Subtyping of Escherichia coli O157:H7 Strains Isolated from Human Infections and Healthy Cattle in Argentina

Abstract

Shiga toxin–producing Escherichia coli (STEC) cause nonbloody (NBD) and bloody diarrhea (BD), and hemolytic uremic syndrome (HUS). Cattle have been described as their main reservoir. STEC O157:H7 is recognized as the predominant serotype in clinical infections, but much less is known about the dominant subtypes in humans and animals or their genetic relatedness. The aims of this study were to compare the STEC O157 subtypes found in sporadic human infections with those in the bovine reservoir using stx-genotyping, phage typing, and XbaI–pulsed-field gel electrophoresis (PFGE), and correlate the subtypes with the severity of clinical manifestations. The 280 STEC O157:H7 strains collected included in this study were isolated from HUS (n=122), BD (n=69), and NBD (n=30) cases, and healthy carriers (n=5), and from bovines (n=54) in the abattoirs. The stx-genotyping showed that stx ₂/stx _2c(vh-a) was predominant in human (76.1%) and in bovine strains (55.5%), whereas the second more important genotype was stx ₂ (20.8%) in human and stx _2c(vh-a) (16.7%) in cattle strains. In human strains, PT4 (37.6%), PT49 (24.3%), and PT2 (18.6%) were the most frequent PTs (80.5%). In bovine isolates, PT2 (26%), PT39 (16.7%), and PT4 and PT49 (11.1% each) were predominant. By XbaI-PFGE, all 280 strains yielded 148 patterns with 75% similarity, and 169 strains were grouped in 37 clusters. Identical PT-PFGE-stx profile combinations were detected in strains of both origins: PT4-AREXH01.0011-stx ₂/stx _2c(vh-a) (12 humans and one bovine), PT4-AREXH01.0543-stx ₂/stx _2c(vh-a) (one human and four bovines), PT2-AREXH01.0076-stx ₂/stx _2c(vh-a) (one human and four bovines), PT49-AREXH01.0175-stx ₂/stx _2c(vh-a) (seven humans and one bovine), and PT49-AREXH01.0022-stx ₂/stx _2c(vh-a) (seven humans and one bovine). No correlation was found among the stx-genotypes, the phage type, and the clinical symptoms.

Introduction

S higa toxin–producing Escherichia coli (STEC) are important foodborne pathogens that cause nonbloody (NBD) and bloody diarrhea (BD), and hemolytic uremic syndrome (HUS) in humans (Griffin and Tauxe, 1991). STEC O157:H7 is the dominant serotype associated with sporadic cases and outbreaks in different parts of the world (Rangel et al., 2005; Pennington, 2010).

Cattle have been recognized as the main reservoir of E. coli O157 worldwide (Gyles, 2007), including Argentina (Masana et al., 2010; Tanaro et al., 2010). The main vehicles for acquiring the infection are the consumption of contaminated food, especially undercooked ground beef, fresh produce, water, and direct contact with infected persons or animals (Caprioli et al., 2005).

Disease-associated human isolates of E. coli O157:H7 are characterized by the presence of specific sets of virulence genes, including those encoding Shiga toxins (stx ₁, stx ₂), intimin (eae), and hemolysin (ehxA) (Karmali, 2004).

In Argentina, post-enteric HUS is endemic, and approximately 400 new cases are reported annually, with an estimated rate of 12 cases per 100,000 children under 5 years old in 2009 (Rivas et al., 2011), and STEC O157 is the serogroup most commonly identified (Rivas et al., 2006).

There are a number of genotyping methods used for epidemiological studies of STEC O157, such as polymerase chain reaction (PCR) typing of several virulence factors, lineage-specific polimorphism assay using six markers (LSPA-6), restriction fragment length polymorphism (RFLP), pulsed-field gel electrophoresis (PFGE), and variable number tandem repeat analysis (MLVA) (Thomson-Carter, 2001; Yang et al., 2004; Hyytia-Trees et al., 2006). Some studies have reported differences in the genotype and virulence markers of STEC O157 strains from human patients and cattle (Eklund et al., 2002; Mora et al., 2004; Roldgaard et al., 2004; Barker et al., 2007; Nakamura et al., 2008; Aspán and Erikson, 2010; Lee et al., 2011; Kawano et al., 2012).

In this study, PFGE, phage typing, and stx-PCR-RFLP analysis of DNA fragments obtained by PCR were used to compare the STEC O157 subtypes found in sporadic human infections with those in the Argentine bovine reservoir, and to correlate the subtypes with the severity of clinical manifestations.

Methods

Bacterial strains

A total of 280 STEC O157 strains were included in this study. Between November 2006 and April 2008, 226 strains received at the National Reference Laboratory as part of the surveillance of HUS and diarrheal diseases were studied. The strains were isolated from HUS (n=122), BD (n=69), NBD (n=30) cases, and healthy carriers (n=5). For comparison purposes, 54 STEC O157 strains isolated from healthy beef cattle in a survey carried out in beef abattoirs during the same period (Masana et al., 2010) were included as representative of the Argentine bovine reservoir.

Phenotypic and genotypic characterization of isolates

Confirmation of isolates as E. coli was performed through biochemical tests (Ewing, 1986) and serotyping (Ørskov and Ørskov, 1984) using the somatic and flagellar antisera provided by the Instituto Nacional de Producción de Biológicos–ANLIS “Dr. Carlos G. Malbrán.”

