JAC Advance Access published online on March 10, 2008
Journal of Antimicrobial Chemotherapy, doi:10.1093/jac/dkn094
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Heterogeneity of susceptibility to fluoroquinolones in Bartonella isolates from Australia reveals a natural mutation in gyrA
1 Unité des Rickettsies, CNRS UMR 6020, IFR 48, Faculté de Médecine et de Pharmacie Université de la Méditerranée, 27 Bd Jean Moulin, 13385 Marseille Cedex 05, France 2 Public Health Virology, Forensic and Scientific Services, 39 Kessels Road, Coopers Plains Q 4108, Australia
* Corresponding author. Tel: +33-491-38-55-17; Fax: +33-491-83-03-90; E-mail: jm.rolain{at}medecine.univ-mrs.fr
Received 24 October 2007; returned 17 December 2007; revised 29 January 2008; accepted 16 February 2008
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Objectives: Bartonella sp. are intracellular bacteria associated with an increasing number of clinical manifestations but with few published data on in vitro susceptibility testing of antibiotics. Our objective was to evaluate in vitro antibiotic susceptibilities of 20 new Bartonella isolates from animals in Australia.
Methods: MICs were determined using Etest assay on Columbia agar supplemented with 5% horse blood. The presence of mutations in the quinolone-resistance-determining region (QRDR) of gyrA was searched for after PCR amplification and DNA sequencing using specific oligonucleotide primers.
Results: Bartonella isolates from Australia were susceptible to rifampicin, tetracyclines, β-lactam and macrolide compounds but were resistant to vancomycin. We found heterogeneity of susceptibility for fluoroquinolones with ciprofloxacin being more effective (MICs from 0.06 to 0.5 mg/L) than ofloxacin (MICs from 0.5 to 4 mg/L). This heterogeneity was linked to a natural mutation Ser-83
Ala (Escherichia coli numbering) in the QRDR. Surprisingly, this mutation was also present in the QRDR of Bartonella henselae, Bartonella quintana and Bartonella bacilliformis.
Conclusions: Etest is a sensitive and reliable assay for evaluation of antibiotic susceptibility in the genus Bartonella. The higher sensitivity of this method allowed us to detect heterogeneity of susceptibility among fluoroquinolones that was associated with natural mutation in the QRDR of the DNA gyrase. Because a high level of resistance to fluoroquinolones due to a second mutation may be obtained easily in vitro, we believe that fluoroquinolone compounds should be avoided for the treatment of any Bartonella-related diseases.
Key Words: DNA gyrase , Etest diffusion assay , resistance
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Bartonella species are Gram-negative bacilli that belong to the
2 subgroup of Proteobacteria.1 Over the last decade, the number of identified Bartonella sp. has increased rapidly and the family Bartonellaceae now contains 22 validated species and 3 subspecies. To date, 11 species have been associated with an increasing spectrum of clinical syndromes in humans including cat scratch disease (Bartonella henselae), Bacillary angiomatosis, Peliosis hepatitis (B. henselae, Bartonella quintana), bacteraemia and/or endocartitis (B. henselae, B. quintana, Bartonella vinsonii subsp. arupensis, Bartonella alsatica and Bartonella rochalimae), Carrions disease (Bartonella bacilliformis), trench fever (B. quintana), retinitis (B. henselae, Bartonella grahamii), myocarditis (Bartonella washoensis) and splenomegaly (B. rochalimae).2 Treatment of human bartonellosis depends on clinical disease, and current recommendations for treatment are based on a few case reports and very limited data from a few clinical studies.2 Moreover, due to the fastidious nature of these bacteria and the limited number of available isolates worldwide, few data on in vitro susceptibility have been reported. Although successful treatment of Bartonella infections with fluoroquinolones has been reported, there have also been failures and relapses with these drugs.2
Fluoroquinolone antibiotics exert their antibacterial effects by inhibition of DNA gyrase (topoisomerase II) and topoisomerase IV. Bacterial resistance to quinolones can essentially develop either by point mutations in the quinolone-resistance-determining region (QRDR) of both gyrase (gyrA) and topoisomerase IV (parC), or by efflux mechanisms.
In the present study, we have determined the MICs of 11 antimicrobial agents by Etest for 20 new Bartonella strains from Australia isolated from the blood of different animals including kangaroos and rodents. Moreover, we have determined the sequences of the QRDR of gyrA gene for these isolates and compared them with the available sequences of Bartonella reference strains in order to explain the molecular basis of the heterogeneity of susceptibility to fluoroquinolone compounds.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Bartonella strains
Twenty-four strains of Bartonella species were tested including 20 strains of Bartonella sp. isolated from the blood of wild animals in Australia and from four reference strains (Table 1). All animal trapping and blood collection was carried out in accordance with ‘The Australian Code of Practice for the Care and Use of Animals for Scientific Purposes’.
|
Drugs
Etest strips for erythromycin, clarithromycin, ceftriaxone, tetracycline, azithromycin, ofloxacin, amoxicillin, imipenem, vancomycin, ciprofloxacin and rifampicin were purchased from AB BIODISK (Solna, Sweden).
