Molecular epidemiology of quinolone resistance in Acinetobacter spp.

ORIGINAL ARTICLE

Molecular epidemiology of quinolone resistance in Acineto bacter spp. Rebecca J. Seward and Kevin J. Towner D e p a r t m e n t of Microbiology a n d PHLS Laboratory, University Hospital, Queen’s M e d i c a l C e n t r e , N o t t i n g h a n i , UK

Objective: To determine whether similar mutations to quinolone resistance in the gyrA subunit of DNA gyrase and the parC subunit of topoisomerase IV are occurring independently in genotypically unrelated clinical isolates of Acinetobacter spp., or whether worldwide clonal spread of particular resistant strains is occurring. Methods: The genotypic relationships of 25 nosocomial isolates of Acinetobacterspp. from 15 locations in 11 different countries worldwide were examined by randomly amplified polymorphic DNA analysis. Quinolone resistancedetermining regions of gyrA and parCwere amplified by PCR and mutations were analyzed by restriction digestion with Hinfl and DNA sequencing. Results: Twenty-four of the 25 Acinetobacter isolates were genotypically heterogeneous and 12 were resistant to both nalidixic acid and ciprofloxacin. Analysis of conserved gyrA and parC regions showed that all isolates with a ciprofloxacin MIC of 4 mg/L had a substitution of Ser83 with either Leu or Phe i n the GyrA protein. Five of six isolates with ciprofloxacin MlCs of 64 mg/L had additional substitutions of Ser8O with Leu in the ParC protein.

Conclusions: Similar mutations to quinolone resistance, predominantly at codons 82-83 of gyrA, are occurring independently in genotypically distinct isolates of Acinetobacterspp. from different worldwide locations. Most isolates with high ciprofloxacin MlCs also exhibited secondary mutations in parC at codons 79-80.

Key words: Acinetobacter, quinolones, resistance, molecular epidemiology

generation cephalosporins. Until 1988, the fluoroquinolones had excellent activity against Acirietobnrtcr spp., but widespread use ha? resulted in the eniergence of resistance [5,6]. This resimnce has been attributed to chromosomal mutations that alter either DNA g r a s e or topoisomerase IV, the cellular targets of the fluoroquinolones, or to mutations that reduce the intracellular drug concentration by one of several mechanisms [ 7 ] . DNA =rase (topoisomerase 11) is the primary target of the fluoroquiriolones in Gram-negative bacteria and is essential for LINA synthesis. Structural alterations in this enzyme have been shown in a wide range of bacterial species to be associated with a decrease in susceptibility to the fluoroquinolones 181. T h e enzyme is composed of two A subunits and two U subunits, encoded by the g)’rA and XyrB genes, respectively. Specific mutations have been found in quinolone-resistant EicIier;cliia moli, resulting in substitutions of amino acids SerX3 and Asp87 in the A subunit [ 91. Mutations in2))rBappear to contribute to decreased susceptibility, but are rare [ 10,l 1 1.

INTRODUCTION Opportunistic infections caused by Acirietobarrer spp., predominantly A. Dairri~arrriii,include bacterernia, urinary tract infections and meningitis [ 1]. These organisms also play a significant role in intensive care units, where they are associated predominantly with nosocoinial pneunionia [2-41. Such infections are often difficult to treat because of the organism’s ability to become resistant rapidly to multiple groups of antibiotics, including the aniinoglycosides and third-

Corresponding author and reprint requests: K. J. Towner, Department of Microbiology and PHLS Laboratory, University Hospital, Queen’s Medical Centre, Nottingham NG7 ZUH, UK

Tel: t 4 4 115 970 9163 Fax: + 4 4 115 942 2190 E-mail: [email protected] Accepted 28 December 1997 248

