Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by King, M. R. Articles by Vimr, E. R. Search for Related Content PubMed PubMed Citation Articles by King, M. R. Articles by Vimr, E. R.

 Previous Article  |  Next Article 

Journal of Bacteriology, September 2007, p. 6447-6456, Vol. 189, No. 17
0021-9193/07/$08.00+0     doi:10.1128/JB.00657-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Escherichia coli K1-Specific Bacteriophage CUS-3 Distribution and Function in Phase-Variable Capsular Polysialic Acid O Acetylation ^,

Michael R. King,¹ Ross P. Vimr,¹ Susan M. Steenbergen,¹ Lodewijk Spanjaard,² Guy Plunkett III,³ Frederick R. Blattner,³ and Eric R. Vimr^1*

Laboratory of Sialobiology and Comparative Metabolomics, Department of Pathobiology, University of Illinois at Urbana-Champaign, Urbana, Illinois,¹ Netherlands Reference Laboratory for Bacterial Meningitis, Academic Medical Center, Amsterdam, The Netherlands,² Department of Genetics and Genome Center of Wisconsin, University of Wisconsin, Madison, Wisconsin³

Received 26 April 2007/ Accepted 18 June 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Escherichia coli K1 is the leading cause of human neonatal sepsis and meningitis and is important in other clinical syndromes of both humans and domestic animals; in this strain the polysialic acid capsule (K1 antigen) functions by inhibiting innate immunity. Recent discovery of the phase-variable capsular O acetylation mechanism indicated that the O-acetyltransferase gene, neuO, is carried on a putative K1-specific prophage designated CUS-3 (E. L. Deszo, S. M. Steenbergen, D. I. Freedberg, and E. R. Vimr, Proc. Natl. Acad. Sci. USA 102:5564-5569, 2005). Here we describe the isolation and characterization of a CUS-3 derivative (CUS-3a), demonstrating its morphology, lysogenization of a sensitive host, and the distribution of CUS-3 among a collection of 111 different K1 strains. The 40,207-bp CUS-3 genome was annotated from the strain RS218 genomic DNA sequence, indicating that most of the 63 phage open reading frames have their closest homologues in one of seven different lambdoid phages. Translational fusion of a reporter lacZ fragment to the hypervariable poly- domain facilitated measurement of phase variation frequencies, indicating no significant differences between switch rates or effects on rates of the methyl-directed mismatch repair system. PCR analysis of poly- domain length indicated preferential loss or gain of single 5'-AAGACTC-3' nucleotide repeats. Analysis of a K1 strain previously reported as "locked on" indicated a poly- region with the least number of heptad repeats compatible with in-frame neuO expression. The combined results establish CUS-3 as an active mobile contingency locus in E. coli K1, indicating its capacity to mediate population-wide capsule variation.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Escherichia coli strains can be subdivided into three groups on the basis of host interactions: (i) harmless commensals colonizing the gut and oropharynx, (ii) frank or intestinal pathogenic E. coli often serving as delivery vehicles for diarrheagenic toxins, or (iii) extraintestinal pathogenic E. coli (ExPEC) that may harmlessly colonize the lower intestine but can spread to other tissues, causing diseases such as urinary tract infections, sepsis, meningitis, and pneumonia, to name just a few. The morbidity, mortality, and economic costs associated with ExPEC infections make them one of the most significant causes of human and domestic animal bacterial disease (36). Although most ExPEC can still be controlled by standard antibiotic therapy, the sheer number of infections warrants increased research to better determine the mechanisms of the host-microbe interactions. Such studies should suggest new targets for therapeutic development.

ExPEC strains differ from harmless commensals by containing a megabase or more of genomic DNA encoding virulence factors such as pili, capsules, iron acquisition systems, toxins, and mammalian cell invasion factors (51). Not all ExPEC strains express the same set of virulence genes, yet pathogenic potential seems to be correlated with the overall number of virulence traits in a given strain (5). Although the roles of pili as adhesins and iron for nutrition are reasonably clear, toxins like hemolysins and cytotoxic necrotizing factor, as well as invasins, have less clear functions in pathogenesis. In contrast, capsular polysaccharides are among the longest-known and best-understood virulence factors in terms of resistance to innate immunity (40). When polysaccharides are tethered to protein carriers, the conversion from a T-cell-independent to a -dependent antigen has resulted in many of the safest and most efficacious vaccines developed over the past two decades (35). In contrast, the polysialic acid capsules of E. coli K1 and Neisseria meningitidis group B are structurally similar to the polysialic acid moiety of the mammalian neural cell adhesion molecule. This structural mimicry limits immune recognition and raises concern about cross-reactivity should vaccines ever be developed that break tolerance (43). Therefore, understanding the structures of polysialic acids is essential for targeting them as potential vaccine candidates.

Unlike the meningococcal or neural cell adhesion molecule polysialic acids, some E. coli strains express a modified capsule with O-acetyl esters variably present as covalent modifications of the individual sialic acid residues comprising the capsular polysaccharide chains (11, 17, 23, 30). The most common sialic acid, N-acetylneuraminic acid, may be acetylated at the hydroxyls on carbon positions 7 or 9 (30). These modifications are primarily controlled by the phase-variable contingency locus, neuO, a sialyl-specific O-acetyltransferase that is part of a ca. 40,000-bp accretion domain resembling lambdoid bacteriophage (phage) because of homologous immunity regions (7). Acetylase expression is controlled by a simple translational switch involving random loss or gain of a 5'-AAGACTC-3' heptanucleotide as the basic unit comprising the microsatellite designated the poly- domain (11, 46). If functional phage particles are produced from lysogenized K1 strains, neuO may spread throughout the population by standard mechanisms of phage induction and infection of sensitive hosts during coinfection by lysogenic and nonlysogenic strains. Because acetylation affects polysialic acid antigenicity as well as the physiochemical properties of the capsule, neuO has the capacity to alter the host-microbe association at every level of interaction (23, 46).

