Occurrence of potentially human-pathogenic Escherichia coli O103 in

Similar documents
Extra-intestinal pathogenic E. coli carrying the. shiga toxin gene stx2

Toxins 2011, 3, manuscripts; doi: /toxins Short Note

The surveillance and control programme

Surveillance and control programmes for terrestrial and aquatic animals in Norway

The surveillance programme for Campylobacter spp. in broiler flocks in Norway 2013

The surveillance programme for Campylobacter spp. in broiler flocks in Norway 2015

Annual Reports Norwegian Veterinary Institute. Norwegian Veterinary Institute

The surveillance and control programme

E. coli 0157:H7. By Christopher Tong

The surveillance programme for Campylobacter spp in broiler flocks in Norway 2017

Alberta Provincial Laboratory for Public Health, Edmonton, Alberta, Canada 1 ;

Virulence profiles of Shiga toxin 2e-producing Escherichia coli isolated from healthy pig at slaughter

Foodborne Outbreak of E. coli Infections and Hemolytic Uremic Syndrome in Germany, 2011

Title: The Operon Encoding SubAB a Novel Cytotoxin is Present in US STEC Isolates ACCEPTED

Received 4 August 2004/Accepted 6 January 2005

Shiga Toxin producing E. coli (STEC) in food which serotypes are important? Lothar Beutin

IDENTIFICATION OF THE asta GENE IN ENTEROTOXIGENIC ESCHERICHIA COLI STRAINS RESPONSIBLE FOR DIARRHEA IN PIGS

Human Infections with Non-O157 Shiga Toxin producing Escherichia

Update on infections with and clinical lab guidelines for Shiga toxin-producing E. coli (STEC) in the United States

Serotypes, Virulence Genes, and Intimin Types of Shiga Toxin (Verotoxin)-Producing Escherichia coli Isolates from Healthy Sheep in Spain

Enterovirulent Escherichia coli. Tom Cheasty Laboratory of Enteric Pathogens

Enterohemorrhagic Escherichia coli in Human Infection: In Vivo Evolution of a Bacterial Pathogen

NATIONAL REFERENCE CENTRE FOR SHIGA TOXIN/VEROTOXIN- PRODUCING ESCHERICHIA COLI (NRC STEC/VTEC)

Medical Microbiology Coursework Essay High Class 1 essay. North Germany. This outbreak caused the highest frequency of HUS cases caused by

The surveillance programme for bluetongue in Norway 2014

Surveillance and control programmes for terrestrial and aquatic animals in Norway

International Registries: The Government-Driven Model

Multi-clonal origin of macrolide-resistant Mycoplasma pneumoniae isolates. determined by multiple-locus variable-number tandem-repeat analysis

Key words : Verotoxin-producing Escherichia coli, VTEC, EHEC, imported meat

Received 3 September 2010/Accepted 7 January 2011

Genetic Diversity among Clonal Lineages within Escherichia coli O157:H7 Stepwise Evolutionary Model

The surveillance and control programme for maedi in Norway

The surveillance and control programme for maedi in Norway

Escherichia coli diagnostics

Shiga toxin-producing Escherichia coli

Serotypes of verocytotoxigenic Escherichia coli isolated from cattle and buffalo calf diarrhoea

Received 23 February 2011/Returned for modification 28 March 2011/Accepted 5 April 2011

Epidemiology of Verotoxigenic E. coli O157 in Ireland, 2003 Patricia Garvey and Paul McKeown

EPIDEMIOLOGY OF SHIGA TOXIN-PRODUCING E. COLI (STEC) IN THE FINISHING PIGS AND HUMANS. Marion Tseng

Nucleotide Polymorphisms within the O-antigen Gene Cluster of Escherichia coli O26,

Robert Tauxe, MD, MPH

In May 2011 there was a large scale outbreak of Haemolytic Uraemic Syndrome (HUS) in

The surveillance and control programme for specific virus infections in swine herds in Norway

Phenotypic and genotypic characterization of Escherichia Coli O111 serotypes

PT18 Identification and typing of STEC and other pathogenic E. coli

Prevalence and Some Properties of Verotoxin (Shiga-Like Toxin)- Producing Escherichia coli in Seven Different Species

Subtype-Specific Suppression of Shiga Toxin 2 Released from Escherichia coli upon Exposure to Protein Synthesis Inhibitors

Serotypes, Virulence Genes, and Intimin. Patients: Prevalence in Lugo, Spain, from 1992 through 1999

Verocytotoxin (vtx)-producing Escherichia coli (VTEC),

Michael Weizenegger Laboratory Group Limbach, Heidelberg, Germany, Dept. of Microbiology and Molecular Genetic

Surveillance and outbreak investigation of Shiga toxin-producing Escherichia coli using whole genome sequencing- time for a change!

Improving the Detection of Shiga Toxin- Producing Escherichia coli (STEC)

Discrimination of enterohemorrhagic E. coli (EHEC) from non-ehec strains. based on detection of various combinations of non-lee-encoded type III

3/08/2012. EHECO104: Lessons for Australia from the German outbreak. E. coli Pathotypes. EHEC Reservoirs & Transmission. EHEC Virulence Markers

The surveillance and control programme for specific virus infections in swine herds in Norway 2012

JCM Accepts, published online ahead of print on 20 March 2013 J. Clin. Microbiol. doi: /jcm

Applicability of Phylogenetic Methods for Characterizing the Public Health Significance of Verocytotoxin-Producing Escherichia coli Strains

Escherichia coli: Great Diversity around a Common Core

OPINION 1 / 19. AFSSA Request No SA-0031 Related Request No SA Maisons-Alfort, 27 May 2010

stx 1c Is the Most Common Shiga Toxin 1 Subtype among Shiga Toxin-Producing Escherichia coli Isolates from Sheep but Not among Isolates from Cattle

Association of Enterohemorrhagic Escherichia coli Hemolysin with Serotypes of Shiga-Like-Toxin-Producing Escherichia coli of Human and Bovine Origins

Prevalence and Properties of Diarrheagenic Escherichia coli among Healthy Individuals in Osaka City, Japan

Edinburgh Research Explorer

Associations between Virulence Factors of Shiga Toxin-Producing Escherichia coli and Disease in Humans

3/18/ Update: STEC Diagnosis and Surveillance in Wisconsin. Objectives. Objectives. Shiga toxin-producing Escherchia coli (STEC)

2014 Update: STEC Diagnosis and Surveillance in Wisconsin

Investigation of a spatiotemporal cluster of verotoxinproducing Escherichia coli O157 infections in eastern England in 2007

Certifications and training for mammography screening services and professionals in Norway

Escherichia coli Verotoxigenic Infections

IPCVA, Buenos Aires - 7 December Infections with Shiga toxin producing E.coli (STEC): emerging issues and reflections on the global food trade

Campylobacter spp. and Yersinia enterocolitica in Growing Pigs in Iowa and North Carolina: A Pilot Study

eaea Genes in Escherichia coli Derived from Japanese Patients with Sporadic Diarrhea

Are all VTEC created Equal?