Antimicrobial susceptibility to amikacin, ampicillin, ciprofloxacin, cloramphenicol, colistin, gentamicin, nalidixic acid, nitrofurantoin, streptomycin, tetracycline, and trimethoprim-sulfamethoxazole was established by standard methods (CLSI, 2011).

In all isolates, stx ₁, stx _2, and rfb _O157 genes were detected by a multiplex PCR as described by Leotta et al. (2005), whereas eae, ehxA, and fliC _H7 genes were detected by PCR as described by Karch et al. (1993), Schmidt et al. (1995), and Gannon et al. (1997), respectively.

To determine Stx production, cytotoxicity assays on Vero cells were performed (Karmali et al., 1985). Enterohemolysis was determined on sheep blood agar plates (Beutin et al., 1989).

Subtyping of isolates

The analysis of stx ₁ variants was conducted according to Zhang et al. (2002). Genotyping of stx ₂ variants was done by RFLP analysis of the B-subunit-encoding DNA fragments obtained by PCR (Tyler et al., 1991).

Phage typing was performed with the method described by Ahmed et al. (1987) and extended by Khakhria et al. (1990), employing a set of 16 phages provided by the Canadian Centre for Human and Animal Health, Winnipeg, Manitoba, Canada, which allow the differentiation of 87 phage types (PTs).

PFGE protocols and data analysis of E. coli O157:H7 isolates were performed using the restriction endonucleases XbaI and BlnI (Promega, Madison, WI) following the standardized PulseNet methods from the Centers for Disease Control (Ribot et al., 2006). The Dice coefficient and the unweighted pair group method with arithmetic mean (UPGMA) were used to generate dendrograms with 1.5% tolerance values. The strains were grouped in a cluster when they showed identical XbaI-PFGE pattern (100% similarity).

Statistical analysis

The statistical analysis of the frequency of stx-genotypes and PTs between human and cattle strains was performed by the two-tailed Fisher's exact test, using InStat version 3.05 (GraphPad Software, San Diego, CA). A p-value of <0.05 was considered statistically significant.

Results

Characterization of isolates

All 280 STEC O157:H7 strains were non-sorbitol-fermenting and β-d-glucuronidase negative. Among the strains of both origin, the biotype C (rhamnose+/dulcitol+) was predominant (93.6%), whereas 14 (5%) strains (12 humans and two bovines) belonged to biotype B (rhamnose−/dulcitol+), and four (1.4%) human strains belonged to biotype D (rhamnose+/dulcitol−).

All human and bovine strains harbored eae, ehxA, rfbO₁₅₇, and fliC _H7 genes. Most of the strains that carried the ehxA gene were hemolytic on sheep blood agar (95%); however, this activity was not detected in 14 strains of bovine origin. Expression of toxicity on Vero cells assays and flagella presence by slide agglutination was demonstrated in all isolates. Two hundred and nineteen (96.9%) of 226 human strains and 48 (89%) of 54 bovine strains were susceptible to all antibiotics assayed.

Subtyping of STEC O157 isolates

All 280 strains harbored the stx ₂ gene, and 14 (5%) also carried the stx ₁ gene. The stx-genotype analysis revealed that stx ₂/stx _2c(vh-a) was predominant (72.1%) in strains of both origin, with a higher frequency in human (76.1%) than in bovine (55.5%) strains (p<0.05). The stx ₂ variant was detected in both human (20.8%) and bovine (9.2%) strains. In human isolates, the stx _2c(vh-a) variant was detected in a low frequency (0.9%), whereas in bovine strains it was the second stx-genotype found (16.7%) (p<0.0001). The stx ₁ gene in combination with other stx genes was more frequently detected in animal (16.6%) than in human (2.2%) strains (Table 1).

Table 1.

Frequency of stx-Genotypes of Escherichia coli O157:H7 Strains Isolated from Humans and Healthy Bovines

	No. of strains (%)
	Total	Human	Bovine
stx-genotype	(n=280)	(n=226)	(n=54)
stx ₂ /stx _2c(vha)	202 (72.1)	172 (76.1)	30 (55.5)^a
stx ₂	52 (18.6)	47 (20.8)	5 (9.2)
stx _2c(vha)	11 (3.9)	2 (0.9)	9 (16.7)^a
stx₁ /stx ₂/stx _2c(vha)	9 (3.2)	5 (2.2)	4 (7.4)
stx₁ /stx ₂	4 (1.4)	0	4 (7.4)^a
stx₁ /stx _2c(vha)	1 (0.4)	0	1 (1.8)
stx ₂UT	1 (0.4)	0	1 (1.8)

Frequencies between human and bovine strains are statistically

significant (p<0.05).

UT, untypeable.

A total of 20 PTs were identified among the 280 strains. Two hundred and fourteen (94.7%) human strains were categorized into 15 different PTs, and 12 (5.3%) strains belonged to the react but do not conform (RDNC) category of the current phage typing scheme. PT4 (37.6%), PT49 (24.3%), and PT2 (18.6%) were the most frequent (80.5%) PTs found. Bovine isolates were grouped into 12 PTs, and two (3.7%) strains were RDNC. PT2 (26%), PT39 (16.7%), and PT4 and PT49 (11.1% each) were predominant, representing 64.9% of the total. PT4 was strongly associated to human strains and PT39 to bovine strains (p<0.0001) (Table 2). PT14, PT24, PT36, PT37, PT40, PT47, PT50, and PT54 were only detected in human strains, whereas PT21, PT31, PT33, PT43, and PT51 were only found in bovine strains.

Table 2.

