- Email: [email protected]
Characterization of Shiga toxin producing E. coli and O157 serotype E. coli isolated in France from healthy domestic cattle
International Journal of Food Microbiology 63 (2001) 217–223 www.elsevier.nl / locate / ijfoodmicro
Characterization of Shiga toxin producing E. coli and O157 serotype E. coli isolated in France from healthy domestic cattle ´ Franc¸ois Rogerie a , Armelle Marecat a , Stephanie Gambade b , Francis Dupond a , b a, Pierre Beaubois , Marc Lange * a
´ ` de Microbiologie des Ecosystemes , 1 Rue du Professeur Calmette, BP 245 -59019 Institut Pasteur de Lille, Service R& D, Departement Lille Cedex, France b SOCOPA Villefranche, 03430 Villefranche d’ Allier, France Received 13 November 1999; received in revised form 31 July 2000; accepted 12 August 2000
Abstract A study was carried out in France in collaboration with the meat industry to investigate the occurrence and characteristics of Shiga toxin-producing E. coli (STEC) and O157 E. coli in a population of healthy bovines representative of French livestock. A total of 851 animals belonging to three bovine classes (106 young bulls, 374 dairy cows and 371 meat cows) were included in the study. Samples of feces and of the corresponding carcasses were collected from March 97 to August 97 in seven abattoirs spread throughout the national territory. STEC cultures from the 1702 samples were screened using PCR for the presence of stx genes. Positive samples were further subjected to colony blot hybridization and to O157-specific immunomagnetic separation. Probe-positive colonies and O157 colonies were then analyzed for the presence of virulence genes and phenotypic characters (serotype, Stx production). In 154 (18.1%) feces and 91 (10.7%) carcass samples stx genes were detected. Two hundred and twenty-two STEC colonies were isolated from 67 (7.9%) feces and 16 (1.9%) carcass samples, with 183 STEC isolated from feces and 39 from carcasses. Only eight O157 isolates were collected from feces samples. None of these O157 E. coli isolates presented stx genes and thus could not be considered as pathogenic regarding hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS). In 3.2% of STEC isolated from feces and in 10.2% of STEC from carcasses eae genes were detected. In 17% of STEC from feces and in 30.7% from carcasses ehx genes were detected. Using these data, the 222 STEC colonies could be classified in 11 different ‘virulence patterns’ (presence / absence of stx1, stx2, eae and ehx genes), showing that more than 77% of isolates presented only one virulence factor. Only three STEC on 222 colonies (1.3%) presented the three virulence factors stx, eae and ehx in association, none of them reacting with antisera specific for enterohemorrhagic E. coli. (EHEC). These data, together with the fact that only five isolates on the 222 (2.2%) reacted with such antisera (three O111 and two O26 isolates) demonstrated that the natural bacterial populations isolated during this study were clearly distinct from EHEC. 2001 Elsevier Science B.V. All rights reserved. Keywords: France; Healthy cattle; Epidemiology; Shiga toxin producing E. coli
*Corresponding author. E-mail address: [email protected] (M. Lange). 0168-1605 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0168-1605( 00 )00422-0
218
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
1. Introduction Enterohemorrhagic Escherichia coli (EHEC) were recognized as important human pathogens in the US in 1982 (Riley et al., 1983). EHEC strains are defined by a combination of the symptoms they produce and the virulence factors they possess. They are associated with a variety of clinical manifestations, watery diarrhea, hemorrhagic colitis (HC) that may lead to more severe complications, hemolytic uremic syndrome (HUS) and thrombotic thrombocytopenic purpura. EHEC infections are mainly caused by E. coli belonging to the O157:H7 serotype, which has been implicated in large foodborne outbreaks all over the world (USA, Canada, Scotland and Japan, Nataro and Kaper, 1998). Other EHEC serotypes (0111, O26, O103, etc.) have also been associated with small to large outbreaks of HC or with human sporadic infections (Johnson et al., 1996). EHEC strains associated with human disease have genetic properties in common. The first is that they possess the stx genes coding for the Shiga toxins Stx1 and / or Stx2, found on temperate lambdoid bacteriophages into the bacterial chromosome. The second is that most EHEC strains have additional virulence factors, mainly the eae gene coding for intimin, located in the LEE locus associated to the Attaching and Effacing phenotype, and the ehx gene located on a large plasmid (90 kb), coding for the synthesis of entero-hemolysin. These three virulence genes stx, eae and ehx are found in association in most of clinical isolates (Barrett et al., 1992; Karch et al., 1997; Gyles et al., 1998). The natural population of Shiga toxin-producing E. coli (STEC) found in the intestine of domestic animals is considered as the natural reservoir of EHEC, and cattle have been implicated as the source of EHEC pathogens in a number of foodborne outbreaks associated with the consumption of undercooked ground beef or raw milk products (Nataro and Kaper, 1998). In Europe, surveys made to estimate the prevalence and properties of STEC in non-diarrheic healthy domestic animal species (Beutin et al., 1993, 1995, 1997; Blanco et al., 1996, 1997), revealed that cattle frequently harboured such STEC strains. These studies, performed on selected herds or populations of farm animals, have shown that most of the STEC bovine isolates lacked the
main additional virulence factor eae and ehx, and thus would not represent a potential danger to human health. However, in a recent study on 10 Dutch dairy farms focused on STEC strains of 0157 serotype, Heuvelink et al. (1998a) revealed that most of the isolates possessed the eae virulence gene. In France, no data were available until now on the fate and characteristics of STEC isolated from healthy domestic cattle. The aim of the present study, initiated by the meat Industry, was to determine the prevalence of STEC and O157 E. coli isolated from feces of a selected population of healthy bovines representative of the entire French livestock, and also to compare their phenotypic, serological and genetic traits to those characteristic of EHEC. Concurrently, the corresponding carcasses were submitted to the same microbiological analysis in order to assess quality of the slaughtering procedures.
2. Material and methods
2.1. Collection of samples and selective enrichment The size of the sampling has been fixed to 851 animals, which is 1 / 10 000 of the total number of bovines bred in France in 1997. Sampling was carried out on three categories of animals in proportions representative of the French breeding system, with 106 (12.5%) young bulls, 374 (44%) dairy cows and 371 (43.5%) meat cows. Samples, collected from March 1997 to August 1997, were taken weekly from seven abattoirs spread throughout the national territory. At each sampling site, the animals were selected randomly and identified by a number from one to 851. The numbers were made to trace the bovine class, the birth location, and the week and site of slaughtering. Immediately after slaughter, samples of the rectal contents were collected aseptically and placed into sterile plastic pots. The corresponding samples of carcasses, also identified by a number from one to 851, were collected after chilling the day after. In each case about 25 g of meat were taken by excision in three different areas (neck, shoulder and backside) and placed into sterile bags. Samples, kept cold with melting ice, were sent to the laboratory by express transport, and stored at 48C prior to microbiological analysis, which started
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
within 72 to 96 h after sampling, as previously described (Uyttendaele et al., 1998). One gram of feces was used to inoculate 10 ml of modified E. coli broth (mEC) (Merck, Nogent sur Marne, France) supplemented with novobiocin (20 mg / l, Sigma, St. Quentin Fallavier, France). The 25 g of carcass meat were added to 225 ml of the same medium mEC 1 n, and homogenized in a stomacher for 1 min at low speed. Feces and carcass enrichment cultures were then incubated for 18 h at 378C.
