Abstract

Background

Concomitant with the recent emergence of CTX-M-type extended-spectrum β-lactamases (ESBLs), Escherichia coli has become the enterobacterial species most affected by ESBLs. Multiple locales are encountering CTX-M-positive E. coli, including specifically CTX-M-15. To gain insights into the mechanism underlying this phenomenon, we assessed clonality and diversity of virulence profiles within an international collection of CTX-M-15-positive E. coli.

Methods

Forty-one ESBL-positive E. coli isolates from eight countries and three continents (Europe, Asia and North America) were selected for study based on suspected clonality. Phylogenetic group, ERIC2 PCR profile, O H serotype, AmpC variant and antibiotic susceptibility were determined. Multilocus sequence typing (MLST) and PFGE provided additional discrimination. Virulence potential was inferred by detection of 46 virulence factor (VF) genes.

Results

Thirty-six (88%) of the 41 E. coli isolates exhibited the same set of core characteristics: phylogenetic group B2, ERIC2 PCR profile 1, serotype O25:H4, AmpC EC6, ciprofloxacin resistance and MLST profile ST131. By PFGE, the 36 isolates constituted one large cluster at the 68% similarity level; this comprised 17 PFGE groups (defined at 85% similarity), some of which included strains from different countries. The 36 isolates exhibited highly (91% to 100%) similar VF profiles.

Conclusions

We describe a broadly disseminated, CTX-M-15-positive and virulent E. coli clonal group with highly homogeneous virulence genotypes and subgroups exhibiting highly similar PFGE profiles, suggesting recent emergence. Understanding how this clone has emerged and successfully disseminated within the hospital and community, including across national boundaries, should be a public health priority.

Introduction

Escherichia coli, a universal commensal of humans and several animal species, is also one of the most common enterobacterial species causing extraintestinal infections in these same hosts. E. coli infections are becoming increasingly difficult to treat because of emerging antimicrobial resistance, most recently to expanded-spectrum cephalosporins, which is usually due to the production of extended-spectrum β-lactamases (ESBLs).1 The earliest ESBLs, which were first reported in 1985, consisted of plasmid-mediated TEM-1, TEM-2 and SHV-1 derivatives and were primarily a hospital-based problem.1 However, since 2000, ESBLs increasingly have also appeared in the community.2 This phenomenon coincided with the emergence of a new group of plasmid-mediated ESBLs, namely the CTX-M enzymes, which seem to be taking over as the main ESBL type in some locales.3,4

Multiple locales are encountering CTX-M-positive E. coli clinical isolates, including specifically CTX-M-15, which is one of the more than 60 variants described in this enzyme group and is able to efficiently hydrolyse not only cefotaxime but also ceftazidime.3–8 The widespread occurrence of CTX-M-15-positive E. coli could have two alternative explanations. That is, the corresponding plasmids or other mobile genetic elements surrounding the plasmid-mediated blaCTX-M-15 gene may be moving from strain to strain through the E. coli population.3,9 Alternatively, the strains themselves may be spreading in a clonal fashion, as has been described for methicillin-resistant Staphylococcus aureus, penicillin-resistant Streptococcus pneumoniae and certain clonal groups of trimethoprim/sulfamethoxazole-resistant E. coli (i.e. ‘clonal group A’ and E. coli O15:K52:H1).10–13

While some articles have reported on the similarity of CTX-M-15-encoding plasmids harboured by strains in different locations, in the present study, we focused on strain genetic background to assess the extent of clonality within a collection of international CTX-M-positive E. coli isolates that we suspected were clonally related to a group of French CTX-M-15 isolates of serogroup O25 previously studied in our laboratory.6,14,15 We also sought to assess these strains’ molecularly inferred virulence potential, a possible contributor (in addition to antimicrobial resistance) to their recent emergence and dissemination as successful pathogens.

Materials and methods

Isolate collection

While we observed in France that CTX-M-15 was the most common CTX-M enzyme and that the great majority of epidemiologically unrelated, CTX-M-15-positive E. coli isolates displayed an identical genetic background, we also observed that different studies performed in different countries found CTX-M-15 as the most common CTX-M-type enzyme in E coli. These concomitant observations pushed us to see whether this worldwide outbreak of CTX-M-15-producing E. coli was due to the spread of a clonal strain as found in France. Therefore, a total of 41 recent human E. coli isolates that were known (n = 34) or presumed (n = 7) to produce CTX-M-15, from three continents (Europe, Asia and North America) and eight countries (France, Portugal, Spain, Switzerland, Lebanon, India, Korea and Canada) were studied. They were selected because they either were known to be clonal (the 13 French isolates) or, if from outside France, were suspected of being related to the French isolates based on CTX-M-15 production, the O25 antigen, the phylogenetic group B2 and/or ciprofloxacin resistance. Previously published isolates included those from Canada (n = 6), India (n = 2), Korea (n = 2), Lebanon (n = 4) and Portugal (n = 5).1620 As indicated in Table 1, they included community-, hospital- and nursing-home-acquired isolates. Except for the Lebanese and two French strains that were digestive tract colonizers, the isolates were obtained from clinical samples: primarily urine but also blood, sputum, intra-abdominal pus and ascites (Table 1).

