JAC Advance Access originally published online on April 7, 2006
Journal of Antimicrobial Chemotherapy 2006 57(6):1210-1214; doi:10.1093/jac/dkl127
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Phenotypic and genotypic characterization of antimicrobial resistance in Escherichia coli O111 isolates
1 Federal Institute for Risk Assessment (BfR), Diedersdorfer Weg 1 D-12277 Berlin, Germany 2 Universidade Federal de São Paulo, Escola Paulista de Medicina CEP 04023-062, São Paulo, Brazil
*Corresponding author. Tel: +49-30-8412-2057; Fax: +49-30-8412-2000; E-mail: b.guerra{at}bfr.bund.de
Received 1 December 2005; returned 31 January 2006; revised 15 March 2006; accepted 16 March 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Objectives: The aim of this study was to generate baseline data on the prevalence and molecular basis of antimicrobial resistance in Escherichia coli O111 isolates.
Methods: A total of 105 epidemiologically unrelated E. coli O111 isolates from humans and cattle (isolated between 1983 and 2003) were tested for susceptibility to 17 antimicrobial agents by broth microdilution. Resistant isolates were screened by molecular methods for resistance genes, class 1 and 2 integrons and mutations in the quinolone-resistance determining regions.
Results: Resistance was found in 76% of the isolates, with a prevalence of 72% for multiresistance. The most prevalent resistances were to streptomycin, sulfamethoxazole and tetracycline (72–68%), followed by spectinomycin, ampicillin and kanamycin/neomycin (39–25%). For each antimicrobial agent, the predominant resistance genes were ampicillin, blaTEM (94%); chloramphenicol, catA1 (100%); gentamicin, aac(3)-IV and aac(3)-II (50% each); kanamycin, aphA1 (100%); streptomycin, aadA1- like (66%); sulfamethoxazole, sul1 (59%); tetracycline, tet(A) (86%); and trimethoprim, dfrA1-like (83%). Class 1 integrons were found in 41% of the isolates. They carried aadA1, dfrA1-aadA1 and dfrA15-aadA1. A class 2 integron (dfrA1-sat1-aadA1) was found in one isolate. Only three isolates (3%) were resistant to nalidixic acid (reduced susceptibility to ciprofloxacin), with a single mutation in the gyrA gene.
Conclusions: E. coli O111 strains exhibit a wide repertoire of genetic elements to sustain antimicrobial pressure. Two specific antimicrobial resistance pheno/genotypes, [STR-SPT]-SUL-TET/aadA1-sul1-tet(A) and STR-SUL-TET-AMP-[KAN-NEO]/strA/B-sul2-tet(A)-blaTEM-aphA1, are predominant.
Keywords: STEC , molecular epidemiology , resistance genes , resistance determinants , integrons
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Over the past years, the emergence and spread of antimicrobial resistance has become a major public health concern. Several monitoring programmes have been initiated to generate baseline data about the prevalence of resistance in different bacterial species, including Escherichia coli.1 The genetic mechanisms that lead to bacterial resistance are manifold and their spread in different bacterial populations is enabled by highly efficient transfer systems of mobile genetic elements.2 During recent years, the importance of integrons (mobile gene expression systems) for the dissemination of resistance in E. coli has been established.3–7 The characterization of resistance mechanisms provides additional information about the epidemiology of resistant clones.1
Because of its epidemiological importance, the prevalence and nature of antimicrobial resistance in zoonotic Shiga toxin-producing E. coli (STEC) has been the subject of many studies.4–9 Diarrhoeal and other severe human diseases (e.g. haemolytic uraemic syndrome, HUS) caused by non-O157 STEC serotypes have increased worldwide.10 Previous studies have suggested that certain serotypes of STEC were more associated with antimicrobial drug resistance than others.4–9 Among these, multiresistant STEC O111 isolates were detected in humans and animals from several European and American countries.5–9 STEC O111 has previously been identified as a group of clonally related bacteria. On the basis of multilocus sequence analysis the O111:H8 strains were members of the enterohaemorrhagic E. coli EHEC-2 clone complex, together with serotypes O26:H11 and O118:[H16].4 We had previously shown that within O118:[H16] isolates antimicrobial resistance was widespread (98%).4 A high prevalence of resistance seems to be a characteristic of O111 (85–100%) and O26 (60–71%) strains as well.5,7,8 This study elucidates the prevalence and molecular basis of antimicrobial resistance in E. coli O111.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
A total of 105 epidemiologically unrelated E. coli O111 isolates (1983–2003) were investigated. Of these, 97 originated from humans with enteric disease, and 8 from cattle; 90 were isolated in 24 German diagnostic laboratories and 15 originated from 7 other countries (Table 1). All of them were maintained at the National Reference Laboratory for E. coli (BfR, Berlin). The isolates were serologically tested as described previously.10 Non-motile O111 isolates were investigated for their flagellar (fliC) genes by PCR–RFLP.10 The O111 isolates were also investigated for the production of Shiga toxins (Stx) by the Vero cell toxicity test and Stx-producing isolates investigated for their Stx-types by PCR-amplification.10
|
The MICs of 17 antimicrobial agents were determined using the CLSI (formerly the NCCLS) broth microdilution method.3 In the phenotypic analysis, isolates with intermediate MICs were not considered as resistant.
