JAC Advance Access originally published online on September 4, 2008
Journal of Antimicrobial Chemotherapy 2008 62(5):1078-1085; doi:10.1093/jac/dkn358
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Comparison of oritavancin versus vancomycin as treatments for clindamycin-induced Clostridium difficile PCR ribotype 027 infection in a human gut model
1 Department of Microbiology, Institute of Molecular and Cellular Biology, University of Leeds, Leeds LS2 9JT, UK 2 Department of Microbiology, Leeds Teaching Hospitals NHS Trust, The General Infirmary, Old Medical School, Leeds, Leeds LS1 3EX, UK
* Correspondence address. Department of Microbiology, The General Infirmary, Old Medical School, Leeds LS1 3EX, UK. Tel: +44-113-3926818; Fax: +44-113-3435649; E-mail: mark.wilcox{at}leedsth.nhs.uk
Received 1 July 2008; returned 21 July 2008; revised 5 August 2008; accepted 6 August 2008
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Objectives: To compare the efficacy of oritavancin and vancomycin in the treatment of Clostridium difficile infection (CDI) using an in vitro human gut model.
Methods: We induced CDI by instilling clindamycin into an in vitro gut model primed with pooled human faeces and C. difficile ribotype 027 spores. Oritavancin and vancomycin were instilled in separate experiments at levels equivalent to those expected in the faeces (vancomycin) of patients or levels limited by the solubility of the drug (oritavancin).
Results: Clindamycin exposure elicited C. difficile proliferation and high-level cytotoxin production in both experiments. Vancomycin instillation reduced vegetative C. difficile numbers within 1 day but did not affect the numbers of C. difficile spores. Oritavancin instillation markedly reduced C. difficile vegetative numbers and spores to below the limits of detection within 2 days. Cytotoxin titres in both experiments declined to the limits of detection after instillation with oritavancin or vancomycin, but did so more quickly (within 5 days) in the vancomycin experiment. Cessation of vancomycin instillation was associated with further C. difficile proliferation and high-level cytotoxin production. Conversely, toxin recrudescence was not observed following cessation of oritavancin.
Conclusions: Both oritavancin and vancomycin were effective in treating clindamycin-induced CDI in a human gut model, but only oritavancin appeared active against spore forms of C. difficile. Furthermore, recurrence of high-level cytotoxin production was observed following vancomycin instillation but not oritavancin. Oritavancin therapy may be more effective in treating CDI than vancomycin, possibly because it may prevent recrudescence of C. difficile spores.
Keywords: spores , chemostat , faeces
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Clostridium difficile is a major cause of morbidity in the hospitalized elderly and is almost exclusively associated with antimicrobial therapy. C. difficile infection (CDI) may range in severity from mild antibiotic-associated diarrhoea/colitis to life-threatening pseudomembranous colitis. Treatment strategies for CDI have changed little over the past two decades. Oral metronidazole (400–500 mg three times daily) or vancomycin (125 mg four times daily) are most commonly used to treat CDI.1 Early studies demonstrated little difference between metronidazole and vancomycin in terms of response or recurrence rates,2,3 although response time was faster with the latter.4 More recent reports have questioned the efficacy of metronidazole therapy for CDI,5,6 particularly for disease attributable to apparently hypervirulent C. difficile PCR ribotype 027 (NAP1/BI). Whether more severe disease associated with C. difficile PCR ribotype 027 is a consequence of enhanced virulence in this strain,7,8 or alternatively due to reduced efficacy of antimicrobial therapy remains to be determined definitively. Despite recent evidence of reduced susceptibility to metronidazole in C. difficile PCR ribotype 001, similar impaired activity has not been seen for C. difficile PCR ribotype 027.9 Studies demonstrating increased severity of C. difficile PCR ribotype 027 associated CDI have reinforced the need for evaluation of the activity of new therapeutic antimicrobial agents. We have previously evaluated a triple-stage chemostat human gut model as an in vitro system to study both antimicrobial predisposition to induction of CDI10–12 and also antimicrobial efficacy in the treatment of antibiotic-induced CDI.7,13 In this study, we used the same in vitro human gut model to compare in separate experiments the efficacy of oritavancin, a semi-synthetic lipoglycopeptide antibiotic, and vancomycin for the treatment of clindamycin-induced CDI caused by epidemic C. difficile PCR ribotype 027.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
C. difficile strains
A single isolate of C. difficile PCR ribotype 027 was investigated in these experiments. This isolate was a clinical strain recovered during an epidemic of CDI at the Maine Medical Centre (Portland, USA) in 2005. It was supplied courtesy of Dr Rob Owens and ribotyped at the Anaerobe Reference Laboratory (Cardiff, Wales, UK) by Dr Jon Brazier.