Table 1

Clinical, bacterial and molecular characteristics of the 41 studied isolates of ESBL-producing E. coli

        Antimicrobial susceptibility 
Country/isolate Sample/acquisition site Epidemic (Ep) or sporadic (Sp) CTX-M-type ERIC2 PCR profile Phylogenetic group AmpC type Serotype CIP GEN AMK TET CHL SXT 
France              
 MECB5 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 Vlab2 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 VA1 catheter/C Sp M15 1 B2 EC6 O25:H4 R 
 VB6 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 VB8 rectal swab/C Sp M15 1 B2 EC6 O25:H4 R 
 VB9 rectal swab/C Sp M15 1 B2 EC6 O25:H4 R 
 HBS1 urine/H Sp M15 1 B2 EC6 O25:H4 R 
 HBS4 ascites/H Sp M15 1 B2 EC6 O25:H4 R 
 TNN (TE2) urine/H Ep-AC-1 M15 1 B2 EC6 O25:H4 R 
 TE1 urine/H Ep-LTC-1 M15 1 B2 EC6 O25:H4 R R R R S S 
 HDE1 urine/H Ep-LTC 2a M15 1 B2 EC6 O25:H4 R 
 HDE2 urine/H Ep-LTC-2b M15 1 B2 EC6 O25:H4 R 
 HDE3 urine/H Ep-LTC-2c M15 1 B2 EC6 O25:H4 R 
Switzerland              
 EcS1 urine/NA Sp M15 B2 EC30 NT 
 3756 EcS2 urine/NA Sp M15 B2 EC6 O25:H4 
Spain              
 FV7561 urine, blood/H Sp M15 B2 EC6 O25:H4 
 FV7563 urine/C Sp M15 B2 EC6 O25:H4 
 FV7569, FV7588, FV7593, FV7595 urine/H Sp M15 B2 EC6 O25:H4 
 FV7591 urine/H Sp M1 EC68 O25:H4 
Portugal              
 5753 urine/C Sp M15 B2 EC6 O25:H4 
 5754 blood/C Sp M15 B2 EC6 O25:H4 
 5800 sputum/H Sp M15 B2 EC6 O25:H4 
 5936 urine/C Sp M15 B2 EC6 O25:H4 
 6373 urine/C Sp M15 B2 EC6 O25:H4 
Korea              
 KUMC KN1604 urine/H Sp M15 B2 EC6 O25:H4 
 KUMC KN1608 urine/H Sp M15 EC68 NT 
India              
E. coliurine/H Sp M15 EC66 O101 
E. coliurine/H Sp M15 B1 EC74 NT 
Lebanon              
 AH8, AH9, AH10, AH15 faeces/C Sp M15 B2 EC6 O25:H4 
Canada              
 1100 urine/C Sp M15 B2 EC6 O25:H4 R 
 17102 urine/C Sp M15 B2 EC6 O25:H4 R 
 15802 intra abd pus/H Sp M15 B2 EC6 O25:H4 R 
 19502 urine/NH Sp M15 B2 EC6 O25:H4 R 
 8501 urine/NH Sp M15 B2 EC6 O25:H4 R 
 16102 urine/C Sp M15 B2 EC6 O25:H4 R 
        Antimicrobial susceptibility 
Country/isolate Sample/acquisition site Epidemic (Ep) or sporadic (Sp) CTX-M-type ERIC2 PCR profile Phylogenetic group AmpC type Serotype CIP GEN AMK TET CHL SXT 
France              
 MECB5 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 Vlab2 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 VA1 catheter/C Sp M15 1 B2 EC6 O25:H4 R 
 VB6 urine/C Sp M15 1 B2 EC6 O25:H4 R 
 VB8 rectal swab/C Sp M15 1 B2 EC6 O25:H4 R 
 VB9 rectal swab/C Sp M15 1 B2 EC6 O25:H4 R 
 HBS1 urine/H Sp M15 1 B2 EC6 O25:H4 R 
 HBS4 ascites/H Sp M15 1 B2 EC6 O25:H4 R 
 TNN (TE2) urine/H Ep-AC-1 M15 1 B2 EC6 O25:H4 R 
 TE1 urine/H Ep-LTC-1 M15 1 B2 EC6 O25:H4 R R R R S S 
 HDE1 urine/H Ep-LTC 2a M15 1 B2 EC6 O25:H4 R 
 HDE2 urine/H Ep-LTC-2b M15 1 B2 EC6 O25:H4 R 
 HDE3 urine/H Ep-LTC-2c M15 1 B2 EC6 O25:H4 R 
Switzerland              
 EcS1 urine/NA Sp M15 B2 EC30 NT 
 3756 EcS2 urine/NA Sp M15 B2 EC6 O25:H4 
Spain              
 FV7561 urine, blood/H Sp M15 B2 EC6 O25:H4 
 FV7563 urine/C Sp M15 B2 EC6 O25:H4 
 FV7569, FV7588, FV7593, FV7595 urine/H Sp M15 B2 EC6 O25:H4 
 FV7591 urine/H Sp M1 EC68 O25:H4 
Portugal              
 5753 urine/C Sp M15 B2 EC6 O25:H4 
 5754 blood/C Sp M15 B2 EC6 O25:H4 
 5800 sputum/H Sp M15 B2 EC6 O25:H4 
 5936 urine/C Sp M15 B2 EC6 O25:H4 
 6373 urine/C Sp M15 B2 EC6 O25:H4 
Korea              
 KUMC KN1604 urine/H Sp M15 B2 EC6 O25:H4 
 KUMC KN1608 urine/H Sp M15 EC68 NT 
India              
E. coliurine/H Sp M15 EC66 O101 
E. coliurine/H Sp M15 B1 EC74 NT 
Lebanon              
 AH8, AH9, AH10, AH15 faeces/C Sp M15 B2 EC6 O25:H4 
Canada              
 1100 urine/C Sp M15 B2 EC6 O25:H4 R 
 17102 urine/C Sp M15 B2 EC6 O25:H4 R 
 15802 intra abd pus/H Sp M15 B2 EC6 O25:H4 R 
 19502 urine/NH Sp M15 B2 EC6 O25:H4 R 
 8501 urine/NH Sp M15 B2 EC6 O25:H4 R 
 16102 urine/C Sp M15 B2 EC6 O25:H4 R 

