Changes in the Archaea microbial community when the biogas fermenters are fed with protein-rich substrates

(1)

Changes in the Archaea microbial community when the biogas fermenters are fed with protein-rich substrates

Norbert Ács

^a

, Etelka Kovács

^a

, Roland Wirth

^a

, Zoltán Bagi

^a

, Orsolya Strang

^a

, Zsóﬁa Herbel

^a

, Gábor Rákhely

^a

, Kornél L. Kovács

^a,b,^⇑

aDepartment of Biotechnology, University of Szeged, H-6726 Szeged, Közép fasor 52, Hungary

bInstitute of Biophysics, Biological Research Center, Hungarian Academy of Sciences, H-6726 Szeged, Temesvári krt. 52, Hungary

h i g h l i g h t s

"Laboratory-scale CSTR AD was done with casein or pig blood as sole substrate.

"The mesophilic methanogenic microbial community acclimated to these substrates.

"T-RFLP and sequencing ofmcrAand 16S rRNA gene identiﬁed methanogenic Archea.

"The methanogens responded to the change in substrate.

a r t i c l e i n f o

Article history:

Received 22 August 2012

Received in revised form 18 December 2012 Accepted 19 December 2012

Available online 27 December 2012

Keywords:

Biogas Methanogen Protein mcrAgene 16S rRNA gene

a b s t r a c t

Terminal restriction fragment length polymorphism (T-RFLP) was applied to study the changes in the composition of the methanogens of biogas-producing microbial communities on adaptation to protein- rich monosubstrates such as casein and blood. Specially developed laboratory scale (5-L) continuously stirred tank reactors have been developed and used in these experiments. Sequencing of the appropriate T-RF fragments selected from a methanogen-specific (mcrAgene-based) library revealed that the methanogens responded to the unconventional substrates by changing the community structure. T-RFLP of the 16S rDNA gene confirmed the findings.

1. Introduction

The scarcity of cheap fossil energy carriers and the global environmental changes associated with their extensive consumption are posing a threat to the maintenance of sustainable living conditions and giving rise to economic and social crises on a global scale.

This situation has led to the initiation of worldwide efforts with the goal of the replacement of fossil resources with renewable and environmentally friendly ones. Biogas, a product of the anaerobic degradation (AD) of organic material, is one of the most promising renewable energy carriers, in part because of the wide range of

substrates that can be utilized in the AD process (De Paoli et al., 2011; Ferreira et al., 2012). This is the only biotechnological approach that effectively combines the treatment and elimination of organic waste streams with direct energy production. The process is carried out by a microbial community involving several hundred individual species. The biogas product of AD is a mixture of CH4, CO2 and traces of other gaseous components. Biogas is either utilized directly for heat generation through burning of the raw gas, or is partially puriﬁed before conversion to heat and elec- tric power. In addition, there is enormous potential in using biom- ethane, the thoroughly puriﬁed biogas, as a biofuel (Murphy et al., 2004; Osorio and Torres, 2009).

Biogas is formed under strict anaerobic conditions by a unique microbial consortium that can be divided into three functional groups (Gerardi, 2003).

The microbial food chain leading to biogas production is very complex, a multitude of methods are therefore used to follow and understand the relationships among the members of this community. In everyday practice, this includes the measurement of 0960-8524/$ - see front matterÓ2012 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.biortech.2012.12.134

⇑ Corresponding author at: Department of Biotechnology, University of Szeged, H-6726 Szeged, Közép Fasor 52, Hungary. Tel.: +36 62546930.

E-mail addresses:acsn@brc.hu(N. Ács),kovacse@brc.hu(E. Kovács),roland.w@

freemail.hu (R. Wirth), bagiz@brc.hu (Z. Bagi), strango@brc.hu (O. Strang), herbelzs@brc.hu(Z. Herbel),rakhely@brc.hu(G. Rákhely),Kovacs.kornel@brc.mta.

hu(K.L. Kovács).

Contents lists available atSciVerse ScienceDirect

Bioresource Technology

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / b i o r t e c h

(2)

various chemical parameters (pH, redox potential, FOS/TAC, con- ductance, etc.), which indirectly reﬂect the functional status of the biogas reactor. Cultivation-independent molecular biological tools have been developed including denaturing gradient gel electrophoresis (Boon et al., 2002), automated ribosomal intergenic spacer analysis (Pecchia et al., 1998) and single strand conforma- tion polymorphism (Quéméneur et al., 2010). Terminal restriction fragment length polymorphism (T-RFLP) is one of the molecular biological approaches that can be used to identify the most pre- dominant microbes in a biogas producing digester (Liu et al., 1997; Osborn et al., 2000). This technique was chosen for the present study, because it is widely employed for the semiquantitative detection of microbes from environmental samples (Horz et al., 2000; Schütte et al., 2008), including those from biogas reactors (Collins et al., 2003; Klocke et al., 2007; Kobayashi et al., 2009).

T-RFLP is a fingerprinting tool mainly targeting the small sub- unit (16S and 18S) ribosomal RNA genes from the total DNA of the community. Polymerase chain reactions (PCRs) have been developed, wherein one or both primers are labeled with a special- ized fluorescent dye. The PCR product is then digested with restriction enzymes, which have four base-pair recognition sites, and the fragments are visualized, e.g. in a capillary gel electrophoresis system. The nascent short DNA sequences, often called T-RFs (terminal restriction fragments) represent the dominant species in the examined sample. Typically, T-RFLP analysis requires a clone library for identification of the assortment of DNA sequences. The members of the clone library are screened and those that yield a substantial individual T-RF are sequenced via the Sanger method.

