Theoretical analysis of mutation hotspots and their DNA sequence context specificity

Mutation frequencies vary significantly along nucleotide sequences such that mutations often concentrate at certain positions called hotspots. Mutation hotspots in DNA reflect intrinsic properties of the mutation process, such as sequence specificity, that manifests itself at the level of interaction between mutagens, DNA, and the action of the repair and replication machineries. The hotspots might also reflect structural and functional features of the respective DNA sequences. When mutations in a gene are identified using a particular experimental system, resulting hotspots could reflect the properties of the gene product and the mutant selection scheme. Analysis of the nucleotide sequence context of hotspots can provide information on the molecular mechanisms of mutagenesis. However, the determinants of mutation frequency and specificity are complex, and there are many analytical methods for their study. Here we review computational approaches for analyzing mutation spectra (distribution of mutations along the target genes) that include many mutable (detectable) positions. The following methods are reviewed: derivation of a consensus sequence, application of regression approaches to correlate nucleotide sequence features with mutation frequency, mutation hotspot prediction, analysis of oligonucleotide composition of regions containing mutations, pairwise comparison of mutation spectra, analysis of multiple spectra, and analysis of "context-free" characteristics. The advantages and pitfalls of these methods are discussed and illustrated by examples from the literature. The most reliable analyses were obtained when several methods were combined and information from theoretical analysis and experimental observations was considered simultaneously. Simple, robust approaches should be used with small samples of mutations, whereas combinations of simple and complex approaches may be required for large samples. We discuss several well-documented studies where analysis of mutation spectra has substantially contributed to the current understanding of molecular mechanisms of mutagenesis. The nucleotide sequence context of mutational hotspots is a fingerprint of interactions between DNA and DNA repair, replication, and modification enzymes, and the analysis of hotspot context provides evidence of such interactions.     

Study of the DNA primary structure effect on induction of mutations by alkylating agents

Analysis and comparison of mutation spectra is one of the major tasks of molecular biology, since mutation spectra often reveal important properties of various mutagens and proteins involved in the repair/replication systems. Mutability is known to vary significantly along the nucleotide sequence. Mutations are abundant at certain positions (mutation hotspots). In this work, we applied regression analysis based on the basic logic patterns to understand the role of the nucleotide sequence context in mutation induction. The spectra of mutations induced by various alkylating agents were studied. The nucleotide bases at positions -2, -1, +1 and +2 were shown to have the most significant effect in G:C-->A:T replacements.     

The effect of the Primary Structure of DNA on Induction of Mutations by Alkylating Agents

Analysis and comparison of mutation spectra is one of the major tasks of molecular biology, since mutation spectra often reveal important properties of various mutagens and proteins involved in the repair/replication systems. Mutability is known to vary significantly along the nucleotide sequence. Mutations are abundant at certain positions (mutation hotspots). In this work, we applied regression analysis based on the basic logic patterns to understand the role of the nucleotide sequence context in mutation induction. The spectra of mutations induced by various alkylating agents were studied. The nucleotide bases at positions –2, –1, +1 and +2 were shown to have the most significant effect in G : C A : T replacements.     

Regression trees for analysis of mutational spectra in nucleotide sequences.

MOTIVATION: The study and comparison of mutational spectra is an important problem in molecular biology, because these spectra often reveal important features of the action of various mutagens and the functioning of repair/replication enzymes. As is known, mutability varies significantly along nucleotide sequences: mutations often concentrate at certain positions in a sequence, otherwise termed 'hotspots'. RESULTS: Herein, we propose a regression analysis method based on the use of regression trees in order to analyse the influence of nucleotide context on the occurrence of such hotspots. The REGRT program developed has been tested on simulated and real mutational spectra. For the G:C-->T:A mutational spectra induced by Sn1 alkylating agents (nine spectra), the prediction accuracy was 0. 99. AVAILABILITY: The REGRT program is available upon request from V.Berikov.     

