(2007) xxx–xxx

+ MODEL

MIMET-02700; No of Pages 10

www.elsevier.com/locate/jmicmeth

ARTICLE IN PRESS

Journal of Microbiological Methods xx

Improved reliability of Pseudomonas aeruginosa PCR detection by the useof the species-specific ecfX gene target

R. Lavenir, D. Jocktane, F. Laurent, S. Nazaret, B. Cournoyer ⁎

Research group on ≪Bacterial Opportunistic Pathogens and Environment≫ Université de Lyon, Lyon, F-69003, FranceUniversité Lyon 1, CNRS, UMR5557, Ecologie Microbienne, Villeurbanne, F-69622, France

Ecole Nationale Vétérinaire de Lyon, Marcy L'étoile, F-69280, France

Received 22 December 2006; received in revised form 14 March 2007; accepted 20 March 2007

Abstract

Reliability of the most widely used PCR screenings for the human opportunistic pathogen Pseudomonas aeruginosa was evaluated. Specificityanalyses showed the gyrB, toxA, and 16S–23S rDNA internal transcribed spacer (ITS) but not the 16S rDNA, oprI, oprL, and fliC PCRscreenings to discriminate P. aeruginosa cells from a collection of fifteen Pseudomonas species. Sensitivity analyses showed all these PCR exceptthe toxA one to be reliable for 100% of the P. aeruginosa strains tested in this study. Specificity of the ITS and gyrB PCR screenings were furtherinvestigated on 9 soils and 29 freshwater DNA extracts of different origins, and on DNA extracted from 3 horse manures. The ITS PCR showedthe highest efficacy on water and soil DNA extracts but only the gyrB one detected P. aeruginosa DNA in horse manure. DNA sequence analysesof ITS and gyrB PCR products revealed uncertainties and false positive results in these P. aeruginosa identification schemes. A novel PCRscreening, targeting the ecfX gene, was thus developed. ecfX encodes an ECF (extracytoplasmic function) sigma factor which is restricted to P.aeruginosa, and might play a role in haem-uptake and virulence. Specificity and sensitivity analyses showed the ecfX PCR screening to be highlyreliable, giving PCR products of the expected size for all P. aeruginosa strains tested and not amplifying DNA from any of the other Pseudo-monas species tested. The ecfX PCR screening was validated on environmental DNA extracts. DNA sequence analyses of the ecfX PCR productsconfirmed their identity and allocation to P. aeruginosa. These investigations suggest a preferential colonization of water rather than soilenvironments by P. aeruginosa. Detection limits of P. aeruginosa in environmental samples were improved by the ecfX PCR screening.© 2007 Elsevier B.V. All rights reserved.

Keywords: Cystic fibrosis; Nosocomial infections; PA1300; Soil; Sputum; Water

1. Introduction

Pseudomonas aeruginosa is a major human opportunisticpathogen. It is responsible of infections in patients with cysticfibrosis (CF) (Saiman and Siegel, 2004), and in burned,mechanically ventilated or immuno-compromised individuals(McManus et al., 1985; Richard et al., 1994). It can cause seriousinfections including septicemia, pneumonia, endocarditis, otitisand keratitis. P. aeruginosa is also an environmental bacteriumgrowing in a wide range of niches including those generated by

⁎ Corresponding author. Mailing address: UMR CNRS 5557 EcologieMicrobienne, Mendel Bldg., 5th floor, Université Lyon 1, 69 622 VilleurbanneCedex, France. Tel.: +33 4 72 43 14 95; fax: +33 4 72 43 12 23.

E-mail address: cournoye@biomserv.univ-lyon1.fr (B. Cournoyer).

0167-7012/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.mimet.2007.03.008

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

human activities. It can be isolated from rivers (Pellett et al.,1983), sea water (Kimata et al., 2004), bottled and tap waters(Pellett et al., 1983; Hunter, 1993; Romling et al., 1994; Ganguliand Tripathi, 1999; Aoi et al., 2000), andwastewaters (Filali et al.,2000). It is also found in soils (Cavalca et al., 2000) and cancolonize plants (Green et al., 1974; Morales et al., 1996) oranimals (Marlier et al., 2000; Martin Barrasa et al., 2000; Lashevand Lasarova, 2001). Its ability at colonizing all sorts ofenvironment tend to favour its contact with weakened individualsleading to several cases per year of community-acquiredinfections. However, these properties can also allow them tocolonize several hospital niches, including most moist environ-ments like rooms dedicated to patients' hygienic care. They cancolonize taps, drains, water pipes and several other devices. Allthese niches can lead to nosocomial infections. Several studies

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

Table 1Pseudomonas strains used in this study

Pseudomonas aeruginosa OthersPseudomonas

From soils7NSK2 a, ATCC21776, ATCC 31479, CFBP 5036,CFBP 5037, CIP 104590, DSMZ 6195, LMD 50.34,LMD 68.7.

P. agaraciCFBP 2063P. alcaligenesCFBP 2437T

P. aspleniiCFBP3279P. balearicaCIP 105297T

From watersLMD 25.27, LMG 9009. P. chlororaphis

CFBP2132T

P. cichoriiCFBP 2101P. citronellolisCIP 104381T

From plantsATCC 14425, ATCC 33818, CFBP 5031, CFBP 5032,CFBP 5033, CFBP 5034, LMG 1272, LMG 2107,LMG 5031, LMG 5032, LMG 5033, LMG 6855,LMG 10643.

P. flavescensCIP 104204T

P. fluorescensCFBP 2102 b

P. fragiCFBP 4556 T

P. pseudoalcaligenesCFBP 2435T

P. putidaCFBP 2066P. tolaasiiCFBP 2068

From patientsATCC 15691, ATCC 27853 c, CIP 72.26 c, PA5 c,PA6 c, PA12 c, PAK6AR2d, PAO1(1)R, e, PAO1(2)R, f,PAO1(3)R, c, UCBPP-PA14 b.

P. stutzeriCFBP 2443 b

P. viridiflavaCFBP 2107T

Other originA14 c, ATCC27014, ATCC33988, CFBP 2466T,CFBP 5035, CIP 58.35 c, CIP 58.39 c, CIP 58.41 c,CIP 58.46 c, CIP 59.20 c, CIP 59.33 c, CIP 59.34, CIP59.35, CIP 59.37, CIP 59.38, CIP 59.39, CIP 59.40,CIP 59.43, CIP 59.44, CIP 59.45, CIP 60.92 c, CIP62.3 c, CIP 100720T, LMG 15153.

ATCC: American Type Culture Collection, Rockville, Maryland, USA. CFBP:Collection Française de Bactéries Phytopathogènes, INRA, Angers, France.CIP: Collection de l'Institut Pasteur, Paris, France. DSMZ: Deutsche Sammlungvon Mikroorganismen und Zellkulturen Gmbh, Braunschwieg, Deutschland.LMG/BCCM: Laboratorium Microbiology Gent Culture Collection/BelgianCoordinated Collections of Microorganisms, Gent, Belgium. NCBB: TheNetherland Culture Collection of Bacteria (in the past LMD and Phabagen),Utrecht, Netherlands.TType strain. RGenomic and genetic reference strain.a Collection of M. Höfte, University of Gent, Gent, Belgium.b Collection of L.G. Rahme, Harvard medical school, Boston, USA.c Collection of J.M. Meyer, University of Strasbourg 1-Louis Pasteur, Strasbourg,

France.d Collection of J. Wang, University of Arkansas for medical sciences, Little Rock,

USA.e Collection of R. Levesques, University of Laval, Laval, Canada.f Collection of A. Lazdunski, University of Luminy, Marseille, France.

