Abstract

The high-pathogenicity island (HPI) present in pathogenic Yersinia and encoding the siderophore yersiniabactin, has recently been identified in the asnT tRNA region of various Escherichia coli pathotypes, especially those responsible for bacteremia and urosepsis. Most E. coli strains causing such extra-intestinal infections belong to phylogenetic groups B2 and D. In this study we investigated (i) the distribution and localization of HPI among the different E. coli phylogenetic groups, using the ECOR reference collection; and (ii) the prevalence of HPI among a set of 124 phylogenetically characterized E. coli strains responsible for neonatal meningitis. Ninety-three percent of the ECOR strains belonging to groups B2 and D harbored HPI. In contrast, the island was present in 32% and 25% of strains belonging to groups A and B1, respectively, which are considered to be non-pathogenic. HPI was found in 100% of the neonatal meningitis strains, 13 of which belonged to groups A and B1, suggesting that HPI might contain virulent factors required for the development of neonatal meningitis. Moreover, we observed for the first time that HPI can be inserted in a site different from the asnT tRNA region.

Introduction

Escherichia coli is both the most common commensal bacterium and the most frequent cause of community-acquired bacterial infections. Phylogenetic and epidemiological studies of virulence factors (VF) in well-characterized clinical collections have thrown light on the reasons for this ‘duality’. E. coli is divided into four main phylogenetic groups, designated A, B1, B2 and D [1,2]. Most E. coli strains responsible for urinary tract infections, septicemia and neonatal meningitis belong to group B2 or, to a lesser extent, to group D [3–5]. This may be in part explained by the fact that group B2 and D strains harbor numerous VF lacking in group A and B1 strains [3–6].

VF involved in iron transport systems play a key role in extra-intestinal E. coli infections. The best illustration of the relationship between VF involved in iron metabolism and phylogenic groups is the newly described heme transporter system represented by the gene chuA [7]. We recently demonstrated the presence of chuA in all strains of groups B2 and D, and its absence from all group A and B1 strains [8], a finding that may have implications for phylogenetic grouping methods and diagnosis. The Yersinia high-pathogenicity island (HPI) contains an integrase gene (int) as well as the genes irp1, irp2, and fyuA, which are involved in iron uptake [9,10]. HPI was recently identified in E. coli and is integrated in the asnT tRNA gene [11–15]. Initial epidemiological studies indicate a high prevalence of HPI in E. coli isolated by blood culture, and also intriguingly in certain pathotypes such as entero-aggregative and Shiga toxin-producing E. coli [4,13–15].

We investigated the distribution of HPI relative to the previously studied iron uptake systems chuA and iucC [3,8] in the different phylogenetic groups of E. coli, using ECOR reference strains [16]. We also analyzed the prevalence of HPI in a set of 124 phylogenetically characterized E. coli strains responsible for neonatal meningitis.

Materials and methods

Bacterial strains

The 72 E. coli strains of the ECOR collection [16] were kindly provided by R. Selander (Department of Biology, University of Rochester, Rochester, NY, USA). These reference strains, isolated from a variety of hosts and geographic locations, are representative of the range of genotypic variation in the species. Sixty-eight belong to the four main phylogenetic groups (A, B1, B2, D), and the remaining four are unclassified [1,2]. We also examined a set of 124 E. coli strains causing neonatal meningitis (ECNM). The phylogenetic group distribution and the prevalence of the chuA and iucC genes have previously been reported in 86 of these strains [3,8]. The remaining 38 strains were characterized with previously published methods [3,8]. The uropathogenic E. coli strain 536 (HPI+) [12] and the laboratory E. coli K12 strain MG1655 (HPI−) [17] were used as positive and negative PCR controls, respectively.

