Induction of the SOS regulon of Haemophilus influenzae does not affect phase variation rates at tetranucleotide or dinucleotide repeats

Wendy A. Sweetman, E. Richard Moxon and Christopher D. Bayliss

Paediatric Molecular Infectious Diseases Group, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DS, UK

Correspondence
Wendy A. Sweetman
wendy.sweetman{at}paediatrics.ox.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Haemophilus influenzae has microsatellite repeat tracts in 5' coding regions or promoters of several genes that are important for commensal and virulence behaviour. Changes in repeat number lead to switches in expression of these genes, a process referred to as phase variation. Hence, the virulence behaviour of this organism may be influenced by factors that alter the frequency of mutations in these repeat tracts. In Escherichia coli, induction of the SOS response destabilizes dinucleotide repeat tracts. H. influenzae encodes a homologue of the E. coli SOS repressor, LexA. The H. influenzae genome sequence was screened for the presence of the minimal consensus LexA-binding sequence from E. coli, CTG(N)10CAG, in order to identify genes with the potential to be SOS regulated. Twenty-five genes were identified that had LexA-binding sequences within 200 bp of the start codon. An H. influenzae non-inducible LexA mutant (lexANI) was generated by site-directed mutagenesis. This mutant showed increased sensitivity, compared with wild-type (WT) cells, to both UV irradiation and mitomycin C (mitC) treatment. Semi-quantitative RT-PCR studies confirmed that H. influenzae mounts a LexA-regulated SOS response following DNA assault. Transcript levels of lexA, recA, recN, recX, ruvA and impA were increased in WT cells following DNA damage but not in lexANI cells. Induction of the H. influenzae SOS response by UV irradiation or mitC treatment did not lead to any observable SOS-dependent changes in phase variation rates at either dinucleotide or tetranucleotide repeat tracts. Treatment with mitC caused a small increase in phase variation rates in both repeat tracts, independently of an SOS response. We suggest that the difference between H. influenzae and E. coli with regard to the effect of the SOS response on dinucleotide phase variation rates is due to the absence of any of the known trans-lesion synthesis DNA polymerases in H. influenzae.


Abbreviations: mitC, mitomycin C; TLS, trans-lesion synthesis


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Pathogenic bacteria are faced with the challenge of surviving in hostile host environments. Heritable adaptation to environmental challenges is achieved through acquisition of mutations in the DNA sequence. Some pathogenic bacteria have developed mechanisms that exploit the infidelities of the replication machinery to achieve elevated mutation rates at specific sites of repetitive DNA sequence (microsatellites) associated with loci that are important for adaptation to changes in the host environment (contingency loci) (Moxon et al., 1994). Replication slippage that leads to changes in the number of repeats at these loci can cause phase-variable switching of expression of the genes, thus aiding rapid adaptation to changing host environments.

Haemophilus influenzae is a Gram-negative, obligate commensal of the upper human respiratory tract with the potential to cause diseases such as meningitis, septicaemia, epiglottitis, pneumonia and otitis media. H. influenzae has microsatellite regions associated with the promoter or 5' coding region of several contingency loci that have roles in virulence and disease. A dinucleotide (TA) repeat tract is present in the promoter region of the pilin locus (van Ham et al., 1993), heptanucleotide repeats are associated with the HMW adhesin genes (Dawid et al., 1999), whilst tetranucleotide repeat sequences are present in loci whose functions include iron acquisition, biosynthesis of LPS epitopes and a putative type III restriction–modification (mod) gene (Hood et al., 1996).

Several studies have sought to determine factors that affect the frequency of replication slippage in microsatellite sequences. Cis factors (number of repeats) and trans factors (Pol I, RNase H, mismatch repair mechanisms) have been shown to have an effect on phase variation frequencies in H. influenzae (De Bolle et al., 2000; Bayliss et al., 2002, 2004, 2005). Interestingly, Morel et al. (1998) showed that, in Escherichia coli, induction of the SOS response greatly influenced the replication slippage rates of a dinucleotide repeat tract, increasing slippage rates by 8- to 22-fold, depending on the repeat sequence.

The SOS response was first described in E. coli and is a temporary, co-ordinated increase in transcription of a network of genes, the products of which are involved in DNA repair, replication restart and cell-division control. More than 40 genes have now been shown to be induced during the SOS response in E. coli (Fernández de Henestrosa et al., 2000; Courcelle et al., 2001; Khil & Camerini-Otero, 2002). Two proteins, RecA, the positive regulator, and LexA, the repressor, control induction of the SOS response. Single-stranded (ss) DNA accumulates at stalled replication forks following DNA damage (Sassanfar & Roberts, 1990) and is bound by RecA, forming a nucleoprotein filament that then stimulates LexA to undergo self-cleavage (Little, 1993). Under non-inducing conditions, LexA binds within the promoter regions of genes to a site referred to as the SOS box (Little & Mount, 1982; Walker, 1984).

Recently, Erill et al. (2003) looked in silico for the presence of SOS boxes in the published genome sequences of several gammaproteobacterial species. Only three genes (lexA, recA and recN) were found to have an SOS box present in the promoter region in all the organisms analysed, indicating that, although the SOS box sequence and RecA/LexA controlling mechanism are conserved within this subclass, the number of genes included in the LexA regulon varies considerably such that individual species may differ in the way in which they survive and repair DNA damage.

Zulty & Barcak (1993) showed that the recA gene of H. influenzae was induced approximately 3-fold in response to SOS-activating signals, but no detailed studies of the LexA regulon of H. influenzae have been performed. Some differences between the SOS regulons of H. influenzae and E. coli are apparent. For example, the SOS repair mechanisms of E. coli are associated with increased mutational frequencies due to the induction of three trans-lesion synthesis (TLS) polymerases, Pol II, IV and V, respectively encoded by the polB, dinB and umuDC genes (Napolitano et al., 2000; Wagner et al., 2002). Notani & Setlow (1980) noted that prophage induction could be stimulated in H. influenzae by exposure to UV light but, in contrast to E. coli, with no associated increase in mutation rates. Annotation of the entire H. influenzae genome sequence (Fleischmann et al., 1995) reveals the apparent absence of any TLS polymerases.

The aim of this work was to determine whether or not induction of the SOS response influences phase variation rates in H. influenzae. In order to achieve this aim, the published H. influenzae Rd KW20 genome sequence (Fleischmann et al., 1995) was first screened for genes that have the E. coli minimal consensus SOS-box sequence CTG(N10)CAG in their promoter region in order to identify genes with the potential to be LexA regulated. Secondly, a non-inducible LexA mutant (lexANI) was constructed and shown to have increased sensitivity to both irradiation with a UV light source and treatment with mitomycin C (mitC). Thirdly, mRNA levels of candidate SOS genes in wild-type (WT) RM118 and lexANI mutant strains of H. influenzae were analysed following exposure to UV and mitC, confirming the presence of a LexA-dependent SOS response in WT H. influenzae strains and disabling of the SOS response in the engineered mutants. Finally, we investigated whether induction of the SOS response in H. influenzae had an effect on phase variation events, mediated by slippage at either dinucleotide (AT) or tetranucleotide (AGTC) repeat tracts.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains.
H. influenzae strain RM118 (a derivative of strain Rd) was used to generate the isogenic lexANI mutant strains G91D-1 and G91D-2. The isogenic recA– strain, in which the recA gene is disrupted by a tetracycline-resistance cassette, has been described elsewhere (Bayliss et al., 2002). All strains were grown at 37 °C in liquid brain heart infusion (BHI) broth supplemented with haemin (10 µg ml–1) and NAD (2 µg ml–1) or on BHI agar plates supplemented with Levinthal's reagent. When appropriate, tetracycline was added to cultures and plates at a concentration of 3 µg ml–1. X-Gal was added to plates at 40 µg ml–1. E. coli strain JM109 was used to propagate plasmids and was grown at 37 °C in Luria–Bertani broth supplemented with tetracycline at a concentration of 12 µg ml–1.

