The Autotransporter Esterase EstA of Pseudomonas aeruginosa Is Required for Rhamnolipid Production, Cell Motility, and Biofilm Formation

Bacterial strains and growth conditions.The strains and plasmids used in this study are listed in Table 1. Escherichia coli was routinely grown at 37°C in Luria-Bertani (LB) medium or on LB agar plates (53). P. aeruginosa was grown at 30°C in LB medium or in rhamnolipid-promoting PPGAS medium (0.02 M NH₄Cl₂, 0.02 M KCl, 0.12 M Tris-HCl, 1.6 mM MgSO₄, 0.5% [wt/vol] glucose, 1% peptone; adjusted to pH 7.2) (62). Agrobacterium tumefaciens was grown at 28°C in AB minimal medium (12) supplemented with 0.2% glucose (ABG).

When required, antibiotics were added to the media at the following concentrations: tetracycline (Tc), 25 μg ml⁻¹ (E. coli) or 100 μg ml⁻¹ (P. aeruginosa); ampicillin, 100 μg ml⁻¹ (E. coli); carbenicillin, 500 μg ml⁻¹ (P. aeruginosa); streptomycin (Sm) and spectinomycin (Sp), 25 μg ml⁻¹ each (E. coli) or 100 μg ml⁻¹ each (P. aeruginosa); chloramphenicol, 50 μg ml⁻¹ (E. coli) or 300 μg ml⁻¹ (P. aeruginosa); gentamicin, 30 μg ml⁻¹ (A. tumefaciens).

General DNA manipulations.Plasmid DNA was purified as described by Birnboim and Doly (6), followed by anion-exchange chromatography on QIAGEN GmbH (Hilden, Germany) tips. Recombinant DNA techniques were performed essentially as described by Sambrook et al. (53). Restriction endonuclease reactions and bacteriophage T4 DNA ligase treatments were done as recommended by the manufacturers. DNA fragments were analyzed on 0.4 to 1% (wt/vol) agarose gels.

Construction of an estA-negative mutant.The suicide plasmid pMEestAΩ, which harbors a ΩSm^r/Sp^r cassette inserted into the estA gene, was transformed into E. coli S17-1. This strain was used to transfer the plasmid into P. aeruginosa by diparental mating (55). For counterselection of E. coli S17-1 donor cells from P. aeruginosa after conjugation, Irgasan was used at a concentration of 25 μg ml⁻¹. The resulting Sm/Sp-resistant P. aeruginosa clones were tested for the loss of Tc resistance as an indication for successful allelic exchange of estA into ΔestA::Ω-Sm/Sp. Enrichment for Tc-sensitive cells was done using cycloserine at a concentration of 1.0 mg ml⁻¹ and Tc at 20 μg ml⁻¹ Successful inactivation was verified by PCR, Western blot analysis, and esterase activity assays.

Construction of an inactive EstA variant by site-directed mutagenesis.In order to inactivate the enzyme activity of the esterase, we changed the codon of active-site serine into an alanine coding triplet by overlap extension PCR (22) using the mutagenesis primers listed in Table 1. The resulting fragment was used to replace an internal EcoNI fragment of the wild-type estA gene in pBBX+, generating pBBXmut. The mutation was verified by DNA sequencing (QIAGEN GmbH, Hilden, Germany).

Overexpression and purification of EstA for antibody generation.Overexpression was done using pBR22estA in E. coli BL21(DE3). This plasmid is a derivative of pBR22b into which the open reading frame of the estA gene was inserted at the NdeI/XhoI sites. Expression was induced by addition of IPTG (isopropyl-β-d-thiogalactopyranoside; 0.4 mM) at an optical density at 500 nm (OD₅₈₀) of 0.5, and cells were harvested 2 h after induction. EstA accumulated in inclusion bodies, which were collected by centrifugation of cell extracts after sonification of the cells. Chemically denatured (8 M urea) EstA was partly purified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (5% stacking and 10% separating gels) (34) and used for the generation of a rabbit polyclonal anti-EstA serum (Eurogentec, Belgium). The specificity of the polyclonal serum was tested and found to be ideal at a dilution of 1:85,000.

