Mol Gen Genet (1992) 236:125-134 MGXI © Springer-Verlag 1992

Nucleotide sequence analysis and comparison of the lexA genes from Salmonella typhimurium, Erwinia carotovora, Pseudomonas aeruginosa and Pseudomonas putida Xavier Garriga, Sebastifin Calero, and Jordi Barb6

Department of Genetics and Microbiology, Autonomous University of Barcelona, Bellaterra, 08193 - Barcelona, Spain

Received March 6, 1992 / Accepted May 18, 1992

Summary. The complete nucleotide sequences of the lexA genes from Salmonella typhimurium, Erwinia caro- tovora, Pseudomonas aeruginosa and Pseudomonas putida were determined; the DNA sequences of the lexA genes from these bacteria were 86%, 76%, 61% and 59% simi- lar, respectively, to the Escherichia coli K12 gene. The predicted amino acid sequences of the S. typhimurium, E. carotovora and P. putida LexA proteins are 202 resi- dues long whereas that of P. aeruginosa is 204. Two putative LexA repressor binding sites were localized up- stream of each of the heterologous genes, the distance between them being 5 bp in S. typhimurium and E. caro- tovora, as in the lexA gene of E. coli, and 3 bp in P. putida and P. aeruginosa. The first lexA site present in the lexA operator of all five bacteria is very well con- served. However, the second lexA box is considerably more variable. The Ala-84-Gly-85 bond, at which the LexA repressor of E. coli is cleaved during the induction of the SOS response, is also found in the LexA proteins of S. typhimurium and E. carotovora. Likewise, the amino acids Ser-119 and Lys-156 are present in all of these three LexA repressors. These residues also exist in the LexA proteins of P. putida and P. aeruginosa, but they are displaced by 4 and 6 residues, respectively. Furthermore, the structure and sequence of the DNA- binding domain of the LexA repressor of E. coli are highly conserved in the S. typhimurium, E. carotovora, P. aeruginosa and P. putida LexA proteins.

Key words: lexA gene - Salmonella typhimurium - Erwin- ia carotovora - Pseudomonas aeruginosa - Pseudomonas putida

Introduction

The SOS response to DNA damage in Escherichia coli is mediated through the r e c A - lexA regulon (Little and

Correspondence to . J. Barb6

Mount 1982; Walker 1984). The lexA gene encodes a repressor that binds the operators of at least 20 different genes, called damage-inducible (din) or SOS genes, which include recA and lexA (Walker 1984). Interruption of replication or damage to the DNA generates an induc- tion signal, which activates the constitutive level of RecA with the help of the Ssb proteins (Lieberman and Witkin 1981; Weinstock and McEntee 1981) and then promotes cleavage of LexA (Craig and Roberts 1980; Little et al. 1980). The characteristics of this inducing signal are still unknown, although it has been proposed that it could be single-stranded DNA regions that would arise after DNA damage was processed during DNA replication (Sassanfar and Roberts 1990). However, RecA appears to play an indirect role in LexA cleavage because it has been reported that the LexA protein may self-cleave in vitro under some conditions (Little 1984). For this rea- son, it has recently been suggested that the term copro- tease should be adopted to emphasize this indirect role of RecA protein in the LexA hydrolysis process (Little 1991). The RecA coprotease of E. coli also collaborates in the cleavage of the UmuD protein to a form (UmuD*) that is active for mutagenesis (Burckhardt et al. 1988; Nohmi et al. 1988; Shinagawa et al. 1988), as well as in the hydrolysis of some prophage repressors.

The LexA repressor of E. coli is a protein of 202 resi- dues, which is composed of two domains linked by a hinge region. The first 84 amino-terminal amino acids are sufficient to confer substantial DNA binding ability (Hurstel et al. 1986). The entire repressor is likely to bind operator DNA as a dimer (Schnarr et al. 1985) through a two-step process in which two monomers seem to bind sequentially (Kim and Little 1992). This binding is at a specific region present upstream of the SOS genes, which has been called the SOS box, its con- sensus sequence being CTGTATATATATACAG (Wertman and Mount 1985). However, some SOS genes have more than one SOS box in the promoter. Thus, the lexA gene itself has two lexA boxes (Horii et al. 1981a), whereas the caa and recN genes contain two and three, respec- tively (Lloubes et al. 1991 ; Schnarr et al. 1991). The car-

126

boxy-terminal domain of the LexA repressor stabilizes D N A binding to the SOS boxes and is further involved in the induction of the SOS system through the RecA- promoted cleavage reaction between amino acids Ala-84 and Gly-85 (Slilaty and Little 1987). This carboxy-termi- nal domain is fairly homologous to the UmuD protein, the SOS-mediated proteolytic cleavage of which occurs between a cysteine and a glycine residue (Burckhardt et al. 1988; Nohmi et al. 1988; Shinagawa et al. 1988), as well as to repressors of three inducible phages, 2, 434 and P22 (Sauer et al. 1982). The mechanism of au- todigestion of the LexA protein of E. coli is similar to that of a serine protease and requires Lys-156 and Ser- 119 (Slilaty and Little 1987).

The existence of similar D N A damage-inducible re- sponses in other bacterial species has been widely re- ported (Miller and Kokjohn 1990). Bacterial recA-like genes may be easily isolated directly by using as a selec- tive phenotype the resistance to DNA damage that a plasmid harbouring the heterologous recA gene confers on recA mutants of E. coli. In fact, the recA gene of many bacterial species has been cloned by this method (Miller and Kokjohn 1990; Roca and Cox 1990). Wheth- er a LexA repressor also exists in species and genera containing recA-like genes is not in all cases clear. The D N A sequences of some recA-analogous genes show possible LexA-binding sites, but others do not, although it has been shown that many species of gram-negative bacteria are able to regulate the expression of the recA gene of E. coli (Fernandez de Henestrosa et al. 1991).

Nevertheless, because the lexA gene has a negative regu- latory role in the SOS pathway, a system to isolate lexA- like genes of bacteria directly had not been developed until a short time ago (Calero et al. 1991). However, a DNA-damage inducible regulation gene of Bacillus subtilis (dinR) has recently been identified that has a sequence that shows 34% identity with the le;rA gene of E. coli (Raymond-Denise and Guillen 1991). The DinR protein also shows similarity with some of the functional regions of the LexA protein of E. coli. In this report, we shall present the complete nucleotide se- quence of the lexA genes from two species of gram- negative bacteria that are closely related to E. coli (Sal- monella typhimurium and Erwinia carotovora), and two others that are not (Pseudomonas aeruginosa and Pseu- domonas putida).

Materials and methods

Bacterial strains, plasmids and growth conditions. The bacterial strains and plasmids used in this work are listed in Table 1. All of the strains were normally grown at 37 ° C in LB rich medium (Miller 1972) or AB minimal medium (Clark and Maaloe 1967) and supplemented with glucose (0.2% wt/vol) or casamino acids (0.4% wt/ vol).

