Methanogenic Pathway and Archaeal Community Structure in the Sediment of Eutrophic Lake Dagow: Effect of Temperature

K. Glissmann1, K.-J. Chin1, P. Casper2 and R. Conrad1

(1) Max-Planck-Institute for Terrestrial Microbiology, Karl-von-Frisch Str., 35043 Marburg, Germany (2) Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Alte Fischerhutte 2, 16775 Stechlin-Neuglobsow, Germany Received: 28 May 2003 / Accepted: 08 August 2003 / Online publication: 29 June 2004

Abstract

Methanogenic degradation of organic matter is an important microbial process in lake sediments. Temperature may affect not only the rate but also the pathway of CH4 production by changing the activity and the abundance of individual microorganisms. Therefore, we studied the function and structure of a methanogenic community in anoxic sediment of Lake Dagow, a eutrophic lake in north-eastern Germany. Incubation of sediment samples (in situ 7.5C) at increasing temperatures (4, 10, 15, 25, 30C) resulted in increasing production rates of CH4 and CO2 and in increasing steadystate concentrations of H2. Thermodynamic conditions for H2/CO2-dependent methanogenesis were only exergonic at 25 and 30C. Inhibition of methanogenesis with chloroform resulted in the accumulation of methanogenic precursors, i.e., acetate, propionate, and isobutyrate. Mass balance calculations indicated that less CH4 was formed via H2 at 4C than at 30C. Conversion of 14 CO2 to 14CH4 also showed that H2/CO2-dependent methanogenesis contributed less to total CH4 production at 4C than at 30C. [2)14C]Acetate turnover rates at 4C accounted for a higher percentage of total CH4 production than at 30C. Collectively, these results showed a higher contribution of H2-dependent methanogenesis and a lower contribution of acetate-dependent methanogenesis at high versus low temperature. The archaeal community was characterized by cloning, sequencing, and phylogenetic analysis of the 16S rRNA genes retrieved from the sediment. Sequences were afliated with Methanosaetaceae, Methanomicrobiaceae, and three deeply branching euryarchaeotal clusters, i.e., group III,

Rice cluster V, and a novel euryarchaeotal cluster, the LDS cluster. Terminal restriction fragment length polymorphism (T-RFLP) analysis showed that 16S rRNA genes afliated to Methanosaetaceae (2030%), Methanomicrobiaceae (3555%), and group III (1025%) contributed most to the archaeal community. Incubation of the sediment at different temperatures (430C) did not result in a systematic change of the archaeal community composition, indicating that change of temperature primarily affected the activity rather than the structure of the methanogenic community.

Introduction

Present address of K.-J. Chin: Department of Microbiology, University of Massachusetts, Amherst, MA 01003, USA Correspondence to: R. Conrad; E-mail: conrad@staff.uni-marburg.de

In lake sediments methanogenesis is the most important terminal processes in the anaerobic degradation of organic material. Organic matter is hydrolyzed and fermented, and the fermentation products are converted to acetate and H2, which in turn serve as methanogenic substrates [42, 49]. During summer stratication, inorganic electron acceptors such as nitrate, ferric iron, and sulfate are depleted in the lake sediment. Otherwise, nitrate reducers, iron reducers, and sulfate reducers compete successfully for methanogenic substrates, especially in the surface layers of the sediment [6, 7, 32, 44]. Various lake sediments have been studied with respect to the functioning of methanogenic pathways in the absence of inorganic electron acceptors other than CO2. Under balanced conditions methane is produced from either acetate or H2/CO2 at a ratio of 2:1, but frequently this ratio was found to be higher (reviewed by [15]). Sediment processes can be affected by the activity of individual microorganisms, but also by the composition of the microbial community. Studies of the structure of the methanogenic community together with its functioning (i.e., acetate versus H2-dependent methanogenesis) are to our knowledge extremely scarce [36, 50]. In the
389

DOI: 10.1007/s00248-003-2027-2

Volume 48, 389399 (2004)

Springer Science+Business Media, Inc. 2004

390

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

sediment of Swiss Lake Rotsee, for example, H2-dependent methanogenesis was only detected in the surface sediment layers and CH4 production was generally dominated by acetoclastic methanogenesis [50]. The methanogenic community was accordingly dominated by acetoclastic Methanosaeta species and hydrogenotrophic Methanomicrobiaceae, but other euryarchaeal clones, characterized as Rice cluster V, were also detected [50]. In the sediment of Lake Kinneret (Israel), on the other hand, CH4 was produced from both acetate and H2/CO2 [36]. The archaeal community consisted of hydrogenotrophic Methanomicrobiaceae, Methanobacteriaceae, and Euryarchaeota of group III, while acetoclastic methanogens were not detected [36]. There is circumstantial evidence that in this sediment fermentatively produced acetate is degraded by a syntrophic mechanism rather than acetoclastic methanogenesis [36, 38]. The molecular community structure of the methanogens and of other archaea has been studied in some other lake sediments, demonstrating the presence of particularly Methanomicrobiaceae and Methanosaetaceae [10, 26, 30], but occasionally also of Methanobacteriaceae and Methanosarcinaceae [10, 34]. Members of other archaeal clusters with unkown phenotype, e.g., euryarchaeotal group III or crenarchaeotal groups I.1a and I.2, have also been detected [30, 34]. Schulz and Conrad [40] observed a change in the methanogenic degradation pattern of organic matter when sediment samples from oligotrophic Lake Constance were incubated at increasing temperatures from 4 to 20C, resulting in higher contribution of H2/CO2 to CH4 production. The authors hypothesized that at low temperature organic matter degradation may be dominated by homoacetogenesis followed by acetate-dependent methanogenesis. At higher temperature, on the other hand, other fermentation pathways, followed by syntrophic oxidation of fatty acids to acetate, CO2, and H2, would become increasingly important, thus resulting in the theoretically expected ratio of acetate to H2dependent CH4 production of about 2:1. This hypothesis is in agreement with a similar temperature-dependent shift of degradation pattern in anoxic rice eld soil [11] and with the observation that enrichment cultures on H2/ CO2 usually result in the dominance of chemolithotrophic homoacetogens at low temperatures and of methanogens at high temperatures [16, 31, 35]. Results obtained with profundal sediment of dystrophic Lake Plusee were equivocal, although they did not disprove this hypothesis [37]. In contradiction to the hypothesis, incubation experiments with sediment from White Oak River Estuary showed no change of the methanogenic degradation pathway with increasing temperature [1, 2]. Nevertheless, more methanogenic sediments need to be studied in order to nd out whether the hypothesis is a general principle or not. In particular, it is necessary to

