Evidence for the lack of base-change and small-deletion mutation induction by trichloroethylene in lacZ transgenic mice.

Trichloroethylene (TCE) is a widely used industrial solvent employed mainly for degreasing and cold-cleaning metal parts. It is also used for dry cleaning, and in the production of a number of chemical products. It has been shown to induce liver and lung tumors in rodents, and have a variety of positive and negative results using in vitro and in vivo mutagenicity tests. In order to assist in the interpretation of the mechanism of carcinogenicity, TCE was tested for the ability to induce gene mutations and small deletions using the lacZ transgenic mouse model (MutaMouse). Male and female animals were exposed by inhalation to 0, 203, 1153, and 3141 ppm TCE, 6 h per day for 12 days. 14 and 60 days following the last exposure, animals were sacrificed and the mutation frequency in bone marrow, kidney, spleen, liver, lung, and testicular germ cells determined. The results of this study indicate that TCE did not induce base-change or small-deletion mutations as detected in this assay in any of the tissues examined. Environ. Mol. Mutagen. 34: 190-194, 1999. Published 1999 Wiley-Liss, Inc.

Analysis of tissue-specific lacZ mutations induced by N-nitrosodibenzylamine in transgenic mice.

N-Nitrosodibenzylamine (NDBzA) is a contaminant found frequently in rubber baby bottle nipples and pacifiers. To evaluate more fully the mutagenic potential and analyse the molecular nature of possible mutations induced in vivo, we have studied the mutagenicity of NDBzA in vivo using the MutaMouse system. NDBzA, suspended in olive oil, was administered orally once to male mice at different doses (0, 30, 100, 425 and 750 mg/kg) and the mice were killed 30 and 90 days after treatment. As a positive control, and to compare relative mutagenicity, N-nitrosodimethylamine (NDMA) was also administered to animals in the same experiment at doses of 0, 2, 6 and 10 mg/kg. Mutant frequencies were increased in both 30 and 90 day liver samples, but not in bone marrow, after both NDBzA and NDMA treatment. However, NDBzA was >100 times less mutagenic than NDMA. A total of 81 mutants obtained from liver samples of treated animals (750 mg/kg) were characterized by DNA sequencing. While spontaneous mutations in transgenic mice have been characterized previously by a preponderance of G:C-->A:T transitions, mainly at 5'-CpG-3' dinucleotide sites, the predominant type of NDBzA-induced mutation in this study was transversion, mainly G:C-->T:A changes. The molecular characteristics of mutations induced by NDBzA indicate that they may arise from specific unidentified DNA adducts and benzylation appears to be the primary mechanism involved in formation of these DNA adducts.

Toward an understanding of the use of transgenic mice for the detection of gene mutations in germ cells.

Recently-developed transgenic models have provided unprecedented access to rodent somatic and germ line tissues for the study of gene mutation in vivo. While mutations in germ cells are considered an important aspect of any regulatory assessment of the risks posed by chemicals, currently-available conventional tests, which involve the study of thousands of offspring make it impractical to test large numbers of chemicals, for the induction of inherited gene mutations. When effects in germ cells per se, rather than offspring are acceptable targets, transgenic mouse assays may provide a practical alternative. As part of an international collaborative study to begin to determine the reliability, efficacy, and role of such assays, lacZ transgenic mice (Muta Mouse) were treated with single i.p. doses of ethylnitrosourea (ENU), methyl methanesulfonate (MMS), and isopropyl methanesulfonate (iPMS), and mutant frequencies determined using phenyl-beta-D-galactoside (p-gal) positive selection. For studies using germ cells, the selection of sampling times and target cells is crucial. Spermatagonial stem cells and cells in post-spermatagonial stem cell stages are the critical target cell populations of regulatory importance. Cell populations within these categories were studied by sampling germ cells isolated from seminiferous tubules and spermatozoa from the epididymis at 91 days and 25 days after treatment. The data show that ENU and iPMS induced mutations in post-spermatagonial stem cells and spermatagonial stem cells. However, MMS did not induce mutations in either cell type, or at either sampling time, at doses approaching lethality. This result is possibly because MMS induces preferentially large lesions and chromosomal aberrations (as opposed to point mutations), which are not readily detectable with bacteriophage-based shuttle vectors. Since MMS-induced specific locus and dominant lethal mutations are induced only after the mid-spermatid stage, it is also possible that the timing used missed this effect. While the ENU and iPMS data in this study demonstrate the suitability of the lacZ male transgenic mice for the study of gene mutations in post-spermatagonial stem cells and spermatagonial stem cells by sampling cells isolated from seminiferous tubules at selected times after treatment, the MMS results do not answer fully whether transgenic mouse mutation assays can detect mutations resulting from lesions induced after the mid-spermatid stage when most cellular processing is retarded. Nevertheless, it appears clear from presently available information, that the bacteriophage-based lacZ transgenic model is suitable for the detection of gene mutations in spermatogonial stem cells, spermatocytes, and early spermatids.

