Rhamnolipid Surfactant Production Affects Biofilm Architecture in Pseudomonas aeruginosa PAO1
Department of Microbiology and Immunology, Dartmouth Medical School, Hanover, New Hampshire 03755q/#s*, 百拇医药
Received 10 June 2002/ Accepted 6 October 2002q/#s*, 百拇医药
ABSTRACTq/#s*, 百拇医药
In response to certain environmental signals, bacteria will differentiate from an independent free-living mode of growth and take up an interdependent surface-attached existence. These surface-attached microbial communities are known as biofilms. In flowing systems where nutrients are available, biofilms can develop into elaborate three-dimensional structures. The development of biofilm architecture, particularly the spatial arrangement of colonies within the matrix and the open areas surrounding the colonies, is thought to be fundamental to the function of these complex communities. Here we report a new role for rhamnolipid surfactants produced by the opportunistic pathogen Pseudomonas aeruginosa in the maintenance of biofilm architecture. Biofilms produced by mutants deficient in rhamnolipid synthesis do not maintain the noncolonized channels surrounding macrocolonies. We provide evidence that surfactants may be able to maintain open channels by affecting cell-cell interactions and the attachment of bacterial cells to surfaces. The induced synthesis of rhamnolipids during the later stages of biofilm development (when cell density is high) implies an active mechanism whereby the bacteria exploit intercellular interaction and communication to actively maintain these channels. We propose that the maintenance of biofilm architecture represents a previously unrecognized step in the development of these microbial communities.
INTRODUCTIONl4!2h, 百拇医药
Although bacteria are commonly viewed as solitary life forms, these organisms are more typically colonial creatures. In their natural settings, bacteria persist within microbial communities, where they exploit elaborate systems of intercellular interaction and communication to adjust to changing environmental parameters. Moreover, biofilm formation has also been linked to the emergence of a variety of opportunistic human pathogens (5). For example, organisms such as Staphylococcus epidermidis and Pseudomonas aeruginosa form biofilms on implants and dead or living tissue, thereby contributing to a variety of persistent infections.l4!2h, 百拇医药
Thinking about bacterial populations as connected organisms capable of concerted multicellular activities has provided researchers with novel insights into microbial biology. The key to such multicellular behavior lies in the ability of each individual cell to sense and respond to information from nearby cells, and this behavior requires a certain population size or quorum of cells. The development of biofilms is a process that involves both a quorum of cells and multicellular behavior (6). Single-species biofilms are of particular interest due to their clinical importance and the monospecies biofilms formed by P. aeruginosa has become a prominent model for studying this aspect of microbial biology.
The complex structure of microbial biofilms has only recently been determined. Detailed analysis by scanning confocal laser microscopy has shown that biofilms of P. aeruginosa formed on solid surfaces and exposed to a continuous flow of fresh nutrients are open, highly hydrated structures consisting of cells embedded in an extracellular matrix filled with large void spaces (14). These void spaces, or channels, allow fluids to flow throughout the biofilm, resulting in the distribution of nutrients and oxygen. In addition, the channels between macrocolonies may also provide a means of removing metabolic end products.gk\!, 百拇医药
It has been hypothesized that open-channel formation is not a stochastic process but instead represents an active process that creates a preferable structural design. For example, under some conditions P. aeruginosa mutants unable to make quorum-sensing molecules initiate biofilm formation but are unable to make normally structured biofilms (7). Other P. aeruginosa mutants, including those deficient in flagellum and pilus synthesis, as well as those with mutations in global regulators such as gacA and crc, have defects in the initial colonization of a surface (9, 22, 24, 27). Taken together, these data suggest that formation of the distinctive architecture of P. aeruginosa biofilms, including the macrocolonies surrounded by fluid-filled channels, is a regulated developmental process requiring distinct genetic surface structures and regulatory elements (23, 29, 30, 35). However, mechanisms by which the bacteria maintain these channels once they form have not been investigated. Our data indicate P. aeruginosa not only regulates development of its distinctive biofilm architecture but, once channels form, this organism utilizes rhamnolipid surfactants to actively maintain the void spaces surrounding macrocolonies. That is, we propose that rhamnolipids are not required for the formation of macrocolonies and channels but participate in the maintenance of channels once they are formed.
MATERIALS AND METHODS(a!3a6, http://www.100md.com
Bacterial strains, plasmids, and media. P. aeruginosa PAO1 was the wild-type strain used in the present study (13), and the rhlA::Tn5 mutant strain was described previously (19). Plasmid pSMC21 (GFP+ Apr Knr) constitutively expresses gfp and was used to label the strains for microscopy studies (1), and plasmid pMH520 contains a rhlA::gfp transcriptional fusion (a gift from Matt Parsek, Northwestern University). All flow cell experiments were performed in EPRI medium as described previously (7), except that morpholinepropanesulfonic acid served as the buffer (2.2%) and the medium was supplemented with phosphate (0.0019% KH2PO4, 0.0063% K2HPO4). Dirhamnolipids were purified from the supernatant of P. aeruginosa as described previously (37) and were a gift of Thea Norman at Microbia, Inc. (Cambridge, Mass.).(a!3a6, http://www.100md.com
Microtiter dish assay for biofilm formation and flow chamber experiments. For the biofilm formation assay, M63 medium (100 µl/well) containing arginine (0.05%) and MgSO4 (100 µM) is inoculated (1:50) with an overnight L-broth-grown culture as described previously (25). The microtiter plates were then incubated at 37°C for the times indicated. The biofilm is macroscopically visualized by addition of 100 µl of a 0.1% solution of crystal violet to each well. Biofilm formation was quantified by the addition of 125 µl of 100% ethanol to each crystal violet-stained microtiter dish well; after 10 min 100 µl was transferred to a new microtiter dish, and the absorbance was determined with a plate reader at 550 nm.
Biofilms were also cultivated in flow chambers with channel dimensions of 5 by 1 by 30 mm. The flow system was assembled as described previously (4). The flow cell was inoculated from overnight L-broth-grown cultures diluted 10-fold in EPRI medium (final concentration, 108/ml). The medium flow was turned off prior to inoculation and for 1 h after inoculation. Thereafter, medium was pumped through the flow cell at a constant rate (1.8 ml/h) for the duration of the experiment. The flow was controlled with a PumpPro MPL (Watson-Marlow).56itn, 百拇医药
Microscopy and staining. A Leica DM IRB inverted microscope (Leica Microsystems, Wetzlar, Germany) equipped with a cooled charge-coupled device digital camera and a x10 or x63 PL Flotar objective lens was used for epifluorescence and phase-contrast microscopy analyses. With this instrument, digital images were captured and processed by using a G4 Macintosh computer with OpenLab software package (Improvision, Coventry, England). The images were processed for publication by using Photoshop software (Adobe, Mountain View, Calif.). Quantitative analysis of the flow cell-grown biofilms was performed with the COMSTAT image analysis software package (11, 12).
Mixing experiments. Mixing experiments were performed by inoculating ~ 107 bacteria of both the wild type (carrying the pSMC21 gfp+ plasmid) and the rhlA mutant into the same channel of the flow cell. Biofilms were allowed to form for 6 days, and then the cells were stained with 5-cyano-2,3-di-4-tolyl-tetrazolium chloride (CTC; Polysciences, Inc., Warrington, Pa.), a metabolic stain that renders metabolically active cells red. Cells expressing green fluorescent protein (GFP) quenched the red fluorescence and thus allowed the wild type and the rhlA mutants to be distinguished. Channels inoculated with the wild type alone or with the rhlA mutant alone were included in each experiment as controls. No difference in architecture or staining was observed when the rhlA mutant carried the GFP-expressing plasmid (data not shown).0mg3%, 百拇医药
Invasion experiments. The wild-type strain was allowed to form a biofilm in a flow cell for 6 days. The biofilm was stained with CTC for 1 h in the absence of flow, and then flow resumed for 2 h to remove the residual dye. This preformed, stained biofilm was "invaded" with ~ 107 wild-type cells carrying the pSMC21 gfp+ plasmid, and the flow was stopped for 1 h. After the resumption of flow for an additional hour to remove the unattached bacteria, the biofilm was examined for red and green fluorescence, and the individual images were merged to form a composite of preformed biofilm cells and GFP-labeled invading cells.
RESULTS'#u#ds, 百拇医药
Rhamnolipids are required to maintain biofilm architecture. It was shown previously that the lasR-lasI quorum-sensing system in P. aeruginosa is involved in the later stages of biofilm differentiation, when cell density is sufficient to constitute a quorum (7). A biofilm produced by the lasI mutant is undifferentiated, i.e., the cells appear to grow as a continuous sheet on the surface lacking the characteristic water channels, indicating that P. aeruginosa has one (or more) systems to control the formation of macrocolony structures. Because lasI controls a whole suite of genes, the specific gene products contributing to this phenotype were unknown. One function controlled by the lasI-lasR system (indirectly via the rhlI-rhlR system) is rhamnolipid biosynthesis. Therefore, we investigated the biofilm formation phenotype of a mutant strain unable to synthesize rhamnolipids. The quorum-sensing-controlled rhlA gene codes for a rhamnosyltransferase whose only known function is in rhamnolipid synthesis. An rhlA mutant does not make rhamnolipids (20, 21).
