You are on page 1of 10

Res. Microbiol.

150 (1999) 323332

Elsevier, Paris

Heterogeneity of the eae genes in attaching/effacing Escherichia coli

from cattle: comparison with human strains
Bernard China, Etienne Jacquemin, Anne-Catherine Devrin,
Vinciane Pirson, Jacques Mainil*
Chaire de bactriologie et de pathologie bactrienne, Facult de mdecine vtrinaire, Universit de Lige,
Sart Tilman, Lige, B-4000, Belgium
(Submitted 12 December 1998; accepted 10 March 1999)

Abstract Enteropathogenic (EPEC) and enterohaemorrhagic (EHEC) Escherichia coli isolated from cattle
were studied by DNA colony hybridization to subtype their intimin-encoding (eae) gene with probes derived
from the variable parts of the eae gene of the human EPEC strain E2348/69, the eae gene of the human
O157:H7 EHEC strain ATCC43888, and the eae gene of the bovine O26:H- EHEC strain 193, whose eae
gene was first cloned and sequenced during this work. The EPEC and EHEC had been isolated from
diarrhoeic calves (143 EPEC and 48 EHEC) and from healthy animals at the slaughterhouse (10 EPEC and 34
EHEC). The 191 bovine EPEC and EHEC isolated from diseased calves were positive with the Eae probe (55
and 27% respectively) and with the Eae probe (9 and 73% respectively), whereas 52 EPEC (36%) were
negative with the Eae, Eae, and Eae probes. The results were different for the 44 bovine EPEC and EHEC
isolated from healthy cattle at slaughterhouses: most tested positive with the Eae probe (80 and 82%
respectively) and the remaining (20 and 18% respectively) with the Eae probe. Nine O26 human EHEC tested
positive with the Eae probe and seven O111 with the Eae probe. The bovine and human EPEC and EHEC
belonging to these two serogroups gave identical results: the 18 bovine and human O26 isolates tested
positive with the Eae probe, whereas the 13 O111 isolates were positive with the Eae probe. In contrast, the
isolates belonging to other serogroups (O5, O15, O18, O20, and O118) gave more variable results. The eae and
eae, but not the eae, variants were thus distributed amongst bovine EPEC and EHEC. The eae variant
seemed to be more frequently associated with the presence of clinical signs in calves, but one third of EPEC
from diarrhoeic calves carried an eae gene variant other than the , , or variants. In addition, the use of these
gene probes did not enable differentiation between bovine and human EHEC belonging to the same O
serogroup. Elsevier, Paris
eae gene / Escherichia coli / locus of enterocyte effacement / cloning / gene probe

1. Introduction
Enteropathogenic (EPEC) and enterohaemorrhagic (EHEC) Escherichia coli cause enteritis
* Correspondence and reprints
Tel.: 32 4 366 4090
Fax: 32 4 366 4056
Abbreviations: AE, attaching/effacing; AEEC, attaching/
effacing E. coli; BEHEC, bovine EHEC; BEPEC, bovine
EPEC; DEPEC, dog EPEC; EHEC, enterohaemorrhagic
E. coli; EPEC, enteropathogenic E. coli; HEHEC, human
EHEC; HEPEC, human EPEC; IPTG, isopropyl-betathiogalactoside; LEE, locus of enterocyte effacement; PEPEC, pig EPEC; REPEC, rabbit EPEC.

and haemorrhagic enterocolitis in man and in

various animal species [7, 8, 14, 24, 25, 27, 31].
They are often regrouped under the name of
attaching/effacing E. coli (AEEC) on the basis of
production of a common histopathological lesion, characterized by the effacement of the
microvilli of the enterocytes, by intimate attachment to the enterocyte membrane, and by accumulation of actin filaments under the area of
adherence of the bacteria [26]. In addition,
EHEC, but not EPEC, produce Shiga toxins and
harbour a ca. 60-MDa plasmid [27]. Production
of the attaching/effacing (AE) lesion is con-


China et al.