Esterase assay.Cells harvested by centrifugation were washed in KP_i buffer (100 mM, pH 7.2) and disrupted by sonification, and the resulting cell extract was used to determine the esterase activity. The substrate (25 μl p-nitrophenyl-caproate [Sigma] dissolved in 5 ml ethanol) was added to 100 ml of KP_i (100 mM)-MgSO₄ (10 mM) buffer (pH 7.2). This test solution (1.5 ml) and cell extracts (5 to 20 μl) were mixed, and ΔOD₄₁₀ min⁻¹ was monitored at 30°C.

Cultivation of P. aeruginosa biofilms.Biofilms were grown on glass slides (Superior; Marienfeld Company) in flow chambers with individual channel dimensions of 3 by 8 by 54 mm for 72 h at 30°C in modified PPGAS medium (0.05% glucose, 0.1% peptone, 2 mM NH₄Cl, 2 mM KCl, 100 mM Tris, 10 mM MgSO₄, 10 mM KNO₃, pH 7.5) or modified alginate-promoting medium (100 ml of a solution containing 100 mM Na-d-gluconate, 100 ml of a solution containing 100 mM KNO₃, and 50 ml of a solution containing 10 mM MgSO₄ and 750 ml phosphate buffer [1.03 g NaH₂PO₄, 2.93 g K₂HPO₄, pH 7.5]). All strains used for biofilm formation assays were chromosomally tagged with gfp as a fluorescent label for confocal laser scanning microscopy (31). The flow system was assembled and prepared as described previously (13). The cultures for inoculation of the flow chambers were prepared as follows. Overnight liquid cultures of P. aeruginosa were plated on PPGAS agar plates including the appropriate antibiotic and incubated overnight at 30°C. Thereafter bacteria from single colonies were resuspended in medium, cells were counted in a Abbe-Zeiss counting cell chamber, and the flow chambers were inoculated by injecting 1 × 10⁷ cells/ml in 5 ml flow medium. After inoculation flow channels were left without flow for 1 h for initial adhesion of the inoculum to the glass surface. Medium flow was kept at a constant rate of 20 ml/h, equivalent to a mean flow velocity of 0.02 mm/s.

Microscopy and image acquisition.All microscopic observations and image acquisitions were performed on a confocal laser scanning microscope (LSM 510; Carl Zeiss Jena, Jena, Germany) equipped with detectors and filter sets for monitoring green fluorescent protein fluorescence.

Motility.Surface motility of P. aeruginosa was monitored on PPGAS agar plates, which had agar concentrations of 0.5% for the swarm plates, 0.3% for the swim plates, and 1.5% for the twitching motility plates. The swarm agar plates contained glutamate (0.05%) instead of NH₄Cl. In the case of swimming and swarming the bacteria were staked on top of the agar, whereas for monitoring of twitching motility the bacteria where staked through the agar to colonize the interphase between the petri dish and agar.

Rhamnolipid analysis.Cultures were adjusted to an initial OD₅₈₀ of 0.05 and grown in 200 ml PPGAS medium supplemented with the appropriate antibiotics in 2-liter Erlenmeyer flasks at 30°C and 200 rpm for 48 h.

The orcinol assay is a colorimetric test to determine the amount of hexose sugars. This assay was used for the quantification of rhamnolipid (3-deoxy-hexose) in the culture supernatant. A volume of the culture supernatant (10 to 300 μl) was diluted with water to attain a volume of 300 μl, which was extracted twice with 600 μl diethylether. The pooled ether fractions were evaporated to dryness, and the remainder was dissolved in 100 μl distilled water and mixed with 100 μl 1.6% orcinol and 800 μl of 60% sulfuric acid. After heating to 80°C and shaking at 175 rpm for 30 min, the OD₄₂₁ was measured. The content of rhamnose in the samples was determined by comparing with rhamnose standards with a defined concentration. Rhamnolipid concentration was calculated based on the assumption that 1 μg of rhamnose corresponds to 2.5 μg of rhamnolipid (43).

