Environmental Microbiology (2008) 10(5), 1310–1319
    
    doi:10.1111/j.1462-2920.2007.01544.x
    
    Detection of novel Fibrobacter populations in landfill sites and determination of their relative abundance via quantitative PCR
    James E. McDonald, Robert J. Lockhart, Michael J. Cox, Heather E. Allison and Alan J. McCarthy* Microbiology Research Group, School of Biological Sciences, BioSciences Building, University of Liverpool, Crown Street, Liverpool L69 7ZB, UK. Summary Members of the bacterial genus Fibrobacter have long been considered important components of the anaerobic cellulolytic community in the herbivore gut, but their presence and activity in other environments is largely unknown. In this study, a specific polymerase chain reaction (PCR) primer set, targeting the 16S rRNA gene of Fibrobacter spp., was applied to community DNA from five landfill sites followed by temporal thermal gel electrophoresis (TTGE) analysis of cloned amplification products. Phylogenetic analysis of clone sequences indicated the presence of novel clusters closely related to the genus Fibrobacter. There are two named species, Fibrobacter succinogenes and F. intestinalis, and only two of the 58 sequenced clones were identified with them, and both were F. succinogenes. The clone sequences from landfill were recovered in five distinct clusters within the Fibrobacter lineage, and four of these were novel. Quantitative PCR (qPCR) assays of reversetranscribed community RNA from landfill leachates and rumen fluid samples indicated that the abundance of Fibrobacter spp. relative to total bacteria varied from 0.2% to 40% in landfill, and 21% to 32% in the rumen, and these data demonstrate that fibrobacters can be a significant component of the microbial community in landfill ecosystems. This is the first evidence for Fibrobacter spp. outside the gut ecosystem, and as the only cultivated representatives of this group are actively cellulolytic, their diversity and abundance points to a possible role in cellulose hydrolysis in landfill, and perhaps other anaerobic environments also.
    Received 2 October, 2007; accepted 27 November, 2007. *For correspondence. Email aj55m@liv.ac.uk; Tel. (+44) 151 795 4574; Fax (+44) 151 795 4410.
    
    Introduction Since Hungate first isolated Fibrobacter succinogenes (formerly Bacteroides succinogenes) from the rumen in 1947, members of the genus Fibrobacter have been considered to be major degraders of cellulosic plant biomass in the herbivore gut (Hungate, 1966; Stewart and Bryant, 1988). The genus Fibrobacter was established on the basis of 16S rRNA gene sequence data to encompass two species, F. succinogenes and F. intestinalis (Montgomery et al., 1988). Phenotypically these organisms are closely related, being cellulolytic and producing succinic acid as a major fermentation endproduct. However, the genetic diversity of Fibrobacter isolates has been well documented (Amann et al., 1992; Lin et al., 1994; Lin and Stahl, 1995). Beyond the herbivore gut, cellulose is hydrolysed under anoxic conditions in various environments that include freshwater lakes, waterlogged soil, sludge digesters and landfill sites. It has been suggested that clostridia are the predominant cellulose degraders in landfill (Van Dyke and McCarthy, 2002), and this appears to be supported by the absence of Fibrobacter clones in the 16S rRNA gene library of a landfill leachate bioreactor (Burrell et al., 2004). Although the predominance of fibrobacters in the rumen is established, their nucleic acid sequences are often poorly represented in general 16S rRNA gene clone libraries (Tajima et al., 1999; 2000; 2001; Daly et al., 2001) and more recently in ribosomal intergenic spacer clone libraries (Larue et al., 2005). While members of the genus Fibrobacter have never been specifically detected outside the herbivore gut, recent molecular ecological data have demonstrated the presence of Fibrobacter-related organisms in other environments, i.e. cloned 16S rRNA gene sequences belonging to the Acidobacteria–Fibrobacter division have frequently been observed in libraries generated from soil (Nusslein and Tiedje, 1999; Saul et al., 2005) and freshwater ponds (Wise et al., 1997). In addition, members of a lineage related to the genus Fibrobacter have been detected in DNA libraries from the Atlantic and Pacific oceans (Gordon and Giovannoni, 1996). The occurrence and distribution of members of the Fibrobacteres phylum has recently been extended to include termite intestinal contents where cellulose is again the primary carbon
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd
    
    Detection of novel fibrobacters in landfill 1311
    Table 1. 16S rRNA gene primers used for PCR amplification. E. coli numbering 8–28 1522–1542 153–168 1492–1507 153–168 998–1017 341–361 518–536 1369–1386 1492–1510 153–172 238–257 Annealing temperature (°C) 55 48 60 68 60c 60c
    
    Primer set pA pH′ Fibro Bact1512bR Fib1F Fib2AR gc-pC pD′ 1369F Prok 1492R FibroQ153F FibroQ238R
    
    Sequence: 5′-3′a AGAGTTTGATCCTGGCTCAG AAGGAGGTGATCCAGCCGCA CCGTGCCAACGCGCGG TACCTTGTTACGACTT CCGKSCCAACGSSCGG ATCTCTCGCYGCGGCGWTYCC CTACGGGAGGCAGCAGTGGGb GTATTACCGCGGCTGCTG CGGTGAATACGTTCYCGG GGWTACCTTGTTACGACTT CCGKSCCAACGSSCGGHTAA CSCCWACTRGYTAATCRGAC
    
    Specificity General bacteria General bacteria Fibrobacter General bacteria Fibrobacter Fibrobacter General bacteria General bacteria General bacteria General bacteria Fibrobacter Fibrobacter
    
    Reference Edwards et al. (1989) Edwards et al. (1989) Lin and Stahl (1995) Lin and Stahl (1995) This study This study Edwards et al. (1989) Lane et al. (1985) Suzuki et al. (2000) Suzuki et al. (2000) This study This study
    
    a. Ambiguities: K = (G or T); S = (G or C); W = (A or T); Y = (C or T); H = (A, C or T); R = (A or G). b. GC-clamp CGCCCGCCGCGCCCCGCGCCCGGCCCGCCGCCCCCGCCCC (Sheffield et al., 1989). c. QuantiFastTM SYBR® Green PCR assay (Qiagen) uses the same annealing temperature (60°C) for all primer sets.
    
