JOURNAL OF BACTERIOLOGY, Jan. 2005, p. 37–44 0021-9193/05/$08.00⫹0 doi:10.1128/JB.187.1.37–44.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 187, No. 1

Timing and Localization of Rhamnolipid Synthesis Gene Expression in Pseudomonas aeruginosa Biofilms Yannick Lequette and E. P. Greenberg* Department of Microbiology and W. M. Keck Microbial Communities and Cell Signalling Program, Carver College of Medicine, University of Iowa, Iowa City, Iowa Received 19 May 2004/Accepted 16 August 2004

Pseudomonas aeruginosa biofilms can develop mushroom-like structures with stalks and caps consisting of discrete subpopulations of cells. Self-produced rhamnolipid surfactants have been shown to be important in development of the mushroom-like structures. The quorum-sensing-controlled rhlAB operon is required for rhamnolipid synthesis. We have introduced an rhlA-gfp fusion into a neutral site in the P. aeruginosa genome to study rhlAB promoter activity in rhamnolipid-producing biofilms. Expression of the rhlA-gfp fusion in biofilms requires the quorum-sensing signal butanoyl-homoserine lactone, but other factors are also required for expression. Early in biofilm development rhlA-gfp expression is low, even in the presence of added butanoylhomoserine lactone. Expression of the fusion becomes apparent after microcolonies with a depth of >20 ␮m have formed and, as shown by differential labeling with rfp or fluorescent dyes, rhlA-gfp is preferentially expressed in the stalks rather than the caps of mature mushrooms. The rhlA-gfp expression pattern is not greatly influenced by addition of butanoyl-homoserine lactone to the biofilm growth medium. We propose that rhamnolipid synthesis occurs in biofilms after stalks have formed but prior to capping in the mushroom-like structures. The differential expression of rhlAB may play a role in the development of normal biofilm architecture. Pseudomonas aeruginosa is an opportunistic pathogen that can cause acute infections or chronic biofilm infections (3, 5, 6, 16, 35). In part because it is an emerging pathogen (33), the biology of P. aeruginosa biofilms has received increasing attention recently. At least under specific laboratory conditions, P. aeruginosa biofilm development involves a number of discrete steps. First, individual cells attach to a surface. This is followed by the formation of microcolonies on the surface. Finally, the microcolonies mature into mushroom-like structures in which the cells are embedded in a self-produced extracellular polymeric matrix (6, 20). We have begun to learn about genes involved in the development of normal biofilm structures. The development of mushroom-like structures is dependent upon the available carbon and energy source (19). A recent report showed that the mushroom-like structures consist of stalks on which caps form. Cap production requires a form of surface movement called twitching motility (18). Apparently, cells in the stalk do not move and a subpopulation of cells use twitching motility to migrate up the stalk and form a cap. Quorum sensing influences the ability of P. aeruginosa to develop stalks, so that quorum-sensing mutants form a thin, rather uniform layer on surfaces (8). Furthermore, P. aeruginosa produces extracellular rhamnolipid surfactants, and the synthesis of these rhamnolipids is controlled by quorum sensing (25). Rhamnolipid synthesis mutants form abnormal unstructured biofilms and may show enhanced dispersal from surfaces (7). The involvement of rhamnolipids in biofilm formation in part could explain why quorum sensing influences biofilm development. We have focused this study on rhamnolipid synthesis gene

expression in biofilms, because quorum sensing controls rhamnolipid synthesis and because rhamnolipid synthesis influences biofilm development. There are two P. aeruginosa acyl-homoserine lactone (HSL) quorum-sensing systems that govern the expression of hundreds of genes (12, 31, 36). LasR is a transcription factor that responds to the LasI-generated signal, N(3-oxoododecanyl)-HSL (3OC12-HSL), and RhlR is a transcription factor that responds to the RhlI-generated signal, N-butanoyl-HSL (C4-HSL) (11, 21, 27, 28, 37, 40). The rhamnolipid synthesis operon rhlAB is among the functions controlled by quorum sensing (22, 25, 29). The rhlAB operon shows a large response to C4-HSL but the specificity is not absolute, and it also shows a significant response to 3OC12HSL (31, 38). Furthermore, the expression of the rhlAB operon occurs as planktonic cultures enter stationary phase, and the lag in expression is not reduced by the addition of 3OC12HSL and C4-HSL (22, 29, 31, 39). Here we investigate the timing and location of rhlAB expression in P. aeruginosa biofilms. A previous study indicated that both lasI and rhlI were preferentially expressed in cells near the base of mature biofilms (9). We did not understand that mushroom-like structures consisted of stalks and caps at the time of this earlier publication, but the results are consistent with expression of these genes in stalks. These genes are not only required for acyl-HSL synthesis but also they are themselves controlled by quorum sensing (30). This suggested that we might observe differential expression of the rhlAB operon in biofilms. We present evidence indicating this operon is expressed primarily in the stalks of mushroom-like structures. The transcription of the rhlAB operon seems to commence at the time stalks are beginning to form.

