Out of 8000 candidates from a genetic screening for Pseudomonas putida KT2442 mutants showing defects in biofilm formation, 40 independent mutants with diminished levels of biofilm were analyzed. Most of these mutants carried insertions in genes of the lap cluster, whose products are responsible for synthesis, export and degradation of the adhesin LapA. All mutants in this class were strongly defective in biofilm formation. Mutants in the flagellar regulatory genes fleQ and flhF showed similar defects to that of the lap mutants. On the contrary, transposon insertions in the flagellar structural genes fliP and flgG, that also impair flagellar motility, had a modest defect in biofilm formation. A mutation in gacS, encoding the sensor element of the GacS/GacA two-component system, also had a moderate effect on biofilm formation. Additional insertions targeted genes involved in cell envelope function: PP3222, encoding the permease element of an ABC-type transporter and tolB, encoding the periplasmic component of the Tol-OprL system required for outer membrane stability. Our results underscore the central role of LapA, suggest cross-regulation between motility and adhesion functions and provide insights on the role of cell envelope trafficking and maintenance for biofilm development in P. putida.
INTRODUCTION
In natural environments, bacteria are most often found associated with interfaces in highly structured communities known as biofilms (Costerton et al. 1995; Davey and O'Toole 2000). Development of such communities in which cells are encased in a shared self-produced polymeric extracellular matrix, endows bacteria with improved environmental fitness, including increased resistance to water limitation, presence of toxicants or predation (Costerton, Stewart and Greenberg 1999; van de Mortel and Halverson 2004; Matz and Kjelleberg 2005; Szomolay et al. 2005).
Pseudomonas putida is a soil bacterium that has been deeply characterized because of its wide metabolic versatility, including biodegradation of xenobiotic and aromatic pollutants (Martin dos Santos et al. 2004). The model strain P. putida KT2440 and its Rifampicin-resistant derivative KT2442 have been shown to form biofilms efficiently on diverse surfaces (Espinosa-Urgel, Salido and Ramos 2000; López-Sánchez et al. 2013). Biofilm development has been described to occur as a series of highly regulated steps: attachment, microcolony formation, maturation and dispersal in multiple species. Upon initial contact with a surface, bacteria attach apically to the surface in a reversible fashion, and then progress into irreversible lateral interaction with the surface. Tightly attached cells proliferate to form clonal microcolonies and to produce the biofilm matrix, composed by exopolysaccharide (EPS), proteins and extracellular DNA. By proliferation and flagella dependent migration processes, microcolonies evolve to a mature biofilm, formed by macrocolonies separated by channels that allow the transport of nutrients, water and waste (Monds and O'Toole 2009).
Initial attachment of P. putida cells has been proposed to involve flagella (Yang, Menge and Cooksey 1994; Turnbull et al. 2001; Yousef-Coronado, Travieso and Espinosa-Urgel 2008), while the high molecular weight adhesin protein LapA is required for irreversible attachment to both biotic and abiotic materials (Espinosa-Urgel, Salido and Ramos 2000). Some of the matrix-forming exopolysaccharides and the extracellular DNA, although are not essential to biofilm development, have been suggested to have a role in P. putida biofilm stability (Steinberger and Holden 2005; Nilsson et al. 2011). In addition, the adhesin LapF has been described to take part in the build-up of structured biofilms (Martínez-Gil, Yousef-Coronado and Espinosa-Urgel 2010).
In P. putida, nutrient availability controls the presence of LapA in the outer membrane. Activity of the LapA-specific periplasmic protease LapG is inhibited by the transmembrane protein LapD in the presence of high cyclic diguanylate (c-di-GMP) levels. Nutrient starvation induces c-di-GMP hydrolysis, resulting in the release of LapG inhibition by LapD, and the subsequent proteolytic cleavage of LapA causes biofilm dispersal (O'Toole, Kaplan and Kolter 2000; Gjermansen et al. 2005; Klausen et al. 2006). The c-di-GMP phosphodiesterase BifA was recently described as one of the proteins involved in c-di-GMP turnover that controls biofilm dispersal in response to nutrient availability (Jiménez-Fernández et al. 2015). The GacS/GacA two components system and the flagellar master regulator FleQ have been proposed to be involved in biofilm formation by transcriptional control of the genes coding for the adhesins LapA and LapF, but the mechanistic details are largely unknown (Martínez-Gil, Ramos-González and Espinosa-Urgel 2014).
