INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Illegitimate recombination (IR) takes place between DNA sequences that have little or no homology and is a major cause of chromosomal aberrations, such as deletions or translocations. IR can be classified into two types, short-homology-independent IR (SHIIR) and short-homology-dependent IR (SHDIR). SHIIR occurs between sequences having virtually no homology and is mediated by DNA topoisomerases (2, 3, 24). SHDIR is induced by UV irradiation or other DNA-damaging agents and requires short regions of homology between recombination sites (24, 25, 28).

SHDIR usually takes place at a low frequency, but it is greatly enhanced by genotoxic agents (11, 18). It is known that this type of recombination is enhanced by RecE and RecJ and is suppressed by DNA polymerase I, SbcB, RecQ, and UvrAB (1, 7, 9, 10, 25, 27). Previous studies have suggested that DNA ends are processed by 5'-to-3' exonucleases, such as RecE or RecJ, and as a result DNAs with a 3' overhang are formed. These molecules anneal with other molecules carrying a complementary overhang, and then end joining takes place. It has also been found that DNA-binding proteins affect this type of IR. Fis and IHF (integration host factor) enhance SHDIR, and H-NS suppresses it (21, 22).

It is known that a double-stranded break (DSB) can result from arrest of a replication fork, which may be caused by DNA secondary structure, DNA lesions, or DNA-bound proteins (5, 6, 8). The involvement of Rep and DnaB helicases in DSB formation indicates that there is a link between DNA replication stoppage and the formation of DSBs (13). In addition, RuvABC, which catalyze branch migration and cleavage of Holliday junctions, are responsible for DSB formation at stalled replication forks (20).

In this study, to determine the role of DNA replication in UV-induced IR, we examined the effects of dnaB(Ts) mutations on IR. One of the mutants used, a dnaB14 mutant, had a defect in the elongation process during DNA replication (12), and another mutant, a dnaB252 mutant, had a defect in initiation of DNA replication (19). We found that the frequency of UV-induced IR was reduced by the elongation-deficient mutation in the dnaB14 mutant but not by the initiation-deficient mutation in the dnaB252 mutant. Thus, we concluded that processivity of DnaB helicase is a prerequisite for UV-induced IR. Below we discuss a model for the possible function of DnaB in UV-induced IR.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains. The Escherichia coli strains used in this study are described in Table 1. Most of the strains used in this study are derivatives of E. coli K-12 which contain one unit of the  cI857 prophage; the exceptions are Ymel and the P2 lysogen. Ymel was used for titration of the total number of  phage, and the P2 lysogen was used for titration of the number of  Spi phage.

Measurement of the frequency of  Spi phage induced spontaneously or by UV irradiation. E. coli  cI857 or a derivative of this strain was grown at 30°C in  YP broth. If necessary, 2 ml of the culture was irradiated with a UV lamp (15 W) with a wavelength of 253.6 nm. Thermal induction of  prophage was carried out by incubation at 42°C for 15 min. The culture was then incubated at 37°C for 2 to 4 h. The titer of  Spi phage was measured on a lawn of E. coli WL95 (P2). The number of all  phage was determined on a lawn of E. coli Ymel. The frequency of IR was calculated by dividing the number of  Spi phage by the total number of all  phage (11).

Measurement of the frequency of  Spi phage in a lysogen carrying  xis1 prophage. E. coli  cI857 xis1 or a derivative of this strain was thermally induced, and phage lysate was prepared as described above. The frequency of IR was calculated by dividing the number of  Spi phage by the number of cells before prophage induction (21).

Independent isolation of bio transducing phages induced by UV irradiation. E. coli  lysogen was irradiated with UV as described above. The culture was then divided into 50 small tubes. Each tube containing 0.5 ml of the culture was then incubated at 42°C for 15 min. The cultures were then incubated at 37°C for 2 h. Phage lysates were plated on a lawn of E. coli WL95. A plaque derived from each tube was picked, suspended in M9 buffer, and plated on a lawn of Ymel to isolate a single clone.

