Shiga toxin-producing Escherichia coli (STEC) are a group of highly pathogenic foodborne zoonotic pathogens, causing diarrhea, hemorrhagic colitis (HC), and hemolytic uremic syndrome (HUS) in humans. The common feature and main virulence factors of STEC are two phageencoded cytotoxins, called Shiga toxins (stx1 and stx2), which are directly correlated with human pathogenicity (Lindgren et al., 1993). Adherence factors are also critical features of STEC pathogenicity: human pathogenic STEC known to cause severe intestinal disease can attach to intestinal epithelial cells and form attaching and effacing lesions through an outer membrane protein called intimin, which is encoded by the eae gene. Furthermore, as shown by the enteroaggregative E. coli (EAEC) O104:H4 strain that caused the large outbreak in Germany in 2011, other means of attachment such as the factors coded by the aggR regulatory plasmid gene and the chromosomal aaiC gene, when coupled with the production of stx2, can have severe consequences (Beutin and Martin, 2012). More than 200 virulent STEC serotypes have been isolated from human infections (Coombes et al., 2008). Although E. coli O157: H7 is the serotype that has been linked to most outbreaks of food-borne diseases and brought the largest number of cases of HUS, in recent years a growing number of non-O157 STEC strains have been isolated from human clinical cases and outbreaks (Caprioli et al., 2005). The lower intestinal tract of ruminants is considered to be the main natural reservoir of STEC. In most human infections, transmission occurs primarily by ingestion of contaminated food of bovine origin, though few outbreaks have been associated to the consumption of other food products, including pork products contaminated by O157 (CDC, 1995; Williams et al.,2000; MacDonald et al., 2004; Conedera et al., 2007; Trotz-Williams et al., 2012; Honish et al., 2017) and by O111 (Paton et al., 1996). However, the epidemiology and virulence characteristics of STEC carried by on-farm pigs remain largely unknown. The hypothesis that swinederived STEC strains are similar to humanderived STEC strains and have the potential to contribute to human infections needs to be further investigated (Tseng et al., 2014a).
Unfortunately, epidemiological data on STEC prevalence in pigs and in food products made of pork are collected at the European level only by a few countries and they are not always comparable (EFSA and ECDC, 2016). Following E. coli O104:H4 outbreak occurred in Germany in May 2011, Emilia Romagna Region instituted in 2012 a testing program for O157, O26, O111, O103, O145 and O104:H4 STEC, both at production and at retail level, for foodstuff considered at risk of contamination. During years 2012-2013 monitoring plan STEC virulence genes were detected at high frequency (19%) in fresh pork sausage and one verocytotoxic E. coli O103 strain was isolated from pork meat (Bardasi et al., 2015).
Because of the limited epidemiologic data on STEC prevalence in swine, the increasing evidence of pork food STEC contamination, and the rising role of non-O157 STEC in human outbreaks, the aim of this study is to investigate on STEC occurrence both in caecal content samples collected from pigs to the slaughterhouse and from pork food products.
Materials and methods
During the years 2015 and 2016, 83 swine individual faecal samples were collected from the rectum of animals at slaughter in Emilia Romagna Region. Swine were reared in 18 different farms located in Piedmont, Emilia Romagna, Tuscany and Lombardy, 1 to 7 swine from each farm were included.
