Temporal dynamics of ammonia oxidizer (amoA) and denitrifier (nirK) communitieem from Tai Lak

Applied Soil Ecology 48 (2011) 210–218Contents lists available at ScienceDirectApplied SoilEcologyj 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 /a p s o ilTemporal dynamics of ammonia oxidizer (amoA )and denitrifier (nirK )communities in the rhizosphere of a rice ecosystem from Tai Lake region,ChinaQaiser Hussain a ,Yongzhuo Liu a ,Zhenjiang Jin a ,Afeng Zhang a ,Genxing Pan a ,∗,Lianqing Li a ,David Crowley b ,Xuhui Zhang a ,Xiangyun Song a ,Liqiang Cui aa Institute of Resources,Ecosystem and Environment of Agriculture,Nanjing Agricultural University,1Weigang,Nanjing 210095,China bDepartment of Environmental Sciences,University of California Riverside,CA 92521,USAa r t i c l e i n f o Article history:Received 20October 2010Received in revised form 28February 2011Accepted 9March 2011Keywords:Temporal dynamicsAmmonia oxidizer (amoA )Denitrifier (nirK )Rice rhizosphere DGGE qPCRa b s t r a c tA field experiment was conducted to investigate the abundance and dynamics of ammonia oxidizer (ammonia oxidizing archaea –AOA and ammonia oxidizing bacteria –AOB)and denitrifier communi-ties in the rhizosphere at four growing stages of rice using PCR-denaturing gradient gel electrophoresis (DGGE)and real-time PCR approaches.Rice plantation promoted greater abundance of amoA (AOA and AOB)and nirK (denitrifiers)genes in the rhizosphere than in the bulk soil,showing a profound rhizosphere effect.Rice growing stages significantly affected the structures and abundances of AOB and denitrifier (nirK )communities in the rhizosphere,whereas no effect was observed on the community structure and abundance of AOA in the rhizosphere.Moreover,the amoA gene copy numbers of AOA were more than those of AOB in all soil samples.However,denitrifier (nirK )generally dominated the ammonia oxidizer (amoA )in the rhizosphere during all growth stages,suggesting better adaptability of denitrifier in the rice rhizosphere environment.These results further suggest that AOB and denitrifier (nirK )communities associated with rice rhizosphere are highly dynamic in response to prevailing plant and soil conditions over a rice crop season,whereas AOA showed higher stability throughout the rice growing period.© 2011 Elsevier B.V. All rights reserved.1.IntroductionRice is the most important staple food crop for human consump-tion in the world and 75%of the world’s rice production comes from irrigated rice fields.Rice paddy soils are also known as appro-priate model systems to investigate essential aspects of microbial ecology,such as dynamics of microbial community structures,abundances and functional relationship between/among microbial groups (Liesack et al.,2000).Plant rhizosphere provides distinct microhabitats with respect to microorganisms as compared to the surrounding bulk soil (Bais et al.,2006).Nonetheless,the rhizo-sphere is highly dynamic as soil micro-biota respond quickly to changes in quantities and chemical composition of root exudates,which are considered to be crop varieties and growing stage spe-cific (Whipps,2001;Rengel,2002).Many field or microcosm studies have been focused on the influence of plants on the overall micro-bial structures in the rhizosphere based on 16S rRNA gene analysis (Smalla et al.,2001;Wieland et al.,2001;Kennedy et al.,2004).Alternation of water-logging and drainage condition,as unique water regime for rice production,allows shifts of redox potential∗Corresponding author.Tel.:+86254396027;fax:+86254396027.E-mail addresses:gxpan@ ,pangenxing@ (G.Pan).and various biochemical processes such as ammonium oxidation and nitrification in paddy soils (Kikuchi et al.,2007).Moreover,oxic or partially oxic niches are formed due to the diffusion of oxy-gen in upper few millimeters