OUP user menu

Selection and partial characterization of a Pseudomonas aeruginosa mono-rhamnolipid deficient mutant

Marina Wild, Alma Delia Caro, Ana Lilia Hernández, Raina M Miller, Gloria Soberón-Chávez
DOI: http://dx.doi.org/10.1111/j.1574-6968.1997.tb12586.x 279-285 First published online: 1 August 1997


Pseudomonas aeruginosa produces rhamnolipids which are tenso-active compounds with potential industrial and environmental applications. There are two main types of rhamnolipids produced in liquid cultures, rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate (mono-rhamnolipid) and rhamnosyl-rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate (di-rhamnolipid). In this work we report the selective isolation of a rhamnolipid deficient mutant (IBT8), which does not accumulate mono-rhamnolipid while still producing di-rhamnolipid. IBT8 was selected after random mutagenesis with Tn501; yet, its mono-rhamnolipid deficiency was found associated neither with its Tn501 insertion nor with a possible alteration in the rhlABRI genes for rhamnosyl-transferase 1 synthesis. Different possibilities to explain IBT8 phenotype are discussed.

  • Pseudomonas aeruginosa
  • Rhamnolipid
  • rhl gene

1 Introduction

Rhamnolipids are compounds produced by Pseudomonas aeruginosa which reduce water surface tension and emulsify oil. These compounds are biodegradable and have potential industrial and environmental applications [14]. Recently rhamnolipids have been found to have important antagonistic effects on economically important zoosporic plant pathogens, thus opening their use as biocontrol agents [5].

The rhamnolipids produced by P. aeruginosa in liquid cultures are mainly rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate (mono-rhamnolipid) and rhamnosyl-rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate (di-rhamnolipid). Rhamnolipid biosynthesis was first described by Burger et al. in 1963 [6] as a series of rhamnose transfers from TDP-rhamnose to β-hydroxydecanoyl-β-hydroxydecanoate. The enzyme rhamnosyl-transferase 1 (Rt 1) catalyzes the synthesis of mono-rhamnolipids while rhamnosyl-transferase 2 (Rt 2) synthesizes di-rhamnolipid from TDP-rhamnose and mono-rhamnolipid.

Genes coding for biosynthesis, regulation and induction of Rt 1 enzyme are organized in tandem in the rhlABRI gene cluster around minute 38 of the P. aeruginosa chromosome [7]. Mutations in any of these rhl genes cause a rhamnolipid minus phenotype which indicates that mono-rhamnolipid is an obligate precursor of di-rhamnolipid. The genes encoding Rt 2 are yet to be described.

In this study we report the selection and characterization of a rhamnolipid deficient mutant (IBT8) which was derived from a P. aeruginosa PAO1 derivative (strain 29-1 [8], Table 1). This mutant is deficient in the production of mono-rhamnolipid, but is still able to produce di-rhamnolipid. We also report that the gene affected in mutant IBT8 is different from the previously described rhlABRI genes involved in Rt 1 enzyme biosynthesis.

View this table:

Strains and plasmids used in this study

Strain or plasmidRelevant characteristicsaReference
Pseudomonas aeruginosa
29-1PAO1 mutant, met-9020 catA1 puuA1 lip1[7]
IBT829-1::Tn501 mono-rhamnolipid deficient mutantThis work
29-229-1 Strr spontaneous mutantThis work
ATCC 9027Wild-type strain produces only mono-rhamnolipid[2]
Escherichia coli
UW937thr-1, leu-6, thi-1, lacY-1, supE44, tonA21, λ, gyrA[9]
pUW942RP4-ColE1 Tra+, Tn501, Hgr[9]
pUO58Contains the rhlABRI genes, Apr[6]
pJB3JIRP4 derivative, CMA, Apr, Tcr[18]
pNH2198Cosmid clone with P. aeruginosa DNAThis work
  • aAbbreviations used: Resistance to: ampicillin (Apr), kanamycin (Kmr), mercury (Hgr), streptomycin (Strr) and tetracycline (Tcr). Self transmissible (Tra+), promotes chromosome mobilization (CMA).

