Spontaneous and induced chromosomal aberrations and gene mutations in human lymphoblasts: mitomycin C, methylnitrosourea, and ethylnitrosourea

Error-corrected duplex sequencing enables direct detection and quantification of mutations in human TK6 cells with strong inter-laboratory consistency

Error-corrected duplex sequencing (DS) enables direct quantification of low-frequency mutations and offers tremendous potential for chemical mutagenicity assessment. We investigated the utility of DS to quantify induced mutation frequency (MF) and spectrum in human lymphoblastoid TK6 cells exposed to a prototypical DNA alkylating agent, N-ethyl-N-nitrosourea (ENU). Furthermore, we explored appropriate experimental parameters for this application, and assessed inter-laboratory reproducibility. In two independent experiments in two laboratories, TK6 cells were exposed to ENU (25-200 µM) and DNA was sequenced 48, 72, and 96 h post-exposure. A DS mutagenicity panel targeting twenty 2.4-kb regions distributed across the genome was used to sample diverse, genome-representative sequence contexts. A significant increase in MF that was unaffected by time was observed in both laboratories. Concentration-response in the MF from the two laboratories was strongly positively correlated (r = 0.97). C:G>T:A, T:A>C:G, T:A>A:T, and T:A>G:C mutations increased in consistent, concentration-dependent manners in both laboratories, with high proportions of C:G>T:A at all time points. The consistent results across the three time points suggest that 48 h may be sufficient for mutation analysis post-exposure. The target sites responded similarly between the two laboratories and revealed a higher average MF in intergenic regions. These results, demonstrating remarkable reproducibility across time and laboratory for both MF and spectrum, support the high value of DS for characterizing chemical mutagenicity in both research and regulatory evaluation.

Expression of miR-34a is a sensitive biomarker for exposure to genotoxic agents in human lymphoblastoid TK6 cells

miR-34a has been identified as a tumor suppressor microRNA (miRNA) involved in the P53 network. Its expression levels correlate to carcinogenesis, which are generally lower in tumor tissue and higher in response to DNA damage. In this study, the response of miR-34a from exposure to genotoxic agents in human lymphoblastoid TK6 cells was evaluated to assess whether the expression of this miRNA could be used as an early indicator for genotoxic damage in mammalian cells. TK6 cells were treated with seven genotoxic agents with different mode-of-actions (cisplatin, N-ethyl-N-nitrosourea, etoposide, mitomycin C, methyl methanesulphonate, taxol, and X-ray radiation) and a non-genetic toxin (usnic acid) at different concentrations for four hours (except for X-rays) and the expression levels of miR-34a were measured 24 h after the beginning of the treatments. The expression levels of miR-34a were significantly increased by these genetic toxins in a dose-dependent manner, while no significant change in miRNA expression was found in the usnic acid-treated cells. These results suggest that miR-34a can respond to genotoxic insults sensitively; thus, miR-34a expression has the potential to be used to evaluate genotoxicity of agents.

Molecular analysis of chemically-induced mutations in mammalian cell assays

Two mammalian cell mutation assays, the HPRT/V79 assay and the TK/mouse lymphoma assay, were compared for their ability to respond to the genotoxic chemicals ethyl methanesulfonate (EMS) and mitomycin C (MMC). Whereas EMS induced a high mutant frequency at both loci, MMC produced few mutants at the hprt locus, but induced a large number of mutants at the tk locus. Southern blotting analysis showed that this difference was due to the type of genetic damage induced by the two chemicals. Intragenic changes ranging from point mutations to loss of the entire gene were recovered as viable mutants at both the hprt and tk loci. Thus, EMS which causes mainly intragenic mutations induced similar mutant frequencies at both loci. The large multilocus deletions induced by MMC, in which the damage was assumed in many cases to extend into a gene essential for growth since most TK mutants were slow-growing, could not be recovered at the hprt locus. Whereas both loci will detect intergenic mutations, mutants carrying large-scale damage are recoverable only at the heterozygous tk locus. At the hemizygous hprt locus no homologous chromosome exists to provide the function of essential genes if these are lost along with hprt in multilocus deletions. Most human cancers develop as a highly complex process involving both gene and multilocus mutations in oncogenes and tumour suppressor genes. Thus the TK/mouse lymphoma assay is a more appropriate in vitro test for the detection of chemicals capable of causing the types of DNA lesions important in human cancer.

