• enterohemorrhagic Escherichia coli;
  • multilocus variable-number tandem-repeat analysis;
  • serogroup O26;
  • serogroup O111


  1. Top of page

Enterohemorrhagic Escherichia coli (EHEC), a food- and waterborne pathogen, causes diarrhea, hemorrhagic colitis, and life-threatening HUS. MLVA is a newly developed and widely accepted genotyping tool. An MLVA system for EHEC O157 involving nine genomic loci has already been established. However, the present study revealed that the above-mentioned MLVA system cannot analyze EHEC O26 and O111 isolates—the second and third most dominant EHEC serogroups in Japan, respectively. Therefore, with several modifications to the O157 system and the use of nine additional loci, we developed an expanded MLVA system applicable to EHEC O26, O111, and O157. Our MLVA system had a relatively high resolution power for each of the three serogroups: Simpson's index of diversity was 0.991 (95% CI = 0.989–0.993), 0.988 (95% CI, 0.986–0.990), and 0.986 (95% CI, 0.979–0.993) for O26, O111, and O157, respectively. This system also detected outbreak-related isolates; the isolates collected during each of the 12 O26 and O111 outbreaks formed unique clusters, and most of the repeat copy numbers among the isolates collected during the same outbreak exhibited no or single-locus variations. These results were comparable to those of cluster analyses based on PFGE profiles. Therefore, our system can complement PFGE analysis—the current golden method. Because EHEC strains of three major serogroups can be rapidly analyzed on a single platform with our expanded MLVA system, this system could be widely used in molecular epidemiological studies of EHEC infections.

List of Abbreviations: 

amplified fragment-length polymorphism


confidence interval


enterohemorrhagic Escherichia coli


hemorrhagic colitis


hemolytic uremic syndrome


locus of enterocyte effacement


multilocus sequence typing


multilocus variable-number tandem-repeat analysis


pulsed-field gel electrophoresis


repetitive element PCR


Shiga toxin-producing E. coli


variable-number tandem-repeat

Enterohemorrhagic Escherichia coli (EHEC), also called STEC, is a food- and waterborne pathogen that causes diarrhea, HC, and life-threatening HUS (1). Shiga toxin is the main virulence factor of EHEC and exerts cytotoxic effects on host cells. Other virulence factors such as the LEE-encoded type III secretion system also contribute to the pathogenicity of EHEC (2).

EHEC isolates have been extensively monitored in Japan since 1996, when many outbreaks of EHEC O157:[H7] infection occurred (3, 4). All the culture-positive cases of EHEC infection, irrespective of the serogroups isolated, are reported to the National Institute of Infectious Diseases. At present, the most dominant serotype is O157:[H7], followed by O26:[H11] and O111:H-. These three serotypes account for more than 95% of the EHEC isolated in Japan (5).

Recent advances in microbial genome sequencing have enabled the establishment of new methods for subtyping bacterial isolates. Among these, MLVA is one of the most widely accepted and useful methods (6). MLVA was successfully used to elucidate the molecular epidemiology of EHEC O157:[H7], and an MLVA system involving nine genomic loci was established for this serogroup (7). However, this system has not yet been applied for EHEC serogroups other than O157. In the present study, we first investigated whether the MLVA system for O157 can be applied to EHEC O26 and O111 and found that it cannot. Therefore, on the basis of the genome sequences of EHEC O26 and O111 (8), we developed an expanded MLVA system that is applicable not only to the O157 serogroup but also to the O26 and O111 serogroups. Furthermore, our study revealed that cluster analysis based on the MLVA profiles is comparable to that based on PFGE profiles in outbreak investigations.


  1. Top of page

Bacterial strains

A total of 641 EHEC isolates (153 O157:H7/-, 355 O26:H11/-, and 133 O111:H- isolates) were examined in the present study. All these isolates were collected by the staff of local public health institutes between 2005 and 2007. Among these, 145 O26 and 39 O111 isolates had been collected during nine and three outbreaks, respectively, and were used to evaluate the efficacy of our new MLVA system in detecting outbreak-related strains. A strain set comprising 469 isolates (153 O157, 219 O26, and 97 O111 isolates, referred to as ‘representative isolates’) was used to evaluate the discriminatory power of the MLVA system. This included isolates from apparently independent sporadic cases and those representing each outbreak (one isolate from one outbreak).


