Quantification of Methanosaeta Species in Anaerobic Bioreactors Using Genus- and Species-Specific Hybridization Probes

D. Zheng, L. Raskin

Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA

Received: 23 July 1999; Accepted: 8 December 1999; Online Publication: 5 May 2000

A BSTRACT To evaluate the role of Methanosaeta spp. in a variety of anaerobic environments, small-subunit rRNA targeted oligonucleotide hybridization probes were developed and experimentally characterized. The probes were designed to be genus specific for Methanosaeta and species specific for Methanosaeta concilii and Methanosaeta thermophila. The temperature of dissociation was determined for each probe. Probe specificities were determined using a diverse collection of Archaea and through an evaluation of probe nesting using samples from a variety of anaerobic bioreactors. Cell fixation and hybridization conditions for fluorescence in situ hybridizations were also evaluated. Although permeability of methanogens was variable, M. concilii cells could be permeabilized using a range of paraformaldehyde and ethanol based fixation conditions. Using the newly designed probes together with previously designed probes for methanogens, it was determined that Metha- nosaeta spp. were the dominant aceticlastic methanogens in a variety of anaerobic bioreactors when acetate concentrations were low. Their levels were higher in bioreactors with granular sludge than in those with flocculent sludge. In lab-scale upflow anaerobic sludge blanket reactors, the levels of M. concilii rRNA were as high as 30% of the total rRNA.

Introduction than acetate, such as methanol, methylamines, and H2 and

CO2 [10]. Methanosaeta spp. have higher affinities for ac- Only two genera of aceticlastic methanogens have been de- etate but lower growth rates than Methanosarcina spp. [73]. scribed, i.e., Methanosaeta and Methanosarcina. Acetate is Methanosaeta spp. are present in many environments, such the only known energy source for Methanosaeta spp., as anaerobic digesters [53, 55], anaerobic biofilms [42, 58], whereas Methanosarcina spp. can utilize substrates other sediments [12], paddy soil samples [24, 33], contaminated aquifers [16], and anaerobic granular sludge found in upflow anaerobic sludge blanket (UASB) reactors [31, 32]. The taxonomy of the genus Methanosaeta has been some- Correspondence to: L. Raskin; Fax (217) 333-6968; E-mail: [email protected] what controversial [9, 30, 50, 51, 66]. We herein adopt the Oligonucleotide Hybridization Probes for Methanosaeta spp. 247

Table 1. Sequences available in the Ribosomal Database Project (RDP) [40] for Methanosaeta spp.a

RDP ID Strain Sequenceb Reference

3Ј CGATCCACAGCCGGTGCCACGCT5Ј M. concilii ||||||||||||||||||||||| Mst.conci1 FE 5Ј GCUAGGUGUCGGCCACGGUGCGA3Ј [56] Mst.conci1 Opfikon ...... [17] Mst.conci2 Opfikon ...... -...... [30]

Mst.conci4 UA ...... -...... [30] Mst.conci5 PM ...... -...... [30] Mst.conci3 GP6 ...... -...... [30] M. thermophila Mst.thermo CALS-1 ...... [56] Mst.CALS-1 CALS-1 ...... -...... [30] Mst.thermo2 Z-517 ...... -.....NN.. [30]

Mst.sp.2 PT ...... -...... [30] a The sequences for probe S-F-Msae-0825-a-A-23 [54] and for its target site are given. b A hyphen (-) indicates a deletion; a dot (.) indicates the same nucleotide as in the preceding sequence.

taxonomy proposed by Boone et al. [10], who accepted the A small-subunit (SSU) rRNA targeted oligonucleotide genus name Methanosaeta and recognized two species within probe (S-F-Msae-0825-a-A-23) was previously designed for this genus. One is the mesophilic Methanosaeta concilii, the family Methanosaetaceae [54], which currently contains which has an objective synonym Methanothrix concilii and a only the genus Methanosaeta [10]. Some SSU rRNA se- subjective synonym Methanothrix soehngenii [10]. Currently quences of Methanosaeta spp. have a deletion in the target known M. concilii strains include Opfikon [28, 70], VNBF site of this probe at position 838 (based on Escherichia coli