Antibiotic susceptibility testing
The bacteria were grown on Columbia 5% sheep blood agar plates and incubated at 37°C in a 5% CO2-enriched atmosphere. Antibiotic susceptibility testing was carried out with the same agar medium containing an Etest strip and observed daily for growth as previously reported.3 MICs were read after an incubation of 5–12 days. Susceptibility testing was performed twice for each strain and antimicrobial agent in order to confirm the results.
Staphylococcus aureus ATCC 29213 and Escherichia coli ATCC 25922 were used as controls.
Total genomic DNA was extracted from Bartonella isolates with a QIAamp Tissue kit (Qiagen, Hilden, Germany), as described by the manufacturer. Samples were handled under sterile conditions to avoid the risk of cross-contamination.
DNA amplification and sequencing of gyrA gene
Bartonella strains from Australia were screened for the presence of mutations in the QRDR of gyrA gene by PCR amplification and DNA sequencing using specific oligonucleotide primers BagyrAR and BagyrAF, published by Biswas et al.4 The obtained nucleotide sequences of candidate genes were compared using the CLUSTAL W program (http://www.ebi.ac.uk/clustalw/) to look at possible mutations known to be associated with antibiotic resistance. Nucleotide sequences were translated into protein sequences using the DNA TO PROTEIN TRANSLATION program (http://bio.lundberg.gu.se/edu/translat.html). The protein sequences obtained were compared with those of B. henselae (GenBank accession no. BH10120), B. quintana (GenBank accession no. BQ07850), B. bacilliformis (GenBank accession no. BARBAKC583_0737) and E. coli (GenBank accession no. JW2225) QRDR. Chi-square test was performed using Epi info 6 software (Centers for Disease Control and Prevention, Atlanta, GA, USA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
All Bartonella strains required 5 days of incubation, except B. bacilliformis which requires 12 days, in order to record MICs. Using the Etest assay, MICs for the β-lactams tested ranged from 0.004 to 0.03 mg/L for amoxicillin and from 0.004 to 0.016 mg/L for ceftriaxone. The macrolides were highly effective, with MIC ranging from 0.008 to 0.12 mg/L for azithromycin, from 0.004 to 0.016 mg/L for clarithromycin and from 0.008 to 0.06 mg/L for erythromycin. All except one (strain Nh4 MIC = 3 mg/L) had MICs of imipenem ranging from 0.03 to 0.5 mg/L. MICs of tetracycline ranged from 0.008 to 0.06 mg/L. Heterogeneity of susceptibility against fluoroquinolone compounds was found, with a median MIC of ciprofloxacin of 0.25 mg/L (0.06–0.5 mg/L) versus a median MIC of ofloxacin of 2 mg/L (0.5–4 mg/L) (P = 0.005). Rifampicin MICs ranged from 0.002 to 0.06 mg/L. Finally, vancomycin MICs ranged from 1 to 16 mg/L. Table 1 illustrates the MIC results obtained by Etest assay.
Comparison of the QRDR of the 20 Bartonella strains from Australia with those of B. henselae, B. quintana and B. bacilliformis revealed a natural mutation at position 83 of the QRDR (Ser-83Ala, E. coli numbering) for all Bartonella strains (Figure 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
In this study, 20 Bartonella sp. isolates from animals from Australia were tested for their antibiotic susceptibilities to 11 antimicrobial agents using the Etest assay. Evaluation of susceptibility to antibiotics in Bartonella genus has been previously assessed either in the presence of eukaryotic cells or in axenic media especially by the agar dilution technique.2 Recently, we and others have reported the reliability, sensitivity and usefulness of the Etest assay for determination of MICs for the Bartonella genus.4,5 This assay was used in the present study with consistent and similar results when compared with other Bartonella species.2 The Bartonella isolates from Australia were susceptible to aminoglycosides, macrolides, doxycycline and rifampicin, and were resistant to vancomycin.
Imipenem was effective against all the strains except one (Nh4) that had an MIC of 3 mg/L. In this strain, we suspected the presence of a metallo-β-lactamase (MBL) gene either chromosomally mediated or encoded by transferable genes. MBL genes has been found in many Gram-negative bacteria including B. bacilliformis (http://www.genome.jp/kegg/).