investigate the temperature effect with respect not only to function but also to microbial community structure. However, little is known about the microbial composition of methanogenic soils and sediments and how it is affected by temperature. The objective of our study was to investigate both structure and function of the methanogenic archaeal community in the anoxic profundal sediment of a eutrophic lake, i.e., Lake Dagow. Another objective was to test the hypothesis that hydrogenotrophic methanogenesis becomes increasingly important when temperature increases. Finally, we checked whether a temperature-induced change of the methanogenic pathway would be reected in the composition of the methanogenic archaeal community considering the major phylogenetic groups.
Materials and Methods
Study Site and Sampling. The eutrophic Lake Dagow is located in northern Brandenburg, Germany. Some of the main morphological and limnological parameters were described by Casper [9]. Samples were taken in the middle of the lake at a depth of 8.5 m in July 2000 and July 2001 with a grab sediment sampler type petite Ponar (Wildco Company, Michigan, USA). At both sampling dates the water near the sediment had a temperature of 7.5C. The overlying water was O2-free, as determined with a Oxi 197 probe (WTW, Weilheim, Germany). After sampling, the upper 10 cm of the sediment was placed into air-tight bottles without any gas headspace and stored at 4C until use for physiological experiments. The incubation experiments (with and without inhibitors) were started on the next day. The radioactive experiments were started 6 weeks later, after checking that the CH4 production rates were the same as immediately after sampling. Additional samples were stored at )18C (fresh material). The sediment samples obtained in 2000 were used for exploratory T-RFLP analysis. The sediment samples collected in 2001 were used for physiological experiments and T-RFLP analysis, and for creation of a clone library. For comparison, sediment samples of dystrophic Lake Plusee were obtained in summer 1999 [36] and used to create a clone library. Preparation and Incubation of Samples. Sediment slurries were prepared from 10 mL homogenized sediment (approximately 010 cm depth) plus 10 mL anoxic distilled water in 120-mL serum bottles, which were closed with butyl rubber stoppers (for measuring gases at all temperatures and dissolved compounds at 4C) or with 5 mL sediment and 5 mL anoxic water in 60-mL bottles (for measuring dissolved compounds at 30C),

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

391

respectively. In order to remove all dissolved CH4, the bottles were evacuated several times and ushed with O2free N2, then shaken at 120 rpm for 1 h in the dark, again evacuated, ushed with N2, and nally pressurized to 1.2 bar. Separate portions were incubated in triplicate at 4, 10, 15, 25, or 30C in the dark without shaking. The mixing ratios of CH4, CO2, and H2 in the gas phase were repeatedly measured by taking gas samples (0.2 mL) with gas-tight syringes. Prior to sampling the bottles were shaken vigorously to achieve equilibrium between the gas and liquid phase. The concentrations of CH4, CO2, and H2 were determined by gas chromatography as described before in detail [25]. At the end of the incubation sediment samples were removed from the bottles and stored at )18C for analysis of dissolved fatty acids and T-RFLP.
Inhibition and Tracer Experiments. Methanogenesis in sediment slurries was inhibited after 15 (4C) and 27 days (30C) of incubation by addition of chloroform at a nal concentration of 200 lM. After addition the samples were shaken vigorously and then incubated at 4 and 30C in the dark without shaking. The sediment slurries exhibited a pH of 7.5 0.1 (4C) and 7.2 0.2 (30C) over the entire incubation time. Gas samples were taken repeatedly as described above. After the last gas sampling the slurry was transferred to Eppendorf tubes. The samples were centrifuged at 13,000 g for 15 min at 4C. The supernatant was removed and stored at )18C until analysis. After ltration through 0.2-lm membrane lters (regenerated cellulose, Sartorius, Gottingen, Germany) the supernatant was analyzed for dissolved compounds by high-pressure liquid chromatography as described before [25]. For the radioactive experiments NaH14CO3 (54 mCi mmol)1) and [2-14C]acetate (58 mCi mmol)1) (both Amersham, Braunschweig, Germany) were added to sediment slurries to give a nal radioactivity of 1.7 (5.6 104 Bq) and 1.57 lCi (5.2 104 Bq), respectively. Gas samples were taken from the headspace and analyzed for nonradioactive and radioactive CH4 and CO2 as described previously [25]. The respiratory index of [2)14 C]acetate turnover was determined at the end of incubation after acidication of the sediment slurry, using RI = 14CO2 /(14CO2 + 14CH4). Calculations. For mass balance calculations, the metabolites that accumulated in the inhibited incubations were compared to the CH4 formed in the uninhibited control as described in detail by Glissmann and Conrad [25]. The amounts of accumulated metabolites were converted into equivalents of CH4 according to the stoichiometry of the respective syntrophic pathway [11]. These equivalents of CH4 were compared to the CH4, which was formed in the control.