of flooded water,leaving the bulk soil anoxic.In planted soil the rice aerenchymatous tissues are also responsible for the leakage of oxygen creating an oxic rhizosphere within the anoxic bulk soil (Revsbech et al.,1999).The existence of oxic and anoxic microhabitats provides a favorable environment for nitrification and denitrification in rice paddy field soils (Revsbech et al.,1999).Hence,oxygen and carbon-releasing aerenchymatous rice plants may affect the composition of the ammonia oxidizer (AOA and AOB)and denitrifying bacterial communities in water-logged paddy soil.Ammonia oxidizer and denitrifier communities are widely recognized as models for ecology studies and are intrin-sically linked to agroecosystem functioning such as the nitrogen global cycling and N 2O emission (Kowalchuk and Stephen,2001;Philippot and Hallin,2005).Ammonia oxidization to nitrite is the initial and rate limiting step in nitrification which is carried out by AOB (Jackson et al.,2008;Malchair et al.,2010)and/or AOA (Nicol and Schleper,2006;Wuchter et al.,2006).Denitrifying bacteria are a crucial group of microbes involved in denitrification and respon-sible for nitrogen losses as well as N 2O emission from agricultural systems (Philippot et al.,2007).Therefore,in depth knowledge of the abundance and dynamics of ammonia-oxidizing prokaryotes0929-1393/$–see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.apsoil.2011.03.004Q.Hussain et al./Applied Soil Ecology48 (2011) 210–218211and denitrifying bacteria in the rhizosphere of rice paddy soils is essential for understanding the functioning of rice ecosystem and predicting the impact of prevailing plant and soil condi-tions.The ammonia monooxygenase␣-subunit(amoA)and copper nitrite reductase(nirK)genes have been used as functional mark-ers for cultivation independent studies of the ammonia-oxidizing prokaryotes(Cavagnaro et al.,2008;Nguan et al.,2009)and deni-trifying bacteria(Lardy et al.,2010;Smith et al.,2010),respectively. While recent studies have provided insight into the relative preva-lence of AOA and AOB in rice paddy(Chen et al.,2008;Wang et al.,2009),the knowledge is still limited on the dynamics of the individual functional groups of ammonia-oxidizing archaea(AOA), ammonia-oxidizing bacteria(AOB)and denitrifying bacteria in the rhizosphere with rice growing stages underfield conditions.PCR-denaturing gradient gel electrophoresis(DGGE)is a culture independentfingerprinting technique used to monitor the spatial and temporal changes in microbial communities and the dominant microbial species within a sample(Liu et al.,2008).Qualitative DGGE analysis in combination with quantitative real-time PCR (qPCR)is capable to provide a deep insight into soil microbial com-munity dynamics(Ahn et al.,2009).Real-time PCR is considered as aflexible,simple and rapid promising tool for the quantification of soil microbial communities though it may have some impor-tant limitations,including DNA extraction bias,judicious primer design,heterogeneity in ribosomal operon number,the availability of sequence data,and adequate preparation of inhibitor-free target DNA(Fierer et al.,2005).The objectives of this study were(I)to evaluate the temporal prevalence of AOA,AOB and denitrifying bacterial communities in the rhizosphere of rice plant,(II)to investigate the relative abundances of AOA to AOB and denitrifying bacteria to ammonia oxidizer(AOA+AOB)over a period of rice plant growth,and(III)to compare the ammonia oxidizer and denitrifier community struc-tures in bulk and rhizosphere soil of rice plant.The community structures and abundances of ammonia-oxidizing prokaryotes and denitrifying bacteria were characterized by PCR-DGGE and qPCR of functional gene amoA and nirK targets,respectively.2.Materials and methods2.1.Site description and experiment