Arsenite is a cocarcinogen with solar ultraviolet radiation for mouse skin: an animal model for arsenic carcinogenesis.

Although epidemiological evidence shows an association between arsenic in drinking water and increased risk of skin, lung, and bladder cancers, arsenic compounds are not animal carcinogens. The lack of animal models has hindered mechanistic studies of arsenic carcinogenesis. Previously, this laboratory found that low concentrations of arsenite (the likely environmental carcinogen) which are not mutagenic can enhance the mutagenicity of other agents, including ultraviolet radiation (UVR). This enhancing effect appears to result from inhibition of DNA repair by arsenite. Recently we found that low concentrations of arsenite disrupted p53 function and upregulated cyclin D1. These results suggest that the failure to find an animal model for arsenic carcinogenesis is because arsenite is not a carcinogen per se, but rather acts as an enhancing agent (cocarcinogen) with a genotoxic partner. We tested this hypothesis with solar UVR as carcinogenic stimulus in hairless Skh1 mice. Mice given 10 mg/l sodium arsenite in drinking water for 26 weeks had a 2.4-fold increase in yield of tumors after 1.7 KJ/m(2) UVR three times weekly compared with mice given UVR alone. No tumors appeared in mice given arsenite alone. The tumors were mostly squamous cell carcinomas, and those occurring in mice given UVR plus arsenite appeared earlier and were much larger and more invasive than in mice given UVR alone. These results are consistent with the hypothesis that arsenic acts as a cocarcinogen with a second (genotoxic) agent by inhibiting DNA repair and/or enhancing positive growth signaling.

Genes upregulated in lead-resistant glioma cells reveal possible targets for lead-induced developmental neurotoxicity.

Identifying genes upregulated in lead-resistant cells should give insight into lead toxicity and cellular protective mechanisms and may also result in identification of proteins that may be useful as biomarkers. Glial cells are thought to protect neurons against heavy metals. Rat glioma C6 cells share many properties of normal glial cells. To identify and analyze genes upregulated in a lead-resistant variant, PbR11, suppression subtractive hybridization (SSH) between mRNAs of wild-type and PbR11 cells was performed. Sequencing and database searches identified three genes, thrombospondin-1, heparin sulfate 6-sulfotransferase, and neuropilin-1, which play important roles in angiogenesis and axon growth during development. Two genes, HSP90 and UBA3, are involved in the ubiquitin-proteosome system. One gene was identified as that of a rat endogenous retrovirus and another, 2C9, is a transcript expressed in fos-transformed cells. PbR11 also overexpresses c-fos. Expression of these genes and effects of short-term lead exposure (24 h, up to 600 microM) on their expression in C6 cells was examined. The rat endogenous retrovirus and 2C9 are expressed only in PbR11 cells, and show no expression, either constitutive or lead-induced, in wild-type C6 cells. HSP90 is expressed at low level constitutively in C6 cells, but can be induced in a dose-dependent manner by lead. In contrast, thrombospondin-1 is repressed in a dose-dependent manner by lead. The other genes (HS6ST, neuropilin, and UBA3) show low constitutive expression and are neither upregulated nor downregulated by exposure to lead. We suggest that neuropilin-1, heparin sulfate 6-sulfotransferase, and thrombospondin-1 may be important targets for lead-induced developmental neurotoxicity.

Reduction of spontaneous mutagenesis in mismatch repair-deficient and proficient cells by dietary antioxidants.