Determination of the locations of recombination junctions in  Spi phages. The locations of recombination junctions were determined by PCR by using multiple combinations of primers. bio transducing phages were identified by using a mixture of several sets of primers. The method used in this study has been described previously by Ukita and Ikeda (25). The recombination junctions were then sequenced and analyzed with an ABI PRISM 310 genetic analyzer (Perkin-Elmer).

Sequence analysis of the mutation site in the dnaB14 mutant. The mutant dnaB gene was amplified by PCR performed with oligomers B1FH (5'-CAA TTC GTG TTG CCA TGT G-3') and B4RB (5'-TCA CAA CAG TTG CCG CTT G-3') and template DNA extracted from a dnaB mutant. The amplified 1.6-kb fragment contained a region from 100 bp upstream of the ATG codon to 50 bp downstream of the stop codon of the dnaB gene. The fragment was directly sequenced by using the oligomers mentioned above, as well as B3F (5'-TGG AGA TGC CAT CAG AAC AC-3') and B2R (5'-TTC CAG CGT ACG GTA ATC CGA AAG C-3').

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Characterization of the dnaB mutations. To study the role of DNA replication in IR, we examined the effects of two temperature-sensitive mutations on IR. One mutation, the dnaB14 mutation, has a defect in elongation of DNA replication (12), and the other mutation, the dnaB252 mutation, has a defect in initiation of DNA replication (19). Because the dnaB14 mutation site is not known yet, we determined the nucleotide sequence of the mutant gene and found that it contains a point mutation consisting of the following base substitution: the 46th G from the first ATG is changed to an A. This finding indicates that the mutated DnaB protein has a D16K amino acid substitution. On the other hand, the dnaB252 mutation has been demonstrated previously to be a base substitution consisting of a change of the 869th G to an A. The mutant protein therefore has a G299E amino acid substitution (19).

Effects of the dnaB mutations on IR. We next tested the effects of the dnaB mutations on IR during prophage induction. To do this, the wild-type strain and the strain with a heat-sensitive mutation in the dnaB gene were lysogenized by infection of  cI857 and subjected to the  Spi assay. This assay utilizes the Spi phenotype as a marker for bio transducing phages (11). Specialized transducing phages generated from the  prophage by IR usually contain the E. coli genes gal and bio, which are adjacent to the phage genome. Thus, most bio transducing phages are defective in the red-gam region of  phage DNA and can form plaques on an E. coli P2 lysogen lawn (Spi phenotype), whereas normal  phage cannot. Thus, it is possible to select  Spi phage from the phage pool. The number of  Spi phage is assumed to be the same as the number of bio transducing phages since previous experiments have shown that most  Spi phages are bio phages (11, 28).

View larger version (48K): [in this window] [in a new window]   FIG. 1.   Effects of dnaB(Ts) mutations on UV-induced IR. An E. coli  cI857 lysogen (HI1756 or HI2154) or a dnaB(Ts) mutant (HI2151 or HI2155) was treated with several different doses of UV irradiation. A  lysogen was then thermally induced at 42°C and incubated at 37°C to prepare phage lysate. The frequency of IR was measured as described in Materials and Methods. The values are averages based on four independent measurements. The error bars indicate standard errors. (A) dnaB14; (B) dnaB252.

Effect of the dnaB14 mutation on formation of  Spi phages in an E. coli  xis1 lysogen. The results described above imply that DnaB helicase directly participates in IR. However, it is also possible that the transient defect in DnaB function during prophage induction may temporarily stop DNA replication. As a result, Int-mediated excision of  prophage may take place during this stop in replication, resulting in the loss of  prophage and a reduction in the formation of bio transducing phage. To rule out this possibility, we examined the effect of the dnaB14 mutation on formation of  Spi phage in an E. coli strain carrying an excision-deficient  prophage. Since Int-mediated excision is not observed in the  xis1 mutant,  xis1 prophage should not be lost during thermal induction even in the dnaB(Ts) mutant (21). The results obtained indicated that the frequency of IR in the dnaB14 mutant carrying  xis1 prophage was lower than that in the wild type, confirming that DnaB helicase is directly involved in IR (Fig. 2).

