Evaluation of enrichment-free PCR-based detection on the rfbE gene of Escherichia coli O157—Application to municipal wastewater

Available at www.sciencedirect.com else P 7 ven Nıˆm a r t i c l e i n f o a b s t r a c t in natural environment. Human pathogens from enteric origin et al., 2002), notably in biofilm forms (Castonguay et al., 2006). pathogenicity factors, are: E. coli enterotoxigenic (ETEC), E. coli AEC), asive sible emic coding the Shiga-like toxins or verotoxins. E. coli O157:H7 ARTICLE IN PRESS WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 2 8 0 – 1 2 8 6 �Corresponding author. Tel.: +33 4 66 27 95 60; fax: +33 4 66 78 27 01. Coliforms (including E. coli) are used as bacterial indicators of fecal contamination of food and water. It can even be representative of the global presence of pathogenic bacteria, notably salmonella (Edberg et al., 2000). remain the main serotype incriminated in HUS in Europe (Bitzan et al., 1993). For example from 1997 to 2000, 57% of HUS in Germany and Austria was linked to E. coli O157 (Gerber et al., 2002). 0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2006.11.027 E-mail address: [email protected] (R. Bertrand). 2000), soil (Gagliardi and Karns, 2002), surface water (Akashi et al., 1994), wastewater (Arana et al., 2001; Vernozy-Rozand syndrome (HUS) (Haeghebaert et al., 2000). As Shiga toxin-producing E. coli (STEC), this strain has stx gene are therefore a potential contamination risk transmitted through soil, agriculture, water or sediments. E. coli, enteric bacteria, can survive for long periods in the intestinal tracts of warm-blooded animals, including humans but also in other environments as faeces (Wang et al., 1996), slurry and manure (Chapman et al., enteropathogenic (EPEC), E. coli enteroaggregative (E E. coli enterohemorrhagic (EHEC) and E. coli enteroinv (EIEC). E. coli O157:H7 is the main E. coli serotype respon for hemorrhagic colitis (Tarr, 1995) and hemolytic ur 1. Introduction Fecal contamination is one of sources of environmental bacterial contamination and is responsible for the presence of pathogens Some pathogenic strains of E. coli are known to be responsible for gastro-enteritis in children or for ‘‘traveller’s diarrhoea’’ (Dupont, 1995). The main intestinal pathotypes, described according to observed clinical signs and expressed Article history: Received 22 December 2005 Received in revised form 9 November 2006 Accepted 20 November 2006 Available online 12 January 2007 Keywords: E. coli O157 PCR rfbE Wastewater detection Escherichia coli O157 strains have emerged as important human enteric pathogens. Strains that express the O-antigen 157 are commonly associated with severe clinical manifesta- tions, including bloody diarrhea, hemorrhagic colitis, and hemolytic uremic syndrome. E. coli O157 strains may be transmitted in a variety of ways, including food, water and person- to-person or animal-to-person contact. Fecal contamination is one source of environ- mental contamination and is responsible for the presence of these pathogens in the environment. We used a specific and sensitive PCR assay based on the rfbE gene to detect low levels of these pathogens in wastewater. The set of primers used was designed to amplify an intragenic segment of the rfbE gene. The amplification assay detected 200 CFU of E. coli O157 in pure water. The prevalence of E. coli O157 in the effluents of 44 wastewater treatment plants was determined (7%). & 2006 Elsevier Ltd. All rights reserved. journal homepage: www. Evaluation of enrichment-free rfbE gene of Escherichia coli O15 wastewater Romain Bertranda,�, Benoit Roiga,b aBiodiagnostic Department, Ecole des Mines d’Ale´s, Centre LGEI, 6 A bJE2479, Centre Universitaire de Nıˆmes, rue du Dr G. Salan, 30021 vier.com/locate/watres CR-based detection on the —Application to municipal ue de Clavie`res, 30319 Ales, France es Cedex 1, France on appropriate medium. sodium acetate 3 M (pH 4.5) and 2.5 volume of ice-cold ARTICLE IN PRESS (2 Although the predominant mode of transmission to humans seems to be contaminated meat or meat products, infection from contaminated water has also been described. An E. coli O157:H7 infection outbreak was described in Ontario (May 2000), in an extensive cattle breeding zone (Anonymous, 2000). More than 2000 cases