Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Characterization and Comparison of Extended-Spectrum β-Lactamase (ESBL) Resistance Genotypes and Population Structure of Escherichia coli Isolated from Franklin's Gulls (Leucophaeus pipixcan) and Humans in Chile

Characterization and Comparison of Extended-Spectrum β-Lactamase (ESBL) Resistance Genotypes and... We investigated the general level of antibiotic resistance with further analysis of extended-spectrum beta-lactamase (ESBL) prevalence, as well as the population structure of E. coli in fecal flora of humans and Franklin’s gulls (Leucophaeus pipixcan) in central parts of Chile. We found a surprisingly high carriage rate of ESBL-producing E. coli among the gulls 112/372 (30.1%) as compared to the human population 6/49 (12.2%.) Several of the E. coli sequence types (STs) identified in birds have previously been reported as Multi Drug Resistant (MDR) human pathogens including the ability to produce ESBLs. This means that not only commensal flora is shared between birds and humans but also STs with pathogenic potential. Given the migratory behavior of Franklin’s gulls, they and other migratory species, may be a part of ESBL dissemination in the environment and over great geographic distances. Apart from keeping the antibiotic use low, breaking the transmission chains between the environment and humans must be a priority to hinder the dissemination of resistance. Citation: Hernandez J, Johansson A, Stedt J, Bengtsson S, Porczak A, et al. (2013) Characterization and Comparison of Extended-Spectrum b-Lactamase (ESBL) Resistance Genotypes and Population Structure of Escherichia coli Isolated from Franklin’s Gulls (Leucophaeus pipixcan) and Humans in Chile. PLoS ONE 8(9): e76150. doi:10.1371/journal.pone.0076150 Editor: Axel Cloeckaert, Institut National de la Recherche Agronomique, France Received June 18, 2013; Accepted August 20, 2013; Published September 30, 2013 Copyright:  2013 Hernandez et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Anders Johansson would like to thank the Laboratory for Molecular Infection Medicine Sweden for supporting him by a Clinical Research Fellowship. This study was funded by the Swedish Research Council FORMAS (2008-326), the Karin Korsner’s and Olle Engkvist Byggma¨stare Foundations and the Medical Faculty, Uppsala University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] . These authors contributed equally to this work. in Chile are still few [8], and the occurrence of these bacteria Introduction among humans and in the environment is unclear. Extended spectrum b-lactamases of CTX-M type causing Clinical and environmental dissemination of CTX-M ESBLs antibiotic resistance in gram-negative bacteria emerged at several seem to be linked, and several reports have shown similarities locations worldwide in the 1980s [1], and have since become a between clinical human samples and wild avian samples [9,10]. major threat to modern medicine. Spread of CTX-M type ESBL Previous studies have often focused on avian species with close was in many parts of the world preceded by an initial spread of human contact, mainly gulls of different species that may be SHV and TEM ESBL type, but in South America these did not migratory or non-migratory depending on the specific population, have the same early dissemination [2]. Instead CTX-M ESBLs to investigate the dissemination potential. The Chilean coast is emerged early and are now widely distributed in countries such as visited every austral summer by hundreds of thousands of Argentina and Brazil from which most reports are available Franklin’s gulls Leucophaeus pipixcan. These gulls are mainly [2,3,4]. In Chile, the usage of antibiotics is large and uncontrolled, restricted to the coast and have an opportunistic feeding behavior, self-prescription is common, and antibiotics are sold over the seeking their food along the shoreline, offshore, but also in human counter [2]. Fish farming also consumes vast amounts of garbage. After their wintering period in Chile, they return to the antibiotics and Chile is among the top consumers of antibiotics central parts of Canada where they breed. Hence the contact with in the world [5]. Reports have confirmed an increase in bacterial humans and a potential to spread resistant bacteria both within antibiotic resistance in demersal and pelagic fish captured at the large bird colonies locally and through migration between Chilean coast [6,7]. However, reports on CTX-M dissemination PLOS ONE | www.plosone.org 1 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile continents make comparisons of antibiotic resistance markers in center the samples were submerged in bacterial storage medium, fecal bacteria in humans and gulls interesting. transported and stored as described above for avian samples. CTX-M ESBLs have high dissemination potential in bacterial populations due to the location of such resistance genes on Ethics statement transferable plasmids. In epidemiological investigations of plasmid The human sampling took place in San Antonio, Chile in bound resistance, the typing of replicons, which are genetic cooperation with Hospital Claudio Vicun ˜ a, San Antonio, Chile. structures encoding replication control mechanisms, has been well The ethics committe at the Hospital Claudio Vicun ˜ a, San Antonio described and frequently used [11]. It is believed that features of Chile has issued a formal written waiver which state as follows resistance plasmid replicons define their dissemination capabilities. (English translation): The most widespread plasmid replicon variants (or Inc-families) "After the acceptance of the Director of the Hospital Claudio are in the IncF family (FIA, FIB, FIC and FII), a heterogeneous Vicun ˜ a in 2010 the project "Presence of Beta-lactamase producing group of plasmids commonly found in collections of Escherichia coli bacteria in Birds and Humans in Chile" with the applicant Doctor [12,13]. These replicon types may also be found together in the Sr: Jorge Hernandez, University of Linneus and University of same plasmids, then designated multireplicons, one very common Uppsala, Hospital Director Dr. Milton Egan ˜ a, Claudio Vicun ˜a combination being FII-FIB [14]. A PCR-based replicon typing authorizes and assumes the ethical responsibility of the establishing (PBRT) scheme identifying 18 of the 27 major plasmid incom- and the proceeding to human fecal sampling of the population of patibility groups (Inc) found in Enterobacteriaceae [15] was developed San Antonio consulting primary care centers, since that the and introduced by Carattoli et al. and is now widely used [16]. procedure was adjusted to the principles and protocols in place at When investigating the zoonotic features and dissemination the Hospital Ethics Committee, i.e., in free decision with informed potential of E. coli between humans and the environment, an consent and under the commitment to withhold personal data of people who voluntarily agree to participate in the project". The understanding of the population genetics of the bacterium is highly important. Using multi locus sequence typing (MLST), it has participation was voluntarily and a written informed consent was collected from every participant. If the participant was below the previously been shown that E. coli sequence types (STs) that are age of 18, a written informed consent was collected from a parent. common among humans also occur in avians, precluding a clear separation of human and environmental E. coli populations A written permission to perform the gull sampling on the sites visited (only public places) was issued from the department of [9,10,17]. Agriculture, ‘‘Ministerio de Agricultura, Gobierno de Chile’’. Here we present a study of two species that live in close Since the fecal samples were taken from the ground there was no proximity to each other: humans and Franklin’s gulls at the interaction with the gulls at all. Chilean coast. Using a large material of both humans and Franklin’s gulls, we were able to compare their E. coli populations (as determined by MLST) as well as resistance genotypes (through Random selection of E. coli isolates from human and replicon and resistance genotyping) This work illustrates the avian samples zoonotic potential of E. coli and that there has been an exchange of Each fecal sample was inoculated on UriSelect 4 agar plates antibiotic resistance genes between birds and humans. (Bio-Rad Laboratories, Marnes-La-Coquette, France) for isolation of putative E. coli. When growth was present, one randomly Materials and Methods selected colony was isolated from each sample. E. coli species identity was confirmed by conventional biochemical testing. Collection of Franklin’s gull samples In January 2009, we collected 370 fresh fecal samples from adult Antibiotic susceptibility analysis Franklin’s gulls at their wintering sites in Chile, South America. The antibiotic susceptibility of the E. coli isolates was tested 197 samples were collected at the Aconcagua river delta and 173 against a set of antibiotic agents including tetracycline, ampicillin, samples near the Bio-Bio river delta. The distance between the streptomycin, chloramphenicol, nalidixic acid, cefadroxil, tigecy- Aconcagua and the Bio-Bio rivers is approximately 600 km and cline, Trimethoprim/sulfamethoxazole, nitrofurantoin and mecil- both rivers run through highly populated regions before they enter linam. These antibiotics were selected to represent commonly used the Pacific Ocean at the city of Conco ´ n and Talcahuano, agents against E. coli infections in human and veterinary medicine. respectively. Avian fecal material was submerged in bacterial Resistance was determined by antibiotic disk diffusion (discs from storage medium (Luria broth: BD, Sparks, USA, in phosphate- Oxoid, Basingstoke, UK) on Mueller-Hinton agar in accordance buffered saline with the addition of 0.45% Na-citrate, 0.1% with recommendations from EUCAST (The European Commit- MgSO4, 1% (NH4) 2SO4 and 4.4% glycerol). After sampling, tee on Antimicrobial Susceptibility Testing), using E. coli ATCC samples were transported and stored at 270uC for later 25922 as reference strain in all assessments. For antibiotics lacking examination. EUCAST-defined breakpoints for E. coli (tetracycline and strep- tomycin), the NRI method [18] also used by EUCAST, was Collection of