Response of Syntrophic Propionate Degradation to Ph Decrease And

J. Microbiol. Biotechnol. (2016), 26(8), 1409–1419 http://dx.doi.org/10.4014/jmb.1602.02015 Research Article Review jmb

Response of Syntrophic Propionate Degradation to pH Decrease and Microbial Community Shifts in an UASB Reactor Liguo Zhang1,2*, Qiaoying Ban2, Jianzheng Li1, and Ajay Kumar Jha3

1State Key Laboratory of Urban Water Resource and Environment, School of Municipal and Environmental Engineering, Harbin Institute of Technology, Harbin 150090, P.R. China 2College of Environmental and Resource Sciences, Shanxi University, Taiyuan 030006, P.R. China 3Central Campus Pulchowk, Institute of Engineering, Tribhuvan University, Lalitpur 44700, Nepal

Received: February 11, 2016 Revised: April 27, 2016 The effect of pH on propionate degradation in an upflow anaerobic sludge blanket (UASB) Accepted: May 3, 2016 reactor containing propionate as a sole carbon source was studied. Under influent propionate of 2,000 mg/l and 35ºC, propionate removal at pH 7.5-6.8 was above 93.6%. Propionate conversion was significantly inhibited with stepwise pH decrease from pH 6.8 to 6.5, 6.0, 5.5,

First published online 5.0, 4.5, and then to 4.0. After long-term operation, the propionate removal at pH 6.5-4.5 May 9, 2016 maintained an efficiency of 88.5%-70.1%, whereas propionate was hardly decomposed at pH

*Corresponding author 4.0. Microbial composition analysis showed that propionate-oxidizing bacteria from the Phone: +86-351-7011-932; genera Pelotomaculum and Smithella likely existed in this system. They were significantly Fax: +86-351-7011-932; reduced at pH ≤5.5. The methanogens in this UASB reactor belonged to four genera: E-mail: [email protected] Methanobacterium, Methanospirillum, Methanofollis, and Methanosaeta. Most detectable hydrogenotrophic methanogens were able to grow at low pH conditions (pH 6.0-4.0), but the acetotrophic methanogens were reduced as pH decreased. These results indicated that propionate-oxidizing bacteria and acetotrophic methanogens were more sensitive to low pH (5.5-4.0) than hydrogenotrophic methanogens. pISSN 1017-7825, eISSN 1738-8872

Introduction (>2,500 mg/l) may inhibit the activity of propionate- oxidizing bacteria and methanogens, even if pH is Anaerobic digestion technology is widely used for treating maintained near neutral [8]. Therefore, the degradation of all kinds of organic wastes, including municipal sewage, propionate is considered as a rate-limiting step in anaerobic organic wastewater, animal waste, and agricultural wastes, digestion systems [1]. and simultaneously generates methane as an energy source Under methanogenic conditions, propionate is converted [2, 6, 48, 52]. During the anaerobic degradation of these by cooperation between propionate-oxidizing bacteria and organic matters, many intermediates, including propionate, methanogens [32]. Propionate-oxidizing bacteria decompose are formed [32, 40]. The main sources of propionate in the propionate into acetate and H2/CO2 to supply substrate bioreactors are odd-numbered fatty acids from fat, oil, and for methanogens. In return, the methanogens relieve carbohydrate [7]. Propionate is easily accumulated when feedback inhibition for the propionate-oxidizing bacteria the anaerobic digestion system is disturbed by temperature by transforming acetate and H2/CO2 into methane. fluctuation, toxic substances, or high organic load rate [32, Propionate-oxidizing bacteria are often difficult to isolate 40]. The accumulation of propionate would decrease the and culture owing to their stringent metabolism requirements system pH and can lead to process failure and instability [32]. So far, only a few slow-growing propionate-oxidizing [13, 15, 42]. In addition, a high concentration of propionate bacteria were identified [32]. These bacteria can degrade

August 2016 ⎪ Vol. 26⎪ No. 8 1410 Zhang et al.

propionate to acetate, CO2, and hydrogen in the absence of electron acceptors and belong to four genera: Syntrophobacter, Smithella, Pelotomaculum, and Desulfotomaculum. Syntrophobacter and Smithella are grouped into the class δ-proteobacteria in the phylum Proteobacteria, whereas Desulfotomaculum and Pelotomaculum belong to the phylum Firmicutes. Propionate- oxidizing bacteria are extensively present in several anaerobic ecosystems, including flooded soils, freshwater sediments, tundra, wet-wood of trees, landfills, anaerobic granular sludge, and sewage digesters [23, 35, 39, 43, 44]. Their activities and distributions can be influenced by various factors, including temperature, hydraulic retention time (HRT), and hydrogen partial pressure [9, 31, 32]. For instance, the activities of propionate-oxidizing bacteria were significantly inhibited at temperature less than 20ºC Fig. 1. A schematic of the UASB reactor. in an UASB reactor in which P. schinkii was the dominant propionate-oxidizing bacteria [9, 11]. A large amount of pumped to the reactor by a peristaltic pump. The reactor was P. thermopropionicum is distributed in the internal layers of wrapped by an electrothermal wire and was maintained at 35ºC the thermophilic (55ºC) granule [26]. In addition, propionate by a temperature controller. The evolved biogas was collected degradation ability was increased when the HRT was separately from the upper part of the reactor and entered a water decreased to 4 h from 10 h in an anaerobic reactor, in which lock. All the water locks and wet gas meters were filled with the genus Syntrophobacter was the dominant propionate- water of pH 3.0 to prevent dissolution of the biogas. oxidizing bacteria [12]. However, Ariesyady et al. [4] found high proportions of Smithella spp. and lower proportions of Feed Stock and Operation Syntrophobacter spp. in anaerobic sludge, but they did not Synthetic wastewater containing propionate as a sole carbon determine Pelotomaculum spp. Thus, the activities and source was fed into a UASB reactor from startup. The propionate species distributions of propionate-oxidizing bacteria vary concentration in the influent was maintained at 2,000 mg/l. The in different systems. basic medium used in the reactor was as described previously [9]. Although some studies of propionate degradation have The UASB reactor was used for treating propionate synthetic wastewater for 6 months under different temperature conditions been performed, there is little information about the effects [9]. Then the operational temperature was gradually adjusted to of pH on the activity and distribution of propionate- 35 ± 1ºC. The pH of the reactor was maintained at around oxidizing bacteria. It is worth to investigate the effect of pH 7.5 ± 0.5. The HRT was controlled to 6 h using a peristaltic pump. decrease on bacterial activity and determine whether it Since anaerobic sludge is focused on the sludge blanket in an can cause a shift of microorganisms. The current study UASB reactor, the pH in the sludge blanket was considered as a investigated the response pattern of propionate degradation control parameter. The pH was measured three times daily and on pH decrease in an upflow anaerobic sludge blanket was adjusted constantly by adding NaHCO3 (pH > 7.0) or HCl (UASB) reactor containing propionate as a sole carbon (pH < 7.0). The pH in the sludge blanket was gradually decreased source. In addition, changes in the population distributions in eight stages from 7.1 to 6.8, 6.5, 6.0, 5.5, 5.0, 4.5, and finally 4.0. of propionate-oxidizing bacteria and methanogens with pH The pH at each stage was maintained until a stable propionate decrease were analyzed using polymerase chain reaction- removal was achieved. At each stable state, sludge samples were denaturing gradient gel electrophoresis (PCR-DGGE). collected for microbial community analysis.