Novel Radiation Response Genes Identified in Gene-Trapped MCF10A Mammary Epithelial Cells

We have used a gene-trapping strategy to screen human mammary epithelial cells for radiation response genes. Relative mRNA expression levels of five candidate genes in MCF10A cells were analyzed, both with and without exposure to radiation. In all five cases, the trapped genes were significantly down-regulated after radiation treatment. Sequence analysis of the fusion transcripts identified the trapped genes: (1) the human androgen receptor, (2) the uncharacterized DREV1 gene, which has known homology to DNA methyltransferases, (3) the human creatine kinase gene, (4) the human eukaryotic translation elongation factor 1 beta 2, and (5) the human ribosomal protein L27. All five genes were down-regulated significantly after treatment with varying doses of ionizing radiation (0.10 to 4.0 Gy) and at varying times (2-30 h after treatment). The genes were also analyzed in human fibroblast and lymphoblastoid cell lines to determine whether the radiation response being observed was cell-type specific. The results verified that the observed radiation response was not a cell-type-specific phenomenon, suggesting that the genes play essential roles in the radiation damage control pathways. This study demonstrates the potential of the gene-trap approach for the identification and functional analysis of novel radiation response genes.

Genetic instability of cancer cells is proportional to their degree of aneuploidy

Genetic and phenotypic instability are hallmarks of cancer cells, but their cause is not clear. The leading hypothesis suggests that a poorly defined gene mutation generates genetic instability and that some of many subsequent mutations then cause cancer. Here we investigate the hypothesis that genetic instability of cancer cells is caused by aneuploidy, an abnormal balance of chromosomes. Because symmetrical segregation of chromosomes depends on exactly two copies of mitosis genes, aneuploidy involving chromosomes with mitosis genes will destabilize the karyotype. The hypothesis predicts that the degree of genetic instability should be proportional to the degree of aneuploidy. Thus it should be difficult, if not impossible, to maintain the particular karyotype of a highly aneuploid cancer cell on clonal propagation. This prediction was confirmed with clonal cultures of chemically transformed, aneuploid Chinese hamster embryo cells. It was found that the higher the ploidy factor of a clone, the more unstable was its karyotype. The ploidy factor is the quotient of the modal chromosome number divided by the normal number of the species. Transformed Chinese hamster embryo cells with a ploidy factor of 1.7 were estimated to change their karyotype at a rate of about 3% per generation, compared with 1.8% for cells with a ploidy factor of 0.95. Because the background noise of karyotyping is relatively high, the cells with low ploidy factor may be more stable than our method suggests. The karyotype instability of human colon cancer cell lines, recently analyzed by Lengnauer et al. [Lengnauer, C., Kinzler, K. W. & Vogelstein, B. (1997) Nature (London) 386, 623-627], also corresponds exactly to their degree of aneuploidy. We conclude that aneuploidy is sufficient to explain genetic instability and the resulting karyotypic and phenotypic heterogeneity of cancer cells, independent of gene mutation. Because aneuploidy has also been proposed to cause cancer, our hypothesis offers a common, unique mechanism of altering and simultaneously destabilizing normal cellular phenotypes.

Chromosome abnormalities in human beings

Constitutional chromosomal abnormalities are an important cause of miscarriage, infertility, congenital anomalies and mental retardation in humans. Most human constitutional chromosomal imbalance results from aneuploidy, a condition that appears to be much more frequent in humans than in any other species studied. Chromosomal rearrangements and segmental deletions and duplications also occur in humans, but much less often. Although treatment of human somatic cells with some environmental agents produces chromosomal damage, no measurable increase in the frequency of constitutional chromosomal abnormalities has been unequivocally demonstrated among the children of parents exposed to any agent. Recent work has provided insight into a variety of mechanisms by which chromosomal abnormalities can arise during gametogenesis and early embryogenesis. Mechanisms have also been recognized that can correct or partially compensate for chromosomal imbalance, sometimes permitting survival of conceptuses that would otherwise be lost early in gestation. This improved understanding can be used to refine future studies of the cytogenetic effects of environmental exposures.

