APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2010, p. 4675–4683 Vol. 76, No. 140099-2240/10/$12.00 doi:10.1128/AEM.00480-10Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Structural and Functional Characterization of Diffusible SignalFactor Family Quorum-Sensing Signals Produced by

Members of the Burkholderia cepacia Complex�

Yinyue Deng,1,2# Ji’en Wu,1# Leo Eberl,3 and Lian-Hui Zhang1,2*Institute of Molecular and Cell Biology, Proteos, 61 Biopolis Drive, Singapore 138673, Republic of Singapore1; Department of

Biological Sciences, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Republic of Singapore2;and Department of Microbiology, University of Zurich, Zollikerstrasse 107, CH-8008 Zurich, Switzerland3

Received 23 February 2010/Accepted 19 May 2010

Previous work has shown that Burkholderia cenocepacia produces the diffusible signal factor (DSF) familysignal cis-2-dodecenoic acid (C12:�2, also known as BDSF), which is involved in the regulation of virulence. Inthis study, we determined whether C12:�2 production is conserved in other members of the Burkholderia cepaciacomplex (Bcc) by using a combination of high-performance liquid chromatography, mass spectrometry, andbioassays. Our results show that five Bcc species are capable of producing C12:�2 as a sole DSF family signal,while four species produce not only C12:�2 but also a new DSF family signal, which was identified ascis,cis-11-methyldodeca-2,5-dienoic acid (11-Me-C12:�2,5). In addition, we demonstrate that the quorum-sens-ing signal cis-11-methyl-2-dodecenoic acid (11-Me-C12:�2), which was originally identified in Xanthomonascampestris supernatants, is produced by Burkholderia multivorans. It is shown that, similar to 11-Me-C12:�2 andC12:�2, the newly identified molecule 11-Me-C12:�2,5 is a potent signal in the regulation of biofilm formation,the production of virulence factors, and the morphological transition of Candida albicans. These data provideevidence that DSF family molecules are highly conserved bacterial cell-cell communication signals that playkey roles in the ecology of the organisms that produce them.

The Burkholderia cepacia complex (Bcc) comprises a groupof currently 17 formally named bacterial species that, althoughclosely related, are phenotypically diverse (17, 22, 23). Strainsof the Bcc are ubiquitously distributed in nature and have beenisolated from soil, water, the rhizosphere of plants, industrialsettings, hospital environments, and infected humans. SomeBcc strains have emerged as problematic opportunistic patho-gens in patients with cystic fibrosis or chronic granulomatousdisease, as well as in immunocompromised individuals (17).The clinical outcome of Bcc infections ranges from asymptom-atic carriage to a fulminant and fatal pneumonia, the so-called“cepacia syndrome” (12, 17). Although all Bcc species havebeen isolated from both environmental and clinical sources, B.cenocepacia and B. multivorans are most commonly found inclinical samples (16).

Many bacterial pathogens have evolved a cell-cell commu-nication mechanism known as quorum sensing (QS) to coor-dinate the expression of virulence genes. In spite of their ge-netic differences, most Bcc species produce N-acylhomoserinelactone (AHL) QS signals (25). More recently, another QSsignal molecule, cis-2-dodecenoic acid (BDSF), has been iden-tified in B. cenocepacia (3). Subsequent studies showed thatBDSF plays a role in the regulation of bacterial virulence (6,20). Interestingly, the two QS systems appear to act in con-junction in the regulation of B. cenocepacia virulence, as a set

of the AHL-controlled virulence genes are also positively reg-ulated by BDSF (6). Furthermore, mutation of Bcam0581,which is required for BDSF biosynthesis, results in substan-tially retarded energy production and impaired growth in min-imal medium (6), highlighting the dual roles of the QS systemin the physiology of and infection by B. cenocepacia.

BDSF is a structural analogue of cis-11-methyl-2-dodecenoicacid, which is a QS signal known as diffusible signal factor(DSF) originally identified from the plant bacterial pathogenXanthomonas campestris pv. campestris (2, 24). Evidence isaccumulating that DSF-type fatty acid signals represent a newfamily of QS signals, which are widespread among Gram-neg-ative bacteria (10, 24). For example, DSF and seven structuralderivatives were identified in supernatants of Stenotrophomo-nas maltophilia (8, 11), 12-methyl-tetradecanoic acid wasshown to be produced by Xylella fastidiosa (18), and cis-2-decenocic acid was found to be synthesized by Pseudomonasaeruginosa (5). In addition, DSF-like activity has also beenreported in a range of Xanthomonas species, including X.oryzae pv. oryzae and X. axonopodis pv. citri (1, 2, 4, 24), but thechemical structures of these DSF analogues remain to be de-termined. Unlike other known QS signals, such as AHL andAI-2 family signals, DSF and its analogues, including BDSF,are fatty acids and these fatty acid signals were collectivelydesignated DSF family signals for the convenience of discus-sion (10). Considering the fact that the list of DSF family signalis expanding, we propose to designate cis-11-methyl-2-dodece-noic acid (DSF) 11-Me-C12:�2 and cis-2-dodecenoic acid(BDSF) C12:�2. This nomenclature is based on one of the fattyacid nomenclatures (13, 19) where the methyl (Me) substitu-tion and its position are indicated first (for example, 11-Meindicates a methyl group on C-11 of the fatty acid carbon

* Corresponding author. Mailing address: Institute of Molecularand Cell Biology, 61 Biopolis Drive, Singapore 138673, Republic ofSingapore. Phone: 65-6872-7400. Fax: 65-6779-1117. E-mail: lianhui@imcb.a-star.edu.sg.

