Recombination between repeats in Escherichia coli by a recA-independent, proximity-sensitive mechanism

We have examined the influence of proximity on the efficiency of recombination between repeated DNA sequences in Escherichia coli. Our experiments have employed a plasmid-based assay to detect deletions between direct repeats of 100 bp. The rate of deletion of the juxtaposed direct repeats was reasonably high at 6 × 10^–5 per cell. A comparison of recA+ and recA mutant strains showed that these deletion events are primarily the result of recA-independent recombination at these homologous sequences. Random restriction fragments of yeast or E. coli genomic DNA were used to separate the two repeats. Deletion rates decreased over two orders of magnitude with increasing separation of up to 7 kb. There was a surprisingly strong effect of even short sequence separations, with insertions of a few hundred base pairs exhibiting 10-fold reductions of deletion rates. No effect of recA on the efficiency of deletion was observed at any distance between repeats.     

Recombination between repeats in Escherichia coli by a recA-independent,proximity-sensitive mechanism

We have examined the influence of proximity on the efficiency of recombination between repeated DNA sequences in Escherichia coli. Our experiments have employed a plasmid-based assay to detect deletions between direct repeats of 100 bp. The rate of deletion of the juxtaposed direct repeats was reasonably high at 6 × 10^?5 per cell. A comparison of recA+ and recA mutant strains showed that these deletion events are primarily the result of recA-independent recombination at these homologous sequences. Random restriction fragments of yeast or E. coli genomic DNA were used to separate the two repeats. Deletion rates decreased over two orders of magnitude with increasing separation of up to 7 kb. There was a surprisingly strong effect of even short sequence separations, with insertions of a few hundred base pairs exhibiting 10-fold reductions of deletion rates. No effect of recA on the efficiency of deletion was observed at any distance between repeats.     

Single Strand Annealing Plays a Major Role in RecA-Independent Recombination between Repeated Sequences in the Radioresistant Deinococcus radiodurans Bacterium

The bacterium Deinococcus radiodurans is one of the most radioresistant organisms known. It is able to reconstruct a functional genome from hundreds of radiation-induced chromosomal fragments. Our work aims to highlight the genes involved in recombination between 438 bp direct repeats separated by intervening sequences of various lengths ranging from 1,479 bp to 10,500 bp to restore a functional tetA gene in the presence or absence of radiation-induced DNA double strand breaks. The frequency of spontaneous deletion events between the chromosomal direct repeats were the same in recA+ and in ΔrecA, ΔrecF, and ΔrecO bacteria, whereas recombination between chromosomal and plasmid DNA was shown to be strictly dependent on the RecA and RecF proteins. The presence of mutations in one of the repeated sequence reduced, in a MutS-dependent manner, the frequency of the deletion events. The distance between the repeats did not influence the frequencies of deletion events in recA ⁺ as well in ΔrecA bacteria. The absence of the UvrD protein stimulated the recombination between the direct repeats whereas the absence of the DdrB protein, previously shown to be involved in DNA double strand break repair through a single strand annealing (SSA) pathway, strongly reduces the frequency of RecA- (and RecO-) independent deletions events. The absence of the DdrB protein also increased the lethal sectoring of cells devoid of RecA or RecO protein. γ-irradiation of recA ⁺ cells increased about 10-fold the frequencies of the deletion events, but at a lesser extend in cells devoid of the DdrB protein. Altogether, our results suggest a major role of single strand annealing in DNA repeat deletion events in bacteria devoid of the RecA protein, and also in recA ⁺ bacteria exposed to ionizing radiation.     

Molecular mechanisms of deletion formation in Escherichia coli plasmids. I. Deletion formation mediated by long direct repeats.

Summary Derivatives of plasmid pBR327 with the tet gene interrupted by 165 pb or 401 by direct repeats were constructed. In cells harboring these plasmids, deletions which restored the wild-type tet gene gave rise to tetracycline-resistant colonies, thereby allowing a simple phenotypic test for deletion formation. The frequencies of deletions in these plasmids were measured in Escherichia coli strains proficient or deficient in general recombination. The structure of plasmid DNA isolated from tetracycline-resistant transformants was analyzed by agarose gel electrophoresis, restriction mapping and sequencing. The data presented here demonstrate that deletion formation is always associated with dimerization of plasmid DNA. Dimeric plasmids were of two types. Those which carried both a deletion and a compensating duplication were the major type in a Rec⁺ background and were rare in recA, recF, recJ and recO backgrounds. Dimers of the second type contained deletions, but no compensating duplications, and their formation was RecA-independent. The data presented demonstrate that deletion formation mediated by long direct repeats is mainly the result of unequal crossing-over between two plasmid molecules.     

