w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/watres
Relationship between phenol degradation efficiency and microbial community structure in an anaerobic SBR F. Rosenkranz a,b,c,*, L. Cabrol a, M. Carballa b, A. Donoso-Bravo a, L. Cruz a, G. Ruiz-Filippi a,c, R. Chamy a,c, J.M. Lema b a
School of Biochemical Engineering, Pontificia Universidad Cato´lica de Valparaı´so, General Cruz 34, Valparaı´so, Chile b Department of Chemical Engineering, Institute of Technology, University of Santiago de Compostela, E-15782 Santiago de Compostela, Spain c Fraunhofer Chile Research, Mariano Sa´nchez Fontecilla 310, Las Condes, Santiago, Chile
article info
abstract
Article history:
Phenol is a common wastewater contaminant from various industrial processes, including
Received 18 May 2013
petrochemical refineries and chemical compounds production. Due to its toxicity to microbial
Received in revised form
activity, it can affect the efficiency of biological wastewater treatment processes. In this study,
21 August 2013
the efficiency of an Anaerobic Sequencing Batch Reactor (ASBR) fed with increasing phenol
Accepted 2 September 2013
concentrations (from 120 to 1200 mg L1) was assessed and the relationship between phenol
Available online 13 September 2013
degradation capacity and the microbial community structure was evaluated. Up to a feeding
Keywords:
concentration (up to 180 mg L1 d1) and the elimination capacity remained relatively con-
Anaerobic digestion
stant around 27 mg phenol removed∙gVSS1 d1. Operation at higher concentrations
Phenol
(1200 mg L1) resulted in a still efficient but slower process: the elimination capacity and the
ASBR
initial degradation rate decreased to, respectively, 11 mg phenol removed∙gVSS1 d1 and
Inhibition
154 mg L1 d1. As revealed by Denaturing Gradient Gel Electrophoresis (DGGE) analysis, the
Community structure
increase of phenol concentration induced level-dependent structural modifications of the
DGGE
community composition which suggest an adaptation process. The increase of phenol con-
concentration of 800 mg L1, the initial degradation rate steadily increased with phenol
centration from 120 to 800 mg L1 had little effect on the community structure, while it involved drastic structural changes when increasing from 800 to 1200 mg L1, including a strong community structure shift, suggesting the specialization of the community through the emergence and selection of most adapted phylotypes. The thresholds of structural and functional disturbances were similar, suggesting the correlation of degradation performance and community structure. The Canonical Correspondence Analysis (CCA) confirmed that the ASBR functional performance was essentially driven by specific community traits. Under the highest feeding concentration, the most abundant ribotype probably involved in successful phenol degradation at 1200 mg L1 was affiliated to the Anaerolineaceae family. ª 2013 Elsevier Ltd. All rights reserved.
* Corresponding author. School of Biochemical Engineering, Pontificia Universidad Cato´lica de Valparaı´so, General Cruz 34, Valparaı´so, Chile. Tel.: þ56 (0)322274829. E-mail addresses:
[email protected],
[email protected],
[email protected] (F. Rosenkranz), juan.
[email protected] (J.M. Lema). 0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.watres.2013.09.004
6740
1.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Introduction
Phenol is an important chemical compound with a large number of applications and its global production is estimated to reach 6 million tons per year (Busca et al., 2008). Therefore it is present in wastewaters from various industrial processes including petrochemical refineries, coke production and chemical compounds production such as herbicides, pesticides, insecticides, antioxidants and paper additives, among others. Concentration of phenol in wastewaters depends on the industrial activities, from 6 to 500 mg L1 in petroleum refinery, to 28e3900 mg L1 in coke and 2.8e1220 mg L1 in petrochemical wastewaters (Leve´n et al., 2012). Phenolic compounds have been shown to be toxic to humans, animals and plants (Lin and Wu, 2011). A phenol concentration of 1 mg L1 affects aquatic life, and therefore, unacceptably high concentrations have the potential to contaminate receiving waters (Chang et al., 1995; Tay et al., 2001). For this reason, local legislations impose maximal limits of phenol concentration in discharge waters after treatment. For example, in Chile and Spain this limit is set between 0.5 and 1 g L1 in receiving ecosystems, depending on the industrial sector considered (Ministry Secretariat General of the Presidency, Chile; Ministry of Construction and Urban Development, Spain). Phenolic wastewaters may be treated by physicochemical (thermal decomposition and absorption) and biological processes (Chan et al., 2009). Thermal decomposition can be used to treat wastes with very high concentrations (higher than 15,000 mg L1), but it is only used at small scale due to the high energy demand (Portela et al., 2001). Absorption is an effective method for treating diluted phenolic wastewaters. However, due to the relatively high cost of activated carbon or other sorbents, adsorption cannot be used to treat high concentrated phenolic wastewaters (Mollah and Robinson, 1996; Halhouli et al., 1997). Under aerobic or anaerobic conditions, phenol can be degraded to harmless compounds by microorganisms (Chan et al., 2009). Biological aerobic treatment of phenol wastewater has been mostly based on conventional activated sludge systems, which have been reported to be affected in case of either fluctuating phenol loads or high phenol loading rates exceeding 1 kg phenol m3 d1 (Watanabe et al., 1996, 1999). Jiang et al. (2003) first documented the successful cultivation of aerobic granules with phenol as the sole carbon source at a loading rate of 1.5 kg phenol m3 d1. These phenol-degrading aerobic granules demonstrated a quite high tolerance to phenol, and the kinetic data indicated that they have the potential to treat industrial wastewaters with high phenol loads. However, these granules required a start-up period of three months, including two months for conditioning the municipal activated sludge seed through phenol incubation, and one month for the development of stable granules within the reactor (Tay et al., 2005). Different anaerobic technologies have been implemented to cope with the presence of phenol because of the advantages they offer over other biological operations: withstanding high organic loading rates and low sludge generation, in addition to energy production. However, due to its inhibitory effects on the activity of microbial degraders, phenol can induce adverse
