This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saxton, K.
Right arrow Articles by Wilcox, M. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saxton, K.
Right arrow Articles by Wilcox, M. H.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, February 2009, p. 412-420, Vol. 53, No. 2
0066-4804/09/$08.00+0     doi:10.1128/AAC.00306-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Effects of Exposure of Clostridium difficile PCR Ribotypes 027 and 001 to Fluoroquinolones in a Human Gut Model{triangledown}

Katie Saxton,1 Simon D. Baines,1 Jane Freeman,2 Rachael O'Connor,1 and Mark H. Wilcox1,2*

Institute of Molecular and Cellular Biology, Department of Microbiology, University of Leeds, Leeds LS2 9JT, United Kingdom,1 The General Infirmary, Old Medical School, Leeds LS1 3EX, United Kingdom2

Received 5 March 2008/ Returned for modification 19 May 2008/ Accepted 11 August 2008


arrow
ABSTRACT
 
The incidence of Clostridium difficile infection is increasing, with reports implicating fluoroquinolone use. A three-stage chemostat gut model was used to study the effects of three fluoroquinolones (ciprofloxacin, levofloxacin, and moxifloxacin) on the gut microbiota and two epidemic C. difficile strains, strains of PCR ribotypes 027 and 001, in separate experiments. C. difficile total viable counts, spore counts, and cytotoxin titers were determined. The emergence of C. difficile isolates with reduced antibiotic susceptibility was monitored with fluoroquinolone-containing medium, and molecular analysis of the quinolone resistance-determining region was performed. C. difficile spores were quiescent in the absence of fluoroquinolones. Instillation of each fluoroquinolone led to C. difficile spore germination and high-level cytotoxin production. High-level toxin production occurred after detectable spore germination in all experiments except those with C. difficile PCR ribotype 027 and moxifloxacin, in which marked cytotoxin production preceded detectable germination, which coincided with isolate recovery on fluoroquinolone-containing medium. Three C. difficile PCR ribotype 027 isolates and one C. difficile PCR ribotype 001 isolate from fluoroquinolone-containing medium exhibited elevated MICs (80 to ≥180 mg/liter) and possessed mutations in gyrA or gyrB. These in vitro results suggest that all fluoroquinolones have the propensity to induce C. difficile infection, regardless of their antianaerobe activities. Resistant mutants were seen only following moxifloxacin exposure.


arrow
INTRODUCTION
 
Clostridium difficile is a major nosocomial pathogen and a significant cause of diarrhea in hospitalized elderly individuals. C. difficile infection is precipitated almost exclusively by antimicrobial therapy and is thought to arise following impairment of host colonization resistance within the colon. This is hypothesized to allow C. difficile to proliferate and produce toxins (toxins A and B) that lead to mucosal inflammation and diarrhea. Other factors likely to be important in the development and the severity of C. difficile infection include the virulence of the infecting C. difficile strain and the host immune response to the microorganism and its toxins (25). Antimicrobials with a known propensity to induce C. difficile infection include broad-spectrum agents such as clindamycin, expanded-spectrum cephalosporins, and aminopenicillins (20). However, antibiotics with equivalent antimicrobial spectra do not necessarily have the same propensity to induce C. difficile infection. For example, piperacillin-tazobactam has marked deleterious effects on the predominant members of the colonic microbiota but is less likely to induce C. difficile infection than cefotaxime in vivo and in vitro (4, 21, 41, 44). Thus, a simple relationship between antibiotic-mediated depletion of the colonic microbiota and the induction of C. difficile infection is not evident.

Recent reports have implicated a previously uncommon and seemingly more virulent strain of C. difficile, characterized as PCR ribotype 027 (strain NAP1), in large outbreaks of C. difficile infection (8, 27, 31, 39, 48, 51). It has been suggested that some of these outbreaks are linked to the administration of fluoroquinolones (24, 39). Treatment with fluoroquinolones was previously considered to result in a low risk for the development of C. difficile infection, with evidence of an association with C. difficile infection risk largely being confined to occasional case reports (20). We sought to determine the response of two epidemic C. difficile strains, strains of PCR ribotypes 027 and 001, to exposure to three fluoroquinolones (ciprofloxacin, levofloxacin, and moxifloxacin) using a triple-stage chemostat human gut model that has been validated to simulate C. difficile infection. Historically, C. difficile PCR ribotype 027 was susceptible to fluoroquinolones with enhanced antianaerobe activities (34); however, isolates associated with recent outbreaks have been characterized as being fluoroquinolone resistant (34), as is C. difficile PCR ribotype 001 (53).


arrow
MATERIALS AND METHODS
 
C. difficile strains. C. difficile PCR ribotype 027 was isolated during an outbreak of C. difficile infection in 2005 at the Maine Medical Center in Portland and was supplied by Robert Owens.

C. difficile PCR ribotype 001 was isolated at the Leeds General Infirmary (Leeds, United Kingdom) from a patient with a symptomatic C. difficile infection. Both strains were DNA fingerprinted by a standard method (45) to confirm the ribotypes.

Triple-stage chemostat model. We have previously described the use of a three-stage chemostat gut model to study the interactions between antimicrobial agents, the gut microbiota, and C. difficile (4, 21, 22). The gut model was validated against the physicochemical and microbiological measurements from the colonic contents of individuals who had died suddenly (33). Each gut model consists of three (proximal, medial, and distal) glass fermentation vessels top fed by growth medium at a controlled rate. The system has a total retention time of 66.7 h (for vessel 1, 16.7 h; for vessels 2 and 3, 25 h each), and the dilution rate is 0.015 h–1. The gut model was primed with a fecal emulsion prepared from pooled C. difficile-negative feces (~10% [wt/vol] in prereduced phosphate-buffered saline) (21) from five healthy elderly volunteers (age, >65 years) with no history of antimicrobial therapy in the preceding 3 months. Growth medium was prepared in 5-liter volumes, sterilized by autoclaving, and sparged with oxygen-free nitrogen. The medium consisted of peptone water (2.0 g/liter), yeast extract (2.0 g/liter), NaCl (0.1 g/liter), K2HPO4 (0.04 g/liter), KH2PO4 (0.04 g/liter), MgSO4·7H2O (0.01 g/liter), CaCl2·2H2O (0.01 g/liter), NaHCO3 (2.0 g/liter), hemin (0.005 g/liter), cysteine HCl (0.5 g/liter), bile salts (0.5 g/liter), arabinogalactan (1 g/liter), pectin (2 g/liter), and starch (3 g/liter). Liquid additions were made as follows: vitamin K at 10 µl/liter and Tween 80 at 0.2%. After the medium was autoclaved, glucose (0.4 g/liter) and starch (3 g/liter) solutions were added to the medium through a sterile filtration device. Resazurin anaerobic indicator was also added at 0.005 g/liter. Each vessel operated at a controlled pH to reflect the increasing alkalinity of the colon from the proximal to the distal end. Vessel 1 (280 ml) operated at low pH (5.5) and had a high substrate availability, whereas vessels 2 and 3 (300 ml each) operated at more neutral pHs (6.2 and 6.8, respectively) and had lower substrate availabilities. The contents were maintained at 37°C and were sparged with O2-free nitrogen to maintain an anaerobic environment.

