rhlA is required for the production of a novel biosurfactant promoting swarming motility in Pseudomonas aeruginosa: 3-(3-hydroxyalkanoyloxy)alkanoic acids (HAAs), the precursors of rhamnolipids

Eric Déziel{dagger}, François Lépine, Sylvain Milot and Richard Villemur

INRS-Institut Armand-Frappier, Laval, Québec, Canada H7V 1B7

Correspondence
Eric Déziel
deziel{at}molbio.mgh.harvard.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Pseudomonas aeruginosa produces extracellular glycolipids composed of L-rhamnose and 3-hydroxyalkanoic acid called rhamnolipids. Although these compounds are usually regarded as biosurfactants or haemolysins, their exact physiological function is not well understood. Rhamnolipids are synthesized by a rhamnosyltransferase, encoded by the rhlAB operon, which catalyses the transfer of TDP-L-rhamnose to 3-(3-hydroxyalkanoyloxy)alkanoic acid (HAA) moieties of various lengths. RhlB is the catalytic protein of the rhamnosyltransferase. rhlA is indispensable for rhamnolipid synthesis, but its function is unknown. Using a liquid chromatography/mass spectrometry method, the production of extracellular HAAs by P. aeruginosa was detected previously and it was demonstrated that they are the actual precursors of rhamnolipid biosynthesis. In this report, evidence is presented indicating that rhlA is required for production of HAAs and that these HAAs display potent surface-active properties. P. aeruginosa can colonize surfaces by swarming motility, a form of organized translocation requiring the production of wetting agents. Using rhlA and rhlB mutants it was observed that swarming requires the expression of the rhlA gene but does not necessitate rhamnolipid production, as HAAs act as surfactants. Finally, it was shown that the use of ammonium instead of nitrate as source of nitrogen and an excess of available iron both decrease rhlA expression and swarming motility.


Abbreviations: HAA, 3-(3-hydroxyalkanoyloxy)alkanoic acid; LC/MS, liquid chromatography/mass spectrometry; MSM, mineral salts medium

{dagger}Present address: Department of Surgery, Molecular Surgery Laboratory, Massachusetts General Hospital, Boston, MA 02114, USA.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Pseudomonas aeruginosa is a ubiquitous and versatile Gram-negative bacterium capable of three types of motility: flagellum-mediated swimming, type IV pilus-dependent twitching, and swarming, which was only recently reported in this species (Déziel et al., 2001; Köhler et al., 2000; Rashid & Kornberg, 2000). Swarming is a form of organized surface translocation allowing rapid colonization of surfaces that depends on extensive flagellation and cell-to-cell contact (Fraser & Hughes, 1999; Harshey, 1994). It is a multicellular phenomenon requiring cell differentiation and co-operation, two features also involved in biofilm formation and bacterial virulence (Harshey, 1994; Sharma & Anand, 2002).

Swarming cells need to overcome the strong surface tension of water surrounding the colony to efficiently colonize the surface (Matsuyama & Nakagawa, 1996). This surface conditioning is often achieved by the production of surface-active compounds, which act as wetting agents (Lindum et al., 1998; Matsuyama et al., 1992, 1995; Mendelson & Salhi, 1996; Toguchi et al., 2000). Spreading growth of Serratia marcescens depends on the production of various extracellular lipopeptides named serrawetins (Matsuyama et al., 1986, 1992, 1995). Synthesis of serrawettin W2, a cyclic lipodepsipentapeptide required for swarming motility of Serratia liquefaciens, is regulated in a cell-density-dependent manner by an acyl-homoserine lactone-based quorum sensing system (Lindum et al., 1998). Swarming of Bacillus subtilis relies on the production of surfactin (Mendelson & Salhi, 1996), a cyclic lipopeptide biosurfactant whose synthesis is also controlled by quorum sensing via two pheromone signal peptides, ComX and CSF (Lazazzera et al., 1997; Solomon et al., 1996).