Shiga Toxin, Cytolethal Distending Toxin, and Hemolysin Repertoires in Clinical Escherichia coli O91 Isolates

Improved methodology for isolating Shiga toxin-producing E. coli (STEC) in food

EHEC O111 O157

OCCURRENCE AND CHARACTERISTICS OF VIRULENCE GENES OF ESCHERICHIA COLI STRAINS ISOLATED FROM HEALTHY DAIRY COWS IN INNER MONGOLIA, CHINA

Analysis of phenotypic variants of the serogroup C ET-15 clone of Neisseria meningitidis by pulsed-field gel electrophoresis.

Global food trade and emerging foodborne pathogens: The example of STEC O104

Escherichia coli Pathotypes Associated with Diarrhea in Romanian Children Younger than 5 Years of Age

Evaluation of Automated Ribotyping System for Characterization and Identification of Verocytotoxin-Producing Escherichia coli Isolated in Japan

Phenotypic and Genotypic Characteristics of Shiga Toxinproducing Escherichia coli Strains Isolated from Children in

Non-O157 Shiga toxin-producing E. coli: An emerging pathogen of public health importance

7th annual workshop of the NRL for E. coli in the EU Rome, 8th -9th November 2012

Whole genome sequencing

E. coli O157:H7 shedding in beef cattle. Jane Heller, Geraldine Lammers and Craig McConnel

Received 3 September 2007/Returned for modification 26 October 2007/Accepted 30 April 2008

Enhancing animal health security and food safety in organic livestock production

New genomic typing method MLST

A foodborne outbreak of entero-invasive E.coli (EIEC) infection in Italy, April 2012

Atypical Enteropathogenic Escherichia coli Infection and Prolonged Diarrhea in Children

Pig digest: Bacteriology Manakorn Sukmak

Bacterial Enteric Infections Detected by Culture-Independent Diagnostic Tests FoodNet, United States,

Scottish E. coli O157/Shiga toxin-producing E. coli Reference Laboratory

The Relationship between O-Antigens and Pathogenic Genes of Diarrhea-Associated Escherichia coli

ON-GOING STUDIES ON VTEC IN SWEDEN

E. coli O157:H7 Outbreak Associated with Consumption of Unpasteurized Milk, Kentucky, 2014

E.Coli 0157 Antigen Detection (In Food)

Trends in Toxin Profiles of Human Shiga Toxin- Producing Escherichia Coli (STEC) O157 Strains, United States,

Transcription:

AEM Accepts, published online ahead of print on 27 September 2013 Appl. Environ. Microbiol. doi:10.1128/aem.01825-13 Copyright 2013, American Society for Microbiology. All Rights Reserved. 1 2 Occurrence of potentially human-pathogenic Escherichia coli O103 in Norwegian sheep 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Camilla Sekse 1, Marianne Sunde 1, Petter Hopp 1, Torkjel Bruheim 2, Kofitsyo Sewornu Cudjoe 1, Bjørg Kvitle 1 and Anne Margrete Urdahl 1# 1 Norwegian Veterinary Institute, P.O. Box 750 Sentrum, NO-0106 Oslo, Norway 2 Norwegian Veterinary Institute, P.O. Box 5695 Sluppen, 7485 Trondheim, Norway # Corresponding author: Norwegian Veterinary Institute, P.O. Box 750 Sentrum, NO-0106 Oslo, Norway. Tel. +47 23 21 63 68, Fax. +47 23 21 64 85 E-mail; anne-margrete.urdahl@vetinst.no Running title: E. coli O103 in Norwegian sheep Keywords; STEC, EPEC, E. coli O103, prevalence, sheep, risk factors 1

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 ABSTRACT The investigation of an outbreak of hemorrhagic uremic syndrome in Norway in 2006 indicated that the outbreak strain Escherichia coli O103:H25 could originate from sheep. A national survey of the Norwegian sheep population was performed, with the aim of identifying and describing a possible reservoir of potentially human-pathogenic E. coli O103, in particular of the H-types H2 and H25. The investigation of fecal samples from 585 sheep flocks resulted in 1222 E. coli O103 isolates that were analyzed for the presence of eae and stx genes, while a subset of 369 isolates was further examined for flagellar antigens (Htyping), stx subtypes, bfpa and asta, and molecular typing by pulsed-field gel electrophoresis (PFGE). The total ovine E. coli O103 serogroup was genetically diverse by number of H-types, virulotypes and PFGE banding patterns identified, although a tendency of clustering towards serotypes was seen. The flocks positive for potentially human-pathogenic E. coli O103 were geographically widely distributed and no association could be found to county or geographical region. The survey showed that eae - stx - E. coli O103, probably non-pathogenic to humans, is very common in sheep with 27.5% positive flocks. Moreover, the study documented a low prevalence (0.7%) of potentially human-pathogenic Shiga toxin-producing E. coli O103:H2, while STEC O103:H25 was not detected. However, 3.1% and 5.8% of the flocks were positive for Enteropathogenic E. coli O103 belonging to H-type H2 and H25, respectively. These are of concern as potentially human pathogens by themselves, but more important as possible precursors for human-pathogenic STEC. 2

42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 INTRODUCTION Escherichia coli is mainly found as an intestinal commensal, although several groups of E. coli may cause disease such as diarrhea in humans and animals (1, 2). Shiga toxin-producing (STEC) and atypical enteropathogenic E. coli (aepec) are both zoonotic E. coli pathogroups with animal reservoirs. An array of virulence characteristics have been described for both STEC and aepec, and some, such as the adherence factor intimin (eae), is common in both groups (2, 3). In addition to intimin, STEC also possess one or more variants of Shiga toxin (Stx) that are encoded by stx 1 and/or stx 2 or variants thereof, and which are the major determinants of STEC pathogenicity (2). E. coli O103:H2 is one of the most common non-o157 STEC serotypes isolated from human cases in Europe. Sporadic cases of human infections caused by E. coli O103 belonging to other H-types than H2 have also been reported, such as E. coli O103:H11 in Japan (4) and + in Canada (5). A severe human outbreak was reported in Norway in 2006, caused by an stx 2 E. coli O103 with the rare H-type; H25 (6). Human cases of STEC O103:H25 have sporadically been reported (7-12), but in contrast to the Norwegian O103:H25 outbreak strain, these have been reported as stx + 1 (when stx properties have been given). Although several studies of E. coli O103 in animal have been conducted (reviewed in (13), most of these have been limited, either by using small sample sizes or by focusing only on the isolation of STEC O103 and not E. coli O103 in general. Also, most studies have been performed on material from cattle, and consequently the knowledge of aepec O103 in other potential animal reservoirs, such as sheep, is limited. Furthermore, possible occurrence of STEC O103 of other H-types than H2 is unknown as H-typing has not always been conducted. In the previously mentioned Norwegian outbreak, sheep were incriminated as the original reservoir for the outbreak strain of STEC O103:H25. This was because fermented sausages mainly consisting of mutton were shown to be the outbreak food source (6). In 3