* Corresponding author. Mailing address: 540 EMRB, Roy and Lucille Carver College of Medicine, University of Iowa, Iowa City, IA 52242. Phone: (319) 335-7775. Fax: (319) 335-7949. E-mail: everett [email protected].

MATERIALS AND METHODS Bacterial strains, plasmids, and media. The bacterial strains and plasmids used in this study are listed in Table 1. Planktonic cultures were grown in a tryptic

37

38

LEQUETTE AND GREENBERG

J. BACTERIOL. TABLE 1. Bacterial strains and plasmids

Strain or plasmid

Relevant characteristicsa

Source or reference

P. aeruginosa strains PAO1 PDO100 YL101 YL100 YL120

Wild type rhlI::Hgr PAO1 derivative PAO1 with an rhlA-gfp fusion inserted in attB PDO100 with an rhlA-gfp fusion inserted in attB PAO1 with chromosomal aracC-Pbad-gfp fusion (stable gfp) on the chromosome in attB

17 2 This study This study This study

E. coli strains DH5␣ SM10

F⬘/endA1 hsdR17 supE44 thi-1 recA1 gyrA relA1 ⌬(lacZYA-argF) U169 deoR [␾80 dlacZ-M15 recA1] Kmr, thi-1 thr leu tonA lacY supE recA::RP4-2Tc::Mu

41 10

Plasmids pUC19 pUCP18 pPROBE-AT Mini-CTX-lacZ pFLP2 pRHLI-2 pUO103 pYL120 pYL121 pYL122 pYL125 pYL126 pYL182 PYL183 pmRFP1

Ampr, E. coli cloning vector Cbr, pUCP18 vector containing a P. aeruginosa origin of replication GFP translational fusion vector with P. aeruginosa and E. coli replication origins Tetr, integration-proficient vector for chromosomal insertion at the attB site Ampr flp recombinase plasmid Gmr, pBBR1MCS5 with ptac-rhlI Kmr Smr, pJRD215 containing rhlI pUC19 with an rhlAB promoter fragment Ampr, rhlAB-gfp transcription fusion in pPROBE-AT Ampr, rhlA-gfp transcription fusion in mini-CTX-lacZ pUCP18 with rhlI gene pUCP18 with ptac-rhlI Ampr, araC-Pbad-gfp transcription fusion in pPROBE-AT Gmr, araC-Pbad-gfp transcription fusion in mini-CTX-gm DsRed in pRESTB

42 32 24 1 14 26 25 This This This This This This This 4

study study study study study study study

a Hgr, mercury resistance; Kmr, kanamycin resistance; Ampr, ampicillin resistance; Cbr, carbanicillin resistance; Tetr, tetracycline resistance; Gmr, gentamicin resistance.

soy broth (Difco) at 37°C. Antimicrobial agents were used as appropriate at the following concentrations (per milliliter): for P. aeruginosa, carbenicillin (150 ␮g), tetracycline (200 ␮g), and HgCl (15 ␮g); for Escherichia coli, ampicillin (50 ␮g), tetracycline (50 ␮g), and kanamycin (50 ␮g). Where indicated, 10 ␮M C4-HSL (unless otherwise specified) or 0.2% L-arabinose was added to the culture medium and was present throughout the course of an experiment. P. aeruginosa does not use L-arabinose as an energy source. Furthermore, inclusion of Larabinose in the medium for growth of biofilms containing reporters other than the arabinose-controlled gfp did not influence results (data not shown). The rhlA-gfp fusion strains were constructed as follows: a fragment containing the rhlAB promoter was amplified from P. aeruginosa PAO1 chromosomal DNA by PCR with 5⬘-CCTGGGCAAGAGCACCTACG-3⬘ and 5⬘-TGGTTGTCTGAA AGCGCGGC-3⬘ as primers. The PCR product was digested with BglII and SmaI and ligated with BamHI-SmaI-digested pUC19 to give pYL120. The cloned fragment is 628 bp and extends from 551 bp upstream of the rhlA translational start. An EcoRI-HindIII fragment of pYL120 was cloned in front of the promoterless gfp in pPROBE AT (24) to generate the rhlA-gfp fusion vector pYL121. A pYL121 HindIII-SspI fragment containing the rhlA-gfp fusion was cloned into HindIII-SspI-digested mini-CTX-lacZ (1) to yield pYL122, which was used to insert the rhlA-gfp fusion into the chromosomal attB sites of the P. aeruginosa strains used in this study as described previously (14, 15). To construct the araC-Pbad-gfp fusion, a 1.35-kb SalI-EcoRI fragment containing araC and the pbad promoter was cloned into pPROBE AT, yielding pYL182. A 2.1-kb NsiI fragment was cloned into PstI-digested mini-CTX-Gm to yield pYL183, which was used to insert the fusion in the chromosome. rhlA-gfp and araC-Pbad-gfp direct the synthesis of a stable, bright green fluorescent protein (GFP) called GreenTIR (24). To construct pYL125, we inserted a 1.4-kb EcoRI-HindIII fragment from pUO103 into EcoRI-HindIII-digested pUCP18. To construct pYL126, we inserted an 899-bp rhlI-containing BamHI-HindIII fragment from pRhlI-2 into BamHI-HindIII-digested pUCP18. Biofilm experiments. Biofilms were grown in flow chambers as described elsewhere (8). The biofilm medium was 1% tryptic soy broth. The flow cells were inoculated with 107 cells in 500 ␮l of 1% tryptic soy broth. These cells were from an overnight culture. The flow was initiated at a rate of 0.17 ml per min after 1 h at room temperature. Where indicated SYTO 62 (Molecular Probes, Eugene, Oreg.) was used to stain biofilms. Most images of biofilms were obtained by using a Radiance 2100 scanning confocal laser microscope (SCLM) system (Bio-Rad, Hercules, Calif.) with a

