A P M l S 107: 1 / 0 0 16, IYYY Priiircrl In Dcwrmirk . All r i ~ h t s rrseriwl

Copyright 0 A P M l S IYYY

LA1 F'I i d lN ISSN 0003-4641

Rapid identification of mutations in a multidrug efflux pump

SHAH JALAL,' GUNNAR WRETLIND,' NAOMASA GOTOH' and BENGT WRETLIND'

'Division of Clinical Bacteriology, Huddinge University Hospital, Huddinge, Stockholm, Sweden and 'Department of Microbiology, Kyoto Pharmaceutical University, Yamashina, Kyoto, Japan

Jalal S, Wretlind G, Gotoh N & Wretlind B. Rapid identification of mutations in a multi-drug efflux pump in Pseudomonas aeruginosa. APMIS 1999;107:1109-16.

The gene mexR regulates negatively the expression of the MexA-MexB-OprM efflux pump in Pseudomonas arruginosa, and mutations in mexR cause a multiple antibiotic resistance phenotype. Five hundred and forty resistant clones of P. aeruginosa PA0503 were isolated after selection for resistance to chloramphenicol or tetracycline. All isolates showed similar phenotypes and were resis- tant to tetracycline, chloramphenicol and norfloxacin. Nineteen randomly selected isolates were ana- lyzed. Since mutational analysis by direct sequencing of all regions of interest in several strains is time- consuming and expensive, a screening method, Non-Isotopic RNase Cleavage Assay (NIRCATM), was applied to identify mutant genes so that they could be targeted for DNA sequencing. NIRCA is a simple but rapid method for mutational analysis and can be performed in 3 4 h. Results of NIRCA analysis were compared with DNA sequencing. Both NIRCA and DNA sequencing analysis showed mexR gene mutations in 1 1 of 19 isolates but no alterations in 8 strains. An immunoblot assay showed overexpression of OprN, a component of another multidrug efflux pump, MexE-MexF-OprN, in those eight isolates. Nucleotide sequencing of quinolone resistance-determining regions of DNA gyrase ( g y r A ) or topoisomerase IV @arc) showed no alterations in any of the 19 mutants. The data indicate that two efflux pump systems, MexA-MexB-OprM and MexE-MexF-OprN, were involved in multid- rug resistance including quinolones and that NIRCA is a sensitive method for screening mutations.

Key words: NIRCA; PCR; multidrug efflux pump; antibiotic resistance.

Shah Jalal, Division of Clinical Bacteriology, F 82, Huddinge University Hospital, S-141 86 Huddinge, Sweden.

Alterations in the gyrA subunit of DNA gyrase and the purC subunit of DNA topoisomerase IV have a central role in conferring fluoro- quinolone resistance in bacteria. Strains with high levels of resistance to quinolones have multiple mutations in DNA gyrase (gyrA), topoisomerase IV @arC) and efflux pumps (1 , 2). At least three different multidrug efflux- pump systems active against quinolones have been identified in Pseudomonas aeruginosa: MexAB-OprM, MexCD-OprJ and mexEF- OprN, regulated by mexR (null?), nfxl? and

Received June 10, 1999. Accepted August 19, 1999.

mexT (nfxC), respectively (3-10). Mutations in regulatory genes induce overexpression of the respective efflux-pump system and contribute to multiple drug resistance (MDR) in P. aerugino- sa (1 1-14).

In this study, 540 resistant clones of P. aerugi- nosa PA0503 were isolated after selection for resistance to chloramphenicol or tetracycline. All the isolates showed similar resistance pheno- types and were cross-resistant to norfloxacin. Nineteen randomly selected isolates were ana- lyzed for multidrug-resistant mutations.

Mutational analysis by direct sequencing of all regions of interest in a large number of strains is time-consuming and expensive. There-

1109

JALAL et a/.

fore, the aim of this study was to evaluate a screening method, Non-Isotopic RNase Cleav- age Assay (NIRCATM, Ambion, Austin, TX, USA), to identify mutations and compare the results with DNA sequencing (15, 16).

