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) 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 Ballmer, K. Articles by Hächler, H. Search for Related Content PubMed PubMed Citation Articles by Ballmer, K. Articles by Hächler, H.

 Previous Article  |  Next Article 

Journal of Clinical Microbiology, February 2007, p. 370-379, Vol. 45, No. 2
0095-1137/07/$08.00+0     doi:10.1128/JCM.01361-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Fast DNA Serotyping of Escherichia coli by Use of an Oligonucleotide Microarray

Karin Ballmer,¹ Bozena M. Korczak,¹ Peter Kuhnert,¹ Peter Slickers,² Ralf Ehricht,² and Herbert Hächler^1*

Institute of Veterinary Bacteriology, NENT, Vetsuisse-Faculty, University of Berne, CH-3001 Berne, Switzerland,¹ Clondiag Chip Technologies GmbH, D-07743 Jena, Germany²

Received 3 July 2006/ Returned for modification 9 August 2006/ Accepted 6 November 2006

Top Abstract Introduction Materials and Methods Results Discussion References

 
Classical antibody-based serotyping of Escherichia coli is an important method in diagnostic microbiology for epidemiological purposes, as well as for a rough virulence assessment. However, serotyping is so tedious that its use is restricted to a few reference laboratories. To improve this situation we developed and validated a genetic approach for serotyping based on the microarray technology. The genes encoding the O-antigen flippase (wzx) and the O-antigen polymerase (wzy) were selected as target sequences for the O antigen, whereas fliC and related genes, which code for the flagellar monomer, were chosen as representatives for the H phenotype. Starting with a detailed bioinformatic analysis and oligonucleotide design, an ArrayTube-based assay was established: a fast and robust DNA extraction method was coupled with a site-specific, linear multiplex labeling procedure and hybridization analysis of the biotinylated amplicons. The microarray contained oligonucleotide DNA probes, each in duplicate, representing 24 of the epidemiologically most relevant of the over 180 known O antigens (O antigens 4, 6 to 9, 15, 26, 52, 53, 55, 79, 86, 91, 101, 103, 104, 111, 113, 114, 121, 128, 145, 157, and 172) as well as 47 of the 53 different H antigens (H antigens 1 to 12, 14 to 16, 18 to 21, 23 to 34, 37 to 43, 45, 46, 48, 49, 51 to 54, and 56). Evaluation of the microarray with a set of defined strains representing all O and H serotypes covered revealed that it has a high sensitivity and a high specificity. All of the conventionally typed 24 O groups and all of the 47 H serotypes were correctly identified. Moreover, strains which were nonmotile or nontypeable by previous serotyping assays yielded unequivocal results with the novel ArrayTube assay, which proved to be a valuable alternative to classical serotyping, allowing processing of single colonies within a single working day.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Escherichia coli is a commensal bacterium of the intestinal tract of humans and various animal species. Among the many harmless strains, pathogenic isolates exist; and such strains can be harmful, especially for children but even for fully immunocompetent adults as well as animals (9). Disease-causing, virulent E. coli strains can be divided into two major categories, intestinal and extraintestinal pathogens, with the latter comprising mainly uropathogenic variants and those causing neonatal meningitis. The first category comprises various pathotypes, including enterotoxigenic E. coli, enteropathogenic E. coli, enteroinvasive E. coli, enteroaggregative E. coli, and Shiga-toxin producing E. coli (STEC), with the last group including the subgroup of enterohemorrhagic E. coli. In particular, strains in the STEC group have the potential to inflict major damage on different organs, depending on the presence of several virulence factors, and STEC strains are therefore of interest for epidemiological surveillance (3, 6). Some STEC strains have a zoonotic potential, with a reservoir mainly in ruminants (2, 15, 16).

Methods for the diagnosis of E. coli infections include biochemical methods as well as methods that detect specific virulence genes by PCR or hybridization, allowing determination of the pathotype. However, serotyping takes a central place in the differentiation of various pathogenic and nonpathogenic E. coli types, because specific serogroups are consistently associated with certain clinical syndromes (9). Serotyping with the O antigen, a polymerized side chain bound to the core of the lipopolysaccharide, and the H antigen, which is a flagellar protein, yields important epidemiological information; but it is not suitable for use for routine diagnostics because it is too expensive and labor-intensive.