    source for the host organisms (Hongoh et al., 2005; 2006). In view of the under-representation of fibrobacters in rumen clone libraries and the difficulties in isolating these obligately anaerobic organisms, it is possible that their apparent absence from landfill sites is artefactual. Recently, molecular methods were developed to indicate that anaerobic gut fungi (Neocallimastigales) were in fact present in landfill waste sites in which cellulose was the principle carbon source (Lockhart et al., 2006). This implies a role for anaerobic fungi beyond the herbivore gut, and other microbial groups regarded as ‘gut’ organisms may similarly have a wider ecological distribution. The design and application of Fibrobacter genus-specific polymerase chain reaction (PCR) primers would be one approach to circumvent the inherent problems of isolation/cultivation of obligately anaerobic bacteria and the rarity of Fibrobacter 16S rRNA gene sequences in general clone libraries. Here, we have taken this approach to detect and recover 16S rRNA gene sequences from landfill leachates, by nested PCR followed by cloning, sequencing and phylogenetic analysis. Three of the sampling sites (Bromborough Dock, Bidston Moss and Risley) are conventional municipal solid waste (MSW) landfill sites. Drigg is the UK’s principal low-level radioactive waste disposal site, in which cellulose is the main carbon and energy source (Humphreys et al., 1997; Lockhart et al., 2006). The Brogborough site comprised six test cells, each designed to accommodate 15 000 tonnes of waste maintained under different treatment regimes (Caine et al., 1999). Following demonstration of the presence of Fibrobacter DNA, quantitative PCR (qPCR) assays were performed on reverse-transcribed landfill and rumen RNA to determine relative abundance.
    
    Results Nested PCR amplification of landfill DNA extracts with primers Fibro and Bact1215bR Community DNA extracted from landfill leachate samples taken from five separate landfill sites was subjected to nested PCR in which an initial round of amplification with the general bacterial 16S rRNA gene PCR primers pA and pH′ (Table 1) was followed by a second amplification step with the internal Fibrobacter genus-specific primer (Fibro) and the non-specific eubacterial reverse primer Bact1215bR as described by Lin and Stahl (1995). DNA extracted from bovine rumen fluid and a pure culture of F. succinogenes S85 were used as positive controls in all PCR reactions. Amplification products of the expected size (~1354 bp) were obtained from two landfill site leachate samples, excised from agarose gels, purified and cloned into competent Escherichia coli JM109 cells. Of the 17 sequences analysed, nine were found to be chimeric and were therefore not analysed further. All eight clones clustered with members of the genus Fibrobacter, and two clones from separate sites were located within the F. succinogenes lineage (Fig. 1). The use of a general reverse primer in combination with a forward Fibrobacterspecific primer probably accounted for the large proportion of chimeric products obtained, but this approach did nevertheless provide eight bona fide sequences for inclusion in the alignment used to design new specific Fibrobacter oligonucleotides. Nested amplification of landfill DNA extracts with Fib1 and Fib2A Fibrobacter genus-specific PCR primers The Fib1 and Fib2A primer set were designed using 16S rRNA gene sequences for all cultured Fibrobacter isolates on the RDP website, plus the landfill Fibrobacter
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    1312 J. E. McDonald et al.
    Fig. 1. Maximum-likelihood tree of 16S rRNA gene sequences amplified from landfill sites and bovine rumen fluid using Fibrobacter genus-specific PCR primers. An alignment was produced using CLUSTALX (v. 1.8; Thompson et al., 1997) and imported into ARB (beta v. 2003-08-22; Ludwig et al., 2004) where it was manually optimized. Positions corresponding to PCR primers, gaps and hypervariable bases were removed leaving a final alignment of 536 bp corresponding to the regions of the 16S rRNA gene between 153 and 1017 (E. coli numbering). Alignments were exported and used to calculate phylogenetic trees using a range of different techniques. The above tree shows the output of two of these methods, a neighbour-joining tree with LogDet correction and 100 bootstrap samplings and a maximum-likelihood tree with HKY correction and 100 bootstrap samplings. Both trees were imported to ARB, rooted with members of the Bacteroidetes and the tree topologies and bootstrap values compared. Nucleotide sequence accession numbers are displayed in parentheses. Sequences obtained from the five landfill sites, and the bovine rumen fluid and F. succinogenes S85 control DNA used in this study are displayed in bold. The clustering of sequences was reproducible for both maximum-likelihood and neighbour-joining treeing methods (data not shown). Nodes in which bootstrap values >95% were obtained by both treeing methods are shown as filled circles and those between 75% and 95% as unfilled circles. Semi-filled circles denote nodes at which one treeing algorithm provided a bootstrap value >95% and the other between 75% and 95%. The scale bar indicates 0.1 base substitutions per nucleotide.
    
    clone sequences generated with primers Fibro and Bact1215bR. In addition, the primers were specific for a large number of the uncultured rumen bacterial sequences classified in the genus Fibrobacter by RDP, excluding those database sequences that were found to be chimeric. Landfill DNA extracts were again amplified via nested PCR, the second round with the Fib1 and Fib2A primer set (Table 1). Single amplification products of the expected size (~855 bp) were obtained from DNA extracted from two conventional landfill sites (Bidston Moss and Bromborough Dock), the radioactive waste site at Drigg and all six Brogborough landfill test cells, in addition to the rumen fluid and F. succinogenes S85 DNA preparations. Temporal thermal gel electrophoresis (TTGE) analysis of landfill clones Amplification products were cloned into E. coli JM109 cells and a total of 82 clones from the five sample sites screened using temporal thermal gel electrophoresis (TTGE). Differences in the gel position at which clone templates denatured were observed (data not shown), and this criterion was used to select novel bands for DNA sequencing. A temperature range of 50–54°C and a ramp rate of 0.2°C h-1 provided the greatest discrimination of bands. In total, 50 clones were chosen for sequencing in both directions on the basis of their TTGE banding pattern. Phylogenetic analysis of clone sequences None of the 50 sequenced clones were found to be chimeric. Clone sequences and their closest BLASTN matches were subjected to phylogenetic analysis alongside the RDP Fibrobacter sequences. Trees were constructed using maximum-likelihood and neighbourjoining methods; the branching patterns of the sequences were unaffected by the clustering algorithm used (data not shown). Sequences belonging to Fibrobacteres subphylum 2, which have been recently identified in termite gut environments in the phylogenetic
    