Analysis Materials and Methods The biogas volume was measured daily by wet gas meters

(LML-1; Changchun Filter Co., Ltd., China). The fraction of CH4 Bioreactor System was measured by gas chromatograph (SP-6800A; Shandong Lunan The study was performed in a lab-scale UASB reactor made of Instrument Factory, China) equipped with a thermal conductivity transparent plexiglass. The effective volume was 4 L. As shown in detector [9]. The liquid samples from the influent, sampling ports, Fig. 1, three evenly distributed sampling ports and a solid outlet and effluent were centrifuged at 9,000 ×g for 5 min and then were installed over the height of the column. The feed was treated by 6 M HCl acidification, before assaying for volatile fatty

J. Microbiol. Biotechnol. Syntrophic Propionate Degradation Response to pH Decrease 1411

acids (VFAs) using a gas chromatograph (SP6890; Shandong accession numbers KT896487- KT896493 and KT960984- Lunan Instrument Factory, China) equipped with a flame KT960994. ionization detector [9]. The pH and MLVSS (mixed liquor volatile suspended solid) Results and Discussion were measured according to the procedures described in the standard methods [3]. Effects of pH Decrease on the Propionate Degradation Chemical oxygen demand (COD) mass balance was calculated Under an organic loading rate of 12 kg COD/m3/day and with the equation below, where COD (mgCOD/l) and COD final initial 35ºC conditions, the effects of pH decrease on propionate (mgCOD/l) are the COD in the effluent and influent, respectively. degradation were investigated in an UASB reactor. The and CH (mgCOD/l) and MLVSS (mgCOD/l) represent 4 increased UASB reactor was continuously operated for 240 days with equivalent COD of the produced methane and increased MLVSS for treating 1 L of propionate synthetic wastewater. a HRT of 6 h. Fig. 2 shows the daily variations in pH, propionate removal, methane content, biogas production, COD balance (%) = and acetate content. As the only independent variable, the []COD () mg COD⁄ l ++MLVSS () mg COD⁄ l CH ()mg COD⁄ l ------final increased ------4 pH was reduced gradually with decreased influent pH. COD ()mg COD⁄ l initial However, the pH was increased from bottom to top in the × 100 same operational stage. In addition, the pH in the three DNA Extraction, PCR, and DGGE sampling ports was similar when the pH of the sludge Genomic DNA was extracted using a DNA extraction kit (MO Bio Laboratories, USA) according to the manufacturer’s instructions. The universal bacterial primers used to perform the polymerase chain reaction are as follows: for bacteria, 341F (5’- CCTACGGGAGGCAGCAG-3’, with a GC clamp) and 907R (5’- CCGTCAATTCMTTTGAGTTT-3’); for archaea, 344F (5’-ACG GGGYGCAGCAGGCGCGA-3’, with a GC clamp) and 915R (5’- GTGCTCCCCCGCCAATTCCT-3’). The PCR amplification was conducted in a 50 μl volume containing 5 μl 10× Ex Taq buffer, 4 μl dNTP mixture (2.5 mM), 1 μl forward primer (20 μM) and 1 μl reverse primer (20 μM), 2.5 ng DNA template, and 0.15 U Ex Taq DNA polymerase (Takara, China). The amplification conditions started with an predenaturation of DNA for 10 min at 94°C, followed by 30 cycles for 1 min at 94°C, 1 min at 55°C, and 45 sec at 72°C, followed by a final extension of 10 min at 72°C. The PCR products were separated using the DCode system (Bio-Rad Laboratories, USA). Polyacrylamide gels with 40%-60% vertical denaturing gradient were prepared. The 15 μl PCR products after making to the same concentration were loaded and electrophoresed at 120 V and 60°C for 10 h. Gels were silver stained as described previously [14]. Denaturing gradient gel electrophoresis bands were excised and dissolved in 30 μl of 1× TE at 40°C for 3 h, and then centrifuged at 12,000 rpm for 3 min. The 3 μl supernatant was used as the template for PCR amplification under the conditions as above described using the same primers without the GC clamp. These PCR products were purified by a Gel extraction mini kit (Watson Biotechnologies, China), ligated into the pMD18-T vector (Takara, China), and then cloned into E. coli DH5α. Three white colonies from each sample were randomly selected for PCR detection, and two positive colonies were selected for sequencing by Sangon Biotech., Ltd. (China), and partial 16S rRNA gene sequences were analyzed Fig. 2. Performance data (pH, propionate removal, methane using the BLAST program of NCBI. After analysis, 20 partial 16S content, biogas production, acetate content) of the UASB rRNA gene sequences were submitted to GenBank under the reactor.