MATERIALS AND METHODS

Bacterial strains The fluoroquinolone-resistant mutants of P. aerug-

inosa PA0503 (met-9011) used in this study are listed in Table 1. The laboratory-derived mutants were created after treatment of strain PA0503 with ethyl methanesulfonate (EMS) (Sigma, St. Louis, MO, USA) as described previously (17). The mutants were selected on Isosensitest agar (Oxoid, Basingstoke, UK) plates containing chloramphenicol (1 25 mg/l, 250 mg/l or 500 mg/l) or tetracycline (50 mg/l or 100 mg/l).

Susceptibility testing Minimum inhibitory concentrations (MIC) were

determined by agar dilution using Isosensitest agar medium. Inoculum size of bacterial cells was about 10,000 cfulspot. MIC was defined as the lowest con- centration inhibiting visible cell growth, as evaluated after 18 h of incubation at 37°C. Chloramphenicol, tetracycline and norfloxacin were from Sigma.

Primer design for NIRCA screening of mexR gene mutation

NIRCA is a method using RNase treatment of RNA transcripts to detect cleavage products. The protocol (MisMatch Detect 11, Ambion, Austin, TX, USA) suggested by the manufacturer was followed with slight modifications. The primer pair was de- signed to amplify the mexR gene of P. aeruginosa by single-step polymerase chain reaction (PCR) and to hybridize the sequences of PCR-amplified product. Primers contain bacteriophage promoter sequences, which were added to the 5’ends. By convention, the

TABLE 1. NZRCA analysis of the mexR gene and recording of Mismatch data. The extent of RNase cleavage of Mismatches from barely detectable to essentially complete

Target Wild-type sense Strand/experimental Wild-type antisense Strand Inter- duplex: antisense Strand (T) experimental sense Strand (S) pretation PA0 RNase digestion RNase digestion RNase digestion RNase digestion strains Condition 1 Condition 2 Condition 1 Condition 2

503-WTb 503-K1 - SC - €3 mutation

- 03 mutation 503-K3 - - €3 mutation

€3 503-K2

mutation €3

503-K4 ?d - 503-K2 1 -

503-K22 503-K23 503-K24 503-K25 503-TI 503-T2

%3 - - mutation - €3 €3 mutation

503-T3 03 503-T4 503-T5 - - €3 €3 mutation 503-T6 - - %3 €3 mutation

mutation

- - mutation - -+ mutation

- - - - e - 503-WTa - - - - -

-

- €3 -

- - ? - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - -

-

- €3 -

- - €3

03

- 503-T7 503-T8 503-T9 - 503-T10 The P.aeruginosa PA0503-mutants K, selected on chloramphenicol, and T, selected on tetracycline. a Wild-type control 1, treated with RNases; Wild-type control 2, not treated with RNases; Definite cleavage;

Barely detectable difference between resistant mutant and sensitive wild-type control; (Negative), no detect- able difference between resistant mutant and wild-type. Interpretation of recorded MisMatch data: The experimental samples were generally scored as positive for alterations (mutation) if a score of €3 or + appeared in at least one of the four columns or if a score of t appeared in at least two columns. Samples containing no distinct bands (except for the possibility of full-sized bands that are the PCR products used as transcription template) were scored as negative.

1110

- -

? -

RAPID IDENTIFICATION OF MUTATIONS IN PSEUDOMONAS AERL'GINOSA

TABLE 2. Oligonucleotide primers used in PCR screening, NIRCA, and DNA sequencing Primers Sequences Reference T7-mexRU

SP6-mexRL

5'-TAA TAC GAC TCA CTA TAG GG TCG GCC

5'-ATT TAG GTG ACA CTA TAG AA GGG TGA

This study

This study (41 mer)

(41 mer)

AAA CCA ATG AAC TAC-3'

GCG GGG CAA ACA ACT-3' Underlining indicates the T7 for Sp6 promoter sequences added to previously used mexR-U and mexR-L primers, and the respective nucleotide positions are 265-285 and 629-649 in the mexR gene (8).

T7 promoter was added to the sense primer (T7- mexRU) and the SP6 promoter to the antisense primer (SP6-mexRL) (Table 2).