Screening of the H. influenzae genome sequence for LexA-binding boxes.
Each occurrence of the minimal E. coli SOS-box consensus sequence, CTG(N10)CAG, in the genome sequence of H. influenzae Rd KW20 was determined from a search of the published sequence (Fleischmann et al., 1995) using a search algorithm within an Artemis database. Each SOS box was checked to determine its position relative to the nearest open reading frames (ORFs) in order to assign it as being located in a promoter (here defined as being within 200 bp upstream of an initiation codon), coding or non-coding region.

Generation of LexANI mutants and reporter strains.
The lexA gene, together with approximately 800 bp of 5' DNA and 950 bp of 3' DNA, was amplified from strain RM118 genomic DNA as two fragments using the pairs of primers lex1EcoRI/lex5HindIII and lex6HindIII/lex7BamHI, respectively (Table 1; Fig. 1a). The HindIII sites generated by primers lex5HindIII and lex6HindIII were restricted with HindIII and the two fragments were ligated together to generate the full-length gene construct with an engineered HindIII site immediately downstream of the stop codon of the lexA coding region. This full-length gene construct was then restricted with the appropriate enzymes and cloned into the EcoRI and BamHI sites of plasmid pUC19. A HindIII fragment containing a tetracycline-resistance cassette (derived from plasmid pHVT1; Danner & Pifer, 1982) was then inserted into the engineered HindIII site generating plasmid pLexAWTtet.


View this table:
[in this window]
[in a new window]
 
Table 1. Primers used in these studies

Locus numbers are as given on the TIGR microbial database (http://www.tigr.org).

 


View larger version (12K):
[in this window]
[in a new window]
 
Fig. 1. (a) Generation of plasmid pLexAWTTet and subsequent site-directed mutagenesis to generate plasmid pLexAG91DTet. Boxes represent coding regions and the thin single line shows the genomic DNA strand. Horizontal arrows show the direction of amplification from primers; raised portions of lines indicate non-homologous regions of primers used to engineer restriction sites. Single-letter amino acid codes represent relative positions of conserved amino acids at the active site and cleavage site of the LexA protein. Details of primer sequences and cloning methods are given in Table 1 and Methods. (i) Amplification of lexA and flanking regions as two fragments generating HindIII sites; (ii) restriction of PCR products with HindIII and subsequent cloning and ligation into pUC19 to generate lexA with HindIII site immediately 3' of the stop codon; (iii) insertion of TetR cassette to generate pLexAWTTet; and (iv) use of pLexAWTTet as the template for site-directed mutagenesis (SDM) to generate pLexAG91DTet. Mismatches in site-directed mutagenesis primers that result in the sequence change leading to replacement of Gly by Asp at the cleavage site are shown as inflections in the primer line. (b) Nucleotide and amino acid sequence changes resulting from site-directed mutagenesis. Nucleotides 262–279 of the H. influenzae lexA gene and the corresponding amino acid sequence (residues 88–93) of the LexA protein are given for Rd strain KW20, RM118 and the lexANI mutant strains G91D-1 and G91D-2 generated in this study. Underlined nucleotides in RM118 indicate where this sequence differs from the published KW20 sequence. This polymorphism would result in replacement of Arg by Ala (italics). Bases altered by site-directed mutagenesis to generate lexANI mutants are shown in bold. These changes result in replacement of Gly by Asp at the cleavage site.

 
pLexAWTtet was used as the template for site-directed mutagenesis with the Stratagene Quikchange site-directed mutagenesis kit and primers GDF and GDR (Table 1) to engineer the substitution of Asp for Gly91 in the lexA coding region. The plasmid pLexAG91Dtet was retrieved following the mutagenesis reaction and the insert was sequenced to confirm that the only mutation was the desired mutation at Gly91 (data not shown). Automated DNA sequencing was performed using Big-Dye (Perkin Elmer) sequencing kits and an ABI Prism 377 (Perkin Elmer) autosequencer.

pLexAG91Dtet was linearized by digestion with SalI and used to transform strain RM118 by the method of Herriott et al. (1970). Transformants were selected on BHI tetracycline plates and allelic replacement of the WT lexA gene with the mutated lexANI allele was confirmed by PCR, Southern analyses and DNA sequencing (data not shown). Two independent lexANI mutants, G91D-1 and G91D-2, were selected for use in these studies. Reporter plasmids containing the coding region of the H. influenzae mod that is 5' of the repeat site, fused, in-frame, to a repeat region consisting of AGTC or AT repeats fused, in turn, to a lacZ gene have been described elsewhere (De Bolle et al., 2000; Bayliss et al., 2002). Slippage at the repeat tracts in these constructs results in gain or loss of expression from the lacZ gene. The lexANI strains G91D-1 and G91D-2 were further transformed with reporter plasmids that contained either 38 AGTC or 20 AT repeats in order to create reporter strains for phase variation experiments that also carried the lexANI mutation. The plasmids were linearized by digestion with SalI prior to transformation. Transformants G91D-1AGTC38, G91D-1AT20, G91D-2AGTC38 and G91D-2AT20 were selected for use in phase variation experiments.

Determination of UV sensitivity.
Overnight bacterial cultures were diluted 10-fold in BHI broth and grown to mid-exponential phase (OD650 0·2–0·4). Cells were diluted 1 : 105 in BHI and 25–50 µl aliquots were spread onto BHI plates. Following recovery at 37 °C for 1 h, plates were exposed, with lids off, to a model UV6-54 mineralight lamp (UV Products Inc.) 254 nm UV light source at a distance of 16 cm for 1–5 s (calibration using a Blak-Ray model J-22S short-wave UV meter showed this to be equivalent to a dose of 4 J m–2 s–1). The number of colonies that survived the exposure to UV light was counted after overnight incubation at 37 °C and percentage survival was calculated relative to the number of colonies obtained from unexposed samples of the same cultures.

mitC-sensitivity experiments.
Overnight cultures were diluted 10-fold to a final volume of 20 ml BHI broth, incubated for 2 h at 37 °C (estimated to be mid-exponential phase) and split into four 5 ml aliquots. A 2 mg mitC ml–1 stock in DMSO was diluted to a working concentration of 0·02 mg ml–1 in DMSO and added to the split cultures to give final concentrations of 0, 0·1, 0·05 and 0·01 µg ml–1. Cultures were then incubated at 37 °C for 30 min, serial dilutions were prepared and aliquots of 10–4 and 10–5 dilutions were plated onto BHI plates. The number of colonies that survived treatment was counted after overnight incubation at 37 °C and percentage survival was calculated relative to the number of colonies in the untreated sample.

