lable at ScienceDirect

Anaerobe 30 (2014) 193e198

Contents lists avai

Anaerobe

journal homepage: www.elsevier .com/locate/anaerobe

Molecular biology, genetics and biotechnology

Biofilms of Clostridium species

V�eronique Pantal�eon, Sylvie Bouttier, Anna Philibertine Soavelomandroso, Claire Janoir,Thomas Candela*

EA4043, Facult�e de Pharmacie, Universit�e Paris Sud, 5 Rue Jean Baptiste Cl�ement, 92296 Chatenay-Malabry Cedex, France

a r t i c l e i n f o

Article history:Available online 19 September 2014

Keywords:Clostridium speciesClostridium difficileClostridium perfringensBiofilmRegulation

* Corresponding author.E-mail address: thomas.candela@u-psud.fr (T. Can

http://dx.doi.org/10.1016/j.anaerobe.2014.09.0101075-9964/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

The biofilm is a microbial community embedded in a synthesized matrix and is the main bacterial way oflife. A biofilm adheres on surfaces or is found on interfaces. It protects bacteria from the environment,toxic molecules and may have a role in virulence. Clostridium species are spread throughout both en-vironments and hosts, but their biofilms have not been extensively described in comparison with otherbacterial species. In this review we describe all biofilms formed by Clostridium species during both in-dustrial processes and in mammals where biofilms may be formed either during infections or associatedto microbiota in the gut. We have specifically focussed on Clostridium difficile and Clostridium perfringensbiofilms, which have been studied in vitro. Regulatory processes including sporulation and germinationhighlight how these Clostridium species live in biofilms. Furthermore, biofilms may have a role in thesurvival and spreading of Clostridium species.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

A biofilm is a bacterial community developed on a surface (thiscould be an interface, like air/liquid) producing a polymeric matrixallowing bacteria to attach together. Biofilms are the main bacterialway of life [1]. The first step is an initial bacterial adhesion that caneither be reversible or irreversible. Secondly, bacteria grow and soare embedded in the matrix that they synthesize. Thereafter thebiofilm matures and its structure changes. All such events areessential for biofilm formation. Bacterial biofilms are widelydistributed and play important roles in many environments.Indeed, living in biofilms helps bacteria survive and may play a rolein virulence.

In this review, we will describe the biofilm formed by theClostridium species, which are obligatory anaerobic, endospore-forming, Gram-positive bacteria. These species form the Clos-tridium genus classed in the Clostridiales order and Clostridiaceaefamily. However, according to the rRNA sequences, many speciesnow belong to other families and even other classes [2]. Tosummarise, the Clostridium genus belongs to the Firmicutesphylum that encloses three different classes: Bacilli, Clostridia andErysipelotrichi. A total of 139 Clostridium species were identified

dela).

as belonging either to the Clostridia class in Clostridiaceae (71species), Lachnospiraceae (32 species), Peptostretococcaceae (13species), Ruminococcaceae (16 species), Incertae Sedis (2 species)families or to the Erysipelotrichi class in Erysipelotrichaceae family(5 species). In the interests of clarity we shall focus this review onall published biofilms formed by a Clostridium species, indepen-dently of class or family. Surprisingly, only a few papers werefound covering this field, in comparison with biofilms formed byother bacteria belonging to the Firmicutes phylum, such as Ba-cillus or Staphylococcus species. This raises the question whetherClostridium species are really able to develop a proper biofilm.Most of the described Clostridium species may form biofilms thatmay be found in humans and formed by pathogenic bacteria suchas Clostridium difficile, Clostridium botulinum and Clostridiumperfringens or by non-pathogenic bacteria, involved with themicrobiota in the gut such as Clostridium clostridioforme andClostridium malenominatum. They may also being involved in thecellulose degradation processes with Clostridium thermocellumand Clostridium acetobutylicum during industrial production orrecycling. In all cases, biofilms formed by Clostridium species inthe environment are composed of many bacterial species. How-ever, such multi-species biofilms are difficult to study. Therefore,most often, mono-species biofilm are preferred to investigate. Wewill review here the capacity of some Clostridium species to forma mono-species biofilm and/or to be part of the multi-speciesbiofilm.

V. Pantal�eon et al. / Anaerobe 30 (2014) 193e198194

2. Clostridium species biofilms in industry

In industrial production or sewage plants, many Clostridiumspecies are found to participate in multi-species biofilms (Table 1).All the studied biofilms were grown in the presence of organicwaste containing cellulose. For example, Clostridium celerecrescenswas identified in microbial fuel cells, devices that exploit micro-organisms to generate electrical power [3]. This bacterium wasfound with specialized populations composed mostly of Rhizobialesand is able to degrade cellulose. In similar cellulose fed biogasproduction systems, C. thermocellum and Clostridium cellulolyticumwere identified using metagenomics as the predominant species inthe hydrolytic phase [4]. Cellulose is one of the most commonsubstrates found in the biomass feed-stock. This may explain thehigh presence of bacteria able to degrade this polymer in industryand in biofilm reactors where plants are the main substrate. Indeed,Clostridia is the major bacterial class found in moving bed biofilmreactor systems treating municipal sewage [5]. Moreover, Clos-tridium ljungdahlii and Clostridium drakei were predominant inHollow-fibermembrane biofilm reactors [6]. All of these industrialClostridium species grow well in the presence of cellulose, whichmay promote growth and biofilm formation.

