diversity and functional analysis of luxr-type transcriptional … · are regulators that are...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2009, p. 4753–4761 Vol. 75, No. 140099-2240/09/$08.00�0 doi:10.1128/AEM.00575-09Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Diversity and Functional Analysis of LuxR-Type TranscriptionalRegulators of Cyclic Lipopeptide Biosynthesis in

Pseudomonas fluorescens�†I. de Bruijn and J. M. Raaijmakers*

Laboratory of Phytopathology, Wageningen University, Wageningen, The Netherlands

Received 10 March 2009/Accepted 11 May 2009

Cyclic lipopeptides (CLPs) are produced by many Pseudomonas species and have several biological functions,including a role in surface motility, biofilm formation, virulence, and antimicrobial activity. This study focusedon the diversity and role of LuxR-type transcriptional regulators in CLP biosynthesis in Pseudomonas speciesand, specifically, viscosin production by Pseudomonas fluorescens strain SBW25. Phylogenetic analyses showedthat CLP biosynthesis genes in Pseudomonas strains are flanked by LuxR-type regulators that contain aDNA-binding helix-turn-helix domain but lack N-acylhomoserine lactone-binding or response regulator do-mains. For SBW25, site-directed mutagenesis of the genes coding for either of the two identified LuxR-typeregulators, designated ViscAR and ViscBCR, strongly reduced transcript levels of the viscABC biosynthesisgenes and resulted in a loss of viscosin production. Expression analyses further showed that a mutation ineither viscAR or viscBCR did not substantially (change of <2.5-fold) affect transcription of the other regulator.Transformation of the �viscAR mutant of SBW25 with a LuxR-type regulatory gene from P. fluorescens strainSS101 that produces massetolide, a CLP structurally related to viscosin, restored transcription of the viscABCgenes and viscosin production. The results further showed that a functional viscAR gene was required forheterologous expression of the massetolide biosynthesis genes of strain SS101 in strain SBW25, leading to theproduction of both viscosin and massetolide. Collectively, these results indicate that the regulators flanking theCLP biosynthesis genes in Pseudomonas species represent a unique LuxR subfamily of proteins and thatviscosin biosynthesis in P. fluorescens SBW25 is controlled by two LuxR-type transcriptional regulators.

Cyclic lipopeptides (CLPs) have diverse functions for soil-and plant-associated Pseudomonas species, including a role inmotility, biofilm formation, virulence, and defense againstcompeting (micro)organisms (34). CLPs are synthesized bylarge, nonribosomal peptide synthetases (NRPS), which forman assembly line of various enzymatic steps, resulting in themodification and stepwise incorporation of amino acids in thepeptide moiety (20, 39). For Pseudomonas species, the biosyn-thesis genes for syringomycin, syringopeptin, syringafactins,amphisin, arthrofactin, viscosin, massetolides, and putisolvinare now sequenced (2, 11, 12, 15, 22, 34). In comparison to theunderstanding of CLP biosynthesis, however, knowledge aboutthe genetic regulation is limited and fragmentary.

For the various Pseudomonas strains studied to date, theGacA/GacS two-component system is a key regulator of CLPbiosynthesis, as a mutation in either of the two encoding genesresults in a loss of CLP production (11, 12, 14, 27, 28). Forplant pathogenic Pseudomonas fluorescens strain 5064 and sa-prophytic Pseudomonas putida strain PCL1445, N-acylhomo-serine lactone (N-AHL)-mediated quorum sensing was shownto be required for viscosin and putisolvin biosynthesis, respec-tively (10, 16). In many other pathogenic and saprophyticPseudomonas strains, however, CLP production is not regu-

lated via N-AHL-mediated quorum sensing (1, 11, 18, 26).Downstream of the Gac system, the heat shock proteins DnaKand DnaJ were shown to regulate putisolvin biosynthesis in P.putida strain PCL1445 (14). Although their exact role is not yetresolved, it was suggested that DnaK and DnaJ may be re-quired for proper folding or activity of other regulators or areinvolved in the assembly of the peptide synthetase complex.Downstream of the Gac system in plant pathogenic Pseudomonassyringae pv. syringae, two genes, designated salA and syrF, wereidentified as LuxR-type transcriptional regulators of syringomycinand syringopeptin biosynthesis (27, 30, 31, 44). In the salA mu-tant, syringomycin/syringopeptin production was completely abol-ished, whereas production of these CLPs was reduced by 88% inthe syrF mutant, based on the results of activity assays with Geotri-chum candidum (30). LuxR-type transcriptional regulators werealso shown to be involved in the biosynthesis of syringafactins andputisolvin: mutations in syrF or psoR resulted in the loss of syrin-gafactin and putisolvin production in P. syringae pv. tomato andP. putida, respectively (2, 15).

The LuxR superfamily consists of transcriptional regulatorsthat contain a DNA-binding helix-turn-helix (HTH) motif inthe C-terminal region. Based on their activation mechanism,LuxR-type regulators can be divided into two major groups, asfollows: (i) regulators that belong to a two-component sensorytransduction system and are activated upon phosphorylation,e.g., FixJ of Sinorhizobium meliloti (3), and (ii) regulators thatare activated via binding to an autoinducer like N-AHL, e.g.,LuxR of Vibrio fischeri (21). In addition, there are variousLuxR-type regulators that contain the C-terminal HTH motifbut do not belong to these two subfamilies. Examples of these

* Corresponding author. Mailing address: Laboratory of Phyto-pathology, Droevendaalse Steeg 1, 6708 PD Wageningen, TheNetherlands. Phone: 31 317 483427. Fax: 31 317 483412. E-mail:[email protected].

† Supplemental material for this article may be found at http://aem.asm.org/.

� Published ahead of print on 15 May 2009.