D N A procedures. The conditions used for restriction en- donuclease digestion, agarose gel electrophoresis, isola-

Table 1. Bacterial strains and plasmids used Strain or plasmid Relevant characteristics Source or

reference

Escherichia coli

DH5c~ JL2301

Plasmids pUA166

pUA167

pUA168

pUA170

pUA193

pUA197

pUA212

pUA219

fecAl hsdR17 gyrA

lexA300 (Del) recA + sulA : : laeZ

pUAI65 containing a 6.2 kb fragment carrying the Pseudomonas putida lexA gene pUA165 containing a 7.4 kb fragment carrying the Erwinia carotovora lexA gene pUAI65 containing a 8.3 kb fragment carrying the Salmonella typhimurium lexA gene pUA165 containing a 6.2 kb fragment carrying the Pseudomonas aeruginosa lexA gene pBSK containing a 1.8 kb EcoRI- BamHI fragment carrying the S. typhimurium lexA gene pBSK containing a 2.0 kb BglI--EcoRI fragment carrying the E. carotovora lexA gene pBSK containing a 1.8 kb Pst-SphI fragment carrying the P. putida lexA gene pBSK containing a 2.1 kb SalI-PstI fragment carrying the P. aeruginosa lexA gene

This laboratory Hill and Little (1988)

Calero et al. (•99•) Calero et al. (1991) Calero et al. (•99•) Calero et al. (1991) This work

This work

This work

This work

127

A Sail Oindllt BOlt I EcoRI ?, '

L e x A activity

4-

C

Sail Slhl Salll Pstll P]t l IEcoRI

Sma I Sp~i S~a, SmalBa~__HlLeXA a~+

+

B

H/r~llll Hinc ll Hinoll Hirn ll

Barn Pst I EcoRI ima i . . , si" ff' Sm81 [ PI" - -

4-

4-

1Kb

Fig. 1A-D. Physical maps of plasmids pUA167 (A), pUA168 (B), pUA166 (C) and pUA170 (D) and their derivatives carrying sub- fragments of the lexA region. The symbols ( + / - ) indicate the

ability or otherwise to complement the lexA3OO(Del) mutation of strain JL2301. The black boxes indicate the polylinker region of pUA165

tion and ligation of DNA fragments and transformation into E. coli followed standard methods (Sambrook et al. 1989). Restriction enzymes, T4 ligase, alkaline phospha- tase and the Klenow fragment of DNA polymerase ! were purchased from Boehringer Mannheim.

Cloning of the lexA-Iike genes from S. typhimurium, E. carotovora, P. aeruginosa and P. putida in plasmid pBluescriptSK. Plasmids pUA166, pUA167, pUA168, and pUA170 containing the lexA genes of P. putida, E. carotovora, S. typhimurium and P. aeruginosa, respec- tively, have been described previously (Calero et al. 1991). Subcloning using pUC-based plasmids as vectors was performed to locate the lexA gene more precisely (see Fig. 1). For nucleotide sequencing the smallest re- striction fragment of each that retained lexA activity was cloned into the appropriate restriction sites of pBluescriptSK (Stratagene) to obtain pUA193, pUA197, pUA212 and pUA219.

To confirm that the cloned fragments from pUA166, pUA167, pUA168 and pUA170 contained a functional lexA-like gene, the several plasmids obtained were trans- formed into strain JL2301, which completely lacks the lexA gene sequence and carries a fusion of the lexA- regulable suIA promoter to the laeZ gene (Hill and Little 1988). The presence of recombinant plasmids harbour- ing a functional lexA-like gene gave rise to a dramatic decrease in basal expression of the sulA gene as revealed by the /%galactosidase asssay (Miller 1972), indicating that the plasmid-encoded LexA-like protein repressed the sulA :: lacZ fusion in this strain. However, recombin-

ant plasmids containing a truncated lexA-like gene did not show any decrease in the basal expression level of the sulA::lacZ fusion, indicating that the repression is LexA specific.

Determination of nucleotide sequence. The sequencing strategies used for each of these lexA genes are summa- rized in Fig. 2. In all cases more than 200 bp of both 5' and 3' flanking sequences was determined in addition to the coding regions.

The nucleotide sequence was determined using the dideoxy-chain termination method (Sanger et al. 1977). The starting materials for DNA sequencing were pUA193, pUA197, pUA2!2 and pUA219. Unidirection- al 5' to 3' deletions within lexA-like genes and flanking DNA were made with the Erase-a-Base kit (Promega Biotech). The polylinker region of the plasmids was di- gested with two enzymes, one of which produces a site sensitive to exonuclease III (ExoIII) while the second renders the other end resistant to ExoIII. The plasmids were digested with ExoIII for various times and then treated with S1 nuclease and Klenow DNA polymerase to remove overhangs introduced by the previous treat- ments and allow ligation of the shortened molecules. Derivatives of pUA193, pUA197, pUA212 and pUA219 containing overlapping deletions differing in size by ap- proximately 220 bp were chosen for sequencing. DNA sequencing was done with the T7 Sequencing kit (Phar- macia), with double-stranded plasmid DNA as the tem- plate and [e-35S] dATP (Amersham) as the label. In all cases DNA sequence was obtained from both strands.

128

A Sma I H/ndl II HJncll

I mm

lexA

pl.3 pl.4 pl.5 pl.6 pl.7 pt.8

111

p2.2 ~__ p2.3 ~_ p2.5 ~__ p2.6

O Sph 5 Sat I IKsp i, Xmn i Dra I Eco R

lexA

p5.3 p5,4 p5.5 p5.6

Pstl

L w

p6.3 ~__ p6.5 ~. p6,6

p6.7 p6.8

B

p3.3 p3.4 p3.5 _~ p3.6 p3.7 p3.8

Sail Hindll l y o lexA

p4.3 ~ p4.4

p4.5 ~__ p4.6

p4.7 400bp

Fig. 2A-D. Maps of lexA regions showing relevant restriction sites and exonuclease III deletion derivatives of pUA193 (A), pUA197 (B), pUA212 (C) and pUA219 (D) used for sequencing. The arrow below the map indicates the extent of the lexA gene and the tran-

Sma I Hincll

ri p7.1 p7.2 p7.3 p7.4 p7.5

ml

lexA

Pstl Xmn t Sac I

I IL

p8.5 p8.7 p8.8 p8.9

~_ p8.10 ~- p8.11

scription direction of this gene. Each arrow below this represents the length of a clone used for sequencing and the reading direction of the nucleotide sequence. The black boxes indicate the polylinker region of pBSK

When necessary, 7-deaza-dGTP, 7-deaza-dITP (Phar- macia), and Taq DNA polymerase (Promega) were em- ployed.

Analysis of nucleotide and protein sequence. DNA and protein sequences were analysed using the PC/GENE package (Version 6.26; IntelliGenetics). CLUSTAL was used to obtain the multiple alignments of the nucleotide and amino acid sequences. The secondary structure of the predicted LexA proteins was obtained by means of the GARNIER program and the hydropathy analysis of heterologous LexA proteins was performed using SOAP. HARPIN was used to search for any repeat se- quences and CHARGPRO was employed to calculate the theoretical isoelectric point.

Nucleotide sequence accession numbers. The nucleotide sequences of all lexA genes reported in this paper will appear in the EMBL, GenBank and DDBJ nucleotide sequence data bases under their accession number: X63002 (S. typhimurium), X63189 (E. carotovora), X63018 (P. aeruginosa) and X63017 (P. putida).

