Cloning and nucleotide sequence of the Pseudomonas aeruginosa nfxB gene, conferring resistance to new quinolones

FEMSMicrobiologyLetters97 (1992) 197-202 © 1992Federationof European MicrobiologicalSocieties0378-1097/92/$05.00 Pablished by Elsevier

197

FEMSLE05089

Cloning and nucleotide sequence of the Pseudomonas aeruginosa nfxB gene, conferring resistance to new quinolones Takashi Okazaki and Keiji Hirai

CentralResearchLaboratories,KyorinPharmaceuticalCo. Ltd., Shimotsuga.gun, Tochigi,Japan Received9 June 1992 Revisionreceived 13July 1992 Accepted 20 July 1992 Key words: Pseudomonas aeruginosa; New quinolone antimicrobial agent; Membrane permeability; DNA sequence

1. SUMMARY The gene nfxB is one of the genes which affect the cell membrane permeability of quinolones in Pseudomonas aeruginosa PAO. Both wild-type nfxB and a mutant nfxB (nfxl3E) were cloned and the DNA sequences were determined. The wild-type gene was dominant in PAO strains. The nfxB mutation was a point mutation (cytosine - , guanine) which generates an amino acid exchange (arginine --* glycine) in the putative nfxB product. The amino acid sequence of the wild-type NfxB protein reveal¢d that it has a helix-turn-helix motif which may be responsible for the ability to bind in a sequence-specific manner to DNA. This finding indicated that the NfxB protein may regulate the expression of genes that are associated with cell permeability of drugs in P. aeruginosa. The position of the amino acid substitution between the NfxB protein and the Nfx13E protein

Correspondenceto: K. Hirai, Central Research Laboratories, KyorinPharmaceuticalCo. Ltd., 2399-1,Nogi-machi.Shimotsuga-gun,Tochigi,329-01,Japan.

was located within a possible DNA-binding domain, suggesting that the mutant protein (Nfxl3E) may have lost DNA binding ability and regulator activity.

2. INTRODUCTION New quinolones, such as norfloxacin and ciprofloxacin, have potent in vitro and in vivo antibacterial activity against Gram-positive and Gram-negative bacteria [1]. It is known that quinolone-resistance in Gram-negative bacteria, including Pseudomonas aeruginosa, involves alterations in DNA gyrase and in the bacterial cell membrane permeability [2-6]. The nfxB mutation in P. aeruginosa PAO is one of the quinolone-resistant mutations affecting cell membrane permeability. The mutant showed an increase in resistance to new quinolones and hypersusceptibility to ~-Iactam and aminoglycoside antibiotics [5]. A new 54-kDa outer membrane protein appeared in the nfxB mutants of P. aeruginosa PAO [5].

To gain more detailed information on the resistance mechanism to new quinolones in P. aeruginosa we have analysed the nfxB mutation. Previously, we had cloned a D N A fragment that can complement the nfxB mutation in P. aeruginosa P A O [7]. Using this fragment as a probe, we have isolated the nfxB mutant allele. In this report, we describe the sequence and some characteristics of these D N A fragments and discuss the tentative functions of the NfxB protein as deduced from its amino acid sequence.

3.3. Preparation of chromosomal and plasmid DNA and transformation Preparation of chromosomal and p)asmid D N A and transformation were performed by the methods described previously [7]. Restriction nucleases and D N A modification enzymes were purchased from Takara Shuzo Co. (Kyoto, Japan). T h e D N A fragments were extracted from the gel by D N A P R E P (Asahi Glass, Tokyo, Japan), which is a D N A extraction system using glass powder.

3.4. Nucleotide sequence analysis 3. M A T E R I A L S A N D M E T H O D S

3.1. Bacterial strains and plasmids Bacterial strains and vector plasmids used in this study are shown in Table 1.

3.2. Antimicrobial agents and drug-susceptibility tests Norfloxacin was synthesized in our laboratories. Ampicillin and carbenicillin were obtained from commercial sources. Antimicrobial susceptibility ~vas measured by agar dilution method described previously [4,5].

