Download The rapidly emerging ESBL-producing Escherichia coli O25

RESEARCH LETTER
The rapidly emerging ESBL-producing Escherichia coli O25-ST131
clone carries LPS core synthesis genes of the K-12 type
Valéria Szijártó1,2, Tibor Pal3, Gabor Nagy1, Eszter Nagy1, Akela Ghazawi3, Mohammed al-Haj3,
Sylvia El Kurdi4 & Agnes Sonnevend3
1
Arsanis Biosciences GmbH, Vienna, Austria; 2Department of Microbiology and Immunology, Faculty of Medicine, University of Pécs, Pécs,
Hungary; 3Department of Microbiology and Immunology, Faculty of Medicine and Health Sciences, United Arab Emirates University, Al Ain,
United Arab Emirates; and 4Microbiology Laboratory, Tawam Hospital, Al Ain, United Arab Emirates
Received 29 February 2012; revised 12 April
2012; accepted 23 April 2012.
Final version published online 21 May 2012.
DOI: 10.1111/j.1574-6968.2012.02585.x
MICROBIOLOGY LETTERS
Editor: Stephen Smith
Keywords
O25; ST131; LPS core; K-12; extraintestinal;
Escherichia coli.
Abstract
The clone Escherichia coli O25 ST131, typically producing extended-spectrum
beta-lactamases (ESBLs), has spread globally and became the dominant type
among extraintestinal isolates at many parts of the world. However, the reasons
behind the emergence and success of this clone are only partially understood.
We compared the core type genes by PCR of ESBL-producing and ESBLnonproducing strains isolated from urinary tract infections in the United
Arab Emirates and found a surprisingly high frequency of the K-12 core
type (44.6%) among members of the former group, while in the latter one,
it was as low (3.7%), as reported earlier. The high figure was almost
entirely attributable to the presence of members of the clone O25 ST131
among ESBL producers. Strains from the same clone isolated in Europe also
carried the K-12 core type genes. Sequencing the entire core operon of an
O25 ST131 isolate revealed a high level of similarity to known K-12 core
gene sequences and an almost complete identity with a recently sequenced
non-O25 ST131 fecal isolate. The exact chemical structure and whether and
how this unusual core type contributed to the sudden emergence of ST131
require further investigations.
Introduction
In Escherichia coli, the core oligosaccharide (OS) part
of the lipopolysaccharide (LPS) molecule occurs in five different
types:
R1–4
and
K-12,
respectively
(Muller-Loennies et al., 2007). The core has a crucial role
in maintaining the structure of the cell wall, although to
what extent and how its specific type affects the colonizing capacity or the virulence of a pathogen remains to be
elucidated. Nevertheless, earlier studies consistently found
a highly disproportional distribution of these core types
among commensal and clinical E. coli isolates (Gibb
et al., 1992; Appelmelk et al., 1994; Amor et al., 2000;
Gibbs et al., 2004). Among strains recovered from
extraintestinal infections, the frequency of R1 core type
reached 61.0–81.0%, while that of the K-12 type was
found the least or the second least common (2.2–5.6%)
FEMS Microbiol Lett 332 (2012) 131–136
(Gibb et al., 1992; Appelmelk et al., 1994; Amor et al.,
2000). These frequencies were well reflected by the distribution of core-type-specific antibodies in the population
(Gibbs et al., 2004).
In the past decade, the spread of extended-spectrum
beta-lactamase (ESBL)-producing E. coli strains considerably altered the epidemiology and treatment options of
extraintestinal infections (Woodford et al., 2011; Van der
Bij et al., 2012). A significant percentage of these isolates
belong to a limited number of clones, some considerably
differing in their panel of virulence factors from those
described earlier (Totsika et al., 2011; Van der Bij et al.,
2012). As no data were available on the core types of
these fast spreading clones, we investigated whether the
increased frequency of ESBL-producing organism among
urinary tract isolates has had any impact on the core type
distribution of E. coli isolates.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
Correspondence: Agnes Sonnevend, MD,
PhD, Department of Microbiology and
Immunology, Faculty of Medicine and Health
Sciences, United Arab Emirates University, Al
Ain, PO Box 17666, United Arab Emirates.