In this set of experiments the changes in the composition of the biogas-producing microbial community after the biogas fermenter was fed solely with a protein-rich substrate were investigated. Pro- tein-rich materials, which frequently accumulate as waste by- products to be disposed of in food processing, are generally considered toxic for the biogas fermentation process because of the high level of ammonia released upon protein degradation. We have recently demonstrated that the microbial community acclimates to a high concentration of protein without any additional co-substrate, and the modiﬁed community is capable of an elevated level of biogas production (Kovacs et al. to be published). The changes in the microbial composition involved practically all of the genera partic- ipating in biomass degradation. The present study focuses on the methanogenic Archaea by targeting a part of the methyl coenzyme-M reductase gene (mcrA) with universal primers designed earlier (Luton et al., 2002). The results are complemented with an analysis of the V3–V5 variable region of the 16S rRNA gene, with the use of universal primers (Baker et al., 2003).

2. Methods

2.1. Fermentation conditions

All anaerobic fermentations were carried out in 5-L continuously stirred tank reactors (CSTRs) (Kovacs et al. Journal of Biomed- icine and Biotechnology, in press) designed and constructed by Biospin Ltd., Hungary, and installed at the Department of Biotech- nology, University of Szeged. Experiments were run in triplicate.

Two protein-rich substrates were tested: casein (C) is a by-product of milk processing, while pig blood (B) is a waste material from slaughterhouses. The adaptation was started with an inoculum from an operational biogas plant, in which substrates of agricul- tural origin, i.e. mixtures of pig manure and maize silage, are trea- ted. Steady-state biogas production was attained after 4–6 weeks of operation and daily feeding. From this time point on the fermenters were fed only with the protein-rich substrate, an increas- ing amount of the material being supplied daily (Kovacs et al. to be

published and HU patent P1100510). Temperature was maintained at 37 ± 1.0°C. The pH was kept between 7 and 8, and the redox potential was <500 mV. Gas volume was measured with thermal mass ﬂow devices (DMFC, Brooks) attached to each gas exit port.

2.2. DNA extraction

Two milliliter of samples was withdrawn from the fermenters after the ﬁrst week of the acclimation to the protein-rich substrates (C1 and B1) and after 5 weeks (C5 and B5). A cetyltrimethylammo- nium bromide-based buffer was used to extract the total genomic DNA (gDNA) (Minas et al., 2011). Phenol: chloroform (1:1) puriﬁca- tion was employed (Roose-Amsaleg et al., 2001). gDNA concentration was determined spectrophotometrically (NanoDrop ND-1000 Technologies, Washington, USA). DNA purity was assessed by agarose gel-electrophoresis.

2.3. PCR amplification of the partial mcrA and 16S rRNA gene In order to amplify the approximately 500 bp fragment of the methyl coenzyme-M reductase coding gene, a fluorescent labeled forward primer (TET-mcrA_F) and a reverse primer (mcrA_R) were used (Luton et al., 2002). An approximately 550 bp fragment of the 16S rRNA gene was amplified by using the primer pair F344 and R934 (Baker et al., 2003;Raskin et al., 1994), in this system the forward primer was also labeled with the same fluorescent dye. PCR reactions were carried out in a volume of 30

l

L, containing approximately 50 ng of gDNA, 1X DreamTaq reaction buffer, 100

l

mol of dNTP, 3 mM MgCl2, 10

l

M of each primer (Sigma, St. Louis, MI, USA) and 1.5 U of DreamTaq DNA polymerase. Chemicals, exclud- ing the primers, and enzymes were purchased from Fermentas (St. Leon-Rot, Germany). The PCR reactions were performed in an ABI 9600 Fast Thermal cycler (Applied Biosystem, Foster City, CA, USA), where the reaction profile was as follows: initial denaturation at 96°C for 3 min, followed by 30 cycles of denaturation at 96°C for 30 s, annealing at 52°C for 30 s and elongation at 72°C for 1 min. To avoid amplification of the host 16S rRNA gene, a nested PCR reaction was used in the case of the clones carrying the partial 16S rDNA. In the first step, the vector specific primer pair (M13F and M13R) was applied to multiply the insert. In the second step, the product of the previous reaction was used as tem- plate for the primer pair F344 and R934. The PCR protocol was as described, except that the annealing temperature was raised to 58°C in the case of the primer pair F344 and R934, due to the dif- ferentTmof the oligonucleotides. A final extension step for 10 min was added in order to allow the polymerase to finish incomplete PCR products.

The PCR products were separated by electrophoresis in 1% agarose gel using TRIS-acetate buffer (Green and Sambrook, 2012), and visualized with ethidium bromide under UV light. The DNA fragments were puriﬁed by using the PCR clean-up kit (Viogene-Biotek, New Taipei, Taiwan) following the recommendations of the manufacturer; the recovered DNA was eluted in 30

l

L, and stored at 20°C.