MLVA was carried out as described in our previous study (9). The genome sequences of four EHEC strains (two O157, one O26, and one O111 strain) were searched for tandem repeats in silico (8, 10, 11). Finally, 18 loci, including the nine loci used in the current MLVA system for O157, were used to analyze the isolates in the present study. The primers were designed so that amplification reactions could be carried out in two multiplex mixtures. The primers used in this study are shown in Table 1. The O157-9 reverse primer for O26 and O111 was different from the original primer for O157, because the sequence corresponding to the primer in O26 and O111 differed from that in O157, as described below. One primer of each primer pair was labeled at its 5′ end with 6-FAM, NED, VIC, or PET (Applied BioSystems, Foster City, CA). The PCR mixture contained 0.04 or 0.08 μM of each primer, a DNA template, and 1× multiplex PCR mixture (Qiagen KK, Tokyo, Japan). The PCR conditions were as follows: an initial denaturation step at 95°C for 15 min; 35 cycles of denaturation at 95°C for 20 sec, annealing at 60°C for 90 sec, and extension at 72°C for 60 sec; and the final extension step at 72°C for 10 min.

Table 1.  Primers and characteristics of the loci used in the present study
 Locus Sequence (5′–3′)Concentration (μM)DyeLocation in the genome sequencesRepeat size (bp)Offset
  1. †, same location as Ref. 7

  2. n, not found.

  3. §, 135 for O26.

RGTGTCAGGTGAGCTACAGCCCGCTTACGCTC0.08 1305030269672734907063558035  
RGCGCTGAAAAGACATTCTCTGTTTGGTTTACAC0.04 1450853141394115208161605911  
RAGGCATTAATAGCAGATGTTC0.08 5508780523004153318295361832  
RGTCAACGCTGACCTCTTCCGGT0.04 4611473435019444327984500025  
RCCTTGTGCATTGAGTTCTGTACATAG0.04 3449828316147233165183386352  
RGCAAGTCGAGTGRGCTCTGCGGGG0.04 5364102372134245899464658929  
RCGCGGCTGCCGGAGTATC0.04 4797171466266047815404850302  
RACGCTGGTCCGGGAGATTAT0.04 252797250261252432252433  
RCGGGCAGGGAATAAGGCCACCTGTTAAGC0.04 2938574275446228622102932441  

The PCR products were diluted and separated with an ABI 3130 genetic analyzer, using GeneScan LIZ 600 (Applied Biosystems) as the size standard. The size of each PCR product was converted to a repeat copy number by using the Gene Mapper software (Applied Biosystems). The data were incorporated into the BioNumerics software (Applied Maths, Sint-Martens-Latem, Belgium) and analyzed as previously described (7). Repeat copy number for the null allele, namely, when no PCR product was obtained, was designated as −2. Simpson's index of diversity (D) and 95% CI were calculated according to formulas described in a previous report (12). The number of alleles indicates the number of variations detected in the repeat copy numbers at a locus and is hereafter referred to as the ‘allele number’.


PFGE was carried out according to the PulseNet protocol developed at the Centers for Disease Control and Prevention by using Salmonella enterica serovar Braenderup H9812 strain as a standard for normalization (4, 13). DNA was digested with XbaI and separated using a CHEF DR III apparatus (Bio-Rad Laboratories, Hercules, CA) under the following conditions: switching time from 2.2 to 54.2 sec at 6 V/cm for 21 hr at 14°C. After the gels were stained with ethidium bromide, they were imaged using Gel Doc EQ and Quality One System (Bio-Rad Laboratories). Cluster analysis was carried out using the BioNumerics software as previously described (14).


