Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 1.
Published in final edited form as: Toxicol In Vitro. 2025 May 26;108:106085. doi: 10.1016/j.tiv.2025.106085

Development of a TK6-derived cell line expressing four human cytochrome P450s for genotoxicity testing

Xilin Li a, Yuhan Wang a, Hannah Xu a, Xiaobo He a, Si Chen a, Xiaoqing Guo a, Mugimane G Manjanatha a, Tong Zhou b, Jessica Bonzo c, Nan Mei a,*
PMCID: PMC12359084  NIHMSID: NIHMS2104545  PMID: 40436327

Abstract

Metabolism is essential for in vitro genotoxicity testing. We previously developed a panel of TK6 cell lines, each expressing one of 14 human cytochrome P450 (CYP) enzymes, demonstrating their ability to effectively bioactivate indirect genotoxicants without relying on a rodent liver S9 fraction. In the present study, we extended this work by developing a TK6 cell line co-expressing four human CYP enzymes, including CYP2A6, CYP2E1, CYP2C19, and CYP3A4 (designated as TK6-4CYP), and subsequently assessed its capability to metabolize and activate pro-genotoxicants. Human lymphoblastoid TK6 cells were sequentially transduced with lentiviral vectors carrying CYP2A6/2E1 and CYP2C19/3A4, resulting in more than a 210-fold increase in mRNA expression levels for each CYP compared to parental cells. RNA sequencing revealed selective upregulation of the four CYPs. Their protein expression and enzymatic activities were also confirmed. TK6-4CYP cells were subsequently tested with four CYP-metabolized pro-genotoxicants, including N-nitroso-diethylamine (NDEA) metabolized by CYP2A6, N-nitroso-dimethylamine (NDMA) by CYP2E1, N-nitroso-propranolol (NNP) by CYP2C19, and riddelliine by CYP3A4, in the micronucleus assay, cell cycle analysis, and comet assay. Significant increases were observed in the percentage (%) of micronuclei induction, G2/M phase arrest, and % DNA in tails with all compounds except riddelliine, which showed increases in % micronuclei induction and G2/M phase arrest but no positive response in the comet assay. This study establishes proof-of-concept for using a TK6 cell model co-expressing multiple drug-metabolizing enzymes for genotoxicity evaluation.

Keywords: TK6 cells, Cytochrome P450, Biotransformation, DNA damage, Chromosomal damage

1. Introduction

In vitro genotoxicity tests are extensively used by regulatory agencies worldwide to identify chemicals that induce genetic damage. Since chemicals can be converted into DNA reactive metabolites in vivo, current guidelines from the U.S. Food and Drug Administration (FDA) (FDA, 2007), the Organization for Economic Co-operation and Development (OECD) (OECD, 2016c), the International Council for Harmonization (ICH) (ICH, 2011), and the International Cooperation on Harmonization of Technical Requirements for Registration of Veterinary Medicinal Products (VICH) (VICH, 2014) require the incorporation of a source of metabolic activation for all in vitro mammalian cell assays. Accordingly, chemical-induced liver S9 fractions from rodents, along with co-factors, have been routinely used in the battery of standard in vitro genotoxicity tests as a metabolic activation system (Hashizume et al., 2010).

Due to some well-recognized limitations of using exogenous rodent liver S9 for metabolism, human cell models expressing various phase I or phase II drug-metabolizing enzymes have been increasingly adopted for in vitro genotoxicity testing (Pfuhler et al., 2021; Seo et al., 2019; Seo et al., 2023b). Since its establishment in the late 1970s, the human lymphoblastoid TK6 cell line has been widely used for conducting genotoxicity assays, a major application of its use is in conducting the in vitro TK and HPRT gene mutation assays (OECD TGs 476 and 490) (OECD, 2015, 2016a). In addition, TK6 cells are one of the standard cell lines for conducting the in vitro chromosome aberration test (OECD TG 473) (OECD, 2016b), micronucleus (MN) test (OECD TG 487) (OECD, 2023), and comet assay. Whole genome sequencing analysis indicated negligible genetic variability between TK6 and the human reference genome (Revollo et al., 2016). TK6 cells, however, exhibit negligible expression of major drug metabolizing enzymes responsible for chemical biotransformation. To address this limitation, tissue homogenates, typically rat liver S9, are added to assays for metabolizing test substances that require metabolic activation. Recently, experts at the 7th International Workshop on Genotoxicity Testing (IWGT) evaluated the state-of-the-science with respect to (1) technologies and innovations for improving existing assays using human lymphoblastoid TK6 cells, and (2) novel and emerging technologies and approaches for in vitro mammalian cell mutagenicity test systems (Gollapudi et al., 2019). They recommended that any novel test system should be metabolically competent, thereby eliminating the need for exogenous metabolic activation (Evans et al., 2019).

Cytochrome P450 (CYP) enzymes were first discovered in 1954, and since then, numerous CYP proteins have been identified and found to be widespread throughout the human body (McDonnell and Dang, 2013). A review article reported that five CYPs (1A2, 2C9, 2C19, 2D6, and 3A4) account for over 75 % of P450-mediated drug metabolism in humans, while six CYPs (1A1, 1A2, 1B1, 2A6, 2E1, and 3A4) contribute to approximately 77 % of P450 activation reactions for chemicals classified as carcinogens (Rendic and Guengerich, 2021). Additionally, five CYPs (1A1, 1A2, 2A6, 2E1, and 3A4) are predominantly responsible for generating activated metabolites from natural products. Earlier studies identified twelve CYPs (1A1, 1A2, 1B1, 2A6, 2B6, 2C8, 2C9, 2C18, 2C19, 2D6, 2E1, and 3A4) as being involved in xenobiotic metabolism, with five of them (1A1, 1A2, 1B1, 2A6, and 2E1) primarily responsible for procarcinogen activation, while CYP2B6 and 2D6 have minor roles (He and Feng, 2015). Furthermore, seven CYPs (1A1, 1A2, 1B1, 2A6, 2A13, 2E1, and 3A4) have been identified as major enzymes involved in the activation of environmental carcinogens, as demonstrated by the umu genotoxicity assay and the Ames test (Shimada, 2017).

In response to the IWGT’s recommendation, we have previously developed fourteen TK6-derived cell lines that individually express one of the 14 human CYPs (CYP1A1, 1A2, 1B1, 2A6, 2B6, 2C8, 2C9, 2C18, 2C19, 2D6, 2E1, 3A4, 3A5, and 3A7), and demonstrated that this system is effective at bioactivating various indirect-acting compounds into genotoxicants and mutagens without the addition of a rodent liver S9 fraction (Li et al., 2020a; Li et al., 2020b; Li et al., 2023). In addition, since TK6 cells have negligible endogenous CYP expression, we are able to identify the specific CYP enzymes that account for the bioactivation of pro-genotoxicants, providing critical mechanistic information (Li et al., 2020b; Li et al., 2024b). Although these cell lines are valuable for in vitro genotoxicity screening, one limitation is that testing chemicals in each individual CYP-expressing cell line is labor-intensive. Therefore, it would be advantageous to develop metabolically competent TK6-derived cell lines that co-express multiple phase I enzymes. As such, we propose to further develop metabolically competent TK6 cell lines that co-express multiple human CYPs.