RT-PCR analysis of H. influenzae SOS genes.
Induction of the SOS response, prior to RNA extraction and RT-PCR, was achieved by exposure to a UV light source or treatment with mitC as follows. For UV exposure, overnight bacterial cultures were diluted 10-fold in 20 ml BHI broth, incubated at 37 °C and harvested at mid- to late-exponential phase (OD650 0·4–0·8) by centrifugation at 1800 g for 10 min in a Sorvall RT7 centrifuge. The cell pellet was resuspended in 1–2 vols ice-cold PBS (10 µM phosphate buffer, 27 µM KCl, 137 µM NaCl)/10 % (v/v) BHI and 4 ml aliquots of this suspension were put into 8·5 cm Petri dishes and placed on a gently rotating platform beneath the UV light source. The dishes were exposed to UV light (4 J m–2 s–1), with lids off, for 4 s. Cells were allowed to recover at 37 °C for 20, 30 or 40 min before cells were again recovered by centrifugation as above. Cell pellets were resuspended in 100 µl TE (10 mM Tris/HCl, 1 mM EDTA, pH 7·9) and RNA was extracted using the Promega SV total RNA isolation kit. For mitC treatment, overnight bacterial cultures were diluted and regrown in the same way as for UV exposure. The 20 ml culture was split into two and mitC added to one half to a final concentration of 0·1 µg ml–1. Cultures were then incubated at 37 °C for a further 90 min. Aliquots (0·25 ml) from each culture were then taken, pelleted at 13 000 g for 3 min and resuspended in 100 µl TE. RNA was extracted as for UV exposure.

For RT-PCR analysis, 500 ng isolated RNA was reverse-transcribed with 1 µl random primers (500 µg ml–1; Promega) and 200 U Moloney murine leukaemia virus reverse transcriptase (Invitrogen) in a 25 µl total reaction volume. The equivalent of 1 µl undiluted cDNA was used in PCRs. PCR amplifications were performed in 1x PCR buffer (Invitrogen) containing 200 µM each of dATP, dCTP, dGTP and dTTP (Invitrogen), 1 µl of a 20 µM dilution of each of the target SOS gene primers and 1 µl of a 6 µM dilution of each of the control gene (frdB) primers (primer sequences are listed in Table 1). Cycling conditions were 94 °C for 1 min, 58 °C for 1 min and 72 °C for 1 min for 30 cycles. Amplifications were performed on untranscribed RNA samples to monitor for DNA contamination. If DNA was found to be present, RNA samples were treated with DNase I (Promega) and rescreened for DNA contamination prior to PCR amplification.

PCR products were separated on 2 % agarose gels. Gels were post-stained in 0·5 µg ethidium bromide ml–1 and band intensities were measured using Imagemaster1D image analysis software. Induction ratios were calculated from gel-band intensity measurements in the following manner. For each gel lane, the target gene band intensity was first divided by the frdB gene band intensity to obtain an expression ratio. Induction ratios were then calculated as the expression ratio of UV-exposed or mitC-treated cells divided by the expression ratio obtained from control, unexposed cells.

Northern blot analysis.
Overnight bacterial cultures were diluted 10-fold in 60–80 ml BHI and regrown to mid-exponential phase as before. Each culture was split into two and one half was treated with mitC (0·2 µg ml–1) for 1 h. RNA was extracted using an adaptation of the hot acid phenol method (Miller, 1972). In brief, cultures were pelleted at 6000 g at 4 °C for 5 min, washed once in cold PBS/azide (120 mM NaCl, 2·7 mM KCl, 10 mM phosphate buffer, 20 mM NaN3, pH 7·0) and then resuspended in 10 ml lysis buffer (20 mM sodium acetate, 0·5 % w/v SDS, 1 mM EDTA, pH 5·4). Twenty millilitres of phenol equilibrated in lysis buffer and pre-heated to 65 °C was added and mixed thoroughly. The extracts were then centrifuged at 6000 g at 20 °C for 10 min and the upper, aqueous phase was removed to a second clean, sterile tube. One millilitre of 1 M KCl was added with 2 vols ice-cold ethanol. Precipitates were allowed to form at –20 °C for 1–12 h and then pelleted at 10 000 g at 4 °C for 15 min. Pellets were drained and resuspended in 250 µl water treated with 0·1 % diethyl pyrocarbonate.

RNA (10–20 µg; 5–10 µl) was run on 1·2 % agarose, 3·7 M formaldehyde gels in 1x MOPS buffer (0·4 M MOPS/NaOH, 0·1 M sodium acetate, 10 mM EDTA, pH 7·2) and then transferred to nylon filters. Filters were placed in 25 ml pre-hybridization solution (5x SSPE, 50 % v/v formamide, 5x Denhardt's solution, 1 % w/v SDS, 100 µg heat-denatured salmon sperm DNA ml–1) at 42 °C for several hours. DNA probes were synthesized from the appropriate gel-purified PCR product using the Megaprime DNA-labelling system and [{alpha}-32P]dCTP (Amersham). PCR products were amplified from RM118 genomic DNA using primer pairs listed in Table 1 and the amplification conditions stated above. Hybridization was performed at 42 °C overnight in 10 ml hybridization solution (5x SSPE, 50 % v/v formamide, 5 % w/v dextran sulfate, 5x Denhardt's solution, 1 % w/v SDS). Filters were washed in 2x SSC at 20 °C for 15 min and then in 2x SSC, 0·1 % SDS (w/v) at 65 °C for 30 min before exposure to X-ray film.

Phase variation experiments.
The method for estimating phase variation frequencies using the mod–repeat–lacZ reporter gene constructs has been described previously (De Bolle et al., 2000; Bayliss et al., 2002). Adaptations of the published method were as follows: to determine the effects of UV induction, strains to be investigated were plated onto BHI plates containing 40 µg X-Gal ml–1 and incubated overnight at 37 °C. A single blue (lacZ-expressing) colony was picked, diluted and plated onto BHI/X-Gal plates. These plates were exposed to a UV light source (4 J m–2 s–1) for 4 s. Exposure occurred either immediately after or several hours after plating. To determine the effects of mitC induction, the single blue colony that was selected after overnight incubation was diluted and plated onto plates containing 0·01 µg mitC ml–1and incubated overnight at 37 °C. The concentration of mitC that was used was determined experimentally to give an observable reduction in colony size but still allow sufficient growth of the colonies for estimates of phase variation to be made. Induction of the SOS response by the mitC treatment was confirmed from RT-PCR analysis of RNA extracts prepared from cells harvested from mitC plates. Harvested cells were resuspended in 100 µl TE and RNA was extracted from them as described above.

