International Journal of Food Microbiology 174 (2014) 31–38

Contents lists available at ScienceDirect

International Journal of Food Microbiology

j ourna l homepage: www.e lsev ie r .com/ locate / i j foodmicro

Prevalence and molecular characterization of sorbitol fermenting andnon-fermenting Escherichia coli O157:H7+/H7– isolated from cattle atslaughterhouse and slaughterhouse wastewater

Naim Deniz Ayaz a,⁎, Yilmaz Emre Gencay a, Irfan Erol b

a Kirikkale University Faculty of Veterinary Medicine, Department of Food Hygiene and Technology, 71450 Yahsihan, Kirikkale, Turkeyb Republic of Turkey Ministry of Food Agriculture and Livestock, General Directorate of Food and Control, Lodumlu, Ankara, Turkey

⁎ Corresponding author. Tel.: +90 318 3573301; fax: +E-mail address: naimdenizayaz@kku.edu.tr (N.D. Ayaz

0168-1605/$ – see front matter © 2014 Elsevier B.V. All rhttp://dx.doi.org/10.1016/j.ijfoodmicro.2014.01.002

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 July 2013Received in revised form 31 December 2013Accepted 2 January 2014Available online 9 January 2014

Keywords:Sorbitol fermenting E. coli O157:H7CattleSlaughterhouse wastewaterShiga toxin variantsIntimin variants

The prevalence and seasonal distribution of E. coli O157:H7+/H7– in an array of aged cattle at slaughter and itsdissemination with slaughterhouse wastewater over a two year period in Turkey were investigated. For thispurpose, a total of 720 samples (240 rectoanal mucosal swap [RAMS], 240 carcass sponge and 240 bile samples)of 240 cattle categorized according to age, gender, breed and sampling site were collected along with additional24 wastewater samples and were subjected to immunomagnetic separation based cultivation technique to effi-ciently isolate E. coliO157 from the background flora. Identification (rfbEO157, fliCh7), detection ofmajor virulencefactors (stx1, stx2, eaeA, hly, lpfA1-3 and espA), intimin variants (eae-α1, eae-α2, eae-β, eae-β1, eae-β2, eae-γ1 andeae-γ2/θ) and shiga toxin variants (stx1c, stx1d, stx2c, stx2d, stx2e, stx2f and stx2g) of all the isolates were assessed byPCR. From10 (4.2%) of RAMS and 11 (4.6%) of carcass sponge samples and 5 (20.8%) of slaughterhousewastewa-ter samples, a total of 102 colonies (99 sorbitol negative and 3 sorbitol positive) were isolated. Overall, 17 (7.1%)and 15 (6.3%) of 240 sampled cattle were shown to harbor E. coli O157 and E. coliO157:H7, respectively either intheir RAMS or carcass sponge samples analyzed. Statistically significant differences between categories; season,age, gender and breed of cattle were not observed (p N 0.05). None of the isolated E. coli O157:H7+/H7– strainsharbored any of the investigated intimin types other than eaeγ1 or shiga toxin variants stx1d, stx2e, stx2f or stx2gwhile all were lpfA1-3+ except 5 E. coli O157:H7– strains. Intimin variant eaeγ1 and shiga toxin 1 variant stx1cwere detected from all of the eaeA+ (97/102, 95.1%) and stx1

+ (32/102, 31.3%) strains, respectively while fromstx2

+ (80/102, 78.4%) isolates, both stx2c (68/80, 85.0%) and stx2d (12/80, 15.0%) variants were determined. Inthe last decade, prevalence of E. coliO157:H7has an increasing trend in cattle. Slaughterhouses are the significantsources of environmental contamination with E. coli O157:H7. Isolation and molecular characterization of sorbi-tol fermenting E. coli O157:H7 are a novel finding and may lead to a revision of reference isolation procedure ofE. coli O157:H7 in future.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Escherichia coli O157:H7 has emerged as a pathogen of significantpublic health concern worldwide, and it is recognized as the majoretiologic agent of hemorrhagic colitis (HC) and hemolytic uremic syn-drome (HUS) in humans. Primarily cattle and other ruminants cancarry the agent in their gastrointestinal tracts as asymptomatic reser-voirswhile feces and hide are themajor sources of bacterial carcass con-tamination at slaughterhouse (Elder et al., 2000; Gun et al., 2003;McEvoy et al., 2000). Epidemiological studies indicated that the trans-mission of E. coliO157:H7occurs through consumption of contaminatedraw or undercooked meats of especially bovine origin such as mincedmeat and related products (Meng et al., 2001). In 2011–2012 alone,1.4 million tons of bovine meat originating from culture, crossbred

90 318 3573304.).

ights reserved.

and native cattle were produced in Turkey (TSI, 2013), but the publichealth risk arising from E. coli O157:H7 contamination is fairly unclear.

Age, gender, breed, weaning period, housing, transportation, feedcomposition of the cattle, season and strain specificity of E. coli O157:H7 were all identified as risk factors for carriage and infection of cattle(Callaway et al., 2009; Kulow et al., 2012). In previous studies, un-weaned calves (b4–8 months) generally showed higher prevalencethan adult cattle (N24 months) while adult cattle showed lower preva-lence of E. coli O157:H7 than young stock (12–24 months) as well.Higher prevalence was also observed in female than male cattle(Albihn et al., 2003; Nielsen et al., 2002; Rugbjerg et al., 2003; Yilmazet al., 2002; Van Donkersgoed et al., 1999). Some of the previousworks had speculated the possible influence of genetic difference of cat-tle on E. coli O157:H7 prevalence (Rugbjerg et al., 2003; Widiasih et al.,2003) while a more recent work showed the genetic and physiologicalinfluence of the animal in persistence of the pathogen in cattle (Jeonet al., 2013). In general the prevalence of E. coliO157:H7 is higher during

32 N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

summer and autumn than winter (Hussein and Bollinger, 2005) conse-quently resulting in higher rates of human infections in summer andautumn (Vugia et al., 2007).

Shiga toxins and intimin are considered as themost critical virulencefactors of E. coli O157:H7, involved in development of HUS andattaching-effacing (AE) lesions, respectively (Gyles, 2007). Accumulatedreports showed differences in clinical outcomes (Friedrich et al., 2002)and toxicities of shiga toxin variants on different cell cultures (Lefebvreet al., 2009) while intimin variants were shown to affect tissuetropism and consequently the colonization site (Fitzhenry et al., 2002;Mundy et al., 2007). Furthermore, additional factors such as; EspA(E. coli secreted protein A) and Lpf (Long polar fimbria)were determinedto be important in tropism, attachment, persistence and virulence (Abeet al., 1998; Farfan and Torres, 2011; Torres et al., 2009).

By the present study, we aimed to provide scientific data on the in-fluence of season, age, gender, breed and sampling site on the preva-lence of E. coli O157:H7+/H7– in an array of aged cattle at slaughterover a two years period in Kirikkale, Turkey. Additionally, prevalenceand seasonal distribution of slaughterhouse wastewater contaminationwere assessed for estimation of environmental contamination and theE. coli O157:H7 risk that might arise for public health. Furthermore,major and contributor virulence factors of sorbitol fermenting andnon-fermenting Escherichia coli O157:H7 isolates were investigated.