were identified among 4600 inhabitants. Of 65 inpatients, 27 developed HUS and 7 died (Anonymous, 2000). Epidemiologic research suggested that the drinking water network had been responsible (water resources can be contaminated with surface waters, particu- larly in flood periods). Outbreaks associated with contami- nated drinking water have also been reported in the US (Olsen et al., 2002), Japan (Akashi et al., 1994) and Europe (Paunio et al., 1999). In France, STEC detection assays were carried out on sludge from urban wastewater treatment plants (Vernozy-Rozand et al., 2002). The E. coli O157:H7 strain was found in a wastewater aerator. However, information about the exact prevalence of STEC and notably E. coli O157 in urban waste- water is still lacking. Infective doses for humans are very low (o10 cells) and a small number of cells (about 10–100) in the environment can have major consequences on health (Keene et al., 1994). Serotype O157 can be detected only by specialized labora- tory methods, through inoculation in the Sorbitol-MacConkey agar culture medium (SMAC) (Lingwood et al., 1987). Con- firmation tests use immunologic and genetic methods. For each specific serotype, seroreaction against the verotoxins or against the lipopolysaccharide of the O157 strain can also be used. There is also the reference method (ISO IN 16654), which only provides results after 3 days. For detection in water, the method recommended by the Public Health Laboratory Service (UK) requires 44 h. Sensitivity and speed in detection are required in order accurately to evaluate public health risks (sensitivity) and quickly to identify infected water so as to minimize human exposure (speed). There is therefore a major need for new methods responding to these constraints. In this context, genetic methods are increasingly being used. But the current methods require a pre-treatment enrichment step to detect 1 colony forming units g�1 (CFU g�1). Cui et al. (2003) detected 1000 CFU g�1 in ground beef without this step, and 1 CFU g�1 with only 6 h enrichment. Abdulmawjood et al. (2004) detected �2000 cells per reaction with 100% probability (European project ‘‘Validation and Standardization of Diagnostic PCR for Detection of Foodborne Pathogens’’). Genetic methods target mainly the uidA, fliC or rfbE gene because they are specific to the O157:H7 serotype. rfbE is particularly used because all strains expressing this antigen are associated with severe clinical symptoms. Some PCR assays based on rfbE sequences have been published, with various sensitivities and selectivities (Desmarchelier et al., 1998; Paton and Paton, 1998; Maurer et al., 1999; Abdulmaw- jood et al., 2004). This paper presents the development of a rapid, enrich- ment free, sensitive and specific method, for detecting the pathogenic E. coli O157 strain in water and wastewater by targeting a fragment of the rfbE gene. To achieve this purpose, WAT E R R E S E A R C H 41 DNA was extracted from water or wastewater samples using an original procedure, and gene analysis and specificity tests ethanol. The mixture was centrifuged at 15,700� g for 20 min. Pellets were washed with 70% ethanol and vacuum- dried, then dissolved in 50ml TE (10 mM Tris-HCl+1 mM EDTA) and stored at �20 1C for later use. To assess the relative efficiency of this procedure, DNA isolation was performed following others simple methods. The DNA quantity obtained and its quality were determined spectrophotometrically 2.2. PCR template preparation Homogenized samples (1 l) were filtered through a polycarbo- nate membrane (Millipore, 47 mm, 0.45mm). The filtrates and membranes were re-suspended in pure sterile water (1.8 ml). DNA was extracted from filtrates and membranes by a direct lysis procedure developed for this study. The filtrate and membrane were boiled for 5 min at 100 1C. Two hundred milliliters of a 10 mg l�1 Lysozyme/Proteinase K solution in TE10X buffer (100 mM Tris-HCl+10 mM EDTA) was added to the filtrate and membrane mixture (1.8 ml) giving a 1 mg ml�1 final concentration of Lysozyme and Proteinase K. The lysis mix was incubated for 15 min at 37 1C. The mixture (2 ml) was used for DNA extraction and purification with an equal volume of tris-buffered phenol–chloroform–isoamyl alcohol (2/3 phenol+1/3 chloroform, pH 8, QBiogene), followed by extraction with an equal volume of chloroform. This volume of phenol/chloroform is required for complete dissolution of the filters in the organic phase. Nucleic acids were precipitated via the addition of 1/10 volume of were carried out to determine the molecular target. Original sequences were designed and evaluated for the characteriza- tion of these pathogens by a molecular method. The method was applied to the detection of the E. coli O157 strain in wastewater from Urban WasteWater-Treatment Plant (UWWTP) samples. 