[21], GP6 [49], FE [66], UA and PM [31], and MTKO [47]. numbering) [30, 54]. As shown in Table 1, there is some A new strain of M. concilii, VeAc9, was isolated recently from uncertainty about this deletion since it is not consistent with anoxic rice paddy soil [24]. The other species within the sequences reported by different researchers even for the genus Methanosaeta is the thermophilic Methanosaeta ther- same strains (Opfikon and CALS-1). A specificity study us- mophila, which has an objective synonym Methanothrix ther- ing membrane hybridizations indicated that the deletion mophila and subjective synonyms Methanothrix thermoace- could be real for strain CALS-1 [54]. In this previous study, tophila and Methanosaeta thermoacetophila [10]. Currently the same amounts of RNA extracted from pure cultures of known M. thermophila strains include CALS-1 [74, 75], PT strains CALS-1 and FE were blotted onto hybridization [32], and Z-517 [45]. membranes, but the hybridization signal from strain CALS-1 A variety of methods have been used for the detection was much lower than the signal from strain FE with probe and quantification of Methanosaeta spp. These include (1) S-F-Msae-0825-a-A-23 [54]. However, it cannot be ruled culture-based methods, such as enrichments and most prob- out that the lower signal was due to partial degradation of able number (MPN) estimates (e.g., [69]); (2) morphology- the target site of strain CALS-1. Because of this problem, we based methods, such as light microscopy and scanning or designed a new genus-specific probe for Methanosaeta as transmission electron microscopy (e.g., [37, 69]); (3) immu- well as species-specific probes for M. concilii and M. ther- nostaining with antibody probes (e.g., [7, 67]); (4) lipid mophila. The newly designed probes were used together with component analysis (e.g., [43, 46]), (5) oligonucleotide previously designed probes for methanogens to characterize probe hybridizations (e.g., [23, 53, 55]), and (6) a combi- biomass samples from a variety of anaerobic bioreactors. nation of two or more of these methods (e.g., [26, 36]). In this study, we focus on ribosomal RNA (rRNA) targeted Materials and Methods oligonucleotide probe hybridizations since they can be used Microorganisms and Nucleic Acid Extraction to identify and quantify microorganisms at various levels of The microorganisms used in this study are listed in Fig. 2. These specificity in a complex microbial community without prior microorganisms were obtained from the Deutsche Sammlung von cultivation [5, 52]. Mikroorganismen und Zellkulturen (DSMZ, Braunschweig, Ger- 248 D. Zheng, L. Raskin many), the American Type Culture Collection (ATCC, Manassas, tridges (OPC; Applied Biosystems) by the University of Illinois VA), the Oregon Collection of Methanogens (OCM, Portland, Biotechnology Center Genetic Engineering Facility. The probes Ј 32 OR), and various collections at the University of Illinois at Urbana- were 5 -end labeled with P using bacteriophage T4 polynucleotide Champaign. Strains were grown as recommended by the culture kinase and [␥-32P] ATP [54]. Oligonucleotide probes for FISH collections and were harvested in the mid-log growth phase. RNA were synthesized by Operon Technologies, Inc. (Alameda, CA) and was isolated from cell pellets using a low-pH hot-phenol extraction Genosys (The Woodlands, TX). method slightly modified from Stahl et al. [64]. Cells were dis- To determine the post-hybridization wash temperature (Tw), rupted using zirconium beads, instead of glass beads, with a Mini- the temperature of dissociation (Td) was determined for each probe beadbeater (Biospec Products, Bartlesville, OK) for 2 min, followed using an elution method [71]. A variety of RNAs extracted from by incubation at 60°C for 10 min, and bead-beating for another 2 microorganisms with zero, one, two, or three mismatches were min. included in these Td studies. The specificities of the probes were evaluated by applying 25 ng RNAs extracted from 24 microorganisms representing different Environmental Samples phylogenetic groups in the archaeal domain (Fig. 2) to nylon membranes (Magna Charge; Micron Separation, Inc., Wesboro, MA). Grab samples were taken from the following anaerobic bioreactors: The membranes were hybridized with 32P labeled oligonucleotide (1) two lab-scale anaerobic digesters treating a mixture of munici- probes [54]. The T values derived from the T study (Table 2) pal solid waste and biosolids [23]; (2) two lab-scale thermophilic w d were used as the final post-hybridization wash temperatures, and anaerobic bioreactors operated at Iowa State University, including membrane images were obtained with a PhosphorImager (Molecu- an anaerobic sequencing batch reactor (ASBR) treating nonfat dry lar Dynamics, Sunnyvale, CA). milk and a temperature phased anaerobic digester (TPAD) fed For quantitative membrane hybridizations, sample RNA and a waste activated sludge collected from a plant treating a mixture of dilution series of pure culture RNA were applied to nylon mem- municipal and paper manufacturing wastewater; (3) a full-scale branes and hybridizations were performed as described previously UASB reactor treating the waste stream of a corn wet milling plant; [52]. The hybridization signal intensities were quantified with the and (4) two lab-scale UASB reactors treating synthetic wastewaters ImageQuant software package (Molecular Dynamics). containing glucose or glucose and propionate [72]. The samples were shipped on ice to the laboratory. For RNA extractions, 2–4 ml of the samples was centrifuged at 14,000 × g for 10 min at 4°C. The Cell Fixation, FISH, and Microscopy supernatant was removed and cell pellets were quickly frozen on dry ice and stored at -80°C until RNA extractions were performed To determine the optimal fixation method for Methanosaeta spp., as described above. The quality of isolated RNA was evaluated five fixation conditions were evaluated. Cells of M. concilii strain using polyacrylamide gel electrophoresis [71]. MTKO were fixed at 4°C in 4% PFA for 5 min, 2 h, and 19 h, and Granular sludge for fluorescence in situ hybridizations (FISH) in 50% ethanol for 2 h and 19 h [15]. Methanosaeta concilii strain was fixed with 4% paraformaldehyde (PFA, in phosphate-buffered GP6 was fixed with 4% PFA for 5 min, 2 h, and 19 h. Nontarget saline [PBS, 130 mM NaCl, 10 mM sodium phosphate, pH 7.2]) at organisms (Methanoculleus thermophilicus and Methanogenium 4°C for 19 h. The granules were washed twice with PBS and stored cariaci) were also fixed at 4°C in 4% PFA for 5 min, 2 h, and 19 h. in 1× storage buffer (10 mM Tris-HCl, pH 7.5, 0.1% Nonidet P-40 Cells of M. cariaci were treated with mutanolysin (Sigma, St. Louis, [NP40]) and 50% ethanol at -20°C. Immediately before embed- MO) before hybridization. Cells were applied to Teflon-coated mi- ding, the granules were dehydrated with 70% ethanol in water at croscope slides (Cel-Line Associates, Inc., Newfield, NJ), air-dried, 4°C for 19 h. The ethanol was replaced by tertiary butanol and and dehydrated through 50, 80, and 96% ethanol for 3 min each. subsequently by Paraplast embedding medium (Sigma, St. Louis, Three µl of mutanolysin (5,000 U ml-1 in 0.1 M phosphate buffer, MO) as described by Harmsen et al. [27]. The embedded granules pH 6.8) was added to each well, and the slide was incubated in a were sectioned at the Histological Laboratory of Diagnostic Medi- moisture chamber for 30 min at 35°C, followed by another dehy- cine (College of Veterinary Medicine, University of Illinois at Ur- dration [18]. bana-Champaign) with a thickness of 5 µm. The ribbons were The permeability of all fixed cells was tested using probe S-D- stretched in 50°C water and transferred to microscope slides (posi- Arch-0915-a-A-20 labeled with Oregon Green (Molecular Probes, tively charged slides coated with silane). The slides were dried at Inc., Eugene, OR). Hybridization and wash conditions were de- 42°C for 19 h and deparaffinated as described previously [27]. scribed previously [6]. After washing, the slides were stained with 4’,6-diamidino-2-phenylindole (DAPI) (6.26 µg ml-1 in 0.1 M Tris- Design and Characterization of Oligonucleotide Probes HCl and 0.9 M NaCl, pH 7.2) for 1 min and rinsed with ice-cold distilled water [15]. The slides were air-dried and mounted in Citi- Oligonucleotide probes were designed using SSU rRNA sequences fluor (Citifluor Ltd., London, United Kingdom). Fluorescent sig- available from the Ribosomal Database Project (RDP) [40]. The nals from hybridized cells were observed using an epifluorescence probes and their target groups are listed in Table 2. The probes microscope (Axioskop; Carl Zeiss, Germany) equipped with filter were synthesized with a DNA synthesizer (Applied Biosystems, Fos- sets 41001, 41002, 31000, and 41008 (Chroma Tech. Corp., Brat- ter City, CA) and purified with oligonucleotide purification car- tleboro, VT). The images were captured using a liquid-cooled Oligonucleotide Hybridization Probes for Methanosaeta spp. 249

Table 2. Newly designed probes for Methanosaeta spp., their target regions, the experimentally determined dissociation temperature (Td), and the final wash temperature (Tw)

a b c d Probe and organisms Probe/target sequence Td (°C) Tw (°C) S-G-Msae-0733-a-A-22 3ЈGTTCCCTGCTTTCGATCCCCGT5Ј 60.1–61.0 N/A |||||||||||||||||||||| Methanosaeta concilii GP6 5ЈCAAGGGACGAAAGCUAGGGGCA3Ј Methanosaeta thermophila CALS-1 ...... Methanosarcina acetivorans C2A UG...... G...... Methanolobus tindarius UG...... G...... Methanosarcina sp.WH1 UG...... G...... Methanosarcina mazeii UG...... G......

S-G-Msae-0781-a-A-22 3ЈTTTGCTATGAGCGATCCACAGC5Ј 48.6–52.0 N/A |||||||||||||||||||||| Methanosaeta concilii GP6 5ЈAAACGAUACUCGCUAGGUGUCG3Ј Methanosaeta thermophila CALS-1 ...... Methanosarcina acetivorans C2A ...... G...... A Methanolobus tindarius ...... G...... A Methanosarcina sp.WH1 ...... G...... U Methanosarcina mazeii ...... G...... A

S-G-Msae-0332-a-A-22e 3ЈTTAGGTCCGGGATGCXCCACGT5Ј 55.0–55.8 59 |||||||||||||||||||||| Methanosaeta thermophila PT 5ЈAAUCCAGGCCCUACGGGGUGCA3Ј Methanosaeta thermophila CALS-1 ...... Methanosaeta thermophila Z-517 ...... Methanosaeta concilii GP6 ...... Methanoculleus thermophilicus ...... C... Methanogenium cariaci .U...... C... Methanosarcina barkeri ...... NN... Methanolobus tindarius ...... NNN.. Methanosarcina sp.WH1 ...... NC...

S-S-M.con-0381-a-A-22 3ЈGTTGTCGCTATTCCCTTGGAGC5Ј 53.3–54.1 55 |||||||||||||||||||||| Methanosaeta concilii GP6 5ЈCAACAGCGAUAAGGGAACCUCG3Ј Methanoculleus thermophilicus ....C.U...... Methanosaeta thermophila CALS-1 . . . . . C ...... G......

S-S-M.the-0396-a-A-22 3ЈCTGGAGCTCACGACCCAATGTT5Ј 54.4–54.9 54 |||||||||||||||||||||| Methanosaeta thermophila PT 5ЈGACCUCGAGUGCUGGGUUACAA3Ј Methanosaeta thermophila CALS-1 ...... Methanosaeta thermophila Z-517 ...... Methanosaeta concilii GP6 A...... CA...... a Probes are named according to the Oligonucleotide Probe Database [1]. b X = 5-nitroindole; a dot (.) indicates the same nucleotide as in the preceding sequence. c Td = experimentally determined temperature of dissociation d Tw = final wash temperature after hybridization. N/A, not applicable. e This probe needs to be used with an unlabeled version of a competitor probe (see text).

