Skip to main content
NCBI home page
Journal of Clinical Microbiology logoLink to Journal of Clinical Microbiology
. 2014 Aug;52(8):2757–2763. doi: 10.1128/JCM.00420-14

Characterization of Enterohemorrhagic Escherichia coli O111 and O157 Strains Isolated from Outbreak Patients in Japan

Masanori Watahiki a,✉, Junko Isobe a, Keiko Kimata a, Tomoko Shima a, Jun-ichi Kanatani a, Miwako Shimizu a, Akihiro Nagata b,*, Keiko Kawakami c, Mikiko Yamada d, Hidemasa Izumiya e, Sunao Iyoda e, Tomoko Morita-Ishihara e, Jiro Mitobe e, Jun Terajima e,*, Makoto Ohnishi e, Tetsutaro Sata a
Editor: D J Diekema
PMCID: PMC4136153  PMID: 24829231

Abstract

In April and May 2011, there was a serious food-poisoning outbreak in Japan caused by enterohemorrhagic Escherichia coli (EHEC) strains O111:H8 and O157:H7 from raw beef dishes at branches of a barbecue restaurant. This outbreak involved 181 infected patients, including 34 hemolytic-uremic syndrome (HUS) cases (19%). Among the 34 HUS patients, 21 developed acute encephalopathy (AE) and 5 died. Patient stool specimens yielded E. coli O111 and O157 strains. We also detected both EHEC O111 stx2 and stx-negative E. coli O111 strains in a stock of meat block from the restaurant. Pulsed-field gel electrophoresis (PFGE) and multilocus variable-number tandem-repeat analysis (MLVA) showed that the stx-negative E. coli O111 isolates were closely related to EHEC O111 stx2 isolates. Although the EHEC O157 strains had diverse stx gene profiles (stx1, stx2, and stx1 stx2), the PFGE and MLVA analyses indicated that these isolates originated from a single clone. Deletion of the Stx2-converting prophage from the EHEC O111 stx2 isolates was frequently observed during in vitro growth, suggesting that strain conversion from an EHEC O111 stx2 to an stx-negative strain may have occurred during infection.

INTRODUCTION

Enterohemorrhagic Escherichia coli (EHEC) strains cause a variety of human illnesses, such as uncomplicated diarrhea, hemorrhagic colitis, hemolytic-uremic syndrome (HUS), and related acute encephalopathy (1). Shiga toxin 1 (Stx1) and Stx2 are EHEC virulence factors that cause endothelial cell damage with consecutive systemic thrombotic microangiopathy, resulting in hemorrhagic colitis and subsequent renal failure and involvement of other organs (1–4).

EHEC infection is a category III notifiable infectious disease in Japan, according to the Law Concerning the Prevention of Infectious Diseases and Medical Care for Patients of Infections (Infectious Diseases Control Law). All EHEC cases must be reported by the physician who made the diagnosis. Prefectural and municipal public health institutes (PHIs) conduct EHEC isolation, serotyping, and verotoxin (VT) typing and report their results to the Infectious Disease Surveillance Center (IDSC) of the National Institute of Infectious Diseases (NIID), Japan. Approximately 4,000 cases are reported annually. O157 is the most common EHEC serogroup in gastrointestinal tract infections, accounting for 60 to 70% of the reported EHEC infections. Among non-O157 EHEC serogroups, O26 is the second most common serogroup, accounting for 20 to 25% of the EHEC cases, followed by serogroups O111, O121, and O103.

Hemolytic-uremic syndrome (HUS) is an illness characterized by acute kidney injury, thrombocytopenia, and microangiopathic hemolytic anemia. Approximately 100 HUS cases associated with EHEC infections are reported annually in Japan, corresponding to 3 to 4% of the symptomatic EHEC infections. EHEC O157 is the most prevalent EHEC serogroup causing HUS, accounting for approximately 90% of the HUS cases identified among EHEC isolates. However, a wide variety of EHEC non-O157 serogroups might also cause HUS.

The EHEC O111 serogroup is the etiological agent of approximately 4% of the EHEC cases in Japan (5). During 2006 to 2010, there were 83 EHEC outbreaks in Japan, in which 10 or more EHEC-positive cases were reported. Six of these outbreaks were caused by EHEC O111 strains (6–10): three by EHEC O111 stx1 strains, and the other three by EHEC O111 stx1 stx2 strains. EHEC O111 stx2 isolates are very rare in Japan, consistent with data from other countries (11–13). Although some serious outbreaks due to EHEC O111 isolates have been reported in Western countries (14–16), EHEC O111 outbreaks with serious complications, like HUS, have been rare in Japan.

The important EHEC virulence factors are Stx1 and Stx2 and the locus of enterocyte effacement (LEE) element that is responsible for intimate adhesion to host intestinal cells. The stx1 and stx2 genes are encoded in different but similar prophage genomes. The loss and gain of these prophages must affect EHEC virulence. Since Stx-converting phages have been reported to be lost during in vitro culture (17, 18), they may also be lost under in vivo conditions, although no conclusive in vivo data are currently available.

There was a large EHEC outbreak in Japan in April and May 2011. A case-control study showed that raw beef dishes consumed at a chain of barbecue restaurants were the vehicles for these infections (Y. Yahata, T. Misaki, Y. Ishida, M. Nagira, M. Watahiki, J. Isobe, J. Terajima, S. Iyoda, J. Mitobe, M. Ohnishi, T. Sata, K. Taniguchi, Y. Tada, N. Okabe, and E. coli Outbreak Investigation Team, unpublished data). A total of 941 individuals ate at the restaurant chain from 19 April to 4 May 2011, and 181 presented as outbreak-related cases, including 34 HUS cases. Only 55 of the 181 infections were confirmed by laboratory isolation to be from the EHEC O111 and/or O157 serogroups. Of the 34 HUS cases, 21 developed acute encephalopathy and 5 died. Here, we report studies characterizing the E. coli O111 and O157 strains isolated from the cases in this outbreak.