Intraplasmid recombination in Streptomyces lividans 66

Summary An Escherichia coli-Streptomyces shuttle plasmid pIF132 containing two direct mel repeats was constructed. While pIF132 replicated relatively stably in E. coli (Rec⁺ or recA), its structure was unstable in S. lividans: recombination between the mel repeats resulted in a smaller plasmid, pIF138. Furthermore, pIF132 formed oligomers extensively in E. coli but not in S. lividans.     

Functional characterization of the prokaryotic mobile genetic element IS26

Summary IS26L and IS26R are the 820 bp long elements found as direct repeats at both ends of the kanamycin resistance transposon Tn2680. They can mediate cointegration in E. coli K12 which contains no IS26 in its chromosome. Cointegration occurs in rec ⁺ or recA ^- strains with similar frequency. Upon cointegration mediated by either IS26R or IS26L, the element is duplicated and integrated into one of many different sites. Both IS26L and IS26R carry 14 bp perfect terminal inverted repeats and generate 8 bp direct repeats at their target sequences. Deletion formation mediated by IS26R was also observed. These functional and structural features of IS26 are characteristic of a prokaryotic mobile genetic element.     

Genetic control of intrachromosomal recombination

Intrachromosomal recombination between direct repeats can occur either as gene conversion events, which maintain exactly the number of repeat units, or as deletions, which reduce the number of repeat units. Gene conversions are classical recombination events that utilize the standard chromosome recombination machinery. Spontaneous deletions between direct repeats are generally recA-independent in E. coli and RAD52-independent in S. cerevisiae. This independence from the major recombination genes does not mean that deletions form through a nonrecombinational process. Deletions have been suggested to result from sister chromatid exchange at the replication fork in a recA-independent process. The same type of exchange is proposed to be RAD52-independent in Saccharomyces cerevisiae. RAD52-dependent events encompass all events that involve the initial steps of a recombination reaction, which include strand invasion to form a heteroduplex intermediate.     

Chromosomal mobilization from a recA mutant of Pseudomonas putida.

Summary A methyl methane sulfonate sensitive mutant of P. putida strain PpG1 is also extremely sensitive to UV-rays, compared to parent wild type cells. This mutant behaves typically as recombination less (recA) mutants of Escherichia coli and Pseudomonas aeruginosa, since as a recipient, it exhibits extremely low frequency of recombination following conjugational, transductional, and transformational gene transfer. Sex factor plasmids such as K-XYL or TOL can mobilize chromosomal genes equally well both from recA ⁺ and recA801 donor cells, suggesting that host recombination functions are not necessary for mobilization of chromosomal genes by such plasmids.     

Roles of Parental and Progeny DNA in Two Mechanisms of Phage S13 Recombination

THERE are at least two mechanisms for genetic recombination of phage S13–primary and secondary. They are distinguished by the amount of phage recombination observed in recombination-deficient (recA) bacterial hosts. An S13 cross performed in either of the recA hosts, Escherichia coli JC1553¹ or AB2463², yields a recombination frequency that is greatly reduced from that found in a rec⁺ host^{3, 4}; this suggests that the recA gene is required for the primary mechanism of S13 recombination. But even in a recA host the phage undergoes a small residual amount of recombination³ which has been attributed to a minor, secondary mechanism. Apparently the secondary mechanism functions in a recA cell and is only revealed when the primary mechanism is eliminated.     

Relocation of a cytoplasmic yeast linear plasmid to the nucleus is associated with circularization via nonhomologous recombination involving inverted terminal repeats

The linear plasmid pCLU1 from the yeast Kluyveromyces lactis normally replicates in the cytoplasm, with the aid of the helper linear plasmid pGKL2, using terminal protein (TP) as a primer. However, it relocates to the nucleus when selection is applied for the expression of a plasmid-borne nuclear marker. Migration to the nucleus occurred in K. lactis at a frequency of about 10⁻³/cell ten or more times higher than the rate observed in Saccharomyces cerevisiae. The nuclear plasmids existed only in a circularized form in K. lactis, while in S. cerevisiae a telomere-associated linear form is also found. Sequence analysis showed that circularization in K. lactis was caused by non-homologous recombination between the inverted terminal repeat (ITR) at the ends of the linear form and non-specific internal target sites in pCLU1. No sequence similarity existed among the junction sites, indicating that the free ITR end plays a crucial role in circularization. In S. cerevisiae, circular plasmids were generated not only by non-homologous recombination, but also by homologous recombination between short direct repeats within pCLU1. Circularization via the ITR end was observed independently of RAD52 activity. Sequences highly homologous to ARS core elements, 5′-ATTTATTGTTTT-3′ for K. lactis and 5′-(A/T)TTTAT(T/G)TTT(A/T)-3′ for S. cerevisiae, were detected at multiple sites in the nuclear forms of the plasmids. Received: 25 October 1999 / Accepted: 13 March 2000     

