- Email: [email protected]
Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing human disease in Germany: Diversity, prevalence, and outbreaks
Accepted Manuscript Title: Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing human disease in Germany: diversity, prevalence, and outbreaks Author: Angelika Fruth Rita Prager Erhard Tietze Wolfgang Rabsch Antje Flieger PII: DOI: Reference:
S1438-4221(15)00083-1 http://dx.doi.org/doi:10.1016/j.ijmm.2015.08.020 IJMM 50982
To appear in:
Please cite this article as: Fruth, A., Prager, R., Tietze, E., Rabsch, W., Flieger, A.,Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing human disease in Germany: diversity, prevalence, and outbreaks, International Journal of Medical Microbiology (2015), http://dx.doi.org/10.1016/j.ijmm.2015.08.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
MINI-REVIEW
2 3
Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing
4
human disease in Germany: diversity, prevalence, and outbreaks
5
Angelika Fruth, Rita Prager, Erhard Tietze, Wolfgang Rabsch, and Antje Flieger
ip t
6 7
Division of Enteropathogenic Bacteria and Legionella, National Reference Centre for
9
Salmonella and other Enteric Bacterial Pathogens, Robert Koch Institute, Wernigerode Branch, Burgstr. 37, 38855 Wernigerode, Germany
an
11
us
10
cr
8
12
M
13 14
ed
15
ce pt
16 17 18
Ac
19 20
Corresponding author:
21
Antje Flieger
22
Phone: +49 (0)30 18754 2522
23
Fax: +49 (0)30 18754 4207
24
e-mail: [email protected]
25 26 27 1 Page 1 of 26
Summary
29
Infections by intestinal pathogenic Escherichia coli (E. coli) are among those causing a high
30
mortality and morbidity due to diarrheal disease and post infection sequelae worldwide. Since
31
introduction of the Infection Protection Act in Germany 2001, these pathogens rank third
32
among bacterial infections of the gastrointestinal tract. As a major pathovar Shiga toxin-
33
producing E. coli (STEC) which include enterohemorrhagic E. coli (EHEC) play a leading
34
role in occurrence of sporadic cases and disease outbreaks. An outstanding example is the
35
large outbreak in spring 2011 caused by EHEC/EAEC O104:H4. To monitor and trace back
36
STEC infections, national surveillance programs have been implemented including activities
37
of the German National Reference Centre for Salmonella and other Enteric Bacterial
38
Pathogens (NRC). This review highlights advances in our understanding of STEC in the last
39
20 years of STEC surveillance by the NRC. Here important characteristics of STEC strains
40
from human infections and outbreaks in Germany between 1997 and 2013 are summarized.
an
us
cr
ip t
28
41
Key words
43
STEC, EHEC, HUS, serovar, phage type, diversity, outbreak investigations, antibiotic
44
resistance
ed
M
42
45
Introduction
47
The German National Reference Centre for Salmonella and other Enteric Bacterial Pathogens
48
(NRC) hosted by the Robert Koch Institute Division of Enteropathogenic Bacteria and
49
Legionella has been surveilling important bacterial diarrheal pathogens causing disease in
50
humans
51
http://www.rki.de/DE/Content/Infekt/NRZ/Salmonellen/salmo_node.html;jsessionid=1D05B5
52
E97ACCB8543EF4E41721E3E0E4.2_cid381). Our intention is to obtain an overview about
53
the strain types and their prevalence, virulence, antibiotic resistance, and persistence. Further,
54
changes in the pathogen types are recognized which may be indicative of new emerging or
55
disappearing strain variants and may give a hint on outbreaks. Outbreak detection is
56
performed in cooperation with the department of infection epidemiology of the Robert Koch
57
Institute and other partners, such as the Consultant Laboratory for Hemolytic Uremic
58
Syndrome (HUS) at the University of Münster, the Public Health Authorities of the German
59
Federal States and the Federal Institute of Risk Assessment (BfR). Upon strain comparison
60
and by means of specific strain properties also infection sources can be traced together with
ce pt
46
1993
(Homepage:
Ac
since
2 Page 2 of 26
61
other public health, food or veterinary authorities. The NRC further is involved in the
62
development and improvement of diagnostic procedures.
63
Within the 20 years of the work of the NRC, more than 12,000 Shiga toxin-producing E. coli
65
(STEC) isolates were collected and thoroughly analyzed yielding in a comprehensively
66
studied strain collection. The term STEC is used throughout this article for strains of both,
67
typical enterohemorrhagic E. coli (EHEC) serovars which are highly associated with sequelae
68
like hemorrhagic colitis (HC) and HUS, involving thrombocytopenia, haemolytic anemia, and
69
acute renal failure, and many other serovars which are only rarely associated with severe
70
human disease. With respect to intestinal pathogenic E. coli microbiology, pathogenicity,
71
clinical issues, and evolution, the reader is referred to a recently published comprehensive
72
review (Croxen et al., 2013) and therefore, we here will concentrate on EHEC surveillance at
73
NRC. Within the period of STEC surveillance, the strains analyzed at NRC represent about
74
60-80% of the reported cases in Germany ranging from mild or bloody diarrhea up to HUS.
75
To the purpose of strain characterization, bacterial isolates are mainly sent from public health
76
authorities and in some cases from primary diagnostic laboratories to the NRC for subtyping.
77
At the NRC, a variety of classical phenotypic as well as DNA-based methods are used.
M
an
us
cr
ip t
64
ed
78
In the case of STEC, strain isolation is currently rarely performed in primary diagnostic
80
laboratories. The procedure is very labour intensive and in most cases the determination of the
81
presence of the Shiga toxin or its genes from stool samples or enrichment cultures is sufficient
82
for therapeutic decisions and case reporting according to the German Infection Protection Act.
83
Therefore, isolation of the pathogen and further subtyping is usually performed in specialized
84
or reference laboratories. Phenotypic and DNA-based subtyping is performed on STEC
85
isolates for E. coli pathovar determination and for clone differentiation for epidemiological
86
purposes. A broad spectrum of established mostly PCR-based methods for the detection of
87
virulence determinants, such as adhesin or toxin genes and their variants, is employed for
88
pathovar characterization (see table 1; Flieger et. al., 2013). These PCR-based detections of
89
gene regions are important methods for epidemiological analysis of STEC however cannot
90
evaluate functionality of gene expression or of the corresponding proteins since non-
91
functional degenerated gene mutants have been described (Abu-Ali et al., 2010). Strain
92
subtyping by means of serotyping either via classical or genoserotyping and by
93
biochemotyping, such as sorbitol fermentation, is carried out for all STEC isolates sent to the
94
NRC. Further, international standardized pulsed-field gel electrophoresis (PFGE) after DNA
Ac
ce pt
79
3 Page 3 of 26
95
macrorestriction is still the method of choice for reliable identification of outbreak-associated
96
STEC clones. Serotyping, PCR-based virulence gene profiling, and PFGE analysis were
97
important tools applied during the 2011 EHEC/EAEC O104:H4 outbreak for characterization
98
of the outbreak strain and for tracing of infection clusters (Bielaszewska et al., 2011; Frank et
99
al., 2011; Sin et al., 2013; Robert Koch Institute, 2011). In this review, we present a summary of important characteristics of STEC isolated from
101
human infections in Germany between 1997 and 2013. Used abbreviations and gene names
102
with respect to STEC subtyping are explained and referenced in Table 1.
cr
103
ip t
100
STEC O serogroups causing human disease
105
Serotyping is an important method for STEC subtyping. At the NRC classical serotyping is
106
performed as described by Ørskov and Ørskov (Ørskov and Ørskov, 1984). So far 186
107
O serogroups (O1-O192 except O31, O47, O67, O72, O93 and O94) and 53 associated
108
H forms (H1-56 except H13, H22, H50) are currently described (see WHO Center for
109
Klebsiella and Escherichia coli, Copenhagen, Denmark for further information). Within the
110
here reported period, 139 different O serogroups were detected among STEC strains. Most
111
commonly found O serogroups among STEC were O157, O91, O26 and O103 which
112
represent about 40 to 50% of the STEC strains analyzed at NRC starting from 1999 (Fig. 1).
113
Excluding 2011, this shows that there are no major changes in O serogroup distribution and
114
consequently in their significance in human disease. It is also obvious that the percentage of
115
O157 strains decreased from about 25% in 1999 to about 10% in 2013. This may reflect the
116
growing awareness of other STEC serogroups in diagnostics rather than a significant decrease
117
in O157 infections. Accordingly, the fraction of O91 strains increased from about 5% in 1999
118
to about 15% in 2012 and 2013. The large outbreak due to EHEC/EAEC O104:H4 in 2011
119
highly influenced O serogroup distribution by representing about 50% of the strains analyzed
120
in that year. Most frequent H type combinations of the most important O serogroups were
121
O157:Hnm/H7, O91:Hnm/H14, O26:Hnm/H11, and O103:H2 (Table 2). It is further
122
interesting to note that about 10% of the strains yearly are either untypable or cannot be
123
attributed to a known O serogroup which shows a limitation of classical serotyping compared
124
to DNA-based methods addressing coding genes for expression of serotype-related traits.
Ac
ce pt
ed
M
an
us
104
125 126
In summary, O26, O91, O103, and O157 were the most common O serogroups occurring
127
among STEC from 1997 to 2013 and only slight changes in their distribution can be observed. 4 Page 4 of 26
128
Except in 2011, the large outbreak due to EHEC/EAEC O104:H4 dominated O serogroup
129
percentages (Fig. 1).
130
STEC and association with stx genotypes, with HUS, and occurrence of hybrid
132
pathovars
133
In the early 1990s, the historical approach to assign the pathovar STEC to strains of specific
134
serovars needed certain revisions. Because of the dynamic nature of the Shiga toxin-
135
harboring phages and the limited chance to detect these pathogens later during infection,
136
different assays based on detection of Shiga toxin and Shiga toxin genes improved the capture
137
of diverse STEC. By application of such specific virulence gene PCR assays, the NRC
138
collected a great variety of STEC serovars and virulence gene profile since 1999 (Fig. 1 and
139
Table 3). In addition to known correlations of most frequent serovars with stx1 (such as O26 or
140
O91 strains) or stx2 (such as O157 strains), changes in stx gene type distribution were
141
observed. For example, stx2 gene positive isolates of serovar O26:H11/Hnm emerged with an
142
increasing tendency to harbour different stx2 genes. Only 7% of STEC O26:H11/Hnm
143
harboured stx2 in 1999 but 59% in 2013 (Fig. 2; Bielaszewska et al., 2013).
M
an
us
cr
ip t
131
ed
144
Due to the awareness of severe outcome of STEC infection, the HUSEC strain collection was
146
established and published by Mellmann et al. (2008). It represents the most common strains
147
isolated from patients suffering from HUS. Various HUS-associated isolates belonging to
148
additional serovars were identified by the NRC (Table 3). Among these are isolates of
149
serovars O177:H11, O156:H25, O109:Hnt, and O80:H2 carrying stx2a and eaeA genes.
150
Further, this includes the first description of an O181:H49 isolate with a variety of virulence-
151
associated genes (stx1a, stx2a, stx2c, stx2d, sub and saa).
Ac
152
ce pt
145
153
Genomic plasticity and horizontal gene transfer enable emergence of STEC strains with
154
additionally acquired virulence properties of intestinal or extraintestinal pathogenic E. coli
155
(Müller et al., 2007; Orth et al., 2007). Such hybrid forms of STEC and other pathovars were
156
targeted by PCR analysis of additional virulence genes at the NRC. Accordingly, two novel
157
STEC strains of serovars O59:Hnm and Orough:Hnm were identified harbouring virulence
158
genes involved in aggregative adhesion characteristic for EAEC (Table4; Prager et al., 2014).
