DNA repair modulates the vulnerability of the developing brain to alkylating agents.

Neurons of the developing brain are especially vulnerable to environmental agents that damage DNA (i.e., genotoxicants), but the mechanism is poorly understood. The focus of the present study is to demonstrate that DNA damage plays a key role in disrupting neurodevelopment. To examine this hypothesis, we compared the cytotoxic and DNA damaging properties of the methylating agents methylazoxymethanol (MAM) and dimethyl sulfate (DMS) and the mono- and bifunctional alkylating agents chloroethylamine (CEA) and nitrogen mustard (HN2), in granule cell neurons derived from the cerebellum of neonatal wild type mice and three transgenic DNA repair strains. Wild type cerebellar neurons were significantly more sensitive to the alkylating agents DMS and HN2 than neuronal cultures treated with MAM or the half-mustard CEA. Parallel studies with neuronal cultures from mice deficient in alkylguanine DNA glycosylase (Aag(-/-)) or O(6)-methylguanine methyltransferase (Mgmt(-/-)), revealed significant differences in the sensitivity of neurons to all four genotoxicants. Mgmt(-/-) neurons were more sensitive to MAM and HN2 than the other genotoxicants and wild type neurons treated with either alkylating agent. In contrast, Aag(-/-) neurons were for the most part significantly less sensitive than wild type or Mgmt(-/-) neurons to MAM and HN2. Aag(-/-) neurons were also significantly less sensitive than wild type neurons treated with either DMS or CEA. Granule cell development and motor function were also more severely disturbed by MAM and HN2 in Mgmt(-/-) mice than in comparably treated wild type mice. In contrast, cerebellar development and motor function were well preserved in MAM-treated Aag(-/-) or MGMT-overexpressing (Mgmt(Tg+)) mice, even as compared with wild type mice suggesting that AAG protein increases MAM toxicity, whereas MGMT protein decreases toxicity. Surprisingly, neuronal development and motor function were severely disturbed in Mgmt(Tg+) mice treated with HN2. Collectively, these in vitro and in vivo studies demonstrate that the type of DNA lesion and the efficiency of DNA repair are two important factors that determine the vulnerability of the developing brain to long-term injury by a genotoxicant.

Role of nucleotide- and base-excision repair in genotoxin-induced neuronal cell death.

Base-excision (BER) and nucleotide-excision (NER) repair play pivotal roles in protecting the genomes of dividing cells from damage by endogenous and exogenous agents (i.e. environmental genotoxins). However, their role in protecting the genome of post-mitotic neuronal cells from genotoxin-induced damage is less clear. The present study examines the role of the BER enzyme 3-alkyladenine DNA glycosylase (AAG) and the NER protein xeroderma pigmentosum group A (XPA) in protecting cerebellar neurons and astrocytes from chloroacetaldehyde (CAA) or the alkylating agent 3-methyllexitropsin (Me-Lex), which produce ethenobases or 3-methyladenine (3-MeA), respectively. Neuronal and astrocyte cell cultures prepared from the cerebellum of wild type (C57BL/6) mice or Aag(-/-) or Xpa(-/-) mice were treated with 0.1-50 microM CAA for 24h to 7 days and examined for cell viability, DNA fragmentation (TUNEL labeling), nuclear changes, and glutathione levels. Aag(-/-) neurons were more sensitive to the acute (>20 microM) and long-term (>5 microM) effects of CAA than comparably treated wild type neurons and this sensitivity correlated with the extent of DNA fragmentation and nuclear changes. Aag(-/-) neurons were also sensitive to Me-Lex at comparable concentrations of CAA. In contrast, Xpa(-/-) neurons were more sensitive than either wild type or Aag(-/-) neurons to CAA (>10 microM), but less sensitive than Aag(-/-) neurons to Me-Lex. Astrocytes from the cerebellum of wild type, Aag(-/-) or Xpa(-/-) mice were essentially insensitive to CAA at the concentrations tested. These studies demonstrate that BER and NER are required to protect neurons from genotoxin-induced cell death.