Genetic approaches that can be used to support or replace the classical serotyping method have already been offered. A PCR-restriction fragment length polymorphism assay covering 147 E. coli O serogroups (4) and at least two similar tests covering 48 (11) and 53 (12) different H serogroups have been developed. The availability of these tests has resulted in significant progress in research on the epidemiology of E. coli. However, since O and H antigens must be treated separately and the generation and the interpretation of results are quite challenging, they have not so far been implemented as part of routine procedures. Therefore, for the coverage of all antigens conventional serotyping by agglutination must still be performed; and this requires a major set of costly antisera, time, as well as personnel and is thus restricted to specialized laboratories. Moreover, agglutination results are not always beyond doubt, as the quality of batches of sera may vary; cross-reactions may occur; and a minority of strains are nontypeable, possibly due to masking of capsular antigen, among other reasons. Finally, there are nonmotile strains that lack a flagellar antigen due to corrupted expression, and therefore, only information about the O antigen can be gained. This is where genetic serotyping is clearly superior: it should allow (i) O serotyping of "rough" strains, (ii) H-antigen serotyping of nonmotile strains, and (iii) detection of new H serotypes. Despite the drawbacks of the classical agglutination technique mentioned above, serotyping per se is a very important method which allows early assessment of outbreak situations and possibly even allows tracing of their origins.

Recently, diagnostic microarrays based on the ArrayTube format of Clondiag Chip Technologies GmbH (Jena, Germany) were devised for the detection of E. coli virulence determinants, as well as for the protein-based serotyping of E. coli (1, 10). On the basis of the results of those experiments, we conceived a genetic approach for serotyping, using the ArrayTube technology, since a tool simpler and better than classical serotyping by agglutination is desirable to improve E. coli infection diagnosis and epidemiology. We present an oligonucleotide-based DNA microarray as a powerful tool for the detection of genes critical for the expression of H and O antigens.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Bacterial strains, growth conditions, and genomic DNA extraction. The microarray was evaluated with a set of reference strains that had previously been serotyped by conventional agglutination through various reference centers, including the WHO Reference Center in Copenhagen, Denmark (Table 1). The strains were cultivated on tryptone soy agar (Oxoid, Wesel, Germany) with 5% sheep blood, and genomic DNA was extracted with an E.Z.N.A. DNA extraction kit (peqLab Biotechnologie GmbH, Erlangen, Germany). If necessary, DNA was concentrated to at least 130 ng/µl by using a Speedvac centrifuge. In order to speed up the procedure, a simple single-tube lysis protocol was used for the preparation of crude DNA. In this case a single colony of bacteria was picked from solid medium, resuspended in 100 µl lysis buffer (0.1 M Tris HCl, pH 8.5, 0.05% Tween 20, 240 µg/ml proteinase K), incubated at 60°C for 1 h followed by 15 min at 97°C, and then chilled on ice. Alternatively, a single colony resuspended in 100 µl water was used as the starting material for labeling.

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

TABLE 1. Test strains

Multiplex linear DNA amplification and labeling for hybridization to prepared ArrayTubes. For multiplex linear DNA amplification, a set of 103 primers was used (Table 2). These primers are located on the complementary strand, upstream of the position of the oligonucleotide probe. The labeling of the DNA was accomplished during the linear amplification step by using dUTP linked biotin as a marker, thereby allowing site-specific internal labeling of the corresponding target region. For one reaction the following components were mixed: 1 µl 10x ThermoPol amplification buffer (New England Biolabs, Allschwil, Switzerland), 1 µl deoxynucleoside triphosphate mixture (1 mM each dATP, dCTP, and dGTP; 0.65 mM TTP), 0.35 µl biotin-16-dUTP (Roche Applied Science, Rotkreuz, Switzerland), 1 µl mixture of 103 primer oligonucleotides (final concentration of each oligonucleotide, 0.135 µM; Metabion, Martinsried, Germany), 0.1 µl Therminator DNA polymerase (New England Biolabs), and template (1 µg E. coli DNA or, alternatively, 3 µl crude lysate or bacterial suspension) in a total reaction volume of 10 µl. The linear amplification steps were 5 min of initial denaturation at 96°C, followed by 45 cycles with 20 s of annealing at 62°C, 40 s of elongation at 72°C, and 60 s of denaturation at 96°C. (The proper functioning of this step was monitored and proven by extensive evaluation [see Results section], in that the presence or the absence of signals was always as expected and the signal strengths of the various spots were within a narrow range.)