PCR and incorporation of promoter sequence to the mexR gene

The genomic DNA was prepared from strains of P. aeruginosa with guanidium thiocyanate as de- scribed by Pitcher et al. (18) and was subjected to PCR amplification. A master mix was prepared on ice; it contained I X UlTma buffer (Gene Amp@, Per- kin Elmer Cetus), 200 pM of each dNTP, 0.6 pmol/ pl of each primer (Scandinavian Gene Synthesis AB, Koping, Sweden), 1.5 mM MgC12, 1.5 U/50 pl UlTma DNA polymerase, an enzyme with proof- reading activity (Gene Amp@ Perkin Elmer Cetus, Foster City, CA, USA) and -5.0 ng/pl genomic DNA from P. aeruginosa. The master mix was aliquoted to a volume of 50 p1 and denatured at 97°C for 10 min. The samples were cycled at 94°C for 90 s, 64°C for 30 s, and 72°C for 2 min for 30 cycles, with a final 7 min extension at 72°C. The DNA amplification with the incorporation of bacteriophage promoter was done in a DNA thermal cycler, GenAmp PCR Sys- tem 2400 (Perkin Elmer). (5 pl) PCR product was analyzed by electrophoresis on 1% (wthol) agarose gel, stained with ethidium bromide, and visualized under UV light. The PCR products were 425-bp frag- ments of mexR gene with the incorporation of T7 and Sp6 promoter sequences.

In vitro transcription of the PCR products to make sense and antisense RNA probes

Both strands of each promoter PCR (pPCR) prod- uct from experimental and wild-type samples were transcribed in separate reactions to generate the sense and antisense RNA probes. The sense strand was transcribed with T7 RNA polymerase (T7 transcrip- tion mix) and the antisense strand with SP6 RNA polymerase (SP6 transcription mix). Both transcrip- tion mixtures, prepared on ice, contained 1 X tran- scription buffer, 0.5 mM rNTP mix and 2 U/p1 RNA polymerase (Ambion, Austin, TX). The mixtures were aliquoted to volumes of 8 pl for each experimen- tal and wild-type reaction, and 2 p1 pPCR product was added. The samples were incubated for 1 h at 37°C. The in vitro transcription reactions were ter- minated by addition of an equal volume of Mis- Match hybridization buffer (Ambion), mixed briefly,

spun down, and heated for 3 rnin at 95°C. The final volume of each reaction was then 20 pl. The reaction proceeded to hybridization on the same day.

Hybridization of experimental and control transcripts An equal volume of the wild-type T7 transcript (T-

W) was added to each SP6 experimental transcript and an equal volume of wild-type SP6 transcript (S- W) was added to each T7 experimental transcript, giving a volume of 40 p1 of each T and S reaction. At least 4 p1 of each T7-W and S-W was mixed to make the no-mismatch control duplex (TS-W). Reac- tions were heated for 3 min at 95°C and cooled to room temperature for about 3 min. The hybridized transcripts could be stored at -20°C for several months.

Treatment of the hybridized samples with RNase and electrophoresis

Four microliters of each hybridized sample (T, S and TS-W) was treated with 16 p1 of 1:lOO diluted mismatch detection RNase digestion buffer #1 and #2 (Ambion) and incubated at 37°C for 45 min. The first and second RNase digestion conditions were car- ried out using buffer #I and a mixture of buffers #I and #2. After cooling to room temperature, 4.5 p1 of gel loading solution was added to each sample. The samples were analyzed by gel electrophoresis in l x TBE using 2% (wthol) high resolution agarose-1000 (Life Technologies, NY, USA). The gel was stained with pre-added ethidium bromide in loading solu- tion, visualized under UV light, and photographed. A schematic diagram of the NIRCA method is shown in Fig. 1.

Isolation of cell envelope, SDS-PAGE and immunob- lot assay of OprN

Cells grown in L broth were harvested by centri- fugation at 5,000 X g for 10 min at 4°C. The cells were resuspended in 10 mM Tris-HC1, pH 8.0, and were broken by sonication. After removing unbroken cells by centrifugation, total membranes (cell envel- opes) were pelleted from the resulting supernatant by centrifugation at 20,000 X g for 30 min. SDS-PAGE was performed using 10% wlv acrylamide in the run- ning gel. Proteins fractionated by SDS-PAGE were electrophoretically transferred to nitrocellulose paper (0.45-pm pore size, Bio-Rad, Hercules, CA, USA) as described previously (10). Binding of the primary

