Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 29.
Published in final edited form as: Arch Toxicol. 2022 Jul 26;96(11):3077–3089. doi: 10.1007/s00204-022-03347-6

Genotoxicity evaluation of nitrosamine impurities using human TK6 cells transduced with cytochrome P450s

Xilin Li 1, Xiaobo He 1, Yuan Le 1, Xiaoqing Guo 1, Matthew S Bryant 1, Aisar H Atrakchi 2, Timothy J McGovern 2, Karen L Davis-Bruno 2, David A Keire 2, Robert H Heflich 1, Nan Mei 1
PMCID: PMC12392337  NIHMSID: NIHMS2104551  PMID: 35882637

Abstract

Many nitrosamines are recognized as mutagens and potent rodent carcinogens. Over the past few years, nitrosamine impurities have been detected in various drugs leading to drug recalls. Although nitrosamines are included in a ‘cohort of concern’ because of their potential human health risks, most of this concern is based on rodent cancer and bacterial mutagenicity data, and there are little data on their genotoxicity in human-based systems. In this study, we employed human lymphoblastoid TK6 cells transduced with human cytochrome P450 (CYP) 2A6 to evaluate the genotoxicity of six nitrosamines that have been identified as impurities in drug products: N-nitrosodiethylamine (NDEA), N-nitrosoethylisopropylamine (NEIPA), N-nitroso-N-methyl-4-aminobutanoic acid (NMBA), N-nitrosomethylphenylamine (NMPA), N-nitrosodiisopropylamine (NDIPA), and N-nitrosodibutylamine (NDBA). Using flow cytometry-based assays, we found that 24-h treatment with NDEA, NEIPA, NMBA, and NMPA caused concentration-dependent increases in the phosphorylation of histone H2A.X (γH2A.X) in CYP2A6-expressing TK6 cells. Metabolism of these four nitrosamines by CYP2A6 also caused significant increases in micronucleus frequency as well as G2/M phase cell-cycle arrest. In addition, nuclear P53 activation was found in CYP2A6-expressing TK6 cells exposed to NDEA, NEIPA, and NMPA. Overall, the genotoxic potency of the six nitrosamine impurities in our test system was NMPA > NDEA ≈ NEIPA > NMBA > NDBA ≈ NDIPA. This study provides new information on the genotoxic potential of nitrosamines in human cells, complementing test results generated from traditional assays and partially addressing the issue of the relevance of nitrosamine genotoxicity for humans. The metabolically competent human cell system reported here may be a useful model for risk assessment of nitrosamine impurities found in drugs.

Keywords: Nitrosamine impurities, TK6-derived cell lines, Bioactivation, DNA damage, Chromosomal damage

Introduction

Nitrosamines, chemical substances possessing a nitroso group bonded to an amine, belong to a chemical class that includes potent rodent carcinogens. Currently, more than ten nitrosamines are categorized by the International Agency for Research on Cancer (IARC) as probable or possible human carcinogens (Group 2). Beginning in June 2018, regulatory agencies became aware of the presence of the nitrosamine impurity N-nitrosodimethylamine (NDMA, one of the smallest and most potent nitrosamines) in products containing Valsartan (FDA 2020; WHO 2019). Subsequently, NDMA and other nitrosamines were found in several drugs that are widely used to treat type II diabetes and high blood pressure. In a recent report, the U.S. Food and Drug Administration (FDA) confirmed the observation of five nitrosamine impurities (Fig. 1) in various drug products: NDMA, N-nitrosodiethylamine (NDEA), N-nitroso-N-methyl-4-aminobutanoic acid (NMBA), N-nitrosoethylisopropylamine (NEIPA or NIPEA), and N-nitrosomethylphenylamine (NMPA) (FDA 2020). Two additional nitrosamines, N-nitrosodiisopropylamine (NDIPA) and N-nitrosodibutylamine (NDBA), were proposed to be present in certain drug formulations based on process risk. Evidence in some published reports suggests that human exposure to nitrosamine contaminants in food may be associated with stomach cancers, while other publications reported no association (Jakszyn and Gonzalez 2006; Larsson et al. 2006). Occupational exposures to nitrosamines in the American and British rubber industries were associated with cancer in multiple organs (Fajen et al. 1979; Hidajat et al. 2019). Therefore, the identification of nitrosamine impurities in drugs has led to recalls over concerns of potential carcinogenic and other adverse effects in patients.

Fig. 1.

Fig. 1

Open in a new tab

Structures of six nitrosamine impurities in drugs. MW molecular weight

Overwhelming evidence indicates that nitrosamines induce cancers via a genotoxic mode of action (MoA). NDMA and NDEA are the two most comprehensively studied nitrosamines and both are positive in the Ames test (Aiub et al. 2003). In bacteria and animal studies, NDMA and NDEA metabolites form pro-mutagenic O6-alkylguanine DNA adducts, adenine adducts, and induce DNA strand breakage (Arimoto-Kobayashi et al. 1997; Brendler et al. 1992). NDBA has also tested positive in the Ames test and generates O6-butylguanine DNA adducts in rats (Bonfanti et al. 1990). However, other than data from bacterial mutagenicity assays, genotoxicity data on other nitrosamine impurities are lacking.

Most nitrosamines require metabolic activation by cytochrome P450s (CYPs) to form DNA adducts that result in mutagenicity. More specifically, CYP2E1 has been associated with the metabolism of very small size nitrosamines such as NDMA (molecular weight 74 g/mol), whereas CYP2A6 is important in the bioactivation of small-medium size nitrosamines. Generally, with an increase in the number of carbons in the alkyl sidechains, the importance of CYP2A6 in bioactivation increases. For example, NDEA (molecular weight 102 g/mol) appears to be metabolized by both CYP2A6 and CYP2E1, whereas CYP2A6 becomes predominant for the bioactivation of NMPA and NDBA (Fujita and Kamataki 2001; Kushida et al. 2000; Yamazaki et al. 1992). Notably, four nitrosamines (NMBA, NDIPA, NEIPA, NMPA) that have been identified as drug impurities or potential impurities, have much less genotoxicity data compared to NDMA and NDEA, and all have a larger size than NDEA and can be considered “small-medium” in size. Other CYPs, such as those in the CYP2C and CYP3A families, appear to be important in the bioactivation of large nitrosamines with bulky side chains (e.g., drug-related nitrosamines) (Cross and Ponting 2021).

