Serotyping, stx₂ Subtyping, and Characterization of the Locus of Enterocyte Effacement Island of Shiga Toxin-Producing Escherichia coli and E. coli O157:H7 Strains Isolated from the Environment in France

Abstract

Twenty-seven Shiga toxin-producing Escherichia coli (STEC) strains were isolated from 207 stx-positive French environmental samples. Ten of these strains were positive for stx₁, and 24 were positive for stx₂ (10 were positive for stx_2vh-a or stx_2vh-b, 19 were positive for stx_2d, and 15 were positive for stx_2e). One strain belonged to serotype O157:H7, and the others belonged to serogroups O2, O8, O11, O26, O76, O103, O113, O121, O141, O166, and O174. The environment is a reservoir in which new clones of STEC that are pathogenic for humans can emerge.

Shiga toxin-producing Escherichia coli (STEC) can survive over long periods in slurries, farmyard manure, and sludge as well as on pasture land and in associated water systems. This observation has important implications for the spread of STEC to crops by direct application of manure or irrigation with infected water or the transmission of these strains directly to humans by contact with animals or contaminated soil (17, 25). Data on the presence of STEC in sewage or polluted water and slurries are limited and difficult to compare (11, 14, 16).

In contrast to a previous study that determined STEC prevalence in manure, slurries, and sewage sludge in France (24), the objective of the present work was to characterize STEC isolates. The detection of the virulence genes and the stx₂ subtypes enabled some speculation about the public health significance of STEC. The study of serotypes combined with that of the locus of enterocyte effacement island (LEE) allowed a better appreciation of the heterogeneity of the isolates and helped to determine whether the environment could be one of the main reservoirs in which new clones pathogenic for humans could emerge.

Four different types of sources were sampled: municipal sewage sites, waste storage lagoons, pig farms, and dairy cattle herds. Nine hundred eighty-eight samples were investigated: 95 farmyard manure samples, 201 bovine feces samples, 144 slurry samples, 10 compost samples, and 538 samples from outflows of sewage wastewater treatment plants. Twelve samples were taken from each source, and each source was sampled only once. The samples were collected over 1 year from two different geographical regions that are representative of France: Bourgogne and Rhone-Alpes. The bacterial enrichment procedure, the detection and isolation of STEC strains, and the identification of the isolates were performed as described by Vernozy-Rozand et al. (24).

All STEC strains were investigated for motility and biochemical properties (such as fermentation of sorbitol in 24 h, β-d-glucuronidase activity, and enterohemolytic phenotype) and were serotyped for their O and H antigens as previously described (3, 15, 18). The Vero cell toxicity assay was performed according to a protocol described previously (4, 5). Vero cells were cultivated in Dulbecco's modified Eagle medium (Gibco BRL). Genetic characterization of the isolates with regard to the presence of stx₁, stx₂, subtypes of stx₂, and types of eae, tir, espA, espB and the insertion of the LEE at selC, pheU, or pheV, was performed as previously described (1, 2, 8, 9, 21; V. Livrelli, Y. Bertin, N. Pradel, N. Blanchet, B. Joly, and C. Martin, unpublished data).

Twenty-one percent (207 of 988) of the samples were stx positive by PCR. Samples from clarifiers of waste storage lagoons and wastewater treatment plants had contamination rates of 80 and 48%, respectively. These results are in accordance with those obtained by Holler et al. (13), who sampled the inflow and outflow of sewage treatment plants in northern Germany and found 90% of the samples to be positive for stx₁ or stx₂.