    analysis of Hongoh and colleagues (2006), were also included. The clone sequences generated using the genus-specific primer set reported in this study formed four clusters that were distinct from those containing rumen fibrobacters, and this fundamental separation was supported by bootstrap values of > 95%. Fourteen clone sequences failed to generate a band on TTGE gels and all were recovered in a discrete cluster (Fig. 1) whose closest relative was the F. intestinalis branch of the ‘gut’ fibrobacters. TTGE profiles of individual clone sequences demonstrated real predictive value for the phylogenetic identity of the Fibrobacter sequences. For example, all Drigg clones (K1–K8) and clone I10 (Bidston Moss) showed similar TTGE gel patterns which were very different from all of the other clones, and these sequences occupy a distinct lineage (Fig. 1). The largest cluster contained the highest number of sequenced clones drawn from across the sampling sites, including six of the eight clone sequences initially obtained with the semi-specific Fibro and Bact1215bR primer set (Fig. 1). Although there does appear to be some correlation between sample sites and clustering of clones, this has not been the systematic study that would be required to compare Fibrobacter population structures between sites. Clones Brom F12 and Risley H13, generated previously with the Fibro and Bact1215bR primer set, were the only landfill sequences that clustered with F. succinogenes. We also sequenced 10 clones obtained from the Fib1/ Fib2A PCR control DNA (eight amplified from bovine rumen fluid and two from F. succinogenes S85) and all clustered within the F. succinogenes branch of the Fibrobacter lineage (Fig. 1). qPCR quantification of Fibrobacter spp. in environmental samples Community RNA extracts from seven leachate samples from different locations across two landfill sites, and two ovine rumen fluid samples, were reverse transcribed and subjected to qPCR assay. Two primer sets, one Fibrobacter specific and one targeting general bacteria
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    Detection of novel fibrobacters in landfill 1313
    Key Bidston Moss / Bromborough Dock / Brogborough / Risley cluster Bromborough Dock cluster Drigg / Bidston Moss cluster Fibrobacter succinogenes Bidston Moss cluster Fibrobacter intestinalis
    Brom Dock J29 (EF186251) Brog Cell 6 M1 (EF186264) Brom Dock J2 (EF186250) Brom Dock F11 (EF190824) Brog Cell 6 M12 (EF186267) Brom Dock F10 (EF190823) Risley G2 (EF190829) Brog Cell 6 M9 (EF186293) Brom Dock J12 (EF186284) Brom Dock D1 (EF190822) Brom Dock J7 (EF186282) Brog Cell 6 M2 (EF186265) Brom Dock F17 (EF190827) Brom Dock J1 (EF186249) Brog Cell 1 L14 (EF186263) Brog Cell 1 L12 (EF186261) Brog Cell 1 L1 (Ef186259) Brom Dock J9 (EF186283) Brog Cell 6 M8 (EF186292) Brom Dock J25 (EF186286) Brog Cell 1 L5 (EF186291) Brom Dock J31 (EF186253) Brog Cell 6 M3 (EF186266) Brog Cell 1 L2 (EF186260) Brog Cell 1 L13 (EF186262) Brom Dock F7 (EF190825) Bidston I4 (EF186247) Bidston I1 (EF186244) Bidston I8 (EF186270) Bidston I16 (EF186278) Brom Dock J15 (EF186287) Brom Dock J13 (EF186285) Brom Dock J30 (EF186252) Bidston I10 (EF186272) Drigg K4 (EF186254) Drigg K6 (EF186256) Drigg K3 (EF186290) Drigg K2 (EF186289) Drigg K1 (EF186288) Drigg K7 (EF186257) Drigg K5 (EF186255) Drigg K8 (EF186258) F. succinogenes strain BL2 (AJ505937) F. succinogenes succinogenes strain B1 (M62684) Bovine rumen 4 (EF186239) Bovine rumen 5 (EF186240) Bovine rumen 2 (EF186237) F. succinogenes succinogenes strain A3C (M62683) Bovine rumen 3 (EF186238) Bovine rumen 1 (EF186236) F. succinogenes strain AS213 (AB275489) F. succinogenes strain R (AJ505938) F. succinogenes S85 strain ATCC 19169 (AJ496032) F. succinogenes strain S85 (M62696) F. succinogenes strain RS223 (AB275509) F. succinogenes strain RS225 (AB275511) F. succinogenes S85 2 (EF186235) Risley H13 (EF190828) F. succinogenes strain S85 (M62696) Bovine rumen 8 (EF186243) Bovine rumen 7 (EF186242) F. succinogenes strain AS220 (AB275491) F. succinogenes S85 1 (EF186234) F. succinogenes strain AS226 (AB275494) Brom Dock F12 (EF190826) F. succinogenes strain HM2 (M62689) F. succinogenes strain AS211(AB275485) Bovine rumen 6 (EF186241) Fibrobacter sp.(L35548) Fibrobacter sp.(L35547) Bidston I9 (EF186271) Bidston I2 (EF186245) Bidston I14 (EF186276) Bidston I7 (EF186269) Bidston I21 (Ef186281) Bidston I18 (EF186279) Bidston I15 (EF186277) Bidston I5 (EF186248) Bidston I3 (EF186246) Bidston I12 (EF186274) Bidston I6 (EF186268) Bidston I20 (EF186280) Bidston I11 (EF186273) Bidston I13 (EF186275) F. intestinalis strain NR9 (M62695) F. intestinalis strain JG1 (M62690)
    
    1 2 3 4 5 6
    
    1
    
    2
    
    3
    
    4
    
    5
    
    8
    
    Fibrobacteres subphylum 2
    
    6
    
    Nitrospira sp. strain RC7 (Y14640) Acidobacteriaceaebacterium KP3 (AY765992) Terriglobus roseus strain KBS 63 (DQ660892) Acidobacteriaceaebacterium Gsoil 969 (AB245337) 18 Candidate division TG3 Treponema sp. strain VI:G:G47 (AF056343) Borrelia burgdorferi (X85189) Clostridium cellulovorans strain DSM 3052 (X73438) Clostridium butyricum strain MW8 (AJ002592) Eubacterium barkeri (M23927) Bacillus sp. strain TKSP21(AB017591) Bacillus sp. strain KPU 0013 (AB067810) Bacillus cereus (X55063) Holophaga foetida strain TMBS4-T (X77215) Geothrix fermentans (GFU41563) Capnocytophagaochracea (L14635) Flavobacterium gondwanense (M92278) Cytophaga fermentans (M58766) Bacteroides fragilis strain ATCC 25285T (X83935) Bacteroides fragilis strain TAL3636 (X83941)
    
    0.10
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    1314 J. E. McDonald et al.
    