Eight colonies were picked from either the UV-exposed or mitC-containing plates after overnight incubation and analysed as described previously (De Bolle et al., 2000; Bayliss et al., 2002) to allow estimation of phase variation frequencies. In brief, serial dilutions were made from selected colonies and aliquots of the 10–2–10–5 dilutions were plated. The number of c.f.u. present in the 10–4 and 10–5 dilutions was used to estimate total cell number in each of the selected colonies, whilst the total number of revertants was estimated from duplicate platings of the higher dilutions. The frequency of revertants for each of the eight colonies was calculated from the mean number of revertants divided by the total number of cells. The median value and range of frequencies from the eight colonies were recorded and used to calculate mutation rates according to the method of Drake (1991) and 95 % confidence intervals according to Kokoska et al. (1998). The changes in repeat number that resulted in the revertant phenotype were analysed as described previously (Bayliss et al., 2002). In brief, revertant colonies were picked and the repeat regions were PCR-amplified using fluorescent primers. PCR products were separated and sized by electrophoresis of the fragments using an ABI Prism 377 automatic sequencer and the ABI GeneScan 3.1 program (Perkin-Elmer).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Identification of potential LexA-regulated genes in the H. influenzae genome
A search of the published H. influenzae Rd KW20 genome sequence (Fleischmann et al., 1995) for the minimal E. coli SOS-box consensus sequence CTG(N10)CAG revealed 150 sites of exact homology. From the investigations of Fernández de Henestrosa et al. (2000), we determined that the minimum and maximum distances between functional SOS boxes and start codons of genes in E. coli were 24 bp for lexA and 170 bp for ssb, the gene encoding ssDNA-binding protein. We examined the H. influenzae SOS boxes and found that 26 were located within 200 bp of the start codon of a total of 25 genes. The majority of these boxes were in non-coding regions and are therefore likely to be within or close to promoter regions. These 25 genes are potential members of the H. influenzae SOS regulon and are listed in Table 2.


View this table:
[in this window]
[in a new window]
 
Table 2. Location of putative SOS boxes and LexA-regulated genes in the H. influenzae Rd KW20 genome

Locus names and numbers are as given on the TIGR microbial database (http://www.tigr.org); hypo, hypothetical protein; chypo, conserved hypothetical protein.

 
The seven H. influenzae genes listed in Table 2(a) (lexA, recA, recN, uvrA, ssb, ruvA and uvrD) also form part of the E. coli SOS response. They encode products involved either directly in repair of DNA damage or in regulation of DNA-repair systems. The remaining 18 putative H. influenzae SOS genes are listed in Table 2(b). Eight of these are annotated as hypothetical ORFs of unknown function (HI0325, HI0633, HI0650, HI0696, HI0858, HI0940, HI1254 and HI1423) and so it is difficult to assess the significance of their association with SOS boxes. Four of the remaining 10 genes (himD, recF, recJ and mfd) have known DNA processing and repair functions. Of the remaining six, tgt (involved in tRNA modification) was shown to be upregulated in a LexA-independent manner in E. coli following UV-induced damage of DNA in the microarray studies of Courcelle et al. (2001) and HI1079 shows homologies to the E. coli yecS gene. HI1079 and yecS both encode putative permease proteins and the downstream genes in each species encode ATP-binding proteins. In E. coli, yecS has a non-consensus SOS box but is not upregulated in response to DNA damage (Fernández de Henestrosa et al., 2000). The possible inclusion of fdhD, aroK and sprT in a LexA regulon is not intuitive from their functions: fdhD is involved in anaerobic metabolism and aroK in amino acid biosynthesis and sprT belongs to an uncharacterized family of metallopeptidases. The final gene, HI1546, is annotated as impA. This gene was first identified in an operon of three genes (impCAB), isolated from the I incompatibility group plasmid TP110 (Glazebrook et al., 1986). This operon encodes an error-prone SOS-dependent DNA polymerase that shows homologies to umuDC. With no other error-prone polymerases present in the H. influenzae genome, it is intriguing that a gene showing homology to a single subunit of this polymerase should be present and potentially under the control of LexA. Erill et al. (2003) also suggested mfd and impA as additions to the LexA regulon in H. influenzae from their in silico studies.

Engineering H. influenzae lexANI mutants
Four amino acid residues are essential for the self-cleavage reaction of LexA in E. coli. These are Ala84 and Gly85 at the cleavage site and Ser119 and Lys156 at the active site (Lin & Little, 1988). Mutation of any of these amino acids can prevent cleavage of LexA and result in mutants showing a non-inducible SOS phenotype (lexANI). The published Rd KW20 lexA sequence shows 67 % identity to the E. coli lexA sequence. The Rd KW20 lexA sequence differs from the E. coli sequence in that the conserved Ala and Gly residues that form the cleavage site are at positions 90 and 91, respectively, of the H. influenzae LexA protein. Sequencing of the lexA gene from strain RM118 used in these studies showed that there was a further polymorphism in this strain relative to the published strain KW20 sequence that would result in the substitution of Arg for Ala at position 89 (the E. coli sequence also has Arg at the equivalent position 83). Primers GD1F and GD1R (Table 1) were designed to accommodate this polymorphism and were used in site-directed mutagenesis reactions to generate a mutation in lexA that would result in Gly91 at the cleavage site being replaced by Asp (Fig. 1b). This mutation has previously been described in E. coli, resulting in a non-inducible LexA phenotype (lexANI) (Lin & Little, 1988). Following site-directed mutagenesis and transformation, two independent lexANI mutants, G91D-1 and G91D-2, in which the WT lexA allele had been replaced by the mutated allele, were selected for use in studies of the H. influenzae SOS response. The allelic replacement was confirmed by PCR, Southern blots and sequencing (data not shown).

Sensitivity of H. influenzae lexANI mutants to DNA damage
Fig. 2 shows that the H. influenzae lexANI mutants were more sensitive than WT cells to both UV exposure and mitC treatment but were less sensitive to these DNA-damaging treatments than recA– mutants. The recA– mutant was more sensitive (13·5 % survival) than the WT strain RM118 (88·3 % survival) following only 1 s exposure to UV light, whilst the lexANI mutants showed increased sensitivity only after >3 s of exposure, at which time the survival rates were similar to recA– mutants (37, 0·5, 0·44 and 0·7 % survival for strains RM118, G91D-1, G91D-2 and recA–, respectively). The lexANI and recA– mutants were more sensitive to mitC than WT, as shown by both a decrease in the number of surviving colonies and a reduction in the size of the colonies following overnight growth. The difference in survival rates between the WT and lexANI or recA– mutants is less marked following mitC treatment (4·2 % survival for WT and 0·52, 0·81 and 0·62 % survival for strains G91D-1, G91D-2 and recA–, respectively, using 0·05 µg mitC ml–1).



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 2. Comparison of sensitivities of RM118, recA– and lexANI strains to UV light and mitC. Graphs indicate percentage survival of Rd strain RM118, lexANI strains G91D-1 and G91D-2 and an RM118 recA strain following exposure to a UV light source (4 J m–2 s–1) (upper panel) or treatment with mitC (lower panel). Mid-exponential-phase cells were diluted and plated onto BHI plates, allowed to recover for 1 h and then exposed to the UV light source for the time indicated; mitC was added to mid-exponential-phase cultures, to the final concentration indicated, and cultures were incubated for a further 30 min before diluting and plating to allow estimation of numbers of surviving colonies. Error bars show standard deviations of the means of four (UV) or three (mitC) independent measurements.