MATERIALS AND METHODS

Case definition.

The case definition for the 181 patients in this outbreak was a person who developed illness >10 h after eating at one of the restaurants in the implicated restaurant chain in April 2011, and who presented with at least one of the following symptoms: (i) bloody stools, (ii) more than two gastrointestinal symptoms, such as diarrhea, nausea or vomiting, abdominal pain, and tenesmus, (iii) one of the gastrointestinal symptoms in addition to at least two additional symptoms, such as fever (temperature of >37.5°C), malaise, or headache, or (iv) a stool culture positive for E. coli O111, EHEC O111, or EHEC O157.

Strains.

A total of 104 EHEC O111, EHEC O157, and stx-negative E. coli O111 isolates were collected from three public health centers in Toyama Prefecture, Japan, and sent to the Toyama Institute of Health (TIH). EHEC strains were also obtained from public health institutes in Fukui Prefecture, Ishikawa Prefecture, Kanazawa City, and Yokohama City, where related cases were detected. EHEC O111 strain 11128 (19) and EHEC O157 strain Sakai (20), used as reference strains, were obtained from the National Institute of Infectious Diseases (Tokyo, Japan) and the Research Institute for Microbial Diseases, Osaka University (Osaka, Japan), respectively.

Isolation of EHEC and serotyping.

Specimens were collected from seven public health and medical laboratories (listed in Acknowledgments). Stool specimens (n = 188) from patients, samples from leftover food from the implicated restaurants (n = 20), and smear samples from the implicated restaurant kitchens (n = 14) were collected and sent to the TIH for analysis. For EHEC isolation, the specimens were analyzed by enrichment culture, immunomagnetic separation (IMS) (21) with O111 or O157 lipopolysaccharide (LPS) antibodies, and acid treatment before plating on selective agar medium, as described below (22). The enriched samples were spread on cefixime-tellurite (CT)-MacConkey sorbitol agar, CT-MacConkey sorbose agar, CHROMagar O157 TAM, CT-CHROMagar O157 TAM, or CHROMagar Shiga toxigenic E. coli (STEC). After the plates were incubated overnight at 35°C, the isolated colonies were transferred to Trypticase soy agar (TSA) plates (Becton, Dickinson and Company), and these plates were incubated at 35°C overnight. The colonies (>4) isolated from the TSA plates were tested by PCR for the presence of the stx2 gene and serotyped using anti-E. coli O111 and O157 antisera (Denka Seiken Co., Ltd., Tokyo, Japan). Determination of the flagellar antigen type of the EHEC O111 outbreak strain was carried out by fliC typing with PCR-restriction fragment length polymorphism (23). The PCR amplicons were digested with HhaI (Nippon Gene Co., Ltd., Toyama, Japan) and separated by 2% agarose gel electrophoresis.

Pulsed-field gel electrophoresis.

Pulsed-field gel electrophoresis (PFGE) was performed using the PulseNet protocol (24). Genomic DNA in agarose plugs was digested overnight with 30 U XbaI (Nippon Gene Co. Ltd., Toyama, Japan) at 37°C. Electrophoresis of the XbaI-treated plugs was performed with the CHEF Mapper system (Bio-Rad Laboratories) using pulsed-field certified agarose (Bio-Rad Laboratories) with 0.5× Tris-borate-EDTA (TBE) running buffer. The electrophoretic conditions were as follows: 6 V/cm for 19 h, pulse time ranging from 2.2 to 54.2 s, and 0.5× TBE buffer at 12°C. Salmonella enterica subsp. enterica serovar Braenderup genomic DNA in an agarose plug was also digested overnight with 30 U XbaI and used as a molecular size marker. After electrophoresis, the gels were stained with ethidium bromide (final concentration, 50 ng/ml), destained by washing with distilled water, and photographed with ChemiDoc XRS (Bio-Rad Laboratories).

Multilocus variable-number tandem-repeat analysis.

Multilocus variable-number tandem-repeat analysis (MLVA) was carried out as described in previous reports (25, 26). The PCR products labeled at the 5′ termini of the target loci for MLVA were separated using an ABI 3130xl Genetic Analyzer (Applied Biosystems). The repeat copy number for a null allele (i.e., when no PCR product was obtained) was designated −2.

Detection of virulence-related genes.

Test strains were selected from 10, nine, five, two, and two representative isolates of the EHEC O111:H8 stx2, stx-negative E. coli O111, EHEC O157 stx1 stx2, EHEC O157 stx1, and EHEC O157 stx2 isolates, respectively. The E. coli strains were resuspended in 200 μl of 5% (wt/vol) Chelex 100 resin (27) and heated for 10 min at 100°C. After centrifugation, the supernatants were quantified by using a NanoDrop ND-1000 (Thermo Fisher Scientific). The DNA preparations were diluted to a final concentration of 10 ng/μl and used as a template for PCR. The PCR primers used to detect stx genes were commercially purchased EVC-1 and EVC-2 to simultaneously detect common stx genes, EVT-1 and EVT-2 to detect stx1, and EVS-1 and EVS-2 to detect stx2 (TaKaRa Bio, Inc.). Other virulence-related genes were analyzed by a multiplex PCR-based protocol (28). The target genes were stx1, stx2, eae, CVD432, aggR, invE, elt (labile toxin [LT] gene), esth (saitohin [STh] gene), estp (sulfotransferase [STp] gene), bfp, EAF, and astA. The Stx-encoding genes, stx1 and stx2, of the EHEC isolates in this outbreak were subtyped using PCR (29).