  1. Top of page

Evaluation of the 18 loci selected for the MLVA of EHEC O157, O26, and O111 isolates

Our initial analysis of the genome sequences of the O26 and O111 strains (8) revealed that among the nine loci that are routinely used for analyzing O157 (O157-3, O157-9, O157-10, O157-17, O157-19, O157-25, O157-34, O157-36, and O157-37), five and four loci are not present in the O26 and O111 strains, respectively (Table 1). This finding indicates that additional genomic loci are required for MLVA of the O26 and O111 strains. Therefore, we selected nine additional loci on the basis of the results obtained after analyzing the genome sequences of the O26 and O111 strains and comparing their genome sequence to that of O157; moreover, we developed a system by which these 18 loci can be simultaneously analyzed, as described previously (Table 1). By using this system and the 469 representative EHEC isolates (153 O157, 219 O26, and 97 O111 isolates), we examined whether these 18 loci can be used for MLVA of the O26 and O111 isolates, as well as the O157 isolates (Fig. 1).


Figure 1. Allele numbers and diversity values for the 18 loci evaluated in the present study. Allele numbers per locus are indicated by bars. Simpson's diversity index (D) values are indicated by lines. (a) Values for each serogroup are shown. □, O157 isolates, (n= 153); ▪, O26, (n= 219); inline image, O111, (n= 97); ▵, O157; □, O26; and ○, O111. Allele number of O157-10 for the O157 isolates (45) is out of range. Asterisks indicate the loci with only a null allele in the indicated serogroups. (b) Values obtained by integrating the data of all the isolates belonging to the three serogroups (n= 469) are shown.

Download figure to PowerPoint

Of the nine loci that are currently used for analyzing the O157 isolates, four (O157-3, O157-10, O157-17, and O157-36) were not detected in any of the O26 or O111 isolates. Although the O157-9 locus was found in the genome sequences of O26 and O111, the downstream sequences in the O26 and O111 strains were different from that in O157, and the position of the sequence corresponding to the original O157-9 reverse primer in O26 and O111 was different from that in the O157 genome (Table 1). Therefore, for amplifying the O157-9 locus of the O26 and O111 serogroups, we designed a new reverse primer to equate the size of the offset sequence from the O26/O111 isolates with that from O157. By using this new reverse primer, we found that the O157-9 locus of the O26 and O111 isolates exhibited high allele numbers (11 and 12, respectively) and high D values (0.81 and 0.87, respectively) (Fig. 1a).

Two loci (O157-19 and O157-25) were also present in the genome sequences of O26 and O111, but showed no repeat copy number variation between the O26 and O111 isolates.

There were some problems associated with the O157-34 locus. Re-inspection of the sequence of the O157-34 locus revealed that O157 contained two repeats in this locus in addition to those described in a previous study (15) (Fig. 2). Furthermore, although the sequenced O26 and O111 strains contain one and three repeats, yielding PCR products of 153 bp and 195 bp, respectively, a sequence variation, including a 6-bp deletion, was found in the O157-34 locus-flanking region of the O26 genome sequence. Therefore, we set the offset size for O157 and O111 at 141 bp and that for O26 at 135 bp.


Figure 2. Sequence of the O157-34 locus of EHEC O157 Sakai. Sequence alignment of the 11 repeat units (18-bp long) together with the 39-bp upstream and 102-bp downstream sequences is shown. The repeat region is shown in bold, and the dots indicate that the nucleotides are the same as those in the first repeat sequence.

Download figure to PowerPoint

To summarize, of the nine loci that are currently used for analyzing the O157 isolates, eight were not suitable for analyzing the O26 and O111 isolates when the original primers were used (Fig. 1a). Only the O157-37 locus could be used for the O26 and O111 isolates, which exhibited D values of 0.25 and 0.93, respectively. When a new O157-9 reverse primer was used for the O26/O111 isolates, the O157-9 locus in both the O26 and O111 isolates exhibited high D values.

