Gut solutions to a gut problem: bacteriocins, probiotics and bacteriophage for control of Clostridium difficile infection

Transcription

1 Journal of Medical Microbiology (2013), 62, DOI /jmm Gut solutions to a gut problem: bacteriocins, probiotics and bacteriophage for control of Clostridium difficile infection Mary C. Rea, 1,2 Debebe Alemayehu, 1,2 R. Paul Ross 1,2 and Colin Hill 2,3 Correspondence R. Paul Ross paul.ross@teagasc.ie 1 Teagasc Food Research Centre, Moorepark, Fermoy, County Cork, Ireland 2 Alimentary Pharmabiotic Centre, University College, Cork, Ireland 3 Department of Microbiology, University College, Cork, Ireland Received 13 February 2013 Accepted 16 May 2013 Clostridium difficile infection (CDI) is a major cause of morbidity and mortality among hospitalized patients and imposes a considerable financial burden on health service providers in both Europe and the USA. The incidence of CDI has dramatically increased in recent years, partly due to the emergence of a number of hypervirulent strains. The most commonly documented risk factors associated with CDIs are antibiotic usage leading to alterations of the gut microbiota, age.65 years and long-term hospital stay. Since standard therapies for antibiotic-associated diarrhoea and CDI have limited efficacy, there is now an urgent need for alternative therapeutics. In this review, we outline the current state of play with regard to the potential of gut-derived bacteriocins, probiotics and phage to act as antimicrobial agents against CDI in the human gut. INTRODUCTION Since the introduction of antibiotic-based antimicrobial therapies in the middle of the 20th century, antibioticassociated diarrhoea (AAD) has emerged as an increasing problem. Related to this, one of the more common causes of nosocomially acquired intestinal infections is Clostridium difficile, the aetiological agent responsible for virtually all cases of pseudomembranous colitis and up to % of cases of AAD or hospital-acquired diarrhoea (Pochapin, 2000; Bartlett, 2002). Mortality due to C. difficile (2 42 %) doubled from 1999 to 2004 and continued to rise until 2007, after which a decrease was seen in the UK (Wiegand et al., 2012). However, it is still a significant cause of AAD, posing a considerable financial burden on health service providers in both Europe and the USA (Ghantoji et al., 2010; Wiegand et al., 2012). It is now well recognized that C. difficile infection (CDI) is intrinsically linked to the use of broadspectrum antibiotics, especially clindamycin, cephalosporins, penicillins and, more recently, fluoroquinoles (Warren & Guerrant, 2011). The argument that it is the disruption of the colonic flora that predisposes an individual to CDI has been reinforced by the ability of normal faecal flora to reverse the symptoms of CDI in very refractory cases when applied via nasogastric feeding/colonoscope (Aas et al., 2003; Rohlke et al., 2010; Grehan et al., 2010). Treatment of CDI almost exclusively involves the use of the broadspectrum antibiotics vancomycin and metronidazole and, while the resistance to these antibiotics is low, there is a Abbreviations: AAD, antibiotic-associated diarrhoea; CDI, Clostridium difficile infection; GIT, gastrointestinal tract. trend towards decreased susceptibility and an increase in recurrences of infection. More recently, a novel oral macrolide antibiotic, fidaxomicin, with a narrow inhibition spectrum was approved in Europe and the USA for the treatment of CDI (Johnson & Wilcox, 2012). Owing to the frequent use of broad-spectrum antibiotics, the spread of antibiotic resistance, particularly in the hospital environment, is now posing significant problems for healthcare professionals. The emergence of pathogens with resistance to multiple antibiotics is likely to be a major challenge of the 21st century with antibiotics formally used routinely to treat pathogens, such as Staphylococcus aureus, Enterococcus faecalis, Mycobacterium tuberculosis and C. difficile, no longer being effective (Arias & Murray, 2009; Carlet, 2012; Riley et al., 2012). Few new antibiotics have emerged in the recent past due, in part, to the high cost of production and the associated financial risks of the development of such antibiotics (Cotter et al., 2013). Currently, vancomycin, once termed the antibiotic of last resort, and metronidazole are the main antimicrobials used to treat CDI. However, while resistance to these antimicrobials is low, there is a trend towards decreased susceptibility, possibly resulting in an increase in recurrences of infection (Baines et al., 2008; Brazier et al., 2008; Huang et al., 2009; Freeman et al., 2010). It is universally accepted that broad-spectrum antibiotic therapy results in perturbation of the gut flora, which predisposes individuals to CDI. The elderly are most at risk from CDI with carriage rates of C. difficile in hospitals and care homes for the elderly ranging from 2 % to 20 % (Rupnik et al., 2009; Ryan G 2013 SGM Printed in Great Britain 1369

2 M. C. Rea and others et al., 2010; Rea et al., 2012). However, while CDI is a disease induced by broad-spectrum antibiotic usage, broad-spectrum antimicrobials are also used as therapeutics to treat CDI, which in turn may cause further dysbiosis in the distal colon. Indeed, in a murine study by Wlodarska et al. (2011), metronidazole was shown to reduce the depth of the mucus layer, resulting in a reduction of its barrier function, thereby predisposing the host to Citrobacter rodentium-induced colitis. The changes in the intestinal microbiota as a result of antibiotic treatment disrupted intestinal homeostasis and the integrity of intestinal defences, which normally protect the host against invading pathogens and intestinal inflammation (Wlodarska et al., 2011). It has also been demonstrated in mice that combinations of neomycin, vancomycin and metronidazole can induce infection by vancomycin-resistant enterococci (Brandl et al., 2008). In this article, we explore novel therapies that may have potential as alternative/adjunct therapies for the treatment of CDI, namely bacteriocins, probiotics and bacteriophage therapy. BACTERIOCINS That antibiotics used to treat CDI can cause collateral damage to the gut microbiota prompts the question of what avenues are now open for the development of therapies for the treatment of CDIs. One option currently being explored is the use of bacteriocins, small ribosomally synthesized antimicrobial peptides that have antimicrobial activity against closely related species (narrow spectrum) or across genera (broad spectrum). For a bacteriocin to play a role as an alternative therapeutic for CDI, it would need to be at least as effective as currently used antimicrobials, but ideally would combine high specificity for C. difficile with a narrow spectrum of inhibition, resulting in minimal collateral damage to the gut flora. In addition, resistance development by C. difficile and related species should be minimal, while the bacteriocin should also be non-toxic for the host. To date, a number of studies have shown in vitro activity of a number of bacteriocins active against