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 addition, stx negative strains of E. coli O103:H25 were isolated both from fermented sausages, ovine meat and ovine fecal samples during the outbreak investigation. The aim of the present study was to describe the occurrence of E. coli O103 in Norwegian sheep, and to identify a possible reservoir of potentially human-pathogenic E. coli O103 such as STEC and EPEC. In addition, risk factors for flocks being positive for potentially human-pathogenic E. coli O103 were to be investigated. MATERIAL & METHODS Study design The study was designed as a cross sectional study with sheep flock as the unit. A total of 620 flocks that had at least 30 sheep older than 1 year were randomly selected from the Register of Production Subsidies (RPS). This register includes more than 95% of all commercial sheep flocks in Norway. One hundred flocks were selected from flocks registered in the RPS as of 01.01.2006 and 520 were selected from flocks registered as of 01.01.2007 as previously described (14). Sampling was conducted during the autumns of 2006 and 2007, respectively. From each flock, 50 single fecal samples from the youngest animals (lambs first, then oneyear olds etc.) were requested. A two-page questionnaire on flock characteristics and management factors was filled in at the time of sampling. Fecal samples Fecal samples were collected by digital rectal retrieval and transported in coolers to the laboratory where they arrived the following day. Approximately 2.5 to 5 g fecal material from 10 animals per flock were pooled to give a maximum of five samples per flock. If the number of samples from a single flock was not a multiple of 10, the last pooled sample consisted of samples from the remaining one to nine individual samples. The following analyses were 4

92 93 either started immediately upon arrival at the laboratory or the samples were frozen at -80 C until analyzed. 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 Detection and isolation of E. coli O103 The pooled samples were diluted 1:10 with 37 C pre-warmed buffered peptone water (Difco, Detroit, MI) and pre-enriched for 18 to 24 h at 41.5 ± 1 C. Detection of E. coli O103 was performed by automated immunomagnetic separation-enzyme-linked immunosorbent assay (AIMS-ELISA) and plating of ELISA positive samples onto MacConkey agar (Difco, Detroit, MI) with 4% cefixime-tellurite supplement (Dynal Invitrogen, Oslo, Norway) and washed sheep blood agar containing 10 mm CaCl 2 as previously described (14). Up to 10 colonies showing a typical E. coli appearance were tested by slide agglutination with E. coli O103 antiserum (Sifin, Berlin, Germany). Up to five of the positive colonies were subcultured onto blood agar plates before being tested further by slide agglutination to rule out autoagglutination. For species identification, the indole reaction, oxidase test and Rapid One test kit (Remel Inc., Atlanta, GA) were used. Serotyping and virulence characterization A conventional serotyping method with O103 antisera (Statens Serum Institut, Copenhagen, Denmark) was used to confirm presumptive E. coli O103 isolates. Confirmed E. coli O103 isolates were screened for the virulence genes stx 1, stx 2 and eae with a multiplex polymerase chain reaction (PCR) assay using 16S rrna gene as a control (15). Subtyping of stx was performed as described by Scheutz et al. (16). A selection of isolates was further examined for the virulence genes asta (17) and bfpa (18) using conventional PCR. The selection was based on the following criteria: one isolate from each pooled sample was selected (a maximum of five possible from each flock). 5

117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 If isolates from the same pooled sample had different virulence patterns as defined by PCR for eae and stx, then more than one isolate was further included, so that all virulotypes from a sample should be represented. Flagellar antigens were investigated by amplifying part of the flic gene as described by Machado et al. (19). Flagellar antigens H2 and H25 were identified by evaluation of CfoI restriction pattern of the amplicons as previously described (20). Other H-types were investigated by sequencing the PCR product, and designing specific primers for flic H16, flic H21 and flic H56 for use in PCR. One flic amplicon from each of the H-types H16, H21 and H56 were sequenced and the sequence confirmed by Basic local alignment search tool before used as positive controls in the PCRs for flic H16, flic H21 and flic H56. Boiled bacterial lysates were used as DNA template in all PCRs. Primers and positive controls for each PCR are listed in the supplemental material (Table S1). Pulsed-field gel electrophoresis PFGE was carried out using XbaI (Sigma, St. Louis, MO) restriction endonuclease as described in Sekse et al. (14). PFGE banding patterns were compared using a combination of visual inspection and the Bionumerics software program, version 6.1 (Applied Math NV, Ghent, Belgium). Dendrograms were generated using band-based Dice similarity coefficient and the unweighted pair group method using geometric average (UPGMA) with 1.1% position tolerance and 0.8% optimization. A cut-off level of 97% similarity was used to define a PFGE profile. Multi-dimensional scaling (MDS) was used to show a three-dimensional representation of the PFGE similarity matrix. The algorithm used calculating MDS in BioNumerics is a Principal Coordinates Analyses. Statistical analyses 6

142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 The crude county prevalences of E. coli O103 with various H-types and virulence profiles, and the corresponding 95% confidence interval (CI), were estimated assuming a binomial distribution by using the function binom.test in R software (R-2.15.3 for Windows; R Development Core Team [http://www.r-project.org/]). The relationship between the occurrence of eae + E. coli of serotypes O103:H2 and O103:H25 in a sheep flock and potential risk factors was analyzed using flock as the statistical unit and flocks having at least one positive sample for eae + E. coli O103:H2 or O103:H25, respectively, regarded as positive. Potential risk factors considered were; feed the last two weeks (concentrates, hey, silage, pasture), housing the last two weeks (outdoors, slatted floor, straw bedding, and non-slatted concrete or wooden floor), other animal species on the farm (cattle, goat, pig or poultry), sharing pasture with other animals (cattle, goat, other sheep flocks), purchase of animals within the last four months, being a member of a ram circle, flock size, and geographical region. All variables were initially run by univariable unconditional logistic regression analysis. Variables with Wald 2 p-values < 0.20 were selected for multivariable analysis. The multivariable logistic regression was performed by backward stepwise deletion of variables with p-values > 0.05. For each step the single least significant term was removed until there was no significant difference between the full and the reduced models. Adjusted odds ratio estimate was used as a measure of association between the response variable and the explanatory variable. The statistical analysis was performed using proc logistic in SAS 9.1.3 for Windows (SAS Institute Inc., Cary, NC). 7

