Full Transcriptomic Response of Pseudomonas aeruginosa to an Inulin-Derived Fructooligosaccharide

Pseudomonas aeruginosa is an ubiquitous gram-negative opportunistic human pathogen which is not considered part of the human commensal gut microbiota. However, depletion of the intestinal microbiota (Dysbiosis) following antibiotic treatment facilitates the colonization of the intestinal tract by Multidrug-Resistant P. aeruginosa. One possible strategy is based on the use of functional foods with prebiotic activity. The bifidogenic effect of the prebiotic inulin and its hydrolyzed form (fructooligosaccharide: FOS) is well established since they promote the growth of specific beneficial (probiotic) gut bacteria such as bifidobacteria. Previous studies of the opportunistic nosocomial pathogen Pseudomonas aeruginosa PAO1 have shown that inulin and to a greater extent FOS reduce growth and biofilm formation, which was found to be due to a decrease in motility and exotoxin secretion. However, the transcriptional basis for these phenotypic alterations remains unclear. To address this question we conducted RNA-sequence analysis. Changes in the transcript level induced by inulin and FOS were similar, but a set of transcript levels were increased in response to inulin and reduced in the presence of FOS. In the presence of inulin or FOS, 260 and 217 transcript levels, respectively, were altered compared to the control to which no polysaccharide was added. Importantly, changes in transcript levels of 57 and 83 genes were found to be specific for either inulin or FOS, respectively, indicating that both compounds trigger different changes. Gene pathway analyses of differentially expressed genes (DEG) revealed a specific FOS-mediated reduction in transcript levels of genes that participate in several canonical pathways involved in metabolism and growth, motility, biofilm formation, β-lactamase resistance, and in the modulation of type III and VI secretion systems; results that have been partially verified by real time quantitative PCR measurements. Moreover, we have identified a genomic island formed by a cluster of 15 genes, encoding uncharacterized proteins, which were repressed in the presence of FOS. The analysis of isogenic mutants has shown that genes of this genomic island encode proteins involved in growth, biofilm formation and motility. These results indicate that FOS selectively modulates bacterial pathogenicity by interfering with different signaling pathways.

Pseudomonas aeruginosa is an ubiquitous gram-negative opportunistic human pathogen which is not considered part of the human commensal gut microbiota. However, depletion of the intestinal microbiota (Dysbiosis) following antibiotic treatment facilitates the colonization of the intestinal tract by Multidrug-Resistant P. aeruginosa. One possible strategy is based on the use of functional foods with prebiotic activity. The bifidogenic effect of the prebiotic inulin and its hydrolyzed form (fructooligosaccharide: FOS) is well established since they promote the growth of specific beneficial (probiotic) gut bacteria such as bifidobacteria. Previous studies of the opportunistic nosocomial pathogen Pseudomonas aeruginosa PAO1 have shown that inulin and to a greater extent FOS reduce growth and biofilm formation, which was found to be due to a decrease in motility and exotoxin secretion. However, the transcriptional basis for these phenotypic alterations remains unclear. To address this question we conducted RNAsequence analysis. Changes in the transcript level induced by inulin and FOS were similar, but a set of transcript levels were increased in response to inulin and reduced in the presence of FOS. In the presence of inulin or FOS, 260 and 217 transcript levels, respectively, were altered compared to the control to which no polysaccharide was added. Importantly, changes in transcript levels of 57 and 83 genes were found to be specific for either inulin or FOS, respectively, indicating that both compounds trigger different changes. Gene pathway analyses of differentially expressed genes (DEG) revealed a specific FOS-mediated reduction in transcript levels of genes that participate in several canonical pathways involved in metabolism and growth, motility, biofilm formation, β-lactamase resistance, and in the modulation of type III and VI secretion systems; results that have been partially verified by real time quantitative PCR measurements. Moreover, we have identified a genomic island formed by a

