Metal mutagenesis in transgenic Chinese hamster cell lines.

Metals are toxic agents for which genotoxic effects are often difficult to demonstrate. To study metal mutagenesis, we have used two stable hprt/gpt+ transgenic cell lines that were derived from Chinese hamster V79 cells. Both the G12 and G10 cell lines are known to be very sensitive to clastogens such as X-rays and bleomycin, with the mutagenic response of the integrated xanthine guanine phosphoribosyl transferase (gpt) gene in G10 usually exceeding that of the same gene in the transgenic G12 cells. In studies with carcinogenic insoluble nickel compounds, a high level of mutagenesis was found at the gpt locus of G12 cells but not at the endogenous hypoxanthine phosphoribosyl transferase (hprt) locus of V79 cells. We have since demonstrated the similar recovery of a high frequency of viable G12 mutants with other insoluble nickel salts including nickel oxides (black and green). The relative mutant yield for the insoluble nickel compounds (G12 > G10) is the opposite of that obtained with nonmetal clastogens (G10 > G12). In the G12 cells, nickel mutagenesis may be related to the integration of the gpt sequence into a heterochromatic region of the genome. For some of the insoluble nickel compounds, significant inhibition of both cytotoxicity and mutant yield resulted when the G12 cells were pretreated with vitamin E. In comparison with the nickel studies, the mutagenic responses to chromium compounds in these cell lines were not as dramatic. Mutagenesis of the gpt target could not be demonstrated with other metals such as mercury or vanadium.


Introduction
For many metals, toxicity, genotoxicity, and carcinogenicity have been readily established. To define the mechanisms that produce these deleterious effects, gene-specific mutation analysis can be particularly useful. With metals, however, it has often been difficult to obtain mammalian mutagenesis levels that are high enough to allow definition of the mutagenic spectra. Nickel compounds, for example, are generally weak mutagens in mammalian cells (1)(2)(3)(4) In comparison, mutagenesis can usually be demonstrated for chromium compounds in most bacterial and mammalian cell systems; however, chromium does not elicit a high mutagenic response at the hprt gene in Chinese hamster V79 cells (5,6).
One of the rationales for developing transgenic V79 cell lines was to enhance our ability to detect and characterize genespecific, metal-induced mutations. It was known that integrated plasmid-derived shuttle vector sequences were often more mutable than endogenous mammalian genes such as hprt (7,8). Additionally, the introduction of a small bacterial gene target into Chinese hamster V79 cells could provide a readily recoverable mutagenesis target sequence (7). This would facilitate mutation analysis by molecular methods induding PCR and DNA sequencing. The enzyme encoded by the E. coli gpt gene is functionally equivalent to the enzyme produced by the mammalian hprt gene. The bacterial gpt gene has previously been introduced into several mammalian host cell lines, generating transgenic cells which have been very useful in mutagenesis studies (9)(10)(11)(12). Prior to the transfection of the pSV2gpt plasmid into our Chinese hamster V79 cells, these host cells were mutated by UV to generate an irreversible hprr phenotype. The successful recovery of numerous stable gpt+ transfectants of V79 cells was subsequently described (13).
Two hprv/gpt+ transgenic V79 cell lines, G12 and G10, have since been characterized in extensive mutagenesis studies (13,14). Each of these cell lines maintains a single integration site for the plasmid sequence in different locations of the Chinese hamster genome. Both the G 12 and G 10 cell lines have been shown to be very sensitive to clastogens such as X-rays and bleomycin, and the mutagenic response of the integrated gpt gene in G10 is greater than that of the same gene in the transgenic G12 cells (14). For both cell lines, X-ray mutagenesis of the integrated gpt locus exceeds that which can be recovered at the endogenous hprt locus of the parental Chinese hamster V79 cells. These cell lines are not generally hypermutable, however, as demonstrated by spontaneous and alkylation mutagenesis of the endogenous Na+K+/ATPase locus (14). Having established the utility of the transgenic cell lines for mutagenesis studies, the G12 and G 10 cells were employed to investigate whether significant levels of gpt mutagenesis could be induced by carcinogenic and toxic metals. This manuscript summarizes data that was obtained from our studies of several metals, including soluble and insoluble nickel compounds, chromate, vanadium, and mercury.