To detect the norV gene (30), a primer pair was designed from the EHEC O111 genome norV sequence (GenBank accession no. AP010960): norV-337f (5′-CAT ACC TCA CCG AGT G-3′) and norV-914r (5′-GAG CGG AAG ACA TTG GTC AGG-3′). To detect the ospG gene (31), the primer pair was ospG-F (5′-CCA TTT GAG AAT AAT AAT TCT CAT GCT G-3′) and ospG-R (5′-GCA TTT GTA ATC GTC GGT CGA TAA TC-3′).

Titration of Stx in EHEC cultures.

To detect Shiga toxin in the culture medium of the EHEC isolates, the strains were grown in Casamino Acid-yeast extract (CA-YE) medium (Denka Seiken Co., Ltd., Tokyo, Japan) overnight with shaking at 35°C. Each culture was centrifuged (900 × g for 15 min), and the supernatant was used for the Stx1 and Stx2 assays. Each toxin was detected in a 2-fold dilution series of the supernatant by a verotoxin E. coli reversed passive latex agglutination assay (VTEC-RPLA) (Denka Seiken Co., Ltd., Tokyo, Japan), according to the manufacturer's specifications. The toxin titers were expressed as the maximum dilution with a positive reaction.

RESULTS

Isolation of EHEC from patients.

During this outbreak, we identified 34 patients with HUS cases and, in the early phase of the outbreak, stx-negative E. coli O111-positive cases were identified. Therefore, we reexamined the stool specimens that had given culture-negative results. After enrichment using anti-LPS (O111 or O157) antibody-conjugated magnetic beads, multiple colonies (usually 4 or more; at most, 800 colonies) were checked for stx genes. A total of 55 patients were found to be positive for EHEC O111 with fliCH8 and/or O157 with fliCH7. These laboratory-confirmed cases were divided into three groups: only EHEC O111 isolated (group 1), both EHEC O157 and O111 isolated (group 2), and only EHEC O157 isolated (group 3). The groups contained 25, 12, and 18 cases, respectively (Table 1). Of the 37 EHEC O111-positive cases (groups 1 and 2), 17 developed HUS, including 13 cases of acute encephalopathy. However, of the 30 EHEC O157-positive cases (groups 2 and 3), only 9 cases developed HUS, including 6 cases of acute encephalopathy. Therefore, the HUS rate in patients infected with group 1 EHEC was 28%, with group 2 EHEC was 67%, and with group 3 EHEC was 6%.

TABLE 1.

Numbers of patients in the groups defined by isolation of EHEC in stool specimens

Clinical presentationa No. of patients in groups based on isolation of EHEC serotypesb
 Total no.
Group 1 Group 2 Group 3 Group 4
HUS + AE, death 3 0 0 2 5
HUS + AE 4 6 0 6 16
HUS 2 2 1 8 13
Non-HUSc 16 4 17 110 147
Total no. of patients 25 12 18 126 181
Open in a new tab
a

HUS, hemolytic uremic syndrome; AE, acute encephalopathy.

b

Group 1, only EHEC O111 isolated from these patients; group 2, both EHEC O111 and EHEC O157 isolated from these patients; group 3, only EHEC O157 isolated from these patients; and group 4, neither EHEC O111 nor EHEC O157 isolated from these patients.

c

More than one gastrointestinal symptom.

We did not isolate any EHEC strains from 126 of the 181 cases (Table 1, group 4). Of these 126 cases from group 4, 16 developed HUS, for an HUS rate of 13%, and 8 of these 16 HUS cases developed acute encephalopathy. In the serum samples of 10 patients of the 16 group 4 HUS cases, the anti-O111 antibody was detectable in all samples, whereas the antibody to O157 was only detectable in two samples. These data indicate that EHEC O111 isolates caused most infections, but these patients were no longer shedding EHEC (32).

An interesting feature of this outbreak was that stx-negative E. coli O111 strains were isolated from patient stool specimens. In addition, EHEC O157 strains with several types of toxins were isolated. These results are summarized in Table 2. stx-negative E. coli O111 organisms were isolated from 28 of the 55 EHEC-positive cases (groups 1, 2, and 3) and from 24 of the 126 EHEC-negative cases (group 4). Further analysis was carried out on 37 EHEC O111 stx2, 49 stx-negative E. coli O111, 25 EHEC O157 stx1, 10 EHEC O157 stx2, and 4 EHEC O157 stx1 stx2 strains, in addition to an EHEC O111 stx2 and a stx-negative E. coli O111 isolate from a beef sample from the same lot as the suspected contaminated food, the raw beef dish yukhoe.

TABLE 2.

Profiles of serotypes and toxins of EHEC strains isolated from outbreak patients

Groupa No. of cases Serogroup O111
 Serogroup O157
stx2 stx− stx1 stx2 stx1 stx2
1 10 +b −b − − −
15 + + − − −
2 5 + + + − −
2 + + + + −
3 + − + − −
1 + − + − +
1 + + − + −
3 3 − + + − −
2 − + + + −
5 − − + − −
1 − − + + −
3 − − + − +
4 − − − + −
4 102 − − − − −
24 − + − − −
Total no. of positive isolates 37 52c 25 10 4
Open in a new tab
a

Groups 1 to 4 are defined in the Table 1 footnotes.

b

Symbols indicate that the strain was isolated (+) or not isolated (−).

c

Of these 52 isolates, 3 were not available and were therefore excluded from further analysis.

Molecular typing of E. coli isolates.