2.4. Generation of the mcrA and 16S rDNA clone libraries

For the preparation of the clone library, the pGEM-T vector system was employed (Promega, Madison, WI, USA). The PCR product was used for the ligation, except that non-fluorescent forward primers were employed. NovaBlue chemical competent cells (NovaGene, Billerica, MA, USA) were utilized for the transformation. Cloned inserts were amplified by PCR as above, using vector-specific M13F (5⁰-TGT AAA ACG ACG GCC AGT-3⁰) and M13R (5⁰-CAG GAA ACA GCT ATG ACC-3⁰) primers. Clones carrying an insert of correct size were identified by agarose gel electrophoresis.

(3)

Small portions of the colonies were suspended in 30

l

L distilled water and heated at 95°C for 5 min. The cell debris was collected by centrifugation at 12,000 rpm for 5 min.

2.5. T-RFLP analysis and screening

Restriction digestion of the puriﬁed PCR products was carried out with the endonuclease RsaI (Fermentas, St. Leon-Rot, Ger- many). The digestion mixture contained 10

l

L of puriﬁed PCR product (100 ng), 2

l

L of Tango buffer (Fermentas, St. Leon-Rot, Germany), 3.3 U of RsaI and bi-distilled water to a ﬁnal volume of 20

l

L. The reaction was carried out in a heating block at 37°C for 3 h. The enzyme was inactivated by incubation at 80°C for 20 min, followed by ethanol precipitation. 3

l

L of the digested DNA was mixed with 12

l

L of Hi-Di formamide and 0.3

l

L of TAM- RA 500 size standard (http://www.invitrogen.com/1/1/11192- genescan-500-tamraTM-size-standard.html) (Applied Biosystems, Foster City, CA, USA). The lengths of the T-RFs were determined by comparison with the size standard in an automated ABI PRISM310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) with the help of the GeneScan^Ò3.1 analysis software (Applied Biosystems, Foster City, CA, USA). Both the mixed PCR products and the cloned amplicons were processed as described above. The restriction pattern of the community PCR was compared with the fragments of the positive clones through use of the GeneScan 3.1 program. Every run was conducted in triplicate in order to mini- mize the bias during electrophoresis. In order to eliminate the occurrence of pseudo T-RFs (Egert and Friedrich, 2003), mung bean nuclease (New England Biolabs, Ipswich, MA, USA) digestion was also performed on parallel samples. The mung bean nuclease-trea- ted samples were compared with the untreated ones and the peaks that could only be found in the untreated samples were excluded from subsequent analyses. The T-REX software (http://trex.bio- hpc.org/) helped to generate the restriction fragment size patterns.

The T-RF lengths corresponding to the capillary electrophoresis peaks were adjusted to round numbers and noise-ﬁltered. Peak intensities with an area lower than 2% of the total ﬂuorescence were excluded. The cloned fragments that could be matched with the dominant peaks in the restriction pattern were selected for sequencing and the DNA sequences of the entire inserts were determined.

2.6. Sequencing the clones

Several clones were chosen from the clone library that could be correlated with the dominant peaks of the digested total community gDNA. Plasmid DNA was puriﬁed by means of the GeneElute plasmid miniprep kit (Sigma, St. Louis, MI, USA) according to the manufacturer’s instructions. The DNA sequences were determined with the Sanger sequencing method combined with capillary gel electrophoresis. The sequences were then compared with GenBank entries with the Basic Local AlignmentS Tool (BLAST) at NCBI. The sequences of the 16S rDNA clone library were also aligned against the entries of the Ribosomal Database Project (RDP Release 10, Up- date 29;http://rdp.cme.msu.edu/).

3. Results and discussion 3.1. Biogas production

Analytical parameters (pH, redox potential and temperature) were continuously monitored in the computer-controlled fermenters and stable operation of the process was maintained. Ammonia content and volatile fatty acid (VFA) composition were measured repetitively on every third day (data not shown). The ammonia

concentration of around 1 g/L at the beginning of the fermentation increased to a value of 6 g/L during the adaptation period. The total VFA content ﬂuctuated in the interval 2–4 g/L. The CH4content of the biogas was 50–60%. The overall gas production yield was 800–

900 L/kg organic total solids (oTS), which signiﬁcantly exceeded the yields with commonly employed biogas substrates such as animal manure and/or maize silage (Schwab, 2010). A detailed evaluation of the biogas fermentation results will be presented elsewhere (Kovacs et al. to be published and HU patent P1100510).

3.2. T-RFLP restriction patterns of the mcrA PCR products

The normalized and noise-ﬁltered patterns from the fermenters fed with casein (C1) or blood (B1) at the beginning of the adaptation are presented in Fig. 1. Basically the same T-RF distribution appeared in the two cases, although the abundances of the various T-RFs slightly differed. This similarity between the C1 and B1 samples is not surprising in view of the fact that both sets of fermenters had the same history, i.e. they were fed with the same mixture of pig manure and maize silage. However, when the system was fed with either blood or casein exclusively for 5 weeks, the microbiological community structure changed substantially. The most pronounced transformation was observed in the Bacteria domain, i.e. the microbes responsible for the anaerobic decomposition of the complex feedstock, including proteins (Kovacs et al. to be published and HU patent P1100510). Interestingly, in line with the results of the metagenomic approach based on new generation sequencing of the total community gDNA, our results indicated that the methanogenic population also responded to the introduction of the new substrate. The diversity of the species comprising the archaeal community decreased, with fewer methanogens dominating in the population.