2. Materials and methods

2.1. Sampling design and sample collection

Between July 2011 and June 2013, within a two-year period, 744cattle and wastewater samples (240 rectoanal mucosal swap, 240 car-cass sponge, 240 bile and 24 wastewater samples) were analyzed forthe detection of E. coli O157 by immunomagnetic separation (IMS)based cultivation technique, which enables efficient isolation of thepathogen from the background flora, and PCR. For this purpose, during24months period,weekly visits to slaughterhousewere done to achievea total of 31 samples permonth. Samples for eachmonth consisted of 10rectoanal mucosal swaps (RAMS), 10 carcass sponges and 10 bile sam-ples of 10 different, randomly chosen cattle and one slaughterhousewastewater sample collected from a slaughterhouse serving to the

Table 1Distribution of E. coli O157:H7+/H7– positivity according to age, breed, gender and season in c

Agea n (%) Breed n (%) Genderb n (%

Y ≤ 24 months 158 (65.8) Culture 66 (27.5) M 64 (F 2 (0

Crossbred 55 (22.9) M 46 (F 9 (3

Native 37 (15.4) M 30 (F 7 (3

24 months b M ≤ 4 years 45 (18.8) Culture 7 (2.9) M 3 (1F 4 (1

Crossbred 23 (9.6) M 12 (F 11 (

Native 15 (6.3) M 8 (3F 7 (2

4 years b O 37 (15.4) Culture 15 (6.3) M 0 (0F 15 (

Crossbred 12 (5.0) M 0 (0F 12 (

Native 10 (4.1) M 0 (0F 10 (

a Y, young; M, mature; O, old.b M, male; F, female.c R, rectoanal mucosal swap; CS, carcass sponge; R/CS, both.d S, warm months (May–October); W, cold months (November–April).e One sample positive for E. coli O157:H7–.

region, located in Kirikkale, in the middle of Anatolia, near Ankara inTurkey. While samples were being collected, specifications of the cattlesuch as age (young [≤24 months], mature [N24 months, ≤4 years] orold [N4 years]), gender (male or female) and breed (native [AnatolianBlack and Turkish Grey steppe], culture [Holstein, Brown Swiss andSimmental] or crossbreds) were documented (Table 1) to compareand assess the influence of these parameters to the prevalence ofE. coli O157:H7. Since Turkey is in subtropical climate and continentalwarm to hot weather dominates through early May to late October inmiddle Anatolia, seasonal distribution was specified accordingly.

A total of 240 cattlewere sampled during this two year period. RAMSsamples were collected by sterile cotton swaps after bleeding and wereaseptically transferred to 3 ml of Tryptone Soya Broth (TSB, OxoidCM0129, Hampshire, UK) as essentially described (Rice et al., 2003). Atotal of minimum 2400 cm2 carcass sponge samples were obtainedfollowing evisceration using cellulose sponge-sticks (3 M, SSL100,Minnesota, USA) pre-moistened with 10 ml of buffered peptone water(BD Difco, New Jersey, USA). From each carcass, after evisceration aminimum of 400 cm2 from inside round area and a minimum of2000 cm2 from navel-plate-brisket-foreshank areas were sampledwith at least 5 passes of one side each of a single sponge-stick accordingto Arthur et al. (2004) and FSIS (1996). Ten ml of bile samples wasobtained from undamaged gallbladders of cattle following eviscerationof gastrointestinal tract, using sterile syringes (Goncuoglu et al., 2010).RAMS, carcass sponge and bile samples were coded with same numberfor each cattle.

A total of 24 slaughterhouse wastewater samples were collectedthroughout the sampling period of the study. Fifty ml of wastewatersamples was taken into sterile flask from the canal that leads wastewa-ter to Kizilirmak River. All of the collected RAMS, carcass sponge, bileand wastewater samples were kept in an ice bag and transported tolaboratory in less than two hours and analyzed immediately.

2.2. E. coli O157 isolation

Immunomagnetic separation (IMS) based cultivation techniquewasused for the isolation of E. coli O157 (Byrne et al., 2003). RAMS sampleswere incubated in 3 ml of TSB (Oxoid CM0129) at 37 °C for 18 h (NüveEN120, Ankara, Turkey) (Rice et al., 2003). Ninety ml EC broth (Oxoid

attle samples.

) E. coli O157:H7+/H7– positivity n (%) Positive samplec Seasond

26.6) 6 (2.5)e Re, CS, R, R, R, CS S, S, S, W, S, W.8)19.1) 2 (0.8) R, R W, W.8) 1 (0.4) R/CS S12.5) 2 (0.8) R/CS, CS S, W6.3)

Total: 11 (4.6).3).7) 1 (0.4)e CSe W5.0) 1 (0.4) R/CS S4.6).3).9) 1 (0.4) CS W

Total: 3 (1.3).0)6.3).0)5.0) 1 (0.4) R/CS W.0)4.1) 2 (0.8) CS, CS S, S

Total: 3 (1.3)

33N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

CM0853) supplemented with novobiocin (20 mg/l; Sigma N-1628, St.Louis, MO, USA) was added to each carcass sponge bags and 10 ml ofeach bile and wastewater samples that were transferred to sterilebags, was homogenized (Seward Laboratory, Labblender 400 stomach-er, London, UK) and incubated at 37 °C for 18 h (Jeong et al., 2007).Following the incubation period, IMSwas performedwith 1 ml of selec-tively enriched samples with 20 μl of anti-E. coli O157 Dynabeads(Invitrogen, Dynal AS, Dynabeads anti-E. coli O157, Cat no. 710–04,Oslo, Norway) according to themanufacturer's protocol usingmagneticparticle concentrator (Dynabeads MPC-S). After the IMS procedure,50 μl of resuspended IMSmixturewas plated on Cefixime-Tellurite sup-plemented Sorbitol MacConkey Agar (CT-SMAC, Oxoid CM0813; OxoidSR0172). After 24 h incubation at 42 °C, up to five suspected colonieswere tested for the O157 antigen by latex agglutination (E. coli LatexTest, Oxoid DR0620). Then, latex agglutination positive colonies werestored at −86 °C (Thermo Scientific, Revco Elite Plus 2586-6-v, USA)in cryovials (BHI broth supplemented with 20% glycerol) for PCRanalysis after ensuring their purity by consecutive streaks on CT-SMAC.

2.3. PCR analysis

For identification andvirulence genedetermination of E. coliO157 iso-lates DNA extraction was performed by Chelex-100 (Bio-Rad, Hercules,CA, USA) resin based technique, using proteinase K (20 mg/ml;

Table 2Primer pairs and PCR protocols used for identification and determination of virulence genes an

