HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sakai, A. Articles by Maki, H. Search for Related Content PubMed PubMed Citation Articles by Sakai, A. Articles by Maki, H.

Department of Molecular Biology, Graduate School of Biological Sciences, Nara Institute of Science and Technology, 8916-5 Takayama-cho, Ikoma, Nara 630-0101, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
Reactive oxygen species (ROS) are potent oxidants that attack chromosomal DNA and free nucleotides, leading to oxidative DNA damage that causes genetic alterations. To avoid the ROS-mediated mutagenesis, cells have elaborate mechanisms including powerful antioxidant components and repair pathways that eliminate oxidative DNA damage. Because of the effective anti-mutagenic functions, it has been unclear to what extent the ROS contribute to spontaneous mutagenesis. Here we show that a significant portion of spontaneous mutations is actually caused by the ROS in aerobically growing Escherichia coli cells. Using the rpsL gene as a mutational target sequence, we established an experimental procedure to analyze spontaneous mutations occurring under a strictly anaerobic condition. Strong mutator phenotypes of cells defective in both mutM and mutY genes or ones lacking mutT gene were completely suppressed under the anaerobic condition, indicative of an absence of hydroxyl radicals in the cells. From a series of analyses with wild-type E. coli cells grown under different redox conditions, it appeared that 89% of base substitutions were caused by the ROS, especially hydroxyl radicals, in cells growing in the atmosphere. The ROS-mediated spontaneous mutations included highly site-specific base substitutions, two types of randomly occurring transversions, G:CC:G and A:TT:A, and –1 frameshifts at non-iterated base sequences.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Aerobically growing cells are exposed to endogenous reactive oxygen species (ROS), including singlet oxygen, superoxide, hydrogen peroxide and hydroxyl radical. The ROS can damage intracellular components such as lipids, proteins and DNA (Imlay 2003). Recently, it was suggested that the ROS are the most probable cause of aging-related diseases, such as Parkinson's disease and cancer (Sakumi et al. 2003; Fukae et al. 2005), in which spontaneous mutations induced by ROS might be accumulated during the aging process. Several lines of in vivo and in vitro evidence demonstrate that oxidative damage of DNA and free nucleotides induce various kinds of mutations (Bjelland & Seeberg 2003). Among the oxidative damage, 8-oxo-guanine (8-oxo-7,8-dihydroguanine) nucleotides in DNA and deoxynucleoside triphosphate can pair almost equally with adenine and cytosine nucleotides and, thus, highly efficiently induce specific types of base-substitution mutations, G:CT:A by 8-oxodG in DNA and A:TC:G by 8-oxodGTP (Maki & Sekiguchi 1992; Michaels et al. 1992; Hsu et al. 2004). Most other oxidative base damage results in DNA-replication block, which lead to base substitutions and single-base frameshifts involving translesion DNA synthesis. Despite the highly mutagenic nature of the oxidative damage, it does not necessarily result in spontaneous mutations. This is because most organisms possess elaborate mechanisms detoxifying the ROS, removing the oxidative DNA damages and correcting the resulting mismatches.

Hydroxyl radical (·OH) is thought to play a major role in damaging DNA and nucleotides in cells because of its high reactivity. The cellular level of ·OH is affected by intracellular concentrations of H₂O₂ and Fe⁺⁺ ion, both of which promote the Harber-Weiss/Fenton reaction (Nunoshiba et al. 1999). Imlay and colleagues recently found that AhpCF protein plays a major role in elimination of H₂O₂ and maintains a very low H₂O₂ concentration, about 20 nM or lower in E. coli cells growing in the atmosphere (Seaver & Imlay 2001; Park et al. 2005). The concentration of intracellular iron is also maintained at a low level by a system involving Fur protein, a negative regulator of iron uptake (Ferric ion uptake regulator) in E. coli cells (Hantke 2001). Therefore, intracellular concentration of ^·OH is probably very low. However, oxidative DNA damage is present at a detectable level in the E. coli cells. This notion is based on in vivo and in vitro studies of three mutator genes of E. coli, which suppress the spontaneous mutations caused by 8-oxo-guanine nucleotide. MutM protein removes 8-oxodG from oxidatively damaged DNA, and MutY protein excises adenine from 8-oxoG:A mispairs in DNA. Cells lacking both MutM and MutY proteins show a strong mutator phenotype specific for G:CT:A base substitution, implying that the cells produce 8-oxoG at a level corresponding to the high frequency of G:CT:A mutation if not repaired (Tchou et al. 1991; Michaels et al. 1992; Tajiri et al. 1995). Similarly, cells defective in MutT protein that eliminates 8-oxodGTP from the nucleotide pool show an extremely high frequency of A:TC:G base substitution (Maki & Sekiguchi 1992; Fowler & Schaaper 1997). Although 8-oxo-guanine is the most efficiently produced oxidative nucleotide-damage, more than 20 different kinds of nucleotide damages are generated by ·OH radical (Bjelland & Seeberg 2003). Therefore, it seems very likely that the intracellular ·OH concentration is maintained at a physiologically low level, but still high enough to produce various kinds of oxidative nucleotide damage in DNA as well as the nucleotide pool at readily detectable levels. On the other hand, since G:CT:A and A:TC:G are less frequent types of spontaneous base-substitution in wild-type E. coli cells, it seems probable that the cellular defense mechanisms almost completely suppress the mutations caused by the oxidative nucleotide damage. From the circumstances described above, it has been unclear to what extent the ROS contribute to the generation of spontaneous mutations.