Comparison of the differential context-dependence of DNA deamination by APOBEC enzymes: correlation with mutation spectra in vivo

To investigate the extent to which in vivo mutation spectra might reflect the intrinsic specificities of active mutators, genetic and biochemical assays were used to analyse the DNA target specificities of cytidine deaminases of the APOBEC family. The results reveal the critical importance of nucleotides immediately 5' of the targeted C for the specificity of all three enzymes studied (AID, APOBEC1 and APOBEC3G). At position -1, APOBEC1 showed a marked preference for dT, AID for dA/dG and APOBEC3G a strong preference for dC. Furthermore, AID and APOBEC3G showed distinct dependence on the nucleotide at position -2 with dA/dT being favoured by AID and dC by APOBEC3G. Most if not all activity of the recombinant deaminases on free dC could be attributed to low-level contamination by host enzymes. The target preference of APOBEC3G supports it being a major but possibly not sole contributor to HIV hypermutation without making it a dominant contribution to general HIV sequence variation. The specificity of AID as deduced from the genetic assay (which relies on inactivation of sacB of Bacillus subtilis) agrees well with that deduced by Pham et al. using an in vitro assay although we postulate that major intrinsic mutational hotspots in immunoglobulin V genes in vivo might reflect favoured sites of AID action being generated by proximal DNA targets located on opposite DNA strands. The target specificity of AID also accords with the spectrum of mutations observed in B lymphoma-associated oncogenes. The possibility of deaminase involvement in non-lymphoid human tumours is hinted at by tissue-specific differences in the spectra of dC transitions in tumour-suppressor genes. Thus, the patterns of hypermutation in antibodies and retroviruses owe much to the intrinsic sequence preferences of the AID/APOBEC family of DNA deaminases: analogous biases might also contribute to the spectra of cancer-associated mutation.     

iMARS--mutation analysis reporting software: an analysis of spontaneous cII mutation spectra

The sensitivity of any mutational assay is determined by the level at which spontaneous mutations occur in the corresponding untreated controls. Establishing the type and frequency at which mutations occur naturally within a test system is essential if one is to draw scientifically sound conclusions regarding chemically induced mutations. Currently, mutation-spectra analysis is laborious and time-consuming. Thus, we have developed iMARS, a comprehensive mutation-spectrum analysis package that utilises routinely used methodologies and visualisation tools. To demonstrate the use and capabilities of iMARS, we have analysed the distribution, types and sequence context of spontaneous base substitutions derived from the cII gene mutation assay in transgenic animals. Analysis of spontaneous mutation spectra revealed variation both within and between the transgenic rodent test systems Big Blue Mouse, MutaMouse and Big Blue Rat. The most common spontaneous base substitutions were G:C-->A:T transitions and G:C-->T:A transversions. All Big Blue Mouse spectra were significantly different from each other by distribution and nearly all by mutation type, whereas the converse was true for the other test systems. Twenty-eight mutation hotspots were observed across all spectra generally occurring in CG, GA/TC, GG and GC dinucleotides. A mutation hotspot at nucleotide 212 occurred at a higher frequency in MutaMouse and Big Blue Rat. In addition, CG dinucleotides were the most mutable in all spectra except two Big Blue Mouse spectra. Thus, spontaneous base-substitution spectra showed more variation in distribution, type and sequence context in Big Blue Mouse relative to spectra derived from MutaMouse and Big Blue Rat. The results of our analysis provide a baseline reference for mutation studies utilising the cII gene in transgenic rodent models. The potential differences in spontaneous base-substitution spectra should be considered when making comparisons between these test systems. The ease at which iMARS has allowed us to carry out an exhaustive investigation to assess mutation distribution, mutation type, strand bias, target sequences and motifs, as well as predict mutation hotspots provides us with a valuable tool in helping to distinguish true chemically induced hotspots from background mutations and gives a true reflection of mutation frequency.     

Base damage,local sequence context and TP53 mutation hotspots: a molecular dynamics study of benzo[a]pyrene induced DNA distortion and mutability

The mutational pattern for the TP53 tumour suppressor gene in lung tumours differs to other cancer types by having a higher frequency of G:C>T:A transversions. The aetiology of this differing mutation pattern is still unknown. Benzo[a]pyrene,diol epoxide (BPDE) is a potent cigarette smoke carcinogen that forms guanine adducts at TP53 CpG mutation hotspot sites including codons 157, 158, 245, 248 and 273. We performed molecular modelling of BPDE-adducted TP53 duplex sequences to determine the degree of local distortion caused by adducts which could influence the ability of nucleotide excision repair. We show that BPDE adducted codon 157 has greater structural distortion than other TP53 G:C>T:A hotspot sites and that sequence context more distal to adjacent bases must influence local distortion. Using TP53 trinucleotide mutation signatures for lung cancer in smokers and non-smokers we further show that codons 157 and 273 have the highest mutation probability in smokers. Combining this information with adduct structural data we predict that G:C>T:A mutations at codon 157 in lung tumours of smokers are predominantly caused by BPDE. Our results provide insight into how different DNA sequence contexts show variability in DNA distortion at mutagen adduct sites that could compromise DNA repair at well characterized cancer related mutation hotspots.     

