Environ. Sci. Technol. 2011, 45, 508–513
Predominant Contribution of Syntrophic Acetate Oxidation to Thermophilic Methane Formation at High Acetate Concentrations ¨ , PIN-JING HE,* LI-PING HAO, FAN LU LEI LI, AND LI-MING SHAO State Key Laboratory of Pollution Control and Resources Reuse, Key Laboratory of Yangtze River Water Environment, College of Environmental Science and Engineering, Tongji University, 1239 Siping Road, Shanghai 200092, P.R. China
Received July 2, 2010. Revised manuscript received December 5, 2010. Accepted December 7, 2010.
To quantify the contribution of syntrophic acetate oxidation to thermophilic anaerobic methanogenesis under the stressed condition induced by acidification, the methanogenic conversion process of 100 mmol/L acetate was monitored simultaneously by using isotopic tracing and selective inhibition techniques, supplemented with the analysis of unculturable microorganisms. Both quantitative methods demonstrated that, in the presence of aceticlastic and hydrogenotrophic methanogens, a large percentage of methane (up to 89%) was initially derived from CO2 reduction, indicating the predominant contribution of the syntrophic acetate oxidation pathway to acetate degradation at high acid concentrations. A temporal decrease of the fraction of hydrogenotrophic methanogenesis from more than 60% to less than 40% reflected the gradual prevalence of the aceticlastic methanogenesis pathway along with the reduction of acetate. This apparent discrimination of acetate methanization pathways highlighted the importance of the syntrophic acetate-oxidizing bacteria to initialize methanogenesis from high organic loadings.
Introduction Anaerobic digestion of biomass for methane (CH4) production is attractive for its potential to substitute for fossil fuels and reduce CO2 emissions (1). Fast fermentation of easily biodegradable biomass like food waste and slow methanogenesis can lead to accumulation of volatile fatty acids (VFAs) in the methanogenic bioreactors, which is widely acknowledged as a common problem inducing process deterioration (2). As both the substrate and an inhibitor, metabolism of VFAs is a key link in the CH4 production chain, with crucial influence on the stability of anaerobic digestion (3). Concentration of VFAs is recognized as a major selecting pressure that strongly influences methanogenic community structures and even leads to a metabolic pathway shift for Methanosarcina (4). In-depth knowledge of the degradation mechanism of VFAs at high levels will be of great interest to obtain insight into the initiation centers for methanogenesis (5) and the recovery process after deterioration, thus helping to improve operating strategy and to accelerate restarting of anaerobic digestion. Two pathways could be responsible for the methanization of acetate, the most dominant intermediate in anaerobic * Corresponding author e-mail:
[email protected]. 508
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 45, NO. 2, 2011
digestion of organic matter (6-8). The first one is aceticlastic methanogenesis (AM), which is carried out by Methanosarcinaceae or Methanosaetaceae (6). The second process encompasses two reactions, the syntrophic acetate oxidation (SAO) to H2 and CO2 and the subsequent conversion of these products to CH4 by hydrogenotrophic methanogenesis (HM). Thermodynamically, acetate oxidation is inferior to the aceticlastic pathway (SI Table S1) and can only proceed if H2 partial pressures are kept low by coupling with H2-consuming methanogenesis. It is generally assumed that most of the CH4 produced (up to 67%) should be directly derived from acetate (4), hence the operation and optimization of the biogas reactors are currently based on maintaining aceticlastic methanogens, with the role of the SAO pathway neglected (6). However, some studies have recently found that SAO became significant in thermophilic or stressed environmental conditions, and the syntrophs of acetateoxidizing bacteria and their partner hydrogenotrophic methanogens outcompeted the aceticlastic methanogens to be the dominant acetate utilizers (6-12). Besides the artificial methanogenic reactors, CH4 formation via tandem reactions of syntrophic acetate oxidation and hydrogenotrophic methanogenesis (SAO-HM) has also been suggested to occur in natural environments such as lake sediments, oil reservoirs, and soils (7). The diversity of environments where SAO has been found invoked the recognition of potential role of this pathway in commercial gas generation and waste reduction in anaerobic digesters. Although SAO is not scarce in anaerobic environments, the ecological nature of the habitats for syntrophic acetateoxidizing bacteria and the quantitative contribution of SAO pathway to the total CH4 production, especially to the initiation of methanogenesis in acidic environments, are still unclear. The two acetate conversion