In the present study, we analyzed spontaneous mutations occurring in E. coli cells growing under a strictly anaerobic condition. Comparing anaerobic data with data obtained under an aerobic condition, we found that a significant portion of base substitutions and single-base frameshifts occurring in the aerobically growing cells depended on the oxygen. In particular, three hot-spot type of base substitutions on the rpsL target sequence used for the mutation assay appeared to be exclusively oxygen-dependent mutations. Two types of less frequently and randomly occurring base substitutions, G:CC:G and A:TT:A, as well as –1 frameshift mutations at non-iterated base sequences on the rpsL target gene also depended on the oxygen. From patterns of mutations occurring in wild-type cells treated with H₂O₂ and those found in aerobically growing fur mutant cells lacking fur (ferric ion uptake regulator) gene, we concluded that a very frequent hot-spot type of base substitution and non-hot-spot type of G:CC:G mutations were caused by intracellular OH radical. Non-hot-spot type of A:TT:A mutations and –1 frameshifts at non-iterated base sequences were also likely caused by the ·OH radical, whereas two other hot-spot type of base substitutions might be induced by ROS other than ·OH. Taken together, these data indicate that 89% of the base substitutions spontaneously occurring within the rpsL sequence were caused by ROS, especially hydroxyl radical, in aerobically growing cells.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
Establishment of a strictly anaerobic condition for E. coli growth

To examine effects of the intracellular ROS on spontaneous mutagenesis, we compared the frequency and pattern of spontaneous mutations occurring in E. coli cells growing under an anaerobic condition to those determined with cells growing in the atmosphere.

The chromosome-based rpsL forward mutation assay that we recently developed was used for the analyses. An E. coli strain, MK811, carries two almost identical 0.6 kb DNA segments containing the rpsL gene at different loci 520 kb apart from each other on the chromosome. One rpsL gene is a streptomycin-resistant (Str^r) allele (AC at site 128) at the authentic locus (59 min in the E. coli linkage map), and the other is a trans-gene of the wild-type rpsL allele placed at 73 min and has a polymorphism at –22 (G for the trans-gene vs. A for the authentic gene). Because of the recessive nature of Str^r phenotype, MK811 is unable to grow on LB plates containing streptomycin. Genetic alterations leading to malfunction of the rpsL trans-gene make cells resistant to streptomycin. Thus, mutant clones with rpsL^– forward mutations can be easily selected by plating MK811 cells on LB plates containing streptomycin. Gene-conversion type of recombination events between the two rpsL genes were also detected by the chromosome-based rpsL system (our unpublished observations). To detect and isolate clones with spontaneous mutations occurring under an anaerobic condition, all operations for growth of E. coli cells and the following mutation assay were performed in an anaerobic chamber. In addition, cells were grown on LB plates deoxygenated by an anti-oxidant agent, Oxyrase, and placed in a gas-tight plastic bag with oxygen-trapping materials. After an appropriate incubation period, colonies formed on the plates were dissolved in a small volume of deoxygenated LB medium and the cells were subjected to the mutation assay with similarly deoxygenated plates and gas-tight plastic bag.

To provide evidence that our experimental procedures give strictly anaerobic circumstances, and that oxidative damage to DNA and free nucleotides is not produced in cells grown in these circumstances, we measured spontaneous mutation frequencies of MK6805, a derivative of MK811, in which both mutM and mutY mutator genes are hampered. When grown in the atmosphere, the mutM mutY11 double mutator mutant showed a sharply increased frequency of G:CT:A mutation, about 1000-fold higher than that determined with the wild-type strain. This strong and specific mutator effect was almost completely suppressed when the mutM mutY11 strain was grown under the anaerobic condition (Table 1). Similarly, 10 000-fold elevated frequency of A:TC:G mutation in aerobically growing cells of MK6806, a mutT mutator strain, was decreased to the wild-type level when grown under the anaerobic condition. From these data, we concluded that 8-oxo-dG in DNA and 8-oxodGTP were not produced at all in cells growing under the anaerobic condition developed in the present study. Therefore, it seemed very likely that the anaerobically growing cells would neither generate the hydroxyl radical as well as other ROS nor suffer from oxidative DNA damage. In addition, the data clearly demonstrated that in the aerobically growing cells, 8-oxo-guanine nucleotides are generated by intracellular ROS, and that the MutM, MutY, and MutT proteins are actually acting against the mutagenesis caused by the 8-oxo-guanine nucleotides. In wild-type cells, frequencies of G:CT:A and A:TC:G mutations were very low and unchanged whether the cells were grown under aerobic or anaerobic conditions, implying that the anti-mutagenic actions of MutM, MutY and MutT proteins are complete.

