122

MICROBIAL DRUG RESISTANCEVolume 11, Number 2, 2005© Mary Ann Liebert, Inc.

In Vitro Susceptibility of Helicobacter pullorum Strains to Different Antimicrobial Agents

LIESBETH CEELEN, ANNEMIE DECOSTERE, LUC A. DEVRIESE, RICHARD DUCATELLE, and FREDDY HAESEBROUCK

ABSTRACT

The in vitro activity of 13 antimicrobial agents against 23 Helicobacter pullorum strains from poultry (21) andhuman (two) origin, and one human H. canadensis strain was tested by the agar dilution method. With theH. pullorum strains, monomodal distributions of Minimum Inhibitory Concentrations (MICs) were seen withlincomycin, doxycycline, gentamicin, tobramycin, erythromycin, tylosin, metronidazole, and enrofloxacin inconcentration ranges considered as indicating susceptibility in other bacteria. The normal susceptibility levelfor nalidixic acid was situated at or slightly above the MIC breakpoints proposed for Campylobacteriaceae.Ampicillin, ceftriaxone, and sulphamethoxazole-trimethoprim showed poor activity against H. pullorum. Forthe H. canadensis strain, a similar susceptibility pattern was seen, except for nalidixic acid and enrofloxacin,whose MIC of �512 and 8 �g/ml, respectively, indicated resistance of this agent. With spectinomycin, a bi-modal distribution of the MICs was noted for the tested strains; eight H. pullorum isolates originating fromone flock showed acquired resistance (MIC � 512 �g/ml).

INTRODUCTION

HELICOBACTER PULLORUM was first described by Stanley etal. in 1994.23 The species constitutes a gram-negative,

non-spore forming, gently curved, slender rod with monopolarnon-sheathed flagella.2,18,23 H. pullorum is associated with en-teritis in broiler chickens, vibrionic hepatitis in laying hens, andenteritis and diarrhea in humans.7,23,28 Several authors havesuggested that H. pullorum is involved in the pathogenesis ofchronic liver diseases in humans.1,7,9 Broilers appear to consti-tute the source of infection for humans, due to carcass conta-mination with intestinal contents during the slaughter process.2,7

A significant portion of cases of human diarrhea could havebeen misdiagnosed in the past, because of the specific requi-sites for isolation of the species on the one hand and biochem-ical similarities between the Helicobacter genus and Campy-lobacter genus on the other hand.2,9

Despite the increasing number of reports emphasizing thesignificance of H. pullorum in human beings, hardly any dataabout the antibiotic sensitivity of H. pullorum are available inthe literature. H. pullorum is naturally sensitive to polymyxinB, a phenotypic characteristic distinguishing this species fromthe other Helicobacter spp.2 Resistance to cefalotin and cefop-erazone has been reported.18,23 Different resistance percentages

exhibited by H. pullorum to nalidixic acid were encountered byseveral research groups. On18 and Atabay et al.2 reported 6%and 28% in vitro resistance, respectively, while antimicrobialsusceptibility assays showed 55% resistance to this antimicro-bial agent among the tested strains in a study of Melito et al.15

Thus far, no susceptibility studies comprising widely used an-tibiotics with H. pullorum strains have been reported.

In the present study, the susceptibility of 23 H. pullorumstrains and one H. canadensis strain against 13 antimicrobialagents was investigated by the agar dilution method.

MATERIALS AND METHODS

Bacterial strains

A total of 23 strains of H. pullorum were tested: 21 poultrystrains and two human strains. The poultry strains originatedfrom five different flocks in Belgium and Switzerland. One H.canadensis strain (CCUG 47163) was tested likewise (Table 1).The H. pullorum strains from the Belgian flocks were isolatedfrom cecal samples using the filter technique of Steele and Mc-Dermott.24 The identity of H. pullorum was presumed on thebasis of the bacteria being gram-negative, slightly curved and

Department of Pathology, Bacteriology and Avian Diseases, Faculty of Veterinary Medicine, Ghent University, Merelbeke, Belgium.

rod-shaped, oxidase and catalase positive, indoxyl acetate es-terase negative, and sensitive to polymyxin B. The correct iden-tity of the species was confirmed using a PCR assay developedby Stanley et al.23 Escherichia coli (ATCC 25922) and Staphy-lococcus aureus (ATCC 29213) were included as controlstrains. The Helicobacter strains were inoculated on brain–heartinfusion (BHI) agar (Oxoid, Basingstoke, England), and incu-bation occurred for at least 3 days in a microaerophilic envi-ronment (5% H2, 5% CO2, 5% O2, and 85% N2) at 37°C.

