Formation of benzo[a]pyrene diol epoxide-DNA adducts at specific guanines within K-ras and p53 gene sequences: stable isotope-labeling mass spectrometry approach

The mutagenicity of a prominent tobacco carcinogen, benzo[a]pyrene (B[a]P), is believed to result from chemical reactions between its diol epoxide metabolite, (+)-anti-7r,8t-dihydroxy-c9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE), and DNA, producing promutagenic lesions, e.g., (+)-trans-anti-7R,8S,9S-trihydroxy-10S-(N(2)-deoxyguanosyl)-7,8,9,10-tetrahydrobenzo[a]pyrene (N(2)-BPDE-dG). Previous studies used the DNA repair enzyme UvrABC endonuclease in combination with ligation-mediated PCR (LMPCR) to demonstrate an increased reactivity of BPDE toward guanine nucleobases within codons 157, 248, and 273 of the p53 tumor suppressor gene (Denissenko, M. F., Pao, A., Tang, M., and Pfeifer, G. P. Science 274, 430-432). These sites are also "hot spots" for mutations observed in lung tumors of smokers, suggesting an involvement of B[a]P in the initiation of lung cancer. However, the LMPCR approach relies on the ability of the repair enzyme to excise BPDE-induced lesions, and thus the slowly repaired lesions may escape detection. Furthermore, BPDE-DNA adduct structure and stereochemistry cannot be determined. In the present work, we performed a direct quantitative analysis of N(2)-BPDE-dG originating from specific guanine nucleobases within p53- and K-ras-derived DNA sequences by using a stable isotope labeling-mass spectrometry approach recently developed in our laboratory. (15)N-labeled dG was placed at defined positions within DNA sequences derived from the K-ras proto-oncogene and p53 tumor suppressor gene, the two genes most frequently mutated in smoking-induced lung cancer. (15)N-labeled DNA was annealed to the complementary strands, followed by BPDE treatment and liquid chromatography-electrospray ionization tandem mass spectrometry analysis (HPLC-ESI-MS/MS) of N(2)-BPDE-dG lesions. The extent of adduct formation at (15)N-labeled guanine was determined directly from the HPLC-ESI-MS/MS peak area ratios of (15)N-N(2)-BPDE-dG and N(2)-BPDE-dG. BPDE-induced guanine adducts were produced nonrandomly along K-ras and p53 gene-derived DNA sequences, with over 5-fold differences in adduct formation depending on sequence context. N(2)-BPDE-dG yield was enhanced by the presence of 5-Me substituent at the cytosine base-paired with the target guanine nucleobase, an endogenous DNA modification characteristic for CpG dinucleotides within the p53 gene. In the K-ras-derived DNA sequence, the majority of N(2)-BPDE-dG adducts originated from the first position of the codon 12 (GGT), consistent with the large number of G --> T transversions observed at this nucleotide in smoking-induced lung cancer. On the contrary, the pattern of N(2)-BPDE-dG formation within the p53 exon 5 sequences did not correlate with the mutational spectrum in lung cancer, suggesting that factors other than N(2)-BPDE-dG formation are responsible for these mutations. The stable isotope labeling HPLC-ESI-MS/MS approach described in this work is universally applicable to studies of modifications to isolated DNA by other carcinogens and alkylating drugs.     

p53 mutational spectra and the role of methylated CpG sequences

The occurrence of tumor-specific mutational spectra in the p53 mutation database provides indirect evidence that implicates certain exogenous and possibly endogenous mutagenic events in human carcinogenesis. In some cases, the distribution of DNA damage along the p53 gene caused by environmental carcinogens can be correlated with the mutational spectra, i.e. hotspots and types of mutations of certain cancers, most notably for nonmelanoma skin cancers and lung cancers in smokers. This concept has been validated by experiments with sunlight and cigarette smoke components representing the polycyclic aromatic hydrocarbon class of carcinogens. A disproportionally high number of mutations in p53 (and other genes) are found at methylated CpG dinucleotides. These sequences are particularly prone to mutagenesis involving endogenous events as well as modification by exogenous carcinogens.     