human clinical samples implemented to define a local breakpoint. In March 2010, 49 human fecal samples from randomly selected volunteers (28 females and 21 males) aged 0 to 82 years Isolation of ESBL-producing bacteria (mean 36.5; median 38) were collected. The volunteers were All fecal samples of human or avian origin were also enriched in primary health care (PHC) patients at 18 different centers in 5 brain heart infusion broth (Becton Dickinson, Franklin Lakes, NJ, different counties visiting for non-infection related causes. All USA) supplemented with vancomycin (16 mg/L, ICN Biomedi- centers were located in the San Antonio area between the cals Inc. Aurora, OH, USA)) for 18h at 37uC, and subsequently TM Aconcagua river delta in Con-Con and the Maipo river delta in inoculated and cultured overnight at 37uC on chromID ESBL Llo-Lleo. Samples were taken at home and patients were plates (bioMe ´rieux, Marcy LEtoile, France), according to manu- instructed to cool the samples before deposit them at the PHC facturer’s instructions. Colonies were isolated and species identity facility the same or the following day. Upon delivery at the PHC confirmed by biochemical testing. ESBL production was con- PLOS ONE | www.plosone.org 2 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile firmed by cefpodoxime/cefpodoxime+clavulanic acid double disk Plasmid replicon tying test (MAST Diagnostics, Bootle, UK), before genetic characterisa- All samples selected for MLST analysis were also analysed for tion. plasmid replicon types. Plasmids were classified according to incompatibility groups using the PBRT scheme described by Genetic characterisation of ESBL variants Carattoli et al.[16]. PCR reactions were performed on the The presence of different bla genotypes was determined GeneAmp 9700 (Applied Biosystem, Foster City, USA) with CTX-M using a multiplex real-time PCR protocol [19] that display group JumpStart Taq polymerase (Sigma-Aldrich, Saint Louis Missouri, designation (CTX-M-1/-2/-8/25 and -9) of bla positive USA). Amplicons were visualised on 3% ReadyAgarose gels in CTX-M isolates. A subset of ESBL-producing isolates, that was also TBE buffer (Bio-Rad Laboratories Inc., Hercules CA, USA). investigated by MLST (see below), was further characterised by nucleotide sequencing of the CTX-M-1 and CTX-M-9 genes as Determination of conjugational transfer of ESBL described previously [20,21], and for the presence of bla and TEM harboring plasmids bla genes using primers from Pitout et al [22] and a previously SHV All ESBL-producing isolates selected for MLST analysis were described SYBRHGreen based real-time PCR protocol [10]. The also analysed for conjugational transfer of ESBL resistance. E. coli bla and bla detected were nucleotide sequenced using the TEM SHV J53 [25] was used as the recipient for detection of conjugation, and amplification primers. mating was carried out overnight at 37uC in LB broth (BD, Le Pont de Claix, France) including sodium azide (100 mg/ml) and Multi locus sequence typing cefotaxime (4 mg/ml). Transconjugants were cultured on TSA All randomly selected human E. coli isolates (n = 45), as well as agar (Oxoid, Basingstoke, England) including above concentra- all human ESBL-producing E. coli isolates (n= 5) were analysed by tions of sodium azide and cefotaxime. Conjugation was confirmed MLST. Among the avian E. coli isolates, a simple random selection by testing for presence of ESBL genes by PCR analysis. Plasmid of 50 ESBL-producing and 50 naı ¨ve (lacking resistance pheno- carriage of resistance genotypes was confirmed by conjugational types) isolates were chosen for MLST analysis. If an ESBL- transfer, which was positive for all but one isolate. producing isolate was selected from a sample that carried several different ESBL-producing isolates, no further isolates could be Results selected from that sample. The MLST scheme by Wirth et al. [23] was applied for analysing Antibiotic susceptibility of randomly selected E. coli seven gene fragments according to the protocols found at http:// isolates from human and avian samples mlst.ucc.ie/mlst/dbs/Ecoli/documents/primersColi_html as pre- Antibiotic susceptibility of randomly selected E. coli towards a viously described [9]. The MLST allele designations were panel of commonly used antimicrobial agents was assessed in determined via the online MLST database and all novel sequence order to receive a general resistance profile of human and avian alleles and ST designations discovered during this study were enterobacteria. From 49 human samples, 45 E. coli were isolated, provided by the curator of the database (http://mlst.ucc.ie/mlst/ and they showed an overall high susceptibility for most tested dbs/Ecoli). Each new sequence allele was independently verified by agents. The highest resistance frequencies were found for the curator of the database by evaluation of a minimum of two tetracycline and ampicillin, with frequencies of 17.8% for both, nucleotide sequence traces submitted by one of the authors (S.G) followed by nalidixic acid resistance in 6.7% of samples. All along with the information on the new reference strain for the public isolates were susceptible to cefadroxil, tigecycline and nitrofuran- database. The allele data was analysed by the minimum spanning toin, and the remaining tested agents had a resistance frequency of tree algorithm implemented in Bionumerics v.5.1 (Applied Maths approximately 2 to 4% (mecillinam in 2.2%, and streptomycin, NV) with priority rules set at first link genotypes that have chloramphenicol and trimethoprim -sulphamethoxazole, all in maximum numbers of single-locus variants and then maximal 4.4%, respectively, of isolates). numbers of single-locus variants and double-locus variants. From the 370 avian samples, 267 E. coli were isolated (160 from Concatenated nucleotide datasets of the seven genes were used for Conco ´ n and 107 from Talcahuano). These showed a general constructing a neighbour-joining tree based on the number of pair- resistance profile similar to that of human samples, with ampicillin wise differences among strains using the Pearson coefficient. Forty- resistance in 10.1%, tetracycline resistance in, 8.2%, and seven strains of the ECOR collection previously characterised by streptomycin resistance in 6.0% of the samples. Only nitrofuran- MLST were used as phylogroup references [23,24] The E. coli strain toin displayed full susceptibility in all samples, and for cefadroxil Z205/ST125 [23] was included to root the inferred phylogeny. there was a 1.1% resistance frequency. Other agents ranged Simpsons indices of diversity were calculated using the on-line tool between approximately 0.4 to 5% resistance rate (tigecycline in at the website http://darwin.phyloviz.net/ComparingPartitions/ 0.4%, mecillinam in 1.1%, chloramphenicol in 2.2%, and index.php?link = Home trimethoprim –sulphamethoxazole in 3.7% of isolates). There was no significant difference in resistance levels between Determination of gyrA quinolone resistance mutations Tapacalhuano and Concon samples or between human and avian All isolates selected for MLST analysis were analysed for the samples, as calculated by Fisher’s test. presence of mutations in the positions 248 and 259/260 of the gyrA gene sequence because these positions are known to confer Characterization of resistance genes encoding ESBLs in fluoroquinolone resistance. We designed the primers gyrAF, 5’- human and avian isolates CGTGTCGTTGGTGACGTAAT-3’ and gyrAR, 5’-CCACGC- Out of the 49 human samples, 6 were found to carry ESBL- GTTTTTCTTTTACC-3’ for amplification of 658 nucleotides producing bacteria; 5 isolates of E. coli and one Enterobacter cloacae that were sequenced on both strands using the amplification isolate. The isolates contained resistance genes of the group 1 primers and gyrARseq 5’-AGGAATTTTGGTTGGCATGA-3’. CTX-M type, namely bla (in two E. coli isolates), bla CTX-M-1 CTX- (two E. coli isolates), and bla (one E. coli and one E. M-15 CTX-M-30 PLOS ONE | www.plosone.org 3 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile cloacae isolate). No SHV or TEM type resistance genes were Table 2. CTX-M encoding genes and the presence of bla TEM identified. in the MLST-subset of 50 ESBL-producing isolates from Out of 372 avian samples, 112 samples carried ESBL-producing Franklins gulls. bacteria, some samples carried two (n= 15) or three (n= 1) phenotypically different isolates. Hence, a total of 129 ESBL- producing isolates were collected, all of which were E. coli. 122 bla (presence/ TEM isolates contained resistance genes of the group 1 CTX-M type No. of isolates CTX-M subtype absence) and some of them also contained SHV or TEM type resistance 33 CTX-M-1 - genes (Table 1). Four isolates contained the group 9 CTX-M and 6 CTX-M-1 + two the group 2 CTX-M resistance gene type. 6 CTX-M-15 + The 50 ESBL-producing E. coli isolates from birds included in the MLST analysis showed that isolates with enzymes of the group 2 CTX-M-15 - 1 CTX-M type harbored bla (n= 39), bla (n= 8), CTX-M-1 CTX-M-15 2 CTX-M-3 - or bla (n= 2), and that one isolate with group 9 CTX-M CTX-M-3 1 CTX-M-14 + type harbored bla resistance genes. One isolate harbored CTX-M-14 only bla , while eight CTX-M harboring isolates also SHV-12 doi:10.1371/journal.pone.0076150.t002 harbored the non-ESBL bla . (Table 2) TEM-1 MLST analysis of human and avian isolates Plasmid replicon types of human and avian isolates In E. coli isolated from birds, 67 different STs among the 100 Among the 150 (both ESBL and non-ESBL) isolates selected for isolates were identified (31 STs among 50 ESBL isolates, 40 STs MLST analysis 102 (68%) contained at least one type of replicon, among 50 non-ESBL isolates). Twenty-three new STs were and the proportion was similar among human (35/50; 70%) and identified in E. coli from birds (10 STs among ESBL isolates, 13 avian (67/100; 67%) isolates (Table 3). In the total material, STs among non-ESBL isolates; Figure 1). The most common STs presence of one or two replicons was common (n=57 and n= 37, were ST10 (10 isolates), ST1106, ST93, ST131, ST167, ST48 and respectively), but isolates with three and four replicons were also ST367 (3 isolates in each). ST10 and ST48 were found both found (n = 4 for both). There was a significant difference among ESBL as well as in non-ESBL E. coli. (P = 0.0006, as calculated by Fisher’s test) in