P. aeruginosa produces extracellular glycolipids composed of L-rhamnose and 3-hydroxyalkanoic acid (rhamnolipids) (Hauser & Karnovsky, 1957; Jarvis & Johnson, 1949). Although their exact physiological function is still unclear, these amphiphilic molecules are usually considered biosurfactants, acting as solubilizing agents promoting the uptake of hydrophobic substrates, especially n-alkanes (Beal & Betts, 2000; Itoh et al., 1971). Moreover, rhamnolipids are virulence factors found in high concentrations in sputa of P. aeruginosa-colonized cystic fibrosis patients (Kownatzki et al., 1987). Rhamnolipids interfere with the normal tracheal ciliary function (Read et al., 1992), inhibit the phagocytic response of macrophages (McClure & Schiller, 1996) and act as heat-stable extracellular haemolysins (Johnson & Boese-Marrazzo, 1980). In liquid cultures, they are produced as a complex mixture of congeners containing one or two 3-hydroxy fatty acids of various length, linked to a mono- or dirhamnose moiety (Déziel et al., 1999, 2000). In general, the two more abundant rhamnolipids are L-rhamnosyl-3-hydroxydecanoyl-3-hydroxydecanoate and L-rhamnosyl-L-rhamnosyl-3-hydroxydecanoyl-3-hydroxydecanoate (Lang & Wullbrandt, 1999; Maier & Soberón-Chávez, 2000) (Fig. 1; m, n=6). According to the biosynthetic pathway proposed by Burger et al. (1963), rhamnolipid synthesis proceeds by two sequential glycosyl transfer reactions, each catalysed by a different rhamnosyltransferase. The first rhamnosyltransferase, which catalyses the transfer of TDP-L-rhamnose to 3-(3-hydroxyalkanoyloxy)alkanoic acid (HAA; Fig. 1), is encoded by the rhlAB operon (Ochsner et al., 1994a, b, 1995). Both genes, co-expressed from the same promoter, are essential for rhamnolipid synthesis but, whereas rhlB is known to encode the catalytic subunit of the rhamnosyltransferase, the function of rhlA is still unresolved. RhlA is probably an inner-membrane-bound protein (Rahim et al., 2001), presumably involved in the synthesis or transport of rhamnosyltransferase precursor substrates or in the stabilization of the RhlB protein (Ochsner et al., 1994a). Environmental factors, especially nutritional conditions, influence rhamnolipid production (Guerra-Santos et al., 1986). Furthermore, cell-to-cell signalling regulates the expression of the rhlAB operon (Ochsner et al., 1994b; Ochsner & Reiser, 1995; Pearson et al., 1997; Pesci et al., 1997). This quorum sensing system is composed of rhlI, the N-butyrylhomoserine lactone autoinducer synthase gene, and rhlR, which encodes the transcriptional activator (Ochsner et al., 1994b; Ochsner & Reiser, 1995). The second rhamnosyltransferase, encoded by rhlC, has been characterized and its expression shown to be co-ordinately regulated with rhlAB by the same quorum sensing system (Rahim et al., 2001). Köhler et al. (2000) reported that cell-to-cell signalling, and both flagella and type IV pili, are required for swarming motility of P. aeruginosa. They observed that an rhlA mutant was unable to swarm and therefore concluded that rhamnolipid production is required for swarming motility of P. aeruginosa.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 1. Putative synthetic pathway of rhamnolipids in P. aeruginosa. The proposed position of RhlA is based on its homology with PhaG. n, m=4, 6 or 8. Pha, Poly-3-hydroxyalkanoate.

 
Using a liquid chromatography/mass spectrometry (LC/MS) method developed to directly analyse rhamnolipids in culture supernatants (Déziel et al., 1999, 2000), we recently detected the production of extracellular HAAs by P. aeruginosa 57RP (Lépine et al., 2002). We demonstrated that these compounds were not rhamnolipid degradation products, but the actual intermediates of rhamnolipid biosynthesis postulated by Burger et al. (1963). In this report, we present evidence indicating that rhlA is required for production of HAAs and that these HAAs display potent surface-active properties. We show that swarming motility is tightly linked to the expression of the rhlA gene, but does not necessitate rhamnolipid production, as HAAs can act as wetting agents. Finally, we provide new insights about the influence of the nutritional environment of the cells on the expression of rhlA and swarming motility.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and culture media.
Three strains of P. aeruginosa used in this study were kindly provided by Dr Urs Ochsner (Department of Microbiology, University of Colorado, USA): wild-type strain PG201, and Tn5-Gm-induced mutants rhlA- UO299 and rhlB- UO287.

Expression of the rhlAB operon was investigated by monitoring {beta}-galactosidase activity of the wild-type strain PG201 containing an rhlA'-lacZ translational fusion (plasmid pECP60) (Pesci et al., 1997). pECP60 was introduced into strain PG201 by electroporation (Smith & Iglewski, 1989) and PG201(pECP60) transformants were selected with carbenicillin (200 µg ml-1). X-Gal (40 µg ml-1) was added to culture media when required.

Swarm plates were composed of 0·5 % Bacto-agar and 8 g nutrient broth l-1, both from Difco, supplemented with 5 g glucose l-1 and dried overnight at room temperature before use (Rashid & Kornberg, 2000). Cells were point-inoculated with a sterile toothpick or 2 µl of an overnight culture and the plates were incubated at 30 °C for 24–48 h. For liquid swarm medium, agar was omitted.

An iron-limited mineral salts medium (MSM) designed to promote rhamnolipid production (Déziel et al., 1996, 2000), supplemented with 2 % (w/v) carbon source, was also used. The composition was (g l-1): KH2PO4, 0·7; Na2HPO4, 0·9; NaNO3, 2·0; MgSO4.7H2O, 0·4; CaCl2.2H2O, 0·1; FeSO4.7H2O, 0·001. The final pH was 6·7.

The effects of the nitrogen source (25 µM NH4Cl or NaNO3, or 12·5 µM both) or addition of iron (0, 1, 5 or 10 µM FeSO4.7H2O) on growth and rhlA expression of PG201(pECP60) were studied on swarm agar plates containing 1·5 % agar and carbenicillin (300 µg ml-1). The plates were spot-inoculated in triplicate with 2 µl from an overnight culture and incubated for 20 h at 30 °C. The colonies were then recovered and resuspended in 1 ml PBS. {beta}-Galactosidase activity was assayed according to Miller (1972).