Methylation of CpG dinucleotides in the lacI gene of the Big Blue® transgenic mouse

Cytosine residues at CpG dinucleotides can be methylated by endogenous methyltransferases in mammalian cells. The resulting 5-methylcytosine base may undergo spontaneous deamination to form thymine causing G/C to A/T transition mutations. Methylated CpGs also can form preferential targets for environmental mutagens and carcinogens. The Big Blue® transgenic mouse has been used to investigate tissue and organ specificity of mutations and to deduce mutational mechanisms in a mammal in vivo. The transgenic mouse contains approximately 40 concatenated lambda-like shuttle vectors, each of which contains one copy of an Escherichia coli lacI gene as a mutational target. lacI mutations in lambda transgenic mice are characterized by a high frequency of spontaneous mutations targeted to CpG dinucleotides suggesting an important contribution from methylation-mediated events. To study the methylation status of CpGs in the lacI gene, we have mapped the distribution of 5-methylcytosines along the DNA-binding domain and flanking sequences of the lacI gene of transgenic mice. We analyzed genomic DNA from various tissues including thymus, liver, testis, and DNA derived from two thymic lymphomas. The mouse genomic DNAs and methylated and unmethylated control DNAs were chemically cleaved, then the positions of 5-methylcytosines were mapped by ligation-mediated PCR which can be used to distinguish methylated from unmethylated cytosines. Our data show that most CpG dinucleotides in the DNA binding domain of the lacI gene are methylated to a high extent (>98%) in all tissues tested; only a few sites are partially (70–90%) methylated. We conclude that tissue-specific methylation is unlikely to contribute significantly to tissue-specific mutational patterns, and that the occurrence of common mutation sites at specific CpGs in the lacI gene is not related to selective methylation of only these sequences. The data confirm previous suggestions that the high frequency of CpG mutations in lacI transgenes is related to the presence of 5-methylcytosine bases.     

A comprehensive survey of human polymorphisms at conserved splice dinucleotides and its evolutionary relationship with alternative splicing

Background  

Alternative splicing (AS) is a key molecular process that endows biological functions with diversity and complexity. Generally, functional redundancy leads to the generation of new functions through relaxation of selective pressure in evolution, as exemplified by duplicated genes. It is also known that alternatively spliced exons (ASEs) are subject to relaxed selective pressure. Within consensus sequences at the splice junctions, the most conserved sites are dinucleotides at both ends of introns (splice dinucleotides). However, a small number of single nucleotide polymorphisms (SNPs) occur at splice dinucleotides. An intriguing question relating to the evolution of AS diversity is whether mutations at splice dinucleotides are maintained as polymorphisms and produce diversity in splice patterns within the human population. We therefore surveyed validated SNPs in the database dbSNP located at splice dinucleotides of all human genes that are defined by the H-Invitational Database.     

Stable isotope labeling-mass spectrometry analysis of methyl- and pyridyloxobutyl-guanine adducts of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in p53-derived DNA sequences

The p53 tumor suppressor gene is a primary target in smoking-induced lung cancer. Interestingly, p53 mutations observed in lung tumors of smokers are concentrated at guanine bases within endogenously methylated (Me)CG dinucleotides, e.g., codons 157, 158, 245, 248, and 273 ((Me)C = 5-methylcytosine). One possible mechanism for the increased mutagenesis at these sites involves targeted binding of metabolically activated tobacco carcinogens to (Me)CG sequences. In the present work, a stable isotope labeling HPLC-ESI(+)-MS/MS approach was employed to analyze the formation of guanine lesions induced by the tobacco-specific lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) within DNA duplexes representing p53 mutational "hot spots" and surrounding sequences. Synthetic DNA duplexes containing p53 codons 153-159, 243-250, and 269-275 were prepared, where (Me)C was incorporated at all physiologically methylated CG sites. In each duplex, one of the guanine bases was replaced with [1,7,NH(2)-(15)N(3)-2-(13)C]-guanine, which served as an isotope "tag" to enable specific quantification of guanine lesions originating from that position. After incubation with NNK diazohydroxides, HPLC-ESI(+)-MS/MS analysis was used to determine the yields of NNK adducts at the isotopically labeled guanine and at unlabeled guanine bases elsewhere in the sequence. We found that N7-methyl-2'-deoxyguanosine and N7-[4-oxo-4-(3-pyridyl)but-1-yl]guanine lesions were overproduced at the 3'-guanine bases within polypurine runs, while the formation of O(6)-methyl-2'-deoxyguanosine and O(6)-[4-oxo-4-(3-pyridyl)but-1-yl]-2'-deoxyguanosine adducts was specifically preferred at the 3'-guanine base of 5'-GG and 5'-GGG sequences. In contrast, the presence of 5'-neighboring (Me)C inhibited O(6)-guanine adduct formation. These results indicate that the N7- and O(6)-guanine adducts of NNK are not overproduced at the endogenously methylated CG dinucleotides within the p53 tumor suppressor gene, suggesting that factors other than NNK adduct formation are responsible for mutagenesis at these sites.     