Nikon Eclipse E600 microscope. For GFP the excitation and the emission wavelengths were 488 and 515 nm (⫾15 nm), respectively. For red fluorescent protein (RFP), excitation was at 543 nm and the emission wavelength was collected at 590 ⫾ 35 nm. The excitation wavelength for SYTO 62 was 637 nm, and emission over 660 nm (far-red fluorescence) was collected with a 660LP filter. The image acquisition software was LaserSharp 2000 (Bio-Rad). Images were analyzed with Volocity software (Improvision, Lexington, Mass.). All SCLM biofilm experiments were repeated a minimum of five times. The patterns shown were observed consistently. For confirmation of SCLM results on rhlA-gfp induction in biofilms, we employed a multiphoton laser microscope system (University of Iowa Carver College of Medicine Central Microscopy Core). Experiments with nonbiofilm planktonic cultures. Cultures of each strain were grown in tryptic soy broth. Growth measurements and fluorescence are the averages of triplicate experiments. Fluorescence was measured with a Tecan microtiter plate fluorometer. The excitation wavelength was 435 nm, and the emission wavelength was 535 nm. The relative fluorescent units represent the fluorescence values corrected for background (PAO1 without gfp).

RESULTS Expression of rhlA-gfp in planktonic cultures. To avoid issues related to plasmid copy number and stability in biofilms, we used strains with an rhlA-gfp reporter inserted as a single copy in the attB site on the chromosome (see Materials and Methods). As a first step in the analysis, we followed expression of gfp in planktonic cultures and we compared the expression patterns to the patterns of rhlAB operon expression known from previous studies (22, 25, 29). As expected, rhlA-gfp was not expressed in early and mid-logarithmic phase in the parent strain P. aeruginosa YL101. Fluorescence began to rise during the transition between logarithmic and stationary phase and continued to increase in stationary phase. The addition of C4-HSL did not advance the onset of gfp expression (Fig. 1A). These results are consistent with previous reports that the

VOL. 187, 2005

FIG. 1. Expression of rhlA-gfp in planktonic cultures of P. aeruginosa. (A) The quorum-sensing signal-producing strain YL101; (B) the rhlI mutant YL100. Growth (optical density at 600 nm) without added C4-HSL ({), with added C4-HSL (䊐), YL100 with the rhlI expression vector pYL125 (E), YL100 with the tac promoter-controlled rhlI expression vector pYL126 (⌬). Transcription of rhlA-gfp monitored as relative fluorescence (fluorescence in 50 ␮l of culture corrected for background as described in Materials and Methods). }, without added C4-HSL; ■, with C4-HSL; F, YL100 with pYL125; Œ, YL100 with pYL126.

rhlAB operon is not expressed until late in growth and that the expression pattern is not influenced by addition of C4-HSL to P. aeruginosa PAO1 (22, 25, 39). Also consistent with previous reports, expression of the rhlA-gfp fusion was dependent on C4-HSL. To show this dependence, we examined gfp expression in an rhlI mutant strain that is incapable of producing C4-HSL, P. aeruginosa YL100 (Fig. 1B). This strain showed little or no fluorescence in the absence of added C4-HSL, and it showed the typical pattern of fluorescence when C4-HSL was added to the culture medium or when the rhlI mutation was complemented with a plasmid-borne rhlI (either under control of its own promoter or under tac promoter control). For reasons that are unclear, when the rhlI mutant was supplied with