Frequency of Predominant Phage Types of Escherichia coli O157:H7 Strains Isolated from Humans and Healthy Cattle

	No. of strains (%)
	Total	Human	Bovine
Phage type	(n=280)	(n=226)	(n=54)	P
PT2	56 (20)	42 (18.6)	14 (26)	NS
PT4	91 (32.5)	85 (37.6)	6 (11.1)	<0.0001
PT39	10 (3.6)	1 (0.4)	9 (16.7)	<0.0001
PT49	61 (21.8)	55 (24.3)	6 (11.1)	<0.05
PT54	10 (3.6)	10 (4.4)	0	NS

Relationships between stx-genotype and PT in STEC O157:H7 strains of both origins are shown in Table 3. In human strains, the stx ₂/stx _2c(vh-a) genotype was mainly associated with PT49 (52/172, 30.2%), followed by PT4 (50/172, 29.1%) and PT2 (39/172, 22.7%), whereas in bovine strains this stx-genotype was mainly related to PT2 (12/30, 40%). It is important to highlight that the stx ₂ genotype was strongly associated with PT4 (32/47, 68.1%) in human strains, and stx _2c(vh-a) to PT39 (6/9, 66.7%) in bovine strains. All four stx ₁/stx ₂ animal strains belonged to PT21.

Table 3.

Relationship Between stx-Genotypes and Predominant Phage Types in Escherichia coli O157:H7 Strains of Human and Bovine Origin

stx-genotype	Origin	No.	Phage type (%)
stx ₂ /stx _2c(vha)	Human	172	PT49 (30.2), PT4 (29.1), PT2 (22.7)
	Bovine	30	PT2 (40), PT4 and PT49 (20, each one), PT26 (6.7)
stx ₂	Human	47	PT4 (68.1), PT2 (6.4), PT47, PT49, and PT54 (4.3, each one)
	Bovine	5	PT2, and PT33 (40, each one), PT8 (20)
stx _2c(vha)	Human	2	PT8
	Bovine	9	PT39 (66.7), PT43 (22.2), PT8 (11.1)
stx ₁/stx ₂/stx _2c(vha)	Human	5	PT4 (60), PT26, and PT49 (20, each one)
	Bovine	4	PT10, PT26, PT31, PT39
stx ₁/stx ₂	Human	0
	Bovine	4	PT21
stx ₁/stx _2c(vha)	Human	0
	Bovine	1	PT39
stx ₂UT	Human	0
	Bovine	1	PT33

UT, untypeable.

By XbaI-PFGE, all 280 STEC O157 strains generated 148 different patterns with at least 75% similarity. One hundred and sixty-nine strains were grouped in 37 clusters, and 111 strains showed unique patterns. Twenty-four clusters grouped exclusively human strains, eight were specific for bovine strains, and five clusters (III, VI, XVIII, XXIII, XXIX) included both human (38) and bovine (12) strains (Table 4).

Table 4.

Distribution of Escherichia coli O157:H7 Strains from Human and Bovine Origin in XbaI-PFGE Clusters

