In recent years, detection of fluoroquinolone (FQ) resistance determinants in Escherichia coli (E. coli) isolated from animals indicated that this is an important public health issue and can create a high risk for the treatment of infectious diseases at the recommended available dosage regimens.1 In Gram-negative bacteria, resistance to FQs primarily occurs from gene mutations in the quinolone resistance-determining region (QRDR) of the genes encoding the drug target enzymes (DNA gyrase and topoisomerase IV).2 In addition, the pentapeptide repeat proteins (QnrA, QnrB and QnrS) increase MIC of FQ for E. coli (0.125–16 μg ml−1) by protecting DNA gyrase from inhibition.3, 4

FQs have been approved for use in the treatment of infectious diseases around the world. Danofloxacin, a member of second-generation FQs, is a synthetic FQ with broad-spectrum antibacterial activity. It is used in the treatment of respiratory disease in chickens, cattle and pigs. Orbifloxacin is a member of third-generation FQs and developed for use in companion animal medicine. In canine practice, orbifloxacin is indicated for the treatment of various infections, including urinary, skin and otitis infections.

There are a few reported drug interactions among FQs that have veterinary significance.5, 6, 7 Interactions between main compounds (enrofloxacin and ibafloxacin) and their metabolites (ciprofloxacin and 8-hydroxy-ibafloxacin) against E. coli, Staphylococcus spp. and Pseudomonas aeruginosa (P. aeruginosa) were investigated in these studies. In addition, Pankey and Ashcraft8 showed that there was a synergistic interaction between ciprofloxacin and gatifloxacin against P. aeruginosa. The objective of this work was to identify if a synergistic interaction between danofloxacin and orbifloxacin against FQ-resistant E. coli isolates from animals occurs. The drug combination studies were carried out using the checkerboard and time-kill methods.

Seven E. coli isolates carrying gyrA mutations or qnr genes from the Laboratory of Molecular Pharmacology were selected for the checkerboard and time-kill studies. The gyrA mutant E. coli isolates were obtained from three healthy fowl (E224, E245, E246), the qnr-containing E. coli isolates were from one healthy cow (E101), one cow (E103) and one sheep (E248) with gastroenteritis, and one healthy dog (E300).

Broth microdilution testing was performed to determine the MICs of the compounds according to the guidelines of the Clinical Laboratory Standards Institute.9 E. coli ATCC25922 was used as control for antimicrobial susceptibility testing.

QRDR and plasmid-mediated quinolone resistance (PMQR) genes were amplified using specific primers10, 11 and PCR products of gyrA were sequenced by Macrogen Inc. (Korea). The DNA sequences of gyrA were analyzed using the BLAST program (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Presence of the qnrA and qnrS genes was determined by PCR amplification, as described previously by Robicsek et al.12 and Cengiz et al.1 The primers used are as follows: gyrA, 5′-ACGTACTAGGCAATGACTGG-3′ (forward) and 5′-AGAAGTCGCCGTCGATAGAAC-3′ (reverse); qnrA, 5′-ATTTCTCACGCCAGGATTTG-3′ (forward) and 5′-GATCGGCAAAGGTTAGGTCA-3′ (reverse); and qnrS, 5′-ACGACATTCGTCAACTGCAA-3′ (forward) and 5′-TAAATTGGCACCCTGTAGGC-3′ (reverse).

Fractional inhibitory concentration index/indices (FIC index/indices) of danofloxacin and orbifloxacin were determined using checkerboard method.13 Danofloxacin concentrations ranged from 0.064 to 256 μg ml−1 and orbifloxacin concentrations ranged from 0.128 to 512 μg ml−1. FIC index/indices were calculated as follows:

FICA=MIC drug A in combination/MIC drug A alone

FICB=MIC drug B in combination/MIC drug B alone

FIC index/ΣFIC=FICA+FICB

The FIC index was interpreted as follows: synergy=FIC index⩽0.5; indifference=0.5⩽4; antagonism=FIC index>4.

Time-kill experiments were slightly modified from the method described by Petersen et al.14 A liquid overnight bacterial culture of the gyrA mutant and qnr-containing E. coli isolates was diluted with Mueller–Hinton Broth (Becton, Dickinson and Company, Sparks, MD, USA) and drug stock solutions to achieve an initial inoculum of ∼106 c.f.u. ml−1. Each 10 ml culture was incubated at 37 °C, and samples were withdrawn for the determination of bacterial counts at 0, 6 and 24 h. Colony counts were determined by plating 100 μl of each diluted sample onto Plate Count Agar (Becton, Dickinson and Company) with an automated spiral plater (WASP; Don Whitley Scientific Ltd., Shipley, UK) and then counting using an colony counter (UVITEC Cambridge, Cambridge, UK). Synergy was defined as a ⩾2 log10 decrease in colony count at 6 or 24 h with the combination compared with the initial inoculum. The drug combination was considered to be antagonist if there was a ⩾2 log10 increase in c.f.u. ml−1 and indifference was the interpretation of a <2 log10 change in c.f.u. ml−1.