Contingency loci, such as neuO, that affect bacterial carbohydrate metabolism usually encode enzymes required for the synthesis or modification of cell surface polysaccharides (27). These loci are said to be contingent because their variable (stochastic) expression alters the cell surface in ways that are of selective advantage only in particular environments, such that at any given time at least a portion of the population is better suited (more fit) for survival under an altered environmental condition. Most contingency loci have been described in obligate commensals and facultative pathogens like Haemophilus influenzae, Neisseria meningitidis, and Campylobacter jejuni or related organisms, where these loci have known or suspected functions in host-microbe interactions (27). In most of these bacteria, phase variation between the "on" and "off" forms is controlled by slipped-strand DNA synthesis involving changes in the iteration of simple nucleotide or oligonucleotide repeats (microsatellites) located in 5' coding regions of structural genes or their promoters. The relatively high frequency of frameshifts caused by strand slippage determines whether full-length (active) or, presumably, truncated (inactive) gene products are produced during translation. Although the DNA synthesis and repair machinery necessary for slipped-stand DNA synthesis is known to operate on contingency loci heterologously expressed in Escherichia coli (45), no example of an endogenous contingency locus affecting an extracellular antigen in this organism had been described until neuO (11). If CUS-3, the designation given to the putative phage described above, were functional, K1 capsule variation would be bicontingent, depending on both neuO phase and whether a given K1 strain is a CUS-3 lysogen. Here we demonstrate that CUS-3 is an active, or reactivatable, K1-specific phage with morphology similar to other podoviruses and genetic organization similar to lambdoid phages (7). We also describe the distribution of neuO and the poly- domain in a collection of clinical K1 isolates and the construction of a neuO translational reporter fusion that facilitated quantification of phase variation frequencies under in vitro conditions.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Bacterial strains, plasmids, phages, and growth conditions. The bacterial strains, plasmids, and phages used in this study are given in Table 1. P1 vir (hereafter referred to as P1) was used for all transductions as previously described (34). Bacteria were routinely cultivated with rotary mixing at 37°C in LB (22). The antibiotics kanamycin (Km), ampicillin (Ap), tetracycline (Tc), rifampin (Rif), and nalidixic acid (Nal) were used at 50, 100, 10, 50 and 100, and 20 µg/ml, respectively. The chromogenic indicator substrates 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside and 3,4-cyclohexeneoesculetin-ß-D-galactopyranoside were used for qualitative detection of the neuO2-'lacZ translational reporter fusion as previously described (46). Plasmid pET21a was used to construct isopropyl-thio-ß-D-galactopyranoside (IPTG)-inducible neuO expression vectors with neuO inserts amplified from a pSX785 template using forward (5'-CGTAGGATACTCATATGTTAAGACTC-3') and reverse (5'-CCGCTCGAGTTGCGTGAGCTTCGCATGATAGC-3') primers, where underlined nucleotides indicate NdeI and XhoI sites, respectively. The resulting inserts were sequenced to determine the number of AAGACTC repeats in each construct. Plasmids were expressed in strain BL21(DE3) as previously described (34). Cloning of neuO into the overexpression vector was carried out by C & P Biotech Corp., Thornhill, Ontario, Canada. Strain EV36 was used as an indicator of CUS-3 infection because it was previously shown not to contain CUS-3 prophage (11).

View this table: [in this window] [in a new window]

TABLE 1. Bacterial strains, plasmids, and phages used in this study

Reporter strain construction. A reporter strain expressing a translational fusion of the neuO poly- domain (11) to lacZ absent its ribosome binding site and N-terminal coding sequence ('lacZ) was constructed by a series of transductions and transformations designed to eliminate subcloning steps. The proC677 marker from strain BW13635 was transduced into strain MC4100 by selecting Tc^r and screening for proline auxotrophy. The resulting linked (argF-lac)U169 deletion in strain EV719 was then transduced into strain EV291 by selecting Tc^r and screening for Lac^– on MacConkey-lactose agar to generate strain EV720. An in-frame deletion of the neuO catalytic hexapeptide repeat region (11) was constructed by electroporation of strain EV720 harboring pKD46 with PCR product from pKD13, selecting for Km^r to generate strain EV721. Preparation of electrocompetent cells for this step was carried out by a modified version of the method of Murphey and Campellone (28), with the inclusion of a 42°C heat shock for 15 min followed by swirling in ice-cold water for 10 min after the induction of Red functions with L-arabinose. The forward (neuO-F-PS4) and reverse (neuO-R-PS1) primers used to amplify the pKD13 template were 5'-GACTGCATGATAGCAAGAGATGTTATTTTGCGTGCATCAattccggggatccgtcgacc-3' and 5'-TTGCGTGAGCTTCGCATGATAGCTGGAGCATCTCTTGTCtgtaggctggagctgcttcg-3', respectively, containing 39 nucleotides (capitalized) targeted to the neuO catalytic domain and 20 or 21 nucleotides, respectively, complementary to the template-specific regions (shown in lower case), previously designated as priming sites P1 and P4 (9). Diagnostic Lac and Kan primers have been previously described (13). The Km^r insertion in strain EV721 was eliminated by transformation with pCP20, followed by selection for Km^r after transformation with pCE40 carrying an FRT site and in-frame 'lacZY region (13), which when cointegrated by homologous recombination into the chromosomal FRT site generated the desired 'lacZ fusion strain, designated EV722. The expected in-frame junctions between neuO2, the FRT-scar region, and 'lacZ were confirmed by DNA sequencing of the respective regions amplified by PCR with Lac and neuO-F26 primers (11, 13). The expected phase-variable phenotypes of strain EV722 on chromogenic indicator plates have been documented previously (46). Strain EV722 was used for estimating phase-variable neuO mutation using the method of Enomoto and Stocker (14). A mutL derivative of EV722 was constructed by transduction of strain EV723 with P1 grown on strain ES1301, selecting for Tc^r and screening for an elevated frequency of Rif^r mutants on selective medium, generating strain EV724.