Materials and Methods
Celi CWukre The hprt/lgpt+ transgenic cell lines G12 and GIO, and the parental Chinese hamster V79 cells, are cultured in F12 medium (GIBCO, Grand Island, NY) with 5% fetal bovine serum (GIBCO) and 1% peni-Environmental Health Perspectives cillin/streptomycin (GIBCO) at 37'C and 5% CO2 (13,14). To ensure low spontaneous mutagenesis levels, the G12 and G10 cultures are initiated every 6 weeks from frozen stocks and are supplemented with HAT (100 pM hypoxanthine, 1 pM aminopterin, 100 pM thymidine). V79 cells are also frequently defrosted, but are not supplemented with HAT. Mutagenesis The mutagenesis protocol for detecting 6thioguanine (6-TG) resistance as an indicator of genetic alteration or nonexpression of the gpt gene in the transgenic cell lines has been previously reported (13,14). The protocol is a standard reseeding assay which is essentially similar to that used to detect hprtmutants among V79 cells. Optimum mutant yields can be obtained when the treated cells are reseeded into selection media (F 12 with 10 pg/ml 6-TG) 6 days after treatment, and the mutants are permitted to grow for an additional 13 days. The reseeding plating efficiency is also determined at this time, by seeding a small number of cells into F12 without 6-TG. At the end of the selection period, the mutants are simultaneously fixed and stained with Giemsa (50% Giemsa solution, 50% methanol). Mutant frequencies are calculated by totaling the number of mutants obtained on 10 mutagenesis plates per metal dose, and dividing that number by the reseeding plating efficiency of each treated and recovered population of cells. The cytotoxicity of each treatment is measured by a colony-forming assay in which 300 or 500 cells are seeded in triplicate in 60 mm dishes. Following cell attachment, the cells are treated, rinsed, supplemented with fresh F12, and allowed to grow undisturbed for 7 days before staining and counting.
Nickel mutagenesis studies were performed as previously described (15). Briefly, V79 cells and the transgenic cells were exposed to sterilized, ground (<5 pm) particles of insoluble nickel compounds (nickel sulfide, nickel subsulfide, nickel oxides; INCO, Ltd., Toronto, Ontario) for 24 hr in complete media. The nickel doses are expressed as pg nickel/cm2 surface area of the tissue culture dishes, since these insoluble particles tend to settle out during the 24-hr treatment (16,17). The nickel doses chosen for the mutagenesis studies were determined by preliminary cytotoxicity studies, such that all the different nickel compounds were investigated at generally equitoxic dose ranges. As a positive mutagenesis control, N-methyl-NM-nitro-N-nitrosoguanidine (MNNG, Sigma Chemical Co., St. Louis, MO) was routinely used. Potassium chromate (K2CrO4, Fisher Scientific, Chemical Division, Fair Lawn, NJ) treatments were for 2 to 5 hr (18), at 37°C in either Earle's Balanced Salt Solution (EBSS; GIBCO,Grand Island, NY) or magnesium-supplemented salts glucose media (50 mM HEPES buffer, 100 mM NaCl, 5 mM KCI, 2 mM CaCl2, 0.8 mM MgSO4, pH 7.2). With chromium oxide (Fisher), the cells were exposed 1-6 pM Cr03 for 24 hr in complete media. Vanadium exposures were with ammonium metavanadate, NH4VO3 U.T. Baker Chemicals, Phillipsburg, NJ), for 2 or 24 hr in serum-free F12, SGM, or EBSS as previously reported (5). For the mercury studies, V79, G12, and GIO cells were treated with 3 to 50 pg/mL mercuric chloride (Sigma Chemical Co.) or methyl mercury(II) chloride (Johnson Matthey Electronics, Ward Hill, MA) for 2 hr in EBSS.

Results and Discussion
In preliminary studies with nickel compounds, a high level of 6-TG was found to be induced in G12 cells with insoluble nickel sulfide (NiS) but not with soluble nickel sulfate (NiSO4) (16). These compounds were not mutagenic at the hprt locus of the parental V79 cells. In another study, a high level of gpt mutagenesis was also found with insoluble Ni3S2 in G12 cells (17). We have since confirmed these findings, and have recently shown similar levels of induction of 6-TG resistance following the exposure of G12 cells to insoluble nickel oxides (15). We have also investigated the nickel-induced response of the integrated gpt gene in the GO0 cell line. Figure 1 shows that the cytotoxicity of NiS is similar for the two transgenic cell lines, G12 and G10. These data agree with those from our studies with X-rays, bleomycin, UV, and MNNG (14) in which large cellline differences in cytotoxicity are not generally observed, whereas the site of gpt integration and its mutability differ. Figure 2 compares the relative toxicity of insoluble NiS and Ni3S2 in the G12 cells. We have studied all the nickel compounds over essentially equitoxic dose ranges (10-100% survival), and find that much higher particle concentrations are required for the nickel oxides than for the nickel sulfides (Figures 3,4). Green nickel oxide was significantly less toxic than the black oxide particles. The high levels of induction of 6-TG resistance by nickel sulfide and subsulfide in G12 cells are shown in Figure 3A; while similar results with the nickel oxides are shown in Figure 3B. With soluble NiCl2, mutagenesis of the gpt locus did not exceed 2 to 4 times the spontaneous levels in either G12 or GlO cells (15). This is in agreement with the data obtained previously for soluble NiSO4 in G12 cells (16,17), and