Seward a n d Towner: Quinolone resistance i n Acinetobacter

Topoisomerase IV, with a structure similar to DNA gyrase, is a secondary target for the fluoroquinolones in E. coli [12]. The two subunits of this enzyme are encoded by parC and parE. It has been shown in E . coli that mutations in the region of parC equivalent to the quinolone resistance-determining region ( Q R D R ) of gyrA are also involved in the acquisition of fluoroquinolone resistance [13,14]. Mutations in parE are not found commonly in ciprofloxacin-resistant isolates of E . coli [ 9 ] . Other mechanisms reported to be involved in quinolone resistance include changes in the cell envelope as a result of loss of outer-membrane porins and alterations in lipopolysaccharide that are associated with intracellular accumulation of antibiotic [ 151. Expression of the inducible multiple antibiotic resistance (mar) locus confers resistance to multiple drug types via an active efflux system and may also play a role in quinolone resistance [ 16,171. So far as Acinetobacter spp. are concerned, mutations resulting in amino acid substitutions in GyrA and ParC, most commonly at Ser83 and Ser80, respectively, have been associated with fluoroquinolone resistance in 15 resistant isolates of A. baumannii collected from three hospitals in Spain [18,19]. The aim of the present study was to investigate whether similar independent mutations in XyrA and parC are resulting in quinolone resistance emerging in worldwide, genotypically unrelated clinical isolates of Acinetobacter spp., or whether clonal spread of particular quinolone-resistant isolates has occurred in response to increasing selection pressures.

MATERIALS AND METHODS Bacteria and susceptibility testing

In total, 25 nosocomial Acinetobarter isolates from 15 different locations in 11 countries were studied (Table 1). All had been identified previously to the genomic species level by the tDNA profile method [20]. Minimal inhibitory concentrations (MICs) were determined by an agar dilution method in which -lo4 colony-forming units (CFU) were inoculated onto Isosensitest Agar (Oxoid, UK) containing doubling dilutions of nalidixic acid (Sigma, Poole, Dorset, UK) or ciprofloxacin (Bayer, Newbury, Berkshire, UK). The MIC was taken as the lowest concentration of drug required to totally inhibit growth after incubation at 30°C for 18 h. The isolates were stored at -70°C in Skim Milk Powder (Oxoid, Basingstoke, Hampshire, UK) 10% w/v in sterile distilled water and then subcultured onto Cysteine Lactose Electrolyte Deficient Agar (Oxoid, UK) and incubated overnight at 30°C before further molecular analysis.

249

Determination of genotypic relationships by DNA fingerprinting

The genotypic relationships between the isolates were investigated by analysis with D E N D R O N v. 2.3 (Solltech, Oakdale, IA, USA) of randomly amplified polymorphic DNA (RAPD) fingerprints, obtained following amplification with either M13 core region or DAF4 primers, as described in detail previously [21]. The DENDRON program identifies the positions and intensities of the components of a DNA fingerprint and then calculates a similarity coefficient (Sm) for each individual pair of strains. The SABvalues are presented in a matrix and then used to generate a dendrogram showing the genotypic relationships of the collection of strains studied [21]. Release of genomic DNA and amplification of the QRDR of gyrA and parC by polymerase chain reaction (PCR)

Total genomic DNA was released from bacteria by resuspending a loopful of culture in 50 pL of sterile water and heating at 95 “C for 15 min. The crude DNA suspension was cooled on ice, diluted 1:10 with sterile distilled water, centrifuged for 1 min and stored at -20°C until required. PCRs for the Q R D R s ofgyrA and parC, respectively, were performed with the following primer sets [18,19]: GYR1, 5’-AAATCTGCCCGTGTCGTTGGT-3’ and GYR2, 5’-GCCATAACCTACGGCGATACC-3’; and PAR1 , 5’-AAACCTTGTTCAGCGCATT-3’ and PAR2, 5‘-GTGGTGGCCGTTAAGCAAA-3’. All primers were custom-synthesized by Oswel DNA Service (Southampton, UK). Individual reaction mixtures (25 pL) for gyrA and parC contained 1 x buffer, 200 pM deoxynucleoside triphospbates, 3.0 mM magnesium chloride, 0.6 U Taq polymerase (Promega, Southampton, UK), 2.5 pmol of each primer and 5 pL DNA extract. Amplification profiles were as described previously [ 18,191. Amplified DNA products were separated on agarose 1.5% W / V gels, stained with ethidium bromide and visualized on a UVtransilluminator. Digestion of the PCR product with Hinfl Previous studies of the conserved QRDRs of gyrA and parC of A. baunzannii have identified the presence of H i n f f restriction sites (GANTC) at the codons for amino acids Asp-Ser 82-83 and 79-80, respectively [18,19]. Portions (25 pL) of the P C R products were incubated at 37°C for 1 h with 10 U of Hinfl (Boehringer Mannheim, Lewes, East Sussex, UK) in the buffer recommended by the manufacturer. The digestion products were separated by electrophoresis in agarose 1.5% w/v gels and visualized as above. DNA