    30 25 20
    
    Fibrobacter Slope = -3.372 Amp. Eff. = 1.98 y-intercept = 39.752 r2 = 0.9961
    
    Ct
    
    Fig. 2. Standard curves generated from triplicate serial dilutions of known 16S rRNA gene copies (3 ¥ 108 to 3 ¥ 104) of F. succinogenes S85 using general bacterial ( , solid line) and Fibrobacter genus-specific ( , dashed line) primer sets. Each point represents mean Ct (threshold cycle) values of at least two of the serial dilutions. Error bars represent standard deviations.
    
    15 10 5 0 4
    
    General bacteria Slope = -3.433 Amp. Eff. = 1.96 y-intercept = 38.846 r2 = 0.9997
    
    5
    
    6
    
    7
    
    8
    
    9
    
    Log 16S rRNA Gene Copy Number
    
    (Table 1), were used to generate standard curves of known 16S rRNA gene copy numbers of F. succinogenes S85 control DNA. The amplification efficiencies of the control DNA sample were found to be comparable for each primer set (Fig. 2). As recommended by Smith and colleagues (2006), we report that in all assays performed using the general bacterial primer set, specific products were observed in no template control (NTC) assays at a Ct value of 27.08 1.43 and none of the standard dilutions or samples were within this range. 16S rRNA gene copy numbers for samples with each primer set were obtained from the associated standard curves and used to determine relative abundance as a percentage. Fibrobacters in landfill samples were generally in the range 0.2–1.5%. The Bromborough Dock riser 4 sample gave a reproducible relative abundance value of 40% (Fig. 3) and significantly, this was the only landfill leachate sample from which PCR amplification products were obtained using the Fib1F and Fib2AR primer set directly on DNA extracts. Nested PCR was required to generate detectable amplification products for all other landfill samples (data not shown). The abundance of Fibrobacter in the ovine rumen samples A and B was 32% and 21%, respectively (Fig. 3), supporting the view that Fibrobacter spp. are predominant members of the rumen microbial community. The unusually high relative abundance of Fibrobacter RNA in the Bromborough 4 sample was confirmed in an experiment in which qPCR was performed on community DNA extracts. Only the ovine rumen samples and the Bromborough 4 landfill sample contained sufficient template DNA for quantitative detection of amplification products (data not shown). Similarly, DNA from rumen fluid samples always gave amplification products by direct PCR with the Fib1/Fib2A primer pair without the need to
    
    carry out a primary amplification step with universal bacterial 16S rRNA gene primers.
    
    Discussion Molecular detection methods targeting the 16S rRNA gene were applied here to detect and quantify Fibrobacter spp. in a number of different landfill sites. To our knowledge, this is the first report of members of the genus
    
    100
    
    Relative Abundance
    
    10
    
    1
    
    0.1
    
    Ov i ne Ov Ru ine me Ru n A m en B Br om Br Ris o m er Br Ris 1 o m er Br Ris 2 om er Br Ris 3 om e r Ri 4 se r5
    
    Fig. 3. Percentage relative abundance of 16S rRNA genes of Fibrobacter spp. compared with total bacteria in landfill and rumen samples. Community RNA was reverse-transcribed and each sample assayed in triplicate via qPCR with both primer sets and on two separate PCR plates. Data are means of six assays with the exception of Brom Dock Riser 2 (two assays) and Riser 1 (three assays). Error bars show the standard deviations.
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    Ri sle R i y 8. sle 2 y1 0B
    
