Functional rigidity of a methane biofilter during the temporal microbial succession
Post on 23-Dec-2016
Functional rigidity of a methane biofilterduring the temporal microbial succession
Tae Gwan Kim & So-Yeon Jeong & Kyung-Suk Cho
Received: 15 July 2013 /Revised: 28 October 2013 /Accepted: 30 October 2013# Springer-Verlag Berlin Heidelberg 2013
Abstract Temporal microbial succession was investigated inrelation to the performance of a methane biofilter. Alaboratory-scale biofilter packed with perlite was operatedfor 108 days, without a deliberate biomass control. The systemperformance was stable over the period with a mean elimina-tion capacity of 1,563 g m3 day1, despite a temporal deteri-oration (4556 days). Ribosomal-tag pyrosequencing showedthat bacterial communities at days 1428 were distinct fromthose of days 68108. The accumulation of nonviable sub-stances strongly coincided with the community change (R2>0.97). Rhodobacter, Hydrogenophaga , and Methylomonaswere dominated in the earlier period, while Methylocaldumand Methylococcus were abundant in the later period. Themethanotrophic proportion gradually increased to 41 %, andtype I methanotrophs became predominant over time.However, community structure and methanotrophic popula-tion density stably retained over time, allowing the system tokeep the similar performance. Therefore, the perlite biofiltersystem was functionally rigid against the temporal microbialsuccession.
Keywords Methanotrophic biofilter . Microbial community .
Functional rigidity . Succession . Perlite
Various strategies have been developed to mitigate methaneemissions, as methane is a potent greenhouse gas. Biologicalfiltration is used for methane mitigation from landfills, coal
mines, and animal husbandries where methane is emitted.Aerobic methane-oxidizing bacteria (methanotrophs) utilizemethane as their sole carbon and energy source (Semrau et al.2010) and are used to degrade methane during methane filtra-tion. Previous studies of methane biofiltration have mainlyfocused on abiotic factors, such as bed materials, temperature,loading rate, pH, etc. (Gebert et al. 2003; Nikiema et al. 2005;Melse and van der Werf 2005; Nikiema and Heitz 2009;Nikiema and Heitz 2010). For instance, a number of materials,e.g., perlite, granular activated carbon, and compost, havebeen evaluated as filter beds for methane removal (Melseand van derWerf 2005; Kim et al. 2013a). Recently, biologicalaspects, e.g., microbial community, have received increasingattention in methane biofiltration studies (Gebert et al. 2008;Kim et al. 2013a).
It has been observed that system performance is associatedwith the microbial community in biological engineering sys-tems, e.g., methane biofiltration (Kim et al. 2013a), microbialfuel cells (Wrighton et al. 2010), activated sludge (Hoshinoet al. 2006), phosphorus removal (Okunuki et al. 2004), andanaerobic digestion (Hori et al. 2006). Most of the associatedstudies showed that changes of external factors, such as nutri-ent and pH, resulted in alterations of the microbial communi-ties, which, in turn, altered the system performances. Forinstance, Kim et al. (2013a) reported that volatile organiccompounds modified the development of microbial communi-ties, and differences in microbial communities were responsi-ble for variations in the performance of methanotrophicbiofilters. Besides external factors, temporal variation of mi-crobial community takes place in a biological engineeringsystem, and can result in variation of the system performance(Hoshino et al. 2006). Many studies have examined microbialcommunities in methanotrophic biofiltration systems (Nikiemaet al. 2005; Gebert et al. 2008; Kim et al. 2013a), but no studiesof temporal community changes have previously been con-ducted. Therefore, there are a lack of information on temporal
T. G. Kim : S.
microbial succession, and little knowledge of correlations be-tween the functioning stability and temporal community suc-cession in methane biofiltration systems.