The MICs of danofloxacin and orbifloxacin for E. coli ATCC25922 were 0.032 and 1 μg ml−1, respectively. Microbiological activity (MIC90) of danofloxacin and orbifloxacin to E. coli isolated from animals was reported as 0.015–0.25 and 0.5 μg ml−1, respectively.15 E. coli isolates presented an alteration in gyrA (E224, E245, E246: Ser-83→Leu; E224: Asp-87→Asn, E245: Asp-87→Glu) and the qnr genes detected were qnrA1 (E101) and qnrS1 (E103, E248, E300). The amino-acid substitutions in gyrA were at the most frequently identified site (codon 83).2 The MICs of the compounds and FIC values of the combination for gyrA mutant and qnr-containing E. coli isolates are shown in Table 1. FIC index of the combination for resistant E. coli isolates ranged from 0.09 to 1. The incidence of synergy and additivity/indifference was 71% and 29%, respectively. Antagonism was not detected for any of E. coli isolates by checkerboard method.

Table 1 Resistance mechanisms and MICs of E. coli isolates, checkerboard and time-kill data with the interpretations
Full size table

By using the time-kill method, the in vitro activity of the combination against gyrA mutant and qnr-containing E. coli isolates are shown in Table 1. At 6-h incubation, the combination resulted ⩾2 log10 reduction in viable counts against all E. coli isolates and it showed synergic activity. At 24-h incubation, this was also achieved for E103, E224 and E245 isolates. However, regrowth was observed for four of seven E. coli isolates after 24 h incubation.

There are increasing numbers of antibiotic-resistant infections, especially by Gram-negative bacteria, which are innately multi-drug resistant.16 In addition, Gram-negative bacteria such as E. coli are increasingly resistant to the few effective agents available for treatment via the acquisition of transmissible elements, and when isolated from animals have multiple and different mechanisms of antibiotic resistance.1, 17 Therefore, to restore the efficacy of licensed veterinary FQs against resistant Gram-negative bacteria has become important.

The two methods used most commonly to assess antimicrobial interactions in vitro are the checkerboard and time-kill assays.7 In this study, these methods were used to assess synergy of danofloxacin+orbifloxacin combination against seven clinical isolates of E. coli. Synergy using the checkerboard method was detected for five of seven E. coli isolates with 0.09–0.5 ΣFIC. Pankey and Ashcraft8 used Etest synergy method and found a similar interaction between ciprofloxacin and gatifloxacin against P. aeruginosa with 19% incidence. Enrofloxacin is unique in that it is partially metabolized to ciprofloxacin and both active drugs circulate in treated animals.5 Lautzenhiser et al.7 showed that for staphylococcal and E. coli isolates, FIC indices of enrofloxacin+ciprofloxacin combination were between 0.5 and 4.0, indicating that the combination acted additively in vitro. The combination of ibafloxacin plus its major active metabolite 8-hydroxy-ibafloxacin had synergistic action in two E. coli isolates and additive effects in E. coli ATCC25922.6 The synergistic activity of the active metabolite contributes additionally to the antimicrobial activity of the parent compound.5, 6, 7 In this study, by the time-kill method, synergy was mainly shown after 6 h of exposure for all isolates accompanied by regrowth after 24 h for four of them. These results showed that for E. coli synergy incidence detected by the time-kill method is higher than checkerboard method. Elipoulos and Moellering13 indicated that in contrast to the checkerboard technique, which typically provides only inhibitory data, the killing-curve technique measures the microbicidal activity of the combination being tested. For this reason, it is presumably more relevant for clinical situations in which bactericidal effect.13

Cengiz et al.18 showed that the genetic mechanisms of FQ resistance were determinative for the bactericidal activity of enrofloxacin alone against E. coli. The results of this study clearly indicated that danofloxacin (second-generation FQs) and orbifloxacin (third-generation FQs) can exert synergistic activity against some strains FQ-resistant E. coli isolates, and this combination could be considered for augmenting of their efficacy in vivo.

This study was financed by the Scientific and Technological Research Council of Turkey-TUBITAK (TOVAG-110O478) and supported by COST Action BM0701 ‘ATENS’.