1111

0 in vltru transcription

u hybridization

RNA sclspore

11111 1111 11111 1111

Electrophoresis a 111c1

111c 1111

11111 ri - - Schematic diagram of NIRCA method

Fig. 1. RNA was transcribed from both strands of the PCR product using T7 RNA polymerase for the sense strand and SP6 RNA polymerase for the anti- sense strand from both the control wild-type (WT) and experimental mutant (Mu) separately. Sense WT/ antisense Mu and antisense WT/sense Mu strands were cross-hybridized and treated with RNases (RNA scissors). The RNase cleavage product was electrophoresed. The box shows an uncleaved full- size WT band and cleaved smaller size Mu bands. For the sake of simplification the method is shown for one sense and one antisense strand.

antibody (Anti OprN antibody, TN005) was detected using alkaline phosphatase-conjugated goat anti- bodies to rabbit IgG (Cappel, Durham, NC, USA) or alkaline phosphatase-conjugated goat antibodies to mouse IgG (Cappel) as the secondary antibodies, and AP conjugate substrate Kit (Bio-Rad) for color development (10).

Nucleotide sequencing Primers and free nucleotides from PCR products

were removed using the QIAquick-spin PCR purifi- cation kit (Qiagen Inc., Chatsworth, CA.). The puri- fied PCR products were directly processed for DNA

sequencing using ABI PRISM BigDye Terminator and capillary electrophoresis technology in an ABI 310 Genetic Analyzer (Perkin Elmer). PCR primers were used as sequencing primers. Nucleotide and de- duced amino acid sequences were analyzed using Ma- cintosh DNA programs FACTURATM, EditView 1 .O. 1 (Applied Biosystems, Perkin Elmer), ClustalW Interactive Multiple Sequence alignment at the Euro- pean Bioinformatics Institute, UK (http://www2.ebi.- ac.uk) and the ExPASy Molecular Biology Server at the Swiss Institute of Bioinformatics (SIB), Geneva, Switzerland (http://www.expasy.ch) ( 1 ).

Gen Bunk accmsion

sequence of the niesR gene is U23763. The GenBank accession number for the nucleotide

R ES U LTS

Isola t ion of niu tan t s and an t ihio t ic suswpt ihilitjt The frequency of isolation of EMS-treated

culture (6X lo8 cfu/ml) on Isosensitest agar plates with chloramphenicol and tetracycline was 1:2.5X lop4 and 1:1.5XlO-', respectively. One hundred and eight colonies for each con- centration of chloramphenicol ( 125 mgll, 250 mgll or 500 mgll) or tetracycline (50 mg/l or 100 mgll), total 540 colonies, were isolated and tested for MICs of chloramphenicol, tetracy- cline and norfloxacin. All mutants showed a similar resistance phenotype and were cross-re- sistant to norfloxacin. The MICs of chloram- phenicol, tetracycline and norfloxacin were 1000 mgll, 200 mg/l and 8 mg/l. Corresponding MICs for PA0503 (wild-type strain) were 32 mdl, 10 mgll and 0.25 mg/l, respectively. Nineteen ran- domly selected isolates were analyzed for muta- tions to multidrug resistance using NIRCA, im- munoblot and DNA sequencing.

NIR CA Mis Mutch detection Both strands of pPCR fragments of the nze.uR

gene were analyzed using the NIRCA method. Eleven of nineteen mutants showed definite alterations in the mc..\-R gene and no alterations in eight mutants (Fig. 2 and Table 1).

DNA sequence analysis of' mexR gcwes The rest of the pPCR products of the nzesR

gene from the NIRCA assay were purified and processed for automated DNA sequencing. Both strands of PCR-amplified fragments were sequenced twice to exclude possible errors in-