In this study, we created a TK6-derived cell line that simultaneously expresses four CYPs, including CYP2A6, 2E1, 2C19, and 3A4. The mRNA and protein expression, as well as enzymatic activity of each CYP in this cell line were confirmed and compared to the parental TK6 cells, and all were comparable with those in each corresponding single-CYP cell line. The performance of genotoxicity detection was further verified by several commonly used assays on four pro-genotoxicants that have been identified to be metabolized by CYP enzymes.

2. Materials and methods

2.1. Test chemicals

NDMA (CAS# 62–75-9) was purchased from Chem Service (West Chester, PA). NDEA (CAS# 55–18-5) was obtained from Sigma Aldrich (St. Louis, MO). Riddelliine (CAS# 23246–96-0) was purchased from PhytoLab GmbH & Co. KG (Vestenbergsgreuth, Germany). N-nitroso-propranolol (NNP, CAS# 84418–35-9) was purchased from LKT Lab (St. Paul, MN). While NDMA and NDEA were readily dissolved in distilled water, NNP and riddelliine were dissolved in dimethyl sulfoxide (DMSO) before the treatments.

2.2. Construction of TK6 cell line co-expressing four CYPs

Human lymphoblast TK6 cells were purchased from ATCC (Manassas, VA) and cultured as previously described (Li et al., 2020a). Cells were routinely maintained at a density of 2 × 105 to 15 × 105 cells/ml. Human embryonic kidney (HEK) 293 T cells required for lentiviral packaging were obtained from Biosettia Inc. (San Diego, CA) and propagated in high-glucose Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10 % fetal bovine serum (FBS), 2 mM l-glutamine, 1 mM sodium pyruvate, and 0.1 mM MEM non-essential amino acids. All cells were grown at 37 °C in a humidified atmosphere with 5 % CO2.

Two lentiviral expression vectors that each carries two human CYP cDNAs of interest were used for the establishment of TK6 cells that co-express four human CYPs (Fig. 1). CYP2A6 and 2E1 cDNAs were subcloned into the first lentiviral expression vector pLV-EF1α-CYP2E1-F2A-CYP2A6-IRES-neo (Biosettia) with G418 as selection antibiotic; CYP2C19 and 3A4 cDNAs were subcloned into the second lentiviral expression vector pLV-EF1α-CYP2C19-F2A-CYP3A4-IRES-puro (Biosettia) with puromycin as selection antibiotic. The two CYP cDNAs were separated by F2A peptides on each one of the bi-cistronic vector. The identities and structures of the recombinant CYP expression vectors were confirmed by nucleotide sequencing using EF1α-3′ and IRES-5′ primers for the 5′-and 3′-ends of the CYP gene inserts, respectively. The production of lentivirus was conducted as previously described (Li et al., 2020a). Briefly, 293 T cells were cultured overnight before lentivirus package. Each of the two recombinant lentiviral CYP-expression vectors were separately co-transfected with three lentiviral packaging vectors (pVSV-G, pRSV-Rev, and pMDLg/pRRE) into the 293 T cells using Lipofectamine 2000 transfection reagent (ThermoFisher Scientific, Waltham, MA). At 48 h post-transfection, lentiviral-containing supernatants were harvested and stored at −80 °C. Lentiviral titers were determined using a functional lentivirus tittering kit from Biosettia.

Fig. 1.

Fig. 1.

Open in a new tab

Lentiviral transduction pipeline to generate TK6 cells expressing four CYPs.

For lentiviral transduction, 5 × 105 TK6 cells were collected and resuspended in 2 ml medium without antibiotics. The cells were then mixed with 1 ml lentiviruses and polybrene (8 μg/ml final concentration) in a 6-well plate and centrifuged at 540 ×g for 90 min at room temperature (Li et al., 2024a). Subsequently, 8 ml medium was added into the wells and the cells were expanded for additional 48 h before selection. Specifically, we first transduced wild-type TK6 cells using the lentivirus carrying pLV-EF1α-CYP2E1-F2A-CYP2A6-IRES-neo, and G418 sulfate (ThermoFisher) at the concentration of 500 μg/ml was used to select TK6 cells (6-day selection) that stably express CYP2E1 and CYP2A6. Subsequently, we transduced the TK6 cells expressing CYP2E1 and CYP2A6 using the lentivirus carrying pLV-EF1α-CYP2C19-F2A-CYP3A4-IRES-puro, and further selected TK6 cells that co-express four CYPs (designated as TK6-4CYP) with puromycin at 1 μg/ml (3-day selection). The concentration and time of antibiotic treatments were titrated for TK6 cells before selection.

2.3. RNA extraction and quantitative real-time PCR

A total of 2 × 106 TK6 cells were used for RNA extraction using the RNeasy Mini kit (Qiagen, Valencia, CA). The quantity and purity of RNA were measured with a NanoDrop 8000 spectrophotometer (ThermoFisher). cDNA was synthesized from 2 μg of total RNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Quantitative real-time PCR (qPCR) was used to examine the expression of each CYP in TK6-4CYP cells at the mRNA level. Reactions were performed in a 20-μl volume according to the manufacturer’s protocol of FastStart Universal Probe Master (Rox) (Roche Applied Science, Indianapolis, IN). The following TaqMan probes (Applied Biosystems) were used: human CYP2A6 (Hs00711162_s1), human CYP2C19 (Hs00426380_m1), human CYP2E1 (Hs00559368_m1), human CYP3A4 (Hs00604506_m1), and human GAPDH (Hs02758991_g1). Ct values were used to determine the relative expression levels of target genes and Ct values above 35 were deemed non-detectable (McCall et al., 2014). In order to calculate the relative mRNA expression abundance of a gene, the gene expression value is determined by assuming the expression level of the reference gene (GAPDH) to be 10,000 copies. The gene expression value of each gene was defined using the equation: E=2−(Ctoftestgene−CtofGAPDH)×10,000

2.4. RNA sequencing and data analysis

To further characterize the gene expression of newly generated TK6-4CYP cells compared to their parental TK6 cells, we conducted RNA-seq experiments to examine the transcriptomic profile in these two cell lines. The RNA samples extracted were examined by an Agilent Technologies 2100 Bioanalyzer (Agilent, Santa Clara, CA) for integrity. Only samples with an RNA integrity number > 9.5 were used for subsequent mRNA-seq (poly-A pull down). mRNA-Seq library preparation and sequencing were performed at Indiana University Center for Genomics and Bioinformatics (Bloomington, IN). The library was prepared using Illumina Stranded mRNA Prep protocol and analyzed by Agilent 4200 TapeStation (Agilent). The libraries were pooled and loaded on to a NextSeq 1000/2000 P2 Reagents (100 Cycles) v3 flow cell (catalog no. 20046811) configured to generate 2 × 59 nt paired end reads on a NextSeq 2000 (Illumina, San Diego, CA). About 20 million read pairs were generated for each sample. The demultiplexing of the reads was performed using bcl2fastq, version 2.20.0. Reads were trimmed with fastp version 0.23.2 and mapped with HISAT2 version 2.2.1 against v45 of GENCODE. The reads were counted with featureCounts version 2.0.3 against v45 of GENCODE.