A total of 562 pork products were collected during January 2014 to August 2016 in Emilia Romagna Region from retail outlets, large retailers and processing plants. Food samples comprised 465 not-ready-toeat (NRTE) products to be consumed after cooking (62 pork meat, 109 pork minced meat, 294 fresh meat pork sausages and processed meat products) and 97 ready-toeat (RTE) samples (65 salami, 26 dry-cured ham, 2 mortadella, 2 pancetta, 2 coppa)
The samples were analysed according to ISO TS 13136:2012 (ISO, 2012); 25 g of each food sample were diluted ten-fold (w/v) in modified Tryptone Soya Broth (mTSB) supplemented with 16 mg/L of novobiocin (mTSB+N) and incubated at 37±1°C for 21±3 h. Five grams of each faecal sample was diluted ten-fold (w/v) in Tryptone Soya Broth and incubated at 37±1°C for 21±3 h. Bacterial DNA was extracted from 1 mL of enriched broth using Gen elute™ bacterial genomic DNA kit (Sigma-Aldrich, St. Louis, MO, USA) as described by the manufacturer. All primers and probes used in this study are reported in ISO 13136:2012 and published previously (Perelle et al., 2004; Nielsen and Andersen, 2003; ISO, 2012). Multiplex real time polymerase chain reaction (PCR) targeting the virulence genes eae, stx1 and stx2 was conducted in a 25 μL reaction volume using the following reaction mixture: 1 X Taqman®Universal PCR Master mix (Applied Biosystems, Foster City, CA, USA), 450nM each of the forward and reverse primers, 100 nM of each labeled probe and 4 μL DNA template. A commercially available TaqMan® Exogenous Internal Positive Control (Applied Biosystems) was included in each PCR reaction. Real time-PCR thermal cycling was conducted using a StepOne Plus system (Applied Biosystems). The cycling parameters were: 95°C hold for 10min for initial denaturation of the DNA and activation of the hot-start Taq polymerase, followed by 40 cycles of amplification of 95°C for 15s, and 60°C for 60 s. Samples resulted positive for the presence of stx1 and/or stx2 gene were tested for E. coliO 104 serogroupassociated gene. Sample positive for the presence of stx1 and/or stx2 in association with eae gene were tested for the detection of E. coli O103, O111, O145, O157, O26, serogroup-associated genes. Serogroup specific PCR reactions were conducted in a 25 μL reaction volume using the following reaction mixture: 1 X Taqman®Universal PCR Master mix (Applied Biosystems), 900 nM each of the forward and reverse primers, 250 nM of the labeled probe and 4 μL DNA template. The PCR instrument and program were the same used for the previous reaction. When stx1 and/or stx2 genes were detected, the isolation of the strain from the enrichment sample broth was attempted. Enriched samples were plated on Tryptone Bile X-Glucuronide (TBX) agar and incubated for 18-24 h at 37±1°C. Up to 50 colonies with E. coli morphology were picked up and point-inoculated on Nutrient agar (NA). Pools of 10 colonies were tested by real time PCR for the presence of virulence genes eae, stx1 and stx2, afterward colonies from positive pools were tested singularly in order to identify STEC strain.
STEC stains without eae gene were tested for the presence of aaiC and aggR genes by real time PCR assay described by EU Reference Laboratory for E. coli (EURL VTEC, web site: http://www.iss.it/binary/vtec/cont/EU_RL_VTEC_Method_05_Rev_1.pdf).
Out of a total of 83 swine faecal samples tested by real time PCR for the presence of stx1, stx2 and eae, 92.8 % (77/83) resulted stx-positive with stx2 gene more frequently detected than the stx1 gene (75/83 equals to 90.4% vs 7/83 equals to 8.4%). Two out of 83 (2.4%) samples tested positive for stx1 associated with eae gene, 13/83 samples (15.7%) resulted positive only for the stx2 gene, 57/83 (68.7%) tested positive for the stx2 gene in association with eae gene and 5/83 samples (6.0%) resulted positive for stx1, stx2 and eae (Table 1). Serogroup O26, O103, O104, O111, O145 and O157 and associated genes were detected respectively in 52/64 (81.3%), 51/64 (79.7%), 48/77 (62.3%), 23/64 (35.9%), 51/64 (79.7%) and 50/64 (78.1%) faecal samples. Out of a total of 562 pork products tested, no one of the 97 RTE sample resulted positive for stx genes and among the 465 NRTE samples, 65 (14.0%) resulted stx-positive. The stx2 gene was detected more frequently than the stx1 gene (62/465 equals to 13.3% vs 6/465 equals to 1.3%). One sample out of 465 (0.2%) resulted positive only for stx1 gene, 2/465 (0.4%) samples were positive for stx1 associated with eae gene, 22/465 (4.7%) samples tested positive only for stx2 gene, 37/465 (8.0 %) were positive for the stx2 gene in association with eae gene and 3/465 (0.7%) samples resulted positive for stx1, stx2 and eae. Serogroup O26, O103, O104, O111, O145 and O157 associated genes were detected respectively in 32/42 (76.2%), 19/42 (45.2%), 5/65 (7.7%), 8/42 (19.0%), 37/42 (88.1%) and 35/42 (83.3%) NRTE pork samples.