Methane biofilters that contain methanotrophs and non-methanotrophs are densely populated and highly complex,although methanotrophs are the first producers within thesystem (Kim et al. 2012a, b, 2013a). Previous studies havereported that Hydrogenophaga , Pseudoxanthomonas , andHyphomicrobium genera are common non-methanotrophicinhabitants in methanotrophic biofilters and biocovers (Kimet al. 2013a, b; Nikiema et al. 2005). We have recentlyobserved that microbial communities of methanotrophicbiofilters are prone to the influence of external factors (e.g.,trace amount of volatile organic compounds) (Kim et al.2013a). Methanotrophic biofilters accumulate biomass duringtheir operation due to the growth of the biofilm. Excessbiomass can cause deterioration of system performance inair biofiltration systems (Ryu et al. 2010). The biomass pro-duced may include microbial cells, extracellular polymericsubstances, short carbon compounds, lysed cells, etc., ofwhich nonviable substances can be used as a growth substratefor non-methanotrophic microorganisms. This may alter masstransfer rates of molecules such as O2, CO2, and CH4 throughthe biofilm matrix (Stewart 2003), which can influence thegrowth and activity of non-methanotrophs as well as those ofmethanotrophs. Therefore, it was hypothesized that biomassaccumulation, as an intrinsic factor, can strongly impact themicrobial community and performance of a methanotrophicbiofilter.
The objective of this study was to characterize temporalchange of microbial community in relation to the performancein a methane biofiltration system, and to test whether accu-mulated biomass can impact the microbial community andmethanotrophic activity. Perlite was used as a packing mate-rial in this study, since it was shown to be a promising bedmaterial for methane biofiltration (Kim et al. 2013a). A perlitebiofilter was operated over more than 100 days without ma-nipulation of the biomass (e.g., deliberately washing out ex-cess biomass from the packing material). The microbial com-munity was analyzed using a combination of pyrosequencingand quantitative real-time PCR of the 16S rRNA gene.
Materials and methods
Lab-scale perlite biofilter
A lab-scale biofilter was made of cylindrical acrylic resin. Itconsisted of three parts: a packing section (actual reactor),irrigation system, and drain container, as illustrated in Fig. 1.Their heights and inner diameters were 100 and 8 (a workingvolume of 5 L), 20 and 8, and 20 and 15 cm (3.5 L), respec-tively. The packing section had a perforated plate at the
bottom in to allow gases to evenly spread and sampling portslocated on the side. They were assembled by rubber packingand six bolts/nuts (top-medium irrigation system-packingsection-medium container-bottom). The gas tightness of thereactor was verified with water and compressed air.
Perlite (Hyuga pumice, Japan) with a diameter of 48 mmwas used as a packing material. Perlite was thoroughlywashed with tap water and was air-dried before it was usedto fill the packing section. The water holding capacity, pH,bulk density, skeletal density, porosity, surface area, and in-trusion volume of perlite were 64.612.4 %, 6.00.0, 0.4700.001 g cm3, 2.0840.217 g cm3, 70.46.7 %, 32.17.0 m2 g1, and 1.0830.233 mL g1, respectively (Jeonget al. 2013).
An inoculum was obtained from a biofilter that had onlybeen used for methane removal. The modified nitrate mineralsalts (NMS) medium, used for a week, was collected from themedium container of the biofilter. The medium (2.5 L) wasadded into the medium container of the new biofilter, and wascirculated four times per day for 7 days to allow for cellattachment to the packing material. Methane (99.9 %, SeoulGas, Seoul Korea) and compressed air were passed through a50-cm long humidifier filled with water. Gas flows werecontrolled with commercial flow-meters (Dwyer, Michigancity, USA and Kofloc, Kyoto, Japan). The synthetic gas ofair and methane (50,000 ppm) was continuously fed at a flowrate of 250 ml min1 (a space velocity of 3 h1). The biofilterwas operated at 205 C for 108 days with the modified NMSmedium. The modif ied NMS medium containedMgSO47H2O 1 g L
1, CaCl22H2O 0.295 g L1, KNO3
10 g L1, KH2PO4 2.6 g L1, and Na2HPO42H2O
4.1 g L1. CuSO4 was added to a final concentration of30M. The modified NMSmediumwas circulated four timesper day, and was replaced every week.