trolled by several genes grouped together on

the bacterial chromosome to form a pathogenicity island, the locus of enterocyte effacement, or
LEE [20]. These genes code for a type III secretion system, for several type III-secreted proteins, and for an outer membrane protein, intimin, which mediates the intimate adherence of
the bacteria to the enterocyte cytoplasmic membrane [27].
The eae gene which codes for intimin was first
cloned from the human EPEC (HEPEC) strain
E2348/69 of serotype O127:K-:H6 [17]. Other eae
genes have been cloned from O157:H7 human
EHEC (HEHEC) strains [6, 40], from rabbit
EPEC (REPEC) strains, including the O15:Hstrain RDEC-1 [3], from the dog EPEC (DEPEC)
strain 4221 [4], and from the O45 porcine EPEC
(PEPEC) strain 861390 [4]. These eae genes are
highly homologous in their proximal two
thirds, but much less so in their terminal part.
A recent study [2] has even identified five
variants of the eae gene in HEPEC and HEHEC,
which are to some extent serotype-associated:
in EPEC strains of O55:H6, O127:H6 (including
the reference strain E2348/69), O142:H6, and
O142:H34 serotypes; in EPEC strains of
O26:H-, O111:H-, O114:H2, O119:H2, O119:H6,
and O128:H2 serotypes and in HEHEC strains
of O26:H11 and O111:H8 serotypes; in EPEC of
O55:H- and O55:H7 serotypes and in EHEC
strains of O157:H7 serotype; in EPEC strains
of O86:H34 serotype, and a still untyped one in
EPEC strains of O127:H40 serotype. According
to their sequence the eae genes of the REPEC
and PEPEC strains belong to group , whereas
the eae gene of the DEPEC strain belongs to
group [24]. Bovine EPEC (BEPEC) and EHEC
(BEHEC) strains also harbour eae-related
genes [5, 9, 11, 12, 16, 23, 34, 37, 39], but none
has been cloned so far, nor of course has their
heterogeneity been studied.
The purpose of this study was as follows: i) to
clone and sequence the eae gene from an O26
BEHEC strain isolated from a calf with diarrhoea; ii) to derive gene probes from the variable parts of different eae genes; and iii) to type
and compare the eae genes of a collection of

EPEC and EHEC strains isolated from diarrhoeic calves, healthy adult cattle, and humans.

2. Materials and methods

2.1. Bacteria

The eae gene was cloned from the O26:K-:HBEHEC strain 193 originally isolated in 1962
from a diarrhoeic calf in the USA [22]. Collections of E. coli strains were subsequently studied by colony hybridization to determine the
variants of the eae gene present: 143 BEPEC and
48 BEHEC strains isolated from 2- to 8-week-old
calves with clinical diarrhoea [11, 23]; 10 BEPEC
and 34 BEHEC strains isolated from healthy
cattle at slaughterhouses [11]; and 16 HEHEC
strains received from Dr D. Pirard (AZ-VUB,
Brussels, Belgium). The definition of EPEC
must, however, be taken with some caution, as
EHEC can lose their stx genes in vitro or in
vivo [18, 19, 23]. Twenty-one BEPEC and BEHEC strains from diarrhoeic calves and 18 from
healthy cattle had been serotyped previously:
they belonged to the O5, O15, O18, O20, O26,
O111, and O118 serogroups [23] (unpublished
data). The HEHEC strains belonged to O26
(nine isolates) or O111 (seven isolates) serogoups (Pirard, personal communication).
Three strains were used as positive controls
for the gene probes: the HEPEC strain
E2348/69, carrying an eae gene, the REPEC
strain RDEC1, carrying an eae gene, and the
O157:H7 HEHEC strain ATCC43888 strain, carrying an eae gene [2]. The BEHEC strain 193
carrying a still uncharacterized eae gene was the
fourth control strain. Strain HS was the negative
control for all probes [22, 23, 28].
2.2. Cloning and sequencing of the eae gene

Total DNA of strain 193 was extracted

(Qiagen genomic-tip, Qiagen, Hilden, Germany) and submitted to partial restriction by
the Sau3A enzyme. The restriction fragments
from 5 to 15 kb were eluted (Geneclean, Bio101,
Vista, USA) and ligated into the BamHI site of
the pUC19 plasmid. After electroporation of the

eae Genes in bovine attaching/effacing E. coli

E. coli DH5 strain, the recombinant clones

were selected on LB agar supplemented
with 5-bromo-4-chloro-3-indolyl-beta-D-galacto-sidase (X-gal), isopropyl-beta-thio-galactoside (IPTG) and ampicillin (100 g/mL). White
colonies were subsequently assayed by DNA
hybridization with the gene probe derived from
the constant part of the eae gene [17]. Subclones
were derived from a probe-positive recombinant clone to obtain a restriction map and the
sequence of the bovine eae gene using the
Sanger method [35] and a T7 sequencing kit
(Pharmacia, Uppsala, Sweden).
2.3. Gene probes