 
Northern blot and RT-PCR analysis of the SOS response
In order to confirm that DNA damage caused induction of expression of putative SOS genes in H. influenzae and that this induction did not occur in the lexANI strains, the expression levels of a total of nine genes were investigated by analysis of changes in mRNA levels. Total RNA was isolated from cells following exposure to UV light or mitC and was analysed using Northern blot and RT-PCR techniques. The results of the investigations are shown in Fig. 3 and are summarized in Table 2.



View larger version (50K):
[in this window]
[in a new window]
 
Fig. 3. Northern blot and RT-PCR analysis of putative H. influenzae SOS gene expression levels. (a) Northern blot analysis of recA (upper panel) and ruvA (lower panel) transcript levels from strain RM118 and a lexANI mutant following mitC induction of the SOS response. –, Samples from uninduced cell; +, samples from cells that were exposed to 0·2 µg mitC ml–1. (b) Representative RT-PCR gel profiles obtained for putative LexA-regulated H. influenzae genes. The upper band in each lane is the result of amplification of the control gene frdB. The lower band is the target gene (indicated below each set of four lanes). The first two lanes in each set of four show the RT-PCR products obtained using RNA isolated from strain RM118 either without induction (lanes 1) or following UV induction (lanes 2; lexA, himD, recN and recF) or mitC induction (lanes 2; recA, recX, recJ and impA). The second two lanes in each set of four show the RT-PCR products obtained using RNA from non-inducible SOS mutants. Lanes 3 (lexA, himD, recN and recF) are from unexposed strain recA–; lanes 4 (lexA, himD, recN and recF) from strain recA– following UV induction; lanes 3 (recA, recX, recJ and impA) from uninduced lexANI mutant; and lanes 4 (recA, recX, recJ and impA) from lexANI mutant following mitC induction. (c) Induction ratios calculated for seven putative H. influenzae SOS genes as described in Methods. Values are plotted for RM118 following UV exposure (RM118 UV) and mitC exposure (RM118 mitC), for recA– strain following UV exposure (recA– UV) and for the lexANI strains following UV exposure (lexANI UV) and mitC exposure (lexANI mitC). Results of mitC exposure from culture and plates have been combined. The number of data points obtained for each gene is indicated below the x-axis. Error bars show the standard deviation of the induction ratios where three or more readings were averaged to obtain the value plotted.

 
Eight of the genes listed in Table 2 were chosen for investigation. Four genes, lexA, recA, recN and ruvA, were selected from the seven homologues of the known E. coli LexA-regulated genes. From the 18 additional H. influenzae genes associated with SOS boxes, recF, recJ and himD were selected because of their known roles in DNA processing and repair and impA was investigated because of its homology to a subunit of the UmuDC family of TLS polymerases. The recX gene that is situated 161 bp downstream of the H. influenzae recA gene (an organization that is observed in several bacterial species) was also investigated. In E. coli, recX is co-transcribed with recA (Pagès et al., 2003) and inhibits recombinase, co-protease and ATP hydrolysis activities of RecA (Venkatesh et al., 2002; Pagès et al., 2003; Stohl et al., 2003), indicating that recX may play an important role in SOS responses.

Northern blots were performed with RNA extracted from WT strain RM118 and the lexANI strains, grown in the presence or absence of mitC, and were probed to analyse expression patterns of recA and ruvA. Fig. 3(a) shows that message levels for recA and ruvA were increased following treatment with mitC in strain RM118 but not the lexANI strains, indicating that transcription of these genes is regulated in a LexA-dependent manner (the results from only one of the independent mutants is shown; the second mutant gave the same result). Analysis of other transcripts by this method, however, was not successful, and RT-PCR was adopted as the preferred method of analysis.

Typical gel profiles obtained from the RT-PCRs are shown in Fig. 3(b). For each putative SOS gene analysed, RT-PCRs were performed on mRNA extracted from WT strain RM118 either unexposed or exposed to UV light or mitC and compared to RT-PCRs performed on RNA extracted from either the recA– strain or one of the lexANI strains that were unexposed or exposed to the DNA-damaging treatments. Products from these four RT-PCRs were loaded in adjacent lanes on agarose gels (see Fig. 3b for examples). The ‘housekeeping’ gene frdB (fumarate reductase B) was amplified in each PCR (upper band in each track) as an internal control for the RT-PCR efficiency. Amplification of the frdB gene gave consistent amounts of product following growth under all the conditions used in these studies (see Figs 3b and 4c) and preliminary microarray data have also shown that the transcript levels of this gene are not changed following SOS induction (data not shown). Target gene bands of increased intensity in the post-exposure lanes relative to the unexposed lanes indicate an increase in transcription or stability of transcripts of the target gene in response to DNA damage, as is evident in Fig. 3(b) for lexA, recN, impA, recA and recX of strain RM118.



View larger version (38K):
[in this window]
[in a new window]
 
Fig. 4. Analysis of recAX readthrough transcript. (a) Boxes represent coding regions and the thin single line shows genomic DNA strand. Thin arrows show positions of primers used to amplify the 3' end of each gene or readthrough product from recA to recX (320 bp recA, 324 bp recX, 898 bp recAX readthrough). The frdB gene product is 578 bp. (b) Representative gels showing RT-PCR products obtained from RNA isolated from strains RM118 and the G91D-1 lexANI mutant either before (RM118, lanes 1; lexANI, lanes 3) or after (RM118, lanes 2; lexANI, lanes 4) treatment with mitC. (c) Mean band intensities obtained in RT-PCRs from three independent experiments with RM118 and the lexANI mutant either with or without mitC treatment. Error bars show standard deviations obtained for each gene for these three experiments except for frdB from RM118 unexposed, where only two readings were obtained.

 
RT-PCR analysis was performed from a total of six independent UV-induction experiments for RM118 and two independent experiments for each of the isogenic lexANI mutants G91D-1 and G91D-2 and also two independent experiments for the recA– strain. Changes in gene transcript levels following mitC induction in culture were monitored from three independent experiments for strains RM118 and the lexANI mutant G91D-1. Additional data for the effect of mitC exposure on transcript levels were obtained from cell cultures used in phase variation frequency analysis experiments.

Fig. 3(c) shows the results of the semi-quantitative analysis of the RT-PCRs. The results obtained from the two isogenic lexANI mutants pre- and post-UV exposure have been combined, as have the results of the effect of mitC treatment on RM118 and the lexANI mutant G91D-1 following exposure in culture or on plates. The efficiency of the RT-PCRs showed some variation both within and between experiments and thus incomplete datasets were obtained for some genes. The values plotted are the means of all successful analyses for each gene in each strain and the number of successful analyses is indicated below the axis in Fig. 3(c).

An induction ratio greater than 1·0 indicates that transcript levels are increased following exposure to DNA-damaging treatments (induction ratios were calculated from gel-band intensity measurements as described in Methods). The induction ratios of lexA, recN, recX and recA are all greater than 1·0 following either UV exposure or mitC treatment of strain RM118. The induction ratios of these genes were not increased in the same way in the recA– or lexANI mutants following UV exposure or mitC treatment, indicating that induction of these genes is LexA dependent.

