Cloning, sequence and regulation of expression of the 1exA gene of Aeromonas hydrophila

Gene, 154 (1995) 71-75 ©1995 Elsevier Science B.V. All rights reserved. 0378-1119/95/$09.50

71

GENE 08625

Cloning, sequence and regulation of expression of the lexA gene of Aeromonas hydrophila (Recombinant DNA; repressor; RecA; SOS box)

J o a n R i e r a a n d J o r d i t];arb6 Department of Genetics and Microbiology, Autonomous University of Barcelona, Bellaterra, 08193 Barcelona, Spain Received by H.M. Krisch: 4 July 1994; Revised/Accepted: 20 September/3 October 1994; Received at publishers: 18 November 1994

SUMMARY

The lexA gene of Aeromonas hydrophila (Ah) has been isolated by using a specific one-step cloning system. The Ah LexA repressor is able to block Escherichia coli (Ec) SOS gene expression and is likely to be cleaved by the activated RecA protein of this bacterial species after DNA damage. Ah lexA would encode a protein of 207 amino acids (aa), which is 75% identical to the LexA repressor of Ec. Two Ec-like SOS boxes have been located upstream from Ah lexA, the distance between them being 4 bp, whereas this same distance in Ec lexA is 5 bp. The structure and sequence of the DNA-binding domain of ~LheLexA repressor of Ec, as well as the region at which its hydrolysis occurs, are highly conserved in Ah LexA. Moreover, a residue of the region implicated in the specific cleavage reaction, and which is present in all known RecA-cleavable repressors, is changed in the Ah LexA. Expression of Ah lexA is DNA-damage inducible in both the Ah and Ec genetic backgrounds to the same extent. In contrast, Ec lexA is poorly induced in DNA-injured Ah cells.

INTRODUCTION

Exposure of Escherichia coli (Ec) to agents that threaten the structure or synthesis of DNA results in the induction of a diverse se! of DNA repair and cellular survival functions, designated as the SOS response (Walker, 1984). Induction of the SOS response is triggered by a metabolic signal that activates the RecA protein to cause proteolytic cleavage of the LexA protein, Correspondence to: Dr. J. Barb6, Department of Genetics and Microbiology, Faculty of Sciences, Autonomous University of Barcelona, Bellaterra, 08193 Barcelona, Spain. Tel. (34-3) 581-1837; Fax (34-3) 581-2387; e-mail: [email protected] Abbreviations: aa, amino acid(s); Ah, Aeromonas hydrophila; 13Gal, 13-galactosidase; bp, base pair(s); Def, defective; Ec, Escherichia coli; kb, kilobase(s) or 1000 bp; LexA, repressor of SOS genes; lexA, gene encoding LexA; nt, nucleotide(s); RecA, coprotease of LexA repressor; recA, gene encoding RecA; SD, Shine-Dalgarno sequence; SOS, error-prone DNA repair system; XGal, 5-bromo-4-chloro-3-indolyl-13-o-galactoside; A, deletion; ::, novel junction (fusion or insertion).

SSDI 0378-1119(94)00836-1l

the repressor of at least 28 SOS genes (Little, 1991). Activated RecA protein has apoprotease activity which facilitates autocatalytic clavage of the LexA repressor (Little, 1991). The existence of similar DNA damage-inducible responses in other bacterial species has been widely reported (Miller and Kokjohn, 1990). By either interspecies functional complementation of Ec recA mutants or by using DNA probes, the recA genes of more than 50 species of both Gram- and Gram + bacteria have been isolated and sequenced. The identification and the analysis of the variability of the several functional domains of the RecA protein have been possible through the availability of these recA genes (Miller and Kokjohn, 1990). Nevertheless, the bacterial LexA repressor has been poorly analyzed from this comparative point of view. The reason for this could be attributed to the difficulty in cloning the lexA gene, since it produces a repressor whose overproduction sensitizes the Ec cells to DNA damage. A system to directly isolate lexA genes of Gram-

72 bacteria has been developed in our laboratory (Calero et al., 1991). By this method, the lexA genes of the enterobacteria Salmonella typhimurium, Erwinia carotovora and Providencia rettgeri, and those from the pseudomonad Pseudomonas aeruginosa and Pseudomonas putida have been isolated and sequenced (Garriga et al., 1992; Riera and Barb6, 1993). In this context, we have embarked upon a study of both the variability and the regulation of the expression of the lexA genes from several G r a m bacteria. For this reason we have approached the cloning and sequencing of the lexA gene of the animal and human pathogen Aeromonas hydrophila (Ah) This bacterium belongs to the Vibrionaceae family, from which no gene has been shown to be DNA damage-inducible so far.