The mono-species biofilm is often a model used to study biofilmformation since it is easier to analyse a biofilm formed by only onebacterium. Only 9 out of 139 Clostridium species were studied inmono-species biofilms. Among them, 3 Clostridium species are usedin industrial production because of their capacity to degrade cel-lulose components and produce valuable chemicals. The tech-niques to study mono-species biofilms are well established.Biofilms are usually observed by microscopic techniques such asscanning electron microscopy, confocal laser scanning microscopyand are quantified by counting methods and colorimetric assayusing violet crystal. Yet, the biofilms formed by C. acetobutylicum,Clostridium phytofermentans and C. thermocellum are not well

Table 1Clostridium species found in biofilms.

Species Biofilm Surface or car

C. acetobutylicum Mono-species Fed-batch fermbiofilm buildupumice, glass

C. baratii Mono- and dual species 96-well flat boC. bifermentans Multi-species Within the ep

Mono- and dual species 96-well flatboC. botulinum Multi-species Tooth apical foC. butyricum Multi-species Mucosal (biofiC. celerecrescens Multi-species Microbial fuelC. cellulolyticum Multi-species Anode chambC. clostridioforme Multi-species Mucosal (biofi

using vesselsC. difficile Mono- and dual species 96-well flatbo

Mono-species 24-well polystpolycarbonate

C. drakei Multi-species Hollow-fibre mC. fallax Mono- and dual species 96-well flat boC. ljungdahlii Multi-species Hollow-fiber mC. malenominatum Multi-species Two stage conC. perfringens Multi-species Mucosal (biofi

culture systemThree-species PolycarbonateMono- and dual species 96-well flatboMono species 24- or 6-well

C. phytofermentans Mono-species Submerged-stC. sordelli Multi-species Within the epC. thermocellum Mono-species Cellulose cottoClostridium spp close to

C. chartatabidumMono-species Resistant starc

Undefined Clostridium spp. Multi-species Surfaces of orafibres or two-

characterised. C. acetobutylicum is used in biofilm reactors to opti-mise acetone, butanol and ethanol production [7e9]. This bacte-rium is able to use hemicellulosic pentoses xylose and arabinosethat come from lignocellulosic materials. C. acetobutylicum is amajor bacterium of industry and was extensively studied. However,with the exception of a biofilm observed by Liu et al. using scanningelectron microscopy, no other observations showedC. acetobutylicum biofilms in industrial conditions. C. phyto-fermentans is mesophilic and able to use complex carbohydrates toproduce ethanol. Using scanning electron microscopy, C. phyto-fermentans was shown in one study to form a biofilm on switch-grass [10].The third bacteria forming a mono-biofilm in industry, C.thermocellum, was shown to form a biofilm, by confocal laserscanning microscopy, on cellulose fibres growing as a single bac-terial layer [11,12]. Noteworthy, in the Wang et al. study, aC. thermocellum mature biofilm was observed to form crater likedepressions after 44 h of incubation, showing the biofilm's well-organized structures [12]. The capacity of all these “industrial”species to form a biofilm is not extensively studied although there isincreasing interest. Indeed, biofilms are potentially more resistantto toxic compounds than planktonic culture (see below) and maytherefore improve the industrial production. However, some in-vestigations on the biofilm adherence and matrix composition arestill missing and should be performed in order to improve biofilmformation by these species.

3. Clostridium species isolated from in vivo biofilm

Multi-species biofilms are also found in vivo, particularly in thegut environment since Clostridium species are part of the dominantanaerobic digestive microbiota. Bacteria associated with the guthave been described forming microcolonies or even biofilms.However, in the human gut, it is not totally established if theobserved bacterial layer is a biofilm or not. Some studies suggest

rier References

entation in biofilm reactors, reactor with glass beads forp and fixed bed reactor with different carriers (silica gel,beads, silica sands, Tygon®, Teflon, PVC)

[6e9]

ttomed plastic tissue culture plate [21]idermidis and biliary stent [19,47]ttomed plastic tissue culture plate [21]ramen and external radicular surface [53]lm suggested) [14]cells [3]er and Two-phase leach-bed biogas reactor system [4,54]lm suggested) and Two stage continous culture system [14,15]

ttomed plastic tissue culture plate [21]yrene plates, vented tissue culture flasks (25 cm3),membrane

[23,24,26,52]

embrane biofilm reactor [6]ttomed plastic tissue culture plate [21]embrane biofilm reactor [6]tinuous culture system using vessels [15]lm suggested), biliary stent and two stage continuoususing vessels

[14,15,19]

membrane in a drip-flow reactor [18]ttomed plastic tissue culture plate [21]polystyrene tissue culture plates [25,27]ate fermentation bottles [10]idermidis [20]n fibres, cellulose filter paper [11,55]h [16]

l squamous cell carcinomas, zeolite particles, grass silagephase leach-bed biogas reactor system

[4,56,57]