View this table: [in this window] [in a new window]   Table 1 Specific mutator effects of the mutM mutY11 and mutT strains were completely suppressed under the anaerobic growth condition. MK811 (wild-type), MK6805 (mutM mutY11) and MK6806 (mutT) were grown under aerobic (+O2) or anaerobic (–O2) conditions and were subjcted to the rpsL mutation assay as described in the Experimental procedures. Five different experiments were carried out for each strain aerobically or anaerobically grown, and about 96 independent samples were chosen from rpsL– clones obtained in each experiment and analyzed in their sequence alterations. For MK6805 and MK6806 cells anaerobically grown, 25 independent samples were taken from each experiment and analyzed. From the mutational analysis, frequencies of G:CT:A and A:TC:G mutations were calculated, and the averages of five experiments were indicated

MK811 cells were either aerobically or anaerobically grown on LB plates for periods that gave the same number of cell division cycles for both conditions (12 h and 72 h for aerobic and anaerobic growth conditions, respectively) and subjected to the chromosome-based rpsL mutation assay. Selection of streptomycin resistant clones from cells anaerobically grown was performed in the anaerobic condition, while the mutation assay for cells aerobically grown was done in the atmosphere. We confirmed that the same results were obtained when the selection of streptomycin resistant clones from the aerobically grown cells was carried out in the anaerobic condition. Since the mutation frequencies were very low and thus showed a fluctuating nature, averages of the frequencies were calculated from six and five independent experiments for aerobic and anaerobic growth conditions, respectively. About 96 mutants were randomly picked up from each experiment and their sequence changes were examined by DNA sequencing.

As shown in Fig. 1A, cells anaerobically grown showed a spontaneous mutation frequency about twofold higher than that determined with cells aerobically grown. This slight increase in the mutation frequency was due to a significantly elevated frequency of allelic recombination in cells grown under the anaerobic condition. The increment of the recombination frequency was about fivefold, and the allelic recombination accounted for 79% of total genetic alterations found in Str^r clones from the anaerobically grown wild-type cells. Large deletions, insertions of IS elements and duplications were also more frequent in the anaerobically grown cells, and the increments were 7.3-, 3.6- and 17-fold, respectively. On the other hand, when cells were grown in the absence of oxygen, the frequency of base substitutions dropped to 6.4 times lower than that determined with aerobically growing cells. Base substitutions were the most frequent type of Str^r genetic alterations in aerobically growing cells and accounted for 48% of the alterations. In anaerobically growing cells, however, only 3.1% of Str^r genetic alterations were base substitutions. No statistically significant change was observed in the occurrence of single-base frameshifts, the least frequent type of mutation in aerobically growing cells. Fourteen and 13 single-base frameshifts were found in 576 aerobic and 480 anaerobic rpsL^– mutants, respectively. Among these, –1 frameshift mutations at non-iterated base sequences, which we referred as non-run, were the most frequent type (5 of 14) of frameshifts in the aerobically growing cells. Nevertheless, no –1 frameshift at non-run was present in the 13 frameshift mutants recovered from the anaerobically growing cells. Therefore, the total frequencies of Str^r genetic alterations during aerobic and anaerobic growth were almost the same, but the contents of the genetic alterations were quite different. Compared to the spontaneous mutagenesis under the aerobic condition, allelic recombination events and other chromosomal rearrangements occurred more frequently, and point mutations, especially base substitutions, were much less frequent in the anaerobically growing E. coli cells.

View larger version (33K): [in this window] [in a new window]   Figure 1  Spontaneous rpsL– mutations in wild-type E. coli cells grown under the aerobic and anaerobic conditions. (A) Class distribution of genetic alterations leading to Strr phenotype. (B) Distribution of hot-spot and non-hot-spot types of base substitutions. MK811 cells were subjected to the rpsL mutation assay as described in the Experimental procedures. Six and five independent experiments were carried out for the aerobic and anaerobic mutation assays, respectively. About 96 mutants were randomly picked up from each experiment and their sequence changes were examined by DNA sequencing. Frequency of total or each class of mutation was calculated as an average of six or five independent experiments and is indicated as a pillar. Bars above the pillars indicate the standard deviations. Jackpot mutations were excluded from the calculation. Others include large deletions, duplications, and IS elements. +O2 and –O2 indicate the aerobic and the anaerobic growth conditions, respectively. The mini-graph super-imposed is the same figure but fivefold expanded in the Y-axis.

The most remarkable difference in spontaneous mutagenesis between aerobic and anaerobic conditions was the occurrence of base substitutions. Detailed analyses of the base substitutions recovered from aerobically and anaerobically growing cells revealed that site- and class-distributions of base substitutions were strikingly different between the mutations occurring under the different growth conditions (Figs 1B and 2).

In aerobically growing cells, base substitutions occurred in a site-distribution pattern consisting of two distinct classes of mutations; one class was strongly site-dependent and very frequent (hot-spot mutations), and the other was more randomly occurring and less frequent (non-hot-spot mutations). There were three strong hot-spot mutations occurring at two hot-spot sites in the aerobic spectrum of base substitutions (Fig. 2). G:CT:A mutations at site 82 (referred to as 82 CA) accounted for 5% of total base substitutions recovered from aerobically growing cells. The other two were A:TT:A and A:TC:G mutations at site 245 (referred to as 245 TA and 245 TG, respectively), accounting for 70% and 5% of the total aerobic base-substitutions, respectively. The remaining non-hot-spot type base substitutions corresponded to only 20% of the aerobic base-substitutions but were found at 25 different sites randomly located within the rpsL sequence. Among 15 base substitutions recovered from anaerobically growing cells, no 82 CA, 245 TA and 245 TG were found. Thus, all the anaerobic base-substitutions were non-hot-spot type mutations detected at 13 different sites. This clearly indicated that all the hot-spot type of aerobic base substitutions were exclusively oxygen-dependent mutations. Furthermore, the aerobic and the anaerobic base-substitutions were distinct in their class-distributions of the non-hot-spot mutations (Fig. 1B). Frequencies of two types of non-hot-spot transversions, G:CC:G and A:TT:A, were significantly lower in the anaerobically growing cells, while those of other classes of base substitutions were unchanged. Therefore, the effect of oxygen on the spontaneous base substitutions was not confined only to the generation of hot-spot mutations but also to the sequence-independent induction of G:CC:G and A:TT:A transversions, accounting for 5% and 4% of the total aerobic base-substitutions, respectively. From these data, it appeared that 89% (80% for the hot-spot mutations +5% for the random G:CC:G transversions +4% for the random A:TT:A transversions) of base substitutions occurring in the aerobically growing cells were dependent on the presence of oxygen during the cell growth.