Triple-stage chemostat human gut model
The gut model was designed to allow the study of the intestinal microflora in the low pH, carbohydrate-excess conditions of the proximal colon and the carbohydrate-depleted, non-acidic conditions of the distal colon.14 Microbiological and physicochemical measurements within the gut model vessels were validated against the intestinal contents of sudden death victims.14 Each gut model consists of three fermentation vessels connected in a weir cascade system top-fed with growth medium at a controlled rate (D = 0.015 h–1). Vessels are sparged with oxygen-free nitrogen to ensure anaerobiosis, heated via a water-jacketed system (37°C) and maintained at specific pH using fermentor controller units (Biosolo 3, Brighton Systems, UK) delivering 0.5 M HCl/NaOH. Vessel 1 (280 mL) operates at pH 5.5 and high substrate availability, thus reflecting the conditions within the proximal colon, while vessels 2 and 3 (300 mL) operate at pH 6.2 and 6.8, respectively, with low substrate availability, thus reflecting the conditions within the distal colon. The gut model is primed with faecal slurry 
10% (w/v) and allowed to equilibrate with respect to bacterial populations for 14 days.
Faecal samples were collected from five healthy elderly (>65 years) volunteers and immediately transported to the laboratory under anaerobic conditions (GasPak, Oxoid, Basingstoke, UK). Stools were confirmed as C. difficile culture-negative on Brazier's CCEY agar incorporating 5 mg/L lysozyme (Sigma-Aldrich, UK) as reported previously.10,12 C. difficile-negative faeces were pooled and a coarse-filtered slurry (10% w/v) in sterile pre-reduced phosphate-buffered saline was prepared. Vessels of the gut models were filled to approximately two-thirds volume, and the growth medium pumps were started.
Enumeration of gut microflora and C. difficile
Major culturable components of the indigenous gut microflora and C. difficile were enumerated by viable counting (log10 cfu/mL) on selective and non-selective agars as reported previously.13 Bacterial groups enumerated were: total facultative anaerobes, facultatively anaerobic lactose fermenters, total anaerobes, bifidobacteria, Bacteroides fragilis group, total clostridia, lactobacilli, enterococci, total C. difficile and C. difficile spores. C. difficile cytotoxin production was quantified using a Vero cell cytotoxicity assay as described previously.12 Cytotoxin titres were expressed in log10 relative units (RU). Only C. difficile total bacterial counts, spore counts and cytotoxin titres were enumerated in vessel 1 of the gut model.
The use of the gut model to evaluate therapeutic interventions for CDI has been described previously.7,13 Briefly, following inoculation of each gut model with the faecal slurry, no further interventions were made for 13 days (period A). During period A, bacterial populations were enumerated every 2 days. Next (day 14), vessel 1 of each gut model was inoculated with a single inoculum of 107 cfu C. difficile PCR ribotype 027 spores,12 with no further interventions for 7 days (period B). From this point onwards, bacterial populations were enumerated daily. A further single inoculum of C. difficile PCR ribotype 027 spores was instilled into vessel 1 of each gut model (day 21) in addition to 33.9 mg/L clindamycin (Pfizer, USA) four times daily for 7 days (period C). This instillation regimen aimed to maintain clindamycin levels within the gut model approximately equivalent to those observed in vivo following a single 600 mg dose.15 Following cessation of clindamycin instillation, no further interventions were made (period D) until high-level cytotoxin production (
4 RU) was observed for at least 2 consecutive days. Instillation of vancomycin (single experiment, 125 mg/L four times daily) and oritavancin (single experiment, 64 mg/L two times daily) was initiated on day 39 in each experiment for 7 days (period E). Vancomycin (Sigma-Aldrich) was prepared in distilled water and oritavancin (Targanta Therapeutics, Cambridge, USA) was prepared in 0.002% (v/v in distilled water) polysorbate-80 (Sigma-Aldrich) and both antimicrobial solutions were sterilized by filtration (0.22 µm) prior to instillation into the gut model. The vancomycin instillation regimen aimed to achieve antibiotic concentrations reflective of those observed in vivo following a standard course of therapy in humans.16 The antimicrobial instillation regimen for oritavancin was suggested by Targanta Therapeutics taking into account the solubility limits of the drug in buffered culture media (G. Moeck, Targanta Therapeutics, Quebec, Canada, personal communication). Following cessation of therapeutic antimicrobial instillation, bacterial populations and cytotoxin titres were followed for a further 15 days (period F).