The conversion of H14CO to 14CH4 was used to 3 determine the fraction (fH2) of CH4 produced from H2/ CO2 [19]. The respiratory index (RI) for the degradation of [2)14C]acetate, the turnover time (Tt) and turnover rate (VAc) of [2)14C]acetate, and the rate of CH4 production from acetate (PAc) were calculated from the conversion of [2)14C]acetate to 14CH4 and 14CO2 according to [37]. The concentrations of gases and dissolved compounds in the slurry incubations were used to calculate the Gibbs free energies (DG) of methanogenesis and homoacetogenesis [18]. The standard Gibbs free energies (DG) and standard reaction enthalpies (DH) were calculated from the standard Gibbs free energies and standard reaction enthalpies of formation (DGf, DHf) of the reactants and products [45]. The standard Gibbs free energies were corrected for 4 and 30C by the Vant Hoff equation (see [20]).
DNA Extraction from Sediment Slurries. The samples for molecular analysis were taken from sediment slurries that had been incubated for 73 days at 4, 10, 15, 25, and 30C. In addition, fresh sediment samples (stored frozen) were analyzed from Lake Dagow and Lake Plusee. The DNA extraction protocol has already been described [13] and was a modication of a previously described one [41]. Briey, a slurry sample (0.5 g) was mixed with 0.5 mL of sodium phosphate buffer (pH 8.0, 120 mM), 120 lL of sodium dodecyl sulfate (10%), and 1 g of glass beads (0.1 mm in diameter). After incubation (10 min, 65C) and two 30-s cycles of bead beating at 2500 rpm (Micro-Dismembrator; B. Braun Biotech. International.), the slurry was centrifuged (10 min, 13,000 g). The DNA was extracted from the supernatant with phenolchloroform (1:1, v/v) and chloroformisoamyl alcohol (24:1, v/v), precipitated from the aqueous phase using 0.1 volume of sodium acetate (3 M, pH 5.3) and 2 volumes of ethanol, and then puried by cesium chloride as described elsewhere [41]. Amplication of Archaeal 16S rRNA Genes and TThe 16S rDNA fraction of the DNA RFLP Analysis.

samples was amplied by PCR with the archaeal groupspecic primers described by Grosskopf et al. [27], which amplify from positions 109 to 934 (Escherichia coli 16S rRNA numbering [5]) according to a previously described protocol [13]. The thermal prole used for amplication of DNA included 20 (for cloning) and 24 to 29 (for T-RFLP analysis) cycles of denaturation at 94C (45 s), primer annealing at 52C (1 min), and primer extension at 72C (1 min). The principle of the T-RFLP analysis has been described by Liu et al. [33]. The reverse primer was labeled 5-terminal with FAM (5-carboxyuorescein). Aliquots of the puried 16S rDNA amplicons were digested by TaqI (Promega Mannheim,

392

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

Germany), and subsequent T-RFLP analysis was performed as described before [13].
Cloning, sis. Sequencing, and Phylogenetic Analy-

A clone library of the amplied archaeal 16S rDNA was constructed using the DNA extracts from freshly sampled sediment of Lake Dagow and Lake Plusee, as described by Chin et al. [13]. A total of 52 and 12 clones, respectively, were randomly selected from these two sediment samples, and sequence analysis was carried out following Chin et al. [13]. The ARB software package and its databases [43] were used to phylogenetically analyze the sequence data as described before [13]. In order to ensure the correct phylogenetic placement of the obtained sequences, new sequences were added to the ARB database consisting of 792 complete or partial (>400 nucleotides) archaeal SSU rDNA sequences from public databases. Phylogenetic placement was done in comparison to reference sequences for the main lines of descent within the archaeal kingdom Euryarchaeota and Crenarchaeota [48], as well as Korarchaeota [3]. A 50% invariance criterion for inclusion of individual nucleotide sequence positions in the analysis was used to avoid possible treeing artifacts. To avoid reconstruction artifacts arising from partial sequences, an initial phylogenetic tree was constructed from almost complete SSU rDNA sequences using FITCH distance matrix analysis and veried by neighbor joining and maximum likelihood methods. Sequences from this study (640 to 771 nucleotides in length) were added to this tree using the ARB parsimony tool without altering the global tree topology. The nal tree topology was evaluated by performing maximum likelihood (fastDNAml) analysis.
Accession Numbers. The sequences of SSU rDNA clones obtained in this study have been deposited in the EMBL, GenBank, and DDBJ nucleotide sequence databases under the following accession numbers: Lake Dagow sediment clones LDS1 to LDS52, AY133896 AY133947; Lake Plusee sediment clones PSS1 to PSS12, AY133884AY133895.

Results

Figure 1. Partial pressures of H2 (A) and production rates of CH4

The production of CH4 and CO2 was constant from day 8 for the rest of the incubation time (73 days in total) at any temperature. Production rates were calculated by linear regression (r2 > 0.95) and are given per cm3 sediment (Fig. 1). At 4C, production rates of CH4 and CO2 were similar. The rates increased with increasing temperature, but CH4 production rates increased relatively faster and at 30C, CH4 production rates were almost twice as high as CO2 production rates (Fig. 1). Note that production rates of
Incubation Experiments.