View larger version (25K): [in this window] [in a new window]   Figure 2  Site distribution of rpsL– base substitutions occurring in wild-type cells grown under the aerobic and anaerobic conditions. Sequence changes found in 206 rpsL– mutants from cells grown aerobically and those found in 15 mutants from cells grown anaerobically are shown below and above the rpsL sequence, respectively. Nucleotide positions starting with the first position of the initiation codon are shown on the right side of the sequence. The promoter region (–35 and –10), the Shine-Dalgano sequence (SD), and the initiation and termination codons are underlined.

Oxygen-dependent mutations that we found in the spontaneous rpsL^– mutations seemed to be caused by oxidative DNA damage formed during the aerobic growth. This was supported by findings that mutations caused by 8-oxodG in DNA and 8-oxodGTP disappeared under the anaerobic condition, where the oxygen-dependent mutations also vanished. To obtain further evidence that the ROS is involved in the generation of oxygen-dependent mutations, we examined whether the frequencies of oxygen-dependent mutations could be increased when cells are treated with H₂O₂. Although the hydrogen peroxide is a relatively stable oxygen-radical, it readily penetrates into cells and produces the hydroxyl radical, much more chemically potent than H₂O₂, via the Fenton reaction with intracellular Fe⁺⁺ ion.

We treated cells with 1 mM H₂O₂ for 30 min and examined the rpsL^– mutations after several rounds of propagation of the cells. When this treatment was applied to the mutM mutY11 mutant strain, the frequency of G:CT:A mutations increased to double the frequency determined with untreated mutant cells (data not shown). This indicated that the H₂O₂ treatment was sufficient to increase the cellular level of ·OH radical and to produce efficiently 8-oxodG in the chromosome DNA of the cells. As shown in Fig. 3A, the frequency of rpsL^– mutations was elevated about 13-fold in the wild-type cells after H₂O₂ treatment. Base substitutions, single-base frameshifts, and allelic recombination events were increased 9.3-, 66- and 22-fold, respectively, upon the H₂O₂ treatment. These increasements of various genetic alterations were very likely due to various kinds of DNA lesion caused by OH radical. The H₂O₂-induced single-base frameshifts included all types of frameshifts (+1 and –1 frameshifts within run and non-run) with a significant bias to –1 frameshifts. Among the oxygen-dependent hot-spot mutations, only 245TA was affected by the H₂O₂ treatment (Fig. 3B). Although the elevation was only twofold, the same increment was obtained for the 245TA mutation in five independent experiments. Two types of non-hot-spot oxygen-dependent mutations, G:CC:G and A:TT:A, were well induced by the H₂O₂ treatment. It should be noticed that the extent of increment for the 245TA mutation (0.19 x 10^–6/site) was much greater than that for the non-hot-spot A:T T:A mutations (0.014 x 10^–6/site). These data suggested that the 245TA hot-spot mutation and G:CC:G and A:TT:A non-hot-spot mutations were caused by OH radical in aerobically growing cells. Two other oxygen-dependent hot-spot mutations, 82CA and 245TG, were probably caused by ROS other than ·OH radical.

View larger version (30K): [in this window] [in a new window]   Figure 3  rpsL– mutations induced upon the H2O2 treatment of wild-type cells. (A) Class distribution of genetic alterations leading to Strr phenotype. (B) Distribution of hot-spot and non-hot-spot types of base substitutions. MK811 cells were treated with 1 mM H2O2 and subjected to the rpsL mutation assay as described in the Experimental procedures. The survival index after the H2O2 treatment was 20%. Frequency of total or each class of mutation was calculated as an average of five independent experiments and is indicated as a pillar. Bars above the pillars indicate the standard deviations. Forty-eight independent samples from each experiment were examined for their sequence alterations. Jackpot mutations were excluded from the calculation. Others include large deletions, duplications, and IS elements. "none" indicate control experiments carried out in the same way but treated without H2O2.