PCR and dot blotting

The PCR reaction was carried out in a 20-μl volume, with 2 μl of 10× buffer (supplied with Taq polymerase), 20 pmol of each primer, 200 μM each dNTP, 2.5 U of Taq polymerase (ATGC Biotechnologie) and 200 ng of genomic DNA. PCR was performed with an I-Cycler (Bio-Rad) in the following conditions: denaturation 5 min at 94°C, 30 cycles of 10 s at 94°C, 20 s at 55°C and 30 s at 72°C (2 min for fyuA.1/yeeJ amplification), and a final extension step of 7 min at 72°C. The primers used for PCR are shown in Table 1. They were chosen among previously published sequences or were specifically designed for this study, with the following aims: to demonstrate (i) the presence of HPI by amplification of the fyuA and irp2 genes, and (ii) the adjacency of HPI to the asnT tRNA gene, by testing the amplification of the left and right junctions (Table 1), and the DNA region between asnT and yeeJ, which are contiguous in E. coli K12 [17] and separate in E. coli strains harboring HPI (Fig. 1) [12–14].

1

PCR primers used to study HPI

Primer designation Primer sequence Target Size of PCR product (bp) Primers reference 
fyuA.1 5′-TGATTAACCCCGCGACGGGAA-3′ fyuA 780 [4
fyuA.2 5′-CGCAGTAGGCACGATGTTGTA-3′   [13
irp2.1 5′-AAGGATTCGCTGTTACCGGAC-3′ irp2 280 [13
irp2.2 5′-TCGTCGGGCAGCGTTTCTTCT-3′   [13
asnT 5′-GACAGACAAGGTACCGCTAA-3′ asnT-yeeJ junction as in E. coli K12 790 [14
yeeJ 5′-CGGGACATCCAGTTCATCAC-3′   This study 
asnT 5′-GACAGACAAGGTACCGCTAA-3′ HPI left junction 1200 or 1500 [14
int.2 5′-TGCTTCCAGATAATCCGACCAC-3′   [14
fyuA.1 5′-TGATTAACCCCGCGACGGGAA-3′ HPI right junction 2460 [4
yeeJ 5′-CGGGACATCCAGTTCATCAC-3′   This study 
Primer designation Primer sequence Target Size of PCR product (bp) Primers reference 
fyuA.1 5′-TGATTAACCCCGCGACGGGAA-3′ fyuA 780 [4
fyuA.2 5′-CGCAGTAGGCACGATGTTGTA-3′   [13
irp2.1 5′-AAGGATTCGCTGTTACCGGAC-3′ irp2 280 [13
irp2.2 5′-TCGTCGGGCAGCGTTTCTTCT-3′   [13
asnT 5′-GACAGACAAGGTACCGCTAA-3′ asnT-yeeJ junction as in E. coli K12 790 [14
yeeJ 5′-CGGGACATCCAGTTCATCAC-3′   This study 
asnT 5′-GACAGACAAGGTACCGCTAA-3′ HPI left junction 1200 or 1500 [14
int.2 5′-TGCTTCCAGATAATCCGACCAC-3′   [14
fyuA.1 5′-TGATTAACCCCGCGACGGGAA-3′ HPI right junction 2460 [4
yeeJ 5′-CGGGACATCCAGTTCATCAC-3′   This study 
1

Schematic representation of the HPI region. Genes are indicated by large black arrows (HPI section) or by open arrows (K12-like section). The PCR primers used for the study are indicated by small arrows: 1, asnT; 2, int.2; 3, irp2.1; 4, irp2.2; 5, fyuA.1; 6, fyuA.2; and 7, yeeJ (see Table 1).

1

Schematic representation of the HPI region. Genes are indicated by large black arrows (HPI section) or by open arrows (K12-like section). The PCR primers used for the study are indicated by small arrows: 1, asnT; 2, int.2; 3, irp2.1; 4, irp2.2; 5, fyuA.1; 6, fyuA.2; and 7, yeeJ (see Table 1).