Comparison of the Ames test and a newly developed assay for detection of mutagenic pollution of marine environments

Mismatch repair (MMR) genes, such as Msh2, are classified as "mutator" genes, responsible for the microsatellite instability identified in many tumors. In the current study, the mutation frequency and mutational spectrum in thymic lymphoma arising in Msh2 deficient mice are investigated. Thymic lymphoma developed in Msh2-/- background displayed an eight to nine-fold increase in mutation frequency compared to the normal thymi in Msh2 deficient animals. Sequencing demonstrated significantly different mutational spectra between normal thymus tissue and thymic lymphomas in Msh2-/- mice (P=0.02). The tumor mutational spectrum is characterized by an increase in base substitutions occurring at A:T sites, and multiple mutations, as well as a minor increase in -1 frameshifts. We analyzed mutations in different parts of the tumors, and different regional hotspots could be identified. Several hotspot mutations that are a rare event in normal tissues were identified in the tumor tissues. We conclude that thymic lymphomas arising in Msh2 deficient genetic background are hypermutable and the altered mutational spectrum might be an indication of infidelity of DNA replication during tumorigenesis.     

IS10 transposase mutations that specifically alter target site recognition.

IS10 inserts preferentially into particular hotspots. We describe here mutations of IS10 transposase, called 'ATS' that confer Altered Target Specificity. These mutations yield a general relaxation in target specificity but do not affect other aspects of transposition. Thus, the preference for specific nucleotide sequences at the target site can be cleanly separated from other steps of the transposition reaction. Eleven ATS mutations identified in a genetic screen occur at only two codons in transposase, one in each of two regions of the protein previously implicated in target site interactions (Patch I and Patch II). Genetic analysis suggests that mutations at the two ATS codons affect the same specific function of transposase, thus raising the possibility that Patch I and Patch II interact. For wild-type IS10, insertion specificity is determined in part by a specific 6 bp consensus sequence and in part by the immediately adjacent sequence context of the target DNA. The ATS mutations do not qualitatively alter the hierarchy with which base pairs are recognized in the consensus sequence; instead, sites selected by ATS transposase exhibit a reduction in the degree to which certain base pairs are preferred over others. Models for the basis of this phenotype are discussed.     

Influence of DNA repair on mutation spectra in Salmonella

This paper reviews the influence of DNA repair on spontaneous and mutagen-induced mutation spectra at the base-substitution (hisG46) and -1 frameshift (hisD3052) alleles present in strains of the Salmonella (Ames) mutagenicity assay. At the frameshift allele (mostly a CGCGCGCG target), ΔuvrB influences the frequency of spontaneous hotspot mutations (−CG), duplications, and deletions, and it also shifts the sites of deletions and duplications. Cells with pKM101+ΔuvrB spontaneously produce complex frameshifts (frameshifts with an adjacent base substitution). The spontaneous frequency of 1-base insertions or concerted (templated) mutations is unaffected by DNA repair, and neither mutation is inducible by mutagens. Glu-P-1, 1-nitropyrene (1NP), and 2-acetylaminofluorene (2AAF) induce only hotspot mutations and are unaffected by pKM101, whereas benzo(a)pyrene and 4-aminobiphenyl induce only hotspot in pKM101⁻, and hotspot plus complex in pKM101⁺. At the base-substitution allele (mostly a CC/GG target), the ΔuvrB allele increases spontaneous transitions in the absence of pKM101 and increases transversions in its presence. The frequency of suppressor mutations is decreased 4× by ΔuvrB, but increased 7.5× by pKM101. Both repair factors cause a shift in the proportion of mutations to the second position of the CC/GG target. With UV light and γ-rays, the ΔuvrB allele increases the proportion of transitions relative to transversions. pKM101 is required for mutagenesis by Glu-P-1 and 4-AB, and the types and positions of the substitutions are not altered by the addition of the ΔuvrB allele. Changes in DNA repair appear to cause more changes in spontaneous than in mutagen-induced mutation spectra at both alleles. There is a high correlation (r²=0.8) between a mutagen's ability to induce complex frameshifts and its relative base-substitution/frameshift mutagenic potency. A mutagen induces the same primary class of base substitution in TA100 (ΔuvrB, pKM101) as it does in Escherichia coli, mammalian cells, or rodents as well as in the p53 gene of human tumors associated with exposure to that mutagen. Thus, a mutagen induces the same primary class of base substitution in most organisms, reflecting the conserved nature of DNA replication and repair processes.     