Results

DNA nucleotide sequence comparison oflexA genes

The DNA sequence of the four heterologous lexA genes is shown in Fig. 3. The open reading frame was 606 bp

in length in S. typhimurium, E. carotovora, and P. putida, as well as in E. coli, and 612 in P. aeruginosa. The simi- larity of the sequence of S. typhimurium and E. caroto- vora to that of E. coli (86% and 76%, respectively) was higher than that of P. aeruginosa and P. putida (61% and 59% respectively). Likewise, the lexA genes of P. aeruginosa and P. putida show close identity (82%). These data agree with previous results obtained in hy- bridization experiments (Calero et al. 1991). It is worth noting that neither S. typhimurium nor E. carotovora show any additions or deletions in their respective lexA genes when compared with E. coli. On the other hand, P. aeruginosa and P. putida have an additional 18 and 12 bp, respectively, at position 204. Furthermore, these two bacterial species have a deletion of 12 bp just at the end of the gene. However, no other additions or deletions are present elsewhere in these lexA genes. The substitutions in the lexA gene of S. typhimurium do not seem to occur uniformly withrespect to the E. coli gene, but rather, as shown in Fig. 3, greater sequence diver- gence is evident in the carboxyl-terminal coding region. On the other hand, the lexA gene of E. carotovora has a uniform distribution of substitutions.

Comparison of deduced sequences of LexA proteins

The predicted amino acid sequences of the S. typhimur- ium, E. carotovora and P. putida LexA proteins have

Ec ATGAAAGCGT TAACGGCCAG GCAACAAGAG GTGTTTGATC TCATCCGTGA 50

St ............................................... G.. 50

Ecc ....... TA ..... A..A ..... G..GC .... T.A...C..G ..... C.. 5O

pp ,..TTGAAAC .G...C,AC. C...GCC..A A.TC,C.CGT ..... AAGCG 5o

Pa ...C.GAA.C .G...C..C.C...GCC... A.CC.CTCCT ..... AAGCG 5o

Ec TCACATCAGC CAGACAGGTA TGCCGCCGAC GCGTGCGGAA ATCGCGCAGC 1 oo

St ........................ A ......................... i00

Ecc . . .T..TGCG . .A..C..A ........ A ..... G ........ T..T..A. i00

pp CTG.C.GGAA G,C.AC..CT .C ..... T,. C..C..C..G . .T. ,T...G i00

Pa CTG.C.GGAA G.CCAC..CT .T ........ C..G ........... C...G tO0

Ec GTTT~%~f CCGTTCCCCA AACGCGGCTG AAGAACATCT GAAGGCGCTG 150

St ............................ G ........ C.. T..A ...... 150

Ecc AAC.A ..... T..C..T..C . .T .................... A..A... 150

pp AGC .... C.. .AAG..G..C . .T.,C..C. .G..G..C.. C ..... C..T 150

AAC.C..C.. .AAG..G..G ..... C..C..G..G..C.. C ......... 150 Pa

Ec GCACGCAAAG GCGTTATTGA AATTGTTTCC GGCGCATCAC

St . .G ........ G. ,GC ....... C .............. G.

Ecc . .G..T ..... T..G ........... A..G ..... G..T.

Pp • .C ..... G .... CG..C ..... GACGC. G ..... T..C.

Pa ........ G .... CC..C ..... GAC.C. G ..... C..T.