The probe corresponding to the constant part

of the eae gene (Eae probe) was derived by
restriction as described before [23] from the
pCVD434 recombinant plasmid carrying the
eae gene from the HEPEC strain E2348/69 [17].
The probe for the variable part of the eae gene of
the O26:K-:H- BEHEC strain 193 (later identified as an Eae probe) was derived as described
in the Results section. These two probes were
recovered after digestion by appropriate restriction endonucleases, agarose gel electrophoresis,
and elution (Geneclean, Bio1O1, Vista, USA).
The probes for the variable parts of the eae
gene (Eae probe) and of the eae gene (Eae
probe) were derived by polymerase chain reaction (PCR) (figure 1) from the O127:K-:H6
HEPEC strain E2348/69 and from the O157:H7
HEHEC strain ATCC43888, respectively, with
the following primers: B75 (TAAAGTGATGAAGGGGGAT) and B76 (ATTGGGTGGAGAGAAAAGC) for the Eae probe and B73
probe (Gibco BRL Life Technologies, Merelbeke,
The PCR mixture included one unit of Dynazyme (Finnzymes, Espoo, Finland), five L of
2mM dNTPs (Pharmacia, Uppsala, Sweden),
five L of 10 buffer (Tris-HCl 100mM pH8.8,
MgCl2 15mM, KCl 500 mM, Triton X-100 1%),
0.5 L of each primer (40mM), and 5 L of DNA
template in a total volume of 50 L, and was
covered by 30 L of mineral oil (Sigma-Aldrich,


Bornem, Belgium). The reaction conditions were

the following: 94 C for 5 min followed by 30
cycles at 94 C for 30 s, 50 C for 30 s, and 72 C
for 30 s, in a DNA thermocycler (Perkin Elmer
Cetus, Norwalk, UK).
The DNA fragment amplified with primers
B73 and B74 represented the Eae probe and
had a size of ca. 0.8 kb. The DNA fragment
amplified with primers B75 and B76 had a size
of ca. 0.95 kb and was subsequently digested
with the HinfI restriction enzyme (figure 1). The
two generated fragments were separated by
electrophoresis in agarose gel. The largest restriction fragment (ca. 0.7 kb), corresponding to
the Eae probe, was eluted from the gel
(Geneclean, Bio1O1, Vista, USA).
2.4. Colony hybridization

The gene probe fragments were radioactively

labelled with a-32P-dCTP (Amersham, Litthe
Challont, UK) by random priming with dCTP
labelling beads (Pharmacia, Uppsala, Sweden).
The colony hybridization assays during the
cloning of the eae gene of the BEHEC strain and
for the study of the subtype(s) of eae genes
among the collections of EPEC and EHEC
strains were performed as previously described [23].

3. Results
3.1. Cloning of the eae gene
from O26:K-:H-BEHEC strain 193

Two of the recombinant clones obtained

tested positive with the Eae probe (pCUR3 and
pCUR4). After sequencing of the extremities of
the cloned DNA fragments and comparison
with the published sequences of eae genes, one
of them was considered as carrying the whole
eae gene (pCUR3), whereas the second one
carried only the 3 end of the gene (pCUR4)
(figure 1). The pCUR3 insert was a 10-kb DNA
fragment from which subclones were derived
using restriction endonucleases according to
already published restriction maps of eae genes.
The restriction map of the pCUR3 insert had


China et al.

Figure 1. Restriction map and subclones of the pCUR3 recombinant plasmid DNA insert carrying the eae gene of the O26 BEHEC
strain 193 (1 = start codon <ATG>; 939 = stop codon <TAA>) and derivation of the Eae probe. Restriction sites above the line of
pCUR3 were conserved compared to the restriction maps of the eae gene of the HEPEC strain E2348/69 and of the eae gene of
O157:H7 HEHEC strains; restriction sites below the line were not. The pCUR25H fragment was used as an Eae probe; the Eae
(amplification with primers B75 and B76 and restriction by HinfI) and Eae (amplification with primers B73 and B74) probe fragments
are also presented.

similarities with the restriction maps of the

other eae genes for approximately 2 kb (figure 1).
3.2. Sequencing of the eae gene
from O26:K-:H-BEHEC strain 193

From pCUR5, pCUR6, pCUR22, pCUR25,

and pCUR27 (figure 1), overlapping subclones
were derived using AluI, HincII, PstI, and RsaI
restriction endonucleases to generate short
DNA fragments which were subsequently sequenced. An open reading frame of 2 814 bp
was observed corresponding to a protein of 938
amino acids (Genebank Access Number
AF043226). Comparison with the other eae gene
sequences published revealed 99.9% overall

identity with the eae gene, but only 83% with

the eae and eae genes (table I). In the proximal
two-thirds (1.8 kb), the BEHEC eae gene had 94,
100, and 93% identity with the eae, eae, and
eae genes. The degree of identity dropped to
6277% in the distal third (1.0 kb) with the eae,
eae, and eae genes, but not with the eae gene
(table I, figure 2).
3.3. Derivation of a gene probe
from the eae gene of O26:K-:H-BEHEC strain 193

The probe for the variable part of the eae gene

of strain 193 was thus named Eae. This probe
was a ca. 0.4-kb long DNA fragment derived
from the pCUR25 recombinant plasmid after

eae Genes in bovine attaching/effacing E. coli


Table I. Identity between the eae gene of the O26 BEHEC strain 193 and the eae [17], eae [3], eae [6, 40], and eae [4] gene
% Identity with the respective eae variant
eae Gene

E2348/69 (HEHEC)
4221 (DEPEC)

(2.8 kb)

(1.8 kb)

(1.0 kb)

83 %
99.9 %
83 %

94 %
100 %
93 %

62 %
98 %
65 %
77 %

NA, not accessible.