162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 RESULTS Received samples Fecal samples from 592 (95.5%) of the selected 620 sheep flocks were received by the laboratory; 94 in 2006 and 498 in 2007. Samples from seven flocks were regarded unsuited for analysis, leaving samples from 585 flocks for further investigation. The exact number of 50 individual samples was received from 449 flocks (76.8%). For the remaining flocks, the number of samples varied from 9 to 51 with a mean of 47. There was a deviation of number of flocks examined from a few counties compared to what was planned, and although this may have created a bias, we do not expect it to have had any major effect on the final conclusions. E. coli O103 isolation, H-typing and virulence characteristics Of the 585 flocks, 185 (31.6%, 95% CI [27.9-35.6%]) were identified as positive for E. coli O103. A total of 1222 E. coli O103 were isolated from these flocks and subjected to PCR for eae and stx. The occurrence of positive ELISA results without a following isolation of E. coli O103 was only seen in a few cases and the freezing of samples was therefore not regarded to have had any major effect on the overall conclusions. Only five of the 1222 isolates were both eae + and stx +, and all contained stx 1 identified as subtype stx 1a. These originated from four flocks (0.7% [0.2-1.7 %]). Isolates from 52 flocks were characterized as eae + stx - E. coli O103 (8.9% [6.7-11.5%]). In addition, eae - stx - E. coli O103 was detected from 143 flocks (24.4% [21.0-28.1%], in which eae + stx - O103 isolates and eae + stx + E. coli O103 isolates had been detected in twelve and one of these flocks, respectively. A subset of 369 E. coli O103 isolates were further investigated for H-type, asta and bfpa, and subjected to PFGE. The eae + stx + isolates were all identified as E. coli O103:H2, 8

187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 while the eae + stx - E. coli O103 isolates possessed either the H2 or the H25 antigen. The eae - stx - E. coli O103 isolates belonged to the following H-types: H2, H6, H8, H14, H16, H19, H21, H38, H45 and H56. None of the isolates harbored the bfpa gene, while 30 of them possessed the asta gene. Only one of the asta + isolates also possessed eae, and this was an E. coli O103:H2. The remaining asta + isolates were negative for the other tested virulence genes and were of different H-types; H2, H38, H45 and H56. Table 1 shows an overview of serotypes and virulence genes identified among the 369 isolates. Occurrence of E. coli O103:H2 and E. coli O103:H25 The five eae + stx + E. coli O103:H2 isolates originated from four flocks (0.7% [0.2-1.7 %]) from different parts of Norway (Figure 1). From 18 flocks (3.1% [1.8-4.8%]), eae + stx - E. coli O103:H2 isolates were detected, i.e. 27 of the subset of 369 selected isolates. These positive flocks positive were scattered around in Norway (Figure 1). The majority of the E. coli O103:H2 (170/369) were eae - stx - and originated from 84 flocks (14.4% [11.6-17.5%]), giving a total of 103 flocks positive for E. coli O103:H2 (i.e. independent of detected virulence genes). Table 2 shows the number of positive flocks per region and county. As shown in Table 2, no stx + E. coli O103:H25 were detected in the survey (0% [0-0.6 %]). However, all the E. coli O103:H25 isolates detected were identified as eae + and originated from 34 flocks giving a flock level prevalence of 5.8% [4.1-8.0 %]. The 34 flocks were located in ten different counties (Table 2 and Figure 1). Molecular typing of E. coli O103 by PFGE Of the 369 E. coli O103 isolates, 349 were genotyped by PFGE, while the remaining 20 isolates (one of serotype O103:H38, one of O103:H45 and 18 of O103:H2) were nontypeable using XbaI restriction endonuclease due to DNA degradation. A total of 202 PFGE 9

212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 profiles were detected among the 349 E. coli O103 isolates. With a few exceptions, isolates of the same H-type and virulence profile from the same flock usually produced identical banding patterns. For instance, PFGE comparison of E. coli O103:H2 from 16 flocks showed minor differences (one to three band differences) between the isolates in ten of these flocks, while the six remaining flocks had isolates with more distinct banding patterns. PFGE comparison of E. coli O103:H25 from one flock showed two band differences in two isolates, while a third isolate produced a more distinct banding pattern compared to the other two isolates. On the other hand, identical PFGE profiles were identified for isolates from different flocks; nine occasions for E. coli O103:H2, one occasion for E. coli O103:H25, five occasions for E. coli O103:H16, two occasions for E. coli O103:H21 and one for E. coli O103:H14. The number of PFGE profiles per serotype is shown in Table 1, and a dendrogram showing PFGE profiles of all 349 isolates is included as a supplemental file. Figure 2 shows a multi-dimensional scaling of the similarity matrix of the 349 PFGE profiles. Risk factors for eae + E. coli O103 No risk factors were found to be associated with the occurrence of sheep flocks with animals positive for eae + E. coli O103:H2 (stx + and stx - ) in the final multivariable assessment. When considering eae + E. coli O103:H25, "Sharing of pasture with own cattle" was found to decrease the occurrence of eae + E. coli O103:H25 (Odds ratio 9.3 [1.2-68.8], p-value = 0.03), while "Sharing of pasture with goats owned by others" was found to increase the occurrence (Odds ratio 0.29 [0.09-0.93], p-value = 0.04). The number of flocks positive for eae + stx + E. coli O103:H2 was not sufficient to perform statistical analyses on possible risk factors for the occurrence of this virulotype. DISCUSSION 10