pathways can be quantified by the application of stable isotopic tracers or inhibitors of aceticlastic methanogenesis (13). Qu et al. (4) monitored the evolution of the methanogenic pathways using the stable carbon isotopic signature of CH4/CO2, and the results coincided well with the observed changes in methanogenic community structure using 16S rRNA gene fingerprinting techniques. Methyl fluoride (CH3F), as a rather specific inhibitor of AM (14), has also been used to monitor changes in carbon flow in methanogenic environments, especially in the rice field system (15, 16). However, little is known of this aspect as applied to artificial biogas reactors. The present work quantified the temporal changes of the thermophilic methanogenic pathways at high-level acetate conditions. The stable carbon isotopic signature and the specific inhibitor of AM were employed in parallel, combined with the PCR-denaturing gradient gel electrophoresis (PCRDGGE) to investigate the microbial community structure. The experimental results demonstrated that SAO predominated in the initial period of acetate conversion at high concentrations, which suggests that the syntrophic acetateoxidizing bacteria played an important role in the initiation centers for methanogenesis.
Material and Methods Biomass, Sampling, and Culture Conditions. All experiments were performed with freshly collected methanogenic granular sludge cultivated in a laboratory-scale anaerobic sequenced batch reactor (ASBR). The ASBR was operated at 55 °C and used glucose and acetate (at ratio 80:20, calculated as chemical oxygen demand, COD) as the substrate at a loading rate of 2000 mg-COD/(L · d). Sampled sludge was rinsed with 10.1021/es102228v
2011 American Chemical Society
Published on Web 12/16/2010
FIGURE 1. Temporal change in (a) acetate concentration (closed symbols) and accumulated methane (open symbols) in the control: CL-1 (b, O), OP-1 (9, 0), CH3F incubations: CL-2 (1, 3), OP-2 ([, ]) and blank: BL (2, ∆) (b) methane produced via syntrophic acetate oxidation-hydrogenotrophic methanogenesis (closed symbols) and aceticlastic methanogenesis (open symbols) in the closed system (b, O) and open system (1, ∆). anaerobic preheated (55 °C) medium to remove any residual carbon source. Specific metabolic activities, including aceticlastic methanogenic activity, hydrogenotrophic methanogenic activity, and acetate oxidizing activity (SI Figure S1), were analyzed for the inoculum according to a method modified from Uemura et al. (17) (SI Section 3). Experimental Setup. Twenty milliliters of sludge mixture was transferred into each 1.2-L reactor with 450 mL basal medium containing (per liter): 1.0 g of NH4Cl, 0.4 g of K2HPO4 · 3H2O, 0.2 g of MgCl2 · 6H2O, 0.08 g of CaCl2 · 2H2O, 10 mL of trace element solution, and 10 mL of stock vitamin solution. The stock trace element and vitamin solutions were made according to Chen et al. (18). Na2S · 9H2O at 200 mg/L was used, and 30 mL concentrated acetate solution was added as the substrate to reach a final concentration of 100 mmol/ L. The sludge concentration was 3 g of volatile suspended solids (VSS) per liter. The range in pH was 6.8-7.8, regulated with 5 mol/L H3PO4. The gas phase was exchanged by several cycles of applying a vacuum and refilling with oxygen-free N2. There were five sets of experiments: CL-1, CL-2, OP-1, OP-2 and BL, with closed system (CL) and open system (OP) simulated by different gas regulating methods. Aceticlastic methanogenesis was inhibited by injection of 2.3% (v/v) CH3F (99%) to the headspace in CL-2 and OP-2, after CH3F addition, the reactors were vigorously shaken for 2 min to homogenize the gas distribution. Controls without inhibitor were termed CL-1 and OP-1. Endogenous methanogenesis was studied in BL as the blank without acetate addition. All reactors were incubated statically at 55 °C in darkness. Liquid and gas samples were taken periodically for analysis until the added acetate was exhausted. The above-mentioned closed system for the CL group and open system for the OP group with their different gas regulating methods were used to assess whether the H2 and CO2 partial pressures influenced the SAO pathway. For OP-1 and OP-2, the headspace was replaced with pure N2 by several cycles of removing the residual gas and refilling with oxygenfree N2 by applying a vacuum after each gas sampling event, to simulate the open system with product output. 2.3% CH3F was injected into OP-2 after gas replacement. For CL-1 and CL-2, the produced biogas was allowed to accumulate in the headspace, to simulate a closed system without matter exchange. For CL-2, loss of CH3F in sampling was complemented to maintain the inhibiting concentration. In this way, we could calculate stable carbon isotopic signatures using both approaches. All experiments were done in duplicate and data are given as arithmetic mean ( standard deviation.