Distribution of neuO and CUS-3 in clinical E. coli K1 isolates. A total of 111 K1 strains representing 18 different somatic (O antigen) serotypes as well as O^– (rough), nontypeable, and autoagglutinatable strains were isolated from the blood, cerebrospinal fluid, or urine of infected human neonates, young children, and adults. The presence of neuO, the relative length of the poly- domain in these strains, and open reading frame 22 (ORF22) specific to CUS-3 were detected by agarose or polyacrylamide gel electrophoresis after PCR amplification as previously described (11). For determination of the size of the poly- domain, the forward and reverse primers used were 5'-GGTAAAATAACGTAGGATACTAATATG-3' and 5'-GACCCATTATCATCAACGGAAAAC-3', respectively. The forward and reverse primers used to amplify the entire neuO gene were 5'-AGCACTAAATGTTTCGTTGGCGTC-3' and 5'-AATATTGGTAATATGTCTGCATGATG-3', respectively.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Isolation and morphological characterization of CUS-3a. Attempts to isolate viable CUS-3 particles by determining spot titers of culture supernatants from strain RS218 on EV36 indicator cells were unsuccessful. However, by adding 0.5 to 1.0 µg/ml of mitomycin C to liquid cultures when cells reached an A₆₀₀ of 0.1 and allowing growth to continue with vigorous aeration for 3 to 6 h, spot titers for cell-free culture supernatants resulted in identification of three independent plaques with similar turbid centers (see Fig. S1 in the supplemental material). When a portion of phage from one of these plaques was used to reinfect EV36, CUS-3 titers of >10⁷ PFU/ml were consistently obtained from lysates after treatment with chloroform. The most likely reason for the initial difficulty in isolating CUS-3 was that the prophage in RS218 contains a reversible mutation limiting detection to a few revertant particles. Resuscitation of phage using a sensitive host has been described for cryptic prophage ST64B from Salmonella enterica (15). Plaques with turbid centers were surrounded by halos resulting from an altered refractive index caused by phage depolymerase (endo-N) diffusion and capsule hydrolysis of cells surrounding the plaques (see Fig. S1 in the supplemental material). As shown previously for the K1-specific virulent phage K1F, diffusible endo-N represents overproduced tailspike proteins that did not attach to virions prior to lysis (32, 50). Further evidence for the strain specificity of the isolated phage was resistance of the CUS-3 lysogen, EV291, to infection and the resistance of EV36 acapsular mutants EV136 and EV93, which have defects in capsule synthesis or export, respectively (Table 1). The combined results strongly suggested that the phage originally isolated from RS218, and subsequently grown on EV36, is a CUS-3 derivative, which we have designated CUS-3a to indicate potential reversion of an unknown mutation that limited induction of infective virions.

On the basis of the putative CUS-3 genomic organization (11), we predicted that phage particles would resemble Podoviridae, with characteristic short tails and radiating tailspike proteins. CUS-3a isolated from infected EV36 was prepared by differential centrifugation and stained with ammonium molybdate for transmission electron microscopic (TEM) analysis (6). Figure 1A shows the heads with short tails expected of podoviruses. With increased magnification, the short tail and spike proteins were evident (Fig. 1B). By analogy with the phage K1F endo-N tailspike protein (32, 44), we suggest that the equivalent CUS-3a tailspikes mediate binding and depolymerization of the polysialic acid capsule during infection. This organelle thus defines host range, limiting CUS-3 to infection of E. coli K1 expressing capsular polysialic acid. Both virtual and real restriction analyses confirmed that the CUS-3a particles examined by TEM were derived from CUS-3 (Fig. 2).

View larger version (105K): [in this window] [in a new window]

FIG. 1. TEM of negatively stained CUS-3 phage particles. Magnification: x150,000 (A) and x200,000 (B). The white arrow points to the tip of the portal with emanating polysialic acid depolymerase protein endo-N. Particle morphology is similar to that of the lytic K1-specific phage K1F (32), as well as podoviruses in general.

View larger version (64K): [in this window] [in a new window]

FIG. 2. Virtual and real endonuclease digestion of CUS-3 DNA confirms phage isolation. CUS-3 particles from an EV36 lysate were collected by differential centrifugation, and the DNA was subjected to digestion with EcoR1 (lane 2). The virtual digestion profile of DNA from nucleotides 2621662 through 2661871 of the strain RS218 chromosomal contig is identical (lane 4) to the actual digestion product (lane 2). Lambda HindIII (lane 1) and Promega 1-kb ladder (lane 3) are shown for comparison. The sizes (in bp) of the virtual digestion products are given on the right.

To determine if other naturally occurring CUS-3 lysogens were generally incapable, like the prototypic strain RS218, of producing high phage titers after mitomycin C induction, we compared the titers of nine strains that were both previously shown to produce enzymatically active NeuO and that tested positive for two different CUS-3 open reading frames by PCR analysis (11). Four (44%) of the isolates yielded titers of <10² PFU/ml, similar to RS218, while the other isolates produced titers ranging from 6.0 x 10² to 1.3 x 10⁶ PFU/ml (see Table S1 in the supplemental material). In contrast, lysogens EV36-1 and EV36-2, which were isolated after infection with CUS-3a, yielded titers of 1.9 x 10⁶ and 1.6 x 10⁶ PFU/ml, respectively (see Table S1). Natural E. coli K1 isolates lysogenized by CUS-3 are thus capable of producing phage titers over at least a 4-log range, while nearly half may carry defective prophage capable of reactivation like CUS-3a. It seems clear that because any selective advantage to fitness conferred by the prophage would be lost upon induction due to cell death, there could be a tendency for the population to accumulate defective lysogens, including deletions of CUS-3 ORFs other than neuO (11).

Genetic and functional organization of the CUS-3 genome. Previous annotation of the CUS-3 genomic DNA sequence suffered from sequence errors in the relevant region of the chromosome bracketed by the duplicated argW insertion site (11, 46). The RS218 genome was recently completed, defining three contigs that include a (i) chromosome, containing the CUS-3 genome, (ii) large F-like plasmid, and (iii) replicative form of filamentous prophage CUS-1 (http://www.genome.wisc.edu/sequencing/rs218.htm). The E. coli uropathogenic strain UT189 genome includes a region homologous to CUS-3, suggesting that this strain is also a CUS-3 lysogen (GenBank accession number CP000243). Annotation of the 40,207-bp CUS-3 DNA sequence is given in Table S2 in the supplemental material (GenBank accession number CP000711). As shown in Fig. 3, most of the 63 predicted ORFs have their closest homologue in one of seven different lambdoid phage or phage remnants, indicating the evident mosaic architecture and likely extensive recombinational events that generated CUS-3 during evolution. Five different boundary sequences (bsq) indicate regions of high nucleotide identity to lambdoid regions (see Table S2 in the supplemental material), which may represent ancestral recombination events between different phage modules (8). The high sequence identity of bsq1 extends to the 5' region of endo-N (ORF3), which is required for the attachment of the tailspike to the tail regions of K1-specific phages (38, 44). Except in the regions of the host range determinants, high nucleotide similarity with HK620 is evident from the identity diagonal shown in Fig. 3B. Although the exact functions of most CUS-3 ORFs are unknown (see Table S2 in the supplemental material), the putative ant (ORF4) and mnt (ORF6) genes are reminiscent of the phage P22 immI or lysogeny region, which regulates the lysis decision by affecting the activity of the immC (c2) repressor (ORF44) by binding antirepressor Ant protein. Interestingly, the putative lysogeny region of CUS-3 lacks an arc repressor homologue, suggesting that CUS-3 would behave, especially during infection, like a P22 arc mutant if the Ant homologue were overproduced, thereby reducing lysogenization (25). A description of the CUS-3 integration site and associated int gene (ORF1) has been reported previously (46). The proximity of the host range endo-N determinant (ORF3) to neuO (ORF2) suggests that O acetylation might have evolved as a mechanism of receptor modification to prevent infection of CUS-3 lysogens by other phages. Alternatively, or in addition, it may have evolved by providing a fitness advantage to K1 lysogens during host-microbe interactions. As detailed below, the data support the second hypothesis.