Altogether 359 clones were selected from the clone libraries containing the restriction fragments of themcrAgene. 173 clones were analyzed from the DNA samples isolated after 1 week, and 186 clones from the samples derived from anaerobic digestions of the proteinaceous substrates for 5 weeks. Following the PCR ampliﬁcation of the clonedmcrAgene fragments, restriction digestion revealed the T-RFs of the clones. A total of 68 clones were selected for sequencing, 41 from samples C1 and B1, and 27 from samples C5 and B5. The BLAST analysis results are summarized inTable 1.

3.2.1. T-RF at 57 bp operational taxonomic unit (OTU-1)

The smallest identiﬁed T-RF in all four restriction patterns was at 57 bp. In the case of the non-acclimated samples, the relative abundances were about 23% (C1) and 8% (B1). There was a pronounced change by 5 weeks of adaptation to the protein-rich substrate. The increase in abundance of the T-RF was 41% in the casein sample (C5) and nearly 73% in the fermenter fed with pig blood (B5). Consequently, the species containing thismcrAgene fragment are the group of microorganisms which play the most important role in biogas formation from protein feedstock. The sequencing of 14 clones which contained this T-RF from the two clone libraries yielded a straightforward conclusion after BLAST analysis: all the sequences were closely related (86-88%) toMethanoculleus marisnigri, the only species in the genus Methanomicrobiaceae whose complete genome sequence is known at present. This Class II methanogen has frequently been observed in anaerobic digesters, and was found to be the dominant species at the beginning of acclimation in our parallel experiments when next-generation sequencing was used (Kovacs et al. to be published and HU patent P1100510; Bagi et al., 2007; Herbel et al., 2010; Schlüter et al., 2008; Wirth et al., 2012).

(4)

Fig. 1.Restriction pattern of samples taken from anaerobic fermentation of casein (C) or pig blood (B) after 1 week (C1 and B1) and 5 weeks (C5 and B5) of acclimation to the respective monosubstrate and analyzed using themcrAor 16S rRNA gene speciﬁc primers.

Table 1

BLAST results on the sequenced clones (mcrAclone library), and their abundance in the examined samples (;indicates a decrease, whereas"indicates an increase in abundance).

OTU (bp) Closest species Similarity (%) No. of clones Abundance C1; B1 (%) Abundance C5; B5 (%)

OTU-1 (57) Methanculleus marisnigri 88 14 23; 8 41"; 73"

OTU-2 (62) Methanosarcina mazei 94 2 10; 14 3.5;; 0

UAC rlm_R_380Bc 100 5

OTU-3 (92) Methanocorpusculum parvum 98 3 21; 33 37"; 16;

Methanomassiliicoccus luminyensis 84 4

Methanoculleus bourgensis 98 2

OTU-4 (160) Methanobacterium formicicum 86 7 0.5; 1.5 2.5"; 3"

OTU-5 (189) UAC MCR-HID-UND-38 100 5 35; 40 16;; 8.5

UAC barb-M-21 100 2

UAC VAL25 100 6

OTU-6 (198) Methanoculleus chikugoensis 94 2 6.5; 0.5 0

OTU-7 (309) Methanobrevibacter ruminantium 88 3 4; 3 0

(5)

3.2.2. T-RF at 62 bp (OTU-2)

The relative abundance of this peak was initially around 10%

(C1) and 14% (B1). These strains apparently faded away to below the detection threshold during the adaptation period, which suggests that they could not acclimate to the new, protein-rich monosubstrate environment.

Some of the clones from the corresponding clone libraries were identiﬁed asMethanosarcina mazei, a well-known methanogen in biogas-producing communities, which frequently occurs in biogas plants fed with plant biomass (Klocke et al., 2007). It should be noted that some of the cloned sequences displayed a match to an- other methanogenic archaeon, which was marked as ‘‘unclassiﬁed’’

in the database. This suggests that more than one archaeal strain contributed to this T-RF peak (Engebretson and Moyer, 2003).

3.2.3. T-RF at 92 bp (OTU-3)

This peak had a relative abundance of 21% and 33% in the case of the C1 and B1 samples, respectively. The abundance of this T-RF varied, depending on the protein source used in the adaptation process. When the substrate was casein (sample C5), the area of the peak increased to almost 37%, while the relative abundance with blood as substrate (sample B5) decreased to 16%.

Twelve clones carrying this insert size were selected for sequencing: 6 of them originated from the clone library of samples C1 and B1, while the other 6 were chosen from the clone library of samples C5 and B5. The results indicated that at least 3 distinct microbes were responsible for this T-RF peak.

The most abundant of them was closely related toMethanocor- pusculum parvum, a member of the Methanocorpusculaceae family (Zellner et al., 1987). Two other microbes could be identiﬁed via the BLAST search, which exhibited a lower representation in the clone libraries. These Archaea wereMethanomassiliicoccus luminyensis, andMethanoculleus bourgensis, with a reliability conﬁdence of 84% and 99%, respectively.

It is noteworthy that the importance of these species seemed to increase during the adaptation process as 4 of 6 sequences in the clone library could be linked toM. luminyensis, and 2 of 4 inserts toM. bourgensisafter 5 weeks of adjustment of the microbial community to the protein-rich substrates.