Spontaneously immortalized cell lines obtained from adult Atm null mice retain sensitivity to ionizing radiation and exhibit a mutational pattern suggestive of oxidative stress.

The study of Ataxia-telangiectasia (A-T) has benefited significantly from mouse models with knockout mutations for the Atm (A-T mutation) locus. While these models have proven useful for in vivo studies, cell cultures from Atm null embryos have been reported to grow poorly and then senesce. In this study, we initiated primary cultures from adult ears and kidneys of Atm homozygous mice and found that these cultures immortalized readily without loss of sensitivity to ionizing radiation and other Atm related cell cycle defects. A mutational analysis for loss of expression of an autosomal locus showed that ionizing radiation had a mutagenic effect. Interestingly, some spontaneous mutants exhibited a mutational pattern that is characteristic of oxidative mutagenesis. This result is consistent with chronic oxidative stress in Atm null cells. In total, the results demonstrate that permanent cell lines can be established from the tissues of adult mice homozygous for Atm and that these cell lines will exhibit expected and novel consequences of this deficiency.

Somatic cell mutations: can they provide a link between aging and cancer?

Cancers increase during aging in mammals, and an accumulating body of evidence suggests that mutational events too do likewise. Mutational events are intimately involved in the malignant process. One current view is that mutator phenotypes are required in malignant cells for a sufficient number of critical target genes to be affected. These mutator phenotypes are believed to result from underlying deficiencies in genes necessary to maintain genomic stability. This review will provide a framework for a discussion of cancer and aging by detailing with a pair of wise approach studies that address the relations between aging, cancer, and mutations. Results from these studies will be used to suggest that a mutator phenotype develops in the cells of older individuals in the absence of an underlying genetic deficiency. Instead, it is proposed that a mixture of chromosomal aberrations, DNA damage, and chronic exposure to genotoxic forces, including oxidative stress, provide the basis for this age-accelerated mutator phenotype.

Biallelic methylation and silencing of mouse Aprt in normal kidney cells.

Heritable gene silencing is an important mechanism of tumor suppressor gene inactivation in a variety of human cancers. In the present study, we show that methylation-associated silencing of the autosomal adenine phosphoribosyltransferase (Aprt) locus occurs in primary mouse kidney cells. Aprt-deficient cells were isolated from mice that were heterozygous for Aprt, i.e., they contained one wild-type Aprt allele and one targeted allele bearing an insertion of the bacterial neo gene. Although silencing of the wild-type allele alone was sufficient for the cells to become completely Aprt-deficient, biallelic methylation of the promoter region was found to occur. Moreover, despite the absence of selective pressure against the targeted allele, phenotypic silencing of the inserted neo gene accompanied silencing of the wild-type Aprt allele. A potential role for allelic homology in these events is discussed.

Tandem B1 elements located in a mouse methylation center provide a target for de novo DNA methylation.

A cis-acting methylation center that signals de novo DNA methylation is located upstream of the mouse Aprt gene. In the current study, two approaches were taken to determine if tandem B1 repetitive elements found at the 3' end of the methylation center contribute to the methylation signal. First, bisulfite genomic sequencing demonstrated that CpG sites within the B1 elements were methylated at relative levels of 43% in embryonal stem cells deficient for the maintenance DNA methyltransferase when compared with wild type embryonal stem cells. Second, the ability of the B1 elements to signal de novo methylation upon stable transfection into mouse embryonal carcinoma cells was examined. This approach demonstrated that the B1 elements were methylated de novo to a high level in the embryonal carcinoma cells and that the B1 elements acted synergistically. The results from these experiments provide strong evidence that the tandem B1 repetitive elements provide a significant fraction of the methylation center signal. By extension, they also support the hypothesis that one role for DNA methylation in mammals is to protect the genome from expression and transposition of parasitic elements.

Fully expanded FMR1 CGG repeats exhibit a length- and differentiation-dependent instability in cell hybrids that is independent of DNA methylation.