Gene Primers Nucleotide Sequence (5′–3′

rfbEO157 RfbE-a CTACAGGTGAAGGTGGAATGRfbE-b ATTCCTCTCTTTCCTCTGCGG

fliCh7 FLICH7-F GCGCTGTCGAGTTCTATCGAFLICH7-R CAACGGTGACTTTATCGCCAT

stx1 SLT1-F TGTAACTGGAAAGGTGGAGTSLT1-R GCTATTCTGAGTCAACGAAA

stx2 SLTII-F GTTTTTCTTCGGTATCCTATTCSLTII-R GATGCATCTCTGGTCATTGTA

eaeA AE22 ATTACCATCCACACAGACGGAE20-2 ACAGCGTGGTTGGATCAACC

hly MFS1-F ACGATGTGGTTTATTCTGGAMFS1-R CTTCACGTCACCATACATAT

lpfA1-3 LPFA1-CF GGTTGGTGACAAATCCCCGLPFA1-CR1 CGTCTGGCCTTTACTCAGA

espA EspAa CACGTCTTGAGGAAGTTTGGEspAb CCGTTGTTAATGTGAGTGCG

eae-α1 EAE-FB AAAACCGCGGAGATGACTTCEAE-A CACTCTTCGCATCTTGAGCT

eae-α2 IH2498aF AGACCTTAGGTACATTAAGTIH2498aR TCCTGAGAAGAGGGTAATC

eae-β EAE-FB AAAACCGCGGAGATGACTTCEAE-B CTTGATACACTTGATGACTGT

eae-β1 EA-B1-F CGCCACTTAATGCCAGCGEAE-B CTTGATACACCTGATGACTGT

eae-β2 EA-B2-F CCCGCCACTTAATCGCACGTEAE-B CTTGATACACCTGATGACTGT

eae-γ1 EAE-FB AAAACCGCGGAGATGACTTCEAE-C1 AGAACGCTGCTCACTAGATG

eae-γ2/θ EAE-FB AAAACCGCGGAGATGACTTCEAE-C2 CTGATATTTTATCAGCTTCA

stx1c Stx1c-1 TTTTCACATGTTACCTTTCCTStx1c-2 CATAGAAGGAAACTCATTAG

stx1d VT1AvarF CTTTTCAGTTAATGCGATTGCVT1AvarR AACCCCATGATATCGACTGC

stx2c Stx2c-a GCGGTTTTATTTGCATTAGTStx2c-b AGTACTCTTTTCCGGCCACT

stx2d Stx2d-a GGTAAAATTGAGTTCTCTAAStx2d-b CAGCAAATCCTGAACCTGAC

stx2e Stx2e-a ATGAAGTGTATATTGTTAAAStx2e-b AGCCACATATAAATTATTTCG

stx2f Stx2f-a TGTCTTCAGCATCTTATGCAGStx2f-b CATGATTAATTACTGAAACAG

stx2g 209 F GTT ATATTTCTGTGGATATC781R GAATAACCGCTACAGTA

AppliChem GmbH, Darmstadt, Germany), thermo-shaker (ALLS, Msc-100, China) and centrifuge (Beckman Coulter 22R Centrifuge, CA, USA)as essentially described (Goncuoglu et al., 2010).

All of the primer pairs (EurofinsMWGOperon, Ebersberg, Germany)and PCR protocols used in identification and virulence gene determina-tion of E. coli O157 isolates were given in Table 2. Ten μl aliquot of eachresultant PCR product was further analyzed by agarose gel (1.5%Agarose-Basica LE, Prona, Spain) electrophoresis (CSL MSMixi-Duo,Corston, UK), stained with 0.1 μg/ml ethidium bromide (BioChemicaGmbH, Darmstadt, Germany), at 100 V for 1 h and visualized by a geldocumentation and analysis system (SyngeneIngenius, Cambridge,UK) (Fig. 1).

2.3.1. Verification of E. coli O157 isolates and detection of H7 gene by PCRLatex agglutination positive (E. coli O157 Latex Test) colonies were

tested for verification and identification of E. coli O157:H7 by the detec-tion of rfbEO157 gene and flagellar H7 (fliCh7) gene with RfbE-a/RfbE-b(Wang et al., 2002) and FLICH7-F and FLICH7-R (Fratamico et al.,2000) primer pairs by PCR assay, respectively. E. coli O157:H7 ATCC43895 (rfbEO157+ , fliCh7+) was used as positive control.

2.3.2. Detection of virulence genes by multiplex and conventional PCRVerified and identified E. coli O157:H7 colonies were subjected to

multiplex or conventional PCR for detection of various virulence genes

d gene variants of E. coli O157:H7+/H7–.

) Size (bp) Reference [PCR protocol]

G 327 Wang et al. (2002)

GC 625 Fratamico et al. (2000)TCCATACA 210 Fratamico et al. (2000)AATAACC 484 Fratamico et al. (2000)TTACT 397 Fratamico et al. (2000)T

166 Fratamico et al. (2000)

244 Torres et al. (2009)

299 McNally et al. (2001)

820 Blanco et al. (2004)

AAGC 517 Blanco et al. (2004)

830 Blanco et al. (2004)

811 Blanco et al. (2004)

807 Blanco et al. (2004)

804 Blanco et al. (2004)TC

808 Blanco et al. (2004)

498 Zhang et al. (2002)GT 192 Bürk et al. (2003)

124 Wang et al. (2002)[Osek, 2003]

GTAT 175 Wang et al. (2002)[Osek, 2003]G

GTGGA 303 Imberechts et al. (1992)[Osek, 2003]T

150 Wang et al. (2002)[Osek, 2003]AAAC

573 Leung et al. (2003)

Fig. 1. Agarose gel electrophoresis of virulence gene DNA fragments amplified by PCR from selected positive controls. LaneM: 100 bp DNA marker; Lanes 1 and 2: E. coli O157:H7 ATCC43895 (rfbEO157+,fliCh7

+); Lane 3: E. coli O157:H7 ATCC 43895 (stx1+, stx2

+, eaeA+, hly+); Lane 4: E. coli O157:H7 NCTC 12900 (stx1−, stx2

−, eaeA+, hly+, lpfA1-3+, espA+); Lane 5: E. coli O157:H7 isolate (stx1

−, stx2+, eaeA+, hly+); Lane6:Negative control; Lanes 7, 8 and 9: E. coli O157:H7ATCC43895 (eaeγ1

+,espA+, lpfA1-3+); Lane 10: E. coli O157:H7ATCC43895 (stx1c+); Lane 11:E. coli O157:NM 137/98 (stx2c

+); Lane 12: E. coli O62:H– 551/98 (stx2d+); Lane 13: E. coli O139:K12 107/86 (stx2e

+): Lane 14: E. coli O–:H18 214/125 (stx2f+).

34 N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

such as; stx1, stx2, eaeA, hly (Fratamico et al., 2000), espA (McNally et al.,2001) and lpfA1-3 (Torres et al., 2009) genes. E. coli O157:H7 ATCC43895 (stx1+, stx2+, eaeA+, hly+, lpfA1-3+, espA+) was used as positivecontrol for all investigated genes while E. coli O157:H7 NCTC 12900(stx1−, stx2−, eaeA+, hly+, lpfA1-3+, espA+) as negative control forshiga toxin genes.

2.3.3. Detection of intimin (eae) variants of E. coli O157:H7+/H7−eaepositive isolates

Intimin variants α1, α2, β, β1, β2, γ1 and γ2/θ were tested ineaeA gene detected intimin harboring isolates by previously publishedprimer pairs and PCR conditions (Blanco et al., 2004) (Table 2). E. coliO157:H7 ATCC 43895 (eaeγ1

+) was used as positive control.

2.3.4. Detection of stx1 and stx2 variants from shiga toxigenic E. coli O157:H7 isolates

Shiga toxigenic E. coli O157:H7 isolates (stx1+ and/or stx2+) weresubjected to consecutive multiplex and conventional PCR assays for de-termination of stx1 variants (stx1c (Zhang et al., 2002) and stx1d (Bürket al., 2003)) and/or stx2 variants (stx2c, stx2d, stx2e, stx2f (Osek, 2003)and stx2g (Leung et al., 2003)) (Table 2). E. coli O157:H7 ATCC 43895(stx1c+), strains E. coli O157:NM 137/98 (stx2c+), E. coli O62:H– 551/98(stx2d+), E. coli O139:K12 107/86 (stx2e+), E. coli O–:H18 214/125 (stx2f+)kindly provided by J. Osek in the Department of Microbiology, NationalVeterinary Research Institute, Pulawy, Poland; and E. coli O2:H25 S86(stx2g+) kindly provided by C. Garcia-Aljaro in Faculty of Biology,Department of Microbiology, University of Barcelona, Barcelona, Spainwere used as positive controls while E. coli O157:H7 NCTC 12900 (stx1−

and stx2−) as negative control for shiga toxin genes.