CH4 concentrations were measured at the inlet and outlet tomonitor the methane removal in the biofilter. In order todetermine conversion rate of CH4 to CO2 in the system, CO2concentrations were periodically measured at the inlet andoutlet. Mole ratios of produced CO2 to consumed CH4 werecalculated over time in the biofilter.
Methane was monitored using gas chromatography (GC,6850 N, Agilent Technologies, Santa Clara, USA), equippedwith a flame ionization detector and a wax column (30 m0.32 mm0.25 m, Supelco, Bellefonte, USA). The oven,injector, and detector temperatures were set at 100, 230, and230 C. CO2 concentrations were monitored using GC(6890N, Agilent) equipped with a HP-PLOT/Q column(30 m0.53mm40m) and a thermal conductivity detector.The oven, injector, and detector temperatures were set at 60,
Appl Microbiol Biotechnol
100, and 250 C, respectively. Pure CH4 and CO2 gases werecommercially purchased, and were used as the standard gases.
Sampling and DNA extraction
Samples of the packingmaterials were collected on days 7, 14,28, 68, and 108 from the middle of the packing sectionthrough a sampling port. For biomass estimation, volatilesolids (VS) of packing materials were measured (n =3).Samples were placed into porcelain crucibles and dried indry oven at 105 C for 2 h. After cooling, they were placedin a muffle furnace at 550 C for 0.5 h. The volatile solids(burned out at 550 C) were weighted. At each sampling time,a 10-g sample was added to a sterile cone-tube containing20 ml of sterile 0.9 % saline solution. A horn tip (6 mmdiameter) of the Q500 ultrasonic processor (Qsonica LLC,Newton, USA) was immersed about 1 cm into the solution ofthe tube. It was sonicated for 5 min at 10 W (20 kHz). After abrief hand-shaking, 1-mL suspensions were immediately col-lected from the bottle, and were transferred to 1.5-mLmicrotubes. They were then centrifuged at 16,000g for5 min, and the supernatant was discarded from the tubes.For DNA extraction from the inoculum, 1.5-mL suspensionswere collected and transferred to 1.5-mL microtubes. Theywere centrifuged at 16,000g for 5 min, and the supernatantwas discarded from the tubes.
DNA was extracted using the NucleoSpin Soil kit(Macherey-Nagel GmbH, Dren, Germany), with a modifica-tion that samples were disrupted using a BeadBeater-8 system(BioSpec, Bartlesville, USA) at 5,000 rpm for 30 s. DNAwascollected in 100 L of the elution buffer, and was quantified
using an ASP-2680 spectrophotometer (ACTGene,Piscataway, USA).
Quantitative PCR (qPCR) was performed using the primer set:340F (5-TCCTACGGGAGGCAGCAG-3) and 805R (5-GACTACHVGGGTATCTAATCC-3) to quantify the bacte-rial population (Kim et al. 2012b) (n =5). DNA quantificationwas performed using the Applied Biosystems 7300 real timePCR (Applied Biosystems, Carlsbad, USA). qPCR reactionswere performed in 25-L volumes. The reaction mixtureconsisted of 2.5 L of 10 PCR buffer, 0.125 L of Taqpolymerase (Qiagen, Valencia, USA), 0.5 L of 50 SYBRgreen I, 0.5 L of 340F primer (10 M) and 1 L of 805Rprimer (10 M), 0.5 L of 50 ROX, as a reference dye, and2 L of template DNA. Control reactions contained the samemixtures, with 2 L of sterile water replacing the DNAtemplate. PCR was initiated at 95 C for 3 min, followed by35 cycles at 95 C for 30 s, 50 C for 40 s, and 72 C for 30 s.The 16S rRNA gene of E. coli was amplified using the 340F/805R primer set and purified, which was used to prepare thestandard (a tenfold dilution series) for quantitative detection.