The main objective of this work is to contribute to the expansion of the available knowledge in biofilm formation by the isolation and characterization of biofilm formation-deficient mutants in P. putida.
MATERIALS AND METHODS
Media and growth conditions
Planktonic cultures of Escherichia coli DH5α [φ80dlacZΔM15 Δ(lacZYA-argF)U169 recA1 endA1 hsdR17 (rk− mk+) supE44 thi-1 gyrA relA1] (Hanahan 1983) and P. putida strains were routinely grown in Luria-Bertani (LB) broth (Sambrook, Fristch and Maniatis 1989) at 37°C and 30°C respectively, with 180 rpm shaking. For solid media, Bacto-Agar (Difco) was added to a final concentration of 18 g l−1. Antibiotics and other additions were used, when required, at the following concentrations: kanamycin (25 mg l−1) and rifampicin (10 mg l−1). All reagents were purchased from Sigma-Aldrich.
Mutagenesis and screening
The pUTminiTn5-Km plasmid bearing the miniTn5-Km transposon was transferred to P. putida KT2442 [mt-2 hsdR1 (r− m+) Rif r] (Franklin et al. 1981) by triparental mating using pRK2013 plasmid (Figurski and Helinski 1979) as helper, as previously described in de Lorenzo et al. (1990). Kanamycin-resistant colonies bearing transposon insertions were toothpick-inoculated into the wells of microtiter plates and grown overnight at 25°C with moderate shaking (150 rpm). A total of 50-fold diluted overnight cultures were incubated at the same conditions for 6 h, and planktonic and biofilm growth were quantified essentially as described in O'Toole et al. (1999) (Fig. 1A). To sequence the flanking regions of the 40 independent miniTn5-km insertions, a variation of the arbitrarily primed PCR method (O'Toole et al. 1999; López-Sánchez et al. 2013) was used.
Screening strategy for the isolation of biofilm formation deficient mutants. (A) Steps of the screening procedure. (B) Photograph of a representative dish from an actual screening showing several biofilm deficient mutant candidates. (C) Chart of biofilm versus planktonic growth, showing the values obtained for the mutants in a representative microtiter plate from an actual screening.
Screening strategy for the isolation of biofilm formation deficient mutants. (A) Steps of the screening procedure. (B) Photograph of a representative dish from an actual screening showing several biofilm deficient mutant candidates. (C) Chart of biofilm versus planktonic growth, showing the values obtained for the mutants in a representative microtiter plate from an actual screening.
Serial dilution-based growth curves
Serial dilution-based growth curves were performed as described in López-Sánchez et al. (2013). For each experiment, at least three biological replicates were assayed in octuplicate.
Swimming assays
Swimming assays were adapted from Parkinson (1976). Tryptone motility plates containing 0.3% Bacto-agar (Difco) were toothpick-inoculated with fresh colonies and incubated for 12 h at 30°C. Digital photographs were taken, and the swimming zone diameter was measured and normalized to that of the wild-type. At least three biological replicates were assayed for each strain.
Phase-contrast microscopy
Phase-contrast microscopy of surface adhesion was performed on a Leica DMI4000B inverted microscope using 20× objective and 1.6× ocular magnification. Cells from mid-exponential (A600 = 0.2–0.3) LB cultures of the selected P. putida strains were serially diluted (102–104) in LB and samples were transferred to wells of Costar 96 microtiter polystyrene plates (Corning). Attachment was allowed to proceed for 30 min, planktonic cells were removed by washing twice with 150 μl LB, and then 50 μl LB was added to the wells. For short-term assessment of surface attachment, plates were incubated at room temperature for 3 h, and then cells deposited on the plane of the well surface were video recorded for 1 min.
