FEMS Microbiology Ecology Advance Access published December 9, 2015

Clonal spread and interspecies transmission of clinically relevant ESBL-producing Escherichia coli of ST410 – another successful pandemic clone?

Katharina Schaufler1#, Torsten Semmler2, Lothar H. Wieler3, Michael Wöhrmann1, Ramani Baddam1,4, Niyaz Ahmed4, Kerstin Müller5, Axel Kola6, Angelika Fruth7, Christa Ewers8, Sebastian Guenther1

1

Institute of Microbiology and Epizootics, Veterinary Faculty, Freie Universität Berlin, Germany

2

NG 1 - Microbial Genomics, Robert Koch Institute, Berlin, Germany

3

Robert Koch Institute, Berlin, Germany

4

Pathogen Biology Laboratory, Department of Biotechnology and Bioinformatics, School of Life Sciences, University of Hyderabad, India

5

Clinic of Small Animals, Veterinary Faculty, Freie Universität Berlin, Germany

6

Institute of Hygiene and Environmental Medicine, Charité Universitätsklinikum, Berlin, Germany

7

Department for Infectious Diseases, Division of Bacterial Infections and National Reference, Centre for Salmonella and other Bacterial Enteric Pathogens,

Robert Koch Institute, Wernigerode, Germany 8

Institute of Hygiene and Infectious Diseases of Animals, Veterinary Faculty, Justus-Liebig-Universität Giessen, Germany

# Corresponding author: Katharina Schaufler Centre for Infection Medicine, Institute of Microbiology and Epizootics

Robert von Ostertag-Str. 7-13, 14163 Berlin, Germany Phone: ++49 30 83851898, Fax: ++49 30 838451851 E-mail: [email protected]

Running title Transmission of an ESBL-producing E. coli ST410 clone

Keywords ESBL-producing Escherichia coli, ST410, Next-generation sequencing, “One Health” approach, SNP-analysis, Clone

Abstract Clinically relevant ESBL-producing multi-resistant Escherichia coli have been on the rise for years. Initially restricted to mostly a clinical context, recent findings prove their prevalence in extra-clinical settings independent of the original occurrence of antimicrobial resistance in the environment. To get further insights into the complex ecology of potentially clinically relevant ESBL-producing E. coli, 24 isolates from wild birds in Berlin, Germany, and forty ESBL-producing human clinical E. coli isolates were comparatively analyzed. Isolates of ST410 occurred in both sample groups (six). In addition, three ESBL-producing E. coli isolates of ST410 from environmental dog feces and one clinical dog isolate were included. All ten isolates were clonally analyzed showing almost identical macrorestriction patterns. They were chosen for whole genome sequencing revealing that the whole genome content of these ten E. coli isolates showed a very high genetic similarity, differing by low numbers of single nucleotide polymorphisms only. This study gives initial evidence for a recent interspecies transmission of a new successful clone of ST410 E. coli between wildlife, humans, companion animals, and the environment. The results underline the zoonotic potential of clinically relevant multi-resistant bacteria found in the environment as well as the mandatory nature of the “One Health” approach.

Introduction Extra-intestinal pathogenic Escherichia (E.) coli (ExPEC) bacteria cause a wide range of diseases including meningitis, bacteremia, urinary tract, and soft-tissue infections (Russo & Johnson, 2000; Naseer & Sundsfjord, 2011). They frequently produce extended-spectrum beta-lactamase (ESBL)-enzymes, which hydrolyze different kinds of beta-lactams, including newer agents like third-generation cephalosporins and monobactams (Knothe et al., 1983). Limitations in antimicrobial therapies result from a multi-drug resistant phenotype of these bacteria (Beceiro et al., 2013), which are often additionally resistant against fluoroquinolones, tetracyclines, aminoglycosides, and other classes of antimicrobials (Pitout, 2010; Pitout, 2012). The occurrence of antimicrobial resistance in the environment is an ancient phenomenon (D'Costa et al., 2011). Human respectively animal pathogens and environmental bacteria share large parts of their resistome (Forsberg et al., 2012; Finley et al., 2013; Bengtsson-Palme & Larsson, 2015). One assumes that the acquisition of environmental resistance genes by clinically relevant pathogens happened through horizontal gene transfer influenced by a broad application of antimicrobials (Davies & Davies, 2010). While transmission of clinical multi-resistant isolates and resistance genes from clinical settings back into the environment seems likely (Pruden et al., 2013), it is still unclear where they initially originated. In the clinical context, ESBL-genes were first documented in 1983 and a rapid, worldwide dissemination across human clinical settings has followed (Bernard et al., 1992; Radice et al., 2002). While ESBL-producing E. coli have contributed significantly to human community-acquired infections, they are no longer restricted solely to the clinics (Dubois et al., 2009). One of the first clinical ESBL-producing isolates in animals was detected in a dog from Spain in 1998 (Teshager et al., 2000) and ESBL-enzymes have been found in livestock shortly after in multiple countries (Brinas et al., 2003; Smet et al., 2010). The broad appearance of potentially clinically relevant ESBL-producing isolates in wildlife as well as environmental samples underlines this success (Dolejska et al., 2009; Allen et al., 2010; Guenther et al., 2011; Sousa et al., 2011).