Analysis of the Samples. Gas composition (CH4, CO2, and H2) and volume were analyzed, and the stable carbon isotopic signatures of CH4 (δ13CH4) and CO2 (δ13CO2) were periodically monitored. The liquid samples were analyzed for pH, total organic carbon (TOC), total inorganic carbon (TIC), and VFAs. Structure of the dominant microbiota was assessed using PCR-DGGE. A detailed description of the methods for these analyses is given in SI Section 4. Experimental Methods for the Determination of fmc. fmc, defined as the fraction of CH4 derived from CO2 to total CH4, was used to quantify the HM, which also represented the SAO pathway of acetate conversion in our study. We differentiated the calculated fmc which were obtained by the isotopic tracer and the inhibitor method as fmc-is and fmc-in, respectively. fmc-in was defined as the ratio between the production rates of CH4 from H2/CO2 and both acetate and H2/CO2 as follows fmc-in )
dCH4-H2/CO2 /dt dCH4-total /dt
Calculation of fmc-is and fmc-in is described in SI Section 5, and thermodynamic analysis is in SI Section 6.
Results Specific Metabolic Activity of the Microbial Consortium. Specific metabolic activity assays showed that the aceticlastic and hydrogenotrophic methanogenic activities of the inoculum were 272 and 104 µmol-CH4/(h · g-VSS), respectively. The results indicated that acetate-cleaving and H2-utilizing methanogens were both enriched in the studied consortium. In the acetate oxidizing activity test, during 133 h of incubation, hydrogen partial pressure (pH2) increased to 72 ( 5 Pa in the test vial, while that of the control was always 18 mmol/L) in the uninhibited controls. During the period 0-44 h, there was no apparent difference in acetate consumption rate in the presence or absence of AM inhibitor. With further acetate degradation, the conversion rate in the AM-inhibited sets became much lower than in the controls. When the added acetate was exhausted, the interdiction of AM in CL-2 and OP-2 caused a CH4 yield reduction of 44 and 46%, respectively, compared VOL. 45, NO. 2, 2011 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
509
FIGURE 2. (a) Evolution of δ13C of CH4 (closed symbols) and CO2 (open symbols) in the closed system (b, O), open system (1, ∆) and the calculated newly formed CH4 (9) and CO2 (0) without CH3F; (b) apparent fractionation factor rc in the controls of the closed system (b) and open system (3); fraction of the hydrogenotrophic methanogenesis calculated by isotope fmc-is (b) and inhibitor fmc-in (2) versus residual acetate concentration in (c) the closed system and (d) the open system. with those in the uninhibited CL-1 and OP-1. Gradual differentiating of acetate conversion rates due to AM inhibition demonstrated that AM and SAO both participated in conversion of 100 mmol/L acetate, and the contribution of these two pathways changed with incubation time and acetate concentration. In contrast to the immediate start of SAO-HM, AM was initiated slowly after a lag phase of 20 h (Figure 1b), and the fraction of the total CH4 production it consumed gradually increased before the acetate level decreased to the limiting concentration (5 mmol/L, see definition in SI Section 7). In the present study, no acetate was detected throughout in BL, and a very small amount of CH4 (≈0.02 mmol and