INTRODUCTION
The human pathogen Pseudomonas aeruginosa causes a wide array of life-threatening acute and chronic infections, particularly in immunocompromised, cancer, burn wound, and cystic fibrosis patients (Juhas, 2015). This bacterium is moreover one of the leading causes of nosocomial infections affecting hospitalized patients (Buhl et al., 2015) and mortality associated with hospitalacquired P. aeruginosa infectious like ventilator-associated pneumonia or bacteremia is above 35% (Lynch et al., 2017).
Moreover, under continuous antibiotic treatment the intestinal microbiota integrity is compromised and bears depletion of the intestinal microbiota (Dysbiosis), hence, physiological colonization resistance subsequently facilitates the establishment of the Pseudomonas aeruginosa in the intestinal ecosystem which might be considered an important internal source for P. aeruginosa infection (Ohara and Itoh, 2003;Von Klitzing et al., 2017). It is important to note that pathological alterations of the intestinal microbiota (dysbiosis) is related with continuous antibiotic treatment, obesity, diabetes and fatty liver, and of course alterations of the intestinal barrier function (IBF) as in inflammatory bowel disease and metabolic syndrome (Cano et al., 2013;Miura and Ohnishi, 2014).
The severity and permanence of these infections are related to the ability of P. aeruginosa to resist the effect of antibiotics through the formation of biofilms (Mah et al., 2003;Hoiby et al., 2010;Taylor et al., 2014). Important research efforts have been made to study the molecular mechanisms related to the formation maturation and subsequent dispersion of the biofilm (Stoodley et al., 2002;Flemming et al., 2007). A number of surface proteins and appendages, including flagella and type IV pili, were found to be associated with biofilm formation (Klausen et al., 2003;Anyan et al., 2014). Furthermore, this species is characterized by its ability to synthesize the virulent factors exotoxin A and pyocyanin (Ortiz-Castro et al., 2014) that block protein synthesis consequently leading to cell death (Gaines et al., 2007). Treatment of P. aeruginosa infections can be particularly challenging because this bacterium is intrinsically resistant to multiple antibiotics and can easily acquire new resistances (Breidenstein et al., 2011). In fact, over the past three decades, antibiotic resistance among P. aeruginosa has escalated globally, via the global dissemination of several multidrug-resistant epidemic clones (Miyoshi-Akiyama et al., 2017). Pseudomonas aeruginosa infections thus represent a severe threat to human health worldwide and the World Health Organization has declared this bacterium the second priority pathogen for research and development of new strategies to fight it (WHO, 2017). Besides conventional treatments, one possible strategy is based on the use of functional foods with prebiotic activity which are non-digestible foods (mostly oligosaccharides) that selectively stimulate the growth of a limited number of hostfriendly colonic bacteria (Froebel et al., 2019). Thus, from a chemical standpoint, resistance to human digestive enzymes and low absorption are key for these compounds to reach the distal parts of the gut, where they can be fermented by the microbiota, which in turn is selectively modified in the process. These additional actions of prebiotics tend to enhance the capacity of the mucosa to contain luminal microorganisms and their components, i.e., intestinal barrier function (IBF). Normally, passage of microorganisms and/or their components such as Lipopolysaccharides (LPS) to the mucosa and from there to the bloodstream (translocation) is minimal, and the immune system develops tolerance to the microbiota, without inflammation. Conversely, when IBF is compromised translocation ensues, depending on the nature of the dysfunction and the physiological/pathological context. Therefore, inflammation of the intestine is considered to stem from augmented translocation, which engages the adaptive immune system, ultimately resulting in uncontrolled inflammation. Thus, reinforcing IBF may be protective and is viewed as therapeutic in this context (Natividad and Verdu, 2012;Duseja and Chawla, 2014). A significant number of natural compounds have been found to inhibit bacterial growth, although their mechanisms of action frequently remain unclear (Amer et al., 2010). Fructooligosaccharides (FOS) are short-chain oligosaccharides that are generated by hydrolysis of the polysaccharide inulin, which is composed of two to 60 fructose monomers. Inulin is found in different nutrients such as wheat, onion, garlic and banana (Lattimer and Haub, 2010) and is the most common used fiber in prebiotics that, when used in combination with other probiotics, is able to promote the growth of specific beneficial gut bacteria such as bifidobacteria (Gibson et al., 1995;Bosscher et al., 2006). A number of studies illustrate that FOS and inulin exert a number of different effects on humans and animals. For example, most oraly delivered plant substrate supplements that prevent gastrointestinal infections such as prebiotin TM and symbioram TM contain inulin and FOS, indicating that these compounds are also able to reduce bacterial infection. Moreover, oligosaccharids and in particular inulin and FOS were found to have beneficial effects on intestinal immunity and intestinal barrier function (IBF) . Another study has shown that oligosaccharides from goats milk as well as galactooligosaccharides, modulate cytokine production by intestinal epithelial cells and monocytes via a mechanism involving Toll-like receptor 4 (TLR4) . TLRs are located on the cell membrane and in endosomes, where they recognize components of cell membranes (TLR2/6, TLR2/TLR4), nucleic acids (TLR3, 7, 8, and 9) and flagellin (TLR5). TLR4 is one of the non-pathogen recognition receptors (PRRs), which are key elements in the communication between the host and the microbiota (Sánchez de . However, further clinical oral applications will require studies on the potential effects of these natural substrates on the human body, which corresponds to the research need addressed in this article. In addition to the bacterial growth promoting role of inulin and FOS, we reported that FOS inhibited bacterial growth and biofilm formation of P. aeruginosa PAO1 (Ortega-González et al., 2014). Additionally, both compounds caused opposing effects on bacterial motility. While FOS inhibited motility, an increased motility was observed in the presence of inulin. Moreover, in co-cultures with eukaryotic cells (macrophages) FOS, and to a lesser extent inulin, reduced the secretion of the inflammatory cytokines IL-6, IL-10, and TNF-α. We were also able to show that the reduction in cytokine secretion is due to a FOSmediated modulation of the NF-κβ signal transduction pathway (Ortega-González et al., 2014). To gain insight into the detailed molecular processes triggered by FOS and inulin, we report here results from RNA-seq studies.