August 2016 ⎪ Vol. 26⎪ No. 8 1412 Zhang et al.

Table 1. The operational performance under different pH conditions. Propionate removal Acetate content in effluent Biomass Specific methane production rate Stage (%) (mg/l) (gMLVSS/l) (L CH4 /kg CODremoved·day) pH 7.5 98.6 ± 1.5 55.6 ± 6.9 9.2 ± 0.6 312.8 ± 18.5 pH 7.1 95.1 ± 2.3 69.4 ± 10.5 9.5 ± 0.5 317.7 ± 15.6 pH 6.8 93.6 ± 1.6 25.4 ± 4.3 10.0 ± 0.7 330.9 ± 30.7 pH 6.5 88.5 ± 1.3 94.6 ± 12.6 10.3 ± 0.1 307.6 ± 20.5 pH 6.0 86.1 ± 2.5 58.3 ± 11.5 9.6 ± 0.4 283.0 ± 16.8 pH 5.5 82.1 ± 0.9 42.3 ± 7.2 7.3 ± 0.7 267.9 ± 15.9 pH 5.0 79.3 ± 2.7 43.7 ± 9.1 6.5 ± 0.8 257.6 ± 19.6 pH 4.5 70.1 ± 1.8 43.1 ± 8.6 5.6 ± 0.3 190.7 ± 20.1 pH 4.0 0 0 4.9 ± 0.6 0 blanket was higher than 6.5. However, when the pH of the operation at pH 6.0, 5.5, and 5.0, the contribution rate of sludge blanket was lower than 6.0, the pH in sampling propionate degradation in the suspended layer was 31.6%, ports 2 and 3 was always higher than the pH in sampling 40.6%, and 54.1%, respectively (Fig. 3). The higher level of port 1. propionate degradation was likely related to the higher pH The total propionate removal reached 93.6% at pH 7.5- in the suspended layer (sampling 2 and 3) of about 6.0-5.0, 6.8 and most of the propionate (81.5%-90%) was degraded a pH range that most propionate-oxidizing bacteria can by the sludge blanket (Table 1 and Fig. 3). This is the endure [32]. Under the cooperation between the sludge previously reported favorable pH range for most propionate- blanket and the outside of the sludge blanket, the total oxidizing bacteria [32]. When the pH in the sludge blanket propionate removal of the UASB reactor at pH 6.0, 5.5, and was swiftly lowered to 6.5, the total propionate removal 5.0 was 86.1%, 82.1%, and 79.3%, respectively. When pH was rapidly reduced to 75.9% and the propionate removal was decreased from 5.0 to 4.5, the propionate removal of of the sludge blanket was only 60.6% (Figs. 2 and 3). After the sludge blanket was quickly reduced and reached to a 20 days of operation, the propionate removal reached steady state with propionate removal of 12.3% (Table 1). 67.6% in the sludge blanket and the total propionate Propionate degradation mainly occurred in the suspended removal stabilized at 88.5% (Table 1 and Fig. 3). When the layer at pH 4.5, composing 75.2% of total propionate pH was further decreased to 6.0, 5.5, and 5.0, the propionate removal (Fig. 3). The final pH decrease from 4.5 to 4.0 removal of the sludge blanket declined dramatically to resulted in barely detectable propionate decomposition 49.1%, 33.8%, and 16.7%, respectively. These results indicated (Fig. 2). that the activities of propionate-oxidizing bacteria were The results in Figs. 2 and 3 indicate that propionate markedly inhibited by decreasing pH. After a certain time degradation is closely related to pH, and a low pH of ≤6.5 of operation, propionate removal in the sludge blanket was significantly inhibited the metabolic ability of propionate- gradually increased and reached a steady state with oxidizing bacteria. Similarly, Dhaked et al. [19] also propionate removal levels of 56.9%, 45.5%, and 26.2% for reported faster propionate anaerobic conversion at neutral the pH conditions of 6.0, 5.5, and 5.0. During the steady or weak alkaline pH (7.0 to 8.0) compared with conversion

Fig. 3. The contribution rate of propionate removal in each stage.

J. Microbiol. Biotechnol. Syntrophic Propionate Degradation Response to pH Decrease 1413

Table 2. COD mass balance under different pH conditions. COD COD COD removal Propionate Acetate VSS CH COD balance Stage initial final final final increased 4 (mg/l) (mg/l) (%) (mg COD/l) (mg COD/l) (mg COD/l)a (mg COD/l) (%) pH 7.5 3,017 ± 50.5 102.2 ± 16.9 96.7 ± 1.5 42.7 ± 2.5 59.5 ± 8.9 102.4 ± 8.6 2,607 ± 80.9 93.2 ± 2.1 pH 7.1 3,056 ± 75.6 222.5 ± 30.4 92.7 ± 2.3 148.2 ± 3.7 74.3 ± 10.7 532.5 ± 10.5 2,571 ± 25.7 108.9 ± 2.5 pH 6.8 3,010 ± 45.2 217.9 ± 41.5 91.6 ± 1.6 190.7 ± 10.5 27.2 ± 9.5 537.9 ± 12.9 2,607 ± 90.3 111.7 ± 3.2 pH 6.5 3,005 ± 50.3 445.3 ± 32.6 85.2 ± 1.3 344.1 ± 13.6 101.2 ± 11.6 287.8 ± 21.7 2,250 ± 101.7 99.3 ± 1.1 pH 6.0 3,020 ± 80.7 480.9 ± 43.7 84.1 ± 2.5 418.5 ± 50.3 62.4 ± 7.5 − 2,053 ± 77.9 83.9 ± 4.6 pH 5.5 3,153 ± 66.9 581.5 ± 57.2 81.4 ± 0.9 536.3 ± 25.6 45.2 ± 5.8 − 1,964 ± 115.7 80.7 ± 5.7 pH 5.0 2,995 ± 77.5 737.5 ± 35.1 75.4 ± 2.7 690.9 ± 35.7 46.7 ± 8.7 − 1,660 ± 98.7 80.1 ± 6.8 pH 4.5 2,936 ± 65.3 903.7 ± 45.3 69.2 ± 1.8 897.6 ± 40.9 6.1 ± 9.1 − 1,107 ± 95.1 68.3 ± 3.9 pH 4.0 3,075 ± 46.9 3,078 ± 80.8 0 3,078 ± 80.8 0 − 0 100 ± 4.3 a1 mg VSS = 142 mgCOD [31]

