通过实时PCR技术定量检测环境样品中的新型硝酸盐还原细菌

Quantification of a novel group of nitrate-reducing bacteriain the environment by real-time PCRJuan C.Lo´pez-Gutie ´rrez,Sonia Henry,Ste ´phanie Hallet,Fabrice Martin-Laurent,Ge´rard Catroux,Laurent Philippot *UMR 1229INRA-Universite´de Bourgogne,Microbiologie et Ge ´ochimie des Sols,17,rue Sully,B.P .86510,21065Dijon Cedex,France Received 3February 2004;received in revised form 13February 2004;accepted 13February 2004Available online 20March 2004AbstractNitrate reduction is performed by phylogenetically diverse bacteria.Analysis of narG (alpha subunit of the membrane boundnitrate reductase)trees constructed using environmental sequences revealed a new cluster that is not related to narG gene from known nitrate-reducing bacteria.In this study,primers targeting this as yet uncultivated nitrate-reducing group were designed and used to develop a real-time SYBR R Green PCR assay.The assay was tested with clones from distinct nitrate-reducing groups and applied to various environmental samples.narG copy number was high ranging between 5.08Â108and 1.12Â1011copies per gram of dry weight of environmental sample.Environmental real-time PCR products were cloned and sequenced.Data was used to generate a phylogenetic tree showing that all environmental products belonged to the target group.Moreover,16S rDNA copy number was quantified in the different environments by real-time PCR using universal primers for Eubacteria .16S rDNA copy number was similar or slightly higher than that of narG ,between 7.12Â109and 1.14Â1011copies per gram of dry weight of environmental sample.Therefore,the yet uncultivated nitrate-reducing group targeted in this study seems to be numerically important in the environment,as revealed by narG high absolute and relative densities across various environments.Further analysis of the density of the nitrate-reducing community as a whole by real-time PCR may provide insights into the correlation between microbial density,diversity and activity.D 2004Elsevier B.V .All rights reserved.Keywords:Nitrate reductase;narG ;Real-time PCR;Soil;16S rDNA1.IntroductionNitrate is the main nitrogen source for plant growth but can also be a contaminant in the envi-ronment.A taxonomically wide group of bacteria can reduce nitrate into nitrite,which can be reducedinto gaseous nitrogen compounds by denitrificationor into NH 4+by dissimilatory nitrate reduction (Tiedje,1988).The ability of facultative anaerobic bacteria to respire nitrate when oxygen is limiting has been ascribed mainly to the activity of a mem-brane bound nitrate reductase encoded by the narGHJI operon (Philippot and Hojberg,1999).Because of the importance of nitrate-reducing bacte-ria in the nitrogen cycle,this functional group has been studied either by cultivation (Nijburg and0167-7012/$-see front matter D 2004Elsevier B.V .All rights reserved.doi:10.1016/j.mimet.2004.02.009*Corresponding author.Tel.:+33-3-80-69-33-46;fax:+33-3-80-69-32-24.E-mail address:philippo@dijon.inra.fr (L.Philippot)/locate/jmicmethJournal of Microbiological Methods 57(2004)399–407Laanbroek,1997;Nijburg et al.,1997)or cultivation independent(Cheneby et al.,2003;Gregory et al., 2000;Philippot et al.,2002)microbiological approaches.In recent studies,generating narG se-quence data from the environment by clone libraries construction,narG trees showed the predominance of a new cluster representing between41%and52% of the total number of clones which is not related to any narG gene characterized from cultivated nitrate-reducing bacteria(Cheneby et al.,2003;Mounier et al.,2004;Philippot et al.,2002).Whereas this group may take significant part to the nitrate-reducing activity in the environment,no information could be deduced from these studies on its abundance. Traditional approaches using viable cell enumeration on