View larger version (24K): [in this window] [in a new window]

FIG. 3. Genetic organization and homologies of CUS-3. A. ORFs are shown as numbered, color-coded arrows indicating the predicted directions of transcription. Each color refers to the indicated phage with the closest homologue shown below the line, which indicates the CUS-3 region of the RS218 chromosome from attL to attR. ORFs with uncertain identity are shown in black, as are nucleotide boundary sequences (bsq). The white box in ORF3 indicates continuation of the bsq1 sequence defining the N terminus of the encoded endo-N structural gene product. B. Pustell matrix analysis (MacVector 9.0.2) of CUS-3 versus HK620. The identity diagonal from the upper left to lower right corner indicates nucleotide similarity between the respective phage genomes shown with the same orientation as the CUS-3 genome in panel A. Analysis parameters for a window size of 30 nucleotides were as follows: minimum score of 65%, hash value of 6, jump of 3.

Lysogenization of strain EV36 and susceptibility of lysogens to K1-specific phage infection. To determine whether CUS-3a can lysogenize a K1 host, survivors of infection were recovered from lysates by plating for single colonies on rich medium. Thirteen survivors were single-colony isolated after two passages to remove residual CUS-3a. These strains were then cross-streaked against either virulent K1-specific phage K1F or CUS-3a. The two most likely reasons for surviving the original infection are spontaneous loss of capsule or lysogenization. Sensitivity of most of the survivors to K1F ruled out loss of capsule as an explanation, while resistance to CUS-3a indicated lysogenization and subsequent immunity. PCRs with neuO-specific primers indicated that the CUS-3-resistant survivors harbored the phage genome, as confirmed by positive PCR amplification of CUS-3 ORF22 (not shown). Sensitivity of most survivors to K1F indicates that the function of neuO is not primarily to block receptor recognition or cleavage by other K1-specific phages. Although we have not carried out a one-step growth curve analysis, the kinetics of phage K1F and PK1E phage binding to lysogenized EV36 was unaffected by the prophage, and plaque size was unaltered compared to unlysogenized EV36, strongly suggesting equivalent burst sizes. The rapid phase variation of neuO is also inconsistent with functions limited to exclude other K1-specific phages (46). Finally, strain EV36-3 (Table 1), which was recovered as a lysogen after infection with CUS-3 from strain EV708 and that contains the neuO1 deletion replaced by a Km^r cassette (Table 1), was Km^r and harbored CUS-3 ORF22 as detected by PCR. Therefore, isolation and characterization of EV36-3 indicates, as expected, that neuO is not required for CUS-3 maturation, infection, or lysogenization.

Frequency of neuO phase variation in the wild type and an isogenic methyl-directed mismatch repair mutant. To determine the frequencies of neuO switching between "on" (B, black or blue) or "off" (W, white) phases, we constructed a translational fusion of 'lacZ to the poly- domain (neuO2) and used indicator plates to score colonies for ß-galactosidase (see Materials and Methods). This approach distinguishes between loss or gain of 5'-AAGACTC-3' heptad repeats, which results in translationally full-length (active) NeuO when present in multiples of three repeats. Evidence that gene inactivation results from frameshift mutation and synthesis of truncated gene products is shown by the overexpression of three independent wild-type neuO subclones bearing poly- domains with two different out-of-frame lengths (Fig. 4). These strains overproduced truncated polypeptides with the predicted molecular masses (Fig. 4, lanes 2, 4, and 8). Although we did not detect obvious overproduction of the predicted in-frame gene product (Fig. 4, lane 6), suggesting frameshifting to the "off" form (or some other problem with neuO⁺ expression), the induced extract from this strain was the only neuO clone with detectable transacetylase activity, indicating synthesis of active transacetylase from at least a minority of plasmids in which the in-frame domain (length determined by DNA sequencing) was retained. As shown in Table 2, the frequency of switching from B to W was not significantly different (P = 0.12, Mann-Whitney U test) from the opposite direction despite there being twice as many out-of-frame than in-frame switches possible to generate this phenotype. This lack of switch bias suggested, in contrast to H. influenzae (33), that the mechanism controlling poly- stability in E. coli K1 favors microsatellite expansion over contraction. In total, 44/49 (89.8%) of the B-to-W switches occurred via gain of a single AAGACTC repeat (Fig. 5). Although our electrophoretic method does not accurately discriminate between a switch involving one or two additions or deletions (11), that most B revertants resulted from contraction strongly suggests a bias favoring gain or loss of one repeat unit (Fig. 5). Occasionally, we observed a dramatic contraction (Fig. 5) but never an expansion of similar magnitude. Qualitatively identical results were observed when following the variation from W to B. A similarly strong bias in favor of expansion was observed for a tetranucleotide repeat in a heterologous E. coli system (12), whereas contraction was favored in H. influenzae (33), suggesting species-specific differences in the mechanisms of DNA synthesis or repair involved in slipped-strand mispairing.

View larger version (81K): [in this window] [in a new window]

FIG. 4. Poly- domain length dictates the neuO reading frame. Strains harboring plasmids overexpressing neuO with the indicated numbers of AAGACTC repeats (poly-) were induced with IPTG (even-numbered lanes). Whole-cell extracts of uninduced (odd-numbered lanes) and induced strains were fractionated by polyacrylamide gel electrophoresis, and polypeptides were visualized by staining with Coomassie dye as previously described (34). Sizes of molecular mass markers (lane M) are given by on the left. Relative NeuO activity was measured in induced extracts as described previously, where a minus sign indicates no detectable transacetylation of a model substrate (11). Asterisks indicate overproduced, truncated gene products with molecular masses expected from the neuO sequence (accession number AY779018). The predicted mass of active NeuO (not visible in lane 6) is ca. 32 kDa. Each pair of extracts, starting from lane 1, represents the results obtained with plasmids pSX788, pSX789-1, pSX790, and pSX789-2, respectively.

View this table: [in this window] [in a new window]

TABLE 2. Mutation frequencies in strain with the neuO2-'lacZ reporter fusion

View larger version (58K): [in this window] [in a new window]

FIG. 5. Poly- variation in successive BW transitions. The schematic diagram at the top indicates repeated passage of a founder B colony, chosen at random and found to have a domain length of 18 repeats (third lane from left). W colonies were chosen randomly and analyzed after successive switches. Poly- domain lengths were estimated as previously described (11), with X174 HinFI and HaeIII size markers (first two left lanes, respectively) shown in bp on the left. Note that with the exception of the W colony at step 1, there was one predominant domain amplified after each successive switch.

