Molecular analysis of mutS expression and mutation in natural isolates of pathogenic Escherichia coli

Baoguang Li1, Ho-Ching T. Tsui2,{dagger}, J. Eugene LeClerc1, Manashi Dey1, Malcolm E. Winkler2,{dagger} and Thomas A. Cebula1

1 Center for Food Safety and Applied Nutrition, Food and Drug Administration, Laurel, MD 20708, USA
2 Department of Microbiology and Molecular Genetics, University of Texas Houston Medical School, Houston, TX 77030, USA

Correspondence
Thomas A. Cebula
tac{at}cfsan.fda.gov


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Deficiencies in the MutS protein disrupt methyl-directed mismatch repair (MMR), generating a mutator phenotype typified by high mutation rates and promiscuous recombination. How such deficiencies might arise in the natural environment was determined by analysing pathogenic strains of Escherichia coli. Quantitative Western immunoblotting showed that the amount of MutS in a wild-type strain of the enterohaemorrhagic pathogen E. coli O157 : H7 decreased about 26-fold in stationary-phase cells as compared with the amount present during exponential-phase growth. The depletion of MutS in O157 : H7 is significantly greater than that observed for a laboratory-attenuated E. coli K-12 strain. In the case of stable mutators, mutS defects in strains identified among natural isolates were analysed, including two E. coli O157 : H7 strains, a diarrhoeagenic E. coli O55 : H7 strain, and a uropathogenic strain from the E. coli reference (ECOR) collection. No MutS could be detected in the four strains by Western immunoblot analyses. RNase T2 protection assays showed that the strains were either deficient in mutS transcripts or produced transcripts truncated at the 3' end. Nucleotide sequence analysis revealed extensive deletions in the mutS region of three strains, ranging from 7·5 to 17·3 kb relative to E. coli K-12 sequence, while the ECOR mutator contained a premature stop codon in addition to other nucleotide changes in the mutS coding sequence. These results provide insights into the status of the mutS gene and its product in pathogenic strains of E. coli.


Abbreviations: DEC, diarrhoeagenic Escherichia coli; ECOR, Escherichia coli reference; MMR, methyl-directed mismatch repair

{dagger}Present address: Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, IN 46285, USA.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Mutators arise in bacterial populations at frequencies of 10-5 to 10-6, as shown in studies using laboratory strains of Escherichia coli (Mao et al., 1997; Boe et al., 2000) and Salmonella typhimurium (LeClerc et al., 1998). Surveys of natural isolates of E. coli and Salmonella showed that mutators occur in the environment at much higher frequencies of around 1–5 % (LeClerc et al., 1996). This rise of mutator alleles in natural populations implies an important role for mutators in the evolution of microbes, such as a source of genetic variants that adapt to unstable environments (Cox, 1976). While the hypermutable phenotype may produce the beneficial mutations that increase the frequency of linked mutator alleles in populations, we called attention to the role that promiscuous recombination could play in evolution (LeClerc et al., 1996; LeClerc & Cebula, 1997).

Notably, although spontaneous mutation rates are enhanced by mutation in at least 25 separate genetic loci (Miller, 1998; Horst et al., 1999), the overwhelming majority of mutators found thus far among natural strains are due to defects in the mutS, mutH, mutL or uvrD gene (LeClerc et al., 1996; Matic et al., 1997). These genes define the methyl-directed mismatch repair (MMR) pathway, a DNA repair system that not only corrects base mismatches in newly replicated DNA but is also the main barrier for recombination of mismatched heteroduplexes in DNA (i.e. homeologous recombination) (Modrich & Lahue, 1996). That is, MMR defects relax the recombination barrier, allowing DNA exchange between species that normally do not mate. This promiscuous property, unique to MMR mutants, may explain the emergence and persistence of MMR mutants in natural populations. Most importantly, it implicates a role for horizontal gene transfer in gaining the benefits for survival and growth. Since we reported the high incidence of MMR-defective mutators among E. coli and Salmonella pathogens (LeClerc et al., 1996), the accounts of mutator subpopulations among pathogenic isolates of E. coli (Matic et al., 1997; Denamur et al., 2002), Pseudomonas aeruginosa (Oliver et al., 2000), Neisseria meningitidis (Richardson et al., 2002) and Streptococcus pneumoniae (Negri et al., 2002) bolster the view that these mutators are positively selected in nature, most likely because of the beneficial mutations they carry.

Since the earliest observations of mutators in natural populations (Jyssum, 1960; Gross & Siegal, 1981), several features of their incidence in the feral setting have been puzzling. That is, although laboratory experiments have shown that a high mutation rate can be advantageous, as evidenced by the successful competition of mutators against their non-mutator counterparts in chemostat cultures (Gibson et al., 1970; Nestman & Hill, 1973; Cox & Gibson, 1974; Tröbner & Piechocki, 1981; Chao & Cox, 1983) and the invasion of non-mutator clones by mutators in continuous growth experiments (Sniegowski et al., 1997), mutators have not overtaken natural populations of bacteria. Moreover, most organisms appear to have evolved a seemingly constant mutation rate (Drake, 1991; Ochman et al., 1999). Finally, since deleterious mutations arise at rates roughly 104 times those of beneficial mutations (Cebula & LeClerc, 2000), a heightened mutation rate should merely hasten the mutator to extinction in a stable environment. Yet, our studies demonstrated that MMR mutator alleles persist among natural clones at frequencies 10- to 1000-fold greater than expected for deleterious mutations (LeClerc et al., 1996). Our results suggested, therefore, that the mutator state must be a temporary situation; i.e. a mutator allele rises to prominence with a beneficial trait that it spawned (i.e. hitch-hikes) and then falls when selection again sweeps the population (see Shaver et al., 2002). Haploidy dictates, after all, that if the trait is to be maintained in perpetuity, the bacterium must ‘lose’ (by reversion or recombination) or otherwise quiet (by suppression) the mutator allele.