2.4. Phenotypic confirmation of H7 phenotype and motility in sorbitolfermenting (SF) E. coli O157:H7 isolates

For confirmation of H7 phenotype in SF E. coliO157:H7, isolates weresubjected to serological tube agglutination test with commercially avail-able E. coli Antiserum H7 (BD Difco, Cat. No. 221591, USA) according tothe manufacturer's protocol. Agglutination with a clear supernatant

was considered as positive. Motility in SF E. coli O157:H7 isolates wasdetermined with inoculation of 2 μl of fresh cultures (OD600: 1.0) on LBplates containing 0.25% agar in triplicate. After incubation at 37 °Cfor 6 h, swarming zones were measured and compared with motilitypositive (E. coli O157:H7 NCTC 12900) and negative (E. faecalis ATCC29212) reference strains.

2.5. Statistical analysis

The statistical analysis for determination of the significance of age,gender, breed and season on the prevalence of E. coli O157:H7 in cattlewas performed with chi-square and binary logistic regression (SPSS,version 16.0).

3. Results

3.1. Isolation and identification of E. coli O157:H7 from cattle andwastewater samples

A total of 744 samples including, 240 of each RAMS, carcass spongeand bile samples belonging to 240 cattle and 24 slaughterhouse waste-water samples were analyzed for the detection of E. coliO157:H7 over atwo year period. From 10 (4.2%) of 240 RAMS, 11 (4.6%) of 240 carcasssponges and 5 (20.8%) of 24 slaughterhousewastewater samples, a totalof 102 colonies (99 sorbitol negative and 3 sorbitol positive) were iso-lated as E. coliO157 by IMS based cultivation technique and latex agglu-tination. All of the 102 colonieswere verified by the detection of rfbEO157gene, while fliCh7 gene was detected in 96 of the colonies using PCR.Thus, 96 of the colonies were identified as E. coli O157:H7 while theremaining 6 as E. coli O157:H7–. From only one (0.4%) of the 240RAMS and one (0.4%) of the 240 carcass samples E. coli O157:H7– wasdetected. From 4 of the cattle, E. coli O157:H7 was isolated from bothRAMS and carcass sponges (Table 1) while none of the 240 bile sampleswere E. coli O157 positive. Interestingly, three colonies that were isolat-ed from one of the carcass sponge samples (25 K [A-C]) were suspectedof being mix colonies and were attempted to be purified on CT-SMAC.Theywere latex agglutination and sorbitol positive (sorbitol fermenting

35N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

[SF]) on consecutive purification streaks on CT-SMAC and were as wellidentified as E. coli O157:H7 by PCR. Furthermore, they showed strongagglutination with H7 antiserum and were motile in motility assay.

3.2. Seasonal distribution of E. coli O157 and influence of age, breed andgender of cattle on prevalence

Overall, 17 (7.1%) and 15 (6.3%) of 240 sampled cattle were shownto harbor E. coli O157 and E. coli O157:H7, respectively either in theirRAMS or carcass sponge samples analyzed. The prevalence of E. coliO157 in cattle was slightly higher in warmer months (May–October)(9/120, 7.5%) than colder months (November–April) (8/120, 6.6%). Fur-thermore, when prevalence of both cattle and wastewater positivitywas considered according to the seasonal distribution of two years, astatistically not significant (p N 0.05) increase in prevalence duringwarmer months was observed as prevalences were 4/66 (6.1%, 3 cattleand 1 wastewater), 7/66 (10.6%, 5 cattle and 2 wastewater), 8/66(12.1%, 7 cattle and 1 wastewater), and 3/66 (4.5%, 2 cattle and 1 waste-water) during spring (March–May), summer (June–August), autumn(September–November), andwinter (December–February), respectively.

In our study, the prevalence of E. coli O157 was higher (p N 0.05) inold (3/37, 8.1%) than young (11/158, 7.0%) andmature (3/45, 6.7%) cat-tle. Additionally, E. coli O157 was more prevalent (p N 0.05) in culture(7/88, 8.0%) and native breeds (5/62, 8.1%) than crossbreds (5/90,5.6%) and in female (6/77, 7.8%) than male (11/163, 6.7%) cattle.When only RAMS findings were taken into consideration, the preva-lence of E. coli O157 was higher (p N 0.05) in young (8/158, 5.1%) thanold (1/37, 2.7%) and mature (1/45, 2.2%) cattle. Also, E. coli O157 wasmore prevalent (p N 0.05) in crossbreds (5/90, 5.6%) than culture

Table 3Distribution of major virulence genes profiles of E. coli O157:H7+/H7– isolated from cattle and

Straina n Seasonb Agec Breedd Gendere

Cattle3R (A–C) 3 S Y N M3 K (A, C, D, E) 4 S Y N M19R (A–E) 5 S Y C1 M25 K (A–C)f 3 S Y C2 M34R (A–E) 5 S Y C1 M36 K (A–C) 3 S O N F44R (A–D) 4 W O Cr F44 K (A–E) 5 W O Cr F68R (A, B) 2 W Y C1 M69R (A–E) 5 W Y Cr M91 K (A–D) 4 W Y N M120R (A–C) 3 S Y Cr F120 K (A, B, D, E) 4 S Y Cr F120 K (C) 1 S Y Cr F135R (A–D) 4 S Y C1 M143R (A–E) 5 S M Cr M143 K (A–D) 4 S M Cr M163 K (A–E) 5 W M N F168 K (A–E) 5 W Y C3 M210 KB 1 W M C3 F219R (A–C) 3 W Y Cr M236 KB 1 S O N F236KE 1 S O N F

WastewaterM1A 1 SM1 (C, D, E) 3 SM14 (A–D) 4 SM17 (A–E) 5 WM18 (A–E) 5 WM21 (A, C, D, E) 4 W

a Sample cattle number; sample (R: rectoanal mucosal swap; K: carcass sponge); A–E: colonb S:Warm months (May–October); W: Cold months (November–April).c Y: Young (Y ≤ 24 months);M:Mature (24 months b M ≤ 4 years); O: Old (4 years b O).d C: Culture (C1: Holstein; C2: Brown Swiss; C3: Simmental); Cr: Crossbred; N: Native (Anate M:Male; F: Female.f Sorbitol positive colonies on CT-SMAC (see Section 3.1.).

(4/88, 4.5%) and native breeds (1/62, 1.6%) and in male (8/163, 4.9%)than female (3/77, 3.9%) cattle.

3.3. Virulence gene profile of E. coli O157:H7+/H7– from cattle andwastewater samples

Distribution of major virulence gene profiles of 102 E. coli O157:H7+/H7– isolated from cattle and wastewater samples was given inTable 3. Out of 96 E. coli O157:H7 isolates, 81 (84.4%) were carrying atleast one, 31 (32.3%)were carrying bothwhile 15 (15.6%)were negativefor both of the shiga toxin genes. In all of the 96 E. coli O157:H7 isolateseaeA, hly and lpfA1-3 genes were found but only in 3 SF E. coli O157:H7colonies espA was not detected. Absence of espA besides shiga toxingenes was the main genetic difference observed between SF and non-sorbitol fermenting E. coli O157:H7 colonies. None of the six E. coliO157:H7– strains were harboring shiga toxin genes while 5 were onlyhly+, whereas the remaining one was eaeγ1

+, hly+ and lpfA1-3+. Noneof the isolated E. coliO157:H7+/H7– strains harbored any of the investi-gated intimin variants other than eaeγ1 or shiga toxin variants stx1d,stx2e, stx2f or stx2g. Intimin variant eaeγ1 and shiga toxin 1 variant stx1cwas detected from all of the eaeA+ (97/102, 95.1%) and stx1

+ (32/102,31.3%) strains, respectively. From stx2

+ (80/102, 78.4%) isolates, bothstx2c (68/80, 85.0%) and stx2d (12/80, 15.0%) variants were determined.