Cluster	XbaI-PFGE patterns	stx-genotypes	Phage type	Origin (n)
I	AREXHX01.0390	stx ₂/stx _2c(vh-a)	2	Human (5)
II	AREXHX01.0267	stx ₂/stx _2c(vh-a)	49	Bovine (3)
III	AREXHX01.0022	stx ₂/ stx _2c(vh-a)	49	Human (7), bovine (1) ^a
		stx _2c(vh-a)	4	Human (1)
		stx ₂/stx _2c(vh-a)	14	Human (3)
IV	AREXHX01.0093	stx ₂/stx _2c(vh-a)	49	Human (2)
V	AREXHX01.0189	stx ₂/stx _2c(vh-a)	49	Human (2)
VI	AREXHX01.0175	stx ₂/ stx _2c(vh-a)	49	Human (7), bovine (1)
VII	AREXHX01.0153	stx ₂/stx _2c(vh-a)	49	Human (10)
		stx ₂	4	Human (2)
		stx ₂/stx _2c(vh-a)	RDNC	Human (1)
VIII	AREXHX01.0049	stx ₂	2	Human (1)
		stx ₂/stx _2c(vh-a)	RDNC	Human (5)
IX	AREXHX01.0507	stx _2c(vh-a)	39	Bovine (6)
X	AREXHX01.0514	stx _2c(vh-a)	43	Bovine (2)
XI	AREXHX01.0200	stx ₂/stx _2c(vh-a)	4	Human (2)
		stx ₂	4	Human (5)
XII	AREXHX01.0193	stx ₂/stx _2c(vh-a)	4	Human (2)
XIII	AREXHX01.0341	stx ₂/stx _2c(vh-a)	4	Human (3)
XIV	AREXHX01.0038	stx ₂/stx _2c(vh-a)	4	Human (2)
XV	AREXHX01.0469	stx ₂	54	Human (1)
		stx ₂	8	Human (1)
XVI	AREXHX01.0057	stx ₂/stx _2c(vh-a)	4	Human (3)
		stx ₂	4	Human (2)
		stx ₂/stx _2c(vh-a)	49	Human (2)
XVII	AREXHX01.0007	stx ₂/stx _2c(vh-a)	4	Human (2)
XVIII	AREXHX01.0011	stx ₂ /stx _2c(vh-a)	4	Human (12), bovine (1)
		stx ₂/stx _2c(vh-a)	RDNC	Bovine (1)
		stx ₂	4	Human (5)
		stx ₂/stx _2c(vh-a)	14	Human (1)
XIX	AREXHX01.0064	stx ₂	4	Human (15)
		stx ₁/stx ₂/stx _2c(vh-a)	4	Human (2)
XX	AREXHX01.0012	stx ₂	4	Human (1)
		stx ₂/stx _2c(vh-a)	4	Human (1)
XXI	AREXHX01.0086	stx ₂/stx _2c(vh-a)	2	Human (1)
XXII	AREXHX01.0458	stx ₁/stx ₂/stx _2c(vh-a)	26	Bovine (1)
		stx ₂/stx _2c(vh-a)	26	Bovine (2)
XXIII	AREXHX01.0543	stx ₂/ stx _2c(vh-a)	4	Human (1), bovine (4)
XXIV	AREXHX01.0058	stx ₂/stx _2c(vh-a)	26	Human (1)
		stx ₂/stx _2c(vh-a)	4	Human (1)
XXV	AREXHX01.0517	stx ₂	33	Bovine (2)
		stx ₂ UT	33	Bovine (1)
XXVI	AREXHX01.0454	stx ₂/stx _2c(vh-a)	37	Human (1)
		stx _2c(vh-a)	8	Human (1)
XXVII	AREXHX01.0143	stx ₂/stx _2c(vh-a)	2	Human (2)
XXVIII	AREXHX01.0095	stx ₂/stx _2c(vh-a)	2	Human (4)
XXIX	AREXHX01.0076	stx ₂/ stx _2c(vh-a)	2	Human (1), bovine (4)
XXX	AREXHX01.0001	stx ₂/stx _2c(vh-a)	2	Human (1)
		stx ₂/stx _2c(vh-a)	24	Human (1)
		stx ₂/stx _2c(vh-a)	49	Human (1)
XXXI	AREXHX01.0427	stx ₂	4	Human (1)
		stx ₁/stx ₂/stx _2c(vh-a)	4	Human (1)
XXXII	AREXHX01.0460	stx ₂/stx _2c(vh-a)	2	Bovine (3)
XXXIII	AREXHX01.0013	stx ₂/stx _2c(vh-a)	49	Human (1)
		stx ₂	49	Human (1)
XXXIV	AREXHX01.0449	stx ₂/stx _2c(vh-a)	2	Human (2)
XXXV	AREXHX01.0145	stx ₂/stx _2c(vh-a)	54	Human (2)
XXXVI	AREXHX01.0545	stx ₂/stx _2c(vh-a)	2	Bovine (2)
XXXVII	AREXHX01.0544	stx ₁/stx ₂	21	Bovine (4)

Identical XbaI-PFGE-stx-PT profile combinations detected in strains of both origins are shown in boldface.

RDNC, react but does not conform to the current phage typing scheme; UT, untypeable.

Using the three subtyping techniques, the following PT-XbaI-PFGE-stx profile combinations were detected in strains of both origins: PT4-AREXH01.0011-stx ₂/stx _2c(vh-a) (12 humans and one bovine), PT4-AREXH01.0543-stx ₂/stx _2c(vh-a) (one human and four bovines), PT2-AREXH01.0076-stx ₂/stx _2c(vh-a) (one human and four bovines), PT49-AREXH01.0175-stx ₂/stx _2c(vh-a) (seven humans and one bovine), and PT49-AREXH01.0022-stx ₂/stx _2c(vh-a) (seven humans and one bovine) (Fig. 1).

FIG. 1.

Clonal relatedness by XbaI- and BlnI-PFGE, and phage types in human and bovine Escherichia coli O157:H7 stx2/stx2c (vh-a) strains with identical profile combinations. HUS, hemolytic uremic syndrome; BD, bloody diarrhea; NBD, nonbloody diarrhea; HC, healthy carriers.

Using BlnI as second enzyme, all human and bovine strains of PT49-AREXH01.0175-stx ₂/stx _2c(vh-a) and PT49-AREXH01.0022-stx ₂/stx _2c(vh-a) profiles were not discriminated. The bovine strain belonging to the PT4-AREXH01.0011-stx ₂/stx _2c(vh-a) profile combination showed identical BlnI-PFGE pattern with five of 12 human strains. Human strains belonging to PT4-AREXH01.0543-stx ₂/stx _2c(vh-a) and PT2-AREXH01.0076-stx ₂/stx _2c(vh-a) profile combinations could be discriminated from the bovine strains, but showed patterns with 91.9% and 89.5% similarity, respectively (Fig. 1).

The analysis of the most frequent PTs and stx-genotypes revealed an equal distribution among strains associated with different clinical symptoms (Table 5). STEC O157 strains harboring stx ₂/stx _2c(vh-a) genes belonging to PT4, PT49, and PT2 were predominant.

Table 5.

Relationship Between Predominant Phage Types, stx-Genotypes, and Clinical Symptoms

		Percentage by clinical symptom
		HUS	BD	NBD
Phage type	stx-genotype	(n=122)	(n=69)	(n=30)
PT4	stx ₂ /stx _2c(vha)	23.8	21.7	20.0
	stx ₂	18.0	14.5	0
	stx₁ /stx ₂/stx _2c(vha)	0.8	1.4	3.3
PT49	stx ₂ /stx _2c(vha)	23.0	20.3	23.3
	stx ₂	0	1.4	3.3
	stx₁ /stx ₂/stx _2c(vha)	0	1.4	0
PT2	stx ₂ /stx _2c(vha)	16.4	15.9	20.0
	stx ₂	1.6	0	3.3
PT54	stx ₂ /stx _2c(vha)	4.1	4.3	0
	stx ₂	0.8	0	3.3
PT14	stx ₂ /stx _2c(vha)	4.1	3.1	3.3

HUS, hemolytic uremic syndrome; BD, bloody diarrhea; NBD, nonbloody diarrhea.