The differential gene expression analysis was conducted using DESeq2 (v 1.24.0). A total of 13,169 genes were included in the analyses after removing low count genes (counts per million reads <0.1). A gene was defined differentially expressed with an absolute log2 fold change (shrunken) equal to or greater than 1 and a corrected p value smaller than 0.05.

2.5. Western blot analysis

A total of 1 × 107 cells were lysed in RIPA buffer with Halt Protease Inhibitor Cocktail (ThermoFisher) for protein extraction. The protein concentrations of cell lysates were determined using the bicinchoninic acid (BCA) assay (ThermoFisher). Equivalent amounts (20 μg) of total protein were used and standard Western blot procedures were performed. Primary antibodies against CYP2A6 (ab3570, Abcam; Cambridge, UK), CYP2C19 (ab137015, Abcam), CYP2E1 (ab28146, Abcam), CYP3A4 (sc53850, Santa Cruz; Dallas, TX), and GAPDH (sc365062, Santa Cruz) were diluted in blocking reagent at a concentration of 1:1,000. Secondary antibodies conjugated with horseradish peroxidase (HRP) (Santa Cruz) were incubated at a dilution of 1:10,000. The protein signals were determined by FluroChem E System and quantified by AlphaView SA (ProteinSimple, San Jose, CA).

2.6. CYP activity measurement by UPLC-MS

Cytochrome P450 activities in TK6-4CYP cells and their parental TK6 cells were measured by ultra-performance liquid chromatography coupled with mass spectrometry (UPLC–MS) as described previously (Li et al., 2024a). Briefly, 5 × 105 cells were incubated with 1 ml of FBS free medium containing the substrate for each corresponding CYP: 20 μM coumarin (CYP2A6), 20 μM omeprazole (CYP2C19), 50 μM chlorzoxazone (CYP2E1), and 50 μM midazolam (CYP3A4), at 37 °C for 24 h. At the end of the incubation, the medium was collected, and then 40 μl of the medium was diluted by 160 μl of ice-cold acetonitrile (i.e., 1:4 dilution) followed by centrifugation at 17,000 ×g for 5 min to precipitate proteins. A Waters Acquity ultra-performance liquid chromatograph coupled with Xevo TQ-XS mass spectrometric detector (Waters Corporation, Milford, MA) was used for quantitative analysis of four CYP metabolites: 7-OH-coumarin, 5-OH-omeprazole, 6-OH-chlorzoxazone, and 1-OH-midazolam. An ACQUITY UPLC HSS T3 C18 column (2.1 × 50 mm, 1.8 μm) with a VanGuard HSS T3 pre-column (2.1 × 5 mm, 1.8 μm) was used for chromatographic separation. Column oven temperature was set at 40 °C. The mobile phase consisted of solvent A (water containing 0.1 % formic acid) and B (acetonitrile containing 0.1 % formic acid). The gradient was as follows: 0–0.1 min, 0 % B; 0.1–2 min, 0–95 % B; 2–2.1 min, 95–0 % B; 2.1–3.5 min, 0 % B for column re-equilibration. The mobile phase flow rate was 0.4 ml/min. Sample injection volume was 1 μl.

Mass spectrometric analysis was conducted using the Waters Xevo TQ-XS equipped with an electrospray ionization (ESI) source, operating in positive ion mode for 7-OH-coumarin, 5-OH-omeprazole, 1-OH-midazolam, acetaminophen-d4 (internal standard), and in negative ion mode for 6-OH-chlorzoxazone. The ion source settings for both positive and negative ion mode were as follows: ion-spray voltage 1 kV; desolvation temperature 500 °C; source temperature 150 °C; nitrogen desolvation gas 1000 L/h; cone gas 150 L/h; collision gas 0.15 ml/min. Multiple reaction monitoring (MRM) was employed to monitor the following ion transitions: m/z 163 > 107 for 7-OH-coumarin, m/z 362.1 > 214.1 for 5-OH-omeprazole, m/z 184 > 120 for 6-OH-chlorzoxazone, m/z 342.1 > 203.1 for 1-OH-midazolam, and m/z 156.1 > 114.1 for acetaminophen-d4 (internal standard). Data acquisition and quantification were performed using MassLynx 4.2 software.

2.7. In vitro micronuclei assay

The in vitro MN assay was conducted using a FACSCanto II flow cytometer equipped with a High Throughput Sampler (BD Biosciences, San Jose, CA). The OECD Test Guideline 487 has indicated that flow cytometry-based MN assay utilizes one of the automated scoring systems, in addition to microscopic visual analysis (OECD, 2023). Cells were seeded in 24-well plates at the density of 2 × 105 cells/ml. After 24 h of treatment, cells were washed by phosphate-buffered saline (PBS) and transferred to a round-bottom 96-well plate (about 1 × 105 cells/well). Subsequently, the cells were stained and lysed following the protocol described in the In Vitro MicroFlow Kit (Litron Laboratories, Rochester, NY). For the flow cytometry analysis, the stopping gate was set to record 10,000 intact nuclei and threshold parameters were set as described previously. The percentage of micronuclei (%MN) was calculated as the ratio of MN events to the total number of nucleated events. A positive response is defined when a test compound induces a concentration-related increase in %MN with at least one concentration showing a statistically significant increase compared to the concurrent vehicle control and its relative cytotoxicity (as calculated by comparing the events of nucleated cell count of treated cells to that of vehicle controls during a fixed time frame) of smaller than 50 %. In order to further compare the relative genotoxic potency of chemicals tested, the MN data in TK6-4CYP cells were analyzed using the Bayesian Benchmark Dose (BBMD) modeling system as described previously (Li et al., 2024c).

2.8. CometChip assay

CometChip gels with micropores (35 μm in diameter) were created by embedding a polydimethylsiloxane (PDMS) stamp into 1 % of agarose in 1× DPBS on GelBond Film. CometChip gels were put into the 96-Well CometChip System (R&D Systems, Minneapolis, MN), resulting in ~400 micropores per well. TK6 cells were then loaded into the 96-well CometChip System for 50 mins, and the rest of the experiment was conducted in the dark. After cell loading, the CometChip was briefly washed with 1× DPBS to remove unloaded cells, covered with 1 % low melting point agarose and lysed overnight with chilled CometAssay Lysis Solution (Trevigen, Gaithersburg, MD) at 4 °C. Gel electrophoresis was conducted at 21–22 V and 300–350 mA (1 V/cm) for 50 mins after the gel was immersed in freshly prepared alkaline solution (pH > 13) for 40 mins. The gel was subsequently neutralized and equilibrated in 0.4 M and 0.2 M Tris buffer, for 15 and 30 mins, respectively. The CometChip gel was then stained with SYBR Gold Nucleic Acid Gel Stain for 1 h and the comet images were acquired using Cytation 5 Cell Imaging Multi-Mode Reader (BioTek, Winooski, VT). The percentage of DNA in tail was analyzed in more than 100 cells per replicate for each concentration using Trevigen Comet Analysis software.