From 102 E. coli O157:H7+/H7– colonies, 8 different virulence pro-files were observed. E. coli O157:H7 (eaeγ1, hly

+, stx1−, stx2c

+, lpfA1-3+,espA+) was the most prevalent (38/102, 37.3%) profile, followed by(eaeγ1, hly

+, stx1c+, stx2c

+, lpfA1-3+, espA+) (30/102, 29.4%); (eaeγ1,hly+, stx1

−, stx2−, lpfA1-3+, espA+) (12/102, 11.8%) and (eaeγ1, hly

+,stx1

−, stx2d+, lpfA1-3+, espA+) (11/102, 10.8%). From each of the two car-

cass sponge samples (120 K and 236 K) and one wastewater sample

wastewater samples.

O157 H7 eae hly stx1 stx2 lpf espA

+ + γ1 + − d + ++ + γ1 + − d + ++ − − + − − − −+ + γ1 + − − + −+ + γ1 + − c + ++ + γ1 + − c + ++ + γ1 + c c + ++ + γ1 + c c + ++ + γ1 + − c + ++ + γ1 + − c + ++ + γ1 + − d + ++ + γ1 + c c + ++ + γ1 + c c + ++ + γ1 + c − + ++ + γ1 + − c + ++ + γ1 + − c + ++ + γ1 + − c + ++ + γ1 + − c + ++ + γ1 + − c + ++ − γ1 + − − + −+ + γ1 + c c + ++ + γ1 + c c + ++ + γ1 + − − + +

+ + γ1 + c d + ++ + γ1 + − − + ++ + γ1 + − − + ++ + γ1 + c c + ++ + γ1 + c c + ++ + γ1 + − − + +

y code.

olian Black).

36 N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

(M1), two different virulence profile harboring E. coli O157:H7 colonieswere isolated, while E. coliO157:H7 strains thatwere isolated frombothRAMS and carcass sponges of each individual four cattle (3R-3 K, 44R-44 K, 120R-120 K and 143R-143 K) showed the exact same virulenceprofile.

4. Discussion

In this study, over a two year period, we attempted determiningE. coli O157:H7 prevalence at the slaughterhouse and slaughterhousewastewater, its seasonal distribution and the effects of age, breed andgender of cattle on its distribution. Furthermore, pathogenic potentialof the isolated strains was investigated by determination of the majorvirulence genes. In the last decade, limited number of studies on theprevalence of E. coli O157:H7 in cattle at abattoir environment havebeen reported in Turkey (Akkaya et al., 2007; Gun et al., 2003; Yilmazet al., 2002), but in none of them seasonal distribution and extended ge-nomic characterization over a relatively long term period were evaluat-ed. The prevalence of E. coliO157:H7 found in our studywas higher thanthefindings reported fromSweden (Albihn et al., 2003), Ireland (Carneyet al., 2006), Serbia (Nastasijevik et al., 2009), Greece (Pinaka et al.,2013), Mexico (Varela-Hernandez et al., 2007) and Turkey (Akkayaet al., 2007; Gun et al., 2003; Sarimehmetoglu et al., 2009; Yilmazet al., 2002) while lower than reported in the USA (Barkocy-Gallagheret al., 2003; Brichta-Harhay et al., 2008), Canada (Van Donkersgoedet al., 1999) and Jordan (Osaili et al., 2013) which could be attributedto geographical discrepancies, variations in sampling and isolationtechniques.

Higher frequency of E. coli O157:H7 in warmer months than coldermonths due to temperature variation has long been recognized(Barkocy-Gallagher et al., 2003; Gyles, 2007; Van Donkersgoed et al.,1999), though more recent hypotheses relating this phenomenon today length (Edrington et al., 2006) or levels of thyroid hormones(Schultz et al., 2005) were published. Contrary to Scottish cattle(Ogden et al., 2004), in our study prevalencewas higher inwarmmonthsthan cold months, but this did not reveal a statistically significant differ-ence, as previously observed by Brichta-Harhay et al. (2008), whichmight be attributed to relatively small number of E. coli O157:H7positivity.

In previous reports, there is a decline in E. coliO157:H7prevalence asthe cattle gets older. In weaned calves and young stock (5–24 months),prevalence was as high as 12% while in adult stock (N2 years) it wasmuch lower (2%) (Rugbjerg et al., 2003). A likewise scenario wasobserved by other researchers (Albihn et al., 2003; Yilmaz et al., 2002)as well. This difference in prevalence might be resulting from the directeffects of amount, frequency and/or duration of organisms being shed(Renter and Sargeant, 2002) yet, consumption of old cattlemeats, main-ly cows, preferably asmincedmeat and products due to the lower qual-ity they present, continues to be a human health risk (Faith et al., 1996).Though prevalence was higher inmale cattle, no significant influence ofgender of cattle was seen on prevalence of E. coli O157:H7 at slaughterin our study. As the age of cattle increases however, a shift in higher pos-itivity of male to female cattle was observed but this could be attributedto the decrease in number of male cattle presented to slaughter. Higherprevalence inmale (Yilmaz et al., 2002) or female (Rugbjerg et al., 2003)cattle was observed by other groups, as some observed non-significantdifferences (Nielsen et al., 2002; Van Donkersgoed et al., 1999). Sincetoo many varieties of culture, crossbred or native cattle are brought toslaughter; we aimed to investigate if these categories have any influ-ence on the prevalence of E. coliO157:H7. Prevalence in crossbred cattlewas non-significantly higher than culture or native cattle. AlthoughAlbihn et al. (2003) reported lower prevalence in crossbred than dairyor beef breeds, other researchers' results were contradictory (Jeonet al., 2013; Rugbjerg et al., 2003). So in order to enlighten the influenceof these categories, further works are needed.

Colonization site and shedding dynamics of E. coli O157:H7 in cattlehave long been debated and studies for determination of the most sen-sitive sampling techniques were reported. Even though it is possible toisolate E. coli O157:H7 throughout the gastrointestinal tract of cattle(Callaway et al., 2009; Keen et al., 2010), lymphoid follicle-denserectoanal mucosa at terminal rectum was shown to be the primarycolonization site (Low et al., 2005) and proven to be more sensitiveand less laborious in sampling (Lim et al., 2007; Rice et al., 2003). Gallbladder/bile has also been implicated as an additional possible coloniza-tion site by several studies (Jeong et al., 2007; Goncuoglu et al., 2010;Stoffregen et al., 2004) but it was not proven yet (Lim et al., 2007;Reinstein et al., 2007). Our results also showed that gall bladder/bile isnot a prevalent colonization site. Investigation of hide contaminationwas not the scope of this study, but determination of E. coli O157:H7,each showing the exact same virulence profile in RAMS and carcasssponge samples in only 4 cattle out of 11 carcass sponge positive cattle,reveals the importance of hide contamination and indirect transmission(Callaway et al., 2009).

Previously, Friedrich et al. (2002) associated presence of stx1, stx2and stx2 variants in E. coli isolates with asymptomatic individuals orpatients of clinical manifestations of either HUS or diarrhea withoutHUS. In the study, they find out that the presence of stx2cwasmore like-ly to cause HUS while the presence of stx2d may manifest a milder case(Friedrich et al., 2002). Furthermore, Kawano et al. (2008) have also as-sociated HUS development with the presence of stx2. A higher in vitrocytotoxicity was also reported for stx2c harboring E. coli O157:H7 thanstx1-stx2 or stx1-stx2c harboring strains (Lefebvre et al., 2009). Accordingto these studies, it is possible to speculate that many of our bovine orwastewater originated isolates are highly virulent and genetically har-bor the ability to cause human infections withmanifestation of diarrheaand/or HUS.