V. Pantal�eon et al. / Anaerobe 30 (2014) 193e198 195

that biofilm is present in the human gut. For example, Swidsinskiet al. showed, by in situ-hybridization and microscopy analyses,that a biofilm is present in patients with inflammatory bowel dis-ease whereas biofilms were only found sporadically in healthypatients [13]. Moreover, in patients with ulcerative colitis, thecomposition of the mucosal communities differed from healthypatients. Noteworthy, Clostridium butyricum and C. clostridioformeare reduced in patients with ulcerative colitis as compared tohealthy patients, whereas C. perfringens is found in similar quantityin the two groups [14]. Biofilms in the gut may be found in themucus layer localised close to the epithelium surface of the intes-tine [15] or on the nutrients present in the intestinal lumen [16].Several Clostridium species were indeed reported in biofilms asso-ciated with plant biomass found in the herbivore gastrointestinaltract [17]. Moreover, a new species of Clostridiumwas discovered tobe attached on resistant starch in human feces, forming a rosette-like structure around the starch and subsequently a biofilm [16].This novel Clostridium species (not named in the study) is close toClostridium chartatabidum, which is a cellulolytic bacteriumdependent for growth on resistant starch. It was difficult to isolatefrom other bacteria and this may be due to the formation of multi-species biofilm on the resistant starch.

Multi-species biofilms are also described in models trying tomimic in vivo situations. A two stage continuous system, mimickingthe human gut, was built using fresh feces from a healthy donor andporcine mucin as the main substrate. This system may representthe gut where the wall of the vessels mimicked the epithelial layerand the vessels the intestinal lumen. In this system, three Clos-tridium species were isolated [15]. Noteworthy, C. clostridioformeand C. malenominatumwere observed in the vessels and in biofilmformed on the vessel wall whereas C. perfringens was only found inthe vessels. In this model, C. perfringens was therefore not able toform a biofilm. In contrast, in another model, C. perfringens wasreported to be part of a biofilm formed in a drip-flow reactor [18].This suggests that C. perfringens ability to form a biofilm depends onthe environmental conditions. In the drip-flow reactor model,C. perfringens survival was due to the presence of two other bac-terial species, Pseudomonas aeruginosa and Staphylococcus aureus.Indeed, these two bacteria allowed the C. perfringens anaerobicgrowth in the presence of an aerobic atmosphere, probably byutilizing the oxygen molecules. C. perfringens biofilm was detectedin many types of multi-species biofilm including biliary stents [19](Table 1). Other Clostridium species were observed forming biofilmsoutside of the gut during infections. For instance, 6 species ofClostridia including Clostridium bifermentans and Clostridium sor-delli are found to contribute to the super infection associated withtungiasis, forming a biofilm within the epidermidis [20].

Thus Clostridium species, at least for those presented here, maygrow inmulti-species biofilms in environment and in vivo (Table 1).However, their role in biofilm formation is not clear. The questionwhether they are actively involved in biofilm formation or recruitedinto a pre-existing biofilm remains to be answered. Alternatively,survival of C. perfringens in the presence of oxygenwas allowed in amulti-species biofilm due to other bacteria, suggesting that inparticular cases Clostridium species may take advantage of alreadyformed biofilm structures for growth. This observation may alsohighlight the opportunity of Clostridium as anaerobes to grow inunusual environments like in aerobic conditions. Multi-speciesbiofilms may be robust biofilms. For instance, C. perfringens, Clos-tridium baratii, C. bifermentans, Clostridium fallax and C. difficilewere isolated from explanted biliary stents and were furtherstudied in vitro as mono- and dual-species biofilms [21]. Mono-species biofilms were formed in 96 well-plates and analysed us-ing violet crystal staining and by field emission scanning electronmicroscopy. They all showed a strong biofilm formation, except for

C. difficile, which only formed a moderate biofilm. According to thisstudy, C. difficile appears to be a poor mono-species biofilm former[21]. In multi-species conditions, C. difficile was not contributing tothe formation of biofilm in a chemostat gut model [22], as nogermination and multiplication was observed in this model. How-ever, Donelli et al. described that C. difficile formed a strong biofilmwhen associated with a Finegoldia magna strain [21], suggestingthat they were able to cooperate. A scanning electron microscopystudy showed a strong interaction between the two bacteriapossibly mediated by a matrix. Thus the multi-species compositionmay be important for C. difficile growth in biofilm, suggesting thatdepending on its composition, microbiota could promote survivaland multiplication of C. difficile in the gut as part of a multi-speciesbiofilm.

4. Biofilm formed in vitro: C. difficile and C. perfringens

The previous presented reports highlight the possibility thatsome pathogenic Clostridium species develop or participate in vivoin a biofilm, like C. perfringens or C. difficile (Table 1). However, thesetwo species were also shown to formmono-species biofilms in vitro[23e27]. Most of the studies reported concern C. difficile andC. perfringens, probably because they are pathogens. Indeed,C. difficile is responsible for 15e20% of post-antibiotic diarrhoea andnearly all cases of pseudomembranous colitis (for a review seeRef. [28]) and C. perfringens is a major cause of enteric diseases dueto the production of several toxins (for review see Ref. [29]). C.difficile and C. perfringens were recently studied regarding biofilmformation and several mutants were tested [23e27]. C. difficile andC. perfringens bacteria formed different biofilms according to thestrains and conditions for culture. Biofilms of both species areformed at the bottom of microplate wells at 37 �C. Moreover, at25 �C, C. perfringens was described to develop a non-adherentpellicle, which is a biofilm-like structure. Interestingly, the mainvirulence factors of C. difficile, toxins TcdA and TcdB, were identifiedin the C. difficile matrix of 3 and 6 day old biofilms [26]. Similarly,several toxin genes in C. perfringens biofilms were expressed andtightly controlled by CtrAB (toxin regulator), Spo0A, and AbrB [25].The toxin presence in both biofilms formed by C. difficile andC. perfringens highlights the possibility of a role in virulence ifbiofilms are formed in vivo. However, no in vivo studies have beenreported. The following part of this review will focus on the bio-films formed by these two bacteria.