effects in biological treatment processes, such as a decrease of the elimination rate and longer treatment times. Variations of operating parameters may induce changes of microbial community structure (density, diversity, activity) which in turns result in distinct functional performance (Cabrol and Malhautier, 2011). The negative effect of phenol on degradation efficiency can be physiological (metabolic inhibition) as well as structural (modification of community composition), and both effects can be tightly linked. Razo-Flores et al. (2003) reported a COD-removal efficiency higher than 90% in an UASB reactor operating at an organic loading rate (OLR) of 7 kg COD m3 d1 and treating a mixture of phenol and p-cresol as the main carbon and energy source. However, the treatment of wastewaters containing phenolic compounds as the sole substrates at high concentrations in UASB systems faced several limitations (relatively long acclimation period, small granule size and decrease in phenol removal efficiency at high loads, sensitivity to temperature and loading shocks, and long recovery periods after shocks, among others), thus requiring process improvements (Veeresh et al., 2005). Anaerobic Sequencing Batch Reactors (ASBR) have gained increasing popularity in recent years given their various advantages such as operational simplicity, efficient quality control of the effluent and flexibility of use (Tauseef et al., 2013). Specifically, ASBRs enable to uncouple the reaction time and the sludge retention time, thus favoring biomass retention. In addition, ASBRs offer some kinetic advantages over continuous reactors, such as a higher reaction rate at the beginning of the reaction since the substrate concentration is high, unlike in continuous reactor which always works at the outlet concentration (Ma et al., 2012; Singh and Srivastava, 2011; Donoso-Bravo et al., 2009). The technological potential of ASBR has been assessed for several types of effluents, including the ones generated by dairies, dye and textile industries, breweries, pulp mills, tanneries, petrochemical industries and other sources (Suresh et al., 2011; Mohan et al., 2005; Pasukphu and Vinitnantharat, 2002).Due to its particular working characteristic, ASBR can counteract the inhibitory effects of phenol and favor its degradation. When comparing different ASBR configuration modes for the treatment of phenol as sole carbon source (210 mg L1), Donoso-Bravo et al. (2009) obtained higher efficiency with fed-batch operation, reaching a complete phenol removal in a reaction time of 10 days. The aim of this paper is to study the efficiency of Anaerobic Sequencing Batch Reactors (ASBR) fed with mediumehigh phenol concentrations (from 120 to 1200 mg L1) and to assess the relationship between phenol degradation capacity and the structure and composition of the microbial community developed in the reactor.
2.
Materials and methods
2.1.
Biochemical methane potential (BMP)
BMP tests of phenol were carried out in triplicate in glass bottles of 400e500 mL-effective volume, under mesophilic
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
conditions (37 C), following the protocol of Soto et al. (1993). Two different types of carbon sources were used: only phenol and phenol þ glucose. The initial phenol and glucose concentrations were 200 8.5 mg L1 (0.48 0.02 g COD L1) and 1.4 7.9 g L1 (1.5 0.5 g COD L1), respectively. After phenol depletion, the medium was supplemented with a second and a third addition of phenol at 200 mg L1 (0.48 g COD L1). The initial biomass concentration was 1.5 gVSS L1 in all experiments. Two different inocula were used: one coming from a brewery wastewater treatment plant (WWTP) and another from a tobacco WWTP, with a methanogenic activity of 0.2e0.3 and 0.15e0.2 g CODCH4 m3 d1, respectively. Biogas production and composition were measured daily during the first six days after phenol addition, and every 2e3 days afterwards. Liquid samples were taken weekly for determination of the soluble chemical oxygen demand (CODs), phenol and volatile fatty acids (VFA) concentrations. The maximum degradation rate was calculated as the slope of the phenol concentration curve, within the linear zone (i.e. after the initial lag time).
2.2.
ASBR reactor
A laboratory-scale ASBR (15 cm-diameter and 34 cm height, representing a total volume of 6 L) with effective liquid and headspace volumes of 5 L and 1 L, respectively, was used. The digester was equipped with an external water jacket which allowed a constant temperature operation (37 C). The ASBR operation cycle included four steps: filling (20 min); reaction (variable duration enabling 90% of phenol degradation); settling (1 h); and discharge (20 min) with an exchange volume of 60%. During the reaction step, the liquid phase was recycled at a rate of 0.43 L min1 in order to enhance contact between biomass and liquid. Peristaltic pumps were used to control feed, discharge and recirculation flows. The reactor was fed with synthetic wastewater containing phenol at different concentrations (Table 1), 21.4 mg L1 of potassium dihydrogen phosphate and 4.3 mg L1 of ammonium chloride (Donoso-Bravo et al., 2009) and sodium bicarbonate to maintain alkalinity in a range 2.5e3.0 g CaCO3 L1. Sludge from a tobacco factory WWTP (20 gVSS L1) was used as inoculum. Biomass concentration in the reactor was kept approximately constant during the whole experiment at 12 g VSS L1 by doing sludge purges, every two weeks.
Table 1 e Substrate concentration and operation time of ASBR reactor for the different tested periods. Period Start up (Glucose 5 g COD L1) Acclimation (Phenol 120e240 mg L1) I (Phenol 120 60 mg L1) II (Phenol 240 4 mg L1) III (Phenol 500 50 mg L1) IV (Phenol 800 7 mg L1) V (Phenol 1200 43 mg L1)
Operation time (d) 0e21 (21 days) 0e80 (80 days) 81e120 121e139 140e161 162e195 196e281
(39 days) (18 days) (21 days) (33 days) (85 days)
Number of cycles 15e20 20 22 7 6 7 6
6741
Three different phases can be considered: (a) startup; (b) acclimation to phenol; and (c) testing period under increasing phenol concentrations. During startup, biomass was activated with glucose during 21 days (5 g COD L1). For acclimation, the reactor was fed with 120e240 mg$L-1of phenol during 80 days until it reached a steady degradation. Afterwards, phenol inlet concentration was progressively increased (120, 240, 500, 800 and 1200 mg L1) being each concentration maintained for 18e85 days, depending on the time required by the biomass to degrade at least 90% of the substrate (Table 1). ASBR reactor samples were taken daily during all operation cycles for the determination of phenol, CODs and VFA concentrations. Phenol degradation profile was further thoroughly determined by sampling every hour during the last operation cycle, for each phenol concentration tested.
2.3.
Analytical methods
CODs and alkalinity (total and partial) were measured following Standard Methods (APHA, 1995). VFAs and phenol were detected and quantified by a gas chromatograph GC-8A (Shimadzu, Kyoto, Japan), equipped with a 3000 4 mm ID packed column GP 60/80 Carbopack C/0.3% Carbowax 20M/ 0.1% H3PO4 (Sigma Aldrich, St Louis, MO, US). The analysis was carried out at 120 C, with nitrogen as carrier gas (50 mL min1) and FID detector (200 C). Methane concentration in the biogas was measured by gas chromatography using a Clarus 500 Gas Chromatograph (Perkin Elmer, Waltham, MA, US) equipped with filled Teflon column Hayesep-Q, Supelco; 4 m 1/800 OD, using FID detector. The specific phenol Elimination Capacity (EC) was calculated as the total degraded phenol amount divided by the total degradation time, relative to biomass content (mg 1 phenol.g1 VSS d ). In addition the initial phenol degradation rate was calculated as the slope of the first linear region of the degradation profiles (i.e. during the initial lag time).
