Suppression of gamma ray-induced illegitimate recombination in Escherichia coli by the DNA-binding protein H-NS

To study the mechanism of gamma-ray-induced illegitimate recombination, we examined the formation of lambdabio transducing phage in Escherichia coli after gamma-ray irradiation. We show that gamma-ray irradiation enhances the formation of lambdabio transducing phage during prophage induction. Moreover, an hns mutation synergistically enhanced the incidence of lambda-ray-induced illegitimate recombination. Next we determined the sequences at the recombination junctions of the lambdabio transducing phages induced by gamma-ray irradiation. Most of the recombination sites coincided with known hotspots. Among them, hotspot I accounted for 67% and 77% of gamma-ray-induced lambdabio transducing phages in the wild type and the hns mutant, respectively. Therefore, the recombination sites appear to occur mostly at hotspot I or at other hotspots, but rarely at non-hotspot sites. These results suggest that types of DNA damage other than the double-strand breaks induced at random sites are mainly responsible for the introduction of the site-specific or region-specific DNA double strand breaks that lead to recombination at the hotspots. The results also showed that the recombination events took place between DNA sequences possessing short stretches of homology. H-NS protein, which binds to curved DNA, suppresses illegitimate recombination in the presence and absence of gamma-ray irradiation. Models for gamma-ray-induced illegitimate recombination are discussed.     

The RecJ DNase strongly suppresses genomic integration of short but not long foreign DNA fragments by homology-facilitated illegitimate recombination during transformation of Acinetobacter baylyi

Homology-facilitated illegitimate recombination (HFIR) promotes genomic integration of foreign DNA with a single segment homologous to the recipient genome by homologous recombination in the segment accompanied by illegitimate fusion of the heterologous sequence. During natural transformation of Acinetobacter baylyi HFIR occurs at about 0.01% of the frequency of fully homologous recombination. The role of the 5' single-strand-specific exonuclease RecJ in HFIR was investigated. Deletion of recJ increased HFIR frequency about 20-fold compared with wild type while homologous recombination was not affected. Illegitimate fusion sites were predominantly located within 360 nucleotides away from the homology whereas in wild type most fusion sites were distal (500-2500 nucleotides away). RecJ overproduction reduced the HFIR frequency to half compared with wild type, and transformants with short foreign DNA segments were diminished, leading to on average 866 foreign nucleotides integrated per event (682 in wild type, 115 in recJ). In recJ always the 3' ends of donor DNA were integrated at the homology whereas in wild type these were 3' or 5'. RecJ apparently suppresses HFIR by degrading 5' non-homologous DNA tails at the post-synaptic stage. We propose that the RecJ activity level controls the HFIR frequency during transformation and the amount of foreign DNA integrated per event.     

Formation of lambda transducing phage in vitro: involvement of DNA gyrase

We found that transducing phages carrying the gal or bio regions of the Escherichia coli genome were formed during in vitro packaging of endogenous lambda DNA. Structural analysis of the transducing phage genomes indicated that they were formed by abnormal excision of lambda prophage. Formation of transducing phages was stimulated by oxolinic acid, an inhibitor of DNA gyrase, implying that DNA gyrase participates in the abnormal excision of lambda prophage. When pBR322 DNA was added to the reaction mixture, transducing phages into which pBR322 had been inserted were produced at a high frequency. This reaction was also stimulated by oxolinic acid. Sequence analyses revealed that pBR322 is inserted into the sites of abnormal excision of the prophage. These results show that transducing phages can be formed by DNA gyrase-dependent illegitimate recombination in an in vitro system and that secondary recombination takes place frequently at the site where the first recombination occurs.     

Roles of RecJ,RecO, and RecR in RecET-mediated illegitimate recombination in Escherichia coli

We analyzed effects of overexpression of RecE and RecT on illegitimate recombination during prophage induction in Escherichia coli and found that frequencies of spontaneous and UV-induced illegitimate recombination are enhanced by coexpression of RecE and RecT in the wild type, but the enhanced recombination was reduced by recJ, recO, or recR mutation. The results indicated that RecET-mediated illegitimate recombination depends on the functions of RecJ, RecO, and RecR, suggesting that the RecE and RecJ exonucleases play different roles in this recombination pathway and that the RecO and RecR proteins also play important roles in the recombination. On the other hand, the frequency of the RecET-mediated illegitimate recombination was enhanced by a recQ mutation, implying that the RecQ protein plays a role in suppression of RecET-mediated illegitimate recombination. It was also found that RecET-mediated illegitimate recombination is independent of the RecA function with UV irradiation, but it is enhanced by the recA mutation without UV irradiation. Based on these results, we propose a model for the roles of RecJOR on RecET-mediated illegitimate recombination.     