EXPRESSION OF rhlAB IN P. AERUGINOSA BIOFILMS

39

C4-HSL or an rhlI gene, fluorescence was about twice that of the parent strain (Fig. 1). Timing of rhlA-gfp expression in a developing biofilm. We followed biofilm development in flow chambers. In a first experiment we used P. aeruginosa YL101, the PAO1 derivative with an rhlA-gfp fusion inserted in the chromosomal attB site. The cells used to inoculate the flow chamber were fluorescent, but after 1 to 2 days cell fluorescence was lost (Fig. 2). Over the first 48 h, microcolonies formed but fluorescence of cells in the microcolonies remained low. After an additional 24 h we observed larger microcolonies that were 20 to 40 ␮m in height. These microcolonies showed elevated fluorescence (Fig. 2, 3 days). After 24 to 48 h of further growth we observed biofilms with typical mushroom-like structures. The stalk regions of these structures were brightly fluorescent, but the caps showed only low levels of GFP fluorescence (Fig. 2). Expression but not timing of rhlA-gfp in biofilms depends on C4-HSL. As is true of planktonic cultures (Fig. 1), biofilms of P. aeruginosa YL100, the rhlI mutant containing an rhlA-gfp chromosomal fusion, did not express GFP unless C4-HSL was added to the biofilm growth medium or a functional rhlI was introduced on a plasmid (data not shown). Expression of GFP in strain YL100 with pYL125 (containing rhlI controlled by its own promoter) appeared similar to expression of GFP in the rhlI wild-type strain YL101, except that in mushrooms GFP was not only expressed in the stalk region but it was also expressed in cells around the periphery of the caps (Fig. 3D and J). The timing of GFP expression was similar to that in the rhlI wild type. There was little fluorescence until microcolonies of at least 20 ␮m in thickness had developed (data not shown). When we added C4-HSL (at concentrations of 10 or 20 ␮M) to the medium flowing over a developing P. aeruginosa YL100 biofilm or when we grew biofilms of YL100 with a plasmidborne rhlI (pYL126) under constitutive tac promoter control, the timing and location of GFP expression were identical to the timing and location of GFP expression with this strain containing the rhlI expression plasmid pYL125 (Fig. 3). We conclude that C4-HSL is required for rhlA expression in biofilms and that the delay in the onset of rhlA expression is not due to C4-HSL limitation. Furthermore, biofilms of cells that overexpress rhlI show some GFP expression in the periphery of the caps in addition to fluorescence in the stalks (Fig. 3). This suggests that the lack of GFP fluorescence in the mushroom caps of the rhlI wild-type strain YL101 is not simply an artifact of the SCLM detection. There are many possible explanations as to why GFP is not expressed in the interior of the mushroom caps. As examples, perhaps cells within the caps are not metabolically active or perhaps there are unknown regulatory features. Spatial control of rhlA transcription in biofilms. The experiments shown in Fig. 2 and 3 suggest that rhlA-gfp is preferentially expressed in the nontwitching stalks of mushroom-like structures. However, the images might represent an SCLM artifact resulting from incomplete penetration of light through the thick biofilm. To address this concern we performed three separate experiments. First, we examined the fluorescence of biofilms when we used a P. aeruginosa strain carrying an arabinose promoter-controlled gfp (YL120). Second, we counterstained the biofilm with SYTO 62, which will cause all cells to fluoresce. Third, we followed fluorescence in a P. aeruginosa

40

LEQUETTE AND GREENBERG

J. BACTERIOL.

FIG. 2. Expression of rhlA-gfp in P. aeruginosa YL101 biofilm cells. GFP fluorescence was followed over a 5-day period by SCLM. Images of a horizontal section near the center of the biofilm for all time points are shown, and x-y reconstructions for day-3 to -5 images are also shown. On day 5 the biofilm was stained with SYTO 62. (A) GFP fluorescence; (B) red SYTO 62 fluorescence; (C) a merged image of the green and red fluorescence, where overlapping green and red appears yellow. Bars, 50 ␮m.