Clinical Microbiology and Infection, V o l u m e 4 Number 5

250

Table 1 Source and susce~tibilitvto quinolones of Ariwtobatrr isolates studied MIC (mg/L) Iiolatc

cw 1 CW5

cw11 CW12 cw13 CW14 CWZO CW22 CW27 CW28 CW31 CW32 (;1

(;5 S2 S4

Al A10 13 1 I3i 111

112 ESP41 ESP77 113.3

Origin

Nottingham, U K Nottinghani, UK 13risto1, U K 13ri\toI. UK Kotterdam, The Netherlmck Venlo, The Netherlands Paris. France I’aris, France Cardit?-, UK East Glainorgan, U K Pcr-th. Australia Perth, Australia Chpr Town. South Afric‘i Capc Town. South Africa Singap or? Singapore Buenoi Airei, Argentina 13uciios Aircs. Argcntinci Berlin, Gcrmany I3erlin, (;el inany Freiburg, Gcriiiany Frciburg, Gcrmany ~ ~ ~ l r c e ~Spain ona, Barccloiid, Spdiii Tricste, Italy

Identification‘

A. A

bnrr~~tarrrrii

X

barri~rnr~riii

32

Ari~rctoburrcrsp. 3

Acriietobartcr sp. 3 A ~in14~~1mr1ii

A.

bannrar~iiri

A.iiaurtraiiiiii A . bnirinaririii

A .~ > a u l i i ~ l l l l i i A.

Nalidixic acid

hniinraririii

Aririrtnbartrv ip. 3 A bniirrrarrirri

8

4 > 1024 > 1025 > l(l2.1 256 > 1024 > 1024

Ciproflosaciii

0.5 (1

-5

0.5 11.5 I (1 61

61 4 32 32

X

1

4

11.5

‘4.bnumarrrirr

4

11.5

‘4

4

(1.5

> 1024 > 1024

64 (1 5 il.5 (14 11.5 lJ.5

b,iiirnariiiri

A bnrrnrarrriii A.

barimaririii

A.bn~rn~arrriri A harrrrinririri il barimarrrrir

.4

bnrtrnnrrriii

.4.bniiir~arrrrii ‘4. lar~niuriiiir

A

barrirtniririi

.4

bniinimriii

A

bauirtari~iir

8 > 1024

32

32 8 > 1024 > 1024 32 5

(1.5

61 1h 0,s

11.5

’ Identified by tI1NA tingerprinting iiiethod [XI]

size marker?, run on the same gel, comprised a 100-bp DNA ladder (Promega, USA). Sequencing of PCR products Primers and free nucleotides were removed from I’CR products by incubating with 10 U exonuclease I and 3 U shrimp alkaline phosphatase (Cambridge Bioscience, Cambridge, UK) at 37 “C for 15 min, followed by heat inactivation at 70°C for 10 min. The purified product was then processed for DNA sequencing with an ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit incorporating AmpliTaq DNA polymerase, FS (Perkin Elmer, Warrington, UK), used according to the manufacturer’s instructions, followed by analysis in an ABI Prism 31 0 Gene Analyser (Perkin Elmer). Cycle sequencing was performed with either GYKl or PAK1, as appropriate.

