Background
Clostridium difficile infection (CDI) is a major healthcare concern, especially in the European Union and North America [
1]. In recent years, the formation of biofilms by
C. difficile, which could contribute to antibiotic resistance and CDI treatment failures has been highlighted [
2,
3]. Indeed, increased resistance of
C. difficile to vancomycin and metronidazole and recurrence of disease due to treatment failure have already been reported [
4]. Thus, there is a growing need to find alternative solutions to combat this pathogen. Two of the most problematic
C. difficile biofilm producers are strains 20291 R027 and strain 630 [
5]. Recent studies have provided insights into the mechanisms by which
C. difficile forms biofilms [
2,
3,
5]. Dawson et al. demonstrated that the ability of
C. difficile to produce biofilms involves the formation of several layers of bacteria in a complex matrix which is composed of polysaccharides, proteins and DNA [
2]. Furthermore, Dapa et al. showed that a quorum sensing regulator, LuxS, flagellae and Cwp84 are all needed for optimal
C. difficile biofilm formation [
5]. The authors of the study also found that there could be a potential link between the ability to form biofilms and sporulation, as
C. difficile strains with a mutation in the gene encoding SpoOA, which is a transcription factor needed for sporulation, lacked the ability to form biofilms [
5].
In this study, we provide insights into the potencies of various antimicrobials and antimicrobial combinations, when combined at relatively low concentrations against various C. difficile biofilms and compare and contrast the effects of these combinations against planktonic cells of the same strains. We note that thuricin CD is effective at targeting the biofilms of R027, R106 and DPC6350 at relatively low concentrations, while paired combinations of this bacteriocin with the antibiotics teicoplanin, tigecycline, vancomycin or rifampicin are potent at attenuating the viability of adherent biofilms formed by C. difficile strains 20291 R027 and DPC6350.
Discussion
The aim of the study was to evaluate the efficacy of one bacteriocin (thuricin CD) and 5 antibiotics (teicoplanin, tigecycline, vancomycin, rifampicin and nitazoxanide) when used independently and when combined at low concentrations together against C. difficile biofilms, as well as against planktonic cells of the same strains. As antibiotics are expensive to use in clinical settings, we attempted to find alternative antimicrobial combinations which work well together at relatively low concentrations (2x MIC each), in an attempt to diminish the viability of such biofilms. Furthermore, since biofilms often show resistance to antibiotics used on their own, our objective was to search for effective antimicrobial combinations that could potentiate each other’s effects to circumvent these problems of antimicrobial resistance. We postulated that using antimicrobial combinations at low concentrations would provide insights into effective therapeutic options with a view to targeting C. difficile biofilms.
It was particularly noteworthy there were differences amongst
C. difficile strains in their abilities to form strongly adherent biofilms and variations in the sensitivities of biofilms of these different
C. difficile strains that we tested to the different antibiotics used either independently or in paired combinations. Various factors are likely to be involved in governing the strengths of biofilms formed, including the presence of glucose, sporulation and even sub-lethal doses of antibiotics [
3,
5,
19,
20]. This potential stimulation of
C. difficile biofilm formation by sub-lethal concentrations of antibiotics may occur in treatment regimens involving pulsed or tapered dosing of antibiotics against CDI, whereby antibiotics are likely to be present at sub-lethal doses and thus can contribute to treatment failure [
3,
19].
Clostridium difficile R20291 R027 and DPC6350 were the strongest biofilm formers amongst the strains we tested in this study.
C. difficile R20291 R027 has also previously been shown to be a relatively strong biofilm former in other studies [
2,
3,
5]. A stronger ability to form adherent biofilms may be a contributing factor involved in greater degrees of colonization of this epidemic-associated strain in the gut [
3,
5].