2 R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

demonstrated the close genetic proximity of clinical andenvironmental isolates, and strongly suggested an environmentalorigin of the isolates encountered in several infections (Romlinget al., 2005). Hospital and environmental populations of P.aeruginosa represent an important public health concern.

One major issue remaining poorly documented about P.aeruginosa is its ecology outside the hospital. Reports showingthe presence of these bacteria in the environment have beenmade but only few data concerning their growth or the activitiesthey perform in field situations have been compiled. There ispractically no data on the effects of environmental selectiveforces on the emergence of novel genotypes and properties in P.aeruginosa. In order to begin such investigations, one key issueis the determination of P. aeruginosa preferential niches. Toachieve this goal, a fast and highly reliable detection method isrequired. This method should be sensitive enough to detect fewindividuals including those which are viable but not culturableon classical media. It needs to be applicable on environmentalDNA samples. PCRmethods appear to fulfil these requirements.Several PCR screenings have been used to allocate isolatesrecovered on selective media to the P. aeruginosa species. Mostsignificant examples are those targeting toxA, gyrB, or theribosomal operon (Tyler et al., 1995; Kimata et al., 2004;Kurupati et al., 2005). Some of these screenings were shownapplicable directly on human DNA samples like skin DNA (DeVos et al., 1997). These gene targets could also be used for FISH(fluorescent in situ hybridization) detections of P. aeruginosa inCF sputum (Hogardt et al., 2000), and in quantitative real timePCR on blood (Jaffe et al., 2001) and wound biopsy samples(Pirnay et al., 2000). A few studies applied these PCR screeningsin P. aeruginosa detection scheme from aquatic environments orwastewaters (Khan and Cerniglia, 1994; Schwartz et al., 2006).Such screenings were not applied on soil samples.

The great versatility of P. aeruginosa, making it an apparentlygood colonizer of several ecological niches, could be partly due to ahigh number of regulatory genes among its genome, allowing aquick acclimation to changing environments. Some of these genesencode factors of the sigma 70 family which are involved inpromoter recognition and initiation of gene transcription by theRNApolymerase, in a wide variety of growing conditions (Lonettoet al., 1994; Ishihama, 2000). The extracytoplasmic function (ECF)sigma factors are a sub-group of this family, and are characterizedby their ability to co-ordinate transcription in response toextracytoplasmic stimuli (Raivio and Silhavy, 2001). ECF genescan be found in up to 60–65 copies per genome. Some of thesegenes appear to have a limited distribution, being restricted to asingle or a few species. ECF paralogs have no significant DNAsequence identity and their deduced proteins share around 40 to60% identity. Orthologs of a same ECF lineage among a bacterialspecies can show high percentages of sequence divergences. Forexample, hrpL (hypersensitive response and pathogenicity) ECFfactor gene sequences divergences could be used to infer infra-specific phylogenetic groupings among Pseudomonas syringae(Cournoyer et al., 1996). All these features make these genes goodcandidates for the development of species-specific PCR screenings.

In this study, we evaluated the sensitivity of the most frequentlyused P. aeruginosa PCR screenings on a collection of clinical and

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

environmental P. aeruginosa strains, and their specificity on acollection ofPseudomonas species closely related toP. aeruginosa.Their sensitivity and specificity were also evaluated on soil and

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

3R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

water DNA extracts. DNA sequence analyses of the PCR productsobtained from environmental DNA extracts revealed uncertaintiesand false positive results in these identification schemes. This led usto investigate the use of ECF gene sequences for the developmentof a more specific screening. Analysis of the P. aeruginosa PAO1genome revealed a species-specific ECF gene that was then usedfor the design of a novel P. aeruginosa PCR screening. Thisscreening was found highly reliable, making possible assessmentsabout P. aeruginosa prevalence in complex environments such assoils and freshwaters.

2. Materials and methods

2.1. Bacterial strains and growth conditions

Fifty-nine bacterial strains belonging to P. aeruginosa and15 other Pseudomonas species were obtained from the ATCC(American Type Culture Collection, Rockville, Maryland,USA), CFBP (Collection Française de Bactéries Phytopatho-gènes, INRA, Angers, France), DSMZ (Deutsche von Mik-roorganismen und Zellkulturen Gmbh, Braunschwieg,Deutschland), LMG/BCCM (Belgian Coordinated Collectionsof Microorganisms, Gent, Belgium) and NCBB (The Nether-land Collection of Bacteria, Ultrecht, Nederland) bacterialcollections (Table 1). Clinical isolates coming from nosocomial(72% of the collected isolates) or community-acquired (28% ofthe isolates) infections (ears, urine, blood, wound, peritonealliquid and bronchial aspiration) were collected in October 2003by the French “Collège de Bactériologie, de Virologie etd'Hygiène des Hôpitaux”, and completed this collection. Allbacteria were stored at −80 °C in LB broth containing 20%glycerol. Bacteria were grown overnight at 37 °C in Luria-Bertani (LB) broth with shaking at 180 rpm.

2.2. P. aeruginosa isolation procedure from environmental samples

Twenty-three surface freshwater samples were collected inFebruary 2005 in sterile 500 ml bottles, in the Bourg-en-Bresse

Table 2PCR primers, their target and annealing temperature, of the P. aeruginosa PCR scre

Primer Oligonucleotide sequence (5′–3′)(# of nucleotides)

Annealitempera

PA-SS-F GGGGGATCTTCGGACCTCA (19) 58PA-SS-R TCCTTAGAGTGCCCACCCG (19)OPR-1 GCTCTGGCTCTGGCTGCT (18) 58OPR-2 AGGGCACGCTCGTTAGCC (18)Fla1 GCCTGCAGATCGCCAACC (18) 56Fla2 GGCAGCTGGTTGGCCTG (17)PAL1 ATGGAAATGCTGAAATTCGGC (21) 64PAL2 CTTCTTCAGCTCGACGCGACG (21)ETA1 GACAACGCCCTCAGCATCACCAGC (24) 66ETA2 CGCTGGCCCATTCGCTCCAGCGCT (24)GyrPA-398 CCTGACCATCCGTCGCCACAAC (22) 66GyrPA-620 CGCAGCAGGATGCCGACGCC (20)PA1 TCCAAACAATCGTCGAAAGC (20) 58PA2 CCGAAAATTCGCGCTTGAAC (20)ECF1 ATGGATGAGCGCTTCCGTG (19) 58ECF2 TCATCCTTCGCCTCCCTG (18)

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

sub-urban area (Ain, France): 1 brook, 9 rivers, 6 ponds, 1 watertank, 3 road ditches and 3 wastewater treatment lagoons. Sixwater samples were collected in February 2007 from the Yzeronriver (Oullins, France) of the “Field Observatory for UrbanWater Management” (http://www.graie.org/othu/). Nine soilsamples were collected : one under a permanent pasture field(Montrond, Rhône-Alpes, France), five under cultivated crops(La Côte Saint André and St Victor in Rhône-Alpes, Auxonneand Châteaurenard in Burgundy, France, and Ô Mon, Vietnam),one from an industrial site contaminated with hydrocarbons(Nancy, Lorraine, France), one from a tropical forest (St-Elie,French Guyana), and one from the savana (Lamto, Ivory Coast).The samples were collected from the upper layer (0–20 cm) andwere sieved (2 mm mesh). Three samples of horse manuregoing from the freshly delivered horse-dung to a 3 monthscomposted horse manure were also collected.

