-induced hepatotoxicities by in vitro assays remain very challenging to the pharmaceutical industry. ^1,2 Our current understanding of mechanisms of drug-induced hepatoxicity suggests 2 major types of mechanisms: immunological and metabolic. ^3–6 It is desirable to have a screening assay that incorporates metabolic activation, is reproducible, has a low cost, and is amenable to high-throughput methods. Hepatocytes are frequently used for metabolic activation and provide a valuable early screen. However, using fresh hepatocytes suffers from a number of technical challenges: frozen cells have limited shelf life, poor availability of human liver, high cost, and significant variability among human hepatocyte preparations. ⁷ Although mammalian cell lines are readily available, the principal drawback of mammalian cell lines is that they lack metabolic activation; thus, they are unable to detect metabolism-mediated toxicity. ⁸ Problems with metabolic competence can be overcome by adding liver S-9 fractions and cofactors. Liver S-9 fractions are subcellular fractions that contain drug-metabolizing enzymes including the cytochromes P450, flavin monooxygenases, and UDP glucuronyl transferases. Liver S-9 fractions are a major tool for studying xenobiotic metabolism.

Several publications have described the development of mammalian cell-based genotoxicity test systems in which the DNA damage inducible promoter of the GADD45α gene fused to the green fluorescent protein (GFP). ^1,9 Both GADD45α and GADD45β proteins show strong sequence similarity to bacterial RecA, which promotes the pairing of homologous DNA molecules and strand exchange reactions that initiate general genetic recombination. ¹⁰ GADD45β belongs to the GADD45 nuclear protein family that is expressed ubiquitously in mammalian tissues. ¹¹ In mammalian cells, GADD proteins have an important role in the maintenance of genomic stability. ^12,13 The GADD genes represent a subset of cDNA which were originally isolated on the basis of rapid induction in hamster cells by UV radiation (scheme 1). ^12,14 Recently, it has been shown that NF-κB, but not the MAPK signaling pathways, is involved in the in vivo regulation of GADD45β expression. ¹¹ Thus, NF-κB signaling involves induction of GADD45β expression, which supports the proposed role of GADD45β in protecting cells against DNA damage under various stress conditions. ¹¹ Given the above, GADD45β is thought to be a good predictive hepatic stress biomarker for DNA repair, cell cycle checkpoints, and apoptosis in response to extrinsic stress. ^14–16 In the current study, we have used GADD45β as the biomarker of inducible extrinsic stressors and the CMV promoter as the constitutive control in a dual luciferase reporter assay.

For several reasons, GFP is a nonideal reporter for this assay. Although mammalian cells expressing a GFP tag offer a fluorescence assay for gene expression, GFP signal will invariably be contaminated with endogenous cell or medium-derived autofluorescence. ^17–19 In addition, the limited metabolic capability of mammalian cells requires the use of autofluorescent nicotinamide nucleotides and liver subcellular fraction (S-9) for the activation of most promutagens. Due to such autofluorescence interference, certain kinds of promutagens cannot be detected by GADD45α-GFP without extra precipitation and washing steps. ^9,20 In order to overcome the autofluorescence limitations, we employed dual luciferase genes as the reporters. This dual luciferase assay provides several advantages for measuring gene expression in HepG₂ cells. First, the sensitivity of the luciferase assay allows the reporter to be integrated into the genome and still provide measurable signals. Second, the presence of a control reporter in the genome, expressed from a constitutive promoter, provides a control for cell number. Finally, it has been shown that exposure of S-9 greater than 0.4% in culture medium for 4 days was cytotoxic to all types of neurons in the cultures. ²¹ Due to this cytotoxicity effect, most of the metabolism-mediated assays are using 1% S-9 with extra washing steps in the early stages of drug discovery. ²⁰ However, the amount of toxic metabolite formed might be too low to cause any measurable toxic effect because of the metabolic activity for 1% S-9 is low. In order to overcome the cytotoxicity limitations, we used the Transwell permeable support plate (scheme 2) (Corning, Acton, Massachusetts), and we did not observe cytotoxicity effect up to 20% S-9 treatment.

