Effect of Substrate Concentration on Carbon Isotope Fractionation during Acetoclastic Methanogenesis by Methanosarcina barkeri and M. acetivorans and in Rice Field Soil

Abstract

Methanosarcina is the only acetate-consuming genus of methanogenic archaea other than Methanosaeta and thus is important in methanogenic environments for the formation of the greenhouse gases methane and carbon dioxide. However, little is known about isotopic discrimination during acetoclastic CH₄ production. Therefore, we studied two species of the Methanosarcinaceae family, Methanosarcina barkeri and Methanosarcina acetivorans, and a methanogenic rice field soil amended with acetate. The values of the isotope enrichment factor (ɛ) associated with consumption of total acetate (ɛ_ac), consumption of acetate-methyl (ɛ_ac-methyl) and production of CH₄ (ɛ_CH4) were an ɛ_ac of −30.5‰, an ɛ_ac-methyl of −25.6‰, and an ɛ_CH4 of −27.4‰ for M. barkeri and an ɛ_ac of −35.3‰, an ɛ_ac-methyl of −24.8‰, and an ɛ_CH4 of −23.8‰ for M. acetivorans. Terminal restriction fragment length polymorphism of archaeal 16S rRNA genes indicated that acetoclastic methanogenic populations in rice field soil were dominated by Methanosarcina spp. Isotope fractionation determined during acetoclastic methanogenesis in rice field soil resulted in an ɛ_ac of −18.7‰, an ɛ_ac-methyl of −16.9‰, and an ɛ_CH4 of −20.8‰. However, in rice field soil as well as in the pure cultures, values of ɛ_ac and ɛ_ac-methyl decreased as acetate concentrations decreased, eventually approaching zero. Thus, isotope fractionation of acetate carbon was apparently affected by substrate concentration. The ɛ values determined in pure cultures were consistent with those in rice field soil if the concentration of acetate was taken into account.

Methane (CH₄) is the most abundant reduced gas in the earth's atmosphere and is an important greenhouse gas with a high global-warming potential (7). It is presently a matter of discussion whether the contribution of CH₄ to the greenhouse effect will increase in the future (3, 23). This has made it necessary and more urgent to understand natural processes that lead to the production of CH₄.

Methanogenesis, the microbial formation of CH₄, is the final step in the degradation of organic matter in anoxic environments like natural wetlands, lake sediments, and flooded rice fields. The most important precursors for the production of CH₄ are acetate (equation 1) and CO₂ (equation 2) with the following reactions (8):

(1)

(2)

Acetate is the most important substrate since it contributes more than 67% to microbial methanogenesis during anoxic degradation of polysaccharides. In methanogenic environments only two genera of archaea, Methanosaeta and Methanosarcina, are capable of using acetate (2). While Methanosaeta can be considered a specialist that uses only acetate, Methanosarcina can use a wide range of substrates besides acetate, for example, H₂/CO₂, methanol, methylamines, and methylated sulfides. Among methanogens, Methanosarcinaceae also display the largest environmental distribution. They can be found in freshwater sediments and soil, marine habitats, landfills, and animal gastrointestinal tracts (46).

Additionally, differences between Methanosarcina and Methanosaeta were found for isotope fractionation of stable carbon. The fractionation factor (α) or, equivalently, the enrichment factor (ɛ) during acetoclastic methanogenesis in Methanosarcina barkeri strains typically ranges from an α of 1.021 to 1.027 or an ɛ of −27‰ to −21‰ (14, 27, 48), whereas isotope fractionation in Methanosaeta spp. is weaker, i.e., an α of 1.007 (ɛ = −7‰) for Methanosaeta thermophila (43) and an α of 1.010 (ɛ = −10‰) for Methanosaeta concilii (34). It is suggested that the two archaeal genera differ in isotope fractionation due to differences in their biochemical activation of acetate to acetyl-coenzyme A (acetyl-CoA) (34). However, isotopic data for acetoclastic methanogens are rare. For instance, all data for Methanosarcina refer to only one species, namely M. barkeri.

Hence, in this study we investigated whether differences in carbon isotope fractionation within the genus Methanosarcina occur. Therefore, we determined isotope ratios of stable carbon in cultures of the acetoclastic species M. barkeri and Methanosarcina acetivorans. Second, we were interested if these data, obtained from pure cultures, could also be applied to understand natural environments. For that reason, we determined isotope fractionation during acetoclastic methanogenesis in the model system rice field soil. Furthermore, we discuss the effect of substrate concentration on carbon isotope fractionation and the importance of monitoring isotope fractionation during the course of acetate consumption.

MATERIALS AND METHODS

Growth conditions and soil incubations.