Cells lacking mismatch repair (MMR) exhibit elevated levels of spontaneous mutagenesis. Evidence exists that MMR is involved in repair of some DNA lesions besides mismatches. If some oxidative DNA lesions are substrates for MMR, then the excess mutagenesis in MMR(-) cells might be blocked by dietary antioxidants. Effects of the dietary antioxidants ascorbate, alpha-tocopherol, (-)-epigallocatechin gallate (EGCG) and lycopene on spontaneous mutagenesis were studied using mismatch repair-deficient (hMLH1(-)) human colon carcinoma HCT116 cells and HCT116/ch3 cells, in which normal human chromosome 3 has been added to restore mismatch repair. HCT116 cells have a 22-fold higher spontaneous mutation rate compared with HCT116/ch3 cells. HCT116 cells cultured in 1% fetal bovine serum (FBS) have twice the spontaneous mutation rate of those cultured in 10% FBS, most likely due to reduction in serum antioxidants in the low serum medium. As expected, alpha-tocopherol (50 microM) and ascorbate (284 microM) reduced spontaneous mutagenesis in HCT116 cells growing in 1% serum more dramatically than in cells cultured in 10% serum. The strongest antimutagenic compound was lycopene (5 microM), which reduced spontaneous mutagenesis equally (about 70%) in HCT116 cells growing in 10 and 1% FBS and in HCT116/ch3 cells. Since lycopene was equally antimutagenic in cells growing in low and high serum, it may have another antimutagenic mechanism in addition to its antioxidant effect. Surprisingly, EGCG (10 microM) was toxic to cells growing in low serum. It also reduced spontaneous mutagenesis equally (nearly 40%) in HCT116 and HCT116/ch3 cells. The large proportion of spontaneous mutagenesis that can be blocked by antioxidants in mismatch repair-deficient cells support the hypothesis that a major cause of their excess mutagenesis is endogenous oxidants. Blocking spontaneous mutagenesis, perhaps with a cocktail of antioxidants, should reduce the risk of cancer in people with a genetic defect in mismatch repair as well as other individuals.

Effects of arsenite on p53, p21 and cyclin D expression in normal human fibroblasts -- a possible mechanism for arsenite's comutagenicity.

Arsenite, the most likely environmental carcinogenic form of arsenic, is not significantly mutagenic at non-toxic concentrations, but is able to enhance the mutagenicity of other agents. Evidence suggests that this comutagenic effect of arsenite is due to inhibition of DNA repair, but no specific repair enzyme has been found to be sensitive to low (<1 microM) concentrations of arsenite. To determine whether arsenite affects signaling which might alter DNA repair, this study assesses the effect of arsenite on p53-related signal transduction pathways after ionizing radiation. Long-term (14 day) low dose (0.1 microM) arsenite caused a modest increase in p53 expression in WI38 normal human fibroblasts, while only toxic (50 microM) concentrations increased p53 levels after short-term (18 h) exposure. When cells were irradiated (6 Gy), p53 and p21 protein concentrations were increased after 4h, as expected. Both long-term, low dose and short-term, high dose exposure to arsenite greatly suppressed the radiation-induced increase in p21 abundance. In addition, long-term, low dose (but not short-term, high dose) exposure to arsenite resulted in increased expression of cyclin D1. These results show that in cells treated with arsenite, p53-dependent increase in p21 expression, normally a block to cell cycle progression after DNA damage, is deficient. At the same time, low (non-toxic) exposure to arsenite enhances positive growth signaling. We suggest that the absence of normal p53 functioning, along with increased positive growth signaling in the presence of DNA damage may result in defective DNA repair and account for the comutagenic effects of arsenite.

Cloning genes whose levels of expression are altered by metals: implications for human health research.

When cells are exposed to toxicants, changes in gene expression ensue. To date, there is little information on gene expression changes induced by metals in mammalian cells. The basic methods for identifying altered gene expression of both a temporary and a permanent nature are outlined, with examples drawn mostly from what is known about metal-induced changes in gene expression. The application of this information in the development of new biomarkers of exposure and effect, in identifying individuals with altered susceptibility to metal compounds, and in the choice of genes for microarrays is discussed.

Expression cloning for arsenite-resistance resulted in isolation of tumor-suppressor fau cDNA: possible involvement of the ubiquitin system in arsenic carcinogenesis.

Arsenic is a human carcinogen whose mechanism of action is unknown. Previously, this laboratory demonstrated that arsenite acts as a comutagen by interfering with DNA repair, although a specific DNA repair enzyme sensitive to arsenite has not been identified. A number of stable arsenite-sensitive and arsenite-resistant sublines of Chinese hamster V79 cells have now been isolated. In order to gain understanding of possible targets for arsenite's action, one arsenite-resistant subline, As/R28A, was chosen as a donor for a cDNA expression library. The library from arsenite-induced As/R28A cells was transfected into arsenite-sensitive As/S5 cells, and transfectants were selected for arsenite-resistance. Two cDNAs, asr1 and asr2, which confer arsenite resistance to arsenite-hypersensitive As/S5 cells as well as to wild-type cells, were isolated. asr1 shows almost complete homology with the rat fau gene, a tumor suppressor gene which contains a ubiquitin-like region fused to S30 ribosomal protein. Arsenite was previously shown to inhibit ubiquitin-dependent proteolysis. These results suggest that the tumor suppressor fau gene product or some other aspect of the ubiquitin system may be a target for arsenic toxicity and that disruption of the ubiquitin system may contribute to the genotoxicity and carcinogenicity of arsenite.

