Mutagenic activities of heterocyclic amines in Chinese hamster lung cells in culture.

A mutation assay system with Chinese hamster lung cells (CHL) using diphtheria toxin resistance as a selective marker has been established. The mutagenic activities of heterocyclic amines, originally isolated from pyrolyzates of amino acids and proteins, broiled fish and fried beef were assayed in cultured CHL cells in the absence and presence of a metabolic activation system, with diphtheria toxin resistance as a marker. All the heterocyclic amines tested except 3-amino-1,4-dimethyl-5H-pyrido [4,3-b]indole (Trp-P-1) required the presence of a metabolic activation system for mutagenicity on CHL cells. 3-Amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) was the most mutagenic among the heterocyclic amines tested. Other compounds were also mutagenic in the following order of decreasing potency: Trp-P-1, 2-amino-3,4-dimethylimidazo[4,5-f]quinoline (MeIQ), 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2-amino-9H-pyrido[2,3-b]indole (A alpha C), 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), 2-amino-6-methyldipyrido[1,2-a:3',2'-d]imidazole (Glu-P-1) and 2-aminodipyrido[1,2-a:3',2'-d]imidazole (Glu-P-2).

A mutation assay system with Chinese hamster lung cells (CHL) using diphtheria toxin resistance as a selective marker has been established. The mutagenic activities of heterocyclic amines, originally isolated from pyrolyzates of amino acids and proteins, broiled fish and fried beef were assayed in cultured CHL cells in the absence and presence of a metabolic activation system, with diphtheria toxin resistance as a marker. All the heterocyclic amines tested except 3-amino-1,4-dimethyl-5H-pyrido [4,3-blindole (Trp-P-1) required the presence of a metabolic activation system for mutagenicity on CHL cells. 3-Amino-i-methyl-5H-pyrido [4,3-b]indole (Trp-P-2) was the most mutagenic among the heterocyclic amines tested. Other compounds were also mutagenic in the following order of decreasing potency: Trp-P-1, 2-amino-3,4-dimethylimidazo [4,5- Mutation testing using Salmonella typhimurium has successfully detected and identified mutagens in our daily foods. The application of a microbial mutation test for screening carcinogens is based on the reports showing overlapping ofmutagenicity with carcinogenicity (1)(2)(3)(4)(5). Various mutagenic heterocyclic amines were isolated from pyrolyzates of amino acids, proteins, and cooked foods (6,7). All of these mutagenic heterocyclic amines, with which long-term animal carcinogenicity experiments have been completed, proved to be carcinogenic (6,7).
CHL cells were transferred to 5.5-cm2 culture tubes with flat bottoms (A/S Nunc, Denmark). After 40 to 48 hr, the medium was changed to 1 mL of VAMEM with 3.5% fetal bovine serum or 1 mL of S9 mix medium, and the cells were cultured with a test compound for 3 hr in the CO2 incubator at 37°C. The S9 mix medium consisted of 4.27 mM G6P, 0.75 mM NADP in VAMEM with 3.5% fetal bovine serum, and different concentrations of the S9 fraction from liver of rats treated with polychlorinated biphenyls. The protein concentration of the S9 fraction was about 35 mg/mL, and the optimum concentration of S9 was determined for each compound before assays. After exposure to a test compound, 2 to 4 x 105 cells were transferred after trypsinization to a 25-cm2 flask containing 10 mL of VAMEM with 10% fetal bovine serum. After an expression time of 7 to 8 days, 2.5 x 105 cells were transferred to 100-mm dishes with VAMEM with 10% fetal bovine serum, containing 0.1 Lf/mL of diphtheria toxin. Simultaneously, 200 to 400 cells were transferred to 5 mL of toxin-free medium in 60-mm dishes for determination of plating efficiency.
After incubation for 7 to 8 days in medium in the presence and absence of the toxin, the cells were fixed with methanol and stained with Giemsa. Mutation frequency was expressed as the number of DT' cells per 2.5 x 105 survivors. Concentrations of 0.01 to 1 Lf/mL diphtheria toxin were cytotoxic to CHL cells, and 0.1 Lf/mL diphtheria toxin was used to select DT' cells. Under the assay conditions described, 7 to 9 days of expression periods were optimum for induction of DT' cells.
The mutagenicities of alkylating agents, such as ethyl methanesulfonate (EMS), methyl methanesulfonate (MMS), methylnitrosourea (MNU), and N-methyl-N'nitro-N-nitrosoguanidine (MNNG) were determined. With these compounds, there was a dose-dependent increase in the number of DT' cells.