A recent report indicated that, although nitrosamines span up to 5 orders of magnitude in their tumorigenic potency, only about 18% were considered non-carcinogenic based on the Ames test and rodent data (Thresher et al. 2020). Therefore, given the high proportion of nitrosamines that are carcinogenic, it is challenging for regulatory agencies and industry to dismiss the risk of any particular nitrosamine impurity, even when there are negative genotoxicity and carcinogenicity findings. Because of the importance in evaluating the risks associated with drug impurities, researchers and regulators have developed in silico models to predict the toxicity of impurities. The International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) M7 guideline outlines approaches for the initial assessment of drug impurity genotoxicity using quantitative structure–activity relationship (QSAR) models (ICH 2017). However, high-quality experimental data are needed as the training sets used by these models for providing reliable prediction with high confidence. Since there are hundreds of nitrosamines that require testing, relatively high-throughput methods would be beneficial for establishing robust predictive models. In addition, most of the genotoxicity testing on nitrosamines has been conducted in bacteria (Ames test) and rodents, and their genotoxic potential and the underlying mutagenic and carcinogenic mechanisms in human cells are not well understood. The risk assessment of chemicals should always employ a weight-of-evidence approach that evaluates multiple test results. Genotoxicity data for nitrosamine impurities in human cells could provide key evidence contributing to regulatory decision making.

We have established a panel of human TK6 cell lines that each expresses one of 14 different human CYPs and have tested the performance of this panel for evaluating genotoxicity using the high-throughput micronucleus (MN) test (Li et al. 2020a, b). In the current study, we used TK6 cells transduced with CYP2A6 to evaluate the genotoxicity of six nitrosamine drug impurities. This study provides novel data on the genotoxicity of these nitrosamine impurities in human cells expressing human CYP enzymes.

Methods and materials

Chemicals

NDMA (CAS# 62–75-9), NDEA (CAS# 55–18-5), benzo[a] pyrene [B(a)P, CAS# 50–32-8], and mitomycin C (CAS# 50–07-7) were purchased from Sigma Aldrich (St. Louis, MO). NEIPA (CAS# 16339–04-1) and NMPA (CAS# 614–00-6) were purchased from Enamine (Monmouth Jct., NJ). NMBA (CAS# 61445–55-4) was purchased from Chemspace (Monmouth Junction, NJ), NDBA (CAS# 924–16-3) from TCI America (Portland, OR), and NDIPA (CAS# 601–77-4) from ChemService (West Chester, PA). Lasiocarpine (CAS# 303–34-4) was purchased from PhytoLab (Vesten-bergsgreuth, Germany). All chemicals were stored as recommended by the vendor. Chemical solutions were freshly prepared before treatments.

Cell culture and treatment

Human lymphoblast TK6 cells (referred to here as ‘wild-type’ TK6 cells) were purchased from ATCC (Manassas, VA). TK6 cells expressing CYP1A1, 2A6, 2E1, and 3A4 were generated as described in our previous reports (Li et al. 2020a, b). All cells were cultured at 37 °C in a humidified atmosphere of 5% CO2 in air, using RPMI 1640 medium supplemented with L-glutamine (Thermo Fisher Scientific, Waltham, MA), 100 U/ml penicillin (Thermo Fisher), 100 μg/ml streptomycin (Thermo Fisher), and 10% fetal bovine serum (FBS) (Atlanta Biologicals, Lawrenceville, GA). Cells were routinely maintained at a density of 2 × 105 to 1.5 × 106 cells/ml for a maximum of 10 passages and were treated at the density of 2 × 105 cells/ml in microwell plates unless stated otherwise. Of note, some of the nitrosamines (e.g., NDMA and NMPA) are considered highly volatile with > 1 mm Hg vapor pressure at 20 °C. We noticed that the controls and low-concentration groups could be influenced by high-concentration groups when included in the same multiwell plate. Therefore, the controls, low concentration groups, and the high concentrations groups were treated using separate microplates.

MultiFlow DNA damage assay

The MultiFlow flow-cytometry-based assay (Litron Laboratories, Rochester, NY) was used to measure DNA damage. Cells were seeded in 24-well plates at a concentration of 2 × 105 cells/ml and exposed to the test substance or controls. After 24 h, chemical-treated cells were resuspended and 25 μl of the cell suspension were added to the wells of a separate 96-well round bottom plate containing 50 μl/well of complete labeling solution prepared following the manufacturer’s instructions. The labeling solution contained anti-γH2A.X-Alexa Fluor 647, anti-phospho-histone H3-PE, and anti-P53-FITC antibodies to detect DNA damage and mitotic cells. After a 1-h incubation at room temperature, cells were analyzed using a BD FACSCanto II Cell Analyzer equipped with a highthroughput sampler (BD Biosciences, San Jose, CA). The flow cytometric gating and data analysis employed methods described previously (Bryce et al. 2016).

In vitro micronucleus assay

Wild-type and CYP-expressing TK6 cells were exposed to nitrosamines and mitomycin C continuously for 24 h. The doubling time of TK6 cells is estimated to be 14 h, and, therefore, the 24-h time point allows the negative control cells to go through 1.5–2 cell cycles, as recommended for MN assays by the Organization for Economic Co-operation and Development (OECD) Test Guideline 487 (OECD 2016). After the treatment, cells were washed with phosphate-buffered saline (PBS) and cooled on ice for 20 min. For flow cytometric analysis, cells were stained and lysed following the protocol described in the In Vitro MicroFlow Kit (Litron Laboratories). Ethidium monoazide (EMA) and SYTOX Green were used to stain apoptotic/necrotic cells and chromatin, respectively. Cells were analyzed using a BD FACSCanto II Cell Analyzer equipped with a highthroughput sampler. The stopping gate was set to record 10,000 intact nuclei. Cell cycle information was generated in the same run by retrieving histograms of SYTOX Green nucleic acid stain fluorescence. The cell cycle analysis was conducted using FlowJo v10.1 software (BD Biosciences, Ashland, OR).

Measurement of NDBA by UPLC-MS/MS

To further explore the metabolism of NDBA in our system, wild-type TK6 cells and TK6 cells transduced with CYP2A6, 1A1, 2E1, and 3A4 were exposed to 12.5 μM NDBA for 24 h. Aroclor 1254 induced rat or hamster S9 mix (2%, MolTox, NC) with magnesium and calcium was prepared as previously described (O’Neill et al. 1982). One volume of S9 mix was added to four volumes of regular cell culture medium to make the final S9-containing culture medium. Wild-type TK6 cells were also incubated with 12.5 μM NDBA for 24 h in medium containing rat or hamster S9 to serve as a positive control for NDBA metabolism. After 24-h incubation, cell culture medium was collected and the cell pellets were washed three times by warm PBS. Lights were dimmed during processing the samples.

A Waters Acquity ultra-performance liquid chromatograph coupled with a Quattro Premier XE tandem mass spectrometric detector (UPLC-MS/MS, Waters Corporation, Milford, MA) was used for quantitative analysis of NDBA. For sample preparation, 240 μl acetonitrile were added into 80 μl cell culture medium or cell lysate for protein precipitation. After centrifugation at 12,000 × g for 5 min, the supernatant was subjected to UPLC-MS/MS analysis. 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. The column oven temperature was set at 40 °C. The mobile phase was composed 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, 30% B; 0.1–1.5 min, 30–90% B; 1.5–1.6 min, 90–30% B; 1.6–3 min, 30% B for column re-equilibration. The flow rate of the mobile phase was 0.45 ml/min; the sample injection volume was 5 μl.