Biosurfactant production and analysis.
Cultures were grown in 50 ml iron-limited MSM supplemented with 2 % (w/v) mannitol in 250 ml Erlenmeyer flasks and incubated at 30 °C with gyratory shaking at 200 r.p.m. (Déziel et al., 1999, 2000). The cell-free supernatant was analysed after 6 days incubation at 30 °C and 200 r.p.m. Surface and wetting activities were qualitatively compared with the drop-collapsing test (Jain et al., 1991) and surface tension was measured by the ring method with a du Nouy tensiometer (Fisher Scientific). LC/MS analyses were performed as described previously for rhamnolipids and HAAs (Déziel et al., 2000; Lépine et al., 2002). Culture samples were centrifuged at 16 000 g for 5 min to remove the bacteria and the supernatant was filtered through a 0·2 µm filter. An internal standard (16-hydroxyhexadecanoic acid) was added before injecting the sample into the mass spectrometer to allow quantitative measurements. All the analyses were performed in triplicate with a triple quadrupole mass spectrometer Quattro II (Micromass) equipped with a Z-spray interface using electrospray ionization in negative mode. The spectrometer was interfaced to an HP 1100 HPLC (Agilent Technologies) equipped with a 150 mmx4 mm Zorbax C8 reverse phase column (particle size 5 µm). Quantification was performed by integration of the pseudomolecular and the proper fragment ions.

Determination of biosurfactant production on swarm plates was performed as follows. Each strain was inoculated on five MSM agar plates with 2 % mannitol and after 48 h of incubation the whole content was mixed with 100 ml 1 % KHCO3 (pH 9), incubated overnight at 4 °C and centrifuged for 10 min at 1800 g to remove solids. The supernatant was then acidified to pH 4 with concentrated HCl and extracted three times with 40 ml ethyl acetate. The organic fractions were finally pooled, dried and evaporated. The residue was resuspended in an aqueous solution containing 36 % acetonitrile and 4 mM ammonium acetate.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Production of HAAs requires the expression of rhlA
P. aeruginosa 57RP, a strain isolated from hydrocarbon-contaminated soil (Déziel et al., 1996), excretes HAAs, the lipidic precursor of rhamnolipids, along with rhamnolipids (Lépine et al., 2002). With the goal to increase the production of HAAs, we cultivated an rhlA- mutant ({Delta}rhlA : : Km) of this strain. Such mutation also prevents rhlB transcription (Ochsner et al., 1994a). Because rhlA is considered part of the rhamnosyltransferase 1 complex (Ochsner et al., 1994a), we anticipated that inactivation of this gene would inhibit the glycosylation of HAAs and lead to an increase in the concentration of HAA intermediates in culture supernatants. This mutant did not produce rhamnolipids as expected but, surprisingly, did not produce HAAs either.

We further investigated this observation with the reference strain PG201, which was originally used to isolate the rhlAB operon, and the rhlA- UO299 and rhlB- UO287 mutants, both derived from PG201 (Ochsner et al., 1994a). UO299 does not express both rhlA and rhlB, whereas rhlA is still expressed in UO287 (Ochsner et al., 1994a). Cultures of the mutant UO299 contained no trace of HAAs or rhamnolipids, as observed with mutant 57RP ({Delta}rhlA : : Km). However, UO287, while not producing rhamnolipids, was still secreting HAAs. These results indicate that rhlA is involved in the synthesis of HAAs, the lipidic precursors of rhamnolipids and substrate of the RhlB rhamnosyltransferase. Thus, dimers of 3-hydroxyalkanoic acids are directly excreted in the extracellular milieu or coupled to rhamnose by the rhamnosyltransferase encoded by rhlB to produce rhamnolipids.

The RhlB rhamnosyltransferase prefers longer chain and saturated HAAs
We determined the concentrations of free HAAs congeners, along with the mono- and dirhamnolipids, in the supernatant of a PG201 culture (Table 1). The free HAA profile in strain PG201 is similar to the one previously observed for strain 57RP (Lépine et al., 2002). Notably, the supernatant was proportionally depleted in C10-C10 HAA (23·6 %) relative to the mono- (64·4 %) and dirhamnolipids (60·8 %) and contained a large proportion of C8-C8 HAA (22·1 %), while no rhamnolipids containing C8-C8 were detected. In contrast, the total free HAA concentration in the rhlB- mutant UO287 supernatant was almost nine times higher than in the wild-type strain PG201 supernatant (Table 1). Proportions of the various free HAAs differed between strains UO287 and PG201. Most notably, the proportion of free HAAs represented by the C10-C10 congener goes from 23·6 % for PG201 to 56·6 % for strain UO287. This value is very close to the proportion of the C10-C10 congener observed in mono- (64·4 %) and dirhamnolipids (60·8 %) in PG201. The percentage of free HAAs represented by the C8-C8 congener goes from 22·1 % for PG201, to only 0·5 % for strain UO287, while no C8-C8 congeners are observed in mono- and dirhamnolipids of PG201. The very high percentage of free HAAs represented by the C10-C12 : 1 congener (31·3 %) in PG201 becomes 8·6 % in UO287, close to the values observed for the C10-C12 : 1 congeners in mono- and dirhamnolipids of PG201. These results further strengthen our hypothesis that the RhlB rhamnosyltransferase has a preference for longer chain and saturated HAAs (Lépine et al., 2002), leaving an HAA pool enriched in shorter chain and unsaturated congeners, as seen in the free HAAs of PG201. The residual free HAA pool of PG201 is small because the RhlB rhamnosyltranferase utilizes most of the available HAAs. The larger UO287 pool is therefore representative of the initial HAAs pool before rhamnolipid synthesis. This explains why the pool of free HAAs in UO287 cultures is very similar to the HAA profile of mono- and dirhamnolipids observed with PG201.