Materials
Inulin and FOS were purchased from BENEO-Orafti (Tienen Belgium). Stock solutions at 200 g/L in modified M9 minimal medium were sterilized using 0.22 µm cut-off filters and aliquots were stored at −20 • C.

Culture and Growth Conditions
Pseudomonas aeruginosa PAO1 was grown overnight at 37 • C in minimum M9 medium. The resulting cultures were then used to inoculate 50 ml of minimum M9 medium supplemented by 5 mM of citrate (MM9) (in 250 ml Erlenmeyer flasks) to an initial OD 600 of 0.01 and incubated with shaking at 200 rpm at 37 • C. When cultures reached OD 600 = 0.05, FOS or inulin were added to a final concentration of 20 mg/ml and cultures were harvested for analysis 1 h later.

RNA Extraction, Library Preparation and RNA Sequencing
Total RNA was extracted with the TRI reagent (Ambion) using the manufacturer's instructions. The RNase inhibitor RiboLock (Fermentas) was added to the samples and DNA was removed by treatment with DNase I (Fermentas). The integrity of the RNA samples was assessed with an Agilent 2100 Bioanalyzer (Agilent Technologies). Subsequently, the 23S, 16S, and 5S rRNAs were removed by subtractive hybridization using the MICROBExpress kit (Ambion) following the protocol reported by Gómez-Lozano et al. (2014). Removal of the rRNA was confirmed by an analysis with an Agilent 2100 Bioanalyzer (Agilent Technologies). Sequencing libraries were prepared using the TruSeq Stranded mRNA Sample Preparation kit (Illumina). After each step, the samples were validated using an Agilent 2100 Bioanalyzer (Agilent Technologies) and the final RNA concentration was measured using a Qubit 2.0 Fluorometer (Invitrogen). The libraries were sequenced using the Illumina HiSeq2000 platform with a paired-end protocol and read lengths of 100 nucleotides.