Enumeration of gut bacteria and C. difficile. The gut bacterial populations and C. difficile isolates were enumerated in triplicate, as described previously (22). Briefly, 2-ml samples were removed from each gut model vessel and 0.5 ml underwent serial 10-fold dilution in prereduced peptone water to 10–7 in an anaerobic cabinet. Twenty microliters of an appropriate dilution was then used to inoculate a variety of selective and nonselective agars to enumerate the populations of lactose fermenters, total facultative anaerobes, total anaerobes, total clostridia, bacteroides, bifidobacteria, enterococci, and lactobacilli in vessels 2 and 3 (22). Total C. difficile populations were enumerated by serial dilution and were inoculated onto Brazier's CCEY (cycloserine cefoxitin egg yolk) agar base (Bioconnections, Leeds, United Kingdom) supplemented with 5 mg/liter lysozyme (Sigma, United Kingdom) and 2% lysed horse blood (E&O Labs, Bonnybridge, Scotland). C. difficile spores were enumerated in the same way, after prior treatment of 0.5 ml of sample with an equal volume of 99.6% ethanol and incubation at room temperature for 1 h. Colonies were counted at the dilution at which 30 to 300 well-separated colonies were visible, and viable counts were expressed as log10 CFU/ml.

C. difficile cytotoxin quantification. Cytotoxin titers (relative units [RU]) were determined by a Vero cell cytotoxicity assay, as described previously (22). Briefly, 500 µl of each 2-ml sample was centrifuged at 16,000 x g and the supernatant was removed. The supernatant was then serially diluted to 10–7 in phosphate-buffered saline (pH 7.4). Twenty microliters of each dilution was added to Vero cell culture monolayers prepared in 96-well microtiter trays. The cell culture toxin assay trays were incubated at 37°C in air with 5% CO2 and examined at 24 h and 48 h under an inverted microscope. A positive reaction was indicated by cell rounding. The specific action of C. difficile cytotoxin was confirmed by parallel neutralization with Clostridium sordellii antitoxin (ProLabs Diagnostics). Cytotoxin titers were expressed as log10 RU.

Experimental design. Each vessel of the gut model was inoculated with fecal emulsion, and the medium pump was started (day 0). No further interventions were made for 13 days (period A). During this period, the gut microbiota were enumerated every 2 days. C. difficile spores (~107 CFU) were prepared as described previously (22) and were added to vessel 1 on day 14 (period B). Bacterial populations were enumerated daily, and no further interventions were performed for 7 days. On day 21 (period C), a second inoculum of C. difficile (~107 CFU) spores was added to vessel 1, in addition to the commencement of antimicrobial instillation, with the dosing regimens determined from published data from studies with patients or volunteers (Table 1) to achieve a minimum of two-thirds of the published fecal levels (9, 18, 19). Ciprofloxacin was instilled every 12 h for 7 days, and levofloxacin and moxifloxacin were instilled once daily for 7 and 10 days, respectively. The duration of each experiment was at least 48 days, with each combination of C. difficile strain and antibiotic evaluated in a separate experiment.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Schematic representation of gut model experiments and dosing regimens

Detection of C. difficile with a reduced susceptibility to fluoroquinolones. The emergence of C. difficile isolates with reduced fluoroquinolone susceptibility was evaluated with CCEY agar (Bioconnections) supplemented with 5 mg/liter lysozyme (Sigma) and a fluoroquinolone (FQCM). Two concentrations of fluoroquinolone were used: one was approximately the fluoroquinolone MIC for each strain, and the other was higher than the MIC for each strain. For C. difficile PCR ribotype 027, the selective agars contained 32 mg/liter and 64 mg/liter moxifloxacin, 400 mg/liter and 500 mg/liter ciprofloxacin, and 300 mg/liter and 400 mg/liter levofloxacin. For C. difficile PCR ribotype 001, the respective agar antibiotic concentrations were 64 mg/liter moxifloxacin, 100 mg/liter and 200 mg/liter ciprofloxacin, and 300 mg/liter and 400 mg/liter levofloxacin. These antibiotic concentrations were determined from the MIC of the fluoroquinolone for each C. difficile strain and preliminary experiments with unexposed C. difficile isolates (data not shown). Any C. difficile isolate recovered from FQCM was subcultured onto fresh antibiotic-containing agar to confirm its decreased susceptibility. The isolates were then stored at –20°C in 30% glycerol containing 0.25x MIC of the fluoroquinolone.

Antimicrobial bioassay. Samples from the gut model were centrifuged (16,000 x g), and the supernatants were removed and frozen at –20°C. Moxifloxacin concentrations were determined by an in-house large-plate bioassay on antibiotic medium 1 (pH 7.8 to 8.0; Oxoid, Basingstoke, United Kingdom) with the indicator organism Korcuria rhizophila ATCC 9341 (32). Ciprofloxacin and levofloxacin bioassays were performed on Iso-Sensitest agar (Oxoid) with the indicator organism Escherichia coli ATCC 4004 (52). For all bioassays, culture supernatants and fluoroquinolone calibrators (moxifloxacin, 1.125 to 72 mg/liter; ciprofloxacin and levofloxacin, 8 to 512 mg/liter) were filter sterilized and randomly assigned to 9-mm-diameter wells in a bioassay plate. The bioassay plates were incubated aerobically at 37°C overnight. All assays were performed in duplicate. Following incubation, the zone diameters (in mm) of the calibrators were measured, and a calibration line was produced by plotting the square of the zone diameter against the log2 concentration of the fluoroquinolone. Log2 concentrations of the fluoroquinolone in samples from the gut model were read from each respective calibration line. Log2 concentrations were converted to actual concentrations by using a conversion of 2x (where x is the log2 concentration), and the mean antimicrobial concentrations were calculated.

MIC determination. Moxifloxacin MICs were determined by spiral gradient endpoint determination (38). Briefly, each plate for inoculation contained Brazier's CCEY agar base (Bioconnections) supplemented with 5 mg/liter lysozyme (Sigma) and 2% lysed horse blood (E&O Labs) to give a total volume of 25 ml agar in each petri dish. Antimicrobial gradients were created with a Wasp spiral plater (Don Whitley Scientific, Shipley, United Kingdom). A stock solution of the antimicrobial being tested was prepared in sterile water with a concentration of 10,000 mg/liter, to give an MIC range of 6 to 180 mg/liter on the plate. After antibiotic deposition, the plates were left at room temperature for 2 h to allow the antibiotic to be absorbed by the agar. C. difficile isolates were cultured overnight in Schaedler's anaerobic broth (Oxoid), and each culture was streaked radially from the outer edge of the plate to a distance of 13 mm from the center by using a sterile cotton swab. Each plate was inoculated with three test isolates and an unexposed C. difficile PCR ribotype 001 strain, which acted as a control. The plates were inoculated in triplicate and were incubated for 48 h at 37°C under anaerobic conditions. After incubation, the distance from the commencement of antibiotic deposition (13 mm from the center) to the endpoint of growth was determined with calipers. The depth of the agar was confirmed with a cork borer and calipers following incubation. The MIC for each isolate was then calculated by using a previously described formula (38). Ciprofloxacin and levofloxacin MICs were determined by using agar incorporation on Wilkins-Chalgren agar (Oxoid) (53). This method was chosen instead of spiral gradient endpoint determination, as it required a much less concentrated stock solution of antibiotic (the relative insolubility of ciprofloxacin and levofloxacin meant that a stock solution of 10,000 mg/liter was unachievable).