Among the nine additional genomic loci that we used in the present study, three were previously used for O157 analysis (EH157-12, EHC-1, and EHC-2, designated as O157-13, O157-11, and O157-2, respectively, in the previous report (15)) and six were newly developed (EH26-7, EH111-8, EH111-11, EH111-14, EHC-5, and EHC-6). Of these nine loci, EHC-1 was very useful for genotyping all the serogroups: the D values were 0.83, 0.91, and 0.85 for the O26, O111, and O157 isolates, respectively. EHC-2 was also useful for all the serogroups, especially for the O26 isolates that exhibited an extremely high D value (0.92). EH157-12 was suitable mainly for O157 and exhibited moderate D values for the O26 and O111 isolates, despite the low allele numbers in these two serogroups. EHC-5 and EHC-6 also yielded high or moderate D values for all the serogroups. Although these five loci are not included in the current MLVA system for O157, they can be used for analyzing the O157 isolates, as well as the O26 and O111 isolates.

Of the remaining four loci, one (EH26-7) was present only in the O26 isolate. However, this locus exhibited a D value of 0.43 with an allele number of seven and thus significantly contributed to the genotyping of the O26 isolates. As such, three loci (EH111-8, EH111-11, and EH111-14) were specifically present in O111 but were of a certain level of usefulness for this serogroup because they exhibited moderate D values (0.21, 0.24, and 0.17, respectively). Our results indicate that these four loci can be used for genotyping the O26 and O111 isolates.

Figure 1b shows the results of our evaluation of the 18 loci for the isolates belonging to all the three serogroups together. The allele numbers ranged from 3 to 45, and the D values ranged from 0.34 to 0.92. In this analysis, six loci (EH157-12, O157-34, O157-37, O157-9, EHC-1, and EHC-2) exhibited higher D values than did the other loci. The overall D values were 0.991 (95% CI = 0.989–0.993), 0.988 (95% CI = 0.986–0.990), and 0.986 (95% CI = 0.979–0.993) for the O26, O111, and O157 isolates, respectively. These values indicate that our system is useful for genotyping EHEC isolates of not only the O157, but also the O26 and O111 serogroups.

Applicability to an outbreak investigation

As the results mentioned above indicated that our expanded MLVA system was useful for genotyping the O26 and O111 isolates, we next carried out cluster analyses of the O26 and O111 isolates by using the new MLVA system. In this analysis, we included the isolates collected during nine O26 outbreaks and three O111 outbreaks, as well as assessing the applicability of our system for detecting outbreak-related strains in these two serogroups.

As shown in Figure 3, the isolates collected during each of the 12 outbreaks formed unique clusters. Isolates from three outbreaks (26OB5, 26OB6, and 111OB3 outbreaks) did not exhibit any repeat copy number variations for all 18 loci. With regard to the other nine outbreaks, variations were observed for some loci in a few isolates obtained during the same outbreak (Table 2). However, in eight of the nine outbreaks, variations were mainly found in the O157-37 and/or EHC-6 loci, both of which are located in large plasmids, such as pO157, suggesting that entire plasmids may have been lost or parts of these plasmids may have been deleted in some strains during the outbreaks or after strain isolation. These results indicate that the MLVA system can be useful for detecting outbreaks of the EHEC strains belonging to the O26 and O111 serogroups.


Figure 3. Cluster analysis of (a) EHEC O26 and (b) O111 isolates based on the MLVA profiles. In addition to the representative isolates, outbreak-related isolates collected during nine O26 and three O111 outbreaks were included in this analysis. The numbers at the top indicate similarity (%). Outbreak-related isolates collected during the 12 outbreaks are indicated by bars with the outbreak number.

Download figure to PowerPoint

Table 2.  Variations of repeat-numbers in outbreak-related EHEC O26 and O111 isolates
  1. –, indicates the same value as in the column above.


The O26 and O111 isolates were also subjected to cluster analyses based on PFGE profiles (Fig. 4). Each of the outbreaks formed a unique cluster, as shown in Figure 3. The relative positions of the PFGE-based clusters, however, did not always match those of the MLVA-based clusters. For example, the positions of the clusters of 26OB3 and 26OB7 in the PFGE analysis were closely matched; however, their positions were completely different in the MLVA. Moreover, the subtypes within a cluster defined in each method did not completely match. For example, a single subtype was detected in the 26OB2 and 26OB8 clusters in the PFGE analysis; however, two and three subtypes were detected in the 26OB2 and 26OB8 clusters, respectively, in the MLVA. Alternatively, a single subtype was detected in the 26OB5 and 26OB6 clusters in the MLVA, whereas five and three subtypes were detected in the 26OB5 and 26OB6 clusters, respectively, in the PFGE analysis. Nevertheless, most of these results were consistent with each other, as in the case of O111OB3, where all the isolates exhibited 100% similarity in both the analyses.


