ORIGINAL ARTICLE

Analysis of Microbial Community during BiofilmDevelopment in an Anaerobic WastewaterTreatment Reactor

Nuria Fernández & Emiliano Enrique Díaz &

Ricardo Amils & José L. Sanz

Received: 23 April 2007 /Accepted: 2 October 2007 /Published online: 23 November 2007# Springer Science + Business Media, LLC 2007

Abstract The formation, structure, and biodiversity of amultispecies anaerobic biofilm inside an Upflow AnaerobicSludge Bed (UASB) reactor fed with brewery wastewater wasexamined using complementary microbial ecology methodssuch us fluorescence in situ hybridization (FISH), denaturinggradient gel electrophoresis (DGGE), and cloning. Thebiofilm development can be roughly divided into threestages: an initial attachment phase (0–36 h) characterized byrandom adhesion of the cells to the surface; a consolidationphase (from 36 h to 2 weeks) defined by the appearance ofmicrocolonies; and maturation phase (from 2 weeks to2 months). During the consolidation period, proteobacteriawith broad metabolic capabilities, mainly represented bymembers of alpha-Proteobacteria class (Oleomonas,Azospirillum), predominated. Beta-, gamma-, delta- (bothsyntrophobacteria and sulfate-reducing bacteria) and epsilon-(Arcobacter sp.) Proteobacteria were also noticeable.Archaea first appeared during the consolidation period. AMethanospirillum-like methanogen was detected after 36 h,and this was followed by the detection of Methanosarcina,after 4 days of biofilm development. The mature biofilmdisplayed a hill and valley topography with cells embeddedin a matrix of exopolymers where the spatial distribution ofthe microorganisms became well-established. Compared to

the earlier phases, the biodiversity had greatly increased.Although alpha-Proteobacteria remained as predominant,members of the phyla Firmicutes, Bacteroidete, and Ther-motogae were also detected. Within the domain Archaea, theacetoclastic methanogen Methanosaeta concilii becomedominant. This study provides insights on the trophic weband the shifts in population during biofilm development inan UASB reactor.

Introduction

Biofilms are structured microbial communities made up ofgroups of cells suspended in a self-produced hydratedpolymeric matrix of variable density and permeated bychannels [13, 14]. In most natural and engineered environ-ments, a multispecies microbial community is the prevail-ing life form [58]. Although many species are implicated,biofilm development has been mainly studied usingsystems composed by one or two species which have ledthe formulation of a development model, in which theformation of biofilms occurs in multiple steps: (1) approachof microbes to a surface, (2) initial attachment mainlygoverned by van der Waals and electrostatic forces; (3)formation of microcolonies; and (4) biofilm maturation inwhich microcolonies are stabilized by increased cell-surfaceadhesion due to the accumulation of extracellular polymersubstance (EPS) [21, 51, 57, 58]. Mature biofilms show acomplex structure (the mushroom or the tulip model) withextensive presence of extracellular polymeric substances(EPS) full of channels through which a liquid phase is freeto move [34, 60].

Biofilms play an important role in wastewater treatmentas they form the basis of diverse aerobic and anaerobicreactors (trickling filters, rotating biological contactors...)

Microb Ecol (2008) 56:121–132DOI 10.1007/s00248-007-9330-2

N. Fernández : E. E. Díaz : R. Amils : J. L. SanzDepartment of Molecular Biology,Universidad Autónoma de Madrid,Madrid, 28049, Spain

J. L. Sanz (*)Dpto. de Biología Molecular, Universidad Autónoma de Madrid,Campus de Cantoblanco, Ctra. de Colmenar, Km.15,Madrid, C.P., 28049, Spaine-mail: joseluis.sanz@uam.es

and are characterized by their feasibility and efficiency.Pollutants are anaerobically processed and eliminated bymeans of the complex food chain established within thebiofilm [30]. Consequently, the process efficiency is theresult of the biofilm microbial diversity. Nevertheless,studies of biofilm growth in wastewater treatment systemshave focused mainly on the influence of operationalparameters, physicochemical factors, and the properties ofthe supports on biofilm development [26, 56]. On the otherhand, activity, adhesion, biomass, and other conventionalparameters have been measured to assess the microbialcommunity [22, 29, 52]. Reports based on microbialecology techniques are scarce [8, 19, 49].

Conventional cultivation-dependent microbiologicaltechniques fail to give an indication of biodiversity (about99% of the bacterial cells in biofilms cannot be cultured onstandard media [60]) or the architecture of a biofilm.During the last 15 years, molecular techniques based on16S rRNA/rDNA have been successfully applied tomicrobial ecology research. Denaturing gradient gel elec-trophoresis (DGGE) [39, 40], fluorescence in situ hybrid-ization (FISH) [7, 50, 54], together with molecular cloning,have opened up new perspectives for the study of microbialecosystems [5]. FISH combined with confocal laserscanning microscopy (CLSM) makes it possible to studybiofilms in natural [34], industrial [35], and engineered [15]environments without destroying their critical architecture.An excellent review of the application of molecular ecologytechniques to wastewater treatment systems has recentlybeen published [46].

The aim of this study was to investigate in detail theformation of a multispecies anaerobic biofilm inside areactor. The upflow anaerobic sludge bed (UASB) reactorwas chosen as it is widely used throughout the world.Scanning electron microscopy (SEM) and CLSM combinedwith FISH were used to monitor the development andstructure of the biofilm. Biodiversity was evaluated withthree rRNA-based complementary methods: FISH, DGGE,and cloning and sequencing of 16S rDNA.

Methods

Experimental Set-up

A laboratory scale UASB reactor (0.9 l) was operated for5 months. Intact and crushed granular sludge (2.5 g VSS l−1)from a full-scale UASB reactor treating brewery wastewater(MAHOU, Guadalajara, Spain) was used as inoculum. Thereactor was fed with industrial wastewater from a brewery(4,000 mg l−1 COD; 2,400 mg l−1 BOD; 1.3 g l−1 TSS;100 mg l−1 TNK; 15 mg l−1 NH4

+, 15 mg l−1 P), pH 7.0, andoperated with a hydraulic retention time of 24 h at 30°C. The

typical composition of brewery wastewater is (in mM per gof COD): 1.6 acetate, 1 propionate, 6–8 ethanol, and lessthan 0.2 of butyrate, lactate, succinate, and glucose. TheVFA in the effluent after UASB treatment is around 10% oftheir content in the inffluent [23, 61].