Molecular analysis of C. difficile strains. Genomic DNA was extracted from overnight cultures in Schaedlers' anaerobic broth with a DNA purification kit (Promega, United Kingdom). PCR amplification and sequencing of the quinolone resistance-determining region of gyrA and gyrB were performed with previously designed primers (2, 12). The same regions of the parent strains were also sequenced for comparison. Amplicons were purified with a QIAquick PCR purification kit (Qiagen GmbH, Hilden, Germany) and sequenced by the Integrated Genomic Analysis Facility at the University of Leeds. For comparison, nucleotide sequence alignments were produced with the Clustal W program (10).

Stability of mutations characterized in DNA gyrase. To investigate the genetic stability of the moxifloxacin resistance, isolates found to possess a mutation in the quinolone resistance-determining region were repeatedly subcultured onto fluoroquinolone-free CCEY agar (Bioconnections) supplemented with 5 mg/liter lysozyme (Sigma) every 24 h for 14 days and tested for resistance by the spiral gradient endpoint determination method, as described above. After MIC determination, the gyrA and gyrB regions were sequenced to confirm that the previously characterized mutations remained.


arrow
RESULTS
 
Gut bacterial populations. (i) Ciprofloxacin administration. In the experiments with ciprofloxacin, the bacterial populations were largely stable throughout periods A and B. As the effects of ciprofloxacin on the gut microbiota were similar in vessels 2 and 3 in both experiments, the results only for vessel 3 of the experiment with C. difficile PCR ribotype 001 are shown (Fig. 1A). After ciprofloxacin instillation (period C), the counts of Bacteroides spp., bifidobacteria, and lactose fermenters declined by between 3 and 5 log10 units, to below the limit of detection (~2 log10 units). The counts of lactobacilli, total clostridia, and enterococci declined by between 1 and 4 log10 units (Fig. 1A), with the total facultative and obligate anaerobe populations being relatively unaffected. The bacterial populations generally recovered to or exceeded steady-state (period A) levels within 7 days of administration of the final ciprofloxacin dose, with the exception of bifidobacteria in the experiment with C. difficile PCR ribotype 001, which remained below the limits of detection until 17 days after dosing ended, before rapidly recovering to predosing levels (Fig. 1A).


Figure 1
View larger version (33K):
[in this window]
[in a new window]

  		FIG. 1. Mean (± standard error) total viable counts (log10 CFU/ml) of selected gut bacterial populations in vessel 3 of the gut model in the experiment evaluating C. difficile PCR ribotype 001 and ciprofloxacin (A), moxifloxacin (B), and levofloxacin (C). Lines indicate the last day of the preceding experimental time period.

(ii) Moxifloxacin administration. In the experiments with moxifloxacin, the bacterial populations were largely stable throughout periods A and B. As the effects of moxifloxacin on the gut microbiota were similar in vessels 2 and 3 in both experiments, only the results for vessel 3 of the C. difficile PCR ribotype 001 experiment are shown (Fig. 1B). Moxifloxacin instillation had a marked effect on many components of the gut microbiota enumerated, eliciting declines in Bacteroides spp. (6 log10 units), lactose fermenters (6 log10 units), and enterococci (4 log10 units) to below the limits of detection, with the enterococci showing only limited recovery by the end of the experiments. The Bacteroides numbers increased during period D but failed to recover to steady-state levels. The counts of bifidobacteria, lactobacilli, and total clostridia and the total counts of the facultative and obligate anaerobe populations remained relatively stable after moxifloxacin administration (Fig. 1B).

(iii) Levofloxacin administration. In the experiments with levofloxacin, the bacterial populations were largely stable throughout periods A and B. As the effects of levofloxacin on the gut microbiota were similar in vessels 2 and 3 in both experiments, only the results for vessel 3 of the experiment with C. difficile PCR ribotype 001 are shown (Fig. 1C). The instillation of levofloxacin had a lesser effect on the bacterial populations compared with the effects of moxifloxacin and ciprofloxacin, with a 1-log10-unit decline in Bacteroides spp. and bifidobacteria being observed. Lactose fermenters were reduced to the limits of detection during the dosing period and recovered to the predosing levels in period D. Lactobacilli, total clostridia, enterococci, and total facultative and obligate anaerobe populations were relatively unaffected by levofloxacin administration. All populations recovered rapidly after the cessation of levofloxacin instillation (Fig. 1C).

Fluoroquinolone concentrations in the gut models. The levels of fluoroquinolones achieved in the gut model are presented in Table 2. Although there were some variations in the peak fluoroquinolone concentrations between the gut model experiments, the levels achieved were still within the ranges previously reported from in vivo studies (Table 2) (9, 18, 19). The ciprofloxacin concentrations were below the limits of detection (2 mg/liter) within 10 days of the cessation of dosing. The moxifloxacin concentrations were below the limits of detection (4.5 mg/liter) within 7 days of the cessation of antimicrobial instillation. The levofloxacin concentrations were below the limits of detection (2 mg/liter) within 6 days of the cessation of instillation.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Published fecal fluoroquinolone levels and concentrations of fluoroquinolones achieved in vessels 1 to 3 of the gut model in experiments with C. difficile PCR ribotypes 001 and 027

C. difficile populations and cytotoxin production. (i) Ciprofloxacin administration. C. difficile remained as spores in the absence of ciprofloxacin. After ciprofloxacin administration, the germination of C. difficile PCR ribotype 027 occurred 5 days after the end of the antimicrobial instillation period. This was seen as a marked divergence of the C. difficile total count and spore count and was followed by high-level cytotoxin production 3 days later (Fig. 2B). The germination of C. difficile PCR ribotype 001 was observed 10 days after the end of the antimicrobial instillation period, and high-level cytotoxin production was detected 3 days later (Fig. 2A). The total count of C. difficile PCR ribotype 001 remained markedly above the spore count for the duration of the experiment, while in the experiment with C. difficile PCR ribotype 027, the total count and cytotoxin titers began to decline, with the cytotoxin titer being 2 RU when the experiment was terminated. There was no discernible growth of C. difficile on the ciprofloxacin-containing medium in either experiment.


Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 2. Mean (± standard error) total viable counts, spore counts (log10 CFU/ml), cytotoxin titers (RU), and ciprofloxacin (CIP) levels (mg/liter) in vessel 3 in the experiments with ciprofloxacin and C. difficile PCR ribotype 001 (A) and C. difficile PCR ribotype 027 (B).