layoutThe experiment was carried out on a rice farm located in Yifeng village,Yixing Municipality,Jiangsu Province,China(31◦24.26 N, 119◦41.36 E).Derived from lacustrine deposit,the soil was a typ-ical high-yielding paddy soil classified as a hydroagric Stagnic Anthrosols(Gong,1999)and an entic Halpudept(Soil Survey Staff,1994).A subtropical monsoon climate prevailed in the area with mean annual temperature and precipitation of15.7◦C and 1177mm,respectively.The basic properties of the studied top-soil are as follows:pH(H2O)6.7,CEC18.05cmol kg−1,bulk density 1.1g cm−3,total organic carbon20.2g kg−1and total N2.99g kg−1, and available N276.1mg kg−1.For experiment,one month old rice (Oriza sativa)seedlings of Wugeng13cultivar were transplanted at a plant density of20plants/m2in a randomly selected plot of 4m×5m in area.The experiment used a randomized block design with three replicate plots.No chemicals were applied for plant protection and the plots were weeded by hand.Calcium biphos-phate,KCl and urea were applied as basal fertilizers at a rate of 125kg P2O5ha−1,125kg K2O ha−1and120kg N ha−1,respectively. The water regime was managed using an alternatingflooding and drainage cycle through the whole growing season.Soil samples were collected at different rice growing stages:before plantation (S0,just before paddy water logging and fertilizer application),45 days after planting(S1,tillering stage,flooded condition),81days after planting(S2,grainfilling stage,wet condition)and107days after planting(S3,ripening stage,moist condition).Unplanted bulk soil,S0,was used as the control for the effect of the plant rhizo-sphere on microbial communities.2.2.Soil samplingFor sampling of the rhizosphere soil,ten rice plants with root–soil systems were randomly excavated10cm deep from the same replicate plot at each growth stage.Following Butler et al. (2003)and Liu et al.(2008),the soil separated gently by hand from root–soil systems of the ten excavated plants was considered as bulk soil(non-rhizosphere soil).The remaining about1cm thick soils tightly attached to the root system was used as rhizosphere soil,which was carefully removed from the roots with a probe and forceps.The rhizosphere and bulk soil samples separated from ten plants of a single replicate plot were then pooled to form a composite sample,respectively.Soil samples were sieved(<2mm) immediately after collection and stored at−20◦C(not more than one week)for DNA extraction.For comparative purposes,the sam-pling depth was the same throughout the rice growing stages.All soil samples were taken at same time(9:30AM)on each sampling day to limit diurnal effects.2.3.Physico-chemical analysisAll physico-chemical properties of the soil were analyzed according to the protocols described by Lu(2000).Soil organic carbon and total nitrogen(TN)of dried samples were measured using a CNS Macro Elemental Analyzer(Elementar Analysen System GmbH,Germany)after treatment with HCl(10%,v/v)to remove car-bonates if any in the samples.The moisture content of the samples was determined by oven-drying at105◦C for24h.Soil pH(H2O) was measured by Mettler–Toledo pH meter with a soil:water ratio of1:2.5.Cation exchange capacity(CEC)was measured with the ammonium acetate(1mol L−1,pH7)leaching method.2.4.DNA extraction and real time PCR assayThree DNA extractions from each soil sample of a single repli-catedfield plot were performed.Each DNA extraction was made from0.5g soil using a PowerSoil TM DNA Isolation Kit(Mo Bio Lab-oratories Inc.,CA)following the manufacturer’s instructions.The three DNA extracts of the same soil sample of a single replicated plot were then pooled for analysis.The copy numbers of amoA (AOA),amoA(AOB)and nirK genes in all soil samples were deter-mined in triplicate using an iCycler IQ5Thermocycler(Bio-Rad, Hercules,CA).The