The nucleotide sequence of nfxB was determined by the dideoxynucleotide chain termination method of Sanger et al. [8] using the SEQ U E N A S E sequencing kit ( U n i t e d State Biochemical Co., USA).

3.5. Expression o f the nfxB gene T h e 1.1-kb D N A fragment containing the nfxB g e n t in pNF253 was expressed in an in vitro translation system using a prokaryotic DNA-directed translation kit ( A m e r s h a m International, UK) according to the manufacturer's manual. The molecular masses of the products were checked

Table I Bacterial strains and plasmids Strain or p l a s m i d

Description

Source [reference]

Strains

P. aeruginosa PAO PAOI PAO2142 PAO4009 KH4013E PKI013E

Prototroph

ilr.9001, lys-12, met-9011, tyr-9009 FP5 *. leu.9018, nir.9006 nfxB mutant (nfx13E) of PAO4009 nfxl3E transconjugant of PAO2142

H. Matsumoto [5] H. Matsumoto [5) This laboratory [5] This laboratory [7]

E. coli HB10I

F-. hsd20, ara-14,proA2, lacYl, gaIK2, rpsL20,xyl-5, mtl-l, supE44

Plasmids

P. aeruginosa pMT45 pNF225

Cb r, insertion of a multiple cloning site from pUCII9 (derived from pME294) Cb r. insertion of 2.2-kb DNA fragment of PAOI that complements nfxB mutation (derived from pMT45)

E. coli pUC119

Ap r

Abbreviations: Cbr, carbenicillin resistance: Ap r. ampicillin resistance.

This laboratory [7] This laboratory [7]

199

by sodium dodecyl suifate-polyacrylamide gel electrophoresis (SDS-PAGE).

l~r~t~on

to

pNF2$3

The GenBank accession number for the nucleotide sequence of the P. aeruginosa nfxB gene shown in Fig. 2 is X65646 in the EMBL Data Library.

~227 ~2~ pNn4~5 pNX653

We previously cloned a 2.2-kb DNA fragment from a wild-type strain (PAOI) that complements the nfxB mutation into a plasmid named pNF225 [7]. In order to "determine whether this DNA fragment contains the wild-type nfxB, we cloned a region from the chromosomal DNA of the nfxB mutant, KH4013E, that hybridized to pNF225. An EcoRI digest of KI-14013E chromosomal DNA was separated in a 1% agarose gel by electrophoresis and the region containing DNA of about 2.2-kb was cut out. The DNA fragments were extracted from the gel by DNA PREP. The 2.2-kb EcoRI fragments from KH4013E chromosomal DNA were ligated to pUC119 (EcoRI site) and introduced to E. coli HBI01 by transformation. About 200 ampicillin-resistant transformants were cultured and plasmid DNA was prepared from each transformant. These plasmids were digested with EcoRI and XhoI, and analysed by agarose gel electrophoresis. Plasmids having the same size restriction fragments as the pNF225 insert (1.5-kb and 0.7-kb, Fig. 1) were screened by hybridization with the 2.2-kb insert of pNF225. A plasmid, pNR25, containing a 2.2-kb fragment from the KH4013E chromosome had EcoRI-Xhol fragments of 1.5 kb and 0.7 kb and it was strongly hybridized by the pNF225 insert. The 2.2-kb insert of pNR25 was ligated into EcoRI-digested pMT45, resulting in the plasmids designated pNR425 and pNR426. These plasmids had different orientations of the insert relative to the vector. The plasm,.'ds pNR425 and pNR426 were introduced into strain PK1013E, a nfxB mutant of PAO2142, by transformation. Susceptibifity of PK1013E to norfloxacin was restored by introduc-

Coml~'men4otiml

~F22S

3.6. Nucleotide sequence accession number

4. RESULTS AND DISCUSSION

map

E;

.

.

E..

.

.

,

~ip,

E.s.a

+

...~

~

i

.... g

g ~

,,~

E .