We isolated 27 STEC strains from the 207 stx-positive samples (13%) by colony hybridization with stx₁ and stx₂ gene probes. Three of the 27 STEC isolates belonged to serogroups O26, O103, and O157. The other serotypes are given in Table 1. Serogroups O26, O55, O111, O103, and O157:H7 are the STEC serogroups isolated most frequently from French human patients with hemorrhagic colitis and hemolytic uremic syndrome (12). Twenty-four STEC strains possessed the stx₂ gene, and 10 possessed the stx₁ gene. Seven strains (26%) possessed both stx₁ and stx₂. Seven (26%) strains were eaeA positive, and 18 (66%) were ehxA positive. Nine STEC strains carried three different virulence factors (stx, eae, and ehxA) (Table 1). These strains should be considered potentially pathogenic for humans.

TABLE 1.

Phenotypic and genotypic characterization of isolates^a

Origin of strain	Strain no.	Sample type	Serotypeb	Hemolysin phenotypec	Status regarding:					stx2 variant(s)
					eae gene	ehxA gene	Verocytotoxin production	stx1 gene	stx2 gene
Wastewater treatment plant	V1	Slurry	O166:H28	ehly	N	P	P	N	P	stx2e, stx2d-Ount
	V2	Clarifier	O113:H4	N	P	P	P	N	P	stx2-NV206, stx2d
	V3	Slurry pit	O2:H27	N	N	P	P	N	P	stx2-EDL933, stx2d
	V4	Aerator	O76:H19	ehly	P	P	P	P	P	stx2e, stx2d-Ount
	V5	Clarifier	O26:H11	ehly	P	P	P	P	N	N
	V6	Sludge	O76:H19	ehly	N	P	P	P	P	stx2e, stx2vh-a, stx2d-Ount
	V7	Clarifier	O76:H19	ehly	P	P	P	P	P	stx2e, stx2vh-a, stx2d-Ount
	V8		O76:H19	ehly	N	P	P	N	P	stx2-EDL933
	V9	Aerator	O157:H7	ehly	P	P	P	P	P	stx2vh-a, stx2d-Ount
	V10	Clarifier	O11:H43	N	N	N	P	N	P	N
Pig farms	V11	Feces	Ont:H4	alpha	N	N	P	N	P	stx2e
	V12	Feces	O141:H4	alpha	N	N	P	N	P	stx2e
	V13	Feces	O8:H19	−N	N	N	P	N	P	stx2-NV206, stx2e, stx2d
	V14	Feces	O103:H2	ehly	P	P	P	N	P	stx2e, stx2d
	V15	Feces	Ont:H19	N	N	P	P	P	P	stx2vh-b, stx2e, stx2d-Ount
	V16	Feces	O121:H−	N	N	N	P	N	P	stx2e
Dairy cattle herd	V17	Feces	Ont:H38	N	P	P	P	N	P	stx2vh-a, stx2d-Ount
	V18	Feces	Orough:H25	ehly	P	P	P	N	P	stx2e, stx2vh-a, stx2d-Ount
	V19	Feces	Orough:H25	alpha	P	P	P	N	P	stx2vh-a, stx2d-Ount
	V20	Feces	O113:H4	N	N	N	P	N	P	stx2-NV206, stx2e, stx2d
	V21	Manure	O174:H21	N	N	N	P	N	P	stx2vh-b, stx2e, stx2d-Ount
	V22	Manure	O174:H21	N	N	N	P	N	P	stx2vh-b, stx2e, stx2d-Ount
	V23	Manure	O174:H21	ehly	N	P	P	P	P	stx2e, stx2-EDL933, stx2d
	V24	Slurry pit	O8:H19	ehly	N	P	P	P	N	N
	V25		Ont:H8	N	N	P	P	P	N	N
	V26	Feces	O174:H−	N	N	N	P	N	P	stx2vh-a, stx2d-Ount
	V27	Manure	O8:H19	ehly	N	P	P	P	P	stx2-EDL933, stx2d

^aP, positive; N, negative.

^bOrough, rough; Ont, O type not corresponding to any serogroup between O1 and O174.

^cehly, hemolysis after a 22-h incubation step; alpha, hemolysis after 3- to 4-h incubation step.