Among the samples tested for O-group associated genes, 87.5% (56/64) faecal samples and 90.5% (38/42) food samples resulted positive to more than one serogroup. Percentages of samples tested simultaneously positive to 1 up to 6 serogroups are reported in Table 2.
Six STEC strains were isolated from faecal samples giving a culture positive STEC rate of 7.8% (6/77) for stx-positive samples and 7.2% (6/83) for all samples: one stx1+ strain and 5 stx2+ strains were isolated, two strains stx2+ being isolated from swine reared in the same farm.
Seven STEC strains were isolated from NRTE food samples giving a culture positive STEC rate of 10.8% (7/65) for stx-positive samples and 1.5% (7/465) for all samples. The strains were all stx2+: 4 were isolated from minced meat and 3 from fresh sausages. None of the isolated STEC strains belonged to one of the Top five or O104 serogroups. None of the isolated strains carried aggR and aaiC genes.
Epidemiological studies performed in European countries have reported a STEC contamination rates in faecal samples of slaughtered pigs estimates by PCR for stx genes of 22% out of 630 samples analysed in Switzerland (Kaufmann et al., 2006), 23.8% (24/101) in Belgium (Botteldoorn et al., 2002), 31% (56/182) in France (Bouvet et al., 2002) and 38.6% (81/210) in Italy, Umbria and Marche regions (Ercoli et al., 2016) with a STEC isolation rate ranging from 7.9% (8/101) (Botteldoorn et al., 2002) to 12.4% (26/210) (Ercoli et al., 2016). In Germany and Italy, swine population has been investigated resulting in 10.1% of 475 animals and none of 102 animals positive for STEC isolation, respectively (EFSA and ECDC, 2016). In our study, 92.8% (77/83) swine faecal samples resulted stx-positive, and a STEC isolate was recovered in 7.2% (6/83) of samples. STEC contamination rate assessed by real time PCR for stx genes resulted higher while the isolation rate resulted comparable to that reported in other European countries. However, it is very difficult to make comparisons among different studies since sampling collection method, sample size, geographic location and different protocols applied for STEC PCR detection and isolation can significantly influence the results (Fratamico et al., 2004; Tseng et al., 2014b). As reported by EFSA and ECDC (2016), 859 samples of pork meat have been tested in EU during 2015 by ISO13136:2012 with a STEC isolation rate of 2.56% (22/859). In Italy STEC contamination rate in fresh pork sausages assessed by PCR for stx-genes varied from 16% (20/126) (Villani et al., 2005) to 19% (41/213) (Bardasi et al., 2015) and the analysis of 675 samples including both fresh and dried products revealed 19 (2.8%) stx-positive fresh sausage samples (Ercoli et al., 2016). The STEC isolation rate ranged from 0% (Ercoli et al., 2016) to 10% (13/126) (Villani et al., 2005). In this study, no one of 97 RTE samples and 65 (14.0%) of 465 NRTE samples resulted stx positive by real time PCR with an isolation rate of 1.5% (7/465) among NRTE samples.
It is worth highlighting that all the 97 RTE samples included in the analysis resulted stx-negative. Various production processes were comprised in the panel of the RTE tested samples: heat-curing (mortadella), dry-curing (ham, pancetta and coppa), curing of fermented and air-dried meat (salami). Our data indicate that these manufacturing processes may have rendered the tested samples safe with respect to contamination by STEC.