C. difficile (Table 1). The bacteriocins lacticin 3147 and nisin (and bioengineered nisin variants), which are lantibiotics produced by strains of Lactococcus lactis, have been shown to be effective in vitro against clinically relevant C. difficile strains. However, they both have the disadvantage of having a broad spectrum of activity, especially towards Gram-positive bacteria (Bartoloni et al., 2004; Rea et al., 2007; Field et al., 2010). Additionally, we have shown in ex vivo studies using a model of the distal colon that the broad-spectrum lacticin 3147 causes a profound proportional shift in the microbiota from Firmicutes to Proteobacteria (Fig. 1) (Rea et al., 2011b). As nisin is also active against a broad range of Gram-positive bacteria, it is likely that it too would have a similar effect on the gut microbiota. The semi-synthetic broad-spectrum thiopeptide LFF571 inhibits C. difficile and demonstrates lower inhibitory activity against Lactobacillus and Bidifobacterium species than against other Gram-positive bacteria (Citron et al., 2012). A promising semi-synthetic derivative of the lantibiotic actagardine A with a narrow spectrum of inhibition has also been synthesized and inhibits a range of Gram-positive pathogens, including C. difficile (Boakes et al., 2012). However, in an attempt to overcome the problems associated with broad-spectrum antimicrobials for the treatment of CDI, our research group embarked on a search within the gut microbiota for bacteria that produce antimicrobials that specifically target C. difficile. Faecal samples were screened for the presence of spore-forming strains that could inhibit C. difficile in an overlay assay. This led to the discovery of the sactibiotic thuricin CD, produced by a strain of Bacillus thuringiensis. Thuricin CD exhibits MICs comparable to vancomycin and metronidazole when tested in vitro against clinically significant strains, including ribotypes 160, 001 and 027 (Rea et al., 2010, 2011a, b). More significantly, in a model of the distal colon that had been spiked with C. difficile, we showed that thuricin CD did not significantly impact on the wider gut microbiota, unlike vancomycin and metronidazole for which there was a major increase in abundance of Enterobacteriaceae at the expense of other families. We also demonstrated that thuricin is very effective in an environment containing a high microbial load as it was able to reduce the C. difficile population in the faecal environment ~100 fold (Fig. 2) (Rea et al., 2011b). For peptide antimicrobials such as thuricin CD to be effective as a therapeutic many challenges lie ahead, not least in terms of understanding their mode of action, scaling up production and developing stable delivery mechanisms to ensure delivery of active peptides to the distal colon. However, given the pressures currently on the available therapeutics the potential for the commercialization of bacteriocins or their bioengineered derivatives as targeted therapeutics for treatment of gastrointestinal infections such as C. difficile is very promising. PROBIOTICS The evidence for the beneficial role of the gut microbiota in health and disease is mounting. In this respect, it is recognized that perturbations of the intestinal microflora may, in turn, lead to a range of diseases, including inflammatory immune diseases such as Crohn s disease, insulin resistance, obesity, cancer and infectious diseases such as CDI (Scanlan et al., 2006; Davis & Milner, 2009; Shanahan, 2011; Rea et al., 2012). The fetal gut is sterile and is initially colonized during birth, the microbiota depending largely on the mode of delivery and the hygiene of the surrounding environment, which subsequently develops into the complex ecosystem of the adult gastrointestinal tract (GIT) (Fouhy et al., 2012). In adults, the human microbiota contain ~10 14 bacteria comprising between 1000 and 1150 different species (Qin et al., 2010) 1370 Journal of Medical Microbiology 62

3 Biological solutions for the control of C. difficile infection Table 1. Referenced antimicrobial peptides with activity against C. difficile Bacteriocin Description Spectrum of inhibition Producing strain Reference Nisin A Lantibiotic Broad Lactococcus lactis Bartoloni et al. (2004) Gram-positive Nisin V Lantibiotic, bioengineered Broad Lactococcus lactis Field et al. (2010) derivative of nisin A Gram-positive Thuricin CD Sactibiotic Narrow Bacillus thuringiensis DPC 6431 Rea et al. (2010, 2011b) Lacticin Lantibiotic Broad Lactococcus lactis Rea et al. (2007) Gram-positive LFF571 Thiopeptide Broad Planobispora rosea Citron et al. (2012) Gram-positive ATCC Actagardine A Semi-synthetic derivative of Narrow Actinoplanes liguriea Boakes et al. (2012) (DAB) lantibiotic of actagardine ATCC that exist in a mutualistic relationship with the host and aid in digestion, modulation of the immune system and manufacture of essential nutrients. Disturbances of the gut flora through antibiotic usage can have a profound effect on the health of the host, as is illustrated with CDI. Post-antibiotic treatment, the gut microbiota are altered and can no longer outcompete intestinal pathogens, allowing C. difficile to proliferate and produce toxins. In a study of C. difficile Log c.f.u. ml % 12 % 1.30 % % 0.21 % % 6.50 % 3.80 % 2.70 % carriage in an elderly population (Fig. 3) the gut flora of two patients with CDI, from whom ribotype 027 was isolated, showed a much reduced microbial diversity when compared with those of an asymptomatic carrier of the same ribotype (Rea et al., 2012). Both CDI patients had a previous history of antibiotic use, in contrast to the asymptomatic carrier, which suggests that the gut microbiota exerted a protective effect on the subject who did not develop CDI (Rea et al., 2012). Researchers have investigated whether the dysbiosis that occurs as a result of antibiotic usage can be prevented or reduced by the consumption of probiotics and have provided % 68 % % Log c.f.u. ml Bacteroidetes Proteobacteria Firmicutes Actinobacteria Bacteriodetes Spirochaetes Firmicutes Lentisphaerae 10 6 Proteobacteria Ternicutes Actinobacteria *** T 0 Control (T 12 ) Lacticin 3147 (T 12 ) 0 0 Control Van (90 µm) Met (90 µm) Thuricin (90 µm) Fig. 1. Reduction in phylum level diversity in a model of the distal colon when the model is treated with lacticin 3147 compared with the control (pie charts). The results are expressed as a percentage of the total population of assignable tags. The corresponding bar charts show the decrease in C. difficile numbers in the presence of lacticin T 0 5time zero; T 12 5after 12 h incubation in the absence (control) or presence of lacticin [Redrawn from Rea et al. (2010).] Fig. 2. The effects of vancomycin (Van), metronidazole (Met) and thuricin CD (all at 90 mm) on the viability of C. difficile in a distal colon model (bar chart). The corresponding pie charts show the effect of vancomycin, metronidazole and thuricin CD on the phylum level diversity in the distal colon. The antibiotic-treated model shows a reduction in microbial diversity compared to the control or thuricin CD. [Redrawn from Rea et al. (2011b).]