# Both authors contributed equally to the manuscript.� Published ahead of print on 28 May 2010.

4675

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

chain), followed by the length of the fatty acid carbon chain(C12 represents a 12-carbon fatty acid chain), and then theposition of the double bond in the fatty acid chain (�2 indicatesa double bond in the cis configuration at site 2, i.e., betweenC-2 and C-3 of the fatty acid carbon chain). In this way, it isconvenient to say that 11-Me-C12:�2 and C12:�2 have iden-tical 12-carbon fatty acid chains with a cis bond at the samesite but differ in a methyl substitution on C-11. Followingthis nomenclature system, 12-methyl-tetradecanoic acid andcis-2-decenocic acid can be referred to as 12-Me-C14 andC10:�2, respectively.

DSF family signals have emerged as important factors in theregulation of virulence and biofilm formation in a wide rangeof bacterial pathogens (10). In this study, we have investigatedthe production of the DSF family signals in nine Bcc species. Itis demonstrated that C12:�2 is conserved in members of theBcc and that 11-Me-C12:�2 and a novel DSF family signal werealso produced by some, but not all, of the Bcc strains investi-gated. This new signal was identified as cis,cis-11-methyldo-deca-2,5-dienoic acid (11-Me-C12:�2,5) by nuclear magneticresonance (NMR) analysis and mass spectrometry. We havealso investigated the biological significance of 11-Me-C12:�2,5

in intraspecies and interspecies communication.

MATERIALS AND METHODS

Bacterial strains and growth conditions. The Bcc strains used in this work arelisted in Table 1. These strains were grown at 28°C or 37°C, as indicated, withshaking at 250 rpm in Luria-Bertani (LB) broth. X. campestris pv. campestrisstrain 8004 and its rpfF deletion mutant 8004dF were described previously (9,

24). X. campestris pv. campestris strains were maintained at 30°C in YEB medium(26). B. cenocepacia J2315 and its Bcam0581 deletion mutant d0581 were de-scribed previously (3). For the cultivation of static biofilms of B. cenocepacia,bacteria were grown at 30°C in basal salt medium (pH 7.2) containing 20 mMcitrate and 0.5% Casamino Acids (3). The following antibiotics were added assupplements when necessary: rifampin, 50 �g ml�1; gentamicin, 100 �g ml�1;tetracycline, 10 �g ml�1; trimethoprim, 400 �g ml�1 (B. cenocepacia) or 1.5 mgml�1 (Escherichia coli). Candida albicans SC5314 was grown in GMM mediumconsisting of 6.7 g of Bacto yeast nitrogen base (Difco, Sparks, MD) and 0.2%glucose (pH 7.2) (3). The DSF family signal molecules were added to themedium at a final concentration of 5 �M unless indicated otherwise.

Thin-layer chromatography (TLC) and DSF bioassay analysis. Overnightbacterial culture supernatants (250 ml) were extracted with acidified ethyl ace-tate at a 1:1 ratio. The organic phase was dried using a rotary evaporator, and theresidues were dissolved with 200 �l of methanol. An aliquot of 5 �l extracts wasspotted onto a silica gel TLC plate (20 by 20 cm; Merck) and separated with ethylacetate-hexane (20:80, vol/vol) as running solvents. Subsequently, the plates weredried under airflow and overlaid with 100 ml of NYG medium (20 g glycerol, 5 gpeptone, and 3 g yeast extract per liter), which was supplemented with 0.8 gagarose, 250 �g of 5-bromo-4-chloro-3-indolyl-�-D-glucoside, and 4 ml of thebiosensor strain FE58 at an optical density at 600 nm (OD600) of 1.8 (24). TheTLC plate was incubated overnight at 28°C, and DSF activity was indicated bythe presence of a blue spot.

Purification and structural analysis of DSF family molecules. To isolate andidentify C12:�2 and its analogues from supernatants, 1-liter cultures of Bccstrains were grown to an OD600 of about 3.0 and centrifuged. The supernatantswere acidified to a pH of 4.0 with diluted HCl and extracted with ethyl acetate(1.0, vol/vol) twice. Following evaporation of the ethyl acetate, the residue wasdissolved in methanol, subjected to flash chromatography on normal-phase silicagel, and eluted consecutively with 2 bed volumes of hexane, 2 bed volumes of10% ethyl acetate in hexane, and 4 bed volumes of 25% ethyl acetate in hexane.The active fractions, which were detected using the DSF sensor FE58 (24), werecombined for high-performance liquid chromatography (HPLC) profiling anal-ysis on a reverse-phase column (Phenomenex Luna, 5 �m C18, 250 by 4.60 mm)and eluted with 80% methanol in H2O at a flow rate of 1 ml min�1. Peaks were

TABLE 1. Bacterial strains used in this study

Strain Source or characteristic(s) Growthtemp (°C) Source or reference(s)

B. lata 383 Soil, Trinidad 37 E. Mahenthiralingam laboratoryB. multivorans ATCC 17616 Soil, United States 37 E. Mahenthiralingam laboratoryB. cenocepacia J2315 Cystic fibrosis isolate, United Kingdom 37 ATCCB. stabilis LMG 14086 Respirator, United Kingdom 28 BCCMtma

B. vietnamiensis G4 Industrial waste treatment plant 37 E. Mahenthiralingam laboratoryB. dolosa LMG 18941 Cystic fibrosis isolate, United States 28 BCCMtmB. ambifaria AMMD