Congo Red binding assay
To assess cellulose-like polysaccharide production, the Congo Red (CR) binding assay was adapted from An, Wu and Zhang 2010. Bacterial cells from 2 ml overnight cultures in LB were collected by centrifugation and resuspended in 1 ml 1% bacto-tryptone (Difco) broth containing 40 μg ml−1 CR. The mixtures were incubated at 30°C with shaking at 180 rpm for 180 min. The bacterial cells and bound CR were removed by centrifugation (1 min, 13 000 rpm), and the amount of CR remaining in the supernatants was determined by measuring the absorbance of the supernatant at 490 nm, using the same medium supplemented with 40 μg ml−1 CR as a control.
c-di-GMP levels measurement
c-di-GMP levels were estimated by using a fluorescence-based reporter consisting in a gfp transcriptional fusion to the c-di-GMP responsive-promoter PcdrA, pCdrA::gfpC (Rybtke et al. 2012). This plasmid was transferred to the receipt strains by electroporation as described by Choi, Kumar and Schweizer (2006). Fluorescence and planktonic growth were measured on a Tecan Spark 10 M fluorimeter. Cells from early-exponential (A600 = 0.1) LB cultures of the selected P. putida strains were 100-fold diluted and samples were incubated in Costar 96 microtiter polystyrene plates (Corning) at 30°C for 23 h. Measurements of three biological replicates were performed at 15 min intervals.
RESULTS AND DISCUSSION
Isolation of biofilm formation-defective mutants
We have designed a screening method to isolate mutants defective in biofilm formation based on the fact that P. putida is able to grow and form microcolonies on microtiter dish wells during the exponential growth phase (Gjermansen et al. 2005; López-Sánchez et al. 2013). Biofilm formation-defective mutants are thus distinguished as those showing low levels of crystal violet-stainable biofilm biomass in conditions that support high levels of biofilm growth of the wild-type strain. In order to isolate biofilm formation-defective mutants, P. putida KT2442 was mutagenized with miniTn5-Km and candidates were assessed for planktonic and biofilm growth (see Materials and Methods for details) (Fig. 1A). Most of the ∼8000 mutants analyzed showed biofilm and planktonic growth levels similar to those in the wild-type. Phenotypes of biofilm formation-defective mutants were evident by visual inspection (Fig. 1B) and by plotting biofilm biomass against planktonic growth (Fig. 1C). Thus, 40 candidates showing decreased biofilm growth and planktonic growth comparable to that of the wild-type were selected for further analysis. Phenotypes of all 40 mutants were confirmed by an end-point biofilm formation assay, in which biofilm and planktonic biomass were assessed from each mutant grown in octuplicate in the same conditions used for the screening (data not shown).
The transposon insertion in each of the mutants was mapped by sequencing the flanking regions and comparing with the published sequence of the P. putida KT2440 genome. The locations of the 40 independent miniTn5-Km insertions are listed in Table 1. More than half (29) of the selected biofilm formation mutants were found to bear insertions in genes involved in synthesis, export and maintenance of the outer membrane protein LapA. Not surprisingly, 24 of them carried independent insertions in lapA, the largest gene in the P. putida KT2440 genome (26 kbp). Three additional insertions mapped in the lapBC operon, encoding two subunits of a LapA-specific secretion system, and two more in lapD, encoding a negative regulator of LapA proteolysis. As LapA was previously characterized as a critical determinant for biofilm formation (Espinosa-Urgel, Salido and Ramos 2000), the isolation of these mutants corroborates our approach as a valid method for the isolation of biofilm formation deficient mutants. No insertions were identified to map in the lapF gene, the second largest gene in the P. putida genome encoding LapF, the second outer membrane adhesin protein involved in biofilm development. The absence of mutations in lapF could be explained by the fact that biofilm growth on microtiter plate wells does not progress through the biofilm maturation stage, in which the role of LapF is relevant (Martínez-Gil, Yousef-Coronado and Espinosa-Urgel 2010).
Location of the miniTn5-Km insertions in the mutants isolated in this work.
| ORF ID | Gene | Number of insertions |
|---|---|---|
| PP0165 | lapD | 2 |
| PP0166 | lapC | 1 |
| PP0167 | lapB | 2 |
| PP0168 | lapA | 24 |
| PP1222 | tolB | 1 |
| PP1650 | gacS | 1 |
| PP3222 | ? | 1 |
| PP4343 | flhF | 1 |
| PP4355 | fliP | 1 |
| PP4373 | fleQ | 5 |
| PP4385 | flgG | 1 |
| PP4519 | tolC | 1 |
| ORF ID | Gene | Number of insertions |
|---|---|---|
| PP0165 | lapD | 2 |
| PP0166 | lapC | 1 |
| PP0167 | lapB | 2 |
| PP0168 | lapA | 24 |
| PP1222 | tolB | 1 |
| PP1650 | gacS | 1 |
| PP3222 | ? | 1 |
| PP4343 | flhF | 1 |
| PP4355 | fliP | 1 |
| PP4373 | fleQ | 5 |
| PP4385 | flgG | 1 |
| PP4519 | tolC | 1 |
Location of the miniTn5-Km insertions in the mutants isolated in this work.