(ii) Moxifloxacin administration. In the absence of moxifloxacin, there was no evidence of germination of C. difficile spores, although low-level cytotoxin production was measured in the experiment with C. difficile PCR ribotype 001 (Fig. 3A). Moxifloxacin administration led to the germination of both C. difficile PCR ribotype 027 (after 7 days of moxifloxacin dosing) and C. difficile PCR ribotype 001 (after 4 days of moxifloxacin dosing) (Fig. 3A and B). For C. difficile PCR ribotype 027, significant cytotoxin production (>2 RU) was observed 2 days prior to the observed increase in the total C. difficile counts (Fig. 3B) and corresponded to the isolation of C. difficile from vessels 2 and 3 on selective agar containing 32 mg/liter moxifloxacin. The numbers of C. difficile PCR ribotype 027 colonies isolated from the moxifloxacin-containing medium increased daily until it was a substantial component (53% on day 33) of the total count of C. difficile PCR ribotype 027 (data not shown). Sporadic colonies were also isolated on the 64-mg/liter moxifloxacin-containing agar on two occasions after antimicrobial instillation commenced. In the experiment with C. difficile PCR ribotype 001, marked cytotoxin production occurred after the detectable germination and proliferation of C. difficile (Fig. 3A). C. difficile PCR ribotype 001 was sporadically isolated on moxifloxacin-containing agar (four single colonies), but only from vessels 1 and 2. The total count of C. difficile PCR ribotype 001 remained 1 to 2 log10 units above the spore count for the duration of the experiment, while the total count and the spore count of C. difficile PCR ribotype 027 converged during period D. Evidence of a second increase in C. difficile PCR ribotype 027 numbers was observed toward the end of period D, and the toxin titers were 3 RU for both strains of C. difficile when the experiments were terminated.


Figure 3
View larger version (36K):
[in this window]
[in a new window]

  		FIG. 3. Mean (± standard error) total viable counts, spore counts (log10 CFU/ml), cytotoxin titers (RU) and moxifloxacin (MXF) levels (mg/liter) in vessel 3 in the experiments with moxifloxacin and C. difficile PCR ribotype 001 (A) and C. difficile PCR ribotype 027 (B).

(iii) Levofloxacin administration. In the absence of levofloxacin, there was no evidence of spore germination, although low-level cytotoxin production was measured in the experiment with C. difficile PCR ribotype 001 (Fig. 4A). After the administration of levofloxacin, C. difficile spore germination was observed 9 days after antimicrobial instillation commenced for C. difficile PCR ribotype 027 and 4 days after antimicrobial instillation commenced for C. difficile PCR ribotype 001 (Fig. 4A and B). Marked cytotoxin production by C. difficile PCR ribotype 027 (>2 RU) was observed from 2 days after germination, while substantial toxin production by C. difficile PCR ribotype 001 occurred 3 days after germination. By the time that the experiment was terminated, the total count and the spore count of C. difficile PCR ribotype 027 had converged, while the total count of C. difficile PCR ribotype 001 remained ~1 log10 unit above the spore count. The level of cytotoxin production by both strains was ≤3 RU when the experiments were terminated. There was no discernible growth on the levofloxacin-containing medium in either experiment.


Figure 4
View larger version (32K):
[in this window]
[in a new window]

  		FIG. 4. Mean (± standard error) total viable counts, spore counts (log10 CFU/ml), cytotoxin titers (RU) and levofloxacin (LEV) levels (mg/liter) in vessel 3 in the experiments with levofloxacin and C. difficile PCR ribotype 001 (A) and C. difficile PCR ribotype 027 (B).

MICs and sequence analysis of gyrA and gyrB. The MICs of ciprofloxacin, moxifloxacin, and levofloxacin for C. difficile PCR ribotype 027 were 300 mg/liter, 32 mg/liter, and 256 mg/liter, respectively. The corresponding values for C. difficile PCR ribotype 001 were 64 mg/liter, 45 mg/liter, and 256 mg/liter, respectively. C. difficile isolates recovered from FQCM underwent MIC determination to determine their moxifloxacin susceptibilities compared with those of the parent strains. Of the C. difficile PCR ribotype 027 colonies tested (n = 62), 3 were found to have increased moxifloxacin MICs (80 to ≥180 mg/liter) compared with the MIC for the parent strain (32 mg/liter). Of the C. difficile PCR ribotype 001 isolates tested (n = 4), 1 was found to have an increased moxifloxacin MIC (≥180 mg/liter). The gyrA and gyrB amplicons from these four isolates were sequenced. Sequence analysis of gyrA revealed a point mutation in all four isolates compared with the sequence of C. difficile reference strain 630 that led to the amino acid change Thr-82 -> Ile (2, 12, 15). This substitution was also present in the C. difficile PCR ribotype 027 and the C. difficile PCR ribotype 001 parent strains. The single C. difficile PCR ribotype 001 isolate was also found to have a second point mutation in gyrA that resulted in the amino acid change Ala-83 -> Val compared to the sequence of the parent strain. Sequence analysis of the quinolone resistance-determining region of gyrB revealed a point mutation in the three C. difficile PCR ribotype 027 isolates with a raised moxifloxacin MIC compared with the sequence of the parent strain that led to the amino acid change Glu-466 -> Lys. The C. difficile PCR ribotype 001 isolate with a raised moxifloxacin MIC had no gyrB nucleotide substitutions compared with the sequence of the parent strain. The remaining 59 C. difficile PCR ribotype 027 colonies investigated did not exhibit an elevated moxifloxacin MIC, and no nucleotide substitutions in gyrB of four colonies picked at random were observed.

All C. difficile isolates with mutations in the quinolone resistance-determining region underwent repeated passage on antibiotic-free medium to assess the stability of the moxifloxacin resistance. The MIC for one C. difficile PCR ribotype 027 isolate was found to revert to the MIC for the parent strain, and sequencing revealed a genetic reversion to the wild type. All three remaining isolates consistently exhibited MICs of ≥180 mg/liter and retained the point mutations in the quinolone resistance-determining region.


arrow
DISCUSSION
 
While the antibiotic-mediated induction of C. difficile infection is well accepted, the reasons for this association remain unclear. Previous studies of the predisposition of individuals treated with antimicrobials to C. difficile infection have employed either batch culture (test tube) or hamster models, both of which have limitations when they are used to study the behavior of C. difficile within the human colon. In contrast, the gut model is a reproducible and gut-reflective experimental system that can provide information about the risk of C. difficile infection in vivo following antimicrobial exposure (4, 21, 22). We have used the gut model to demonstrate that the duration of toxin production by C. difficile PCR ribotype 027 (strain NAP1) markedly exceeds that of ribotype 001 (23). These findings may help to explain the increased severity of symptoms and the higher case-fatality ratio associated with infections due to C. difficile ribotype 027 (40). We have here extended these findings by determining the propensity of ciprofloxacin, levofloxacin, and moxifloxacin (which has enhanced antianaerobe activity) to induce toxin production by these two epidemic C. difficile strains.

Addition of either ciprofloxacin or moxifloxacin had a marked effect on the enumerated components of the normal microbial flora in the gut model, while levofloxacin dosing had a lesser impact. The concentrations of ciprofloxacin achieved in the gut model were in excess of the MICs of most aerobic and anaerobic bacteria (11), leading to marked declines in the counts of enterococci and lactobacilli and reducing the counts of the Bacteroides spp. and bifidobacteria to the limits of detection. Moxifloxacin administration caused a marked reduction in the counts of enterococci and lactose fermenters, while the counts of lactobacilli and bifidobacteria were mostly unaffected, generally consistent with previous observations by Edlund (19). Moxifloxacin has previously been shown to have good activity against Bacteroides spp. both in vitro and in vivo (1, 43), which is consistent with the results that we obtained with the gut model. However, other reports from in vivo studies with humans did not report such a marked suppression of Bacteroides spp. following fluoroquinolone administration (19, 46). This may be due to the large variations in the gut fluoroquinolone concentrations achieved in patients after fluoroquinolone administration. Levofloxacin had a lesser effect on the normal flora than the other fluoroquinolones but had a marked effect on lactose fermenters, while all other components of the microbiota enumerated declined ≤2 log10 units. The populations rapidly recovered to the predosing levels after antibiotic dosing ceased. These observations are consistent with those presented in previous reports, in which it was reported that levofloxacin elicited a selective reduction in the gut microbiota and predominantly affected gram-negative aerobic bacteria (18).