(LMG 19182)Rhizosphere, United States 37 E. Mahenthiralingam laboratory

B. anthina LMG 16670 Rhizosphere, United Kingdom 28 BCCMtmB. pyrrocinia LMG 14191 Soil, Japan 37 BCCMtmd0581 DSF-negative mutant derived from J2315 with Bcam0581 deleted 37 3d0581(0581) Mutant d0581 harboring expression construct pMLS7-Bcam0581 37 3d0581(5121) Mutant d0581 harboring expression construct pMLS7-Bmul5121 37 This studyJ2315(egfp) Wild type harboring expression construct pMLS7-egfp 30 This studyd0581(egfp) Mutant d0581 harboring expression construct pMLS7-egfp 30 This studyJ2315(PzmpA-lacZ) Wild type harboring reporter construct pMLS7PzmpA-lacZ 37 6J2315(pMSL7-lacZ) Wild type harboring promoterless fusion construct pMLS7-lacZ 37 6d0581(PzmpA-lacZ) Mutant d0581 harboring reporter construct pMLS7PzmpA-lacZ 37 6d0581(pMSL7-lacZ) Mutant d0581 harboring promoterless fusion construct pMLS7-lacZ 37 68004 Wild-type strain of X. campestris pv. campestris 30 98004dF DSF-negative mutant derived from 8004 with rpfF deleted 30 9DH5� E. coli 37 Laboratory collectionDH5�(0581) DH5� harboring expression construct pMSL7-Bcam0581 37 This studyDH5�(5121) DH5� harboring expression construct pMSL7-Bmul5121 37 This studyFE58 Mutant 8004dF containing construct pLAFR3-PengXCA-gusA, in

which the coding region of E. coli �-glucuronidase was placedunder the control of DSF-inducible promoter PengXCA; this is aspecific biosensor for DSF family signals

30 3, 24

C. albicans SC5314 Clinical isolate 30 Y. Wang laboratory

a BCCMtm, Belgian Coordinated Collections of Microorganisms.

4676 DENG ET AL. APPL. ENVIRON. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

FIG. 1. The presence of DSF synthase genes, as well as the production of DSF-like activity, is conserved in many Bcc species. The strains andthe corresponding GenBank accession numbers (where applicable) of the DSF synthases are as follows: B. lata 383 (B.lat), ABB12683; B.multivorans ATCC 17616 (B.mul), ABX18791; B. cenocepacia J2315 (B.cen), ABK10294; B. vietnamiensis G4 (B.vie), ABO57014; B. dolosaAUO158 (B.dol), EAY71442; B. ambifaria AMMD (B.amb), ABI90833; B. stabilis LMG 14086 (B.sta); B. anthina LMG 16670 (B.ant); B. pyrrociniaLMG 14191 (B.pyr). (A) Sequence alignment of Bcam0581 homologues of members of the Bcc. The different amino acid residues are indicatedby black shading, and the residues with similar physicochemical properties are shown by gray shading. (B) Genomic organization of Bcam0581homologues in various Bcc species. The gray-shaded arrows indicate the Bcam0581 homologues. The other filled arrows indicate the genes flankingthe Bcam0581 homologues in the different species. (C) TLC analysis of the DSF-like activity in the crude extracts of culture supernatants of variousBcc species using the DSF sensor strain FE58. Synthetic C12:�2 was included as a positive control.

4677

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

monitored with a UV detector (� � 210 and 254 nm). Fractions were collectedat 1-min interval and assayed using the DSF biosensor FE58.

The 1H, 13C, and heteronuclear multiple quantum coherence (HMQC) NMRspectra in CDCl3 solution were obtained using a Bruker DRX500 spectrometeroperating at 500 MHz for 1H or 125 MHz for 13C. High-resolution electrosprayionization mass spectrometry (ESI-MS) was performed on a Finnigan/MATMAT 95XL-T mass spectrometer under the conditions described previously (24).

Complementation of B. cenocepacia strain d0581 and heterologous expressionof Bcam0581 and Bmul5121 in E. coli. The coding region of Bmul5121 wasamplified from B. multivorans via PCR using primers BMUL5121-F (5�-TGCTCTAGAGCAATGCAGCTCCAATCACATCCC) and BMUL5121-R (5�-CCCAAGCTTGGGTCACACCGTGCGCAACTTC). The product was digested withXbaI and HindIII and ligated at the same enzyme sites under the control of theS7 ribosomal protein promoter in plasmid vector pMSL7 (15). After sequenceverification, the resulting construct, pMSL7-Bmul5121, was introduced into theDSF-negative mutant d0581 by triparental mating. Transconjugants were se-lected on LB agar plates supplemented with gentamicin and trimethoprim. Like-wise, plasmid pMSL7-Bcam0581 was used to complement mutant d0581. Plas-mids pMSL7-Bcam0581 and pMSL7-Bmul5121 were also used to heterologouslyexpress Bcam0581 and Bmul5121 in E. coli strain DH5�.

Extracellular polysaccharides and biofilm analysis. For quantification of ex-tracellular polysaccharide (EPS) production, 10-ml volumes of overnight YEBcultures at an OD600 of 3.0 were centrifuged at 12,000 rpm for 20 min. Thesupernatants were mixed with 2.5 volumes of absolute ethanol, and the mixturewas incubated at 4°C for 30 min. The precipitated EPS was isolated by centrif-ugation and dried overnight at 55°C before the determination of dry weights.