Repair mechanisms involved in prophage reactivation and UV reactivation of UV-irradiated phage lambda

Survival of UV-irradiated phage λ is increased when the host is lysogenic for a homologous heteroimmune prophage such as λimm434 (prophage reactivation). Survival can also be increased by UV-irradiating slightly the non-lysogenic host (UV reactivation).Experiments on prophage reactivation were aimed at evaluating, in this recombination process, the respective roles of phage and bacterial genes as well as that of the extent of homology between phage and prophage.To test whether UV reactivation was dependent upon recombination between the UV-damaged phage and cellular DNAs, lysogenic host cells were employed. Such hosts had thus as much DNA homologous to the infecting phage as can be attained. Therefore, if recombination between phage and host DNAs was involved in this repair process, it could clearly be evidenced.By using unexposed or UV-exposed host cells of the same type, prophage reactivation and UV reactivation could be compared in the same genetic background.The following results were obtained: (1) Prophage reactivation is strongly decreased in a host carrying recA mutations but quite unaffected by mutation lex-I known to prevent UV reactivation; (2) In the absence of the recA⁺ function, the red⁺ but not the int⁺ function can substitute for recA⁺ to produce prophage reactivation, although less efficiently; (3) Prophage reactivation is dependent upon the number of prophages in the cell and upon their degree of homology to the infecting phage. The presence in a recA host of two prophages either in cis (on the chromosome) or in trans (on the chromosome and on an episome) increases the efficiency of prophage reactivation; (4) Upon prophage reactivation there is a high rate of recombination between phage and prophage but no phage mutagenesis; (5) The rate of recombination between phage and prophage decreases if the host has been UV-irradiated whereas the overall efficiency of repair is increased. Under these conditions UV reactivation of the phage occurs as in a non-lysogen, as attested by the high rate of mutagenesis of the restored phage.These results demonstrate that UV reactivation is certainty not dependent upon recombination between two pre-existing DNA duplexes. The hypothesis is offered that UV reactivation involves a repair mechanism different from excision and recombination repair processes.     

Asymmetric crossing over in the spontaneous formation of large deletions in the tonB-trp region of the Escherichia coli K-12 chromosome.

The chromosomal tonB gene of Escherichia coli was used as a target for the detection of spontaneous deletion mutations. The deletions were isolated in both recA ⁺ and recA ⁻ cells, and mutants carrying large deletions were identified because they also lacked part or all of the trp operon. The frequencies of tonB-trp deletion were 1.79 × 10⁻⁹ and 1.09 × 10⁻⁹ for recA ⁺ and recA ⁻ cells, respectively. We analyzed 12 deletions from recA ⁺ and 10 from recA ⁻ cells by cloning and direct sequencing. The deletions ranged in size from 5612 bp to 15142 bp for recA ⁺ and from 5428 bp to 13289 for recA ⁻ cells. Three deletions from recA ⁺ cells and five deletions from recA ⁻ cells were found to have occurred between short sequence repeats at the termini of the deletion, leaving one copy of the repeat in the mutant sequence. Seven deletions from recA ⁺ cells and three deletions from recA ⁻ cells did not have repeats at their termini; in these cases, the DNA sequences that are adjacent to the deletion termini in the wild-type are characterized by short (2–4 bp) repeats. From these results, a model is presented for the generation of deletion mutations which involves formation of an asymmetric crossover mediated by repeated sequences of 2- to 4-bp. Received: 14 September 1998 / Accepted: 22 December 1998     

Plasmid deletion formation between short direct repeats in Bacillus subtilis is stimulated by single-stranded rolling-circle replication intermediates

