1 23

Applied Microbiology andBiotechnology ISSN 0175-7598 Appl Microbiol BiotechnolDOI 10.1007/s00253-012-4454-9

Characterization of rhamnolipids producedby wild-type and engineered Burkholderiakururiensis

Luiz F. D. Tavares, Patrícia M. Silva,Magno Junqueira, DaniellyC. O. Mariano, Fábio C. S. Nogueira,Gilberto B. Domont, et al.

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag Berlin Heidelberg. This e-offprint is

for personal use only and shall not be self-

archived in electronic repositories. If you

wish to self-archive your work, please use the

accepted author’s version for posting to your

own website or your institution’s repository.

You may further deposit the accepted author’s

version on a funder’s repository at a funder’s

request, provided it is not made publicly

available until 12 months after publication.

BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING

Characterization of rhamnolipids produced by wild-typeand engineered Burkholderia kururiensis

Luiz F. D. Tavares & Patrícia M. Silva &

Magno Junqueira & Danielly C. O. Mariano &

Fábio C. S. Nogueira & Gilberto B. Domont &Denise M. G. Freire & Bianca C. Neves

Received: 3 September 2012 /Revised: 17 September 2012 /Accepted: 18 September 2012# Springer-Verlag Berlin Heidelberg 2012

Abstract Biosurfactants are a class of functional moleculesproduced and secreted by microorganisms, which play im-portant roles in cell physiology such as flagellum-dependentor -independent bacterial spreading, cell signaling, and bio-film formation. They are amphipathic compounds and com-prise a variety of chemical structures, including rhamnolipids,typically produced by Pseudomonas spp. and also reportedwithin other bacterial genera. The present study is focusedon Burkholderia kururiensis KP23T, a trichloroethylene(TCE)-degrading, N-fixing, and plant growth-promotingbacterium. Herein, we describe the production of rham-nolipids by B. kururiensis, and its characterization byLTQ-Orbitrap Hybrid Mass Spectrometry, a powerfultool that allowed efficient identification of molecular subpo-pulations, due to its high selectivity, mass accuracy, and

resolving power. The population of rhamnolipids producedby B. kururiensis revealed molecular species commonly ob-served in Pseudomonas spp. and/or Burkholderia spp. Inaddition, this strain was used as a platform for expression oftwo Pseudomonas aeruginosa biosynthetic enzymes: RhlA,which directly utilizes β-hydroxydecanoyl-ACP intermedi-ates in fatty acid synthesis to generate the HAA, andRhlB, the rhamnosyltransferase 1, which catalyzes thetransfer of dTDP-L-rhamnose to β-hydroxy fatty acidsin the biosynthesis of rhamnolipids. We show that rham-nolipid production by the engineered B. kururiensis wasincreased over 600 % when compared to the wild type.Structural analyses demonstrated a molecular populationcomposed mainly of monorhamnolipids, as opposed towild-type B. kururiensis and P. aeruginosa in whichdirhamnolipids are predominant. We conclude that B.kururiensis is a promising biosurfactant-producing or-ganism, with great potential for environmental and biotechno-logical applications due to its non-pathogenic characteristicsand efficiency as a platform for metabolic engineering andproduction of tailor-made biosurfactants.

Keywords Burkholderia kururiensis . Pseudomonasaeruginosa . Biosurfactants . Rhamnolipids . Heterologousexpression . LTQOrbitrap HybridMass Spectrometer

Introduction

Biosurfactants are a class of compounds produced by bac-teria and fungi, and play essential roles in cell physiologywhich include swarming motility of microorganisms, sig-naling, differentiation, and biofilm formation (Van Hammeet al. 2006). They are functional amphipathic molecules,

L. F. D. Tavares : P. M. Silva :D. C. O. Mariano :F. C. S. Nogueira :G. B. Domont :D. M. G. Freire : B. C. NevesDepartamento de Bioquímica, Instituto de Química, UniversidadeFederal do Rio de Janeiro (UFRJ),Rio de Janeiro, Brazil

G. B. DomontUnidade Proteômica,Rio de Janeiro, Brazil

M. JunqueiraDepartamento de Biologia Celular, Instituto de CiênciasBiológicas, Universidade de Brasília (UnB),Brasília, Brazil

B. C. Neves (*)Laboratório de Microbiologia Molecular e Proteômica, Instituto deQuímica, Departamento de Bioquímica, Universidade Federal doRio de Janeiro (UFRJ),Avenida Athos da Silveira Ramos, 149, Bloco A-Lab. 541,CEP: 21941-909, Cidade Universitária—Rio de Janeiro, RJ, Brazile-mail: bcneves@iq.ufrj.br

Appl Microbiol BiotechnolDOI 10.1007/s00253-012-4454-9

Author's personal copy

which comprise a variety of chemical structures such asglycolipids, lipopeptides, polysaccharide–protein com-plexes, phospholipids, fatty acids, and neutral lipids(Ward 2010).

Rhamnolipids are amongst the most extensively studiedbiosurfactants and are typically produced by Pseudomonasaeruginosa, being composed by one or two α-L-rhamnopyranosyl linked to one or two β-hydroxyalkanoyl-β-hydroxydecanoates (HAAs). More than 50 species of rham-nolipids have been identified to date within Pseudomonas andBurkholderia species together. The most abundant speciesproduced by P. aeruginosa are α-L-rhamnopyranosyl-α-L-rhamnopyranosyl-β-hydroxydecanoyl-β-hydroxydecanoate(Rha-Rha-C10-C10) and α-L-rhamnopyranosyl-α-L-rhamnopyranosyl-β-hydroxydecanoate (Rha-Rha-C10;Déziel et al. 1999; Abdel-Mawgoud et al. 2011).

Biosynthesis of rhamnolipids in P. aeruginosa involvesthree sequential reactions. The first reaction is catalyzed byRhlA, which employs β-hydroxy fatty acids as substrates,followed by rhamnosyltransferase 1 (RT1) that uses dTDP-L-rhamnose as its substrate, generating monorhamnolipids(Ochsner et al. 1994). Monorhamnolipids and dTDP-L-ramnose are substrates for rhamnosyltransferase 2 (RT2),yielding dirhamnolipids (Rahim et al. 2001). RT1 comprisesa membrane-associated protein, RhlB, while RT2 corre-sponds to a single-chain protein, RhlC (Rahim et al.2001). The first genes identified were the rhlAB operonand both genes are essential for rhamnolipid formation(Ochsner et al. 1994; Zhu and Rock 2008). The idea pre-sented in this study was that rhlAB gene products formed therhamnosyltransferase I complex involving an RhlAB heter-odimer. However, a more recent analysis revealed that whileboth rhlA and rhlB mutants fail to produce rhamnolipids,rhlA mutants did produce HAA, indicating that RhlA is alsoinvolved in the synthesis of the fatty acid dimer of rhamno-lipids and free HAAs (Déziel et al. 2003). The role of RhlBwas therefore assigned as the rhamnosyl transferase 1.Genes rhlA and rhlB are organized in a bicistronic operon,upstream of the previously identified rhlR regulatory gene(Pearson et al. 1997). P. aeruginosa contains two transcrip-tion regulators (LasR and RhlR) that, when complexed withtheir specific autoinducers (3-oxo-dodecanoyl-homoserinelactone and butanoyl-homoserine lactone, respectively) acti-vate transcription of different physiology- and virulence-associated traits (Fuqua et al. 2001; Reis et al. 2011; Witherset al. 2001). Those include the rhlAB operon whose expres-sion is promoted by the complex RhlR-butanoyl-homoserinelactone (Reis et al. 2011).