159
Further, novel strains carrying stx2g and estIa were described which represent a hybrid variant
160
of STEC and ETEC (Table 4; Prager et al., 2011). 5 Page 5 of 26
Phage typing of STEC
162
Typing of E.coli by lysis pattern caused by specific bacteriophages is described for O157 and
163
other serovars (Milch, 1978). Additionally for the serovar O157, a Canadian system was
164
developed by Ahmed et al. (1987) and extended by Khakhria et al. (1990). This scheme was
165
used widely as a tool for epidemiological investigations in Canada, the United States,
166
England, Wales, and Scotland. From 2001 till 2009, this Canadian scheme which describes 88
167
distinct phage types by application of sixteen different phages (Duck et al., 1997) was used at
168
the NRC. 826 tested strains of serovar O157 from Germany collected between 2001 and 2009
169
yielded in 30 different phage types. During this time, the dominant phage types were PT8
170
(30%), PT88 (20%), PT14 (15%), PT 2 (12 %), PT 4 (5%), and 18% belonged to 25 different
171
PTs (Fig. 3). The distribution of stx1 and stx2 phages among the different phage types was
172
interesting. For example, the strains of sorbitol-fermenting (sf) O157:Hnm PT88 carried in
173
100% and of O157 PT2 in 96% the stx2 phage only.
an
us
cr
ip t
161
174
Phage typing was also helpful for epidemiological analysis in outbreak investigations. In
176
2001, an outbreak in the German Federal State Saxony-Anhalt caused by sf O157/Hnm STEC
177
of PT88 was observed in a kindergarden. Among 12 patients several HUS cases occurred.
178
STEC sf O157:Hnm of PT 88 have been observed in several outbreaks in Germany, yet their
179
reservoir and routes of transmission remain unknown (Alpers et al., 2009). In 2008 an
180
outbreak by STEC O157 PT14 was traced back to raw milk and children who visited a farm in
181
Lower Saxony were affected (Kirchner et al., 2013).
ed
ce pt
182
M
175
With the introduction of new molecular subtyping methods, the future of STEC phage typing
184
is under debate. But for a scientific view on STEC prophages which carry additional genes
185
(e.g. virulence genes like stx1, stx2) or for the application of the next generation of
186
bacteriophage therapy (Lu and Koeris, 2011), the available sets of phages may be helpful.
187
Ac
183
188
Antibiotic resistance among STEC
189
Antibiotic treatment of EHEC infections is problematic and controversial and is not regularly
190
recommended especially during the acute diarrhoeal phase but is for certain cases, such as
191
pathogen-shedding without clinical symptoms, under debate (Serna et al., 2008, Nitschke et
192
al., 2012). Nevertheless, at the NRC, the susceptibility of STEC towards 16 antimicrobial
193
substances is determined by a broth microdilution method routinely to obtain the antibiogram 6 Page 6 of 26
of the pathogens in order to use it as an epidemiological marker. The breakpoints for the
195
classification “resistant” are mainly according to DIN and were not adapted to any of the
196
more recent updates by CLSI or EUCAST, respectively, in order to maintain consistency in
197
our epidemiological surveillance of antibiotic resistance development. The substances tested
198
include tetracycline (4 mg/l), ampicillin (8), mezlocillin (16), the combination of mezlocillin
199
with sulbactam (16), chloramphenicol (8), the combination of trimethoprim and
200
sulfamethoxazole – cotrimoxazole – (16), streptomycin (16), kanamycin (16), gentamicin (4),
201
amikacin (16), nalidixic acid (16) ciprofloxacin (2), cefotiam (4), cefoxitin (16), cefotaxime
202
(8) and ceftazidime (16). Since 1997, among all of the STEC sent to the NRC, a rather
203
constant yearly proportion of around 70% is tested susceptible to all of these substances (Fig.
204
4). Between 15 and 20% are resistant to one or two antibiotics, and about 10% of STEC
205
isolates are resistant to three and more - rarely up to 13 - of the substances tested. The most
206
frequently observed resistance phenotypes are resistance towards streptomycin, tetracycline
207
and ampicillin (about 15 to 20% of all resistant STEC) followed by resistance towards
208
chloramphenicol (about 10% of all resistant STEC). Resistance of STEC towards
209
fluorchinolones varied over the years below one percent of all resistant STEC reaching 0.7 %
210
cumulated for the years 2012 and 2013. A clearly increasing trend is observed for resistance
211
towards cotrimoxazole. While between 1997 and 2004 about 7% of all resistant STEC
212
showed this resistance phenotype, the respective proportion of cotrimoxazole-resistant STEC
213
rose to over 12% during the period between 2005 and 2011 and reached about 30% cumulated
214
for the years 2012 and 2013. Another trend that seems to be looming is a rising resistance
215
quotient among STEC for cephalosporins as mediated by extended spectrum beta lactamases
216
(ESBL). While up to the year 2010 the proportion of STEC with an ESBL phenotype
217
(reduced susceptibility or resistance towards cefotaxime and/ or ceftazidime) remained below
218
1%, this proportion rose up to 4% of all resistant STEC cumulated for the years 2012 and
219
2013. A very prominent example of an STEC with an ESBL phenotype is the O104:H4 strain
220
which caused the large EHEC outbreak in Germany in 2011. This strain harbored the gene for
221
the ESBL enzyme CTX-M-15 on a plasmid (Robert Koch Institute, 2011). Plasmid-encoded
222
CTX-M-15 is widely distributed in many E. coli strains of different pathovars (Rocha-Garcia
223
et al., 2015, Ewers et al., 2014). Other STEC isolates with an ESBL phenotype were from
224
sporadic cases and were serotyped as O80:H2, O179:H8, O80:Hnt and O16:Hnt. Thus,
225
although selection pressure towards resistant STEC by therapeutic use of antibiotics in
226
clinical cases can be assumed to be low, the surveillance of antibiograms at the NRC reveals
227
the continuous prevalence of antibiotic resistant STEC in Germany over the past 16 years.
Ac
ce pt
ed
M
an
us
cr
ip t
194
7 Page 7 of 26
228
One must assume, therefore, that these pathogens are subject to ecological processes creating
229
and selecting antibiotic resistant strains in their reservoirs.
230
STEC surveillance and outbreak investigations
232
The NRC uses for the surveillance, outbreak investigations and identification of new
233
emerging STEC types different molecular approaches such as virulence gene PCR assays,
234
PFGE, MLST, MLVA and plasmid profile analysis. Because there is no single combination of
235
virulence genes that defines STEC as potential EHEC, a broad range of virulence genes is
236
investigated by PCR. After the 2011 EHEC O104:H4 outbreak, the panel of the routinely
237
tested stx and eaeA genes was extended by the EAEC genes aatA and aggR.
us
cr
ip t
231
238
For surveillance and to predict risks of STEC infections, the knowledge of the stx subtypes
240
and colonization factors and adhesins is essential. The NRC uses the stx subtyping strategies
241
and nomenclature as described by Scheutz et al. (2012). The NRC as national public health
242
reference laboratory also participates every year in the EQA ring trials coordinated by the
243
ECDC, Stockholm, covering the detection of stx1, stx2, eaeA, aggR and aaiC genes and
244
subtyping of stx1 into three known subtypes (stx1a, stx1c, stx1d) and stx2 genes into seven
245
known subtypes (stx2a, stx2b, stx2c, stx2d, stx2e, stx2f, stx2g). It is well known that some stx
246
subtypes, especially variant stx2a, in combination with eaeA or with aatA/aggR are often
247
associated with severe clinical outcomes (e.g. sf O157:Hnm or O104:H4; Tables 3 and 5;
248
Werber et al., 2003). But other virulence gene combinations may also be associated with
249
severe disease in humans, including HUS (see Table 3). Therefore predicting the emergence
250
of „new‟ highly pathogenic STEC types based only on the presence of the stx genes, or by
251
focusing on a restricted panel of serogroups alone is not possible (Orth et al., 2006; Prager et
252
al., 2005; Fasel et al., 2014). The STEC virulence plasmids (pO157, psfO157 or pO113) also
253
play an important role in severe clinical outcomes. Especially eaeA-negative isolates
254
possessing plasmids of type pO113 (Srimanote et al., 2002) in combination with variant stx2d
255
were more often described in correlation with HUS (see Tables 3 and 5). Therefore the NRC
256
uses the plasmid profile analysis as an important tool for surveillance and outbreak
257
investigations, not only for analysis of STEC. Thus it could be shown by the NRC very early
258
that the EAEC virulence plasmids and the plasmid profiles in the outbreak strain EHEC
259
O104:H4 and HUSEC O104:H4 were different (EHEC O104:H4
260
Erregers sowie Hinweise und Hilfestellungen des RKI zur Diagnostik des gegenwärtig
Ac
ce pt
ed
M
an
239
- Eigenschaften des
8 Page 8 of 26
261
zirkulierenden
262
http://www.rki.de/DE/Content/InfAZ/E/EHEC/EHEC_O104/EHEC_O104_Diagnostik.pdf?_
263
_blob=publicationFile; Mellmann et al., 2011).
264
DNA macrorestriction and pulsed-field gel electrophoresis (PFGE) for the subtyping of STEC
265
is widely used to compare strains for epidemiological purposes. The method currently
266
represents the gold standard for the investigation of outbreaks and source-tracing. The NRC
267
uses the standardized PFGE protocol developed by the CDC Atlanta which is part of EQA’s
268
coordinated by the ECDC Stockholm. The NRC as national public health reference laboratory
269
has been successfully participating every year in these EQA’s covering the gel quality and gel
270
analysis by means of BioNumerics (BioNumerics, Applied Math, Ghent, Belgium). This
271
standardized method is a basic requirement to contribute to the international molecular typing
272
pilot project of the ECDC with the aim to detect molecular clusters. The NRC takes an active
273
part in this pilot project. Serovar-specific PFGE library databases based on the restriction with
274
XbaI were established at the NRC for the most important STEC serovars showing severe
275
clinical outcomes and involvement in outbreaks, such as (sf) O157:H7/Hnm, O26:H11/Hnm,
276
O145:Hnm, O103:H2/Hnm. This is an essential precondition for the detection of outbreaks,
277
molecular clusters, or new clones. Altogether, by PFGE 500 (sf and non-sorbitol-fermenting
278
(nsf)) O157:H7/Hnm isolates were subtyped into 265 different patterns, 250 O26:H11/Hnm
279
isolates into 142 patterns, 120 O103:H2/Hnm isolates into 53 patterns and 48 O145:Hnm
280
isolates into 30 patterns. In case of an outbreak, PFGE analysis is done with a second enzyme
281
(BlnI) as recommended by CDC (PulseNet International, 2013). During outbreak
282
investigations or for cluster detections, the NRC works together with the department of
283
Infectious Disease Epidemiology of the Robert Koch Institute and the National Reference
284
Laboratory for E. coli at the German Federal Institute for Risk Assessment (BfR), especially
285
for source tracing. The universal applicability of PFGE for subtyping of isolates other than the
286
most common serovars was shown (Prager et al., 2009, 2011; Bielaszewska et al., 2011;
287
Frank et al., 2011). In addition to PFGE, the NRC also uses for the investigations of genetic
288
relationships MLVA and MLST analysis and in the future the application of whole genome
289
sequencing and the verification of the method in outbreak analysis will be compassed (Prager
290
et al., 2009, 2011).
Available:
Ac
ce pt
ed
M
an
us
cr
ip t
Ausbruchstammes,
291 292
Conclusions
293
We here presented an overview on the characteristics of the STEC strains isolated from
294
human disease in Germany during the last 17 years. STEC comprise strains with a high 9 Page 9 of 26
potential to cause severe symptoms in humans and outbreaks in association with consumption
296
of contaminated food. At the beginning of that period, classical serotyping with a focus on
297
O157 strain detection was a major method for STEC risk assessment. Today, analysis of a
298
wide array of virulence or phylogenetic markers by DNA-based methods together with
299
serotyping allows a more complete view on the pathogen’s virulence profile and phylogeny.
300
In the future, a more complete picture will arise from the implementation of transcriptomics,
301
proteomics, and whole genome sequencing. Especially the latter is currently in progress at
302
NRC and has the potential to revolutionize STEC surveillance and outbreak analysis.
303
However, the established methods will certainly remain in use for the near future, both to
304
serve the diagnostic and epidemiological needs of the public health system and to complement
305
the genome-based approaches in the process of their evaluation.
us
cr
ip t
295
306
Acknowledgements
308
We thank the team members of the Robert Koch Institute Division of Enteropathogenic
309
Bacteria and Legionella for their contributions to subtyping of bacteria and preparation of the
310
manuscript. We further are indebted to all the diagnostic laboratories providing strains for
311
epidemiological analysis.
ed
312
M
an
307
References
314
Abu-Ali, G.S., Ouellette, L.M., Henderson, S.T., Lacher, D.W., Riordan, J.T., Whittam, T.S.,
315
Manning, S.D., 2010. Increased adherence and expression of virulence genes in a lineage
316
of Escherichia coli O157:H7 commonly associated with human infections. PLoS One 5
317
(4).e10167.