| ORF ID | Gene | Number of insertions |
|---|---|---|
| PP0165 | lapD | 2 |
| PP0166 | lapC | 1 |
| PP0167 | lapB | 2 |
| PP0168 | lapA | 24 |
| PP1222 | tolB | 1 |
| PP1650 | gacS | 1 |
| PP3222 | ? | 1 |
| PP4343 | flhF | 1 |
| PP4355 | fliP | 1 |
| PP4373 | fleQ | 5 |
| PP4385 | flgG | 1 |
| PP4519 | tolC | 1 |
| ORF ID | Gene | Number of insertions |
|---|---|---|
| PP0165 | lapD | 2 |
| PP0166 | lapC | 1 |
| PP0167 | lapB | 2 |
| PP0168 | lapA | 24 |
| PP1222 | tolB | 1 |
| PP1650 | gacS | 1 |
| PP3222 | ? | 1 |
| PP4343 | flhF | 1 |
| PP4355 | fliP | 1 |
| PP4373 | fleQ | 5 |
| PP4385 | flgG | 1 |
| PP4519 | tolC | 1 |
A total of eight additional mutants were found to contain insertions in genes related to flagellar biogenesis and function. Five of these mapped in fleQ, encoding the top regulator of the flagellar cascade in P. aeruginosa (Dasgupta et al. 2003) and P. putida (Jiménez-Fernández, unpublished data). Single insertions were shown to be located in fliP, whose product is a component of the flagellar export apparatus (Minamino and Macnab 1999), flgG, encoding the major flagellar component of the distal part of the basal-body rod (Homma, DeRosier and Macnab 1990), and flhF, encoding a determinant of flagellar location (Kazmierczak and Hendrixson 2013). According to the Database of PrOkaryotic OpeRons (DOOR) operon prediction algorithm (Mao et al. 2009), fliP and flgG are predicted to be part of the fliK-pp4360-fliLMNOPQR-flhB and flgFGHIJKL operons, respectively, and therefore insertions in these genes are expected to be polar on multiple genes encoding diverse flagellar components. Although Pandza et al. (2000) has proposed that flhF is a part of a large operon and the DOOR algorithm predicts that flhF forms an operon with the upstream flhA and downstream fleN and fliA genes, RT-PCR analysis showed that two distinct transcriptional units (flhAF and fleNfliA) occur in this region in P. putida (Rodríguez-Herva et al. 2010) (also see below).
An insertion was identified as being located in tolB (PP1222), encoding the periplasmic component of the Tol-OprL system, involved in outer membrane maintenance. According to DOOR, tolB is part of the ybgC-tolQRAtolB-oprL-ybgF operon. Experimental results confirmed this transcriptional organization, although an internal promoter is present between tolB and oprL (Llamas, Ramos and Rodríguez-Herva 2003). We also identified transposons insertion in gacS (PP1650), a part of the gacS-ldhA-pp1648-1647 operon also encoding a lactate dehydrogenase and two hypothetical proteins, whose product is the sensor kinase of the GacS/GacA two-component regulatory system, and in PP3222, encoding the permease element of an ABC-type transporter, encoded by the pp3226-3225-3224-3223-3222-3221-3220 operon. Transposon insertions in PP3222 are predicted to have polar effect on PP3221 and PP3220 transcription, encoding other subunits of the ABC-type transporter.
