Kaohsiung J Med Sci June 2007 • Vol 23 • No 6 273
The rapid spread of antimicrobial-resistant bacteria is an alarming and increasing problem which often com-plicates treatment of infections. Such complications often result from rapid dissemination of antibiotic-resistant genes carried by plasmids, transposons, and integrons [1,2]. Integrons are potentially mobile genetic elements frequently located on transposons and have been identified at loci in which site-specific incorpora-tion and excision of gene cassettes frequently occur [3].
Most of the integrons characterized to date contain gene cassettes coded for resistance to antimicrobial agents. Since integrons can act as natural expression vectors for any gene cassettes inserted, recent studies have examined the role of integrons in the carriage and dissemination of antimicrobial resistance genes [4].
As an increasing number of bacterial isolates dem-onstrate resistance to a wide spectrum of antimicrobial agents [5] in both clinical and community settings, the dissemination of antimicrobial-resistant genes between bacteria is of great concern. Several studies have doc-umented the widespread prevalence of integrons in clinical isolates [6–9]. However, the stability of these gene cassettes after insertion into integrons requires further investigation to clarify the development of resistance.
Received: August 1, 2006 Accepted: December 6, 2006 Address correspondence and reprint requests to: Professor Lin-Li Chang, Department of Microbiology, Kaohsiung Medical University, 100 Shih-Chuan 1stRoad, Kaohsiung 807, Taiwan. E-mail: [email protected]
V
ARIABLE
G
ENE
C
ASSETTE
P
ATTERNS OF
C
LASS
1
I
NTEGRON
-
ASSOCIATED
D
RUG
-
RESISTANT
E
SCHERICHIA COLI IN
T
AIWAN
Lin-Li Chang, Tsung-Ming Chang, and Chung-Yu Chang
Department of Microbiology, Faculty of Medicine, Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan.
This study characterized class 1 integrons in Escherichia coli in Taiwan. The stability and changes in gene cassettes inserted into integrons were also evaluated. The study included 436 clinical strains of E. coli isolated in 2002. Class 1 integrons were characterized by polymerase chain reaction and direct sequencing. Genetic localization of class 1 integrons was determined by conjugal transfer and Southern hybridization. The results indicated that 64% of E. coli isolates carried class 1 inte-grons. Molecular analysis revealed that the class 1 integrons harbored 13 different antimicrobial resistance gene cassettes and two unknown gene cassettes; the predominant cassettes were aadA and dfrA. Novel gene cassettes first recovered from E. coli were aacA4 and linF. Cassette arrays
orfD-aacA4-catB8 and aadA1-linF were also observed. Gene cassette dfrA12-orfF-aadA2 was stable.
The class 1 integron and dfrA17-aadA5 gene cassette were located on the same transferable plas-mids and were capable of transmission. Therefore, the increased drug resistance of clinical iso-lates may be explained by antibiotic selective pressure and widespread presence of integrons. Under antibiotic selective pressure, gene cassette-mediated resistance may not be easily lost. The potential role of integrons in the uptake and dissemination of resistance genes by plasmid between species of bacteria may decrease the therapeutic effectiveness of antibiotics.
Key Words:aadA1-linF, conjugative plasmid, dfrA17-aadA5, Escherichia coli, orfD-aacA4-catB8
In Taiwan, Escherichia coli has shown a high overall rate of resistance to commonly used “first-line” antibio-tics including ampicillin, cephalothin, gentamicin, and trimethoprim/sulfamethoxazole [10]. Because of this selective pressure, it would be of interest to character-ize the antimicrobial-resistant gene cassettes located on the class 1 integrons of E. coli isolated in 2002. Fur-thermore, in comparison with our previous study [11], the stability and changes in gene cassettes over a 10-year period in Taiwan were also evaluated.