Bioassay of active antibiotic concentrations
Concentrations of clindamycin were not determined in the present studies. Vancomycin and oritavancin concentrations within the vessels of each gut model were determined using an in-house large plate bioassay. Briefly, samples (1 mL) from all vessels of each gut model were centrifuged (16 000 g) and stored at –20°C prior to bioassay. One hundred millilitres of Muller–Hinton agar (Oxoid, UK) supplemented with 1 M para-amino benzoic acid was sterilized by autoclaving, cooled to 50°C and inoculated with 1 mL of the indicator organism Staphylococcus aureus ATCC 29 213 (at a turbidity equivalent to that of a 0.5 McFarland standard). Molten agar was transferred aseptically into 245 mm2 Petri dishes (Fisher Scientific, Loughborough, UK) and allowed to set. We evaluated whether zone diameters differed with calibrators diluted in distilled water or sterile pH-adjusted (5.5, 6.2, 6.8) gut model fluid. As there was no appreciable difference in zone diameter in the different diluents, deionized water containing 0.002% polysorbate-80 was used for all oritavancin calibrators. Vancomycin calibrators were prepared in sterile deionized water. Samples from the gut model were sterilized by filtration (0.22 µm). Twenty-five wells (9 mm diameter) were removed from the agar using a number 5 cork borer. Twenty microlitres of each doubling dilution of vancomycin calibrator (1–512 mg/L), oritavancin calibrator (1–128 mg/L) or sample from the gut model was assigned randomly to each well in triplicate. Oritavancin was filtered at 1280 mg/L as filtration at lower concentrations leads to significant proportional losses of drug owing to saturable surface binding (G. Moeck, Targanta Therapeutics, Quebec, Canada, personal communication). Agar plates were refrigerated (4°C) for 5 h to allow antimicrobial diffusion while minimizing bacterial growth, following which bioassay plates were incubated aerobically at 37°C for 48 h. Zone diameters were measured using callipers accurate to 0.01 mm and calibration lines were produced by plotting diameter squared against log2 concentration of antimicrobial. Unknown antimicrobial concentrations were read from the calibration line for each plate, and converted into actual concentrations using an inverse log2 function. Mean antimicrobial concentrations (mg/L) were averaged from the three replicates. Limits of detection for oritavancin and vancomycin bioassays were 2 and 8 mg/L, respectively.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Inter-experiment variations in viable counts of indigenous gut bacteria and C. difficile PCR ribotype 027 between vessels 2 and 3 were very small. Therefore, only the results from vessel 3 are presented. Meaningful differences in observations between vessels 2 and 3 of the gut models will be highlighted where appropriate. Viable counts of enumerated components of the indigenous gut microflora were stable throughout period A in both oritavancin (Figure 1a) and vancomycin (Figure 1b) experiments. Instillation of C. difficile PCR ribotype 027 spores (day 14, period B) did not substantially affect viable counts of any enumerated components of the gut microflora in either experiment.
|
Effects of clindamycin exposure on gut microflora
Instillation of clindamycin (period C) elicited marked declines in populations of bifidobacteria (6 log10 cfu/mL), which were below the limits of detection (2 log10 cfu/mL) by the end of period C in both experiments (Figure 1a and b). Viable counts of Bacteroides and lactobacilli declined by 1–2 log10 cfu/mL in both experiments, while enterococcal viable counts increased by 2 log10 cfu/mL. Following cessation of clindamycin instillation (period D), bifidobacterial populations remained below the limits of detection prior to instillation of oritavancin and vancomycin. All other components of the indigenous gut microflora recovered to or exceeded their steady-state (period A) concentrations.
