Aerobic granulation strategy for bioaugmentation treating high strength pyridine wastewater

Journal of Hazardous Materials 295(2015)153–160

Contents lists available at ScienceDirect

Journal of Hazardous

Materials

j o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /j h a z m a

t

Aerobic granulation strategy for bioaugmentation of a sequencing batch reactor (SBR)treating high strength pyridine wastewater

Xiaodong Liu a ,Yan Chen a ,Xin Zhang a ,b ,Xinbai Jiang a ,Shijing Wu a ,Jinyou Shen a ,?,Xiuyun Sun a ,Jiansheng Li a ,Lude Lu a ,Lianjun Wang a ,??

a

Jiangsu Key Laboratory for Chemical Pollution Control and Resources Reuse,School of Environmental and Biological Engineering,Nanjing University of Science and Technology,Nanjing 210094,Jiangsu Province,China b

Suzhou Institute of Architectural Design Co.,Ltd,Suzhou 215021,Jiangsu Province,China

g r a p h i c a l

a b s t r a c

t

a r t i c l e

i n f o

Article history:

Received 26August 2014

Received in revised form 6March 2015Accepted 9April 2015

Available online 13April 2015

Keywords:

Aerobic granule Bioaugmentation Biodegradation

High-throughput sequencing

a b s t r a c t

Aerobic granules were successfully cultivated in a sequencing batch reactor (SBR),using a single bacterial strain Rhizobium sp.NJUST18as the inoculum.NJUST18presented as both a good pyridine degrader and an ef?cient autoaggregator.Stable granules with diameter of 0.5–1mm,sludge volume index of 25.6±3.6mL g ?1and settling velocity of 37.2±2.7m h ?1,were formed in SBR following 120-day cultivation.These granules exhibited excellent pyridine degradation performance,with maximum volumetric degradation rate (V max )varied between 1164.5mg L ?1h ?1and 1867.4mg L ?1h ?1.High-throughput sequencing anal-ysis exhibited a large shift in microbial community structure,since the SBR was operated under open condition.Paracoccus and Comamonas were found to be the most predominant species in the aerobic granule system after the system had stabilized.The initially inoculated Rhizobium sp.lost its dominance during aerobic granulation.However,the inoculation of Rhizobium sp.played a key role in the start-up process of this bioaugmentation system.This study demonstrated that,in addition to the hydraulic selection pressure during settling and ef?uent discharge,the selection of aggregating bacterial inocula is equally important for the formation of the aerobic granule.

?2015Elsevier B.V.All rights reserved.

?Corresponding author.Tel.+862584303965;fax:+862584303965.??Corresponding author.Tel.+862584315941;fax:+862584315941.

E-mail addresses:shenjinyou@https://www.sodocs.net/doc/248402305.html, (J.Shen),wanglj@https://www.sodocs.net/doc/248402305.html, (L.Wang).

https://www.sodocs.net/doc/248402305.html,/10.1016/j.jhazmat.2015.04.0250304-3894/?2015Elsevier B.V.All rights reserved.

154X.Liu et al./Journal of Hazardous Materials295(2015)153–160

1.Introduction

Aerobic granules have been employed in treating toxic and recalcitrant organics,such as phenol,p-nitrophenol and2,4-dichlorophenol,exhibiting high resistance to these toxic and recalcitrant compounds[1].The high toxic resistance and degra-dation ability of the aerobic granules are partially attributed to the unique structure of granules,where the dense microbial cells and extracellular polymeric substances(EPS)set a barrier for mass transfer and lower the concentration of toxics on the inner cells [2].Considering the compact structure,large microbial density and high toxic resistance,aerobic granules might be the core of bioaug-mentation[1].In the formation of aerobic granules,hydraulic selection pressure to select and retain compact microbial aggre-gates plays a key role.In addition,cell aggregation including autoaggregative or coaggregative interactions may be promoted for aerobic granulation.In the previous studies,aerobic granu-lation from coaggregative or autoaggregative phenol-degrading strains has been reported for enhanced phenol biodegradation [3–5].However,detailed information regarding the roles that coag-gregation/autoaggregation played in aerobic granulation has not yet been fully investigated.

Although,several works have been reported to isolate ef?cient microbial species for pyridine biodegradation[6],most of the work reported on this heterocyclic pollutant degradation pertained to ?ask level experiments.Only limited reports with technological orientation are available by now[7].Reports on aerobic granula-tion aiming at the bioremediation of pyridine waste are rare[8]. In addition,formation of pyridine-degrading aerobic granules from pure cultures capable of degrading pyridine has not previously been explored to our knowledge.