Molecular Analysis of λbio Transducing Phage Produced by Oxolinic Acid-Induced Illegitimate Recombination in Vivo

To study the mechanism of DNA gyrase-mediated illegitimate recombination in Escherichia coli, we examined the formation of λ Spi(-) phage during prophage induction. The frequency of Spi(-) phage was two to three orders of magnitude higher in the presence of oxolinic acid, an inhibitor of DNA gyrase A subunit, than in the absence of the drug, while it was very low in nalA(r) bacteria with the drug. RecA function is not required for the formation of these phages, indicating that this enhancement is not caused by the expression of SOS-controlled genes. Analyses of att region and recombination junctions of Spi(-) phages revealed that they have essentially the same structures as λbio transducing phages but are classified into two groups with respect to recombination sites. In the majority class of the transducing phages, there were not more than 3-bp homologies bewteen the parental E. coli bio and λ recombination sites. In the minority class of the transducing phages, on the other hand, 9-10-bp homologies were found between the parental recombination sites. These results suggested that oxolinic acid-induced illegitimate recombination takes place by two variants of a DNA gyrase-dependent mechanism.     

Host RecJ is required for growth of P22 erf bacteriophage.

Growth of bacteriophage P22 erf is known to require host RecA recombination function. We show that the RecA function is necessary but not sufficient to restore the plaque-forming ability of phage P22 erf; such mutant phage also requires host RecJ function. The residual efficiency of plaquing of P22 erf in a recJ background (0.03%) is completely abolished in recJ recB hosts (< 0.001%), suggesting that the RecBCD nuclease can provide an alternative function allowing phage growth. One tentative explanation is that circularization of P22 erf DNA mostly proceeds through the RecF pathway of recombination; however, less efficient circularization via the RecBCD pathway may also occur. In a recJ background, lysates obtained upon induction of an erf prophage show reduced yield (10%), suggesting that growth of P22 erf may require host RecJ in a step(s) other than circularization of phage DNA.     

A hotspot of spontaneous and UV-induced illegitimate recombination during formation ofλbio transducing phage

To study the mechanism of spontaneous and UV-induced illegitimate recombination, we examined the formation of theλbio specialized transducing phage inEscherichia coli. Because mostλbio transducing phages have double defects in thered andgam genes and have the capacity to form a plaque on anE. coli P2 lysogen (Spi^? phenotype), we selectedλbio transducing phage by their Spi^? phenotype, rather than using thebio marker. We determined sequences of recombination junctions ofλbio transducing phages isolated with or without UV irradiation and deduced sequences of parental recombination sites. The recombination sites were widely distributed onE. coli bio andλ DNAs, except for a hotspot which accounts for 57% of UV-inducedλbio transducing phages and 77% of spontaneously inducedλbio transducing phages. The hotspot sites onE. coli andλ DNAs shared a short homology of 9 bp. In addition, we detected direct repeat sequences of 8 by within and near both thebio andλ hotspots. ArecA mutation did not affect the frequency of the recombination at the hotspot, indicating that this recombination is not a variant ofrecA-dependent homologous recombination. We discuss a model in which the short homology as well as the direct repeats play essential roles in illegitimate recombination at the hotspot.     

Evidence that SbcB and RecF pathway functions contribute to RecBCD-dependent transductional recombination.