Ec TCTG .................. TTGCAGGA AGAGGAAGAA

St .... - ................. . .A ........ A..G..C

Ecc .... - ................. C.. AT ...... A.. GACG

Pp CA.CCCTGGC CTGGAAGCCA AGGCT ...... . .A .... CC

Pa CA.TCCCGGC TTCGAACCGC ATGCCGCCA. C..C..T..G

GCGGGATTCG 2OO

.A, .T..C.. 2OO

.T..T ..... 2OO

.... C..C.. 2OO

. . . . C..C.. 200

C~TTGCCGC 232

• ,A..A .... 232

• .TA.T..T. 232

• .CC .... CA 244

• .CC ..... G 25O

Ec TGGTAGGTCG T G ~ T G C C GGTGAACCAC TTCTGCCGCA ACAGCATAY'P 282

St .T..C..G ..... C..G..G ........ G ........... G ......... 282

Ecc .... T ..... C ..... C..A ........ G. .G ........ GG.A..C..C 282

pp .CA.C..C.. G..C .......... CG..GA .... T..CG ....... C..C 294

Pa . .A.C..A.. G..C, .C ..... C.C...GA .C..C..CG ..... A.C..C 300

Ec GAAGGTCA%~2 ATCAGGTCGA TCCq~fCCTTA TTCAAGCCGA ATGCTGAT~ 332

St ..... C ..... C .............. GC.G ..... A..C. GC ........ 332

Ecc ...T...GC ........ T.. C...G.GA.G ..... A..C. GC ........ 332

pp . .GCAATCC. GCA.CA..A. C,..G .... C ...C.C. ,CC .A, .C..G.A 344

Pa . .G.AATCC. GC.G.A..A .... CG .... C ..... T..TC GC..C..C.A 350

Ec CCTGCTGCGC GTCAGCGGGA TGTCGATGAA AGATATCC~ ATTATGGATG 3 8 2

St ............ G ..... T .............. C ..... T .......... 382

Ecc TT ....... G . .G ..... C ............ A ....... T .......... 382

Pp TT .......... ACA.. ,C. . .AGC ..... G..CG ....... CT.C. ,C. 394

Pa .... T ....... GC .... C. , .AGC ..... G..C ......... C.C,.C. 400

Ec GTGACTTGCT GGCAGTGCAT AAAACTCAGG ATGTACGTAA CGGTCAGGTC 432

St .G, .T ........ O. ,A ........ G ..... C, .C ..... T. ,C ..... G 432

Ecc .C..T..A ..... C ........... AG.A ..... G..A.. T ...... A.T 432

Pp .C...C ....... G ..... C .CCTGC.GT. .A. CC..C ..... C...A.. 444

Pa .C...C ....... C..C. ,C GTC, .C.GC. .A.CG..C ..... C ..... G 450

Ec GTTGTCGCAC GTATTGATGA CGAAGTTACC GTTAAGCGCC TGAAAAAACA 482

St ........ T. .C ........ T ..... G,.A. .A..A ........... G.. 482

Ecc ..C ..... G ........ C.. T ..... G..G. .G .............. G.. 482

Pp ..... G..G. .C..C.G ......... C ..... C ...... T .C..GCGCG. 494

Pa ..G ..... G. .G..C.GC.. G..G..C..G ..G..A...T .C..GCGCG. 500

Ec ~ A A T A A A GTCGAACTGT TGCCAGAAAA TAGCGAGTTT AAACCAATTG 532

St ............ G..G...C .... G ..... C .......... C...G..A. 532

Ecc ........ CG ..AC.C..CC .CG.T ...... GAA ...... GCC..G .... 532

Pp A .... GC ...... TGG...C .TG.C ..... CCC...A, .C C~C. ,C..C. 544

Pa A .... CC..G ...TGG...C . .G,G ..... CCCT ..... C GCT..G..C. 550

Ec TCGTTGACCT TCGTCAGCAG AGC~TCACCA TTGAAGGGCT GGCGGTTGGG 582

St .G..G..T ..G..CG.A..A ..... T..G ................. A..C 582

Ecc .T..C ..... G ........ A ..... TT.G. .A ..... CT. A ........ C 582

Pp AA..C ..... GAAAG.A... GAGC.GGTG..C.,G..CT..AGC..C..C 594

Pa AA..C..T,. GAAGG ..... GAAC.G.T.. .C ..... CT. .AGC..C..C 600

Ec GTTATTCGCA ACGGCGACTG GCTGTAACAT ATCTCTGAGA CCGCGTGCCG 632

St . .C .............. A.. .T..TAGTCT CTTTTTAATC TCCTTGTAAG 632

Ecc . .G ..... T. . .A ........ AGCT _ECA TTCGCAGAGA TGCACTGCTC 632

Pp . .C ...... C G.TGA ............ TCC AGGAGGCGTC ATGCAGCAGT 632

Pa . .G..C..AC G.TGA ............ CAG GAGATACCAT GCAGACCTCC 63s

Ec CCTGGCGTCG CGGTTTGTTT TTCATCTCTC TTCATCAGGC TTGTCTGCAT 682

St CCGCCATCCG GCAATCGTGT AGCCTGATGG CGCTGCGCTT ATCAGGCCTA 682

Ecc GTCTGTCTGG GTCAGTGCAT CATGTCCTGT ATTCATCGTT TAGCGTGTTA 682

Pp TCATTCACGC ACCCGAGCAA GCCCAGTTGC CCCTGTTCGA AGCATTCC~E 682

Pa CACTCGCTGC CCAGCGCCCA GTTGCCACTG TTCCAGGAAG CGTTCTGGGC 688

Ec GGCATTCCTC ACTTCATCTG ATAAAGCACT C ~ A T C T C GCCTTACCCA 732

St CGGGAATGCA GTTCCTGAGA TGATTAATTT GTA~CGGA TAAGGCGTTA 732

Ece ATCTGCTAAC CATATATATT TAGT~ACATT TCGCGCGCAT TTTCTACGAT 732

Pp GCCCAGCCCG TGCTGCCAGG CCTGAAAGCC AGGGAACCGG CGCGCAAGAG 732

Pa CAGCAACGCgZ GCTCCCT~GC TCGACGATGT CATCGACAGC CCTTCCAGCG 738

Ec TGATTTTCTC CAATATCACC GTTC 756

St CGTCGCCATC CGGCAATGCG CTCG 756

Ece CCCTATTACC CTCTGTTTTT TCAC 756

Pp CAGCCAGCCC GAGCTG~TCA GCGA 756

Pa CCTCCATCGA GGAACCCGCT GCCT 760

Fig. 3. D N A sequence comparison of the coding region of the lexA genes and the sequences immediately downstream from the termi- nation codons of Escherichia coli (Ec), Salmonella typhimurium (St), Erwinia carotovora (Ecc), Pseudomonas putida (Pp) and Pseu- domonas aeruginosa (Pa). The nucleotide sequence of the lexA gene

129

EC MKALTARQQE VFDLIRDHIS QTGMPPTRAE IAQRLGFRSP NAAEEHLKAL 50

St .................................................. 5O

ECC ..V ...... Q .Y ....... A ............. Q ................ 5o

Pp .LK..P-.A. ILAF.KRCLE DN.F ......... E...K ............ 5O

Pa .QK..P..A. ILSF.KRCLE DH.F ......... E...K ............ 5o

Ec ARKGVIEIVS GASRGIRL ...... LQEEEE GLPLVGRVAA GEPLLAQQHI 94

St ..... L ............ - ..... . .... D .................... 94

Ecc .................. - ..... .M . .T I ................ E.. 94

Pp .... A..MTP ....... IPG LEAKA--..A ...II ...... A.I..E... 98

Pa .... A..MTP ....... IPG FEPHAANDD .... VI ...... A.I..E.N. lOO

Ec EGHYQVDPSL FKPNADFLLR VSGMSMZ<DIG IMDGDLLAVH KTQDVRNGQV ~_44

St ............. S .................................... 144

Ecc .CR ..... AM ...S ............. N ........... . ..E ...... I 144

Pp .QSCNIN.AF .H.Q..Y .... H ...... V. .F ........ TCREA .... I 148

Pa . ESCRIN.AF .N.R..Y .... R ......... L ........ V.REA ..... 150

Ec VVARIDDEVT VKRLKKQGNK VELLPENSEF KPIVVDLRQQ SFTIEGLAVG 194

St .............................. T ....... E ........... 194

ECC ................... T .H..A..E.. A ........... S ....... 194

Pp ..... G ....... F.RE.S. .W..A..P.. A..E...KE. ELV .... S.. 198

Pa ..... GE ...... F.RE.S. .W..A..P.. A..E...KE. ELI .... S.. 20o

Ec VIRNGDWL 202

St ..... E.. 202

Ecc .... S..S 202

Pp ...R .... 202

Pa ...R .... 2o4

Fig. 4. Deduced amino acid sequence of the LexA proteins of Escherichia coli (Ec), Salmonella typhimurium (St), Erwinia caroto- vora (Ecc), Pseudomonas putida (Pp) and Pseudomonas aeruginosa (Pa). The predicted protein sequences of the heterologous LexA proteins are shown in comparison with that of E. coli. identical residues are indicated by a dot and amino acid substitutions are indicated. Dashes indicate insertions needed to maintain the align- ment

202 residues (Fig. 4), as does the LexA protein of E. co~i, whereas that of P. aeruginosa has 204. The deduced mo- lecular weight of all these proteins was practically the same: 22358 Da for E. eoli (Horii et al. 1981a; Mark- ham et al. 1981), 22320 for E. carotovora, 22307 for S. typhimurium, 22506 for P. aeruginosa and 22147 for P. putida, their isoelectric points being 6.24, 5.95, 5.7, 6.13 and 6.24 respectively. The amino acid sequence sim- ilarity of these proteins to the LexA repressor of E. coIi was 97% for S. typhimurium, 87.6% for E. carotovora, 64% for P. putida and 64% for P. aeruginosa. The degree of identity between the LexA proteins of both members of the Pseudomonodaceae family was 89%. The hydro- pathy profiles of the five LexA proteins did not differ significantly from each other (data not shown).