P. aeruginosa isolations from environmental samples wereperformed using the Pseudomonas Agar Base medium (Oxoïd,Dardilly, France) supplemented with cetrimide (200 mg l−1)and nalidixic acid (15 mg l−1). One hundred ml of watersamples were filtered using 0.22 μm filters which were thentransferred onto solid medium. Bacterial cells from soils andhorse manure were extracted by blending samples with 50 ml ofa saline solution (NaCl 0.8%) for 90 s in a Waring Blender(Eberbach Corporation). Homogenized soil and horse manuresuspensions were serially diluted in sterile saline solution, and100 μl of the 10−1 to 10−5 dilutions were spread on the agarplates. Enrichment assays were performed by transferring 2 g ofsoil into 20 ml of a salt solution supplemented with acetamide,as described previously (Green et al., 1974). Inoculatedenrichment broths were incubated for 3 days at 28 °C withshaking at 180 rpm. Serial dilutions were performed and plated.In all cases, three plates were inoculated per dilution. Plateswere incubated at 37 °C for 72 h. From 5 to 10 colonies werechosen at ramdom from the selective plates, and confirmed as P.aeruginosa with the elastase and oxydase tests, and the toxAPCR screening (shown below to be highly specific but toneglect 5% of P. aeruginosa diversity, see Section 2.5).

enings used in this study

ngture (°C)

Target and size of theexpected product (pb)

Reference

16S rDNA, 956 Spilker et al. (2004)

oprI, 197 Qin et al., 2003

fliC, 1000 or 1300 Spangenberg et al. (1996)

oprL, 504 De Vos et al. (1997)

toxA, 367 Khan and Cerniglia, 1994

gyrB, 222 Qin et al., 2003

16S–23S rDNA ITS, 181 Tyler et al. (1995)

ecfX, 528 This work

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

4 R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

2.3. DNA extractions from environmental samples

For water samples, 100 ml (for wastewaters) or 500 ml (otherwaters) (Table 5) were filtered using 0.22 μm HA membranefilters (Millipore, Molsheim, France). Filters were then groundin liquid nitrogen, and DNA was extracted as describedpreviously (Ranjard et al., 2000). For soil samples, DNAextractions from La Côte Saint André, Montrond, St-Elie andLamto sites were performed using a direct lysis method(Ranjard et al., 2000). This procedure involved samplehomogenization and cell disruption in liquid nitrogen followedby enzymatic lysis. DNA extraction of the other soils wasperformed using the FastDNA® SPIN® Kit (For Soil) (BIO 101,Inc., Carlsbad, CA). Extractions were performed on threesamples, and the extracted DNA were pooled.

2.4. PCR screenings

PCR amplifications of the ecfX, 16S rDNA, gyrB, oprI,oprL, toxA, 16S–23S rDNA ITS and fliC genes were performedin 50 μl using the Expand High Fidelity Taq polymerase(Roche, Neuilly-sur-Seine, France). The reaction mix contained2.5 U Taq polymerase, 200 μM dNTP, 1X PCR reaction buffer,2 mM MgCl2, 10% DMSO (except for the oprL target), and0.5 μM of each primer. For soil DNA extracts, 0.5 μg T4 gene32 protein (Roche) was added to the PCR mix. PCR reactionswere performed using 5 μl of cells from frozen stocks (forstrains listed in Table 1) or 100 ng of water and soil DNAextracts. PCR amplifications were done using a RapidCycler(MJ Research Inc. PTC-100, USA) with an initial denaturationstep of 5 min at 95 °C, 35 cycles of 45 s at 94 °C, 45 s at theappropriate annealing temperature (Table 2), and 45 s or 1 min30 s (16S rDNA and fliC targets) at 72 °C, and a final elon-gation step of 5 min at 72 °C. PCR products were visualised byelectrophoresis using a 2% agarose gel stained with ethidiumbromide. Each PCR was, at least, duplicated.

2.5. DNA cloning, PCR products endonuclease restrictionprofiles and DNA sequencing

The ecfX, 16S–23S rDNA ITS and gyrB PCR productsamplified from environmental DNA samples were agarose gelpurified using the Qiaquick gel extraction kit (Qiagen,Courtaboeuf, France) according to the manufacturer's instruc-tions. Purified PCR products were cloned in E. coli DH5αcompetent cells (Invitrogen, Cergy Pontoise, France) using thepGEMT easy vector system (Promega, Charbonnières, France).For each DNA clonings, six clones were randomly picked, andtheir pGEMT inserts restricted, respectively, by TaqI for theecfX products, Bsp143I for the 16S–23S rDNA ITS ones, andAluI for the gyrB ones. All restriction enzymes were usedaccording to the manufacturer's instructions (Euromedex,Mundolsheim, France). Restricted inserts were analysed using2.5% metaphor gels (TebuBio, Le Perray en Yvelines, France).Clones were divided into RFLP groups, and three pGEMTsubclones were fully sequenced per group (Genome Express,Meylan, France).

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

2.6. DNA blots of SmaI restricted Pseudomonas total DNAs

To validate the specificity of the ecfX PCR screening, anelectrophoresis agarose gel of SmaI restricted total DNAs of thefifteen Pseudomonas species, listed in Table 1, was transferredonto a nylon membrane. This gel was transferred using a VacuumBlotter model 785 of BIORAD (Hertfordshire, England),according to the manufacturer's instructions. This blot was thenhybridized using 32P-labeled ecfX PCR fragments amplified fromP. aeruginosa strain PAO1. Radioactive labelings of the purifiedPCR products were performed using the random priming DNAlabeling kit (Amersham). Hybridization conditions were asdescribed previously (Cournoyer et al., 1995).

2.7. Computer analyses

A selection of Pseudomonas sigma 70 factor sequencesavailable at GenBank (http://www.ncbi.nih.gov) was alignedusing the CLUSTALW software (Higgins and Sharp, 1988).Neighbour-Joining trees were computed through the PHYLO_-WIN graphic tool (Galtier et al., 1996). All BLAST analyseswere run at NCBI (http://www.ncbi.nlm.nih.gov).

2.8. Nucleotide sequences accession numbers

DNA sequences of the PCR products are available at GenBankunder the following accession numbers: ecfX sequences(#DQ996551 and DQ99552 for soil DNA; DQ99553 toDQ996559 for water DNA), gyrB sequences (#DQ996547 andDQ996548 for soil DNA; DQ996549 and DQ996550 for waterDNA), and 16S–23S rDNA ITS sequences (#DQ996560 for soilDNA; DQ996561–DQ996566 for water DNA).

3. Results and discussion

3.1. Efficacy of P. aeruginosa PCR screenings on isolatedbacteria

PCR screenings using the following DNA targets wereevaluated: (1) the 16S rDNA, (2) the internally transcribed16S–23S rDNA spacer (ITS), (3) gyrB encoding the DNA gyrasesubunit B, (4) oprI encoding the lipoprotein I, (5) oprL encodingan outer-membrane peptidoglycan-associated lipoprotein, (6) fliCencoding the C-terminus flagellin, and (7) toxA encoding theexotoxin A precursor (Table 2). Most of these targets are found inother bacterial species, with the exception of toxA which is P.aeruginosa-specific (Khan and Cerniglia, 1994). The oprI gene isconserved among the pseudomonads (De Vos et al., 1997). TheoprL gene has been detected in several other genera includingthe Burkholderia (Plesa et al., 2004). Sensitivity of the PCRscreenings was estimated on a panel of 59 P. aeruginosa strains.Their specificity was estimated on a panel of 15 Pseudomonasspecies closely related to P. aeruginosa (see Table 1).