Once the reactor reached steady state (organic loadingrate: 1 g COD l−1 day−1; sludge loading rate 0.35 g COD gVSS−1 day−1; efficiency of COD removal: 85–90%, novolatile fatty acids accumulated), biofilms were allowed todevelop on vertical glass slides submerged in the reactor.

Growth Curve

Once the reactor reached steady state, a growth curveduring the 10 first days of biofilm development wasobtained to provide insights on the initial period of biofilmdevelopment. At the beginning of the experiment, severalFalcon 3911 MicroTest III assay plates were verticallyplaced inside the reactor. Subsequently, three of them wereremoved at each sampling time. The biofilm formation wasdetermined by optical density at 560 nm as previouslydescribed [25].

Fluorescent in situ Hybridization

Supports consisting of glass slides with 10-wells (75×25 mm, with wells of 5 mm diameter) (Superior, Marienfeld,Germany) were hung vertically by a nylon line at adistance of a few centimeters over the sludge bed. All theslides were introduced at the start of the biofilm formationexperiment, and they were removed at different times (1,2, 4, 12, 24, and 36 h, and 2, 4, 8, 18, 25, 30 and 60 days)for in situ hybridization. Samples were rinsed with filteredwater to remove loosely attached planktonic forms andimmediately fixed with ethanol for Gram-positive bacteriadetection or with 4% paraformaldehyde in phosphatebuffered saline solution (PBS) during 4 h at 4°C forGram-negative bacteria. The samples were then washed inPBS, and stored in PBS: Ethanol (1:1) solution at −20°C.All samples were further dehydrated by immersion in 50,80, and 100% ethanol solutions for 3 min each time.Hybridization was performed following the protocol de-scribed elsewhere [3, 6, 32]. The probes used in this workare listed in Table 1. The NON338 probe was used asnegative control. The total cells present in the samples weredetermined by direct counting of 4′,6′-diamin phenylindol(DAPI, 1 mg/ml) stained cells when possible. Samples wereexamined under a Zeiss Axiovert 200 microscope. Struc-tural and morphological studies on intact biofilms werecarried out with a Confocal Radiance 2000 scanning systemcoupled to a Zeiss Axiovert S100 TV confocal laser-scanning microscope.

122 N. Fernández et al.

Scanning Electron Microscopy

Samples for scanning electron microscopy were prepared asfollows: biofilms were grown on glass slides (Φ 0.5 cm,FEDELCO, ER-308). The location of the slides andsampling protocol were similar to FISH procedure. Oncea sample was taken out from the reactor, it was fixed withglutaraldehyde (2.5% v/v) in 0.2 M sodium cacodylatebuffer (pH 7.1) and dehydrated with graded ethanolsolutions (10, 30, 50, 70, 90, and 100% ethanol). Thesamples were dehydrated by the critical point dryingmethod and coated with gold. Micrographs were takenwith a Phillips XL30 EDAX DX4i SEM.

DNA Extraction and 16S rRNA Amplification

Biofilms, 2- and 60-day old were detached from the slidesusing Triton X-100 (0.25%) solution. DNA was extractedusing FastDNA kit for soils BIO101 according to themanufacturer’s protocol. The 16S rRNA genes from mixedmicrobial DNAwere amplified by polymerase chain reaction(PCR). To obtain almost complete 16S rRNA gene, twooligonucleotide primer pairs were used: 27F and 1492R(annealing T: 56°C) for the domain Bacteria [28] and 25Fand 1492R (annealing T: 52°C) for the domain Archaea[28]. For subsequent DGGE analysis, a fragment of DNAwas amplified with two primer pairs: 341F-GC and 907R(annealing T: 52°C) for the domain Bacteria [11] and 622F-GC and 1492R (annealing T: 42°C) for the domain Archaea[12] (GC clamp: 5′-CGC CCG CCG CGC CCC GCG CCCGTC CCG CCC CCG CCC-3′). PCR reactions were

performed with the following thermocycler program: pre-denaturation at 94°C for 5 min, 30 cycles of denaturation at94°C for 1 min, annealing at corresponding temperature for1 min, and elongation at 72°C for 3 min (1 min for DGGEuse); and post-elongation at 72°C for 10 min.

Denaturing Gradient Gel Electrophoresis

The PCR products of the same length were separated byDGGE [39], which was performed according to the Dcode-System (BioRad, Germany). Polyacrylamide gels 6% (w/v,acrylamide-bisacrylamide 37.5:1) were prepared with de-naturing gradients ranging from 30 to 60% (in which 100%denaturant contained 7 M urea and 40% v/v formamide)and were run at 60°C and 80 V for 15 h. Bands detected byfluorescence using a UV transilluminator were excised andre-amplified for sequencing.

Clone Libraries, ARDRA Analysis, and Sequencing

For a further comparison between young and mature biofilmcommunities, 2- and 60-day old biofilms were analyzed byclone libraries. The 16S rRNA gene amplificates (length1,465 and 1,467 bp for Bacteria and Archaea, respectively)were cloned using TOPO Cloning Kit (Invitrogen Corpora-tion, San Diego, California) and then transformed intocompetent Escherichia coli cells. Plasmid inserts werescreened by amplified ribosomal DNA restriction analysis(ARDRA) using the enzyme Sau3AI (BioLabs Inc., NewEngland). Fragments were separated by 2% (w/v) agarose(Pronadisa, Madrid) gel electrophoresis and visualized by

Table 1 rDNA oligonucleotide probes used in this study

Probe Specificity Probe sequence (5’-3’) Reference

EUB338 Bacteria GCTGCCTCCCGTAGGAGT [4]ARC915 Achaea GTGCTCCCCCGCCAATTCCT [50]NON338 Negative control ACTCCTACGGGAGGCAGC [55]ALF968 α-Proteobacteria GGTAAGGTTCTGCGCGTT [42]BET42a β-Proteobacteria GCCTTCCCACTTCGTTT [32]GAM42a γ-Proteobacteria GCCTTCCCACATCGTTT [32]SRB385 Sulfate-reducing bacteria CGGCGTCGCTGCGTCAGG [4]DSS658 Desulfosarcina, Desulfococcus TCCACTTCCCTCTCCCAT [33]DSV698 Desulfovibrio sp. GTTCCTCCAGATATCTACGG [33]SYN835 Syntrophobacter GCAGGAATGAGTACCCGC [24]BAC1080 Bacteroides GCACTTAAGCCGACACCT [17]LGC354 Gram-positive bacteria with low G + C content TGGAAGATTCCCTACTGC [36]HGC69A Gram-positive bacteria with high G + C content TATAGTTACCACCGCGT [44]MEB859 Methanobacteriales (except Methanothermaceae) GGACTTAACAGCTTCCCT [9]MC1109 Methanococcales GCAACATAGGGCACGGGTCT [43]MG1200 Methanomicrobiales CGGATAATTCGGGGCATGCTG [43]MSSH859 Methanosarcinales CTCACCCATACCTCACTCGGG [9]MS1414 Methanosarcinales (except Methanosaeta) CTCACCCATACCTCACTCGGG [43]MX825 Methanosaeta TCGCACCGTGGCCGACACCTAGC [43]