YL100 derivative carrying a plasmid-borne constitutive rfp gene in the presence of C4-HSL. When we added L-arabinose to the medium, the arabinose promoter-controlled gfp was expressed throughout 3-day-old biofilms regardless of microcolony thickness (Fig. 4, 3 days), and GFP expression was more uniform in 5-day mushroom structures (Fig. 4A) than it was in 5-day mushroom structures of rhlA-gfp strains. The fluorescence was not entirely uniform, and pockets near the tops of the caps appeared relatively dim. This might represent an artifact resulting from limited penetration of light through the biofilm to the tops of the caps. Nevertheless, the images were quite distinct from images with rhlA promoter-controlled gfp (Fig. 2A and 3D, E, and F). Red fluorescence from the constitutive rfp was evident in both stalks and caps (Fig. 3I), as was red fluorescence when biofilms were counterstained with SYTO 62 (Fig. 3G and H and 4B). All of these control experiments were consistent with the conclusion that transcription from the rhlAB promoter occurs in the stalks prior to cap formation and that transcription from the rhlAB promoter is relatively low in the cells that form the cap. In a further effort to validate our conclusion that rhlA-gfp is expressed preferentially in the stalks compared to araC-Pbadgfp, we made additional measurements and we used a different

imaging technology that reduces light-quenching artifacts. With images from SCLM we measured fluorescence intensity at different depths in biofilms of P. aeruginosa with either the rhlA-gfp or the Pbad-gfp control reporter (Fig. 5A). With the control there was a steady drop in fluorescence as the distance from the surface increased, but compared to the rhlA-gfpcontaining strain the fluorescence was more uniform throughout the biofilm. With the rhlA-gfp reporter the fluorescence decreased markedly as the biofilm reached a height of between 20 and 50 ␮m. One advantage that the newer imaging technology, multiphoton laser microscopy, has over SCLM is that fluorescence quenching is at a minimum. Thus, we used multiphoton laser microscopy to examine biofilms of P. aeruginosa carrying either rhlA-gfp or arabad-gfp (Fig. 5B and C, respectively). As expected, the green fluorescence in the rhlA-gfp strain was restricted to the stalk area of mushroom-like structures, whereas the Pbad-gfp biofilm fluorescence was more uniform. In fact, the lack of fluorescence in the caps of the strain carrying the rhlA-gfp reporter was even more obvious with multiphoton microscopy than it was with SCLM, while fluorescence in biofilms of the strain carrying the araC-Pbad-gfp fusion reporter was more uniform (compare Fig. 4A and 5C). We assume that this is because there is less noise with mul-

FIG. 3. Expression of rhlA-gfp in biofilms of the rhlI mutant P. aeruginosa YL100 containing rhlI expression vectors or exposed to C4-HSL. For P. aeruginosa YL100 with pmRFP1, we added C4-HSL (10 ␮M) to the medium. Without added C4-HSL, this construct did not show GFP fluorescence (data not shown). The red fluorescence in panels G and H and the composites J and K is from SYTO 62. There was no SYTO 62 staining in panels I and L, where the red fluorescence was from the plasmid-encoded RFP. Bar, 50 ␮m. Images are as described in the legend to Fig. 2. 41

42

LEQUETTE AND GREENBERG

FIG. 4. Expression araC-PBad-gfp in a P. aeruginosa PAO1 biofilm. The biofilm was grown in the presence of 0.2% L-arabinose. The images are from 3- and 5-day biofilms. (A) GFP fluorescence; (B) SYTO 62 fluorescence; (C) a merged image from panels A and B. Regions of overlapping green and red fluorescence appear yellow in panel C. Bar, 50 ␮m.

J. BACTERIOL.

FIG. 5. Measurement of GFP fluorescence in biofilms. (A) GFP fluorescence from rhlA-gfp in YL101 (E) or aracC-Pbad-gfp fluorescence in YL120 (Œ). Measurements were at intervals from the base (stalk) to the top (mushroom heads). Ten images were used to obtain average intensities. The ratios in parentheses were calculated by dividing average intensities between 0 and 20 ␮m by average intensities between 80 and 100 ␮m. (B to D) Biofilm grown in the presence of 0.2% L-arabinose. The images are from 5-day biofilms. In panels B and C, GFP fluorescence was acquired with a multiphoton laser system. Panels D and E are merged images from SYTO 62 dye fluorescence and GFP fluorescence. Regions of overlapping green and red fluorescence appear yellow in panels D and E. Bar, 50 ␮m.

EXPRESSION OF rhlAB IN P. AERUGINOSA BIOFILMS

VOL. 187, 2005

FIG. 6. Biofilm structure of the rhlI mutant. Strain PD100 harboring gfp under the constitutive ptac promoter (8) was grown in a biofilm without C4-HSL signal. The images represent constitutive GFP expression in a 5-day biofilm. Bar, 50 ␮m.