(B) and CO2 (C) in sediment slurries incubated at different temperatures (bars = SE, n = 3).

gaseous CO2 were not corrected for dissolved inorganic carbon. Production rates of dissolved inorganic carbon were probably similar to those of gaseous CO2. The partial pressure of H2 in the headspace of the incubation vessels stayed constant after 4 days of incubation with values increasing with increasing temperatures (Fig. 1). The H2 partial pressures at 4, 10, and 15C did not allow

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

393

for exergonic H2/CO2-dependent methanogenesis (DG > +6 kJ mol)1 CH4). Only at 25C (DG = )9 kJ mol)1 CH4) and 30C (DG = )13 kJ mol)1 CH4) did the reaction become slightly exergonic. The sediment was analyzed for dissolved compounds at the end of the incubation. Only acetate and formate were detected at low concentrations (<10 lM), with no difference between the different incubation temperatures. Despite the low acetate concentration, acetoclastic methanogenesis was always exergonic with DG values increasing from )28 kJ mol)1 CH4 at 4C to )18 kJ mol)1 CH4 at 30C.
Inhibition Experiments. For inhibition experiments chloroform was added to sediment slurries, which were incubated subsequently at 4C and 30C. Inhibition of methanogenesis resulted in the accumulation of H2, acetate, propionate, and isobutyrate (only in 30C incubations). Formate was also detected, but did not accumulate. Therefore, it was not considered as a major precursor of methanogenesis. The amounts of the intermediates that accumulated in the absence of methanogenesis were balanced against the CH4 that was produced in the uninhibited control (Table 1). At both temperatures acetate was the most important intermediate, accounting for more CH4 at 4C (59%) than at 30C (37%). Hydrogen, on the other hand, accounted for more CH4 at 30C (9%) than at 4C (2%). Propionate was also an important intermediate in both incubations, whereas isobutyrate was only found at 30C. A comparison of the CH4 formed from all accumulated intermediates and from calculated equivalents of acetate and H2 showed that at 4C about 87% and at 30C about 76% of the CH4 was formed via acetate (Table 1). Radiotracer Experiments. The fraction (fH2) of CH4 produced from H2/CO2 was determined in samples incubated at 4 and 30C. After addition of NaH14CO3, radioactive and nonradioactive CH4 and CO2 were followed by repeated sampling (n = 5) between days 12 and 21, when sufcient 14CH4 had been produced. The value of fH2 was then constant. H2/CO2 contributed 24 4% to CH4 production at 4C and 40 3% at 30C. The conversion of [2)14C]acetate to 14CH4 was measured at 4C and 30C. [2)14C]Acetate was completely turned over within 21 h, with $3040% being converted to gaseous products, predominantly CH4 as indicated by the low RI (Table 2). The remainder was recovered in the sediment and was presumably adsorbed to sediment particles or assimilated into microbial biomass. The turnover time of [2)14C]acetate and the acetate-dependent production (PAc) of CH4 were faster at 30C than 4C (Table 2). In relation to the total CH4 production rates, acetate turnover accounted for a 2.5 times higher percentage (24%) at 4C

Table 1. Percentage amounts of metabolites accumulated in the

presence of an inhibitor (chloroform) during the incubation of sediment at 4 and 30Ca Metabolite H2 Acetate Propionate Isobutyrate Total CH4 from metabolites Via acetateb Via H2b 4C 21 59 6 22 8 0 88 15 87 3 13 3 30C 9 37 13 6 65 76 24 2 2 1 0 4 2 2

a Amounts of metabolites were converted to equivalent amounts of H2, acetate and CH4 according to the stoichiometries of conversion, and balanced against the CH4 in the uninhibited control. Rates of CH4 production were similar to those shown in Fig. 1. The table shows mean values ( SE) measured at three different time points during the incubation. b Equivalent CH4 produced from equivalent acetate or H2 balanced against total CH4 from metabolites.

than at 30C (9%) (Table 2). However, acetate turnover rates added up with H2-dependent CH4 production rates to only $50% of total CH4 production. We assume that acetate turnover rates were systematically underestimated because of compartmentalization of acetate concentrations [14, 39].
Structure of the Archaeal Communities. The archaeal community structure in the sediment was analyzed by T-RFLP analysis targeting the 16S rDNA. Six major terminal restriction fragments (T-RF) were detected, which were assigned to different archaeal lineages (Table 3). The assignment was based on a library of 16S rDNA clones retrieved from Lake Dagow sediment. Randomly selected clones (n = 52) were both analyzed by T-RFLP and se-

Table 2. Turnover of [2)14C]acetate and rate of methanogenesis

in sediment of Lake Dagow incubated at 4 and 30C; mean SE; n=3 Parameter/substrate Maximum fraction of [2)14C]acetate converted to 14CH4 and 14CO2 Tt, turnover time (min) Cacetate, substrate concentration (nmol cm)3) Vac, turnover rate of acetate (nmol h)1 cm)3) RI Pac, CH4 production calculated from acetate turnover (nmol h)1 cm)3) CH4 production rate (nmol h)1 cm)3) Pac in relation to actual CH4 production rate 4C 0.28 0.03 30C 0.39 0.07

46.4 16.7 13.3 1.6 0.09 0.04 0.14 0.05 0.08 0.02

28.0 4.9 38.0 2.4 0.56 0.16 0.2 0.03 0.45 0.14

0.33 0.19 24%

4.84 0.56 9%

394

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

Table 3. Lengths of restriction fragments of T-RFLP analysis of the different archaeal 16S rDNA clones obtained from methanogenic Lake Dagow sediment (LDS) and afliation to a distinct phylogenetic lineage by 16S rDNA sequence analysis of the clones as shown in Fig. 2

Clones LDS25, 30 LDS7, 8, 11, 17, 26, 36, 37, 44, 46, 50, 51, 52 LDS 10, 27, 32, 48 LDS28, 29, 31, 38, 39, 47, 49 LDS1, 2, 4, 5, 6, 12, 13, 14, 16, 18, 19, 20, 21, 22, 23, 24, 33, 34, 35, 42, 43, 45 LDS3, 9, 15, 40, 41
a

Number of clones 2 12 4 7 22 5

Restriction fragments (bp) 63 75 91 283 392 >700

Phylogenetic lineage LDS cluster (novel Euryarchaeota) Rice cluster V LDS cluster (novel Euryarchaeota) Methanosaetaceae Methanomicrobiaceae Group IIIa (Thermoplasma and relatives)

As termed by Jurgens et al. [30].