Summary The effects of the rolling-circle mode of replication and the generation of single-stranded DNA (ss DNA) on plasmid deletion formation between short direct repeats in Bacillus subtilis were studied. Deletion units consisting of direct repeats (9, 18, or 27 bp) that do or do not flank inverted repeats (300 bp) were introduced into various plasmid replicons that generate different amounts of ss DNA (from 0% to 40% of the total plasmid DNA). With ss DNA-generating rolling-circle-type plasmids, deletion frequencies between the direct repeats were 3- to 13-fold higher than in plasmids not generating ss DNA. When the direct repeats flanked inverted repeats the deletion frequencies in ss DNA-generating plasmids were increased by as much as 20- to 140-fold. These results support models for deletion formation based on template-switching errors during complementary strand synthesis of rolling-circle-type plasmids. The structural instability (deletion formation between short direct repeats) of the ss DNA-generating plasmid pTA1060 in B. subtilis was very low in the presence of a functional initiation site for complementary strand synthesis (minus origin). This observation suggests that it will be possible to develop stable host-vector cloning systems for B. subtilis.     

Evidence of abortive recombination in ruv mutants of Escherichia coli K12

Summary Genetic recombination in Escherichia coli was investigated by measuring the effect of mutations in ruv and rec genes on F-prime transfer and mobilization of nonconjugative plasmids. Mutation of ruv was found to reduce the recovery of F-prime transconjugants in crosses with recB recC sbcA strains by about 30-fold and with recB recC sbcB sbcC strains by more than 300-fold. Conjugative plasmids lacking any significant homology with the chromosome were transferred normally to these ruv mutants. Mobilization of the plasmid cloning vectors pHSG415, pBR322, pACYC184 and pUC18 were reduced by 20- to 100-fold in crosses with ruv rec ⁺ sbc ⁺ strains, depending on the plasmid used. Recombinant plasmids carrying ruv ⁺ were transferred efficiently. With both F-prime transfer and F-prime cointegrate mobilization, the effect of ruv was suppressed by inactivating recA. It is proposed that the failure to recover transconjugants in ruv recA ⁺strains is due to abortive recombination and that the ruv genes define activities which function late in recombination to help convert recombination intermediates into viable products.     

Analysis of DNA repeats in bacterial plasmids reveals the potential for recurrent instability events

Structural instability has been frequently observed in natural plasmids and vectors used for protein expression or DNA vaccine development. However, there is a lack of information concerning hotspot mapping, namely, DNA repeats or sequences identical to the host genome. This led us to evaluate the abundance and distribution of direct, inverted, and tandem repeats with high recombination potential in 36 natural plasmids from ten bacterial genera, as well as in several widely used bacterial and mammalian expression vectors. In natural plasmids, we observed an overrepresentation of close direct repeats in comparison to inverted ones and a preferential location of repeats with high recombination potential in intergenic regions, suggesting a highly plastic and dynamic behavior. In plasmid vectors, we found a high density of repeats within eukaryotic promoters and non-coding sequences. As a result of this in silico analysis, we detected a spontaneous recombination between two 21-bp direct repeats present in the human cytomegalovirus early enhancer/promoter (huCMV EEP) of the pCIneo plasmid. This finding is of particular importance, as the huCMV EEP is one of the most frequently used regulatory elements in plasmid vectors. Because pDNA integration into host gDNA can have adverse consequences in terms of plasmid processing and host safety, we also mapped several regions with high probability to mediate integration into the Escherichia coli or human genomes. Like repeated regions, some of these were located in non-coding regions of the plasmids, thus being preferential targets to be removed.     

Investigation of a phage resistantSerratia entomophila strain (BC4B), establishment of generalised transduction and construction ofS. entomophila RecA mutants

ArecA clone was isolated from a cosmid library ofSerratia entomophila constructed in theEscherichia coli strain HB101. Subcloning and transposon mutagenesis were used to identify a 1.36 kb fragment containing therecA gene. A clonedrecA mutation, generated by transposon mutagenesis and the replacement of a portion of therecA gene with an antibiotic resistance cassette, was introduced into the chromosome via a marker exchange technique. TherecA strains created were deficient in DNA repair, homologous recombination and both the spontaneous and UV induction of prophages.S. entomophila recA strains showed continued pathogenicity towards the New Zealand grass grub,Costelytra zealandica. Simple procedures for further construction ofS. entomophila recA strains have been demonstrated.     