The formation of biofilms was investigated as follows. Cultures of the X.campestris pv. campestris wild-type strain and the DSF-negative mutant 8004dFwere grown overnight in 5 ml of YEB medium with or without a signal molecule.Methanol was used as a solvent control. After overnight incubation, bacterialcells were visualized by phase-contrast microscopy (Olympus BX50) and imageswere taken with an Olympus DP70 digital camera.

Analysis of static biofilms and measurement of �-galactosidase activity. Plas-mid pMLS7-egfp (15) was used to tag B. cenocepacia J2315 and its DSF-negativederivative d0581 with green fluorescent protein. Five-microliter samples of bac-terial cultures grown overnight to an OD600 of about 3.0 were inoculated induplicate into sterile six-well tissue culture plates containing 3 ml of basal saltmedium. The plates were incubated without agitation at 30°C for 3 days. Thebiofilms that formed at the air-liquid interface were analyzed by confocal scan-ning laser microscopy using a Carl Zeiss LSM510-Axiovert 100M confocal mi-croscope.

To test the effects of signal molecules on the expression of the zmpA gene,which encodes a zinc metalloprotease important for pathogenicity, a PzmpA-lacZgene fusion was employed as described previously (6) and the promoterlessfusion construct pMLS7-lacZ was used as the negative control (6). For measure-ment of �-galactosidase activity, bacterial cells were grown in LB medium at37°C with shaking at 250 rpm. When required, signal molecules were added to afinal concentration of 5 �M as indicated. Bacterial cells were harvested, and the�-galactosidase activities were assayed as described previously (14).

Microscopic analysis and quantification of germ tube formation in C. albicans.To test the effect on C. albicans germ tube formation, an overnight culture of C.albicans strain SC5314 grown in GMM medium was diluted 20-fold in freshGMM medium (3). Signal molecules were then added separately as indicated,and the cells were induced for 3 h at 37°C. Visualization and quantification ofgerm tube formation were performed using a phase-contrast microscope (Olym-pus BX50) by counting about 400 fungal cells per sample. Microphotography wasdone with an Olympus DP70 digital camera.

RESULTS

DSF-like signal molecules are produced by many Bccstrains. Our previous study has shown that BCAM0581 of B.cenocepacia J2315 is the enzyme responsible for the synthesisof C12:�2, which shares about 37% identical amino acids withthe DSF synthase RpfF of X. campestris pv. campestris (3). ABLAST search revealed that BCAM0581 is conserved in B.lata, B. multivorans, B. cenocepacia, B. vietnamiensis, B. dolosa,and B. ambifaria. Sequence alignment of the BCAM0581 ho-mologues showed that they are highly conserved, with morethan 94% amino acid identity (Fig. 1A). A synteny analysis

FIG. 2. Characterization of the DSF family signal molecules pro-duced by various Bcc strains. (A) DSF-like molecules produced bymembers of the Bcc. The strain abbreviations are defined in the legendto Fig. 1. The relative amounts of signal molecules were calculated onthe basis of their peak areas. For convenient comparison, peak b(C12:�2) of B. multivorans was arbitrarily defined as 100% and used tonormalize the signal ratios of the different species. (B) HPLC analysisof the B. multivorans supernatant extracts after flash chromatography.(C) Production of C12:�2 in DSF-negative mutant d0581 by in transexpression of Bcam0581 from B. cenocepacia and its homologueBmul5121 from B. multivorans. (D) Production of 11-Me-C12:�2 andC12:�2 in E. coli DH5� by heterologous expression of Bcam0581 fromB. cenocepacia and its homologue Bmul5121 from B. multivorans.

4678 DENG ET AL. APPL. ENVIRON. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

showed that the downstream region is highly conserved in allsix strains while the upstream region is variable, with only oneconserved gene, Bcam0582 (Fig. 1B). Sequence analyses of theconserved neighboring genes suggested that Bcam0580 en-codes a PAS-GGDEF-EAL multidomain fusion protein,Bcam0578 encodes a putative 5-oxoprolinase, and Bcam0582encodes a transglutaminase. These orthologous genes showabout 80 to 91% amino acid identity.

To test whether the strains are capable of producing C12:�2,we analyzed the crude solvent extracts of nine Bcc strains byTLC. After separation, the DSF-like signals were visualized byoverlaying the TLC plate with the DSF sensor strain FE58,which contains a GUS gene under the control of a DSF-induc-ible promoter (24). A blue spot with an Rf value similar to thatof C12:�2 was detected in all nine strains, suggesting that theyproduce a DSF-like signal(s) (Fig. 1C).

Purification and structural analysis of DSF-like signal mol-ecules. For purification, the solvent extracts of the culturesupernatants were first subjected to flash chromatography. Theactive fractions, as identified by aid of the DSF sensor strainFE58, were then combined for reverse-phase HPLC analysis.Bioactive peaks with a retention time indicative of C12:�2 (3)were observed with all of the strains investigated (Fig. 2A).Intriguingly, additional peaks with DSF-like activity were ob-served with B. multivorans, B. stabilis, B. anthina, and B. pyr-rocinia (Fig. 2A). The structures of these DSF-like moleculeswere identified as exemplified for B. multivorans, which pro-duces three peaks with DSF-like activity (Fig. 2B). Active frac-tions were collected and analyzed by ESI-MS. Fractions a, b,and c showed peaks at m/z values of 211.27, 197.27, and 209.27,respectively (Fig. 3A to C). These m/z values are in agreementwith the molecular formulas C13H23O2, C12H21O2, andC13H21O2, respectively. The compounds in fractions a and bwere unambiguously characterized as cis-11-methyl-2-decenoicacid (11-Me-C12:�2) and cis-2-dodecenoic acid (C12:�2) byNMR analysis (Fig. 3D; Table 2). These molecules were orig-inally identified in culture supernatants of X. campestris and B.cenocepacia (3, 24).