charge coupled device (CCD) camera (Photometrics Ltd., Tucson, nontarget cells using hybridization buffers and wash buffers at dif- AZ) and IPLab Spectrum software (Signal Analytics, Vienna, VA), ferent formamide (FA) concentrations [15]. The hybridization and were exported to Adobe Photoshop 3.0 (Adobe, Seattle, WA) buffer contained 0 to 70% FA in a solution of 0.1% sodium dodecyl for printing [15]. sulfate (SDS), 0.9 M NaCl, and 20 mM Tris-HCl, pH 7.2. The wash The optimal stringency for the newly designed probes was de- buffer contained X M NaCl in 0.1% SDS and 20 mM Tris-HCl, pH termined by comparing hybridization signals from both target and 7.2, where X represents the concentration necessary to obtain a 250 D. Zheng, L. Raskin stringency equivalent to the stringency of the hybridization buffer estimated by RDP [40]) than the desired value of 50–55°C, as calculated by an empirical formula [63]. The hybridization and based on Td values of previously designed probes and spe- wash temperatures were 46°C and 48°C, respectively. cies-specific probes (see below). This high Td value would make it difficult to use this probe together with other probes in the same FISH experiment [52]. Therefore, to reduce the Results and Discussion Td value, a universal nucleotide, 5-nitroindole (N5) [35], was Probe Design and Optimization of Wash Temperatures substituted for a cytosine at position 347 (E. coli numbering)

(Table 2). N5 behaves almost indiscriminately toward the Among the 11 described Methanosaeta strains, full-length five natural nucleotides but reduces target-duplex stability,

SSU rRNA sequences are only available for strains FE, Opfi- thus resulting in a lower Td value [35; Oerther and Raskin, kon, and CALS-1 [17, 56]; partial sequences (E. coli num- unpublished data]. Since this probe has only one mismatch bering, 290–492, 701–901, 1195–1380) are available for a with some nontarget sequences, a competitor probe (S-G- Ј number of other strains [30]. To ensure probe specificity, Msae-0322-p-A-22, 5 TGC GCCN5CGTAGGGCCTGGATT efforts were made to design probes within the regions se- 3Ј) that matches the nontarget sequences was designed. Dur- quenced by Kamagata et al. [30]. The specificity of each ing hybridization, an equal amount of an unlabeled version probe was evaluated using the CHECK_PROBE command of the competitor probe was added together with the 32P in the RDP and the Oligonucleotide Probe Database (OPD) labeled probe S-G-Msae-0332-a-A-22. The Td study demon- [1], and the BLAST network service [2]. strated that probe S-G-Msae-0332-a-A-22 could distinguish Initially, two genus-specific probes, S-G-Msae-0733-a-A- target sequences from nontarget sequences when used to-

22 and S-G-Msae-0781-a-A-22, were designed. According to gether with probe S-G-Msae-0332-p-A-22 (Fig. 1c). The Td the SSU rRNA sequences available in RDP, both probes are for the five target RNAs tested ranged from 55.0°C to 55.8°C, specific for target organisms and have at least two mis- with an average of 55.4°C. The RNA of M. cariaci, which has matches with nontarget organisms (Table 2). However, for two mismatches with the probe, had a Td of 51.0°C. The probe S-G-Msae-0733-a-A-22, the experimentally deter- RNA of M. thermophilicus, which has only one mismatch mined Td values of nontarget organisms with three mis- with probe S-G-Msae-0332-a-A-22, had an even lower Td matches were only 1–4°C lower than the Td values of target (49.2°C). This apparent discrepancy is probably due to the organisms (Fig. 1a). In addition, the slopes of the elution fact that the target site of M. thermophilicus perfectly curves were relatively low. Therefore, it was not possible to matches the competitor probe, which further reduces the select a Tw that would result in reasonable hybridization amount of probe S-G-Msae-0332-a-A-22 that binds even at signals with target RNA and negligible signals with nontarget low stringencies. The other three organisms used in this Td RNA. Similarly, for probe S-G-Msae-0781-a-A-22, it was not study all have unknown nucleotide(s) (N) within the probe possible to select a reasonable Tw due to the similar Td values target site. Their Td values were between 51.0°C and 51.6°C. for target and nontarget RNAs (Fig. 1b). It is unclear why Species-specific probes were designed for M. concilii (S- these two probes did not behave as expected. A sequencing S-M.con-0381-a-A-22) and M. thermophila (S-S-M.the- error is not likely, since the sequences were obtained inde- 0396-a-A-22) (Table 2). Probe S-S-M.con-0381-a-A-22 has pendently by three research groups. Secondary or higher- at least two mismatches with nontarget RNA, including M. order structures of the rRNA did not seem to prohibit the thermophilicus and M. thermophila RNA (Table 2). The Td binding of the probes, since the amount of probe bound to values of this probe determined for two strains of M. concilii, the target site during hybridization was sufficiently high to MTKO and GP6, were 53.3°C and 54.1°C, respectively. The be consistent with successful hybridizations. Despite these Td values were 48.1°C for M. thermophilicus and 48.9°C for problems, we report our results to demonstrate the impor- M. thermophila (Fig. 1d). Using a Tw of 55°C should distin- tance of experimental probe characterization before probes guish M. concilii RNA from the nontarget RNAs. Similarly, are used in microbial ecology studies. the Td values for three target strains for probe S-S-M.the- A third probe for the genus Methanosaeta was designed. 0396-a-A-22 were between 54.4°C and 54.9°C (Fig. 1e). The This probe had a high G+C content, and a length of 22 nontarget RNA had at least three mismatches with this probe nucleotides was necessary to make the probe specific (Table (Table 2). The Td values for two nontarget strains, M. concilii

2), which resulted in a higher estimated Td value (60°C, GP6 and MTKO, were 44.9°C and 45.0°C, respectively. lgncetd yrdzto rbsfor Probes Hybridization Oligonucleotide Methanosaeta p.251 spp.

Fig. 1. Temperature of dissociation (Td ) studies for target and nontarget microorganisms. Numbers in parentheses indicate the numbers of mismatches between probe and target sequences. The SSU rRNA sequence of strain MTKO is not available (n/a). All the elution curves were obtained by averaging the results of duplicate experiments. The differences in Td values obtained with duplicate experiments were always lower than 1°C, and in most cases, lower than 0.5°C. 252 D. Zheng, L. Raskin

Fig. 2. Probe specificity study. Images were obtained with a PhosphorImager and were imported to Adobe Photoshop for printing. (a) Organisms; (b) template; (c) probe S-*-Univ- 1390-a-A-18; (d) probe S-S-M.con-0381-a-A-22; (e) probe S-S-M.the-0396-a-A-22; (f) probe S-G-Msae-0332-a-A-22.

Therefore, M. thermophila can be differentiated from non- signals (Figs. 2d–2f). Application of the genus probe S-G- Msae-0332-a-A-22 resulted in weak signals for some nontarget RNAs when a Tw of 54°C is used. target species when the post-hybridization wash was con-

ducted at 56°C (data not shown). Consequently, the Tw was Probe Specificity Studies increased to 59°C. As shown in Fig. 2f, signals from nontarget RNA were eliminated. Probe specificities were initially assessed by a traditional Probe specificities were further assessed through an specificity study using RNA extracted from 24 pure cultures evaluation of probe “nesting” [52]. Theoretically, the sum of of target and nontarget methanogens. Hybridization with the hybridization results obtained at one taxonomic level the universal probe S-*-Univ-1390-a-A-18 [71] demon- (e.g., species) should equal the hybridization result of the strated that similar amounts of rRNA were applied to the taxon at a higher level (e.g., genus). Meeting this nesting membranes (Fig. 2c). Hybridization results with the newly requirement for a variety of environments shows that the designed probes indicated that only the rRNA extracted probes are likely specific. On the other hand, if the hybrid- from target microorganisms resulted in strong hybridization ization result obtained for the higher taxonomic level is Oligonucleotide Hybridization Probes for Methanosaeta spp. 253