Figure 4. Cluster analysis of (a) EHEC O26 and (b) O111 isolates based on PFGE profiles. The same isolates as mentioned in the legend to Figure 3 were included in this analysis. The numbers at the bottom indicate similarity (%). Outbreak-related isolates collected during the 12 outbreaks are indicated by bars with the outbreak number.

Download figure to PowerPoint


  1. Top of page

Genotyping is a powerful and useful tool for epidemiological investigation; for example, during outbreaks of infectious diseases. MLVA is a newly developed genotyping method for bacterial infectious diseases and is based on differences between the isolates with regard to the repeat copy numbers in certain genomic loci. Dozens of bacterial species, including EHEC O157, have been studied using this method (6, 7). Owing to its simplicity and discriminating power, it is considered one of the methods of the next generation to PFGE, which is currently the golden method of genotyping. MLVA can be accomplished through PCR and electrophoresis. The results are converted to digitalized repeat copy numbers, which can be clearly evaluated for each isolate. MLVA is also a rapid method—the results can be obtained within several hours after isolation (16). MLVA, however, requires high-quality electrophoresis facilities, such as an automatic sequencer, which has a high cost of implementation. Further, for the start-up process, genome sequences of target bacterial agents are required, and the efficacy of an MLVA system can be affected by information on the genome sequences analyzed. That is, increasing availability of the genome sequences of a given bacterial species increases the efficiency of MLVA. In the present study, we developed and evaluated the efficiency of an expanded MLVA system that was designed for analyzing the EHEC O26 and O111 isolates as well as the EHEC O157 isolates. The three serogroups account for more than 95% of the EHEC isolated in Japan (5).

The results of evaluation of the MLVA system that is now being routinely used for analyzing EHEC O157 isolates (7) indicate that it is not applicable to the EHEC O26 and O111 isolates. Most loci were not amplified by PCR, even if any amplification occurred, the repeat copy numbers exhibited less variation among the EHEC O26 and O111 isolates (Fig. 1). Comparison and re-inspection of the genome sequences also resulted in correction of interpretation of the O157-34 locus (Fig. 2). By modifying the O157-9 primer and including nine additional loci, six of which were newly developed in the present study, we finally developed an improved MLVA system that can be used for genotyping EHEC O157, O26, and O111. All the loci adopted in this study exhibited D values of more than 0.3 when applied to all the isolates, and could aid the differentiation of all the three serogroups (Fig. 1b, and data not shown). The D values of EHEC O26 and O111 were comparable to the D value of EHEC O157 that was already proven to be useful in epidemiological analyses (14); the findings of this study suggest a sufficient discriminating power of the MLVA system.

In the present study, the new MLVA system was also useful for detecting outbreak-related isolates, and this is one of the most prioritized objectives of genotyping (Fig. 3; Table 2). Most of the outbreak-related isolates did not exhibit any, or exhibited only single-locus, variations within each outbreak (Table 2). The cluster analysis based on the MLVA profiles revealed that each outbreak could form a unique cluster. This was also true for the cluster analysis based on the PFGE profiles. Further, consistent results were obtained by both these methods (Figs 3, 4). However, the relationships between the clusters observed in one method differed from those observed in the other method because of the differences in the two methods with regard to the targets; MLVA discriminates isolates by repeat copy numbers of specific loci, whereas PFGE differentiates them by restriction fragment length polymorphisms of the entire DNA. Moreover, either PFGE or MLVA can be superior to the other method for discriminating isolates in some outbreaks. These results indicate that MLVA can complement PFGE analysis. Considering that the procedure of MLVA is simpler and more rapid than that of PFGE, MLVA can be applied for the first screening of isolates in outbreak investigations before the results can be confirmed by PFGE.

