Acetogens and Acetoclastic Methanosarcinales Govern Methane Formation in Abandoned Coal Mines

Samples were collected in May 2007 in sealed compartments of coal mines closed in the 1960s. The mines harbor hard coal (“rock coal”) belonging to the fat coals according to the German classification (reference 33 and references therein). Briefly, large pieces of coal (2 to 15 cm in diameter) and mine timber (2 to 10 cm in diameter) were collected aseptically in glass bottles that were immediately flushed with N₂ and stored at 4°C until further processing. In situ temperatures were 35 to 36°C, with 100% air humidity. Coal and mine timber samples were processed in an anaerobic chamber under a nitrogen atmosphere to prevent oxidation. Samples were homogenized and distributed into Hungate tubes containing 500 ml of sulfate-free mineral medium (36) with a salinity of 15 practical salinity units (PSU), according to in situ values. Controls were supplemented with 10 mM 2-bromoethanesulfonate (BES) to exclude abiotic degassing by inhibiting methanogenesis. Enrichment cultures were amended with 10 mM fully ¹³C-labeled acetate (Campro Scientific) or [¹³C]bicarbonate (Campro Scientific) plus H₂. All incubated samples were frozen at −80°C after an incubation time of 6 months until further processing for DNA-SIP. All incubations were carried out at least in triplicate. Subsamples of all incubated samples were taken at the beginning and at month 3 of incubation. The increases in the amounts of methane and hydrogen in the headspace, as well as the stable isotopes of methane, were monitored continuously over 6 months and were analyzed by gas chromatography-mass spectrometry (GC-MS) as described previously by Krüger et al. (13). Concentrations of acetate were analyzed by high-performance liquid chromatography (Agilent Technologies) using a Zorbax Eclipse Plus C₈ USP L7 column (Agilent Technologies) at 60°C. The eluent was a 5 mM H₂SO₄-methanol gradient at 1 ml/min. Acetate was detected by a diode array detector (DAD; Agilent Technologies).

DNA was extracted from 0.5 g of incubation slurry after 0, 3, and 6 months using phenol-chloroform extraction as described by Lueders et al. (18). Three parallel extractions were carried out, and extracts were pooled for each incubation treatment. DNA was checked by standard agarose gel electrophoresis and was quantified using PicoGreen staining according to the method of Wilms et al. (37).

Gradient preparation, isopycnic centrifugation, and gradient fractionation were performed as described by Lueders et al. (17) with minor modifications. Each gradient consisted of 6.3 ml of CsCl (approximately 1.72 g ml⁻¹; Calbiochem) and ca. 1 ml of gradient buffer (100 mM Tris-HCl [pH 8.0] liter⁻¹, 100 mM KCl liter⁻¹, 1 mM EDTA liter⁻¹) including 2,000 ng of DNA. Prior to centrifugation, the average density of the centrifugation medium was controlled refractometrically and was adjusted to 1.84 g cm⁻³. The samples were centrifuged in 6.3-ml Quick-Seal Polyallomer tubes (Beckman) in an NVT 65 near-vertical rotor (Beckman) using an LE-70 ultracentrifuge (Beckman Instruments). Centrifugation was performed at 20°C for 36 h at 44,500 rpm (184,000 × g). Gradients were fractionated as described previously by Neufeld et al. (21). Briefly, the gradients were fractionated from bottom to top into 12 equal fractions (400 μl). A precisely controlled flow rate was achieved by displacing the gradient medium with water at the top of the tube using a Graseby 3100 syringe pump at a flow rate of 1 ml min⁻¹. The density of each collected fraction (a small aliquot of 100 μl) was measured by determining the refractory index using a digital refractometer (model AR-20; Reichert Analytical Instruments, Depew, NY). Subsequently, the DNA was precipitated using polyethylene glycol 6000 (Aldrich Chemistry). The DNA pellet was washed once with 70% ethanol and was dissolved in 25 μl of elution buffer.

DNA was precipitated from gradient fractions and was quantified fluorometrically and by quantitative PCR (qPCR) using the Ar109f/Ar912rt and Ba27F/907R primer systems described by Lueders et al. (17). The qPCR mixtures contained 12.5 μl of the premix solution of a DyNAmo HS SYBR Green qPCR kit (New England Biolabs, Inc., Hitchin, United Kingdom), 1 μl of each primer, and 10 μl of the standard or DNA extract as a template in a final reaction mixture volume of 25 μl. The PCR was carried out in a Rotor-Gene 3000 cycler (Corbett Research, Sydney, Australia). After initial denaturation at 95°C for 15 min, 50 cycles followed. Each cycle consisted of denaturation for 30 s at 94°C, annealing for 20 s at 52°C, elongation for 30 s at 70°C, and fluorescence measurement at 70°C. To check the amplification specificity, fluorescence was also measured at the end of each cycle for 20 s at 80°C. After the last cycle, a melting curve was recorded by increasing the temperature from 50°C to 99°C (1°C every 10 s). The numbers of bacterial and archaeal 16S rRNA gene targets were calculated from the DNA concentrations according to the work of Süss et al. (30). DNA standards for quantitative (real-time) PCR were prepared as described by Wilms et al. (37) and Engelen et al. (6). Bacterial and archaeal targets were measured in at least three different dilutions of DNA extracts (1:10 to 1:1,000) and in triplicate.