4753

on April 29, 2020 by guest

http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

are regulators that are activated upon ligand binding, e.g.,MalT of Escherichia coli, induced by maltotriose (36), andso-called autonomous effector domain LuxR-type regulators,e.g., GerE, which is involved in regulating spore formation inBacillus subtilis (17). With respect to CLP biosynthesis, twoLuxR-type regulators have been identified to date. These in-clude PpuR, a LuxR-type regulator involved in putisolvin pro-duction by P. putida PCL1445 and which harbors a putativeN-AHL-binding domain (16), and the LuxR-type regulatorsSalA, SyrF, and SyrG in P. syringae pv. syringae; these latterLuxR-type regulators do not contain N-AHL-binding domainsor response regulator domains and, therefore, do not belong toeither of the two major LuxR-type subfamilies (27, 30, 31, 44).

This study focused on the function of specific LuxR-typeregulators in P. fluorescens SBW25, a strain isolated from thephyllosphere of the sugar beet (43) and extensively used as amodel for bacterial evolution and adaptation (35). StrainSBW25 produces the CLP viscosin, which plays an importantrole in swarming motility, biofilm formation, and activityagainst oomycete plant pathogens (12). Viscosin biosynthesis isgoverned by the NRPS genes viscA, viscB, and viscC. In con-trast to the organization of most other CLP gene clustersdescribed to date, the visc genes are not genetically linked, withviscA located approximately 1.62 Mb from viscBC on a linearmap of the SBW25 genome (12). The biosynthesis of viscosinis regulated by the Gac system but not by N-AHL-mediatedquorum sensing (11, 12, 44). Sequence analysis of the genesflanking the viscosin biosynthesis genes in strain SBW25 re-vealed the presence of two genes encoding LuxR-type tran-scriptional regulators designated ViscAR and ViscBCR. Bothregulators were analyzed for specific motifs and domains, in-cluding HTH motifs and N-AHL domains, and for their re-latedness to other LuxR-type transcriptional regulators. Therole of ViscAR and ViscBCR in viscosin production for strainSBW25 was investigated by site-directed mutagenesis, followedby gene expression and biochemical analyses.

MATERIALS AND METHODS

Bacterial strains and culture conditions. P. fluorescens strains SBW25 andSS101 were grown on Pseudomonas agar F (Difco) plates or in liquid King’smedium B (KB) at 25°C. The �viscA, �viscB, and �viscC biosynthesis mutantswere obtained as described previously (12). Escherichia coli strain DH5� wasused as a host for the plasmids used in site-directed mutagenesis and geneticcomplementation. E. coli strains were grown on Luria-Bertani (LB) plates or inLB broth amended with the appropriate antibiotics.

Bioinformatic analyses. Identification of the genes flanking the CLP biosyn-thesis genes was performed by either Blastx or Blastp analysis. For phylogeneticanalyses, alignments were made with ClustalX (version 1.81), and trees wereinferred by neighbor joining, using 1,000 bootstrap replicates. Autoinducer andresponse regulator domains were identified by Pfam analysis (http://pfam.sanger.ac.uk/search?tab�searchSequenceBlock).

Site-directed mutagenesis. Site-directed mutagenesis of the viscAR and vis-cBCR genes was performed based on the method described by Choi and Schwei-zer (9), generating �viscAR::Gmr (referred to as �viscAR) and �viscBCR::Gmr

(referred to as �viscBCR). The primers used for amplification are described inTable 1. The 5�- and 3�-end fragments were chosen in such a way that whenhomologous recombination in Pseudomonas takes place, the FRT-Gm-FRT cas-sette is inserted approximately 383 bp after the start of the viscAR and 192 bpafter the start of viscBCR open reading frames. The FRT-Gm-FRT cassette wasamplified with pPS854-GM, a derivative of pPS854 (24), and FRT-F and FRT-Rwere used as primers (Table 1). The first-round PCR was performed with KODpolymerase (Novagen), according to the manufacturer’s protocol. The programused for the PCR consisted of 2 min of denaturation at 95°C, followed by fivecycles subsequently of 95°C, 55°C, and 68°C, each for 20 s. The PCR amplifica-

tion was preceded with 25 cycles, with annealing at 60°C. All fragments were runon a 1% agarose gel and purified with a NucleoSpin kit (Macherey-Nagel). Thesecond-round PCR was performed by addition of equimolar amounts of the5�-end fragment, FRT-Gm-FRT, and 3�-end fragments, and the PCR programcontained three cycles subsequently of 95°C, 55°C, and 68°C for 20, 30, and 60 s,respectively. In the third extension cycle, 1.5 �l of the Up forward and Dn reverseprimers (10 �M) was added. The PCR amplification was preceded with 25 cyclessubsequently of 95°C, 58°C, and 68°C for 20, 20, and 120 s, respectively, and PCRfragments were purified as described above. The fragments were digested withBamHI or KpnI and cloned into BamHI- or KpnI-digested NucleoSpin-purifiedplasmid pEX18Tc. E. coli DH5� was transformed with the obtained pEX18Tc-viscAR and pEX18Tc-viscBCR plasmids by heat shock transformation, accordingto the work of Inoue et al. (25), and transformed colonies were selected on LBmedium supplemented with 25 �g/ml gentamicin (Sigma). Integration of theinserts was verified by PCR analysis with pEX18Tc primers (Table 1) and byrestriction analysis of isolated plasmids. The inserts were verified with sequenc-ing by BaseClear (Leiden, The Netherlands). The correct pEX18Tc-viscAR andpEX18Tc-viscBCR constructs were subsequently electroporated into strainSBW25. Electrocompetent cells were obtained according to the work of Choiet al. (8) by washing the cells three times with 300 mM sucrose from a 6-mlovernight culture and finally dissolving the cells in 100 �l of 300 mM sucrose.Electroporation occurred at 2.4 kV and 200 �F, and after incubation in SOCmedium (2% Bacto tryptone [Difco], 0.5% Bacto yeast extract [Difco], 10 mMNaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4, 20 mM glucose [pH 7]) for2 h at 25°C, cells were plated on KB supplemented with gentamicin (25 �g/ml).Obtained colonies were grown in LB medium for 1 h at 25°C and plated on LB

