Genome-Wide Analysis Reveals a RhlA-Dependent Modulation of Flagellar Genes in Pseudomonas Aeruginosa PAO1

Background : Pseudomonas aeruginosa is an opportunistic pathogen and an important model organism for the study of bacterial group behaviors, including cell motility and biofilm formation. Rhamnolipids play a pivotal role on biofilm formation and motility phenotypes in P. aeruginosa , possibly acting as wetting agents and mediating chemotactic stimuli. However, no biochemical mechanism or gene regulatory network has been investigated in regard to rhamnolipids’ modulation of those group behaviors. Results : Using DNA microarrays, we investigated the transcriptomic profiles in the stationary phase of growth of wild-type P. aeruginosa PAO1 and a rhlA -mutant strain, unable to produce rhamnolipids. A total of 134 genes were differentially expressed, comprising different functional categories, indicating a significant physiological difference between the rhamnolipid-producing and non-producing strains. Interestingly, several flagellar genes are repressed in the mutant strain, which directly relates to the non-motile phenotype of the rhlA -minus strain. Swarming motility was restored with the addition of exogenous rhamnolipids obtained from the wild-type strain. Conclusions : Our results show significant evidence that rhamnolipids and/or their precursors, 3-(3-hydroxyalkanoyloxy) alkanoic acids, the major biosynthetic products of rhlABC pathway, seem to modulate gene expression in P. aeruginosa . Swarming motility assays support this hypothesis, since the non-motile rhlA -mutant strain had its swarming ability restored by the addition of exogenous rhamnolipids.


Background
Pseudomonas aeruginosa is a ubiquitous bacterium capable of causing opportunistic infection in immunocompromised individuals and is the major pathogen in cystic fibrosis [1]. P. aeruginosa is a model organism for the study of bacterial social behaviour, or 'sociomicrobiology' [2], biofilm development [3] and bacterial motility patterns [4][5][6]. Multiple factors are involved in bacterial motility and biofilm formation during surface interactions, in which rhamnolipids seem to play a central role [7].
Rhamnolipids are typically produced during the stationary phase of growth, as the rhlAB operon is tightly regulated by QS systems [11,12], and post-transcriptionally regulated by small non-coding RNAs, NrsZ [13] and PhrD [14].
There is evidence that rhamnolipids are necessary for the three-dimensional architecture of biofilm structures and the initial formation of P. aeruginosa PAO1 microcolonies during growth in controlled flow systems [3,7]. They are usually associated to the maintenance of channels between the typical mushroom-like structures and the biofilm dispersion [3,15]. Zheng et al., (2017) reported that synthetic analogs of rhamnolipids restored structured biofilm formation by a rhlA-mutant strain, in a concentration-dependent manner. In a similar fashion, one of those compounds promoted swarming motility in the rhamnolipid-deficient strain [16].
Surface-associated bacterial motility is also critical for biofilm formation and spatial arrangement, including well-characterized pores and channels [17]. P. aeruginosa has the ability to move through complex motility patterns such as twitching, swimming and swarming [4][5][6]. Evidence in the literature suggests that rhamnolipids, flagella and type IV pilus are fundamental for motility in P. aeruginosa, particularly for the swarming motility [18][19][20][21]. However, the specific molecular mechanisms involved in rhamnolipid modulation of swarming motility and formation of structured biofilms are poorly understood.
In this study, we describe a transcriptome analysis to compare the gene expression profile of P. aeruginosa PAO1 and its rhlA-knockout derivative, using Affymetrix GeneChip TM P. aeruginosa genome arrays and real-time qPCR analysis. We tested the hypothesis that rhlA-derived rhamnolipids affect gene expression, under dynamic growth conditions, at the stationary phase of growth. We show that 134 genes were differentially expressed in the rhlA-mutant strain, 91 positively and 43 negatively regulated.
Interestingly, the repression of flagellar genes, ultimately involved with swarming motility, was observed. Supporting these findings, the non-motile rhlA-deficient strain had the swarming motility restored when supplemented with rhamnolipids extracted from the wild-type strain. Therefore, we gain mechanistic insight into the role of rhamnolipids in P. aeruginosa motility.