EXPERIMENTALAND DISCUSSION (a) Cloning strategy of A. hydrophila (Ah) lexA Chromosomal DNA from Ah was partially digested with Sau3AI, and fragments in the range of 5 to 10 kb were purified and ligated to plasmid pUA165 (Calero et al., 1991), which had previously been digested with BamHI. The ligation products were transformed by electroporation in strain UA4794 (lexA (Def) sulAr e c A - ). Two kinds of clones were detected: one of them presented a high basal level of expression of the suIA::lacZ fusion contained in strain UA4794, whereas the second showed a lower expression of this fusion when plated on XGal plates. Plasmid DNA isolation and restriction analysis of both kinds of colonies showed that the first type was a spontaneous deletion derivative of plasmid pUA165 lacking a fragment of the sulA gene. The second kind contained a pUA165 plasmid derivative carrying a heterologous fragment of DNA. One of these plasmids containing a 5.5-kb fragment was designated pUA403 and used for further experiments. pUA403 was transformed in both the UA4793 (lexA71::Tn5) and JL2301 (AlexA300) strains of Ec to analyze their effects on the basal level of suIA expression in two different lexA(Def) recA ÷ backgrounds. The suIA::lacZ expression was practically the same in UA4793 and JL2301 cells containing pUA403 (223 and 145 units of 13Gal, respectively) or the same plasmid vector carrying the Ec lexA gene (215 and 230 units of [3Gal, respectively). Mitomycin-C-treated cultures of the UA4793 strain containing the pUA403 plasmid showed an increase in the expression of the suIA::lacZ fusion (Fig. 1A), indicating that the Ah LexA repressor may be hydrolyzed in recA ÷ cells of Ec. It is known that in Ec the presence of a multicopy plasmid harboring lexA enhances the sensitivity of the Ec cells to DNA damage as a consequence of stronger

(B)

lOO

10

i

0 Time (mm)

5

10

15

UV dose (J / m 2 )

Fig. 1. Effect of the Ah lexA gene on DNA-damaged Ec cells. (A) Induction of a sulA::lacZ fusion in Ec UA4793 (lexA71::Tn5 recA+) carrying the lexA gene of either Ah (•) or Ec (A) after addition of 20 gg mitomycinC/ml. The induction factor is the ratio between units of 13Galof both mitomycin-C-treatedand untreated cells. (B) Survival of EcUA4793 (lexA71::Tn5 recA+) carrying the lexA gene of Ah (•), Ec (A), or neither (O) after UV irradiation at differentdoses. All data were reproducibleto within an error of __+10%. repression of SOS genes, such as uvrABD, which directly participate in DNA repair (Sedgwick and Goodwin, 1985). Thus, plasmid pUA403 causes UV sensitization of the UA4793 strain (Fig. 1B).

(b)Nucleotide sequence of A. hydrophila (Ah) lexA By digestion with different restriction enzymes and complementation experiments, the Ah lexA gene was located on a 1.4-kb XbaI-SalI fragment of plasmid pUA403. The nt sequence of this fragment was determined b y the chain termination procedure (Sanger et al., 1977) after a set of nested deletions was created by using the Erase-a-Base kit (Promega, Madison, WI, USA). The Ah lexA consisted of 624 bp, including the stop codon, encoding a protein of 207 aa (Fig. 2). This is the longest lexA gene known. This increase in its length is due to the insertion of 15 bp at nt 221 in comparison with the sequence of Ec. This insertion is located in the hinge region which links two functional domains of the LexA repressor (Schnarr et al., 1991). There is a probable SD beginning at nt 377. The use of FastA to search the databases revealed that LexA of Ah shares a 75% identity over a 202-aa overlap with Ec LexA, a 73% identity over a 205-aa overlap with P. rettgeri LexA and a 67% identity with P. aeruginosa LexA over a 204-aa overlap. All G r a m - lexA genes sequenced so far have two tandem consensus LexA-binding sites upstream from the lexA ORF which are used for autocontrol of LexA protein levels; Ah lexA also presented two such binding sites. Nevertheless, the distance between them is 4bp, this value being of 5 bp or 3 bp for either enterobacterial or pseudomonadad lexA genes described (Garriga et al., 1992). The first SOS box of Ah lexA differs by two residues from that of Ec, whereas the second presents four differences in comparison with that of Ec (Fig. 3A). These data are in agreement with the hypothesis that the first SOS box is the most important one in the process of the