In the case of fraction c, the 1H spectrum indicated that

FIG. 3. ESI-MS analysis of purified fractions from B. multivorans.(A) ESI-MS spectrum of 11-Me-C12:�2 (peak a in Fig. 2B).(B) ESI-MS spectrum of C12:�2 (peak b in Fig. 2B). (C) ESI-MSspectrum of 11-Me-C12:�2,5 (peak c in Fig. 2B). (D) Predicted chem-ical structures of 11-Me-C12:�2,5, 11-Me-C12:�2, and C12:�2.

TABLE 2. NMR data for DSF family signals producedby Bcc members

Carbonno.a

Value (ppm) for:

11-Me-C12:�2 C12:�2 11-Me-C12:�2,5

1H 13C 1H 13C 1H 13C

1 172.2 172.0 169.92 5.82 119.1 5.81 119.0 5.81 118.33 6.38 153.6 6.38 153.7 6.30 151.14 2.69 29.0 2.69 29.0 3.45 27.75 1.44 29.2 1.46 29.3 5.38 125.36 1.20–1.35 29.3 1.20–1.35 29.4 5.47 132.57 1.20–1.35 29.5 1.20–1.35 29.6 2.06 27.48 1.20–1.35 29.8 1.20–1.35 29.7 1.35 29.89 1.20–1.35 27.4 1.20–1.35 29.2 1.27 27.010 1.25 39.0 1.20–1.35 33.8 1.17 38.911 1.55 28.0 1.20–1.35 22.7 1.52 27.912 0.86 22.6 0.89 14.1 0.86 22.613 0.86 22.6 0.86 22.6

a The carbon number is based on Fig. 3D.

VOL. 76, 2010 DSF FAMILY SIGNALS FROM B. CEPACIA COMPLEX MEMBERS 4679

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

there are two pairs of ethylenic protons (Table 2). The cou-pling constants between the protons in each pair are lower than11 Hz. This indicates that the two double bonds are both in thecis configuration. The two methylene protons at H 3.45 sug-gest that this methylene carbon is connected with the twodouble bonds. The overlapping signals of two doublet methylgroups at H 0.87 indicate a branched structure similar to thatof 11-Me-C12:�2 (24). 13C spectra revealed that one of thedouble bonds is conjugated with a carbolic acid (Table 2).Therefore, the second double bond in the molecule is at C-5(Table 2). Collectively, the 1H, 13C, and HMQC data estab-lished the structure of this novel DSF family member as cis,cis-11-methyldodeca-2,5-dienoic acid, which is structurally identi-cal to 11-Me-C12:�2, except for an extra double bond betweenC-5 and C-6 (Table 2). For consistency and convenience, thisnewly identified molecule was designated 11-Me-C12:�2,5. Us-ing a combination of HPLC analysis and bioassays, we showedthat 11-Me-C12:�2 was only produced by B. multivorans,whereas 11-Me-C12:�2,5 was detectably produced by thethree Bcc species B. stabilis, B. anthina, and B. pyrrocinia(Fig. 2A).

The difference in DSF signal spectrum in the Bcc is notrelated to the variation of Bcam0581 homologues. Quantitativeanalysis showed that the Bcc species investigated here differednot only in the quantity of but also in the variety of DSF-likemolecule species they produced (Fig. 2A). Given that the Bccspecies we investigated have similar but not identical BDSFsynthase genes (Fig. 1A), it was of importance to test whether

the ability to produce different DSF-like signals is related tovariations in the BDSF synthase genes. To this end, the BDSFsynthase gene Bcam0581 of B. cenocepacia and its homologueBmul5121 of B. multivorans were cloned and expressed in theBcam0581 deletion mutant d0581. As expected, overexpressionof Bcam0581 in d0581 rescued C12:�2 biosynthesis (Fig. 2C).Interestingly, in trans expression of Bmul5121 in d0581 also ledto the production of C12:�2 but not of 11-Me-C12:�2 or 11-Me-C12:�2,5 (Fig. 2C). However, heterologous expression ofboth Bcam0581 and Bmul5121 in E. coli DH5� resulted theproduction of 11-Me-C12:�2 and C12:�2 (Fig. 2D). Thesedata suggest that it is the genetic background of the Burk-holderia strains used rather than variation in the BDSFsynthase genes that affects the spectrum of DSF signal mol-ecules.

C12:�2 and 11-Me-C12:�2,5 are functional analogues of 11-Me-C12:�2. To evaluate the biological relevance of C12:�2 and11-Me-C12:�2,5, we tested whether they can serve as signalmolecules in interspecies communication. In the plant patho-gen X. campestris, 11-Me-C12:�2 is required as an antiaggrega-tion factor (7). While the DSF-producing wild-type strain 8004grows planktonically, with cells being well dispersed (Fig. 4A),the DSF-negative mutant 8004dF forms large aggregates (9)(Fig. 4B). Similar to the effect of 11-Me-C12:�2 (Fig. 4C),addition of 5 �M C12:�2 and 11-Me-C12:�2,5 completely dis-persed these cell aggregates (Fig. 4D and E).