A transient mutator state might also result, however, because of an altered expression of MMR proteins. As downregulation of MutS, and to a lesser extent MutH, is known to occur in stationary phase and in starved bacteria (Feng et al., 1996), and the amount of MutS in E. coli K-12 is already near limiting for MMR in exponentially growing cells, the decreases of MMR proteins observed in stationary-phase cells and in nutrient-deprived cells are of especial interest. If similar downregulation occurs in natural isolates, as observed in laboratory cultures of E. coli K-12, the decreased amounts of MutS and MutH could contribute significantly to increased mutagenesis and homeologous recombination as cells enter and leave the stationary phase or otherwise stressed environments.

To evaluate mechanisms that may be involved in the emergence of a mutator phenotype, we determined the molecular basis for MMR deficiencies in natural isolates. To do this, we analysed a wild-type isolate of enterohaemorrhagic E. coli O157 : H7 for levels of MutS, MutH and MutL proteins in growing and stationary-phase cultures. We also determined the molecular defects in stable mutators identified among pathogenic isolates of E. coli: two E. coli O157 : H7 isolates, EC503 and EC535; a diarrhoeagenic E. coli of the O55 : H7 serotype, DEC5A; and a uropathogenic E. coli from the E. coli reference (ECOR) strain collection, ECOR48. Genetic complementation studies showed that normal mutability could be restored in each of these mutator strains by plasmid copies of a wild-type mutS gene (LeClerc et al., 1996), hence molecular analyses were carried out to define the nature of their MutS defects.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains.
E. coli MG1655, used for E. coli genome sequence analysis (Blattner et al., 1997), and E. coli W3110 are K-12 strains described in Bachmann (1987). E. coli EC536, EC535 and EC503 are O157 : H7 strains from the Food and Drug Administration (FDA) bacterial pathogen collection and have been described previously (LeClerc et al., 1996). E. coli DEC5A is an O55 : H7 strain from the reference collection of diarrhoeagenic E. coli (DEC) described by Whittam et al. (1993). ECOR48 is a uropathogenic E. coli strain from the ECOR collection described by Ochman & Selander (1984).

Growth conditions.
For routine growth, cells from frozen stocks of each bacterial strain were grown in LB medium (Miller, 1972) with shaking at 37 °C. For RNase protection assays and Western blotting experiments, bacterial stocks were added to 5 ml enriched minimal salts/glucose [EMMG; Vogel-Bonner (1xE) minimal salts, 0·01 mM FeSO4, 0·4 % (w/v) glucose and 0·5 % (w/v) vitamin-assay Casamino acids (Difco Laboratories)]. Cultures were grown overnight with shaking (300 r.p.m.) at 37 °C, and the overnight cultures were diluted 200-fold with fresh EMMG and grown with shaking. Samples for RNase T2 protection assays were collected during exponential growth at a turbidity of 50 Klett (660 nm) units ({approx}5x108 cells per ml determined for MG1655). Samples for Western analysis were collected from cultures grown to mid-exponential phase (a turbidity of 50–65 Klett units), transitional phase (200 Klett units) and stationary phase (24 and 48 h after inoculation).

Western blotting.
Samples for Western blotting were prepared as described previously (Feng et al., 1996) or with the following modifications. Portions of cultures (20, 7 and 3 ml) of strains MG1655 and EC536 were removed at culture turbidity values of 50–65 Klett units and 200 Klett units, or at 24 and 48 h after inoculation. After centrifugation at 5000 g for 7 min at room temperature, pellets were washed once with 20 ml of 1xE salts and washed a second time in 1 ml of 1xE salts in microfuge tubes. The final pellets were resuspended in 300 µl buffer containing 0·125 M Tris/HCl, pH 6·8, and 4 % SDS and boiled for 10 min. Total protein concentrations were determined by using the Bio-Rad Dc protein assay kit on samples diluted 20-fold in water and on BSA standards in the same diluted buffer. For analysis on SDS-PAGE gels, samples were diluted in a solution containing 2 M DTT, 40 % (v/v) glycerol and 0·1 % bromophenol blue, so that final protein samples were resuspended in Laemmli buffer (Laemmli, 1970; 2 % SDS, 100 mM DTT, 10 % glycerol, 62·5 mM Tris/HCl, pH 6·8).

Quantitative Western blotting assays to detect MutS, MutL and MutH were carried out as described previously (Tsui et al., 1997) or with the changes described below. Antisera against hexahistidine (His-6)-tagged E. coli MutS, MutL and MutH proteins were produced as described previously (Feng & Winkler, 1995) and affinity-purified with CNBr-activated Sepharose 4B (Pharmacia Biotech) coupled to His-6-tagged E. coli MutS, His-6-tagged E. coli MutL or His-6-tagged E. coli MutH as described previously (Harris et al., 1997). In gel analysis, the His-6 tag (about 2 kDa) did not appreciably change the migration of the tagged proteins. To further reduce background, the antibodies were pre-adsorbed to lysates prepared from mutS, mutL or mutH insertion mutants (Feng et al., 1996). Alkaline-phosphatase-conjugated secondary antibodies or ECL Western blotting detection reagents (Amersham) were used for detection.