The WatersQuattro Premier XE mass spectrometer equipped with a ESI source was operated in positive ion mode. The ion source settings were optimized as follows: ion-spray voltage, 1.5 kV; desolvation temperature, 500 °C; source temperature, 150 °C; nitrogen desolvation gas flow, 1000 L/hr; cone gas 150 L/hr; and collision gas flow, 0.2 ml/min. Muiltiple reaction monitoring (MRM) was employed, monitoring ion transions of m/z 159.1 > 57.3 for NDBA. Data acquisition and quantification were performed with MassLynx 4.1 software (Waters Corporation). The limit of detection for NDBA was 6.8 nM.

Bayesian benchmark dose analysis

Bayesian Benchmark Dose (BBMD) modeling was conducted using a web-based BBMD system (https://benchmarkdose.com) (Shao and Shapiro 2018) to evaluate the relative genotoxic potency of nitrosamine impurities in TK6 cells. The concentration–response data from the MN assay and the γH2A.X assay were used as the genotoxicity metrics. Briefly, Markov chain Monte Carlo (MCMC) sampling was conducted in Pystan (version 2.19.0.0) with 30,000 iterations per chain, 50% warmup percent, and 0–99,999 random seed. The concentration–response data were fitted by eight standard models (Exponential 2, Exponential 3, Exponential 4, Exponential 5, Hill, Linear, Michaelis Menten, and Power) provided by the BBMD system for continuous data. The goodness-of-fit indicators were evaluated and models with posterior predictive P-values < 0.05 or > 0.95 were excluded. Assuming a critical effective size (CES) of 0.5, which represents 50% increase in chromosomal damage (MN) or DNA damage (γH2A.X) over the background (vehicle controls), respectively (Zeller et al. 2017), a BMD50 was calculated for each dose response. The BBMD modeling system gives a model-averaged BMD value that represents an ensemble of all eight models (Shao and Shapiro 2018).

Statistical analysis

All data are presented as mean ± standard deviation (SD) of data from at least three independent experiments. Statistical analyses were performed using GraphPad Prism version 6.0 (GraphPad Software, La Jolla, CA). Either one-way ANOVA followed by Dunnett’s post hoc test or two-way ANOVA followed by the Bonferroni post hoc test was used to compare groups, with statistical significance set at P < 0.05.

Results

As part of the flow-cytometric DNA damage and MN assays, the relative cytotoxicity of each nitrosamine was assessed by comparing the nucleated cell count events for the treated groups to that of vehicle controls. None of the tested nitrosamines, except NDBA at 1 mM, caused any cytotoxicity in wild-type TK6 cells, while NDEA, NEIPA, NMPA, and NMBA caused 15–30% cytotoxicity at the highest concentration tested in CYP2A6-expressing TK6 cells (data not shown). The 1 mM NDBA treatment decreased the relative nuclei count in CYP2A6-expressing TK6 cells to 50%. NDIPA was not cytotoxic at concentrations up to 1 mM.

Nitrosamines caused γH2A.X formation in CYP2A6-expressing TK6 cells

The phosphorylation of histone H2A.X (γH2A.X) is considered a hallmark of DNA damage. Results from a MultiFlow DNA damage assay that measures γH2A.X formation are shown in Fig. 2. As DNA fragmentation in apoptotic cells also can increase the phosphorylation of H2A.X (Mukherjee et al. 2006), apoptotic and necrotic cells were gated out during the flow-cytometric analysis to minimize misleading false positive results. All tested nitrosamine impurities induced concentration-dependent increases in γH2A.X fluorescence in CYP2A6-expressing TK6 cells while they had no effects on wild-type TK6 cells (Fig. 2). In CYP2A6-expressing cells, NDEA produced a statistically significant increase (1.5-fold, P < 0.0001) in γH2A.X even at the lowest concentration tested (6.25 μM). At 100 μM NDEA, the γH2A.X fluorescence signal increased by 2.7-fold over the controls. The DNA-damaging potency of NEIPA, which has an additional methyl group on the sidechain compared to NDEA, was very similar to that of NDEA; NEIPA induced a concentration-dependent increase in γH2A.X in CYP2A6-expressing TK6 cells. NMPA appeared to be the most potent among the nitrosamines investigated, inducing a 1.3-fold increase (P = 0.001) in γH2A.X at concentrations as low as 0.625 μM. NMBA was weaker compared to NDEA, NEIPA, and NMPA, producing increased γH2A.X signals only at concentrations above 100 μM in CYP2A6-expressing cells. Last, NDIPA and NDBA had the weakest DNA-damaging capability in CYP2A6-expressing TK6 cells. For both compounds, the maximum fold change was found at 500 μM, but not at 1 mM (the highest concentration tested) (Fig. 2). Although γH2A.X induction was statistically significant, the maximum fold change was only 1.2 and 1.3 for NDIPA and NDBA, respectively. Mitomycin C at 25 and 50 ng/ml was used as the concurrent positive control, and both concentrations caused a similar, robust increase in γH2A.X in both wild-type and CYP2A6-expressing TK6 cells.

Fig. 2.

Fig. 2

Open in a new tab

Nitrosamine-induced γH2A.X formation in CYP2A6-expressing TK6 cells. The data points represent the means ± SD from at least three independent experiments. Mitomycin C was used as a positive control. *P < 0.05 comparing nitrosamine-treated groups to the vehicle control for each cell line

NDEA, NEIPA, and NMPA induced P53 activation in CYP2A6-expressing TK6 cells

TK6 cells contain wild-type human P53, which is considered an advantage for increasing the specificity of genotoxicity testing (Fowler et al. 2012). The MultiFlow assay also measures nuclear P53 content, which is an indicator of P53 activation in response to genotoxic stress (Bryce et al. 2018). As shown in Fig. 3, none of the six nitrosamines caused any P53 changes in wild-type TK6 cells, while 25 and 50 ng/ml mitomycin C (positive controls) induced 1.7- and 2.1-fold increases in nuclear P53, respectively. With bioactivation by CYP2A6, NDEA, NEIPA, and NMPA treatments led to concentration-dependent increases in nuclear P53 content. The lowest observed adverse effect level (LOAEL) was 25, 12.5, and 1.25 μM for NDEA, NEIPA, and NMPA, respectively. Compared to the vehicle control, NDEA and NEIPA induced 1.7- and 1.5-fold increases (P < 0.0001) for nuclear P53 at 100 μM, respectively. On the other hand, NMBA, NDIPA, and NDBA did not cause an increase in nuclear P53 at the concentrations studied. Considering that these three nitrosamines also showed weaker γH2A.X responses in CYP2A6-expressing TK6 cells (Fig. 2), nuclear P53 response was a less sensitive biomarker of nitrosamine-induced DNA damage than γH2A.X.