To study the role rhamnolipids played in the formation of surface-attached communities, we monitored biofilm development of wild-type P. aeruginosa PAO1 and a rhlA-null mutant. Early biofilm formation was assayed in the microtiter dish assay. The wild-type strain forms a biofilm that peaks at 8 h and then declines at 24 h . Previous studies indicated that the P. aeruginosa biofilm formed in a 96-well microtiter dish declines after ca. 10 h of incubation, probably due to starvation in this batch culture system (25). In contrast, the biomass of the rhlA mutant biofilm continued to increase over this 24-h period to a final A550 value twofold greater than the maximum observed for the wild type. There was no discernible difference in the planktonic growth of these two strains. These data suggest that the rhlA mutant is more proficient at early colonization than is the wild-type strain.?*dul2, 百拇医药
fig.ommitted?*dul2, 百拇医药
A rhamnolipid mutant produces biofilms lacking characteristic architecture. (A) Biofilm formation by the wild type and rhlA mutant in microtiter dishes. The biofilm formation phenotype of the wild type (solid bars) and rhlA mutant (open bars) was quantitated over 24 h by using the microtiter dish assay, as reported elsewhere (26). The A550 value represents crystal violet-stained bacteria attached to the walls of the microtiter dish and is an indirect measure of the biofilm formed. The data represent two experiments, each performed in triplicate. (B) Flow cell architecture of mature biofilms. The wild-type and rhlA::Tn5 strains carrying plasmid pSMC21, which constitutively expresses GFP, were inoculated into flow cells fed by minimal salts EPRI medium supplemented with glucose as the carbon and/or energy source. The top-down views shown here were acquired by epifluorescent microscopy by using the x63 objective lens, which was used to monitor biofilm development in flow cells over 6 days. The large green macrocolonies are surrounded by dark areas, which are the sparsely colonized channel regions. At day 4, there is no difference between the wild type and the rhlA mutant, thus demonstrating that rhamnolipids are not required for the formation of macrocolonies. For the rhlA mutant, partial filling of the channels can be observed by day 5 and the channels are completely filled in this strain by day 6. The wild-type strain at day 6 displays the characteristic architecture of a mature P. aeruginosa biofilm. Scale bars are included and labeled on the figure. (C) Flow cell architecture of mature biofilms visualized under low magnification. This panel shows a day 6 biofilm obtained at low magnification (with a x10 objective lens versus the x63 objective lens used for the images in panel B). Scale bars are included and are labeled on the figure.
We followed further biofilm maturation over the course of 6 days by using a once-through flow cell system that provides a continuous supply of fresh nutrients to the biofilm. To monitor the bacteria by epifluorescence microscopy, the strains were transformed with plasmid pSMC21, which constitutively expresses the GFP. We had shown previously that this plasmid is maintained in the absence of antibiotic selection and confers no detectable metabolic load on the cells (1).^, 百拇医药
Macrocolony formation by the wild-type and rhlA strains appeared to be identical after 4 days of incubation . Both strains formed well-defined fluorescent (green) colonies separated by dark, fluid-filled channels. This observation indicates that the small advantage gained by the rhlA mutant in initial colonization in a microtiter dish assay does not translate to an observable phenotype in the architecture of the 4-day-old flow cell-grown biofilm. By day 6, wild-type architecture was comprised of characteristic macrocolonies surrounded by open channels. The rhlA mutant, however, produced a biofilm that was a thick, uniform mat of bacterial cells . Visualizing the biofilm at a lower magnification confirmed the difference in architecture . These data indicated that the rhlA mutant was able to form the channels surrounding the macrocolonies but was unable to maintain these channels.
Quantitative analysis of biofilm architecture. Visual inspection of biofilms produced by wild-type and the rhamnolipid mutant at day 6 indicated an altered biofilm structure for the mutant. To confirm this observation, we applied the COMSTAT image analysis program to perform a quantitative analysis of biofilm architecture (11, 12). As shown in , four variables (mean thickness, substratum coverage, roughness coefficient, and surface/volume ratio) were used to evaluate biofilm architecture. The mean thickness of the biofilm and the percent substratum coverage were much higher in the biofilms formed by the rhamnolipid mutant, a finding consistent with the images shown in. The roughness coefficient was much lower in the mutant, demonstrating that biofilms produced in the absence of rhamnolipids are much less heterogeneous. In addition, the surface/volume ratio was significantly lower in the biofilm produced by the mutant, indicating that a smaller fraction of the biomass is exposed to the nutrient flow. These data confirm the visual observation that the rhlA mutant produces a biofilm consisting of a thick, uniform mat of bacterial cells.
fig.ommittedlk;v\y$, http://www.100md.com
Quantitative analysis of biofilm architecturelk;v\y$, http://www.100md.com
Partial rescue of the rhlA biofilm architecture defect by the wild-type strain. If the defect in biofilm development is the result of a lack of rhamnolipid production, it follows that providing rhamnolipids should restore wild-type biofilm architecture. Rhamnolipids are likely produced by a certain subset of cells in response to specific environmental cues, resulting in surfactant production at a specific time and place, as has been reported elsewhere (8). Consistent with this idea, providing spent supernatants of P. aeruginosa cultures was insufficient to rescue the architecture defect of a rhlA mutant (data not shown). As an alternative approach to test the ability of rhamnolipids to rescue the phenotype of a rhlA mutant, we monitored biofilm development in a flow cell inoculated with a mixture of equal numbers of wild-type PAO1 (green cells) and the rhlA::Tn5 mutant (red cells). As seen in, this mixing experiment resulted in a biofilm that consisted of macrocolonies made up in large part of a mixture of wild-type and rhlA mutant bacteria. Despite the presence of wild-type bacteria in many macrocolonies, the resulting biofilm is a mixture of typical wild-type architecture (comprised of macrocolonies surrounded by open areas) and dense compact areas where the channels were occupied by cells, a phenotype that is reminiscent of the rhlA mutant. These data indicate that the presence of the wild type could only partially rescue the biofilm architecture defect of the rhlA mutant.
fig.ommitted)63t.ne, http://www.100md.com
Partial rescue of the rhlA mutant architecture phenotype by coculture with the wild-type strain. A mixing experiment was done to evaluate the ability of the wild type to rescue the phenotype of the rhlA mutant. The image is a top-down epifluorescent view of a 6-day-old biofilm formed in a flow cell where wild-type cells was mixed 50:50 with the rhlA mutant obtained with a x10 objective lens. The rhlA mutants cells were visualized by CTC staining (red cells, see Materials and Methods), and the wild-type strain expresses the GFP from a plasmid (green cells). Regions of overlapping wild-type and mutant bacteria are yellow.)63t.ne, http://www.100md.com
Expression of rhlA during biofilm development. To evaluate where and when rhamnolipids are synthesized in the biofilm, we examined the expression of the rhlA gene during development. Expression of rhlA was determined indirectly by using plasmid prhlA::gfp, which harbors a transcriptional fusion of the rhlA promoter region to the GFP. As shown in, on day 1, small microcolonies of 10 to 30 cells associated with the early stages in biofilm development can be seen by phase-contrast microscopy; however, no rhlA-gfp-dependent fluorescence was observed at this time point. However, by day 2, larger colonies comprised of thousands of fluorescent cells could be detected, indicating expression of the rhlA gene. Expression of this fusion was found to coincide with macrocolony formation and continued throughout biofilm development in areas where the cell density was high. The expression of rhlA only in macrocolonies is consistent with the observation that rhamnolipids are not required for the initiation of biofilm development. In control experiments, we could observe the fluorescence of individual batch grown planktonic, stationary-phase cells carrying the prhlA-gfp plasmid (not shown). These data demonstrate that the digital camera used to obtain images was sufficiently sensitive to detect an individually fluorescing bacterium.
fig.ommitted]o9l|, 百拇医药
Expression pattern of an rhlA::gfp transcriptional fusion during biofilm formation. The images are top-down views of individual cells and colonies at day 1 (A) and day 2 (B) for flow cell-grown bacteria (obtained with the x63 objective lens). Both phase-contrast and fluorescent images of the same view are shown. Small clusters of bacteria produced no detectable fluorescence; however, larger colonies (high cell density) resulted in expression of the rhlA::gfp fusion.]o9l|, 百拇医药
Overproduction of rhamnolipids inhibits biofilm development. Based on their surface-active properties, we predicted that rhamnolipids could affect the interactions of bacteria with surfaces and with each other. To address this hypothesis directly, we examined the effects of rhamnolipid overexpression on the interactions of bacteria with a surface. First, the effects of rhamnolipids on early biofilm formation were determined by growing bacteria under conditions leading to high-level expression of rhamnolipids. Rhamnolipids are typically synthesized during late-exponential or stationary phase of growth when the cell density is high, and this production is greatly enhanced by nutrient limitation (10). However, it has been shown previously that growth in a phosphate-limited complex medium induces the production of high levels of rhamnolipids (~ 100 µM in planktonic cultures) in the absence of high cell density (17). The formation of biofilms was observed in flow cells fed with phosphate-limited PPGAS medium (~ 0.03 mM inorganic phosphate, low-phosphate conditions). As shown in , when grown in low-phosphate, rhamnolipid-inducing conditions, wild-type cells started to attach at day 1, but by day 3 very few bacteria remained attached to the surface of the flow cell. In contrast, the rhlA mutant, which is unable to produce any rhamnolipids, formed a dense biofilm on the surface by day 3. Supplementing the PPGAS complex medium with 2.5 mM potassium phosphate (high-phosphate conditions) restored the ability of the wild-type strain to form a biofilm in the flow cell (data not shown). Therefore, expressing high levels of rhamnolipids can impede the formation of biofilms.