The detection of antimicrobial resistance genes, class 1 integrons and related genes, and mutations in the quinolone-resistance determining regions was performed by PCR-amplification and DNA-sequencing as described previously.3,4 All isolates showing full or intermediate resistance to a certain antimicrobial were tested for the corresponding resistance genes (Table 1). Those isolates that were PCR-negative for the selected genes were screened for additional resistance determinants. All resistant isolates were screened for the selected streptomycin-resistance genes.
Information on the isolates, primers and conditions used can be requested from the authors.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
Resistance was found in 76% of the isolates, with a prevalence of 72% for multiresistance (2–9 resistance determinants). The most prevalent resistances were to streptomycin, sulfamethoxazole and tetracycline (72–68%), followed by spectinomycin, ampicillin and kanamycin/neomycin (39–25%) and trimethoprim and trimethoprim/sulfamethoxazole (17%). Less than 10% were resistant to chloramphenicol, amoxicillin/clavulanic acid, gentamicin or nalidixic acid (8–3%). No resistance against ceftiofur (two isolates MICs = 1–2 mg/L), ciprofloxacin (three isolates MIC = 0.5 mg/L), colistin or florfenicol could be detected. The 80 drug-resistant isolates were grouped into 30 resistance phenotypes, with only 3 of them represented by more than 2 isolates (Table 1). A total of 86 isolates were STEC (92% resistant, 89% multiresistant) and 19 were non-STEC (63% resistant, 58% multiresistant). The most frequently observed multiresistance profiles [STR-SPE]-SUL-TET and STR-SUL-TET-AMP-[KAN-NEO] (23% and 16% of all isolates and 30% and 21% of the resistant isolates, respectively) were only associated with STEC O111 (most of them were H8 or non-motile carrying a fliC-H8 gene) but not with non-STEC O111 isolates (more heterogeneous with regard to their H-types). Two pheno/genotypes were only shown by Brazilian isolates, and the second most frequent pheno/genotype was only found among German isolates (Table 1). Despite the low number of O111 isolates from cattle, it was seen that cattle and human isolates shared three pheno/genotypes (Table 1).
It appears possible that multidrug-resistant STEC O111 strains have evolved very early and spread globally.5 In contrast, the acquisition of multiresistance in serotype O157 seems to be a recent event.5 The prevalence of specific phenotypes/genotypes seems to be related to different serotypes of STEC. In fact, the two most frequent phenotypes shown by the O111 isolates were more rarely (8–2%) found among the STEC O118 isolates analysed in a previous work.4 On the other hand, STEC O118 showed two major phenotypes [STR-(SPT)]-SUL-TET-AMP-CHL-[KAN-NEO] and [STR-(SPT)]-SUL-TET-TMP-SXT-AMP-CHL-[KAN-NEO]; both were present in 23% of the isolates and appeared only in 1% and 2% of the O111 isolates, respectively.