We evaluated GADD45β-based induction of luminescence using 57 model compounds. The test compounds included hepatotoxicants and DNA-damaging agents, many of which are well-known reagents employed as controls in hepatotoxicity tests. Our data show that the assay has a high level of concordance with in vivo assays and suggests that this mammalian dual luciferase reporter system affords a new tool to identify metabolically activated hepatotoxicants and genotoxicants.

Materials and Methods

All the positive controls were handled in accordance with National Institutes of Health (NIH) Guidelines for the Laboratory Use of Chemical Carcinogens. ²² All solvents and chemicals were purchased from Aldrich Chemicals (Milwaukee, Wisconsin) or Sigma (St Louis, Missouri), and the purities are greater than 99%. Aroclor1254-induced rat liver S-9 homogenate was obtained from Molecular Toxicology Inc (Boone, North Carolina). The dual luciferase assay kit (No. E1960) was purchased from Promega Corp (Madison, Wisconsin).

Cell Lines and Growth Medium

The human HepG₂ cell line was purchased from ATCC (Manassas, Virginia) and maintained in MEM supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, Carlsbad, California) and 1% penicillin-streptomycin (Invitrogen) at 5% CO₂ at 37°C. The cells were grown for 24 to 48 hours to achieve logarithmic growth and treated with various concentrations of test compounds or DMSO in fresh medium. Compounds were freshly dissolved in DMSO, and the final DMSO concentration was 0.5%.

Luciferase Reporter Plasmids

The HepG₂ assay contains 2 plasmid vectors pGL4.82 [hRluc/Puro] and pGL4.17 [luc2/Neo] ordered from the Promega Corp. The vector pGL4.82 [hRluc/Puro] was digested with BglII, and pGL4.17 [luc2/Neo] was digested with XhoI and HindIII. To generate a CMVpGL4.82 [hRluc/Puro] promoter fusion construct, the CMV gene sequence was amplified by PCR from pRL-CMV plasmids (Promega Corp). The CMV primers used were forward: 5’-CCTCGAGGATATCAAGATCTTCAATATTGGCCATTAGCCAT-3’ and reverse: 5’-CCGAGGCCAGATCTCCTGTGGAGAGAAAGGCAAAGT-3’. Both the forward and reverse primers incorporate a BglII site into the amplified product. The GADD45β-pGL4.17 [luc2/Neo] reporter construct was amplified by PCR from HepG₂ genomic DNA using primers by introducing an XhoI site at −0.9 kb of the GADD45β 5’ NTS and taking advantage of a HindIII site at the third codon. The primers used were forward: 5’-TCGCTAGCCTCGAGGGTGACAGCTGATGTGTATT-3’ and reverse: 5’-GGATTGCCAAGCTTCGTCATGTTGCAATTATAATCCAC-3’. The final vectors bear the CMV promoter in front of the Renilla luciferase reporter, and the GADD45β gene fused with firefly luciferase reporter. All the sequences were verified using the dye terminator cycle sequencing method on an ABI PRISM (Applied Biosystems, Foster City, California).

Construction of Luciferase Reporter Stably Transfected HepG₂ Cells

Two 6-well plates of 50% confluent HepG₂ cells (passage 10) each were transfected with the plasmids CMV-pGL4.82 [hRluc/Puro] (1 μg/well) and GADD45β-pGL4.17 [luc2/Neo] (1 μg/well). Transfection of the plasmids into the HepG₂ cells was performed sequentially using Fugene 6 (Roche Diagnostics, Basel, Switzerland), according to the manufacturer’s instructions. The transfection mix (100 μL) was carried out overnight (15 hours). After selection, cells were cultured in maintenance medium, which consisted of standard medium supplemented with 200 μg/mL geneticin and 1 μg/mL puromycin.