RNase T2 protection assays of chromosomal transcripts.
RNA was prepared by adding portions of bacterial cultures directly to lysis solutions without intervening steps as described previously (Feng et al., 1996). RNase T2 protection assays of transcripts from the bacterial chromosome were conducted as described by Tsui et al. (1994). RNA probes for detecting mutS transcripts (see Fig. 2) were synthesized using the following phage RNA polymerases and linearized plasmid templates: SP6 and EcoRI-treated pTX414 (5' end of mutS transcripts); T7 and HincII-treated pTX541 (3' end of mutS transcripts); and SP6 and EcoRI-treated pTX541 (antisense transcripts at the 3' end of mutS) (Tsui et al., 1997). A series of labelled, undigested probes of known length were used as size standards to determine the lengths of the protected fragments (standard errors of 5–10 %) (Tsui et al., 1994). Each hybridization reaction contained 25 µg total RNA from cells grown in EMMG. Radioactivity in bands was quantified by using an InstantImager (Packard).



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 2. Western immunoblots to determine the cellular amounts of MutS, MutL and MutH in EC536 (E. coli O157 : H7) or MG1655 (E. coli K-12) grown to mid-exponential phase (lane 1), transitional phase (lane 2) or stationary phase (24 and 48 h after inoculation, lanes 3 and 4, respectively). Values under each lane are amounts (ng) of MutS, MutL or MutH per 50 µg total protein extract loaded per lane. Quantification standards were obtained by adding known amounts (ng indicated under each lane) of MutS, His-6-MutL or MutH to 50 µg total protein from the respective null mutants [MG1655 mutS : : {Omega}, MG1655 mutL : : {Omega} or MG1655 mutH : : Tn5 (Tsui et al., 1997)]. MW, molecular mass standards.

 
PCR.
Standard PCRs and long PCRs were performed according to the manufacturer's directions using the AmpliTaq DNA Polymerase Kit and the GeneAmp XL PCR Kit (Perkin Elmer), respectively. Primers used to produce amplification products for sequencing the mutS region of E. coli strains are listed in Table 1. In the experiment shown in Fig. 4, amplification products of 17·8 and 14·1 kb are expected to be produced on templates prepared from wild-type strains of E. coli K-12 (W3110) and E. coli O157 : H7 (EC536), respectively, using primers F21 and BL260 (Table 1).


View this table:
[in this window]
[in a new window]
 
Table 1. Primers for DNA amplification by PCR

 


View larger version (27K):
[in this window]
[in a new window]
 
Fig. 4. Long PCR analysis of deletions in the E. coli mutS mutants. The PCR products shown are from the wild-type O157 : H7 strain EC536 (lane 1) and from mutants EC503 (lane 2), DEC5A (lane 3) and EC535 (lane 4). PCR was carried out using the primer pair F21/BL260, producing a 14·1 kb amplicon on O157 : H7 wild-type DNA and hence covering the largest deletion. The BRL molecular size markers are the HMW Marker (H) and 1 kb DNA ladder (M).

 
Cloning the mutS gene of ECOR48.
Chromosomal DNA was prepared from an overnight culture of ECOR48 using the Purigene DNA Isolation Kit (Gentra) according to the manufacturer's directions. DNA, digested with SalI and BglII, produced a 3·9 kb fragment containing an intact mutS gene and flanking regions as determined by restriction digestion of PCR product from the mutS region of ECOR48. DNA, purified from the 3·9 kb region of an agarose gel, was ligated with SalI- and BglII-digested vector pTZ72 (Promega) and electroporated into ElectroMax DH10B cells (Gibco-BRL). A 32P-end-labelled oligonucleotide, specific for the mutS region of ECOR48 (5'-GCCGGAGAGCCATACTCTCC), was used to probe transformed cells by colony filter hybridization as described previously (Cebula & Koch, 1990). Plasmid DNA was prepared from probe-positive clones, and the presence of the mutS gene from ECOR48 was confirmed by PCR and restriction digestion analysis.

For sequencing the mutS gene of strain ECOR48, the ECOR48 mutS gene clone was grown in LB medium at 37 °C overnight and subjected to a Wizard Plus Midiprep DNA Purification System (Promega), according to the manufacturer's instructions. Plasmid DNA was further purified by phenol extraction and ethanol precipitation. The 3·9 kb DNA fragment, covering the mutS gene and its flanking sequences, was sequenced from plasmid DNA using a commercial service (Lark Technologies).

Nucleotide sequencing and analysis.
To determine the exact end points of mutS deletions in strains EC503, EC535 and DEC5A, PCR products containing the junction fragments were sequenced. DNA sequencing was performed using the T7 Sequenase Version 2.0 DNA Sequencing Kit (Amersham) according to the manufacturer's instructions. Programs from the Wisconsin Sequence Analysis Package (Genetics Computer Group) were used for sequence analyses.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Detection of MutS in wild-type and mutant strains
Western analysis of the MutS proteins from E. coli K-12 strains and pathogenic E. coli strains is shown in Fig. 1. MutS was detected in the wild-type E. coli O157 : H7 strain EC536; the intensity of the MutS-specific band in EC536 is similar to that of the K-12 prototrophic strains MG1655 and W3110. It should be noted that protein bands on gels often appeared blurred and/or shifted upward in the case of extracts from the natural pathogen EC536, possibly because these extracts contain considerably more outer-membrane components, such as lipopolysaccharides (data not shown). Consistent with previous genetic complementation results (LeClerc et al., 1996), MutS was undetectable in the mutS mutant strains ECOR48, EC503, EC535 and DEC5A.