The P. aeruginosa oprI, oprL, 16S rDNA and fliC PCRscreenings were found highly sensitive, giving the expectedPCR products for all the strains of the P. aeruginosa panel(Table 3). However, they showed a poor specificity giving oprI

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

Table 3Sensitivity and specificity of P. aeruginosa PCR screenings on panels ofP. aeruginosa strains and species of pseudomonads closely related toP. aeruginosa a

P. aeruginosatargeted gene

Number of positive reactions/number of strains tested

Sensitivity(%)

Specificity(%)

P. aeruginosa Pseudomonas sp.

16S rDNA 59/59 1/15 100 94oprI 59/59 3/15 100 80fliC 59/59 13/15 100 13oprL 59/59 2/15 100 86toxA 55/59 0/15 95 100gyrB 59/59 0/15 100 100

16S–23S rDNAITS

59/59 0/15 100 100

ecfX b 59/59 0/15 100 100a Strains are listed in Table 1.b Type strains of two additional species, P. oleovorans (CIP 59.11) and

P. resinovorans (CIP 61.9), did not yield any PCR products with this screening.

5R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

PCR products for P. fragi, P. citronellolis and P. viridiflava,oprL products for P. balearica and P. citronellolis, 16S rDNAproducts for P. fragi, and fliC products for almost all thePseudomonas species tested in this study, except P. fluorescensand P. chlororaphis (Table 3). The oprL, oprI, and 16S rDNAP. aeruginosa PCR screenings were previously shown to behighly sensitive (De Vos et al., 1997; Qin et al., 2003; Spilkeret al., 2004), and sufficiently specific for use on clinical samples.However, our screening of environmental Pseudomonas speciessuggests that they would not be sufficiently specific to be usedon environmental DNA extracts. The fliC PCR screening used inthis study was previously defined to investigate fliC geneticdiversity among P. aeruginosa CF isolates (Spangenberg et al.,1996), and not to detect P. aeruginosa in various samples. It wasthus not surprising to observe a high sensitivity but a poorspecificity of this screening. The oprL, oprI, 16S rDNA, andfliC false positive PCR products confirm the great geneticconservation of these genes among the pseudomonads but showthe difficulties in building up a highly reliable screening withsuch targets. It is clear from these data that the more the analysedsample will have an important diversity in terms of speciescontent, themore false positive PCR products will be obtained. Itthus appears reasonable to suggest avoiding the use of thesescreenings to confirm the taxonomic status of P. aeruginosa-likecolony forming units (CFU) on selective media or to detectP. aeruginosa DNA in various samples including CF sputum.

However, PCR screenings making use of two other highlyconserved genetic loci, gyrB and the 16S–23S rDNA ITS, gaveencouraging results indicative of high sensitivity and specific-ity. The P. aeruginosa gyrB and ITS PCR screenings gave theexpected PCR products for all P. aeruginosa strains of our panel(100% sensitivity), and did not yield any products for the otherPseudomonas species tested (100% specificity) (Table 3).

A widely used PCR target for the characterization ofP. aeruginosa-like CFU, which is not distributed in other eubac-teria, is the toxA gene. However, the toxA PCR screening used inthis study did not yield PCR products for all the P. aeruginosa

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

strains of our panel (Table 3). It showed a sensitivity of 95%.A similar result was obtained by Khan and Cerniglia (1994)who previously observed that other pseudomonads and someP. aeruginosa isolates would not yield PCR products with thetoxA PCR screening. From this study and others, the toxA genetarget thus appeared to introduce a bias in the P. aeruginosaidentification process by neglecting about 5% of its diversity.

3.2. Efficacy of P. aeruginosa PCR screenings on environ-mental DNA extracts

Outdoor environments are known to hide a much greaterdiversity in terms of number of bacterial species than the oneobserved in most tissues/organs infected or colonized byP. aeruginosa including lungs of cystic fibrosis (CF) patients.These differences in diversity can be appreciated by comparingbacterial community profiles generated by an approach like theterminal restriction fragment (T-RF) length polymorphism one.This latter approach allowed to deduce a species richness ofabout 13.4±6.7 species per adult CF patient (Rogers et al.,2004). Similar estimations using DNA reassociation kineticsdata suggested a soil species richness of about one millionspecies per gram (Gans et al., 2005). The best approach tovalidate the reliability of the gyrB and 16S–23S rDNA ITSP. aeruginosa PCR screenings thus appeared to test them onDNA extracts from soil and water environments. Twenty nineDNA extracts of water samples collected from the Yzeron river(Oullins) and the Bourg-en-Bresse sub-urban area (France), 9DNA extracts of soil samples coming from various countries,and 3 DNA extracts of horse manures coming from EcoleNationale Vétérinaire de Lyon (Marcy l'Etoile, France) wereused to perform these tests (Table 4). The presence andabundance of P. aeruginosa in these environmental sampleswere verified using selective culture media (Table 4). Theseplatings revealed P. aeruginosa-like CFU in the three sampledwastewater treatment lagoons (32 to 200 CFU ml−1), in the soilpolluted with hydrocarbons (325 CFU g−1 of dry soil) and inthe horse manure composted for three months (3040 CFU g−1)(Table 4). Status of these CFU was confirmed by variousbiochemical and molecular tests (see Materials and methods).

The DNA extracts of samples found to contain culturableP. aeruginosa cells were used to pursue our testing of thesensitivity and specificity of the gyrB and 16S–23S rDNA ITSPCR screenings. The gyrB PCR screening did not yield theexpected PCR products from DNA extracts of one of the 3wastewater treatment lagoons, and from the one extracted froma hydrocarbon contaminated soil (Table 4). The 16S–23S rDNAITS PCR did not yield PCR products from the DNA extract ofthe 3-month old horse manure shown to contain more than3000 CFU g−1 of P. aeruginosa (Table 4). The poor sensitivityof these screenings on these environmental DNA extracts couldbe due to Taq polymerase inhibitors like humic acids that areoften co-extracted with DNA (Wilson, 1997). Nevertheless, inthree instances, these PCR screenings revealed the presence ofP. aeruginosa-like DNA in extracts from samples that had notshown P. aeruginosa CFU on selective media. The gyrB PCRscreening detected P. aeruginosa-like DNA in the extract of a

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

Table 4Detection of P. aeruginosa in environmental samples using the PCR screeningsdescribed in Table 2, and a selective culture medium approach

Environmental sample Selectivemedium(CFU/ml)

PCR products (% identity withPAO1 sequence)

ecfX 16S–23S rrnITS

gyrB

Freshwater samplesWastewater treatment lagoon1

200 +(99%)

+ (100%) +(100%)

Wastewater treatment lagoon2

54 +(100%)

+ (100%) +(100%)

Wastewater treatment lagoon3

32 +(99%)

+ (100%) −

Pond 1 0 +(99%)

+ (100%) −

Ponds 2 to 6 0 − − −Road ditches 1 to 3 0 − − −Brooks 1 to 5 0 − − −River 1 0 +

(91%)− −

Rivers 2 to 4 0 − − −River 5 0 +

(100%)+ (98%) −

Water tank 0 +(99%)

+ (100%) −

Yzeron river 1 to 6 (OTHU) 0 − − −

Soil samples (management) CFU/g drysoil

La Côte Saint André, France(cultivated)

0⁎ − − −

Montrond, France (pasture) nd − − −Lamto, Ivory Coast (savanna) nd − − −Ô mon, Vietnam (paddy field) nd − − −Auxonne, France (cultivated) nd − − −Châteaurenard, France(cultivated)

nd − − −

St Victor, France (cultivated) nd − − −St Elie, French Guyana(forest)

nd − − −

Nancy, France(polluted by hydrocarbons)

325 +(100%)

+ (100%) −

Horse manure samples CFU/g drymanure

Freshly delivered horse-dung 0⁎ − − +(86%)

3 days-old horse manure 0⁎ − − −3 months-old horse manure 3040 +

(100%)− +

(100%)

+, positive; − negative; ⁎ even after acetamide enrichment; nd: not done.