When the standard PCR protocol was negative, long-range PCR was performed with the Expand Long Template PCR System (Roche) to amplify the DNA region between asnT and yeeJ using the same primers. The PCR reaction was carried out according to the manufacturer's instructions in a 30-μl volume, with 3 μl of buffer 3, 12 pmol of each primer, 500 μM each dNTP, 2.5 U of Taq mix and 200 ng of genomic DNA.

Dot blotting for the fyuA and irp2 genes was performed on DNA extracted from each strain and spotted on positively charged nylon membranes. Hybridization was performed at 65°C in 1% SDS/1 M NaCl/Tris 50 mM HCl, pH 7.5/1% blocking reagent (Roche). The membranes were washed in 2×SSC for 15 min at room temperature, then in 2×SSC/0.1% SDS for 30 min at 65°C, and finally in 0.1×SSC for 5 min at room temperature. Chemiluminescence detection was performed according to the manufacturer's instructions (DIG Luminescent Detection kit for nucleic acids, Roche). The probes were produced by PCR according to the manufacturer's instructions (PCR DIG Probe Synthesis kit, Roche), and using the primers and amplification procedure described above.

Results

The distributions of fyuA and irp2 in the ECOR collection and in the set of strains causing neonatal meningitis are shown in Table 2. All the PCR results were confirmed by dot blot hybridization (not shown). The distribution of other iron transport systems, the aerobactin system (iucC) and the heme captation system (chuA) is also presented for comparison. When positive, fyuA and irp2 were always associated. In the ECOR collection, HPI was the most frequent iron transport system relative to the chuA and iucC genes, with a global prevalence of 54%, 42% and 32%, respectively. As regards the phylogenetic groups, HPI was more frequent in groups B2 and D (100% and 83%, respectively) than in group A (32%) or B1 (25%). Interestingly, 100% of our 124 neonatal meningitis strains harbored HPI, whatever their phylogenetic group.

2

Distribution of the iron uptake systems among the different E. coli phylogenetic groups in the ECOR collection and a collection of strains causing neonatal meningitis (ECNM)

Genes Group A (%) Group B1 (%) Group D (%) Group B2 (%) Reference 
 ECOR n=25 ECNM n=9 ECOR n=16 ECNM n=4 ECOR n=12 ECNM n=22 ECOR n=15 ECNM n=89  
chuA 100 100 100 100 [8], This study 
iucC 24 100 12 50 67 91 32 79 [3], This study 
fyuA/irp2 32 100 25 100 83 100 100 100 This study 
Genes Group A (%) Group B1 (%) Group D (%) Group B2 (%) Reference 
 ECOR n=25 ECNM n=9 ECOR n=16 ECNM n=4 ECOR n=12 ECNM n=22 ECOR n=15 ECNM n=89  
chuA 100 100 100 100 [8], This study 
iucC 24 100 12 50 67 91 32 79 [3], This study 
fyuA/irp2 32 100 25 100 83 100 100 100 This study 

It was recently reported that the HPI in E. coli is adjacent to the asnT tRNA gene (Fig. 1) [12–14]. We analyzed the location of HPI in the ECOR collection by using PCR primers specifically designed (i) to demonstrate the interruption of the asnT region by HPI insertion, and (ii) to amplify the left and right junctions between HPI and the asnT region as defined in Table 1. Different PCR products from representative strains are shown in Fig. 2. All but one of the 39 HPI-positive strains had an interrupted asnT region. Amplification of the right junction was successful in all these 38 strains, and amplification of the left junction was successful in all but three strains. Amplification of the left junction yielded products of two different sizes (1200 bp in ECOR 2, 9, 10, 24 belonging to group A; ECOR 68, 70, 71 belonging to group B1; and the unclassified ECOR 43 and 1500 bp in the other strains) (Fig. 2). These results suggested that HPI was effectively inserted in the asnT region of these 38 strains, and that left junctions were not identical in all strains. ECOR 72 was the only HPI-positive strain with an intact asnT region and in which the left and right junctions could not be amplified, demonstrating that HPI was inserted in another site in this strain. Surprisingly, three group A strains (ECOR 15, 22, 23) devoid of HPI were not amplified by standard PCR protocol in the region between asnT and yeeJ, suggesting that asnT was interrupted; moreover, the left junction was amplified in all three strains, with a 1500-bp product. In these strains, long-range PCR between asnT and yeeJ yielded a 15-kb product for ECOR 15 and ECOR 23, and a 4.5-kb product for ECOR 22 (Fig. 2).

