Mutagenicity and genotoxicity evaluation of 15 nitrosamine drug substance-related impurities in human TK6 cells

Abstract

Nitrosamine drug substance-related impurities (NDSRIs) are a sub-category of N-nitrosamine drug impurities that share structural similarity to the corresponding active pharmaceutical ingredient. The mutagenicity of NDSRIs is poorly understood. We previously tested a series of NDSRIs using the Enhanced Ames Test (EAT). In this follow-up study, we further examined the genotoxicity and mutagenicity of 15 of these NDSRIs in human TK6 cells. Seven EAT-positive NDSRIs, including N-nitroso-nortriptyline, N-nitroso-fluoxetine, N-nitroso-desmethyl-diphenhydramine, N-nitroso-duloxetine, N-nitroso-lorcaserin, N-nitroso-varenicline, and N-nitroso-sertraline, induced concentration-dependent increases in micronuclei after bioactivation with hamster liver S9. These NDSRIs were also mutagenic in the TK and HPRT gene mutation assays, consistent with their positive EAT results. In the presence of hamster liver S9, the eight EAT-negative NDSRIs were negative in the micronucleus assay and negative for mutation induction. Using TK6 cells endogenously expressing a single human cytochrome P450 (CYP), we found that CYP2C19, CYP2B6, CYP2A6, and CYP3A4 are key enzymes activating the genotoxicity and mutagenicity of these NDSRIs. Overall, the hamster S9-mediated TK6 cell mutagenicity results agreed with those observed in the EAT, indicating consistency in the mutagenic responses produced by NDSRIs across different testing systems. These data support the use of EAT for hazard identification and safety assessment of NDSRIs.

Keywords: Chromosomal damage; DNA damage; Gene mutation; Hamster liver S9; NDSRIs; Nitrosamine impurities.

Fig. 1.. Structures of 15 nitrosamine drug substance-related impurities

(NDSRIs, in the order of their molecular weight).

Fig. 2.. NDSRI-induced micronucleus formation in TK6 cells in the presence of hamster liver S9.

Cells were treated with indicated concentrations of seven Ames-positive NDSRIs (A) and eight Ames-negative NDSRIs (B) for 4 h. After each treatment, micronucleus frequency (%) was determined by the high-throughput micronucleus assay using flow cytometry. The red line indicates relative cytotoxicity (right y-axis), and the open bars indicate the percentage of micronuclei (left y-axis). Lasiocarpine (2.5 μM) was used as the positive control (black bars). The data points represent the means ± SD from at least three independent experiments. * denotes statistical significance compared to the vehicle control (P < 0.05, fold-change >2.0).

Fig. 3.. NDSRI-induced micronucleus formation in TK6 cells in the absence of S9 activation.

Cells were treated with indicated concentrations of seven Amespositive NDSRIs (A) and eight Ames-negative NDSRIs (B) for 24 h. After each treatment, micronucleus frequency (%) was determined by the high-throughput micronucleus assay using flow cytometry. The red line indicates relative cytotoxicity (right y-axis), and the open bars indicate the percentage of micronuclei (left y-axis). Mitomycin C (50 ng/mL) was used as the positive control (black bars). The data points represent the means ± SD from at least three independent experiments. * denotes statistical significance compared to the vehicle control (P < 0.05, fold-change >2.0).

Fig. 4.. NDSRI-induced cell cycle changes in TK6 cells.

Cells were treated with seven NDSRIs, that induced micronucleus formation in the presence of hamster liver S9 (see Fig. 2), for 4 h (A and B) and six NDSRIs, that induced micronucleus formation in the absence of S9 activation (see Fig. 3), for 24 h (C and D) and cell cycle was measured using a flow cytometer at the 24-h time point. Representative flow cytometry plots are shown from one experiment (A and C), and the histograms show cell cycle changes of G1 and G2/M phases (B and D). The data points (B and D) represent the means ± SD from at least three independent experiments. *P < 0.05 compared with the vehicle control.

Fig. 5.. Identification of specific CYPs accounting for the bioactivation of Ames-positive NDSRIs in CYP-expressing TK6 cells.

The TK6 cells transduced with nine different CYPs (see Supplementary Figs. 2A–2G) were treated with indicated concentrations of seven Ames-positive NDSRIs for 24 h. After each treatment, micronucleus frequency (%) was determined by the high-throughput micronucleus assay using flow cytometry, and the data (A–G) with significant increases in two different CYP-expressing TK6 cell lines (CYP2A6, 2B6, 2C19, or 3A4) are shown and compared with the results in wild-type TK6 cells (WT). The red line indicates relative cytotoxicity (right y-axis), and the bars indicate the percentage of micronucleus formation (left y-axis). The data points represent the means ± SD from at least three independent experiments. * denotes statistical significance compared to the vehicle control (P < 0.05, fold-change >2.0).