SEARCH

SEARCH BY CITATION

Keywords:

  • c-di-GMP;
  • Pel;
  • pellicle;
  • SigX;
  • sucrose

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Pseudomonas aeruginosa biofilm formation was increased by addition of sucrose to Luria–Bertani medium, whereas addition of NaCl to a final similar osmolarity and use of maltose instead of sucrose, were ineffective. In a previous study, we showed that the extracytoplasmic sigma factor SigX is activated in the presence of sucrose. The sucrose-mediated pellicle increase was abolished in a sigX mutant strain. Sucrose addition led to an increase in pel expression and cyclic-diguanylate (c-di-GMP) pool level production. Interestingly, these two phenotypes were strongly decreased in a sigX mutant. Since pel is not known as a SigX-target, we suspect SigX to be involved in the c-di-GMP production. We found that expression of the diguanylate cyclase PA4843 gene was increased in the presence of sucrose at least partly through SigX activity. Our study shows that sucrose itself rather than osmolarity favours the biofilm mode of P. aeruginosa through the activation of SigX.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Pseudomonas aeruginosa is a ubiquitous Gram-negative bacterium capable of surviving in a broad range of natural environments including water and soils, although it is best known as an opportunistic human pathogen associated with antibiotic-resistant hospital-acquired infections and chronic infections (Filloux & Vallet, 2003; Buchanan et al., 2009). Chronic infections have been associated with the formation of antibiotic-resistant biofilms. Biofilms are surface-associated microbial communities encased within an extracellular matrix that is critical for building and maintaining the biofilm structure, and which functions as a structural scaffold and/or a protective barrier against harsh environments (Mah & O'Toole, 2001; Sutherland, 2001; Branda et al., 2005; Ryder et al., 2007; Harmsen et al., 2010). The matrix is composed of extracellular DNA, proteins, vesicles and exopolysaccharides. Pseudomonas aeruginosa is able to produce at least three exopolysaccharides types: Psl, Pel, and alginate. Alginate is a capsular polysaccharide virulence factor that confers the characteristic mucoid phenotype of P. aeruginosa in the cystic fibrosis airway (Wozniak et al., 2003; Ryder et al., 2007). In nonmucoid strains, Psl and/or Pel are the major exopolysaccharides components of the biofilm matrix. Psl contains a repeating pentasaccharide consisting of d-mannose, d-glucose and l-rhamnose (Byrd et al., 2009), which serves as a scaffold and as a signalling molecule to stimulate biofilm formation (Irie et al., 2012). Noticeably, Psl has been shown to be produced only by certain P. aeruginosa strains, such as PAO1 but not PA14 (Friedman & Kolter, 2004b). In the latter strain, the major exopolysaccharides produced is the glucose-rich and cellulase-sensitive Pel. Pel is required for the formation of pellicles, which develop at the air–liquid interface in standing cultures. It is acknowledged that transcription of pel and psl gene clusters is highly regulated, at least via the cyclic-diguanylate (c-di-GMP) pool level, which is required to increase biofilm formation (Mikkelsen et al., 2011; Romling et al., 2013).

Sucrose is one of the carbohydrates commonly found in the rhizosphere, creating an attractive niche for many microorganisms (Berg et al., 2005; Reid & Abratt, 2005), and also a major component of the dental plaque, in which it causes major biochemical and physiological changes during the process of biofilm formation, leading to dental caries (Paes Leme et al., 2006). As high sucrose concentrations have been associated with increased medium osmolarity, it has been suggested that osmoadaptation is an important parameter for successful bacterial colonization (Miller & Wood, 1996). Despite the huge distribution of sucrose, little is known about the effects of high sucrose concentrations on bacteria. In this study, we show that sucrose (but not osmolarity) promotes biofilm formation of P. aeruginosa through a mechanism leading to increased c-di-GMP production at least partly via the extracytoplasmic function (ECF) sigma factor SigX.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Bacterial strains and culture conditions

The strains and plasmids are listed in Supporting Information, Table S1. Bacteria were grown at 37 °C on a rotary shaker (180 r.p.m.) in Luria–Bertani (LB) broth containing 171 mM NaCl. LB was supplemented with 171 mM NaCl (LBN), 1, 2, 5 or 11.7 % (342 mM, LBS) sucrose or with 11.7% maltose (LBM). Cultures were inoculated at an initial OD580 nm of 0.08. For Escherichia coli, 10 μg mL−1 gentamycin (Gm), 100 μg mL−1 ampicillin (Ap), or 15 μg mL−1 tetracycline (Tc) were used when required. For P. aeruginosa, 50 μg mL−1 Gm, 300 μg mL−1 carbenicillin (Cb) or 250 μg mL−1 Tc were used when required. Pseudomonas aeruginosa sucrose utilization was assayed on the mineral medium ‘Ayers, Rupp and Johnson’ (ARJ), supplemented with 0.2 % glucose or sucrose.