Carcinogens stimulate intrachromosomal homologous recombination at an endogenous locus in human diploid fibroblasts

Mitotic recombination is believed to play an important role in the development of many cancers. An improved system has been developed to detect reversion of an intragenic DNA duplication, as a model for intrachromosomal homologous recombination. The 'LNtd' strain of human fibroblasts, derived from a Lesch-Nyhan donor, produces no detectable hypoxanthine phosphoribosyltransferase (HPRT) activity due to a 13.7-kilobase-pair DNA insertion duplicating exons 2 and 3 of the HPRT locus. These cells are therefore sensitive to selection in HAT medium, against cells lacking functional HPRT enzyme. Clonal reversion to HAT resistance occurs spontaneously at 1-3 x 10(-5)/cell/generation, and can be induced by brief exposure to a variety of carcinogenic agents. Six known carcinogens, including two (diethylstilbestrol and nickel chloride) which were non-mutagenic in Salmonella by Ames HIS-reversion tests, showed dose-dependent induction of LNtd reversion by a maximum of 2.4- to > 11-fold over controls (each p < 0.01). In contrast, 5 non-carcinogenic agents, including two 'Ames-positive' chemicals, sodium azide and 8-hydroxyquinoline, evoked no more than a 1.7-fold increase in reversion (not significant). The molecular events associated with reversion to HAT-resistance were characterized, relative to the parental strain, in HATR clones derived from either untreated or carcinogen-treated cells. Both the intron-3:intron-1 junction situated between the duplicated HPRT segments in LNtd cells (amplified by polymerase chain reaction), and a restriction fragment corresponding to the duplicated HPRT DNA (assessed by Southern-blot hybridization), were lost from the majority of HATR revertant clones, whether they arose spontaneously or following exposure to Cr(VI) or ultraviolet light. These results imply that HATR reversion is induced in LNtd cells by carcinogenic treatments, through a mechanism consistent with homologous recombination, and is highly concordant with induction of in vivo carcinogenesis by the same agents.

Mutagenic damage to mammalian cells by therapeutic alkylating agents

Cytotoxic alkylating agents used as therapeutics include nitrogen mustards, ethyleneimines, alkyl sulfonates, nitrosoureas and triazenes. Their reactivity with DNA, RNA and proteins can cause cell death. Side-effects of treatment include tissue toxicity and secondary malignancies, likely due to the genetic damage induced. The full mutagenic potential of alkylating agents may only be realised after they undergo metabolic activation, principally by cytochromes P450. Mutagenicity is related to the ability of alkylating agents to form crosslinks and/or transfer an alkyl group to form monoadducts in DNA. The most frequent location of adducts in the DNA is at guanines. Expressed mutations involve different base substitutions, including all types of transitions and transversions. The mutational spectra of alkylating agents on mammalian cells is distinct from that induced in bacterial cells, reflecting the different codon usage by bacteria and differences in DNA repair and replication enzymes. Mutations are induced by busulfan, chlorambucil (CAB), cyclophosphamide (CP, or its metabolite), dacarbazine, mechlorethamine, melphalan, mitomycin-C (MMC), nitrosoureas and thiotepa. Although dose-dependent, the relationship is not always linear. The molarities at which alkylating agents induce cell killing and mutations vary over three orders of magnitude. The mutagenic efficiency, of alkylating agents also varies, with some agents inducing three times more mutations for equivalent cell killing. The induction of micronuclei, sister chromatid exchanges, or chromosome aberrations is variable, but has been observed for CP, CAB, MMC, melphalan and triethylenemelamine. There is insufficient information to determine whether any synergistic effects of alkylating agents used in combination will influence the cytotoxic and mutagenic damage equally. Understanding the potential synergy of alkylating agents at the cellular and molecular level should allow improvement of the therapeutic efficacy of alkylating agents without increasing the unwanted mutation induction.