Gene expression profile of a rhlA-deficient strain of P. aeruginosa in stationary phase of growth
To identify genes possibly regulated in response to secreted biosurfactants in P. aeruginosa, we investigated global gene expression in the stationary phase of growth, using a P. aeruginosa genome microarray to compare the transcript profiles of a rhlA mutant with its parental strain, PAO1. We identified 134 differentially expressed genes, 91 genes were upregulated and 43 were downregulated ( Figure S1; Table S1).
Hierarchical clustered heatmaps showed that differentially expressed genes could distinguish between the rhlA-mutant and the control, wild-type strain PAO1 (Table 1; Figure 1). The majority of differentially expressed genes are upregulated in the strain lacking the enzyme RhlA, responsible for production of HAAs and rhamnolipids.
We investigated functional categories of differentially expressed loci that correspond to 134 genes in the previously published annotation of the P. aeruginosa genome [1]. As depicted in Figure 2, gene functional classes were obtained through COG database [22], showing an expressive number of genes that encode proteins involved with signaling and basic metabolism mechanisms being positively regulated. Interestingly, we can see that a relevant portion of genes encoding proteins involved in cellular motility are being repressed. On the other hand, genes that encode proteins related to defense mechanisms, which include a type VI secretion system (T6SS) and some virulence factors are being positively regulated (Table 1). Several genes encoding hypothetical proteins with unknown functions have also had their gene expression induced. Table S1 summarizes the expression data from genome microarray analyses of rhlA-mutant P. aeruginosa versus the wild-type strain, PAO1. Among those 134 genes that showed a response to the lack of RhlA of at least 1.5-fold, several were identified as motility-related genes. Interestingly, down-regulation of key flagellar genes was particularly significant, and included class II, III and IV flagellar genes [23,24]: fleN, fliD, fliT, fliS (class II), flgL (class III), and fliC (class IV). These findings characterize a pattern of repression of flagellar genes, which can be directly related to the non-motile phenotype of rhlA-deficient strains, that has been previously described [18][19][20][21]. In a similar manner, protein PA3731, annotated as a Phage Shock Protein, for having a predicted secondary structure similar to that of PspA from Escherichia coli, was repressed in the rhlA-mutant strain (-1.57). The PA3731 gene was found to have a positive effect on biofilm formation and swarming motility in P. aeruginosa [25].
On the other hand, upregulation of pilO was observed, a gene that encodes a protein involved in type IV pilus function (T4P) [26], essential to surface adhesion and twitching motility. Other functional gene categories have responded to the absence of RhlA and its biosynthetic products in a similar fashion. Remarkably, hcp1, mapping within an antiprokaryotic type VI secretion system (H1-T6SS) [27,28] of P. aeruginosa, showed a positive regulation (Table 1) genes is repressed [29]. However, a subset of prophage genes can be expressed as an evolutionary adaptation, conferring a physical conditioning advantage to the bacterial host [30].
We have confirmed this approach by measuring gene expression of some selected genes with RT-qPCR, in the wild-type strain and its rhlA-minus derivative strain (see below). Class III (flgL), and Class IV (fliC) [24,31]. In addition, flagellar expression and assembly play a pivotal role on phenotypical traits that also depend on production of HAA and rhamnolipids, especially swarming motility [18][19][20][21].

Genes repressed in the
The present assays show that gene repression of the fliC (PA1092), flgL (PA1087), fliD (PA1094) and fliT (PA1096) in the rhlA-deficient strain compared to PAO1 was statistically significant (Figure 3a). The fliC gene showed 91% of inhibition (Table S3), while genes flgL, fliD, and fliT showed downregulation of 66, 60 and 86%, respectively (Table S3). The FlgL, FliD and FliC proteins make up the structure of the flagellum and are present in the extracellular medium [32][33][34]. FliT is a chaperone that plays an important role in the proper exportation and assembly of proteins that compose the flagellar machinery, including FliD, which is a filament-capping protein [35,36]. The single polar flagellum of P. aeruginosa is an important virulence and colonization factor of this opportunistic pathogen, having been attributed fundamental roles in the formation of biofilm and bacterial motility [37,38].
Expression of rhlA and rhlB was also assessed by RT-qPCR, since rhlA has been mutated (ΔrhlA::Gm R ) [39] and rhlB maps within the same operon. In fact, we found that the rhlA-mutation had a polar effect that is shown both in the microarrays (Table 1) and RT-qPCR (Figure 4), as rhlB is substantially repressed.

