Characterisation of Shiga toxin-producing Escherichia coli (STEC) isolated from seafood and beef

Abstract

Shiga toxin-producing Escherichia coli (STEC) strains isolated in Mangalore, India, were characterised by bead-enzyme-linked immunosorbent assay (bead-ELISA), Vero cell cytotoxicity assay, PCR and colony hybridisation for the detection of stx₁ and stx₂ genes. Four strains from seafood, six from beef and one from a clinical case of bloody diarrhoea were positive for Shiga toxins Stx1 and Stx2 and also for stx₁ and stx₂ genes. The seafood isolates produced either Stx2 alone or both Stx1 and Stx2, while the beef isolates produced Stx1 alone. The stx₁ gene of all the beef STEC was found to be of recently reported stx₁c type. All STEC strains and one non-STEC strain isolated from clam harboured EHEC-hlyA. Interestingly, though all STEC strains were negative for eae gene, two STEC strains isolated from seafood and one from a patient with bloody diarrhoea possessed STEC autoagglutinating adhesion (saa) gene, recently identified as a gene encoding a novel autoagglutinating adhesion.

STEC, Bead ELISA, Colony hybridization, stx_1c

1 Introduction

Shiga toxin-producing Escherichia coli (STEC) are considered a major cause of gastrointestinal disease in developed countries [1,2]. The nature of illness can range from mild form of diarrhea to more severe forms known as hemorrhagic colitis (HC) and hemolytic uremic syndrome (HUS). Contaminated food and water are major sources of STEC infections [3]. The food types most commonly associated with outbreaks of food poisoning are of bovine origin, in particular minced beef and beef burgers [4,5] and unpasteurised milk [6]. Most of the outbreaks previously reported, where food is thought to be the source of infection, have been associated with the consumption of beef [7,8,9,10,11,12]. However, it has been increasingly recognized that fresh vegetables and fruits other than beef or beef-product can be the sources of STEC infection [13,14]. Though the serotype O157:H7 is predominantly associated with human infections in US, UK, Canada and Japan, many other serotypes are prevalent in other part of the world and known to cause diseases similar in severity to that caused by O157:H7. Several recent outbreaks have been attributed to non-O157 STEC [15]. Non-O157 STEC is more often isolated from foods and animal feces [5,15,16,17]. However, not all STEC serotypes have been reported to be associated with severe human disease.

Of several virulence determinants reported to have been associated with STEC, the production of one or more Shiga toxins is considered to be most important [18]. Shiga toxins are of two types, Stx1 and Stx2 encoded by stx₁ and stx₂ genes, respectively. Variants of Stx2 have been reported which include Stx2c [19], Stx2d [20], Stx2e [21] and Stx2f [22]. Though originally believed to be a non-variant toxin, subtypes of Stx1 namely Stx1c and Stx1OX3 have recently been reported [23,24]. Other accessory virulence factors of STEC comprise production of a 94-kDa outer membrane protein called intimin, encoded by eae gene present on a 34-kb chromosomal pathogenecity island termed the locus for enterocyte effacement (LEE) [25] and an enterohemolysin (EHEC-hlyA) encoded by a large plasmid-borne ehxA gene [26,27]. In addition, a novel autoagglutinating adhesin, designated Saa, encoded by the saa gene present on a megaplasmid has recently been found in a LEE-negative non-O157 STEC strain that was isolated from a patient of HUS [28].

To date, more than 200 serotypes of E. coli are known to produce Shiga toxin(s) [15]. However, it is not possible to categorize STEC as pathogenic or non-pathogenic by serotyping alone. Within STEC, serotypes with pathogenic potential and serotypes that have not been associated with human disease exist. Although, Stx is considered the primary virulence factor, a combination of virulence factors is required for full virulence [15,29,30]. Since there is a need to determine the clinical significance of STEC strains associated with foods, it is important to identify the various virulence determinants commonly reported in them. Here, we report the characterization of STEC strains isolated in Mangalore, India and the usefulness of a combination of methods that include bead-ELISA, Vero cell cytotoxicity assay, PCR and colony hybridisation in typing STEC strains from foods.

2 Materials and methods

2.1 Escherichia coli strains

Escherichia coli strains used in this study were the representative serotypes isolated from seafood and beef in Mangalore [31]. The clinical isolate (ONT:H21) used in this study was isolated from a patient with bloody diarrhea. Briefly, 25 g of the samples were enriched in 225 ml modified EC broth (mEC) containing 20 μg ml⁻¹ of novobiocin. Following 6 h incubation, streaking was done on sorbitol MacConkey agar (SMAC) and both the sorbitol fermenting and non-fermenting colonies were selected for biochemical identification of E. coli. E. coli O145:H-isolated from diarrhoeal patient in 1985 in Japan was used as positive control for stx₁, stx₂, eae and ehxA genes. E. coli O91:H21 strain B2F1 was used as a positive control for saa gene. The test organisms were grown overnight in Luria Bertani (LB) broth in a shaker incubator at 37 °C. The standard E. coli strains carrying cloned stx₁ and stx₂ genes were grown in LB-medium supplemented with 50 μg ml⁻¹ of ampicillin.