Table 3. Previously developed probes used in this study is usually used as a cutoff to define species, it is likely that clone WCHD3-03 represents a new species in the genus Tw Probe Target group (°C)a Reference Methanosaeta. S-*-Univ-1390-a-A-18 Virtually all 44 71 The specificities of the three new probes were further organisms evaluated by checking probe nesting requirements within the S-D-Arch-0915-a-A-20 Archaea 56 63 order Methanosarcinales. This order contains two known M-O-Mmic-1200-a-A-21 Methanomicrobiales 53 54 S-O-Msar-0860-a-A-21 Methanosarcinales 60 54 families, Methanosarcinaceae and Methanosaetaceae [10]. S-F-Mbac-0310-a-A-22 Methanobacteriaceae 57 54 Within the Methanosarcinaceae, only the genus Methanosar- S-F-Mcoc-1109-a-A-20 Methanococcaceae 55 54 cina is expected to be present in anaerobic bioreactors, since S-G-Msar-0821-a-A-24 Methanosarcina 60 54 the other genera are halophilic. Methanosaeta is the only a Tw = final wash temperature after hybridization genus in the family Methanosaetaceae. Therefore, the sum of the abundances of the genera Methanosaeta and Methano- sarcina in most anaerobic bioreactors should equal the level larger than expected, a novel population may be present of Methanosarcinales. As shown in Fig. 3b and Fig. 4b, the [34]. In order to evaluate probe nesting at different taxo- sum of the levels of genera Methanosaeta and Metanosarcina nomic levels, previously designed probes (Table 3) together (Sum-2) was much higher than the level of Methanosarci- with the probes designed in this study were used to charac- nales (detected by the order-specific probe S-O-Msar-0860- terize biomass samples from a variety of lab-scale anaerobic a-A-21) for most of the samples. Unless probe S-O-Msar- bioreactors. 0860-a-A-21 does not target all Methanosarcinales, it is likely As discussed above, the genus Methanosaeta comprises that probe S-G-Msae-0332-a-A-22 hybridized nonspecifi- only two species, M. concilii and M. thermophila. Thus, the cally with some RNAs. Figures 3b and 4b also compare the sum of the levels of M. concilii and M. thermophila should level of Methanosarcinales to the sum of M. concilii, M. ther- equal the level of the genus Methanosaeta. This nesting re- mophila, and Methanosarcina spp. (Sum-3). Except for quirement was met relatively well for most of the lab-scale samples U1, U2, and U3, the sums (Sum-3) match well with UASB samples (Fig. 4a). All other samples exhibited much the Methanosarcinales population levels (detected by probe higher levels of the genus Methanosaeta (detected by probe S-O-Msar-0860-a-A-21). This further suggests that probe S-G-Msae-0332-a-A-22) than the sum of M. concilii and M. S-G-Msae-0332-a-A-22 is not specific, while probes S-S- thermophila (Sum-1) (detected by probes S-S-M.con-0381- M.con-0381-a-A-22 and S-S-M.the-0396-a-A-22 are spe- a-A-22 and S-S-M.the-0396-a-A-22, respectively) (Fig. 3a cific. In addition, it is unlikely that a potentially new Metha- and Fig. 4a). nosaeta sp., represented by the sequence of clone WCHD3- This observation may be explained if the genus probe 03 [16], is present in significant levels in these samples. For hybridizes to nontarget RNAs that are not included in da- the full-scale UASB reactor (samples U1, U2, and U3), the tabases and in the specificity study. Alternatively, unknown total Methanosarcinales levels were much higher than the strains of M. concilii and M. thermophila or new Methano- sum of M. concilii, M. thermophila, and Methanosarcina spp. saeta spp. may exist that are targeted by probe S-G-Msae- This may indicate that other families within the order 0332-a-A-22, but are not targeted by the two species probes. Methanosarcinales, which are halophilic and utilize methyl-

Recently, two new Methanosaeta sequences have been pub- ated C1 compounds, are important in this reactor (see be- lished: the sequence of M. concilii strain VeAc9 [24] and a low). cloned sequence obtained from a contaminated aquifer The specificity of probe S-O-Msar-0860-a-A-21 can be (clone WCHD3-03) [16]. Probe S-G-Msae-0332-a-A-22 tar- evaluated by checking probe nesting in the archaeal domain. gets both of these sequences. Probe S-S-M.con-0381-a-A-22 In anaerobic bioreactors, it is expected that most or all Ar- perfectly matches the SSU rRNA of strain VeAc9 but has one chaea are methanogens [25, 59]. Of the five orders of metha- mismatch with clone WCHD3-03. Clone WCHD3-03 exhib- nogens, Methanopyrales are extremely thermophilic and are its 97% similarity to the SSU rRNA of M. concilii strain not likely to be present in anaerobic bioreactors. For the Opfikon [16]. Since organisms with SSU rRNA similarities other four orders, probes are available that target most less than 97.5% never have DNA reassociation values higher methanogens within these orders except for some thermo- than 70% [62] and since a DNA reassociation value of 70% philes. As shown in Figs. 3c and 4c, for all the samples 254 D. Zheng, L. Raskin

Fig. 3. Methanogenic composition in bioreactors with flocculent sludge: (a) M. concilii, M. thermophila and Methanosaeta genus; (b) aceticlastic methanogens; (c) methanogens. M1, M17, and M38 indicate samples taken at days 1, 17, and 38 from a mesophilic digester; T1, T7, and T38 represent samples taken from a thermophilic digester on days 1, 7, and 38; I1 was taken from a thermophilic ASBR; and I2 was taken from the thermophilic phase of a TPAD. Sum-1 = S-S-M.the-0396-a-A-22 + S-S-M.con-0381-a-A-22; Sum-2 = S-G-Msae-0332-a-A-22 + S-G- Msar-0821-a-A-24; Sum-3 = S-S- M.the-0396-a-A-22 + S-S-M.con- 0381-a-A-22 + S-G-Msar-0821-a- A-24; Sum-4 = S-O-Msar-0860-a- A-21 + S-F-Mcoc-1109-a-A-20 + S-F-Mbac-0310-a-A-20 + S-O- Mmic-1200-a-A-21. The error bars indicate standard deviations obtained by triplicate application of the same RNA extract to hybridization membranes.

analyzed in this study, the sum of the levels of methanogens Optimization of Conditions for FISH detected by these four probes (Sum-4) matches the level of Archaea (quantified by probe S-D-Arch-0915-a-A-20), sug- The first step of FISH is fixation, which opens the cell to gesting that these probes are specific, including probe S-O- allow penetration of probes. Since poor fixation can be the Msar-0860-a-A-21. reason behind unsuccessful FISH, it is recommended to Oligonucleotide Hybridization Probes for Methanosaeta spp. 255

Fig. 4. Methanogenic composition in UASB reactors with granular sludge: (a) M. concilii, M. thermophila, and Methano- saeta genus; (b) aceticlastic methanogens; (c) methanogens. U1, U2, and U3 represent samples taken from three different heights (bottom, middle, and top) from a full-scale UASB reactor treating wastewater from an ethanol producing plant; G203, G224, and G286 indicate samples obtained from a lab- scale UASB reactor treating glucose on days 203, 224, and 286; GP203, GP224, and GP286 represent samples taken on the same days from a lab-scale UASB reactor treating glucose and propionate. Sum-1, Sum-2, Sum-3, and Sum-4 are defined as in the legend for Fig. 3. The error bars indicate standard deviations obtained by triplicate application of the same RNA extract to hybridization membranes. evaluate the permeability of the cell wall using previously extent that probes cannot penetrate [41]. However, excessive characterized and well-labeled universal or domain-specific cross-linking can be avoided by shortening the PFA fixation probes [5]. time [15]. Some gram-positive cells need treatment with Fixatives can be precipitants (e.g., ethanol) or cross- lytic enzymes (mutanolysin or lysozyme), hydrophobic sol- linking reagents (e.g., PFA). In general, 50% ethanol offers vents (toluene or diethylether), or acids for proper fixation good results for fixation of gram-positive cells, whereas 4% [3, 18]. Burggraf et al. [11] demonstrated that 4% PFA was PFA works well for gram-negative cells [5]. Gram-negative a suitable fixative for most Archaea tested in their study. cells can also be fixed with 50% ethanol, but the fixed cells Since optimal fixation conditions appear to be species de- were found to have low stability [41]. Paraformaldehyde is pendent even within a group of closely related species [15, not suitable for gram-positive cells since it can cause cross- 38], it is difficult to fix all cells in a complex community with linkage of proteins within the gram-positive cell wall to the a single treatment. Therefore, different fixation methods 256 D. Zheng, L. Raskin