R, resistant; S, susceptible; C, community-acquired; H, hospital-acquired; NH, nursing home; intra abd, intra-abdominal; AC, acute care; LTC, long-term care; NA, not available; NT, non-typeable; CIP, ciprofloxacin; GEN, gentamicin; AMK, amikacin; TET, tetracycline; CHL, chloramphenicol; SXT, co-trimoxazole.

Italic font indicates the bacteriological characteristics that were already known when the strains were selected for this study.

Relevant characteristics of the isolates that were known prior to this study are italicized in Table 1. Notably, the French isolates [including a strain (TE1) previously reported as responsible for an outbreak in a long-term care facility] were previously shown to produce CTX-M-15 and to exhibit the same genomic PCR profile, O antigen (O25), chromosomal cephalosporinase variant (AmpC EC6) and ciprofloxacin phenotype (resistant).15,21

β-Lactamase determination

CTX-M-type β-lactamase genes were identified as previously described.21 Briefly, two sets of primers were used, allowing amplification and sequencing of any type of blaCTX-M gene (primer set 1) versus specific blaCTX-M variants (primer set 2). To determine an isolate's ampC variant, DNA amplification was done using primers VL1A (5′-TGCACGATCTGAAAATCCAC-3′) and VL2A (5′-AGCAGGCGCATAAATGTTTC-3′) under standard PCR conditions, with a Tm of 42°C, which yielded a fragment of 1398 bp. Direct sequencing of the PCR product was performed using these PCR primers and two additional primers, VL1S (5′-TATCTTCAATGGTCG-3′) and VL2S (5′-TGCATGGGCTCCAGG-3′). The ampC nucleotide sequences and deduced protein sequences were analysed by using software available at the Biosupport web site (http://bioinfo.hku.hk/). These were then compared with those available in GenBank by using Blast sequence software (htpp://www.ncbi.nlm.nih.gov). The new AmpC peptide sequences were named EC66 and EC68 and their corresponding genes were deposited in GenBank under accession numbers EF507686 and EF507687, respectively.

ERIC2 PCR profiles

ERIC2 PCR profiles, which are strain-specific banding patterns obtained by amplifying multiple anonymous regions of the genome using repetitive element-based primers, were generated as previously described, with bacterial lysates used as template DNA.22 Profiles were defined as different when they exhibited at least one high intensity band difference according to visual inspection.

Phylogenetic group

Determination of major E. coli phylogenetic group (A, B1, B2 and D) was done by multiplex PCR.23

Serotyping

The determination of O and H antigens was carried out by using the method previously described by Guinée et al.,24 in which all available O (O1–O185) and H (H1–H56) antisera were tested. All antisera were obtained and absorbed with the corresponding cross-reacting antigens to remove the non-specific agglutinins. The O25 antigen was also determined by a PCR-based method.25

Sequence type (ST) determination

Multilocus sequence typing (MLST) was carried out as previously described.26 Gene amplification and sequencing were performed by using the primers specified at the E. coli MLST web site (http://web.mpiib-berlin.mpg.de/mlst/dbs/Ecoli) except for mdh, icd and recA. Both the forward and reverse strands of mdh were sequenced using primers mdh SF: 5′-CCAGGCGCTTGCACTACTGTTAA-3′ and mdh SR: 5′-GCGATATCTTTCTTCAGCGTATC-3′, respectively, whereas the forward strand of icd and the reverse strand of recA were sequenced with the primers icd SF: 5′-CGGCAAACTCAACGTTCC-3′ and recA SR: 5′-CTGACGCTGCAGGTGAT-3′, respectively. Allelic profile and ST determinations were as per the E. coli MLST web site scheme.

PFGE profiles

XbaI PFGE analysis was performed as previously described.27 Profiles were compared digitally using BioNumerics software (Applied Maths). Cluster analysis of Dice similarity indices based on the unweighted pair group method with arithmetic mean (UPGMA) was used to generate a dendrogram describing the relationships among PFGE profiles. Isolates were considered to belong to the same PFGE group if their Dice similarity index was ≥85%.28

Virulence genotypes

Forty-six extraintestinal virulence-associated genes were detected by multiplex PCR, as previously described.29 These included 16 adhesin-encoding genes (papAH, papC, papEF, papG and its 3 alleles, sfa/focDE, sfaS, focG, afa/draBC, afaE8, iha, bmaE, gafD, F17, clpG, fimH and hra), 8 toxin-encoding genes (hlyA, hlyF, cnf1, cdtB, sat, pic, tsh and astA) and 4 siderophore-related genes (iroN, fyuA, ireA and iutA). They also included 10 protectin/invasin-encoding genes (kpsM II, kpsMT III, the K1, K2, K5 and K15 kps variants, rfc, traT, ibeA and iss) and 7 pathogenicity island markers and miscellaneous genes (cvaC, usp, ompT, clbB, clbN, fliC H7 and malX). A UPGMA-based dendrogram was constructed depicting similarity relationships among the isolates according to composite virulence gene profiles.