GADD45β Dual Luciferase Assay

Assays were carried out in 96-well white microplates (Dynex Technologies, Chantilly, Virginia). Before treatment with hepatotoxicants, genotoxicants, and negative controls, HepG₂ cells were collected and reseeded at 5 × 10⁴ cells/mL (either 100 μL for without S-9 treatment or 175 μL for with S-9 treatment) into a 96-well, clear plate (Becton Dickinson Labware, Franklin Lakes, New Jersey). All the test compounds were dissolved in 10% DMSO to make 2 × 10⁴, 1 × 10⁴, 2 × 10³, 1 × 10³, 2 × 10², and 1 × 10² μg/mL. For treatment in the absence of S-9 activation, the test compounds were added directly into HepG₂ cells (10 μL of compound dilutions in 10% DMSO to make final concentrations of 1000, 500, 100, 50, and 10 μg/mL in the incubation systems) combined with 90 μL of MEM growth medium. For treatment with S-9 activation, 10 μL of test compounds were added to the upper chamber of Transwell permeable support plates (scheme 2) (Corning) combined with 15 μL of 10% S-9 mix (2.7 mg/mL isocitrate acid, 1.5 mg/mL NADP, 1.5 μL of 20 mg/mL S-9, and MEM medium). Cultured HepG₂ cells were plated in the lower chambers. Because previous results have shown that DMSO increases yeast DEL recombination, ²³ the final concentration for DMSO in the medium was only 0.5% in both experiments with and without S-9. After 24- and 48-hour treatment, cells were harvested in 100 μL passive lysis buffer (Promega luciferase kits). Luciferase activity was determined with 90 μL aliquots of cell lysates mixed with 25 μL of the luciferase assay solution and read on a Tecan Ultra-384 (Tecan UK Ltd, Theale, United Kingdom) in a 96-well plate format. Changes in induction ratio (fold changes) were calculated using the equation (F = firefly, R = Renilla): $$\text{Fold} \text{Change}=\frac{{\text{F}}_{\text{treated}}/{\text{R}}_{\text{treated}}}{{\text{F}}_{\text{DMSO}}/{\text{R}}_{\text{DMSO}}}.$$ The results are expressed as the mean ± SD.

Determination of Catalytic Activity of Aroclor1254-Induced Rat Liver S-9

The HepG₂ GADD45β cells with S-9 mix were treated with 2-aminoanthracene (2AA) and 2-acetylaminofluorene (2AAF) for 6 hours, and the metabolites from the lower compartment medium were collected and analyzed by LC-MSⁿ. The HPLC analysis was performed using a Luna C18 column (2.0 × 50 mm, 5 μm; Phenomenex, Torrance, California). A gradient with buffer A (0.1% HCOOH in water) and buffer B (0.1% HCOOH in acetonitrile) was used at a flow rate of 0.5 mL/min. The gradient was held at 3% B for the first 2 minutes, then linearly increased from 3% to 97% B over 14 minutes, then held at 96% B for 2 minutes, and then from 97% to 3% B over 2 minutes, followed by re-equilibration at 3% B for 2 minutes. The MSⁿ spectra of 2AA and 2AAF metabolites were obtained using an LTQ mass spectrometer (Thermo Electron Corporation, Waltham, Massachusetts). The mass spectrometer was operated in positive ion electrospray mode with a capillary temperature of 300°C. The MSⁿ analyses were performed with relative collision energy of 25%.

Data Analysis

For each compound, data from 3 independent experiments were collected, and the mean and standard deviation were calculated. If the test compounds showed potential cytotoxicity (both firefly and Renilla signals are lower than 2-fold of background) or precipitation in more than 3 concentrations (totally, 6 concentrations were screened), further dilutions were tested. A compound was considered hepatotoxic in the luciferase assay if it caused a significant GADD45β induction (P < .001) as compared to the DMSO control and the observed increase was concentration dependent. A compound was considered equivocal if the significant induction (P < .001) was only shown at the highest concentrations. A compound was considered negative if it did not satisfy the above mentioned criteria but led to a cytotoxic effect or precipitation was found in the medium within the treated concentration range. Methylmethane sulfonate (MMS), at concentrations of 17 and 65 μg/mL, served as a positive control for experiments performed in the absence of metabolic activation. Cyclophosphamide (CP) at concentrations of 18 and 75 μg/mL was used as positive control for studies in the presence of metabolic activation.