LapA function is essential for biofilm formation
In order to further characterize the biofilm formation defect showed by the mutants isolated in this work, we performed serial dilution-based growth curves in LB medium (López-Sánchez et al. 2013). In this method, a dilution series is used to recapitulate the time-course of planktonic and biofilm growth on microtiter plates. The behavior displayed by the wild-type strain in all the assays was equivalent to that previously described: biofilm developed to reach a maximum level coincident with the end of the exponential phase and dispersal was observed during stationary phase. As shown in Fig. 2A–D, representative mutants harboring miniTn5 insertions in lapA, lapB, lapC and lapD genes were unable to produce significant amounts of biofilm in this assay. However, planktonic growth was similar to that of the wild-type strain in all cases. Defects in flagellar function of these mutants were assessed by a soft agar swim plate assay. None of these insertions provoked a detectable effect on swimming motility, as the diameter of the swimming halo was similar to that of the wild-type strain in all cases (Fig. 3). LapA belongs to a group of large surface proteins containing repetitive tandem sequences that share functional and structural features (Lasa and Penades 2006). The presence of LapA on the cell surface has been previously described to be essential for biofilm formation and depends on an ABC transporter system formed by LapB, LapC and LapE (Hinsa et al. 2003) and the control of the LapG protease activity by the c-diGMP sensor LapD (Gjermansen et al. 2005, 2010; Hinsa and O'Toole 2006; Newell et al. 2011). The biofilm formation defect showed by the lapD mutant is consistent with the persistent biofilm behavior as previously described for lapG and lapD mutants (Gjermansen et al. 2005). Taken together, these results support previous data on the relevance of the LapA/LapBCE/LapGD system in biofilm formation in P. putida (Gjermansen et al. 2010).
Dilution-based growth curves of the selected mutants. 20 h serial dilution-based growth curves for lapA− (A), lapB− (B), lapC− (C), lapD− (D), fleQ− (E), flgG− (F), flhF− (G), fliP− (H), gacS− (I), tolB− (J) and PP3222− (K) mutants in LB. Planktonic growth (open symbols) and biofilm growth (filled symbols) are plotted against A600 at the time of inoculation. The wild-type strain KT2442 was assayed in every experiment for comparison purposes (circles). Average from eight experimental replicates and standard deviation are shown and each plot represents an experiment out of three biological replicates.
Dilution-based growth curves of the selected mutants. 20 h serial dilution-based growth curves for lapA− (A), lapB− (B), lapC− (C), lapD− (D), fleQ− (E), flgG− (F), flhF− (G), fliP− (H), gacS− (I), tolB− (J) and PP3222− (K) mutants in LB. Planktonic growth (open symbols) and biofilm growth (filled symbols) are plotted against A600 at the time of inoculation. The wild-type strain KT2442 was assayed in every experiment for comparison purposes (circles). Average from eight experimental replicates and standard deviation are shown and each plot represents an experiment out of three biological replicates.
Swimming assays of the biofilm formation mutants. Swimming assays on 0,3%. Triptone-Agar incubated for 12 h at 30°C. Each picture is a representative swim plate out of at least three biological replicates with the wild-type (top) and mutants strain (bottom).
Swimming assays of the biofilm formation mutants. Swimming assays on 0,3%. Triptone-Agar incubated for 12 h at 30°C. Each picture is a representative swim plate out of at least three biological replicates with the wild-type (top) and mutants strain (bottom).
Mutants in flagellar genes are defective in biofilm formation
Dilution series-based growth curves in LB were also used to better characterize the biofilm formation phenotypes of the flagella-related mutants. The general planktonic growth behavior was clearly slower compared to the wild-type strain, but two different biofilm development defects were observed: the fliP and flgG mutants showed delayed biofilm formation and a modest decrease in maximum biofilm biomass, whereas fleQ and flhF mutants displayed a stronger defect in biofilm formation (Fig. 2E–H). fliP and flgG encode a component of the flagellar export apparatus (Minamino and Macnab 1999) and the major flagellar component of the distal part of basal-body rod (Homma, DeRosier and Macnab 1990), respectively. The biofilm phenotype showed by the fliP and flgG mutants has been previously observed for mutants affected in the structural elements of the flagellum FliF, FliG and FliN (López-Sánchez, unpublished data). These results support previous data on the role of flagella in attachment to surfaces in different Pseudomonas species (DeFlaun et al. 1990; Yang, Menge and Cooksey 1994; O'Toole and Kolter 1998; Turnbull et al. 2001; Toutain et al. 2007; Yousef-Coronado, Travieso and Espinosa-Urgel 2008; Duque et al. 2013), but indicate that flagellar integrity or function is not essential to develop a biofilm. This is consistent with the fact that a non-flagellated mutant is able to form mature biofilms in a flow-chamber system (Gjermansen et al. 2005).