Therefore,the present study aimed to investigate the feasibil-ity of aerobic granulation from Rhizobium sp.NJUST18,which was both a functional pyridine degrader and a good autoaggregator. The effectiveness of the aerobic granules in treating high strength pyridine wastewater was evaluated,and the bacterial community structure of the aerobic granules was also included.

2.Materials and methods

2.1.Microorganism and synthetic wastewater

The isolation and cultivation of Rhizobium sp.NJUST18was described in our previous study[9].The16S rRNA sequence (comprising1381nucleotides)was deposited in the GenBank database under accession no.JN106368.The synthetic in?uent of the sequential batch reactor(SBR)was as follows:Na2HPO4?12H2O (3.057g L?1),KH2PO4(0.743g L?1),NH4Cl(1g L?1),MgSO4?7H2O (0.89g L?1),KCl(0.35g L?1),CaCl2(0.20g L?1),FeCl3(0.03g L?1), pyridine and sodium acetate at desired concentrations,and 10mL L?1trace element solution SL-4[10].

2.2.Aerobic granules cultivation

The aerobic granules were cultivated in a column-type sequen-tial batch reactor(SBR)with inner diameter of6cm,height of 100cm and effective volume of2.2L.The sequential operation of the reactors was automatically controlled by the timers.During the ?lling phase,the synthetic in?uent was introduced through ports located at the bottom of the bioreactor.While,?ne air bubbles for aeration were supplied by means of air bubble diffusers placed at the bottom of the SBR at volumetric?ow rate of0.2m3h?1and super?cial gas velocity of0.02m s?1.The ef?uent was withdrawn through the outlet ports positioned at the medium height in this column reactor,resulting in volumetric exchange ratio of50%.The bioreactor was operated in a thermostatic bath to keep cultivation temperature stable at30?C.

The SBR was initially seeded with2g(measured in dry weight) Rhizobium sp.NJUST18pure culture and fed with the synthetic in?uent.After inoculation,the SBR was operated in open condition, with the in?uent and the air introduced without sterilization.Oper-ational parameters for different phases of the cultivation procedure are presented in Table1.During the start-up period,i.e.,stage I (1–35days),the bioreactor was operated at a cycle of24h:5min of feeding without stirring,23.5h of aerobic reaction,12min of settling,3min of ef?uent withdrawal and10min of idling.For the?rst week,in addition to1000mg L?1pyridine,1000mg L?1 acetate sodium was added into synthetic in?uent as supplementary substrate for biomass growth stimulation.One week later,acetate sodium in synthetic in?uent was reduced progressively,and?nally reduced to0mg L?1on day30.Thereafter,in?uent pyridine con-centration increased to4000mg L?1step by step,while the cycle time and settling time progressively decreased to6h and2min, respectively.

120days later,mature granules were observed in SBR,with stable operation of the SBR achieved.In order to investigate the effect of pyridine concentration on pyridine biodegradation in the bioaugmented SBR,1000–8000mg L?1pyridine was added into the synthetic in?uent to make initial pyridine in SBR over the concen-tration range between500and4000mg L?1.

In order to reveal the key role of Rhizobium sp.NJUST18 during the startup of the bioaugmented SBR,a SBR inoculated with common activated sludge was operated as a control.The common activated sludge used was taken from the secondary sed-imentation tank of a municipal sewage treatment plant located in Nanjing.The control SBR was initially seeded with6g acti-vated sludge(measured in dry weight)and was operated under the same conditions as the bioaugmented SBR,except for the inocula used.Speci?cally,1000mg L?1pyridine and1000mg L?1 acetate sodium was added into the synthetic in?uent as the car-bon sources for the?rst week.One week later,acetate sodium in synthetic in?uent was reduced to800mg L?1while pyri-dine remained the same as the?rst week.However,due to the complete failure,the control SBR stopped running after2 weeks.

2.3.Analytical methods

Pyridine was quanti?ed by HPLC(Waters2996,Waters Incor-poration,USA)through authentic standard.Before analysis,water samples were passed through a0.22?m?lter.The HPLC anal-ysis was conducted at room temperature using a Waters RP18 column(5?m,4.6×250mm)and a UV–vis detector.The mobile phase was a mixture of70%methanol and30%water pumped at a?ow rate of 1.00mL/min.The analysis was performed at 254nm,with column temperature at35?C.Measurements of the parameters,such as sludge volume index(SVI),mixed liquor sus-pended solid(MLSS)and mixed liquor volatile suspended solids (MLVSS),were carried out according to APHA standard methods [11].The settling velocity of the sludge was measured by record-ing the traveling time of individual granules from the top to the bottom of a cylinder[12].The granule size was manually mea-sured using a?ne-scale ruler[13].Scanning electron microscopy (SEM)observation was carried out according to Ho et al.[4]. Transmission electron microscopy(TEM)observation was carried out according to Bhattacharya et al.[14].The ef?uent samples for the determination of the residual TOC concentration were sampled2h after the complete exhaustion of the pyridine.TOC concentration was measured using a Germany Elementar vario TOC analyzer.