4 M. C. Rea and others Ruminococcus 9.0 % Parabacteroides 15 % (a) (b) (c) Bacteroides Bacteroides Enterococcus 19 % 0.3 % 2.0 % Bifidobacterium 3.0 % Clostridium 6.0 % Bacteroides 31 % Clostridium 1.0 % Faecalibacterium 18 % Eubacterium 10 % Parabacteroides 97 % Parabacteroides 62 % Enterococcus 3.0 % Fig. 3. Genus level diversity of gut communities of three subjects carrying C. difficile ribotype 027. (a) Asymptomatic carrier. (b) and (c) Patients with active CDI at the time of sampling. The results are expressed as a percentage of the total population of assigned tags. [Redrawn from Rea et al. (2012).] evidence that probiotics have potential as prophylactics in the prevention of CDI. Probiotics are defined as live microorganisms which when consumed in adequate amounts confer a health benefit on the host (Pineiro & Stanton, 2007). Probiotics may play a role in improving epithelial barrier function, secretion of inhibitory substances (such as bacteriocins and hydrogen peroxide), immunomodulation, inhibition of expression of virulence factors and competitive exclusion possibly through colonization resistance (McFarland, 2000; Corr et al., 2009). Most probiotics are lactic acid bacteria, belonging mainly to the genera Lactobacillus and Bifidobacterium. However,in the case of CDI, Saccharomyces boulardii has also been suggested as a potential probiotic. The mechanism of action of most probiotics in the context of prevention of gastrointestinal infection is poorly understood, albeit that some evidence exists for a limited number of probiotic strains. For example, Corr et al. (2007) showed that mice were protected against infection by the intestinal pathogen Listeria monocytogenes when orally fed a strain of Lb. salivarius producing the bacteriocin APB118. That the inhibition of infection was due to the bacteriocin-producing capability of the strain in the GIT was shown through the failure of an isogenic nonbacteriocin-producing variant of the same strain to prevent infection by L. monocytogenes. In addition, when the infecting L. monocytogenes strain was made immune to APB118 no protection against the infection was observed. Interestingly, this strain also was effective in preventing Salmonella infection, albeit through a different mechanism, as the non-producing variant was also effective in preventing infection (Corr et al., 2007). Three possible mechanisms for the probiotic effect of bacteriocin production in vivo have been proposed. These include the bacteriocin (1) acting as colonizing peptides by successfully allowing the probiotic strain to compete with the resident flora, (2) killing peptides, resulting in the eliminating of the pathogen, or (3) acting as signalling peptides through recruitment of other bacteria in the gut or the immune system to fight and eliminate the infectious organism (Dobson et al., 2012). Bacteriocinproducing probiotic cultures may therefore function by facilitating the introduction of a producer into an established niche, directly inhibiting the invasion of competing strains or pathogens into an established community, or modulating the composition of the microbiota and the host immune system. Therefore, the identification of probiotic strains capable of producing bacteriocins within the GIT targeting specific pathogens, such as C. difficile, may be one avenue in the search for probiotics for CDI. Currently, the efficacy of probiotics for the treatment or prevention of CDI is controversial to say the least. In 2008, a Cochrane review was published on the use of probiotics for the treatment of C. difficile-associated colitis (Pillai & Nelson, 2008). The inclusion criteria for this analysis were randomized prospective studies using probiotics alone or in conjunction with conventional antibiotics for the treatment of documented C. difficile colitis. Only four studies from articles published from 1966 to 2007 were deemed eligible for this study (Table 2). Criticisms of the studies included small sample size, considerable heterogeneity relating to antibiotic usage and initial disease state. The limitations were such that the authors concluded there was insufficient evidence for the use of probiotics as adjuncts to conventional antibiotic therapy and no evidence to support the use of probiotics alone as therapeutics for CDI. In an interesting study (n557 test group, n556 placebo group) Hickson et al. (2007) showed that the consumption of a probiotic drink containing Lb. casei, Lb. bulgaricus and Streptococcus thermophilus twice a day during the course of antibiotic treatment and for 1 week following treatment reduced the incidence of AAD and CDI in hospital patients. In the largest study (255 subjects) to date, a combination of Lb. acidophilus and Lb. casei reduced the risk of hospitalized patients on antibiotics contracting CDI (Gao et al., 2010). Other reviewers have concluded that there may be some merit in the use of probiotics for the treatment of CDI, but again point to 1372 Journal of Medical Microbiology 62

5 Biological solutions for the control of C. difficile infection Table 2. Published studies used by Pillai & Nelson (2008) for the systematic review on the use of probiotics for treatment of CDI Study Probiotic c.f.u. g 1 No. subjects Additional treatment Lawrence et al. (2005) Lactobacillus rhamnsosus GG twice daily during antibiotic therapy and 21 days post-cessation 15 Vancomycin or metronidazole (concentration at discretion of clinician) McFarland et al. Saccharomyces boulardii daily for 4 weeks 124 Vancomycin or metronidazole (1994) Surawicz et al. (2000) Saccharomyces boulardii 1 g day 21 for 30 days 168 Vancomycin (high dose: 2 g day 21 or low dose: 500 mg day 21 )or metronidazole Wullt et al. (2003) Lactobacillus plantarum 229v for 38 days 21 Metronidazole 400 mg three times a day small sample sizes and inadequate study design, indicating the need for large-scale well-designed clinical trials across a number of sites (McFarland, 