A

1awsaA8bs©

K

1

p2cbofniopf

0d

International Journal of Antimicrobial Agents 30 (2007) 222–228

Complete nucleotide sequence of the gyrA gene of Helicobacterpullorum and identification of a point mutation leading to

ciprofloxacin resistance in poultry isolates

Frederique Pasquali a,∗, Mirko Rossi b, Gerardo Manfreda a, Renato Zanoni b

a Department of Food Science, Alma Mater Studiorum—University of Bologna, San Giacomo 9, 40127 Bologna, Italyb Department of Veterinary Public Health and Animal Pathology, Alma Mater Studiorum—University of Bologna,

via Tolara di Sopra 50, 40064 Ozzano Emilia (BO), Italy

Received 19 January 2007; accepted 29 April 2007

bstract

To assess the molecular basis of nalidixic acid and ciprofloxacin resistance in Helicobacter pullorum, the gyrA gene of H. pullorum CIP04787T was sequenced. In addition, 9 isolates (2 susceptible to ciprofloxacin and resistant to nalidixic acid, 3 susceptible and 4 resistant to bothntibiotics) were selected from 44 poultry isolates and the nucleotide sequences of their quinolone resistance-determining regions (QRDRs)ere compared. The 2490 bp gyrA gene showed an open reading frame encoding a polypeptide of 829 amino acids. The deduced amino acid

equence of gyrA showed ≥72% identity to Helicobacter hepaticus, Helicobacter pylori and Wolinella succinogenes. Moreover, ≥98% aminocid sequence identity was found comparing the QRDR of the H. pullorum type strain with the QRDRs of the aforementioned bacterial species.ll ciprofloxacin-resistant poultry isolates showed an ACA → ATA (Thr → Ile) substitution at codon 84 of gyrA, corresponding to codons 86,

7 and 83 of Campylobacter jejuni, H. pylori and Escherichia coli gyrA genes, respectively. This substitution was functionally confirmed toe associated with the ciprofloxacin-resistant phenotype of poultry isolates. This is the first report describing the complete 2490 bp nucleotideequence of H. pullorum gyrA and confirming the involvement of the Thr84Ile substitution of GyrA in ciprofloxacin resistance of H. pullorum.

2007 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

ce

v[tdatwicgo

eywords: Helicobacter pullorum; gyrA; QRDR; Fluoroquinolone resistan

. Introduction

The genus Helicobacter, belonging to the class Epsilon-roteobacteria, was created in 1989 and currently comprises3 validly published species of microaerobic, Gram-negative,urved, spiral or fusiform bacteria. Helicobacter spp. haveeen found in the intestinal tract, oral cavity and internalrgans of man and animals and may be associated with dif-erent diseases, generally related to their natural ecologicaliche [1]. Helicobacter pullorum, a urease-negative organ-sm with unsheathed flagella, was classified as a new species

f Helicobacter by Stanley et al. on the basis of 16S rRNAhylogenetic analysis [2]. This organism has been isolatedrom the livers and intestinal contents of laying hens with

∗ Corresponding author. Tel.: +39 051 209 4223; fax: +39 051 251 936.E-mail address: fpasquali@disa.unibo.it (F. Pasquali).

ccphui

924-8579/$ – see front matter © 2007 Elsevier B.V. and the International Societyoi:10.1016/j.ijantimicag.2007.04.017

ibrionic hepatitis and from the caeca of broiler chickens2–4]. In Belgium, one-third of live chickens were reportedo be colonised by H. pullorum [5]. An Italian study con-ucted on 42 farms showed a prevalence of H. pullorum >78%mong 209 chicken caecal contents analysed [4]. Helicobac-er pullorum has also been isolated from the faeces of humansith gastroenteritis [6,7], and its DNA has been detected

n livers from patients with primary sclerosing cholangitis,irrhosis and hepatocellular carcinoma [8,9]. In a recent Bel-ian study, Ceelen et al. detected H. pullorum DNA in 4.3%f patients with gastrointestinal disease and from 4.0% oflinically healthy persons [10]. As poultry carcasses can beontaminated by H. pullorum during slaughtering [3], the

otential role of this bacteria as an emerging food-borneuman pathogen needs to be considered, although it wasnclear whether the organism had a causal role in humannfections [11].

of Chemotherapy. All rights reserved.