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

TABLE 2. Reference strains, chip layout, and evaluation results

Hybridization of E. coli ArrayTubes. The ArrayTubes, which were spotted with the probes for serotyping, were produced by Clondiag Chip Technologies. They were placed in a thermomixer (Eppendorf, Hamburg, Germany) and subsequently washed with 500 µl of deionized water for 5 min at 55°C and 550 rpm and with 500 µl hybridization buffer 1 (Clondiag) for 5 min at 30°C and 550 rpm. Liquid was always drawn completely with a soft plastic pipette to avoid scratching of the chip surface at the bottom of the tube. In a separate tube, 10 µl of the labeled DNA was diluted with 90 µl hybridization buffer 1, denatured for 5 min at 95°C, cooled on ice for 2 min, and then added to the ArrayTubes. The hybridization was carried out at 55°C and with shaking at 550 rpm for 1 h.

After hybridization the ArrayTubes were washed once with 500 µl 2x SSC (10x SSC is 1.5 M NaCl plus 0.15 M sodium citrate, pH 7.0) containing 0.2% sodium dodecyl sulfate at 40°C and once with 500 µl 2x SSC at 30°C, followed by the final washing step with 500 µl 0.2x SSC at 30°C, with each step performed for 5 min at 550 rpm. The arrays were blocked with 100 µl 6x SSPE solution (6x SSPE solution is 60 mM sodium phosphate, 1.08 M NaCl, 6 mM EDTA, pH 7.4) containing 0.005% Triton X-100 and 2% (wt/vol) milk powder for 15 min at 30°C and 550 rpm. Peroxidase-streptavidin conjugate (1 µg/µl; Sigma, St. Louis, MO) was diluted 1:3,000 in 6x SSPE-Triton X-100. A total of 100 µl of the dilution was added to each tube, and the mixture was incubated for 15 min at 30°C and 550 rpm. Afterwards, washing was carried out at 550 rpm with 500 µl 2x SSC-0.01% Triton X-100 at 30°C, 500 µl 2x SSC at 20°C, and 500 µl 0.2x SSC at 20°C, with each step performed for 2 min, in duplicate. The visualization of hybridization was achieved by adding 100 µl of peroxidase substrate (True Blue; Kirkegaard & Perry Laboratories, Gaithersburg, MD) to the ArrayTubes, and signals were detected with the ATR01 ArrayTube reader (Clondiag). Signals were recorded at 25°C for 5 min and analyzed by using the IconoClust, version 2.2, software (Clondiag).

Top Abstract Introduction Materials and Methods Results Discussion References

 
Target gene selection and construction of array. The O antigen is determined by a polysugar side chain bound to the core of the lipopolysaccharide in the outer membrane. The components of this side chain are specific triple-sugar repeat units, which are synthesized in the cytoplasm. They are then secreted into the outer membrane and finally polymerized. Many proteins are involved in these complex steps, but two of them have proved to be highly determinative for the appropriate O-antigen serotype, namely, the O-antigen polymerase which is encoded by the gene wzy and which polymerizes the specific trisaccharide repeat units, and the O-antigen flippase, which is encoded by the gene wzx and which is involved in the secretion process (5, 7, 8). The use of wzx or wzy sequences depended on the availability of the respective sequences in public sequence databases (GenBank, EMBL Bank). Since there is no significant similarity between different wzx and wzy sequences either at the DNA level or at the protein level, homology searches were inadequate for the retrieval of additional variants. Therefore, two additional strategies were derived. On the one hand, the annotations of all DNA sequence entries were searched for terms such as "wzx," "wzy," or acronyms related to these specific functions; and on the other hand, different genes that had previously been described to be O-antigen specific, such as wzm, rfbU, rfbE, isla29, sil-inv, sil-1, sil-2, wbdA, wbdH, wbdM, and wbdU, were evaluated as alternatives to wzx and wzy.