Rhamnolipids have also been demonstrated as extracel-lular products from Burkholderia, a genus closely related toPseudomonas (Toribio et al. 2010). Häubler et al. (1998)have characterized rhamnolipids produced by Burkholderiapseudomallei, a causative agent of a systemic infectious

disease of humans and animals. Rhamnolipids have alsobeen reported and characterized in Burkholderia plantarii(Hörmann et al. 2010; Howe et al. 2006). More recently,Dubeau et al. (2009) have identified mono- and dirhamno-lipid homologues from Burkholderia thailandensis.These studies demonstrate that the ability to producerhamnolipid seems to be a common feature withinBurkholderia spp. which are related to either animal orplant-associated infections.

Burkholderia kururiensis KP23T is a TCE-degradingbacterium isolated from an aquifer at a TCE-polluted sitein Japan (Zhang et al. 2000). Although the ability to fixnitrogen is not widespread amongst known Burkholderia,this strain has been identified as a diazotrophic bacterium(Estrada-De-Los-Santos et al. 2001). Burkholderia spp.have been associated to a wide geographic distribution,showing the ability to colonize diverse host plants at dis-tantly related environments (Elliott et al. 2007; Perin et al.2006). Mattos et al. (2008) reported the ability of B. kurur-iensis to endophytically colonize rice and promote plantgrowth. More recently, a B. kururiensis strain was isolatedfrom rhizosphere soil of tobacco plants and identified as anovel subspecies, distinct from the type strain KP23T

(Anandham et al. 2009). Altogether, those studies demon-strate a remarkable environmental adaptation of B. kurur-iensis, formerly described as “Burkholderia brasilensis”(Baldani et al. 1997) when first isolated from soil in Brazil.

Legal regulations concerning environmental impact as aresult of industrial activities strongly encourage the replace-ment of synthetic chemical surfactants by their biodegrad-able counterparts, such as rhamnolipids (Fraser 2000).Although presenting several advantages, the cost of micro-bial surfactants is prohibitive to upscale applications. Mucheffort has taken place in process optimization and engineer-ing in order to provide a lower cost of production of thesecompounds (Makkar and Cameotra 2002). In the presentstudy, the strategy adopted to improve biosurfactant produc-tion was the construction of a modified B. kururiensis strain,as an initial step to provide market competitiveness againstchemical surfactants. The fact that B. kururiensis KP23T hasnever been associated to any disease and is a BiosafetyLevel 1 microorganism makes it an attractive target forindustrial applications and metabolic engineering for im-proved rhamnolipid production. Expression of genes rhlAand rhlB from P. aeruginosa in B. kururiensis was expectedto provide a functional recombinant RhlA and RT1 to cata-lyze and enhance the synthesis of rhamnolipids.

We report herein the structural characterization and rela-tive abundance of rhamnolipids from wild-type and engi-neered B. kururiensis, accomplished by nanospray directinfusion mass spectrometric analysis using a linear quadru-pole ion trap and an Orbitrap mass analyzer (LTQ-OrbitrapHybrid Mass Spectrometry) through multi-stage mass

Appl Microbiol Biotechnol

Author's personal copy

spectrometric fragmentations of precursor and fragment ions(MS2) using collision-induced dissociation (CID) andhigher energy collision-induced dissociation (HCD) as com-plementary fragmentation methods. Because of the specific-ity of MS2 measurements and the ability to eliminateinterferences through the initial mass selection stage, qual-itative analysis of rhamnolipids could be accomplished.Accurate ion mass measurements were very important forstructural elucidation of the rhamnolipids, helping the dis-tinction between isobaric fragments and their assignment toindividual species. Considering the wide range of rhamno-lipid species described in the literature, LTQ-OrbitrapHybrid Mass Spectrometry (Hu et al. 2005) emerges as apowerful tool for providing improved, rapid identification,and structural characterization of novel biosurfactants.Interestingly, our study demonstrates, for the first time, theability of an N-fixing bacterium to produce biosurfactants.In addition, we demonstrate the feasibility of B. kururiensisKP23T as a heterologous platform for expression of P.aeruginosa genes by constructing an engineered strain withsubstantially enhanced rhamnolipid production levels.

Materials and methods

Bacteria and culture media

The strains used in this work are described in Table 1.Bacteria were stored in Luria-Bertani (LB) broth (Invitrogen)supplemented with 20 % glycerol (v/v) at −80 °C.Mineral salts employed for biosurfactant productionwere purchased from Merck (Darmstadt, Germany) andSigma-Aldrich (USA).

Molecular techniques

DNA manipulations were performed using standardtechniques as described by Sambrook and Russell(2001). Restriction endonucleases and T4 DNA poly-merase were purchased from Fermentas Life Sciences.T4 DNA ligase was purchased from New EnglandBiolabs. Plasmids were introduced into Escherichia colior B. kururiensis by electroporation using a Gene Pulser(Bio-Rad, Richmond, CA).

Cloning procedures required amplification of DNA frag-ments by polymerase chain reaction (PCR) and were per-formed with the proofreading DNA polymerase Pfu(Fermentas Life Sciences), and the clones were verified bysequencing. The synthetic oligonucleotides used as primersin the PCR reactions were obtained from Integrated DNATechnologies (IDT, USA) and are listed below. DNA anal-ysis was performed with DNA Star and programs availableat http://www.expasy.ch/.

Cloning of rhl genes from P. aeruginosa

PCR was used to generate a 2,233-bp DNA fragment con-taining the genes rhlA and rhlB, with primers RhlAPAO1-F(5′ CCGGAATTCATGCGGCGCGAAAGTCTGTTG 3′)a n d R h l B PA O 1 - R ( 5 ′ CCGGAATTCTCAGGACGCAGCCTTCAGCCA 3′) and genomic DNA fromwild-type P. aeruginosa strain PAO1 (ATCC BAA-47TM,Table 1) as DNA template. Cycling conditions were 95 °Cfor 1 min; 30 cycles of 95 °C for 1 min, 60 °C for 1 min, and72 °C for 8 min; and 72 °C for 6 min. The amplifiedfragment was inserted into pGEM-T vector (Promega).The cloned PCR product was digested with EcoRI andcloned into EcoRI-digested pTrc99A (Table 1). ThepTrc99A-based expression construct was transformed intowild-type B. kururiensis strain KP23T (ATCC 700977TM,Table 1) by electroporation.

Biosurfactant production

Bacteria were initially grown overnight in LB broth at 28 °Cfrom −80 °C stocks. Preliminary tests were conducted in LBbroth to access the initial potential of the engineered strain.Overnight cultures were inoculated into 1 L flasks contain-ing 300 mL of pre-production medium (K2HPO4 7.0 g/L,KH2PO4 3.0 g/L, NaNO3 1.0 g/L, MgSO4.7H2O 0.2 g/L,glycerol 30.0 g/L, yeast extract 5.0 g/L, Bacto-Peptone5.0 g/L; Santos et al. 2002) and incubated at 28 °C/170rpm on a rotary shaker for 40 h. Bacteria were then har-vested by centrifugation at 6,000×g for 25 min at 15 °C.Bacteria were inoculated at a final concentration of 1 g/Linto 1 L flasks containing 500 mL of a chemically opti-mized, mineral salt production (MSP) medium (K2HPO4

7.0 g/L, KH2PO4 3.0 g/L, NaNO3 1.4 g/L, MgSO4.7H2O0.2 g/L, glycerol 30.0 g/L; Santos et al. 2002), with amedium/flask volume ratio (MV/FV) of 0.5. The productionof biosurfactants was conducted for 5 days, at 28 °C andagitation at 170 rpm. For rhamnolipid purification, cultiva-tion in glycerol-supplemented MSP was conducted for10 days.