presence of replicons Among E. coli isolated from humans there were 29 different STs in ESBL harbouring isolates versus non-ESBL isolates among among the 50 isolates (3 STs among 5 ESBL isolates; 29 STs avian samples (42/50 versus 25/50). Due to the small sample sizes among 45 non-ESBL isolates). Eight new STs were identified in E. (5/5 versus 30/45), this could not be established among human coli from humans (all of them among non-ESBL isolates; Figure 1). samples (Table 3). The most frequent STs were ST10 (14 isolates), ST58 and ST361, Inc FII (n = 47) and Inc I1 (n = 41) were the most frequently (3 isolates in each) (Figure 1 and 2). Phylogenetic analysis of detected replicon types among both ESBL/non-ESBL isolates, nucleotide sequences of seven gene fragments showed a diverse and human/avian isolates, although IncI1 replicons were more population of E. coli (Figure 1). Comparison with the reference frequent in the ESBL isolates (n = 33) (Table 4). The most sequences for E. coli of the ECOR-collection showed that all common combination of replicons was FII-FIB, present in 18 of 45 phylogroups were represented in the material. E. coli from gulls as isolates with multiple replicons. Most avian bla harboring CTX-M-15 well as from humans were spread evenly across the phylogroups isolates also displayed the FII-FIB combination, while human and specific STs, with no evident preference of human or avian bla harboring isolates did not. Further, the FII-FIB CTX-M-15 isolates, respectively. Both ESBL and non-ESBL isolates were replicon was present in several, both avian and human, non-ESBL found evenly distributed in the phylogenetic tree (Figure 1). Nine isolates. of a total of 87 STs were found in both human and avian samples (ST10, ST48, ST58, ST69, ST155, ST349, ST398, ST540 and Conjugational transfer of ESBL harboring plasmids ST641) (Figure 1). The most common STs were ST10 (24 isolates), Conjugational transfer was observed for all but one avian ST48 and ST58 (4 isolates each) (Figure 1). There were no obvious ESBL-harboring isolate. All human ESBL-producing isolates phylogeographic patterns, i.e., an analysis by the geographic displayed conjugational transfer, although one had a low efficiency source of the isolates showed little correlation to the genetic and was sensitive to variation in culturing conditions. Table 1. Distribution of b-lactamase genotypes among 129 ESBL-producing isolates from Franklins gulls. b-lactamase genotypes No. of isolates Group of CTX-M SHV (presence/absence) TEM (presence/absence) 101 CTX-M-1 - - 19 CTX-M-1 - + 2 CTX-M-1 + - 2 CTX-M-2 - - 2 CTX-M-9 - - 2 CTX-M-9 - + 1 None detected + - doi:10.1371/journal.pone.0076150.t001 PLOS ONE | www.plosone.org 4 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile Table 3. Total prevalence and the number of distinct replicon types identified in the MLST-subset of isolates from Franklins gulls and humans. Sample type ESBL status No. of isolates No. isolates with a given no. of replicon types (percent) $11 2 3 4 Franklins gulls ESBL 50 42 (84) 16 (32) 20 (40) 3 (6) 3 (6) Non-ESBL 50 25 (50) 15 (30) 10 (20) 0 (0) 0 (0) Humans ESBL 5 5 (100) 5 (100) 0 (0) 0 (0) 0 (0) Non-ESBL 45 30 (67) 21 (47) 7 (16) 1 (2) 1 (2) Total 150 102 (68) 57 (38) 37 (25) 4 (3) 4 (3) doi:10.1371/journal.pone.0076150.t003 analyses (Figure 2). Some STs contained isolates from all three that the mutations have appeared independently at several time sampling locations (ST10 and ST48) while several others points and is not (only) spread clonally. contained isolates from two of the locations. In Talcahuano, Among the five human ESBL-producing E. coli none was NAL- 35% (11/31) of the isolates were ESBL positive. In Con-con, the R. Among the human non-ESBL isolates three proved to be NAL- ratio was 57% (39/69). Some differences in genetic diversity R, but none of them had any of the investigated gyrA resistance among isolates from gulls and humans were indicated by crude mutations. analysis but with overlapping confidence intervals: Simpson’s index of diversity was 0.986 (CI 0.977-0.995) in bird isolates and Discussion 0.925 (CI 0.866-0.985) in isolates from humans. Simpson’s index We investigated the general level of resistance, with further in- of diversity was 0.991 (0.982-0.999) and 0.976 (0.962-0.991) depth analysis of ESBL prevalence in, as well as the phylogeny of among resistance naı ¨ve avian isolates and ESBL positive avian E. coli in fecal normal flora of humans and Franklin’s gulls in isolates, respectively. central parts of Chile. There were generally low levels of resistance in randomly selected (one from each sample) human and avian E. Analysis of gyrA mediated quinolone resistance coli isolates. By selective culturing of for identification of ESBL Out of the 50 bird ESBL isolates investigated by MLST, 11 isolates, however, we found a surprisingly high number of samples were resistant to NAL and the sequence of the gyrA gene was harboring ESBL-producing E. coli especially in the avian isolates. determined. Of the 11 NAL-R isolates, one isolate displayed a This illustrates that ESBL-producers in normal flora was not resistance mutation in position 83 (Ser83Leu) whereas 10 isolates dominant in frequency, as they were rarely picked when cultured had two mutations (Ser83Leu, and Asp87Asn). Although several unselectively, but that the bacterial population as a whole is well isolates of the same ST exhibited the same mutation (ST131/ prepared for degradation of cephalosporins. This is to our n = 2, ST167/n = 4, ST617/n = 2), the fact that the same knowledge the first time the fecal carriage of ESBL has been quinolone resistance mutation was found in different STs suggests compared between humans and wildlife from the same area. Table 4. The diversity of replicon types identified in the MLST-subset of 150 isolates from Franklins gulls and humans. Replicon type No. of isolates containing the indicated replicon types Total no. of each replicon type Franklins gulls Humans ESBL Non-ESBL ESBL Non-ESBL HI2 1 0 0 0 1 I1 28 1 5 7 41 N3 0 0 1 4 FIA 40 00 4 FIB 13 9 0 7 29 Y7 7 0 4 18 P1 0 0 0 1 FIC 31 00 4 A/C 0 0 0 1 1 FII 17 18 0 12 47 K0 0 0 5 5 B/O 0 0 0 5 5 The replicon types HI1, X, L/M, W, T, and FIIC were not detected in any of the isolates. doi:10.1371/journal.pone.0076150.t004 PLOS ONE | www.plosone.org 5 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile PLOS ONE | www.plosone.org 6 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile Figure 1. A neighbor joining tree based on seven concatenated gene sequences obtained by MLST. Genetic relationships between different STs and the distribution of the isolates between humans and Franklin's gulls are shown. The new and previously undescribed STs are marked by number 1711 or a higher number. The reference strain Z205 was used to root the tree. Forty-seven strains of the ECOR collection were included in the analysis as references. doi:10.1371/journal.pone.0076150.g001 Unexpectedly, the carriage rate of ESBL-producing bacteria interesting and could indicate an influx from poultry since CTX- among the gulls 112/372 (30.1%) were more than twice as high as M-1 has been the dominating ESBL in poultry production at least in the human population 6/49 (12.2%). We found mainly ESBLs in Europe [27,28]. of bla and bla genotypes that has been extensively Replicon typing identified the FII-FIB replicon combination in CTX-M-1 CTX-M -15 described in E. coli causing human infections, but also some rare a majority of the avian bla harboring isolates, most likely CTX-M -15 examples of bla and . A few occurred in combination representing a multireplicon contained on a single plasmid. The CTX_M-3,-14 -30 with SHVs or TEMs, but the larger part was only carrying CTX- FII-FIB replicon combination has previously been described to M type ESBLs. This confirms the rapid dissemination of different carry bla in E. coli from humans [14], and our findings in CTX-M -15 types of CTX-M ESBLs in Chile, both in comparison to the isolates from birds indicates a zoonotic transfer of these ESBL overall general low resistance profile of this material, but also when variants. Interestingly, none of the human bla isolates in CTX-M -15 compared to the low level of SHVs and TEMs, which emerged this study were of FII-FIB replicon type. Instead bla was CTX-M -15 prior to CTX-M ESBLs in Europe and North America [2,26]. found in combination with the I1 replicon type, the overall most The high proportion of CTX-M-1 in the avian samples is common type in this material, mostly found unpaired with other Figure 2. Minimal spanning tree calculated from the allelic differences between the isolates. The tree shows the relationship between the genetic analysis and the geographical source of the isolates. doi:10.1371/journal.pone.0076150.g002 PLOS ONE | www.plosone.org 7 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile replicon types. This is in line with previous studies describing E. human or avian), with no subgrouping present, further supporting coli isolates of both animal and human origin, harboring plasmids the zoonotic behavior of E. coli, regardless of ESBL carriage. of the Inc I1 group and carrying ESBL genes of both bla Considering the potential dissemination of E. coli by Franklin’s CTX-M -15 and bla as well as other genotypes [14]. Further, we found gulls to nesting places in the central parts of Canada, a recent CTX-M -1 more replicons among avian ESBL isolates than among avian non- publication on ESBLs in human clinical isolates from the central ESBL isolates. This is in agreement with a previous study of 127 Canadian region is of interest since ST10, ST617, ST38, ST131 human uropathogenic E. coli isolates lacking phenotypic resistance, and ST405 was common in this material ranging from year 2000 showing that such isolates, in contrast to antibiotic resistant to 2010 [34]. Apparently in line with prevalence patterns isolates, frequently are plasmid naıve and contains few replicon elsewhere, these STs were all present in our avian isolates and types [29]. Our findings are similar and suggest that although the similarly most often carrying bla CTX-M-15. detected replicon types are present in both ESBLs and non- There was no evidence for local epidemics of any particular E. ESBLs, they occur more frequently in ESBL harboring isolates coli ESBL clone among humans or gulls in this