2009; Johnson et al., 2012). One very critical review by Miller (2009) contends that probiotics play no role at all in the prevention or therapy of CDI and advise against the use of Sc. boulardii as a probiotic due to the danger of fungaemia not only in the patient being treated but also in patients in close proximity in intensive care facilities (McFarland, 2009; Miller, 2009; Johnson et al., 2012). However, the question needs to be posed as to why particular probiotic strains were chosen for these trials. Most likely, they were chosen without any sound rationale for their selection other than that these strains were easy to produce or were commercially available. There is no evidence that in vivo trials were carried with the probiotics selected in animal models of CDI. As stated previously, current probiotics for human use belong almost exclusively to the genera Lactobacillus and Bifidobacterium as they are considered to contribute significantly to human health and have the added advantage of a long history of safe use, thereby overcoming many of the hurdles that may exist for other genera when included in foods for human nutrition. However, it is now recognized that probiotic properties being assigned to these genera are not only species specific but are also much more likely to be strain specific. The importance of strain selection was highlighted in a recent murine study investigating the effect of two strains of B. breve, NCIMB and DPC 6330, on gut microbiota composition which showed that only B. breve DPC 6630 modulated the gut microflora (Wall et al., 2012). The future of probiotic design for CDI will involve a deeper understanding of the role of the commensal flora of the gut and how alterations in the host microbiota relationship as a result of antibiotic therapy impact on the health of the host. It is unlikely that a single probiotic strain of Lactobacillus or Bifidobacterium will be sufficient to redress the imbalance of gut microflora, and a more effective approach may be to use a cocktail of strains from a number of genera. This will pose a challenge in terms of identification of the most suitable mixes and the technological hurdles that will have to be overcome to produce stable probiotic mixes containing anaerobic organisms. PHAGE THERAPY Given the rise in antibiotic resistance, non-antibiotic therapies such as phage therapy are now being revisited as a possible therapeutic. Advantages of using phage for the control of infectious agents include (1) their ability to attack specifically target cells and replicate inside the cell by hijacking the DNA replication machinery, and ultimately killing the host, (2) the multiplying effect (given that they are capable of replicating in the presence of the target organism), (3) the specificity of phage against the target pathogen and (4) their inability to attack mammalian host cells or tissue. Phage therapy has been practised in Eastern European countries for decades, although largely neglected by the more-developed world after the advent of antibiotics in the middle of the 20th century. Although work on phage therapy continues in Eastern Europe, this was not conducted to a standard allowing it to support clinical uses in areas regulated by the European Medicines Agency or the US Food and Drug Administration (Harper et al., 2011). Bacteriophages have been used successfully to control infections in animals and animal models of infection. Efficacy of bacteriophage to control Pseudomonas infections in dogs, humans and mice has been reported (Wright et al., 2009; Debarbieux et al., 2010; Burrowes et al., 2011; Harper et al., 2011; Saussereau & Debarbieux, 2012). Very recently, we demonstrated the effective use of bacteriophage to control Pseudomonas lung infections (a major cause of lung infection in patients with cystic fibrosis) in a validated mouse model and on biofilms formed on human airway cells by using a cocktail of two phages, QMR299-2 and QNH-4 (Alemayehu et al., 2012). Efficacies of bacteriophages to control the E. coli O157 in sheep, steers and on cowhides have also been reported (Raya et al., 2006; Rozema et al., 2009; Coffey et al., 2011). In 2011 in Germany, a large outbreak of haemolytic uraemic syndrome and bloody diarrhoea caused by the Shiga toxinproducing E. coli O104:H4 was reported. The strain

6 M. C. Rea and others responsible for the epidemic was resistant to all penicillins and cephalosporins (Merabishvili et al., 2012). Recently, bacteriophages that efficiently lyse the E. coli O104:H4 outbreak strains have been suggested as potential therapeutics against such antibiotic-resistant pathogens (Merabishvili et al., 2012; Muniesa et al., 2012). Experimental data have also shown that both bacteriophages and their lysins could be effective in the treatment of methicillin-resistant St. aureus, one of today s most prevalent pathogens associated with nosocomial infections (Fischetti, 2008; Fenton et al., 2010; Borysowski et al., 2011). Recently, the potential of bacteriophage therapy against experimental bubonic plague was highlighted (Filippov et al., 2012). The report shows that administration of phage QA1122 provided 40 % protection to BALB/c mice against 1000 LD 50 of Yersinia pestis CO92 and extended the mean time to death by 84 %, showing that QA1122 bacteriophage is a promising alternative therapy against multidrug-resistant plague strains (Filippov et al., 2012). Studies on C. difficile bacteriophages, however, are limited and early reports deal mainly with their use in strain typing and morphological characterization (Fortier & Moineau, 2007). Recent work has focused more on the isolation, morphotyping, molecular characterization and genome sequencing of C. difficile bacteriophages (see Table 3 for a list of C. difficile bacteriophages whose genome sequences are available). Almost all C. difficile phages reported so far were isolated exclusively from the host cell following mitomycin