2

Agarose gel showing PCR products from representative strains. K12, E. coli K12 strain MG1655 (HPI−, asnT/yeeJ+); E536, E. coli strain 536 (HPI+, asnT/yeeJ −); E2, ECOR 2 (HPI+, asnT/yeeJ −, truncated integrase); E72, ECOR 72 (HPI+, asnT/yeeJ+); E15, ECOR 15; E22, ECOR 22; E23, ECOR 23; M, molecular mass marker. AsnT/yeeJ PCR products were obtained by long-range PCR (see Section 2).

2

Agarose gel showing PCR products from representative strains. K12, E. coli K12 strain MG1655 (HPI−, asnT/yeeJ+); E536, E. coli strain 536 (HPI+, asnT/yeeJ −); E2, ECOR 2 (HPI+, asnT/yeeJ −, truncated integrase); E72, ECOR 72 (HPI+, asnT/yeeJ+); E15, ECOR 15; E22, ECOR 22; E23, ECOR 23; M, molecular mass marker. AsnT/yeeJ PCR products were obtained by long-range PCR (see Section 2).

Discussion

The main advances in our understanding of extra-intestinal E. coli virulence are the discovery of new VF and the relationship between phylogenetic groups and virulence. Numerous extra-intestinal VF are known to be organized in blocks called pathogenicity islands (PAIs) [18] and are particularly frequent in uropathogenic and neonatal meningitis strains [19,20]. These PAIs may be selectively distributed among the phylogenetic groups. For example, PAIs containing the Pap adhesin are principally found in groups B2 and D [6], while the PAI containing Sfa adhesin is almost exclusively found in group B2 [3,6]. To identify a possible relation between HPI and phylogenetic grouping in E. coli, we investigated the distribution and localization of a recently described PAI designated HPI, which is involved in iron uptake. We focused our investigations on two well-characterized E. coli strain collections.

HPI was highly prevalent (93%) in group B2 and D strains of ECOR collection, further supporting its probable key role in extra-intestinal infections. The results for group B1 and especially group A may appear puzzling. The HPI prevalence of 32% in group A was unexpected in a group considered non-pathogenic. The results support the hypothesis that HPI was acquired by groups B2 and D, which are known to be sister groups [21], soon after their emergence. Some group A and B1 strains may then have acquired HPI by horizontal transfer. An other hypothesis consists in the acquisition of HPI by a common ancestor and its subsequently loss by the majority of strains belonging to groups A and B1. However, the paradoxical distribution of HPI in group A remains to be explained.

The probable major role of HPI in the pathogenesis of extra-intestinal infections is supported by our results with neonatal meningitis strains. The highest prevalence previously reported in E. coli was 93%[4,13]. To our knowledge, this is the first time that HPI has been found in 100% of strains of a given E. coli responsible of a specific pathology. Our previous studies of ECNM showed no such prevalence of other iron uptake systems (iucC and chuA). This may in part be explained by the higher iron affinity constant of yersiniabactin compared to aerobactin (1030 versus 1023) [22]. Our most striking result is the presence of HPI in all neonatal meningitis strains belonging to groups A and B1; this 100% prevalence rate was significantly higher than that found in the ECOR strains of the same groups (P <0.05), and suggests that HPI might contain VF required for the development of neonatal meningitis.