If an EHEC strain loses its Stx prophage during infection, the resulting strain is an stx-negative eae-positive strain. Therefore, PFGE molecular typing was carried out to investigate the genetic relationships between the stx-positive and -negative EHEC O111 strains in this study. The PFGE patterns are shown in Fig. 1. All EHEC O111 stx2 strains, including an isolate from a beef sample, had the same XbaI digestion pattern (Fig. 1A, lane 1), except one strain that had a similar pattern but with a three-band difference (Fig. 1A, lane 3). Of the 50 stx-negative EHEC O111 strains, 48 (96%), including an isolate from a beef sample, had the pattern shown in Fig. 1A, lane 2. There was only a two-band difference between the patterns of almost all stx-positive and -negative EHEC O111 strains (cf. Fig. 1A, lanes 1 and 2). A comparison of these PFGE patterns identified an approximately 550-kb band in stx-positive EHEC O111 strains that was not present in the patterns of stx-negative EHEC O111 strains, as well as an approximately 490-kb band in stx-negative EHEC O111 strains that was not present in the patterns of stx-positive EHEC O111 strains. The size difference of these bands, 60 kb, corresponds to the genome size of Stx-converting phages (33).

FIG 1.

FIG 1

Open in a new tab

PFGE patterns of the E. coli O111 and E. coli O157 isolates and toxin types present. (A) Serogroup O111 isolates: lanes 1 and 3, stx2-positive strains; lanes 2, 4, and 5, stx-negative strains. (B) Serogroup O157 isolates: lane 1, 16 stx1 stx2, 10 stx1, and four stx2-positive strains: lane 2, eight stx1 stx2-positive strains; lane 3, one stx1 stx2-positive strain. Each lane shows a representative PFGE pattern for each strain derived from the photographs of the separated gels. The tables in panels A and B show the number of strains with identical PFGE patterns and list a breakdown of the number of strains with each toxin type. Lane M, molecular markers; stx−, stx-negative.

To confirm the genetic relationship of the EHEC O111 isolates from the outbreak, MLVA was carried out. Of the 88 E. coli O111 isolates, including the isolate from a beef sample, 80 (including both stx-positive and -negative isolates) had an identical MLVA profile. The remaining eight EHEC O111 isolates had similar MLVA profiles, but each had a repeat number variation at one locus among the 18 loci. These data strongly suggest that the E. coli O111 strains isolated during this outbreak were genetically closely related. The isolates from beef had a PFGE type (Fig. 1A, lane 1) and MLVA type (MLVA-O111) that appeared to be identical to those of the most prevalent stx-positive and -negative E. coli O111 isolates from clinical specimens, supporting the hypothesis that yukhoe beef was the vehicle for this outbreak.

The EHEC O157 isolates in this study were also analyzed by PFGE and MLVA to investigate the genetic relatedness of the three different toxin types present in these strains. There were three groups of EHEC O157 isolates based on a PFGE analysis of their XbaI restriction digest patterns, with one- to three-band differences (Fig. 1B). Of the 39 O157 isolates, 30 showed an identical PFGE pattern (Fig. 1B, lane 1). This pattern was found in all the EHEC O157 isolates carrying stx1 stx2, stx1, and stx2. An analysis of the MLVA results showed that the 39 EHEC O157 strains could be divided into five MLVA types (Fig. 2B). MLVA-O157 was the most commonly found MLVA type in 31 of the 39 strains and included strains with all three toxin types: stx1 stx2, stx1, and stx2. The remaining eight EHEC O157 strains were single-locus variants of MLVA-O157, each with a repeat number variation at locus 10 (see Fig. 2, table for E. coli O157 in Fig. 2). Therefore, although EHEC O157 strains possessing different toxin types were isolated during this outbreak, all of the strains may have originated from a single common strain.

FIG 2.

FIG 2

Open in a new tab

Minimum spanning trees of MLVA of the E. coli O111 and E. coli O157 isolates. Shown are serogroup O111 isolates with stx2 (black circle) or stx-negative E. coli (circle with dots) (A) and serogroup O157 isolates with stx1 stx2 (white circle), stx1 (gray circle), or stx2 (black circle) (B). One circle represents one MLVA type, and the size of the circle is proportional to the number of isolates. The tables in panels A and B show the MLVA profiles for E. coli O111 and E. coli O157 strains, with the gray columns indicating the locus detected single-locus variants (SLVs). Each SLV represents each satellite circle with the number of isolates and its locus type.

Shiga toxin production and other virulence markers.

This outbreak produced an extraordinary number of cases of HUS and encephalopathy. In this outbreak, proinflammatory cytokines were induced in most patients with acute encephalopathy and severe HUS (34, 35). In addition to E. coli LPS, Stx1 and Stx2 can induce expression and synthesis of cytokines in Caco-2 cells, a human colon epithelial cell line (36). Therefore, we attempted to estimate the amount of Stx1 and Stx2 present in the culture media of EHEC strains isolated from outbreak patients. The 8 isolates of EHEC O111 stx2 and 3 isolates of EHEC O157 stx1 stx2 produced Stx2 titers of 1:64 and 1:32, respectively, which were lower than those produced by EHEC O111 strain 11128 and EHEC O157 strain Sakai (1:128 and 1:256, respectively). The Stx1 titer of the EHEC O157 stx1 stx2 strains was also lower than that of the EHEC O157 strain Sakai (1:32 versus 1:128).

To investigate additional potential virulence factors, like the hybrid type of STEC isolated in Germany in 2011 (36), the presence of major diarrheagenic E. coli virulence factors was examined by PCR. EHEC O111 stx2, toxin-negative E. coli, and EHEC O157 isolates carried eae, hlyA, and ospG but were negative for CVD432, aggR, invE, elt, esth, bfp, EAF, and astA (data not shown). EHEC O111 strains had an intact norV gene, but EHEC O157 strains had a deletion in the norV gene (data not shown).