Mutation frequency and specificity with age in liver, bladder and brain of lacI transgenic mice

Mutation frequency and specificity were determined as a function of age in nuclear DNA from liver, bladder, and brain of Big Blue lacI transgenic mice aged 1.5-25 months. Mutations accumulated with age in liver and accumulated more rapidly in bladder. In the brain a small initial increase in mutation frequency was observed in young animals; however, no further increase was observed in adult mice. To investigate the origin of mutations, the mutational spectra for each tissue and age were determined. DNA sequence analysis of mutant lacI transgenes revealed no significant changes in mutational specificity in any tissue at any age. The spectra of mutations found in aging animals were identical to those in younger animals, suggesting that they originated from a common set of DNA lesions manifested during DNA replication. The data also indicated that there were no significant age-related mutational changes due to oxidative damage, or errors resulting from either changes in the fidelity of DNA polymerase or the efficiency of DNA repair. Hence, no evidence was found to support hypotheses that predict that oxidative damage or accumulation of errors in nuclear DNA contributes significantly to the aging process, at least in these three somatic tissues.     

Escherichia coli mutator (Delta)polA is defective in base mismatch correction: the nature of in vivo DNA replication errors

We constructed a set of Escherichia coli strains containing deletions in genes encoding three SOS polymerases, and defective in MutS and DNA polymerase I (PolI) mismatch repair, and estimated the rate and specificity of spontaneous endogenous tonB(+)-->tonB- mutations. The rate and specificity of mutations in strains proficient or deficient in three SOS polymerases was compared and found that there was no contribution of SOS polymerases to the chromosomal tonB mutations. MutS-deficient strains displayed elevated spontaneous mutation rates, consisting of dominantly minus frameshifts and transitions. Minus frameshifts are dominated by warm spots at run-bases. Among 57 transitions (both G:C-->A:T and A:T-->G:C), 35 occurred at two hotspot sites. PolI-deficient strains possessed an increased rate of deletions and frameshifts, because of a deficiency in postreplicative deletion and frameshift mismatch corrections. Frameshifts in PolI-deficient strains occurred within the entire tonB gene at non-run and run sequences. MutS and PolI double deficiency indicated a synergistic increase in the rate of deletions, frameshifts and transitions. In this case, mutS-specific hotspots for frameshifts and transitions disappeared. The results suggested that, unlike the case previously known pertaining to postreplicative MutS mismatch repair for frameshifts and transitions and PolI mismatch repair for frameshifts and deletions, PolI can recognize and correct transition mismatches. Possible mechanisms for distinct MutS and PolI mismatch repair are discussed. A strain containing deficiencies in three SOS polymerases, MutS mismatch repair and PolI mismatch repair was also constructed. The spectrum of spontaneous mutations in this strain is considered to represent the spectrum of in vivo DNA polymerase III replication errors. The mutation rate of this strain was 219x10(-8), about a 100-fold increase relative to the wild-type strain. Uncorrected polymerase III replication errors were predominantly frameshifts and base substitutions followed by deletions.     

Detection of single DNA base mutations with mismatch repair enzymes.