M. barkeri (DSM 804) and M. acetivorans (DSM 2834) were obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (Braunschweig, Germany). Both species were grown under N₂/CO₂ (80:20) as suspensions of single cells (40) in high-salt medium (31) with 20 mM acetate as an energy and carbon source. M. barkeri and M. acetivorans were incubated in glass bottles (500 ml; Ochs, Bovenden-Lenglern, Germany), without shaking, at 30°C and 37°C, respectively. For experiments cultures (10% inocula) in the late exponential phase were transferred, resulting in a final volume of 250 ml. Samples from the headspace and the liquid phase were removed to determine pH, solute concentrations, and the carbon isotope compositions of acetate, methane, and carbon dioxide. All experiments were performed in triplicate.

Rice field soil was collected in 2006 from rice paddies of the Italian Rice Research Institute near Vercelli in the valley of the River Po, Italy. The characteristics of this site were described by Schütz et al. (37, 38). The soil was air dried and stored in polyethylene vats at room temperature. Afterwards, residues of straw and roots were coarsely ground using a jaw crusher (Retsch and Dietz-Motoren GmbH & Co. KG, Dettingen unter Teck, Germany), and the rice field soil was sieved (≤1-mm mesh size). For experiments, rice field slurries (soil and demineralized water, 1:1 [wt/wt]) amended with rice straw (1 g per kg of slurry; coarsely ground with an A11 Basic Analytical Mill [IKA Werke, Staufen, Germany]) were preincubated at 25°C for at least 4 weeks, when electron acceptors such as available iron, sulfate, or nitrate were completely depleted. The rice slurry was then distributed into 27-ml pressure tubes (Ochs, Bovenden-Lenglern, Germany); each tube was filled with 10 g of slurry and sealed with a butyl rubber stopper. Thereafter, the incubation vessels were repeatedly flushed and evacuated with N₂ for 10 min, and a final overpressure of 0.5 × 10⁵ Pa was adjusted inside the tubes. Then, 5 mM acetate was added under sterile conditions, and the incubation was started at 25°C in the dark. At each sampling day three tubes were harvested to determine the concentrations and isotope ratios of substrates and products and for DNA extraction.

Extraction of DNA and PCR amplification of 16S rRNA genes.

DNA extraction from rice field soil was performed using a FastDNA spin kit for soil (Qbiogene, Heidelberg, Germany), according to the manufacturer's instructions, and an additional treatment with guanidine thiocyanate to remove humic acids (21).

Archaeal 16S rRNA genes were amplified using the forward primer A109f (5′-ACKGCTCAGTAACACGT-3′) (17) and the 5-carboxyfluorescein-labeled (5′-terminal) backward primer A915b (5′-GTGCTCCCCCGCCAATTCCT-3′) (41). In a total volume of 50 μl, the PCR mixture contained 10 μl of 5× Green Go Taq Flexi buffer (Promega, Hilden, Germany), 1 U of Taq DNA polymerase (Invitrogen GmbH, Karlsruhe, Germany), a 200 μM concentration of each deoxynucleoside triphosphate (Amersham Pharmacia Biotech, Freiburg, Germany), 1.5 mM MgCl₂ (Promega), 10 μg of bovine serum albumin (Roche, Mannheim, Germany), and a 3.3 nM concentration of each primer (Sigma-Aldrich, Taufkirchen, Germany). A volume of 1 μl of DNA solution was added as a template. The amplification was performed using a Primus cycler (MWG Biotech, Ebersberg, Germany) with an initial denaturation step (3 min at 94°C), followed by 29 cycles of denaturation (45 s at 94°C), annealing (45 s at 52°C), and elongation (90 s at 72°C); a terminal elongation step was performed for 5 min at 72°C.

T-RFLP analysis.

The principle of terminal restriction fragment length polymorphism (T-RFLP) analysis of archaeal 16S rRNA has been described by Chin et al. (6). Gel electrophoresis was carried out as a visual control for a successful amplification of 16S rRNA genes. Afterwards, fluorescently labeled 16S rRNA gene amplicons were purified by the use of a GenElute PCR Clean-up Kit (Sigma-Aldrich) according to the instructions of the manufacturer. DNA concentrations of purified 16S rRNA gene fragments were determined by standard UV photometry (Biophotometer; Eppendorf, Hamburg, Germany) at 260 nm. Restriction digestion was performed in a total volume of 10 μl containing approximately 80 ng of 16S rRNA gene amplicons. The latter were restricted with 5 U of enzyme TaqI (Fermentas, St. Leon-Rot, Germany) and 1 μl of the appropriate incubation buffer and incubated for 3 h at 65°C. The restriction digestion was purified using a postreaction clean-up spin column kit (Sigma-Aldrich) according to the manufacturer's instructions. To prepare the samples for the T-RFLP analysis, 3 μl of the purified restriction digestions were mixed with 0.3 μl of an internal lane standard (MapMarker 1000; 50 to 1,000 bp; X-rhodamine labeled [BioVentures Inc.]) and 11 μl of HiDi formamide (Applied Biosystems, Weiterstadt, Germany) and denatured for 3 min at 95°C. The analysis of the digested amplicons was performed by separation using capillary electrophoresis with an automatic sequencer (3130 Genetic Analyzer; Applied Biosystems) for 50 min at 15 kV and 9 μA. The injection time per sample was 6 s. After capillary electrophoresis, the lengths of the fluorescently labeled terminal restriction fragments (T-RFs) were identified by comparison to the internal standard using the GeneMapper software (version 4.0; Applied Biosystems). The relative abundance of a detected T-RF within a given T-RFLP pattern was calculated as the respective signal area of the peak divided by the peak area of all peaks of the T-RFLP pattern, starting from a fragment size of 50 bp to exclude T-RFs caused by primers.