quantification was based on thefluorescent dye SYBR-green1,which binds to double stranded DNA during PCR amplification.Primers and the thermal cycling conditions are men-tioned in Table1.The DNA concentration of all soil samples was measured at260nm using a UV Spectrophotometer(Bio Photome-ter,Eppendorf,Germany)and then adjusted to10ng␮l−1.Each reaction was performed in a25␮l volume containing10ng of DNA, 0.2mg ml−1BSA,0.2␮M of each primer and12.5␮l of SYBR premix EX Taq TM(Takara Shuzo,Shinga,Japan).Melting curve analy-sis of the PCR products was conducted following each assay to confirm that thefluorescence signal originated from specific PCR products and not from primer–dimers or other artifacts.PCR prod-ucts were checked for correct size by comparison to a standardized molecular weight ladder by electrophoresis on1.5%agarose gel.A plasmid standard containing the target region was generated for each primer set(AOA,AOB and denitrifying bacteria)using total DNA extracted from the soil samples.The amplified PCR products of amoA(AOA),amoA(AOB)and nirK genes were purified using PCR solution purification kit(Takara),ligated into p-GEM T easy vec-212Q.Hussain et al./Applied Soil Ecology48 (2011) 210–218Table1Primer sets and thermal profiles used for the absolute quantification of functional target genes involved in nitrogen turnover.Target gene Primer set Size Thermal cycling profile ReferenceamoA(AOB)amoA-1FamoA-2R 490bp94◦C(10min);40cycles of94◦C(30s),53◦C(60s),and72◦C(60s).Data acquisition temperature at83◦CMcTavish et al.(1993)amoA(AOA)Arch-amoA FArch-amoA R 635bp94◦C(10min);40cycles of94◦C(30s),53◦C(30s),and72◦C(45s).Data acquisition temperature at83◦CFrancis et al.(2005)nirK nirK876nirK1040165bp95◦C(10min);40cycles of94◦C(30s),58◦C(60s),and72◦C(60s).Data acquisition temperature at80◦CHenry et al.(2004)AOB:ammonia oxidizing bacteria;AOA:ammonia oxidizing archaea.tor(Promega,Madison,WI)and cloned into Escherichia coli DH5␣. Clones containing correct inserts were chosen as the standards for real-time PCR(qPCR).Plasmid DNA was isolated using plasmid extraction kit(Takara)and DNA concentrations were determined by spectrophotometer as mentioned above.As the size of the vector and PCR inserts were known,the copy numbers of amoA genes(AOA and AOB)and nirK gene were directly calculated from the concen-tration of extracted plasmid DNA.Standard curves were generated using triplicate10-fold dilutions of plasmid DNA ranging from 1.35×102to1.35×108copies for amoA(AOA)gene,1.03×102to 1.03×108copies for amoA(AOB)gene and4.89×102to4.89×108 copies of template for nirK gene per assay.High amplification effi-ciencies of99%(AOA),109%(AOB)and nirK(93%)were obtained using the slopes−3.35,−3.11and−3.51of standard curve,respec-tively.2.5.PCR-DGGE ammonia oxidizers and denitrifying bacterial community analysisTotal extracted DNA of each soil sample was amplified with the Arach-amoA F-GC and Arach-amoA R primer set specific for AOA (Francis et al.,2005),the amoA-1F-GC and amoA-2R set specific for the AOB(McTavish et al.,1993)and the nirK876-GC and nirK1040 set for the denitrifying bacterial communities(Henry et al.,2004). The GC clamp(5 -CCGCCGCGCGGCGGGCGGGGCGGGGGCACGGGG-3 )described by Muyzer et al.(1997)was added to5 end of primer to stabilize the melting behaviour of the DNA fragments.PCR reac-tion was performed in an Eppendorf Autothermer Cycler(Bio-Rad, USA)using25␮l reaction volume.The reaction mixture contained 12.5␮l Go Taq®Green Master Mix(Promega,Madison,WI),1␮l of 20␮M of each primer,and1␮l of DNA template.For DGGE analy-sis,PCR products were separated on8%(w/v)polyacrylamide gels (acrylamide–bisacrylamide[37.5:1])containing denaturing gradi-ents of45–65%for AOA,45–70%for AOB and35–65%for nirK using the Bio-Rad D-Code universal mutation detection system.A100% denaturant