fig.ommitted'jp, 百拇医药
Effect of rhamnolipid overexpression on biofilm formation. The images are fluorescent top-down views of flow cell-grown biofilms formed by wild type and the rhamnolipid mutant at days 1 and 3. PPGAS medium, a low-phosphate medium, induced the synthesis of high levels of rhamnolipids in the wild-type strain and resulted in a sparsely colonized surface. The rhlA mutant was unable to synthesize any rhamnolipids even in low-phosphate medium and made a dense biofilm by day 3.'jp, 百拇医药
Rhamnolipids disrupt both cell-to-cell and cell-to-surface interactions. The data presented above indicated that production of rhamnolipids could help maintain the channels formed in a biofilm. How can rhamnolipids carry out such a function? As mentioned above, rhamnolipids are surfactants, or surface-active molecules, and a general property of surfactants involves their ability to alter the physical and/or chemical properties at interfaces (18). We hypothesized, therefore, that the rhamnolipid surfactants could influence P. aeruginosa cell-to-cell and/or cell-to-surface interactions.
To test these hypotheses, purified rhamnolipids were assessed for their effects on the initiation of biofilm development by using the previously described microtiter dish assay. Purified rhamnolipid (250 µM) added to bacteria at the time of inoculation into the microtiter dishes completely blocked biofilm formation. Partial inhibition of biofilm formation was observed at lower concentrations of rhamnolipid (not shown).998e^|{, 百拇医药
fig.ommitted998e^|{, 百拇医药
Effect of purified rhamnolipids on cell-surface and cell-cell interactions. (Upper panels) Cell-surface interactions. The ability to form a biofilm on polyvinylchloride plastic in the presence or absence of rhamnolipids was examined. The rhamnolipid was added at the time of inoculation, and the biofilm formation was examined after 8 h of incubation. The addition of 250 µM rhamnolipid completely blocked biofilm formation (right panel). (Lower panels) Cell-cell interactions. The effect of rhamnolipids on rafts or pellicles of wild-type P. aeruginosa cells was examined. The disruption of cell-to-cell interactions caused by adding rhamnolipids was captured by phase-contrast microscopy immediately after addition of the compound (right panel). The addition of the same volume of water had no effect on biofilm formation or disruption of the rafts (left panels).
We also examined the effect of purified rhamnolipids on cell-to-cell interactions. P. aeruginosa typically forms a pellicle or raft of cells on the surface of a liquid culture. As shown in , the addition of 100 µM dirhamnolipid disrupted cell-to-cell interactions, and this disruption occurred immediately after the addition of the compound. Taken together, these data demonstrate that rhamnolipids can disrupt both cell-to-cell and cell-to-surface interactions. Interestingly, neither purified rhamnolipid nor rhamnolipid-containing spent supernatants of P. aeruginosa cultures had any discernible effect on a preformed 6-day-old biofilm (not shown). These data suggest that rhamnolipids can only interfere with earlier stages of biofilm development.48c2v/, 百拇医药
Invasion of a preformed biofilm. Based on the evidence that biofilm channel structure is maintained over time, we predicted that few planktonic cells invading a preformed biofilm would be capable of attaching to this community. We performed an invasion experiment by introducing GFP-tagged P. aeruginosa cells (~ 107 cells) into a flow cell containing a preformed mature 5-day-old biofilm, and the tagged cells were allowed to attach for 1 h under static conditions . After this period of attachment, flow was resumed for an additional hour to remove unattached bacteria before microscopic examination at low and high magnifications. As shown in, very few of the invading GFP-tagged cells attached to the biofilm. These data are consistent with the hypothesis that preformed biofilms have the means, including but not necessarily limited to the production of rhamnolipids, to control the attachment of planktonic bacteria.
fig.ommitted3, 百拇医药
Biofilm invasion assay. A biofilm of non-GFP-labeled wild-type bacteria was allowed to form for 5 days in a flow cell and then stained with CTC. (a and c) The flow was turned off, ca. 107 GFP-labeled wild-type P. aeruginosa cells were injected into the flow cell, and the flow chamber was incubated without flow for 1 h. (b and d) Flow was resumed for 1 h to remove planktonic bacteria, and the biofilm was examined by fluorescence microscopy. A merged composite image of the fluorescence micrographs is shown. The images in panels a and b were captured by using a x63 objective lens, and the images in panels c and d were captured by using a x10 objective lens. Scale bars are included in the images.3, 百拇医药
DISCUSSION3, 百拇医药
Our studies show that in situ production of rhamnolipid surfactants by P. aeruginosa cells affects biofilm architecture but that the loss of rhamnolipid production does not appear to affect the initial development of macrocolonies and open channels. However, the rhlA mutant is incapable of maintaining open channels, eventually forming a relatively homogeneous layer of cells as thick as, or thicker than, the wild-type biofilm . Such a role for surface-active compounds in the structuring of microbial populations at interfaces is not unprecedented in the microbial world. Surfactants are known to play a role in the emergence of aerial hyphae in both fungi and the filamentous bacterium Streptomyces coelicolor (31, 36) and in the formation of hydrophobic air channels in fruiting bodies of fungi (15). In addition, a recent study of fruiting body formation in the bacterium Bacillus subtilis showed that a mutant (sfp) deficient in surfactin production formed atypical surface-associated pellicle (biofilm) structures (2).
How does the production of rhamnolipids maintain the open channels formed during biofilm development? At this point, the answer to this question is not clear. As has been reported, high cell density planktonic growth induces the synthesis of quorum-sensing-dependent rhamnolipid production (3, 21, 28), and similar density-dependent transcription of the RhlI-RhlR system and rhlA is also seen in biofilms (8) . We propose that surfactant production may inhibit planktonic cells from attaching to the preformed biofilm. This contention is supported by the "invasion assay" shown in . Alternatively, rhamnolipids may cause the detachment of cells or microcolonies from the biofilm, preventing the accumulation of new biomass in the channels. These models are not mutually exclusive. Finally, our data indicate that exposing a preformed biofilm to rhamnolipids has no significant effect on established biofilm structure, suggesting that the production of surfactants by preformed macrocolonies would not necessarily cause wholesale detachment of the biofilm.
The ability to block colonization of a preformed biofilm is likely due in large part to the well-known properties of surfactants at interfaces. Surfactants have the potential to affect both cell-to-cell and cell-to-surface interactions (18), and our data support the hypothesis that rhamnolipids produced by P. aeruginosa do affect these interface interactions. We showed that purified rhamnolipids, the only characterized surfactants produced by P. aeruginosa, are capable of disrupting interactions between cells and each other and a surface. Hence, rhamnolipids are able to modulate cell-to-surface and cell-to-cell interactions. The ability of surfactants to disrupt bacterial attachment has been observed for the surfactant produced by Lactobacillus spp., known as surlactin, which blocked initial surface adhesin of uropathogenic Enterococcus faecalis, Escherichia coli, and Staphylococcus epidermidis, as well as two yeast strains (32-34). The surfactant produced by B. subtilis, called surfactin, also inhibits biofilm formation by Salmonella enterica, E. coli, and Proteus mirabilis (16). Studies in our laboratory indicate that purified P. aeruginosa rhamnolipids can block initial adherence by fluorescent pseudomonads and E. coli (M. E. Davey and G. A. O'Toole, unpublished data), suggesting that surfactants may have a broad and relatively nonspecific ability to interfere with cell-to-cell and cell-to-surface interactions. Therefore, P. aeruginosa rhamnolipid may prevent colonization not only of "self" but also of other planktonic microbes attempting to take up residence within the channels of its biofilm.