2. Materials and methods 2.1. Bacterial strains and culture The E. coli O157:H7 ATCC reference strain 43895 was used in all experiments. Bacteria were inoculated in LB Broth medium, incubated at 37 1C with stirring for 18–24 h then spread on LB agar plates and conserved at 4 1C. These colonies were used to prepare solutions with various cell concentra- tions (controlled by plate counts on LB agar medium). The strains of EPEC, uropathogenic E. coli (UPEC), E. coli K1 (NMEC), Listeria monocytogenes, Staphylococcus aureus, Legionella pneumophila, Enterococcus faecium, Klebsiella pneumoniae, Strep- tococcus pyogenes, Pseudomonas aeruginosa were obtained from the strain collection of the University Hospital Center, Nıˆmes, France. The E. coli XL10 strain was obtained from Stratagene Europe, Netherlands (ultracompetent cells). The Vibrio cholerae strain (CIP 56.38) was obtained from Biological Resource Centres of Pasteur Institute. All these strains were cultured 007) 1280– 1286 1281 at 260 nm and after gel electrophoresis by comparison to a DNA ladder. 2.3. PCR detection of the E. coli O157 rfbE gene The rfbE gene is responsible for biosynthesis of the O157 antigen. The rfBE sequence from E. coli 0157:H7 86-24 strain was available on NCBI databanks, GenBank number S83460. By means of sequence recoveries, analysis and alignments from the rfbE gene, we estimated inter- and intra-specific homologies and differences with other bacterial strains, other genera, and other kingdoms. The rfbE gene sequence was taken from the GenBank database; using NCBI and INFOBIOGEN BLASTN (nucleotide–nucleotide blast, search for equivalent sequences in a given sequence), other bacterial genera and other close-kingdom organism sequences for this gene were also used. After sequence alignment using the Multalin multiple-sequence alignment software (Corpet, 1988), the homology and various regions were analyzed. This analysis enabled the specification of a specific gene target region. Primers were determined on the basis of this analysis. confirmed compared with 500 ng of pUC19 DNA/MspI Marker (Fermentas). 2.4. Detection sensitivity One liter of ultra-pure sterile water samples were spiked with different concentrations of E. coli O157:H7 (10-fold dilution series, 2�105–2 cells l�1). An exponential phase culture was used to make these dilutions. It was serially diluted and the number of CFU of bacteria in each dilution was determined by plating on LB agar plates. Homogenized samples were filtered through a polycarbonate membrane (Millipore, 47 mm, 0.45mm). Filtrates and membranes were re-suspended in ultra-pure sterile water (1.8 ml) and resultant solutions were used to extract the DNA. The effect of the filtration procedure was determined by processing different concentrations of E. coli O157:H7 without filtration. 500ml of different concentrations of E. coli O157:H7 O157:H7 DNA solution (extracted according to the procedure ARTICLE IN PRESS oca 676– 949– 577– 058– 509– 747– 918– 321– 574– 051– t b ain coli WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 2 8 0 – 1 2 8 61282 They were situated in the O157 specific region (see Table 1 for primers description). After designation of the primers, oligonucleotides were synthesized by Proligo SA. The set of primers thus defined was tested by BLASTN on the GenBank to estimate its specificity. The primers developed generated a 296-bp frag- ment. The targeted DNA fragment was generated using an Eppendorf Mastercycler. Two microliters of DNA template were added to a PCR mix containing 1� PCR Buffer (50 mM KCl+10 mM Tris-HCl pH 8.8+0.08% Nonidet P40), 2.5 mM of MgCl2, 0.4mM of forward primer and 0.4 mM of reverse primer, 200mM of dNTPs, and 1 ml of Taq DNA polymerase (all PCR reagents obtained from Fermentas), giving