We then quantitatively compared the biological activity ofC12:�2 and 11-Me-C12:�2,5 with that of their analogue on the

FIG. 4. Growth characteristics of X. campestris pv. campestris wild-type (WT) strain 8004 (A) and DSF-negative mutant 8004dF (B). Additionof 5 �M 11-Me-C12:�2 (C), C12:�2 (D), or 11-Me-C12:�2,5 (E) to cultures of 8004dF restored planktonic growth, as well as EPS production (F).The data shown are the means of three repeats, and error bars indicate standard deviations.

4680 DENG ET AL. APPL. ENVIRON. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

production of EPS in the DSF-negative X. campestris rpfFmutant 8004dF. These experiments showed that addition of 5�M 11-Me-C12:�2, C12:�2, and 11-Me-C12:�2,5 to cultures of8004dF increased their EPS production to 78%, 69%, and78.9% of the wild-type level, respectively (Fig. 4F). Theseresults establish C12:�2 and 11-Me-C12:�2,5 as an effectivesignal in bacterial interspecies communication and suggestthat the three signal molecules are functionally interchange-able.

11-Me-C12:�2,5 is a potent signal in bacterium-fungus in-terkingdom communication. Previous studies in our laboratoryhave shown that C12:�2 and 11-Me-C12:�2 are able to modu-late the morphological transition of C. albicans (3, 24). To testwhether the extra double bond in 11-Me-C12:�2,5 influences itspotency in bacterium-fungus communication, this newly iden-tified signal molecule was added to fresh fungal yeast cells. Weused 11-Me-C12:�2 and C12:�2 as positive controls along withmethanol as a solvent control. After incubation at 37°C for 3 h,the majority of the C. albicans cells in the solvent controlformed germ tubes (Fig. 5A), whereas the fungus grew mainlyin the form of yeast cells when the culture medium wasamended with 5 �M 11-Me-C12:�2,5, 11-Me-C12:�2, or C12:�2

(Fig. 5B to D). A quantitative analysis using serial dilutions ofthe signal molecules showed that C12:�2 was the most potentinhibitor of C. albicans germ tube formation, followed by 11-Me-C12:�2,5 and 11-Me-C12:�2 (Fig. 5E).

11-Me-C12:�2 and 11-Me-C12:�2,5 are functional analoguesof C12:�2 in the regulation of B. cenocepacia biofilm formationand virulence factor production. The fact that 11-Me-C12:�2

plays a key role in the negative regulation of cell aggregateformation by X. campestris (7, 9) encouraged us to examine therole of the DSF family signals in biofilm formation by B. ceno-cepacia. When grown statically, B. cenocepacia formed a thinlayer of pellicle-like biofilm at the liquid-air interface. Micro-scopic examination of the biofilms formed by the wild-typestrain B. cenocepacia J2315 revealed a smooth surface withonly a few small cell aggregates (Fig. 6A). However, the sur-face of the biofilms formed by the DSF-negative mutant wasuneven, with several large protrusions, as indicated by thelumpy x- and y-axis cross sections (Fig. 6B). The biofilm struc-ture was restored to that of the wild-type strain when themutant was grown in the presence of 11-Me-C12:�2, C12:�2, or11-Me-C12:�2,5 (Fig. 6C to E, respectively).

Our previous study has shown that inactivation of the DSFsynthase gene Bcam0581 in B. cenocepacia J2315 resulted indecreased expression of virulence genes and that this reductioncould be rescued by the addition of C12:�2 (6). To test whether11-Me-C12:�2,5 and 11-Me-C12:�2 are functional analogues ofC12:�2, cultures of the DSF-negative B. cenocepacia mutantd0581 harboring a fusion of the zmpA metalloprotease pro-moter with the lacZ reporter gene on a plasmid were supple-mented with DSF family signal molecules. Compared with thatin the wild type, the zmpA promoter activity was decreased by53% in mutant d0581 (Fig. 6F). This reduction was rescued notonly by the addition of C12:�2 but also by supplementation with11-Me-C12:�2,5 and 11-Me-C12:�2 at the same concentration(Fig. 6F).

FIG. 5. Effects of DSF family signals on C. albicans germ tube formation. C. albicans cells were grown at 37°C in the presence of(A) methanol (in the same amount as used as a solvent for DSF compounds) or 5 �M (B) 11-Me-C12:�2, (C) C12:�2, or (D) 11-Me-C12:�2,5.Microphotographs were taken 3 h after induction. (E) Measurement of the inhibitory activities of the DSF family signal molecules on germtube formation by C. albicans. The data shown are the means of two experiments, and at least 400 cells were counted per treatment. Errorbars show standard deviations.

VOL. 76, 2010 DSF FAMILY SIGNALS FROM B. CEPACIA COMPLEX MEMBERS 4681

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

DISCUSSION

The results of this study showed that C12:�2 is a conservedsignal molecule in the Bcc, with at least nine species producingC12:�2 as the major DSF family signal molecule (Fig. 2A).Whether the eight more recently described Bcc species (22, 23)also produce DSF-like molecules remains to be determined. Inaddition to its role in interspecies signal communication (3),C12:�2 has recently been shown to play a role in the regulationof B. cenocepacia virulence gene expression (6, 20). In addi-tion, evidence is emerging that B. cenocepacia appears to uti-lize both AHL- and DSF-dependent QS systems to coordinatethe expression of virulence factors (6). Interestingly, similar toC12:�2, the AHL-type QS signal N-octanoyl-L-homoserine isalso produced by many members of the Bcc (21, 25). Thesefindings, together with the results of this study, suggest that thetwo QS systems may have coevolved in the Bcc to concertedlymodulate bacterial physiology and virulence.