The genes encoding drug resistance appeared to be very homogeneous in the E. coli O111 group (Tables 1 and 2). In most of the cases one phenotype was due to a single genotype. For ampicillin, chloramphenicol and kanamycin a single gene (blaTEM, catA1 or aphA1, respectively) was responsible for the resistance in 95–100% of the isolates. Two genes each were found for gentamicin, streptomycin–spectinomycin, sulfamethoxazole, tetracycline and trimethoprim resistance (Table 2). Some of the genes (blaTEM, catA1, aphA1, aadA1-like, tet(A) and dfrA1-like) were widely spread (frequency > 60%) among the resistant isolates. Furthermore, in 16% of the resistant isolates, the same resistance was encoded by more than one gene [strA/B and aadA1-like in 10, sul1 and sul2 in 5, and tet(A) and tet(B) in 5 isolates].
|
For streptomycin resistance, silent genes in E. coli strains considered as streptomycin-susceptible have been described.6 Consequently we screened all resistant isolates for selected streptomycin resistance genes. One isolate carried the gene tandem strA/B disrupted by a dfrA14 gene. Such disrupted genes have been described for African E. coli isolates (AJ884725 [GenBank] and AJ313522 [GenBank] ). None of the Salmonella Genomic Island I genes (blaPSE-1 and tet(G) or floR genes), or the recently described sul3 gene, could be detected among the E. coli O111 isolates investigated. Both floR and sul3 have been found in other series of E. coli isolates (B. Guerra, unpublished data, 2005).2
In recent years, the increasing incidence of resistance to extended-spectrum ß-lactams and quinolones/fluoroquinolones has become a public health concern.3,9 Table 1 shows that only three German O111:H8 human isolates, carrying 7–11 resistances, were resistant to nalidixic acid (decreased susceptibility to ciprofloxacin, mutation in the gyrA gene affecting Ser-83). No extended-spectrum ß-lactamases or AmpC genes could be detected. This is in contrast to other studies on E. coli isolates from poultry in which a quinolone-resistance prevalence of 20–50% has been detected.3,9 For STEC O111 strains (also for O157, O118, O26, etc.) cattle are the main reservoir and the resistance patterns found among our series are related to the antimicrobial agents approved for bovine treatment.2
The multiple-resistance phenotypes observed are related to the spread of mobile genetic elements such as plasmids, transposons and integrons. Several class 1 and 2 integrons have been found in E. coli and they seem to be widely disseminated among O111 strains.5–7 Class 1 integrons were detected in 41% of the isolates (45% in STEC isolates, 21% in non-STEC isolates). They carried aadA1 (33 isolates), dfrA1-aadA1 (8 isolates), dfrA15-aadA1 (1 isolate) or no (1 isolate) gene cassettes in their variable region. Integrons carrying the aadA1 gene are the most widely spread among Enterobacteriaceae, and this is related to the dissemination of Tn21 transposon.3,5 All these isolates carried the class 1 integron-associated genes intI1, qacE
1 and sul1, except one isolate which carried a dfrA1-aadA1 integron lacking the sul1 gene. A class 2 integron carrying the dfrA1-sat1-aadA1 gene cassettes was found in one isolate. The MIC values of streptomycin for these isolates were dependent on the location of aadA genes in the integrons. strA/B-negative isolates (strA/B confers high resistance, MIC > 64 mg/L) carrying the aadA1 after a dfrA gene in a class 1 integron (three isolates) showed a low MIC of 32 mg/L. In contrast, the isolates that carried the aadA1-like genes not located on integrons (6 isolates), and most of the isolates (24 of 30) carrying aadA1 as a single gene cassette on the integrons, had high MICs (
64 mg/L).
The results from this study provide a solid background to follow the trends of multidrug resistance in E. coli O111. They give broad information on antimicrobial resistance in the E. coli O111 serotype, associating phenotypes with genotypes, and the presence of Stx toxins, H-types, etc. The data show that STEC O111 strains have a wide genetic repertoire to survive under antimicrobial pressure. This is a good example of bacteria which are normally not treated in human infections but are nevertheless highly resistant, indicating that resistance originates from other selective pressures. The data are relevant to public health and contribute to future risk assessment of antimicrobial resistance in zoonotic bacteria.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We thank B. Hoog (NRL-Salm, Federal Institute for Risk Assessment, BfR, Berlin) and S. Zimmermann (Robert Koch Institut, Berlin) for their very helpful technical assistance and H. Steinrück and G. Krause (NRL-E. coli, Federal Institute for Risk Assessment, BfR, Berlin) for their support. This work was supported by grants from the German Ministry of Consumer Protection and Agriculture (BMVEL, AZ:1000-WK-17/00) and the BfR (ref. 1322–136 and 1322–196).