5. Biofilm cohesion and resistance

Biofilm cohesion is mainly due to the presence of the matrix,which is a key structure of biofilms. The matrix common feature ofbacterial biofilms is the presence of extracellular polymeric sub-stance (EPS) that may be composed of proteins, DNA and poly-saccharides. The C. perfringens biofilm matrix is at least composedof proteins including type IV pili, DNA, and may contain poly-saccharides [25,27]. In C. difficile, the matrix is composed of poly-saccharide II, a surface carbohydrate polymer found in all strains,proteins and DNA [24,30]. Polysaccharide II is a common antigenfound at the surface of all C. difficile species [30]. The polysaccharideII or another carbohydratewas detected using concanavalin A in thematrix of several strains of C. difficile [26]. Moreover, a similarprotein profile, found in the matrix of 7 different C. difficile strains,was analysed and showed the presence of 7 proteins involved inmetabolism: formate-tetrahydrofolate ligase, acetyl-CoA acetyl-transferase, 2-hydroxyisocaproate CoA-transferase, NAD-specificglutamate dehydrogenase, 3-hydroxybutyryl-CoA dehydrogenase,fructose-bisphosphate aldolase and formate-tetrahydrofolateligase [26]. The presence of these proteins suggests that a high

V. Pantal�eon et al. / Anaerobe 30 (2014) 193e198196

level of metabolism occurs in biofilms and that the lysis rate withinthe biofilm may be important. Lysis may also reflect a cannibalisticphenomenon within biofilm [31].

On the other hand, biofilm allows protection of bacteria fromtoxic compounds. This was demonstrated in C. difficile and inC. perfringens. PCR-ribotype 027 C. difficile strain R20291 in a biofilmdisplayed increased survival in the presence of subinhibitory andinhibitory concentration of antibiotic. It is noteworthy that in the630 strain a vanGCd cluster has been reported to be functional andexpressed, but was not able to promote vancomycin resistance [32].This cluster is also present in the R20291 strain and may contributeto the observed survival in biofilm in presence of vancomycin.Moreover, C. difficile biofilms increased the bacterial resistance tometronidazole up to 100-fold [26]. The biofilm is also able to pro-tect the R20291 C. difficile strain from oxygen stress [23] and thisphenomenon is accentuated in a spo0A mutant. RegardingC. perfringens that is tolerant to oxygen, its biofilm allowed a sur-vival increase in the presence of H2O2 [27]. Thus, biofilms of Clos-tridium species may protect the embedded bacteria from toxicmolecules coming from the environment.

6. Role of the motility and bacterial surface in biofilmformation

Motility is required for the early stage of biofilm formation formany bacterial species and, therefore a flagella mutant (maincomponent or motor) is impaired for biofilm formation (for reviewsee Ref. [33]). Motility is thereafter inhibited in the mature biofilm.fliC encodes the main structural component of the flagella filamentin C. difficile. Ethapa et al. described in C. difficile that, surprisingly,the fliC mutant is impaired in the later stages of biofilm formation(5 days) [24]. As already shown in B. cereus, this may be related tothe potential of some bacterial swimmers to form transient poreswhich increase nutrient flow and improve the bacterial fitness intothe biofilm [34]. Thus, the motility of C. difficile in mature biofilmsmay be crucial. Other such regulatory processes driving motility inbiofilms are described below (Section 7). Concerning C. perfringens,mutants in pilC and pilT (encoding components of type IV pili) areimpaired for motility and biofilm formation [27].

The bacterial surface interacts directly or through the matrix tothe adhesion surface. A change of this structure may thereforemodify many processes in biofilm formation. The impact of the S-layer through the study of Cwp84 was therefore investigated inC. difficile. Cwp84 is a cysteine protease that harbours a Cwpdomain (cell wall-binding motif; pfam 04122) and is involved in S-layer maturation. Its absence leads to a non-cleaved SlpA (uniquecomponent of the S-layer) resulting in aberrant retention of someCwp proteins [35]. A cwp84mutant in the C. difficile 630Derm straingrew more slowly than the parental strain and had a propensity toaggregate [36]. In contrast, Ethapa et al. showed that the cwp84mutant in the R20291 background grew at the same level of theparental strain and is impaired for biofilm formation. Aggregation isoften linked to the capacity to form a biofilm, but this was notinvestigated by Ethapa et al. [24]. The apparent contradiction be-tween these two results may be due to the different strains (R20291and 630Derm) used in each study.