Epidemiological studies, together with in vivo and in vitro experiments, have revealed that stx₂ is the most important virulence factor associated with severe human disease (6, 19). The genes encoding these toxins (stx genes) are encoded by lambdoid bacteriophages. At least 12 stx₂ gene variants have been described (1, 2, 23). Some of these variants, which are associated with STEC strains isolated from specific hosts such as sheep (stx_2d) and pigs (stx_2e), are probably less pathogenic for humans (10, 26). The type of variant could thus reflect not only the origins and relationships but also the virulences of the different STEC strains.

In our collection of environmental STEC strains isolated in France, the most frequent stx₂ subtype was stx_2d (70% of the isolates), followed by stx_2e (55%) and stx_2vh-a or stx_2vh-b (37%). The stx_2e variants were isolated from pig samples but also from a dairy cattle herd, a finding which is not in accordance with data given in the literature (7, 20) and suggests the spread of STEC from pigs to cattle.

The presence of more than one stx₂ gene, which was observed in 70% of our STEC isolates, could contribute to the level of toxin production and thus to the virulence of environmental isolates.

In a previous study, an absolute correlation between the presence of eae and the presence of the other genes of the LEE was found (2). The LEE gene was not detected in eae-negative strains. On the basis of these findings, we used the eae gene to screen for LEE-positive STEC isolates. EspA, EspB, Tir, and intimin proteins encoded by different LEEs show high degrees of sequence polymorphism. To date, 10 distinct variants of eae and three variants of espA, espB, and tir have been described (22, 27). Previous studies have indicated that some serotypes are highly associated with a particular intimin variant (8, 9). These associations were confirmed in our study; the O157 serogroup is associated with eaeγ, the O26 serogroup is associated with eaeβ, and the O103 serogroup is associated with eaeɛ. No other eae variants, including eaeα, were found. In this study, we detected only three distinct pathotypes, and except for the O103 STEC strain, tir, espA, and espB genes were always of the same type within a pathotype. We also observed a high degree of heterogeneity among the seven eae-positive isolates from our collection relative to the LEE integration site as well as the sequences of the right and left ends of the LEE (Table 2).

TABLE 2.

Characterization of the LEE

Origin of strain	Strain no.	Serotype	Subtype of gene				PCR analysis result fora:
							selC site			pheU site				pheV site
			eae	espA	espB	tir	Intact (K260/K295)	Right side (K260/K255)	Left side (K295/K296)	Intact (K913/K916)	Right side		Left side (P54/P15)	pheV (913/914)	Intact (pheVF/pheVR)	Right side (P64/P86)
											K913/K917	P119/P32
Wastewater treatment plant	V4	O76:H19	β	β	β	β	N	N	N	N	P	N	N	N	N	P
	V5	O26:H11	β	β	β	β	N	N	N	N	P	N	N	N	N	P
	V9	O157:H7	γ	γ	γ	γ	N	P	P	P	N	N	N	P	N	P
Pig farm	V14	O103:H2	ɛ	βv	β	β	P	N	N	N	N	N	N	N	N	N
Dairy cattle herd	V17	Ont:H38	β	β	β	β	N	N	N	N	P	N	N	N	N	N
	V18	Orough:H25	β	β	β	β	N	N	N	N	P	N	P	N	N	P
	V19	Orough:H25	β	β	β	β	N	N	N	N	P	N	N	N	N	N

^aN, negative; P, positive. K260, K295, K255, K296, K913, K916, K917, P119, P32, P54, P15, 913, 914, pheVF, pheVR, P64, and P86 are primers.

This study demonstrated that the environment is a reservoir in which new clones of STEC that are pathogenic for humans could emerge. A comparison with STEC strains isolated from patients with hemolytic uremic syndrome will be undertaken in the near future.

Our results emphasize that appropriate handling and use of manure, slurry, and sewage sludge are necessary so that contamination of the environment and food by STEC can be prevented.