Surprisingly, while five of the Bcc species investigated,namely, B. lata, B. cenocepacia, B. vietnamiensis, B. dolosa, andB. ambifaria, only produced C12:�2, B. multivorans, B. stabilis,B. anthina, and B. pyrrocinia were shown to also synthesize11-Me-C12:�2,5 (Fig. 2A), which is a new member of the familyof DSF-like signal molecules. Moreover, B. multivorans wasfound to also produce 11-Me-C12:�2 (Fig. 2A and B), whichwas originally identified in supernatants of the plant pathogenX. campestris pv. campestris (10, 24). What may account for thestructural diversity of DSF family molecules in these bacterialspecies? The findings of this study are consistent with the

notion that the variation in the DSF synthases is not respon-sible for the differences in the product spectra. First, only theC12:�2 molecule was detected when the DSF synthase genesfrom B. cenocepacia or B. multivorans were overexpressed in aDSF-negative mutant of B. cenocepacia (Fig. 2C). Second,heterologous expression of Bcam0581, as well as Bmul5121, inE. coli DH5� resulted in the production of 11-Me-C12:�2 andC12:�2 (Fig. 2D). Importantly, a bioinformatic analysis of theB. multivorans genome sequence with the protein sequence ofthe DSF synthase Bmul5121 did not reveal the presence ofparalogues. Taken together, these data suggest that the ob-served variations in the production of DSF molecules is likelydue to the availability of different precursors in the differentBcc species.

Structural analysis characterized 11-Me-C12:�2,5 as cis,cis-11-methyldodeca-2,5-dienoic acid (Fig. 3; Table 2), which dif-fers from 11-Me-C12:�2 by an extra double bond in the cisconfiguration at the C-5–C-6 position (Fig. 3D). Functionalcharacterization of this newly identified signal moleculeshowed that 11-Me-C12:�2,5 is not only active in interspeciescommunication (Fig. 4 and 5) but also potent in the regulationof virulence gene expression and biofilm development in B.cenocepacia (Fig. 6). In agreement with the previous findingthat the methyl group substitution at C-11 contributes to thebiological activity in the regulation of virulence gene expres-sion (24), we found that 11-Me-C12:�2,5 was superior to C12:�2

in the induction of EPS production in X. campestris pv. campes-tris (Fig. 4F). The conserved distribution of DSF family signals

FIG. 6. Effects of exogenous addition of DSF family signal molecules on biofilm formation and virulence gene expression. Shown are surfaceimages of static biofilms of wild-type B. cenocepacia J2315 (A), DSF-negative mutant d0581 (B), d0581 supplemented with 11-Me-C12:�2 (C),d0581 supplemented with C12:�2 (D), and d0581 supplemented with 11-Me-C12:�2,5 (E). Scanning confocal microscopic images of the surface ofstatic biofilms were taken using a 40 objective. (F) Virulence gene expression was determined by measuring the activities of a PzmpA-lacZtranscriptional fusion in the B. cenocepacia d0581 background. The DSF family signal molecules were added separately at a final concentrationof 5 �M. The data shown are the means of three repeats, and error bars indicate standard deviations.

4682 DENG ET AL. APPL. ENVIRON. MICROBIOL.

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

in the Bcc provides further evidence that this group of relatedmolecules represents a novel class of widely conserved bacte-rial cell-cell communication signals (10).

ACKNOWLEDGMENT

The funding for this work was provided by the Biomedical ResearchCouncil, Agency of Science, Technology, and Research (A*Star), Sin-gapore.

REFERENCES

1. Andrade, M. O., M. C. Alegria, C. R. Guzzo, C. Docena, M. C. Rosa, C. H.Ramos, and C. S. Farah. 2006. The HD-GYP domain of RpfG mediates adirect linkage between the Rpf quorum-sensing pathway and a subset ofdiguanylate cyclase proteins in the phytopathogen Xanthomonas axonopodispv citri. Mol. Microbiol. 62:537–551.

2. Barber, C. E., J. L. Tang, J. X. Feng, M. Q. Pan, T. J. Wilson, H. Slater, J. M.Dow, P. Williams, and M. J. Daniels. 1997. A novel regulatory systemrequired for pathogenicity of Xanthomonas campestris is mediated by a smalldiffusible signal molecule. Mol. Microbiol. 24:555–566.

3. Boon, C., Y. Deng, L. H. Wang, Y. He, J. L. Xu, Y. Fan, S. Q. Pan, and L. H.Zhang. 2008. A novel DSF-like signal from Burkholderia cenocepacia inter-feres with Candida albicans morphological transition. ISME J. 2:27–36.

4. Chatterjee, S., and R. V. Sonti. 2002. rpfF mutants of Xanthomonas oryzae pv.oryzae are deficient for virulence and growth under low iron conditions. Mol.Plant Microbe Interact. 15:463–471.

5. Davies, D. G., and C. N. Marques. 2009. A fatty acid messenger is respon-sible for inducing dispersion in microbial biofilms. J. Bacteriol. 191:1393–1403.

6. Deng, Y., C. Boon, L. Eberl, and L. H. Zhang. 2009. Differential modulationof Burkholderia cenocepacia virulence and energy metabolism by quorumsensing signal BDSF and its synthase. J. Bacteriol. 191:7270–7278.

7. Dow, J. M., L. Crossman, K. Findlay, Y. Q. He, J. X. Feng, and J. L. Tang.2003. Biofilm dispersal in Xanthomonas campestris is controlled by cell-cellsignaling and is required for full virulence to plants. Proc. Natl. Acad. Sci.U. S. A. 100:10995–11000.

