Substrate-induced gene expression (SIGEX) screenin

Taku Uchiyama 1&Kazuya Watanabe 2–4

Institute for Biological Resources and Functions,National Institute of Advanced Industrial Science and Technology (AIST),Tsukuba,Ibaraki 305-8561,Japan.

2Laboratory

of Applied Microbiology,Marine Biotechnology Institute,3-75-1,Heita,Kamaishi,Iwate,Japan.3Hashimoto Light Energy Conversion Project,Exploratory

Research for Advanced Technology,Japan Science and Technology Agency,Hongo,Bunkyo-ku,Tokyo 113-8656,Japan.4Research Center for Advanced Science and Technology,The University of Tokyo,Komaba,Meguro-ku,Tokyo 153-04,Japan.Correspondence should be addressed to T.U.(uchiyama-taku@aist.go.jp).Published online 3July 2008;doi:10.1038/nprot.2008.96

Substrate-induced gene-expression screening (SIGEX)has been developed for isolating novel catabolic genes from environmental metagenomes,particularly genes that are difficult to obtain using conventional gene-cloning methods.In SIGEX,restriction enzyme-digested metagenome fragments are ligated into an operon-trap vector (e.g.,p18GFP),and a library is constructed in a liquid culture by transforming a cloning host (e.g.,Escherichia coli ).The library is subjected to a substrate-dependent gene-induction assay,and positive cells are selected by detecting activity of a co-expressed marker (e.g.,GFP)encoded in the vector.High-throughput screening is possible if fluorescence-activated cell sorting (FACS)is used to select GFP-expressing cells.The abovementioned SIGEX procedure requires B 17d.In this protocol,a widely applicable SIGEX scheme is presented along with typical experimental results.

INTRODUCTION

Prokaryotes exist in every part of our planet and contribute to its geochemistry,cycling of elements and breakdown of wastes.This ability of prokaryotes relies on their huge genetic and metabolic diversity,which also infers great potential for their application to biotechnological purposes,for example,production of pharma-ceutical drugs,enzymes,polymers and microbial agents.Nowadays,many institutes and companies have established collections of prokaryotes (mainly bacteria)that have been isolated from a wide range of environments.These prokaryotes have been used for screening because of their abilities to produce medicinally relevant metabolites and industrially useful enzymes,for example.However,recent developments in molecular microbial ecology have shown that the prokaryotes that have been isolated and cultivated by conventional microbiological techniques represent only a small portion of the estimated total.For example,scientists have estimated that cultivable prokaryotes form o 1%of total prokar-yotes present in the environment 1,2,indicating that the huge genetic diversity of prokaryotes in the environment has not yet been explored sufficiently.

Given the limitations of current culture-based approaches for exploring natural genetic diversity,researchers have started using ‘metagenomic approaches’3,4.A metagenome represents a mixture of genomes of multiple organisms,particularly mixed microbial genomes extracted from an environmental sample.In this approach,a metagenome is directly used to screen for obtaining genes of interest without isolating and cultivating individual microorganisms.In some cases,the environmental sample was subjected to enrichment culture to increase the abundance ratio of microbes with a desired function,followed by a metagenomic screening of the enrichment sample 5.Current metagenome approaches have used either of the two conventional screening methods for isolating catabolic genes,namely,enzyme activity-based screening 6and nucleotide sequence-base screening 7.In addition to these methods,we have recently introduced a third option for isolating novel catabolic operons,which uses substrate-induced gene-expression screening (SIGEX)8.

SIGEX is a method to screen a metagenome library for obtaining catabolic genes whose expression is induced in response to environmental stimuli,such as the occurrence of a chemical com-pound 8.This is based on knowledge that catabolic gene expression is generally induced by relevant compounds (substrates or metabo-lites)and,in many cases,controlled by regulatory elements situated proximate to catabolic genes.T able 1summarizes advantages and limitations of the three screening strategies (namely,enzyme activ-ity-based,nucleotide sequence-based and gene expression-based (SIGEX)strategies).As presented in this table and discussed else-where 9,10,each screening method has intrinsic biases,indicating that there is no single method that can isolate a wide variety of genes from environmental metagenomes.It is therefore important for researchers to sufficiently understand these advantages and limita-tions and select an appropriate screening method.SIGEX is con-sidered to be suitable for obtaining novel catabolic genes whose enzymatic activities are difficult to detect in a cloning host (e.g.,those that require supplemental factors)and also for isolating novel genes whose genetic sequences may be largely different from known genes.The SIGEX scheme is also applicable to screening of a genomic library for isolated bacteria,particularly when target genes are difficult to obtain by conventional gene-cloning methods;this application has been reported with an example of isolating phenol hydroxylase genes from a phenol-degrading strain 8.The outline and time schedule of the SIGEX scheme are presented in Figure 1,and the procedure is described in detail below.

Experimental design

A crucial step in the SIGEX procedure is the design of the operon-trap vector that should be suitable for shotgun cloning and high-throughput screening (in particular,when a metagenome library is screened).It is also important to bear in mind that the selection of a cloning host (thereby the selection of a replication system)is pivotal to achieve successful application of the SIGEX concept because there will be preferences between transcriptional regulators (encoded in a metagenome fragment)and RNA polymerases (in a

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h 1202|VOL.3NO.7|2008|NATURE PROTOCOLS

cloning host).Before starting the metagenome treatment,experi-mental systems (selection of a cloning host,vector construction,developments of conditions for a gene-expression induction and high-throughput screening,e.g.,fluorescence-activated cell sorting (FACS))should be established and fully checked.

Choice of fluorescence protein.GFP and most of its spectral variants can be used in combination with FACS,and high-fluorescence variants are useful for mining weakly expressed genes.Table 2sum-marizes the fluorescence proteins com-monly used in FACS.

Cloning host.We have performed SIGEX

using Escherichia coli cells as the cloning host,while other organisms may also be applicable to SIGEX.We have found that SIGEX tends to select metagenome frag-ments whose original hosts (in the environ-ment)are related to a SIGEX cloning host.Use of another bacterium as a cloning host may be useful for obtaining as-yet-unex-plored genes from metagenomes.Cloning vector.An operon-trap vector (used to clone a gene fragment that induces the expression of a downstream marker gene (e.g.,gfp ))is used for constructing a metagenome library.An example is p18GFP (Fig.2),which is quite suitable for this purpose.This vector includes genes for GFP that are used as a coexpressed marker whose activity (fluorescence)is used to detect clones that contain metagenome fragments with gene-expression activities in response to a substrate.It also includes the lac promoter (P lac )that is used to detect clones harboring the self-ligation vector.In this vector,a cloning site (the Bam HI site)exists in front of three-frame stop codons,a

ribosome-binding site (RBS)and the start codon of gfp ,which prevents formation of GFP-fusion proteins.T o construct a custom SIGEX vector,one should take the following three points into account:(i)multicopy plasmids have to be used as a base construct;(ii)a commonly used inducible promoter is inserted in front of gfp .This is used for eliminating clones harboring self-ligated plasmids;(iii)introduce a cloning site in front of the three-frame stop condons,RBS and gfp gene.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h TABLE 1|Advantages and limitations of metagenome screening methods.Method