    Detection of novel fibrobacters in landfill 1315 Fibrobacter outside the gut ecosystem. In ruminant nutrition, where F. succinogenes is established as one of the predominant cellulose-degrading microorganisms (Hungate, 1966; Stewart and Bryant, 1988), cell numbers obtained by anaerobic culture of rumen microbes suggest a less significant contribution to cellulose hydrolysis. Fibrobacter spp. are difficult to isolate, and are poorly represented in clone libraries generated from universal or general bacterial 16S rRNA gene PCR primers (Tajima et al., 2001). In a recent study by Larue and colleagues (2005), using ribosomal intergenic spacer analysis (RISA) of microorganisms adherent to plant biomass in the herbivore gastrointestinal tract, the absence of Fibrobacter sequences in clone libraries was reported. However, genus-specific DGGE for Fibrobacter spp. confirmed the presence of these organisms in all community DNA samples used to generate the libraries, and the F. succinogenes sequences were found to have no mismatches with the oligonucleotides used to produce the library (Larue et al., 2005). Thus, the presence of members of the Fibrobacter genus in anaerobic environments such as landfill sites may not have been recorded previously due to these culture and PCR amplification difficulties. A number of studies have attempted to characterize the indigenous cellulolytic microbial populations in landfill sites. However, with the exception of one culture-based study (Westlake et al., 1995) in which 37 strains of cellulolytic bacteria were isolated, other studies have focused on the generation of 16S rRNA gene clone libraries (Burrell et al., 2004; Huang et al., 2004; 2005). This has led to the accepted dogma that clostridia are the predominant cellulolytic microorganisms in landfill. Although this may certainly be the case, the application of genusspecific PCR primers for fibrobacters resulting in their detection here raises the question of whether these organisms have a role in cellulose hydrolysis in landfill environments, and to what extent. It may be that, as in the rumen, the clostridia are not the most important cellulose degraders, but are readily isolated and well represented in general bacterial clone libraries. Although cultured Fibrobacter strains are all actively cellulolytic, there is as yet no evidence that members of the genus detected only by phylogenetic analysis of environmental DNA sequences possess the cellulolysis phenotype. The data presented here show that members of the genus Fibrobacter or their close relatives occupy a much wider ecological range than is currently acknowledged, and are not restricted to the gut environment. TTGE and phylogenetic analysis of clone inserts demonstrated a high degree of diversity in landfill sequences. With the exception of two sequences, all clones formed clusters distinct from the known rumen Fibrobacter spp. indicating differences between rumen and landfill fibrobacters. None of the sequences from landfill samples were recovered in Fibrobacteres subphylum 2 (Fig. 1), named by Hongoh and colleagues (2006) to better describe the diversity of termite gut Fibrobacteres. In contrast, the clusters identified from landfill samples here are alongside the F. succinogenes and F. intestinalis lineages (Fig. 1) designated as Fibrobacteres subphylum 1 by Hongoh and colleagues (2006). These clusters represent new centres of variation within the genus Fibrobacter, as currently defined, although it has been suggested that fibrobacters probably comprise a suprageneric taxon (Amann et al., 1992), and this may certainly be the case. Detection of Fibrobacter sequences, particularly by nested PCR, does not necessarily indicate the presence of a significant and active population. Quantitative PCR was used here on cDNA produced from RNA extracts to ensure that sufficient template was available and to possibly avoid any poor amplification efficiencies due to DNAassociated molecules, the only hypothesis offered thus far (Tajima et al., 2001) to explain the poor representation of Fibrobacter sequences in general bacterial 16S rRNA gene libraries. The reverse transcriptase qPCR assays were validated by the data in Fig. 2 and by using ovine rumen RNA to obtain relative abundance values for fibrobacters that compare favourably with the range of values reported elsewhere for herbivore gut contents (Lin et al., 1994; Lin and Stahl, 1995; Ziemer et al., 2000; Daly and Shirazi-Beechey, 2003). Although the relative abundance values are understandably much lower for landfill samples (Fig. 3), they do confirm the presence of an indigenous Fibrobacter community in landfill sites. Application of the new primer sets and the techniques described here will enable the detection and quantification of Fibrobacter populations in the herbivore gut, and possibly elsewhere. There is always biotechnological interest in novel strains of bacteria as a source of new cellulases, and this work provides an impetus to attempt the isolation and cultivation of those fibrobacters now known to be present in landfill sites. Experimental procedures Sampling
    Leachate samples (1 l) from Bidston Moss, Risley and Bromborough Dock municipal waste landfill sites in the northwest of England were obtained in October 2005. DNA was extracted from leachate, as described below, immediately upon return to the laboratory. Drigg is the UK’s principal disposal facility for solid low-level radioactive waste and lacks readily putrescible organic matter, but possesses a significant cellulose content. Samples of sediment/leachate were collected (June 2002) from gas vent pipes which are driven into the waste at Drigg to allow monitoring of gas production. Leachate samples (1 l) were also obtained from the six Brogborough test cells initiated in 1986 by the UK Department of
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    1316 J. E. McDonald et al.
    Energy. Each cell contained 15 000 tonnes of waste and was 40 m long, 25 m wide and 20 m deep. The objective of the test cells was to investigate the application of different waste treatment methods to accelerate waste degradation, ultimately enhancing energy production from landfill gas (Caine et al., 1999). amplification. Polymerase chain reaction products were visualized using 1% agarose gel electrophoresis and product size determined using GeneRuler™ 100 bp DNA Ladder Plus (MBI Fermentas).
    
    Cloning and sequencing of 16S rRNA gene fragments
    Amplification products obtained using Fibrobacter-specific primers from the above nested PCR reactions were extracted and purified from agarose gels using a GenEluteTM Gel Extraction Kit (Sigma). Ligation and cloning of 16S rRNA gene PCR products into competent E. coli JM109 (Promega) cells was performed using a pGEM®-T Easy Vector System I (Promega) according to the manufacturer’s protocol. Plasmid DNA was extracted from overnight clone cultures (LB broth and 100 mg ml-1 ampicillin) using a QIAprep® Spin Miniprep Kit (Qiagen) following the manufacturer’s protocol and sequenced in both directions by MWG Biotech or Macrogen.
    
    DNA extraction
    The 1 l volumes of landfill leachate were centrifuged at 12 000 g for 1 h and pellets re-suspended in a small volume of leachate. DNA was extracted using an UltraCleanTM Soil DNA Kit (MOBIO) or FastDNA® SPIN® kit for soil (BIO 101) according to the manufacturer’s protocol and stored at -80°C. DNA was similarly extracted from Drigg samples as described previously (Lockhart et al., 2006).
    
    Design of Fibrobacter genus-specific PCR primers
    Using the semi-specific primer set (Fibro and Bact1512bR) designed by Lin and Stahl (1995), we were able to detect and recover Fibrobacter sequences from landfill sites for use in the design of the new genus-specific primer set (Fib1 and Fib2A) described below. These were used in a nested PCR protocol after initial amplification with general bacterial 16S rRNA gene primers, pA and pH′ (Table 1). CLUSTALX (Thompson et al., 1997) was used to align all 25 cultured Fibrobacter isolate sequences available on the Ribosomal Database Project site in addition to the eight landfill fibrobacter clones generated using the Fibro and Bact1512bR primer set. Four degeneracies were added to the Fibro (forward) primer sequence of Lin and Stahl (1995) and the reverse Fibrobacter-specific primer was designed to span the V3 variable region and U5 conserved region (E. coli positions 998–1017) of the 16S rRNA gene. Primer specificity was confirmed using the RDP Probe Match facility (Cole et al., 2005) and subsequently by sequence analysis of cloned amplification products.
    