VSSincreased (mg COD/l) = [(VSSt-VSSt-1) (g/l)×V(L) ×1000×142 mgCOD]/[(Time (day)) ×(Total influent/day)] (t, Present stage; t-1, Previous stage; V, Working volume of reactor)

“−”, VSSincreased (mg COD/l) could not be calculated because the VSS was reduced. at weak acid (pH 6.0). biogas production reached a steady state with 18.1 l/day at pH 5.0 and 14.5 l/day at pH 4.0. During the final operation Methane Production Response to pH Decrease stage (pH 4.0), there was little detectable production of The data in Fig. 2 show that the effects of pH also biogas. decreased methane production. Methane content and It is obvious from Figs. 2 and 3 and Table 1 that both biogas production were similar at pH 6.8-7.5, a favorable propionate-oxidizing bacteria and methanogens exhibited pH range for most methanogens [34]. Methane content and high activity at neutral (6.8 to 7.5) conditions. At acidic (pH biogas production at pH 6.8-pH 7.5 were maintained at 6.5-4.5) conditions, methanogens still showed relatively 55.2%-69.3% and 22.8-26.7 l/day, respectively. When the higher metabolic ability with a specific methane production pH in the sludge blanket was decreased to 6.5 from 6.8, the rate of 190.7-307.6 L CH4 /(kgCODremoved·day). However, methane content was hardly changed but the biogas propionate removal was significantly inhibited at pH 6.5- production was reduced by 27.7%. The biogas production 4.0. This indicates that propionate-oxidizing bacteria are gradually increased and reached to 23.0 l/day at steady more sensitive to acid environment than methanogens in state. Two pH decreases, from 6.5 to 6.0 and then 5.5, led to this experimental reactor. reductions in methane contents by 5.1% and 33.7%, Investigation of carbon mass balance enables monitoring respectively. The biogas production was deceased to and distribution of the metabolic products involved in an 15.3 l/day with the pH decreased to 6.0 from 6.5 and the anaerobic fermentation process [16]. The amount of biogas production was reduced to 19.0 l/day after 9 days metabolic products formed from propionate during the of operation at pH 5.5. After a long-term operation, the steady-state periods of COD mass balance are listed in methane contents at pH 6.0 and 5.5 were recovered to Table 2. The closure of the COD balances at 68.3%-111.7% 51.4% and 46.2%, respectively. The biogas production verifies the reliability of the data. The data in Table 2 show similarly gradually increased to the previous level. Although that COD balance was decreased as pH decreased from pH hydrogen and acetate did not markedly accumulate in this 6.0, likely due to sludge loss. During the experiment, the study, some studies reported that the feasible pH range for suspended substances in the effluent increased with pH acetotrophic methanogens is 6.6-7.3 and methanogenesis decrease. is obviously inhibited when the pH is lower than 6.2 [18]. The explanation for this requires further investigation. Microbial Community Shift with pH Decrease When the pH in the sludge blanket was further decreased Bacteria. The performance of an anaerobic biological from 5.5 to 5.0 and then 4.5, the methane content hardly treatment system is primarily linked to the structure of the changed but the biogas production was reduced by 22.7% microbial community in the reactor. To investigate the and 32.8%, respectively. After about 20 days running, the bacterial composition in anaerobic sludge at each operational

August 2016 ⎪ Vol. 26⎪ No. 8 1414 Zhang et al.

stage, PCR-DGGE of the partial 16S rRNA gene sequence rRNA gene sequences of three bands were related to was performed (Fig. 4A). DGGE-based molecular analysis previously identified propionate-oxidizing bacteria. Two has been extensively applied to monitor the changes in the DGGE bands belonged to the genus Pelotomaculum in the microbial diversity in organics in anaerobic treatment phylum Firmicutes. The DGGE band B3 showed 96% systems [24, 46, 50]. Understanding changes in community sequence similarity with P. schinkii, and band B4 was distributions provides insight into the metabolic function related to P. propionicicum. These two propionate-oxidizing of microorganisms in biological treatment systems and bacteria from Pelotomaculum are obligately syntrophic allows the optimization of operational conditions [46]. As bacteria, which grow only in co-culture with methanogens shown in Fig. 4A, seven bands were obtained from the [17, 27]. The band B6 was related to the genus Smithella DGGE gel and sequenced. The phylogenetic placement within phylum Proteobacteria with 99% sequence similarity based on the partial 16S rRNA gene sequences determined to S. propionica. P. propionicum and S. propionica can grow in from these bands and reference 16S rRNA gene sequences a pH range of 6.5-7.5 and 6.3-7.8, respectively. However, obtained from the NCBI database was constructed by the pH range of P. schinkii for growth is unknown [33]. MEGA 3.1 software (Fig. 4B). Pelotomaculum spp. degrade propionate by the methylmalonyl- As shown in Fig. 4B, all of the obtained sequences were coenzyme A pathway. S. propionica oxidizes propionate to grouped into three groups in the phylogenetic tree. 16S acetate and butyrate via an integration of two molecules of

Fig. 4. The DGGE map (A) and phylogenetic relationship (B) of bacteria in different pH conditions. All reference 16S rRNA gene sequences are from the NCBI database. The tree was constructed using the neighbor-joining method. The bar represents five substitutions per 100 nucleotide positions. Bootstrap probabilities are indicated at branch nodes.