7. Sporulation and germination in biofilm

Obana et al. showed in C. perfringens that the regulators Spo0Aand CtrAB are required for non-adherent pellicle formation(observed at 25 �C) whereas the regulator AbrB is required forbiofilm formation (observed at 37 �C) [25]. Indeed, theC. perfringens spo0A mutant was not impaired in biofilm formationat 37 �C. In contrast, in C. difficile, Spo0A was involved in biofilm

formation at this temperature [23,24,37]. Spo0A is the masterregulator of sporulation of Gram positive sporulating bacteria andwhen phosphorylated, Spo0A modifies the transcription of morethan 500 genes in Bacillus subtilis [38]. Similarly, in B. subtilis, aspo0A mutant is impaired for sporulation and biofilm formation.The number of viable bacteria in the C. difficile and B. subtilis bio-films for the corresponding spo0A mutant is similar to the parentalstrain, suggesting a real role for the Spo0A regulation pathway inbiofilm formation [23,24,39]. As a major difference, C. difficile pos-sesses neither the tasA gene nor the exopolysacharide locus(encoding matrix components) that are regulated by the spo0Aphosphorelay in B. subtilis. This suggests that the regulationmediated through the spo0A pathway may be similar betweenB. subtilis and C. difficile, yet the targeted genes are different.Another result supports the role of Spo0A in biofilm formation; thecomplemented C. difficile spo0Amutant was not able to sporulate atthe same level although forming a similar biofilm as the parentalstrain. The sporulation rate was suggested to be at a higher levelwhen C. difficile is grown in biofilm formed on polycarbonatemembranes than on a plastic surface [26]. However, the sporequantity within a 6 day old biofilm varied from 10% to more than60% depending on the strains. Moreover, spores harboured unusualstructures interacting with the coat as a shroud [26]. Interestingly,the germination rate is lower in the spores isolated from the biofilmin comparison with the spores purified in planktonic culture. Thislower rate of germination may be due to the presence of thedescribed shroud present around the spores formed within thebiofilm. This may prevent germination in the biofilm. However,germination may also impact the biofilm formation. Indeed, a sleCmutant affects biofilm formation of C. perfringens. SleC was firstdescribed in C. perfringens [40] and is a protein involved in thespore germination process especially in the cortex lysis, in thisspecies and in C. difficile [41,42]. A sleC mutant formed biofilm-likestructures, yet it was not able to form a biofilm as thick as theC. difficile R20291 parental strain [24]. But, intriguingly, the sleCmutant spore in the R20291 strain is affected by heat, in contrast tothe sleC mutant of the 630Derm strain [41].

8. Regulation of mono-species biofilm

Compiling the data from Clostridium species grown in biofilm,some common features may be reported. Whereas it is believedthat nutrient limitation induces biofilm formation [43], the majorcarbon source allowing biofilm formation in Clostridium species isglucose. Cellulose is a polymer of glucose and is the main resourcefor Clostridium species grown in industrial biofilms [11]. Moreover,most mono-species biofilm studies used media containing glucose.Glucose is even added in very rich media for C. difficile biofilmformation [24] and was shown to promote biofilms of most strainsof C. perfringens except the SM101 strain [27]. In C. perfringens, ccpAencodes a key regulator of the response to carbohydrate limitationsuch as glucose. CcpA (catabolite control protein) was shown to be aglobal transcription factor involved in the regulation of virulence ofmany bacterial species. It forms a complex with the Hpr phos-phorylated protein, allowing CcpA to bind DNA. In B. subtilis, itnegatively controls several operons involved in catabolism of sec-ondary carbon sources and represses biofilm formation in thepresence of high concentrations of glucose [44]. Varga et al.investigated therefore whether ccpA was essential for biofilm for-mation of C. perfringens [27]. In the SM101 C. perfringens strain,CcpA played a role in regulating the ratio of biofilm to planktoniccells at low glucose concentration. However, in contrast with theother tested strains in the study, the SM101 strain formed the bestbiofilm in the absence of glucose. The role of CcpA in these otherC. perfringens strains was therefore not clear. ccpA has been studied

V. Pantal�eon et al. / Anaerobe 30 (2014) 193e198 197

in several Clostridium species such as C. difficile and C. cellulolyticum[45]. It has been observed that CcpA can be involved in the regu-lation of toxin production, sporulation, and heat stress responsegenes [27,45,46]. It may be interesting to investigate the impact of accpA mutant in those Clostridium species biofilms.

Quorum sensing was also shown to be involved in biofilm for-mation. In C. difficile, LuxS is involved in the synthesis of the autoinducer 2 (AI-2) that is driving quorum sensing-mediated regula-tion. AI-2 has been shown to control a variety of cellular processes(for review see Ref. [47]). When luxS is inactivated in most bacteriabiofilm formation is affected. Similarly in C. difficile a luxS mutantwas not able to form a single bacterial layer on a glass surface [24].Yet controversially, LuxS may also be involved in toxin regulation[48,49].