X.Liu et al./Journal of Hazardous Materials295(2015)153–160155 Table1

Operational parameters applied during different cultivation phases.

Stage I II III IV V VI

Time(day)1–3536–4546–5556–6566–8586–120 In?uent pyridine concentration(mg L?1)100015002000200040004000

In?uent NaAC concentration(mg L?1)1000–000000

Cycle duration(h)241212866 Settling time(min)12105332 Aeration time(min)1410692697459339340 Pyridine loading rate(kg m?3d?1)0.5 1.5 2.0 3.08.08.0

2.4.DNA extraction,PCR and High-throughput sequencing

The microbial community structure of the aerobic granules in the SBR system was pro?led using high-throughput sequenc-ing technology.About1g sample of mature aerobic granules was collected at the end of stage VI.DNA of the aerobic granules was extracted from the aerobic granule sample using the FastDNA?SPIN for soil kit(MP Biomedicals,CA,USA)according to the manufac-turer’s instruction.The concentration and purity of genomic DNA were measured using NanoDrop?ND-1000(NanoDrop

Technolo-

Fig.1.TEM of Rhizobium sp.NJUST18(a),biomass2weeks after inoculation(b),biomass14weeks after inoculation(c),biomass18weeks after inoculation(d),and SEM of mature granule(e and f).

156X.Liu et al./Journal of Hazardous Materials295(2015)

153–160

Fig.2.The MLSS,MLVSS and settling time pro?les during120days’operation. gies,Willmington,DE,USA)to ensure the sample concentration higher than20ng/?L and A260/A280between1.80and2.00.

The extracted DNA was ampli?ed by polymerase chain reaction(PCR)using TaKaRa Ex Taq?(TaKaRa Bio Japan) according to the manufacturer’s instruction.Primers 27F(5 -AGAGTTTGATYMTGGCTCAG-3 )and338R(5 -TGCTGCCTCCCGTAGGAGT-3 )which targeted V1/V2hyper-variable regions of bacterial16S rRNA genes were selected.In order to tag the PCR products from each sample,speci?c8bases long identi?er were included in both forward and reverse primers. For sequencing the PCR products,speci?c sequencing adaptors were attached to both primers.The extracted DNA was diluted to 20ng/?L and used as the template DNA.Thermal cycling conditions were as follows:an initial denaturation at98?C for5min,and20 cycles at98?C for30s,50?C for30s,and72?C for40s,with a?nal extension at72?C for10min.

Four parallel PCR products were pooled and puri?ed by OMEGA E.Z.N.A Cycle-Pure Kit(Omega Bio-Tek Norcross,USA).The purity of puri?ed PCR product was measured using NanoDrop?ND-1000 once again to ensure its A260/A280between1.70and1.90.The puri?ed PCR product was quanti?ed precisely using Qubit2.0Flu-orometer(Life Technologies,Grand Island,USA).

The puri?ed library was diluted,denatured,rediluted,mixed with PhiX(equal to30%of?nal DNA amount)as described in the Illumina library preparation protocols,and then applied to an Illu-mina Miseq system in Jiangsu Zhongyijinda Analytical&Testing Co., Ltd.for sequencing with the Reagent Kit v22×250bp as described in the manufacture manual.DNA library building and data analysis were performed in according to Liang et al.[15]and Pala-Ozkok et al.[16].

3.Results

3.1.Formation of pyridine aerobic degrading granule

TEM image of Rhizobium sp.NJUST18(Fig.1a)showed that strain NJUST18was a rod-shaped bacterium with?agella,which were assumed to correspond closely to the noted capability of autoag-gregation.One week after the inoculation of the autoaggregative NJUST18,both biomass increase and obvious aggregation in SBR were observed.Two weeks later,the biomass in SBR showed?uffy, irregular,loose-structure morphology(Fig.1b).Some small and white granules could be observed,however,these granules were easily broke up into small pieces if placed under vigorous shaking. 14weeks after inoculation,with the increase of in?uent pyridine concentration,decrease of the cycle time and settling time,the ?ocs-like sludge gradually decreased and was replaced by

small

Fig.3.The settling velocities and SVI pro?les during120days’operation. granules with diameter of0.2–0.5mm,while the color of the gran-ules gradually changed from white to yellowish brown(Fig.1c).In the following4weeks,the granules became denser and more reg-ular in shape under high shear force which induced the biomass aggregates to secrete more exopolysaccharides[17,18].Through the control of settling time,more?occulent sludge was washed out from the SBR,resulting in the accumulation of the aerobic granules with high settling velocity.After18weeks of inoculation,mature granules(Fig.1d)with size of0.5–1mm were formed,leading to a stable operation of the SBR.The mature granules turned out to be smooth,with a solid surface(Fig.1d and e).Through SEM obser-vation,rod-shaped bacteria were found to be dominated at the granule surface(Fig.1f),indicating that rod-shaped strains played an important role in the granule formation and granule stabiliza-tion.