1112

RAPID IDENTIFICATION OF MUTATIONS IN PSEUDOMONAS AERUGINOSA

M K25 K24 K23 K22 K21 K4 K3 K2 K1 WT' M T10 T9 T8 T7 T6 T5 T4 T3 T2 T1 M

PCR product

WT sense/ Exp antisense strand

WT sense/ Exp antisense strand

a

WT K25 K24 K23 K22 K21 K4 K3 K2 K1 WT' T10 T9 T8 T7 T6 T5 T4 T3 T2 T1 WT

b

WT' WT K25 K24 K23 K22 K21 K4 K3 K2 K1 T10 T9 T8 T7 T6 T5 T4 T3 T2 T1

C

K1 K2 K3 K4 K21 K22 K23 K24 K25 WT T10 T9 T8 T7 T6 T5 T4 T3 T2 T1

d

WT' WT K25 K24 K23 K22 K21 K4 K3 K2 K1 T10 T9 T8 T7 T6 T5 T4 T3 T2 T1

e Fig. 2. A panel of 20 mexR gene targets amplified from genomic DNA from quinolone-resistant strains was transcribed to RNA. Sense and antisense experimental (Exp) RNA probes were hybridized with complementary wild-type (WT) control probes and treated with two RNases. Cleavage of mismatches was analyzed on 2%) high-resolution agarose gel electrophoresis. Two or more bands and/or one band smaller than the wild-type bands represent mismatch cleavages. WT*, wild-type PCR product; M, molecular weight marker. Fig. a, Pre- transcribed 425-bp PCR products (5 plllane) of the mexR gene. Fig. b & d, hybridized transcripts cleaved using RNase digestion condition 1 (shown in columns 1 and 3, Table 2) . Fig. c and e, sample cleaved using RNase digestion condition 2 (shown in columns 2 and 4, Table 2) .

1113

JALAL et al.

TABLE 3. NIRCA screening and nucleotide changes and predicted amino acid changes in PCR-amplijied 425-hp fragments of the mexR gene and expression of OprN in non-quinolone selection of quinolone-resistant I? aerugino-

sa P A 0 mutants PA0 strains Mutational analysis

NIRCA analysis of mexR gene DNA sequencing (mexR) Immunoblot assay of OprN - - - 503-WT

503-K1 + Arg (CGC) 83+His (CAC) ND 503-K2 + Arg (CGC) 83+His (CAC) ND 503-K3 + Arg (CGC) 91+His (CAC) ND 503-K4 + Leu (CTG) 13+Met (ATG) ND

+ + + + + + +wk

- - 503-K21 503-K22 503-K23 503-K24 503-K25 503-T1 503-T2

- - - -

- - - - - -

- -

503-T3 + Gly (GGA) 58+Glu (GAA) ND 503-T4 + Leu (CTC) 95+Phe (TTC) ND 503-T5 + Gln (CAG) 60+Stop (TAG) ND 503-T6 + Thr (ACC) 69 j I l e (ATC) ND 503-T7 + Arg (CGG) 21+Trp (TGG) ND

503-T9 + Leu (CTG) 80+Pro (CCG) ND 503-T10 + Arg (CGG) 7O+Trp (TGG) ND NIRCA mutational analysis of mexR gene as illustrated in Fig. 2; +, mutation present; -, mutation absent. Production level of OprN was determined from the intensity of the specific signal demonstrated compared to WT shown in Fig. 3. +, production; +wk, weak production; -, non-production and ND, not done.

+ - - 503-T8

troduced during PCR despite using the proof- reading UlTma DNA polymerase. Sequence analysis of DNA showed that 11 of 19 mutants had mexR gene mutations (Table 3). We have also analyzed all of the 19 mutants using DNA sequencing of the quinolone resistance-deter- mining region (QRDR) of gyrA and parC and the entire sequence of nfxB genes, but did not find any alterations in these genes (data not shown).

Immunoblot assay of O p r N We found no alterations in mexR genes in 8

of 19 mutants using either NIRCA or DNA se- quencing. Immunoblot assay showed that these eight strains overexpressed OprN compared to PA0503-WT (Table 3).

DISCUSSION

Antibiotic resistance presents an emerging health problem with respect to the treatment of infectious diseases. In the last couple of years,

1114

it has been assumed that mutations in the gyrA subunit of topoisomerase I1 and the p a r C sub- unit of topoisomerase IV are the major cause of quinolone resistance in bacteria. The relevance of mutations in gyrA and parC is well estab- lished, but the relevance of efflux-pump systems in the selection of quinolone-resistant bacteria is emerging. Previous observations in E. coli showed that mutants obtained after three con- secutive selection steps on quinolone carry both gyrA and p a r C mutations (19).