Intimin variants were shown to influence the tissue tropism and col-onization site (Fitzhenry et al., 2002; Mundy et al., 2007) and in accor-dance with previous reports (Blanco et al., 2004; Oswald et al., 2000)all our intimin positive E. coli O157:H7+/H7– strains were indeedeaeγ1 and were as well positive for lpfA1-3. Besides the 6 E. coli O157:H7– isolates and 3 SF E. coli O157:H7 isolates, all of our isolates wereespA positive indicating their capability in initial binding and formationof attaching–effacing lesions (Kenny et al., 1996). By late 80s, SF E. coliO157:H– had emerged in Germany and since then, has been isolatedfrom human clinical cases and meats (Karch and Bielaszewska, 2001;Sallam et al., 2013). Interestingly, 3 of our isolates were SF E. coliO157:H7 rather than SF E. coli O157:H– resembling themodel predicted(Bono et al., 2012) and further identified (Jenke et al., 2012) SF E. coliO157:H7 descending from STEC O55:H7 progenitor. However, morework on our SF E. coli O157:H7 isolates is needed before any furtherspeculations. SF E. coli O157:H7 strains isolated in this study were mo-tile, reactedwith H7 antiserum and did not harbor any of the investigat-ed shiga toxin genes. However, since excision of shiga toxin-encodinglambdoid prophages with varying exogenous factors that stress E. coliO157:H7 cells is possible (Shaikh and Tarr, 2003), it is alarming withregards to use of SF dependent E. coli O157 isolation techniques inwhich the presence of shiga toxigenic SF E. coli O157:H7 colonies willbe unnoticed.

Since E. coli O157:H7 survival in aquatic environments vary fromtwo weeks to 10 months and transmission to humans through the useof untreated irrigation water for fresh produce is evident (Chekababet al., 2013), isolation of shiga toxigenic E. coli O157:H7 from slaughter-house wastewater with a high prevalence (20.8%) is alarming. Thisstudy reveals the importance of ensuring establishment of water treat-ment facilities to slaughterhouse wastewater efflux by governmentalauthorities.

In conclusion, from 744 cattle and slaughterhouse wastewater sam-ples 4.2% of the RAMS, 4.6% of the carcass (6.3% of the cattle) and 20.8%of the wastewater samples were found contaminated with virulentE. coli O157:H7+/H7–. When we compared the findings to the previous

37N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

reports it is clear that in the last decade, prevalence of E. coli O157:H7has an increasing trend in cattle. This is probably because of the highlycontaminated grasslands and cross contaminations during slaughtering.We believe that slaughterhouses are the significant sources of environ-mental contamination with E. coli O157:H7. In this study age, gender,breed and season did not show significant effects on the prevalence ofE. coli O157:H7 in cattle and thus, further studies that focus on alterna-tive effective factors are necessary. Detection and molecular characteri-zation of SF E. coli O157:H7 is a novel finding andmay lead to a revisionof reference isolation procedure of E. coli O157:H7 in future.

Acknowledgments

Thisworkwas supported by TUBITAK (The Scientific and Technolog-ical Research Council of Turkey) grant no. 110R013. We are grateful toJ. Osek and C. Garcia-Aljaro for kindly supplying the reference strains.

References

Abe, B.A., Heczko, U., Hegele, R.G., Finlay, B.B., 1998. Two enteropathogenic Escherichia colitype III secreted proteins, EspA and EspB, are virulence factors. J. Exp. Med. 188,1907–1916.

Akkaya, L., Cetinkaya, Z., Alisarli, M., Telli, R., Gök, V., 2007. The prevalence of E. coli O157/O157:H7, L. monocytogenes and Salmonella spp. on bovine carcasses in Turkey.J. Muscle Foods 19, 420–429.

Albihn, A., Eriksson, E., Wallen, C., Aspan, A., 2003. Verotoxinogenic Escherichia coli (VTEC)O157:H7— a nationwide Swedish survey of bovine faeces. Acta Vet. Scand. 44, 43–52.

Arthur, T.M., Bosilevac, J.M., Nou, X., Shackelford, S.D., Wheeler, T.L., Kent, M.P., Jaroni, D.,Pauling, B., Allen, D.M., Koohmaraie, M., 2004. Escherichia coli O157 prevalence andenumeration of aerobic bacteria, Enterobacteriaceae, and Escherichia coli O157 atvarious steps in commercial beef processing plants. J. Food Prod. 67, 658–665.

Barkocy-Gallagher, G.A., Arthur, T.M., Rivera-Betancourt, M., Nou, X., Shackelford, S.D.,Wheeler, T.L., Koohmaraie, M., 2003. Seasonal prevalence of shiga toxin-producingEscherichia coli, including O157:H7 and nonO157 serotypes, and Salmonella incommercial beef processing plants. J. Food Prod. 66, 1978–1986.

Blanco, M., Blanco, J.E., Mora, A., Dahbi, G., Alonso, M.P., Gonzalez, E.A., Bernardez, M.I.,Blanco, J., 2004. Serotypes, virulence genes, and intimin types of shiga toxin(verotoxin)-producing Escherichia coli isolates from cattle in Spain and identificationof a new intimin gene (eae-î). J. Clin. Microbiol. 42, 645–651.

Bono, J.L., Smith, T.P.L., Keen, J.E., Harhay, G.P., McDaneld, T.G., Mandrell, R.E., Jung, W.K.,Besser, T.E., Gerner-Smidt, P., Bielaszewska, M., Karch, H., Clawson, M.L., 2012.Phylogeny of shiga toxin-producing Escherichia coli O157 isolated from cattle andclinically ill humans. Mol. Biol. Evol. http://dx.doi.org/10.1093/molbev/mss072.

Brichta-Harhay, D.M., Guerini, M.N., Arthur, T.M., Bosilevac, J.M., Kalchayanand, N.,Shackelford, S.D., Wheeler, T.L., Koohmaraie, M., 2008. Salmonella and Escherichiacoli O157:H7 contamination on hides and carcasses of cull cattle presented forslaughter in the United States: an evaluation of prevalence and bacterial loads byimmunomagnetic separation and direct plating methods. Appl. Environ. Microbiol.74, 6289–6297.

Bürk, C., Dietrich, R., Açar, G., Moravek, M., Bülte, M., Märtlbauer, E., 2003. Identificationand characterization of a new variant of shiga toxin 1 in Escherichia coli ONT:H19 ofbovine origin. J. Clin. Microbiol. 41, 2106–2112.

Byrne, C.M., Erol, I., Call, J.E., Kaspar, C.W., Buege, D.R., Hiemke, C.J., Fedorka-Cray, P.J.,Benson, A.K., Wallace, F.M., Luchansky, J.B., 2003. Characterization of Escherichia coliO157:H7 from downer and healthy dairy cattle in the Upper Midwest Region of theUnited States. Appl. Environ. Microbiol. 69, 4683–4688.

Callaway, T.R., Carr, M.A., Edrington, T.S., Anderson, R.C., Nisbet, D.J., 2009. Diet, Escherichiacoli O157:H7, and cattle: a review after 10 years. Curr. Issues Mol. Biol. 11, 67–79.

Carney, E., O'Brien, S.B., Sheridan, J.J., McDowell, D.A., Blair, I.S., Duffy, G., 2006. Prevalenceand level of Escherichia coli O157 on beef trimmings, carcasses and boned head meatat a beef slaughter plant. Food Microbiol. 23, 52–59.

Chekabab, S.M., Paquin-Veillette, J., Dozois, C.M., Harel, J., 2013. The ecological habitat andtransmission of Escherichia coli O157:H7. FEMS Microbiol. Lett. 341, 1–12.