RNA-seq Analysis
The quality of sequenced reads was assessed using FastQC software, version 0.11.5 (Andrews and Fast, 2010). Single-end reads were aligned to the reference genome of P. aeruginosa PAO1 (GenBank accession number: AE004091.2) using SAMtools v 0.1.19 (Li et al., 2009). BAM files from SAM tools were used as input for the feature counts function (Liao et al., 2014) from the Rsubread package (Liao et al., 2013) of Bioconductor version 3.5 to generate a matrix of annotated genes with their corresponding raw counts. An average of 84.5% reads were successfully mapped to the reference genome. The count data were then analyzed to look for differential gene expression levels and statistical significance using DEseq2 (Love et al., 2014;Kimberly and Stephen, 2015).
The threshold to define differences in transcript levels was a statistical Log2 fold change. Genes were considered significantly differentially expressed when p-values were below 0.05.

RNA-Sequencing Data Registration Number
The sequence reads have been deposited in the GEO database under accession N • : GSE124468. The following secure token has been created to allow for the review of record GSE124468: glshocwuzvkprop. The above table shows the list of the significant down regulated genes and control with a log2 fold change cut off ≥ 0.5 by taking significant p-value < 0.05.
was retrotranscribed following the protocol provided in the manufacturer protocol (iScript BioRad, Alcobendas, Spain) and DNA sequences were amplified with a MX3005P real time PCR instrument (Stratagene) using the primers listed in Supplementary Table S3. The genes of interest were amplified by PCR using the Go Taq@qPCR Master Mix (Promega, Madison, WI, United States) as well as 1 µl of the cDNA template and the primers listed in Supplementary Table S3. Forty PCR cycles were conducted using an annealing temperature of 61 • C. The cycle threshold values were normalized to that of the reference transcript, 16S RNA, and data were normalized to the control.

Generation of Mutants in Genes of the Genomic Island PA0643, PA0644, and PA0646
To generate the PA0643:Gm, PA0644:Gm, and PA0646:Gm mutants 656, 241, and 636 pb DNA fragments, respectively, covering the central part of the genes were amplified by PCR from P. aeruginosa PA01 genomic DNA. The resulting products were cloned into plasmid pMBL to yield plasmids pMBL:PA0643, pMBL:PA0644, and pMBL:PA0646. Subsequently, the resulting plasmids were digested with BamHI, which liberated the PA0643, PA0644, and PA0646 fragment. The plasmid pCHESI was also digested with BamHI, to liberate the gentamicin resistance gene (Gm) in order to ligate it with the three DNA fragments. The resulting chimeric DNA was cloned into pMBL digested with BamHI, to yield pMBL:PA0643 Gm, pMBL:PA0644 Gm and pMBL:PA0646 Gm. The resulting plasmids were electroporated into P. aeruginosa PA01 for homologous double recombination. Mutant strains were selected on Gm plates and the correctness of the mutation was verified by Southern blotting (Sambrook et al., 1989;Molina-Fuentes et al., 2015).

Semi-Quantitative Determination of Biofilm Formation
Semi-quantitative determination of biofilm formation were performed as previously described (Christensen et al., 1985). P. aeruginosa PA0643, PA0644, and PA0646 mutants were tested in the biofilm-forming capacities in Minimum medium supplemented with 5 mM citrate. The determination of biofilm production was performed after 2, 4, and 6 h of growth by dissolving crystal violet from the biofilm with an ethanol-acetone mixture (70:30) and the absorbance measure at 590 nm.

Motility Assays
Assays were carried out to determine the effect of the PA0643, PA0644, and PA0646 deletion gene on swimming, twitching and swarming. For swimming assays, bacteria were placed with the help of a sterile tooth-pick at the center of plates containing a 5 mm layer of LB medium with 0.3% (w/v) Bacto agar, 0.2% casamino acids (w/v), and 30 mM glucose. Plates were incubated at 37 • C for 24 h and the radial diffusion of bacteria, due to swimming, was inspected. To monitor twitching motility, bacteria were placed with a toothpick into a 2 mm thick layer containing 1.5% (w/v) Bacto agar, 0.2% (w/v) casamino acids,  The above table shows the list of the significant up regulated genes and control with a log2 fold change cut off ≥ 0.5 by taking significant p-value < 0.05.
Frontiers in Microbiology | www.frontiersin.org FIGURE 3 | Functional comparison of differentially expressed genes in the presence of FOS and inulin as compared to the control. Functions are organized into three groups: Cellular components, Molecular Function, and Biological Process. The functional groups with the main differences between FOS and inulin are "Molecular transducer," "Biological adhesion" "Developmental process" and "locomotion system." and 30 mM glucose. After incubation at 37 • C for 24 h, the expansion of bacteria on the plate was observed. For swarming assays, 5 µl of an overnight culture of bacteria were placed in the center of swarm plates, which are made of 0.5% (w/v) Bacto agar supplemented with 0.2% (w/v) casamino acids and 30 mM glucose. Plates were incubated at 37 • C for 24 h, followed by an inspection of the surface movement of the bacteria. All motility assays were performed in triplicate.