2.9. Statistical analysis

All data were expressed as mean ± 1 standard deviation (SD) from at least three independent experiments. A two-tailed Student’s t-test was used to compare treated groups to controls on the percentage of DNA in tails in the comet assay. One-way ANOVA followed by Dunnet’s post hoc test was used to compare the differences between groups for the rest of the assays. These analyses were performed using GraphPad Prism version 6.0 for windows (GraphPad Software, La Jolla, CA). For all tests, a p < 0.05 was deemed statistically significant.

3. Results

3.1. Determine the expression levels of four transduced CYPs in TK6-4CYP cells

Similar to our previous work using a lentivirus-based methodology (Li et al., 2024a), this study used two lentiviral expression vectors, each carrying two human CYP cDNAs of interest, to establish TK6 cells co-expressing four human CYPs (Fig. 1). To determine the expression levels of the four transduced CYPs in the newly developed TK6-4CYP cell line, we used qPCR, Western blot, and UPLC-MS to confirm the co-expression of CYP2A6, 2E1, 2C19, and 3A4 in TK6-4CYP cells at the gene expression, protein, and enzymatic activity levels, respectively. Previously, we observed that wild-type TK6 cells (ATCC, lot # 64048674) had non-detectable mRNA levels for most CYPs, except for CYP1A1, 2A6, 2B6, 2D6, and 2E1, which were expressed at low levels detectable by qPCR (Li et al., 2020a; Li et al., 2020b). As shown in Fig. 2A, wild-type TK6 cells had undetectable mRNA levels for CYP2C19 and 3A4, and very low levels of CYP2A6 and 2E1, given about 2 μg cDNA as input for qPCR. In contrast, the newly established TK6-4CYP cells demonstrated at least a 210-fold increase in mRNA expression for all four transduced CYPs compared to wild-type TK6 cells, indicating the effectiveness of lentiviral transduction. In addition, the gene expression level of the four CYPs in the TK6-4CYP cells were almost similar, varying by less than 2-fold.

Fig. 2.

Fig. 2.

Open in a new tab

Comparison of mRNA level, protein expression, and enzymatic activity in wild-type TK6 cells and TK6-4CYP cells. (A) mRNA levels of each CYP in TK6-4CYP cells were measured by quantitative real-time PCR. All gene expression values were normalized to the level of GAPDH. (B) Protein levels of CYP isoforms in TK6-4CYP cells were detected by Western blotting using GAPDH (37 kDa) as a loading control. (C) Enzymatic activities of various CYPs in the TK6-4CYP cells were determined by mass spectrometric analysis after a 24 h incubation with each substrate.

Subsequently, we measured the protein levels of CYP2A6, 2E1, 2C19, and 3A4 in TK6-4CYP cells by Western blot (Fig. 2B). Wild-type TK6 cells had negligible CYP protein expression, whereas the TK6-4CYP cells strongly exhibited expression of CYP2A6, 2E1, 2C19, and 3A4. At last, to confirm the metabolic functionality of these CYPs in TK6-4CYP cells, we measured their enzymatic activity using corresponding substrate probes. Fig. 2C shows representative UPLC-MS chromatograms for enzymatic activity. While wild-type TK6 cells had no activity for the enzymes studied, TK6-4CYP cells efficiently hydroxylated the substrate probes into their respective metabolites. Taken together, we used three different endpoints to ensure that our recombinant lentivirus system efficiently transduced CYP2A6, 2E1, 2C19, and 3A4 into TK6 cells.

Since lentiviral vectors integrate randomly into the host genome, we conducted a whole transcriptomic analysis to compare TK6-4CYP cells with wild-type TK6 cells and assess whether the transduction procedure altered the expression of non-target genes. About 26,000 genes were quantified, and 13,169 genes showed an average count per million >0.1 across samples. Using a log2 fold change >1 and a corrected p value <0.05 as the cutoff, only the four target genes were found to be differentially expressed (Table 1), indicating the specificity of the transduction. Notably, other CYP genes with high sequence homology to the target genes (e.g., CYP2A6 and CYP2A7, which share about 94 % homology) were unaffected.

Table 1.

Log2 fold-changes in expression of genes comparing TK6-4CYP cells with their parental TK6 cells based on transcriptomic data.

Gene Log2 FC Gene Log2 FC
CYP2A6 6.1 CYP19A1 −0.03
CYP2E1 6.3 CYP51A1 −0.16
CYP2C19 7.1 CYP1A1 < 0.1 CPM
CYP3A4 7.1 CYP1A2 < 0.1 CPM
CYP2A7 0.04 CYP1B1 < 0.1 CPM
CYP2A13 0.07 CYP2C8 < 0.1 CPM
CYP2C9 0.05 CYP2C18 < 0.1 CPM
CYP3A7 0.05 CYP2D6 < 0.1 CPM
CYP2B6 −0.15 CYP3A5 < 0.1 CPM
Open in a new tab

FC: fold change. CPM: counts per million reads.

To further evaluate the stability of CYP expression, we conducted a time-course experiment to monitor the expression levels of the four transduced CYPs. TK6-4CYP cells were subcultured daily, and RNA was collected from a subset of cells on Days 2, 5, 8, 11, 14, 17, and 20. Results showed that the expression of the four genes remained relatively stable over the 20-day period (about 30 doublings, based on the in-house doubling time of 16 h) (Fig. 3). For subsequent experiments, we limited the doubling number to 20 (~14 days of subculture) to maintain relatively consistent CYP expression.

Fig. 3.

Fig. 3.

Open in a new tab

Longitudinal analysis of mRNA expression for the four transduced CYPs. The stability of mRNA expression levels for each CYP in TK6-4CYP cells was determined at the indicated timepoints during a 20-day culture period by qPCR. All gene expression values were normalized to GAPDH levels. Data points represent the means ± SD from at least three independent experiments.

In addition, we compared the mRNA and enzymatic activity levels of each CYP in the newly developed TK6-4CYP cells with those in our previously established TK6 cells, each transduced with a single CYP. At the mRNA expression level, the expression of the four CYPs in the TK6-4CYP cells was comparable to that in the single-CYP TK6 cells (Table 2, left panel). However, relatively lower enzymatic activity levels of each CYP were observed in the TK6-4CYP cells compared to the single-CYP TK6 cells (Table 2, right panel). Despite this, the TK6-4CYP cells efficiently hydroxylated four substrate probes into their respective metabolites, confirming the metabolic activity of each enzyme (Fig. 2C).