View larger version (38K):
[in this window]
[in a new window]
 
Fig. 1. Western immunoblots to detect MutS proteins. One-hundred micrograms of total protein extract from the strains indicated were loaded in each lane and immunoreactive bands were detected using the alkaline phosphatase method. W3110 and MG1655, E. coli K-12 prototrophic strains; TX3496, the MutS null mutant MG1655 mutS : : {Omega} (Tsui et al., 1997). In the second lane from the top, 15 ng purified MutS were added to total protein from TX3496.

 
Quantification of the MutS, MutH and MutL proteins in wild-type strains
We have previously reported that the amounts of MutS and MutH are reduced in K-12 strain MG1655 as cells enter stationary phase (Feng et al., 1996). To investigate whether similar downregulation occurs in the pathogenic E. coli O157 : H7 strain, we performed Western analysis of the MutS, MutL and MutH proteins using protein samples obtained from EC536 and MG1655 strains that were grown to mid-exponential, transitional or stationary phase. Similar to results reported previously (Feng et al., 1996), MutS amounts in MG1655 decreased 4·5-fold as cells entered stationary phase (24 and 48 h after inoculation, lanes 3 and 4, Fig. 2) when compared to cells grown to mid-exponential phase (lane 1) or transitional phase (lane 2). As shown in Fig. 2, MutS amounts decreased more in EC536 than in MG1655 after cells entered stationary phase. While MutS amounts in the exponential phase and transitional phase were similar in the two strains, MutS amounts were almost non-detectable in the stationary-phase EC536 strain (lanes 3 and 4). From a curve generated from the quantification standards with different amounts of purified MutS spiked into extracts of null strains (Fig. 2), we estimated that the amount of MutS in EC536 was decreased about 26-fold in stationary-phase cells compared to that in exponential-phase cells. MutH amounts decreased about twofold in EC536 as cells entered stationary phase (Fig. 2), similar to that observed with MG1655 (Fig. 2, and Feng et al., 1996). MutL amounts remained relatively constant in the different growth phases in strains EC536 and MG1655 (Fig. 2, and Feng et al., 1996).

Mapping and measurement of mutS transcripts
To determine whether the transcriptional organization of the mutS gene in E. coli O157 : H7 is similar to that in MG1655, we mapped the transcripts by RNase T2 protection assays using RNA probes specific for the 5' or 3' end of the mutS gene (Fig. 3a). EC536, the wild-type O157 : H7 strain, showed similar patterns of 5' and 3' ends of mutS transcripts as in MG1655 (Fig. 3b). Three major protected transcripts of 520, 480 and 428 nt were seen using the 5'-end mutS probe on RNA from MG1655 and EC536. The 5' ends of the 520 and 480 nt protected species were mapped to positions approximately 74 and 34 nt, respectively, upstream from the mutS start codon of MG1655 (Tsui et al., 1997). The 428 nt protected species is a partial mutS transcript (Tsui et al., 1997). Two major protected transcripts of 990 and 920 nt were seen using the 3'-end mutS probe on RNA from MG1655 and EC536. These bands correspond to mutS transcripts with 3' ends at approximately 97 and 29 nt downstream from the mutS stop codon. Together, these results show that mutS in EC536 (O157 : H7) is in a single-gene operon, as was reported previously for MG1655 (K-12) (Tsui et al., 1997), and that the transcription starts, stops and possible processing sites appear similar for the enterohaemorrhagic and laboratory strains of E. coli.



View larger version (42K):
[in this window]
[in a new window]
 
Fig. 3. Mapping of mutS transcripts in E. coli strains. (a) Structure (drawn to scale) and transcription of the mutS genomic region in E. coli K-12 (Tsui et al., 1997). Putative promoters, PmutS and Po218b, and terminators, Term1 and Term2, were described by Tsui et al. (1997). {Delta}, start of deletion in the 3' end of mutS in strain EC503. Lines indicate RNA probes used in RNase T2 assays. (b) mutS transcripts in E. coli strains. RNase T2 protection assays (see Methods) were performed using RNA probes specific to the 5' or 3' ends of mutS, or antisense transcripts at the 3' end of mutS.

 
Mapping of mutS transcripts was carried out on the mutator strains EC503, EC535, DEC5A and ECOR48 to examine the nature of their mutS defects (Fig. 3b). No protected band, corresponding to the 5' end of mutS transcripts, was observed in strains EC535 and DEC5A, indicating an absence of mutS mRNA in these strains. In strain EC503, hybridization of the 5' probe showed patterns and amounts of 5' mutS transcript similar to RNA from the non-mutator EC536. However, the protected transcripts of EC503 at the 3' end were much less abundant and shorter than those of EC536. The 620 nt protected band from EC503 corresponds to a transcript with a 3' end about 270 nt upstream from the mutS stop codon. In strain ECOR48, mutS transcript was detected using the 5'-end probe, but not with the 3'-end probe. The results from analyses of EC503 and ECOR48 transcripts suggested that mutations in the mutS structural genes of these strains affected transcript length or stability. The presence of antisense transcripts was also investigated in order to test for DNA contamination in RNA preparations and to determine if antisense transcription was identical in wild-type and mutant strains. It is notable that protected antisense transcripts that correspond to the 3' end of the mutS gene were detected in EC503 (Fig. 3b), but not in EC536 (Fig. 3b) or MG1655 (data not shown).