2.4. Characterization of the microbial community structure For microbial community characterization, samples were taken from the ASBR reactor at the end of the last cycle of operation under each phenol concentration tested. After centrifugation for 10 min at 10,000 rpm, DNA was extracted from 450 mg of sludge pellet using Powersoil DNA Isolation Kit, MO BIO Laboratories (Carlsbad, CA, US). After gel electrophoresis and DNA quantification (BioSpec-Nano, Shimadzu), the hypervariable V3 region of bacterial 16S rDNA was PCRamplified using the primers w49F and w104R (Delbes et al., 2000) with a 40-bp GC clamp at the 50 -end of the forward primer (Muyzer et al., 1993). PCR amplification was carried out in 50 mL with 1 U Platinum Taq DNA Polymerase (Invitrogen, Life Technology, Carlsbad, CA, US), 1.5 mM MgCl2, 0.2 mM each primer, 0.5 mMdNTP and 100 ng of template DNA, in a MJ-Mini thermocycler (Bio-Rad Laboratories Inc, Hercules, CA, US) through an initial denaturation at 94 C for 2 min, followed by 30 cycles at 94 C (30 s), 57 C (40 s) and 72 C (40 s), and a final elongation at 72 C for 2 min. After gel electrophoresis and quantification (Quant-iTPico Greends DNA, Invitrogen), PCR
6742
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
2.5.
Statistical analysis
DGGE gel photograph was analyzed with Gel Compar II software (Applied Maths, Sint-Martens-Latem, Belgium) to obtain the matrix of relative band intensities according to band position. The pair-wise similarity between community profiles was calculated by the BrayeCurtis coefficient (Legendre and Legendre, 1998) and samples were hierarchically clustered by UPGMA linking dendrogram on the basis of their BrayeCurtis similarity. The potential correlations between the bacterial community structure (based on the band intensity matrix) and various environmental variables (EV, after standardization) was investigated by Canonical Correspondence Analysis (CCA), assuming an unimodal distribution of species (terBraak, 1986). CCA was computed with XLSTAT (Addin soft) and the relationship significance was tested by a permutation test (500 permutations). Moreover, a bubble representation was superimposed on the CCA plot, by allocating to each sample position a symbol whose size was proportional to the inlet phenol concentration at the corresponding sampling date. In the CCA biplot, the EVs are represented as arrows whose direction and length indicate, respectively, (i) the direction in which the EV explains most of the bacterial patterns and (ii) the magnitude of contribution of that EV in explaining variation in bacterial profiles.
3.
Results and discussion
3.1.
BMP tests
BMP tests were conducted as a preliminary evaluation of phenol biodegradation and assessment of inoculum and cosubstrate effects. Fig. 1 shows phenol concentration in BMP tests using tobacco inoculum and phenol as sole carbon source.
250 200 Phenol (mg·L-1)
products (500 ng) were separated by Denaturing Gradient Gel Electrophoresis (DGGE) on a 10% polyacrylamide gel with a linear gradient ranging from 25% to 65%, according to the protocol of Muyzer et al. (1993). The migration was carried out for 16 h at 120 V and 60 C, in 1X TAE buffer, using the DCode System (Bio-Rad Laboratories Inc, Hercules, CA, US). DGGE gels were stained with 1X Sybr Green I (Invitrogen, Life Technology, Carlsbad, CA, US). DGGE bands of interest were excised from the gels with a sterile razor blade, crushed in 40 ml ultrapure water, and kept overnight at 4 C. DNA was purified with the Favor Prep Gel/ PCR purification kit (Favorgen, Taiwan), and PCR-amplified as before. New DGGE was performed to verify the single-band profile of each PCR product. If necessary, the excision procedure was repeated until obtaining a single band which was finally excised from the gel and PCR-amplified without GCclamp. The PCR products were sent to sequencing (Macrogen Inc, Seoul, Korea) in aABI 3730XL sequencer with ABI PRISM BigDye Terminator Cycle sequencing kit and Ampli Taq DNA polymerase (Applied Biosystems). The NCBI BLAST software was used to identify putative close phylogenetic relatives within non-redundant Gen Bank database.
150 100 50 0
0
5
10
15
20
25 30 35 Time (d)
40
45
50
55
Fig. 1 e Phenol concentration in BMP tests after the first (B), second (-) and third (:) addition of phenol, using inoculum from a tobacco processing WWTP. Mean values and standard deviations were calculated from triplicates.
A lag time of 20e25 days was observed after the first phenol addition, and the degradation was completed after 40e45 days. After the second addition, the lag time decreased to 10 days and the degradation was completed in 20e38 days, thus revealing that biomass did adapt to phenol presence by exhibiting a faster consumption response. After the third phenol addition, the lag and total times were further reduced to 4 and 11 days, respectively. The progressive adaptation to repeated phenol additions resulted in a significant reduction of lag time period and increasing degradation rates (up to three times higher than the initial ones) after only three phenol additions. A similar degradation trend was obtained for brewery inoculum, with very similar lag times, and similar adaptation behavior to repeated substrate additions (Table 2). Both inocula enabled complete phenol degradation, although with different maximum and specific degradation rates (Table 2). As the inoculum from tobacco factory WWTP displayed a higher activity, it was selected for the following experiments in ASBR reactor. When glucose was fed as co-substrate in the first addition, phenol degradation rate and degradation times did not significantly change. For inocula from brewery and tobacco WWTP, lag time and total degradation were, respectively, 20/ 45 d and 25/45 d, while EC was 4.5 0.3 and 7.0 0.2 mg 1 phenol$g1 VSS d , respectively. For this reason, phenol was used as the sole carbon source for the following assays. The mass balance has been calculated for the brewery inoculum. It is closed with app 10e20% of error, with or without glucose. Anaerobic phenol degradation was studied in batch by Fang et al. (2006) at 55 C with concentrations varying from 50 to 1000 mg$L-1and using an inoculum from a UASB reactor treating phenol (630 mg L1). After an adaptation time (up to 20 days at 1000 mg L1), phenol was completely degraded in 30 days. At 200 mg phenol L1 (in the range of the present study), phenol degradation rate was 2.6 mg phenol L1 d1, after a lag phase of 5 days. Our results are similar to those of Fang et al. (2006) in terms of rapid inoculum adaptation to increasing phenol concentration, although the degradation
6743
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Table 2 e Lag time and total degradation time, and maximum and specific degradation rates in batch assays inoculated with WWTP biomass from brewery or tobacco WWTP, after the first, second and third phenol additions.