Gel profiles for impA (Fig. 3b) showed it to be very tightly regulated by LexA. No message was detected in RNA samples from unexposed strain RM118 but a strong product band was obtained from RT-PCR following exposure. recA– and lexANI strains showed no detectable message in either pre- or post-exposure samples. The lack of detectable transcript in the mutant and unexposed WT samples meant that an induction ratio could not be determined.

Analysis of the recJ transcript showed an increase in transcription of this gene only in strain G91D-1 following mitC induction. The reasons for this result are not clear, but we note that RecJ is an ssDNA-specific exonuclease and it may be that recJ is upregulated in the G91D-1 mutant because it permits survival of mitC but not UV damage through a process that is achieved in WT cells by another SOS-regulated gene product. The remaining genes, himD, recF and HI1222, did not show an increase in transcription after induction using this analysis method.

Transcription of recX was further investigated to determine whether it was co-transcribed with recA, as it is in E. coli (Pagès et al., 2003). Fig. 4(b) shows representative results of RT-PCR amplifications performed on RNA extracts isolated from strain RM118 and the lexANI mutant G91D-1 before and after treatment with mitC in culture. A strong product was produced by amplification with primers spanning recA and recX, indicating the presence of a recAX readthrough transcript. The mean band intensities obtained following mitC treatment (Fig. 4c) show that the recA, recX and recAX transcripts were all more abundant in strain RM118 but not in the lexANI strain. The frdB control gene was amplified separately in these experiments and Fig. 4(c) demonstrates that the transcript level of this gene was constant between strains and induction conditions (see also Fig. 3b). It is not possible from these experiments to draw conclusions about the relative levels of recA or recX transcription relative to the co-transcript, as the lower intensities obtained for the recAX transcript may be due to a reduced amplification efficiency of the longer product.

Phase variation experiments
The lexANI strains G91D-1 and G91D-2 were transformed with plasmids containing either 38AGTC or 20AT repeats in a mod–repeat–lacZ reporter construct, as described in Methods, to generate strains G91D-1AGTC38, G91D-1AT20, G91D-2AGTC38 and G91D-2AT20, in which the mod–repeat–lacZ construct replaces the WT mod allele. Phase variation rates were first measured for the reporter constructs in the lexANI background under normal culture conditions, i.e. overnight growth on BHI agar at 37 °C. The results, given in Table 3, show that there was no significant difference in the phase variation frequencies measured for the reporter construct in the lexANI background compared with those for the reporter construct in the RM118 strain, indicating that the lexANI mutation has no detectable effect on the repeat-mediated phase variation frequency of the reporter construct under conditions where the SOS response is not induced.


View this table:
[in this window]
[in a new window]
 
Table 3. Effect of DNA-damaging agents on the phase variation frequencies of reporter constructs in RM118 and lexANI mutants

Phase variation frequencies (f) were obtained from measurements on the numbers of colonies given in brackets. Mutation rates were calculated according to Drake (1991) and 95 % confidence intervals (CI) were derived according to Kokoska et al. (1998).

 
Phase variation frequencies for the reporter construct in the lexANI and RM118 strains were also measured following induction of the SOS response by exposure to UV light or mitC treatment. Cells were exposed to UV light either immediately after plating or at the stage of small colonies, whilst mitC was added to the plates (see Methods). UV exposure was, therefore, a transient event from which the cells and colonies would be expected to recover, whilst mitC exposure was more prolonged. No significant differences in phase variation frequencies between the RM118 and lexANI strains were observed for either the dinucleotide repeat- or tetranucleotide repeat-containing constructs, following either UV exposure or mitC treatment (Table 3). The types of slippage events (i.e. insertions or deletions of one or more repeat units) that gave rise to the revertant phenotype were analysed for all experiments using the method previously described by Bayliss et al. (2002), in which the repeat region is amplified using fluorescently labelled primers and the products are sized using the ABI Prism 377 automatic sequencer and the ABI GeneScan 3.1 program. The types of slippage events were found not to differ between WT and lexANI strains under any of the culture or induction conditions, suggesting that no change in the mechanism by which slippage occurs is caused by the lexANI mutation or induction of the SOS response (data not shown).

However, although there was no significant difference in phase variation rates of the reporter constructs observed between RM118 and lexANI strains following either UV or mitC induction of the SOS response, a significant (approx. 2-fold) increase in phase variation rates was observed between mitC-treated and untreated cells for the 20AT reporter construct in RM118 (P=0·0298), the 38AGTC reporter construct in RM118 (P=0·0156) and the 38AGTC reporter construct in the lexANI strains (P=0·0002). The difference in the phase variation rate of the 20AT reporter construct in the lexANI strains was just outside the level of significance (P=0·0501) (P values obtained from Mann–Whitney non-parametric rank sum tests). The mechanism by which mitC treatment could give rise to an increase in phase variation frequency is not clear, but may be due to the constant DNA damage resulting from the prolonged exposure leading to repeated interruptions of DNA replication and hence more opportunities for the occurrence of slippage events.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The main aim of this study was to investigate whether the SOS response of H. influenzae could affect phase variation rates of contingency loci in this bacterial pathogen. In order to achieve this aim, we characterized the H. influenzae SOS response and constructed mutants in which the SOS response was not induced following DNA damage. The first step was to identify genes of the H. influenzae genome with the potential to be LexA regulated and so possibly form part of the SOS response, by means of a genomic sequence analysis of the Rd KW20 genome for the E. coli consensus LexA-binding sequence CTG(N)10CAG. This analysis revealed that 25 genes had the consensus LexA-binding sequence within 200 bp of the transcription start site. Seven of these 25 genes (lexA, recA, recN, uvrA, ssb, ruvA and uvrD) form part of the LexA regulon in E. coli and a further four have recognized repair and DNA-processing functions (himD, recF, recJ and mfd). Eight of the remaining SOS-box-associated genes in H. influenzae are of unknown function, whilst a further five have functions that would not intuitively place them in a LexA regulon (fdhD, aroK, sprT, tgt and HI1079). The remaining gene is annotated as impA. impA forms part of the impCAB operon in plasmid TP110 which belongs to the UmuDC-like family of plasmid-borne polymerases that confer UV resistance on their hosts (Strike & Lodwick, 1987). In this polymerase, impA encodes a umuD homologue. The H. influenzae impA gene sequence, however, lacks several of the sequence motifs shown to be important for UmuD to form a functional component of Pol V in E. coli, e.g. the conserved serine and lysine residues needed for the RecA-dependent self-cleavage reaction of UmuD to UmuD', which is analogous to the self-cleavage reaction of LexA, and sequences shown to be involved in dimerization and ClpXP tethering and processing of UmuD/D' complexes (Sutton et al., 2002; Neher et al., 2003). The function of this gene, which is very tightly regulated by LexA in H. influenzae, is therefore somewhat of a puzzle. However, the observation that H. influenzae is non-mutable by UV (Notani & Setlow, 1980) suggests that, whatever its role, impA does not form part of a functional polymerase in H. influenzae.