tiphoton technology. These experiments are consistent with the conclusion that rhlA-gfp is expressed primarily in stalks of mushroom-like structures. A previous report showed that biofilms of rhamnolipid synthesis mutants were abnormal, with a relatively flat architecture that did not have mushroom-like structures (7). An obvious hypothesis based on this finding and our findings is that rhamnolipid synthesis is required to allow migration of a motile subpopulation up the stalk to form caps. However, our investigators previously reported that an rhlI mutant produced structured biofilms in comparison to a lasI mutant (8). Because we have used different conditions than those used previously, because the imaging technology has improved, and because we now understand more about the subtleties of biofilm development, we have examined biofilms of an rhlI mutant containing a constitutively expressed stable gfp (Fig. 6). The biofilms we observed were not flat like the biofilms of rhamnolipid synthesis mutants described previously, nor did they appear similar to the wild-type strain. Rather, there were structures that had a pillar-like appearance. We believe these might represent uncapped stalks. DISCUSSION Previous investigations have established rhamnolipids as a key factor in defining P. aeruginosa biofilm structure (7). We also know that in planktonic cells genes for rhamnolipid synthesis are quorum sensing controlled. Thus, we were interested to begin to learn about when and where in P. aeruginosa biofilms rhamnolipid gene expression might occur. In planktonic cultures, the rhamnolipid synthesis operon rhlAB is controlled primarily by C4-HSL through its interaction with RhlR (23, 29). Moreover, there is additional poorly understood transcriptional control that manifests itself in a delay in rhlAB expression until late logarithmic-early stationary phase, at least in complex media, regardless of C4-HSL concentration (29, 31). By using RhlAB⫹ strains with an rhlA-gfp fusion inserted at a neutral site in the chromosome, we were able to follow transcription from an rhlA promoter during P. aeruginosa bio-

43

film development. A first point is that, as expected, rhlA-gfp transcription appeared to be dependent upon C4-HSL. Second, as is the case with cells in planktonic cultures, transcription from the rhlAB promoter was delayed during early biofilm development regardless of C4-HSL availability (Fig. 2). The biofilm flow chambers were inoculated with fluorescent cells, but fluorescence of attached cells was lost within hours and cell fluorescence remained relatively low until microcolonies with a thickness of 20 ␮m or greater had developed. When we added 3OC12-HSL in addition to C4-HSL in experiments with P. aeruginosa YL100, the pattern of rhlA-gfp expression appeared similar to the pattern we observed with C4-HSL alone (data not shown). The rhlAB promoter region is complex, and planktonic cells show RpoN-dependent rhlAB expression (13, 29). We found that GFP fluorescence in biofilms of an rpoN mutant containing the rhlA-gfp chromosomal fusion was indistinguishable from that in the wild type (data not shown). Thus, we conclude that the delay in rhlA-gfp expression of biofilm cells and its restriction to stalks are not related to rpoN expression, at least in the P. aeruginosa strains used in our studies. Our images indicate that rhlA expression begins in microcolonies, and GFP fluorescence can be observed not only in microcolonies but also in the stalks of mushroom-like structures as the biofilm continues to grow. However, the fluorescence seems to be restricted to the stalk, and there is very little fluorescence in the mushroom caps even when C4-HSL is added to the medium flowing over the biofilms (Fig. 2 and 3). Because we used a stable GFP, we do not have information on whether transcription from the rhlA promoter was transient, perhaps restricted to a stage of development prior to cap formation, or whether it persisted in the stalks even after cap formation had commenced. We chose to use a stable GFP because we believed this type of reporter, which gives a history of gene expression, would best reflect where rhamnolipids might exist in the biofilm. We were quite concerned that images showing GFP fluorescence almost exclusively in the stalk region (Fig. 2 and 3) might result from SCLM artifacts, and we performed a number of control experiments to address this issue. These included controls with gfp transcribed from a promoter other than the rhlA promoter, counterstaining with a red fluorescent bacterial cell stain (Fig. 3 and 4), and fluorescence measurements and acquisition of images with a multiphoton laser microscope system (Fig. 5). All of our experiments led to the view that rhlA is expressed primarily in the stalk region of mushroom-like structures. Collectively, our data and other reports are consistent with the following model for the development of mushroomlike structures by P. aeruginosa growing under flow on a glass surface. Cells of P. aeruginosa attach to and move on a surface by means of twitching motility. Often after cell division the two daughter cells stop twitching and begin to form a microcolony (34). We do not know whether the cessation of twitching motility results from a loss of the twitching motility apparatus or whether it might result from cells adhering to each other or the surface in a special manner. The microcolonies develop into stalks that contain bacteria that do not exhibit twitching motility (18). Our current evidence suggests that rhamnolipids may be synthesized specifically in the growing stalks. A subpopulation of cells will migrate up the stalks by twitching motility to form the mushroom cap (18). We suggest that rham-