Lag time/total degradation time (d)
Brewery Tobacco
Maximum degradation rate (mg phenol L1 d1)
Brewery Tobacco
Specific degradation rate (mg phenol$gVSS1 d1)
Brewery Tobacco
1st
2nd
3rd
20/45 25/40 6.3 0.2 10.8 0.3 4.2 0.2 7.2 0.3
10/38 10/20 7.6 0.4 13.5 0.3 5.1 0.4 9.0 0.3
ND 4/11 ND 35.5 0.5 ND 23.7 0.5
ND (not determined).
3.2.
Phenol degradation in ASBR
A) 1400 1200 1000 800 600 400 200
0 0
20
40
0
5
10
60
80
100 120 140 160 180 200 220
B) 1000
800
Phenol (mg L-1)
During the start-up phase, glucose degradation time was less than 24 h (data not shown). Afterwards, acclimation to phenol was carried out during 20 cycles with a total duration of 80 days. In the last cycle, phenol (120 mg L1) was completely depleted in 24 h, indicating a high capacity of the biomass to cope with phenol as the sole carbon source. No VFA accumulation was detected during this period and pH kept around 7.5e8.0. Once the biomass was adapted to phenol, the initial phenol concentration in the reactor was increased stepwise up to 1200 43 mg phenol L1. The anaerobic biomass required approximately 4e6 cycles of operation to adapt to each phenol concentration, as observed by the decrease of the reaction time during the first
cycles (Fig. 2). Afterwards, the reaction time remained quite constant during the following cycles at each inlet concentration. Fig. 3 shows the degradation profiles of phenol during the reaction phase of the last cycle of operation at each phenol concentration. Increasing phenol concentration resulted in a lengthening of the reaction time necessary to degrade at least 90% of the inlet phenol (Fig. 4). The reaction time increased linearly up to a feeding concentration of 800 mg L1. For phenol concentration above 800 mg L1 up to 1200 mg L1, the reaction time increased much more (higher slope on Fig. 4),
Phenol (mg L-1)
rates obtained in both studies depended upon the inoculum source. In previous studies, the presence of glucose as co-substrate had contradictory effects on phenol degradation: in various cases glucose did not improve phenol degradation (Tay et al., 2001; Donoso-Bravo et al., 2009), while others suggest that the presence of easily degradable substrates may facilitate, by cometabolism, the degradation of recalcitrant substrates such as phenol (Kennes et al., 1997; Mehrotra et al., 2003; Veeresh et al., 2005). In the present study, the co-metabolism effects may be negligible, since the degradation times of both substrates were very different: less than 1 day for glucose and at least 11 days for phenol.
600
400
200
0 15
20
25
30
35
40
45
50
55
Time (h)
Fig. 2 e Phenol concentration in consecutive ASBR cycles at different feeding concentrations (120, 240, 500, 800 and 1200 mg phenol LL1).
Fig. 3 e Phenol concentration profiles during the last cycle of ASBR operation at different feeding phenol concentrations: 120 (B), 240 (-), 500 (:), 800 (C) and 1200 (A) mg phenol LL1. A) All phenol concentrations; B) Zoom of low and intermediate phenol concentrations.
6744
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Fig. 4 e OLR (-) and total degradation time (D) during the last cycle at different feeding phenol concentrations in ASBR reactor. The dotted line represents the average OLR.
probably due to an inhibitory effect, which required to increase the reaction period and operate the ASBR at lower OLR. For all tested concentrations, two degradation rates ranges could be distinguished: an initial rate was observed during the initial period (lag time), different from the subsequent one observed during the linear range. During the first experiments at 120 ad 240 mg L1, the initial rate was strongly lower than the subsequent one (Fig. 3), indicating that an initial adaptation was necessary. For higher phenol concentrations up to 800 mg L1, the difference between the initial rate and the subsequent one got attenuated, as a result of the progressive increase of the initial degradation rate with increasing phenol concentration (Fig. 5). However, when phenol reached 1200 mg L1, the initial rate drastically decreased. Lag phase for all phenol concentrations was observed, which could be explained by the following hypothesis: i) If phenol is not the preferential substrate, the biomass can adopt a stressresponse strategy: storage of carbon/energy sources during phenol degradation of cycle n; consumption of this stored reserves during the lag phase of cycle n þ 1; beginning of highrate phenol degradation once the stored reserves are depleted. ii) If the enzymes responsible for phenol hydrolysis are extracellular, they may be washed during the reactor discharge step between each cycle, implying that they have to be synthesized again at the beginning of each cycle.
Fig. 5 e Phenol Elimination Capacity (A) and initial degradation rate (B) during the last cycle of ASBR operation at different feeding phenol concentrations. The dotted line represents the average EC.
The Elimination Capacity remained relatively constant 1 (around 26 5 mg phenol removed$g1 VSS d ) for the low and intermediate phenol concentrations up to 800 mg L1 (Fig. 5). Afterwards it decreased down to 11 mg phenol remov-1 1 ed$g1 VSS$d at the highest phenol concentration (1200 mg L ). This suggests the existence of a threshold value between 800 and 1200 mg L1 after which a slowdown of the reaction rate took place. Most of the anaerobic treatments of phenolic effluents have been carried out in continuous systems, mainly in UASB reactors (Fang et al., 2004, 2006) and in Expanded Granular Sludge Blanket (EGSB) reactors (Scully et al., 2006). The duration of the start-up is reported to range between 6 weeks and 10 months, depending on the nature of inocula and the operational strategy (Veeresh et al., 2005). The acclimation time obtained in the present ASBR reactor (80 days) was comparable to the low range of the previously reported acclimation times in continuous reactors. The 80 days of acclimation period in the present ASBR were sufficient to favor the adaptation of biomass to higher substrate concentrations, enabling the elimination of 90% of phenol. Muftah et al. (2010) evaluated the aerobic SBR technology with Pseudomonas putida immobilized on polyvinyl alcohol (PVA), reaching phenol degradation rates of 1152, 2059 and 1800 mg phenol L1 d1 at initial phenol concentrations of 40, 100 and 190 mg L1, respectively. These values are higher than those obtained in the present study, which may be explained by the operation with a pure culture and at lower concentrations of phenol. Moreover, Muftah et al. (2010) observed a strong inhibition at 190 mg phenol L1(similar to the concentration used in the acclimation period of this study), which indicates the higher efficiency but the lower tolerance of that pure culture.
