Search

Homologous recombination is a crucial process for the maintenance of genome integrity. The two main recombination pathways in Escherichia coli (RecBCD and RecF) differ in the initiation of recombination. The RecBCD enzyme is the only component of the RecBCD pathway which acts in the initiation of recombination, and possesses all biochemical activities (helicase, 5'-3' exonuclease, χ cutting and loading of the RecA protein onto single-stranded (ss) DNA) needed for the processing of double stranded (ds) DNA breaks (DSB). When the nuclease and RecA loading activities of the RecBCD enzyme are inactivated, the proteins of the RecF recombination machinery, i.e., RecJ and RecFOR substitute for the missing 5'-3' exonuclease and RecA loading activity respectively. The above mentioned activities of the RecBCD enzyme are regulated by an octameric sequence known as the χ site (5'-GCTGGTGG-3'). One class of recC mutations, designated recC*, leads to reduced χ cutting in vitro. The recC1004 strain (a member of the recC* mutant class) is recombination proficient and resistant to UV radiation. In this paper, we studied the effects of mutations in RecF pathway genes on DNA repair (after UV and γ radiation) and on conjugational recombination in recC1004 and recC1004 recD backgrounds. We found that DNA repair after UV and γ radiation in the recC1004 and recC1004 recD backgrounds depends on recFOR and recJ gene products. We also showed that the recC1004 mutant has reduced survival after γ radiation. This phenotype is suppressed by the recD mutation which abolishes the RecBCD dependent nuclease activity. Finally, the genetic requirements for conjugational recombination differ from those for DNA repair. Conjugational recombination in recC1004 recD mutants is dependent on the recJ gene product. Our results emphasize the importance of the canonical χ recognition activity in DSB repair and the significance of interchange between the components of two recombination machineries in achieving efficient DNA repair., (Copyright © 2012 Elsevier Masson SAS. All rights reserved.)

In Escherichia coli, homologous recombination initiated at double-stranded DNA breaks requires the RecBCD enzyme, a multifunctional heterotrimeric complex that possesses processive helicase and exonuclease activities. Upon encountering the DNA regulatory sequence, chi, the enzymatic properties of RecBCD enzyme are altered. Its helicase activity is reduced, the 3'-->5'nuclease activity is attenuated, the 5'-->3' nuclease activity is up-regulated, and it manifests an ability to load RecA protein onto single-stranded DNA. The net result of these changes is the production of a highly recombinogenic structure known as the presynaptic filament. Previously, we found that the recC1004 mutation alters chi-recognition so that this mutant enzyme recognizes an altered chi sequence, chi*, which comprises seven of the original nucleotides in chi, plus four novel nucleotides. Although some consequences of this mutant enzyme-mutant chi interaction could be detected in vivo and in vitro, stimulation of recombination in vivo could not. To resolve this seemingly contradictory observation, we examined the behavior of a RecA mutant, RecA(730), that displays enhanced biochemical activity in vitro and possesses suppressor function in vivo. We show that the recombination deficiency of the RecBC(1004)D-chi* interaction can be overcome by the enhanced ability of RecA(730) to assemble on single-stranded DNA in vitro and in vivo. These data are consistent with findings showing that the loading of RecA protein by RecBCD is necessary in vivo, and they show that RecA proteins with enhanced single-stranded DNA-binding capacity can partially bypass the need for RecBCD-mediated loading.

To study roles of Rep helicase in short-homology-dependent illegitimate recombination, we examined the effect of a rep mutation on illegitimate recombination and found that the frequency of spontaneous illegitimate recombination is enhanced by the rep mutation. In addition, illegitimate recombination was synergistically enhanced by the rep mutation and UV irradiation, showing that Rep helicase plays a role in suppression of spontaneous as well as UV-induced illegitimate recombination. The defect in RecQ helicase also has a synergistic effect on the increased illegitimate recombination in the rep mutant. It was also found that the illegitimate recombination induced by the rep mutation is independent of the RecA function with or without UV irradiation. Nucleotide sequence analyses of the recombination junctions showed that the illegitimate recombination induced by the rep mutation mostly takes place between short homologous sequences. Based on the fact that the defect of Rep helicase induces replication arrest during replication, resulting in the formation of DNA double-strand breaks, we propose a model for illegitimate recombination, in which double-strand breaks caused by defect of Rep helicase promotes illegitimate recombination via short-homology-dependent-end-joining. In addition, the mechanism of synergistic action between the rep mutation and UV irradiation on illegitimate recombination is discussed.