PFGE analysis is currently the golden method for subtyping bacterial pathogens. (13). Other researchers reported that subtyping methods, such as AFLP, rep-PCR and MLST, could be useful for analyzing EHEC O157, but PFGE was the best method to discriminate isolates, for example, in outbreak investigations (17, 18). In this study, the results of MLVA were similar to those of PFGE analysis in outbreak investigations; this suggests that the discriminating power of MLVA is greater than that of the above-mentioned methods, although it might be necessary to evaluate the discriminating power of them for EHEC non-O157 strains, as described below. Furthermore, other methods are more time-consuming than MLVA. The results of the other methods, except MLST, are deduced from anonymous banding patterns, which can lead to ambiguous typing, whereas the results of MLVA are deduced from known loci and can be controlled by direct sequencing of the amplified products (19).

Recently, infection with EHEC serogroups other than O157 has raised concerns not only in Japan but also in other countries: EHEC O26:[H11], O103:H2, O111:[H8], and O145:[H28] are frequently associated with HC and HUS (20). Although PFGE is the first line of choice for subtyping, most of the methods mentioned above have not yet been evaluated for analyzing EHEC non-O157 strains. A molecular subtyping method might be specific for a serotype or serogroup. Indeed, in the present study, the current MLVA system for O157 was proven to be specific for O157. Modifications in this study enabled it to be applied for the analysis of, at least, EHEC O26 and O111. Other methods, therefore, might also need to be evaluated and modified so they can be applied for the analysis of EHEC non-O157 strains.

In conclusion, by using the MLVA system developed in this study, the EHEC strains of three major serogroups, such as O157, O26 and O111, can be analyzed on a single platform. Therefore, this system could be widely used for molecular epidemiological studies of EHEC infections.


  1. Top of page

We thank the staff of all the municipal and prefectural public health institutes for providing the EHEC isolates. We thank Ms Nobuko Takai, Ms Tamayo Kudo, and Ms Lee Jiyoung for their technical assistance. This work was partly supported by grants-in-aid from the Ministry of Health, Labour and Welfare of Japan (H21-Shokuhin-Ippan-005, H21-Shokuhin-Ippan-013, H20-Shinko-Ippan-013, and H20-Shinko-Ippan-015).