solid media are biased by the need of cultivation, which is undesirable to estimate the abundance of this unknown group of nitrate-reducing bacteria.Real-time PCR is a method based on the use of fluorogenic probes(Nitsche et al.,1999)or dyes (Morrison et al.,1999)to quantify the copy number of target DNA in a sample.This technology,which is not limited by the culturability of bacteria,has been already successfully applied to quantify functional genes as markers for bacteria in the environment. For instance,peptidase and methane monooxygenase genes were used to quantify proteolytic bacteria (Bach et al.,2002)and methanotrophs(Kolb et al., 2003),respectively,in environmental samples.Thus, the aim of this study was to use real-time PCR for culture-independent quantification of the copy num-ber of these new nitrate reductase encoding genes for as yet uncultivated bacteria in the environment in relation to the number of16S rDNA copies.For this purpose,a new set of partially degenerated narG primers was designed and tested in vitro and applied on various environmental samples.2.Materials and methods2.1.Environmental samplesLa Bouzule(LB)soil(15.4%sand,51.3silt%, 33.3%clay,pH5.8and15.3%C)was obtained from the first20cm of the top layer of a wheat planted plot in an experimental field of the ENSAIA domain of La Bouzule(Nancy,France).Vannecourt(V)soil(33.2%sand,44.3%silt,22.5%clay,pH6.0,1.6%C and 0.2%N)was obtained from a winter wheat agricul-tural field in Vannecourt(Moselle,North East France). Vezin le Coquet(VLC)soil(12%sand,75%silt,13% clay and pH6.4)and two earthworm modified soil compartments,cast(VLCC)and gallery(VLCG), were obtained from Vezin Le Coquet(Bretagne, France)(Kersante,2003).Freshwater sediment(S) was obtained from the surface2cm of sediment of Ille et Vilaine river in Rennes,France.Paris(P)soil (77%sand,10%silt,13%clay,pH7.7,1.1%C and 0.09%N)was obtained from garden soils close to Paris(Baize et al.,2003).Finally,Auxonne(A)soil (7%clay,8%silt and85%sand,pH6.4,1.0%C and 0.05%N)was obtained from a cultivated field in Auxonne(Burgundy,France).Samples were trans-ported at4j C and kept atÀ20j C until use.2.2.DNA extractionDNA was extracted from three250-mg aliquots of samples according to the method of Martin-Laurent et al.(2001).Briefly,samples were homogenized in1ml of extraction buffer for30s at1600rpm in a mini-bead beater cell disrupter(Mikro-Dismembrator S;B. Braun Biotech International).Soil and cell debris were removed by centrifugation(14,000Âg for5min at4 j C).Afterwards,5M sodium acetate precipitation was used to remove proteins.Nucleic acids were then precipitated with cold isopropanol,washed with70% ethanol,and purified through a Sepharose4B spin column.DNA quality and size was checked by1% (wt/v)agarose gel electrophoresis.Finally,DNA was quantified spectrophotometrically at260nm using a BioPhotometer(Eppendorf,Hamburg,Germany),and by comparison to DNA standards in1%(wt/v)aga-rose gel electrophoresis.2.3.Primer design and testingSequences from a previously described narG environmental cluster(Cheneby et al.,2003;Mou-nier et al.,2004;Philippot et al.,2002)were aligned and scanned for conserved regions that could provide suitable primer target sites.Two degenerated primers, narG1960m2f(5V-TA(CT)GT(GC)GGG CAG GA(AG)AAA CTG-3V)and narG2050m2r(5V-CGT AGA AGA AGC TGG TGC TGT T-3V)wereJ.C.Lo´pez-Gutie´rrez et al./Journal of Microbiological Methods57(2004)399–407 400designed to amplify a 110bp narG fragment fromthe targeted cluster.The primers were tested by real-time PCR amplification of two Eubacteria and clones from a previously isolated collection (Mou-nier et al.,2004;Philippot et al.,2002)belonging to the same and other narG cluster groups (Table 2)using DNA concentrations corresponding to narG copy numbers two logs above the real-time PCR detection