induction (Table 3). Shan et al. (2012) performed an analysis of prophage carriage and diversity on 16 clinically isolated C. difficile ribotypes and found that 15 of the ribotypes were myovirus positive whereas two carry additional siphovirus prophages, while only one ribotype was negative for prophages. Recently, Nale et al. (2012) reported that they induced prophages from 91 clinical C. difficile 027 isolates and identified myophages and siphophages in 63 and three of the isolates, respectively, whereas dual phage type carriage was observed only in four isolates. Both reports showed convincing data regarding the very lysogenic nature of C. difficile phages and their coexistence through integration into the host genome. Studies on functional analysis of prophages revealed that they can influence the phenotype of their host, e.g. expression of potent toxins to make the pathogen more aggressively virulent. This is the case, for instance, with the Shiga toxin in E. coli, cholera toxin in Vibrio cholerae and the botulinum neurotoxins in C. botulinum (Brussow et al., 2005) where their expression is positively affected by the presence of prophage. A possible role of prophages in influencing the TcdA and TcdB exotoxin expression in C. difficile has also been reported (Goh et al., 2005; Sekulovic et al., 2011). Introducing the genome of phage CD38-2, a pac-type temperate siphophage, into the C. difficile CD274 host stimulated TcdA and TcdB expression over twofold, demonstrating a possible role for the phage genome in toxin production in C. difficile (Sekulovic et al., 2011). Table 3. Referenced C. difficile phages with complete genome sequence data Phage ID Family Genome size (kb) G+C (%) Source Reference GenBank accession number QMMP02 Myoviridae C. difficile infected Meessen-Pinard et al. (2012) JX QMMP04 Myoviridae Human stool (viral genome located JX in C. difficile host) Q38-2 Siphoviridae Mitomycin induced Sekulovic et al. (2011) YP_ QCD6356 Siphoviridae Mitomycin induced Horgan et al. (2010) ZP_ QCD27 Myoviridae Mitomycin induced Mayer et al. (2008) YP_ QC2 Myoviridae Mitomycin induced Goh et al. (2007) DQ QCD630 Myoviridae Mytomycin induced Sebaihia et al. (2006) YP_ QCD119 Myoviridae Mitomycin induced Govind et al. (2006) YP_ Journal of Medical Microbiology 62

7 Biological solutions for the control of C. difficile infection Contrary to this, however, a very recent study by Meessen- Pinard et al. (2012) reported a natural induction of prophage in CDIs. In the study, four Myoviridae phages, QMMP01, QMMP02, QMMP03 and QMMP04, were recovered as free viral particles in filter-sterilized stool supernatants of patients suffering from CDI, which provides evidence that prophages are naturally induced and can kill the host (C. difficile) during episodes of CDI (Meessen-Pinard et al., 2012). The current knowledge regarding the use of phage therapy for the control of CDI is limited. This is mainly due to the lysogenic nature of C. difficile bacteriophages studied and isolated to date. Thus far, no lytic phage specific for C. difficile has been isolated, with all reports to date showing that phage were recovered only after induction of the host with mitomycin C (Table 3). The resulting phage can integrate easily during infection of a new host, limiting the lytic cycle and resulting in a very low phage titre in vitro. It has been suggested that the temperate lifestyle of C. difficile phage is due to (1) the high incidence of prophage genes which may impart resistance to further infections and (2) the spore-forming nature of C. difficile which would also favour phage that integrate into the genome of the host (Meader et al., 2010). Revathi et al. (2011) also reported recently the lysogenization of phage CD119 into the genome of C. difficile during infection in the hamster model and highlighted the importance of lysogeny of C. difficile phages as an evolutionary adaptation for survival. Despite the lysogenic nature of C. difficile phages, anti-c. difficile bacteriophages and their active endolysins have potential as effective antimicrobial agents. Administration of bacteriophage intragastrically to hamsters challenged with C. difficile protected the animals against infection, whereas control animals that did not receive phage died within 96 h (Ramesh et al., 1999). By using an ex vivo faecal fermentation model (that simulated the conditions in the human colon), Meader et al. (2010) also demonstrated how effectively phage therapy can reduce/prevent C. difficile toxin production and/or colonization. To overcome the problem of the lysogenic nature of C. difficile bacteriophage, cloned biologically active phage endolysins have been suggested as an alternative avenue for the development of phage-based anti-c. difficile therapeutics. The cloned endolysin CD27L was active against a diverse range of C. difficile strains, including ribotype 027 (Mayer et al., 2008, 2011). We reported previously the isolation and genetic characterization of phage QCD6356, the first report of a C. difficile temperate phage belonging to the Siphoviridae family (Horgan et al., 2010). This endolysin has been cloned, and the partially purified protein was shown to have high lytic activity against C. difficile ribotype 027 and resulted in a reduced cell count of 2 3 log units (Burke et al., 2012). While lytic bacteriophage have shown potential in their ability to kill a wide range of pathogens involved in infectious diseases, the lysogenic nature of the C. difficile phages poses a formidable challenge for their use as therapeutics, and may, in fact, prove counterproductive if they integrate into the genome of the host and impact on toxin production. However, the ability to clone active phage endolysins may be the way forwards to developing a very specific anti-c. difficile therapeutic, which may go some way towards overcoming the increasing problem of developing drug resistance within the gut microflora. Overall