Growth monitoring

Bacterial cells were quantified by plating serially dilutedculture samples onto LB agar plates and counting thecolony-forming units (CFU).

Biosurfactant purification and characterization

Extraction and purification of biosurfactants

Culture supernatants were obtained under the conditionsdescribed above and served as starting material for the

Appl Microbiol Biotechnol

Author's personal copy

purification process. A culture in MSP medium supple-mented with glycerol was carried out for 10 days with thedifferent bacteria to be tested. The cultures were centrifuged at6,000×g for 25 min at 15 °C and cell-free supernatants wereused for following purification steps. Supernatants were acid-ified with 1.0 N HCl to pH 3.5 and processed for directextraction at a 1:3 ratio of culture supernatant and ethylacetate, respectively. Organic phases were collected and driedwith a rotary evaporator (HetoDrywinner Heto-Holten,Gydevank, Denmark). The resulting extracts were then dis-solved with methanol, lyophilized and stored at −20 °C.

Structural characterization of rhamnolipids by LTQ-OrbitrapHybrid Mass Spectrometry

Chemicals and reagents

All chemicals and reagents were obtained from Merck(Darmstadt, Germany) and were of analytical or highergrade.

Sample preparation

All samples were prepared in methanol–chloroform–2-prop-anol at a ratio of 1:2:4 (v/v/v) containing 7.5 mM ammoniumacetate and further centrifuged at 12,000×g for 5 min toallow precipitation of any impurity.

Mass spectrometric analysis

Biosurfactant samples were analyzed by direct infusionon an LTQ OrbitrapVelos (Thermo Fisher Scientific,Bremen, Germany) equipped with a nanoelectrosprayion source using nanoelectrospray capillaries (PicoTipEmitter, Glass Tip, coating 1P-4P, New Objective,MA, USA), with 2.0 kV ionization voltage and capillarytransfer temperature at 180 °C. Each data-dependentacquisition cycle consisted of a survey scan on theOrbitrap under a mass resolving power of 60,000(FWHM, at m/z 400) in the m/z range of 100–2,000 Da.

MS2 fragmentation of the most abundant precursor ions wasperformed by collision-induced dissociation (CID) in the lin-ear trap or in the higher energy collision-induced dissociation(HCD) cell. Precursor ions were analyzed within m/z of 400–800 Da and fragments were detected in the Orbitrap at a60.000 resolution. Normalized collision energies of 35 % inCID and 50 % in HCD were used for the MS2 fragmentations.

MS data processing

The calculation of chemical formulae from the m/z valuesand data processing of extracted ion current (EICs) werecarried out with Xcalibur 2.0 (Thermo Fisher Scientific Inc.Co.). High resolution EICs were obtained by processing thefull-scan MS data using potential rhamnolipid ions with15 ppm mass tolerance. The potential candidate ions wereobtained from isotopic mass calculation in ISOPRO v3.1.The relative abundances of rhamnolipid ions were assignedby their respective peak areas within the full-scan MS datawith Xcalibur 2.0.

Determination of surface tension properties

Determination of surface tension (ST, mN m−1) and criticalmicelle concentration (CMC) of the crude extract werecarried out on an Aqua-Pi tensiometer (Kibron Inc.,Helsinki) at 25 °C. Surface tension measurements werebased on the Du Noüy method, utilizing small probes. Allsample solutions were prepared with a solution composed of10 mM NaCl and 15 mM NaCO3. Prior to each experiment,the instrument was calibrated by measuring the surfacetension of pure water.

Quantification of total rhamnolipids

Considering that biosurfactants secreted by B. kururiensiswere rhamnolipids, the quantification of extracellular rham-nolipids was performed in triplicates by measuring the L-rhamnose contents with a colorimetric assay (orcinol reac-tion; Koch et al. 1991). Mass spectrometric analyses

Table 1 Bacterial strains and plasmids

Strain/plasmid Description Origin or reference

P. aeruginosa PAO1 Wild-type strain ATCC BAA-47TM; Stover et al. (2000)

B. kururiensis KP23T Type strain, isolated from an aquifer in Japan ATCC 700977TM; Zhang et al. (2000)

E. coli JM109 recA1 supE44 endA1 hsdR17 gyrA96relA1 thiΔ(lac-proAB) New England Biolabs

LMM21 B. kururiensis KP23T harboring pLMM12 This study

pGEM-T PCR cloning vector; AmpR Promega

pTrc99A Expression vector containing the tac promoter; AmpR Amann et al. (1988)

pLMM11 pGEM-T with full-length rhlA and rhlB genes from P. aeruginosa This study

pLMM12 pTrc99A with full-length rhlA and rhlB genes from P. aeruginosa This study

Appl Microbiol Biotechnol

Author's personal copy

indicated the major molecular populations, allowing deter-mination of a conversion factor (CF) for each rhamnolipid-producing microorganism. Each CF was determined by themass ratio between the sugar moieties and the full rhamno-lipid molecules, based on the most representative ions ineither rhamnolipid-producing microorganism. CF valueswere 2.00, 2.69, and 2.31 for the rhamnose equivalentsproduced by B. kururiensis KP23T, B. kururiensis LMM21,and P. aeruginosa PAO1, respectively.

Results

Structural characterization of a biosurfactant producedby B. kururiensis with ESI-LTQ Orbitrap Hybrid MS

The secretion of surface-active compounds by B. kurur-iensis was strongly suggested by its ability to spread onsemi-solid media (data not shown), although the natureof the putative surfactant still remained unknown.Therefore, to elucidate the chemical structure of theunknown surfactant, cultivation in glycerol-containingMSP medium was carried out and the cell-free superna-tant was processed for preparation of biosurfactantextracts. The resulting ethyl acetate extract appeared asa yellowish viscous paste.

Analysis of B. kururiensis extracts resulted in the identi-fication of the surface-active compounds as rhamnolipids.The structural data lead to identification of 23 congeners andelucidated the general population of rhamnolipids producedby B. kururiensis strain KP23T. Table 2 shows the congenersand the diagnostic ions obtained after fragmentation on CIDor HCD collision cells. It is observed that the majority ofrhamnolipids are dirhamnolipids. Many rhamnolipid speciescontain lipid moieties ranging from C8 to C18 (Table 2). It isinteresting to note the presence of some unprecedentedrhamnolipid congeners, such as Rha-Rha-C10C16 and Rha-Rha-C10C18 (Table 2) which are present as subpopulations,with lower ionizations. Table 3 shows that the majority ofthe rhamnolipid population produced by B. kururiensisKP23T is composed by dirhamnolipids (88.70 %). Figure 1shows the fragmentation of dirhamnolipid Rha-Rha-C10C10,the major rhamnolipid congener.