Statistical Analysis
All results are expressed as means from three cultures with the corresponding standard deviations. Data were analyzed for statistical significance using the one-way ANOVA analysis and a posteriori least significance test. All analyses were carried out with the SigmaStat 2.03 program (Jandel Corporation, San Rafael, CA, United States). Fitting of dose-response curves was done using Origin 7.0 (OriginLab Corporation, Northampton, MA, United States). Differences were considered significant at p < 0.05.

FOS and Inulin Induce Differential Changes in Pseudomonas aeruginosa Transcript Levels
To understand the cellular response of P. aeruginosa PAO1 to FOS and inulin treatment, we conducted RNA-seq studies. Transcriptomic changes were determined in duplicate cultures grown in the absence and in the presence of either FOS or inulin at final concentrations of 20 mg/ml. Between 5,300,000 and up to 7,500,000 reads were obtained for inuline and FOS respectively, of which approximately 85% could be assigned to the 6,322 coding regions of the P. aeruginosa PAO1 reference genome ( Table 1).
The heat map shown in Figure 1A illustrates genes with the most important alterations in transcript levels in the presence of FOS/inulin as compared to the control. For both compounds the number of genes with increased transcript levels were superior to those with decreased levels (see Figure 1B). FOS and inulin induced changes in the transcript level of 217 and 258 genes respectively, compared to the control to which no polysaccharide was added (Figure 2). Importantly, down changes in transcript levels of 57 and 83 genes were found to be specific for inulin or FOS, respectively (Figure 2A). Moreover, 201 and 134 genes showed an increase in transcript levels by inulin and FOS, respectively ( Figure 2B) indicating that both compounds trigger different changes.
Supplementary Tables S1 and S2 show the list of genes for which the expression level significantly increased or decreased in the presence of 20 mg/ml of inulin (Supplementary Table S1) and FOS (Supplementary Table S2). Analysis of the expression pattern showed that only 22 genes had decreased levels in the presence of both compounds (Figure 2A and Table 2). Among these genes the most prominent changes were observed for genes involved in: (1) Organic acid transport such as PA1342 (aatj), which encodes a C4-dicarboxylate transport protein and PA1183 (dctA), (2) Central metabolism, like a PA0795 (prpC), which regulates a citrate synthase, PA2008 (fahA) that controls a fumarylacetoacetase, and PA1585 (sucA) encoding a 2-oxoglutatate dehydrogenase, (3) Oxidative stress such as superoxide dismutase PA4366 and (4) Virulence system like  The above table shows the list of the significant up regulated genes with a log2 fold change cut off ≥ 0.5 by taking significant p-value < 0.05.
the Type VI secretion ATPase (PA0090). These genes were outnumbered by the genes for which both compounds caused an increase in transcript levels (Figure 2 and Table 3). A significant number of these genes appear to be involved in sensing (transcriptional regulators, sensor kinases, and chemotaxis transducers), motility, glucose metabolism as well as control of transcription and protein synthesis (Figure 2 and Table 3).