l of An

aiioivabtnogmlfo

capp

poabHtg

2

2

Pr

tancEma(

2

ca(CCdfirSCiwfmwsUsb77bf

TC

I

CHHHHHHHHH

ME

F. Pasquali et al. / International Journa

Few data are currently available on the incidence ofntibiotic resistance in H. pullorum [12]. Among the var-ous antimicrobial resistances, fluoroquinolone resistances of particular interest since clinical use of this antibi-tic both in human and veterinary medicine has led to anncrease in the prevalence of fluoroquinolone resistance inarious Epsilonproteobacteria [13,14]. Among Helicobacter-ceae, only ciprofloxacin-resistant H. pylori isolates haveeen described and characterised [15,16]. In this species,he fluoroquinolone resistance character was associated withucleotide point mutations at positions Asn87 and Asp91f the quinolone resistance-determining region (QRDR) ofyrA coding for subunit A of DNA gyrase. These pointutations correspond to codons 86 and 90 of Campy-

obacter jejuni gyrA. They have been reported as the mostrequently detected mutations associated with high-level flu-roquinolone resistance in Campylobacter [17].

The gyrA nucleotide sequences of Helicobacter hepati-us, Helicobacter acinonychis and Wolinella succinogenesre available in the GenBank database, whereas the com-lete nucleotide sequence of H. pullorum gyrA has never beenublished.

In the present study, we cloned and sequenced gyrA of H.ullorum CIP 104787T and sequenced the QRDRs of gyrAf a total of 9 poultry isolates (2 susceptible to ciprofloxacinnd resistant to nalidixic acid, 3 susceptible and 4 resistant tooth antibiotics) selected from among 44 poultry isolates of. pullorum in order to investigate point mutations leading

o quinolone and fluoroquinolone resistance in this microor-anism.

. Materials and methods

.1. Bacterial strains

Helicobacter pullorum CIP 104787T (Institut Pasteur,aris, France) as well as nine poultry isolates of H. pullo-um were tested (Table 1). The isolates were selected from a

3iaC

able 1haracteristics of Helicobacter pullorum isolates

solate MIC (mg/L)

CIP NA AMP CHL TET

IP 104787 (type strain) 0.12 8 4 4 0.25P1 0.12 8 2 4 0.25P5 0.25 64 4 16 4P6 0.25 4 2 2 1P8 <0.015 4 2 8 0.5P9 0.25 32 8 16 2P2 32 32 1 1 0.25P3 32 32 1 2 0.25P4 32 32 4 4 0.5P7 128 128 8 8 128

IC, minimum inhibitory concentration; CIP, ciprofloxacin; NA, nalidixic acid; AMRY, erythromycin; –, no change.

timicrobial Agents 30 (2007) 222–228 223

otal of 44 field isolates, including 2 resistant to nalidixic acidnd 22 resistant to both nalidixic acid and ciprofloxacin. Theine isolates were recovered during 2005 from caecal contentollected from nine different broiler chicken flocks in Italy.ach poultry isolate of H. pullorum was identified by poly-erase chain reaction (PCR) assay as described by Stanley et

l. [2] and by PCR-restriction fragment length polymorphismRFLP) [18].

.2. Antimicrobial susceptibility

The minimum inhibitory concentrations (MICs) ofiprofloxacin (Bayer AG, Leverkusen, Germany), nalidixiccid (Sigma Chemical Co., St Louis, MO), ampicillinSigma Chemical Co.), chloramphenicol (Sigma Chemicalo.), tetracycline (Sigma Chemical Co.), gentamicin (Sigmahemical Co.) and erythromycin (Sigma Chemical Co.) wereetermined for all isolates according to a slightly modi-ed protocol of the agar dilution method for C. jejuni andelated species recommended by the Clinical and Laboratorytandards Institute (CLSI) (formerly known as the Nationalommittee for Clinical Laboratory Standards) [19]. As all the

solates did not grow on Muller–Hinton agar, nutrient agar,hich is the medium recommended by On and Holmes [20]

or tolerance tests for campylobacters, was used to deter-ine the MIC values. In particular, the original methodas modified as follows: (i) the medium used was repre-

ented by Nutrient Broth No. 2 (Oxoid Ltd., Basingstoke,K) supplemented with 1.5% Bacto Agar (Becton, Dickin-

on and Company, Sparks, MD) and 5% defibrinated sheeplood; and (ii) the plates were incubated at 37 ± 1 ◦C for2 h under an increased H2 microaerobic atmosphere (ca.% H2, 7% CO2, 5% O2 and 81% N2) obtained as describedy Bolton et al. [21]. The antibiotic concentrations rangedrom 0.015 mg/L to 128 mg/L. Campylobacter jejuni ATCC