TLC.The extracted rhamnolipids were dissolved in chloroform-methanol (9:1) in the concentration of 10 mg/ml, and 2 μl of the sample was spotted onto thin-layer chromatography (TLC) plates (silica gel 60 F₂₅₄; Merck). After development in chloroform-methanol-acetic acid (65:15:2), the TLC plates were dipped in the reagent 15% H₂SO₄ in ethanol and incubated for 2 to 5 min at 100°C.

Acyl-homoserine lactone (AHL) detection by an A. tumefaciens-based bioassay (9).Supernatants from P. aeruginosa stationary-phase cultures (500 μl) were extracted twice with 400 μl dichloromethane and dissolved in 5 μl of ethylacetate. Wells were punched in 1.5% agar plates with a sterile inoculating loop, and the extracts were filled in. After 15 min the ethylacetate was evaporated. Afterwards the plates were overlaid with 5 ml overnight culture of A. tumefaciens NTL4(pZLR4) in soft agar (soft ABG agar [0.7%] containing 40 μg 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside [X-Gal]). The plates were incubated overnight at 30°C.

Phenotypic characterization of an estA-negative P. aeruginosa mutant.An estA-negative mutant of P. aeruginosa PAO1 was constructed by deletion of an internal fragment of the estA structural gene and insertion of a resistance marker cassette. This construct was then cloned into a suicide vector, and the resulting plasmid was used for homologous recombination, resulting in allelic exchange of estA in P. aeruginosa. As expected the estA mutant showed no esterase activity in preparations of the outer membrane, and Western blot analysis demonstrated that the EstA protein was completely absent. Preliminary characterization of the P. aeruginosa esterase-negative mutant revealed no obvious physiological differences compared to the wild type concerning the growth behavior under different culture conditions and in different growth media (data not shown). This indicates a physiological role for EstA which is not essential under standard laboratory conditions. For complementation purposes and moderate overexpression of the wild-type estA gene in P. aeruginosa the broad-host-range plasmid pBBX+ (63) was used.

EstA is required for rhamnolipid production.EstA has been shown to possess esterase activity, but the natural substrate of the enzyme is presently not known as well as the physiological circumstances under which its function is relevant for P. aeruginosa. However, it was not unrealistic that EstA, as a surface-exposed esterase, might be somehow involved in hydrolysis of ester bonds in compounds present on the cell surface or in the culture medium. Taking into account that EstA, as an autotransporter protein, was suspected of having virulence-related functions, we tentatively suspected rhamnolipids of being potential substrates for EstA because rhamnolipids are virulence determinants secreted by P. aeruginosa. The major rhamnolipids of P. aeruginosa contain one or two rhamnose moieties linked to two units of decanoic acids connected by an ester bond and thus might be cleavable by EstA.

In order to test the assumption that EstA might somehow be involved in the turnover or modification of rhamnolipid, its extracellular concentration was analyzed. Therefore, cultures of wild-type P. aeruginosa PAO1 and the estA mutant were grown in PPGAS medium, which is known to promote rhamnolipid production (62). Culture supernatants were collected after 48 h of growth, and rhamnolipids were extracted and subjected to TLC analysis. Indeed, we could detect a clear difference in rhamnolipid production between the two strains. As shown in Fig. 1 and 2 the amounts of both mono- and dirhamnolipid were affected in the estA mutant compared to the wild type. Unexpectedly, the estA mutant was completely unable to produce both major rhamnolipids (Fig. 2), which is contradictory to the assumption that inactivation of a potential rhamnolipid-hydrolyzing enzyme should result in accumulation of this compound. However, these findings suggested a potential role for EstA in rhamnolipid production rather than in its turnover by hydrolysis although its physiological role in this process was not obvious. The biosynthesis pathway of rhamnolipid is largely characterized, and there is no obvious requirement for an esterase activity in this pathway (56). At present it is completely unknown how rhamnolipids are secreted into the environment, and it can be speculated that this process may be mediated by a pore-forming protein of the outer membrane.

• Open in new tab
• Download powerpoint

FIG. 1.

Amounts of rhamnolipids in culture supernatants of different P. aeruginosa PAO1 strains. The amounts of rhamnolipids in culture supernatants were determined by an indirect assay (orcinol test). Wild-type (wt) P. aeruginosa PAO1, the estA mutant, and mutants overexpressing estA from plasmid pBBX+ and pBBXmut (inactive variant) were compared, and the rhlA mutant and strains containing the empty vector pBBR1MCS (V) served as controls. Culture supernatants were collected after 48 h of growth. The results were plotted into two diagrams because of the significant differences in the rhamnolipid amounts. Standard deviations were derived from three independent experiments.