To study the involvement of DNA replication in UV-induced illegitimate recombination, we examined the effect of temperature-sensitive dnaB mutations on illegitimate recombination and found that the frequency of illegitimate recombination was reduced by an elongation-deficient mutation, dnaB14, but not by an initiation-deficient mutation, dnaB252. This result indicates that DNA replication is required for UV-induced illegitimate recombination. In addition, the dnaB14 mutation also affected spontaneous or UV-induced illegitimate recombination enhanced by the recQ mutation. Nucleotide sequence analyses of the recombination junctions showed that DnaB-mediated illegitimate recombination is short homology dependent. Previously, Michel et al. (B. Michel, S. Ehrlich, and M. Uzest, EMBO J. 16:430--438, 1997) showed that thermal treatment of the temperature-sensitive dnaB8 mutant induces double-stranded breaks, implying that induction of illegitimate recombination occurs. To explain the discrepancy between the observations, we propose a model for DnaB function, in which the dnaB mutations may exhibit two types of responses, early and late responses, for double-stranded break formation. In the early response, replication forks stall at damaged DNA, resulting in the formation of double-stranded breaks, and the dnaB14 mutation reduces the double-stranded breaks shortly after temperature shift-up. On the other hand, in the late response, the arrested replication forks mediated by the dnaB8 mutation may induce double-stranded breaks after prolonged incubation.

RecBCD has two conflicting roles in Escherichia coli. (i) As ExoV, it is a potent double-stranded (ds)DNA exonuclease that destroys linear DNA produced by restriction of foreign DNA. (ii) As a recombinase, it promotes repair of dsDNA breaks and genetic recombination in the vicinity of chi recombination hot-spots. These paradoxical roles are accommodated by chi-dependent attenuation of RecBCD exonuclease activity and concomitant conversion of the enzyme to a recombinase. To challenge the proposal that chi converts RecBCD from a destructive exonuclease to a recombinogenic helicase, we mutated the nuclease catalytic centre of RecB and tested the resulting mutants for genetic recombination and DNA repair in vivo. We predicted that, if nuclease activity inhibits recombination and helicase activity is sufficient for recombination, the mutants would be constitutive recombinases, as has been seen in recD null mutants. Conversely, if nuclease activity is required, the mutants would be recombination deficient. Our results indicate that 5' --> 3' exonuclease activity is essential for recombination by RecBCD at chi recombination hot-spots and at dsDNA ends in recD mutants. In the absence of RecB-dependent nuclease function, recombination becomes entirely dependent on the 5' --> 3' single-stranded (ss)DNA exonuclease activity of RecJ and the helicase activity of RecBC(D).

The present study was conducted to clarify the lethality of Auger cascades induced by the K-shell photoabsorption of phosphorus in Escherichia coli. Killing of wild-type and radiation-sensitive mutants of E. coli was examined. Three x-ray energies were chosen for irradiation; at 2.153 keV: the resonance peak of K-shell photoabsorption of phosphorus; at 2.146 and 2.160 keV: off-peak. Enhancement ratio, which was defined as the ratio of the killing sensitivity of 2.153 keV to that at 2.146 keV, were 1.32 to 1.54 for examined strains. Increment of absorbed energy calculated in entire cells for 2.153 keV radiation could not explain the degree of observed enhancement of killing. Lethality of Auger cascades depended on the killing sensitivity with x-rays which did not induce Auger cascades. The lethality for wild-type was lower than that for recombination repair-deficient mutants. It was concluded that one part of damages produced by Auger cascades was repaired in wild-type strains.

The effect of the Escherichia coli RecA protein on mitotic recombination in the diploid D7 strain of Saccharomyces cerevisiae damaged by UV radiation was investigated. The D7 strain was transformed by two modified versions of the pNF2 plasmid: one, containing the ADH-1 promoter, and the other containing the recA gene tandemly arranged behind the ADH-1 promoter region. Immunological analysis proved the presence of the 38-kDa RecA protein in D7/pNF2ADHrecA transformants. We observed a positive effect of recA gene expression on mitotic gene conversion, mainly at higher doses of UV radiation. The results indicate that a RecA-like activity could participate in steps preceding mitotic conversion events in yeast.