The fragile X syndrome is characterized at the molecular level by expansion and methylation of a CGG trinucleotide repeat located within the FMR1 locus. The tissues of most full mutation carriers are mosaic for repeat size, but these mutational patterns tend to be well conserved when comparing multiple tissues within an individual. Moreover, full mutation alleles are stable in cultured fibroblasts. These observations have been used to suggest that fragile X CGG repeat instability normally is limited to a period during early embryogenesis. DNA methylation of the repeat region is also believed to occur during early development, and some experimental evidence indicates that this modification may stabilize the repeats. To study the behavior of full mutation alleles in mitotic cells, we generated human-mouse somatic cell hybrids that carry both methylated and unmethylated full mutation FMR1 alleles. We observed considerable repeat instability and analyzed repeat dynamics in the hybrids as a function of DNA methylation, repeat length and cellular differentiation. Our results indicate that although DNA methylation does correlate with stability in primary human fibroblasts, it does not do so in the cell hybrids. Instead, repeat stability in the hybrids is dependent on repeat length, except in an undifferentiated cellular background where large alleles are maintained with a high degree of stability. This stability is lost when the cells undergo differentiation. These results indicate that the determinants of CGG repeat stability are more complex than generally believed, and suggest an unexpected role for cellular differentiation in this process.

The establishment and maintenance of DNA methylation patterns in mouse somatic cells.

Somatic cell DNA methylation patterns in mammals are established during embryonic development and are then maintained somewhat faithfully for the remainder of the individual's lifetime. Pattern formation can be divided into a series of linked steps that include demethylation, de novo methylation, methylation spreading, methylation blocking, and maintenance methylation. In this review, these steps will be combined to present a model for the formation and maintenance of a methylation pattern in the 5' region of the mouse Aprt gene. This model suggests that an apparently 'stable' methylation pattern results from a dynamic equilibrium between forces that promote and inhibit methylation spreading.

Solid tissues removed from ATM homozygous deficient mice do not exhibit a mutator phenotype for second-step autosomal mutations.

The presence of increased frequencies of blood-derived and solid tumors in ataxia-telangiectasia (A-T) patients, coupled with a role for the ATM (A-T mutation) protein in detecting specific forms of DNA damage, has led to the assumption of a mutator phenotype in A TM-deficient cells. Supporting this assumption are observations of increased rates of chromosomal aberrations and intrachromosomal homologous recombinational events in the cells of A-T patients. We have bred mice with knockout mutations for the selectable Aprt (adenine phosphoribosyltransferase) locus and the Atm locus to examine the frequency of second-step autosomal mutations in Atm-deficient cells. Two solid tissues were examined: (a) the ear, which yields predominately mesenchymal cells; and (b) the kidney, which yields predominately epithelial cells. We report here the lack of a mutator phenotype for inactivating autosomal mutations in solid tissues of the Atm-deficient mice.

Hypomethylation of an expanded FMR1 allele is not associated with a global DNA methylation defect.

The vast majority of fragile-X full mutations are heavily methylated throughout the expanded CGG repeat and the surrounding CpG island. Hypermethylation initiates and/or stabilizes transcriptional inactivation of the FMR1 gene, which causes the fragile X-syndrome phenotype characterized, primarily, by mental retardation. The relation between repeat expansion and hypermethylation is not well understood nor is it absolute, as demonstrated by the identification of nonretarded males who carry hypomethylated full mutations. To better characterize the methylation pattern in a patient who carries a hypomethylated full mutation of approximately 60-700 repeats, we have evaluated methylation with the McrBC endonuclease, which allows analysis of numerous sites in the FMR1 CpG island, including those located within the CGG repeat. We report that the expanded-repeat region is completely free of methylation in this full-mutation male. Significantly, this lack of methylation appears to be specific to the expanded FMR1 CGG-repeat region, because various linked and unlinked repetitive-element loci are methylated normally. This finding demonstrates that the lack of methylation in the expanded CGG-repeat region is not associated with a global defect in methylation of highly repeated DNA sequences. We also report that de novo methylation of the expanded CGG-repeat region does not occur when it is moved via microcell-mediated chromosome transfer into a de novo methylation-competent mouse embryonal carcinoma cell line.