    TTGE analysis of clones
    Clone insert PCR products obtained using Fib1 and Fib2A were diluted 100-fold and 1 ml of diluted product was added to a 50 ml PCR reaction containing 0.2 mM each of primers gc-pC and pD (Table 1), 0.2 mM each dNTP, 1¥ Phusion HF buffer (Finnzymes), 3% DMSO, 1¥ BSA, 1 unit Phusion™ High-Fidelity DNA Polymerase (Finnzymes) and ddH2O. Polymerase chain reaction cycling conditions were as follows: 98°C for 30 s, 30 cycles of 98°C for 10 s, 68°C for 30 s, 72°C for 20 s and a final extension of 72°C for 8 min. Polymerase chain reaction products were visualized via 1% agarose gel electrophoresis, and 250 ng of PCR product for each clone sequence was subjected to TTGE analysis performed using the DecodeTM universal mutation detection system (Bio-Rad). Gels consisted of 6% acrylamide (37:1 acrylamide: bisacrylamide), 7 M urea, 20% deionized formamide, 2% glycerol and 1.25¥ TAE buffer [50 mM Tris, 25 mM acetic acid, 1.25 mM Na2EDTA (pH 8)]. Products were run at 65 V for 16–18 h along a temperature gradient of 50–54°C and ramp rate of 0.2°C h-1. Gels were stained using SYBR® green I nucleic acid stain (Molecular Probes) 1:50 000 (v/v) in 1.25 TAE buffer (pH 8) for 30 min and visualized using a STORM 860 optical scanner and its associated ImageQuant software (Molecular Dynamics).
    
    PCR amplification
    Polymerase chain reactions were performed in 50 ml volumes containing: 0.2 mM each primer, 0.2 mM each dNTP, 1¥ SuperTaq Buffer (HT Biotechnologies), 0.5 mM MgCl2, 1¥ BSA, approximately 50 ng of DNA template, 1 unit SuperTaq DNA Polymerase (HT Biotechnologies) and ddH2O. Polymerase chain reaction cycling conditions were as follows: initial denaturation at 94°C for 5 min, 35 cycles of 94°C for 1 min, 1 min at the specific annealing temperature for each primer set (Table 1), 72°C for 1.5 min. Final extension was performed at 72°C for 10 min. Initial amplification of bacterial 16S rRNA genes was performed using the primers pA and pH′ (Edwards et al., 1989). A second round of PCR amplification was performed on 16S rRNA gene amplification products (1 ml of PCR product) using the Fibrobacter genus-specific primer set. In PCR reactions with the Fibro and Bact1215bR primer set, an increased MgCl2 concentration of 1.5 mM was used. Primers Fib1 and Fib2A were used at a concentration of 0.4 mM each. Only 30 cycles were used in the second ‘nested’ round of PCR
    
    Nucleotide sequence accession numbers
    These sequence data have been submitted to the GenBank database under Accession No. EF186234–EF186293 and EF190822–EF190829.
    
    Phylogenetic analysis of 16S rRNA gene sequences
    Forward and reverse clone insert DNA sequences were assembled into contigs using PreGap 4 and Gap 4 software (Staden, 1996) and base calling was visually checked using the sequencing traces. Assembled contigs were subjected to two chimera check packages, RDP Chimera Check (Cole et al., 2005) and Pintail (Ashelford et al., 2005). Sequences were aligned using CLUSTALX (v. 1.8, Thompson et al., 1997)
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    Detection of novel fibrobacters in landfill 1317
    and imported into ARB (beta v. 2003-08-22; Ludwig et al., 2004) where the alignment was manually optimized. Phylip (v. 3.65; Felsenstein, 2005) was used to calculate a neighbour-joining tree with 100 bootstrap samplings. The Seqboot, Dnadist, Neighbor and Consense packages were used in sequence and Dnadist was run with the LogDet substitution model for evolutionary change. A maximumlikelihood tree was obtained using PhyML (v. 2.4.4, Guindon et al., 2005) run as a standalone package with 100 bootstrap samplings, the HKY substitution model, empirical base frequency estimates, fixed transition-to-transversion (Ts/tv) ratio (4.00), fixed proportion of invariable sites (p-inv = 0.00), and four substitution rate categories with fixed gamma distribution (alpha = 2.00). sisted of 95°C for 5 min, followed by 45 cycles of 95°C for 10 s, and 60°C for 30 s. Fluorescence was detected during the combined annealing extension step. A dissociation step was included at the end of every run to confirm the presence of single amplification products. Each sample was run in triplicate (reproducibility) for each primer set on the same plate and each plate was run in duplicate to confirm the repeatability of assays. In addition to dissociation curves, amplification products from one of the two duplicate plates were visualized on 2% agarose gels to further confirm the presence of single amplification products. The relative abundance of Fibrobacter spp. in community nucleic acid extracts was determined using amplified 16S rRNA genes from F. succinogenes S85 control DNA. Primers pA and pH′ were used to amplify the almost full-length 16S rRNA gene using PhusionTM Taq polymerase (Finnzymes), as described previously. The specific amplification product was excised from an agarose gel and purified using a Perfectprep® Gel Cleanup kit (Eppendorf). The DNA concentration of control DNA was determined using a Quant-iTTM PicoGreen® dsDNA kit (Invitrogen) and the number of 16S rRNA genes per microlitre calculated. Standard curves of the F. succinogenes S85 16S rRNA gene were generated in triplicate using 10-fold serial dilutions of 3 ¥ 108 to 3 ¥ 104 gene copies, and all three serial dilutions were run on each plate with both primer sets. Ct values for each dilution were plotted against log gene copy number to generate standard curves for each primer set. A linear line of best fit was used to determine the y-intercept, amplification efficiency and r2 value of the standard with each primer set (Pfaffl, 2001). Standard curves for general bacterial and Fibrobacterspecific primer sets demonstrated comparable amplification efficiencies for the target gene (Fig. 2). Relative abundances (%) were determined by dividing the number of gene copies per sample (Fibrobacter-specific primers) by the average gene copy per sample (general bacterial primers), using the relevant standard curve for each primer set (Smits et al., 2004).
    
    Design of Fibrobacter-specific qPCR primers
    Primer FibroQ238R (Table 1) was designed manually using the Arb alignment of all sequences in Fig. 1. The primer targets semi-conserved region S1b of the 16S rRNA gene, corresponding to E. coli positions 238–257. The ‘Probe match’ feature of the RDP release 9 was used to test primer specificity, and showed that in addition to the genus Fibrobacter, the FibroQ238R primer targets some members of the Proteobacteria, Firmicutes, Bacteroidetes and Cyanobacteria in addition to a small proportion of sequences from other phyla. Primer Fib1F was optimized for qPCR using the Arb alignment generated in this study by the addition of four bases at the 3′ end. The optimized primer FibroQ153F (Table 1) is a 20 mer, and the RDP ‘Probe match’ function confirmed it was specific only for members of the genus Fibrobacter.
    