73 A ~ATCCTCGC~GCACTGTTG~AGGGTCATC~GGTCGGTGA~CGAGCTGGT~TTCAGGGCG~ AGACGATGGGTCTGG£CGGATCCAGCTTGAGCTCGGTGATGGGATTTTCCGGCAGGCTCT TGTGATTGACCAAGCCGCTGACCGGCCATTGCAA6ATGGCTCTGGAAATGTTCTGTCCAA GGGACATGTGCGTCTGCTCTCCAAAAAAATGCGGCGTGAGCATAGCAAATCACCGACTGC TTGTCGCCGAGAGGTACGATGAGTCGCTGCGCGGGATCAAATCTGCAGCAAATGAGTGCT GGGCCGCGACAAATGTCAGTGCGCCTTGCCTTGGCCACTGCACTGTATATACTGCCAGTG

60 120

-35 Ec

T GCACTTTATGGTTCCAAAATCGCCTTTTG

2~ 300 3~

Ah

TGTCAGTGCGC CTTC-CCTTG-GCCACTGCA

TGCTGTATAAAAAGA(~[~'~'~CCATGCTATGAAACCCCTTACTCCCCGCCAGGCCGAAG M K P L T P R Q A E V

420

EC

C T G T A T A T A C T C A C A C ,C A T A A C T G T A T A T A C A C C C A G G G G G C G - - - G A A T G

Ah

CTGTATATACTGCCAGTGTG-CTGTATAAAAAG&CAGGTGCCCATGCTATG

TGCTGGAGTTGATCAAGGCGAACATGAACGAAACIGGCATGCCGCCGACTCGCGCCGAGA L E t I K A N M N E T G M P P T R A E I

480

TAGCCCAGAAGCTTGGTTTCAAGTCGGCCAATGC~GCCGAGGAACATCTGAAGGCCCTGG A Q K L G F K S A N A A E E H L K A L A

540 51

CCAAAAAGGGGGTCATCGAGATCATGCCGGGCACGTCCCGCGGCATCCGCCTGTTGCTCG K K G V I E I M P G T S R G I R L L L E

600 71

AAGAG~AGAGCCGCTG~AGAGAGCGGTCTGCCGCTGATCGGCAAGGTGGCCGCCGGCG E E E P L E E S G L P L I G K V A A G E

660 91

AGTCCATCCTGGCGCAGGAGCATATCGAGAGTCACTACCAGGTGGATCCGGCCCTGTTCC S I L A Q E H I E S H Y Q V D P A L F H

720 IIi

ACCCGCGGGCCGACTTCCTGCTGCGGGTACAGGGCATGAGCATGAAGAACATCGGCATTC P R A D F L L R V Q G M S M K N I G I L

780 131

TG~TGGCGATCTGCTGGCGGTACACAAGACCCAGGAGGTGCGCAACGGCCAGGTGGTGG D G D L L A V H K T Q E V R N G Q V V V

840 151

Ah

TGGCACGGCTCGACGAGGATGTCACCGTCAAACGCTTCCAGCGCAAGGGCAGCCAGGTGT A R L D E D V T V K R F Q R K G S Q V W

~0

Region

GGCTGCTGCCGGAGAACGAAGAGCTCTCCCCCATCGAGGTGGATCTCTCCTGCCAGCAGC L L P E N E E L S P I E V D L S C Q Q L

960 191

TCACCATCGAGGGGCTGGCGGTCGGCGTCATTCGCAACGCCGACTGGATGTAATCCCTGC T I E G L A V G V I R N A D W M *

207

CATGAAACGGCC~CTGCGG~GGCCGTTTCGTTTCTGTACTCCTTTCCCCATGATTTCA

1080

180

ii

-10

31

lexA1 box

lexA2 box

B Region

ATCAGATCCTGAT~CCCGCC~TTTTGTCTGTGCGCGGTGACCGCAGGCGATTCATTTCA GATATATCATTTGGTAAGGCGGACTCGTTGGCTTGAGCAGCTCCCGCTGCTCGCCATGAC GCGCGTTATCCA~TCGCTGATAGCAGGAACCCCGAGGGCAGCAGGGAAAAACAACAGCC TCGTGCGGTGTGCAGCGCACACCTGGCAAAGAGCCGCTTGCGGGATAGAAGGGAGAGTGG TCAGCAGGGCGCCGGCAGTCGGCCCCGCAGCGGCACACTCGCGCA