Unlike the fliP and flgG mutants, mutants affected in fleQ or flhF, encoding regulatory elements of flagellar biogenesis were unable to form a biofilm (Fig. 2E and F). This defect was confirmed by serial dilution-based growth curves performed with an extended incubation time (30 h) (data not shown). As the loss of flagellar function is not sufficient to explain the phenotypes displayed by fleQ and flhF mutants, these data suggest a more general role of these regulators in biofilm development. FleQ has been described as the master regulator of flagella gene expression in P. aeruginosa (Arora et al. 1997; Dasgupta et al. 2003) and has been shown to regulate transcription of genes involved in biofilm formation in response to levels of the secondary metabolite c-di-GMP, including the pel and pls operons, involved in the synthesis of the major EPS in P. aeruginosa (Dasgupta et al. 2003) and the cdrA gene, that codes for a key biofilm formation adhesin in the same organism (Borlee et al. 2010). In a recent work, FleQ has been proposed to stimulate the production of LapA in response to increased concentrations of c-diGMP in P. putida KT2442 (Martínez-Gil, Ramos-González and Espinosa-Urgel 2014).
On the other hand, FlhF has been described a signal recognition particle-like protein required for polar flagellar placement in P. putida, P. aeruginosa and Vibrio alginolyticus (Pandza et al. 2000; Kusumoto et al. 2006; Murray and Kazmierczak 2006). No other regulatory functions has been described for FlhF, but flhF is located downstream of flhA and upstream of fleN and fliA. As the results published on the transcriptional organization of this region are conflicting (Pandza et al. 2000; Rodríguez-Herva et al. 2010), it remains possible that the cause of the phenotype displayed by de flhF::miniTn5-Km mutant could be the altered expression of the downstream gene fleN. FleN, which is an ATPase, has been described to work together with FleQ, being necessary for full expression of FleQ-dependent flagellar genes, pel, psl and cdr operons in P. aeruginosa (Hickman and Harwood 2008). We hypothesize that analogous elements regulated by FleQ in P. putida, like the adhesin LapA and Class II flagellar genes, could be similarly regulated by FleN.
As expected, none of these mutants showed a motility halo when tested for swimming by a soft agar swim plate assay (Fig. 3), thus confirming that inactivation of fleQ, flgG, flhF or fliP results in loss of flagellar function. Taken together, these data suggest that the strong defect in biofilm formation showed by the fleQ and flhF mutants is due not only to the loss of flagella, but to the direct or indirect effect over the expression of essential elements for biofilm formation.
Other elements involved in biofilm formation
GacS, the sensor component of the GacS/GacA two components system, is one of the elements that has been identified by our formation-defective mutants screening to be related to biofilm formation in P. putida. Serial dilution-based growth curves showed that planktonic growth of the gacS mutant is equivalent to the wild-type and although it is able to form biofilm, the mutant fails in producing the same biofilm amount than the wild-type strain (Fig. 2I). In contrast with the hypermotile phenotype described in P. fluorescens by Martínez-Granero et al. (2012), but in agreement with the wild-type phenotype showed in P. syringae by Kinscherf and Willis (1999), the gacS::miniTn5-Km mutant isolated in this work did not show a defect in swimming motility (Fig. 3).
The GacS-GacA two-component system controls the production of exoproducts and virulence factors in several Pseudomonas species (Heeb and Haas 2001; Lapouge et al. 2008). Upon activation, GacA switches on the transcription of sRNA genes that bind RsmA/CsrA translational repressor proteins. Thus, by producing the sRNAs, the GacS/GacA signal transduction pathway upregulates the expression of genes repressed by RsmA/CsrA. The GacS/GacA regulon varies among different species, but includes biofilm-related genes like the pgaABCD operon in E. coli, responsible for the synthesis of a polysaccharide adhesin (Wang et al. 2005) or genes for alginate production in Azotobacter vinelandii and P. syringae (Castañeda et al. 2001; Willis, Holmstadt and Kinscherf 2001). Sequence inspection revealed that, in addition to the GacS/GacA system, the sRNAs rsmX, rsmY and rsmZ and their target CsrA are conserved in P. putida KT2442 and GacS was recently proposed to influence the biofilm formation by controlling the expression of the two high molecular weight biofilm adhesins LapA and LapF (Martínez-Gil, Ramos-González and Espinosa-Urgel 2014). These published results suggest that, although the transposon insertion in gacS is predicted to be polar on ldhA, PP1648 and PP1647, the biofilm formation defect observed is likely due to the lack of GacS.