Chemical and isotopic analysis.

CH₄ and CO₂ were analyzed by gas chromatography (GC) using a flame ionization detector (Shimadzu, Kyoto, Japan). CO₂ was detected after conversion to CH₄ with a methanizer (Ni-catalyst at 350°C; Chrompack, Middelburg, The Netherlands).

Stable isotope analysis of ¹³C/¹²C in gas samples was performed using a GC combustion isotope ratio mass spectrometer (GC-C-IRMS) system that was purchased from Finnigan (Thermo Fisher Scientific, Bremen, Germany). The principle operation was described by Brand (4). The CH₄ and CO₂ in the gas samples (30 to 400 μl) were first separated in a Hewlett Packard 6890 GC using a Pora Plot Q column (27.5-m length, 0.32-mm internal diameter, and 10-μm film thickness; Chrompack, Frankfurt, Germany) at 30°C and He (99.996% purity; 2.6 ml/min) as the carrier gas. After conversion of CH₄ to CO₂ in the Finnigan Standard GC Combustion Interface III, the isotope ratio of ¹³C/¹²C was analyzed in the IRMS (Finnigan MAT Delta^plus). The isotope reference gas was CO₂ (99.998% purity) (Air Liquide, Düsseldorf, Germany), calibrated with the working standard methylstearate (Merck). The latter was intercalibrated at the Max Planck Institute for Biogeochemistry, Jena, Germany (courtesy of W. A. Brand) against the NBS-22 and USGS-24 standards and reported in the delta notation versus Vienna Pee Dee Belemnite notation:

(3)

with R = ¹³C/¹²C of sample (sa) and standard (st), respectively. The precision of repeated analysis was ± 0.2‰ when 1.3 nmol of CH₄ was injected.

Isotopic measurements and quantification of acetate were performed on a high-performance liquid chromatography (HPLC) system (Spectra System P1000 [Thermo Fisher Scientific, San Jose, CA] and Mistral [Spark, Emmen, The Netherlands], respectively) equipped with an ion-exclusion column (Aminex HPX-87-H; Bio-Rad, Munich, Germany) and coupled to a Finnigan LC IsoLink (Thermo Fisher Scientific, Bremen, Germany), as described previously (26). Isotope ratios were detected on an IRMS (Finnigan MAT Delta^plus Advantage). The isotope reference gas was CO₂ calibrated as described above.

Off-line pyrolysis was performed to determine the δ¹³C of the methyl group of acetate (δ_ac-methyl). Acetate in the liquid sample was purified using HPLC by collecting the acetate fraction from each run. The purified sample was added to a strong NaOH solution and dried in a Pyrex tube under vacuum. The dried reactants were pyrolysed under vacuum at 400°C, converting the carboxyl carbon to CO₂ and the methyl carbon to CH₄ (1). Gas samples were taken and analyzed by GC-C-IRMS as described above.

The analysis of the δ¹³C of biomass was carried out at the Centre for Stable Isotope Research and Analysis at Goettingen University, Goettingen, Germany, with an elemental analyzer (EA)-IRMS system consisting of an EA (NA 2500; CE Instruments, Rodano, Italy) and an IRMS (Finnigan MAT Delta^plus), coupled via an interface (ConFlo III; Thermo Fisher Scientific). The samples and the laboratory reference compound acetanilide were applied as solid samples in tin capsules (IVA, Meerbusch, Germany). The standardization scheme of the EA-IRMS measurements as well as the measurement strategy and the calculations for assigning the final δ¹³C values on the Vienna Pee Dee Belemnite scale were analogous to those described by Werner and Brand (45) on an EA-IRMS. The precision of repeated analysis was ± 0.18‰ when 0.4 to 1.5 mg of acetanilide was injected.