A role for the RecF, RecJ, and SbcB proteins in the RecBCD-dependent recombination pathway is suggested on the basis of the effect of null recF, recJ, and sbcB mutations in Salmonella typhimurium on a "short-homology" P22 transduction assay. The assay requires recombination within short (approximately 3-kb) sequences that flank the selected marker and lie at the ends of the transduced fragment. Since these ends are subject to exonucleolytic degradation, the assay may demand rapid recombination by requiring that the exchange be completed before the essential recombining sequences are degraded. In this assay, recF, recJ, and sbcB null mutations, tested individually, cause a small decrease in recombinant recovery but all pairwise combinations of these mutations cause a 10- to 30-fold reduction. In a recD mutant recipient, which shows increased recombination, these pairwise mutation combinations cause a 100-fold reduction in recombinant recovery. In a standard transduction assay (about 20 kb of flanking sequence), recF, recJ, and sbcB mutations have a very small effect on recombinant frequency. We suggest that these three proteins promote a rate-limiting step in the RecBC-dependent recombination process. The above results were obtained with a lysogenic recipient strain which represses expression of superinfecting phage genomes and minimizes the contribution of phage recombination functions. When a nonlysogenic recipient strain is used, coinfecting phage genomes express functions that alter the genetic requirements for recombination in the short-homology assay.     

A hotspot of spontaneous and UV-induced illegitimate recombination during formation ofλbio transducing phage

To study the mechanism of spontaneous and UV-induced illegitimate recombination, we examined the formation of thebio specialized transducing phage inEscherichia coli. Because mostbio transducing phages have double defects in thered andgam genes and have the capacity to form a plaque on anE. coli P2 lysogen (Spi^– phenotype), we selectedbio transducing phage by their Spi^– phenotype, rather than using thebio marker. We determined sequences of recombination junctions ofbio transducing phages isolated with or without UV irradiation and deduced sequences of parental recombination sites. The recombination sites were widely distributed onE. coli bio and DNAs, except for a hotspot which accounts for 57% of UV-inducedbio transducing phages and 77% of spontaneously inducedbio transducing phages. The hotspot sites onE. coli and DNAs shared a short homology of 9 bp. In addition, we detected direct repeat sequences of 8 by within and near both thebio and hotspots. ArecA mutation did not affect the frequency of the recombination at the hotspot, indicating that this recombination is not a variant ofrecA-dependent homologous recombination. We discuss a model in which the short homology as well as the direct repeats play essential roles in illegitimate recombination at the hotspot.     

Mechanism of DNA gyrase-mediated illegitimate recombination: characterization of Escherichia coli gyrA mutations that confer hyper-recombination phenotype.

To study the mechanism of DNA gyrase-mediated illegitimate recombination in Escherichia coli, we isolated temperature-sensitive gyrA mutants that confer spontaneous illegitimate recombination and spontaneous induction of lambda prophage at higher frequencies than that in the wild-type. After reconstruction of single mutations by targeted mutagenesis, we confirmed that two single mutations, gyrAL492P and gyrAL488P, and a double mutation, gyrAI203V+gyrAI205V, show the same properties as those described above. With respect to the phenotypes of hyper-recombination and higher induction of lambda prophage, these mutations were dominant over the wild-type. Analysis of recombination junctions of lambdabio transducing phages formed spontaneously in these mutants showed that the parental E. coli bio and lambda recombination sites have a homologous sequence of only 0. 7 base-pair on average, indicating that homology is not required for this illegitimate recombination. Analysis of nucleotide sequences of mutant gyrA genes revealed that the gyrAL492P and gyrAL488P mutations contain amino acid substitutions of Leu492-->Pro and Leu488-->Pro, respectively, which correspond to the alpha18 helix in the breakage-reunion domain of DNA gyrase A subunit. The gyrAI203V and gyrAI205V mutations contain Ile203-->Val and Ile205-->Val, respectively, which correspond to the alpha10' helix, also in the breakage-reunion domain of DNA gyrase A subunit. Biochemical analysis indicated that the GyrA63 protein that contains the L492P mutation has an apparently normal supercoiling activity, but it also produces a small amount of linear DNA in the absence of DNA gyrase inhibitor during the supercoiling reaction, suggesting that the mutant DNA gyrase may have a defect at the step of religation or a defect in the subunit interaction. These results suggest that the recombination is induced by defects of religation and/or dimer formation in the mutant DNA gyrases, implying that two alpha helices, alpha10' and alpha18, of DNA gyrase A subunit have crucial roles in subunit interaction and/or resealing of DNA.     