SD

1

E c 80/GRV._AAA G EPLLAQQHIE/95 A h 85/.K . . . . . SI...E.../105 * #* . Region

2

E C II2/LLRVSGM S MKDIGIMDGDLLAVHK/135 117/ .... Q . . . . . N...L ......... /140 # * . 3

171

1020

1140

12~ 1268 1328

1362

Fig. 2. Nucleotide sequence of :he Ah lexA. The putative ShineDalgarno (SD) site is boxed. The additional 15 bp present in the Ah lexA in comparison to the Ec sequence are underlined. A possible transcription terminator is marked by arrows in the 3' non-coding region. The accession No. of the nt sequence in the EMBL/GenBank is X77263.

LexA repressor binding to the lexA gene (Schnarr et al., 1991 ). In fact, the sequence of the first nine bp of the first SOS box is entirely conserved in all G r a m - lexA genes isolated so far. The A h lexA is the first gene of a m e m b e r of the Vibrionaceae family for which the presence of an Ec SOS-like box has been described, since the Vibrio cholerae recA does not possess this regulatory sequence upstream from its coding region ( G h o s h et al., 1992). It is k n o w n that the N-terminal portion of Ec LexA is involved in D N A binding. This d o m a i n contains three s-helices comprised of aa Gln 8 to Ser 2°, Arg 28 to Leu 35 and Asn 41 to Gly 54 (Schna:rr et al., 1991). In the ~1 helix of the Ah LexA repressor there are eight substitutions at aa 9, 12, 13, 16, 17, 18, 19 and 20, whereas the Qt2 and ~3 helices showed only one change at aa 34 and 52, respectively. The high n u m b e r of substitutions in the ~1 helix gives support to previous suggestions a b o u t the unimportance of this region in the D N A - b i n d i n g ability of the LexA repressor (Oertel-Buchheit et al., 1990). The hydrophobic pocket formed by Ala 32, Ala 42, and Ala 43 of Ec

E c 152/EVTV K RLKKQGNKVELLPE~SEFKPIVVDLRQ~SF/186 157/D . . . . . FQRK. SQ.W ..... E.LS..E...SC.QL/191 . . # . # ## # # .# # ## # .

Ah

Fig. 3. Comparison of Ah and Ec lexA genes. (A) Y-regulatory regions of the lexA gene of both Ah and Ec. The - 35 and - 10 regions and the SD sequence are underlined. LexA-binding regions are indicated. Dashes indicate insertions needed to maintain the alignment. (B) Alignment of the deduced LexA protein sequences of Ec and Ah corresponding to the three regions in which all lexA (Ind-) mutations obtained in E. coli are clustered. The symbols indicate whether a position in the alignment is perfectly conserved (.), similar (*) or is not conserved (#). The number indicates the position in the aa sequence. Alas4, Gly85, Ser119 and Lys156 are in bold type. The residue which is conserved in all the rest of the Gram- lexA genes as well as in other RecA-cleavable proteins, but not in the Ah LexA, is indicated by an arrow. Positions of Ind- mutations isolated in Ee are underlined.

LexA is fully preserved in Ah, including Phe 37. Moreover, the fragment that spans aa 61-67, which has been proposed to be involved in D N A - b i n d i n g ability (OertelBuchheit et al., 1990), presents a Thr 62 ~ A l a 62 substitution. This is the first change found in this segment of all G r a m - lexA genes whose sequence is known. It has been shown that the sites required for the cleavage reaction of the LexA repressor lie within the C-terminal two-thirds of this protein. In this respect, there are three regions in which all lexA ( I n d - ) mutations obtained are clustered (Lin and Little, 1988). Region 1 includes aa 80 t h r o u g h 95, region 2 includes aa 114 t h r o u g h 133 and region 3 includes Thr ls4, Lys 156 and Gln 184. In region 1 there is the Ala84-Gly 85 bond, which is the point where the cleavage of the LexA protein takes place (Horii et al., 1981; M a r k h a m et al., 1981). Regions 2 and 3 contain, respectively, residues Ser 119 and Lys 156, which participate in the cleavage of Ala84-Gly 8s by a mechanism similar to that of a serine protease (Slilaty