Effects of oritavancin and vancomycin instillation on gut microflora
Instillation of oritavancin (period E) was followed by minor deleterious effects on the indigenous gut microflora that we enumerated. Bacteroides and enterococcal populations were the only bacterial groups adversely affected by oritavancin instillation, declining by 1 and 2 log10 cfu/mL, respectively (Figure 1a). Declines in Bacteroides and enterococcal populations following vancomycin instillation were 6 and 1 log10 cfu/mL, respectively (Figure 1b). Bifidobacterial populations remained below the limits of detection during period E in both experiments. Following cessation of oritavancin instillation (period F), all bacterial populations recovered to steady-state (period A) concentrations, except bifidobacteria which remained below the limits of detection (Figure 1a). Following cessation of vancomycin instillation, all indigenous gut bacterial populations recovered to steady-state (period A) concentrations, except bifidobacteria, which were 2 log10 cfu/mL lower (Figure 1b).
C. difficile were not recovered during steady state (period A) in either experiment. In the absence of clindamycin instillation (period B), C. difficile PCR ribotype 027 remained as spores in all vessels of the gut model (data for vessels 1 and 2 not shown) in both experiments (Figure 2a and b). C. difficile numbers declined by 1 log10 cfu/mL in vessel 3 and at a similar rate during period B in both experiments and cytotoxin production was not detected.
|
Effects of clindamycin exposure on C. difficile
C. difficile remained as spores during clindamycin instillation (period C) in both oritavancin and vancomycin experiments (Figure 2a and b). Additionally, cytotoxin production was not detected during this period. Following cessation of clindamycin instillation (period D), C. difficile remained as spores, which were at the limits of detection 5 days after cessation of clindamycin instillation in the oritavancin experiment (Figure 2a). In the same period during the vancomycin experiment, C. difficile spore numbers declined for the first 2 days of period D, following which C. difficile spore numbers increased by 1 log10 cfu/mL (Figure 2b). Germination of C. difficile spores was detected 6 and 7 days after cessation of clindamycin instillation in oritavancin and vancomycin experiments, respectively. Vegetative C. difficile numbers increased sharply to peak viable counts of 6 log10 cfu/mL in both experiments. Cytotoxin production was detected on day 35 in both experiments and reached maximal titres of 5 RU in both experiments. Instillation of treatment antimicrobial agents commenced on day 39 in both experiments. C. difficile spore germination, proliferation and high-level cytotoxin production were not observed in vessel 1 of the gut model in either oritavancin or vancomycin experiments during period D (data not shown).
Effects of oritavancin instillation on C. difficile
C. difficile total counts were 2 log10 cfu/mL above spore counts (6 log10 cfu/mL) when instillation of oritavancin commenced (period E, Figure 2a). On day 2 of oritavancin instillation, both C. difficile total counts and spore counts declined by 2 log10 cfu/mL and were below the limits of detection (1.22 log10 cfu/mL) 1 day later, remaining so for the rest of period E. Cytotoxin titres declined by 3RU during oritavancin instillation (period E). Oritavancin concentrations peaked at 128, 109 and 52 mg/L in vessels 1, 2 and 3, respectively, i.e. 8-fold lower than those achieved in the vancomycin gut model. It is possible that the concentration of oritavancin demonstrated in culture samples from the gut model may be under-represented due to potential loss following filtration as concentrations were <1280 mg/L (G. Moeck and F. Arhin, Targanta Therapeutics, Quebec, Canada, personal communication). All calibration line R2 values were >0.95.
Effects of vancomycin instillation on C. difficile
C. difficile total counts declined by 1.5 log10 cfu/mL after 1 day of vancomycin instillation (period E, Figure 2b). C. difficile spore counts were unaffected by vancomycin instillation and C. difficile remained predominantly as spores for the remainder of period E. Cytotoxin titres declined by 3 RU during vancomycin instillation. Vancomycin concentrations peaked at 957, 800 and 423 mg/L in vessels 1, 2 and 3 of the gut model, respectively. All calibration line R2 values were >0.99.
Events following cessation of oritavancin instillation
C. difficile was isolated sporadically at the limits of detection for the remainder of the experiment (period F, Figure 2a). Centrifugation and washing of culture samples, in addition to exposure to activated charcoal (20–40 g/L) in an effort to minimize oritavancin carry-over, did not enhance the recovery of C. difficile (data not shown). Cytotoxin titres continued to decline and were below the limits of detection within 10 days. Oritavancin concentrations were below the limits of detection 11 days after cessation of oritavancin instillation.