A novel signature mutation for oxidative damage resembles a mutational pattern found commonly in human cancers.

To determine the types of mutations induced by oxidative damage, a kidney cell line with a heterozygous deficiency for the autosomal Aprt (adenine phosphoribosyltransferase) gene was tested for its mutagenic response to hydrogen peroxide. Aprt-deficient cells were selected and scored for loss of heterozygosity (LOH) for 11 microsatellite loci on mouse chromosome 8. On the basis of the LOH analysis, spontaneous mutants (n = 38) were distributed into four classes: apparent point mutation, mitotic recombination, chromosome loss, and large interstitial deletion. However, 9 of 20 (45%) hydrogen peroxide-induced mutants exhibited a novel class of mutations characterized by "discontinuous LOH" for one or more of the microsatellite loci. Interestingly, mutations resembling discontinuous LOH are commonly observed in a wide variety of human cancers. Our data suggest that discontinuous LOH is a signature mutational pattern for oxidative damage and further suggest that such genetic damage is widespread in cancer.

The primary function of a redundant Sp1 binding site in the mouse aprt gene promoter is to block epigenetic gene inactivation.

The promoter region of the mouse adenine phosphoribosyltransferase (aprt) gene contains one non-consensus Sp1 binding site at its 5' end followed by three consensus Sp1 binding sites. The two 3'-most binding sites are sufficient for maximal expression of aprt , suggesting that the non-consensus and consensus binding sites at the 5' end are redundant. However, the two 3' sites are not sufficient to block epigenetic inactivation, which led to the hypothesis that the redundant consensus and/or non-consensus 5' Sp1 binding sites are required to block inactivation events. To test this hypothesis, promoter region constructs were made in which the two 5' Sp1 binding sites were mutated alone or in tandem, and then each construct was tested for its ability to withstand epigenetic inactivation. A cis -acting methylation center that is normally located 1.2 kb upstream of the promoter was used to induce inactivation. The results demonstrate that the presence of the redundant consensus Sp1 binding site is required to block methylation-associated gene inactivation. Therefore, the Sp1 binding sites comprising the mouse aprt promoter have evolved two distinct functions, one to promote transcription and the other to block epigenetic inactivation.

Estimation of mutation frequencies in normal mammalian cells and the development of cancer.

A current controversy in cancer research is whether the rate of accumulation of mutations in normal somatic cells is sufficient to account for the number of mutations in malignant cells. This review will focus on the types of mutations that can occur in mammalian somatic cells and the frequency at which some of these mutations have been shown to occur in vivo. Human and mouse mutation detection systems will be highlighted. The possibility that distinct mutational events can arise from a common precursor will be discussed, as will the possibility that inherited and sporadic cancers can acquire mutator phenotypes at different times in tumor development. Consideration will also be given to a potential relationship between an age-related accumulation of mutations and cancer.

Formation of methylation patterns in the mammalian genome.

Cytosine methylation in mammals is an epigenetic modification required for viability of the developing embryo. It has been suggested that DNA methylation plays important roles in X-chromosome inactivation, imprinting, protection of the genome from invasive DNA sequences, and compartmentalization of the genome into active and condensed regions. Despite the significance of DNA methylation in mammalian cells, the mechanisms used to establish methylation patterns during development are not understood. This review will summarize the current state of knowledge about potential roles for cis- and trans-acting factors in the formation of methylation patterns in the mammalian genome.

A cell line selected for resistance to ionizing radiation exhibits cross resistance to other genotoxic agents and a mutator phenotype for loss of heterozygosity events.