.~~

_

oz~

_

Fig. I. The 2.2-kb DNA fragment containing nfrB: restriction map. subcloning strategy and DNA fragment exchange. B. F, Sa, Sp and X indicate restriction sites for B a m H L EcoRI. Sail, Sphl and ,1Otoi. respectively. White and black bars represent the fragment from P A P ! (wild-type) and KH4013E (nfxB mutant), respectively. The symbol+indicates that the plasmid possesses the ability to restore norfloxacin resistance of PKI013E (nfxB mutant) to the normal level. Transformants with pNX653 and pNX654 were cultured in 4 0 0 / t g / m l of carbenicillin, because these plasmids were unstable in PK1013E

tion of pNF225, but pNR425 and pNR426 could not restore quinolone-susceptibility of PK1013E (Fig. 1). TIt~: result indicates that the 2.2-kb DNA fragment from KH4013E has one or more mutations. To restrict the nfxB region in the pNF225 insert, various deletions of the 2.2-kb EcoRI fragment on oNF225 were prepared by Exonuclease III digestion (Fig. 1). One of the resultant plasmids, pNR253, having a 1.l-kb insert of the 2.2-kb fragment of pNF225, had complementation activity to the nfxB mutation in PK1013E, indicating that nfxB may exist in this 1.1-kb region. To further restrict the differences between wild-type and mutant nfxB, plasmids oNX653 and pNX654 were constructed from pNR253 by exchanging wild-type and mutant DNA fragments (Fig. 1). In the case of pNX653, the 0.6-kb SalI-SalI fragment of pNF253 was replaced by the same region from pNR425. For pNX654, the XhoI-SalI 0.1-kb fragment was replaced by that from pNR425. Neither of these plasmids could complement the nfxB mutation (Fig. 1). This result demonstrated that the mutation(s) of nfxl3E (nfxB mutation of KH4013E) e',dsted in the 0.1-kb XhoI-Sall region.

200

The DNA sequence of the l.l-kb region in pNF253, and the same region of pNR425, were determined. The DNA sequences are shown in Fig. 2. There was one base difference between wild-type nfxB and nfxl3E in the XhoI-Sall 0.1-kb region. Computer analysis revealed that there are two open reading frames (ORFs) which are in opposite directions, one (ORF1) extending from position 271 to 831 and second (ORF2) from position 807 to 202 in 1.1-kb DNA fragment (Fig. 2). The ORF1 was considered to encode the NfxB polypeptide, because the nucleotide substitution (C to G at position 394) did not cause an amino acid change between NfxB and Nfx13E polypeptides in ORF2. The nfxl3E mutation is a transition from C to G at position 394, which

results in an amino acid chance from Arg to Giy at amino acid 42 in the ORF1. As a putative promoter of ORFI, we predict that the sequence "ITGTCT at nucleotides 175180 and GCTAAT at nucleotides 198-203 are assigned to the - 3 5 region and the - 1 0 region, respectively. These sequences are homologous to the consensus sequences (TrGACA and TATAAT) for the - 3 5 and - 1 0 regions of E. coil promoters [9]. A putative Shine-Dalgarno sequence GGGA is found 7 bp upstream of the initiation codon of ORF1. ORF1 encoded a polypeptide of 187 amino acids with a predicted M, value of 21076. An in vitro transcription-translation experiment using an E. coli extract revealed that the pNF253 insert

30 O0 AAG•TTGcATG•CCGG•G•••T••TGT•G•T•TT••GC•G••••A•AA•C•GCCAAC•C•ATGG•CATA•CCAAC•CCCCGAT••TTCcT

90

120

150 180 ATTGCACGCAA~TC~GCC~T~ACACACCCGACC~TTGATTG~AACG~TCGA~T~TACTTTTTC~CT~ATT~AGTCAAT~TyGT~T -35

210 24O 270 CAAAT~TCTTTTGACAG~?TCCTTT~G~CGCGA~C~TTTTCT~CAC~AT~CGCA~ATCA~A~CC~ACCGGGACCCkTC~ -IO SO 300

330

360

AT~ACC~TGATTTCCCATGAC~GCGACT~kTC~^GG~GCT~GCAGTCG~T~T~ST~G~CCGC~CGC~AGCG~GCT~A~G~CTGG~C N~tThrLeu~SerHisAspG|uAr~Leu~LysA~aL~uA~a¥a|A~1~ev||AspAr¢PP~ArzA|~Thr~LeuLysGiuLeuA~e