View this table:
[in this window]
[in a new window]
 
Table 1. Concentration of HAAs, and mono- and dirhamnolipids in PG201 and UO287 culture supernatants after 169 h of incubation

Numbers in parentheses are the SD. Samples were analysed in triplicate. Numbers in brackets are the concentration percentage of a given congener within the same family of compounds. ND, Not detected.

 
The biosynthetic pathway of HAAs is not completely elucidated. Campos-García et al. (1998) have described an NADPH-dependent ketoacyl reductase, RhlG, which is presumably responsible for draining the fatty acid precursors of HAAs from the general de novo biosynthetic pathway towards HAA synthesis (Fig. 1). Although it is essential for rhamnolipid production and presumed to be part of the rhamnosyltransferase 1 complex, the exact function of the rhlA gene product is unknown. A role for the RhlA protein in rhamnolipid precursor biosynthesis had previously been suggested, but not demonstrated (Ochsner et al., 1994a). The amino acid sequence of the rhlA gene product exhibits 41–48 % identity and 60–65 % similarity with 3-hydroxyacyl ACP : CoA transferases (PhaG) identified in various members of the Pseudomonas genus (Pseudomonas fluorescens, Pseudomonas putida, Pseudomonas aeruginosa, Pseudomonas oleovorans, Pseudomonas syringae and Pseudomonas sp. 61-3). PhaG directly links de novo fatty acid biosynthesis to poly-3-hydroxyalkanoate synthesis (Rehm et al., 1998). Thus, based on its homology with PhaG, RhlA could potentially be an acyltransferase catalysing the transfer of the 3-hydroxyacyl moiety from the ACP thioester to CoA. HAAs would result from the condensation of two of these 3-hydroxacyl-CoA residues. It is important to note that the rhamnolipid and poly-3-hydroxyalkanoate biosynthesis pathways are competing for common intermediates (Fig. 1) (Rehm et al., 2001). Future work will be required to determine the exact function of rhlA in HAA synthesis.

HAAs are surfactants
Besides the lack of concordance between the composition of the initial HAA pool and the preference of the RhlB rhamnosyltransferase, presence of HAAs in cultures of the wild-type strain indicates that their release by the rhlB- mutant is not simply the result of overproduction. The mixture of hydroxylated fatty acids is actually excreted under normal growth conditions, suggesting that free HAAs play a specific role of their own in the bacteria. The excretion of anionic biosurfactants, such as rhamnolipids, is revealed by the formation of a dark blue halo around colonies growing on SW Blue agar plates (Siegmund & Wagner, 1991). As expected, the rhlA mutant UO299 did not produce a halo on these plates. Interestingly, the rhlB mutant UO287 produced a small halo, indicating that the cells released anionic amphiphilic compounds other than rhamnolipids.

Accordingly, significant surface activity was detected in the extracellular fluids of strains PG201 and UO287 cultivated in a liquid medium designed to promote the production of rhamnolipids (Table 2). A positive drop-collapsing test and a surface tension below 45 mN m-1 usually denote the presence of surface-active agents. The surface tension of rhlB- mutant cultures was very low and consistently lower than wild-type cultures, suggesting that HAAs display a potent surface-tension-lowering activity. Total rhamnolipids and HAAs from PG201 and UO287 cultures, respectively, were purified, diluted in water over a range of concentrations and the surface tension of these solutions was measured. Fig. 2 shows that HAAs can decrease the surface tension to a lower value than rhamnolipids, confirming that HAAs have excellent tensioactive properties. Thus, HAAs represent a new class of biosurfactants released by P. aeruginosa in addition to rhamnolipids.


View this table:
[in this window]
[in a new window]
 
Table 2. Properties of P. aeruginosa PG201 and its mutants UO299 and UO287

 


View larger version (15K):
[in this window]
[in a new window]
 
Fig. 2. Surface tension of the total mixtures of rhamnolipids or HAAs. Squares, rhamnolipid mixture purified from a PG201 culture; circles, HAA mixture extracted from a UO287 culture. Each data point shows the mean of triplicate measurements from two different experiments. Error bars represent the SD.

 
HAAs act as wetting agents for swarming motility
Bacterial species that are capable of swarming motility often require the production of an extracellular wetting agent (Sharma & Anand, 2002). Köhler et al. (2000) concluded that rhamnolipid production was required for swarming motility because an rhlA- mutant was unable to swarm. We investigated further the relationship between swarming motility and biosurfactant production in P. aeruginosa PG201. Swarming was observed with this strain at 21, 30 and 37 °C in the medium described by Rashid & Kornberg (2000), for Bacto-agar concentrations between 0·4 and 0·8 %. In comparison, swarming was completely abrogated in the rhlA- mutant, but the rhlB- mutant was still able to swarm (Fig. 3).