Swarming motility of a rhlA-deficient P. aeruginosa is restored by addition of exogenous Rhamnolipids
Many studies have demonstrated that P. aeruginosa swarming is a complex adaptation process to solid surfaces, in a viscous milieu, that is controlled by a substantial number of cooperating genes, including rhlA [40,41]. To further understand the role of RhlA biosynthetic products on swarming motility, a plate assay was performed with the wild-type P. aeruginosa PAO1 and its isogenic rhlA-negative mutant. To determine if the non-motile rhlA-mutant strain respond to exogenous rhamnolipids and/or HAAs, the cells were preconditioned with a rhamnolipid extract obtained from the wild-type strain, prior to inoculation onto the swarm plates. Figure 5 shows agar gels with the swarming migration patterns of wild-type P. aeruginosa PAO1, a rhlA mutant, which is unable to synthesize rhamnolipids, and the same rhlA mutant after supplementation with an exogenous rhamnolipid extract. The observation of the migration patterns indicates that the rhamnolipid extract modulates the swarming by the RhlA-deficient strain, which responds with the formation of tendrils at the swarm fronts. Therefore, rhamnolipids and/or HAAs may act as inducers of swarming cells.

P. aeruginosa is among the most common causes of hospital-acquired infections
and is often associated with lung infection in cystic fibrosis patients [42]. Some P.
aeruginosa strains have developed resistance to most antimicrobial agents and are listed by WHO as one of the most critical threats to human health [43]. It produces a wide variety of virulence factors, including specialized protein secretion systems, the ability to form adherent biofilms and typical motility patterns, which account to its adaptability to many different hosts and environments [44,45].
Among its many phenotypical hallmarks, P. aeruginosa produces large amounts of rhamnolipids, surface-active amphipathic molecules produced as blends and composed by one or two rhamnose moieties linked to a dimer of R-3-hydroxy fatty acids with chain lengths ranging from C8 to C12 [46]. As reviewed by Reis et al. (2011) rhamnolipid biosynthesis in P. aeruginosa is directly controlled by quorum sensing, through the transcriptional regulator RhlR, which activates the rhlAB operon when complexed to C4-HSL [47]. Most functions attributed to rhamnolipids relate to their well-characterized physicochemical properties. However, there is a lack of comprehensive studies investigating any possible biochemical mechanisms involved in the phenotypes promoted by rhamnolipids in P. aeruginosa.
Although the mechanistic role of rhamnolipids in physiology and pathogenesis of P. aeruginosa is not fully understood, there is mounting evidence showing their relation with important virulence traits, such as biofilm formation and flagelar-driven motility [40,48]. Rhamnolipids have been associated to P. aeruginosa biofilms and play a central role in maintaining the multicellular structures and the dispersion of sessile biofilm cells [3,7,15]. In a similar manner, the hallmark swarming motility pattern of P. aeruginosa is fully dependant on the presence of rhamnolipids [18][19][20][21].
DNA microarrays and RT-qPCR have been widely used to analyze the global gene expression of P. aeruginosa within its complex adaptative behaviors, including comparative analyses of biofilm versus planktonic cells and the development of a specialized swarming behaviour [40,[49][50][51][52]. In the present study, we assessed the genome-wide expression profiles of a wild-type P. aeruginosa and its rhamnolipiddeficient derivative, using independent experiments ( Figure 1) designed at the stationary phase of growth. Interestingly, the absence of the rhlABC biosynthetic pathway in the mutant strain revealed a broad impact on gene expression profile, including 134 loci in the previously published annotation of the P. aeruginosa genome [1], comprising different functional categories ( Figure 2). The differential expression of a diverse group of genes indicates that there is a significant physiological difference between the rhamnolipid-producing and non-producing strains. The set of genes was mostly related to motility, prophages, transcription, and translation. Remarkably, a number of genes that encoded signaling proteins were positively regulated, while a relevant portion of genes involved in flagellar motility and chemotaxis were significantly repressed ( Table 1, Table   S2). Repressed motility-related genes included those encoding type b flagellin (fliC), two flagellar secretion chaperones (fliS and fliT), a fagellum number regulator protein (fleN) and a putative chemotaxis transducer (PA2788). On the other hand, the microarray analyses indicated that pilO, related to the biogenesis and functioning of the Type IV pili [26], was upregulated. Some loci encoding two-component systems also had their expression induced, including a putative sensor kinase (PA1180) and NarL (Table S1), which responds to nitrate sensing and modulates motility and virulence [53,54].
Based on the down-regulation profile of flagellar genes observed on the P. aeruginosa microarrays and considering their pivotal role on many phenotypes, we selected flagellar genes of different regulatory categories, including class II, III and IV [23,24] for further assessment with RT-qPCR assays. Hence, we further demonstrate a Previous reports suggest that planktonic and sessile subpopulations of P.
aeruginosa interact via lasIrhlI quorum sensing signaling, promoting their movement to the top of the adherent microcolonies [37]. This process, apparently mediated by flagella and rhamnolipids, is believed to contribute to the formation of typical mushroom-shaped structures, observed in mature P. aeruginosa biofilms [7,37]. In fact, Pamp & Tolker-Nielsen (2007) demonstrated the complete absence of mushroom-like structures in biofilms formed by a rhlA-knockout strain, deficient in the production of rhamnolipids.
However, the lack of expression of flagellar genes in the rhlA-mutant strain was not suggested at the time, an evidence that we demonstrate in the present study, for the first time.
In a similar fashion, the ability to produce rhamnolipids has been associated to virulence, using in vitro and in vivo infection models, as a rhamnolipid-deficient P.