TABLE 1. Primers used in this study

Fragment usedfor indicated

testingOrientation Primer sequencesa

Site-directedmutagenesis

FRT Forward 5�-CGAATTAGCTTCAAAAGCGCTCTGA-3�Reverse 5�-CGAATTGGGGATCTTGAAGTTCCT-3�

viscAR-Up Forward 5�-TCAAGCAAGCGGATCCGAAAAGATCCGCACGCTGGA-3�

Reverse 5�-tcagagcgcttttgaagctaattcgGCGCTGTGTGATTTCGTTCT-3�

viscAR-Dn Forward 5�-aggaacttcaagatccccaattcgCAGCAGCTTGATATCGAGCAC-3�

Reverse 5�-TCAAGCAAGCGGATCCCGGGAGGAATGCTTCATA AG-3�

viscBCR-Up Forward 5�-TCAAGCAAGCGGTACCCGACACGGCAGATGAAACTG-3�

Reverse 5�-tcagagcgcttttgaagctaattcgGAATCGGTGTAGATCGGGCT-3�

viscBCR-Dn Forward 5�-aggaacttcaagatccccaattcgATTGCCGATGGTGGAAAAGC-3�

Reverse 5�-TCAAGCAAGCGGTACCGCGGTATCGGGGTGATGAAT-3�

pEX18Tc Forward 5�-CCTCTTCGCTATTACGCCAG-3�Reverse 5�-GTTGTGTGGAATTGTGAGCG-3�

Heterologousexpression

massAR Forward 5�-TG CTC CAG GGC GCT GTA GAG-3�Reverse 5�-CAT GCC GAG GGT GCA CAG-3�

Q-PCRviscA Forward 5�-GGACATCTGGCTCGA CCAA-3�

Reverse 5�-AGCCGCCGATGTTGTACAG-3�viscB Forward 5�-GCCGTCGCCCTGTACGT-3�

Reverse 5�-GTCTTTTTCCGCGAGATGCT-3�viscC Forward 5�-GGTTCTCCAAGGCGGTTTG-3�

Reverse 5�-GACCTTGGCGATGACTTTGC-3�rpoD Forward 5�-GCAGCTCTGTGTCCGTGATG-3�

Reverse 5�-TCTACTTCGTTGCCAGGGAATT-3�viscAR Forward 5�-AACGAAATCACACAGCGCC-3�

Reverse 5�-CCTCGCACCAGTGCAACC-3�viscBCR Forward 5�-CGCAGGAGCGCAGCAT-3�

Reverse 5�-CCATCGGCAATAGCAACGT-3�

a The 5� ends of the Up reverse and Dn forward primers for site-directedmutagenesis contain a 25-bp sequence (lowercase letters) complementary to theFRT-F and FRT-R primers for overlap extension in the second-round PCR. The5� ends of the Up forward and Dn reverse primers contain a restriction site(underlined) for BamHI, which is required for cloning into pEX18Tc.

4754 DE BRUIJN AND RAAIJMAKERS APPL. ENVIRON. MICROBIOL.


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

medium supplemented with gentamicin (25 �g/ml) and 5% sucrose to select forrecombinants that do not carry an integrated sacB gene as a result of a doublecrossover. Colonies were restreaked onto LB medium supplemented with gen-tamicin and 5% sucrose and onto LB medium supplemented with tetracycline(25 �g/ml). Colonies that did grow on LB medium with gentamicin but not on LBmedium with tetracycline were selected and subjected to colony PCR to confirmthe presence of the gentamicin resistance cassette and the absence of the tetra-cycline resistance cassette. Positive colonies were confirmed by sequencing thePCR fragments obtained with the Up forward and Dn reverse primers. Theobtained �viscAR and �viscBCR mutants were tested for viscosin production bya drop collapse assay and reverse-phase high-performance liquid chromatogra-phy (RP-HPLC), as described in the work of de Bruijn et al. (11).

Construction of pME6031-based vectors for heterologous expression. Gener-ation of the pME6031-massA and pME6031-massBC constructs is described inthe work of de Bruijn et al. (11). The pME6031-massAR construct (previouslyreferred to as pME6031-luxR-mA) was generated as described previously (13).The correct pME6031-massA, pME6031-massBC, and pME6031-massAR con-structs were subsequently electroporated into strain SBW25 and the �viscA,�viscB, �viscC, �viscAR, and �viscBCR mutants. Electrocompetent cells wereobtained by washing the cells three times with 1 mM MOPS (morpholinepro-panesulfonic acid) and 15% glycerol from a 5-ml overnight culture and finallydissolving the cells in 100 �l of the washing buffer. Electroporation occurred asdescribed above, and selection of transformants was performed on KB mediumsupplemented with tetracycline (25 �g/ml). Verification of transformation wasperformed by PCR analysis using one primer specific for the insert and oneprimer specific for the pME6031 vector. Viscosin and massetolide A productionin the transformed SBW25 strain and mutants was tested with a drop collapseassay, followed by RP-HPLC and liquid chromatography-tandem mass spectrom-etry (LC-MS/MS) analyses, as described previously (11).

Surface tension measurements and transcriptional analysis. Cells were grownin a 24-well plate, with 1.25 ml KB broth per well, and shaken at 220 rpm at 25°C.At specific time points, growth was determined by measuring 100 �l in a 96-wellplate in a Bio-Rad 680 microplate reader at 600 nm. From each culture, 1 ml wascollected and spun down. The cells were frozen in liquid N2 and stored at �80°C.For the RNA isolations and cDNA synthesis, four biological replicates were usedfor each time point. Biosurfactant production was measured qualitatively by thedrop collapse assay and quantitatively by tensiometric analysis of the cell-freesupernatant (K6 tensiometer; Kruss GmbH, Hamburg, Germany) at room tem-perature. To get the sufficient volume for the tensiometric analysis, the super-natants of four biological replicates were collected and pooled for each timepoint. The surface tension of each sample was measured in triplicate. RNAisolation and quantitative PCR (Q-PCR) analysis were performed, as describedpreviously (11). The threshold cycle (CT) value for viscA was corrected for thatof the housekeeping gene, as follows: �CT � CT (viscA) � CT (rpoD). The sameformula was used for the other genes. The relative quantification (RQ) valueswere calculated with the formula RQ � 2�[�CT (mutant) � �CT (wild type)]. Theprimers used for the Q-PCR are described in Table 1. Q-PCR analysis wasperformed in duplicate (technical replicates) on four independent RNA isola-tions (biological replicates). Statistically significant differences were determinedfor log-transformed RQ values by analysis of variance (P � 0.05), followed by theBonferroni post hoc multiple comparisons.