conclusions While the discovery of antibiotics in the first half of the 20th century undoubtedly heralded a new age in our ability to fight infections, the emergence of organisms like C. difficile highlights many of their limitations and side effects. There is a clear need for a new approach to fight gastrointestinal infections and one that takes into account the need to maintain the integrity of the gut microbiota. What better place to look for such therapeutics than in the complexity of the human gut itself? In this review, we have outlined the current state of play with regard to the potential of gut-derived probiotics, phages and bacteriocins to act as antimicrobial agents against CDI in the human gut. ACKNOWLEDGEMENTS The authors and their work were supported by the Science Foundation of Ireland (SFI) and funded by the Centre for Science, Engineering and Technology grant 02/CE/B124. The Alimentary Pharmabiotic Centre is a research centre funded by the SFI. REFERENCES Aas, J., Gessert, C. E. & Bakken, J. S. (2003). Recurrent Clostridium difficile colitis: case series involving 18 patients treated with donor stool administered via a nasogastric tube. Clin Infect Dis 36, Alemayehu, D., Casey, P. G., McAuliffe, O., Guinane, C. M., Martin, J. G., Shanahan, F., Coffey, A., Ross, R. P. & Hill, C. (2012). Bacteriophages QMR299-2 and QNH-4 can eliminate Pseudomonas aeruginosa in the murine lung and on cystic fibrosis lung airway cells. MBio 3, e Arias, C. A. & Murray, B. E. (2009). Antibiotic-resistant bugs in the 21st century a clinical super-challenge. N Engl J Med 360, Baines, S. D., O Connor, R., Freeman, J., Fawley, W. N., Harmanus, C., Mastrantonio, P., Kuijper, E. J. & Wilcox, M. H. (2008). Emergence of reduced susceptibility to metronidazole in Clostridium difficile. J Antimicrob Chemother 62, Bartlett, J. G. (2002). Antibiotic-associated diarrhea. N Engl J Med 346, Bartoloni, A., Mantella, A., Goldstein, B. P., Dei, R., Benedetti, M., Sbaragli, S. & Paradisi, F. (2004). In-vitro activity of nisin against clinical isolates of Clostridium difficile. J Chemother 16, Boakes, S., Ayala, T., Herman, M., Appleyard, A. N., Dawson, M. J. & Cortés, J. (2012). Generation of an actagardine A variant library through saturation mutagenesis. Appl Microbiol Biotechnol 95, Borysowski, J., Lobocka, M., Międzybrodzki, R., Weber-Dabrowska, B. & Górski, A. (2011). Potential of bacteriophages and their lysins in

8 M. C. Rea and others the treatment of MRSA: current status and future perspectives. BioDrugs 25, Brandl, K., Plitas, G., Mihu, C. N., Ubeda, C., Jia, T., Fleisher, M., Schnabl, B., DeMatteo, R. P. & Pamer, E. G. (2008). Vancomycinresistant enterococci exploit antibiotic-induced innate immune deficits. Nature 455, Brazier, J. S., Raybould, R., Patel, B., Duckworth, G., Pearson, A., Charlett, A., Duerden, B. I. & HPA Regional Microbiology Network (2008). Distribution and antimicrobial susceptibility patterns of Clostridium difficile PCR ribotypes in English hospitals, Euro Surveill 13 (41) 3. Brussow, H., Canchaya, C. & Hardt, W. (2005). Phages and the evolution of bacterial pathogen: from genomic rearrangements to lysogenic conversion. Microbiol Mol Biol Rev 68, Burke, D., Alemayehu, D., Rea, M. C., McAuliffe, O., Cooney, C. J., Hill, C. & Ross, R. P. (2012). Cloning and expression of a biologically active endolysin from the Clostridium difficile bacteriophage QCD6356. Abstract Poster Presentation 195. Viruses of Microbes II, July 2012, Brussels, Belgium. Burrowes, B., Harper, D. R., Anderson, J., McConville, M. & Enright, M. C. (2011). Bacteriophage therapy: potential uses in the control of antibiotic-resistant pathogens. Expert Rev Anti Infect Ther 9, Carlet, J. (2012). The gut is the epicentre of antibiotic resistance. Antimicrob Resist Infect Control 1, 39. Citron, D. M., Tyrrell, K. L., Merriam, C. V. & Goldstein, E. J. (2012). Comparative in vitro activities of LFF571 against Clostridium difficile and 630 other intestinal strains of aerobic and anaerobic bacteria. Antimicrob Agents Chemother 56, Coffey, B., Rivas, L., Duffy, G., Coffey, A., Ross, R. P. & McAuliffe, O. (2011). Assessment of Escherichia coli O157:H7-specific bacteriophages e11/2 and e4/1c in model broth and hide environments. Int J Food Microbiol 147, Corr, S. C., Li, Y., Riedel, C. U., O Toole, P. W., Hill, C. & Gahan, C. G. (2007). Bacteriocin production as a mechanism for the antiinfective activity of Lactobacillus salivarius UCC118. Proc Natl Acad Sci U S A 104, Corr, S. C., Hill, C. & Gahan, C. G. (2009). Understanding the mechanisms by which probiotics inhibit gastrointestinal pathogens. Adv Food Nutr Res 56, Cotter, P. D., Ross, R. P. & Hill, C. (2013). Bacteriocins a viable alternative to antibiotics? Nat Rev Microbiol 11, Davis, C. D. & Milner, J. A. (2009). Gastrointestinal microflora, food components and colon cancer prevention. J Nutr Biochem 20, Debarbieux, L., Leduc, D., Maura, D., Morello, E., Criscuolo, A., Grossi, O., Balloy, V. & Touqui, L. (2010). Bacteriophages can treat and prevent Pseudomonas aeruginosa lung infections. J Infect Dis 201, Dobson, A., Cotter, P. D., Ross, R. P. & Hill, C. (2012). Bacteriocin production: a probiotic trait? Appl Environ Microbiol 78, 1 6. Fenton, M., Casey, P. G., Hill, C., Gahan, C. G., Ross, R. P., McAuliffe, O., O Mahony, J., Maher, F. & Coffey, A. (2010). The truncated phage lysin CHAP(k) eliminates Staphylococcus aureus in the nares of mice. Bioeng Bugs 1, Field, D., Quigley, L., O Connor, P. M., Rea, M. C., Daly, K., Cotter, P. D., Hill, C. & Ross, R. P. (2010). Studies with bioengineered Nisin peptides highlight the broad-spectrum potency of Nisin V. Microb Biotechnol 3, Filippov, A. A., Sergueev, K. V., He, Y., Huang, X. Z., Gnade, B. T., Mueller, A. J., Fernandez-Prada, C. M. & Nikolich, M. P. (2012). Bacteriophage therapy of experimental bubonic plague in mice. In Advances in Yersinia Research (Advances in Experimental Medicine and Biology vol. 954), pp