View larger version (148K):
[in this window]
[in a new window]
 
Fig. 3. Swarm plate incubated for 24 h at 30 °C with P. aeruginosa PG201 (wild-type strain), and with UO287 rhlB- and UO299 rhlA- mutants. Under these experimental conditions, the swarming colonies display a concentric pattern instead of the familiar dendritic pattern.

 
We then checked directly for biosurfactant release in swarm plates. Extraction of the whole content of MSM swarm agar plates, after removal of the cells, confirmed that HAAs were produced under these growth conditions by PG201 and UO287 (Table 3). These results indicated that RhlB rhamnosyltransferase activity, hence production of rhamnolipids, is not required for swarming of P. aeruginosa, but that rhlA is essential. It also suggests that HAAs act as a wetting agent promoting swarming motility. Although HAAs alone seemed sufficient, the wild-type strain always produced larger swarming zones than the rhlB- mutant (Fig. 3), indicating that rhamnolipids also contribute to the swarming process. Serratia rubidaea, which is capable of swarming motility, similarly produces concurrently two types of surface-active extracellular lipids: rubiwettin RG1, a mixture of glycolipids, and rubiwettin R1, a mixture of its glucose-free fatty acids counterpart displaying a potent wetting activity (Matsuyama et al., 1990).


View this table:
[in this window]
[in a new window]
 
Table 3. Quantification of biosurfactant (rhamnolipids and HAAs) production in swarm agar plates by LC/MS

The whole agar content of five MSM swarm plates was extracted after 48 h growth.

 
Production of biosurfactants correlates with swarming motility
Nitrate promotes, whereas ammonium and excess iron repress, rhamnolipid production (Guerra-Santos et al., 1984; Mulligan & Gibbs, 1989). The effect on rhlAB expression and swarming of the addition of an equimolar concentration of NaNO3 or NH4Cl, or the addition of FeSO4.7H2O was investigated. We monitored the {beta}-galactosidase activity of strain PG201 containing an rhlA'-lacZ translational fusion. Addition of nitrate strongly stimulated rhlA'-lacZ expression and swarming (Fig. 4), whereas ammonium had the opposite effect. Interestingly, the stimulating effect of NaNO3 seemed to supersede the repressing effect of NH4Cl when they were added together at the same molar concentration (Fig. 4b. Like most bacteria (Merrick & Edwards, 1995), P. aeruginosa favours ammonium over nitrate as a source of nitrogen (Fig. 4c). Swarming motility was repressed by increasing concentrations of iron in the growth medium (Fig. 5a), and this also correlated with the downregulation of rhlA'-lacZ expression (Fig. 5b). On the other hand, supplemented iron promotes growth (Fig. 5c). These results demonstrate that growth conditions promoting biosurfactant production also favour swarming motility.



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 4. Effect of various nitrogen sources on swarming motility and rhlA expression in P. aeruginosa. (a) Strain PG201 containing a rhlA'-lacZ translational fusion grown on swarm agar supplemented with NaNO3 or NH4Cl for 20 h at 30 °C. X-Gal allows the visual detection of rhlA'-lacZ expression. (b) Expression of rhlA'-lacZ in swarming zones by monitoring {beta}-galactosidase activity (Miller units, MU). (c) Growth of strain PG201 in liquid swarm medium supplemented with NaNO3 (circles) or NH4Cl (squares).

 


View larger version (28K):
[in this window]
[in a new window]
 
Fig. 5. Effect of increasing concentrations of iron on swarming motility and rhlA expression in P. aeruginosa. (a) Strain PG201 containing a rhlA'-lacZ translational fusion grown on swarm agar supplemented with various concentrations of FeSO4.7H2O for 20 h at 30 °C. X-Gal allows the visual detection of rhlA'-lacZ expression. (b) Expression of rhlA'-lacZ in swarming zones by monitoring {beta}-galactosidase activity (Miller units, MU). (c) Growth of strain PG201 in liquid swarm medium supplemented with various concentrations of FeSO4.7H2O.

 
Since is preferred over as a source of nitrogen, it appears that a less attractive nitrogen source promotes biosurfactant production and swarming. Furthermore, the observation that excess iron prevented swarming suggests that less favourable nutritional conditions in general may elicit swarming motility, presumably as a means to find a new niche with more propitious nutrient supplies. Mulligan & Gibbs (1989) have found a direct relationship between enhanced rhamnolipid production and increased glutamine synthetase activity. Synthesis of this enzyme, which is upregulated under nitrogen-limiting conditions, is controlled by the RpoN {sigma} factor ({sigma}54; Totten et al., 1990). It is noteworthy that {sigma}54 is also required for transcription of rhlAB genes (Ochsner et al., 1994a; Pearson et al., 1997) and the synthesis of pili (Ishimoto & Lory, 1989) and flagella (Totten et al., 1990), all features required for swarming motility (Köhler et al., 2000). These observations offer further evidence that biosurfactant production in P. aeruginosa is closely associated with motility, apparently reflecting a general role for survival in the natural environment. The expression of the rhlAB operon and the production of rhamnolipids are regulated by quorum sensing (Ochsner & Reiser, 1995; Pearson et al., 1997) and also by nutritional factors (Guerra-Santos et al., 1986). Our results suggest that nutritional conditions supersede cell-to-cell communication and high cell density does not necessarily correlate with up-regulation of quorum-sensing-controlled genes such as rhlAB. Bollinger et al. (2001) reached a similar conclusion when the expression of sodA, a quorum-sensing-regulated gene encoding the manganese-cofactored superoxide dismutase, was induced by iron limitation independently of quorum sensing. Further studies will be required to elucidate the interplay between nutrition-based and cell-density-based gene regulation in P. aeruginosa.