Spontaneous mutations occurring in aerobically growing fur mutant cells

It has been known that the intracellular level of ·OH radical is increased in fur mutant cells (Nunoshiba et al. 1999). In addition to the H₂O₂ treatment, we took advantage of this mutant strain to further confirm involvement of the ROS in the oxygen-dependent mutagenesis. We found that the rpsL^– mutation frequency in a mutY11 strain was increased about tenfold when fur mutation was introduced into the strain, suggesting an elevated level of ·OH radical in the cells. When aerobically grown, the frequency of rpsL^– mutations in the fur mutant cells was 8.3 times as high as that in the wild-type cells (Fig. 4A). Similarly to H₂O₂ treatment, the absence of Fur protein in cells increased the frequency of base substitutions, single-base frameshifts and allelic recombination events. As shown in Fig. 4B, the frequency of 245TA hot-spot mutations was greatly increased, while the 82CA and 245TG hot-spot mutations were unaffected in the aerobically growing fur mutant cells. This was consistent with the observation when cells were treated with H₂O₂. All types of non-hot-spot mutations except A:TT:A were stimulated in the mutant cells. Interestingly, the fur mutation showed a mutator effect on A:TC:G transversions, which were not significantly induced by the H₂O₂ treatment. Thus, the mutator effect of fur mutation on the non-hot-spot base substitutions was somewhat different from the mutagenicity of the H₂O₂ treatment. It seems probable that some repair function might suppress the A:TT:A mutations during the growth of fur mutant cells, whereas the H₂O₂ treatment might produce oxidative DNA damage leading to the A:TT:A mutations in an amount exceeding the capacity of such a repair function. On the other hand, a temporary production of 8-oxodGTP upon the H₂O₂ treatment would not result in the elevation of A:TC:G frequency, but a constantly increased 8-oxodGTP level in the nucleotide pool would lead to the mutator effect in the fur mutant cells.

View larger version (29K): [in this window] [in a new window]   Figure 4  Spontaneous rpsL– mutations in fur mutant cells aerobically grown. (A) Class distribution of genetic alterations leading to Strr phenotype. (B) Distribution of hot-spot and non-hot-spot types of base substitutions. MK6168 (fur) and MK811 (wild-type) cells were grown on LB plates at 37 °C in the atmosphere and subjected to the rpsL mutation assay as described in the Experimental procedures. Frequency of total or each class of mutation was calculated as an average of six independent experiments and is indicated as a pillar. Bars above the pillars indicate the standard deviations. Forty-eight (for MK6168) and 240 (for MK811) independent samples from each experiment were examined for their sequence alterations. Jackpot mutations were excluded from the calculation. Others include large deletions, duplications, and IS elements.

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
As natural and potent DNA damaging substances, reactive oxygen species have been extensively studied, especially their mutagenic effects and processes (Imlay 2003). Using redox agents such as H₂O₂ and paraquat, biological and biochemical effects of the ROS at unusually increased cellular levels were analyzed (Carlioz & Touati 1986; Imlay & Linn 1986). Various mutant strains that are hyper-sensitive or resistant to the redox agents were helpful in solving multiple pathways counteracting the ROS and the resulting oxidative DNA damages. To this end, the chemistry and mutagenic nature of each kind of oxidative DNA damage have now become well understood, and numerous pathways of ROS-detoxification, repair and tolerance of the oxidative DNA damages, and sanitization of the oxidized nucleotide pool have been elucidated (Bjelland & Seeberg 2003). The findings that organisms possess such elaborate mechanisms to counteract the ROS clearly tell us the potential harmfulness of the intracellular ROS inevitable for aerobically respiring life. Since cells defective in any of the functions counteracting the ROS showed an elevated level of spontaneous mutation, it seemed conceivable that the ROS could be more or less a source of spontaneous mutagenesis. Using protocols for anaerobic cultivation of E. coli cells, we provided evidence for the first time that 8-oxo-guanine nucleotides are actually generated in DNA and the nucleotide pool when cells grow in the atmosphere. This supports the notion that aerobic organisms are threatened by the enormous amount of oxidative nucleotide damage continuously generated in the cell.

Nevertheless, for the following reasons, it has been difficult to estimate to what extent the ROS contribute to spontaneous mutagenesis. First of all, the anti-mutagenic mechanisms are so effective that the frequency of spontaneous mutation is several orders of magnitude lower than that obtained when oxidative damages are unrepaired. Secondly, there are other possible sources of spontaneous mutagenesis, which include mismatched base-pairs formed as errors during DNA replication and spontaneous DNA damage caused by cellular metabolites other than the ROS (Maki 2002). Some of them occur as frequently as the oxidative DNA damage, and almost all of them are corrected very efficiently by the anti-mutagenic functions specific for each source. This situation makes it hard to define the genuine source of spontaneous mutations. In the present study, we demonstrated for the first time that a significant portion of spontaneous mutations is caused by the ROS in aerobically growing E. coli cells. As far as base substitutions are concerned, 89% of those occurring within the rpsL target sequence are oxygen-dependent, and most of them are caused by the intracellular ·OH radical.

It is the hot-spot type of base substitutions that were exclusively dependent on the aerobic growth condition. Uneven site-distribution is a characteristic generally observed in spontaneous base-substitution mutagenesis, resulting in hot-spot and non-hot-spot types of mutations in the mutational spectrum on a given target sequence. Within the rpsL target sequence, three hot-spot type base-substitutions occurred at two sites and accounted for 80% of the spontaneous base-substitutions when cells were grown under the aerobic condition. Because all of the hot-spot mutations disappeared in the anaerobically growing cells, the impact of oxygen on the base-substitution mutagenesis appeared to be quite significant at least for the rpsL target sequence. However, hot-spot type of base substitutions would occur at different frequencies within different target sequences. Furthermore, hot-spot type of base substitutions might not be necessarily caused only by oxidative damage, although all the three hot-spot base substitutions within the rpsL sequence were oxygen-dependent. To generalize the impact of oxygen on the hot-spot mutagenesis, further studies using various target DNA sequences other than the rpsL sequence are needed. On the other hand, two types of non-hot-spot base substitutions, G:CC:G and A:TT:A, and –1 frameshifts at non-run were oxygen-dependent mutations that occur randomly within the rpsL target sequence. Therefore, it seems that in general these particular types of non-hot-spot mutations are caused by ROS, most likely the ·OH radical, in aerobically growing cells. It should be noted that other types of non-hot-spot base-substitutions and single-base frameshifts were not oxygen-dependent. While origins of these oxygen-independent mutations are yet undefined, most of them are probably derived from spontaneous DNA damages other than the oxidative ones (Maki 2002). Taken all together, the ROS-mediated spontaneous mutations are confined to types of point mutations, which consist of two categories; one is highly sequence-dependent specific base-substitutions, and the other is sequence-independent G:CC:G and A:TT:A base substitutions and –1 frameshifts at non-run.