correlates well with carcinogenicity studies in which the insoluble nickel compounds are the most active due to their phagocytic uptake (19). In our studies with the G12 cells, we could microscopically visualize phagocytized intracellular particles of NiS and Ni3S2 at the end of the 24-hr treatment, but could not see phagocytized particles of NiO (15). Although nickel oxide particles may not be phagocytized as readily as the nickel sulfides (20), there are indications that these particles may remain associated with the cell membranes for long periods (21). Since the treated cells are not disturbed in our assay for up to 6 days after the end of  the treatment,, the opportunity exists for useful for defining the mechanisms of prolonged cellular association of the nickel nickel carcinogenesis. oxides. The transgenic G10 cells did not What is the nature of these nickel-respond to the insoluble nickel compounds induced mutants, and why do they arise at with the same intensity as seen above for such a high frequency? Preliminary studies G 12 cells (Figure 4), in contrast to our data of the mutagenic alterations within the gpt with X-rays and bleomycin (14). Nickel sequence of several 6-TG-resistant G12 mutagenesis could only be induced to colonies have thus far failed to show base about three times that of the spontaneous changes or deletions involving the coding frequency for the G10 cell line. The cell region of the gptgene (17). In studies with line differences in gpt mutagenesis are another set of mutants, preliminary data clearly evident as summarized on the tables strongly suggest that most of the nickel-for both nonmetal mutagens (Table 1) and induced 6-TG-resistant colonies were vari-metals (Table 2). In the G 12 cells, the ants, whose phenotype could be reverted to nickel effects may be related to the integra-6-TG sensitivity by subsequent treatment tion of the gpt sequence into a heterochrowith nickel or 5-azacytidine (22). Nickel matic region of the genome. The specificity appears to be unique in this regard, since for nickel interactions with heterochro-DNA sequence changes, as well as total or matin are well documented (19,24). Using partial deletions of the gpt gene sequence, fluorescent in situ hybridization of a biotinamong spontaneous, X-ray, bleomycin, and labeled gpt probe to the metaphase chroeven chromate-induced 6-TG-resistant mosomes of G12 cells, we have localized mutants of both G 12 and G10 cells have the integrated gpt sequence to a subtelombeen observed (14,23). In fact, the gpt eric region of chromosome 1 (25). Based locus in the G 12 cells may be an important on the C-banding pattern of Chinese hamnickel-sensitive locus, and studies of the ster V79 chromosomes (25), the gpt modulation of gpt gene expression may be sequence of the V79-derived G 12 cells tne process or pnagocytosos, ana tne ongoing generation of reactive oxygen radicals, may play important roles in nickel genotoxicity.
In comparison with the nickel studies, the mutagenic responses to chromium compounds in the transgenic cell lines were not as dramatic. Relative to spontaneous mutagenesis levels, the maximal chromateinduced mutant yield is only three times the background yield for both G12 and   Chromium mutagenesis studies, however, are complicated for several reasons. We have observed that the prolonged exposure of attached cells to chromium compounds often results in cellular detachment from the treatment dishes (18). As a result, premutant cells are likely to be lost from the chromium-treated populations. It is also difficult to define an effective chromium treatment dosing regime, since chromium compounds are often mutagenic over a very narrow dose range (5,6,18,28). This may be partly related to residual toxicity that could result from trivalent Cr that is trapped within the cells (30). Residual chromate toxicity, as indicated by reduced plating efficiency, has been observed up to one week after the treatment of V79, G12 and G10 cells (18). Mutagenesis of the gpt targets in G12 and GIO cells could not be demonstrated with other metals such as mercury or vanadium. Figure 5 shows the toxicity and mutagenesis data obtained in experiments with methyl mercury chloride. As noted in Figure 5A, the G0 cells seem to exhibit a slight resistance to the toxicity of this agent. This effect has not been further studied. Mutagenesis could not be induced at the gpt locus of either G 12 or G10 cells, nor could it be demonstrated at the hprt gene in the parental V79 cells. These findings suggest that mercury compounds are generally too toxic to permit cell survival as required for mutagenesis, and are consistent with the literature on mercury genotoxicity (31,32). In studies with vanadate ( Table 2), weak mutagenesis was demonstrated at the hprt gene of V79 cells, and at the gpt locus of G12 cells, but was not investigated with G 10 cells (5).
In conclusion, metal mutagenesis is difficult to quantitate in mammalian cell systems. For most of the metal studies summarized here, the level of induced gpt mutagenesis was less than five times the spontaneous frequency. The transgenic G 12 cell line, however, is sensitive to insoluble nickel compounds, and will be used in future studies of nickel modulation of gpt gene expression. These studies may ultimately be useful in determining the mechanisms involved in nickel carcinogenesis.