Radiotracer experiments were done to determine the fraction of CH₄ and CO₂ produced from the methyl group of acetate. For that, 10 μCi/ml (0.37 MBq/ml) of Na-[2-¹⁴C]acetate (Amersham, Braunschweig, Germany) was added in a volume of 1.0-ml to 120-ml bottles (Ochs, Bovenden-Lenglern, Germany) which were filled with 50 ml of culture liquid. The origin, specific radioactivity, and the quantity of the added tracer were 2.1 GBq mmol⁻¹ and 3.7 MBq, respectively. The ¹⁴C-labeled acetate was added after CH₄ production was observed. Total and radioactive CH₄ and CO₂ were analyzed in a GC equipped with a flame ionization detector, reduction column, and a RAGA radioactivity detector (9). Total and radioactive acetate were analyzed in the liquid phase in an HPLC system equipped with a refraction index detector and a Ramona radioactivity detector (25). The respiratory index (RI) was determined at the end of the incubation after the addition of 2.0 ml of 2.5 M H₂SO₄ per bottle to liberate CO₂ (CO₂ + bicarbonate):

(4)

Calculations.

Fractionation factors for a reaction A → B are defined after Hayes (19) as

(5)

also expressed as ɛ ☰ 10³ (1 − α). The isotope enrichment factor ɛ associated with acetoclastic methanogenesis was determined as described by Mariotti et al. (30) from the residual reactant, calculated as

(6)

and from the product formed, calculated as

(7)

where δ_ri is the isotope composition of the reactant (either acetate or acetate-methyl) at the beginning, δ_r and δ_p are the isotope compositions of the residual acetate and the pooled CH₄, respectively, at the instant when f was determined, and f is the fractional yield of the products based on the consumption of acetate (0 < f < 1). Linear regression of δ_r against ln(1 − f) and of δ_p against (1 − f)[ln(1 −f)]/f gives ɛ as the slope of best-fit lines. To analyze ɛ of the carboxyl group of acetate (ɛ_ac-carboxyl), values for δ_ac-carboxyl were calculated using the following equation:

(8)

Because total oxidized carbon was distributed among different carbon species (gaseous CO₂, dissolved CO₂, HCO₃⁻, and CO₃²⁻), δ¹³C of total inorganic carbon (δ_TIC) could not be determined directly. This value was calculated by the following mass-balance equation:

(9)

where X is the mole fraction and δ is the isotopic composition of the C of gaseous CO₂ (g), dissolved CO₂ (d), HCO₃⁻ (b), and CO₃²⁻ (c). The distribution of carbon among these species was calculated using solubility and equilibrium constants (42). δ_g was measured directly; the remaining isotopic compositions were calculated from the relevant equilibrium isotope fractionation factors (11, 33):

(10)

(11)

(12)

RESULTS

Methane production by M. barkeri was observed after 9 days of incubation (Fig. 1A) and started immediately after inoculation in M. acetivorans (Fig. 1C). Acetate was consumed completely, leading to an increase in pH. Concentrations of CO₂ are not shown because the high concentrations of bicarbonate in the medium made it impractical to measure yields of this product. During the fermentation the preferred consumption of [¹²C]acetate caused an enrichment of the heavier isotope ¹³C in the remaining acetate (Fig. 1B and D). Likewise, this led to an increasing production of ¹³C-CH₄. The initial high δ¹³C value of CH₄ in M. barkeri resulted from the transfer of dissolved CH₄ during inoculation. CO₂ (illustrated in Fig. 1 as total inorganic carbon [TIC]) became slightly depleted in ¹³C with time but was not used for determination of isotope fractionation since, as mentioned above, the high background of bicarbonate did not allow precise quantification of the δ¹³C of the newly formed TIC. Carbon isotope fractionation during acetoclastic methanogenesis was determined for total acetate (both carbon atoms), acetate-methyl, acetate-carboxyl, and CH₄ using equations 6 and 7, based on Rayleigh distillation (Fig. 2 and Table 1). The isotope enrichment factors for acetate, acetate-methyl, and acetate-carboxyl were calculated for three different phases of acetate consumption: until 50% of acetate was consumed (f values between 0.0 and 0.5), between 50 and 80% consumption (0.5 < f < 0.8), and from 80% to maximum consumption of acetate (0.8 < f < 1.0). As carbon isotope fractionation decreased after the first phase, ɛ values for the range of 0 < f < 0.5 were finally used to determine and compare isotope fractionation (Fig. 2B and E). Isotope fractionation in CH₄ was linear during the complete experiments (Fig. 2C and F). Isotope enrichment in CH₄ (ɛ_CH4) and acetate-methyl (ɛ_ac-methyl) during the initial phase agreed within error since the major part of the methyl group of acetate was converted to CH₄ (see below).

FIG. 1.