44

LEQUETTE AND GREENBERG

nolipids on the stalk surface function as a surfactant that enables a population moving by twitching motility to migrate on the stalk and form the cap. This might at least in part provide a plausible explanation for why rhamnolipid synthesis mutants form abnormal biofilm structures (7). ACKNOWLEDGMENTS We thank Michael Jacobs, Colin Manoil, and Matthew Parsek for providing P. aeruginosa strains ahead of publication, and we thank Pradeep Singh for helpful discussions. We thank Thomas Moninger and the University of Iowa Central Microscopy Core for help and support with multiphoton microscopy. Support was provided by grants from the National Institute of General Medicine (GM59026) and from the W.M. Keck Foundation. REFERENCES 1. Becher, A., and H. P. Schweizer. 2000. Integration-proficient Pseudomonas aeruginosa vectors for isolation of single-copy chromosomal lacZ and lux gene fusions. BioTechniques 29:948–950. 2. 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. 3. Burns, J. L., B. W. Ramsey, and A. L. Smith. 1993. Clinical manifestations and treatment of pulmonary infections in cystic fibrosis. Adv. Pediatr. Infect. Dis. 8:53–66. 4. Campbell, R. E., O. Tour, A. E. Palmer, P. A. Steinbach, G. S. Baird, D. A. Zacharias, and R. Y. Tsien. 2002. A monomeric red fluorescent protein. Proc. Natl. Acad. Sci. USA 99:7877–7882. 5. Costerton, J. W. 1995. Overview of microbial biofilms. J. Ind. Microbiol. 15:137–140. 6. Costerton, J. W., P. S. Stewart, and E. P. Greenberg. 1999. Bacterial biofilms: a common cause of persistent infections. Science 284:1318–1322. 7. Davey, M. E., N. C. Caiazza, and G. A. O’Toole. 2003. Rhamnolipid surfactant production affects biofilm architecture in Pseudomonas aeruginosa PAO1. J. Bacteriol. 185:1027–1036. 8. 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. 9. 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. 10. De Lorenzo, V., and K. N. Timmis. 1994. Analysis and construction of stable phenotypes in gram-negative bacteria with Tn5- and Tn10-derived minitransposons. Methods Enzymol. 235:386–405. 11. Fuqua, C., and E. P. Greenberg. 2002. Listening in on bacteria: acyl-homoserine lactone signalling. Nat. Rev. Mol. Cell Biol. 3:685–695. 12. Hentzer, M., H. Wu, J. B. Andersen, K. Riedel, T. B. Rasmussen, N. Bagge, N. Kumar, M. A. Schembri, Z. Song, P. Kristoffersen, M. Manefield, J. W. Costerton, S. Molin, L. Eberl, P. Steinberg, S. Kjelleberg, N. Hoiby, and M. Givskov. 2003. Attenuation of Pseudomonas aeruginosa virulence by quorum sensing inhibitors. EMBO J. 22:3803–3815. 13. Heurlier, K., V. Denervaud, G. Pessi, C. Reimmann, and D. Haas. 2003. Negative control of quorum sensing by RpoN (␴54) in Pseudomonas aeruginosa PAO1. J. Bacteriol. 185:2227–2235. 14. Hoang, T. T., R. R. Karkhoff-Schweizer, A. J. Kutchma, and H. P. Schweizer. 1998. A broad-host-range Flp-FRT recombination system for site-specific excision of chromosomally located DNA sequences: application for isolation of unmarked Pseudomonas aeruginosa mutants. Gene 212:77–86. 15. Hoang, T. T., A. J. Kutchma, A. Becher, and H. P. Schweizer. 2000. Integration-proficient plasmids for Pseudomonas aeruginosa: site-specific integration and use for engineering of reporter and expression strains. Plasmid 43:59–72. 16. Hoiby, N. 1993. Cystic fibrosis and endobronchial Pseudomonas infection. Curr. Opin. Pediatr. 5:247–254. 17. Holloway, B. W., V. Krishnapillai, and A. F. Morgan. 1979. Chromosomal genetics of Pseudomonas. Microbiol. Rev. 43:73–102. 18. Klausen, M., A. Aaes-Jorgensen, S. Molin, and T. Tolker-Nielsen. 2003. Involvement of bacterial migration in the development of complex multicellular structures in Pseudomonas aeruginosa biofilms. Mol. Microbiol. 50:61– 68. 19. Klausen, M., A. Heydorn, P. Ragas, L. Lambertsen, A. Aaes-Jorgensen, S. Molin, and T. Tolker-Nielsen. 2003. Biofilm formation by Pseudomonas

J. BACTERIOL.

20. 21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32. 33. 34.

35. 36.

37.

38.

39.

40.

41.