Microsatellite repeats ranging in length from 1 to 13 nucleotides were destabilized by mismatch repair defects in yeast (39). However, whereas bacterial mono- and dinucleotide repeats were unstable in mut strains with defects in methyl-directed mismatch repair, tetranucleotides were not (2), suggesting uncoupling of mismatch repair from the mutation rates of repeats composed of three or more nucleotides. Although the frequencies of poly- switching in a mutL background were significantly different from wild type, the magnitude of the effects was less than fourfold (Table 2). In contrast, the frequency of spontaneous Rif^r EV724 mutants was increased 50- to 100-fold by mutL (not shown). Furthermore, the effects of mutL were biased toward the "off" phase, suggesting a contribution to the frequencies of switching from spontaneous inactivation of the lacZ reporter. When taken together, our results indicate that the methyl-directed mismatch repair system has little effect on the stability of the poly- domain in isogenic strains.

The enigma of strain C375. E. coli K1 strain C375 was independently reported to have the lowest transacetylase specific activity and least amount of capsule acetylation of any phase-variable K1 strain examined (17, 30). Furthermore, C375 appeared to be locked into the on phase (switch frequency < 1/5,000), whereas other strains were observed to vary between 2 and 3% (30). These observations suggested there was either an alternative mechanism of capsule modification or that the poly- domain in strain C375 did not undergo expansion or contraction at a detectable frequency. Although we obtained several vials of freeze-dried strain C375 from its source (30), none contained viable bacteria. However, PCR analysis indicated a poly- length of three repeats (Fig. 6), the least number of heptanucleotides, with the exception of a perfect deletion, compatible with in-frame translation of neuO. Because microsatellite length is positively correlated with mutation frequency (10), our results indicate that the failure of C375 to detectably vary was due to its relatively short poly- domain. DNA sequencing confirmed that the poly- domain with the predicted length shown in Fig. 6 contains three repeats of the AAGACTC oligonucleotide. Furthermore, a viable strain (see Table S3 in the supplemental material, strain no. 4), with three AAGACTC repeats, had a switch rate comparable to that of C375 (30). Although an artificial neuO construct lacking all repeats still produced active transacetylase (11), our results and those of others (17, 30) suggest complex strain-dependent effects of poly- length on the degree of acetylation. This correlation suggests an effect of the poly- domain on enzyme localization, stability, or catalytic efficiency (11, 42, 46).

View larger version (59K): [in this window] [in a new window]

FIG. 6. Poly- domain length in strain C375. Primers amplifying the poly- domain (lane 1) or complete neuO gene (lane 2) were used to amplify DNA recovered from nonviable, freeze-dried samples of strain C375. The bracket at left indicates the region of the gel that would detect poly- domains with 1 to 40 5'-AAGACTC-3' repeats. Lanes 3 and 4 represent X174 HinFI and HaeIII size markers, respectively, with lengths given in bp on the right.

Distribution of CUS-3 and neuO. The capability of 111 different E. coli K1 clinical isolates to vary capsule form was analyzed by PCR. The presence of CUS-3 was then correlated with serotype and poly- length in all positive strains (see Table S3 in the supplemental material). There was 100% concordance between O18 (16/16) and O45 (5/5) strains with CUS-3/neuO, while only 1/8 O7 strains was not positive for neuO. The absence of CUS-3 ORF22 but presence of neuO in some strains could indicate loss of phage genes unrelated to fitness while retaining the ability to phase vary. The correlation between neuO and typically "high-virulence" O18 and O45 serotypes (5) is consistent with the importance of CUS-3 in the K1 host-microbe interaction (23). Interestingly, every serotype shown in Table S3 of the supplemental material that was represented by at least two strains included at least one CUS-3 lysogen, while overall neuO prevalence was 61% with 11% of these strains lacking ORF22, presumably indicating inactivation of CUS-3 by spontaneous deletion events that remove phage genes while retaining neuO. Our results predict that the incidence of CUS-3 will rise in O83 strains, which are emerging in cases of meningitis in The Netherlands (4). Poly- lengths in neuO⁺ strains varied from 3 to 67 AAGACTC repeats, and length was not correlated with serotype (see Fig. S2 in the supplemental material). The wide distribution of microsatellite lengths has been observed previously in epidemiologically distinct H. influenzae strains, suggesting that repeat number reflects the equilibrium between host environment and strain, i.e., "best fitness" (27). It will be interesting to determine the dynamics of poly- variation during model infections. These studies are in progress.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Previous results led to the hypothesis that polysialic acid capsule modification coevolved during the association between CUS-3 and E. coli K1 by conferring a fitness advantage to the bacterium during colonization of animal hosts (11, 46). Assuming that CUS-3 evolved relatively recently, two predictions of the hypothesis are that the K1 population is composed of both lysogenized and nonlysogenized (CUS-3-sensitive) hosts and that CUS-3 is an active or reactivatable instead of defective prophage. The results of this communication confirm both of these predictions, indicating that CUS-3a can infect and lysogenize a sensitive K1 host. Our analysis of CUS-3 and neuO distribution among a diverse set of clinical K1 serotypes indicates 100% prevalence in common disease isolates with the O18 and O45 antigens but reduced prevalence or absence in low-virulence or less commonly isolated serotypes. Interestingly, the O83 serotype is isolated mainly in The Netherlands (4), but the incidence of neuO was only 20% in these strains (see Table S3 in the supplemental material). Our results suggest that the incidence of O83 lysogens will increase over time, which we presume occurs during cocolonization with CUS-3⁺ strains. Both the extent of acetylation and the position of O-acetyl groups influences polysaccharide immunogenicity in a wide range of microbial pathogens, including Shigella flexneri, Neisseria meningitidis, group B Streptococcus, Staphylococcus aureus, Salmonella enterica serovar Typhi, and Cryptococcus neoformans (reference 24 and references therein).