Biodiversity and Structure of Anaerobic Biofilms 123123

ethidium bromide staining. Clones were grouped accordingto their restriction patterns defining different operationaltaxonomic units (OTUs). Subsequently, two clones of eachOTU were amplified by PCR using the M13 primer set(Invitrogen). Automated DNA sequencing was performedwith an ABI model 377 sequencer (Applied Biosystems).

Sequence Analysis

All sequences obtained in this work were compared withthe databases by using Basic Local Alignment Search Tool(BLAST) [1] to identify the closest sequence. Sequencedata were aligned and analyzed with the ARB programpackage [31]. Parsimony was used to construct phyloge-netic trees.

Nucleotide Sequence Accession Numbers

The sequences obtained in this study have been depositedin the GenBank database under accession numbersAY692039 to AY692074.

Results

Three complementary approaches were used to monitor thekinetics of biofilm formation, their phylogenetic diversity,and the spatial distribution of the populations: SEM, FISH,and DNA sequencing after direct cloning of the 16S rRNAgene and resolution by DGGE of amplified fragments ofthis gene.

Figure 1 Micrographs of a bio-film over time using scanningelectron microscopy. ×1000magnification for all the picturesexcept for 60 days (200×).Details shown in the innersquare are ×4000.

124 N. Fernández et al.

SEM

Microscopic examination showed an erratic colonizationsequence during the first few hours of biofilm formation.The cell density on the surface changed continuouslywithout a trend, and the microorganisms were widelyspaced on the surface of the support (Fig. 1, 3–24 h). After2 days, the growth tended to stabilize, the production of amatrix of exopolymers began, and the first microcoloniescould be observed (Fig 1, 4 days). The microcolonies thenprogressively spread to form a biofilm (Fig. 1, 8 –14 days).SEM images showed that the mature biofilm consisted ofboth densely populated and less dense areas during thegrowth period (Fig. 1, 60 days), results that were similar tothose previously reported [52].

These data were corroborated by monitoring the adheredcells during the first 10 days (Fig. 2). The growth curveshowed that adhesion behavior was random during the first36 h, corresponding to the period before growth ofmicrocolonies was observed by SEM. Subsequently, aconstant rate of adhesion was observed which tended tostabilize in the final stage of the studied period.

This microscopic analysis also revealed extensive mor-phological diversity including different kinds of spirillum,straight and curved rods, and small coccoid cells (Fig. 1,4 days detail). Two methanogens could be identified due totheir particular shape: Methanospirillum-like cells appearedafter 1 day of growth, whereas Methanosaeta-like cellswere clearly identified after 4 days.

FISH

Based on SEM and growth curve results, to determine thespatial distribution of the microorganisms, the developmentof the biofilm was divided roughly into three stages: initial

attachment (0–36 h), consolidation (from 36 h to 2 weeks),and maturation (from 2 weeks to 2 months).

Different groups belonging to the domain Bacteria (α,β, γ-Proteobacteria, Syntrophobacter, sulfate-reducingbacteria, Bacteroides, and Gram-positive bacteria) wereanalyzed by FISH and quantified for each stage. Duringinitial attachment, the number of microorganisms reached106 cells/cm2 (total DAPI stained cells), with a percentageof hybridized cells between 50 and 70% (probes EUB338plus ARC915 vs DAPI stained cells). Of those, 85–95%corresponded to the domain Bacteria. During this period,α-Proteobacteria (35–55% of detected bacterial cells,Fig. 3a) was the most representative group. β- and γ-Proteobacteria (5–15% each) were also detected, alwaysassociated to colonies of α-Proteobacteria (Fig. 3b and c).The presence of Syntrophobacter (5–10%) and sulfate-reducing bacteria (5–8% each) was also noticeable (Fig. 3dand e). The presence of archaea could be detected after only36 h, although these microorganisms were always scarce incomparison to bacteria. These cells had the form of longbowed rods that formed small Methanospirillum-likefilaments, which hybridized with the MG1200 (Methano-microbiales) probe (Fig. 3f), confirming the results of SEM.

The consolidation stage of the biofilm was marked by theformation of colonies that began to interconnect but stillappeared as independent entities (Fig. 3h). β-Proteobacteriawere specifically located on the edges of the colonies.These colonies were mainly made up of α-Proteobacteriatogether with the other groups detected: γ-Proteobacteria,Syntrophobacter, and sulfate-reducing bacteria which hadno tendency to occupy specific positions within thecolonies. Archaea were basically represented by Methano-spirillum-like cells (probe MG1200: Methanomicrobiales),as large rods and small filaments scattered throughout thebacterial colonies forming a network. Methanosarcinagenus (probe MS1414) were observed after 4 days ofgrowth, which were dense and packed within the bacterialcolonies (Fig. 3h and i), although in smaller amounts thanMethanospirillum-like cells. In addition to this, Methano-bacteriales group (probe MEB859, Fig. 3g) was found invery small quantities. The presence of Methanosaeta wasnot detected.

Mature biofilm covered almost the entire surface of thesupport, with cells embedded in an exopolymeric matrix.3D reconstructions based on biofilm sections obtainedusing confocal microscopy revealed their spatial distribu-tion. The groups detected were not very different fromprevious stages. The main body was formed mostly ofbacteria, with the archaeal cells clearly defined within it(Fig. 3j). They were basically α-Proteobacteria, with β-Proteobacteria always located on the edge of the biofilmand γ- and δ-Proteobacteria scattered within it. Bacter-oides and Gram-positives were also detected at this stage.

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

0.040

0 50 100 150 200 250

Time (h)

O.D

.

Figure 2 Growth curve of a biofilm during the first 10 days ofdevelopment.

Biodiversity and Structure of Anaerobic Biofilms 125125

Archaea appeared as individual cells or short filaments thatspread forming a network throughout the biofilm; however,it was mainly composed of Methanosaeta filaments. Itappears that Methanospirillum was displaced by Methano-saeta. As noted before, densely packed groups of verybright Methanosarcina were normally located inside thecolonies (Fig. 3j).