8. Fouhy, Y., K. Scanlon, K. Schouest, C. Spillane, L. Crossman, M. B. Avison,R. P. Ryan, and J. M. Dow. 2007. Diffusible signal factor-dependent cell-cellsignaling and virulence in the nosocomial pathogen Stenotrophomonas mal-tophilia. J. Bacteriol. 189:4964–4968.

9. He, Y. W., M. Xu, K. Lin, Y. J. Ng, C. M. Wen, L. H. Wang, Z. D. Liu, H. B.Zhang, Y. H. Dong, J. M. Dow, and L. H. Zhang. 2006. Genome scaleanalysis of diffusible signal factor regulon in Xanthomonas campestris pv.campestris: identification of novel cell-cell communication-dependent genesand functions. Mol. Microbiol. 59:610–622.

10. He, Y. W., and L. H. Zhang. 2008. Quorum sensing and virulence regulationin Xanthomonas campestris. FEMS. Microbiol. Rev. 32:842–857.

11. Huang, T. P., and A. C. Lee Wong. 2007. Extracellular fatty acids facilitateflagella-independent translocation by Stenotrophomonas maltophilia. Res.Microbiol. 158:702–711.

12. Isles, A., I. Maclusky, M. Corey, R. Gold, C. Prober, P. Fleming, and H.Levison. 1984. Pseudomonas cepacia infection in cystic fibrosis: an emergingproblem. J. Pediatr. 104:206–210.

13. IUPAC-IUB Commission on Biochemical Nomenclature. 1977. The nomen-clature of lipids: recommendations, 1976. Eur. J. Biochem. 79:11–21. http://www3.interscience.wiley.com/cgi-bin/fulltext/119625380/PDFSTART.

14. Jeffrey, H. M. 1992. A short course in bacterial genetics. Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.

15. Lefebre, M. D., and M. A. Valvano. 2002. Construction and evaluation ofplasmid vectors optimized for constitutive and regulated gene expression inBurkholderia cepacia complex isolates. Appl. Environ. Microbiol. 68:5956–5964.

16. Mahenthiralingam, E., A. Baldwin, and C. G. Dowson. 2008. Burkholderiacepacia complex bacteria: opportunistic pathogens with important naturalbiology. J. Appl. Microbiol. 104:1539–1551.

17. Mahenthiralingam, E., T. A. Urban, and J. B. Goldberg. 2005. The multi-farious, multireplicon Burkholderia cepacia complex. Nat. Rev. Microbiol.3:144–156.

18. Newman, K. L., R. P. Almeida, A. H. Purcell, and S. E. Lindow. 2004.Cell-cell signaling controls Xylella fastidiosa interactions with both insectsand plants. Proc. Natl. Acad. Sci. U. S. A. 101:1737–1742.

19. Rigaudy, J., and S. P. Klesney. 1979. Nomenclature of organic chemistry.Pergamon Press, Oxford, United Kingdom.

20. Ryan, R. P., Y. McCarthy, S. A. Watt, K. Niehaus, and J. M. Dow. 2009.Intraspecies signaling involving the diffusible signal factor BDSF (cis-2-do-decenoic acid) influences virulence in Burkholderia cenocepacia. J. Bacteriol.191:5013–5019.

21. Sokol, P. A., R. J. Malott, K. Riedel, and L. Eberl. 2007. Communicationsystems in the genus Burkholderia: global regulators and targets for novelantipathogenic drugs. Future Microbiol. 2:555–563.

22. Vanlaere, E., A. Baldwin, D. Gevers, D. Henry, E. De Brandt, J. J. Lipuma,E. Mahenthiralingam, D. P. Speert, C. Dowson, and P. Vandamme. 2009.Taxon K, a complex within the Burkholderia cepacia complex, comprises atleast two novel species, Burkholderia contaminans sp. nov. and Burkholderialata sp. nov. Int. J. Syst. Evol. Microbiol. 59:102–111.

23. Vanlaere, E., J. J. Lipuma, A. Baldwin, D. Henry, E. De Brandt, E. Ma-henthiralingam, D. Speert, C. Dowson, and P. Vandamme. 2008. Burkhold-eria latens sp. nov., Burkholderia diffusa sp. nov., Burkholderia arboris sp. nov.,Burkholderia seminalis sp. nov. and Burkholderia metallica sp. nov., novelspecies within the Burkholderia cepacia complex. Int. J. Syst. Evol. Microbiol.58:1580–1590.

24. Wang, L. H., Y. He, Y. Gao, J. E. Wu, Y. H. Dong, C. He, S. X. Wang, L. X.Weng, J. L. Xu, L. Tay, R. X. Fang, and L. H. Zhang. 2004. A bacterialcell-cell communication signal with cross-kingdom structural analogues.Mol. Microbiol. 51:903–912.

25. Wopperer, J., S. T. Cardona, B. Huber, C. A. Jacobi, M. A. Valvano, and L.Eberl. 2006. A quorum-quenching approach to investigate the conservationof quorum-sensing-regulated functions within the Burkholderia cepacia com-plex. Appl. Environ. Microbiol. 72:1579–1587.

26. Zhang, H. B., L. H. Wang, and L. H. Zhang. 2002. Genetic control ofquorum-sensing signal turnover in Agrobacterium tumefaciens. Proc. Natl.Acad. Sci. U. S. A. 99:4638–4643.

VOL. 76, 2010 DSF FAMILY SIGNALS FROM B. CEPACIA COMPLEX MEMBERS 4683

on Decem

ber 17, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from