With respect to the antimicrobials used independently in our study, the glycopeptide vancomycin was found to be ineffective against biofilms of four of the five strains we assessed. Similar to our findings, Dapa and co-workers had also previously reported that
C. difficile biofilms display attenuated sensitivity to vancomycin [
5
]. Indeed, biofilms are generally more resistant to antibiotics and studies by Semenyuk et al. had reported that
C. difficile biofilms also exhibit increased resistance to the antibiotic metronidazole, relative to
C. difficile planktonic cells grown in liquid broth media [
21]. The study by Dapa et al. indicates that the structure and organisation of the matrix in
C. difficile biofilms is likely to play a role in mediating resistance to antibiotics, as such matrices can prevent penetration of the antimicrobial reaching the bacteria within [
5,
22]. In addition, the physiological state of the bacteria present in the biofilm as well as the presence of persister cells are likely to have a role in attenuated sensitivity to antibiotics [
23]. Interestingly, the semi-synthetic glycopeptide, teicoplanin, was found to be effective only against biofilms of
C. difficile strain DPC6350, while being ineffective against the other strains in our study. Thus, it appears that the glycopeptide group of antibiotics can be ineffective at targeting biofilms of a number of
C. difficile strains. The thiazolide antimicrobial, nitazoxanide, was effective at reducing the biofilm viability of three of the five strains we tested in this study and it was determined that tigecycline (a member of the glycylcycline group of antibiotics) was potent against 20291 R027 and DPC6350, while being ineffective against biofilms of strains TL178 R002, Liv022 R106 and VPI10463. It may be the case that tigecycline has a greater ability to target strongly-adherent biofilms, such as those formed by R027, DPC6350, and is ineffective against weakly-adherent biofilms. Similar to tigecycline, rifampicin (belonging to the rifamycin group of antibiotics) also exhibited more potency against the stronger biofilms of R027 and DPC6350, while only being marginally effective against Liv022 R106 biofilms and ineffective against the weak biofilms of TL178 R002 and VPI10463. It is plausible that altered growth rates of cells within these
C. difficile biofilms caused by genetic mutations leads to attenuated sensitivity of certain strains to teicoplanin, tigecycline and rifampicin. Thus, it is clear that biofilms of different
C. difficile strains display varying degrees of sensitivity to the different antibiotics when used on their own but the precise mechanisms of sensitivity remain largely unknown.
With respect to antimicrobial combinations used in this study, variations in sensitivities of biofilms of different strains to the antimicrobial combinations were also apparent. The most striking difference was that while thuricin CD combined with either teicoplanin, tigecycline, vancomycin or rifampicin was effective against stronger biofilms of strains 20291 R027 and DPC6350, such combinations were largely ineffective against the relatively weaker biofilms of TL178 R002, Liv022 R106 and VPI10463. Interestingly, 2x thuricin CD-2x nitazoxanide combinations were particularly ineffective against R027 biofilms. It may be the case that the addition of thuricin CD to nitazoxanide at low concentrations somehow interferes with the ability of nitazoxanide to target the pyruvate:ferredoxin oxidoreductase system in
C. difficile R027 at such concentrations. Alternatively, it could be the case that nitazoxanide prevents thuricin CD from reaching a target receptor in
C. difficile, thus leading to a lack of efficacy against
C. difficile R027 biofilms when combined together, resulting in apparent antagonistic effects at these concentrations. Furthermore, while higher concentrations of these antimicrobials combined (8x MIC thuricin CD with 8x MIC nitazoxanide) were effective against biofilms of R027, DPC6350 and Liv022 R106, this combination at higher concentrations was still ineffective against the weakest biofilms of TL178 R002 and VPI10463. Thus, it is apparent that there are strain-specific as well as concentration-dependent variations with regards to sensitivities of
C. difficile biofilms to different antimicrobials and antimicrobial combinations. Since the XTT-menadione reduction assay determines the level of viability of
C. difficile biofilms, it provides insights into the amount of a biofilm that is still metabolically active, subsequent to antimicrobial challenges and thus is likely to provide a realistic picture regarding the potency of antimicrobials against biofilms in vivo [
24‐
27]. The precise mechanism of action involved in thuricin CD-nitazoxanide against