was defined as8%acrylamide containing7M urea and 40%deionized formamide.DGGE was performed using20␮l of the PCR product in1×TAE buffer at60◦C,200V for5min,then150V for7h(AOA and AOB)and110V for12h(nirK).Gels were silver stained(Sanguinetti et al.,1994)and scanned with gel document system(Bio-Rad,USA).2.6.Sequencing and phylogenetic analysisSome bands from DGGE gel of AOA,AOB and denitrifying com-munities were detected and numbered on the basis of their relative intensity or specific positions across all treatments.The num-bered bands with same mobility in the different lanes of DGGE gel of each gene were excised in triplicates.The excised bands were left to diffuse passively for24h at4◦C in30␮l sterilized dd H2O to elute the DNA.2␮l of recovered DNA was used as template for the PCR amplification under the same conditions as described above.The PCR amplified products were subjected to DGGE again to confirm their identity and ensure that all retrieved DGGE bands are single bands.The confirmed bands were fur-ther re-amplified and cloned to E.coli described as above in qPCR assay section and white colonies were selected for sequencing. After sequencing,we found that bands with same mobilities in the DGGE gel of each gene had same sequences.Therefore we used one sequence of each numbered band for phylogenetic anal-ysis.Sequences retrieved from the DGGE profiles of amoA(AOA and AOB)and nirK genes were compared with GenBank data base sequences using BLAST(Basic Local Alignment Search Tool) (http://www.ncbi.nlm.nih/gov/blast/)to search for best matches. The sequences of DGGE bands have been deposited in GenBank under the accession numbers HQ012641–HQ012645amoA(AOA), HQ020334–020341amoA(AOB)and HQ012630–HQ012634(nirK).2.7.DGGE profile analysisDGGE profiles of amoA(AOA and AOB)and nirK genes of all three replicate plots revealed highly reproducible results for each treat-ment(data not shown);therefore,the results for only one replicate are shown in the DGGE patterns(Figs.3–5).However,principal component analyses(PCA)of DGGE profiles have been made on three replicates to elucidate the microbial community structures based on relative band intensity and positions.Digitized DGGE images were analyzed with Quantity One image analysis software (Version4.0,Bio-Rad,USA).This software identifies the bands with the same position in the different lanes of the gel and also measures the intensity of identified bands.2.8.Data processing and statistical analysisA non-parametric analysis(Kruskal–Wallis)was performed to test the overall effect of plant growth stages on gene abundances. The use of this statistical method was justified by the small sample size and by the heterogeneity of the data.The Dunn procedure was used as post hoc test to check the differences between growing stages of rice plant(P<0.05)using Minitab v.15.3.Results3.1.Ammonia oxidizers(AOA and AOB)abundance and relative AOA:AOB ratiosThe copy numbers of amoA(AOA)gene in the paddy soil,ranging from1.2×106to4.5×106g−1dry weight of soil,were greater than those of amoA(AOB)gene,ranging5.5×105to3.1×106g−1dry weight of soil,during all growing stages(Fig.1).The AOB abundance in the rhizosphere varied significantly in response to rice grow-ing stages(Kruskal–Wallis test,P<0.05)while the AOA abundance in rhizosphere and bulk soil was unchanged over all rice growing stages(Kruskal–Wallis test,P>0.05).Population size of AOA and AOB increased in the rhizosphere soon after the rice transplanting with the maximum abundances recorded at the grainfilling stage (S2).Compared to bulk soil,archaeal amoA gene and bacterial amoA gene abundances in rhizosphere were1.7,3.3and2.5,and2.6,4.8 and4.6times higher respectively at S1,S2and S3stages.The rela-tive AOA:AOB ratios ranged from1.46to3.8for overall soil samples,Q.Hussain et al./Applied Soil