RsaI restriction and cloned into the SmaI site of

pUC19 plasmid (pCUR25H) (figure 1). The
probe fragment was obtained by restriction of
pCUR25H with BamHI and EcoRI.
3.4. Study of the heterogeneity of the eae genes
among BEPEC, BEHEC, and HEHEC strains

All strains were first tested with the Eae

probe corresponding to the constant part of the
eae genes and subsequently with the three
probes derived from the variable parts of the eae
genes coding for the intimins , , and . The
Eaea probe hybridized with all 235 EPEC and
EHEC isolates from cattle and 16 from humans.
The Eaea probe tested positive with none of the
isolates except for the positive control, but the
Eae and the Eae probes tested positive with
most bovine and all human isolates (table II) in
addition to the positive controls. No crosshybridization results were observed.

The 191 BEPEC and BEHEC isolated from

diseased calves were mainly positive with the
Eae probe (78 EPEC, 55%; 13 EHEC, 27%) and
half of the remaining with the Eae (13 EPEC,
9%; 35 EHEC, 73%). This means that 52 EPEC
(36%) were negative with the Eae, Eae, and
Eae probes. The results were different for the
44 BEPEC and BEHEC isolated from healthy
cattle at slaughterhouses: most tested positive
with the Eae probe (80 and 82%, respectively)
and the remaining (20 and 18%, respectively)
with the Eae probe. The 16 HEHEC were also
divided into two groups: nine were positive
with the Eae probe and seven with the Eae
probe (table II).
3.5. Association between serogroups and eae gene
subtype among BEPEC, BEHEC, and HEHEC

The strains belonging to serogroups common

to humans and cattle gave identical results

Figure 2. Phylogenetic tree representation of the sequences of the distal third (1.0 kb) of the different eae gene variants. Gene bank
access numbers: RDEC1 = U60002; 193 = AF043226; 4221 = U66102; E2348/69 = AF022236; O157:H7 = AF071034.


China et al.

Table II. Colony hybridization results of bovine and human EPEC and EHEC with gene probes derived from the variable 3 ends of the
eae genes coding for intimins (Eae probe), (Eae probe), and (Eae probe).
Type of isolate
(no. isolates)

No. isolates negative

with the three probes

No. isolates positive with the probe (%)







Diarrhoeic calves (191)

EPEC (143)
EHEC (48)

78 (55)
13 (27)

13 (9)
35 (73)

52 (36)

Healthy cattle (44)

EPEC (10)
EHEC (34)

2 (20)
6 (18)

8 (80)
28 (82)

Humans (16)

EHEC (16)

9 (55)

7 (45)

(table III): the O26 bovine (nine isolates) and

human (nine isolates) EPEC and EHEC tested
positive with the Eae probe, whereas the O111
bovine (six isolates) and human (seven isolates)
EPEC and EHEC were positive with the Eae
probe. On the other hand, the isolates from
other serogroups, regarded as cattle-specific (O5
and O118) or less frequently identified (O15,
O18, O20), gave more variable results. These
BEPEC and BEHEC were often positive with the
Eae probe when isolated from diarrhoeic
calves and with the Eae probe when isolated
from healthy cattle (table III).

4. Discussion
AEEC, either enterohaemorrhagic (EHEC) or
enteropathogenic (EPEC), cause enteric diseases

in humans and in animals. The zoonotic risk of

EPEC has never really been estimated. However, their host specificity is generally believed
to be high, since most EPEC from humans and
from the various animal species belong to specific serotypes (groups) and/or produce specific
colonization factors [7, 8, 14, 23, 25, 27, 31].
The situation is different for EHEC, which are
mainly isolated from humans and ruminants;
human infections have indeed been traced to
contaminated beef and dairy products in several cases [21, 27]. Based on serotyping, three
categories of EHEC strains could be previously
distinguished: those mainly associated with disease in calves (e.g., serogroups O5, O118), those
associated with disease in calves and humans
(e.g., serogroups O26, O103, O111), and those

Table III. Correlation between serogroup O of bovine and human AEEC strains and probe hybridization results (strains are EHEC
unless otherwise stated).
Serogroup O