Role of DNA ligase in the illegitimate recombination that generates lambdabio-transducing phages in Escherichia coli

We studied the role of DNA ligase in illegitimate recombination in Escherichia coli. A temperature-sensitive mutation in the lig gene reduced the frequency with which lambdabio-transducing phages were generated to 10-14% of that of wild type under UV irradiation. Reintroduction of the lig gene into this mutant restored the frequency of recombinant phage generation to that of wild type. Furthermore, overexpression of DNA ligase enhanced illegitimate recombination by 10-fold with or without UV irradiation. In addition, when DNA ligase was present in only limited amounts, UV-induced or spontaneous illegitimate recombination occurred exclusively at hotspot sites that have relatively long sequences of homology (9 or 13 bp). However, when DNA ligase was overexpressed, most of the illegitimate recombination took place at non-hotspot sites having only short sequences of homology (<4 bp). Thus, the level of ligase activity affects the frequency of illegitimate recombination, the length of sequence homology at the recombination sites, and the preference for recombination at hotspots, at least after UV irradiation. These observations support our hypothesis that the illegitimate recombination that generates lambdabio-transducing phages is mediated by the DNA break-and-join mechanism.     

Molecular analysis of the recombination junctions of lambda bio transducing phages.

Summary To examine the mechanism of recombination involved in the formation of specialized transducing phage during the induction of bacteriophage we have determined the nucleotide sequences of the recombination junctions of bio phages. The results indicate that abnormal excision takes place at many sites on both bacterial and phage genomes and that the recombination sites have short regions of homology (5–14 bp). Some of the sequences of the recombination sites were similar to the consensus sequences of DNA gyrase-cleavage sites and repetitive extragenic palindromic (REP) sequences. These results showed that abnormal excision is a type of illegitimate recombination. The possible involvement of DNA gyrase in this recombination is discussed.     

Suppression of recA deficiency in plasmid recombination by bacteriophage lambda beta protein in RecBCD- ExoI- Escherichia coli cells.

Plasmid recombination, like other homologous recombination in Escherichia coli, requires RecA protein in most conditions. We have found that the plasmid recombination defect in a recA mutant can be efficiently suppressed by the beta protein of bacteriophage lambda. beta protein is required for homologous recombination of lambda chromosomes during lytic phage growth in a recA host and is known to have a strand-annealing activity resembling that of RecA protein. The bioluminescence recombination assay was used for genetic analysis of beta-protein-mediated plasmid recombination. Efficient suppression of the recA mutation by beta protein required the absence of the E. coli nucleases exonuclease I and RecBCD nuclease. These nucleases inhibit a RecA-mediated plasmid recombination pathway that is more efficient than the pathway functioning in wild-type cells. Like RecA-mediated plasmid recombination in RecBCD- ExoI- cells, beta-protein-mediated plasmid recombination depended on concurrent DNA replication and on the activity of the recQ gene. However, unlike RecA-mediated plasmid recombination, beta-protein-mediated recombination in RecBCD- ExoI- cells was independent of recF and recJ activities. We propose that inactivation of exonuclease I and RecBCD nuclease stabilizes a recombination intermediate that is involved in RecA- and beta-protein-catalyzed homologous pairing reactions. We suggest that the intermediate may be linear plasmid DNA with a protruding 3' end, since these nucleases are known to interfere with the synthesis of such linear forms. The different recF and recJ requirements for beta-protein-dependent and RecA-dependent recombinations imply that the mechanisms of formation or processing of the putative intermediate differ in the two cases.     

Neisseria gonorrhoeae recJ mutants show defects in recombinational repair of alkylated bases and UV-induced pyrimidine dimers