C. difficile planktonic cells and biofilms remains unclear however and merits further investigation in future studies.
With respect to utilising FIC values against
C. difficile planktonic cells as a predictor of effective combinations against biofilms of the same strains, the partial synergistic and additive effects seen with thuricin CD-teicoplanin, thuricin CD-rifampicin and thuricin CD-tigecycline combinations against
C. difficile R027 planktonic cells are consistent with additive effects against R027 biofilms (Fig.
1; Table
2). Combinations of the sactibiotic thuricin CD with vancomycin, rifampicin, tigecycline and teicoplanin may enable one of the antimicrobials to gain access through the complex matrix and allow the antimicrobial to exert its killing effect on the R027 biofilm, thereby leading to a reduction in biofilm viability. In contrast, there doesn’t appear to be a correlation between the additive effects (FIC 1.0) obtained with thuricin CD-vancomycin combinations against planktonic cells of Liv022 R106 and the combination against biofilms of the same strain. In addition, even though thuricin CD combined with rifampicin resulted in indifferent effects (FIC 1.01–2) against planktonic cells of DPC6350, it is apparent that this combination displayed ameliorated potency against DPC6350 biofilms, compared to either 2x MIC thuricin CD or 2x MIC rifampicin. Finally, while thuricin CD-nitazoxanide combinations resulted in additive effects (FIC 1.0) against planktonic cells of DPC6350, combinations of 2x MIC thuricin CD with 2x MIC nitazoxanide did not exhibit enhanced activity when compared to 2x MIC thuricin CD used independently against biofilms of this strain (P > 0.05). Similar trends were noted against planktonic cells and biofilms of strains TL178 R002 and VPI10463. The variations between planktonic cells and biofilms could be due to differences in the mechanisms of action of thuricin CD-nitazoxanide combinations against planktonic cells, as opposed to the two antimicrobials targeting a complex multi-layered matrix that exists in a
C. difficile biofilm.
Conclusions
In conclusion, this is the first study assessing the antimicrobial effect of a sactibiotic bacteriocin and the antibiotics nitazoxanide, tigecycline and teicoplanin against
C. difficile biofilms. Overall, it is encouraging to note from this study that a number of
C. difficile strains have a relatively weak ability to form strongly adherent biofilms, in comparison to other pathogens such as
Streptococcus mutans,
Staphylococcus aureus and
Pseudomonas aeruginosa, which are notorious biofilm producers [
28‐
30]. It is also encouraging to note that four of the five antimicrobial combinations we tested (with the exception of thuricin CD-nitazoxanide) appeared to be highly effective against two of the stronger
C. difficile biofilm formers, 20291 R027 and DPC6350. It is apparent from this study that, not only are there variations between different
C. difficile strains in their abilities to form biofilms, there are also variations in terms of sensitivities of biofilms of different strains to several antimicrobial treatment options. Such variations seen with regards to the potencies of the different antimicrobials/antimicrobial combinations against these different strains may be due to the relative strengths of the biofilms that the antimicrobials target. Furthermore, the mechanism of action involved in effective antimicrobial combinatorial therapy is likely to be different against planktonic cells, compared to cells in biofilms. While it is clear that much remains to be elucidated with respect to the precise mechanisms governing antimicrobial sensitivity and antimicrobial resistance within
C. difficile biofilms, overall, this study could form the basis for the development of successful antimicrobial combination therapy strategies, with a view to targeting
C. difficile biofilms. The findings presented in this study could have implications with regards to reducing the recurrence rates of CDI, particularly in cases where a strong
C. difficile biofilm former is contributing to increased recurrence rates.
Authors’ contributions
HM, MRC, CH, PDC and RPR designed the study and wrote the manuscript. HM conducted the experiments, interpreted and analysed the data. All authors read and approved the final manuscript.