Table 2.

Comparing the levels of mRNA expression and enzymatic activity of each CYP between TK6-4CYP cells and TK6 cells transduced with a single CYP.

CYP mRNA expression (GAPDH = 10,000)
 Enzymatic activity (ng/ml of metabolite in medium)
TK6-4CYP TK6-single CYP TK6-4CYP TK6-single CYP
CYP2A6 3733.0 ± 191.1 5995.6 ± 236.4 481.8 ± 26.9 3134.9 ± 31.3
CYP2E1 2542.4 ± 232.8 2851.8 ± 494.5 110.0 ± 3.3 115.7 ± 0.3
CYP2C19 2251.2 ± 145.3 2703.9 ± 342.5 68.3 ± 4.9 164.7 ± 0.5
CYP3A4 2234.2 ± 212.4 1747.2 ± 187.8 36.3 ± 1.1 132.0 ± 0.6
Open in a new tab

3.2. Validate the performance of TK6-4CYP cells using four prototypical pro-genotoxicants

The primary application of this newly developed cell line is for genotoxicity testing. To evaluate its performance, we selected four pro-genotoxicants, NDMA, NDEA, NNP, and riddelliine, that are known to be bioactivated by human CYP2E1, 2A6, 2C19, and 3A4, respectively. Our previous studies have demonstrated that these four compounds are not genotoxic in wild-type TK6 cells without the addition of exogenous metabolic activation using rodent liver S9 (Li et al., 2020b; Li et al., 2022; Li et al., 2023). As shown in Fig. 4, all four pro-genotoxicants induced a concentration-dependent increase in %MN formation in TK6-4CYP cells. The lowest observed adverse effect levels (LOAELs) for NDEA, NDMA, N-nitroso propranolol, and riddelliine were determined to be 12.5 μM, 5 μM, 0.5 μM, and 5 μM, respectively. Additionally, these compounds exhibited cytotoxicity in TK6-4CYP cells, with the highest tested concentrations causing 20–40 % growth inhibition. These results showed that this newly established TK6-4CYP cell line is capable of detecting the genotoxicity of various pro-genotoxicants without the need for exogenous metabolic activation or co-factors. In addition, the MN data in TK6-4CYP cells were further used for the BBMD analysis. Each of four chemicals showed similar BMD100 values in TK6-4CYP cells and TK6 cells transduced with a single CYP (Table 3), with overlapping 90 % confidence intervals for BMD estimates (BMD100L-BMD100U) between the two systems.

Fig. 4.

Fig. 4.

Open in a new tab

Induction of micronuclei by pro-genotoxicants in TK6-4CYP cells. TK6-4CYP cells were treated with different concentrations of four chemicals bioactivated by different CYPs, for 24 h. Micronucleus frequency was determined as the percentage of micronuclei relative to intact nuclei (left y-axis and open bars). Cytotoxicity was calculated as the percentage of relative nuclei count compared to control (right y-axis and red line). Data points represent the means ± SD from at least three independent experiments and * indicates p < 0.05 for a treatment group vs. the corresponding control. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 3.

Comparing the BMD estimates (μM) between TK6-4CYP cells and TK6 cells transduced with a single CYP.

Chemical TK6-4CYP
 TK6 transduced with a single CYP
BMD100 BMD100L-BMD100U U/L BMD100 BMD100L-BMD100U U/L
NDEA 49.2 38.0–65.6 1.7 60.1 (CYP2A6) 47.6–72.9 1.5
NDMA 4.9 2.7–10.6 3.9 4.7 (CYP2E1) 2.5–7.8 3.1
NNP 0.3 0.1–1.5 13.2 0.1 (CYP2C19) 0.07–0.13 1.8
Riddelliine 2.5 2.2–3.3 1.5 4.4 (CYP3A4) 3.1–6.4 2.1
Open in a new tab

BMDL: lower bound of the 90 % confidence interval of the BMD (μM).

BMDU: upper bound of the 90 % confidence interval of the BMD (μM).

U/L: the ratio of the BMDU to BMDL.

Cell cycle changes, especially G2/M phase cell cycle arrest, are commonly associated with genotoxic response. In our study, all four pro-genotoxicants induced a concentration-dependent increase in the cell populations in the G2/M phase, accompanied by a concurrent decrease in the G1 and S phase populations (Fig. 5). For example, at the highest tested concentrations, NDEA (100 μM), NDMA (40 μM), NNP (2 μM), and riddelliine (10 μM) increased the % cells in the G2/M phase by 7.5 %, 9.1 %, 16.9 %, and 19.1 %, respectively.

Fig. 5.

Fig. 5.

Open in a new tab

Induction of G2/M phase cell cycle arrest by pro-genotoxicants in TK6-4CYP cells. TK6-4CYP cells were treated with different chemicals for 24 h. (A) Representative images are shown cell cycle distribution from one experiment analyzed by flow cytometry. (B) Histograms show cell cycle changes at the indicated concentrations of each chemical, represented as the percentage of cells in each phase. Data were expressed as means ± SD from at least three independent experiments. * indicates p < 0.05 comparing treated groups to the corresponding control.

To further assess DNA damage potential, we conducted the CometChip assay to evaluate the four pro-genotoxicants in TK6-4CYP cells (Fig. 6). The results showed that NDMA (20 μM), NDEA (50 μM), and N-nitroso propranolol (2 μM) induced significant 4.1-, 2.1-, and 5.8-fold increases in %DNA in tails compared to the controls, with the relative cell viability of >70 %. In contrast, no significant changes were observed following treatment with 10 μM riddelliine; this is expected as pyrrolizidine alkaloids are well-documented to induce micronuclei formation rather than direct DNA strand breaks in various cell types. In fact, riddelliine has been shown to have DNA crosslink activity that can inhibit DNA migration and reduce comet tails induced by hydrogen peroxide (Hadi et al., 2021).

Fig. 6.

Fig. 6.

Open in a new tab

Induction of DNA damage by pro-genotoxicants in TK6-4CYP cells. TK6-4CYP cells were exposed to different chemicals at the indicated concentrations for 24 h. (A) Representative arrayed comet images following the treatment in the CometChip assay. (B) DNA damage was measured as the percentage of DNA in the tail. Data are expressed as means ± SD from at least three independent experiments. * indicates p < 0.05 comparing treated groups to the corresponding control.