J. Microbiol. Biotechnol. Syntrophic Propionate Degradation Response to pH Decrease 1415

propionate, followed by syntrophic β-oxidation of butyrate Although propionate was used as the sole carbon source, to acetate [33]. Butyrate degradation is thermodynamically some non-propionate-oxidizing bacteria were also present easier than propionate so that product inhibition is easily in this system. These non-propionate-oxidizing bacteria eliminated, leading to the promotion of propionate might utilize the products of microbial disintegration or oxidation by S. propionica [38]. This unique metabolic intermediates of propionate degradation as substrates for pathway of S. propionica may be favorable for propionate growth. As shown in Fig. 4B, bands B1 and B7 showed 99% degradation under acidic condition. sequence similarity with Petrimonas sulfuriphila BN3 from The present study showed that the genera Smithella and the phylum Bacteroidetes. P. sulfuriphila BN3 is a mesophilic Pelotomaculum may be the main propionate-oxidizing bacteria fermentative bacterium isolated from a biodegraded oil in this UASB system. Similarly, Smithella dominated in a reservoir in Canada, whose favorable pH for growth is 7.2 mesophilic anaerobic reactor fed with different concentrations [22]. Band B5 was closely related to the Bacteroidetes of propionate under neutral conditions [5]. In chemostat QEEB2AH08 (CU918164), a bacterium whose function has experiments containing propionate as the sole carbon source, not been characterized. Interestingly, Syntrophomonas spp. high dilution rates were found to promote Pelotomaculum were also observed in this UASB reactor. Band B2 was and low dilution rates promoted Syntrophobacter spp. at similar to S. zehnderi OL-4, which is a mesophilic syntrophic pH 7.0 [45]. In contrast, Worm et al. [51] found that fatty-acid-oxidizing bacterium. It can degrade C4-C8 Syntrophobacter was stable in UASB reactors treating compounds when in co-culture with methanogens [47]. The propionate synthentic wastewater at pH 7.0. However, existence of this bacterium might be related to the presence Smithella and Pelotomaculum competed with Syntrophobacter of Smithella propionica in the reactor. Intermediate butyrate for propionate consumption when the UASB reactor was can be produced during the S. propionica oxidization of not supplied with molybdenum, tungsten, and selenium propionate, and butyrate can be used as a substrate for for a long term [51]. S. zehnderi [33]. As shown in Fig. 4, propionate-oxidizing bacteria relevant Archaea. Propionate anaerobic oxidation depends on the band B3 was enhanced with the pH decrease from 7.1 to syntrophy of propionate-oxidizing bacteria and methanogens 6.0, indicating that this propionate degrader (band B3) is in methanogenic environments [9, 37]. Methanogens are slightly acid tolerant. In contrast, the signals of band B4 important for eliminating the products from propionate and band B6 decreased as pH decreased. Overall, the pH oxidation and then driving the propionate degradation decrease reduced propionate-oxidizing bacteria, resulting process. In this study, methanogens allowed the acetate in a decrease in propionate removal by the sludge blanket concentration to always be less than 140 mg/l in the (Fig. 3). These propionate-oxidizing bacteria (bands B3, B4, effluent. Moreover, the hydrogen content was lower than and B6) were also present in the suspended layer at pH the detection limit of the gas chromatograph during the 5.5-4.5, likely because the pH in the suspended layer was whole operational period (Fig. 2). Therefore, it is significant higher than that of the sludge blanket (Figs. 2 and 4). Other to know the change in the population of the methanogens bands have not shown obvious succession with decrease in as pH decreases in the UASB system (Figs. 5 and 6). pH from 7.5 to 6.0 and then to be reduced. The 13 excised DGGE bands were successfully sequenced

Fig. 5. The DGGE map of archaea in different pH conditions.

August 2016 ⎪ Vol. 26⎪ No. 8 1416 Zhang et al.

Fig. 6. Phylogenetic relationships of methanogens in different pH conditions. All references 16S rRNA gene were sequenced from the NCBI database. The tree was constructed using the neighbor-joining method. The bar represents two substitutions per 100 nucleotide positions. Bootstrap probabilities are indicated at branch nodes. and a similar search using the BLAST program in the NCBI variety of anaerobic reactors with low acetate concentrations database was performed. As shown in Fig. 6, all of the [54]. The presence of Methanosaeta species usually leads to obtained sequences grouped into three orders in the an improved granulation process and results in a more phylogenetic tree. The DGGE bands A1-A4 and A9 were stable bioreactor performance [29, 49]. Bands A8 and A12 affiliated with genus Methanobacterium within order with other 16S rRNA gene sequences formed a unique Methanobacteriales, with 97%-100% sequence similarity to branch in the phylogenetic tree, corresponding to the order Methanobacterium sp. MO-MB1, M. beijingense 4-1, and Methaomicrobiales. The band A8 was matched with

M. petrolearium. They can use H2/CO2 and formate as Methannospirillum hungatei, a hydrogenotrophic methanogen, substrate for growth at pH 6.5-8.6 [28, 36, 41]. Four whose optimal pH range for growth was 6.6-7.4 [20]. Band bands (A5-A7 and A10) were grouped into the order A12 was related to Methanofollis liminatans, which was