The H antigen is determined by the flagellar protein. There are currently 53 known H antigens, but as the designations do not strictly follow sequential numbers, designations up to H56 appear in the literature. Because flagellar genes contained conserved parts at both ends, these parts could be used for BLAST homology searches. Most E. coli strains carry only a single structural flagellar gene, but some strains contain a second flagellin gene at a different locus on the chromosome (13, 14). The most prominent is the fliC locus, where 43 of 53 alleles map. Alternative loci were named fllA, flmA, and flkA. If a strain contains two flagellin genes, only one is expressed at a time, because the fllA, flmA, and flkA loci also comprise a repressor of the fliC gene, which is called fljA.

The 23- to 30-bp oligonucleotide probes were selected from parts of the determinative genes that had proved, by multiple-sequence alignment (with the ClustalX program), to be as diverse as possible and to contain about the same G+C content to ensure similar annealing temperatures. Probes for the H-antigen (fliC) genes were selected from the variable middle region of the open reading frame. Additionally, some probes were taken from the conserved amino- or carboxy-terminal parts of the H antigens as positive controls (Table 2, spots 37 to 41). O-antigen-specific probes were selected in a similar manner by using the whole open reading frame to select discriminative probes. The probe sequences were designated with the gene name, the serotype number, and an index number symbolizing multiple probe positions within the sequence, e.g., fliC-H01_11 (Table 2). Sometimes multiple probes were drawn in order to find the most discriminative one, and they were given the index numbers _12, _13, etc. Putative fliC genes were detected in two GenBank entries by homology searches. Since no documentation was available, they were spotted on the array and designated fl (spots 35 and 36). The fliC-H28-H46 probe covers H28 as well as H46 (spot 76), but individual, specific spots for H28 and H46 are present, too. The array was loaded with a total of 111 probes. Each probe was spotted in duplicate, along with two spots with DNA-free spotting buffer as a negative control (Fig. 1).

View larger version (61K): [in this window] [in a new window]

FIG. 1. Layout of the oligonucleotide probes on the microarray. Details are given in Table 2.

Evaluation of the diagnostic microarray. The performance of the serotyping ArrayTube was tested for each spot with a series of reference strains (Table 1). Three examples of processed arrays are shown in Fig. 2. The system detected the available sequences for O and H antigens with a high degree of reliability (Table 2). All of the 24 O-antigen groups were correctly identified with at least one reference strain. Only a few reference strains failed to hybridize to the corresponding spots, e.g., both strains O9:K9:H12 and O9:H19 (Table 1) at both spots 154 and 181 (Table 2) and O172:H– (Table 1) at spots 141 and 167 (Table 2). Cross-reactions were also rare and were limited to a small number of spots, predominantly to those representing O-antigen-specific genes other than wzx or wzy. All of the 47 H-antigen groups were correctly identified. Twenty-two O serotypes and 46 H serotypes were detected with each of the appropriate test strains. For 21 O serotypes and 43 H serotypes there was a clear positive signal without any cross-reaction, resulting in a sensitivity of 96% and a specificity of 90%. Importantly, for three strains that were reported to be nonmotile (O101:H–, O172:H–, and O172:K–:H–) we found clear results for a fliC gene, which obviously was not being expressed. Moreover, the microarray could even identify two strains that had been correctly serotyped but confused when they were dispatched by the supplier. Eight target sequences (isla29-O145_11, sil-inv-O145_11, rfbU-O157_11, sil-inv-O157_11, sil1-O157_11, sil2-O157_11, fliC-H01_11, and fliC-H46_11) showed strong cross-reactions with about 20% of all strains. Furthermore, seven target sequences (fliC-H01_12, fliC-H01_13, fliC-H06_11, fliC-H12_11, wzx-O104_11, wzy-O104_11, and fljA_11) showed a false-positive signal in only one case. The presence of four target sequences (fl-H-NM_11 [NM indicates nonmotile], fliC-H01_11, wzx-O9_11, and wzy-O9_11) could never be confirmed, and the presence of eight target sequences (fliC-H28_11, wbdA-O9_11, wbdA-O9a_11, wzx-O172_11, wzy-O172_11, wzx-O113_11, wzx-O8_11, and wzy-O8_11) could not be confirmed with every test strain. However, since the O-antigen serotype was mostly represented by more than one target sequence, confirmation of the appropriate serotype was possible in each case. All strains showed signals with the positive control probes on the ArrayTube. In summary, the majority of target sequences were correctly identified.