Other regulatory mechanisms have been described in bacterialbiofilms. Bis-(30-50)-cyclic dimeric guanosine monophosphate [50](c-di-GMP) is a bacterial secondary messenger controlling diversebacterial phenotypes mostly known to be involved in the transitionfrom free-living, motile to biofilm lifestyle in Gram-negative bac-teria. The balance of c-di-GMP is regulated by its synthesis anddegradation. c-di-GMP is synthesized by several diguanylate cy-clases that harbour a GGDEF domain. This molecule is degraded byseveral phosphodiesterases that harbour either an EAL (PDEA) orHD-GYP domain. Some enzymes may have both activities and aretherefore able to degrade and synthesize c-di-GMP molecules. InC. difficile, up to 37 genes encode enzymes harbouring a GGDEF oran EAL domain. 31 genes are conserved among C. difficile species[50]. Among them dccA (CD1420, diguanylate cylclase) has beenoverexpressed in C. difficile 630, increasing c-di-GMP concentration,thus inducing aggregation through riboswitches (type I and II), andleading to a decreased motility in C. difficile [51]. Soutourina et al.confirmed these results and demonstrated that a high concentra-tion of c-di-GMP increased biofilm formation [52]. They alsoshowed that, when c-di-GMP concentrationwas increased, flgB andfliC gene expression was reduced whereas CD3513 (encoding aputative pilin) and CD2831 (encoding a precursor of a collagen-binding protein) were positively regulated. These results mayexplain why biofilm formation is increased with a higher concen-tration of c-di-GMP. Noteworthy, although motility is inhibited viathe c-di-GMP regulation pathway it remains necessary for C. difficilebiofilm maturation, as mentioned before [24]. c-di-GMP inC. difficile regulates a putative pilin (CD3513) potentially involved inadhesion, investigation whether the C. difficile pilin influencesadhesion and biofilm may therefore be interesting.

9. Conclusion

Many studies are focused on the biofilms formed by multiplebacteria (Table 1), but they reveal little data regarding biofilmformation and structures of Clostridium species. Most of the clos-tridial studies regarding mono-species biofilms are in contrastfocused on two major pathogenic bacteria, C. difficile andC. perfringens. Many results on regulation, sporulation, and biofilmresistance have been obtained in these studies, however, it seemshazardous to extrapolate for all the Clostridium species. All thestudies highlight that Clostridium species are probably able to growin a multi-species biofilms. The high level of regulation withinC. difficile and C. perfringens biofilms, already described in otherFirmicutes, suggests a high possibility of adaptation during envi-ronmental stresses. Moreover, C. difficile and C. perfringens bio-films, if formed in vivo, are able to produce toxins that mayparticipate in virulence and produce spores that may promote newinfections, or even relapses for C. difficile. Thus, biofilms for thesetwo species may be involved in environmental adaptation andresistance, virulence and spreading. Deciphering the role of

biofilm formation in the pathogenesis of C. difficile andC. perfringenswill be an important challenge for the next few years,and may modify treatments and prevention of diseases related tothese bacteria.

Acknowledgements

V.P. and A-P.S. were funded by a doctoral fellowship from theFrench Ministry of Higher Education and Research.

References

[1] Hall-Stoodley L, Costerton JW, Stoodley P. Bacterial biofilms: from the naturalenvironment to infectious diseases. Nat Rev Microbiol 2004;2:95e108.

[2] Ludwig W, Karl-Heinz S, Whitman W. Revised road map to the phylum fir-micutes. In: Paul De Vos GMG, Jones Dorothy, Krieg Noel R, Ludwig Wolfgang,Rainey Fred A, Schleifer Karl-Heinz, et al., editors. Bergey's Manual® of sys-tematic bacteriology. springer; 2009.

[3] Ishii S, Shimoyama T, Hotta Y, Watanabe K. Characterization of a filamentousbiofilm community established in a cellulose-fed microbial fuel cell. BMCMicrobiol 2008;8:6.

[4] Rademacher A, Zakrzewski M, Schluter A, Schonberg M, Szczepanowski R,Goesmann A, et al. Characterization of microbial biofilms in a thermophilicbiogas system by high-throughput metagenome sequencing. FEMS MicrobiolEcol 2012;79:785e99.

[5] Biswas K, Turner SJ. Microbial community composition and dynamics ofmoving bed biofilm reactor systems treating municipal sewage. Appl EnvironMicrobiol 2012;78:855e64.

[6] Zhang F, Ding J, Shen N, Zhang Y, Ding Z, Dai K, et al. In situ hydrogenutilization for high fraction acetate production in mixed culture hollow-fibermembrane biofilm reactor. Appl Microbiol Biotechnol 2013;97:10233e40.

[7] Liu D, Chen Y, Ding FY, Zhao T, Wu JL, Guo T, et al. Biobutanol production in aClostridium acetobutylicum biofilm reactor integrated with simultaneousproduct recovery by adsorption. Biotechnol Biofuels 2014;7:5.

[8] Napoli F, Olivieri G, Russo ME, Marzocchella A, Salatino P. Butanol productionby Clostridium acetobutylicum in a continuous packed bed reactor. J IndMicrobiol Biotechnol 2010;37:603e8.

[9] Raganati F, Olivieri G, Procentese A, Russo ME, Salatino P, Marzocchella A.Butanol production by bioconversion of cheese whey in a continuous packedbed reactor. Bioresour Technol 2013;138:259e65.

[10] Jain A, Morlok CK, Henson JM. Comparison of solid-state and submerged-statefermentation for the bioprocessing of switchgrass to ethanol and acetate byClostridium phytofermentans. Appl Microbiol Biotechnol 2013;97:905e17.

[11] Dumitrache A, Wolfaardt G, Allen G, Liss SN, Lynd LR. Form and function ofClostridium thermocellum biofilms. Appl Environ Microbiol 2013;79:231e9.

[12] Wang ZW, Lee SH, Elkins JG, Morrell-Falvey JL. Spatial and temporal dynamicsof cellulose degradation and biofilm formation by Caldicellulosiruptor obsi-diansis and Clostridium thermocellum. AMB Express 2011;1:30.