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionTransparency declarationsReferences |
|---|
1 Aarestrup FM. (2004) Monitoring of antimicrobial resistance among food animals: principles and limitations. J Vet Med B Infect Dis Vet Public Health 51:380–8.[Web of Science][Medline]
2 Schwarz S and Chaslus-Dancla E. (2001) Use of antimicrobials in veterinary medicine and mechanisms of resistance. Vet Res 32:201–25.[CrossRef][Web of Science][Medline]
3
Guerra B, Junker E, Schroeter A, et al. (2003) Phenotypic and genotypic characterization of antimicrobial resistance in German Escherichia coli isolates from cattle, swine and poultry. J Antimicrob Chemother 52:489–92.
4
Maidhof H, Guerra B, Abbas S, et al. (2002) A multiple-anitibiotic-resistant clone of Shiga toxin-producing Escherichia coli O118:[H16] is spread in cattle and humans over different European countries. Appl Environ Microbiol 68:5834–42.
5 Morabito S, Tozzoli R, Caprioli A, et al. (2002) Detection and characterization of class 1 integrons in enterohemorrhagic Escherichia coli. Microb Drug Resist 8:85–91.[CrossRef][Web of Science][Medline]
6
Singh R, Schroeder CM, Meng J, et al. (2005) Identification of antimicrobial resistance and class 1 integrons in Shiga toxin-producing Escherichia coli recovered from humans and food animals. J Antimicrob Chemother 56:216–19.
7 White DG, Zhao S, McDermott PF, et al. (2002) Characterization of antimicrobial resistance among Escherichia coli O111 isolates of animal and human origin. Microb Drug Resist 8:139–46.[CrossRef][Web of Science][Medline]
8 Mora A, Blanco JE, Blanco M, et al. (2005) Antimicrobial resistance of Shiga toxin (Verotoxin)-producing Escherichia coli O157:H7 and non-O157 strains isolated from humans, cattle, sheep and food in Spain. Res Microbiol 156:793–806.[Medline]
9 Schroeder CM, Meng J, Zhao S, et al. (2002) Antimicrobial resistance of Escherichia coli O26, O103, O111, O128, and O145 from animals and humans. Emerg Infect Dis 8:1409–14.[Web of Science][Medline]
10
Beutin L, Krause G, Zimmermann S, et al. (2004) Characterization of Shiga toxin-producing Escherichia coli strains isolated from human patients in Germany over a 3-year period. J Clin Microbiol 42:1099–108.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Bibbal, V. Dupouy, M. F. Prere, P. L. Toutain, and A. Bousquet-Melou Relatedness of Escherichia coli Strains with Different Susceptibility Phenotypes Isolated from Swine Feces during Ampicillin Treatment Appl. Envir. Microbiol., May 15, 2009; 75(10): 2999 - 3006. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ziebell, M. Steele, Y. Zhang, A. Benson, E. N. Taboada, C. Laing, S. McEwen, B. Ciebin, R. Johnson, and V. Gannon Genotypic Characterization and Prevalence of Virulence Factors among Canadian Escherichia coli O157:H7 Strains Appl. Envir. Microbiol., July 15, 2008; 74(14): 4314 - 4323. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Nwaneshiudu, T. Mucci, D. J. Pickard, and I. N. Okeke A Second Large Plasmid Encodes Conjugative Transfer and Antimicrobial Resistance in O119:H2 and Some Typical O111 Enteropathogenic Escherichia coli Strains J. Bacteriol., August 15, 2007; 189(16): 6074 - 6079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Beutin and E. Strauch Identification of Sequence Diversity in the Escherichia coli fliC Genes Encoding Flagellar Types H8 and H40 and Its Use in Typing of Shiga Toxin-Producing E. coli O8, O22, O111, O174, and O179 Strains J. Clin. Microbiol., February 1, 2007; 45(2): 333 - 339. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||