Addition of all three fluoroquinolones led to germination and high-level toxin production by both C. difficile PCR ribotypes 001 and 027 in the gut model. Previous studies with the gut model have indicated that gut antimicrobial concentrations play an important role in the induction of C. difficile germination and cytotoxin production (4, 21, 22). In the present study, the moxifloxacin and levofloxacin concentrations in vessel 3 were subinhibitory for the duration of the experiments due to the relatively high MICs of the C. difficile strains being investigated. Ciprofloxacin was subinhibitory for C. difficile PCR ribotype 027, which has an MIC of ~300 mg/liter, throughout the experiment. For C. difficile PCR ribotype 001 (MIC, ~64 mg/liter), the levels were inhibitory by day 3 of ciprofloxacin dosing but declined to subinhibitory concentrations 5 days after the final dose was administered. Addition of either moxifloxacin or levofloxacin led to the germination of both C. difficile strains either during or within a day of the end of the antibiotic instillation period. After ciprofloxacin administration, there was a greater lag before germination and toxin production occurred, particularly for C. difficile PCR ribotype 001, which remained quiescent until 10 days after antibiotic dosing finished. This difference in the timing of germination after ciprofloxacin exposure may be related to the markedly different susceptibilities of the two C. difficile strains to ciprofloxacin. There was no consistent relationship between changes in the gut flora populations and the germination of C. difficile. Germination occurred as the obligate anaerobe counts began to decline in the experiments with moxifloxacin, in contrast to the counts seen following ciprofloxacin dosing, i.e., when the gut bacterial populations had recovered to predosing levels. It is interesting that all fluoroquinolones investigated facilitated C. difficile spore germination, growth, and toxin production, despite the markedly reduced effect of levofloxacin on the gut microbiota. This suggests that factors other than the loss of colonization resistance, such as a direct stimulatory effect of the antimicrobial, may induce C. difficile proliferation and toxin production and, thus, the onset of C. difficile infection. Alternatively, components of the indigenous gut microbiota potentially important in colonization resistance to C. difficile infection may not have been enumerated in the present studies due to the limited scope of bacterial enumeration by determination of viable counts.

In the experiments with moxifloxacin and levofloxacin and C. difficile PCR ribotype 001, low-level cytotoxin production (≤2 RU) was observed before antimicrobial dosing commenced. The two experiments in which this phenomenon was observed ran simultaneously and thus used the same fecal inocula and C. difficile spore preparations. We have previously recorded similar findings (22), and this may be due to low levels of vegetative cells being present in the spore preparation, which could have lysed, releasing intracellular cytotoxin (28). After ciprofloxacin and levofloxacin administration, germination was followed by marked cytotoxin production within 2 to 3 days, whereas high-level cytotoxin production by C. difficile PCR ribotype 027 occurred before detectable germination and proliferation were seen following moxifloxacin instillation. This observation is intriguing, as C. difficile has previously been demonstrated to produce toxin as it passes from logarithmic to stationary or decline phase (50). We repeated this experiment and found that the phenomenon of marked early cytotoxin production following moxifloxacin exposure was reproducible for C. difficile PCR ribotype 027 (data not shown); this significant cytotoxin production was not seen when C. difficile PCR ribotype 001 was exposed to moxifloxacin. We have previously demonstrated that the duration of cytotoxin production by C. difficile PCR ribotype 027 was markedly longer than that of C. difficile PCR ribotype 001 (23 days and 13 days, respectively) but that the peak cytotoxin titers were similar for both strains (23).

Early toxin production could indicate the emergence of resistance and the clonal expansion of C. difficile PCR ribotype 027 or the presence of a heteroresistant C. difficile PCR ribotype 027 population. A subpopulation of C. difficile with reduced susceptibility to moxifloxacin may have been able to germinate and produce cytotoxin while the antimicrobial levels in the gut model were still inhibitory to the predominant population. This is supported by the observation of growth on FQCM containing 32 mg/liter moxifloxacin in the experiment with C. difficile PCR ribotype 027 and moxifloxacin after dosing commenced. However, the C. difficile PCR ribotype 027 mutants with a gyrB mutation that we detected were isolated only sporadically and at a low frequency; hence, on all days apart from day 32, when the mutation frequency in vessel 3 was 2.4 x 10–5, we did not detect moxifloxacin-resistant mutants. An alternative resistance mechanism, multidrug efflux, has been reported to mediate fluoroquinolone resistance in other Clostridium spp. (42). The activity of such efflux pumps in C. difficile remains unclear, although cloning and expression of the C. difficile cdeA gene in E. coli has been found to correlate with fluoroquinolone resistance (13). The lower level of recovery of C. difficile PCR ribotype 001 (n = 4) from FQCM compared to that of C. difficile PCR ribotype 027 (n = 62) may be due to differences in the MICs of the two strains in relation to the concentrations of fluoroquinolones included in the moxifloxacin breakpoint plates.

The mutation in gyrB seen in three C. difficile PCR ribotype 027 strains, which led to the amino acid change Glu-466 -> Lys, is novel in this species, although homologous mutations mediating fluoroquinolone resistance in other bacteria have been reported (37). All parent strains and resistant mutants also had a mutation in gyrA that led to the amino acid change Thr-82 -> Ile. This mutation has previously been identified by several groups (2, 12, 15) and was found in all moxifloxacin-resistant C. difficile strains tested but was absent in the susceptible strains. Thr-82 in C. difficile corresponds to Ser-83 in E. coli; and an amino acid substitution at this location has also been associated with fluoroquinolone resistance in several other bacterial species, including Pseudomonas aeruginosa, Campylobacter jejuni, and Enterobacter aerogenes (26). A second gyrA mutation, Ala-83 -> Val, found in one C. difficile PCR ribotype 001 isolate recovered from FQCM, has not previously been reported in C. difficile (14). Its location in the quinolone resistance-determining region suggests that it may play a role in resistance (isolate MIC ≥ 180 mg/liter), as DNA gyrase has been shown to be the primary intracellular target for fluoroquinolones in C. difficile (36).