The findings presented here indicate that open-channel maintenance is not a stochastic process but represents an active process by the bacterial community that stabilizes a preferable biofilm architecture. Recent studies in P. aeruginosa suggest a role for a number of genes in the formation of biofilm architecture, indicating that a distinct set of genetic factors is required for the development of biofilm architecture in P. aeruginosa. We suggest that the production of rhamnolipid surfactants is one mechanism employed by P. aeruginosa for the active maintenance of biofilm architecture. We believe that other, as-yet-unidentified mechanisms may also contribute to the maintenance of architecture. For example, a decrease or cessation of growth in large macrocolonies may slow or stop the filling of channels between these structures. Finally, we propose that the maintenance of biofilm architecture is a previously undescribed stage in the development of P. aeruginosa biofilms.ay9@[, 百拇医药
ACKNOWLEDGMENTS
We thank Thea Norman at Microbia, Inc., for providing purified dirhamnolipid, Urs Ochsner for the rhlA::Tn5 strain, Matt Parsek for the rhlA-gfp fusion plasmid, and John P. Connolly for computer expertise.4q, http://www.100md.com
This research was funded by grants from Microbia, Inc., and The Pew Charitable Trusts (to G.A.O.). G.A.O is a Pew Scholar in the Biomedical Sciences. M.E.D. was supported by an NIH training grant (5 T32 AI07519).4q, http://www.100md.com
REFERENCES4q, http://www.100md.com
Bloemberg, G. V., G. A. O'Toole, B. J. J. Lugtenberg, and R. Kolter. 1997. Green fluorescent protein as a marker for Pseudomonas spp. Appl. Environ. Microbiol. 63:4543-4551.4q, http://www.100md.com
Branda, S. S., J. E. Gonzalez-Pastor, S. Ben-Yehuda, R. Losick, and R. Kolter. 2001. Fruiting body formation by Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98:11621-11626.4q, http://www.100md.com
Brint, J. M., and D. E. Ohman. 1995. Synthesis of multiple exoproducts in Pseudomonas aeruginosa is under the control of RhlR-RhlI, another set of regulators in strain PAO1 with homology to the autoinducer-responsive LuxR-LuxI family. J. Bacteriol. 177:7155-7163.
Christensen, B. B., C. Sternberg, J. B. Andersen, R. J. Palmer, Jr., A. T. Nielsen, M. Givskov, and S. Molin. 1999. Molecular tools for study of biofilm physiology. Methods Enzymol. 310:20-42.qx, 百拇医药
Costerton, J. W., P. S. Stewart, and E. P. Greenberg. 1999. Bacterial biofilms: a common cause of persistent infections. Science 284:318-322.qx, 百拇医药
Davey, M. E., and G. O. O'Toole. 2000. Microbial biofilms: from ecology to molecular genetics. Microbiol. Mol. Biol. Rev. 64:847-867.qx, 百拇医药
Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298.qx, 百拇医药
De Kievit, T. R., R. Gillis, S. Marx, C. Brown, and B. H. Iglewski. 2001. Quorum-sensing genes in Pseudomonas aeruginosa biofilms: their role and expression patterns. Appl. Environ. Microbiol. 67:1865-1873.qx, 百拇医药
Drenkard, E., and F. M. Ausubel. 2002. Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature 416:740-743.
Guerra-Santos, L., O. Käppeli, and F. Fiechter. 1986. Dependence of Pseudomonas aeruginosa continuous culture biosurfactant production on nutritional and environmental factors. Appl. Microbiol. Biotechnol. 24:443-448.j, 百拇医药
Heydorn, A., B. K. Ersboll, M. Hentzer, M. R. Parsek, M. Givskov, and S. Molin. 2000. Experimental reproducibility in flow-chamber biofilms. Microbiology 146:2409-2415.j, 百拇医药
Heydorn, A., A. T. Nielsen, M. Hentzer, C. Sternberg, M. Givskov, B. K. Ersboll, and S. Molin. 2000. Quantification of biofilm structures by the novel computer program COMSTAT. Microbiology 146:2395-2407.j, 百拇医药
Holloway, B. W., V. Krishnapillai, and A. F. Morgan. 1979. Chromosomal genetics of Pseudomonas. Microbiol. Rev. 43:73-102.j, 百拇医药
Lawrence, J. R., D. R. Korber, B. D. Hoyle, J. W. Costerton, and D. E. Caldwell. 1991. Optical sectioning of microbial biofilms. J. Bacteriol. 173:6558-6567.j, 百拇医药
Lugones, L. G., H. A. B. Wosten, K. U. Birkenkamp, K. A. Sjollema, J. Zagers, and J. G. H. Wessels. 1999. Hydrophobins line air channels in fruiting bodies of Schizophyllum commune and Agaricus bisporus. Mycol. Res. 103:635-640.
Mireles, J. R., II, A. Toguchi, and R. M. Harshey. 2001. Salmonella enterica serovar Typhimurium swarming mutants with altered biofilm-forming abilities: surfactin inhibits biofilm formation. J. Bacteriol. 183:5848-5854.71z@f, 百拇医药
Mulligan, C. N., G. Mahmourides, and B. F. Gibbs. 1989. The influence of phosphate metabolism on biosurfactant production by Pseudomonas aeruginosa. J. Biotechnol. 12:199-210.71z@f, 百拇医药
Neu, T. R. 1996. Significance of bacterial surface-active compounds in interaction of bacteria with interfaces. Microbiol. Rev. 60:151-166.71z@f, 百拇医药
Ochsner, U. A., A. Fiechter, and J. Reiser. 1994. Isolation, characterization, and expression in Escherichia coli of the Pseudomonas aeruginosa rhlAB genes encoding a rhamnosyltransferase involved in rhamnolipid biosurfactant synthesis. J. Biol. Chem. 269:19787-19795.71z@f, 百拇医药
Ochsner, U. A., A. K. Koch, A. Fiechter, and J. Reiser. 1994. Isolation and characterization of a regulatory gene affecting rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. J. Bacteriol. 176:2044-2054.
Ochsner, U. A., and J. Reiser. 1995. Autoinducer-mediated regulation of rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 92:6424-6428.^iddch], http://www.100md.com
O'Toole, G. A., K. A. Gibbs, P. W. Hager, P. V. Phibbs, Jr., and R. Kolter. 2000. The global carbon metabolism regulator Crc is a component of a signal transduction pathway required for biofilm development by Pseudomonas aeruginosa. J. Bacteriol. 182:425-431.^iddch], http://www.100md.com
O'Toole, G. A., H. Kaplan, and R. Kolter. 2000. Biofilm formation as microbial development. Annu. Rev. Microbiol. 54:49-79.^iddch], http://www.100md.com
O'Toole, G. A., and R. Kolter. 1998. Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol. Microbiol. 30:295-304.^iddch], http://www.100md.com
O'Toole, G. A., and R. Kolter. 1998. The initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signaling pathways: a genetic analysis. Mol. Microbiol. 28:449-461.^iddch], http://www.100md.com
O'Toole, G. A., L. A. Pratt, P. I. Watnick, D. K. Newman, V. B. Weaver, and R. Kolter. 1999. Genetic approaches to the study of biofilms. Methods Enzymol. 310:91-109.
Parkins, M. D., H. Ceri, and D. G. Storey. 2001. Pseudomonas aeruginosa GacA, a factor in multihost virulence, is also essential for biofilm formation. Mol. Microbiol. 40:1215-1226.54ocf, http://www.100md.com
Pearson, J. P., E. C. Pesci, and B. H. Iglewski. 1997. Roles of Pseudomonas aeruginosa las and rhl quorum-sensing systems in control of elastase and rhamnolipid biosynthesis genes. J. Bacteriol. 179:5756-5767.54ocf, http://www.100md.com
Sauer, K., and A. K. Camper. 2001. Characterization of phenotypic changes in Pseudomonas putida in response to surface-associated growth. J. Bacteriol. 183:6579-6589.54ocf, http://www.100md.com
Sauer, K., A. K. Camper, G. D. Ehrlich, J. W. Costerton, and D. G. Davies. 2002. Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. J. Bacteriol. 184:1140-1154.54ocf, http://www.100md.com
Tillotson, R. D., H. A. Wosten, M. Richter, and J. M. Willey. 1998. A surface active protein involved in aerial hyphae formation in the filamentous fungus Schizophillum commune restores the capacity of a bald mutant of the filamentous bacterium Streptomyces coelicolor to erect aerial structures. Mol. Microbiol. 30:595-602.
Velraeds, M. M., B. van de Belt-Gritter, H. C. van der Mei, G. Reid, and H. J. Busscher. 1998. Interference in initial adhesion of uropathogenic bacteria and yeasts to silicone rubber by a Lactobacillus acidophilus biosurfactant. J. Med. Microbiol. 47:1081-1085.*, 百拇医药
Velraeds, M. M., H. C. van der Mei, G. Reid, and H. J. Busscher. 1996. Inhibition of initial adhesion of uropathogenic Enterococcus faecalis by biosurfactants from Lactobacillus isolates. Appl. Environ. Microbiol. 62:1958-1963.*, 百拇医药
Velraeds, M. M., H. C. van der Mei, G. Reid, and H. J. Busscher. 1997. Inhibition of initial adhesion of uropathogenic Enterococcus faecalis to solid substrata by an adsorbed biosurfactant layer from Lactobacillus acidophilus. Urology 49:790-794.*, 百拇医药
Whiteley, M., M. G. Bangera, R. E. Bumgarner, M. R. Parsek, G. M. Teitzel, S. Lory, and E. P. Greenberg. 2001. Gene expression in Pseudomonas aeruginosa biofilms. Nature 413:860-864.*, 百拇医药
Wosten, H. A., M. Richter, and J. M. Willey. 1999. Structural proteins involved in emergence of microbial aerial hyphae. Fungal Genet. Biol. 27:153-160.*, 百拇医药
Zhang, Y., and R. M. Miller. 1992. Enhanced octadecane dispersion and biodegradation by a Pseudomonas rhamnolipid surfactant (biosurfactant). Appl. Environ. Microbiol. 58:3276-3282.(Mary E. Davey Nicky C. Caiazza and George A. O'Toole)
Received 10 June 2002/ Accepted 6 October 2002q/#s*, 百拇医药
ABSTRACTq/#s*, 百拇医药
In response to certain environmental signals, bacteria will differentiate from an independent free-living mode of growth and take up an interdependent surface-attached existence. These surface-attached microbial communities are known as biofilms. In flowing systems where nutrients are available, biofilms can develop into elaborate three-dimensional structures. The development of biofilm architecture, particularly the spatial arrangement of colonies within the matrix and the open areas surrounding the colonies, is thought to be fundamental to the function of these complex communities. Here we report a new role for rhamnolipid surfactants produced by the opportunistic pathogen Pseudomonas aeruginosa in the maintenance of biofilm architecture. Biofilms produced by mutants deficient in rhamnolipid synthesis do not maintain the noncolonized channels surrounding macrocolonies. We provide evidence that surfactants may be able to maintain open channels by affecting cell-cell interactions and the attachment of bacterial cells to surfaces. The induced synthesis of rhamnolipids during the later stages of biofilm development (when cell density is high) implies an active mechanism whereby the bacteria exploit intercellular interaction and communication to actively maintain these channels. We propose that the maintenance of biofilm architecture represents a previously unrecognized step in the development of these microbial communities.