View larger version (88K): [in this window] [in a new window]

FIG. 2. Microphotographs of microarrays hybridized with genomic DNA of E. coli O157:H7 (A), O53:H3 (B), and O172:K–:H– (C). The numbers correspond to the spot identifiers given in Table 2 and Fig. 1.

Optimization of the protocol. In order to allow fast analysis using the microarray technology, we optimized the procedure and tested whether the step of DNA isolation, which is still time-consuming, could be circumvented. At the same time we wanted to achieve the ability to perform a single-colony analysis. For this purpose we tested whether a simple lysate would be sufficient as the template preparation for the labeling reaction. The results derived from purified DNA and a simple lysate were comparable. To further cut down on the time required for the assay, we tested direct labeling from a bacterial suspension. Figure 3 shows that those results were comparable to those obtained with a lysate or genomic DNA. The detection limit was found to be a genome copy number of 10⁶, which is the amount that must be added to the labeling reaction mixture. Indeed, this corresponds to a normal sized single colony resuspended in 100 µl liquid, of which 3 µl is taken for labeling.

View larger version (87K): [in this window] [in a new window]

FIG. 3. Microphotographs of microarrays of E. coli O157:H43 for which 1 µg of genomic DNA (A), 3 µl of a single colony lysate with a 100-µl total volume (B), and 3 µl of a single colony resuspended in 100 µl water (C) was used for labeling. The numbers correspond to the spot identifiers given in Table 2 and Fig. 1.

Top Abstract Introduction Materials and Methods Results Discussion References

 
We present a diagnostic microarray developed for E. coli serotyping; its validation against "gold standard" agglutination procedures; and its evaluation as a fast, accurate, and easy diagnostic tool. The validation of the microarray, in its present state, showed a high sensitivity of 96% and a specificity of 90% for the O and H antigens covered. The genes selected for detection of the O antigen, wzx and wzy, are the most valuable target sequences for an oligonucleotide DNA microarray-based typing approach, the results of which showed a strong correlation with those of the classical agglutination method. The correlation of the results of the assay with wzx, wzy, and fliC with those of serotyping is such that the microarray technology may well replace serotyping as the gold standard, once relevant sequence data for all serotypes are available. When this goal is achieved, archives of serotype data will smoothly be continued and molecular data will be fully comparable epidemiologically and compatible with classical data. Therefore, long-term-trend studies will not be interrupted upon migration from the classical to the molecular serotyping methods. The alternative genes included in the study proved to be less specific, often showing cross-hybridization signals with other unrelated serotypes, indicating that those genes of the regions selected for probe design are less appropriate than the highly determinative wzx and wzy genes. The same is true for alternative molecular typing methods, such as pulsed-field gel electrophoresis (PFGE). PFGE analysis will never allow prediction of the serotype with sufficient accuracy (data not shown).

Comparison of the microarray method with the traditional serotyping method revealed several advantages of the microarray method. Labeling with specific primers located in proximity to the target oligonucleotides improved the signal-to-background ratio compared to that obtained with random primers (data not shown). Furthermore, it resulted in a higher specificity and shorter labeling times. Together with the short template preparation protocol, it is possible to obtain clear array results within a few hours starting from a single colony. Hence, the chip allows fast and parallel detection of serotypes in a single experiment. Moreover, the microarray is able to detect unexpressed potential H antigens even in the case of a phenotypically nonmotile strain. Therefore, the microarray provides a powerful tool for further characterization and epidemiological tracing of motile and nonmotile strains, a feature that renders it clearly superior to phenotypic serotyping. Furthermore, when the H antigen is expressed by a gene other than fliC, the chip provides additional information by indicating the silent fliC gene, as well as the repressor gene, which then provides unambiguous results. Evaluation revealed that there is room for improvement for future versions of the present chip. For the next generation of the chip, spots showing strong cross-reactions (fliC-H01_11, fliC-H46_11, isla29-O145_11, sil-inv-O145_11, rfbU-O157_11, sil-inv-O157_11, sil1-O157_11, and sil2-O157_11) as well as the spot fl-H-NM, which yielded no result at all, can be eliminated. These modifications will increase the specificity of the microarray without loosing sensitivity. The important serotype O157 is covered by specific spots wzx-O157_11 and rfbE-O157_11. Serotype O145 could be covered with the recently published target sequences for the O-antigen flippase and the O-antigen polymerase (GenBank accession no. AY863412). The few target sequences that yielded false-positive results with only one reference strain, as well as the target sequences whose presence could not be confirmed with all of the appropriate tests strains, should be tested with more reference strains to show whether the cross-reactions encountered were exceptional or whether these target sequences will have to be modified. Such fine-tuning of single probes is, however, beyond the scope of this feasibility study. It will be carried out with advanced future versions of the microarray, and field strains may then also be included in the test series.