Comparison of the molecular characteristics of lacZ transgenic mouse mutations detected by visual and positive selection.

lacZ gene mutations in the transgenic Muta Mmouse can be detected by two different selection systems. While mutant frequencies recovered by phenyl-beta-D-galactoside (P-gal) selection are comparable with those obtained using 5-bromo-4-chloro-3-indolyl beta-D-galactopyranoside (X-gal) as substrate for beta-galactosidase, there may still be differences at the molecular level in the mutations detected by these two methods. Accordingly, we have examined a spectrum of mutants recovered from the X-gal system by regrowing these mutants on the P-gal plates. All colourless X-gal mutants grew normally on the P-gal plates. However, 11 out of 53 single light blue mutants, which express partial beta-galactosidase activity, produce few or no plaques on the P-gal plates, indicating the possible loss of some mutations using the positive selection system. Further analysis of mutant phenotypes and base changes indicates that such loss is not mutation-type specific. Our data suggest that the positive selection method can detect the majority of lacZ mutations detectable by visual selection and is more efficient at detecting mutants within a reduced range of beta-galactosidase activity.

Temporal and molecular characteristics of lacZ mutations in somatic tissues of transgenic mice.

In order to help establish criteria for optimizing protocols for in vivo mutation studies, lacZ transgenic mice (Muta mouse) were treated with five consecutive daily doses of ethylnitrosourea (50 mg/kg), sampled at times up to 55 days after treatment, and mutant frequencies and DNA sequences determined for liver and bone marrow. In the bone marrow, the mutant frequency rose very rapidly in the first 5 days after treatment to 34 times the control frequency. Subsequently, there was a brood peak where the mutant frequency did not vary significantly, although it did appear to begin to decline after 45 days. In contrast, in the liver, the peak mutant frequency (11 times the control frequency) was not achieved until 35 days, after which there appeared to be a slow decline up to 55 days, which was not statistically significant. Once the maximum mutant frequency was reached, the mutation spectra in the two tissues were indistinguishable. In contrast to the G:C-->A:T transitions in 5'-CpG sites characteristic of untreated mice, A:T-->T:A transversions and A:T-->G:C transitions were prominent in both liver and bone marrow of ENU-treated mice, suggesting the involvement of unrepaired O2- and O4-ethylthymine adducts. In addition, G:C-->T:A transversions were induced in liver. This study demonstrates the possibility that although tissues may have different mutation fixation times, a single mutation fixation time equal to the longest time may be appropriate for in vivo mutation studies, provided that the mutation frequency does not decline appreciably after the peak is reached. This study also illustrates the necessity of ensuring that mutation characteristics are determined after optimal fixation has occurred.

Temporal and molecular characteristics of mutations induced by ethylnitrosourea in germ cells isolated from seminiferous tubules and in spermatozoa of lacZ transgenic mice.

The lacZ transgenic mouse (Muta mouse) model was used to examine the timing of ethylnitrosourea (ENU)-induced mutations in germ cells. The spectrum of mutations was also determined. Animals received five daily treatments with ENU at 50 mg/kg and were sampled at times up to 55 days after treatment. In mixed germ-cell populations isolated from seminiferous tubules, there was little increase in the mutant frequency 5 days after treatment; subsequently, there was a continuous increase until the maximum (17.5-fold above background) was reached by approximately 35 days. In the spermatozoa, an increase in mutant frequency was not seen until 20 days after treatment, with the maximum (4.3-fold above background) being achieved no sooner than approximately 35 days. Based on the timing of sampling, these data demonstrate the detection of both spermatogonial and postspermatogonial, mutations. The most prominent feature of the ENU-induced base-pair mutations in testicular germ cells sampled 55 days after treatment is that 70% are induced in A.T base pairs, compared to only 16% in spontaneous mutations. These findings are consistent with comparable data from ENU studies using assays for inherited germ-cell mutations in mice. This study has demonstrated the utility and potential of the transgenic mouse lacZ model (Muta mouse) for the detection and study of germ-cell mutations and provides guidance in the selection of simplified treatment and sampling protocols.

Evidence for in vivo non-mutagenicity of the carcinogen hydrazine sulfate in target tissues of lacZ transgenic mice.