Methanosarcinales, showing 99-100% similarity with isolated from a fluidized bed reactor and can use H2/CO2 genus Methanosaeta, in which species are specialists in and formate as substrates [53]. Two bands (A11 and A13) utilizing acetate [34]. Methanosaeta species were previously showed 99% sequence similarity with Methanosarcina identified as the dominant acetotrophic methanogens in a barkeri, which can use H2/CO2 and formate as substrates,

J. Microbiol. Biotechnol. Syntrophic Propionate Degradation Response to pH Decrease 1417

and acetate with low affinity [21]. detectable hydrogenotrophic methanogens can endure a low It was obvious from Figs. 5 and 6 that the methanogenic pH (pH 5.5-4.0), but acetotrophic methanogens deceased composition was remarkably changed by pH decrease, significantly at pH ≤5.5. especially the hydrogenotrophic methanogens. In the sludge blanket, the bands A1, A2, and A9 (Methanobacterium) Acknowledgments were detected at pH 5.0-4.0, pH 6.0-4.0 and pH 6.8-5.0, respectively. The signal of band A8 (Methanospirillum) was This work was supported by the National Natural enhanced as pH decreased from 7.5 to 6.0 and then Science Foundation (51508316), Open Project of State Key stabilized. Bands A3 and A4 (Methanobacterium) maintained Laboratory of Urban Water Resource and Environment, similar signal intensity during the whole operational Harbin Institute of Technology (QA201523), and the Shanxi period. The Methanofollis species group (band A12) was Basic Research Program (2015021136). unique at pH 4.5 in the suspended layer. Although the pure cultures about hydrogenotrophic methanogens can grow at References pH 6.5-8.6, this study indicated that these methanogens can grow at pH as low as 4.5 [28, 36, 41]. These results 1. Amani T, Nosrati M, Mousavi SM, Kermanshahi RK. 2011. showed that most detectable hydrogenotrophic methanogens Study of syntrophic anaerobic digestion of volatile fatty can grow at low pH conditions. Similarly, Kim et al. [30] acids using enriched cultures at mesophilic conditions. Int. reported that the hydrogen-utilizing methanogens were more J. Environ. Sci. Technol. 8: 83-96. tolerant to acidic conditions than the other methanogens. 2. Ariunbaatar J, Panico A, Espositoa G, Pirozzi F, Lens PNL. 2014. Pretreatment methods to enhance anaerobic digestion Some hydrogenotrophic methanogens could be observed at of organic solid waste. Appl. Energy 123: 143-156. the acidogenic phase (pH 6.0-6.2) in an anaerobic baffled 3. APHA. 1995. Standard Methods for the Examination of Water reactor with four compartments treating sugar refinery and Wastewater. American Public Health Association. [10]. In addition, hydrogenotrophic methanogenesis existed 4. Ariesyady HD, Ito T, Okabe S. 2007. Functional bacterial in an acidic (pH 4.4) peat and attempts to obtain acetotrophic and archaeal community structures of major trophic groups methanogenic enrichment was unsuccessful [25]. Presumably, in a full-scale anaerobic sludge. Water Res. 41: 1554-1568. it is because the conversion of H2/CO2 to CH4 is energetically 5. Ariesyady HD, Ito T, Yoshiguchi K, Okabe S. 2007. Phylogenetic more favorable than that of acetate [34]. and functional diversity of propionate-oxidizing bacteria in an For acetotrophic methanogens, all members of the UASB digester sludge. Appl. Microbiol. Biotechnol. 75: 673-683. Methanosaeta species group (bands A5, A6, A7, and A10) 6. Astals S, Nolla-Ardèvol V, Mata-Alvarez J. 2012. Anaerobic were reduced with pH decrease. In the sludge blanket, the co-digestion of pig manure and crude glycerol at mesophilic bands A5, A6, and A7 disappeared at pH 5.5, 6.5, or 5.0, conditions: biogas and digestate. Bioresour. Technol. 110: 63-70. 7.Bai J, Liu H, Yin B, Ma H. 2015. Modeling of enhanced separately. However, bands A5 and A7 were observed at VFAs production from waste activated sludge by modified pH 5.5-4.5 in the suspended layer, consistent with a higher ADM1 with improved particle swarm optimization for pH (Fig. 2). The shift of band A11 (Methanosarcina) was parameters estimation. Biochem. Eng. J. 103: 22-31. similar to band A10, whereas band A13 was only detected 8. Ban Q, Li J, Zhang L, Zhang Y, Jha AK, Ai B. 2013. Effect of in the suspend layer. Although acetotrophic methanogens propionate concentration on degradation characteristics of a were reduced with pH decrease, the acetate concentration propionate enriched culture. J. Harbin Institute Technol. 45: in the effluent remained low. Acetate might be syntrophically 43-47. oxidized to CO2 by the concerted activity of acetate- 9. Ban Q, Li J, Zhang L, Jha AK. 2013. Syntrophic propionate oxidizing anaerobes and hydrogenotrophic methanogens degradation response to temperature decrease and microbial in this system [25]. community shift in an UASB reactor. J. Microbiol. Biotechnol. A better understanding of the microbial community 23: 382-389. succession with pH change in the UASB reactor will allow 10.Ban Q, Li J, Zhang L, Jha AK, Loring N. 2013. Linking performance with microbial community characteristics in an improvement of propionate removal and stability during anaerobic baffled reactor. Appl. Biochem. Biotechnol. 169: anaerobic digestion. The stepwise pH decrease from 7.5 to 1822-1836. 4.0 significantly inhibited propionate removal (98.6% to 11. Ban Q, Li J, Zhang L, Jha AK, Zhang Y. 2013. Quantitative 70.1%). PCR-DGGE analysis confirmed that propionate- analysis of previously identified propionate-oxidizing bacteria oxidizing bacteria (Pelotomaculum and Smithella) were and methanogens at different temperatures in an UASB remarkably reduced at low pH (≤5.5) conditions. Most reactor containing propionate as a sole carbon source. Appl.