Analysis of the 5' regulatory regions of the lexA genes

In E. coli, the lexA gene is regulated by its own product, which recognizes two LexA-binding boxes located imme- diately upstream of the initiator ATG codon (Brent and Ptashne 1981 ; Little et al. 1981). The sequences of the 5' regulatory regions of the four heterologous lexA genes are shown in Fig. 5A. Alignment of these D N A se- quences with that of the E. coli lexA gene revealed that

of E. eoti is shown for comparison. The nucleotides are numbered starting from the first nucleotide of the ATG initiation codon. The termination codon is at about residue 606. Both these codons are underlined. Identical residues in the open reading frame are depicted by a dot and nucleotide substitutions are indicated. Dashes indicate insertions needed to maintain the alignment

130

A - 3 5

EC ATTCGATAAA TCTCT--GGTTT ATTGTGCAGT TTATGGTTCC AAAATCGCCT St ATTTTATAAA TCTCT--TGTTT TTTGCGCCGT TTGTGGTTCC AA/hATCACCT Ecc AATGAGTAAA AGTCA--TGGGC TTTGCCGATA TCTTCATTCC CAATTGTGCC Pp AAGCG-CCAC -CTCTTCTGTTT ATTGCGCCTT TTGGC-TTGC AGCG-CACCC Pa AT-CG-TGAC GCAGGACGAAAT GTCATGCGGA AAATCATTGC CACG-CCTCG

lexA BOX 1 lexA BOX 2

Eo TTT~CTGTAT ATAc~eAcAG[CAT~CTGTA TATACACCC~CGGa--ATG

St TT ................ CAT T ............. CGGA--ATG

Ecc TTA I ................ I CAAGI ....... A... A..~CGGA--ATG

Pa cTc~ t . . . . . . . . . A..c... l--~c~...G...A.~..A...~ccacG~-cATG 8 IexAI box e T G T A T A T A c T c a C A G

lexA2 box c T G t A t A t A c T c c C A G

Fig. 5. A Comparison of the 5' regulatory regions of the heterolo- gous lexA genes. The DNA sequences immediately upstream of the coding regions are shown for Escherichia coli (Ec), Salmonella typhimurium (St), Erwinia carotovora (Ecc), Pseudomonas putida (Pp) and Pseudomonas aeruginosa (Pa). Canonical sequences for the Pribnow box(- 10) and the recognition site (-35) are indicated (Harley and Reynolds 1987; Deretic et al. 1989) as well as the ribosome-binding site (SD) (Shine and Dalgarno 1975). LexA-bind- ing regions are boxed. Identical residues in the lexA boxes are indicated by a dot and nucleotide substitutions are indicated. Dashes indicate insertions needed to maintain the alignment. B Consensus sequence of lexAi and lexA2 boxes of E. coli, S. typhi- murium, E. carotovora, P. putida and P. aeruginosa

the LexA operator sites are present at almost the same positions in all five bacterial strains. There are 5 bp be- tween the two lexA operators of the three enterobacteria and 3 bp between those of the pseudomonads.

The affinity of LexA for SOS operators can be ana- lysed by a statistical-mechanical approach in terms of the heterology index (2E), which is a measure of how far a particular sequence departs from the consensus sequence (Berg 1988). The lower the 2E value, the closer the fit to this consensus sequence. When the 2E of the SOS boxes of the heterologous lexA genes was calculated in comparison with the SOS consensus sequence of E. coli (Table 2), the left half-site of the first SOS box of the lexA genes showed lower values than the other half- sites. Moreover, with the exception of E. coli, the 2E of the first lexA box was also lower than the AE of

the second. All these data indicate that all five lexA genes have a very similar lexA1 box. In fact, the sequence of the first nine bp of the lexAi box is entirely conserved in all five lexA genes (Fig. 5 B). On the other hand, the consensus sequence of the lexA2 box is less constrained than that of lexAi (Fig. 5B). With the exception of the lexA2 operator of S. typhimurium, which has a change at the first position, the CTGN10CAG motif is con- served in all lexA operators.

Comparison o f domains involved in the cleavage of LexA proteins

It has been shown that the sites required for cleavage of the LexA repressor lie within the carboxy-terminal two-thirds of this protein (Lin and Little 1988). Related to this, there are three regions in which all lexA ( Ind- ) mutations so far obtained are clustered. Region 1 in- cludes amino acid residues 80 to 85, region 2 includes amino acid residues 114 to 120, and region 3 includes Thr-154 and Lys-156. The A l a - 8 4 - Gly-85 bond, which is the point where the cleavage of the LexA protein takes place (Horii et al. 1981 b), is found in region 1. Regions 2 and 3 contain, respectively, residues Ser-119 and Lys- 156, which participate in the cleavage of Ala -84-Gly-85 by a mechanism similar to that of a serine protease (Sli- laty and Little 1987). Figure 6 shows the sequence of these regions in the LexA proteins of S. typhimurium, E. carotovora, P. aeruginosa and P. putida. It must be pointed out that the S. typhimurium and E. carotovora proteins not only conserve Ala-84 and Gly-85 but the residues upstream and downstream of this point are also the same. Moreover, P. aeruginosa and P. putida also contain the A l a - G l y bond, but at positions 90-91 and 88-89, respectively. It has recently been reported that alteration of Gln-92 to Tyr, Phe or Trp in E. coli gives rise to an increase in the rate of LexA cleavage (Smith et al. 1991). The LexA protein of S. typhimurium has Gln-92, whereas E. carotovora has a Glu residue in this position. However, a change of Gln-92 to Glu does not affect the behaviour of the LexA repressor of E. coli (Smith et al. 1991). The P. aeruginosa and P. putida se-

Table 2. Heterology index for the lexA1 and lexA2 boxes of Escheriehia coli, Sal- monella typhimurium, Erwinia carotovora, Pseudomonas putida and Pseudomonas aer- uginosa"

Bacterial species Box Sequence 2E1 b let ° 2E d

E. carotovora lexA1 E. carotovora lexA2 E. coli lexA1 E. coli lexA2 P. aeruginosa lexA1 P. aeruginosa lexA2 P. putida lexA1 P. putida lexA2 S. typhimurium lexA1 S. typhimurium lexA2

ACCTGTATATACTCACAGCA 1.7 6A 7.8 GACTGTATATACACCCAGGG 4.0 6.2 10.2 TGCTGTATATACTCACAGCA 2.0 5.7 7.7 AACTGTATATACACCCAGGG 0.5 6.5 7.0 CACTGTATATAATCCCAGTC 1.5 7.6 9.1 CACTGGATAAAAACACAGAG 4.6 4.8 9.4 TACTGTATATAATTCCAGTC 0.0 6.3 6.3 CACTGTACAAAAAGACAGAG 5.3 8.4 13.7 AACTGTATATACTCACAGCA 0.5 6.1 6.6 GATTGTATATACACCCAGGG 6.3 6.2 12.5

a All values were calculated according to Berg (1988) b Heterology index for the left half of the site ° Heterology index for the right half of the site a Heterology index for the whole site

ct2 ct3 Region 1

Ec 80 / GRVAAGEPLLAQQHIE / 95 St 80/GRVAAGEPLLAQQH IE / 95 Ecc 80/GRVAAGEPLLAQEHIE / 95 pp 84/GRVAAGAP ILAEQH IE / 99 Pa 86/GRVAAGAP I LAEQNIE / 101