As reported by EFSA and ECDC (2016), among the 859 samples of pork meat tested in EU in 2015 one O157 STEC strain (0.12%) and no O26, O103, O145, O111 STEC strains were isolated. Ercoli et al.(2016), reported high percentage of detection of Top five serogroup associatedgenes both in pork samples (O157: 42.1%, O145: 94.7%, O103: 78.9%, O26: 36.8%) and in swine faecal samples (O157: 86.4%, O145: 9.9%, O103: 17.3%, O26: 38.3%, O111: 1.2%), but no STEC strains belonging to these serogroups has been isolated.
Verocytotoxin-producing E. coli O157 has been isolated from swine faecal samples at low frequencies in Europe, ranging from 0% (0/630) in Switzerland (Kaufmann et al., 2006), France (0/182) (Bouvet et al., 2002), Belgium (0/101) (Botteldoorn et al., 2003) and United Kingdom (0/1000) (Chapman et al., 1997), 0.08% (2/2446) in Sweden (Eriksson et al., 2003), 0.1% (2/1976) in Norway (Johnsen et al.,2001) and 0.7% (1/150) in Italy (Bonardi et al., 2003).
The genes associated with the top five and O104 serogroups were detected at high frequency in this study, in particular O26, O145 and O157 serogroups resulted positive with high percentage in both faecal and food samples. O26 resulted the most frequent serogroup in faecal samples (81.3% of the stx and eae positive samples) and O145 in pork products (88.1% of the stx and eae positive samples). The O157 serogroup resulted positive in 78.1% and 83.3% of the faecal samples and pork samples tested for O-group associated genes, respectively. Furthermore, 61 out of the 64 (95.3%) faecal samples and all the 42 pork samples tested for the six serogroups (O26, O103, O104, O111, O145 and O157) resulted contaminated with one or more serogroups, the higher percentage of samples being positive to all six serogroups in faecal samples and to four serogroups in pork products. Nevertheless STEC strains belonging to the six serogroups have not been isolated in this study. Additionally, neither strains harboring both stx and eae genes, nor strains harboring stx gene in association with aaiC and aggR genes, have been isolated. The characteristics of the isolated strains indicate that the virulence and the serogroups-associated genes detected at high frequencies by real time PCR in the enriched samples, are presumably located on different bacterial strains. This hypothesis is also supported by the isolation from faecal samples of 19 strains belonging to O26, O103, O145, O111 and O104 serogroups (5 belonging to serogroup O26, 3 to O103, 2 to O104, 1 to O111, 8 to O145, none to serogroup O157) but not characterised by the presence of the stx and/or eae genes and of two E. coli strains only harboring the eae gene. An extensive study carried out in U.S.A. on 181 STEC strains recovered from swine faecal samples reported comparable results: none of the STEC strains characterised by molecular methods carried eae gene or belonged to O26, O103, O111, O157 and O104 serogroups, only one strain being positive to O145 serogroup-associated gene (Baranzoni et al., 2016).
According to the molecular classification scheme proposed by EFSA (2013) STEC can be categorised according to potential risk for consumers health as group I (high potential risk) through to group III (unknown risk). An isolate of STEC serogroups O157, O26, O103, O145, O111, O104 in combination with stx and eae or aaiC and aggR genes (group I) should be considered as presenting a potentially high risk for diarrhoea and HUS. For any other serogroups in combination with the same genes (group II), the potential risk is regarded as high for diarrhoea, but currently unknown for HUS. In the absence of these genes (group III), there is uncertainty whether or not they are able to cause disease and a scientific judgement based on current knowledge of virulence characteristics cannot be done (EFSA, 2013). Remarkably, the STEC strains isolated in this study all belong to group III. Actually, swine are not viewed as an important STEC reservoir given the rare incidence of cases of severe human illness associated with STEC of swine origin (Tseng et al., 2014b).
In conclusion, in this study STEC stains have been isolated both in swine feces at slaughter and in NRTE pork products, while all RTE samples resulted stx-negative. The isolated strains are not the ones correlated with high risk for diarrhoea and HUS; nevertheless their human pathogenic potential is not yet fully defined.