RESULTS AND DISCUSSION

Genetic analysis of LuxR-type regulators flanking CLP bio-synthesis genes. Analysis of gene clusters and genome se-quences of various Pseudomonas species and strains known to

FIG. 1. Schematic presentation of various CLP biosynthesis clusters in Pseudomonas species and strains. The CLP biosynthesis genes and theirflanking genes, including the LuxR-type transcriptional regulators (open arrows), are indicated. SBW25, viscosin-producing P. fluorescens strain;SS101, massetolide-producing P. fluorescens strain; Pf-5, orfamide-producing P. fluorescens strain; MIS38, arthrofactin-producing Pseudomonas sp.;PCL1445, putisolvin-producing P. putida strain; DC3000, syringafactin-producing P. syringae pv. tomato strain; B301D, syringomycin- andsyringopeptin-producing P. syringae pv. syringae strain. The presence of a LuxR-type transcriptional regulator downstream of the arthrofactinbiosynthesis cluster is not known due to a lack of sequence information.

VOL. 75, 2009 LuxR-TYPE REGULATORS IN CLP BIOSYNTHESIS 4755


http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

produce CLPs revealed that LuxR-type transcriptional regula-tors are positioned up- and downstream of the CLP biosynthe-sis genes (Fig. 1). Subsequent alignment and Pfam analysisshowed that the C-terminal regions of these LuxR-type regu-lators are relatively conserved and contain the typical DNA-binding HTH motif found in the entire family of LuxR-typetranscriptional regulators (see Fig. S1 in the supplemental ma-terial). However, they do not harbor the N-AHL-binding do-main found in LuxR of Vibrio fischeri (23, 38, 40) or the re-sponse regulator domain in FixJ, which activates the proteinupon phosphorylation (3) (see Fig. S1 in the supplemental

material). Phylogenetic analysis of these LuxR-type proteinsresulted in several distinct clusters, as follows: LuxR-typeregulators located upstream of the CLP biosynthesis genesclustered separately from the LuxR-type regulators locateddownstream of the CLP biosynthesis genes, except for theLuxR-type regulators SalA, SyrG, and SyrF from P. syringaepv. syringae, which were dispersed among the two clusters(Fig. 2). All LuxR-type regulators flanking the CLP biosynthe-sis genes clustered distantly from other well-known LuxR-typeregulators, including GerE from B. subtilis, FixJ from S. me-liloti, RhlR and LasR from Pseudomonas aeruginosa, and LuxR

FIG. 2. Phylogenetic analysis of the LuxR-type regulators flanking the Pseudomonas CLP biosynthesis genes (shaded in gray). Also includedin the analysis are other LuxR-type regulators, including LuxR of Vibrio fischeri (GenBank accession no. AAQ90196), LasR and RhlR ofPseudomonas aeruginosa (GenBank accession no. BAA06489 and NP_252167), PpuR of Pseudomonas putida (GenBank accession no. AAZ80478),PhzR of Pseudomonas chlororaphis (GenBank accession no. ABR21211), the autonomous effector domain protein GerE of Bacillus subtilis(GenBank accession no. NP_390719), the phosphorylation-activated response regulator FixJ of Sinorhizobium meliloti (GenBank accession no.NP_435915), NarL of Escherichia coli (GenBank accession no. CAA33023), UhpA of Salmonella enterica serovar Typhimurium (GenBankaccession no. NP_462689), and the ligand-binding MalT of E. coli (GenBank accession no. AAA83888). The numbers at the nodes indicate thelevel of bootstrap support, based on neighbor joining using 1,000 resampled data sets. The bar indicates the relative number of substitutions persite.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

from V. fischeri (Fig. 2), suggesting that they can be classified asa separate subfamily of LuxR-type regulators.

Functional analysis of LuxR-type regulatory genes in visco-sin biosynthesis. The CLP viscosin is produced by plant patho-genic and beneficial P. fluorescens strains (5, 10, 12). In theplant growth-promoting strain SBW25, viscosin biosynthesis isgoverned by the NRPS genes viscA, viscB, and viscC (Fig. 1)(12). In contrast to the organization of most other CLP geneclusters described to date, the visc genes are not geneticallylinked (Fig. 1) (12), with viscA located approximately 1.62 Mbfrom viscBC on a linear map of the SBW25 genome (12). Inspite of this unusual organization, viscosin biosynthesis instrain SBW25 does obey the colinearity rule of nonribosomalpeptide biosynthesis (12). Sequence analysis showed that theviscABC genes are also flanked by two LuxR-type regulatorgenes, one located upstream of viscA, designated viscAR, andone downstream of viscBC, designated viscBCR (Fig. 1).Genomic inactivation by site-directed mutagenesis of viscARor viscBCR resulted in viscosin deficiency, as determined bydrop collapse assays (data not shown) and RP-HPLC analyses(Fig. 3A and B). Transcriptional analysis by Q-PCR furthershowed that a mutation in either viscAR or viscBCR stronglyreduced transcript levels of viscA, viscB, and viscC (Fig. 3C).These results confirm and extend the results obtained previ-ously for P. syringae (2, 27, 30) and P. putida (15). In P. syringaepv. tomato DC3000, however, only the LuxR-type regulatorSyfR upstream of the syringafactin genes appeared to be es-sential, whereas a mutation of the LuxR-type regulator locateddownstream did not affect syringafactin biosynthesis (2). In P.putida PCL1445, only the LuxR-type regulator psoR, locatedupstream of the putisolvin genes, was investigated for its role inputisolvin production (15), but the function of the other LuxR-type regulator downstream of the putisolvin genes has, to ourknowledge, not been resolved yet. Q-PCR analyses furthershowed that a mutation in either viscAR or viscBCR did notsubstantially (change of �2.5-fold) affect transcription of theother regulator (Fig. 3D), suggesting that they do not tran-scriptionally regulate each other. In P. syringae pv. syringae, theLuxR-type regulators SalA and SyrF were shown to be essen-tial for syringomycin/syringopeptin biosynthesis (27, 30, 31).Subsequent gel shift analysis revealed that SalA binds to thepromoter region of SyrF, whereas SyrF binds to the syr-syp boxlocated around the �35 region of the syr-syp cluster (44, 45).These results elegantly showed that the control of expressionof the syr-syp genes by SalA is mediated through SyrF. Wang