******.*. **...**

tt

Region 2

EC 105 / FKPNADFLLRVSGMSMKDI GXMDGDLLAVHK/135 St 105 / FKPSADFLLRVSGMSMKDI GIMDGDLLAVHK / 135 Ecc 105 / FKP SADFLLRVSGMSMKNIGIMDGDLLAVHK / 135 pp 109 / FHPQADYLLRVHGMSMKDVGI FDGDLLAVHT / 139 Pa 111 / FNPRADYLLRVRGMSMKD IGILDGDLLAVHV / 141

t

Region 3

Ec 150/DDEVTVKRLKKQGNKVELLPENSE FKP IVVDLRQQS FT IEG/190 St 150/DDEVTVKRLKKQGNKVELLPENSEFTPIWDLREQSETIEG/190 Ecc 150/DDEVTVKRLKKQGNTVHLLAENEE FAP IVVDLRQQS FS IEG / 190 Pp 154/GDEVTVKRFKREGSKVWLLAENPEFAPIEVDLKEQELVIEG/194 Pa 156/GEEVTVKRFKREGSKVWLLAENPEFAP IEVDLKEQEL I IEG / 196

• .******.*..*..*#**.**#**#**#***..*..#***

t

Fig. 6. Alignment of the deduced LexA protein sequences of Escherichia coli (Ec), Salmonella typhimurium (St), Erwinia caroto- vora (Ecc), Pseudomonas putida (Pp) and Pseudomonas aeruginosa (Pa) 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 (*), simi- lar (.) or is not conserved (@). The number indicates the position in the amino acid sequence. Residues Ala-84, Gly-85, Ser-119 and Lys-156 are indicated (arrows). The positions of Ind- mutations isolated in E. coli are overlined

quences also show a Gln residue 7 amino acids from the A l a - G l y bond. Furthermore, the LexA repressors of S. typhimurium and E. carotovora have a Ser-ll9 with an identical surrounding region, whereas in P. putida and P. aeruginosa the corresponding amino acid is at positions 123 and 125, respectively. Likewise, the region containing Lys-156 shows a very conserved sequence in all five proteins, although this residue is at positions 160 and 162 in P. putida and P. aeruginosa, respectively. Thus, regardless of the precise position of the A l a - Gly bond and of the Lys and Ser residues, the distance be- tween these three points remains unaltered in all LexA proteins. It is also worth noting that the Glu-152 residue, which when changed to Ala produces a LexA protein with an increased rate of in vivo cleavage (Slilaty and Vu 1991), is conserved in all five organisms although its position is slightly displaced in P. aeruginosa and P. putida.

Conservation of the DNA binding domain in LexA proteins

The amino-terminal DNA binding domain of the LexA protein of E. coli shows some homology with those DNA binding proteins containing a 'helix-turn-helix' motif (Pabo and Sauer 1984). Indeed the LexA repressor of E. coli is known to have three e-helices, in the peptide segments comprising amino acids 8 20 (helix ~1), 28-35 (helix ~2) and 41 54 (helix ~3; Lamerichs et al. 1989). Furthermore, the region located between residues 61-67

131

ctl 20 28 35 41 54 Ec QQEVFDLIRDHI~TGMPr TRAE IAQRL 3FRS~ NAAEEHLKALARKG

1 St QQEVFDLIRDHI~TGMP~ TRAEIAQRL GFRS~ NAAEEHLKALARKG

1 pp QAE i LAF IKRCLE]DNGFP~ TRAE IAQEL 3FKS ~ NAAEEHLKALARKG

Pa )AE ILSFIKRCL~HGEP~ PRAE IAQEL 3FKS~ NAAEEHLKALARKG

Ecc QQQVYDLIRDHIA~TGMPI PRAEIAQQL 3FRS~ NAAEEHLKALARKG * . . . # . . * . # # . . . # * . * * * * * * * * * # * * * . * * * * * * * * * * * * * * * *

E1

Fig. 7. Comparison of the three c~-helices involved in the DNA- binding ability of the LexA proteins of Escherichia coli (Ec), Salmo- nella typhimurium (St), Erwinia earotovora (Ecc), Pseudomonas puti- da (Pp) and Pseudomonas aeruginosa (Pa). E1 indicates the presence of a possible extended conformation The symbols indicate whether a position in the alignment is perfectly conserved (*), similar (.) or is not conserved (~)

also seems to be involved in the DNA-binding capacity of the LexA protein of E. coli (Oertel-Buchheit et al. 1990). Comparison of all these regions in LexA of E. carotovora, S. typhimurium, P. aeruginosa, P. putida and E. coli (Fig. 7) shows that the conformation of re- gions ~2 and c~3 is identical in the five bacterial species studied here. In other words, helices ~2 and ~3 are pres- ent in the LexA protein of all five bacteria. Additionally, ~3 helix is perfectly conserved in all cases, whereas the cd region is 100% identical between E. coli and S. typhi- murium, and 88% between E. coli and E. carotovora, P. putida and P. aeruginosa. Likewise, the secondary structure of the segment comprising the el helix is the same for E. coli, S. typhimurium, P. putida and P. aeru- ginosa. However, the similarity of the sequence of this :d helix between E. coli, P. putida and P. aeruginosa is only 23%. On the other hand, computer analysis sug- gests that this region may have an extended conforma- tion in the LexA protein of E. carotovora (Fig. 7). The fragment that spans positions 61-67 does not differ in the five LexA proteins (Fig. 4). Finally, it is also worth noting that residues of the LexA repressor of E. coli in which LexA(Def) mutations have been isolated, such as Thr-5, Gly-23, Pro-25, Pro-26 and Pro-40, are fully retained in the five LexA repressors.

Discussion

The results presented in this paper demonstrate that the lexA gene and protein are highly conserved in gram- negative bacteria. In the closely related S. typhimurium bacterium, the DNA sequence of the lexA gene differed from that of E. coli K12 by 83 base changes, which only produced 6 amino acid substitutions in the LexA pro- tein. The more distantly related gram-negative bacteri- um E. carotovora encodes a LexA protein that, neverthe- less, is remarkably similar to that of E. coli K12. Fur- thermore, the lexA gene from P. putida and P. aerugino- sa is 59% and 61% identical to that of E. coli. These similarity values are not usual between these genera, be- cause, for instance, the degree of identity of the tufA gene of E. coli with that of S. typhimurium, E. caroto- vora, P. aeruginosa and P. putida is 85%, 40%, 25%

132

and 25%, respectively (Filer et al. 1981). Likewise, the similarity of the TrpC, TrpD, TrpE and TrpG proteins of P. aeruginosa and P. putida with these same proteins of E. coli is about 30% in all cases (Essar et al. 1990). Thus, the similarity of the lexA genes of S. typhimurium, E. carotovora, P. aeruginosa and P. putida with the IexA gene of E. coli is considerably higher than the average relatedness among other homologous genes in all these bacteria. The identity values of the lexA gene of E. caro- tovora and P. aeruginosa cited above are very similar to those found between the recA gene and protein of these bacteria and those of E. coli (80% and 91%, and 65% and 71%, respectively) (Zhao and McEntee 1990; Sancar et al. 1980; Sano and Kageyama 1987).

The lexA boxes of the lexA genes reported here show strong conservation of the regions involved in the con- trol of lexA gene expression in the five bacteria. Further- more, our results demonstrate a high level of conserva- tion of the first lexA box and are in agreement with those previously published indicating that the left half of this first lexA box is very important in the process of LexA repressor binding (Schnarr et al. 1991). The lexA gene of S. typhimurium has a substitution at posi- tion + 1 of the lexA2 operator, although the significance and importance of this change, if any, remain to be eluci- dated. Moreover, it is interesting to note that in the P. aeruginosa and P. putida lexA genes there are only 3 bp between the two lexA boxes, whereas in E. coli, E. carotovora, and S. typhimurium this distance is 5 bp. The possible effect of this on the in vivo expression of the lexA gene in the two members of the Pseudomonas genus requires further investigation.