Metabolic engineering of B. kururiensis for enhancedproduction of biosurfactants

Cloning and expression of rhamnosyltransferase 1 genesfrom P. aeruginosa in B. kururiensis

PCR was used to generate a 2,233-bp fragment containingthe rhlAB genes from genomic DNA from wild-type P.

aeruginosa strain PAO1 (Stover et al. 2000). In this study,we constructed an artificial rhlAB operon under the controlof the tac promoter (Amann et al. 1988) which providesconstitutive or inducible expression. A rhlAB-containingfragment was generated by digestion of plasmid pLMM11with EcoRI and ligated into pTrc99A, generating plasmidpLMM12. Expression plasmid pLMM12 was then trans-formed into E. coli for replication and assessment of induc-ible expression (data not shown). Subsequently, pLMM12was introduced into wild-type B. kururiensis KP23T, gener-ating strain LMM21, designed for expression of rhlA andrhlB from P. aeruginosa.

Structural characterization of rhamnolipids from engineeredstrain B. kururiensis LMM21

Culture of B. kururiensis LMM21 in MSP was carried out,along with the rhamnolipid-producing prototype strain P.aeruginosa PAO1 as internal control. Cell-free supernatantswere processed for extraction of biosurfactants, and theresulting ethyl acetate extract appeared as a yellowish vis-cous paste in both cases.

Analysis of rhamnolipid extracts from strain LMM21by ESI-LTQ-Orbitrap MS lead to the identification of28 congeners. Table 4 shows the rhamnolipid congeners,as well as the diagnostic ions obtained after CID orHCD fragmentation. Structural analysis revealed 61 %of monorhamnolipids within the rhamnolipid population(Table 5), unlike those observed for wild-type strain B.kururiensis KP23T, where only 22 % of the populationwas composed by monorhamnolipids (Table 3). Manyrhamnolipid congeners were identified, containing lipidmoieties with lengths between C8 and C18. It is alsointeresting to observe the identification of congenerRha-Rha-C10C16 and its isotopomer, Rha-Rha-C16C10,also identified in the wild-type strain, as describedabove, which are unprecedented in the literature(Fig. 2). Relative abundance of the molecular popula-tions was determined by measuring their respective peakareas, through the full-scan MS analysis, as shown inTables 3 and 5.

Rhamnolipid production by engineered strain B. kururiensisLMM21

Rhamnolipid production by B. kururiensis strain LMM21was initially assessed by growth in LB broth for 24 h, at28 °C/170 rpm, using wild-type B. kururiensis KP23T and P.aeruginosa PAO1 strains as controls. This preliminary testrevealed enhanced rhamnolipid yield of strain LMM21compared to wild-type strain B. kururiensis KP23T. Theresults demonstrated that strain LMM21 produced substan-tially higher rhamnolipid levels, equivalent to that of P.

Appl Microbiol Biotechnol

Author's personal copy

aeruginosa (data not shown), and was therefore employedfor further experiments.

Based on results obtained from the preliminary testsperformed with LB broth, a production medium optimized for

P. aeruginosa was used to assess rhamnolipid production byB. kururiensis. Figure 3 shows the results observed after120 h of cultivation in MSP medium supplementedwith glycerol as the sole carbon source. Results pre-sented in mg of rhamnolipids per CFU confirm thatstrain LMM21 produced higher rhamnolipid yields whencompared to the reference rhamnolipid-producer strain P.aeruginosa PAO1. These results show that strainLMM21 produced 100 % higher rhamnolipid yieldwhen compared to the wild-type B. kururiensis, asshown in Fig. 3. When compared solely on the basisof production rates, strain LMM21 produces 626 %more rhamnolipids (5.67 g/L) than the wild type(0.78 g/L) and 38 % more than the referencerhamnolipid-producing strain P. aeruginosa PAO1 (4.10 g/L).Furthermore, we observed that a 240-h fermentationwith strain LMM21 yielded 7.40 g/L of rhamnolipids(data not shown).

Surface tension properties of rhamnolipids producedby wild-type or engineered B. kururiensis strains

The CMC was determined for the purified rhamnolipidextracts, as shown in Fig. 4. Rhamnolipid extract from B.kururiensis KP23T produced a CMC of 180–220 mg/L,

Table 2 Rhamnolipids pro-duced by B.kururiensis KP23T Congener Molecular formula Molecular ion [M-H]- Observed fragments (m/z)

Rha-C8C10 C24H44O9 475.2895 163, 169, 305, 311

Rha-C10C8 C24H44O9 475.2895 141, 163, 311, 333

Rha-C10C10 C26H48O9 503.3207 169, 333, 339

Rha-C10C12:1 C28H50O9 529.3363 169, 333

Rha-C10C12 C28H52O9 531.3519 333, 367

Rha-Rha-C8 C20H36O11 451.2169 141, 163

Rha-Rha-C10 C22H40O11 479.2481 147, 163, 169

Rha-Rha-C12 C24H44O11 507.2793 163, 379

Rha-Rha-C8C8 C28H50O13 593.3159 451

Rha-Rha-C8C10 C30H54O13 621.3471 163, 169, 451

Rha-Rha-C10C8 C30H54O13 621.3471 141, 163, 479

Rha-Rha-C10C10:1 C32H56O13 647.3627 163, 479

Rha-Rha-C10:1C10 C32H56O13 647.3627 163, 169

Rha-Rha-C8C12:1 C32H56O13 647.3627 163, 195, 451

Rha-Rha-C12:1C8 C32H56O13 647.3627 141, 163, 505

Rha-Rha-C10C10 C32H58O13 649.3783 163, 169, 339, 479

Rha-Rha-C10C12:1 C34H60O13 675.3939 163, 195, 479

Rha-Rha-C12:1C10 C34H60O13 675.3939 163, 169, 505

Rha-Rha-C10C12 C34H62O13 677.4095 163, 197, 479

Rha-Rha-C12C10 C34H62O13 677.4095 163, 169, 507

Rha-Rha-C10C14:1 C36H64O13 703.4251 479

Rha-Rha-C10C16 C38H70O13 733.4719 479

Rha-Rha-C10C18 C40H74O13 761.5031 163, 479

Table 3 Relative abundance of rhamnolipid congeners produced by B.kururiensis KP23T

Molecularion (m/z)

Congener Molecularformula

Relativequantification (%)

Monorhamnolipids

475,2909 Rha-C8c10 C24H44O9 4.73Rha-C10C8

503,3221 Rha-C10C10 C26H48O9 5.54

529,3363 Rha-C10C12:1 C28H50O9 1.02

Total 11.30 %

Dirhamnolipids

451,2169 Rha-Rha-C8 C20H36O11 7.95

479,2481 Rha-Rha-C10 C22H40O11 25.47

621,3471 Rha-Rha-C8C10 C30H54O13 21.26Rha-Rha-C10C8

649,3783 Rha-Rha-C10C10 C32H58O13 33.11

675,3939 Rha-Rha-C10C12:1 C34H60O13 0.91Rha-Rha-C12:1C10

Total 88.70 %

Appl Microbiol Biotechnol

Author's personal copy

while the extract from B. kururiensis LMM21 showed asignificantly lower CMC of 80–100 mg/L of rhamnoli-pids, very similar to the CMC found when P. aerugi-nosa PAO1 extracts were analyzed, leading to a CMCof 88–110 mg/L. Synthetic surfactants sodium dodecylsulfate (SDS) and Tween 20 were also tested for com-parison, and yielded CMC values of 800 and 500 mg/L,respectively.