material. Rather a where they supposedly can facilitate and increase the spread of scenario emerges where the E. coli population as a whole is well resistance genes regardless of a human or avian host. prepared for degrading cephalosporins by containing a vast array The MLST data of this study support that humans and birds are of different ESBL-genes distributed among many different STs. sharing E. coli STs, i.e., there has been exchange of E coli between The fact that the ESBL carriage rate was more than twice as high birds and humans. We found nine STs (ST10, ST48, ST58, ST69, among the gulls compared to the humans underline that the ST155, ST349, ST398, ST540 and ST641) present in both avian antibiotic resistance problem must be seen in a large ecological and human populations (Figure 2). Several of the E. coli STs context. We have come to a situation when bacteria both in the detected in birds have previously been reported as MDR human environment and in the commensal flora have the capacity to pathogens including the ability to produce ESBLs. This means respond to almost any antibiotic used for treatment. Apart from that not only commensal flora is shared between humans and birds keeping the antibiotic use low, breaking the transmission chains but also STs with pathogenic potential. We found many STs between the environment (e.g. in gulls) and humans must be a previously described as major CTX-M carriers [30], including priority to hinder the dissemination of resistance. ST10, ST23, ST38, ST46, ST59, ST69, ST101, ST131, ST155, and ST405, of which only ST59 was unique to a human sample. Acknowledgments Interestingly, one avian ST38 isolate was found to carry bla CTX-M- We would like to acknowledge Dr. George Jacoby, Lahey Clinic, , although this ST has been reported to display preference for Burlington,MA, USA, for kindly providing E. coli J53, and Prof. Alessandra CTX-M group 9 genes [30,31,32]. There were many new E. coli Carattoli (Istituto Superiore di Sanita ´ , Rome, Italy for kindly providing STs detected in this study from both humans and in birds, and positive controls used in the replicon typing, Jonas Waldenstro ¨ m, Linnaeus several of the new STs from birds exhibited ESBL-production. In University for valuable comments on the manuscript, Project FONDE- addition, with the caveat that it is difficult to ensure full coverage CYT Nr. 1070494 and Pamela Ortega, Hospital Claudio Vicun ˜ a, San Antonio, Chile for logistic support. of the current literature, a number of previously described STs found in our avian material were new to wild animal hosts. Some ESBL-producing STs identified, e.g., ST10, ST48 and ST58 have Author Contributions previously been described in both humans and poultry [33]. For a Conceived and designed the experiments: JH BO JB MD AJ SG AP SB. full list of STs identified in this study, consult Figure 1. It should be Performed the experiments: JH JS AJ SG SB AP DG BO JB MD. noted, that the STs were spread evenly among the isolate Analyzed the data: JH AJ JB MD SG. Contributed reagents/materials/ populations (four combinations of ESBL or non-ESBL, and analysis tools: AJ BO SG MD. Wrote the paper: AJ MD JB. References 1. Bonnet R (2004) Growing group of extended-spectrum beta-lactamases: the isolates from Kalmar, on the southeast coast of Sweden. J Antimicrob CTX-M enzymes. Antimicrob Agents Chemother 48: 1–14. Chemother 65: 1939–1944. 2. Villegas MV, Kattan JN, Quinteros MG, Casellas JM (2008) Prevalence of 11. Novick RP (1987) Plasmid incompatibility. Microbiol Rev 51: 381–395. extended-spectrum beta-lactamases in South America. Clin Microbiol Infect 14 12. Datta N, Hughes VM (1983) Plasmids of the same Inc groups in Enterobacteria Suppl 1: 154–158. before and after the medical use of antibiotics. Nature 306: 616–617. 3. Bauernfeind A, Casellas JM, Goldberg M, Holley M, Jungwirth R, et al. (1992) 13. Johnson TJ, Wannemuehler YM, Johnson SJ, Logue CM, White DG, et al. A new plasmidic cefotaximase from patients infected with Salmonella (2007) Plasmid replicon typing of commensal and pathogenic Escherichia coli typhimurium. Infection 20: 158–163. isolates. Appl Environ Microbiol 73: 1976–1983. 4. Chagas TP, Alves RM, Vallim DC, Seki LM, Campos LC, et al. (2011) Diversity 14. Carattoli A, Garcia-Fernandez A, Varesi P, Fortini D, Gerardi S, et al. (2008) of genotypes in CTX-M-producing Klebsiella pneumoniae isolated in different Molecular epidemiology of Escherichia coli producing extended-spectrum beta- hospitals in Brazil. Braz J Infect Dis 15: 420–425. lactamases isolated in Rome, Italy. J Clin Microbiol 46: 103–108. 5. Millanao BA, Barrientos HM, Gomez CC, Tomova A, Buschmann A, et al. 15. Carattoli A (2009) Resistance plasmid families in Enterobacteriaceae. Anti- (2011) [Injudicious and excessive use of antibiotics: public health and salmon microb Agents Chemother 53: 2227–2238. aquaculture in Chile]. Rev Med Chil 139: 107–118. 16. Carattoli A, Bertini A, Villa L, Falbo V, Hopkins KL, et al. (2005) Identification 6. Cabello FC (2004) [Antibiotics and aquaculture in Chile: implications for human of plasmids by PCR-based replicon typing. J Microbiol Methods 63: 219–228. and animal health]. Rev Med Chil 132: 1001–1006. 17. Moura RA, Sircili MP, Leomil L, Matte MH, Trabulsi LR, et al. (2009) Clonal 7. Miranda CD, Zemelman R (2001) Antibiotic resistant bacteria in fish from the relationship among atypical enteropathogenic Escherichia coli strains isolated Concepcion Bay, Chile. Mar Pollut Bull 42: 1096–1102. from different animal species and humans. Appl Environ Microbiol 75: 7399– 8. Moreno MA, Teshager T, Porrero MA, Garcia M, Escudero E, et al. (2007) 7408. 18. Kronvall G, Kahlmeter G, Myhre E, Galas MF (2003) A new method for Abundance and phenotypic diversity of Escherichia coli isolates with diminished normalized interpretation of antimicrobial resistance from disk test results for susceptibility to expanded-spectrum cephalosporins in faeces from healthy food animals after slaughter. Vet Microbiol 120: 363–369. comparative purposes. Clin Microbiol Infect 9: 120–132. 9. Bonnedahl J, Drobni M, Gauthier-Clerc M, Hernandez J, Granholm S, et al. 19. Birkett CI, Ludlam HA, Woodford N, Brown DF, Brown NM, et al. (2007) (2009) Dissemination of Escherichia coli with CTX-M type ESBL between Real-time TaqMan PCR for rapid detection and typing of genes encoding CTX-M extended-spectrum beta-lactamases. J Med Microbiol 56: 52–55. humans and yellow-legged gulls in the south of France. PLoS One 4: e5958. 10. Bonnedahl J, Drobni P, Johansson A, Hernandez J, Melhus A, et al. (2010) 20. Bouallegue-Godet O, Ben Salem Y, Fabre L, Demartin M, Grimont PA, et al. Characterization, and comparison, of human clinical and black-headed gull (2005) Nosocomial outbreak caused by Salmonella enterica serotype Livingstone (Larus ridibundus) extended-spectrum beta-lactamase-producing bacterial producing CTX-M-27 extended-spectrum beta-lactamase in a neonatal unit in Sousse, Tunisia. J Clin Microbiol 43: 1037–1044. PLOS ONE | www.plosone.org 8 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile 21. Edelstein M, Pimkin M, Palagin I, Edelstein I, Stratchounski L (2003) Prevalence 29. Bengtsson S, Naseer U, Sundsfjord A, Kahlmeter G, Sundqvist M (2012) and molecular epidemiology of CTX-M extended-spectrum beta-lactamase- Sequence types and plasmid carriage of uropathogenic Escherichia coli devoid of producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. phenotypically detectable resistance. J Antimicrob Chemother 67: 69–73. Antimicrob Agents Chemother 47: 3724–3732. 30. Naseer U, Sundsfjord A (2011) The CTX-M conundrum: dissemination of 22. Pitout JD, Thomson KS, Hanson ND, Ehrhardt AF, Moland ES, et al. (1998) plasmids and Escherichia coli clones. Microb Drug Resist 17: 83–97. beta-Lactamases responsible for resistance to expanded-spectrum cephalosporins 31. Oteo J, Diestra K, Juan C, Bautista V, Novais A, et al. (2009) Extended- in Klebsiella pneumoniae, Escherichia coli, and Proteus mirabilis isolates spectrum beta-lactamase-producing Escherichia coli in Spain belong to a large recovered in South Africa. Antimicrob Agents Chemother 42: 1350–1354. variety of multilocus sequence typing types, including ST10 complex/A, ST23 23. Wirth T, Falush D, Lan R, Colles F, Mensa P, et al. (2006) Sex and virulence in complex/A and ST131/B2. Int J Antimicrob Agents 34: 173–176. Escherichia coli: an evolutionary perspective. Mol Microbiol 60: 1136–1151. 32. Suzuki S, Shibata N, Yamane K, Wachino J, Ito K, et al. (2009) Change in the 24. Ochman H, Selander RK (1984) Standard reference strains of Escherichia coli prevalence of extended-spectrum-beta-lactamase-producing Escherichia coli in from natural populations. J Bacteriol 157: 690–693. Japan by clonal spread. J Antimicrob Chemother 63: 72–79. 25. Jacoby GA, Han P (1996) Detection of extended-spectrum beta-lactamases in 33. Leverstein-van Hall MA, Dierikx CM, Cohen Stuart J, Voets GM, van den clinical isolates of Klebsiella pneumoniae and Escherichia coli. J Clin Microbiol Munckhof MP, et al. (2011) Dutch patients, retail chicken meat and poultry 34: 908–911. share the same ESBL genes, plasmids and strains. Clin Microbiol Infect 17: 873– 26. Canton R, Coque TM (2006) The CTX-M beta-lactamase pandemic. Curr Opin Microbiol 9: 466–475. 34. Peirano G, van der Bij AK, Gregson DB, Pitout JD (2012) Molecular 27. Overdevest I, Willemsen I, Rijnsburger M, Eustace A, Xu L, et al. (2011) epidemiology over an 11-year period (2000 to 2010) of extended-spectrum Extended-spectrum beta-lactamase genes of Escherichia coli in chicken meat beta-lactamase-producing Escherichia coli causing bacteremia in a centralized and humans, The Netherlands. Emerg Infect Dis 17: 1216–1222. Canadian region. J Clin Microbiol 50: 294–299. 28. Girlich D, Poirel L, Carattoli A, Kempf I, Lartigue MF, et al. (2007) Extended- spectrum beta-lactamase CTX-M-1 in Escherichia coli isolates from healthy poultry in France. Appl Environ Microbiol 73: 4681–4685. PLOS ONE | www.plosone.org 9 September 2013 | Volume 8 | Issue 9 | e76150 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PLoS ONE Unpaywall

Characterization and Comparison of Extended-Spectrum β-Lactamase (ESBL) Resistance Genotypes and Population Structure of Escherichia coli Isolated from Franklin's Gulls (Leucophaeus pipixcan) and Humans in Chile

Download PDF
9 pages

Loading next page...