2.2. PCR-based detection of stx genes One milliliter of enrichment culture was transferred to a 1.5-ml microtube and centrifuged 3 min at 12 000 3 g. The supernatant was removed and the pellet was washed in PBS buffer (pH 7.4, Sigma). The DNA from pelleted cells was released by boiling and purified using Instagen DNA purification matrix (Biorad, Ivry sur Seine, France). Ten microliters of the DNA were used as template for PCR detection of stx genes. Oligonucleotides ES149 and ES151 described by Read et al. (1992) amplified a conserved sequence of the stx1 and stx2 genes. The conditions for the PCR amplification were the same as those described by Uyttendaele et al. (1998).
2.3. Isolation of STEC by selective plating and colony hybridization with stx gene probes When PCR products of the expected size were obtained, STEC were tentatively isolated by colony blot hybridization, either directly from feces or from feces and carcass enrichment culture. Direct isolation was made from 1 to 2 g of feces suspended in physiological solution. One hundred microliters (v / v) were plated onto Plate count agar (PCA, Difco laboratories, Osi, Maurepas, France) supplemented with 1.5 g / l of bile salts No. 3 (Difco), and incubated 24 h at 42.58C. Ten microliters of feces or carcass enrichment were streaked on plates and incubated overnight at 42.58C. In each case, after growth, plates containing from 50 to 100 colonies were replicated onto Nylon membrane of 82 mm of diameter (Pall-Gelman, Champ sur Marne, France). Lysis of bacterial cells and binding of DNA to Nylon membrane were achieved as described by Sambrook et al. (1989), except that prior to DNA fixation, the membrane was treated 1 h at 378C with Proteinase K
219
(2 mg / ml, Boehringer-Mannheim, Meylan, France). DNA probes specific for stx genes were generated with the PCR DIG probe synthesis kit (BoehringerMannheim). Two degenerate primers 772 and 849 described by Karch and Meyer (1989) were used to amplify DIG labeled stx genes from reference strains LCDCH19 and LCDC902380 (Laboratory Center for Disease Control, Ottawa, Canada), where the stx1 and the stx2 genes are present, respectively. The hybridization, washing and detection reactions were performed according to the manufacturer’s instructions.
2.4. Isolation of O157 by IMS Standard IMS with magnetic beads coated with ` antibody to O157 antigen (Dynal, Compiegne, France) was performed with 1 ml of enrichment culture where stx genes were detected. The concentrates were inoculated onto sorbitol–MacConkey agar (Difco) supplemented with cefixime (0.05 mg / l) and potassium tellurite (2.5 mg / l), to test the sorbitol fermentation ability. Concentrates were also plated onto chromagar O157 (Dynal) to examine the b-glucuronidase activity. The sorbitol non-fermenting and / or b-glucuronidase negative colonies were tested for agglutination with an E. coli O157 latex test kit (Oxoid, Basingstoke, UK). Positive isolates were tested for the H7 antigen by agglutination (Difco) and transferred in mannitol-nitrate-motility medium (Sanofi, Marnes la Coquette, France) to check for their motility.
2.5. Biochemical traits of isolates STEC and O157 were inoculated on KliglerHajna, indole-urea, Rapid’E. coli (Difco) and SMAC (Sanofi) media and incubated 24 h at 378C. Isolates with atypical combination of characters were further confirmed by using API20E biochemical test strips ´ (Biomerieux, Marcy l’Etoile, France).
2.6. Serotyping of STEC O grouping was carried out by slide agglutination of living bacteria with antisera specific for the major enterohemorrhagic E. coli groups O26, O111 (Sanofi) and O103 (provided by Dr. F. Scheutz, Copenhagen, Denmark). O157 and H7 determination
220
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
were performed with Difco antisera. Reference strains for serotype determination were kindly pro´ vided by Professor E. Bingen (Hopital Robert Debre, Paris, France).
2.7. Virulence traits of isolates All isolates were investigated for production of Shiga toxins (Stx) by Vero cell toxicity as described by Konowalchuck et al. (1977). Stx release was induced by Polymixin B treatment as described by Evanz et al. (1974). The presence of stx1 and stx2 genes was detected by PCR with conserved ES149– ES151 primers coupled to dot hybridization with oligonucleotides specific to stx1 or to stx2 gene. Two oligonucleotides were designed within the 323 bp amplified from primers ES149–ES151. Oligo HSLTI (59-CATTGTCTRGTGACAGTAGCT-39) is specific to stx1, and oligo SLTIIF (59-CCATGACRACGGACAGCAGTT-39) is specific to the different variants of stx2 gene. STLIIF was chosen after sequence analyzing of all known stx2 gene variants and derived from a conserved region. STLIIF was further tested with our own collection of EHEC E. coli. The labeling of oligos and the hybridizationdetection reactions were performed according to the manufacturer’s recommendations of the DIG oligonucleotide tailing kit (Boehringer-Mannheim). The presence of the eae gene was examined using specific primers eae1 /eae2 described by Germani (1996). These primers are derived from conserved regions of the eae gene and allow amplification of a 490-bp DNA fragment. Isolates were additionally tested for the presence of the ehx gene by PCR with the primers hlyA1 / hlyA4 described by Schmidt et al. (1995).