RESULTS Table 1 shows the identification, source and quinolone susceptibility of the 25 Acinetobactev isolates included in the study. The isolates varied in their Fensitivity to the quinolone7, with MICs of nalidixic acid and ciprofloxacin between 4 and > 1024 mg/L and 0.5 and 64 ing/L, respectively.

RAPL) fingerprint profiles, obtained following amplification with the M13 core primer, for the 11 isolates with a nalidixic acid MIC of >64 mg/L c o n prised up to I 6 amplification products. The dendrogram derived from these profiles following analysis by the D E N D R O N program is shown in Figure 1. Although the two isolates from Wales (isolates CW27 and CW2X) appeared to be identical, the reinainder of these isolates with high levels of quinolone resistance were shown to be genotypically heterogeneous. This was confirmed by the similar results obtained following analysis of KAPD fingerprints generated with the DAF4 primer (results not shown). Amplification by PCR of the conserved Q R D K of the gyvA gene yielded a PCR product of343 bp from each isolate, as predicted by a previous study confined to isolates from three hospitals in Spain 1181. Hinfl digestion of the P C R products from the isolates with nalidixic acid MICs of 6 3 2 mg/L and ciprofloxacin MICs of 0.5 nig/L produced two fi-agrnents of 201 bp and 52 bp, except for one isolate (CW31) with a nalidixic acid MIC of 8 ing/L that failed to cut. Lligestion of the PCK products from the 11 isolates with a nalidixic acid MIC of >64 mg/L also failed to yield two fragments, indicating abolition of the restriction site at the codons for amino acids 82-83.

Seward and Towner: Quinolone resistance in Acinetobacter

251

CW13

I ' X CW2B CW28 CW27 cw22 CW14 D2

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

'AB

Figure 1 Dendrogram, based on RAPD fingerprints obtained with the M13 core primer, illustrating the similarity coefficients (Sm values) for the 11 isolates of Acinetobacter spp, with a nalidixic acid MIC of >64 mg/L.

Sequencing ofthe PCR products obtained fi-om CW31 And the 11 isolates with a high nalidixic acid MIC revealed a change from Ser to Leu at amino acid 83 in 11 of the 12 isolates. Isolate CW27 had a change at Ser83 to Phe. The 13 other isolates with no change at Ser83 had codon TCA for this amino acid. The resistant isolates with the Ser83 to Leu substitution had a transversion of C+T in codon TCA, and CW27 had a transversion of A+C in this codon. S2 had an additional substitution at amino acid 77 from Tyr to Cys. These results are summarized in Figure 2A. Amplification by P C R with primers for the conserved Q R D R of pavC also gave the predicted product of 327 bp [19]. Products of 206 bp and 121 bp were visualized after Hinfl digestion of the amplification products from those isolates with ciprofloxacin MICs of 6 3 2 mg/L. The P C R product from isolate S4 (ciprofloxacin MIC of 64 mg/L) also yielded two fragments following Hinfl digestion. The remaining isolates (ciprofloxacin MICs of 64 rng/L) had mutations a t the codons for amino acids 79-80, since Hinfl digestion of the P C R products from these isolates was unsuccessfi~l.Sequencing showed that these high-level ciprofloxacin-resistant isolates had a substitution at Ser80 with Leu. All contained a transversion of C-+T in the corresponding codon (Figure 2B). No other mutations were found. No parC mutation occurred without a concomitant mutation in gyr.4.

DISCUSSION Previous studies in 15 clinical isolates of A. baumannii from three hospitals in Spain have associated mutations in gyti4 and parC with fluoroquinolone resistance 118,191. The aim of this study was to determine whether similar mutations were occurring independently in isolates or whether resistant phenotypes are spreading clonally. O f the 25 isolates of Aci~etabacterspp. studied, cluster analysis results following DNA fingerprinting showed that 10 of the highly resistant isolates were genotypically heterogeneous. Nevertheless, all 10 had a substitution from Ser to Leu at codon 83 of GyrA, and the final isolate (CW27, genotypically identical to CW28) had a different substitution from Ser to Phe at codon 83. This mutation found in CW27 has not been reported previously in Acinetobacter spp., but is common in other quinolone-resistant Gram-negative bacteria [22,23]. A second mutation was found in the gyrA gene of isolxe S2 at codon 77. One of the relatively sensitive isolates (CW31), belonging to Acinetobacter sp. 3, also had a substitution in GyrA of SerX3 with Leu, but had nalidixic acid and ciprofloxacin MICs of only 8 mg/L and 1 mg/L, respectively. O f the 11 isolates with resistance to ciprofloxacin (MIC 4 mg/L), five had additional mutations at codon 80 in parC leading to substitution of Ser with Leu. These isolates all exhibited high-level resistance