FIG. 3. RP-HPLC analyses of extracts obtained from cell-free cul-ture supernatants of wild-type Pseudomonas fluorescens SBW25(A) and the �viscAR or �viscBCR mutant of strain SBW25 (B). Ex-tracts of each of the mutants were obtained and analyzed separately;the RP-HPLC chromatogram shown here is representative for each of

the mutants. The main peak, with a retention time of approximately 26min, is viscosin; the other peaks, with retention times between 16 and24 min, are structural derivatives of viscosin. (C) Transcript levels ofviscA, viscB, and viscC in cells of P. fluorescens SBW25, and the �vis-cAR and �viscBCR mutants obtained from mid-exponential growthphase. (D) Transcript levels of viscAR and viscBCR in the �viscAR and�viscBCR mutants in mid-exponential growth phase. The transcriptlevels of each of the genes was corrected for the transcript level of thehousekeeping gene rpoD [�CT � CT (gene x) � CT (rpoD)] andpresented relative to the transcript levels in wild-type SBW25 (logRQ), with RQ equaling 2�[�CT (mutant) � �CT (wild type)]. Mean values offour biological replicates are given; error bars represent the standarderrors of the means. Asterisks indicate statistically significant (P �0.05) differences relative to wild-type SBW25.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

et al. (44) further demonstrated that SalA and SyrF formhomodimers in vitro (44), a process that has been shown to becritical for DNA binding by other various transcriptional reg-ulators (17, 33, 46). Whether ViscAR and ViscBCR specificallybind to the promoter regions of the viscA and/or viscBC bio-synthesis genes was not addressed in this study. Also, the roleof dimerization of these LuxR-type regulators in the activationof the viscABC genes remains to be investigated.

Heterologous expression of LuxR-type transcriptional reg-ulator. Like the physical organization of the viscABC genes instrain SBW25, the massetolide biosynthesis genes massABCare not genetically linked in P. fluorescens strain SS101 (Fig. 1)(11). Blastp and phylogenetic analyses showed that ViscARand ViscBCR from strain SBW25 are most similar (82% and81% amino acid identity, respectively) to MassAR and Mass-BCR, the two transcriptional regulators in strain SS101 locatedupstream of massA and downstream of massBC, respectively(Fig. 1 and 2; Table 2). In spite of these similarities, however,the growth characteristics of strains SBW25 and SS101 and,concomitantly, the dynamics of viscosin and massetolide pro-duction differ, with a significantly higher growth rate for strainSBW25 (Fig. 4A). Tensiometric analysis of the cell-free culturesupernatants, which is indicative for viscosin and massetolideproduction (11, 12), further showed that viscosin is alreadyproduced by strain SBW25 after 8 h of growth, whereas mas-setolide production by strain SS101 is detectable after 12 h ofgrowth (Fig. 4B). To determine if a LuxR-type regulator fromstrain SS101 can direct viscosin biosynthesis in strain SBW25,the massAR gene of strain SS101 was introduced into the�viscAR mutant of SBW25. The results showed that massARdid not affect growth of the �viscAR mutant of SBW25 (Fig.4C) and restored the viscosin production level to the wild-typelevel, as confirmed by tensiometric and RP-HPLC analyses ofthe cell-free supernatants (Fig. 4D and E). Moreover, Q-PCRanalysis also revealed that the transcript levels of viscA, viscB,and viscC were restored to wild-type levels in the massAR-transformed �viscAR mutant of strain SBW25 (Fig. 4F). Nocomplementation of viscosin production was observed for theempty vector control (data not shown). These results indicatethat LuxR-type transcriptional regulators can be exchanged

among different Pseudomonas strains, thereby regulating the bio-synthesis of structurally different CLPs. However, transforma-tion of the �viscBCR mutant of strain SBW25 with the massARgene from strain SS101 did not restore viscosin production (seeFig. S2 in the supplemental material), suggesting that yet un-known characteristics (e.g., specific residues or domains) areessential for successful heterologous expression of these LuxR-type regulatory genes.

Heterologous expression of CLP biosynthesis genes. Sinceviscosin (m/z 1126) produced by strain SBW25 and massetolide(m/z 1140) produced by strain SS101 differ only in the fourthamino acid of the peptide ring (11, 12), experiments wereconducted to determine (i) if the mass genes could comple-ment viscosin deficiency of the �visc mutants of strain SBW25and (ii) if a functional viscAR gene is required for successfulexpression of the mass genes. We expected that geneticcomplementation of the �viscA mutant of SBW25 with themassA gene from SS101 would restore viscosin production. Incontrast, genetic complementation of the �viscB mutant withthe massB gene, harboring the adenylation domain of thefourth amino acid with a signature sequence for isoleucineinstead of valine (see Table S1 in the supplemental material),should result in massetolide production instead of that of vis-cosin. The results showed that introduction of the massA ormassBC gene from strain SS101 did not affect growth of eachof the �visc mutants of strain SBW25 (data not shown). Sub-sequent RP-HPLC analyses of cell-free culture supernatantsshowed that introduction of the massA gene in the �viscAmutant of SBW25 restored viscosin production to the wild-typelevel (Fig. 5A and C). Introduction of the massBC genes in the�viscB mutant did not restore viscosin production but resultedin massetolide production, as predicted and confirmed by LC-MS/MS analysis (Fig. 5B and D). Introduction of the massBCgenes in the �viscC mutant or in wild-type strain SBW25 re-sulted in the production of both viscosin and massetolide (Fig.5E and F). Transformation of the �viscAR mutant of SBW25with massA or massBC did not restore viscosin production, nordid it lead to massetolide production, indicating that a func-tional viscAR gene is required for heterologous expression ofthe mass genes. These results also showed that large NRPS