View larger version (34K): [in this window] [in a new window]   FIG. 2.   Effect of the dnaB14 mutation on UV-induced IR during induction of  xis1 lysogen. E. coli  cI857 xis1 lysogen (HI2158) or a dnaB(Ts) mutant (HI2159) was irradiated with UV, and phage lysate was prepared after thermal induction. The frequency of IR was determined as described in Materials and Methods.

Distribution and nucleotide sequences of junctions formed by UV-induced IR in the dnaB14 mutant. We determined the distribution of recombination junctions of  Spi phages derived from the wild type and the dnaB14 mutant. As shown in a previous study (28), approximately one-half of the recombinants are formed by recombination at hotspot I sites. These sites are located in the bioC gene of E. coli and the gam gene of  DNA. In addition, the relative rates of recombination at hotspots II and III were 13 and 8%, respectively (Table 2). On the other hand, in the recombinants derived from the dnaB14 mutant, the relative rates of recombination at hotspots I and II were 17 and 66%, respectively (Table 2). Recombination at hotspot III was not detected in the phages derived from this mutant. These results indicate that the dnaB14 mutation preferentially suppressed IR at hotspots I and III.

View this table: [in this window] [in a new window]   TABLE 2.   Distribution of recombination sites of bio transducing phages formed during UV irradiationa

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Distribution of lengths of overlapping nucleotides at recombination junctions. The y axis indicates the number of bio transducing phages, and the x axis indicates the number of overlapping nucleotides at the recombination junction (i.e., the number of overlapping nucleotides in the  and E. coli genomes). (A) Distribution of the lengths of overlapping nucleotides at the junctions derived from wild-type strain HI1756. (B) Distribution of the lengths of overlapping nucleotides at the junctions derived from dnaB14 mutant HI2151.

Effect of the dnaB14 recQ double mutation on IR. In a previous study, we showed that the frequency of IR is increased by the recQ mutation with or without UV irradiation (10). This result indicates that RecQ is a suppressor of IR. To analyze the functional relationship between RecQ helicase and DnaB helicase in IR, we determined the frequency of IR in the recQ dnaB14 double mutant. The frequency of IR in the double mutant was lower than that in the recQ single mutant, with or without UV irradiation (Fig. 4). These results indicate that the increased IR in the recQ mutant is suppressed by the dnaB14 mutation irrespective of UV irradiation.

View larger version (38K): [in this window] [in a new window]   FIG. 4.   Effect of the recQ dnaB14 double mutation on IR. The frequency of  Spi phages was measured in a recQ single mutant, HI2700, and in a recQ dnaB14 double mutant, HI2701, with or without UV irradiation as described in the legend to Fig. 1.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The dnaB14 mutation is located at amino acid 16 (Glu is changed to Lys), which occurs in the amino-terminal fragment (fragment III) that is about 12 kDa long (amino acid 15 to amino acid 126 or 128). This domain is known to be essential for DnaB helicase activity and for DnaB interaction with certain primases, such as DnaC, and other primosomal proteins (14, 15). It is also known that the dnaB14 mutation causes a defect in elongation during DNA replication (12) and that the dnaB252 mutation, which contains a G299E amino acid substitution, has a defect in initiation of DNA replication (19). Upon thermal treatment of bacteria for a short time, the dnaB14 mutation did not significantly affect the yield of all  phages but greatly affected the frequency of UV-induced IR. On the other hand, the dnaB252 mutation did not affect UV-induced IR. These results indicated that DnaB is involved in UV-induced IR, implying that progression of the replication fork and collision of the replication fork with a damaged region of DNA are essential for IR.

DnaB interacts with many replication proteins, such as DnaC, DnaG, DNA polymerase III, and  P protein, resulting in the formation of a replisome. When a DNA lesion is present on a chromosome, DNA polymerase III stalls if the lesion is located on the leading strand during DNA synthesis. On the lagging strand, however, DNA polymerase III arrested at the lesion relocates to the next RNA primer site, resulting in continued DNA synthesis (23). In the replisome, the role of DnaB is to unwind DNA at replication forks. An important property of this process is that the helicase activity of DnaB is not suppressed by the presence of DNA lesions (16). It is therefore possible for a long single-stranded DNA region to be formed at the stalled replication fork. Such a single-stranded DNA region could be cleaved by an endonuclease. The reduced frequency of UV-induced IR in the dnaB14 mutant suggests that the mutant DnaB protein has a low processivity and therefore a low probability of collision with damaged DNA, which would result in lower levels of DSBs.