Antibiotic susceptibility

Susceptibility to the following non-β-lactam molecules was determined by disc diffusion: ciprofloxacin, gentamicin, amikacin, tetracycline, chloramphenicol and co-trimoxazole. Isolates were defined as resistant or susceptible according to the standards of the French Antibiogram Committee.30

Results

The primary strain set comprised 13 epidemiologically diverse French E. coli isolates, all from the Paris area except one (strain MECB5, from the south of France). All were known to be characterized in terms of CTX-M-15 production, phylogenetic group B2, ERIC2 PCR profile 1, serogroup O25, AmpC variant 6 and ciprofloxacin resistance (Table 1). H antigen determined in this study was found to be H4.

These characteristics (when unknown) were newly assessed for 28 other ESBL-positive E. coli isolates from seven other countries, representing three continents. Twenty-three (82%) of these 28 isolates were found to be identical to the 13 French isolates with respect to all 6 core characteristics (CTX-M-15, group B2, ERIC2 PCR profile 1, serotype O25:H4, AmpC EC6 and ciprofloxacin resistance). These 23 isolates included 6 (86%) of 7 from Spain and 17 (81%) of 21 from the other six countries, including all 15 from Lebanon, Portugal and Canada, plus 1 each from Switzerland and Korea (Table 1).

In contrast, five of the isolates (both isolates from India, and one each from Spain, Switzerland and Korea) were found to exhibit non-1 ERIC profiles (Figure 1) and non-EC6 AmpC types, and proved to be mostly non-O25 and non-B2 (Table 1). Moreover, the Spanish isolate (FV7591) exhibited CTX-M-1 rather than CTX-M-15 (Table 1).

Figure 1

ERIC2 PCR profiles of seven Spanish and three French isolates of ESBL-producing E. coli. Lane 1, molecular weight marker. Lanes 2–8, Spanish isolates with strain FV7591 in lane 6. Lanes 9–11, French isolates [strain HBS1 (lane 9), strain HDE2 (lane 10) and strain TE1 (lane 11)]. A uniform profile was found among these isolates except for the Spanish strain FV7591 (CTX-M-1).

Figure 1

ERIC2 PCR profiles of seven Spanish and three French isolates of ESBL-producing E. coli. Lane 1, molecular weight marker. Lanes 2–8, Spanish isolates with strain FV7591 in lane 6. Lanes 9–11, French isolates [strain HBS1 (lane 9), strain HDE2 (lane 10) and strain TE1 (lane 11)]. A uniform profile was found among these isolates except for the Spanish strain FV7591 (CTX-M-1).

These findings suggested that most (82%) of the non-French isolates, like the 13 French isolates, represented a geographically dispersed, group B2-derived, serotype O25:H4, AmpC variant EC 6 clonal group of E. coli, characterized by CTX-M-15 and ciprofloxacin resistance.

MLST results

To more rigorously assess phylogenetic relationships within this collection, all 37 O25:H4 isolates (including the 36 CTX-M-15-positive isolates and the single CTX-M-1 isolate from Spain) underwent seven-locus MLST. Irrespective of geographical origin, the 36 CTX-M-15-positive O25:H4 isolates exhibited the same combination of alleles across the seven sequenced loci, corresponding to an established ST, ST131. In contrast, the Spanish CTX-M-1-positive O25:H4 isolate (phylogenetic group D, ERIC2 PCR profile 2, ciprofloxacin-susceptible) exhibited a novel combination of alleles and was assigned to a new ST, ST648. This confirmed the clonality and distinctness of the CTX-M-15 isolates.

PFGE profiles

Finer resolution of clonal relationships was obtained by PFGE analysis. Figure 2 shows PFGE analysis results for the 36 ST131 strains and the ST648 Spanish strain FV7591. The 36 ST131 strains constituted one large cluster (defined at the 68% similarity level), which was tied to the ‘outgroup’ strain FV7591 at <40% similarity. The ST131 cluster, in turn, comprised 17 separate PFGE groups, as defined at the 85% similarity level. These PFGE groups corresponded inconsistently with geographical origin. That is, the 13 French strains were classified into seven PFGE groups, the 6 Canadian strains into five groups, the 5 Portuguese strains into four groups and the 4 Lebanese strains into two groups. The six Spanish CTX-M-15-producing strains represented the only example of all isolates from a given country being classified into the same PFGE group. (Of note, the single Korean strain and the single Swiss strain were the sole representatives of their respective PFGE groups.) Likewise, multiple countries were represented within certain PFGE groups, including PFGE group I (France and Canada), group V (France, Canada and Portugal) and group XIII (France and Portugal). Nonetheless, frankly indistinguishable PFGE profiles were encountered only among strains from the same country, including two strains each from France (VB6 and HBS1), Lebanon (AH15 and AH10) and Spain (FV7569 and FV7595).

Figure 2

XbaI-PFGE dendrogram for 36 CTX-M-15-positive E. coli isolates from ST131 and a Spanish strain from ST648. The dendrogram for the 37 isolates, as produced by the UPGMA algorithm based on Dice similarity coefficients, included 18 PFGE groups, as defined based on ≥85% similarity of PFGE profiles.