Pellicle formation assays

Pellicles were allowed to form as previously described (Friedman & Kolter, 2004a) with minor modifications. Standing cultures were grown in 5 mL of medium in glass tubes, and the pellicles were evaluated by visual inspection of the air–liquid interface after 96 h of growth at 37 °C.

Biofilm formation assays

Quantification of biofilm was performed in 24-well polystyrene microtitre plates (Nunc). LB medium (2 mL per well) was inoculated to a final OD580 nm of 0.08 and incubated at 37 °C without shaking for 24 h. After removal of the standing culture, the wells were washed with distilled water and then stained with 3 mL of 0.4% crystal violet (v/v) for 20 min at room temperature, before the residual stain was removed. After washing, the stained biofilms were resuspended in 3 mL of 100% ethanol and absorbance was measured at 595 nm.

pelA inactivation

The ΔpelA mutant of P. aeruginosa PAO1 was obtained by allelic exchange as previously described (Quénée et al., 2005; Bazire et al., 2010; Bouffartigues et al., 2012), using primers pelA1-pelA2 and pelA3-pelA4, leading to a gene deletion of 2779 bp.

pel transcriptional fusion

The 500-bp promoter region of pelA was PCR-amplified using upelFsacI/upelRspeI primers, digested by SacI and SpeI, and fused to the promoter-less luxCDABE cassette in pAB133, yielding pABpel (Bazire et al., 2005). The insert was verified by DNA sequencing with SpAB and ASpAB primers.

Bioluminescence assays

Pseudomonas aeruginosa strains containing pAB133-derived plasmids were inoculated in LB or LBS to a final OD580 nm of 0.01 in white 96-well optiplates with a flat transparent bottom (BD Falcon, San Jose, CA). Bioluminescence and absorbance at 580 nm were measured simultaneously to monitor both the promoter activity and bacterial growth, respectively, using a multimode plate reader (Xenius, SAFAS) as previously described (Bouffartigues et al., 2012).

Quantitative RT-PCR

RNA extraction, cDNA synthesis and real-time PCR were achieved as previously described (Gicquel et al., 2013) using the primers listed in Table S2 (Fito-Boncompte et al., 2011).

c-di-GMP assays

The c-di-GMP-responsive cdrA promoter-gfp fusion-encoding vector was used to monitor the c-di-GMP level in H103 and PAOSX strains (Rybtke et al., 2012). Transformed strains were grown to stationary phase in LB or LBS supplemented with 50 μg mL−1 Gm. The fluorescence was measured on 109 cells in a 96-black well microtitre plate (Nunc), as arbitrary fluorescence intensity units (FIU) using a Xenius multimode plate reader (SAFAS, excitation: 490 nm, emission: 515 nm).

Statistical analyses

All experiments were carried out at least three times. Significances of differences between mean values were assessed using Student's t-test with significance set at *P < 0.05, **< 0.01 and ***P < 0.001.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Sucrose, but not NaCl or maltose, increases biofilm production