3.3.
Bacterial community structure
The analysis of bacterial community by DGGE is limited by the quantitative bias inherent to the different steps of the methodology (non-exhaustive DNA extraction, preferential PCR amplification, DGGE detection limit and saturation, multiple operon sequences and co-migration) (Muyzer et al., 1993; Muyzer and Smalla, 1998). However, despite these limitations, DGGE complies with our objective to compare the community structures-based on dominant members-between samples undergoing the same methodological steps (Loisel et al., 2006). In the ASBR under study, the total community structure revealed by DGGE was dynamic with time and clustered according to time and feeding phenol concentration (Figs. 6 and 7). The three samples corresponding to the Acclimation phase (day < 80) clustered together, strongly apart from the other ones. The acclimation phase was the period with most dynamism and highest bacterial diversity, in terms of DGGE band number (Fig. 6). This segregation evidenced that the acclimation to phenol was a structural process (not only physiological), driven by deep changes of community diversity and composition. Between the acclimation period and the test period, the DGGE band number decreased from 31 3 to 25 3 (equivalent to Shannon index from 3,1 0,1 to 2,6 0,2), due to the disappearance of the most sensitive species.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
6745
Fig. 6 e UPGMA clustering based on BrayeCurtis similarity from 16S-DGGE profiles of ASBR community samples. The numbers indicate the sampling day, and the testing phase is indicated in parenthesis: A: Acclimation phase; I: 120 ± 60 mg phenol LL1; II: 240 ± 4 mg phenol LL1; III: 500 ± 50 mg phenol LL1; IV: 800 ± 7 mg phenol LL1; V: 1200 ± 43 mg phenol LL1. The letters on the DGGE gel indicate the excised and sequenced bands of interest.
Afterwards, during the test phase, the diversity kept rather constant (25 3 DGGE bands, equivalent to Shannon index 2,6 0,2). However, the community structure was variable. Three different groups could be discriminated on the dendrogram representation, according to the feeding
Fig. 7 e Canonical Correspondence Analysis (CCA) of microbial community patterns generated by 16S-DGGE analysis of 11 ASBR community samples (circles). The size of the circles was proportional to the inlet phenol concentration. The numbers indicate the sampling day, and the testing phase is indicated in parenthesis (as in Fig. 6). The first and second axes represent resp. 40.9% and 37.6% of the total variance. The selected environmental variables (arrows) explained 59% of the inertia of the DGGE data set. Among all the individual variables (DGGE bands) only the excised and sequenced ones are represented (diamond symbols and letter as in Fig. 6).
concentration: the samples corresponding to the lowest phenol concentration (120 mg L1, period I), to intermediate phenol concentrations (240e800 mg L1, periods II to IV), and to the highest phenol concentration (1200 mg L1, period V) (Fig. 6). The increase of phenol concentration was accompanied by a strong community structure shift, suggesting the specialization of the community through the emergence and selection of most adapted phylotypes, which can be either simply tolerant species (no consuming phenol) or degrader species (playing a significant role in the maintenance of functional capacity). On the CCA representation (Fig. 7), the samples corresponding to low and intermediate phenol concentrations (120e800 mg L1, period I to IV) were closely grouped together, and only the samples corresponding to the highest phenol concentration (1200 mg L1, period V) got strongly separated. The influence of phenol concentration on the community structure was level-dependent: the increase of phenol concentration from 120 to 800 mg$L-1had no significant effect on the community structure, while it involved drastic structural changes when increasing from 800 to 1200 mg L1. The threshold of structural disturbance was similar to the functional threshold observed previously for the functional parameters (degradation time, Elimination Capacity and initial degradation rate) (Figs. 4 and 5), suggesting that the destabilization of degradation performance between 800 and 1200 mg L1 was related to a destabilization of the community structure.
3.4. Relationships between bacterial community structure and process functioning The set of environmental variables (EV) tested for CCA analysis was: the feeding phenol concentration, the total degradation time, the initial degradation rate, and the elimination capacity. The CCA permutation test revealed a significant
6746
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
linear relation between the abundance data (community structure) and the EVs (p-value 0.014). Among the most significant relationships, the structure of the community which developed under 1200 mg phenol L1 was positively correlated to the degradation time (longer) and negatively correlated to the elimination capacity. On the contrary, the structures of the communities which developed between 120 and 800 mg phenol L1 were equitably correlated to both the elimination capacity and the initial degradation rate. The influence of environmental conditions, mainly the contaminant concentration, on bacterial community structure is now well-established in various ecosystem. Previous studies have shown that repeated phenol additions have effects on community structure (Guieysse et al., 2001) Microbial succession and population specialization have been reported under the selection pressure of other toxic aromatic compounds at increasing concentrations (Trably et al., 2008). However the reverse (i.e., to what extent the community characteristics drive the macroscopic function) is still a critical question for ecologists (Cabrol and Malhautier, 2011). Here, we bring clear evidence that the higher or lower degradation performance (in terms of phenol degradation rate and elimination capacity) were essentially driven by specificities of the community structure (species composition and relative abundances).
3.5.
Species composition
As a first approximation, the five major bands of the DGGE profiles corresponding to different phenol concentrations were sequenced. The successive band excision and reamplification from DGGE profiles provided pure band patterns after a second DGGE run, and high-quality unambiguous sequences were obtained by Sanger sequencing. The BLAST search revealed that four out of the five band sequences were closely related (>97% sequence similarity) to known families and/or genera (Table 3). In most cases (4/5), the best matching sequences according to BLAST program were retrieved from anaerobic digestion processes, many of them related with hydrocarbon degradation. Band A was the most intense band in DGGE profiles corresponding to the highest phenol concentration (days 249e271, period V) and was barely detected in other profiles corresponding to lower phenol concentrations. It showed 99% sequence-similarity with an uncultured Anaerolineaceae bacterium clone (within the Chloroflexi phylum) identified in an anaerobic toluene-degrading microbial community, but not responsible for toluene degradation (Sun and Cupples, 2012).