Fig. 3.

Fig. 3

Open in a new tab

Nitrosamine-induced nuclear P53 activation in CYP2A6-expressing TK6 cells. Increased p53 fluorescence intensity is indicative of genotoxicant-induced P53 responses. The data points represent the means ± SD from at least three independent experiments. Mitomycin C was used as a positive control. *P < 0.05 comparing nitrosamine-treated groups to the vehicle control for each cell line

NDEA, NEIPA, NMBA, and NMPA caused micronucleus formation in CYP2A6-expressing TK6 cells

Micronucleus (MN) induction is a regulatory endpoint for genotoxicity evaluations, conducted both in vitro and in vivo. The lymphocyte MN assay is the most widely used method for occupational health biomonitoring and MN formation has been shown to be a reliable marker of cancer risk in humans (Bonassi et al. 2011; Mišík et al. 2021). As far as using human TK6 cells for MN testing, genotoxicity data using TK6 cells have a high correlation with results obtained in primary human lymphocytes (Fowler et al. 2014). We measured MN formation using a flow cytometry-based method. None of the six nitrosamines induced MN formation in wild-type TK6 cells (Fig. 4), while 25 and 50 ng/ml mitomycin C (positive controls) induced robust 3.8- and 5.9-fold increases in %MN over the vehicle control, respectively. In CYP2A6-expressing TK6 cells, NDEA, NEIPA, NMBA, and NMPA caused concentration-dependent increases in %MN. Similar to the DNA damage endpoints, NMPA was the most potent nitrosamine impurity, producing a statistically significant increase (2.0-fold, P = 0.02) in %MN at 2.5 μM. At the highest NMPA concentration tested (10 μM), the %MN in treated cells was 2.8-fold that of controls. NDEA and NEIPA produced 2.6- and 3.6-fold increases in %MN at 100 μM, respectively. On the other hand, NDIPA and NDBA did not cause any significant increase at concentrations up to 1 mM, a finding that correlated with the weak γH2A.X response (Fig. 2) and the negative results for nuclear P53 (Fig. 3) produced by these two compounds.

Fig. 4.

Fig. 4

Open in a new tab

Nitrosamine-induced micronucleus formation in CYP2A6-expressing TK6 cells. The data points represent the means ± SD from at least three independent experiments. Mitomycin C was used as a positive control. *P < 0.05 comparing nitrosamine-treated groups to the vehicle control for each cell line

Nitrosamines induced cell cycle changes in CYP2A6-expressing TK6 cells

G2/M phase cell cycle arrest is a common response to DNA damage (Taylor and Stark 2001). We have observed that many prototypical genotoxicants, such as mitomycin C, methyl-methanesulfonate, and pyrrolizidine alkaloids, cause G2/M phase cell cycle arrest in TK6 cells (Li et al. 2020a, b). In the current study, we found that treatment with NDEA, NMPA, NEIPA, or NMBA resulted in G2/M phase cell cycle arrest in CYP2A6-expressing TK6 cells, with a concurrent decrease in the percentage of cells in S phase (Fig. 5). More specifically, the percentage of cells in the G2/M phase in vehicle controls was 21.6%. After 24-h treatment with 100 μM NDEA, 10 μM NMPA, 100 μM NEIPA, and 400 μM NMBA, the percentage of cells in G2/M phase increased to 25.1%, 32.9%, 27.0%, and 24.3%, respectively (P < 0.05). Treatment with 1 mM NDBA did not cause G2/M phase cell cycle arrest. On the contrary, 1 mM NDBA decreased the percentage of cells in G2/M phase and significantly increased the percentage of cells in the sub-G1 phase, likely due to NDBA’s cytotoxicity at this concentration. No change was observed in wild-type TK6 cells treated by any of the nitrosamines (data not shown).

Fig. 5.

Fig. 5

Open in a new tab

Nitrosamine-induced cell cycle changes in CYP2A6-expressing TK6 cells. The data points represent the means ± SD from at least three independent experiments. *P < 0.05 comparing nitrosamine-treated groups to the vehicle control

NDBA was not metabolized in TK6 cells expressing CYP2A6, 1A1, 2E1, or 3A4

Based on its chemical structure, we anticipated a weak/negative genotoxic response for NDIPA in TK6 cells expressing CYP2A6. It was surprising, however, to observe that NDBA did not increase MN frequencies in CYP2A6-expressing TK6 cells even at concentrations as high as 1 mM (Fig. 4), considering that NDBA is positive in the Ames test with metabolic activation by rodent S9 (Inami et al. 2009).

To determine if NDBA was metabolized in our system, we performed LC–MS/MS analysis to compare the level of NDBA in wild-type and CYP2A6-expressing TK6 cells after 24-h incubation (the same length of exposure used for our genotoxicity assays). We also included three additional TK6 cell lines that express either CYP1A1, CYP2E1, or CYP3A4, since these enzymes have been suggested to bioactivate various nitrosamines in different biological systems (Fujita and Kamataki 2001; Kushida et al. 2000). Rat and hamster S9 were used as positive controls for NDBA metabolism. As shown in Fig. 6A, B, none of the CYP-expressing TK6 cell lines significantly decreased the amount of NDBA in the medium when compared to the wild-type TK6 cells, suggesting that the CYP2A6, 1A1, 2E1, and 3A4 expressed in the cells did not efficiently metabolize NDBA over a period of 24 h. In contrast, when rodent S9 was added into the system, the level of NDBA decreased to less than 0.5% when compared to medium without cells and medium with wild-type TK6 cells, confirming that rodent S9 is a valid extra-cellular system for the biotransformation of NDBA.

Fig. 6.

Fig. 6

Open in a new tab

N-Nitrosodibutylamine (NDBA) is not efficiently metabolized in TK6 cells expressing CYPs. (A) Representative LC–MS/MS chromatograms showing the NDBA level in cell culture medium. Cells were exposed to 12.5 μM NDBA for 24 h. (B) Relative quantification of NDBA level in cell culture medium. The medium with wild-type TK6 cells was used as a reference (100%). (C) Micronucleus formation after the bioactivation by CYP1A1, 2E1, and 3A4 in TK6 cells. Benzo(a)pyrene [B(a)P] at 1 μM, N-nitrosodimethylamine (NDMA) at 10 μM, lasiocarpine at 0.5 μM were used as positive controls for CYP1A1, 2E1, and 3A4 bioactivation, respectively. *P < 0.05 comparing nitrosamine-treated groups to the vehicle control

Furthermore, we performed the MN assay after NDBA treatments in TK6 cells expressing CYP1A1, 2E1, or 3A4. Positive control chemicals were used for each cell line that require metabolic activation by the specific CYP [i.e., B(a) P for CYP1A1, NDMA for CYP2E1, and lasiocarpine for CYP3A4]. We found that relatively low concentrations of the positive controls all produced significant increases in MN formation without S9 in their corresponding cell line. However, NDBA had no effect on the %MN frequency in any of these cell lines, suggesting that NDBA is not efficiently metabolized and does not cause chromosomal damage in our system (Fig. 6C).