Tel.: 00 971 3 7137 481; fax:
00 971 3 7671 966; e-mail:
[email protected]
V. Szijártó et al.
132
Materials and methods
Strain collection and phenotypic
microbiological methods
Molecular typing of the isolates
The phylogenetic type of isolates was established according to (Clermont et al. (2000). Macrorestriction analysis
of the strains was carried out by pulsed field gel electrophoresis (PFGE) using a CHEF-Mapper system (BioRad, Hercules, CA) subsequent to the digestion of the
genom by XbaI (Gautom, 1997). The macrorestriction
patterns were compared according to Dice similarity
index (1–1% tolerance interval) using the GELCOMPARE II
software (Applied Maths, Sint-Martens-Latem, Belgium).
A pulsotype was arbitrarily defined as a cluster of strains
exhibiting macrorestriction banding patterns with
80% similarity. Twenty-four selected isolates representing all pulsotypes were also submitted to multilocus
sequence typing (MLST) (Wirth et al., 2006). The MLST
type of strain SE15 was established in silico, based on
published sequences [GenBank No. AP009378 (Toh
et al., 2010)].
The core type of the isolates was determined by PCR
using primers targeting genes in the core operon and specific the R1–4 and K-12 core types, respectively (Amor
et al., 2000). All strains were also subjected to a PCR
detecting the rfbO25b gene specific to the 25b O serogroup
(Blanco et al., 2009).
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
Genomic DNA of strain 81009 was purified with Wizard
Genomic DNA purification kit (Promega). About 1- to 3-kb
overlapping fragments between genes kbl and coaD were
amplified with KlenTaq LA DNA Polymerase Mix
(Sigma), visualized in 1% agarose gels, purified with
QIAquick Gel Extraction Kit (Qiagen), and sequenced at
Eurofins MWG Operon (Germany). Sequences were
assembled with CLC Main Workbench 6.0.2.
Results and discussion
Comparing the distribution of core-specific genes in
groups of ESBL-producing (n = 121) and ESBL-nonproducing (n = 109) urinary E. coli isolates in the former
group, we detected a surprisingly high rate (44.6%) of
positivity with a K-12 core-specific PCR (Table 1). This
considerably exceeded the rate reported by previous studies (Gibb et al., 1992; Appelmelk et al., 1994; Amor et al.,
2000; Gibbs et al., 2004). As the core type distribution
among non-ESBL-producing strains (3.7%) was similar to
those found earlier (Table 1), and as the production of
ESBL is, at least partly, a clonal phenomenon (Woodford
et al., 2011), the possible clustering of the 58 K-12 core
PCR-positive isolates was investigated. We found that 54
of these strains (93.1%) carried the rfbO25b gene with the
O25 serogroup also confirmed by slide agglutination. All
strains belonged to the B2 phylogenetic group. All isolates, except two, were ESBL-producing strains. Fifty-two
of the 54 K-12 core and rfbO25b-positive strains were
typable by PFGE exhibiting 18 pulsotypes (10 clusters with
2–11 members and 8 singletons) (Fig. 1). Twenty-four
selected isolates representing all pulsotypes were submitted to MLST and found to belong to the rapidly spreading, often multidrug resistant ST131 clone (Fig. 1).
To rule out that the presence of the K-12 core-specific
genes was restricted to Emirati UTI isolates of the O25
ST131 group, ten independent representatives of this
clone isolated in Hungary from UTI (five strains) and
BSI (five strains) in 2008 and 2009, respectively, were also
tested. Importantly, all these strains were also positive
with the K-12 core-specific PCR (Fig. 1).