3.2.Biomass pro?le and settling properties of pyridine aerobic degrading granules

Since the settling time is a critical parameter and signi?-cantly in?uences the granulation process,a short settling time was generally considered favorable for the granule formation [2].Fig.2shows the MLSS,MLVSS during the development period of the pyridine aerobic degrading granules.During the stage I,the settling time was?xed at12min.MLSS and MLVSS increased from initial878.2±30.8mg L?1and853.6±45.3mg L?1 to3582.0±63.8mg L?1and3451.1±194.5mg L?1on day30, respectively.Decreasing the settling time from12min to10min on stage II and then to5min on stage III caused the slight decrease of MLSS and MLVSS,probably due to the biomass washout in the ef?uent.Then with the increase of in?uent pyridine con-centration from2000mg L?1to4000mg L?1on stage V,MLSS and MLVSS increased continuously to5197.4±132.8mg L?1and 4608.6±171.1mg L?1on day120,respectively.As the sludge?ocs were transformed into the granules,the SVI decreased all the time, from125.1±12.5mL g?1on day15to?nal25.6±3.6mL g?1on day 120(Fig.3).However,settling velocity increased all the time,from initial1.4±0.3m h?1to?nal37.2±2.7m h?1on day120(Fig.3). The settling velocity of the mature granular sludge developed in this study fell in the range of36.6±8.8m h?1reported by Su and Yu[19].

3.3.Pyridine removal ef?ciency of pyridine aerobic degrading granules

The removal ef?ciency of pyridine in the SBR system from begin-ning until the end of granules development period is illustrated in

X.Liu et al./Journal of Hazardous Materials295(2015)153–160

157

Fig.4.Pyridine degradation during120days’operation.

Fig.4.At stage I of the operation,the pyridine concentration in the ef?uent decreased from432.9±30.0mg L?1on day2to0mg L?1on day5,presumably due to the adapting process of the biomass with pyridine containing wastewater.36days later,with the increase of in?uent pyridine concentration from1000mg L?1to1500mg L?1 and the decrease of the cycle duration from24h to12h,the pyri-dine removal ef?ciency was?uctuating,with the ef?uent pyridine concentration increased to the maximal356.6±35.3mg L?1on day 36.From then on,pyridine concentration in the ef?uent decreased, with the ef?uent quality improved greatly.On day56,the ef?uent pyridine concentration increased to166.7±18.4mg L?1suddenly. The deterioration of the reactor performance could be attributed to the wash out of the biomass from the SBR,which was caused by the sudden decrease of the settling time from5min to3min. Thereafter,as the evolution of?occulent sludge into granular sludge gradually took place in the bioreactor system,the degra-dation ability for pyridine removal had been improved,with removal ef?ciency at around100%,although the in?uent pyridine concentration was increasing all the time.3.4.Enhanced pyridine biodegradation by aerobic granules

The aerobic granules could ef?ciently degrade pyridine over ini-tial concentration up to4000mg L?1(Fig.5a).At initial pyridine concentration around500mg L?1,degradation kinetics followed closely zero-order kinetics without obvious lag phase.With the increase of initial pyridine concentration,obvious lag phase was observed,which could be attributed to the high toxicity and recalcitrance of pyridine.However,around4000mg L?1pyridine could be completely degraded within7.5h.For the pure Rhizobium sp.NJUST18,although2600mg L?1pyridine could be completely removed in batch reactor,the incubation time needed was as long as240h(Fig.5b).The maximum volumetric degradation rate (V max,mg L?1h?1)of pyridine was modeled with the integrated Gompertz equation,according to Shen et al.[9].The calculated V max for pure Rhizobium sp.NJUST18was in the range from 4.2mg L h?1to32.4mg L?1h?1as the initial pyridine concentra-tions within the range of100–2600mg L?1.However,in this aerobic granular system,as the initial pyridine concentration within the range of500–4000mg L?1,the calculated V max varied between 1164.5mg L?1h?1and1867.4mg L?1h?1,which was rather high compared with that for pure Rhizobium sp.NJUST18.