Among the oxygen-dependent hot-spot mutations, 245TA is particularly interesting. This is the most frequent mutation that accounts for 70% of the aerobic base-substitutions. From analyses of mutations induced by the H₂O₂ treatment and those occurring in aerobically growing fur cells, we concluded that the 245TA is caused by intracellular OH radicals. Consistent with this notion, we observed that the frequency of 245TA was unchanged in a mutS strain (our unpublished observations). This clearly indicates that the mismatch repair is unable to suppress the 245TA mutagenesis, and that the hot-spot mutation is not derived from the replication error. The frequency of 245TA was also unchanged in a uvrA strain and a polB dinB umuCD strain (our unpublished observations). Therefore, oxidative DNA damage leading to the 245TA is not subjected to the nucleotide excision repair and none of the translesion DNA polymerases participate in suppressing or inducing the 245TA mutagenesis. While unidentified, the 245TA mutation might be derived from an infrequent type of oxidative DNA damage that could escape the repair mechanisms but would be fixed to the mutation by the replicative DNA polymerase. We further postulate that the rare kind of oxidative DNA damage could be generated in a manner highly dependent on the DNA structure surrounding the site 245 within the rpsL sequence. Most recently, we found that the 5'-GATC-3' sequence in which the 245TA mutation occurs at the second residue was essential for the mutagenesis (our unpublished observations), suggesting that the methyl-adenine produced by Dam methylase might be involved in generating the unknown oxidative DNA damage. This possibility is now under investigation in our laboratory. Other oxygen-dependent mutations might also be derived from infrequent and unrepairable oxidative DNA damages. 8-oxodG and other frequent type of oxidative base damages are apparently well repaired in E. coli cells, and mutations that should be specifically induced by these oxidative damages were not found to be oxygen-dependent in the present study. Probably, organisms have evolved to eliminate all the frequent types of oxidative damage in order to efficiently reduce the resulting mutations, whereas they would have been evolutionally unable to develop repair mechanisms for the rare kinds of oxidative damage. For example, guanidinohydantoin and spiroiminodihydantoin, two major products of 8-oxoG oxidation, were recently shown to be potent mutagenic DNA lesions that induce G:CC:G mutations (Henderson et al. 2003). It is of interest but difficult to identify such an infrequent type of oxidative DNA damage in cells aerobically growing.

In the present study, we have unexpectedly observed that some kinds of genetic alterations were significantly induced under the anaerobic growth condition. In particular, frequency of chromosomal gross rearrangements including allelic recombination events, large deletions, and duplications, was apparently higher in anaerobically growing cells than aerobically growing ones. A possible explanation for this phenomenon could be a longer incubation time for anaerobic growth of the cells. To grow cells to the same size (diameter) of colony under both growth conditions took 7.5 times longer for the anaerobically growing compared to aerobically growing cells. If the allelic recombination event and other rearrangement types of genetic alterations were dependent on the total incubation time, the frequencies of these genetic alterations in the anaerobic culture would be about 7.5 times higher than those in the aerobic culture. Another possibility could be anaerobic-dependent induction of spontaneous DNA damage other than the oxidative damages. For example, alkylation and deamination of DNA by intracellular nitrosating agents were shown to be stimulated when concentration of oxygen lowered at stationary phase of cell growth (Weiss 2001). From this point of view, the oxidative respiration results in induction of a part of spontaneous point-mutations on one hand, but suppression of chromosomal gross rearrangements on the other.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
Bacterial strains

The bacterial strains used in this study were all derivatives of E. coli MG1655. MK611 (mutM:: km^r mutY11 zgd::tet) was previously described (Tajiri et al. 1995). MA313 (mutT:: km^r) was a gift from Dr M. Akiyama (Akiyama et al. 1987). QC1732 (fur::km^r) was kindly provided by Dr T. Nunoshiba (Nunoshiba et al. 1999). The strain MK811, in which a 735 bp DNA fragment containing the wild-type rpsL gene was inserted at cysE locus on E. coli chromosome, was used as a tester strain for chromosome-based rpsL mutation assay. Strains MK6804 (mutY11), MK6805 (mutM mutY11), MK6168 (fur), MK6183 (mutY11 fur) and MK6806 (mutT) were constructed from MK811 by P1-mediated transduction with MK611, MA313 and QC1732 as donors.