2 Materials and methods

2.1 Microbiological procedures

Strains and plasmids used in this work are shown in Table 1. P. aeruginosa strains were routinely grown at 29°C on one of the following media: Pseudomonas Isolation Agar (PIA, Difco), PPGAS [2] (NH4Cl2 0.02 M; KCl 0.02 M; Tris-HCl 0.12 M; MgSO4 0.0016 M; glucose 0.5%; peptone 1%, adjusted to pH 7.2) or PPSW (PPGAS medium supplemented with 2.5 μg/ml of methylene blue and 200 μg/ml of the surfactant cetyl-trimethyl-ammonium-bromide (CTAB)). Antibiotic and mercury concentrations (μg/ml) used for P. aeruginosa and Escherichia coli, respectively, were: carbenicillin (Cb) 300 and 30, mercury (Hg) 10 and 10, streptomycin (Str) 200 and 200, and tetracycline (Tc) 150 and 30. Transformation of P. aeruginosa was done as described by Olsen et al. [9]. Matings were performed mixing a 1:1 proportion of donor:recipient strains and incubating them overnight at 29°C onto agar LB medium. Appropriate dilutions were made on selective media in order to isolate transconjugants. Controls of individual donor and recipient strains were treated similarly to the mixture. Mutagenesis with the transposon Tn501 was carried out by selecting Hgr transconjugants on PIA medium from the cross between E. coli UW937(pUW942) [10] and P. aeruginosa 29-1. Rhamnolipid deficient mutants were selected on PPSW medium by lack of halo formation.

2.2 Rhamnolipid determination

Culture supernatants of 96 h PPGAS cultures at 29°C and shaken at 250 rpm were used for all rhamnolipid determinations. Total rhamnolipid concentration was determined from cell culture supernatants by measuring rhamnose concentration after acid hydrolysis using the orcinol method [11]. Interfacial tension between water and hexadecane was determined by the ‘drop weight’ method [12]. Thin layer chromatography (TLC) was carried out on silica gel G plates treated as indicated by Matsuyama et al. [13], rhamnolipids being revealed with a 25 mg/ml solution of α-naphthol. The samples for TLC were prepared by extracting 333 μl of culture supernatant twice with two volumes of diethyl ether and suspending in 10 μl of water. Unrevealed upper spots were scratched off from a parallel untreated TLC plate and were extracted twice with diethyl ether, suspended in 1 ml of phosphate buffer and analyzed by high performance liquid chromatography (HPLC). A reverse phase nova-pack C18 3.9 per 150 mm column was used. HPLC analysis was isocratic using 40% acetonitrile and 60% 0.1 M phosphate buffer pH 6 with a flow rate of 1 ml/min.

2.3 Construction of the genomic library

Total genomic DNA from P. aeruginosa IGB83 [14] was partially digested with Sau3A and ligated to cosmid pCP13 [15] digested with BamHI endonuclease. Cosmids were packaged onto λ heads and transfected as described previously [16]. About 5000 clones were obtained, which represent at least 10 times the size of the bacterial genome.

2.4 Nucleic acid procedures

DNA isolation and cloning, Southern blotting and nick translation procedures were carried out as described previously [16].

3 Results and discussion

3.1 Detection of rhamnolipid production on plates

SW medium was originally developed for the selection of P. aeruginosa rhamnolipid deficient mutants on plates [17]. It contains the cationic dye methylene blue which complexes with the slightly anionic rhamnolipid molecules in the presence of the surfactant CTAB. This complex is visualized in the form of a blue precipitate surrounding the rhamnolipid producing colonies. However, in our experiments, P. aeruginosa colonies formed a blue pigment which interfered with the blue precipitate formed by rhamnolipids in SW medium. Therefore, we developed a slightly different medium for enhanced visualization of precipitated rhamnolipids. In this medium CTAB and Congo red or methylene blue were combined with PPGAS, a phosphate-limited medium used for rhamnolipid production. Using these media it was found that a halo was formed around the P. aeruginosa colonies that produce rhamnolipids or around the purified rhamnolipids themselves (Fig. 1). A comparison of methylene blue and Congo red showed the former to be slightly better.