an S-layer and a pseudomurein layer, a sheath and an S-layer layer, or a pseudomurein layer only [61]. In addition, proteins in the S-layer can be either glycosylated or nonglycosylated. The cell wall of Methanosarcina mazeii contains a methanochondroitin layer and an S-layer. Sørensen et al. [60] found that fixation in 4% PFA for 16 h disrupted the cell morphology of M. mazeii and recommended fixation of the M. mazeii cells in saline–formaldehyde (1.6% formalde- Fig. 5. Effects of fixation treatments on Methanosaeta concilii hyde, 0.85% NaCl). Fixation with 4% PFA for 3 h at room strain MTKO. FISH was conducted at 0% FA using Oregon Green temperature works well with Methanospirillum hungatei, labeled probe S-D-Arch-0915-a-A-20. Bar = 10 µm. whose cell wall contains a sheath and a glycosylated S-layer [11]. Likewise, we observed that M. concilii, which has the same type of cell wall as M. hungatei, can be fixed properly should be used according to the cell wall properties of the by 4% PFA for 5 min, 2 h, and 19 h at 4°C. Fixation with 4% populations that are to be detected [41]. PFAfor3hatroom temperature was found to be too mild To determine optimal fixation conditions for Methano- to open cell walls of Methanopyrus kandleri and Methano- saeta spp., an archaeal-specific probe S-D-Arch-0915-a-A-20 thermus fervidus, which have a regularly structured S-layer (labeled with Oregon Green) was used with M. concilii and a pseudomurein layer, and Methanobacterium thermo- MTKO and GP6 cells fixed using five and three fixation autotrophicum with only a pseudomurein layer [11]. How- conditions, respectively. All fixation conditions resulted in ever, we observed that fixation with 4% PFA at 4°C for 5 similar hybridization intensities, except that strain MTKO min, 2 h, and 19 h was suitable for Methanobacterium bry- cells fixed with ethanol for 19 h exhibited a slightly lower antii and Methanobrevibacter arboriphilicus, which also have signal (Fig. 5; data for strain GP6 are not shown). The pseudomurein cell walls (Zheng and Raskin, unpublished Methanosaeta probes were developed to study biomass char- data). For methanogens that have a regularly structured acteristics in anaerobic bioreactors, some of which contain nonglycosylated S-layer, including Methanococcus thermo- granular sludge. Because of possible diffusion limitations, lithotrophicus and M. vannielii, Burggraf et al. [11] found long fixation times are preferred to allow the fixative to that fixation with 4% PFA for 3 h at room temperature can penetrate into the granules. Therefore, cells were fixed for 19 open the cell walls. In contrast, in this study, we found that h with 4% PFA in subsequent FISH experiments. M. cariaci, which has the same type of cell wall, was resistant Fixation conditions were also evaluated for the nontarget to 4% PFA fixation at 4°C for 5 min, 2 h, and 19 h. Since the species Methanocelleus thermophilicus and Methanogenium cell walls of Methanogenium spp. can be lysed easily by an- cariaci using the archaeal-specific probe. For M. thermophili- ionic detergents [39], it might be possible to find an efficient cus, fixing for 19 h with 4% PFA resulted in better signals fixation condition for M. cariaci using anionic detergents at than the other two conditions (5 min or 2 h with 4% PFA) low concentrations. On the other hand, 4% PFA fixation at (data not shown). Fixation of M. cariaci cells with 4% PFA 4°C for 5 min, 2 h, and 19 h was found to work well for M. for 5 min did not result in a noticeable hybridization signal. thermophilicus, which has a regularly structured glycosylated When the fixation time was extended to2hor19h,probe S-layer. These results again demonstrate that it is difficult to S-D-Arch-0915-a-A-20 appeared to have bound to the sur- find one fixation condition that can permeabilize all cells in face of the cells resulting in O-ring shaped images (data not a microbial community. shown). Treatment with mutanolysin did not open the cells Autofluorescence is often a problem for FISH. For in- fixed for any of the three fixation times, but the treatment stance, Harmsen et al. [27] observed autofluorescence at 520 prevented the probe from binding to the cell wall after fixa- and 580 nm from granule sections and suspected that the tion for2hor19h(data not shown). autofluorescence was caused by the presence of inorganic Cell wall structures among methanogens vary signifi- precipitates in the granules. Sørensen et al. [60] found that cantly and can be divided into five groups (with the excep- autofluorescence exhibited by Methanosarcina mazeii cells tion of the cell wall of Methanoplasma elizabethii), including was enhanced when the hybridization buffer contained a regularly structured paracrystalline proteinaceous surface EDTA. They also observed autofluorescence from homog- array (S-layer), a methanochondoritin layer and an S-layer, enized granules and suggested the use of specific narrow- Oligonucleotide Hybridization Probes for Methanosaeta spp. 257

Fig. 6. Autofluoroscence by M. concilii GP6 (a) and M. barkeri (b) with filter sets 41002, 41001, and 41008. Bar = 10 µm. band filters to avoid it [60]. In this study, we evaluated autofluorescence associated with M. concilii GP6, M. barkeri, M. cariaci, M. thermophilicus, M. bryantii, and M. arboriphi- lus with three filter sets (41001, 41002, and 41008). Metha- nogens were fixed with 4% PFA at 4°C for 19 h and were processed as for FISH but without addition of probe. The cells were stained with DAPI before observation. M. concilii GP6 and M. barkeri exhibited weak autofluorescence when filter 41002 (exciter 535/50 and emitter 610/75) was used, and M. barkeri exhibited strong autofluorescence with filter sets 41001 (exciter 480/40 and emitter 535/50) (Fig. 6). The other four methanogens did not exhibit autofluorescence with any of the three filter sets used (data not shown). The observed autofluorescence can interfere with FISH when a Fig. 7. Optimization of FISH conditions for probe S-S-M.con- sample contains high levels of organisms exhibiting auto- 0381-a-A-22 (a) and S-G-Msae-0332-a-A-22 (b). The FA concen- fluorescence (e.g., M. barkeri) and the probe is labeled with trations were 0, 5, 10, and 15%. Bar = 10 µm. a dye that is excited at a similar wavelength. Based on the limited data available, filter 41008 (exciter 620/60 and emit- signals were still observed for M. thermophilicus. When the ter 700/75) appears to be a good choice for avoiding auto- FA concentration was increased to 15%, the signals from M. fluorescence. concilii cells were reduced and those from M. thermophilicus Optimal FISH conditions were estimated with eight FA were not detectable. Therefore, 0% FA appears to be the concentrations ranging from 0 to 70% for probes S-G-Msae- optimal concentration for hybridization and wash buffers 0332-a-A-22 and S-S-M.con-0381-a-A-22. For both probes, for probe S-S-M.con-0381-a-A-22. We have observed an in- the optimal stringencies were between 0 and 10% FA (data crease in hybridization signals when increasing the FA con- not shown). A second stringency study was conducted in a centration from 0% to 5% or from 0% to 10% FA for a narrower range, i.e., 0% to 15% FA. The hybridization sig- number of probes (unpublished data). It is possible that the nals from target and nontarget organisms are shown in Fig. presence of low levels of FA (versus the absence of FA) 7. For probe S-S-M.con-0381-a-A-22 (Fig. 7a), 0% FA did reduces the secondary structure of the target site and makes not result in a signal for the nontarget organism (M. ther- it more accessible to probes. mophilicus), while target cells (M. concilii GP6) were suc- For probe S-G-Msae-0332-a-A-22 (Fig. 7b), 15% FA was cessfully visualized. The signal for M. concilii GP6 increased found to be the optimal stringency, since signals were de- slightly when the FA concentration was increased from 0% tected with the nontarget organism, M. thermophilicus, at to 5%. However, at 5% FA, weak signals were also observed lower FA concentrations. Similarly to probe S-S-M.con-381- for the nontarget organism. At 10% FA, hybridization sig- a-A-22, hybridizations with 5% and 10% FA resulted in nals from M. concilii were similar to those at 0% FA, but higher signals than with 0% FA. 258 D. Zheng, L. Raskin