6 R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

fresh horse-dung (Table 4). The ITS PCR screening detectedP. aeruginosa-like DNA in extracts from water samples of ariver, a pond, and a water tank (Table 4). The observed varia-tions in sensitivity between these PCR screenings could be dueto the number of copies per genome of the two DNA targetsused which is of four for the rrn operon and one for the gyrBlocus (Botes et al., 2003). Some of the differences observedbetween P. aeruginosa detections by selective culture mediaand PCR screenings could be due to dead or viable but notculturable P. aeruginosa cells in the environmental samples.

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

To confirm that the PCR products obtained from environmen-tal DNA extracts truly revealed the presence of P. aeruginosa,RFLP and DNA sequence analyses of these products wereperformed. AluI and Bsp143I RFLP analyses revealed identicalprofiles for the gyrB and 16S–23S rDNA ITS PCR productscoming from all samples. PCR products coming from a samesample were found to have identical sequences. However, DNAsequences of the 16S–23S rDNA ITS coming from differentsamples showed from 98 to 100% identity with the ITS DNAsequence of P. aeruginosa strain PAO1 (Table 4). BLASTnanalyses of these ITS sequences highlighted, in the GenBankdatabase, the presence of a P. fluorescens (accession #AY818674)DNA sequence showing 98% identity with the ITS of the PAO1strain. These analyses thus brought uncertainties on the specificityof the P. aeruginosa ITS PCR screening. Similarly, a DNAsequence analysis of the gyrB PCR products obtained from a freshhorse-dung DNA (accession #DQ996547.1) extract showed 86%identity with the P. aeruginosa strain PAO1 gyrB sequence(accession #DQ996547.1). This important percentage of diver-gence between sequences amplified by this P. aeruginosa PCRscreening, and the fact that P. mendocina gyrB (accession#AB039480) shows 85% identity with the one of strain PAO1,contributed to bring some doubts on the specificity of thisscreening. These sequences were further analysed by molecularphylogeny. A gyrB Neighbour-Joining phylogenetic tree ofsequences coming from P. aeruginosa and its closest relativesshowed this horse-dung sequence to group significantly withgyrB from P. mendocina and Pseudomonas nitroreducens(accession #AB176844.1) (for 94% of the bootstrap replicates),apart from the P. aeruginosa sequences (data not shown). Thesedata thus suggest again that the use of highly conserved DNAtargets (among eubacterial species) could lead to false positivePCR products.

It thus appears that the gyrB and 16S–23S rDNA ITSshowed 100% specificity and sensitivity on bacterial culturecollections but could not be used with high confidence on waterand soil DNA extracts. False positive PCR products couldbe obtained with these screenings. It was also observed thatsome of these screenings on environmental DNA extracts couldlead to false negative results, even though large numbers of P.aeruginosa CFUs could be detected, from these environments,on selective media. These observations led us to investigate thepossibility of designing a more reliable PCR screening. For thisscreening, the use of genes encoding sigma 70 factors wasinvestigated because several of these determinants appear to bespecies-specific (Menard et al., 2005; Lavire et al., 2004).

3.3. The ecfX P. aeruginosa PCR screening

The σ70 gene family varies in size from one bacterial speciesto another. The largest σ70 family, made of 65 paralogs, hasbeen reported for Streptomyces ambofaciens (Bentley et al.,2002), and the smallest one, made of a single gene, for Myco-plasma spp. (Himmelreich et al., 1996; Dandekar et al., 2000).The P. aeruginosa strain PAO1 genome was found to contain 23σ70 paralogs (Martinez-Bueno et al., 2002). This family can bedivided into four groups (Lavire et al., 2004): (1) the essential

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

7R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

primary sigma factors, involved mainly in exponential growth,(2) the stationary phase factors, (3) the factors involved insporangium development, flagellin synthesis and heat shockresponse, and (4) the extra-cytoplasmic function sigma factors

Fig. 1. Molecular phylogeny of 132 Pseudomonas sigma factors. Distances are esequences. Bootstrap values over 85% are indicated. Amino acids sequences are namnumber. PA: Pseudomonas aeruginosa; Pflu: Pseudomonas fluorescens; PP: Pseudomsyringae; Psyrpisi: Pseudomonas syringae pv. pisi; PSPTO: Pseudomonas syringae pin this study (see Table 4). Bold lineages indicate P. aeruginosa ECF factors signifispecies used in this analysis. Known homologs of P. aeruginosa ECF sigma factorsarrow, and appear under the term “neighbour”. See text for their description.

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

(termed ECF) (see Fig. 1). The most important number ofparalogs among this family are found in the ECF group. 19 ECFsigma factors have been detected in the P. aeruginosa PAO1genome; several of these showing only distant similarities with

xpressed in percentage of observed divergence between pairs of amino acidsed according to their genome sequencing project gene number or their accessiononas putida; Psyr: Pseudomonas syringae; Psyrsyr: Pseudomonas syringae pv.v. tomato; ECFX: environmental EcfX deduced amino acids sequences obtainedcantly apart (N40% observed divergence) from other lineages of Pseudomonasare indicated in brackets. Some genes surrounding ecf loci are indicated by an

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

8 R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

those of the closely related P. putida genome (Martinez-Buenoet al., 2002).