As predicted from annotation of the complete CUS-3 genome, phage particle morphology resembles podoviruses, while the predicted phage immunity region places CUS-3 within the lambdoid group (7). CUS-3 appears to be a close relative of HK620, acquiring both endo-sialidase and neuO genes in order to expand host range. Acquisition of endo-N results in the observed host range specificity of CUS-3 for E. coli K1, just as a lytic T7-like phage acquired the endo-sialidase gene to gain K1 host specificity in phage K1F (32, 38). Although the exact functions conferred by polysialic acid O acetylation to the host-pathogen interaction are not yet completely defined, polysaccharide modification genes carried on prophages significantly affect pathogenic potential. For example, the glucosyltransferase GtrV, encoded by a temperate S. flexneri phage, affects epithelial cell invasion by modulating the outer membrane activity of the bacterial type III secretion apparatus (52). Similarly, the variable addition of bulky acetyl groups to the polysialic acid chains is known to affect antigenicity of the K1 capsule, making it likely that the modification also affects bacterial interactions with host epithelial or endothelial cells and the innate immune system (23). Interestingly, the S. flexneri phages responsible for serotype conversion add either acetyl or glycan units, or both, to different Shigella lipopolysaccharides (1). Like CUS-3, S. flexneri phages insert into the argW tRNA gene using an integrase nearly identical to that of CUS-3 (see Table S2 in the supplemental material). We suggest that so-called phage "receptor-modifying enzymes" play an important role in a variety of host-microbe interactions despite the possibility that they might have originally evolved to exclude other phages.

Methyl-directed mismatch repair is a mechanism for strand discrimination, excision, and repair synthesis of mismatched nucleotides in DNA. Recognition of the mismatches is initiated by MutS binding to one to three unpaired nucleotides, poorly to four unpaired nucleotides, and not at all to five (31), followed by recruitment of MutL and MutH, excision of mismatches, and DNA repair (37). Repeated attempts to introduce a mutS::Tn10 mutation from strain ES1481 into reporter strain EV723 were unsuccessful, resulting in our focus on the effects of a mutL defect on poly- mutation frequency in EV724. Previous investigators found no effect of mut defects on stability of microsatellites composed of a tetranucleotide or greater tandem repeats in bacteria (3, 33). Our results are consistent with this conclusion, suggesting that the effect of mut on repeat stability is confined to microsatellites with three or fewer nucleotide repeats. This conclusion suggests that mutator strains do not play a significant role in the emergence of pathogenic clones involving microsatellites with >4 nucleotides in their repeats. Such microsatellites presumably evolved by conferring high localized mutation rates independent of cell mutator status or direct effects on protein function when cells are in the "on" phase (11). That elimination of all repeats does not block NeuO activity is not evidence that repeat length has no effect on enzymatic activity (11, 27). However, the almost complete absence of in-phase bacterial triplet repeats argues against a general effect of long repeats, such as that of the poly- domain, on enzyme function.

Despite a variety of results correlating capsule phase variation with pathogenesis (23), it is currently impossible to precisely define all of the functions of CUS-3, probably because of the importance of capsule O acetylation at every stage of the host-pathogen interaction. In humans, E. coli K1 may harmlessly colonize the neonatal large intestine but sometimes causes diarrhea when expressing verotoxin and, possibly, cdt (18, 29). From the intestine, it may go on to colonize the bladder or kidneys (cystitis and pyelonephritis), progressing to urosepsis in some patients (19, 20). Another common pathogenic scenario is breaching of the intestinal mucosa followed by propagation to high numbers systemically (bacteremia and sepsis) and subsequent invasion of the blood-brain barrier (meningitis). In poultry, E. coli K1 colonizes the lungs (pneumonia), from where it may spread systemically (colibacillosis) (26). In all of these diseases, the capsule plays a crucial role in pathogenesis. Indeed, polysialic acid is even implicated as a factor during colonization of Acanthamoeba, with which it may share an environmental niche (21). In addition to the known function of capsule acetylation in intestinal sialidase resistance and avoidance of specific immune responses, capsule modification may affect eukaryotic cell binding and invasion and resistance to innate immunity (23). The diverse functions of capsule acetylation and its modulation by phase variation are currently being investigated.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 
Using an in vitro approach, Bergfeld et al. (A. K. Bergfeld, H. Claus, U. Vogel, and M. Mühlenhoff, J. Biol. Chem. 282:22217-22227, 2007) recently confirmed the conclusion that the relative trans-acetylase activity of NeuO increases with increasing poly- domain length (11, 46), consistent with previous results (17, 30) and the results shown in Fig. 6.

 
This research was supported by NIH grant AI43025 and an American Society for Microbiology Undergraduate Research Fellowship and James Scholar Award to M.R.K.

We are grateful to Dean Scholl for helpful discussions during the isolation of CUS-3. We thank James Johnson, Scott Weissman, and James Slauch for kindly providing some of the bacterial strains used in this study.

 
* Corresponding author. Mailing address: 2522 VMBSB, 2001 South Lincoln Avenue, Urbana, IL 61802. Phone: (217) 333-8502. Fax: (217) 244-7421. E-mail: ervimr{at}uiuc.edu

Published ahead of print on 29 June 2007.

Supplemental material for this article may be found at http://jb.asm.org/.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADDENDUM IN PROOF REFERENCES

 