Dilution-based growth and biofilm curves showed that the tolB mutant grows slower than the wild-type in LB, but forms and disperses the biofilm with a similar timing (Fig. 2J). However, the total amount of biofilm observed in the tolB mutant is lower throughout the whole growth curve, suggesting a role of the protein TolB in the stability of the biofilm. The tolB mutant was also strongly deficient in swimming motility (Fig. 3). TolB encodes a periplasmic protein of the Tol-OprL system and a similar non-motile phenotype was previously showed for mutants in each component of this system in P. putida. These mutants were described having a total deficiency in swarming motility assays and, in addition, being shorter than the parental ones and having altered permeability barrier functions of the outer membrane (Llamas, Ramos and Rodríguez-Herva 2000). This proposed role in the maintenance of the structure of cellular envelope could be responsible for the biofilm formation defect displayed by the tolB mutant (Fig. 2J). The tolB gene is transcribed as part of mRNA species that includes the two downstream genes oprL and orf2, which also encode proteins of the Tol-OprL system (Llamas, Ramos and Rodríguez-Herva 2000). This transcript contributes about only 10%–15% of the total OprL protein (Llamas, Ramos and Rodríguez-Herva 2003). Thus, a possible polar effect on the expression of downstream genes oprL and orf2 should be irrelevant.
The last element that has been identified by our screening for biofilm formation deficient mutants was PP3222, encoding a putative permease of an ABC transporter. As shown in Fig. 2K, the mini-transposon insertion in PP3222 causes a null biofilm phenotype, nevertheless planktonic growth was almost similar to the wild-type. No defect in swimming motility was observed for the PP3222 mutant (Fig. 3). PP3222 seems to be located into an operon with PP3220, PP3221 and PP3223, encoding for an ATP-binding protein, a second permease and a substrate binding protein, suggesting that this ATP-binding cassette transporter may act as a solute uptake system. These transporters take up a wide variety of substrates, including from small sugars, amino acids and small peptides to metals, anions and siderophores (Davidson and Chen 2004).
Considering that the most novel discovery in the screening is the involvement of the permease PP3222 in biofilm formation and that the role of Tol-OprL in biofilm development has not been extensively studied, we decided to further characterize the phenotype of these two mutants.
Further characterization of the tolB− and PP3222− mutants
To determine whether the transposon insertions in tolB or PP222 have an impact on adhesion to the surface, we used phase-contrast microscopy to analyze the adhesion phase of biofilm formation. To this end, cells of the wild-type, tolB− and PP3222− strains were allowed to attach to the bottom of polystyrene microtiter plate wells, and were then recorded in 1-min videos (Fig. 4). Observation of the images obtained revealed that the majority of the wild-type and tolB− mutant cells were immobilized on the polystyrene surface, strongly suggesting that both strains are capable of irreversible attachment. In contrast, the PP3222− mutant cells were not immobilized in these conditions. This result indicates that PP3222 is required for irreversible interaction with the surface and in agreement, this mutant in unable to form biofilm (Fig. 2K). In contrast, TolB is not related with the initial surface attachment as supported by the ability of the tolB− mutant to develop a biofilm.
Adhesion assays of PP3222− and tolB− mutants. Phase-contrast micrographs of the wild-type strain KT2442 (top) and its tolB (center) and PP3222 (bottom) mutant derivatives. Two frames of the same field were taken in a 1-min interval (left and center). The right panel shows an overlay of the two images of each strain digitally colored red (t = 0) or green (t = 1 min)
Adhesion assays of PP3222− and tolB− mutants. Phase-contrast micrographs of the wild-type strain KT2442 (top) and its tolB (center) and PP3222 (bottom) mutant derivatives. Two frames of the same field were taken in a 1-min interval (left and center). The right panel shows an overlay of the two images of each strain digitally colored red (t = 0) or green (t = 1 min)
The lower amount of biofilms produced by the tolB mutant suggest a defect in biofilm stability and some of the exopolysaccharides produce by P. putida has been related with the maintenance of the biofilm structure (Nilsson et al. 2011). We performed CR binding assays to measure the quantity of EPS produced by the tolB− and PP3222− mutants. Levels of CR retained by the wild-type and the tolB− mutant were equivalent (Fig. 5), ruling out a role of the Tol-OprL system in EPS production. Interestingly, CR retention by the PP3222 mutant was 2-fold reduced relative to the wild-type (Fig. 5).