Spontaneous mutagenesis in mammalian cells is caused mainly by oxidative events and can be blocked by antioxidants and metallothionein.

Little is known about endogenous processes causing spontaneous mutagenesis in mammalian cells. To study this problem, a mathematical model and method developed previously in our laboratory was used to measure the spontaneous mutation rate in mammalian cells at the transgenic gpt locus in Chinese hamster G12 cells. We found that spontaneous mutagenesis increased when cells were cultured in low (<0.25%) serum. These cells also contained higher oxidant levels, measured by dichloroflourescein (DCF) fluorescence, suggesting that the elevated spontaneous mutagenesis resulted from endogenous oxidants which are normally quenched by serum antioxidants. This was found to be the case. Spontaneous mutagenesis was significantly reduced in serum-depleted as well as control cells when catalase (100 ng/ml) or the antioxidants ascorbate (50 microg/ml) or mannitol (100-500 microg/ml) were added to the medium. Overexpression of metallothionein in these cells also suppressed spontaneous mutagenesis and mutagenesis induced by oxygen radical-generating compounds. Cells expressing metallothionein antisense RNA become mutators. Taken together, these results suggest that the major cause of spontaneous mutagenesis in mammalian cells is endogenously-generated oxidative DNA damage which can be blocked by metallothionein or by dietary antioxidants carried by the blood supply.

Isolation and properties of lead-resistant variants of rat glioma cells.

Glial cells are thought to protect neurons from heavy-metal toxicity. To gain a better understanding of mechanisms of protection against lead compounds, a number of lead-resistant C6 rat glioma cell sublines have been isolated. After 8 mo of growth in the absence of lead nitrate, three sublines still maintain their lead-resistant phenotype. None of the lead-resistant sublines are cross-resistant to Cd(II) or Ni(II), but all are cross-resistant (in varying degrees) to Hg(II), As(III), Sb(III), and Sn(II), and one is resistant to trimethyl tin. No inducible lead resistance is seen in any glioma line. One subline has been used to create cell-cell hybrids with wild-type cells. The hybrids exhibit dominance of the lead-resistant phenotype. To identify and analyze altered gene expression at the mRNA level in the lead-resistant sublines, the differential display technique was used. Numerous differences are seen between amplified fragments from wild-type and lead-resistant cells. Candidate clones are now being analyzed to confirm the differential expression and to isolate cDNAs that confer lead resistance.

Human cells lack the inducible tolerance to arsenite seen in hamster cells.

Chinese hamster V79 cells and their arsenite-resistant variants were found to have an arsenite- and antimonite-inducible tolerance mechanism which protects against the subsequent cytotoxic effects of arsenate, arsenate and antimonite. Inducible tolerance requires de novo mRNA and protein synthesis, and is independent of the heat shock response. In contrast, we report that the arsenite hypersensitive variant line As/S27D lacks the inducible tolerance response. Numerous attempts were made to detect an inducible tolerance response to arsenite in a variety of human cells. An assay based on Neutral red uptake was used in order to study inducible tolerance in cells with poor clonability. Neither normal diploid cells nor human tumor cells of different origins were found to elicit an inducible tolerance response to arsenite. This finding may help to explain why rodents do not develop tumors after exposure to arsenite, while humans do. In addition, all human cell lines tested were much more sensitive to arsenite compared to Chinese hamster cells. Human keratinocytes were especially sensitive. In general, human cells resemble arsenic hypersensitive Chinese hamster As/R27D cells, which have lost a protective mechanism found in wild-type Chinese hamster cells.