Description

Advantage

Limitation

Nucleotide sequence-based screening

Primers and probes used for screening are designed from known gene sequences

(mostly those cloned from easily cultivable bacteria)

High-throughput PCR cloning is possible

Only genes homologous to known genes can be obtained Genes obtained are partial

Enzyme activity-based screening

An activity expressed by a trans-formed host cell (e.g.,an enzyme activity)is detected and used for selecting positive clones Gene fragments that are sufficient to express enzymatic activities can be obtained

Many enzymes are difficult to be expressed in a heterogeneous host as an active form It is generally laborious

Gene expression-based screening (substrate-induced gene-expression screening)

A gene-expression activity of a metagenome fragment in a clon-ing host is detected using an activity of coexpressed marker encoded in a cloning vector

High-throughput fluorescence flow sorting is possible

Catabolic genes that are distant from a relevant transcriptional regulator cannot be obtained Genes obtained may be partial

(2 d)(6 d)

(4 d)

(2 d)

Sorting

Sorting 1 mM benzoate No benzoate

Figure 1|Schematic representation of the substrate-induced gene expression screening (SIGEX)scheme with an example of cloning of a benzoate-degradative operon fragment from the metagenome from groundwater microbial flora 19.Benzoate was used as an induction substrate.Time schedule is also presented.dLB medium,diluted Luria–Bertani medium;IPTG,isopropyl b -D -1-thiogalactopyranoside.

NATURE PROTOCOLS |VOL.3NO.7|2008|1203

Choice of an induction substrate

A wide variety of chemicals can be used as induction substrates in the SIGEX scheme to activate gene expression,but these should be

chemicals that can migrate into the cytoplasm of a cloning host.A substrate of a target catabolic-enzyme reaction is commonly used as an induction substrate,although the expression of catabolic genes is not always induced by a substrate of the enzyme.We recommend to use a series of related compounds as induction substrates of SIGEX.It is also important to be aware of enzymes whose expression is induced by a product of a relevant catalytic reaction.

MATERIALS

REAGENTS

.E.coli JM109cells (TaKaRa,cat.no.9022)

.Plasmid vectors and constructs:p18GFP vector for SIGEX and relevant

control constructs 8

.Bacto-tryptone (BD Diagnostics,cat.no.211705).Bacto yeast extract (BD Diagnostics,cat.no.212750).Bacto agar (BD Diagnostics,cat.no.214010).NaCl (Wako,cat.no.191-01665).HCl (Wako,cat.no.080-01066).NaOH (Wako,cat.no.198-13765).KCl (Wako,cat.no.163-03545)

.MgCl 2Á6H 2O (Wako,cat.no.135-00165).Glucose (Wako,cat.no.041-00595).Maltose (Wako,cat.no.130-00615).Na 2HPO 4(Wako,cat.no.194-02875).NaH 2PO 4(Wako,cat.no.198-14505).KH 2PO 4(Wako,cat.no.169-04245)

.Hexadecyltrimethylammonium bromide (CTAB;Wako,cat.no.575-28741)(see REAGENT SETUP)

.SDS (Wako,cat.no.192-13981)(see REAGENT SETUP)

.Ampicillin (Wako,cat.no.015-10382)(see REAGENT SETUP).Proteinase K 20mg ml –1solution (Wako,cat.no.162-22751).Lysozyme,from chicken egg white (Wako,cat.no.122-02673)(see REAGENT SETUP)

.Sodium acetate (Wako,cat.no.195-13915)(see REAGENT SETUP).Glacial acetic acid (Wako,cat.no.516-33981)

.Phenol for molecular biology grade (Wako,cat.no.167-22441).8-Hydroxyquinoline (Wako,cat.no.087-01691).2-Mercaptoethanol (Wako,cat.no.131-14572).Chloroform (Wako,cat.no.038-02606).Isoamyl alcohol (Wako,cat.no.130-15245)

.Paraformaldehyde (Wako,cat.no.160-16061)(see REAGENT SETUP).Ethanol (Wako,cat.no.057-00451)

.FACSClean (BD Biosciences,cat.no.340345)

.Isopropyl b -D -1-thiogalactopyranoside (IPTG;Wako,cat.no.099-02534)(see REAGENT SETUP)

.DNA ligation kit LONG (TaKaRa,cat.no.6024)

.Alkaline phosphatase solution,from shrimp (Wako,cat.no.544-02291).Bam HI (TaKaRa,cat.no.1010A).Eco RI (TaKaRa,cat.no.1040A).Pst I (TaKaRa,cat.no.1073A).Sau 3AI (TaKaRa,cat.no.1082A)

.Luria–Bertani (LB)medium (see REAGENT SETUP).Diluted LB medium (see REAGENT SETUP).LB agar plates (see REAGENT SETUP).SOC medium (see REAGENT SETUP).PBS (see REAGENT SETUP)

.Tris–HCl (see REAGENT SETUP).EDTA (see REAGENT SETUP)

.DNA extraction buffer (see REAGENT SETUP).Tris–EDTA (TE)buffer (see REAGENT SETUP).TE-saturated phenol (see REAGENT SETUP)

.Phenol–chloroform–isoamyl alcohol (PCI)(see REAGENT SETUP)EQUIPMENT

.RECOCHIP (TaKaRa,cat.no.9039)

.Micro slide glass (76Â26mm;MATSUNAMI).Micro cover glass (22Â22mm;MATSUNAMI)

.Calibration beads for flow cytometers (Fluoresbrite calibration grade 6.0micron YG microspheres;Polysciences,cat.no.18862).Petri dishes (94Â16;Greiner)

.Sterile filter size 0.22m m (Millipore)

.Incubator/shaker programmable to different temperatures 16–371C

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h TABLE 2|Fluorescence proteins used in fluorescence-activated cell sorting.Protein

Filters

Excitation laser (nm)Dichromatic mirror (nm)Band pass filter (nm)

GFP (EGFP)

Argon (488)Short pass (560)530/30Red fluorescent protein (DsRed)Argon (488)Short pass (560)575/26Yellow fluorescent protein (YFP)

Argon (488)

Short pass (560)

530/30

: Catabolic enzyme Partial benzoate

T ATG GCT

START gfp

: T ranscriptional regulator : T ransporter protein

b

Figure 2|Schematic drawing of the substrate-induced gene expression

screening (SIGEX)vector (p18GFP)and positive clone which was obtained by a benzoate-induction scheme (pbzo26).(a )p18GFP.The gfp gene expression is under the control of the lac promoter.The DNA sequence of the cloning site (the Bam HI site)and downstream region [including three stop codons,ribosome-binding site (RBS)and the start codon of the gfp gene]are also presented.(b )pbzo26.This clone was obtained by SIGEX with benzoate as an induction substrate 8.Putative genes for catabolic enzymes and those for transcriptional regulators are indicated with different colors (see the bottom of the figure).The clone contains open-reading frames (ORFs)homologous to genes in catechol-degradative and benzoate-degradative operons and benR encoding a transcriptional regulator 18.