Clinical M i c r o b i o l o g y a n d Infection, Volume 4 Number 5

252

Figure 2 Partial DNA sequences showing the conserved QKDKs of (A) the gyrA and (B) the p n r C genes of quinolonesusceptible and -resistant isolates of A . haumannii. The locations of the H i d restriction sites are marked with arrows. The deduced amino acid sequence is given below the DNA sequence for a susceptible isolate (CW1). Other susceptible isolates also shared the consensus seauence. Deviations from the consensu? sequence are shown for the resistant isolates. ~

(Ai

76

75

77

78

79

~~

xo

81

82

83

i

84

85

86

87

88

89

90

CGT AAA TAT CAC CCG CAT GGT GAC TCA GCT GTT TAT GAA ACC ATT GTT (Scncitive strain) Glv Lys Tyr Hir Pro Hir Gly Asp Ser Ala Val Tyr Glu Thr Ile Val

cw1*

cw13

( X 4 c:w20 CW22

TTA Leu .......................... ..TTA Lru . . . . . . . . . . . . . . . . . . . . . . . . . TTA Lcu . . . . . . . . . . . . . . . . . . . . . . . . . . TTA Leu ............................

MIC5 ( nig/L) (Nalidisic acid Ciprofloxa~in

8

1(1J

........................ ..........................

........................... ..........................

..TK ..........................

cw27

. . . . . . . . . . . . . . . . . . . . . . .

CW28

. . . . . . . . . . . . . . . . . . . . . . . . .

S2

Phe TTA . . . . . . . . . . . . . . . . . . . . . . . . . . . Leu . . . . . . T G T . . . . . . . . . . . . . . . . . .TTA . . . . . . . . . . . . . . . . . . . . . . . . . . .

S4

cy\ ...........................

A10

.............................

1j 2

..............................

ESP41

. . . . . . . . . . . . . . . . . . . . . .

Leu TTA Lru TTA Let1 TTA Leu TTA

......................... ..........................

. . . . . . . . . . . . . . . . . . . ..........................

Lru

M)

(:wI*

73

72

74

75

76

77

78

70

81)

1

81

82

83

84

X i

86

87

MICs ( ina/L) (Nalidis~cacid (:ipi-ofloxacin

GGT AAA TAT CAC CCa C.4T GGT GAC TCG GCA TGT TAT GAA GC(: ATG GTA Tyr Hi\ Pro His Gly A ~ ISrr Ala Val CyS G l u Ala Met Val

X

<:w13

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

> 1!124

16

(:wI4

..............................

TT(; . . . . . . . . . . . . . . . . . . . . . . . . . .

> 1 u24

64

>1!124

64

2.5 6

4

>lo 2 4

32

> 11’124

S2

...

21024

04

..

>lll24

(14

..TTG . . . . . . . . . . . . . . . . . . . . Leu . . . . . . . . . . . . . . . . . . . . . . . . . TTG . . . . . . . . . . . . . . . . . . . . . . . . . Leu

>1!124

64

>1024

04

111124

16

50.5

(Semitive strdin) Gly Lys

Lcu

CW21)

cw22

TTG . . . . . . . . . . . . . . . . . . . . . LCU ........................................................

. . . . . . . . . . . . . . . . . . . . . . . . .

cw27

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CW28

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

32

. . .

TTG

Leu

S4 A10

112 ESP4 I

... .........................