Figure 2

XbaI-PFGE dendrogram for 36 CTX-M-15-positive E. coli isolates from ST131 and a Spanish strain from ST648. The dendrogram for the 37 isolates, as produced by the UPGMA algorithm based on Dice similarity coefficients, included 18 PFGE groups, as defined based on ≥85% similarity of PFGE profiles.

Virulence profiles

Extended virulence profiles were determined for the 36 ST131 isolates to assess the extent of within-group diversity and the virulence potential of the clonal group. Of the 46 virulence genes tested, 16 (35%) were detected in at least 1 isolate each. Isolates contained from 7 to 14 genes each (Table 2). Five different virulence genes were uniformly present in all 36 isolates, including fimH (type I fimbriae), sat (secreted auto-transporter toxin), fyuA (yersiniabactin receptor), usp (uropathogenic specific protein) and malX (pathogenicity island marker) (Table 2). Four other genes were present in >90% of the isolates, including iha (adhesin-siderophore receptor: 91%), kpsM II (group 2 capsule synthesis: 94%), iutA (aerobactin receptor: 97%) and ompT (outer membrane protease T: 97%) (Table 2), with the K5 and K2 kpsM II variants being detected in 53% and 39% of the isolates, respectively. Intermediate prevalence virulence genes included traT (serum resistance associated: 75%) and afa/draBC (afimbrial Dr-binding adhesins: 22%). In contrast, three genes occurred in <12% of strains each, typically together. These included hlyF (haemolysin F: 8%), iss (increased serum survival: 8%) and iroN (siderophore receptor: 11%) (Table 2).

Table 2

Virulence genotype of 36 CTX-M-15-producing Escherichia coli strains of clone ST131

 Adhesin Toxin Siderophore Protectin, invasin, pathogenicity island marker, miscellaneous 
Country/isolate afa/draBC iha fimH hlyF sat iroN fyuA iutA kpsM II K5 kps variant K2 kps variant usp traT ompT iss malX 
France                 
 MECB5 − − 
 VA1 − − − 
 Vlab2, VB8, HBS4, TNN (TE2), TE1, HDE1 − − − − − 
 VB9 − − − − − − − 
 HDE2 − − − − − − 
 HDE3 − − − − − − 
 VB6, HBS1 − − − − − 
Switzerland                 
 3756 EcS2 − − − − − 
Spain                 
 FV7561, FV7563 − − − − 
 FV7569, FV7588, FV7593, FV7595 − − − − − 
Portugal                 
 5753 − − − − − − − − − 
 5936 − − − − − − − − 
 5754, 5800, 6373 − − − − − 
Korea                 
 KUMC KN1604 − − − − − 
Lebanon                 
 AH9, AH8, AH15, AH10 − − − − − 
Canada                 
 1100, 8501, 16102 − − − − − 
 17102 − − − − − − 
 15802 − − − − 
 19502 − − 
 Adhesin Toxin Siderophore Protectin, invasin, pathogenicity island marker, miscellaneous 
Country/isolate afa/draBC iha fimH hlyF sat iroN fyuA iutA kpsM II K5 kps variant K2 kps variant usp traT ompT iss malX 
France                 
 MECB5 − − 
 VA1 − − − 
 Vlab2, VB8, HBS4, TNN (TE2), TE1, HDE1 − − − − − 
 VB9 − − − − − − − 
 HDE2 − − − − − − 
 HDE3 − − − − − − 
 VB6, HBS1 − − − − − 
Switzerland                 
 3756 EcS2 − − − − − 
Spain                 
 FV7561, FV7563 − − − − 
 FV7569, FV7588, FV7593, FV7595 − − − − − 
Portugal                 
 5753 − − − − − − − − − 
 5936 − − − − − − − − 
 5754, 5800, 6373 − − − − − 
Korea                 
 KUMC KN1604 − − − − − 
Lebanon                 
 AH9, AH8, AH15, AH10 − − − − − 
Canada                 
 1100, 8501, 16102 − − − − − 
 17102 − − − − − − 
 15802 − − − − 
 19502 − − 

afa/draBC, afimbrial Dr-binding adhesins; iha, adhesin-siderophore receptor; fimH, type I fimbriae; hlyF, haemolysin F; sat, secreted auto-transporter toxin; iroN, siderophore receptor; fyuA, yersiniabactin receptor; iutA, aerobactin receptor; kpsM II, group 2 capsule synthesis (variant K5 and K2); usp, uropathogenic specific protein; traT, serum resistance associated; ompT, outer membrane protease; iss, increased serum survival; malX, pathogenicity island marker.

Overall, virulence profile similarity among the 36 isolates was high, ranging from 91% to 100% (Figure 3). Only eight isolates exhibited a unique virulence profile. Indeed, 14 isolates (from Canada, France, Portugal, Korea and Switzerland) had an identical 11-gene virulence profile and four other groups of 2–6 strains each exhibited uniform VF profiles (Figure 3 and Table 2). Although some geographical segregation of virulence profiles was evident, virulence profiles corresponded inconsistently with PFGE type or locale (Figure 3), suggesting ongoing evolution of virulence genotypes.

Figure 3

Virulence profile dendrogram for 36 CTX-M-15-positive E. coli isolates from ST131. The dendrogram was produced by the UPGMA algorithm based on extended virulence gene profiles for the 36 strains from ST131. Virulence profile similarity varied from 91% to 100%.