Catabolism of acetate in pure cultures of M. barkeri (A and B) and M. acetivorans (C and D). (A and C) Acetate consumption, CH₄ production, and pH. (B and D) Isotope signatures of total acetate, acetate-methyl, CH₄, and CO₂ (illustrated as TIC). ▪, acetate; □, acetate-methyl; •, CH₄; ⋄, TIC; dotted line without symbols, pH. The concentrations are given as mmol per bottle; values are means ± standard errors (n = 3). d, days.

FIG. 2.

Isotope enrichment during acetoclastic methanogenesis by M. barkeri (A to C) and M. acetivorans (D to F). Equations derived by Mariotti et al. (30) have been used to calculate ɛ values from fractional yields and isotope compositions of (total) acetate (▪), acetate-methyl (□), acetate-carboxyl (▵) (A, B, D, and E), and CH₄ (•) (C and F). Panels B and E show magnifications of the boxed segments in panels A and D, respectively. Isotope enrichment in acetate, acetate-methyl, and acetate-carboxyl was calculated for three different phases of acetate consumption (A and D): solid lines show substrate levels between 0 and −0.6 on the ln(1 − f) scale (corresponding to up to 50% acetate consumption), dashed lines show levels between −0.6 and −1.6 (50 to 80%), and dotted lines show levels between −1.6 and −3.0 (from 80% to maximum consumption of acetate). The values are means ± standard errors (n = 3).

TABLE 1.

Isotope enrichment factors for acetate, acetate-methyl, acetate-carboxyl, and CH₄ during acetoclastic methanogenesis by Methanosarcina spp. and in rice field soil^a

Studied system	f	Acetate concn (mM)	ɛac (‰)	ɛac-methyl (‰)	ɛac-carboxyl (‰)b	ɛCH4 (‰)
M. barkeri	0.0-0.5	23.0-11.5	−30.53 ± 1.44	−25.61 ± 2.39	−34.66 ± 3.45	−27.40 ± 0.37c
	0.5-0.8	11.5-4.6	−17.00 ± 0.34	−11.10 ± 4.25	−22.90 ± 3.29
	0.8-1.0	4.6-0	−9.57 ± 0.63	5.60 ± 0.72	−26.10 ± 1.95
M. acetivorans	0.0-0.5	21.0-0.5	−35.29 ± 1.04	−24.77 ± 0.90	−47.74 ± 2.65	−23.81 ± 0.69c
	0.5-0.8	10.5-4.2	−25.70 ± 0.09	−14.40 ± 2.66	−37.00 ± 2.83
	0.8-1.0	4.2-0	−13.69 ± 0.80	13.56 ± 4.64	−40.77 ± 3.54
Rice field soil	0.0-0.5	5.3-2.6	−18.70 ± 0.43	−16.90 ± 1.95	−20.85 ± 1.02	−22.12 ± 0.64d
	0.5-0.8	2.6-1.1	−7.21 ± 0.31	−2.41 ± 0.59	−12.32 ± 1.59
	0.8-1.0	1.1-0	−2.01	5.47	−9.5	−3.32 ± 1.06

^aValues are means of triplicates.

^bδ_ac-carboxyl = 2δ_ac − δ_ac-methyl.

^cMariotti plots showed no significant changes during acetate consumption over time.

^dThe value of ɛ was calculated for the range of 0.0 < f < 0.8 due to a smaller amount of data.

Isotope signatures of total acetate, acetate-methyl, CH₄, and CO₂ observed during the catabolism of acetate in M. barkeri and M. acetivorans followed similar trends. In both methanogens the continuous preferential consumption of [¹²C]acetate caused an enrichment of ¹³C in the remaining acetate and in CH₄, as expected for a closed system. Nevertheless, minor differences between the two archaeal species occurred in isotope fractionation of stable carbon (Table 1). The fractionation of acetate (ɛ_ac) and acetate-carboxyl (ɛ_ac-carboxyl) was stronger in M. acetivorans than in M. barkeri by 4.8‰ and 13.0‰, respectively, and weaker for CH₄ (ɛ_CH4) by 3.6‰ during the first phase of acetate consumption (0 < f < 0.5). Isotope enrichment in acetate-methyl (ɛ_ac-methyl) was identical for both cultures.

Also the δ¹³C of the biomass was determined at the end of the experiments and was found to be slightly ¹³C enriched in both M. barkeri (δ¹³C_biomass = −19.1‰ ± 0.1‰) and M. acetivorans (δ¹³C_biomass = −18.5‰± 0.8‰), compared to the initial δ¹³C_ac values of total acetate (−25.9‰ ± 0.1‰ and −25.7‰ ± 0.3‰, respectively). Additionally, radiotracer experiments with [2-¹⁴C]acetate were carried out to determine the fraction of CH₄ and CO₂ produced from acetate-methyl. Confirming literature data (44), incubations with radiolabeled acetate resulted in the production of mostly ¹⁴CH₄ for both Methanosarcina strains. Nevertheless, ¹⁴CO₂ was also produced from acetate-methyl. For both strains the calculated RI was 0.11, according to a ¹⁴CO₂ production of 11%.