Discussion

Previous studies indicate that the clinical outcome of STEC infection depends on the stx genotype of the infecting strain (Jelacic et al., 2003; Persson et al., 2007), and that there exists an increased risk for developing HUS when both stx ₂ and eae genes are present (Böerlin et al., 1999). By stx-genotyping, we found that the stx ₂/stx _2c(vh-a) genotype was predominant in strains isolated from both human (76.1%) and bovine (55.5%). Moreover, stx ₂ was detected in human and bovine strains (20.8% vs. 9.2%), whereas the stx _2c(vh-a) gene was predominant in the animal reservoir (16.7%) but not in humans (0.9%). The stx ₁ gene was always detected together with other genes. The stx ₁/stx ₂/stx _2c(vh-a) genotype was harbored by nine (3.2%) strains: five of them isolated from HUS (n=2), BD (n=2), and NBD (n=1) cases.

A number of studies have documented that stx ₂/stx _2c(vh-a) genotype is more often associated with HUS than other stx ₂ genotypes (Friedrich et al., 2002; Eklund et al., 2002). Our findings that stx ₂/stx _2c(vh-a) is prevalent in strains of both origins is relevant because this stx-genotype was also predominant in other countries. In Sweden, Aspán and Erikson (2010) reported that STEC O157:H7 strains characterized as being PT4:stx ₂/stx _2c(vh-a) were the cause of most cattle-to-human transmitted STEC in the 1996–2002 period. In Finland, Eklund et al. (2002) found that 64% of human STEC O157 isolates carried stx ₂/stx _2c genes, and the virulence profile PT2:stx ₂/stx _2c/eae/ehxA was significantly more frequently associated with HUS and BD than other profiles. Moreover, Manning et al. (2008) found that this stx-genotype was predominant among strains of the clade 8 lineage associated with two unusually severe outbreaks linked to fresh produce. Recent studies performed in Argentina using LSPA-6 to identify lineages (Yang et al., 2004) and clade typing (Manning et al., 2008) show that lineage I/II-clade 8 strains predominate in human isolates (>80%) and may explain the high incidence of HUS in our country (unpublished data).

In contrast, Nakamura et al. (2008) in Japan found that the predominant stx-genotype in human patients was stx ₂ (68.7%), whereas the strains from asymptomatic carriers possessed only stx _2c(vh-a) (47.9%), and the stx ₁ gene was found in a few isolates (1.5%). Meanwhile, cattle harbored mainly stx _2c(vh-a) (58.4%), followed by stx ₂ (26.0%) and stx ₂/stx _2c(vh-a) (10.4%).

We found 20 different known PTs among the 280 isolates of STEC O157 studied, but only three types were predominant in human strains, PT4 (37.6%), PT49 (24.3%), and PT2 (18.6%), whereas four were predominant in bovine strains, PT2 (26%), PT39 (16.7%), and PT4 and PT49 (11.1% each). In Denmark, Roldgaard et al. (2004) found that PT2 (19%) and PT4 (28%) were also predominant in human isolates, but these PTs were only identified in 2% and 8%, respectively, of the bovines isolates. In the same study, PT8 and PT14 constituted approximately 35–40% of isolates from both origins. PT8 was responsible for 27% of human cases, and PT14 was the most predominant PT in cattle (22%). In the present study, PT8 was found in a very low frequency in human (1.3%) and bovine (3.7%) isolates, and PT14 was only found in clinical isolates (3.5%). Rivas et al. (2006) found one PT8:stx ₂/stx _2c(vh-a) strain isolated from a BD case who developed HUS, and three PT14:stx ₂/stx _2c(vh-a) strains isolated from BD cases, during a prospective case-control study carried out in 2001–2002. Chinen et al. (2001) found 2/11 (18.2%) strains of PT14 harboring the stx ₂/stx _2c(vh-a) and stx _2c(vh-a) genotypes, which were isolated from ground beef and fresh sausage, respectively, during a survey performed in 136 retail meats. Moreover, 2/20 (10%) E. coli O157:H7 strains characterized as PT8:stx ₂/stx _2c(vh-a) and PT14:stx ₁/stx_2c(vh-a) were isolated from undercooked beef burger during sampling procedures in fast food restaurants (Chinen et al., 2009). In Spain (Mora et al., 2004), Germany (Beutin et al. 2002), and Denmark (Roldgaard et al., 2004), PT8 strains showed a high association with stx ₁ and were mainly non-motile. In Argentina, all 280 strains were motile, and this could be another important virulence factor.

Between 1998 and 2004, PT21/28 was described as the most frequent PT in the United Kingdom, comprising over 50% of the positive cattle isolates and reported human cases, suggesting a link between bovine shedding and human infection (Pearce et al., 2009). In Argentina, PT21/28 has never been found.

We found a marked overlap between the PT-stx profiles in strains isolated from cattle and those recovered from human diseases. The profiles PT49:stx ₂/stx _2c(vh-a) (30.2% vs. 20%), PT4:stx ₂/stx _2c(vh-a) (29.1% vs. 20%), and PT2:stx ₂/stx _2c(vh-a) (22.7% vs. 40%) were predominant in strains of both origins (Table 3). However, some PT-stx-genotype profiles were exclusively found in the bovine reservoir. These findings are consistent with the notion that only a subset of bovine O157:H7 is implicated in human disease.