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

to ciprofloxacin, with MICs of > 64 mg/L. N o purC mutation was found without a mutation in gyrA. Isolate S4 had a ciprofloxacin MIC of 64 mg/L, but no

concomitant mutation in p r C was found. A mutation i n another region of the purC gene that was not amplified may have been responsible for the increase in

Seward and Towner: Quinolone resistance i n Acinetobacter

resistance. Although such mutations are uncommon in E. coli, this isolate may have had further mutations in gyrB or parE which lead to an increase in MIC. Alternatively, drug-permeability and -efflux mechanisms may have been involved. However, a separate study has shown that diminished quinolone accumulation is unlikely to account for the high-level resistance of the A. baumannii isolates studied [24]. Overall, this work confirmed earlier findings that a change in Ser83 of GyrA is the predominant mutation responsible for quinolone resistance in A . baurnarmii and generally confers ciprofloxacin MICs of 3 4 mg/L [18]. However, this mutation was also found in an isolate of Acinetobacter genospecies 3 which had a ciprofloxacin MIC of 1 mg/L. A double mutation in gyrA and parC has also been shown previously to be required for high-level resistance to ciprofloxacin (MICs of 64 mg/L) [19]. This was also found to be the case in this study. However, in contrast, the present study found that the gyrA mutation alone is sufficient to confer resistance to nalidixic acid at a concentration of >I024 mg/L. As described above, minor variations in resistance between isolates may reflect differences in drug-permeability and -efflux mechanisms. However, this study has shown that substitutions at Ser83 of GyrA and at Ser80 of Pare appear to be primarily responsible for quinolone resistance in the isolates of Acinetobacter spp. studied, and that these mutations are occurring independently in genotypically distinct worldwide isolates. Similar findings have been reported for E. coli and Klebsieh ynetrmoniae [25]. Further work is required to assess the contributions that mutations in other subunits of the proteins and other general resistance mechanisms make to quinolone resistance in occasional unusual isolates and also to study the relative effects that the mutations have among different genospecies of Acinetobacter. Ac knowledgrnents

We are indebted to our colleagues who contributed strains of Acinetobacter to this study and to Dr J. Vila for helpful discussions.

References 1. Bergogne-BkrCzin E, Towner KJ. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidenliological features. Clin Microbiol Rev 1996; 9: 148-65. 2. Buxton AE, Anderson RL, Werdegard D, Atlas E. Nosocomial respiratory tract infection and colonization with Ahetobacttv mlcoaceticns: epidemiologic characteristics. Am J Med 1978; 65: 507-13. 3 . Cefai C, Richards J, Gould FK, McPeake I? An outbreak of Acinetobacter respiratory tract infection resulting from