319
Ahmed, R., Bopp, C., Borczyk, A., Kasatiya, S., 1987. Phage-typing scheme for E. coli 0157:
Ac
318
ce pt
313
H7. J. Infect. Dis. 155, 806-809.
320
Alpers, K., Werber, D., Frank, C., Koch, J., Friedrich, A.W., Karch, H., an der Heiden, M.,
321
Prager, R., Fruth, A., Bielaszewska, M., Morlock, G., Heissenhuber, A., Diedler, A.,
322
Gerber, A., Ammon, A., 2009. Sorbitol-fermenting enterohaemorrhagic Escherichia coli
323
O157:H- causes another outbreak of haemolytic uraemic syndrome in children. Epidemiol.
324
Infect. 137, 389-395.
325
Bielaszewska, M., Mellmann, A., Zhang, W., Köck, R., Fruth, A., Bauwens, A., Peters, G.,
326
Karch, H., 2011. Characterisation of the Escherichia coli strain associated with an outbreak 10 Page 10 of 26
327
of haemolytic uraemic syndrome in Germany, 2011: a microbiological study. Lancet
328
Infect. Dis. 11(9):671-676. Bielaszewska, M., Mellmann, A., Bletz, S., Zhang, W., Köck, R., Kossow, A., Prager, R.,
330
Fruth, A., Orth-Höller, D., Marejková, M., Morabito, S., Caprioli, A., Piérard, D., Smith,
331
G., Jenkins, C., Curová, K., Karch, H., 2013. Enterohemorrhagic Escherichia coli
332
O26:H11/H-: a new virulent clone emerges in Europe. Clin. Infect. Dis. 2013 56(10), 1373-
333
1381.
ip t
329
Boisen, N., Struve, C., Scheutz, F., Krogfelt, K.A., Nataro, J.P., 2008. New adhesin of
335
enteroaggregative Escherichia coli related to the Afa/Dr/AAF family. Infect. Immun.
336
76(7), 3281-3292.
339
us
338
Croxen, M.A., Law, R.J., Scholz, R., Keeney, K.M., Wlodarska, M., Finley, B.B., 2013. Recent advances in understanding enteric pathogenic Escherichia coli. CMR 26, 822-880.
an
337
cr
334
Duck, D., Demczuk, W., McKinnon, J., Mulvey, M., Clarke, C., Woodward, D., Khakhria, R., Ahmed, R., 1997. Escherichia coli O157: H7 phage typing. Natl . Lab. Enteric Pathogens,
341
Lab. Centre Dis. Control, Ottawa, Ontario, Canada. 4/1/97.
M
340
Dudley, E.G., Thomson, N.R., Parkhill, J., Morin, N.P., Nataro, J.P., 2006. Proteomic and
343
microarray characterization of the AggR regulon identifies a pheU pathogenicity island in
344
enteroaggregative Escherichia coli. Mol. Microbiol. 61(5), 1267-1282.
ed
342
Ewers, C., Stamm, I., Pfeifer, Y., Wieler, L.H., Kopp, P.A., SchØnning, K., Prenger-
346
Berninghoff, E., Scheufen, S., Stolle, I., Günther, S., Bethe, A., 2014. Clonal spread of
347
highly successful ST15-CTX-M-15 Klebsiella pneumoniae in companion animals and
348
horses. J. Antimicrobial. Chemother. 69 (10), 2676-2680.
ce pt
345
Fasel, D., Mellmann, A., Cernela, N., Hächler, H., Fruth, A., Khanna, N., Egli, A., Beckmann,
350
C., Hirsch, H.H., Goldenberger, D., Stephan, R., 2014. Hemolytic uremic syndrome in a
351
65-Year-old male linked to a very unusual type of stx2e- and eae-harboring O51:H49 shiga
352
toxin-producing Escherichia coli. J. Clin. Microbiol. 52 (4), 1301-1303.
Ac
349
353
Flieger, A., Mielke, M., Tietze, E., 2013. Role of pathogen surveillance and subtyping in
354
outbreak recognition of food-borne bacterial infections. Microbiological perspective -
355
Aims,
356
Gesundheitsforschung Gesundheitsschutz. 56(1), 42-46.
methods
and
prospects
of
pathogen
subtyping.
Bundesgesundheitsblatt
11 Page 11 of 26
357
Frank, C., Werber, D., Cramer, J.P., Askar, M., Faber, M., an der Heiden, M., Bernard, H.,
358
Fruth, A., Prager, R., Spode, A., Wadl, M., Zoufaly, A., Jordan, S., Kemper, M.J., Follin,
359
P., Müller, L., King, L.A., Rosner, B., Buchholz, U., Stark, K., Krause, G., HUS
360
Investigation Team, 2011. Epidemic profile of Shiga-toxin-producing Escherichia coli
361
O104:H4 outbreak in Germany. New Engl. J. Med. 365(19), 1771-1780. Frost, J.A., Smith, H.R, Willshaw, G.A., Scotland, S.M., Gross, R.J., Rowe, B., 1989.
363
Phagetyping of Verocytotoxin (VT) producing E. coli O157 isolated in the United
364
Kingdom. Epidemiol. Infect. 103, 73- 81.
cr
366
Khakhria, R., Duck, D., Lior, H., 1990. Extended phage typing scheme for E. coli O157:H7. Epidemiol. Infect. 105, 511-520.
us
365
ip t
362
Kirchner, M., Dildei, C., Runge, M., Brix, A., Claussen, K., Weiss, U., Fruth, A., Prager, R.,
368
Mellmann, A., Beutin, L., Miko, A., Wichmann-Schauer, H., Pulz, M., Dreesman, J., 2013.
369
Outbreak of non-sorbitol-fermenting shiga toxin-producing E. coli O157:H7 infections
370
among school children associated with raw milk consumption in Germany. Archiv für
371
Lebensmittelhygiene 64, 68-74.
374 375
M
the hag gene encoding flagellin of Escherichia coli. J. Bacteriol. 168, 1479–1483.
ed
373
Kuwajima, G., Asaka, J.-I., Fujiwara, T., Node, K., Kondo. E.,1986. Nucleotide sequence of
Lu, T.K., Koeris, M.S., 2011. The next generation of bacteriophage therapy. Curr. Opin. Microbiol. 14, 524-31.
ce pt
372
an
367
Lu, Y., Iyoda, S., Satou, H., Satou, H., Itoh, K., Saitoh, T., Watanabe, H., 2006. A new
377
immunoglobulin-binding protein, EibG, is responsible for the chain-like adhesion
378
phenotype of locus of enterocyte effacement-negative, shiga toxin-producing Escherichia
379
coli. Infect. Immun. 74(10), 5747-5755.
380 381
Ac
376
Milch, H., 1978. Phage Typing of Escherichia coli. In: Bergan, T., Norris, J.R. (Eds.), Microbial Methods, CHAPTER III, Academic Press London, pp. 87-156.
382
Mellmann, A., Bielaszewska, M., Koeck, R., Friedrich, A.W., Fruth, A., Middendorf B.,
383
Harmsen, D., Schmidt, M.A., Karch , H., 2008. Serotypes and Phylogenetic Analysis of
384
the Collection of Hemolytic Uremic Syndrome-Associated enterohemorrhagic Escherichia
385
coli (HUSEC). EID 14(8), 1287-1290.
386
Mellmann, A., Harmsen, D., Cummings, C.A., Zentz, E.B., Leopold, S.R., Rico, A., Prior, K.,
387
Szczepanowski, R., Ji, Y., Zhang, W., McLaughlin, S.F., Henkhaus, J.K., Leopold, B., 12 Page 12 of 26
388
Bielaszewska, M., Prager, R., Brzoska, P.M., Moore, R., Guenther, S., Rothberg, J.M.,
389
Karch, H., 2011. Prospective genomic characterization of the German enterohemorrhagic
390
Escherichia coli O104:H4 outbreak by rapid next generation sequencing technology. PloS
391
One 6(7), e22751. Müller, D., Greune, L., Heusipp, G., Karch, H., Fruth, A., Tschäpe, H., Schmidt, M.A., 2007.
393
Identification of unconventional intestinal pathogenic Escherichia coli isolates expressing
394
intermediate virulence factor profiles by using a novel single-step multiplex PCR. Appl.
395
Environ. Microbiol., 73(10), 3380-3390.
cr
ip t
392
Nada, R.A., Armstrong, A., Shaheen, H.I., Nakhla, I., Sanders, J.W., Riddle, M.S., Young, S.,
397
Sebeny, P., 2013. Phenotypic and genotypic characterization of enterotoxigenic
398
Escherichia coli isolated from U.S. military personnel participating in Operation Bright
399
Star, Egypt, from 2005 to 2009. Diagn. Microbiol. Infect. Dis. 76(3), 272-277.
an
us
396
Nielsen, S., Frank, C., Fruth, A., Spode, A., Prager, R., Graff, A., Plenge-Bönig, A. Loos, S.,
401
Lütgehetmann, M., Kemper, M.J., Müller-Wiefel, D.E., Werber, D., 2011. Desperately
402
seeking diarrhoea: outbreak of haemolytic uraemic syndrome caused by emerging sorbitol-
403
fermenting shiga toxin-producing Escherichia coli O157:H-, Germany, 2009. Zoonoses
404
Public Health. 58(8), 567-572.
ed
M
400
Nitschke, M., Sayk, F., Härtel, C., Roseland, R.T., Hauswaldt, S., Steinhoff, J., Fellermann,
406
K., Derad, I., Wellhöner, P., Büning, J., Tiemer, B., Katalinic, A., Rupp, J., Lehnert, H.,
407
Solbach, W., Knobloch, J.K., 2012. Association between azithromycin therapy and
408
duration
409
enteroaggregative Escherichia coli O104:H4. JAMA 307 (10), 1046-1052.
of
ce pt
405
bacterial
shedding
among
patients
with
Shiga
toxin-producing
Ørskov, F., Ørskov, I.,1984. Serotyping of Escherichia coli. Methods Microbiol. 14, 43-112.
411
Orth, D., Grif, K., Khan, A.B., Naim, A., Dierich, M.P., Würzner, R., 2007. The Shiga toxin
412
genotype rather than the amount of Shiga toxin or the cytotxicity of Shige toxin in vitro
413
correlates with the apparence of the hemolytic syndrome. Diagn. Microbiol. Infect. Dis. 59,
414
235-242.
Ac
410
415
Orth, D., Grif, K., Fisher, I., Fruth, A., Tschäpe, H., Scheutz, F., Dierich, M.P., Würzner, R.,
416
2006. Emerging Shiga toxin-producing Escherichia coli serotypes in Europe: O100:H- and
417
O127:H40. Curr. Microbiol. 53(5), 428-429.
13 Page 13 of 26
418
Prager, R., Annemüller, S., Tschäpe, H., 2005. Diversity of virulence patterns among shiga
419
toxin-producing Escherichia coli from human clinical cases – need for more detailed
420
diagnostics. Int. J. Med. Microbiol. 295, 29-38. Prager, R., Fruth, A., Siewert, U., Strutz, U., Tschäpe, H., 2009. Escherichia coli encoding
422
Shiga toxin 2f as an emerging human pathogen. Int. J. Med. Microbiol. 299(5), 343-353.
423
Prager, R., Fruth, A., Busch, U., Tietze, E., 2011. Comparative analysis of virulence genes,
424
genetic diversity, and phylogeny of Shiga toxin 2g and heat-stable enterotoxin STIa
425
encoding Escherichia coli isolates from humans, animals, and environmental sources. Int.
426
J. Med. Microbiol. 301(3), 181-191.
cr
ip t
421
Prager, R., Lang, C., Aurass, P., Fruth, A., Tietze, E., Flieger, A., 2014. Two Novel
428
EHEC/EAEC Hybrid Strains Isolated from Human Infections. PLoS One. 9(4), e95379.
429
PulseNet International, 2013. Standard operating procedure for PulseNet PFGE of
430
Escherichia coli O157:H7, Escherichia coli non-O157 (STEC), Salmonella serotypes,
431
Shigella sonnei and Shigella flexneri. PNL05. PulseNet International, Atlanta, GA.