74 and Little, 1987). Fig. 3B shows that, as predicted, Ah LexA also contains the Ala-Gly bond, but at aa 89-90 as a consequence of the 15-bp insertion mentioned above. Likewise, the residues upstream and downstream from this point whose substitution in Ec LexA generates a LexA(Ind-) phenotype (Lin and Little, 1988) are also preserved. It is worth noting that, regardless of the precise position of the Ala-Gly bond, the distance between this point and that of the Lys and Ser residues remains unaltered in the Ah LexA repressor. The region containing the Ser 119 is at position 124 in the Ah LexA and is highly conserved (Fig. 3B). Nevertheless, there is a significant change around the residue functionally equivalent to the Lys 156 of Ec LexA (Lys t61 at the Ah LexA). So, and whereas other important residues of region 3 are fully preserved (Fig. 3B), the Glu 152, which is maintained in all the rest of G r a m - lexA genes, as well as in other RecA-cleavable proteins such as UmuD and X, 434 and P22 repressors (Lin and Little, 1988), is substituted by Asp (Fig. 3B). (c) Construction and behaviour of chromosomal lexA::lacZ fusions Data presented above clearly indicated that the Ah lexA presents two Ec-like SOS-boxes upstream from the encoding region. Nevertheless, several genes are known in which, despite the presence of these kinds of sequences, they are not DNA-damage inducible in Ec. Besides, there are several recA genes from different G r a m - bacteria which have been shown not to be DNA-damage inducible (Miller and Kokjohn, 1990). For these reasons, and to confirm that Ah lexA is really inducible in both Ec and its own genetic background, a fusion between lacZ and this gene has been constructed. The promoter and the 5'-coding region of the Ah lexA located as described above were cloned in plasmid pUJ8 (de Lorenzo et al., 1990) upstream from the promoterless trp'-'lacZ region, giving rise to plasmids pUA466. Afterwards, a NotI fragment of this plasmid containing the lecA::lacZ fusion was inserted into the unique NotI cloning site of the miniTn5Km (de Lorenzo et al., 1990). The minitransposon constructed as described above and containing the lecA::lacZ fusions is located on a R6Kbased suicide delivery plasmid that provides the IS50R transposase tnp gene in cis, but external to the mobile element and whose conjugal transfer to bacterial recipients is mediated by RP4 mobilization functions in the donor (de Lorenzo et al., 1990). Once this lecA::lacZ plasmid fusion had been constructed, it was transformed into E. coli S17-1(Xpir) and then introduced into the chromosome of Rif R derivatives of Ec MC1061, and Ah. 20 independent exconjugants of each mating were analyzed, and one clone of each recipi-

ent strain with a basal level belonging to the most representative value was selected for further study. About 90% of the exconjugants tested in each case showed a similar 13Gal activity. Basal levels of the lexA::lacZ fusion of Ah, when inserted into either its own chromosome or into that of E. coli were of 190 or 45 units of 13Gal, respectively. Moreover, the basal expression of the lecA::lacZ fusion of Ec, when inserted in either its own chromosome or in that of Ah were of 175 or 1200 units of 13Gal, respectively. These data indicate that the affinity of the Ec LexA repressor must be approximately the same for its own operators as for that of the Ah lexA gene. On the other hand, the fact that the level of the basal expression of the Ec lecA::lacZ fusion was so high in Ah indicates a low affinity of the Ah LexA repressor to the operators of the Ec lexA gene. The DNA damage-mediated expression of both lecA::lacZ fusions suggests that in Ec cells the extent of induction was about twofold higher for the Ah lexA gene than that of Ec (Fig. 4). In Ah cells the induction of the Ah lexA was about fivefold higher than that of Ec. The induction of the Ah lexA gene was always higher than that of Ec. Such behaviour could perhaps be attributed to the difference in the distance between two LexAbinding sequences of the lexA genes of Ec (5 bp) and Ah (4 bp). In fact, similar data have been obtained with the Ec lexA when inserted into the chromosome of both P. aeruginosa and P. putida whose lexA genes present a distance of 3 bp between the first and the second SOS boxes

6"~ (A) 5-

g ~ e.

0-

(8)

8-

~ 6 .~_

.

~

4-

2-

0

0

J

i

J

I

30

60

90

120

Time (rain)

Fig. 4. Induction by 20 tag mitomycin C/ml of the lecA::lacZ fusion from Ec (11) and Ah (O) inserted into the chromosomeof Ah (A) and Ec (B). The induction factor is the ratio betweenunits of [3Galof both mitomycin-C-treatedand untreated cells.

75 (Calero et al., 1993). Mitomycin C-mediated induction of Ah lexA was also detected when analyzed by Northern experiments using the internal 0.9-kb HindlII-SalI fragment as a probe (data not shown).