August 2016 ⎪ Vol. 26⎪ No. 8 1418 Zhang et al.

Biochem. Biotechnol. 171: 2129-2141. Hydrogenotrophic methanogenesis by moderately acid-tolerant 12. Ban Q, Zhang L, Li J. 2015. Shift of propionate-oxidizing methanogens of methane-emitting acidic peat. Appl. Environ. bacteria with HRT decrease in an UASB reactor containing Microbiol. 69: 74-83. propionate as a sole carbon source. Appl. Biochem. Biotechnol. 26. Imachi H, Sekiguchi Y, Kamagata Y, Ohashi A, Harada H. 175: 274-286. 2000. Cultivation and in situ detection of a thermophilic 13. Barredo MS, Evison LM. 1991. Effect of propionate toxicity bacterium capable of oxidizing propionate in syntrophic on methanogen-enriched sludge. Appl. Environ. Microbiol. 57: association with hydrogenotrophic methanogens in a 1764-1769. thermophilic methanogenic granular sludge. Appl. Environ. 14. Bassam BJ, Caetano-Anollés G, Gresshoff PM. 1991. Fast and Microbiol. 66: 3608-3615. sensitive silver staining of DNA in polyacrylamide gels. 27. Imachi HS, Sakai S, Ohashi A, Harada H, Hanada S, Anal. Biochem. 196: 80-83. Kamagata SY, Sekiguchi Y. 2007. Pelotomaculum propionicicum 15. Boone DR, Xun LY. 1987. Effects of pH, temperature, and sp. nov., an anaerobic, mesophilic, obligately syntrophic, nutrients on propionate degradation by a methanogenic propionate-oxidizing bacterium. Int. J. Syst. Evol. Microbiol. enrichment culture. Appl. Environ. Microbiol. 53: 1589-1592. 57: 1487-1492. 16. Chen X, Sun Y, Xiu Z, Li X, Zhang D. 2006. Stoichiometric 28. Imachi H, Aoi K, Tasumi E, Saito Y, Yamanaka Y, Saito Y, et analysis of biological hydrogen production by fermentative al. 2011. Cultivation of methanogenic community from bacteria. Int. J. Hydrogen Energy 31: 539-549. subseafloor sediments using a continuous-flow bioreactor. 17. de Bok FAM, Harmsen HJM, Plugge CM, de Vries MC, ISME J. 5: 1913-1925. Akkermans ADL, de Vos WM, Stams AJM. 2005. The first 29. Keyser M, Witthuhn RC, Lamprecht C, Coetzee MP, Britz true obligately syntrophic propionate-oxidizing bacterium, TJ. 2006. PCR-based DGGE fingerprinting and identification Pelotomaculum schinkii sp. nov., co-cultured with Methanospirillum of methanogens detected in three different types of UASB hungatei, and emended description of the genus Pelotomaculum. granules. Syst. Appl. Microbiol. 29: 77-84. Int. J. Syst. Evol. Microbiol. 55: 1697-1703. 30. Kim IS, Hwang MH, Jang NJ, Hyun SH, Lee ST. 2004. Effect 18. Demirel B, Scherer P. 2008. The roles of acetotrophic and of low pH on the activity of hydrogen utilizing methanogen hydrogenotrophic methanogens during anaerobic conversion in bio-hydrogen process. Int. J. Hydrgen Energy 29: 1133-1140. of biomass methane: a review. Rev. Environ. Sci. Biotechnol. 7: 31. Kosaka T, Kat S, Shimoyama T, Ishii S, Abe T, Watanabe K. 173-190. 2008. The genome of Pelotomaculum thermopropionicum reveals 19. Dhaked RK, Waghmare CK, Alam SI, Kamboj DV, Singh L. niche associated evolution in anaerobic microbiota. Genome 2003. Effect of propionate toxicity on methanogenesis of Res. 18: 442-448. night soil at psychrophilic temperature. Bioresour. Technol. 32. Li J, Ban Q, Zhang L, Jha AK. 2012. Syntrophic propionate 87: 299-303. degradation in anaerobic digestion: a review. Int. J. Agric. 20. Ferry JG, Smith PH, Wolfe RS. 1974. PCR-based DGGE Biol. 14: 843-850. fingerprinting and identification of methanogens detected in 33. Liu Y, Balkwill DL, Aldrich HC, Drake GR, Boone DR. 1999. three different types of UASB granules. Int. J. Syst. Bacteriol. Characterization of the anaerobic propionate-degrading 24: 465-469. syntrophs Smithella propionica gen. nov., sp. nov. and 21. Fischer R, Thauer RK. 1990. Ferredoxin-dependent methane Syntrophobacter wolinii. Int. J. Syst. Bacteriol. 49: 545-556. formation from acetate in cell extracts of Methanosarcina 34. Liu Y, Whitman WB. 2008. Metabolic, phylogenetic, and barkeri (strain MS). FEBS Lett. 269: 368-372. ecological diversity of the methanogenic archaea. Ann. NY 22. Grabowski A, Tindall BJ, Bardin V, Blanchet D, Jeanthon C. Acad. Sci. 1125: 171-189. 2005. Petrimonas sulfuriphila gen. nov., sp. nov., a mesophilic 35. Lueders T, Pommerenke B, Friedrich MW. 2004. Stable-isotope fermentative bacterium isolated from a biodegraded oil probing of microorganisms thriving at thermodynamic reservoir. Int. J. Syst. Evol. Microbiol. 55: 1113-1121. limits: syntrophic propionate oxidation in flooded soil. Appl. 23. Harmsen HJH, Kengen MP, Akkermans ADL, Stams AJM, Environ. Microbiol. 70: 5778-5786. de Vos WM. 1996. Detection and localization of syntrophic 36. Ma K, Liu X, Dong X. 2005. Methanobacterium beijingense sp. propionate-oxidizing bacteria in granular sludge by in situ nov., a novel methanogen isolated from anaerobic digesters. hybridization using 16S rRNA-based oligonucleotide probes. Int. J. Syst. Evol. Microbiol. 55: 325-329. Appl. Environ. Microbiol. 62: 1656-1663. 37. Masahiro T, Takashi M, Yoshiyuki U, Masafumi G, Koji S. 24. Hong SW, Lim IK, Kim YW, Shin SM, Chung KS. 2013. 2008. Methanogenesis from acetate and propionate by Denaturing gradient gel electrophoresis and culture-based thermophilic down-flow anaerobic packed-bed reactor. analysis of the bacterial community in cheonggukjang, a Bioresour. Technol. 99: 4786-4795. Korean traditional fermented soybean food from Gangwon 38. McInerney MJ, Struchtemeyer CG, Sieber J, Mouttaki H, province. Korean J. Food Sci. Technol. 45: 515-520. Stams AJM, Schink B, et al. 2008. Physiology, ecology, 25. Horn MA, Matthies C, Küsel K, Schramm A. 2003. phylogeny, and genomics of microorganisms capable of