Figure 3

Virulence profile dendrogram for 36 CTX-M-15-positive E. coli isolates from ST131. The dendrogram was produced by the UPGMA algorithm based on extended virulence gene profiles for the 36 strains from ST131. Virulence profile similarity varied from 91% to 100%.

Antimicrobial susceptibility patterns

To assess the multidrug resistance of the 36 ST131 isolates, susceptibility to non-β-lactam antimicrobials was tested (Table 1). Eighty-three per cent of the isolates were resistant to tetracycline, 77% to amikacin, 53% to co-trimoxazole and 50% to gentamicin, but only 0.3% to chloramphenicol.

Discussion

Our findings provide novel evidence of a recently emerged, broadly disseminated, CTX-M-15-positive E. coli clonal group as a cause of multidrug-resistant extraintestinal infections on at least three continents. This lineage exhibits a fairly robust virulence gene profile, implying substantial extraintestinal pathogenic potential. In most study locales, it accounted for a large proportion of ESBL-positive E. coli that were CTX-M-15 and/or O25-positive. The emergence of a new multidrug-resistant extraintestinal pathogen that may be spreading rapidly through the population while continuing to evolve appears to pose a significant public health threat in need of urgent attention.

The clonality of the ST131 strains is evident from their homogeneity with respect to phylogenetic group, seven-gene MLST allele combination and ERIC2 PCR profile. Clonality is further supported by the isolates’ uniform serotype (O25:H4), β-lactamase repertoire (CTX-M-15 and AmpC EC6) and ciprofloxacin phenotype, and their >90% similar virulence gene profiles. Furthermore, the considerable similarity of PFGE profiles observed among certain isolates indicates quite recent divergence from a common ancestor, whereas the occurrence in different locales of isolates with similar PFGE profiles suggests recent or ongoing transmission.

The virulence of the O25:H4-ST131 isolates can be inferred from two lines of evidence. First, most isolates were from samples submitted to clinical microbiology laboratories from inpatients and outpatients, so likely caused extraintestinal infections; this certainly was true for the blood and ascites isolates. Second, the number and types of virulence genes present in these strains (7–14 per isolate; coding for adhesins, siderophores, toxins, protectins and pathogenicity island markers) imply a robust virulence capability.29,31 Although these virulence profiles are not so extensive as those of typical antimicrobial-susceptible pathogenic E. coli from phylogenetic group B2, they nonetheless are more extensive than is usually observed among fluoroquinolone- or extended-spectrum cephalosporin-resistant E. coli, including human clinical isolates.32 Thus, these strains appear to pose the double threat of multidrug resistance (including to first-line therapeutic agents for Gram-negative infections) and substantial extraintestinal virulence capability. This makes their emergence and dissemination particularly concerning and suggests a need to identify their origins, reservoirs and transmission pathways so that appropriate interventions can be implemented. Better definition of the extent of this problem is needed, to clarify how great a public health threat these strains actually pose, so that resources can be allocated accordingly.

Dissemination of antimicrobial resistance genes by spread of the particular clone(s) in which they reside differs from the established paradigm for the emergence of ESBLs, which involves transfer of resistance-encoding plasmids rather than the host bacteria per se.9 However, clonal dispersal of drug-resistant pathogens has precedent in other species, such as methicillin-resistant S. aureus and penicillin-resistant S. pneumoniae.11,12 It also has been documented in E. coli, as exemplified by the localized outbreaks and international dissemination observed with multidrug-resistant clonal groups such as (group D-derived) ‘clonal group A’ and serotype O15:K52:H1, including the recent detection of clonal group A isolates in wastewater effluents from geographically dispersed areas of the United States.10,13,33 The present CTX-M-15-positive E. coli clonal group was previously shown to have caused what appeared to be localized outbreaks involving specific healthcare institutions (France) or geographical regions (Calgary).20,21 Our findings suggest that some of these seemingly isolated occurrences are actually linked, a principle that may apply broadly to drug-resistant extraintestinal infections. From this point of view, it would be relevant to determine whether the serogroup O25, CTX-M-15-positive E. coli previously published, notably in the UK, and not included in this study also belong to clone ST131.34 Recognition that geographically distant infection episodes may be caused by the same bacterial clone, arising from a common source, is the basis for the CDC's PulseNet surveillance system.35 Whereas that system focuses mainly on diarrhoeal pathogens, a similar system may be needed for extraintestinal infections.

In summary, we have characterized a broadly disseminated, CTX-M-15-positive, multidrug-resistant, virulent E. coli clonal group with highly homogeneous virulence genotypes and subgroups exhibiting highly similar PFGE profiles, suggesting recent emergence. Understanding how this clone has emerged and successfully disseminated within the hospital and community, including across national boundaries, should be a public health priority.

Funding

This study was supported by a grant (AOR 04016) from La Direction de le Recherche Clinique de l'Assistance Publique-Hôpitaux de Paris (M.-H. N.-C.), a grant (PI052023-PI051481) from Fondo de Investigacion Sanitaria (FIS), Instituto de Salud Carlos III, Spanish Ministerio de Sanidad y Consumo (J. B.) and Office of Research and Development, Medical Research Service, Department of Veterans Affairs (J. R. J.).

Acknowledgements

We are indebted to Professor Patrice Nordmann, Professor Guillaume Arlet and Dr Florence Doucet-Populaire for providing us with the Indian and Swiss strains, strain TNN (TE2) and the Lebanese strains, respectively. We also are grateful to Dr Azucena Mora, Dr Jesus Blanco, Dr Miguel Blanco, Mrs Ghizlane Dahbi and Mrs Cecilia Lopez for their contribution to this study.