Furthermore, PCR subtyping of the Stx1 and Stx2 toxins from both the EHEC O111 and O157 strains in this outbreak showed that the toxins were types Stx1a and Stx2a.

Stability of the stx2 gene during EHEC subculture.

The EHEC isolates from some clinical specimens appeared to have lost the stx2 gene during isolation and growth on the agar plates. For some EHEC O111 strains, we noted that the PFGE patterns, which were confirmed by PCR to be stx2 positive, sometimes changed to the PFGE pattern of stx-negative E. coli O111 strains. To investigate the stability of the Stx2 prophage, 10 O111 colonies (1 to 10) isolated from a stool specimen were picked, and each colony was spread on a TSA plate. A portion of each overnight subculture was transferred to a new TSA plate, and then this procedure was repeated to produce two subcultures. After the overnight culture, the subcultures were tested using colony-sweep PCR for stx2, and the five stx2-positive colonies were selected from the primary isolates for the second subculture (Fig. 3). After the first subculture, the amount of PCR amplicon produced from isolate 6 was less than that produced from the other isolates. After the second subculture, no PCR stx2 amplicon was seen in two of the five isolates (Fig. 3, isolates 1 and 6). The E. coli O111 strain, which converted to stx2 negative in this experiment, had a PFGE pattern that was identical to that of the stx-negative E. coli O111 strains (data not shown), indicating that an Stx2 prophage had been lost during subculture. Collectively, these data indicate that some Stx2 prophages in the EHEC O111 strains isolated in this outbreak were unstable during in vitro cultivation, and they suggest that loss of the Stx2 prophage may have occurred in the infected patients.

FIG 3.

FIG 3

Open in a new tab

Stability of stx2 genes of the five EHEC O111 stx2 strains in successive subcultures. Ten O111 colonies (1 to 10) isolated from a stool specimen were picked, and each colony was spread on a TSA plate (A). A portion of each overnight subculture was transferred to a new TSA plate, and then this procedure was repeated to produce two subcultures. Following overnight incubation, the subcultures were tested using colony-sweep PCR for stx2 (A, B, and C), and the five stx2-positive colonies (1, 2, 4, 5, and 6) from the primary isolates were selected for the second round of subculture (C). The stx2 and 16S rRNA gene amplicons were 404 bp and 544 bp, respectively. Sakai, EHEC O157 stx1 stx2 Sakai strain as a positive control; NC, template DNA-free reaction as a negative control.

DISCUSSION

Several reports have identified EHEC-associated HUS cases that no longer shed EHEC but do shed stx-negative EHEC organisms that presumably lost stx during infection (18, 37–39). Stx is the most important EHEC virulence factor, causing the specific clinical features of EHEC infection (1–4). Therefore, stx-negative E. coli organisms may be produced by excision of the Stx-converting phage during infection. In agreement with this suggestion, more stx-negative EHEC strains have been isolated at follow-up testing of EHEC-infected patients (38). In HUS patients who shed only stx-negative EHEC, the etiology of HUS can be missed using current bacteriological methods based on detecting only the stx gene or Stx. This can hamper epidemiological investigations and lead to inappropriate clinical management, especially with a cluster of sporadic cases.

The etiology of the EHEC O111/O157 outbreak studied here was confirmed in only 18 of the 34 HUS cases by isolating EHEC O111 (9 cases), EHEC O157 (1 case), and both EHEC O111 and O157 (8 cases). The remaining 16 HUS cases were EHEC negative, even after extensive efforts to isolate EHEC, with up to 800 E. coli colonies examined after O111 enrichment using immunomagnetic beads. Since serologically positive results for the E. coli O111 antigen were found from all but one of the EHEC-negative HUS cases (32), EHEC O111 may be the primary cause of the severe HUS complications. An stx-negative E. coli O111 strain was isolated from most EHEC-positive HUS cases (13 of 18 HUS cases) and from a few EHEC-negative HUS cases (3 of 16 HUS cases). We also isolated EHEC O111 stx2 and stx-negative E. coli O111 organisms from a beef sample from the same lot as the suspected infection vehicle. Therefore, conversion from stx-positive to stx-negative E. coli O111 may occur in contaminated food, during infection, and/or bacteriological testing. The subculture data presented here show that the EHEC O111 stx2 isolates in this study can lose the Stx2 prophage during in vitro subculture (Fig. 3). This may have biased our bacteriological analyses, resulting in an underestimation of the scale of the outbreak. Since the loss of the stx gene was not consistent in all EHEC O111 isolates in this study, with some isolates remaining stx2 positive during subculture and other isolates losing stx2, the molecular mechanism and factors affecting the loss of the stx2 phage require further study.

Bielaszewka et al. (40) reported the isolation of stx-negative EHEC organisms from approximately 5% of the HUS patients. In that report, the majority of the stx-negative/eae-positive isolates belonged to serogroups O26, O103, O145, and O157:H7/NM, but no stx-negative/eae-positive E. coli O111 isolates were reported. The absence of stx-negative/eae-positive E. coli O111 isolates in the HUS patients may be explained by the fact that a majority of the EHEC O111 isolates possessed stx1 solely or in combination with stx2 (13), and the Stx1 prophage of EHEC O111 is thought to be defective, resulting in immobilization and stability in the EHEC chromosome (41). In contrast to a previous investigation (39), there were 16 HUS cases (47.1% [16 in 34 HUS cases]) in this outbreak from patients whose stools contained stx-negative E. coli O111 strains. It is probably because EHEC O111 in this outbreak possessed the Stx2 prophage only, which was unstable and progressively lost from the genome. An HUS outbreak in Italy in 1992 may have involved a similar loss of Stx phages during EHEC O111 infection (15). That HUS outbreak had nine HUS cases, including 6 cases that were diagnosed by detecting a serum antibody to E. coli O111 LPS. Stx-producing E. coli was isolated from a stool specimen in only one case. Unfortunately, there were no data on the isolation of stx-negative strains.