Intramolecular homologous recombination of linearized plasmids in Escherichia coli K12

Summary The efficacy of linear DNA as a substrate for general homologous recombination was demonstrated using BamHI-linearized pKLC8.5, a plasmid that carries internal direct repeats flanking the unique BamHI site. An analogous plasmid, pKLC2.31, was used in a parallel and comparative study of intramolecular homologous recombination in circular DNA substrates. When the rec ⁺ wild-type strain, AB1157, and its isogenic rec ^– derivatives were transformed with linear pKLC8.5 DNA, intramolecular homologous recombination was independent of recA, recB, recN, recO and exonuclease III (xth-1) functions. Although the recBCsbcA and recBCsbcBC cells were both very recombination proficient, only linear but not circular DNA was used as substrate for intramolecular homologous recombination in the recBCsbcA cells. In both the recBCsbcA and recBCsbcBC genetic backgrounds, the recombination frequencies for linearized pKLC8.5 DNA were 100%. A notable difference between the two strains was that none of the recBCsbcA transformants obtained with circular pKLC8.5 DNA were Tc^s recombinants, whereas 11% of the corresponding recBCsbcBC transformants were Tc^s recombinants. The sbcB mutation was responsible for the recombination proficiency of the recBCsbcBC cells. Unlike the case in recBCsbcA cells, intramolecular homologous recombination of linear DNA in the recBCsbcBC cells was dependent on recA and recF as well as recN and recO gene functions, but was independent of recJ and reeL gene functions.     

Different efficiency of UmuDC and MucAB proteins in UV light induced mutagenesis in Escherichia coli

Summary Two multicopy plasmids carrying either the umuDC or the mucAB operon were used to compare the efficiency of UmuDC and MucAB proteins in UV mutagenesis of Escherichia coli K12. It was found that in recA ⁺ uvr ⁺bacteria, plasmid pIC80, mucAB ⁺mediated UV mutagenesis more efficiently than did plasmid pSE117, umuDC ⁺. A similar result was obtained in lexA51(Def) cells, excluding the possibility that this was due to a differential regulation by LexA of the umuDC and mucAB operons. We conclude that some structural characteristic of the UmuDC and MucAB proteins determines their different efficiency in UV mutagenesis. This characteristic could be also responsible for the observation that in the recA430 mutant, pIC80 but no pSE117 can mediate UV mutagenesis. In the recA142 mutant, pIC80 also promoted UV mutagenesis more efficiently than pSE117. In this mutant, the recombination proficiency, the protease activity toward LexA and the mutation frequency were increased by the presence of adenine in the medium. In recA ⁺ uvrB5 bacteria, plasmid pSE117,umuDC caused both an increase in UV sensitivity as well as a reduction in the mutation frequency. These nagative effects resulting from the overproduction of UmuDC proteins were higher in recA142 uvrB5 than in recA ⁺ uvrB5 cells. In contrast, overproduction of MucAB proteins in excision-deficient bacteria containing pIC80 led to a large increase in the mutation frequency. We suggest that the functional differences between UmuDC and MucAB proteins might be due to their different dependence on the direct role of RecA protease in UV mutagenesis.     

DNA repair and the evolution of transformation IV. DNA damage increases transformation

Natural genetic transformation in the bacterium Bacillus subtilis provides a model system to explore the evolutionary function of sexual recombination. In the present work, we study the response of transformation to UV irradiation using donor DNAs that differ in sequence homology to the recipient's chromosome and in the mechanism of transformation. The four donor DNAs used include homologous-chromosomal-DNA, two plasmids containing a fragment of B. subtilis trp⁺ operon DNA and a plasmid with no sequence homology to the recipient cell's DNA. Transformation frequencies for these DNA molecules increase with increasing levels of DNA damage (UV radiation) to recipient cells, only if their transformation requires homologous recombination (i.e. is recA⁺-dependent). Transformation with non-homologous DNA is independent of the recipient's recombination system and transformation frequencies for it do not respond to increases in UV radiation. The transformation frequency for a selectable marker increases in response to DNA damage more dramatically when the locus is present on small, plasmid-borne, homologous fragments than if it is carried on high molecular weight chromosomal fragments. We also study the kinetics of transformation for the different donor DNAs. Different kinetics are observed for homologous transformation depending on whether the homologous locus is carried on a plasmid or on chromosomal fragments. Chromosomal DNA- and non-homologous-plasmid-DNA-mediated transformation is complete (maximal) within several minutes, while transformation with a plasmid containing homologous DNA is still occurring after an hour. The results indicate that DNA damage directly increases rates of homologous recombination and transformation in B. subtilis. The relevance of these results and recent results of other labs to the evolution of transformation are discussed.