M
ATERIALS ANDM
ETHODSClinical isolates and antimicrobial
susceptibility profiles
Four hundred and thirty-six E. coli strains were ob-tained from 436 inpatients at Kaohsiung Medical University Hospital in southern Taiwan in 2002. The
E. coli strains were isolated from clinical specimens
taken by various hospital departments, including urine (n=233), pus (n=85), blood (n=45), sputum (n=25), bile (n= 8), genital tract samplings (n = 10), and other sam-ples (n=30). Once identified, the isolates were preserved at −70°C in Tryptic Soy Broth (Difco Laboratories, Detroit, MI, USA) containing glycerol (15%, v/v).
Antimicrobial susceptibility was analyzed by disk diffusion methods in Mueller-Hinton agar (Difco Laboratories). Results were interpreted according to CLSI criteria [12]. E. coli strain ATCC 25922 was used as a control strain.
Incidence of class 1 integron
The template DNA for polymerase chain reaction (PCR) were prepared as described by Bass et al [13]. Integrons were detected by PCR with primers intI1 (F: 5′-CCTCCCGCACGATGATC-3′, R: 5′-TCCACG-CATCGTCAGGC-3′), which hybridize to conserved regions of integron-encoded integrase genes intI1 [14]. The PCR reactions were performed at 94°C for 1 minute and then for 30 cycles at 94°C for 30 seconds, 58°C for 20 seconds and 72°C for 30 seconds. Plasmid pUB2401, which harbors transposon Tn21 containing the In2 class 1 integron, was used as control [1].
DNA sequencing
To determine the possibility of inserting of antimicro-bial-resistant gene cassettes, primers specific for the 5′-conserved (5′-GGCATCCAAGCAGCAAG-3′) and
3′-conserved segment (5′-AAGCAGACTTGACCTGA-3′) were used to amplify the entire integron cassette insertion region. The PCR products were purified using the Wizard PCR kit (Promega, Madison, WI, USA) and used as templates for direct nucleotide sequencing. The Prism Ready Reaction DyeDeoxy Termination Cycle Sequencing kit (Applied Biosystems, Foster City, CA, USA) was used according to the manu-facturer’s instructions. The samples were run on a 377 DNA sequencer (Applied Biosystems), and the sequences were analyzed by Sequence Navigator software (Applied Biosystems).
Conjugation, plasmid DNA isolation, and
Southern hybridization
Five clinical isolates carrying class 1 integron and
dfrA17-aadA5 cassette were randomly selected for use
as a donor; E. coli K12 20R 764 Rifrwas used as the recipient for the conjugation experiment [15]. In accor-dance with the method described by Kado and Liu [16], plasmid DNA was extracted, electrophoresed on 1% agarose gels, then blotted onto nylon membrane. The membranes were hybridized with digoxigenin-labeled probe and then auto-radiographed in accor-dance with the manufacturer’s instructions (Boehringer Mannheim, Mannheim, Germany). Two DNA probes were amplified by PCR. Primers 5 ′-AGTGTCAAA-GAACGGAATTTCAAGCTCA-3′ and 5′-GGATAGCG-CCAAGGCACTAC-3′, including part of the dfrA17 gene and the aadA5 gene were used for dfrA17-aadA5 cassette detection. A 727 bp fragment was amplified. The amplification reaction consisted of 35 cycles of 1 minute of denaturation at 94°C, 30 seconds of anneal-ing at 58°C, and 90 seconds of extension at 72°C. The other probe used was integrase gene intI1.