Environmental Samples spp. in thermophilic environments, samples were obtained from two thermophilic bioreactors. Sample I1 was collected Samples obtained from a number of anaerobic bioreactors from a thermophilic (55°C) ASBR treating synthetic waste- containing either flocculent or granular sludge were ana- water that contained nonfat dry milk (12 g chemical oxygen lyzed with probes designed in this study to evaluate the demand [COD] L-1). The pH in the reactor was between 7.1 importance of Methanosaeta spp. in these engineered sys- and 7.3 and the COD removal was approximately 80%. tems. Two lab-scale anaerobic digesters, treating a mixture Sample I2 was collected from the thermophilic phase (55°C) of municipal solid waste and biosolids operated at meso- of a TPAD fed waste activated sludge from a plant treating a philic (35°C) and thermophilic (55°C) conditions, were mixture of municipal and paper manufacturer wastewaters. sampled during startup [23]. The levels of Methanosaeta in The pH in this reactor ranged from 5.5 to 6.5. The total these digesters were previously quantified using probe S-F- concentration of volatile fatty acids (VFA) in the effluent was Msae-0825-a-A-23 [23], which targets a site with a potential 754 mg L-1 as acetate. The levels of M. thermophila in deletion (see above). Therefore, the probes designed in this samples I1 and I2 were below the detection limit (0.2%), study were used to re-evaluate these samples. The M. concilii while M. concilii levels were about 0.5%. In both samples, level was about 4.5% in the mesophilic digester on day 1 Methanosarcina spp. were more abundant than M. thermo- (Fig. 3a). The level decreased to less than 1% on day 17 when phila and M. concilii (Fig. 3b), presumably due to high ac- the acetate concentration was more than 2,000 mg L-1.On etate concentrations in the reactors. Since the ASBR effluent day 38, when digester performance had become stable and contained approximately 2.4 g COD L-1, the acetate concen- acetate levels had decreased, M. concilii had increased again tration in this reactor was likely high. Similarly, the total to 2%. Methanosaeta thermophila was detected in the meso- VFA concentration of 754 mg L-1 as acetate in the thermophilic digester at levels higher than the detection limit philic phase of the TPAD corresponded to a high acetate (0.2%) (Fig. 3a). Methanosaeta thermophila levels in the concentration. The low overall methanogen levels in sample thermophilic digester were higher (1.1–1.7%) than in the I2 (Fig. 3c) were probably due to the unfavorable pH in this other thermophilic and mesophilic reactors evaluated (see reactor. below). The mesophilic M. concilii was present in the ther- Granular sludge samples were taken from a full-scale me- mophilic digester at a level of about 1% (detection limit was sophilic (38°C) UASB reactor treating wastewater from a 0.1%) (Fig. 3a). corn wet milling plant. The levels of M. concilii in this UASB For most samples (except M1), the sum of M. concilii and reactor were between 2.2 and 2.8% (Fig. 4a), which is higher M. thermophila levels were found to be higher than those than the levels of M. concilii in flocculent sludge, but much detected by probe S-F-Msae-0825-a-A-23 [23]. The deletion lower than those in granules from lab-scale UASB reactors in the target site of probe S-F-Msae-0825-a-A-23 for some (see below). Methanosaeta thermophila and Methanosarcina Methanosaeta spp., as discussed earlier, might be responsible levels were below 1% (Fig. 4a and 4b). Since Methanosarci- for this underestimation. nales were relatively abundant (7.4–27.9%), genera other Regardless of the slightly higher levels of Methanosaeta than Methanosaeta and Methanosarcina within the order of obtained in this study, results on methanogenic population Methanosarcinales (Methanolobus, Methanococcoides, dynamics obtained here were generally similar to those in Methanohalophilus, Methanohalobium, or Methanosalsus) the previous study [23]. Methanosarcina spp. were the pre- might be present in this reactor. Currently known strains of dominant aceticlastic methanogens in both reactors during these genera are halophilic and use methylated C1 com- periods of high acetate concentration and acetate mover. pounds as their substrates. It is possible that the waste Their levels increased from approximately 0.5% on day 1 to stream treated in this reactor has an elevated salt concentra- 17% on day 17 in the mesophilic digester and to 12% in the tion due to the use of sodium hydroxide as a cleaning agent. thermophilic digester on day 7 (Fig. 3b), when the systems In addition, some corn protein may end up in the waste started to remove acetate. This agrees with the fact that stream and fermentation of proteins can produce methylat-

Methanosarcina spp. are more competitive at higher acetate ed C1 compounds to support the growth of these methano- concentration [73]. On day 38, when the digesters were gens. stable and acetate levels had decreased, their levels had Granular sludge samples were also taken from two lab- dropped to 8.4% in the mesophilic digester and 5.4% in the scale mesophilic UASB reactors treating synthetic wastewa- thermophilic digester (Fig. 3b). ters containing glucose as the only energy source (Reactor G) To further determine the abundance of Methanosaeta or a mixture of glucose and propionate (Reactor GP) [72]. Oligonucleotide Hybridization Probes for Methanosaeta spp. 259