[13] Swidsinski A, Weber J, Loening-Baucke V, Hale LP, Lochs H. Spatial organi-zation and composition of the mucosal flora in patients with inflammatorybowel disease. J Clin Microbiol 2005;43:3380e9.

[14] Fite A, Macfarlane S, Furrie E, Bahrami B, Cummings JH, Steinke DT, et al.Longitudinal analyses of gut mucosal microbiotas in ulcerative colitis inrelation to patient age and disease severity and duration. J Clin Microbiol2013;51:849e56.

[15] Macfarlane S, Woodmansey EJ, Macfarlane GT. Colonization of mucin byhuman intestinal bacteria and establishment of biofilm communities in atwo-stage continuous culture system. Appl Environ Microbiol 2005;71:7483e92.

[16] Sharp R, Macfarlane GT. Chemostat enrichments of human feces with resistantstarch are selective for adherent butyrate-producing clostridia at high dilutionrates. Appl Environ Microbiol 2000;66:4212e21.

[17] Larue R, Yu Z, Parisi VA, Egan AR, Morrison M. Novel microbial diversityadherent to plant biomass in the herbivore gastrointestinal tract, as revealedby ribosomal intergenic spacer analysis and rrs gene sequencing. EnvironMicrobiol 2005;7:530e43.

[18] Woods J, Boegli L, Kirker KR, Agostinho AM, Durch AM, Delancey Pulcini E,et al. Development and application of a polymicrobial, in vitro, wound biofilmmodel. J Appl Microbiol 2012;112:998e1006.

[19] Leung JW, Liu Y, Chan RC, Tang Y, Mina Y, Cheng AF, et al. Early attachment ofanaerobic bacteria may play an important role in biliary stent blockage.Gastrointest Endosc 2000;52:725e9.

[20] Feldmeier H, Heukelbach J, Eisele M, Sousa AQ, Barbosa LM, Carvalho CB.Bacterial superinfection in human tungiasis. Trop Med Int Health 2002;7:559e64.

[21] Donelli G, Vuotto C, Cardines R, Mastrantonio P. Biofilm-growing intestinalanaerobic bacteria. FEMS Immunol Med Microbiol 2012;65:318e25.

[22] Crowther GS, Chilton CH, Todhunter SL, Nicholson S, Freeman J, Baines SD,et al. Development and validation of a chemostat gut model to study bothplanktonic and biofilm modes of growth of Clostridium difficile and humanmicrobiota. PloS one 2014;9:e88396.

V. Pantal�eon et al. / Anaerobe 30 (2014) 193e198198

[23] Dawson LF, Valiente E, Faulds-Pain A, Donahue EH, Wren BW. Characterisa-tion of Clostridium difficile biofilm formation, a role for Spo0A. PloS one2012;7:e50527.

[24] Ethapa T, Leuzzi R, Ng YK, Baban ST, Adamo R, Kuehne SA, et al. Multiplefactors modulate biofilm formation by the anaerobic pathogen Clostridiumdifficile. J Bacteriol 2013;195:545e55.

[25] Obana N, Nakamura K, Nomura N. A sporulation factor is involved inmorphological change of Clostridium perfringens biofilm in response to tem-perature. J Bacteriol 2014 Apr;196(8):1540e50. http://dx.doi.org/10.1128/JB.01444-13 [Epub 2014 Feb 7].

[26] Semenyuk EG, Laning ML, Foley J, Johnston PF, Knight KL, Gerding DN, et al.Spore formation and toxin production in Clostridium difficile biofilms. PloS one2014;9:e87757.

[27] Varga JJ, Therit B, Melville SB. Type IV pili and the CcpA protein are needed formaximal biofilm formation by the gram-positive anaerobic pathogen Clos-tridium perfringens. Infect Immun 2008;76:4944e51.

[28] Blanckaert K, Coignard B, Grandbastien B, Astagneau P, Barbut F. Update onClostridium difficile infections. Rev Med Interne 2008;29:209e14.

[29] Smedley 3rd JG, Fisher DJ, Sayeed S, Chakrabarti G, McClane BA. The enterictoxins of Clostridium perfringens. Rev Physiol Biochem Pharmacol 2004;152:183e204.

[30] Ganeshapillai J, Vinogradov E, Rousseau J, Weese JS, Monteiro MA. Clostridiumdifficile cell-surface polysaccharides composed of pentaglycosyl and hex-aglycosyl phosphate repeating units. Carbohydr Res 2008;343:703e10.

[31] Lopez D, Vlamakis H, Losick R, Kolter R. Cannibalism enhances biofilmdevelopment in Bacillus subtilis. Mol Microbiol 2009;74:609e18.

[32] Ammam F, Meziane-Cherif D, Mengin-Lecreulx D, Blanot D, Patin D,Boneca IG, et al. The functional vanGCd cluster of Clostridium difficile does notconfer vancomycin resistance. Mol Microbiol 2013;89:612e25.

[33] Guttenplan SB, Kearns DB. Regulation of flagellar motility during biofilmformation. FEMS Microbiol Rev 2013;37:849e71.

[34] Houry A, Gohar M, Deschamps J, Tischenko E, Aymerich S, Gruss A, et al.Bacterial swimmers that infiltrate and take over the biofilm matrix. Proc NatlAcad Sci U S A 2012;109:13088e93.