Genotoxic potency of nitrosamine impurities ranked by BBMD modeling

The relative genotoxic potency of nitrosamine impurities was investigated by BBMD modeling of concentration–response data for DNA damage (γH2A.X) and chromosomal damage (MN formation) in CYP2A6-expressing TK6 cells. All eight models for continuous data were included for concentration–response model fitting with prior weight set to 0.125 for each model. The model with the highest posterior weight for fit to each nitrosamine concentration response is shown in Fig. 7A, B. Because NDIPA and NDBA did not increase %MN and the maximum fold-change of γH2A.X was smaller than 1.5 (i.e., a CES = 0.5), their BMD50s could not be estimated. Model averaging was used to estimate BMD values to address model uncertainty.

Fig. 7.

Fig. 7

Open in a new tab

The genotoxic potency of nitrosamine impurities evaluated after the bioactivation by CYP2A6. The micronucleus (MN) data (A) from Fig. 4 and γH2A.X data (B) from Fig. 2 in CYP2A6-expressing TK6 cells exposed to nitrosamines were used for Bayesian Benchmark Dose (BBMD) modeling. (A, B) The model with the highest weight (i.e., the best fitting curve) for each nitrosamine is presented. (C) The BMD50s that estimate a 50% increase above the controls are summarized

The results demonstrated that NMPA had the lowest BMD estimates for both genotoxicity endpoints: the BMD50 was 2.2 and 1.1 μM for MN frequency and γH2A.X induction, respectively (Fig. 7C). In other words, 24-h treatment with 2.2 μM or 1.1 μM NMPA produced a 50% increase in the %MN or γH2A.X response above the background level in CYP2A6-expressing TK6 cells. BMD50 estimates for NDEA and NEIPA were similar for both endpoints and indicated that these two compounds were more than tenfold less genotoxic than NMPA. NMBA was the weakest genotoxicant among the four chemicals for which BMD50s could be estimated, more than 100-fold weaker than NMPA.

Discussion

Although the mutagenicity and genotoxicity of some nitrosamines, especially NDMA and NDEA, have been widely studied, the structural diversity of other nitrosamines makes compound-specific risk assessment challenging. The emerging issue of nitrosamine contamination in human drugs, such as valsartan and other tetrazole-containing angiotensin II receptor blockers, not only is a major safety concern but also spotlights the question of whether all nitrosamines (including N-nitrosodialkylamines and those derived from drug molecules) have a significant level of genotoxic risk. Yet, hazard identification for nitrosamines mainly relies on data from the Ames test and rodent bioassays, and the question remains as to whether hazards detected in these assays fully capture the hazard to humans. In addition, while many of these studies were conducted in the past, cancer bioassays have become very costly and are decidedly low throughput, making a thorough evaluation of not-yet-characterized nitrosamine impurities challenging.

In the current study, we used a human-based system (TK6 lymphoblastoid cells expressing human CYPs) to measure γH2A.X, P53, micronucleus formation, and cell cycle changes, in order to identify DNA damage induced by nitrosamine impurities. The high-throughput flow-cytometry-based analytical methods described here can be used as a screening tool for hazard identification that potentially provides specific information on associated carcinogenicity hazard to humans. The current study showed that NDEA, NEIPA, NMBA, and NMPA produced concentration-dependent increases in both γH2A.X formation and MN induction, after metabolic activation by endogenous CYP2A6 (Figs. 2 and 4). In contrast, none of the nitrosamines showed positive responses in wild-type TK6 cells. Notably, the concentrations tested in the current study were at least tenfold lower than those in previous genotoxicity studies that used rodent S9 fractions as an exogenous metabolism system (Liviac et al. 2011; O’Neill et al. 1982). Our observations confirm the genotoxic risk of these compounds using a human-based system. In addition, the data suggest the importance, at least for some nitrosamines, of generating DNA reactive metabolites within the cell compartment by endogenous CYPs.

First, we provided new data on the genotoxicity of nitrosamines in a human-based mammalian cell system. For example, there are little data on the genotoxicity and carcinogenicity of NEIPA, which have been identified as a contaminant in drug substances (FDA 2020). The results from our human cell assays suggest that the genotoxic potency of NEIPA is similar to that of NDEA (Fig. 7), a potent rodent carcinogen. It is reasonable to hypothesize that NEIPA also is a rodent carcinogen based on the structural similarity between NEIPA and NDEA and the available genotoxicity data. Therefore, the potential risks of human exposure to NEIPA should be considered carefully and in vivo follow-up studies are warranted.

NDIPA, which is similar in structure to NEIPA (with only an additional carbon on the sidechain), was much less genotoxic than NEIPA in our study. NDIPA also is a weak carcinogen in Sprague–Dawley rats (Lijinsky and Taylor 1979). Such relatively weak effects are likely due to the branched alkyl sidechain that may partially inhibit α-hydroxylation by CYPs to form a reactive derivative. Our data are concordant with the rodent carcinogenicity results: we observed that NDIPA weakly increased DNA damage in TK6 cells and had no significant MN induction over the 24-h time window of the assay. γH2A.X increased moderately at relatively low concentrations of NDIPA, and then plateaued (Fig. 2). This response pattern may be due to low CYP-mediated metabolism efficiency because of the branched alkyl sidechain on NDIPA. Therefore, more substrate (higher NDIPA concentrations) did not cause an increase in the genotoxic response. The negative response for MN induction may be due to the DNA lesions induced by NDIPA being partially repaired, considering that TK6 cells harbor wild-type P53. For repeated treatments in vivo, a persistently activated DNA damage-repair mechanism may result in repair-errors that lead to mutations. Notably, in the rodent bioassay, NDIPA induced a late tumor onset—the mean induction periods were 770 and 430 days for the 25 and 50 mg/kg dosing groups, respectively (Lijinsky and Taylor 1979).

NMBA caused only bladder cancer in rats given large doses administered in drinking water; NMBA was the weakest among seven N-nitrosomethyl-N-propylamine derivatives that have been tested for carcinogenicity (Lijinsky et al. 1983). There is no published literature on the genotoxicity of NMBA in vitro. Although not a potent genotoxicant in our studies, NMBA induced DNA damage (e.g., γH2A.X formation) and chromosomal damage (e.g., MN induction) in CYP2A6-expressing TK6 cells.