PCR and pyrosequencing
A total of 10 pyrosequencing libraries were constructed in thisstudy. There were two replicates per sample of the packingmaterial on days 14, 28, 68, and 108 and the initial inoculum.For PCR, ten composite primer sets were made based on the340F805R set for multiplex pyrosequencing as previouslydescribed (Kim et al. 2012a). Four independent PCR reactions
Fig. 1 Schematic diagram of the biofilter
Appl Microbiol Biotechnol
were prepared in parallel to avoid PCR bias for each DNAsample. Twenty five-cycle PCR reactions were performedwith 100150 ng of template DNA, and PCR products werepurified as previously described (Kim et al. 2012a). The sameamounts of purified DNAs were combined in a single tube toproduce pyrosequencing libraries, and were then sent toMacrogen Incorporation (Seoul, Korea) to be run on aGenome Sequencer 454 FLX Titanium system (RocheDiagnostics, Mannheim, Germany).
Data analysis for bacterial community
The pyosequencing libraries that were constructed contained4,64914,893 sequences with average read lengths of 414419 bp. The pyrosequencing libraries were deposited into theDNA Data Bank of Japan (DDBJ) Sequence Read Archive(http://trace.ddbj.nig.ac.jp/dra) under the accession numberDRA000988. Sequences were filtered (length
was stable at about 6063%.Meanmethane RE over the wholeexperimental period was 61.310.2 %. Elimination capacity(EC) was 1,563365 g m3 day1 at an inlet load of 2,557251 g m3 day1. Conversion rate of CH4 to CO2 (mole ratio ofCH4 consumption and CO2 production in the system) increasedby day 45 and then slightly reduced over time (Fig. 2c), and themean conversion rate was 74.79.9 %.
During the operation, the water in the drain container wascirculated every 6 h to provide moisture and minerals. Thefilter bed was not deliberately manipulated in any waythroughout the 108-d experimental period, other than throughthe circulation of water although viscous biomass grew overtime. Volatile solids (VS) increased to 0.320.02 g g1 of drypacking material by day 68 (p
the CA plot explained 58.8 and 27.5 % of the communitycomposition variation, respectively. The CA plot showed thatcommunities at days 14 and 28 were different from those atdays 68 and 108, as also shown by the OTU-based CA plot(Fig. 3). The genera enclosed by a continuous line in Fig. 4were more abundant in the communities of days 14 and 28,while the genera enclosed by a dotted line were more abun-dant in the later communities. For instance, Rhodobacter
(17.223.0 % of the community), Hydrogenophaga (16.618.8 %), Methylomonas (7.919.1 %), and Methylosinus(3.54.8 %) were more abundant in the earlier communities.Methylocaldum (15.432.2 %), Methylococcus (11.011.2 %), and Flavitalea (3.74.4 %) were more abundant inthe later communities. Hyphomicrobium (18.049.3 %) andOhtaekwangia (4.67.3 %) were abundant during the opera-tional period.