Low mutagenic effects of mitomycin C in undifferentiated embryonic P19 cells are correlated with efficient cell cycle control

Pluripotent undifferentiated embryonic carcinoma cells of line P19 and their differentiated progeny, epithelioid ectoderm-like EPI-7 cells, showed different responses to mitomycin C (MMC) with respect to induction of micronuclei, mutations at the HPRT-locus and cell cycle control. Cytotoxic effects of MMC after a 5-h treatment were lower in undifferentiated P19 cells than in differentiated EPI-7 cells with IC50 values of 1.3 and 0.25 microM for P19 and EPI-7 cells, respectively. MMC did not induce 6-thioguanine-resistant mutants in P19 cells but significantly increased the mutation frequency in EPI-7 cells with concentrations of 0.25, 0.5 and 1.0 microM MMC. Micronuclei determined by flow-cytometry were induced by MMC in both cell lines at equitoxic concentrations of 4.5 (P19) and 0.75 (EPI-7) microM, reducing the viability in both cell lines to 10%. Whereas the induction of micronuclei in P19 cells was maximal 28 h after treatment and declined thereafter, micronucleus induction peaked 48 h post treatment in EPI-7 cells and remained significantly increased even 67 h after the treatment. Flow-cytometric determination of the distribution of MMC-treated P19 and EPI-7 within the cell cycle revealed a distinct G2/M-block in P19 cells, whereas EPI-7 cells showed normal progression through S-phase and a negligible G2/M-block. Therefore, we conclude that the lower effectivity of MMC to induce gene mutations and micronuclei in P19 cells seemed to be correlated with a more efficient cell cycle control in undifferentiated compared to differentiated EPI-7 cells.

Lower mutation frequencies are induced by ENU in undifferentiated embryonic cells than in differentiated cells of the mouse in vitro

The pluripotent embryonic carcinoma cells of line P19 established from undifferentiated cells of the early mouse embryo and their differentiated progeny, the epithelioid ectoderm-like EPI-7 cells, were investigated for the induction of mutations at the HPRT locus by the alkylating agent N-ethyl-N-nitrosourea (ENU). We showed that the cytotoxic effects of ENU after a 5-h treatment were lower in undifferentiated P19 cells than in differentiated EPI-7 cells. The IC50 values of ENU in the two cell lines amounted to 0.6 mg/ml and 0.09 mg/ml for P19 and EPI-7 cells, respectively. The induction of 6-thioguanine-resistant mutants by ENU (1.0 mg/ml) determined after an expression time of 8 days for both cell lines resulted in similar mutation frequencies. Using expression times of 8 days for P19 and 11.75 days for EPI-7 cells, taking into account the longer generation time of differentiated EPI-7 cells (13.7 +/- 3.6 h) in comparison to undifferentiated P19 cells (9.3 +/- 0.9 h), ENU induced significantly higher mutant frequencies in EPI-7 cells (4865 mutants/10(6) cells) than in P19 cells (282 mutants/10(6) cells). Our results and data from the literature on UV irradiation-induced repair support the idea that the induction of lower mutation frequencies in embryonic cells may correlate with different proliferation capacities, cell cycle parameters and/or different mechanisms of DNA repair in embryonic stem cells and differentiated cells, respectively.

N-nitroso-N-methylurea-induced rat mammary tumors arise from cells with preexisting oncogenic Hras1 gene mutations

GGA to GAA mutations in the 12th codon of the Hras gene are frequently observed in rat mammary tumors induced by N-nitroso-N-methylurea (NMU). We developed an assay to measure point mutations present in tissues at a frequency of 10(-5) and have now applied this assay to measure the specific G to A transition of the Hras gene in rat mammary epithelium. We find that (i) 70% of untreated rats contain detectable levels of Hras mutants; (ii) these mutants are clustered within the gland as sectors in a manner consistent with their origin as a mutation arising during early organ development; and (iii) treatment with a carcinogenic dose of NMU did not result in a significant increase in the number of such mutants, the fraction of organ sectors with mutant cells, or the fraction of animals containing detectable levels of ras mutants. We conclude that the NMU-induced mammary tumors carrying the G to A transition at the 12th codon of the Hras gene arose from preexisting ras mutants and that an independent effect of NMU was directly or indirectly responsible for tumor formation.