    Determination of the relative abundance of Fibrobacter spp. in landfill and rumen environments using qPCR
    Community DNA and RNA was co-extracted in duplicate from landfill leachate and ovine rumen fluid samples using the method of Griffiths and colleagues (2000). One nucleic acid extract was RNase treated and used as the community DNA sample. Each sample was diluted to a final concentration of 50 ng DNA ml-1 and 1 ml of extract (50 ng template) was subsequently added to the qPCR assay. The remaining nucleic acid extract was DNase treated using a Turbo DNA-free™ Kit (Ambion) for community RNA analysis. DNase-treated RNA samples (1 mg) were reverse transcribed using BioScript RNase H Low (Bioline), following the manufacturer’s protocol. For reverse-transcribed RNA, 1 ml of the reverse transcription (RT) reaction was added to the qPCR and a no-RT control consisting of an equivalent amount of RNA from the initial extract for each sample was also included to confirm the absence of contaminating DNA. Quantitative PCR was performed in triplicate with general bacterial and Fibrobacter-specific PCR primer sets (1369F, Prok 1492R and FibroQ153F, FibroQ238R respectively) on the same 96-well plate using the 7500 FAST Real-Time PCR system (Applied Biosystems). Each reaction contained either 50 ng of DNA template or 1 ml of an RT reaction, 12.5 ml of 2¥ QuantiFastTM SYBR® Green PCR Master Mix (Qiagen), forward and reverse primers to a final concentration of 1 mM and water to a final volume of 25 ml. Cycling conditions con-
    
    Acknowledgements
    This research was funded by the Natural Environment Research Council (NERC) and Nexia Solutions, both in the UK. The authors would like to thank the landfill site operators for providing the leachate samples analysed in this study. We are also grateful to Colin Stewart and Sylvia Duncan (Rowett Research Institute, Aberdeen) for providing F. succinogenes S85 biomass, and Dr Matthew Spencer (University of Liverpool) for advice on phylogenetic analysis.
    
    References
    Amann, R.I., Lin, C.H., Key, R., Montgomery, L., and Stahl, D.A. (1992) Diversity among Fibrobacter isolates – towards a phylogenetic classification. Syst Appl Microbiol 15: 23–31. Ashelford, K.E., Chuzhanova, N.A., Fry, J.C., Jones, A.J., and Weightman, A.J. (2005) At least 1 in 20, 16S rRNA sequence records currently held in public repositories is estimated to contain substantial anomalies. Appl Environ Microbiol 71: 7724–7736.
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    1318 J. E. McDonald et al.
    Burrell, P.C., O’Sullivan, C., Song, H., Clarke, W.P., and Blackall, L.L. (2004) Identification, detection, and spatial resolution of Clostridium populations responsible for cellulose degradation in a methanogenic landfill leachate bioreactor. Appl Environ Microbiol 70: 2414–2419. Caine, M., Campbell, D., and van Santen, A. (1999) The landfill gas timeline: the Brogborough test cells. Waste Manag Res 17: 430–442. Cole, J.R., Chai, B., Farris, R.J., Wang, Q., Kulam, S.A., McGarrell, D.M., et al. (2005) The Ribosomal Database Project (RDP-II): sequences and tools for high-throughput rRNA analysis. Nucleic Acids Res 33: D294–D296. Daly, K., and Shirazi-Beechey, S.P. (2003) Design and evaluation of group-specific oligonucleotide probes for quantitative analysis of intestinal ecosystems: their application to assessment of equine colonic microflora. FEMS Microbiol Ecol 44: 243–252. Daly, K., Stewart, C.S., Flint, H.J., and Shirazi-Beechey, S.P. (2001) Bacterial diversity within the equine large intestine as revealed by molecular analysis of cloned 16S rRNA genes. FEMS Microbiol Ecol 38: 141–151. Edwards, U., Rogall, T., Blocker, H., Emde, M., and Bottger, E.C. (1989) Isolation and direct complete nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal RNA. Nucleic Acids Res 17: 7843–7853. Felsenstein, J. (2005) PHYLIP (Phylogeny Inference Package), Version 3.6. Distributed by the author. Seattle, WA, USA: Department of Genome Sciences, University of Washington. Gordon, D.A., and Giovannoni, S.J. (1996) Detection of stratified microbial populations related to Chlorobium and Fibrobacter species in the Atlantic and Pacific Oceans. Appl Environ Microbiol 62: 1171–1177. Griffiths, R., Whiteley, A., O’Donnell, A., and Bailey, M. (2000) Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal DNAand rRNA-based microbial community composition. Appl Environ Microbiol 66: 5488–5491. Guindon, S., Lethiec, F., Duroux, P., and Gascuel, O. (2005) PHYMLOnline – a web server for fast maximum likelihoodbased phylogenetic inference. Nucleic Acids Res 33: W557–W559. Hongoh, Y., Deevong, P., Inoue, T., Moriya, S., Trakulnaleamsai, S., Ohkuma, M., et al. (2005) Intra- and interspecific comparisons of bacterial diversity and community structure support coevolution of gut microbiota and termite host. Appl Environ Microbiol 71: 6590–6599. Hongoh, Y., Deevong, P., Hattori, S., Inoue, T., Noda, S., Noparatnaraporn, N., et al. (2006) Phylogenetic diversity, localization, and cell morphologies of members of the candidate phylum TG3 and a subphylum in the phylum Fibrobacteres, recently discovered bacterial groups dominant in termite guts. Appl Environ Microbiol 72: 6780– 6788. Huang, L.N., Zhou, H., Zhu, S., and Qu, L.H. (2004) Phylogenetic diversity of bacteria in the leachate of a fullscale recirculating landfill. FEMS Microbiol Ecol 50: 175– 183. Huang, L.N., Zhu, S., Zhou, H., and Qu, L.H. (2005) Molecular phylogenetic diversity of bacteria associated with the leachate of a closed municipal solid waste landfill. FEMS Microbiol Lett 242: 297–303. Humphreys, P., McGarry, R., Hoffmann, A., and Binks, P. (1997) DRINK: a biogeochemical source term model for low level radioactive waste disposal sites. FEMS Microbiol Rev 20: 557–571. Hungate, R.E. (1966) The Rumen and Its Microbes. New York, USA: Academic Press. Lane, D.J., Pace, B., Olsen, G.J., Stahl, D.A., Sogin, M.L., and Pace, N.R. (1985) Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc Natl Acad Sci USA 82: 6955–6959. Larue, R., Yu, Z.T., Parisi, V.A., Egan, A.R., and Morrison, M. (2005) Novel microbial diversity adherent to plant biomass in the herbivore gastrointestinal tract, as revealed by ribosomal intergenic spacer analysis and rrs gene sequencing. Environ Microbiol 7: 530–543. Lin, C., and Stahl, D.A. (1995) Taxon-specific probes for the cellulolytic genus Fibrobacter reveal abundant and novel equine-associated populations. Appl Environ Microbiol 61: 1348–1351. Lin, C.Z., Flesher, B., Capman, W.C., Amann, R.I., and Stahl, D.A. (1994) Taxon specific hybridization probes for fiber-digesting bacteria suggest novel gut-associated Fibrobacter. Syst Appl Microbiol 17: 418–424. Lockhart, R.J., Van Dyke, M.I., Beadle, I.R., Humphreys, P., and McCarthy, A.J. (2006) Molecular biological detection of anaerobic gut fungi (Neocallimastigales) from landfill sites. Appl Environ Microbiol 72: 5659–5661. Ludwig, W., Strunk, O., Westram, R., Richter, L., Meier, H., Yadhukumar, B., et al. (2004) ARB: a software environment for sequence data. Nucleic Acids Res 32: 1363–1371. Montgomery, L., Flesher, B., and Stahl, D. (1988) Transfer of Bacteroides succinogenes (Hungate) to Fibrobacter GenNov as Fibrobacter succinogenes Comb Nov and description of Fibrobacter Intestinalis Sp-Nov. Int J Syst Bacteriol 38: 430–435. Nusslein, K., and Tiedje, J.M. (1999) Soil bacterial community shift correlated with change from forest to pasture vegetation in a tropical soil. Appl Environ Microbiol 65: 3622– 3626. Pfaffl, M.W. (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: e45. Saul, D.J., Aislabie, J.M., Brown, C.E., Harris, L., and Foght, J.M. (2005) Hydrocarbon contamination changes the bacterial diversity of soil from around Scott Base, Antarctica. FEMS Microbiol Ecol 53: 141–155. Sheffield, V.C., Cox, D.R., Lerman, L.S., and Mayers, R.M. (1989) Attachment of a 40-base-pair G+C rich sequence (GC-clamp) to genomic DNA fragments by polymerase chain reaction results in improved detection of single-base changes. Proc Natl Acad Sci USA 86: 232–236. Smith, C.J., Nedwell, D.B., Dong, L.F., and Osborn, A.M. (2006) Evaluation of quantitative polymerase chain reaction-based approaches for determining gene copy and gene transcript numbers in environmental samples. Environ Microbiol 8: 804–815. Smits, T.H.M., Devenoges, C., Szynalski, K., Maillard, J., and Holliger, C. (2004) Development of a real-time PCR method for quantification of three genera Dehalobacter,
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319
    