Table 2 Bacterial community composition at the genus level. Only genera (>0.5 % of the sequencing effort) observed in either library are shown
Phylum Genus Inoculum Operation day
14 28 68 108
Acidobacteria Acidobacteria 1.320.10 0.110.09 0.200.15 0.670.06 1.160.04
Proteobacteria Brevundimonas a 3.040.28 0.120.05 0.330.27 1.530.09 2.940.48
Devosia a 1.230.08 0.100.08 0.000.00 0.040.03 0.200.05
Hyphomicrobium a 15.440.99 18.040.28 32.751.16 49.310.15 30.360.07
Methylocystis a,b 0.400.07 0.550.14 0.570.25 0.340.01 0.030.04
Methylosinus a,b 5.780.59 4.840.19 3.540.20 0.010.02 0.000.00
Paracoccus a 2.840.12 0.140.05 0.380.17 0.040.01 0.000.00
Rhodobacter a 9.010.31 17.201.67 23.010.92 2.860.28 1.390.18
Sphingomonas a 0.310.06 0.090.01 0.060.02 0.060.02 0.470.07
Sphingosinicella a 0.040.01 0.030.04 0.030.01 0.350.01 1.420.29
Tepidamorphus a 0.080.00 0.090.06 0.030.01 1.090.00 0.460.01
Comamonas c 0.380.20 0.000.00 0.020.03 0.000.00 0.030.00
Hydrogenophaga c 1.520.44 18.840.56 16.551.29 3.090.13 0.380.06
Arenimonas d 0.590.43 0.000.00 0.000.00 0.090.02 0.220.01
Methylocaldum b,d 3.700.05 0.070.04 0.040.06 15.410.11 32.152.96
Methylococcus b,d 14.111.31 1.090.37 3.250.61 10.980.15 11.241.03
Methylomonas b,d 6.400.59 19.084.04 7.940.29 1.440.15 0.110.06
Pseudoxanthomonas d 0.110.06 1.510.14 0.510.14 0.450.03 0.220.17
Nannocystis e 0.170.09 3.070.34 0.900.25 0.040.02 0.010.02
Plesiocystis e 0.020.02 0.000.00 0.000.00 0.590.09 1.170.05
Bacteroidetes Ferruginibacter 0.010.01 0.840.05 0.240.10 0.010.02 0.000.00
Flavitalea 0.160.05 0.930.46 0.630.10 3.660.41 4.350.05
Niabella 0.220.08 1.340.23 0.800.16 0.370.08 0.040.05
Nubsella 0.550.17 0.120.05 0.000.00 0.000.00 0.000.00
Terrimonas 0.130.02 0.560.06 0.430.18 0.270.04 0.400.06
Ohtaekwangia 1.350.18 7.300.02 4.570.40 5.200.13 7.030.49
Proteiniphilum 2.500.78 0.000.00 0.000.00 0.010.01 0.000.00
Flavobacterium 20.194.74 1.810.28 0.710.09 0.010.02 1.270.14
Spirochaetes Leptonema 1.300.15 0.120.02 0.070.04 0.010.01 0.050.03
Verrucomicrobia Opitutus 0.590.17 0.030.04 0.020.03 0.020.03 0.000.00
Sum 93.500.83 98.000.13 97.620.26 97.950.04 97.080.54
Others 6.50 2.00 2.38 2.05 2.92
Appl Microbiol Biotechnol
Methanotrophic community change
Seven methanotrophic genera were observed from the inocu-lum and biofilter. Relative abundances of methanotrophs areshown in Fig. 5a. The relative abundance of methanotrophs inthe inoculum was about 25 %. The relative abundance de-creased to 15 % on day 28 as the bacterial population increasedby one order, and then increased to 41 % over time (p
II) was 3.8 % in the biofilm. The relative abundance ofMethylosinus (18.7 %, II) was 19.1 and 23.1 % at days 14and 28, respectively, and was present in very low concentra-tions in the later time period.
CA was performed to distinguish between themethanotrophic communities. The first and second axes ofthe CA plot explained 84.1 and 14.0 % of the communitycomposition variation, respectively (Fig. 6b). The CA plotshowed the temporal change in the methanotrophic commu-nity. Biofilter methanotrophic communities were categorizedinto two distinct groups, consistent with the total bacterialcommunity result.
Relationship between the biomass and community change
VS increased for the first 68 days and was then maintainedover time, while bacterial density peaked at day 28, followedby a reduction (Fig. 2d). It is therefore assumed that nonviablesubstances were accumulated on the bed material. Figure 7ashows the relationship between the VSlog10(16S rDNA copynumber)1 and scores on the first ordination axis of the CAplots of total bacteria (Fig. 3) and methanotrophs (Fig. 6b).The difference between scores indicates the extent of commu-nity dissimilarity. Variations in total bacterial community andmethanotrophic community were greatest between days 28
) in th
100Inoculumd 14d 28d 68d 108
1.0 1.3 2.1 3.