limit (see Section 2.4).Primers were then used for real-time PCR amplification in environmen-tal samples.For 16S rDNA amplification two universal primers specific for members of Eubacteria 341f (5V -CCT ACG GGA GGC AGC AG-3V )and 515r (5V -ATT CCG CGG CTG GCA-3V )were used to amplify a 174bp region.2.4.Real-time PCR assayReal-time PCRs were carried out in a Smart Cycler (Chypheid R ,USA).The 25A l reaction mixtures contained 0.5A M of each primer for narG ,and 0.3A M for 16S rDNA amplification,12.5A l of SYBR Green PCR master mix,including HotStar Taq k DNA Polymerase,Quanti Tec SYBR Green PCR Buffer,dNTP mix with dUTP,SYBR Green I,ROX and 5mM MgCl 2(QuantiTect k SYBR R Green PCR Kit,QIAGEN,France),2.5A l of DNA diluted tem-plate corresponding to 25ng of total DNA,and RNase-free water to complete the 25A l volume.The conditions for narG real-time PCR were 120s at 50j C for carryover prevention,900s at 95j C for enzyme activation;afterwards six pairs of touchdown cycles were added as follows:15s at 95j C for denaturation 30s at 63j C for annealing,30s at 72j C for extension and 15s at 81j C (temperature above the melting point of primer dimers)for a final data acquisition step,progressively decreasing the anneal-ing temperature by 1j C down to 58j C,to finally replicate the last step 40times.The conditions for 16S rDNA real-time PCRs were 120s at 50j C for carryover prevention,900s at 95j C for enzyme activation and 30cycles of 15s at 95j C,30s at 60j C and 30s at 72j C for denaturation,annealing and extension steps,respectively.Fluorescence produced by the binding of the SYBR Green fluorochrome to the double-strand DNA was measured at the end of each data acquisition step for narG and each exten-sion step for 16S rDNA (excitation 450–495nm,emission 505–537nm).An additional step from 60j C to 95j C (0.2j C s À1)was added in both cases to obtain a denaturation curve specific for each amplified sequence.Purity of amplified products was confirmed by the observation of a fluorescence fall at the fusion temperature of the product and the presence of a band of the correct size in a 2%(wt/v)agarose gel stained with ethidium bromide.The increment in fluorescence versus reaction cycle was plotted and the threshold cycle (Ct)was obtained by manually positioning the threshold base-line at 30fluorescence units of the first derivative curve.Ct could be related to the log of the target gene copy number by the following formula:log T 0=Àlog E ÂCt +log K ,where E is the PCR efficien-cy,T 0is the initial amount of DNA and K is the calculated initial amount of DNA for a Ct value of0.Fig.1.Agarose gel (2%wt/v)electrophoresis of real-time PCR amplification products using partially degenerated narG primers specific for the target group in DNA extracted from different environments (see Table 2for abbreviations).A 25bp ladder was used as DNA size marker.J.C.Lo ´pez-Gutie ´rrez et al./Journal of Microbiological Methods 57(2004)399–407401J.C.Lo´pez-Gutie´rrez et al./Journal of Microbiological Methods57(2004)399–407 402Standard curves were obtained by plotting Ct as a function of log of the copy number of target DNA. For this purpose recombinant pGEM-T Easy Vector (Promega)plasmids containing the standard sequen-ces were linearized using10U of Sal I and incubat-ing at37j C for1h(Appligene-Oncor,Illkirch, France).Tenfold serial dilutions of the linearized plasmids ranging from101to108were used as template,by triplicate,to determine the calibration curves.Three points of the standards were used as positive controls in each reaction.Fivefold dilution of environmental DNA was also used for determina-tion of the detection limit of our narG real-time PCR assay.In addition,1Â106copies of the target gene in environmental were added in serial dilutions of environmental DNA to check for real-time PCR