Eae probe

No. positive isolates/no. isolates tested

Diarrhoeic calves

Healthy cattle














Both are EPEC. bFive are EPEC and EHEC is strain 193.

eae Genes in bovine attaching/effacing E. coli

carried by healthy cattle but associated with

disease in humans (serogroup O157) [21, 27].
But the first two categories are becoming less
and less clearly defined as the range of serogroups and serotypes isolated from both cattle
and humans widens continuously.
Other criteria than the serotype must be used
to compare AEEC from identical serotypes isolated from humans and cattle: the genetic determinants of the AE lesion and of the cytotoxins
such as Shiga toxins and enterohaemolysins, or
the whole bacterial genome [21, 27, 36]. Comparison of the whole bacterial genome is timeconsuming and needs sophisticated equipment.
Comparison of the Shiga toxins and of the
enterohaemolysins is easier, but their encoding
genes are phage- and plasmid-located, respectively, and can be lost in vitro and in vivo [18,
19, 23]. Comparison of the genes located within
the LEE is another possibility, as several variants of the eae gene coding for the intimin and of
different type III-secreted proteins have been
recently recognized [1, 2, 13, 29, 32].
The eae gene and its product, intimin, play a
central role in the pathogenesis of AEEC and in
the production of the AE lesion by mediating
the intimate attachment of the bacteria to the
enterocyte cytoplasmic membrane and, at least
in part, the host and tissue specificity of the
infection, and in the diagnosis of AEEC which is
based upon the detection of the eae gene in the
bacteria either by gene probe or by PCR [10, 15,
17, 27]. Since the recognition of several variants [24, 6, 17, 40], the eae gene could also serve
as a molecular epidemiological marker in the
public health problem of tracing human infections and of comparing AEEC from different
This study thus sought to answer two questions: i) are the different variants of the eae gene
present in bovine AEEC strains? and ii) do
bovine and human AEEC belonging to identical
serotypes carry the same variant(s) of the eae
The cloning and sequencing of the eae gene
from one BEHEC indicates that strain 193 carries a variant, like the REPEC. The existence of
the variant amongst bovine AEEC was con-


firmed by the colony hybridization assay using

a gene probe derived from the variable part of
this gene (pCUR25H or Eae probe). Conversely, a collection of REPEC [33] tested positive with this probe (unpublished data). Keeping in mind the always possible loss of the stx
genes by some of the strains here defined as
EPEC, the eae variant is actually more often
present in EPEC from diarrhoeic calves (55%)
than in EHEC from diarrhoeic calves (27%) and
EPEC and EHEC from healthy cattle (20 and
18% respectively). Given the results with the
and probes on EPEC from diarrhoeic calves (0
and 9% of positive isolates respectively), one
third of them possess still another variant, possibly the eae variant, which was not included in
this work, or the recently described eaee variant
(Oswald, personal communication). Hence,
with the exception of EPEC from diarrhoeic
calves, the eae variant is the most frequently
detected (73 to 82%) and all isolates are positive
for either the eae variant or the eae variant
(table II).
Although the number of strains with serogrouping results is small, bovine and human
AEEC belonging to the same serogroup give
identical results irrespective of their origin: the
O26 isolates were positive for the variant and
the O111 isolates for the variant. No O157
strains, which are also positive for the variant
according to published results [2], were tested,
because O157 EHEC are rarely isolated from
humans or from cattle or beef products in
Belgium (Daube and Pirard, personal communications). Apart from the O26 and O111 isolates, specific associations between serogroups
and intimin types do not seem to exist amongst
bovine AEEC (table III), in contrast to human
AEEC [2]. Complete serotyping may, however,
reveal such specific associations.
If these results are confirmed on a larger
number of bovine AEEC, they could mean
either that the and the untyped intimins are
more cattle-specific than the intimin as far as
disease production is concerned, in contrast to
humans, or that the three intimin types are
involved in disease in calves, with possible
differing tissue specificity (small intestine or


China et al.