DGGE Profiles

Changes in microbial diversity during biofilm formation werestudied using DGGE profiling. Figure 4 shows the bandpatterns resolved with DGGE after partial 16S rRNA geneamplification using specific primers for the domains Archaeaand Bacteria. Archaea were not detected during the first 24 hof development. Since then and during the rest of the studied

period, a stable pattern was maintained for this domain(Fig. 4a). The bacterial patterns were highly variable duringthe first 2 days reflecting large changes in diversity in thisdomain at the beginning of the growth (Fig. 4b). After that,the bacterial diversity remained fairly constant.

All visible bands were excised from the DGGE finger-prints, re-amplified, purified and sequenced. A total of 11bacterial bands and 3 archaeal bands yielded sequences thatwere analyzed using the BLAST program. The taxonomicaffiliations of the 16S rRNA partial sequences are shown inTable 2. All sequenced bacterial bands belonged to Proteo-bacteria, Firmicutes, Actinobacteria, and Bacteroidetesphyla. Some of them were present only during the initialperiod: B5 (Flavobacteriaceae), B7 (Hydrogenophilaceae),and B11 (Nocardiaceae). Bands B2 (Acetobacteraceae), B3(Rhodocyclaceae), B9 (Syntrophomonadaceae), and B10

Figure 3 FISH of differentdevelopment stages of a biofilmviewed by epifluorescence andCLS microscopies. a, b, c, d, e,f 36 h. g, h, i 4 days. j 60 days.The samples a, b, d. e, f, i weresimultaneously stained withDAPI (blue) and specific-groupCy3-labeled probes (red).a α-Proteobacteria (ALF338-Cy3). b β-Proteobacteria (BE-T42a-Cy3). d Sulfate-reducingbacteria (SRB385-Cy3). e Syn-trophobacter (SYN835-Cy3).f Methanomicrobiales(MG1200-Cy3), with Methano-spirillum-like red rods. g Over-lay of Methanobacteriales(MEB859-Cy3, red) and Ar-chaea (ARC915-fluos, green).The samples c and h were si-multaneously hybridized withthe universal bacterial probe(EUB338-fluos, green) and oth-er probes: c γ-Proteobacteria(GAM42a-Cy3, yellow) andh Archaea (ARC915-Cy3, red).i Detail of Methanosarcinagroup (ARC915-Cy3). jOrthogonal CLMS section ofthe XZ plane of a mature bio-film hybridized with specificprobes for Archaea and Bacteria(ARC915-Cy3, red andEUB338-fluos, green). All themicrographs have a ×630magnification with the excep-tion of i which has a ×1000magnification.

126 N. Fernández et al.

(Campylobacteraceae) were also detected in the consolida-tion period. The remaining bands were found in theconsolidation and mature stage: B4 (Comamonadaceae),B1 (Syntrophobacteraceae), B6, and B8 (order Clostri-diales). In the case of the domain Archaea, the bandsbelonged to Methanosarcinales (A1 and A2) and Methano-microbiales orders from the Methanomicrobia class.

Clone Libraries

Two specific periods during the biofilm formation werestudied by cloning: after 2 days, at the beginning of theconsolidation stage when the microorganisms can beconsidered to be specifically attached to the support, andafter 2 months, when the biofilm is mature. Almostcomplete 16S rRNA sequences were amplified from totalDNA extracted from the biofilm using universal primers forthe bacterial and archaeal domains. ARDRA analysis of the90 (bacterial Domain) and 85 (archaeal Domain) clones forthe 2-day-old biofilm and 77 (bacterial Domain) and 96(archaeal Domain) clones for the 60-day-old biofilmallowed us to group them and to define different OTUs

Figure 4 DGGE fingerprint for Archaea (a) and Bacteria (b)throughout the experiment.

Table 2 Sequences from NCBI database with the highest similarity to each band

Banda Pb Closest relative (Accession number) Sc (%)

B1 c,m Syntrophobacter fumaroxidans (X82874) 92B2 i,c Clone I79 (AY692039), this work 99

Oleomonas sagaranensis (D45202) 97B3 i,c Uncultured IMCC1716 (DQ664239), freshwater bacteria 99

Azonexus fungiphilum (AJ630292) 97B4 c,m Uncultured clone SsB12 (AB291302), UASB reactor 92

Hydrogenophaga defluvii (AJ585993) 90B5 i Clone I73 (AY692051), this work 99

Flavobacterium frigoris (AJ557887) 93B6 c,m Clone M77 (AY692049), this work 99

Catabacter hongkongensis (AY574991) 91B7 i Dechloromonas aromatica (CP000089) 98B8 c,m Clone M78 (AY692048), this work 99

Aminobacterium colombiense (AF069287) 90B9 i,c Clone BA128 (AF323770), methanogenic consortium 99

Thermovirga lienii (DQ071273) 96B10 i,c Clon I92 (AY692047) this work 99

Arcobacter sp. R-28314 (AM084114) 98B11 i Rhodococcus opacus (AY027583) 99A1 c,m Uncultured CLONG74 (DQ478747), anaerobic sludge 98

Metanosarcina mazei (AY196685) 94A2 c,m Clone M1 (AY692055), this work 98

Methanosaeta concilii (X51423) 96A3 c,m Uncultured methanospirillum (DQ478753), UASB reactor 100

Methanoculeus palmolei (Y16382) 95

aA Archaea, B bacteriab Period, i Initial, c consolidation, m maturec SimilarityFor uncultured microorganisms a brief description of the environment of origin is given

Biodiversity and Structure of Anaerobic Biofilms 127127

which were formed from clones with the same restrictionband pattern (Table 3).

After 2 days of biofilm formation, five different OTUswere detected for Bacteria domain and one for Archaeadomain, while after 60 days, six and three OTUs werefound, respectively. The 16S rRNA gene sequences wereaffiliated with mainly uncultured bacteria from differenthabitats and only remotely related to known bacterialspecies (Table 3).

Three methanobacteria were identified inside the domainArchaea (Table 3, Fig. 5). Two could be included in theMethanosarcinales order: Methanosaeta (Ca1 and Ca2) andMethanosarcina genera (Ca3). The other sequencesbelonged to the Methanomicrobiales order, probably tothe Methanospirillum genera (Ca4).