There are several possible explanations why ESBL-producing bacteria occur in various settings and hosts. Numerous ESBL-genes within the most common ESBL-families blaCTX-M, blaSHV, blaTEM and blaOXA are encoded on plasmids, which are rapidly transmitted among bacterial species (Carattoli, 2011). A broad dissemination is likely once these plasmids or plasmid-carrying bacteria cross from the clinical to extra-clinical setting and vice versa as they seem not necessarily restricted to settings with high antimicrobial pressures (Guenther et al., 2011). Beyond the plasmid-driven spread of ESBL, studies have shown that the clonal dissemination of certain clinically relevant phylogenetic lineages, distinguished by their sequence types (STs), occur frequently among ESBLproducing E. coli (Ewers et al., 2012). Interestingly, examining this phylogenetic distribution has revealed that several STs, including ST410, ST131, ST648, ST617 and others, are shared by humans, domestic animals and the environment (Guenther et al., 2011). It remains unclear, however, where these STs, which combine pathogenicity and resistance, first appeared, and whether repeating transmission cycles occur or if ESBL-producing strains once transmitted to a host develop independently. To investigate the hypothesis of an occurrence of clonal or closely related antimicrobial resistant clinically relevant E. coli isolates from different hosts and settings and to contribute to the understanding of transmission scenarios, we analyzed ESBL-producing E. coli isolates sampled between 2008 and 2013 from one geographic area. We used molecular characterization methods including next-generation sequencing (NGS) applications, which allowed the highest possible resolution of the phylogenetic relationship as far as the detection of similar clones.

Methods Origin of samples Avian bacterial isolates originated from a screening for ESBL-producing E. coli during the entry examination of rescued wild birds in the small animal clinic of the Freie Universität Berlin, Germany. Overall 320 cloacal swabs (supplement Fig. 1) from 40 different avian species were examined from 2011 until 2014. The

birds were found sick or injured in Berlin and Brandenburg, Germany. Each animal was sampled once and the cloacal swab (MASTASWAB containing Amies medium, Mast Diagnostics Reinfeld, Germany) was directly delivered to our microbiological laboratory (Institute of Microbiology and Epizootics, Berlin, Germany). Forty ESBL-producing human clinical E. coli isolates were selected randomly from in total 60 ESBL-producing human clinical isolates (supplement Fig. 1) from blood cultures collected in a university medical center in Berlin, from 2008 until 2010 (Leistner et al., 2014). In addition, three ESBL-producing E. coli of ST410 from environmental dog feces and one isolate from a clinical dog sample characterized in a previous study (Schaufler et al., 2015) were included (supplement Fig. 1). Isolation of E. coli (avian cloacal samples) Cloacal swabs were streaked on CHROMagarTM Orientation plates (with and without 4 µg/ml cefotaxime; Mast Diagnostica, Reinfeld, Germany) and incubated overnight to isolate E. coli and to preselect for cefotaxime-resistant E. coli. One colony per sample with coliform appearance on CHROMagarTM was further processed and bacterial species identification was carried out using the automated VITEK®2 system (BioMérieux, Germany). Phenotypic characterization of cefotaxime-resistant E. coli (avian ESBL-producing E. coli) E. coli isolates showing growth on CHROMagarTM containing cefotaxime were confirmed as ESBL-producers using the phenotypic confirmatory test for ESBLproduction, performed and interpreted according to CLSI guideline M31-A3 (CLSI, 2008). Genotypic characterization of ESBL-producing E. coli (avian ESBL-producing E. coli) The genomic composition of the confirmed ESBL-producing E. coli was characterized using established PCR protocols with amplification and subsequent sequencing for common E. coli resistance genes including blaCTX-M, blaSHV, blaoxa, blaTEM, tetA/B/C, sul1, sul2, sul3, strA/B, aadA1-like, and aacC4 (Maguire et al., 2001; Robicsek et al., 2006; Jouini et al., 2007; Park et al., 2007; Rodriguez et al., 2009; Ewers et al., 2010). Mutations in gyrA and parC genes were determined by PCR and sequence analysis (Ewers et al., 2014). Plasmid-mediated AmpC beta-lactamase genes were analyzed using a multiplex PCR (PerezPerez & Hanson, 2002).

Multi-locus sequence typing (MLST) and phylogenetic grouping by structure analysis (all ESBL-producing E. coli) MLST determination was carried out according to the protocol developed by Wirth et al. (Wirth et al., 2006). Gene amplification and sequencing were performed by using primers specified on the E. coli MLST website (University of Warwick website, http://mlst.warwick.ac.uk/mlst/). Sequences were analyzed by the software package Ridom SeqSphere 0.9.39 (Ridom website, http://www3.ridom.de/seqsphere) and STs were computed automatically. The phylogenetic group of the E. coli isolates was determined using the software structure 2.3.4 based on the concatenated sequences of the seven housekeeping genes used for MLST (University of Chicago website http://pritch.bsd.uchicago.edu/structure.html). Macrorestriction analysis and pulsed-field gel electrophoresis (PFGE) (ten ST410 ESBL-producing E. coli) Based on MLST results ten selected ESBL-producing isolates (supplement Fig. 1) were examined with macrorestriction analysis using a CHEF DRIII System (BioRad, Munich, Germany) with the following conditions: 14 °C, 20.5 hours running time, 200 volts, 6 volts/cm2 and angle 120 ° (Ewers et al., 2004). PFGE profiles generated by restriction of chromosomal DNA with XbaI after digestion of the bacterial cells with proteinase K were compared digitally using BioNumerics software (Version 7.1, Applied Maths, Belgium). Cluster analysis of Dice similarity indices based on the unweighted pair group method with arithmetic mean (UPGMA) was applied to generate dendrograms depicting the relationships among PFGE profiles. Isolates with a similarity of > 85 % were assigned to one clonal group (Tenover et al., 1995). O and H antigen typing (ten ST410 ESBL-producing E. coli) Determination of O and H antigens of the ten selected ST410 isolates (IMT28707, IMT28764, IMT30467, IMT31351, IMT31352, IMT31359, IMT31487, IMT33180, IMT33181 and IMT33204) was carried out by conventional serotyping at the Robert Koch Institute; the National Reference Centre for Salmonella and other enteric pathogens.