Finally, an extended version of the present array may contain not only more targets for additional serotypes but also the most important target sequences for classification of the four most common pathotypes, enteropathogenic, enterotoxigenic, and enteroinvasive E. coli and STEC. Oligonucleotide target sequences for these hallmark virulence genes have already been described by Anjum et al. (M. Anjum et al., personal communication). Such a chip with a combination of sequences would be a fast and powerful tool that could replace common methods of E. coli pathotyping and, in addition, would offer epidemiological information. This would allow, for example, the assessment of the importance of the O157:H7 serotype within the group of highly virulent STEC strains.

In summary, the microarray for serotyping of E. coli has proved to be a helpful tool for epidemiological surveillance of pathogenic E. coli strains and is an easy and fast technique. The capacity of the microarray is limited, at the moment, by the number of published antigen target sequences, but further amendments will render it even more versatile and more powerful.

 
We thank R. Stephan, as well as S. Monecke and M. Anjum, for kindly providing reference strains and Elke Müller and Jana Sachtschal for technical assistance.

This work was supported by a grant from the Department Communicable Diseases, Swiss Federal Office of Public Health, and by the research fund of the Institute of Veterinary Bacteriology, Vetsuisse-Faculty, Berne, Switzerland.

 
* Corresponding author. Mailing address: NENT, Institute for Medical Microbiology, Cantonal Hospital of Lucerne, CH-6000 Lucerne 16, Switzerland. Phone: 0041 (0)41 205 34 56. Fax: 0041 (0)41 205 37 05. E-mail: herbert.haechler{at}ksl.ch.