Transgenic mouse models permit the confirmation of in vitro mutagenicity in vivo without the constraints in the selection of tissues imposed by other in vivo assays. This feature is of particular importance in the determination of mutagenicity in the target tissues of carcinogens, especially those that are in vitro mutagens. Such information is critical in the determination of whether a chemical is carcinogenic via a genotoxic or non-genotoxic mechanism. Hydrazine sulfate is an in vitro mutagen that induces lung and liver tumours in mice. Transgenic mice from strain 40.6 (Mutamouse) were administered single oral doses up to a toxic concentration (400 mg/kg). No dose induced any lacZ mutations in lung, liver or bone marrow. Since the highest single dose used is higher than the cumulative dose that induced tumours in previous studies, it may be that either hydrazine sulfate is genotoxic in target tissues in vivo only when given in multiple doses or that it is a non-genotoxic carcinogen.

Sequence spectra of spontaneous lacZ gene mutations in transgenic mouse somatic and germline tissues.

As a critical step in determining whether transgenic mouse gene mutation systems are suitable models for the detection and quantification of induced gene mutations in vivo, spontaneous mutant frequencies and mutation spectra have been characterized for liver, bone marrow, and male germ cells of the lacZ transgenic mouse strain 40.6. The lacZ transgene is carried on a recombinant bacteriophage lambda shuttle vector that is recovered from mouse genomic DNA, and analysed in vitro for mutations that occurred in the mouse tissues. Mutations are detected visually as clear or pale blue plaques when X-gal is the substrate for beta-galactosidase; whereas, the wild-type plaques are dark blue. There was no statistical difference in the mutant frequency among the three tissues studied, the pooled mutant frequency being 2.23 +/- 0.41 per 10(5) pfu. The predominant type of mutation was GC-->AT transitions, with most occurring in 5'-CpG dinucleotides, suggesting that the deamination of 5-methylcytosine is the main mechanism of mutagenesis. There was, however, a statistically significant difference in the base pair substitution mutation spectrum for the liver and bone marrow when mutations were grouped according to GC or AT base-pairs. The proportion of transition versus transversion mutations was also statistically different among the three tissues, resulting mainly from the fact that germ cells were different from both bone marrow and liver. A lower number of spontaneous transitions in male germ cells was accompanied by an increase in transversions, with the proportion of GC-->AT transitions in 5'-CpG sites also declining.(ABSTRACT TRUNCATED AT 250 WORDS)

Mutagenicity of peroxyacetyl nitrate (PAN) in vivo: tests for somatic mutations and chromosomal aberrations.

A series of experiments was conducted in which Chinese hamsters inhaled PAN, an ubiquitous pollutant that is present in the atmosphere at concentrations that are as high as, or higher than, other known genotoxic agents. The animals were exposed to PAN in air at concentrations of approximately 3 ppm for up to 1 month and then examined for somatic mutations and chromosomal aberrations. Mutations were assayed by measuring the frequency of thioguanine-resistant lung fibroblasts (isolated de novo and cultured). Chromosomal aberrations were assayed by measuring the frequency of micronuclei in either the bone marrow (polychromatic erythrocytes) or the lungs (binucleate lung fibroblasts cultured in the presence of cytochalasin B). The results for the test animals were compared to those from animals exposed similarly, but without PAN. Although in each experiment the mutation frequencies for the test animals were higher than the corresponding controls, the mutation frequencies were not significantly different from the concurrent negative controls (P > .05) or the historical controls, except for experiment C. In experiment C, there was a significant regression of mutation frequency versus dose (P < 0.001) if all of the historical controls for pooled animals are included at zero dose. No reproducible evidence of chromosomal breakage was found in either lung or bone marrow. Thus, although PAN has been found to be a bacterial mutagen, we did not find statistically significant evidence of mutagenicity in vivo. The toxicity of PAN limited the exposure concentration that could be used. When all of the PAN data were used, the best estimate of the mutagenic potency proved to be comparable to that of ethylene dibromide, a carcinogenic atmospheric pollutant.

Detection of somatic mutations in vivo in lung fibroblasts, I. Spontaneous frequencies in Chinese hamsters and F344 rats.

The frequency of mutant cells observed in a series of experiments in which primary cells were isolated from untreated or solvent control animals are reported. The mutants detected are thioguanine-resistant lung fibroblasts isolated de novo from Chinese hamsters or Fischer 344 rats. The results in the two species were very similar. The distribution of mutant colonies in cells isolated from untreated animals is not random (Poissonian) but rather shows an excess of mutant clusters. No significant difference was detected between males and females. The results provide the information necessary to define the appropriate conditions for the negative controls for a routine assay for mutations induced in vivo.

Concurrent detection of gene mutations and chromosomal aberrations induced in vivo in somatic cells.