Discussion

Biosurfactants such as rhamnolipids have attractedgrowing interest in several industrial sectors, notablythe oil industry, which are increasingly focused ongreen and sustainable policies (Fraser 2000). However,the large-scale application of rhamnolipids are still ham-pered by its high production costs and have triggeredenormous research efforts towards the discovery of nov-el, appropriate biosurfactant-producing microorganisms,and metabolic engineering. The aim of this work was toobtain surface-active rhamnolipid compounds from thenon-pathogenic, N-fixing bacterium B. kururiensisKP23T. The present study also highlighted the applica-tion of the LTQ-Orbitrap Hybrid Mass Spectrometry forstructural elucidation of new biosurfactants and wasessential since the wild-type B. kururiensis strain pro-duced very low levels of those compounds. Most impor-tantly, metabolic engineering of B. kururiensis KP23T

resulted in a highly improved rhamnolipid production,with an increase of ca. 600 %. Noteworthy, monorham-nolipids were predominant in the engineered strain, ac-counting for 61 % of the congeners.

The use of a powerful analytical tool, such as theESI-LTQ-Orbitrap Hybrid MS, with direct injectionthrough a nanoflow capillary system, has given a greatboost in characterization of unknown compounds (Perryet al. 2008). Herein, we show, for the first time, theapplication of HCD fragmentation as a complement ofCID fragmentation for improved identification of bio-surfactants. Its high selectivity and resolving powerprovided better FT mass spectra than the methods wehad previously tested, LC-ESI-MS or Maldi-ToF-MS2

(data not shown), which are commonly used for struc-tural characterization of biosurfactants. In addition, thisstrategy proved to be less time-consuming than theprevious methods.

The observation of unprecedented rhamnolipid spe-cies (Rha-Rha-C10C16 and Rha-Rha-C10C18; Table 2)highlights the potential of a LTQ-Orbitrap HybridMass Spectrometer for determination of unknown com-pounds, since those congeners have not been reportedto date. The analyses of B. kururiensis KP23T rhamno-lipids show a slightly different composition of lipidmoieties, when compared to those previously reportedwithin Burkholderia. However, a typical distributionpattern of mono- and dirhamnolipids is observed whena relative quantification based on the peak areas from

Fig. 1 Mass spectrum of thefragmented Rha-Rha-C10C10

from B. kururiensis KP23T. In-set shows a schematic repre-sentation of the fragmentationpattern of the dirhamnolipid.Asterisks highlight the ion frag-ments within the spectrum

Appl Microbiol Biotechnol

Author's personal copy

the full MS spectra was performed. The predominance ofdirhamnolipid species observed in the rhamnolipid extract ofB. kururiensis KP23T (Table 3) corroborates previous reportsof rhamnolipid production by Burkholderia, in which dirham-nolipids comprise the majority of the congeners (Abdel-Mawgoud et al. 2011).

(Pajarron et al. 2003) identified new dirhamnolipids pro-duced by Pseudomonas glumae strains, later renamed asBurkholderia glumae. In that case, the most abundant con-gener contained a C14-C14 lipid moiety. The same congenerwas also predominant in other Burkholderia, B. pseudomal-lei and B. plantarii (Häubler et al. 1998; Howe et al. 2006).Other mono- and dirhamnolipid congeners were also iden-tified, containing C12-C14, C14-C12, C14-C16, and C16-C14

lipid moieties. More recently, Dubeau et al. (2009) identi-fied, in addition to the congeners already described withinBurkholderia, other congeners of mono- and dirhamnoli-pids: C10-C12, C12-C12, and C16-C16.

In this study, we demonstrate that the populations ofrhamnolipids produced by B. kururiensis KP23T contain

not only congeners described as typical of Burkholderiaspp., but also monorhamnolipids containing C8-C10,C10-C8, C10-C10, and C10-C12:1, and dirhamnolipids con-taining C8, C10, C12, C8-C8, C8-C10, C10-C10 (Fig. 1),C12-C12:1, C10-C10:1, C10:1-C10, C8-C12:1, C12:1-C8, andC10-C14:1 moieties, which are typically reported for P.aeruginosa (Abdel-Mawgoud et al. 2011), in addition tonovel species containing C10-C16 and C10-C18, as de-scribed above (Table 2). When first reported, B. kurur-iensis presented mainly fatty acid chains with 13 ormore carbon atoms (Zhang et al. 2000). However, a morerecent study reported a significant contribution of smaller fattyacid chains within B. kururiensis, such as C10:0 and C11:0

(Anandham et al. 2009). Interestingly, all the biosurfactant-producing Burkholderia reported to date are either animalpathogens (e.g., “pseudomallei” group) or closely related,and are phylogenetically distant from members of theBeneficial-Plant-Environmental (PBE) Burkholderia clus-ter, which include B. kururiensis (Suárez-Moreno et al.2012). The structural profile of the rhamnolipid congeners

Table 4 Rhamnolipids pro-duced by B. kururiensis LMM21 Congener Molecular formula Molecular ion [M-H]- Fragmentation ions (m/z)

Rha-C8C8 C22H40O9 447.2583 141, 163, 283, 305

Rha-C8C10 C24H44O9 475.2895 141, 163, 169, 305, 311

Rha-C10C8 C24H44O9 475.2895 141, 163, 311, 333

Rha-C10C10 C26H48O9 503.3208 163, 169, 333, 339

Rha-C10C12:1 C28H50O9 529.3360 169, 195, 333, 365

Rha-C12:1C10 C28H50O9 529.3360 169,195, 359, 365

Rha-C10C12 C28H52O9 531.3519 163, 169, 197, 333, 367

Rha-C12C10 C28H52O9 531.3519 163, 169, 197, 361, 367

Rha-C10C14 C30H56O9 559.3824 163, 169, 225, 333, 395

Rha-C14C10 C30H56O9 559.3824 163, 169, 225, 389, 395

Rha-C12C12 C30H56O9 559.3824 163, 197, 361, 395

Rha-C12C14 C32H60O9 587.4143 163, 197, 225

Rha-C10C16 C32H60O9 587.4143 163, 169, 253

Rha-C14C14 C34H64O9 615.4455 451

Rha-Rha-C8C10 C30H54O13 621.3471 141, 163, 169, 311, 451

Rha-Rha-C10C8 C30H54O13 621.3471 141, 163, 169, 311, 479

Rha-Rha-C10C10 C32H58O13 649.3783 163, 169, 339, 479

Rha-Rha-C10C12 C34H62O13 677.4095 103, 119, 163, 169, 197, 479

Rha-Rha-C12C10 C34H62O13 677.4095 103, 119, 163, 169, 197, 507

Rha-Rha-C10C14:1 C36H64O13 703.4251 163, 169, 223, 393, 479, 533

Rha-Rha-C12C12:1 C36H64O13 703.4251 163, 197, 393, 507

Rha-Rha-C10C14 C36H66O13 705.4407 163, 169, 225, 395, 479

Rha-Rha-C14C10 C36H66O13 705.4407 163, 169, 225, 395, 535

Rha-Rha-C12C12 C36H66O13 705.4407 163, 197, 395, 507

Rha-Rha-C10C16 C38H70O13 733.4719 163, 169, 253, 422, 479

Rha-Rha-C16C10 C38H70O13 733.4719 163, 169, 253, 422, 563

Rha-Rha-C12C14 C38H70O13 733.4719 163, 197, 225, 422, 507

Rha-Rha-C14C12 C38H70O13 733.4719 163, 197, 225, 422, 535

Appl Microbiol Biotechnol

Author's personal copy

produced by B. kururiensis might possibly relate to its uniquemetabolic features, such as the ability to fix atmosphericnitrogen (Estrada-De-Los-Santos et al. 2001). In fact, ourstudy is the first to demonstrate the ability of a N-fixingbacterium to produce biosurfactants.