aeruginosa strain was eradicated more rapidly and significantly, when compared to the wild-type parental strain [55]. As far as the swarming motility is concerned, in the light of the present study, this observation could be related to the abrogated motility observed in the rhlA-defective strain ( Figure 5) and the downregulation of flagellar genes (Table 1, Figure 3). Supporting this hypothesis, Overhage et al. (2008) provided that under swarming conditions, P. aeruginosa PA14 exhibited the upregulation of many virulence-related genes, including genes for the Type III Secretion System (T3SS) and its effector proteins [40] required for colonizing host mucosal surfaces.
Expression of type IV fimbriae by P. aeruginosa is also associated with its ability to form adherent biofilms onto biotic and abiotic surfaces, twitching and swarming motility. The swarming motility by P. aeruginosa is dependent on cell-cell signaling and requires flagella and type IV fimbriae, in addition to the production of rhamnolipids [18,20,56]. However, no biochemical mechanism or function has been specifically demonstrated for rhamnolipids in the context of biofilms and motility patterns. It has been proposed that P. aeruginosa requires flagella during swarming to overcome adhesive interactions mediated by fimbriae type IV [57]. In fact, swarming motility is abolished in the absence of rhamnolipid-type surfactants, when the rhlA gene is not functional [18][19][20][21]. In the present study, we show for the first time that the swarming motility can be restored to a rhlA-mutant strain, with addition of exogenous rhamnolipids obtained from the wild-type strain PAO1 ( Figure 5).
Adding to the discussion regarding surface interactions, a previous study demonstrated there is an inverse relationship between the production of exopolysaccharides (EPS) and the swarming motility in P. aeruginosa [58]. The knockout of sadC, which codes for diguanylate cyclase SadC, results in a hyperswarmer phenotype, while multicopy expression of this gene promotes a phenotype often associated with overproduction of EPS [58]. Thus, understanding how EPS and motility influence the different P. aeruginosa phenotypes can also contribute to the development of strategies against chronic and persistent infections caused by this opportunistic pathogen.
Therefore, we also investigated the expression profile of genes related to the EPSs biosynthesis of P. aeruginosa PAO1. According to our RT-qPCR results (data not shown), we observed an induction of expression of EPS genes (pel and psl components) in the non-motile rhlA-mutant strain, deficient in the production of rhamnolipids, when compared to expression in the wild-type strain. A recently published study also lists the contrast that we see in our analyzes, as null flagellar mutants overexpressed exopolysaccharides Pel and Psl in biofilms of P. aeruginosa, in the laboratory and in biofilms isolated from cystic fibrosis (CF) infection [59]. via SadC [62] or HptB branch [63], which affect the levels of the secondary messenger c-di-GMP. In addition, the Wsp chemosensory system triggers a signal transduction array in response to surface sensing, by controlling the synthesis of c-di-GMP, which promotes the formation of biofilms and decreases the expression of the flagellar genes [64][65][66]. The main target of c-di-GMP within this pathway is the transcriptional regulator FleQ, which positively regulates the expression of flagellar genes while repressing EPS genes, such as pel. Upon interaction with c-di-GMP, FleQ is inhibited, leading to reduced expression of flagellar genes and upregulation of EPS genes [64,67]. Therefore, the GacS network and the Wsp signalling pathway represent systems that are compatible to the rhamnolipiddependent modulation of flagellar expression observed in the present study. Interestingly, within the Wsp chemosensory system, flagellar and EPS genes are inversely regulated, a behaviour we have consistently observed in regard to rhamnolipids. FleQ, the master regulator of flagelar biogenesis, is the major player in the Wsp system [67]. Whether rhamnolipids would act directly upon any sensory system, or indirectly, through its wetting activity, controlling bacteria's surface sensing, remains to be understood.
We have also identified the positive regulation of hcp1, a gene that encodes a component of the type VI protein secretion system (H1-T6SS) [27] and some prophage genes (Table 1, Figure 2). Prophages play an important role in the evolution of bacterial genomes and their pathogenicity [68] and the expression of phage genes has also been shown to cause negative impacts on bacterial motility. A study revealed that flagella are reduced on the surface of cells that overexpress phage genes in strains PAO1 and PA14, in contrast to the observed hyperpiliation by type IV pili [30].
Regarding the data obtained through the DNA microarrays, the present study also revealed that the expression of genes responsive to iron, among which we can highlight PA4221 and PA4220, were negatively regulated in the rhlA-mutant strain ( Figure 1 and Table 1). It has been reported that the reduction in iron uptake, in mutants deficient in its acquisition mediated by pyoverdines, may impair the formation of biofilms. However, this deficiency can favor twitching motility in P. aeruginosa. Thus, under iron-limiting conditions P. aeruginosa presents greater motility and forms unstructured flat biofilms [69,70].
In the last decades, many different approaches are being used for a better understanding of the regulatory mechanisms involved in the formation of biofilm in P.
aeruginosa. Through a proteomic analysis, a hypothetical protein encoded by the PA3731 gene was identified. The PA3731 gene was related to swarming motility and the synthesis of rhamnolipids. The PA3731-mutant were deficient in biofilm formation compared to the reference strain PAO1 [25]. Corroborating with those data, in our transcriptomic analyses, we also observed repression of the PA3731 gene in the rhlA-knockout strain, compared to the wild-type strain.
Taken together, these results show significant evidence that rhamnolipids and/or their HAA precursors, the major biosynthetic products of the rhlABC pathway, seem to modulate gene expression in P. aeruginosa. Further investigation is required to understand the biochemical and/or biophysical mechanisms involved in the process.