PCR and sequence analysis of EC503, EC535 and DEC5A mutant DNAs
In EC536, mutS is adjacent on its 5' side to fhlA and on its 3' side to the o218–yclD–yclC–yclB–slyA–rpoS gene cluster (cf. Fig. 3). Attempts to PCR-amplify an intact mutS gene from the EC503, EC535 and DEC5A mutators failed, suggesting that primer-annealing sites were disrupted by deletions or rearrangements of the gene in these strains. To determine the extent of possible mutS deletions, the mutants were probed under low stringency conditions using a series of 32P-end-labelled oligonucleotides designed to anneal at 1·5 kb intervals throughout a 12 kb region surrounding mutS from either E. coli K-12 or O157 : H7. The results of colony hybridization indicated that EC503 contained a deletion of the 3' end of the mutS gene and open reading frames (ORFs) between the mutS and rpoS genes; EC535 and DEC5A carried extensive deletions encompassing mutS and neighbouring 5'-fhlA gene and extending at least 3 kb to the rpoS gene (data not shown).

Analysis of the EC503 mutant by PCR amplification of DNA from the 5' end of the mutS gene to the middle of the rpoS gene followed by sequencing of the PCR product confirmed that the 3' end of the mutS gene and ORFs between the mutS and rpoS genes were deleted in this mutant. Initial PCR analysis of the more extensive deletions in EC535 and DEC5A involved amplification of short products on each side of the deletions in order to determine their approximate extent. Primers, established for productive PCR, were then used in long PCRs for amplification of products containing the deletion junctions. Sequence analysis of these PCR products defined the deletions in mutant DNAs. Table 2 gives the sizes and positions of the deletions and shows wild-type sequence present at the deletion end points based on the K-12 genome. Sequences of 6–8 base pairs are repeated at the end points of the deletions, leaving one intact copy in the cases of EC503 and EC535 or deleting both copies in DEC5A. Fig. 4 shows the PCR products amplified from primers at fixed sites in the three mutant DNAs. In the experiment of Fig. 4, an amplification product of 14·1 kb is expected to be produced on template prepared from a wild-type strain of E. coli O157 : H7 (EC536), while products of 10·6, 4·2 and 0·8 kb are observed on EC503, DEC5A and EC535 DNAs, respectively.


View this table:
[in this window]
[in a new window]
 
Table 2. Deletion end points in mutS mutants of E. coli

 
Sequence analysis of ECOR48 DNA
As shown in Fig. 5, PCR analysis of the mutS region of the ECOR48 mutator did not exhibit gross disruption of the mutS gene relative to K-12 and O157 : H7 sequences, but it did show an insertion of approximately 0·5 kb upstream from the gene. For further examination of mutS, the gene from ECOR48 was cloned and sequenced (see Methods: GenBank accession no. AY216262). The 2562 bp structural gene is 96·1 % similar to that of E. coli K-12 and 96·2 % similar to the E. coli O157 : H7 sequence. Ninety-three of the 101 base-pair changes relative to the K-12 sequence are synonymous third-position changes; two codons contain silent mutations at both the first and third positions. Among the four non-synonymous base-pair changes found are three amino acid substitutions and a GC->TA transversion at Glu454 that creates a TAG stop codon, located 1·2 kb from the 3' end of the 2·6 kb mutS gene.



View larger version (41K):
[in this window]
[in a new window]
 
Fig. 5. PCR analysis of the mutS region in the ECOR48 mutant and wild-type strains. The left side of the panel shows a size comparison of the mutS gene in the O157 : H7 wild-type EC536 (lane 1), K-12 wild-type MG1655 (lane 2) and ECOR48 mutS (lane 3) strains, amplified using primer pair BL262/BL100. The right side of the panel shows a comparison of the region between the fhlA and mutS genes in EC536 (lane 4), MG1655 (lane 5), ECOR48 (lane 6) and Shigella dysenteriae type 1 strain SD567 (lane 7), amplified using the primer pair BL95/BL189. M, 1 kb DNA ladder.

 
The insertion upstream from the mutS gene of ECOR48 was investigated by sequencing the PCR product amplified from the 3' end of the fhlA gene to the 5' end of the mutS gene (Fig. 5). Compared with the K-12 sequence, an insertion of 479 bp was located one base pair downstream from the fhlA termination codon. The sequence is nearly identical (98·3 %) to that found at the same location in Shigella dysenteriae type 1. The lack of hybridization of a 32P-end-labelled 21-mer of this sequence to the genomes of E. coli K-12 and O157 : H7 is consistent with the absence of the sequence from their genomes (data not shown). Hybridization analysis showed the presence of the sequence in 33 out of 72 strains of the ECOR collection, however, and these were strains primarily of human and other primate origin (Brown et al., 2001b).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We report here that the amount of MutS in a natural E. coli pathogen is depleted to barely detectable levels when cells are in the stationary phase. As compared with levels in exponentially growing cells, the MutS protein in enterohaemorrhagic E. coli O157 : H7 in stationary phase was decreased about 26-fold. While the cellular amounts of MutS during exponential- and transitional-phase growth were similar in O157 : H7 and a laboratory-attenuated K-12 strain, the decline in MutS levels in O157 : H7 was nearly sixfold greater than in the K-12 strain. Comparison of protein levels in the two strains is useful here, because calculation of MutS binding to DNA mismatches in exponentially growing K-12 cells showed the amount of MutS to be nearly limiting for mismatch repair (Feng et al., 1996). These data suggest that stationary phase depletion of MutS in cells in the natural habitat could have a greater impact than recognized from studies using laboratory-attenuated strains. Such a comparison with the non-pathogenic K-12 strain suggests, for instance, that the pathogen may be predisposed to an advantageous transient mutator phenotype in the stationary-phase condition.