The P. aeruginosa mutants used in this study exhibited similar multidrug resistance pheno- types, including cross-resistance to norfloxacin. The Non-Isotopic RNase cleavage assay (NIR- CA) was applied as a preliminary screening method so that only samples containing muta- tions would be targeted for DNA sequencing.

The NIRCA method is an RNA-based analy- sis method, but does not require strict adher- ence to RNase-free techniques. This is because most of the time RNA is double stranded dur- ing the procedure and less vulnerable to degra- dation. PCR products used as transcription template have less potential for introducing

RAPID IDENTIFICATION OF MUTATIONS IN PSEUDOMONAS AERUGINOSA

RNase contamination. It is a simple method, does not require any special equipment, and can be performed in 3 4 h after PCR. In the NIR- CA method, the electrophoresis pattern was ob- served by RNase digestion of the RNA/RNA duplex from samples and control strains. The score from the electrophoresis pattern was then interpreted as presence or absence of mutation.

NIRCA analysis of mexR gene showed the presence of mutations in 11 of 19 strains, which was confirmed by DNA sequence analysis of the mexR gene. The eight remaining strains over- expressed another efflux pump-associated pro- tein, OprN (7).

Since the mutants were resistant to norflox- acin, we also analyzed quinolone resistance-de- termining regions of gyrA and parC by DNA sequencing, but did not find any alterations. MexR-MexAB-OprM and NfxC-MexEF-OprN efflux pumps are alternative resistance mechan- isms, phenotypically similar, with the same MICs to the tested antibiotics. Mutations in either of them seem to occur with about the same frequency after selection for different un- related antibiotics.

The original NIRCA assay was based on dual amplification steps of PCR. Better primer de- sign and successful optimization of the PCR allowed simplification of the method to a single- step PCR. Primer design is an important factor in successful PCR amplifications. Primers that have complementary 3' ends may fail to amplify the desired target because of mispriming, espe- cially when the primer length is short and the GC content high. Although there are ways of dealing with this problem, such as preboiling primers and adding Taq polymerase after the denaturing steps of PCR, the disadvantages are risk of contamination or low product output. The methods used are also laborious. Often, nested PCR is required to achieve sufficient am- plification (20, 21). We encountered this prob- lem when we attempted to use PCR to amplify gyrA, parC, mexR and nfxB genes of P. aerugi- nosa, an organism with high GC content. Using UlTma DNA polymerase, a highly thermo- stable enzyme (half-life 40 min at 97.5"C) and heating the PCR mix at 97°C for 10 min at the initial denaturation step resulted in successful DNA amplification in single-step PCR.

In conclusion, mutations in regulatory genes for the two efflux systems MexAB-OprM and

MexEF-OprN, cause resistance to at least three unrelated antibiotics, and NIRCA is a con- venient, simple and sensitive method for the de- tection of such mutations.

This work was supported by a grant from the Scandi- navian Society for Antimicrobial Chemotherapy (SSAC).

REFERENCES

1. Jalal S, Wretlind B. Mechanisms of quinolone re- sistance in clinical strains of Pseudomonas aerugi- nosa. Microb Drug Resist 1998;4:257-61.

2. Martinez JL, Alonso A, Gomez-Gomez JM, Ba- quero E Quinolone resistance by mutations in chromosomal gyrase genes. Just the tip of the ice- berg? J Antimicrob Chemother 1998;42:683-8.

3 . Fukuda H, Hosaka M, Iyobe S, Gotoh N, Nishi- no T, Hirai K. nfxC-type quinolone resistance in a clinical isolate of Pseudomonas aeruginosa. Antimicrob Agents Chemother 1995;39:790-2.

4. Gotoh N, Tsujimoto H, Nomura A, Okamoto K, Tsuda M, Nishino T. Functional replacement of OprJ by OprM in the MexCD-OprJ multidrug efflux system of Pseudomonas aeruginosa. FEMS Microbiol Lett 1998;165:21-7.

5. Gotoh N, Tsujimoto H, Tsuda M, Okamoto K, Nomura A, Wada T, Nakahashi M, Nishino T. Characterization of the MexC-MexD-OprJ mul- tidrug efflux system in DeltamexA-mexB-oprM mutants of Pseudomonas aeruginosa. Antimicrob Agents Chemother 1998;42:193843.