Conclusions
The genome-wide transcriptomic analysis revealed that a rhlA-mutant strain, unable to produce rhamnolipids, had a substantially different gene expression profile when compared to the wild-type PAO1. Notably, several key flagellar genes were repressed in the mutant strain, suggesting a possible gene regulation pattern mediated by rhamnolipids and/or their HAA precursors. Swarming motility assays support this hypothesis, since the non-motile rhlA-mutant strain had its swarming ability restored by the addition of exogenous rhamnolipids.
The strains were grown in 250 mL erlenmeyer flasks containing 100 mL of LB medium, until the stationary growth phase (OD600 2.0), with shaking (170 rpm), at 30 °C.
An aliquot of the culture was collected for total RNA extraction.

RNA extraction and cDNA synthesis
Total RNA was obtained from bacterial cultures at the the stationary growth phase Genes were considered differentially expressed in the rhlA-knockout mutant compared to the wild type strain. Expression differences were assessed using comparative and paired analyzes between desired groups, with fold changes ≤ -1.5 or ≥ 1.5. The data obtained were processed from the Transcriptome Analysis Console (TAC) Software, version 4.02 (Affymetrix, Thermo Fisher Scientific, Waltham MA, USA). For this purpose, the Gene Level Differential Expression Analysis approach and the ANOVA test were used, with statistical significance for p < 0.05 and FDR (False Discovery Rate) of 5%. The DIAMOND [72] and PseudoCAP [73] tools were also used to identify proteins.
Functional categories were determined using the COG database [22].

Confirmation of differentially expressed genes in the microarray by RT-qPCR
Differentially expressed genes in the microarrays, fliC, flgL, fliD and fliT, related to flagellar motility, were selected for RT-qPCR quantification of expression. The gene used to normalize the experiments was 16S rRNA and the expression values of the wild-type strain (PAO1) were adopted as a basal reference (1.0). RT-qPCR was performed with SYBR Green PCR fluorophore Master Mix (Applied Biosystems, Foster City, USA) and the primers were designed with the aid of the online tool Primer3Plus [74] and the Beacon Designer software [75] (Table S3). RT-qPCR was also performed for genes rhlA and rhlB: primers for the 16S rRNA, rhlA and rhlB genes were used with TaqMan probes synthesized by Life Technologies (Applied Biosystems, Foster City, USA). The relative quantification analyses of rhlA and rhlB gene expression were performed with the 2(-Delta Delta C(T)) method [76]. The RT-qPCR reactions were conducted on a CFX96 Real-Time PCR Detection Systems platform (Bio-Rad Laboratories, Hercules, USA).