The prevalence of HUS (19% [34 of 181 EHEC-infected patients]) was unexpectedly high in the 2011 outbreak, even if the suspected cases are included in the calculation. The most probable explanation for this relatively high HUS prevalence is that the concentration of EHEC in the EHEC-contaminated food was high. However, the EHEC O111 and EHEC O157 concentrations in the vehicles for the outbreak studied here are not known. We were unable to examine the raw beef dish, yukhoe, which was the vehicle for EHEC O111, and thus far, the vehicle for EHEC O157 has not been confirmed. The cytokine profiles of the patients with serious complications in this outbreak indicate that massive induction of proinflammatory cytokines may have contributed to the development of serious complications (34, 35). The high level of cytokine induction may have been due to LPS release in the intestinal tract, although it remains unclear how LPS-dependent induction might have occurred in this outbreak. In addition, although Stx1 and Stx2 may induce cytokines, including tumor necrosis factor alpha, the amount of Stx2 production in the EHEC O111 strains, as well as Stx1 and Stx 2 production in the EHEC O157 strains, in this outbreak was low or similar to that in EHEC O111 strain 11128 and EHEC O157 Sakai by in vitro testing. Unfortunately, we did not determine the reasons for this high HUS prevalence using the in vitro bacteriological testing alone. Another possibility is that the instability of the Stx2 prophages in the EHEC O111 strains may have played a role in the high HUS prevalence in this outbreak. Mellmann et al. (42) suggested that stx-negative EHEC strains might be the recipients of Stx-converting phages from isogenic stx-positive strains. In fact, we detected many Stx2 phage plaques in the culture lysates of some EHEC O111 isolates following incubation with mitomycin C, as well as in some of the bloody stool specimens (data not shown). Almost all of these plaques were shown to be stx2 positive by the PCR assay, and these phages were found to be functional Stx2-converting phages. This would suggest a highly dynamic system that converts in both directions by the loss or gain of Stx2 phages. This cycling lifestyle might be enhanced by a mixture of stx-positive EHEC organisms and its isogenic stx-negative strain in the contaminated vehicle or during infection. Further studies of integrated prophages in both the EHEC O111 and EHEC O157 strains isolated during this outbreak need to be carried out, because the dynamic conversion of these strains might influence the outcome of disease.

ACKNOWLEDGMENTS

We thank the public health centers of Takaoka, Tonami, and Toyama City, the medical laboratories of Tonami General Hospital, Kouseiren Takaoka Hospital, Takaoka City Hospital, and Toyama University Hospital for collecting clinical specimens during the 2011 EHEC outbreak, and Yoshihiro Takada of the Toyama Institute of Health for continuing encouragement in this study.

This work was supported by a Special Research grant (H23-TOKUBETU-SHITEI-004 and H24-SHINKO-IPPAN-012) and a Health Labor Sciences Research grant from the Ministry of Health, Labor, and Welfare, Japan.

We declare that we have no conflicts of interest.