In addition, we determined carbon isotope fractionation in the anoxic model system rice field soil to investigate whether fractionation factors obtained from pure culture studies may also be used for environmental systems. Incubations with paddy soil, which were amended with acetate, showed a complete degradation of acetate. Simultaneously, CH₄ was produced (Fig. 3), indicating an active population of acetoclastic methanogens. Their presence was confirmed by T-RFLP analysis targeting archaeal 16S rRNA genes. The relative amplicon frequencies of T-RFs assigned according to Chin et al. (5) on incubation days 0, 6, and 10 were highly similar. The most abundant T-RF was found at 185 bp, representing Methanosarcinaceae, which accounted for more than 60% of the total archaeal community. The other archaeal family capable of utilizing acetate, Methanosaetaceae, was detected only at the beginning of the incubation with a low frequency of 1% (284-bp T-RF). Other abundant T-RFs were related to hydrogenotrophic methanogens, namely, Methanomicrobiaceae at 83 bp, Methanobacteriaceae at 91 bp, and Rice Cluster I (recently described as Methanocellales [36]) at 392 bp. As in the cultures of acetoclastic M. barkeri and M. acetivorans, acetate consumption in rice field soil also exhibited discrimination against ¹³C. The data resulted in the following fractionation factors: an ɛ_ac of −18.70‰ and an ɛ_ac-methyl of −16.90‰ for the range 0 < f < 0.5 and an ɛ_CH4 of −22.12‰ for the range 0 < f < 0.8 (Fig. 4 and Table 1).

FIG. 3.

Anaerobic incubation of rice field soil amended with acetate. (A) Acetate consumption, CH₄, and CO₂ production. (B) Isotope signatures of (total) acetate, acetate-methyl, CH₄, and CO₂. ▪, acetate; □, acetate-methyl; •, CH₄; ⋄, CO₂. The concentrations are given as mmol per tube; values are means ± standard errors (n = 3). d, days.

FIG. 4.

Carbon isotope fractionation during consumption of acetate in rice field soil. The plots are based on equations derived by Mariotti et al. (30) to determine isotope fractionation of acetate (▪), acetate-methyl (□), and acetate-carboxyl (▵) (A) and of CH₄ (•) (B). Isotope enrichment factors were calculated for different phases of acetate consumption. In panel A solid lines show ln(1 − f) values between 0 and −0.6, dashed lines indicate values between −0.6 and −1.6, and dotted lines show values between −1.6 and −3.0; in panel B solid lines represent values of (1 − f) ln(1 − f)/f between −1.0 and −0.6, and dashed lines indicate values between −0.6 and 0. The values are means ± standard errors (n = 3).

DISCUSSION

Carbon isotope fractionation during the course of acetate consumption.

Our results for isotope fractionation of stable carbon in CH₄ by M. barkeri disagree to some extent with previous data. Krzycki et al. (27) and Gelwicks et al. (14) determined ɛ_ac values between −23.1 and −24.5‰ and ɛ_ac-methyl values between −21.2 and −24.0‰, whereas our average values were −30.5‰ for ɛ_ac and −25.6‰ for ɛ_ac-methyl. Hence, the isotope fractionation (in terms of α) observed in our experiments was slightly larger. However, the strains tested were not the same. The M. barkeri strain Fusaro (DSM 804) was used in this study, whereas Krzycki et al. (27) worked with M. barkeri MS (DSM 800), and Gelwicks et al. (14) worked with M. barkeri 228 (DSM 1538). These different strains may express slightly different carbon isotope fractionations. An additional, more important aspect is that most of the earlier studies (27, 28) were based on initial and end point measurements and did not monitor substrate consumption over time. Therefore, the fractionation factors could not be determined over the course of acetate consumption. The determinations of ɛ values by Gelwicks et al. (14) also depended on only a few data points. Consequently, it was not possible to differentiate fractionation in acetate and acetate-methyl between different stages of acetate consumption. We observed strong changes in isotope enrichment occurring during the degradation process and, thus, divided the data points into three different stages (Fig. 2A and D). Fractionation factors near the end of the incubation were generally smaller than those determined in an earlier stage. These discontinuities, particularly those at substrate levels when >80% of the substrate was consumed, were also observed in other recent studies (15, 24) and may result from the fact that substrate concentrations were no longer saturating. If acetate concentrations are saturating, fractionation is fully expressed. If acetate concentrations are limiting, on the other hand, each molecule is processed by the microbes irrespective of its isotopic composition so that isotope fractionation is no longer expressed (i.e., α = 1 and ɛ = 0). We assume that the slightly lower fractionation factors (lower α and higher ɛ) in the earlier studies than in the present study were because fractionation factors decreased along with acetate concentrations during the course of the experiment.