Edrington, T.S., Callaway, T.R., Ives, S.E., Engler, M.J., Looper, M.L., Anderson, R.C., Nisbet,D.J., 2006. Seasonal shedding of Escherichia coliO157:H7 in ruminants: a new hypoth-esis. Foodborne Pathog. Dis. 3, 413–421.

Elder, R.O., Keen, J.E., Siragusa, G.R., Barkocy-Gallacher, G.A., Koohmaraie, M., Laegreid,W.W., 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence infeces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci.U. S. A. 97, 2999–3003.

Faith, N.G., Shere, J.A., Brosch, R., Arnold, K.W., Ansay, S.E., Lee, M.S., Luchansky, J.B.,Kaspar, C.W., 1996. Prevalence and clonal nature of Escherichia coli O157:H7 ondairy farms in Wisconsin. Appl. Environ. Microbiol. 62, 1519–1525.

Farfan, M.J., Torres, A.G., 2011. Molecular mechanisms mediating colonization of shigatoxin-producing Escherichia coli strains. Infect. Immun. 80, 903–913.

Fitzhenry, R.J., Pickard, D.J., Hartland, H.L., Reece, S., Dougan, G., Phillips, A.D., Frankel, G.,2002. Intimin type influences the site of human intestinal mucosal colonization byenterohaemorrhagic Escherichia coli O157:H7. Gut 50, 180–185.

Fratamico, P.M., Bagi, L.K., Pepe, T., 2000. A multiplex polymerase chain reaction assay forrapid detection and identification of Escherichia coli O157:H7 in foods and bovinefeces. J. Food Prot. 63, 1032–1037.

Friedrich, A.W., Bielaszewska, M., Zhang, W.L., Pulz, M., Kuczius, T., Ammon, A., Karch, H.,2002. Escherichia coli harboring shiga toxin 2 gene variants: frequency and associa-tion with clinical symptoms. J. Infect. Dis. 185, 74–84.

FSIS (Food Safety and Inspection Service), 1996. Pathogen reduction, hazard analysis andcritical control points (HACCP) systems. Final rule, USDA, 9 CFR Part 304,pp. 38806–38989.

Goncuoglu, M., Erol, I., Ayaz, N.D., BilirOrmanci, F.S., Kaspar, C.W., 2010. Isolation andgenomic characterization of Escherichia coli O157:H7 in bile of cattle. Ann. Microbiol.60, 293–297.

Gun, H., Yilmaz, A., Turker, S., Tanlasi, A., Yilmaz, H., 2003. Contamination of bovinecarcasses and abattoir environment by Escherichia coli O157:H7 in Istanbul. Int.J. Food Microbiol. 84, 339–344.

Gyles, C.L., 2007. Shiga toxin-producing Escherichia coli: an overview. J. Anim. Sci. 85, 45–62.Hussein, H.S., Bollinger, L.M., 2005. Prevalence of shiga toxin-producing Escherichia coli in

beef cattle. J. Food Prod. 68, 2224–2241.Imberechts, H., De Greve, H., Lintermans, P., 1992. The pathogenesis of edema disease in

swine. Vet. Microbiol. 31, 221–233.Jenke, C., Leopold, S.L., Weniger, T., Rothgänger, J., Harmsen, D., Karch, H., Mellmann, A.,

2012. Identification of intermediate in evolutionary model of enterohemorrhagicEscherichia coli O157. Emerg. Infect. Dis. 18, 582–588.

Jeon, S.J., Elzo, M., DiLorenzo, N., Lamb, G.C., Jeong, K.C., 2013. Evaluation of animal geneticand physiological factors that affect the prevalence of Escherichia coli O157 in cattle.PLoS One 8, e55728 (1–9).

Jeong, K.C., Kang, M.Y., Heimke, C., Shere, J.A., Erol, I., Kaspar, C.W., 2007. Isolation ofEscherichia coli O157:H7 from the gall bladder of inoculated and naturally-infectedcattle. Vet. Microbiol. 119, 339–345.

Karch, H., Bielaszewska, M., 2001. Sorbitol-fermenting shiga-toxin-producing Escherichiacoli O157:H(−) strains: epidemiology, phenotypic and molecular characteristics,and microbiological diagnosis. J. Clin. Microbiol. 39, 2043–2049.

Kawano, K., Okada, M., Haga, T., Maeda, K., Goto, Y., 2008. Relationship between pathoge-nicity for humans and stx genotype in shiga toxin-producing Escherichia coli serotypeO157. Eur. J. Clin. Microbiol. Infect. Dis. 27, 227–232.

Keen, J.E., Laegreid, W.W., Chitko-McKown, C.G., Durso, L.M., Bono, J.L., 2010. Distributionof shiga-toxigenic Escherichia coli O157 in the gastrointestinal tract of naturallyO157-shedding cattle at necropsy. Appl. Environ. Microbiol. 76, 5278–5281.

Kenny, B., Lai, L.C., Finlay, B.B., Donnenberg, M.S., 1996. EspA, a protein secreted byenteropathogenic Escherichia coli, is required to induce signals in epithelial cells.Mol. Microbiol. 20, 313–323.

Kulow, M.J., Gonzales, T.K., Pertzborn, K.M., Dahm, J., Miller, B.A., Park, D., Gautam, R.,Kaspar, C.W., Ivanek, R., Döpfer, D., 2012. Differences in colonization and sheddingpatterns after oral challenge of cattle with three Escherichia coli O157:H7 strains.Appl. Environ. Microbiol. 78, 8045–8055.

Lefebvre, B., Diarra, M.S., Vincent, C., Moisan, H., Malouin, F., 2009. Relative cytotoxicity ofEscherichia coli O157:H7 isolates from beef cattle and humans. Foodborne Pathog. Dis.6, 357–364.

Leung, P.H.M., Peiris, J.S.M., Ng,W.W.S., Robins-Browne, R.M., Bettelheim, K.A., Yami,W.C.,2003. A newly discovered verotoxin variant, VT2g, produced by bovineverocytotoxigenic Escherichia coli. Appl. Environ. Microbiol. 69, 7549–7553.

Lim, J.Y., Li, J., Sheng, H., Besser, T.E., Potter, K., Hovde, C.J., 2007. Escherichia coli O157:H7colonization at the rectoanal junction of long-duration culture-positive cattle. Appl.Environ. Microbiol. 73, 1380–1382.

Low, J.C., McKendrick, I.J., McKechnie, C., Fenlon, D., Naylor, S.W., Currie, C., Smith, D.G.,Allison, L., Gally, D.L., 2005. Rectal carriage of enterohemorrhagic EscherichiacoliO157 in slaughtered cattle. Appl. Environ. Microbiol. 71, 93–97.

McEvoy, J.M., Doherty, A.M., Finnerty, M., Sheridan, J.J., McGuire, L., Blair, I.S., McDowell,D.A., Harrington, D., 2000. The relationship between hide cleanliness and bacterialnumbers on beef carcasses at a commercial abattoir. Lett. Appl. Microbiol. 30, 390–395.

McNally, A., Roe, A.J., Simpson, S., Thomson-Carter, F.M., Hoey, D.E.E., Currie, C.,Chakraborty, T., Smith, D.G.E., Gally, D.L., 2001. Differences in levels of secretedlocus of enterocyte effacement proteins between human-disease associated andbovine Escherichia coli O157. Infect. Immun. 69, 5107–5114.

Meng, J., Doyle, M.P., Zhao, T., Zhao, S., 2001. Enterohemorrhagic Escherichia coli O157:H7.In: Doyle, M.P., Beuchat, L.R., Montville, T.J. (Eds.), Food Microbiology: Fundamentalsand Frontiers. ASM Press, Washington D.C., pp. 193–213.