3.2.4. T-RF at 160 bp (OTU-4)

This T-RF peak reﬂected strains with relatively low abundance.

The corresponding capillary gel electrophoresis peak was almost negligible in samples taken at the time when the substrate was practically exclusively animal manure and plant biomass (samples C1 and B1). The relative representation of the strains characterized by thismcrAT-RF comprised only about 3% in the samples adapted to casein (C5) and pig blood (B5), suggesting that these members of the community might have some secondary function in the biogas- producing community.

Seven clones carrying the DNA fragment that produced this T-RF were identiﬁed in the clone libraries and sequenced. All of them were related toMethanobacterium formicicumwith an 86% conﬁ- dence value. This hydrogenotrophic and formate-producing methanogen, which belongs to the genus Methanobacterium, has been found in various biogas-producing consortia (Rastogi et al., 2008).

3.2.5. T-RF at 189 bp (OTU-5)

This peak displayed a high abundance in both samples taken at the beginning of acclimation to the protein-rich substrates. Unfor- tunately, on BLAST analysis none of the clones selected for sequencing gave clear-cut hits to species identiﬁed in the database.

The most likely relatives in the databank entries were ‘‘uncultured archaeon’’ sequences. The corresponding strain may have been detected in biogas microbial communities, but its identity remains unknown at present. Nevertheless, the insert DNA sequences of

the clones demonstrated a high degree of similarity, suggesting a taxonomic relationship.

3.2.6. T-RF at 198 bp (OTU-5)

This OTU was apparently present only during the start-up phase of the adaptation process (samples C1 and B1) with a relative abundance of 6.5% and 0.5%, respectively. Reﬂecting the low abundance of this peak, inserts of the corresponding size were found only in the clone library of sample C1. On the basis of sequence similarities, these clones were identiﬁed as close relatives of the twoMethanoculleusspecies already detected in the shorter mcrA T-RFs.In silicostudies suggested that a third member of the Met- hanomicrobiaceae genus may also possibly account for this T-RF, e.g. Methanoculleus chikugoensis, on which only very limited knowledge is available (Tang et al., 2005).

3.2.7. T-RF at 309 bp (OTU-6)

This was the largest T-RF in the capillary electrophoresis pattern. It was detected in relatively low abundance (3–4%) in samples C1 and B1 and the corresponding microbes were likely to diminish as the adaptation to the protein substrate progressed. Nevertheless 3 members in the clone library prepared from sample C5 displayed a restriction fragment of the same size. The sequencing results for all 3 clones pointed toMethanobrevibacter ruminantium, although with fairly low (88%) conﬁdence. The Methanobrevibacter genus comprises the majority of rumen methanogens, andM. ruminanti- umis a hydrogenotrophic methanogen that also forms CH4 from formate (Leahy et al., 2010).

3.3. T-RFLP restriction pattern of the 16S rDNA gene products In order to acquire a more detailed view of the archaeal community and validate the results of themcrA-based identiﬁcations, in a separate set of experiments we tested a second primer pair, designed to the V3–V5 variable segment of the 16S rRNA gene.

The universal primers F344 and R934 permitted the amplification of a relatively short DNA sequence from the gDNA pool. The com- positions of the microbial communities in samples taken on the first day of feeding with protein-rich substrates (C1 and B1) were compared. The restriction patterns (Fig. 1) were determined and two clone libraries were created from the samples, each library containing around 100 individual clones. A total of 181 clones were screened by restriction analysis (RsaI), which resulted in 56 promising clones. Some of the capillary electrophoresis peaks were sit- uated close to each other, which made screening difficult. A detailedin silicoanalysis was therefore also carried out on the corresponding sequence of the selected clones and the precise size of the T-RF was determined. The results are presented inTable 2. In brief, the restriction patterns of samples B1 and C1 were similar.

This is not surprising since the same inoculum was used to start the anaerobic digestion process. A significant proportion of the sequenced individual clones did not relate to methanogens, as the primer pair F344 and R934 are not specific for this group of microbes, but target all Archaea groups. However, in consequence of the nature of this environment, the majority of the identified strains were involved in CH4production.

3.3.1. T-RF at 62 bp and T-RF at 142 bp (OTU-1)

As a result of the lack of matching clones in the clone libraries, available information about this T-RF is limited. Several members were found in the corresponding clone library of sample C1, but only one matching clone was found with 142 bp T-RF. The single microorganism identiﬁed wasMethanobacterium formicicum. This methanogen was also found in themcrAclone library of samples C1 and B1. It is possible that a faulty restriction reaction resulted in the occurrence of the larger T-RF (142 bp); this would explain

(6)

the lack of matching clones after screening. In silico restriction analysis veriﬁed that the insert of the clone yielding the 142 bp T-RF carried the correct size of the shorter T-RF (62 bp). These T-RFs therefore could be regarded as the same OTU.

3.3.2. T-RF at 284 (287) bp (OTU-2)

The length of this T-RF was determined experimentally as 284 bp althoughin silicorestriction digestion of the clones carrying the same T-RF gave 287 bp for this OTU. This fragment displayed the highest abundance in both samples. The relative proportion in the overall community was 40% (C1) and 47% (B1), respectively.