TABLE 2. Percent amino acid identity of ViscAR and ViscBCR, the two LuxR-type transcriptional regulators of viscosin biosynthesis inPseudomonas fluorescens SBW25, with other LuxR-type regulators flanking the CLP biosynthesis genes in other

Pseudomonas species and strains

Species Strain CLP biosynthesis cluster(s) LuxR-type transcriptionalregulator

% Identity

ViscAR ViscBCR

P. fluorescens SBW25 Viscosin ViscAR 100 42ViscBCR 42 100

P. fluorescens SS101 Massetolide MassAR 82 41MassBCR 44 81

P. fluorescens Pf-5 Orfamide LuxR-PFL2143 63 41LuxR-PFL2150 36 63

Pseudomonas sp. MIS38 Arthrofactin ORF1-MIS38 62 32P. syringae pv. tomato DC3000 Syringafactins SyfR 56 37

LuxR-pspto2833 41 44P. putida PCL1445 Putisolvin PsoR 54 29

LuxR-PCL1445 40 52P. syringae pv. syringae B301D Syringomycin/syringopeptin SyrG 49 30

SyrF 48 30SyrA 27 38



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

genes encoding CLPs can be exchanged among differentPseudomonas strains, leading to the production of nonnativecompounds.

Concluding remarks. Over the past decade, the biosynthesisgenes of at least eight CLPs produced by Pseudomonas specieshave been identified and sequenced (2, 11, 12, 15, 22, 34). Incontrast, very little is known about the genetic network in-volved in regulation of CLP biosynthesis. For plant growth-promoting P. fluorescens strain SBW25, only the gacS gene has

been identified to date as a regulator of viscosin biosynthesis(12). This study established that viscosin production in strainSBW25 is also regulated by two LuxR-type genes, designatedviscAR and viscBCR. Phylogenetic and domain analysesshowed that the LuxR-type regulators flanking CLP biosynthe-sis genes in Pseudomonas, including ViscAR and ViscBCR,belong to a separate LuxR subfamily of proteins: they containthe typical HTH domain in the C-terminal region but lack theN-AHL-binding or response regulator domains found in other

FIG. 4. (A) Growth of Pseudomonas fluorescens strains SBW25 and SS101 in KB at 25°C; at each time point, cell density was measuredspectrophotometrically (600 nm), and mean values from four replicates are given; error bars represent the standard deviations of the means.(B) Surface tension of cell-free culture supernatant of strains SBW25 and SS101 given in legend to panel A. (C) Growth of Pseudomonas fluorescensSBW25, its �viscAR mutant, and the �viscAR mutant plus pME6031-massAR at 25°C; at each time point, cell density was measured spectropho-tometrically (600 nm), and mean values from four replicates are given; error bars represent the standard deviations of the means. (D) Surfacetension of cell-free culture supernatant of strain SBW25 and the different mutants given in legend to panel C. (E) RP-HPLC analysis of the�viscAR mutant plus pME6031-massAR. (F) Transcript levels of viscA, viscB, and viscC in cells of wild-type SBW25, the �viscAR mutant, and the�viscAR mutant plus pME6031-massAR obtained from mid-exponential growth phase. The transcript levels of each of the genes was correctedfor the transcript level of the housekeeping gene rpoD [�CT � CT (gene x) � CT (rpoD)] and presented relative to the transcript levels inwild-type SBW25 (log RQ), with RQ equaling 2�[�CT (mutant) � �CT (wild type)]. For each time point, mean values from four biological replicatesare given; error bars represent the standard errors of the means. Asterisks indicate statistically significant (P � 0.05) differences relative towild-type SBW25.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

LuxR-type regulators (3, 23, 38, 40). The results showed, forthe first time, that these LuxR-type transcriptional regulatorscan be exchanged among different Pseudomonas strains,thereby regulating the biosynthesis of structurally differentCLPs. Genetic engineering of NRPS and polyketide synthasebiosynthesis genes has received considerable interest in naturalproduct research, as it provides a means to generate novelderivatives with more specific or enhanced activities (4, 6, 29,42). For example, modification of genes involved in the bio-synthesis of daptomycin and carotenoids resulted in a library ofnovel derivatives and generated valuable information on struc-tural features required for antibacterial activity (32, 37). Workby Eppelmann et al. (19) showed that expression of the com-plete bacitracin biosynthesis cluster of the slow-growing Bacil-lus licheniformis in a surfactin-deficient (srfA) mutant of B.subtilis resulted in elevated levels of bacitracin compared tothose of the natural producer. In many cases, however, yieldsof nonnative or structurally novel NRPS/polyketide synthase

products are relatively low (6, 29, 42). Whether this is relatedto inefficient transcriptional regulation, poor communicationbetween the encoded NRPS, or limited efflux of the final prod-ucts remains to be addressed.

ACKNOWLEDGMENTS

We are grateful to Teris van Beek and Frank Claassen for theirassistance in RP-HPLC and LC-MS/MS. We thank Teresa Sweat(USDA, Corvallis, OR) and Dimitri Mavrodi (USDA, Pullman, WA)for providing the plasmids (pEX18Tc and pPS854-Gm), protocols, andadvice for the site-directed mutagenesis.

This work was funded by the Dutch Technology Foundation (STW),the applied science division of NWO.

REFERENCES

1. Andersen, J. B., B. Koch, T. H. Nielsen, D. Sorensen, M. Hansen, O. Nybroe,C. Christophersen, J. Sorensen, S. Molin, and M. Givskov. 2003. Surfacemotility in Pseudomonas sp. DSS73 is required for efficient biological con-tainment of the root-pathogenic microfungi Rhizoctonia solani and Pythiumultimum. Microbiology 149:37–46.

FIG. 5. RP-HPLC chromatograms of cell-free culture extracts of wild-type SBW25, SS101, and the �viscA plus pME6031-massA, �viscB pluspME6031-massBC, �viscC plus pME6031-massBC, and SBW25 plus pME6031-massBC strains. Wild-type strain SBW25 produces viscosin(retention time of approximately 27 to 28 min; m/z 1126) and other derivatives of viscosin (peaks with retention times ranging from 16 to 24 min).Wild-type strain SS101 produces massetolide A (retention time of approximately 36 to 37 min; m/z 1140) and various other derivatives ofmassetolide A (peaks with retention times ranging from 21 to 33 min), differing in amino acid composition of the peptide moiety (11).