Molecular analysis of mutations at the tk locus of L5178Y mouse-lymphoma cells induced by ethyl methanesulphonate and mitomycin C

Mutations at the tk locus of mouse-lymphoma L5178Y cells were induced by treatment with ethyl methanesulphonate (EMS), primarily a point mutagen and mitomycin C (MMC), a potent clastogen. Mutant colony size was distinctly bimodal with 35% of spontaneous mutants growing as small colonies and 65% large. The proportion of small colonies increased only slightly to 41% in EMS-treated cultures but to 64% after MMC treatment. Mutations were analysed by Southern and Northern blotting. Digestion of DNA with the restriction enzyme, Nco I, revealed that many mutants had lost a 6.3-kb fragment which constituted the loss of the entire tk gene. Almost all of the EMS-induced large-colony mutants analysed (9/10) retained the tk+ allele suggesting the presence of an intragenic mutation. Of the small-colony mutants, half (6/12) had lost the tk+ gene and presumably other genes affecting growth and half retained the tk+ allele suggesting point mutations in both the tk gene and other sites in the genome affecting growth. A very different spectrum of mutation was induced with MMC. Only 1/12 of the large-colony mutants were due to intragenic mutation, the remaining large-colony mutants having lost the tk+ allele while all the small-colony mutants had lost the tk+ gene presumably with the deletion extending to genes essential for normal growth. Northern blot analysis showed no changes in the size of tk transcript in any mutants. Alterations in the amount of tk mRNA were not detectable since all mutants produced an mRNA of similar size and amount, which may indicate the production of an abnormal mRNA from the tk- allele. Unlike cell-mutation assays that use hemizygous loci (such as hprt+/0) for detecting potential chemical carcinogens, the mouse-lymphoma tk+/- assay allows the recovery of both intragenic and intergenic mutations thus enabling the detection of both point mutagens such as EMS and potent clastogens like MMC.

Molecular analysis of mutation at the hprt locus of Chinese hamster V79 cells induced by ethyl methanesulphonate and mitomycin C

Mutations at the hprt locus of Chinese hamster V79 cells were induced by treatment with ethyl methanesulphonate (EMS), considered primarily a point mutagen and mitomycin C (MMC), a potent clastogen. EMS gave a dose-dependent induction of mutants while MMC induced a poor mutagenic response. Mutations were analysed using Southern and Northern blotting. Analysis of 9 EMS-induced and 4 spontaneous mutants yielded no detectable alterations in the hprt locus after digestion of DNA with 6 restriction enzymes. Mutants without detectable changes carried presumptive point mutations. In contrast, 4 out of 12 MMC-induced mutants had detectable alterations. 2 of these appeared to have lost the entire hprt gene while the other 2 had probable partial deletions. For these 4 deletion mutants no hprt mRNA was detected. 3 MMC-induced and 1 EMS-induced mutants had reduced levels of hprt mRNA. All the other mutants showed normal levels of hprt mRNA and the message detected was always of the correct size. It is suggested that the poor mutagenic response induced by MMC may be due to the lethal nature of large deletions involving both the hemizygous hprt locus and adjacent essential genes. This may lead to an underestimate of the mutagenicity of clastogenic agents such as MMC in the V79 HPRT mutation assay.

Studies of mutagenicity and clastogenicity of 5-azacytidine in human lymphoblasts and Salmonella typhimurium

5-Azacytidine (5-AzaC) induced mutation in the TK+/- human lymphoblastoid line, TK6, at both the thymidine kinase (tk) locus as measured by resistance to trifluorothymidine (F3TdR), and the hypoxanthine-guanine phosphoribosyltransferase (hgprt) locus, as measured by resistance to 6-thioguanine (6TG). F3TdRR and 6TGR mutant fractions induced by 5-AzaC were observed after a normal phenotypic expression time and remained stable. Interestingly, 5-AzaC was 5-10 times more mutagenic at the tk locus than the hgprt locus. However, F3TdRR colonies from 5-AzaC-treated cultures behaved like TK-deficient mutants induced by other chemical mutagens. The TK or HGPRT phenotype had no effect on the toxicity of 5-AzaC, thus eliminating differential toxicity as a potential cause for the observed higher mutability at the tk locus. 5-AzaC did not induce F3TdRR cells in the parental TK+/+ lymphoblastoid line, indicating that 5-AzaC-induced F3TdRR variants were not due to a dominant alteration in gene expression. 5-AzaC did not induce chromosomal aberrations in TK6 cells, eliminating clastogenic events as a potential cause for the higher mutability at the tk locus. 5-AzaC was also found to be mutagenic in a forward mutation assay to 8-azaguanine resistance in Salmonella typhimurium.