3560 (DSMZ, Braunschweig, Germany) was used as a qual-ty control strain and showed MIC values on nutrient agarlways within the quality control limits indicated by theLSI [19]. Since no breakpoints are currently available for

gyrA mutation

GEN ERY Codon Base change Amino acidchange

0.25 0.5 – – –0.12 0.5 – – –0.25 8 – – –0.25 2 – – –0.12 1 – – –0.5 0.25 – – –0.12 >128 84 ACA → ATA Thr → Ile0.12 >128 84 ACA → ATA Thr → Ile0.12 0.25 84 ACA → ATA Thr → Ile0.25 >128 84 ACA → ATA Thr → Ile

P, ampicillin; CHL, chloramphenicol; TET, tetracycline; GEN, gentamicin;

224 F. Pasquali et al. / International Journal of An

Table 2Primers used in the study

Primer Sequence (5′ → 3′) Amplicon size (bp)

dHpgyrAF CCTGTKCATAGRMGWATYTT 545dHpgyrAR TARATDATYCCACCWGTWGGHpgyrA24F TGCAATGTATGAGCTTGGT 498HHH

Hbn

2

Qgn(oJAw1daO(a(3a14s

cTaw(u(pwb

guTecwH

wd

(lsaacda

aaQb

Ca

2

lwDw1doom3fTp

lu

2

tCftLt(

2

pgyrA521R TCTGGACCTTCAACAAATTGpgyrA74F CAAGAATCGTGGGTGATG 351pgyrA424R GTGGAATATTTGTCGCCA

. pullorum, we tentatively used Enterobacteriaceaereakpoints (ciprofloxacin resistance breakpoint ≥4 mg/L;alidixic acid resistance breakpoint ≥32 mg/L) [22].

.3. Molecular cloning of the gyrA gene

A degenerate primer set was designed to amplify theRDR of H. pullorum CIP 104787T gyrA. The primers (dHp-yrAF and dHpgyrAR) (Table 2) were selected based onucleotide sequence alignment of gyrA conserved regionsfrom ca. codon 42 to codon 232), including the QRDRf gyrA of H. hepaticus ATCC 51449, Helicobacter pylori99 and C. jejuni NCTC 11168 (GenBank accession nos.E017125, AE001439 and AL111168, respectively). PCRas carried out in a 50 �L reaction mixture containing00 ng of DNA template, 1× HotMaster Taq buffer (Eppen-orf AG, Hamburg, Germany), 0.1 mg/mL of bovine serumlbumin (BSA) (Fermentas International Inc., Burlington,N, Canada), 200 �M of each deoxynucleoside triphosphate

dNTP) (Eppendorf AG), 2 �M of each primer (dHpgyrAFnd dHpgyrAR) and 2 U of HotMaster Taq DNA polymeraseEppendorf AG). Thermocycling conditions were 94 ◦C formin followed by 5 cycles of 94 ◦C for 1 min, 30 ◦C for 1 minnd 72 ◦C for 1 min, 5 cycles of 94 ◦C for 1 min, 35 ◦C formin and 72 ◦C for 1 min, and 20 cycles of 94 ◦C for 1 min,0 ◦C for 1 min and 72 ◦C for 1 min, with a final extensiontep of 72 ◦C for 10 min.

The PCR product of H. pullorum CIP 104787T wasloned into pCR®-4-TOPO® plasmid vector using a TOPO-A cloning kit (Invitrogen Life Technologies, Carlsbad, CA)ccording to the manufacturer’s instructions. Plasmid DNAas isolated using Quantum prep® Plasmid Miniprep Kit®

Bio-Rad Laboratories, Hercules, CA) according to the man-facturer’s instructions and was sequenced by Primm srlMilan, Italy) on both strands using M13 forward and reverserimers. The deduced amino acid sequence was comparedith GyrA amino acid sequences of other Gram-negativeacteria.

The upstream and downstream sequences of the QRDR ofyrA were amplified by cassette ligation-mediated PCR [23]sing LA PCR in vitro cloning kit (Takara, Shiga, Japan).his method is a nested PCR based on two subsequent PCRs,

ach one performed using a QRDR-specific primer and aassette-specific primer. Based on the sequence amplifiedith degenerate PCR, four specific primers (HpgyrA24F,pgyrA521R, HpgyrA74F and HpgyrA424R) (Table 2)

pg

timicrobial Agents 30 (2007) 222–228

ere drawn and used in the cassette-ligation PCR proce-ure.

Total genomic DNA was digested with PstI and EcoRIFermentas International Inc.). PstI and EcoRI digests wereigated with PstI and EcoRI cassettes, respectively, both cas-ettes containing complementary sequences of cassette C1nd cassette C2 primers. The ligation products were useds template for PCR. The first PCRs were performed withassette C1 and HpgyrA24F primers to amplify the QRDRownstream, and cassette C1 and HpgyrA521R primers tomplify the QRDR upstream.