Published ahead of print on 15 November 2006.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Anjum, M. F., J. D. Tucker, K. A. Sprigings, M. J. Woodward, and R. Ehricht. 2006. Use of miniaturized protein arrays for Escherichia coli O serotyping. Clin. Vaccine Immunol. 13:561-567.[Abstract/Free Full Text]
2
2. Blanco, J., M. Blanco, J. E. Blanco, A. Mora, E. A. Gonzalez, M. I. Bernardez, M. P. Alonso, A. Coira, A. Rodriguez, J. Rey, J. M. Alonso, and M. A. Usera. 2003. Verotoxin-producing Escherichia coli in Spain: prevalence, serotypes, and virulence genes of O157:H7 and non-O157 VTEC in ruminants, raw beef products, and humans. Exp. Biol. Med. (Maywood) 228:345-351.[Abstract/Free Full Text]
3
3. Boerlin, P., S. A. McEwen, F. Boerlin Petzold, J. B. Wilson, R. P. Johnson, and C. L. Gyles. 1999. Associations between virulence factors of Shiga toxin-producing Escherichia coli and disease in humans. J. Clin. Microbiol. 37:497-503.[Abstract/Free Full Text]
4
4. Coimbra, R. S., F. Grimont, P. Lenormand, P. Burguiere, L. Beutin, and P. A. D. Grimont. 2000. Identification of Escherichia coli O-serogroups by restriction of the amplified O-antigen gene cluster (rfb-RFLP). Res. Microbiol. 151:639-654.[Medline]
5
5. DebRoy, C., E. Roberts, J. Kundrat, M. A. Davis, C. E. Briggs, and P. M. Fratamico. 2004. Detection of Escherichia coli serogroups O26 and O113 by PCR amplification of the wzx and wzy genes. Appl. Environ. Microbiol. 70:1830-1832.[Abstract/Free Full Text]
6
6. Ethelberg, S., K. E. Olsen, F. Scheutz, C. Jensen, P. Schiellerup, J. Enberg, A. M. Petersen, B. Olesen, P. Gerner-Smidt, and K. Molbak. 2004. Virulence factors for hemolytic uremic syndrome, Denmark. Emerg. Infect. Dis. 10:842-847.[Medline]
7
7. Feng, L., W. Han, Q. Wang, D. A. Bastin, and L. Wang. 2005. Characterization of Escherichia coli O86 O-antigen gene cluster and identification of O86-specific genes. Vet. Microbiol. 106:241-248.[CrossRef][Medline]
8
8. Fratamico, P. M., C. E. Briggs, D. Needle, C. Y. Chen, and C. DebRoy. 2003. Sequence of the Escherichia coli O121 O-antigen gene cluster and detection of enterohemorrhagic E. coli O121 by PCR amplification of the wzx and wzy genes. J. Clin. Microbiol. 41:3379-3383.[Abstract/Free Full Text]
9
9. Kaper, J. B., J. P. Nataro, and H. L. Mobley. 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2:123-140.[CrossRef][Medline]
10
10. Korczak, B., J. Frey, J. Schrenzel, G. Pluschke, R. Pfister, R. Ehricht, and P. Kuhnert. 2005. Use of diagnostic microarrays for determination of virulence gene patterns of Escherichia coli K1, a major cause of neonatal meningitis. J. Clin. Microbiol. 43:1024-1031.[Abstract/Free Full Text]
11
11. Machado, J., F. Grimont, and P. A. D. Grimont. 2000. Identification of Escherichia coli flagellar types by restriction of the amplified fliC gene. Res. Microbiol. 151:535-546.[Medline]
12
12. Prager, R., U. Strutz, A. Fruth, and H. Tschape. 2003. Subtyping of pathogenic Escherichia coli strains using flagellar (H)-antigens: serotyping versus fliC polymorphisms. Int. J. Med. Microbiol. 292:477-486.[CrossRef][Medline]
13
13. Tominaga, A. 2004. Characterization of six flagellin genes in the H3, H53 and H54 standard strains of Escherichia coli. Genes Genet. Syst. 79:1-8.[CrossRef][Medline]
14
14. Wang, L., D. Rothemund, H. Curd, and P. R. Reeves. 2003. Species-wide variation in the Escherichia coli flagellin (H-antigen) gene. J. Bacteriol. 185:2936-2943.[Abstract/Free Full Text]
15
15. Zweifel, C., J. E. Blanco, M. Blanco, J. Blanco, and R. Stephan. 2004. Serotypes and virulence genes of ovine non-O157 Shiga toxin-producing Escherichia coli in Switzerland. Int. J. Food Microbiol. 95:19-27.[CrossRef][Medline]
16
16. Zweifel, C., S. Schumacher, M. Blanco, J. E. Blanco, T. Tasara, J. Blanco, and R. Stephan. 2005. Phenotypic and genotypic characteristics of non-O157 Shiga toxin-producing Escherichia coli (STEC) from Swiss cattle. Vet. Microbiol. 105:37-45.[CrossRef][Medline]

---

Journal of Clinical Microbiology, February 2007, p. 370-379, Vol. 45, No. 2
0095-1137/07/$08.00+0     doi:10.1128/JCM.01361-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Miller, M. B., Tang, Y.-W. (2009). Basic Concepts of Microarrays and Potential Applications in Clinical Microbiology. Clin. Microbiol. Rev. 22: 611-633 [Abstract] [Full Text]  
• Clark, C. G., Kropinski, A. M., Parolis, H., Grant, C. C. R., Trout-Yakel, K. M., Franklin, K., Ng, L.-K., Paramonov, N. A., Parolis, L. A. S., Rahn, K., Tabor, H. (2009). Escherichia coli O123 O antigen genes and polysaccharide structure are conserved in some Salmonella enterica serogroups. J Med Microbiol 58: 884-894 [Abstract] [Full Text]  
• Anjum, M. F., Mafura, M., Slickers, P., Ballmer, K., Kuhnert, P., Woodward, M. J., Ehricht, R. (2007). Pathotyping Escherichia coli by Using Miniaturized DNA Microarrays. Appl. Environ. Microbiol. 73: 5692-5697 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) 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 Ballmer, K. Articles by Hächler, H. Search for Related Content PubMed PubMed Citation Articles by Ballmer, K. Articles by Hächler, H.