Nucleotide sequences called Chi (5'-GCTGGTGG-3') enhance homologous recombination near their location by the RecBCD enzyme in Escherichia coli (Chi activation). A partial inhibition of Chi activation measured in lambda red gam mutant crosses was observed after treatment of wild-type cells with DNA-damaging agents including UV, mitomycin, and nalidixic acid. Inhibition of Chi activation was not accompanied by an overall decrease of recombination. A lexA3 mutation which blocks induction of the SOS system prevented the inhibition of Chi activation, indicating that an SOS function could be responsible for the inhibition. Overproduction of the RecD subunit of the RecBCD enzyme from a multicopy plasmid carrying the recD gene prevented the induced inhibition of Chi activation, whereas overproduction of RecB or RecC subunits did not. It is proposed that in SOS-induced cells the RecBCD enzyme is modified into a Chi-independent recombination enzyme, with the RecD subunit being the regulatory switch key.

Genetic recombination of nonreplicating phage lambda-DNA, during infection of homoimmune lysogenic bacteria, was previously observed to be dramatically stimulated by prior uv irradiation of the phages, even when the Escherichia coli hosts lacked the major uv-photo-product excision-repair system (UvrABC). UvrABC-independent recombination of circular phage molecules depends on host MutHLS functions and on undermethylation of adenines at GATC sites in the phage DNA, and thus appears to be the result of "mismatch repair" of uv photoproducts. Recombinant frequencies pass through a relatively sharp maximum at 20 J/m2 and decrease at higher doses, whereas most plausible models for the process predict monotonic increases with dose, or a plateau at high uv doses. A uv-dose-dependent loss of biological activity (restriction) of all intracellular phage DNA was also observed previously. In order to provide a framework for testing possible explanations for the unusual recombinant-frequency vs uv-dose curve, a statistical model was constructed. This model includes probability terms for all possible one-exchange and two-exchange recombination processes, and incorporates the assumption that dimer recombinants are more susceptible to restriction than monomer parents (or recombinants), because of their larger target size. By adjustment of model parameters, particularly epsilon, the efficiency per photoproduct of initiation of a recombinational exchange, a theoretical dose-response curve that agreed well with experiment was obtained. The best fit corresponded to epsilon = 0.035, close to the previously observed restriction efficiency of 0.053. In the calculations, the value for h0, the average length of heteroduplex DNA, was taken to be 0.5 lambda units, i.e., about 25 kilobase pairs. This estimate for h0 was obtained here by analysis of the density distributions of the progeny of crosses between nonreplicating density-labeled lambda-phage chromosomes, published by others [M. S. Fox, C. S. Dudney and E. J. Sodergren (1979) Cold Spring Harbor Symposium on Quantitative Biology, Vo. 43, pp. 999-1007].

Direct and indirect UV-stimulated homologous genetic recombination was investigated in Escherichia coli strains blocked in several host-encoded functions. Genetic recombination was assayed by measuring beta-galactosidase produced after recombination between two noncomplementing lacZ ochre alleles. Both types of stimulation (direct and indirect) were found to be primarily RecF pathway-mediated. In a rec+ background, both direct and indirect stimulation were found to be dependent on uvrD (coding for helicase II). In a recB21 sbcB15 background, direct and indirect stimulation were uvrD dependent only when the strain was additionally deficient in the UvrABC excision repair pathway. Indirect but not direct stimulation was also dependent on recJ (coding for a 5'-to-3' exonuclease specific for single-stranded DNA) regardless of sbcA or sbcB configuration. The methyl-directed mismatch repair system (mutSLH) also appeared to play an important role in stimulation. On the basis of these findings, we suggest that excision of UV-induced DNA damage is a prelude to UV-mediated stimulation of genetic recombination.

Plasmid DNA was used to study gamma-radiation-induced recombination and mutagenesis in Escherichia coli host cells. Plasmid pBRP1, a derivative of pBR322 containing the lac operon of E. coli, was irradiated with 60Co gamma rays prior to transformation into E. coli strains of different recA and lac genotypes. Plasmid-chromosome recombination was assayed in lacY1 host cells, whereas plasmid mutagenesis was assayed in delta lac host cells lacking chromosomal sequences homologous to the plasmid. Both recombinant and mutant plasmids were identified by the phenotypic changes in lactose utilization, and confirmed by restriction analysis of isolated plasmids. Plasmid-chromosome recombination was induced to high levels (about 20% of survivors at 700 Gy) and was dependent on the host recA gene. Plasmid mutagenesis occurred at lower levels (about 1.5% of survivors at 600 Gy) and was relatively independent of the recA gene. Plasmid survival was unaffected by the presence or absence of host recA mutations or the potential for plasmid-chromosome recombination.