inhibition.2.5.Cloning and sequencing of real-time PCR narG productsReal-time PCR products from one replicate of each of the different environments studied were cloned using pGEM-T Easy Vector System(Prom-ega,France).Three to six nucleotide sequences fromeach environment were determined using DTCS-1kit (Beckman,Coulter)with primer T7in a Ceq2000 XL Sequencer,according to manufacturer instruc-tions(Beckman,Coulter).Sequences obtained were verified by BLAST,aligned using CLUSTALX soft-ware V.1.0.1(Thompson et al.,1997)and con-structed into a phylogenetic tree using the neighbor-joining method.All resulting sequences were deposited in Genbank under accession numbers AY53175AY531109.2.6.Statistical analysisWe performed one way analysis of variance (ANOV A)in each environment to observe differ-ences in narG and16S rDNA copy number and ratios.Data were log or square root transformed to meet the ANOVA premises.Means were compared using the least significant difference(LSD)test at P<0.05.3.Results3.1.Primer specificityThe primer sets used generated the expected amplicons of110bp for narG and174bp for 16S rDNA,as corroborated by gel electrophoresis analysis(data not shown for16S rDNA).Gel electrophoresis analysis of narG real-time PCR products in environmental samples(Fig.1)showed only bands of the expected size,with the exception of some replicates(VLCC-2,V-2and V-3),how-ever,the presence of non-specific faint bands may be attributed to primer dimerization(Becker et al., Table1Real-time PCR amplicons using partially degenerated narG primers in clones and bacteria from different phylogenetic groupsClone a or bacteria Phylogeneticgroup bSpecific generationof ampliconE611ÀA722ÀE682ÀD882ÀD143ÀEscherichia coli JM1013ÀC304ÀE384ÀD134ÀD714ÀPseudomonasfuorescens C7R125ÀD605ÀGH106+GG536+A736+C436+D446+a Clones belong to previously PCR libraries amplified from agricultural soils(Mounier et al.,in press;Philippot et al.,2002).b Phylogenetic groups correspond to the cluster groups defined in Fig.2.Fig.2.Phylogenetic analysis based on environmental narG nucleotide sequences obtained in this study(text in bold,see Table2for abbreviations),previous studies(Cheneby et al.,2003;Mounier et al.,2004;Philippot et al.,2002)and closely related bacteria.Phylogenetic distances were determined by neighbor-joining method.J.C.Lo´pez-Gutie´rrez et al./Journal of Microbiological Methods57(2004)399–4074032000;Kolb et al.,2003)and did not modify narGcopy number quantification.We further tested the specificity of the narG primers by real-time PCR using two Bacteria and previously isolated clones (Mounier et al.,2004;Philippot et al.,2002)from the same and different phylogenetic cluster groups (Fig.2).Table 1shows that only clones from cluster 6resulted in a positive Ct value,i.e.,clones from our target,not yet cultivated,nitrate-reducing group were selectively amplified by our designed narG primer set.3.2.Sensitivity of the real-time PCR assay and detection limitStandard curves generated for 16S rDNA (log (16S rDNA)=Àlog0.287Ct +12.129,r 2=0.997)and narG (log (narG )=Àlog0.272ÂCt +16.893,r 2=0.993)target molecules were linear from 1Â104to 1Âparisons between real-time PCR results obtained with different primer sets can be achieved if PCR efficiency is similar among the different reac-tions (Devers et al.,2004).Both curves had a highcorrelation coefficient and similar slopes,and thus similar PCR efficiencies (E ),which allowed the use of the curves as standards and the comparison among the results obtained with both primer sets.Analysis of real-time PCR amplification on serial dilutions of extracted environmental DNA corroborat-ed that the detection limit was 104for environmental samples (data not shown).By adding 1Â106copies of the target gene in environmental DNA extracts diluted bellow the detection limit,we were able