432
http://www.pulsenetinternational.org/assets/PulseNet/uploads/pfge/PNL05_Ec-Sal-
433
ShigPFGEprotocol.pdf.
an
M
Robert Koch Institute, 2011. Characteristics of the pathogen and information and assistance
ed
434
us
427
435
by
436
http://www.rki.de/EN/Home/EHECO104.pdf?__blob=publicationFile.
438
RKI
in
diagnosis
of
the
currently
circulating
outbreak
strain.
ce pt
437
the
Robert Koch-Institut, 2011. Bakteriologische Untersuchungen im Rahmen des Ausbruchs mit E. coli O104:H4. Epid. Bull. 35, 325-329. Rocha-Garcia, R.C., Cortés-Cortés, G., Lozano-Zarain, P., Belo, F., Martínez-Laguna, Y.,
440
Torres, C., 2014. Faecal Escherichia coli isolates from healthy dogs harbour CTX-M-15
441
and CMY-2 ß-lactamases. Vet. J. doi: 10.1016/j.tvjl.2014.12.026.
Ac
439
442
Scheutz, F., Teel, L.D., Beutin, L., Piérard, D., Buvens, G., Karch, H., Mellmann, A.,
443
Caprioli, A., Tozzoli, R., Morabito, S., Strockbine, N.A., Melton-Celsa, A.R., Sanchez, M.,
444
Persson, S., O'Brien, A.D., 2012. Multicenter evaluation of a sequence-based protocol for
445
subtyping Shiga toxins and standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951-
446
2963.
447 448
Serna, A. 4th, Boedeker, E.C., 2008. Pathogenesis and treatment of Shiga toxin-producing Escherichia coli infections. Curr. Opin. Gastroenterol. 24 (1), 38-47. 14 Page 14 of 26
449
Sin, M.A., Takla, A., Flieger, A., Prager, R., Fruth, A., Tietze, E., Fink, E., Korte, J., Schink,
450
S., Höhle, M., Eckmanns, T., 2013. Carrier prevalence, secondary household transmission,
451
and long-term shedding in 2 districts during the Escherichia coli O104:H4 outbreak in
452
Germany, 2011. J. Infect. Dis. 207(3), 432-438. Srimanote, P., Paton, A.W., Paton, J.C., 2002. Characterization of novel type IV pilus locus
454
encoded on the large plasmid of locus of enterocyte effacement-negative Shiga-toxigenic
455
Escherichia coli strains that are virulent for humans. Infect. Immun. 70 (6), 3094-3100.
456
Toma, C., Martínez Espinosa, E., Song, T., Miliwebsky, E., Chinen, I., Iyoda, S., Iwanaga,
457
M., Rivas, M. 2004. Distribution of putative adhesins in different seropathotypes of Shiga
458
toxin-producing Escherichia coli. J. Clin. Microbiol. 42(11):4937-46.
us
cr
ip t
453
Werber, D., Fruth, A., Buchholz, U., Prager, R., Kramer, M.H., Ammon, A., Tschäpe, H.,
460
2003. Strong association between shiga toxin-producing Escherichia coli O157 and
461
virulence genes stx2 and eae as possible explanation for predominance of serogroup O157
462
in patients with haemolytic uraemic syndrome. Eur. J. Clin. Microbiol. Infect. Dis. 22(12),
463
726-730.
M
an
459
Ac
ce pt
ed
464
15 Page 15 of 26
Table 1. Gene names and abbreviations used Descriptions
Reference
stx (stx1a,1c,1d,2a,2b,2c,2d,2e,2f,g)
shiga toxin gene(s)
Scheutz et al. (2012)
aggR, aatA, agg3A, agg3C, aggA, aggC, hda
genes associated with aggregative adhesion
Boisen et al. (2008)
aaiC
Type six secretion substrate gene
Dudley et al. (2006)
estIa
heat-stable enterotoxin gene
Nada et al. (2013)
eaeA
Intimin gene
Schmidt et al. (1993)
fliC
flagellin gene
Kuwajima et al. (1986)
saa
autoagglutinating adhesin gene
eibG
immunoglobulin-binding protein gene Lu et al. (2006)
sfpA
major subunit of Sfp fimbriae gene
Friedrich et al. (2004)
sub
subtilase cytotoxin gene
Paton et al. (2004)
toxB
virulence gene involved in adhesion located on pO157 plasmid
Toma et al. (2004)
O-antigen
surface antigen
Ørskov and Ørskov (1984)
H-antigen
flagellar antigen
nm
non motile
-
nt
not typeable
-
sf/nsf 466 467 468
cr
Paton et al. (2001)
us
an
M
ed
ce pt
PT
ip t
Genes / Abbreviations
Ørskov and Ørskov (1984)
phage type
-
sorbitol fermenting/ non-sorbitol fermenting
-
Ac
465
16 Page 16 of 26
Table 2. Most frequently observed combinations of O and H antigens among human STEC isolates between 1997 and 2013 (NRC collection)
M
an
us
cr
11/40 4/8/10/14/21/25/28/49 21 6/12/31 7/nt nm/21/32 21 nm/11/18 nm/4/7/21 8 6/11/21/28 nm/8 nm/12 nm/2/10 25/28/34 nm/8/10/31 nm/1/7 nm/21/24/28/35 45
ip t
other H antigens
ce pt
5 8 26 55 76 80 91 103 104 111 113 115 118 128 145 146 156 157 174 177
H-antigen most frequently observed nm nm/19 nm/11 nm/7 nm/19 2 nm/14 2 unknown nm nm/4 10 16 2 nm 21/28 25 nm/7 2/8 nm/11
ed
O-antigen
Ac
469 470
17 Page 17 of 26
ip t cr
stx1
stx2
eaeA
ehxA
O103:H2 O103:H2
+/1a +/1a
+/2a +/2d
+ +
+ +
O104:H4
-
+/2a
-
-
O104:H4
-
+/2a
-
-
O109:Hnt O111:Hnm O113:H21 O128:H2 O132:H2
+/1a +/1c +/1a
+/2a +/2a +/2a +/2b -
+ + +
O145:Hnm
-
+/2a
O156:H25
-
O157:Hnm
-
O157:H7 O157:H7 O177:H11 O181:H49 O26:H11
+/1a +/1a +/1a
Selected accessory virulence genes*
d
M
an
Serovar
us
Table 3. HUS-associated serovars and their main virulence gene profile among human STEC isolates between 1997 and 2013 (NRC collection), * Selected accessory virulence genes were determined in certain strains
aggR, aatA, agg3A, agg3C aggR, aatA, aggA, aggC
Related strains in HUSEC Collection HUSEC 7: O103:H2 HUSEC 7: O103:H2 HUSEC 41: O104:H4 HUSEC 41: O104:H4
saa, sub
HUSEC 12: O111:Hnm HUSEC 26: O113:H21 HUSEC 28: O128:H2
+
+
toxB
HUSEC 22: O145:Hnm, HUSEC 36: O145:Hnm
+ /2a
+
+
+/2c
+
+
sfpA
HUSEC 6: O157:Hnm, HUSEC 4: O157:Hnm
+/2a +/2a+2c +/2a +/2a+2c+2d +/2a
+ + + +
+ + + + +
toxB toxB toxB saa, sub,
Ac
ce
pt e
+ + + -
HUSEC 3: O157:H7
HUSEC 17: O26:H11 18 Page 18 of 26
ip t stx2
eaeA
ehxA
O26:H11
-
+/2a
+
+
O55:H7
-
+/2a
+
-
O80:H2 O91:H14 O91:H21 O98:Hnm
+/1a +/1a
+/2a +/2d+2c -
+ -
+ + + +
d
M
an
HUSEC 18: O26:H11
pt e
+/2a +/2b +/2a
+ +
eibG saa aatA
HUSEC 5: O55:H7
HUSEC 34: O91:H21 HUSEC 30: O98:Hnm
saa
ce
+
Related strains in HUSEC Collection
Ac
Ont:Hnt Orauh:Hnm +/1a Orauh:Hnm -
cr
stx1
us
Serovar
Selected accessory virulence genes*
19 Page 19 of 26
ip t cr
Table 4. Observed hybrid pathovars among human STEC isolates from 1997 to 2013 and associated virulence gene pattern (NRC strain collection) Virulence pattern
Serovar
EHEC/EAEC
stx2a , aggR, aatA, aggA, aggC
O104:H4
EHEC/EAEC
stx2a , aggR, aatA, agg3A, agg3C
O104:H4
EHEC/EAEC
stx2a , aggR, aatA, hdaA, hdaC
O59:Hnm
Prager et al. (2014)
STEC/EAEC
stx2b , aatA
Orough:Hnm
Prager et al. (2014)
STEC/ETEC
stx2e , estIa
O100:Hnm, Ont:H21, Ont:Hnm
This study
STEC/ETEC
stx2g , estIa
O2:H28, O15:H16, O136:Hnm, O175:H28, Ont:H16, Ont:Hnm
This study and Prager et al. (2011)
References Robert Koch Institute (2011) Mellmann et al. (2008)
Ac
ce
pt e
d
M
an
us
Pathovar combinations
20 Page 20 of 26
ip t cr
Year Epidemiological cluster / suspected source of infection
an
Serovar and shigatoxin-type (phage type) O146:H21 stx2
us
Table 5. Selected outbreak events in Germany, data of NRC in cooperation with State Public Health Institutes and Federal Institute for Risk Assessment, BfR D/BD
HUS
Death Reference
4
0
0
NRC data
7
0
0
NRC data
23
2
0
Kirchner et al. (2013)
private party / unknown
O103:H2 stx1
2008
kindergarden / unknown
O157:H7 (PT 14) stx2
2008
tent camp / consumption of raw milk
O55:H7 stx1
2008
kindergarden / unknown
7
0
0
NRC data
O76:H19 stx1
2009
private party / consumption of mutton
3
0
0
NRC data
sf O157:Hnm (PT 88) stx2
2009
school and kindergarden / human-to-human-transmission
2
7
1
Nielsen et al. (2009)
O26:H11 stx2
2010
regional cluster /consumption of contaminated soft cheese
0
3
0
NRC data
O55:H7 stx2 O113:Hnt stx2
d
pt e
ce
Ac
O104:H4 stx2
M
2008
2011
outbreak / fenugreek seeds
2987
855
53
Frank et al. (2011)
2012
family cluster / unknown
4
1
0
NRC data
2013
family cluster / visit of a petting zoo
2
1
1
NRC data
21 Page 21 of 26
Figure Legends Figure 1. Major O serogroup distribution of human STEC isolates between 1997 and 2013 (NRC strain collection) Figure 2. Occurrence of different stx gene subtypes for most common STEC serovars between 1997 and 2013 (NRC strain collection), *O26:H21 and O91:H21 represent rarely
ip t
found O/H combinations, see also Table 2
Figure 3. Occurrence of different phage types among STEC O157 strains between 2001 and
cr
2009 (NRC strain collection), *possible bias due to outbreaks in following years: 2001,
Saxony-Anhalt, sf O157:Hnm , PT 88; 2006, Northern Germany, sf O157:Hnm, PT 88; 2007,
us
Saxony and Bavaria, O157:Hnm, PT 8; 2008, Lower Saxony, O157:H7, PT 14; 2009, Lower
an
Saxony and Hamburg, sf O157:Hnm, PT 88; 2009, Baden-Wuerttemberg, O157:Hnm, PT8
Figure 4. Percentage of antibiotic resistant strains among STEC isolates from human stool
M
samples and food sources (NRC strain collection); numbers for 2011 adjusted for outbreak
Ac
ce pt
ed
isolates, the strains from food sources contributed to a maximum of about 10% of total strains
22 Page 22 of 26
i cr us 739
892
840
M an
367
946
777
735
582
365
337
390
352
2352
619
466
ce pt
ed
278
Ac
n = 24
Page 23 of 26
86
120
133
128
137
118
81
65
53
50
61
29
100
56
40
28
39
47
62
59
32
38
32
145
107
M an
us
69
cr
i
O157:Hnm/H7
n= 3
37
O26:Hnm/H11/H21*
45
103
125
76
87
27
75
91
112
106
73
84
72
47
Ac
ce pt
ed
n= 9
n= 1
O91:H14/Hnm/H21* 20
79
85
76
36
42
69
Page 24 of 26
i cr us M an 128
137
ed
133
118
81
65
53
50
61
Ac
ce pt
n=
Page 25 of 26
i cr us 416
816
911
M an
278
885
964
837
783
601
381
319
392
335
819
552
438
Ac
ce pt
ed
n = 24
Page 26 of 26
S1438-4221(15)00083-1 http://dx.doi.org/doi:10.1016/j.ijmm.2015.08.020 IJMM 50982
To appear in:
Please cite this article as: Fruth, A., Prager, R., Tietze, E., Rabsch, W., Flieger, A.,Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing human disease in Germany: diversity, prevalence, and outbreaks, International Journal of Medical Microbiology (2015), http://dx.doi.org/10.1016/j.ijmm.2015.08.020 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
MINI-REVIEW
2 3
Molecular epidemiological view on Shiga toxin-producing Escherichia coli causing
4
human disease in Germany: diversity, prevalence, and outbreaks
5
Angelika Fruth, Rita Prager, Erhard Tietze, Wolfgang Rabsch, and Antje Flieger
ip t
6 7
Division of Enteropathogenic Bacteria and Legionella, National Reference Centre for
9
Salmonella and other Enteric Bacterial Pathogens, Robert Koch Institute, Wernigerode Branch, Burgstr. 37, 38855 Wernigerode, Germany
an
11
us
10
cr
8
12
M
13 14
ed
15
ce pt
16 17 18
Ac
19 20
Corresponding author:
21
Antje Flieger
22
Phone: +49 (0)30 18754 2522
23
Fax: +49 (0)30 18754 4207
24
e-mail: [email protected]
25 26 27 1 Page 1 of 26
Summary
29
Infections by intestinal pathogenic Escherichia coli (E. coli) are among those causing a high
30
mortality and morbidity due to diarrheal disease and post infection sequelae worldwide. Since
31
introduction of the Infection Protection Act in Germany 2001, these pathogens rank third
32
among bacterial infections of the gastrointestinal tract. As a major pathovar Shiga toxin-
33
producing E. coli (STEC) which include enterohemorrhagic E. coli (EHEC) play a leading
34
role in occurrence of sporadic cases and disease outbreaks. An outstanding example is the
35
large outbreak in spring 2011 caused by EHEC/EAEC O104:H4. To monitor and trace back
36
STEC infections, national surveillance programs have been implemented including activities
37
of the German National Reference Centre for Salmonella and other Enteric Bacterial
38
Pathogens (NRC). This review highlights advances in our understanding of STEC in the last
39
20 years of STEC surveillance by the NRC. Here important characteristics of STEC strains
40
from human infections and outbreaks in Germany between 1997 and 2013 are summarized.