In order to identify species-specific factors, the phylogeneticproximity of P. aeruginosa ECF factors (protein sequences)with those of P. putida, P. chlororaphis, P. oleovorans, P.tolaasii, P. syringae and P. fluorescens was investigated. Onehundred and thirty two ECF-like sequences from thesepseudomonads were aligned, and a Neighbour-Joining phylo-genetic tree was computed (Fig. 1). Six P. aeruginosa ECF-likesigma factors (PA1363, PA1300, PA2093, PA4896, PA1912 andPA3410) showed observed divergence percentages, with otherfactors of the Pseudomonas spp., of more than 40% (Fig. 1),and were kept for further analyses. An analysis of genessurrounding these loci was performed. The PA1300 locus isfound upstream a gene encoding an HxuC-like product. HxuCwas previously shown to play a role in the utilization of haem–haemopexin complexes and low levels free haem (Cope et al.,1995). Similarly, the PA3410 locus is upstream has geneswhich have been shown involved in haem-uptake (Létoffé et al.,1999). The PA1912 locus encodes a FecI-like factor most likelyinvolved in Fe-uptake. PA1363 is upstream a cirA-like gene thatmight be involved in Fe-transport. cirA was found related toPIG genes which are Fe-regulated (Ochsner and Vasil, 1996).Concerning the other loci, no clear function could be inferredfor their surrounding genes. However, the deduced proteins ofPA2093 and PA4896 loci were respectively showing similaritieswith PrhI-like (involved in quorum sensing) and PupI (involvedin the synthesis of siderophores) sigma factors. All thesededuced sigma factors are grouped in a same phylogenetic

Table 5BLASTp most closely related GenBank homologs of a selection of P.aeruginosa ECF sigma factors, and sensitivity of P. aeruginosa PCRscreenings targeting their respective open reading frame (ORF)

P. aeruginosatargeted ORF

BLASTpresult a (%identity)

Species (accessionnumber)

# of PCR positive/# ofP. aeruginosa tested b

PA1363 51 Ralstoniasolanacearum(ZP_00944553)

8/10

PA1300 51 Yersinia mollaretii(ZP_00826317)

10/10

PA2093 58 Pseudomonasfluorescens(ABA74864)

9/10

PA4896 79 Pseudomonasentomophila(YP_606375)

6/10

PA1912 70 Nitrosomonaseuropaea(CAD85010)

6/10

PA3410 59 Pseudomonasentomophila(YP_610209)

9/10

a BLASTn did not detect significant hits in the GenBank database other thanP. aeruginosa orthologs of these sequences. All these targets were found highlyconserved in the sequenced P. aeruginosa genomes.b P. aeruginosa strains used in this screening: ATCC 14425, ATCC 31479,

CFBP 5033, CFBP 5035, CIP 58.46, CIP 60.92, CIP 72.26, DSMZ6195,LMG5031 and PA01(1) (see Table 1).

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

cluster (Fig. 1) which was found to include all P. aeruginosafactors showing similarities with FecI or other factors involvedin Fe-uptake like PvdS, which regulates the synthesis ofpyoverdin (Leoni et al., 2000). BLASTp searches showedPA1363 and PA1300 deduced proteins to be the most specificP. aeruginosa ECF, with b51% identity conserved with sigmafactors from other eubacteria available at GenBank (Table 5).No significant BLASTn hit was detected in GenBank using thesix selected ecf loci, apart from those of the P. aeruginosaorthologs available in this database. PCR primers were designedto test the specificity and sensitivity of the selected targets, inthe frame of the development of a more reliable P. aeruginosa-specific PCR screening. These primers were designed to matchthe beginning and the end of these sequences. PCR screeningsusing these primers were initially tested on 10 P. aeruginosastrains (ATCC 14425, ATCC 31479, CFBP 5033, CFBP 5035,CIP 58.46, CIP 60.92, CIP 72.26, DSMZ6195, LMG5031 andPAO1). PCR products of the expected size could only beobtained repeatedly for the PA1300 gene target; giving a 100%sensitivity (Table 5). The PA1300 target was kept for furtheranalyses, and renamed ecfX because of its obvious role intranscription and position in the ECF group of σ70 factors but,so far, undemonstrated role in haem-uptake and virulence. Thedefined primers for this target were named ECF1 and ECF2(Table 2).

The sensitivity of the ecfX PCR screening was further testedon the panels of 59 P. aeruginosa strains and 15 Pseudomonasspecies defined for this study (Table 1). Two additional species,Pseudomonas oleovorans (CIP 59.11) and Pseudomonasresinovorans (CIP 61.9), were included in these tests. Theexpected PCR products were obtained for all P. aeruginosastrains, giving a 100% sensitivity (Table 3). No PCR product wasobtained from the other Pseudomonas species, giving a 100%specificity (Table 3). Sensitivity was further investigated on anadditional set of 241 clinical P. aeruginosa strains of variousorigin (pus from ears and skin, urine, blood, and peritoneal andbronchial aspiration fluids). PCR products of the expected sizewere obtained for all these isolates, suggesting that the ecfX genetarget is highly conserved among P. aeruginosa.

The ecfX PCR screening was tested on the 29 water, 9 soil and3 horse manure DNA extracts presented in the above section(Table 4). This screening yielded the expected PCR products fromall DNA extracts of samples previously shown to containculturable P. aeruginosa, using selective media. It also allowedto detect P. aeruginosa-like DNA in the samples that were notcontaining culturable P. aeruginosa but gave positive PCRproducts using the 16S–23S ITS P. aeruginosa PCR screening(section above). Interestingly, the ecfX PCR screening did notyield a PCR product using the fresh horse-dung DNA extract(Table 4). This extract had previously yielded a gyrB PCRproductwhich was shown to have an 86% identity with the one of P.aeruginosa PAO1. The ecfX PCR screening thus confirmed ourconclusion that this gyrB PCR product was not coming from a P.aeruginosa strain but rather from another Pseudomonas species.

DNA sequence analyses of the ecfX-like PCR products wereperformed. From 91% to 100% identity between these PCRproducts and the ecfX PAO1 ortholog was observed (Table 4).

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

9R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

BLASTn analysis against the GenBank database did not showDNA sequences from bacterial species other than P. aeruginosamatching significantly with these ecfX sequences. Molecularphylogenetic analysis grouped all deduced EcfX amino acidssequences into a single significant cluster, apart from other ECFfactors from the pseudomonads (Fig. 1). A DNA blot analysis ofrestricted total DNA of the panel of Pseudomonas species usedto test the specificity of the P. aeruginosa PCR screenings wasperformed, using a 32P-labeled ecfX (from strain PAO1) DNAprobe. Positive hybridization signals could only be obtainedfromP. aeruginosa restricted DNA, suggesting either an absenceof the ecfX genetic determinant in the genome of these closelyrelated Pseudomonas species or a divergence of more than 20%(theoretical limit to detect a hybridization signal) between thisgene sequence and homologous sequences found in thesegenomes. These data are in line with the high sequencedivergences observed between ECF orthologs. For example,BLAST2 sequences analyses of hrpL, a sigma factor involved inthe induction of the plant hypersensitive response and bacterialvirulence, showed up to 11% nucleotide substitutions among theP. syringae genomovar I (Cournoyer et al., 1996), a speciesdefined according to genomic DNA analyses. This range ofpercentage of observed divergence among a single species isvery similar to the one found between the ecfX orthologsdetected in P. aeruginosa. hrpL orthologs were also shown tohave around 17% differences when compared between closelyrelated genomospecies belonging to the P. syringae complexlike P. syringae pathovar tomato (accession #AF232004) and P.syringae pathovar maculicola (accession #AB016341). Allthese data and observations thus tend to suggest that the ecfXPCR screening is P. aeruginosa specific and highly sensitive.

The ecfX PCR screenings performed in this study suggestwater rather than soil environments as the preferential biotopes ofP. aeruginosa. The next step in this study will be to use the ecfXPCR screening on a larger scale and large number ofenvironmental samples, to confirm this trend. As reportedelsewhere (Ganguli and Tripathi, 1999; Cavalca et al., 2000;Filali et al., 2000), our data confirmed the presence of P.aeruginosa in hydrocarbon polluted soils. P. aeruginosarecovered from petroleum contaminated samples were previouslyfound able to use saturated C12 to C22 alkanes as a carbon source(Belhaj et al., 2002). Some of these isolates were also shown toencode an alkane-1-monooxygenase (alkane hydroxylase) allow-ing the metabolism of C6 to C10 alkanes. Our data support theidea that P. aeruginosa strains naturally encode several enzymepathways allowing their growth in alkane polluted sites.