For denaturing gradient gel electrophoresis (DGGE), an 803-bp fragment of the archaeal 16S rRNA gene was amplified by using primers Ar109f (5′-AC KGC TCA GTA ACA CGT-3′) and Ar912rt (5′-GTG CTC CCC CGC CAA TTC CTT TA-3′). For the analysis of bacterial composition, primers BA27F (5′-AGA GTT TGA TCM TGG CTC AG-3′) and 907R (5′-CCG TCA ATT CCT TTG AGT TT-3′) were used to amplify an 880-bp fragment of the bacterial 16S rRNA gene. At the 5′ end of each forward primer, an additional 40-nucleotide GC-rich sequence (GC-clamp) was added to achieve a stable melting point for the DNA fragments in the DGGE according to the method of Muyzer et al. (20). PCR amplification was performed using an Eppendorf thermal cycler system (Mastercycler; Eppendorf, Hamburg, Germany) as follows: 2 μl (1 to 100 ng) of template DNA, 1 U of Taq DNA polymerase, the manufacturer's recommended buffer supplied with the polymerase enzyme, 10 mM deoxynucleoside triphosphates (dNTPs), 50 μM (each) appropriate primers, and 10 mM bovine serum albumin (BSA) were adjusted to a total volume of 50 μl with PCR-grade water (Ampuwa; Fresenius, Bad Homburg, Germany). The PCR program for archaeal and bacterial DNA included an initial denaturation step for 5 min at 94°C, followed by 30 cycles of 30 s at 94°C, 30 s at 52°C, and 1 min at 72°C. Primer extension was carried out for 5 min at 72°C. Aliquots (5 μl) of the PCR products were analyzed by agarose gel electrophoresis in 1.5% (wt/vol) agarose gels and ethidium bromide (0.8 ng ml⁻¹) staining for 20 min on a UV transilluminator as described previously (37).

DGGE was performed using an INGENYphorU-2 system (Ingeny, Goes, Netherlands). PCR products and loading buffer (40% [wt/vol] glycerol, 60% [wt/vol] 1× Tris-acetate-EDTA [TAE], and bromphenol blue) were mixed in a 1:2 ratio. The PCR amplicons were applied directly to 6% (wt/vol) polyacrylamide gels with a linear gradient of 30 to 80% denaturant for archaeal PCR products and 50 to 70% denaturant for bacterial PCR products (100% denaturant corresponds to 7 M urea and 40% [vol/vol] formamide). Electrophoresis was carried out in 1× TAE buffer (40 mM Tris-acetate [pH 7.4], 20 mM sodium acetate, 1 mM Na₂ EDTA) at a constant voltage of 100 V and a temperature of 60°C for 20 h. After electrophoresis, the gels were stained for 2 h in 1× SYBR Gold solution (Molecular Probes, Eugene, OR) in 1× TAE and were washed for 20 min with distilled water. The gel was digitized using a digital imaging system (BioDocAnalyze; Biometra, Göttingen, Germany) with UV transillumination (302 nm).

DGGE bands were excised for sequencing using a sterile scalpel and were treated as described by Del Panno et al. (3). Briefly, the bands were transferred into 50 μl of PCR-grade water and were incubated for 48 h at 4°C. For reamplification, 2 μl of the supernatant was taken as a template using the reaction mixture described above with a final volume of 50 μl. The PCR protocol adjusted for the reamplification was the same as that described above, with the following minor modifications: the total number of cycles was 25, and the final elongation was carried out at 72°C for 10 min. PCR products were purified by using the QIAquick PCR purification kit (Qiagen GmbH) and were eluted in 30 μl of PCR-grade water. DNA yields were estimated fluorometrically in a microtiter plate reader (FLUOstar Optima; BMG Labtechnologies, Offenburg, Germany) using a 1:200-diluted PicoGreen reagent according to a modified manufacturer′s protocol (Molecular Probes, Eugene, OR) as described in detail by Wilms et al. (37). Sequence analyses were performed by GATC Biotech AG (Konstanz, Germany). Sequences were compared to those in GenBank using the BLAST tool of the National Center for Biotechnology Information server (1).