an
us
cr
ip t
28
41
Key words
43
STEC, EHEC, HUS, serovar, phage type, diversity, outbreak investigations, antibiotic
44
resistance
ed
M
42
45
Introduction
47
The German National Reference Centre for Salmonella and other Enteric Bacterial Pathogens
48
(NRC) hosted by the Robert Koch Institute Division of Enteropathogenic Bacteria and
49
Legionella has been surveilling important bacterial diarrheal pathogens causing disease in
50
humans
51
http://www.rki.de/DE/Content/Infekt/NRZ/Salmonellen/salmo_node.html;jsessionid=1D05B5
52
E97ACCB8543EF4E41721E3E0E4.2_cid381). Our intention is to obtain an overview about
53
the strain types and their prevalence, virulence, antibiotic resistance, and persistence. Further,
54
changes in the pathogen types are recognized which may be indicative of new emerging or
55
disappearing strain variants and may give a hint on outbreaks. Outbreak detection is
56
performed in cooperation with the department of infection epidemiology of the Robert Koch
57
Institute and other partners, such as the Consultant Laboratory for Hemolytic Uremic
58
Syndrome (HUS) at the University of Münster, the Public Health Authorities of the German
59
Federal States and the Federal Institute of Risk Assessment (BfR). Upon strain comparison
60
and by means of specific strain properties also infection sources can be traced together with
ce pt
46
1993
(Homepage:
Ac
since
2 Page 2 of 26
61
other public health, food or veterinary authorities. The NRC further is involved in the
62
development and improvement of diagnostic procedures.
63
Within the 20 years of the work of the NRC, more than 12,000 Shiga toxin-producing E. coli
65
(STEC) isolates were collected and thoroughly analyzed yielding in a comprehensively
66
studied strain collection. The term STEC is used throughout this article for strains of both,
67
typical enterohemorrhagic E. coli (EHEC) serovars which are highly associated with sequelae
68
like hemorrhagic colitis (HC) and HUS, involving thrombocytopenia, haemolytic anemia, and
69
acute renal failure, and many other serovars which are only rarely associated with severe
70
human disease. With respect to intestinal pathogenic E. coli microbiology, pathogenicity,
71
clinical issues, and evolution, the reader is referred to a recently published comprehensive
72
review (Croxen et al., 2013) and therefore, we here will concentrate on EHEC surveillance at
73
NRC. Within the period of STEC surveillance, the strains analyzed at NRC represent about
74
60-80% of the reported cases in Germany ranging from mild or bloody diarrhea up to HUS.
75
To the purpose of strain characterization, bacterial isolates are mainly sent from public health
76
authorities and in some cases from primary diagnostic laboratories to the NRC for subtyping.
77
At the NRC, a variety of classical phenotypic as well as DNA-based methods are used.
M
an
us
cr
ip t
64
ed
78
In the case of STEC, strain isolation is currently rarely performed in primary diagnostic
80
laboratories. The procedure is very labour intensive and in most cases the determination of the
81
presence of the Shiga toxin or its genes from stool samples or enrichment cultures is sufficient
82
for therapeutic decisions and case reporting according to the German Infection Protection Act.
83
Therefore, isolation of the pathogen and further subtyping is usually performed in specialized
84
or reference laboratories. Phenotypic and DNA-based subtyping is performed on STEC
85
isolates for E. coli pathovar determination and for clone differentiation for epidemiological
86
purposes. A broad spectrum of established mostly PCR-based methods for the detection of
87
virulence determinants, such as adhesin or toxin genes and their variants, is employed for
88
pathovar characterization (see table 1; Flieger et. al., 2013). These PCR-based detections of
89
gene regions are important methods for epidemiological analysis of STEC however cannot
90
evaluate functionality of gene expression or of the corresponding proteins since non-
91
functional degenerated gene mutants have been described (Abu-Ali et al., 2010). Strain
92
subtyping by means of serotyping either via classical or genoserotyping and by
93
biochemotyping, such as sorbitol fermentation, is carried out for all STEC isolates sent to the
94
NRC. Further, international standardized pulsed-field gel electrophoresis (PFGE) after DNA
Ac
ce pt
79
3 Page 3 of 26
95
macrorestriction is still the method of choice for reliable identification of outbreak-associated
96
STEC clones. Serotyping, PCR-based virulence gene profiling, and PFGE analysis were
97
important tools applied during the 2011 EHEC/EAEC O104:H4 outbreak for characterization
98
of the outbreak strain and for tracing of infection clusters (Bielaszewska et al., 2011; Frank et
99
al., 2011; Sin et al., 2013; Robert Koch Institute, 2011). In this review, we present a summary of important characteristics of STEC isolated from
101
human infections in Germany between 1997 and 2013. Used abbreviations and gene names
102
with respect to STEC subtyping are explained and referenced in Table 1.
cr
103
ip t
100
STEC O serogroups causing human disease
105
Serotyping is an important method for STEC subtyping. At the NRC classical serotyping is
106
performed as described by Ørskov and Ørskov (Ørskov and Ørskov, 1984). So far 186
107
O serogroups (O1-O192 except O31, O47, O67, O72, O93 and O94) and 53 associated
108
H forms (H1-56 except H13, H22, H50) are currently described (see WHO Center for
109
Klebsiella and Escherichia coli, Copenhagen, Denmark for further information). Within the
110
here reported period, 139 different O serogroups were detected among STEC strains. Most
111
commonly found O serogroups among STEC were O157, O91, O26 and O103 which
112
represent about 40 to 50% of the STEC strains analyzed at NRC starting from 1999 (Fig. 1).
113
Excluding 2011, this shows that there are no major changes in O serogroup distribution and
114
consequently in their significance in human disease. It is also obvious that the percentage of
115
O157 strains decreased from about 25% in 1999 to about 10% in 2013. This may reflect the
116
growing awareness of other STEC serogroups in diagnostics rather than a significant decrease
117
in O157 infections. Accordingly, the fraction of O91 strains increased from about 5% in 1999
118
to about 15% in 2012 and 2013. The large outbreak due to EHEC/EAEC O104:H4 in 2011
119
highly influenced O serogroup distribution by representing about 50% of the strains analyzed
120
in that year. Most frequent H type combinations of the most important O serogroups were
121
O157:Hnm/H7, O91:Hnm/H14, O26:Hnm/H11, and O103:H2 (Table 2). It is further
122
interesting to note that about 10% of the strains yearly are either untypable or cannot be
123
attributed to a known O serogroup which shows a limitation of classical serotyping compared
124
to DNA-based methods addressing coding genes for expression of serotype-related traits.
Ac
ce pt
ed
M
an
us
104
125 126
In summary, O26, O91, O103, and O157 were the most common O serogroups occurring
127
among STEC from 1997 to 2013 and only slight changes in their distribution can be observed. 4 Page 4 of 26
128
Except in 2011, the large outbreak due to EHEC/EAEC O104:H4 dominated O serogroup
129
percentages (Fig. 1).
130
STEC and association with stx genotypes, with HUS, and occurrence of hybrid
132
pathovars
133
In the early 1990s, the historical approach to assign the pathovar STEC to strains of specific
134
serovars needed certain revisions. Because of the dynamic nature of the Shiga toxin-
135
harboring phages and the limited chance to detect these pathogens later during infection,
136
different assays based on detection of Shiga toxin and Shiga toxin genes improved the capture
137
of diverse STEC. By application of such specific virulence gene PCR assays, the NRC
138
collected a great variety of STEC serovars and virulence gene profile since 1999 (Fig. 1 and
139
Table 3). In addition to known correlations of most frequent serovars with stx1 (such as O26 or
140
O91 strains) or stx2 (such as O157 strains), changes in stx gene type distribution were
141
observed. For example, stx2 gene positive isolates of serovar O26:H11/Hnm emerged with an
142
increasing tendency to harbour different stx2 genes. Only 7% of STEC O26:H11/Hnm
143
harboured stx2 in 1999 but 59% in 2013 (Fig. 2; Bielaszewska et al., 2013).
M
an
us
cr
ip t
131
ed
144
Due to the awareness of severe outcome of STEC infection, the HUSEC strain collection was
146
established and published by Mellmann et al. (2008). It represents the most common strains
147
isolated from patients suffering from HUS. Various HUS-associated isolates belonging to
148
additional serovars were identified by the NRC (Table 3). Among these are isolates of
149
serovars O177:H11, O156:H25, O109:Hnt, and O80:H2 carrying stx2a and eaeA genes.
150
Further, this includes the first description of an O181:H49 isolate with a variety of virulence-
151
associated genes (stx1a, stx2a, stx2c, stx2d, sub and saa).
Ac
152
ce pt
145
153
Genomic plasticity and horizontal gene transfer enable emergence of STEC strains with
154
additionally acquired virulence properties of intestinal or extraintestinal pathogenic E. coli
155
(Müller et al., 2007; Orth et al., 2007). Such hybrid forms of STEC and other pathovars were
156
targeted by PCR analysis of additional virulence genes at the NRC. Accordingly, two novel
157
STEC strains of serovars O59:Hnm and Orough:Hnm were identified harbouring virulence
158
genes involved in aggregative adhesion characteristic for EAEC (Table4; Prager et al., 2014).