1204|VOL.3NO.7|2008|NATURE PROTOCOLS

.DNA electrophoresis equipment .Fluorescence microscope (Olympus)

.Fluorescence spectrophotometer (SpectraMAX Gemini XS;Molecular Devices)

.Electroporator (MicroPulser;Bio-Rad)

.Flow cytometer (FACSVantage SE;BD Biosciences)

REAGENT SETUP

LB medium,1l 10g Bacto-tryptone,5g bacto yeast extract and 5g NaCl.Weigh the reagents and dissolve in 950ml distilled water.Adjust to pH 7.3by adding either HCl or NaOH solution if necessary.Make up volume to 1l with the distilled water.Sterilize by autoclaving for 15min at 1211C.LB can be stored at 41C for 1month.

SOC medium,1l 20g Bacto-tryptone,5g bacto yeast extract,0.58g NaCl and 0.19g KCl.Weigh the reagents and dissolve in 950ml distilled water.Adjust to pH 7.0by adding either HCl or NaOH solution if necessary.Make up volume to 970ml with distilled water.After the medium is autoclaved for 15min at 1211C,it is cooled to below 601C and supplemented with 10ml sterile solution (by autoclaving)of 1M MgCl 2and 20ml sterile solution (by passing through a 0.22-m m filter)of 1M glucose.SOC can be stored at 41C for 1month.

dLB medium,1l 1g Bacto-tryptone,0.5g bacto yeast extract,1g NaCl and 2g maltose.Weigh the reagents and dissolve them in 950ml distilled water.Adjust to pH 7.3with either HCl or NaOH.Make up volume to 990ml with distilled water.After the medium is autoclaved for 15min at 1211C,it is cooled down to 601C and supplemented with 10ml sterile solution (by autoclaving)of 1M MgSO 4.dLB can be stored at 41C for 1month.

Selective media LB and dLB medium containing the ampicillin at a suitable final concentration constitutes the selective media (e.g.,100m g ml –1).Ampicillin stock solutions should be gently thawed immediately before use.Selective media can be stored at 41C for 1month.

LB agar plates Add 15g bacto agar to 1-l LB medium.After autoclaving,it is cooled down to 601C and supplemented with 100m g ml –1ampicillin.Dispense it into Petri dishes.These plates can be stored at 41C for 1month.

Ampicillin solution Dissolve ampicillin in distilled water at 100mg ml –1,pass the solution through a 0.22-m m pore filter and store it at –201C for 1year.IPTG IPTG used for induction of protein expression is dissolved in distilled water at 1mmol ml –1and sterilized by passing through a 0.22-m m pore filter.It can be stored at –201C for 1year.

PBS Dissolve 8g NaCl,0.2g KCl,1.1g Na 2HPO 4and 0.2g KH 2PO 4in distilled water,adjust the volume to 1l and adjust pH to 7.4by adding HCl solution.Sterilize it by autoclaving for 15min at 1211C,pass it through a 0.22-m m pore filter and store it at room temperature (201C)for 6months.

1.53PBS Dissolve 12g NaCl,0.3g KCl,1.65g Na 2HPO 4and 0.3g KH 2PO 4in distilled water,adjust the volume to 1l and adjust pH to 7.4by adding HCl solution.Sterilize it by autoclaving for 15min at 1211C,pass it through a 0.22-m m pore filter and store it at room temperature for 6months.

1M Tris–HCl Dissolve 121.1g of Tris in 800ml of distilled water.Adjust the pH to the desired value by adding HCl solution and adjust the volume to 1l.Sterilize it by autoclaving for 15min at 1211C and store it at room temperature for 6months.

0.5M EDTA Add 186.1g of disodium EDTA Á2H 2O to 800ml distilled

water and adjust the pH to 8.0with NaOH pellets.The disodium salt of EDTA will not dissolve until the pH of the solution is adjusted to 8.0by adding NaOH.Adjust the volume to 1l,sterilize it by autoclaving for 15min at 1211C and store it at room temperature for 6months.

10%CTAB Dissolve 100g of CTAB in 900ml of distilled water and heat to 601C to assist dissolution.Adjust the volume to 1l and store it at room temperature for 6months.20%SDS Dissolve 200g of SDS in 900ml distilled water and heat to 601C to assist dissolution.Adjust the volume to 1l and store it at room temperature for 6months.

DNA extraction buffer Dissolve 13.5g of Na 2HPO 4,0.g of NaH 2PO 4and 87.7g of NaCl in 500ml of distilled water and adjust the volume to 600ml.Sterilize it by autoclaving for 15min at 1211C.After autoclaving,cool to room temperature and add 100ml of 1M Tris–HCl (pH 8.0),200ml of 0.5M EDTA and 100ml of 10%CTAB.This solution can be stored at room temperature for 1week.

Lysozyme solution Dissolve 0.2g of lysozyme in 10ml of 0.25M Tris–HCl (pH 8.0)immediately before use.

3M Sodium acetate (pH 5.2)Dissolve 408.3g of sodium acetate 3H 2O in 800ml of distilled water and adjust the pH to 5.2with glacial acetic acid.Adjust the volume to 1l,sterilize it by autoclaving for 15min at 1211C and store it at room temperature for 6months.

TE-saturated phenol Melt phenol for molecular biology grade at 651C.Add 10ml of 0.5M Tris–HCl (pH 8.0)and 0.03g of 8-hydroxyquinoline to 40g of liquified phenol and stir the mixture for 15min.Stop stirring and,when the two phases have separated,remove the upper (aqueous)phase.Repeat the extractions until the pH of the phenolic phase is 47.8as measured with pH paper.After the pH of the phenolic phase is 47.8and the final aqueous phase have been removed,add 10ml of 0.1M Tris–HCl (pH 8.0),2m l of 2-mercaptoethanol and 80m l of 0.5M EDTA and stir the mixture.This solution can be kept in a light-tight bottle at 41C for up to 1month.!CAUTION Phenol is harmful.W ear gloves to avoid direct contact with the solution.Always use in a chemical fume hood.PCI Mix 50ml of TE-saturated phenol,48ml of chloroform,2ml of isoamyl alcohol and 10ml of 0.1M Tris–HCl (pH 8.0).This solution can be kept in a light-tight bottle at 41C for up to 1month.!CAUTION Phenol,chloroform and isoamyl alcohol are harmful.Wear gloves to avoid direct contact with the solution.Always use in a chemical fume hood.