Mundy, R., Schüller, S., Girard, F., Fairbrother, J.M., Phillips, A.D., Frankel, G., 2007.Functional studies of intimin in vivo and ex vivo: implications for host specificityand tissue tropism. Microbiology 153, 959–967.

Nastasijevik, I., Mitrovic, R., Buncic, S., 2009. The occurrence of E. coli O157 in/on faeces,carcasses and fresh meats from cattle. Meat Sci. 82, 101–105.

Nielsen, E.M., Tegtmeier, C., Andersen, H.J., Grønbaek, C., Andersen, J.S., 2002. Influence ofage, sex and herd characteristics on the occurrence of Verocytotoxin-producingEscherichia coli O157 in Danish dairy farms. Vet. Microbiol. 88, 245–257.

Ogden, I.D., MacRae, M., Strachen, N.J.C., 2004. Is prevalence and shedding of E. coli O157in beef cattle in Scotland seasonal? FEMS Microbiol. Lett. 233, 297–300.

Osaili, T.M., Alaboudi, A.R., Rahahlah,M., 2013. Prevalence and antimicrobial susceptibilityof Escherichia coli O157:H7 on beef cattle slaughtered in Amman abattoir. Meat Sci.93, 463–468.

Osek, J., 2003. Development of a multiplex PCR approach for the identification of shigatoxin-producing Escherichia coli strains and their major virulence factor genes.J. Appl. Microbiol. 95, 1217–1225.

Oswald, E., Schmidt, H., Morabito, S., Karch, H., Marches, O., Caprioli, A., 2000. Typing ofintimin genes in human and animal enterohemorrhagic and enteropathogenicEscherichia coli: characterization of a new intimin variant. Infect. Immun. 68, 64–71.

38 N.D. Ayaz et al. / International Journal of Food Microbiology 174 (2014) 31–38

Pinaka, O., Pournaras, S., Mouchtouri, V., Plakokefalos, E., Katsiaflaka, A., Kolokythopoulou,F., Barboutsi, E., Bitsolas, N., Hadjichristodoulou, C., 2013. Shiga toxin-producingEscherichia coli in central Greece: prevalence and virulence genes of O157:H7 andnon-O157 in animal feces, vegetables, and humans. Eur. J. Clin. Microbiol. Infect.Dis. http://dx.doi.org/10.1007/s10096-013-1889-6.

Reinstein, S., Fox, J.T., Shi, X., Nagaraja, T.G., 2007. Prevalence of Escherichia coli O157:H7in gallbladders of beef cattle. Appl. Environ. Microbiol. 73 (3), 1002–1004.

Renter, D.G., Sargeant, J.M., 2002. Enterohemorrhagic Escherichia coli O157: epidemiologyand ecology in bovine production environments. Anim. Health Res. Rev. 3, 83–94.

Rice, D.H., Sheng, H.Q.,Wynia, S.A., Hovde, C.J., 2003. Rectoanalmucosal swab culture ismoresensitive than fecal culture and distinguishes Escherichia coli O157:H7– colonized cattleand those transiently shedding the same organism. J. Clin. Microbiol. 41, 4924–4929.

Rugbjerg, H., Nielsen, E.M., Andersen, J.S., 2003. Risk factors associated with faecal shed-ding of verocytotoxin-producing Escherichia coli O157 in eight known-infected Dan-ish dairy herds. Prev. Vet. Med. 58, 101–113.

Sallam, K.I., Mohammed, M.A., Ahdy, A.M., Tamura, T., 2013. Prevalence, genetic charac-terization and virulence genes of sorbitol-fermenting Escherichia coli O157:H– andE. coli O157:H7 isolated from retail beef. Int. J. Food Microbiol. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.05.024.

Sarimehmetoglu, B., Aksoy, M.H., Ayaz, N.D., Ayaz, Y., Kuplulu, O., Kaplan, Y.Z., 2009.Detection of Escherichiacoli O157:H7 in ground beef using immunomagneticseparation and multiplex PCR. Food Control 20, 357–361.

Schultz, C.L., Edrington, T.S., Schroeder, S.B., Hallford, D.M., Genovese, K.J., Callaway, T.R.,Anderson, R.C., Nisbet, D.J., 2005. Effect of the thyroid on faecal shedding of E. coliO157:H7 and Escherichia coli in naturally infected yearling beef cattle. J. Appl.Microbiol. 99, 1176–1180.

Shaikh, N., Tarr, P.I., 2003. Escherichia coli O157:H7 shiga toxin-encoding bacteriophages:integrations, excisions, truncations, and evolutionary implications. J. Bacteriol. 185,3596–3605.

SPSS, 2013. Statistical Analysis Package Program. Version 16.0.Stoffregen, W.C., Pohlenz, J.F.L., Dean-Nystrom, E.A., 2004. Escherichia coli O157:H7 in

gallbladders of experimentally infected calves. J. Vet. Diagn. Investig. 16, 79–83.

Torres, A.G., Blanco, M., Valenzuela, P., Slater, T.M., Patel, S.D., Dahbi, G., Lopez, C., Barriga,X.F., Blanco, J.E., Gomes, T.A.T., Vidal, R., Blanco, J., 2009. Genes related to long polarfimbriae of pathogenic Escherichia coli strains as reliable markers to identify virulentisolates. J. Clin. Microbiol. 47, 2442–2451.

TSI, Turkish Statistical Institute, 2013. Number of slaughtered animals and the quantity ofmeat production. http://www.turkstat.gov.tr/VeriBilgi.do?alt_id=46 (Accessiondate: 12.06.2013).

Van Donkersgoed, J., Graham, T., Gannon, V., 1999. The prevalence of verotoxins,Escherichia coli 0157:H7, and Salmonella in the feces and rumen of cattle at process-ing. Can. Vet. J. 40, 332–338.

Varela-Hernandez, J.J., Cabrera-Diaz, E., Cardona-Lopez, M.A., Ibarra-Velazquez, L.M.,Rangel-Villalobos, H., Castillo, A., Torres-Vitela, M.R., Ramirez-Alvarez, A., 2007.Isolation and characterization of Shiga toxin-producing Escherichia coli O157:H7 andnon-O157 from beef carcasses at a slaughter plant in Mexico. Int. J. Food Microbiol.113, 237–241.

Vugia, D., Cronquist, A., Hadler, J., Tobin-D'Angelo, M., Blythe, D., Smith, K., Lathrop, S.,Morse, D., Cieslak, P., Jones, T., Holt, K.G., Guzewich, J.J., Henao, O.L., Scallan, E.,Angulo, F.J., Griffin, P.M., Tauxe, R.V., 2007. Preliminary FoodNet data on the incidenceof infection with pathogens transmitted commonly through food — 10 states, 2006.Morb. Mortal. Wkly. Rep. 56, 336–339.

Wang, G., Clark, C.G., Rodgers, F.G., 2002. Detection in Escherichia coli of the genesencoding the major virulence factors, the genes defining the O157:H7 serotype,and components of the type 2 shiga toxin family by multiplex PCR. J. Clin. Microbiol.40, 3613–3619.

Widiasih, D.A., Ido, N., Omoe, K., Sugii, S., Shinagawa, K., 2003. Duration andmagnitude offaecal shedding of shiga toxin-producing Escherichia coli from naturally infected cat-tle. Epidemiol. Infect. 132, 67–75.

Yilmaz, A., Gun, H., Yilmaz, H., 2002. Frequency of E. coli O157:H7 in Turkish cattle. J. FoodProt. 65, 1637–1640.

Zhang, W., Bielaszewska, M., Kuczius, T., Karch, H., 2002. Identification, characterization,and distribution of a shiga toxin 1 gene variant (stx1c) in Escherichia coli strainsisolated from humans. J. Clin. Microbiol. 40, 1441–1446.