Conclusion
The finding of a second class of biosurfactants produced by P. aeruginosa as well as the complex regulation of rhlAB expression indicate that swarming motility plays a critical role in the environmental adaptation of this bacterium. The capacity for surface-colonizing migration is increasingly recognized as a widespread trait in eubacteria (Fraser & Hughes, 1999; Harshey, 1994). The multicellular and cooperative nature of swarming motility clearly associates this phenomenon with the natural propensity of micro-organisms to form biofilms (Eberl et al., 1999; Mireles et al., 2001; Sharma & Anand, 2002). Interestingly, Singh et al. (2002) recently reported that very low available iron concentrations, resulting from sequestering, markedly stimulated surface motility, thus preventing biofilm development by P. aeruginosa. It is therefore intriguing that we also observed increased surface motility when limiting iron conditions prevailed. These results support the hypothesis that biosurfactant production and surface motility are hyper-expressed under unfavourable nutritional conditions, presumably to prevent P. aeruginosa from settling and forming a biofilm. This hypothesis is strengthened by the very recent report that rhamnolipid production is involved in the maintenance of biofilm architecture (Davey et al., 2003).


   ACKNOWLEDGEMENTS
 
We thank Urs Ochsner for providing the PG201, UO287 and UO299 strains, and Luciano Passador for the pECP60 plasmid.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Beal, R. & Betts, W. B. (2000). Role of rhamnolipid biosurfactants in the uptake and mineralization of hexadecane in Pseudomonas aeruginosa. J Appl Microbiol 89, 158–168.[CrossRef][Medline]

Bollinger, N., Hassett, D. J., Iglewski, B. H., Costerton, J. W. & McDermott, T. R. (2001). Gene expression in Pseudomonas aeruginosa: evidence of iron override effects on quorum sensing and biofilm-specific gene regulation. J Bacteriol 183, 1990–1996.[Abstract/Free Full Text]

Burger, M. M., Glaser, L. & Burton, R. M. (1963). The enzymatic synthesis of a rhamnose-containing glycolipid by extracts of Pseudomonas aeruginosa. J Biol Chem 238, 2595–2602.[Free Full Text]

Campos-García, J., Caro, A. D., Nájera, R., Miller-Maier, R. M., Al-Tahhan, R. A. & Soberón-Chávez, G. (1998). The Pseudomonas aeruginosa rhlG gene encodes an NADPH-dependent {beta}-ketoacyl reductase which is specifically involved in rhamnolipid synthesis. J Bacteriol 180, 4442–4451.[Abstract/Free Full Text]

Davey, M. E., Caiazza, N. C. & O'Toole, G. A. (2003). Rhamnolipid surfactant production affects biofilm architecture in Pseudomonas aeruginosa PAO1. J Bacteriol 185, 1027–1036.[Abstract/Free Full Text]

Déziel, É., Paquette, G., Villemur, R., Lépine, F. & Bisaillon, J.-G. (1996). Biosurfactant production by a soil Pseudomonas strain growing on polycyclic aromatic hydrocarbons. Appl Environ Microbiol 62, 1908–1912.[Abstract]

Déziel, E., Lépine, F., Dennie, D., Boismenu, D., Mamer, O. A. & Villemur, R. (1999). Liquid chromatography/mass spectrometry analysis of mixtures of rhamnolipids produced by Pseudomonas aeruginosa strain 57RP grown on mannitol or naphthalene. Biochim Biophys Acta 1440, 244–252.[Medline]

Déziel, E., Lépine, F., Milot, S. & Villemur, R. (2000). Mass spectrometry monitoring of rhamnolipids from a growing culture of Pseudomonas aeruginosa strain 57RP. Biochim Biophys Acta 1485, 145–152.[Medline]

Déziel, E., Comeau, Y. & Villemur, R. (2001). Initiation of biofilm formation by Pseudomonas aeruginosa 57RP correlates with the emergence of hyperpiliated and highly adherent phenotypic variants deficient in swimming, swarming and twitching motilities. J Bacteriol 183, 1195–1204.[Abstract/Free Full Text]

Eberl, L., Molin, S. & Givskov, M. (1999). Surface motility of Serratia liquefaciens MG1. J Bacteriol 181, 1703–1712.[Free Full Text]

Fraser, G. M. & Hughes, C. (1999). Swarming motility. Curr Opin Microbiol 2, 630–635.[CrossRef][Medline]

Guerra-Santos, L. H., Käppeli, O. & Fiechter, A. (1984). Pseudomonas aeruginosa biosurfactant production in continuous culture with glucose as carbon source. Appl Environ Microbiol 48, 301–305.[Medline]

Guerra-Santos, L. H., Käppeli, O. & Fiechter, A. (1986). Dependence of Pseudomonas aeruginosa continuous culture biosurfactant production on nutritional and environmental factors. Appl Microbiol Biotechnol 24, 443–448.