 
/lp/unpaywall/characterization-and-comparison-of-extended-spectrum-lactamase-esbl-6FT09JKoMw

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1932-6203
DOI
10.1371/journal.pone.0076150
Publisher site
See Article on Publisher Site

Abstract

We investigated the general level of antibiotic resistance with further analysis of extended-spectrum beta-lactamase (ESBL) prevalence, as well as the population structure of E. coli in fecal flora of humans and Franklin’s gulls (Leucophaeus pipixcan) in central parts of Chile. We found a surprisingly high carriage rate of ESBL-producing E. coli among the gulls 112/372 (30.1%) as compared to the human population 6/49 (12.2%.) Several of the E. coli sequence types (STs) identified in birds have previously been reported as Multi Drug Resistant (MDR) human pathogens including the ability to produce ESBLs. This means that not only commensal flora is shared between birds and humans but also STs with pathogenic potential. Given the migratory behavior of Franklin’s gulls, they and other migratory species, may be a part of ESBL dissemination in the environment and over great geographic distances. Apart from keeping the antibiotic use low, breaking the transmission chains between the environment and humans must be a priority to hinder the dissemination of resistance. Citation: Hernandez J, Johansson A, Stedt J, Bengtsson S, Porczak A, et al. (2013) Characterization and Comparison of Extended-Spectrum b-Lactamase (ESBL) Resistance Genotypes and Population Structure of Escherichia coli Isolated from Franklin’s Gulls (Leucophaeus pipixcan) and Humans in Chile. PLoS ONE 8(9): e76150. doi:10.1371/journal.pone.0076150 Editor: Axel Cloeckaert, Institut National de la Recherche Agronomique, France Received June 18, 2013; Accepted August 20, 2013; Published September 30, 2013 Copyright:  2013 Hernandez et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Anders Johansson would like to thank the Laboratory for Molecular Infection Medicine Sweden for supporting him by a Clinical Research Fellowship. This study was funded by the Swedish Research Council FORMAS (2008-326), the Karin Korsner’s and Olle Engkvist Byggma¨stare Foundations and the Medical Faculty, Uppsala University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected] . These authors contributed equally to this work. in Chile are still few [8], and the occurrence of these bacteria Introduction among humans and in the environment is unclear. Extended spectrum b-lactamases of CTX-M type causing Clinical and environmental dissemination of CTX-M ESBLs antibiotic resistance in gram-negative bacteria emerged at several seem to be linked, and several reports have shown similarities locations worldwide in the 1980s [1], and have since become a between clinical human samples and wild avian samples [9,10]. major threat to modern medicine. Spread of CTX-M type ESBL Previous studies have often focused on avian species with close was in many parts of the world preceded by an initial spread of human contact, mainly gulls of different species that may be SHV and TEM ESBL type, but in South America these did not migratory or non-migratory depending on the specific population, have the same early dissemination [2]. Instead CTX-M ESBLs to investigate the dissemination potential. The Chilean coast is emerged early and are now widely distributed in countries such as visited every austral summer by hundreds of thousands of Argentina and Brazil from which most reports are available Franklin’s gulls Leucophaeus pipixcan. These gulls are mainly [2,3,4]. In Chile, the usage of antibiotics is large and uncontrolled, restricted to the coast and have an opportunistic feeding behavior, self-prescription is common, and antibiotics are sold over the seeking their food along the shoreline, offshore, but also in human counter [2]. Fish farming also consumes vast amounts of garbage. After their wintering period in Chile, they return to the antibiotics and Chile is among the top consumers of antibiotics central parts of Canada where they breed. Hence the contact with in the world [5]. Reports have confirmed an increase in bacterial humans and a potential to spread resistant bacteria both within antibiotic resistance in demersal and pelagic fish captured at the large bird colonies locally and through migration between Chilean coast [6,7]. However, reports on CTX-M dissemination PLOS ONE | www.plosone.org 1 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile continents make comparisons of antibiotic resistance markers in center the samples were submerged in bacterial storage medium, fecal bacteria in humans and gulls interesting. transported and stored as described above for avian samples. CTX-M ESBLs have high dissemination potential in bacterial populations due to the location of such resistance genes on Ethics statement transferable plasmids. In epidemiological investigations of plasmid The human sampling took place in San Antonio, Chile in bound resistance, the typing of replicons, which are genetic cooperation with Hospital Claudio Vicun ˜ a, San Antonio, Chile. structures encoding replication control mechanisms, has been well The ethics committe at the Hospital Claudio Vicun ˜ a, San Antonio described and frequently used [11]. It is believed that features of Chile has issued a formal written waiver which state as follows resistance plasmid replicons define their dissemination capabilities. (English translation): The most widespread plasmid replicon variants (or Inc-families) "After the acceptance of the Director of the Hospital Claudio are in the IncF family (FIA, FIB, FIC and FII), a heterogeneous Vicun ˜ a in 2010 the project "Presence of Beta-lactamase producing group of plasmids commonly found in collections of Escherichia coli bacteria in Birds and Humans in Chile" with the applicant Doctor [12,13]. These replicon types may also be found together in the Sr: Jorge Hernandez, University of Linneus and University of same plasmids, then designated multireplicons, one very common Uppsala, Hospital Director Dr. Milton Egan ˜ a, Claudio Vicun ˜a combination being FII-FIB [14]. A PCR-based replicon typing authorizes and assumes the ethical responsibility of the establishing (PBRT) scheme identifying 18 of the 27 major plasmid incom- and the proceeding to human fecal sampling of the population of patibility groups (Inc) found in Enterobacteriaceae [15] was developed San Antonio consulting primary care centers, since that the and introduced by Carattoli et al. and is now widely used [16]. procedure was adjusted to the principles and protocols in place at When investigating the zoonotic features and dissemination the Hospital Ethics Committee, i.e., in free decision with informed potential of E. coli between humans and the environment, an consent and under the commitment to withhold personal data of people who voluntarily agree to participate in the project". The understanding of the population genetics of the bacterium is highly important. Using multi locus sequence typing (MLST), it has participation was voluntarily and a written informed consent was collected from every participant. If the participant was below the previously been shown that E. coli sequence types (STs) that are age of 18, a written informed consent was collected from a parent. common among humans also occur in avians, precluding a clear separation of human and environmental E. coli populations A written permission to perform the gull sampling on the sites visited (only public places) was issued from the department of [9,10,17]. Agriculture, ‘‘Ministerio de Agricultura, Gobierno de Chile’’. Here we present a study of two species that live in close Since the fecal samples were taken from the ground there was no proximity to each other: humans and Franklin’s gulls at the interaction with the gulls at all. Chilean coast. Using a large material of both humans and Franklin’s gulls, we were able to compare their E. coli populations (as determined by MLST) as well as resistance genotypes (through Random selection of E. coli isolates from human and replicon and resistance genotyping) This work illustrates the avian samples zoonotic potential of E. coli and that there has been an exchange of Each fecal sample was inoculated on UriSelect 4 agar plates antibiotic resistance genes between birds and humans. (Bio-Rad Laboratories, Marnes-La-Coquette, France) for isolation of putative E. coli. When growth was present, one randomly Materials and Methods selected colony was isolated from each sample. E. coli species identity was confirmed by conventional biochemical testing. Collection of Franklin’s gull samples In January 2009, we collected 370 fresh fecal samples from adult Antibiotic susceptibility analysis Franklin’s gulls at their wintering sites in Chile, South America. The antibiotic susceptibility of the E. coli isolates was tested 197 samples were collected at the Aconcagua river delta and 173 against a set of antibiotic agents including tetracycline, ampicillin, samples near the Bio-Bio river delta. The distance between the streptomycin, chloramphenicol, nalidixic acid, cefadroxil, tigecy- Aconcagua and the Bio-Bio rivers is approximately 600 km and cline, Trimethoprim/sulfamethoxazole, nitrofurantoin and mecil- both rivers run through highly populated regions before they enter linam. These antibiotics were selected to represent commonly used the Pacific Ocean at the city of Conco ´ n and Talcahuano, agents against E. coli infections in human and veterinary medicine. respectively. Avian fecal material was submerged in bacterial Resistance was determined by antibiotic disk diffusion (discs from storage medium (Luria broth: BD, Sparks, USA, in phosphate- Oxoid, Basingstoke, UK) on Mueller-Hinton agar in accordance buffered saline with the addition of 0.45% Na-citrate, 0.1% with recommendations from EUCAST (The European Commit- MgSO4, 1% (NH4) 2SO4 and 4.4% glycerol). After sampling, tee on Antimicrobial Susceptibility Testing), using E. coli ATCC samples were transported and stored at 270uC for later 25922 as reference strain in all assessments. For antibiotics lacking examination. EUCAST-defined breakpoints for E. coli (tetracycline and strep- tomycin), the NRI method [18] also used by EUCAST, was Collection of human clinical samples implemented to define