Antimicrobial agents

The following antibiotics were tested: ampicillin, lin-comycin, doxycycline, spectinomycin, gentamicin, tobramycin,ceftriaxone, metronidazole, erythromycin, tylosin, nalidixicacid, and sulphamethoxazole-trimethoprim, all purchased fromSigma (St. Louis, MO) and enrofloxacin (Bayer, Brussels, Bel-gium). Drug solutions were prepared immediately prior to use.

Agar dilution method

The Minimum Inhibitory Concentrations (MICs) of thetested antimicrobial agents were determined by the agar dilu-tion method using BHI agar, containing doubling dilutions ofthe above stated antimicrobials with final concentrations rang-ing from 0.03 to 512 �g/ml. Inocula of the Helicobacter strainswere prepared in phosphate buffered saline by adjusting bacte-rial suspensions directly derived from the culture plate to a den-sity of three on the McFarland turbidity scale. The controlstrains were utilized at a density of 0.5 McFarland standard andconsequently 1:10 diluted. The various strains were seeded on

the antibiotic-containing plates and on antibiotic-free controlplates with a Steers inoculum replicator (MAST, London, UK).Incubation followed in a microaerophilic environment for 3days at 37°C. The MIC was defined as the lowest concentra-tion that almost entirely inhibited growth. The MIC tests wereperformed in duplicate.

RESULTS

The results of the MIC tests are summarized in Table 2. Amonomodal distribution for the MICs was seen for all antimi-crobial agents, except for spectinomycin having a bimodal ap-pearance with a second peak at �512 �g/ml, that is, the MICvalue for all isolates retrieved from flock II. These strains mayhence be designated as having acquired resistance to this an-tibiotic. Lowered susceptibility or resistance to ampicillin, cef-triaxone, and sulphamethoxazole-trimethoprim of all strainswas noted. Susceptibility to lincomycin, doxycycline, gentam-icin, tobramycin, erythromycin, tylosin, metronidazole, en-rofloxacin, and nalidixic acid may be assumed. H. canadensis,however, was resistant to enrofloxacin and nalidixic acid.

DISCUSSION

This study constitutes the first report of in vitro susceptibil-ity testing of H. pullorum and H. canadensis strains against var-ious antimicrobial agents. The selection of these antimicrobialswas based on current recommendations of the National Com-

IN VITRO SUSCEPTIBILITY OF H. PULLORUM TO ANTIMICROBIAL AGENTS 123

TABLE 1. BACTERIAL STRAINS USED IN THIS STUDY

Strain number Source

Reference strains of H. pullorum CCUG 33837 Broiler, SwitzerlandCCUG 33838 Human stool, SwitzerlandCCUG 33839 Human stool, SwitzerlandCCUG 33840 Laying hen, Switzerland

Field strains of H. pullorumisolated from boilers in Belgium

Flock I CE I 1 Cecal dropsCE I 2 Cecal dropsCE I 3 Cecal dropsCE I 4 Cecal dropsB I 1 Farmer’s bootsCE II 1 CecumCE II 2 CecumCE II 3 CecumCE II 4 Cecum

Flock II CE II 5 CecumCE II 6 CecumCE II 7 CecumCE II 8 CecumCE III 1 CecumCE III 2 Cecum

Flock III CE III 3 CecumCE III 4 CecumCE III 5 CecumCE III 6 Cecum

Type strain of H. canadensis CCUG 47163 Human stool, Canada

TA

BL

E2.

MIC

DIS

TR

IBU

TIO

NO

FV

AR

IOU

SA

NT

IMIC

RO

BIA

LS

FO

RH

. P

UL

LO

RU

MA

ND

H.

CA

NA

DE

NSI

SST

RA

INS

Num

ber

of H

. pu

lloru

man

d H

. C

anad

ensi

sst

rain

s w

ith

MIC

of (

�g/

ml)

Ant

imic

robi

al�

0.03

0.06

0.12

0.25

0.5

12

48

1632

6412

825

651

2�

512

Am

pici

llin

810

c2

21

1L

inco

myc

in3

414

3c

Dox

ycyc

line

37

94

1c

Gen

tam

icin

11

1210

c

Spec

tinom

ycin

112

c1

11

8T

obra

myc

in1

49

7c3

Cef

tria

xone

a2c

25

13M

etro

nida

zole

313

7c1

Enr

oflo

xaci

n20

31c

Ery

thro

myc

in11

12c

1T

ylos

in3

57

6c3

Sulp

ham

etho

xazo

le-t

rim

etho

prim

b4

16c

12

Nal

idix

ic a

cid

21

69

51c

a MIC

val

ues

of t

wo

stra

ins

wer

e no

t de

term

ined

.b M

IC v

alue

of

one

stra

in w

as n

ot d

eter

min

ed.