Harshey, R. M. (1994). Bees aren't the only ones: swarming in Gram-negative bacteria. Mol Microbiol 13, 389–394.[Medline]

Hauser, G. & Karnovsky, M. L. (1957). Rhamnose and rhamnolipid biosynthesis by Pseudomonas aeruginosa. J Biol Chem 224, 91–105.[Free Full Text]

Ishimoto, K. S. & Lory, S. (1989). Formation of pilin in Pseudomonas aeruginosa requires the alternative {sigma} factor (RpoN) subunit of RNA polymerase. Proc Natl Acad Sci U S A 86, 1954–1957.[Abstract]

Itoh, S., Honda, H., Tomita, F. & Suzuki, T. (1971). Rhamnolipids produced by Pseudomonas aeruginosa grown on n-paraffin (mixture of C12, C13 and C14 fractions). J. Antibiot 24, 855–859.[Medline]

Jain, D. K., Collins-Thompson, D. L., Lee, H. & Trevors, J. T. (1991). A drop-collapsing test for screening surfactant-producing microorganisms. J Microbiol Methods 13, 271–279.[CrossRef]

Jarvis, F. G. & Johnson, M. J. (1949). A glycolipid produced by Pseudomonas aeruginosa. J Am Chem Soc 71, 4124–4126.

Johnson, M. K. & Boese-Marrazzo, D. (1980). Production and properties of heat-stable extracellular hemolysin from Pseudomonas aeruginosa. Infect Immun 29, 1028–1033.[Medline]

Köhler, T., Curty, L. K., Barja, F., Van Delden, C. & Pechère, J.-C. (2000). Swarming of Pseudomonas aeruginosa is dependent on cell-to-cell signaling and requires flagella and pili. J Bacteriol 182, 5990–5996.[Abstract/Free Full Text]

Kownatzki, R., Tummler, B. & Doring, G. (1987). Rhamnolipid of Pseudomonas aeruginosa in sputum of cystic fibrosis patients. Lancet 1, 1026–1027.

Lang, S. & Wullbrandt, D. (1999). Rhamnose lipids – biosynthesis, microbial production and application potential. Appl Microbiol Biotechnol. 51, 22–32.

Lazazzera, B. A., Solomon, J. M. & Grossman, A. D. (1997). An exported peptide functions intracellularly to contribute to cell density signaling in B. subtilis. Cell 89, 917–925.[Medline]

Lépine, F., Déziel, E., Milot, S. & Villemur, R. (2002). Liquid chromatographic/mass spectrometric detection of the 3-(3-hydroxyalkanoyloxy)alkanoic acid precursors of rhamnolipids in Pseudomonas aeruginosa cultures. J Mass Spectrom 37, 41–46.[CrossRef][Medline]

Lindum, P. W., Anthoni, U., Christophersen, C., Eberl, L., Molin, S. & Givskov, M. (1998). N-acyl-L-homoserine lactone autoinducers control production of an extracellular lipopeptide biosurfactant required for swarming motility of Serratia liquefaciens MG1. J Bacteriol 180, 6384–6388.[Abstract/Free Full Text]

Maier, R. M. & Soberón-Chávez, G. (2000). Pseudomonas aeruginosa rhamnolipids: biosynthesis and potential applications. Appl Microbiol Biotechnol 54, 625–633.[CrossRef][Medline]

Matsuyama, T. & Nakagawa, Y. (1996). Surface-active exolipids: analysis of absolute chemical structures and biological functions. J Microbiol Methods 25, 165–175.[CrossRef]

Matsuyama, T., Murakami, T., Fujita, M., Fujita, S. & Yano, I. (1986). Extracellular vesicle formation and biosurfactant production by Serratia marcescens. J Gen Microbiol 132, 865–875.

Matsuyama, T., Kaneda, K., Ishizuka, I., Toida, T. & Yano, I. (1990). Surface-active novel glycolipid and linked 3-hydroxy fatty acids produced by Serratia rubidaea. J Bacteriol 172, 3015–3022.[Medline]

Matsuyama, T., Kaneda, K., Nakagawa, Y., Isa, K., Hara-Hotta, H. & Yano, I. (1992). A novel extracellular cyclic lipopeptide which promotes flagellum-dependent and -independent spreading growth of Serratia marcescens. J Bacteriol 174, 1769–1776.[Abstract]

Matsuyama, T., Bhasin, A. & Harshey, R. M. (1995). Mutational analysis of flagellum-independent surface spreading of Serratia marcescens. J Bacteriol 177, 987–991.[Abstract]

McClure, C. D. & Schiller, N. L. (1996). Inhibition of macrophage phagocytosis by Pseudomonas aeruginosa rhamnolipids in vitro and in vivo. Curr Microbiol 33, 109–117.[CrossRef][Medline]

Mendelson, N. H. & Salhi, B. (1996). Patterns of reporter gene expression in the phase diagram of Bacillus subtilis colony forms. J Bacteriol 178, 1980–1989.[Abstract]