Transparency declarations

All of the authors except one have none to declare. J. R. J. is a consultant for the following companies: Bayer, Ortho-McNeil, Merck, Wyeth-Ayerst, and Procter and Gamble.

References

1
Paterson
D
Bonomo
R
Extended-spectrum β-lactamases: a clinical update
Clin Microbiol Rev
 , 
2005
, vol. 
18
 (pg. 
657
-
86
)
2
Pitout
JD
Nordmann
P
Laupland
KB
, et al.  . 
Emergence of Enterobacteriaceae producing extended-spectrum β-lactamases (ESBLs) in the community
J Antimicrob Chemother
 , 
2005
, vol. 
56
 (pg. 
52
-
9
)
3
Canton
R
Coque
TM
The CTX-M β-lactamase pandemic
Curr Opin Microbiol
 , 
2006
, vol. 
9
 (pg. 
466
-
75
)
4
Bonnet
R
Growing group of extended-spectrum β-lactamases: the CTX-M enzymes
Antimicrob Agents Chemother
 , 
2004
, vol. 
48
 (pg. 
1
-
14
)
5
Gangoue-Pieboji
J
Miriagou
V
Vourli
S
, et al.  . 
Emergence of CTX-M-15-producing enterobacteria in Cameroon and characterization of a blaCTX-M-15 carrying element
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
441
-
5
)
6
Lavollay
M
Mamlouk
K
Frank
T
, et al.  . 
Clonal dissemination of a CTX-M-15 β-lactamase-producing Escherichia coli strain in the Paris area, Tunis and Bangui
Antimicrob Agents Chemother
 , 
2006
, vol. 
50
 (pg. 
2433
-
8
)
7
Livermore
DM
Canton
R
Gniadkowski
M
, et al.  . 
CTX-M: changing the face of ESBLs in Europe
J Antimicrob Chemother
 , 
2007
, vol. 
59
 (pg. 
165
-
74
)
8
Pallecchi
L
Malossi
M
Mantella
A
, et al.  . 
Detection of CTX-M-type β-lactamase genes in fecal Escherichia coli isolates from healthy children in Bolivia and Peru
Antimicrob Agents Chemother
 , 
2004
, vol. 
48
 (pg. 
4556
-
61
)
9
Fierer
J
Guiney
D
Extended-spectrum β-lactamases: a plague of plasmids
JAMA
 , 
1999
, vol. 
281
 (pg. 
563
-
4
)
10
Manges
AR
Johnson
JR
Foxman
B
, et al.  . 
Widespread distribution of urinary tract infections caused by a multidrug-resistant Escherichia coli clonal group
N Engl J Med
 , 
2001
, vol. 
345
 (pg. 
1007
-
13
)
11
McGee
L
McDougal
L
Zhou
J
, et al.  . 
Nomenclature of major antimicrobial-resistant clones of Streptococcus pneumoniae defined by the pneumococcal molecular epidemiology network
J Clin Microbiol
 , 
2001
, vol. 
39
 (pg. 
2565
-
71
)
12
Oliveira
DC
Alexander
T
de Lencastre
H
Secrets of success of a human pathogen: molecular evolution of pandemic clones of methicillin-resistant Staphylococcus aureus
Lancet Infect Dis
 , 
2002
, vol. 
2
 (pg. 
180
-
9
)
13
Prats
G
Navarro
F
Mirelis
B
, et al.  . 
Escherichia coli serotype O15:K52:H1 as a uropathogenic clone
J Clin Microbiol
 , 
2000
, vol. 
38
 (pg. 
201
-
9
)
14
Arpin
C
Coulange
L
Dubois
V
, et al.  . 
Extended-spectrum β-lactamase-producing Enterobacteriaceae in various types of private health care centers
Antimicrob Agents Chemother
 , 
2007
, vol. 
51
 (pg. 
3040
-
4
)
15
Prince
N
Leflon-Guibout
V
Doit
C
, et al.  . 
Genetic background of CTX-M-producing Escherichia coli isolates reveals a great strain diversity and the emergence of a clone in the community and the hospital
Abstracts of the Forty-sixth Interscience Conference on Antimicriobial Agents Chemotherapy
2006
San Francisco, CA
Washington, DC, USA
American Society for Microbiology
 