colon for instance), as already observed for the

and intimins in humans and in animal
models [38]. It is also possible, however, that
neither hypothesis is correct, since intimin
specificity may reside in a few amino acid
sequences and because only DNA sequencing,
and not the typing methods used so far (probes,
PCR), could discriminate between bovine and
human and intimins. However, it is possible
that O26 and O111 BEHEC and HEHEC are
clonally related and that no method will distinguish between them. Indeed, evidence for
clonality was recently observed between O26,
O103, O111, and O157 HEHEC within the same
serogroup [36].
In summary, the answer to the first question
is that the eae and eae variants, but not the
eae variant, are distributed amongst bovine
AEEC. In addition, one third of the EPEC from
diarrhoeic calves carry another variant of the eae
gene. As to the second question, O26 and O111
AEEC harbour, respectively, the same eae gene
variants irrespective of their origin (cattle or
humans). Using gene probes derived from the
variable part of the genes, the typing of the eae
gene variants was thus unable to differentiate
between BEHEC and HEHEC within the same
serogroup. Other techniques should be used
and/or parameters should be looked at: for
example, other genes of the LEE or other targets
such as the cytotoxin-encoding genes.
Moreover, in order to be practical, a typing
method must not only be reproducible but also
easy and fast, and multiplex PCRs should be
developed for genes located within the LEE, just
as some have already been developed for shigatoxigenic E. coli [10, 15, 30].

eae dune souche EHEC bovine O26:H- (193), dont

le gne eae a t clon et squenc au cours de ce
travail. Ces souches EPEC et EHEC avaient t
isoles de veaux malades (143 EPEC et 48 EHEC) ou
danimaux sains labattoir (10 EPEC et 34 EHEC).
Les 191 souches EPEC et EHEC bovines isoles de
veaux malades sont positives avec la sonde Eae (55
et 27 % respectivement) et avec la sonde Eae (9 et
73 % respectivement), mais 52 souches EPEC (36 %)
donnent des rsultats ngatifs avec les sondes Eae,
Eae et Eae. Les 44 souches EPEC et EHEC isoles
danimaux sains labattoir donnent des rsultats
diffrents : la majorit (80 et 82 % respectivement)
sont positives avec la sonde Eae et les autres (20 et
18 % respectivement) avec la sonde Eae. Neuf souches EHEC humaines O26 sont positives avec la
sonde Eae et sept O111 avec la souche Eae. Les
souches EPEC et EHEC bovines et humaines appartenant ces deux srogroupes donnent des rsultats
dhybridation identiques : les dix-huit souches appartenant au srogroupe O26 sont positives avec la
sonde Eae, tandis que les treize souches appartenant au srogroupe O111 le sont avec la sonde Eae.
Inversement, les souches appartenant dautres srogroupes donnent des rsultats variables. Les variants eae et eae, mais non le variant eae, sont donc
prsents parmi les souches EPEC et EHEC bovines.
Le premier semble plus frquemment associ avec la
prsence de troubles cliniques chez de jeunes veaux.
Cependant, un tiers des souches EPEC isoles de
veaux malades contiennent un variant du gne eae
diffrent des variants , ou . De plus, lutilisation
de ces sondes gntiques na pas permis de diffrencier les souches EHEC bovines et humaines appartenant au mme srogroupe O. Elsevier, Paris

Rsum Htrognit des gnes eae dans des

souches bovines dEscherichia coli : comparaison
avec des souches humaines. Des souches dEscherichia coli entropathognes (EPEC) et entrohmorragiques (EHEC) isoles de bovins ont t tudies par
hybridation ADN sur colonies pour le typage de
leurs gnes eae, au moyen de sondes drives des
parties variables du gne eae de la souche EPEC
humaine E2348/69, du gne eae dune souche
EHEC humaine O157:H7 (ATCC43888) et du gne

The authors wish to thank Dr Georges Daube

from the Dpartement des denres alimentaires
dorigine animale, Facult de mdecine vtrinaire, Universit de Lige, Lige, Belgium, Dr
Eric Oswald from the Laboratoire associ Inra
de microbiologie molculaire, Ecole nationale
vtrinaire, Toulouse, France for helpful discussions, and Dr Denis Pirard from the Akamedisch Ziekenhuis, Vrije Universiteit, Brussels,
Belgium for providing the HEHEC strains and

gne eae / Escherichia coli / locus deffacement des

entrocytes / clonage / sondes gntiques


eae Genes in bovine attaching/effacing E. coli

their serotypes. This work was financially supported by the Ministre fdral des classes
moyennes et de lagriculture, Brussels, Belgium
(research contract 5740), the Fonds national de
la recherche scientifique (Crdit aux chercheurs,
1995), and the University of Lige (Crdit du
conseil de la recherche, 1996).