R
ESULTSThe percentage of isolates containing class 1 specific integron was 64% (280/436). The integron cassette region could not be amplified by PCR in 54 of the class 1 integron-containing isolates. No cassette inserts were detected in two isolates (Figure 1, lane 9); however, the other 224 strains of E. coli carried gene cassettes con-ferring antimicrobial resistance. Of the 224 isolates, 215 yielded one amplicon, and nine yielded two amplicons of different sizes (Table 1). The amplicon lengths pres-ent within each integron ranged in size from 700 bp
to 3 kb (Figure 1). The integrons were classified into nine groups according to the length and the numbers of amplicons yielded by a single isolate (Table 1). Groups 1, 2, 6 and 7 were further divided into sub-groups based on the kinds of gene cassette types
identified. As Table 1 illustrates, 15 different gene cas-settes were found, including genes encoding resistance to aminoglycosides (aadA1, aadA2, aadA5, aadB, aacA4), chloramphenicol (cmlA, catB8), trimethoprim (dfrA1,
dfrA5, dfrA7, dfrA12, dfrA17), lincosamide (linF) and
unknown genes (orfD, orfF). The most common type of cassette, aadA, carried by class 1 integrons were those conferring resistance to streptomycin and spectinomycin, which represented 92% (207/224) of all cassettes. In the second most prevalent cassette,
dfrA, resistance to trimethoprim was observed in 76%
(170/224). In the present study, novel gene cassettes
aacA4 and linF were recovered, and cassette arrays orfD-aacA4-catB8 and aadA1-linF were found.
The dfrA12-orfF-aadA2 cassette array conferring re-sistance to streptomycin/spectinomycin and to tri-methoprim was frequently found in class 1 integrons containing clinical isolates of E. coli (33%, 74/224) (Table 1). The cassette array dfrA17-aadA5 was found in 81 (36%) class 1-containing isolates (Table 1). Southern hybridization with probes specific for the intI1 and
dfrA17-aadA5 genes revealed that class 1 integron
con-taining dfrA17-aadA5 genes were located on the same plasmids and conjugatively transferable, ranging in size from 126 kb to 168 kb (Figure 2, lanes 2, 4, 6). Additionally, nonconjugative plasmid from isolates
100 bp 200 bp 300 bp 500 bp 1,000 bp 1,500 bp 3,000 bp M 1 2 3 4 5 6 7 8 9
Figure 1.Polymerase chain reaction (PCR) amplification of the integron-variable regions with 5′-CS and 3′-CS primers. The amplicons were separated by electrophoresis through an agarose 2% gel. Lane M, size marker; lanes 1–9, amplicons from clinical E. coli isolates. Lane 9, no cassette inserts into integron when 150 bp PCR product was found.
Table 1.Amplicons and corresponding gene cassettes inserted in class 1 integrons and antibiotic resistance phenotype of gene cassettes of Escherichia coli isolates in 2002
Integron Approximate Identity and order of Resistance Number of isolates length of
group
amplicons (kb) inserted gene cassettes phenotype in 2002
1a 1 aadA1 STR 26 1b 1 aadA2 STR 5 2a 1.6 dfrA1-aadA1 TMP-STR 7 2b 1.6 dfrA17-aadA5 TMP-STR 72 3 1.9 dfrA12-orfF-aadA2 TMP-STR 69 4 3 aadB-aadA1-cmlA GEN-STR-CHL 18 5 1.6 and dfrA17-aadA5 TMP-STR 5 1.9 dfrA12-orfF-aadA2 TMP-STR 6a 0.7 aadB GEN 2 6b 0.7 dfrA5 TMP 6 6c 0.7 dfrA7 TMP 7 7a 2 aadA1-linF STR-LIM 1 7b 2 orfD-aacA4-catB8 GEN-CHL 2 8 1 and aadA1 STR 4 1.6 dfr17-aadA5 TMP-STR 9 0.15 Empty integron – 2 Unidentified – 54 Total 280
(Figure 2, lane 5) was also hybridized with the
dfr17-aadA5 and intI1 specific probes.
Resistance of clinical isolates to almost every class of antimicrobial agent was noted. The resistance to ampicillin and extended-spectrum penicillin (e.g. pi-peracillin) was 87% and 67%, respectively. Addition-ally, resistance to penicillins combined with β-lactamase
inhibitors was also observed, e.g. amoxicillin/clavu-lanic acid (21%) and ticarcillin/clavuamoxicillin/clavu-lanic acid (15%). However, resistance to third generation cephalosporins was observed in less than 10% of the isolates (Table 2). Integron-positive clinical isolates tended to have greater antibiotic resistance than integron-negative clinical isolates. Interestingly, integrons containing isolates
A 1 2 3 4 5 6 B 1 2 3 4 5 6
119 kb 148 kb
Figure 2. (A) Agarose gel electrophoresis and (B) Southern hybridization of plasmid DNA of clinical E. coli isolates. Southern hybridization was performed with two probes specific for the intI1 and dfrA17-aadA5 gene, respectively. Lane 1, 119 kb conjugative plasmid pEC1072 containing dfrA17-aadA5 was used as positive control; lanes 2–6, E. coli clinical isolates. The two probes were hybridized to the same conjugative plasmids in lanes 2, 4, 6, and to the nonconjugative plasmid in lane 5.