Next, we determined the DNA sequence of the entire
waa locus (Heinrichs et al., 1998) of one of the O25ST131 isolates from our collection (#81009). The resulting
> 16-kb sequence (GenBank JQ241150) covered the 15
K-12 core genes (Muller-Loennies et al., 2007) between
the kbl and the coaD genes flanking the waa locus. As
expected, based on the PCR results, individual gene
sequences displayed extensive homology to their respective homologues in the prototype K-12 commensal strain,
MG1655 (Table 2). Comparison of the deduced amino
FEMS Microbiol Lett 332 (2012) 131–136
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
All nonrepeat, clinically significant, ESBL-producing
E. coli (n = 121) strains isolated from urine samples in
Tawam Hospital, Al Ain, United Arab Emirates,
between May 2008 and April 2009 were studied and
compared to a pool of matching number of ESBLnonproducing urine isolates (n = 109) collected during
the same period of time. From our strain collection, 10
representatives of the E. coli ST131 clone isolated in
Hungary from urinary tract infection (UTI) (5 strains)
and from bloodstream infection (BSI) (five strains) in
2008 and 2009, respectively, were also tested. Isolates were stored in glycerol at 80 °C. Strains were
identified, and the initial antibiotic susceptibility test
was carried out by the VITEK 2 automated system
(Biomérieux). ESBL production was phenotypically confirmed according to the CLSI standards (CLSI, 2010)
using ceftazidime and cefotaxime discs with and without clavulanic acid. Expression of the O25 cell wall
antigen was determined by slide agglutination using
specific antibodies purchased from the MAST Group
Ltd, Boottle, UK, according to the manufacturer’s
instructions.
Sequencing of the waa locus
K-12 core type synthesis genes of Escherichia coli O25 ST131
133
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
Fig. 1. Pulsotypes of phylogenetic group B2, rfbO25, and K-12 core PCR-positive isolates. *All strains except 306-0838 and 270-2067 were ESBL
producers. MLST – results are shown for all strains typed by MLST. Origin – HUN UTI: strain isolated from UTI in Hungary. HUN BSI: strain
isolated from blood stream infection in Hungary. Strains without marks are all strains isolated from UTIs in Al Ain, UAE. Note: The two
nontypable isolates, one of them also identified as ST131, are not shown.
acid sequences of the various Waa proteins of the ST131
O25 strain #81009 revealed 90% identities with their
counterparts in MG1655 with the exception of WaaQ,
FEMS Microbiol Lett 332 (2012) 131–136
exhibiting a 71% homology, only (Table 2). This enzyme
of strain #81009, however, was 99% identical to its counterparts found in strains with core types R1, R3, and R4
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
V. Szijártó et al.
134
Table 1. Core type distribution and frequency of the rfbO25b locus among urinary tract Escherichia coli isolates
Positive with PCR n (%)
Core types
ESBL (n = 121)
non-ESBL (n = 109)
rfbO25b
K-12
R1
R2
R3
R4
52 (43.0)
2 (1.8)
54 (44.6)
4 (3.7)
21 (17.4)
69 (63.3)
11 (9.1)
11 (10.1)
11 (9.1)
17 (15.6)
24 (19.8)
8 (7.3)
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
from those of the other E. coli LPS core types (MullerLoennies et al., 2007). However, it remains to be
elucidated whether the 4–10% nonidentity of the LPS
synthesis enzymes of the tested ST131 strain and the prototype K-12 MG1655 strain is reflected in any differences
in the chemical composition of the outer core. It is interesting to note that an unusual glycoform composition of
the K-12 core was recently described in a strain isolated
from bovine mastitis, although no sequence of the encoding locus has been made available for comparison (Duda
et al., 2011).