Ecology 48 (2011) 210–21821312345(a m o A g e n e c o p i e s g -1 d r y s o i l × 106)AOAAOB Fig.1.Abundance of AOA (white)and AOB (shaded)amoA gene in rhizosphere (R)and bulk (B)soil at four rice growing stages (unplanted soil,S0;tillering,S1;grainfilling,S2;ripening,S3).Ratios of AOA to AOB amoA copies are shown in bold at top of columns with each treatment.Different capital letters indicate statistically significant differences among the rice growing stages for AOA (n =3;error bars are ±SD).Different small letters indicate statistically significant differences among rice growing stages for AOB (n =3;error bars are ±SD).showing the predominance of AOA throughout the period of rice growth (Fig.1).3.2.Denitrifying bacterial (nirK)abundance and relative nirK:amoA (AOA +AOB)ratiosThe denitrifier nirK gene abundance in the rhizosphere varied significantly with rice growing stages (Kruskal–Wallis test,P <0.05)while copy numbers of nirK gene in the paddy soil ranged from 2.4×107to 6.1×107g −1dry weight of soil (Fig.2).The nirK gene abundance was 2.6,2.0and 1.9times higher in the rhizosphere than that of the unplanted soil (S0)at S1,S2and S3growth stages,respectively.The nirK gene abundance was significantly higher in the rhizosphere at S1compared to S2,however no significant differ-ence was observed between S2and S3.The nirK gene copy numbers of the rhizosphere were 1.8,1.9and 2.1times higher compared to corresponding bulk soil at S1,S2and S3growth stages.The relative nirK :amoA (AOA +AOB)ratios ranged from 6.0to 13.9for overall soil samples,indicating that denitrifying bacterial populations were dominant relative to ammonia oxidizers in the paddy soil (Fig.2).The denitrifying bacterial (nirK )abundance relative to ammoniaoxidizers (amoA )decreased significantly in rhizosphere at repro-ductive stages (grain filling and ripening)compared to vegetative stage of tillering.3.3.Ammonia oxidizer (amoA)and denitrifying bacterial (nirK)communities structuresPrincipal component analysis (PCA)of DGGE profiles of AOA and AOB in rhizosphere and bulk soil at all growth stages (S1,S2and S3)gave good summaries of data,as 74.2%(AOA)and 78.7%(AOB)of the total variability was explained by the first two compo-nents (Figs.3A1and 4B1).PCA of AOA and AOB clearly showed that the unplanted soil (S0)was distinct from the other plant growth stages (S1,S2and S3).AOA community profiles of rhizosphere and bulk soil were indistinguishable among all growth stages,whereas the AOB patterns of the vegetative stage (tillering stage)were well separated from reproductive stage (grain filling and ripen-ing).Moreover,both rhizosphere and bulk soil showed no distinct separation in the grainfilling and ripening stages.PCA of DGGE profile further elucidated differences in denitri-fying bacterial (nirK )community structure betweenrhizosphere20406080( g e n e c o p i e s g -1d r y s o i l × 106)nirK amoA (AOA+AOB)Fig.2.Abundance of nirK gene (white)and amoA (AOA +AOB)gene in rhizosphere (R)and bulk (B)soil at four rice growing stages (unplanted soil,S0;tillering,S1;grainfilling,S2;ripening,S3).Ratios of nirK to amoA (AOA +AOB)copies are shown in bold at top of columns with each treatment.Different capital letters indicate statistically significant differences among the rice growing stages for amoA gene (n =3;error bars are ±SD).Different small letters indicate statistically significant differences among rice growing stages for nirK gene (n =3;error bars are ±SD).214Q.Hussain et al./Applied Soil Ecology48 (2011) 210–218Fig.3.DGGE profiles(A)and principal component analysis(A1)of archaeal-amoA gene fragments from rhizosphere(R)and bulk(B)soil at S0(no plant,NP),S1(tiller-ing),S2(grainfilling)and S3(ripening)stages.M:100bp ladder marker.Arrows indicate the excised bands(A1–A5)for sequencing on the basis of their