Neisseria gonorrhoeae lacks several common DNA repair pathways found in other organisms. As recent evidence had indicated that gonococci use recombinational repair to repair UV-induced DNA lesions, this study examined whether the gonococcal RecJ homologue contributes in this repair capacity. The recJ gene from strain MS11 was cloned and sequenced and was found to show a considerable degree of identity to its Escherichia coli homologue. A N. gonorrhoeae delta recJ mutant was constructed and tested for recombinational proficiency as well as for defects in DNA repair. In the absence of the RecJ exonuclease, DNA transformation and pilin switching occurred at wild type levels, indicating that the efficiency of recombination remained unimpaired. In contrast, N. gonorrhoeae delta recJ mutants showed extreme sensitivity to low levels of UV irradiation and to exposure to DNA-alkylating reagents [e.g. ethyl methanesulfonate (EMS) and methyl methanesulfonate (MMS)]. Complementation of the gonococcal recJ mutant in cis restored resistance to low-level UV, indicating that the gonococcal RecJ protein is involved in recombinational repair, and can act independently of other single-strand-specific exonucleases. Furthermore, transformation competence was not required for RecJ-dependent DNA repair. Overall, the data show that N. gonorrhoeae recJ mutants present a unique phenotype when compared to their E. coli recJ counterparts, and further support the contention that RecORJ-dependent recombinational repair is a major DNA repair pathway in the genus Neisseria.     

Isolation and characterization of specialized lambda transducing bacteriophage carrying the metBJF methionine gene cluster.

Secondary attachment site lysogens of Deltaatt(lambda)Deltappc-argECBH strains of Escherichia coli with lambdacI857 integrated into the bfe gene (88 min) were isolated. Of 20 such lysogens examined, 2 produce lysates with transducing phage containing the metBJF gene cluster (87 min). Reintroduction of the ppc-argECBH chromosome segment (which lies between the bfe and met genes) into these strains virtually abolishes the production of met transducing phage. All of the phage examined have lost essential genes from the left arm of the lambda chromosome. Approximately 85% of the phage appear to have the same genetic composition, containing the metBJF gene cluster, but not the closely linked gene cytR, and having lost phage genes G and J. Analytical CsCl density gradient centrifugation of five representatives of this major class of phage shows four of them to have identical densities (lighter than lambda), while the fifth cannot be resolved from lambda. The four apparently identical phage were isolated from three separate lysates, which suggests the existence of preferred sites for illegitimate recombination on the bacterial and phage chromosomes. Three specialized transducing phage that carry cytR in addition to metB, metJ, and metF have also been studied. Each of these viruses has a different amount of phage deoxyribonucleic acid. Two of them have less deoxyribonucleic acid than lambda, whereas the third has about the same amount. The metB, metF, and cytR genes of the transducing phage have been shown to function in vivo. The phage-borne metB and metF genes are subject to metJ-mediated repression.     

Exonuclease requirements for recombination of lambda-phage in recD mutants of Escherichia coli

Recombination of lambda red gam phage in recD mutants is unaffected by inactivation of RecJ exonuclease. Since nucleases play redundant roles in E. coli, we inactivated several exonucleases in a recD mutant and discovered that 5'-3' exonuclease activity of RecJ and exonuclease VII is essential for lambda-recombination, whereas exonucleases of 3'-5' polarity are dispensable. The implications of the presented data on current models for recombination initiation in E. coli are discussed.     

Function and biochemical characterization of RecJ in Deinococcus radiodurans

The single-stranded DNA-specific nuclease RecJ is found in most bacteria where it is involved in the RecFOR double-stranded break (DSBs) repair pathway. DSBs repair mainly occurs via the RecFOR pathway in Deinococcus radiodurans, a well-known radiation-resistant bacterium. A recJ null mutant was constructed to investigate the role of recJ in D. radiodurans. recJ inactivation caused growth defects and sensitivity to high temperatures. However, the radiation resistance of the recJ mutant was only moderately decreased. The full-length D. radiodurans RecJ (DrRecJ) protein was expressed and purified to further characterize its biochemical properties. DrRecJ possessed a Mn(2+) concentration-dependent nuclease activity where the optimal Mn(2+) concentration was 0.1mM. DrRecJ had a similar activity profile after adding 10mM Mg(2+) to reactions with different Mn(2+) concentrations, indicating that Mn(2+) is a RecJ regulator. Escherichia coli RecJ has no activity on 5' ssDNA tails shorter than 6-nt, but DrRecJ could effectively degrade DNA with a 4-nt 5' ssDNA tail, suggesting that DrRecJ may have a wider range of DNA substrates. Moreover, SSB in D. radiodurans stimulated the DrRecJ exonuclease activity, whereas DdrB inhibited it and provided protection to ssDNA. Overall, our results indicate that recJ is a nonessential gene in D. radiodurans and that the activity of DrRecJ is regulated by Mn(2+) and SSB-DdrB.     