Taxonomic similarities and phylogenetic relationshipsshowed that most of the bacteria belonged to the phylumProteobacteria (Table 3, Fig. 6): two OTUs could beincluded in the α class, one was related to Acetobacteraceae(Cb1) and the other to Rhodospirillaceae (Cb2); another onein the β class was related to Rhodocyclaceae (Cb3); oneOTU in the δ class was related to sulfate-reducing bacteria(Cb4); and one OTU in the ε class was related to the

Arcobacter genus (Cb5). Many of the remaining sequenceswere members of Gram-positive bacteria: three OTUsbelonging to the Clostridiales (Cb6, Cb7 and Cb8).Members of the Flavobacterium–Cytophaga group (Cb9)and Thermotogae phylum (Cb10) were also identified.

0.10

Ca2 (AY692055)

Ca1 (AY692053)

Methanosaeta concillii (M59146)

Methanosaeta thermoacetophila (AF334401)

Methanosarcina mazei (AF028691)

Ca3 (AY692059)

Methanosarcina barkeri (AF028692)

Ca4 (AY692060)

Methanomicrobium mobile (M59142)

Methanobacterium formicicum (AF028689)

Methanospirillum hungatei (M60880)

Figure 5 Archaeal phylogenetic tree based on almost complete 16SrRNA gene sequences retrieved from the cloning analysis. The barscale represents 10 nucleotide substitutions per 100 nucleotides. Thetree was constructed using parsimony.

Table 3 Abundance of operational taxonomic units (OTUs) determined with each clone library and sequences from NCBI database with thehighest similarity to each OTU

OTUa Pb Abundance (%) Closest relative (accession number) Sc (%)

α-ProteobacteriaCb1 c 16.7 Oleomonas sagaranensis (D45202) 97Cb2 c 5.6 Azospirillum brasilense (X79733) 96β-ProteobacteriaCb3 c 27.8 Dechloromonas sp. LT-1 (AY124797) 98δ-ProteobacteriaCb4 m 36.3 Uncultured bacterium PL-37B10 (AY570628), oil reservoir. 99ε-ProteobacteriaCb5 c 38.9 Uncultured Arcobacter sp. DS081 (DQ234164), mangrove bacterioplankton. 99ClostridiaCb6 m 27.3 Uncultured bacterium CLONG96 (DQ478749), UASB reactor. 95Cb7 m 18.2 Uncultured bacterium SJA-136 (AJ009493), anaerobic trichlorobenzene-transforming

microbial consortium96

Cb8 m 9.1 Uncultured bacterium SHA-74 (AJ306755), dechlorinate consortium 94FlavobacteriaCb9 c 11.1 Flavobacterium frigidarium (AY771722) 93ThermotogaeCb10 m 9.1 Uncultured bacterium (AB195923), anaerobic reactor. 99MethanomicrobiaCa1 c 100 Methanosaeta concilii (X51423) 99Ca2 m 91.7 Methanosaeta concilii (X51423) 98Ca3 m 6.2 Methanosarcina mazei (AE008384) 98Ca4 m 2.1 Methanospirillum hungatei (M60880) 96

aCa Archaea, Cb bacteriab Period, c consolidation, m maturec SimilarityFor uncultured microorganisms, a brief description of the environment of origin is given

128 N. Fernández et al.

Discussion

The main phases described for single-species biofilmformation were also observed during the development of amultispecies anaerobic biofilm. It is known that microbialcells might be affected by the adsorption–desorptionprocesses caused by the electrostatic and shearing forcesthat take place between surface charges and charges on thebacterial surface [30]. This could explain the behaviorobserved during the initial stage of development in whichthe microbial adhesion to the support was a randomprocess, as indicated by SEM (Fig. 1) and supported byoptical density measurements (Fig. 2), FISH (Fig. 3a to f),and the band patterns generated with DGGE (Fig. 4). In thisstudy, the influence of physicochemical and operationalconditions on the growth trend was minimized, as thereactor was operated at steady-state throughout the entireexperiment, and the environmental conditions wereexpected to be fairly constant. It should be noted thatneither archaea nor Gram-positive bacteria appeared duringthe first 36 h. It is plausible that the chemical characteristicsof their external envelopes, and the lower number offimbriae of Gram-positive with respect to the Gram-negative bacteria could be critical factors during the initialcolonization and could be implicated in their delay incolonizing the surface.

According to the accepted model, once beyond thisinitial stage, the influence of the stochastic processes andphysicochemical conditions waned, and the nature of thebacteria–surface interaction could be determined by theattachment of bacterial fimbriae [18] and by the excretionof an exopolymeric matrix [18, 53]. Such a matrix wasobserved after 1–2 days and marked the beginning of theconsolidation stage. From this point on, the physicalstructure of biofilm evolved, changing from isolated micro-colonies to a mature stage in which the cells wereembedded in the matrix, adopting rounded shapes stabilizedby EPS (Fig. 1, 60th day). In addition, the bacterialdiversity of the biofilm, confirmed by FISH and DGGEband patterns, remained fairly constant in comparison withthe initial stage. This implies that the basis of the microbialcommunity and, therefore, the main pathways of the trophicweb were formed during the consolidation stage.

Once irreversible adhesions to the surface had occurredand the microcolonies started to form, the first communitymainly comprised α-Proteobacteria according to FISH.Oleomonas sagaranensis (Cb1, B2) is able to formaggregates [27] producing some EPS, so, facilitating itsadhesion to the surface. This characteristic and its highmetabolic versatility [27]—are advantages that make thismicroorganism a pioneer in the colonization of the surface.At the same time, Azospirillum (Cb2), which also has abroad metabolic capability, appeared. These two α-Proteobacteria could participate in the breakdown oforganic polymers.