Direcci6 General d'Universitats de la Generalitat de Catalunya for grants for the purchase of equipment.

REFERENCES

(d) Conclusions In the present work we have isolated and characterized the lexA gene from Ah. The principal properties of this gene and its product are: (1) The Ah LexA protein is able to repress Ec SOS genes. (2) The Ah LexA protein is likely to be cleaved by the Ec RecA coprotease. (3) Upstream of the encoding region of the Ah lexA gene there are two Ec-like SOS boxes. These two regulatory sequences are separated by 4 bp instead of 5 bp in Ec.

(4) The Ah LexA repressor contains 207 aa and is 75% identical to that of Ec. The structure and sequence of the DNA-binding domain of the Ec LexA repressor, as well as the region at which its hydrolysis occurs, are highly conserved in the Ah LexA protein. Nevertheless, a residue of the region implicated in the specific cleavage reaction, and which until now had be,en found to be fully preserved in all known RecA-cleavable proteins, is substituted in the Ah LexA repressor. (5) Expression of the AI* lexA gene is DNA damage inducible in both Ah and Ec genetic backgrounds, practically to the same extent. Nevertheless, the Ec lexA gene is poorly induced in DNA-injured Ah cells. This is the first gene of a member of the Vibrionaceae family which has been shown to be DNA damage inducible.

ACKNOWLEDGEMENTS

This work was supported by grant PB91-0470 of the Comisirn Interministerial de Ciencia y Tecnologia (CICYT), Spain. We would also like to acknowledge the

Calero, S., Garriga, X. and Barbr, J.: One-step cloning system for isolation of bacterial lexA-like genes. J. Bacteriol. 173 (1991) 7345 7350 Calero S., Garriga, X. and Barbr, J.: Analysis of the DNA damagemediated induction of Pseudomonas putida and Pseudomonas aeruginosa lexA genes. FEMS Microbiol. Lett. 110 (1993) 65-70. de Lorenzo, V., Herrero, M., Jakubzik, U. and Timmis, K.N.: MiniTn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in Gramnegative eubacteria. J. Bacteriol. 172 (1990) 6568-6572. Garriga, X., Calero, S. and Barbr, J.: Nucleotide sequence analysis and comparison of the lexA genes from Salmonella typhimurium, Erwinia carotovora, Pseudomonas aeruginosa and Pseudomonas putida. Mol. Gen. Genet. 236 (1992) 125-134. Ghosh, S.K., Biswas, S.K., Paul, K. and Das, J.: Nucleotide and deduced amino acid sequence of the recA gene of Vibrio cholerae. Nucleic Acids Res. 20 (1992) 372. Lin, L.L. and Little, J.W.: Isolation and characterization of noncleavable ( I n d ) mutants of the LexA repressor of Escherichia coil K-12. J. Bacteriol. 170 (1988) 2163-2173. Little, J.W.: Mechanism of specific LexA cleavage: autodigestion and the role of RecA coprotease. Biochimie 73 (1991) 411 422. Miller, R.V. and Kokjohn, T.A.: General microbiology of recA. Environmental and evolutionary significance. Annu. Rev. Microbiol. 44 (1990) 365-394 Oertel-Buchheit, P., Lamerichs, R.M.J.N., Schnarr, M. and GrangerSchnarr, M.: Genetic analysis of the LexA repressor: isolation and characterization of LexA (Def) mutant proteins. Mol. Gen. Genet. 223 (1990) 40-48 Riera, J. and Barbr, J.: Sequence of the Providencia rettgeri lexA gene and its control region. Nucleic Acids Res. 21 (1993) 2256. Sanger, F., Nicklen, S. and Coulson, A.: DNA sequencing with chainterminating inhibitors. Proc. Natl. Acad. Sci. USA 74 (1977) 5463-5467 Schnarr, M., Oertel-Buchheit, P., Kazmaier, M. and Granger-Schnarr, M.: DNA binding properties of the LexA repressor. Biochimie 73 (1991) 423-431 Sedgwick, S.G. and Goodwin, P.A.: Interspecies regulation of the SOS response by the Escherichia coli lexA ÷ gene. Mutation Res. 145 (1985) 103-106. Slilaty, S.N. and Little, J.W.: Lysine-156 and serine-ll9 are required for LexA repressor cleavage: a possible mechanism. Proc. Natl. Acad. Sci. USA 84 (1987) 3987-3991. Walker, G.C.: Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coll. Microbiol. Rev. 48 (1984) 60-93