Michel et al. (13) showed that thermal treatment of the temperature-sensitive dnaB8 mutant induces DSBs in DNA. This observation apparently contradicts the results of the present work. However, we suggest that thermal treatment of the dnaB8 mutant may have resulted in irreversible arrest of the replication forks for long period at a nonpermissive temperature, consequently causing DSB formation at the stalled replication sites. On the other hand, a reduced frequency of UV-induced IR in the dnaB14 mutant was detected immediately after temperature shift-up. Therefore, we propose a model in which the dnaB mutants may exhibit two types of responses, early and late responses, for DSB formation. In the early response, replication forks stall at damaged DNA, resulting in DSB formation, and the dnaB14 mutation affects the DSBs shortly after temperature shift-up. On the other hand, in the late response, arrested replication forks mediated by the dnaB8 mutation may induce DSBs after prolonged incubation.

Two major models, the break and join model and the slipped mispairing model, have been proposed for IR. Both models have been used to explain SHDIR in general. In the latter model, a replication fork slips down a region of the DNA strand during DNA replication, resulting in a deletion or other mutation. The former model consists of three steps: DNA DSB, DNA end processing, and end joining. Recently, Onda et al. (17) showed that reduction or augmentation of ligase activity affects the frequency of UV-induced SHDIR, as well as the lengths of homologous sequences at the recombination junctions in SHDIR. The results of these authors strongly support the idea that UV-induced IR is mediated by the break and join mechanism. It is therefore likely that the role of DnaB in promotion of UV-induced IR is formation of DSBs in DNA.

Since it is known that linear chromosomal DNA accumulates spontaneously in a recB or recC mutant (13), enhanced IR might be expected in such mutants. However, this is not the case. The frequency of IR in a recB recC double mutant with or without UV irradiation was comparable to that in the wild type (Y. Ogata and H. Ikeda, unpublished results). It is therefore possible that linear chromosomal DNA is not an appropriate substrate for bio phage formation. Since it is also known that low-molecular-weight DNAs are abundant in UV-induced cells and the number of DSBs per genome is between 20 and 30 at a UV dose of 100 J/m² (4), there is another possibility, that the low-molecular-weight DNAs are substrates for IR. Wang and Smith (26) proposed a model in which low-molecular-weight DNAs may be produced as a result of breaks in the parental DNA opposite unrepaired DNA daughter strand gaps. The role of DnaB in formation of DSBs or low-molecular-weight DNAs is being investigated in our laboratory.

There are two types of IR that lead to formation of bio transducing phages, SHDIR and SHIIR. Nucleotide sequence analyses of the recombination junctions in the wild type showed that one-half of the junctions were formed at the hotspot I site and the other half were formed at nonhotspot sites, as described previously (28). Most junctions contain short regions of homology in the recombination sites, indicating that DnaB plays a role in promoting SHDIR. We also found that the relative rate of UV-induced recombination in the dnaB14 mutant is reduced at hotspots I and III and increased at hotspot II. This result suggests that DSBs that occur at hotspot I and III sites may be relatively dependent on DnaB. The reason for the preference for DSB sites remains to be determined.

The experiment with the recQ dnaB14 double mutant showed that the effect of the dnaB14 mutation is dominant over the effect of the recQ mutation in the IR pathway. We have previously shown that the recQ mutation enhances IR. Based on the properties of RecQ helicase, we previously proposed that this protein may disrupt recombination intermediates produced by annealing of complementary single-stranded ends (10). These results imply that DnaB helicase plays a role in the introduction of DSBs into DNA at an early step of the IR pathway, while RecQ helicase may act as a suppressor of IR in a later step of this pathway. Moreover, we showed that overproduction of DnaB increased the frequency of IR and that there is a synergy between overproduction of DnaB and the recQ mutation for enhancement of recombination (29). This result is consistent with the proposed functions of DnaB and RecQ in our model of IR (10).