The second nested PCRs were carried out with cassette C2nd HpgyrA74F primers to amplify the QRDR downstream,nd cassette C2 and HpgyrA424R primers to amplify theRDR upstream. The second PCR products were sequencedy primer walking strategy (Primm srl).

The complete nucleotide sequence of gyrA of H. pullorumIP 104787T was deposited in the GenBank database underccession no. DQ836338.

.4. PCR amplification of the QRDR

The QRDRs of gyrA of nine H. pullorum poultry iso-ates were amplified, sequenced and compared by alignmentith the QRDR of gyrA of H. pullorum CIP 104787T. ANA fragment of 498 bp containing the QRDR was amplifiedith specific primers in a 50 �L reaction mixture containing00 ng of DNA template, 1× HotMaster Taq Buffer (Eppen-orf AG), 200 �M of each dNTP (Eppendorf AG), 0.5 �Mf each primer (HpgyrA24F and HpgyrA521R) and 2.5 Uf HotMaster Taq DNA polymerase (Eppendorf AG). Ther-ocycling conditions were 94 ◦C for 3 min followed by

0 cycles of 94 ◦C for 1 min, 50 ◦C for 1 min and 72 ◦Cor 1 min with a final extension step of 72 ◦C for 10 min.he amplicons were sequenced by Primm srl using PCRrimers.

The 120 bp QRDR nucleotide sequence of gyrA of H. pul-orum CIP 104787T was deposited in the GenBank databasender accession no. DQ675017.

.5. Transformation of H. pullorum

Transformation of H. pullorum was performed accordingo a previously reported protocol [15]. Helicobacter pullorumIP 104787T was chosen as the recipient cell. The DNA used

or transformation was obtained following purification withhe Quantum Prep PCR Clean Spin Columns Kit® (Bio-Radaboratories) of the PCR products obtained by amplifica-

ion of the QRDRs of gyrA of isolates HP2, HP3 and HP7Table 1).

.6. Sequence analysis

Sequence analysis was performed with the BLASTrograms BLASTn and BLASTp (http://www.ncbi.nlm.nih.ov/BLAST) and with the Open Reading Frame (ORF)

l of An

FhweuB

3

3

tMcowra3rcc

wttenwwwat(cn

3

fafpQArasn5

tH

wE

3slr

snsOp

7Hio

istcaHi

3a

hcttto

nHoatcs(

3p

m

F. Pasquali et al. / International Journa

inder program (http://www.ncbi.nlm.nih.gov/gorf/gorf.tml). Alignments of nucleotide and amino acid sequencesere performed with the ClustalW program (http://www.

bi.ac.uk/clustalw/index.html). Putative bacterial promoterspstream of H. pullorum gyrA were analysed using thePPROM program (http://www.softberry.com).

. Results

.1. Antimicrobial susceptibility

The MIC values of the nine isolates tested and of theype strain H. pullorum CIP 104787T are shown in Table 1.

onomodal courses for the MICs were found for ampicillin,hloramphenicol and gentamicin. Bimodal courses werebserved for erythromycin, tetracycline and nalidixic acidith a second peak at >128 mg/L, 128 mg/L and 32 mg/L,

espectively, whilst the MICs values for ciprofloxacin showedmultimodal course with peaks at <0.015 mg/L, 0.25 mg/L,2 mg/L and 128 mg/L. These results suggest an acquiredesistance to erythromycin, tetracycline, nalidixic acid andiprofloxacin and susceptibility to ampicillin, chlorampheni-ol and gentamicin.

Based on CLSI breakpoints for Enterobacteriaceae [22],e may assume that the type strain as well as all five poul-

ry isolates susceptible to ciprofloxacin were also susceptibleo ampicillin, chloramphenicol, tetracycline, gentamicin andrythromycin (Table 1). Among all four ciprofloxacin- andalidixic acid-resistant isolates, two isolates (HP2 and HP3)ith ciprofloxacin and nalidixic acid MIC values of 32 mg/Lere also resistant to erythromycin (MIC = 256 mg/L),hereas the resistant isolate HP7, showing ciprofloxacin

nd nalidixic acid MICs of 128 mg/L, was also resis-ant to tetracycline (MIC = 128 mg/L) and erythromycinMIC = 256 mg/L) (Table 1). Isolates HP5 and HP9 were sus-eptible to ciprofloxacin (MIC = 0.25 mg/L) but resistant toalidixic acid (MIC = 64 mg/L and 32 mg/L, respectively).