Media

LB contained 1% (w/v) Bactotryptone (Difco), 0.5% (w/v) yeast extract (Difco), and 1% (w/v) NaCl. LB plates were solidified with 1.5% Bacto agar (Difco). K medium (1% glucose, 1% casamino acids, 1 µg/mL thiamine hydrochloride, 1 mM MgSO₄·7H₂O, 0.1 mM CaCl₂, M9 salts) were used for hydrogen peroxide treatment of E. coli cells (Imlay & Linn 1986). To ensure the anaerobic environment, we added a commercial enzyme additive, Oxyrase for broth (Oxyrase, Inc., Mansfield, OH, USA) to LB and Oxyrase for agar to LB plates in accordance with the supplier's manual. LB with Oxyrase for broth was prepared one day before use, and LB plates with Oxyrase for agar was prepared three days before use. These media were placed in a gas-tight plastic bag with oxygen-absorbing materials (Anaeropack with KENKI, Mitsubishi Gas Chemical Co., Tokyo, Japan) and kept in an anaerobic chamber (Sheldon Manufacturing Inc., Cornelius, OR, USA) until they were used. The aerobic chamber was filled with mixed gas (5% CO₂, 5% nitrogen, 90% hydrogen), and oxygen was eliminated by a catalyst that converts oxygen and hydrogen to water in the anaerobic chamber. During the anaerobic incubation, oxygen was always monitored using an oxygen-indicator sheet for Anaeropack and a mechanical oxygen-detector for the anaerobic chamber. To select streptomycin-resistance clones, we used LB plates containing 0.5 mg/mL or 1.0 mg/mL streptomycin for aerobic or anaerobic incubation, respectively.

rpsL mutation assay

On the same basis as for the genetic selection of mutation in the plasmid-based rpsL mutation assay (Mo et al. 1991; Fujii et al. 1999; Yoshiyama et al. 2001; Yoshiyama & Maki 2003), we have recently developed chromosome-based rpsL mutation assay (our unpublished observations). We used MK811 and its derivatives, which carry two rpsL genes, one with a base substitution leading to the Str^r phenotype and the other with the wild-type sequence. Cells were fully grown in LB at 37 °C in the atmosphere and approximately 100 cells were plated on a deoxygenated LB plate. The plate was placed in the Anaeropack with KENKI and incubated in the anaerobic chamber at 37 °C for three days until cells were grown up to 5 x 10⁶ cells/colony. Cells were collected in 5 mL of LB, and appropriate dilutions were spread on LB plates to determine the number of total cells and on LB plates containing streptomycin to determine rpsL^– cells. All the operations were carried out in the anaerobic chamber. Mutation frequency for each cell collective was calculated by dividing the number of mutant cells by the total number of cells. To measure the rpsL^– mutation frequency in aerobically growing cells, we performed almost the same assay as for the anaerobic mutations except that all the operations were done in the atmosphere. The rpsL coding region and its surrounding sequence (from position –130 to position 385) in each mutant cell were determined using an automated DNA sequencer (Megabase 1000, Amersham Pharmacia Biotech).

Hydrogen peroxide treatment of E. coli cells

Some modifications were added to the method of Imlay (Imlay & Linn 1986). Cells were grown in K medium at 37 °C with vigorous shaking and challenged with H₂O₂ at a density of 1–4 x 10⁷ cells/mL in 1 mL of K medium for 30 min at 37 °C. The challenge was terminated by the addition of 2 µg of catalase. These cells were spun down and resuspended in 5 mL of LB containing catalase. The wash was repeated twice. Then, cells were grown at 37 °C with shaking to OD₆₀₀ = 1.0 and subjected to the rpsL mutation assay.

	Acknowledgements

 
We acknowledge the financial support of Grants-in-Aid for Scientific Research on Priority Areas (12213082 to H. M.) from the Ministry of Education, Culture, Sports, Science and Technology of Japan. A. S. was supported by the 21st Century Center of Excellence Program (Graduate School of Biological Sciences, NAIST) from Japan Society for the Promotion of Science.

	Footnotes

 
Communicated by: Fumio Hanaoka

^* Correspondence: E-mail: maki{at}bs.naist.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Akiyama, M., Horiuchi, T. & Sekiguchi, M. (1987) Molecular cloning and nucleotide sequence of the mutT mutator of Escherichia coli that causes A:T to C:G transversion. Mol. Gen. Genet. 206, 9–16.[CrossRef][Medline]

Bjelland, S. & Seeberg, E. (2003) Mutagenicity, toxicity and repair of DNA base damage induced by oxidation. Mutat. Res. 531, 37–80.[Medline]

Carlioz, A. & Touati, D. (1986) Isolation of superoxide dismutase mutants in Escherichia coli: is superoxide dismutase necessary for aerobic life? EMBO J. 5, 623–630.[Medline]

Fowler, R.G. & Schaaper, R.M. (1997) The role of the mutT gene of Escherichia coli in maintaining replication fidelity. FEMS Microbiol. Rev. 21, 43–54.[CrossRef][Medline]

Fujii, S., Akiyama, M., Aoki, K., et al. (1999) DNA replication errors produced by the replicative apparatus of Escherichia coli. J. Mol. Biol. 289, 835–850.[CrossRef][Medline]

Fukae, J., Takanashi, M., Kubo, S., et al. (2005) Expression of 8-oxoguanine DNA glycosylase (OGG1) in Parkinson's disease and related neurodegenerative disorders. Acta Neuropathol. (Berl) 109, 256–262.[CrossRef][Medline]

Hantke, K. (2001) Iron and metal regulation in bacteria. Curr. Opin. Microbiol. 4, 172–177.[CrossRef][Medline]