Chinese hamster cells expressing antisense to metallothionein become spontaneous mutators.

The functions of metallothioneins (MTs) have been debated for at least a decade. Because it seems unlikely that they evolved only to protect cells against exogenous heavy metals, it has been suggested that MTs have roles in scavenging reactive intermediates, controlling zinc and copper homeostasis, and controlling transfer of zinc to transcription factors and other proteins. Previously, we demonstrated that Chinese hamster G12 cells which overexpress MT have greatly reduced spontaneous mutation rates, suggesting that MT evolved to prevent spontaneous mutagenesis induced by free nuclear zinc ions. We have now isolated G12 transfectants which express antisense RNA to MT. Immunofluorescent staining reveals MT protein in both the nucleus and the cytoplasm in parental cells. A clone expressing high levels of antisense RNA (AMT30) shows reduced basal and induced levels of MT protein. AMT30 cells are hypersensitive to cadmium, zinc, copper and mercury chlorides as well as to menadione. Glutathione levels in AMT30 and G12 cells do not differ. AMT30 cells are spontaneous mutators, showing a spontaneous mutation rate 5-10 times that of G12 cells or G12 cells transfected with vector alone. Only transfectants which show a high level of MT antisense expression (i.e., AMT30) had greatly elevated spontaneous mutation rates. These results support our hypothesis that a major role of MT is to act as an endogenous antimutagen probably via scavenging of reactive intermediates in the nucleus. AMT30 cells should be useful in delineating the sources of spontaneous mutagenesis.

Maximum likelihood estimation of spontaneous mutation rates from large initial populations.

When estimating a spontaneous mutation rate from either a single culture (C=1) or from the C parallel cultures (C>1) of a fluctuation experiment, the use of a large initial population size N0 to seed each culture will permit a gaussian approximation for the probability distribution of the number M of mutants at the time when the culture(s) has (have) grown to size N=N02g, i.e., experienced g doublings. Using this gaussian approximation we find that the maximum likelihood estimate mu of the expected number mu of mutants present in a culture in generation g is (exactly) (equation: see text) where r = 2g / g and M 2 is the average of the squares of the C mutant counts. The maximum likelihood estimate p of the unknown mutation rate p is p = 2 mu / gN assuming an 'ideal' experiment and that there were no mutants in the initial population. A well-behaved maximum likelihood estimate is known to be efficient in large samples and we illustrate by Monte Carlo simulation that indeed p is better (has smaller mean squared error) than our previous (Rossman et al., 1995) estimator (equation: see text) (M is the average mutant count) provided N0 is of the order 1/p or larger. This advantage exists even without a fluctuation experiment, i.e., for C = 1.

Efflux-mediated resistance to arsenicals in arsenic-resistant and -hypersensitive Chinese hamster cells.

Several Chinese hamster V79 cell line variants resistant to arsenite and one arsenite-hypersensitive variant have been isolated. The basis for the variation in arsenite sensitivity was studied by transport experiments using radiolabeled arsenite. Two arsenite-resistant variants (As/R7 and As/R27) exhibited decreased accumulation of arsenite, and the hypersensitive variant (As/S5) exhibited increased arsenite accumulation compared with the parental line. Cells depleted of endogenous energy reserves were loaded with radiolabeled arsenite, and the rate of arsenic efflux was measured. Arsenite-resistant variants exhibited an increased rate of efflux, while the hypersensitive variant exhibited a decreased efflux rate. Efflux was decreased in cells incubated with the protonophore carbonyl cyanide m-chlorophenylhydrazine, demonstrating its energy dependence. Two inhibitors of glutathione S-transferase also decreased arsenite efflux, suggesting the involvement of an arsenite-glutathione complex. However, separation of the products of extrusion and the intracellular arsenic species by paper chromatography followed by autoradiography failed to show the appearance of an arsenite-glutathione complex in either case. Rather, all label in the product of the transport reaction appeared to be arsenite whether cells were loaded with arsenate or arsenite, indicating first that intracellular reduction of As(V) to As(III) had occurred and second that the arsenite was transported as an unconjugated species. All intracellular label was associated with high-molecular-weight material, possibly protein. Our results demonstrate the existence of an energy-dependent arsenical efflux pump in mammalian cells and show that arsenic is extruded as arsenite.