In this study, it was not possible to establish a relationship between the stx-genotypes found in human strains and the clinical symptoms, probably due to the high rate of stx ₂/stx _2c(vh-a) genotype (>60%) observed in Argentina throughout the years (Rivas et al., 2011). Other methods, such as stx expression and the amount of toxin produced, could be used to further relate the severity of the infections to particular subtypes.

The molecular surveillance of STEC has been implemented through PulseNet Latin America and The Caribbean. From 1988 to 2010, a total of 780 XbaI-PFGE patterns corresponding to 1,954 STEC O157 strains isolated from human infections (n=1,693), food (n=185), animals (n=191), and environmental samples (n=35) were included in the Argentine Database for O157.

In the present work, E. coli O157 strains showed a high degree of diversity. A total of 148 different XbaI-PFGE patterns were established among the 280 strains studied, with the strains that yielded identical patterns grouped in clusters. The five common XbaI-PFGE patterns identified in both human and bovine strains in the present study had been previously included in the National Database. These patterns corresponded to strains isolated from HUS (n=120), BD (n=94), NBD (n=52), cooked and uncooked ground meat (n=18), sausages (n=2), and water streams (n=1), indicating that they are widespread in human population, and are also detected in food and in the environment. Two of these common patterns, AREXH01.0011 and AREXH01.0022, are predominant in the Argentine Database of E. coli O157, representing 23.1% and 13.7%, respectively.

The AREXH01.0011 pattern is identical to SMI-H and EXH01.0047 patterns, described as predominant in Sweden and the United States, respectively. The Swedish type has also the other characteristics of the Argentine type, i.e., PT4 and stx ₂/stx _2c(vh-a) genes (Löfdahl, 2008). It would be of great value to perform the surveillance of this type more carefully in the rest of the world.

A total of 28 of 226 (12.4%) human strains showed identical PT-XbaI-PFGE-stx profile combinations as those found in 11 of 54 (20.4%) cattle strains. Some STEC strains of both origins could be discriminated using BlnI as second enzyme. Although they must be considered as different, they presented a certain degree of similarity that should be evaluated in the epidemiologic context. This overlap of human and bovine strains indicates that cattle are important reservoir for humans. It was possible to link 12% of reported human infections to the bovine reservoir by analyzing a very small proportion from the total cattle at slaughter in Argentina, estimated as approximately 21 million heads in the period under study (IPCVA, 2011).

The STEC prevalence in cattle and the high consumption of beef in Argentina are a risk for human population. Recently, a more extensive investigation on STEC O157 in nine selected beef exporting abattoirs of Argentina showed a prevalence of 4.1% (CI 95%, 5.6–2.9%) in the fecal content and 2.6% (CI 95%, 3.9–1.6%) in the carcasses of 811 bovines at slaughter (Masana et al., 2010). From these data, and slaughter statistics, it could be roughly estimated that a proportion of around 38,000 bovines were contaminated with STEC O157 in their feces at slaughter per each reported human case. This should be taken into account to determine control and prevention guidelines aimed at reducing the burden of these diseases in our country.

Conclusion

This study showed that the use of phage typing, PFGE, and stx-genotyping techniques as epidemiological tools that may help to detect reservoirs, and establish temporal and geographical variations of newly emerging clones. However, other typing methods must be employed to relate the subtypes to the severity of the associated-illnesses.

Footnotes

Acknowledgments

We are grateful to Ariela Baschkier and Elisabet Vilacoba, for their technical assistance. We also thank all participating laboratories for their contribution to the National Health Surveillance. This project was partially supported by a grant from the Research Program of the Instituto de Promoción de la Carne Vacuna Argentina (Argentine Beef Promotion Institute), and public funds of the Research Project AETA3692 of the Instituto Nacional de Tecnología Agropecuaria de Argentina (National Institute for Agricultural Technology, INTA).

Disclosure Statement

No competing financial interests exist.

References

Ahmed

, Bopp

, Borczyk

. Phage-typing scheme for Escherichia coli serotype O157:H7. J Infect Dis, 1987; 155:806–809.

Aspán

, Erikson

. Verotoxigenic Escherichia coli O157:H7 from Swedish cattle; isolates from prevalence studies versus strains linked to human infection—A retrospective study. BCM Vet Res, 2010; 6:7–16.

Baker

, Moxley

, Steele

, LeJeune

, Christopher-Hennings

, Chen

, Hardwidge

, Francis

. Differences in virulence among Escherichia coli O157:H7 strains isolated from humans during disease outbreaks and from healthy cattle. Appl Environ Microbiol, 2007; 73:7338–7346.

Beutin

, Montenegro

, Ørskov

, Prada

, Zimmermann

, Stephan

. Close association of verotoxin (Shiga-like toxin) production with enterohemolysin production in strains of Escherichia coli. J Clin Microbiol, 1989; 27:2559–2564.

Beutin

, Kaulfuss

, Cheasty

, Brandenburg

, Zimmermann

, Gleier

, Willshaw

, Smith

. Characteristics and association with disease of two major subclones of Shiga toxin (verocytotoxin)–producing strains of Escherichia coli (STEC) O157 that are present among isolates from patients in Germany. Diagn Microbiol Infect Dis, 2002; 44:337–346.

Böerlin

, McEwen

, Böerlin-Petzold

, Wilson

, Johnson

, Gyles

. Association between virulence factors of Shiga toxin–producing Escherichia coli and disease in humans. J Clin Microbiol, 1999; 37:497–503.