Fluoroquinolones may have contributed to the recent spread of C. difficile PCR ribotype 027, although there are conflicting data. Several outbreaks of C. difficile infection have been associated with fluoroquinolone use, but it is important to note that most such reports have not controlled for confounding factors, especially exposure to C. difficile (8, 24, 27, 39, 48, 55). Intervention studies that provide convincing evidence of a cause and an effect between the increased incidence of C. difficile infection and fluoroquinolone use are currently lacking. In an outbreak of C. difficile infection that followed a formulary switch from levofloxacin to moxifloxacin, the reversion to levofloxacin was not associated with a decrease in the incidence of C. difficile infection (7). In Montreal, Quebec, Canada, fluoroquinolone use was retrospectively identified as a significant risk factor during a large outbreak of C. difficile infection (39). C. difficile infection rates subsequently decreased in association with antimicrobial restriction of narrow- (–21%), expanded- (–93%), and broad-spectrum (–79%) cephalosporins; clindamycin (–87%); macrolides (–78%); and ciprofloxacin (–29%). However, the rates of use of respiratory fluoroquinolones (predominantly moxifloxacin) and piperacillin-tazobactam increased by +79% and +114%, respectively, as the outbreak was controlled (47). Following an outbreak of C. difficile infection, Muto et al. used a package of measures to first reduce the incidence of infection from peak levels and then eventually to return to the baseline rates of C. difficile infection at a university hospital in the United States (35). In addition to the use of multiple infection control measures, antimicrobial usage was altered, including an increase in the rates of prescription of moxifloxacin and ciprofloxacin (levofloxacin was removed from the formulary). It was recently reported that moxifloxacin promoted increased levels of growth and toxin production by C. difficile in the cecal contents of mice compared with those achieved in ciprofloxacin- or levofloxacin-treated animals. However, when they were administered at higher concentrations, levofloxacin and ciprofloxacin also stimulated C. difficile growth and toxin production (3).

To put our results in context with clinical findings, it is likely that all fluoroquinolones can induce C. difficile infection, but it remains unclear whether some are associated with a higher risk. Resistance to fluoroquinolones may act as a selective pressure for C. difficile. While this trait has been suggested as being significant in the emergence of C. difficile PCR ribotype 027, it is noteworthy that earlier epidemic strains (including C. difficile PCR ribotype 001) shared this phenotype (53). Antibiotic polypharmacy and a long duration of therapy (6, 54), as well as nonantimicrobial risk factors, including exposure to other cases of C. difficile infection (16), influence the likelihood of infection. Additionally, the host response has been show to influence the occurrence of symptomatic disease (29, 30). The in vitro gut model does not mimic the immunological or secretory events that occur within the colon, but it has been shown to be a reliable indicator of whether a drug may have a propensity to induce C. difficile infection in vivo (4, 21). Following fluoroquinolone instillation, the antibiotic concentrations achieved in vessels 2 and 3 were often markedly lower than those achieved in vessel 1. This may be due to antibiotic inactivation by components of the normal flora or binding of the antimicrobial to fecal material, which has previously been demonstrated for other fluoroquinolones (17, 49). Previous investigations into the inactivation of antimicrobial agents, including nalidixic acid, neomycin, and trimethoprim, found evidence of rapid biological inactivation by intestinal contents (49). Edlund et al. also found that norfloxacin reversibly binds to fecal material, potentially reducing its activity in vivo (17). Reduced fluoroquinolone activity in vivo may be relevant, as toxin production in the gut model correlates with sub-MICs of antibiotics (22).

In conclusion, all three fluoroquinolones facilitated germination and cytotoxin production by C. difficile PCR ribotypes 027 and 001, despite differences in the extent of inhibition of the gut flora. These results suggest that all of the fluoroquinolones tested have the propensity to induce C. difficile infection. The detection of isolates with a reduced susceptibility to moxifloxacin is interesting and implies the potential for the selection of such isolates in vivo. The fact that early toxin production was observed only for C. difficile PCR ribotype 027 may suggest some strain-specific differences in the response of C. difficile to fluoroquinolone exposure.


arrow
ACKNOWLEDGMENTS
 
We thank Rob Owens for providing the C. difficile PCR ribotype 027 strain used in this study.

This work was partially supported by a research grant from Bayer HealthCare.

M.H.W. has received honoraria for consultancy work, financial support to attend meetings, and research funding from Astra-Zeneca, Bayer, Cerexa, Genzyme, Nabriva, Pfizer, Targanta, Vicuron, and Wyeth.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Microbiology, Old Medical School, Leeds General Infirmary, and University of Leeds, Leeds LS1 3EX, United Kingdom. Phone: 44 113 3926818. Fax: 44 113 3435649. E-mail: Mark.Wilcox{at}Leedsth.nhs.uk Back