.2. Sequence analysis of H. pullorum gyrA gene

Using the degenerate PCR assay, a 545 bp fragmentrom H. pullorum CIP 104787T chromosomal DNA wasmplified. The deduced amino acid sequence of the 545 bpragment showed 78% and 86% sequence identity in com-arison with an analogous 545 bp fragment containing theRDR of gyrA of H. hepaticus ATCC 51449 (accession no.E017125) and H. pylori UC946 (accession no. L29481),

espectively. This sequence also showed 83% and 73% aminocid sequence identity to W. succinogenes DSM 1740 (acces-ion no. NC 005090) and C. jejuni NCTC 11168 (accessiono. AL111168), respectively. These results confirm that the

45 bp PCR product contains the QRDR of H. pullorum gyrA.

Sequence analysis of the 545 bp fragment allowedwo specific primer sets to be designed, HpgyrA24F-pgyrA521R and HpgyrA74F-HpgyrA424R (Table 2), that

lctt

timicrobial Agents 30 (2007) 222–228 225

ere included in a cassette-ligation PCR following PstI andcoRI digestion of the genomic DNA.

PstI cassette-ligation PCR amplified two fragments of ca.kb and 4 kb corresponding to the upstream and downstream

equences of the QRDR of H. pullorum gyrA. EcoRI cassette-igation PCR failed to amplify QRDR surrounding nucleotideegions.

The complete nucleotide and deduced amino acidequence of gyrA of H. pullorum CIP 104787T (accessiono. DQ836338) is shown in Fig. 1. The 2490 bp gyrA genehowed an ORF encoding 829 amino acids. Upstream of thisRF, a putative ribosome binding site [24] and a putativeromoter were found (Fig. 1).

The deduced amino acid sequence of this ORF showed6% and 72% sequence identity with the GyrA protein of. hepaticus and H. pylori, respectively. Moreover, sequence

dentity of 83% and 68% was found compared with the GyrAf W. succinogenes and C. jejuni, respectively.

The deduced QRDR of the GyrA of H. pullorum wasdentical to those of H. hepaticus and W. succinogenes, andhowed 98%, 83% and 80% amino acid sequence identity tohe QRDR of the GyrA of H. pylori, C. jejuni and Escherichiaoli K12 (accession no. U00096), respectively. The aminocid sequences of the QRDRs of H. pullorum, H. hepaticus,. pylori, C. jejuni, W. succinogenes and E. coli are shown

n Fig. 2.

.3. Characterisation of gyrA mutations in H. pullorumnimal isolates

The primer set HpgyrA24F-HpgyrA521R was shown to beighly specific for amplification of a 498 bp DNA fragmentontaining the QRDR of H. pullorum gyrA. In particular,his primer set was able to amplify all H. pullorum isolatesested, whereas no amplification was observed for Helicobac-er canadensis CCUG 471/63 (Culture Collection, Universityf Goteborg, Goteborg, Sweden) (data not shown).

In comparison with CIP 104787T, ciprofloxacin- andalidixic acid-susceptible poultry isolates (HP1, HP6 andP8) showed wild-type QRDRs of gyrA (Table 1). More-ver, no gyrA mutations were found in both isolates (HP5nd HP9) showing susceptibility to ciprofloxacin and resis-ance to nalidixic acid (Table 1). All isolates resistant to bothiprofloxacin and nalidixic acid (HP2, HP3, HP4 and HP7)howed an ACA → ATA (Thr → Ile) substitution at codon 84Table 1).

.4. Transformation of fluoroquinolone resistance in H.ullorum

To confirm functionally that the ACA → ATA (Thr84Ile)utation detected in ciprofloxacin-resistant poultry iso-

ates leads to ciprofloxacin resistance in H. pullorum, theiprofloxacin-susceptible strain H. pullorum 104787T wasransformed with the PCR products obtained from amplifica-ion of the QRDR of gyrA of isolates HP2, HP3 and HP7. The

226 F. Pasquali et al. / International Journal of Antimicrobial Agents 30 (2007) 222–228

F IP 1047g ing sitet isk. The

pcamfmmH

4

p

ig. 1. Nucleotide sequence of the gyrA gene of Helicobacter pullorum CyrA gene containing the putative promoter and the putative ribosome bindhe 2490 bp gyrA gene. The translational stop codon is indicated by an aster

resence of the ACA → ATA point mutation at codon 84 ofiprofloxacin-resistant transformants was confirmed by PCRnd subsequent sequencing of the QRDR of gyrA of transfor-ants. All nine randomly tested resistant transformants (three

or each DNA donor isolate) showed ACA → ATA pointutation at codon 84, functionally confirming the Thr84Ileutation as being responsible for ciprofloxacin resistance in. pullorum.

slpq

87T (accession no. DQ836338): (a) nucleotide sequence upstream of the(RBS); and (b) nucleotide sequence and deduced amino acid sequence ofquinolone resistance-determining region is highlighted in grey.