Results

Background

Firefly luciferase, under the control of various promoters, has been employed in cell-based assays for the detection of heavy metals, such as cadmium and lead, and aromatic organics. ^24,25 The high sensitivity (subattomole level), broad dynamic range (7–8 orders of magnitude), simplicity, and no endogenous activity in mammalian cells are primary advantages of firefly luciferase reporter. ²⁴ An important criterion for many reporter assays is the background signal. To evaluate background luminescence resulting from the spontaneous oxidation of luciferin with components in the cell lysate and S-9, we performed both firefly and Renilla luciferase assays using the HepG₂ cells. The medium background signals for the firefly luciferase assay (mean ± SD) with and without S-9 were 88 ± 33 and 106 ± 41 relative light units (RLU), respectively. Renilla luciferase assay yielded 96 ± 41 RLU with S-9 and 81 ± 28 RLU without S-9. The background signals for the HepG₂ cells in lysis buffer without adding luciferase substrate were 117 ± 29 RLU without S-9 and 109 ± 37 RLU with S-9.

Sensitivity and Reproducibility

Other important criteria for reporter assays are sensitivity, reproducibility, and linearity. The luciferase cells were grown to exponential phase in MEM growth medium at 37°C. The cell numbers were quantified by hemocytometer, serial dilutions were prepared, and between 500 and 6000 cells were used per assay for both firefly and Renilla luciferase activities. Our data showed that both the firefly and Renilla luciferase activities increased linearly with increasing numbers of cells (Figure 1A and B), and the correlation of firefly to Renilla luciferase activities was very strong (Figure 1C). In general, the responses from lower cell concentrations would fluctuate more. Therefore, 4000 to 6000 cells per well were employed for validation studies. The HepG₂ dual luciferase assay was also tested with different concentrations of S-9 (0%, 1%, 2%, 5%, 10%, and 20%) with 8 replicates for 48 hours (Figure 1D and E). The selection of concentration range for S-9 is based on those used in the Ames assay. ²⁶ Although adding S-9 fractions might raise concerns about the reproducibility of the firefly/Renilla ratios, our data demonstrate that those theoretical concerns do not cause any practical problems with this assay (Figure 1D and E). Therefore, the dual luciferase reporter assay is useful as an early screen for predicting the risk of hepatotoxicity and genotoxicity studies with respect to sensitivity, reproducibility, and linearity.

Effect of 2AA on Induction of Firefly Luciferase

We selected 2AA as a model compound for optimizing the metabolic activation conditions using Aroclor1254-induced liver S-9. 10% and 20% S-9 caused similar inductions of luminescence with test concentrations of 2AA (Figure 2A). Thus, 10% S-9 was chosen as the working concentration. Figure 2B showed that 10% S-9 and 2AA caused a dose-dependent stimulation of GADD45β gene expression. The LC-MSⁿ results showed a similar trend as found in the luciferase experiment. At 10% S-9, both 2AA and 2AAF exhibited the oxidative metabolites (data not shown).

Effect of MMS and CP on Induction of Firefly Luciferase

The system was optimized by using the alkylating agent MMS, which mutates DNA and promotes activation of the intra-S-phase checkpoint response. ^27,28 We selected CP as a validation compound for the metabolic activation conditions using 10% Aroclor1254-induced liver S-9. The mutagenic potencies of MMS and CP were measured by serial dilutions of MMS or CP in the exponentially growing cells, with or without S-9 activation. Figure 3A and B demonstrate that both MMS and CP cause a dose-dependent stimulation of GADD45β gene expression.