Members of the Anaerolineae class are obligate anaerobes widely distributed in various types of natural and artificial anaerobic ecosystems. Some cultivable strains could be isolated from anaerobic digesters and rice paddy soil, but phenol as sole substrate (1 mM) did not support the growth of these cultured strains (Yamada et al., 2007). Members of the Anaerolineaceae family were significantly enriched in oil-amended microcosms and involved in oil degradation under sulfatereducing conditions (Sherry et al., 2013). In soils from a diesel-contaminated railway site, the presence of contamination correlated with increased relative abundances of the phyla Chloroflexi and detection of Anaerolineae-like OTUs, which are known to be associated with anaerobic degradation of oil-related compounds (Sutton et al., 2013). Bands B and C were also present in DGGE profiles corresponding to the V period, although they were more intense in DGGE profiles corresponding to acclimation and I period. Band B showed 97% sequence-similarity with a Spirochaetaceae clone (within the Spirochaetes phylum) identified from the early stages of enrichment cultivation of crude oil-contaminated soil under alkane amendment, but its concerted function could not be ruled out (Cheng et al., 2013). In addition, band B also displayed 97% similarity with various uncultured bacterium clones which, although not closely related to known species, were identified from an anaerobic fluidized bed reactor treating 2,4,6-trichlorophenol and phenol (accession number DQ306716), and from a methanogenic phenoldegrading enrichment culture (accession number GQ377468, Leven et al., 2010); however, this clone was not considered to be the main responsible of phenol degradation in the latter enrichment culture. Band C showed 100% sequence-similarity with an uncultured Clostridium sp. clone (within the Firmicutes phylum) identified from a petroleum hydrocarbon contaminated aquifer (Table 3). Band C had also 99% sequence similarity with a Clostridium clone identified among the most frequent community members in sediments from an infiltration basin receiving highway runoff (accession number JQ177694, Rotaru et al., 2012). Clostridiales are fermentative organisms thought to use hydrocarbon intermediates to generate organic acid precursors (e.g. acetate, formate) and have been significantly enriched in oil-amended microcosms where they were involved in oil degradation (Sherry et al., 2013). Clostridia members were identified (by Stable Isotope Probing) as the putative toluene degraders in sulfate-amended agricultural soil microcosms (Sun and Cupples, 2012). Clostridiaceae were also identified among the most abundant OTU enriched from an oil-utilising methanogenic consortium (Gieg
Table 3 e Blast affiliation of the sequences from the five dominant bands excised from the DGGE profiles as indicated on Fig. 6. DGGE band
Closest BLAST match (accession number)
Similarity %
Identification source
A B
Uncultured Anaerolineaceae bacterium clone (JN806351) Spirochaetaceae bacterium clone (HQ689210)
99 97
C D
Uncultured Clostridium sp. clone (JQ087143) Uncultured bacterium clone (FR823667)
100 100
E
Uncultured Syntrophaceae bacterium clone (JN806341)
100
Anaerobic toluene-degrading microbial communities. Methanogenic consortium from crude oil-contaminated soil. Hydrocarbon-contaminated aquifer. Anaerobic methanogenic reactor for brewery wastewater treatment. Anaerobic toluene-degrading microbial communities.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
et al., 2008), and as the most efficient microorganisms responsible for the anaerobic primary oxidation of benzene in an iron-reducing enrichment culture (Kunapuli et al., 2007). Bands D and E were low-intensity bandsin the profiles corresponding to period V. Band D showed 100% sequence similarity with uncultured and unclassified bacterium clones identified from various anaerobic digestion processes treating municipal sludge (accession number CU918284, Riviere et al., 2009) or food-processing wastes (accession number GU389559, Nelson et al., 2012), among others. Band E showed 100% sequence similarity with an uncultured Syntrophaceae bacterial clone (within the Deltaproteobacteria class) identified from an anaerobic toluene-degrading microbial community, but not responsible for toluene degradation (Sun and Cupples, 2012). Band E had also 99% sequence similarity with an uncultured Deltaproteobacteria clone which was probably responsible for the syntrophic degradation of 4-methylbenzoate in a methanogenic consortium (accession number AF254389, Wu et al., 2001). In addition, band E displayed 99% similarity with various uncultured bacterium clones which, although not closely related to known species, were identified from anaerobic diethyl phthalate-and benzoate-degrading UASB reactor (accession number EF053105) and a phenol-degrading consortium from an anaerobic reactor treating industrial wastewater (accession number EF590284) (Liang et al., 2007). Although the exact function of the community members corresponding to each DGGE band cannot be reliably ruled out from 16S rDNA-DGGE band sequencing, we could distinguish between two groups of dominant species: the three community members corresponding to bands C, D, E (Clostridium-like Unclassified and Syntrophaceae-like) were putative phenol degraders under low phenol concentrations, with high phenol degradation rate but low tolerance above 800 mg L1; the two community members corresponding to bands A and B (Spirochaetaceae-like and Anaerolineaceae-like) were putative phenol degraders under high phenol concentrations, with low phenol degradation rate but high tolerance above 800 mg L1. We can assume that the Anaerolineaceae-like most abundant ribotype present in ASBR under the highest feeding concentration was most likely involved in successful phenol degradation at 1200 mg L1, and emerged through a progressive adaptation to increasing phenol concentrations, as it was not dominant under lower phenol concentrations.
4.
Conclusions
The glucose co-metabolism hypothesis for phenol degradation is discarded in our system. The ASBR appeared to be an attractive technological option for anaerobic phenol degradation. The high removal performance obtained, despite the high phenol concentration, may be a result of the operation strategy based on the progressive increase of feeding phenol concentration. This operation strategy could be applied at full scale for effective treatment of high load phenolic wastewaters. Biomass adaptation to increasing phenol concentration was not only physiological but also structural. After strong community shifts and diversity decrease during the acclimation phase, the community structure got rather stable up
6747
to 800 mg L1 of phenol, allowing in turn high degradation performance to be reached. Above a phenol threshold around 1 g L1, a strong composition change occurred, selecting for Anaerolineaceae-like species tolerant to this high phenol concentration (usually reported as inhibitory), which correlated to a still efficient but lower degradation rate and elimination capacity.