relative intensity or specific positions over a course of plant growth.Similar symbols with same color in PCA plot indicate the replicate samples.and bulk soil at the growth stages(Fig.5N1).Thefirst two principal components(PC1and PC2)could explain76.3%of the total variance for soil denitrifying bacterial community structures in rice paddy over all growing periods.The denitrifying community structures of the rhizosphere at ripening stage(R3)showed clearly divergence from the profiles generated for the corresponding bulk soil.More-over,the denitrifying community structures of the unplanted soil (S0)were well separated from the rhizosphere and bulk soils of all growth stages(Fig.5N1).Denitrifying bacterial community struc-tures in the rhizosphere and the corresponding bulk soil at tillering stage(S1)were close to grainfilling(S2)but well separated from ripening stage(S3).3.4.Phylogenetic analysisSome specific bands of amoA(AOA and AOB)and nirK genes in DGGE profiles were selected and numbered(A1–A5,B1–B8and N1–N5)on the basis of their relative intensity and positions across all treatments(Figs.3–5).Although some bands were present in the profiles sampled at all growing stages,their intensity varied among treatments.The DGGE profile of AOA,AOB and nirK revealed that bands A1,A2,A3,A4,B1and B8were present in all grow-ing stages except the unplanted(S0),while A5,B5,B6,N1andN4Fig.4.DGGE profiles(B)and principal component analysis(B1)of bacterial-amoA gene fragments from rhizosphere(R)and bulk(B)soil at S0(no plant,NP),S1(tiller-ing),S2(grainfilling)and S3(ripening)stages.M:100bp ladder marker.Arrows indicate the excised bands(B1–B8)for sequencing on the basis of their relative intensity or specific positions over a course of plant growth.Similar symbols with same color in PCA plot indicate the replicate samples.were found in the profiles of all the soil samples.The intensity of band B3was strong in S1,while slight in other stages(S2and S3). The band N3was detected at all sampling time periods except S1. The band N5was present in all samples;however the intensity of band N5became strong with the development of plant at S3.The obtained sequences were subjected to BLAST search in the Gen-Bank database,which confirmed that all sequenced clones of DGGE profile represented amoA(AOA and AOB)and nirK like sequences (Table2).The BLAST analysis of the amoA(AOA)sequences obtained from DGGE gel bands showed high similarity(>97%)with uncul-tured Crenarchaeote and majority of those belonged to rice paddy soil.All gene sequences retrieved from our DGGE profile of amoA (AOB)were related to the class of‘␤-proteobacteria’.Most of amoA (AOB)sequences had the highest similarity(>97%)database hits to the uncultured AOB bacteria,isolated from soils and river sed-iments,affiliated with Nitrosospira or Nitrosospira-like sequences. Moreover,all sequences of denitrifying bacterial nirK showed their best matches(>82%)in the NCBI(National Center for Biotechnology Information)GenBank database with different uncultured strains from soil source(Table2).4.Discussions4.1.Rhizosphere effect on microbial communities involved in N turnoverA major focus in microbial ecology is to understand whether and how microbial communities in ecosystems interact.Q.Hussain et al./Applied Soil Ecology48 (2011) 210–218215Table2Nucleotide sequence BLAST results of DGGE amplicons generated using amoA(AOA),amoA(AOB)and nirK genes specific primers.Representative sequence Close NCBI blast matchBand Accession number Species name Accession number Similarity(%)Ammonia oxidizing archaea(amoA gene)A1HQ012641Uncultured Crenarchaeote a rice soil FN56252199A2HQ012642Uncultured Crenarchaeote a rice soil FN56253199A3HQ012643Uncultured Crenarchaeote a rice soil FN56251899A4HQ012644Uncultured Crenarchaeote a rice soil FN56252299A5HQ012645Uncultured