GADD45β Induction by Noncarcinogenic Pharmaceuticals

In order to examine the ability of the dual luciferase assay to detect hepatotoxicants with and without metabolic activation, we assessed the levels of GADD45β induction using a variety of reference compounds. Tables 1 and 2 list chemicals that were tested in the HepG₂ cell line. The dual luciferase assay was performed in parallel, with and without metabolic activation (Aroclor1254-induced rat liver S-9), using the same cell clone as the tester cells. MMS was employed as a positive control for the induction of GADD45 in the absence of S-9 metabolic activation. CP was used as a positive control for the S-9 metabolic activation luciferase studies. The background luminescence was measured using 0.5% DMSO treatment. Our data showed that among the hepatotoxic drugs tested, valproic acid, tacrine, albendazole, carbamazepine, amiodarone, digoxin, cyclosporine A, gemfibrozil, nefazodone, phenytoin, and troglitazone showed significant (P < .001) induction of GADD45β. On the other hand, chlorpromazine, acetaminophen, and haloperidol showed GADD45β induction with S-9 metabolic activation but no induction without S-9 activation. These results were consistent with the previous studies that chlorpromazine, acetaminophen, and haloperidol can form toxic metabolites in vivo and further cause liver injury. ^2,29–31 For the pharmaceutical classes of antiestrogens, GADD45β was assessed with mifepristone, tamoxifen, and raloxifene. All 3 compounds showed equivocal results as the high levels of cytotoxicity interfered with the assessment of GADD45β induction and prevented the ability to obtain clear results. Equivocal results were also obtained for quinidine, flutamide, propranolol, terfenadine, and tetracycline. No induction was observed for isoniazid and dapsone.

GADD45β Induction by Other DNA-damaging Agents

Genotoxicants and nongenotoxicants were also tested in the HepG₂ dual luciferase assay (Tables 3 and 4). All of the direct-acting mutagens including MMS, EMS, MNNG, sodiumazide, 2-nitrofluorene (2-NF), and 4-nitroquinoline-N-oxide (4NQO1) were shown to be positive in our system. The cross-linking agents (mitomycin C and cisplatin) were also evaluated and showed GADD45β induction. We also tested clastogens that are negative in the Ames assay, including etoposide, vinblastin, arsenic III oxide, bleomycin, taxol, araC, chlorambucil, cytochalasin B, and actinomycin D. All of these compounds exhibited a dose-dependent induction of GADD45β (Tables 3 and 4). Because the dual luciferase assay could detect promutagens, we selected 2AA, 2AAF, 4-acetylaminofluorene (4AAF), dimethylnitrosamine, benzo[a]pyrene (BaP), and CP as model compounds for metabolic activation evaluation. All of these compounds except BaP exhibited a dose-dependent induction of GADD45β (Figure 4). Finally, HCl, NaOH, MeOH, urethane, caffeine, benzyl chloride, butylated hydroxytoluene, nalidixic acid, methyl viologen, and DMSO did not show any GADD45β induction in the dual luciferase assay.

Discussion

Drug-induced hepatotoxicity is of great concern to the pharmaceutical industry. The biotransformation of xenobiotics to reactive or toxic metabolites is considered to be a major cause of adverse drug reactions observed in the clinic. ³² Of the 16 drugs withdrawn from the market in recent years, 4 (25%) were due to hepatotoxic effects. ³³ Also, it has been estimated that liver injury is the principal reason for toxicity-related attrition in clinical drug trials and withdrawal of drugs postmarketing. ³⁴ As a result of this, both regulatory authorities as well as the pharmaceutical industry are currently paying significant attention to metabolism-mediated toxicity. ^2,32,35 Therefore, it is important to identify the risk of hepatotoxicity in the early discovery phase of compound development to facilitate compound selection decisions. ³⁴

Altered expression of specific genes is one of the hallmarks of the DNA-damage response in living cells and plays an essential role in cell survival and maintaining the stability of the genetic material. ¹⁵ Both eukaryotic and prokaryotic cells activate transcription of many genes following treatment with agents that cause stress and DNA damage. A rapidly growing application of gene expression reporter assays is to study the effects of chemical exposure on gene regulation by monitoring expression of toxicity markers such as those associated with tumorigenesis, cytokine release, and transcriptional activation which relate to carcinogenicity, mutagenicity, inflammation, and endocrine disruption. ^36,37 The advantage of in vitro cell-based reporter assays is their high specificity, selectivity, and rapid reaction times. ³⁸ Expression of a reporter gene produces a measurable signal, which can be readily distinguished from background. However, despite their potential, reporter assays are limited in their capabilities as they are usually conducted in vitro and therefore cannot predict the interactions that may occur in vivo. Moreover, many responses are tissue, species, and time specific, and therefore a single in vitro reporter gene assay may not accurately predict the responses observed in vivo. ^39,40 Despite the pitfalls of reporter gene assays, the advance of toxicogenomic technologies and understanding will continue to evolve, allowing development of more sensitive and simpler reporter gene vectors, thereby increasing our ability to design improved predictive models for application of early genetic toxicity determination.