Plasmid DNA substrates were used to study ultraviolet (UV)-induced recombination events in Escherichia coli host cells. Plasmids derived from pBR322, containing all or part of the lac operon of E. coli, were irradiated with ultraviolet light before transformation into E. coli strains of different recA and lacY genotypes. Recombinational exchanges were identified by phenotypic changes in lactose utilization and were confirmed by restriction analysis of isolated plasmids. Ultraviolet-induced reciprocal plasmid-chromosome recombination occurred at a slightly higher frequency then non-reciprocal chromosome-to-plasmid recombination, and at a much higher frequency than non-reciprocal plasmid-to-chromosome recombination. These frequencies did not depend on segregative mechanisms. The asymmetry of non-reciprocal exchange was not due to the particular arrangement of wild-type and lacY1 alleles because the same results were observed when these were interchanged. The host recA gene was required for plasmid-chromosome recombination, and slightly enhanced plasmid survival. Evidence for plasmid replication prior to recombination was found in reciprocal recombinants, but rarely in the non-reciprocal recombinants analyzed. Irradiation of competent bacterial host cells prior to transformation did not effectively induce plasmid-chromosome recombination.

In the radiation-resistant Gamr444 mutant the inheritance frequency of long F' episomes ORF1 (purE+ tsx+ procC+ lac+) and F'14 (ilv+--argE+) is lower, and the frequencies of chromosome mobilization and integrative suppression of temperature-sensitive dnaA46 mutation by the sex factor F are much higher than those in the wild-type strain AB1157 and another radiation-resistant mutant Gamr445. In this respect, the mutant Gamr444 is very similar to the recRC sbcB mutant (RecF-pathway of recombination).

The question of whether induction of the SOS response in Escherichia coli increases the efficiency of excision repair was addressed by measuring repair of UV-damaged nonreplicating lambda phage DNA in previously irradiated bacteria. Prior UV irradiation of lex+ bacteria enhanced both the rate of regeneration of infective phage DNA (about 10-fold) and the rate of cyclobutane dimer removal early in repressed infections. Indirect induction of SOS-regulated repair activities by the nonreplicating irradiated phage DNA itself seemed negligible. Prior bacterial irradiation reduced the frequency of recombination (loss of a tandem chromosomal duplication) of nonreplicating UV-irradiated DNA. In this respect UV-stimulated recombination of nonreplicating DNA differs from RecF-dependent recombination processes that are stimulated by increased SOS expression. Surprisingly, prior UV irradiation of lexA3 bacteria caused a small but reproducible increase in the regeneration of infective phage DNA.

Genetically disabled lambda plac5 transducing phage derivatives were used to study the recB dependence of recombination during specialized transduction. The frequency of transduction was normalized to colony-forming units, and the end product of recombination was monitored by scoring for addition and substitution transductants. When a chromosomal lac gene was the recipient DNA substrate molecule, both the normalized transduction frequency and the proportion of addition and substitution transductants showed essentially no recB dependence. There was a pronounced recB dependence for both normalized transduction frequency and recombination end product formation when F42 lac was the recipient DNA substrate. recB appears to have no significant role in the recombination that occurs between the two lac regions in an addition transductant. UV irradiation of the transducing phages increased the absolute level of both addition and substitution transductants obtained with a chromosomal lac gene but resulted in a considerable change in the relative frequency of addition versus substitution transductants.

It was shown that the presence of colicinogenis plasmid ColIb-P9 increased the survival of UV-irradiated bacteriophage lambda cI857 in non-irradiated cells of Escherichia coli K-12. The effect of this plasmid was retained in the polA and recB mutants, being sharply reduced in the uvrA and recB recC sbcB recF mutants. This effect strongly depended on recA+ and lexA+ genotype. The W-reactivation efficiency was slightly higher in the cells containing ColIb-P9 than in those lacking the plasmid. No significant effect of the plasmid on recombination during transduction, after conjugation under usual conditions and in the case when a conjugation mixture or recipient cells were irradiated, was observed. The data demonstrate that the effect of ColIb-P9 plasmid on DNA repair is not mediated by its influence on recombination.