CG sites and their nucleotide environment in the human gamma-delta-beta-globin gene cluster: distribution, frequency and possible frame for differential methylation

Available restriction endonucleases including CG dinucleotides in their target sequences (most of them being unable to cut the DNA when the cytosine of the CG sequence is methylated) have been used to map cloned DNA covering the human gamma-delta-beta globin gene cluster. Since the human DNA fragments were cloned in Escherichia coli, only the internal cytosine in the sequence CCAT GG could be methylated. Thus, any recognized "CG enzyme" site can be detected since they are unmethylated. Results show that frequencies of "CG enzyme" sites regularly decrease from the gamma-globin region to the beta-globin region, the latter being very poor in "CG enzyme"' sites. The array of enzymes used here detects 4 times more CG sites than the classical MspI/HpaII system. Examination of previously sequenced parts of the gamma-delta-beta globin gene cluster shows that CG dinucleotides correspond to an average frequency of 1 out of 104 nucleotides in the gamma-globin region and 1 out of 138 nucleotides in the beta-globin region. In the gamma-globin region, 1 CG out of 4 or 5 may be detected by the enzymes used; the detected frequency is less than 1 out of 10 CG in the beta-region. Analysis of nucleotide environment around CG dinucleotides shows occurrence of local differences, the main sequences being CGG in the 5' side flanking the gamma genes and ACG in the corresponding area of the beta gene. The results presented introduce some new considerations about analysis of cytosine methylation which has been previously proposed as playing a role in the control of the activity of gamma, delta and beta genes respectively.     

Methylation and repeats in silent and nonsense mutations of p53

All exonic CG sequences in p53 are methylated; this epigenetic modification is correlated with frequent G:C-->A:T transitions in p53. Recent reports reveal the presence in p53 of non-CG methylation in CC and CCC sequences, complementary to sites of selective guanosine adduct formation (GG and GGG), and the association of genetic instability with methylation at repetitive sequences. We presently investigated the distribution of methylation sites and repetitive elements in silent and nonsense p53 mutations (2051) among the IARC's TP53 somatic mutation database for exons 5-8. Silent mutations are nonrandom, but mostly involve G:C-->A:T transitions (62%); in particular C-->T mutations (39% of all silent mutations) are mostly correlated with CC and CCC sequences, while G-->A mutations with GG sequences. Sequence analysis of all non-G:C-->A:T silent mutations reveals the frequent formation of new methylation sites (CG), new CCC and GGG sequences in the resulting sequence, refinement of symmetry elements at interrupted microsatellite-like sequences and formation of small repeats (55.3%). The G:C-->A:T silent mutations characterize cancers associated with cigarette smoking (e.g. bladder or lung and bronchus cancer versus colorectal cancer); on the contrary, non-G:C-->A:T silent mutations have similar frequencies in most cancers. Nonsense mutations in exons 5-8, all resulting in mutants lacking amino acids 307-393, which are crucial for p53 activity, were also analyzed. The frequency of nonsense mutations is higher at methylated sites or repeats 1-2 nucleotides removed from methylation sites. Frameshift mutations are also more frequent at repeated sequences. The frequent G:C-->A:T silent mutations could indicate that CC and CCC sequences of exons 5-8 are occasionally targets of non-CpG methylation of cytosine. This process of de novo methylation in the presence of microsatellite-like sequences and small repeats might influence the genetic stability of a variety of genes.     