What is more,there is signi?cant difference in terms of resid-ual TOC concentrations between aerobic granular system and pure Rhizobium sp.NJUST18system(Fig.6).As the initial pyri-dine concentrations within the range of100–2600mg L?1,the residual TOC concentration varied between65.7±4.2mg L?1and 156.8±8.9mg L?1in the pure Rhizobium sp.NJUST18system.As for the aerobic granular system,as the initial pyridine concentrations within the range of500–4000mg L?1,the residual TOC concen-tration varied between17.2±4.6mg L?1and74.5±11.8mg L?1, which was rather low compared with that for pure Rhizobium sp. NJUST18system.

3.5.Microbial community structure and dominant species

analysis

In this study,the microbial community structure of the mature aerobic granules in the SBR system was analyzed by high through-put sequencing based on Illumina Miseq system,with the

dominant https://www.sodocs.net/doc/248402305.html,parison of pyridine degradation performance between aerobic granules(a)and pure Rhizobium sp.NJUST18(b).

158X.Liu et al./Journal of Hazardous Materials295(2015)

153–160

https://www.sodocs.net/doc/248402305.html,parison of TOC removal between aerobic granules and pure Rhizobium sp.NJUST18.

bacteria revealed.As shown in Fig.7,bacterial sequences af?liated with Proteobacteria(53.41%)were the most abundant,followed by the sequences af?liated with Bacteroidetes(25.12%),unclassi?ed phylum(9.99%),Chlorobium(4.69%),Minor phyla(2.34%),Chlo-ro?exi(2.29%)and Actinobacteria(2.17%).At the genus level,the majority of dominant populations belonged to Paracoccus(55.54%), Comamonas(26.78%),Paludibacter(3.44%),Gemmatimonas(1.65%), Rhizobium(1.21%)and Propionicimonas(1.12%).The bacterial rich-ness and diversity was relatively low,probably due to the selective pressure placed on the bacterial community by high loading of pyridine.

4.Discussion

The?agella of NJUST18cells were assumed to correspond closely to the noted capability to autoaggregate.High autoaggrega-tion of Acinetobacter calcoaceticus was also observed by Adav and Lee[3],the authors attributed the high autoaggregation potential of A.calcoaceticus to the interconnecting?brils.Strains with?ag-ella,interconnecting?brils or mycelium were commonly found in the aggregates[20,21].The inoculation of the aggregating bacte-rial strains enhanced the granulation process,probably due to the formation of primary matrixes through aggregation[3,5].Liu and Tay[22]had pointed out that the appearance of primary matrixes was a very crucial step in initiating activated sludge granulation. Sludge granulation could be accelerated if the primary matrix could be provided rapidly.Thus,it could be inferred that the autoaggre-gation activity of Rhizobium sp.NJUST18cells played a signi?cant role in aerobic granulation,probably due to the small and white granules formed within2weeks after the startup of this aerobic granular system.In fact,in the pure Rhizobium sp.NJUST18system, the small and white granules could also be observed,which might served as the primary matrixes during granulation.In addition to the hydraulic selection pressure during settling and ef?uent dis-charge,the selection of aggregating bacterial inocula was equally important on the aerobic granule formation.

The settling property of the mature aerobic granules formed in this study was rather excellent,as was indicated by the low SVI and the high settling velocity values observed after the system had stabilized.The SVI value of25.6±3.6mL g?1at the end of stage VI(day120)was relatively low,compared with that in the aero-bic granule systems described in the literature,which varied in the range of40–120mL g?1[12,23].This result indicated that the set-tling properties of the aerobic granules developed in this study were rather good,which was favorable for the operation of the wastew-ater treatment plant.The high settling velocity of37.2m h?1had given signi?cant impact on the biomass retention in the SBR reac-tor[18].Despite the short settling time,the high settling velocity possessed by the developed microbial granular sludge enabled the granules to escape from being?ushed out during the decanting phase.Such conditions had caused more microbial granular sludge to be retained in the system and resulted in the increase of biomass concentration in the reactor.

The excellent pyridine degradation performance of the mature aerobic granules formed in this study was revealed by both the high V max and the high tolerance toward pyridine.The excellent degra-dation performance in terms of the rather high pyridine removal ef?ciency indicated the high biological activity occurred during microbial aerobic degradation process of pyridine wastewater in the SBR,even when the in?uent pyridine concentration was as high as4000mg L?1.The V max for mature aerobic granules was much higher than the V max for the pure Rhizobium sp.NJUST18.In addi-tion,the residual TOC concentration in the ef?uent of the aerobic granular system was much lower than that of the pure Rhizobium sp. NJUST18system.Low residual TOC concentration observed in the aerobic granular system suggested that intermediate metabolite leakage was not evident.It was probably due to the large micro-bial density,high toxic resistance and relatively high community diversity of aerobic granules.In addition,the compact structure of aerobic granules protected the microbes inside from the inhibitory effects of the target compounds[1].