INTRODUCTIONl4!2h, 百拇医药
Although bacteria are commonly viewed as solitary life forms, these organisms are more typically colonial creatures. In their natural settings, bacteria persist within microbial communities, where they exploit elaborate systems of intercellular interaction and communication to adjust to changing environmental parameters. Moreover, biofilm formation has also been linked to the emergence of a variety of opportunistic human pathogens (5). For example, organisms such as Staphylococcus epidermidis and Pseudomonas aeruginosa form biofilms on implants and dead or living tissue, thereby contributing to a variety of persistent infections.l4!2h, 百拇医药
Thinking about bacterial populations as connected organisms capable of concerted multicellular activities has provided researchers with novel insights into microbial biology. The key to such multicellular behavior lies in the ability of each individual cell to sense and respond to information from nearby cells, and this behavior requires a certain population size or quorum of cells. The development of biofilms is a process that involves both a quorum of cells and multicellular behavior (6). Single-species biofilms are of particular interest due to their clinical importance and the monospecies biofilms formed by P. aeruginosa has become a prominent model for studying this aspect of microbial biology.
The complex structure of microbial biofilms has only recently been determined. Detailed analysis by scanning confocal laser microscopy has shown that biofilms of P. aeruginosa formed on solid surfaces and exposed to a continuous flow of fresh nutrients are open, highly hydrated structures consisting of cells embedded in an extracellular matrix filled with large void spaces (14). These void spaces, or channels, allow fluids to flow throughout the biofilm, resulting in the distribution of nutrients and oxygen. In addition, the channels between macrocolonies may also provide a means of removing metabolic end products.gk\!, 百拇医药
It has been hypothesized that open-channel formation is not a stochastic process but instead represents an active process that creates a preferable structural design. For example, under some conditions P. aeruginosa mutants unable to make quorum-sensing molecules initiate biofilm formation but are unable to make normally structured biofilms (7). Other P. aeruginosa mutants, including those deficient in flagellum and pilus synthesis, as well as those with mutations in global regulators such as gacA and crc, have defects in the initial colonization of a surface (9, 22, 24, 27). Taken together, these data suggest that formation of the distinctive architecture of P. aeruginosa biofilms, including the macrocolonies surrounded by fluid-filled channels, is a regulated developmental process requiring distinct genetic surface structures and regulatory elements (23, 29, 30, 35). However, mechanisms by which the bacteria maintain these channels once they form have not been investigated. Our data indicate P. aeruginosa not only regulates development of its distinctive biofilm architecture but, once channels form, this organism utilizes rhamnolipid surfactants to actively maintain the void spaces surrounding macrocolonies. That is, we propose that rhamnolipids are not required for the formation of macrocolonies and channels but participate in the maintenance of channels once they are formed.
MATERIALS AND METHODS(a!3a6, http://www.100md.com
Bacterial strains, plasmids, and media. P. aeruginosa PAO1 was the wild-type strain used in the present study (13), and the rhlA::Tn5 mutant strain was described previously (19). Plasmid pSMC21 (GFP+ Apr Knr) constitutively expresses gfp and was used to label the strains for microscopy studies (1), and plasmid pMH520 contains a rhlA::gfp transcriptional fusion (a gift from Matt Parsek, Northwestern University). All flow cell experiments were performed in EPRI medium as described previously (7), except that morpholinepropanesulfonic acid served as the buffer (2.2%) and the medium was supplemented with phosphate (0.0019% KH2PO4, 0.0063% K2HPO4). Dirhamnolipids were purified from the supernatant of P. aeruginosa as described previously (37) and were a gift of Thea Norman at Microbia, Inc. (Cambridge, Mass.).(a!3a6, http://www.100md.com
Microtiter dish assay for biofilm formation and flow chamber experiments. For the biofilm formation assay, M63 medium (100 µl/well) containing arginine (0.05%) and MgSO4 (100 µM) is inoculated (1:50) with an overnight L-broth-grown culture as described previously (25). The microtiter plates were then incubated at 37°C for the times indicated. The biofilm is macroscopically visualized by addition of 100 µl of a 0.1% solution of crystal violet to each well. Biofilm formation was quantified by the addition of 125 µl of 100% ethanol to each crystal violet-stained microtiter dish well; after 10 min 100 µl was transferred to a new microtiter dish, and the absorbance was determined with a plate reader at 550 nm.
Biofilms were also cultivated in flow chambers with channel dimensions of 5 by 1 by 30 mm. The flow system was assembled as described previously (4). The flow cell was inoculated from overnight L-broth-grown cultures diluted 10-fold in EPRI medium (final concentration, 108/ml). The medium flow was turned off prior to inoculation and for 1 h after inoculation. Thereafter, medium was pumped through the flow cell at a constant rate (1.8 ml/h) for the duration of the experiment. The flow was controlled with a PumpPro MPL (Watson-Marlow).56itn, 百拇医药
Microscopy and staining. A Leica DM IRB inverted microscope (Leica Microsystems, Wetzlar, Germany) equipped with a cooled charge-coupled device digital camera and a x10 or x63 PL Flotar objective lens was used for epifluorescence and phase-contrast microscopy analyses. With this instrument, digital images were captured and processed by using a G4 Macintosh computer with OpenLab software package (Improvision, Coventry, England). The images were processed for publication by using Photoshop software (Adobe, Mountain View, Calif.). Quantitative analysis of the flow cell-grown biofilms was performed with the COMSTAT image analysis software package (11, 12).
Mixing experiments. Mixing experiments were performed by inoculating ~ 107 bacteria of both the wild type (carrying the pSMC21 gfp+ plasmid) and the rhlA mutant into the same channel of the flow cell. Biofilms were allowed to form for 6 days, and then the cells were stained with 5-cyano-2,3-di-4-tolyl-tetrazolium chloride (CTC; Polysciences, Inc., Warrington, Pa.), a metabolic stain that renders metabolically active cells red. Cells expressing green fluorescent protein (GFP) quenched the red fluorescence and thus allowed the wild type and the rhlA mutants to be distinguished. Channels inoculated with the wild type alone or with the rhlA mutant alone were included in each experiment as controls. No difference in architecture or staining was observed when the rhlA mutant carried the GFP-expressing plasmid (data not shown).0mg3%, 百拇医药
Invasion experiments. The wild-type strain was allowed to form a biofilm in a flow cell for 6 days. The biofilm was stained with CTC for 1 h in the absence of flow, and then flow resumed for 2 h to remove the residual dye. This preformed, stained biofilm was "invaded" with ~ 107 wild-type cells carrying the pSMC21 gfp+ plasmid, and the flow was stopped for 1 h. After the resumption of flow for an additional hour to remove the unattached bacteria, the biofilm was examined for red and green fluorescence, and the individual images were merged to form a composite of preformed biofilm cells and GFP-labeled invading cells.
RESULTS'#u#ds, 百拇医药
Rhamnolipids are required to maintain biofilm architecture. It was shown previously that the lasR-lasI quorum-sensing system in P. aeruginosa is involved in the later stages of biofilm differentiation, when cell density is sufficient to constitute a quorum (7). A biofilm produced by the lasI mutant is undifferentiated, i.e., the cells appear to grow as a continuous sheet on the surface lacking the characteristic water channels, indicating that P. aeruginosa has one (or more) systems to control the formation of macrocolony structures. Because lasI controls a whole suite of genes, the specific gene products contributing to this phenotype were unknown. One function controlled by the lasI-lasR system (indirectly via the rhlI-rhlR system) is rhamnolipid biosynthesis. Therefore, we investigated the biofilm formation phenotype of a mutant strain unable to synthesize rhamnolipids. The quorum-sensing-controlled rhlA gene codes for a rhamnosyltransferase whose only known function is in rhamnolipid synthesis. An rhlA mutant does not make rhamnolipids (20, 21).