Plasmid profile analysis (ten ST410 ESBL-producing E. coli) Determination of plasmid-carriage of the ten selected ST410 isolates was analyzed using established plasmid profile analysis protocols (Green & Sambrook, 2012; Schaufler et al., 2013). After alkaline lysis of the bacteria plasmid DNA was separated from chromosomal DNA by classic differential precipitation. Plasmid preparation was visualized using a 0.4 % agarose gel with TAE buffer, running for 16 hours at 25 volt.

Whole genome sequencing and analysis (ten ST410 ESBL-producing E. coli) Ten selected isolates (supplement Fig. 1, Fig. 1A) were whole genome sequenced using MiSeq Illumina 300 bp paired-end sequencing and a coverage greater than 50 x was obtained (except for IMT30467 with a 27 x coverage). The sequence read data was first subjected to quality control using the NGS tool kit (Patel & Jain, 2012). Reads with a minimum of 70 % of bases having a phred score of greater than 20 were defined as high quality reads. De novo assembly of resulting high quality filtered reads into contiguous sequences (contigs) was administered using Velvet (Zerbino & Birney, 2008). The optimum genome assembly was chosen based on the following parameters: N50, maximum contig size, number of contigs and approximate size of the genome assembly. Contigs (> 500 bp) obtained at the optimum hash length were considered for further analysis. Assembled draft genomes of all ten isolates were then subjected to annotation using RAST (Aziz et al., 2008). The determination of the maximum common genome (MCG) alignment was done as described previously (von Mentzer et al., 2014). We therefore clustered the coding sequences based on the two parameters sequence similarity (minimum 70 %) and coverage (minimum 90 %) and defined those genes as MCG that were present in each of the ten genomes and which fulfilled the threshold parameters. In a next step the allelic variants (or single nucleotide polymorphisms [SNPs]) of these genes were extracted from all genomes by a BLAST based approach, aligned individually for each gene and then concatenated, which resulted in an alignment of overall 3,577 Mbp for these 10 strains. This alignment was used to generate a phylogenetic tree (Fig. 1B) with RAxML 8.1 (Stamatakis, 2014). The percentage of divergence between two strains was calculated based on the number of allelic variants compared to the overall number of base pairs in their MCG. Due to the highest quality of assembly (N50 = 111 kbp; overall 138 contigs with a maximum contig size of 215 kbp and an approximate genome assembly size of 4.8 Mbp) among all ten isolates PLACNET was applied for isolate IMT31351. PLACNET is a method to identify and analyze plasmid sequences retrieved from whole genome sequence data by creating a network of plasmid and chromosomal contig interactions (Lanza et al., 2014). First, the original network was generated by PLACNET mainly utilizing the information obtained from the (a) genome assembly (length and coverage of contigs), (b) scaffold links identified from the SAM (Sequence Alignment/Map format) file, which was produced by mapping reads back to contigs, (c) BLAST analysis with the reference database including both plasmids and chromosomes (downloaded from NCBI FTP site [ftp://ftp.ncbi.nlm.nih.gov/genomes]), and (d) identification of plasmid-encoded relaxase (REL)/plasmid replication initiator protein (RIP) sequences. Second, the manual pruning of the original network mainly involved identification of hubs and their duplication based on links, coverage information and BLASTx analysis as described previously (Lanza et al., 2014). Third, results were visualized in Cytoscape (Shannon et al., 2003) to identify both the plasmid and chromosomal contents of IMT31351.

Comparison of the plasmid sequences of IMT31351 (retrieved via PLACNET) and the whole genome sequences of the other nine isolates was performed using BLAST Ring Image Generator (BRIG) (Alikhan et al., 2011). BRIG displays concentric circles to visualize the similarity of a central reference sequence (here: IMT31351) and compared other sequences with indicated sequence identities.