4. Discussion

The current study draws significant inspiration from the MCL-5 cell line, a well-established human lymphoblastoid cell line co-expressing multiple CYPs, including CYP1A1, 1A2, 2A6, 2E1, and 3A4 (Crespi et al., 1991). In MCL-5 cells, two separate vectors were used to stably transfect CYP1A2, 2A6, 2E1, and 3A4, while CYP1A1 is constitutively expressed in the parental AHH-1 cells. Several studies have showed that metabolically competent MCL-5 cells effectively detect the mutagenicity and genotoxicity of N-nitrosamines, such as NDMA, NDEA, and smoke-related nitrosamines (Dobo et al., 1998; Krause et al., 1999). Like their parental AHH-1 cells, one drawback of the MCL-5 model is its heterozygous mutation at the TP53 locus; therefore, MCL-5 cells tend to show increased genotoxicity and decreased cytotoxicity when compared to TK6 cells, which harbor wild-type P53 (Brüsehafer et al., 2016; Guest and Parry, 1999).

Human TK6 cells are currently the most widely used human cell model for genotoxicity testing in the standard test battery (FDA, 2007; ICH, 2011; OECD, 2016c) and many commonly used tools (that are specifically developed based on this cell line), such as TGX-DDI (require wild-type P53) (Li et al., 2017), MultiFlow DNA damage assays (ideally with functional P53) (Bryce et al., 2016), and CometChip assay (Ge et al., 2014). Therefore, creating a TK6 cell line that co-expresses multiple CYPs while retaining compatibility with these tools offers a significant advantage. Furthermore, our recent findings suggest that certain nitrosamine drug substance-related impurities (NDSRIs) may require CYPs distinct from those typically involved in the bioactivation of environmental carcinogens (Li et al., 2024b). For instance, CYP2C19 appears to be the primary enzyme responsible for the bioactivation of NNP (Li et al., 2023), which further motivated the development of a new TK6-derived cell line as an alternative tool for genotoxicity screening, especially for NDSRIs.

For the cell line establishment, we utilized F2A peptide sequences, one of the most commonly used 2A peptide sequences, to co-express two different CYP coding sequences in a single open reading frame (Fig. 1). Bi-cistronic vectors incorporating 2A sequences enable protein separation through a ribosomal skipping mechanism, which results in the absence of a peptide bond without halting translation (Lengler et al., 2005). Initially, we designed a bi-cistronic lentiviral vector containing four CYPs and one UDP-glucuronosyltransferase (UGT) coding sequences, using F2A peptides to separate each cDNA (e.g., pLV-EF1α-CYP3A4-F2A-CYP2E1-F2A-CYP2A6-F2A-CYP1A1-F2A-UGT1A1-IRES-Bsd). However, while the first two or three CYPs were expressed well, the last two proteins showed poor expression, as indicated by Western blot analysis (data not shown). This likely occurred because the F2A peptides may cause ribosome drop-off during translation, with the extent of drop-off varying between constructs. Consequently, coding sequences located near the C-terminus have a lower chance of being translated, and the decline in translation efficiency correlates with the length of the vector (Liu et al., 2017; Sin et al., 2016). Since a previous study showed that CYP1B1 protein linked with F2A peptides (EGFP-F2A-CYP1B1) exhibited lower, but still acceptable, enzymatic activity compared to unmodified CYP1B1 (Lengler et al., 2005), we subsequently adopted bi-cistronic vectors carrying only two CYP cDNAs (Fig. 1) to minimize significant reductions in activity. As a result, the TK6-4CYP cells successfully showed significant increased gene expression, protein expression, and enzymatic activities of all four CYPs compared to the parental TK6 wild-type cells (Table 1, Fig. 2).

To verify the functions of this newly developed TK6-4CYP cell line, we compared its mRNA level and enzymatic activity of each CYP with those in our previously established single CYP-expressing TK6 cells (Li et al., 2020a; Li et al., 2020b). While the mRNA levels in the TK6-4CYP cells were comparable to those in the TK6 cells transduced with a single CYP, the enzymatic activities of each CYP in the TK6-4CYP cells were relatively lower (Table 2). This result is not surprising, as the construction of bi-cistronic lentiviral vectors can reduce protein expression levels depending on gene positioning within the vector (Kim et al., 2004), leading to lower overall enzymatic activity in TK6-4CYP cells compared to single-CYP TK6 cells. However, when genotoxicity testing was conducted using four pro-genotoxicants (NDEA, NDMA, NNP, and riddelliine), all of these compounds significantly induced concentration-dependent increases in MN formation as well as in the proportion of cells in the G2/M phase. These results indicate that the TK6-4CYP cells are effective in detecting chemical-induced genotoxicity. Moreover, their responses to these four chemicals were consistent with our previous results using single-CYP TK6 cell lines expressing CYP2A6, 2E1, 2C19, and 3A4 (Li et al., 2020b; Li et al., 2022; Li et al., 2023). For quantitative comparisons, results from the MN assay were analyzed using BBMD analysis. Usually, the lower and upper limits derived from the BMD estimates are used to differentiate between the responses based on non-overlapping confidence intervals. Our results indicated that all four compounds exhibited overlapping 90 % confidence intervals for BMD estimates (BMD100L-BMD100U) between the two systems (Table 3), indicating comparable effectiveness. Nonetheless, the concentrations of these pro-genotoxicants required to produce a positive response in TK6-4CYP cells were substantially lower than those needed when using an exogenous metabolic activation system (e.g., rodent liver S9).

Importantly, our objectives do not focus on the ambitious task of co-expressing CYP enzymes at physiological levels comparable to primary human hepatocytes. Instead, our aim is to explore the mechanistic intricacies of CYP-mediated metabolism and identify the specific CYP enzymes responsible for genotoxicity within a cell-based context. This approach prioritizes mechanistic insights over replicating the full complexity of hepatocyte functions. Currently, HepaRG cells are among the most promising metabolically competent models for genotoxicity assessment (Guo et al., 2020). These cells exhibit the expression of many CYP enzymes and phase II metabolic enzymes crucial for chemical biotransformation (Seo et al., 2023a). However, while we support the use of HepaRG cells, their limited proliferation post-differentiation poses challenges for conducting standard gene mutation assays. Furthermore, using HepaRG cells often provides limited information about the specific enzymes involved in bioactivation. Therefore, we consider different cell models to be complementary rather than mutually exclusive in advancing our research objectives.

In summary, we developed a TK6-derived cell line that co-expresses four human CYP enzymes (CYP2A6, 2C19, 2E1, and 3A4) and evaluated its ability to metabolize and activate pro-genotoxicants. Using four CYP-metabolized pro-genotoxicants such as NDEA, NDMA, NNP, and riddelliine, we demonstrated that TK6-4CYP cells can effectively detect their genotoxicity through several commonly used genotoxicity assays. In the future, a broader set of chemicals should be tested to fully validate this cell line for genotoxicity screening.

Acknowledgments

This work was funded by the U.S. Food and Drug Administration (FDA). YW was supported by an appointment to the Postgraduate Research Program at the National Center for Toxicological Research (NCTR) administered by the Oak Ridge Institute for Science Education through an interagency agreement between the U.S. Department of Energy and the U.S. FDA. We thank Drs. Ji-Eun Seo and Yuan Le. for their critical review of this manuscript.