Table 2.Association between antibiotic resistance and integrons in Escherichia coli isolates
Antibiotic Integron (+) (n = 280) Integron (−) (n = 156) p*
% (no. of resistant isolates) % (no. of resistant isolates)
Trimethoprim 90 (251) 30 (48) < 0.01 Streptomycin 94 (263) 70 (109) < 0.01 Chloramphenicol 74 (206) 35 (55) < 0.01 Ampicillin 95 (265) 72 (113) < 0.01 Amikacin 5 (14) 0.7 (1) NS Aztreonam 12 (35) 0.7 (1) < 0.01 Ceftazidime 11 (32) 0 (0) < 0.01 Gentamicin 17 (48) 0.6 (1) < 0.01 Moxalactam 10 (29) 0 (0) < 0.01 Cefalothin 13 (38) 4 (5) < 0.01 Trimethoprim/sulfamethoxazole 86 (241) 18 (28) < 0.01 Piperacillin 80 (223) 44 (68) < 0.01 Cefoperazone 18 (50) 4 (6) < 0.01 Cefazolin 24 (67) 10 (15) < 0.01 Amoxicillin/clavulanic acid 28 (77) 10 (16) < 0.01 Ceftriaxone 13 (39) 1 (2) < 0.01 Imipenem 0.3 (1) 0 (0) NS Ticarcillin/clavulanic acid 20 (56) 6 (9) < 0.01 Ofloxacin 27 (75) 3 (5) < 0.01 Piperacillin/tazobactam 3 (8) 0 (0) NS Cefepime 2 (7) 0.7 (1) NS All sensitive† 0 (0) 10 (15) < 0.01
had a significantly high resistance to most of the antibiotics except amikacin, imipenem, cefepime, and piperacillin/tazobactam (Table 2).
D
ISCUSSIONThe increasing drug resistance of clinical isolates may be explained by antibiotic selective pressure and wide-spread presence of integrons. This study documented an increased prevalence of class 1 integrons from 52% [11] to 64% (this study) in E. coli isolates recovered in Taiwan over a 10-year period. Increased prevalence of the class 1 integron has also been reported else-where: 43% in Norway in 2000–2001 [17], 43% in Western and Central Europe in 1996–1997 [8], 50% in The Netherlands in 1994 [18], 59% in France in 1992 [7], 54% in Korea in 1980–2002 [19], and 86% in China in 2005 [20].
In the present study, most of the isolates with class 1 integrons contained at least one aadA cassette (aadA1, aadA2 or aadA5), alone or in combination with other cassettes. However, the predominance of gene cassettes encoding resistance to streptomycin and spectinomycin (aadA) was unanticipated, as the clinical use of the drugs in Taiwan has been minimal. However, streptomycin-resistant bacteria can be iso-lated from animals, probably as a result of strepto-mycin and spectinostrepto-mycin use in animal husbandry [21–25]. Furthermore, coliform bacteria isolated from the aquatic environment have also revealed a resis-tance to aminoglycosides [26]. It is therefore likely that humans became colonized with streptomycin-resistant bacteria via the food chain in a contami-nated environment [27–29]. Another possibility is that integrons transferred from animal E. coli to human
E. coli while transiently passing through the human
intestine. A third possibility is that even when anti-biotics cease to be used therapeutically, genes encoding resistance to these antibiotics were not easily lost [30].
aadA was the most common cassette carried by
class 1 integron; however, cassette aadA1 apparently decreased gradually during the 10-year period in Taiwan. Similar results have been reported in E. coli isolates in Korea [19]. Nevertheless, multigene cassette array aadB-aadA1-cmlA was found to increase from 2% in 1993 to 6% in 2002 in isolates in this study. The trimethoprim-resistant (dfrA) gene cassette was also fre-quently found in the E. coli strains under investigation.