No antisense transcript to mutS was detected in EC536 during exponential or early stationary phase (data not shown). Curiously, an antisense transcript was observed in the EC503 deletion mutant. Although antisense transcription from rpoS and other downstream genes from mutS could provide an exquisite mechanism for effecting a transient mutator phenotype, there is no evidence that an antisense mechanism operates to mediate MutS amount under the conditions tested so far.

To date, studies have investigated increased mutagenesis caused by decreases of MutS in stationary-phase cells (Schaaper & Radman, 1989; Zhao & Winkler, 2000), though increased homeologous recombination caused by MutS depletion may be the more significant occurrence in resting or non-dividing cells in nature. Examining recombination in ‘starved’ bacteria may be especially important not only because MutS expression is compromised severely under nutrient-deprived conditions (Feng et al., 1996) but also because cells in the natural habitat constantly experience a ‘feast or famine’ existence. Such conditions may signal the need for a transient hypermutable and/or hyper-recombinagenic phenotype.

Analysis of molecular defects in the stable E. coli mutators EC503, EC535 and DEC5A revealed large deletions incapable of direct reversion to the wild-type allele. This outcome is compatible with findings on defects in natural P. aeruginosa mutators, which were frameshift, insertion, deletion or multiple substitution mutations (Oliver et al., 2002). While the nonsense mutation in the mutS gene of the ECOR48 mutator is susceptible to simple reversion, other base changes in the gene – the expected consequence for an inactive gene in a mutator background – also likely lead to its inactivation. The question then remains how an adaptive mutant can break its linkage with the mutator locus and thereby escape the deleterious effects of a heightened mutation rate. Besides reversion at the mutator locus, the mutator phenotype may be suppressed by mutation at other loci, as evidenced by results from long-term growth experiments using a mutT strain of E. coli (Tröbner & Piechocki, 1984). In these studies, isolates taken after 2200 generations of chemostat growth had lower mutation frequencies than the starting culture, but transduction analyses showed that the mutT allele itself had not changed. The lower mutation rates of the isolates were likely due to secondary mutations that suppressed the mutator phenotype (Tröbner & Piechocki, 1984).

Another option, one particularly feasible in promiscuous MMR mutators in feral settings, is repair of the mutator defect by homeologous recombination. An intact mut locus could be inherited from a Mut+ cell after horizontal transfer of sequence from either closely or distantly related species. Phylogenetic analysis of mutS-positive alleles from naturally occurring E. coli strains (including pathogens and strains of the ECOR collection) showed that mutS alleles indeed have recombined, evidenced by a phylogeny of mutS alleles different from that of the E. coli strains in which they reside (Denamur et al., 2000; Brown et al., 2001a, b). Such differences between gene evolution and strain evolution are indicative of horizontal exchange, a conclusion also supported by evidence for specific transfer events in the mutS region of the E. coli chromosome (LeClerc et al., 1999; Culham & Wood, 2000; Reid et al., 2000). The identification of mutH among a class of genes inherited from horizontal transfer (Médigue et al., 1991) suggests that defective mutH alleles may also have been repaired by homeologous recombination.

The occurrence of strains carrying deletions that encompass multiple loci, as many as 15 genes and ORFs in the EC535 mutant, is surprising in view of the environmental challenges of natural conditions. We have argued that the preponderance of mutS mutants among mutators in nature is indicative of the benefits of adaptive change generated by increased recombination and mutagenesis (LeClerc et al., 1996). The deletion mutants analysed here also affect the rpoS gene, however, and many genes needed for survival during environmental stress are regulated by the RpoS sigma factor. There are rpoS mutants of E. coli that have a selective advantage over wild-type cells as they emerge from stationary- to exponential-phase growth (Zambrino et al., 1993) and rpoS mutants are common in aged cultures (Sutton et al., 2000). It is possible that other, compensatory changes in these mutator strains overcome the limitations of growth in the absence of RpoS regulation.