Footnotes

CRediT authorship contribution statement

Xilin Li: Writing – review & editing, Writing – original draft, Validation, Resources, Methodology, Investigation, Data curation. Yuhan Wang: Writing – original draft, Validation, Methodology, Investigation, Data curation. Hannah Xu: Methodology, Investigation, Data curation. Xiaobo He: Methodology, Investigation, Data curation. Si Chen: Writing – review & editing, Validation, Methodology, Investigation. Xiaoqing Guo: Writing – review & editing, Methodology, Investigation, Data curation. Mugimane G. Manjanatha: Writing – review & editing, Validation, Resources, Project administration. Tong Zhou: Writing – review & editing, Validation, Conceptualization. Jessica Bonzo: Writing – review & editing, Resources, Funding acquisition, Conceptualization. Nan Mei: Writing – review & editing, Writing – original draft, Supervision, Project administration, Investigation, Funding acquisition, Conceptualization.

Declaration of competing interest

This article reflects the views of the authors and does not necessarily reflect those of the U.S. Food and Drug Administration (FDA). Any mention of commercial products is for clarification only and is not intended as approval, endorsement, or recommendation. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.

References

  1. Brüsehafer K, Manshian BB, Doherty AT, Zaïr ZM, Johnson GE, Doak SH, Jenkins GJ, 2016. The clastogenicity of 4NQO is cell-type dependent and linked to cytotoxicity, length of exposure and p53 proficiency. Mutagenesis 31, 171–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bryce SM, Bernacki DT, Bemis JC, Dertinger SD, 2016. Genotoxic mode of action predictions from a multiplexed flow cytometric assay and a machine learning approach. Environ. Mol. Mutagen 57, 171–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Crespi CL, Gonzalez FJ, Steimel DT, Turner TR, Gelboin HV, Penman BW, Langenbach R, 1991. A metabolically competent human cell line expressing five cDNAs encoding procarcinogen-activating enzymes: application to mutagenicity testing. Chem. Res. Toxicol 4, 566–572. [DOI] [PubMed] [Google Scholar]
  4. Dobo KL, Eastmond DA, Grosovsky AJ, 1998. Sequence specific mutations induced by N-nitrosodimethylamine at two marker loci in metabolically competent human lymphoblastoid cells. Carcinogenesis 19, 755–764. [DOI] [PubMed] [Google Scholar]
  5. Evans SJ, Gollapudi B, Moore MM, Doak SH, 2019. Horizon scanning for novel and emerging in vitro mammalian cell mutagenicity test systems. Mutat. Res 847, 403024. [DOI] [PubMed] [Google Scholar]
  6. FDA, 2007. Guidelines for Industry and Other Stakeholders: Toxicological Principles for the Safety Assessment of Food Ingredients. U.S. Food and Drug Administration. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/guidance-industry-and-other-stakeholders-redbook-2000#TOC (accesseed on 13 May 2025). [Google Scholar]
  7. Ge J, Prasongtanakij S, Wood DK, Weingeist DM, Fessler J, Navasummrit P, Ruchirawat M, Engelward BP, 2014. CometChip: a high-throughput 96-well platform for measuring DNA damage in microarrayed human cells. J. Vis. Exp 92, e50607. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Gollapudi BB, White PA, Honma M, 2019. The IWGT in vitro mammalian cell gene mutation (MCGM) assays working group-introductory remarks & consensus statements. Mutat. Res 848, 403061. [DOI] [PubMed] [Google Scholar]
  9. Guest RD, Parry JM, 1999. P53 integrity in the genetically engineered mammalian cell lines AHH-1 and MCL-5. Mutat. Res 423, 39–46. [DOI] [PubMed] [Google Scholar]
  10. Guo X, Seo JE, Li X, Mei N, 2020. Genetic toxicity assessment using liver cell models: past, present, and future. J. Toxicol. Environ. Health B Crit. Rev 23, 27–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hadi NSA, Bankoglu EE, Schott L, Leopoldsberger E, Ramge V, Kelber O, Sievers H, Stopper H, 2021. Genotoxicity of selected pyrrolizidine alkaloids in human hepatoma cell lines HepG2 and Huh6. Mutat. Res. Genet. Toxicol. Environ. Mutagen 861–862, 503305. [DOI] [PubMed] [Google Scholar]
  12. Hashizume T, Yoshitomi S, Asahi S, Uematsu R, Matsumura S, Chatani F, Oda H, 2010. Advantages of human hepatocyte-derived transformants expressing a series of human cytochrome p450 isoforms for genotoxicity examination. Toxicol. Sci 116, 488–497. [DOI] [PubMed] [Google Scholar]
  13. He X, Feng S, 2015. Role of metabolic enzymes P450 (CYP) on activating Procarcinogen and their polymorphisms on the risk of cancers. Curr. Drug Metab 16, 850–863. [DOI] [PubMed] [Google Scholar]
  14. ICH, 2011. Guidance on genotoxicity testing and data interpretation for pharmaceuticals intened for human use S2(R1). In: The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use. https://database.ich.org/sites/default/files/S2_R1_Guideline.pdf (accessed on 13 May 2025).
  15. Kim KJ, Kim HE, Lee KH, Han W, Yi MJ, Jeong J, Oh BH, 2004. Two-promoter vector is highly efficient for overproduction of protein complexes. Protein Sci. 13, 1698–1703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Krause G, Garganta F, Vrieling H, Scherer G, 1999. Spontaneous and chemically induced point mutations in HPRT cDNA of the metabolically competent human lymphoblastoid cell line, MCL-5. Mutat. Res./Fund. Mol. Mech. Mutagen 431, 417–428. [DOI] [PubMed] [Google Scholar]
  17. Lengler J, Holzmüller H, Salmons B, Günzburg WH, Renner M, 2005. FMDV-2A sequence and protein arrangement contribute to functionality of CYP2B1-reporter fusion protein. Anal. Biochem 343, 116–124. [DOI] [PubMed] [Google Scholar]