In general, the microorganisms that initiated the forma-tion of the biofilm have a broad metabolic flexibility thatfacilitated invasion of the surface. The intermediatesproduced after the lysis of macromolecules could bedegraded by proteobacteria: Oleomonas (Cb1, B2), Azo-nexus (B3), Azospirillum (Cb2), a γ-Proteobacteria (enter-obacteria were detected by FISH), Arcobacter (Cb5, B10),or sulfate reducers (FISH data) [10], or by the firmicutesThermovirga (B9). Syntrophobacteria (B1) was anexpected member of the microbial community because itis implicated in the degradation of fatty acids and otherintermediate products during fermentation to acetate andhydrogen. However, Syntrophobacteria appeared in lowproportions (FISH data) in the first stages. We mustemphasize that Azonexus is able to grow under micro-aerophilic conditions, in line with the observation that theβ-Proteobacteria were always located in the exterior zonesof the microcolonies where they had access to traces ofoxygen present in the reactor. The presence of a microor-ganism related to the class Flavobacteria (Cb9, B5) seemsodd because they are typical of aerobes environments.Nonetheless, sequences similar to Cytophaga have beenfound in anaerobic environments such as sulfate-reducingenrichment cultures [41] and in granular sludge [12]. It has

Azospirillum brasilense (X79733)

Cb2 (AY692040)

Rhodospirillum rubrum (M55597)

Oleomonas sagaranensis (D45202)

Cb1 (AY692039)

Rhodocylus tenuis (D16209)

Dechloromonas sp. LT-1 (AY124797)

Cb3 (AY692041)

Arcobacter butzleri (AF314538)

Cb5 (AY692046)

Desulfuromonas michiganensis (AF357915)

Cb4 (AY692042)

Flavobacterium frigidarium (AF162266)

Cb9 (AY692051)

Bacteroides fragilis (X83959)

Aminobacterium colombiense (AF069287)

Cb6 (AY692048)

Ruminococcus hansenii (M59114)

Clostridium aminobutyricum (X76161)

Cb8 (AY692050)

Catabacter hongkongensis (AY574991)

Cb7 (AY692049)

Cb10 (AY692052)

Fervidobacterium islandicum (AY434670)

Thermotoga maritima (M21774)

0.10

Figure 6 Bacterial phylogenetic tree based on complete 16S rRNAgene sequences retrieved from the cloning analysis. The bar scalerepresents 10 nucleotide substitutions per 100 nucleotides. The treewas constructed using parsimony.

Biodiversity and Structure of Anaerobic Biofilms 129129

been suggested that the Cytophaga–Flavobacterium clusterplay an important role in the degradation of complexorganic matter in anaerobic marine sediments [45]. Thisrole might be extended to all anaerobic environments.

According to the FISH results, major microbial groupswithin the biofilm maintained the same architecture duringbiofilm maturation. Nevertheless, with time, new membersappeared, thus, increasing the complexity of the ecosystem.In the maturation stage, some new sequences were retrieved(Cb6, Cb7, Cb8, B6, B8). Based on the closest genusdescribed (Aminobacterium, Catabacter and Clostridium),these sequences belong to the Gram-positive class Clos-tridia, which together with Bacteroidetes members (FISHdata), could be involved not only in hydrolysis but also infermentation steps. Some closely related sequences havebeen found in methanogenic sludge (AF323770, notpublished) and granular sludge [16], suggesting that thesemicroorganisms are extremely widespread in anaerobicsystems and that they must have an important role inanaerobic digestion even though, an interesting observationwas that microorganisms from the phylum Thermotogaewere detected in the mature biofilm. Mesophilic Thermo-togae have been isolated from anaerobic ecosystems [37,59], and also, several similar sequences have been reportedin an anaerobic reactor (AB195923, not published).

With regard to the domain Archaea, all microorganismsidentified in the biofilm corresponded to methanogenicarchaea. They were absent until 36 h of biofilm develop-ment. During the consolidation stage, the number ofmethanogens grew in parallel with increase in microcolonysize, mainly represented by Methanospirillum sp., a hydro-genotrophic methanogen that began to form an increasinglycomplex network within the bacterial colonies (FISH data,A3). Because of the energy released, the biomass yield islower for acetate-consuming methanobacteria than forhydrogen-consumers favoring initial predominance ofMethanospirillum over Methanosarcina and Methanosaeta.The increased number of fermentative microorganismsimplied a higher availability of acetate for acetoclasticmethanogens, which appeared later than hydrogenotrophicones. The first acetoclastic methanogen to appear was thegenus Methanosarcina (Ca3,A1), which was found in thebiofilm as dense, bright packs. In the case of granularsludge, it has been reported [20] that substrates that areeasily fermented—such as the brewery wastewater used tofeed our reactor—are degraded mainly on the surface ofgranular sludge, while the intermediate products areconverted into acetate in the middle layers. This wouldexplain the position in the biofilm of Methanosarcina,which has a high Ks (4.02 mmol l−1) for acetate. DGGEand clone libraries showed the presence of Methanosaetagenus (Ca1, Ca2, A2) in the former steps, but it wasimpossible to detect by FISH. This implies that a few of

these microorganisms were present (less than 1%), but theirmetabolic role was negligible.

In the evolution to a mature biofilm, Methanospirillumhungateii was displaced over time by Methanosaetaconcilii. This archaea is the dominant methanogen in theanaerobic granular sludge reactors [2, 47, 48], which wasused to inoculate our reactor. M. concilii is an exclusivelyacetoclastic organism with a slow growth rate (Vmax=0.11 days−1). This could explain its early detection bymeans of PCR-based technologies, whereas with FISH, itonly appeared in abundance in the mature stages of thebiofilm. It is well known [38] that the complexity of thetrophic web of an engineered ecosystem leads to the prev-alence of acetate over hydrogen as the final product of thefermentation, resulting in overgrowth of Methanosaetacompared to Methanospirillum.

The techniques based on the 16S rRNA gene allowed usto describe the development and diversity in the microbialcommunity of an anaerobic biofilm. The insight gained onthe formation of biofilms during anaerobic wastewatertreatment can be applied to control anaerobic bioreactortechnology for a wide range of applications.

Acknowledgments This work was supported by grant CTM2006–04131/TECNO from the SpanishMinistry of Education and Science. TheCBM is also indebted to an institutional grant from the Fundación Areces.

References

1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990)Basic local alignment search tool. J Mol Biol 215:403–410

2. Alphenaar PA (1994) Anaerobic granular sludge: characterization,and factors affecting its functioning. Ph.D. thesis, Landbouwuni-versiteit, Wageningen, The Netherlands

3. Amann RI, Krumholz I, Stahl DA (1990) Fluorescent oligonucle-otide probing of whole cells for determinative, phylogenetic, andenvironmental studies in microbiology. J Bateriol 172:762–770

4. Amann RI, Binder BJ, Olson BJ, Chisholm SW, Devereux R,Stahl DA (1990) Combination of 16S rRNA-targeted oligonucle-otide probes with flow cytometry for analyzing mixed microbialpopulations. Appl Environ Microbiol 56:1919–1925

5. Amann RI, Ludwig W, Schleifer KH (1995) Phylogeneticidentification and in situ detection of individual microbial cellswithout cultivation. Microbiol Rev 59:143–169