Axis 1 (84.1%)
Fig. 6 Comparison ofmethanotrophic communities inthe biofilter. Proportion (inpercent) of methanotrophicgenera in the methanotrophiccommunity (a) andmethanotrophic communitycomparison by correspondenceanalysis (b)
Appl Microbiol Biotechnol
and 68 as VSlog10(16S rDNA copy number)1 rapidly in-
creased. VSlog10(16S rDNA copy number)1 significantly
coincided with the scores of the CA plots (R2=0.9740, p 0.97) with the scores of the first ordi-nations of the total bacterial and methanotrophic CA plots(Fig. 7). This strong coincidence suggests that the accumula-tion of nonviable substances can be a driving force for chang-es of bacterial and methanotrophic communities in a methanebiofilter. Methanotrophic performance deteriorated from days45 to 56 in the biofilter (Fig. 2), which was coincided with thedistinct shift of both the communities of total bacteria andmethanotrophs, but not consistent with the temporal stabilityof methanotrophic population level. Thus, the substantialmicrobial succession was likely responsible for the temporaldeterioration of methane removal performance. Similarly, pre-vious studies reported that phosphorus (Okunuki et al. 2004),volatile fatty acids (Hori et al. 2006), and volatile organiccompounds (Kim et al. 2013a) shifted the microbial commu-nity in biological engineering systems, which was linked to avariation in the system performance.
The microbial composition temporally varied in thebiofilter system. However, the system exhibited a similarperformance over the experimental period after an acclimationperiod despite the 11-day deterioration period in the middle.
Appl Microbiol Biotechnol
These results indicated that the microbial community of thebiofilter had functional similarity although it succeeded overtime. The term functional similarity is defined as the ability ofmicrobial communities to similarly conduct a functional pro-cess although the community compositions are not the same(Allison and Martiny 2008). Therefore, the perlite biofilterwas functionally rigid during the temporal succession of totalbacterial and methanotrophic communities. The stable micro-bial community structure (richness, diversity, and evenness)and methanotrophic population density were responsible forthe functional rigidity of the system.
It was hypothesized that biomass accumulation stronglyinfluences both the microbial community and the performanceof the methane biofilter. We found marked changes in themicrobial community but little evidence suggesting that theperformance of the biofilter was limited. The biofilter perfor-mance was apparently stable (1,563365 g m3 day1) al-though the microbial composition dramatically changed. Thesystem favored the growth of Rhodobacter,Hydrogenophaga ,and Methylomonas in the first 28 days, and Methylocaldumand Methylococcus in 68108 days. Type I methanotrophsbecame predominant, and the proportion of this methanotrophtype increased in the microbial community over time.However, the system allowed the microbial communitystructure and methanotrophic population density to remaintemporally stable, so that a similar performance wasmaintained over time. These findings indicate the func-tional rigidity of the perlite biofilter against microbialsuccession.
Acknowledgments This research was supported by the Basic ScienceResearch Program through the National Research Foundation of Korea(NRF), funded by the Ministry of Education, Science and Technology(MEST) (NRL program, R0A-2008-000-20044-0). This work was alsosupported by the National Research Foundation of Korea (NRF) grantfunded by the Korea government (MSIP) (no. 2012R1A2A03046724).
Conflict of interest The authors declare that they have no conflict ofinterest.