a local breakpoint. In March 2010, 49 human fecal samples from randomly selected volunteers (28 females and 21 males) aged 0 to 82 years Isolation of ESBL-producing bacteria (mean 36.5; median 38) were collected. The volunteers were All fecal samples of human or avian origin were also enriched in primary health care (PHC) patients at 18 different centers in 5 brain heart infusion broth (Becton Dickinson, Franklin Lakes, NJ, different counties visiting for non-infection related causes. All USA) supplemented with vancomycin (16 mg/L, ICN Biomedi- centers were located in the San Antonio area between the cals Inc. Aurora, OH, USA)) for 18h at 37uC, and subsequently TM Aconcagua river delta in Con-Con and the Maipo river delta in inoculated and cultured overnight at 37uC on chromID ESBL Llo-Lleo. Samples were taken at home and patients were plates (bioMe ´rieux, Marcy LEtoile, France), according to manu- instructed to cool the samples before deposit them at the PHC facturer’s instructions. Colonies were isolated and species identity facility the same or the following day. Upon delivery at the PHC confirmed by biochemical testing. ESBL production was con- PLOS ONE | www.plosone.org 2 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile firmed by cefpodoxime/cefpodoxime+clavulanic acid double disk Plasmid replicon tying test (MAST Diagnostics, Bootle, UK), before genetic characterisa- All samples selected for MLST analysis were also analysed for tion. plasmid replicon types. Plasmids were classified according to incompatibility groups using the PBRT scheme described by Genetic characterisation of ESBL variants Carattoli et al.[16]. PCR reactions were performed on the The presence of different bla genotypes was determined GeneAmp 9700 (Applied Biosystem, Foster City, USA) with CTX-M using a multiplex real-time PCR protocol [19] that display group JumpStart Taq polymerase (Sigma-Aldrich, Saint Louis Missouri, designation (CTX-M-1/-2/-8/25 and -9) of bla positive USA). Amplicons were visualised on 3% ReadyAgarose gels in CTX-M isolates. A subset of ESBL-producing isolates, that was also TBE buffer (Bio-Rad Laboratories Inc., Hercules CA, USA). investigated by MLST (see below), was further characterised by nucleotide sequencing of the CTX-M-1 and CTX-M-9 genes as Determination of conjugational transfer of ESBL described previously [20,21], and for the presence of bla and TEM harboring plasmids bla genes using primers from Pitout et al [22] and a previously SHV All ESBL-producing isolates selected for MLST analysis were described SYBRHGreen based real-time PCR protocol [10]. The also analysed for conjugational transfer of ESBL resistance. E. coli bla and bla detected were nucleotide sequenced using the TEM SHV J53 [25] was used as the recipient for detection of conjugation, and amplification primers. mating was carried out overnight at 37uC in LB broth (BD, Le Pont de Claix, France) including sodium azide (100 mg/ml) and Multi locus sequence typing cefotaxime (4 mg/ml). Transconjugants were cultured on TSA All randomly selected human E. coli isolates (n = 45), as well as agar (Oxoid, Basingstoke, England) including above concentra- all human ESBL-producing E. coli isolates (n= 5) were analysed by tions of sodium azide and cefotaxime. Conjugation was confirmed MLST. Among the avian E. coli isolates, a simple random selection by testing for presence of ESBL genes by PCR analysis. Plasmid of 50 ESBL-producing and 50 naı ¨ve (lacking resistance pheno- carriage of resistance genotypes was confirmed by conjugational types) isolates were chosen for MLST analysis. If an ESBL- transfer, which was positive for all but one isolate. producing isolate was selected from a sample that carried several different ESBL-producing isolates, no further isolates could be Results selected from that sample. The MLST scheme by Wirth et al. [23] was applied for analysing Antibiotic susceptibility of randomly selected E. coli seven gene fragments according to the protocols found at http:// isolates from human and avian samples mlst.ucc.ie/mlst/dbs/Ecoli/documents/primersColi_html as pre- Antibiotic susceptibility of randomly selected E. coli towards a viously described [9]. The MLST allele designations were panel of commonly used antimicrobial agents was assessed in determined via the online MLST database and all novel sequence order to receive a general resistance profile of human and avian alleles and ST designations discovered during this study were enterobacteria. From 49 human samples, 45 E. coli were isolated, provided by the curator of the database (http://mlst.ucc.ie/mlst/ and they showed an overall high susceptibility for most tested dbs/Ecoli). Each new sequence allele was independently verified by agents. The highest resistance frequencies were found for the curator of the database by evaluation of a minimum of two tetracycline and ampicillin, with frequencies of 17.8% for both, nucleotide sequence traces submitted by one of the authors (S.G) followed by nalidixic acid resistance in 6.7% of samples. All along with the information on the new reference strain for the public isolates were susceptible to cefadroxil, tigecycline and nitrofuran- database. The allele data was analysed by the minimum spanning toin, and the remaining tested agents had a resistance frequency of tree algorithm implemented in Bionumerics v.5.1 (Applied Maths approximately 2 to 4% (mecillinam in 2.2%, and streptomycin, NV) with priority rules set at first link genotypes that have chloramphenicol and trimethoprim -sulphamethoxazole, all in maximum numbers of single-locus variants and then maximal 4.4%, respectively, of isolates). numbers of single-locus variants and double-locus variants. From the 370 avian samples, 267 E. coli were isolated (160 from Concatenated nucleotide datasets of the seven genes were used for Conco ´ n and 107 from Talcahuano). These showed a general constructing a neighbour-joining tree based on the number of pair- resistance profile similar to that of human samples, with ampicillin wise differences among strains using the Pearson coefficient. Forty- resistance in 10.1%, tetracycline resistance in, 8.2%, and seven strains of the ECOR collection previously characterised by streptomycin resistance in 6.0% of the samples. Only nitrofuran- MLST were used as phylogroup references [23,24] The E. coli strain toin displayed full susceptibility in all samples, and for cefadroxil Z205/ST125 [23] was included to root the inferred phylogeny. there was a 1.1% resistance frequency. Other agents ranged Simpsons indices of diversity were calculated using the on-line tool between approximately 0.4 to 5% resistance rate (tigecycline in at the website http://darwin.phyloviz.net/ComparingPartitions/ 0.4%, mecillinam in 1.1%, chloramphenicol in 2.2%, and index.php?link = Home trimethoprim –sulphamethoxazole in 3.7% of isolates). There was no significant difference in resistance levels between Determination of gyrA quinolone resistance mutations Tapacalhuano and Concon samples or between human and avian All isolates selected for MLST analysis were analysed for the samples, as calculated by Fisher’s test. presence of mutations in the positions 248 and 259/260 of the gyrA gene sequence because these positions are known to confer Characterization of resistance genes encoding ESBLs in fluoroquinolone resistance. We designed the primers gyrAF, 5’- human and avian isolates CGTGTCGTTGGTGACGTAAT-3’ and gyrAR, 5’-CCACGC- Out of the 49 human samples, 6 were found to carry ESBL- GTTTTTCTTTTACC-3’ for amplification of 658 nucleotides producing bacteria; 5 isolates of E. coli and one Enterobacter cloacae that were sequenced on both strands using the amplification isolate. The isolates contained resistance genes of the group 1 primers and gyrARseq 5’-AGGAATTTTGGTTGGCATGA-3’. CTX-M type, namely bla (in two E. coli isolates), bla CTX-M-1 CTX- (two E. coli isolates), and bla (one E. coli and one E. M-15 CTX-M-30 PLOS ONE | www.plosone.org 3 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile cloacae isolate). No SHV or TEM type resistance genes were Table 2. CTX-M encoding genes and the presence of bla TEM identified. in the MLST-subset of 50 ESBL-producing isolates from Out of 372 avian samples, 112 samples carried ESBL-producing Franklins gulls. bacteria, some samples carried two (n= 15) or three (n= 1) phenotypically different isolates. Hence, a total of 129 ESBL- producing isolates were collected, all of which were E. coli. 122 bla (presence/ TEM isolates contained resistance genes of the group 1 CTX-M type No. of isolates CTX-M subtype absence) and some of them also contained SHV or TEM type resistance 33 CTX-M-1 - genes (Table 1). Four isolates contained the group 9 CTX-M and 6 CTX-M-1 + two the group 2 CTX-M resistance gene type. 6 CTX-M-15 + The 50 ESBL-producing E. coli isolates from birds included in the MLST analysis showed that isolates with enzymes of the group 2 CTX-M-15 - 1 CTX-M type harbored bla (n= 39), bla (n= 8), CTX-M-1 CTX-M-15 2 CTX-M-3 - or bla (n= 2), and that one isolate with group 9 CTX-M CTX-M-3 1 CTX-M-14 + type harbored bla resistance genes. One isolate harbored CTX-M-14 only bla , while eight CTX-M harboring isolates also SHV-12 doi:10.1371/journal.pone.0076150.t002 harbored the non-ESBL bla . (Table 2) TEM-1 MLST analysis of human and avian isolates Plasmid replicon types of human and avian isolates In E. coli isolated from birds, 67 different STs among the 100 Among the 150 (both ESBL and non-ESBL) isolates selected for isolates were identified (31 STs among 50 ESBL isolates, 40 STs MLST analysis 102 (68%) contained at least one type of replicon, among 50 non-ESBL isolates). Twenty-three new STs were and the proportion was similar among human (35/50; 70%) and identified in E. coli from birds (10 STs among ESBL isolates, 13 avian (67/100; 67%) isolates (Table 3). In the total material, STs among non-ESBL isolates; Figure 1). The most common STs presence of one or two replicons was common (n=57 and n= 37, were ST10 (10 isolates), ST1106, ST93, ST131, ST167, ST48 and respectively), but isolates with three and four replicons were also ST367 (3 isolates in each). ST10 and ST48 were found both found (n = 4 for both). There was a significant difference among ESBL as well as in non-ESBL E. coli. (P = 0.0006, as