The strength of the wastewaters was 4 g COD L−1 for both reactors, and the loading rate was 6 g COD L-1d-1. Regardless of the composition of the feed, methanogen population levels were similar in the two reactors (Fig. 4). The levels of M. concilii in both reactors were 10 to 20 times higher than in the other samples analyzed in this study (Fig. 3a and 4a). Their levels were equivalent to those of Methanosarcinales, indicating that M. concilii was the predominant species in this order. The dominance of M. concilii in granules agrees with observations from other researchers [20, 26, 27, 29, 36]. Their presence in granules is believed to be critical for good performance of UASB reactors [68], and it has been hypoth- esized that they play an important role in the granulation process [37, 68]. The levels of M. thermophila were lower than the detection limit (0.2%) in all samples. Granules obtained from the two lab-scale UASB reactors were analyzed with FISH using probe S-S-M.con-0381-a-A- 22, together with probes S-D-Bact-0338-a-A-18 (Bacteria) and S-D-Arch-0915-a-A-20 (Archaea). Approximately 20 granules from each reactor were observed, and representa- tive micrographs are shown in Fig. 8. The granules from Reactor G consisted of two layers. The outer layer was thin, and contained mostly bacterial cells mixed with small amounts of archaeal cells (Fig. 8a). The inner core accounted for a large portion of the total granule volume and contained almost only Archaea. Methanosaeta concilii was well distrib- uted in this portion. The layered structure agrees with the model proposed by MacLeod et al. [37]. The substrate (glu- Fig. 8. Micrograph obtained after FISH with sliced granular cose) is degraded at the surface of the granule by fermenters sludge from Reactor G (a) and Reactor GP (b) on Day 224. Three (Bacteria) before it can diffuse into the granule. The pro- probes were used: S-S-M.con-0381-a-A-22 labeled wiht Cy5 (blue), duced hydrogen and acetate then diffuse into the granule S-D-Arch-0915-a-A-20 labeled with Oregon Green (green), and S-D-Bact-0338-a-A-18 labeled with Cy3 (red). Bar = 50 µm. where they are converted to methane by methanogens (Ar- chaea). The granules from Reactor GP also exhibited a layered structure (Fig. 8b). Methanosaeta concilii cells were found aerobic bioreactors. A detailed evaluation of probe nesting at dispersed throughout the granules but were more abundant several taxonomic levels suggests that the species-specific in the cores. They were also observed in close association probes developed in this study (S-S-M.con-0381-a-A-22 and with Bacteria at the surface of the granules, consistent with S-S-M.the-0396-a-A-22) as well as probes that were designed previous reports of observations of Methanosaeta at the sur- previously [4, 54] are specific. On the other hand, probe face of propionate-fed granules [22]. Furthermore, a sub- S-G-Msae-0332-a-A-22 may not be as specific as expected surface layer of mixed archaeal and bacterial cells was ob- based on the analysis of available sequences. Furthermore, served, which likely consisted of syntrophic consortia re- we demonstrated, using a number of bioreactor samples, sponsible for the degradation of propionate (40% of the that Methanosaeta spp. were more abundant in granular influent COD was propionate). sludge than in flocculent sludge, and that Methanosarcina spp. outcompeted Methanosaeta spp. when acetate concentrations were high (>200 mg L-1). We also detected the ther- Conclusions mophilic M. thermophila in mesophilic bioreactors and Oligonucleotide probes were developed and characterized found the mesophilic M. concilii in thermophilic bioreactors. for evaluating the significance of Methanosaeta spp. in an- Several studies have shown that thermophilic methanogens 260 D. Zheng, L. Raskin exist in mesophilic environments [13, 19, 23, 44, 48, 65], but Ecology, Physiology, Biochemistry and Genetics. Chapman & few studies have demonstrated the existence of mesophilic Hall, New York, pp 35–80 methanogens in thermophilic environments [8, 14, 57]. Our 11. Burggraf S, Mayer T, Amann R, Schandhauser S, Woese CR, Stetter KO (1994) Identifying members of the domain Ar- observations indicate that methanogens can survive and chaea with rRNA-targeted oligonucleotide probes. Appl Envi- maintain low population levels in environments that are not ron Microbiol 60:3112–3119 optimal for growth. 12. Cairo J, Macario AJL, Bardulet M, Conway de Macario E, Paris JM (1991) Psychrophilic ecosystems of interest for wastewater treatment: Microbiologic and immunologic elucidation of Acknowledgments their methanogenic flora. Syst Appl Microbiol 14:85–92 13. Chen M (1983) Adaptation of mesophilic anaerobic sewage fermentor populations to thermophilic temperatures. Appl We thank Biswarup Mukhopadhyay for providing cultures Environ Microbiol 45:1271–1276 of Methanosarcina barkeri and Methanobacterium bryantii, as 14. Davydova CIA, Kuznetsova VG, Mityshina LL, Belyaev SS well as for his advice on methanogen culturing. We also (1992) Methanogenic rod-shaped bacteria from the oil fields thank Shihwu Sung for providing biomass samples and re- of Tataria and western Siberia. Mikrobiologiya 61:299–305 lated reactor operating data. 15. de los Reyes FL, Ritter W, Raskin L (1997) Group-specific small-subunit rRNA hybridization probes to characterize fila- mentous foaming in activated sludge system. Appl Environ Microbiol 63:1107–1117 References 16. Dojka MA, Hugenholtz P, Haack SK, Pace NR (1998) Micro- bial diversity in hydrocarbon- and chlorinated-solvent con- 1. Alm EW, Oerther DB, Larsen N, Stahl DA, Raskin L (1996) taminated aquifer undergoing intrinsic bioremediation. Appl The oligonucleotide probe database. Appl Environ Microbiol Environ Microbiol 64:3869–3877 62:3557–3559 17. Eggen R, Harmsen H, Geerling A, de Vos WM (1989) Nucleo- 2. Altschul SF, Gish W, Miller W, Myer EW, Lipman DJ (1990) tide sequence of a 16S rRNA encoding gene from the Archae- Basic local alignment search tool. J Mol Biol 215:403–410 bacterium Methanothrix soehngenii. Nucleic Acids Res 17:9469 3. Amann RI (1995) Fluorescently labeled, rRNA-targeted oligo- 18. Erhart R, Bradford D, Seviour RJ, Amann R, Blackall LL nucleotide probes in the study of microbial ecology. Mol Ecol (1997) Development and use of fluorescent in situ hybridiza- 4:543–554 tion probes for the detection and identification of “Microthrix 4. Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R, parvicella” in activated sludge. Syst Appl Microbiol 20:310–318 Stahl DA (1990) Combination of 16S rRNA-targeted oligo- 19. Fang HHP, Lau IWC (1996) Startup of thermophilic (55°C) nucleotide probes with flow cytometry for analyzing mixed UASB reactors using different mesophilic seed sludges. Water microbial populations. Appl Environ Microbiol 56:1919–1925 Sci Technol 34:4455–452 5. Amann RI, Ludwig W, Schleifer K-H (1995) Phylogenetic 20. Fang HHP, Li YY, Chui HK (1995) Performance and sludge identification and in situ detection of individual microbial characteristics of UASB process treating propionate-rich cells without cultivation. Microbiol Rev 59:143–169 wastewater. Water Res 29:895–898 6. Amann RI, Zarda B, Stahl DA, Scheifer KH (1992) Identifi- 21. Fathepure BZ (1983) Isolation and characterization of an ace- cation of individual prokaryotic cells by using enzyme-labeled, ticlastic methanogen from a biogas digester. FEMS Microbiol rRNA-targeted oligonucleotide probes. Appl Environ Micro- Lett 19:151–156 biol 58:3007–11 22. Fukuzaki S, Nishio N, Sakurai H, Nagai S (1991) Character- 7. Archer DB (1984) Detection and quantitation of methanogens istics of methanogenic granules grown on propionate in an by enzyme-linked immunosorbent assay. Appl Environ Mi- upflow anaerobic sludge blanket (UASB) reactor. J Ferment crobiol 48:797–801 Bioeng 71:50–57 8. Belyaev SS, Rozanova EP, Borzenkov IA, Charakhchyan IA, 23. Griffin ME, McMahon KD, Mackie RI, Raskin L (1998) Miller YM, Sokolov MY, Ivanov MV (1990) Specificity of the Methanogenic population dynamics during startup of anaero- microbiological processes in a water-flooded oil field located bic digesters treating municipal solid waste and biosolids. Bio- in the Middle Ob (Russian SFSR, USSR) region. Microbi- technol Bioeng 57:342–355 ologiya 59:1075–1082 24. Grosskopf R, Janssen PH, Liesack W (1998) Diversity and 9. Boone DR (1991) Strain GP6 is proposed as the neotype strain structure of the methanogenic community in anoxic rice of Methanothrix soehngenii VP pro synon. Methanothrix con- paddy soil microcosms as examined by cultivation and direct cilii VP and Methanosaeta concilii VP: Request for an opinion. 16S rRNA gene sequence retrieval. Appl Environ Microbiol Int J Syst Bacteriol 41:588–589 64:960–969 10. Boone DR, Whitman WB, Rouvie`re P (1993) Diversity and 25. Godon JJ, Zumstein E, Dabert P, Habouzit F, Moletta R taxonomy of methanogens. In: Ferry JG (ed) Methanogenesis: (1997) Molecular microbial diversity of an anaerobic digestor Oligonucleotide Hybridization Probes for Methanosaeta spp. 261