Allison SD, Martiny JBH (2008) Resistance, resilience, and redundancyin microbial communities. Proc Natl Acad Sci U S A 105:1151211519
Amaral JA, Archambault C, Richards SR, Knowles R (1995)Denitrification associated with groups I and II methanotrophsin a gradient enrichment system. FEMS Microbiol Ecol 18:289298
Atlas RM, Bartha R (1997) Microbial ecology: fundamentals and appli-cations. Benjamin/Cummings Science Publishing, Melon Park
Cbron A, Bodrossy L, Stralis-Pavese N, Singer AC, Thompson IP,Prosser JI, Murrell JC (2007) Nutrient amendments in soil DNAstable isotope probing experiments reduce the observedmethanotroph diversity. Appl Environ Microbiol 73:798807
Contzen M, Moore ERB, Blmel S, Stolz A, Kmpfer P (2000)Hydrogenophaga intermedia sp. nov., a 4-aminobenzene-sulfonatedegrading organism. Syst Appl Microbiol 23:487493
Gebert J, Groengroeft A, Miehlich G (2003) Kinetics of microbial landfillmethane oxidation in biofilters. Waste Manage 23:609619
Gebert J, Stralis-Pavese N, Alawi M, Bodrossy L (2008) Analysis ofmethanotrophic communities in landfill biofilters using diagnosticmicroarray. Environ Microbiol 10:11751188
Girard M, Ramirez AA, Buelna G, Heitz M (2011) Biofiltration ofmethane at low concentrations representative of the piggery indus-tryinfluence of the methane and nitrogen concentrations. ChemEng J 168:151158
Graham DW, Chaudhary JA, Hanson RS, Arnold RG (1993) Factorsaffecting competition between type I and type II methanotrophs intwo-organism, continuous-flow reactors. Microb Ecol 25:117
Hanson R, Hanson T (1996) Methanotrophic bacteria. Microbiol Rev 60:439471
Hatamoto M, Miyauchi T, Kindaichi T, Ozaki N, Ohashi A (2011)Dissolved methane oxidation and competition for oxygen indown-flow hanging sponge reactor for post-treatment of anaerobicwastewater treatment. Bioresource Technol 102:1029910304
Henckel T, Roslev P, Conrad R (2000) Effects of O2 and CH4 on presenceand activity of the indigenous methanotrophic community in ricefield soil. Environ Microbiol 2:666679
Ho A, Vlaeminck SE, Ettwig KF, Schneider B, Frenzel P, Boon N (2013)Revisiting methanotrophic communities in sewage treatment plants.Appl Environ Microbiol 79:28412846
Hori T, Haruta S, Ueno Y, Ishii M, Igarashi Y (2006) Dynamic transitionof a methanogenic population in response to the concentration ofvolatile fatty acids in a thermophilic anaerobic digester. ApplEnviron Microbiol 72:16231630
Hoshino T, Terahara T, Yamada K, Okuda H, Suzuki I, Tsuneda S, HirataA, Inamori Y (2006) Long-term monitoring of the succession of amicrobial community in activated sludge from a circulation flushtoilet as a closed system. FEMS Microbiol Ecol 55:459470
Jahnke LL, Nichols PD (1986) Methyl sterol and cyclopropane fatty acidcomposition of Methylococcus capsulatus grown at low oxygentensions. J Bacteriol 167:238242
Jeong S-Y, Yoon H-Y, Kim TG, Cho K-S (2013) Effect of tobermolite,perlite and polyurethane packing materials on methanotrophic ac-tivity. Korean J Microbiol Biotechnol 41:215220
Keer JT, Birch L (2003) Molecular methods for the assessment ofbacterial viability. J Microbiol Meth 53:175183
Kim Y-J, Kim MK, Weon H-Y, Kim H-B, Yang D-C (2010)Hydrogenophaga temperata sp. nov., a betaproteobacteriumisolated from compost in Korea. J Gen Appl Microbiol 56:419425
Kim TG, Lee E-H, Cho K-S (2012a) Microbial community analysis of amethane-oxidizing biofilm using ribosomal tag pyrosequencing. JMicrobiol Biotechnol 22:360370
Kim TG, Yi T, Lee E-H, RyuHW, Cho K-S (2012b) Characterization of amethane-oxidizing biofilm using microarray, and confocal micros-copy with image and geostatic analyses. Appl Microbiol Biotech 95:10511059