[35] de la Riva L, Willing SE, Tate EW, Fairweather NF. Roles of cysteine proteasesCwp84 and Cwp13 in biogenesis of the cell wall of Clostridium difficile.J Bacteriol 2011;193:3276e85.

[36] Kirby JM, Ahern H, Roberts AK, Kumar V, Freeman Z, Acharya KR, et al. Cwp84,a surface-associated cysteine protease, plays a role in the maturation of thesurface layer of Clostridium difficile. J Biol Chem 2009;284:34666e73.

[37] Heap JT, Pennington OJ, Cartman ST, Carter GP, Minton NP. The ClosTron: auniversal gene knock-out system for the genus Clostridium. J MicrobiolMethods 2007;70:452e64.

[38] Molle V, Fujita M, Jensen ST, Eichenberger P, Gonzalez-Pastor JE, Liu JS, et al.The Spo0A regulon of Bacillus subtilis. Mol Microbiol 2003;50:1683e701.

[39] Hamon MA, Lazazzera BA. The sporulation transcription factor Spo0A isrequired for biofilm development in Bacillus subtilis. Mol Microbiol 2001;42:1199e209.

[40] Miyata S, Moriyama R, Miyahara N, Makino S. A gene (sleC) encoding a spore-cortex-lytic enzyme from Clostridium perfringens S40 spores; cloning,sequence analysis and molecular characterization. Microbiology 1995;141(Pt10):2643e50.

[41] Burns DA, Heap JT, Minton NP. SleC is essential for germination of Clostridiumdifficile spores in nutrient-rich medium supplemented with the bile salttaurocholate. J Bacteriol 2010;192:657e64.

[42] Gutelius D, Hokeness K, Logan SM, Reid CW. Functional analysis of SleC fromClostridium difficile: an essential lytic transglycosylase involved in sporegermination. Microbiology 2014;160:209e16.

[43] Allan VJ, Callow ME, Macaskie LE, Paterson-Beedle M. Effect of nutrient lim-itation on biofilm formation and phosphatase activity of a Citrobacter spp.Microbiology 2002;148:277e88.

[44] Tobisch S, Zuhlke D, Bernhardt J, Stulke J, Hecker M. Role of CcpA in regulationof the central pathways of carbon catabolism in Bacillus subtilis. J Bacteriol1999;181:6996e7004.

[45] Antunes A, Camiade E, Monot M, Courtois E, Barbut F, Sernova NV, et al. Globaltranscriptional control by glucose and carbon regulator CcpA in Clostridiumdifficile. Nucleic Acids Res 2012;40:10701e18.

[46] Varga J, Stirewalt VL, Melville SB. The CcpA protein is necessary for efficientsporulation and enterotoxin gene (cpe) regulation in Clostridium perfringens.J Bacteriol 2004;186:5221e9.

[47] Federle MJ, Bassler BL. Interspecies communication in bacteria. J Clin Invest2003;112:1291e9.

[48] Carter GP, Purdy D, Williams P, Minton NP. Quorum sensing in Clostridiumdifficile: analysis of a luxS-type signalling system. J Med Microbiol 2005;54:119e27.

[49] Lee AS, Song KP. LuxS/autoinducer-2 quorum sensing molecule regulatestranscriptional virulence gene expression in Clostridium difficile. BiochemBiophys Res Commun 2005;335:659e66.

[50] Bordeleau E, Fortier LC, Malouin F, Burrus V. c-di-GMP turn-over in Clos-tridium difficile is controlled by a plethora of diguanylate cyclases and phos-phodiesterases. PLoS Genet 2011;7:e1002039.

[51] Purcell EB, McKee RW, McBride SM, Waters CM, Tamayo R. Cyclic diguanylateinversely regulates motility and aggregation in Clostridium difficile. J Bacteriol2012;194:3307e16.

[52] Soutourina OA, Monot M, Boudry P, Saujet L, Pichon C, Sismeiro O, et al.Genome-wide identification of regulatory RNAs in the human pathogenClostridium difficile. PLoS Genet 2013;9:e1003493.

[53] Signoretti FG, Endo MS, Gomes BP, Montagner F, Tosello FB, Jacinto RC.Persistent extraradicular infection in root-filled asymptomatic human tooth:scanning electron microscopic analysis and microbial investigation afterapical microsurgery. J Endod 2011;37:1696e700.

[54] Ren Z, Steinberg LM, Regan JM. Electricity production and microbial biofilmcharacterization in cellulose-fed microbial fuel cells. Water Sci Technol2008;58:617e22.

[55] Dumitrache A, Wolfaardt GM, Allen DG, Liss SN, Lynd LR. Tracking the cellu-lolytic activity of Clostridium thermocellum biofilms. Biotechnol Biofuels2013;6:175.

[56] Nagy KN, Sonkodi I, Szoke I, Nagy E, Newman HN. The microflora associatedwith human oral carcinomas. Oral Oncol 1998;34:304e8.

[57] Weiss S, Lebuhn M, Andrade D, Zankel A, Cardinale M, Birner-Gruenberger R,et al. Activated zeoliteesuitable carriers for microorganisms in anaerobicdigestion processes? Appl Microbiol Biotechnol 2013;97:3225e38.