Developmental differences in methylation of human Alu repeats.

Alu repeats are especially rich in CpG dinucleotides, the principal target sites for DNA methylation in eukaryotes. The methylation state of Alus in different human tissues is investigated by simple, direct genomic blot analysis exploiting recent theoretical and practical advances concerning Alu sequence evolution. Whereas Alus are almost completely methylated in somatic tissues such as spleen, they are hypomethylated in the male germ line and tissues which depend on the differential expression of the paternal genome complement for development. In particular, we have identified a subset enriched in young Alus whose CpGs appear to be almost completely unmethylated in sperm DNA. The existence of this subset potentially explains the conservation of CpG dinucleotides in active Alu source genes. These profound, sequence-specific developmental changes in the methylation state of Alu repeats suggest a function for Alu sequences at the DNA level, such as a role in genomic imprinting.     

Statistical analysis of DNA sequences nearby splicing sites

Recognition of coding regions within eukaryotic genomes is one of oldest but yet not solved problems of bioinformatics. New high-accuracy methods of splicing sites recognition are needed to solve this problem. A question of current interest is to identify specific features of nucleotide sequences nearby splicing sites and recognize sites in sequence context. We performed a statistical analysis of human genes fragment database and revealed some characteristics of nucleotide sequences in splicing sites neighborhood. Frequencies of all nucleotides and dinucleotides in splicing sites environment were computed and nucleotides and dinucleotides with extremely high\low occurrences were identified. Statistical information obtained in this work can be used in further development of the methods of splicing sites annotation and exon-intron structure recognition.     

The loss of dinucleotides CpG from DNA. IV. Methylation and divergence of genes and pseudogenes of small nuclear RNA

The frequency of neighboring base pairs in nucleotide sequences of over 80 genes and pseudogenes of low molecular weight RNAs U1-U8, 4.5S and 7S in different eukaryotes was determined. The probable frequency of CpG----TpG + CpA substitutions, caused as a result of the deamination of 5-methylcytosine residues in DNA, was determined. It was found that the genes of small RNAs do not reveal a single level of CpG methylation for all the species studied. In most cases CpG in the genes of U1, 4.5S and 7S RNA are methylated, whereas in the genes of U2-U6-RNA these sites must have never been subjected to methylation. Nearly all the investigated pseudogenes of different small RNAs are strongly methylated due to a considerable lack of CpG. It was established that CpG----TpG + CpA transitions may amount to as much third of all the mutations accumulated in the genes of the same RNAs in different species. Such transitions in pseudogenes may account for 40% of all the nucleotide substitutions. This disproportionately high level of mutations in CpG dinucleotides (3-5-fold higher than in other DNA dupletes) must be the direct result of the methylation of these sites. Consequently, CpG methylation causes a dramatic acceleration of the divergence rate of DNA sequences. It has been concluded that protection of most vital genes against methylation is one of the essential conditions for sustaining the high level of stability of the macromolecular structure and for the reliability of macromolecular functioning in a cell.     

Mutational spectra of human cancer

The purpose of this review is to summarize the evidence that can be used to reconstruct the etiology of human cancers from mutations found in tumors. Mutational spectra of the tumor suppressor gene p53 (TP53) are tumor specific. In several cases, these mutational spectra can be linked to exogenous carcinogens, most notably for sunlight-associated skin cancers, tobacco-associated lung cancers, and aristolochic acid-related urothelial tumors. In the TP53 gene, methylated CpG dinucleotides are sequences selectively targeted by endogenous and exogenous mutagenic processes. Recent high-throughput sequencing efforts analyzing a large number of genes in cancer genomes have so far, for the most part, produced mutational spectra similar to those in TP53 but have unveiled a previously unrecognized common G to C transversion mutation signature at GpA dinucleotides in breast cancers and several other cancers. Unraveling the origin of these G to C mutations will be of importance for understanding cancer etiology.