Functional Analysis of Pseudomonas aeruginosa Transcriptome Following Exposure to FOS and Inulin
GO terms (WEGO) enrichment analyses were conducted to characterize the DEG (Differentially expressed genes) profiles and K-means clustering was performed to further investigate their biological function. We found that the differentially expressed genes can be classified into 32 categories that belonged to three gene ontology (GO) categories, i.e., the biological process, the cellular component or the molecular function (Figure 3). There were more genes classified into biological processes than the other two categories and most genes were predicted to have a binding function, as these genes are primarily involved in protein metabolism (Figure 3).
The over-expressed DEGs were assigned to 15 GO categories based on biological processes (Figure 3) and the results showed that the response to the stimulus, metabolic process, biological regulation, the establishment of localization and pigmentation were among the most highly represented groups in the biological process category in the presence of inulin or FOS. However, "biological adhesion" and "developmental process" showed a drastic decrease in the number of genes between inulin or FOS samples. While in the "locomotion process" most of the inhibited genes (43%) are annotated in FOS samples (Figure 3).
Furthermore, DEGs were assigned to seven GO categories based on cellular component and the result showed that "Cell" and "Cell part" are similar and highly represented groups for FOS and inulin samples (Figure 3). However, the "extracellular region" and the "organelle part" are lower and distinctly represented in the presence of inulin or FOS (Figure 3).
Genes with altered transcript levels could be grouped into 10 GO terms with different molecular functions of which the categories "antioxidant, " "binding, " "electron carrier, " "transcription regulator, " "structural molecular", and "transcription regulator" were most populated (Figure 3). It is worth noting that, in the "molecular transducer" category a significantly higher number of genes were noted for inulin as compared to FOS. The individual genes that   The above table shows the list of the significant up regulated genes with a log2 fold change cut off ≥ 0.5 by taking significant p-value < 0.05.
were classified into these different GO terms are provided in Supplementary Tables S1, S2.
Surprisingly, in addition to increasing the expression of genes involved in different metabolic pathways necessary for bacterial growth, we found that transcript levels of many genes that are related to P. aeruginosa motility were increased in the presence of FOS (Table 5), exemplified by genes involved in twitching motility (pilG), components of the flagellar motor or chemoreceptors including CtpL that mediates specific taxis to inorganic phosphate -a key regulator of P. aeruginosa virulence (Zaborin et al., 2009;Bains et al., 2012). In the same manner, transcript levels of many genes associated within the glucose metabolism were also increased including genes eda, zwf, gapA, glk, edd, or gltR (Table 5) suggesting that P. aeruginosa can catabolize these compounds.

Effect of FOS on Virulence Related Gene Transcript Levels
All the DEGs were mapped to KO terms in the KEGG database to identify FOS modulated genes that play a role in bacterial virulence, motility, or sensitivity to antibiotics and the corresponding genes are provided in Table 6.
Moreover, FOS treatment resulted in lower transcript levels of the dctA gene (PA1183) which is associated with the normal growth of P. aeruginosa (Valentini et al., 2011) and of the icmP gene (PA4370) encoding an metalloproteinase outer membrane protein, which has been shown to degrade the plasminogen activator (Christian et al., 2015) and plays a key role in the Pseudomonas aeruginosa pathogenicity. These FOS mediated alterations in transcript levels have been confirmed by rt-qPCR studies (see Table 8). Altogether, data suggest that FOS acts FIGURE 5 | Location of the genomics island with genes widely repressed by FOS-treatment in Pseudomonas aeruginosa PAO1 genome. Figure were made using IslandViewer4 (Bertelli et al., 2017). Genes ids with a red dot represent genes repressed in presence of FOS treatment. Genes with blue dot represent genes identified in a genomic island in P. aeruginosa PAO1 genome.
as a signal molecule that modulates bacterial virulence through distinct signaling pathways.

Confirmation That FOS Reduces Expression of Structural Proteins of Secretion System III and VI
To confirm that FOS mediates changes in secretion system genes, we have conducted rt-qPCR experiments to determine the influence of FOS and inulin on the expression of 4 genes that encode proteins that are part of type III and VI secretion systems. The products of the pcrV and exsA genes control the activation of the type III secretion system (Lee et al., 2010), whereas the proteins encoded by hcp1 and vgrG1 are necessary for the type VI secretion system (Hachani et al., 2011).
We found that inulin caused a significant increase in pcrV and exsA transcript levels, whereas those of hcp1 and vgrG1 did not vary ( Figure 4A). In contrast, the expression of exsA and hcp1 were dramatically down regulated by factors of approximately 20 and 7, respectively, in the presence of FOS (Figure 4B). FOS but not inulin down regulated the expression of two components of the type III and VI secretion system.