159
Further, novel strains carrying stx2g and estIa were described which represent a hybrid variant
160
of STEC and ETEC (Table 4; Prager et al., 2011). 5 Page 5 of 26
Phage typing of STEC
162
Typing of E.coli by lysis pattern caused by specific bacteriophages is described for O157 and
163
other serovars (Milch, 1978). Additionally for the serovar O157, a Canadian system was
164
developed by Ahmed et al. (1987) and extended by Khakhria et al. (1990). This scheme was
165
used widely as a tool for epidemiological investigations in Canada, the United States,
166
England, Wales, and Scotland. From 2001 till 2009, this Canadian scheme which describes 88
167
distinct phage types by application of sixteen different phages (Duck et al., 1997) was used at
168
the NRC. 826 tested strains of serovar O157 from Germany collected between 2001 and 2009
169
yielded in 30 different phage types. During this time, the dominant phage types were PT8
170
(30%), PT88 (20%), PT14 (15%), PT 2 (12 %), PT 4 (5%), and 18% belonged to 25 different
171
PTs (Fig. 3). The distribution of stx1 and stx2 phages among the different phage types was
172
interesting. For example, the strains of sorbitol-fermenting (sf) O157:Hnm PT88 carried in
173
100% and of O157 PT2 in 96% the stx2 phage only.
an
us
cr
ip t
161
174
Phage typing was also helpful for epidemiological analysis in outbreak investigations. In
176
2001, an outbreak in the German Federal State Saxony-Anhalt caused by sf O157/Hnm STEC
177
of PT88 was observed in a kindergarden. Among 12 patients several HUS cases occurred.
178
STEC sf O157:Hnm of PT 88 have been observed in several outbreaks in Germany, yet their
179
reservoir and routes of transmission remain unknown (Alpers et al., 2009). In 2008 an
180
outbreak by STEC O157 PT14 was traced back to raw milk and children who visited a farm in
181
Lower Saxony were affected (Kirchner et al., 2013).
ed
ce pt
182
M
175
With the introduction of new molecular subtyping methods, the future of STEC phage typing
184
is under debate. But for a scientific view on STEC prophages which carry additional genes
185
(e.g. virulence genes like stx1, stx2) or for the application of the next generation of
186
bacteriophage therapy (Lu and Koeris, 2011), the available sets of phages may be helpful.
187
Ac
183
188
Antibiotic resistance among STEC
189
Antibiotic treatment of EHEC infections is problematic and controversial and is not regularly
190
recommended especially during the acute diarrhoeal phase but is for certain cases, such as
191
pathogen-shedding without clinical symptoms, under debate (Serna et al., 2008, Nitschke et
192
al., 2012). Nevertheless, at the NRC, the susceptibility of STEC towards 16 antimicrobial
193
substances is determined by a broth microdilution method routinely to obtain the antibiogram 6 Page 6 of 26
of the pathogens in order to use it as an epidemiological marker. The breakpoints for the
195
classification “resistant” are mainly according to DIN and were not adapted to any of the
196
more recent updates by CLSI or EUCAST, respectively, in order to maintain consistency in
197
our epidemiological surveillance of antibiotic resistance development. The substances tested
198
include tetracycline (4 mg/l), ampicillin (8), mezlocillin (16), the combination of mezlocillin
199
with sulbactam (16), chloramphenicol (8), the combination of trimethoprim and
200
sulfamethoxazole – cotrimoxazole – (16), streptomycin (16), kanamycin (16), gentamicin (4),
201
amikacin (16), nalidixic acid (16) ciprofloxacin (2), cefotiam (4), cefoxitin (16), cefotaxime
202
(8) and ceftazidime (16). Since 1997, among all of the STEC sent to the NRC, a rather
203
constant yearly proportion of around 70% is tested susceptible to all of these substances (Fig.
204
4). Between 15 and 20% are resistant to one or two antibiotics, and about 10% of STEC
205
isolates are resistant to three and more - rarely up to 13 - of the substances tested. The most
206
frequently observed resistance phenotypes are resistance towards streptomycin, tetracycline
207
and ampicillin (about 15 to 20% of all resistant STEC) followed by resistance towards
208
chloramphenicol (about 10% of all resistant STEC). Resistance of STEC towards
209
fluorchinolones varied over the years below one percent of all resistant STEC reaching 0.7 %
210
cumulated for the years 2012 and 2013. A clearly increasing trend is observed for resistance
211
towards cotrimoxazole. While between 1997 and 2004 about 7% of all resistant STEC
212
showed this resistance phenotype, the respective proportion of cotrimoxazole-resistant STEC
213
rose to over 12% during the period between 2005 and 2011 and reached about 30% cumulated
214
for the years 2012 and 2013. Another trend that seems to be looming is a rising resistance
215
quotient among STEC for cephalosporins as mediated by extended spectrum beta lactamases
216
(ESBL). While up to the year 2010 the proportion of STEC with an ESBL phenotype
217
(reduced susceptibility or resistance towards cefotaxime and/ or ceftazidime) remained below
218
1%, this proportion rose up to 4% of all resistant STEC cumulated for the years 2012 and
219
2013. A very prominent example of an STEC with an ESBL phenotype is the O104:H4 strain
220
which caused the large EHEC outbreak in Germany in 2011. This strain harbored the gene for
221
the ESBL enzyme CTX-M-15 on a plasmid (Robert Koch Institute, 2011). Plasmid-encoded
222
CTX-M-15 is widely distributed in many E. coli strains of different pathovars (Rocha-Garcia
223
et al., 2015, Ewers et al., 2014). Other STEC isolates with an ESBL phenotype were from
224
sporadic cases and were serotyped as O80:H2, O179:H8, O80:Hnt and O16:Hnt. Thus,
225
although selection pressure towards resistant STEC by therapeutic use of antibiotics in
226
clinical cases can be assumed to be low, the surveillance of antibiograms at the NRC reveals
227
the continuous prevalence of antibiotic resistant STEC in Germany over the past 16 years.
Ac
ce pt
ed
M
an
us
cr
ip t
194
7 Page 7 of 26
228
One must assume, therefore, that these pathogens are subject to ecological processes creating
229
and selecting antibiotic resistant strains in their reservoirs.
230
STEC surveillance and outbreak investigations
232
The NRC uses for the surveillance, outbreak investigations and identification of new
233
emerging STEC types different molecular approaches such as virulence gene PCR assays,
234
PFGE, MLST, MLVA and plasmid profile analysis. Because there is no single combination of
235
virulence genes that defines STEC as potential EHEC, a broad range of virulence genes is
236
investigated by PCR. After the 2011 EHEC O104:H4 outbreak, the panel of the routinely
237
tested stx and eaeA genes was extended by the EAEC genes aatA and aggR.
us
cr
ip t
231
238
For surveillance and to predict risks of STEC infections, the knowledge of the stx subtypes
240
and colonization factors and adhesins is essential. The NRC uses the stx subtyping strategies
241
and nomenclature as described by Scheutz et al. (2012). The NRC as national public health
242
reference laboratory also participates every year in the EQA ring trials coordinated by the
243
ECDC, Stockholm, covering the detection of stx1, stx2, eaeA, aggR and aaiC genes and
244
subtyping of stx1 into three known subtypes (stx1a, stx1c, stx1d) and stx2 genes into seven
245
known subtypes (stx2a, stx2b, stx2c, stx2d, stx2e, stx2f, stx2g). It is well known that some stx
246
subtypes, especially variant stx2a, in combination with eaeA or with aatA/aggR are often
247
associated with severe clinical outcomes (e.g. sf O157:Hnm or O104:H4; Tables 3 and 5;
248
Werber et al., 2003). But other virulence gene combinations may also be associated with
249
severe disease in humans, including HUS (see Table 3). Therefore predicting the emergence
250
of „new‟ highly pathogenic STEC types based only on the presence of the stx genes, or by
251
focusing on a restricted panel of serogroups alone is not possible (Orth et al., 2006; Prager et
252
al., 2005; Fasel et al., 2014). The STEC virulence plasmids (pO157, psfO157 or pO113) also
253
play an important role in severe clinical outcomes. Especially eaeA-negative isolates
254
possessing plasmids of type pO113 (Srimanote et al., 2002) in combination with variant stx2d
255
were more often described in correlation with HUS (see Tables 3 and 5). Therefore the NRC
256
uses the plasmid profile analysis as an important tool for surveillance and outbreak
257
investigations, not only for analysis of STEC. Thus it could be shown by the NRC very early
258
that the EAEC virulence plasmids and the plasmid profiles in the outbreak strain EHEC
259
O104:H4 and HUSEC O104:H4 were different (EHEC O104:H4
260
Erregers sowie Hinweise und Hilfestellungen des RKI zur Diagnostik des gegenwärtig
Ac
ce pt
ed
M
an
239
- Eigenschaften des
8 Page 8 of 26
261
zirkulierenden
262
http://www.rki.de/DE/Content/InfAZ/E/EHEC/EHEC_O104/EHEC_O104_Diagnostik.pdf?_
263
_blob=publicationFile; Mellmann et al., 2011).
264
DNA macrorestriction and pulsed-field gel electrophoresis (PFGE) for the subtyping of STEC
265
is widely used to compare strains for epidemiological purposes. The method currently
266
represents the gold standard for the investigation of outbreaks and source-tracing. The NRC
267
uses the standardized PFGE protocol developed by the CDC Atlanta which is part of EQA’s
268
coordinated by the ECDC Stockholm. The NRC as national public health reference laboratory
269
has been successfully participating every year in these EQA’s covering the gel quality and gel
270
analysis by means of BioNumerics (BioNumerics, Applied Math, Ghent, Belgium). This
271
standardized method is a basic requirement to contribute to the international molecular typing
272
pilot project of the ECDC with the aim to detect molecular clusters. The NRC takes an active
273
part in this pilot project. Serovar-specific PFGE library databases based on the restriction with
274
XbaI were established at the NRC for the most important STEC serovars showing severe
275
clinical outcomes and involvement in outbreaks, such as (sf) O157:H7/Hnm, O26:H11/Hnm,
276
O145:Hnm, O103:H2/Hnm. This is an essential precondition for the detection of outbreaks,
277
molecular clusters, or new clones. Altogether, by PFGE 500 (sf and non-sorbitol-fermenting
278
(nsf)) O157:H7/Hnm isolates were subtyped into 265 different patterns, 250 O26:H11/Hnm
279
isolates into 142 patterns, 120 O103:H2/Hnm isolates into 53 patterns and 48 O145:Hnm
280
isolates into 30 patterns. In case of an outbreak, PFGE analysis is done with a second enzyme
281
(BlnI) as recommended by CDC (PulseNet International, 2013). During outbreak
282
investigations or for cluster detections, the NRC works together with the department of
283
Infectious Disease Epidemiology of the Robert Koch Institute and the National Reference
284
Laboratory for E. coli at the German Federal Institute for Risk Assessment (BfR), especially
285
for source tracing. The universal applicability of PFGE for subtyping of isolates other than the
286
most common serovars was shown (Prager et al., 2009, 2011; Bielaszewska et al., 2011;
287
Frank et al., 2011). In addition to PFGE, the NRC also uses for the investigations of genetic
288
relationships MLVA and MLST analysis and in the future the application of whole genome
289
sequencing and the verification of the method in outbreak analysis will be compassed (Prager
290
et al., 2009, 2011).
Available:
Ac
ce pt
ed
M
an
us
cr
ip t
Ausbruchstammes,
291 292
Conclusions
293
We here presented an overview on the characteristics of the STEC strains isolated from
294
human disease in Germany during the last 17 years. STEC comprise strains with a high 9 Page 9 of 26
potential to cause severe symptoms in humans and outbreaks in association with consumption
296
of contaminated food. At the beginning of that period, classical serotyping with a focus on
297
O157 strain detection was a major method for STEC risk assessment. Today, analysis of a
298
wide array of virulence or phylogenetic markers by DNA-based methods together with
299
serotyping allows a more complete view on the pathogen’s virulence profile and phylogeny.
300
In the future, a more complete picture will arise from the implementation of transcriptomics,
301
proteomics, and whole genome sequencing. Especially the latter is currently in progress at
302
NRC and has the potential to revolutionize STEC surveillance and outbreak analysis.
303
However, the established methods will certainly remain in use for the near future, both to
304
serve the diagnostic and epidemiological needs of the public health system and to complement
305
the genome-based approaches in the process of their evaluation.
us
cr
ip t
295
306
Acknowledgements
308
We thank the team members of the Robert Koch Institute Division of Enteropathogenic
309
Bacteria and Legionella for their contributions to subtyping of bacteria and preparation of the
310
manuscript. We further are indebted to all the diagnostic laboratories providing strains for
311
epidemiological analysis.
ed
312
M
an
307
References
314
Abu-Ali, G.S., Ouellette, L.M., Henderson, S.T., Lacher, D.W., Riordan, J.T., Whittam, T.S.,
315
Manning, S.D., 2010. Increased adherence and expression of virulence genes in a lineage
316
of Escherichia coli O157:H7 commonly associated with human infections. PLoS One 5
317
(4).e10167.