  18. Li HH, Chen R, Hyduke DR, Williams A, Frötschl R, Ellinger-Ziegelbauer H, O’Lone R, Yauk CL, Aubrecht J, Fornace AJ Jr., 2017. Development and validation of a high-throughput transcriptomic biomarker to address 21st century genetic toxicology needs. Proc. Natl. Acad. Sci. USA 114, E10881–e10889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Li X, Chen S, Guo X, Wu Q, Seo JE, Guo L, Manjanatha MG, Zhou T, Witt KL, Mei N, 2020a. Development and application of TK6-derived cells expressing human cytochrome P450s for genotoxicity testing. Toxicol. Sci 175, 251–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Li X, He X, Chen S, Guo X, Bryant MS, Guo L, Manjanatha MG, Zhou T, Witt KL, Mei N, 2020b. Evaluation of pyrrolizidine alkaloid-induced genotoxicity using metabolically competent TK6 cell lines. Food Chem. Toxicol 145, 111662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li X, He X, Le Y, Guo X, Bryant MS, Atrakchi AH, McGovern TJ, Davis-Bruno KL, Keire DA, Heflich RH, Mei N, 2022. Genotoxicity evaluation of nitrosamine impurities using human TK6 cells transduced with cytochrome P450s. Arch. Toxicol 96, 3077–3089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Li X, Le Y, Seo JE, Guo X, Li Y, Chen S, Mittelstaedt RA, Moore N, Guerrero S, Sims A, King ST, Atrakchi AH, McGovern TJ, Davis-Bruno KL, Keire DA, Elespuru RK, Heflich RH, Mei N, 2023. Revisiting the mutagenicity and genotoxicity of N-nitroso propranolol in bacterial and human in vitro assays. Regul. Toxicol. Pharmacol 141, 105410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li X, Chen S, He X, Wu Q, Guo L, Mei N, 2024a. Use of lentivirus-based method for establishing TK6 human cell lines expressing cytochrome P450 and its applications in genotoxicity testing. Curr. Protoc 4, e1003. [DOI] [PubMed] [Google Scholar]
  24. Li X, Le Y, Guo X, King ST, Dorsam RT, Atrakchi AH, McGovern TJ, Davis-Bruno KL, Keire DA, Heflich RH, Mei N, 2024b. Mutagenicity and genotoxicity evaluation of 15 nitrosamine drug substance-related impurities in human TK6 cells. Regul. Toxicol. Pharmacol 154, 105730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li X, Le Y, Li Y, Chen S, Guo L, Fu X, Manjanatha MG, Mei N, 2024c. Evaluation of weak genotoxicity of hydroxychloroquine in human TK6 cells. Toxicol. Lett 393, 84–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Liu Z, Chen O, Wall JBJ, Zheng M, Zhou Y, Wang L, Vaseghi HR, Qian L, Liu J, 2017. Systematic comparison of 2A peptides for cloning multi-genes in a polycistronic vector. Sci. Rep 7, 2193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McCall MN, McMurray HR, Land H, Almudevar A, 2014. On non-detects in qPCR data. Bioinformatics 30, 2310–2316. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. McDonnell AM, Dang CH, 2013. Basic review of the cytochrome p450 system. J. Adv. Pract. Oncol 4, 263–268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. OECD, 2015. In vitro mammalian cell gene mutation tests using the thymidine kinase gene. In: OECD Guideline for Testing of Chemicals, No. 490. http://www.oecd-ilibrary.org/environment/test-no-490-in-vitro-mammalian-cell-gene-mutation-tests-using-the-thymidine-kinase-gene_9789264242241-en (Accessed on 13 May 2025). [Google Scholar]
  30. OECD, 2016a. In vitro mammalian cell gene mutation tests using the Hprt and xprt genes. In: OECD Guideline for Testing of Chemicals, No. 476. https://www.oecd.org/en/publications/test-no-476-in-vitro-mammalian-cell-gene-mutation-tests-using-the-hprt-and-xprt-genes_9789264264809-en.html (Accessed on 13 May 2025). [Google Scholar]
  31. OECD, 2016b. In vitro mammalian chromosomal aberration test. OECD Guideline for Testing of Chemicals. No. 473. https://www.oecd.org/en/publications/2016/2007/test-no-2473-in-vitro-mammalian-chromosomal-aberration-test_g2011g2016ed2010e.html#:~:text=The%2020purpose%2020of%2020the%2020in,two%2020types%2013A%2020chromosome%2020or%2020chromatid (Accessed on 13 May 2025). [Google Scholar]
  32. OECD, 2016c. OECD Guidelines for the Testing of Chemicals, Section 4. Organisation for Economic Co-operation and Development. https://www.oecd-ilibrary.org/environment/oecd-guidelines-for-the-testing-of-chemicals-section-4-health-effects_20745788 (accessed on 13 May 2025).
  33. OECD, 2023. In vitro mammalian cell micronucleus test. In: OECD Guideline for Testing of Chemicals, No. 487. https://www.oecd.org/en/publications/2023/2007/test-no-2487-in-vitro-mammalian-cell-micronucleus-test_g2021g2026fb2022a.html (Accessed on 13 May 2025). [Google Scholar]
  34. Pfuhler S, Pirow R, Downs TR, Haase A, Hewitt N, Luch A, Merkel M, Petrick C, Said A, Schäfer-Korting M, Reisinger K, 2021. Validation of the 3D reconstructed human skin comet assay, an animal-free alternative for following-up positive results from standard in vitro genotoxicity assays. Mutagenesis 36, 19–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rendic SP, Guengerich FP, 2021. Human family 1-4 cytochrome P450 enzymes involved in the metabolic activation of xenobiotic and physiological chemicals: an update. Arch. Toxicol 95, 395–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Revollo J, Petibone DM, McKinzie P, Knox B, Morris SM, Ning B, Dobrovolsky VN, 2016. Whole genome and normalized mRNA sequencing reveal genetic status of TK6, WTK1, and NH32 human B-lymphoblastoid cell lines. Mutat. Res. Genet. Toxicol. Environ. Mutagen 795, 60–69. [DOI] [PubMed] [Google Scholar]
  37. Seo JE, Tryndyak V, Wu Q, Dreval K, Pogribny I, Bryant M, Zhou T, Robison TW, Mei N, Guo X, 2019. Quantitative comparison of in vitro genotoxicity between metabolically competent HepaRG cells and HepG2 cells using the high-throughput high-content CometChip assay. Arch. Toxicol 93, 1433–1448. [DOI] [PubMed] [Google Scholar]
  38. Seo JE, Li X, Le Y, Mei N, Zhou T, Guo X, 2023a. High-throughput micronucleus assay using three-dimensional HepaRG spheroids for in vitro genotoxicity testing. Arch. Toxicol 97, 1163–1175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Seo JE, Yu JZ, Xu H, Li X, Atrakchi AH, McGovern TJ, Bruno KLD, Mei N, Heflich RH, Guo X, 2023b. Genotoxicity assessment of eight nitrosamines using 2D and 3D HepaRG cell models. Arch. Toxicol 97, 2785–2798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shimada T, 2017. Inhibition of carcinogen-activating cytochrome P450 enzymes by xenobiotic chemicals in relation to antimutagenicity and anticarcinogenicity. Toxicol. Res 33, 79–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sin C, Chiarugi D, Valleriani A, 2016. Quantitative assessment of ribosome drop-off in E. coli. Nucleic Acids Res. 44, 2528–2537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. VICH, 2014. Studies to evaluate the safety of residues of veterinary drugs in human food: Genotoxicity testing, VICH GL23(R). In: International Cooperation on Harmonization of Technical Requirements for Registration of Veterinary Medicinal Products (VICH). https://vichsec.org/guidline/studies-to-evaluate-the-safety-of-residues-of-veterinary-drugs-in-human-food-genotoxicity-testing/ (accessed on 13 May 2025).

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data will be made available on request.

RESOURCES