TE buffer Mix 10ml of 1M Tris–HCl buffer (pH 8.0)and 2ml of 0.5M EDTA solution (pH 8.0)in distilled water,and adjust the volume to 1l.Sterilize it by autoclaving for 15min at 1211C and store it at room temperature for 6months.1%Paraformaldehyde Freshly prepare 100ml of a 1%(wt/vol)paraformal-dehyde solution in PBS.T o dissolve the paraformaldehyde efficiently,the

solution should be heated up to 701C under constant stirring with a magnetic stirrer in a fume hood.After the paraformaldehyde solution is cooled,filter it to remove precipitates.!CAUTION Paraformaldehyde is toxic.Wear gloves to avoid direct contact with the solution.EQUIPMENT SETUP

Flow cytometer The procedure described below is optimized for the

FACSVantage SE flow cytometer and will need to be modified for other flow cytometers.The flow cytometer should be equipped with an argon laser for excitation at 488nm.Laser output power of 0.5W is used for measuring forward scatter (FSC),side scatter (SSC)and green fluorescence intensity (FL1).For measuring GFP fluorescence,a 530/30-band pass filter is used.Fluorescent calibration beads are used for the laser alignment setting.The nozzle tip is 70m m in diameter,and the sheath pressure is adjusted to 11psi.All data are analyzed using CellQuest software (BD Biosciences).The flow cytometer needs to be equipped with a sort unit that can be used for separating GFP-expressing cells from nonfluorescent cells.For cell sorting,flow cytometer water lines should be sterile;the lines were pre-treated with a bleach solution (FACSClean)and 70%ethanol solution,the sheath fluid (1.5ÂPBS)was filter steriled and autoclaved.The drop drive frequency is set at B 26kHz,the ‘Normal-R’mode is chosen and droplet delay is set between 12and 18.The system threshold rate is kept under 3,000s –1.For sorting out GFP-expressing cells,use the 488-nm band of the argon laser for excitation.

PROCEDURE

Construction of a SIGEX vector

1|Construct an appropriate operon-trap gfp -expression vector (SIGEX vector,e.g.,p18GFP)by considering the cloning host and screening system to be used (see Experimental design for further details).Figure 2illustrates the SIGEX vector that we are using (i.e.,p18GFP).

Check of the p18GFP system

2|Transform E.coli JM109competent cells with a SIGEX vector by an electroporation method 11.Plate them out on appropriate selective LB plates.Incubate at 371C overnight.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h NATURE PROTOCOLS |VOL.3NO.7|2008|1205

3|On the next day,pick a single colony from the LB agar plates,inoculate it into 5ml of a liquid LB medium containing ampicillin (LB-amp)and incubate the medium at 301C with vigorous shaking (200cycles min –1on a reciprocal shaker)until the optical density at 600nm (OD 600)of the culture reaches 0.6–0.8.

4|Induce GFP expression by adding IPTG (final concentration of 0.1mM).Incubate the culture at 301C overnight with vigorous shaking (200cycles min –1on a reciprocal shaker).

5|On the next day,centrifuge the cells at 23,180g for 1min and remove the supernatant.

6|Wash the cells with PBS two times,suspend them in 5ml PBS and immediately use them for GFP-expression check (see below).

Alternatively,the cells can be fixed and stored as described in Box 1and analyzed later.

m CRITICAL STEP Washing is essential to remove the medium because it will cause auto-fluorescence in the GFP-expression check.7|Check GFP expression by microscopy (option A),fluorescence spectrophotometry (option B)and/or flow cytometery

(option C).Figure 3shows typical results of microscopy and fluorescence spectrophotometry analyses (cells transformed with p18GFP or pbzo26).Figure 4shows typical results of flow cytometry analyses (cells transformed with p18GFP or pbzo26).(A)Observation of GFP expression by fluorescence microscopy (i)Put a drop of the cell suspension (B 3m l)on a slide.

(ii)Immediately cover the sample with a cover glass.Gently press the cover glass to remove air bubbles and wipe away excess cell suspension with tissue paper.

(iii)Examine the glass slide under a microscope.Set a filter suitable for GFP excitation (460–490nm)and GFP emission

(510–550nm).Microscopy images can be taken using fluorescence microscope equipped with a camera.Figure 3a shows typical results of this observation.

(B)GFP-expression check by fluorescence spectrophotometer (i)Dilute a cell suspension with PBS down to OD 600of r 1.0.

(ii)Set up a fluorescence spectrophotometer:excitation wavelength at 488nm and cutoff filter at 515nm;record the

fluorescence emission spectra in the range of 520–600nm.The photomultiplier tube (PMT)sensitivity parameter was set at ‘Automatic’and ‘Readings 6’.

m CRITICAL STEP The optimal values of parameters such as excitation/emission,gain and recording speed should be determined empirically because these values depend on both the sample to be analyzed and the type of fluorescence spectrophotometer.Excitation and emission wavelengths are also dependent on the fluorescence protein used.(iii)Measure each sample in triplicate.Use PBS for blank reading.Figure 3b shows typical results of this observation.(C)GFP-expression check by flow cytometry

(i)Dilute the cell suspension with PBS to an approximate concentration (ranging from 5Â105to 1Â107cells ml –1).

m CRITICAL STEP Dilution prevents cells from aggregation and prevents the narrow bore tube of the flow cytometer from clogging.

(ii)Check the flow cytometer

parameters by comparing the

fluorescence properties of GFP

nonexpressing E.coli (negative control)and GFP-expressing E.coli (positive-control)cells.First,using the negative-control sample,check whether the event rate is o 3,000events s –1,which

is necessary for high-resolution

analyses.Event rates can be

lowered by decreasing the flow rate and/or by diluting a sample.

(iii)Determine flow cytometer

parameters (FSC,SSC and fluores-cent green light (FL1))by checking the light scattering and p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h BOX 1|FIXING CELLS FOR LATER ANALYSIS

This step ensures a constant fluorescence signal during analysis to stop the bacterial growth.However,we do not fix cells when cell sorting experiments are performed (Steps 47and 53).1.Re-suspend the cells in an equal volume (to the original culture medium)of PBS containing 1%(vol/vol)formaldehyde.2.They can be stored at 41C in the dark for r 5d without a significant loss of GFP fluorescence.

p18GFP + 0.1 mM IPTG Phase-contrast image a b Fluorescence image 20,000F l u o r e s c e n c e (A U )

16,00012,000

8,000

4,0000

500

520540560580+ 1,000 µM benzoate

+ 100 µM benzoate + 10 µM benzoate

No benzoate 600Wavelength (nm)pbzo26

no benzoate pbzo26+ 1 mM benzoate Figure 3|GFP-expression analysis.(a )Phase-contrast microscopy and fluorescence microscopy images of

GFP-expressing and nonexpressing cells.Induction conditions are given in the left side.(b )Fluorescence-emission spectra of pbzo26-harboring cells.A correlation between fluorescence-signal intensity and substrate concentration (0–1mM benzoate)is shown.