319
Ahmed, R., Bopp, C., Borczyk, A., Kasatiya, S., 1987. Phage-typing scheme for E. coli 0157:
Ac
318
ce pt
313
H7. J. Infect. Dis. 155, 806-809.
320
Alpers, K., Werber, D., Frank, C., Koch, J., Friedrich, A.W., Karch, H., an der Heiden, M.,
321
Prager, R., Fruth, A., Bielaszewska, M., Morlock, G., Heissenhuber, A., Diedler, A.,
322
Gerber, A., Ammon, A., 2009. Sorbitol-fermenting enterohaemorrhagic Escherichia coli
323
O157:H- causes another outbreak of haemolytic uraemic syndrome in children. Epidemiol.
324
Infect. 137, 389-395.
325
Bielaszewska, M., Mellmann, A., Zhang, W., Köck, R., Fruth, A., Bauwens, A., Peters, G.,
326
Karch, H., 2011. Characterisation of the Escherichia coli strain associated with an outbreak 10 Page 10 of 26
327
of haemolytic uraemic syndrome in Germany, 2011: a microbiological study. Lancet
328
Infect. Dis. 11(9):671-676. Bielaszewska, M., Mellmann, A., Bletz, S., Zhang, W., Köck, R., Kossow, A., Prager, R.,
330
Fruth, A., Orth-Höller, D., Marejková, M., Morabito, S., Caprioli, A., Piérard, D., Smith,
331
G., Jenkins, C., Curová, K., Karch, H., 2013. Enterohemorrhagic Escherichia coli
332
O26:H11/H-: a new virulent clone emerges in Europe. Clin. Infect. Dis. 2013 56(10), 1373-
333
1381.
ip t
329
Boisen, N., Struve, C., Scheutz, F., Krogfelt, K.A., Nataro, J.P., 2008. New adhesin of
335
enteroaggregative Escherichia coli related to the Afa/Dr/AAF family. Infect. Immun.
336
76(7), 3281-3292.
339
us
338
Croxen, M.A., Law, R.J., Scholz, R., Keeney, K.M., Wlodarska, M., Finley, B.B., 2013. Recent advances in understanding enteric pathogenic Escherichia coli. CMR 26, 822-880.
an
337
cr
334
Duck, D., Demczuk, W., McKinnon, J., Mulvey, M., Clarke, C., Woodward, D., Khakhria, R., Ahmed, R., 1997. Escherichia coli O157: H7 phage typing. Natl . Lab. Enteric Pathogens,
341
Lab. Centre Dis. Control, Ottawa, Ontario, Canada. 4/1/97.
M
340
Dudley, E.G., Thomson, N.R., Parkhill, J., Morin, N.P., Nataro, J.P., 2006. Proteomic and
343
microarray characterization of the AggR regulon identifies a pheU pathogenicity island in
344
enteroaggregative Escherichia coli. Mol. Microbiol. 61(5), 1267-1282.
ed
342
Ewers, C., Stamm, I., Pfeifer, Y., Wieler, L.H., Kopp, P.A., SchØnning, K., Prenger-
346
Berninghoff, E., Scheufen, S., Stolle, I., Günther, S., Bethe, A., 2014. Clonal spread of
347
highly successful ST15-CTX-M-15 Klebsiella pneumoniae in companion animals and
348
horses. J. Antimicrobial. Chemother. 69 (10), 2676-2680.
ce pt
345
Fasel, D., Mellmann, A., Cernela, N., Hächler, H., Fruth, A., Khanna, N., Egli, A., Beckmann,
350
C., Hirsch, H.H., Goldenberger, D., Stephan, R., 2014. Hemolytic uremic syndrome in a
351
65-Year-old male linked to a very unusual type of stx2e- and eae-harboring O51:H49 shiga
352
toxin-producing Escherichia coli. J. Clin. Microbiol. 52 (4), 1301-1303.
Ac
349
353
Flieger, A., Mielke, M., Tietze, E., 2013. Role of pathogen surveillance and subtyping in
354
outbreak recognition of food-borne bacterial infections. Microbiological perspective -
355
Aims,
356
Gesundheitsforschung Gesundheitsschutz. 56(1), 42-46.
methods
and
prospects
of
pathogen
subtyping.
Bundesgesundheitsblatt
11 Page 11 of 26
357
Frank, C., Werber, D., Cramer, J.P., Askar, M., Faber, M., an der Heiden, M., Bernard, H.,
358
Fruth, A., Prager, R., Spode, A., Wadl, M., Zoufaly, A., Jordan, S., Kemper, M.J., Follin,
359
P., Müller, L., King, L.A., Rosner, B., Buchholz, U., Stark, K., Krause, G., HUS
360
Investigation Team, 2011. Epidemic profile of Shiga-toxin-producing Escherichia coli
361
O104:H4 outbreak in Germany. New Engl. J. Med. 365(19), 1771-1780. Frost, J.A., Smith, H.R, Willshaw, G.A., Scotland, S.M., Gross, R.J., Rowe, B., 1989.
363
Phagetyping of Verocytotoxin (VT) producing E. coli O157 isolated in the United
364
Kingdom. Epidemiol. Infect. 103, 73- 81.
cr
366
Khakhria, R., Duck, D., Lior, H., 1990. Extended phage typing scheme for E. coli O157:H7. Epidemiol. Infect. 105, 511-520.
us
365
ip t
362
Kirchner, M., Dildei, C., Runge, M., Brix, A., Claussen, K., Weiss, U., Fruth, A., Prager, R.,
368
Mellmann, A., Beutin, L., Miko, A., Wichmann-Schauer, H., Pulz, M., Dreesman, J., 2013.
369
Outbreak of non-sorbitol-fermenting shiga toxin-producing E. coli O157:H7 infections
370
among school children associated with raw milk consumption in Germany. Archiv für
371
Lebensmittelhygiene 64, 68-74.
374 375
M
the hag gene encoding flagellin of Escherichia coli. J. Bacteriol. 168, 1479–1483.
ed
373
Kuwajima, G., Asaka, J.-I., Fujiwara, T., Node, K., Kondo. E.,1986. Nucleotide sequence of
Lu, T.K., Koeris, M.S., 2011. The next generation of bacteriophage therapy. Curr. Opin. Microbiol. 14, 524-31.
ce pt
372
an
367
Lu, Y., Iyoda, S., Satou, H., Satou, H., Itoh, K., Saitoh, T., Watanabe, H., 2006. A new
377
immunoglobulin-binding protein, EibG, is responsible for the chain-like adhesion
378
phenotype of locus of enterocyte effacement-negative, shiga toxin-producing Escherichia
379
coli. Infect. Immun. 74(10), 5747-5755.
380 381
Ac
376
Milch, H., 1978. Phage Typing of Escherichia coli. In: Bergan, T., Norris, J.R. (Eds.), Microbial Methods, CHAPTER III, Academic Press London, pp. 87-156.
382
Mellmann, A., Bielaszewska, M., Koeck, R., Friedrich, A.W., Fruth, A., Middendorf B.,
383
Harmsen, D., Schmidt, M.A., Karch , H., 2008. Serotypes and Phylogenetic Analysis of
384
the Collection of Hemolytic Uremic Syndrome-Associated enterohemorrhagic Escherichia
385
coli (HUSEC). EID 14(8), 1287-1290.
386
Mellmann, A., Harmsen, D., Cummings, C.A., Zentz, E.B., Leopold, S.R., Rico, A., Prior, K.,
387
Szczepanowski, R., Ji, Y., Zhang, W., McLaughlin, S.F., Henkhaus, J.K., Leopold, B., 12 Page 12 of 26
388
Bielaszewska, M., Prager, R., Brzoska, P.M., Moore, R., Guenther, S., Rothberg, J.M.,
389
Karch, H., 2011. Prospective genomic characterization of the German enterohemorrhagic
390
Escherichia coli O104:H4 outbreak by rapid next generation sequencing technology. PloS
391
One 6(7), e22751. Müller, D., Greune, L., Heusipp, G., Karch, H., Fruth, A., Tschäpe, H., Schmidt, M.A., 2007.
393
Identification of unconventional intestinal pathogenic Escherichia coli isolates expressing
394
intermediate virulence factor profiles by using a novel single-step multiplex PCR. Appl.
395
Environ. Microbiol., 73(10), 3380-3390.
cr
ip t
392
Nada, R.A., Armstrong, A., Shaheen, H.I., Nakhla, I., Sanders, J.W., Riddle, M.S., Young, S.,
397
Sebeny, P., 2013. Phenotypic and genotypic characterization of enterotoxigenic
398
Escherichia coli isolated from U.S. military personnel participating in Operation Bright
399
Star, Egypt, from 2005 to 2009. Diagn. Microbiol. Infect. Dis. 76(3), 272-277.
an
us
396
Nielsen, S., Frank, C., Fruth, A., Spode, A., Prager, R., Graff, A., Plenge-Bönig, A. Loos, S.,
401
Lütgehetmann, M., Kemper, M.J., Müller-Wiefel, D.E., Werber, D., 2011. Desperately
402
seeking diarrhoea: outbreak of haemolytic uraemic syndrome caused by emerging sorbitol-
403
fermenting shiga toxin-producing Escherichia coli O157:H-, Germany, 2009. Zoonoses
404
Public Health. 58(8), 567-572.
ed
M
400
Nitschke, M., Sayk, F., Härtel, C., Roseland, R.T., Hauswaldt, S., Steinhoff, J., Fellermann,
406
K., Derad, I., Wellhöner, P., Büning, J., Tiemer, B., Katalinic, A., Rupp, J., Lehnert, H.,
407
Solbach, W., Knobloch, J.K., 2012. Association between azithromycin therapy and
408
duration
409
enteroaggregative Escherichia coli O104:H4. JAMA 307 (10), 1046-1052.
of
ce pt
405
bacterial
shedding
among
patients
with
Shiga
toxin-producing
Ørskov, F., Ørskov, I.,1984. Serotyping of Escherichia coli. Methods Microbiol. 14, 43-112.
411
Orth, D., Grif, K., Khan, A.B., Naim, A., Dierich, M.P., Würzner, R., 2007. The Shiga toxin
412
genotype rather than the amount of Shiga toxin or the cytotxicity of Shige toxin in vitro
413
correlates with the apparence of the hemolytic syndrome. Diagn. Microbiol. Infect. Dis. 59,
414
235-242.
Ac
410
415
Orth, D., Grif, K., Fisher, I., Fruth, A., Tschäpe, H., Scheutz, F., Dierich, M.P., Würzner, R.,
416
2006. Emerging Shiga toxin-producing Escherichia coli serotypes in Europe: O100:H- and
417
O127:H40. Curr. Microbiol. 53(5), 428-429.
13 Page 13 of 26
418
Prager, R., Annemüller, S., Tschäpe, H., 2005. Diversity of virulence patterns among shiga
419
toxin-producing Escherichia coli from human clinical cases – need for more detailed
420
diagnostics. Int. J. Med. Microbiol. 295, 29-38. Prager, R., Fruth, A., Siewert, U., Strutz, U., Tschäpe, H., 2009. Escherichia coli encoding
422
Shiga toxin 2f as an emerging human pathogen. Int. J. Med. Microbiol. 299(5), 343-353.
423
Prager, R., Fruth, A., Busch, U., Tietze, E., 2011. Comparative analysis of virulence genes,
424
genetic diversity, and phylogeny of Shiga toxin 2g and heat-stable enterotoxin STIa
425
encoding Escherichia coli isolates from humans, animals, and environmental sources. Int.
426
J. Med. Microbiol. 301(3), 181-191.
cr
ip t
421
Prager, R., Lang, C., Aurass, P., Fruth, A., Tietze, E., Flieger, A., 2014. Two Novel
428
EHEC/EAEC Hybrid Strains Isolated from Human Infections. PLoS One. 9(4), e95379.
429
PulseNet International, 2013. Standard operating procedure for PulseNet PFGE of
430
Escherichia coli O157:H7, Escherichia coli non-O157 (STEC), Salmonella serotypes,
431
Shigella sonnei and Shigella flexneri. PNL05. PulseNet International, Atlanta, GA.
432
http://www.pulsenetinternational.org/assets/PulseNet/uploads/pfge/PNL05_Ec-Sal-
433
ShigPFGEprotocol.pdf.
an
M
Robert Koch Institute, 2011. Characteristics of the pathogen and information and assistance
ed
434
us
427
435
by
436
http://www.rki.de/EN/Home/EHECO104.pdf?__blob=publicationFile.