{triangledown} Published ahead of print on 18 August 2008. Back


arrow
REFERENCES
 
    1
  1. Ackermann, G., R. Schaumann, B. Pless, M. C. Claros, E. J. C. Goldstein, and A. C. Rodloff. 2000. Comparative activity of moxifloxacin in vitro against obligately anaerobic bacteria. Eur. J. Clin. Microbiol. Infect. Dis. 19:228-232.[CrossRef][Medline]
    2. 2
  2. Ackermann, G., Y. J. Tang-Feldman, R. Schaumann, J. P. Henderson, A. C. Rodloff, J. Silva, and S. H. Cohen. 2003. Antecedent use of fluoroquinolones is associated with resistance to moxifloxacin in Clostridium difficile. Clin. Microbiol. Infect. 9:526-530.[CrossRef][Medline]
    3. 3
  3. Adams, D. A., M. M. Riggs, and C. J. Donskey. 2007. Effect of fluoroquinolone treatment on growth of and toxin production by epidemic and nonepidemic Clostridium difficile strains in the cecal contents of mice. Antimicrob. Agents Chemother. 51:2674-2678.[Abstract/Free Full Text]
    4. 4
  4. Baines, S. D., J. Freeman, and M. H. Wilcox. 2005. Effects of piperacillin/tazobactam on Clostridium difficile growth and toxin production in a human gut model. J. Antimicrob. Chemother. 55:974-982.[Abstract/Free Full Text]
    5. 5
  5. Baines, S. D., K. Saxton, J. Freeman, and M. H. Wilcox. 2006. Tigecycline does not induce proliferation or cytotoxin production by epidemic C. difficile in a human gut model. J. Antimicrob. Chemother. 58:1062-1065.[Abstract/Free Full Text]
    6. 6
  6. Bignardi, G. E. 1998. Risk factors for Clostridium difficile infection. J. Hosp. Infect. 40:1-15.[Medline]
    7. 7
  7. Biller, P., B. Shank, L. Lind, M. Brennan, L. Tkatch, G. Killgore, A. Thompson, and L. C. McDonald. 2007. Moxifloxacin therapy as a risk factor for Clostridium difficile-associated disease during an outbreak: attempts to control a new epidemic strain. Infect. Control Hosp. Epidemiol. 28:198-201.[CrossRef][Medline]
    8. 8
  8. Brierley, R. 2005. Clostridium difficile—a new threat to public health? Lancet Infect. Dis. 5:535.[Medline]
    9. 9
  9. Brismar, B., C. Edlund, M. Malmbourg, and C. E. Nord. 1990. Ciprofloxacin concentrations and impact of colon microbiota of patients undergoing colorectal surgery. Antimicrob. Agents Chemother. 34:481-483.[Abstract/Free Full Text]
    10. 10
  10. Chenna, R., H. Sugawara, T. Koike, T. J. Gibson, D. G. Higgins, and J. D. Thompson. 2003. Multiple sequence alignment with the Clustal series of programs. Nucleic Acids Res. 31:3497-3500.[Abstract/Free Full Text]
    11. 11
  11. Chin, N.-X., and H. C. Neu. 1984. Ciprofloxacin, a quinolone carboxylic acid compound active against aerobic and anaerobic bacteria. Antimicrob. Agents Chemother. 25:319-326.[Abstract/Free Full Text]
    12. 12
  12. Dridi, L., J. Tankovic, B. Burghoffer, F. Barbut, and J. Petit. 2002. gyrA and gyrB mutations are implicated in cross resistance to ciprofloxacin and moxifloxacin in Clostridium difficile. Antimicrob. Agents. Chemother. 46:3418-3421.[Abstract/Free Full Text]
    13. 13
  13. Dridi, L., J. Tankovic, and J. Petit. 2004. CdeA of Clostridium difficile, a new multidrug efflux transporter of the MATE family. Microb. Drug Resist. 10:191-196.[Medline]
    14. 14
  14. Drudy, D., T. Quinn, R. O'Mahony, L. Kyne, P. O'Gaora, and S. Fanning. 2006. High-level resistance to moxifloxacin and gatifloxacin associated with a novel mutation in gyrB in toxin A-negative, toxin B-positive Clostridium difficile. J. Antimicrob. Chemother. 58:1264-1267.[Abstract/Free Full Text]
    15. 15
  15. Drudy, D., L. Kyne, R. O'Mahoney, and S. Fanning. 2007. gyrA mutations in fluoroquinolone resistant Clostridium difficile PCR-027. Emerg. Infect. Dis. 13:504-505.[Medline]
    16. 16
  16. Dubberke, E. R., K. A. Reske, Y. Yan, M. A. Olsen, L. C. McDonald, and V. J. Fraser. 2007. Clostridium difficile associated disease in a setting of endemicity: identification of novel risk factors. Clin. Infect. Dis. 45:1543-1549.[Medline]
    17. 17
  17. Edlund, C., L. Lindqvist, and C. E. Nord. 1988. Norfloxacin binds to human faecal material. Antimicrob. Agents Chemother. 32:1869-1874.[Abstract/Free Full Text]
    18. 18
  18. Edlund, C., S. Sjöstedt, and C. E. Nord. 1997. Comparative effects of levofloxacin and ofloxacin on the normal oral and intestinal microflora. Scand. J. Infect. Dis. 29:383-386.[Medline]
    19. 19
  19. Edlund, C. 2000. Comparative effects of moxifloxacin and clarithromycin on the normal intestinal microflora. Scand. J. Infect. Dis. 32:81-85.[CrossRef][Medline]
    20. 20
  20. Freeman, J., and M. H. Wilcox. 1999. Antibiotics and Clostridium difficile. Microbes Infect. 1:377-384.[CrossRef][Medline]
    21. 21
  21. Freeman, J., F. J. O'Neill, and M. H. Wilcox. 2003. Effects of cefotaxime and desacetylcefotaxime upon Clostridium difficile proliferation and toxin production in a triple-stage chemostat model of the human gut. J. Antimicrob. Chemother. 52:96-102.[Abstract/Free Full Text]
    22. 22
  22. Freeman, J., S. D. Baines, D. Jabes, and M. H. Wilcox. 2005. Comparison of the efficacy of ramoplanin and vancomycin in both in vitro and in vivo models of clindamycin-induced Clostridium difficile infection. J. Antimicrob. Chemother. 56:717-725.[Abstract/Free Full Text]
    23. 23
  23. Freeman, J., S. D. Baines, K. Saxton, and M. H. Wilcox. 2007. Effect of metronidazole on growth and toxin production by epidemic Clostridium difficile PCR ribotypes 001 and 027 in a human gut model. J. Antimicrob. Chemother. 60:83-91.[Abstract/Free Full Text]
    24. 24
  24. Gaynes, R., D. Rimland, E. Killum, H. K. Lowery, T. M. Johnson II, G. Killgore, and F. C. Tenover. 2004. Outbreak of Clostridium difficile infection in a long term nursing facility: association with gatifloxacin use. Clin. Infect. Dis. 38:640-645.[CrossRef][Medline]
    25. 25
  25. George, W. L., R. D. Rolfe, and S. M. Finegold. 1982. Clostridium difficile and its cytotoxin in feces of patients with antimicrobial agent-associated diarrhea and miscellaneous conditions. J. Clin. Microbiol. 15:1049-1053.[Abstract/Free Full Text]
    26. 26
  26. Hooper, D. C. 1999. Mechanisms of fluoroquinolone resistance. Drug Resist. Updat. 2:38-55.[CrossRef][Medline]
    27. 27
  27. Joseph, R., D. Demeyer, D. Vanrenterghem, R. van den Berg, E. Kuijper, and M. Delmée. 2005. First isolation of Clostridium difficile PCR ribotype 027, toxinotype III in Belgium. Euro. Surveill. 10:E051020.4. http://www.eurosurveillance.org/ew/2005/051020.asp#4.
    28. 28
  28. Ketley, J. M., S. C. Halsam, T. J. Mitchell, J. Stephen, D. C. Candy, and D. W. Burdon. 1984. Production and release of toxins A and B by C. difficile. J. Med. Microbiol. 18:385-391.[Abstract/Free Full Text]
    29. 29
  29. Kyne, L., M. Warny, A. Qamar, and C. P. Kelly. 2000. Asymptomatic carriage of Clostridium difficile and serum levels of IgG antibody against toxin A. N. Engl. J. Med. 342:390-397.[Abstract/Free Full Text]
    30. 30
  30. Kyne, L., M. Warny, A. Qamar, and C. P. Kelly. 2001. Association between antibody response to toxin A and protection against recurrent Clostridium difficile diarrhoea. Lancet 357:158-159.[CrossRef][Medline]
    31. 31
  31. Loo, V. G., L. Poirier, M. A. Miller, M. Oughton, M. D. Libman, S. Michaud, A. M. Bourgault, T. Nguyen, C. Frenette, M. Kelly, A. Vibien, P. Brassard, S. Fenn, K. Dewar, T. J. Hudson, R. Horn, P. René, Y. Monkzak, and A. Dascal. 2005. A predominantly clonal multi-institutional outbreak of Clostridium difficile associated diarrhoea with high morbidity and mortality. N. Engl. J. Med. 353:2442-2449.[Abstract/Free Full Text]