Kim TG, Lee E-H, Cho K-S (2013a) Effects of nonmethane volatileorganic compounds on microbial community of methanotrophicbiofilter. Appl Microbiol Biotech 97:65496559
Kim TG, Moon K-E, Yun J, Cho K-S (2013b) Comparison of RNA- andDNA-based bacterial communities in a lab-scale methane-degradingbiocover. Appl Microbiol Biotech 97:31713181
Mackenzie C, Eraso JM, Choudhary M, Roh JH, Zeng X, Bruscella P,Pusks , Kaplan S (2007) Postgenomic adventures withRhodobacter sphaeroides . Annu Rev Microbiol 61:283307
Melse RW, van der Werf AW (2005) Biofiltration for mitigation ofmethane emission from animal husbandry. Environ Sci Tech 39:54605468
Appl Microbiol Biotechnol
Nikiema J, Heitz M (2009) The influence of the gas flow rate duringmethane biofiltration on an inorganic packing material. Can J ChemEng 87:136142
Nikiema J, Heitz M (2010) The use of inorganic packing materials duringmethane biofiltration. Int J Chem Eng 2010:18
Nikiema J, Bibeau L, Lavoie J, Brzezinski R, Vigneux J, Heitz M (2005)Biofiltration of methane: an experimental study. Chem Eng J 113:111117
Nikiema J, Girard M, Brzezinski R, Heitz M (2009) Biofiltration ofmethane using an inorganic filter bed: influence of inlet load andnitrogen concentration. Can J Civ Eng 36:19031910
Nikiema J, Brzezinski R, Heitz M (2010) Influence of phosphorus,potassium, and copper on methane biofiltration performance. CanJ Civ Eng 37:335345
Okunuki S, Kawaharasaki M, Tanaka H, Kanagawa T (2004) Changes inphosphorus removing performance and bacterial community struc-ture in an enhanced biological phosphorus removal reactor. WaterRes 38:24332439
Perez-Garcia O, Escalante FME, de-Bashan LE, Bashan Y (2011)Heterotrophic cultures of microalgae: metabolism and potentialproducts. Water Res 45:1136
Puhar E, Lorencez I, Fiechter A (1983) Influence of partialpressure of oxygen and carbon dioxide on Methylomonas
clara in continuous culture. Eur J Appl Microbiol 18:131134
Ryu HW, Cho K-S, Chung DJ (2010) Relationships between biomass,pressure drop, and performance in a polyurethane biofilter.Bioresource Technol 101:17451751
Semrau JD, DiSpirito AA, Yoon S (2010) Methanotrophs and copper.FEMS Microbiol Rev 34:136
Stewart PS (2003) Diffusion in biofilms. J Bacteriol 185:14851491Thierry S, Macarie H, Iizuka T, Geissdorfer W, Assih EA, Spanevello M,
Verhe F, Thomas P, FudouR,MonroyO, LabatM,OuattaraAS (2004)Pseudoxanthomonas mexicana sp. nov. and Pseudoxanthomonasjaponensis sp. nov., isolated from diverse environments, and emendeddescriptions of the genus Pseudoxanthomonas Finkmann et al.2000 and of its type species. Int J Syst Evol Microbiol 54:22452255
Wittebolle L,Marzorati M, Clement L, Balloi A, Daffonchio D, Heylen K,De Vos P, Verstraete W, Boon N (2009) Initial community even-ness favours functionality under selective stress. Nature 458:623626
Wrighton KC, Virdis B, Clauwaert P, Read ST, Daly RA, Boon N, PicenoY, Andersen GL, Coates JD, Rabaey K (2010) Bacterial communitystructure corresponds to performance during cathodic nitrate reduc-tion. ISME J 4:14431455
Appl Microbiol Biotechnol
Functional rigidity of a methane biofilter during the temporal microbial successionAbstractIntroductionMaterials and methodsLab-scale perlite biofilterGas analysisSampling and DNA extractionQuantitative PCRPCR and pyrosequencingData analysis for bacterial communityStatistical analysis
ResultsBiofilter performance and biomass changeTotal bacterial community change on the basis of OTUTotal bacterial community change on the basis of taxaMethanotrophic community changeRelationship between the biomass and community change