1206|VOL.3NO.7|2008|NATURE PROTOCOLS

fluorescence properties of a negative-control sample.Initial instrument settings (threshold parameter,PMT voltage and detector sensitivity settings)should be as follows:Threshold:SSC,value 50

FSC:logarithmic amplification

SSC:280V,logarithmic amplification FL1:600V,logarithmic amplification

m CRITICAL STEP Set FSC,SSC and FL1detector sensitivity to logarithmic amplification settings so that wide ranges of bacterial sizes and fluorescence intensities can be seen,making the gate setting easier.

(iv)Optimize the instrument settings.First,set the SSC threshold,and adjust PMT voltages and threshold levels.Cell count

versus log FSC and cell counts versus log SSC histograms should be checked to make sure that the edges of the bell-shaped peak are not cut off on the histogram.Sometimes,a flat peak is observed in the cell count versus log FSC histograms.This may be caused by the sample analyzed;E.coli (and other bacteria)may change cell size,cell shape and aggregation trend,which can be checked by microscopy.

(v)Generate a dot plot of the SSC versus FSC data.Set a gate around the bacterial population R1(Fig.4a ).Statistical information for R1is provided in the CellQuest program.?TROUBLESHOOTING

(vi)For GFP fluorescence-intensity measurements,the FL1PMT voltage has to be adjusted using nonfluorescent E.coli cells

(Fig.4a ).Show a histogram (log FL1fluorescence versus cell count)to analyze fluorescence levels of nonfluorescent cells.Set the PMT voltage,so that the mean fluorescence value of the peak is below 3(the whole peak area should stay below 10).Acquire a total of 20,000events for R1.Flow cytometer settings used for detecting bacterial cells in this study are as follows:

Threshold (SSC):value 180SSC:310V FL1:600V

(vii)After the measurements,detach the sample tube from the instrument and wash the sample lines by flowing the sheath

fluid for 15s to remove residual bacterial cells.

(viii)Next,analyze positive-control GFP-expressing E.coli cells (Fig.4b ).Use the same settings as established by negative control

in Step 7C(vi).Acquire a total of 20,000events for R1.Statistical information for R1is provided in the CellQuest program.(ix)Compare the histograms of the GFP-expressing and nonexpressing cells (Fig.4c ),and confirm the adequacy of the flow

cytometer analysis.Clear separation of these histograms indicates that the flow cytometer analysis is successful.At this point,a histogram marker can be placed around the positive control to designate positive events and differentiate them from negative events using the CellQuest software.Statistical information is provided by this program,for example,mean fluorescence intensity.

(x)Finally,analyze the fluorescence of test samples using the same settings as established by negative control in Step 7C(vi).The mean fluorescence-intensity value of test samples is correlated with the level of gene-induction activity encoded in the cloned fragment.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t

h c

100

101

102103

104

FL1

250No benzoate 1 mM benzoate

200C o u n t s

150100500250b

a

200C o u n t s

C o u n t s

S S C

S S C

15010050FSC

FSC

SSC

FL1

FL1

SSC

0250200

150100500250

200C o u n t s

150100500

250

200C o u n t s

150100500

250200C o u n t s

150100500250200C o u n t s

150100500

FSC

R1

R1

100

101102103104

FSC

100

10

101102103

10410

101102103104

101102103104

100101

102103104100101

102103104100101

102103104100

101102103104

100

101102103104

100101102103104

Figure 4|Gating in the flow cytometer analysis and sorting of bacterial populations.Cells were analyzed in forward scatter (FSC)and side scatter (SSC)

histograms (upper two panels in a and b ),and also in FSC versus SSC dot plots (lower left panels in a and b ).To select appropriate sizes of particles (i.e.,those equivalent to bacterial cells),an appropriate gate (R1)was set,and fluorescence intensities (FL1)of particles in the gate were analyzed (lower right panels in a and b ).(a )GFP nonexpressing cells.The mean value of FL1intensity was 2.2.(b )GFP-expressing cells.The mean value of FL1intensity was 1,004.(c )Check of specific GFP induction in a positive clone (a clone having pbzo26).The mean FL1intensity under a noninduced condition was 1.7,while it was 450after induction with benzoate.

NATURE PROTOCOLS |VOL.3NO.7|2008|1207

Preparation of metagenome DNA

8|Take 5g of soil sample in a 50-ml tube.Add 13.5ml DNA-extraction buffer and 0.1ml proteinase K solution and then suspend it by vortexing.

9|Incubate the suspension at 371C for 30min by horizontal shaking at 200cycles min –1on a reciprocal shaker.

10|Add 1.5ml of 20%SDS and 750m l of lysozyme solution.Incubate the suspension at 651C for 1.5h by horizontal rocking at 60cycles min –1on a reciprocal shaker.

11|Stop the rocking and incubate at 651C for 1.5h.

12|Centrifuge the suspension at 6,000g for 20min.Transfer the supernatant into a 50-ml tube.

m CRITICAL STEP Use a wide-bore pipette tip (0.3-cm diameter orifice)to transfer the supernatant into a fresh tube.

13|Add an equal volume of TE-saturated phenol to the supernatant and rock the organic and aqueous phases for 30min by horizontal rocking at 60cycles min –1on a reciprocal shaker.

14|Centrifuge the suspension at 6,000g for 20min.Transfer the aqueous phase into a 50-ml tube.

m CRITICAL STEP Use a wide-bore pipette tip (0.3-cm diameter orifice)to transfer the aqueous phase into a fresh tube.15|Add an equal volume of PCI and rock the organic and aqueous phases for 30min by horizontal rocking at 60cycles min –1on a reciprocal shaker.

16|Centrifuge the suspension at 6,000g for 20min.Transfer the aqueous phase into a fresh tube.

m CRITICAL STEP Use a wide-bore pipette tip (0.3-cm diameter orifice)to transfer the aqueous phase into a fresh tube.17|Add 0.1volume of 3M sodium acetate (pH 5.2)and 2volumes of ethanol.Mix the tube gently until the ethanol solution is thoroughly mixed.The DNA immediately forms a white precipitate.

18|Remove the precipitate from the ethanol solution with a Pasteur pipette whose end has been sealed.

19|Wash the DNA precipitate two times with 5ml of 70%ethanol and collect the DNA by centrifugation at 6,000g for 5min.Store the pellet of DNA in an open tube until the ethanol has evaporated.

m CRITICAL STEP Do not allow the pellet of DNA to dry completely because completely dried DNA is very difficult to dissolve.20|Add 1ml of TE buffer and rock the solution at 41C for 12–24h by horizontal rocking at 60cycles min –1on a reciprocal shaker.

m CRITICAL STEP Metagenome DNA samples should be stored for further analysis,which may include post-SIGEX analyses.For instance,when a gene obtained by SIGEX is partial,a flanking genome fragment needs to be obtained (e.g.,using inverse affinity nested-PCR 12).

’PAUSE POINT DNA solution can be stored at 41C for 1year.