Events following cessation of vancomycin instillation
C. difficile remained as spores until 12 and 13 days after cessation of vancomycin instillation in vessels 2 (data not shown) and 3, respectively (period F, Figure 2b), following which recurrent germination, proliferation and high-level cytotoxin production were observed. C. difficile total counts were 6 log10 cfu/mL and cytotoxin titres were 5 RU by the end of the experiment.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
We have previously reported the use of the gut model in both the evaluation of antimicrobial predisposition to induction of CDI10–12 and also evaluation of therapeutic interventions for CDI.7,13 Observations within the gut model largely reflected clinical observations, i.e. exposure of C. difficile to antimicrobials with a known propensity to induce CDI in vivo (e.g. cefotaxime and clindamycin) facilitated germination, proliferation and high-level cytotoxin production within the gut model.7,12,13 Conversely, antimicrobial agents not readily associated with the development of CDI in vivo (e.g. piperacillin/tazobactam and tigecycline) failed to facilitate sustained C. difficile germination, proliferation and high-level cytotoxin production.10,11 Prior gut model experiments examining antimicrobial agents for treating clindamycin-induced CDI within the gut model have included evaluations of metronidazole (C. difficile PCR ribotypes 001 and 027), vancomycin (C. difficile PCR ribotype 001) and ramoplanin (C. difficile PCR ribotype 001).7,13
In the present study, clindamycin was utilized to facilitate C. difficile germination, proliferation and high-level cytotoxin production due to the high degree of reproducibility observed in prior gut model studies. All prior gut model studies using clindamycin demonstrated that C. difficile spores remain quiescent prior to and during clindamycin instillation and clindamycin elicited almost identical alterations of indigenous gut microfloras. These observations were reflected in the present studies. Despite the fact that both of these studies were performed once only, we believe that the gut model is a highly reproducible model system to study the interaction between C. difficile, antimicrobial agents and the human colonic microflora. Clindamycin concentrations were not assayed in the present studies; however, in all prior experiments, C. difficile spore germination, proliferation and high-titre cytotoxin production occurred only after clindamycin concentrations declined to sub-MIC levels for the C. difficile strain under evaluation. The timing of C. difficile PCR ribotype 027 proliferation and cytotoxin production following clindamycin instillation in the present study suggested that this was also the case in experiments evaluating oritavancin and vancomycin. Cytotoxin production was observed in both experiments 1 day after detectable C. difficile spore germination, which was earlier (1–2 days) in the growth cycle than was observed with C. difficile PCR ribotype 001.7,13 This may indicate that C. difficile PCR ribotype 027 releases toxins A and B earlier in the growth cycle than other C. difficile strains,8 although the present studies employed a once-daily sampling regimen, which is not ideal for growth/kinetic studies. Warny et al.8 reported quantitatively higher toxin production by C. difficile PCR ribotype 027 than non-epidemic (toxinotype 0) C. difficile PCR ribotypes, but failed to acknowledge that a statistically significantly greater cell density in PCR ribotype 027 cultures compared with non-epidemic C. difficile may have been responsible for the observed ‘elevated’ toxin production.17 The present studies clearly demonstrate that peak cytotoxin titres produced by C. difficile PCR ribotype 027 (maximal titre 5 RU) do not exceed those of other C. difficile PCR ribotypes in the gut model (maximal C. difficile PCR ribotype 001 cytotoxin titre 6 RU), as reported previously.7,13
Changes in the indigenous gut microflora cultured in the present studies following clindamycin instillation largely reflected those reported previously, i.e. reduced obligate anaerobes and increased or maintained populations of facultative anaerobes.7,13 Peak vancomycin concentrations in the present study were 4-fold greater than those reported previously in our gut model,13 but nevertheless were still reflective of those concentrations observed in vivo following a standard therapeutic course.16 The reason for this intra-experiment variation in antimicrobial levels is unclear but may be a consequence of differing methodologies for antimicrobial concentration determination between studies. Peak vancomycin concentrations in vessel 3 of the gut model were 15-fold greater than the MIC90 for B. fragilis.18 Therefore, the profoundly deleterious effect of vancomycin against this bacterial group observed in the present study was not surprising and reflected prior gut model studies.13 Oritavancin solubility limitations in buffered culture media have precluded the determination of definitive MICs for B. fragilis,18 but the slow decline in viable counts observed in the present study suggests that bactericidal concentrations of oritavancin were not attained within the vessels of the gut model. Bifidobacterium spp. failed to recover to steady-state (period A) concentrations following the cessation of oritavancin instillation, which contrasted with observations following vancomycin instillation. Similarly, bifidobacteria failed to recover following instillation of ramoplanin in a prior gut model study.13 The significance of these observations is unclear. Whether prolonged perturbation of the gut microflora following oritavancin (or ramoplanin) therapy is likely in vivo remains to be determined. Ideally, data from in vitro studies should be combined with clinical experience to better understand the effect of antimicrobial agents on the human gut microflora.