Edited by A. M. P. de Almeida & N. C. Leal. New York: Springer. Fischetti, V. A. (2008). Bacteriophage lysins as effective antibacterials. Curr Opin Microbiol 11, Fortier, L. C. & Moineau, S. (2007). Morphological and genetic diversity of temperate phages in Clostridium difficile. Appl Environ Microbiol 73, Fouhy, F., Ross, R. P., Fitzgerald, G. F., Stanton, C. & Cotter, P. D. (2012). Composition of the early intestinal microbiota: knowledge, knowledge gaps and the use of high-throughput sequencing to address these gaps. Gut Microbes 3, Freeman, J., Bauer, M. P., Baines, S. D., Corver, J., Fawley, W. N., Goorhuis, B., Kuijper, E. J. & Wilcox, M. H. (2010). The changing epidemiology of Clostridium difficile infections. Clin Microbiol Rev 23, Gao, X. W., Mubasher, M., Fang, C. Y., Reifer, C. & Miller, L. E. (2010). Dose response efficacy of a proprietary probiotic formula of Lactobacillus acidophilus CL1285 and Lactobacillus casei LBC80R for antibiotic-associated diarrhea and Clostridium difficile-associated diarrhea prophylaxis in adult patients. Am J Gastroenterol 105, Ghantoji, S. S., Sail, K., Lairson, D. R., DuPont, H. L. & Garey, K. W. (2010). Economic healthcare costs of Clostridium difficile infection: a systematic review. J Hosp Infect 74, Goh, S., Riley, T. V. & Chang, B. J. (2005). Isolation and characterization of temperate bacteriophages of Clostridium difficile. Appl Environ Microbiol 71, Goh, S., Ong, F. P., Song, P. K., Rikey, V. T. & Chang, J. B. (2007). The complete genome sequence of Clostridium difficile phage wc2 and comparisons to wcd119 and inducible prophages of CD630. Microbiology 153, Govind, R., Fralick, J. A. & Rolfe, R. D. (2006). Genomic organization and molecular characterization of Clostridium difficile bacteriophage fcd119. J Bacteriol 188, Grehan, M. J., Borody, T. J., Leis, S. M., Campbell, J., Mitchell, H. & Wettstein, A. (2010). Durable alteration of the colonic microbiota by the administration of donor fecal flora. J Clin Gastroenterol 44, Harper, D. R., Anderson, J. & Enright, M. C. (2011). Phage therapy: delivering on the promise. Therapeutic Delivery 2 (7), Hickson, M., D Souza, A. L., Muthu, N., Rogers, T. R., Want, S., Rajkumar, C. & Bulpitt, C. J. (2007). Use of probiotic Lactobacillus preparation to prevent diarrhoea associated with antibiotics: randomised double blind placebo controlled trial. BMJ 335, 80. Horgan, M., O Sullivan, O., Coffey, A., Fitzgerald, G. F., van Sinderen, D., McAuliffe, O. & Ross, R. P. (2010). Genome analysis of the Clostridium difficile phage PhiCD6356, a temperate phage of the Siphoviridae family. Gene 462, Huang, H., Weintraub, A., Fang, H. & Nord, C. E. (2009). Antimicrobial resistance in Clostridium difficile. Int J Antimicrob Agents 34, Johnson, A. P. & Wilcox, M. H. (2012). Fidaxomicin: a new option for the treatment of Clostridium difficile infection. J Antimicrob Chemother 67, Johnson, S., Maziade, P. J., McFarland, L. V., Trick, W., Donskey, C., Currie, B., Low, D. E. & Goldstein, E. J. (2012). Is primary prevention of Clostridium difficile infection possible with specific probiotics? Int J Infect Dis 16, e786 e792. Lawrence, S. J., Korzenik, J. R. & Mundy, L. M. (2005). Probiotics for recurrent Clostridium difficile disease. J Med Microbiol 54, Journal of Medical Microbiology 62

9 Biological solutions for the control of C. difficile infection Mayer, M. J., Narbad, A. & Gasson, M. J. (2008). Molecular characterization of a Clostridium difficile bacteriophage and its cloned biologically active endolysin. J Bacteriol 190, Mayer, M. J., Garefalaki, V., Spoerl, R., Narbad, A. & Meijers, R. (2011). Structure-based modification of a Clostridium difficiletargeting endolysin affects activity and host range. J Bacteriol 193, McFarland, L. (2000). Normal flora: diversity and functions. Microb Ecol Health Dis 12, McFarland, L. V. (2009). Evidence-based review of probiotics for antibiotic-associated diarrhea and Clostridium difficile infections. Anaerobe 15, McFarland, L. V., Surawicz, C. M., Greenberg, R. N., Fekety, R., Elmer, G. W., Moyer, K. A., Melcher, S. A., Bowen, K. E., Cox, J. L. & other authors (1994). A randomized placebo-controlled trial of Saccharomyces boulardii in combination with standard antibiotics for Clostridium difficile disease. JAMA 271, Meader, E., Mayer, M. J., Gasson, M. J., Steverding, D., Carding, S. R. & Narbad, A. (2010). Bacteriophage treatment significantly reduces viable Clostridium difficile and prevents toxin production in an in vitro model system. Anaerobe 16, Meessen-Pinard, M., Sekulovic, O. & Fortier, L. C. (2012). Evidence of in vivo prophage induction during Clostridium difficile infection. Appl Environ Microbiol 78, Merabishvili, M., De Vos, D., Verbeken, G., Kropinski, A. M., Vandenheuvel, D., Lavigne, R., Wattiau, P., Mast, J., Ragimbeau, C. & other authors (2012). Selection and characterization of a candidate therapeutic bacteriophage that lyses the Escherichia coli O104:H4 strain from the 2011 outbreak in Germany. PLoS ONE 7, e Miller, M. (2009). The fascination with probiotics for Clostridium difficile infection: lack of evidence for prophylactic or therapeutic efficacy. Anaerobe 15, Muniesa, M., Hammerl, J. A., Hertwig, S., Appel, B. & Brüssow, H. (2012). Shiga toxin-producing Escherichia coli O104:H4: a new challenge for microbiology. Appl Environ Microbiol 78, Nale, J. Y., Shan, J., Hickenbotham, P. T., Fawley, W. N., Wilcox, M. H. & Clokie, M. R. (2012). Diverse temperate bacteriophage carriage in Clostridium difficile 027 strains. PLoS ONE 7, e Pillai, A. & Nelson, R. (2008). Probiotics for treatment of Clostridium difficile-associated colitis in adults. Cochrane Database Syst Rev (1), CD Pineiro, M. & Stanton, C. (2007). Probiotic bacteria: legislative framework requirements to evidence basis. J Nutr 137 (Suppl 2), 850S 853S. Pochapin, M. (2000). The effect of probiotics on Clostridium difficile diarrhea. Am J Gastroenterol 95 (Suppl), S11 