The decrease in the fractionation factor (increase of ɛ) during the final stages of acetate consumption were observed in all of our experiments, i.e., in M. barkeri, M. acetivorans, and methanogenic rice field soil (Table 1). The decrease was especially pronounced with regard to the isotope composition of acetate-methyl. After consumption of >80% of the acetate, ɛ values even became positive (i.e., α < 1), indicating a nonnormal fractionation. We hypothesize that at this stage a kinetic isotope effect was no longer expressed and that the slightly positive ɛ values (about +5‰) were caused by a relatively small equilibrium isotope effect (EIE), which is normally masked by the much stronger kinetic isotope effect. An EIE of similar magnitude was recently reported for acetate dissimilation by sulfate reducers through the tricarboxylic acid cycle (15). These authors hypothesized that the EIE is caused by the equilibrium between ionized and protonated acetate and the exclusive uptake of protonated acetate into the microbial cells. Such an EIE should apply not only to acetate-dissimilating sulfate reducers but also to acetate-dissimilating methanogens.

The decrease in the fractionation factor in Methanosarcina spp. with time was observed only in the substrate acetate (ɛ_ac and ɛ_ac-methyl) and not in the product methane (ɛ_CH4). We think that the reason for this difference is that acetate limitation affects the fractionation only during biochemical activation of the substrate. In addition, the concentration of the pooled CH₄ at a late incubation time was very high compared to that of the instantaneously formed CH₄ so that the δ¹³C of the newly formed CH₄ was strongly diluted by the previously accumulated CH₄. Hence, determinations of ɛ_CH4 values were sensitive only during early stages of growth. In rice field soil, on the other hand, the ɛ_CH4 value also exhibited a shift when >80% of the available acetate was consumed. The shift may be caused by an increasing contribution of hydrogenotrophic methanogenesis to total CH₄ production (see below).

Different levels of fractionation in acetate-methyl and total acetate.

Results of the experiments with both methanogens indicated a stronger fractionation of ¹³C in total acetate than in the methyl carbon at all phases of acetate consumption, as observed previously in M. barkeri (14) and Methanosaeta concilii (34). Exchange of the acetate-carboxyl with CO₂ (10) is not an explanation for the stronger fractionation in acetate-carboxyl versus acetate-methyl, since δ¹³C values of acetate-carboxyl eventually reached much higher values (δ¹³C of >10 to 20‰) than in TIC (δ¹³C of <−15‰). Therefore, the stronger fractionation in acetate-carboxyl versus acetate-methyl must be due to an isotope effect. During activation of acetate to acetyl-CoA, the actual bond formation occurs at the carboxyl carbon, and, hence, the carboxyl carbon might experience a stronger isotope effect than the methyl carbon. As suggested by Zyakun (47), the carbonyl group of the resulting acetyl-CoA would be depleted in ¹³C relative to the carboxyl group of the residual acetate. Alternatively, the carbon isotope fractionation may be expressed during cleavage of acetate by the multienzyme CO dehydrogenase/acetyl-CoA synthase complex (12, 13, 16). This multienzyme complex catalyzes the cleavage of acetyl-CoA, the oxidation of the carbonyl group to CO₂, and the eventual transfer of the methyl group of acetate to CH₃-S-CoM (12, 16). Possibly, this multienzyme complex is the rate-limiting step and responsible for the observed fractionation (14).

Differences in isotope fractionation within the genus Methanosarcina.

Previous studies have shown that carbon isotope fractionation is different between the two acetoclastic genera Methanosarcina and Methanosaeta (see introduction). In this study we observed differences even within the genus Methanosarcina. While the enrichment factors for acetate-methyl (−25.6‰ and −24.8‰, respectively) were nearly identical in M. barkeri and M. acetivorans, the values for ɛ_ac (−30.5‰ and −35.3‰) and ɛ_CH4 (−27.4‰ and −23.8‰) disagreed. Although both methanogens use the same enzymes for acetate activation, i.e., acetate kinase and phosphotransacetylase, other enzymes involved in acetate dissimilation are different. Guss et al. (18) reported that the Ech hydrogenase (a ferredoxin-reducing membrane-bound [Ni/Fe] hydrogenase with sequence similarity to energy-conserving NADH:quinone oxidoreductase [32]) is essential for growth on acetate in M. barkeri while this enzyme is not present in M. acetivorans. In contrast to M. barkeri, M. acetivorans cannot grow on H₂ plus CO₂ and thus cannot reduce the methyl group of acetate with H₂. Furthermore, it has been observed that M. barkeri in contrast to M. acetivorans lacks genes encoding acetyl-CoA synthetase (29). It cannot be ruled out that these differences also affect isotope fractionation.