Only 12 of the 31 E. coli SOS genes reported by Fernández de Henestrosa et al. (2000) have homologues in the H. influenzae genome. Those that are absent include the TLS polymerases (Pols II, IV and V) that are responsible for increased mutation rates associated with induction of the SOS response in E. coli (Napolitano et al., 2000; Wagner et al., 2002).

We next created a mutation in the lexA gene of H. influenzae that gave a non-SOS-inducible phenotype (lexANI) and showed that strains with the mutation were more sensitive than WT cells to DNA-damaging agents but not as sensitive as recA– mutants. These results demonstrate that the SOS response is compromised in the lexANI and recA– mutants and suggest that the H. influenzae lexA gene mediates survival following DNA damage in a process that is mechanistically similar to that mediated by E. coli lexA. Transcriptional analysis of nine of the candidate H. influenzae SOS genes identified from the genome analysis showed increased message levels for lexA, recA, recN, ruvA, recX and impA following induction of the SOS response. These induced genes include the core response genes (lexA, recA and recN) identified in the work of Erill et al. (2003). However, although the sensitivity of the semi-quantitative RT-PCR analysis method was sufficient to determine that these six genes are upregulated as part of the SOS response, it did not permit accurate comparisons of the relative induction levels of these genes or a definitive demonstration that other genes analysed (himD, recF, recJ) are not induced. A comprehensive investigation using full genome microarrays will clarify whether these three genes are upregulated by H. influenzae in response to DNA damage and whether this response includes some of the additional members of the potential H. influenzae LexA regulon identified herein but not analysed (Table 2) or indeed other genes regulated by non-consensus SOS boxes or by alternative induction mechanisms.

We have shown evidence that the H. influenzae recX is co-transcribed with recA. In E. coli, only 5–10 % of the recA transcripts are recArecX co-transcripts, because the majority of transcripts terminate at a palindromic repeat sequence that is located between these two genes (Pagès et al., 2003). Whilst we have not accurately determined the relative levels of the recA and recX transcripts, it is interesting to note that there are two inverted copies of the H. influenzae uptake sequence 19 bp 3' of the H. influenzae recA stop codon, which could form a stem–loop and thereby terminate transcription between recA and recX (Zulty & Barcak, 1993). Additionally, it was recently shown that the E. coli RecX competes with DinI (an E. coli SOS-regulated protein that is absent from the H. influenzae genome) to modulate RecA activity; RecX being a negative regulator and DinI positive (Lusetti et al., 2004). It will be of interest to investigate the role of recX further in H. influenzae, given the extreme sensitivity of the H. influenzae recA– mutant (Fig. 2) and the absence of the SOS-regulated dinI gene from the H. influenzae genome (Table 2).

The results of our phase variation studies showed that, in contrast to the findings of Morel et al. (1998), induction of the SOS response in H. influenzae did not affect phase variation frequencies mediated by either an AT dinucleotide or AGTC tetranucleotide repeat tract in the mod locus. Phase variation frequencies at a second locus, lic2A, which contains CAAT tetranucleotide repeats fused to the lacZ reporter gene, were also measured for strain RM118 with and without mitC treatment (data not shown). A similar result to that obtained for the AGTC repeats following mitC treatment was observed, confirming that the tetranucleotide repeat tracts in H. influenzae are not destabilized by the SOS response. One interpretation of our findings is that the destabilization of dinucleotide repeats observed by Morel et al. (1998) was due to expression of one or more of the E. coli SOS genes that are absent in H. influenzae; particularly strong candidates are the TLS polymerases. Each of the three TLS polymerases in E. coli can cause –1 or –2 bp frameshifts, dependent upon the nature of the lesion and the combination of polymerases present (reviewed by Tippin et al., 2004).

In our studies, treatment with mitC caused an increase in phase variation rates at both dinucleotide and tetranucleotide repeats, independently of the SOS response. The reasons for the small increase in phase variation rates observed with mitC are unclear, but we have suggested that they may be due to repeated interruptions of DNA replication resulting from prolonged exposure to mitC allowing more opportunities for the occurrence of slippage events. Alternatively, DNA-repair pathways that are not LexA controlled may be responsible for this observed increase.

In summary, we have shown that H. influenzae has a LexA-dependent SOS response to DNA damage that aids survival following DNA assault and have demonstrated that at least five genes involved in DNA repair and recombination (lexA, recA, recX, recN and ruvA) are expressed as part of this response. Induction of the H. influenzae SOS response did not affect slippage rates at an AT dinucleotide or tetranucleotide repeat tract. The observed stability of H. influenzae repeats during an SOS response may be due to the absence of TLS polymerases in the H. influenzae genome.


   ACKNOWLEDGEMENTS
 
The authors wish to acknowledge and thank Dr Derek Hood in particular, for his invaluable advice whilst this work was ongoing and during preparation of this manuscript, as well as the rest of our colleagues for their discussion and support. This work was funded by a Wellcome Trust program grant.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bayliss, C. D., van de Ven, T. & Moxon, E. R. (2002). Mutations in polI but not mutSLH destabilize Haemophilus influenzae tetranucleotide repeats. EMBO J 21, 1465–1476.[Abstract/Free Full Text]

Bayliss, C. D., Sweetman, W. A. & Moxon, E. R. (2004). Mutations in Haemophilus influenzae mismatch repair genes increase mutation rates of dinucleotide repeat tracts but not dinucleotide repeat-driven pilin phase variation rates. J Bacteriol 186, 2928–2935.[Abstract/Free Full Text]

Bayliss, C. D., Sweetman, W. A. & Moxon, E. R. (2005). Destabilization of tetranucleotide repeats in Haemophilus influenzae mutants lacking RnaseHI or the Klenow domain of PolI. Nucleic Acids Res 33, 400–408.[Abstract/Free Full Text]

Courcelle, J., Khodursky, A., Peter, B., Brown, P. O. & Hanawalt, P. C. (2001). Comparative gene expression profiles following UV exposure in wild-type and SOS-deficient Escherichia coli. Genetics 158, 41–64.[Abstract/Free Full Text]

Danner, D. B. & Pifer, M. L. (1982). Plasmid cloning vectors resistant to ampicillin and tetracycline which can replicate in both E. coli and Haemophilus cells. Gene 18, 101–105.[CrossRef][Medline]

Dawid, S., Barenkamp, S. J. & St Geme, J. W., III (1999). Variation in expression of the Haemophilus influenzae HMW adhesions: a prokaryotic system reminiscent of eukaryotes. Proc Natl Acad Sci U S A 96, 1077–1082.[Abstract/Free Full Text]

De Bolle, X., Bayliss, C. D., Field, D., van de Ven, T., Saunders, N. J., Hood, D. W. & Moxon, E. R. (2000). The length of a tetranucleotide repeat tract in Haemophilus influenzae determines the phase variation rate of a gene with homology to type III DNA methyltransferases. Mol Microbiol 35, 211–222.[CrossRef][Medline]

Drake, J. W. (1991). A constant rate of spontaneous mutation in DNA-based microbes. Proc Natl Acad Sci U S A 88, 7160–7164.[Abstract/Free Full Text]