253

incomplete disinfection of ventilatory equipment. J Hosp Infect 1990; 15: 177-82. 4. Bel-gogne-BCrkzin E, Joly-Guillou M-L. Hospital infection with Acinetobacter spp.: an increasing problem. J Hosp Infect 1991; 18(~upplA): 250-5. 5. Joly-Gudlou M-L, Bergogne-Btrkzin E. In vitro activity of sparfloxacin, pefloxacin, ciprofloxacin and temafloxacin against clinical isolates of Acinetobacter spp. J Antimicrob Chcmother 1993; 32: 1285-8. 6. Fass RJ, Barnishan J, Ayers L. Emergence of bacterial resistance to imipenem and ciprofloxacin in a university hospital. J Antimcrob Chemother 1995; 36: 343-53. 7. Wolfson JS, Hooper DC. Fluoroquinolone antimicrobial agents. Clin Microbiol Rev 1989; 2: 378-424. 8. Tillotson GS, Dorrian I, Blondeau J. Fluoroquinolone resistance: mechanisms and epidemiology. J Med Microbiol 1997; 46: 457-61. 9. Everett MJ, Fang Jin Y, Ricci V, Piddock LJV. Contributions of individual mechanisms to fluoroquinolone resistance in 36 Esclierichin coli strains isolated from humans and animals. Antimicrob Agents Chemother 1996; 40: 2380-6. 10. Nakamura S, Nakamura M, Kojuna T, Yoshida H. C y r A and gyrEl mutations in quinolone resistant strains of Escherichia coli. Antimicrob Agents Chemother 1989; 33: 254-5. 11. Vila J, Ruiz T, Marco F, et al. Association between double mutation in gyrA of ciprofloxacin-resistantclinical isolates of Eschwrichia colt and MICs. Antiniicrob Agents Chemother 1994; 38: 2471-9. 12. Hoshino K, Kitamura A, Morrisey I, Sato K, Ikeda H. Cornparison of inhibition of Escherichia coli topoisomerase IV by quinolones with DNA gyrase inhibition. Antimicrob Agents Chemother 1994; 38: 2623-7. 13. Kuniagai Y, Kayo J-I, Hoshino K, Akasaka T, Sat0 K, Ikeda H. Quinolone-resistant mutants of Escherichia coli DNA topoisomerase IV parC gene. Antimicrob Agents Chemother 1996; 40: 710-14. 14. Heirig P. Genetic evidence for a role of parC mutations in development of high-level fluoroquinolone resistance in Escherichia coli. Antimicrob Agents Chemother 1996; 40: 879--85. 15. Hirai K, Aoyama H, Suzue S, Irikera J, Iyobe S, Mitsuhashi S. 1,;olation and characterisation of norfloxacin-resistant mutants of Escherichia coli K-12. Antimicrob Agents Chemother 1986; 31: 582-6. 16. Yoshimura F, Nikaido H. Permeability of Pseudomonas aevuginosa outer membrane to hydrophil~csolutes.J Bacteriol 19821; 152: 636-42. 17. Goltlnian JD, White DG, Levy SB. Multiple antibiotic resistance (mar) locus protects Escherichia coli from rapid cell killing by fluoroquinolones. Antimicrob Agents Chemother 1996; 40: 1266-9. 18. Vila J, Ruiz J, Goiii P, Marcos A, Jiminez de Anta T. Mutation in the gyrA gene of quinolone-resistant clinical isolates ofAcinetobacter baumannii. Antimicrob Agents Chemother 1995; 39: 1201-3. 19. Vila J, Ruiz J, Gofii P, Jiminez de Anta T. Quinoloneresistance mutations in the topoisomerase IV parC gene of Acinctobacter baumannii. J Antimicrob Chemother 1997; 39: 757- 62.

254

Clinical Microbiology a n d Infection, Volume 4 Number 5

20. Ehrenstein €3, Uernardr AT, Ilijkshoorii L, et al. Acinetohncter

rpecies identification using t K N A fingerprinting. J Clin Microbiol 1996; 34: 2414-20. 21. Seward RJ, Ehrenstein B, Grundmann HJ, Towner KJ. Direct comparison of two commercially available computer program$ for analysing DNA fingerprinting gels. J Med Microbiol 1997; 46: 311-20. 22. Deguchi T, Fakuoka A, Yasuda M , et al. Alterations in the GyrA rubunit of DNA gyrase and the ParC subunit of topoisonierase IV in quinolone-resistant clinical irolates of Klrhsidln pweumonrar. Antiniicrob Agents Cheniother 1997; 41: 699-701. 23. Ruiz J. Cactro 11, Gofii P, Santamaria JA, Borrego JJ, Vila J. Analysis of the mechanism of quinolone resistance i n

nalidinic acid-resiytant clinicd isolates of Saliiroriella rerotypt. Typhimurium. J Med Microbiol 1997; 46: 623-8. 24. Moreau NJ, Houot S , Jolp-Guillou M-L, Bergognr-Bbrtziii E. Characterisation of D N A gyrase and nieasurenient of drug accuniulation in clinical isolates of Acirretobactrr hniirirannii resistant to fluoroquinolones. J Antimicrob Chemother 1996; 38: 1079-83. 25. Bauernfeind A, Abele-Horn M , Enimerling P, Jungwirth R. Multiclonal emergence of ciprofloxacin retiqtant clinical isolate$ of Exherichin roli m d Klchsiella pnciruzoniae. J Antimicrob Chemother 1994: 34: 1074-6.