S13. Qin, J., Li, R., Raes, J., Arumugam, M., Burgdorf, K. S., Manichanh, C., Nielsen, T., Pons, N., Levenez, F. & other authors (2010). A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464, Ramesh, V., Fralick, J. A. & Rolfe, D. (1999). Prevention of Clostridium difficile-induced ileocecitis with bacteriophage. Anaerobe 5, Raya, R. R., Varey, P., Oot, R. A., Dyen, M. R., Callaway, T. R., Edrington, T. S., Kutter, E. M. & Brabban, A. D. (2006). Isolation and characterization of a new T-even bacteriophage, CEV1, and determination of its potential to reduce Escherichia coli O157:H7 levels in sheep. Appl Environ Microbiol 72, Rea, M. C., Clayton, E., O Connor, P. M., Shanahan, F., Kiely, B., Ross, R. P. & Hill, C. (2007). Antimicrobial activity of lacticin 3147 against clinical Clostridium difficile strains. J Med Microbiol 56, Rea, M. C., Sit, C. S., Clayton, E., O Connor, P. M., Whittal, R. M., Zheng, J., Vederas, J. C., Ross, R. P. & Hill, C. (2010). Thuricin CD, a posttranslationally modified bacteriocin with a narrow spectrum of activity against Clostridium difficile. Proc Natl Acad Sci U S A 107, Rea, M. C., Cotter, P. D., Hill, C. & Ross, R. P. (2011a). Classification of bacteriocins from Gram positive bacteria. In Prokaryotic Antimicrobial Peptides. From Genes to Applications, pp Edited by D. Drider & S. Rebuffat. New York: Springer. Rea, M. C., Dobson, A., O Sullivan, O., Crispie, F., Fouhy, F., Cotter, P. D., Shanahan, F., Kiely, B., Hill, C. & Ross, R. P. (2011b). Effect of broad- and narrow-spectrum antimicrobials on Clostridium difficile and microbial diversity in a model of the distal colon. Proc Natl Acad Sci U S A 108 (Suppl 1), Rea, M. C., O Sullivan, O., Shanahan, F., O Toole, P. W., Stanton, C., Ross, R. P. & Hill, C. (2012). Clostridium difficile carriage in elderly subjects and associated changes in the intestinal microbiota. J Clin Microbiol 50, Revathi, G., Fralick, J. A. & Rolfe, R. D. (2011). In vivo lysogenization of a Clostridium difficile bacteriophage QCD119. Anaerobe 17, Riley, M. A., Robinson, S. M., Roy, C. M., Dennis, M., Liu, V. & Dorit, R. L. (2012). Resistance is futile: the bacteriocin model for addressing the antibiotic resistance challenge. Biochem Soc Trans 40, Rohlke, F., Surawicz, C. M. & Stollman, N. (2010). Fecal flora reconstitution for recurrent Clostridium difficile infection: results and methodology. J Clin Gastroenterol 44, Rozema, E. A., Stephens, T. P., Bach, S. J., Okine, E. K., Johnson, R. P., Stanford, K. & McAllister, T. A. (2009). Oral and rectal administration of bacteriophages for control of Escherichia coli O157:H7 in feedlot cattle. J Food Prot 72, Rupnik, M., Wilcox, M. H. & Gerding, D. N. (2009). Clostridium difficile infection: new developments in epidemiology and pathogenesis. Nat Rev Microbiol 7, Ryan, J., Murphy, C., Twomey, C., Paul Ross, R., Rea, M. C., MacSharry, J., Sheil, B. & Shanahan, F. (2010). Asymptomatic carriage of Clostridium difficile in an Irish continuing care institution for the elderly: prevalence and characteristics. Ir J Med Sci 179, Saussereau, E. & Debarbieux, L. (2012). Bacteriophages in the experimental treatment of Pseudomonas aeruginosa infections in mice. Adv Virus Res 83, Scanlan, P. D., Shanahan, F., O Mahony, C. & Marchesi, J. R. (2006). Culture-independent analyses of temporal variation of the dominant fecal microbiota and targeted bacterial subgroups in Crohn s disease. J Clin Microbiol 44, Sebaihia, M., Wren, B. W., Mullany, P., Fairweather, N. F., Minton, N. & other authors (2006). The multidrug-resistant human pathogen Clostridium difficile has a high mobile, mosaic genome. Nat Genet 38, Sekulovic, O., Meessen-Pinard, M. & Fortier, L. C. (2011). Prophagestimulated toxin production in Clostridium difficile NAP1/027 lysogens. J Bacteriol 193, Shan, J., Patel, K. V., Hickenbotham, P. T., Nale, J. Y., Hargreaves, K. R. & Clokie, M. R. (2012). Prophage carriage and diversity within clinically relevant strains of Clostridium difficile. Appl Environ Microbiol 78, Shanahan, F. (2011). The colonic microflora and probiotic therapy in health and disease. Curr Opin Gastroenterol 27, Surawicz, C. M., McFarland, L. V., Greenberg, R. N., Rubin, M., Fekety, R., Mulligan, M. E., Garcia, R. J., Brandmarker, S., Bowen, K. & other authors (2000). The search for a better treatment for

10 M. C. Rea and others recurrent Clostridium difficile disease: use of high-dose vancomycin combined with Saccharomyces boulardii. Clin Infect Dis 31, Wall, R., Marques, T. M., O Sullivan, O., Ross, R. P., Shanahan, F., Quigley, E. M., Dinan, T. G., Kiely, B., Fitzgerald, G. F. & other authors (2012). Contrasting effects of Bifidobacterium breve NCIMB and Bifidobacterium breve DPC 6330 on the composition of murine brain fatty acids and gut microbiota. Am J Clin Nutr 95, Warren, C. A. & Guerrant, R. L. (2011). Pathogenic C difficile is here (and everywhere) to stay. Lancet 377, 8 9. Wiegand, P. N., Nathwani, D., Wilcox, M. H., Stephens, J., Shelbaya, A. & Haider, S. (2012). Clinical and economic burden of Clostridium difficile infection in Europe: a systematic review of healthcare-facilityacquired infection. J Hosp Infect 81, Wlodarska, M., Willing, B., Keeney, K. M., Menendez, A., Bergstrom, K. S., Gill, N., Russell, S. L., Vallance, B. A. & Finlay, B. B. (2011). Antibiotic treatment alters the colonic mucus layer and predisposes the host to exacerbated Citrobacter rodentium-induced colitis. Infect Immun 79, Wright, A., Hawkins, C. H., Anggård, E. E. & Harper, D. R. (2009). A controlled clinical trial of a therapeutic bacteriophage preparation in chronic otitis due to antibiotic-resistant Pseudomonas aeruginosa; a preliminary report of efficacy. Clin Otolaryngol 34, Wullt, M., Hagslätt, M. L. & Odenholt, I. (2003). Lactobacillus plantarum 299v for the treatment of recurrent Clostridium difficileassociated diarrhoea: a double-blind, placebo-controlled trial. Scand J Infect Dis 35, Journal of Medical Microbiology 62