Five dfrA gene type cassettes were identified; these gene cassettes were not unexpected since trimethoprim-sulfamethoxazole is commonly used to treat bacterial infections in Taiwan. Such specific selective pres-sure may favor the acquisition and maintenance of a trimethoprim-resistant cassette by class 1 integrons containing sul1 in the 3′ region.
Gene cassette aacA4, which codes for amikacin, netilmicin and tobramycin resistance was previously found in pan-resistant Gram-negative clinical iso-lates of Pseudomonas aeruginosa [31] and Acinetobacter
baumannii [32]. Multigene cassette arrays aacA4-catB8
found only in A. baumannii [32,33] were also found in
E. coli in the present study. Although amikacin resistant E. coli was rarely isolated (3%, Table 2) from the
clini-cal samples, class 1 integron carrying orfD-aacA4-catB8 gene cassette was found in two isolates. An additional finding was a novel aadA1-linF cassette which differed from the aadA2-linF cassette previously identified in a Norway study [17]. These results indicate that resist-ant gene cassettes might disseminate through the inte-gron between different bacteria species under antibiotic selective pressure. Furthermore, the same cassette may combine with other cassettes in different geographi-cal areas. Most of the class 1 integrons identified in this study carried the dfr12-orfF-aadA2 cassette array. This pattern was also reported in urinary E. coli iso-lates in a Korean study [19,22], and in Shigella strains isolated in Finland but originating from Asia [34]. This observation indicates that this combination of gene cassettes has achieved stable integration.
Furthermore, dfr17-aadA5 located on a 119 kb con-jugative plasmid pEC1072 was isolated from an E. coli strain in 1993 in our previous study [35]. The quanti-tative presence of the same cassette arrays with the predominant class 1 integrons found on transferable plasmids in E. coli rapidly increased in 2002 in this work. The same dfrA17-aadA5 cassettes were also de-tected in clinical isolates in Korea [19,22] and Australia [36] as well as in E. coli isolated from dogs, pigs, and other domestic livestock [28,29]. These data reveal that the dfrA17-aadA5 cassette can disseminate by self-transferable plasmids in humans or animals and in different areas worldwide.
Of the class 1 integron-containing isolates recovered in 2002, the integron cassette region could not be am-plified by PCR in 54 isolates. Possible causes are: (1) the number of inserted genes in the cassette exceeded the PCR extension capacity; (2) lack of the 3′ conserved
segment or insufficient homology to the 3′ conserved segment primer to produce a product [37].
Previous studies have not clarified whether impru-dent or inappropriate use of first-line antimicrobials in Taiwan promotes cassette array formation. How-ever, this study revealed a steady increasing preva-lence of antibiotic resistant E. coli in Taiwan over a 10-year period. It is suggested that this increased resis-tance is partly attributable to the acquisition, dissem-ination and stable maintenance of a class 1 integron.
In conclusion, the extended survey period revealed a predominance of dfrA and aadA resistant gene cas-settes conferring resistance to trimethoprim and amino-glycoside. However, genes conferring resistance to recently introduced antibiotics which were already part of the gene cassettes of pan-drug resistant bacte-ria were found in E. coli. Therefore, the potential role of integrons in the uptake and dissemination of resis-tant genes may be a continuing threat to the effective-ness of certain antibiotic therapeutic agents both in Taiwan and globally.
A
CKNOWLEDGMENTSThis work was supported by a grant from the National Science Council (91-2320-B-037-058).
R
EFERENCES1. Hall RM, Collis CM. Mobile gene cassettes and inte-grons: capture and spread of genes by site-specific recombination. Mol Microbiol 1995;15:593–600.