3. Results and discussion
3.1. Prevalence and characteristics of natural colonizing STEC isolated from healthy domestic cattle in France 3.1.1. Virulence genes The stx genes were detected in 154 (18.1%) feces and 91 (10.7%) carcass samples; stx-positive colonies were obtained from 67 (7.9%) feces and 16 (1.9%) carcass samples. From these 83 samples, 22
allowed isolation of only one colony. As an average of two to four STEC colonies were taken from the other 61 positive samples, a total of 222 STEC colonies were collected, 183 from feces and 39 from carcasses. No clear correlation was found between PCR or STEC isolation data and the slaughterhouse’s locations, the date of birth of animals or with any other parameters linked to the breeding of animals (results not shown). Only 195 (87%) colonies of the 222 produced an active Shiga toxin (results not shown). Such a lack of expression of stx genes has already been observed by Heuvelink et al. (1996) in STEC isolated from retail raw meats. However, all of the 222 isolates were considered as STEC and we have thus estimated the prevalence of STEC into French domestic cattle to be 18.1% in feces (10.7% in carcasses) using stx PCR data and 7.9% in feces (1.9% in carcasses) based on isolation data (yield of isolation is around 40%, results not shown). The lower values observed for carcasses (10.7% using PCR data and 1.9% using isolation data) may be indicative of the quality of the slaughtering procedures used in the seven abattoirs selected in this study. These data are however difficult to compare with those of other European studies because obtaining them may be greatly influenced by factors such as the target population, the sampling strategy, the methodology of screening and isolation and also by possible seasonal variation in prevalence. In fact, our study showed that the carriage rate of VTEC E. coli in bovines was not influence by the birth, breeding or abattoir place, by the abattoir season or by the aged groups (data not shown). Nevertheless, our estimation using stx PCR data lies within the European average, where the carriage rate of STEC in cattle was shown to range from 11 to 21% (Caprioli and Tozzi, 1998). Each of the 222 STEC isolates was studied for the presence of the different virulence genes (stx1, stx2, eae and ehx genes). This typing scheme allowed us to differentiate the 222 colonies in 11 distinct ‘virulence patterns’ (Table 1). The characterization of stx genes by oligonucleotide-typing showed that STEC colonies with only one virulence gene (stx1 or stx2 or stx1 / 2) represented more than 77% of the isolates. A majority of STEC colonies harboured stx2 gene alone (60% of STEC). Beutin et al. (1997) have also shown that stx2 positive STEC were more
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223 Table 1 Virulence patterns of STEC isolated from healthy cattle Gene(s)
stx1 stx2 stx1 stx2 stx1 ehx stx2 ehx stx1 stx2 ehx stx1 eae stx2 eae stx1 stx2 eae stx1 eae ehx stx2 eae ehx
No. of STEC from (%) Feces (n 5 183)
Carcass (n 5 39)
37 (20) 98 (53) 14 (7.6) 2 (1) 16 (9) 10 (5) 2 (1) 0 (0) 1(1) 1 (1) 2 (2)
3 (7.7) 19 (48) 1 (2.5) 0 (0) 2 (2.5) 10 (25) 2 (5) 2 (5) 0 (0) 0 (0) 0 (0)
frequently isolated from healthy cattle. Furthermore, we observed differences concerning the main type of stx gene harboured by STEC from bovines of different age categories. stx2 gene was the dominant Shiga toxin coding gene in STEC from feces of adult cows (68%), and stx1 gene was slightly more abundant in STEC from young animals (52%). Similar observations were made in Spain (Blanco et al., 1996, 1997), where adult cows preferentially harboured stx2-positive STEC and young cows stx1positive STEC, suggesting that the age of animals may have an influence on the dynamic of STEC colonization. As also shown in Table 1, the proportion of colonies with the eae gene is low, 3.2% in feces and 10% in carcass. These values are in good agreement with the value observed for European healthy cattle (Beutin et al., 1997) or for STEC isolated from raw ´ meat (Pierard et al., 1997). The low prevalence of the eae gene is characteristic of STEC isolated from healthy domestic cattle. Conversely, eae genes were found very frequently in STEC from diarrheagenic cattle or infected humans (Wieler et al., 1996; Gyles et al., 1998). On the other hand, we found a higher proportion of STEC isolates with the ehx gene (17% in feces, 30.7% in carcasses). These values are much lower than that (51%) published in a study by Gyles et al. (1998) investigating the occurrence of the stx gene in bovine STEC strains belonging to serotypes less commonly or not associated with disease in humans (in our study, three STEC O111 and two
221
O26 have been isolated, none of them possessing the ehx gene, see below). Only three STEC isolated from three different samples of feces were characterized by the presence in association of the three virulence markers, none of these colonies reacting with the antisera specific for the main EHEC serotypes — O157, O103, O111 and O26 — encountered in France and Europe (Decludt, 1997, Caprioli and Tozzi, 1998).When similar typing schemes were made with STEC commonly associated with disease in humans, the majority of the isolates were characterized by the presence of stx, eae and ehx genes in association (Barrett et al., 1992). In a study by Karch et al. (1997) on children with HUS in Germany, the data demonstrated that 94% of the STEC from patients with HUS were positive for the presence of eae and ehx genes. In the same way, Gyles et al. (1998) have studied a collection of human enterohemorrhagic E.coli and shown that 89% of them harboured ehx genes and 92% the eae gene. Finally, we could also compare colonies isolated from the same sample (61 out of 83 samples are concerned) and thus observed that isolates from 14 of theses samples had different virulence patterns. These results revealed that cattle could harbour at the same time different STEC strains, regarding their virulence patterns. Besides, when colonies were obtained from feces and carcass of the same animal (only two animals were concerned), they were characterized by different virulence types, with the exception of two STEC. However, these two colonies could be discriminated by ribotyping analysis (data not shown).
3.1.2. Phenotypic traits All STEC were confirmed to be E. coli by biochemical identification with the exception of three isolates. Two isolates were characterized as E. fergusonii and one as Citrobacter freundii. The O grouping was carried out by choosing the serotypes which were mostly associated with HUS in France (Decludt, 1997) as well as in North America and Australia (Nataro and Kaper, 1998). Thus, serotyping of the 222 STEC colonies showed that none of them was of the O157 serotype. Besides, only five colonies (2.2%) reacted with three other selected antisera (O103, O111 and O26) specific for enterohemorrhagic E. coli. Three isolates from three
222
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
different samples of feces reacted with the O111 antiserum. None of these three isolates presented the eae gene, which is present in all the O111 STEC commonly associated with disease in human (Gyles et al., 1998). The two other isolates from the same carcass, were serotyped O26 and harboured the eae gene, not the ehx gene. Additional molecular typing data for comparison of these isolates to O26 clinical isolates are now needed to give a clear conclusion on their possible pathogenicity. However they have been isolated in only one sample among the 1702 analyzed in the study.
3.2. Prevalence and characteristics of natural colonizing O157 from healthy domestic cattle in France The O157 STEC were searched by the IMS method from feces and carcass enrichment positive for the presence of stx genes. Eight O157 colonies were isolated from four different samples of feces, none from carcasses.
3.2.1. Virulence genes The eight O157 isolates were tested for stx genes by PCR and displayed none of them. This results was confirmed by colony hybridization with cloned stx genes as described in Section 2 (data not shown). Thus these isolates could not be considered as pathogenic regarding HC and HUS. This situation is clearly different from that described by Heuvelink et al. (1998a) in Dutch dairy farms where the majority of the O157 strains isolated from cattle were stx1, stx2 and eae positive. Five of the eight O157 isolates had no virulence gene at all. Two strains were eae and ehx positive, while the other one was only eae positive. 3.2.2. Phenotypic traits Of the eight isolates, one had the ability to ferment sorbitol and five showed a b-glucuronidase activity. Three strains displayed the phenotype generally associated to pathogenic O157 / H7 strains, i.e., they were negative for sorbitol fermentation and bglucuronidase activity. Only one strain was positive for the H7 antigen. The other seven strains, with only one non-mobile, were not tested for other H antigens as they were not displaying stx genes. As four of the 154 tested feces samples allowed isolation of O157,
we could calculate that such O157 are present in 2.6% of feces samples, assuming that O157 are equally distributed in all feces samples, positive or negative for the presence of stx genes. This value lies in the range of 0–10% that has been estimated for European (Caprioli and Tozzi, 1998) and North American cattle (Armstrong et al., 1996).
4. Conclusion In this study, the first global survey ever performed in France of the STEC bovine reservoir, we have shown that the prevalence of STEC and O157 E. coli in French cattle are similar to the reported prevalence in European and North American healthy cattle. More importantly, the genotypic and phenotypic analysis of the 222 STEC and of the eight O157 isolates has shown that these natural populations are clearly distinct from pathogenic EHEC strains. However, the present survey may only be representative of the risk of EHEC contamination at the precise period of investigation. In fact, it was shown in follow up studies (Heuvelink et al., 1998b; Shere et al., 1998) that carriage of a given STEC strain by cattle may be intermittent and of short duration. It is thus advisable to further survey the STEC bovine reservoirs following procedures based on the characterization of isolates with genotypic (virulence traits) and phenotypic (serotype) markers.