a final volume of 25ml. The program consisted of an initial template denaturation step at 94 1C for 3 min, followed by 48 cycles of denaturation at 94 1C for 15 s, primer annealing at 64 1C for 30 s, and primer extension at 72 1C for 40 s. The PCR products were analyzed by electrophoretic migration (50 V for 20 min) in 2% QA-Agarose TM gel (QBiogene) containing 0.5 mg/ml ethidium bromide (MO-BIO) in TBE 0.5� (Tris-Borate+EDTA). The amplicon size (296 bp) was Table 1 – Nucleotides sequences of primers Primers Sequences (50-30) Tm (1C) GC (%) L rfbE-for CAGGTGAAGGTGGAATGGTTGTC 57 52 rfbE-rev TTAGAATTGAGACCATCCAATAAG 51 33 GiO157-II ATTGCGCTGAAGCCTTTG 48 50 GiO157-I CGAGTACATTGGCATCGTG 51 53 1 O157F CGGACATCCATGTGATATGG 52 50 O157R TTGCCTATGTACAGCTAATCC 50 43 O157PF8 CGTGATGATGTTGAGTTG 46 44 O157PR8 AGATTGGTTGGCATTACTG 47 42 1 O157AF AAGATTGCGCTGAAGCCTTTG 52 48 O157AR CATTGGCATCGTGTGGACAG 54 55 1 a Selectivity of primer set against a selection of target and non-targe b Test of specificity of PCR assay with O157 strains and non-O157 str c Cross-reactions described by Chapman et al. (2001) with various E. d Excluding cross-reaction with Vibrio spp, not verified. e Evaluation against a more extensive strain list than Maurer et al. (199 acteria. s (Table 2). non-O157 strains. described above and quantified spectrophotometrically at 260 nm) were amplified using the two different sets of primers (for the GiO157 set of primers, using the optimized PCR conditions described by Abdulmawjood et al., 2004). 2.5. Description of the wastewater treatment plant samples Wastewaters from 44 urban wastewater treatment plants (UWWTPs) in the Gard (France) were sampled. One-liter sample were collected from the clarifiers of each waste- water-treatment plant. Upon arrival at the laboratory, all tion Gene detected False positivea Ref. 698 rfbE Nob This study 972 (296 bp) 594 rfbE Noc Abdulmawjood et al. (2004) 1076 (500 bp) 528 rfbE Yesd Paton and Paton (1998) 767 (259 bp) 935 rfbA/rfbB ) Noe Maurer et al. (1999) 1339 (422 bp 594 rfbe Yesd Desmarchelier et al. (1998) 1070 (497 bp) cells (10-fold dilution series in pure sterile water, 4�105–4 cells ml�1) were directly used to template the DNA extraction. The sensitivity of the primers designed in this study for the detection of E. coli O157: H7 DNA was compared with that of the primers designed by Abdulmawjood et al. (2004). To evaluate the relative sensitivity, dilution series of E. coli 9) and Chapman et al. (2001) including 28 Vibrio spp. samples were either analyzed immediately or stored at 4 1C prior to analysis (maximum 48 h after sampling). 3. Results 3.1. PCR template preparation The DNA quantity obtained by the lysis procedure developed in this study was higher than those obtained using other DNA isolation procedures with only Lysozyme or Proteinase K, with Guanidine Thiocyanate (Chomczynski, 1993), with only one boiling step (Abdulmawjood et al., 2004), or with different alkaline solutions (data not shown). The pre-boiling step (5 min at 100 1C) increased the DNA quantity obtained with the new procedure (around 10%). The Lysozyme and Protei- nase K incubation time at 37 1C was set at 15 min (5% is lost with 10 and 20 min of incubation time). 3.2. Detection of the rfbE gene and specificity of the PCR assay The multi-alignment analysis of the sequences of the first fifteen results described above shows all these O157 strains are identical for this gene and the rfbE gene does not exhibit polymorphism. Moreover, the rfbE gene sequences of the eight non-H7 O157 strains obtained (two O157:H16, O157:H12, O157:H-, O157:H3, O157:H38 and two O157:NM) show total equivalence except for one replacement (guanine in place of adenine in position 6651 for the O157:H- strain) and two deletions (in positions 6550 and 6557 for the O157:H38 strain). Based on this rfbE gene analysis, a specific set of primers was designed to carry out a specific identification of the E. coli O157 strains. Amplification of other bacterial strain DNA templates was performed in order to assess and verify the specificity of these primers (Table 2). The E. coli O157:H7 86-24 strain gives positive results. For all the non-0157 E. coli and the others strains no amplification was observed. 