Our data also provide support for the idea that the CTGT motif at the 5' end of the lexA box is of primary importance for the LexA-mediated repression of SOS genes (Schnarr et al. 1991) since all lexA boxes found, with the sole exception of lexA2 in S. typhimurium pos- sess these 4 bp. Besides, as in E. coli, the lexA operators of all lexA genes overlap the promoter of this gene.

The LexA repressor of E. coli is composed of two domains linked by a hinge region that is thought to be rather flexible (Little 1991). Residues required for the cleavage reaction must be located in the carboxy- terminal two-thirds of this protein, since proteolytic fragments containing this portion catalyze the cleavage reaction as efficiently as does the intact protein (Little 1984). Other evidence indicates that the carboxy-termi- nal domain is also important for repressor dimerization (Schnarr et al. 1988), and that the amino-terminal por- tion, comprising the first 84 amino acids, is the DNA-

:binding domain (Hurstel et al. 1986). Cleavage of the LexA repressor of E. coli takes place in the hinge, be- tween amino acids Ala-84 and Gly-85, separating the dimerization domain from the DNA-binding domain and inactivating repressor function. This DNA-binding domain of the LexA protein of E. coli contains at least three e-helices. Helix cd, extending from Gln-8 to Ser- 20; helix c~2, from Arg-28 to Leu-35; and helix c~3, from Ash-41 to Gly-54 (Lamerichs et al. 1989). The first helix folds over helix ~2 and helix c~3, since Val-ll is close to Leu-35, Phe-37 and His-46, but the exact spatial rela-

tionship has not as yet been determined (Schnarr et al. 1991). Our data showing that the e3 helix is present in all five LexA repressors with the same sequence in all of them indicate that this region must play a very important role in the DNA-binding process, despite there being only a similarity of 40% with the analogous region of DinR of B. subtilis (Raymond-Denise and Guillen 1991). However, because the DNA-binding spe- cifities of LexA and DinR are not the same (Raymond- Denise and Guillen 1991), significant differences be- tween them are not unexpected. Helix e2 is also present in all bacteria although its sequence homology between E. coli on the one hand, and E. carotovora, P. aerugino- sa, and P. putida on the other, is slightly lower (about 87%). The similarity with DinR at these points is only 20%. Moreover, the region cd does not seem to be as important since, despite the fact that all bacteria, with the exception of E. carotovora, have a helical conforma- tion in this fragment, the similarity of the sequence ranges from 76% for E. carotovora to 23% in P. aerugin- osa and P. putida. However, the sequence of ~1 from these last two bacteria show 46% similarity with the corresponding region of the DinR protein. Moreover, it is worth noting that the Thr-5, Gly-23, Pro-25, Pro-26 and Pro-40 residues, which most frequently give rise to a lexA(Def) phenotype when mutated (Oertel-Buchheit et al. 1990; Thliveris et al. 1991), are fully conserved in all five LexA proteins, although only positions Gly-23, Pro-25 and Pro-26 are preserved in DinR. Similarly, the region comprising amino acids 61 67 of the LexA pro- tein, which has been proposed to be involved in DNA- binding ability (Oertel-Buchheit et al. 1990), has the same amino acid sequence in E. coli, S. typhimurium, E. carotovora, P. aeruginosa and P. putida, strongly sug- gesting that these residues have a functional role. Next to this highly conserved region, the LexA proteins of P. aeruginosa and P. putida have an insertion of four and six amino acids, respectively, relative to LexA pro- teins from enterobacteria. It is interesting that the DinR protein also has an insertion of six residues at this same point. The fact that these additions are, in all cases, far from positions 68 to 74, is in agreement with the idea that the LexA repressor has two clearly delimited domains located at each end ot the protein, these do- mains being linked by a non-functional region. In this connection, it has already been suggested that the N- terminal domain of the LexA repressor of E. coli ends at or near residue 67 (Little 1984). As for the carboxy- terminal domain, Ind- mutations in the lexA gene have been found to cluster in three regions : around the cleav- age site, and around Set-119 and Lys-156 (Lin and Little 1988). Our data show that the amino acid sequence of these regions is fully conserved in LexA proteins of E. carotovora and S. typhimurium. Similarly, these features are also present in P. aeruginosa and P. putida although their precise positions are slightly displaced. All these data demonstrate that the several regions of the lexA gene involved either in the control of expression or en- coding the cleavage and DNA-binding sites are highly conserved in bacteria from two distantly related families such as Enterobacteriaceae and Pseudomonadaceae.

133

Acknowledgements. This work was supported by grants PB88-0246 and PB91-0470 of the Comisi6n Interministerial de Ciencia y Tec- nologia (CICYT), Spain. Sebasti/m Calero and Xavier Garriga were supported by predoctoral fellowships from the Spanish Min- istry of Education and Science and the Autonomous University of Barcelona, respectively.

References

Berg OG (1988) Selection of DNA binding sites by regulatory proteins : the LexA protein and the arginine repressor use differ- ent strategies for functional specificity. Nucleic Acids Res 16:5089-5105

Brent R, Ptashne M (1981) Mechanism of action of the lexA gene product. Proc Natl Acad Sci USA 78:4204-4208

Burekhardt SE, Woodgate R, Scheuermann RH, Echols H (1988) UmuD mutagenesis protein of Escherichia coli: overproduc- tion, purification and cleavage by RecA. Proc Natl Acad Sci USA 85:1811-1815

Calero S, Garriga X, Barb6 J (1991) One-step cloning system for isolation of bacterial lexA-like genes. J Bacterio1173:7345-7350

Clark D J, Maaloe O (1967) DNA replication and the division cycle of Escherichia coll. J Mol Biol 23:99-112

Craig NL, Roberts JW (1980) E. coli recA protein-directed cleavage of phage 2 repressor requires polynucleotide. Nature 283 : 26-30

Deretic V, Konyecsni WM, Mohr CD, Martin DW, Hibler NS (1989) Common determinators of promoter control in Pseudo- monas and other bacteria. Bio/Technology 7 : 1249 1254

Essar DW, Eberly L, Crawford IP (1990) Evolutionary differences in chromosomal locations of four early genes of the tryptophan pathway in fluorescent pseudomonads: DNA sequences and characterization of Pseudomonas putida trpE and trpGDC. J Bacteriol 172:867-883

Fernandez de Henestrosa AR, Calero S, Barb6 J (1991) Expression of the recA gene of Eseherichia coli in several species of gram- negative bacteria. Mol Gen Genet 226: 503-506

Filer D, Dhar R, Furano AV (1981) The conservation of DNA sequences over very long periods of evolutionary time. Eur J Biochem 120:69-77

Harley CB, Reynolds RP (1987) Analysis of E. eoli promoter se- quences. Nucleic Acids Res 15 : 2343-2361

Hill SA, Little JW (1988) Allele replacement in Escheriehia coli by use of a selectable marker for resistance to spectinomycin: replacement of the lexA gene. J Bacteriol 170:5913-5915