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

2. Berti, A. D., N. J. Greve, Q. H. Christensen, and M. G. Thomas. 2007.Identification of a biosynthetic gene cluster and the six associated lipopep-tides involved in swarming motility of Pseudomonas syringae pv. tomatoDC3000. J. Bacteriol. 189:6312–6323.

3. Birck, C., M. Malfois, D. Svergun, and J. Samama. 2002. Insights into signaltransduction revealed by the low resolution structure of the FixJ responseregulator. J. Mol. Biol. 321:447–457.

4. Bode, H. B., and R. Muller. 2005. The impact of bacterial genomics onnatural product research. Angew. Chem. Int. Ed. Engl. 44:6828–6846.

5. Braun, P. G., P. D. Hildebrand, T. C. Ells, and D. Y. Kobayashi. 2001.Evidence and characterization of a gene cluster required for the productionof viscosin, a lipopeptide biosurfactant, by a strain of Pseudomonas fluore-scens. Can. J. Microbiol. 47:294–301.

6. Cha, M., N. Lee, M. Kim, M. Kim, and S. Lee. 2008. Heterologous produc-tion of Pseudomonas aeruginosa EMS1 biosurfactant in Pseudomonas putida.Bioresour. Technol. 99:2192–2199.

7. Reference deleted.8. Choi, K. H., A. Kumar, and H. P. Schweizer. 2006. A 10-min method for

preparation of highly electrocompetent Pseudomonas aeruginosa cells: ap-plication for DNA fragment transfer between chromosomes and plasmidtransformation. J. Microbiol. Methods 64:391–397.

9. Choi, K. H., and H. P. Schweizer. 2005. An improved method for rapidgeneration of unmarked Pseudomonas aeruginosa deletion mutants. BMCMicrobiol. 5:30.

10. Cui, X., R. Harling, P. Mutch, and D. Darling. 2005. Identification of N-3-hydroxyoctanoyl-homoserine lactone production in Pseudomonas fluorescens5064, pathogenic to broccoli, and controlling biosurfactant production byquorum sensing. Eur. J. Plant Pathol. 111:297–308.

11. de Bruijn, I., M. J. de Kock, P. de Waard, T. A. van Beek, and J. M.Raaijmakers. 2008. Massetolide A biosynthesis in Pseudomonas fluorescens.J. Bacteriol. 190:2777–2789.

12. de Bruijn, I., M. J. D. de Kock, M. Yang, P. de Waard, T. A. van Beek, andJ. M. Raaijmakers. 2007. Genome-based discovery, structure prediction andfunctional analysis of cyclic lipopeptide antibiotics in Pseudomonas species.Mol. Microbiol. 63:417–428.

13. de Bruijn, I., and J. M. Raaijmakers. 2009. Regulation of cyclic lipopeptidebiosynthesis in Pseudomonas fluorescens by the ClpP protease. J. Bacteriol.191:1910–1923.

14. Dubern, J.-F., E. L. Lagendijk, B. J. J. Lugtenberg, and G. V. Bloemberg.2005. The heat shock genes dnaK, dnaJ, and grpE are involved in regulationof putisolvin biosynthesis in Pseudomonas putida PCL1445. J. Bacteriol.187:5967–5976.

15. Dubern, J. F., E. R. Coppoolse, W. J. Stiekema, and G. V. Bloemberg. 2008.Genetic and functional characterization of the gene cluster directing thebiosynthesis of putisolvin I and II in Pseudomonas putida strain PCL1445.Microbiology 154:2070–2083.

16. Dubern, J. F., B. J. J. Lugtenberg, and G. V. Bloemberg. 2006. The ppuI-rsaL-ppuR quorum-sensing system regulates biofilm formation of Pseudomo-nas putida PCL1445 by controlling biosynthesis of the cyclic lipopeptidesputisolvins I and II. J. Bacteriol. 188:2898–2906.

17. Ducros, V. M., R. J. Lewis, C. S. Verma, E. J. Dodson, G. Leonard, J. P.Turkenburg, G. N. Murshudov, A. J. Wilkinson, and J. A. Brannigan. 2001.Crystal structure of GerE, the ultimate transcriptional regulator of sporeformation in Bacillus subtilis. J. Mol. Biol. 306:759–771.

18. Dumenyo, C. K., A. Mukherjee, W. Chun, and A. K. Chatterjee. 1998.Genetic and physiological evidence for the production of N-acyl homoserinelactones by Pseudomonas syringae pv. syringae and other fluorescent plantpathogenic Pseudomonas species. Eur. J. Plant Pathol. 104:569–582.

19. Eppelmann, K., S. Doekel, and M. A. Marahiel. 2001. Engineered biosyn-thesis of the peptide antibiotic bacitracin in the surrogate host Bacillussubtilis. J. Biol. Chem. 276:34824–34831.

20. Finking, R., and M. A. Marahiel. 2004. Biosynthesis of nonribosomal pep-tides. Annu. Rev. Microbiol. 58:453–488.

21. Fuqua, C., S. C. Winans, and E. P. Greenberg. 1996. Census and consensusin bacterial ecosystems: the LuxR-LuxI family of quorum-sensing transcrip-tional regulators. Annu. Rev. Microbiol. 50:727–751.

22. Gross, H., V. O. Stockwell, M. D. Henkels, B. Nowak-Thompson, J. E.Loper, and W. H. Gerwick. 2007. The genomisotopic approach: a system-atic method to isolate products of orphan biosynthetic gene clusters.Chem. Biol. 14:53–63.

23. Hanzelka, B. L., and E. P. Greenberg. 1995. Evidence that the N-terminalregion of the Vibrio fischeri LuxR protein constitutes an autoinducer-bindingdomain. J. Bacteriol. 177:815–817.

24. Hoang, T. T., R. R. Karkhoff-Schweizer, A. J. Kutchma, and H. P. Schweizer.1998. A broad-host-range Flp-FRT recombination system for site-specific

excision of chromosomally-located DNA sequences: application for isolationof unmarked Pseudomonas aeruginosa mutants. Gene 212:77–86.