Results Twenty-four (7.5 %) of the 320 bacterial samples collected from wild birds were positive for ESBL-producing E. coli (supplement Fig. 1) with a dominance of blaCTX-M-1 (60 %) and blaCTX-M-15 (40 %). In addition, resistance genes like blaTEM-1, blaOXA-1, tetA/B, sul1, sul2, sul3, strA/strB and mutations in gyrA/parC appeared commonly (Tab. 1). In summary, the majority (14/24) of the tested ESBL-producing E. coli isolates were multi-resistant resulting in overall 4 % (14/320) multi-resistant isolates from wild birds. The highest ESBL-producing E. coli carriage rate in avian hosts was detected in mute swans (Cygnus olor) 26.5 % (34 animals tested) (Tab. 1). The isolation rates were equally distributed during the year (comparing winter and summer months) and throughout the whole sampling period. MLST of the avian ESBL-producing E. coli revealed 19 different STs: ST410 (3x), ST617 (3x), ST405 (2x), ST88, ST115, ST167, ST131, ST224, ST373, ST398, ST648, ST1167, ST1204, ST1304, ST1670, ST1730, ST1968, ST4306, and ST4307 (Tab. 1). The detection of ESBL-producing ST410 E. coli in the avian samples and its occurrence in human clinical samples as well as in dog feces isolates recently published by us (Schaufler et al., 2015) led to the decision to further focus the analysis on this ST by characterizing the relationship of these ten potentially clinically relevant ST410 isolates in depth. To elucidate their clonal relatedness, macrorestriction analysis was performed. Unexpectedly, these ten isolates showed almost identical macrorestriction patterns (similarity index > 94 % (Tenover et al., 1995)) (Fig. 1A). In addition, they all expressed identical O and H surface antigens, namely O20:H9. Plasmid profile analysis revealed the presence of similarly large putative ESBL-plasmids (> 100 kbp; with the exception of IMT28707 and IMT28764, which appeared slightly smaller). Only two isolates (IMT28764 and IMT33204) harbored additional smaller plasmids (supplement Fig. 2). For an even higher phylogenetic and clonal resolution, the whole genomes of all ten ST410 isolates were sequenced (supplement Fig. 1). The MCG for these isolates consists of 4186 orthologous genes and the resulting alignment consists of 3,582,101 sites with 1,839 phylogenetically informative SNP sites. Based on their whole genomes, all ten isolates were confirmed to be highly clonally related (phylogenetic tree [Fig. 1B]). In the phylogenetic tree, two main clades (Cluster I and Cluster II) were identified: the first one including isolates from humans (IMT33181, IMT33204 and IMT33180), wild birds (IMT28764 and IMT28707) and dog feces (IMT31352) and the other clade with the canine clinical isolate (IMT31487), two dog feces isolates (IMT31359 and IMT31351) and one wild bird

isolate (IMT30467). The number of SNPs ranged from 0 (between IMT31351 and IMT31487) to 1767 (between IMT28764 and IMT30467) (supplement Tab. 1). Overall, the average amount of SNPs was very low, within each clade always less than 100 and less than 1800 between the clades. The number of orthologous genes shared by all isolates as well as the clustering into two clades based on the amount of shared genes was confirmed by using OrthoMCL (Li et al., 2003) (data not shown). PLACNET analysis was performed to separate chromosomal from plasmid sequences. Figure 2 shows the PLACNET network for IMT31351 obtained after manual pruning. All contigs were assigned either to the chromosomal or plasmid network and those with hits to plasmid-encoded REL/RIP proteins are highlighted differently in pink and yellow (Fig. 2). In total, 25 contigs were assigned to the plasmid constellation, thereby constituting about 128 kbp of plasmid (minimum size as it is not closed). We could not resolve the network completely into disjoint components due to difficulties posed by plasmid repetitive sequences. BRIG circular visualization comparison of plasmid sequences of isolate IMT31351 (central reference in the image) with the whole genome sequences of the other nine isolates revealed identical plasmid sequences for IMT31351 and IMT31487 (both from dog feces). Similar plasmid sequences were obtained for the other isolates (Fig. 3). Missing parts are most likely explainable due to assembly errors of common plasmid repetitive elements.

Discussion Potentially clinically relevant ESBL-producing E. coli occur in both clinical as well as extra-clinical settings. Until now, it remains mostly unclear how closely related these isolates sampled from different environments are and what role transmission scenarios play, regardless if originating from either environmental or human/animal clinical infection sources. Recent literature discusses transmission possibilities of clinical multi-resistant bacteria mainly based on MLST and macrorestriction analyses (Bonnedahl et al., 2014; Jamborova et al., 2015) both limited in their informative value concerning clones. Techniques with higher resolution have rarely been applied but might provide more detailed information on the environmental dimension of antimicrobial resistant, potentially clinically relevant isolates. The ESBL-producing E. coli STs present in avian hosts from this study included ST131, ST410, ST617 and ST648 frequently occurring in human and veterinary clinical cases (Wang et al., 2009; Guenther et al., 2011; Guenther et al., 2012; Sherchan et al.) underlining the potential clinical relevance of ESBL-producing E. coli found in wildlife. In general, shared STs might give a first hint for the phylogenetic relatedness as well as transmission potential of certain multi-resistant bacteria between different hosts and settings. One cannot assume, however, that isolates within one ST belong to one clonal group. A ST occurring in all of the