quenced. The sequences were phylogenetically analyzed by tree construction (Fig. 2). T-RFLP analysis resulted in a characteristic fragment length for each clone. The clones fell into the following major groups with characteristic T-RF (Table 3): a novel cluster of Euryarchaeota, tentatively called LDS (after Lake Dagow sediment) cluster; Rice cluster V; Methanosaetaceae; Methanomicrobiaceae; and Group III (Thermoplasma and relatives [30]). Some additional minor T-RF were detected at 83, 172, 676, and 686 bp, which were not represented in the clone library, and therefore were not assigned. Also included in the phylogenetic tree (Fig. 2) were sequences of the 16S rDNA obtained from Lake Plusee sediment. Eleven of the clones were related to those obtained from the Lake Dagow sediment and clustered with Group III (Thermoplasma and relatives), Rice cluster V, and LDS cluster. One clone (PSS10) was related to Desulfurococcus mobilis, Sulfolobus solfataricus, and Thermolum pendens (Fig. 2). The sediment samples (taken in 2001) of Lake Dagow that had been incubated at different temperatures were analyzed by T-RFLP. The relative frequency of the individual T-RF among the total archaeal community was determined from their relative peak areas (Fig. 3). The T-RF characteristic for the Methanosaetaceae (283 bp; <27%), the Methanomicrobiaceae (392 bp; <52%), and the Group III (>700 bp; <23%) were the most dominant in the sediment at all temperatures (Fig. 3). The same pattern was obtained in fresh sediment that had been stored frozen. The pattern was also similar in sediment samples from the year 2000, which had been used for preliminary experiments at 4, 10, 15, and 30C (data not shown). The different incubation temperatures did not cause a systematic trend in the relative frequency of the individual T-RF. However, there may have been trends within the different phylogenetic groups that could not be detected with the primers used in this study.

Discussion

Our study showed that the methanogenic pathway in Lake Dagow sediment was dominated by acetoclastic methanogenesis at low temperatures. The contribution of turnover of acetate relative to other fatty acids and of H2dependent methanogenesis increased when temperature was increased. Thus, our results are in agreement with the hypothesis of Schulz and Conrad [40] that the pathway of organic matter degradation changes with increase of temperature from domination by homoacetogenesis and acetoclastic methanogenesis to increasing contribution of other fermentation pathways and H2-dependent methanogenesis. Since the in situ temperature at the sediment surface is always <10C, acetoclastic methanogenesis is most probably the dominant pathway in this lake. The measured methanogenic activity was in the range (about 16 nmol d)1 cm)3) as formerly described for in situ conditions [8, 9]. In the absence of inorganic electron acceptors other than CO2, methanogenesis is the last reaction in a chain of processes that ultimately result in the degradation of organic matter to CH4 and CO2 [42, 49]. The degradation starts with the hydrolysis of organic matter followed by the fermentation of monomers to various alcohols and fatty acids including homoacetogenic fermentation. The alcohols and fatty acids are further degraded to acetate and H2 by the syntrophic bacteria that are thus called, since they rely on the activity of methanogenic archaea that utilize acetate and H2 and thus keep the concentrations of these compounds, H2 in particular, at a level that is thermodynamically permissive. At steady state, the reactions involving production and consumption of H2 seem to be under thermodynamic homeostasis irrespective of temperature [24, 28, 47]. However, the relative contribution of H2 within the overall degradation process seems to increase with temperature [24]. The key for this increase is most likely the total production of H2

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

395

Figure 2. Phylogenetic tree (maximum likelihood tree) showing 16S rDNA sequences of the clones obtained from sediments of Lake Dagow (LDS clones shown in bold) and Lake Plusee (PSS clones shown in bold) in relation to known sequences of Euryarchaeota and Crenarchaeaota, including environmental sequences (given by accession numbers) retrieved from Italian rice eld soil and rice roots (ABS3: AJ227951; ABS9: AJ227954; ABS12: AJ227955; ABS23: AJ227959; ARR16: AJ227929; S307: AJ236515; ST116: AJ236467) and other environmental sequences (VAL2: AJ131264; VAL47: AJ131268; Eel-TA1e6: AF134389; WHARN: M88078; WCHD330: AF050612). The 16S rDNA sequence of Aquifex pyrophilus was used as outgroup reference. The scale bar indicates the estimated number of base changes per nucleotide sequence position.

increasing with temperature, i.e., fermentation reactions other than homoacetogenesis becoming more important with increasing temperature. This assumption is reason-

able since the entropy change of homoacetogenic glucose fermentation (no H2 production; reaction 7, Table 4) is much smaller than that of other glucose fermentation

396

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

Figure 3. Relative abundance of T-RF (number of base pairs) used as a measure of the composition of the archaeal microbial community in sediment samples incubated anaerobically at different temperatures. For assignment of the individual T-RF to phylogenetic groups, see Table 3. T-RF that could not be assigned are summarized as diverse. The gene frequencies were calculated from parallels using the areas of all detected fragments (bars = SE, n = 3).

reactions that do produce H2 (Table 4). Consequently, the energetic advantage of homoacetogenesis relatively to other fermentation reactions decreases when temperature increases. At steady-state conditions, hydrolysis of organic matter followed by fermentation should be the crucial rate-limiting step [4, 21]. The low concentrations of H2, acetate, and other fatty acids measured in Lake Dagow sediments incubated at different temperatures indeed indicate limitation of overall degradation and of CH4 production by hydrolysis and fermentation. It is

noteworthy that apparent steady-state conditions were reached at all the incubation temperatures varying between 4 and 30C, although in situ sediment temperatures never reach values >10C. This observation shows a surprisingly high adaptability of the resident microbiota. The steady-state H2 partial pressures increased with increasing temperature, similarly as observed before in microbial cultures [20] and natural environments [24, 28, 47]. However, the H2 partial pressures were rather low. At low temperatures, they were so low that the DG cal-