25. Inoue, H., H. Nojima, and H. Okayama. 1990. High efficiency transformationof Escherichia coli with plasmids. Gene 96:23–28.

26. Kinscherf, T. G., and D. K. Willis. 1999. Swarming by Pseudomonas syringaeB728a requires gacS (lemA) and gacA but not the acyl-homoserine lactonebiosynthetic gene ahlI. J. Bacteriol. 181:4133–4136.

27. Kitten, T., T. G. Kinscherf, J. L. McEvoy, and D. K. Willis. 1998. A newlyidentified regulator is required for virulence and toxin production in Pseudo-monas syringae. Mol. Microbiol. 28:917–929.

28. Koch, B., T. H. Nielsen, D. Sorensen, J. B. Andersen, C. Christophersen, S.Molin, M. Givskov, J. Sorensen, and O. Nybroe. 2002. Lipopeptide produc-tion in Pseudomonas sp. strain DSS73 is regulated by components of sugarbeet seed exudate via the Gac two-component regulatory system. Appl.Environ. Microbiol. 68:4509–4516.

29. Lee, Y. K., B. D. Yoon, J. H. Yoon, S. G. Lee, J. J. Song, J. G. Kim, H. M. Oh,and H. S. Kim. 2007. Cloning of srfA operon from Bacillus subtilis C9 and itsexpression in E. coli. Appl. Microbiol. Biotechnol. 75:567–572.

30. Lu, S. E., B. K. Scholz-Schroeder, and D. C. Gross. 2002. Characterizationof the salA, syrF, and syrG regulatory genes located at the right border of thesyringomycin gene cluster of Pseudomonas syringae pv. syringae. Mol. Plant-Microbe Interact. 15:43–53.

31. Lu, S. E., N. Wang, J. Wang, Z. J. Chen, and D. C. Gross. 2005. Oligonu-cleotide microarray analysis of the salA regulon controlling phytotoxin pro-duction by Pseudomonas syringae pv. syringae. Mol. Plant-Microbe Interact.18:324–333.

32. Nguyen, K. T., D. Ritz, J. Q. Gu, D. Alexander, M. Chu, V. Miao, P. Brian,and R. H. Baltz. 2006. Combinatorial biosynthesis of novel antibiotics relatedto daptomycin. Proc. Natl. Acad. Sci. USA 103:17462–17467.

33. Pristovsek, P., K. Sengupta, F. Lohr, B. Schafer, M. W. von Trebra, H.Ruterjans, and F. Bernhard. 2003. Structural analysis of the DNA-bindingdomain of the Erwinia amylovora RcsB protein and its interaction with theRcsAB box. J. Biol. Chem. 278:17752–17759.

34. Raaijmakers, J. M., I. De Bruijn, and M. J. D. De Kock. 2006. Cycliclipopeptide production by plant-associated Pseudomonas spp.: diversity, ac-tivity, biosynthesis, and regulation. Mol. Plant-Microbe Interact. 19:699–710.

35. Rainey, P. B., and K. Rainey. 2003. Evolution of cooperation and conflict inexperimental bacterial populations. Nature 425:72–74.

36. Schlegel, A., A. Bohm, S. J. Lee, R. Peist, K. Decker, and W. Boos. 2002.Network regulation of the Escherichia coli maltose system. J. Mol. Microbiol.Biotechnol. 4:301–307.

37. Schmidt-Dannert, C., D. Umeno, and F. H. Arnold. 2000. Molecular breed-ing of carotenoid biosynthetic pathways. Nat. Biotechnol. 18:750–753.

38. Shadel, G. S., R. Young, and T. O. Baldwin. 1990. Use of regulated cell lysisin a lethal genetic selection in Escherichia coli: identification of the autoin-ducer-binding region of the LuxR protein from Vibrio fischeri ATCC 7744. J.Bacteriol. 172:3980–3987.

39. Sieber, S. A., and M. A. Marahiel. 2005. Molecular mechanisms underlyingnonribosomal peptide synthesis: approaches to new antibiotics. Chem. Rev.105:715–738.

40. Slock, J., D. VanRiet, D. Kolibachuk, and E. P. Greenberg. 1990. Criticalregions of the Vibrio fischeri luxR protein defined by mutational analysis. J.Bacteriol. 172:3974–3979.

41. Reference deleted.42. Stachelhaus, T., A. Schneider, and M. A. Marahiel. 1995. Rational design of

peptide antibiotics by targeted replacement of bacterial and fungal domains.Science 269:69–72.

43. Timms-Wilson, T. M., R. J. Ellis, A. Renwick, D. J. Rhodes, D. V. Mavrodi,D. M. Weller, L. S. Thomashow, and M. J. Bailey. 2000. Chromosomalinsertion of phenazine-1-carboxylic acid biosynthetic pathway enhances ef-ficacy of damping-off disease control by Pseudomonas fluorescens. Mol.Plant-Microbe Interact. 13:1293–1300.

44. Wang, N., S. E. Lu, A. R. Records, and D. C. Gross. 2006. Characterizationof the transcriptional activators SalA and SyrF, which are required forsyringomycin and syringopeptin production by Pseudomonas syringae pv.syringae. J. Bacteriol. 188:3290–3298.

45. Wang, N., S. E. Lu, Q. Yang, S. H. Sze, and D. C. Gross. 2006. Identificationof the syr-syp box in the promoter regions of genes dedicated to syringomycinand syringopeptin production by Pseudomonas syringae pv. syringae B301D. J.Bacteriol. 188:160–168.

46. Zhang, R. G., T. Pappas, J. L. Brace, P. C. Miller, T. Oulmassov, J. M.Molyneaux, J. C. Anderson, J. K. Bashkin, S. C. Winans, and A. Joachimiak.2002. Structure of a bacterial quorum-sensing transcription factor complexedwith pheromone and DNA. Nature 417:971–974.



http://aem.asm

.org/D

ownloaded from

http://aem.asm.org/

diversity and functional analysis of luxr-type transcriptional … · are regulators that are...

Documents