. Discussion

Currently, no published data are available on the com-lete nucleotide sequence of H. pullorum gyrA. In the present

tudy, we cloned and sequenced the gyrA gene of H. pul-orum CIP 104787T and screened nine poultry isolates foroint mutations associated with their different levels ofuinolone and fluoroquinolone resistance. The gyrA gene of

F. Pasquali et al. / International Journal of Antimicrobial Agents 30 (2007) 222–228 227

Fig. 2. Alignment of deduced amino acid sequences of the quinolone resistance-determining regions (QRDRs) of the GyrA protein of Helicobacter pullorumC T es DSMa 90, L2h

HHtoa

aDta

rfictH

lic8dEqihr[aTftstrib

istnram

iodp

tm[

tssQcsmrap

aHoihtaHiotcnh

n1qtt

IP 104787 , Helicobacter hepaticus ATCC 51449, Wolinella succinogennd Escherichia coli K12 (accession nos. DQ836338, AE017125, NC 0050ighlighted in grey.

. pullorum showed a higher similarity in comparison with

. hepaticus than with H. pylori. The amino acid sequence ofhe QRDR is homologous (100% sequence identity) to thosef H. hepaticus and W. succinogenes, confirming the QRDRs a conservative region.

The high sequence identity found between H. pullorumnd W. succinogenes agrees with Stanley et al. [2] andewhirst et al. [25] who described a close relationship of

he two species based on whole-cell protein profile analysisnd phylogenetic analysis of 23S rRNA, respectively.

Primers used in this study for amplification of the QRDRegion of gyrA of H. pullorum showed a very high speci-city for H. pullorum and failed to amplify the QRDR of H.anadensis. This finding suggests the possible application ofhese primers for discrimination of the two closely relatedelicobacter species.Only one type mutation was detected in all poultry iso-

ates resistant to both ciprofloxacin and nalidixic acid. Thesesolates showed Thr → Ile substitution at codon 84. Thisodon corresponds to codon 86 of gyrA of C. jejuni, codon7 of gyrA of H. pylori and codon 83 of gyrA of Pseu-omonas aeruginosa, E. coli, Salmonella enterica serovarsnteritidis and Typhimurium. A Thr86Ile substitution has fre-uently been found associated with fluoroquinolone-resistantsolates of Campylobacter [26,27]. Thr87Ile and Thr83Ileave been functionally confirmed as associated with fluo-oquinolone resistance in H. pylori [15] and P. aeruginosa28], respectively. Ser83Ala in E. coli [29] and Ser83Phe,n amino acid substitution analogous to Ser83Ala, in S.yphimurium [30] and S. Enteritidis [31] were identified andunctionally confirmed as associated with quinolone resis-ance and decreased fluoroquinolone susceptibility. In ourtudy, the Thr84Ile mutation was also functionally confirmedo be associated with ciprofloxacin resistance in H. pullo-um, confirming the relevance of mutations at this codonn acquired fluoroquinolone resistance of Gram-negativeacteria.

Interestingly, no mutations in the QRDR of gyrA weredentified in H. pullorum poultry isolates (HP5 and HP9)howing resistance to nalidixic acid and susceptibilityo ciprofloxacin. This finding suggests that resistance to

alidixic acid alone arises independently from ciprofloxacinesistance, as already described for C. jejuni [32]. Nalidixiccid resistance of these isolates may be due to: (i) pointutations in gyrA outside the QRDR; (ii) point mutations

pT

1740, Helicobacter pylori UC946, Campylobacter jejuni NCTC 111689481, AL111168 and U00096, respectively). Amino acid substitutions are

n other target genes (i.e. gyrB); or (iii) point mutations inperons leading to overexpression of efflux pump systems orownregulation of genes coding for membrane proteins (i.e.orins).

No mutation was found at codon 88 (corresponding tohe Asp91 residue in H. pylori), frequently reported as a

utation leading to fluoroquinolone resistance in H. pylori15,16].

No correlation can be drawn between the MICs of poul-ry isolates and the number or type of gyrA mutationsince: (i) resistant isolates with different MICs showed theame nucleotide substitution; and (ii) no mutations in theRDR of gyrA were detected in isolates with decreased

iprofloxacin susceptibility associated with nalidixic acidusceptibility or resistance. This observation suggests thatutations in other target genes or other mechanisms (i.e.

educed membrane permeability) may be involved in thecquisition of quinolone and fluoroquinolone resistance in H.ullorum.