Acknowledgements

Raphaël Lavenir was supported by a PhD fellowship from the“Prospective” Rhône-Alpes Region (France) program. We thankthe CNRS, Université Lyon 1, and Rhône-Alpes Region (France)for having financially supported parts of this work. This workwasalso partly funded by the Agence Nationale de la Recherche(ANR) SEST project #0010705, and the AFSSET #ES-2005-008project. This work is part of an OTHU (“Field Observatory forUrbanWaterManagement”) research action.We thank the French

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

“Collège de Bactériologie, de Virologie et d'Hygiène desHôpitaux”, and particularly Dr Allouch, for having madeavailable a collection of clinical Pseudomonas aeruginosa. Wethank L. Ranjard for having kindly provided soil samples fromBurgundy and Provence. We thank Claire Monnez, Bruno Tillyand Elisabeth Brothier (UMR 5557; CNRS, Université Lyon 1and ENVL) for their technical support.

References

Aoi, Y., Nakata, H., Kida, H., 2000. Isolation of Pseudomonas aeruginosa fromUshubetsu River water in Hokkaido, Japan. Jpn. J. Vet. Res. 48, 29–34.

Belhaj, A., Desnoues, N., Elmerich, C., 2002. Alkane biodegradation in Pseu-domonas aeruginosa strains isolated from a polluted zone: identification ofalkB and alkB-related genes. Res. Microbiol. 153, 339–344.

Bentley, S.D., Chater, K.F., Cerdeno-Tarraga, A.M., Challis, G.L., Thomson, N.R.,James, K.D., et al., 2002. Complete genome sequence of the modelactinomycete Streptomyces coelicolor A3(2). Nature 417, 141–147.

Botes, J., Williamson, G., Sinickas, V., Gurtler, V., 2003. Genomic typing ofPseudomonas aeruginosa isolates by comparison of Riboprinting andPFGE: correlation of experimental results with those predicted from thecomplete genome sequence of isolate PAO1. J. Microbiol. Methods 55,231–240.

Cavalca, L., Di Gennaro, P., Colombo, M., Andreoni, V., Bernasconi, S., Ronco,I., Bestetti, G., 2000. Distribution of catabolic pathways in somehydrocarbon-degrading bacteria from a subsurface polluted soil. Res.Microbiol. 151, 877–887.

Cope, L.D., Yogev, R., Muller-Eberhard, U., Hansen, E.J., 1995. A gene clusterinvolved in the utilization of both free heme and heme:hemopexin byHaemophilus influenzae type b. J. Bacteriol. 177, 2644–2653.

Cournoyer, B., Sharp, J.D., Astuto, A., Gibbon, M.J., Taylor, J.D., Vivian, A.,1995. Molecular characterization of the Pseudomonas syringae pv. pisiplasmid-borne avirulence gene avrPpiB which matches the R3 resistancelocus in pea. Mol. Plant–Microb. Interact. 8, 700–708.

Cournoyer, B., Arnold, D., Jackson, R., Vivian, A., 1996. Phylogenetic evidencefor a diversification of Pseudomonas syringae pv pisi race 4 strains into twodistinct lineages. Phytopathology 86, 1051–1056.

Dandekar, T., Huynen, M., Regula, J.T., Ueberle, B., Zimmermann, C.U.,Andrade, M.A., et al., 2000. Re-annotating the Mycoplasma pneumoniaegenome sequence: adding value, function and reading frames. Nucleic AcidsRes. 28, 3278–3288.

De Vos, D., Lim Jr., A., Pirnay, J.P., Struelens, M., Vandenvelde, C., Duinslaeger,L., et al., 1997. Direct detection and identification ofPseudomonas aeruginosain clinical samples such as skin biopsy specimens and expectorations bymultiplex PCR based on two outer membrane lipoprotein genes, oprI andoprL. J. Clin. Microbiol. 35, 1295–1299.

Filali, B.K., Taoufik, J., Zeroual, Y., Dzairi, F.Z., Talbi, M., Blaghen, M., 2000.Waste water bacterial isolates resistant to heavy metals and antibiotics. Curr.Microbiol. 41, 151–156.

Galtier, N., Gouy, M., Gautier, C., 1996. SEAVIEW and PHYLO_WIN: twographic tools for sequence alignment and molecular phylogeny. Comput.Appl. Biosci. 12, 543–548.

Ganguli, A., Tripathi, A.K., 1999. Survival and chromate reducing ability of Pseu-domonas aeruginosa in industrial effluents. Lett. Appl. Microbiol. 28, 76–80.

Gans, J., Wolinsky, M., Dunbar, J., 2005. Computational improvements reveal greatbacterial diversity and high metal toxicity in soil. Science 309, 1387–1390.

Green, S.K., Schroth, M.N., Cho, J.J., Kominos, S.K., Vitanza-jack, V.B., 1974.Agricultural plants and soil as a reservoir for Pseudomonas aeruginosa.Appl. Microbiol. 28, 987–991.

Higgins, D.G., Sharp, P.M., 1988. CLUSTAL: a package for performingmultiple sequence alignment on a microcomputer. Gene 73, 237–244.

Himmelreich, R., Hilbert, H., Plagens, H., Pirkl, E., Li, B.C., Herrmann, R.,1996. Complete sequence analysis of the genome of the bacterium Myco-plasma pneumoniae. Nucleic Acids Res. 24, 4420–4449.

Hogardt, M., Trebesius, K., Geiger, A.M., Hornef, M., Rosenecker, J.,Heesemann, J., 2000. Specific and rapid detection by fluorescent in situ

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.

10 R. Lavenir et al. / Journal of Microbiological Methods xx (2007) xxx–xxx

ARTICLE IN PRESS

hybridization of bacteria in clinical samples obtained from cystic fibrosispatients. J. Clin. Microbiol. 38, 818–825.

Hunter, P.R., 1993. The microbiology of bottled natural mineral waters. J. Appl.Bacteriol. 74, 345–352.

Ishihama, A., 2000. Functional modulation of Escherichia coli RNA polymerase.Annu. Rev. Microbiol. 54, 499–518.

Jaffe, R.I., Lane, J.D., Bates, C.W., 2001. Real-time identification ofPseudomonasaeruginosa direct from clinical samples using a rapid extraction method andpolymerase chain reaction (PCR). J. Clin. Lab. Anal. 15, 131–137.

Khan, A.A., Cerniglia, C.E., 1994. Detection of Pseudomonas aeruginosa fromclinical and environmental samples by amplification of the exotoxin A geneusing PCR. Appl. Environ. Microbiol. 60, 3739–3745.

Kimata, N., Nishino, T., Suzuki, S., Kogure, K., 2004. Pseudomonas aeruginosaisolated from marine environments in Tokyo Bay. Microb. Ecol. 47, 41–47.

Kurupati, P., Kumarasinghe, G., Laa Poh, C., 2005. Direct identification ofPseudomonas aeruginosa from blood culture bottles by PCR-enzymelinked immunosorbent assay using oprI gene specific primers. Mol. Cell.Probes 19, 417–421.

Lashev, L., Lasarova, S., 2001. Pharmacokinetics and side-effects of gentamicinin healthy and Pseudomonas aeruginosa infected sheep. J. Vet. Pharmacol.Ther. 24, 237–240.