Acknowledgements This work was supported financially by the Association Nationale Interprofessionnelle des Viandes ´ et du Betail (INTERBEV), by the Office National Interprofessionnel des Viandes de l’Elevage et de l’Aviculture (OFIVAL). We would like to thank Dr. Barbara DUFOUR, from the Agence Franc¸aise de ´ ´ Sanitaire des Aliments (AFSSA) and Dr. Securite Corinne GRANGETTE (Institut Pasteur de Lille) for helpful discussions.
References Armstrong, G.L., Hollingsworth, J., Morris, Jr. J.G., 1996. Emerging foodborne pathogens: Escherichia coli O157:H7 as a
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223 model of entry of a new pathogen into the food supply of the developed world. Epidemiol. Rev. 18, 29–51. Barrett, T.J., Kaper, J.B., Jerse, A.E., Wachsmuth, I.K., 1992. Virulence factors in Shiga-like toxin producing Escherichia coli isolated from humans and cattle. J. Infect. Dis. 165, 979–980. ¨ Beutin, L., Geier, D., Steinruck, 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. Beutin, L., Geier, D., Zimmermann, S., Karch, H., 1995. Virulence markers of Shiga-like toxin-producing Escherichia coli strains originating from healthy domestic animals of different species. J. Clin. Microbiol. 33, 631–635. Beutin, L., Geier, D., Zimmermann, S., Aleksic, S., Gillespie, H.A., Whittam, T.S., 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. Blanco, M., Blanco, J.E., Blanco, J., Blanco, E.A., Gonzalez, E.A., Mora, A., Prado, C. et al., 1996. Prevalence and characteristics of Escherichia coli serotype O157:H7 and other verotoxin-producing E. coli in healthy cattle. Epidemiol. Infect. 117, 251–257. Blanco, M., Blanco, J.E., Blanco, J., Mora, A., Prado, C., Alonso, M.P. et al., 1997. Distribution and characterization of faecal verotoxin-producing Escherichia coli (VTEC) isolated from healthy cattle. Vet. Microbiol. 54, 309–319. Caprioli, A., Tozzi, A.E., 1998. Epidemiology of Shiga toxinproducing Escherichia coli infections in continental Europe. In: Kaper, J.B., O’Brien, A.D. (Eds.), Escherichia Coli O157:H7 and Other Shiga Toxin-producing E. coli Strains. American Society for Microbiology, Washington, DC, pp. 38–48. ´ ´ Decludt, B., June 1997. Syndromes hemolytiques et uremiques en ´ ´ France, epidemiologie et agents responsables (April 1995– ´ March 1996). Reseau National de Sante´ Publique, Saint Meaurice, France, 83 pp. Evanz, Jr. D.J., Evans, D.G., Gorbach, S.L., 1974. Polymyxin B-induced release of molecular weight, heat-labile enterotoxin from Escherichia coli. Infect. Immun. 10, 1010–1017. Germani, Y., 1996. Diagnostic des Escherichia coli agents d’e´ nterites. Bull. Soc. Fr. Microbiol. 11, 216–224. ´ Gyles, C., Johnson, R., Gao, A., Ziebell, K., Pierard, D., Aleksic, S., Boerlin, P., 1998. Association of enterohemorrhagic Escherichia coli hemolysin with serotypes of Shiga-like-toxinproducing Escherichia coli of human and bovine origins. Appl. Environ. Microbiol. 64, 4134–4141. Heuvelink, A.E., Wernars, K., De Boer, E., 1996. Occurrence of Escherichia coli O157 and other verocytotoxin-producing E. coli in retail raw meats in The Netherlands. J. Food Prot. 59, 1267–1272. Heuvelink, A.E., Van Den Biggelaar, F.L.A.M., De Boer, E., Herbes, R.G., Melchers, W.J.G., Huis In’T Veld, J.H.J., Monnens, L.A.H., 1998a. Isolation and characterization of ver-
223
ocytotoxin-producing Escherichia coli O157 strains from dutch cattle and sheep. J. Clin. Microbiol. 36, 878–882. Heuvelink, A.E., Van Den Biggelaar, F.L.A.M., ZwartkruisNahuis, J.T.M., Herbes, R.G., Huynen, R., Nagelkerbe, N. et al., 1998b. Occurrence of verocytotoxin-producing Escherichia coli O157 on dutch dairy farms. J. Clin. Microbiol. 36, 3480– 3487. Johnson, R.P., Clarke, R.C., Wilson, J.B., Read, S.C., Rahn, K., Renwick, S.A. et al., 1996. Growing concerns and recent outbreaks involving non-O157:H7 serotypes of verotoxigenic Escherichia coli. J. Food Prot. 59, 1112–1122. Karch, H., Meyer, T., 1989. Evaluation of oligonucleotide probes for identification of Shiga-like-toxin producing Escherichia coli. J. Clin. Microbiol. 27, 1180–1186. Karch, H., Huppertz, H.I., Bockenmuhl, J., Schmidt, H., Schwarzkopf, A., Lissner, R., 1997. Shiga toxin-producing Escherichia coli infections in Germany. J. Food Prot. 60, 1454– 1457. Konowalchuck, J., Speirs, J.I., Stavric, S., 1977. Vero response to a cytotoxin of Escherichia coli. Infect. Immun. 18, 775–779. Nataro, J.P., Kaper, J.B., 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11, 142–201. ´ Pierard, D., Van Damme, L., Moriau, L., Stevens, D., Lauwers, S., 1997. Virulence factors of verocytotoxin-producing Escherichia coli isolated from raw meats. Appl. Environ. Microbiol. 63, 4585–4587. Read, S.C., Clarke, R.C., Martin, A., De Grandis, S.A., Hiis, J., McEwen, S., Gyles, C.L., 1992. Polymerase chain reaction for detection of verocytotoxigenic Escherichia coli isolated from animal and food sources. Mol. Cell Probes 6, 153–161. Riley, L.W., Remis, R.S., Helgerson, S.D., McGee, H.B., Wells, J.G., Davis, B.R. et al., 1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. New Engl. J. Med. 308, 681–685. Sambrook, J., Fritsch, E.F., Maniatis, T. (Eds.), 1989. Molecular Cloning. A Laboratory Manual, 2nd Edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Schmidt, H., Beutin, L., Karch, H., 1995. Molecular analysis of plasmid-encoded hemolysin of Escherichia coli O157:H7 strain EDL 933. Infect. Immun. 63, 1055–1061. Shere, G., Bartlett, K.J., Kaspar, C.W., 1998. Longitudinal study of Escherichia coli O157:H7 dissemination on four dairy farms in Wisconsin. Appl. Environ. Microbiol. 64, 1390–1399. Uyttendaele, M., Grangette, C., Rogerie, F., Pasteau, S., Debevere, J., Lange, M., 1998. Influence of cold stress on the preliminary enrichment time needed for detection of enterohemorrhagic Escherichia coli in ground beef by PCR. Appl. Environ. Microbiol. 64, 1640–1643. ¨ Wieler, L.H., Vieler, E., Erpenstein, C., Schlapp, T., Steinruck, R., Bauerfeind, et al., 1996. Shiga toxin-producing Escherichia coli strains from bovines: Association of adhesion with carriage of eae and other genes. J. Clin. Microbiol. 34, 2980– 2984.