Serum deprivation, but not inhibition of growth per se, induces a hypermutable state in Chinese hamster G12 cells.

Spontaneous mutagenesis is thought to play a crucial role in spontaneous carcinogenesis. We recently described a new mathematical model for estimation of the spontaneous mutation rate (mutation/gene/generations) based on the assumption that mutations are fixed in the S-phase of the cell cycle. With this definition, the spontaneous mutation rate should be independent of the growth rate. In the present study, we tested this hypothesis, using cell line G12, a transgenic Chinese hamster V79 derivative, which contains a single copy of the Escherichia coli gpt gene as a target for mutagenesis. The growth rate was modulated by varying the serum concentration or the seeding density, or by addition of suramin, transforming growth factor beta, or dichlorobenzimidazole riboside to the medium. Significant increases in the spontaneous mutation rate occurred when cell proliferation was blocked by serum deprivation. Density-dependent inhibition of growth and inhibition of growth by suramin, transforming growth factor beta, or dichlorobenzimidazole riboside did not result in significant increases in spontaneous mutation rates. The level of oxidants in cells cultivated in the presence of low concentrations of serum was higher compared to control cells, suggesting that the increases in the spontaneous mutation rates under low serum conditions may be partly a result of oxidative stress due to a lack of serum antioxidants. This was shown to be the case, because spontaneous mutation rates were significantly reduced in serum-depleted cells when antioxidants were added to the medium. We suggest that during carcinogenesis, when tumors are in a prevascularized state, the spontaneous mutation rate may be elevated, and this process may contribute to the genetic instability of the tumor cells.

Mutations and infinity: improved statistical methods for estimating spontaneous rates.

Certain mathematical artifacts which had been appended by others to Luria and Delbrück's [Genetics 28: 491-511, 1943] model of spontaneous mutagenesis in bacterial populations have added confusion to the modeling and measurement of spontaneous mutation rates. Additional confusion arises when models which had been tuned for experiments with bacterial cultures grown from a small inoculum are adapted for use with mammalian cell cultures grown from a large initial population. As one consequence, biologists still tend to grow the large number of parallel cultures required by the fluctuation test in order to avoid large errors due to the high variability in the number of mutants in a growing culture. By avoiding models with infinite mean values and certain mathematical approximations that lead to conceptual and practical difficulties, the large variance of the number of mutants can be avoided (and the precision of the estimated mutation rate controlled) through the use of sufficiently large initial cell populations. A direct consequence is that simpler experiments with fewer cultures may suffice. In this paper, after a discussion of the confusions, we extend our previous approach [Rossman et al.: Mutat Res 328:21-30, 1995] by giving improved formulas for the standard error of the estimated mutation rate. The improvement results from using a more inclusive model based on consideration of the variability due to both the biological phenomenon of the growing culture (growth and mutation) and the protocols used for selection (sampling and plating efficiency). Also included is the situation where the initial cell population is not assumed to be free of mutants but the initial mutant fraction is measured instead. These standard error formulas are useful in planning experiments that yield mutation rate estimates with planned precision and for comparing and testing hypotheses about mutation rates in two or more populations which are grown under different conditions.

Modeling and measurement of the spontaneous mutation rate in mammalian cells.

The study of spontaneous mutation rates in mammalian cells has been hampered by the lack of an alternative to the cumbersome Luria and Delbrück fluctuation test. A brief review of mathematical treatments of spontaneous mutagenesis, along with some of the limitations of the fluctuation test, is presented. A new experimental method and a simple mathematical model for deriving the spontaneous mutation rate are described. Data from the transgenic Chinese hamster G12 cell line growing at two different rates is analyzed according to this model. The results support the concept that, at least for growing cells, the spontaneous mutation rate is independent of the growth rate, and the mutant fraction increases in a linear fashion with the number of generations.

A role for metallothionein and zinc in spontaneous mutagenesis.