CR adsorption phenotypes of PP3222− and tolB− mutants. CR adsorption of the wild-type (KT2442), PP3222− and tolB− strains. Data are normalized to the wild-type, set to 100%. Bars represent the averages and standard deviation of at least three biological replicates.
CR adsorption phenotypes of PP3222− and tolB− mutants. CR adsorption of the wild-type (KT2442), PP3222− and tolB− strains. Data are normalized to the wild-type, set to 100%. Bars represent the averages and standard deviation of at least three biological replicates.
The signal molecule c-di-GMP has been shown to stimulate surface adhesion, EPS production and biofilm formation (Gjermansen, Ragas and Tolker-Nielsen 2006; Matilla et al. 2007, 2011). The fact that PP3222 is required for a phenotype set associated with biofilm formation induced by c-di-GMP, suggests the involvement of PP3222 in c-di-GMP regulation. To test this hypothesis, we measured the c-di-GMP levels during the growth curve using a fluorescence-based reporter consisting in a gfp transcriptional fusion to the c-di-GMP responsive-promoter PcdrA from P. aeruginosa. The expression of cdrA promoter is induced by the presence of c-diGMP and subjected to negative control by FleQ (Rybtke et al. 2012). As an experimental control, fluorescence from PcdrA-gfpc in a fleQ deletion mutant (unpublished) was simultaneously monitored in the wild-type and PP3222− mutant. As expected, florescence levels displayed by the ΔfleQ mutant were higher throughout the complete growth curve. On the contrary, no significant differences were observed in the PP3222− mutant relative to the wild-type (Fig. 6). This result indicates that the absence of biofilm formation (Fig. 2K), the inability to attach to surfaces (Fig. 4) and the reduced CR retention of the PP3222− mutant (Fig. 6) are not a consequence of lower intracellular c-diGMP levels.
C-di-GMP levels of PP3222− mutant. PcdrA-gfp fluorescence (open circles) and A600nm (filled circles) of KT2442 (black), PP3222− (gray) and ΔfleQ (light-gray) measured each 15 min for 23 h. Curves represent the average and standard deviation of three biological replicates.
C-di-GMP levels of PP3222− mutant. PcdrA-gfp fluorescence (open circles) and A600nm (filled circles) of KT2442 (black), PP3222− (gray) and ΔfleQ (light-gray) measured each 15 min for 23 h. Curves represent the average and standard deviation of three biological replicates.
Since the only biofilm-related phenotype shown by the tolB mutant is a slight decrease in the amount of biofilm produced, we propose that TolB has a minor role in biofilm formation. The partial loss of biofilm stability could be associated to the altered permeability barrier functions of the outer membrane. On the contrary, although further studies are necessary to clarify the nature of the substrate of the ABC transport system formed by PP3220, PP3221, PP3222 and PP3223, and the relationship between uptake and biofilm development in P. putida, phenotypes displayed by the PP3222 mutant clearly indicate a role of this ABC transporter in biofilm formation.
Phenotype characterization of the newly isolated biofilm formation mutants described in this work provides new insights on the early biofilm development in P. putida. Our results underscore the relevance of LapA, suggest cross-regulation between flagellar biogenesis process and biofilm formation involving FleQ and auxiliary regulatory proteins, provide evidence of the GacA/GacS system involvement in the regulation of biofilm formation, indicates a role of outer membrane stability and point out solute transport as an essential process for biofilm development in P. putida.
We wish to thank all members of the Govantes and Santero laboratories at Centro Andaluz de Biología del Desarrollo for critical discussion.
FUNDING
This work was funded by Grant and of the Spanish Ministerio de Ciencia e Innovación and Spanish Ministerio de Economía y Competitividad, respectively.
Conflict of interest. None declared.