Pseudomonas aeruginosa can form robust pellicles at the air–liquid interface of a standing liquid culture. The addition of 342 mM sucrose to LB led to the production of a markedly thicker pellicle (Fig. 1a, LBS), which was not observed when adding 171 mM NaCl instead of sucrose (Fig. 1a, LBN). Similar results were observed when biofilm was grown in microtitre wells and stained with crystal violet (Fig. 1b). This suggests that sucrose rather than osmolarity triggers the increase of biofilm formation in our conditions. Using maltose instead of sucrose at the same concentration (342 mM) did not increase the pellicle, suggesting that it was the sucrose that led to the increased pellicle formation in our conditions (Fig. 1a and b, LBM). As pellicle formation has been associated with pel and psl gene cluster expression in P. aeruginosa PAO1 (Friedman & Kolter, 2004b), the transcription levels of pelB and pslB were compared using qRT-PCR in P. aeruginosa H103 grown in LB and LBS (Fig. 1c). The pelB mRNA level was increased almost 3.5-fold in the presence of 342 mM sucrose, suggesting that sucrose promotes pel transcription (Fig. 1c). Conversely, the pslB mRNA levels were found to be very similar in LB and LBS, suggesting that the psl gene cluster was not mainly involved in the sucrose-related biofilm production increase (Fig. 1c). As a control, the effect of sucrose on pellicle production was assayed in a pelA mutant strain (ratio LBS/LB). As shown on Fig. 1d, the pelA mutant strain produced about twofold more pellicle in LBS than LB, and the isogenic parental strain produced about fourfold more pellicle in LBS than LB. These data suggest that Pel polysaccharide was partly involved in the sucrose-mediated pellicle production. However, this also indicates that another part of the sucrose effect on biofilm production remained pel-independent. Biofilm-enhanced production by exogenous polysaccharides has been described previously. Sucrose is the most cariogenic carbohydrate, leading to elevated matrix exopolysaccharides production by oral bacteria and accumulation on the tooth surface (Paes Leme et al., 2006). One of these bacteria, Streptococcus mutans, was shown to form well-defined and tightly adherent biofilms in the presence of 1% of sucrose (Duarte et al., 2008), but not in high salinity media (Kawarai et al., 2009). In the plant growth-promoting bacterium Bacillus subtilis, biofilm formation is triggered by plant-produced polysaccharides such as arabinogalactan, pectin or xylan at very low concentrations (0.5% or even 0.05% of plant arabinogalactan) (Kearns et al., 2005; Beauregard et al., 2013). Plant polysaccharides stimulate biofilm formation in B. subtilis by acting as environmental signals that induce matrix gene expression, and as substrates that are processed and incorporated into the exopolysaccharides biofilm matrix (Beauregard et al., 2013). Interestingly, it was shown previously that addition of glucose at a concentration of 1–5% in LB medium increased production of the biofilm-related polysaccharide alginate in P. aeruginosa PAO1 (Ma et al., 1997). In our study, we assayed sucrose at a very high concentration, that is 342 mM, corresponding to 11.7%. We further assayed lower sucrose concentrations to investigate the putative signalling effect of sucrose on biofilm formation. As shown on Fig. 2a, the sucrose-related effect on biofilm formation occurred only with elevated sucrose concentrations (higher than 5%), suggesting that sucrose did not act as a signal in our conditions. Pseudomonas aeruginosa is furthermore unable to metabolize sucrose due to the lack of the sucrose-hydrolyzing corresponding enzymes (Fig. 2b), suggesting that sucrose is thus unlikely to be used as a carbohydrate source for the building of P. aeruginosa matrix. The mechanisms by which sucrose might increase P. aeruginosa biofilm formation will be further discussed below.

image

Figure 1. Sucrose-increased biofilm formation of Pseudomonas aeruginosa H103. (a) Pellicle and crystal violet (CV)-stained biofilms formed in microtitre wells by P. aeruginosa H103 grown in LB, LBS (LB + 342 mM sucrose), LBN (LB + 171 mM NaCl) or LBM (LB + 342 mM maltose) media. (b) Measurement of the biofilm produced in LBS (grey bar), LBN (white bar) and LBM (light grey bar) relative to that formed in LB (dotted line). (c) qRT-PCR assays on pelB and pslB in P. aeruginosa H103 grown in LB and LBS media. PCR reactions were performed in triplicate and the standard deviations were lower than 0.15 Ct. (d) Quantification of CV-stained biofilms formed by the wild-type strain PAO1 (grey bar) and its isogenic pelA mutant strain (white bar) grown in LBS compared with LB. Results are given as the ratio LBS/LB. Statistical analysis used pairwise strain comparisons (t-test). *P < 0.05; **< 0.01; ns, non-significant.

Download figure to PowerPoint

image

Figure 2. Effect of sucrose concentration on Pseudomonas aeruginosa biofilm formation. (a) Pellicle at the air–liquid interface and quantification of CV-stained biofilms by P. aeruginosa H103 grown in LB or in LB supplemented with 1, 2, 5 or 11.7% sucrose, relative to that formed in LB. (b) Growth of P. aeruginosa H103 in mineral medium ARJ containing sucrose or glucose as a carbon source. Statistical analysis used pairwise strain comparisons (t-test). *< 0.05; **< 0.01; ns, non-significant.