G12, a transgenic Chinese hamster V79 cell derivative which contains a single copy of the Escherichia coli gpt gene as a target for mutagenesis, has little constitutive metallothionein (MT) expression. It was transfected with a vector containing the mouse MT-I gene, and MT-I-overproducing lines were isolated. MT-I transfectants had lower spontaneous mutation frequencies compared with the G12 parental cell line. Mutagenesis by alkylating agents was unchanged. MT expression in G12 and MT transfectants could be modulated by exposure to Zn(II) or Cd(II). The spontaneous mutation frequencies in Zn(II)- and Cd(II)-treated cells was inversely related to MT expression. In G12 cells grown in concentrations of Zn(II) up to 12 microM, a significant dose-dependent increase in spontaneous mutagenesis was observed. At higher (but subtoxic) concentrations in which endogenous MT was induced, a dramatic decrease in spontaneous mutagenesis was observed. In contrast, MT-I transfectants exhibited much lower spontaneous mutagenesis after growth in all concentrations of Zn(II). These data demonstrate a possible role for MT in modulating spontaneous mutagenesis and point to a role for Zn(II) in contributing to spontaneous mutagenesis. Because there is variability in human MT expression, low MT expression might be a risk factor for cancer.

Induction of arsenite tolerance and thermotolerance by arsenite occur by different mechanisms.

Both V79 and As/R28A cells (an arsenite-resistant Chinese hamster V79 cell variant) show increased resistance to toxic concentrations of arsenite after pretreatment with a nontoxic concentration. The induced tolerance can be completely inhibited by actinomycin D or cycloheximide. Pretreatment with a nontoxic heat shock (45 degrees C, 10 min) resulted in a clear increased thermotolerance in both cell lines but failed to induce arsenite tolerance in either cell line. Pretreatment with arsenite induced a thermotolerance in V79 cells but not in As/R28A cells. These results are consistent with a model whereby the signal for induction of arsenite tolerance involves binding of arsenite to a protein effector which is amplified in the As/R28A line, thereby preventing action of arsenite in the regulation of heat shock factor which induces the heat shock response.

Large-scale supercoiled plasmid preparation by acidic phenol extraction.

A novel method for large-scale plasmid preparation is described. Crude extracts are subjected to acidic phenol extraction to remove any contaminants present in the aqueous phase. The supercoiled plasmid DNA, which preferentially remains in the organic phase and inter-phase, is extracted back into the aqueous phase with 1.5 M TRIZMA base, from which it is precipitated. The resultant plasmid DNA is highly pure and satisfactory for any subsequent procedures. The method is extremely economical and takes only 3-4 h.

Transgenic gpt+ V79 cell lines differ in their mutagenic response to clastogens.

Several gpt+ transgenic cell lines were derived from hprt V79 cells to study mutagenesis mechanisms in mammalian cells. The G12 cell line was previously shown to be hypermutable by X-rays and UV at the gpt locus compared to the endogenous hprt gene of the parental V79 cells (Klein and Rossman, 1990), and is now shown to be highly mutable by the clastogenic anti-tumor agent bleomycin sulfate. A second transgenic cell line G10, which has a different gpt insertion site, was studied in comparison with G12. Both G12 and G10 cell lines carry the stable gpt locus at a single integration site in the Chinese hamster genome, and neither spontaneously deletes the integrated gpt sequence at a high frequency. Although spontaneous mutation to 6-thioguanine resistance in G10 cells is 3-4 times higher than in G12 cells, the cell lines differ to a much greater extent when mutated by clastogens. In comparison to G12 cells, the gpt locus in G10 cells is up to 13 times more sensitive to bleomycin mutagenesis and 5 times more responsive to X-ray mutagenesis. In contrast, there is much less difference in UV-induced mutagenesis and no differences in mutagenesis induced by alkylating agents such as N-methyl-N'-nitro-N-nitrosoguanidine (MNNG). The dose-dependent decrease in survival of the transgenic cells is the same for all mutagens tested, and does not differ from that of V79 cells. Neither transgenic cell line is generally hypermutable, since mutagenesis at an endogenous gene, Na+K+/ATPase, is similar to that of the parental V79 cell line. Although both cell lines can be induced to delete the transgene following clastogen exposure, deletions are not the only recovered mutations, and the cell lines can also be used to study mutations within the PCR recoverable gpt gene. The utility of these transgenic cells to investigate genome position effects related to mammalian mutagenesis mechanisms is discussed.