to recover the expected number of copies in all en-vironments studied.We obtained an average of 8.5Â105F 8.33Â105narG copies per reaction out of the 1.01Â106narG copies expected.3.3.narG and 16S copy number and ratios in different environments16S rDNA target molecules,ranging from 5.06Â103to 1.00Â105copies per nanogram of DNA and 7.12Â109to 1.14Â1011copies per gram of dry weight of environmental sample,were signifi-cantly different between some of the environments.Table 2Amplicon nucleotide identity,gene copy numbers and ratios of narG from target group and 16S rDNA from total Eubacteria in environmental samples determined by real-time PCR Environment Nucleotide Gene copy number per nanogram DNA Ratio Gene copy number per gram dry weight Ratioidentity a (%)narG 16S rDNA narG /16S bnarG 16S rDNA narG /16S c La Bouzule (LB)80 1.01Â103a4.20Â104b 0.02a5.36Â109b 2.34Â1011d 0.02a (0.21Â103)(3.05Â104)(1.95Â109)(2.01Â1011)Vannecourt (V)77–78 5.95Â104e6.67Â104b 0.89c 1.12Â1011d 1.14Â1011cd 0.98c (3.67Â104)(7.58Â104)(0.71Â1011)(1.08Â1011)Vezin le Coquet 81–849.92Â103cd8.70Â103a 1.14bc 3.63Â1010c 3.12Â1010b 1.16bc (VLC)(1.72Â103)(2.75Â103)(0.47Â1010)0.34Â1010Vezin le Coquet 80–85 2.96Â103b 5.49Â103a 0.54b 1.95Â1010c 3.77Â1010b 0.52b cast (VLCC)(0.87Â103)(0.72Â103)(0.10Â1010)(0.86Â1010)Vezin le Coquet 77–86 2.94Â103b 5.06Â103a 0.58bc 4.59Â109b 7.98Â109a 0.58bc Gallery (VLCG)(1.62Â103)(1.48Â103)(2.30Â109)(2.14Â109)Sediment (S)80–82 1.71Â104d 2.44Â104b 0.70bc 3.70Â1010c 6.10Â1010bc 0.61bc (0.95Â104)(0.45Â104)(1.57Â1010)(2.91Â1010)Paris (P)83–85 5.39Â104e 4.69Â104b 1.15bc 7.96x 109b 7.12Â109a 1.12bc (1.59Â104)(0.37Â104)(1.19Â109)(0.96Â109)Auxonne (A)81–856.04Â103bc 1.00Â105c 0.06a5.08Â108a 8.47Â109a 0.06a(1.10Â103)(0.04Â105)(0.95Â108)(0.77Â109)Values followed by the same letter within columns do not significantly differ according to LSD test (P <0.05).aRange of the percentage of nucleotide identity between environmental narG sequences and narG from P .aeruginosa PA01.bTaking into account gene copy number per nanogram of extracted DNA.cTaking into account gene copy number per gram of dry weight of environmental sample.J.C.Lo´pez-Gutie ´rrez et al./Journal of Microbiological Methods 57(2004)399–407404narG target molecules were nearly as abundant as the 16S rDNA,ranging from1.01Â103to5.95Â104 copies per nanogram of DNA and 3.94Â108to 1.12Â1011copies per gram of dry weight of environ-mental sample.They were also significantly different among some of the environments studied.In fact,16S rDNA copy number per gram of dry weight of envi-ronmental sample were significantly higher in agricul-tural soils(LB and V)than in sediment(S)and sandy soil environments(P and A),while narG copy number did not follow the same pattern.Moreover,the pro-portion of narG with respect to16S rDNA copy number significantly differed across environments, ranging from0.02to1.15for gene copy number per weight of extracted DNA,0.02to1.16for gene copy number per gram of dry weight of environmental sample(Table2).3.4.Environmental narG sequence analysisThe35sequences obtained from the cloning of environmental real-time PCR products clustered to-gether with known narG sequences of the targeted phylogenetic cluster group(Fig.2).Phylogenetic tree in Fig.2shows that all the checked sequences belong to cluster6.The percentage of nucleotide identity between environmental isolated sequences and the characterized narG sequence of Pseudomonas aeru-ginosa PA01ranged between77%and86%in the different environments(Table2).4.DiscussionIn this study,a set of partially degenerated primers was designed for real-time PCR amplification of narG of an as yet uncultivated nitrate-reducing target group. The assay allowed the detection of as little as104 narG target molecules and amplification was not significantly inhibited in any of the environments