Aerobic granules reported in this study were able to degrade pyridine at initial concentration up to4000mg L?1within

only Fig.7.Phylogenetic distribution of sequences assigned on phylum(a)and genus(b).

X.Liu et al./Journal of Hazardous Materials295(2015)153–160159

7.5h,which was an advantage for the treatment of pyri-dine containing wastewater at relatively high concentrations. In a completely mixed activated sludge process seeded with P. pseudoalcaligenes-KPN capable of degrading pyridine,an increase of the in?uent pyridine concentration to400mg L?1resulted in the reactor failure to degrade pyridine[24].In an acclimated pyri-dine biodegradation system based on aerobic granules initially for phenol degradation,at pyridine concentration higher than 3000mg L?1,pyridine degradation rate was rather low[8].The V max reported in this literature was as low as63.7mg L?1h?1,imply-ing a strong inhibitory effect of pyridine on the granules.Although Liu et al.[25]have reported that4250mg pyridine L?1could be degraded completely in an aerobic immobilized microbial system, the hydraulic retention time(HRT)required was as long as36h. During the whole operation period of about6months in this study, stable treatment performance was observed in the SBR dominated by aerobic granules.This further corroborated the robustness of the granular technology for application in the treatment of wastewa-ters containing highly toxic and recalcitrant organics.

Since the bioaugmented SBR was operated in open condition, other strains from the air or the in?uent would survive in the bioaugmented SBR.Thus,the signi?cant shift of the microbial com-munity structure was very likely,although the pure Rhizobium sp. NJUST18was initially inoculated in to the SBR as the sole inocula. It was worth noting that the initially inoculated pyridine degrader, Rhizobium sp.NJUST18,was not the most predominant strain of the pyridine degrading aerobic granules after the system had stabilized. The loss of the dominance of the inoculated strain was consistent with several previous studies.For instance,Wen et al.[26]found that the seeded pyridine degrading strain Paracoccus denitri?cans W12was not the dominant species in a bioaugmented MBR treating pharmaceutical wastewater.Bai et al.[27]also did not?nd the inoc-ulated bacteria capable of degrading pyridine and quinoline in the bioaugmented zeolite-BAF for coking wastewater treatment.How-ever,in a biaugmented aerobic granular system for2-?uorphenol degradation,the initially seeded strain capable of degrading2-?uorphenol,i.e.,Rhodococcus sp.strain FP1,could be successfully recovered after months of exposure to2-?uorphenol[28].Thus,the fate and dynamics of the inoculated degrading bacteria could be the integrated result of various environmental parameters,which would be rather complex[26].However,considering the rapid and successful start-up of the SBR based aerobic granules,it could be inferred that Rhizobium sp.NJUST18played a key role in the start-up of this bioaugmentation system.The success of the bioaug-mentation by Rhizobium sp.NJUST18was also con?rmed by the complete failure of a control experiment carried out in our lab using the common activated sludge as the sole inoculum for pyridine wastewater treatment.Almost no pyridine was degraded in the control SBR inoculated with activated sludge for the whole exper-iment period of2weeks.In addition,serious biomass?ushing was observed in the control SBR,as was indicated by the decreasing MLSS.The complete failure of the control SBR was probably due to the poor biodegradability of pyridine and the high toxicity of pyridine toward the activated sludge.

The genera Paracoccus,which was the most predominant in the mature aerobic granules,might be mainly responsible for the pyridine degradation because pyridine was effectively removed in the aerobic granular system at rather high pyridine loading.Pre-vious studies had revealed that Paracoccus sp.was an excellent pyridine degrader[29–31].The dominance of the genera Coma-monas was probably related with its strong aggregation ability. The genus Comamonas had polar?agella and may be contributing to the noted granule stability[32].Aerobic granulation and phe-nol degradation in the reactor bioaugmented with Comamonas sp. PG-08was signi?cantly enhanced compared with a control reactor [5].Comamonas sp.was also found to be the dominant species of the bacterial community in a bioaugmented MBR treating pyridine-containing pharmaceutical wastewater[26].In fact,the presence of the aggregation-induced strains could help the formation of micro-bial aggregates and the removal of recalcitrant compounds,even if they did not participate in critical degradation steps.

In our laboratory,six bacterial strains,including Paracoccus thio-philus NJUST24and Comamonas sp.NJUST25,have been isolated from the mature aerobic granules and used as the inocula for aer-obic granulation.Further study is needed to examine the dynamic change of the microbial community structure during the start-up and long-term operation of this aerobic granluation system, with the role of Paracoccus sp.and Comamonas sp.emphasized.In addition,other identi?ed bacteria such as Paludibacter and Gemma-timonas,which was also found from a pyridine anoxic degradation system in our laboratory,should also play an important role in pyri-dine biodegradation.Their detailed function needs to be further studied through selective degradation experiments.