6. Hosaka M, Gotoh N, Nishino T. Purification of a 54-kilodalton protein (OprJ) produced in NfxB mutants of Pseudomonas aeruginosa and produc- tion of a monoclonal antibody specific to OprJ. Antimicrob Agents Chemother 1995;39: 1731-5.

7. Kohler T, Michea-Hamzehpour M, Henze U, Gotoh N, Curty LK, Pechere JC. Chardcteriza- tion of MexE-MexF-OprN, a positively regulated multidrug efflux system of Pseudomonas aerugi- nosa. Mol Microbiol 1997;23:345-54.

8. Li XZ, Nikaido H, Poole K. Role of mexA- mexB-oprM in antibiotic efflux in Pseudomonas aeruginosa. Antimicrob Agents Chemother 1995; 39:1948-53.

9. Masuda N, Gotoh N, Ohya S, Nishino T. Quanti- tative correlation between susceptibility and OprJ production in NfxB mutants of Pseudomon- as aeruginosa. Antimicrob Agents Chemother

10. Poole K, Gotoh N, Tsujimoto H, Zhao Q, Wada A, Yamasaki T, Neshat S, Yamagishi J, Li XZ, Nishino T. Overexpression of the mexC-mexD- oprJ efflux operon in nfxB-type multidrug-resis- tant strains of Pseudomonas aeruginosa. Mol Microbiol 1996;21:713-24.

1996;40:909-13.

1115

JALAL cr al.

11. Poole K, Krebes K, McNally C, Neshat S. Multiple antibiotic resistance in Pseudomonas aeruginosa: evidence for involvement of an efflux operon. J Bacteriol 1993; 175:7363-72.

12. Poole K, Tetro K, Zhao Q, Neshat S, Heinrichs DE, Bianco N. Expression of the multidrug resis- tance operon mexA-mexB-oprM in Pseudomonas aeruginosa: mexR encodes a regulator of operon expression. Antimicrob Agents Chemother 1996;

13. Srikumar R, Kon T, Gotoh N, Poole K. Express- ion of Pseudomonas aeruginosa multidrug efflux pumps MexA-MexB-OprM and MexC-MexD- OprJ in a multidrug-sensitive Escherichia coli strain. Antimicrob Agents Chemother 1998;42:

14. Wong KK, Poole K, Gotoh N, Hancock RE. In- fluence of OprM expression on multiple anti- biotic resistance in Pseudomonas aeruginosa. Antimicrob Agents Chemother 1997;41:2009-12.

15. Goldrick MM, Kimball GR, Liu Q, Martin LA, Sommer SS, Tseng JY: NIRCA. A rapid robust method for screening for unknown point muta- tions. Biotechniques 1996;21: 106-12.

16. Haseyama K, Kudoh T, Yoto Y, Suzuki NA, Chi- ba s. Analysis of genetic diversity in the VPI

40: 202 1-8.

65-7 1.

unique region gene of human parvovirus B19 using the mismatch detection method and direct nucleotide sequencing. J Med Virol 1998;56:205- 9.

17. Wretlind B, Pavlovskis OR. Genetic mapping and characterization of Pseudomonas aeruginosu mutants defective in the formation of extracellu- lar proteins. J Bacteriol 1984; 158:801-8.

18. Pitcher D, Saunders Na, Owen R. Rapid extrac- tion of bacterial genomic DNA with guanidium thiocyanate. Lett Appl Microbiol 1989;8: 151-6.

19. Heisig I? Genetic evidence for a role of parC mutations in development of high-level fluoro- quinolone resistance in Escherichiu coli. Anti- microb Agents Chemother 1996;40:879-85.

20. Kureishi A, Bryan LE. Pre-boiling high GC con- tent, mixed primers with 3’ complementation allows the successful PCR amplification of Pseudomonas ueruginosa DNA. Nucleic Acids Res 1992; 20: 1 155.

21. Kureishi A, Diver JM, Beckthold B, Schollaardt T, Bryan LE. Cloning and nucleotide sequence of Pseudomonus aeruginosa DNA gyrase gyrA gene from strain PA01 and quinolone-resistant clin- ical isolates. Antimicrob Agents Chemother 1994;38: 194452.

1116