The use of quinolones in poultry can potentially favourn increase in quinolone-resistant pathogens. Among them,. pullorum is of particular interest owing to the isolationf this pathogen both in poultry and humans [4–7], suggest-ng the possible transfer of this pathogen from poultry toumans. In this study, we observed that ciprofloxacin resis-ance in H. pullorum is due to Thr84Ile substitution of gyrAnd that this mutation can be transferred in vitro to other. pullorum recipient cells via natural transformation. Hor-

zontal transfer of antibiotic resistance genes was recentlybserved in vitro from H. pylori to C. jejuni [33], suggestinghe possible in vivo spread of different antibiotic resistanceharacters from Helicobacter to other Epsilobacteria viaatural transformation or conjugation, both in poultry andumans.

In conclusion, this is the first report on the completeucleotide sequence of the gyrA gene of H. pullorum CIP04787T and on QRDR point mutations leading to fluoro-uinolone resistance in this microorganism. Based on naturalransformation, the Thr → Ile substitution at codon 84 leadso acquired ciprofloxacin resistance in H. pullorum.

Funding: This study was financed within the Euro-

ean Project FOOD-CT-200X-007076, named POUL-RYFLORGUT (www.poultryflorgut.org).

Competing interests: None to declared.Ethical approval: Not required.

2 l of An

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

28 F. Pasquali et al. / International Journa

eferences

[1] On SL, Lee A, O’Rourke JL, et al. Genus I. Helicobacter. In: BrennerDJ, Krieg NR, Staley JT, Garrity GM, editors. Bergey’s manual ofsystematic bacteriology, vol. 2. New York, NY: Springer Science andBusiness Media; 2005. Part C.

[2] Stanley J, Linton D, Burnens AP, et al. Helicobacter pullorumsp. nov. —genotype and phenotype of new species isolated frompoultry and from human patients with gastroenteritis. Microbiology1994;140:3441–9.

[3] Atabay HI, Corry JE, On SLW. Identification of unusualCampylobacter-like isolates from poultry products as Helicobacterpullorum. J Appl Microbiol 1998;84:1017–24.

[4] Manfreda G, Rossi M, Sanguinetti V, et al. Prevalence of Helicobacterpullorum in broiler chickens reared in intensive and extensive farms.Worlds Poult Sci J 2006;62:551–2.

[5] Ceelen L, Decostere A, Van den Bulck K, et al. Helicobacter pullorumin chickens, Belgium. Emerg Infect Dis 2006;12:263–7.

[6] Burnens AP, Stanley J, Morgenstern R, Nicolet J. Gastroenteritis asso-ciated with Helicobacter pullorum. Lancet 1994;344:1569–70.

[7] Steinbrueckner B, Haerter G, Pelz K, et al. Isolation of Heli-cobacter pullorum from patients with enteritis. Scand J Infect Dis1997;29:315–8.

[8] Pellicano R, Mazzaferro V, Grigioni WF, et al. Helicobacter speciessequences in liver samples from patients with and without hepatocel-lular carcinoma. World J Gastroenterol 2004;10:598–601.

[9] Ponzetto A, Pellicano R, Leone N, et al. Helicobacter infection andcirrhosis in hepatitis C virus carriage: is it an innocent bystander or atroublemaker? Med Hypotheses 2000;54:275–7.

10] Ceelen L, Decostere A, Verschraegen G, Ducatelle R, HaesebrouckF. Prevalence of Helicobacter pullorum among patients with gas-trointestinal disease and clinically healthy persons. J Clin Microbiol2005;43:2984–6.

11] Gibson JR, Ferrus MA, Woodward D, Xerry J, Owen RJ. Geneticdiversity in Helicobacter pullorum from human and poultry sourcesidentified by an amplified fragment length polymorphism technique andpulsed-field gel electrophoresis. J Appl Microbiol 1999;87:602–10.

12] Ceelen L, Decostere A, Devriese LA, Ducatelle R, Haesebrouck F.In vitro susceptibility of Helicobacter pullorum strains to differentantimicrobial agents. Microb Drug Resist 2005;11:122–6.

13] Glocker E, Stueger HP, Kist M. Quinolone resistance in Helicobacterpylori in Germany. Antimicrob Agents Chemother 2006;51:346–9.

14] Luber P, Wagner J, Hahn H, Bartelt E. Antimicrobial resistance inCampylobacter jejuni and Campylobacter coli strains isolated in 1991and 2001–2002 from poultry and humans in Berlin, Germany. Antimi-crob Agents Chemother 2003;47:3825–30.