Figure 1

Halo formation in PPSW plates associated with rhamnolipid production. A: Strain 29-1, B: mutant IBT8, and C: purified rhamnolipids from strain 29-1.

The lack of halo formation on plates of PPGAS supplemented with methylene blue and CTAB (PPSW medium) was used as a criterion to select rhamnolipid deficient mutants after Tn501 mutagenesis. One thousand 29-1::Tn501 insertion mutants were screened on PPSW medium and one colony devoid of a halo (IBT8) was further characterized (Fig. 1).

3.2 Rhamnolipids production by IBT8

Total rhamnolipid production by mutant IBT8 was nearly 10-fold lower than that of its parental strain 29-1 (Table 2). Lower interfacial tension values confirmed that rhamnolipids were still present in the IBT8 culture supernatant, but in less quantity (Table 2). Interestingly TLC analysis of an IBT8 supernatant showed that the IBT8 di-rhamnolipid spot remained visible while the mono-rhamnolipid species could not be detected (Fig. 2A). As control we used purified rhamnolipids from the culture supernatants of strain 29-1 (wild-type) and strain ATCC 9027 which accumulates only mono-rhamnolipid [2]. In order to rule out that the lack of mono-rhamnolipid detection in mutant IBT8 was due to a problem of TLC sensibility, we ran different dilutions of the rhamnolipids produced by strain 29-1. We found that in all the dilutions where di-rhamnolipid was detected in the 29-1 supernatant, the spot corresponding to the mono-rhamnolipid was clearly visible (Fig. 2B). The 1:10 dilution of the 29-1 supernatant approximately corresponds to the rhamnolipid concentration produced by mutant IBT8, in this condition the mono-rhamnolipid spot is apparent, but the di-rhamnolipid spot is hardly seen (Fig. 2B, lane 4). We cannot rule out that small amounts of mono-rhamnolipid are still produced by mutant IBT8, but it is clear that this mutant produces a considerable decreased ratio of the mono- to the di-rhamnolipid species than the wild-type strain.

View this table:

Production of rhamnolipids and interfacial tension of culture supernatants of different P. aeruginosa strains

StrainRhamnolipids(mg rhl/mg protein)Interfacial tension(din/cm)
– (broth)19.82
Figure 2

Detection of rhamnolipids in TLC plates. A: Lanes correspond to: 1: strain 29-1, 2: mutant IBT8, 3: mutant IBT8(pUO58), and 4: strain ATCC 9027. B: Different dilutions of the rhamnolipids produced by strain 29-1, lanes correspond to: 1: undiluted, 2: 1:2, 3: 1:4, and 4: 1:10.

Identification of the upper migrating band of the parental strain supernatant as mono-rhamnolipid (the one lacking in IBT8 supernatant), was confirmed by its similar HPLC elution profile to that of strain ATCC 9027 supernatant (Fig. 3), previously characterized as authentic mono-rhamnolipid by mass spectroscopy [3].

Figure 3

HPLC elution profiles of purified mono-rhamnolipid from P. aeruginosa strains ATCC 9027 (▪) and 29-1 (—).

3.3 The mutation in IBT8 does not affects the rhlABRI genes

Transcomplementation of IBT8 mono-rhamnolipid deficiency by the rhlABRI genes was undertaken by transferring plasmid pUO58 into IBT8 [7] (Table 1); only partial complementation was apparent when rhamnolipids were measured (Table 2, Fig. 2), but halo formation was not restored (data not shown). The lack of halo formation by strain IBT8(pUO58) might be due to a lower sensitivity of this test for rhamnolipid detection than that of the orcinol method. These results suggested that the mutation present in strain IBT8 was not due to a mutation in the rhl genes encoded by plasmid pUO58. This was the expected result since mutations in any of the rhl gene clusters completely abrogate the synthesis of both rhamnolipid species [7].