5.Conclusions

Cultivation of stable aerobic granules was achieved in a SBR inoculated with an autoaggregative pyridine-degrader,Rhizobium sp.NJUST18.The aerobic granules could degrade pyridine at extremely high V max,demonstrating excellent pyridine degra-dation performance.Bacterial community analysis revealed that Paracoccus and Comamonas were the most predominant species in the aerobic granular system after the system had stabilized.The initially inoculated Rhizobium sp.lost its dominance during aero-bic granulation.However,the inoculation of Rhizobium sp.played a key role in the start-up of this bioaugmentation system.In addition to the hydraulic selection pressure during settling and ef?uent dis-charge,the selection of aggregating bacterial inocula was equally important on the aerobic granule formation.

Acknowledgements

This research is?nanced by Major Project of Water Pollu-tion Control and Management Technology of P.R.China(No. 2012ZX07101-003-001),National Natural Science Foundation of China(No.51478225)and Zijin Intelligent Program of NJUST(No. 2013-ZJ-02-19).

References

[1]A.M.Maszenan,Y.Liu,W.J.Ng,Bioremediation of wastewaters with

recalcitrant organic compounds and metals by aerobic granules,Biotechnol.

Adv.29(2011)111–123.

[2]L.Liu,G.P.Sheng,W.W.Li,Z.H.Tong,R.J.Zeng,J.X.Liu,J.Xie,S.C.Peng,H.Q.Yu,

Cultivation of aerobic granular sludge with a mixed wastewater rich in toxic organics,Biochem.Eng.J.57(2011)7–12.

[3]S.S.Adav,D.J.Lee,Single-culture aerobic granules with Acinetobacter

calcoaceticus,Appl.Microbiol.Biotechnol.78(2008)551–557.

[4]K.L.Ho,B.Lin,Y.Y.Chen,D.J.Lee,Biodegradation of phenol using

Corynebacterium sp.DJ1aerobic granules,Bioresour.Technol.100(2009)

5051–5055.

[5]H.L.Jiang,J.H.Tay,A.M.Maszenan,S.T.L.Tay,Enhanced phenol

biodegradation and aerobic granulation by two coaggregating bacterial

strains,Environ.Sci.Technol.40(2006)6137–6142.

[6]J.Q.Sun,L.Xu,Y.Q.Tang,F.M.Chen,W.Q.Liu,X.L.Wu,Degradation of pyridine

by one Rhodococcus strain in the presence of chromium(VI)or phenol,J.

Hazard.Mater.191(2011)62–68.

[7]K.V.Padoley,S.N.Mudliar,R.A.Pandey,Heterocyclic nitrogenous pollutants in

the environment and their treatment options-an overview,Bioresour.

Technol.99(2008)4029–4043.

[8]S.S.Adav,D.J.Lee,N.Q.Ren,Biodegradation of pyridine using aerobic granules

in the presence of phenol,Water Res.41(2007)2903–2910.

[9]J.Y.Shen,X.Zhang,D.Chen,X.D.Liu,L.B.Zhang,X.Y.Sun,J.S.Li,H.P.Bi,L.J.

Wang,Kinetics study of pyridine biodegradation by a novel bacterial strain, Rhizobium sp.NJUST18,Bioprocess Biosyst.Eng.37(2014)1185–1192. [10]J.Y.Shen,J.F.Zhang,Y.Zuo,L.J.Wang,X.Y.Sun,J.S.Li,W.Q.Han,R.He,

Biodegradation of24,6-trinitrophenol by Rhodococcus sp.isolated from a

picric acid-contaminated soil,J.Hazard.Mater.163(2009)1199–1206.

160X.Liu et al./Journal of Hazardous Materials295(2015)153–160

[11]APHA,Standard Methods for the Examination of Water and Wastewater,19th

edition,American Public Health Association,Washington DC,1998.

[12]K.Muda,A.Aris,M.R.Salim,Z.Ibrahim,A.Yahya,M.C.M.van Loosdrecht,A.

Ahmad,Z.Nawahwi,Development of granular sludge for textile wastewater treatment,Water Res.44(2010)4341–4350.

[13]Y.J.Song,S.Ishii,L.Rathnayake,T.Ito,H.Satoh,S.Okabe,Development and

characterization of the partial nitri?cation aerobic granules in a sequencing batch airlift reactor,Bioresour.Technol.139(2013)285–291.

[14]A.Bhattacharya,A.Gupta,A.Kaur,D.Malik,Ef?cacy of Acinetobacter sp.B9for

simultaneous removal of phenol and hexavalent chromium from

co-contaminated system,Appl.Microbiol.Biotechnol.(2014),

https://www.sodocs.net/doc/248402305.html,/10.1007/s00253-014-5910-5.