different origins examined in this study happened to be ST410, a ST associated with ESBL-production previously reported in different hosts and settings (Guenther et al., 2012; Guenther et al., 2013; Ben Sallem et al.). Macrorestriction analysis determined the ST410’s high clonal relatedness, which is not always the case for strains within one single ST (Ewers et al., 2010). The clonality of isolates was underlined by similar sero- and plasmid profile types. The implemented NGS applications enabled a high resolution determination of the position of clonal isolates originating from different hosts and settings in the overall phylogenetic context of ESBL-producing ST410. The low number of SNPs, especially in each of the two clades and also between different hosts and settings, indicates no host specific association and the limited genetic diversity of the isolates. Detecting different phylogenetic clades within a ST is not surprising and was described in a recently published study for the ST131 clone by Petty et al. (Petty et al., 2014). Few numbers of SNPs were found in IMT28764 from a wild mute swan, compared to IMT33180, from human clinical origin (n=45; 8.6 SNPs/Mbp), representing 0,0013 % divergence. Also remarkable was the difference between the same wild mute swan isolate (IMT28764) and one collected from dog feces (IMT31352), only differing in 50 total SNPs (9.6 SNPs/Mbp; divergence=0,0014 %). Most strikingly, one human clinical isolate (IMT33204) had 24 SNPs (when compared to avian IMT28707, 4.6 SNPs/Mbp) and 29 SNPs (when compared to dog feces isolate IMT31352, 5.5 SNPs/Mbp). This very low number of SNPs per Mbp is only slightly higher than the one described for clonal EHEC strains during the German outbreak (Grad et al., 2012; de Been et al., 2014) (1.8 SNPs/Mbp), clearly pointing towards a recent interspecies transmission event in our study. When comparing human clinical and livestock isolates the authors of the same EHEC study obtained 1263 SNPs/Mbp, which should be noted in addition. They concluded that there is no evidence for a recent interspecies transmission of ESBL-producing clones from livestock to humans and plasmids are the driving force for cephalosporin-resistance dissemination. In contrary, our data show the clonal relatedness between human and animal clinical, environmental as well as wildlife ESBL-producing E. coli isolates. The similar plasmid sequences of IMT31351 compared to those of the other nine isolates as displayed with identity indices (50-100 %) in Fig. 3 indicate a low plasmid flux in this putative new clone of ST410, as opposed to the lineages of ST131, where plasmids were rather diverse (Lanza et al., 2014). Interestingly, the sample period for the ten ESBL-producing ST410 isolates covers several years (2009-2013), so why do almost genetically identical isolates exist in different hosts after this amount of time? It might be that this clinically relevant ESBL-producing ST410 clone goes constantly back and forth between different hosts and settings without the requirement to undergo excessive genetic changes. Isolation dates and number of SNPs (between human and avian as well as environmental dog isolates) might suggest the initial occurrence of this clinically relevant clone in the human clinical setting of Berlin, Germany, underlined by the expected number of SNPs (up to 40) to occur in between the sampling time points (de Been et al., 2014). But based on our data we cannot exclude the

opposite direction namely the first appearance of this clone outside the clinic. Anyway, whether these findings are due to a recent clonal spread indeed or a limited microevolution emphasized by the narrow geographic sampling region needs to be further investigated. Despite the sample origins of different hosts and settings, we found an ESBL-producing E. coli clone of ST410. This shows that potentially clinically relevant isolates are present in the environment even in the absence of clinical antimicrobial selection pressures. Also, the “resistance armory” of these isolates might be further equipped with novel resistance determinants, which they acquired from environmental bacteria, resulting in even more problematic strains affecting human and animal health (Walsh et al., 2011; Bengtsson-Palme & Larsson, 2015). This pilot study, although limited due to a restricted sample size, reinforces efforts to prove putative recent, maybe ongoing transmission scenarios of multiresistant bacteria between different hosts and habitats. In addition, our study underlines the theory of the spread of (potentially clinically relevant) clones of certain STs rather than only relying on plasmid-driven disseminations of multi-resistance. The detailed molecular and clonal characterization of multi-resistant ESBL-producing E. coli, which are of significant importance regarding their zoonotic potential, contributes to the “One Health” approach to integrate the human, animal and environmental health sectors.

Funding This work was supported by a doctoral stipend grant of the SONNENFELD-STIFTUNG to K.S and by a grant of the German Research Foundation entitled “Functional analysis of non-resistance genes of extended-spectrum beta-lactamases associated sequence types of Escherichia coli” [grant GU1283/3-1 to S.G. and EW116/2-1 to C.E.].