Download figure to PowerPoint

ECF sigma factor SigX is involved in the sucrose-related biofilm increase

SigX is an ECF sigma factor that is involved in fatty acid synthesis (Boechat et al., 2013; Blanka et al., 2014) and in biofilm formation (Gicquel et al., 2013; Blanka et al., 2014). As we have previously shown that SigX activity was increased in high sucrose-containing medium (Bouffartigues et al., 2012), we suspected SigX to be involved in the molecular mechanism linking sucrose and biofilm formation. The lack of SigX abolished the sucrose-promoting effect on biofilm formation (Fig. 3a and b). Complementing the sigX mutation (PAOSX+) restored this phenotype, showing that SigX triggers the sucrose-related effect on biofilm production (Fig. 3a and b). Using a transcriptional fusion between the pelA promoter region and the luxCDABE cassette, pelA expression was strongly increased (by about 10-fold) in response to sucrose in the wild-type strain (see qRT-PCR experiments in Fig. 1c). In the PAOSX mutant, expression of pelA was reduced by about twofold. The pelA mRNA level was increased almost 3.4-fold in the presence of 342 mM sucrose in the sigX mutant (Fig. 1c). The data suggest that SigX was mainly involved in the expression of the sucrose-related pel gene cluster, but also that another part of the sucrose effect on pel expression remains SigX-independent. In a previous study, we achieved a comparative transcriptomic study between P. aeruginosa H103 and its isogenic sigX mutant PAOSX grown in M9 minimal medium (Gicquel et al., 2013). In these conditions, the pel gene cluster was not identified as a putative SigX target (Gicquel et al., 2013). In a study integrating chip-on-chip experiments and transcriptomic assays based on P. aeruginosa PA14 and its isogenic sigX mutant, pel was not identified as a direct SigX target (Blanka et al., 2014). Furthermore, we were unable to find a putative SigX consensus binding motif (Blanka et al., 2014) in the promoter region of pelA (data not shown). The data suggest that the sucrose-related increase of pelA expression is indirectly linked to SigX.

image

Figure 3. SigX is involved in sucrose induction of the pellicle production. Pellicle at the air–liquid interface (a) and quantification of CV-stained biofilms (b) formed by Pseudomonas aeruginosa H103 grown in LB (black bars) or LBS (grey bars), relative to that formed in LB by P. aeruginosa H103. (c) Transcriptional activity (reporter assay) of the pelA promoter region in P. aeruginosa H103 or PAOSX (isogenic sigX mutant) grown in LB (black bars) and in LBS (grey bars). Statistical analysis used pairwise strain comparisons (t-test). *P < 0.05; **< 0.01; ***P < 0.001; ns, non-significant.

Download figure to PowerPoint

Sucrose increases the c-di-GMP pool level at least partly through SigX

As pel transcription is positively regulated by c-di-GMP, we hypothesized that growth in LBS could affect the c-di-GMP pool level. We thus assayed c-di-GMP intracellular levels using the reporter constructed by Rybtke et al. (2012) that is based on transcriptional fusion between the c-di-GMP-responsive cdrA promoter and gfp. As shown on Fig. 4a, the activity of this promoter region was increased 1.7-fold in H103 wild-type strain grown in LBS compared with LB, indicating that the c-di-GMP pool level was increased in the presence of sucrose. In addition to pel, psl is also regulated by the c-di-GMP pool level. The question is then, how could c-di-GMP affect pel but not psl transcription (Mikkelsen et al., 2011)? Interestingly, it has been shown recently that the c-di-GMP produced by the diguanylate cyclase WspR specifically affects Pel but not Psl synthesis (Huangyutitham et al., 2013). These authors suggested that (1) phenotypes that are very dominant in P. aeruginosa may not be greatly affected by the small c-di-GMP increase (this could be the case for Psl, which is the exopolysaccharides mostly responsible for attachment and biofilm formation in P. aeruginosa strain PAO1 (Colvin et al., 2012); and (2) the differential effects on c-di-GMP-sensitive processes may be due to differences in the affinities of receptor proteins for c-di-GMP (Hickman & Harwood, 2008; Whitney et al., 2012). However, as mentioned above, a part of the sucrose-mediated effect on biofilm production remained pel-independent, and thus may be also unrelated to c-di-GMP.

image

Figure 4. Sucrose increased the c-di-GMP level via the ECF sigma factor SigX. (a) c-di-GMP level assay (relative FIU) using the cdrA-gfp transcriptional fusion reporter system (Rybtke et al., 2012) in H103 (black bars) and PAOSX (sigX mutant, grey bars) grown in LB or LBS. (b) Quantification of PA4843 mRNA level assayed using qRT-PCR on RNAs extracted from Pseudomonas aeruginosa H103 and PAOSX grown in LB medium. (c) Quantification of PA4843 mRNA level assayed by qRT-PCR on RNAs extracted from P. aeruginosa H103 or PAOSX sigX mutant strain grown in LBS compared with LB. PCR reactions were performed in triplicate and the standard deviations were lower than 0.15 Ct. Statistical analysis used pairwise strain comparisons (t-test). *< 0.05; **< 0.01.