Characterization of Shiga toxin producing E. coli and O157 serotype E. coli isolated in France from healthy domestic cattle ´ Franc¸ois Rogerie a , Armelle Marecat a , Stephanie Gambade b , Francis Dupond a , b a, Pierre Beaubois , Marc Lange * a
´ ` de Microbiologie des Ecosystemes , 1 Rue du Professeur Calmette, BP 245 -59019 Institut Pasteur de Lille, Service R& D, Departement Lille Cedex, France b SOCOPA Villefranche, 03430 Villefranche d’ Allier, France Received 13 November 1999; received in revised form 31 July 2000; accepted 12 August 2000
Abstract A study was carried out in France in collaboration with the meat industry to investigate the occurrence and characteristics of Shiga toxin-producing E. coli (STEC) and O157 E. coli in a population of healthy bovines representative of French livestock. A total of 851 animals belonging to three bovine classes (106 young bulls, 374 dairy cows and 371 meat cows) were included in the study. Samples of feces and of the corresponding carcasses were collected from March 97 to August 97 in seven abattoirs spread throughout the national territory. STEC cultures from the 1702 samples were screened using PCR for the presence of stx genes. Positive samples were further subjected to colony blot hybridization and to O157-specific immunomagnetic separation. Probe-positive colonies and O157 colonies were then analyzed for the presence of virulence genes and phenotypic characters (serotype, Stx production). In 154 (18.1%) feces and 91 (10.7%) carcass samples stx genes were detected. Two hundred and twenty-two STEC colonies were isolated from 67 (7.9%) feces and 16 (1.9%) carcass samples, with 183 STEC isolated from feces and 39 from carcasses. Only eight O157 isolates were collected from feces samples. None of these O157 E. coli isolates presented stx genes and thus could not be considered as pathogenic regarding hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS). In 3.2% of STEC isolated from feces and in 10.2% of STEC from carcasses eae genes were detected. In 17% of STEC from feces and in 30.7% from carcasses ehx genes were detected. Using these data, the 222 STEC colonies could be classified in 11 different ‘virulence patterns’ (presence / absence of stx1, stx2, eae and ehx genes), showing that more than 77% of isolates presented only one virulence factor. Only three STEC on 222 colonies (1.3%) presented the three virulence factors stx, eae and ehx in association, none of them reacting with antisera specific for enterohemorrhagic E. coli. (EHEC). These data, together with the fact that only five isolates on the 222 (2.2%) reacted with such antisera (three O111 and two O26 isolates) demonstrated that the natural bacterial populations isolated during this study were clearly distinct from EHEC. 2001 Elsevier Science B.V. All rights reserved. Keywords: France; Healthy cattle; Epidemiology; Shiga toxin producing E. coli
*Corresponding author. E-mail address: [email protected] (M. Lange). 0168-1605 / 01 / $ – see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S0168-1605( 00 )00422-0
218
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
1. Introduction Enterohemorrhagic Escherichia coli (EHEC) were recognized as important human pathogens in the US in 1982 (Riley et al., 1983). EHEC strains are defined by a combination of the symptoms they produce and the virulence factors they possess. They are associated with a variety of clinical manifestations, watery diarrhea, hemorrhagic colitis (HC) that may lead to more severe complications, hemolytic uremic syndrome (HUS) and thrombotic thrombocytopenic purpura. EHEC infections are mainly caused by E. coli belonging to the O157:H7 serotype, which has been implicated in large foodborne outbreaks all over the world (USA, Canada, Scotland and Japan, Nataro and Kaper, 1998). Other EHEC serotypes (0111, O26, O103, etc.) have also been associated with small to large outbreaks of HC or with human sporadic infections (Johnson et al., 1996). EHEC strains associated with human disease have genetic properties in common. The first is that they possess the stx genes coding for the Shiga toxins Stx1 and / or Stx2, found on temperate lambdoid bacteriophages into the bacterial chromosome. The second is that most EHEC strains have additional virulence factors, mainly the eae gene coding for intimin, located in the LEE locus associated to the Attaching and Effacing phenotype, and the ehx gene located on a large plasmid (90 kb), coding for the synthesis of entero-hemolysin. These three virulence genes stx, eae and ehx are found in association in most of clinical isolates (Barrett et al., 1992; Karch et al., 1997; Gyles et al., 1998). The natural population of Shiga toxin-producing E. coli (STEC) found in the intestine of domestic animals is considered as the natural reservoir of EHEC, and cattle have been implicated as the source of EHEC pathogens in a number of foodborne outbreaks associated with the consumption of undercooked ground beef or raw milk products (Nataro and Kaper, 1998). In Europe, surveys made to estimate the prevalence and properties of STEC in non-diarrheic healthy domestic animal species (Beutin et al., 1993, 1995, 1997; Blanco et al., 1996, 1997), revealed that cattle frequently harboured such STEC strains. These studies, performed on selected herds or populations of farm animals, have shown that most of the STEC bovine isolates lacked the
main additional virulence factor eae and ehx, and thus would not represent a potential danger to human health. However, in a recent study on 10 Dutch dairy farms focused on STEC strains of 0157 serotype, Heuvelink et al. (1998a) revealed that most of the isolates possessed the eae virulence gene. In France, no data were available until now on the fate and characteristics of STEC isolated from healthy domestic cattle. The aim of the present study, initiated by the meat Industry, was to determine the prevalence of STEC and O157 E. coli isolated from feces of a selected population of healthy bovines representative of the entire French livestock, and also to compare their phenotypic, serological and genetic traits to those characteristic of EHEC. Concurrently, the corresponding carcasses were submitted to the same microbiological analysis in order to assess quality of the slaughtering procedures.
2. Material and methods
2.1. Collection of samples and selective enrichment The size of the sampling has been fixed to 851 animals, which is 1 / 10 000 of the total number of bovines bred in France in 1997. Sampling was carried out on three categories of animals in proportions representative of the French breeding system, with 106 (12.5%) young bulls, 374 (44%) dairy cows and 371 (43.5%) meat cows. Samples, collected from March 1997 to August 1997, were taken weekly from seven abattoirs spread throughout the national territory. At each sampling site, the animals were selected randomly and identified by a number from one to 851. The numbers were made to trace the bovine class, the birth location, and the week and site of slaughtering. Immediately after slaughter, samples of the rectal contents were collected aseptically and placed into sterile plastic pots. The corresponding samples of carcasses, also identified by a number from one to 851, were collected after chilling the day after. In each case about 25 g of meat were taken by excision in three different areas (neck, shoulder and backside) and placed into sterile bags. Samples, kept cold with melting ice, were sent to the laboratory by express transport, and stored at 48C prior to microbiological analysis, which started
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
within 72 to 96 h after sampling, as previously described (Uyttendaele et al., 1998). One gram of feces was used to inoculate 10 ml of modified E. coli broth (mEC) (Merck, Nogent sur Marne, France) supplemented with novobiocin (20 mg / l, Sigma, St. Quentin Fallavier, France). The 25 g of carcass meat were added to 225 ml of the same medium mEC 1 n, and homogenized in a stomacher for 1 min at low speed. Feces and carcass enrichment cultures were then incubated for 18 h at 378C.