21|Analyze the quality of the preparation of metagenome DNA by electrophoresis through a 0.5%agarose gel 13,14.Good

quality metagenome DNA should consist of DNA fragments larger than 15kb.Large DNA fragments (415kb)can be used in the next steps to construct a metagenome library.

Construction of a metagenome library

22|Digest the p18GFP vector DNA (5m g)with twofold-excess Bam HI (compared with the amount recommended by the

manufacturer).Incubate the mixture at 301C for 1h.After the digestion,purify the DNA by the standard ethanol-precipitation procedure 15and dissolve it in TE buffer.

’PAUSE POINT DNA can be stored at –201C for 1year.

23|Take 1m g of the linearized vector DNA,add the dephosphorylation buffer (as recommended by the manufacturer)and an appropriate amount (as recommended by the manufacturer)of shrimp alkaline phosphatase,and incubate the mixture at 371C for 1h,followed by incubation of the reaction mixture for 15min at 651C to inactivate the enzyme.24|Add an equal volume of TE-saturated phenol and mix the organic and aqueous phases by vortexing.25|Centrifuge the emulsion at 41C,23,180g for 5min.Transfer the aqueous phase into a fresh tube.26|Add an equal volume of PCI and mix the organic and aqueous phases by vortexing.

27|Centrifuge the emulsion at 41C,23,180g for 5min.Transfer the aqueous phase into a fresh tube.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h

28|Recover the DNA by the standard ethanol-precipitation procedure 15,and dissolve the linearized and dephosphorized vector DNA in TE buffer.

’PAUSE POINT DNA can be stored at –201C for 1year.29|Partially digest metagenome DNA with the restriction enzyme Sau 3AI (Box 2).

m CRITICAL STEP Conditions for the partial digestion should be optimized sample by sample in a small-scale pilot experiment.30|After the digestion,check DNA by agarose gel (typically 0.5%)electrophoresis 13and fluorescence dye staining 14.A good sample exhibits intense fluorescence in a region of desired molecular sizes (5–10kb in our experiments).?TROUBLESHOOTING

31|After electrophoresis,make an incision using a razor edge immediately in front of the DNA bands (+side)of the electro-phoresis gel (a range of B 5–10kb was recovered in our experi-ments).Insert RECOCHIP (treated with the electrophoresis buffer)in the incision.

32|Put the gel back into the electrophoresis chamber and perform electrophoresis again.A few minutes later,stop the electrophoresis and remove the RECOCHIP from the gel.

33|Put the RECOCHIP into a 2-ml tube and centrifuge it at 4,020g for 5s.

34|Recover the DNA solution (30–50m l in volume),purify the DNA by standard ethanol-precipitation procedure 15and finally dissolve the DNA in TE buffer.

’PAUSE POINT The DNA solution can be stored at –201C for 1year.35|Set up a ligation reaction mixture as follows:

Vector DNA (3.4kb,from Step 28):50ng Insert DNA (5–10kb,from Step 34):25ng 10ÂLONG ligation buffer:5m l H 2O:up to 49m l

36|Incubate the above reaction mixture for 3min at 651C and cool immediately on ice.Next,add 1m l of DNA ligase (from DNA ligation kit LONG)to the reaction mixture and incubate them for 15h at 161C.’PAUSE POINT The reaction mixtures can be stored at –201C for several days.

37|Recover the DNA by the standard ethanol-precipitation procedure 15and dissolve it in TE buffer.

’PAUSE POINT Ligated DNA can be stored at –201C for several days.

38|Transform E.coli JM109competent cells with the ligation mixture.Check the competent cells for transformation efficiency (Box 3).

m CRITICAL STEP We recommend the electroporation method 11to achieve a good transformation efficiency.

m CRITICAL STEP It is necessary to set up a negative-control reaction (the digested SIGEX vector alone)to estimate the number of background transformants caused by self-ligation of the vector.

39|After the transformation,transfer the cells into

1ml of SOC medium and incubate them at 371C for 1h with gentle rotation.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h BOX 2|OPTIMIZATION OF PARTIAL DIGESTION

Conditions for the partial digestion should be optimized sample by sample in a small-scale pilot experiment.

1.Set up pilot reaction mixtures from the same batch of

metagenomic DNA.The mixtures contain 1m g of metagenomic DNA (27m l in volume)and 3m l of the appropriate 10Ârestriction buffer.Incubate the tubes on ice.

2.Add 2units of a restriction enzyme (e.g.,Sau 3AI)to one tube.Then,transfer 30m l of this mixture into a new mixture and completely mix it;this procedure is repeated to make serial dilutions.

3.Incubate the mixtures for 1h at 371C.Then,inactivate the Sau 3AI by heating at 701C for 15min.

4.Analyze the mixtures by electrophoresis in 0.5%agarose gel.Compare sizes of the digested DNA with reference to a DNA size marker.

BOX 3|CHECK OF EFFICIENCIES FOR LIGATION AND TRANSFORMATION

Ligation and transformation efficiencies can be checked as follows:

1.Take an aliquot (B 10m l)from the culture in Step 39,and spread it on LB plates containing ampicillin (100m g ml –1)and isopropyl b -D -1-thiogalactopyranoside (0.5mM).Incubate at 371C overnight.

2.Next day,lay the plate on a UV trasilluminator (365nm).Count non-GFP expressing colonies on the plate.

3.When using competent cells with a transformation efficiency of 1Â108cfu m g –1,the above described ligation and electro-poration procedures yield 41Â104non-GFP expressing colo-nies (insert size of 5–10kb).GFP-expressing cells on the plates contain either the self-ligation vector or vectors ligated with constitutively expressing inserts.

40|Add 5ml of LB containing ampicil-lin (100m g ml –1)and incubate the

culture at 371C overnight with vigorous shaking (200cycles min –1on a recipro-cal shaker).

’PAUSE POINT The overnight culture can be stored at –801C for 1year without significant loss of viability in the medium supplemented with 15%(vol/vol)glycerol.?TROUBLESHOOTING

41|Take an aliquot from (B 10m l)the culture in Step 39and spread it on LB plates containing ampicillin

(100m g ml –1)and IPTG (0.5mM).Incubate at 371C overnight.

42|Next day,pick ten white colonies and inoculate them into 1ml of liquid LB media containing ampicillin (100m g ml À1).Incubate at 371C overnight with vigorous shaking (200cycles min –1on a reciprocal shaker).m CRITICAL STEP Pick white colonies as green colonies contain the self-ligation vector.

43|Extract plasmid DNA from cultures from Step 42using a standard plasmid-extraction procedure 16and analyze the size of inserted DNA fragment by restriction enzyme digestion (we usually use Eco RI or Pst I)and agarose-gel electrophoresis 13,14.A good quality library has random inserts.

Screening of positive clones by flow sorting

44|On the next day,take an aliquot of 50m l of the metagenome library from Step 40and inoculate it into 5ml LB medium containing ampicillin (100m g ml À1).Incubate at 301C with vigorous shaking (200cycles min –1on a reciprocal shaker)until the OD 600reaches 0.6–0.8.