6. Amann RI (1995) In situ identification of microorganisms bywhole cell hybridization with rRNA-targeted nucleic acid probes.In: Akkermans ADL, van Elsas JK (eds) Molecular microbialecology manual, section 3.3.6. Kluwer Academic Publishers,London, pp 3.3.6/1–3.3.6/15

7. Amann RI, Fuchs BM, Behrens S (2001) The identification ofmicroorganisms by fluorescence in situ hybridisation. Curr OpinBiotech 12:231–236

8. Araujo JC, Brucha G, Campos JR, Vazoller RF (2000) Monitoringthe development of anaerobic biofilms using fluorescent in situhybridization and confocal laser scanning microscopy. Water SciTechnol 41:69–77

9. Boetius A, Ravenschlag K, Schubert C, Rickert D, Widdel F,Gieseke A, Amann RI, Jørgesen B, Witte U, Pfannkuche O (2000)

130 N. Fernández et al.

A marine microbial consortium apparently mediating anaerobicoxidation of methane. Nature 407:623–626

10. Bergey’s manual of systematic bacteriology, 2nd edn (2005). In:Brenner DJ, Krieg NR, Staley JT (eds) The proteobacteria,Springer.

11. Brosius J, Dull TJ, Sleeter DD, Noller HF (1981) Geneorganization and primary structure of a ribosomal RNA operonfrom Escherichia coli. J Mol Biol 148:107–127

12. Chan O-C, Liu W-T, Fang HHP (2001) Study of microbialcommunity of brewery-treating granular sludge by denaturinggradient gel electrophoresis of 16S rRNA gene. Water Sci Technol43:77–82

13. Costerton JW, Llewandowski Z, DeBeer D, Caldwell D, KorberD, James G (1994) Biofilms, the customized microniche. J.Bacteriol 176:2137–2142

14. Costerton JW, Stewart PS, Greenberg EP (1999) Bacterialbiofilms: a common cause of persistent infections. Science284:1318–1322

15. Díaz EE, Amils R, Sanz JL (2003) Molecular ecology ofanaerobic granular sludge grown at different conditions. WaterSci Technol 48:57–64

16. Díaz EE, Stams AJ, Amils R, Sanz JL (2006) Phenotypicproperties and microbial diversity of methanogenic granules froma full-scale upflow anaerobic sludge bed reactor treating brewerywastewater. Appl Environ Microbiol 72:4942–4949

17. Doré J, Sghir A, Hannequart-Gramet G, Corthier G, Pochart P(1998) Design and evaluation of a 16S rRNA-targeted oligonu-cleotide probe for specific detention and quantification of humanfaecal Bacteroides populations. Syst Appl Microbiol 21:65–71

18. Dunne WM (2002) Bacterial adhesion: seen any good biofilmslately? Clin Microbiol Rev 15:155–166

19. Egli D, Bosshard G, Werlen C, Lais P, Siegrist H, Zehnder AJB,Van der Meer JR (2003) Microbial composition and structure of arotating biological contactor biofilm treating ammonium-richwastewater without organic carbon. Microbial Ecol 45:419–432

20. Fang HHP (2000) Microbial distribution in UASB granules andits resulting effects. Water Sci Technol 42:201–208

21. Fang HHP, Chan K-Y, Xu L-C (2000) Quantification of bacterialadhesion forces using atomic force microscopy (AFM). J Micro-biol Meth 40:89–97

22. García Encina PA, Hidalgo MD (2005) Influence of substrate feedpatterns on biofilm development in anaerobic fluidized bedreactors (AFBR). Process Biochem 40:2509–2516

23. Gonzalez-Gil G, Lens PNL, van Aelst A, van As H, Versprille AI,Lettinga G (2001) Cluster structure of anaerobic aggregates of anexpanded granular sludge bed reactor. Appl Environm Microb678:3683–3692

24. Harmsen HJM, Kengen HMP, Akkermans ADL, Stams AJM, DeVos WM (1996) Detection and localization of syntrophicpropionate-oxidizing bacteria in granular sludge by in situhybridization using 16S rRNA-based oligonucleotide probes.Appl Environ Microb 62:1656–1663

25. Hentzer M, Teitzel GM, Balzer GJ, Heydorn A, Molin S, GivskovM, Parsek MR (2001) Alginate overproduction affects Pseudo-monas aeruginosa biofilm structure and function. J Bacteriol183:5395–5401

26. Hidalgo MD, García-Encina PA (2002) Biofilm development andbed segregation in a methanogenic fluidized bed reactor. WaterRes 36:3083–3091

27. Kanamori T, Rashid N, Morikawa M, Haruyuki A, Imanaka T(2002) Oleomonas sagaranensis gen. nov., sp. nov., represents anovel genus in the α-Proteobacteria. FEMS Microbiol Lett217:255–261

28. Lane DJ (1991) 16S/23S rRNA sequencing. In: Stackebrandt E,Goodfellow M (eds) Nucleic acid techniques in bacterialsystematic. Wiley, New York, NY, pp 115–175

29. Lehtola MJ, Miettinen IT, Keinänen MM, Kekki TK, Laine O,Hirvonen A, Vartiainen T, Martikanen PJ (2004) Microbiology,chemistry and biofilm development in a pilot drinking waterdistribution system with copper and plastic pipes. Water Res38:3769–3779

30. Liu Y, Tay J-H (2002) The essential role of hydrodynamic shearforce in the formation of biofilm and granular sludge. Water Res36:1653–1665

31. Ludwig W, Strunk O, Westram R, Richter L, Meier H,Yadhukumar, Buchner A, Lai T, Steppi S, Jobb G, Förster W,Brettske I, Gerber S, Ginhart AW, Gross O, Grumann S, HermannS, Jost R, König A, Liss T, Lüßmann R, May M, Nonhoff B,Reichel B, Strehlow R, Stamatakis A, Stuckmann N, Vilbig A,Lenke M, Ludwig T, Bode A, Schleifer KH (2004) ARB: asoftware environment for sequence data. Nucleic Acids Res32:1363–1371

32. Manz W, Amann R, Ludwig W, Wagner M, Schleifer KH (1992)Phylogenetic oligodeoxynucleotide probes for the major sub-classes of proteobacteria: problems and solutions. Syst ApplMicrobiol 15:593–600

33. Manz W, Eisenbrecher M, Neu TR, Szewzyk U (1998) Abun-dance and spatial organization of gram-negative sulfate-reducingbacteria in activated sludge investigated by in situ probing withspecific 16S rRNA targeted oligonucleotides. FEMS MicrobiolEcol 25:43–61