Footnotes

Published ahead of print 14 May 2014

REFERENCES

  • 1.O'Brien AD, Kaper JB. 1998. Shiga toxin-producing Escherichia coli: yesterday, today, and tomorrow, p 1–11 In Kaper JB, O'Brien AD. (ed), Escherichia coli O157:H7 and other Shiga toxin-producing E. coli strains. ASM Press, Washington, DC [Google Scholar]
  • 2.Nataro JP, Kaper JB. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142–201 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Karmali MA, Steele BT, Petric M, Lim C. 1983. Sporadic cases of haemolytic-uraemic syndrome associated with faecal cytotoxin and cytotoxin-producing Escherichia coli in stools. Lancet 321:619–620. 10.1016/S0140-6736(83)91795-6 [DOI] [PubMed] [Google Scholar]
  • 4.Mayer CL, Leibowitz CS, Kurosawa S, Stearns-Kurosawa DJ. 2012. Shiga toxins and the pathophysiology of hemolytic uremic syndrome in humans and animals. Toxins (Basel) 4:1261–1287. 10.3390/toxins4111261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.National Institute of Infectious Diseases. 2013. Enterohemorrhagic Escherichia coli infection in Japan as of April 2013. IASR 34:123–124 http://www.nih.go.jp/niid/en/iasr-vol34-e/865-iasr/3570-tps399.html [Google Scholar]
  • 6.National Institute of Infectious Diseases. 2007. Enterohemorrhagic Escherichia coli infection in Japan as of April 2007. IASR 28:131–132 http://idsc.nih.go.jp/iasr/28/327/tpc327.html [Google Scholar]
  • 7.National Institute of Infectious Diseases. 2008. Enterohemorrhagic Escherichia coli infection in Japan as of April 2008. IASR 29:117–118 http://idsc.nih.go.jp/iasr/29/339/tpc339.html [Google Scholar]
  • 8.National Institute of Infectious Diseases. 2009. Enterohemorrhagic Escherichia coli infection in Japan as of April 2009. IASR 30:119–120 http://idsc.nih.go.jp/iasr/30/351/tpc351.html [Google Scholar]
  • 9.National Institute of Infectious Diseases. 2010. Enterohemorrhagic Escherichia coli infection in Japan as of May 2010. IASR 31:151–152 http://idsc.nih.go.jp/iasr/31/364/tpc364.html [Google Scholar]
  • 10.National Institute of Infectious Diseases. 2011. Enterohemorrhagic Escherichia coli infection in Japan as of April 2011. IASR 32:125–126 http://idsc.nih.go.jp/iasr/32/375/tpc375.html [Google Scholar]
  • 11.Brooks JT, Sowers EG, Wells JG, Greene KD, Griffin PM, Hoekstra RM, Strockbine NA. 2005. Non-O157 Shiga toxin-producing Escherichia coli infections in the United States, 1983–2002. J. Infect. Dis. 192:1422–1429. 10.1086/466536 [DOI] [PubMed] [Google Scholar]
  • 12.Beutin L, Krause G, Zimmermann S, Kaulfuss S, Gleier K. 2004. Characterization of Shiga toxin-producing Escherichia coli strains isolated from human patients in Germany over a 3-year period. J. Clin. Microbiol. 42:1099–1108. 10.1128/JCM.42.3.1099-1108.2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Zhang W, Mellmann A, Sonntag AK, Wieler L, Bielaszewska M, Tschäpe H, Karch H, Friedrich AW. 2007. Structural and functional differences between disease-associated genes of enterohaemorrhagic Escherichia coli O111. Int. J. Med. Microbiol. 297:17–26. 10.1016/j.ijmm.2006.10.004 [DOI] [PubMed] [Google Scholar]
  • 14.Elliott EJ, Robins-Browne RM, O'Loughlin EV, Bennett-Wood V, Bourke J, Henning P, Hogg GG, Knight J, Powell H, Redmond D, Contributors to the Australian Paediatric Surveillance Unit 2001. Nationwide study of haemolytic uraemic syndrome: clinical, microbiological, and epidemiological features. Arch. Dis. Child. 85:125–131. 10.1136/adc.85.2.125 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Caprioli A, Luzzi I, Rosmini F, Resti C, Edefonti A, Perfumo F, Farina C, Goglio A, Gianviti A, Rizzoni G. 1994. Community-wide outbreak of hemolytic uremic syndrome associated with non-O157 verocytotoxin-producing Escherichia coli. J. Infect. Dis. 169:208–211. 10.1093/infdis/169.1.208 [DOI] [PubMed] [Google Scholar]
  • 16.Piercefield EW, Bradley KK, Coffman RL, Mallonee SM. 2010. Hemolytic uremic syndrome after an Escherichia coli O111 outbreak. Arch. Intern. Med. 170:1656–1663. 10.1001/archinternmed.2010.346 [DOI] [PubMed] [Google Scholar]
  • 17.Karch H, Meyer T, Rüssmann H, Heesemann J. 1992. Frequent loss of Shiga-like toxin genes in clinical isolates of Escherichia coli upon subcultivation. Infect. Immun. 60:3464–3467 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Friedrich AW, Zhang W, Bielaszewska M, Mellmann A, Köck R, Fruth A, Tschäpe H, Karch H. 2007. Prevalence, virulence profiles, and clinical significance of Shiga toxin-negative variants of enterohemorrhagic Escherichia coli O157 infection in humans. Clin. Infect. Dis. 45:39–45. 10.1086/518573 [DOI] [PubMed] [Google Scholar]
  • 19.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. U. S. A. 106:17939–17944. 10.1073/pnas.0903585106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hayashi T, Makino K, Ohnishi M, Kurokawa K, Ishii K, Yokoyama K, Han CG, 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. 8:11–22. 10.1093/dnares/8.1.11 [DOI] [PubMed] [Google Scholar]
  • 21.Karch H, Janetzki-Mittmann C, Aleksic S, Datz M. 1996. Isolation of enterohemorrhagic Escherichia coli O157 strains from patients with hemolytic-uremic syndrome by using immunomagnetic separation, DNA-based methods, and direct culture. J. Clin. Microbiol. 34:516–519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Fukushima H, Hoshina K, Gomyda M. 2000. Selective isolation of eae-positive strains of Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 38:1684–1687 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Beutin L, Strauch E. 2007. Identification of sequence diversity in the Escherichia coli fliC genes encoding flagellar types H8 and H40 and its use in typing of Shiga toxin-producing E. coli O8, O22, O111, O174 and O179 strains. J. Clin. Microbiol. 45:333–339. 10.1128/JCM.01627-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hunter SB, Vauterin P, Lambert-Fair MA, Van Duyne MS, Kubota K, Graves L, Wrigley D, Barrett T, Ribot E. 2005. Establishment of a universal size standard strain for use with the PulseNet standardized pulsed-field gel electrophoresis protocols: converting the national databases to the new size standard. J. Clin. Microbiol. 43:1045–1050. 10.1128/JCM.43.3.1045-1050.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Izumiya H, Pei Y, Terajima J, Ohnishi M, Hayashi T, Iyoda S, Watanabe H. 2010. New system for multilocus variable-number tandem-repeat analysis of the enterohemorrhagic Escherichia coli strains belonging to three major serogroups: O157, O26, and O111. Microbiol. Immunol. 54:569–577. 10.1111/j.1348-0421.2010.00252.x [DOI] [PubMed] [Google Scholar]
