Abstract
A total of 65 toxigenic Clostridium difficile strains isolated from patients with antibiotic-associated diarrhea were tested for susceptibility to avilamycin, flavomycin, monensin, and salinomycin. Except for flavomycin the substances showed in vitro efficacy comparable to reports of the currently most commonly used drugs for treatment of C. difficile. This indicates that these old compounds may be useful for the treatment of C. difficile infections in man and perhaps for other bacterial causes of diarrhea.
Introduction
C. difficile infections have recently increased in incidence and severity in many regions of North America and many countries in Europe.13,20 C. difficile is intrinsically resistant to many antimicrobial agents, and acquired resistance to the few therapeutically efficient drugs has been rapidly increasing. 18 The current treatment regime is associated with treatment failure in some cases and relapses occur in ∼20%–30% of all clinical cases. 3 This has increased the interest in searching for novel and more efficient treatment options and the development of new antibiotics.5,13
A number of antimicrobial agents, never approved for human therapy, have for several decades been used for nontherapeutic purposes or to control coccidia infections in food animals.1,2 Some of these compounds have shown good in vitro activity against Clostridium perfringens9,22 and have been also able to control infections with this bacterium in poultry19,21 and swine. 14 These antimicrobials are included in the feed and are poorly absorbed in the gut. Thus, we speculated whether these compounds might have activity against C. difficile-caused gastrointestinal infections in humans.
This study was conducted to evaluate the in vitro susceptibility to avilamycin, flavomycin, monensin, and salinomycin against C. difficile isolated from infections in humans as a first step to determine their potential value as human therapeutic drugs.
Materials and Methods
Bacterial isolates
A total of 65 clinical isolates of C. difficile were selected from cases of antibiotic-associated diarrhea at Rigshospitalet in Denmark. The isolates were cultured from feces on cycloserine–cefoxitin–fructose egg yolk agar (Statens Serum Institute) and identified as C. difficile by usual routine fermentation tests and demonstration of volatile fatty acids by gas–liquid chromatography according to the manual of the Virginia Polytechnic Institute. 11 Cytotoxin was demonstrated in a McCoy cell assay established in the laboratory at Rigshospitalet. 12
Antimicrobial susceptibility testing
All isolates were examined for their minimum inhibitory concentrations (MICs) using twofold dilutions on Brucella agar supplemented with 5% laked sheep blood (Statens Serum Institute), 5 mg/L hemin (Sigma-Aldrich), and 1 mg/L vitamin K (Sigma-Aldrich) according to CLSI standards. 8 Isolates were tested to the following concentrations of avilamycin (Elanco Animal Health), flavomycin (Hoechst GmbH), monensin (Elanco Animal Health), and salinomycin (Huvepharma): 0.03, 0.06, 0.125, 0.25, 0.5, 1, 2, 4, 8, and 16 mg/L. Avilamycin and monensin were dissolved in acetone, salinomycin in 99% ethanol, and flavomycin directly in water before being added to the agar. As quality control strain, C. difficile ATTC 17857 was included on all agar plates. The agar plates were incubated under anaerobic conditions at 36°C for 42–48 hours.
Results
The MICs are shown in Table 1. MIC values for avilamycin ranged from ≤0.03 to 0.25 mg/L; two isolates, however, had MIC values of 2 mg/L. MIC values for flavomycin ranged from 0.25 to 8 mg/L, for monensin from ≤0.03 to 1 mg/L, and for salinomycin from ≤0.03 to 0.5 mg/L.
Values in this row indicate MICs (mg/L).
MIC of Clostridium difficile ATCC 17857.
MIC, minimum inhibitory concentration.
Discussion
Infections caused by C. difficile are increasing in prevalence and severity. Treatment of these infections is increasingly difficult because of the development of resistance toward fluoroquinolones and rifampicin and reduced clinical response to metronidazole and vancomycin. The use of vancomycin is further problematic because of the risk of selection for vancomycin-resistant enterococci and staphylococci.
We tested the in vitro susceptibility of 65 C. difficile isolates toward avilamycin, flavomycin, monensin, and salinomycin. All four compounds have been extensively used for decades in food animals and no toxicological reactions, to our knowledge, have been ever recorded. Avilamycin, an oligosaccharide, and another closely related antimicrobial agent, everninomicin, were previously under development for therapeutic use in humans. Avilamycin has been originally shown to bind to the 30S subunit of ribosome and thereby interfere with the protein synthesis function. 23 Everninomycin has, however, been shown to bind exclusively to the 50S subunit. 17 In enterococci, resistance to both compounds is mediated by either point mutations in the L16 50S subunit ribosomal protein 4 or a methyltransferase acting on the 50S subunit. 16 Flavomycin is a phosphoglycopeptide that inhibits peptidoglycan synthesis by inhibiting peptidoglycan polymerases. 7 Several bacterial species exhibit natural resistance to flavomycin, but so far acquired resistance has not been confirmed. 7 Monensin and salinomycin are both ionophores and interfere with the natural ion transport system of both prokaryotic and eukaryotic cells. Staphylococcal and enterococcal isolates with reduced susceptibility has been reported, 3 but the mechanism has never been elucidated. To our knowledge, no substances belonging to these classes have been ever developed for human usage.
All four substances tested had activity against C. difficile; however, flavomycin had relatively high MIC values, whereas avilamycin, monensin, and salinomycin had MIC values comparable to or slightly lower than those reported for metronidazole and vancomycin. 10
These results indicate that some of these old compounds, which for decades have been used for other purposes, might find application in human medicine for the treatment of C. difficile diarrhea and perhaps other chronic diarrhea diseases. Further studies are, however, necessary to determine the therapeutic potential and also to explore whether some negative effects might be associated with a future use of one or more of these compounds in human medicine.
Footnotes
Acknowledgments
The authors are grateful to Jacob Dyring Jensen and Anette Gregersen for technical assistance. This study was funded by a grant (274-05-0117) from the Danish Research Agency.
Disclosure Statement
No competing financial interests exist.