A novel method for identifying DNA point mutations has been developed by using mismatch repair enzymes. The high specificity of the Escherichia coli MutY protein has permitted the development of a reliable and sensitive method for the detection and characterization of point mutations in the human genome. The MutY protein is involved in a repair pathway that can convert A/G or A/C mismatches to C/G or G/C basepairs, respectively. A/G or A/C mismatches formed by hybridization between two amplified genomic DNA samples or between specific DNA probes and target DNA are nicked at the mispaired adenine strand by MutY protein. As little as 1% of the mutant sequence can be detected by the mismatch repair enzyme cleavage (MREC) method in a mixture of normal and mutated DNAs (e.g., mutant cells are only present in 1% of the normal cell background). By using different probes, the assay also can determine the nucleotide sequence of the mutation. We have applied this method to detect single-base substitutions in human oncogenes.     

Replication-dependent and selection-induced mutations in respiration-competent and respiration-deficient strains of Saccharomyces cerevisiae

Adaptive or selection-induced mutations are defined as mutations that occur in non-dividing cells as a response to prolonged non-lethal selective pressure such as starvation for an essential amino acid. In the absence of DNA replication, the processing of endogenous DNA lesions by repair enzymes probably acts as a source of mutations. We are studying selection-induced reversions of frameshift alleles in the eukaryote Saccharomyces cerevisiae. Here we show that respiration-deficient strains, totally devoid of mitochondrial DNA, yield selection-induced mutants at slightly elevated frequencies compared to isonucleic respiration-competent strains. Therefore factors of mitochondrial origin such as reactive oxygen species or hypothetical recombinogenic DNA fragments are unlikely to be mediators of selection-induced nuclear frameshift mutation in yeast. Furthermore we compared sequence spectra of reversions of the +1 hom3-10 frameshift allele and found a strong preference for ?1 deletions in mononucleotide repeats in selection-induced and replication-dependent revertants, indicating slippage errors during DNA repair synthesis as well as during DNA replication. Remarkably, a higher degree of variation in the site of the reverting frameshift and accompanying base substitutions was found among selection-induced revertants.     

DNA damage and mutations produced by chloroacetaldehyde in a CpG-methylated target gene

Chloroacetaldehyde (CAA) is a metabolite of the human carcinogen vinyl chloride. CAA produces several types of DNA adducts including the exocyclic base adducts 3,N(4)-ethenocytosine, 1,N(6)-ethenoadenine, N(2),3-ethenoguanine, and 1,N(2)-ethenoguanine. Adducts of CAA with 5-methylcytosine have not yet been characterized. Here we have analyzed the mutational spectra produced by CAA in the supF gene of the pSP189 shuttle vector when present in either an unmethylated or CpG-methylated state. The vectors were replicated in human nucleotide excision repair-deficient XP-A fibroblasts. The mutational spectra obtained with the unmethylated and methylated supF target genes were generally similar with a preponderance of C/G to T/A transitions and C/G to A/T transversions. CAA-induced DNA adducts were mapped along the supF gene by using thermostable thymine DNA glycosylase (TDG) in conjunction with ligation-mediated PCR or by a Taq polymerase stop assay. Prominent CAA-induced TDG-sensitive sites were seen at several CpG positions but were independent of methylation. Methylated CpG sites were sites of CAA-induced mutations but were not the major mutational hotspots. Taq polymerase arrest sites were observed at numerous sequence positions in the supF gene and reflected the rather broad distributions of mutations along the sequence. We conclude that methylated CpG sites are not preferential targets for chloroacetaldehyde-induced mutagenesis.     

The role of weak specific and nonspecific interactions in recognition and conversation by enzymes of long DNA