  1. Top of page
  • 1
    Tarr P.I., Gordon C.A., Chandler W.L. (2005) Shiga-toxin-producing Escherichia coli and haemolytic uraemic syndrome. Lancet 365: 107386.
  • 2
    Law D. (2000) Virulence factors of Escherichia coli O157 and other Shiga toxin-producing E. coli. J Appl Microbiol 88: 72945.
  • 3
    Watanabe H., Wada A., Inagaki Y., Itoh K., Tamura K. (1996) Outbreaks of enterohaemorrhagic Escherichia coli O157:H7 infection by two different genotype strains in Japan 1996. Lancet 348: 8312.
  • 4
    Izumiya H., Terajima J., Wada A., Inagaki Y., Itoh K.I., Tamura K., Watanabe H. (1997) Molecular typing of enterohemorrhagic Escherichia coli O157:H7 isolates in Japan by using pulsed-field gel electrophoresis. J Clin Microbiol 35: 167580.
  • 5
    National Institute of Infectious Diseases (2009) Enterohemorrhagic Escherichia coli infection as of April 2009. IASR 30: 11920.
  • 6
    Van Belkum A. (2007) Tracing isolates of bacterial species by multilocus variable number of tandem repeat analysis (MLVA). FEMS Immunol Med Microbiol 49: 227.
  • 7
    Hyytia-Trees E., Smole S.C., Fields P.A., Swaminathan B., Ribot E. (2006) Second generation subtyping: A proposed PulseNet protocol for multiple-locus variable-number tandem repeat analysis of Shiga toxin-producing Escherichia coli O157 (STEC O157). Foodborne Pathog Dis 3: 11831.
  • 8
    Ogura Y., Ooka T., Iguchi A., Toh H., Asadulghani M., Oshima K., Kodama T., Abe H., Nakayama K., Kurokawa K., Tobe T., Hattori M., Hayashi T. (2009) Comparative genomics reveal the mechanism of the parallel evolution of O157 and non-O157 enterohemorrhagic Escherichia coli. Proc Natl Acad Sci USA 106: 1793944.
  • 9
    Izumiya H., Tada Y., Ito K., Morita-Ishihara T., Ohnishi M., Terajima J., Watanabe H. (2009) Characterization of Shigella sonnei isolates from travel-associated cases in Japan. J Med Microbiol 58: 148691.
  • 10
    Perna N.T., Plunkett G. 3rd, Burland V., Mau B., Glasner J.D., Rose D.J., Mayhew G.F., Evans P.S., Gregor J., Kirkpatrick H.A., Posfai G., Hackett J., Klink S., Boutin A., Shao Y., Miller L., Grotbeck E.J., Davis N.W., Lim A., Dimalanta E.T., Potamousis K.D., Apodaca J., Anantharaman T.S., Lin J., Yen G., Schwartz D.C., Welch R.A., Blattner F.R. (2001) Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409: 52933.
  • 11
    Hayashi T., Makino K., Ohnishi M., Kurokawa K., Ishii K., Yokoyama K., Han C.G., Ohtsubo E., Nakayama K., Murata T., Tanaka M., Tobe T., Iida T., Takami H., Honda T., Sasakawa C., Ogasawara N., Yasunaga T., Kuhara S., Shiba T., Hattori M., Shinagawa H. (2001) Complete genome sequence of enterohemorrhagic Escherichia coli O157:H7 and genomic comparison with a laboratory strain K-12. DNA Res 8: 1122.
  • 12
    Grundmann H., Hori S., Tanner G. (2001) Determining confidence intervals when measuring genetic diversity and the discriminatory abilities of typing methods for microorganisms. J Clin Microbiol 39: 41902.
  • 13
    Ribot E.M., Fair M.A., Gautom R., Cameron D.N., Hunter S.B., Swaminathan B., Barrett T.J. (2006) Standardization of pulsed-field gel electrophoresis protocols for the subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathog Dis 3: 5967.
  • 14
    Pei Y., Terajima J., Saito Y., Suzuki R., Takai N., Izumiya H., Morita-Ishihara T., Ohnishi M., Miura M., Iyoda S., Mitobe J., Wang B., Watanabe H. (2008) Molecular characterization of enterohemorrhagic Escherichia coli O157:H7 isolates dispersed across Japan by pulsed-field gel electrophoresis and multiple-locus variable-number tandem repeat analysis. Jpn J Infect Dis 61: 5864.
  • 15
    Keys C., Kemper S., Keim P. (2005) Highly diverse variable number tandem repeat loci in the E. coli O157:H7 and O55:H7 genomes for high-resolution molecular typing. J Appl Microbiol 98: 92840.
  • 16
    Olsen J.S., Aarskaug T., Skogan G., Fykse E.M., Ellingsen A.B., Blatny J.M. (2009) Evaluation of a highly discriminating multiplex multi-locus variable-number of tandem-repeats (MLVA) analysis for Vibrio cholerae. J Microbiol Methods 78: 27185.
  • 17
    Hahm B.K., Maldonado Y., Schreiber E., Bhunia A.K., Nakatsu C.H. (2003) Subtyping of foodborne and environmental isolates of Escherichia coli by multiplex-PCR, rep-PCR, PFGE, ribotyping and AFLP. J Microbiol Methods 53: 38799.
  • 18
    Foley S.L., Simjee S., Meng J., White D.G., Mcdermott P.F., Zhao S. (2004) Evaluation of molecular typing methods for Escherichia coli O157:H7 isolates from cattle, food, and humans. J Food Prot 67: 6517.
  • 19
    Lindstedt B.A. (2005) Multiple-locus variable number tandem repeats analysis for genetic fingerprinting of pathogenic bacteria. Electrophoresis 26: 256782.
  • 20
    Beutin L. (2006) Emerging enterohaemorrhagic Escherichia coli, causes and effects of the rise of a human pathogen. J Vet Med B Infect Dis Vet Public Health 53: 299305.