3.3. Sensitivity of the rfbE PCR assay for E. coli O157 detection in water The sensitivity of the PCR assay was assessed using different concentrations of E. coli O157:H7 cells suspended in pure ARTICLE IN PRESS bE Genotype Positive PCR Origin/disease or status pe ran WAT E R R E S E A R C H 41 (2007) 1280– 1286 1283 a n ¼ 4. b Strain was obtained from Stratagene Europe, Netherlands (Ultracom c Strain was obtained from the National Veterinary School of Lyon F Bacterial strain type tested Strain no. Serotype Enterohaemorraghic E. coli (EHEC) 43895b O157:H7 Enteropathogenic E. coli (EPEC) c Non-O157 Uropathogenic E. coli (UPEC) c Non-O157 Escherichia coli K1 (NMEC) c Non-O157 Escherichia coli XL10 d Non-O157 Vibrio cholerae CIP 56.38e Listeria monocytogenes c Staphylococcus aureus c Legionella pneumophila c Enterococcus faecium c Klebsiella pneumoniae c Streptococcus pyogenes c Pseudomonas aeruginosa c The analysis of rfbE sequence from the E. coli O157:H7 86–24 strain and its comparison with those of other closely related bacteria enabled us to observe various homologies that could exist between different bacterial strains. After the BLASTN on GenBank, the first 15 results (presenting a high level of homology, 99% and 98%) all corresponded to O157 E. coli strains. In this analysis, the first non-O157 strains belonging to other bacterial genera show a very low homology percentage (the highest homology between a non-O157 strain and the rfbE gene sequence, 80%, was found for a short part of a genome of the Yersinia genus). Table 2 – Specificity of the PCR assay for detection of the rf d Strains were obtained from the strain collection of the University Hos e Strain was obtained from the Biological Resource Centres of Pasteur In rfb rfbE + + +/+/+/+ Human/diarrhea + � �/�/�/� Human/diarrhea + � �/�/�/� Human/cystitis + � �/�/�/� Neonate/meningitis + � �/�/�/� Stratagene cells + + �/�/�/� Seawater �/�/�/� Food �/�/�/� Food �/�/�/� Network water �/�/�/� Human + �/�/�/� Human �/�/�/� Human + �/�/�/� Human tent cells). ce. water (1 l). The detection limit for cells suspended in pure water was 200 CFU l�1 (Table 3). Moreover, Table 3 shows that the filtration of the sample is not a limitation step in the sensitivity of the method because the same results were found with or without filtration. Table 4 compares the primers designed in this study with previously published primers (Abdulmawjood et al., 2004). A more sensitive detection level was obtained with the primers designed in this study (560 fg against 1125). Each experiment has been performed two times in dupli- cate (n ¼ 4). gene of Escherichia coli O157 a pital Center, Nıˆmes, France. stitute. ARTICLE IN PRESS oli s of 2� +/+ +/+ or D say 57:H7 on fentogram detected by the set of primersa 1120 560 280 +/+/+/+ +/+/+/+ �/�/�/� +/+/+/+ �/�/�/� �/�/�/� a ation project (Abdulmawjood et al., 2004). The selectivity of the primers n-O157, and 85 non-E. coli. It was shown to be highly inclusive (100%) and tection probability of approximately 2� 103 cells per reaction. Table 5 – Characterization of samples with PCR rfbE positive amplification ( 2 n ¼ 4. b Set of primers developed in a multicenter validation and standardiz was evaluated against 155 strains, including 32 E. coli O157, 38 E. coli no exclusive (100%). Under the conditions used, the assay has a 100% de Table 3 – Sensitivity of the rfbE PCR assay for detecting E. c E. coli O157:H7 Number 2�105 2�104 1 +/+/+/+ +/+/+/+ 2 +/+/+/+ +/+/+/+ 1—E. coli O157:H7 cells suspended in 1 l pure water, filtered and used f ultra pure sterile water directly used for DNA template extraction. a n ¼ 4. Table 4 – Comparison of rfBE PCR assay and GiO157 PCR as Set of primers DNA quantity of E. coli O1 4500 2250 rfbE (this study) +/+/+/+ +/+/+/+ GiO157b +/+/+/+ +/+/+/+ WAT E R R E S E A R C H 4 11284 3.4. Application: detection of E. coli O157 in wastewater The detection of E. coli O157 was carried out in 44 clarifier samples from UWWTPs (Table 5). Seven percent (3/44) of the samples gave a positive result using the rfbE PCR-based method (measurement in duplicate). 