calculated by Fisher’s test) in presence of replicons Among E. coli isolated from humans there were 29 different STs in ESBL harbouring isolates versus non-ESBL isolates among among the 50 isolates (3 STs among 5 ESBL isolates; 29 STs avian samples (42/50 versus 25/50). Due to the small sample sizes among 45 non-ESBL isolates). Eight new STs were identified in E. (5/5 versus 30/45), this could not be established among human coli from humans (all of them among non-ESBL isolates; Figure 1). samples (Table 3). The most frequent STs were ST10 (14 isolates), ST58 and ST361, Inc FII (n = 47) and Inc I1 (n = 41) were the most frequently (3 isolates in each) (Figure 1 and 2). Phylogenetic analysis of detected replicon types among both ESBL/non-ESBL isolates, nucleotide sequences of seven gene fragments showed a diverse and human/avian isolates, although IncI1 replicons were more population of E. coli (Figure 1). Comparison with the reference frequent in the ESBL isolates (n = 33) (Table 4). The most sequences for E. coli of the ECOR-collection showed that all common combination of replicons was FII-FIB, present in 18 of 45 phylogroups were represented in the material. E. coli from gulls as isolates with multiple replicons. Most avian bla harboring CTX-M-15 well as from humans were spread evenly across the phylogroups isolates also displayed the FII-FIB combination, while human and specific STs, with no evident preference of human or avian bla harboring isolates did not. Further, the FII-FIB CTX-M-15 isolates, respectively. Both ESBL and non-ESBL isolates were replicon was present in several, both avian and human, non-ESBL found evenly distributed in the phylogenetic tree (Figure 1). Nine isolates. of a total of 87 STs were found in both human and avian samples (ST10, ST48, ST58, ST69, ST155, ST349, ST398, ST540 and Conjugational transfer of ESBL harboring plasmids ST641) (Figure 1). The most common STs were ST10 (24 isolates), Conjugational transfer was observed for all but one avian ST48 and ST58 (4 isolates each) (Figure 1). There were no obvious ESBL-harboring isolate. All human ESBL-producing isolates phylogeographic patterns, i.e., an analysis by the geographic displayed conjugational transfer, although one had a low efficiency source of the isolates showed little correlation to the genetic and was sensitive to variation in culturing conditions. Table 1. Distribution of b-lactamase genotypes among 129 ESBL-producing isolates from Franklins gulls. b-lactamase genotypes No. of isolates Group of CTX-M SHV (presence/absence) TEM (presence/absence) 101 CTX-M-1 - - 19 CTX-M-1 - + 2 CTX-M-1 + - 2 CTX-M-2 - - 2 CTX-M-9 - - 2 CTX-M-9 - + 1 None detected + - doi:10.1371/journal.pone.0076150.t001 PLOS ONE | www.plosone.org 4 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile Table 3. Total prevalence and the number of distinct replicon types identified in the MLST-subset of isolates from Franklins gulls and humans. Sample type ESBL status No. of isolates No. isolates with a given no. of replicon types (percent) $11 2 3 4 Franklins gulls ESBL 50 42 (84) 16 (32) 20 (40) 3 (6) 3 (6) Non-ESBL 50 25 (50) 15 (30) 10 (20) 0 (0) 0 (0) Humans ESBL 5 5 (100) 5 (100) 0 (0) 0 (0) 0 (0) Non-ESBL 45 30 (67) 21 (47) 7 (16) 1 (2) 1 (2) Total 150 102 (68) 57 (38) 37 (25) 4 (3) 4 (3) doi:10.1371/journal.pone.0076150.t003 analyses (Figure 2). Some STs contained isolates from all three that the mutations have appeared independently at several time sampling locations (ST10 and ST48) while several others points and is not (only) spread clonally. contained isolates from two of the locations. In Talcahuano, Among the five human ESBL-producing E. coli none was NAL- 35% (11/31) of the isolates were ESBL positive. In Con-con, the R. Among the human non-ESBL isolates three proved to be NAL- ratio was 57% (39/69). Some differences in genetic diversity R, but none of them had any of the investigated gyrA resistance among isolates from gulls and humans were indicated by crude mutations. analysis but with overlapping confidence intervals: Simpson’s index of diversity was 0.986 (CI 0.977-0.995) in bird isolates and Discussion 0.925 (CI 0.866-0.985) in isolates from humans. Simpson’s index We investigated the general level of resistance, with further in- of diversity was 0.991 (0.982-0.999) and 0.976 (0.962-0.991) depth analysis of ESBL prevalence in, as well as the phylogeny of among resistance naı ¨ve avian isolates and ESBL positive avian E. coli in fecal normal flora of humans and Franklin’s gulls in isolates, respectively. central parts of Chile. There were generally low levels of resistance in randomly selected (one from each sample) human and avian E. Analysis of gyrA mediated quinolone resistance coli isolates. By selective culturing of for identification of ESBL Out of the 50 bird ESBL isolates investigated by MLST, 11 isolates, however, we found a surprisingly high number of samples were resistant to NAL and the sequence of the gyrA gene was harboring ESBL-producing E. coli especially in the avian isolates. determined. Of the 11 NAL-R isolates, one isolate displayed a This illustrates that ESBL-producers in normal flora was not resistance mutation in position 83 (Ser83Leu) whereas 10 isolates dominant in frequency, as they were rarely picked when cultured had two mutations (Ser83Leu, and Asp87Asn). Although several unselectively, but that the bacterial population as a whole is well isolates of the same ST exhibited the same mutation (ST131/ prepared for degradation of cephalosporins. This is to our n = 2, ST167/n = 4, ST617/n = 2), the fact that the same knowledge the first time the fecal carriage of ESBL has been quinolone resistance mutation was found in different STs suggests compared between humans and wildlife from the same area. Table 4. The diversity of replicon types identified in the MLST-subset of 150 isolates from Franklins gulls and humans. Replicon type No. of isolates containing the indicated replicon types Total no. of each replicon type Franklins gulls Humans ESBL Non-ESBL ESBL Non-ESBL HI2 1 0 0 0 1 I1 28 1 5 7 41 N3 0 0 1 4 FIA 40 00 4 FIB 13 9 0 7 29 Y7 7 0 4 18 P1 0 0 0 1 FIC 31 00 4 A/C 0 0 0 1 1 FII 17 18 0 12 47 K0 0 0 5 5 B/O 0 0 0 5 5 The replicon types HI1, X, L/M, W, T, and FIIC were not detected in any of the isolates. doi:10.1371/journal.pone.0076150.t004 PLOS ONE | www.plosone.org 5 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile PLOS ONE | www.plosone.org 6 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile Figure 1. A neighbor joining tree based on seven concatenated gene sequences obtained by MLST. Genetic relationships between different STs and the distribution of the isolates between humans and Franklin's gulls are shown. The new and previously undescribed STs are marked by number 1711 or a higher number. The reference strain Z205 was used to root the tree. Forty-seven strains of the ECOR collection were included in the analysis as references. doi:10.1371/journal.pone.0076150.g001 Unexpectedly, the carriage rate of ESBL-producing bacteria interesting and could indicate an influx from poultry since CTX- among the gulls 112/372 (30.1%) were more than twice as high as M-1 has been the dominating ESBL in poultry production at least in the human population 6/49 (12.2%). We found mainly ESBLs in Europe [27,28]. of bla and bla genotypes that has been extensively Replicon typing identified the FII-FIB replicon combination in CTX-M-1 CTX-M -15 described in E. coli causing human infections, but also some rare a majority of the avian bla harboring isolates, most likely CTX-M -15 examples of bla and . A few occurred in combination representing a multireplicon contained on a single plasmid. The CTX_M-3,-14 -30 with SHVs or TEMs, but the larger part was only carrying CTX- FII-FIB replicon combination has previously been described to M type ESBLs. This confirms the rapid dissemination of different carry bla in E. coli from humans [14], and our findings in CTX-M -15 types of CTX-M ESBLs in Chile, both in comparison to the isolates from birds indicates a zoonotic transfer of these ESBL overall general low resistance profile of this material, but also when variants. Interestingly, none of the human bla isolates in CTX-M -15 compared to the low level of SHVs and TEMs, which emerged this study were of FII-FIB replicon type. Instead bla was CTX-M -15 prior to CTX-M ESBLs in Europe and North America [2,26]. found in combination with the I1 replicon type, the overall most The high proportion of CTX-M-1 in the avian samples is common type in this material, mostly found unpaired with other Figure 2. Minimal spanning tree calculated from the allelic differences between the isolates. The tree shows the relationship between the genetic analysis and the geographical source of the isolates. doi:10.1371/journal.pone.0076150.g002 PLOS ONE | www.plosone.org 7 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile replicon types. This is in line with previous studies describing E. human or avian), with no subgrouping present, further supporting coli isolates of both animal and human origin, harboring plasmids the zoonotic behavior of E. coli, regardless of ESBL carriage. of the Inc I1 group and carrying ESBL genes of both bla Considering the potential dissemination of E. coli by Franklin’s CTX-M -15 and bla as well as other genotypes [14]. Further, we found gulls to nesting places in the central parts of Canada, a recent CTX-M -1 more replicons among avian ESBL isolates than among avian non- publication on ESBLs in human clinical isolates from the central ESBL isolates. This is in agreement with a previous study of 127 Canadian region is of interest since ST10, ST617, ST38, ST131 human uropathogenic E. coli isolates lacking phenotypic resistance, and ST405 was common in this material ranging from year 2000 showing that such isolates, in contrast to antibiotic resistant to 2010 [34]. Apparently in line with prevalence patterns isolates, frequently are plasmid naıve and contains few replicon elsewhere, these STs were all present in our avian isolates and types [29]. Our findings are similar and suggest that although the similarly most often carrying bla CTX-M-15. detected replicon types are present in both ESBLs and non- There was no evidence for local epidemics of any particular E. ESBLs, they