Three aspects of recombination of UV-irradiated nonreplicating lambda phage DNA were addressed: the photoproduct(s) responsible, the role of UvrABC-mediated excision repair, and the dependence on RecF function. Cyclobutane pyrimidine dimers appeared responsible for some recombination because photoreactivation reduced the frequency of 254-nm-stimulated recombination and because photosensitized 313-nm irradiation stimulated recombination. Other photoproducts seemed recombinogenic as well, because high fluences of 254-nm irradiation stimulated recombination considerably more, per cyclobutane dimer induced, than photosensitized 313-nm irradiation, and because photoreactivation did not eliminate 254-nm stimulated recombination. For both treatments, much, but not all, of the recombination was UvrABC-dependent. Recombination was mostly RecF-dependent, but was not affected by recB recC or recE mutations

A mutation in the recF gene of Escherichia coli results in a radiation-sensitive strain. The RecF pathway and the RecBC pathway account for nearly all of the conjugative recombination occurring in E. coli. recBC cells are radiation-sensitive and carry only out a small amount of recombination but these deficiencies are suppressed by an sbcB as recombination is shunted to the RecF pathway. A recBC sbcB recF strain is very radiation-sensitive and is devoid of recombination ability. These deficiencies are suppressed by the srfA mutation; srfA is a recA allele. UV-induced respiration shutoff is a recA+, lexA+ and recBC+ dependent. We report in this paper that respiration does not shutoff in a recF strain at 37 and 30 degrees C. an srfA mutation suppresses this lack of respiration shutoff effect in a recF srfA mutant at 30 degrees C but not at 37 degrees C; no suppression by this mutation occurs at either temperature in a recF recBC sbcB strain. An srfA strain also does not shut off its respiration at 37 degrees C and shows a temperature conditional UV-induced respiration shutoff response at 30 degrees C. The srfA mutation is thought to cause an altered RecA protein to be produced and we suggest that at 37 degrees This altered protein is temperature sensitive. We conclude from the results in this paper that the recF gene product is required for UV-induced respiration shutoff and that the RecA protein plays a special role in the induction process.

The recovery of phage P1 mediated transductants varies with the marker selected in a manner which cannot be fully accounted for by dosage differences in the donor gene population. This variation in transduction frequency is due primarily to recombinational discrimination in the recipient cell. We show here that increasing the intracellular level of recA protein, which might be expected to increase the contribution of recF mediated events to recombinant formation, decreases this discrimination slightly, and that replacing recBC mediated recombination by a recF dependent process, augmented by an additional, as yet uncharacterized mutation, dramatically reduces recombinational discrimination. We conclude that although recBC mediated transductional recombination is selective, recombination which relies on recF need not be so. We also show that UV-damaged DNA can be successfully recombined in the absence of the recB product (even in sbcB+ cells) and that eliminating exonuclease I (the sbcB product) facilitates the recombination of heavily irradiated DNA.

The efficiency of recovery of P1 transductants is marker dependent and normally varies over a 25-fold range. UV irradiation of either transducing lysates for recipient cells results in a selective stimulation of the transduction of markers which are normally transduced poorly. As a result the range in frequency of transduction is reduced to about 3-fold and resembles the gene frequency distribution expected in the donor cells. We conclude that P1 transducing lysates are likely to contain a random sample of donor DNA but that the recombination system of the recipient cell exhibits a preference for the DNA of some regions over that of others. Damage to DNA presumably overrides this specificity.

An insertion of transposon Tn10 was isolated in a gene essential for the RecF pathway of recombination. This mutation and six other previously isolated rec mutations, including one in the recJ gene, were mapped to 62 min on the Escherichia coli chromosome. All seven mutations were examined for their effects on recombination and UV survival in several genetic backgrounds. Although all affected the RecF, and not the RecBC, pathway of recombination, they could be separated into two classes: four severely reduced UV survival and three did not. In addition, whereas the effects of the former class appeared to be nearly equivalent in recB recC sbcA and recB recC sbcB mutant backgrounds, the latter class affected recombination differentially. The latter class was found to be either "leaky" or temperature-sensitive mutations. Transductional complementation assays indicated that representative members of each class were mutations in the same gene, recJ. Possible explanations for the differences in phenotypic effects produced by these recJ mutations are discussed.

From Escherichia coli cells we purified '8'-shaped dimeric molecules in which two circular DNA molecules of bacteriophage lambda were joined at a homologous site. Some of them had a complex junction which we interpreted as being two closely spaced Holliday structures because of (i) superhelicity of the molecule, (ii) the sedimentation rate of the molecule in sucrose gradients, and (iii) the appearance in the electron microscope. Other 'figure-eights's' had two separate homologous junctions, presumably two Holliday bridges. A possible role for these 'double Holliday structures' in UV-stimulated recA-dependent recombination in vivo is discussed.