   ACKNOWLEDGEMENTS
 
We gratefully acknowledge Mr William L. Payne for his dedicated and diligent efforts in the laboratory. In addition to research carried out in FDA laboratories, this work was in part supported by NIH grant RO1-CA77103 to M. W. and by resources at the Lilly Research Laboratories.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bachmann, B. J. (1987). Derivatives and genotypes of some mutant derivatives of Escherichia coli K-12. In Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology, pp. 1190–1219. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Blattner, F. R., Plunkett, G., 3rd, Bloch, C. A. & 14 other authors (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1462.[Abstract/Free Full Text]

Boe, L., Danielsen, M., Knudsen, S., Petersen, J. B., Maymann, J. & Jensen, P. R. (2000). The frequency of mutators in populations of Escherichia coli. Mutat Res 448, 47–55.[Medline]

Brown, E. W., LeClerc, J. E., Kotewicz, M. L. & Cebula, T. A. (2001a). The three R's of bacterial evolution: how replication, repair, and recombination frame the origin of species. Environ Mol Mutagen 38, 248–260.[CrossRef][Medline]

Brown, E. W., LeClerc, J. E., Li, B., Payne, W. L. & Cebula, T. A. (2001b). Phylogenetic evidence for horizontal transfer of mutS alleles among naturally occurring Escherichia coli strains. J Bacteriol 183, 1631–1644.[Abstract/Free Full Text]

Cebula, T. A. & Koch, W. H. (1990). Analysis of spontaneous and psoralen-induced Salmonella typhimurium hisG46 revertants by oligodeoxyribonucleotide colony hybridization: use of psoralens to cross-link probes to target sequences. Mutat Res 229, 79–87.[CrossRef][Medline]

Cebula, T. A. & LeClerc, J. E. (2000). DNA repair and mutators: effects on antigenic variation and virulence of bacterial pathogens. In Virulence Mechanisms of Bacterial Pathogens, 3rd edn, pp. 143–159. Edited by K. A. Brogden and others. Washington, DC: American Society for Microbiology.

Chao, L. & Cox, E. C. (1983). Competition between high and low mutating strains of Escherichia coli. Evolution 37, 125–134.

Cox, E. C. (1976). Bacterial mutator genes and the control of spontaneous mutation. Annu Rev Genet 10, 135–156.[CrossRef][Medline]

Cox, E. C. & Gibson, T. C. (1974). Selection for high mutation rates in chemostats. Genetics 77, 169–184.[Abstract/Free Full Text]

Culham, D. E. & Wood, J. M. (2000). An Escherichia coli reference collection group B2- and uropathogen-associated polymorphism in the rpoSmutS region of the E. coli chromosome. J Bacteriol 182, 6272–6276.[Abstract/Free Full Text]

Denamur, E., Lecointre, G., Darlu, P. & 9 other authors (2000). Evolutionary implications of the frequent horizontal transfer of mismatch repair genes. Cell 103, 711–721.[Medline]

Denamur, E., Bonacorsi, S., Giraud, A. & 8 other authors (2002). High frequency of mutator strains among human uropathogenic Escherichia coli isolates. J Bacteriol 184, 605–609.[Abstract/Free Full Text]

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]

Feng, G. & Winkler, M. E. (1995). Single-step purification of His6-MutH, His6-MutL, and His6-MutS repair proteins of Escherichia coli K-12. Biotechniques 19, 956–965.[Medline]

Feng, G., Tsui, H.-C. T. & Winkler, M. E. (1996). Depletion of the cellular amounts of the MutS and MutH methyl-directed mismatch repair proteins in stationary-phase Escherichia coli K-12 cells. J Bacteriol 178, 2388–2396.[Abstract]

Gibson, T. C., Scheppe, M. L. & Cox, E. C. (1970). Fitness of an Escherichia coli mutator gene. Science 169, 686–688.[Medline]

Gross, M. D. & Siegel, E. C. (1981). Incidence of mutator strains in Escherichia coli and coliforms in nature. Mutat Res 91, 107–110.[Medline]

Harris, R. S., Feng, G., Ross, K. J., Gidhu, R., Thulin, C., Longerich, S., Szigety, S. K., Winkler, M. E. & Rosenberg, S. M. (1997). Mismatch repair protein MutL becomes limiting during stationary-phase mutation. Genes Dev 11, 2426–2437.[Abstract/Free Full Text]

Horst, J.-P., Wu, T. & Marinus, M. G. (1999). Escherichia coli mutator genes. Trends Microbiol 7, 29–36.[CrossRef][Medline]

Jyssum, K. (1960). Observations on two types of genetic instability in Escherichia coli. Acta Pathol Microbiol Scand 48, 113–120.[Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[Medline]

LeClerc, J. E. & Cebula, T. A. (1997). Hypermutability and homeologous recombination: ingredients for rapid evolution. Bull Inst Pasteur 95, 97–106.[CrossRef]

LeClerc, J. E., Li, B., Payne, W. L. & Cebula, T. A. (1996). High mutation frequencies among Escherichia coli and Salmonella pathogens. Science 274, 1208–1211.[Abstract/Free Full Text]

LeClerc, J. E., Payne, W. L., Kupchella, E. & Cebula, T. A. (1998). Detection of mutator subpopulations in Salmonella typhimurium LT2 by reversion of his alleles. Mutat Res 400, 89–97.[CrossRef][Medline]

LeClerc, J. E., Li, B., Payne, W. L. & Cebula, T. A. (1999). Promiscuous origin of a chimeric sequence in the Escherichia coli O157 : H7 genome. J Bacteriol 181, 7614–7617.[Abstract/Free Full Text]

Mao, E. F., Lane, L., Lee, L. & Miller, J. H. (1997). Proliferation of mutators in a cell population. J Bacteriol 179, 417–422.[Abstract]

Matic, I., Radman, M., Taddei, F., Picard, B., Doit, C., Bingen, E., Denamur, E. & Elison, J. (1997). Highly variable mutation rates in commensal and pathogenic Escherichia coli. Science 277, 1833–1834.[Free Full Text]