Download figure to PowerPoint

We next assayed the effects of sucrose on the c-di-GMP level in the sigX mutant strain PAOSX. As shown on Fig. 4a, the activity of the cdrA promoter region was maintained at a similar level in PAOSX grown in the presence or absence of sucrose (Fig. 4a), showing that SigX triggers P. aeruginosa response to sucrose. However, these data suggest also that the c-di-GMP level was not drastically affected by the sigX mutation. This result should be viewed with caution; although it is useful to detect increases in c-di-GMP levels, the reporter system we used to detect c-di-GMP intracellular levels modulation might be less suitable for detecting decreases in fluorescence (Rybtke et al., 2012). Otherwise, the depletion in total intracellular c-di-GMP due to a sigX mutation may be small. Interestingly, a mutant lacking the WspR diguanylate cyclase that is involved in c-di-GMP production only slightly reduced the total intracellular c-di-GMP level (Hickman et al., 2005). Overall, our data show that SigX triggers the P. aeruginosa H103 increase of the c-di-GMP level in response to a high sucrose concentration in LB. The question is then, how could SigX increase the c-di-GMP level? The PA4843 gene encoding a diguanylate cyclase was suggested to be a putative SigX-target (Gicquel et al., 2013; Blanka et al., 2014). As shown on Fig. 4b, PA4843 expression was lowered about twofold in PAOSX mutant compared with H103 strain grown in LB, confirming the involvement of SigX in the expression of PA4843. Sucrose significantly increased PA4843 transcription by about 2.5-fold in H103 (ratio LBS/LB), and 1.5-fold in PAOSX (ratio LBS/LB), showing that SigX was partly involved in the sucrose-related increase of PA4843 expression (Fig. 4c). However, this also indicates that another part of the PA4843 stimulation remains SigX-independent. Although further investigations are needed to confirm the involvement of PA4843 in the c-di-GMP level increase in response to sucrose, our data give a first hint as to the molecular mechanism linking sucrose, the ECF sigma factor SigX and c-di-GMP. Noticeably, in E. coli, at least five of the 29 genes encoding diguanylate cyclases or phosphodiesterases are stationary phase-induced under the control of the general stress response sigma factor RpoS, and also exhibit differential control by additional environmental and temporal signals (Weber et al., 2006; Landini, 2009).

How does sucrose activate SigX?

In this study, we show that sucrose increases biofilm formation and c-di-GMP production at least partly via the ECF sigma factor SigX. The mechanisms by which sucrose increased P. aeruginosa biofilm production appear highly complex, as P. aeruginosa is unable to metabolize sucrose due to the lack of the corresponding sucrose-hydrolyzing enzymes (Fig. 2b). Sucrose is thus unlikely to be used as a carbohydrate source for the building of P. aeruginosa matrix. One possibility would be that sucrose itself may act as a potent signal, as plant polysaccharides in the case of B. subtilis, activating a regulatory pathway including the ECF sigma factor SigX, and increasing the c-di-GMP pool level. However, as shown elsewhere, this effect on P. aeruginosa biofilm production was only observed with high concentrations of sucrose, which are 10–20-fold higher than the concentrations of plant polysaccharides required for signalling in B. subtilis (Beauregard et al., 2013). Thus our data suggest that sucrose may not act as an environmental signal on P. aeruginosa. Alternatively, high sucrose-containing environments may produce a signal that in turn could activate such a regulatory pathway. In a previous study, we have shown that addition of sucrose strongly disturbs the bacterial envelope (Bouffartigues et al., 2012). This hypothesis is strengthened by the greater expression of SigX when the master cell wall stress responsive ECF sigma factor AlgU is activated, as in the case of the P. aeruginosa response to d-cycloserine (Wood et al., 2006), or to a low shear-modelled microgravity (Crabbé et al., 2010). Interestingly, the involvement of SigX in short chain fatty acid synthesis (Boechat et al., 2013; Blanka et al., 2014) further suggests that SigX may be involved in membrane remodelling and fluidity (Duchesne et al., 2013). It is thus tempting to speculate that sucrose might induce a mechanic stress on the cell envelope that leads to SigX activity and c-diGMP production, thus enhancing the bacterial biofilm mode.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We are grateful to T. Tolker-Nielsen for providing the cdrA-gfp fusion containing vector, K. Shioya for pelA mutant construction, and P. Cornelis for helpful discussion. This work was supported by the French Government, the Région Haute Normandie (GRRs Sésa and IRIB), the Grand Evreux Agglomération and European FEDER funds.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
fml12482-sup-0001-TableS1.docWord document27KTable S1. Bacterial strains and plasmids used in this study.
fml12482-sup-0002-TableS2.docWord document28KTable S2. Primers used for cloning and qRT-PCR assays.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.