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 The present survey documented that there is a low prevalence of eae + stx + E. coli O103:H2 in the Norwegian sheep population. This O103 virulotype is well known to be a human pathogen. Moreover, the study did not detect any eae + stx + E. coli O103:H25. However, the relatively high prevalence of eae + stx - E. coli O103 (H2 and H25) is of concern as these may be precursors for STEC. Also, the survey shows that eae - stx - E. coli O103 is a very common and diverse E. coli serogroup in sheep with as much as 27.5% positive flocks. As previously described (14), this study consists of a representative sampling of a high number of sheep flocks from all parts of Norway and, to the authors knowledge, no other countries have conducted comparable national surveys in sheep. One recent Scottish study has, however, investigated sheep presented for slaughter in four Scottish abattoirs (21), giving individual sheep prevalences. Neither this study, nor two previous Norwegian studies that examined for all possible O-groups of stx + E. coli among sheep (22, 23), managed to detect any stx + E. coli O103. In contrast, four eae + stx + E. coli O103:H2/H- were detected among 60 ovine STEC investigated strains from Switzerland (24). However, other studies have also failed in detecting eae + stx + E. coli O103 from sheep (25-32), supporting the results of the present study suggesting that O103 may be an uncommon or low prevalent eae + STEC serogroup in this species. Two studies has been performed in cattle farms in Norway without detecting any eae + stx + E. coli O103 (33, 34), although one of them detected eae - stx + E. coli O103 in 3.2% of the herds. These studies were performed more than 10 years ago and new studies should be performed to see whether the eae + stx + E. coli O103 is low prevalent in Norwegian cattle herds as well. The finding of eae + stx - E. coli O103 in ruminants has occasionally been reported (25, 35, 36), but knowledge of the prevalence of eae + stx - E. coli O103 in the ruminant reservoir is sparse. Evans et al. (21) found only 0.6% of the tested sheep positive for eae + stx - E. coli 11

262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 O103. The present study, however, documents a relatively high flock occurrence of eae + stx - E. coli O103 in Norwegian sheep, i.e. of the two serotypes E. coli O103:H2 and E. coli O103:H25. Similar to the occurrence of E. coli O26:H11 (14), the occurrence of eae + stx - E. coli O103:H25 was higher in middle Norway (10.9%) than in the other regions (south east: 5.1%, west: 4.4% and north: 4.4%). However, in contrast to the occurrence of E. coli O26:H11, the factor region was not significantly associated to the occurrence of neither eae + stx - E. coli O103:H25 nor eae + O103:H2 (stx + and stx - ) in the statistical analyses. In contrast to the results in sheep in Norway, the two previous studies in cattle detected only a few eae + stx - E. coli O103 isolates (33, 34), indicating that eae + stx - E. coli O103 may be less common in Norwegian cattle than in sheep. However, these studies were not as extensive as the current study and caution should be taken making comparisons to these results. The eae + stx - E. coli O103 sheep reservoir is genetically diverse as demonstrated by the number of PFGE profiles found. Identical PFGE profiles were identified from a few flocks, both from contiguous counties and from counties not sharing any border. The authors have no explanation on how a common isolate could have been spread as there is no known epidemiological link between these farms. Feed concentrates used in Norway is heat treated and not considered a risk for spreading E. coli. Also, transport of sheep is not allowed between counties, though contacts between sheep of neighboring counties cannot be excluded as it is common to have the sheep free grazing on common pasture areas. This was, however, not a risk factor found to be associated in the statistical analyses. In the final multivariable assessment, "Sharing of pasture with own cattle" was found to decrease the occurrence and "Sharing of pasture with goats owned by others" was found to increase the occurrence, of eae + stx - E. coli O103:H25. We find it, however, difficult to explain these factors biologically. Analyzing the data with Sharing pasture with cattle and Sharing pasture with goats 12

286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 regardless of owner as possible risk factors, did not result in any statistical association (data not shown). Further investigation would be needed to investigate this aspect thoroughly. During the 2006-outbreak of E. coli O103:H25 in Norway, only eae + stx - E. coli O103:H25 was isolated from food and sheep (6, 37), and it was argued that the detected eae + stx - E. coli O103:H25 probably had lost their stx genes during cultivation. This has also been reported for other STEC (38-40). In the present study, up to five pooled samples per flock of a total of 585 flocks, and several isolates per positive pooled sample were investigated without detecting any stx + of E. coli O103:H25. We consider it unlikely that all the isolated eae + stx - E. coli O103:H25 should have lost their stx genes during enrichment. The findings of eae + stx - E. coli O103:H25 in sheep have also been reported by others (41, 42), while there to the authors knowledge have not been reported any stx + of E. coli O103:H25 from the animal reservoir. However, although not detected in the present nationwide study, it cannot be excluded that a small reservoir of stx + E. coli O103:H25 do exist among sheep in Norway. Whether the detected sheep reservoir of eae + stx - E. coli O103:H2 and O103:H25 have a potential as human pathogens by themselves, is not clear. A previous study from the county Sør-Trøndelag investigating EPEC in children under the age of five (with and without diarrhea), did not report any findings of EPEC O103:H2 or O103:H25 (43). Although in the present study there was a tendency of clustering towards serotypes, the PFGE patterns of both the E. coli O103:H2 and the E. coli O103:H25 were relatively diverse. Beutin et al. (44) compared EPEC and STEC O103:H2 from humans, animals and meat originating from several countries, and found the PFGE patterns of E. coli O103:H2 to be diverse as well. The number of isolates from animals and meat were, however, limited. Also, they report STEC O103:H2, including two EPEC isolates which they conclude to be precursors of STEC, to carry the subvariant eae-, while the ones they call EPEC O103:H2 usually carried the eae- subvariant. This has also been reported in a previous study by Oswald et al. (45). In the 13

311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 present study, subtyping of eae has not been performed. However, 51 of the isolates have been subtyped as part of another study in our laboratory (L. N. Nesse, C. Sekse, K. Berg, K. C. S. Johannesen, H. Solheim, L. Vestby and A. M. Urdahl, unpublished data), indicating that ovine stx - eae + O103:H2 carry the eae- subtype. This is in contrast to the findings of Beutin et al. (44) and Oswald et al. (45), but their studies included only a limited numbers of EPEC O103 and not any ovine isolates. Whether this finding of the eae- subtypes in stx- O103:H2 is of importance for the isolates human pathogenicity, is unknown. The previously mentioned study by Nesse and co-workers, indicate that the ovine E. coli O103:H25 in the present study have the eae- subtype, as also reported by others for both ovine and human isolates (7, 41, 42). However, preliminary comparison of PFGE patterns of the ovine E. coli O103:H25 isolates to human E. coli O103:H25 isolates from the Norwegian 2006 outbreak indicates that only a few of the ovine isolates cluster closely with the human isolates (data not shown). A more thorough genomic comparison, including PFGE data, to investigate the association between ovine and human strains of E. coli O103:H25 is in progress. The detected reservoir of eae + stx - E. coli O103:H2/O103:H25 may, however, be a concern by itself in the context of being precursors for stx positives. Sekse et al. (46) showed that isolates of eae + stx - E. coli O103:H25 may have potential to be infected by stx bacteriophages and become stx +. Recently, L Abée-Lund et al. (47) found that stx - and eae + E. coli O103:H25 isolates from both patients and food carried an stx negative bacteriophage in the same insertion site as the Stx2-encoding bacteriophage of the human eae + stx + E. coli O103:H25 2006 outbreak strain. Whether this is the case for the eae + stx - E. coli O103:H25 from the sheep reservoir as well and whether this would influence the ability to acquire stx bacteriophages, needs to be further investigated. The present survey focused on E. coli O103 in general, giving results on eae - stx - isolates in addition to aepec and STEC. Although there was a tendency of clustering towards 14