Henderson, P.T., Delaney, J.C., Muller, J.G., et al. (2003) The hydantoin lesions formed from oxidation of 7,8-dihydro-8-oxoguanine are potent sources of replication errors in vivo. Biochemistry 42, 9257–9262.[CrossRef][Medline]

Hsu, G.W., Ober, M., Carell, T. & Beese, L.S. (2004) Error-prone replication of oxidatively damaged DNA by a high-fidelity DNA polymerase. Nature 431, 217–221.[CrossRef][Medline]

Imlay, J.A. (2003) Pathways of oxidative damage. Annu. Rev. Microbiol. 57, 395–418.[CrossRef][Medline]

Imlay, J.A. & Linn, S. (1986) Bimodal pattern of killing of DNA-repair-defective or anoxically grown Escherichia coli by hydrogen peroxide. J. Bacteriol. 166, 519–527.[Abstract/Free Full Text]

Maki, H. (2002) Origins of spontaneous mutations: specificity and directionality of base-substitution, frameshift, and sequence-substitution mutageneses. Annu. Rev. Genet. 36, 279–303.[CrossRef][Medline]

Maki, H. & Sekiguchi, M. (1992) MutT protein specifically hydrolyses a potent mutagenic substrate for DNA synthesis. Nature 355, 273–275.[CrossRef][Medline]

Michaels, M.L., Cruz, C., Grollman, A.P. & Miller, J.H. (1992) Evidence that MutY and MutM combine to prevent mutations by an oxidatively damaged form of guanine in DNA. Proc. Natl. Acad. Sci. USA 89, 7022–7025.[Abstract/Free Full Text]

Mo, J.Y., Maki, H. & Sekiguchi, M. (1991) Mutational specificity of the dnaE173 mutator associated with a defect in the catalytic subunit of DNA polymerase III of Escherichia coli. J. Mol. Biol. 222, 925–936.[CrossRef][Medline]

Nunoshiba, T., Obata, F., Boss, A.C., et al. (1999) Role of iron and superoxide for generation of hydroxyl radical, oxidative DNA lesions, and mutagenesis in Escherichia coli. J. Biol. Chem. 274, 34832–34837.[Abstract/Free Full Text]

Park, S., You, X. & Imlay, J.A. (2005) Substantial DNA damage from submicromolar intracellular hydrogen peroxide detected in Hpx^– mutants of Escherichia coli. Proc. Natl. Acad. Sci. USA 102, 9317–9322.[Abstract/Free Full Text]

Sakumi, K., Tominaga, Y., Furuichi, M., et al. (2003) Ogg1 knockout-associated lung tumorigenesis and its suppression by Mth1 gene disruption. Cancer Res. 63, 902–905.[Abstract/Free Full Text]

Seaver, L.C. & Imlay, J.A. (2001) Alkyl hydroperoxide reductase is the primary scavenger of endogenous hydrogen peroxide in Escherichia coli. J. Bacteriol. 183, 7173–7181.[Abstract/Free Full Text]

Tajiri, T., Maki, H. & Sekiguchi, M. (1995) Functional cooperation of MutT, MutM and MutY proteins in preventing mutations caused by spontaneous oxidation of guanine nucleotide in Escherichia coli. Mutat. Res. 336, 257–267.[Medline]

Tchou, J., Kasai, H., Shibutani, S., et al. (1991) 8-oxoguanine (8-hydroxyguanine) DNA glycosylase and its substrate specificity. Proc. Natl. Acad. Sci. USA 88, 4690–4694.[Abstract/Free Full Text]

Weiss, B. (2001) Endonuclease V of Escherichia coli prevents mutations from nitrosative deamination during nitrate/nitrite respiration. Mutat. Res. 461, 301–309.[Medline]

Yoshiyama, K., Higuchi, K., Matsumura, H. & Maki, H. (2001) Directionality of DNA replication fork movement strongly affects the generation of spontaneous mutations in Escherichia coli. J. Mol. Biol. 307, 1195–1206.[CrossRef][Medline]

Yoshiyama, K. & Maki, H. (2003) Spontaneous hotspot mutations resistant to mismatch correction in Escherichia coli: transcription-dependent mutagenesis involving template-switching mechanisms. J. Mol. Biol. 327, 7–18.[CrossRef][Medline]

Received: 23 January 2006
Accepted: 5 April 2006

This article has been cited by other articles:

				I. Wiegand, A. K. Marr, E. B. M. Breidenstein, K. N. Schurek, P. Taylor, and R. E. W. Hancock Mutator Genes Giving Rise to Decreased Antibiotic Susceptibility in Pseudomonas aeruginosa Antimicrob. Agents Chemother., October 1, 2008; 52(10): 3810 - 3813. [Abstract] [Full Text] [PDF]

				K. Driffield, K. Miller, J. M. Bostock, A. J. O'Neill, and I. Chopra Increased mutability of Pseudomonas aeruginosa in biofilms J. Antimicrob. Chemother., May 1, 2008; 61(5): 1053 - 1056. [Abstract] [Full Text] [PDF]

				K. Hasegawa, K. Yoshiyama, and H. Maki Spontaneous mutagenesis associated with nucleotide excision repair in Escherichia coli. Genes Cells, May 1, 2008; 13(5): 459 - 469. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sakai, A. Articles by Maki, H. Search for Related Content PubMed PubMed Citation Articles by Sakai, A. Articles by Maki, H.