In contrast to the uniformly weak responses produced by NMBA, NMPA is a potent esophageal carcinogen in rats (Kroeger-Koepke et al. 1983) and mutagenic in Salmonella typhimurium expressing human CYP2A6 (Kushida et al. 2000). However, NMPA was not mutagenic in a standard Ames test (Koepke et al. 1990) or in Chinese hamster V79 cells with an exogenous microsomal activation system (Kuroki et al. 1977). Thus, it is possible that either the reactive arylating metabolites of NMPA generated exogenously by rat liver microsomes cannot get into the cells and damage DNA or perhaps that the CYP2A enzymes from the microsomes cannot metabolize NMPA to a DNA-damaging derivative. In the current study, NMPA was the most potent nitrosamine impurity among the six that were tested, producing concentration-dependent increases in γH2A.X and %MN in CYP2A6-expressing TK6 cells (Figs. 2 and 4). This result is consistent with previous findings that NMPA was the most mutagenic compound among eight nitrosamines, including NDMA and NDEA, when evaluated in an Ames test using engineered Salmonella typhimurium expressing CYP2A6 or CYP2E1 (Kushida et al. 2000).

Previous reports noted that NDBA had a much lower potency than NDMA or NDEA for inducing DNA damage in rat livers (Brambilla et al. 1981). The estimated carcinogenic potency of NDBA in Sprague–Dawley rats also was the lowest among the six nitrosamines evaluated. Previous studies suggest that NDBA is mutagenic in the Ames test with metabolic activation by CYP2A6 or CYP1A1 (Fujita and Kamataki 2001). Interestingly, although we found that NDBA increased the γH2A.X signal, it was far from a robust increase (Fig. 2). In addition, NDBA did not increase MN formation in our system. To rule out that this negative result was specific to TK6 cells, we performed the MN assay using human HepG2 cells overexpressing CYP2A6 (Chen et al. 2021) and found NDBA did not increase %MN frequencies in these cells either (Supplementary Fig. 1). We also tested the mutagenicity of NDBA in TK6 cells expressing CYP2A6 and found it did not induce TK or HPRT gene mutations (Supplementary Table 1). Our metabolism analysis further showed that none of three CYP-expressing TK6 cells efficiently bioactivated NDBA, while they functioned well in detecting the MN formation induced by other pro-genotoxicants that require metabolic activation to become DNA reactive (Fig. 6).

We postulate that the observed negative results for NDBA in our system can be attributed to two possibilities. First, while it is hypothesized that reactive metabolites generated within the cell compartment are more likely to reach nuclear DNA and cause damage, a prerequisite for metabolism is that the parent compound must be transported into the cells. Our data indicate that most of the NDBA remains in the cell culture medium, and only a very limited amount of NDBA is found in the cell lysate (Supplementary Fig. 2). On the other hand, rodent S9 mix, containing K+, Mg++, Ca++ cations, can directly react with NDBA outside of the cells, generating metabolites that may more readily enter into the cells and cause DNA damage. The second possibility is that other enzymes, or more than one drug-metabolizing enzyme, are needed for the bioactivation of NDBA. This possibility highlights one of the limitations of our system. TK6 cells are largely devoid of both Phase I and Phase II enzymes (Li et al. 2022); therefore, if more than one drug-metabolizing enzyme is involved in the biotransformation of a pro-genotoxicant, the current system will have limited capability for detecting its toxicity.

The structure of nitrosamines appears to play a critical role in their genotoxicity (Honma et al. 2019). Generally, nitrosamines undergo α-hydroxylation by CYPs and form active intermediates, which subsequently may be transformed to carbonium ions that alkylate DNA and cause genotoxicity. This mechanism of action has been studied extensively using NDMA (George et al. 2020). After metabolism by CYP2E1, NDMA is converted to hydroxymethylnitrosamine, which degrades into formaldehyde and N-nitrosomethylamine via non-enzymatic reactions. N-nitrosomethylamine can be further biotransformed into a methyl-carbocation, which is highly unstable and reacts with DNA. It is generally assumed that other nitrosamines will be dealkylated through hydroxylation by CYPs and eventually form carbocations with different numbers of carbon atoms and structures (Bellec et al. 1996). These presumed pathways of nitrosamine metabolic activation were summarized in a recent review article (Li and Hecht 2022). Since different carbocations likely have different capabilities to bind with DNA, the structure of nitrosamines can be informative when assessing their genotoxic potency.

Assuming a similar pathway of biotransformation as NDMA, the relative genotoxicity of the four nitrosamines containing alkyl sidechains that were used in the current study (NDEA, NDBA, NEIPA, NDIPA) can be ordered based on their presumed carbocation derivatives. For example, both NDEA and NEIPA can produce ethyl-carbocations and they showed similar potency for both MN induction and DNA damage in CYP2A6-expressing TK6 cells (Fig. 7C). NDIPA and NDBA presumably generate butyl-carbocations (albeit with different structures) and they are the weakest genotoxicants in our system. These observations are consistent with previous studies indicating that the sidechain length of nitrosamines is inversely correlated with their mutagenicity in the Ames test (Kuroki et al. 1977). NDBA potentially generates butyl-carbocations whereas NDIPA generates the weakest tert-butyl-carbocations, which leads to their different carcinogenic potency in rodents. NDIPA is considered a very weak carcinogen in rodents, causing tumors of the nasal turbinates only at high doses (Lijinsky and Taylor 1979); whereas NDBA, although much weaker than NDEA and N-nitrosodipropylamine (NDPA), causes liver, esophageal, bladder, and forestomach tumors (Lijinsky et al. 1983; Tsuda et al. 1987).

We found that NMPA, which contains an aromatic ring on the side chain, is the most potent nitrosamine after the bioactivation by CYP2A6 (Fig. 7). This result is consistent with a previous study which indicated that NMPA is even more mutagenic than NDMA and NDEA in engineered Salmonella typhimurium expressing CYP2A6 (Kushida et al. 2000). NMPA can undergo α-hydroxylation and form formaldehyde and a benzenediazonium ion after CYP metabolism. The benzenediazonium ion has been shown to interact with adenine and forms a triazene DNA adduct both in vitro and in vivo (Koepke et al. 1990). The benzenediazonium ion also appears to be more stable than methyl- or ethyl-carbocations due to its conjugated structure. Therefore, it may have a greater chance of interacting with DNA and inducing a genotoxic response. On the other hand, NMPA, when compared to other nitrosamines, may be more efficiently metabolized by CYP2A6, which can contribute to its high potency in our cell system.

In summary, the current study used human TK6 cells expressing human CYP2A6 to evaluate the genotoxicity of nitrosamine impurities. Using this system, the genotoxic potency of six nitrosamine impurities were ranked as NMPA > NDEA ≈ NEIPA > NMBA > NDBA ≈ NDIPA, although the genotoxicity of NDBA should be carefully reevaluated due to its poor metabolism in our system. This study provides new information on the genotoxic potential of nitrosamines in human cells and, therefore, can support the risk assessment of these widely found impurities. In addition, our findings serve as a proof-of-concept for the subsequent genotoxicity screening of novel nitrosamine drug substance-related impurities (NDSRIs) in a human-based cell system.

Supplementary Material

Supplementary Tables & Figures

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s00204-022-03347-6.