1
1. Allison, G. E., and N. K. Verma. 2000. Serotype-converting bacteriophages and O-antigen modification in Shigella flexneri. Trends Microbiol. 8:17-23.[CrossRef][Medline]
2
2. Bayliss, C. D., W. A. Sweetman, and E. R. Moxon. 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]
3
3. Bayliss, C. D., T. van de Ven, and E. R. Moxon. 2002. Mutations in polI but not mutSLH destabilize Haemophilus influenzae tetranucleotide repeats. EMBO J. 21:1465-1476.[CrossRef][Medline]
4
4. Bonacorsi, S., and E. Bingen. 2005. Molecular epidemiology of Escherichia coli causing neonatal meningitis. Int. J. Med. Microbiol. 295:373-381.[CrossRef][Medline]
5
5. Bonacorsi, S., O. Clermont, V. Houdouin, C. Cordevant, N. Brahimi, A. Marecat, C. Tinsley, X. Nassif, M. Lange, and E. Bingen. 2003. Molecular analysis and experimental virulence of French and North American Escherichia coli neonatal meningitis isolates: identification of a new virulent clone. J. Infect. Dis. 187:1895-1906.[CrossRef][Medline]
6
6. Carlson, K. 2005. Appendix: working with bacteriophages: common techniques and methodological approaches, p. 437-494. In E. Kutter and A. Sulakvelidze (ed.), Bacteriophage: biology and applications. CRC Press, Boca Raton, FL.
7
7. Casjens, S. R. 2005. Comparative genomics and evolution of the tailed-bacteriophages. Curr. Opin. Microbiol. 8:451-458.[CrossRef][Medline]
8
8. Clark, A. J., W. Inwood, T. Cloutier, and T. S. Dhillon. 2001. Nucleotide sequence of coliphage HK620 and the evolution of lambdoid phages. J. Mol. Biol. 311:657-679.[CrossRef][Medline]
9
9. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.[Abstract/Free Full Text]
10
10. DeBolle, X., C. D. Bayliss, D. Field, T. van de Ven, N. J. Saunders, D. W. Hood, and E. R. Moxon. 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]
11
11. Deszo, E. L., S. M. Steenbergen, D. I. Freedberg, and E. R. Vimr. 2005. Escherichia coli K1 polysialic acid O-acetyltransferase gene, neuO, and the mechanism of capsule form variation involving a mobile contingency locus. Proc. Natl. Acad. Sci. USA 102:5564-5569.[Abstract/Free Full Text]
12
12. Eckert, K. A., and G. Yan. 2000. Mutational analyses of dinucleotide and tetranucleotide microsatellites in Escherichia coli: influence of sequence on expansion mutagenesis. Nucleic Acids Res. 14:2831-2838.
13
13. Ellermeier, C. D., A. Janakiraman, and J. M. Slauch. 2002. Construction of targeted single copy lac fusions using Red and FLP-mediated site-specific recombination in bacteria. Gene 290:153-161.[CrossRef][Medline]
14
14. Enomoto, M., and B. A. D. Stocker. 1975. Integration, at hag or elsewhere, of H2 (phase-2 flagellin) genes transduced from Salmonella to Escherichia coli. Genetics 81:595-614.[Abstract/Free Full Text]
15
15. Figueroa-Bossi, N., and L. Bossi. 2004. Resuscitation of a defective prophage in Salmonella cocultures. J. Bacteriol. 186:4038-4041.[Abstract/Free Full Text]
16
16. Gonzalez, M. D., C. A. Lichtensteiger, and E. R. Vimr. 2001. Adaptation of signature-tagged mutagenesis to Escherichia coli K1 and the infant-rat model of invasive disease. FEMS Microbiol. Lett. 198:125-128.[CrossRef][Medline]
17
17. Higa, H. H., and A. Varki. 1988. Acetyl-coenzyme A: polysialic acid O-acetyltransferase from K1-positive Escherichia coli. J. Biol. Chem. 263:8872-8878.[Abstract/Free Full Text]
18
18. Johnson, J. R., E. Oswald, T. T. O'Bryan, M. A. Kuskowski, and L. Spanjaard. 2002. Phylogenetic distribution of virulence-associated genes among Escherichia coli isolates associated with neonatal bacterial meningitis in The Netherlands. J. Infect. Dis. 185:774-784.[CrossRef][Medline]
19
19. Johnson, J. R., K. L. Owens, C. R. Clabots, S. J. Weissman, and S. B. Cannon. 2006. Phylogenetic relationships among clonal groups of extraintestinal pathogenic Escherichia coli as assessed by multi-locus sequence analysis. Microbes Infect. 8:1702-1713.[CrossRef][Medline]
20
20. Johnson, J. R., S. J. Weissman, A. L. Stell, E. Trintchina, D. E. Dykhuizen, and E. V. Sokurenko. 2001. Clonal and pathotypic analysis of archetypal Escherichia coli cystitis isolate NU14. J. Infect. Dis. 184:1556-1565.[CrossRef][Medline]
21
21. Jung, S.-Y., A. Matin, K. S. Kim, and N. A. Khan. 2007. The capsule plays an important role in Escherichia coli K1 interactions with Acanthamoeba. Int. J. Parasitol. 37:417-423.[CrossRef][Medline]
22
22. Kalivoda, K. A., S. M. Steenbergen, E. R. Vimr, and J. Plumbridge. 2003. Regulation of sialic acid catabolism by the DNA binding protein NanR in Escherichia coli. J. Bacteriol. 185:4806-4815.[Abstract/Free Full Text]
23
23. King, M. R., S. M. Steenbergen, and E. R. Vimr. 2007. Going for baroque at the Escherichia coli K1 cell surface. Trends Microbiol. 15:196-202.[CrossRef][Medline]
24
24. Kubler-Kielb, J., E. Vinogradov, C. Chu, and R. Schneerson. 2007. O-acetylation in the O-specific polysaccharide isolated from Shigella flexneri serotype 2a. Carbohydr. Res. 342:643-647.[CrossRef][Medline]
25
25. Maloy, S. R., V. J. Stewart, and R. K. Taylor. 1996. Genetic analysis of pathogenic bacteria: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
26
26. Moulin-Schouleur, M., C. Schouler, P. Tailliez, M. R. Kao, A. Brée, P. Germon, E. Oswald, J. Mainil, M. Blanco, and J. Blanco. Common virulence factors and genetic relationships between O18:K1:H7 Escherichia coli isolates of human and avian origin. J. Clin. Microbiol. 44:3484-3492.
27
27. Moxon, R., C. Bayliss, and D. Hood. 2006. Bacterial contingency loci: the role of simple sequence DNA repeats in bacterial adaptation. Annu. Rev. Genet. 40:303-333.
28
28. Murphy, K. C., and K. G. Campellone. 2003. Lambda Red-mediated recombinogenic engineering of enterohemorrhagic and enteropathogenic E. coli. BMC Mol. Biol. 4:11-22.[CrossRef][Medline]
29
29. Olesen, B., C. Jensen, K. Olsen, V. Fussing, P. Gerner-Smidt, and F. Scheutz. 2005. VTEC O117:K1:H7. A new clonal group of E. coli associated with persistent diarrhoea in Danish travelers. Scand. J. Infect. Dis. 37:288-294.[Medline]
30
30. Ørskov, F., I. Ørskov, A. Sutton, R. Schneerson, W. Lin, W. Egan, G. E. Hoff, and J. B. Robbins. 1979. Form variation in Escherichia coli K1: determined by O-acetylation of the capsular polysaccharide. J. Exp. Med. 149:669-685.[Abstract/Free Full Text]