Identification of Genomics Islands Widely Repressed by FOS-Treatment in Pseudomonas aeruginosa PAO1
Due to their relevance to human health, extensive efforts have been made to study genomic islands, which are large genetic elements acquired through horizontal transmission (Juhas et al., 2009;Stephen and Julian, 2015;Mao and Lu, 2016).
Our data, using RNAseq sequence analysis, indicates the presence of a genomic island in PAO1 of ∼17 kb comprising 15 genes that had been inserted into the 3 -end of the PA0639 gene ( Figure 5) through a phage encoded R2/F2 pyocin (Chang et al., 2005). Most of its genes were clearly repressed in the presence of FOS. Unfortunately, most of these genes code for proteins with unknown functions. Furthermore, in order to determine the biological role of this genomic island, various isogenic mutants were constructed and submitted to a phenotypic analysis that investigates changes in growth, biofilm formation, and motility.
Interestingly, PA0643, PA0644, and PA0646 mutants showed reduced growth inhibition compared to the wild-type strain (Figure 6). In addition, while the PA0643 mutant strain did not cause any significant changes, the PA0644 and PA0646 isogenic mutant demonstrated a reduction in biofilm formation at 4 and 6 h (Figure 7). Moreover, PA0643, PA0644, and PA0646 mutants were tested for their ability to swim, swarm, and twitch and the results showed that the mutants exhibited differences in all three types of motility (Figure 8). While PA0643, PA0644, and PA0646 mutants caused the same reduced change in swarming motility, it was found that PA0644 and PA0646 mutants significantly inhibited (at least 50-60% of WT) the swimming motility. Interestingly, the twitching motility has been drastically reduced in the case of the PA0646 mutant (Figure 8). Markedly, this is the first study which shows that the deletion of PA0643, PA0644, and PA0646 genes are able to block P. aeruginosa swarming, swimming, and twitching motility. However, further studies are required to elucidate the function of these proteins.