References

Alikhan NF, Petty NK, Ben Zakour NL, et al. BLAST Ring Image Generator (BRIG): simple prokaryote genome comparisons. Bmc Genomics 2011; 12: 402. Allen HK, Donato J, Wang HH, et al. Call of the wild: antibiotic resistance genes in natural environments. Nat Rev Microbiol 2010; 8: 251-259. Aziz RK, Bartels D, Best AA, et al. The RAST server: Rapid annotations using subsystems technology. Bmc Genomics 2008; 9: 75. Beceiro A, Tomas M & Bou G Antimicrobial resistance and virulence: a successful or deleterious association in the bacterial world? Clin Microbiol Rev 2013; 26: 185-230. Ben Sallem R, Ben Slama K, Estepa V, et al. Detection of CTX-M-15-producing Escherichia coli isolates of lineages ST410-A, ST617-A and ST354-D in faecal samples of hospitalized patients in a Mauritanian hospital. J Chemotherapy 2015; 27: 114-116. Bengtsson-Palme J & Larsson DGJ Antibiotic resistance genes in the environment: prioritizing risks. Nat Rev Microbiol 2015; 13: 396. Bernard H, Tancrede C, Livrelli V, et al. A novel plasmid-mediated extended-spectrum beta-lactamase not derived from TEM- or SHV-type enzymes. J Antimicrob Chemother 1992; 29: 590-592. Bonnedahl J, Hernandez J, Stedt J, et al. Extended-spectrum beta-lactamases in Escherichia coli and Klebsiella pneumoniae in Gulls, Alaska, USA. Emerg Infect Dis 2014; 20: 897-899. Brinas L, Moreno MA, Zarazaga M, et al. Detection of CMY-2, CTX-M-14, and SHV-12 beta -lactamases in Escherichia coli fecal-sample isolates from healthy chickens. Antimicrob Agents Chemother 2003; 47: 2056-2058. Carattoli A Plasmids in Gram negatives: Molecular typing of resistance plasmids. Int J Med Microbiol 2011; 301: 654-658. CLSI Performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals; approved standard - Third Edition (2008) p.^pp. CLSI, Wayne, PA, U.S.A. D'Costa VM, King CE, Kalan L, et al. Antibiotic resistance is ancient. Nature 2011; 477: 457-461. Davies J & Davies D Origins and Evolution of Antibiotic Resistance. Microbiol Mol Biol R 2010; 74: 417-433. de Been M, Lanza VF, de Toro M, et al. Dissemination of Cephalosporin Resistance Genes between Escherichia coli Strains from Farm Animals and Humans by Specific Plasmid Lineages. Plos Genet 2014; 10: e1004776. Dolejska M, Bierosova B, Kohoutova L, et al. Antibiotic-resistant Salmonella and Escherichia coli isolates with integrons and extended-spectrum beta-lactamases in surface water and sympatric black-headed gulls. J Appl Microbiol 2009; 106: 1941-1950. Dubois D, Prasadarao NV, Mittal R, et al. CTX-M beta-lactamase production and virulence of Escherichia coli K1. Emerg Infect Dis 2009; 15: 1988-1990. Ewers C, Janssen T, Kiessling S, et al. Molecular epidemiology of avian pathogenic Escherichia coli (APEC) isolated from colisepticemia in poultry. Vet Microbiol 2004; 104: 91-101.

Ewers C, Bethe A, Semmler T, et al. Extended-spectrum beta-lactamase-producing and AmpC-producing Escherichia coli from livestock and companion animals, and their putative impact on public health: a global perspective. Clin Microbiol Infect 2012; 18: 646-655. Ewers C, Grobbel M, Stamm I, et al. Emergence of human pandemic O25:H4-ST131 CTX-M-15 extended-spectrum beta-lactamase-producing Escherichia coli among companion animals. J Antimicrob Chemother 2010; 65: 651-660. Ewers C, Stamm I, Stolle I, et al. Detection of Shiga toxin- and extended-spectrum beta-lactamase-producing Escherichia coli O145:NM and Ont:NM from calves with diarrhoea. J Antimicrob Chemother 2014; 69: 2005-2007. Finley RL, Collignon P, Larsson DGJ, et al. The Scourge of Antibiotic Resistance: The Important Role of the Environment. Clin Infect Dis 2013; 57: 704-710. Forsberg KJ, Reyes A, Bin W, et al. The Shared Antibiotic Resistome of Soil Bacteria and Human Pathogens. Science 2012; 337: 1107-1111. Grad YH, Lipsitch M, Feldgarden M, et al. Genomic epidemiology of the Escherichia coli O104:H4 outbreaks in Europe, 2011. P Natl Acad Sci USA 2012; 109: 5547-5547. Green AM & Sambrook J Molecular Cloning: A Labarotory Manual. Cold Springer Harbor: Cold Springer Harbor Labarotory Press, 2012. Guenther S, Ewers C & Wieler LH Extended-Spectrum Beta-Lactamases Producing E. coli in Wildlife, yet Another Form of Environmental Pollution? Front Microbiol 2011; 2: 246. Guenther S, Bethe A, Fruth A, et al. Frequent combination of antimicrobial multiresistance and extraintestinal pathogenicity in Escherichia coli isolates from urban rats (Rattus norvegicus) in Berlin, Germany. PLoS One 2012; 7: e50331. Guenther S, Wuttke J, Bethe A, et al. Is Fecal Carriage of Extended-Spectrum-beta-Lactamase-Producing Escherichia coli in Urban Rats a Risk for Public Health? Antimicrob Agents Chemother 2013; 57: 2424-2425. Guenther S, Aschenbrenner K, Stamm I, et al. Comparable high rates of extended-spectrum-beta-lactamase-producing Escherichia coli in birds of prey from Germany and Mongolia. PLoS One 2012; 7: e53039. Jamborova I, Dolejska M, Vojtech J, et al. Plasmid-mediated resistance to cephalosporins and fluoroquinolones in various Escherichia coli sequence types isolated from rooks wintering in Europe. Appl Environ Microbiol 2015; 81: 648-657. Jouini A, Vinué L, Slama KB, et al. Characterization of CTX-M and SHV extended-spectrum beta-lactamases and associated resistance genes in Escherichia coli strains of food samples in Tunisia. J Antimicrob Chemother 2007; 60: 1137-1141. Knothe H, Shah P, Krcmery V, et al. Transferable resistance to cefotaxime, cefoxitin, cefamandole and cefuroxime in clinical isolates of Klebsiella pneumoniae and Serratia marcescens. Infection 1983; 11: 315-317. Lanza VF, de Toro M, Garcillan-Barcia MP, et al. Plasmid Flux in Escherichia coli ST131 Sublineages, Analyzed by Plasmid Constellation Network (PLACNET), a New Method for Plasmid Reconstruction from Whole Genome Sequences. Plos Genet 2014; 10: e1004766. Leistner R, Sakellariou C, Gurntke S, et al. Mortality and molecular epidemiology associated with extended-spectrum beta-lactamase production in Escherichia coli from bloodstream infection. Infect Drug Resist 2014; 7: 57-62. Li L, Stoeckert CJ & Roos DS OrthoMCL: Identification of ortholog groups for eukaryotic genomes. Genome Res 2003; 13: 2178-2189. Maguire AJ, Brown DF, Gray JJ, et al. Rapid screening technique for class 1 integrons in Enterobacteriaceae and nonfermenting gram-negative bacteria and its use in molecular epidemiology. Antimicrob Agents Chemother 2001; 45: 1022-1029. Naseer U & Sundsfjord A The CTX-M Conundrum: Dissemination of Plasmids and Escherichia coli Clones. Microb Drug Resist 2011; 17: 83-97.