  • 26.Hyytiä-Trees E, Smole SC, Fields PA, 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. 3:118–131. 10.1089/fpd.2006.3.118 [DOI] [PubMed] [Google Scholar]
  • 27.Walsh PS, Metzger DA, Higuchi R. 1991. Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. Biotechniques 10:506–513 [PubMed] [Google Scholar]
  • 28.Kimata K, Shima T, Shimizu M, Tanaka D, Isobe J, Gyobu Y, Watahiki M, Nagai Y. 2005. Rapid categorization of pathogenic Escherichia coli by multiplex PCR. Microbiol. Immunol. 49:485–492. 10.1111/j.1348-0421.2005.tb03752.x [DOI] [PubMed] [Google Scholar]
  • 29.Scheutz F, Teel LD, Beutin L, Piérard D, Buvens G, Karch H, Mellmann A, Caprioli A, Tozzoli R, Morabito S, Strockbine NA, Melton-Celsa AR, Sanchez M, Persson S, O'Brien AD. 2012. Multicenter evaluation of a sequence-based protocol for subtyping Shiga toxins and standardizing Stx nomenclature. J. Clin. Microbiol. 50:2951–2963. 10.1128/JCM.00860-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kulasekara BR, Jacobs M, Zhou Y, Wu Z, Sims E, Saenphimmachak C, Rohmer CL, Ritchie JM, Radey M, McKevitt M, Freeman TL, Hayden H, Haugen E, Gillett W, Fong C, Chang J, Beskhlebnaya V, Waldor MK, Samadpour M, Whittam TS, Kaul R, Brittnacher M, Miller SL. 2009. Analysis of the genome of the Escherichia coli O157:H7 2006 spinach-associated outbreak isolate indicates candidate genes that may enhance virulence. Infect. Immun. 77:3713–3721. 10.1128/IAI.00198-09 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Nobe R, Nougayrède JP, Taieb F, Bardiau M, Cassart D, Navarro-Garcia F, Mainil J, Hayashi T, Oswald E. 2009. Enterohaemorrhagic Escherichia coli serogroup O111 inhibits NF-κB-dependent innate responses in a manner independent of a type III secreted OspG orthologue. Microbiology 155:3214–3225. 10.1099/mic.0.030759-0 [DOI] [PubMed] [Google Scholar]
  • 32.Isobe J, Shima T, Kanatani J, Kimata K, Shimizu M, Kobayashi N, Tanaka T, Iyoda S, Ohnishi M, Sata T, Watahiki M. 2014. Serodiagnosis using microagglutination assay during the food-poisoning outbreak in Japan caused by consumption of raw beef contaminated with enterohemorrhagic Escherichia coli O111 and O157. J. Clin. Microbiol. 52:1112–1118. 10.1128/JCM.03469-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Plunkett G, III, Rose DJ, Durfee TJ, Blattner FR. 1999. Sequence of Shiga toxin 2 phage 933W from Escherichia coli O157:H7: Shiga toxin as a phage late-gene product. J. Bacteriol. 181:1767–1678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Shimizu M, Kuroda M, Sakashita N, Konishi M, Kaneda H, Igarashi N, Yamahana J, Taneichi H, Kanegane H, Ito M, Saito S, Ohta K, Taniguchi T, Furuichi K, Wada T, Nakagawa M, Yokoyama H, Yachie A. 2012. Cytokine profiles of patients with enterohemorrhagic Escherichia coli O111-induced hemolytic-uremic syndrome. Cytokine 60:694–700. 10.1016/j.cyto.2012.07.038 [DOI] [PubMed] [Google Scholar]
  • 35.Shimizu M, Kuroda M, Inoue N, Konishi M, Igarashi N, Taneichi H, Kanegane H, Ito M, Saito S, Yachie A. 2014. Extensive serum biomarker analysis in patients with enterohemorrhagic Escherichia coli O111-induced hemolytic-uremic syndrome. Cytokine 66:1–6. 10.1016/j.cyto.2013.12.005 [DOI] [PubMed] [Google Scholar]
  • 36.Yamasaki C, Natori Y, Zeng XT, Ohmura M, Yamasaki S, Takeda Y, Natori Y. 1999. Induction of cytokines in a human colon epithelial cell line by Shiga toxin 1 (Stx1) and Stx2 but not by non-toxic mutant Stx1 which lacks N-glycosidase activity. FEBS Lett. 442:231–234. 10.1016/S0014-5793(98)01667-6 [DOI] [PubMed] [Google Scholar]
  • 37.Bielaszewska M, Mellmann A, Zhang W, Köck R, Fruth A, Bauwens A, Peters G, Karch H. 2011. Characterisation of the Escherichia coli strain associated with an outbreak of haemolytic uraemic syndrome in Germany, 2011: a microbiological study. Lancet Infect. Dis. 11:671–676. 10.1016/S1473-3099(11)70165-7 [DOI] [PubMed] [Google Scholar]
  • 38.Feng P, Dey M, Abe A, Takeda T. 2001. Isogenic strain of Escherichia coli O157:H7 that has lost both Shiga toxin 1 and 2 genes. Clin. Diagn. Lab. Immunol. 8:711–717 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Mellmann A, Bielaszewska M, Zimmerhackl LB, Prager R, Harmsen D, Tschäpe H, Karch H. 2005. Enterohemorrhagic Escherichia coli in human infection: in vivo evolution of a bacterial pathogen. Clin. Infect. Dis. 41:785–792. 10.1086/432722 [DOI] [PubMed] [Google Scholar]
  • 40.Bielaszewska M, Köck R, Friedrich AW, von Eiff C, Zimmerhackl LB, Karch H, Mellmann A. 2007. Shiga toxin-mediated hemolytic uremic syndrome: time to change the diagnostic paradigm? PLoS One 2:e1024. 10.1371/journal.pone.0001024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Creuzburg K, Köhler B, Hempel H, Schreier P, Jacobs E, Schmidt H. 2005. Genetic structure and chromosomal integration site of the cryptic prophage CP-1639 encoding Shiga toxin. Microbiology 151:941–950. 10.1099/mic.0.27632-0 [DOI] [PubMed] [Google Scholar]
  • 42.Mellmann A, Lu S, Karch H, Xu J, Harmsen D, Schmidt MA, Bielaszewska M. 2008. Recycling of Shiga toxin 2 genes in sorbitol-fermenting enterohemorrhagic Escherichia coli O157:NM. Appl. Environ. Microbiol. 74:67–72. 10.1128/AEM.01906-07 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES