Suicide vectors for antibiotic marker exchange and
rapid generation of multiple knockout mutants by allelic exchange
in Gram-negative bacteria
Inmaculada Ortiz-Martín, Alberto P. Macho, Lotte Lambersten
1,
Cayo Ramos, Carmen R. Beuzón
⁎
Área de Genética, Departamento de Biología Celular, Genética y Fisiología, Universidad de Málaga, Campus de Teatinos s/n, Málaga, 29071, Spain
Received 10 January 2006; received in revised form 5 April 2006; accepted 18 April 2006 Available online 5 June 2006
Abstract
Allelic exchange is frequently used in bacteria to generate knockout mutants in genes of interest, to carry out phenotypic analysis and learn about their function. Frequently, understanding of gene function in complex processes such as pathogenesis requires the generation of multiple mutant strains. In Pseudomonads and other non-Enterobacteriaceae, this is a time-consuming and laborious process based on the use of suicide vectors and allelic exchange of the appropriate mutant version of each gene, disrupted by a different antibiotic marker. This often implies the generation of a series of mutants for each gene, each disrupted by a different antibiotic marker, in order to obtain all possible double or multiple mutant combinations. In this work, we have modified this method by developing a set of 3 plasmid derivatives from the previously described suicide vector for allelic exchange, pKAS32, to make antibiotic marker exchange easier and thus accelerate the entire process. Briefly, the construction of each single gene knockout mutant is carried out by allelic exchange of the chromosomal gene with a mutant allele disrupted by the insertion of a kanamycin resistance cassette. When a double mutant strain is required, antibiotic marker exchange is performed in either one of the single mutants, using any of the three plasmid derivatives that carry the kanamycin resistance gene disrupted by either a chloramphenicol, gentamycin, or streptomycin resistance cassette. The single mutant strain, carrying now an antibiotic resistance marker other than kanamycin, can be used to introduce a second mutation using the original plasmid constructs, to generate a double mutant. The process can be repeated sequentially to generate multiple mutants. We have validated this method by generating strains carrying different combinations of mutations in genes encoding different transcriptional regulators of the Hrp type III secretion system inPseudomonas syringae. We have also tested the genetic organisation and stability of the resulting mutant strains during growth in laboratory conditions as well asin planta.
© 2006 Elsevier B.V. All rights reserved.
Keywords:Allelic exchange;Pseudomonas syringae; Pathogenesis; Type III secretion system; Virulence
1. Introduction
The availability of more than 150 complete bacterial genome sequences (bacterial genome repository at NCBI http://www.ebi.ac.uk/genomes/bacteria.html)
⁎ Corresponding author. Tel.: +34 952131959; fax: +34 952132001. E-mail address:[email protected](C.R. Beuzón).
1Current address: Department of Bacteriology, Mycology and
Parasitology, Building 43, Room 405A, Statens Serum Institut, Artillerivej 5, 2300 Copenhagen S, Denmark.
0167-7012/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2006.04.011
has provided a great amount of information on the molecular structure and organisation of a multitude of genes whose functions need to be determined. This is usually achieved by reverse genetic analysis of the genes of interest, i.e. knockout of the target gene via allelic exchange followed by phenotypic characterisa-tion of the resulting mutant strain (Donnenberg and Kaper, 1991; Miller and Mekalanos, 1988). Thorough functional characterisation of complex processes often requires the generation of multiple mutant strains. Such is the case for type III secretion systems (TTSS) and their effector proteins, where individual mutations frequently display no phenotype (Ruiz-Albert et al., 2002), or multi-factorial regulatory systems (Beuzón et al., 2000, 2001). Foresight of potentially interesting mutant combinations allows antibiotic marker combi-nation to be taken into account when generating the corresponding single mutants. However, in reverse genetics it is mostly through phenotypic analysis of single mutants how the relevance of a particular combination of mutations becomes apparent. Thus, de novo construction of mutants with the appropriate selective markers is usually required (Shea et al., 1999).
One of the most common strategies to generate knockout mutants in Gram-negative bacteria are suicide vectors for allelic exchange, such as those carrying the replication origin of R6K that replicate only in strains producing the πprotein from λ phage (Miller and Mekalanos, 1988). Maintenance of R6K derivatives under selective pressure requires plasmid integration by homologous recombination. These vectors can be used to carry out allelic exchange of wild-type genes by plasmid-encoded disrupted alleles. This is a two-step procedure with plasmid integration within the target gene by a recombination event, followed by its excision via a second crossover, which renders allelic exchange. Integration of the plasmids into the chromosome can be selected by means of the antibiotic marker used to disrupt the target gene (first recombination event). Excision of the integrated plasmid that results in allelic exchange (second recombination event), may be either selected using counter-selectable markers (Donnenberg and Kaper, 1991; Selbitschka et al., 1993), or screened for on the basis of antibiotic resistance (Miller and Mekalanos, 1988). This method requires cloning and disruption of the target gene into the suicide vector prior to allelic exchange, a time-consuming and laborious process, particularly when generating multiple mutants since it frequently implies de novo disruption. Although one-step mutagenesis methods have recently been
devel-oped that considerably simplify a generation of single and multiple knockout mutant strains in Enterobacter-iaceae (Datsenko and Wanner, 2000), similar methods are not available for Pseudomonads, or other Gram-negative bacteria.
Pseudomonas syringae is a plant pathogenic bacteria of major agricultural and economic concern. The ability of P. syringae to cause infection in susceptible hosts is dependent on a large number of virulence factors that include many regulatory pro-teins, a type III protein secretion system (Hrp TTSS), and a suite of secreted effector proteins (Buell et al., 2003; Feil et al., 2005; Joardar et al., 2005). Molecular analyses of the contribution that each factors plays in the virulence process, as well as the functional relationships between them are essential to understand the pathogenesis of P. syringae, thus requiring phenotypic analysis of multiple mutant strains. The species P. syringae is divided into pathovars largely on the basis of host specificity (Dye et al., 1980). P.
syringae pv. phaseolicola causes halo blight in bean and requires, among other virulence determinants, a functional Hrp type III secretion system (Lindgren et al., 1986). Expression of hrp genes in P. syringae is activated by HrpL, an alternative sigma factor of the ECF (extracytoplasmic factor) family (Xiao et al., 1994). Transcription of hrpL is in turn activated by HrpR and HrpS, a two component regulatory system of the NtrC family that work in concert with σ54 (Grimm et al., 1995; Grimm and Panopoulos, 1989). Lon protease has been reported to degrade HrpR in
hrp-repressing conditions therefore repressing the expression of hrp genes (Bretz et al., 2002). Additionally, HrpV has been reported to down regulate expression of hrp genes in hrp-inducing conditions although it is unclear how it acts (Preston et al., 1998). To date, no analysis of double or multiple mutant in any combination of these regula-tory factors has been carried out.
In this work, we have developed a set of plasmid derivatives of pKAS32, an oriR6K vector (Skorupski and Taylor, 1996), to make antibiotic marker exchange easier and thus accelerate the process of generating multiple mutants. Single gene knockout mutants are generated by allelic exchange of the wild-type alleles with mutant alleles disrupted by anaphAgene (Oka et al., 1981), that confers resistance to kanamycin, and carries its own promoter but no transcriptional termina-tor. When a double mutant strain is required, antibiotic marker exchange is carried out in any of the single mutants using one of three pKAS32 derivatives carrying the aphA gene disrupted by either a chloramphenicol
(Cm), gentamycin (Gm), or streptomycin (Sm) resis-tance gene. Thus, a second mutation can be introduced into the new strain using the original plasmid construct previously applied to the generation of the single mutant. This allows double or multiple mutant strains to be rapidly generated and tested for potentially interesting phenotypes.
We demonstrate the validity of this approach by disruptinghrpL,hrpV, andlongenes ofP.syringaepv.
phaseolicola, as well as hrcC, which encodes a structural component of the type III secretion apparatus (Charkowski et al., 1997), and generating several combinations of double and triple mutations. We also analyse the genetic organisation and stability of the resulting mutant strains when growing in laboratory conditions as well asin planta.
2. Materials and methods
2.1. Bacterial strains and growth conditions
Bacterial strains used in this study are listed in
Table 1. Bacteria were grown at 37 °C (Escherichia coli strains) or 28 °C (P. syringae pv. phaseolicola
strains) with aeration in LB medium supplemented with ampicillin (100μg/ml forE.colistrains; 300μg/ml for liquid cultures and 500 μg/ml for plates, for
P. syringae pv. phaseolicola strains), kanamycin (15 μg/ml), chloramphenicol (30 μg/ml for E. coli
strains; 6 μg/ml for P. syringae pv. phaseolicola
strains), streptomycin (50 μg/ml), or gentamycin (10μg/ml), as appropriate.
2.2. Plasmids
Plasmids used in this work are listed in Table 2. pUC18N-Km is a pUC18Not derivative (Biomedal, Sevilla, Spain; pUC18Not is a pUC18 derivative [Gen Bank/EMBL L09136]) that carries theSphI fragment of pMKm (Murillo et al., 1994) containing a kanamycin resistance cassette (aphA). This fragment was treated to render blunt ends, and cloned into the SmaI site of pUC18Not.
To obtain pIOM14, pIOM15, and pIOM17, three intermediary plasmids, pUC18N–Km–Gm, pUC18N– Km–Sm, and pUC18N–Km–Cm, were constructed. A 2 Kb SalI fragment from each pMGm and pSmUC (Murillo et al., 1994) containing a gentamycin resistance and a streptomycin resistance cassette, respectively, was cloned into the XhoI site of pUC18N–Km, rendering pUC18N–Km–Gm and pUC18N–Km–Sm. TheXhoI site cleaves the
kanamy-cin resistance cassette in two bands of approximately 1 Kb and 0.8 Kb. To generate pUC18N–Km–Cm, a fragment containing a chloramphenicol resistance cartridge was amplified by PCR from pRK600 (Kessler et al., 1992) using the appropriate primers (Table 3), which introduced aXhoI site at both 5′and 3′ends of the fragment. The resulting PCR product was digested with XhoI and cloned into the XhoI site of pUC18N– Km (Fig. 1). pIOM14, pIOM15, and pIOM17 were obtained digesting pUC18N–Km–Gm, pUC18N–Km– Sm, and pUC18N–Km–Cm, withNotI and cloning the fragments corresponding to the antibiotic resistance cartridges into NotI-digested pKAS32 (Skorupski and Taylor, 1996).
Table 1
Bacterial strains used in this studya
Strain Description Source or reference
DH5α F-endA1 hsdR17 supE44 thi-1 recA1 gyrA96 relA1
ΔlacU189 f80
ΔlacZDM15
(Hanahan, 1983)
CC118λpir RifRΔ(ara-leu)araD
ΔlacX74 galE galK phoA20 thi-1 rpsE rpoB argE(Am)recA1,
λpir lisogen
(Herrero et al., 1990)
S17–1λpir Thi-Pro-Hsd-recA-zzz:: RP4-2 (tet::Mu, kan:: Tn7 [TpR, SmR])
(Simon et al., 1983)
HB101 SmRrecA thi por leu
hsdRM+
(Kessler et al., 1992)
P. syringaepv phaseolicola 1448a
Race 6, wild-type strain D. Teversonb
IOM3 ΔhrcC(KmR) This work IOM7 ΔhrpL(KmR) This work IOM9 ΔhrpV(KmR) This work IOM10 Δlon(KmR) This work IOM13 ΔhrpV(GmR) This work IOM16 ΔhrcC(GmR)ΔhrpL
(KmR)
This work
IOM28 Δlon(SmR) This work
IOM29 ΔhrcC(GmR)ΔhrpL
(SmR)
This work
IOM30 ΔhrcC(GmR)ΔhrpL
(SmR)Δlon(KmR)
This work
IOM31 Δlon(SmR)ΔhrpL
(KmR)
This work
IOM33 Δlon(SmR)ΔhrpV
(KmR)
This work
IOM34 ΔhrpL(CmR) This work
a
RifR, KmR, GmR, SmR, CmR, and TpR, indicate resistance to rifampicin, kanamycin, gentamycin, streptomycin, chloramphenicol, and trimetropin, respectively.
The ORFs fromhrcC,hrpL,hrpVandlon, together with their flanking regions were amplified by a PCR using genomic DNA fromP. syringaepvphaseolicola
(Pph) 1448a as template, and the appropriate primers (Table 3). These primers introduced aKpnI restriction site at the 5′ and a SacII site at the 3′ ends of the amplified fragments. PCR products corresponding to
hrpLand lon were digested with KpnI andSacII,and cloned into theKpnI–SacII sites of pKAS32, generating pKAS32–hrpL, and pBluescript SK (+) (Stratagene; La Jolla, California, USA), generating pBSK–lon. Frag-ments corresponding tohrcCand hrpVwere treated to generate blunt ends, and cloned into theEcoRV site of pKAS32, rendering pKAS32–hrcCand pKAS32–hrpV
(Fig. 2A). pKAS32–hrcC, pKAS32–hrpV, pKAS32–
hrpL and pBSK–lon, were amplified by PCR using Expand Long Template Polymerase (Roche; Mannheim, Germany) and the appropriate primers (Table 3). These PCR reactions amplify the entire length of the plasmid except the target gene, since primers are designed to amplify outwards from the ORF. Primers used for PCR of pKAS32–hrcC, pKAS32–hrpVand pKAS32–hrpL, introduced XhoI sites at both 5′ and 3′ ends of the amplified products. Primers used for PCR of pBSK–lon
introduced SphI sites instead. PCR products were
digested with XhoI orSphI as appropriate, and ligated to a 1.7 Kb SalI or SphI fragment from pMKm containing the aphA gene, generating pIOM8, pIOM11, pIOM10 and pBSKlonKm. pIOM16 was generated by cloning a 2.7 Kb KpnI–SacI fragment from pBSKlonKm into pKAS32 digested withKpnI and
SacI (Fig. 2B).
2.3. Conjugations
S17–1λpir (Simon et al., 1983) was used as donor strain for bi-parental conjugations, whereas CC118λpir (Herrero et al., 1990) and HB101 carrying pRK600 (Kessler et al., 1992), were used for tri-parental conjugations, as donor and auxiliary strains, respec-tively. Donor (D), receptor (R) and auxiliary (A) strains were grown overnight in LB at 28 °C or 37 °C with aeration, as appropriate. Cultures were then adjusted to an OD600 nmof approximately 4, combined
in 1:4 or 1:1:4 volume ratios of D:R or A:D:R, respectively, rinsed in fresh LB medium three times, spotted onto sterile filters set on LB plates, and incubated at 28 °C. Following overnight incubation, filters were removed from the plates, placed into NaCl 0.9% and vortexed. The resulting bacterial suspensions
Table 2
Plasmids used in this studya
Plasmids Description Antibiotic resistance Source or reference
pBluescript SK (+) Cloning vector Ap Stratagene (La Jolla, CA)
pUC18Not Cloning vector Ap Genbank accession no. L09136
pKAS32 rpsL, oriRK6,mob Ap (Skorupski and Taylor, 1996)
pMKm aphAgene Km (Murillo et al., 1994)
pMGm aacC1gene Gm (Murillo et al., 1994)
pSmUC aadA gene Sm (Murillo et al., 1994)
pRK600 catgene Cm (Kessler et al., 1992)
pUC18N–Km pUC18 derivative,aphA gene Ap, Km This work
pKAS32–hrcC ContainshrcCORF plus 0,5 Kb on each side Ap This work pKAS32–hrpL ContainshrpLORF plus 0,5 Kb on each side Ap This work pKAS32–hrpV ContainshrpVORF plus 0,5 Kb on each side Ap This work pBSK–lon ContainslonORF plus 0,5 Kb on each side Ap This work
pBSKlonKm ContainsΔlon::aphA Ap, Km This work
pUC18N–Km–Gm ContainsaphA::aacC1 Ap, Gm This work
pUC18N–Km–Sm ContainsaphA::aadA Ap, Sm This work
pUC18N–Km–Cm ContainsaphA::cat Ap, Cm This work
pIOM8 ContainsΔhrcC::aphA Ap, Km This work
pIOM10 ContainsΔhrpL::aphA Ap, Km This work
pIOM11 ContainsΔhrpV::aphA Ap, Km This work
pIOM14 ContainsaphA::aacC1 Ap, Gm This work
pIOM15 ContainsaphA::aadA Ap, Sm This work
pIOM16 Containslon::aphA Ap, Km This work
pIOM17 ContainsaphA::cat Ap, Cm This work
pIOM18b ContainsaphA::catb Ap, Cm This work
a ApR, KmR, GmR, SmRand CmRindicate resistance to ampicillin, kanamycin, gentamycin, streptomycin and chloramphenicol, respectively. b Orientation ofcatin respect toaphAis opposite to that in pIOM17.
were rinsed three times in NaCl 0.9% and plated onto selective media. Replica plates of the resulting colonies were carried out in plates supplemented with ampicillin (500 μg/ml) to determine whether each transconjugant was the result of plasmid integration or allelic exchange.
2.4. Plant experiments
Seeds of Phaseolus vulgaris cultivar Canadian Wonder (Thomas Etty Esquire [Kent, UK]) were germinated and grown with 16 h light–8 h dark cycles. Bacterial lawns were grown for 48 h in LB plates and resuspended in 10 mM MgCl2. Cell densities were
adjusted to an OD600 nm of 0.1 (corresponding to
approximately 5 × 10−7 colony forming units [cfu] per ml) and serial dilutions were carried out to reach the inoculation dose (104 cfu/ml). Eight days-old bean plants were inoculated by infiltrating 200 μl of bacterial suspension into the intracellular spaces of the primary leaves. Infiltration was achieved by pressing the bacterial suspension against the leaf underside with a 2 ml syringe without needle. Symptoms were analyzed 14 days post-inoculation (dpi). Leaf disks were taken with a 7 mm–diameter cork borer from infected tissue at 14 dpi. Ten leaf disks were taken per plant, placed into 1 ml of sterile distilled
water, and homogenized by mechanical disruption. Serial dilutions of the resulting bacterial suspension were plated onto selective plates.
2.5. DNA manipulations
Basic DNA and molecular techniques were per-formed following standard methods (Sambrook et al., 2001). Genomic DNA was extracted using the Jet Flex Extraction Kit (Genomed; Löhne, Germany), and plasmid DNA was extracted using either Nucleobond AX, or NucleoSpin Plasmid Quick Pure, depending on the scale of the extraction (Macherery-Nagel; Düren, Germany). Routine clone analysis was carried out by boiling plasmid extraction (Holmes and Quigley, 1981). DNA gel-purification was carried out using the GFX PCR DNA kit (Amersham; Little Chalfont, Buckin-ghamshire, England). Taq Polymerase or Expand Long Template Polymerase (Roche; Mannheim, Germany) was used as appropriate.
DNA hybridization was performed following stan-dard methods (Sambrook et al., 2001), using a DIG-Nucleic Acid Detection Kit (Roche; Mannheim, Ger-many), following the instructions provided by the manufacturer. DNA was transferred onto a nylon membrane by upward capillary transfer, and cross-linked by UV irradiation. Prehybridization and
Table 3
Primers used in this study
Name Sequence 5′-3′ Locationa Product (bp)
hrpL-F AAT GGT ACC TCG CCT TCA GCT CCA TCT TCC 1504993-5014 1690 (ORF: 555) hrpL-R AAT CCG CGG AAG AGT ATT CGG AGT TGT CG 1506661-81
hrpV-F AAT GGT ACC CTG ATG GTC GAT GGA ACT GC 1493858-77 1335 (ORF: 351) hrpV-R AAT CCG CGG ATT CAT TGC AAG GGT GAG GAC G 1495168-90
lon-F AAT GGT ACC AAG CTG TTC GAG ATG GAA GG 1963969-89 3225 (ORF: 2397) lon-R AAT CCG CGG TCT TAC CAG TCT GAG GGT TGC 1967171-92
hrcC-F AAT GGT ACC ATG GAT TTC AGT GAG TTC GC 1491681-701 2985 (ORF: 2100) hrcC-R AAT CCG CGG CAA AAT CAC GCT GTA CAT CC 1494644-63
hrpLinv-F AAT CCT CGA GAT TGA AGC AGC AGA TTG ACC 1506186-205 5430 hrpLinv-R AAT CCT CGA GTA CAG CTT TTG GCA CAA ACG 1505496-515
hrpVinv-F AAT CCT CGA GAC CAA CCT GGA TGA CAT ACG 1494702-21 5450 hrpVinv-R AAT CCT CGA GAC TTC ACG ACT GTG TAA TCC 1494384-403
loninv-F ATC GGC ATG CTA AAT GGA TTG ACG AAG TCC 1966711-30 5400 loninv-R ATC GGC ATG CGA GAG GCA ATT CAA TAG TGG 1964454-73
hrcCinv-F AAT CCT CGA GAA ATG GCG AAA GAG AGT CGG 1494165-84 5440 hrcCinv-R AAT CCT CGA GGT AAG CCG TGT GTT TCC AGG 1492191-210
Cm600-F TCA CTC GAG TCT GGA TTT GTT CAG AAC GC From pRK600b 800 Cm600-R TCA CTC GAG TCT TTC AC TGA GCC TTT CG
Km–aphA–F ATG AGC CAT ATT CAA CGG From pUC18NKmc 820
Km–aphA–R AGC ATC AAA TGA AAC TGC
aPosition in the annotated genome.
b These primers amplify thecatgene with its own promoter but without the transcriptional terminator. cThese primers amplify theaphAgene with its own promoter but without the transcriptional terminator.
NotI digestion and cloning into pKAS32 NotI
OriR6K
mob BglII/ EcoRV/ KpnI/ NotI/ SacII/ SacI/ EcoRI
MCS
pKAS32
4619 bp
pUC18N-Km-AntR 6386 bp
NotI
NotI
XhoI/SalI
XhoI/SalI
Ligation
NotI
NotI pUC18N-Km 4386 bp
XhoI
SalI
SalI
XhoI
SalI
SalI
XhoI aacC1
aadA
cat (CmR) (SmR) (GmR)
OriR6K
mob NotI
NotI
pIOM14
7450 bp a
OriR6K
mob
NotI NotI
pIOM15
7450 bp
mob OriR6K NotI NotI
pIOM17
6630 bp (ApR, CmR)
(ApR, SmR)
(ApR, GmR)
Fig. 1. Construction of pKAS32 derivatives for antibiotic marker exchange. pUC18N–Km is a pUC18Notderivative into which theaphAgene from pMKm, conferring kanamycin resistance, has been cloned. Three mutant alleles ofaphAwere obtained by cloning in itsXhoI site either aSalI fragment from pMGm containingaacC1(which confers resistance to gentamycin), aSalI fragment from pSmUC containingaadA(which confers resistance to streptomycin), or aXhoI fragment obtained by PCR from pRK600 containingcat(which confers resistance to chloramphenicol). These SalI fragments are collectively referred to as AntRin the figure.NotI fragments containingaphA::aacC1,aphA::aadA, andaphA::catwere obtained from the resulting plasmids and cloned intoNotI-digested pKAS32, rendering pIOM14, pIOM15, and pIOM17, respectively.
hybridization stages were carried out at 65 °C. DNA probe was labelled by PCR reaction with chemilumi-niscent digoxigenin-dNTPs using a DIG Labelling Mix
(Roche; Mannheim, Germany), primers Km–aphA–F and Km–aphA–R (Table 3) and pUC18N–Km as DNA template.
KpnI
pBSK-lon SphI SphI
SacIISacI
OriR6K
hrcCor hrpV
pKAS32-hrcC or pKAS32-hrpV
XhoI XhoI
mob BglII/ EcoRV/ KpnI/ NotI/ SacII/ SacI/ EcoRI
MCS
OriR6K
mob pKAS32
4619 bp
gene KpnI SacII
Amplification of target gene by PCR A
B
KpnI SacII
OriR6K
hrpL
pKAS32-hrpL XhoI XhoI
mob
KpnI-SacI digestion and cloning into KpnI-SacI sites of pKAS32
KpnI
SacII SacI
pBluescript SK (+) 2958 bp
mob OriR6K
KpnI
XhoI/SalI
SacII
XhoI/SalI
pIOM8 pIOM10 pIOM11
mob OriR6K
SacI
SphI
SphI
SacI
pIOM16 pBSKlonKm
KpnISphISphISacSacIII
XhoI or
SphI aphA
SalI or SphI +(ligation)
XhoI or
SphI
SalI or SphI
Fig. 2. Generation of pKAS32 derivatives for deletion ofhrpV,hrcC,hrpL, andlongenes. (A)hrpV,hrcC, andhrpLORFs plus 0.5 kb flanking regions from each side were amplified and cloned into pKAS32, whereaslonORF plus its l.5 Kb flanking regions were amplified and cloned into pBluescript SK (+). PCR reactions were carried out using the resulting plasmids as DNA template and outward facing primers that amplify the entire plasmid, except the ORFs of each gene, and introduce a restriction site. Dotted lines represent the amplified sequences (B) TheaphAgene carrying its own promoter but no transcriptional terminator, was ligated to the inverse PCR products obtained, rendering pIOM8, pIOM10, pIOM11, and pBSK– lon::aphA. The fragment containingΔlon::aphAfrom the former was cloned into pKAS32 460 rendering pIOM16.
3. Results
3.1. Construction of pKAS32 derivatives for antibiotic marker exchange
To facilitate generation of multiple knockout mutants, we constructed a set of suicide vectors for antibiotic marker exchange on mutant strains. We cloned the aphA gene from pMKm (Murillo et al., 1994; Oka et al., 1981), carrying its own promoter but no transcriptional terminator, into pUC18Not. Then, we disrupted aphA with three different selection markers, aacC1, aadA, and cat, which confer gentamycin, streptomycin, and chloramphenicol resis-tance, respectively (Kessler et al., 1992; Murillo et al., 1994). All three markers carry their own promoter, but only aadA carries a transcriptional terminator. The resulting vectors, pIOM14 (aphA::aacC1), pIOM15 (aphA::aadA), and pIOM17 (aphA::cat), contained the selection marker genes flanked by fragments of aphA
(Fig. 1). These flanking regions provide homology to allow the allelic exchange of aphA wild-type copies. These vectors can be introduced into the receptor strain by transformation or conjugation. To be introduced by conjugation, tra functions have to be providedin transeither by the donor or a helper strain since the pKAS32 plasmid backbone carries a mob
locus.
pKAS32 also carries therpsLgene fromE.coli, that encodes ribosomal protein S12. Mutations in rpsL
conferring resistance to streptomycin are recessive in a strain expressing the wild-type protein (Lederberg, 1951), allowingrpsLto be used as a counter-selectable marker for plasmid loss. Resistance to streptomycin due to expression ofaadA is not affected by expression of
rpsL.
3.2. Construction and genetic characterisation of hrpL, hrpV, hrcC, and lon mutants
P. syringae pv. phaseolicola (Pph) is divided into races according to cultivar specificity (Taylor et al., 1996). Race 6 groups a number of strains capable of causing infection in all bean cultivars (Tsiamis et al., 2000). Strain 1448a is the representative strain of race 6 and its genome has been fully sequenced (Joardar et al., 2005). Using the genome sequence, primers were designed to amplify the ORFs plus their flanking regions from hrpL, hrpV, hrcC, and lon. Amplified fragments were cloned and the complete ORFs deleted, and disrupted with theaphAgene (Fig. 2). The resulting plasmids, pIOM8 (ΔhrcC::aphA), pIOM10 (ΔhrpL::
aphA), pIOM11 (ΔhrpV::aphA), and pIOM16 (Δlon::
aphA) were introduced by conjugation into 1448a. Transconjugants were selected for resistance to kanamycin. In these experiments, as in all conjugation experiments carried out throughout this work, isolated transconjugant colonies were screened for sensitivity to ampicillin (pKAS32 derivatives selective marker), in order to identify which transconjugants have undergone allelic exchange, and therefore did not carry the plasmid still integrated (merodiploids for the target gene). Frequencies of both transconjugants and allelic ex-change are shown inTable 4.
In order to verifyaphAdisruption of the correct target gene, genomic DNA from the mutant strains was extracted, digested with the appropriate restriction enzyme, and subjected to Southern blot analysis using
aphA ORF as specific probe. As expected, a single hybridization band was obtained for each sample since restriction enzymes used in each sample cuts neither within aphAORF, nor within the target gene (Fig. 3). Band sizes depend on the location of the restriction sites nearest to each target gene, and corresponded to those predicted from the genome sequence.
Genomic DNA of these strains was also digested using XhoI, which cuts once inside aphA ORF. Using
aphA ORF as a probe, Southern blot analysis showed two hybridization bands for each strain, with sizes corresponding to those predicted for each gene disrup-tion (data not shown).
3.3. Antibiotic marker exchange and generation of multiple knockout mutant strains
Strains IOM7 (ΔhrpL; KmR), IOM9 (ΔhrpV; KmR), and IOM10 (Δlon; KmR) were selected for antibiotic marker exchange experiments, and used as receptor strains for conjugational transfer of plasmids pIOM17 (aphA::cat), pIOM14 (aphA::aacC1), and pIOM15 (aphA::aadA), respectively. Transconjugants were se-lected as CmR, GmR, and SmR, respectively, and checked for ampicillin sensitivity (Table 4). The resulting strains, IOM34 (ΔhrpL; CmR), IOM13 (ΔhrpV; GmR), and IOM28 (Δlon; SmR), respectively, were checked for sensitivity to kanamycin (KmS) to confirm disruption of theaphAgene. Disruption of the
aphA gene with cat, aacC1, and aadA in these three strains was also confirmed using PCR, and Southern blot analyses (Figs. 4, 5 and data not shown). Additionally, we tested the resulting strains for HR induction inN.tabacum.
Strains IOM13 (ΔhrpV; GmR) and IOM28 (Δlon; SmR), were used as receptor strains for the conjugative
transfer of plasmid pIOM10 (ΔhrpL::aphA), to generate double mutant strains. Transconjugants were selected as GmR KmR, and SmR KmR, respectively, and checked for ampicillin sensitivity (Table 4). Disruption of the
hrpLgene in both resulting strains (IOM16 and IOM31, respectively) was confirmed using PCR and Southern blot analyses (Figs. 4, 5and data not shown).
Strains IOM28 (Δlon; SmR) was used as receptor strain for the conjugative transfer of plasmids pIOM11 (ΔhrpV::aphA), to generate a double mutant strain. Transconjugants were selected as SmR KmR, and checked for ampicillin sensitivity (Table 4). Disruption of the target genes in the resulting strains (IOM33) was confirmed by Southern blot analysis (Fig. 4).
Strain IOM16 (ΔhrpV[GmR]; ΔhrpL [KmR]) was used as receptor strain for the conjugative transfer of plasmid pIOM15 (aphA::aadA) to exchange kanamycin resistance for streptomycin resistance. Transconjugants were selected as GmRSmR, and checked for sensitivity to kanamycin and ampicillin to confirm antibiotic marker exchange (Table 4). Antibiotic marker exchange
in the resulting strain (IOM29) was further confirmed using PCR (Fig. 5).
Strain IOM29 (ΔhrpV [GmR]; ΔhrpL [SmR]) was used as receptor strain for the conjugative transfer of plasmid pIOM16 (Δlon::aphA) to generate a triple mutant strain. Transconjugants were selected as GmR SmRKmRand checked for ampicillin sensitivity (Table 4). Disruption of the lon gene in the resulting strain (IOM30) was confirmed using PCR (Fig. 5).
Every time an antibiotic marker exchange is performed a “scar” of aphA sequences is left behind. Thus, upon entry of a plasmid carrying a gene disrupted by the aphA gene, integration of the plasmid can take place either by homologous recombination into the target gene, or by homologous recombination into the disrupted aphA gene. After excision of the integrated plasmid, the former would generate the double mutant strain intended. This strain would be resistant to kanamycin as well as the antibiotic corresponding to the gene used to disruptaphA. However, excision of the plasmid in the second case would result in the allelic
Table 4
Frequency of conjugation and allelic exchange
Receptors Transconjugants Allelic exchanged strain
Strain Description cfu/mla Selection used
Frequencyb T/Rc
Frequencyd K/Te
Description Strain
Generation of single mutants
1448a Wild-type 5 × 108 Km 1.36 × 10−6 10−2 ΔhrcC(KmR) IOM3
1448a Wild-type 5 × 108 Km 6.96 × 10−7 10−2 ΔhrpL(KmR) IOM7
1448a Wild-type 5 × 108 Km 6.44 × 10−7 10−2 ΔhrpV(KmR) IOM9
1448a Wild-type 5 × 108 Km 4.2 × 10−7 1.5 × 10−2 Δlon(KmR) IOM10
Antibiotic marker exchange in single mutants
IOM7 ΔhrpL(KmR) N.d. Cm N.d. 2.5 × 10−1 ΔhrpL(CmR) IOM34
IOM9 ΔhrpV(KmR) 2 × 108 Gm 1.5 × 10−7 3 × 10−2 ΔhrpV(GmR) IOM13 IOM10 Δlon(KmR) 1.6 × 108 Sm 3.12 × 10−7 4 × 10−2 Δlon(SmR) IOM28 Generation of double mutants
IOM13 ΔhrpV(GmR) 2 × 108 Gm, Km 1.35 × 10−2 6 × 10−3 ΔhrpV(GmR)ΔhrpL(KmR) IOM16 IOM28 Δlon(SmR) 8 × 106 Sm, Km 5.5 × 10−5 6 × 10−3 Δlon(SmR)ΔhrpL(KmR) IOM31 IOM28 Δlon(SmR) 1.2 × 108 Sm, Km 4.7 × 10−6 3 × 10−3 Δlon(SmR)ΔhrpV(KmR) IOM33 Antibiotic marker exchange in a double mutant
IOM16 ΔhrpV(GmR)ΔhrpL(KmR) 2 × 108 Gm, Sm 1.03 × 10−6 5 × 10−3 ΔhrpV(GmR)ΔhrpL(SmR) IOM29 Generation of a triple mutant
IOM29 ΔhrpV(GmR)ΔhrpL(SmR) 3.2 × 108 Gm, Sm, Km 2 × 10−6 2 × 10−2 ΔhrpV(GmR)ΔhrpL(SmR)
Δlon(KmR)
IOM30
a
Numbers are the results of one conjugation experiment per strain.
b
FrequencyT/Rreflects both frequency of conjugation and plasmid integration.
c
Cfu ml−1of transconjugants (T) divided by cfu/ml of receptors (R).
d
FrequencyK/Treflects frequency of plasmid excision.
e
exchange of a disruptedaphAallele by a wild-type one, and the resulting strains would therefore revert to be a single mutant strain, only resistant to kanamycin. By applying double or triple antibiotic selection, as appropriate in each case, to obtain the transconjugants, those transconjugants resulting from the allelic ex-change ofaphAwere eliminated. Only a small decrease in the frequencies of transconjugants were detected in these experiments (6 × 10−3to 2 × 10−2) when compared to similar ones carried out to generate the corresponding single mutants (10−2 to 1.5 × 10−2), indicating that of
aphA“scars”do not act as hotspots for allelic exchange (Table 4).
3.4. Genetic stability of multiple mutant strains in vitro and in planta
Double and triple mutant strains generated using this method carry fragments ofaphAin different loci, thus providing fragments of homology that can constitute targets for recombination events. Recombination events between these fragments would give rise to chromo-somal reorganisations, either inversions if recombina-tion took place between fragments in inverted orientation, or deletions if recombination took place between fragments in direct orientation. Thus, recom-bination events between aphA fragments in direct orientation would result in the deletion of one of the
copies of aphA, either the wild-type or the disrupted version, as well as the deletion of the chromosomal region between them, altering the hybridization, PCR profile, and antibiotic resistance of the strain. PCR analyses carried out after either growth in laboratory conditions or growthin plantaof the triple mutant strain (IOM30) did not detect any evidence of a profile that could correspond to such reorganisations in the bacterial population of any of the strains generated (Fig. 4). Furthermore, analysis of antibiotics resistance of this strain after either growth in laboratory conditions or growth in planta, carried out to detect reorganisations that could occur at a frequency below the level of detection of Southern blot analysis, showed that reorganisations occur at the same rate as any other recombination event in these loci (data not shown). These results further support the conclusion that aphA “scars”do not constitute a“hotspot”for recombination events.
1448a IOM7
ΔhrpL KmR
IOM34
ΔhrpL CmR
IOM10
Δlon KmR
IOM28
Δlon SmR
IOM31
Δlon SmR
ΔhrpL KmR
IOM33
Δlon SmR
ΔhrpV KmR
8.0 Kb
4.3 Kb
3.0 Kb
Fig. 4. Southern blot analysis of antibiotic-marker exchanged and double mutant strains. Southern blot analysis of genomic DNA from strains 1448a (wild-type), IOM7 (ΔhrpL; KmR), IOM34 (ΔhrpL; CmR), IOM10 (Δlon; KmR), IOM28 (Δlon; SmR), IOM31 (Δlon, SmR;ΔhrpL, KmR), and IOM33 (Δlon, SmR;ΔhrpV, KmR), digested withEcoRV, usingaphAORF as a probe. No hybridization signal was detected in 1448a. Single bands of 7.5 kb (corresponding toEcoRV fragment containinghrpL::aphA) and 8.4 kb (corresponding to the same fragment plus the size of thecatgene used to disruptaphA) were detected in IOM7 and IOM34, respectively. Single bands of 3.6 kb (corresponding toEcoRV fragment containinglon::aphA), and 5.7 kb (EcoRV fragment containinglon::aphAplus the size of theaadAgene disruptingaphA), were obtained in IOM10 and IOM28, respectively. Bands of 5.7 kb (EcoRV fragment containinglon::aphAplus the size of the aadA gene disrupting aphA) and 7.5 kb (EcoRV fragment containinghrpL::aphA), were detected in IOM31, whereas the same band of 5.7 kb plus a 5.6 kb band (EcoRV fragment containinghrpV:: aphA), were detected in IOM33.
1448a IOM3
hrcC
IOM9
hrpV
IOM7
Δ Δ
Δ hrpL
IOM10
Δlon
23.1 Kb
8.0 Kb
4.3 Kb
3.0 Kb
Fig. 3. Southern blot analysis and hypersensitive response in N. tabacumof single mutant strains. Southern blot analysis of genomic DNA from strains 1448a (wild-type), IOM3 (ΔhrcC), IOM9 (ΔhrpV), IOM7 (ΔhrpL), and IOM10 (Δlon), using aphAORF as a probe, indicate gene disruption of target genes. Genomic DNA from strains IOM3, IOM7, IOM9 (digested withSacII), and IOM10 (digested with EcoRV) render a single hybridization band in each case, with sizes corresponding to those predicted from the genome sequence.
As expected, frequencies were lower after growthin planta, probably reflecting that large chromosomal reorganisation would probably be deleterious under the strong selective pressure to be found within the plant.
4. Discussion
In this work, we have developed a set of plasmid vectors that allows antibiotic marker exchange in a single conjugation step. We have validated the efficacy and calculated the efficiency of the method in single and double mutant strains, as well as established its application to the rapid generation of multiple mutant strains. Once plasmids for gene knockout have been constructed, this set of vectors allows generation of multiple mutant combinations in a number of conjuga-tion steps equal to the number of mutaconjuga-tions to be combined, eliminating any cloning step and decreasing considerably the time usually required for this purpose. pKAS32 carries therpsLgene fromE.coli. Vectors carrying rpsL, which encodes ribosomal protein S12, allow the use of streptomycin as counter-selection of the
presence of the vector, since mutations in rpsL
conferring resistance to streptomycin are recessive in a strain expressing the wild-type protein (Lederberg, 1951). Thus, if streptomycin-resistant strains, due to mutation inrpsL, are used as receptors of these vectors in conjugation or transformation experiments, plasmid integration (first recombination event) will render bacteria sensitive to the antibiotic. Subsequent selection of SmR permits identification of transconjugants that have excised the plasmid (second recombination event). However, given the high frequencies of allelic exchange obtained in our system (4 × 10−2to 5 × 10−3;Table 4), and the fact that streptomycin-resistant 1448a grows considerably slower than 1448a, an undesirable charac-teristic for a strain to be tested for growth in planta
(Zumaquero A. and Beuzón, C.R. unpublished results), we decided against using rpsL as counter-selectable marker for the generation of the mutant strains, screening for them instead on the basis of antibiotic resistance. When knockout mutants were not detected among the transconjugants, they could be found after a round of growth in non-selective medium.