To study the role rhamnolipids played in the formation of surface-attached communities, we monitored biofilm development of wild-type P. aeruginosa PAO1 and a rhlA-null mutant. Early biofilm formation was assayed in the microtiter dish assay. The wild-type strain forms a biofilm that peaks at 8 h and then declines at 24 h . Previous studies indicated that the P. aeruginosa biofilm formed in a 96-well microtiter dish declines after ca. 10 h of incubation, probably due to starvation in this batch culture system (25). In contrast, the biomass of the rhlA mutant biofilm continued to increase over this 24-h period to a final A550 value twofold greater than the maximum observed for the wild type. There was no discernible difference in the planktonic growth of these two strains. These data suggest that the rhlA mutant is more proficient at early colonization than is the wild-type strain.?*dul2, 百拇医药
fig.ommitted?*dul2, 百拇医药
A rhamnolipid mutant produces biofilms lacking characteristic architecture. (A) Biofilm formation by the wild type and rhlA mutant in microtiter dishes. The biofilm formation phenotype of the wild type (solid bars) and rhlA mutant (open bars) was quantitated over 24 h by using the microtiter dish assay, as reported elsewhere (26). The A550 value represents crystal violet-stained bacteria attached to the walls of the microtiter dish and is an indirect measure of the biofilm formed. The data represent two experiments, each performed in triplicate. (B) Flow cell architecture of mature biofilms. The wild-type and rhlA::Tn5 strains carrying plasmid pSMC21, which constitutively expresses GFP, were inoculated into flow cells fed by minimal salts EPRI medium supplemented with glucose as the carbon and/or energy source. The top-down views shown here were acquired by epifluorescent microscopy by using the x63 objective lens, which was used to monitor biofilm development in flow cells over 6 days. The large green macrocolonies are surrounded by dark areas, which are the sparsely colonized channel regions. At day 4, there is no difference between the wild type and the rhlA mutant, thus demonstrating that rhamnolipids are not required for the formation of macrocolonies. For the rhlA mutant, partial filling of the channels can be observed by day 5 and the channels are completely filled in this strain by day 6. The wild-type strain at day 6 displays the characteristic architecture of a mature P. aeruginosa biofilm. Scale bars are included and labeled on the figure. (C) Flow cell architecture of mature biofilms visualized under low magnification. This panel shows a day 6 biofilm obtained at low magnification (with a x10 objective lens versus the x63 objective lens used for the images in panel B). Scale bars are included and are labeled on the figure.
We followed further biofilm maturation over the course of 6 days by using a once-through flow cell system that provides a continuous supply of fresh nutrients to the biofilm. To monitor the bacteria by epifluorescence microscopy, the strains were transformed with plasmid pSMC21, which constitutively expresses the GFP. We had shown previously that this plasmid is maintained in the absence of antibiotic selection and confers no detectable metabolic load on the cells (1).^, 百拇医药
Macrocolony formation by the wild-type and rhlA strains appeared to be identical after 4 days of incubation . Both strains formed well-defined fluorescent (green) colonies separated by dark, fluid-filled channels. This observation indicates that the small advantage gained by the rhlA mutant in initial colonization in a microtiter dish assay does not translate to an observable phenotype in the architecture of the 4-day-old flow cell-grown biofilm. By day 6, wild-type architecture was comprised of characteristic macrocolonies surrounded by open channels. The rhlA mutant, however, produced a biofilm that was a thick, uniform mat of bacterial cells . Visualizing the biofilm at a lower magnification confirmed the difference in architecture . These data indicated that the rhlA mutant was able to form the channels surrounding the macrocolonies but was unable to maintain these channels.
Quantitative analysis of biofilm architecture. Visual inspection of biofilms produced by wild-type and the rhamnolipid mutant at day 6 indicated an altered biofilm structure for the mutant. To confirm this observation, we applied the COMSTAT image analysis program to perform a quantitative analysis of biofilm architecture (11, 12). As shown in , four variables (mean thickness, substratum coverage, roughness coefficient, and surface/volume ratio) were used to evaluate biofilm architecture. The mean thickness of the biofilm and the percent substratum coverage were much higher in the biofilms formed by the rhamnolipid mutant, a finding consistent with the images shown in. The roughness coefficient was much lower in the mutant, demonstrating that biofilms produced in the absence of rhamnolipids are much less heterogeneous. In addition, the surface/volume ratio was significantly lower in the biofilm produced by the mutant, indicating that a smaller fraction of the biomass is exposed to the nutrient flow. These data confirm the visual observation that the rhlA mutant produces a biofilm consisting of a thick, uniform mat of bacterial cells.
fig.ommittedlk;v\y$, http://www.100md.com
Quantitative analysis of biofilm architecturelk;v\y$, http://www.100md.com
Partial rescue of the rhlA biofilm architecture defect by the wild-type strain. If the defect in biofilm development is the result of a lack of rhamnolipid production, it follows that providing rhamnolipids should restore wild-type biofilm architecture. Rhamnolipids are likely produced by a certain subset of cells in response to specific environmental cues, resulting in surfactant production at a specific time and place, as has been reported elsewhere (8). Consistent with this idea, providing spent supernatants of P. aeruginosa cultures was insufficient to rescue the architecture defect of a rhlA mutant (data not shown). As an alternative approach to test the ability of rhamnolipids to rescue the phenotype of a rhlA mutant, we monitored biofilm development in a flow cell inoculated with a mixture of equal numbers of wild-type PAO1 (green cells) and the rhlA::Tn5 mutant (red cells). As seen in, this mixing experiment resulted in a biofilm that consisted of macrocolonies made up in large part of a mixture of wild-type and rhlA mutant bacteria. Despite the presence of wild-type bacteria in many macrocolonies, the resulting biofilm is a mixture of typical wild-type architecture (comprised of macrocolonies surrounded by open areas) and dense compact areas where the channels were occupied by cells, a phenotype that is reminiscent of the rhlA mutant. These data indicate that the presence of the wild type could only partially rescue the biofilm architecture defect of the rhlA mutant.
fig.ommitted)63t.ne, http://www.100md.com
Partial rescue of the rhlA mutant architecture phenotype by coculture with the wild-type strain. A mixing experiment was done to evaluate the ability of the wild type to rescue the phenotype of the rhlA mutant. The image is a top-down epifluorescent view of a 6-day-old biofilm formed in a flow cell where wild-type cells was mixed 50:50 with the rhlA mutant obtained with a x10 objective lens. The rhlA mutants cells were visualized by CTC staining (red cells, see Materials and Methods), and the wild-type strain expresses the GFP from a plasmid (green cells). Regions of overlapping wild-type and mutant bacteria are yellow.)63t.ne, http://www.100md.com
Expression of rhlA during biofilm development. To evaluate where and when rhamnolipids are synthesized in the biofilm, we examined the expression of the rhlA gene during development. Expression of rhlA was determined indirectly by using plasmid prhlA::gfp, which harbors a transcriptional fusion of the rhlA promoter region to the GFP. As shown in, on day 1, small microcolonies of 10 to 30 cells associated with the early stages in biofilm development can be seen by phase-contrast microscopy; however, no rhlA-gfp-dependent fluorescence was observed at this time point. However, by day 2, larger colonies comprised of thousands of fluorescent cells could be detected, indicating expression of the rhlA gene. Expression of this fusion was found to coincide with macrocolony formation and continued throughout biofilm development in areas where the cell density was high. The expression of rhlA only in macrocolonies is consistent with the observation that rhamnolipids are not required for the initiation of biofilm development. In control experiments, we could observe the fluorescence of individual batch grown planktonic, stationary-phase cells carrying the prhlA-gfp plasmid (not shown). These data demonstrate that the digital camera used to obtain images was sufficiently sensitive to detect an individually fluorescing bacterium.
fig.ommitted]o9l|, 百拇医药
Expression pattern of an rhlA::gfp transcriptional fusion during biofilm formation. The images are top-down views of individual cells and colonies at day 1 (A) and day 2 (B) for flow cell-grown bacteria (obtained with the x63 objective lens). Both phase-contrast and fluorescent images of the same view are shown. Small clusters of bacteria produced no detectable fluorescence; however, larger colonies (high cell density) resulted in expression of the rhlA::gfp fusion.]o9l|, 百拇医药
Overproduction of rhamnolipids inhibits biofilm development. Based on their surface-active properties, we predicted that rhamnolipids could affect the interactions of bacteria with surfaces and with each other. To address this hypothesis directly, we examined the effects of rhamnolipid overexpression on the interactions of bacteria with a surface. First, the effects of rhamnolipids on early biofilm formation were determined by growing bacteria under conditions leading to high-level expression of rhamnolipids. Rhamnolipids are typically synthesized during late-exponential or stationary phase of growth when the cell density is high, and this production is greatly enhanced by nutrient limitation (10). However, it has been shown previously that growth in a phosphate-limited complex medium induces the production of high levels of rhamnolipids (~ 100 µM in planktonic cultures) in the absence of high cell density (17). The formation of biofilms was observed in flow cells fed with phosphate-limited PPGAS medium (~ 0.03 mM inorganic phosphate, low-phosphate conditions). As shown in , when grown in low-phosphate, rhamnolipid-inducing conditions, wild-type cells started to attach at day 1, but by day 3 very few bacteria remained attached to the surface of the flow cell. In contrast, the rhlA mutant, which is unable to produce any rhamnolipids, formed a dense biofilm on the surface by day 3. Supplementing the PPGAS complex medium with 2.5 mM potassium phosphate (high-phosphate conditions) restored the ability of the wild-type strain to form a biofilm in the flow cell (data not shown). Therefore, expressing high levels of rhamnolipids can impede the formation of biofilms.
fig.ommitted'jp, 百拇医药
Effect of rhamnolipid overexpression on biofilm formation. The images are fluorescent top-down views of flow cell-grown biofilms formed by wild type and the rhamnolipid mutant at days 1 and 3. PPGAS medium, a low-phosphate medium, induced the synthesis of high levels of rhamnolipids in the wild-type strain and resulted in a sparsely colonized surface. The rhlA mutant was unable to synthesize any rhamnolipids even in low-phosphate medium and made a dense biofilm by day 3.'jp, 百拇医药
Rhamnolipids disrupt both cell-to-cell and cell-to-surface interactions. The data presented above indicated that production of rhamnolipids could help maintain the channels formed in a biofilm. How can rhamnolipids carry out such a function? As mentioned above, rhamnolipids are surfactants, or surface-active molecules, and a general property of surfactants involves their ability to alter the physical and/or chemical properties at interfaces (18). We hypothesized, therefore, that the rhamnolipid surfactants could influence P. aeruginosa cell-to-cell and/or cell-to-surface interactions.