2.2. PCR-based detection of stx genes One milliliter of enrichment culture was transferred to a 1.5-ml microtube and centrifuged 3 min at 12 000 3 g. The supernatant was removed and the pellet was washed in PBS buffer (pH 7.4, Sigma). The DNA from pelleted cells was released by boiling and purified using Instagen DNA purification matrix (Biorad, Ivry sur Seine, France). Ten microliters of the DNA were used as template for PCR detection of stx genes. Oligonucleotides ES149 and ES151 described by Read et al. (1992) amplified a conserved sequence of the stx1 and stx2 genes. The conditions for the PCR amplification were the same as those described by Uyttendaele et al. (1998).
2.3. Isolation of STEC by selective plating and colony hybridization with stx gene probes When PCR products of the expected size were obtained, STEC were tentatively isolated by colony blot hybridization, either directly from feces or from feces and carcass enrichment culture. Direct isolation was made from 1 to 2 g of feces suspended in physiological solution. One hundred microliters (v / v) were plated onto Plate count agar (PCA, Difco laboratories, Osi, Maurepas, France) supplemented with 1.5 g / l of bile salts No. 3 (Difco), and incubated 24 h at 42.58C. Ten microliters of feces or carcass enrichment were streaked on plates and incubated overnight at 42.58C. In each case, after growth, plates containing from 50 to 100 colonies were replicated onto Nylon membrane of 82 mm of diameter (Pall-Gelman, Champ sur Marne, France). Lysis of bacterial cells and binding of DNA to Nylon membrane were achieved as described by Sambrook et al. (1989), except that prior to DNA fixation, the membrane was treated 1 h at 378C with Proteinase K
219
(2 mg / ml, Boehringer-Mannheim, Meylan, France). DNA probes specific for stx genes were generated with the PCR DIG probe synthesis kit (BoehringerMannheim). Two degenerate primers 772 and 849 described by Karch and Meyer (1989) were used to amplify DIG labeled stx genes from reference strains LCDCH19 and LCDC902380 (Laboratory Center for Disease Control, Ottawa, Canada), where the stx1 and the stx2 genes are present, respectively. The hybridization, washing and detection reactions were performed according to the manufacturer’s instructions.
2.4. Isolation of O157 by IMS Standard IMS with magnetic beads coated with ` antibody to O157 antigen (Dynal, Compiegne, France) was performed with 1 ml of enrichment culture where stx genes were detected. The concentrates were inoculated onto sorbitol–MacConkey agar (Difco) supplemented with cefixime (0.05 mg / l) and potassium tellurite (2.5 mg / l), to test the sorbitol fermentation ability. Concentrates were also plated onto chromagar O157 (Dynal) to examine the b-glucuronidase activity. The sorbitol non-fermenting and / or b-glucuronidase negative colonies were tested for agglutination with an E. coli O157 latex test kit (Oxoid, Basingstoke, UK). Positive isolates were tested for the H7 antigen by agglutination (Difco) and transferred in mannitol-nitrate-motility medium (Sanofi, Marnes la Coquette, France) to check for their motility.
2.5. Biochemical traits of isolates STEC and O157 were inoculated on KliglerHajna, indole-urea, Rapid’E. coli (Difco) and SMAC (Sanofi) media and incubated 24 h at 378C. Isolates with atypical combination of characters were further confirmed by using API20E biochemical test strips ´ (Biomerieux, Marcy l’Etoile, France).
2.6. Serotyping of STEC O grouping was carried out by slide agglutination of living bacteria with antisera specific for the major enterohemorrhagic E. coli groups O26, O111 (Sanofi) and O103 (provided by Dr. F. Scheutz, Copenhagen, Denmark). O157 and H7 determination
220
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
were performed with Difco antisera. Reference strains for serotype determination were kindly pro´ vided by Professor E. Bingen (Hopital Robert Debre, Paris, France).
2.7. Virulence traits of isolates All isolates were investigated for production of Shiga toxins (Stx) by Vero cell toxicity as described by Konowalchuck et al. (1977). Stx release was induced by Polymixin B treatment as described by Evanz et al. (1974). The presence of stx1 and stx2 genes was detected by PCR with conserved ES149– ES151 primers coupled to dot hybridization with oligonucleotides specific to stx1 or to stx2 gene. Two oligonucleotides were designed within the 323 bp amplified from primers ES149–ES151. Oligo HSLTI (59-CATTGTCTRGTGACAGTAGCT-39) is specific to stx1, and oligo SLTIIF (59-CCATGACRACGGACAGCAGTT-39) is specific to the different variants of stx2 gene. STLIIF was chosen after sequence analyzing of all known stx2 gene variants and derived from a conserved region. STLIIF was further tested with our own collection of EHEC E. coli. The labeling of oligos and the hybridizationdetection reactions were performed according to the manufacturer’s recommendations of the DIG oligonucleotide tailing kit (Boehringer-Mannheim). The presence of the eae gene was examined using specific primers eae1 /eae2 described by Germani (1996). These primers are derived from conserved regions of the eae gene and allow amplification of a 490-bp DNA fragment. Isolates were additionally tested for the presence of the ehx gene by PCR with the primers hlyA1 / hlyA4 described by Schmidt et al. (1995).