Lavire, C., Blaha, D., Cournoyer, B., 2004. Selection of unusual actinomycetalprimary sigma 70 factors by plant-colonizing Frankia strains. Appl.Environ. Microbiol. 70, 991–998.

Leoni, L., Orsi, N., de Lorenzo, V., Visca, P., 2000. Functional analysis of PvdS,an iron starvation sigma factor of Pseudomonas aeruginosa. J. Bacteriol.182, 1481–1491.

Létoffé, S., Nato, F., Goldberg, M.E., Wandersman, C., 1999. Interactions ofHasA, a bacterial haemophore, with haemoglobin and with its outermembrane receptor HasR. Mol. Microbiol. 33, 546–555.

Lonetto, M.A., Brown, K.L., Rudd, K.E., Buttner, M.J., 1994. Analysis of theStreptomyces coelicolor sigE gene reveals the existence of a subfamily ofeubacterial RNA polymerase sigma factors involved in the regulation ofextracytoplasmic functions. Proc. Natl. Acad. Sci. 91, 7573–7577.

Marlier, D., Mainil, J., Linde, A., Vindevogel, H., 2000. Infectious agentsassociated with rabbit pneumonia: isolation of amyxomatous myxoma virusstrains. Vet. J. 159, 171–178.

Martin Barrasa, J.L., Lupiola Gomez, P., Gonzalez Lama, Z., Tejedor Junco, M.T.,2000. Antibacterial susceptibility patterns of Pseudomonas strains isolatedfrom chronic canine otitis externa. J. Vet. Med. B. Infect. Dis. Vet. PublicHealth 47, 191–196.

Martinez-Bueno, M.A., Tobes, R., Rey, M., Ramos, J.L., 2002. Detection ofmultiple extracytoplasmic function (ECF) sigma factors in the genome ofPseudomonas putida KT2440 and their counterparts in Pseudomonasaeruginosa PAO1. Environ. Microbiol. 4, 842–855.

McManus, A.T.,Mason Jr., A.D.,McManus,W.F., Pruitt Jr., B.A., 1985. Twenty-fiveyear review ofPseudomonas aeruginosa bacteremia in a burn center. Eur. J. Clin.Microbiol. 4, 219–223.

Menard, A., Graindorge, A., Balandreau, J., Thibault, F., Cournoyer, B., 2005.Progress on the characterization of Burkholderia cepacia complex σ70 genefamily. Abstract available at http://go.to/cepacia 10th International BurkholderiacepaciaWorkingGroupMeeting (IBCWG), April 22–24, 2005, Oklahoma, OK,USA.

Morales, A., Garland, J.L., Lim, D.V., 1996. Survival of potentially pathogenichuman-associated bacteria in the rhizosphere of hydroponically grown wheat.FEMS Microbiol. Ecol. 20, 155–162.

Please cite this article as: Lavenir, R. et al. Improved reliability of Pseudomonas aeMicrobiol. Methods (2007), doi:10.1016/j.mimet.2007.03.008

Ochsner, U.A., Vasil, M.L., 1996. Gene repression by the ferric uptake regulatorin Pseudomonas aeruginosa: cycle selection of iron-regulated genes. Proc.Natl. Acad. Sci. 93, 4409–4414.

Pellett, S., Bigley, D.V., Grimes, D.J., 1983. Distribution of Pseudomonasaeruginosa in a riverine ecosystem. Appl. Environ. Microbiol. 45, 328–332.

Pirnay, J.P., De Vos, D., Duinslaeger, L., Reper, P., Vandenvelde, C., Cornelis,P., Vanderkelen, A., 2000. Quantitation of Pseudomonas aeruginosa inwound biopsy samples: from bacterial culture to rapid ‘real-time’polymerase chain reaction. Crit. Care 4, 255–261.

Plesa, M., Kholti, A., Vermis, K., Vandamme, P., Stavroula, P., Winstanley, C.,Cornelis, P., 2004. Conservation of the opcL gene encoding thepeptidoglycan-associated outer-membrane lipoprotein among representa-tives of the Burkholderia cepacia complex. J. Med. Microbiol. 53, 389–398.

Qin, X., Emerson, J., Stapp, J., Stapp, L., Abe, P., Burns, J.L., 2003. Use of real-time PCR with multiple targets to identify Pseudomonas aeruginosa andother nonfermenting gram-negative bacilli from patients with cystic fibrosis.J. Clin. Microbiol. 41, 4312–4317.

Raivio, T.L., Silhavy, T.J., 2001. Periplasmic stress and ECF sigma factors.Annu. Rev. Microbiol. 55, 591–624.

Ranjard, L., Poly, F., Nazaret, S., 2000. Monitoring complex bacterialcommunities using culture-independent molecular techniques: applicationto soil environment. Res. Microbiol. 151, 167–177.

Richard, P., Le Floch, R., Chamoux, C., Pannier, M., Espaze, E., Richet, H.,1994. Pseudomonas aeruginosa outbreak in a burn unit: role ofantimicrobials in the emergence of multiply resistant strains. J. Infect. Dis.170, 377–383.

Rogers, G.B., Carroll, M.P., Serisier, D.J., Hockey, P.M., Jones, G., Bruce, K.D.,2004. Characterization of bacterial community diversity in cystic fibrosislung infections by use of 16S ribosomal DNA terminal restriction fragmentlength polymorphism profiling. J. Clin. Microbiol. 42, 5176–5183.

Romling, U., Wingender, J., Muller, H., Tummler, B., 1994. A major Pseudo-monas aeruginosa clone common to patients and aquatic habitats. Appl.Environ. Microbiol. 60, 1734–1738.

Romling, U., Kader, A., Sriramulu, D.D., Simm, R., Kronvall, G., 2005.Worldwide distribution of Pseudomonas aeruginosa clone C strains in theaquatic environment and cystic fibrosis patients. Environ. Microbiol. 7,1029–1038.

Saiman, L., Siegel, J., 2004. Infection control in cystic fibrosis. Clin. Microbiol.Rev. 17, 57–71.

Schwartz, T., Volkmann, H., Kirchen, S., Kohnen, W., Schon-Holz, K., Jansen,B., Obst, U., 2006. Real-time PCR detection of Pseudomonas aeruginosa inclinical and municipal wastewater and genotyping of the ciprofloxacin-resistant isolates. FEMS Microbiol. Ecol. 57, 158–167.

Spangenberg, C., Heuer, T., Burger, C., Tummler, B., 1996. Genetic diversity offlagellins of Pseudomonas aeruginosa. FEBS Lett. 396, 213–217.

Spilker, T., Coenye, T., Vandamme, P., LiPuma, J.J., 2004. PCR-based assay fordifferentiation ofPseudomonas aeruginosa from other Pseudomonas speciesrecovered from cystic fibrosis patients. J. Clin. Microbiol. 42, 2074–2079.

Tyler, S.D., Strathdee, C.A., Rozee, K.R., Johnson, W.M., 1995. Oligonucle-otide primers designed to differentiate pathogenic pseudomonads on thebasis of the sequencing of genes coding for 16S–23S rRNA internaltranscribed spacers. Clin. Diagn. Lab. Immunol. 2, 448–453.

Wilson, I.G., 1997. Inhibition and facilitation of nucleic acid amplification.Appl. Environ. Microbiol. 63, 3741–3751.

ruginosa PCR detection by the use of the species-specific ecfX gene target. J.