Médigue, C., Rouxel, T., Vigier, P., Hénaut, A. & Danchin, A. (1991). Evidence of horizontal gene transfer in Escherichia coli speciation. J Mol Biol 222, 851–856.[Medline]

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

Miller, J. H. (1998). Mutators in Escherichia coli. Mutat Res 409, 99–106.[Medline]

Modrich, P. & Lahue, R. (1996). Mismatch repair in replication fidelity, genetic recombination, and cancer biology. Annu Rev Biochem 65, 101–133.[CrossRef][Medline]

Negri, M. C., Morosini, M. I., Baquero, M. R., Campo, Rd. R., Blazquez, J. & Baquero, F. (2002). Very low cefotaxime concentrations select for hypermutable Streptococcus pneumoniae populations. Antimicrob Agents Chemother 46, 528–530.[Abstract/Free Full Text]

Nestmann, E. R. & Hill, R. F. (1973). Population changes in continuously growing mutator cultures of Escherichia coli. Genetics 73 (Suppl 73), 41–44.[Medline]

Ochman, H. & Selander, R. K. (1984). Standard reference strains of Escherichia coli from natural populations. J Bacteriol 157, 690–693.[Medline]

Ochman, H., Elwyn, S. & Moran, N. A. (1999). Calibrating bacterial evolution. Proc Natl Acad Sci U S A 56, 12638–12643.[CrossRef]

Oliver, A., Canton, R., Campo, P., Baquero, F. & Blazquez, J. (2000). High frequency of hypermutable Pseudomonas aeruginosa in cystic fibrosis lung infection. Science 288, 1251–1254.[Abstract/Free Full Text]

Oliver, A., Baquero, F. & Blazquez, J. (2002). The mismatch repair system (mutS, mutL and uvrD genes) in Pseudomonas aeruginosa: molecular characterization of naturally occurring mutants. Mol Microbiol 43, 1641–1650.[CrossRef][Medline]

Reid, S. D., Herbelin, J., Bumbaugh, A. C., Selander, R. K. & Whittam, T. S. (2000). Parallel evolution of virulence in pathogenic Escherichia coli. Nature 406, 64–67.[CrossRef][Medline]

Richardson, A. R., Yu, Z., Popovic, T. & Stojiljkovic, I. (2002). Mutator clones of Neisseria meningitidis in epidemic serogroup A disease. Proc Natl Acad Sci U S A 99, 6103–6107.[Abstract/Free Full Text]

Schaaper, R. M. & Radman, M. (1989). The extreme mutator effect of Escherichia coli mutD5 results from saturation of mismatch repair by excessive DNA replication errors. EMBO J 8, 3511–3516.[Abstract]

Shaver, A. C., Dombrowski, P. G., Sweeney, J. Y., Treis, T., Zappala, R. M. & Sniegowski, P. D. (2002). Fitness evolution and the rise of mutator alleles in experimental Escherichia coli populations. Genetics 162, 557–566.[Abstract/Free Full Text]

Sniegowski, P. D., Gerrish, P. J. & Lenski, R. E. (1997). Evolution of high mutation rates in experimental populations of E. coli. Nature 387, 703–705.[CrossRef][Medline]

Sutton, A., Buencamino, R. & Eisenstark, A. (2000). rpoS mutants in archival cultures of Salmonella enterica serovar typhimurium. J Bacteriol 182, 4375–4379.[Abstract/Free Full Text]

Tröbner, W. & Piechocki, R. (1981). Competition growth between Escherichia coli mutL and mut+ in continuously growing cultures. Z Allg Mikrobiol 21, 347–349.[Medline]

Tröbner, W. & Piechocki, R. (1984). Selection against hypermutability in Escherichia coli during long-term evolution. Mol Gen Genet 198, 177–178.[CrossRef][Medline]

Tsui, H.-C. T., Pease, A. J., Koehler, T. M. & Winkler, M. E. (1994). Detection and quantitation of RNA transcribed from bacterial chromosomes. Methods Mol Genet 3, 179–204.

Tsui, H.-C. T., Feng, G. & Winkler, M. E. (1997). Negative regulation of mutS and mutH repair gene expression by the Hfq and RpoS global regulators of Escherichia coli K-12. J Bacteriol 179, 7476–7487.[Abstract]

Whittam, T. S., Wolfe, M. L., Wachsmuth, I. K., Ørskov, F, Ørskov. I. & Wilson, R. A. (1993). Clonal relationships among Escherichia coli strains that cause hemorrhagic colitis and infantile diarrhea. Infect Immun 61, 1619–1629.[Abstract]

Zambrino, M., Siegele, D., Almiron, A. & Kolter, R. (1993). Microbial competition: Escherichia coli mutants that take over stationary phase cultures. Science 259, 1757–1760.[Medline]

Zhao, J. & Winkler, M. E. (2000). Reduction of GC->TA transversion mutation by overexpression of MutS in Escherichia coli K-12. J Bacteriol 182, 5025–5028.[Abstract/Free Full Text]

Received 31 December 2002; accepted 22 January 2003.



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 Li, B.
Articles by Cebula, T. A.
Articles citing this Article
PubMed
PubMed Citation
Articles by Li, B.
Articles by Cebula, T. A.
Agricola
Articles by Li, B.
Articles by Cebula, T. A.


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 © 2003 Society for General Microbiology.