studied.Specificity of our primers was confirmed by the phylogenetic analysis of35nucleotide sequences recovered from the different environments(Fig.2). The high narG copy number(between108and1011 copies per gram of dry weight of environmental sample),quantified by real-time PCR using SYBR R Green as detection system,confirms previous results suggesting that this unknown bacterial group with the genetic potential to reduce nitrate was numerically important in the environment(Cheneby et al.,2003; Mounier et al.,2004;Philippot et al.,2002).Thus,it has been shown that in narG clone libraries from three different agricultural soils,the proportion of clones belonging to this unknown cluster of nitrate reducing bacteria ranged between41%and51%(Cheneby et al.,2003;Mounier et al.,2004;Philippot et al.,2002). In this study,copy number of narG in the different environments tested is2log lower up to the same order of magnitude than16S rDNA copy number which further shows the high density of the group with respect to total bacteria.In addition,narG/16S rDNA ratios between0.02and1.16partially agree with the fact that10–50%of the bacteria present in the environment are able to reduce nitrate into nitrite (Tiedje,1988).Incidentally,a NarG sequence BLAST against bacterial genomes(including unfinished genomes)shows the presence of the gene in23%of the cases.Nevertheless,narG copy number does not have a direct correlation with16S rDNA copy num-ber,i.e.,environments with low copies of16S rDNA (e.g.,sandy soils)do not necessarily have a lower narG copy number and vice versa.Significant differ-ences in narG/16S rDNA ratios corroborate the latter observation.Thus,it may be possible that nitrate-reducing genetic potential is selectively advantageous in some environments and/or specific environmental conditions(Ghiglione et al.,2000).When comparing the16S rDNA copy number per amount of DNA,instead of amount of dry weight,we find slightly lower values than the1.25to1.32Â106 copies found in agricultural soils(Ochsenreiter et al., 2003).Similarly,when taking into account the16S rDNA copy number per gram of dry weight we obtain numbers that agree with bacterial counts obtained by non-molecular methods such as DNA-intercalating, blue fluorescent dye4V,6V-diamidino-2-phenylodole (DAPI).For instance,Uhlirova and Santruckova (2003)reported up to6.3Â109bacterial cells per gram of forest soil;Taylor et al.(2002)found up to 2.74Â1010cells per gram of agricultural soil,while Straub and Buchholz-Cleven(1998)reported 2.9Â1010cells per gram of fresh water sediment.To determine the importance of a specific bacterial functional group we need to consider the actual number of these organisms with respect to the total present in the environment.Gene copy number canJ.C.Lo´pez-Gutie´rrez et al./Journal of Microbiological Methods57(2004)399–407405provide a rough estimate of the actual number of organisms if the number of copies of the genes per genome and the number of genomes per organisms are known(Bach et al.,2002;Hermansson and Lindgren,2001).In the case of narG,the number of genes per organism may be close to the gene copy number obtained by real-time PCR since bacteria are expected to have one to three copies of this gene per genome(Philippot,2002).Nonetheless,the estimation of total bacteria using16S rDNA copy number is more cumbersome since different bacterial groups have different gene copy number ranging from1to 13(Fogel et al.,1999).Environmental bacterial number estimation using real-time PCR may also be biased by the DNA extrac-tion method.DNA isolated from complex environ-ments,such as soil,can contain inhibitory substances that may interfere with PCR amplification(Martin-Laurent et al.,2001).Moreover,since DNA released by organisms can persist(Recorbet et al.,1993)and be absorbed on soil components(Demaneche