2.2 Polynucleotide probes and colony hybridization

stx₁ and stx₂-specific polynucleotide probes were prepared from a 0.9-kb HincII–HpaI fragment of stx₁ gene [32] and a 0.86-kb PstI–SmaI fragment of A subunit of stx₂ gene [33]. These fragments were labeled with [α-³²P]dCTP by random primed labeling method (Amersham). The colony hybridization test was performed as described previously [34].

2.3 PCR

PCR was performed for stx₁, stx₂, eae, saa and ehxA, the virulence genes commonly associated with STEC. The sequences of the primers used for PCR are given in Table 1. An overnight culture diluted 1:10 in TE (10 mM Tris–HCl, pH 8.0, 1 mM EDTA) was boiled for 10 min and a 5-μl portion was used as template for PCR in a 50-μl reaction mix containing a 10× PCR buffer, 200 μM concentration of deoxynucleotide triphosphates, 1 μM concentration of each primer, and 2.5 U of Taq polymerase (Takara, Kyoto, Japan). PCR conditions used for stx-common [35,36], stx₁, stx₂[37], stx_1c[38], eae[39] and ehxA[39] were essentially the same as described previously. The stx₁ and stx₂ primers were used in a single PCR assay. For saa, the conditions for PCR consisted of an initial 94 °C denaturation step for 5 min followed by 30 cycles of 94 °C for 10 s, 58 °C for 30 s and 72 °C for 30 s and a final delay of 5 min at 72 °C. PCR was carried out in a PTC 100 Thermocycler (M.J Research, MA, USA). The amplified products of PCR were electrophoresed on a 2% agarose gel, stained with ethidium bromide (0.5 μg ml⁻¹) and photographed using Bio-Rad 2000 gel imaging system.

1

PCR primers used in this study

Primer	Sequence (5′–3′)	Target	Product size (bp)	Reference
Lin-F	gaacgaaataatttatatgt	stx1 and stx2	900	[35]
Lin-R	tactgacattgttagttt
VTcom-u	gagcgaaataatttatatgtg	stx1 and stx2	518	[36]
VTcom-d	tgatgatggcaattcagtat
VT1-u	caacactggatgatctcag	stx1 family	349	[37]
VT1-d	ccccctcaactgctaata
VT2-u	atcagtcgtcacactggt	stx2 family	110	[37]
VT2-d	ctgctgtcacagtgacaaa
Stx1c-1	ttttcacatgttacctttcct	stx1c	498	[38]
Stx1c-2	catagaaggaaactcattagg
EhxA-u	ggtgcagcagaaaaagttgtag	ehxA	1551	[39]
EhxA-d	tctcgcctgatagtgtttggta
Eae-u	aaacaggtgaaactgttgcc	eae	350	[39]
Eae-d	ctctgcagattaacctctgc
SAA-u	cgtgatgaacaggctattgc	saa	119	[48]
SAA-d	atggacatgcctgtggcaac

2.4 Bead-ELISA

The production of Stx1 and Stx2 by the strains was tested by bead-ELISA. Polyclonal antibodies raised in rabbits against purified Stx1 and Stx2 were used for coating the solid phase beads. The antigen–antibody complex was detected using a Fab'-horse radish peroxidase conjugated second antibody as described previously [36].

2.5 Vero cytotoxicity assay

The biological activity of the toxins was tested on Vero cell line as previously described [40].

2.6 Serotyping

The serotyping was performed by slide agglutination with antisera against O1–O173 and H1–H56 prepared at the Osaka Prefecture Public Health Institute, Osaka, Japan followed as originally described [41].