    Detection of novel fibrobacters in landfill 1319
    Dehalococcoides, and Desulfitobacterium in microbial communities. FEMS Microbiol Ecol 57: 369–378. Staden, R. (1996) The Staden Sequence Analysis Package. Mol Biotechnol 5: 233–241. Stewart, C.S., and Bryant, M.P. (1988) The rumen bacteria. In The Rumen Microbial Ecosystem. Hobson, P.N. (ed.). New York, USA: Elsevier Appl Sci, pp. 21–75. Suzuki, M.T., Taylor, L.T., and DeLong, E.F. (2000) Quantitative analysis of small-subunit rRNA genes in mixed microbial populations via 5′-nuclease assays. Appl Environ Microbiol 66: 4605–4614. Tajima, K., Aminov, R.I., Nagamine, T., Ogata, K., Nakamura, M., Matsui, H., and Benno, Y. (1999) Rumen bacterial diversity as determined by sequence analysis of 16S rDNA libraries. FEMS Microbiol Ecol 29: 159–169. Tajima, K., Arai, S., Ogata, K., Nagamine, T., Matsui, H., Nakamura, M., et al. (2000) Rumen bacterial community transition during adaptation to high-grain diet. Anaerobe 6: 273–284. Tajima, K., Aminov, R.I., Nagamine, T., Matsui, H., Nakamura, M., and Benno, Y. (2001) Diet-dependent shifts in the bacterial population of the rumen revealed with realtime PCR. Appl Environ Microbiol 67: 2766–2774. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., and Higgins, D.G. (1997) The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25: 4876–4882. Van Dyke, M.I., and McCarthy, A.J. (2002) Molecular biological detection and characterization of Clostridium populations in municipal landfill sites. Appl Environ Microbiol 68: 2049–2053. Westlake, K., Archer, D.B., and Boone, D.R. (1995) Diversity of cellulolytic bacteria in landfill. J Appl Bacteriol 79: 73–78. Wise, M.G., McArthur, J.V., and Shimkets, L.J. (1997) Bacterial diversity of a Carolina bay as determined by 16S rRNA gene analysis: confirmation of novel taxa. Appl Environ Microbiol 63: 1505–1514. Ziemer, C.J., Sharp, R., Stern, M.D., Cotta, M.A., Whitehead, T.R., and Stahl, D.A. (2000) Comparison of microbial populations in model and natural rumens using 16S ribosomal RNA-targeted probes. Environ Microbiol 2: 632–643.
    
    © 2008 The Authors Journal compilation © 2008 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 10, 1310–1319

Add Comment

To add a comment, you need to login or signup (you can signup without leaving this page)