Acknowledgments The authors would like to thank Fondecyt Project 1110861 for providing funding to this project, Xunta de Galicia through the postdoctoral contract of Dr. Marta Carballa (Isidro Parga Pondal, IPP-08-37) and Banco Santander-USC through Predoctoral fellowships for researchers from Latinoamerica of Francisca Rosenkranz.
references
APHA, 1995. Methods for Examination of Water and Wastewater, nineteenth ed. American Public Health Association/American Water Works Association/Water Environment Federation, Washington, DC, USA. Busca, G., Berardinelli, S., Resini, C., Arrighi, L., 2008. Technologies for the removal of phenol from fluid streams: a short review of recent developments. Review. J. Hazard. Mater. 160 (2e3), 265e288. Cabrol, L., Malhautier, L., 2011. Integrating microbial ecology in bioprocess understanding: the case of gas biofiltration. Appl. Microbiol. Biotechnol. 90 (3), 837e849. Chan, Y.J., Chong, M.F., Law, C.L., Hassell, D.G., 2009. A review on anaerobiceaerobic treatment of industrial and municipal wastewater. Chem. Eng. J. 155 (1e2), 1e18. Chang, Y.-J., Nishio, N., Nagai, S., 1995. Characteristics of granular methanogenic sludge grown on phenol synthetic medium and methanogenic fermentation of phenolic wastewater in a UASB reactor. J. Ferment. Bioeng. 79 (4), 348e353. Cheng, L., Rui, J., Li, Q., Zhang, H., Lu, Y., 2013. Enrichment and dynamics of novel syntrophs in a methanogenic hexadecanedegrading culture from a Chinese oilfield. Feder. Europ. Microbiol. Soc. Microb. Ecol. 83, 757e766. Delbes, C., Moletta, R., Godon, J.J., 2000. Monitoring of activity dynamics of an anaerobic digester bacterial community using 16S rRNA polymerase chain reaction-single-strand conformation polymorphism analysis. Environ. Microbiol. 2 (5), 506e515. Donoso-Bravo, A., Rosenkranz, F., Valdivia, V., Ruiz-Filippi, G., Chamy, R., 2009. Anaerobic sequencing batch reactor as an alternative for the biological treatment of wine distillery effluents. Water Sci. Technol. 60 (5), 1155e1160. Fang, H.H.P., Liu, Y., Ke, S.Z., Zhang, T., 2004. Anaerobic degradation of phenol in wastewater at ambient temperature. Water Sci. Technol. 49 (1), 95e102. Fang, H., Lianga, D.W., Zhanga, T., Liu, Y., 2006. Anaerobic treatment of phenol in wastewater under thermophilic condition. Water Res. 40 (3), 427e434. General Secretary Ministry of the Presidency of Chile, 2001. Decree N 90. Emission Standard for Control of Pollutants Associated with Liquid Waste Discharges to Surface Waters and Inland Marine. Gieg, L.M., Duncan, K.E., Suflita, J.M., 2008. Bioenergy production via microbial conversion of residual oil to natural gas. Appl. Environ. Microbiol. 74 (10), 3022e3029.
6748
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Guieysse, B., Wikstro¨m, P., Forsman, M., Mattiasson, B., 2001. Biomonitoring of continuous microbial community adaptation towards more efficient phenol-degradation in a fed-batch bioreactor. Appl. Microbiol. Biotechnol. 56, 780e787. Halhouli, K.A., Darwish, N.A., Al-Jahmany, Y., 1997. Effects of temperature and inorganic salts on the adsorption of phenol from multicomponent systems onto a decolorizing carbon. Separ. Sci. Technol. 32 (18), 3027e3036. Jiang, H., Fang, Y., Fu, Y., Guo, Q., 2003. Studies on the extraction of phenol in wastewater. J. Hazard. Mater. 101 (2), 179e190. Kennes, C., Mendez, R., Lema, J.M., 1997. Methanogenic degradation of p-cresol in batch and in continuous UASB reactors. Water Res. 31 (7), 1549e1554. Kunapuli, U., Lueders, T., Meckenstock, R., 2007. The use of stable isotope probing to identify key iron-reducing microorganisms involved in anaerobic benzene degradation. Int. Soc. Microb. Ecol. J. 1, 643e653. Legendre, P., Legendre, L., 1998. Second English Edition. Numerical Ecology, Elsevier B.V, Amsterdam. Leve´n, L., Nyberg, K., Schnu¨rer, A., 2010. Molecular characterisation of two anaerobic phenol-degrading enrichment cultures. Int. Biodeter. Biodegr. 64, 427e433. Leve´n, L., Nyberg, K., Schnu¨rer, A., 2012. Conversion of phenols during anaerobic digestion of organic solid waste e a review of important microorganisms and impact of temperature. J. Environ. Manage. 95, 99e103. Liang, D.W., Zhang, T., Fang, H.H., 2007. Anaerobic degradation of dimethyl phthalate in wastewater in a UASB reactor. Water Res. 41 (13), 2879e2884. Lin, Y., Wu, C., 2011. Sensitivity analysis of phenol degradation with sulfate reduction under anaerobic conditions. Environ. Model. Assess. 16 (2), 213e225. Loisel, P., Harmand, J., Zemb, O., Latrille, E., Lobry, C., Delgene`s, J.-P., Godon, J.-J., 2006. Denaturing gradient electrophoresis (DGGE) and single-strand conformation polymorphism (SSCP) molecular fingerprintings revisited by simulation and used as a tool to measure microbial diversity. Environ. Microbiol. 8 (4), 720e731. Ma, J., Yu, L., Frear, C., Zhao, Q., Li, X., Chen, S., 2012. Kinetics of psychrophilic anaerobic sequencing batch reactor treating flushed dairy manure. Bioresour. Technol. 131, 6e12. Mehrotra, I., Kumar, P., Gali, V., 2003. Treatment of phenolic wastewater using upflow anaerobic sludge blanket reactor. In: Proceedings of National Conference on Biological Treatment of Wastewater and Waste Air, August 28 and 29. Regional Research Laboratory (CSIR), Trivandrum, India. Ministry of Construction and Urban Development, Spain, 1986. Real Decree 849. Annex to Part IV. Class 2. Extraction, Preparation and Agglomeration of Solid Fuels and Coke. Tables of the Characteristic Parameters to Consider, at Least in the Estimation of Discharge Treatment. Mohan, S.V., Rao, N.C., Prasad, K.K., Madhavi, B.T.V., Sharma, P.N., 2005. Treatment of complex chemical wastewater in a sequencing batch reactor (SBR) with an aerobic suspended growth configuration. Proc. Biochem. 40 (5), 1501e1508. Mollah, A.H., Robinson, C.W., 1996. Pentachlorophenol adsorption and desorption characteristic of granular activated carbon. Water Res. 30 (12), 2901e2910. El-Naas, Muftah H., Al-Zuhair, Sulaiman, Makhlouf, Souzan, 2010. Batch degradation of phenol in a spouted bed bioreactor system. J. Indust. Eng. Chem. 16 (2), 267e272. Muyzer, G., Smalla, K., 1998. Application of denaturing gradient gel electrophoresis (DGGE) in microbial ecology. Antonie van Leeuwenhoek 73 (1), 127e141. Muyzer, G., de Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-
amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59 (3), 695e700. Nelson, M., Morrison, M., Schanbacher, F., Yua, S., 2012. Shifts in microbial community structure of granular and liquid biomass in response to changes to infeed and digester design in anaerobic digesters receiving food-processing wastes. Bioresour. Technol. 107, 135e143. Pasukphu, N., Vinitnantharat, S., 2002. Degradation of organic substances and reactive dye in an immobilized-cell sequencing batch reactor operation on simulated textile wastewater. In: 5th International Conference on Small Water and Wastewater Treatment Systems. Portela, J.R., Nebot, E., Mart, E., 2001. Kinetic comparison between subcritical and supercritical water oxidation of phenol. Chem. Eng. J. 81 (1e3), 287e299. Razo-Flores, E., Iniestra-Gonzalez, M., Field, J.A., Olguin- Lora, P., Puig-Grajales, L., 2003. Biodegradation of mixtures of phenolic compounds in an upward-flow anaerobic sludge blanket reactor. J. Environ. Eng. 129 (11), 999e1006. Rivie`re, D., Desvignes, V., Pelletier, E., Chaussonnerie, S., Guermazi, S., Weissenbach, J., Li, T., Camacho, P., Sghir, A., 2009. Towards the definition of a core of microorganisms involved in anaerobic digestion of sludge. Int. Soc. Microb. Ecol. 3, 700e714. Rotaru, C., Woodard, T.L., Choi, S., Nevin, K.P., 2012. Spatial heterogeneity of bacterial communities in sediments from an infiltration basin receiving highway runoff. Microb. Ecol. 64, 461e473. Scully, C., Collins, G., O’Flaherty, V., 2006. Anaerobic biological treatment of phenol at 9.5-15 degrees C in an expanded granular sludge bed (EGSB)-based bioreactor. Water Res. 40 (20), 3737e3744. Sherry, A., Gray, N.D., Ditchfield, A.K., Aitken, C.M., Jones, D.M., Ro¨ling, W.F.M., Hallmann, C., Larter, S.R., Bowler, B.F.J., Head, I.M., 2013. Anaerobic biodegradation of crude oil under sulphate-reducing conditions leads to only modest enrichment of recognized sulphate-reducing taxa. Int. Biodeter. Biodegr. 81, 105e113. Singh, M., Srivastava, R.K., 2011. Sequencing batch reactor technology for biological wastewater treatment: a review. Asia-pacific J. Chem. Eng. 6 (1), 3e13. Soto, M., Mendez, R., Lema, J.M., 1993. Methanogenic and nonmethanogenic activity tests: theoretical basis and experimental setup. Water Res. 27 (8), 1361e1376. Sun, W., Cupples, A.M., 2012. Diversity of five anaerobic toluene degrading microbial communities investigated using stable isotope probing (SIP). Appl. Environ. Microbiol. 78, 972e980. Suresh, S., Tripathi, R.K., Gernal Rana, M.N., 2011. Review on treatment of industrial wastewater using sequential batch reactor. Int. J. Sci. Technol. Manage. 2, 64e84. Sutton, N.B., Maphosa, F., Morillo, J.A., Abu Al-Soud, W., Langenhoff, A.A., Grotenhuis, T., Rijnaarts, H.H., Smidt, H., 2013. Impact of long-term diesel contamination on soil microbial community structure. Appl. Environ. Microbiol. 79 (2), 619e630. Tauseef, S.M., Abbasi, T., Abbasi, S. a, 2013. Energy recovery from wastewaters with high-rate anaerobic digesters. Renew. Sust. Energy Rev. 19, 704e741. Tay, J.H., He, Y.X., Yan, Y.G., 2001. Improved anaerobic degradation of phenol with supplemental glucose. J. Environ. Eng. 127 (1), 38e45. Tay, S.T.-L., Moy, B.Y.-P., Jiang, H.-L., Tay, J.-H., 2005. Rapid cultivation of stable aerobic phenol-degrading granules using acetate-fed granules as microbial seed. J. Biotechnol. 115 (4), 387e395. terBraak, C.J.F., 1986. Canonical correspondence analysis: a new eigenvector technique for multivariate direct gradient analysis. Ecology 67 (5), 1167e1179.
w a t e r r e s e a r c h 4 7 ( 2 0 1 3 ) 6 7 3 9 e6 7 4 9
Trably, E., Batstone, D., Christensen, N., Patureau, D., Schmidt, J., 2008. Microbial dynamics in anaerobic enrichment cultures degrading di-n -butyl phthalic acid ester. Feder. Europ. Microbiol. Soc. Microbiol. Ecol. 66, 472e483. Veeresh, G., Kumar, P., Mehrotra, I., 2005. Treatment of phenol and cresols in upflow anaerobic sludge blanket (UASB) process: a review. Water Res. 39 (1), 154e170. Watanabe, K., Hino, S., Takahashi, N., 1996. Responses of activated sludge to an increase in phenol loading. J. Ferment. Bioeng. 82 (5), 522e524. Watanabe, K., Teramoto, M., Harayama, S., 1999. An outbreak of non flocculating catabolic populations caused the breakdown
6749
of a phenol-digesting activated sludge process. Appl. Environ. Microbiol. 65 (7), 2813e2819. Wu, J., Liu, W., Tseng, I., Cheng, S., 2001. Characterization of a 4-Methylbenzoate-degrading methanogenic consortium as determined by small-subunit rDNA sequence analysis. J. Biosci. Bioeng. 91 (5), 449e455. Yamada, T., Imachi, H., Ohashi, A., Harada, H., Hanada, S., Kamagata, Y., Sekiguchi, Y., 2007. Bellilinea caldifistulae gen Nov., sp. nov. and Longilinea arvoryzae gen. nov., sp. nov., strictly anaerobic, filamentous bacteria of the phylum Chloroflexi isolated from methanogenic propionate-degrading consortia. Int. J. Syst. Evol. Microbiol. 57, 2299e2306.