c Inc

ludi

ng t

he H

. ca

nade

nsis

stra

in.

mittee of Clinical Laboratory Standards (M31-A2).16 This listwas completed with other antibiotics that are fairly frequentlyused in human patients with gastrointestinal disease and poul-try. Nalidixic acid was included in the tests to clarify the al-leged susceptibility of H. pullorum against this antibiotic.

According to the NCCLS, the agar dilution is the method ofchoice for testing Helicobacter pylori, Campylobacter jejuni,and related species (M31-A2).17,22 Since Campylobacter andHelicobacter spp. necessitate comparable incubation circum-stances and time, the agar dilution method was used in the pre-sent study to determine the minimum inhibitory concentrationsfor H. pullorum and H. canadensis.

Hitherto, for Helicobacter spp., no internationally acceptedcriteria for susceptibility testing are available. Part of the ex-planation probably lies in the fact that the specific growth re-quirements and the fastidious nature of Helicobacter make itdifficult for establishing standardized MIC determination pro-cedures. The quality control limits given for nonfastidious bac-teria in aerobic environment are not conforming.11 For Campy-lobacter and Arcobacter spp., breakpoint values for a numberof antimicrobials used in the present study (ampicillin, doxy-cycline, gentamicin, tobramycin, ceftriaxone, sulphamethoxa-zole-trimethoprim, nalidixic acid) have been suggested by sev-eral authors.4,25,26 Based on these breakpoints, one may assumethat the H. pullorum strains are sensitive to doxycycline, gen-tamicin, tobramycin, enrofloxacin, and erythromycin and re-sistant to ampicillin, ceftriaxone, and sulphamethoxazole-trimethoprim. The high MIC values for ampicillin arenoteworthy, since other tested Helicobacter species are mostlysensitive to this antimicrobial agent.5,6,10,13,14 For nalidixicacid, it is difficult to draw clear-cut conclusions about sus-ceptibility versus resistance of the tested H. pullorum strains,based on the breakpoints (16–32 �g/ml) for Campylobacterspp. and Arcobacter spp. given by the authors mentionedabove. Also according to Stanley et al.,23 a H. pullorum strainmay be denoted as being resistant to this antimicrobial whenthe MIC is higher than 32 �g/ml. In the present study, the MICvalues of nalidixic acid for 20 tested H. pullorum strains aresituated at or above 32 �g/ml. This represents the normal sus-ceptibility level of this species to nalidixic acid and does notimply that these strains display acquired resistance to nalidixicacid.

A bimodal frequency distribution of MICs for spectinomycinwas seen. The MIC for all isolates from flock II was higherthan 512 �g/ml, while the MICs of the other tested strainsranged from 8 to 128 �g/ml. The former strains hence may bedesignated as having acquired resistance. Spectinomycin is anaminocyclitol antibiotic that binds to the 30 S ribosomal sub-unit and acts on the protein synthesis during the mRNA–ribo-some interaction by preventing elongation of the polypeptidechain at the translocation step.20 Acquired resistance to spectin-omycin may develop as a result of a single step mutation in thechromosomal gene rpsE, coding for the ribosomal small sub-unit protein S5.3,27 Escherichia coli mutants with alterations atposition 1192 in 16S rRNA, namely G/C changing to G/U basepair, have been described as well.3,21 Plasmid-mediated resis-tance to spectinomycin is uncommon,20 although resistance dueto R-plasmid specified adenylylation and phosphorylation ofthe antibiotic has been described in several bacteria, includingCampylobacter spp.12,19,20

The other tested antimicrobial agents showed a monomodaldistribution of MICs, indicating that MIC values obtained inthis study reflect normal susceptibility levels of this species tothese antibiotics.

In the present study, the MIC of nalidixic acid for the H.canadensis strain was found to be higher than 512 �g/ml; thatof enrofloxacin was 8 �g/ml, which is remarkably higher thanthe MICs recorded for H. pullorum (�0.03–0.06). This con-firms the results of Fox et al.,8 who found four H. canadensisisolates to be resistant to nalidixic acid in a disk diffusion test.Resistance to nalidixic acid is considered to be specific for thespecies H. canadensis.