DISCUSSION
Various factors appear to be implicated in the ability of P. aeruginosa to cause health problems. While surface structures, including pili and the polysaccharide, seem to favor biofilm formation and adhesion of P. aeruginosa to host cells (Pollack, 1984(Pollack, /1992Prince, 1992) promoting colonization, its exotoxins in combination with various secretion systems deteriorate the host's defenses. The identification of signal molecules and the study of their corresponding molecular mechanism, which is associated with these processes, are of utmost interest.
Prebiotics have been shown to exert beneficial effects on human health by altering the intestinal microbiota and also by inhibiting the progression of some pathogenic strains (Knol et al., 2005), due to an indirect effect caused by the selective growth of host friendly bacteria. To our knowledge, antimicrobial properties have been described for a number of oligosaccharides (Daddaoua et al., 2006) and in a previous study, it was shown that FOS as a prebiotic had specific effects on P. aeruginosa, since it reduced growth, FIGURE 8 | Effect of PA0643, PA0644, and PA0646 deletion on the motility of P. aeruginosa PAO1. Motility assays were carried out as described in section "Materials and Methods." Shown are means and standard deviations with n = 3-6; *P < 0.05 vs. WT. limited the formation of biofilm, impaired motility, and reduced the inflammatory response (Ortega-González et al., 2014). Although, inulin and FOS are structurally related oligosaccharides, they differ in chain length and both compounds can be used as a carbon and energy source for P. aeruginosa growth.
The present study demonstrates that FOS induces a number of important changes in P. aeruginosa transcript levels some of which are related to bacterial survival (Figure 1) and provides an initial insight into the corresponding molecular mechanisms. As shown in Figure 2, in the 162 genes with reduced transcript levels by FOS and inulin, only 13% of them were affected by both compounds. Similarly, of the 443 genes with increased levels only 24% of gene transcript levels were altered by both compounds. Therefore, these compounds can be considered as signal molecules; however, their molecular mechanisms remain unknown.
RNA-Seq and rt-qPCR based comparative RNA profiling of P. aeruginosa PAO1 in the presence of FOS or inulin, not only highlighted known functions required for survival metabolism, but revealed a decrease in transcript levels of genes associated with carbohydrate metabolism and growth such as PA1183 (dctA) which has been shown to be associated previously with bacterial growth (Valentini et al., 2011). Thus, the growth reduction may be due to a reduction in transcript levels of genes associated with carboxylic acid transport, carboxylic acid metabolism, and reduction in ribosomal proteins.
Furthermore, these studies revealed a decrease of spuI gene transcript levels (PA0296), confirmed by rt-qPCR (Table 8), which control the expression of polyamine toxicity and of pyocins (PA3866), which are virulence factors that are induced by DNAdamaging agents, such as UV light and mitomycin C (Dean-Scholl and Martin, 2008).
The current analysis clearly shows that the pathogenesis of P. aeruginosa in the presence of FOS is compromised due to a decrease in several virulence-associated genes; i.e., alpB (PA0908), a holin-like protein that is required for lysis and the icmP gene (PA4370), which has been shown to degrade plasminogen activator (Christian et al., 2015) and may play a role in P. aeruginosa pathogenicity (Tables 5, 8).
Secretion of exotoxins through type III secretion systems (T3SSs) (Filloux, 2011) are related to acute infections in P. aeruginosa, while type VI secretion systems are often associated with chronic infections and biofilm formation (Silverman et al., 2012).
Here we show by rt-qPCR that FOS lowers the transcript levels of exsA encoding a transcriptional activator of the secretion system type (T3SS) as well as those of a key protein necessary for the secretion system type VI (T6SS) function, encoded by the hcp gene (Figure 4).
Interestingly, the RNA seq analysis data coincide with a report that demonstrated that the repression of the ptrB gene (PA0612) is implicated in the regulation of the T3SS under DNA damage stress conditions (Weihui and Shouguang, 2005). In addition, this report demonstrates that a repression of clpV (PA0090) regulates the biofilm formation and T6SS (Table 5). Altogether, our study is consistent with the notion that the FOS mediated reduction in pathogenicity is mediated by the secretion systems III and VI.
Further, we obtained initial data suggesting that FOS modulates bacterial resistance to antibiotics, since FOS reduced the expression of ampD (PA0807) transcription levels which leads to the constitutive hyperproduction of the beta-lactamase AmpC and consequently to an increase of the β-lactam resistance (Lindberg et al., 1987). Furthermore, the data shows that FOS lowers transcript levels of alpB (PA0908) a gene product which is related to cell responses to antibiotics (Alvarez-Ortega et al., 2010).
The RNA seq analysis demonstrated the presence of an island with different genes of unknown function, whose expression was completely inhibited in the presence of FOS (Figure 5 and Table 4). The deletion of the PA0643, PA0644, and PA0646 genes caused alterations in bacterial growth (Figure 6), biofilm formation (Figure 7) and motility (Figure 8). The alteration of the transcript levels of this island is thus another possible mechanism by which FOS reduces bacterial pathogenesis.

CONCLUSION
FOS containing supplements are currently being used to prevent gastrointestinal infections (Yasuda et al., 2012), suggesting that it is a valid strategy to combat Pseudomonas infection by potentially including FOS in antimicrobial cocktails. The present study is a contribution to close the gap of knowledge that exists in the corresponding molecular mechanisms.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/Supplementary Material.

AUTHOR CONTRIBUTIONS
JR-G, CS, and MG provided advice on experimental design. ZU and CS collected and assembled the data. AD designed the study, supervised and analyzed the data, and wrote the manuscript. TK and J-LR revised the manuscript. All authors reviewed and commented on the manuscript.

FUNDING
This work was supported by grants from the Spanish Ministry for Economy and Competitiveness (AGL2017-85270-R). CS is funded by the program Juan de la Cierva-Formación (FJCI-2015-23810).