Nucleotide sequence of the Escherichia coli recJ chromosomal region and construction of recJ-overexpression plasmids.

The nucleotide sequence of the recJ gene of Escherichia coli K-12 and two upstream coding regions was determined. Three regions were identified within these two upstream genes that exhibited weak to moderate promoter activity in fusions to the galK gene and are candidates for the recJ promoter. recJ appeared to be poorly translated: the recJ nucleotide sequence revealed a suboptimal initiation codon GUG, no discernible ribosome-binding consensus sequence, and relatively nonbiased synonymous codon usage. Comparison of the sequence of this region of the chromosome with DNA data bases identified the gene immediately downstream of recJ as prfB, which encodes translational release factor 2 and has been mapped near recJ at 62 min. No significant homology between recJ and other previously sequenced regions of DNA was detected. However, protein sequence comparisons with a gene upstream of recJ, denoted xprB, revealed significant homology with several site-specific recombination proteins. Its genetic function is presently unknown. Knowledge of the nucleotide sequence of recJ allowed the construction of a plasmid from which overexpression of RecJ protein could be induced. Supporting the notion that translation of recJ is limiting, a strong T7 bacteriophage promoter upstream of recJ did not, by itself, allow high-level expression of RecJ protein. The addition of a ribosome-binding sequence fused to the initiator GTG of recJ in this construction was necessary to promote expression of high levels of RecJ protein.     

A thermostable single-strand DNase from Methanococcus jannaschii related to the RecJ recombination and repair exonuclease from Escherichia coli

The RecJ protein of Escherichia coli plays an important role in a number of DNA repair and recombination pathways. RecJ catalyzes processive degradation of single-stranded DNA in a 5'-to-3' direction. Sequences highly related to those encoding RecJ can be found in most of the eubacterial genomes sequenced to date. From alignment of these sequences, seven conserved motifs are apparent. At least five of these motifs are shared among a large family of proteins in eubacteria, eukaryotes, and archaea, including the PPX1 polyphosphatase of yeast and Drosophila Prune. Archaeal genomes are particularly rich in such sequences, but it has not been clear whether any of the encoded proteins play a functional role similar to that of RecJ exonuclease. We have investigated three such proteins from Methanococcus jannaschii with the strongest overall sequence similarity to E. coli RecJ. Two of the genes, MJ0977 and MJ0831, partially complement a recJ mutant phenotype in E. coli. The expression of MJ0977 in E. coli resulted in high levels of a thermostable single-stranded DNase activity with properties similar to those of RecJ exonuclease. Despite overall weak sequence similarity between the MJ0977 product and RecJ, these nucleases are likely to have similar biological functions.     

Properties of Escherichia coli expressing bacteriophage P22 Abc (anti-RecBCD) proteins, including inhibition of Chi activity.

Escherichia coli strains bearing plasmids expressing phage P22 anti-RecBCD functions abc1 and abc2 were tested for the presence of recBC-like phenotypes. Abc2 induces moderate sensitivity to UV light in wild-type and recD mutant strains but severely sensitizes both recF and recJ mutants. Abc1 has little effect on UV sensitivity in wild-type or recF or recJ mutant hosts but increases the sensitivity of recD mutants to a UV dose of 20 J/m2 about 10-fold. Abc2 induces E. coli to segregate inviable cells during growth, interferes with the growth of lambda red gam chi+ and chi 0 phage (the effect is greater with chi+ phage), inhibits Chi and Chi-like activity as measured by lambda red gam crosses, and prevents SOS induction in response to nalidixic acid; Abc1 has no effect in these tests. Abc2, alone or with Abc1, does not allow the growth of lambda red gam in the presence of a P2 prophage but does not kill the P2 lysogenic host (as lambda Gam does). Finally, Abc2 inhibits conjugational recombination in wild-type cells to the level seen in recBC mutants. These data suggest that Abc2 inhibits the recombination-promoting ability of RecBCD but leaves the exonuclease functions intact.