3~0 ~ 420 •A•GCGGCCGGC•TAA•CAA•GCC•CCCTGCACCGCTTCT•CGGCACGC•••ACAACCT•GT•CAGATGCTC•A••ACCAC••A•^••CC

450

G~uA~aA~iG~y~a~SerL~sA~aT~rLeuR~s~r~Phe~ysG~ThrAr~As~A~nLeu~|G~NetLeuG|uAspH|sG~G~uThr G17 (in KH4013[) 480 510 540 GTACTGAACCA•ATCATCCAGGCCT•CGAC•TGGAGCATGCCGAGCCTCT•GA•GCGTTGCAGCGCCT•ATCAA•GAACACCTCACCC•C •••LeuAsnG•n••e••eG|nA••C7sAspLeu••u•|sA|aG•vPr•Leu•••A|•LevG•nAr•Lev••eLysG•uHisLeuThrHis

570 000 030 •GCGAGCTG•TGGTATTCCTGGTATT••AGTAC•GC•CGGA•T•C•T•GA•CCGCA•GGCGAAGG•GCA•G•TGG•AGT••TA•:TGGAA Ar~G~uLeuLeu~PheLeu~heG~nT7rAr¢~r~AspPhsLe~AspPh~isG~7~u~yA~r~TrpG~rT~rLeuG~u 600 090 ••GCTGGACG•ATTCTTC•TGC••GGACAG•AGAAA••CGT•TTTCGCATC•A•ATCAC•GC•GCCGTGTT•A•CGAACTGTT•AT•ACC ~L~AspA~PhePh~Le~Ar~G~n~Lys~Ph~Ar~sp~ThrA~A~h~ThrG~uL~uPh~Thr

720

75O ?80 810 CTGGTCTACGGCATGGTC••TGCGGA•CGTCGCGG•CGGGCGGCCAGCTCC•ATTCCGCGCATACCCTG••GC•••T•TTCCTCCATG•C Le~TyrG~7N~t~AspA~sG~Ar~Ar~yAr~A|~S~rS~rAsnS~rA~H~sThrL~u~u~nN~tPh~L~uH~sG~y 840 870 GCCTCCAATCCG•CT••CT•CTGACCCTC•CGCCCGC••CGGCGCCCCTCT•GTC•CCGCATGGCAC•GACC•ATCCGC•TGCCGT••AC AlsSerAsnProAlaArsSerSSS

900

930 960 990 ~CcG~C~A~C~CT~GA~A~AA~A~CA~CC~CA~CC~A~AAA~CC~A~A~cAAAG~ACC~T~

1020 lena ACGCT~GTCGG~G~TACT~G~ACTT~CCTT~CT~CC~GCATCCTAACT~TCTACCG~TCACACCTCCCCCAG~A~TTC

Fig. 2. Nucleotide sequence of the nfxB gene. The deduced amino acid sequence of the nfxBgene product is given below the D N A sequence. The putative - 3 5 and - 1 0 sequences of the promotor, the Shine-Dalgarno sequence are underlined. The possible DNA-blnding domain is boxed. T h e ::rrow indicates the position of mutation site of nfx13E (in the KH4013E strain), and the changed nucleotide and amino acid are also indicated.

Pro~in ~ i t i ~

NfxB

26

L K E L A E

Lacl DeoR

8 ~

L~]v@Ali Y A ~ ~ V S I y ~ TL_..q | L d A E'~I~A G V Sle qlT

v~IR]v v m i~R d I n

AsnC

25

y

i HIv

E

A

S K

q flO

V Sip

T L H~

,l'I

F C G

r

.