Interestingly, we found heterogeneity of susceptibility to fluoroquinolones with ciprofloxacin to be more effective than with ofloxacin. Ofloxacin MICs were 10-fold higher compared with ciprofloxacin. Looking at previous results obtained for other Bartonella strains, we found that ciprofloxacin was also more effective when compared with pefloxacin.2,6,7 We found that this heterogeneity of susceptibility to fluoroquinolones was linked to a natural mutation at position 83 of the QRDR of the DNA gyrase. In vitro studies with E. coli suggest that the first step in selection for decreased susceptibility to fluoroquinolones was an alteration of Ser-83.8 Studies suggest that not all mutations in gyrA confer the same level of resistance to all quinolones, and mutations at codons Ser-83 and Asp-87 appear most frequently and confer higher levels of quinolone resistance than mutations at other codons within the QRDR. Moreover, strains with a single mutation at Ser-83 were significantly more resistant to fluoroquinolones than those with a single mutation at Asp-87 because mutations at this active site may alter the binding of quinolones to the site and lead to reduced susceptibility or resistance to quinolones. According to Heisig and Tschorny,8 a mutation at codon Ser-83Ala in E. coli leads to MICs of ciprofloxacin that are lower than MICs of ofloxacin, whereas E. coli strains with double mutations in Ser-83Ala and Asp-87Gly present a high level of resistance for all compounds. This was also observed for a ciprofloxacin-resistant B. bacilliformis strain obtained in vitro with a double mutation.4 Surprisingly, the same mutation was also found in other Bartonella strains including B. henselae strain Houston, B. henselae strain Marseille, B. bacilliformis and B. quintana strain Oklahoma.
Minnick et al.9 and our team have recently reported gyrA mutations in ciprofloxacin-resistant B. bacilliformis strains obtained in vitro at position Asp-87Asn (E. coli numbering) of the QRDR of the DNA gyrase.4,9 In these two reports, the Ser-83Ala mutation was also present in the QRDR of the DNA gyrase but it was not mentioned (J.-M. R., unpublished data). We believe that the heterogeneity of susceptibility to fluoroquinolones in the Bartonella genus is due to this natural mutation at position 83 but additional factors such as decreased intracellular drug accumulation by drug efflux or plasmid-quinolone resistance gene (qnr) may also occur. This natural mutation at position 83 may also explain heterogeneity of susceptibility to fluoroquinolones reported for other intracellular bacteria including Tropheryma whipplei and bacteria of the genus Brucella.10
Our findings have clinical significance and could explain relapses observed using ciprofloxacin for treatment of Bartonella infections especially for the treatment of Oroya fever or ocular infections.2 Since a high level of resistance by a second mutation in the QRDR could be obtained easily in vitro,4,9 we believe that fluoroquinolones should not be used for the treatment of any Bartonella-related disease.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
No specific funding was received.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
We thank Lee Smythe, Greg Smith, Ina Smith, Meegan Symonds and Michael Dohnt who trapped animals.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
1 . Birtles RJ, Raoult D. Comparison of partial citrate synthase gene (gltA) sequences for phylogenetic analysis of Bartonella species. Int J Syst Bact (1996) 46:891–7.
2
.
Rolain JM, Brouqui P, Koehler JE, et al. Recommendations for treatment of human infections caused by Bartonella species. Antimicrob Agents Chemother (2004) 48:1921–33.
3 . Meghari S, Rolain JM, Grau GE, et al. Anti-angiogenic effect of erythromycin in Bartonella quintana: in vitro model of infection. J Infect Dis (2006) 193:380–6.[CrossRef][Web of Science][Medline]
4
.
Biswas S, Raoult D, Rolain JM. Molecular mechanisms of resistance to antibiotics in Bartonella bacilliformis. J Antimicrob Chemother (2007) 59:1065–70.
5
.
Pendle S, Ginn A, Iredell J. Antimicrobial susceptibility of Bartonella henselae using Etest methodology. J Antimicrob Chemother (2006) 57:761–3.
6 . Ives TJ, Marston EL, Regnery RL, et al. In vitro susceptibilities of Bartonella and Rickettsia spp. to fluoroquinolone antibiotics as determined by immunofluorescent antibody analysis of infected Vero cell monolayers. Int J Antimicrob Agents (2001) 18:217–22.[CrossRef][Web of Science][Medline]
7 . Kordick DL, Papich MG, Breitschwerdt B. Efficacy of enrofloxacin or doxycycline for treatment of Bartonella henselae or Bartonella clarridgeiae infection in cats. Antimicrob Agents Chemother (1997) 41:2448–55.[Abstract]
8
.
Heisig P, Tschorny R. Characterization of fluoroquinolone-resistant mutants of Escherichia coli selected in vitro. Antimicrob Agents Chemother (1994) 38:1284–91.
9
.
Minnick MF, Wilson ZR, Smitherman LS, et al. gyrA mutations in ciprofloxacin-resistant Bartonella bacilliformis strains obtained in vitro. Antimicrob Agents Chemother (2003) 47:383–6.
10
.
Masselot F, Boulos A, Maurin M, et al. Molecular evaluation of antibiotic susceptibility, the Tropheryma whipplei paradigm. Antimicrob Agents Chemother (2003) 47:1658–64.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Angelakis, D. Raoult, and J.-M. Rolain Molecular characterization of resistance to fluoroquinolones in Bartonella henselae and Bartonella quintana J. Antimicrob. Chemother., June 1, 2009; 63(6): 1288 - 1289. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||