In light of the previously found low frequency of the
K-12 core type among E. coli strains, it is intriguing to
contemplate why the highly successful ESBL-producing
ST131 clone carries this type seldom harbored by pathogenic E. coli (Amor et al., 2000). Unlike the strain
MG1655, that is, a phylogenetic group A strains characterized with limited virulence, members of the ST131 clone,
and in general, those of the B2 phylogenetic group are
characterized with considerable extraintestinal pathogenic
potential (Totsika et al., 2011; Van der Bij et al., 2012).
Although the role of anticore antibodies in interfering
with bacterial colonization is still speculative, a hypothesis
was recently proposed regarding their contribution to
prevent mucosal infections, such as the one caused by
E. coli O157 (Currie et al., 2001). Further investigations
should clarify whether the low anti-K12 core antibody
titers found in the population earlier (Gibbs et al., 2004)
could have contributed to a permissive environment
allowing the rapid spread of the K-12 core-containing
strains, such as the members of ST131 clone, in the gut
and in extraintestinal niches.
As most of the epidemiological studies revealing the
frequency of various core types and core-specific antibodies were conducted prior the emergence of the ST131
clone (Gibb et al., 1992; Appelmelk et al., 1994; Amor
et al., 2000; Gibbs et al., 2004), it remains to be seen
whether its recent spread has had any effect on the prevalence of antibodies with the respective specificities. As
our clinical isolates were preselected according to ESBL
production, these data do not allow drawing a direct conclusion regarding the current frequency of strains with a
K-12 core type in UTI. However, as the incidence of
FEMS Microbiol Lett 332 (2012) 131–136
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
(Table 2), while the WaaQ protein encoded by the
MG1655 allele was identical to that of a representative R2
strain, F632. Because the function of this protein as a
heptosyltransferase is completely conserved in all core
types (Muller-Loennies et al., 2007), we surmise that this
sequence variation is unlikely to have functional consequences.
An almost 100% identity (except two nucleotide differences resulting in a single amino acid mismatch in WaaB)
of the entire waa locus of the strain #81009 was found
with that of a commensal fecal isolate SE15 of the B2
phylogenetic group (Toh et al., 2010) (Table 2). Importantly, although SE15 reportedly expressed serotype O150:
H5, its sequence type, as revealed by the in silico MLST
analysis from the published sequence, was ST131.
This is the first report on the complete core operon
sequence of an O25 ST131 isolate. Recently, two groups
reported on the total genome sequences of O25 ST131
(Avasthi et al., 2011; Totsika et al., 2011) and deposited
it in GenBank; however, none of them contained the
complete waa cluster. In strain EC958 (Totsika et al.,
2011), the locus annotated as ‘O-antigen 2’ and available
as parts of two nonoverlapping contigs (GenBank
CAFL01000107.1 and CAFL01000108.1) contained the
waa genes, which, with the exception of a 293-bp-long
fragment missing from the waaR gene, exhibited 100%
identity with the waa operon of strain #81009.
Similarly, the sequences of the waaA, waaQ, waaG,
waaP, waaC, waaF, and waaD genes of another O25
ST131 strain (NA114) (Avasthi et al., 2011) were 100%
identical to the respective genes of our isolate. However,
a large fragment corresponding to the sequence between
4715–12806 bp of our ST131 isolate (GenBank JQ241150)
was missing from the sequence available in the database.
As this represents a considerable part of the waa operon,
including the complete waaB, waaI, waaR, waaY, waaZ,
waaU genes and parts of waaS and waaL genes, an extensive comparison between the waa operons of stains
#81009 and NA114 was not possible.
The high level of similarity in the genetic background
of core synthesis of the ST131 strains to that of strain
MG1655 suggests that it is also likely to be similar to the
known structure of the K-12 core, but definitely different
99
†
†
100
100
71
99
71
99
99
91
100
89
99
99
90
92
86
92
92
‡
‡
92
‡
‡
85
‡
‡
‡
†
93
†
†
†
99
99
99
99
100
†
100
†
MG1655 § (U00096.2)
CFT073 (AE014075.1)
F632 (AF019375)
042 (FN554766.1)
F2513 (AF019746)
*GenBank accession numbers of respective waa loci.