336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 serotypes as visualized by the multi-dimensional scaling, the E. coli O103 serogroup was genetically diverse as demonstrated by H-type, virulotype and PFGE banding pattern. The high prevalence of eae - stx - E. coli O103 including both H2 and other H-types, in addition to 3.1% and 5.8% of eae + stx - E. coli O103:H2 and E. coli O103:H25 positive flocks, respectively, emphasizes the importance of conducting both virulence characteristics and H- typing, and not concluding from detection of O-group by itself. This is in concurrence with the findings of Beutin et al. (44) that from virulence characterization, PFGE analyses and molecular typing of housekeeping genes conclude that the O serogroup as such cannot be taken as an indicator of genetic relationship between E. coli isolates. MDS is not commonly used to visualize the similarity among PFGE profiles, but is popular for analyzing other fingerprinting methods. The authors find MDS especially useful and illustrative for presenting a rather large and diverse PFGE dataset as in the present study. Compared to the traditional dendrogram that expands over several pages, MDS can be presented in one small figure making it easier to see the whole picture. Previous to our survey, there was sparse knowledge on the occurrence and distribution of E. coli O103 in Norwegian sheep flocks. Documentation of the prevalence of potentially human-pathogenic E. coli O103 is of importance for risk assessments. The finding of such relatively high occurrence of eae + stx - E. coli O103 in sheep flocks all over Norway was surprising and the pathogenic potential of these isolates needs to be further investigated by comparison with human isolates. 15

356 357 358 359 360 361 362 ACKNOWLEDGEMENT The authors thank personnel at the Norwegian Food Safety Authority for collecting the samples, Rune Pedersen, Trine Grønbeck, Tone Mathisen Fagereng, Anne Kari Kind and Heidi Solheim at the Norwegian Veterinary Institute for excellent technical assistance, and Attila Tarpai at the Norwegian Veterinary Institute for help with the geographical presentation. This work was supported by the Norwegian Food Safety Authority and by grant 178161/110 from the Research Council of Norway. Downloaded from http://aem.asm.org/ on July 9, 2017 by guest 16

363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 REFERENCES 1. Nataro JP, Kaper JB. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142-201. 2. Kaper JB, Nataro JP, Mobley HL. 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2:123-140. 3. Wong AR, Pearson JS, Bright MD, Munera D, Robinson KS, Lee SF, Frankel G, Hartland EL. 2011. Enteropathogenic and enterohaemorrhagic Escherichia coli: even more subversive elements. Mol. Microbiol. 80:1420-1438. 4. Seto K, Taguchi M, Kobayashi K, Kozaki S. 2007. Biochemical and molecular characterization of minor serogroups of Shiga toxin-producing Escherichia coli isolated from humans in Osaka prefecture. J. Vet. Med. Sci. 69:1215-1222. 5. Spika JS, Michel P, Milley D, Wilson J, Waters J. 1998. Shiga toxin-producing Escherichia coli infections in Canada, p. 23-29. In Kaper JB, O'Brien AD (ed.), Escherichia coli O157:H7 and other Shiga toxin-producing E. coli strains. ASM Press, Washington DC. 6. Schimmer B, Nygard K, Eriksen HM, Lassen J, Lindstedt BA, Brandal LT, Kapperud G, Aavitsland P. 2008. Outbreak of haemolytic uraemic syndrome in Norway caused by stx2-positive Escherichia coli O103:H25 traced to cured mutton sausages. BMC Infect. Dis 8:41. doi:10.1186/1471-2334-8-41 7. Iguchi A, Iyoda S, Ohnishi M. 2012. Molecular Characterization Reveals Three Distinct Clonal Groups among Clinical Shiga Toxin-Producing Escherichia coli of Serogroup O103. J. Clin. Microbiol. 50:7. 8. Couturier MR, Lee B, Zelyas N, Chui L. 2011. Shiga-toxigenic Escherichia coli detection in stool samples screened for viral gastroenteritis in Alberta, Canada. J. Clin. Microbiol 49:574-578. 17

388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 9. Ramotar K, Henderson E, Szumski R, Louie TJ. 1995. Impact of free verotoxin testing on epidemiology of diarrhea caused by verotoxin-producing Escherichia coli. J. Clin. Microbiol. 33:1114-1120. 10. Rivas M, Miliwebsky E, Chinen I, Roldan CD, Balbi L, Garcia B, Fiorilli G, Sosa-Estani S, Kincaid J, Rangel J, Griffin PM. 2006. Characterization and epidemiologic subtyping of Shiga toxin-producing Escherichia coli strains isolated from hemolytic uremic syndrome and diarrhea cases in Argentina. Foodborne Pathog. Dis. 3:88-96. 11. Thompson LH, Giercke S, Beaudoin C, Woodward D, Wylie JL. 2005. Enhanced surveillance of non-o157 verotoxin-producing Escherichia coli in human stool samples from Manitoba. Can. J. Infect. Dis. Med. Microbiol. 16:329-334. 12. Mackenzie AM, Lebel P, Orrbine E, Rowe PC, Hyde L, Chan F, Johnson W, McLaine PN. 1998. Sensitivities and specificities of premier E. coli O157 and premier EHEC enzyme immunoassays for diagnosis of infection with verotxin (Shigalike toxin)-producing Escherichia coli. The SYNSORB Pk Study investigators. J. Clin. Microbiol. 36:1608-1611. 13. Caprioli A, Morabito S, Brugereb H, Oswald E. 2005. Enterohaemorrhagic Escherichia coli: emerging issues on virulence and modes of transmission. Vet. Res. 36:289-311. 14. Sekse C, Sunde M, Lindstedt BA, Hopp P, Bruheim T, Cudjoe K, Kvitle B, Urdahl AM. 2011. Potentially human-pathogenic Escherichia coli O26 in Norwegian sheep flocks. Appl. Environ. Microbiol. 77:4949-4958. 15. Brandal LT, Lindstedt BA, Aas L, Stavnes TL, Lassen J, Kapperud G. 2007. Octaplex PCR and fluorescence-based capillary electrophoresis for identification of 18