Acknowledgements

This work was partly supported by funding from the Center for Drug Evaluation and Research (CDER) Regulatory Science Research program. YL 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. Food and Drug Administration (FDA). We thank Drs. Robert Dorsam, Sruthi King, Naomi Kruhlak from CDER for their valuable comments regarding nitrosamine impurities and Dr. Tao Chen and Ms. Roberta Mittelstaedt for their critical review of this manuscript.

Footnotes

Declarations

Conflict of 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.

References

  1. Aiub CA, Pinto LF, Felzenszwalb I (2003) N-Nitrosodiethylamine mutagenicity at low concentrations. Toxicol Lett 145(1):36–45. 10.1016/s0378-4274(03)00263-7 [DOI] [PubMed] [Google Scholar]
  2. Arimoto-Kobayashi S, Kaji K, Sweetman GM, Hayatsu H (1997) Mutation and formation of methyl- and hydroxylguanine adducts in DNA caused by N-nitrosodimethylamine and N-nitrosodiethylamine with UVA irradiation. Carcinogenesis 18(12):2429–2433. 10.1093/carcin/18.12.2429 [DOI] [PubMed] [Google Scholar]
  3. Bellec G, Dréano Y, Lozach P, Ménez JF, Berthou F (1996) Cytochrome P450 metabolic dealkylation of nine N-nitrosodialkylamines by human liver microsomes. Carcinogenesis 17(9):2029–2034. 10.1093/carcin/17.9.2029 [DOI] [PubMed] [Google Scholar]
  4. Bonassi S, El-Zein R, Bolognesi C, Fenech M (2011) Micronuclei frequency in peripheral blood lymphocytes and cancer risk: evidence from human studies. Mutagenesis 26(1):93–100. 10.1093/mutage/geq075 [DOI] [PubMed] [Google Scholar]
  5. Bonfanti M, Magagnotti C, Galli A et al. (1990) Determination of O6-butylguanine in DNA by immunoaffinity extraction/gas chromatography-mass spectrometry. Cancer Res 50(21):6870–6875 [PubMed] [Google Scholar]
  6. Brambilla G, Cavanna M, Pino A, Robbiano L (1981) Quantitative correlation among DNA damaging potency of six N-nitroso compounds and their potency in inducing tumor growth and bacterial mutations. Carcinogenesis 2(5):425–429. 10.1093/carcin/2.5.425 [DOI] [PubMed] [Google Scholar]
  7. Brendler SY, Tompa A, Hutter KF, Preussmann R, Pool-Zobel BL (1992) In vivo and in vitro genotoxicity of several N-nitrosamines in extrahepatic tissues of the rat. Carcinogenesis 13(12):2435–2441. 10.1093/carcin/13.12.2435 [DOI] [PubMed] [Google Scholar]
  8. 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(3):171–189. 10.1002/em.21996 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bryce SM, Bernacki DT, Smith-Roe SL, Witt KL, Bemis JC, Dertinger SD (2018) Investigating the generalizability of the MultiFlow DNA damage assay and several companion machine learning models with a set of 103 diverse test chemicals. Toxicol Sci 162(1):146–166. 10.1093/toxsci/kfx235 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen S, Wu Q, Li X et al. (2021) Characterization of cytochrome P450s (CYP)-overexpressing HepG2 cells for assessing drug and chemical-induced liver toxicity. J Environ Sci Health C Toxicol Carcinog 39(1):68–86. 10.1080/26896583.2021.1880242 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cross KP, Ponting DJ (2021) Developing structure-activity relationships for N-nitrosamine activity. Comput Toxicol 20:100186. 10.1016/j.comtox.2021.100186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Fajen JM, Carson GA, Rounbehler DP et al. (1979) N-nitrosamines in the rubber and tire industry. Science 205(4412):1262–1264. 10.1126/science.472741 [DOI] [PubMed] [Google Scholar]
  13. FDA (2020) Control of nitrosamine impurities in human drugs. US Food and Drug Administration; https://www.fda.gov/regulatory-information/search-fda-guidance-documents/control-nitrosamine-impurities-human-drugs (accessed 15 Apr. 2022) [Google Scholar]
  14. Fowler P, Smith K, Young J et al. (2012) Reduction of misleading (“false”) positive results in mammalian cell genotoxicity assays. I. Choice of cell type. Mutat Res 742(1–2):11–25. 10.1016/j.mrgentox.2011.10.014 [DOI] [PubMed] [Google Scholar]
  15. Fowler P, Smith R, Smith K et al. (2014) Reduction of misleading (“false”) positive results in mammalian cell genotoxicity assays. III: sensitivity of human cell types to known genotoxic agents. Mutat Res 767:28–36. 10.1016/j.mrgentox.2014.03.001 [DOI] [PubMed] [Google Scholar]
  16. Fujita K, Kamataki T (2001) Role of human cytochrome P450 (CYP) in the metabolic activation of N-alkylnitrosamines: application of genetically engineered Salmonella typhimurium YG7108 expressing each form of CYP together with human NADPH-cytochrome P450 reductase. Mutat Res 483(1–2):35–41. 10.1016/s0027-5107(01)00223-8 [DOI] [PubMed] [Google Scholar]
  17. George J, Tsuchishima M, Tsutsumi M (2020) Metabolism of N-nitrosodimethylamine, methylation of macromolecules, and development of hepatic fibrosis in rodent models. J Mol Med (berl) 98(9):1203–1213. 10.1007/s00109-020-01950-7 [DOI] [PubMed] [Google Scholar]
  18. Hidajat M, McElvenny DM, Ritchie P et al. (2019) Lifetime exposure to rubber dusts, fumes and N-nitrosamines and cancer mortality in a cohort of British rubber workers with 49 years follow-up. Occup Environ Med 76(4):250–258. 10.1136/oemed-2018-105181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Honma M, Kitazawa A, Cayley A et al. (2019) Improvement of quantitative structure-activity relationship (QSAR) tools for predicting Ames mutagenicity: outcomes of the Ames/QSAR International Challenge Project. Mutagenesis 34(1):3–16. 10.1093/mutage/gey031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. ICH (2017) M7(R1) Assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals to limit potential carcinogenic risk. https://www.ich.org/page/multidisciplinary-guidelines#7-2 (Accessed 12 Mar. 2022) [Google Scholar]
  21. Inami K, Ishikawa S, Mochizuki M (2009) Activation mechanism of N-nitrosodialkylamines as environmental mutagens and its application to antitumor research. Genes Environ 31(4):97–104. 10.3123/jemsge.31.97 [DOI] [Google Scholar]