Horii T, Ogawa T, Ogawa H (1981a) Nucleotide sequence of the lexA gene of E. coli. Cell 23 : 689-697

Horii T, Ogawa T, Nakatani T, Hase T, Matsubara H, Ogawa H (1981 b) Regulation of SOS functions: purification of E. coli LexA protein and determination of its specific site cleaved by the RecA protein. Cell 27:515-522

Hurstel S, Granger-Schnarr M, Daune M, Schnarr M (1986) In vitro binding of LexA repressor to DNA: evidence for the in- volvement of the amino-terminal domain. EMBO J 5 : 793-798

Kim B, Little JW (1992) Dimerization of a specific DNA-binding protein on the DNA. Science 255:203-206

Lamerichs RMJN, Padilla A, Boelens R, Kaptein R, Ottleben G, Riiterjans H, Granger-Schnarr M, Oertel P, Schnarr M (1989) the amino-terminal domain of LexA repressor is a-helical but differs from canonical helix-turn-helix proteins: a two-dimen- sional H NMR study. Proc Natl Acad Sci USA 86:6863-6867

Lieberman HB, Witkin EM (1981) DNA degradation. UV sensitiv- ity and SOS-mediated mutagenesis in strains of Escherichia coli deficient in single-strand DNA binding protein: effects of muta- tions and treatments that alter levels of exonuclease V or RecA protein. Mol Gen Genet 190:92-100

Lin LL, Little JW (1988) Isolation and characterization of non- cleavable (Ind-) mutants of the LexA repressor of Eseherichia coli K-12. J Bacteriol 170:2163-2173

Little JW (1984) Autodigestion of LexA and phage 2 repressors. Proc Natl Acad Sci USA 81:1375-1379

Little JW (1991) Mechanism of specific LexA cleavage: autodiges- tion and the role of RecA coprotease. Biochimie 73:411-422

Little JW, Mount DW (1982) The SOS regulatory system of Escher- ichia coli. Cell 29 : 11-22

Little JW, Edmiston SH, Pacelli LZ, Mount DW (1980) Cleavage of the Eseherichia coli LexA protein by the RecA protease. Proc Natl Acad Sci USA 77: 3225-3229

Little JW, Mount DW, Yanisch-Perron CR (1981) Purified lexA protein is a repressor of the recA and lexA genes. Proc Natl Acad Sci USA 78:4199-4203

Lloubes R, Granger-Schnarr M, Lazdunski C, Schnarr M (1991) Interaction of a regulatory protein with a DNA target contain- ing two overlapping binding sites. J Biol Chem 266:2303-2312

Markham BE, Little JW, Mount DW (1981) Nucleotide sequence of the lexA gene of Escherichia coli K-12. Nucleic Acids Res 9:4149-4161

Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Miller RV, Kokjohn TA (1990) General microbiology of recA. Environmental and evolutionary significance. Annu Rev Mi- crobiol 44: 365-394

Nohmi T, Battista JR, Dodson LA, Walker GC (1988) RecA-me- diated cleavage activates UmuD for mutagenesis: Mechanistic relationship between transcriptional derepression and post- translational activation. Proc Natl Acad Sci USA 85:1816-1820

Oertel-Buchheit P, Lamerichs RMJN, Schnarr M, Granger- Schnarr M (1990) Genetic analysis of the LexA repressor: Isola- tion and characterization of LexA(Def) mutant proteins. Mol Gen Genet 223 : 40-48

Pabo CO, Sauer RT (1984) Protein-DNA recognition. Annu Rev Biochem 53 : 293-321

Raymond-Denise A, Guillen N (1991) Identification of dinR, a DNA damage-inducible regulator gene of Bacillus subtilis. J Bacteriol 173:7084-7091

Roca AI, Cox MM (1990) The RecA protein: structure and func- tion. CRC Crit Rev Biochem Mol Biol 25:415 455

Sambrook J, Fritsch EF, Mauiatis T (1989) Molecular cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Sancar A, Stachelek C, Konigsberg W, Rupp WD (1980) Sequences of the recA gene and protein. Proc Natl Acad Sci USA 77:2611 2615

Sanger F, Nicklen S, Coulson A (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74: 5463-5467

Sano Y, Kageyama M (1987) The sequence and function of the recA gene and its protein in Pseudomonas aeruginosa PAO. Mol Gen Genet 208:412-419

Sassanfar M, Roberts JW (1990) Nature of the SOS-inducing signal in Eseherichia coli. J Mol Biol 212:79-96

Sauer RT, Yocum RR, Doolittle RF, Lewis M, Pabo CO (1982) Homology among DNA-binding proteins suggests use of a con- served super-secondary structure. Nature 298:447-451

Schnarr M, Pouyet J, Granger-Schnarr M, Daune M (1985) Large- scale purification, oligomerization equilibria, and specific inter- action of the LexA repressor of Eseherichia coli. Biochemistry 24:2812-2818

Schnarr M, Granger-Schnarr M, Hurstel S, Pouyet J (1988) The carboxy-terminal domain of the LexA repressor oligomerises essentially as the entire protein. FEBS Lett 34:56-60

Schnarr M, Oertel-Buchheit P, Kazmaier M, Granger-Schnarr M (1991) DNA binding properties of the LexA repressor. Biochi- mie 73:423-431

Shinagawa H, Iwasaki H, Kato T, Nakata A (1988) RecA protein- dependent cleavage UmuD protein and SOS mutagenesis. Proc Natl Acad Sci USA 85:1806-1810

Shine J, Dalgarno L (1975) Determinant of cistron specificity in bacterial ribosomes. Nature 254:34-38

134

Slilaty SN, Little JW (1987) Lysine-156 and serine-119 are required for LexA repressor cleavage: a possible mechanism. Proc Natl Acad Sci USA 84:3987-3991

Slilaty SN, Vu KH (1991) The role of electrostatic interactions in the mechanism of peptide bond hydrolysis by a Ser-Lys catalytic dyad. Pr0t Eng 4:919-922

Smith M, Cavenagh MM, Litthe JW (1991) Mutant LexA proteins with an increased rate of in vivo cleavage. Proc Natl Acad Sci USA 88:7356-7360

Thliveris AT, Little JW, Mount DW (1991) Repression of the E. co- li recA gene requires-at least two LexA protein monomers. Biochimie 73 : 449-455

Walker GC (1984) Mutagenesis and inducible responses to deoxyri- bonucleic acid damage in Eseherichia eoli. Microbiol Rev 48 : 60-93

Weinstock GM, McEntee K (1981) RecA protein dependent pro- teolysis of bacteriophage lambda repressor. Characterization of the reaction and stimulation by DNA-binding proteins. J Biol Chem 256:10883-10888

Wertman KF, Mount DW (1985) Nucleotide sequence binding specificity of the LexA repressor of Escherichia coli K-12. J Bacteriol 163 : 376-384

Zhao X, McEntee K (1990) DNA sequence analysis of the reeA genes from Proteus vulgars, Erwinia carotovora, Shigella flexneri and Escherichia coli B/r. Mol Gen Genet 222:369-376

C o m m u n i c a t e d by R. Devore t