34. Manz W, Wendt-Potthoff K, Neu TR, Szewzyk U, Lawrence JR(1999) Phylogenetic composition, spatial structure, and dynamicsof lotic bacterial biofilms investigated by fluorescent in situhybridization and confocal laser scanning microscopy. MicrobialEcol 37:225–237

35. McLeod ES, MacDonald R, Brözel VS (2002) Distribution ofShewanella putrefaciens and Desulfovibrio vulgaris in sulphido-genic biofilms of industrial cooling water systems determined byfluorescent in situ hybridisation. Water SA 28:123–128

36. Meier H, Amann RI, Ludwig W, Schleifer KH (1999) Specificoligonucleotide probes for in situ detection of a major group ofgram-positive bacteria with low DNA G + C content. Syst ApplMicrobiol 22:186–196

37. Miranda-Tello E, Fardeau M-L, Thomas P, Ramirez F, CAsalot L,Cayol J-L, Garcia J-L, Ollivier B (2004) Petrotoga mexicana sp.nov., a novel thermophilic, anaerobic and xylanolytic bacteriumisolated from an oil-producing well in the Gulf of Mexico. Int JSys Evo Microbiol 54:169–174

38. Mujer W, Zehnder AJB (1983) Conversion. processes inanaerobic digestion. Water Sci Technol 15:127–167

39. Muyzer G, Hottenträger S, Teske A, Wawer C (1996) Denaturinggradient gel electrophoresis of PCR-amplified 16S rDNA - A newmolecular approach to analyse the genetic diversity of mixedmicrobial communities. In: Akkermans ADL, van Elsas JK (eds)Molecular microbial ecology manual, Section 3.4.4. KluiwerAcademic Publishers, London, pp 3.4.4./1–3.4.4./23

40. Muyzer G, Smalla K (1998) Application of denaturing gradientgel electrophoresis (DGGE) and temperature gradient gel electro-phoresis (TGGE) in microbial ecology. Mini review. Anton LeeuwInt J G 73:127–141

41. Nakagawa T, Sato S, Yamamoto Y, Fukui M (2002) Successivechanges in community structure of an ethylbenzene-degradingsulfate-reducing consortium. Water Res 36:2813–2823

42. Neef A, (1997) An wendug der in situ-Einzelzell-Identifizierungvon Bakterien zur populations analyse in Komplexen Mikro-biellen biozönsen. Doctoral thesis, Faculty of Biology, Chemistryand Geosciences, Technische Universität München.

43. Raskin L, Strinmley J, Rittmann BE, Stahl DA (1994) Groupspecific 16S RNA hybridization probes to describe naturalcommunities of methanogens. Appl Environ Microb 60:1232–1240

Biodiversity and Structure of Anaerobic Biofilms 131131

44. Roller E, Wagner M, Amann RI, Ludwig W, Schleifer KH (1994)In situ probing of gram-positive bacteria with high DNA G + Ccontent using 23S rRNA targeted oligonucleotides. Microbiology140:2849–2858

45. Roselló-Mora R, Thamdrup B, Schäfer H, Weller R, Amann R(1999) The response of the microbial community of marinesediments to organic carbon input under anaerobic conditions. SistAppl Microbiol 22:237–248

46. Sanz JL, Köchling T (2007) Molecular. biology techniques usedin wastewater treatment: An overview. Process Biochem 42:119–133

47. Schmidt JE, Ahring BK (1995) Granular sludge formation inupflow anaerobic sludge blanket (UASB) reactors. BiotechnolBioeng 49:229–246

48. Sekiguchi Y, Kamagata Y, Nakamura K, Ohashi A, Harada H(1999) Fluorescence in situ hybridization using 16S rRNA-targeted oligonucleotides reveals localization of methanogensand selected uncultured bacteria in mesophilic and thermophilicsludge granules. Appl Environ Microb 65:1280–1288

49. Silyn-Roberts G, Lewis G (2003) Substrata effects on bacterialbiofilm development in a subsurface flow dairy waste treatmentwetland. Water Sci Technol 48:261–269

50. Stahl DA, Amann RI (1991) Development and application ofnucleic acid probes. In: Stackebrandt E, Goodfellow M (eds)Nucleic acid techniques in bacterial systematics.. Wiley Inc, NewYork, pp 205–247

51. Tolker-Nielsen T, Brinch UC, Ragas PC, Andersen JB, JacobsenCS, Molin S (2000) Development and dynamics of Pseudomonassp. biofilms. J Bacteriol 182:6482–6489

52. Thörn M, Mattsson A, Sörensson F (1996) Biofilm developmentin a nitrifying trickling filter. Water Sci Tech 34:83–89

53. Tsuneda S, Aikawa H, Hayashi H, Yuasa A, Hirata A (2003)Extracellular polymeric substances responsible for bacterialadhesion onto solid surface. FEMS Microbiol Lett 223:287–292

54. Wagner M, Horn M, Daims H (2003) Fluorescence in situhybridisation for the identification and characterization of prokar-yotes. Curr Opin Microbiol 6:302–309

55. Wallner G, Amann R, BeiskerW (1993) Optimizing fluorescent in situhybridization with rRNA-targeted oligonucleotide probes for flowcytometric identification of microorganisms. Cytometry 14:136–143

56. Wäsche S, Horn H, Hempel DC (2002) Influence of growthconditions on biofilm development and mass transfer at the bulk/biofilm interface. Water Res 36:4775–4784

57. Watnick PI, Kolter R (1999) Steps in the development of a Vibriocholerae biofilm. Mol Microbiol 34:586–595

58. Watnick P, Kolter R (2000) Minireview. biofilm, city of Microbes. JBacteriol 182:2675–2679

59. Wery N, Lesongeur F, Pignet P, Derennes V, Cambon-BonavitaM-A,Godfroy A, Barbier G (2001)Marinitoga camini gen. nov., sp. nov.,a rod-shaped bacterium belonging to the order Thermotogales,isolated from a deep-sea hydrothermal vent. Int J Sys EvoMicrobiol 51:495–504

60. Wimpenny J, Manz W, Szewzyk U (2000) Heterogeneity inbiofilms. FEMS Microbiol Rev 24:661–671

61. Wu W-M, Hickey RF, Zeikus JG (1991) Characterization ofmetabolic performance of methanogenic granules treating brewerywastewater: role of sulphate-reducing bacteria. Appl EnvironmMicrob 57:3438–3449

132 N. Fernández et al.