Crenarchaeote a unfertilized red soil EF20721497Ammonia oxidizing bacteria(amoA gene)B1HQ020334Uncultured beta proteobacterium a soil aggregate fractions DQ48079399B2HQ020335Nitrosospira sp.CT2F a Cascade Mountains AY18914399B3HQ020336Uncultured beta proteobacterium a river sediment FJ15880999B4HQ020337Uncultured beta proteobacterium a river sediment FJ15876999B5HQ020338Uncultured beta proteobacterium a rice rhizosphere soil GU37731299B6HQ020339Uncultured beta proteobacterium a red soil EU79080797B7HQ020340Uncultured beta proteobacterium a red soil EU79080798B8HQ020341Uncultured beta proteobacterium a river sediment FJ15880498 Denitrifying bacteria(nirK gene)N1HQ012630Uncultured bacterium a ricefield soil AB45366098N2HQ012631Uncultured bacterium a soils AY67550192N3HQ012632Uncultured bacterium a soil from agricultural plots DQ78334582N4HQ012633Uncultured bacterium a red soil GU27052994N5HQ012634Uncultured bacterium a soils AY675477100a Isolated source/habitat;NCBI:National Center for Biotechnology Information;BLAST:Basic Local Alignment SearchTool.Fig.5.DGGE profiles(N)and principal component analysis(N1)of denitrifying bacterial-nirK gene fragments rhizosphere(R)and bulk(B)soil at S0(no plant,NP), S1(tillering),S2(grainfilling)and S3(ripening)stages.M:100bp ladder marker. Arrows indicate the excised bands(N1–N5)for sequencing on the basis of their rel-ative intensity or specific positions over a course of plant growth.Similar symbols with same color in PCA plot indicate the replicate samples.Functional groups of microorganisms like AOA,AOB and deni-trifiers are crucial mediators of N cycling in the rice paddy soil and can thereby affect plant growth by competing for nutrients. Therefore,the interaction between plant and rhizosphere microor-ganisms involved in N turnover is of special interest in this study. In the current study,rice plant stimulated higher AOA,AOB and denitrifying bacterial abundances in rhizosphere and structures of ammonia oxidizers(AOA and AOB)and denitrifying bacte-ria were also different between planted and unplanted soils.In fact,rice plantation may affect the physical–chemical properties and the biological parameters of the rhizosphere by continuously producing and excreting organic compounds through rhizode-position(Hinsinger et al.,2006).The quantity,composition and spectra of root exudates are considered to lead the development of plant-specific microbial communities in root-associated habi-tats(Kowalchuk et al.,2002).The fact that the number of amoA (AOA)gene,amoA(AOB)gene and nirK gene copies in the rhizo-sphere were significantly higher than in the corresponding bulk soil at all growth stages(Figs.1and2)may be attributable to the rhizodeposition of carbohydrates from plant roots favoring micro-bial growth in comparison with that in the bulk,a phenomenon well known as‘rhizosphere effect’(Smalla et al.,2001;Dunfield and Germida,2003).It is also well known that rice roots release oxygen through aerenchymatous tissue at rates sufficient to sup-port aerobic microbial processes in the rhizosphere(Bedford et al., 1991).Moreover,qPCR in combination with PCR-DGGE provided reproducible metric to monitor gross differences and changes in microbial population size and structure between rice planted and unplanted soils.4.2.Ammonia oxidizers dynamics and relative AOA:AOB ratios in the rice rhizosphereRecentfindings have extended the known ammonia-oxidizing prokaryotes from the domain Bacteria to Archaea.However,in the complex rice ecosystem it remains unclear whether AOA or AOB are exclusively or predominantly linked to prevailing plant and soil conditions over a rice crop season.In line with ourfindings,Chen et al.(2008)reported that rice cultivation under microcosm exper-iment led to greater abundance of AOA relative to AOB amoA gene。