    32. 32
  32. Lovering, A. M. 1999. Lincosamides: clindamycin and lincomycin, p. 193-198. In D. S. Reeves, R. Wise, J. M. Andrews, and L. O. White (ed.), Clinical antimicrobial assays. The British Society for Antimicrobial Chemotherapy, Oxford, United Kingdom.
    33. 33
  33. Macfarlane, G. T., S. Macfarlane, and G. R. Gibson. 1998. Validation of a three-stage compound continuous culture system for investigating the effect of retention time on the ecology and metabolism of bacteria in the human colon. Microb. Ecol. 35:180-187.[CrossRef][Medline]
    34. 34
  34. McDonald, L. C., G. E. Killgore, A. Thompson, R. C. Owens, Jr., S. V. Kazakova, S. P. Sambol., S. Johnson, and D. N. Gerding. 2005. An epidemic, toxin gene-variant strain of Clostridium difficile. N. Engl. J. Med. 353:2433-2441.[Abstract/Free Full Text]
    35. 35
  35. Muto, C. A., M. K. Blank, J. W. Marsh, E. N. Vergis, M. M. O'Leary, K. A. Schutt, A. W. Pasculle, M. Pokrywka, J. G. Garcia, K. Posey, T. L. Roberts, B. A. Potoski, G. E. Blank, R. L. Simmons, P. Veldkamp, L. H. Harrison, and D. L. Paterson. 2007. Control of an outbreak of infection with the hypervirulent Clostridium difficile BI strain in a university hospital using a comprehensive ‘bundle’ approach. Clin. Infect. Dis. 45:1266-1273.[CrossRef][Medline]
    36. 36
  36. Oh, H., and C. Edlund. 2003. Mechanism of quinolone resistance in anaerobic bacteria. Clin. Microbiol. Infect. 9:512-517.[CrossRef][Medline]
    37. 37
  37. Ouabdesselam, S., D. C. Hooper, J. Tankovic, and C. J. Soussy. 1995. Detection of gyrA and gyrB mutations in quinolone-resistant clinical isolates of Escherichia coli by single-strand conformational polymorphism analysis and determination of levels of resistance conferred by two different single gyrA mutations. Antimicrob. Agents Chemother. 39:1667-1670.[Abstract]
    38. 38
  38. Paton, J. H., H. A. Holt, and M. J. Bywater. 1990. Measurement of MICs of antibacterial agents by spiral gradient endpoint compared with conventional dilution method. Int. J. Exp. Clin. Chemother. 3:31-38.
    39. 39
  39. Pepin, J., N. Saheb, M. A. Coloumbe, M. E. Alary, M. P. Corriveau, S. Authier, M. Leblanc, G. Rivard, M. Bettez, V. Primeau, M. Nguyen, C. E. Jacob, and L. Lanthier. 2005. Emergence of fluoroquinolones as the predominant risk factor for Clostridium difficile-associated diarrhea: a cohort study during an epidemic in Quebec. Clin. Infect. Dis. 41:1254-1260.[CrossRef][Medline]
    40. 40
  40. Pepin, J., L. Valiquette, and B. Cossette. 2005. Mortality attributable to nosocomial Clostridium difficile associated disease during an epidemic caused by a hypervirulent strain in Quebec. Can. Med. Assoc. J. 173:1037-1042.[Abstract/Free Full Text]
    41. 41
  41. Pultz, N. J., and C. J. Donskey. 2005. Effect of antibiotic treatment on growth of and toxin production by Clostridium difficile in the cecal contents of mice. Antimicrob. Agents Chemother. 49:3529-3532.[Abstract/Free Full Text]
    42. 42
  42. Rafii, F., M. Park, and R. Wynne. 2005. Evidence for active drug efflux in fluoroquinolone resistance in Clostridium hathewayi. Int. J. Exp. Clin. Chemother. 51:256-262.
    43. 43
  43. Schaumann, R., R. Blatz, J. Beer, G. Ackermann, and A. C. Rodloff. 2004. Effect of moxifloxacin versus imipenem/cilastatin treatment on the mortality of mice infected intravenously with different strains of Bacteroides fragilis and Escherichia coli. J. Antimicrob. Chemother. 53:318-324.[Abstract/Free Full Text]
    44. 44
  44. Settle, C. D., M. H. Wilcox, W. N. Fawley, O. J. Corrado, and P. M. Hawkey. 1998. Prospective study of the risk of Clostridium difficile diarrhoea in elderly patients following treatment with cefotaxime or piperacillin-tazobactam. Aliment. Pharmacol. Ther. 12:1217-1223.[CrossRef][Medline]
    45. 45
  45. Stubbs, S. L. J., J. Brazier, G. L. O'Neill, and B. Duerden. 1999. PCR targeted to the 16S-23S rRNA gene intergenic spacer region of Clostridium difficile and construction of a library consisting of 116 different PCR ribotypes. J. Clin. Microbiol. 37:461-463.[Abstract/Free Full Text]
    46. 46
  46. Sullivan, A., C. Edlund, and C. E. Nord. 2001. Effect of antimicrobial agents on the ecological balance of human microflora. Lancet Infect. Dis. 1:101-114.[CrossRef][Medline]
    47. 47
  47. Valiquette, L., B. Cossette, M.-P. Garant, H. Diab, and J. Pepin. 2007. Impact of a reduction in the use of high risk antibiotics on the course of an epidemic of Clostridium difficile associated disease caused by the hypervirulent NAP1/027 strain. Clin. Infect. Dis. 45:S112-S121.[CrossRef][Medline]
    48. 48
  48. van Steenbergen, J., S. Debast, E. van Kregten, R. van den Berg, D. Notermans, and E. Kuijper. 2005. Isolation of Clostridium difficile ribotype 027, toxinotype III in The Netherlands after increase in C. difficile-associated diarrhoea. Euro. Surveill. 10:E050714.1. http://www.eurosurveillance.org/ew/2005/050714.asp#1.
    49. 49
  49. Veringa, E. M., and D. van der Waaij. 1984. Biological inactivation by faeces of antimicrobial drugs applicable in selective decontamination of the digestive tract. J. Antimicrob. Chemother. 14:605-612.[Abstract/Free Full Text]
    50. 50
  50. Voth, D. E., and J. D. Ballard. 2005. Clostridium difficile toxin: mechanism of action and role in disease. Clin. Microbiol. Rev. 18:247-263.[Abstract/Free Full Text]
    51. 51
  51. Warny, M., J. Pepin, A. Fang, G. Killgore, A. Thompson, J. Brazier, E. Frost, and L. McDonald. 2005. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet 366:1079-1084.[CrossRef][Medline]
    52. 52
  52. White, L. O., and A. M. Tobin. 1999 The 4-quinolones, p 193-198. In D. S. Reeves, R. Wise, J. M. Andrews and L. O. White (ed.), clinical antimicrobial Assays. The British Society for Antimicrobial Chemotherapy, Oxford, United Kingdom.
    53. 53
  53. Wilcox, M. H., W. N. Fawley, J. Freeman, and J. Stead. 2000. In vitro activity of new generation fluoroquinolones against genotypically distinct and indistinguishable Clostridium difficile isolates. J. Antimicrob. Chemother. 46:551-556.[Abstract/Free Full Text]
    54. 54
  54. Wistrom, J., S. R. Norrby, E. B. Myhre, S. Eriksson, G. Granström, L. Lagergren, G. Englund, C. E. Nord, and B. Svenungsson. 2001. Frequency of antibiotic associated diarrhoea in 2462 antibiotic treated hospitalised patients: a prospective study. J. Antimicrob. Chemother. 47:43-50.[Abstract/Free Full Text]
    55. 55
  55. Yip, C., M. Loeb, S. Salama, L. Moss, and J. Olde. 2001. Quinolone use as a risk factor for nosocomial Clostridium difficile-associated diarrhoea. Infect. Control Hosp. Epidemiol. 22:572-575.[CrossRef][Medline]

Antimicrobial Agents and Chemotherapy, February 2009, p. 412-420, Vol. 53, No. 2
0066-4804/09/$08.00+0     doi:10.1128/AAC.00306-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Spigaglia, P., Barbanti, F., Louie, T., Barbut, F., Mastrantonio, P. (2009). Molecular Analysis of the gyrA and gyrB Quinolone Resistance-Determining Regions of Fluoroquinolone-Resistant Clostridium difficile Mutants Selected In Vitro. Antimicrob. Agents Chemother. 53: 2463-2468 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Saxton, K.
Right arrow Articles by Wilcox, M. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Saxton, K.
Right arrow Articles by Wilcox, M. H.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»