Table 4. Gibbs free energy, enthalpy and entropy changes of various glucose fermentation reactions under standard conditions per

mol glucose Reaction (1) (2) (3) (4) (5) (6) (7) C6H12O6 2 CH3CHOHCOOH C6H12O6 2 CH3CH2OH + 2 CO2 1.5 C6H12O6 2 CH3CH2COOH + CH3COOH + CO2 + H2O C6H12O6 CH3CH2CH2COOH + 2 CO2 + 2 H2 3 C6H12O6 + 2 H2O 2 CH3CH2CH2COOH + CH3COOH + 6 CO2 + 8 H2 C6H12O6 + 2 H2O 2 CH3COOH + 2 CO2 + 4 H2 C6H12O6 3 CH3COOH DG0 (kJ mol)1) )118.4 )235.0 )231.6 )224.1 )194.9 )136.4 )191.6 DH0 (kJ mol)1) )79.9 )90.6 )188.2 )49.8 )4.8 +85.0 )186.0 DS0 (kJ mol)1 C)1) 0.129 0.484 0.146 0.585 0.638 0.743 0.019

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

397

culated for H2-dependent methanogenesis was positive. As shown by conversion of 14CO2 to 14CH4, H2/CO2dependent methanogenesis was nevertheless operating. Therefore, we have to assume microenvironments in which H2 concentrations were sufciently large for exergonic methanogenesis, presumably microbial consortia [17]. Acetate concentrations may also have been compartmentalized, in a manner similar to that described for other environments [14, 39], thus explaining why acetate turnover rates were lower than expected from H2dependent and total CH4 production. The development of microenvironments could also result from the different age of the sampled bulk sediment, consisting of freshly sedimented material from the surface to material of about 10 years age at 10 cm depth. Nevertheless, acetate turnover accounted for a higher percentage of CH4 production at 4 than at 30C (Table 2) and thus showed a similar trend with temperature as seen in the accumulation of intermediates after inhibition of methanogenesis (Table 1). The latter experiment should not have been affected by compartmentalization. The experiments with radioactive bicarbonate, which showed the opposite trend, i.e., higher contribution of H2/CO2 to CH4 production at 30 than at 4C, should also not have been affected. Another important objective of our study was to follow the relationship between the structure of the methanogenic community and its function. The relative abundance of the different archaeal groups was quantied by T-RFLP analysis. The individual T-RF were assigned to phylogenetic groups after T-RFLP analysis of clones retrieved from the same sediment. It should be noted that the assignment of T-RF in Lake Dagow sediment, the 392 bp-long fragment in particular, gave a slightly different result than that in anoxic rice eld soil [13]. Hence, it is important to use clones from the same environment for the assignment of T-RF. The three most important phylogenetic groups of archaea in Lake Dagow sediment were the Methanomicrobiaceae (392 bp; <57%), Methanosaetaceae (283 bp; <27%), and euryarchaeotal group III (>700 bp; <23%). Additional cloning and T-RFLP data (Chan and Casper, unpublished) conrm this observation. Methanosaetaceae are acetoclastic methanogens, in contrast to Methanosarcinaceae, which have the ability to utilize acetate at low concentrations [29]. Methanosarcinaceae were not detected in Lake Dagow sediments. Probably, acetate concentrations were too low to support populations of this group. In rice eld soil, for example, the relative abundance of Methanosarcinaceae decreased relative to that of Methanosaetaceae with decreasing acetate concentrations [24]. Methanosaetaceae, but not Methanosarcinaceae were also abundant in Lake Rotsee sediment [50]. On decaying rice straw [46] and rice roots [12], on the other hand, where acetate reaches millimolar concentrations,

Methanosarcinaceae were the dominant acetotrophic methanogens. We detected 16S rRNA genes of Methanomicrobiaceae related to Methanospirillum hungatei, to Methanoculleus spp., and to symbiontic methanogens. All these species are hydrogenotrophic methanogens. Hence, hydrogenotrophic methanogens were indeed present in the sediment, thus being consistent with the methanogenic activities detected. Interestingly, other hydrogenotrophic methanogenic groups, e.g., Methanobacteriaceae, were not detected in sediment of Lake Dagow, either by cloning or by T-RFLP analysis. Other methanogenic lake ecosystems also contained mainly Methanomicrobiaceae, but rarely Methanobacteriaceae [10, 26, 30, 36, 50]. The phenotypes of the other archaeal groups detected in Lake Dagow sediment remain elusive. Euryarchaeotal group III and Rice cluster V, whose closest relatives are members of the Thermoplasmales, have been detected in numerous aquatic ecosystems (see [30]). Together they constituted almost 25% of the total archaeal 16S rRNA gene frequency. However, their function remains unknown. The LDS cluster is also related to these groups. This cluster has so far not been described. Only one type of clone (WCHD330; originating from an aquifer) belonging to this cluster has been reported in the databases [22]. Members of the LDS cluster were not uncommon in the 16S rDNA clone library created from sediment of Lake Dagow and constituted <5% of the total archaeal 16S rRNA gene frequency (Fig. 3). In order to check whether this new cluster also occurs in other lake sediments, a clone library obtained from Lake Plusee (>200 km away from Lake Dagow) was tested. Indeed, genes belonging to the LDS cluster were also retrieved from Lake Plusee. Hence, this euryarchaeotal cluster may be widespread in lake sediments. Again, however, the functioning of this group is unknown. As hydrogenotrophic and acetotrophic methanogenic reactions were found to be controlled by the supply of substrates originating from the preceding fermentation reactions (see above), the community structure of the methanogens may have been inuenced by this supply. As acetotrophic versus hydrogenotrophic methanogenesis decreased with increasing temperature, the acetotrophic methanogens might have decreased while the hydrogenotrophic ones might have increased. For example, after incubation of rice eld soil at temperatures higher than 40C, CH4 was exclusively produced from H2/CO2 and hydrogenotrophic methanogens prevailed [23]. However, such a shift was not observed in Lake Dagow sediment. Apparently, the change in the methanogenic pathway was not paralleled by a change of the community structure of the major phylogenetic groups (and guilds) of methanogenic archaea. This does not exclude the possibility that more delicate changes oc-