Interestingly, if further rounds of growth in non-selective medium were carried out, the frequency of knockout mutants dropped (data not shown), presum-ably due to a growth advantage of isolates with a higher
rpsL gene dosage. Nevertheless, streptomycin as counter-selection maker can be useful when generating mutants in strains with lower recombination frequen-cies, or natural resistance to high concentrations of ampicillin.
Antibiotic marker exchange can also be helpful in phenotypic characterisation of single mutants since swapping the antibiotic resistance cassette used in a particular mutation might also imply swapping from a polar to a non-polar mutation or vice versa. Since this method uses two types of antibiotic resistance cassettes, with and without transcriptional terminators, allelic exchange of one type of resistance gene (aphA, no transcriptional terminator) for the other (aadA, tran-scriptional terminator), would allow determination of phenotypic differences due to polarity on downstream genes. Furthermore, ifaphAdisruption of the target gene is designed in order to prevent expression of down-stream genes from theaphApromoter, further disruption of aphA using pIOM18, that carries the cat gene in opposite orientation toaphAtranscription, would cause expression of downstream genes from thecatpromoter, allowing determination of phenotypic differences due to constitutive overexpression of downstream genes. This is illustrated by the results described byPreston et al. (1998), where ahrpT::ntpIImutation caused a reduction
1448a IOM9
ΔhrpV
KmR
3.0 Kb
1.9 Kb
1.4 Kb
1.0 Kb
0.7 Kb
a b
IOM13
ΔhrpV
GmR IOM16
ΔhrpV
GmR
ΔhrpL
KmR IOM29
ΔhrpV
GmR
ΔhrpL
SmR
IOM30
ΔhrpL GmR
ΔhrpV SmR
Δlon KmR
Fig. 5. PCR analysis of antibiotic-marker exchanged, double and triple mutant strains. PCR analysis of strains IOM9 (ΔhrpV; KmR), IOM13 (ΔhrpV; GmR), IOM16 (ΔhrpV, GmR;ΔhrpL, KmR), IOM29 (ΔhrpV, GmR;ΔhrpL, SmR), and IOM30 (ΔhrpV, GmR;ΔhrpL, SmR;Δlon, KmR), using primers to amplifyaphAORF. No PCR product was obtained for strain 1448a; a 0.8 kb amplification product, corresponding to the size of aphAORF was obtained from strain IOM9; a 2.8 kb band corresponding to the size ofaphA::aacC1, was obtained from strain IOM13; two bands of 0.8 kb and 2.8 kb, corresponding to the sizes of the aphA ORF and aphA::aacC1, respectively, were obtained from strain IOM16; a band of 2.8 kb corresponding to the sizes of aphA::aadA and aphA::aacC1, was obtained from strain IOM29; and the same band plus a 0.8 kb band corresponding to the size ofaphAwas obtained from strain IOM30 (a). The same bands were obtained from IOM30 after 14 days of growthin planta(b).
in HrpZ expression through overexpression of hrpT -downstream hrpV gene from the nptII promoter, constituting one of the evidences that indicated that HrpV was a negative regulator of Hrp genes.
We also show that the “scars” of homology left behind in multiple mutants do not constitute a“hotspot” for recombination events, since genetic reorganisations mediated by them occur at the same rate as other recombination events in these loci (Table 4). Thus genetic stability of the resulting strains is sufficient for most phenotypic characterisations, allowing identifica-tion of mutant combinaidentifica-tions of particular interest. Further characterisation of such a mutant combination can then be carried out in a reconstructed mutant strain, reducing the number of de novo disruptions to be generated to those worthy of detailed analysis. Howev-er, although genetic stability of multiple mutants is sufficient for phenotypic analysis, it would not suffice if the intended use of the resulting strain were the isolation of variants, or its use for mutagenesis and screening, since chromosomal reorganisations do occur and could be selected during the screening process.
The method and vectors described in this work have been developed and tested inP.syringae, however they can be potentially used in any genetically amenable Gram-negative bacteria in which gene replacement is possible.
Acknowledgements
This work was supported by a project grant from the Ministerio de Ciencia y Tecnología (Spain) to Carmen R. Beuzón. Carmen R. Beuzón was supported by the
“Ramón y Cajal” Programme from the Ministerio de Ciencia y Tecnología. Inmaculada Ortiz-Martín was supported by a Fellowship from Junta de Andalucía. We are grateful to L. Cruzado, J. Guillamet, and T. Duarte for their practical assistance. We also want to thank Javier Ruiz-Albert and Eduardo R. Bejarano for their helpful discussion and critical reading of the manuscript.
References
Beuzón, C.R., Meresse, S., Unsworth, K.E., Ruiz-Albert, J., Garvis, S., Waterman, S.R., Ryder, T.A., Boucrot, E., Holden, D.W., 2000. Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. EMBO J. 19, 3235–3249.
Beuzón, C.R., Unsworth, K.E., Holden, D.W., 2001. In vivo genetic analysis indicates that PhoP–PhoQ and theSalmonella pathoge-nicity island 2 type III secretion system contribute independently to Salmonella enterica serovar Typhimurium virulence. Infect. Immun. 69, 7254–7261.
Bretz, J., Losada, L., Lisboa, K., Hutcheson, S.W., 2002.Lonprotease functions as a negative regulator of type III protein secretion in Pseudomonas syringae. Mol. Microbiol. 45, 397–409.
Buell, C.R., Joardar, V., Lindeberg, M., Selengut, J., Paulsen, I.T., Gwinn, M.L., Dodson, R.J., Deboy, R.T., Durkin, A.S., Kolonay, J. F., Madupu, R., Daugherty, S., Brinkac, L., Beanan, M.J., Haft, D. H., Nelson, W.C., Davidsen, T., Zafar, N., Zhou, L., Liu, J., Yuan, Q., Khouri, H., Fedorova, N., Tran, B., Russell, D., Berry, K., Utterback, T., Van Aken, S.E., Feldblyum, T.V., D'ascenzo, M., Deng, W.L., Ramos, A.R., Alfano, J.R., Cartinhour, S., Chatterjee, A.K., Delaney, T.P., Lazarowitz, S.G., Martin, G.B., Schneider, D. J., Tang, X., Bender, C.L., White, O., Fraser, C.M., Collmer, A., 2003. The complete genome sequence of the Arabidopsis and tomato pathogen Pseudomonas syringae pv. tomato DC3000. Proc. Natl. Acad. Sci. 100, 10181–10186.
Charkowski, A.O., Huang, H.C., Collmer, A., 1997. Altered localization of HrpZ inPseudomonas syringaepv.syringae hrp mutants suggests that different components of the type III secretion pathway control protein translocation across the inner and outer membranes of Gram-negative bacteria. J. Bacteriol. 179, 3866–3874.
Datsenko, K.A., Wanner, B.L., 2000. One-step inactivation of chromosomal genes inEscherichia coliK-12 using PCR products. Proc. Natl. Acad. Sci. 97, 6640–6645.
Donnenberg, M.S., Kaper, J.B., 1991. Construction of an eae deletion mutant of enteropathogenicEscherichia coliby using a positive-selection suicide vector. Infect. Immun. 59, 4310–4317. Dye, D.W., Bradbury, J.F., Goto, M., Hayward, R.A., Lelliot, R.A.,
Schroth, M.N., 1980. International standards for naming pathovars of phytopathogenic bacteria and a list of pathovar names and pathotype strains. Rev. Plant. Pathol. 59, 153–168.
Feil, H., Feil, W.S., Chain, P., Larimer, F., Dibartolo, G., Copeland, A., Lykidis, A., Trong, S., Nolan, M., Goltsman, E., Thiel, J., Malfatti, S., Loper, J.E., Lapidus, A., Detter, J.C., Land, M., Richardson, P.M., Kyrpides, N.C., Ivanova, N., Lindow, S.E., 2005. Comparison of the complete genome sequences of Pseudomonas syringae pv. syringae B728a and pv. tomato DC3000. Proc. Natl. Acad. Sci. 102, 11064–11069.
Grimm, C., Panopoulos, N.J., 1989. The predicted protein product of a pathogenicity locus fromPseudomonas syringaepv.phaseolicola is homologous to a highly conserved domain of several procaryotic regulatory proteins. J. Bacteriol. 171, 5031–5038.
Grimm, C., Aufsatz, W., Panopoulos, N.J., 1995. ThehrpRSlocus of Pseudomonas syringae pv. phaseolicola constitutes a complex regulatory unit. Mol. Microbiol. 15, 155–165.
Hanahan, D., 1983. Studies of transformation ofEscherichia coliwith plasmids. J. Mol. Biol. 166, 557–580.
Herrero, M., De Lorenzo, V., Timmis, K.N., 1990. Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in Gram-negative bacteria. J. Bacteriol. 172, 6557–6567.
Holmes, D.S., Quigley, M., 1981. A rapid boiling method for the preparation of bacterial plasmids. Anal. Biochem. 114, 193–197. Joardar, V., Lindeberg, M., Jackson, R.W., Selengut, J., Dodson, R., Brinkac, L.M., Daugherty, S.C., Deboy, R., Durkin, A.S., Giglio, M.G., Madupu, R., Nelson, W.C., Rosovitz, M.J., Sullivan, S., Crabtree, J., Creasy, T., Davidsen, T., Haft, D.H., Zafar, N., Zhou, L., Halpin, R., Holley, T., Khouri, H., Feldblyum, T., White, O., Fraser, C.M., Chatterjee, A.K., Cartinhour, S., Schneider, D.J., Mansfield, J., Collmer, A., Buell, C.R., 2005. Whole-genome sequence analysis ofPseudomonas syringaepv. phaseolicola 1448A reveals divergence among pathovars in
genes involved in virulence and transposition. J. Bacteriol. 187, 6488–6498.
Kessler, B., De Lorenzo, V., Timmis, K.N., 1992. A general system to integrate lacZfusions into the chromosomes of Gram-negative eubacteria: regulation of the Pm promoter of the TOL plasmid studied with all controlling elements in monocopy. Mol. Gen. Genet. 233, 293–301.
Lederberg, J., 1951. Streptomycin resistance: a genetically recessive mutation. J. Bacteriol. 61, 549–550.
Lindgren, P.B., Peet, R.C., Panopoulos, N.J., 1986. Gene cluster of Pseudomonas syringaepv.“phaseolicola”controls pathogenicity of bean plants and hypersensitivity of nonhost plants. J. Bacteriol. 168, 512–522.
Miller, V.L., Mekalanos, J.J., 1988. A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants inVibrio cholerae requirestoxR. J. Bacteriol. 170, 2575–2583.
Murillo, J., Shen, H., Gerhold, D., Sharma, A., Cooksey, D.A., Keen, N.T., 1994. Characterization of pPT23B, the plasmid involved in syringolide production by Pseudomonas syringae pv. tomato PT23. Plasmid 31, 275–287.
Oka, A., Sugisaki, H., Takanami, M., 1981. Nucleotide sequence of the kanamycin resistance transposon Tn903. J. Mol. Biol. 147, 217–226.
Preston, G., Deng, W.L., Huang, H.C., Collmer, A., 1998. Negative regulation of hrp genes in Pseudomonas syringae by HrpV. J. Bacteriol. 180, 4532–4537.
Ruiz-Albert, J., Yu, X.J., Beuzón, C.R., Blakey, A.N., Galyov, E.E., Holden, D.W., 2002. Complementary activities of SseJ and SifA regulate dynamics of the Salmonella typhimurium vacuolar membrane. Mol. Microbiol. 44, 645–661.
Sambrook, J., Fritsch, E.F., Maniatis, T., 2001. Molecular cloning: a laboratory manual, second ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
Selbitschka, W., Niemann, S., Puhler, A., 1993. Construction of gene replacement vectors for Gram (−) bacteria using a genetically modified sacRB gene as a positive selection marker. Appl. Microbiol. Biotechnol. 38, 615–618.
Shea, J.E., Beuzón, C.R., Gleeson, C., Mundy, R., Holden, D.W., 1999. Influence of the Salmonella typhimurium pathogenicity island 2 type III secretion system on bacterial growth in the mouse. Infect. Immun. 67, 213–219.
Simon, R., Priefer, U., Puler, A., 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Methods Enzymol. 118, 640–659.
Skorupski, K., Taylor, R.K., 1996. Positive selection vectors for allelic exchange. Gene 169, 47–52.
Taylor, J.D., Teverson, D.M., Allen, D.J., Pastor-Corrales, M.A., 1996. Identification and origin of races ofPseudomonas syringaepv. phaseolicola from Africa and other bean growing area. Plant Pathol. 45, 469–478.
Tsiamis, G., Mansfield, J.W., Hockenhull, R., Jackson, R.W., Sesma, A., Athanassopoulos, E., Bennett, M.A., Stevens, C., Vivian, A., Taylor, J.D., Murillo, J., 2000. Cultivar-specific avirulence and virulence functions assigned toavrPphFinPseudomonas syringae pv.phaseolicola, the cause of bean halo-blight disease. EMBO J. 19, 3204–3214.
Xiao, Y., Heu, S., Yi, J., Lu, Y., Hutcheson, S.W., 1994. Identification of a putative alternate sigma factor and character-ization of a multicomponent regulatory cascade controlling the expression ofPseudomonas syringaepv.syringaePss61hrpand hrmAgenes. J. Bacteriol. 176, 1025–1036.