An intriguing result of this study is the amplification of a left junction of HPI in three group A strains devoid of fyuA and irp2, with a 4.5-kb or 15-kb PCR product between asnT and yeeJ, compared to the 35–45-kb product for HPI [11]. One possible explanation is the loss of a previously acquired HPI. Alternatively, another genomic island might be present at this site, harboring the same HPI integrase. The lack of amplification of ybtP, ybtQ, ybtX, and ybtS, which are known to be localized between the int and irp2 genes [11] (data not shown), is in favor of the second hypothesis.

Knowledge of the insertion of HPI in three different asn tRNA loci named asnT, asnU and asnV in Yersinia pseudotuberculosis [9] prompted Schubert et al. to analyze the location of HPI in E. coli [12–14] and up to now, asnT tRNA has been the unique site incriminated in this species. Here, we demonstrate that HPI may be inserted at a site other than asnT in E. coli. However it must be pointed out that ECOR strain 72 belongs to group B1, a group rarely encountered in human infections, thus the HPI insertion in an atypical site that we notice might be an exception.

Acknowledgements

We thank Elisabeth Carniel for helpful discussion and Grégory Sicard for technical help. This work was supported in part by the Programme de Recherche Fondamentale en Microbiologie, Maladies Infectieuses et Parasitaires (Appel d'offre 1998), ‘Recherche de déterminants génétiques de pathogénicité chez E. coli K1 responsable de méningite néonatale’ and by the Programme Hospitalier de Recherche Clinique (Grant AOM 96069).

References

[1]
Selander
R.K.
Caugant
D.A.
Whittam
T.S.
(
1987
)
Genetic structure and variation in natural populations of Escherichia coli
. In:
Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology
  (
Neidhardt
F.C.
Ingraham
J.L.
Low
K.B.
Magasanik
B.
Schaechter
M.
Umbarger
H.E.
, Eds.), pp.
1625
1648
.
American Society for Microbiology
,
Washington, DC
.
[2]
Herzer
P.J.
Inouye
S
Inouye
M
Whittam
T.S.
(
1990
)
Phylogenetic distribution of branched RNA-linked multi-copy single-stranded DNA among natural isolates of Escherichia coli
.
J. Bacteriol.
 
172
,
6175
6181
.
[3]
Bingen
E
Picard
B
Brahimi
N
Mathy
S
Desjardins
P
Elion
J
Denamur
E
(
1998
)
Phylogenetic analysis of Escherichia coli strains causing neonatal meningitis suggests horizontal gene transfer from a predominant pool of highly virulent B2 group strains
.
J. Infect. Dis.
 
177
,
642
650
.
[4]
Johnson
J.R.
Stell
A.L.
(
2000
)
Extended virulence genotypes of Escherichia coli strains from patients with urosepsis in relation to phylogeny and host compromise
.
J. Infect. Dis.
 
181
,
261
272
.
[5]
Picard
B
Garcia
J.S.
Gouriou
S
Duriez
P
Brahimi
N
Bingen
E
Elion
J
Denamur
E
(
1999
)
The link between phylogeny and virulence in Escherichia coli extraintestinal infection
.
Infect. Immun.
 
67
,
546
553
.
[6]
Boyd
E.F.
Hartl
D.L.
(
1998
)
Chromosomal regions specific to pathogenic isolates of Escherichia coli have a phylogenetically clustered distribution
.
J. Bacteriol.
 
180
,
1159
1165
.
[7]
Mills
M
Payne
S
(
1995
)
Genetics and regulation of haem iron transport in Shigella dysenteriae and detection of an analogous system in Escherichia coli O157: H7
.
J. Bacteriol.
 
177
,
3004
3009
.
[8]
Clermont
O
Bonacorsi
S
Bingen
E
(
2000
)
Rapid and simple determination of the Escherichia coli phylogenetic group
.
Appl. Environ. Microbiol.
 
66
,
4555
4558
.
[9]
Buchrieser
C
Brosch
R
Bach
S
Guiyoule
A
Carniel
E
(
1998
)
The high-pathogenicity island of Yersinia pseudotuberculosis can be inserted into any of the three chromosomal asn tRNA genes
.
Mol. Microbiol.
 