et al.,2001), coextraction and later PCR amplification of such DNA may cause an overestimation of gene copy number. However,in our study,we assumed that the use of gene copy number ratios allows for some of these biases to be negligible because both,narG and16S rDNA quantification,are subjected to the same DNA extrac-tion and amplification biases.In conclusion,the use of partially degenerated narG primers for real-time PCR amplification pro-vided a suitable and rapid cultivation independent method for the quantification of as yet uncultivated group of nitrate-reducing bacteria with unknown physiological features.Application of this technique to various environments confirmed the high density of this target group and emphasized the fact that we need information on its actual activity.Techniques targeting the mRNA,such as reverse transcription real-time PCR amplification,may help to monitor metabolic activity of these nitrate-reducers in the environment(Nogales et al.,2002).Knowledge of the density of functional community is essential for evaluating correlation between microbial diversity, density and activity in the environment.Therefore, it is now important to design primers targeting the other groups of nitrate-reducing bacteria to apply real-time PCR for the quantification of the whole nitrate-reducing community.AcknowledgementsThe authors would like to gratefully acknowledge Lionel Ranjard and Francßoise Binet for providing DNA from Auxonne and Vezin Le Coquet,respec-tively.Juan Carlos Lo´pez Gutie´rrez was funded by the Conseil Re´gional de Bourgogne(France)no.02 514AA02S24.ReferencesBach,H.-J.,Tomanova,J.,Schloter,M.,Munch,J.C.,2002.Enu-meration of total bacteria and bacteria with genes for proteolytic activity in pure cultures and in environmental samples by quan-titative PCR mediated amplification.J.Microbiol.Methods49, 235–245.Baize,D.,Lamy,I.,Oort,F.V.,Bourennane,H.,Chaussod,R., 2003.7th International Conference on the Biogeochemestry of Trace Elements,Uppsala.Becker,S.,Boger,P.,Oehlmann,R.,Ernst,A.,2000.PCR bias in ecological analysis:a case study for quantitative Taq nuclease assays in analysis of microbial communities.Appl.Environ.Microbiol.66,4945–4953.Cheneby,D.,Hallet,S.,Mondon,M.,Martin-Laurent,F.,Germon, J.C.,Philippot,L.,2003.Genetic characterization of the nitrate reducing community based on narG nucleotide sequence analy-sis.Microb.Ecol.46,113–121.Demaneche,S.,Jocteur-Monrozier,L.,Quiquampoix,H.,Simonet, P.,2001.Evaluation of biological and physical protection against nuclease degradation of clay-bound plasmid DNA.Appl.Environ.Microbiol.67,293–299.Devers,M.,Soulas,G.,Martin-Laurent,F.,2004.Real-time PCR reverse transcription PCR analysis of expression of atrazine catabolism genes in two bacterial strains isolated from soil.J.Microbiol.Methods56,3–15.Fogel,G.B.,Collins,C.R.,Li,J.,Brunk,C.F.,1999.Prokariotic genome size and SSU rDNA copy number:estimation of mi-crobial relative abundance from a mixed population.Microb.Ecol.38,93–113.Ghiglione,J.F.,Gourbiere,F.,Potier,P.,Philippot,L.,Lensi,R., 2000.Role of respiratory nitrate reductase in ability of Pseudo-monas fluorescens YT101to colonize the rhizosphere of maize.Appl.Environ.Microbiol.66,4012–4016.Gregory,L.G.,Karakas-Sen,A.,Richardson,D.J.,Spiro,S.,2000.Detection of genes from membrane-bound nitrate reductase in nitrate-respiring bacteria and in community DNA.FEMS Micro-biol.Lett.183,275–279.Hermansson,A.,Lindgren,P.,2001.Quantification of ammonia-oxidizing bacteria in arable soil by real-time PCR.Appl.Envi-ron.Microbiol.67,972–976.Kersante, A.,2003.Roˆle re´gulateur de la macrofaune lombri-cienne dans la dynamique de l’herbicide atrazine en sol cultive´tempe´re´.The`se de doctorat Universite´de Rennes1.166pp.J.C.Lo´pez-Gutie´rrez et al./Journal of Microbiological Methods57(2004)399–407 406。