Twelve clones were selected for sequencing, six from each clone library. With the exception of one clone, all BLAST and RDP search results suggested that Methanoculleus bourgensis constituted this OTU. The genus Methanoculleus was also dominant in the same samples tested for the mcrA gene. An additional methanogen, Methanogenum marinum, (Chong-Song et al., 2002) was also iden- tiﬁed in this OTU, indicating a mixed T-RF, although the abundance of this latter strain was very low.

3.3.3. T-RF at 289–291 bp (OTU-3)

Numerous clones were selected for sequencing from this T-RF, which makes this OTU the second most frequently detected group, with relative abundances of 27% (C1) and 26% (B1), respectively.

Thein silicorestriction patterns suggested the complex nature of this T-RF, and this group therefore could be a mixture of separate OTUs.

The first group ofin silicoT-RFs was at 289 bp (OTU-3.1); 12 of the sequenced clones were related to this group. Four of them identified the methanogenM. luminyensis, with a high (97%) confidence value.

This methanogen was also detected in themcrAlibrary. The remain- ing clones were identified at genus level with the help of the RDP database. They belonged in the Thermogymnomonas genus, which apparently dominated this OTU and the next OTUs. A restriction fragment was also detected in silicoat 290 bp (OTU-3.2). All 11 clones sequenced corroborated the presence ofThermogymnomonas sp. in the fermenters. An additionalin silicoT-RF was detected at 291 bp (OTU-3.3). Two of the five sequenced clones were closely related to the genus Methanobrevibacter; the others pointed to the genus Thermogymnomonas. Little is known about these facultative, anaerobic thermoacidophilic archaea, although this genus has been described being closely related toM. luminyensis, a recently discov- ered member of the human microbiome (Dridi et al., 2011, 2012; Vo- lant et al., 2012). It should be noted that it was virtually impossible to separate the T-RFs comprising OTU 3.1–3.3 in the capillary gel elec- trophoretic system because of the small differences in their sizes. A careful comparison of the estimatedin silicorestriction fragments with the experimentally determined ones resulted in the identification of these microbes in the community.

3.4. Comparison with related data

It is clear from Sections3.2 and 3.3that the data based on the mcrAand the 16S rRNA genes showed a good correlation. We also

have the opportunity to relate these data sets to those obtained in a metagenomic approach (Kovacs et al. to be published and HU patent P1100510). The sequences of the clones generated in this work were compared with the short read raw DNA sequence database obtained from a metagenomic study of samples taken from the same biogas experiments. A detailed analysis of the two data sets is not possible here, but the main ﬁndings can be summarized as follows:

1. The strains identified by the T-RFLP approach using either the mcrAor 16S rRNA genes were present in the microbial consortium according to the metagenomic analysis using the SOLiD^Ò new generation DNA sequencing technology. The metagenomic approach revealed significantly more diversity due to the more extensive database generated, but the strains identified in our T-RFLP work were among the most abundant ones according to both experimental systems.

2. The relative abundances during the acclimation period to the protein rich substrates did vary depending on the experimental approach applied. Some of the strains that appeared to gain dominance as a result of acclimation according to one of the methods did not seem to follow the same pattern when the other approach was employed. There are several possible expla- nations for these divergences. First the samples were taken from the same fermentation experiments but at somewhat different time points. The DNA extracted for the metagenomic analysis was obtained from samples at the very beginning of the acclimation process (week 0), whereas the ﬁrst DNA samples for T-RFLP studies were one week older (week 1) in the course of the experiment. Second, the inherent differences between the two approaches may have been reﬂected in the dissimilar results. T-RFLP involves a PCR step, which is a potential source of systematic bias due to primer preferences even if one uses ‘‘universal’’ primers.

These observations suggest that strain speciﬁc deductions from the analyses of complex, uncultured microbial communities should be considered with precautions.

4. Conclusions

Although Archaea are not directly involved in decomposition of the protein-rich substrates such as blood and casein, the substrate composition does inﬂuence their biogas-evolving performance and abundance. Adaptation was achieved in 5 weeks, indicating the ﬂexibility of the microbial community.

Analysis of themcrAgene led to the determination of the methanogenic archaebacterial activities. The 16S rRNA gene-based analyses conﬁrmed the ﬁndings obtained from themcrAgene libraries.

The use ofin silicorestriction fragment size estimation greatly im- proved the resolution of the T-RFLP method.

It is difﬁcult to establish a putative metabolic context at this stage of understanding a decrease in the diversity of methanogenic Table 2

BLAST and RDP results on the sequenced clones (16S rDNA clone library), and their abundance in the examined samples.

OTU (bp) In silicoT-RF (bp) Closest species Similarity (%) No. of clones Abundance C1; B1

OTU-1 (62, 142) 62 Methanobacterium formicicum 97 2 13; 3.5

Methanobacterium formicicum 95 1 20; 23.5

OTU-2 (284) 287 Methanoculleus bourgensis 99 11 40; 47

Methanogenium marinum 98 1

OTU-3 (289) 289 Methanomassiliicoccus luminyensis 97 4 27; 26

Thermogymnomonassp. 81–98 8

290 Thermogymnomonassp. 87–98 11

291 Methanobrevibactersp. 95 2

Thermogymnomonassp. 85–95 3

(7)

archaea accompany the acclimation to the protein-rich monosubstrates.