412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 human diarrheagenic Escherichia coli and Shigella spp. J. Microbiol. Methods 68:331-341. 16. Scheutz F, Teel LD, Beutin L, Pierard 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. 17. Yamamoto T, Nakazawa M. 1997. Detection and sequences of the enteroaggregative Escherichia coli heat-stable enterotoxin 1 gene in enterotoxigenic E. coli strains isolated from piglets and calves with diarrhea. J. Clin. Microbiol. 35:223-227. 18. Gunzburg ST, Tornieporth NG, Riley LW. 1995. Identification of enteropathogenic Escherichia coli by PCR-based detection of the bundle-forming pilus gene. J. Clin. Microbiol. 33:1375-1377. 19. Machado J, Grimont F, Grimont PA. 2000. Identification of Escherichia coli flagellar types by restriction of the amplified flic gene. Res. Microbiol. 151:535-546. 20. Prager R, Strutz U, Fruth A, Tschape H. 2003. Subtyping of pathogenic Escherichia coli strains using flagellar (H)-antigens: serotyping versus flic polymorphisms. Int. J. Med. Microbiol. 292:477-486. 21. Evans J, Knight H, McKendrick IJ, Stevenson H, Varo BA, Gunn GJ, Low JC. 2011. Prevalence of Escherichia coli O157: H7 and serogroups O26, O103, O111 and O145 in sheep presented for slaughter in Scotland. J. Med. Microbiol. 60:653-660. 22. Urdahl AM, Beutin L, Skjerve E, Wasteson Y. 2002. Serotypes and virulence factors of shiga toxin-producing Echerichia coli isolated from healthy Norwegian sheep. J. Appl. Microbiol. 93:1026-1033. 19

436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 23. Urdahl AM, Beutin L, Skjerve E, Zimmermann S, Wasteson Y. 2003. Animal-host associated differences in shiga toxin-producing Escherichia coli isolated from sheep and cattle on the same farm. J. Appl. Microbiol. 95:92-101. 24. Zweifel C, Blanco JE, Blanco M, Blanco J, Stephan R. 2004. Serotypes and virulence genes of ovine non-o157 Shiga toxin-producing Escherichia coli in Switzerland. Int. J. Food Microbiol. 95:19-27. 25. Aktan I, Sprigings KA, La Ragione RM, Faulkner LM, Paiba GA, Woodward MJ. 2004. Characterisation of attaching-effacing Escherichia coli isolated from animals at slaughter in England and Wales. Vet. Microbiol 102:43-53. 26. Beutin L, Geier D, Steinrück H, Zimmermann S, Scheutz F. 1993. Prevalence and some properties of verotoxin (Shiga-like toxin)- producing Escherichia coli in seven different species of healthy domestic animals. J. Clin. Microbiol. 31:2483-2488. 27. Beutin L, Geier D, Zimmermann S, Aleksic S, Gillespie HA, Whittam TS. 1997. Epidemiological relatedness and clonal types of natural populations of Escherichia coli strains producing Shiga toxins in separate populations of cattle and sheep. Appl. Environ. Microbiol. 63:2175-2180. 28. Bettelheim KA, Bensink JC, Tambunan HS. 2000. Serotypes of Verotoxinproducing (Shiga toxin-producing) Escherichia coli isolated from healthy sheep. Comp. Immunol. Microbiol. Infect. Dis. 23:1-7. 29. Blanco M, Blanco JE, Mora A, Rey J, Alonso JM, Hermoso M, Hermoso J, Alonso MP, Dahbi G, González EA, Bernárdez MI, Blanco J. 2003. Serotypes, virulence genes, and intimin types of Shiga toxin (verotoxin)-producing Escherichia coli isolates from healthy sheep in Spain. J. Clin. Microbiol. 41:1351-1356. 20

459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 30. Kudva IT, Hatfield PG, Hovde CJ. 1997. Characterization of Escherichia coli O157:H7 and other Shiga toxin-producing E. coli serotypes isolated from sheep. J. Clin. Microbiol. 35:892-899. 31. Rey J, Blanco JE, Blanco M, Mora A, Dahbi G, Alonso JM, Hermoso M, Hermoso J, Alonso MP, Usera MA, Gonzalez EA, Bernardez MI, Blanco J. 2003. Serotypes, phage types and virulence genes of Shiga-producing Escherichia coli isolated from sheep in Spain. Vet. Microbiol. 94:47-56. 32. Vettorato MP, de Castro AF, Cergole-Novella MC, Camargo FL, Irino K, Guth BE. 2009. Shiga toxin-producing Escherichia coli and atypical enteropathogenic Escherichia coli strains isolated from healthy sheep of different populations in Sao Paulo, Brazil. Lett. Appl. Microbiol. 49:53-59. 33. Hofshagen M, Nygård K, Hauge K. 2006. Zoonoserapporten 2005 ISSN 1502-5713. 34. Opheim M, Hofshagen M, Wasteson Y, Bruheim T, Kruse H. 2003. Abstr. 5 th International Symposium on "Shiga toxin (Verocytotoxin)-producing E. coli infections",, abstr P100. Faecal carriage of Shigatoxin-Producing Escherichia coli O26, O103, O111, O145 and O157 in Norwegian Beef Cattle. 128. 35. Blanco M, Schumacher S, Tasara T, Zweifel C, Blanco JE, Dahbi G, Blanco J, Stephan R. 2005. Serotypes, intimin variants and other virulence factors of eae positive Escherichia coli strains isolated from healthy cattle in Switzerland. Identification of a new intimin variant gene (eae-eta2). BMC Microbiol. 5:23. 36. Pearce MC, Evans J, McKendrick IJ, Smith AW, Knight HI, Mellor DJ, Woolhouse MEJ, Gunn GJ, Low JC. 2006. Prevalence and virulence factors of Escherichia coli serogroups O26, O103, O111, and O145 shed by cattle in Scotland. Appl. Environ. Microbiol. 72:653-659. 21