Properties of DTr Cells
Diphtheria toxin is composed of an A fragment and a B fragment having molecular weights of 22,000 and 40,000, respectively. Fragment B attaches to receptors on the cells and fragment A enters the cells. Fragment A itself is an enzyme, which catalyzes ADP-ribosylation of elongation factor 2 at a unique modified amino acid, diphthamide (13,14). Once elongation factor 2 is ADPribosylated, the factor loses its activity and protein synthesis stops, leading to killing of the cells. There are at least two possible mechanisms by which the CHL cells are converted to DT' (15). One involves a membrane alteration leading to reduced entry of the toxin into cells, and the other involves an alteration of elongation factor 2 such that ADP-ribosylation by the toxin does not occur. Seventeen DT' clones were isolated in this work. Cytoplasmic fractions containing elongation factor 2 were prepared from the parental and mutant cells, and the toxin-catalyzed ADP-ribosylation of EF-2 was measured using nicotinamide-[U-14C]adenine dinucleotide as a substrate (8). With the parental cells ADPribosylation increased linearly with the amounts of elongation factor 2, and 113 pmole of ADP-ribose per milligram protein of postribosomal supernatant was covalently linked to EF-2. By contrast, there was no detectable ADP-ribosylation of EF-2 from any of the 17 DT' clones. These results showed that almost all the DT' cells induced in the assay were due to alteration of elongation factor 2 itself. The high concentration of the toxin in the selection medium probably did not allow survival of variants with altered membrane permeability.

Mutagenic Activities of Heterocyclic Amines
The optimum concentration for S9 differs for each compound; for benzo(a)pyrene, the optimum concentration of S9 was 12.5 ,uL/mL of S9 mix, while for dimethylnitrosamine it was 100 piL/mL (9). Trp-P-1 showed mutagenic activity towards CHL cells in the absence of S9, and increasing the concentration of S9 resulted in reduced mutagenic activity. With Trp-P-2, the optimum concentration of S9 was 12 ,uL/mL, and there was a dose-dependent increase in the number of DT' cells with increasing concentrations of Trp-P-2. The specific mutagenic activity of Trp-P-2 was the highest among the heterocyclic amines tested.
At the optimal concentration of S9 required for induction of DT' cells by Glu-P-1 and Glu-P-2 (25 ,uL/mL) there was a dose-dependent increase in mutation frequency by both compounds. AaC also showed concentration-dependent mutagenicity with S9 optimized at 25 pLL/mL. With IQ, MeIQ and MeIQx, the maximum number of DT' cells was obtained with S9 concentration of 12 to 50 pLL/mL. With S9 at 25 uL/mL, IQ, MeIQ, and MeIQx showed concentration-dependent mutagenicities. The cytotoxic effects of these compounds on CHL cells were much less than those of other heterocyclic amines tested.
With S. typhimurium, norharman and harman do not have mutagenic activity. But, their addition to certain nonmutagenic carcinogens results in marked mutagenicity in this bacterium. Norharman and harman are called comutagens (16). These compounds, however, were found to have mutagenic activities in CHL cells without S9 mix, using DT' as a marker (10). The mutagenicities of these compounds decreased with increasing concentration of S9. The reasons for the difference in response to these compounds between CHL cells and the bacterial system remain to be clarified.

Comparison of Mutagenic Activities of Various Mutagen-Carcinogens
Mutation assay with the use of S. typhimurium has been widely and successfully employed to monitor and screen for mutagen-carcinogens in our daily life. Bacteria, however, differ from mammalian cells in the organization of genetic material, metabolism, transport, and DNA repair processes. In addition, the S. typhimurium assay measures reverse mutation, while the DT' assay with CHL cells measures forward mutation. We have measured the mutagenicities of various compounds on CHL cells. For the comparison, the specific mutagenic activities were calculated from the linear portions of the concentration-response curve at optimal S9 concentration (when required), and are presented in Table 1. Nitropyrenes are among the most mutagenic compounds on S. typhimurium TA 98 without S9 mix (17). These compounds, ubiquitous in our environment, are produced by internal combustion engines and coalfired power plants. They were also found in chicken grilled with sauce (18). The 1,6-and 1,8-dinitropyrenes without S9 mix were highly mutagenic to CHL cells, 1,3-dinitropyrene and 1,3,6-trinitropyrene being less mutagenic (8). No mutagenic activity was detected using 1-nitropyrene and 1,3,6,8-tetranitropyrene. The mutagenic activity of 1,6-and 1,8-dinitropyrenes was comparable to that of 4-nitroquinoline 1-oxide (4NQO). It should be noted that 1,3-, 1,6-, and 1,8-dinitropyrenes were carcinogenic when injected subcutaneously into mice or rats, while 1-nitropyrene failed to induce tumors upon subcutaneous administration to rats (19,20).
Among the heterocyclic amines tested, Trp-P-2 was the most mutagenic, followed in order by IQ, MeIQ, Trp-P-1, AoaC, MeIQx, Glu-P-1, and Glu-P-2. The order of the specific mutagenic activities of the compounds toward CHL cells differed from that of their activities toward S. typhimurium. It should be noted that mutagenic activity of crude material such as coffee could be measured with this assay system (11). The mutation assay with CHL cells, using diphtheria toxin resistance gives relatively reproducible results and can be recommended for use as a mammalian mutation method.