3. Results and discussion
3.1. Prevalence and characteristics of natural colonizing STEC isolated from healthy domestic cattle in France 3.1.1. Virulence genes The stx genes were detected in 154 (18.1%) feces and 91 (10.7%) carcass samples; stx-positive colonies were obtained from 67 (7.9%) feces and 16 (1.9%) carcass samples. From these 83 samples, 22
allowed isolation of only one colony. As an average of two to four STEC colonies were taken from the other 61 positive samples, a total of 222 STEC colonies were collected, 183 from feces and 39 from carcasses. No clear correlation was found between PCR or STEC isolation data and the slaughterhouse’s locations, the date of birth of animals or with any other parameters linked to the breeding of animals (results not shown). Only 195 (87%) colonies of the 222 produced an active Shiga toxin (results not shown). Such a lack of expression of stx genes has already been observed by Heuvelink et al. (1996) in STEC isolated from retail raw meats. However, all of the 222 isolates were considered as STEC and we have thus estimated the prevalence of STEC into French domestic cattle to be 18.1% in feces (10.7% in carcasses) using stx PCR data and 7.9% in feces (1.9% in carcasses) based on isolation data (yield of isolation is around 40%, results not shown). The lower values observed for carcasses (10.7% using PCR data and 1.9% using isolation data) may be indicative of the quality of the slaughtering procedures used in the seven abattoirs selected in this study. These data are however difficult to compare with those of other European studies because obtaining them may be greatly influenced by factors such as the target population, the sampling strategy, the methodology of screening and isolation and also by possible seasonal variation in prevalence. In fact, our study showed that the carriage rate of VTEC E. coli in bovines was not influence by the birth, breeding or abattoir place, by the abattoir season or by the aged groups (data not shown). Nevertheless, our estimation using stx PCR data lies within the European average, where the carriage rate of STEC in cattle was shown to range from 11 to 21% (Caprioli and Tozzi, 1998). Each of the 222 STEC isolates was studied for the presence of the different virulence genes (stx1, stx2, eae and ehx genes). This typing scheme allowed us to differentiate the 222 colonies in 11 distinct ‘virulence patterns’ (Table 1). The characterization of stx genes by oligonucleotide-typing showed that STEC colonies with only one virulence gene (stx1 or stx2 or stx1 / 2) represented more than 77% of the isolates. A majority of STEC colonies harboured stx2 gene alone (60% of STEC). Beutin et al. (1997) have also shown that stx2 positive STEC were more
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223 Table 1 Virulence patterns of STEC isolated from healthy cattle Gene(s)
stx1 stx2 stx1 stx2 stx1 ehx stx2 ehx stx1 stx2 ehx stx1 eae stx2 eae stx1 stx2 eae stx1 eae ehx stx2 eae ehx
No. of STEC from (%) Feces (n 5 183)
Carcass (n 5 39)
37 (20) 98 (53) 14 (7.6) 2 (1) 16 (9) 10 (5) 2 (1) 0 (0) 1(1) 1 (1) 2 (2)
3 (7.7) 19 (48) 1 (2.5) 0 (0) 2 (2.5) 10 (25) 2 (5) 2 (5) 0 (0) 0 (0) 0 (0)
frequently isolated from healthy cattle. Furthermore, we observed differences concerning the main type of stx gene harboured by STEC from bovines of different age categories. stx2 gene was the dominant Shiga toxin coding gene in STEC from feces of adult cows (68%), and stx1 gene was slightly more abundant in STEC from young animals (52%). Similar observations were made in Spain (Blanco et al., 1996, 1997), where adult cows preferentially harboured stx2-positive STEC and young cows stx1positive STEC, suggesting that the age of animals may have an influence on the dynamic of STEC colonization. As also shown in Table 1, the proportion of colonies with the eae gene is low, 3.2% in feces and 10% in carcass. These values are in good agreement with the value observed for European healthy cattle (Beutin et al., 1997) or for STEC isolated from raw ´ meat (Pierard et al., 1997). The low prevalence of the eae gene is characteristic of STEC isolated from healthy domestic cattle. Conversely, eae genes were found very frequently in STEC from diarrheagenic cattle or infected humans (Wieler et al., 1996; Gyles et al., 1998). On the other hand, we found a higher proportion of STEC isolates with the ehx gene (17% in feces, 30.7% in carcasses). These values are much lower than that (51%) published in a study by Gyles et al. (1998) investigating the occurrence of the stx gene in bovine STEC strains belonging to serotypes less commonly or not associated with disease in humans (in our study, three STEC O111 and two
221
O26 have been isolated, none of them possessing the ehx gene, see below). Only three STEC isolated from three different samples of feces were characterized by the presence in association of the three virulence markers, none of these colonies reacting with the antisera specific for the main EHEC serotypes — O157, O103, O111 and O26 — encountered in France and Europe (Decludt, 1997, Caprioli and Tozzi, 1998).When similar typing schemes were made with STEC commonly associated with disease in humans, the majority of the isolates were characterized by the presence of stx, eae and ehx genes in association (Barrett et al., 1992). In a study by Karch et al. (1997) on children with HUS in Germany, the data demonstrated that 94% of the STEC from patients with HUS were positive for the presence of eae and ehx genes. In the same way, Gyles et al. (1998) have studied a collection of human enterohemorrhagic E.coli and shown that 89% of them harboured ehx genes and 92% the eae gene. Finally, we could also compare colonies isolated from the same sample (61 out of 83 samples are concerned) and thus observed that isolates from 14 of theses samples had different virulence patterns. These results revealed that cattle could harbour at the same time different STEC strains, regarding their virulence patterns. Besides, when colonies were obtained from feces and carcass of the same animal (only two animals were concerned), they were characterized by different virulence types, with the exception of two STEC. However, these two colonies could be discriminated by ribotyping analysis (data not shown).
3.1.2. Phenotypic traits All STEC were confirmed to be E. coli by biochemical identification with the exception of three isolates. Two isolates were characterized as E. fergusonii and one as Citrobacter freundii. The O grouping was carried out by choosing the serotypes which were mostly associated with HUS in France (Decludt, 1997) as well as in North America and Australia (Nataro and Kaper, 1998). Thus, serotyping of the 222 STEC colonies showed that none of them was of the O157 serotype. Besides, only five colonies (2.2%) reacted with three other selected antisera (O103, O111 and O26) specific for enterohemorrhagic E. coli. Three isolates from three
222
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223
different samples of feces reacted with the O111 antiserum. None of these three isolates presented the eae gene, which is present in all the O111 STEC commonly associated with disease in human (Gyles et al., 1998). The two other isolates from the same carcass, were serotyped O26 and harboured the eae gene, not the ehx gene. Additional molecular typing data for comparison of these isolates to O26 clinical isolates are now needed to give a clear conclusion on their possible pathogenicity. However they have been isolated in only one sample among the 1702 analyzed in the study.
3.2. Prevalence and characteristics of natural colonizing O157 from healthy domestic cattle in France The O157 STEC were searched by the IMS method from feces and carcass enrichment positive for the presence of stx genes. Eight O157 colonies were isolated from four different samples of feces, none from carcasses.
3.2.1. Virulence genes The eight O157 isolates were tested for stx genes by PCR and displayed none of them. This results was confirmed by colony hybridization with cloned stx genes as described in Section 2 (data not shown). Thus these isolates could not be considered as pathogenic regarding HC and HUS. This situation is clearly different from that described by Heuvelink et al. (1998a) in Dutch dairy farms where the majority of the O157 strains isolated from cattle were stx1, stx2 and eae positive. Five of the eight O157 isolates had no virulence gene at all. Two strains were eae and ehx positive, while the other one was only eae positive. 3.2.2. Phenotypic traits Of the eight isolates, one had the ability to ferment sorbitol and five showed a b-glucuronidase activity. Three strains displayed the phenotype generally associated to pathogenic O157 / H7 strains, i.e., they were negative for sorbitol fermentation and bglucuronidase activity. Only one strain was positive for the H7 antigen. The other seven strains, with only one non-mobile, were not tested for other H antigens as they were not displaying stx genes. As four of the 154 tested feces samples allowed isolation of O157,
we could calculate that such O157 are present in 2.6% of feces samples, assuming that O157 are equally distributed in all feces samples, positive or negative for the presence of stx genes. This value lies in the range of 0–10% that has been estimated for European (Caprioli and Tozzi, 1998) and North American cattle (Armstrong et al., 1996).
4. Conclusion In this study, the first global survey ever performed in France of the STEC bovine reservoir, we have shown that the prevalence of STEC and O157 E. coli in French cattle are similar to the reported prevalence in European and North American healthy cattle. More importantly, the genotypic and phenotypic analysis of the 222 STEC and of the eight O157 isolates has shown that these natural populations are clearly distinct from pathogenic EHEC strains. However, the present survey may only be representative of the risk of EHEC contamination at the precise period of investigation. In fact, it was shown in follow up studies (Heuvelink et al., 1998b; Shere et al., 1998) that carriage of a given STEC strain by cattle may be intermittent and of short duration. It is thus advisable to further survey the STEC bovine reservoirs following procedures based on the characterization of isolates with genotypic (virulence traits) and phenotypic (serotype) markers.