31
31. Parker, B. O., and M. G. Marinus. 1992. Repair of DNA heteroduplexes containing small heterologous sequences in Escherichia coli. Proc. Natl. Acad. Sci. USA 89:1730-1734.[Abstract/Free Full Text]
32
32. Petter, J. G., and E. R. Vimr. 1993. Complete nucleotide sequence of the bacteriophage K1F tail gene encoding endo-N-acylneuraminidase (endo-N) and comparison to an endo-N homolog in bacteriophage PK1E. J. Bacteriol. 175:4354-4363.[Abstract/Free Full Text]
33
33. Ren, Z., H. Jin, P. W. Whitby, D. J. Morton, and T. L. Stull. 1999. Role of CCAA nucleotide repeats in regulation of hemoglobin and hemoglobin-haptoglobin binding protein genes of Haemophilus influenzae. J. Bacteriol. 181:5865-5870.[Abstract/Free Full Text]
34
34. Ringenberg, M. A., S. M. Steenbergen, and E. R. Vimr. 2003. The first committed step in the biosynthesis of sialic acid by Escherichia coli K1 does not involve a phosphorylated N-acetylmannosamine intermediate. Mol. Microbiol. 50:961-975.[CrossRef][Medline]
35
35. Robbins, J. B., and R. Schneerson. 2004. Future vaccine development at NICHD. Ann. N. Y. Acad. Sci. 1038:49-59.[CrossRef][Medline]
36
36. Russo, T. A., and J. R. Johnson. 2003. Medical and economic impact of extraintestinal infections due to Escherichia coli: focus on an increasingly important endemic problem. Microbes Infect. 5:449-456.[CrossRef][Medline]
37
37. Schofield, M., and P. Hsieh. 2003. DNA mismatch repair: molecular mechanisms and biological function. Annu. Rev. Microbiol. 57:579-608.[CrossRef][Medline]
38
38. Scholl, D., and C. Merril. 2005. The genome of bacteriophage K1F, a T7-like phage that has acquired the ability to replicate on K1 strains of Escherichia coli. J. Bacteriol. 187:8499-8503.[Abstract/Free Full Text]
39
39. Sia, E. A., R. J. Kokoska, M. Dominska, P. Greenwell, and T. D. Petes. 1997. Microsatellite instability in yeast: dependence on repeat unit size and DNA mismatch repair genes. Mol. Cell. Biol. 17:2851-2858.[Abstract]
40
40. Silver, R. P., and E. R. Vimr. 1990. Polysialic acid capsule of Escherichia coli K1, p. 39-60. In B. H. Iglewski and V. Miller (ed.), The Bacteria, vol. XI, molecular basis of bacterial pathogenesis. Academic Press, Inc., San Diego, CA.
41
41. Steenbergen, S. M., and E. R. Vimr. 1990. Mechanism of polysialic acid chain elongation in Escherichia coli K1. Mol. Microbiol. 4:603-611.[Medline]
42
42. Steenbergen, S. M., Y.-C. Lee, W. F. Vann, J. Vionnet, L. F. Wright, and E. R. Vimr. 2006. Separate pathways for O acetylation of polymeric and monomeric sialic acids and identification of sialyl O-acetyl esterase in Escherichia coli K1. J. Bacteriol. 188:6195-6206.[Abstract/Free Full Text]
43
43. Stein, D. M., J. Robbins, M. A. Miller, F.-Y.C. Lin, and R. Schneerson. 2006. Are antibodies to the capsular polysaccharide of Neisseria meningitidis group B and Escherichia coli K1 associated with immunopathology? Vaccine 24:221-228.[CrossRef][Medline]
44
44. Stummeyer, K., D. Schwarzer, H. Claus, U. Vogel, R. Gerardy-Schahn, and M. Mühlenhoff. 2006. Evolution of bacteriophages infecting encapsulated bacteria: lessons from Escherichia coli K1-specific phages. Mol. Microbiol. 60:1123-1135.[CrossRef][Medline]
45
45. Torres-Cruz, J., and M. W. van der Woude. 2003. Slipped-strand mispairing can function as a phase variation mechanism in Escherichia coli. J. Bacteriol. 185:6990-6994.[Abstract/Free Full Text]
46
46. Vimr, E. R., and S. M. Steenbergen. 2006. Mobile contingency locus controlling Escherichia coli K1 polysialic acid capsule acetylation. Mol. Microbiol. 60:828-837.[CrossRef][Medline]
47
47. Vimr, E. R., and F. A. Troy. 1985. Identification of an inducible catabolic system for sialic acids (nan) in Escherichia coli. J. Bacteriol. 164:845-853.[Abstract/Free Full Text]
48
48. Vimr, E. R., and F. A. Troy. 1985. Regulation of sialic acid metabolism in Escherichia coli: role of N-acylneuraminiate pyruvate-lyase. J. Bacteriol. 164:854-860.[Abstract/Free Full Text]
49
49. Vimr, E. R., W. Aaronson, and R. P. Silver. 1989. Genetic analysis of chromosomal mutations in the polysialic acid gene cluster of Escherichia coli K1. J. Bacteriol. 172:1106-1117.
50
50. Vimr, E. R., R. D. McCoy, H. F. Vollger, N. C. Wilkison, and F. A. Troy. 1984. Use of prokaryotic-derived probes to identify poly (sialic acid) in neonatal neuronal membranes. Proc. Natl. Acad. Sci. USA 81:1971-1975.[Abstract/Free Full Text]
51
51. Welch, R. A., V. Burland, G. Plunkett III, P. Redford, P. Roesch, D. Rasko, E. L. Buckles, S. R. Liou, A. Boutin, J. Hackett, D. Stroud, G. F. Mayhew, D. J. Rose, S. Zhou, D. C. Schwartz, N. T. Perna, H. L. Mobley, M. S. Donnenberg, and F. R. Blattner. 2002. Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc. Natl. Acad. Sci. USA 99:17020-17024.[Abstract/Free Full Text]
52
52. West, N. P., P. Sansonetti, J. Mounier, R. M. Exley, C. Parsot, S. Guadagnini, M.-C. Prévost, A. Prochnicka-Chalufour, M. Delepierre, M. Tanguy, and C. M. Tang. 2005. Optimization of virulence functions through glucosylation of Shigella LPS. Science 307:1313-1316.[Abstract/Free Full Text]

---

Journal of Bacteriology, September 2007, p. 6447-6456, Vol. 189, No. 17
0021-9193/07/$08.00+0     doi:10.1128/JB.00657-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Scholl, D., Cooley, M., Williams, S. R., Gebhart, D., Martin, D., Bates, A., Mandrell, R. (2009). An Engineered R-Type Pyocin Is a Highly Specific and Sensitive Bactericidal Agent for the Food-Borne Pathogen Escherichia coli O157:H7. Antimicrob. Agents Chemother. 53: 3074-3080 [Abstract] [Full Text]  
• Toth, I., Nougayrede, J.-P., Dobrindt, U., Ledger, T. N., Boury, M., Morabito, S., Fujiwara, T., Sugai, M., Hacker, J., Oswald, E. (2009). Cytolethal Distending Toxin Type I and Type IV Genes Are Framed with Lambdoid Prophage Genes in Extraintestinal Pathogenic Escherichia coli. Infect. Immun. 77: 492-500 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Supplemental material Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by King, M. R. Articles by Vimr, E. R. Search for Related Content PubMed PubMed Citation Articles by King, M. R. Articles by Vimr, E. R.