Both oritavancin and vancomycin were effective in inhibiting vegetative C. difficile in the present studies. Vancomycin instillation facilitated the inhibition of vegetative C. difficile such that only C. difficile spores remained in the vessels of the gut model 1 day after the start of vancomycin instillation. C. difficile spores were unaffected by the presence of vancomycin concentrations that were >500-fold above the MIC for vegetative forms of C. difficile PCR ribotype 027.19 The failure of vancomycin to elicit inhibitory activity against C. difficile spores at concentrations observed in faeces has been reported previously and supports the data presented in this study.13,20,21 Additionally, following a decline in vancomycin concentrations below the limits of detection in vessels 2 and 3 of the gut model during period F, a further episode of C. difficile spore germination, proliferation and high-level cytotoxin production was observed. Indeed, Walters et al.21 suggested that survival of C. difficile spores within faeces of vancomycin-treated CDI patients may be a factor in the recrudescence of C. difficile spores following cessation of antimicrobial therapy. However, a recent study found that microbiological success, defined as failure to recover C. difficile from faeces, occurred more frequently in vancomycin- versus metronidazole-treated patients.22 Symptomatic recurrence following treatment with vancomycin or metronidazole occurs in 5% to 20% of patients with CDI23 and has been documented to be more commonly observed in metronidazole-treated CDI associated with C. difficile PCR ribotype 027 than with other ribotypes.6 In our prior in vitro gut model study that evaluated vancomycin treatment of clindamycin-induced high-level cytotoxin production due to C. difficile PCR ribotype 001, we did not observe a recurrent episode of proliferation and high-level cytotoxin production after vancomycin was washed out of the system. However, in this prior study, time constraints meant that the post-treatment phase (period F) was shorter than that in the present studies.13 Therefore, the extended recovery period in the present studies may have allowed conditions within the gut model to become more conducive to C. difficile PCR ribotype 027 spore germination, proliferation and high-level cytotoxin production. We plan to evaluate whether there are inter-strain differences with respect to recurrence of toxin production following vancomycin treatment of high-level cytotoxin production in the gut model.
Oritavancin rapidly reduced the numbers of both vegetative and spore forms of C. difficile, despite active peak antimicrobial concentrations 8-fold lower than vancomycin. These active concentrations of oritavancin were 25- and 200-fold greater than the MIC for C. difficile PCR ribotype 027 when measured by agar incorporation and broth macrodilution methods, respectively.19 Despite efforts to remove active oritavancin from culture samples by washing and charcoal adsorption, we were unable to recover C. difficile other than sporadically isolated individual colonies at the limits of detection. Therefore, a marked difference in the effect of oritavancin against C. difficile spores was observed in the present studies in comparison with vancomycin. We reported similar observations when evaluating ramoplanin as a treatment for clindamycin-induced CDI by C. difficile PCR ribotype 001, and we postulated that ramoplanin may possess potential activity against spore germination and/or outgrowth.13 Oritavancin is functionally related to lipid II-binding antimicrobial agents such as nisin, and recent studies indicate that oritavancin may also directly inhibit transglycosylase enzymes involved in polymerizing peptidoglycan.24 Nisin has been demonstrated to inhibit outgrowth of Bacillus spp. and Clostridium sporogenes,25,26 and therefore the potential for similar activity by oritavancin exists. Indeed, we recently observed that inhibition of outgrowth of C. difficile PCR ribotype 027 spores exposed to oritavancin (10 mg/L) persisted after the drug was removed by washing, this effect not being observed for either metronidazole- or vancomycin-exposed spores. No antimicrobial agent tested demonstrated any inhibitory activity against spore germination at supra-MIC concentrations; therefore, we believe that oritavancin may potentially possess activity against C. difficile spore outgrowth (S. Baines, unpublished results). No recrudescence of C. difficile spore germination, proliferation and high-level cytotoxin production was observed in the present study during period F following oritavancin instillation, in direct contrast to vancomycin, even after antimicrobial concentrations declined below the limits of detection. Oritavancin has been demonstrated previously to bind to surfaces, an effect abrogated by 0.002% polysorbate-80.27 Polysorbate-80 was incorporated in the gut model growth medium (0.2% v/v), and thus binding to surfaces within the gut model with resultant continued antimicrobial activity below the limits of bioassay detection is an unlikely explanation for the failure to isolate C. difficile spores for large parts of period F in the present study. Although the decline in C. difficile cytotoxin titres was similar during period E (3 RU) in both oritavancin and vancomycin experiments, it took longer (5 days) for cytotoxin titres to decline to undetectable levels in the oritavancin experiment. Slower washout of cytotoxin in one gut model is puzzling given the similar rates of decline of C. difficile spores during periods A and B as well as identical system flow rates. Whether this observation is an artefact of in vitro experimentation or may potentially be observed in vivo remains to be determined.