3 Results and discussion

Of the E. coli strains isolated from seafood, beef and a clinical sample included in this study, 11 were found to be producing either or both of Stx1 and Stx2 by bead ELISA. Among these, four isolates were from seafood, six from beef and one from a clinical case of bloody diarrhoea. All the Stx positive strains exhibited cytotoxicity to Vero cells. No strain was negative by bead ELISA and positive by Vero cell assay and vice versa thus exhibiting a 100% correlation between the two assays. The results of stx PCR correlated well with bead ELISA and Vero cell assay in the case of seafood and clinical isolates (Table 2). The primers common to stx₁ and stx₂ amplified the genes in four seafood isolates and one clinical isolate that were also positive by one or both of individual primers for stx₁ and stx₂. Two seafood isolates produced both the Shiga toxins and were also positive by stx₁ and stx₂ PCR. Similarly the only clinical isolate which produced Stx1 and Stx2 detected by bead ELISA was also positive by stx₁ and stx₂ PCR. However, differences were noticed in the PCR results of beef isolates. Six strains isolated from beef produced stx₁ when tested by bead-ELISA and were also cytotoxic to Vero cells. In all these strains, the common primers to stx₁ and stx₂ genes [36] could not amplify the stx genes. Interestingly, four strains were positive for stx₁ when tested by stx₁-specific primer set [37], while 2 were negative by both stx common and stx₁ primer sets but still produced Stx1 that was detected by bead-ELISA and cytotoxicity to Vero cells. The beef isolates of STEC were further characterised using stx common primers previously described by Lin et al. [35] and stx_1c primers [38]. All the beef isolates were positive by Lin common primers and also by stx_1c gene specific primers indicating that the stx₁ gene of these strains was an stx_1c variant. In probe hybridisation assay, all the seafood isolates that were positive by PCR for either or both of stx genes were also positive with the corresponding polynucleotide probes for stx genes. No strain was positive by probe hybridization and negative by PCR and vice versa. The clinical isolate was stx₁ and stx₂ probe positive. All the six beef isolates producing Stx1 showed faint probe hybridization signal with stx₁ probe. None of the isolates possessed the eae gene. However, two strains of stx₁ and stx₂-positive E. coli O76:H21 isolated from seafood and one stx₁ and stx₂-positive E. coli ONT:H21 isolated from the bloody stool of a diarrhoeal patient possessed saa gene. All STEC strains were positive by PCR for ehxA gene. The eae gene has been thought to be an important virulence factor in STEC. However, some severe cases of STEC infection recorded sporadically or in outbreaks have been caused by LEE-negative strains [42,43]. Paton et al. [28] reported the saa gene encoding a novel outer membrane protein, which appears to be an autoagglutinating adhesin on the large plasmid of certain strain of LEE-negative but not LEE-positive STEC strains. In this study, saa gene was detected in STEC strains isolated from a patient with severe bloody diarrhea and also from seafood.

2

Results of PCR, colony hybridization and bead ELISA

Strain		Serotype		Source		PCR/hybridization	Bead ELISA
			stx1	stx1c	stx2			ehxA	eae	saa	Stx1 (S/L)	Stx2 (S/L)
A3	ONT:NM	Clam	−	−	+/+		+	−	−	−/−	+/+
A4	ONT:H7	Clam	−	−	−		+	−	−	−/−	−/−
A5	O76:H21	Clam	+/+	−	+/+		+	−	+	+/+	+/+
A6	O76:H21	Fish	+/+	−	+/+		+	−	+	+/+	+/+
A7	ONT:NM	Fish	−	−	+/+	+	−	−	−/−	+/+
B2	O110:NM	Beef	+/+a	+	−	+	−	−	+/+	−/−
B3	O110:NM	Beef	+/+a	+	−	+	−	−	+/+	−/−
B4	O110:NM	Beef	+/+a	+	−	+	−	−	+/+	−/−
B5	O110:NM	Beef	+/+a	+	−	+	−	−	+/+	−/−
B6	O110:NM	Beef	+/+a	+	−	+	−	−	+/+	−/−
B7	O110:NM	Beef	−/+a	+	−	+	−	−	+/+	−/−
S3	ONT:H21	Clinical	+/+	−	+/+	+	−	+	+/+	+/+

S = culture supernatant; L= cells lysed by sonication.

All the strains were positive by stx common primers [35].

a

Indicates weak reaction.

Interesting differences were also observed between seafood isolates and beef isolates. Seafood isolates were positive for either stx₂ alone or for both stx₁ and stx₂. No isolate was positive for only stx₁ and none was stx_1c type. The beef isolates on the other hand were positive only for stx₁ variant gene, stx_1c. The detection of STEC in seafood is of particular importance as this reflects secondary contamination. All the isolates from beef were positive for stx₁ supporting previous hypothesis that bovine isolates are predominantly stx₁ positive. In general, cattle are considered reservoirs of STEC [44]. A few studies in India too suggest the bovine reservoir for STEC [39,45]. The isolation of non-O157 STEC strains from patients has also been reported from India [39,46]. Prevalence of STEC O157:H7 in beef imported to Malaysia from India has been previously reported [11]. The present study strengthens the earlier observations that STEC strains are widely prevalent in meat of animal origin, especially in beef in India. We previously reported [31] the detection of stx genes in supernatants of beef enrichments by direct PCR using the common primers described by Karch and Mayer [47]. But the isolates obtained from PCR positive samples were found to be negative when tested by the same primers. The same strains were included in this study and were found to be negative by common primers [36] but four of them were positive by primers for stx_1c gene. This is the first report of stx_1c gene in STEC strains from India. Though toxin assays such as the highly sensitive bead-ELISA and the Vero cell cytotoxicity assay help to identify strains possessing stx gene variants, such tests can be performed only in specialized laboratories and it is time consuming to perform the assays described in this study when large number of strains have to be tested. The presence of saa gene in non-O157 LEE-negative STEC strains reported from India for the first time is in agreement with some earlier reports of non-O157 strains possessing the saa gene [28]. Further studies are required to confirm the significance of Saa in the severity of STEC infection.

Acknowledgements

H.S.K. was a recipient of UNESCO fellowship for young scientist. The work in India on STEC was done under the Indo-German collaborative project.