The dominant role ofMethanoculleusspecies in mesophilic AD was corroborated.

Acknowledgements

This work was supported by EU projects HUSRB/1002/214/041 IPA, HURO/1001/193/2.2.2 CBC and IEE/10/235 SI2.591589 Green- GasGrids. Domestic funds from TÁMOP-4.2.1/B-09/1/KONV-2010- 0005 and TÁMOP-4.2.2/B-10/1-2010-0012 are gratefully acknowl- edged. The authors thank Dr. Gergely Maróti (Institute of Biochemistry, Biological Research Center, Hungarian Academy of Sciences) for valuable consultations and advices.

References

Bagi, Z., Acs, N., Balint, B., Horvath, L., Dobo, K., Perei, K.R., Rakhely, G., Kovacs, K.L., 2007. Biotechnological intensiﬁcation of biogas production. Applied Microbiology and Biotechnology 76 (2), 473–482.

Baker, G.C., Smith, J.J., Cowan, D.A., 2003. Review and re-analysis of domain-speciﬁc 16S primers. Journal of Microbiological Methods 55 (3), 541–555.

Boon, N., De Windt, W., Verstraete, W., Top, E.M., 2002. Evaluation of nested PCR–

DGGE (denaturing gradient gel electrophoresis) with group-speciﬁc 16S rRNA primers for the analysis of bacterial communities from different wastewater treatment plants. FEMS Microbiology Ecology 39 (2), 101–112.

Chong-Song, C., Liu, Y., Cummins, M., Valentine, D.L., Boone, D.R., 2002.

Methanogenium marinumsp. nov., a H2-using methanogen from Skan Bay, Alaska, and kinetics of H2utilization. Antonie van Leeuwenhoek 81 (1–4), 263–

270.

Collins, G., Woods, A., McHugh, S., Carton, M.W., O’Flaherty, V., 2003. Microbial community structure and methanogenic activity during start-up of psychrophilic anaerobic digesters treating synthetic industrial wastewaters.

FEMS Microbiology Ecology 46 (2), 159–170.

De Paoli, F., Bauer, A., Leonhartsberger, C., Amon, B., Amon, T., 2011. Utilization of by-products from ethanol production as substrate for biogas production.

Bioresource Technology 102 (11), 6621–6624.

Dridi, B., Fardeau, M.L., Ollivier, B., Raoult, D., Drancourt, M., 2011. The antimicrobial resistance pattern of cultured human methanogens reﬂects the unique phylogenetic position of archaea. The Journal of Antimicrobial Chemotherapy 66 (9), 2038–2044.

Dridi, B., Henry, M., Richet, H., Raoult, D., Drancourt, M., 2012. Age-related prevalence of Methanomassiliicoccus luminyensis in the human gut microbiome. Acta Pathologica, Microbiologica et Immunologica Scandinavica 120 (10), 773–777.

Egert, M., Friedrich, M.W., 2003. Formation of pseudo-terminal restriction fragments, a PCR-related bias affecting terminal restriction fragment length polymorphism analysis of microbial community structure. Applied Environmental Microbiology 69 (5), 2555–2562.

Engebretson, J.J., Moyer, C.L., 2003. Fidelity of select restriction endonucleases in determining microbial diversity by terminal-restriction fragment length polymorphism. Applied Environmental Microbiology 69 (8), 4823–4829.

Ferreira, L., Duarte, E., Figueiredo, D., 2012. Utilization of wasted sardine oil as co- substrate with pig slurry for biogas production – a pilot experience of decentralized industrial organic waste management in a Portuguese pig farm.

Bioresource Technology 116, 285–289.

Gerardi, M.H., 2003. The Microbiology of Anaerobic Digesters. John Wiley & Sons, Inc., USA.

Green, M.R., Sambrook, J., 2012. Molecular Cloning: A Laboratory Manual, Fourth ed. Cold Spring Harbor Laboratory Press.

Herbel, Z., Rakhely, G., Bagi, Z., Ivanova, G., Acs, N., Kovacs, E., Kovacs, K.L., 2010.

Exploitation of the extremely thermophilicCaldicellulosiruptor saccharolyticusin hydrogen and biogas production from biomasses. Environmental Technology 31 (8–9), 1017–1024.

Horz, H.-P., Rotthauwe, J.-H., Lukow, T., Liesack, W., 2000. Identiﬁcation of major subgroups of ammonia-oxidizing bacteria in environmental samples by T-RFLP analysis ofamoAPCR products. Journal of Microbiological Methods 39 (3), 197–

204.

Klocke, M., Mähnert, P., Mundt, K., Souidi, K., Linke, B., 2007. Microbial community analysis of a biogas-producing completely stirred tank reactor fed continuously

with fodder beet silage as mono-substrate. Systematic and Applied Microbiology 30 (2), 139–151.

Kobayashi, T., Yasuda, D., Li, Y.-Y., Kubota, K., Harada, H., Yu, H.-Q., 2009.

Characterization of start-up performance and archaeal community shifts during anaerobic self-degradation of waste-activated sludge. Bioresource Technology 100 (21), 4981–4988.