To test these hypotheses, purified rhamnolipids were assessed for their effects on the initiation of biofilm development by using the previously described microtiter dish assay. Purified rhamnolipid (250 µM) added to bacteria at the time of inoculation into the microtiter dishes completely blocked biofilm formation. Partial inhibition of biofilm formation was observed at lower concentrations of rhamnolipid (not shown).998e^|{, 百拇医药
fig.ommitted998e^|{, 百拇医药
Effect of purified rhamnolipids on cell-surface and cell-cell interactions. (Upper panels) Cell-surface interactions. The ability to form a biofilm on polyvinylchloride plastic in the presence or absence of rhamnolipids was examined. The rhamnolipid was added at the time of inoculation, and the biofilm formation was examined after 8 h of incubation. The addition of 250 µM rhamnolipid completely blocked biofilm formation (right panel). (Lower panels) Cell-cell interactions. The effect of rhamnolipids on rafts or pellicles of wild-type P. aeruginosa cells was examined. The disruption of cell-to-cell interactions caused by adding rhamnolipids was captured by phase-contrast microscopy immediately after addition of the compound (right panel). The addition of the same volume of water had no effect on biofilm formation or disruption of the rafts (left panels).
We also examined the effect of purified rhamnolipids on cell-to-cell interactions. P. aeruginosa typically forms a pellicle or raft of cells on the surface of a liquid culture. As shown in , the addition of 100 µM dirhamnolipid disrupted cell-to-cell interactions, and this disruption occurred immediately after the addition of the compound. Taken together, these data demonstrate that rhamnolipids can disrupt both cell-to-cell and cell-to-surface interactions. Interestingly, neither purified rhamnolipid nor rhamnolipid-containing spent supernatants of P. aeruginosa cultures had any discernible effect on a preformed 6-day-old biofilm (not shown). These data suggest that rhamnolipids can only interfere with earlier stages of biofilm development.48c2v/, 百拇医药
Invasion of a preformed biofilm. Based on the evidence that biofilm channel structure is maintained over time, we predicted that few planktonic cells invading a preformed biofilm would be capable of attaching to this community. We performed an invasion experiment by introducing GFP-tagged P. aeruginosa cells (~ 107 cells) into a flow cell containing a preformed mature 5-day-old biofilm, and the tagged cells were allowed to attach for 1 h under static conditions . After this period of attachment, flow was resumed for an additional hour to remove unattached bacteria before microscopic examination at low and high magnifications. As shown in, very few of the invading GFP-tagged cells attached to the biofilm. These data are consistent with the hypothesis that preformed biofilms have the means, including but not necessarily limited to the production of rhamnolipids, to control the attachment of planktonic bacteria.
fig.ommitted3, 百拇医药
Biofilm invasion assay. A biofilm of non-GFP-labeled wild-type bacteria was allowed to form for 5 days in a flow cell and then stained with CTC. (a and c) The flow was turned off, ca. 107 GFP-labeled wild-type P. aeruginosa cells were injected into the flow cell, and the flow chamber was incubated without flow for 1 h. (b and d) Flow was resumed for 1 h to remove planktonic bacteria, and the biofilm was examined by fluorescence microscopy. A merged composite image of the fluorescence micrographs is shown. The images in panels a and b were captured by using a x63 objective lens, and the images in panels c and d were captured by using a x10 objective lens. Scale bars are included in the images.3, 百拇医药
DISCUSSION3, 百拇医药
Our studies show that in situ production of rhamnolipid surfactants by P. aeruginosa cells affects biofilm architecture but that the loss of rhamnolipid production does not appear to affect the initial development of macrocolonies and open channels. However, the rhlA mutant is incapable of maintaining open channels, eventually forming a relatively homogeneous layer of cells as thick as, or thicker than, the wild-type biofilm . Such a role for surface-active compounds in the structuring of microbial populations at interfaces is not unprecedented in the microbial world. Surfactants are known to play a role in the emergence of aerial hyphae in both fungi and the filamentous bacterium Streptomyces coelicolor (31, 36) and in the formation of hydrophobic air channels in fruiting bodies of fungi (15). In addition, a recent study of fruiting body formation in the bacterium Bacillus subtilis showed that a mutant (sfp) deficient in surfactin production formed atypical surface-associated pellicle (biofilm) structures (2).
How does the production of rhamnolipids maintain the open channels formed during biofilm development? At this point, the answer to this question is not clear. As has been reported, high cell density planktonic growth induces the synthesis of quorum-sensing-dependent rhamnolipid production (3, 21, 28), and similar density-dependent transcription of the RhlI-RhlR system and rhlA is also seen in biofilms (8) . We propose that surfactant production may inhibit planktonic cells from attaching to the preformed biofilm. This contention is supported by the "invasion assay" shown in . Alternatively, rhamnolipids may cause the detachment of cells or microcolonies from the biofilm, preventing the accumulation of new biomass in the channels. These models are not mutually exclusive. Finally, our data indicate that exposing a preformed biofilm to rhamnolipids has no significant effect on established biofilm structure, suggesting that the production of surfactants by preformed macrocolonies would not necessarily cause wholesale detachment of the biofilm.
The ability to block colonization of a preformed biofilm is likely due in large part to the well-known properties of surfactants at interfaces. Surfactants have the potential to affect both cell-to-cell and cell-to-surface interactions (18), and our data support the hypothesis that rhamnolipids produced by P. aeruginosa do affect these interface interactions. We showed that purified rhamnolipids, the only characterized surfactants produced by P. aeruginosa, are capable of disrupting interactions between cells and each other and a surface. Hence, rhamnolipids are able to modulate cell-to-surface and cell-to-cell interactions. The ability of surfactants to disrupt bacterial attachment has been observed for the surfactant produced by Lactobacillus spp., known as surlactin, which blocked initial surface adhesin of uropathogenic Enterococcus faecalis, Escherichia coli, and Staphylococcus epidermidis, as well as two yeast strains (32-34). The surfactant produced by B. subtilis, called surfactin, also inhibits biofilm formation by Salmonella enterica, E. coli, and Proteus mirabilis (16). Studies in our laboratory indicate that purified P. aeruginosa rhamnolipids can block initial adherence by fluorescent pseudomonads and E. coli (M. E. Davey and G. A. O'Toole, unpublished data), suggesting that surfactants may have a broad and relatively nonspecific ability to interfere with cell-to-cell and cell-to-surface interactions. Therefore, P. aeruginosa rhamnolipid may prevent colonization not only of "self" but also of other planktonic microbes attempting to take up residence within the channels of its biofilm.
The findings presented here indicate that open-channel maintenance is not a stochastic process but represents an active process by the bacterial community that stabilizes a preferable biofilm architecture. Recent studies in P. aeruginosa suggest a role for a number of genes in the formation of biofilm architecture, indicating that a distinct set of genetic factors is required for the development of biofilm architecture in P. aeruginosa. We suggest that the production of rhamnolipid surfactants is one mechanism employed by P. aeruginosa for the active maintenance of biofilm architecture. We believe that other, as-yet-unidentified mechanisms may also contribute to the maintenance of architecture. For example, a decrease or cessation of growth in large macrocolonies may slow or stop the filling of channels between these structures. Finally, we propose that the maintenance of biofilm architecture is a previously undescribed stage in the development of P. aeruginosa biofilms.ay9@[, 百拇医药
ACKNOWLEDGMENTS
We thank Thea Norman at Microbia, Inc., for providing purified dirhamnolipid, Urs Ochsner for the rhlA::Tn5 strain, Matt Parsek for the rhlA-gfp fusion plasmid, and John P. Connolly for computer expertise.4q, http://www.100md.com
This research was funded by grants from Microbia, Inc., and The Pew Charitable Trusts (to G.A.O.). G.A.O is a Pew Scholar in the Biomedical Sciences. M.E.D. was supported by an NIH training grant (5 T32 AI07519).4q, http://www.100md.com
REFERENCES4q, http://www.100md.com
Bloemberg, G. V., G. A. O'Toole, B. J. J. Lugtenberg, and R. Kolter. 1997. Green fluorescent protein as a marker for Pseudomonas spp. Appl. Environ. Microbiol. 63:4543-4551.4q, http://www.100md.com
Branda, S. S., J. E. Gonzalez-Pastor, S. Ben-Yehuda, R. Losick, and R. Kolter. 2001. Fruiting body formation by Bacillus subtilis. Proc. Natl. Acad. Sci. USA 98:11621-11626.4q, http://www.100md.com
Brint, J. M., and D. E. Ohman. 1995. Synthesis of multiple exoproducts in Pseudomonas aeruginosa is under the control of RhlR-RhlI, another set of regulators in strain PAO1 with homology to the autoinducer-responsive LuxR-LuxI family. J. Bacteriol. 177:7155-7163.