We have recently reported on a yeast-based assay that incorporates metabolic activation and allows for the high-throughput detection of potential genotoxicants and clastogens. ⁴¹ There is controversy concerning the use of yeast compared to the use of mammalian cells. ⁹ Therefore, new approaches using dual luciferase reporters have been created to overcome these problems by using human liver HepG₂ cells. HepG₂ cells were selected as target cells for evaluating liver toxicity because they are derived from human liver and have been extensively used as the test system for the prediction of toxicity, carcinogenicity, and cell mutagenicity in humans. ^2,42 In addition, HepG₂ cells do retain the expression of some enzymes relevant to metabolism, including some phase II enzymes. ⁴³ The transcript levels of CYP1A1, 1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, and 3A4 had been measured in HepG₂ cells with quantitative PCR. ⁴⁴ Results showed that transcripts of all CYPs were present in HepG₂ cells; however, mRNA levels of most CYPs were dramatically lower than in primary human hepatocytes. ⁴⁴ As a consequence of the low levels of CYPs in HepG₂ cells, hepatotoxicity of several compounds might have been missed or underestimated as compared with the primary human hepatocytes. In order to overcome the HepG₂ phase I metabolic activity limitation, we used the Aroclor1254-induced rat liver S-9 with Transwell permeable support plates (scheme 2) (Corning). Compared to primary human hepatocytes, our HepG₂ cell system is a relatively easy-to-handle tool to study the metabolism-mediated hepatotoxicants. In our system, cells were transfected with a DNA damage inducible gene and a control vector. This dual luciferase reporter approach enables a direct measure of stress-related response and potential cytotoxicity in the same human cell lines. Thus, the inherent fundamentals of the biological microenvironment are better conserved. Additionally, as the mammalian cell system is devoid of cell walls, it more closely resembles the in vivo situation as compared to bacterial and yeast systems.

In the present study, 57 compounds were tested for the induction of GADD45β. These compounds have been studied previously for either hepatotoxicity or genotoxicity (Tables 2 and 4). Even though most of the drugs listed in Table 2 gave positive results only at certain levels of cytotoxicity, acetaminophen, albendazole, amiodarone, and cyclosporin A showed GADD45 induction at noncytotoxic levels, which suggests the positive results are not due to the high concentration tested in the system. In Table 4, our data suggest that this dual luciferase reporter system can identify metabolically activated mutagens and genotoxins. Among the chemicals tested, EMS, 4-NQO1, DMN, CP, 4AAF, 2AAF, 2AA, cisplatin, etoposide, arsenic III oxide, and chlorambucil were positive at noncytotoxic levels. An unexpected finding of this study was that BaP did not show any induction in the GADD45β dual luciferase assay. The reasons for lack of agreement with the traditional in vitro assays might arise from the complex mechanisms of DNA damage sensing that require numerous cellular pathways for the final biological consequences. Thus, a single model organism, gene, or cell line can offer only modest predictive power for hepatotoxicity in general; but when used in conjunction with other assays, the majority of toxicity risks can be identified and addressed preclinically.

Finally, the expression of drug transporters within HepG₂ cells is not fully explored and hence may impact upon the disposition and action of test compounds. Another clear direction for future assay development will focus on large-scale validation. The large-scale validation could provide more information and improvement for the criteria applied in this assay, making the assay suitable for the detection of hepatotoxicity in early drug discovery screening.

In summary, induction of the GADD45β gene has been monitored using the firefly luciferase reporter assay. Previous studies have shown that GADD45β expression is controlled at the level of transcription, with levels varying in response to DNA damage and extrinsic stress. ^45,46 Therefore, the stable transfection of HepG₂ cells with dual reporter constructs could be a useful addition for the early predictive of hepatotoxicity. In addition, it may also have value as an additional assay for the detection of genotoxicants in a mammalian-based system. Further work would be required to assess the potential of this assay to detect idiosyncratic toxicity. This simple and reproducible, high-throughput method may be useful in minimizing the number of hepatotoxicants that are advanced into clinical trails.