occur more frequently in ESBL harboring isolates coli ESBL clone among humans or gulls in this material. Rather a where they supposedly can facilitate and increase the spread of scenario emerges where the E. coli population as a whole is well resistance genes regardless of a human or avian host. prepared for degrading cephalosporins by containing a vast array The MLST data of this study support that humans and birds are of different ESBL-genes distributed among many different STs. sharing E. coli STs, i.e., there has been exchange of E coli between The fact that the ESBL carriage rate was more than twice as high birds and humans. We found nine STs (ST10, ST48, ST58, ST69, among the gulls compared to the humans underline that the ST155, ST349, ST398, ST540 and ST641) present in both avian antibiotic resistance problem must be seen in a large ecological and human populations (Figure 2). Several of the E. coli STs context. We have come to a situation when bacteria both in the detected in birds have previously been reported as MDR human environment and in the commensal flora have the capacity to pathogens including the ability to produce ESBLs. This means respond to almost any antibiotic used for treatment. Apart from that not only commensal flora is shared between humans and birds keeping the antibiotic use low, breaking the transmission chains but also STs with pathogenic potential. We found many STs between the environment (e.g. in gulls) and humans must be a previously described as major CTX-M carriers [30], including priority to hinder the dissemination of resistance. ST10, ST23, ST38, ST46, ST59, ST69, ST101, ST131, ST155, and ST405, of which only ST59 was unique to a human sample. Acknowledgments Interestingly, one avian ST38 isolate was found to carry bla CTX-M- We would like to acknowledge Dr. George Jacoby, Lahey Clinic, , although this ST has been reported to display preference for Burlington,MA, USA, for kindly providing E. coli J53, and Prof. Alessandra CTX-M group 9 genes [30,31,32]. There were many new E. coli Carattoli (Istituto Superiore di Sanita ´ , Rome, Italy for kindly providing STs detected in this study from both humans and in birds, and positive controls used in the replicon typing, Jonas Waldenstro ¨ m, Linnaeus several of the new STs from birds exhibited ESBL-production. In University for valuable comments on the manuscript, Project FONDE- addition, with the caveat that it is difficult to ensure full coverage CYT Nr. 1070494 and Pamela Ortega, Hospital Claudio Vicun ˜ a, San Antonio, Chile for logistic support. of the current literature, a number of previously described STs found in our avian material were new to wild animal hosts. Some ESBL-producing STs identified, e.g., ST10, ST48 and ST58 have Author Contributions previously been described in both humans and poultry [33]. For a Conceived and designed the experiments: JH BO JB MD AJ SG AP SB. full list of STs identified in this study, consult Figure 1. It should be Performed the experiments: JH JS AJ SG SB AP DG BO JB MD. noted, that the STs were spread evenly among the isolate Analyzed the data: JH AJ JB MD SG. Contributed reagents/materials/ populations (four combinations of ESBL or non-ESBL, and analysis tools: AJ BO SG MD. Wrote the paper: AJ MD JB. References 1. Bonnet R (2004) Growing group of extended-spectrum beta-lactamases: the isolates from Kalmar, on the southeast coast of Sweden. J Antimicrob CTX-M enzymes. Antimicrob Agents Chemother 48: 1–14. Chemother 65: 1939–1944. 2. Villegas MV, Kattan JN, Quinteros MG, Casellas JM (2008) Prevalence of 11. Novick RP (1987) Plasmid incompatibility. Microbiol Rev 51: 381–395. extended-spectrum beta-lactamases in South America. Clin Microbiol Infect 14 12. Datta N, Hughes VM (1983) Plasmids of the same Inc groups in Enterobacteria Suppl 1: 154–158. before and after the medical use of antibiotics. Nature 306: 616–617. 3. Bauernfeind A, Casellas JM, Goldberg M, Holley M, Jungwirth R, et al. (1992) 13. Johnson TJ, Wannemuehler YM, Johnson SJ, Logue CM, White DG, et al. A new plasmidic cefotaximase from patients infected with Salmonella (2007) Plasmid replicon typing of commensal and pathogenic Escherichia coli typhimurium. Infection 20: 158–163. isolates. Appl Environ Microbiol 73: 1976–1983. 4. Chagas TP, Alves RM, Vallim DC, Seki LM, Campos LC, et al. (2011) Diversity 14. Carattoli A, Garcia-Fernandez A, Varesi P, Fortini D, Gerardi S, et al. (2008) of genotypes in CTX-M-producing Klebsiella pneumoniae isolated in different Molecular epidemiology of Escherichia coli producing extended-spectrum beta- hospitals in Brazil. Braz J Infect Dis 15: 420–425. lactamases isolated in Rome, Italy. J Clin Microbiol 46: 103–108. 5. Millanao BA, Barrientos HM, Gomez CC, Tomova A, Buschmann A, et al. 15. Carattoli A (2009) Resistance plasmid families in Enterobacteriaceae. Anti- (2011) [Injudicious and excessive use of antibiotics: public health and salmon microb Agents Chemother 53: 2227–2238. aquaculture in Chile]. Rev Med Chil 139: 107–118. 16. Carattoli A, Bertini A, Villa L, Falbo V, Hopkins KL, et al. (2005) Identification 6. Cabello FC (2004) [Antibiotics and aquaculture in Chile: implications for human of plasmids by PCR-based replicon typing. J Microbiol Methods 63: 219–228. and animal health]. Rev Med Chil 132: 1001–1006. 17. Moura RA, Sircili MP, Leomil L, Matte MH, Trabulsi LR, et al. (2009) Clonal 7. Miranda CD, Zemelman R (2001) Antibiotic resistant bacteria in fish from the relationship among atypical enteropathogenic Escherichia coli strains isolated Concepcion Bay, Chile. Mar Pollut Bull 42: 1096–1102. from different animal species and humans. Appl Environ Microbiol 75: 7399– 8. Moreno MA, Teshager T, Porrero MA, Garcia M, Escudero E, et al. (2007) 7408. 18. Kronvall G, Kahlmeter G, Myhre E, Galas MF (2003) A new method for Abundance and phenotypic diversity of Escherichia coli isolates with diminished normalized interpretation of antimicrobial resistance from disk test results for susceptibility to expanded-spectrum cephalosporins in faeces from healthy food animals after slaughter. Vet Microbiol 120: 363–369. comparative purposes. Clin Microbiol Infect 9: 120–132. 9. Bonnedahl J, Drobni M, Gauthier-Clerc M, Hernandez J, Granholm S, et al. 19. Birkett CI, Ludlam HA, Woodford N, Brown DF, Brown NM, et al. (2007) (2009) Dissemination of Escherichia coli with CTX-M type ESBL between Real-time TaqMan PCR for rapid detection and typing of genes encoding CTX-M extended-spectrum beta-lactamases. J Med Microbiol 56: 52–55. humans and yellow-legged gulls in the south of France. PLoS One 4: e5958. 10. Bonnedahl J, Drobni P, Johansson A, Hernandez J, Melhus A, et al. (2010) 20. Bouallegue-Godet O, Ben Salem Y, Fabre L, Demartin M, Grimont PA, et al. Characterization, and comparison, of human clinical and black-headed gull (2005) Nosocomial outbreak caused by Salmonella enterica serotype Livingstone (Larus ridibundus) extended-spectrum beta-lactamase-producing bacterial producing CTX-M-27 extended-spectrum beta-lactamase in a neonatal unit in Sousse, Tunisia. J Clin Microbiol 43: 1037–1044. PLOS ONE | www.plosone.org 8 September 2013 | Volume 8 | Issue 9 | e76150 Comparison of ESBL from Gulls and Humans in Chile 21. Edelstein M, Pimkin M, Palagin I, Edelstein I, Stratchounski L (2003) Prevalence 29. Bengtsson S, Naseer U, Sundsfjord A, Kahlmeter G, Sundqvist M (2012) and molecular epidemiology of CTX-M extended-spectrum beta-lactamase- Sequence types and plasmid carriage of uropathogenic Escherichia coli devoid of producing Escherichia coli and Klebsiella pneumoniae in Russian hospitals. phenotypically detectable resistance. J Antimicrob Chemother 67: 69–73. Antimicrob Agents Chemother 47: 3724–3732. 30. Naseer U, Sundsfjord A (2011) The CTX-M conundrum: dissemination of 22. Pitout JD, Thomson KS, Hanson ND, Ehrhardt AF, Moland ES, et al. (1998) plasmids and Escherichia coli clones. Microb Drug Resist 17: 83–97. beta-Lactamases responsible for resistance to expanded-spectrum cephalosporins 31. Oteo J, Diestra K, Juan C, Bautista V, Novais A, et al. (2009) Extended- in Klebsiella pneumoniae, Escherichia coli, and Proteus mirabilis isolates spectrum beta-lactamase-producing Escherichia coli in Spain belong to a large recovered in South Africa. Antimicrob Agents Chemother 42: 1350–1354. variety of multilocus sequence typing types, including ST10 complex/A, ST23 23. Wirth T, Falush D, Lan R, Colles F, Mensa P, et al. (2006) Sex and virulence in complex/A and ST131/B2. Int J Antimicrob Agents 34: 173–176. Escherichia coli: an evolutionary perspective. Mol Microbiol 60: 1136–1151. 32. Suzuki S, Shibata N, Yamane K, Wachino J, Ito K, et al. (2009) Change in the 24. Ochman H, Selander RK (1984) Standard reference strains of Escherichia coli prevalence of extended-spectrum-beta-lactamase-producing Escherichia coli in from natural populations. J Bacteriol 157: 690–693. Japan by clonal spread. J Antimicrob Chemother 63: 72–79. 25. Jacoby GA, Han P (1996) Detection of extended-spectrum beta-lactamases in 33. Leverstein-van Hall MA, Dierikx CM, Cohen Stuart J, Voets GM, van den clinical isolates of Klebsiella pneumoniae and Escherichia coli. J Clin Microbiol Munckhof MP, et al. (2011) Dutch patients, retail chicken meat and poultry 34: 908–911. share the same ESBL genes, plasmids and strains. Clin Microbiol Infect 17: 873– 26. Canton R, Coque TM (2006) The CTX-M beta-lactamase pandemic. Curr Opin Microbiol 9: 466–475. 34. Peirano G, van der Bij AK, Gregson DB, Pitout JD (2012) Molecular 27. Overdevest I, Willemsen I, Rijnsburger M, Eustace A, Xu L, et al. (2011) epidemiology over an 11-year period (2000 to 2010) of extended-spectrum Extended-spectrum beta-lactamase genes of Escherichia coli in chicken meat beta-lactamase-producing Escherichia coli causing bacteremia in a centralized and humans, The Netherlands. Emerg Infect Dis 17: 1216–1222. Canadian region. J Clin Microbiol 50: 294–299. 28. Girlich D, Poirel L, Carattoli A, Kempf I, Lartigue MF, et al. (2007) Extended- spectrum beta-lactamase CTX-M-1 in Escherichia coli isolates from healthy poultry in France. Appl Environ Microbiol 73: 4681–4685. PLOS ONE | www.plosone.org 9 September 2013 | Volume 8 | Issue 9 | e76150

Journal

PLoS ONE – Unpaywall

Published: Sep 30, 2013

There are no references for this article.