398

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

curred on a lower phylogenetic level that was not discerned by the T-RFLP analysis. Nevertheless, adaptation of the methanogens seemed to occur mainly on the level of activity change rather than population change. Possibly, temperature adaptation by population change was more relevant for the fermenting bacteria, but this was outside of the scope of this study.
Acknowledgments

16.

17.

18.

We thank B. Nusslein for supplying the sediment samples of Lake Plusee.

19.

References
1. Avery, GB, Martens, CS (1999) Controls on the stable carbon isotopic composition of biogenic methane produced in a tidal freshwater estuarine sediment. Geochim Cosmochim Acta 63: 10751082 2. Avery, GB, Shannon, RD, White, JR, Martens, CS, Alperin, MJ (2003) Controls on methane production in a tidal freshwater estuary and a peatland: methane production via acetate fermentation and CO2 reduction. Biogeochemistry 62: 1937 3. Barns, SM, Delwiche, CF, Palmer, JD, Pace, NR (1996) Perspectives on archaeal diversity, thermophily and monophyly from environmental rRNA sequences. Proc Natl Acad Sci USA 93: 9188 9193 4. Billen, G (1982) Modelling the processes of organic matter degradation and nutrients recycling in sedimentary systems. In: Nedwell, DB, Brown, CM (Eds.) Sediment Microbiology, Academic Press, New York, pp 15521982 5. Brosius, J, Palmer, ML, Kennedy, PJ, Noller, HR (1978) Complete nucleotide sequence of a 16S ribosomal gene from Escherichia coli. Proc Natl Acad Sci USA 75: 48014805 6. Capone, DG, Kiene, RP (1988) Comparison of microbial dynamics in marine and freshwater sediments: contrasts in anaerobic carbon catabolism. Limnol Oceanogr 33: 725749 7. Cappenberg, T, Prins, RA (1974) Interrelations between sulfatereducing and methane-producing bacteria in bottom deposits of a fresh-water lake. III. Experiments with 14C-labeled substrates. Antonie van Leeuwenhoek 40: 457469 8. Casper, P (1992) Methane production in lakes of different trophic state. Arch Hydrobiol Beih Ergebn Limnol 37: 149154 9. Casper, P (1996) Methane production in littoral and profundal sediments of an oligotrophic and a eutrophic lake. Arch Hydrobiol Beih Ergebn Limnol 48: 253259 10. Chan, OC, Wolf, M, Hepperle, D, Casper, P (2002) Methanogenic archaeal community in the sediment of an articially partitioned acidic bog lake. FEMS Microbiol Ecol 42: 119129 11. Chin, KJ, Conrad, R (1995) Intermediary metabolism in methanogenic paddy soil and the inuence of temperature. FEMS Microbiol Ecol 18: 85102 12. Chin, KJ, Lueders, T, Friedrich, MW, Klose, M, Conrad, R (2004) Archaeal community structure and pathway of methane formation on rice roots. Microb Ecol 47: 5967 13. Chin, KJ, Lukow, T, Conrad, R (1999) Effect of temperature on structure and function of the methanogenic archaeal community in an anoxic rice eld soil. Appl Environ Microbiol 65: 2341 2349 14. Christensen, D, Blackburn, TH (1982) Turnover of 14C-labelled acetate in marine sediments. Mar Biol 71: 113119 15. Conrad, R (1999) Contribution of hydrogen to methane production and control of hydrogen concentrations in methano20.

21. 22.