Abstract K-1508, p. 357
16
Karim
A
Poirel
L
Nagarajan
S
, et al.  . 
Plasmid-mediated extended-spectrum β-lactamase (CTX-M-3 like) from India and gene association with insertion sequence ISEcp1
FEMS Microbiol Lett
 , 
2001
, vol. 
201
 (pg. 
237
-
41
)
17
Kim
Y
Kim
S
Lee
J
, et al.  . 
Nosocomial transmission of CTX-M-15 and OXA-30 β-lactamase-producing Escherichia coli in a neurosurgical intensive care unit
Ann Clin Lab Sci
 , 
2005
, vol. 
35
 (pg. 
297
-
301
)
18
Mendonça
N
Leitao
J
Manageiro
V
, et al.  . 
Spread of clinical extended-spectrum β-lactamase (CTX-M)-producing Escherichia coli isolates in community and nosocomial environments in Portugal
Antimicrob Agents Chemother
 , 
2007
, vol. 
51
 (pg. 
1946
-
55
)
19
Moubareck
C
Daoud
Z
Hakimé
NI
, et al.  . 
Countrywide spread of community- and hospital-acquired extended-spectrum β-lactamase (CTX-M-15)-producing Enterobactericeae in Lebanon
J Clin Microbiol
 , 
2005
, vol. 
43
 (pg. 
3309
-
13
)
20
Pitout
JDD
Laupland
KB
Church
DL
, et al.  . 
Virulence factors of Escherichia coli isolates that produce CTX-M-type extended-spectrum β-lactamases
Antimicrob Agents Chemother
 , 
2005
, vol. 
49
 (pg. 
4667
-
70
)
21
Leflon-Guibout
V
Jurand
C
Bonacorsi
S
, et al.  . 
Emergence and spread of three clonally related virulent isolates of CTX-M-15-producing Escherichia coli with variable resistance to aminoglycosides and tetracycline in a French geriatric hospital
Antimicrob Agents Chemother
 , 
2004
, vol. 
48
 (pg. 
3736
-
42
)
22
Johnson
JR
O'Bryan
TT
Improved repetitive-element PCR fingerprinting for resolving pathogenic and nonpathogenic phylogenetic groups within Escherichia coli
Clin Diagn Lab Immunol
 , 
2000
, vol. 
7
 (pg. 
265
-
73
)
23
Clermont
O
Bonacorsi
S
Bingen
E
Rapid and simple determination of the Escherichia coli phylogenetic group
Appl Environ Microbiol
 , 
2000
, vol. 
66
 (pg. 
4555
-
8
)
24
Guinée
PAM
Jansen
WH
Wadström
T
, et al.  . 
Leeww
PW
Guinée
PAM
Escherichia coli associated with neonatal diarrhoea in piglets and calves
Laboratory Diagnosis in Neonatal Calf and Pig Diarrhoea: Current Topics in Veterinary and Animal Science
 , 
1981
The Hague, Netherlands
Martinus-Nijhoff
(pg. 
126
-
62
)
25
Clermont
O
Johnson
JR
Menard
M
, et al.  . 
Determination of Escherichia coli O types by allele-specific polymerase chain reaction: application to the O types involved in human septicemia
Diagn Microbiol Infect Dis
 , 
2007
, vol. 
57
 (pg. 
214
-
23
)
26
Tartof
SY
Solberg
OD
Manges
AR
, et al.  . 
Analysis of a uropathogenic Escherichia coli clonal group by multilocus sequence typing
J Clin Microbiol
 , 
2005
, vol. 
43
 (pg. 
5860
-
4
)
27
Blanco
M
Blanco
JE
Dahbi
G
, et al.  . 
Typing of intimin (eae) genes from enteropathogenic Escherichia coli (EPEC) isolated from children with diarrhoea in Montevideo, Uruguay: identification of two novel intimin variants (μB and ξR/β2B)
J Med Microbiol
 , 
2006
, vol. 
55
 (pg. 
1165
-
74
)
28
Carrico
JA
Pinto
FR
Simas
C
, et al.  . 
Assessment of band-based similarity coefficients for automatic type and subtype classification of microbial isolates analyzed by pulsed-field gel electrophoresis
J Clin Microbiol
 , 
2005
, vol. 
43
 (pg. 
5483
-
90
)
29
Johnson
JR
Kuskowski
MA
O'Bryan
TT
, et al.  . 
Epidemiological correlates of virulence genotype and phylogenetic background among Escherichia coli blood isolates from adults with diverse-source bacteremia
J Infect Dis
 , 
2002
, vol. 
185
 (pg. 
1439
-
47
)
30
Anonymous
Comité de l'Antibiogramme de la Société Française de Microbiologie report 2003
Int J Antimicrob Agents
 , 
2003
, vol. 
21
 (pg. 
364
-
91
)
31
Johnson
JR
Clermont
O
Menard
M
, et al.  . 
Experimental mouse lethality of Escherichia coli isolates in relation to accessory traits, phylogenetic group, and ecological source
J Infect Dis
 , 
2006
, vol. 
194
 (pg. 
1141
-
50
)
32
Johnson
JR
Kuskowski
MA
Owens
K
, et al.  . 
Phylogenetic origin and virulence genotype in relation to resistance to fluoroquinolones and/or extended-spectrum cephalosporins and cephamycins among Escherichia coli isolates from animals and humans
J Infect Dis
 , 
2003
, vol. 
188
 (pg. 
759
-
68
)
33
Boczek
LA
Rice
EW
Johnston
B
, et al.  . 
Occurrence of antibiotic-resistant uropathogenic Escherichia coli clonal group A in wastewater effluents
Appl Environ Microbiol
 , 
2007
, vol. 
73
 (pg. 
4180
-
4
)
34
Woodford
N
Ward
ME
Kaufmann
ME
, et al.  . 
Community and hospital spread of Escherichia coli producing CTX-M extended-spectrum β-lactamases in the UK
J Antimicrobial Chemother
 , 
2004
, vol. 
54
 (pg. 
735
-
43
)
35
Swaminathan
B
Gerner-Smidt
P
Ng
LK
, et al.  . 
Building PulseNet International: an interconnected system of laboratory networks to facilitate timely public health recognition and response to foodborne disease outbreaks and emerging foodborne diseases
Foodborne Pathog Dis
 , 
2006
, vol. 
3
 (pg. 
36
-
50
)