Caprioli

, Morabito

, Brugère

, Oswald

. Enterohaemorrhagic Escherichia coli: Emerging issues on virulence and modes of transmission. Vet Res, 2005; 36:289–311.

Chinen

, Tanaro

, Miliwebsky

, Lound

, Chillemi

, Ledri

, Baschkier

, Scarpin

, Manfredi

, Rivas

. Isolation and characterisation of Escherichia coli O157:H7 from retails meats in Argentina. J Food Prot, 2001; 64:1346–1351.

Chinen

, Epszteyn

, Melamed

, Aguerre

, Martinez Espinosa

, Motter

, Baschkier

, Manfredi

, Miliwebsky

, Rivas

. Shiga toxin–producing Escherichia coli O157 in beef and chicken burgers, and chicken carcasses in Buenos Aires, Argentina. Int J Food Microbiol, 2009; 132:167–171.

10.

[CLSI] Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing: Twenty-first Informational Supplement. 31,1 Approved Standard M100-S21. Wayne, PA: CLSI, 2011.

11.

Eklund

, Leino

, Siitonen

. Clinical Escherichia coli strains carrying stx genes: stx variants and stx-positive virulence profiles. J Clin Microbiol, 2002; 40:4585–4593.

12.

Ewing

. Edwards and Ewing's Identification of Enterobacteriaceae, 4 ^th. New York: Elsevier, 1986; 92–134.

13.

Friedrich

, Bielaszewska

, Zhang

, Pulz

, Kuczius

, Ammon

, Karch

. Escherichia coli harboring Shiga toxin 2 gene variants: Frequency and association with clinical symptoms. J Infect Dis, 2002; 185:74–84.

14.

Gannon

, D'Souza

, Graham

, King

, Rahn

, Read

. Use of the flagellar H7 gene as a target in multiplex PCR assays and improved specificity in identification of enterohemorrhagic Escherichia coli strains. J Clin Microbiol, 1997; 35:656–662.

15.

Griffin

, Tauxe

. The epidemiology of infections caused by Escherichia coli O157:H7, other enterohemorrhagic E. coli, and the associated hemolytic uremic syndrome. Epidemiol Rev, 1991; 13:60–98.

16.

Gyles

. Shiga toxin–producing Escherichia coli: An overview. J Anim Sci, 2007; 85:E45–E62.

17.

Hyytia-Trees

, Smole

, Fields

, Swaminathan

, Ribot

. Second generation subtyping: A proposed PulseNet protocol for multiple-locus variable-number tandem repeat analysis of Shiga toxin–producing Escherichia coli O157 (STEC O157) Foodborne Pathog Dis, 2006; 3:118–131.

18.

[IPCVA] Instituto de Promoción de la Carne Vacuna Argentina. 2011. www.ipcva.com.ar/estadisticas/vista_faena.php. 2012 February 1.

19.

Jelacic

, Damrow

, Chen

, Jelacic

, Bielaszewska

, Ciol

, Carvalho

, Melton-Celsa

, O'Brien

, Tarr

. Shiga toxin–producing Escherichia coli in Montana: Bacterial genotypes and clinical profiles. J Infect Dis, 2003; 188:719–729.

20.

Karch

, Bohm

, Schmidt

, Gunzer

, Aleksic

, Heesemann

. Clonal structure and pathogenicity of Shiga-like toxin-producing, sorbitol-fermenting Escherichia coli O157:H7. J Clin Microbiol, 1993; 31:1200–1205.

21.

Karmali

, Petric

, Lim

, Cheung

, Arbus

. Sensitive method for detecting low numbers of verotoxin-producing Escherichia coli in mixed cultures by use of colony sweeps and polymyxin extraction of verotoxin. J Clin Microbiol, 1985; 22:614–619.

22.

Karmali

. Infection by Shiga toxin–producing Escherichia coli: An overview. Mol Biotechnol, 2004; 26:117–122.

23.

Kawano

, Ono

, Iwashita

, Kurogi

, Haga

, Maeda

, Goto

. stx genotype and molecular epidemiological analyses of Shiga toxin–producing Escherichia coli O157:H7/H- in human and cattle isolates. Eur J Clin Microbiol Infect Dis, 2012; 31:119–127.

24.

Khakhria

, Duch

, Lior

. Extended phage-typing scheme for Escherichia coli O157:H7. Epidemiol Infect, 1990; 105:511–520.

25.

Lee

, French

, Hara-Kudo

, Iyoda

, Kobayashi

, Sugita-Konishi

, Tsubone

, Kumagai

. Multivariate analyses revealed distinctive features differentiating human and cattle isolates of Shiga toxin–producing Escherichia coli O157 in Japan. J Clin Microbiol, 2011; 49:1495–1500.

26.

Leotta

, Chinen

, Epszteyn

, Miliwebsky

, Melamed

, Motter

, Ferrer

, Marey

, Rivas

. Validación de una técnica de PCR múltiple para la detección de Escherichia coli productor de toxina Shiga. Rev Argent Microbiol, 2005; 37:1–10. (In Spanish.)

27.

Löfdahl

. How global is VTEC? Epidemiology and Transmission of VTEC and Other Pathogenic Escherichia coli. Stockholm: Sweden, 2008; 65–67.

28.