Merrick, M. J. & Edwards, R. A. (1995). Nitrogen control in bacteria. Microbiol Rev 59, 604–622.[Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mireles, J. R., 2nd, Toguchi, A. & Harshey, R. M. (2001). Salmonella enterica serovar typhimurium swarming mutants with altered biofilm-forming abilities: surfactin inhibits biofilm formation. J Bacteriol 183, 5848–5854.[Abstract/Free Full Text]

Mulligan, C. N. & Gibbs, B. F. (1989). Correlation of nitrogen metabolism with biosurfactant production by Pseudomonas aeruginosa. Appl Environ Microbiol 55, 3016–3019.[Medline]

Ochsner, U. A. & Reiser, J. (1995). Autoinducer-mediated regulation of rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 92, 6424–6428.[Abstract]

Ochsner, U. A., Fiechter, A. & Reiser, J. (1994a). Isolation, characterization, and expression in Escherichia coli of the Pseudomonas aeruginosa rhlAB genes encoding a rhamnosyltransferase involved in rhamnolipid biosurfactant synthesis. J Biol Chem 269, 19787–19795.[Abstract/Free Full Text]

Ochsner, U. A., Koch, A. K., Fiechter, A. & Reiser, J. (1994b). Isolation and characterization of a regulatory gene affecting rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. J Bacteriol 176, 2044–2054.[Abstract]

Ochsner, U. A., Hembach, T. & Fiechter, A. (1995). Production of rhamnolipid biosurfactants. Adv Biochem Eng Biotechnol 53, 89–118.

Pearson, J. P., Pesci, E. C. & Iglewski, B. H. (1997). Roles of Pseudomonas aeruginosa las and rhl quorum-sensing systems in control of elastase and rhamnolipid biosynthesis genes. J Bacteriol 179, 5756–5767.[Abstract]

Pesci, E. C., Pearson, J. P., Seed, P. C. & Iglewski, B. H. (1997). Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J Bacteriol 179, 3127–3132.[Abstract]

Rahim, R., Ochsner, U. A., Olvera, C., Graninger, M., Messner, P., Lam, J. S. & Soberon-Chávez, G. (2001). Cloning and functional characterization of the Pseudomonas aeruginosa rhlC gene that encodes rhamnosyltransferase 2, an enzyme responsible for di-rhamnolipid biosynthesis. Mol Microbiol 40, 708–718.[CrossRef][Medline]

Rashid, M. H. & Kornberg, A. (2000). Inorganic polyphosphate is needed for swimming, swarming, and twitching motilities of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A 97, 4885–4890.[Abstract/Free Full Text]

Read, R. C., Roberts, P., Munro, N. & 7 other authors (1992). Effect of Pseudomonas aeruginosa rhamnolipids on mucociliary transport and ciliary beating. J Appl Physiol 72, 2271–2277.[Abstract/Free Full Text]

Rehm, B. H. A., Krüger, N. & Steinbüchel, A. (1998). A new metabolic link between fatty acid de novo synthesis and polyhydroxyalkanoic acid synthesis. The phaG gene from Pseudomonas putida KT2440 encodes a 3-hydroxyacyl-acyl carrier protein-coenzyme A transferase. J Biol Chem 273, 24044–24051.[Abstract/Free Full Text]

Rehm, B. H. A., Mitsky, T. A. & Steinbüchel, A. (2001). Role of fatty acid de novo biosynthesis in polyhydroxyalkanoic acid (PHA) and rhamnolipid synthesis by Pseudomonads: establishment of the transacylase (PhaG)-mediated pathway for PHA biosynthesis in Escherichia coli. Appl Environ Microbiol 67, 3102–3109.[Abstract/Free Full Text]

Sharma, M. & Anand, S. K. (2002). Swarming: a coordinated bacterial activity. Curr Science 83, 707–715.

Siegmund, I. & Wagner, F. (1991). New method for detecting rhamnolipids excreted by Pseudomonas species during growth on mineral agar. Biotechnol Tech 5, 265–268.

Singh, P. K., Parsek, M. R., Greenberg, E. P. & Welsh, M. J. (2002). A component of innate immunity prevents bacterial biofilm development. Nature 417, 552–555.[CrossRef][Medline]

Smith, A. W. & Iglewski, B. H. (1989). Transformation of Pseudomonas aeruginosa by electroporation. Nucleic Acids Res 17, 10509.[Medline]

Solomon, J. M., Lazazzera, B. A. & Grossman, A. D. (1996). Purification and characterization of an extracellular peptide factor that affects two different developmental pathways in Bacillus subtilis. Genes Dev 10, 2014–2024.[Abstract]

Toguchi, A., Siano, M., Burkart, M. & Harshey, R. M. (2000). Genetics of swarming motility in Salmonella enterica serovar Typhimurium: critical role for lipopolysaccharide. J Bacteriol 182, 6308–6321.[Abstract/Free Full Text]

Totten, P. A., Lara, J. C. & Lory, S. (1990). The rpoN gene product of Pseudomonas aeruginosa is required for expression of diverse genes, including the flagellin gene. J Bacteriol 172, 389–396.[Medline]

Received 27 November 2002; revised 3 March 2003; accepted 15 April 2003.