483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 37. Sekse C, O'sullivan K, Granum PE, Rørvik LM, Wasteson Y, Jørgensen HJ. 2009. An outbreak of Escherichia coli O103:H25 - bacteriological investigations and genotyping of isolates from food. Int. J. Food Microbiol. 133:259-264. 38. Karch H, Meyer T, Russmann H, Heesemann J. 1992. Frequent loss of Shiga-like toxin genes in clinical isolates of Escherichia coli upon subcultivation. Infect. Immun. 60:3464-3467. 39. Bielaszewska M, Prager R, Kock R, Mellmann A, Zhang W, Tschape H, Tarr PI, Karch H. 2007. Shiga toxin gene loss and transfer in vitro and in vivo during enterohemorrhagic Escherichia coli O26 infection in humans. Appl. Environ. Microbiol. 73:3144-3150. 40. Joris MA, Pierard D, De ZL. 2011. Occurrence and virulence patterns of E. coli O26, O103, O111 and O145 in slaughter cattle. Vet. Microbiol. 151:418-421. 41. Krause G, Zimmermann S, Beutin L. 2005. Investigation of domestic animals and pets as a reservoir for intimin- (eae) gene positive Escherichia coli types. Vet. Microbiol. 106:87-95. 42. Frohlicher E, Krause G, Zweifel C, Beutin L, Stephan R. 2008. Characterization of attaching and effacing Escherichia coli (AEEC) isolated from pigs and sheep. BMC Microbiol. 8:144. 43. Afset JE, Anderssen E, Bruant G, Harel J, Wieler L, Bergh K. 2008. Phylogenetic backgrounds and virulence profiles of atypical enteropathogenic Escherichia coli strains from a case-control study using multilocus sequence typing and DNA microarray analysis. J. Clin. Microbiol. 46:2280-2290. 44. Beutin L, Kaulfuss S, Herold S, Oswald E, Schmidt H. 2005. Genetic analysis of Enteropathogenic and Enterohemorrhagic Escherichia coli serogroup O103 strains by 22

507 508 509 510 511 512 513 514 515 516 517 518 molecular typing of virulence and housekeeping genes and pulsed-field gel electrophoresis. J. Clin. Microbiol. 43:1552-1563. 45. Oswald E, Schmidt H, Morabito S, Karch H, Marches O, Caprioli A. 2000. Typing of intimin genes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun. 68:64-71. 46. Sekse C, Muniesa M, Wasteson Y. 2008. Conserved Stx2 phages from Escherichia coli O103:H25 isolated from patients suffering from hemolytic uremic syndrome. Foodborne Pathog. Dis. 5:801-810. 47. L'Abee-Lund TM, Jørgensen HJ, O'sullivan K, Bohlin J, Ligard G, Granum PE, Lindback T. 2012. The highly virulent 2006 Norwegian EHEC O103:H25 outbreak strain is related to the 2011 German O104:H4 outbreak strain. PLoS One 7:e31413. Downloaded from http://aem.asm.org/ on July 9, 2017 by guest 23

519 520 521 522 523 524 525 526 527 528 529 530 531 Figure legends: FIG 1 Distribution of flocks positive for A) eae + E. coli O103:H2 and B) eae + E. coli O103:H25. Red dots marks flocks with stx + isolates. Geographical locations of the flocks are set to the centroid of the municipality. The regions, as used in Table 2, are shown by thick lines, and the letters designates the counties; O=Oslo, Ø=Østfold, A=Akershus, HE=Hedmark, OP=Oppland, BD=Buskerud, V=Vestfold, TK=Telemark, AA=Aust-Agder, VA=Vest-Agder, R=Rogaland, H=Hordaland, SF=Sogn og Fjordane, M=Møre og Romsdal, ST=Sør-Trøndelag, NT=Nord-Trøndelag, N=Nordland, T=Troms, F=Finmark. FIG 2 Multi-dimensional scaling of 349 pulsed-field gel electrophoresis profiles of ovine E. coli isolates found in a study of Norwegian sheep flocks in 2006 and 2007. The distance between two isolates reflects the similarity between them. The various serotypes are shown in different colours. Downloaded from http://aem.asm.org/ on July 9, 2017 by guest 24

533 534 TABLE 1 Virulence characteristics and serotypes of 369 ovine E. coli O103 isolates found in a study of Norwegian sheep flocks in 2006 and 2007 N o of N o of N o of PFGE Serotype isolates flocks stx 1 + stx 2 + eae + bfpa + asta + profiles a 535 536 537 538 539 540 O103:H2 eae - 170/202 84 b - - - - 10 75 O103:H2 eae + 32/202 22 b 5-32 - 1 c 22 O103:H6 7 6 - - - - - 7 O103:H8 1 1 - - - - - 1 O103:H14 5 3 - - - - - 2 O103:H16 36 28 - - - - - 21 O103:H19 1 1 - - - - - 1 O103:H21 32 22 - - - - - 19 O103:H25 58 34 - - 58 - - 36 O103:H38 6 2 - - - - 1 1 O103:H45 5 4 - - - - 4 3 O103:H56 16 12 - - - - 14 14 Total 369 185 d 5 0 90 0 30 202 a PFGE profiles among 349 isolates, 20 isolates (one of serotype O103:H38, one of O103:H45 and 18 of O103:H2) were nontypeable using XbaI restriction endonuclease b Both eae + and eae - isolates from three flocks c One eae + stx - isolate d N o of flocks E. coli O103 was detected from. Some flocks were positive for more than one serotype/virulotype, so this column adds up to 219 25

541 542 TABLE 2 Number of sheep flocks positive for E. coli O103 per Norwegian region and county. SE = South-Eastern, W = Western, M = Middle and N = Northern Norway. N o of flocks positive for E. coli O103 H2 H25, Other H- No. of eae+ eae- and eae+ and types, Region County flocks stx1+ stx- stx- stx- eae- and stx- Østfold 2 - - - - - Akershus 13-1 - - - Hedmark 20-2 1 - - Oppland 60-1 4 4 3 SE Buskerud 30-1 1 4 1 Vestfold 4 - - - - - Telemark 12 1 - - - 1 Aust-Agder 3 - - 1 - - Vest-Agder 14-1 6-3 Rogaland 114-1 45 5 30 Hordaland 71 1 2 7 4 7 W Sogn og 64-4 9 2 10 Fjordane Møre og 45 - - 3 7 7 Romsdal M Sør-Trøndelag 32-1 3 3 3 Nord- Trøndelag 33-1 - 2 1 N Nordland 41 2 3 3 1 6 26

Troms 21 - - 1 2 3 Finnmark 6 - - - - 2 Total 585 4 18 84 34 77 Prevalence (%) 0.7 3.1 14.4 5.8 13.2 543 95% CI 0.2-1.7 1.8-4.8 11.6 17.5 4.1 8.0 10.5 16-2 Downloaded from http://aem.asm.org/ on July 9, 2017 by guest 27