Manning

, Motiwala

, Springman

, Qi

, Lacher

, Ouellette

, Mladonicky

, Somsel

, Rudrik

, Dietrich

, Zhang

, Swaminathan

, Alland

, Whittam

. Variation in virulence among clades of Escherichia coli O157:H7 associated with disease outbreaks. Proc Natl Acad Sci USA, 2008; 105:4868–4873.

29.

Masana

, Leotta

, Del Castillo

, D'Astek

, Palladino

, Galli

, Vilacoba

, Carbonari

, Rodríguez

, Rivas

. Prevalence, characterization, and genotypic analysis of Escherichia coli O157:H7/NM from selected beef exporting abattoirs of Argentina. J Food Prot, 2010; 73:649–656.

30.

Mora

, Blanco

, Alonso

, Dhabi

, Thomson-Carter

, Usera

, Bartolomé

, Prats

, Blanco

. Phage types and genotypes of Shiga toxin–producing Escherichia coli O157:H7 isolates from humans and animals in Spain: Identification and characterization of two predominating phage types (PT2 and PT8) J Clin Microbiol, 2004; 42:4007–4015.

31.

Nakamura

, Ogasawara

, Kita

, Hase

, Nishikawa

. Typing of Stx2 genes of Escherichia coli O157 isolates from cattle. Jpn J Infect Dis, 2008; 61:251–252.

32.

Ørskov

, Ørskov

. Serotyping of Escherichia coli. Methods in Microbiology. Bergan

. London: Academic Press, 1984; 43–112.

33.

Pearce

, Chase-Topping

, McKendrick

, Mellor

, Locking

, Allison

, Ternent

, Matthews

, Knight

, Smith

, Synge

, Reilly

, Low

, Reid

SWJ

, Gunn

, Woolhouse

MEJ

. Temporal and spatial patterns of bovine Escherichia coli O157 prevalence and comparison of temporal changes in the patterns of phage types associated with bovine shedding and human E. coli O157 cases in Scotland between 1998–2000 and 2002–2004. BMC Microbiol, 2009; 9:276–288.

34.

Pennington

. Escherichia coli O157. Lancet, 2010; 376:1428–1435.

35.

Persson

, Olsen

, Ethelberg

, Scheutz

. Subtyping methods for Escherichia coli Shiga toxin (verocytotoxin) 2 variants and correlation to clinical manifestations. J Clin Microbiol, 2007; 45:2020–2024.

36.

Rangel

, Sparling

, Crowe

, Griffin

, Swerdlow

. Epidemiology of Escherichia coli O157:H7 outbreaks, United States, 1982–2002. Emerg Infect Dis, 2005; 11:603–609.

37.

Ribot

, Fair

, Gautom

, Cameron

, Hunter

, Swaminathan

, Barrett

. Standardization of pulsed-field gel electrophoresis protocols for the subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathog Dis, 2006; 3:59–67.

38.

Rivas

, Miliwebsky

, Chinen

, Roldán

, Balbi

, García

, Fiorilli

, Sosa-Estani

, Kincaid

, Rangel

, Griffin

. Case-Control Study Group. Characterization and epidemiologic subtyping of Shiga toxin–producing Escherichia coli strains isolated from hemolytic uremic syndrome and diarrhea cases in Argentina. Foodborne Pathog Dis, 2006; 3:88–96.

39.

Rivas

, Chinen

, Miliwebsky

, Galli

, Repetto

, Masana

. Epidemiology of Argentinean STEC. Population Genetics of BacteriaA Tribute to Thomas

, Whittam , Walk

, Feng

PCH

. Washington, DC: ASM Press, 2011; 109–132.

40.

Rolgaard

, Scheutz

, Boel

, Aabo

, Schultz

, Cheasty

, Møller Nielsen

, Olsen

KEP

, Christensen

. VTEC O157 subtypes associated with the most severe clinical symptoms in humans constitute a minor part of VTEC O157 from Danish cattle. Int J Med Microbiol, 2004; 294:255–259.

41.

Schmidt

, Beutin

, Karch

. Molecular analysis of the plasmid-encoded hemolysin of Escherichia coli O157:H7 strain EDL 933. Infect Immun, 1995; 63:1055–1061.

42.

Tanaro

, Leotta

, Lound

, Galli

, Piaggio

, Carbonari

, Araujo

, Rivas

. Escherichia coli O157 in bovine feces and surface water streams in a beef cattle farm of Argentina. Foodborne Pathog Dis, 2010; 7:475–477.

43.

Thomson-Carter

. General recovery, characterisation and typing protocols for VTEC. Verocytotoxigenic E. coli. Duffy

, Garvey

, McDowell

. Trumbull, CT: Food & Nutrition Press, Inc., 2001.

44.

Tyler

, Johnson

, Lior

, Wang

, Rozee

. Identification of verotoxin type 2 variant B subunit genes in Escherichia coli by the polymerase chain reaction and restriction fragment length polymorphism analysis. J Clin Microbiol, 1991; 29:1339–1343.

45.

Yang

, Kovar

, Kim

, Nietfeldt

, Smith

, Moxley

, Olson

, Fey

, Benson

. Identification of common subpopulations of nonsorbitol-fermenting, beta-glucuronidase-negative Escherichia coli O157:H7 from bovine production environments and human clinical samples. Appl Environ Microbiol, 2004; 70:6846–6854.

46.

Zhang

, Bielaszewska

, Thorsten

, Karch

. Identification, characterization, and distribution of a Shiga toxin 1 gene variant (stx1c) in Escherichia coli strains isolated from humans. J Clin Microbiol, 2002; 40:1441–1446.