[1] Abe A., Kenny B., Stein M., Finlay B.B., Characterization of two
virulence proteins secreted by rabbit enteropathogenic Escherichia
coli, EspA and EspB, whose maximal expression is sensitive to host
body temperature, Infect. Immun. 65 (1997) 35473555.
[2] Adu-Bobie J., Frankel G., Bain C., Goncalves A.G., Trabulsi L.,
Douce G., Knutton S., Dougan G., Detection of intimins , , , and
, four intimin derivatives expressed by attaching and effacing
microbial pathogens, J. Clin. Microbiol. 36 (1998) 662668.
[3] Agin T.S., Cantey J.R., Boedeker E.C., Wolf M.K., Characterization
of the eaeA gene from rabbit enteropathogenic Escherichia coli strain
RDEC-1 and comparison to other eaeA genes from bacteria that
cause attaching and effacing lesions, FEMS Microbiol. Lett. 144
(1996) 249258.
[4] An H., Fairbrother J.M., Dubreuil J.D., Harel J., Cloning and cahracterization of the eae gene from a dog attaching and effacing
Escherichia coli strain 4221, FEMS Microbiol. Lett. 148 (1997)
[5] Barrett T.B., Kaper J.B., Jerse A.E., Wachsmuth I.K., Virulence
factors in Shiga-like toxin-producing Escherichia coli isolated from
humans and cattle, J. Infect. Dis. 165 (1992) 979980.
[6] Beebakhee G., Louie M., DeAzavedo J., Brunton J., Cloning and
nucleotide sequence of the eae gene homologue from enterohemorrhagic Escherichia coli serotype O157:H7, FEMS Microbiol. Lett.
91 (1992) 6368.
[7] Beutin L., Escherichia coli as a pathogen in dogs and cats, Vet. Res.
(1999) in press.
[8] Broes A., Les Escherichia coli pathognes du chien et du chat, Ann.
Md. Vt. 137 (1993) 377384.
[9] China B., Pirson V., Jacquemin E., Pohl P., Mainil J., , Pathotypes of
verotoxigenic Escherichia coli isolates producing Attaching/Effacing
(AE) lesions in the ligated intestinal loop assay in rabbits, Adv. Exp.
Med. Biol. 412 (1997) 311316.
[10] China B., Pirson V., Mainil J., Typing of bovine attaching and effacing
Escherichia coli by multiplex in vitro amplification of virulenceassociated genes, Appl. Environ. Microbiol. 62 (1996) 34623465.
[11] China B., Pirson V., Mainil, J., Prevalence and molecular typing of
attaching and effacing Escherichia coli among calf populations in
Belgium, Vet. Microbiol. 63 (1998) 249259.
[12] Cray W.C.J.R., Thomas L.A., Schneider R.A., Moon H.W., Virulence
attributes of Escherichia coli isolated from dairy heifer feces, Vet.
Microbiol. 53 (1996) 369374.
[13] Elliott S.J., Wainwright L.A., McDaniel T.K., Jarvis K.G., Deng Y.K.,
Lai L.C., McNamara B.P., Donnenberg M.S., Kaper, J.B., The complete sequence of the locus of enterocyte effacement (LEE) from
enteropathogenic Escherichia coli E2348/69, Mol. Microbiol. 28
(1998) 14.
[14] Fairbrother J.M., Les colibacilloses du porc, Ann. Md. Vt. 137
(1993) 369375.
[15] Franck S.M., Bosworth B.T., Moon H.W., Multiplex PCR for enterotoxigenic, attaching and effacing, and Shiga toxin-producing Escherichia coli strains from calves, J. Clin. Microbiol. 36 (1998) 17951797.