In conclusion, both oritavancin and vancomycin effectively eliminated vegetative C. difficile from the gut model, but only oritavancin demonstrated potential activity against spores. Recrudescence of C. difficile PCR ribotype 027 spores was observed following the decline in antibiotic levels to below the limits of detection for vancomycin, but this phenomenon was not seen for oritavancin. These findings suggest that oritavancin may have a therapeutic advantage over vancomycin in terms of anti-spore activity; this could translate into a lower likelihood of symptomatic recurrence associated with the former. Further investigation is required to determine the relevance of these in vitro observations in the clinical setting.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
This work was supported by a research grant from Targanta Therapeutics.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
M. H. W. has received honoraria for consultancy work, financial support to attend meetings and research funding from Astra- Zeneca, Bayer, Genzyme, Nabriva, Novacta, Pfizer and Wyeth. S. D. B. has received financial support to attend meetings from Bayer and Targanta Therapeutics. K. S. has received financial support to attend meetings from Bayer. J. F. has received financial support to attend meetings from Bayer and Wyeth. R. O.: none to declare.
|
Acknowledgements |
|---|
Clindamycin was supplied as a gift from Pfizer, for which we are grateful. We also thank Mrs Margaret Freeman for coordinating the collection of faeces donated for this research and also Professor Eileen Ingham for allowing us access to tissue culture facilities at the University of Leeds. We thank Drs F. F. Arhin, D. Lehoux, G. Moeck and T. R. Parr Jr for their helpful discussions in the preparation of this manuscript and also Mr Alan Noel (BCARE, Bristol, UK) for his helpful advice on the bioassay of glycopeptide antimicrobial agents.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
1 Gerding DN, Muto CA, Owens RC. Treatment of Clostridium difficile infection. Clin Infect Dis (2008) 46(Suppl 1):S32–42.[CrossRef][Web of Science][Medline]
2 Teasley DG, Gerding DN, Olson MM, et al. Prospective randomised trial of metronidazole versus vancomycin for Clostridium-difficile- associated diarrhoea and colitis. Lancet (1983) ii:1043–6.
3 Wenisch C, Parschalk B, Hasenhundl M, et al. Comparison of vancomycin, teicoplanin, metronidazole, and fusidic acid for the treatment of Clostridium difficile-associated diarrhea. Clin Infect Dis (1996) 22:813–8.[Web of Science][Medline]
4
Wilcox MH, Howe R. Diarrhoea caused by Clostridium difficile: response time for treatment with metronidazole and vancomycin. J Antimicrob Chemother (1995) 36:673–9.
5 Musher DM, Aslam S, Logan N, et al. Relatively poor outcome after treatment of Clostridium difficile colitis with metronidazole. Clin Infect Dis (2005) 40:1586–90.[CrossRef][Web of Science][Medline]
6 Pepin J, Alary ME, Valiquette L, et al. Increasing risk of relapse after treatment of Clostridium difficile colitis in Quebec, Canada. Clin Infect Dis (2005) 40:1591–7.[CrossRef][Web of Science][Medline]
7
Freeman J, Baines SD, Saxton K, et al. 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 (2007) 60:83–91.
8 Warny M, Pepin J, Fang A, et al. Toxin production by an emerging strain of Clostridium difficile associated with outbreaks of severe disease in North America and Europe. Lancet (2005) 366:1079–84.[CrossRef][Web of Science][Medline]
9
Baines SD, O'Connor R, Freeman J, et al. Emergence of reduced susceptibility to metronidazole in Clostridium difficile. J Antimicrob Chemother (2008) 62:1046–52.