The analysis of B. kururiensis LMM21 rhamnolipidmixture lead to the identification of many congenerspreviously reported as typical of Burkholderia, withmonorhamnolipids presenting C12-C12, C12-C14, andC14-C14 and dirhamnolipids presenting C12-C12:1, C12-C12, C14-C12, and C12-C14 moieties (Table 4). The iden-tification of those congeners could be further confirmeddue to enhanced production by strain LMM21. Themolecular profile of the rhamnolipid mixture presentedseveral similarities with the previously obtained massspectra from wild-type strain KP23T. The data allowedthe identification of monorhamnolipids containing C8-C8, C8-C10, C10-C8, C10-C10, C10-C12:1, C12:1-C10, C10-C12, C12-C10, C10-C14, and C14-C10 lipid moieties anddirhamnolipids containing C8-C10, C10-C8, C10-C10, C10-C12, C12-C10, C10-C14:1,C10-C14, C14-C10, C10-C16, andC16-C10 (Fig. 2), besides those previously reported astypically produced by Burkholderia.

Relative abundance of molecular populations wasdetermined by analysis of the full-scan mass spectra,with measurement of their respective peak areas, asshown in Tables 3 and 5. Interestingly, B. kururiensisKP23T produces mostly dirhamnolipids (88.70 %), whilestrain LMM21 produces 61 % of monorhamnolipids.These findings regarding rhamnolipid structural profilefrom strain LMM21 are consistent with the metabolicengineering strategy used, focused on the heterologousexpression of genes rhlA and rhlB (encoding rhamno-syltransferase 1) from P. aeruginosa PAO1 within B.kururiensis. In addition, the presence of a functionalendogenous rhamnosyltransferase 2 (RhlC) can be in-ferred by the enhanced production of dirhamnolipids, inresponse to heterologous expression of P. aeruginosarhamnosyltransferase 1.

Previous work by our group resulted in the initialproduction of 1.54 g/L by the wild-type strain P. aeru-ginosa PA1 in MSP supplemented with glycerol (SantaAnna et al. 2001). After optimization in flasks, theproduction reached 7.45 g/L, a very similar yield towhat we were able to achieve with engineered B. kur-uriensis LMM21 (7.40 g/L), without any specific opti-mization studies. This strongly indicates the potential ofLMM21 strain for rhamnolipid production, aiming atbiotechnological applications.

The results obtained with engineered strain LMM21demonstrate that it presents substantially higher productionlevels of rhamnolipids than well-known rhamnolipid-producing strains, considering it has only been assayed inagitated flasks. It is the best succeeded strategy of metabolicengineering yet addressed for the production of rham-nolipids by a heterologous host, without any reactorproduction optimization (Table 6). Besides being aBiosafety Level 1, non-pathogenic bacterium and a nat-ural producer of rhamnolipids, B. kururiensis has theadvantage of producing substantial amounts of therhamnolipid precursor dTDP-L-rhamnose, also presentin the exopolysaccharide polymer of B. kururiensisstrain M130 (Mattos et al. 2005). The metabolic engi-neering strategy adopted in the present work includedonly biosynthetic enzymes, exclusively dedicated torhamnolipid biosynthesis (RhlA and RhlB), which werefunctionally expressed. Our strategy does not involveglobal genetic regulation pathways, such as QS system,a regulatory system successfully explored by Cha et al.(2008), with a maximal titer of 6.97 g/L of rhamnoli-pids. Metabolic engineering based on global regulatoryproteins may result in generalized modifications in thehost’s metabolism, making the system difficult to con-trol. In addition, exploring the ability of B. kururiensisto fix atmospheric nitrogen will allow the developmentof novel strategies involving nitrogen-deprived culture

Table 5 Relative abundance of rhamnolipid congeners produced by B.kururiensis LMM21

Molecularion (m/z)

Congener Molecularformula

Relativequantification (%)

Monorhamnolipids

447,2583 Rha-C8C8 C22H40O9 0.40

475,2900 Rha-C8C10 C24H44O9 15.28Rha-C10C8

503,3208 Rha-C10C10 C26H48O9 39.65

529,3363 Rha-C10C12:1 C28H50O9

531,3519 Rha-C10C12 C28H52O9 0.27

Rha-C12C10

Total 61.22 %

Dirhamnolipids

621,3481 Rha-Rha-C8C10 C30H54O13 6.73Rha-Rha-C10C8

649,3788 Rha-Rha-C10C10 C32H58O13 25.64

675,3939 Rha-Rha-C10C12:1 C34H60O13 3.07Rha-Rha-C12:1C10

677,4089 Rha-Rha-C10C12 C34H62O13 3.06Rha-Rha-C12C10

703,4248 Rha-Rha-C10C14:1 C36H64O13 0.19Rha-Rha-C12C12:1

705,4395 Rha-Rha-C10C14 C36H66O13 0.08Rha-Rha-C14C10

Rha-Rha-C12C12

Total 38.78 %

Appl Microbiol Biotechnol

Author's personal copy

media, which may represent an additional advantage inthe use of this bacterium for industrial applications.

The surface-active properties of a rhamnolipid mix-ture will determine its future applications. Those proper-ties can directly interfere in processes such as adsorption,aggregation and micellization, affecting the solubiliza-tion, suspension, dispersion, and other possible applica-tions (Placwa-Plociniczak et al. 2011). When applied to

bioremediation, the composition of a rhamnolipid mix-ture can directly affect its performance as contaminantcarrier and its ability to enhance the bioavailability ofthose compounds in the environment. For instance, dir-hamnolipids show a lower capacity of phenantranesolu-bilization than monorhamnolipids, but can, instead,enhance their availability by micellization. Rhamnolipidspecies with enhanced surface tension-lowering proper-ties can facilitate the transport of polluting agentsadsorbed on soils (Mata-Sandoval et al. 1999). Therhamnolipid mixture currently produced by engineeredstrain LMM21 shall be suitable to applications requiringdetergent properties of biosurfactants. The results ob-served in the present study corroborate previous resultspresented and discussed in the literature, regarding theadvantages of biosurfactants (Abdel-Mawgoud et al.2011) since biosurfactants produced by B. kururiensisstrains, wild type or modified, have lower CMCs com-pared to synthetic counterparts (e.g., SDS or Tween 20);they can also reduce the surface tension to lower values.

The predominance of either mono- or dirhamnolipids canhighly interfere in the surface-active properties of the mix-ture, due to relative balance variations between the hydro-philic sugar and hydrophobic fatty acid moieties. Also, thefatty acid chain size can also affect the mixture properties.Evaluations of each mixture and its individual compositionand properties are essential for the “design” of tailor-madesurfactant molecules and mixtures, with desirable propertiesfor specific applications. High CMC values are expected for

Fig. 3 Yields of rhamnolipids by wild-type and engineered strains.Rhamnolipid yields were assessed after fermentation for 120 hours inMSP supplemented with glycerol as sole carbon source. The graphdepicts wild-type B. kururiensis KP23T, engineered B. kururiensisLMM21 and the prototype rhamnolipid producer P. aeruginosaPAO1. Results are expressed as mg of rhamnolipids/CFU (×106)

Fig. 2 Mass spectrum of thefragmented dirhamnolipidsRha-Rha-C10C16 and Rha-Rha-C16C10 from B. kururiensisLMM21. The MS2 scan of ionproducts (m/z) shows the char-acteristic ion fragments of thesecongeners, as illustrated in theschematic inset. Asterisks high-light the ion fragments withinthe spectrum

Appl Microbiol Biotechnol

Author's personal copy

more hydrophilic compounds and the contrary is expectedfor more hydrophobic species, such as Rha-C10C10 (Abaloset al. 2001).