In conclusion, notwithstanding the lack of standardizedguidelines on MIC testing for the species H. pullorum, this studyallows us to assess the normal in vitro susceptibility of thisspecies for several antimicrobials. Acquired resistance was onlydetected to spectinomycin.

ACKNOWLEDGMENTS

We thank the abattoir Nollens (Kruishoutem, Belgium) forproviding intestinal tracts from poultry and M. Heyndrickx(Centre of Agricultural Research, Melle, Belgium) for provid-ing H. pullorum strains. We appreciate the skilled technical as-sistance provided by J. De Craene and A. Van de Kerckhove.This work was supported by a Ph.D. grant of the Institute forthe Promotion of Innovation by Science and Technology inFlanders (IWT Vlaanderen) to Liesbeth Ceelen.

REFERENCES

1. Ananieva, O., I. Nilsson, T. Vorobjova, R. Uibo, and T. Wad-ström. 2002. Immune responses to bile-tolerant Helicobacterspecies in patients with chronic liver diseases, a randomized pop-ulation group, and healthy blood donors. Clin. Diagn. Lab. Im-munol. 9:1160–1164.

2. Atabay, H.I., J.E.L. Corry, and S.L.W. On. 1998. Identificationof unusual Campylobacter-like isolates from poultry products asHelicobacter pullorum. J. Appl. Microbiol. 84:1017–1024.

3. Bilgin, N., A.A. Richter, M. Ehrenberg, A.E. Dahlberg, andC.G. Kurland. 1990. Ribosomal RNA and protein mutants resis-tant to spectinomycin. EMBO J. 9:735–739.

4. Fera, M.T., M. Maugeri, M. Giannone, C. Gugliandolo, E. LaCamera, G. Blandino, and M. Carbone. 2003. In vitro suscepti-bility of Arcobacter butzleri and Arcobacter cryaerophilus to dif-ferent antimicrobial agents. Int. J. Antimicrob. Agents 21:488–491.

5. Flores, B.M., C.L. Fennell, K.K. Holmes, and W.E. Stamm.1985. In vitro susceptibilities of Campylobacter-like organisms totwenty antimicrobial agents. Antimicrob. Agents Chemother.28:188–191.

6. Fox, J.G. 1997. The expanding genus of Helicobacter: pathogenicand zoonotic potential. Semin. Gastrointest. Dis. 8:124–141.

7. Fox, J.G., F.E. Dewhirst, Z. Shen, Y. Feng, N.S. Taylor, B.J.Paster, R.L. Ericson, C.N. Lau, P. Correa, J.C. Araya, and I.Roa. 1998. Hepatic Helicobacter species identified in bile and gall-bladder tissue from Chileans with chronic cholecystitis. Gastro-enterology 114:755–763.

8. Fox, J.G., C.C. Chien, F.E. Dewhirst, B.J. Paster, Z. Shen, P.L.Melito, D.L. Woodward, and F.G. Rodgers. 2000. Helicobactercanadensis sp. nov. isolated from humans with diarrhea as an ex-ample of an emerging pathogen. J. Clin. Microbiol. 38:2546–2549.

IN VITRO SUSCEPTIBILITY OF H. PULLORUM TO ANTIMICROBIAL AGENTS 125

9. Gibson, J.R., M.A. Ferrus, D. Woodward, J. Xerry, and R.J.Owen. 1999. Genetic diversity in Helicobacter pullorum and poul-try sources identified by an amplified fragment length polymor-phism technique and pulsed-field gel electrophoresis. J. Appl. Mi-crobiol. 87:602–610.

10. Hachem, C.Y., J.E. Clarridge, R. Reddy, R. Flamm, D.G.Evans, S.K. Tanaka, and D.Y. Graham. 1996. Antimicrobial sus-ceptibility testing of Helicobacter pylori. Comparison of E-test,broth microdilution, and disk diffusion for ampicillin, clar-ithromycin, and metronidazole. Diagn. Microbiol. Infect. Dis.24:37–41.

11. Hakanen, A., P. Huovinen, P. Kotilainen, A. Siitonen, and H.Jousimies-Somer. 2002. Quality control strains used in suscepti-bility testing of Campylobacter species. J. Clin. Microbiol.40:2705–2706.

12. Kawabe, H., T. Tanaka, and S. Mitsuhashi. 1987. Streptomycinand spectinomycin resistance mediated by plasmids. Antimicrob.Agents Chemother. 13:1031–1035.