45|Induce GFP expression by adding IPTG (0.5mM).Incubate at 301C overnight with vigorous shaking (200cycles min –1on a reciprocal shaker).

46|On the next day,check the alignment of the flow cytometer and prepare the cell sorter as described in Step 7C.m CRITICAL STEP Use a sheath fluid without antiseptics in order not to kill transformed cells during sorting.

47|Apply cells in the metagenome library (typically containing 10,000independent clones;from Step 45)to the flow

cytometer (Fig.5a ).Set a sorting region for cells with no fluorescence signal (a mean fluorescence value of o 3;R2in Fig.5a ).In this step,fluorescent cells are those that contain either the self-ligation vector or vectors with short inserts.Sort out B 500,000cells.

48|Check the efficiency of the sorting by analyzing a portion of the sorted sample by flow cytometry (Step 7C).In the sorted fraction,a ratio of nonfluorescent cells to total cells should be increased significantly.A typical histogram (cell count versus log FL1)at this step is shown in Figure 5b .?TROUBLESHOOTING

49|Inoculate the sorted cells (B 100,000cells)from Step 47into 5ml of LB liquid media containing ampicillin

(100m g ml À1).Incubate at 371C overnight with vigorous shaking (200cycles min –1on a reciprocal shaker).Check this culture again using the flow cytometer,as described in Steps 7C.If cells with high fluorescence signal (a fluorescence value 480)are abundantly present,repeat Steps 44–49.

’PAUSE POINT An overnight culture can be stored at –801C for 1year without significant loss of viability in the medium containing 15%(vol/vol)glycerol.

50|Take an aliquot (10m l)of the culture from Step 49and inoculate it into 1ml of dLB medium containing ampicillin

(100m g ml À1).Incubate at 301C with vigorous shaking (200cycles min –1on a reciprocal shaker)until the OD 600reaches 0.6–0.8.51|Induce GFP expression by adding an induction substrate (e.g.,benzoate in our previous study 8)at a final concentration of 2mM.Incubate the culture at 301C overnight with vigorous shaking.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h 100

101

102FL1

10

3

104

100

101

102103

104

100

101

102103

104

250a

b

c 200R2

R3

R2

R3

R4

R5

C o u n t s

C o u n t s

C o u n t s 150100FL1

FL1

50025020015010050

0250

20015010050

0Figure 5|Selection of benzoate-positive clones by cell sorting.(a )Histogram of cells in the

metagenome library after induction with isopropyl b -D -1-thiogalactopyranoside (IPTG).Cells in the R2region were identified as non-GFP expressing cells,while those in the R3region were IPTG-induced GFP-expressing cells.(b )Cells in the R2fraction in a were sorted and collected.The sorted sample was subjected to flow cytometry to check the sorting rate.Cells in the R2region were identified as non-GFP expressing cells,while those in the R3region were IPTG-induced GFP-expressing cells.The ratio of

IPTG-induced GFP-expressing cells was decreased.(c )Cells in the R2region in b were induced with 2mM benzoate,and subjected to flow cytometry.Cells in the R5region were sorted out for the subsequent analyses.Cells in the R4region were identified not induced non-GFP-expressing cells.

52|On the next day,check the alignment of the flow cytometer and prepare the cell sorter as described in Step 7C.m CRITICAL STEP Use a sheath fluid without antiseptics.

53|Apply the culture from Step 51(it typically contains 100,000independent clones)to the flow cytometer (Fig.5c ).

In the resulting histogram (cell count versus log FL1),set a sorting region to collect cells expressing GFP (typically the mean fluorescence value 480;R5in Fig.5c ).In our case,this gating sorted out B 500cells.Cells without GFP fluorescence are also sorted out (the R4region in Fig.5c ).

m CRITICAL STEP The R4cells (nonfluorescence cells)are used in the next induction assay with another substrate (Steps 50–53can be repeated with different substrates).

54|Grow the sorted cells on LB plates containing ampicillin (100m g ml À1)and IPTG (0.5mM),and incubate the plates at 371C overnight.Green colonies are contaminated cells that have the self-ligation vector.

55|On the next day,pick all white colonies and inoculate them into 1ml of liquid dLB media containing ampicillin (100m g ml À1).Incubate them at 301C with vigorous shaking (200cycles min –1on a reciprocal shaker)until the OD 600reaches 0.6–0.8.56|Divide the culture from Step 55into two fractions.Add the induction substrate (used in Step 51)to a final concentration of 2mM into one of the two fractions and incubate it at 301C overnight with vigorous shaking (200cycles min –1on a reciprocal shaker).Incubate the other fraction in the same manner but without adding the induction substrate.

57|On the next day,compare fluorescence intensities of the induced and noninduced cultures using the flow cytometer (as described in Step 7C).Check the induction efficiency [efficiency ¼(mean fluorescence intensity of the induced culture)/(mean fluorescence intensity of the noninduced culture)].

58|Extract plasmid DNA from a positive clone using a standard plasmid-extraction procedure 16and analyze the nucleotide sequence of the inserted DNA fragment 17.

m CRITICAL STEP Restriction fragment length polymorphism (RFLP)analysis is useful to check identical clones. TIMING

Steps 2–7,Check of the p18GFP system:3d Steps 8–21,preparation of a metagenome:2d

Steps 22–43,construction of the metagenome library:6d Steps 44–58,screening of positive clones by flow sorting:9d

?TROUBLESHOOTING Step 7C(v)

Problem:Difficult to define the R1region.

Solution:Owing to the small size of bacteria,some problems could occur in the establishment of a recognizable population in the dot plot SSC versus FSC.Analyze a blank sample (cell-suspending buffer)to check the background noise level.Filtering and autoclaving of cell-suspending buffer and sheath fluid used to suspend the bacteria minimize background particle-scatter signal noise.Alternatively,pass a cell suspension through a nylon-mesh filter to remove cell aggregates.

Step 30

Problem:Sample DNA is not digested by a restriction enzyme.

Solution:DNA extracted from an environmental sample may not be pure enough for the restriction enzyme digestion (contamination with proteins,mucoids and humic acids).DNA should be further purified (e.g.,use of a commercial column-chromatography kit).

Step 40

Problem:Many false positives (those having the self-ligation vectors).

Solution:Preparation of the linearized SIGEX vector (Steps 22–28)is not successful.Analyze the preparation by gel electrophor-esis to check whether the digestion is complete.If the digestion is not complete,digest it again with a larger amount of the restriction enzyme.Alternatively,in case the dephosphorylation reaction is not complete,treat it again with a larger amount of alkaline phosphatase.

Step 48

Problem:Low sorting efficiency.

Solution:Lower the event rate down to below 3,000events per s –1.It improves accuracy of cell detection by the sorting machine,but it results in a longer time needed for sorting.