• Open in new tab
• Download powerpoint

FIG. 2.

TLC separation of extracellular and cell-bound rhamnolipids from P. aeruginosa. P. aeruginosa PAO1 (wild type [wt]) and the estA-negative mutant (estA-) were examined for rhamnolipid production. Rhamnolipids were extracted from cultures which were grown for 48 h in PPGAS medium. The rhamnolipid standard (RL) was purified from P. aeruginosa. Rhamnolipids were separated by TLC and stained with a reagent containing orcinol and sulfuric acid in ethanol.

EstA belongs to the autotransporter protein family and is composed of two distinct domains, of which the N-terminal domain harbors the enzymatic activity whereas the C-terminal transporter domain represents a beta-barrel pore in the outer membrane. Although it has not been reported that substrates other than the corresponding passenger domain itself can be secreted across the outer membrane by these domains, one can speculate that the pore-forming domain of the EstA transporter domain might serve as a translocation pore for rhamnolipid.

In order to distinguish whether the pore-forming C-terminal translocator domain of EstA might be the relevant part of EstA for rhamnolipid production and responsible for the rhamnolipid-negative phenotype of the estA mutant rather than the enzyme activity of the passenger domain, we constructed an inactive variant of EstA. The predicted active-site serine₃₈ residue (63) was replaced by an alanine using site-directed mutagenesis. The loss of enzyme activity of the EstA variant (EstA*) was confirmed by expression from plasmid pBBXmut in the estA-negative mutant and subsequent activity measurement, thereby demonstrating that in fact serine₃₈ is the active-site serine. In addition, expression of EstA* and correct localization of the protein in the membrane were verified by Western blot analysis and found to be unaffected by the mutation (data not shown). In addition we know also from our work that even in the heterologous host E. coli another inactive EstA* variant resides on the cell surface (5).

The amount of rhamnolipids produced by the strains was determined by the orcinol assay (10). The cultures were grown as described above; culture supernatants were collected, and rhamnolipids were extracted. The data were corrected according to the minimal variation of cell densities. The results of these experiments shown in Fig. 1 revealed a significant difference in the amounts of rhamnolipids between wild-type, estA-negative, and estA-overexpressing strains. The rhlA-negative strain is known to be completely defective in rhamnolipid synthesis and thus served as a negative control. The rhamnolipid production of the estA mutant could be restored by complementation with the plasmid pBBX+, harboring the wild-type estA gene. In contrast, complementation failed when the inactive estA variant was expressed from the same plasmid (pBBXmut). Interestingly pBBX+ not only restored rhamnolipid production of the mutant strain but also significantly increased the amount of rhamnolipid produced when it was introduced into the wild type. These data indicate that rhamnolipid production of P. aeruginosa essentially depends on EstA, which appears to play the role in rhamnolipid biosynthesis of limiting the yield of rhamnolipids in the wild-type background. Moreover, the failure of the inactive variant in complementation of EstA-dependent rhamnolipid production demonstrates that the enzymatic activity of EstA in fact is the relevant function of the protein for rhamnolipid production. This is consistent with the finding that no accumulation of rhamnolipids within the cell was observed in the estA mutant (Fig. 2).

The production of rhamnolipid in P. aeruginosa is regulated by a quorum-sensing system involving AHLs as signal molecules or “autoinducers.” Riedel and coworkers (51) found that inactivation of an EstA homologue in Serratia liquefaciens resulted in reduced levels of AHLs. However, this effect was observed only under specific artificial experimental conditions, namely, when the growth medium was supplemented with unusually high concentrations of Tween 20 as artificial lipid substrates, whereas it remained uncertain without lipid supplementation (51). In the experiments presented in this report the bacteria were grown without any lipid supplementation. Under these conditions the estA-negative mutant produced signal molecules which could be detected by the A. tumefaciens-based bioassay (Fig. 3). In this bioassay AHLs are detected using a reporter strain which harbors an AHL-responsive promoter fused to the β-galactosidase gene (lacZ::traG), which converts X-Gal, resulting in formation of blue halos in the presence of autoinducers (9). Compared to supernatants of the wild type, those of the estA-negative mutant raised halos of the same size and thus contained equal amounts of AHLs. In contrast supernatants of the lasRI rhlRI quorum-sensing mutant, which served as a negative control, produced no visible halo in the bioassay. This demonstrates that the production of the AHLs detected by the A. tumefaciens bioassay is unaffected in the estA mutant and that the EstA-dependent effects on rhamnolipid production are not the result of alterations of these AHL levels.