15] Moore RA, Beckthold B, Wong S, Kureishi A, Bryan LE.Nucleotide sequence of the gyrA gene and characterization ofciprofloxacin-resistant mutants of Helicobacter pylori. AntimicrobAgents Chemother 1995;39:107–11.

16] Wang G, Wilson TJ, Jiang Q, Taylor DE. Spontaneous mutations thatconfer antibiotic resistance in Helicobacter pylori. Antimicrob Agents

Chemother 2001;45:727–33.

17] Piddock LJ, Ricci V, Pumbwe L, Everett MJ, Griggs DJ. Flu-oroquinolone resistance in Campylobacter species from man andanimals: detection of mutations in topoisomerase genes. J AntimicrobChemother 2003;51:19–26.

[

timicrobial Agents 30 (2007) 222–228

18] Fox JG, Chien CC, Dewhirst FE, et al. Helicobacter canadensis sp.nov. isolated from humans with diarrhea as an example of an emergingpathogen. J Clin Microbiol 2000;38:2546–9.

19] National Committee for Clinical Laboratory Standards. Performancestandards for antimicrobial disk and dilution susceptibility tests forbacteria isolated from animals. 2nd ed. Approved standard M31-A2.Villanova, PA: NCCLS; 2002.

20] On SL, Holmes B. Reproducibility of tolerance tests that areuseful in the identification of campylobacteria. J Clin Microbiol1991;29:1785–8.

21] Bolton FJ, Wareing DRA, Skirrow MB, Hutchinson DN. Identificationand biotyping of campylobacters. In: Board RG, Jones D, Skin-ner FA, editors. Identification methods in applied and environmentalmicrobiology. Oxford, UK: Blackwell Scientific Publications; 1992. p.151–61.

22] National Committee for Clinical Laboratory Standards. Performancestandards for antimicrobial susceptibility testing. Fourteenth Infor-mational Supplement. Approved standard M100-S14. Villanova, PA:NCCLS; 2004.

23] Okuda J, Hayakawa E, Nishibuchi M, Nishino T. Sequence analysis ofthe gyrA and parC homologues of a wild-type strain of Vibrio para-haemolyticus and its fluoroquinolone-resistant mutants. AntimicrobAgents Chemother 1999;43:1156–62.

24] Shine J, Dalgarno L. The 3′-terminal sequence of Escherichia coli16S ribosomal RNA: complementary to nonsense triplets and ribosomebinding sites. Proc Natl Acad Sci USA 1974;71:1342–6.

25] Dewhirst FE, Shen Z, Scimeca MS, et al. Discordant 16S and 23SrRNA gene phylogenies for the genus Helicobacter: implicationsfor phylogenetic inference and systematics. J Bacteriol 2005;187:6106–18.

26] Wang Y, Huang WM, Taylor DE. Cloning and nucleotide sequence ofthe Campylobacter jejuni gyrA gene and characterization of quinoloneresistance mutations. Antimicrob Agents Chemother 1993;37:457–63.

27] Zirnstein G, Helsel L, Li Y, Swaminathan B, Besser J. Characteri-zation of gyrA mutations associated with fluoroquinolone resistancein Campylobacter coli by DNA sequence analysis and MAMA PCR.FEMS Microbiol Lett 2000;190:1–7.

28] Mouneimne H, Robert J, Jarlier V, Cambau E. Type II topoisomerasemutations in ciprofloxacin-resistant strains of Pseudomonas aerugi-nosa. Antimicrob Agents Chemother 1999;43:62–6.

29] Yoshida H, Kojima T, Yamagishi J, Nakamura S. Quinolone-resistantmutations of the gyrA gene of Escherichia coli. Mol Gen Genet1988;211:1–7.

30] Piddock LJ, Griggs DJ, Hall MC, Jin YF. Ciprofloxacin resistance inclinical isolates of Salmonella typhimurium obtained from two patients.Antimicrob Agents Chemother 1993;37:662–6.

31] Griggs DJ, Hall MC, Jin YF, Piddock LJ. Quinolone resistancein veterinary isolates of Salmonella. J Antimicrob Chemother1994;33:1173–89.

32] Jesse TW, Englen MD, Pittenger LG, Cray PJ. Two distinct mutationsin gyrA lead to ciprofloxacin and nalidixic acid resistance in Campy-

lobacter coli and Campylobacter jejuni isolated from chickens and beefcattle. J Appl Microbiol 2006;100:682–8.

33] Oyarzabal OA, Rad R, Backert S. Conjugative transfer of chromo-somally encoded antibiotic resistance from Helicobacter pylori toCampylobacter jejuni. J Clin Microbiol 2007;45:402–8.