42.

aeruginosa wild type, flagella and type IV pili mutants. Mol. Microbiol. 48:1511–1524. Kolter, R., and R. Losick. 1998. One for all and all for one. Science 280: 226–227. Latifi, A., M. K. Winson, M. Foglino, B. W. Bycroft, G. S. A. B. Stewart, A. Lazdunski, and P. Williams. 1995. Multiple homologues of LuxR and LuxI control expression of virulence determinants and secondary metabolites through quorum sensing in Pseudomonas aeruginosa PAO1. Mol. Microbiol. 17:333–343. Medina, G., K. Juarez, and G. Soberon-Chavez. 2003. The Pseudomonas aeruginosa rhlAB operon is not expressed during the logarithmic phase of growth even in the presence of its activator RhlR and the autoinducer N-butyryl-homoserine lactone. J. Bacteriol. 185:377–380. Medina, G., K. Juarez, B. Valderrama, and G. Soberon-Chavez. 2003. Mechanism of Pseudomonas aeruginosa RhlR transcriptional regulation of the rhlAB promoter. J. Bacteriol. 185:5976–5983. Miller, W. G., J. H. Leveau, and S. E. Lindow. 2000. Improved gfp and inaZ broad-host-range promoter-probe vectors. Mol. Plant Microbe Interact. 13: 1243–1250. 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. Parsek, M. R., D. L. Val, B. L. Hanzelka, J. E. Cronan, Jr., and E. P. Greenberg. 1999. Acyl homoserine-lactone quorum-sensing signal generation. Proc. Natl. Acad. Sci. USA 96:4360–4365. Pearson, J. P., K. M. Gray, L. Passador, K. D. Tucker, A. Eberhard, B. H. Iglewski, and E. P. Greenberg. 1994. Structure of the autoinducer required for expression of Pseudomonas aeruginosa virulence genes. Proc. Natl. Acad. Sci. USA 91:197–201. Pearson, J. P., L. Passador, B. H. Iglewski, and E. P. Greenberg. 1995. A second N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 92:1490–1494. 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. Pesci, E. C., J. P. Pearson, P. C. Seed, and B. H. Iglewski. 1997. Regulation of las and rhl quorum sensing in Pseudomonas aeruginosa. J. Bacteriol. 179:3127–3132. Schuster, M., C. P. Lostroh, T. Ogi, and E. P. Greenberg. 2003. Identification, timing, and signal specificity of Pseudomonas aeruginosa quorum-controlled genes: a transcriptome analysis. J. Bacteriol. 185:2066–2079. Schweizer, H. P. 1991. Escherichia-Pseudomonas shuttle vectors derived from pUC18/19. Gene 97:109–121. Sharma, S., P. Sachdeva, and J. S. Virdi. 2003. Emerging water-borne pathogens. Appl. Microbiol. Biotechnol. 61:424–428. Singh, P. K., M. R. Parsek, E. P. Greenberg, and M. J. Welsh. 2002. A component of innate immunity prevents bacterial biofilm development. Nature 417:552–555. Smith, R. S., and B. H. Iglewski. 2003. Pseudomonas aeruginosa quorumsensing systems and virulence. Curr. Opin. Microbiol. 6:56–60. Wagner, V. E., D. Bushnell, L. Passador, A. I. Brooks, and B. H. Iglewski. 2003. Microarray analysis of Pseudomonas aeruginosa quorum-sensing regulons: effects of growth phase and environment. J. Bacteriol. 185:2080–2095. Whitehead, N. A., A. M. Barnard, H. Slater, N. J. Simpson, and G. P. Salmond. 2001. Quorum-sensing in gram-negative bacteria. FEMS Microbiol. Rev. 25:365–404. Whiteley, M., and E. P. Greenberg. 2001. Promoter specificity elements in Pseudomonas aeruginosa quorum-sensing-controlled genes. J. Bacteriol. 183: 5529–5534. Whiteley, M., K. M. Lee, and E. P. Greenberg. 1999. Identification of genes controlled by quorum sensing in Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 96:13904–13909. Winson, M. K., M. Camara, A. Latifi, M. Foglino, S. R. Chhabra, M. Daykin, M. Bally, V. Chapon, G. P. Salmond, B. W. Bycroft, A. Lazdunski, G. S. Stewart, and P. Williams. 1995. Multiple N-acyl-L-homoserine lactone signal molecules regulate production of virulence determinants and secondary metabolites in Pseudomonas aeruginosa. Proc. Natl. Acad. Sci. USA 92:9427– 9431. Woodcock, D. M., P. J. Crowther, J. Doherty, S. Jefferson, E. DeCruz, M. Noyer-Weidner, S. S. Smith, M. Z. Michael, and M. W. Graham. 1989. Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17:3469–3478. Yanisch-Perron, V., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequence of the M13mp18 and pUC19 vectors. Gene 33:103–119.