[15]B.Liang,H.Y.Cheng,J.D.V.Nostrand,J.C.Ma,H.Yu,D.Y.Kong,W.Z.Liu,N.Q.

Ren,L.Y.Wu,A.J.Wang,D.J.Lee,J.Z.Zhou,Microbial community structure and function of nitrobenzene reduction biocathode in response to carbon source switchover,Water Res.54(2014)137–148.

[16]I.Pala-Ozkok,A.Rehman,N.Yagci,E.Ubay-Cokgor,D.Jonas,D.Orhon,

Characteristics of mixed microbial culture at different sludge ages:effect on variable kinetics for substrate utilization,Bioresour.Technol.126(2012)

274–282.

[17]E.Dulekgurgen,N.Artan,D.Orhon,P.A.Wilderer,How does shear affect

aggregation in granular sludge sequencing batch reactors?Relations between shear hydrophobicity and extracellular polymeric substances,Water Sci.

Technol.58(2008)267–276.

[18]N.H.Rosman,A.N.Anuar,I.Othman,H.Harun,M.Z.Sulong,S.H.Elias,M.A.H.M.

Hassan,S.Chelliapan,Z.Ujang,Cultivation of aerobic granular sludge for

rubber wastewater treatment,Bioresour.Technol.129(2013)620–623. [19]K.Z.Su,H.Q.Yu,Formation and characterization of aerobic granules in a

sequencing batch reactor treating soybean-processing wastewater,Environ.

Sci.Technol.39(2005)2818–2827.

[20]S.S.Adav,D.J.Lee,K.Y.Show,J.H.Tay,Aerobic granular sludge:recent

advances,Biotechnol.Adv.26(2008)411–423.

[21]H.L.Wang,L.Li,P.Li,H.Li,G.S.Liu,J.M.Yao,The acceleration of sludge

granulation using the chlamydospores of Phanerochaete sp.HSD,J.Hazard.

Mater.192(2011)963–969.[22]Y.Liu,J.H.Tay,State of the art of biogranulation technology for wastewater

treatment,Biotechnol.Adv.22(2004)533–563.

[23]B.S.McSwain,R.L.Irvine,P.A.Wilderer,The in?uence of settling time on the

formation of aerobic granules,Water Sci.Technol.50(2004)195–202.

[24]K.V.Padoley,A.S.Rajvaidya,T.V.Subbarao,R.A.Pandey,Biodegradation of

pyridine in a completely mixed activated sludge process,Bioresour.Technol.

97(2006)1225–1236.

[25]G.Liu,Z.Ye,H.Li,R.Che,L.Cui,Biological treatment of hexanitrostilbene

(HNS)produced wastewater using an anaerobic–aerobic immobilized

microbial system,Chem.Eng.J.213(2012)118–124.

[26]D.H.Wen,J.Zhang,R.L.Xiong,R.Liu,L.J.Chen,Bioaugmentation with a

pyridine-degrading bacterium in a membrane bioreactor treating

pharmaceutical wastewater,J.Environ.Sci.25(2013)2265–2271.

[27]Y.H.Bai,Q.H.Sun,R.H.Sun,D.H.Wen,X.Y.Tang,Bioaugmentation and

adsorption treatment of coking wastewater containing pyridine and

quinoline using zeolite-biological aerated?lters,Environ.Sci.Technol.45

(2011)1940–1948.

[28]A.F.Duque,V.S.Bessa,M.F.Carvalho,M.K.de Kreuk,M.C.M.van Loosdrecht,

P.M.L.Castro,2-Fluorophenol degradation by aerobic granular sludge in a

sequencing batch reactor,Water Res.45(2011)6745–6752.

[29]Y.H.Bai,Q.H.Sun,C.Zhao,D.H.Wen,X.Y.Tang,Microbial degradation and

metabolic pathway of pyridine by a Paracoccus sp.strain BW001,

Biodegradation19(2008)915–926.

[30]L.Qiao,D.H.Wen,J.L.Wang,Biodegradation of pyridine by Paracoccus sp.

KT-5immobilized on bamboo-based activated carbon,Bioresour.Technol.

101(2010)5229–5234.

[31]L.Qiao,J.L.Wang,Microbial degradation of pyridine by Paracoccus sp.isolated

from contaminated soil,J.Hazard.Mater.176(2010)220–225.

[32]Y.Lv,C.Wan,X.Liu,Y.Zhang,D.J.Lee,J.H.Tay,Freezing of aerobic granules for

storage and subsequent recovery,J.Taiwan Inst.Chem.Eng.44(2013)

770–773.