438
RKI
in
diagnosis
of
the
currently
circulating
outbreak
strain.
ce pt
437
the
Robert Koch-Institut, 2011. Bakteriologische Untersuchungen im Rahmen des Ausbruchs mit E. coli O104:H4. Epid. Bull. 35, 325-329. Rocha-Garcia, R.C., Cortés-Cortés, G., Lozano-Zarain, P., Belo, F., Martínez-Laguna, Y.,
440
Torres, C., 2014. Faecal Escherichia coli isolates from healthy dogs harbour CTX-M-15
441
and CMY-2 ß-lactamases. Vet. J. doi: 10.1016/j.tvjl.2014.12.026.
Ac
439
442
Scheutz, F., Teel, L.D., Beutin, L., Piérard, D., Buvens, G., Karch, H., Mellmann, A.,
443
Caprioli, A., Tozzoli, R., Morabito, S., Strockbine, N.A., Melton-Celsa, A.R., Sanchez, M.,
444
Persson, S., O'Brien, A.D., 2012. Multicenter evaluation of a sequence-based protocol for
445
subtyping Shiga toxins and standardizing Stx nomenclature. J. Clin. Microbiol. 50, 2951-
446
2963.
447 448
Serna, A. 4th, Boedeker, E.C., 2008. Pathogenesis and treatment of Shiga toxin-producing Escherichia coli infections. Curr. Opin. Gastroenterol. 24 (1), 38-47. 14 Page 14 of 26
449
Sin, M.A., Takla, A., Flieger, A., Prager, R., Fruth, A., Tietze, E., Fink, E., Korte, J., Schink,
450
S., Höhle, M., Eckmanns, T., 2013. Carrier prevalence, secondary household transmission,
451
and long-term shedding in 2 districts during the Escherichia coli O104:H4 outbreak in
452
Germany, 2011. J. Infect. Dis. 207(3), 432-438. Srimanote, P., Paton, A.W., Paton, J.C., 2002. Characterization of novel type IV pilus locus
454
encoded on the large plasmid of locus of enterocyte effacement-negative Shiga-toxigenic
455
Escherichia coli strains that are virulent for humans. Infect. Immun. 70 (6), 3094-3100.
456
Toma, C., Martínez Espinosa, E., Song, T., Miliwebsky, E., Chinen, I., Iyoda, S., Iwanaga,
457
M., Rivas, M. 2004. Distribution of putative adhesins in different seropathotypes of Shiga
458
toxin-producing Escherichia coli. J. Clin. Microbiol. 42(11):4937-46.
us
cr
ip t
453
Werber, D., Fruth, A., Buchholz, U., Prager, R., Kramer, M.H., Ammon, A., Tschäpe, H.,
460
2003. Strong association between shiga toxin-producing Escherichia coli O157 and
461
virulence genes stx2 and eae as possible explanation for predominance of serogroup O157
462
in patients with haemolytic uraemic syndrome. Eur. J. Clin. Microbiol. Infect. Dis. 22(12),
463
726-730.
M
an
459
Ac
ce pt
ed
464
15 Page 15 of 26
Table 1. Gene names and abbreviations used Descriptions
Reference
stx (stx1a,1c,1d,2a,2b,2c,2d,2e,2f,g)
shiga toxin gene(s)
Scheutz et al. (2012)
aggR, aatA, agg3A, agg3C, aggA, aggC, hda
genes associated with aggregative adhesion
Boisen et al. (2008)
aaiC
Type six secretion substrate gene
Dudley et al. (2006)
estIa
heat-stable enterotoxin gene
Nada et al. (2013)
eaeA
Intimin gene
Schmidt et al. (1993)
fliC
flagellin gene
Kuwajima et al. (1986)
saa
autoagglutinating adhesin gene
eibG
immunoglobulin-binding protein gene Lu et al. (2006)
sfpA
major subunit of Sfp fimbriae gene
Friedrich et al. (2004)
sub
subtilase cytotoxin gene
Paton et al. (2004)
toxB
virulence gene involved in adhesion located on pO157 plasmid
Toma et al. (2004)
O-antigen
surface antigen
Ørskov and Ørskov (1984)
H-antigen
flagellar antigen
nm
non motile
-
nt
not typeable
-
sf/nsf 466 467 468
cr
Paton et al. (2001)
us
an
M
ed
ce pt
PT
ip t
Genes / Abbreviations
Ørskov and Ørskov (1984)
phage type
-
sorbitol fermenting/ non-sorbitol fermenting
-
Ac
465
16 Page 16 of 26
Table 2. Most frequently observed combinations of O and H antigens among human STEC isolates between 1997 and 2013 (NRC collection)
M
an
us
cr
11/40 4/8/10/14/21/25/28/49 21 6/12/31 7/nt nm/21/32 21 nm/11/18 nm/4/7/21 8 6/11/21/28 nm/8 nm/12 nm/2/10 25/28/34 nm/8/10/31 nm/1/7 nm/21/24/28/35 45
ip t
other H antigens
ce pt
5 8 26 55 76 80 91 103 104 111 113 115 118 128 145 146 156 157 174 177
H-antigen most frequently observed nm nm/19 nm/11 nm/7 nm/19 2 nm/14 2 unknown nm nm/4 10 16 2 nm 21/28 25 nm/7 2/8 nm/11
ed
O-antigen
Ac
469 470
17 Page 17 of 26
ip t cr
stx1
stx2
eaeA
ehxA
O103:H2 O103:H2
+/1a +/1a
+/2a +/2d
+ +
+ +
O104:H4
-
+/2a
-
-
O104:H4
-
+/2a
-
-
O109:Hnt O111:Hnm O113:H21 O128:H2 O132:H2
+/1a +/1c +/1a
+/2a +/2a +/2a +/2b -
+ + +
O145:Hnm
-
+/2a
O156:H25
-
O157:Hnm
-
O157:H7 O157:H7 O177:H11 O181:H49 O26:H11
+/1a +/1a +/1a
Selected accessory virulence genes*
d
M
an
Serovar
us
Table 3. HUS-associated serovars and their main virulence gene profile among human STEC isolates between 1997 and 2013 (NRC collection), * Selected accessory virulence genes were determined in certain strains
aggR, aatA, agg3A, agg3C aggR, aatA, aggA, aggC
Related strains in HUSEC Collection HUSEC 7: O103:H2 HUSEC 7: O103:H2 HUSEC 41: O104:H4 HUSEC 41: O104:H4
saa, sub
HUSEC 12: O111:Hnm HUSEC 26: O113:H21 HUSEC 28: O128:H2
+
+
toxB
HUSEC 22: O145:Hnm, HUSEC 36: O145:Hnm
+ /2a
+
+
+/2c
+
+
sfpA
HUSEC 6: O157:Hnm, HUSEC 4: O157:Hnm
+/2a +/2a+2c +/2a +/2a+2c+2d +/2a
+ + + +
+ + + + +
toxB toxB toxB saa, sub,
Ac
ce
pt e
+ + + -
HUSEC 3: O157:H7
HUSEC 17: O26:H11 18 Page 18 of 26
ip t stx2
eaeA
ehxA
O26:H11
-
+/2a
+
+
O55:H7
-
+/2a
+
-
O80:H2 O91:H14 O91:H21 O98:Hnm
+/1a +/1a
+/2a +/2d+2c -
+ -
+ + + +
d
M
an
HUSEC 18: O26:H11
pt e
+/2a +/2b +/2a
+ +
eibG saa aatA
HUSEC 5: O55:H7
HUSEC 34: O91:H21 HUSEC 30: O98:Hnm
saa
ce
+
Related strains in HUSEC Collection
Ac
Ont:Hnt Orauh:Hnm +/1a Orauh:Hnm -
cr
stx1
us
Serovar
Selected accessory virulence genes*
19 Page 19 of 26
ip t cr
Table 4. Observed hybrid pathovars among human STEC isolates from 1997 to 2013 and associated virulence gene pattern (NRC strain collection) Virulence pattern
Serovar
EHEC/EAEC
stx2a , aggR, aatA, aggA, aggC
O104:H4
EHEC/EAEC
stx2a , aggR, aatA, agg3A, agg3C
O104:H4
EHEC/EAEC
stx2a , aggR, aatA, hdaA, hdaC
O59:Hnm
Prager et al. (2014)
STEC/EAEC
stx2b , aatA
Orough:Hnm
Prager et al. (2014)
STEC/ETEC
stx2e , estIa
O100:Hnm, Ont:H21, Ont:Hnm
This study
STEC/ETEC
stx2g , estIa
O2:H28, O15:H16, O136:Hnm, O175:H28, Ont:H16, Ont:Hnm
This study and Prager et al. (2011)
References Robert Koch Institute (2011) Mellmann et al. (2008)
Ac
ce
pt e
d
M
an
us
Pathovar combinations
20 Page 20 of 26
ip t cr
Year Epidemiological cluster / suspected source of infection
an
Serovar and shigatoxin-type (phage type) O146:H21 stx2
us
Table 5. Selected outbreak events in Germany, data of NRC in cooperation with State Public Health Institutes and Federal Institute for Risk Assessment, BfR D/BD
HUS
Death Reference
4
0
0
NRC data
7
0
0
NRC data
23
2
0
Kirchner et al. (2013)
private party / unknown
O103:H2 stx1
2008
kindergarden / unknown
O157:H7 (PT 14) stx2
2008
tent camp / consumption of raw milk
O55:H7 stx1
2008
kindergarden / unknown
7
0
0
NRC data
O76:H19 stx1
2009
private party / consumption of mutton
3
0
0
NRC data
sf O157:Hnm (PT 88) stx2
2009
school and kindergarden / human-to-human-transmission
2
7
1
Nielsen et al. (2009)
O26:H11 stx2
2010
regional cluster /consumption of contaminated soft cheese
0
3
0
NRC data
O55:H7 stx2 O113:Hnt stx2
d
pt e
ce
Ac
O104:H4 stx2
M
2008
2011
outbreak / fenugreek seeds
2987
855
53
Frank et al. (2011)
2012
family cluster / unknown
4
1
0
NRC data
2013
family cluster / visit of a petting zoo
2
1
1
NRC data
21 Page 21 of 26
Figure Legends Figure 1. Major O serogroup distribution of human STEC isolates between 1997 and 2013 (NRC strain collection) Figure 2. Occurrence of different stx gene subtypes for most common STEC serovars between 1997 and 2013 (NRC strain collection), *O26:H21 and O91:H21 represent rarely
ip t
found O/H combinations, see also Table 2
Figure 3. Occurrence of different phage types among STEC O157 strains between 2001 and
cr
2009 (NRC strain collection), *possible bias due to outbreaks in following years: 2001,
Saxony-Anhalt, sf O157:Hnm , PT 88; 2006, Northern Germany, sf O157:Hnm, PT 88; 2007,
us
Saxony and Bavaria, O157:Hnm, PT 8; 2008, Lower Saxony, O157:H7, PT 14; 2009, Lower
an
Saxony and Hamburg, sf O157:Hnm, PT 88; 2009, Baden-Wuerttemberg, O157:Hnm, PT8
Figure 4. Percentage of antibiotic resistant strains among STEC isolates from human stool
M
samples and food sources (NRC strain collection); numbers for 2011 adjusted for outbreak
Ac
ce pt
ed
isolates, the strains from food sources contributed to a maximum of about 10% of total strains
22 Page 22 of 26
i cr us 739
892
840
M an
367
946
777
735
582
365
337
390
352
2352
619
466
ce pt
ed
278
Ac
n = 24
Page 23 of 26
86
120
133
128
137
118
81
65
53
50
61
29
100
56
40
28
39
47
62
59
32
38
32
145
107
M an
us
69
cr
i
O157:Hnm/H7
n= 3
37
O26:Hnm/H11/H21*
45
103
125
76
87
27
75
91
112
106
73
84
72
47
Ac
ce pt
ed
n= 9
n= 1
O91:H14/Hnm/H21* 20
79
85
76
36
42
69
Page 24 of 26
i cr us M an 128
137
ed
133
118
81
65
53
50
61
Ac
ce pt
n=
Page 25 of 26
i cr us 416
816
911
M an
278
885
964
837
783
601
381
319
392
335
819
552
438
Ac
ce pt
ed
n = 24
Page 26 of 26