13. Kiehlbauch, J.A., D.J. Brenner, D.N. Cameron, A.G. Steiger-walt, J.M. Makowski, C.N. Baker, C.M. Patton, and I.K.Wachsmuth. 1995. Genotypic and phenotypic characterization ofHelicobacter cinaedi and Helicobacter fennelliae strains isolatedfrom humans and animals. J. Clin. Microbiol. 33:2940–2947.

14. Loo V.G., C.A. Fallone, E. De Souza, J. Lavallee and A.N.Barkun. 1997. In vitro susceptibility of Helicobacter pylori toampicillin, clarithromycin, metronidazole and omeprazole. J. An-timicrob. Chemother. 40:881–883.

15. Melito P.L., D.L. Woodward, K.A. Bernard, L. Price, R.Khakhria, W.M. Johnson, and F.G. Rodgers. 2000. Differenti-ation of clinical Helicobacter pullorum isolates from related Heli-cobacter and Campylobacter species. Helicobacter 5:142–147.

16. National Committee for Clinical Laboratory Standards. 2002.Approved standard—Second edition M31-A2. Selection of an-timicrobial agents for routine testing and reporting. NCCLS,Wayne, PA.

17. National Committee for Clinical Laboratory Standards. 2002.Approved standard—Second edition M31-A2. Disk diffusion sus-ceptibility tests. NCCLS, Wayne, PA.

18. On S.L.W. 1996. Identification methods for Campylobacters, Helicobacters, and related organisms. Clin. Microbiol. Rev. 9:405–422.

19. Pinto-alphandary H., C. Mabilat, and P. Courvalin. 1990. Emer-gence of aminoglycoside resistance genes aadA and aadE in thegenus Campylobacter. Antimicrob. Agents Chemother. 34:1294–1296.

20. Prescott J.F. 2000. Aminoglycosides and aminocyclitols. In: J.F.Prescott, J.D. Baggot, and R.D. Walker (eds.), Antimicrobial ther-apy in veterinary medicine, 3rd ed. Iowa State University Press,Ames, pp. 191–228.

21. Sigmund C.D., M. Ettayebi, and E.A. Morgan. 1984. Antibioticresistance mutations in 16S and 23S ribosomal RNA gene of Escherichia coli. Nucleid Acid Res. 12:4653–4663.

22. Solnick J.V. 2003. Clinical significance of Helicobacter speciesother than Helicobacter pylori. Clin Infect Dis. 36:349–354.

23. Stanley J., D. Linton, A.P. Burnens, F.E. Dewhirst, S.L.W. On,A. Porter, R.J. Owen, and M. Costas. 1994. Helicobacter pullo-rum sp. nov.—genotype and phenotype of a new species isolatedfrom poultry and from human patients with gastroenteritis. Micro-biology 140:3441–3449.

24. Steele T.W., and S.N. McDermott. 1984. The use of membranefilters applied directly to the surface of agar plates for the isola-tion of Campylobacter jejuni from faeces. Pathology 16:263–265.

25. ter Kuile B., H. van der Zee, and C. Pellicaan. 2002. In D.J.Mevius and W. van Pelt (eds.), Monitoring of antimicrobial resis-tance and antibiotic usage in animals in the Netherlands in 2002.VWA, The Hague, The Netherlands, pp. 19–46.

26. Van Looveren M., G. Daube, L. De Zutter, J-M. Dumont, C.Lammens, M. Wijdooghe, P. Vandamme, M. Jouret, M. Cor-nelis, and H. Goossens. 2001. Antimicrobial susceptibilities ofCampylobacter strains isolated from food animals in Belgium. J.Antimicrob. Chemother. 48:235–240.

27. Wilcox S.K., G.S. Cavey, and J.D. Pearson. 2001. Single ribo-somal protein mutations in antibiotic-resistant bacteria analyzed bymass spectrometry. Antimicrob. Agents Chemother. 45:3046–3055.

28. Young V.B., C.C. Chien, K.A. Knox, N.S. Taylor, D.B.Schauer, and J.G. Fox. 2000. Cytolethal distending toxin inavian and human isolates of Helicobacter pullorum. J. Infect. Dis.182:620–623.

Address reprint requests to:Dr. Liesbeth Ceelen

Department of Pathology, Bacteriology and Avian DiseasesFaculty of Veterinary Medicine

Ghent UniversitySalisburylaan 133

9820 Merelbeke, Belgium

E-mail: liesbeth.ceelen@UGent.be

CEELEN ET AL.126