  22. Jakszyn P, Gonzalez CA (2006) Nitrosamine and related food intake and gastric and oesophageal cancer risk: a systematic review of the epidemiological evidence. World J Gastroenterol 12(27):4296–4303. 10.3748/wjg.v12.i27.4296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Koepke SR, Kroeger-Koepke MB, Michejda CJ (1990) Evidence for an unstable DNA adduct from N-nitroso-N-methylaniline. Chem Res Toxicol 3(1):17–20. 10.1021/tx00013a003 [DOI] [PubMed] [Google Scholar]
  24. Kroeger-Koepke MB, Reuber MD, Iype PT, Lijinsky W, Michejda CJ (1983) The effect of substituents in the aromatic ring on carcinogenicity of N-nitrosomethylaniline in F344 rats. Carcinogenesis 4(2):157–160. 10.1093/carcin/4.2.157 [DOI] [PubMed] [Google Scholar]
  25. Kuroki T, Drevon C, Montesano R (1977) Microsome-mediated mutagenesis in V79 Chinese hamster cells by various nitrosamines. Cancer Res 37(4):1044–1050 [PubMed] [Google Scholar]
  26. Kushida H, Fujita K, Suzuki A et al. (2000) Metabolic activation of N-alkylnitrosamines in genetically engineered Salmonella typhimurium expressing CYP2E1 or CYP2A6 together with human NADPH-cytochrome P450 reductase. Carcinogenesis 21(6):1227–1232. 10.1093/carcin/21.5.227 [DOI] [PubMed] [Google Scholar]
  27. Larsson SC, Bergkvist L, Wolk A (2006) Processed meat consumption, dietary nitrosamines and stomach cancer risk in a cohort of Swedish women. Int J Cancer 119(4):915–919. 10.1002/ijc.21925 [DOI] [PubMed] [Google Scholar]
  28. Li Y, Hecht SS (2022) Metabolic activation and DNA Interactions of carcinogenic N-nitrosamines to which humans are commonly exposed. Int J Mol Sci 23(9):4559. 10.3390/ijms23094559 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li X, Chen S, Guo X et al. (2020a) Development and application of TK6-derived cells expressing human cytochrome P450s for genotoxicity testing. Toxicol Sci 175(2):251–265. 10.1093/toxsci/kfaa035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Li X, He X, Chen S et al. (2020b) Evaluation of pyrrolizidine alkaloid-induced genotoxicity using metabolically competent TK6 cell lines. Food Chem Toxicol 145:111662. 10.1016/j.fct.2020.111662 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Li X, Li Y, Ning KG et al. (2022) The expression of Phase II drug-metabolizing enzymes in human B-lymphoblastoid TK6 cells. J Environ Sci Health C Toxicol Carcinog 40(1):106–118. 10.1080/26896583.2022.2044242 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lijinsky W, Taylor HW (1979) Carcinogenicity of methylated derivatives of N-nitrosodiethylamine and related compounds in Sprague-Dawley rats2. J Natl Cancer Inst 62(2):407–410. 10.1093/jnci/62.2.407 [DOI] [PubMed] [Google Scholar]
  33. Lijinsky W, Reuber MD, Saavedra JE, Singer GM (1983) Carcinogenesis in F344 rats by N-nitrosomethyl-n-propylamine derivatives. J Natl Cancer Inst 70(5):959–963. 10.1093/jnci/70.5.959 [DOI] [PubMed] [Google Scholar]
  34. Liviac D, Creus A, Marcos R (2011) Genotoxic evaluation of the non-halogenated disinfection by-products nitrosodimethylamine and nitrosodiethylamine. J Hazard Mater 185(2–3):613–618. 10.1016/j.jhazmat.2010.09.062 [DOI] [PubMed] [Google Scholar]
  35. Mišík M, Nersesyan A, Kment M et al. (2021) Micronucleus assays with the human derived liver cell line (Huh6): a promising approach to reduce the use of laboratory animals in genetic toxicology. Food Chem Toxicol 154:112355. 10.1016/j.fct.2021.112355 [DOI] [PubMed] [Google Scholar]
  36. Mukherjee B, Kessinger C, Kobayashi J et al. (2006) DNA-PK phosphorylates histone H2AX during apoptotic DNA fragmentation in mammalian cells. DNA Repair (amst) 5(5):575–590. 10.1016/j.dnarep.2006.01.011 [DOI] [PubMed] [Google Scholar]
  37. OECD (2016) In vitro mammalian cell micronucleus test, OECD Guidelines for the Testing of Chemicals, No. 487. Organisation for Economic Co-operation and Development (Paris, France) 10.1787/9789264264861-en (assessed 15 Apr. 2022) [DOI] [Google Scholar]
  38. O’Neill JP, Machanoff R, San Sebastian JR, Hsie AW (1982) Cytotoxicity and mutagenicity of dimethylnitrosamine in mammalian cells (CHO/HGPRT system); enhancement by calcium phosphate. Environ Mutagen 4(1):7–18. 10.1002/em.2860040103 [DOI] [PubMed] [Google Scholar]
  39. Shao K, Shapiro AJ (2018) A web-based system for Bayesian benchmark dose estimation. Environ Health Perspect 126(1):017002. 10.1289/EHP1289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Taylor WR, Stark GR (2001) Regulation of the G2/M transition by p53. Oncogene 20(15):1803–1815. 10.1038/sj.onc.1204252 [DOI] [PubMed] [Google Scholar]
  41. Thresher A, Foster R, Ponting DJ, Stalford SA, Tennant RE, Thomas R (2020) Are all nitrosamines concerning? A review of mutagenicity and carcinogenicity data. Regul Toxicol Pharmacol 116:104749. 10.1016/j.yrtph.2020.104749 [DOI] [PubMed] [Google Scholar]
  42. Tsuda H, Mera Y, Seki K, Aoki T, Fukushima S, Ito N (1987) Induction of tumors in the liver, urinary bladder, esophagus and forestomach by short-term treatment with different doses of N, N’-dibutylnitrosamine in rats. Jpn J Cancer Res 78(3):227–234. 10.20772/cancersci1985.78.3_227 [DOI] [PubMed] [Google Scholar]
  43. WHO (2019) Update on nitrosamine impurities. World Health Organization; https://www.who.int/news/item/20-11-2019-information-note-nitrosamine-impurities (accessed 15 Apr. 2022) [Google Scholar]
  44. Yamazaki H, Inui Y, Yun CH, Guengerich FP, Shimada T (1992) Cytochrome P450 2E1 and 2A6 enzymes as major catalysts for metabolic activation of N-nitrosodialkylamines and tobacco-related nitrosamines in human liver microsomes. Carcinogenesis 13(10):1789–1794. 10.1093/carcin/13.10.1789 [DOI] [PubMed] [Google Scholar]
  45. Zeller A, Duran-Pacheco G, Guerard M (2017) An appraisal of critical effect sizes for the benchmark dose approach to assess dose-response relationships in genetic toxicology. Arch Toxicol 91(12):3799–3807. 10.1007/s00204-017-2037-3 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Tables & Figures
NIHMS2104551-supplement-Supplementary_Tables___Figures.docx (85.5KB, docx)

RESOURCES