An ionizing radiation resistant derivative was obtained from the mouse P19H22 (aprt hemizygote) embryonal carcinoma cell line by repeated exposure to 137Cs gamma radiation. Ionizing radiation resistance in the 6Gy-R cell line was not correlated with a failure to undergo cell cycle arrest or a loss of the p53 response after exposure to 137Cs gamma radiation. Moreover, the cells did not display increased resistance to bleomycin, a double strand break inducing agent. However, the cells did display increased resistance to ultraviolet radiation, ethyl methanesulfonate, and 95% oxygen. A mutational analysis demonstrated a > 700 fold-fold increase in the frequency of aprt mutants for the 6Gy-R cells, but no change in the frequency of hprt or dhfr mutants. A molecular analysis suggested that the aprt mutations in the 6Gy-R cells arose by recombinational events. A possible association between radiation resistance, DNA repair, and a mutator phenotype for large-scale mutational events is discussed.

A role for certain mouse Aprt sequences in resistance to toxic adenine analogs.

A mouse embryonal carcinoma cell line hemizygous for the adenine phosphoribosyltransferase gene (aprt) was exposed to ultraviolet light (UV) or to the alkylating agent, ethyl methanesulfonate (EMS). Thirty eight cell lines retaining the aprt gene were isolated by selecting for resistance to 2,6-diaminopurine (DAP), an adenine analogue which selects against aprt activity. Of these, six cell lines distinguished by significant levels of aprt enzymatic activity after selection in DAP, were found to carry mutations in the aprt gene affecting the apparent Km of the enzyme for adenine in every cell line, and the apparent Km for phosphoribosylpyrophosphate in two of the six cell lines. The results indicate that the ability of these cells to survive in the presence of toxic adenine analogues while maintaining significant levels of aprt enzyme activity may be due to a reduced affinity for the adenine analogue, DAP. This biochemical analysis along with results obtained from sequencing the aprt gene from 31 DAP resistant cell lines with no detectable aprt activity were used to implicate certain amino acids within aprt in substrate binding. It was also determined that, in contrast to UV, EMS did not appear to exhibit any strand bias in the distribution of mutations.

Genetic alterations distinguish different types of ovarian tumors.

Seventy-four sporadic ovarian tumors were studied for loss of heterozygosity (LOH) and microsatellite instability (MI) with 20 polymorphic markers on chromosome 17 and at least I marker on every other chromosome. Additionally, activation of the K-ras oncogene was examined through mutation analysis of codon 12. A majority of the tumors analyzed were low grade and/or of the mucinous histologic type. A negative correlation between LOH on chromosome 17 and K-ras activation was observed, with the former alteration present in the majority of high grade serous and endometrioid tumors and the latter most commonly found in the mucinous and low malignant potential (LMP) tumors. In 60% of cases where LOH on chromosome 17 was present, it was observed at all informative markers, indicating chromosome loss. In these cases, frequent events of LOH were observed on the other chromosomes. When confined events of LOH were observed on chromosome 17, fewer events of LOH were observed on the other chromosomes. In the absence of LOH on chromosome 17, LOH on other chromosomes was rare. K-ras activation was most commonly observed in tumors with no LOH events. Two endometrioid tumors and 2 mucinous tumors demonstrated MI. Our data support the involvement of different molecular pathways in the development of different types of ovarian tumors.

Genetic alterations on chromosome 17 distinguish different types of epithelial ovarian tumors.

Epithelial tumors of the ovary are the most common ovarian tumors of adult women. They exist in several different histological patterns and exhibit varying degrees of aggressiveness. Molecular genetic studies in epithelial ovarian cancer have shown that loss of heterozygosity (LOH) for regions of chromosome 17 is a common event, probably reflecting the inactivation of one or more tumor suppressor genes present on this chromosome. We examined 87 sporadic epithelial ovarian tumors of different grade and histological type at 16 loci on this chromosome and found that 35% of them showed LOH for chromosome 17. Of these, 84% showed LOH for all informative markers, suggesting that loss of the entire chromosome 17 homologue may have occurred. Interestingly, chromosome 17 loss was observed frequently in serous tumors (49%), was less common in endometrioid tumors (15%), and was rare in mucinous tumors (4%) (P = .01 and P = .0002, respectively). Our findings support the concept that the histological subtypes of epithelial ovarian cancer may be the result of different molecular genetic events.