†
No or only partial CDS available.
‡
Gene not present.
§
Waa proteins of laboratory strains W3110 (AP009048.1), DH10B (NC_010473.1), and BW2952 (CP001396.1) are identical to those of MG1655.
‡
‡
‡
95
95
99
94
98
24
11
11
24
‡
‡
95
98
52
91
54
52
96
39
88
38
40
‡
93
97
50
86
51
49
‡
100
100
100
100
99
100
100
O150:H5 ST131
commensal
fecal isolate
K-12 core
R1 core
R2 core
R3 core
R4 core
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
O25 ST131
O25 ST131
NA114 (CP002797)
EC 958 (CAFL01000107
and CAFL01000108)
SE15 (AP009378.1)
100
100
100
100
100
100
100
†
99
100
WaaA
WaaQ
WaaG
WaaP
100
†
‡
‡
WaaS
WaaB
WaaI
WaaR
‡
‡
‡
‡
WaaY
WaaZ
WaaU
†
WaaL
WaaC
WaaF
WaaD
Features
Strains (accession number) *
% Amino acid identity in proteins
FEMS Microbiol Lett 332 (2012) 131–136
135
third-generation cephalosporin resistance among local
E. coli isolates during the period of strain collection was
23.7% (Al-Kaabi et al., 2011) and because 44.6% of the
ESBL-producing isolates were positive with the K-12 core
PCR, a considerable increase in K-12-type E. coli compared to the figures found earlier, that is, 2.2–5.6% (Gibb
et al., 1992; Appelmelk et al., 1994; Amor et al., 2000),
can be anticipated.
The rapid spread of the ST131 clone and the fact that
it still keeps evolving by acquiring genes as blaKPC-2 or
blaNDM-1 (Morris et al., 2011; Peirano et al., 2011) further
extending its antibiotic resistance emphasize the need to
identify the factors responsible for its fitness and virulence. Revealing the genetic background for its LPS core
OS synthesis may contribute to finding some of the
answers and may even lead to the development of preventive and curative interventions.
Acknowledgements
This work was supported by grants FMHS NP-10/07,
UAEU1636-08-01-10 and 1439-08-02-01.
Disclosure
V.S.Z., G.N. and E.N. are employees of a Arsanis, a biotechnology company. The authors declare no potential
conflict of interest.
References
Al-Kaabi MR, Tariq WU & Hassanein AA (2011) Rising
bacterial resistance to common antibiotics in Al Ain, United
Arab Emirates. East Mediterr Health J 17: 479–484.
Amor K, Heinrichs DE, Frirdich E, Ziebell K, Johnson RP &
Whitfield C (2000) Distribution of core oligosaccharide
types in lipopolysaccharides from Escherichia coli. Infect
Immun 68: 1116–1124.
Appelmelk BJ, An YQ, Hekker TA, Thijs LG, MacLaren DM &
de Graaf J (1994) Frequencies of lipopolysaccharide core
types in Escherichia coli strains from bacteraemic patients.
Microbiology 140(Pt 5): 1119–1124.
Avasthi TS, Kumar N, Baddam R, Hussain A, Nandanwar N,
Jadhav S & Ahmed N (2011) Genome of multidrug-resistant
uropathogenic Escherichia coli strain NA114 from India.
J Bacteriol 193: 4272–4273.
Blanco M, Alonso MP, Nicolas-Chanoine MH et al. (2009)
Molecular epidemiology of Escherichia coli producing
extended-spectrum {beta}-lactamases in Lugo (Spain):
dissemination of clone O25b:H4-ST131 producing CTX-M15. J Antimicrob Chemother 63: 1135–1141.
Clermont O, Bonacorsi S & Bingen E (2000) Rapid and simple
determination of the Escherichia coli phylogenetic group.
Appl Environ Microbiol 66: 4555–4558.