According to a currently accepted model, enzymes engage in high-rate sliding along DNA when searching for specific recognition sequences or structural elements (modified nucleotides, breaks, single-stranded DNA fragments, etc.). Such sliding requires these enzymes to possess sufficiently high affinity for DNA of any sequence. Thus, significant differences in the enzymes' affinity for specific and nonspecific DNA sequences cannot be expected, and formation of a complex between an enzyme and its target DNA unlikely contributes significantly in the enzyme specificity. To elucidate the factors providing the specificity we have analyzed many DNA replication, DNA repair, topoisomerization, integration, and recombination enzymes using a number of physicochemical methods, including a method of stepwise increase in ligand complexity developed in our laboratory. It was shown that high affinity of all studied enzymes for long DNA is provided by formation of many weak contacts of the enzymes with all nucleotide units covered by protein globules. Contacts of positively charged amino acid residues with internucleotide phosphate groups contribute most to such interactions; the contribution of each contact is very small and the full contact interface usually resembles interactions between oppositely charged biopolymer surfaces. In some cases significant contribution to the affinity is made through hydrophobic and/or van der Waals interactions of the enzymes with nucleobases. Overall, depending on the enzyme, such nonspecific interactions provide 5-8 orders of the enzyme affinity for DNA. Specific interactions of enzymes with long DNA, in contrast to contacts of enzymes with small ligands, are usually weak and comparable in efficiency with weak nonspecific contacts. The sum of specific interactions most often provides approximately one and rarely two orders of the affinity. According to structural data, DNA binding to any of the investigated enzymes is followed by a stage of DNA conformation adjustment including partial or complete DNA melting, deformation of its backbone, stretching, compression, bending or kinking, eversion of nucleotides from the DNA helix, etc. The full set of such changes is characteristic for each individual enzyme. The fact that all enzyme-dependent changes in DNA are effected through weak specific rather than strong interactions is very important. Enzyme-specific changes in DNA conformation are required for effective adjustment of reacting orbitals with accuracy about 10-15 degrees, which is possible only for specific DNA. A transition from nonspecific to specific DNA leads to an increase in the reaction rate (kcat) by 4-8 orders of magnitude. Thus, the stages of DNA conformation adjustment and catalysis proper provide the high specificity of enzyme action.     

Replication-dependent and selection-induced mutations in respiration-competent and respiration-deficient strains of Saccharomyces cerevisiae

Adaptive or selection-induced mutations are defined as mutations that occur in non-dividing cells as a response to prolonged non-lethal selective pressure such as starvation for an essential amino acid. In the absence of DNA replication, the processing of endogenous DNA lesions by repair enzymes probably acts as a source of mutations. We are studying selection-induced reversions of frameshift alleles in the eukaryote Saccharomyces cerevisiae. Here we show that respiration-deficient strains, totally devoid of mitochondrial DNA, yield selection-induced mutants at slightly elevated frequencies compared to isonucleic respiration-competent strains. Therefore factors of mitochondrial origin such as reactive oxygen species or hypothetical recombinogenic DNA fragments are unlikely to be mediators of selection-induced nuclear frameshift mutation in yeast. Furthermore we compared sequence spectra of reversions of the +1 hom3-10 frameshift allele and found a strong preference for −1 deletions in mononucleotide repeats in selection-induced and replication-dependent revertants, indicating slippage errors during DNA repair synthesis as well as during DNA replication. Remarkably, a higher degree of variation in the site of the reverting frameshift and accompanying base substitutions was found among selection-induced revertants. Received: 25 May 1998 / Accepted: 20 August 1998     

Effect of sequence context on O(6)-methylguanine repair and replication in vivo.

Understanding the origins of mutational hotspots is complicated by the intertwining of several variables. The selective formation, repair, and replication of a DNA lesion, such as O(6)-methylguanine (m(6)G), can, in principle, be influenced by the surrounding nucleotide environment. A nearest-neighbor analysis was used to address the contribution of sequence context on m(6)G repair by the Escherichia coli methyltransferases Ada or Ogt, and on DNA polymerase infidelity in vivo. Sixteen M13 viral genomes with m(6)G flanked by all permutations of G, A, T, and C were constructed and individually transformed into repair-deficient and repair-proficient isogenic cell strains. The 16 genomes were introduced in duplicate into 5 different cellular backgrounds for a total of 160 independent experiments, for which mutations were scored using a recently developed assay. The Ada methyltransferase demonstrated strong 5' and 3' sequence-specific repair of m(6)G in vivo. The Ada 5' preference decreased in the general order: GXN > CXN > TXN > AXN (X = m(6)G, N = any base), while the Ada 3' preference decreased in the order: NX(T/C) > NX(G/A), with mutation frequencies (MFs) ranging from 35% to 90%. The Ogt methyltransferase provided MFs ranging from 10% to 25%. As was demonstrated by Ada, the Ogt methyltransferase repaired m(6)G poorly in an AXN context. When both methyltransferases were removed, the MF was nearly 100% for all sequence contexts, consistent with the view that the replicative DNA polymerase places T opposite m(6)G during replication irrespective of the local sequence environment.