[16] Goffaux F., Mainil J., Pirson V., Charlier G., Pohl P., Jacquemin E.,
China B., Bovine attaching and effacing Escherichia coli possess a
pathogenesis island related to the LEE of human enteropathogenic
Escherichia coli strain E2348/69, FEMS Microbiol. Lett. 154 (1997)
[17] Jerse A.E., Yu J., Tall B.D., Kaper J.B., A genetic locus of enteropathogenic Escherichia coli necessary for the production of attaching and
effacing lesions on tissue cell cultures, Proc. Nat. Acad. Sci. USA 87
(1990) 78397843.
[18] Karch H., Meyer T., Rssmann H., Heesemann J., Frequent loss of
Shiga-like toxin genes in clinical isolates of Escherichia coli upon
subcultivation, Infect. Immun. 60 (1992) 34643467.
[19] Karch H., Rssmann H., Schmidt H., Schwarzkopf A., Heesemann J.,
Long-term shedding and clonal turnover of enterohaemorrhagic
Escherichia coli O157 in diarrheal disease, J. Clin. Microbiol. 33
(1995) 16021605.
[20] McDaniel T.K., Jarvis K.G., Donnenberg M.S., Kaper J.B., A genetic
locus of enterocyte effacement conserved among diverse enterobacterial pathogens, Proc. Nat. Acad. Sci. USA 92 (1995)
[21] Mainil J.G., Vero/Shigatoxins and vero/shigatoxigenic Escherichia coli
in animals, Vet. Res. (1999) in press.
[22] Mainil J.G., Duchesnes C.J., Whipp S.C., Marques L.R.M., OBrien
A.D., Casey T.A., Moon H.W., Shiga-like toxin production and
attaching and effacing activity of Escherichia coli associated with calf
diarrhea, Am. J. Vet. Res. 48 (1987) 743748.
[23] Mainil J.G., Jacquemin E., Kaeckenbeeck A., Pohl P., Association
between the effacing gene (eae) and the Shiga-like toxin-encoding
genes in Escherichia coli isolates from cattle, Am. J. Vet. Res. 54
(1993) 10641068.
[24] Mainil J., Pohl P., Les souches attachantes et effaantes dEscherichia
coli dorigine bovine, Ann. Md. Vt. 138 (1994) 419429.
[25] Milon A., Oswald E., DeRycke J., Rabbit EPEC: a model for the study
of enteropathogenic Escherichia coli, Vet. Res. (1999) in press.
[26] Moon H.W., Whipp S.C., Argenzio R.A., Levine M.M., Gianella,
R.A., Attaching and effacing activities of rabbit and human enteropathogenic Escherichia coli in pig and rabbit intestines, Infect. Immun. 53 (1983) 13401351.
[27] Nataro J.P., Kaper J.B., Diarrheagenic Escherichia coli, Clin. Microbiol. Rev. 11 (1998) 142201.
[28] OBrien A.D., Laveck G.D., Thompson M.R., Formal S., Production
of Shigella dysenteriae type 1-like cytotoxin by Escherichia coli, J.
Infect. Dis. 146 (1982) 763769.
[29] Paton A.W., Manning P.A., Woodrow M.C., Paton J.C., Translocated
intimin receptors (Tir) of shigatoxigenic Escherichia coli isolates
belonging to serogroups O26, O111, and O157 react with sera
from patients with hemolytic-uremic syndrome and exhibit marked
sequence heterogeneity, Infect. Immun. 66 (1998) 55805586.
[30] Paton A.W., Paton J.C., Detection and characterization of shigatoxigenic Escherichia coli using multiplex PCR assays for stx1, stx2,
enterohemorrhagic Escherichia coli hlyA, rfbO111, and rfbO157, J.
Clin. Microbiol. 36 (1998) 598602.
[31] Peeters J., Les Escherichia coli entropathognes (EPEC) du lapin,
Ann. Md. Vt. 137 (1993) 361368.
[32] Perna N., Mayhew G.F., Posfai G., Elliot S., Donnenberg M.S., Kaper
J.B., Blattner F.R., Molecular evolution of a pathogenicity island from
enterohaemorrhagic Escherichia coli, Infect. Immun. 66 (1998)
[33] Pohl P.H., Peeters J.E., Jacquemin E.R., Lintermans P.F., Mainil, J.G.,
Identification of eae sequences in enteropathogenic Escherichia coli
strains from rabbits, Infect. Immun. 61 (1993) 22032206.
[34] Sandhu K.S., Clarke R.C., McFadden K., Brouwer A., Louie M.,
Wilson J., Lior H., Gyles C.L., Prevalence of the eaeA gene in
verotoxigenic Escherichia coli strains from dairy cattle in Southwest
Ontario, Epidemiol. Infect. 116 (1996) 17.
[35] Sanger F., Nicklen S., Coulson A., DNA sequencing with chain
terminating inhibitors, Proc. Nat. Acad. Sci. USA 74 (1977)


China et al.

[36] Schmidt H., Geitz C., Tarr P.I., Frosch M., Karch H., Non-O157:H7
pathogenic Shiga toxin-producing Escherichia coli: phenotypic and
genetic profiling of virulence traits and evidence for clonality, J.
Infect. Dis. 179 (1999) 115123.
[37] Smith H.R., Willshaw G.A., Scotland S.M., Thomas A., Rowe B.,
Properties of Vero cytotoxin-producing Escherichia coli isolated
from human and non-human sources, Zentralbl. Bakteriol. 278
(1993) 436444.
[38] Tzipori S., Gunzer F., Donnenberg M.S., de Montigny L., Kaper J.B.,
Donohue-Rolfe A., The role of the eaeA gene in diarrhea and

neurological complications in a gnotobiotic piglet model of enterohemorrhagic Escherichia coli infection, Infect. Immun. 63 (1995)
[39] Wieler L.H., Vieler E., Erpenstein C., Sclapp T., Steinrck H.,
Bauerfeind H., Byomi A., Baljer G., Shiga toxin-producing Escherichia coli (STEC) from bovines: association of adhesion with the
carriage of eae and other genes, J. Clin. Microbiol. 34 (1996)
[40] Yu J., Kaper J.B., Cloning and characterization of the eae gene of
enterohaemorrhagic Escherichia coli O157:H7, Mol. Microbiol. 6
(1992) 411417.