4. Discussion The detection of E. coli O157 is an important environmental issue because of the health problems that can be involved. In order to achieve successful detection, we tried to develop a rapid and sensitive method based on gene amplification. We focussed our method on the amplification of the rfbE gene to detect the presence of all E. coli O157 strains. The O157 rfbE PCR assay developed in this study is a selective and rapid method for confirmation of the O157 serogroup. The assay to determine the selectivity of the primers was 100% inclusive and 100% exclusive for all the target and non-target strains, respectively (Table 2). Other tests need to be performed, such as PCR on a more extensive strain list (Abdulmawjood et al., 2004), to confirm the high selectivity of the rfbE primer set designed in this study. Indeed, all sets of published primers designed to detect O157 strains have cross-reactions with non-target strains (Table 1). Chapman et al. (2001) found that the primers of Des- marchelier et al. (1998) and also Paton and Paton (1998) have cross-reactions with non-O157 strains. The rfbE gene is described in E. coli O157 and also in Vibrio spp. These two genes have high homologies (Bilge et al., 1996). It is necessary O157:H7 cells detected by rfbE PCR assaya 103 2�102 2�101 2 /+/+ +/+/+/+ �/�/�/� �/�/�/� /+/+ +/+/+/+ �/�/�/� �/�/�/� NA template extraction.2—E. coli O157:H7 cells suspended in 500ml of sensitivity on E. coli O157:H7 DNA 0 0 7 ) 1 2 8 0 – 1 2 8 6 to evaluate the selectivity of the set of primers designed in the rfbE region with a Vibrio spp. strain list (Abdulmawjood et al., 2004). The primers designed here does not give cross-reaction with the strain Vibrio cholerae (Serogroup non-O:1, non-O:139). These primers study can be compared with the rfbE primers previously published (Table 1). The forward oligonucleotides have higher fusion tempera- ture than other primers (57 1C), which enables a high annealing temperature (64 1C) to be chosen for the optimized PCR conditions, increasing specificity by decreasing non- specific matching. Comparison of PCR sensitivity with primers published by Abdulmawjood et al. (2004) on DNA dilutions of only E. coli O157:H7 86-24 strain showed greater sensitivity for the primers designed in this study (all 0157 strains were tested by Abdulmawjood et al., 2004). PCR assays based on the rfbE gene enabled detection of all E. coli O157 strains. All O157 strains, both Shiga-like-toxin (stx) producing WWTP Capacity (population-equivalent) No. of samples tested PCR rfbE positivesa p5000 18 0 45000 et p10,000 16 1 410,000 et p20,000 7 1 420,000 3 1 Total 44 3 (7%) a n ¼ 2. UWWTPs in the Gard (in discharged effluent). A prevalence of in ground beef. J. Appl. Microbiol. 95 (1), 129–134. ARTICLE IN PRESS (2 7% (X200 CFU l�1) was determined for these pathogens, although there are no installations usually able to produce these bacteria (slaughterhouses, meat renderers, farms or the food processing industry). The presence of the O157 strains could therefore stem from bacterial excretion in human faeces. Table 4 suggests that the probability of E. coli 0157 detection in wastewater increase with the WWTP capacity but the lack of data preclude to discuss this hypothesis. This could have been tested by increasing the number of high- capacity WWTP samples. The data in this study agree with the results of studies concerning STEC prevalence in the environment (Vernozy-Rozand et al., 2002), but information about the presence of these bacteria in wastewater is lacking. This correlates with the study of Arana et al. (2001) which shows that the E. coli survival period in wastewater can be 21 day and that these bacteria can therefore occur in treated effluent, sewage sludge. Consequently, soils and surface or ground water can be contaminated, causing the pathogens to be maintained in livestock herds and eventually enabling transmission to humans. As all DNA-based detection methods, this assay might detect DNA from dead or non-growing cells. But DNA derived from dead cells (from environment with high microbial activity, like wastewater), will be degraded rapidly suggesting that interference by DNA from these cells might be of less importance (England et al., 1998; Schena and Ippolito, 2003). 