Erill, I., Escribano, M., Campoy, S. & Barbé, J. (2003). In silico analysis reveals substantial variability in the gene contents of the gamma proteobacteria LexA-regulon. Bioinformatics 19, 2225–2236.[Abstract/Free Full Text]

Fernández de Henestrosa, A. R., Ogi, T., Aoyagi, S., Chafin, D., Hayes, J. J., Ohmori, H. & Woodgate, R. (2000). Identification of additional genes belonging to the LexA regulon in Escherichia coli. Mol Microbiol 35, 1560–1572.[CrossRef][Medline]

Fleischmann, R. D., Adams, M. D., White, O. & 37 other authors (1995). Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512.[Medline]

Glazebrook, J. A., Grewal, K. K. & Strike, P. (1986). Molecular analysis of the UV protection and mutation genes carried by the I incompatibility group plasmid TP110. J Bacteriol 168, 251–256.[Medline]

Herriott, R. M., Meyer, E. M. & Vogt, M. (1970). Defined nongrowth media for stage II development of competence in Haemophilus influenzae. J Bacteriol 101, 517–524.[Medline]

Hood, D. W., Deadman, M. E., Jennings, M. P., Bisercic, M., Fleischmann, R. D., Venter, J. C. & Moxon, E. R. (1996). DNA repeats identify novel virulence genes in Haemophilus influenzae. Proc Natl Acad Sci U S A 93, 11121–11125.[Abstract/Free Full Text]

Khil, P. P. & Camerini-Otero, R. D. (2002). Over 1000 genes are involved in the DNA damage response of Escherichia coli. Mol Microbiol 44, 89–105.[CrossRef][Medline]

Kokoska, R. J., Stefanovic, L., Tran, H. T., Resnick, M. A., Gordenin, D. A. & Petes, T. D. (1998). Destabilization of yeast micro- and minisatellite DNA sequences by mutations affecting a nuclease involved in Okazaki fragment processing (rad27) and DNA polymerase {delta} (pol3-t). Mol Cell Biol 18, 2779–2788.[Abstract/Free Full Text]

Lin, L. L. & Little, J. W. (1988). Isolation and characterisation of noncleavable (Ind) mutants of the LexA repressor of Escherichia coli K-12. J Bacteriol 170, 2163–2173.[Medline]

Little, J. W. (1993). LexA cleavage and other self-processing reactions. J Bacteriol 175, 4943–4950.[Medline]

Little, J. W. & Mount, D. W. (1982). The SOS regulatory system of Escherichia coli. Cell 29, 11–22.[CrossRef][Medline]

Lusetti, S. L., Drees, J. C., Stohl, E. A., Seifert, H. S. & Cox, M. M. (2004). The DinI and RecX proteins are competing modulators of RecA function. J Biol Chem 279, 55073–55079.[Abstract/Free Full Text]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Morel, P., Reverdy, C., Michel, B., Ehrlich, S. D. & Cassuto, E. (1998). The role of SOS and flap processing in microsatellite instability in Escherichia coli. Proc Natl Acad Sci U S A 95, 10003–10008.[Abstract/Free Full Text]

Moxon, E. R., Rainey, P. B., Nowak, M. A. & Lenski, R. E. (1994). Adaptive evolution of highly mutable loci in pathogenic bacteria. Curr Biol 4, 24–33.[CrossRef][Medline]

Napolitano, R., Janel-Bintz, R., Wagner, J. & Fuchs, R. P. P. (2000). All three SOS-inducible DNA polymerases (Pol II, Pol IV and Pol V) are involved in induced mutagenesis. EMBO J 19, 6259–6265.[Abstract/Free Full Text]

Neher, S. B., Sauer, R. T. & Baker, T. A. (2003). Distinct peptide signals in the UmuD and UmuD' subunits of UmuD/D' mediate tethering and substrate processing by the ClpXP protease. Proc Natl Acad Sci U S A 100, 13219–13224.[Abstract/Free Full Text]

Notani, N. K. & Setlow, J. K. (1980). Inducible repair system in Haemophilus influenzae unaccompanied by mutation. J Bacteriol 143, 516–519.[Medline]

Pagès, V., Koffel-Schwartz, N. & Fuchs, R. P. P. (2003). recX, a new SOS gene that is co-transcribed with the recA gene in Escherichia coli. DNA Repair 2, 273–284.[CrossRef][Medline]

Sassanfar, J. & Roberts, J. W. (1990). Nature of the SOS-inducing signal in Escherichia coli. The involvement of DNA replication. J Mol Biol 212, 79–96.[CrossRef][Medline]

Stohl, E. A., Brockman, J. P., Burkle, K. L., Morimatsu, K., Kowalczykowski, S. C. & Seifert, H. S. (2003). Escherichia coli RecX inhibits RecA recombinase and coprotease activities in vitro and in vivo. J Biol Chem 278, 2278–2285.[Abstract/Free Full Text]

Strike, P. & Lodwick, D. (1987). Plasmid genes affecting DNA repair and mutation. J Cell Sci 6, 303–321.

Sutton, M. D., Guzzo, A., Narumi, I., Costanzo, M., Altenbach, C., Ferentz, A. E., Hubbell, W. L. & Walker, G. C. (2002). A model for the structure of the Escherichia coli SOS-regulated UmuD2 protein. DNA Repair 1, 77–93.[CrossRef][Medline]

Tippin, B., Pham, P. & Goodman, M. F. (2004). Error-prone replication for better or worse. Trends Microbiol 12, 288–295.[CrossRef][Medline]

van Ham, S. M., van Alphen, L., Mooi, F. R. & van Putten, J. P. M. (1993). Phase variation of H. influenzae fimbriae: transcriptional control of two divergent genes through a variable combined promoter region. Cell 73, 1187–1196.[CrossRef][Medline]

Venkatesh, R., Ganesh, N., Guhan, N., Sreedhar Reddy, M., Chandrasekhar, T. & Muniyappa, K. (2002). RecX protein abrogates ATP hydrolysis and strand exchange promoted by RecA: insights into negative regulation of homologous recombination. Proc Natl Acad Sci U S A 99, 12091–12096.[Abstract/Free Full Text]

Wagner, J., Etienne, H., Janel-Bintz, R. & Fuchs, R. P. P. (2002). Genetics of mutagenesis in E. coli: various combinations of translesion polymerases (Pol II, IV and V) deal with lesion/sequence context diversity. DNA Repair 1, 159–167.[CrossRef][Medline]

Walker, G. C. (1984). Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coli. Microbiol Rev 48, 60–93.[Medline]

Zulty, J. J. & Barcak, G. J. (1993). Structural organization, nucleotide sequence, and regulation of the Haemophilus influenzae rec-1+ gene. J Bacteriol 175, 7269–7281.[Abstract]

Received 28 February 2005; revised 19 May 2005; accepted 25 May 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Sweetman, W. A.
Articles by Bayliss, C. D.
Articles citing this Article
PubMed
PubMed Citation
Articles by Sweetman, W. A.
Articles by Bayliss, C. D.
Agricola
Articles by Sweetman, W. A.
Articles by Bayliss, C. D.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.