2. Normark BH, Normark S. Evolution and spread of antibiotic resistance. J Intern Med 2002;252:91–106. 3. Recchia GD, Hall RM. Gene cassettes: a new class of
mobile element. Microbiology 1995;141:3015–27. 4. Sallen B, Rajoharison A, Desvarenne S, et al. Molecular
epidemiology of integron-associated antibiotic resistance genes in clinical isolates of Enterobacteriaceae. Microb
Drug Resist 1995;1:195–202.
5. Neu HC. The crisis in antibiotic resistance. Science 1992; 257:1064–73.
6. Gonzalez G, Sossa K, Bello H, et al. Presence of inte-grons in isolates of different biotypes of Acinetobacter
baumannii from Chilean hospitals. FEMS Microbiol Lett
1998;161:125–8.
7. Hamada K, Oshima K, Tsuji H. Drug resistance genes encoded in integrons and in extra-integrons: their dis-tribution and lateral transfer among pathogenic Entero-bacteriaceae including enterohemorrhagic Escherichia coli
and Salmonella enterica serovars typhimurium and infantis.
Jpn J Infect Dis 2003;56:123–6.
8. Martinez-Freijo P, Fluit AC, Schmitz FJ, et al. Class I integrons in Gram-negative isolates from different European hospitals and association with decreased susceptibility to multiple antibiotic compounds. J
Anti-microb Chemother 1998;42:689–96.
9. Martinez-Freijo P, Fluit AC, Schmitz FJ, et al. Many class I integrons comprise distinct stable structures occurring in different species of Enterobacteriaceae iso-lated from widespread geographic regions in Europe.
Antimicrob Agents Chemother 1999;43:686–9.
10. Lauderdale TL, Clifford McDonald L, Shiau YR, et al. The status of antimicrobial resistance in Taiwan among gram-negative pathogens: the Taiwan surveillance of antimicrobial resistance (TSAR) program, 2000. Diagn
Microbiol Infect Dis 2004;48:211–9.
11. Chang CY, Chang LL, Chang YH, et al. Characterisation of drug resistance gene cassettes associated with class 1 integrons in clinical isolates of Escherichia coli from Taiwan, ROC. J Med Microbiol 2000;49:1097–102. 12. Clinical and Laboratory Standards Institute. Performance
Standards for Antimicrobial Disk Susceptibility Tests; Approv-ed Standard M2-A8, 8thedition. Wayne, PA: Clinical and
Laboratory Standards Institute, 2003.
13. Bass L, Liebert CA, Lee MD, et al. Incidence and characterization of integrons, genetic elements mediat-ing multiple-drug resistance, in avian Escherichia coli.
Antimicrob Agents Chemother 1999;43:2925–9.
14. Zhao S, White DG, Ge B, et al. Identification and char-acterization of integron-mediated antibiotic resistance among Shiga toxin-producing Escherichia coli isolates.
Appl Environ Microbiol 2001;67:1558–64.
15. Lin SR, Chang SF. Drug resistance and plasmid profile of shigellae in Taiwan. Epidemiol Infect 1992;108:87–97. 16. Kado CI, Liu ST. Rapid procedure for detection and
isolation of large and small plasmids. J Bacteriol 1981; 145:1365–73.
17. Heir E, Lindstedt BA, Leegaard TM, et al. Prevalence and characterization of integrons in blood culture Enterobacteriaceae and gastrointestinal Escherichia coli in Norway and reporting of a novel class 1 integron-located lincosamide resistance gene. Ann Clin Microbiol
Antimicrob 2004;3:12.
18. Jones ME, Peters E, Weersink AM, et al. Widespread occurrence of integrons causing multiple antibiotic resis-tance in bacteria. Lancet 1997;349:1742–3.
19. Wu HS, Lee JC, Kang HY, et al. Changes in gene cas-settes of class 1 integrons among Escherichia coli isolates from urine specimens collected in Korea during the last two decades. J Clin Microbiol 2003;41:5429–33.
20. Su J, Shi L, Yang L, et al. Analysis of integrons in clinical isolates of Escherichia coli in China during the last six years. FEMS Microbiol Lett 2006;254:75–80.