5. Conclusions The method developed in this study enabled the detection of 200 cells per reaction (200 CFU) of E. coli O157 in pure water without enrichment. Depending on the presence of inhibitors, this threshold might be higher in wastewater and the use of an internal control is indeed necessary. The infectious dose of E. coli O157 is as low as o10 cells (Keene et al., 1994). Therefore, the PCR should not be used directly on environmental samples without an appropriate pre-enrich- ment step. But the limit of detection without enrichment in this study (tested on E. coli O157:H7 cells only) is better than the 104 cells of Maurer et al. (1999) and the 2�103 cells of Abdulmawjood et al. (2004) and enables a shorter detection strains and strains which do not produce stx genes, gave positive signals. PCR assays for detecting stx genes (Paton and Paton, 1998) can differentiate between the stx producing strains and the non-stx producing strains, enabling a toxi- genic genotype to be confirmed. False-negatives can occur in a PCR assay. To enable the detection of these false-negative results an internal amplification control (IAC) is necessary (Abdulmawjood et al., 2002), and should be integrated in the diagnostic PCR of E. coli O157. A DNA extraction step was implemented before the PCR analysis, giving a good quantity and quality of DNA compared to other simple procedures usually used for extracting nucleic acids. The method was applied to the detection of E. coli O157 in wastewater samples. Spot sampling was carried out on 44 WAT E R R E S E A R C H 41 time to be envisaged, even if an extra step needs to be added. Acknowledgments We thank Dr. Christine Vernozy-Rozand, National Veterinary School of Lyon, France for providing the E. coli O157:H7 strain and Dr. Jean-Philippe Lavigne, University Hospital Center, Nıˆmes, France, for providing the strains list used for primer specificity evaluation. We also thank SAUR France, Ge´ne´rale des Eaux, BRL Exploitation, SDEI, RUAS, and SATESE (French WWTP operating companies) for allowing us to collect wastewater samples from their plants. R E F E R E N C E S Abdulmawjood, A., Roth, S., Bulte, M., 2002. Two methods for construction of internal amplification controls for the detec- tion of E. coli O157 by polymerase chain reaction. Mol. Cell Probes 16 (5), 335–339. Abdulmawjood, A., Bulte, M., Roth, S., Schonenbrucher, H., Cook, N., Heuvelink, A.E., Hoorfar, J., 2004. Development, validation, and standardization of polymerase chain reaction-based detection of E. coli O157. J. AOAC Int. 87 (3), 596–603. Akashi, S., Joh, K., Tsuji, A., Ito, H., Hoshi, H., Hayakawa, T., Ihara, J., Abe, T., Hatori, M., Mori, T., et al., 1994. A severe outbreak of haemorrhagic colitis and haemolytic uraemic syndrome associated with E. coli O157:H7 in Japan. Eur. J. Pediatr. 153 (9), 650–655. Anonymous, 2000. Waterborne outbreak of gastroenteritis asso- ciated with a contaminated municipal water supply, Walk- erton, Ontario, May–June 2000. Can. Commun. Dis. Rep. 26 (20), 170–173. Arana, I., Justo, J.I., Muela, A., Barcina, I., 2001. Survival and plasmid transfer ability of E. coli in wastewater. Water Air Soil Pollut. 126, 223–238. Bilge, S.S., Vary Jr., J.C., Dowell, S.F., Tarr, P.I., 1996. Role of the E. coli O157:H7 O side chain in adherence and analysis of an rfb locus. Infect. Immun. 64 (11), 4795–4801. Bitzan, M., Ludwig, K., Klemt, M., Konig, H., Buren, H., Muller- Wiefel, D.E., 1993. The role of E. coli O 157 infections in the classical (enteropathic) haemolytic uraemic syndrome: results of a Central European, multicentre study. Epidemiol. Infect. 110 (2), 183–196. Castonguay, M.H., van der Schaaf, S., Koester, W., Krooneman, J., van der Meer, W., Harmsen, H., Landini, P., 2006. Biofilm formation by E. coli is stimulated by synergistic interactions and co-adhesion mechanisms with adherence-proficient bac- teria. Res. Microbiol. 157 (5), 471–478. Chapman, P.A., Cornell, J., Green, C., 2000. Infection with verocytotoxin-producing E. coli O157 during a visit to an inner city open farm. Epidemiol. Infect. 125 (3), 531–536. Chapman, P.A., Ellin, M., Ashton, R., Shafique, W., 2001. Compar- ison of culture, PCR and immunoassays for detecting E. coli O157 following enrichment culture and immunomagnetic separation performed on naturally contaminated raw meat products. Int. J. Food Microbiol. 68 (1–2), 11–20. Chomczynski, P., 1993. A reagent for the single-step simultaneous isolation of RNA, DNA and proteins from cell and tissue samples. Biotechniques 15 (3), 532–534, 536-7. Corpet, F., 1988. Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res. 16 (22), 10881–10890. Cui, S., Schroeder, C.M., Zhang, D.Y., Meng, J., 2003. Rapid sample preparation method for PCR-based detection of E. coli O157:H7 007) 1280– 1286 1285 Desmarchelier, P.M., Bilge, S.S., Fegan, N., Mills, L., Vary Jr., J.C., Tarr, P.I., 1998. A PCR specific for E. coli O157 based on the rfb locus encoding O157 lipopolysaccharide. J. Clin. Microbiol. 36 (6), 1801–1804. DuPont, H.L., 1995. Pathogenesis of traveler’s diarrhea. Che- motherapy 41 (Suppl 1), 33–39. Edberg, S.C., Rice, E.W., Karlin, R.J., Allen, M.J., 2000. E. coli: the best biological drinking water indicator for public health protec- tion. Symp. Ser. Soc. Appl. Microbiol. (29), 106S–116S. England, L.S., Holmes, S.B., Trevors, J.T., 1998. Persistence of viruses and DNA in soil. World J. Microbiol. Biotechnol. 14 (2), 163–169. Gagliardi, J.V., Karns, J.S., 2002. Persistence of E. coli O157:H7 in soil and on plant roots. Environ. Microbiol. 4 (2), 89–96. Gerber, A., Karch, H., Allerberger, F., Verweyen, H.M., Zimmer- hackl, L.B., 2002. Clinical course and the role of shiga toxin- producing E. coli infection in the hemolytic-uremic syndrome in pediatric patients, 1997–2000, in Germany and Austria: a prospective study. J. Infect. Dis. 186 (4), 493–500. Haeghebaert, S., Vaillant, V., Decludt, B., Grimont, P.A., 2000. Surveillance of haemolytic uraemic syndrome in children under 15 years of age in France in 1998. Eur. Surveill. 5 (6), 68–73. Keene, W.E., McAnulty, J.M., Hoesly, F.C., Williams Jr., L.P., Hedberg, K., Oxman, G.L., Barrett, T.J., Pfaller, M.A., Fleming, D.W., 1994. A swimming-associated outbreak of hemorrhagic colitis caused by E. coli O157:H7 and Shigella sonnei. New Engl. J. Med. 331 (9), 579–584. Lingwood, C.A., Law, H., Richardson, S., Petric, M., Brunton, J.L., De Grandis, S., Karmali, M., 1987. Glycolipid binding of purified and recombinant E. coli produced verotoxin in vitro. J. Biol. Chem. 262 (18), 8834–8839. Maurer, J.J., Schmidt, D., Petrosko, P., Sanchez, S., Bolton, L., Lee, M.D., 1999. Development of primers to O-antigen biosynthesis genes for specific detection of E. coli O157 by PCR. Appl. Environ. Microbiol. 65 (7), 2954–2960. Olsen, S.J., Miller, G., Breuer, T., Kennedy, M., Higgins, C., Walford, J., McKee, G., Fox, K., Bibb, W., Mead, P., 2002. A waterborne outbreak of E. coli O157:H7 infections and hemolytic uremic syndrome: implications for rural water systems. Emerg. Infect. Dis. 8 (4), 370–375. Paton, A.W., Paton, J.C., 1998. Detection and characterization of Shiga toxigenic E. coli by using multiplex PCR assays for stx1, stx2, eaeA, enterohemorrhagic E. coli hlyA, rfbO111, and rfbO157. J. Clin. Microbiol. 36 (2), 598–602. Paunio, M., Pebody, R., Keskimaki, M., Kokki, M., Ruutu, P., Oinonen, S., Vuotari, V., Siitonen, A., Lahti, E., Leinikki, P., 1999. Swimming-associated outbreak of E. coli O157:H7. Epidemiol. Infect. 122 (1), 1–5. Schena, L., Ippolito, A., 2003. Rapid and sensitive detection of Rosellinia necatrix in roots and soils by real time Scorpion-PCR. J. Plant Pathol. 85 (1), 15–25. Tarr, P.I., 1995. E. coli O157:H7: clinical, diagnostic, and epidemio- logical aspects of human infection. Clin Infect Dis 20 (1), 1–8 quiz 9–10. Vernozy-Rozand, C., Montet, M.P., Lequerrec, F., Serillon, E., Tilly, B., Bavai, C., Ray-Gueniot, S., Bouvet, J., Mazuy-Cruchaudet, C., Richard, Y., 2002. Prevalence of verotoxin-producing E. coli (VTEC) in slurry, farmyard manure and sewage sludge in France. J. Appl. Microbiol. 93 (3), 473–478. Wang, G., Zhao, T., Doyle, M.P., 1996. Fate of enterohemorrhagic E. coli O157:H7 in bovine feces. Appl. Environ. Microbiol. 62 (7), 2567–2570. ARTICLE IN PRESS WAT E R R E S E A R C H 4 1 ( 2 0 0 7 ) 1 2 8 0 – 1 2 8 61286 Evaluation of enrichment-free PCR-based detection on the rfbE gene of Escherichia coli O157--Application to municipal wastewater Introduction Materials and methods Bacterial strains and culture PCR template preparation PCR detection of the E. coli O157 rfbE gene Detection sensitivity Description of the wastewater treatment plant samples Results PCR template preparation Detection of the rfbE gene and specificity of the PCR assay Sensitivity of the rfbE PCR assay for E. coli O157 detection in water Application: detection of E. coli O157 in wastewater Discussion Conclusions Acknowledgments References