J. Microbiol. Biotechnol. Syntrophic Propionate Degradation Response to pH Decrease 1419

syntrophic metabolism. Ann. NY Acad. Sci. 1125: 58-72. acid inhibition of hydrogen fermentation by temperature 39. McMahon KD, Zheng D, Stams AJM, Mackie RI, Raskin L. shift strategy. Int. J. Hydrogen Energy 39: 19232-19241. 2004. Microbial population dynamics during start-up and 47. Sousa DZ, Smidt H, Alves MM, Stams AJM. 2007. overload conditions of anaerobic digesters treating municipal Syntrophomonas zehnderi sp. nov., an anaerobe that degrades solid waste and sewage sludge. Biotechnol. Bioeng. 87: 823-834. long-chain fatty acids in co-culture with Methanobacterium 40. Moertelmaier C, Li C, Winter J, Gallert C. 2014. Fatty acid formicicum. Int. J. Syst. Evol. Microbiol. 57: 609-715. metabolism and population dynamics in a wet biowaste 48.Suresh A, Seo C, Chang HN, Kim Y. 2013. Improved digester during re-start after revision. Bioresour. Technol. volatile fatty acid and biomethane production from lipid 166: 479-484. removed microalgal residue (LRμAR) through pretreatment. 41. Mori K, Harayama S. 2011. Methanobacterium petrolearium sp. Bioresour. Technol. 149: 590-594. nov. and Methanobacterium ferruginis sp. nov., mesophilic 49. Uyanik S. 2003. Granule development in anaerobic baffled methanogens isolated from salty environments. Int. J. Syst. reactor. Turkish J. Eng. Environ. Sci. 27: 131-144. Evol. Microbiol. 61: 138-143. 50. Wan C, Du M, Li DJ, Yang X, Zheng L. 2011. Electrokinetic 42. Narihiro T, Terada T, Ohashi A, Kamagata Y, Nakamura K, remediation and microbial community shift of β-cyclodextrin- Sekiguchi Y. 2012. Quantitative detection of previously dissolved petroleum hydrocarbon-contaminated soil. Appl. characterized syntrophic bacteria in anaerobic wastewater Microbiol. Biotechnol. 89: 2019-2025 treatment systems by sequence-specific rRNA cleavage 51. Worm P, Fermoso FG, Lens PNL, Plugge CM. 2009. method. Water Res. 46: 2167-2175. Decreased activity of a propionate degrading community in 43. Plugge CM, Balk M, Stams AJM. 2002. Desulfotomaculum a UASB reactor fed with synthetic medium without thermobenzoicum subsp. thermosyntrophicum subsp. nov., a molybdenum, tungsten and selenium. Enzyme Microb. Technol. thermophilic, syntrophic, propionate-oxidizing, spore-forming 45: 139-145. bacterium. Int. J. Syst. Evol. Microbiol. 52: 391-399. 52. Yang Y, Guo J, Hu Z. 2013. Impact of nano zero valent iron 44. Sekiguchi Y, Kamagata Y, Nakamura K, Ohashi A, Harada (NZVI) on methanogenic activity and population dynamics H. 1999. Fluorescence in situ hybridization using 16S rRNA- in anaerobic digestion. Water Res. 47: 6790-6800. targeted oligonucleotides reveals localization of methanogens 53. Zellner G, Boone DR, Keswan J, Whitman WB, Woese CR. 1999. and selected uncultured bacteria in mesophilic and thermophilic Reclassification of Methanogenium tationis and Methanogenium sludge granules. Appl. Environ. Microbiol. 65: 1280-1288. liminatans as Methanofollis tationis gen. nov., comb. nov. and 45. Shigematsu T, Era S, Mizuno Y, Ninomiya K, Kamegawa Y, Methanofollis lirninatans comb. nov. and description of a new Morimura S, Kida K. 2006. Microbial community of a strain of Methanofollis liminatans. Int. J. Syst. Bacteriol. 49: mesophilic propionate-degrading methanogenic consortium 247-255. in chemostat cultivation analyzed based on 16S rRNA and 54. Zheng D, Raskin L. 2000. Quantification of Methanosaeta acetate kinase genes. Appl. Microbiol. Biotechnol. 72: 401-415. species in anaerobic bioreactors using genus- and species- 46. Sivagurunathan P, Sen B, Liu CY. 2014. Overcoming propionic specific hybridization probes. Microb. Ecol. 39: 246-262.

August 2016 ⎪ Vol. 26⎪ No. 8