23. 24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

genic soils and sediments [review]. FEMS Microbiol Ecol 28: 193202 Conrad, R, Bak, F, Seitz, HJ, Thebrath, B, Mayer, HP, Schutz, H (1989) Hydrogen turnover by psychrotrophic homoacetogenic and mesophilic methanogenic bacteria in anoxic paddy soil and lake sediment. FEMS Microbiol Ecol 62: 285294 Conrad, R, Phelps, TJ, Zeikus, JG (1985) Gas metabolism evidence in support of juxtapositioning between hydrogen producing and methanogenic bacteria in sewage sludge and lake sediments. Appl Environ Microbiol 50: 595601 Conrad, R, Schink, B, Phelps, TJ (1986) Thermodynamics of H2producing and H2-consuming metabolic reactions in diverse methanogenic environments under in-situ conditions. FEMS Microbiol Ecol 38: 353360 Conrad, R, Schutz, H (1988) Methods of studying methanogenic bacteria and methanogenic activities in aquatic environments. In: Austin, B (Ed.) Methods in Aquatic Bacteriology, Wiley, Chichester, UK, pp 301343 Conrad, R, Wetter, B (1990) Inuence of temperature on energetics of hydrogen metabolism in homoacetogenic, methanogenic, and other anaerobic bacteria. Arch Microbiol 155: 9498 Degens, ET, Mopper, K (1975) Early diagenesis of organic matter in marine salts. Soil Sci 119: 6572 Dojka, MA, Hugenholtz, P, Haack, SK, Pace, NR (1998) Microbial diversity in a hydrocarbon- and chlorinated-solvent-contaminated aquifer undergoing intrinsic bioremediation. Appl Environ Microbiol 64: 38693877 Fey, A, Chin, KJ, Conrad, R (2001) Thermophilic methanogens in rice eld soil. Environ Microbiol 3: 295303 Fey, A, Conrad, R (2000) Effect of temperature on carbon and electron ow and on the archaeal community in methanogenic rice eld soil. Appl Environ Microbiol 66: 47904797 Glissmann, K, Conrad, R (2000) Fermentation pattern of methanogenic degradation of rice straw in anoxic paddy soil. FEMS Microbiol Ecol 31: 117126 Go, YS, Han, SK, Lee, IG, Ahn, TY (2000) Diversity of the domain Archaea as determined by 16S rRNA gene analysis in the sediments of Lake Soyang. Arch Hydrobiol 149: 459466 Grosskopf, R, Janssen, PH, Liesack, W (1998) Diversity and structure of the methanogenic community in anoxic rice paddy soil microcosms as examined by cultivation and direct 16S rRNA gene sequence retrieval. Appl Environ Microbiol 64: 960 969 Hoehler, TM, Alperin, MJ, Albert, DB, Martens, CS (1998) Thermodynamic control on hydrogen concentrations in anoxic sediments. Geochim Cosmochim Acta 62: 17451756 Jetten, MSM, Stams, AJM, Zehnder, AJB (1990) Acetate threshold and acetate activating enzymes in methanogenic bacteria. FEMS Microbiol Ecol 73: 339344 Jurgens, G, Glockner, F, Amann, R, Saano, A, Montonen, L, Likolammi, M, Munster, U (2000) Identication of novel Archaea in bacterioplankton of a boreal forest lake by phylogenetic analysis and uorescent in situ hybridization. FEMS Microbiol Ecol 34: 45 56 Kotsyurbenko, OR, Nozhevnikova, AN, Zavarzin, GA (1993) Methanogenic degradation of organic matter by anaerobic bacteria at low temperature. Chemosphere 27: 17451761 Kuivila, KM, Murray, JW, Devol, AH (1989) Methane production, sulfate reduction and competition for substrates in the sediments of Lake Washington. Geochim Cosmochim Acta 53: 409416 Liu, WT, Marsh, TL, Cheng, H, Forney, LJ (1997) Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Appl Environ Microbiol 63: 45164522

K. GLISSMANN

ET AL.:

METHANOGENIC PATHWAY

AND

ARCHAEAL COMMUNITY STRUCTURE

399

34. MacGregor, BJ, Moser, DP, Alm, EW, Nealson, KH, Stahl, DA (1997) Crenarchaeota in Lake Michigan sediment. Appl Environ Microbiol 63: 11781181 35. Nozhevnikova, AN, Kotsyurbenko, OR, Simankova, MV (1994) Acetogenesis at low temperature. In: Drake, HL (Ed.) Acetogenesis, Chapman & Hall, New York, pp 416431 36. Nusslein, B, Chin, KJ, Eckert, W, Conrad, R (2001) Evidence for anaerobic syntrophic acetate oxidation during methane production in the profundal sediment of subtropical Lake Kinneret (Israel). Environ Microbiol 3: 460470 37. Nusslein, B, Conrad, R (2000) Methane production in eutrophic Lake Plusssee: seasonal change, temperature effect and metabolic processes in the profundal sediment. Arch Hydrobiol 149: 597 623 38. Nusslein, B, Eckert, W, Conrad, R (2003) Stable isotope biogeochemistry of methane formation in profundal sediments of Lake Kinneret (Israel). Limnol Oceanogr 48: 14391446 39. Parkes, RJ, Taylor, J, Joerck-Ramberg, D (1984) Demonstration, using Desulfobacter sp., of two pools of acetate with different biological availabilities in marine pore water. Mar Biol 83: 271 276 40. Schulz, S, Conrad, R (1996) Inuence of temperature on pathways to methane production in the permanently cold profundal sediment of Lake Constance. FEMS Microbiol Ecol 20: 114 41. Smalla, K, Cresswell, N, Mendonca-Hagler, LC, Wolters, A, van Elsas, JD (1993) Rapid DNA extraction protocol from soil for polymerase chain reaction-mediated amplication. J Appl Bacteriol 74: 7885

42. Stams, AJM (1994) Metabolic interactions between anaerobic bacteria in methanogenic environments. Antonie van Leeuwenhoek 66: 271294 43. Strunk, O, Ludwig, W (2000) ARB Software Package Version 2.5b. Technische Universitat Munchen, Germany 44. Thamdrup, B, Caneld, DE (2000) Benthic respiration in aquatic sediments. In: Sala, OE, Jackson, RB, Mooney, HA, Howarth, RW (Eds.) Methods in Ecosystem Science, Springer, New York, pp 86 103 45. Thauer, RK, Jungermann, K, Decker, K (1977) Energy conservation in chemotrophic anaerobic bacteria. Bacteriol Rev 41: 100 180 46. Weber, S, Lueders, T, Friedrich, MW, Conrad, R (2001) Methanogenic populations involved in the degradation of rice straw in anoxic paddy soil. FEMS Microbiol Ecol 38: 1120 47. Westermann, P (1994) The effect of incubation temperature on steady-state concentrations of hydrogen and volatile fatty acids during anaerobic degradation in slurries from wetland sediments. FEMS Microbiol Ecol 13: 295302 48. Woese, CR, Kandler, O, Wheelis, ML (1990) Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci USA 87: 45764579 49. Zehnder, AJB (1978) Ecology of methane formation. In: Mitchell, R (Ed.) Water Pollution Microbiology, Wiley, New York, pp 349 376 50. Zepp Falz, K, Holliger, C, Grosskopf, R, Liesack, W, Nozhevnik ova, AN, Muller, B, Wehrli, B, Hahn, D (1999) Vertical distribution of methanogens in the anoxic sediment of Rotsee (Switzerland). Appl Environ Microbiol 65: 24022408