30
,
965
978
.
[10]
Carniel
E
Guilvout
I
Prentice
M
(
1996
)
Characterization of a large chromosomal ‘high-pathogenicity island’ in biotype 1B Yersinia enterocolitica
.
J. Bacteriol.
 
178
,
6743
6751
.
[11]
Bach
S
de Almeida
A
Carniel
E
(
2000
)
The Yersinia high-pathogenicity island is present in different members of the family Enterobacteriaceae
.
FEMS Microbiol. Lett.
 
183
,
289
294
.
[12]
Schubert
S
Rakin
A
Fischer
D
Sorsa
J
Heesemann
J
(
1999
)
Characterization of the integration site of Yersinia high-pathogenicity island in Escherichia coli
.
FEMS Microbiol. Lett.
 
179
,
409
414
.
[13]
Schubert
S
Rakin
A
Karch
H
Carniel
E
Heesemann
J
(
1998
)
Prevalence of the ‘high-pathogenicity island’ of Yersinia species among Escherichia coli strains that are pathogenic to humans
.
Infect. Immun.
 
66
,
480
485
.
[14]
Karch
H
Schubert
S
Zhang
D
Zhang
W
Schmidt
H
Olschlager
T
Hacker
J
(
1999
)
A genomic island, termed high-pathogenicity island, is present in certain non-O157 Shiga toxin-producing Escherichia coli clonal lineages
.
Infect. Immun.
 
67
,
5994
6001
.
[15]
Schubert
S
Cuenca
S
Fischer
D
Heesemann
J
(
2000
)
High-pathogenicity island of Yersinia pestis in enterobacteriaceae isolated from blood cultures and urine samples: prevalence and functional expression
.
J. Infect. Dis.
 
182
,
1268
1271
.
[16]
Ochman
H
Selander
R.K.
(
1984
)
Standard reference strains of Escherichia coli from natural populations
.
J. Bacteriol.
 
157
,
690
692
.
[17]
Blattner
F.R.
Plunkett
G.I.
Bloch
C.A.
Perna
N.T.
Burland
V
Riley
M
(
1997
)
The complete genome sequence of Escherichia coli K-12
.
Science
 
277
,
1453
1461
.
[18]
Hacker
J
Blum-Oehler
G
Muhldorfer
I
Tschape
H
(
1997
)
Pathogenicity islands of virulent bacteria, structure, function and impact on microbial evolution
.
Mol. Microbiol.
 
23
,
1089
1097
.
[19]
Bonacorsi
S.P.
Clermont
O
Tinsley
C
Le Gall
I
Beaudoin
J.C.
Elion
J
Nassif
X
Bingen
E
(
2000
)
Identification of regions of the Escherichia coli chromosome specific for neonatal meningitis-associated strains
.
Infect. Immun.
 
68
,
2096
2101
.
[20]
Guyer
D.M.
Kao
J.S.
Mobley
H.L.
(
1998
)
Genomic analysis of a pathogenicity island in uropathogenic Escherichia coli CFT073, distribution of homologous sequences among isolates from patients with pyelonephritis, cystitis, and catheter-associated bacteriuria and from fecal samples
.
Infect. Immun.
 
66
,
4411
4417
.
[21]
Lecointre
G
Rachdi
L
Darlu
P
Denamur
E
(
1998
)
Escherichia coli molecular phylogeny using the incongruence length difference test
.
Mol. Biol. Evol.
 
15
,
1685
1695
.
[22]
Gehring
A.M.
DeMoll
E
Fetherston
J.D.
Mori
I
Mayhew
G.F.
Blattner
F.R.
Walsh
C.T.
Perry
R.D.
(
1998
)
Iron acquisition in plague: modular logic in enzymatic biogenesis of yersiniabactin by Yersinia pestis
.
Chem. Biol.
 
5
,
573
586
.