Lethal action of gamma-rays on derivatives of the wild-type strain AB1157 and of two radiation-resistant mutants (Gamr444 and Gamr445) containing additional mutations dnaA46, recB21, recF143, recA56, recA430, lexA3, lexA102 or lexA3 recAo98, was studied. When the mean number of genomes per cell was reduced by means of pre-incubation at 43 degrees C, radioresistance of the strains AB1157 dnaA46 and Gamr445 dnaA46 was not changed, and that of the strain Gamr444 dnaA46 was reduced to the level of the Gamr445 dnaA46 strain. Introduction of additional mutations recB21, recA56 or lexA3 (lexA102) into the genome of the strains Gamr444 or Gamr445 made them as radiosensitive as the corresponding variants of AB1157. Additional mutations recF143 or recA430 (lexB30) significantly decreased the radioresistance of Gamr444 and Gamr445 mutants, although did not level them to corresponding derivatives of AB1157. Operator-constitutive mutation recAo98 enhanced radioresistance of all lexA3 derivatives tested but not to the level of the corresponding lexA+ strains. The role of recombinational repair and the inducible SOS system in enhanced radioresistance of Gamr mutants is discussed. The data of post-irradiation DNA degradation in various derivatives of the strains AB1157 and Gamr suggest that Gamr mutants have a constitutive inhibitor of degradation which does coincide with RecA protein.

F-like plasmids (FBl, FBl drd, F-lac+) were transferred to the recipient strains E. coli, K-12 with different recombination ability. Then the sensitivity of these strains to ultra-violet irradiation and also the stability of the mentioned plasmids under conditions of bacteria treatment with elimination agents was determined. The presence of any F-like plasmids under study failed to influence the ultra-violet sensitivity of the bacterial cells. On the other hand, the research results of spontaneous and inducec elimination of these plasmids indicated considerable differences in relation to the elimination agents. It is supposed that the stability of individual F-like plasmids in the cells largely depended on the genetic differences in rec A gene in the bacterial cells containing them.

Lethal, mutagenic and recombonogenic action of the solar radiation on the model microorganisms--phage T4, bacteria Escherichia coli and ascomycet Aspergillus nidulans--has been studied. A considerable lethal effect of the solar radiation on phage T4 and E. coli was found. An increasing of mutation frequency in E. coli and A. nidulans by sunlight was also revealed. Recombinogenic action of solar radiation has been demonstrated in the experiments with diploid A. nidulans strains. It was shown that the excision and postreplication repair systems took part in recovery of damages induced by sunlight. An important role of ultra-violet region (280-320 nm) solar radiation in induction of lethal and mutagenic effects was demonstrated for all investigated microorganisms.

It has been shown previously that the radiation sensitivity of LexA strains of Escherichia coli K-12 can be suppressed by thermosensitive mutations (designated tsl) that are closely linked to the lexA locus. These are thought to be intragenic suppressors that reduce the activity of the diffusible product that gives rise to the LexA- phenotype (Mount et al., 1973). When a recA mutation is crossed into a suppressed tsl- strain, the extreme radiation sensitivity usually conferred by a recA mutation is considerably reduced without any detectable change in genetic recombination deficiency. Suppression of UV sensitivity depends upon the activity of the uvrA+ product. We propose that at least part of the radiation sensitivity of a recA- strain is due to a DNA repair defect that is different from inability to perform genetic exchanges and depends upon the presence of the lexA+ product. We hypothesize that the lexA+ product is a repressor of the synthesis of repair enzymes. In recA+ cells with DNA lesions, repressor is inactivated leading to enzyme induction but this does not occur in recA- cells. tsl mutations inactivate repressor leading to constitute enzyme synthesis and bypassing the need for recA+ product to inactivate the lexA+ product.

The available data concerning postreplication repair are summarized. In Escherichia coli, recombination is implicated in this repair because the recA+ gene is necessary and because strand exchanges occur that extend over long regions. Other experiments involving phage-induced resistance also point to an interrelation between recombination and repair. In this phenomenon, gene products of lambda bacteriophage are introduced into bacteria, resulting in an increased resistance of the cells when they are subsequently exposed to X rays.