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
Table 2. Amino acid homology of enzymes involved in the LPS core synthesis of Escherichia coli O25 ST131 strain # 81009 compared to corresponding enzymes of E. coli strains carrying various
core types
K-12 core type synthesis genes of Escherichia coli O25 ST131
136
ª 2012 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
belonging to the international ST131 clone. Antimicrob
Agents Chemother 55: 4935–4936.
Muller-Loennies S, Brade L & Brade H (2007)
Neutralizing and cross-reactive antibodies against
enterobacterial lipopolysaccharide. Int J Med Microbiol 297:
321–340.
Peirano G, Schreckenberger PC & Pitout JD (2011)
Characteristics of NDM-1-producing Escherichia coli isolates
that belong to the successful and virulent clone ST131.
Antimicrob Agents Chemother 55: 2986–2988.
Toh H, Oshima K, Toyoda A et al. (2010) Complete genome
sequence of the wild-type commensal Escherichia coli strain
SE15, belonging to phylogenetic group B2. J Bacteriol 192:
1165–1166.
Totsika M, Beatson SA, Sarkar S et al. (2011) Insights into a
multidrug resistant Escherichia coli pathogen of the globally
disseminated ST131 lineage: genome analysis and virulence
mechanisms. PLoS ONE 6: e26578.
Van der Bij AK, Peirano G, Pitondo-Silva A & Pitout JD
(2012) The presence of genes encoding for different
virulence factors in clonally related Escherichia coli
that produce CTX-Ms. Diagn Microbiol Infect Dis 72:
297–302.
Wirth T, Falush D, Lan R et al. (2006) Sex and virulence in
Escherichia coli: an evolutionary perspective. Mol Microbiol
60: 1136–1151.
Woodford N, Turton JF & Livermore DM (2011)
Multiresistant Gram-negative bacteria: the role of high-risk
clones in the dissemination of antibiotic resistance. FEMS
Microbiol Rev 35: 736–755.
FEMS Microbiol Lett 332 (2012) 131–136
Downloaded from http://femsle.oxfordjournals.org/ by guest on June 22, 2016
CLSI (2010) Performance Standards for Antimicorbial
Susceptibility Testing, M100-S20, Vol. 30. Clinical and
Laboratory Standards Institute, Wayne, Pennsylvania, USA.
Currie CG, McCallum K & Poxton IR (2001) Mucosal and
systemic antibody responses to the lipopolysaccharide of
Escherichia coli O157 in health and disease. J Med Microbiol
50: 345–354.
Duda KA, Lindner B, Brade H, Leimbach A, Brzuszkiewicz E,
Dobrindt U & Holst O (2011) The lipopolysaccharide of the
mastitis isolate Escherichia coli strain 1303 comprises a novel
O-antigen and the rare K-12 core type. Microbiology 157:
1750–1760.
Gautom RK (1997) Rapid pulsed-field gel electrophoresis
protocol for typing of Escherichia coli O157:H7 and other
gram-negative organisms in 1 day. J Clin Microbiol 35:
2977–2980.
Gibb AP, Barclay GR, Poxton IR & di Padova F (1992)
Frequencies of lipopolysaccharide core types among clinical
isolates of Escherichia coli defined with monoclonal
antibodies. J Infect Dis 166: 1051–1057.
Gibbs RJ, Stewart J & Poxton IR (2004) The distribution of,
and antibody response to, the core lipopolysaccharide
region of Escherichia coli isolated from the faeces of healthy
humans and cattle. J Med Microbiol 53: 959–964.
Heinrichs DE, Yethon JA & Whitfield C (1998) Molecular
basis of structural diversity in the core region of the
lipopolysaccharide of Escherichia coli and Salmonella
enterica. Mol Microbiol 30: 221–232.
Morris D, Boyle F, Ludden C et al. (2011) Production of
KPC-2 carbapenemase by an Escherichia coli clinical isolate
V. Szijártó et al.