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h

ANTICIPATED RESULTS

In a typical procedure of the SIGEX screening,IPTG-induced cells (those having the self-ligation vector or vectors ligated with constitutive expressing inserts)are removed by the first-flow sorting;cells in the R3region in Figure 5a are removed,while those in R2are sorted out.This separation is checked by flow cytometry of a small portion of the R2fraction (Fig.5b );in our experiment,R2comprised 490%of the total cell count.R2is next used for a substrate-induction assay in Steps 50–53

(Fig.5c ),and cells with FL1signals 480are sorted out.If cells with strong FL1signals are not sufficiently present,the gate value may be lowered (down to 20)to select weakly GFP-expressing cells,although this method may increase false positives.These cells (in the R5region)are isolated by colony formation on the agar plates,and substrate-dependent GFP expression is confirmed by flow cytometry.In our previous experiment 8,20%of the sorted cells formed colonies on LB agar plates,and the colony check identified that 40%of the colonies are positive in benzoate-dependent gene expression.Metagenome fragments in the 58positive colonies are divided into 33different RFLP patterns.Cells in the R4region (nonfluorescent cells)are also sorted out,which are used in subsequent screening with different substrates.In our previous study 8,benzoate-negative cells (the R4cells)were subjected to a naphthalene-induction assay.Alternatively,a metagenome library is divided into several aliquots that are used in gene-induction assays with different substrates.

One of colonies isolated using benzoate as the induction substrate induced a strong fluorescence in the presence of benzoate but not in the absence of benzoate (Fig.4c ).Sequence analysis revealed that this clone contained a metagenome fragment with a putative open-reading frame (ORF)structure shown in Figure 2(pbzo26).An ORF highly homologous to benR

(encoding a positive regulator responding to benzoate 18)was present in front of gfp together with ORFs for putative aromatic hydrocarbon-transforming enzymes.We have developed an improved inverse PCR scheme applicable to cloning of flanking regions of a metagenome-derived fragment 12;this method can be used to clone a DNA fragment flanking the BenR -like ORF from a metagenome.

ACKNOWLEDGMENTS This work was supported by the Japan Society for Promotion of Science.We thank Robert A.Kanaly for critical reading of this manuscript.AUTHOR CONTRIBUTIONS T.U.prepared the experimental data and wrote the paper,and K.W.wrote the paper.

Published online at http://www.natureprotocols.com

Reprints and permissions information is available online at http://npg.nature.com/reprintsandpermissions

1.Amann,R.I.,Ludwig,W.&Schleifer,K.H.Phylogenetic identification and in situ

detection of individual microbial cells without cultivation.Microbiol.Rev.59,143–169(1995).

2.Cowan,D.A.Microbial genomes—the untapped resource.Trends Biotechnol.18,

14–16(2000).

3.Handelsman,J.,Rondon,M.R.,Brady,S.F.,Clardy,J.&Goodman,R.M.Molecular

biological access to the chemistry of unknown soil microbes:a new frontier for natural products.Chem.Biol.5,R245–R249(1998).

4.Lorenz,P.,Liebeton,K.,Niehaus,F.&Eck,J.Screening for novel enzymes

for biocatalytic processes:accessing the metagenome as a resource of novel functional sequence space.Curr.Opin.Biotechnol.13,572–577(2002).

5.Knietsch,A.,Waschkowitz,T.,Bowien,S.,Henne,A.&Daniel,R.Construction

and screening of metagenomic libraries derived from enrichment cultures:

generation of a gene bank for genes conferring alcohol oxidoreductase activity on Escherichia coli .Appl.Environ.Microbiol.69,1408–1416(2003).6.Henne,A.,Daniel,R.,Schmitz,R.A.&Gottschalk,G.Construction of

environmental DNA libraries in Escherichia coli and screening for the presence of genes conferring utilization of 4-hydroxybutyrate.Appl.Environ.Microbiol.65,3901–3907(1999).

7.Okuta,A.,Ohnishi,K.&Harayama,S.PCR isolation of catechol 2,3-dioxygenase

gene fragments from environmental samples and their assembly into functional genes.Gene 212,221–228(1998).

8.Uchiyama,T.,Abe,T.,Ikemura,T.&Watanabe,K.Substrate-induced gene-expression screening of environmental metagenome libraries for isolation of catabolic genes.Nat.Biotechnol.23,88–93(2005).

9.de Lorenzo,V.Problems with metagenomic screening.Nat.Biotechnol.23,1045

(2005).

10.Uchiyama,T.&Watanabe,K.The SIGEX scheme:high throughput screening of

environmental metagenomes for the isolation of novel catabolic genes.Biotechnol.Genet.Eng.Rev.24,107–116(2007).

11.Sambrook,J.&Russell,D.W.Transformation of E.coli by electroporation.in

Molecular Cloning:A Laboratory Manual 3rd edn.Vol.1(eds.Sambrook,J.

&Russell,D.W.)1.119–1.122(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

12.Uchiyama,T.&Watanabe,K.Improved inverse PCR scheme for metagenome

walking.Biotechniques 41,183–188(2006).

13.Sambrook,J.&Russell,D.W.Agarose gel electrophoresis.in Molecular Cloning:A

Laboratory Manual 3rd edn.Vol.3(eds.Sambrook,J.&Russell,D.W.)

5.4–5.13(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

14.Sambrook,J.&Russell,D.W.Detection of DNA in agarose gels.in Molecular

Cloning:A Laboratory Manual 3rd edn.Vol.3(eds.Sambrook,J.&Russell,D.W.)5.14–5.17(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

15.Sambrook,J.&Russell,D.W.Commonly used techniques in molecular cloning.in

Molecular Cloning:A Laboratory Manual 3rd edn.Vol.3(eds.Sambrook,J.&Russell,D.W.)A8.12–A8.16(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

16.Sambrook,J.&Russell,D.W.Preparation of plasmid DNA by alkaline lysis with

SDS.in Molecular Cloning:A Laboratory Manual 3rd edn.Vol.1(eds.Sambrook,J.&Russell,D.W.)1.31–1.42(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

17.Sambrook,J.&Russell,D.W.DNA sequencing.in Molecular Cloning:

A Laboratory Manual 3rd edn.Vol.2(eds.Sambrook,J.&Russell,D.W.)12.1–12.120(Cold Spring Harbor Laboratory Press,Cold Spring Harbor,New York,2001).

18.Cowles,C.E.,Nichols,N.N.&Harwood,C.S.BenR,a XylS homologue,regulates

three different pathways of aromatic acid degradation in Pseudomonas putida .J.Bacteriol.182,6339–6346(2000).

19.Watanabe,K.,Watanabe,K.,Kodama,Y.,Syutsubo,K.&Harayama,S.Molecular

characterization of bacterial populations in petroleum-contaminated

groundwater discharged from underground crude oil storage cavities.Appl.Environ.Microbiol.66,4803–4809(2000).

p u o r G g n i h s i l b u P e r u t a N 8002©n a t u r e p r o t o c o l s

/m o c .e r u t a n .w w w //:p t t h