The sequences determined in this study have been deposited in the GenBank nucleotide sequence database under accession no. FR838951 to FR838999.

CH₄ production was observed in all incubated samples (Fig. 1A), with higher activities in timber than in coal enrichment cultures. The highest CH₄ formation rates (0.13 μmol per g [wet weight] and day) were detected in the [¹³C]acetate enrichment cultures between months 3 and 6 of incubation. The addition of H₂-¹³CO₂ resulted in less stimulation of methanogenesis (maximum rate, 0.05 μmol per g [wet weight] and day). The isotopic signature of CH₄ indicated that methane was formed from the labeled substrates added (data not shown). No methane formation was observed in control samples incubated with the methanogenesis inhibitor 2-bromoethanesulfonate (BES). Therefore, the possibility of abiotic degassing of adsorbed methane from the incubated samples can be excluded.

While H₂ was largely used up after 3 months, acetate was completely depleted after 6 months (Fig. 1B and C). Interestingly, acetate formation was observed in the H₂-¹³CO₂ cultures, while there was no H₂ formation in the samples incubated with acetate. The fact that acetate formation was also detected in the unamended incubated samples indicates the central role of acetate in the process of methanogenesis. The isotopic signature of the acetate formed showed strong labeling, indicating its formation from ¹³CO₂ (data not shown).

In order to identify active community members, samples from the ¹³C-enriched incubated samples were analyzed by SIP. Density-resolved archaeal DNA was detected in gradient fractions using archaeal qPCR (Fig. 2 and 3). The amount of archaeal DNA detected in the heavy fractions increased over the incubation time between months 3 and 6. Only the timber cultures showed ¹³C-labeled archaeal DNA after 3 months, with maximum quantities in DNA gradients from acetate enrichment cultures. After 6 months, “light” and “heavy” DNA fractions substantiated a clear labeling of archaeal DNA in timber and coal cultures amended with [¹³C]acetate or ¹³CO₂ (Fig. 2 and 3).

Next, the archaeal community members detected in “light” and “heavy” gradient fractions were analyzed by denaturing gradient gel electrophoresis (DGGE) and subsequent band sequencing (Tables 1 and 2). In general, the compositions of the total archaeal communities in the [¹³C]acetate and H₂-¹³CO₂ cultures were similar, and these communities were dominated by relatives of Methanosarcina spp., Methanosaeta spp., and uncultured Crenarchaeota (Tables 1 and 2). However, clear ¹³C labeling was evident mainly for relatives of Methanosarcina spp. Increased band intensities reflected larger amounts of ¹³C-labeled DNA in acetate cultures (Fig. 2 and 3). To a minor extent, labeled DNA was also observed for a DGGE band related to Methanosaeta spp. in the acetate enrichment cultures. Members of the Crenarchaeota were abundant in the original coal and timber samples but showed no incorporation of ¹³C-labeled substrates.

Quantities of labeled bacterial 16S rRNA genes increased over the incubation time (Fig. 2 and 3), as found for the Archaea. While strongly labeled bacterial DNA was detected especially in the [¹³C]acetate-amended coal samples after 3 months, H₂-¹³CO₂ cultures showed substantial quantities of ¹³C-labeled DNA only after 6 months of incubation.

In the acetate-amended coal samples, bacterial DNA labeled after 3 months was affiliated with a lineage of uncultured Geobacteraceae and Pelobacter spp., and bacterial DNA labeled after 6 months was additionally affiliated with Pseudoalteromonas and Clostridium spp. Most surprisingly, however, abundant unlabeled and labeled DNA was detected for a relative of a Pseudomonas sp. after 6 months only. In H₂-¹³CO₂-amended coal samples, clearly labeled DNA was evident from gradient fractions in DGGE analysis after 6 months only and was affiliated with Clostridium, Desulfovibrio, and Pelobacter spp., as well as, again, with uncultured Geobacteraceae.

In [¹³C]acetate-amended timber cultures, members of chemoautotrophic bacteria (Hydrogenophaga and Hyphomicrobium spp.) as well as sulfate and sulfur reducers (Desulfovibrio and Desulfuromonas spp.) were detected primarily in “intermediate” and “heavy” gradient fractions. Surprisingly, the DGGE band dominating the highest-density DNA gradient fractions was related to Burkholderia spp. This could be a methodological bias due to the high-GC DNA content of Burkholderia spp. In H₂-¹³CO₂-amended timber microcosms, Hydrogenophaga spp. clearly dominated “heavy” DNA after 6 months, while a relative of Desulfovibrio spp. was detected in “intermediate” gradient fractions after 3 months.