The properties of a rhamnolipid mixture will deter-mine its future applications. When applied to bioreme-diation, the composition of the rhamnolipid mixturescan directly affect their performance as contaminantcarriers and ability to enhance the bioavailability ofthose compounds in the environment. For instance, dir-hamnolipids show a lower capacity of phenanthrene

solubilization than monorhamnolipids, but can, instead,enhance their availability by micellization. Rhamnolipidspecies with enhanced surface tension-lowering proper-ties can facilitate the transport of polluting agentsadsorbed on soils (Mata-Sandoval et al. 1999).

Investigation of other carbon sources and fermenter-based optimization strategies remain to be further addressedand are underway in our research group. Finally, studies onbiosynthetic and regulatory mechanisms involved in therhamnolipid production by B. kururiensis are currently

Fig. 4 Critical micelarconcentration (CMC) of puri-fied rhamnolipid extracts.Extracts were obtained aftercultivation of strains B. kurur-iensis KP23T, LMM21, and P.aeruginosa PAO1 in glycerol-containing MSP

Table 6 Comparative analysis of engineered bacteria for rhamnolipid production

Strains RMLa (g/L) CultivationTime

QPb (mg/L.h−1) Type of expression Reference

P. aeruginosa PG201 2.2 7 days 0.50 Homologous rhlABRI, QSc genes Ochsner et al. (1995)

P. fluorescens ATCC15453

0.25 7 days 0.06 Heterologous rhlAB, QS genes Ochsner et al. (1995)

E. coli W3110 0.120 48 h 2.50 Heterologous rhlAB, rmlBDAC Cabrera-Valladareset al. (2006)

P. aeruginosa PEER02 1.8 4 days 18.75 Homologous rhlAB, controlled byendogenous QS genes

Wang et al. (2007)

E. coli TnERAB 0.175 1 day 7.29 Heterologous rhlAB; inducible Wang et al. (2007)

P. putida KCTC 1067 6.97 7 days 41.50 Controlled by heterologous QS genes Cha et al. (2008)

P. putida KT42C1 1.5 42 h 1.37 Heterologous rhlAB, inducible; reactor scale Wittgens et al. (2011)

B. kururiensis LMM21 7.4 10 days 30.83 Heterologous rhlAB, constitutive This study

a RML: Rhamnolipid production (g/L)bQP: specific production rate constantsc QS: quorum sensing

Appl Microbiol Biotechnol

Author's personal copy

taking place and are expected to contribute to maximizationof rhamnolipid production.

Acknowledgements This study was carried out as part of a projectsupported by PETROBRAS and FAPERJ. L.F.D.T and P.M.S. receivedscholarships from CAPES, and D.C.O.M was supported by ANP.We are also grateful to Gabriel Duarte for invaluable technicalassistance.

References

Abalos A, Pinazo A, Infante MR, Casals M, García F, Manresa A(2001) Physicochemical and antimicrobial properties of newrhamnolipids produced by Pseudomonas aeruginosa AT10 fromsoybean oil refinery wastes. Lagmuir 17:1367–1371

Abdel-Mawgoud AM, Lepine F, Déziel E (2011) Rhamnolipids: diver-sity of structures, microbial origins and roles. Appl MicrobiolBiotechnol 86:1323–1336

Amann E, Ochs B, Abel KJ (1988) Tightly regulated tac promotervectors useful for the expression of unfused and fused proteins inEscherichia coli. Gene 69:301–315

Anandham R, Gandhi PI, Kwon SW, Sa TW, Kim YS, Jee HJ (2009)Mixotrophic metabolism in Burkholderia kururiensis subsp. Thi-ooxidans subsp. nov., a facultative chemolithoautotrophic thiosul-fate oxidizing bacterium isolated from rhizosphere soil andproposal for classification of the type strain of Burkholderiakururiensis as Burkholderia kururisnsis supsp. Kururiensis subsp.nov. Arch Microbiol 191:885–894

Baldani JI, Caruso L, Baldani VLD, Goi SR, Dobereiner J (1997)Recent advances in BNF with non-legume plants. Soil Biol Bio-chem 29:911–922

Cabrera-Valladares N, Richardson AP, Olvera C, Treviño LG, DézielE, Lépine F, Soberón-Chávez G (2006) Monorhamnolipids and 3-(3-hydroxyalkanoyloxy) alkanoic acids (HAAs) production usingEscherichia coli as a heterologous host. Appl Microbiol Biotech-nol 73:187–194

Cha M, Lee N, Kim M, Kim M, Lee S (2008) Heterologous productionof Pseudomonas aeruginosa EMS1 biosurfactant in Pseudomo-nas putida. Bioresour Technol 99:2192–2199

Déziel E, Lépine F, Dennie D, Boismenu D, Mamer OA, Villemur R(1999) Liquid chromatography/mass spectrometry analysis ofmixtures of rhamnolipids produced by Pseudomonas aeruginosastrain 57RP grown on mannitol or naphthalene. Biochim BiophysActa 1440:244–252

Déziel E, Lépine F, Milot S, Villemur R (2003) rhlA is required for theproduction of a novel biosurfactant promoting swarming motilityin Pseudomonas aeruginosa: 3-(3-hydroxyalkanoyloxy) alkanoicacids (HAAs), the precursors of rhamnolipids. Microbiology149:2005–2013

Dubeau D, Déziel E, Woods DE, Lépine F (2009) Burkholderia thai-landensis harbors two identical rhl gene clusters responsible forthe biosynthesis of rhamnolipids. BMC Microbiol 9:263

Elliott GN, Chen WM, Chou JH, Wang HC, Sheu SY, Perin L, ReisVM, Moulin L, Simon MF, Bontemps C, Sutherland JM, Bessi R,de Faria SM, Trinick MJ, Prescott AR, Sprent JI, James EK(2007) Burkholderia phymatum is a highly effective nitrogen-fixing symbiont of Mimosa spp. and fixes nitrogen ex planta.New Phytol 173:168–180

Estrada-De-Los-Santos P, Bustillos-Cristales R, Caballero-Mellado J(2001) Burkholderia, a genus rich in plant-associated nitrogenfixers with wide environmental and geographic distribution. ApplEnviron Microbiol 67:2790–2798

Fraser L (2000) Lipid Lather Removes Metals. Environ Health Per-spect 108:320–323

Fuqua C, Parsek MR, Greenberg EP (2001) Regulation of gene ex-pression by cell-to-cell communication: acyl-homoserine lactonequorum-sensing. Annu Rev Genet 35:439–468

Häubler S, Nimtz M, Domke T, Wray V, Steinmetz I (1998) Purifica-tion and characterization of a cytotoxic exolipid of Burkholderiapseudomallei. Inf Immun 66:1588–1593

Hörmann B, Müller MM, Syldatk C, Hausmann R (2010) Rhamnolipidproduction by Burkholderia plantarii DSM 9509(T). Eur J ofLipid Sci and Technol 112:674–680

Howe J, Bauer J, Andrä J, Schorm AB, Ernst M, Rossle M, ZahringerU, Rademann J, Bradenburg K (2006) Biophysical characteriza-tion of synthetic rhamnolipids. FEBS J 273:5101–5112