Extraction and quantification of rhamnolipids
Culture-free supernatants were obtained by centrifugation at 6,000 ×g for 25 min, at 15 °C and acidified with 1 N HCl into pH 3.5, following the addition of ethyl acetate at a 1:3 ratio. Rhamnolipids were then extracted into the organic phase and dried with a HetoDrywinner rotary evaporator (Heto-Holten, Gydevank, Denmark). The resulting preparation was dissolved in methanol, lyophilized and stored at -20 °C. Prior to use, a 8.0 µg/μl rhamnolipid suspension was prepared in electrolyte water solution (15 mM NaHCO3, 10 mM NaCl). Quantification of rhamnolipids was performed as previously described [78], through quantification of rhamnose with high performance liquid chromatography (HPLC). Briefly, 100 μL of a 10 M sulfuric acid solution were added to 1 mL of rhamnolipids, in electrolyte solution, and heated at 100 ºC for 4 hours. The hydrolyzed solution was neutralized with 10 M sodium hydroxide, filtered with a 0.22 μm membrane and analyzed with HPLC. The HPLC system used in this study was an Agilent 1260 Infinity (Santa Clara, CA, USA) with a refractive index detector. The analytical column was an Aminex HPX-87H (Bio-Rad Laboratories, Hercules, USA), the mobile phase was 5 mM sulfuric acid, with a flow rate of 0.6 mL/min and oven temperature of 45 ºC. The procedure was carried out in triplicate and a mass spectrometric conversion factor of 2.5 was applied to convert the concentration of rhamnose into rhamnolipids [78].

Plate Swarming Motility Test
Swarming assays were performed on plates containing 0.5% agar in M8 medium, supplemented with 0.2% Glucose, 0.5% Tryptone and 1 mM MgSO4 [79]. 2.5 µL of the cultures were added to the center of the plate and incubated at 30 °C for 48 hours. In the case of the rhlA-knockout strain (ΔrhlA::Gm R ) the culture replicates were conditioned with an aqueous rhamnolipid solution, at a final concentration of 4.0 μg/μL immediately before application onto the swarming medium.

Statistical analysis
The samples were compared for expression levels by the control strain (PAO1) and the knockout strain (rhlA-mutant) using the paired t-test of the Graphpad Prism 4.0 program (GraphPad Software Inc, San Diego, CA, USA). Each sample is representative of 3 biological replicates. The significance level established was 95% (p < 0.05).

Ethics approval and consent to participate
Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated during the current study are available in the Gene Expression Omnibus (GEO) repository (https://www.ncbi.nlm.nih.gov/geo/), under the accession number GSE163816.

Competing interests
The authors declare that there is no potential conflict of interest.            Figure 1 Differentially expressed genes. Hierarchical clustered heatmaps of the datasets show a total of 53 genes, with a fold-change ≤ -1.5 or ≥ 1.5 and p-value < 0.05 between samples of P. aeruginosa PAO1 (PAO1 replicas 1, 2 and 3) and rhlA-mutant (rhlA replicas 1, 2 and 3). The relative differences in gene expression are proportional to the intensity of the coloration (heatmap), with blue referring to decreased and red to increased expression.

Figure 2
Functional classi cation of P. aeruginosa genes differentially expressed in the microarrays. rhlA-knockout mutant compared to wild-type parental strain PAO1, with a fold change ≤ -1.5 or ≥ 1.5. The analyses were performed using the COG database.  Expression of rhlA and rhlB in P. aeruginosa PAO1 and rhlA-mutant strain. The picture shows the quanti cation of mRNA of a) rhlA and b) rhlB by RT-qPCR, with 16S rRNA gene as endogenous control.
Expression values of rhlA-mutant strain are relative to the control (PAO1), normalized as 1.0. Bars indicate ± standard error. Relative expression was calculated using the comparative Cq method. Figure 5 Swarming motility assay. a) P. aeruginosa PAO1, b) rhlA-knockout mutant and c) rhlA-knockout mutant supplemented with exogenous rhamnolipids.