Park YJ, Yu JK, Lee S, et al. Prevalence and diversity of qnr alleles in AmpC-producing Enterobacter cloacae, Enterobacter aerogenes, Citrobacter freundii and Serratia marcescens: a multicentre study from Korea. J Antimicrob Chemother 2007; 60: 868-871. Patel RK & Jain M NGS QC Toolkit: a toolkit for quality control of next generation sequencing data. PLoS One 2012; 7: e30619. Perez-Perez FJ & Hanson ND Detection of plasmid-mediated AmpC beta-lactamase genes in clinical isolates by using multiplex PCR. J Clin Microbiol 2002; 40: 2153-2162. Petty NK, Ben Zakour NL, Stanton-Cook M, et al. Global dissemination of a multidrug resistant Escherichia coli clone. Proc Natl Acad Sci U S A 2014; 111: 56945699. Pitout JD Infections with extended-spectrum beta-lactamase-producing enterobacteriaceae: changing epidemiology and drug treatment choices. Drugs 2010; 70: 313-333. Pitout JD Extraintestinal Pathogenic Escherichia coli: A Combination of Virulence with Antibiotic Resistance. Front Microbiol 2012; 3: 9. Pruden A, Larsson DGJ, Amezquita A, et al. Management Options for Reducing the Release of Antibiotics and Antibiotic Resistance Genes to the Environment. Environ Health Persp 2013; 121: 878-885. Radice M, Power P, Di Conza J, et al. Early dissemination of CTX-M-derived enzymes in South America. Antimicrob Agents Chemother 2002; 46: 602-604. Robicsek A, Strahilevitz J, Jacoby GA, et al. Fluoroquinolone-modifying enzyme: a new adaptation of a common aminoglycoside acetyltransferase. Nat Med 2006; 12: 83-88. Rodriguez I, Barownick W, Helmuth R, et al. Extended-spectrum beta-lactamases and AmpC beta-lactamases in ceftiofur-resistant Salmonella enterica isolates from food and livestock obtained in Germany during 2003-07. J Antimicrob Chemother 2009; 64: 301-309. Russo TA & Johnson JR Proposal for a New Inclusive Designation for Extraintestinal Pathogenic Isolates of Escherichia coli: ExPEC. J Infect Dis 2000; 181: 17531754. Schaufler K, Wieler LH, Semmler T, et al. ESBL-plasmids carrying toxin-antitoxin systems can be "cured" of wild-type Escherichia coli using a heat technique. Gut Pathog 2013; 5: 34. Schaufler K, Bethe A, Lubke-Becker A, et al. Putative connection between zoonotic multiresistant extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli in dog feces from a veterinary campus and clinical isolates from dogs. Infect Ecol Epidemiol 2015; 5: 25334. Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 2003; 13: 2498-2504. Sherchan JB, Hayakawa K, Miyoshi-Akiyama T, et al. Clinical epidemiology and molecular analysis of extended-spectrum-beta-lactamase-producing Escherichia coli in Nepal: characteristics of sequence types 131 and 648. Antimicrob Agents Chemother 2015; 59: 3424-3432. Smet A, Martel A, Persoons D, et al. Broad-spectrum beta-lactamases among Enterobacteriaceae of animal origin: molecular aspects, mobility and impact on public health. FEMS Microbiol Rev 2010; 34: 295-316. Sousa M, Torres C, Barros J, et al. Gilthead seabream (Sparus aurata) as carriers of SHV-12 and TEM-52 extended-spectrum beta-lactamases-containing Escherichia coli isolates. Foodborne Pathog Dis 2011; 8: 1139-1141. Stamatakis A RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 2014; 30: 1312-1313.

Tenover FC, Arbeit RD, Goering RV, et al. Interpreting Chromosomal DNA Restriction Patterns Produced by Pulsed-Field Gel-Electrophoresis - Criteria for Bacterial Strain Typing. J Clin Microbiol 1995; 33: 2233-2239. Teshager T, Dominguez L, Moreno MA, et al. Isolation of an SHV-12 beta -lactamase-producing Escherichia coli strain from a dog with recurrent urinary tract infections. Antimicrob Agents Chemother 2000; 44: 3483-3484. von Mentzer A, Connor TR, Wieler LH, et al. Identification of enterotoxigenic Escherichia coli (ETEC) clades with long-term global distribution. Nat Genet 2014; 46: 1321-1326. Walsh TR, Weeks J, Livermore DM, et al. Dissemination of NDM-1 positive bacteria in the New Delhi environment and its implications for human health: an environmental point prevalence study. Lancet Infect Dis 2011; 11: 355-362. Wang Z, Gerstein M & Snyder M RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 2009; 10: 57-63. Wirth T, Falush D, Lan R, et al. Sex and virulence in Escherichia coli: an evolutionary perspective. Mol Microbiol 2006; 60: 1136-1151. Zerbino DR & Birney E Velvet: Algorithms for de novo short read assembly using de Bruijn graphs. Genome Res 2008; 18: 821-829.