A cosmid clone (pNH2198) which fully complemented mutant IBT8 for rhamnolipid production was selected from a P. aeruginosa IGB83 [14] genomic library by selecting the restoration of halo formation in PPSW medium. The Southern blot experiment shown in Fig. 4 shows that plasmid pNH2198 does not hybridize with the rhlABRI genes and that mutant IBT8 and its parental strain 29-1 have the same pattern of hybridization with them. These results confirmed that the mutation in IBT8 is not within the rhlABRI genes. The gene encoded by plasmid pNH2198 which complements mutant IBT8, and is the presumably mutated gene in IBT8 strain, is currently being characterized.

Figure 4

Southern blot analysis using plasmid pUO58 as probe. A: Electrophoretic profile of DNAs restricted with different endonucleases. B: Hybridization of these DNAs with plasmid pUO58 used as probe. Lanes correspond to: 1 and 6: phage λ DNA restricted with HindIII endonuclease, 2: plasmid pNH2198 DNA restricted with PstI endonuclease, 3: total genomic DNA of mutant IBT8 restricted with PstI enzyme, 4: total genomic DNA of strain 29-1 restricted with PstI enzyme, and 5: plasmid pUO58 restricted with EcoRI and HindIII enzymes.

3.4 The mutation in IBT8 strain is not caused by the Tn501 insertion

In order to determine whether the Tn501 insertion in mutant IBT8 was causing its mono-rhamnolipid deficiency, the linkage between Hgr encoded by the transposon and rhamnolipid deficiency was determined. Hgr was mobilized from mutant IBT8 to a 29-1 Strr spontaneous mutant (29–2, Table 1) using plasmid pJB3JI which has chromosome mobilization ability [18]. The frequency of Hgr transconjugants from this cross was at least one order of magnitude higher than the transposition frequency of Tn501 in strain 29-1, thus secondary events of transposition were discarded as being the primary event for the incorporation of the transposon into strain 29-2. Thirteen 29-2 Hgr transconjugants were selected, but none inherited rhamnolipid deficiency. According to the transposition and the recombination frequencies mentioned above, only one to three of these transconjugants could represent new Tn501 transposition events. This experiment strongly suggests the presence in strain IBT8 of a second mutation, independent of the Tn501 insertion, that affects mono-rhamnolipid accumulation.

This is the first report of a P. aeruginosa mutant affected in the production of mono-rhamnolipid while still presenting di-rhamnolipid production. This could be explained by: (1) degradation of the former, and (2) expression of a second metabolic route from substrates other than mono-rhamnolipid. The first alternative would imply synthesis of mono-rhamnolipid from the expression of the rhlABRI gene cluster, its flow to di-rhamnolipid via the Rt 2 pathway and degradation of surplus amounts of mono-rhamnolipids. The second alternative implies both the silencing of an intact rhlABRI gene cluster and the onset of an alternative pathway, whatever its nature. Partial complementation of IBT8 by pUO58 with regard to mono-rhamnolipid accumulation could be interpreted by either positive (as in alternative 1) or negative (as in alternative 2) regulation and thus, does not throw any light as to which of the two alternatives is correct. Only further work along these lines, now in progress, will aid to solve this issue.


We greatly acknowledge the support of Urs A. Ochsner during the development of this research. We thank Rebeca Nájera for her technical assistance and Fernando Bastarrachea for his critical reading of the manuscript. This research was founded in part by the United States-Mexico Foundation for Science, Grant 57-CH197, and by the National Institute of Environmental Health Sciences, Grant P42 ES04940.


View Abstract