Leahy, S.C., Kelly, W.J., Altermann, E., Ronimus, R.S., Yeoman, C.J., Pacheco, D.M., Li, D., Attwood, G.T., 2010. The genome sequence of the rumen methanogen Methanobrevibacter ruminantium reveals new possibilities for controlling ruminant methane emissions. PLoS ONE 5 (1), e8926.

Liu, W.-T., Marsh, T.L., Cheng, H., Forney, L.J., 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied Environmental Microbiology 63 (11), 4516–4522.

Luton, P.E., Wayne, J.M., Sharp, R.J., Riley, P.W., 2002. The mcrA gene as an alternative to 16S rRNA in the phylogenetic analysis of methanogen populations in landﬁll. Microbiology 148 (11), 3521–3530.

Minas, K., McEwan, N.R., Newbold, C.J., Scott, K.P., 2011. Optimization of high throughput CTAB-based protocol for the extraction of qPCR-grade DNA from rumen ﬂuid, plant and bacterial pure cultures. FEMS Microbiology Letters 325 (2), 162–169.

Murphy, J.D., McKeogh, E., Kiely, G., 2004. Technical/economic/environmental analysis of biogas utilisation. Applied Energy 77 (4), 407–427.

Osborn, A., Moore, E., Timmis, K., 2000. An evaluation of terminal-restriction fragment length polymorphism (T-RFLP) analysis for the study of microbial community structure and dynamics. Environmental Microbiology 2 (1), 39–50.

Osorio, F., Torres, J.C., 2009. Biogas puriﬁcation from anaerobic digestion in a wastewater treatment plant for biofuel production. Renewable Energy 43 (10), 2164–2171.

Pecchia, S., Mercatelli, E., Vannacci, G., 1998. PCR ampliﬁcation and characterization of the intergenic spacer region of the ribosomal DNA inPyrenophora graminea.

FEMS Microbiology Letters 166 (1), 21–27.

Quéméneur, M., Hamelin, J., Latrille, E., Steyer, J.-P., Trably, E., 2010. Development and application of a functional CE-SSCP ﬁngerprinting method based on [Fe–

Fe]-hydrogenase genes for monitoring hydrogen-producing Clostridium in mixed cultures. International Journal of Hydrogen Energy 35 (24), 13158–

13167.

Raskin, L., Stromley, J.M., Rittmann, B.E., Stahl, D.A., 1994. Group-speciﬁc 16S rRNA hybridization probes to describe natural communities of methanogens. Applied Environmental Microbiology 60 (4), 1232–1240.

Rastogi, G., Ranade, D.R., Yeole, T.Y., Patole, M.S., Shouche, Y.S., 2008. Investigation of methanogen population structure in biogas reactor by molecular characterization of methyl-coenzyme M reductase A (mcrA) genes.

Bioresource Technology 99 (13), 5317–5326.

Roose-Amsaleg, C.L., Garnier-Sillam, E., Harry, M., 2001. Extraction and puriﬁcation of microbial DNA from soil and sediment samples. Applied Soil Ecology 18 (1), 47–60.

Schlüter, A., Bekel, T., Diaz, N.N., Dondrup, M., Eichenlaub, R., Gartemann, K.-H., Krahn, I., Krause, L.B., Krömeke, H., Kruse, O., Mussgnug, J.H., Neuweger, H., Niehaus, K., Pühler, A., Runte, K.J., Szczepanowski, R., Tauch, A., Tilker, A., Viehöver, P., Goesmann, A., 2008. The metagenome of a biogas-producing microbial community of a production-scale biogas plant fermenter analysed by the 454-pyrosequencing technology. Journal of Biotechnology 136 (1–2), 77–

90.

Schütte, U.M., Abdo, Z., Bent, S.J., Shyu, C., Williams, C.J., Pierson, J.D., Forney, L.J., 2008. Advances in the use of terminal restriction fragment length polymorphism (T-RFLP) analysis of 16S rRNA genes to characterize microbial communities. Applied Microbiology and Biotechnology 80 (3), 365–380.

Schwab, M., 2010. Gasausbeute in landwirtschaftlichen Biogasanlagen. Kuratorium für Technik und Bauwesen in der Landwirtschaft, Darmstadt, Deutschland.

Tang, Y., Shigematsu, T., Morimura, S., Kida, K., 2005. Microbial community analysis of mesophilic anaerobic protein degradation process using bovine serum albumin (BSA)-fed continuous cultivation. Journal of Bioscience and Bioengineering 99 (2), 150–164.

Volant, A., Desoeuvre, A., Casiot, C., Lauga, B., Delpoux, S., Morin, G., Personné, J.C., Héry, M., Elbaz-Poulichet, F., Bertin, P.N., Bruneel, O., 2012. Archaeal diversity:

temporal variation in the arsenic-rich creek sediments of Carnoulès Mine, France. Extremophiles 16 (4), 645–657.

Wirth, R., Kovács, E., Maróti, G., Bagi, Z., Rákhely, G., Kovács, K.L., 2012.

Characterization of a biogas-producing microbial community by short-read next generation DNA sequencing. Biotechnology for Biofuels 5 (41).

Zellner, G., Alten, C., Stackebrandt, E., 1987. Isolation and characterization of Methanocorpusculum parvum, gen. nov., spec. nov., a new tungsten requiring, coccoid methanogen. Archives of Microbiology 147 (1), 13–20.