• Open in new tab
• Download powerpoint

FIG. 3.

Agrobacterium tumefaciens-based bioassay for autoinducer (AI) detection in culture supernatants of P. aeruginosa. Tested were wild-type (WT) P. aeruginosa PAO1 and the estA-negative mutant for the presence of AIs, and an AI-deficient mutant (rhlRI and lasRI deficient) served as a negative control. Shown is the A. tumefaciens-based bioassay, where the formation of blue halos confirmed the presence of AI molecules.

Motility and biofilm formation of the estA mutant.Some phenotypes of P. aeruginosa which are physiologically connected to rhamnolipid production were investigated in the estA mutant. Rhamnolipid production has been reported to promote surface translocation on semisolid surfaces, i.e., swarming motility is facilitated by reduced friction through the action of rhamnolipid detergent (32). In fact, as expected, the esterase-negative mutant strain showed no swarming motility (Fig. 4), which is consistent with the lack of rhamnolipid production.

• Open in new tab
• Download powerpoint

FIG. 4.

Swimming (A), swarming (B), and twitching (C) motility of P. aeruginosa PAO1. Wild-type P. aeruginosa (wt) and estA-negative mutant (estA-) strains were analyzed on agar plates. In addition the estA-negative mutant was transformed with different plasmids, i.e., the empty vector (v) as a negative control, the estA-carrying plasmid pBBX+, and a mutated version of pBBX coding for an inactive esterase (pBBXmut). The agar plates contained different agar concentrations to establish swimming (0.3%) and swarming (0.5%) motility and were incubated overnight at room temperature (swimming) or at 30°C (swarming plus twitching). (D) Motility of the P. aeruginosa estA mutant with the addition of exogenous rhamnolipids (RLs) and the wild type was monitored. Extracted RL was added to the agar plates at a concentration of 0.4 g/liter, which represents the wild-type level of RLs.

Swarming motility of P. aeruginosa depends on the functionality of both type IV pili and flagella (32). To further dissect the swarming motility deficiency phenotype of the EstA mutant, twitching motility and swimming were also analyzed. Interestingly, in addition to swarming, both other forms of surface motility were completely absent in the estA mutant. Motility could not be restored by addition of wild-type amounts of rhamnolipid to the growth medium of the estA mutant, suggesting that the rhamnolipid deficiency is not the direct cause of the nonmotile phenotype (Fig. 4D).

In biofilm formation the ability of bacterial cells to move on surfaces plays an important role in achieving the mature three-dimensional biofilm architecture (29, 30). Since the esterase mutant strain appeared to be unable to display any type of motility, we suspected that the formation or the structure of biofilms might be different from that of wild-type biofilms.

Indeed, differences in biofilm formation and maintenance were observed with this strain. Compared to that of the wild type, the estA mutant biofilms were characterized by a significant reduction in surface coverage. The mutant strain appeared to form biofilms consisting of large aggregates of cells with regions of extremely low cell densities between the individual microcolonies, as expected for nonmotile strains that can expand only by clonal growth (Fig. 5).

• Open in new tab
• Download powerpoint

FIG. 5.

Biofilm confocal laser scanning micrographs of different P. aeruginosa strains. P. aeruginosa PAO1, the estA mutant, and mutants overexpressing estA from plasmids pBBX+ and pBBXmut (inactive variant) were compared, and the strain containing the empty vector pBBR1MCS (V) served as a control. Biofilms were grown in a flow cell, and section pictures were taken after 24 and 72 h of growth with ×630 magnification.