21. Du X, Shen Z, Wu B, et al. Characterization of class 1 integrons-mediated antibiotic resistance among calf pathogenic Escherichia coli. FEMS Microbiol Lett 2005; 245:295–8.
22. Kang HY, Jeong YS, Oh JY, et al. Characterization of antimicrobial resistance and class 1 integrons found in
Escherichia coli isolates from humans and animals in
Korea. J Antimicrob Chemother 2005;55:639–44.
23. Lindstedt BA, Heir E, Nygard I, et al. Characterization of class I integrons in clinical strains of Salmonella
enter-ica subsp. enterenter-ica serovars typhimurium and Enteritidis
from Norwegian hospitals. J Med Microbiol 2003;52: 141–9.
24. McDonald LC, Chen MT, Lauderdale TL, et al. The use of antibiotics critical to human medicine in food-producing animals in Taiwan. J Microbiol Immunol Infect 2001;34:97–102.
25. Ridley A, Threlfall EJ. Molecular epidemiology of antibiotic resistance genes in multiresistant epidemic
Salmonella typhimurium DT 104. Microb Drug Resist
1998;4:113–8.
26. Park JC, Lee JC, Oh JY, et al. Antibiotic selective pres-sure for the maintenance of antibiotic resistant genes in coliform bacteria isolated from the aquatic environment.
Water Sci Technol 2003;47:249–53.
27. Guerra B, Junker E, Schroeter A, et al. Phenotypic and genotypic characterization of antimicrobial resist-ance in German Escherichia coli isolates from cattle, swine and poultry. J Antimicrob Chemother 2003;52: 489–92.
28. Sanchez S, McCrackin Stevenson MA, Hudson CR, et al. Characterization of multidrug-resistant Escherichia
coli isolates associated with nosocomial infections in
dogs. J Clin Microbiol 2002;40:3586–95.
29. Wegener HC, Aarestrup FM, Jensen LB, et al. Use of antimicrobial growth promoters in food animals and
Enterococcus faecium resistance to therapeutic
antimicro-bial drugs in Europe. Emerg Infect Dis 1999;5:329–35. 30. Collis CM, Hall RM. Gene cassettes from the insert
region of integrons are excised as covalently closed circles. Mol Microbiol 1992;6:2875–85.
31. Pagani L, Colinon C, Migliavacca R, et al. Nosocomial outbreak caused by multidrug-resistant Pseudomonas
aeruginosa producing IMP-13 metallo-beta-lactamase. J Clin Microbiol 2005;43:3824–8.
32. Turton JF, Kaufmann ME, Glover J, et al. Detection and typing of integrons in epidemic strains of Acinetobacter
baumannii found in the United Kingdom. J Clin Microbiol
2005;43:3074–82.
33. Wu TL, Ma L, Chang JC, et al. Variable resistance patterns of integron-associated multidrug-resistant Acinetobacter
baumannii isolates in a surgical intensive care unit. Microb Drug Resist 2004;10:292–9.
34. Heikkila E, Skurnik M, Sundstrom L, et al. A novel dihydrofolate reductase cassette inserted in an inte-gron borne on a Tn21-like element. Antimicrob Agents
Chemother 1993;37:1297–304.
35. Chang CY, Chang LL, Chang YH, et al. Two new gene cassettes, dfr17 (for trimethoprim resistance) and aadA4 (for spectinomycin/streptomycin resistance), inserted in an Escherichia coli class 1 integron. J Antimicrob
Chemother 2000;46:87–9.
36. White PA, McIver CJ, Deng Y, et al. Characterisation of two new gene cassettes, aadA5 and dfrA17. FEMS
Microbiol Lett 2000;182:265–9.
37. White PA, McIver CJ, Rawlinson WD. Integrons and gene cassettes in the Enterobacteriaceae. Antimicrob Agents
!"VR= =U==N= !"VS==NO==S= !"#$%&' !"!#$%!& UMT !"#$NMM