10
Baines SD, Freeman J, Wilcox MH. Effects of piperacillin/tazobactam on Clostridium difficile growth and toxin production in a human gut model. J Antimicrob Chemother (2005) 55:974–82.
11
Baines SD, Saxton K, Freeman J, et al. Tigecycline does not induce proliferation or cytotoxin production by epidemic Clostridium difficile strains in a human gut model. J Antimicrob Chemother (2006) 58:1062–5.
12
Freeman J, O'Neill FJ, Wilcox MH. 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 (2003) 52:96–102.
13
Freeman J, Baines SD, Jabes D, et al. 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 (2005) 56:717–25.
14 Macfarlane GT, Macfarlane S, Gibson GR. 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 (1998) 35:180–7.[CrossRef][Web of Science][Medline]
15
Brown RB, Martyak SN, Barza M, et al. Penetration of clindamycin phosphate into the abnormal human biliary tract. Ann Intern Med (1976) 84:168–70.
16 Young GP, Ward PB, Bayley N, et al. Antibiotic-associated colitis due to Clostridium difficile: double-blind comparison of vancomycin with bacitracin. Gastroenterology (1985) 89:1038–45.[Web of Science][Medline]
17 Freeman J, Fawley W, Baines S, et al. Measurement of toxin production by. Clostridium difficile. Lancet (2006) 367:982–3. author reply 983–4.
18 Moeck G, Draghi DC, Arhin FF, et al. In vitro activity profile of oritavancin against a broad spectrum of aerobic and anaerobic bacterial pathogens. Abstracts of the Forty-seventh Interscience Conference on Antimicrobial Agents and Chemotherapy, 2007: Chicago, IL. Washington, DC, USA: American Society for Microbiology. Abstract E-1612, p. 204.
19
O'Connor R, Baines SD, Freeman J, et al. In vitro susceptibility of genotypically distinct and clonal Clostridium difficile strains to oritavancin. J Antimicrob Chemother (2008) 62:762–5.
20
Levett PN. Time-dependent killing of Clostridium difficile by metronidazole and vancomycin. J Antimicrob Chemother (1991) 27:55–62.
21
Walters BA, Roberts R, Stafford R, et al. Relapse of antibiotic associated colitis: endogenous persistence of Clostridium difficile during vancomycin therapy. Gut (1983) 24:206–12.
22 Al-Nassir WN, Sethi AK, Nerandzic MM, et al. Comparison of clinical and microbiological response to treatment of Clostridium difficile-associated disease with metronidazole and vancomycin. Clin Infect Dis (2008) 47:56–62.[CrossRef][Web of Science][Medline]
23 Wilcox MH, Spencer RC. Clostridium difficile infection: responses, relapses and re-infections. J Hosp Infect (1992) 22:85–92.[CrossRef][Web of Science][Medline]
24 Wang T, Walker S, Kahne DE. Probing the mechanism of inhibition of bacterial peptidoglycan glycosyltransferases by glycopeptide analogs. Abstracts of the Forty-seventh Interscience Conference on Antimicrobial Agents and Chemotherapy, 2007: Chicago, IL. Washington, DC, USA: American Society for Microbiology. Abstract C1-1474, p. 86.
25
Chan WC, Dodd HM, Horn N, et al. Structure-activity relationships in the peptide antibiotic nisin: role of dehydroalanine 5. Appl Environ Microbiol (1996) 62:2966–9.
26
Rayman MK, Aris B, Hurst A. Nisin: a possible alternative or adjunct to nitrite in the preservation of meats. Appl Environ Microbiol (1981) 41:375–80.
27
Arhin FF, Sarmiento I, Belley A, et al. Effect of polysorbate 80 on oritavancin binding to plastic surfaces: implications for susceptibility testing. Antimicrob Agents Chemother (2008) 52:1597–603.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  O. M. Williams and R. C. Spencer The management of Clostridium difficile infection Br. Med. Bull., September 1, 2009; 91(1): 87 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Baines, J. Freeman, and M. H. Wilcox Tolevamer Is Not Efficacious in the Neutralization of Cytotoxin in a Human Gut Model of Clostridium difficile Infection Antimicrob. Agents Chemother., May 1, 2009; 53(5): 2202 - 2204. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||