Fig. 3. Comparisonof amino acid sequencesof helix-turn-helixDNA bindingdomain of NfxBprotein and bacterialtranscriptional control proteins. Lacl, DeaR. CytR and AsnC. Amino acid sequence of Lad, DeaR, CytR, and AsnC were taken from helix-turn-helixsequences described by Dadd and Egan [I l]. The amino acid residues of the baclerial transcriptionalcontrol proteins identicalto the correspondingresiduesof the NfxBprotein are shown by capital letter. Identicalresiduesand conservative substitutionsaccordingto the group (I, L, M, V), (D, E) are boxed. The position in the NfxBprotein that was mutated in the Nfxl3 protein (Arg ~ Gly)is encircled. encoded a polypeptide with an estimated M r of approximately 22000 (data not shown). This value agreed with the M r value calculated from the deduced amino acids of the nfxB gene. Analysis of the hydrophobicity of the deduced NfxB polypeptide by the method of Kyte and Doolittle [10] showed that it has no significant hydrophobic membrane-spanning regions, indicating that NfxB protein may not be a membrane-associated protein. However, analysis showed that the NfxB protein had substantial homology to the proteins with a helix-turn-helix motif. This motif is often responsible for the ability of a protein to bind in a sequence-specific manner to DNA [11,12]. By using a systematic method for the detection of A era-like helix-turn-helix DNA-binding motifs in proteins, described by Dodd and Egan [12], we further analysed the amino acid sequence of the NfxB protein. The result demonstrated that the region of amino acid 26-45 of ORF1 is similar to the sequence of helix-turn-helix DNA-binding domains of bacterial regulator (transcriptional control) proteins such as Lacl, DeaR, CytR and AsnC [12] (Fig. 3). Thus, we considered it likely that the NfxB protein regulates gene expression. The wild-type nfxB is dominant in P. aeruginasa, suggesting that NfxB protein negatively regulates the production of outer membrane protein(s) or other membrane component(s). An

arginine residue in the DNA-binding domain was changed to a glycine residue in the nfxB mutant, suggesting that Nfxl3E protein may not be able to bind DNA and thereby lose its regulator function. This de-regulation in the nfxB mutant may cause enhanced expression of the 54-kDa outer membrane protein in P. aeruginosa PAP. The dominance of wild-type nfxB may result from the DNA binding ability of NfxB protein.

ACKNOWLEDGEMENTS We gratefully acknowledge S. lyobe and H. Hashimoto for their interest and comments, M. Fukuda for DNA analysis and homology search. We thank T. Shiba for critical reviews of the manuscript and helpful advice, and H, Fukuda for his excellent collaborations throughout this work.

REFERENCES [1] Wolfson, J.S, and Hopper, D.C. (1985) Antimicrob. Agents Chemother, 28, 581-586. [2l Indue, Y., Sato. K., Fujii,T., Hirai, K., lnoue, M.. lyobe, S. and Mitsuhashi,S. (1987)J, Bacterial. 169.??.322-2325. [3] Hopper, D.C., Wolfson, J.S., Souza, S.. Tang, C., McHugh. G.L. and Swartz, M.N. (1986) Antimicrob. Agents Chemother. 29, 639-644.

202 [4] Hirai. K., Aoyama, H.. Suzue, S., Irikura, T., lyobe, S. and Mitsuhashi, S. (1986) Antimicrob. Agents Chemother. 30. 248-253. [5] Hirai, K., Suzue, S., lrikura, T., lyobe, S. and Mitsuhashi, S. (1987) Antimicrob. Agents Chemother. 3 I, 582-586. [6] Fukuda, H.. Hosaka, M.. Hirai, K. and lyobe, S. (1990) Antimicrob. Agents Chemother. 34. ! 757-1761. [7] Okazaki. T., lyobe. S., Hashimoto, H. and Hirai, K. (1991) FEMS Microbiol. Left. 79, 31-36.

[8] Sanger, F., Nicklen, S. and Coulson, A.R. (1977) Proc. Natl. Acad. Sci. USA 74. 5463-5467. i9] ~cClure, W.R. (1985) Annu. Rev. Biochem. 54, 171-204. ll0] Kyte, J. and Doolittle, R.F. (1982) J. Mol. Biol. 157, 105-132. [11] Dodd, I.B. and Egan, J.B. (1987) J. Mol. Biol. 194, 557-564. [12] Dodd, I.B. and Egan, J.B. (1990) Nucleic Acids Res. 18, 5019-5026.