A simple direct assay for gene mutations in vivo is the missing assay in genetic toxicology. Here we report that both gene mutations and chromosomal aberrations induced in vivo can be reliably detected and quantified in a single cell type. Lung cells were isolated and cultured from Chinese hamsters that had been exposed to model mutagenic carcinogens, and then analysed in culture for genetic alterations. Chromosomal aberrations were assayed by the frequency of micronuclei found in binucleate cells after growth in the presence of cytochalasin B. Mutations were assayed by the frequency of thioguanine-resistant colonies on the 10th day of culture. X-rays and ethyl nitrosourea produced detectable increases in both mutations and chromosomal aberrations. Ethyl methanesulphonate produced large numbers of mutations but no chromosomal aberrations and methyl methanesulphonate produced the inverse: large numbers of chromosomal aberrations but no mutations. Our experience with these agents and the procedure suggests a standard protocol for the assay, and confirms the value of measuring both gene mutations and chromosomal aberrations.

Genotoxicity of two fecal steroids in murine colonic epithelium assessed by the sister chromatid exchange technique.

Two components of human feces are known to induce nuclear anomalies in mice when applied intrarectally, but to be nonmutagenic in Salmonella. We have tested these two compounds for their ability to induce sister chromatid exchanges in the colonic epithelium of mice, the same tissue in which they induce nuclear anomalies when administered by the same route. One, 4-cholesten-3-one, induced sister chromatid exchanges whereas the other, 5-alpha-cholestan-3-one did not, even at the maximum feasible dose. The results suggest that 4-cholesten-3-one is more likely to be a significant factor in human colon cancer than the 5-alpha analog.

Screening for colon carcinogens--a new strategy.

The major defect of in vivo assays for mutagenic carcinogens may be tissue specificity: a cancer bioassay of a single tissue would not be expected to detect all carcinogens, so the failure of a genetic assay in a single tissue to detect all carcinogens should not be surprising. In the search for an environmental carcinogen responsible for a specific cancer in a particular population, however, it may be that tissue specificity can be advantageous. Assays for genotoxicity directly in the target cells may have higher success rates with fewer false positives than assays in tissues of convenience. For example, to facilitate the search for one or more dietary carcinogens responsible for the high rate of colon cancer in North America, assays for genotoxicity in the target cells themselves, the colonic epithelium, may be useful. To this end we have investigated assays for three different endpoints: nuclear anomalies, sister chromatid exchanges, and gene mutations. Our experience may prove useful for others considering a similar strategy.

Induction of sister chromatid exchanges in murine colonic tissue.

As cancers of the large bowel arise primarily in the epithelial cells of the colon, measurement of genetic damage induced in those cells may aid in elucidating contributing factors in the etiology of the disease. Accordingly, procedures for the in vivo measurement of sister chromatid exchanges (SCE) have been adapted for use with colonic cells. Agar-coated 5-bromo-2'-deoxyuridine (BrdUrd) tablets were implanted (s.c.) in mice and colonic tissue prepared for measurement of SCE 48 hr later. When mice were treated by intraperitoneal injection with 20 mg/kg 1,2-dimethylhydrazine (DMH), a colon carcinogen, a significant increase over control SCE frequencies was found when treatment with carcinogen occurred in the interval from 14 hr before to 36 hr following BrdUrd administration. Treatment with DMH 2 hr following implantation of BrdUrd produced a dose-dependent increase in SCE frequencies, with the highest dose (20 mg/kg) resulting in frequencies approximately twice those of controls. Dimethylhydrazine was not found to be effective in inducing SCE in bone marrow cells of mice. As SCE can be induced by carcinogens and mutagens, measurement of SCE in colonic epithelium may be useful in identifying potential colon carcinogens.

Autoradiographic assay of mutants resistant to diphtheria toxin in mammalian cells in vitro.

Diphtheria toxin kills mammalian cells by ribosylating elongation factor 2, a protein factor necessary for protein synthesis. The frequency of cells able to form colonies in the presence of the toxin can be used as an assay for mutation to diphtheria toxin resistance. We report here that resistance to diphtheria toxin can also be detected autoradiographically in cells exposed to [3H]leucine after treatment with the toxin. In cultures of Chinese hamster ovary cells, the frequency of such resistant cells is increased by exposure of the cells to gamma-rays, ultraviolet light, ethylnitrosourea, mitomycin c, ethidium bromide, and 5-bromo-2'-deoxyuridine in a dose- and time-dependent manner. The resistant cells form discrete microcolonies if they are allowed to divide several times before intoxication, which indicates that they are genuine mutants. The assay is potentially adaptable to any cell population that can be intoxicated with diphtheria toxin and labeled with [3H]leucine, whether or not the cells can form colonies. It may be useful, therefore, for measuring mutation rates in slowly growing or nondividing cell populations such as breast, brain, and liver, as well as in cells that do divide but cannot be readily cloned, such as the colonic epithelium.