The fact that Methanosarcina spp. were responsible for methane production in our enrichment cultures is in agreement with our earlier studies showing that the Methanosarcinales are most abundant in in situ coal and timber samples (2). Although Methanosarcina spp. are known to use either hydrogen or acetate, those identified here seem to be strictly adapted to the conditions in this habitat. In the mines studied, acetate seems to be quantitatively more available for the Methanosarcinales. Hydrogen might hardly be formed at low metabolic rates and therefore might not be available for methanogens. Even after an incubation time of 6 months with an adequate supply of hydrogen, coal mine methanogens did not make direct use of hydrogen for methane production. Hydrogen appeared rather to be used by acetogens producing acetate, which then, in turn, was utilized by the Methanosarcinales.

Acetotrophic members of the Geobacteraceae were found to be labeled in the coal enrichment cultures. Their abundance could be increased by amendment with acetate. However, they were also active in the samples incubated with hydrogen. This could be explained by secondary cross-feeding processes, since labeled acetate was formed (15). In our cultures, it is not clear which electron acceptor is used by the Geobacteraceae for acetate oxidation. One possibility could be the utilization of electron acceptors, such as sulfur, directly from the coal, since members of the family Geobacteraceae are reported to be elemental-sulfur reducers (11). Jones et al. (12) also obtained high numbers of Geobacter species from coal, but none of the known electron acceptors was present, suggesting that Geobacter might be capable of coupling the degradation of organics to an electron- or H₂-accepting partner. This could also be proposed in our case. However, in contrast to our strictly anoxic enrichment cultures, coal mines showed low concentrations of oxygen in layers of coal near the surface, and Geobacteraceae constituted the bulk of the overall bacterial community in the original coal samples (F. Gründger, unpublished data).

The active bacterial communities in coal and timber differ. Amendment with acetate or H₂-CO₂ did not have a strong effect on the composition of the active community. Besides the Geobacteraceae, the active community in the coal samples comprised Pelobacter acetylenicus and members of Clostridium and Pseudomonas species. In earlier studies, Pseudomonas stutzeri was isolated from coal samples; this species is potentially able to utilize polycyclic aromatic hydrocarbons (PAHs) (24). This suggestion can be supported by the fact that the first event of coal fragmentation is exoenzymatic hydrolysis into small PAHs (29). Moreover, Pseudomonas stutzeri is even more active when acetate is present as a second electron donor (19), and that could be one reason for its predominance in the coal enrichment cultures amended with acetate. However, oxygen and nitrate were not available as electron acceptors.

Accumulation of acetate was detected in both supplemented and nonsupplemented cultures. The presence of isotopically heavy acetate in samples incubated with H₂-¹³CO₂ indicated its new formation from CO₂. Relatives of Pelobacter and Clostridium species might be involved in this process. For Pelobacter acetylenicus, acetate formation from acetylene has been described as well (23). Which other coal and timber compounds could be feasible substrates for acetogenesis is not known. However, acetate formation appeared to continue, since the levels do not approach zero in most cases.

In the timber enrichment cultures, active bacteria similar to Hydrogenophaga and Clostridium species predominated, suggesting that they might use timber compounds or secondary fermentative products for metabolism. Only the H₂-¹³CO₂-amended coal cultures showed distinct labeling of bacteria after 6 months. The slight heavy shift in the acetate-amended coal cultures might not be an indication of bacterial activity, but the result of an increase in the amount of bacterial DNA with a high GC content, such as Burkholderia species (Fig. 3; Table 2). Like the coal cultures, the H₂-¹³CO₂-enriched timber cultures also showed acetate formation. Obviously, timber provides acetogenic substrates other than those from coal. However, Hydrogenophaga as well as Clostridium species were recently also detected in coal samples from another mine (12).

In connecting the results from our earlier investigations (2, 14) with our new findings in the present study, we present the following arguments supporting the notion that acetoclastic methanogenesis is the main route for methane formation. First, the natural isotopic signatures of methane indicated an acetoclastic origin, supported by the isotopic signatures of acetate that was formed from ¹³CO₂ (14). Second, the highest methane formation rates were observed in the acetate-amended enrichment cultures of coal and timber, while H₂ gave lower activities. Third, acetate was depleted in the acetate cultures but accumulated in the H₂-CO₂- and BES-treated enrichment cultures. Finally, DNA-SIP revealed that relatives of Methanosarcina spp. were responsible for methane production in both [¹³C]acetate- and H₂-¹³CO₂-amended cultures, and in the latter, their activity was coupled to the activity of bacteria related to acetogens.