As expected for a closed system, the isotopic composition of the pooled product (δ¹³C_CH4) at completion almost agreed with the initial substrate (δ¹³C_ac-methyl) (confirmed by experiments with labeled [2-¹⁴C]acetate) in both Methanosarcina spp. (Fig. 1B and D). Also the isotope enrichment in acetate-methyl and CH₄ during the course of acetate consumption agreed within error, at least during the initial phase. Minor deviations could be theoretically caused by assimilation of acetate because a branching of the carbon flow occurs when acetate is converted into biomass instead of CH₄. In our experiments the δ¹³C of the biomass was slightly enriched in ¹³C compared to the initial δ¹³C_ac for both methanogens (δ¹³C_biomass of −18.5 to −19.0‰). Consequently, this might have resulted in a slightly stronger depletion of ¹³C in CH₄ than in acetate, which, however, was not observed in our experiments. However, because of the relatively low level of biomass formation in anaerobic metabolism, we assume that assimilation had no significant influence on fractionation during acetate dissimilation.

Carbon isotope fractionation and archaeal diversity in rice field soil.

The most abundant archaeal family in the paddy soil, detected by T-RFLP analysis, was Methanosarcinaceae. The other acetoclastic family, Methanosaetaceae, was detected only at the beginning of the incubation and only to a minor extent. We assume that Methanosarcinaceae eventually dominated since they can grow faster than Methanosaetaceae at the relatively high acetate concentrations added (22). Since rice field soil had been preincubated under anoxic conditions, electron acceptors such as nitrate, sulfate, and available ferric iron were completely depleted. Under such conditions, acetoclastic methanogenesis is the almost exclusive process of acetate consumption (20, 39). It is noteworthy that the ɛ values determined in rice field soil were generally higher (i.e., fractionation was weaker) than those in pure cultures of Methanosarcina spp. Reduction of CO₂ by the hydrogenotrophic methanogens present in soil could have affected the values of δ¹³C_CH4 and of the resulting ɛ_CH4. An increased contribution of hydrogenotrophic methanogenesis may also explain the observed shift in isotope fractionation of CH₄ after depletion of >80% of the available acetate, resulting in progressively lower values of δ¹³C_CH4. Such a shift was not observed in the pure cultures, in which hydrogenotrophic methanogenesis was absent.

Despite this complication, values of ɛ, in particular those of ɛ_CH4 (−22‰), were still in a range that is consistent with acetate consumption by acetoclastic Methanosarcina spp. (−27‰ to −24‰). Also, ɛ_ac-methyl (−17‰) was still consistent with acetoclastic Methanosarcina spp. (−26‰ to −27‰), in particular, since ɛ_ac-methyl increased to values greater than −11‰ when 50 to 80% of the available acetate was consumed (Table 1). It might be possible that delivery of acetate to the microbial cells was hindered by diffusion in the soil matrix, thus resulting in smaller fractionation than observed during the initial phase of acetate consumption in microbial culture. In addition, acetate concentrations were generally lower in rice field soil (<5 mM) than in microbial cultures (<20 mM). We may compare the experiments with rice soil with similar ones with rice roots (35). The archaeal microflora on rice roots is also dominated by Methanosarcinaceae (5). A previous study using suspensions of excised rice roots resulted in an ɛ_ac-methyl value of −24.7‰ (35). This value is almost the same as found in the pure cultures of M. barkeri and M. acetivorans (Table 1). In the root system delivery of acetate to the microbial cells was probably not hindered by diffusion to the same extent as in the soil system.

Conclusions.

The fractionation factors during acetate consumption in M. barkeri and M. acetivorans were somewhat larger than those previously determined. Thus, the difference between the two acetoclastic genera Methanosarcina and Methanosaeta is even bigger than assumed before. We may therefore assume that under methanogenic conditions, ɛ values of less than −20‰ indicate a predominant activity of Methanosarcinaceae rather than Methanosaetaceae. Consequently, the determination of ɛ_ac and/or ɛ_CH4 may help predict which methanogenic genus is active. However, our results also illustrate that it is important to determine fractionation factors for acetate carbon over the course of acetate depletion since a decrease in acetate concentrations causes a shift in the effective fractionation of the substrate. Therefore, we suggest using the early phase of substrate consumption (e.g., until 50% substrate has been consumed) to determine the isotope fractionation associated with the biological process during a period where transport limitation is negligible or to determine the fractionation in the accumulated product CH₄, which is less affected by substrate limitation.