Hu Q, Noll RJ, Li H, Makarov A, Hardman M, Cook RJ (2005) TheOrbitrap: a new mass spectrometer. J Mass Spectrom 40:430–443

Koch AK, Kappeli O, Fiechter A, Reiser J (1991) Hydrocarbon assim-ilation and biosurfactant production in Pseudomonas aeruginosamutants. J Bacteriol 173:4212–4219

Makkar RS, Cameotra SS (2002) An update on the use of unconven-tional substrates for biosurfactant production and their new appli-cations. Appl Microbiol Biotechnol 58:428–434

Mata-Sandoval JC, Karns J, Torrents A (1999) High-performanceliquid chromatography method for the characterization of rham-nolipid mixtures produced by Pseudomonas aeruginosa UG2 oncorn oil. J Chromatogr A 864:211–220

Mattos KA, Padua VL, Romeiro A, Hallack LF, Neves BC, UlissesTM, Barros CF, Todeschini AR, Previato JO, Mendonca-PreviatoL (2008) Endophytic colonization of rice (Oryza sativa L.) by thediazotrophic bacterium Burkholderia kururiensis and its ability toenhance plant growth. An Acad Bras Cienc 80:477–493

Mattos KA, Todeschini AR, Heise N, Jones C, Previato JO, Men-donça-Previato L (2005) Nitrogen-fixing bacterium Burkholderiakururiensis produces a novel yersiniose A-containing O-polysaccharide. Glycobiol 15(3):313–321

Ochsner UA, Fiechter A, Reiser J (1994) Isolation, characterization,and expression in Escherichia coli of the Pseudomonas aerugi-nosa rhlAB genes encoding a rhamnosyltransferase involved inrhamnolipid biosurfactant synthesis. J Biol Chem 269:19787–19795

Ochsner UA, Reiser J, Fiechter A, Witholt B (1995) Production ofPseudomonas aeruginosa rhamnolipid biosurfactants in heterolo-gous hosts. Appl Environ Microbiol 61:3503–3506

Pajarron AM, Dekoster CG, Heerma W, Schmidt M, Haverkamp J(2003) Structure identification of natural rhamnolipid mixtures byfast-atom-bombardment tandem mass-spectrometry. Glycoconj J10:219–226

Pearson JP, Pesci EC, Iglewski BH (1997) Roles of Pseudomonasaeruginosa las and rhl quorum-sensing systems in control ofelastase and rhamnolipid biosynthesis genes. J Bacteriol179:5756–5767

Perin L, Martinez-Aguilar L, Castro-Gonzalez R, Estrada-de-los-Santos P, Cabellos-Avelar T, Guedes HV, Reis VM, Caballero-Mellado J (2006) Diazotrophic Burkholderia species associatedwith field-grown maize and sugarcane. Appl Environ Microbiol72:3103–3110

Perry RH, Cooks RG, Noll RJ (2008) Orbitrap Mass Spectrometry:instrumentation, ion motion and applications. Mass Spectrom Rev27:661–699

Placwa-Plociniczak M, Plaza GA, Piotrwska-Seget Z, Cameotra SS(2011) Environmental applications of biosurfactants: recentadvances. Int J Mol Sci 12:633–654

Rahim R, Ochsner UA, Olvera C, Graninger M, Messner P, Lam JS,Soberon-Chavez G (2001) Cloning and functional characteriza-tion of the Pseudomonas aeruginosa rhlC gene that encodes

Appl Microbiol Biotechnol

Author's personal copy

rhamnosyltransferase 2, an enzyme responsible for di-rhamnolipid biosynthesis. Mol Microbiol 40:708–718

Reis RS, Pereira AG, Neves BC, Freire DM (2011) Gene regulation ofrhamnolipid production in Pseudomonas aeruginosa – a review.Bioresour Technol 102(11):6377–6384

Sambrook J, Russell DW (2001) Molecular Cloning: A Laboratory.Manual, 3rd edn. Cold Spring Harbor Laboratory Press, ColdSpring Harbor

Santa Anna LM, Sebastian GV, Pereira N Jr, Alves TL, Menezes EP,Freire DM (2001) Production of biosurfactant from a new andpromising strain of Pseudomonas aeruginosa PA1. Appl BiochemBiotechnol 91–93:459–467

Santos AS, Sampaio AP, Vasquez GS, Santa Anna LM, Pereira N Jr.,Freire DM (2002) Evaluation of different carbon and nitrogensources in production of rhamnolipids by a strain of Pseudomonasaeruginosa. Appl Biochem Biotechnol 98-100:1025–1035

Stover CK, Pham XQ, Erwin AL, Mizoguchi SD, Warrener P, HickeyMJ, Brinkman FS, Hufnagle WO, Kowalik DJ, Lagrou M, GarberRL, Goltry L, Tolentino E, Westbrock-Wadman S, Yuan Y, BrodyLL, Coulter SN, Folger KR, Kas A, Larbig K, Lim R, Smith K,Spencer D, Wong GK, Wu Z, Paulsen IT, Reizer J, Saier MH,Hancock RE, Lory S, Olson MV (2000) Complete genome se-quence of Pseudomonas aeruginosa PAO1, an opportunistic path-ogen. Nature 406:959–964

Suárez-Moreno ZR, Caballero-Mellado J, Coutinho BG, Mendonça-Previato L, James EK, Venturi V (2012) Common Features ofEnvironmental and Potentially Beneficial Plant-Associated Bur-kholderia. Microb Ecol 63:249–266

Toribio J, Escalante AE, Soberón-Chavez G (2010) Production ofrhamnolipids in bacteria other than Pseudomonas aeruginosa.Eur J of Lipid Sci and Technol 112:1082–1087

Van Hamme JD, Singh A, Ward OP (2006) Physiological aspects: part1 in a series of papers devoted to surfactants in microbiology andbiotechnology. Biotechnol Adv 24:604–620

Wang QH, Fang XD, Bai BJ, Liang XL, Shuler PJ, Goddard WA, TangYC (2007) Engineering bacteria for production of rhamnolipid asan agent for enhanced oil recovery. Biotechnol Bioeng 98:842–853

Ward OP (2010) Microbial biosurfactants and biodegradation. AdvExp Med Biol 672:65–74

Withers H, Swift S, Williams P (2001) Quorum sensing as an integralcomponent of gene regulatory networks in Gram-negative bacte-ria. Curr Opin Microbiol 4:186–193

Wittgens A, Tiso T, Arndt TT, Wenk P, Hemmerich J, Müller C,Wichmann R, Küpper B, Zwick M, Wilhelm S, Hausmann R,Syldatk C, Rosenau F, Blank LM (2011) Growth independentrhamnolipid production from glucose using the non-pathogenicPseudomonas putida KT2440. Microb Cell Fact 10:80

Zhang H, Hanada S, Shigematsu T, Shibuya K, Kamagata Y,Kanagawa T, Kurane R (2000) Burkholderia kururiensis sp. nov.,a trichloroethylene (TCE)-degrading bacterium isolated from anaquifer polluted with TCE. Int J Syst Evol Microbiol 50:743–749

Zhu K, Rock CO (2008) RhlA converts {beta}-hydroxyacyl-acyl car-rier protein intermediates in fatty acid synthesis to the {beta}-hydroxydecanoyl-{beta}-hydroxydecanoate component of rham-nolipids in Pseudomonas aeruginosa. J Bacteriol 190:3147–3154

Appl Microbiol Biotechnol

Author's personal copy