as determined by small-subunit rDNA sequence analysis. Appl 40. Maidak BL, Olsen GJ, Larsen N, Overbeek R, McCaughey MJ, Environ Microbiol 63:2802–2813 Woese CR (1997) The RDP (Ribosomal Database Project). 26. Grotenhuis JT, Smit M, Plugge CM, Xu Y-S, van Lammeren Nucleic Acids Res 25:109–111 AA, Stams AJ, Zehnder AJ (1991) Bacteriological composition 41. Manz W, Wagner M, Amann R, Schleifer K-H (1994) In situ and structure of granular sludge adapted to different sub- characterization of the microbial consortia active in two strates. Appl Environ Microbiol 57:1942–1949 wastewater treatment plants. Water Res 28:1715–1723 27. Harmsen HJM, Kengen HMP, Akkermans ADL, Stams AJM, 42. Meraz M, Gonzalez-Barrera A, Alvarez-Ramirez J, Monroy O de Vos WM (1996) Detection and localization of syntrophic (1996) Dynamics of the immobilization of anaerobic meso- propionate-oxidization using 16S rRNA-based oligonucleo- philic bacteria on a plastic support. Appl Biochem Biotechnol tide probes. Appl Environ Microbiol 62:1656–1663 57–58:525–534 28. Huser BA, Wuhrmann K, Zehnder AJB (1982) Methanothrix 43. Nishihara M, Akagawa-Matsushita M, Togo Y, Koga Y (1995) soehngenii gen. nov. sp. nov., a new acetotrophic non- Inference of methanogenic bacteria in wastewater digester hydrogen-oxidizing methane bacterium. Arch Microbiol sludges by lipid component analysis. J Ferment Bioeng 132:1–9 79:400–402 29. Kalyuzhnyi SV, Sklyar BI, Davlyatshina MA, Parshina SN 44. Nozhevnikova A, Holliger C, Ammann A, Zehnder AJB (1997) (1996) Organic removal and microbiological features of Methanogenesis in sediments from deep lakes at different UASB-reactor under various organic loading rates. Biore- temperatures (2–70 degrees C). Water Sci Technol 36:57–64 source Technol 55:47–54 45. Nozhevnikova AN, Chudina VI (1985) Morphology of the 30. Kamagata Y, Kawasaki H, Oyaizu H, Nakamura K, Mikami E, thermophilic acetate methane-producing bacterium Methano- Endo G, Koga Y, Yamasato K (1992) Characterization of three thrix thermoacetophila sp. nov. Mikrobiologiya 53:618–624 thermophilic strains of Methanothrix (“Methanosaeta”) ther- 46. Ohtsubo S, Kanno M, Miyahara G, Kohno S, Koga Y, Miura mophila sp nov and rejection of Methanothrix (“Methano- I (1993) A sensitive method for quantification of aceticlastic saeta”) thermoacetophila. Int J Syst Bacteriol 42:463–468 methanogens and estimation of total methanogenic cells in 31. Kamagata Y, Mikami E (1990) Some characteristics of two natural environments based on an analysis of ether-linked morphotypes of Methanothrix soehngenii from mesophilic an- glycerolipids. FEMS Microbiol Ecol 12:39–50 aerobic digesters. J Ferment Bioeng 70:272–274 47. Ohtsubo S, Miyahara H, Demizu K, Kohno S, Miura I (1991) 32. Kamagata Y, Mikami E (1991) Isolation and characterization Isolation and characterization of new Methanothrix strains. Int of a novel thermophilic Methanosaeta strain. Int J Syst Bacte- J Syst Bacteriol 41:358–362 riol 41:191–196 48. Ohtsuki T, Watanabe M, Miyaji Y (1992) Start up of thermo- 33. Kudo Y, Nakajima T, Miyaki T, Oyaizu H (1997) Methanogen philic UASB (Upflow Anaerobic Sludge Blanket) reactors us- flora of paddy soils in Japan. FEMS Microbiol Ecol 22:39–48 ing micro-carrier and mesophilic granular sludge. Water Sci 34. Lin C, Stahl DA (1995) Taxon-specific probes for the cellulo- Technol 26:877–886 lytic genus Fibrobacter reveal abundant and novel equine- 49. Patel GB (1984) Characterization and nutritional properties of associated populations. Appl Environ Microbiol 61:1348–1351 Methanothrix concilii sp nov, a mesophilic, aceticlastic meth- 35. Loakes D, Brown DM (1994) 5-Nitroindole as an universal anogen. Can J Bacteriol 30:1383–1396 base analogue. Nucleic Acids Res 22:4039–4043 50. Patel GB (1992) A contrary view of the proposal to assign a 36. Macario AJL, Visser FA, van Lier JB, Conway de Macario E neotype strain for Methanothrix soehngenii. Int J Syst Bacteriol (1991) Topography of methanogenic subpopulations in a mi- 42:324–326 crobial consortium adapting to thermophilic conditions. J 51. Patel GB, Sprott GD (1990) Methanosaeta concilii gen nov, sp Gen Microbiol 137:2179–2189 nov (“Methanothrix concilii”) and Methanosaeta thermoaceto- 37. MacLeod FA, Guiot SR, Costerton JW (1990) Layered struc- phila nom rev, comb nov. Int J Syst Bacteriol 40:79–82 ture of bacterial aggregates produced in an upflow anaerobic 52. Raskin L, Capman WC, Sharp R, Poulsen LK, Stahl DA (1997) sludge bed and filter reactor. Appl Environ Microbiol Molecular ecology of gastrointestinal ecosystems. In: Mackie 56:1598–1607 RI, White BA, Isaacson RE (eds) Ecology and Physiology of 38. Macnaughton SJ, O’Donnell AG, Embley TM (1994) Perme- Gastrointestinal Microbes, Volume 2: Gastrointestinal Micro- abilization of mycolic-acid-containing actinomycetes for in biology and Host Interactions. Chapman and Hall, New York, situ hybridization with fluorescently labeling oligonucleotide pp 243–298 probes. Microbiol 140:2859–2865 53. Raskin L, Poulsen LK, Noguera DR, Rittmann BE, Stahl DA 39. Maestrojuan GM, Boone DR, Xun L, Mah RA, Zhang L (1990) (1994) Quantification of methanogenic groups in anaerobic Transfer of Methanogenium bourgense, Methanogenium biological reactors by oligonucleotide probe hybridization. marisnigri, Methanogenium olentangyi, and Methanogenium Appl Environ Microbiol 60:1241–1248 thermophilicum to the genus of Methanoculleus gen nov, 54. Raskin L, Stromley JM, Rittmann BE, Stahl DA (1994) Group- emendation of Methanoculleus marisnigri and Methanogenium, specific 16S rRNA hybridization probes to describe natural and description of new strains of Methanoculleus bourgense communities of methanogens. Appl Environ Microbiol and Methanoculleus marisnigri. Int J Syst Bacteriol 40:117–122 60:1232–1240 262 D. Zheng, L. Raskin

55. Raskin L, Zheng D, Griffin ME, Stroot PG, Misra P (1995) 66. Touzel JP, Prensier G, Roustan JL, Thomas I, Dubourguier Characterization of microbial communities in anaerobic bio- HC, Albagnac G (1988) Description of a new strain of Metha- reactors using molecular probes. Antonie van Leeuwenheok nothrix soehngenii and rejection of Methanothrix concilii as a 68:297–308 synonym of Methanothrix soehngenii. Int J Syst Bacteriol 56. Rouvie`re P, Mandelco L, Winker S, Woese CR (1992) A de- 38:30–36 tailed phylogeny for the Methanomicrobiales. Syst Appl Micro- 67. Visser FA, van Lier JB, Macario AJL, Conway de Macario E biol 15:363–371 (1991) Diversity and population dynamics of methanogenic 57. Rozanova EP, Savvichev AS, Miller YM, Ivanov MV (1997) bacteria in a granular consortium. Appl Environ Microbiol Microbial processes in a flooded West Siberian oil field with a 57:1728–1734 complex of organic compounds. Mikrobiologiya 66:852–859 68. Wiegant WM (1988) The “spaghetti theory” on anaerobic 58. Sanchez JM, Arijo S, Munoz MA, Morinigo MA, Borrego JJ granular sludge formation, or the inevitability of granulation. (1994) Microbial colonization of different support materials In: Lettinga G, Zehnder AJB, Grotenhuis JTC, Hulshoff Pol used to enhance the methanogenic process. Appl Microbiol LW (eds) Granular Anaerobic Sludge: Microbiology and Biotechnol 41:480–486 Technology Workshop. Puduc Wageningen, Lunteren, The 59. Sekiguchi Y, Kamagata Y, Syutsubo K, Ohashi A, Harada H, Netherlands, pp 146–153 Nakamura K (1998) Phylogenetic diversity of mesophilic and 69. Wu W-M, Hickey RF, Zeikus JG (1991) Characterization of thermophilic granular sludges determined by 16S rRNA gene metabolic performance of methanogenic granules treating analysis. Microbiol 144:2655–2665 brewery wastewater: Role of sulfate-reducing bacteria. Appl 60. Sørensen AH, Torsvik VL, Torsvik T, Poulsen LK, Ahring BK Environ Microbiol 57:3438–3449 (1997) Whole-cell hybridization of Methanosarcina cells with 70. Zehnder AJB, Huser BA, Brock TD, Wuhrmann K (1980) two new oligonucleotide probes. Appl Environ Microbiol Characterization of an acetate-decarboxylating, non- 63:3040–3050 hydrogen-oxidizing methane bacterium. Arch Microbiol 61. Sprott GD, Beveridge TJ (1993) Microscopy. In: Ferry JG (ed) 124:1–11 Methanogenesis. Chapman & Hall, New York, pp 81–127 71. Zheng D, Alm EW, Stahl DA, Raskin L (1996) Characteriza- 62. Stackebrandt E, Goebel BM (1994) Taxonomic note: A place tion of universal small-subunit rRNA hybridization probes for for DNA-DNA reassociation and 16S rRNA sequence analysis quantitative molecular microbial ecology studies. Appl Envi- in present species definition in bacteriology. Int J Syst Bacte- ron Microbiol 62:4504–4513 riol 4:846–849 72. Zheng D, Raskin L, Monitoring granulation processes in an- 63. Stahl DA, Amann RI (1991) Development and application of aerobic reactors using oligonucleotide hybridization probes. I. nucleic acid probes. In: Stackebrandt E, Goodfellow M (eds) Population dynamics. In preparation Nucleic Acid Techniques in Bacterial Systematics. John Wiley 73. Zinder SH (1993) Physiological ecology of methanogens. In: & Sons, Inc, New York, pp 205–248 Ferry JG (ed) Methanogenesis: Ecology, Physiology, Biochem- 64. Stahl DA, Flesher B, Mansfield HR, Montgomery L (1988) Use istry and Genetics. Chapman & Hall, New York, pp 128–206 of phylogenetically based hybridization probes for studies of 74. Zinder SH, Anguish T, Lobo AL (1987) Isolation and charac- ruminal microbial ecology. Appl Environ Microbiol 54:1079– terization of a thermophilic acetotrophic strain of Methano- 1084 thrix. Arch Microbiol 146:315–322 65. Syutsubo K, Harada H, Ohashi A, Suzuki H (1997) An effec- 75. Zinder SH, Cardwell SC, Anguish T, Lee M, Koch M (1984) tive start-up of thermophilic UASB reactor by seeding meso- Methanogenesis in a thermophilic (58°C) anaerobic digestor: philically-grown granular sludge. Water Sci Technol 36:391– Methanothrix sp. as an important aceticlastic methanogen. 398 Appl Environ Microbiol 47:796–807