Christensen, B. B., C. Sternberg, J. B. Andersen, R. J. Palmer, Jr., A. T. Nielsen, M. Givskov, and S. Molin. 1999. Molecular tools for study of biofilm physiology. Methods Enzymol. 310:20-42.qx, 百拇医药
Costerton, J. W., P. S. Stewart, and E. P. Greenberg. 1999. Bacterial biofilms: a common cause of persistent infections. Science 284:318-322.qx, 百拇医药
Davey, M. E., and G. O. O'Toole. 2000. Microbial biofilms: from ecology to molecular genetics. Microbiol. Mol. Biol. Rev. 64:847-867.qx, 百拇医药
Davies, D. G., M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and E. P. Greenberg. 1998. The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280:295-298.qx, 百拇医药
De Kievit, T. R., R. Gillis, S. Marx, C. Brown, and B. H. Iglewski. 2001. Quorum-sensing genes in Pseudomonas aeruginosa biofilms: their role and expression patterns. Appl. Environ. Microbiol. 67:1865-1873.qx, 百拇医药
Drenkard, E., and F. M. Ausubel. 2002. Pseudomonas biofilm formation and antibiotic resistance are linked to phenotypic variation. Nature 416:740-743.
Guerra-Santos, L., O. Käppeli, and F. Fiechter. 1986. Dependence of Pseudomonas aeruginosa continuous culture biosurfactant production on nutritional and environmental factors. Appl. Microbiol. Biotechnol. 24:443-448.j, 百拇医药
Heydorn, A., B. K. Ersboll, M. Hentzer, M. R. Parsek, M. Givskov, and S. Molin. 2000. Experimental reproducibility in flow-chamber biofilms. Microbiology 146:2409-2415.j, 百拇医药
Heydorn, A., A. T. Nielsen, M. Hentzer, C. Sternberg, M. Givskov, B. K. Ersboll, and S. Molin. 2000. Quantification of biofilm structures by the novel computer program COMSTAT. Microbiology 146:2395-2407.j, 百拇医药
Holloway, B. W., V. Krishnapillai, and A. F. Morgan. 1979. Chromosomal genetics of Pseudomonas. Microbiol. Rev. 43:73-102.j, 百拇医药
Lawrence, J. R., D. R. Korber, B. D. Hoyle, J. W. Costerton, and D. E. Caldwell. 1991. Optical sectioning of microbial biofilms. J. Bacteriol. 173:6558-6567.j, 百拇医药
Lugones, L. G., H. A. B. Wosten, K. U. Birkenkamp, K. A. Sjollema, J. Zagers, and J. G. H. Wessels. 1999. Hydrophobins line air channels in fruiting bodies of Schizophyllum commune and Agaricus bisporus. Mycol. Res. 103:635-640.
Mireles, J. R., II, A. Toguchi, and R. M. Harshey. 2001. Salmonella enterica serovar Typhimurium swarming mutants with altered biofilm-forming abilities: surfactin inhibits biofilm formation. J. Bacteriol. 183:5848-5854.71z@f, 百拇医药
Mulligan, C. N., G. Mahmourides, and B. F. Gibbs. 1989. The influence of phosphate metabolism on biosurfactant production by Pseudomonas aeruginosa. J. Biotechnol. 12:199-210.71z@f, 百拇医药
Neu, T. R. 1996. Significance of bacterial surface-active compounds in interaction of bacteria with interfaces. Microbiol. Rev. 60:151-166.71z@f, 百拇医药
Ochsner, U. A., A. Fiechter, and J. Reiser. 1994. Isolation, characterization, and expression in Escherichia coli of the Pseudomonas aeruginosa rhlAB genes encoding a rhamnosyltransferase involved in rhamnolipid biosurfactant synthesis. J. Biol. Chem. 269:19787-19795.71z@f, 百拇医药
Ochsner, U. A., A. K. Koch, A. Fiechter, and J. Reiser. 1994. Isolation and characterization of a regulatory gene affecting rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. J. Bacteriol. 176:2044-2054.
Ochsner, U. A., and J. Reiser. 1995. Autoinducer-mediated regulation of rhamnolipid biosurfactant synthesis in Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 92:6424-6428.^iddch], http://www.100md.com
O'Toole, G. A., K. A. Gibbs, P. W. Hager, P. V. Phibbs, Jr., and R. Kolter. 2000. The global carbon metabolism regulator Crc is a component of a signal transduction pathway required for biofilm development by Pseudomonas aeruginosa. J. Bacteriol. 182:425-431.^iddch], http://www.100md.com
O'Toole, G. A., H. Kaplan, and R. Kolter. 2000. Biofilm formation as microbial development. Annu. Rev. Microbiol. 54:49-79.^iddch], http://www.100md.com
O'Toole, G. A., and R. Kolter. 1998. Flagellar and twitching motility are necessary for Pseudomonas aeruginosa biofilm development. Mol. Microbiol. 30:295-304.^iddch], http://www.100md.com
O'Toole, G. A., and R. Kolter. 1998. The initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signaling pathways: a genetic analysis. Mol. Microbiol. 28:449-461.^iddch], http://www.100md.com
O'Toole, G. A., L. A. Pratt, P. I. Watnick, D. K. Newman, V. B. Weaver, and R. Kolter. 1999. Genetic approaches to the study of biofilms. Methods Enzymol. 310:91-109.
Parkins, M. D., H. Ceri, and D. G. Storey. 2001. Pseudomonas aeruginosa GacA, a factor in multihost virulence, is also essential for biofilm formation. Mol. Microbiol. 40:1215-1226.54ocf, http://www.100md.com
Pearson, J. P., E. C. Pesci, and B. H. Iglewski. 1997. Roles of Pseudomonas aeruginosa las and rhl quorum-sensing systems in control of elastase and rhamnolipid biosynthesis genes. J. Bacteriol. 179:5756-5767.54ocf, http://www.100md.com
Sauer, K., and A. K. Camper. 2001. Characterization of phenotypic changes in Pseudomonas putida in response to surface-associated growth. J. Bacteriol. 183:6579-6589.54ocf, http://www.100md.com
Sauer, K., A. K. Camper, G. D. Ehrlich, J. W. Costerton, and D. G. Davies. 2002. Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. J. Bacteriol. 184:1140-1154.54ocf, http://www.100md.com
Tillotson, R. D., H. A. Wosten, M. Richter, and J. M. Willey. 1998. A surface active protein involved in aerial hyphae formation in the filamentous fungus Schizophillum commune restores the capacity of a bald mutant of the filamentous bacterium Streptomyces coelicolor to erect aerial structures. Mol. Microbiol. 30:595-602.
Velraeds, M. M., B. van de Belt-Gritter, H. C. van der Mei, G. Reid, and H. J. Busscher. 1998. Interference in initial adhesion of uropathogenic bacteria and yeasts to silicone rubber by a Lactobacillus acidophilus biosurfactant. J. Med. Microbiol. 47:1081-1085.*, 百拇医药
Velraeds, M. M., H. C. van der Mei, G. Reid, and H. J. Busscher. 1996. Inhibition of initial adhesion of uropathogenic Enterococcus faecalis by biosurfactants from Lactobacillus isolates. Appl. Environ. Microbiol. 62:1958-1963.*, 百拇医药
Velraeds, M. M., H. C. van der Mei, G. Reid, and H. J. Busscher. 1997. Inhibition of initial adhesion of uropathogenic Enterococcus faecalis to solid substrata by an adsorbed biosurfactant layer from Lactobacillus acidophilus. Urology 49:790-794.*, 百拇医药
Whiteley, M., M. G. Bangera, R. E. Bumgarner, M. R. Parsek, G. M. Teitzel, S. Lory, and E. P. Greenberg. 2001. Gene expression in Pseudomonas aeruginosa biofilms. Nature 413:860-864.*, 百拇医药
Wosten, H. A., M. Richter, and J. M. Willey. 1999. Structural proteins involved in emergence of microbial aerial hyphae. Fungal Genet. Biol. 27:153-160.*, 百拇医药
Zhang, Y., and R. M. Miller. 1992. Enhanced octadecane dispersion and biodegradation by a Pseudomonas rhamnolipid surfactant (biosurfactant). Appl. Environ. Microbiol. 58:3276-3282.(Mary E. Davey Nicky C. Caiazza and George A. O'Toole)