Acknowledgements This work was supported financially by the Association Nationale Interprofessionnelle des Viandes ´ et du Betail (INTERBEV), by the Office National Interprofessionnel des Viandes de l’Elevage et de l’Aviculture (OFIVAL). We would like to thank Dr. Barbara DUFOUR, from the Agence Franc¸aise de ´ ´ Sanitaire des Aliments (AFSSA) and Dr. Securite Corinne GRANGETTE (Institut Pasteur de Lille) for helpful discussions.
References Armstrong, G.L., Hollingsworth, J., Morris, Jr. J.G., 1996. Emerging foodborne pathogens: Escherichia coli O157:H7 as a
F. Rogerie et al. / International Journal of Food Microbiology 63 (2001) 217 – 223 model of entry of a new pathogen into the food supply of the developed world. Epidemiol. Rev. 18, 29–51. Barrett, T.J., Kaper, J.B., Jerse, A.E., Wachsmuth, I.K., 1992. Virulence factors in Shiga-like toxin producing Escherichia coli isolated from humans and cattle. J. Infect. Dis. 165, 979–980. ¨ Beutin, L., Geier, D., Steinruck, 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. Beutin, L., Geier, D., Zimmermann, S., Karch, H., 1995. Virulence markers of Shiga-like toxin-producing Escherichia coli strains originating from healthy domestic animals of different species. J. Clin. Microbiol. 33, 631–635. Beutin, L., Geier, D., Zimmermann, S., Aleksic, S., Gillespie, H.A., Whittam, T.S., 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. Blanco, M., Blanco, J.E., Blanco, J., Blanco, E.A., Gonzalez, E.A., Mora, A., Prado, C. et al., 1996. Prevalence and characteristics of Escherichia coli serotype O157:H7 and other verotoxin-producing E. coli in healthy cattle. Epidemiol. Infect. 117, 251–257. Blanco, M., Blanco, J.E., Blanco, J., Mora, A., Prado, C., Alonso, M.P. et al., 1997. Distribution and characterization of faecal verotoxin-producing Escherichia coli (VTEC) isolated from healthy cattle. Vet. Microbiol. 54, 309–319. Caprioli, A., Tozzi, A.E., 1998. Epidemiology of Shiga toxinproducing Escherichia coli infections in continental Europe. In: Kaper, J.B., O’Brien, A.D. (Eds.), Escherichia Coli O157:H7 and Other Shiga Toxin-producing E. coli Strains. American Society for Microbiology, Washington, DC, pp. 38–48. ´ ´ Decludt, B., June 1997. Syndromes hemolytiques et uremiques en ´ ´ France, epidemiologie et agents responsables (April 1995– ´ March 1996). Reseau National de Sante´ Publique, Saint Meaurice, France, 83 pp. Evanz, Jr. D.J., Evans, D.G., Gorbach, S.L., 1974. Polymyxin B-induced release of molecular weight, heat-labile enterotoxin from Escherichia coli. Infect. Immun. 10, 1010–1017. Germani, Y., 1996. Diagnostic des Escherichia coli agents d’e´ nterites. Bull. Soc. Fr. Microbiol. 11, 216–224. ´ Gyles, C., Johnson, R., Gao, A., Ziebell, K., Pierard, D., Aleksic, S., Boerlin, P., 1998. Association of enterohemorrhagic Escherichia coli hemolysin with serotypes of Shiga-like-toxinproducing Escherichia coli of human and bovine origins. Appl. Environ. Microbiol. 64, 4134–4141. Heuvelink, A.E., Wernars, K., De Boer, E., 1996. Occurrence of Escherichia coli O157 and other verocytotoxin-producing E. coli in retail raw meats in The Netherlands. J. Food Prot. 59, 1267–1272. Heuvelink, A.E., Van Den Biggelaar, F.L.A.M., De Boer, E., Herbes, R.G., Melchers, W.J.G., Huis In’T Veld, J.H.J., Monnens, L.A.H., 1998a. Isolation and characterization of ver-
223
ocytotoxin-producing Escherichia coli O157 strains from dutch cattle and sheep. J. Clin. Microbiol. 36, 878–882. Heuvelink, A.E., Van Den Biggelaar, F.L.A.M., ZwartkruisNahuis, J.T.M., Herbes, R.G., Huynen, R., Nagelkerbe, N. et al., 1998b. Occurrence of verocytotoxin-producing Escherichia coli O157 on dutch dairy farms. J. Clin. Microbiol. 36, 3480– 3487. Johnson, R.P., Clarke, R.C., Wilson, J.B., Read, S.C., Rahn, K., Renwick, S.A. et al., 1996. Growing concerns and recent outbreaks involving non-O157:H7 serotypes of verotoxigenic Escherichia coli. J. Food Prot. 59, 1112–1122. Karch, H., Meyer, T., 1989. Evaluation of oligonucleotide probes for identification of Shiga-like-toxin producing Escherichia coli. J. Clin. Microbiol. 27, 1180–1186. Karch, H., Huppertz, H.I., Bockenmuhl, J., Schmidt, H., Schwarzkopf, A., Lissner, R., 1997. Shiga toxin-producing Escherichia coli infections in Germany. J. Food Prot. 60, 1454– 1457. Konowalchuck, J., Speirs, J.I., Stavric, S., 1977. Vero response to a cytotoxin of Escherichia coli. Infect. Immun. 18, 775–779. Nataro, J.P., Kaper, J.B., 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11, 142–201. ´ Pierard, D., Van Damme, L., Moriau, L., Stevens, D., Lauwers, S., 1997. Virulence factors of verocytotoxin-producing Escherichia coli isolated from raw meats. Appl. Environ. Microbiol. 63, 4585–4587. Read, S.C., Clarke, R.C., Martin, A., De Grandis, S.A., Hiis, J., McEwen, S., Gyles, C.L., 1992. Polymerase chain reaction for detection of verocytotoxigenic Escherichia coli isolated from animal and food sources. Mol. Cell Probes 6, 153–161. Riley, L.W., Remis, R.S., Helgerson, S.D., McGee, H.B., Wells, J.G., Davis, B.R. et al., 1983. Hemorrhagic colitis associated with a rare Escherichia coli serotype. New Engl. J. Med. 308, 681–685. Sambrook, J., Fritsch, E.F., Maniatis, T. (Eds.), 1989. Molecular Cloning. A Laboratory Manual, 2nd Edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Schmidt, H., Beutin, L., Karch, H., 1995. Molecular analysis of plasmid-encoded hemolysin of Escherichia coli O157:H7 strain EDL 933. Infect. Immun. 63, 1055–1061. Shere, G., Bartlett, K.J., Kaspar, C.W., 1998. Longitudinal study of Escherichia coli O157:H7 dissemination on four dairy farms in Wisconsin. Appl. Environ. Microbiol. 64, 1390–1399. Uyttendaele, M., Grangette, C., Rogerie, F., Pasteau, S., Debevere, J., Lange, M., 1998. Influence of cold stress on the preliminary enrichment time needed for detection of enterohemorrhagic Escherichia coli in ground beef by PCR. Appl. Environ. Microbiol. 64, 1640–1643. ¨ Wieler, L.H., Vieler, E., Erpenstein, C., Schlapp, T., Steinruck, R., Bauerfeind, et al., 1996. Shiga toxin-producing Escherichia coli strains from bovines: Association of adhesion with carriage of eae and other genes. J. Clin. Microbiol. 34, 2980– 2984.