Tab. 1: Overview of avian ESBL-producing E. coli isolates, their host and sampling location (in Berlin and Brandenburg, Germany) as well as genotypic resistance features (Abbreviations: IMT=Institut für Mikrobiologie und Tierseuchen, PG=Phylogenetic group, ST=Sequence type, gyrA=mutations in the gyrase A gene, parC mutations in a gene encoding for a subunit of topoisomerase IV, number gives the position of the resulting amino acid change: A=alanine, D=aspartate, E=glutamate, I=isoleucine, L=leucine, N=asparagine, Q=glutamine, S=serine; V=valine, Y=tyrosine, WT=wild type, resistance genes: blaCTX-M-1/15, TEM-1, OXA-1, LAT1-4, CMY2-7, BIL-1, MIR-1T, ACT-1

[beta-lactams], tetA/B [tetracyclines], sul1-3 [sulfonamides], strA/B [streptomycin], aacC4 [aminoglycosides]). Samples

originate from one geographic area in Berlin and Brandenburg, Germany, and were collected between 2011 and 2014.

IMT number

Host

Sampling location

PG

ST

Resistance genotype

gyrA

parC

28551

mute swan

Berlin-Wannsee

B2

131

blaCTX-M-15, tetA, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I; 84 E-V

28552

sea eagle

Finsterwalde, Brandenburg

B1

88

blaCTX-M-1, blaTEM-1, sul1, sul2, strA/B

WT

WT

28707

mute swan

Körzin, Brandenburg

B1

410

blaCTX-M-15, blaOXA-1-group, tetA, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

28764

mute swan

Rangsdorf, Brandenburg

B1

410

blaCTX-M-15, blaOXA-1-group, tetA, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

29694

crow

Berlin-Charlottenburg

D

405

blaCTX-M-15, blaOXA-1-group, tetA, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

29792

mute swan

Potsdam, Brandenburg

A

617

blaCTX-M-15, blaOXA-1 group, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

29800

goshawk

Berlin-Lichtenberg

AxB1

224

blaCTX-M-1, bla(LAT1-4; CMY2-7; BIL-1)-group, bla(MIR-1T; ACT-1)-group, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

29851

goshawk

Berlin-Kreuzberg

ABD

1968

blaCTX-M-1, sul1, sul2, strA/B

83 S-L

WT

30167

herring gull

Zernsdorf, Brandenburg

A

373

blaCTX-M-1, tetA, sul1, sul2, strA/B

83 S-L; 87 D-Y

WT

30348

sparrow hawk

Tauche, Brandenburg

D

4306

blaCTX-M-1, tetB, sul1, sul2, strA/B

WT

WT

30349

pigeon

Berlin-Friedrichshain

D

1670

blaCTX-M-1, sul1, sul2, strA/B

WT

WT

30467

bean goose

Berlin-Schöneberg

B1

410

blaCTX-M-15, blaOXA-1-group, tetA, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

30763

mute swan

Berlin-Mariendorf

D

405

bla(LAT1-4; CMY2-7; BIL-1)-group, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

30987

sea eagle

Blumenthal, Brandenburg

A

167

blaCTX-M-1, blaTEM-1, aacC4, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I

30988

mute swan

Berlin-Grunewald

B1

1304

blaCTX-M-1, tetA, tetB, sul1, sul2, strA/B

83 S-L

WT

31281

magpie

Berlin-Neukölln

ABD

648

blaCTX-M-1, tetB, sul1, sul2, strA/B

WT

WT

31384

crow

Berlin-Mitte

B2

1730

blaCTX-M-1, tetA, tetB, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I; 84 E-V

31702

grey heron

Berlin-Charlottenburg

A

1167

bla(MOX-1,-2; CMY-1; -8-11)-group, sul1, sul2, strA/B

WT

WT

32636

mute swan

Berlin-Kreuzberg

D

4307

bla(LAT1-4; CMY2-7; BIL-1)-group, tetA, tetB, aacC4, sul1, sul2, strA/B

WT

WT

32637

mute swan

Berlin-Moabit

A

617

blaCTX-M-15, blaOXA-1-group, tetA, tetB, aacC4, sul1, sul2, sul3, strA/B

83 S-L; 87 D-N

80 S-I

32639

marsh harrier

Mellenau, Brandenburg

D

1204

blaCTX-M-1, tetA, sul1, sul2, sul3, strA/B

WT

WT

32660

mute swan

Potsdam, Brandenburg

D

115

blaCTX-M-1, blaOXA-1-group, tetA, tetB, aacC4, sul1, sul2, sul3, strA/B

WT

32661

mute swan

Berlin-Moabit

A

617

blaCTX-M-15, blaOXA-1-group, tetA, tetB, aacC4, sul1, sul2, strA/B

83 S-L; 87 D-N

80 S-I; 87 E-Q; 105 A-V 80 S-I

32662

buzzard

Groß-Kreutz, Brandenburg

AxB1

398

blaCTX-M-1, blaTEM-1, tetA, tetB, aacC4, sul1, sul2, sul3, strA/B

WT

WT